Young talents in polymer science

Böker, Alexander

Wiesbrock, Frank

2017

Multidisciplinary Digital Publishing Institute

Polymers aims to compile the current trends and research directions of internationally renowned and successful young polymer scientists in one dedicated Special Issue. The idea for this Special Issue has arisen from the numerous and, above all, brilliant nominations for our 1st Young Investigator Award. It was the easiest-ever choice for us to publicly share the fascinating and thrilling research successes of the nominated scientists with you, our dear peer readers. For this prestigious Special Issue, the journal Polymers will only accept original research papers that have been invited exclusively by the editors. The contributions to this Special Issue will highlight the current state-of-the-art in the fields of polymerization methods, theory/simulation/modeling, new physical phenomena, advances in characterization techniques, and harnessing of self-assembly and biological strategies for producing complex multifunctional structures. Please enjoy reading these high-end research highlights

Chemistry (General)

English

978-3-03842-459-8

978-3-03842-458-1

polymers

Young Talents in
Polymer Science
Edited by

Alexander Böker and Frank Wiesbrock
Printed Edition of the Special Issue Published in Polymers

www.mdpi.com/journal/polymers

Young Talents in
Polymer Science

Special Issue Editors
Alexander Böker
Frank Wiesbrock

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade

Special Issue Editors
Alexander Böker
Universität Potsdam
Germany

Frank Wiesbrock
Polymer Competence Center Leoben GmbH
Austria

Editorial Office
MDPI AG
St. Alban-Anlage 66
Basel, Switzerland

This edition is a reprint of the Special Issue published online in the open access
journal Polymers (ISSN 2073-4360) from 2016–2017 (available at:
http://www.mdpi.com/journal/polymers/special_issues/young_talents).
For citation purposes, cite each article independently as indicated on the article
page online and as indicated below:
Author 1; Author 2. Article title. Journal Name Year, Article number, page range.
First Edition 2017

ISBN 978-3-03842-458-1 (Pbk)
ISBN 978-3-03842-459-8 (PDF)

Articles in this volume are Open Access and distributed under the Creative Commons Attribution
license (CC BY), which allows users to download, copy and build upon published articles even for
commercial purposes, as long as the author and publisher are properly credited, which ensures
maximum dissemination and a wider impact of our publications. The book taken as a whole is
© 2017 MDPI, Basel, Switzerland, distributed under the terms and conditions of the
Creative Commons license CC BY-NC-ND (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Table of Contents
About the Special Issue Editors ................................................................................................................ v
Preface to “Young Talents in Polymer Science” ..................................................................................... vii

Spencer D. Brucks, Jessica L. Freyer, Tristan H. Lambert and Luis M. Campos
Influence of Substituent Chain Branching on the Transfection Efficacy of Cyclopropenium-Based
Polymers
Reprinted from: Polymers 2017, 9(3), 79; doi: doi:10.3390/polym9030079 ............................................ 1
Lihua Guo, Xinyu Jing, Shuoyan Xiong, Wenjing Liu, Yanlan Liu, Zhe Liu and Changle Chen
Influences of Alkyl and Aryl Substituents on Iminopyridine Fe(II)- and Co(II)-Catalyzed Isoprene
Polymerization
Reprinted from: Polymers 2016, 8(11), 389; doi: 10.3390/polym8110389 ............................................... 10
Kanykei Ryskulova, Anupama Rao Gulur Srinivas, Thomas Kerr-Phillips, Hui Peng, David
Barker, Jadranka Travas-Sejdic and Richard Hoogenboom
Multiresponsive Behavior of Functional Poly(p-phenylene vinylene)s in Water
Reprinted from: Polymers 2016, 8(10), 365; doi: 10.3390/polym8100365 ............................................... 22
Anna P. Constantinou, Hanyi Zhao, Catriona M. McGilvery, Alexandra E. Porter and
Theoni K. Georgiou
A Comprehensive Systematic Study on Thermoresponsive Gels: Beyond the Common
Architectures of Linear Terpolymers
Reprinted from: Polymers 2017, 9(1), 31; doi: 10.3390/polym9010031................................................... 35
Kyriakos Christodoulou, Epameinondas Leontidis, Mariliz Achilleos, Christiana Polydorou
and Theodora Krasia-Christoforou
Semi-Interpenetrating Polymer Networks with Predefined Architecture for Metal Ion
Fluorescence Monitoring
Reprinted from: Polymers 2016, 8(12), 411; doi: 10.3390/polym8120411 ............................................... 52
Siew Yin Chan, Wee Sim Choo, David James Young and Xian Jun Loh
Thixotropic Supramolecular Pectin-Poly(Ethylene Glycol) Methacrylate (PEGMA) Hydrogels
Reprinted from: Polymers 2016, 8(11), 404; doi: 10.3390/polym8110404 ............................................... 65
Tianfeng Shen, Piming Ma, Qingqing Yu, Weifu Dong and Mingqing Chen
The Effect of Thermal History on the Fast Crystallization of Poly(L-Lactide) with Soluble-Type
Nucleators and Shear Flow
Reprinted from: Polymers 2016, 8(11), 431; doi: 10.3390/polym8120431 ............................................... 77
Sai Aditya Pradeep, Hrishikesh Kharbas, Lih-Sheng Turng, Abraham Avalos,
Joseph G. Lawrence and Srikanth Pilla
Investigation of Thermal and Thermomechanical Properties of Biodegradable PLA/PBSA
Composites Processed via Supercritical Fluid-Assisted Foam Injection Molding
Reprinted from: Polymers 2017, 9(1), 22; doi: 10.3390/polym9010022................................................... 89

iii

Jenny Alongi and Federico Carosio
All-Inorganic Intumescent Nanocoating Containing Montmorillonite Nanoplatelets in Ammonium
Polyphosphate Matrix Capable of Preventing Cotton Ignition
Reprinted from: Polymers 2016, 8(12), 430; doi: 10.3390/polym8120430 ............................................... 107
Benjamin Weber, Christine Seidl, David Schwiertz, Martin Scherer, Stefan Bleher, Regine Süss
and Matthias Barz
Polysarcosine-Based Lipids: From Lipopolypeptoid Micelles to Stealth-Like Lipids in Langmuir
Blodgett Monolayers
Reprinted from: Polymers 2016, 8(12), 427; doi: 10.3390/polym8120427 ............................................... 121
Takahiro Sakaue
Dynamics of Polymer Translocation: A Short Review with an Introduction of Weakly-Driven Regime
Reprinted from: Polymers 2016, 8(12), 424; doi: 10.3390/polym8120424 ............................................... 135
Hannah C. Bygd and Kaitlin M. Bratlie
Investigating the Synergistic Effects of Combined Modified Alginates on Macrophage Phenotype
Reprinted from: Polymers 2016, 8(12), 422; doi: 10.3390/polym8120422 ............................................... 147

iv

About the Special Issue Editors
Alexander Böker is Director of the Fraunhofer-Institute for Applied Polymer
Research (IAP) and holds the Chair for Polymer Materials and Polymer
Technology at the University of Potsdam, Germany. He studied chemistry at
the Johannes Gutenberg University, Mainz, and Cornell University, Ithaca, NY.
He received his PhD from Bayreuth University in 2002 working with
Prof. G. Krausch and Prof. Axel H.E. Müller. From 2002–2004, he was a
postdoctoral fellow with Thomas P. Russell at the University of Massachusetts,
Amherst. In 2006, he received a Lichtenberg-Professorship funded by the
VolkswagenStiftung and in 2015 the ERC Consolidator Grant. Alexander Böker
has published more than 130 papers in peer-reviewed journals. The
main research interests of his group include guided self-assembly of block
copolymer systems, hierarchical (bio)nanoparticle assemblies and the control of self-assembly
processes via external fields.
Frank Wiesbrock received his doctorate at the Technical University of Munich,
Germany, in 2003 for his work on metal ß-amino carboxylates under the
supervision of H. Schmidbaur. His subsequent biannual postdoctoral stay with
U.S. Schubert at the Eindhoven Technical University focused on block
copoly(2-oxazoline)s from microwave-assisted synthesis. He worked as a
project manager at Chemspeed Technologies AG in Augst, Switzerland, and
returned to academia in 2007 as a Marie Curie ToK researcher with T.
Calogeropoulou at the National Hellenic Research Foundation in Athens,
Greece. Since 2008, he has been an assistant professor/lecturer at the
Graz University of Technology, Austria, where he completed his Habilitation in
2012. Currently he is employed as a senior researcher at the Polymer
Competence Center Leoben (PCCL), Austria. His research interests comprise microwave-assisted
polymerizations, poly(2-oxazoline)s, biopolymers and biocompatible polymers, and nanocomposites
for electronic applications.

v

Preface to “Young Talents in Polymer Science”
Inspired by the excellent nominations for the 1st Young Investigator Award, the journal Polymers
compiled a selection of publications of these successful young polymer scientists in one Special Issue. It
was an easy choice for us to publish the recent research successes of the nominated high-ranking scientists
as an additional hardcover book to share with you, our dear readers. The Special Issue as well as the book
contains 12 contributions in the form of 11 original research articles and one review, highlighting the
exciting plethora of current research topics in the area of polymer science.
Polymer Synthesis and Macromolecular Architectures. In the article ′Influence of Substituent Chain
Branching on the Transfection Efficacy of Cyclopropenium-Based Polymers′ by the 2016 award winner,
Luis M. Campos [1] and colleagues, the authors correlate the degree of alkyl chain branching of cationic
polymers based on trisaminocyclopropenium ions with their cytotoxicity, transfection efficacy, and their
performance as nonviral vectors in gene therapy [2]. Changle Chen et al. describe a series of
iminopyridine FeII and CoII complexes as catalysts for the polymerization of isoprene in the publication
′Influences of Alkyl and Aryl Substituents on Iminopyridine Fe(II)- and Co(II)-Catalyzed Isoprene
Polymerization′ with particular focus on the cis-1,4-selectivity of the corresponding poly(isoprene)s [3].
Richard Hoogenboom and colleagues report the multiresponsive behavior of different 2,5-substituted
poly(phenylene vinylene)s (with varying amounts of carboxylic acid units) upon stimuli such as pH
changes, temperatures, and the addition of divalent ions in the article ′Multiresponsive Behavior of
Functional Poly(p-phenylene vinylene)s in Water′ [4].
Methacrylate-Based Crosslinked Polymers and (Hydro-)Gels. Theoni K. Georgiou et al. describe
the synthesis of different methacrylate-based diblock terpolymers and correlate the copolymers′
architectures with their thermoresponsive properties and sol–gel transitions in the publication ′A
Comprehensive Systematic Study on Thermoresponsive Gels: Beyond the Common Architectures of
Linear Terpolymers′ [5]. In the article ′Semi-Interpenetrating Polymer Networks with Predefined
Architecture for Metal Ion Fluorescence Monitoring′, Theodora Krasia-Christoforou and coworkers report
the preparation of methacrylate-based semi-interpenetrating networks and their applicability as sorbents
and fluorescence chemosensors [6]. David James Young, Xian Jun Loh and colleagues describe the
grafting of methoxy-poly(ethylene glycol) methacrylate onto pectin and the subsequent formation of the
corresponding supramolecular hydrogels with thixotropic properties in the study ′Thixotropic
Supramolecular Pectin-Poly(Ethylene Glycol) Methacrylate (PEGMA) Hydrogels’ [7].
Polymer Blends and Composites. Piming Ma et al. report the promotion of the crystallization
process of a blend composed of poly(L-lactide) and a phenyloxalamide-based nucleator in the publication
‘The Effect of Thermal History on the Fast Crystallization of Poly(L-Lactide) with Soluble-Type
Nucleators and Shear Flow’ [8]. Srikanth Pilla and colleagues summarize the processing of blends of
poly(lactic acid) and poly(butylene succinate-co-adipate) for the production of biobased polymer foams in
the report ′Investigation of Thermal and Thermomechanical Properties of Biodegradable PLA/PBSA
Composites Processed via Supercritical Fluid-Assisted Foam Injection Molding′ [9]. A completely
inorganic intumescent flame retardant nanocomposite is reported by Jenny Alongi et al. in the study ′AllInorganic Intumescent Nanocoating Containing Montmorillonite Nanoplatelets in Ammonium
Polyphosphate Matrix Capable of Preventing Cotton Ignition′ [10].
Medic(in)al Applications of Polymers. Matthias Barz and coworkers describe the synthesis of
polysarcosine-based lipids by nucleophilic ring-opening polymerizations and evaluate their potential to
substitute poly(ethylene glycol)-based lipopolymers in the publication ′Polysarcosine-Based Lipids: From
Lipopolypeptoid Micelles to Stealth-Like Lipids in Langmuir Blodgett Monolayers′ [11]. A concise
summary of the current state-of-the-art knowledge of polymer translocation and a crossover
scenario connecting unbiased and strongly-driven regimes is provided by Takahiro Sakaue in the
review ′A Short Review with an Introduction of Weakly-Driven Regime′ [12]. The effect of various
chemically modified alginates on the reprogramming capabilities of macrophages is reported in the
publication ′Investigating the Synergistic Effects of Combined Modified Alginates on Macrophage
Phenotype′ by Kaitlin M. Bratlie et al. [13].
vii

Enjoy reading this collection of scientifically high-end research highlights, addressing the current
′hot′ research topics!
Alexander Böker and Frank Wiesbrock
Special Issue Editors
References
1.
2.

3.

4.

5.

6.

7.
8.

9.

10.

11.

12.
13.

Böker, A. Announcement of the 2016 Polymers Young Investigator Award. Polymers 2016, 8, 2,
doi:10.3390/polym8030065.
Brucks, S.D.; Freyer, J.L.; Lambert, T.H.; Campos, L.M. Influence of Substituent Chain Branching on
the Transfection Efficacy of Cyclopropenium-Based Polymers. Polymers 2017, 9, 9,
doi:10.3390/polym9030079.
Guo, L.; Jing, X.; Xiong, S.; Liu, W.; Liu, Y.; Liu, Z.; Chen, C. Influences of Alkyl and Aryl
Substituents on Iminopyridine Fe(II)- and Co(II)-Catalyzed Isoprene Polymerization. Polymers 2016,
8, 12, doi:10.3390/polym8110389.
Ryskulova, K.; Rao Gulur Srinivas, A.; Kerr-Phillips, T.; Peng, H.; Barker, D.; Travas-Sejdic, J.;
Hoogenboom, R. Multiresponsive Behavior of Functional Poly(p-phenylene vinylene)s in Water.
Polymers 2016, 8, 13, doi:10.3390/polym8100365.
Constantinou, A.P.; Zhao, H.; McGilvery, C.M.; Porter, A.E.; Georgiou, T.K. A Comprehensive
Systematic Study on Thermoresponsive Gels: Beyond the Common Architectures of Linear
Terpolymers. Polymers 2017, 9, 17, doi:10.3390/polym9010031.
Christodoulou, K.; Leontidis, E.; Achilleos, M.; Polydorou, C.; Krasia-Christoforou, T. SemiInterpenetrating Polymer Networks with Predefined Architecture for Metal Ion Fluorescence
Monitoring. Polymers 2016, 8, 13, doi:10.3390/polym8120411.
Chan, S.Y.; Choo, W.S.; Young, D.J.; Loh, X.J. Thixotropic Supramolecular Pectin-Poly(Ethylene
Glycol) Methacrylate (PEGMA) Hydrogels. Polymers 2016, 8, 12, doi:10.3390/polym8110404
Shen, T.; Ma, P.; Yu, Q.; Dong, W.; Chen, M. The Effect of Thermal History on the Fast
Crystallization of Poly(L-Lactide) with Soluble-Type Nucleators and Shear Flow. Polymers 2016, 8,
12, doi:10.3390/polym8120431.
Pradeep, S.A.; Kharbas, H.; Turng, L.-S.; Avalos, A.; Lawrence, J.G.; Pilla, S. Investigation of
Thermal and Thermomechanical Properties of Biodegradable PLA/PBSA Composites Processed via
Supercritical
Fluid-Assisted
Foam
Injection
Molding.
Polymers
2017,
9,
18,
doi:10.3390/polym9010022.
Alongi, J.; Carosio, F. All-Inorganic Intumescent Nanocoating Containing Montmorillonite
Nanoplatelets in Ammonium Polyphosphate Matrix Capable of Preventing Cotton Ignition.
Polymers 2016, 8, 14, doi:10.3390/polym8120430.
Weber, B.; Seidl, C.; Schwiertz, D.; Scherer, M.; Bleher, S.; Süss, R.; Barz, M. Polysarcosine-Based
Lipids: From Lipopolypeptoid Micelles to Stealth-Like Lipids in Langmuir Blodgett Monolayers.
Polymers 2016, 8, 14, doi:10.3390/polym8120427.
Sakaue, T. Dynamics of Polymer Translocation: A Short Review with an Introduction of WeaklyDriven Regime. Polymers 2016, 8, 12, doi:10.3390/polym8120424.
Bygd, H.C.; Bratlie, K.M. Investigating the Synergistic Effects of Combined Modified Alginates on
Macrophage Phenotype. Polymers 2016, 8, 16, doi:10.3390/polym8120422.

viii

polymers
Communication

Influence of Substituent Chain Branching
on the Transfection Efficacy of
Cyclopropenium-Based Polymers
Spencer D. Brucks, Jessica L. Freyer, Tristan H. Lambert and Luis M. Campos *
Department of Chemistry, Columbia University, 3000 Broadway, New York, NY 10027, USA;
sdb2147@columbia.edu (S.D.B.); jlf2176@columbia.edu (J.L.F.); tl2240@columbia.edu (T.H.L.)
* Correspondence: lcampos@columbia.edu; Tel.: +1-212-854-9561
Academic Editor: Wei Min Huang
Received: 6 January 2017; Accepted: 21 February 2017; Published: 24 February 2017

Abstract: The realization of gene therapy relies on the development of delivery vectors with high
efficiency and biocompatibility. With a multitude of structures accessible, the core challenge is
precisely tuning vector structure to probe and optimize structure–property relationships. Employing
a modular strategy, two pairs of cationic polymers based on the trisaminocyclopropenium (TAC) ion
were synthesized where the substituents differ in the degree of alkyl chain branching. All TAC-based
polymers exhibited higher transfection efficiencies than the untreated controls, with variable in vitro
toxicities. Considering both cytotoxicity and transfection efficacy, an optimal nonviral vector was
identified. Our studies highlight the importance of exercising precise control over polymer structure,
both in terms of backbone identity and substituent nature, and the necessity of a robust, modular
platform from which to study them.
Keywords: gene delivery; polyelectrolyte; nonviral vectors; structure-property relationships

1. Introduction
The fundamental challenge of gene therapy is the design of efficient, biocompatible delivery
vectors [1–3]. Since the concept of gene delivery was first introduced, both viral and nonviral options have
been explored, with each presenting their own benefits and drawbacks [1,4]. While viral technologies
demonstrate high gene expression, their clinical translation has been limited by immunogenicity, lack of
selectivity, and synthetic difficulty [2,5]. On the contrary, nonviral vectors elicit a reduced immunogenic
response, but thus far have shown lower delivery efficiencies. The development of an optimized
nonviral vector, balancing safety with efficiency, thus remains a critical goal to the realization of gene
therapy [1,2,6].
Cationic polymers are one of the most commonly studied nonviral vectors due to their high stability
and capacity to tune macromolecular composition and architecture [7,8]. Linear polyethylenimine
(PEI) was the first cationic polymer observed to rapidly complex negatively charged nucleic acids into
polyplexes [9]. In the subsequent decades, there has been a great proliferation of cationic structures
shown to bind and transfect nucleic acids. In addition to increasing the library of cationic polymers
available, these research efforts have also uncovered some design principles towards optimizing
polymeric structure for transfection [1,7]. In general, lower molecular weight polymers demonstrate
lowered cytotoxicity than their high molecular weight counterparts [10–13], and increasing the degree
of polymer chain branching appears to heighten both transfection efficacy and toxicity [14–17]. Wang
and coworkers have recently highlighted the great potential of branched and dendritic polymer
topologies to enhance transfection efficiency [18–21]. With the advent of controlled polymerization

Polymers 2017, 9, 79

1

www.mdpi.com/journal/polymers

Polymers 2017, 9, 79

techniques, the challenge has now become fine-tuning polymer structure, size, and architecture to
probe detailed structure–property relationships among the various polymers accessible.
We recently reported a modular platform to functionalize polymers of various sizes and architectures
with a host of bis(dialkylamino)cyclopropenium chloride (BACCl) derivatives [22]. In an efficient
post-polymerization click reaction, polymers bearing pendent or main-chain secondary amines were
quantitatively transformed into the aromatic trisaminocyclopropenium (TAC) ion. We found that some
of the resulting TAC polymers were biocompatible and efficient transfection agents and, furthermore,
the ability to precisely tune polymer structure had significant effects on macromolecular properties,
including transfection efficiency.
Herein, we extend these results by probing the effect of the substituted alkylamino chain’s degree
of branching on cytotoxicity and transfection efficacy. Pendent moieties have been shown to modify the
structure and stability of polymer–DNA polyplexes to facilitate cellular release, with several reports
exploring the effects of fine tuning a substituent alkyl chain length [23–25]. However, comparably few
studies have investigated the nature of branching within an alkyl group and its effect on transfection
efficacy. As our platform is amenable to a wide variety of BACCl derivatives, we synthesized two
that would directly compare the branching of the alkyl chain: n-butyl (Bu; BACBu) and isopropyl
(iP; BACiP). Employing these ionic liquids in the post-polymerization functionalization of two
polymeric backbones, poly(methylaminostyrene) (PMAS) and polyethylenimine (PEI), furnished
a total of four polymers that we investigated as nonviral vectors (Figure 1). We found that there exist
important synergies between the polymeric backbone and the nature of the substituent, and that the
ability to simultaneously manipulate both is instrumental for the optimization of efficient nonviral
gene delivery vectors.

Figure 1. Trisaminocyclopropenium (TAC) polymer structures examined for biocompatibility
and transfection efficacy. PEI: polyethylenimine; PMAS: poly(methylaminostyrene); Bu: n-butyl;
iP: isopropyl.

2. Materials and Methods
2.1. Materials
All materials were purchased from Sigma-Aldrich (St. Louis, MO, USA) and were used without
further purification except as noted below. Deuterated solvents used for NMR spectroscopy were

2

Polymers 2017, 9, 79

purchased from Thermo Fisher Scientific (Waltham, MA, USA). Spectrum Labs dialysis bags were
purchased from VWR (Radnor, PA, USA). Organic solutions were concentrated by use of a Buchi
(New Castle, DE, USA) rotary evaporator.
2.2. Procedure for the Synthesis of 1,2-Bis(dibutylamino)-3-chlorocyclopropenium Chloride (BACBu)
2.2.1. Synthesis of 2,3-Bis(dibutylamino)-1-cyclopropenone
This procedure was performed at ambient conditions. Dibutylamine (24.5 g, 190 mmol, 8.0 equiv)
was slowly added to a solution of pentachlorocyclopropane (5.0 g, 23.6 mmol, 1.0 equiv) in CH2 Cl2
(250 mL) in a 1 L round-bottom flask at 0 ◦ C. The solution turned orange, and was allowed to
warm to room temperature with stirring overnight. The reaction mixture was washed with 1 M
HCl (3 × 100 mL), deionized (DI) water (1 × 100 mL), and brine (1 × 100 mL) dried over magnesium
sulfate, and concentrated in vacuo to yield a crude orange solid. The solid was dissolved in tert-butanol
(50 mL) and to this was added potassium hydroxide (10 g, 178 mmol) in DI water (15 mL). The solution
was heated at 70 ◦ C for 2 h, and then water was removed by rotary evaporation. The resulting solid
was dissolved in CH2 Cl2 and filtered to remove salt. The organic solution was dried with anhydrous
magnesium sulfate, and concentrated in vacuo to a crude yellow oil. The crude material was purified
by silica gel chromatography (100% EtOAc; 5% MeOH in CH2 Cl2 ) to yield the title product as an
orange solid (2.19 g, 7.08 mmol, 30% two-step yield). 1 H NMR (400 MHz, CDCl3 ) δ 3.16 (t, 8H,
NCH2 CH2 CH2 CH3 ), 1.59 (m, 8H, NCH2 CH2 CH2 CH3 ), 1.34 (m, 8H, NCH2 CH2 CH2 CH3 ), 0.94 (t, 12H,
NCH2 CH2 CH2 CH3 ).
2.2.2. Synthesis of 1,2-Bis(dibutylamino)-3-chlorocyclopropenium Chloride (BACBu)
Oxalyl chloride (0.09 mL, 0.9 mmol, 2.0 equiv) was slowly added to a solution of
2,3-bis(dibutylamino)-1-cyclopropenone (0.150 g, 0.45 mmol, 1.0 equiv) in dry CH2 Cl2 (5 mL) at
0 ◦ C under argon. The solution was warmed to room temperature and allowed to react for 1 h.
The solution was then concentrated in vacuo to yield the title product as a brown liquid in quantitative
yield. 1 H NMR (400 MHz, CDCl3 ) δ 3.64 (t, 4H, NCH2 CH2 CH2 CH3 ), 3.50 (t, 4H, NCH2 CH2 CH2 CH3 ),
1.76 (m, 4H, NCH2 CH2 CH2 CH3 ), 1.66 (m, 4H, NCH2 CH2 CH2 CH3 ), 1.40 (m, 8H, NCH2 CH2 CH2 CH3 ),
0.99 (t, 12H, NCH2 CH2 CH2 CH3 ).
2.3. Synthesis of PMAS(Bu)
The procedure was performed open to the atmosphere. Poly(methylaminostyrene) (PMAS)
(50 mg, 0.3 mmol, 1 equiv, DP: 50, Mn : 7400, Mw : 8300, Đ: 1.08), synthesized according to previously
reported procedures [22], was dissolved in CHCl3 (8 mL) in a scintillation vial equipped with a stir
bar. To the vial was added N,N-diisopropylethylamine (115 mg, 0.9 mmol, 3 equiv) and BACBu
(160 mg, 0.46 mmol, 1.5 equiv). The reaction mixture was allowed to stir at 65 ◦ C for 3 h. The resulting
solution was concentrated in vacuo, dissolved in minimum acetone and precipitated once into ethyl
acetate at −78 ◦ C. The resulting powder was dissolved in methanol and transferred to a 3.5k MWCO
Spectrum Labs dialysis bag and dialyzed against methanol followed by concentration under vacuum
to yield a pale brown powder (60 mg, 42% yield). 1 H NMR (500 MHz, CDCl3 ) δ 7.18–6.07 (b, 200H,
ArH), 4.99–4.39 (b, 100H, ArCH2 N), 3.78–2.91 (b, 575H, NCH3 , NCH2 CH2 CH2 CH3 ), 1.98–1.58 (b,
400H, NCH2 CH2 CH2 CH3 ) 1.39–1.08 (b, 650H, NCH2 CH2 CH2 CH3 , ArCHCH2 ) 1.02–0.63 (b, 600H,
NCH2 CH2 CH2 CH3 ).
2.4. Synthesis of PEI(Bu)
This procedure was performed open to the atmosphere. Linear 25k polyethyleneimine (20 mg,
0.46 mmol, 1 equiv) was dissolved in CHCl3 (8 mL) in a scintillation vial equipped with a stir bar.
To the vial was added N,N-diisopropylethylamine (180 mg, 1.4 mmol, 3 equiv) and BACBu (250 mg,
0.70 mmol, 1.5 equiv). The reaction mixture was allowed to stir at 65 ◦ C for 3 h. The resulting solution

3

Polymers 2017, 9, 79

was concentrated in vacuo and precipitated once into ethyl acetate at −78 ◦ C. The resulting powder
was dissolved in methanol and transferred to a 3.5k MWCO Spectrum Labs dialysis bag and dialyzed
against methanol. The resulting solution was concentrated vacuum to yield a yellow-brown powder
(25 mg, 15% yield). 1 H NMR (400 MHz, CDCl3 ) δ 4.45–3.15 (b, 7000H, CH2 CH2 N)581 , NCH2 CH2 CH2 CH3 )
1.71–1.55 (b, 3500H, NCH2 CH2 CH2 CH3 ), 1.43–1.22 (b, 5400H, NCH2 CH2 CH2 CH3 ), 1.00–0.90 (b, 6200H,
NCH2 CH2 CH2 CH3 ).
2.5. Synthesis of 1,2-Bis(diisopropylamino)-3-chlorocyclopropenium Chloride (BACiP), PMAS(iP), and PEI(iP)
The preparations of BAC(iP) and the subsequently derivatized polymers have been previously
reported [22,26].
2.6. Cell Culture
HEK-293T cells (American Type Culture Collection, Manassas, VA, USA) were grown in
Dulbecco’s Modified Eagle Medium (DMEM) with L-glutamine (Gibco, Grand Island, NY, USA)
supplemented with 10% fetal bovine serum (FBS) (Atlanta Biologicals, Flowery Branch, GA, USA) and
1% penicillin/streptomycin (Gibco, Grand Island, NY, USA). Cultures were incubated in humidified
tissue incubators at 37 ◦ C and 5% CO2 .
2.7. Cell Viability
Trypan blue dye exclusion cell counting was performed in triplicate with an automated cell
counter (ViCell, Beckman-Coulter, Brea, CA, USA). Cell viability under experimental conditions is
reported as a percentage relative to untreated cells.
2.8. Cell Transfection and Luciferase Expression
HEK-293T cells were seeded on 12-well plates at a density of 5 × 104 cells/well 24 h prior
to transfection. The media was then evacuated, replaced with fresh, and supplemented with
polymer–pDNA polyplexes. Polyplexes were prepared by adding polymer solutions in RNase-free
water to 3 μg of plasmid DNA (pDNA) (gWiz-Luciferase, Aldevron, Fargo, ND, USA) at indicated
loadings, and vortexing at 1500 rpm for 3 min at room temperature. After 48 h of incubation, cell
viability was measured, and cells were re-plated on 96-well plates at a density of 5 × 103 cells/well.
After 24 h of incubation, cells were analyzed for luciferase activity according to the manufacturer’s
protocol. Briefly, cells were rinsed with phosphate-buffered saline (PBS) and lysed with 20 μL/well
1 × Cell Lysis Buffer (Promega, Madison, WI, USA). To the cell lysates was added 100 μL/well of
Luciferase Assay Reagent (Promega) and the light produced was measured immediately on a plate
reader (PerkinElmer, Waltham, MA). Results were expressed as relative light units (RLU) normalized
to cell counts, with error bars showing the standard deviation of triplicate measurement.
2.9. Hydrodynamic Size and Zeta Potential Measurement
Polyplex size and zeta potential were measured on a Malvern Zetasizer Nano ZS (Malvern
Instruments, Malvern, UK). For all measurements, polyplexes were diluted 1:100 in Milli-Q
water at neutral pH. The reported diameters are the average of three measurements, where each
measurement comprises at least 10 acquisitions, and the zeta potential was calculated according to the
Smoluchowski approximation.
2.10. Gel Electrophoresis Shift Assay
Polyplexes were prepared at different weight ratios by adding 10 μL of polymer in Milli-Q H2 O to
10 μL of pDNA (5 ng/μL), and vortexing at 1500 rpm for 3 min at room temperature. To the polyplex
solution was then added 2 μL of loading dye, for a total volume of 22 μL, which was subsequently

4

Polymers 2017, 9, 79

added to the well. Agarose gels were prepared as 1 wt % in tris-acetate EDTA (TAE) buffer with 2 μL
ethidium bromide and run at 100 V for 20 min. Gels were visualized under UV illumination at 365 nm.
3. Results and Discussion
3.1. Cationic Polymer Synthesis
The candidate nonviral vectors were synthesized from neutral parent precursor polymers
in a post-polymerization functionalization strategy. For this study, we employed two parent
polymers—PMAS (7.5 kg·mol− 1 ), synthesized according to literature [22], and commercially
available PEI (25 kg·mol–1 ; linear)—to compare the effects of backbone structure. Both were
subsequently transformed into TAC derivatives by reaction with a BACCl salt, in a “click” conjugation
reaction proceeding under mild conditions with stoichiometric amounts of reactants. The cationic
nature of cyclopropenium-based polymers prevents their characterization by gel permeation
chromatography [26], so PMAS was shown to have a narrow dispersity less than 1.1 (Supplementary
Materials, Figure S1), and complete functionalization was confirmed by NMR.
We elected to study two BACCl structures comprising dialkylamino substituents differing in
the degree of branching, and thereby “floppiness”, as well as hydrophobicity, as the Bu-derivatized
polymers have one more carbon. Complete functionalization of both BAC(Bu), containing n-butyl
substituents, and BAC(iP), containing isopropyl substituents, was confirmed by proton nuclear
magnetic resonance spectroscopy. Quantitatively functionalizing polymers holds effects of dispersity
and degree of polymerization constant, permitting direct comparisons of subtle structural changes on
macromolecular properties.
3.2. Biocompatibility Studies
Careful engineering of cationic polymers is necessary to enable permeation of the cell membrane
without reducing overall viability. Studies have shown that cationic polymers form pores in the cell
membrane to mediate entry, but they also reduce overall cell viability [27]. Pore formation typically
involves intercalation of the cell membrane’s lipid bilayer by aliphatic groups on the delivery agent;
thus, tuning TAC’s alkyl substituents represents an opportunity to promote cell entry while minimizing
cell death associated with membrane destabilization. Therefore, to probe the impact of alkyl chain
conformation on cell viability and transfection efficiency, the series of TAC-functionalized polymers
were assessed and compared as vectors via cytotoxicity assays and luciferase transfection experiments
in HEK-293T cells.
All four homopolymers were highly water-soluble, permitting their condensation with an aqueous
solution of plasmid DNA (pDNA) containing the firefly luciferase reporter gene. Combining the
polymers at varied loadings with a fixed amount of pDNA, and subsequently incubating in cells for
2 d, revealed the polymers’ biocompatibility as a function of loading. We found that PEI(iP) and
PMAS(Bu) were the most biocompatible with high cell viabilities through loadings of 20 μg·mL− 1
(Figure 2). Surprisingly, their counterparts, PEI(Bu) and PMAS(iP), exhibited notable toxicity at all
loadings tested. All TAC-derived polymers here were highly toxic at loadings of 50 μg·mL− 1 and
greater, similar to linear PEI. This stands in contrast to our previous work, where polymers with more
rigid amino substituents on the TAC ion were still viable in this regime [22]. Thus, rigidity or flexibility
of substituent chains stand as an important parameter to understand for optimal gene transfection.
These results suggest there is a complex interaction between a polymer backbone and its substituent in
the design of biocompatible gene delivery vectors.

5

Polymers 2017, 9, 79

Figure 2. Biocompatibility of TAC-based polymers at various doses in HEK-293T cells following 48 h
incubation. Viability is measured by trypan blue dye exclusion and normalized to untreated cells.
Error bars show the standard deviation of triplicate measurement.

3.3. DNA Binding and In Vitro Gene Transfection
In order to affect transfection, candidate gene delivery vectors must condense nucleic acids
into a polyplex, permeate the cell membrane, escape the endosome, and release their payload [27].
We found that fine tuning the substituent chain branching had a dramatic influence on delivery efficacy.
In order to assess the amount of polymer necessary to completely condense pDNA into a polyplex,
gel electrophoresis shift assays were performed (Figures S2–S5). While all polymers were able to fully
bind the pDNA by a weight ratio of 3:33:1, PEI(iP) was the most efficient, binding at a weight ratio
of only 0.83:1. This corresponds to the lowest polymer loading tested for either biocompatibility or
transfection, and less than 1 TAC unit per phosphate anion of pDNA (Tables S1 and S2). However,
binding efficiency is not a clear indicator of delivery efficiency, as too favorable an interaction can be
detrimental for eventual release of the genetic material.
Luciferase expression assays revealed a significant dependence on the nature of the amino
substituent and polymer backbone for successful gene delivery. While all TAC-based polymers
transfected pDNA significantly better than the untreated controls, PMAS(iP) demonstrated the highest
transfection efficacy (Figure 3). As is the case with unmodified linear PEI, successful delivery of
intact pDNA to cells comes at the cost of significant cytotoxicity. By contrast, the nontoxic PEI(iP)
demonstrated a much lower luciferase activity. Interestingly, PEI(iP) was the most efficient at
compacting pDNA into a polyplex, suggesting that it binds nucleic acids too strongly and never releases
its payload. Converting either of the polymer backbones into a TAC bearing n-butyl chains seemed
to yield successful nonviral vectors capable of both binding and slowly releasing pDNA. This could
potentially be attributed to critical destabilization of the cell and endosomal membranes due to the
long, flexible alkyl substituents. As none of the TAC polymers examined here contain any protonatable
centers, it is unlikely that polyplexes escape the endosome via a proton sponge mechanism [24]. At their
optimal loadings, both PEI(Bu) and PMAS(Bu) demonstrated two orders-of-magnitude improvement
over untreated control cells. Taken together with the cytotoxicity and pDNA-binding data, we conclude
that amongst this family of TAC polymers, PMAS(Bu) is the most potent nonviral vector.

6

Polymers 2017, 9, 79

Figure 3. Luciferase expression in HEK-293T cells transfected with pDNA containing the firefly
luciferase reporter gene using TAC polymers. Luciferase expression is measured after 48 h incubation
with specified polymer loadings (all with pDNA loadings of 3 μg·mL−1 ) and normalized by cell count.
Error bars show the standard deviation of triplicate measurement.

3.4. Hydrodynamic Size and Zeta Potential Measurements
Dynamic light scattering determined the hydrodynamic diameter (DH ) and zeta potential (ζ) of the
polyplexes at their optimal loading for transfection efficacy (Table 1). All four cationic polymers formed
stable polyplexes of small sizes and highly positive charge. The polyplexes all exhibit a hydrodynamic
diameter in the size regime considered optimal for successful gene transfection. Notably, the polymers
modified with BAC(iP) resulted in more positively charged polyplexes than those with BAC(Bu),
which could be a result of enhanced hydrophobic screening of the charge by the longer, flexible n-butyl
chains [20,28]. These data suggest the dramatic differences in cytotoxicity and transfection efficacy
between the four TAC-based polymers are likely a result of structural variations rather than significant
size or surface charge differences.
Table 1. Characterization of transfection agents and polyplexes at optimal transfection efficacy.
Transfection agent 1

MM 2 (kDa)

Charge Ratio 3

DH (nm)

ζ Potential (mV)

PEI(Bu)
PEI(iP)
PMAS(Bu)
PMAS(iP)

215
182
25
21

6:1
35:1
12:1
5:1

110 ± 40
100 ± 40
150 ± 50
160 ± 40

45 ± 7
58 ± 5
31 ± 8
44 ± 6

1

Polyplexes of polymers at the loading corresponding to highest transfection efficacy in Figure 3. 2 Molecular mass
of transfection agent, calculated based on commercial linear 25k PEI; for PMAS(R) materials, PMAS was measured
by gel permeation chromatography (GPC) calibrated using polystyrene (PS) standards of narrow dispersity, then
calculated for the corresponding TAC group. 3 Ratio of TAC to phosphate anions.

4. Conclusions
We have shown that fine-tuning polymer structure can have dramatic effects on macromolecular
properties—in particular, cytotoxicity and gene delivery. Beginning with two discrete parent polymers,
we synthesized two pairs of TAC-based polymers differing solely in alkyl substituent identity,
branching, and polymer backbone. While none of the examined polymers surpass linear PEI in
transfection efficiency, our results demonstrate that there is an important interplay between polymer
backbone and substituent structure, and that both must be carefully considered in the design of
nonviral vectors. Amongst our examined TAC-based polymers, PMAS(Bu) exhibited the best balance
of biocompatibility, efficient DNA binding, and transfection efficacy. While hydrophobic modifications
7

Polymers 2017, 9, 79

of nonviral vectors are frequently reported to promote transfection, our results demonstrate that
a careful balance of hydrophobicity and substituent flexibility must be achieved for optimal gene
delivery. Importantly, this work exemplifies that design of transfection reagents demands precise
control over all aspects of polymer structure and a robust, modular platform from which to study
them. Only through these kinds of systematic studies can an optimized nonviral vector to achieve
successful gene delivery be developed.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/9/3/79/s1,
Figure S1: GPC trace of poly(methylaminostyrene), the parent polymer for both PMAS(Bu) and PMAS(iP);
Figure S2: Gel electrophoresis shift assay of pDNA polyplexes formed with PMAS(Bu) at the indicated polymer:
pDNA weight ratios. All pDNA is bound in polyplexes at a weight ratio of 1.66 PMAS(Bu): 1 pDNA; Figure S3:
Gel electrophoresis shift assay of pDNA polyplexes formed with PEI(Bu) at the indicated polymer:pDNA weight
ratios. All pDNA is bound in polyplexes at a weight ratio of 3.33 PEI(Bu): 1 pDNA; Figure S4: Gel electrophoresis
shift assay of pDNA polyplexes formed with PMAS(iP) at the indicated polymer:pDNA weight ratios. All pDNA
is bound in polyplexes at a weight ratio of 1.66 PMAS(iP): 1 pDNA; Figure S5: Gel electrophoresis shift assay
of pDNA polyplexes formed with PEI(iP) at the indicated polymer:pDNA weight ratios. All pDNA is bound
in polyplexes at a weight ratio of 0.83 PEI(iP): 1 pDNA; Table S1: Polymer loading and pDNA weight ratios for
gel electrophoresis and transfection experiments; Table S2: Conversion of weight ratios to charge ratios for each
tested polymer.
Acknowledgments: This work was funded by the National Science Foundation (NSF CAREER DMR-1351293,
and CHE 1464992), ACS Petroleum Research Fund, and 3M Non-Tenured Faculty Award. S.D.B. is grateful for
NSF GRFP (DGE-16-44869). We are grateful to the Stockwell lab for use of their cell culture hoods and plate reader.
Author Contributions: Spencer D. Brucks, Jessica L. Freyer, Tristan H. Lambert, and Luis M. Campos conceived
and designed the experiments; Spencer D. Brucks and Jessica L. Freyer performed the experiments; The manuscript
was written by Spencer D. Brucks, Jessica L. Freyer, and Luis M. Campos, with contributions from all authors.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.
4.
5.
6.
7.

8.
9.

10.
11.

12.

Yin, H.; Kanasty, R.L.; Eltoukhy, A.A.; Vegas, A.J.; Dorkin, J.R.; Anderson, D.G. Non-viral vectors for
gene-based therapy. Nat. Rev. Genet. 2014, 15, 541–555. [CrossRef] [PubMed]
Mintzer, M.A.; Simanek, E.E. Nonviral vectors for gene delivery. Chem. Rev. 2009, 109, 259–302. [CrossRef]
[PubMed]
Bender, E. Gene therapy: Industrial strength. Nature 2016, 537, S57–S59. [CrossRef] [PubMed]
Li, S.D.; Huang, L. Non-viral is superior to viral gene delivery. J. Control. Release 2007, 3, 181–183. [CrossRef]
[PubMed]
Kay, M.A. State-of-the-art gene-based therapies: The road ahead. Nat. Rev. Genet. 2011, 5, 316–328. [CrossRef]
[PubMed]
Kostarelos, K.; Miller, A.D. Synthetic, self-assembly ABCD nanoparticles: A structural paradigm for viable
synthetic non-viral vectors. Chem. Soc. Rev. 2005, 34, 970–994. [CrossRef] [PubMed]
Samal, S.K.; Dash, M.; Van Vlierberghe, S.; Kaplan, D.L.; Chiellini, E.; van Blitterswijk, C.; Moroni, L.;
Dubruel, P. Cationic polymers and their therapeutic potential. Chem. Soc. Rev. 2012, 21, 7147–7194.
[CrossRef] [PubMed]
Merdan, T.; Kopeček, J.; Kissel, T. Prospects for cationic polymers in gene and oligonucleotide therapy
against cancer. Adv. Drug Deliv. Rev. 2002, 54, 715–758. [CrossRef]
Boussif, O.; Lezoualc’h, F.; Zanta, M.A.; Mergny, M.D.; Scherman, D.; Demeneix, B.; Behr, J.P. A versatile
vector for gene and oligonucleotide transfer into cells in culture and in vivo: Polyethylenimine. Proc. Natl.
Acad. Sci. USA 1995, 92, 7297–7301. [CrossRef] [PubMed]
Godbey, W.T.; Wu, K.K.; Mikos, A.G. Size matters: Molecular weight affects the efficiency of
poly(ethylenimine) as a gene delivery vehicle. J. Biomed. Mater. Res. 1998, 3, 268–275. [CrossRef]
Fischer, D.; Bieber, T.; Li, Y.; Elsässer, H.P.; Kissel, T. A novel non-viral vector for DNA delivery based on low
molecular weight, branched polyethylenimine: Effect of molecular weight on transfection efficiency and
cytotoxicity. Pharm. Res. 1999, 8, 1273–1279. [CrossRef]
Layman, J.M.; Ramirez, S.M.; Green, M.D.; Long, T.E. Influence of polycation molecular weight on
poly(2-dimethylaminoethyl methacrylate)-mediated DNA delivery in vitro. Biomacromolecules 2009, 10,
1244–1252. [CrossRef] [PubMed]
8

Polymers 2017, 9, 79

13.

14.
15.

16.

17.

18.
19.

20.

21.
22.

23.
24.
25.

26.

27.
28.

Bishop, C.J.; Abubaker-Sharif, B.; Guiriba, T.; Tzeng, S.Y.; Green, J.J. Gene delivery polymer structure-function
relationships elucidated via principal component analysis. Chem. Commun. 2015, 51, 12134–12137. [CrossRef]
[PubMed]
Breitenkamp, R.B.; Emrick, T. Pentalysine-grafted ROMP polymers for DNA complexation and delivery.
Biomacromolecules 2008, 9, 2495–2500. [CrossRef] [PubMed]
Synatschke, C.V.; Schallon, A.; Jérôme, V.; Freitag, R.; Müller, A.H.E. Influence of polymer architecture and
molecular weight of poly(2-(dimethylamino)ethyl methacrylate) polycations on transfection efficiency and
cell viability in gene delivery. Biomacromolecules 2011, 12, 4247–4255. [CrossRef] [PubMed]
Raup, A.; Stahlschmidt, U.; Jérôme, V.; Synatschke, C.V.; Müller, A.H.E.; Freitag, R. Influence of polyplex
formation on the performance of star-shaped polycationic transfection agents for mammalian cells. Polymers
2016, 8, 224. [CrossRef]
Zhou, D.; Cutlar, L.; Gao, Y.; Wang, W.; O’Keeffe-Ahern, J.; McMahon, S.; Duarte, B.; Larcher, F.;
Rodriguez, B.J.; Greiser, U.; et al. The transition from linear to highly branched poly(β-amino ester)s:
Branching matters for gene delivery. Sci. Adv. 2016, 2, 1–15. [CrossRef] [PubMed]
Cutlar, L.; Zhou, D.; Gao, Y.; Zhao, T.; Greiser, U.; Wang, W.; Wang, W. Highly branched poly(β-amino esters):
Synthesis and application in gene delivery. Biomacromolecules 2015, 16, 2609–2617. [CrossRef] [PubMed]
Zhou, D.; Gao, Y.; O’Keeffe Ahern, J.; Xu, Q.; Huang, X.; Greiser, U.; Wang, W. Development of branched
poly(5-amino-1-pentanol-co-1,4-butanediol diacrylate) with high gene transfection potency across diverse
cell types. ACS Appl. Mater. Interfaces 2016, 8, 34218–34226. [CrossRef] [PubMed]
Zhou, D.; Gao, Y.; Aied, A.; Cutlar, L.; Igoucheva, O.; Newland, B.; Alexeeve, V.; Greiser, U.; Uitto, J.; Wang, W.
Highly branched poly(β-amino ester)s for skin gene therapy. J. Control. Release 2016, 244, 336–346. [CrossRef]
[PubMed]
Liu, S.; Gao, Y.; Zhou, D.; Greiser, U.; Guo, T.; Guo, R.; Wang, W. Biodegradable highly branched poly(β-amino
ester)s for targeted cancer cell gene transfection. ACS Biomater. Sci. Eng. 2016, 18, 440–451. [CrossRef]
Freyer, J.L.; Brucks, S.D.; Gobieski, G.S.; Russell, S.T.; Yozwiak, C.E.; Sun, M.; Chen, Z.; Jiang, Y.; Bandar, J.S.;
Stockwell, B.R.; et al. Clickable poly(ionic liquids): A materials platform for transfection. Angew. Chem.
Int. Ed. 2016, 55, 12382–12386. [CrossRef] [PubMed]
Gao, Y.; Chen, L.; Zhang, Z.; Gu, W.; Li, Y. Linear cationic click polymer for gene delivery: Synthesis,
biocompatibility, and in vitro transfection. Biomacromolecules 2010, 11, 3102–3111. [CrossRef] [PubMed]
Hemp, S.T.; Allen, M.H., Jr.; Green, M.D.; Long, T.E. Phosphonium-containing polyelectrolytes for nonviral
gene delivery. Biomacromolecules 2012, 13, 231–238. [CrossRef] [PubMed]
Jarz˛ebińska, A.; Pasewald, T.; Lambrecht, J.; Mykhaylyk, O.; Kümmerling, L.; Beck, P.; Hasenpusch, G.;
Rudolph, C.; Plank, C.; Dohmen, C. A Single methylene group in oligoalkylamine-based cationic polymers
and lipids promotes enhanced mRNA delivery. Angew. Chem. Int. Ed. 2016, 55, 9591–9595. [CrossRef]
[PubMed]
Jiang, Y.; Freyer, J.L.; Cotanda, P.; Brucks, S.D.; Killops, K.L.; Bandar, J.S.; Torsitano, C.; Balsara, N.P.;
Lambert, T.H.; Campos, L.M. The evolution of cyclopropenium ions into functional polyelectrolytes.
Nat. Commun. 2015, 6, 5950. [CrossRef] [PubMed]
Vaidyanathan, S.; Orr, B.G.; Banaszak Holl, M.M. Role of cell membrane-vector interactions in successful
gene delivery. Acc. Chem. Res. 2016, 49, 1486–1493. [CrossRef] [PubMed]
Santos, J.L.; Oliveira, H.; Pandita, D.; Rodrigues, J.; Pêgo, A.P.; Granja, P.L.; Tomás, H. Functionalization of
poly(amidoamine) dendrimers with hydrophobic chains for improved gene delivery in mesenchymal stem
cells. J. Control. Release 2010, 144, 55–64. [CrossRef] [PubMed]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

9

polymers
Article

Influences of Alkyl and Aryl Substituents on
Iminopyridine Fe(II)- and Co(II)-Catalyzed
Isoprene Polymerization
Lihua Guo 1 , Xinyu Jing 1 , Shuoyan Xiong 2 , Wenjing Liu 1 , Yanlan Liu 1 , Zhe Liu 1 and
Changle Chen 2, *
1

2

*

School of Chemistry and Chemical Engineering, Qufu Normal University, Qufu 273165, China;
18753770200@163.com (L.G.); m18463758798@163.com (X.J.); liuwenjingaaa@126.com (W.L.);
lyl373740@163.com (Y.L.); liuzheqd@163.com (Z.L.)
Key Laboratory of Soft Matter Chemistry, Chinese Academy of Sciences,
Department of Polymer Science and Engineering, University of Science and Technology of China,
Hefei 230026, China; xiong18@mail.ustc.edu.cn
Correspondence: changle@ustc.edu.cn; Tel.: +86-551-6360-1495

Academic Editor: Alexander Böker
Received: 28 September 2016; Accepted: 28 October 2016; Published: 3 November 2016

Abstract: A series of alkyl- and aryl-substituted iminopyridine Fe(II) complexes 1a–7a and Co(II)
complexes 2b, 3b, 5b, and 6b were synthesized. The activator effect, influence of temperature, and,
particularly, the alkyl and aryl substituents’ effect on catalytic activity, polymer molecular weight,
and regio-/stereoselectivity were investigated when these complexes were applied in isoprene
polymerization. All of the Fe(II) complexes afforded polyisoprene with high molecular weight
and moderate cis-1,4 selectivity. In contrast, the Co(II) complexes produced polymers with low
molecular weight and relatively high cis-1,4 selectivity. In the iminopyridine Fe(II) system, the alkyl
and aryl substituents’ effect exhibits significant variation on the isoprene polymerization. In the
iminopyridine Co(II) system, there is little influence observed on isoprene polymerization by alkyl
and aryl substituents.
Keywords: iminopyridine; Iron(II); Cobalt(II); isoprene polymerization; selectivity

1. Introduction
The polymerization of isoprene can afford polymers with various regio- and/or stereoregularities
such as isotactic or syndiotactic polyisoprene via 1,2 or 3,4 addition, and cis- or trans-1,4 polyisoprene
via 1,4 addition. The structures of polyisoprene strongly influence the properties of the resulting
material. For example, the properties of cis-1,4 polyisoprene is very similar to those of natural rubber [1],
while the properties of trans-1,4 polyisoprene is very close to those of gutta-percha [2]. The development
of highly efficient and highly regio- and stereoselective catalysts plays a key role in the field of
metal-catalyzed polymerization of conjugate dienes [3]. Titanium and rare-earth metal catalysts
can afford cis-1,4 and trans-1,4 polybutadienes and polyisoprenes with up to 98% selectivity [4–15].
In addition, some late transition-metal catalytic systems were successfully applied in olefins [16–34],
butadiene [4,35–45], and isoprene [36,37,46–50] polymerization. Late transition-metal catalysts have
lower Lewis acid characteristics and may possess high tolerance towards functional groups and polar
additives. Special attention was paid to low-cost and earth-abundant iron- and cobalt-based catalysts
with well-defined molecular structures that could be easily prepared.
Recently, Dai et al. [41] showed that an aryl-substituted iminopyridine Co(II) catalyst exhibited
high catalytic activity and cis-1,4-selectivity for 1,3-butadiene polymerization. Raynaud et al. [51]

Polymers 2016, 8, 389

10

www.mdpi.com/journal/polymers

Polymers 2016, 8, 389

reported that the combination of the iminopyridine Fe(II) complexes, alkylaluminum, and dealkylating
reagent [Ph3 C][B(C6 F5 )4 ] can polymerize isoprene with high stereoselectivity. The octyl-substituted
imines favor trans-1,4 insertion, whereas supermesityl-substituted imines favor cis-1,4 insertion.
The authors suggested that higher electron density at the iron center may increase the trans-1,4
selectivity. However, this accidental discovery and studies of only these two catalysts make it difficult
draw any rational conclusions.
Inspired by these works, we became very interested in the influence of iminopyridine
ligand substituents on the selectivity of isoprene polymerization. In this work, various alkyl- and
aryl-substituted iminopyridine Fe(II) and Co(II) complexes were synthesized and employed in isoprene
polymerization when activated using an alkylaluminum (methylaluminoxane (MAO) or AlEtCl2 )
(Scheme 1). The effects of the imine moiety on the catalytic activity, molecular weight, and, particularly,
the regio- and stereoselectivity were investigated.

Scheme 1. Alkyl- and aryl-substituted iminopyridine Fe(II) and Co(II) complexes for isoprene polymerization.

2. Experimental Section
2.1. General Information
All manipulations of air-and-moisture sensitive materials were performed under a dry nitrogen
atmosphere by using standard Schlenk techniques. Nitrogen was purified by passing through a
MnO oxygen-removal column and an activated 4 Å molecular sieve column. 1 H and 13 C NMR
spectra were recorded using CDCl3 as solvent on a Bruker Ascend™ 500 spectrometer (Bruker,
Karlsruhe, Germany) at room temperature unless otherwise stated. The chemical shifts of the 1 H and
13 C NMR spectra (Bruker, Karlsruhe, Germany) were referenced to tetramethylsilane (TMS). Coupling
constants are in units of hertz. Fourier-transform infrared (FTIR) spectrometry was performed on
a Thermo Scientific Nicolet iS5 (Thermo Fisher Scientific Corporation, Waltham, MA, USA) using
the conventional KBr wafer technique. Elemental analysis was performed by the Analytical Center
of the University of Science and Technology of China (Hefei, China). Mass spectra were recorded
on a P-SIMS-Gly of Bruker Daltonics Inc. (EI, Bruker Daltonics Inc., Billerica, MA, USA). X-ray
Diffraction data were collected at 298(2) K on a Bruker Smart CCD area detector (Bruker, Karlsruhe,
Germany) with graphite-monochromated MoKα radiation (λ = 0.71073 Å). Molecular weights and
molecular weight distributions were determined by gel permeation chromatography (GPC, Waters,
Milford, MA, USA) employing a series of two linear Styragel columns (HR2 and HR4) at an oven
temperature of 45 ◦ C. A Waters 1515 pump and Waters 2414 differential refractive index detector
(30 ◦ C) were used. The eluent was tetrahydrofuran (THF) at a flow rate of 1.0 mL·min−1 . A series
of low-polydispersity polystyrene standards was used for calibration. Hexane (Tianjin Fuyu Fine
Chemical Limited Company, Tianjin, China), toluene (Laiyang Fine Chemical Factory, Laiyang, China)
and THF (tetrahydrofuran, Tianjin Fuyu Fine Chemical Limited Company, Tianjin, China) were
refluxed over sodium benzophenone ketyl until the solution turned blue and then distilled before use.
CH2 Cl2 (Tianjin Fuyu Fine Chemical Limited Company, Tianjin, China) was refluxed over phosphorus
pentoxide for 8 h and distilled under a nitrogen atmosphere. Isoprene (Aladdin Industrial Corporation,
Shanghai, China) was dried over CaH2 prior to use in polymerization. Ligands L2 and L5–L7 were
prepared according to reported procedure [51–53]. Complexes 2a, 5a, 6a, and 6b were synthesized
according the reported method [41,51,54]. All other reagents were purchased from commercial sources
and used without purification.
11

Polymers 2016, 8, 389

2.2. General Procedure for the Synthesis of Ligands L1, L3, and L4
A solution of the corresponding amine (30 mmol) in methanol (30 mL) was added to
pyridine-2-carbaldehyde (30 mmol) and a drop of formic acid was subsequently added. The mixture
was stirred at room temperature overnight.
Cyclohexyl(pyridin-2-yl-methylene)amine (L1): the reaction mixture was concentrated under reduced
pressure. The residue was purified by distillation under vacuum to give the colorless oil. Yield: 5.37 g
(95.1%). 1 H NMR (500 MHz, CDCl3 ) δ 8.63 (d, J = 3.9 Hz, 1H), 8.40 (s, 1H, CH=N), 7.99 (d, J = 7.8 Hz,
1H), 7.70 (dd, J = 10.6, 4.1 Hz, 1H), 7.43–7.13 (m, 1H), 3.53–3.19 (m, 1H, N–CH), 2.05–1.52 (m, 7H),
1.53–1.04 (m, 3H). 13 C NMR (126 MHz, CDCl3 ) δ 157.97 (CH=N), 153.79, 147.76, 134.52, 122.80, 119.47,
67.83, 32.84, 24.30, 23.13. Anal. calcd. for C12 H16 N2 : C, 76.55; H, 8.57; N, 14.88; found: C, 76.13; H, 8.44;
N, 14.79.
Adamantyl(pyridin-2-yl-methylene)amine (L3): the reaction mixture was concentrated under reduced
pressure. The residue was purified by distillation under vacuum to give the light-yellow oil which
quickly changed to solid at room temperature. Yield: 6.76 g (93.7%). 1 H NMR (500 MHz, CDCl3 ) δ
8.63 (d, J = 3.6 Hz, 1H), 8.36 (s, 1H, CH=N), 8.01 (t, J = 16.1 Hz, 1H), 7.73 (t, J = 7.3 Hz, 1H), 7.35–7.27
(m, 1H), 2.23–2.13 (m, 3H, CH(CH2 )3 ), 1.83 (s, 6H, CH(CH2 )3 ), 1.79–1.65 (m, 6H, CH(CH2 )3 ). 13 C NMR
(126 MHz, CDCl3 ) δ 156.08 (CH=N), 155.58, 149.17, 136.38, 124.24, 120.82, 58.02, 42.89, 36.44, 29.44.
Anal. calcd. for C16 H20 N2 : C, 79.96; H, 8.39; N, 11.66; found: C, 79.81; H, 8.37; N, 11.72.
Triphenyl(pyridin-2-yl-methylene)amine (L4): the white solid precipitated from the solution and was
separated by filtration. The white solid was washed with methanol (3 × 5 mL). Yield: 8.18 g (78.3%).
1 H NMR (500 MHz, CDCl ) δ 8.61 (d, J = 3.9 Hz, 1H), 8.38 (d, J = 7.9 Hz, 1H, CH=N), 7.99 (s, 1H), 7.81
3
(t, J = 7.5 Hz, 1H), 7.34 (m, 16H, ). 13 C NMR (126 MHz, CDCl3 ) δ 160.95 (CH=N), 155.27, 149.28, 145.25,
136.61, 129.76, 127.85, 126.93, 124.84, 121.32. Anal. calcd. for C25 H20 N2 : C, 86.17; H, 5.79; N, 8.04;
found: C, 86.32; H, 5.63; N, 7.98.
2.3. General Procedure for the Synthesis of Iron Complexes
All complexes were prepared in a similar manner by the reaction of anhydrous FeCl2 with the
corresponding ligands in dichloromethane. A typical synthetic procedure used for complexes 1a, 3a, 4a,
and 7a is as follows. Ligand (1.0 mmol) and FeCl2 (1.0 mmol) were stirred in 10 mL of dichloromethane
overnight at room temperature. The precipitate was collected by filtration, washed with hexane
(10 mL × 2) and dried under vacuum to obtain orange, purple, or burgundy solid.
(Cyclohexyl Iminopyridine)FeCl2 (1a) (purple solid, 0.30 g, 95%): MALDI-TOF-MS (m/z): calcd. for
C12 H16 ClFeN2 : 279.0351, found: 278.9959 [M − Cl]+ . Anal. calcd. for C12 H16 Cl2 FeN2 : C, 45.75;
H, 5.12; N, 8.89; found: C, 46.20; H, 4.99; N, 9.12. IR/cm−1 :1563, ν(C=N).
(Adamantyl Iminopyridine)FeCl2 (3a) (orange solid, 0.35 g, 95%): MALDI-TOF-MS (m/z): calcd. for
C16 H20 ClFeN2 : 331.0664, found: 330.9991 [M − Cl]+ . Anal. calcd. for C16 H20 Cl2 FeN2 : C, 52.35;
H, 5.49; N, 7.63; found: C, 52.55; H, 5.33; N, 7.29. IR/cm−1 :1588, ν(C=N).
(Triphenyl Iminopyridine)FeCl2 (4a) (light-orange solid, 0.46 g, 96%): MALDI-TOF-MS (m/z): calcd. for
C25 H20 ClFeN2 : 439.0664, found: 439.0714 [M − Cl]+ . Anal. calcd. for C25 H20 Cl2 FeN2 : C, 63.19; H, 4.24;
N, 5.90; found: C, 62.88; H, 4.18; N, 5.67. IR/cm−1 :1588, ν(C=N).
(dibenzhydryl Iminopyridine)FeCl2 (7a) (burgundy solid, 0.61 g, 93%): MALDI-TOF-MS (m/z): calcd. for
C39 H32 ClFeN2 : 619.1603, found: 619.0020 [M − Cl]+ . Anal. calcd. for C39 H32 Cl2 FeN2 : C, 71.47; H, 4.92;
N, 4.27; found: C, 71.99; H, 4.87; N, 4.17. IR/cm−1 :1593, ν(C=N).
2.4. General Procedure for the Synthesis of Cobalt Complexes
All complexes were prepared in a similar manner by the reaction of anhydrous CoCl2 with the
corresponding ligands in tetrahydrofuran (THF). A typical synthetic procedure used for complexes
12

Polymers 2016, 8, 389

2b, 3b, and 5b is as follows. Ligand (1.0 mmol) and CoCl2 (1.0 mmol) were stirred in 10 mL of THF
overnight at room temperature. The precipitate was collected by filtration, washed with hexane
(10 mL × 2) and dried under vacuum to obtain a blue or green solid.
(octyl Iminopyridine)CoCl2 (2b) (blue solid, 0.30 g, 87%): MALDI-TOF-MS (m/z): calcd. for
C14 H22 ClCoN2 : 312.0804, found: 311.9917 [M − Cl]+ . Anal. calcd. for C14 H22 Cl2 CoN2 : C, 48.30; H,
6.37; N, 8.05; found: C, 49.41; H, 6.45; N, 7.91. IR/cm−1 :1597, ν(C=N).
(Adamantyl Iminopyridine)CoCl2 (3b) (blue solid, 0.33 g, 90%): MALDI-TOF-MS (m/z): calcd. for
C16 H20 ClCoN2 : 334.0647, found: 333.9984 [M − Cl]+ . Anal. calcd. for C16 H20 Cl2 CoN2 : C, 51.91;
H, 5.45; N, 7.57; found: C, 52.03; H, 5.23; N, 7.88. IR/cm−1 :1595, ν(C=N).
(supermesityl Iminopyridine)CoCl2 (5b) (green solid, 0.49 g, 91%): MALDI-TOF-MS (m/z): calcd. for
C30 H22 ClCoN2 : 504.0804, found: 503.9194 [M − Cl]+ . Anal. calcd. for C30 H22 Cl2 CoN2 : C, 66.68; H,
4.10; N, 5.18; found: C, 66.11; H, 3.96; N,5.31. IR/cm−1 :1597, ν(C=N).
2.5. General Procedure for Isoprene Polymerization
The polymerization of isoprene in toluene was carried out in a 50 mL Schlenk reactor. In a typical
experiment, the reactor was heated, dried in a vacuum, and recharged with nitrogen more than three
times before the required amount of an aluminum coactivator, toluene (7 mL), and isoprene (2 mL) were
added into the reactor. Then, 8.0 μmol of iron or cobalt complex in 1 mL CH2 Cl2 was injected to initiate
the polymerization at the desired temperature. After 2 h, the polymerization was quenched with a
diluted HCl solution of methanol (methanol/HCl = 50/1). The polymer was collected by filtration and
washed with ethanol several times and dried at room temperature for 24 h under vacuum.
2.6. Calculation of Microstructure Contents of Polyisoprenes
According to the calculated area of the characteristic signals at 4.66–4.72 and 5.12 ppm, the molar
content of 3,4 units and 1,4 units based on 1 H NMR spectra can be calculated by Equations (1) and (2)
where I (5.12 ppm) and I (4.66–4.72 ppm) represent signal areas at 5.12 and 4.66–4.72 ppm.

[%1, 4-units] =

[%3, 4-units] =

I (5.12 ppm)
I (5.12 ppm) +

I (5.12

I (4.66−4.72 ppm)
2

I (4.66−4.72 ppm)
2
I (4.66−4.72 ppm)
ppm) +
2

(1)

(2)

According to the calculated area of the characteristic signals at 16.2 and 23.8 ppm, the molar
content of cis-1,4 units and trans-1,4 units based on 13 C NMR spectra can be calculated by Equations (3)
and (4), where I (23.8 ppm) and I (16.2 ppm) represent signal areas at 23.8 and 16.2 ppm.

[%cis-1, 4-units] =

I (23.8 ppm)
I (23.8 ppm) + I (16.2 ppm)

[%trans-1, 4-units] =

I (16.2 ppm)
I (23.8 ppm) + I (16.2 ppm)

(3)

(4)

The microstructures of the polyisoprenes based on the FTIR spectra can be calculated according
to the equations in the literature [50].
A1375 = 24[cis-1, 4-units] L + 32.6[3, 4-units] L

(5)

A890 = 101[3, 4-units] L

(6)

[%cis-1, 4-units] = 100 ×

[cis-1, 4-units]
[cis-1, 4-units] + [3, 4-units]

13

(7)

Polymers 2016, 8, 389

[%3, 4-units] = 100 ×

[3, 4-units]
[cis-1, 4-units] + [3, 4-units]

(8)

where A1375 and A890 are the absorption intensity at 1375 and 890 cm−1 , expressed by the peak height,
[cis-1,4-units] represents the molar content of cis-1,4-units, [3,4-units] represents the molar content of
3,4-units, and L indicates the thickness of the sample.
3. Results and Discussion
3.1. Synthesis and Characterization of the Iron and Cobalt Complexes
The synthetic route for the iminopyridine complexes is shown in Scheme 2. The ligands
were prepared at high yields by acid-catalyzed condensation between corresponding anilines and
2-pyridinecarboxaldehyde in methanol and identified by NMR (See Supplementary Materials,
Figures S1–S6) and elemental analysis. The corresponding Fe(II) and Co(II) complexes (1a–7a, 2b,
3b, 5b, 6b) were prepared from the reaction of the ligands with 1 equiv of anhydrous FeCl2 or
CoCl2 in CH2 Cl2 and THF, respectively. These complexes were characterized by mass spectroscopy
(See Supplementary Materials, Figures S7–S13) and elemental analysis.

Scheme 2. Synthesis of the ligands and the Fe(II) and Co(II) complexes.

The structures of the complexes 1a–7a should be those drawn in Scheme 2. This is supported by the
elemental analysis, mass spectroscopy, and literature results on similar Fe(II) complexes [51]. Multiple
attempts to grow single crystals of complexes 1a–7a failed. However, during this process, single crystals
of complex 7a were obtained and analyzed by X-ray diffraction (Figure 1, See Supplementary Materials,
Tables S2 and S3). Complex 7a probably arises from the oxidation of 7a during the recrystallization
process. This unusual complex of 7a is interesting, and can prove the connectivity of the iminopyridine
ligand to the metal center. The X-ray crystal structure analysis of 7a shows a distorted trigonal
bipyramidal coordination geometry around the Fe(II) center. The steric environment of the ligand
and the blocking of the axial position of the metal center from the dibenzhydryl moiety can be
clearly observed from this molecular structure. Single crystals of pure complex 2b could be obtained
and the X-ray structure is shown in Figure 2. In a solid state, the cobalt center adopts a distorted
tetrahedral coordination geometry with N1–Co–N2 angle of 81.61◦ and Cl1–Co–Cl2 angle of 112.06◦
(See Supplementary Materials, Tables S2 and S4). Complex 2b shows shorter Co–N bond distance
(2.040 and 2.046 Å) than aryl-substituted Co(II) complexes reported in literature [41] (2.044~2.181 Å),
which may be attributed to the strong electron-donating effect of the octyl substituents.

14

Polymers 2016, 8, 389

Figure 1. Molecular structure of complex 7a . (Thermal ellipsoids are shown at the 50% probability
level.) Hydrogen atoms have been omitted for clarity.

Figure 2. Molecular structure of complex 2b. (Thermal ellipsoids are shown at the 50% probability
level.) Hydrogen atoms have been omitted for clarity.

3.2. Isoprene Polymerization Studies
3.2.1. Polymerization of Isoprene with Iron Catalysts
The isoprene polymerization was evaluated using various common alkylaluminum reagents as
cocatalysts. Triisobutylaluminum (TIBA) or AlEt2 Cl cocatalysts were not effective at all. Both AlEtCl2
and MAO were able to activate 2a for isoprene polymerization (Table 1, entries 1 and 2). However,
the 2a/MAO system can generate high molecular polyisoprenes. Therefore, MAO was chosen as the
activator in the iminopyridine Fe(II) system (See Supplementary Materials, Table S1).
Table 1. Isoprene polymerization results with Fe(II) catalyst a .

Entry

Complex

T
(◦ C)

Yield
(%)

Activity c

Mn d
(×10−4 )

PDI d

1b
2
3
4
5
6
7
8
9
10

2a
2a
2a
1a
3a
4a
5a
5a
6a
7a

25
25
−25
25
25
25
−25
25
25
25

83.4
83.1
66.3
64.1
58.2
61.3
81.0
98.1
83.2
85.7

7.1
7.1
5.6
5.4
4.9
5.2
6.9
8.3
7.1
7.3

0.18
6.1
7.9
6.0
7.0
6.1
15.4
10.3
18.0
18.2

4.70
1.57
2.45
2.11
1.82
2.08
2.13
2.05
1.75
1.61

Microstructure c (%) e
cis-1,4
77.5
77.0
77.1
76.8
78.2
63.9
62.7
69.9
71.4

trans-1,4 cis/trans
8.1
8.7
8.9
8.2
7.6
3.0
2.8
4.5
4.8

91:9
90:10
90:10
90:10
91:9
96:4
96:4
94:6
94:6

3,4
14.4
14.3
14.0
15.0
14.2
33.1
34.5
25.6
23.8

a Polymerization conditions: 8.0 μmol of Fe(II) complex; MAO/Fe = 500; 7 mL toluene and 1 mL CH Cl ;
2 2
isoprene = 2 mL; time = 2 h; b activator = AlEtCl2 , Al/Fe = 150; c 104 g of polyisorene (mol of Fe)−1 ·h−1 ;
d determined by gel permeation chromatography (GPC); e determined by 1 H NMR and 13 C NMR.

15

Polymers 2016, 8, 389

The alkyl and aryl moiety significantly influenced the catalytic performances of the complexes.
The aryl-substituted complexes 5a–7a produced polymers at higher yields (83.2%–98.1%) than the
alkyl-substituted complexes 1a–4a (58.2%–83.1%). The aryl moiety is electronically more withdrawing
than the alkyl moiety, which can reduce the electron density on the metal center, leading to better
monomer coordination and faster chain propagation. This is supported by the fact that complex 5a
bears the strongest electron-withdrawing substituent and displays the highest yield. In addition,
the molecular weight of polyisoprenes obtained by aryl-substituted complexes 5a–7a is higher than
alkyl-substituted complexes 1a–4a (10.3 × 104 ~18.2 × 104 vs. 6.0 × 104 ~7.9 × 104 ). Probably, the
steric environment of the aryl moiety retards chain transfer reaction more effectively than the alkyl
moiety (See Supplementary Materials, Figures S14–S20). This is supported by the fact that complex 7a
bears a sterically bulky dibenzhydryl-derived ligand framework and generates polyisoprene with the
highest molecular weight (18.2 × 104 ). The temperature influence on the catalytic performance was
also investigated. Polymerization of isoprene at −25 ◦ C showed lower yields (2a: 66.3% vs. 83.1%; 5a:
81.0% vs. 98.1%) and afforded the polymer with higher molecular weight (2a: 7.9 × 104 vs. 6.1 × 104 ;
5a: 15.4 × 104 vs. 10.3 × 104 ) than those at 25 ◦ C.
The microstructures of the resulting polyisoprenes were analyzed via 1 H NMR and 13 C NMR
(See Supplementary Materials, Figures S23–S26) [51]. The representative 1 H NMR spectra of the
polyisoprenes obtained by the Fe(II) catalysts are shown in Figure 3. The 1,2-unit was not observed.
The polyisoprene obtained by aryl-substituted complex 5a contains 34.5% 3,4-units (Table 1, entry 10),
which was much higher than that of the aryl-substituted complex 3a (15.0%, entry 5). Similar trends
were observed for other alkyl-substituted complexes (14.0%~15.0%) and aryl-substituted complexes
(greater than 23.8%). Interestingly, the R group in the alkyl-substituted complexes only slightly
influenced 3,4-selectivity from 14.0% to 15.0%. However, the selectivity of 3,4-units was increased from
23.8% to 34.5% when the steric hindrance of the aryl-substituted complexes was decreased. The high
3,4 units of polyisoprene can increase the toughness of the synthetic rubber and show outstanding wet
skid resistance and low heat build-up when applied as car tires [55], thus representing a big advantage
of this catalyst system. Additionally, the alkyl-substituted complexes 1a–4a produced polymers with
higher cis-1,4 content (77.1%~78.2%) than the aryl-substituted complexes 5a–7a (62.7%–71.4%). At the
same time, the alkyl-substituted complexes produced polyisoprene with 7.6%~8.9% trans-1,4 content,
which was ca. twice as much as that by aryl-substituted complexes (2.8%~4.8%). However, polymers
generated from the aryl-substituted complexes had the higher cis-1,4/trans-1,4 ratio (e.g., 5a: 96:4,
entry 8) than the alkyl-substituted complexes (e.g., 3a: 90:10, entry 5). These results indicated that the
electron-donating alkyl-substituted complexes tend to polymerize isoprene with trans-1,4-selectivity
when 1,4-addition occurred. It was also observed that the steric hindrance of both kinds of complexes
almost have minimum influence on cis-1,4/trans-1,4 stereoselectivity with 1a–4a (ca. 90:10) and 5a–7a
(ca. 95:5).
Previously, Raynaud et al. used alkylaluminum/[Ph3 C][B(C6 F5 )4 ] cocatalysts to activate the Fe(II)
complexes, and the 1,4-trans/1,4-cis selectivity was affected by the alkyl/aryl substituents and the
alkylaluminum agents (TIBA and AlEt3 ) [51]. In our system, the alkylaluminum/[Ph3 C][B(C6 F5 )4 ]
cocatalysts led to highly unreproducible results, which may originate from the high sensitivity of these
Fe(II) complexes. As a result, MAO was chosen as the cocatalyst. These Fe(II) complexes showed high
activities and high polymer molecular weight when activated using MAO as cocatalyst. Furthermore, in
our Fe(II)/MAO system, the aryl-substituted iminopyridine iron complexes also favor 3,4-insertion and
give rise to higher amounts of 3,4-units than the alkyl-substituted iminopyridine iron complexes, which
is similar to the Fe(II)/alkylaluminum/[Ph3 C][B(C6 F5 )4 ] system. However, there are some notable
differences between these two systems. In our Fe(II)/MAO system, the ratio between 1,4-cis/trans
units was not affected by the aryl or alkyl substituents. Although this difference is not fully understood,
it is clear that the cocatalysts may play an important role in determining the stereoselectivity.

16

Polymers 2016, 8, 389

Figure 3. 1 H NMR spectra of polyisoprenes obtained by Fe(II) catalyst.

3.2.2. Polymerization of Isoprene with Co(II) Catalysts
The polymerization results using Co(II) complexes 2b, 3b, 5b, and 6b are summarized in Table 2.
Four cocatalysts (TIBA, AlEt2 Cl, AlEtCl2 , and MAO) were used in attempts to generate the active
catalysts. Only cocatalyst AlEtCl2 was able to activate Co(II) complex 2b for isoprene polymerization.
Although the yields (greater than 76.9%) of polyisoprene generated from Co(II) complexes are similar
with those of Fe(II) complexes, there are some apparent differences between the Co(II) and the Fe(II)
systems. In sharp contrast to the Fe(II) complexes, the polymers produced by Co(II) complexes were
white powder with molecular weights below 2000 and broad molecular distribution of above 4.76
(See Supplementary Materials, Figures S21 and S22). Moreover, complexes 5b and 6b containing
electron-withdrawing aryl substituents afforded polymers with higher molecular weights (5b: 1700,
6b: 1800) at higher yields (5b: 97.3%, 6b: 94.9%) than those by complexes 2b and 3b containing
electron-donating alkyl substituents (2b: 1400 and 78.2%, 3b: 1500 and 76.9%). This is similar with the
trend observed in the Fe(II) systems.
Table 2. Isoprene polymerization results with Co(II) catalyst a .

Entry

Complex

Yield (%)

Activity b

M n c (×10−3 )

PDI c

1
2
3
4

2b
3b
5b
6b

78.2
76.9
97.3
94.9

6.6
6.5
8.3
8.1

1.4
1.5
1.7
1.8

7.97
4.76
8.05
9.38

Microstructure d (%)
cis-1,4

3,4

91.1
90.8
88.1
89.7

8.9
9.2
11.9
10.3

a Polymerization conditions: 8.0 μmol of Co(II) complex; activator = AlEtCl , Al/Co = 150; 7 mL toluene and
2
1 mL CH2 Cl2 ; isoprene = 2 mL; time = 2 h; b 104 g of polyisorene (mol of Co)−1 ·h−1 ; c determined by GPC;
d determined by Fourier-transform infrared (FTIR) spectroscopy.

The 1 H NMR and 13 C NMR spectra of polyisoprene obtained by Co(II) complexes have the broad
peaks and low resolution because of the low molecular weight of the polymers (See Supplementary
Materials, Figures S27–S30). It was difficult to assign the peaks of these polymers in the 1 H NMR and
13 C NMR spectra, so FTIR measurements were carried out to determine and analyze the microstructures
of the polyisoprenes (See Supplementary Materials, Figures S31–S34). The absorption bands at 1375
and 890 cm−1 correspond to the cis-1,4 and the 3,4-units [50]. The typical bands of trans-1,4 units
are at 845, 1152, 1325, and 1385 cm−1 and the band of 1,2-units is at 911 cm−1 [50]. As shown in
Figure 4, no bands were observed for the trans-1,4 unit or 1,2-unit in the spectrum. Based on the

17

Polymers 2016, 8, 389

equations shown in the experimental section, the polymer generated with the Co(II)/AlEtCl2 system
is composed of predominantly cis-1,4 units (ca. 90%) along with a small amount of 3,4-units (ca. 10%).
Interestingly, the Co(II) system produced polymers with higher cis-1,4 content (ca. 90%) than the Fe(II)
system (65%~85%). The stereoregularity of the polyisoprenes was only slightly influenced by the
ligand environment.

Figure 4. FTIR spectra of polyisoprenes obtained by Co(II) catalysts (entry 6).

4. Conclusions
In conclusion, a series of iminopyridine Fe(II) and Co(II) complexes bearing various alkyl and
aryl substituents was prepared. The aim is to systematically investigate the influence of alkyl and
aryl substituents on the isoprene polymerization. Activated by MAO, the Fe(II) complexes exhibited
moderate cis-1,4 selectivity, generating high molecular weight polyisoprenes. The Fe(II) catalyzed
polymerization of isoprene was relatively sensitive to alkyl and aryl substituents. High 3,4-units
(up to 34.5%) and high molecular weight (10.3 × 104 ~18.2 × 104 ) polyisoprenes can be obtained using
aryl-substituted Fe(II) complexes. Meanwhile, the Co(II)/AlEt2 Cl system exhibited relatively high
cis-1,4-selectivity, affording low molecular weight polyisoprenes. The alkyl and aryl substituents
in Co(II) complexes did not significantly influence the selectivity and molecular weight of the
resulting polymers.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/8/11/389/s1.
Optimization of MAO/Fe Ratio with 3a (Table S1), NMR spectra of the ligands L1, L3 and L4 (Figures S1–S6),
MALDI-TOF-MS of complexes (Figures S7–S13), crystal data of complex 7a (CCDC number: 1503575) and 2b
(CCDC number: 1503576) (Tables S2–S4), GPC curves of polyisoprene samples (Figures S14–S22), NMR spectra of
the representive polyisoprenes (Figures S23–S30) and FTIR spectra of representive polyisoprenes (Figures S31–S34).
Acknowledgments: This work was supported by National Natural Science Foundation of China (NSFC,
21304054, 21374108 and 51522306), Foundation of Qufu Normal University (xkJ201603), National College Students
Innovation Project (201610446029), Anhui Provincial Natural Science Foundation (1408085QB28, 1608085MB29)
and the Recruitment Program of Global Experts.
Author Contributions: Lihua Guo and Changle Chen conceived and designed the experiments; Lihua Guo,
Xinyu Jing, Shuoyan Xiong, Wenjing Liu, Yanlan Liu and Zhe Liu performed the experiments; Lihua Guo and
Changle Chen analyzed the data and wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.

Horne, S.E., Jr.; Kiehl, J.P.; Shipman, J.J.; Folt, V.L.; Gibbs, C.F. Ameripol SN-A cis-1,4-polyisoprene.
Ind. Eng. Chem. 1956, 48, 784–791. [CrossRef]
Natta, G. Progress in five years of research in stereospecific polymerization. SPE J. 1959, 53, 373–382.

18

Polymers 2016, 8, 389

3.
4.

5.
6.

7.

8.

9.

10.
11.
12.

13.

14.

15.

16.
17.
18.
19.
20.

21.

22.
23.

Proto, A.; Capacchione, C. Stereoselective Polymerization with Single-Site Catalysts; Baugh, L.S., Canich, J.A.M.,
Eds.; CRC Press: Boca Raton, FL, USA, 2008.
Ricci, G.; Sommazzi, A.; Masi, F.; Ricci, M.; Boglia, A.; Leone, G. Well-defined transition metal complexes
with phosphorus and nitrogen ligands for 1,3-dienes polymerization. Coord. Chem. Rev. 2010, 254, 661–676.
[CrossRef]
Osakada, K.; Takeuchi, D. Coordination polymerization of dienes, allenes, and methylenecycloalkanes.
Adv. Polym. Sci. 2004, 171, 137–194.
Zhang, L.X.; Suzuki, T.; Luo, Y.; Nishiura, M.; Hou, Z.M. Cationic alkyl rare-earth metal complexes bearing
an ancillary Bis(phosphinophenyl)amido ligand: A catalytic system for living cis-1,4-polymerization and
copolymerization of isoprene and butadiene. Angew. Chem. Int. Ed. 2007, 46, 1909–1913. [CrossRef] [PubMed]
Gao, W.; Cui, D.M. Highly cis-1,4 selective polymerization of dienes with homogeneous Ziegler-Natta
catalysts based on NCN-pincer rare earth metal dichloride precursors. J. Am. Chem. Soc. 2008, 130, 4984–4991.
[CrossRef] [PubMed]
Zhang, L.X.; Nishiura, M.; Yuki, M.; Luo, Y.; Hou, Z.M. Isoprene polymerization with yttrium amidinate
catalysts: Switchingthe regio-and stereoselectivity by addition of AlMe3 . Angew. Chem. Int. Ed. 2008, 47,
2642–2645. [CrossRef] [PubMed]
Li, D.F.; Li, S.H.; Cui, D.M.; Zhang, X.Q. β-diketiminato rare-earth metal complexes. Structures, catalysis,
and active species for highly cis-1,4-selective polymerization of isoprene. Organometallics 2010, 29, 2186–2193.
[CrossRef]
Lv, K.; Cui, D.M. CCC-pincer bis(carbene) lanthanide dibromides. Catalysis on highlycis-1,4-selective
polymerization of isoprene and active species. Organometallics 2010, 29, 2987–2993. [CrossRef]
Nishiura, M.; Hou, Z.M. Novel polymerization catalysts and hydride clusters from rare-earth metal dialkyls.
Nat. Chem. 2010, 2, 257–268. [CrossRef] [PubMed]
Milione, S.; Cuomo, C.; Capacchione, C.; Zannoni, C.; Grassi, A.; Proto, A. Stereoselective polymerization
of conjugated dienes and styrene-butadiene copolymerization promoted by octahedral titanium catalyst.
Macromolecules 2007, 40, 5638–5643. [CrossRef]
Buonerba, A.; Fienga, M.; Milione, S.; Cuomo, C.; Grassi, A.; Proto, A.; Capacchione, C. Binary
copolymerization of p-methylstyrene with butadiene and isoprene catalyzed by titanium compounds
showing different stereoselectivity. Macromolecules 2013, 46, 8449–8457. [CrossRef]
Proto, A.; Avagliano, A.; Saviello, D.; Ricciardi, R.; Capacchione, C. Living, isoselective polymerization of
styrene and formation of stereoregular block copolymers via sequential monomer addition. Macromolecules
2010, 43, 5919–5921. [CrossRef]
Capacchione, C.; Saviello, D.; Ricciardi, R.; Proto, A. Living, isoselective polymerization of 4-methyl-1,3pentadiene and styrenic monomers and synthesis of highly stereoregular block copolymers via sequential
monomer addition. Macromolecules 2011, 44, 7940–7947. [CrossRef]
Johnson, L.K.; Killian, C.M.; Brookhart, M. New Pd(II)- and Ni(II)-based catalysts for polymerization of
ethylene and α-olefins. J. Am. Chem. Soc. 1995, 117, 6414–6415. [CrossRef]
Nakamura, A.; Ito, S.; Nozaki, K. Coordination-insertion copolymerization of fundamental polar monomers.
Chem. Rev. 2009, 109, 5215–5244. [CrossRef] [PubMed]
Chen, E.Y.X. Coordination polymerization of polar vinyl monomers by single-site metal catalysts. Chem. Rev.
2009, 109, 5157–5214. [CrossRef] [PubMed]
Guan, Z.; Cotts, P.M.; McCord, E.F.; McLain, S.J. Chain Walking: A new strategy to control polymer topology.
Science 1999, 283, 2059–2061. [CrossRef] [PubMed]
Vaidya, T.; Klimovica, K.; LaPointe, A.M.; Keresztes, I.; Lobkovsky, E.B.; Daugulis, O.; Coates, G.W. Secondary
alkene insertion and precision chain-walking: A new route to semicrystalline “polyethylene” from α-olefins
by combining two rare catalytic events. J. Am. Chem. Soc. 2014, 136, 7213–7216. [CrossRef] [PubMed]
Takano, S.; Takeuchi, D.; Osakada, K.; Akamatsu, N.; Shishido, A. Dipalladium catalyst for olefin
polymerization: Introduction of acrylate units into the main chain of branched polyethylene. Angew. Chem.
Int. Ed. 2014, 53, 9246–9250. [CrossRef] [PubMed]
Guo, L.H.; Chen, C.L. (α-Diimine) palladium catalyzed ethylene polymerization and copolymerization with
polar comonomers. Sci. China Chem. 2015, 58, 1663–1673. [CrossRef]
Guo, L.H.; Dai, S.Y.; Sui, X.L.; Chen, C.L. Palladium and nickel catalyzed chain walking olefin polymerization
and copolymerization. ACS Catal. 2016, 6, 428–441. [CrossRef]
19

Polymers 2016, 8, 389

24.

25.
26.
27.
28.
29.
30.
31.

32.

33.
34.
35.
36.
37.

38.
39.

40.
41.

42.

43.

44.

Guo, L.; Gao, H.; Guan, Q.; Hu, H.; Deng, J.; Liu, J.; Liu, F.; Wu, Q. Substituent effects of the backbone
in α-diimine palladium catalysts on homo- and copolymerization of ethylene with methyl acrylate.
Organometallics 2012, 31, 6054–6062. [CrossRef]
Guo, L.H.; Sui, X.L.; Dai, S.Y.; Chen, C.L. Ligand electronic effects on α-diimine nickel(II) catalyzed ethylene
polymerization. Polymers 2016, 8, 37. [CrossRef]
Dai, S.Y.; Chen, C.L. Direct synthesis of functionalized high-molecular-weight polyethylene by copolymerization
of ethylene with polar monomers. Angew. Chem. Int. Ed. 2016, 55, 13281–13285. [CrossRef] [PubMed]
Dai, S.Y.; Sui, X.L.; Chen, C.L. Synthesis of high molecular weight polyethylene using iminopyridyl nickel
catalysts. Chem. Commun. 2016, 52, 9113–9116. [CrossRef] [PubMed]
Wang, R.K.; Zhao, M.H.; Chen, C.L. Influence of ligand second coordination sphere effects on the olefin
(co)polymerization properties of α-diimine Pd(II) catalysts. Polym. Chem. 2016, 7, 3933–3938. [CrossRef]
Hu, X.H.; Dai, S.Y.; Chen, C.L. Ethylene polymerization by salicylaldimine nickel(II) complexes containing
dibenzhydryl moiety. Dalton Trans. 2016, 45, 1496–1503. [CrossRef] [PubMed]
Wang, R.K.; Sui, X.L.; Pang, W.M.; Chen, C.L. Ethylene polymerization by xanthene bridged dinuclear
α-diimine Ni(II) complexes. ChemCatChem 2016, 8, 434–440. [CrossRef]
Chen, M.; Yang, B.P.; Chen, C.L. Redox-controlled olefin (co)polymerization catalyzed by ferrocene bridged
phosphine-sulfonate palladium complexes. Angew. Chem. Int. Ed. 2015, 54, 15520–15524. [CrossRef]
[PubMed]
Dai, S.Y.; Sui, X.L.; Chen, C.L. Highly robust Pd(II) α–diimine catalysts for slow-chain-walking polymerization
of ethylene and copolymerization with methyl acrylate. Angew. Chem. Int. Ed. 2015, 54, 9948–9953. [CrossRef]
[PubMed]
Sui, X.L.; Dai, S.Y.; Chen, C.L. Ethylene polymerization and copolymerization with polar monomers by
cationic phosphine phosphonic amide palladium complexes. ACS Catal. 2015, 5, 5932–5937. [CrossRef]
Chen, M.; Zou, W.P.; Cai, Z.G.; Chen, C.L. Norbornene homopolymerization and copolymerization with
ethylene by phosphine-sulfonate nickel catalysts. Polym. Chem. 2015, 6, 2669–2676. [CrossRef]
Ricci, G.; Battistella, M.; Porri, L. Chemoselectivity and stereospecificity of chromium(II) catalysts for
1,3-diene polymerization. Macromolecules 2001, 34, 5766–5769. [CrossRef]
Bazzini, C.; Giarrusso, A.; Porri, L. Diethylbis(2,2 -bipyridine)iron/MAO. A very active and stereospecific
catalyst for 1,3-diene polymerization. Macromol. Rapid Commun. 2002, 23, 922–927. [CrossRef]
Ricci, G.; Morganti, D.; Sommazzi, A.; Santi, R.; Masi, F. Polymerization of 1,3-dienes with iron complexes
based catalysts influence of the ligand on catalyst activity and stereospecificity. J. Mol. Catal. A Chem. 2003,
204–205, 287–293. [CrossRef]
Bazzini, C.; Giarrusso, A.; Porri, L.; Pirozzi, B.; Napolitano, R. Synthesis and characterization of syndiotactic
3,4-polyisoprene prepared with diethylbis(2,2 -bipyridine)iron-MAO. Polymer 2004, 45, 2871–2875. [CrossRef]
Ricci, G.; Forni, A.; Boglia, A.; Sonzogni, M. New chromium(II) bidentate phosphine complexes: Synthesis,
characterization, and behavior in the polymerization of 1,3-butadiene. Organometallics 2004, 23, 3727–3732.
[CrossRef]
Ricci, G.; Forni, A.; Boglia, A.; Motta, T. Synthesis, structure, and butadiene polymerization behavior of
alkylphosphine cobalt(II) complexes. J. Mol. Catal. A Chem. 2005, 226, 235–241. [CrossRef]
Dai, Q.Q.; Jia, X.Y.; Yang, F.; Bai, C.X.; Hu, Y.M.; Zhang, X.Q. Iminopyridine-based cobalt(II) and nikel(II)
complexes: Synthesis, characterization, and their catalytic behaviors for 1,3-butadiene polymerization.
Polymers 2016, 8, 12. [CrossRef]
Gong, D.R.; Dong, W.M.; Hu, Y.M.; Bi, J.F.; Zhang, X.Q.; Jiang, L.S. Syndiotactically enriched 1,2-selective
polymerization of 1,3-butadiene initiated by iron catalysts based on a new class of donors. Polymer 2009, 50,
5980–5986. [CrossRef]
Cámpora, J.; Tabla, L.O.; Palma, P.; Álvarez, E.; Lahoz, F.; Mereiter, K. Synthesis and catalytic activity of
cationic allyl complexes of nickel stabilized by a single n-heterocyclic carbene ligand. Organometallcs 2006,
25, 3314–3316. [CrossRef]
Ricci, G.; Forni, A.; Boglia, A.; Motta, T.; Zannoni, G.; Canetti, M.; Bertini, F. Synthesis and X-ray structure
of CoCl2 (Pi PrPh2 )2 . A new highly active and stereospecific catalyst for 1,2 polymerization of conjugated
dienes when used in association with MAO. Macromolecules 2005, 38, 1064–1070. [CrossRef]

20

Polymers 2016, 8, 389

45.

46.
47.

48.

49.

50.

51.
52.
53.

54.

55.

Ricci, G.; Boglia, A.; Motta, T. Synthesis of new Cr(II) complexes with bidentate phosphine ligands and their
behavior in the polymerization of butadiene influence of the phosphine bite angle on catalyst activity and
stereoselectivity. J. Mol. Catal. A Chem. 2007, 267, 102–107. [CrossRef]
Ricci, G.; Leone, G.; Boglia, A.; Boccia, A.C.; Zetta, L. cis-1,4-alt-3,4 polyisoprene: Synthesis and
characterization. Macromolecules 2009, 42, 9263–9267. [CrossRef]
Ricci, G.; Leone, G.; Boglia, A.; Bertini, F.; Boccia, A.C.; Zetta, L. Synthesis and characterization of
isotactic 1,2-poly(E-3-methyl-1,3-pentadiene). Some remarks about the influence of monomer structure on
polymerization stereoselectivity. Macromolecules 2009, 42, 3048–3056. [CrossRef]
Ricci, G.; Motta, T.; Boglia, A.; Alberti, E.; Zetta, L.; Bertini, F.; Arosio, P.; Famulari, A.; Meille, S.V.
Synthesis, characterization, and crystalline structure of syndiotactic 1,2-polypentadiene: The trans polymer.
Macromolecules 2005, 38, 8345–8352. [CrossRef]
Ricci, G.; Boglia, A.; Motta, T.; Bertini, F.; Boccia, A.C.; Zetta, L.; Alberti, E.; Famulari, A.; Arosio, P.; Meille, S.V.
Synthesis and structural characterization of syndiotactic trans-1,2 and cis-1,2 polyhexadienes. J. Polym. Sci. A
Polym. Chem. 2007, 45, 5339–5353. [CrossRef]
He, A.H.; Wang, G.; Zhao, W.Z.; Jiang, X.B.; Yao, W.; Sun, W.H. High cis-1,4 polyisoprene or cis-1,4/3,4 binary
polyisoprene synthesized using 2-(benzimidazolyl)-6-(1-(arylimino)ethyl)pyridine cobalt(II) dichlorides.
Polym. Int. 2013, 62, 1758–1766. [CrossRef]
Raynaud, J.; Wu, J.Y.; Ritter, T. Iron-catalyzed polymerization of isoprene and other 1,3-dienes. Angew. Chem.
Int. Ed. 2012, 51, 11805–11808. [CrossRef] [PubMed]
Wu, J.Y.; Stanzl, B.N.; Ritter, T. A Strategy for the synthesis of well-defined iron catalysts and application to
regioselective diene hydrosilylation. J. Am. Chem. Soc. 2010, 132, 13214–13216. [CrossRef] [PubMed]
Sun, W.H.; Song, S.J.; Li, B.X.; Redshaw, C.; Hao, X.; Li, Y.S.; Wang, F.S. Ethylene polymerization by
2-iminopyridylnickel halide complexes: Synthesis, characterization and catalytic influence of the benzhydryl
group. Dalton Trans. 2012, 41, 11999–12010. [CrossRef] [PubMed]
Gibson, V.C.; O Reilly, R.K.; Wass, D.F.; White, A.J.P.; Williams, D.J. Iron complexes bearing iminopyridine
and aminopyridine ligands as catalysts for atom transfer radical polymerization. Dalton Trans. 2003.
[CrossRef]
Wang, B.L.; Cui, D.M.; Lv, K. Highly 3,4-selective living polymerization of isoprene with rare earth metal
fluorenyl N-heterocyclic carbene precursors. Macromolecules 2008, 41, 1983–1988. [CrossRef]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

21

polymers
Article

Multiresponsive Behavior of Functional
Poly(p-phenylene vinylene)s in Water
Kanykei Ryskulova 1 , Anupama Rao Gulur Srinivas 2,3 , Thomas Kerr-Phillips 2,3 , Hui Peng 4,5 ,
David Barker 2 , Jadranka Travas-Sejdic 2,3 and Richard Hoogenboom 1, *
1
2

3
4
5

*

Supramolecular Chemistry Group, Department of Organic and Macromolecular Chemistry,
Faculty of Science, Ghent University, Krijgslaan 281 S4, Ghent B-9000, Belgium; kanikey@gmail.com
Polymer Electronics Research Center, School of Chemical Sciences, The University of Auckland,
Private Bag 92019, Auckland, New Zealand; gsanu85@gmail.com (A.R.G.S.);
tker016@aucklanduni.ac.nz (T.K.-P.); d.barker@auckland.ac.nz (D.B.);
j.travas-sejdic@auckland.ac.nz (J.T.-S.)
MacDiarmid Institute for Advanced Materials and Nanotechnology, Victoria University of Wellington,
P.O. Box 600, Wellington, New Zealand
Key Laboratory of Polarized Materials and Devices, Ministry of Education, East China Normal University,
Shanghai 200062, China; h.peng@auckland.ac.nz
Collaborative Innovation Center of Extreme Optics, Shanxi University, Taiyuan 030006, Shanxi, China
Correspondence: Richard.Hoogenboom@ugent.be; Tel.: +32-926-449-98; Fax: +32-926-444-82

Academic Editors: Alexander Böker and Frank Wiesbrock
Received: 18 August 2016; Accepted: 10 October 2016; Published: 18 October 2016

Abstract: The multiresponsive behavior of functionalized water-soluble conjugated polymers (CPs) is
presented with potential applications for sensors. In this study, we investigated the aqueous solubility
behavior of water-soluble CPs with high photoluminescence and with a particular focus on their pH
and temperature responsiveness. For this purpose, two poly(phenylene vinylene)s (PPVs)—namely
2,5-substituted PPVs bearing both carboxylic acid and methoxyoligoethylene glycol units—were
investigated, with different amount of carboxylic acid units. Changes in the pH and temperature of
polymer solutions led to a response in the fluorescence intensity in a pH range from 3 to 10 and for
temperatures ranging from 10 to 85 ◦ C. Additionally, it is demonstrated that the polymer with the
largest number of carboxylic acid groups displays upper critical solution temperature (UCST)-like
thermoresponsive behavior in the presence of a divalent ion like Ca2+ . The sensing capability of these
water-soluble PPVs could be utilized to design smart materials with multiresponsive behavior in
biomedicine and soft materials.
Keywords: polymeric sensor; water-soluble conjugated polymer; fluorescent sensor; reversible
calcium binding; thermoresponsive polymer; upper critical solution temperature

1. Introduction
Polymers that respond to external stimuli such as pH, temperature, ionic strength, light,
and concentration are extensively studied for a broad range of applications, including biomedical
applications like drug delivery [1,2], cell imaging [3–5], as well as biological [6] and optical sensing [7–9].
Polymer-based systems with responsive properties can straightforwardly be designed thanks to recent
advances in polymer synthesis as well as better understanding of physicochemical properties at the
molecular level [10–12]. Such responsive functions arise from the chemical or topological structure
of the polymer and lead to the design of stimuli-responsive smart materials with tailored functions.
In contrast to their small molecular analogs, macromolecular or polymeric sensors exhibit numerous
advantageous features, such as adjustable water solubility, higher detection sensitivity, straightforward
integration into sensing devices, better stability, biocompatibility, and sharpened responses [13].
Polymers 2016, 8, 365

22

www.mdpi.com/journal/polymers

Polymers 2016, 8, 365

The important role of temperature and pH in many biological processes has led to increased
interest in constructing smart materials that respond to these parameters. To develop polymeric sensors
for temperature and/or pH, normally two routes can be followed. Either a water-soluble polymer
is modified with a temperature- or pH-sensitive dye [3,4,7,14], or a temperature- or pH-responsive
polymer is modified with a solvatochromic dye [15–18] to translate the polymer phase transition into
a fluorescent or absorbance output signal. Whereas a variety of polymeric sensors are reported in
literature either for pH [19–21] or temperature[22,23], only a few dual-responsive polymers for both
pH and temperature exist [3,7,24]. A recent emerging strategy to design such dual-responsive system
incorporates a pH-responsive solvatochromic dye, such as Disperse Red 1, into a thermoresponsive
polymer [3,25].
Conjugated polymers (CPs) are widely applied for sensing applications. CPs provide a platform
for prominent detection of chemical and biological entities due to a highly delocalized electronic
backbone structure and optical signal amplification that are sensitive to changes in the structure
induced by minor quantities of analytes [26–29]. For biomedical applications and aqueous sensing,
water-soluble CPs are required, which can be achieved by introducing ionic or polar functional groups
to the side chains, such as sulfonate, phosphate, carboxylate, quaternary ammonium, and ethylene
glycol units [8,13]. Such water-soluble CPs have been utilized to develop biosensors for detection
of DNA strands [30–34] and proteins [35,36] and for cellular imaging [3,37]. Aggregation induced
by the analyte, which results in a change of the spectral signal (absorbance or fluorescence), is the
most common method for analyte sensing with CPs [38,39], although aggregation can also lead to
fluorescence quenching or signal enhancement [40].
Carboxylate-functionalized conjugated polymers developed for ion sensing demonstrated
cation-induced aggregation with divalent cations [38], where quenching is enhanced with cation
concentration increase [41]. Fluorescent carboxylated polyelectrolytes, water-soluble poly(p-phenylene
ethylene) [42] and poly(thiophene) [43], are credited for their high sensitivity in biosensing applications.
Sensing by these polymers of viologen and protein calmodulin, respectively, is based upon Ca2+
binding, which gives a shift in absorption or emission spectra based on aggregation of polymer or
change in conformation.
In our recent report [13], we described the synthesis of water-soluble poly(phenylene vinylene)s
(PPVs) having both carboxylic acid and methoxyoligoethylene glycol pendant groups (Scheme 1).
In the current work, we focused our attention to the multiresponsive sensing behavior of these
materials, as the carboxylic acid moiety should bring pH responsiveness while the oligoethylene glycol
side chains may induce thermoresponsive behavior. Furthermore, the conjugated polymer backbone
may enable a direct visual output signal upon a change in solubility and/or aggregation state of the
polymer. Thermoresponsive oligoethylene glycol-modified polymers are gaining increased attention as
alternatives for the well-known poly(N-isopropylacrylamide) (PNIPAAM) [44–46]. Such oligoethylene
glycol-modified polymers often exhibit lower critical solution temperature (LCST) behavior in
water, meaning that they are water-soluble at lower temperatures and undergo entropy-driven
phase separation at a critical temperature. Opposite thermoresponsive behavior, so-called upper
critical solution temperature (UCST), whereby the polymer is insoluble at lower temperatures and
solubilizes upon heating, is less common in water and requires strong interpolymer interactions [47,48].
UCST behavior can also be induced by combining charged polymers with oppositely charged species,
such as metal ions, metal ligand complexes, or organic compounds. A prime example is the UCST
behavior of alginate in presence of calcium(II) ions [49–54]. Inspired by these systems, we additionally
explored whether the acid-functionalized PPVs under study exhibit UCST behavior in presence of
calcium(II) analogs (Scheme 1).

23

Polymers 2016, 8, 365

Scheme 1. General polymer structure with stimuli-responsive functional units, its response to pH,
temperature, and upper critical solution temperature (UCST)-like behavior when complexed with Ca2+ .

2. Experimental
2.1. Materials
All chemicals were commercially available and used as received unless otherwise stated.
Milli-Q water was obtained from a Sartorius Arium 611(Brussels, Belgium) with a Sartopore 2 150
(0.45 + 0.2 mm pore size) cartridge filter (resistivity less than 18.2 MU cm). NaOH and HCl are from
Acros Organics (Geel, Belgium) and were diluted with Milli-Q water. Functionalized PPVs were
synthesized as previously reported by us [13].
2.2. Instrumentation/Methods
The pH was recorded with a Mettler Toledo FE20 FiveEasy Benchtop pH meter (Brussels, Belgium).
All spectroscopic measurements were carried out in 1 cm quartz cuvettes. The fluorescence emission
spectra were measured on a Cary Eclipse fluorescence spectrometer (Santa Clara, CA, USA) with Peltier
temperature control under stirring. The excitation wavelength was set at 392 nm with photomultiplier
tube voltage at 600 V. The split width of the excitation and emission were both 5 nm. The fluorescence
spectra were recorded with a recording emission range of 400–750 nm. The thermal measurements
were taken between 5 and 85 ◦ C with heating/cooling rate of 1 ◦ C/min. Optical absorption (UV–vis)
spectra were measured using a Varian Cary 300 Bio UV–visible spectrometer (Santa Clara, CA, USA)
equipped with a Cary Peltier temperature control while stirring. Samples were measured in quartz
cuvettes with a pathlength of 1.0 cm in the wavelength range of 200–600 nm. The concentration of
each sample was 1.0 mg/mL in Milli-Q water.
1 H-NMR spectra were recorded on a Bruker Avance 400 or 300 MHz spectrometer (Billerica, MA,
USA) at room temperature in deuterated solvents. Chemical shifts (δ) are given in ppm relative to TMS.
Size-exclusion chromatography (SEC) was performed on an Agilent 1260-series HPLC system
(Santa Clara, CA, USA) equipped with a 1260 online degasser, a 1260 ISO-pump, a 1260 automatic
liquid sampler, a thermostatted column compartment, a 1260 diode array detector (DAD) and
a 1260 refractive index detector (RID). Analyses were performed on a PSS Gram30 column in series
with a PSS Gram1000 column at 50 ◦ C. DMAc containing 50 mM of LiCl was used as eluent at a flow
rate of 0.6 mL/min. The SEC traces were analyzed using the Agilent Chemstation software with the
GPC add-on. Size exclusion chromatography was used to evaluate the number average molecular
weight (Mn ) and dispersity (Ð) against PMMA standards.
Turbidity measurements were performed on a Crystal16 from Technobis Crystallization Systems
(Alkmaar, The Netherlands) at a wavelength of 600 nm. The samples were fully dissolved at
pH = 10 (0.5 mg/mL), after which the samples were placed in the instrument and cooled to
10 ◦ C. The transmittance was measured during at least two controlled cooling/heating cycles with
a cooling/heating rate of 1 ◦ C·min−1 while stirring in PS cuvettes controlled by block temperature probe.
24

Polymers 2016, 8, 365

2.3. Synthesis and Characterization
2.3.1. Synthesis of 1,4-bis(2-(2-(2-Methoxyethoxy)ethoxy)ethoxy)-2,5-divinylbenzene
To a solution of 1,4-bis(2-(2-(2-methoxyethoxy)ethoxy)ethoxy)-2,5-diiodobenzene (1 g, 1.5 mmol)
and vinyltributyltin (1 mL, 3.7 mmol) in DMF (6 mL), triphenylphosphine palladium(0) was added
(0.08 g, 0.07 mmol) and the reaction vessel was immediately sealed and degassed via freeze-pump-thaw
(ca. 5 cycles). The mixture was then heated to 100 ◦ C and stirred for 6 h. The solution was then allowed
to cool before diluting with DCM (ca. 50 mL) and filtering into cold water. The organic layer was
then washed three times with water, once with brine solution, dried (Na2 SO4 ), and reduced via
vacuo. This was then purified by passing through a hexane column to yield the title compound,
in approximately 40% yield, as a dark red oil.
δH (300 MHz; CDCl3 ; Me4 Si): 7.25 (2H, s, ArH), 4.18 (4H, t, J = 6 Hz, CH2 ), 3.82 (4H, t, J = 6 Hz,
CH2 ), 3.80–3.75 (4H, m, CH2 ), 3.74–3.68 (8H, m, CH2 ), 3.66–3.49 (4H, m, CH2 ), 3.38 (6H, s, OCH3 ).
IR: νmax(neat)/cm−1 ; 2875 (CH, aromatic), 1352 (C–O, ether), 1242 (C–O, ether), 1176 (C–O,
ether), 1096 (C–O, ether) 395 (C–I). m/z (CI+) 677 (MH+, 100%). High Resolution (CI+): found (MH+):
677.0100 C22 H31 O8 I2 requires 677.0103. The 1H NMR data was in agreement with literature values [13].
2.3.2. Synthesis of PMEE-PDTriG (Figure 1)
A solution of tri-n-butylamine (0.60 mL, 1.86 mmol), palladium acetate (15.00 mg, 0.02 mmol),
tri-o-tolylphosphine (80.00 mg, 0.25 mmol), 1,4-bis(2-(2-(2-methoxyethoxy)ethoxy)ethoxy)-2,5divinylbenzene (0.2 g, 0.44 mmol) and 1,4-diiodo-2,5-dimethoxybenzene (0.36 mmol), 6,6 -((2,5-diiodo1,4-phenylene)bis(oxy))dihexanoic acid (0.055 g, 0.09 mmol) in DMF (4 mL) under an atmosphere of
nitrogen, was degassed using freeze-thaw cycles (×5), before heating at 90 ◦ C for 24 h. The mixture
was then filtered and dissolved in basic water, this was then dialyzed against deionized water for
2 days with a 6–8 kD MWCO cellulose membrane. Water was removed in vacuum to give the title
compound PMEE-PDTriG, in approximately 50% yield, as a red gel-like solid.
δH (400 MHz; CDCl3 ; Me4 Si): 7.96 (2H, s, ArH), 7.54–7.51 (2H, m, ArH), 7.29–7.20 (4H, m, ArH),
6.92–6.85 (8H, m, CH=CH), 4.24 (8H, m, OCH2 ), 3.92 (6H, bs, OCH3 ), 3.69 (6H, bs, OCH3 ), 3.58–3.19
(m, OCH2 , OCH3 ), 2.29 (4H, m, CH2 ), 1.72–1.50 (12H, m, CH2 ).

Figure 1. Synthetic scheme of PMEE-PDTriG.

2.4. Sample Preparation
Samples for fluorescent measurements first were dissolved in Milli-Q water at pH = 13 and
adjusted to lower pH values by dilute HCl. All samples were prepared at 0.01 mg/mL. For UV–vis
measurements, samples were prepared at pH = 10 (0.5 mg/mL) in Milli-Q water (Milli-Q, resistivity
≤18.2 MΩ·cm).

25

Polymers 2016, 8, 365

3. Results and Discussion
3.1. Synthesis and Characterization
PPVs with functional carboxylic acid and oligoethylene glycol units were previously reported
to show high solubility and high luminescence in aqueous and organic media [13]. Here, specifically
2,5-substituted PPV polymers decorated with methoxytriethylene glycol units and carboxylic acid
groups (denoted as PDTriG) and its more hydrophobic copolymer with methoxytrietylene glycol
and methoxy side groups (denoted as PMEE-PDTriG) were investigated to evaluate their responsive
behavior (Figure 2). PDTriG was synthesized as previously reported [13] whereas the monomer
and polymer of PMEE-PDTriG were synthesized using a slightly modified literature procedure [13].
These polymers were characterized by means of NMR and SEC. Size exclusion chromatography (SEC)
with DMA as a solvent and PMMA as calibration standard was used to obtain the average molecular
mass of PDTriG (Mn = 197,000; Ð= 1.34) and PMEE-PDTriG (Mn = 108,200; Ð = 1.04).
OR1

OR1

OR1
R1O

R1O

OR2

OR2

R2O

OR3

R2O
R1O

R3O

PMEE-PDTriG

PDTriG
R1= CH2CH2OCH2CH2OCH2CH2OCH3
R2= C5H10CO2H
R3= CH3

Figure 2. Structures of the polymers: PDTriG and PMEE-PDTriG.

3.2. Responsive Behavior
3.2.1. pH-Responsive Behavior
It is well known that the optical properties of CPs are governed by the polarity of the polymer,
conformational changes in the backbone, the polarity of the solvent, intramolecular dynamics, and
interchain interactions [55,56]. A change in the conformation of the polymer chain is dictated by
different variables, among which solvent and temperature are prominent. Accordingly, any external
factor effecting backbone configuration influences the photophysical features of CPs. It was shown
previously that the emission spectra of functional PPVs were highly dependent on acidity of the
media [13]. Photophysical properties of both PDTriG and PMEE-PDTriG were studied in aqueous
media as function of pH by fluorescence spectroscopy. It was observed that the polymers did not readily
dissolve in water and in buffer solution, unless the pH was increased to 13. Therefore, the samples
were first fully solubilized at pH = 13 followed by addition of dilute hydrochloric acid solution to
adjust the pH to the desired pH values. When following this procedure the polymers remained in
aqueous solution and no macroscopic precipitate was observed. As shown in the fluorescence spectra
in Figure 3a, PDTriG has a maximum emission at 498 nm and an excitation maximum at 392 nm,
as measured at pH = 13, 0.01 mg/mL. PMEE-PDTriG was found to have an emission λmax of 478 nm
and an excitation λmax of 378 nm when measured at the same concentration and pH (Figure 3b)
indicating both polymers have similar optical properties. Thus, all the emission measurements were
carried out at an excitation maximum wavelength λmax = 392 nm for PDTriG and λmax = 378 nm
for PMEE-PDTriG. The fluorescence intensity of both polymers was found to be dependent on the
pH of the solution, as expected based on the presence of the carboxylic acid groups. While at pH
below 7 quenching of the fluorescence was observed, strong emission was found at higher pH (pH > 7).
This change in emission intensity can be attributed to the enhanced solubility of the polymer at higher
pH values due to deprotonation of the carboxylic acid groups, thereby suppressing aggregation of the
polymer and quenching of fluorescence [57]. Fluorescence intensity increased for both of the polymers
26

Polymers 2016, 8, 365

as pH values increased, without shifting of emission maxima peaks (Figure 3c,d). A slight blue shift is
observed for PDTriG with decreasing acidity, at pH 6 and below, due to aggregation of amphiphilic
CPs as they become protonated [56]. It was, however, observed that the fluorescence intensity of
PDTriG decreased in time at pH 13 and at pH 6 and lower. In contrast, PMEE-PDTriG was found to be
more stable and only showed a decrease in fluorescence intensity at pH 13 and below pH 3. At lower
pH values, protonation of the carboxylic acid groups is presumed to cause insolubility of the polymers;
this was proved by complete loss of emission intensity, which was not restored even after adjusting
pH value back to 10. Especially, the higher stability at lower pH values for PMEE-PDTriG indicates that
the presence of protonated carboxylic acid groups induces degradation and/or insolubility, which is
more pronounced for PDTriG as it has more carboxylic acid groups. Note that for sample preparation,
the polymers were only dissolved in pH 13 for a short time to avoid degradation of the materials,
presumably resulting from the high concentration of hydroxyl anions facilitating degradation of the
polymer backbone.
b

a

800

600

600

excitation

Relative Fluoresent unit

Fluorescence Intensity

700

emission

500
400
300
200

excitation

emission

400

200

100
0
200

300

400

500

600

0
200

700

300

400

c
800

700

pH13
pH10
pH7
pH6
pH5
pH4
pH3

800
700

700

600

600

Emission Intensity

Emission Intensity

600

d

pH 13
pH 10
pH 7
pH 6

900

500
400
300

500
400
300
200

200

100

100
0
400

500

600

0
400

700

500

Wavelength(nm)

600

700

Wavelength(nm)

e

f

900

400

R=0.955

800

R=0.995

350

700
600

Emission Intensity

Emission Intensity

500

Wavelength(nm)

Wavelength(nm)

500
400
300

300

250

200

200
150

100
0

100

4

6

8

10

12

14

pH

2

4

6

8

10

12

14

pH

Figure 3. Emission and excitation spectra of (a) PDTriG; (b) PMEE-PDTriG, measured at pH = 13, 0.01 M.
Fluorescence spectra of (c) PDTriG; (d) PMEE-PDTriG at different pH values at 20 ◦ C, 0.01 mg/mL.
Linear fit of fluorescence intensity of (e) PDTriG (f) PMEE-PDTriG as a function of pH at emission λmax .

27

Polymers 2016, 8, 365

When plotting the fluorescence intensity of PDTriG versus pH, it is clear that no good correlation
can be drawn, possibly due to the low stability of the polymers. However, PMEE-PDTriG revealed
a linear decrease of fluorescence intensity with decreasing pH in between pH 3 and pH 10, making it
suitable as fluorescent pH sensor. This rather broad pH range also facilitates employment in a biological
environment as it covers the physiological pH range.
3.2.2. Thermoresponsive Behavior
As mentioned above, the conformation of the π-conjugated backbone has substantial impact on the
photophysical properties of CPs [8]. Therefore, thermoresponsive behavior of the studied PPVs would
also induce a change in solubility that also strongly influences the backbone conformation, directly
leading to fluorescent output signal. Possible temperature effects on the photoluminescent properties
of the more stable PMEE-PDTriG were studied by investigating the fluorescence intensity as function
of temperatures, where it was anticipated that these polymers may exhibit LCST behavior based on the
methoxyoligoethylene glycol side chains. Fluorescence intensity of the PMEE-PDTriG was measured at
different pH values upon heating from 20 to 80 ◦ C, revealing a minor decrease in fluorescence intensity
with increasing temperature due to increased chain mobility leading to a decrease in conjugation length
(Figure 4). The relative order in fluorescence intensity at different pH values is, however, retained
at different temperatures. Nonetheless, this minor temperature influence will make it necessary to
recalibrate the pH-sensing response at different temperatures.
a

heating1
cooling1
heating2
cooling2
heating3
cooling3

500

400

Emission Intensity

400

Emission Intensity

b

pH13
pH10
pH7
pH6
pH5
pH4
pH3

500

300

200

300

200

100

100

0

0
20

30

40

50

60

70

0

80

10

20

30

50

60

70

80

Temperature ( C)

c

d

heating1
cooling1
heating2
cooling2
heating3
cooling3

500

heating1
cooling1
heating2
cooling2
heating3
cooling3

500

400

Emission Intensity

400

Emission Intensity

40
o

o

Temperature ( C)

300

200

100

300

200

100

0

0

0

10

20

30

40

50

60

70

80

0

o

10

20

30

40

50

60

70

80

o

Temperature ( C)

Temperature ( C)

Figure 4. Fluorescence emission as a function of temperature at different pH values (a) PMEE-PDTriG.
Thermal stability with heating and cooling cycles of PMEE-PDTriG at (b) pH = 13; (c) pH = 10;
(d) pH = 7.

A more critical look at Figure 4 reveals some minor disturbances in fluorescence intensity at lower
temperature, especially at pH 7. Intrigued by this effect and to probe the stability of the polymer, the
fluorescence intensity was recorded during multiple heating–cooling cycles at pH 7, 10, and 13 in
between 10 and 80 ◦ C (Figure 4b–d). First of all, these experiments revealed that PMEE-PDTriG is stable
during multiple cycles; only the first heating shows a slightly different behavior at pH 7 and 10, most
likely related to solubilization since this effect is no longer seen in the second and third heating runs.
28

Polymers 2016, 8, 365

Secondly, it is evident that PMEE-PDTriG reveals distinct thermoresponsive behavior around 10–20 ◦ C.
This abrupt increase in fluorescence emission intensity upon heating is most likely due to LCST-like
behavior of the oligoethylene glycol side chains, as such LCST behavior is an abrupt cooperative
dehydration effect. More specifically, the observed increase in fluorescence emission intensity may be
attributed to (partial) dehydration of the oligoethylene glycol side chains, leading them to collapse.
No macroscopic phase separation is observed, as the PMEE-PDTriG remains stabilized and solubilized
by the charged carboxylate groups. Interestingly, the reverse process of rehydration of the oligoethylene
glycol side chains only occurs around 10 ◦ C, representing some hysteresis in the responsive behavior.
This observation may be related to the presence of the hydrophobic alkyl chains on the carboxylic acid
side chains facilitating the formation of hydrophobic domains with the dehydrated oligoethylene glycol
chains, thereby making it more difficult to rehydrate these oligoethylene glycol chains during cooling.
To gain further insights into the effect of temperature and, especially, dehydration of the
oligoethylene glycol side chains on the polymer chain conformation, full emission spectra were
recorded at different temperatures (Figure 5). Changing the temperature, however, showed no shift
in emission wavelength maxima but only resulted in a decrease in fluorescence intensity. The low
emission intensity at lower temperatures showed a pronounced increase at 9 ◦ C, giving a jump of
emission intensity as depicted in Figure 5b. Therefore, it may be speculated that the collapse of
the oligoethylene glycol side chains leads to an increase in chain rigidity, thereby increasing the
conjugation length that results in enhanced emission intensity. It is noteworthy to mention that the
more hydrophilic PDTriG did not show such thermoresponsive behavior.
a

b
500

300

200

350

Emission Intensity

Emission Intensity

400

477nm

400

0

20 C
0
18 C
0
16 C
0
14 C
0
12 C
0
10 C
0
9C
0
8C
0
7C
0
6C
0
5C
0
4C

300

250

100

0

200
400

450

500

550

600

650

700

2

Wavelength(nm)

4

6

8

10

12

14

16

18

20

Temperature

Figure 5. (a) Fluorescence emission spectra of PMEE-PDTriG at low temperatures, from 4 to 20 ◦ C;
(b) emission intensity at emission maxima, λmax versus temperature plot.

The observed thermoresponsive emission behavior of PMEE-PDTriG may be utilized for aqueous
temperature sensing, whereby the hysteresis gap could lead to a thermal memory of the solution
temperature [58,59].
3.2.3. UCST-Like Behavior with Calcium(II) Ions
Besides studying the potential of these functional PPVs as pH and temperature sensors, we were
interested to exploit the presence of carboxylate groups for inducing UCST-like thermoresponsive
solubility behavior. Therefore, PDTriG was employed as it has a higher load of carboxylic acid
groups. The polymer was solubilized in basic solution so that the carboxylic acid groups were
present as carboxylate groups, which are known to bind with calcium(II) ions in a 2:1 fashion [60–62].
Upon addition of half an equivalent of calcium(II) ions compared to the carboxylate groups, all polymer
precipitated out of solution, ascribed to interpolymer crosslinking by calcium(II)–carboxylate
coordination as schematically depicted in Scheme 2. By heating up the polymer–calcium(II) precipitate,
the supramolecular association is weakened leading to dissolution of the polymer and the calcium(II)
ions, thereby providing reversible binding of calcium(II) as function of temperature.

29

Polymers 2016, 8, 365

Scheme 2. Schematic representation of the reversible binding of PDTriG with CaCl2 in basic media
leading to UCST behavior.

The complexation and precipitation of the polymer with calcium(II) ions is also shown in Figure 6,
indicated by the loss of UV-absorbance as well as by the photographs of the vial. As the precipitated
PDTriG–calcium complex is held together by noncovalent supramolecular interactions, we explored
whether heating would lead to sufficient lowering of the interaction strength to bring the polymer back
in solution. Indeed, heating the sample vial led to regaining of the yellow solution as the precipitate
redissolved. Subsequent cooling led to precipitation demonstrating the reversibility of this UCST-like
thermoresponsive behavior of PDTriG. Temperature response of the PDTriG and PDTriG–calcium
complex is further shown by UV–vis turbidimetry. As depicted in Figure 7, heating the samples from
10 to 70 ◦ C gives no change in the transmittance of the polymer solution, whereas the polymer–calcium
complex goes through UCST-like phase transition.

Figure 6. (a) UV–vis spectra of PDTriG (pH = 10, 0.5 mg/mL) on addition of CaCl2 (1:1 equivalents) in
basic media at 25 ◦ C after 5 min (red), after 30 min (green), after 2 h (blue), after 3 h (light blue), and at
60 ◦ C (purple); (b) visual observation of complexation within time demonstrating the phase separation
of PDTriG in presence of calcium(II) ions.

Transmittance [%]

100

80

PDTriG
PDTriG+CaCl2

60

40

20

20

30

40

50

60

Temperature [°C]

Figure 7. Transmittance as a function of temperature for PDTriG (black) and PDTriG with CaCl2 (red)
at pH = 10 (0.5 mg/mL) measured at a wavelength of 600 nm.

30

Polymers 2016, 8, 365

4. Conclusions
Functional groups on PPV not only give high solubility in water but also provide new properties
for sensing regime of the polymers. In this study, water-soluble PPVs with carboxylic acid groups
and methoxyoligoethylene units as pendant side chains are reported to have high luminescence in
aqueous media with multiresponsive behavior. Higher solubility is attained for the polymers with
increased pH values as more carboxylic acid groups are deprotonated, thereby inhibiting aggregation
of the polymers and enhancing fluorescence intensity. This phenomenon is observed for both of the
polymers, providing pH response behavior of PDTriG in between pH 6–13 and in between pH 3–13
for PMEE-PDTriG. Higher solubility of the reported polymers was also achieved by influence of
temperature, making them polymeric-thermometers as they are thermosensitive in aqueous media.
Photophysical properties at different pH values in a broad temperature window (10–80 ◦ C) shows that
there is a temperature response for both polymers. Moreover, stability of the polymers was maintained
throughout multiple heating–cooling cycles. Furthermore, UCST-like thermoresponsive behavior was
demonstrated for PDTriG with Ca2+ ions, which are known to bind divalently with carboxylate groups.
The combined pH-responsive and thermoresponsive properties of these PPVs are promising features
for the design of sensors, specifically for their use as dual sensors for temperature and pH values.
Such dual sensors are a highly desired addition to the class of polymeric sensors and a related system
based on a temperature-responsive polymer with a pH-responsive solvatochromic dye in the side
chain. The hysteresis gap in the heating–cooling cycles at lower temperatures for PMEE-PDTriG brings
about an optional application of a solution thermometer with memory function.
Acknowledgments: Kanykei Ryskulova and Richard Hoogenboom thank Ghent University (BOF) and FWO
Flanders for financial support. Richard Hoogenboom is grateful to BELSPO (IAP VII/5 Functional Supramolecular
Systems FS2). Thomas Kerr-Phillips thanks MacDiarmid Institute for Advanced Materials and Nanotechnology
for doctoral scholarship.
Author Contributions: Jadranka Travas-Sejdic and Richard Hoogenboom conceived and designed the
experiments and supervised the work; David Barker and Hui Peng designed the polymers and David Barker
supervised the polymer synthesis; Anupama Rao Gulur Srinivas and Thomas Kerr-Phillips synthesized the
polymers; Kanykei Ryskulova performed the experiments and analyzed the data; Kanykei Ryskulova and
Richard Hoogenboom wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.

4.

5.
6.
7.
8.

Schmaljohann, D. Thermo- and pH-responsive polymers in drug delivery. Adv. Drug Deliv. Rev. 2006, 58,
1655–1670. [CrossRef] [PubMed]
Mura, S.; Nicolas, J.; Couvreur, P. Stimuli-responsive nanocarriers for drug delivery. Nat. Mater. 2013, 12,
991–1003. [CrossRef] [PubMed]
Yin, L.Y.; He, C.S.; Huang, C.S.; Zhu, W.P.; Wang, X.; Xu, Y.F.; Qian, X.H. A dual pH and temperature
responsive polymeric fluorescent sensor and its imaging application in living cells. Chem. Commun. 2012, 48,
4486–4488. [CrossRef] [PubMed]
Bao, Y.Y.; De Keersmaecker, H.; Corneillie, S.; Yu, F.; Mizuno, H.; Zhang, G.F.; Hofkens, J.; Mendrek, B.;
Kowalczuk, A.; Smet, M. Tunable ratiometric fluorescence sensing of intracellular pH by aggregation-induced
emission-active hyperbranched polymer nanoparticles. Chem. Mater. 2015, 27, 3450–3455. [CrossRef]
Peng, H.S.; Stolwijk, J.A.; Sun, L.N.; Wegener, J.; Wolfbeis, O.S. A nanogel for ratiometric fluorescent sensing
of intracellular pH values. Angew. Chem. Int. Ed. 2010, 49, 4246–4249. [CrossRef] [PubMed]
Li, K.; Liu, B. Water-soluble conjugated polymers as the platform for protein sensors. Polym. Chem. 2010, 1,
252–259. [CrossRef]
Pietsch, C.; Hoogenboom, R.; Schubert, U.S. Soluble polymeric dual sensor for temperature and pH value.
Angew. Chem. Int. Ed. 2009, 48, 5653–5656. [CrossRef] [PubMed]
Zhu, C.; Liu, L.; Yang, Q.; Lv, F.; Wang, S. Water-soluble conjugated polymers for imaging, diagnosis, and
therapy. Chem. Rev. 2012, 112, 4687–4735. [CrossRef] [PubMed]

31

Polymers 2016, 8, 365

9.
10.

11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.

25.
26.
27.
28.
29.
30.
31.
32.
33.

Cabane, E.; Zhang, X.Y.; Langowska, K.; Palivan, C.G.; Meier, W. Stimuli-responsive polymers and their
applications in nanomedicine. Biointerphases 2012, 7, 9. [CrossRef] [PubMed]
Cheng, R.; Meng, F.H.; Deng, C.; Klok, H.A.; Zhong, Z.Y. Dual and multi-stimuli responsive polymeric
nanoparticles for programmed site-specific drug delivery. Biomaterials 2013, 34, 3647–3657. [CrossRef]
[PubMed]
Gil, E.S.; Hudson, S.M. Stimuli-reponsive polymers and their bioconjugates. Prog. Polym. Sci. 2004, 29,
1173–1222. [CrossRef]
Galaev, I.Y.; Mattiasson, B. “Smart” polymers and what they could do in biotechnology and medicine.
Trends Biotechnol. 1999, 17, 335–340. [CrossRef]
Gulur Srinivas, A.R.; Kerr-Phillips, T.E.; Peng, H.; Barker, D.; Travas-Sejdic, J. Water-soluble anionic
poly(p-phenylene vinylenes) with high luminescence. Polym. Chem. 2013, 4, 2506–2514. [CrossRef]
Enzenberg, A.; Laschewsky, A.; Boeffel, C.; Wischerhoff, E. Influence of the near molecular vicinity on the
temperature regulated fluorescence response of poly(N-vinylcaprolactam). Polymers 2016, 8, 109. [CrossRef]
Zhang, Q.L.; Vancoillie, G.; Mees, M.A.; Hoogenboom, R. Thermoresponsive polymeric temperature sensors
with broad sensing regimes. Polym. Chem. 2015, 6, 2396–2400. [CrossRef]
Pietsch, C.; Vollrath, A.; Hoogenboom, R.; Schubert, U.S. A fluorescent thermometer based on
a pyrene-labeled thermoresponsive polymer. Sensors 2010, 10, 7979–7990. [CrossRef] [PubMed]
Uchiyama, S.; Kawai, N.; De Silva, A.P.; Iwai, K. Fluorescent polymeric AND logic gate with temperature
and pH as inputs. J. Am. Chem. Soc. 2004, 126, 3032–3033. [CrossRef] [PubMed]
Jin, Q.A.; Liu, G.Y.; Ji, J.A. Micelles and reverse micelles with a photo and thermo double-responsive block
copolymer. J. Polym. Sci. Polym. Chem. 2010, 48, 2855–2861. [CrossRef]
Uchiyama, S.; Makino, Y. Digital fluorescent pH sensors. Chem. Commun. 2009, 19, 2646–2648. [CrossRef]
[PubMed]
Korostynska, O.; Arshak, K.; Gill, E.; Arshak, A. Review on state-of-the-art in polymer based pH sensors.
Sensors 2007, 7, 3027–3042. [CrossRef]
Wang, S. Dopamine modified cationic conjugated polymers as a new platform for pH sensing and autophagy
imaging. Adv. Funct. Mater. 2013, 23, 764–769.
Gota, C.; Uchiyama, S.; Ohwada, T. Accurate fluorescent polymeric thermometers containing an ionic
component. Analyst 2007, 132, 121–126. [CrossRef] [PubMed]
Wang, X.-D.; Wolfbeis, O.S.; Meier, R.J. Luminescent probes and sensors for temperature. Chem. Soc. Rev.
2013, 42, 7834–7869. [CrossRef] [PubMed]
Khine, Y.Y.; Jiang, Y.Y.; Dag, A.; Lu, H.X.; Stenzel, M.H. Dual-responsive pH and temperature sensitive
nanoparticles based on methacrylic acid and Di(ethylene glycol) methyl ether methacrylate for the triggered
release of drugs. Macromol. Biosci. 2015, 15, 1091–1104. [CrossRef] [PubMed]
Wang, X.D.; Meier, R.J.; Wolfbeis, O.S. A fluorophore-doped polymer nanomaterial for referenced imaging
of pH and temperature with sub-micrometer resolution. Adv. Funct. Mater. 2012, 22, 4202–4207. [CrossRef]
McQuade, D.T.; Pullen, A.E.; Swager, T.M. Conjugated polymer-based chemical sensors. Chem. Rev. 2000,
100, 2537–2574. [CrossRef] [PubMed]
Feng, X.L.; Liu, L.B.; Wang, S.; Zhu, D.B. Water-soluble fluorescent conjugated polymers and their interactions
with biomacromolecules for sensitive biosensors. Chem. Soc. Rev. 2010, 39, 2411–2419. [CrossRef] [PubMed]
Aydemir, N.; Malmstrom, J.; Travas-Sejdic, J. Conducting polymer based electrochemical biosensors.
Phys. Chem. Chem. Phys. 2016, 18, 8264–8277. [CrossRef] [PubMed]
Zhu, B.; Alsager, O.A.; Kumar, S.; Hodgkiss, J.M.; Travas-Sejdic, J. Label-free electrochemical aptasensor for
femtomolar detection of 17 β-estradiol. Biosens. Bioelectron. 2015, 70, 398–403. [CrossRef] [PubMed]
Liu, B.; Bazan, G.C. Homogeneous fluorescence-based DNA detection with water-soluble conjugated
polymers. Chem. Mater. 2004, 16, 4467–4476. [CrossRef]
Gaylord, B.S.; Heeger, A.J.; Bazan, G.C. DNA detection using water-soluble conjugated polymers and peptide
nucleic acid probes. Proc. Natl. Acad. Sci. USA 2002, 99, 10954–10957. [CrossRef] [PubMed]
Jadranka, T.-S.; AnupamaRaoGulur, S. Optical gene detection using conjugated polymers. In Optical Properties
of Functional Polymers and Nano Engineering Applications; CRC Press: Boca Raton, FL, USA, 2014; pp. 61–86.
Srinivas, A.R.G.; Peng, H.; Barker, D.; Travas-Sejdic, J. Switch on or switch off: An optical DNA sensor based
on poly(p-phenylenevinylene) grafted magnetic beads. Biosens. Bioelectron. 2012, 35, 498–502. [CrossRef]
[PubMed]
32

Polymers 2016, 8, 365

34.
35.
36.
37.
38.
39.
40.
41.
42.
43.

44.
45.
46.
47.
48.
49.

50.

51.
52.

53.
54.
55.

56.

57.

Srinivas, A.R.G.; Barker, D.; Travas-Sejdic, J. “Switch-on” DNA sensor based on poly (p-phenylene vinylenes)
bound tentacle probes. Pure Appl. Chem. 2015, 87, 707–715. [CrossRef]
Giepmans, B.N.G.; Adams, S.R.; Ellisman, M.H.; Tsien, R.Y. Review—The fluorescent toolbox for assessing
protein location and function. Science 2006, 312, 217–224. [CrossRef] [PubMed]
Fan, C.H.; Plaxco, K.W.; Heeger, A.J. High-efficiency fluorescence quenching of conjugated polymers by
proteins. J. Am. Chem. Soc. 2002, 124, 5642–5643. [CrossRef] [PubMed]
Zhang, W.; Sun, H.; Yin, S.Y.; Chang, J.J.; Li, Y.H.; Guo, X.Y.; Yuan, Z. Bright red-emitting polymer dots for
specific cellular imaging. J. Mater. Sci. 2015, 50, 5571–5577. [CrossRef]
Thomas, S.W.; Joly, G.D.; Swager, T.M. Chemical sensors based on amplifying fluorescent conjugated
polymers. Chem. Rev. 2007, 107, 1339–1386. [CrossRef] [PubMed]
Lakowicz, J.R. Principles of Fluorescence Spectroscopy, 2nd ed.; Plenm: New York, NY, USA, 1999.
Tan, C.Y.; Alas, E.; Muller, J.G.; Pinto, M.R.; Kleiman, V.D.; Schanze, K.S. Amplified quenching of a conjugated
polyelectrolyte by cyanine dyes. J. Am. Chem. Soc. 2004, 126, 13685–13694. [CrossRef] [PubMed]
Kim, I.B.; Dunkhorst, A.; Gilbert, J.; Bunz, U.H.F. Sensing of lead ions by a carboxylate-substituted PPE:
Multivalency effects. Macromolecules 2005, 38, 4560–4562. [CrossRef]
Jiang, H.; Zhao, X.Y.; Schanze, K.S. Amplified fluorescence quenching of a conjugated polyelectrolyte
mediated by Ca2+ . Langmuir 2006, 22, 5541–5543. [CrossRef] [PubMed]
Nilsson, K.P.R.; Inganas, O. Optical emission of a conjugated polyelectrolyte: Calcium-induced
conformational changes in calmodulin and calmodulin–calcineurin interactions. Macromolecules 2004, 37,
9109–9113. [CrossRef]
Vancoillie, G.; Frank, D.; Hoogenboom, R. Thermoresponsive poly(oligo ethylene glycol acrylates).
Prog. Polym. Sci. 2014, 39, 1074–1095. [CrossRef]
Weber, C.; Hoogenboom, R.; Schubert, U.S. Temperature responsive bio-compatible polymers based on
poly(ethylene oxide) and poly(2-oxazoline)s. Prog. Polym. Sci. 2012, 37, 686–714. [CrossRef]
Lutz, J.F. Thermo-switchable materials prepared using the OEGMA-platform. Adv. Mater. 2011, 23, 2237–2243.
[CrossRef]
Seuring, J.; Agarwal, S. Polymers with upper critical solution temperature in aqueous solution.
Macromol. Rapid Commun. 2012, 33, 1898–1920. [CrossRef] [PubMed]
Zhang, Q.L.; Hoogenboom, R. Polymers with upper critical solution temperature behavior in alcohol/water
solvent mixtures. Prog. Polym. Sci. 2015, 48, 122–142. [CrossRef]
Plamper, F.A.; Schmalz, A.; Ballauff, M.; Muller, A.H.E. Tuning the thermoresponsiveness of
weak polyelectrolytes by pH and light:
Lower and upper critical-solution temperature of
poly(N,N-dimethylaminoethyl methacrylate). J. Am. Chem. Soc. 2007, 129, 14538–14539. [CrossRef] [PubMed]
Ott, C.; Hoogenboom, R.; Hoeppener, S.; Wouters, D.; Gohy, J.F.; Schubert, U.S. Tuning the morphologies of
amphiphilic metallo-supramolecular triblock terpolymers: From spherical micelles to switchable vesicles.
Soft Matter 2009, 5, 84–91. [CrossRef]
Jia, X.; Chen, D.Y.; Jiang, M. Preparation of PEO-b-P2VPH(+)-S2 O8 2− micelles in water and their reversible
UCST and redox-responsive behavior. Chem. Commun. 2006, 16, 1736–1738. [CrossRef] [PubMed]
Serp, D.; Mueller, M.; von Stockar, U.; Marison, I.W. Low-temperature electron microscopy for the study of
polysaccharide ultrastructures in hydrogels. II. Effect of temperature on the structure of Ca2+ -alginate beads.
Biotechnol. Bioeng. 2002, 79, 253–259. [CrossRef] [PubMed]
Pawar, S.N.; Edgar, K.J. Alginate derivatization: A review of chemistry, properties and applications.
Biomaterials 2012, 33, 3279–3305. [CrossRef] [PubMed]
Morch, Y.A.; Donati, I.; Strand, B.L.; Skjak-Braek, G. Effect of Ca2+ , Ba2+ , and Sr2+ on alginate microbeads.
Biomacromolecules 2006, 7, 1471–1480. [CrossRef] [PubMed]
De Silva, A.P.; Gunaratne, H.Q.N.; Gunnlaugsson, T.; Huxley, A.J.M.; McCoy, C.P.; Rademacher, J.T.; Rice, T.E.
Signaling recognition events with fluorescent sensors and switches. Chem. Rev. 1997, 97, 1515–1566.
[CrossRef] [PubMed]
Van Severen, I.; Breselge, M.; Fourier, S.; Adriaensens, P.; Manca, J.; Lutsen, L.; Cleij, T.J.; Vanderzande, D.
2,5-substituted PPV-derivatives with different polarities: The effect of side chain polarity on solubility,
optical and electronic properties. Macromol. Chem. Phys. 2007, 208, 196–206. [CrossRef]
Lee, K.; Kim, H.J.; Kim, J. Design principle of conjugated polyelectrolytes to make them water-soluble and
highly emissive. Adv. Funct. Mater. 2012, 22, 1076–1086. [CrossRef]
33

Polymers 2016, 8, 365

58.

59.
60.

61.
62.

Sambe, L.; De La Rosa, V.R.; Belal, K.; Stoffelbach, F.; Lyskawa, J.; Delattre, F.; Bria, M.; Cooke, G.;
Hoogenboom, R.; Woisel, P. Programmable polymer-based supramolecular temperature sensor with
a memory function. Angew. Chem. Int. Ed. 2014, 53, 5044–5048.
De la Rosa, V.R.; Hoogenboom, R. Solution polymeric optical temperature sensors with long-term memory
function powered by supramolecular chemistry. Chem. Eur. J. 2015, 21, 1302–1311. [CrossRef] [PubMed]
Custers, J.P.A.; van Nispen, S.F.G.M.; Can, A.; De la Rosa, V.R.; Maji, S.; Schubert, U.S.; Keurentjes, J.T.F.;
Hoogenboom, R. Reversible calcium(II)-ion binding through an apparent pK(a) shift of thermosensitive
block-copolymer micelles. Angew. Chem. Int. Ed. 2015, 54, 14085–14089. [CrossRef] [PubMed]
Custers, J.P.A.; Kelemen, P.; Van den Broeke, L.J.P.; Stuart, M.A.C.; Keurentjes, J.T.F. Reversible binding of
multivalent ions by surfactant self-assembly. J. Am. Chem. Soc. 2005, 127, 1594–1595. [CrossRef] [PubMed]
Mahou, R.; Tran, N.M.; Dufresne, M.; Legallais, C.; Wandrey, C. Encapsulation of Huh-7 cells within
alginate-poly(ethylene glycol) hybrid microspheres. J. Mater. Sci. Mater. Med. 2012, 23, 171–179. [CrossRef]
[PubMed]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

34

polymers
Article

A Comprehensive Systematic Study on
Thermoresponsive Gels: Beyond the
Common Architectures of Linear Terpolymers
Anna P. Constantinou, Hanyi Zhao, Catriona M. McGilvery, Alexandra E. Porter and
Theoni K. Georgiou *
Department of Materials, Imperial College London, Royal School of Mines, Exhibition Road,
London SW7 2AZ, UK; anna.constantinou14@imperial.ac.uk (A.P.C.); hanyi.zhao15@imperial.ac.uk (H.Z.);
catriona.mcgilvery@imperial.ac.uk (C.M.M.); a.porter@imperial.ac.uk (A.E.P.)
* Correspondence: t.georgiou@imperial.ac.uk; Tel.: +44-20-7594-5177
Academic Editor: Alexander Böker
Received: 4 December 2016; Accepted: 16 January 2017; Published: 20 January 2017

Abstract: In this study, seven thermoresponsive methacrylate terpolymers with the same molar
mass (MM) and composition but various architectures were successfully synthesized using group
transfer polymerization (GTP). These terpolymers were based on tri(ethylene glycol) methyl ether
methacrylate (TEGMA, A unit), n-butyl methacrylate (BuMA, B unit), and 2-(dimethylamino)ethyl
methacrylate (DMAEMA, C unit). Along with the more common ABC, ACB, BAC, and statistical
architectures, three diblock terpolymers were also synthesized and investigated for the first time,
namely (AB)C, A(BC), and B(AC); where the units in the brackets are randomly copolymerized.
Two BC diblock copolymers were also synthesized for comparison. Their hydrodynamic diameters
and their effective pKa s were determined by dynamic light scattering (DLS) and hydrogen ion
titrations, respectively. The self-assembly behavior of the copolymers was also visualized by
transmission electron microscopy (TEM). Both dilute and concentrated aqueous copolymer solutions
were extensively studied by visual tests and their cloud points (CP) and gel points were determined.
It is proven that the aqueous solution properties of the copolymers, with specific interest in their
thermoresponsive properties, are influenced by the architecture, with the ABC and A(BC) ones to
show clear sol-gel transition.
Keywords: thermoresponsive polymers; 2-(dimethylamino)ethyl methacrylate; injectable gels;
3-D printing; group transfer polymerization (GTP); terpolymers; complex architectures;
well-defined polymers

1. Introduction
Thermoresponsive polymers are “smart” polymers which are able to respond to temperature [1–3].
The response of the polymers is indicated by a change of their properties. A special class of
thermoresponsive polymers which has become very popular covers polymers that exhibit a Lower
Critical Solution Temperature (LCST) behavior. These polymers become insoluble in aqueous media
when increasing the temperature. This behavior is explained by the “hydrophobic effect”, i.e., the entropy
of water becomes the most dominant factor and forces the polymer to precipitate out of solution [2,4].
It is the same phenomenon that in lower temperatures and concentrations forces the polymers to form
micelles [2,4]. Thermoresponsive polymers with the appropriate structural parameters, and under the
appropriate environmental conditions form 3-D networks of physically-interconnected micelles; these 3-D
networks are known as thermoresponsive gels [5]. As has been well-demonstrated, the architecture, the
molar mass (MM), the composition, and the molar mass distribution (MMD) of the thermoresponsive

Polymers 2017, 9, 31

35

www.mdpi.com/journal/polymers

Polymers 2017, 9, 31

copolymers determine whether, and at which temperature and concentration a gel is formed [1]. In the
case of aqueous solutions, these micelles are connected via well-hydrated bridges [5].
Thermoresponsive gels have been widely studied on account of their interesting applications,
including tissue engineering as injectable gels [6–8] and more recently in 3-D printing [9–12].
For thermoresponsive polymers to be used as injectable gels a sol-gel transition close to body
temperature is required for a minimal invasion administration [13]. On the other hand, for the
3-D printing application the sol-gel transition should occur either close to room or body temperature,
depending on the temperature that a stable printed structure is desirable at [11,14]. In this case, the
thermoresponsive gel should not only possess good mechanical properties in order to maintain the
printed structure, but it should also be characterized by good shear-thinning properties [11,14].
For thermoresponsive polymers to be widely applied and industrialized, their synthesis
should be easy, time- and cost-effective, reproducible, and scalable. Furthermore, a “living” or
“controlled” polymerization technique is required for the synthesis in order to produce copolymers
with well-defined structural parameters [15] since it has been well-demonstrated in the literature that
these parameters can affect the sol-gel transition [1]. Group transfer polymerization (GTP) is an ideal
polymerization method for the synthesis of methacrylate well-defined polymers because: (i) it is a fast
polymerization technique (10–15 min per block) [16]; (ii) it produces polymers with narrow MMD
(usually below 1.2) and well-defined and controllable composition [16,17]; (iii) it can produce polymers
in an industrial scale and (iv) it is cost-effective [17]. GTP is cost-effective for several reasons: it works
at higher concentrations than for example anionic polymerization, it is performed at room temperature
so there is no need to cool down or heat up the reaction, and there is a 100% conversion of the monomer
to the polymer so sequential one-pot polymerization can be easily achieved for block copolymers
without any extra pot(s) or purification steps needed [17]. Therefore, this polymerization technique
has been used in the present study and the aim was to investigate how the polymers’ architecture
affects their polymers’ thermoresponsive behavior.
Our group has previously published five research articles on thermoresponsive gels in which
several polymeric parameters have been varied [18–22]. In four of the studies, the effects of
architecture [18,20], composition [20–22], and MM [21] of triblock and statistical terpolymers on
the thermoresponsive properties have been systematically investigated. These terpolymers were
based on (i) a poly(ethylene glycol) (PEG) methacrylate unit as the hydrophilic and biocompatible
unit (A unit); (ii) an alkyl methacrylate hydrophobic unit (B unit); and (iii) the thermoresponsive and
pH-responsive 2-(dimethylamino)ethyl methacrylate (DMAEMA, C unit). The lengths of the alkyl
and the PEG side-chains have been also systematically varied by using: (i) ethyl, n-butyl, and n-hexyl
methacrylate (EtMA, BuMA, and HexMA, respectively) [18]; and (ii) di-, penta-, and nona(ethylene
glycol) methyl ether methacrylate (DEGMA, PEGMA, and NEGMA, respectively) [22]. It has been
demonstrated that the optimum parameters producing polymers with the clearest sol-gel transition
and gels of good mechanical properties are: (i) the ABC architecture [18,20]; (ii) MM ranging between
7000–10,000 g·mol−1 [21]; (iii) intermediate hydrophobic composition of around 30%–35% w/w [20–22];
(iv) BuMA as the hydrophobic unit [18]; and (v) a PEG methacrylate unit with ethylene glycol groups
between two and five [22].
Given these optimum parameters, new copolymers which are based on the unique combination
of these structural characteristics were designed. Therefore, the copolymers are based on BuMA,
and DMAEMA as B and C units, respectively, while the A unit was based on tri(ethylene glycol)
methyl ether methacrylate (TEGMA). We chose TEGMA instead of the 300 g·mol−1 PEGMA that
we normally use because based on our latest study we think it will improve the thermoresponsive
ability of the terpolymers. The target TEGMA-BuMA-DMAEMA composition was 25%–35%–40%
w/w, while the target MM was 8200 g·mol−1 . Both the composition and the MM values are within
the intermediate range established previously. The aim of this study is not only to use the optimum
structural parameters, but also to examine interesting architectures of terpolymers which have not been
systematically investigated before. More specifically, apart from the ABC, ACB, BAC, and statistical

36

Polymers 2017, 9, 31

architectures that have been previously studied, in the present study diblock terpolymers have also
been synthesized. The three possible combinations that produce different diblock terpolymers are:
(i) (AB)C, A(BC), and B(AC); the units in the brackets have been randomly copolymerized and the
resulting block has been polymerized with the third monomer to form the final diblock terpolymer.
Two BC diblock bipolymers have also been synthesized to mimic: (i) the BuMA:DMAEMA weight
percentages in the terpolymers, and (ii) the hydrophobic:hydrophilic [BuMA:(DMAEMA + TEGMA)]
weight percentages in the terpolymers. To the best of our knowledge this is the first time so many
different architectures have been synthesized and investigated in terms of their thermoresponsive
behavior. Furthermore, it should be noted that the architecture was able to be varied independently
i.e., without altering the MM and the composition that is not easy to achieve, but it was because GTP
was used.
2. Experimental
2.1. Materials
TEGMA (monomer, MM = 232.27 g·mol−1 , 94%), BuMA (monomer, 99%), DMAEMA
(monomer, 98%), activated basic aluminum oxide (Al2 O3 ·KOH), calcium hydride (CaH2 , ≥90%),
2,2-diphenyl-1-picrylhydrazyl (DPPH, free-radical inhibitor), potassium metal, sodium metal,
tetrahydrofuran (THF, polymerization solvent, HPCL grade, ≥99.9%), methyl trimethylsilyl
dimethylketene acetal (MTS, initiator, 95%), deuterated chloroform (chloroform-d, 99.8 atom% D),
sodium hydroxide pellets (NaOH, 97%), concentrated hydrochloric acid (HCl, ACS reagent, 37%),
and hydrochloric acid solution (volumetric, 1M) were purchased from Sigma Aldrich Co Ltd., Irvine,
UK. Tetrabutylammonium hydroxide (40% in water) and benzoic acid were purchased from Acros
Organics—UK distributor Fisher Scientific UK Ltd., Loughborough, UK. Tetrahydrofuran (THF, mobile
phase in GPC, GPC grade) and n-hexane (precipitation solvent) were purchased from Fisher Scientific
UK Ltd. (Loughborough, UK) and VWR International Ltd. (Lutterworth, UK), respectively. Phosphate
buffered saline (PBS, 10x solution) was purchased from Fischer Scientific UK Ltd., Loughborough, UK.
2.2. Purification of the Starting Materials
The monomers, TEGMA, BuMA, and DMAEMA, were passed twice through Al2 O3 ·KOH in order to
remove the inhibitor (monomethyl ether hydroquinone) and any acidic impurities. DPPH was added in
order to prevent free-radical polymerization and the humidity was eliminated by stirring the monomers
over CaH2 for 3 h. The monomers were then kept refrigerated until use. The polymerization solvent,
THF, was dried by refluxing for 3 days over potassium and sodium metals. The initiator, MTS, and the
monomers were distilled under vacuum prior to polymerization. The catalyst, tetrabutylammonium
bibenzoate (TBABB), was previously synthesized by tetrabutylammonium hydroxide and benzoic acid,
as reported by Dicker et al. [23], and it was dried and kept under vacuum until use. All the glassware
was dried overnight at 140 ◦ C and assembled hot under vacuum. The chemical structures of the
monomers, the initiator and the catalyst are shown in Figure 1.
2.3. Copolymer Synthesis
All the copolymers were synthesized using a cost-effective and easy to scale-up anionic
polymerization technique, specifically group transfer polymerization (GTP), which enables the
synthesis of each block in only 10–15 min. The block copolymers were synthesized via sequential
addition of the monomers. As an example, the synthesis of polymer 1, namely TEGMA9 -b-BuMA20 -bDMAEMA21 follows: As a first step, TBABB (~10 mg) was added in a 250 mL round-bottom flask.
This flask was sealed with a rubber septum and purged with argon, followed by the addition of 60 mL
freshly-distilled THF using a syringe. MTS (0.37 mL, 0.32 g, 1.8 mmol) was then syringed into the
flask. The addition of the TEGMA monomer (3.6 mL, 3.7 g, 15.9 mmol) followed and the exothermic
reaction was monitored; the temperature was increased from 24.2 to 28.3 ◦ C. As soon as the reaction

37

Polymers 2017, 9, 31

was complete, two samples of 0.1 mL were obtained for GPC and 1 H-NMR analysis. Following this,
BuMA (5.8 mL, 5.2 g, 36.3 mmol) was added, thus the temperature was increased from 25.2 to 28.9 ◦ C.
Two samples of 0.1 mL were obtained in order to analyze them by GPC and 1 H-NMR. As a last step,
6.3 mL of DMAEMA (5.9 g, 37.5 mmol) was added and an exotherm from 27.6 to 31.8 ◦ C was observed.
GPC and 1 H-NMR samples (0.1 mL each) were extracted for analysis. The polymers were precipitated
in cool n-hexane and dried in a vacuum oven at room temperature. In this study, seven terpolymers of
the same composition and target MM, but different architectures were synthesized. This was achieved
by keeping the amount of monomers the same, while the order of each monomer addition was varied.
More specifically, concerning the diblock terpolymers, the statistical bipolymer was synthesized via
simultaneous addition of the two monomers. The statistical terpolymer was synthesized by adding
all the monomers prior to the addition of the MTS. Two diblock copolymers based only on BuMA
and DMAEMA with different compositions were also synthesized by adding different amounts of
the monomers.

2
Tri(ethylene glycol) methyl
ether methacrylate
(TEGMA)
Monomer

n-Butyl methacrylate
(BuMA)
Monomer

Methyl trimethylsilyl
dimethylketene acetal
(MTS)
Initiator

2-(Dimethylamino)
ethyl methacrylate
(DMAEMA)
Monomer

Tetrabutylammonium bibenzoate
(TBABB)
Catalyst

Figure 1. Chemical structures of the monomers, the initiator, and the catalyst.

2.4. Characterization in Organic Solvents
The MM, the molar mass distribution (MMD), and the composition of all the copolymers and
their linear precursors were determined in organic solvents.
2.4.1. Gel Permeation Chromatography (GPC)
The final copolymers and their precursors were characterized in terms of their MM and MMD
by GPC. For this, an Agilent SECurity GPC system, with a Polymer Standard Service (PSS) SDV
analytical linear M column (SDA083005LIM) was used (Agilent technologies UK Ltd., Shropshire, UK).
This system is equipped with a “1260 Iso” isocratic pump and an Agilent 1260 refractive index (RI)
detector. As a mobile phase, THF with 5% vol triethylamine was used, which was pumped with a flow
rate of 1 mL·min−1 . The calibration curve was plotted by running six different linear poly(methyl
methacrylate) (PMMA) standard samples with MM equal to 2000, 4000, 8000, 20,000, 50,000, and
100,000 g·mol−1 , purchased from Fluka, Sigma Aldrich Co Ltd., Irvine, UK.

38

Polymers 2017, 9, 31

2.4.2. Proton Nuclear Magnetic Resonance Spectroscopy (1 H-NMR)
The final copolymers and their precursors were characterized in terms of their composition
by 1 H-NMR using a 400 MHz Avance Bruker NMR spectrometer (Bruker UK Ltd., Coventry, UK).
The 1 H-NMR spectra were obtained using CDCl3 as the deuterated solvent.
2.5. Characterization in Aqueous Solution
The effective dissociation constants (pKa ), the hydrodynamic diameters (dh ), the cloud points
(CP), and the thermal response of copolymers in aqueous solutions were determined.
2.5.1. Hydrogen Ion Titrations
Hydrogen ion titrations of 1% w/w aqueous polymer solutions were performed by using a
HI98103 pH-checker from Hanna instruments Ltd., Leighton Buzzard, UK. The solutions were titrated
from pH 2 to pH 12 using a 0.25 M NaOH solution. The start- and end-point of the titration of the
DMAEMA units were determined by plotting the first derivative of the titration curve and the effective
pKa was determined as the pH of the solution at which the amino groups were protonated by 50%.
2.5.2. Dynamic Light Scattering (DLS)
DLS measurements of 1% w/w aqueous polymer solutions (pH adjusted to 6 and 7) were
conducted by using a Zetasizer Nano ZSP instrument from Malvern Instruments Ltd., Malvern,
UK. The polymer solutions were filtered using nylon 0.45 μm PTFE syringe filters in order to remove
any dust and bigger aggregates. After filtering, the solutions were allowed to settle in order to ensure
complete bubble removal. Three DLS experiments were performed per sample at room temperature
and the scattered light was collected at a backscatter angle of 173◦ . The dh s reported are the mean
values determined as the diameters corresponding to the peak of maximum intensity.
These experimental dh s were compared to the theoretical ones. Four different theoretical models
were applied depending on the polymer architecture. Concerning the block copolymers, formation
of spherical micelles was assumed and the calculations were based on the projected length of the
methacrylate unit, equal to 0.254 nm, and the experimental degree of polymerization (DP). (1) When
the ABC and BC architectures are concerned, the theoretical diameter was calculated as follows:
d = (DPBuMA + 2 × DPDMAEMA ) × 0.254 nm; (2) When the hydrophobic BuMA unit forms a distinct
block at the end of the polymer chain, i.e., ACB, BAC, and B(AC), the theoretical values were calculated
using the following equation: d = [DPBuMA + 2 × (DPDMAEMA + DPTEGMA )] × 0.254 nm; (3) When the
hydrophobic BuMA unit is randomly copolymerized with either of the hydrophilic monomer (TEGMA
or DMAEMA), while the other hydrophilic monomer is in a block structure, i.e., (AB)C and A(BC),
the following formulae were used: d = (DPBuMA + DPTEGMA + 2 × DPDMAEMA ) × 0.254 nm and
d = (DPBuMA + DPDMAEMA + 2 × DPTEGMA ) × 0.254 nm; (4) When a random copolymer is concerned,
random coil is assumed to be formed, the diameter of which is calculated according to the following
equation: <dg 2 >1/2 = 2 × [2 × 2.20 × (DPTEGMA + DPBuMA + DPDMAEMA )/3]1/2 × 0.154 nm. For these
calculations, the experimental DPs were used, as calculated from GPC and 1 H-NMR results.
2.5.3. Transmission Electron Microscopy (TEM)
The TEM images were recorded using an FEI Titan 80–300 transmission electron microscope
(TEM) (FEI (part of Thermo Fisher Scientific, Hillsboro, OR, USA), equipped with an image corrector.
The instrument was operated at 80 kV to enhance contrast for bright field TEM, and an objective
aperture of 70 μm was used. 1% w/w aqueous copolymer solutions (pH adjusted at 6) was used for
the preparation of the TEM samples. The TEM samples were prepared by pipetting 3.5 μL of solution
onto holey-carbon grids. After two minutes any excess of solvent was removed using filter paper.
To increase contrast in the TEM the samples were then negatively stained by adding 30 μL of 1% w/v

39

Polymers 2017, 9, 31

uranyl acetate solution to the grid, while the TEM grids were held at an angle of 45◦ . Any remaining
excess solution was removed with filter paper, and the grids were left to dry.
2.5.4. Visual Tests
The visual tests were performed using an IKA RCT stirrer hotplate (IKA® England Ltd., Oxford,
UK), equipped with an IKA ETS-D5 temperature controller, and a continuously-stirred water-bath.
For the determination of the CPs, 1% w/w aqueous copolymer solutions were used, whereas for
the construction of the phase diagrams, 1%, 2%, 5%, 10%, 15%, 20%, 25%, and 30% w/w copolymer
solutions in PBS, were tested. In both cases, the DMAEMA units were protonated by 10% to enhance
solubility. The vials were suspended in a water-bath and a thermal response was visually inspected
every one degree from 20 to 80 ◦ C. The CP was determined as the temperature at which the solution
turned cloudy, while the gel point was determined as the temperature at which a stable gel was formed,
which did not flow upon tube inversion.
3. Results and Discussion
In this study, the synthesis of seven terpolymers was achieved using GTP and the architecture
was systematically varied, while the TEGMA-BuMA-DMAEMA composition and the MM were kept
constant at 25%–35%–40% w/w and 8200 g·mol−1 , respectively. More specifically, three triblock
terpolymers (ABC, CAB, and BAC), three diblock terpolymers ((AB)C, A(BC), and B(AC)) and
one statistical copolymer were synthesized in order to investigate the effect of architecture on the
thermoresponsive behavior. For comparison, two BC diblock copolymers were also synthesized, with
the same MM but with the BuMA-DMAEMA weight percentages mimicking the BuMA-DMAEMA
and the hydrophobic-hydrophilic weight ratio in the terpolymers. The structures of the copolymers
are schematically shown in Figure 2. The TEGMA, BuMA, and DMAEMA units are represented by
blue, orange, and green spheres, respectively.

Figure 2. Schematic showing the structures of all the copolymers studied. Blue, orange, and green
colors represent the TEGMA, BuMA, and DMAEMA units, respectively.

3.1. Structural Properties
Table 1 summarizes the structural properties of all the copolymers and their linear precursors.
Specifically, the theoretical MMs, and the experimental MMs and MMDs as resulted from GPC
analysis are shown; the latter are shown as dispersity indices (Ð). The theoretical compositions and the
experimental ones, as resulted by 1 H-NMR analysis, are also listed in Table 1.

40

Polymers 2017, 9, 31

Table 1. Theoretical and experimental molar masses (MMs) and compositions, and molar mass
distributions (MMDs) of the copolymers and their precursors.

Polymer No.

Theoretical polymer structure a

% w/w
TEGMA-BuMA-DMAEMA
Theoretical

1 H-NMR

MM Theor. b
(g·mol−1 )

Mn c
(g·mol−1 )

Ðc

1

TEGMA9
TEGMA9 -b-BuMA20
TEGMA9 -b-BuMA20 -b-DMAEMA21

100-00-00
42-58-00
25-35-40

100-00-00
42-58-00
25-35-40

2125
4960
8200

3020
6600
10,700

1.12
1.12
1.17

2

TEGMA9
TEGMA9 -b-DMAEMA21
TEGMA9 -b-DMAEMA21 -b-BuMA20

100-00-00
38-00-62
25-35-40

100-00-00
38-00-62
26-34-40

2125
5365
8200

3030
7650
10,800

1.11
1.11
1.16

3

BuMA20
BuMA20 -b-TEGMA9
BuMA20 -b-TEGMA9 -b-DMAEMA21

00-100-00
42-58-00
25-35-40

00-100-00
41-59-00
26-35-39

2935
4960
8200

2980
6700
10,100

1.12
1.08
1.13

4

TEGMA9 -co-BuMA20
(TEGMA9 -co-BuMA20 )-b-DMAEMA21

42-58-00
25-35-40

42-58-00
26-35-39

4960
8200

5940
8780

1.09
1.10

5

TEGMA9
TEGMA9 -b-(BuMA20 -co-DMAEMA21 )

100-00-00
25-35-40

100-00-00
26-34-40

2125
8200

2810
9950

1.12
1.14

6

BuMA20
BuMA20 -b-(TEGMA9 -co-DMAEMA21 )

00-100-00
25-35-40

00-100-00
26-37-37

2935
8200

4180
8580

1.10
1.13

7

TEGMA9 -co-BuMA20 -co-DMAEMA21

25-35-40

26-35-39

8200

8760

1.10

8

BuMA26
BuMA26 -b-DMAEMA28

00-100-00
00-46-54

00-100-00
00-47-53

3826
8200

6240
11,000

1.07
1.11

9

BuMA20
BuMA20 -b-DMAEMA34

00-100-00
00-35-65

00-100-00
00-37-63

2935
8200

4040
9380

1.10
1.10

a

TEGMA, BuMA, and DMAEMA stand for tri(ethylene glycol) methyl ether methacrylate, n-butyl methacrylate,
and 2-(dimethylamino) ethyl methacrylate, respectively; b The theoretical MM of the polymer was calculated as
follows: (DPTEGMA × MMTEGMA ) + (DPBuMA × MMBuMA ) + (DPDMAEMA + MMDMAEMA ) + 100 g·mol−1 ; MM and
DP stand for molar mass and degree of polymerization, respectively. The value of 100 g·mol−1 is the part of the MTS
initiator which stays on the polymer chain; c The Mn and Ð were determined by gel permeation chromatography
(GPC). The calibration curve was plotted using six linear poly (methyl methacrylate) (pMMA) standard samples of
MM equal to 2000, 4000, 8000, 20,000, 50,000, and 100,000 g·mol−1 .

3.1.1. Molar Masses and Molar Mass Distributions
As can be seen in Table 1, the number-average MM (Mn ) values of the final copolymers vary
between 8550 and 11,000 g·mol−1 , which are within the desirable range for obtaining a clear sol-gel
transition. When the experimental Mn values of the final copolymers and their precursors are compared
to the theoretical ones, they are slightly higher. This is ascribed to (i) the calibration curve being based
on PMMA standard samples, and (ii) the partial deactivation of the initiator, MTS, i.e., an amount of
the initiator molecules is terminated by the presence of humidity or any other protic impurities. This is
consistent with other studies on polymers synthesized via GTP [18–22,24].
In Table 1, the Ð values are also given, which are satisfactorily close to the ideal value of unity
(varying between 1.07 and 1.17). This confirms the successful “living” GTP, similar to previously
reported studies [18–22,24]. Concerning the Ð values of the final copolymers, it can be generally
observed that lower Ð values were obtained for the BC diblock copolymers; these copolymers
are the BuMA26 -b-DMAEMA28 (Polymer 8) and BuMA20 -b-DMAEMA34 (Polymer 9). This can be
attributed to the absence of the TEGMA units, which are macromonomers with average MM, thus
meaning that they have wider MMD compared to the BuMA and DMAEMA monomers with
well-defined structure and MM. This is supported by other GTP studies in which a PEGMA-based
macromonomer has been incorporated into the polymer structure [22,25]. Also, lower Ð values are
observed in the case of the statistical copolymer TEGMA9 -co-BuMA20 -co-DMAEMA21 (Polymer 7)
and the (TEGMA9 -co-BuMA20 )-b-DMAEMA21 (Polymer 4). To produce the statistical copolymer,
Polymer 7, all monomers were added to the flask prior to the initiator and the flask was cooled down
in a room temperature water bath to avoid the flask overheating because the monomers were not
added dropwise. This resulted in a more controlled polymerization and a synthesized polymer with
41

Polymers 2017, 9, 31

a narrower MMD was produced unlike previous studies where the flask was not cooled down and
the statistical copolymer had a broader Ð [19,22]. Polymer 4 (TEGMA9 -co-BuMA20 )-b-DMAEMA21
also had a very narrow MMD probably due to the first block being a statistical block that contained
TEGMA and its simultaneous polymerization with BuMA assisted a controlled polymerization.
Figure 3 shows the GPC traces of the TEGMA9 -b-BuMA20 -b-DMAEMA21 (Polymer 1) before and
after precipitation, shown in green solid and dashed line, respectively, and its precursors (TEGMA9 and
TEGMA9 -b-BuMA20 , colored in blue and orange, respectively). Figure 3 confirms the successful
sequential GTP since the peak appears at higher MM as the polymerization progresses (from the
homopolymer to the diblock to the triblock). By precipitating, the lower polymer chains are lost
in the precipitation solvent, thus the MM shifts at higher MM. The low intensity shoulder at lower
MM corresponds to the TEGMA homopolymer, which slightly prevented further polymerization.
Also, it is worth-noting that no peak related to the monomers was observed, thus indicating
complete consumption and 100% conversion of the monomers to the polymer. These observations
are the same for all the copolymers, the GPC traces of which can be found in Figure S1 in the
Supplementary Materials.

Normalised RI
Signal

1
0.8
0.6
0.4
0.2
0
500

10500

20500

30500

40500

Molar Mass (g mol-1)
TEGMA9
TEGMA9-b-BuMA20
TEGMA9-b-BuMA20-b-DMAEMA21 (before precipitation)
TEGMA9-b-BuMA20-b-DMAEMA21 (after precipitation)

Figure 3. Gel permeation chromatography (GPC) traces of Polymer 1 (TEGMA9 -b-BuMA20 -bDMAEMA21 ) and its precursors. The TEGMA homopolymer (TEGMA9 ), the diblock (TEGMA9 -bBuMA20 ), and the triblock copolymer before precipitation are shown in blue solid, orange dotted
and green solid lines, respectively. The triblock copolymer after precipitation is indicated by green
dashed line.

3.1.2. Compositions
The 1 H-NMR results show good agreement between the theoretical and experimental
compositions of the final copolymers and their precursors; these values are listed in Table 1.
The experimental compositions were calculated by using the integral of three distinctive peaks
belonging to the three different repeated units (see Figure S2 in the Supplementary Materials, which
shows the NMR spectra of Polymer 1 and its precursors). The distinctive peak of the TEGMA unit
is the one at 3.35 ppm and belongs to the three methoxy protons, whereas the one of BuMA appears
at 3.9 ppm and belongs to the two methylene protons of the side-chain closest to the ester. The peak
at 2.25 ppm is the one used for DMAEMA and it corresponds to the six methyl protons next to
the nitrogen.

42

Polymers 2017, 9, 31

3.2. Aqueous Solution Properties
3.2.1. Hydrodynamic Diameters
Table 2 lists the theoretical hydrodynamic diameters, calculated by assuming spherical micelle
and random coil formation by the block and statistical copolymers, respectively. The experimental
hydrodynamic diameters of the copolymers in aqueous solutions at both pH 6 and 7, as obtained by
DLS, are also summarized in Table 2.
Concerning the theoretical calculations, the amphiphilic block copolymers are assumed to form
spherical micelles with the BuMA-based block (either as homopolymer or random copolymer with one
of the hydrophilic units) to form the core of the micelle. The structures of the spherical micelles, as well
as the random coil configuration assumed to be adopted by the block and the statistical copolymers,
respectively, are illustrated schematically in Figure 4. The TEGMA, BuMA, and DMAEMA units are
represented by blue, orange, and green spheres, respectively.
Table 2. Theoretical and experimental hydrodynamic diameters, effective pKa s and cloud points of
1% w/w aqueous copolymer solutions.
Polymer
No.
1
2
3
4
5
6
7
8
9

Hydrodynamic diameter (nm)
Theoretical polymer structure a
Theoretical
TEGMA9 -b-BuMA20 -b-DMAEMA21
TEGMA9 -b-DMAEMA21 -b-BuMA20
BuMA20 -b-TEGMA9 -b-DMAEMA21
(TEGMA9 -co-BuMA20 )-b-DMAEMA21
TEGMA9 -b-(BuMA20 -co-DMAEMA21 )
BuMA20 -b-(TEGMA9 -co-DMAEMA21 )
TEGMA9 -co-BuMA20 -co-DMAEMA21
BuMA26 -b-DMAEMA28
BuMA20 -b-DMAEMA34

b

20.5 c
26.6 d
24.8 d
19.1 e
18.1 f
20.8 d
2.7 g
28.1 c
25.3 c

Experimental ± 0.5
pH = 7

pH = 6

8.7
32.7
18.2
21.0
25.5
28.2
NA
32.7
21.0

6.5
32.7
13.5
24.4
21.0
18.2
15.7
15.7
15.7

Effective
pKa s ± 0.1
6.7
6.8
6.9
6.9
6.2
6.7
6.1
6.5
6.8

Cloud Points ± 2 ◦ C
0% H+

10% H+

29
NA
NA
No CP
32
NA
NA
NA
NA

No CP
No CP
No CP
No CP
No CP
No CP
NA
No CP
No CP

a

TEGMA, BuMA, and DMAEMA stand for tri(ethylene glycol) methyl ether methacrylate, n-butyl methacrylate,
and 2-(dimethylamino) ethyl methacrylate, respectively; b The theoretical hydrodynamic diameters (dh ) were
calculated by taking into account the experimental degrees of polymerization (DP) as resulted from GPC and
1 H-NMR analysis; c The theoretical values of the ABC triblock terpolymer and the BC diblock bipolymers were
calculated based on the following equation: dh = (DPBuMA + 2 × DPDMAEMA ) × 0.254 nm; d For the calculation
of the theoretical hydrodynamic diameters of the ACB and BAC triblock and the B(AC) diblock terpolymers, the
following equation was used: dh = [DPBuMA + 2 × (DPDMAEMA + DPTEGMA )] × 0.254 nm; e The following formula
was used in the case of the (AB)C diblock terpolymer: dh = (DPBuMA + DPTEGMA + 2 × DPDMAEMA ) × 0.254 nm;
f In the case of the A(BC) diblock terpolymer, the theoretical value was calculated according to the following
equation: dh = (DPBuMA + DPDMAEMA + 2 × DPTEGMA ) × 0.254 nm; g A random coil configuration was assumed
to be adopted by the statistical copolymer and therefore the equation for a random coil was used: <dg 2 >1/2 = 2 ×
[2 × 2.20 × (DPTEGMA + DPBuMA + DPDMAEMA )/3]1/2 × 0.154 nm; NA Testing these copolymer solutions was not
feasible because the copolymers were insoluble at this pH.

As can be seen in Table 2, the experimental hydrodynamic diameters of the block terpolymers
depend on the polymer architecture. At both pH values, the experimental sizes of Polymer 1, and 3,
with ABC and BAC architecture, respectively, are smaller than the theoretical ones, as expected and
observed before [20–22]. This is also valid for the solution of Polymer 6 (with B(AC) architecture) at
pH 6, at which the DMAEMA units are more than 50% protonated (as it is discussed in the section on
effective pKa ). This trend can be attributed to the theoretical model assuming that (i) the methacrylate
backbone is fully extended and (ii) the BuMA-based block fully overlaps. In reality, this is not the
case since (i) the hydrophobic block is in the collapsed state on account of its incompatibility with the
aqueous solvent and (ii) the polymer chains might overlap in a bigger extent, thus reducing the size of
the micelles. The opposite effect is observed in the rest of the cases on block terpolymers, i.e., Polymer
2, 4, 5 with ACB, (AB)C, and A(BC) architecture at both pH values and Polymer 6 [B(AC)] at pH 7.
This can be attributed to several factors including (i) the DPBuMA being higher than in the other studies,
or the difference in hydrophilicity between the repeated units being less pronounced, i.e., at pH 7, thus
enhancing the hydrophobic interactions and aggregation, (ii) the core-forming block overlapping in a
43

Polymers 2017, 9, 31

lesser extent, and (iii) the theoretical model of Polymer 4 and 5 assuming that the hydrophilic either
TEGMA or DMAEMA, which are randomly copolymerized with BuMA, respectively, take part in the
core formation, which might not be the case.

Figure 4. Schematic showing the structures of the spherical micelles adopted by the block copolymers
(Polymers 1–6, 8, and 9). The random coil formed by the statistical copolymer (Polymer 7) is also
shown. The TEGMA, BuMA, and DMAEMA units are colored in blue, orange, and green, respectively.

Concerning the diblock bipolymers, Polymer 9, which is the most hydrophilic, shows the
expected trend at both pH values, i.e., the experimental values are smaller than the theoretical
values. Concerning Polymer 8, the expected trend is observed at pH 6 at which the difference in
the hydrophobicity of BuMA and DMAEMA is more pronounced. However, at pH 7, the opposite
trend is observed which can be attributed to the effect of hydrophobicity. This confirms the explanation
given before about the hydrophobicity enhancing the aggregation to some extent.
The experimental value of the statistical copolymer is almost six times higher than the theoretical
one corresponding to random coil configuration. This is consistent with other studies, in which bigger
structures were detected by DLS [18,22]. This can be ascribed to (i) the presence of the lengthy side
chains which might favor aggregation [26], and (ii) the good compatibility of TEGMA and DMAEMA
with water, thus enhancing the interactions with water and favoring the extension of the polymer
chain [18]. In addition, the experimental value of 15.7 nm is closer to the value of 13.5 nm which
corresponds to the length of one fully extended polymer chain [(DPTEGMA + DPBuMA + DPDMAEMA ) ×
44

Polymers 2017, 9, 31

0.254 nm], rather than then size of the random coil which was calculated at 2.7 nm. The experimental
value is slightly bigger than the length of a fully extended polymer chain. This could be attributed to
the fact that the long side chains of TEGMA units have not been taken into account in the calculations.
As opposed to one of our earliest studies [18], in which the size was small enough to ensure no micelle
or aggregate formation, in this case, this cannot be verified by the DLS results; this is consistent with
our latest study [22]. The difference may be attributed to different compositions but nevertheless,
another characterization technique, specifically TEM, was carried out to provide information about the
state of the statistical copolymer in solution.
3.2.2. Transmission Electron Microscopy Images
To confirm the hypothesis that the block copolymers form spherical micelles and provide more
details concerning the state of the statistical copolymer, TEM images were recorded, which are shown
in Figure 5.

Figure 5. Transmission electron microscopy (TEM) images of 1% w/w negatively-stained copolymer
solutions at pH 6. The scale bar is set at 100 nm. The schematics of the polymer structures are also given;
Blue, orange, and green spheres represent the TEGMA, BuMA, and DMAEMA units, respectively.

As mentioned before, the TEM images at 100 nm scale are shown in Figure 5. As a reminder the
pH of the solutions used to prepare the TEM samples was adjusted to 6 to ensure that all polymers
were soluble in water. From the TEM results, it can be confirmed that the block copolymers form
spherical-like micelles, whereas the statistical copolymer does not. The diameter of these spherical
45

Polymers 2017, 9, 31

micelles in most cases is around 20 nm, with the exception of Polymer 6, the micelles of which appear
to be slightly bigger at around 25–30 nm. This is in a good agreement with the DLS data. The only
significant difference is observed for Polymer 1, for which the DLS measurements showed a size of
6.5 nm. At this point, it should be remembered that the hydrodynamic diameters are determined as
the size corresponding to the maximum intensity on the DLS histograms. In the case of Polymer 1,
a bimodal distribution was obtained by DLS, with a second peak at higher values, which confirms
the TEM results. The minor differences between the DLS and TEM results can be attributed to the
different way of performing the experiments, which was previously discussed [27]. Specifically, DLS is
performed in the aqueous state, in which the hydrophobic BuMA units are fully collapsed and the
hydrophilic TEGMA and DMAEMA units are fully expanded. On the other hand, TEM is conducted
in the dry state [27]. Therefore, it is concluded that the TEM images are not representative of the actual
structure of the micelles in solution, since they are recorded in the dry state. However, they give a
good approximation of the micelles size and shape, i.e., the size is close to the one obtained by DLS,
and formation of spherical-like micelles is revealed by the block copolymers. It is also confirmed by
TEM that no micelles are formed by the statistical copolymer, which complements the DLS results.
3.2.3. Effective pKa s

pKa

The effective pKa s of the copolymers, as determined by hydrogen ion titrations, are listed in Table 2
and are shown in Figure 6. In most of the cases, the effective pKa values vary between 6.7 and 6.9, which
are similar to previously reported pKa values on DMAEMA-based polymers [18–22,28–30]. However,
the TEGMA9 -b-(BuMA20 -co-DMAEMA21 ) (Polymer 5), the TEGMA9 -co-BuMA20 -co-DMAEMA21
(Polymer 7), and the BuMA26 -b-DMAEMA28 (Polymer 8) show lower effective pKa s, specifically
6.2, 6.1, and 6.5, respectively. The low pKa of the statistical copolymer is in agreement with other
studies on similar statistical terpolymers studied by our group and it is attributed to its inability to form
micelles and stabilize itself in solution [18]. This was also proven in this study by TEM. Compared to the
previously-studied PEGMA6 -co-BuMA18 -co-DMAEMA19 , with the same target MM and composition,
the TEGMA-based copolymer shows significantly lower pKa value (6.1 versus 6.7) [22]; this is due
to the more hydrophobic TEGMA unit replacing the PEGMA one. The lower pKa of Polymer 8 is
explained upon considering its higher hydrophobic BuMA content, which decreases the dielectric
constant, similar to other studies [19–22,31,32]. Concerning Polymer 5, this is the first systematic study
in which this type of architecture has been studied. It can be concluded that randomly copolymerizing
the pH-responsive unit DMAEMA with the hydrophobic BuMA one makes the DMAEMA units
weaker, based on their steric hindrance and the surrounding hydrophobic environment. Among the
other copolymers, it can be observed that the more exposed the DMAEMA units are (see micelle
structures in Figure 4), the easier the protonation, thus the stronger the base and the higher the pKa
values are.
7
6.9
6.8
6.7
6.6
6.5
6.4
6.3
6.2
6.1
6

Theor. BuMA = 35 % w/w
Theor. BuMA = 45 % w/w

ABC ACB BAC (AB)C A(BC) B(AC) STAT AB

Polymer Architecture
Figure 6. Effect of the polymer architecture and the theoretical hydrophobic BuMA weight percentage
on the pKa .

46

Polymers 2017, 9, 31

3.2.4. Cloud Points
Table 2 lists the CP of 1% w/w copolymer solutions in DI water when the DMAEMA units were
not protonated and were protonated by 10%. At the initial pH (no protonation), only three of the
solutions were homogeneous and thus they were able to be visually tested; specifically, Polymers 1, 4,
and 5. The rest of the polymer solutions were insoluble. The insolubility can be ascribed to (i) their
architecture (Polymers 2, 3, 6, and 7) and (ii) the increased hydrophobicity of Polymers 8 and 9 (absence
of hydrophilic TEGMA and increased hydrophobic content). While Polymer 1 and 5 present a CP
at 29 and 32 ◦ C, respectively, Polymer 4 did not respond to temperature, presumably on account of
its architecture, which increases the hydrophilicity; the TEGMA groups are distributed in the same
block as the hydrophobic BuMA. On the other hand, when the DMAEMA units were protonated by
10%, none of the solutions presents a CP within the temperature range tested. This shows that the
protonation of the DMAEMA units, even by only 10%, increases the hydrophilicity of the structure
and also prevents aggregation of the micelles on account of the electrostatic repulsion between the
positively charged amino groups. The effect of protonation on the thermoresponse of DMAEMA units
is well-documented [33,34].
3.2.5. Visual Gel Points
Phase diagrams were constructed for seven out of nine polymers. Specifically, the diluted
and concentrated solutions of the following polymers in PBS formed homogeneous mixtures
(either solutions or gels) and they were visually tested for thermoresponse. The phase diagrams
of TEGMA9 -b-BuMA20 -b-DMAEMA21 (Polymer 1), BuMA20 -b-TEGMA9 -b-DMAEMA21 (Polymer 3),
(TEGMA9 -co-BuMA20 )-b-DMAEMA21 (Polymer 4), TEGMA9 -b-(BuMA20 -co-DMAEMA21 ) (Polymer 5),
BuMA20 -b-(TEGMA9 -co-DMAEMA21 ) (Polymer 6), TEGMA9 -co-BuMA20 -co-DMAEMA21 (Polymer 7),
and BuMA20 -b-DMAEMA34 (Polymer 9) are shown in Figure 7. The other two polymers,
TEGMA9 -b-DMAEMA21 -b-BuMA20 (Polymer 2) and BuMA26 -b-DMAEMA28 (Polymer 8), were only
soluble at the lowest concentrations and therefore, the construction of meaningful phase diagrams was
not feasible.
All the copolymer solutions at 1% w/w showed a CP within the temperature range tested,
with the exception of Polymer 1. The non-thermoresponse of Polymer 1 can be attributed to the
significantly smaller experimental hydrodynamic diameter of the micelles formed by Polymer 1; its size
is less than half of the size of the micelles formed by the other polymers. While all the copolymer
solutions showed thermoresponse at 70 ◦ C and above, the solution of Polymer 5 interestingly showed
thermoresponse at 42 ◦ C. This demonstrates an architectural effect which, by randomly copolymerizing
the thermoresponsive unit DMAEMA with the hydrophobic BuMA, while keeping the hydrophilic
TEGMA unit as a distinct block, enhances the thermoresponse. This can be explained by considering
the enhanced hydrophobic environment surrounding the thermoresponsive unit. This is in contrast
with the results obtained by testing the solutions in DI water, during which no thermoresponse was
observed. This is due to the ionic strength effect, i.e., screening of the electrostatic repulsions between
the protonated DMAEMA units. The ionic strength effect on the LCST of DMAEMA units has been
previously observed and discussed by De Souza et al. [33].
Since physical gels have interesting applications, including injectable gels and 3-D printing, the
solutions were visually tested for gelation. The gelation region (if there is one), i.e., temperature and
concentration ranges where a stable gel was formed, is approximately shown by the black dashed
line in Figure 7. The points which correspond to stable transparent and cloudy gel are shown in
yellow circles and squares, respectively. Interestingly, a gelation region was observed for five out of six
polymers. Specifically, Polymers 1, 3, 5, 6, and 9, formed a stable gel, whereas the solutions of Polymers
4 and 7 only presented a CP, regardless of the concentration of the solution; the formation of a stable
gel by the most concentrated solution of the statistical terpolymer can be attributed to reaching the
limit of entanglements. The inability of Polymer 4 to form a gel, even in highly concentrated solutions,
demonstrates, once again, an architectural effect. Particularly, by randomly copolymerizing the
47

Polymers 2017, 9, 31

hydrophobic BuMA unit with the hydrophilic TEGMA unit, while polymerizing the thermoresponsive
DMAEMA in a distinct block, enhances the hydrophilicity of the structure, thus interrupting gelation.

Figure 7. Phase diagrams of the TEGMA-based copolymer solutions in phosphate buffered saline
(PBS) of various architectures, namely ABC (Polymer 1), BAC (Polymer 3), (AB)C (Polymer 4), A(BC)
(Polymer 5), B(AC) (Polymer 6), and the statistical terpolymer (Polymer 7). The phase diagram of the
diblock bipolymer BuMA20 -b-DMAEMA34 (Polymer 9) is also shown. The concentrated copolymer
solutions in PBS of the ACB (Polymer 2), and the most hydrophobic diblock bipolymer (Polymer 8)
were not soluble, thus their phase diagrams are not presented. The transparent and slightly cloudy
runny solutions are presented in white and green circles, respectively, whereas the cloudy ones are
indicated by green squares. The transparent and cloudy viscous solutions are shown in orange circles
and squares, respectively. The transparent and cloudy stable gels are indicated by yellow circles and
squares, respectively. Gel syneresis is represented by white squares with diagonal yellow lines, while
the precipitation is shown in black squares. The gelation region is approximately shown with the black
dashed line. The corresponding polymer structures are schematically given, in which the TEGMA,
BuMA, and DMAEMA units are colored in blue, orange, and green, respectively.

The phase diagrams which include a gelation region can be divided into two categories, which
differ in the thermoresponsive behavior; as is shown in Figure 7. In the first category, two copolymers
are included, specifically Polymers 1 and 5, whereas Polymers 3, 6, and 9 belong to the second
category. When copolymer solutions of the first category are concerned, runny solutions are formed
at low temperatures, regardless of the increased concentrations. These solutions form a stable gel
48

Polymers 2017, 9, 31

upon increasing the temperature, which then gets destabilized and turns back to solution when the
temperature increases further. The ones belonging to the second category, show gelation at higher
temperature ranges and the gels formed mostly remain stable until the highest temperature tested.
The gelation temperature decreases by increasing the concentration of the solution; as expected. It is
worth-noting that in these phase diagrams, the highly concentrated solutions formed either a stable
gel or a highly viscous at room temperature. When a stable gel at room temperature was formed,
these gels either remained stable upon heating or gel syneresis was observed at higher temperatures.
In the case of highly viscous solution, a stable gel is formed upon increasing the temperature, followed
by gel syneresis. The gelation at room temperature by highly concentrated polymer solutions can
be ascribed to reaching the limit of entanglements rather than to thermoresponse. An interesting
feature is that a clear region of transparent gel is not obtained by Polymer 9, which does not contain
the hydrophilic TEGMA units and by Polymer 5, in which the thermoresponsive unit is within a
hydrophobic BuMA-based environment. Regarding the rest of the polymers which formed a gel
(Polymers 1, 3, and 6), a transparent gel was formed which was then transformed into a cloudy one at
higher concentrations; this has been previously observed for triblock copolymers containing similar
monomers [22].
Therefore, in this study, the effect of architecture on the thermoresponsive behavior is clearly
demonstrated. Comparing the behavior of the three triblock terpolymers, the ABC architecture shows
clear sol-gel transition, while the ACB one is insoluble and the BAC one shows gelation at higher
temperatures, thus confirming the trends previously observed [18,20]. Compared to our previous
study in which no gelation was observed for the ACB architecture, the gelation in the present study can
be attributed to the presence of TEGMA units, instead of PEGMA units (around five repeat EG groups),
which increases the hydrophobicity of the structure, thus enhancing the thermoresponse. Concerning
the three diblock terpolymers, which have been systematically studied for the first time, the A(BC)
architecture shows clear sol-gel transition, whereas the (AB)C one does not form gel, and the B(AC)
one forms gel at higher temperatures. On the other hand, no gelation was observed for the statistical
terpolymer, which is consistent with our previous studies [18,20]. Interestingly, gelation is observed
at high temperatures by the AB diblock copolymer which did not show solubility issues, specifically
Polymer 9, as opposed to the solubility issues of Polymer 8. In conclusion, two architectures, namely
ABC with the hydrophobic BuMA forming the central block and A(BC) in which the thermoresponsive
DMAEMA has been randomly copolymerized with the hydrophobic BuMA, show clear sol-gel-sol
transition close to the desirable, body temperature values. Among the two, the ABC architecture is
most promising since it shows wider gelation region, i.e., a stable gel is formed at lower temperatures
and concentration and this gel gets destabilized at higher temperatures.
4. Conclusions
In this study, seven well-defined terpolymers of different architectures were successfully synthesized
via GTP. The hydrophilic TEGMA, the hydrophobic BuMA, and the thermoresponsive DMAEMA were
used as the A, B, and C units, respectively. The architecture of the polymers was systematically varied in
order to investigate its effect on the thermoresponsive behavior of the copolymers. Three architectures,
namely ABC, ACB, BAC, and statistical, that have been previously studied, as well as three novel
architectures, specifically (AB)C, A(BC), and B(AC) were investigated and compared. Two BC diblock
bipolymers mimicking the BuMA:DMAEMA and BuMA:(TEGMA + DMAEMA) ratios of the weight
percentages in the terpolymers were also synthesized. All block copolymers formed spherical micelles
unlike the statistical copolymer. Interestingly, the block architecture and the position of the hydrophobic
groups strongly influence the thermoresponsive behavior with the ABC and A(BC) architectures to
show the desirable, clear sol-gel transition close to body temperature.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/9/1/31/s1.
Figure S1 shows the GPC chromatograms of the triblock terpolymers (Polymers 1–3), the diblock terpolymers
(Polymers 4–6), the statistical terpolymer (Polymer 7), the diblock bipolymers (Polymers 8 and 9), and their linear

49

Polymers 2017, 9, 31

precursors. The GPC traces of the first block and the diblock are indicated by blue solid and orange dotted lines,
whereas the final triblock terpolymer (if there is any) is shown by green solid line. The GPC chromatogram of the
statistical terpolymer is colored in black. Green dashed line represents the GPC traces of the final terpolymers after
precipitation; Figure S2 shows the 1 H NMR spectra of Polymer 1 before precipitation and its linear precursors:
(a) TEGMA9 , (b) TEGMA9 -b-BuMA20 , and (c) TEGMA9 -b-BuMA20 -b-DMAEMA21 are colored in blue, orange,
and green, respectively.
Acknowledgments: The Department of Materials at Imperial College London is acknowledged for funding a
Ph.D. scholarship for Anna P. Constantinou.
Author Contributions: Author Contributions: Anna P. Constantinou performed most of the experiments,
analyzed the data, and guided Hanyi Zhao, who assisted with some of the aqueous characterization;
Anna P. Constantinou wrote the paper; Catriona M. McGilvery performed the TEM experiments and analysis;
Alexandra E. Porter assisted with the TEM analysis and provided the materials for TEM sample preparation;
Theoni K. Georgiou conceived the idea, provided the materials, designed the experiments and supervised
the work.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.
4.
5.

6.
7.
8.

9.

10.

11.

12.

13.

14.
15.

Constantinou, A.P.; Georgiou, T.K. Tuning the Gelation of Thermoresponsive Gels. Eur. Polym. J. 2016, 78,
366–375. [CrossRef]
Ward, M.A.; Georgiou, T.K. Thermoresponsive Polymers for Biomedical Applications. Polymers 2011, 3,
1215–1242. [CrossRef]
Chassenieux, C.; Tsitsilianis, C. Recent Trends in pH/Thermo-Responsive Self-Assembling Hydrogels:
From Polyions to Peptide-Based Polymeric Gelators. Soft Matter 2016, 12, 1344–1359. [CrossRef] [PubMed]
Southall, N.T.; Dill, K.A.; Haymet, A.D.J. A View of the Hydrophobic Effect. J. Phys. Chem. B 2002, 106,
521–533. [CrossRef]
Madsen, J.; Armes, S.P.; Bertal, K.; Lomas, H.; MacNeil, S.; Lewis, A.L. Biocompatible Wound Dressings
Based on Chemically Degradable Triblock Copolymer Hydrogels. Biomacromolecules 2008, 9, 2265–2275.
[CrossRef] [PubMed]
Place, E.S.; George, J.H.; Williams, C.K.; Stevens, M.M. Synthetic Polymer Scaffolds for Tissue Engineering.
Chem. Soc. Rev. 2009, 38, 1139–1151. [CrossRef] [PubMed]
Elstad, N.L.; Fowers, K.D. OncoGel (ReGel/Paclitaxel)—Clinical Applications for a Novel Paclitaxel Delivery
System. Adv. Drug Deliv. Rev. 2009, 61, 785–794. [CrossRef] [PubMed]
Zentner, G.M.; Rathi, R.; Shih, C.; McRea, J.C.; Seo, M.-H.; Oh, H.; Rhee, B.G.; Mestecky, J.; Moldoveanu, Z.;
Morgan, M.; et al. Biodegradable Block Copolymers for Delivery of Proteins and Water-Insoluble Drugs.
J. Control. Release 2001, 72, 203–215. [CrossRef]
Hsieh, F.-Y.; Lin, H.-Y.; Hsu, S.-H. 3D Bioprinting of Neural Stem Cell-Laden Thermoresponsive
Biodegradable Polyurethane Hydrogel and Potential in Central Nervous System Repair. Biomaterials 2015,
71, 48–57. [CrossRef] [PubMed]
Tsai, Y.-C.; Li, S.; Hu, S.-G.; Chang, W.-C.; Jeng, U.-S.; Hsu, S.-H. Synthesis of Thermoresponsive Amphiphilic
Polyurethane Gel as a New Cell Printing Material near Body Temperature. ACS Appl. Mater. Interfaces 2015,
7, 27613–27623. [CrossRef] [PubMed]
Zhang, M.; Vora, A.; Han, W.; Wojtecki, R.J.; Maune, H.; Le, A.B.A.; Thompson, L.E.; McClelland, G.M.;
Ribet, F.; Engler, A.C.; et al. Dual-Responsive Hydrogels for Direct-Write 3D Printing. Macromolecules 2015,
48, 6482–6488. [CrossRef]
Kolesky, D.B.; Truby, R.L.; Gladman, A.S.; Busbee, T.A.; Homan, K.A.; Lewis, J.A. 3D Bioprinting of
Vascularized, Heterogeneous Cell-Laden Tissue Constructs. Adv. Mater. 2014, 26, 3124–3130. [CrossRef]
[PubMed]
Buwalda, S.J.; Boere, K.W.M.; Dijkstra, P.J.; Feijen, J.; Vermonden, T.; Hennink, W.E. Hydrogels in a Historical
Perspective: From Simple Networks to Smart Materials. J. Control. Release 2014, 190, 254–273. [CrossRef]
[PubMed]
Skardal, A.; Atala, A. Biomaterials for Integration with 3-D Bioprinting. Ann. Biomed. Eng. 2015, 43, 730–746.
[CrossRef] [PubMed]
Davis, F.J. Polymer Chemistry: A Practical Approach; Oxford University Press: Oxford, UK, 2004.

50

Polymers 2017, 9, 31

16.

17.
18.
19.
20.
21.
22.
23.
24.

25.
26.

27.
28.

29.

30.
31.

32.
33.
34.

Webster, O.W.; Hertler, W.R.; Sogah, D.Y.; Farnham, W.B.; RajanBabu, T.V. Group-Transfer Polymerization.
1. A New Concept for Addition Polymerization with Organosilicon Initiators. J. Am. Chem. Soc. 1983, 105,
5706–5708. [CrossRef]
Hatada, K.; Kitayama, T.; Vogl, O. Macromolecular Desing of Polymeric Materials; Marcel Dekker, Inc.: New York,
NY, USA, 1997; p. 878.
Ward, M.A.; Georgiou, T.K. Multicompartment Thermoresponsive Gels: Does the Length of the Hydrophobic
Side Group Matter? Polym. Chem. 2013, 4, 1893–1902. [CrossRef]
Ward, M.A.; Georgiou, T.K. Thermoresponsive Gels Based on ABA Triblock Copolymers: Does the
Asymmetry Matter? J. Polym. Sci. A 2013, 51, 2850–2859. [CrossRef]
Ward, M.A.; Georgiou, T.K. Thermoresponsive Terpolymers Based on Methacrylate Monomers: Effect of
Architecture and Composition. J. Polym. Sci. A 2010, 48, 775–783. [CrossRef]
Ward, M.A.; Georgiou, T.K. Thermoresponsive Triblock Copolymers Based on Methacrylate Monomers:
Effect of Molecular Weight and Composition. Soft Matter 2012, 8, 2737–2745. [CrossRef]
Constantinou, A.P.; Georgiou, T.K. Thermoresponsive Gels Based on ABC Triblock Copolymers: Effect of the
Length of the PEG Side Group. Polym. Chem. 2016, 7, 2045–2056. [CrossRef]
Dicker, I.B.; Cohen, G.M.; Farnham, W.B.; Hertler, W.R.; Laganis, E.D.; Sogah, D.Y. Oxyanions Catalyze
Group-Transfer Polymerization to Give Living Polymers. Macromolecules 1990, 23, 4034–4041. [CrossRef]
Patrickios, C.S.; Lowe, A.B.; Armes, S.P.; Billingham, N.C. ABC Triblock Polymethacrylates: Group Transfer
Polymerization Synthesis of the ABC, ACB, and BAC Topological Isomers and Solution Characterization.
J. Polym. Sci. A 1998, 36, 617–631. [CrossRef]
Raduan, N.H.; Horozov, T.S.; Georgiou, T.K. “Comb-Like” Non-Ionic Polymeric Macrosurfactants. Soft Matter
2010, 6, 2321–2329. [CrossRef]
Hadjiyannakou, S.C.; Vamvakaki, M.; Patrickios, C.S. Synthesis, Characterization and Evaluation of
Amphiphilic Diblock Copolymer Emulsifiers Based on Methoxy Hexa(Ethylene Glycol) Methacrylate and
Benzyl Methacrylate. Polymer 2004, 45, 3681–3692. [CrossRef]
Wu, C.; Ying, A.; Ren, S. Fabrication of Polymeric Micelles with Core-Shell-Corona Structure for Applications
in Controlled Drug Release. Colloid Polym. Sci. 2013, 291, 827–834. [CrossRef]
Georgiou, T.K.; Vamvakaki, M.; Patrickios, C.S.; Yamasaki, E.N.; Phylactou, L.A. Nanoscopic Cationic
Methacrylate Star Homopolymers: Synthesis by Group Transfer Polymerization, Characterization and
Evaluation as Transfection Reagents. Biomacromolecules 2004, 5, 2221–2229. [CrossRef] [PubMed]
Georgiou, T.K.; Phylactou, L.A.; Patrickios, C.S. Synthesis, Characterization, and Evaluation as Transfection
Reagents of Ampholytic Star Copolymers: Effect of Star Architecture. Biomacromolecules 2006, 7, 3505–3512.
[CrossRef] [PubMed]
Simmons, M.R.; Patrickios, C.S. Synthesis and Aqueous Solution Characterization of Catalytically Active
Block Copolymers Containing Imidazole. Macromolecules 1998, 31, 9075–9077. [CrossRef]
Emileh, A.; Vasheghani-Farahani, E.; Imani, M. Swelling Behavior, Mechanical Properties and Network
Parameters of pH- and Temperature-Sensitive Hydrogels of Poly((2-Dimethyl Amino) Ethyl Methacrylate-coButyl Methacrylate). Eur. Polym. J. 2007, 43, 1986–1995. [CrossRef]
Philippova, O.E.; Hourdet, D.; Audebert, R.; Khokhlov, A.R. pH-Responsive Gels of Hydrophobically
Modified Poly(Acrylic Acid). Macromolecules 1997, 30, 8278–8285. [CrossRef]
De Souza, J.C.P.; Naves, A.F.; Florenzano, F.H. Specific Thermoresponsiveness of PMMA-Block-PDMAEMA
to Selected Ions and Other Factors in Aqueous Solution. Colloid Polym. Sci. 2012, 290, 1285–1291. [CrossRef]
Fournier, D.; Hoogenboom, R.; Thijs, H.M.L.; Paulus, R.M.; Schubert, U.S. Tunable pH- and TemperatureSensitive Copolymer Libraries by Reversible Addition-Fragmentation Chain Transfer Copolymerizations of
Methacrylates. Macromolecules 2007, 40, 915–920. [CrossRef]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

51

polymers
Article

Semi-Interpenetrating Polymer Networks
with Predefined Architecture for Metal Ion
Fluorescence Monitoring
Kyriakos Christodoulou 1 , Epameinondas Leontidis 2 , Mariliz Achilleos 1 , Christiana Polydorou 1
and Theodora Krasia-Christoforou 1, *
1

2

*

Department of Mechanical and Manufacturing Engineering, University of Cyprus, 1678 Nicosia, Cyprus;
christodoulou.c.kyriakos1@ucy.ac.cy (K.C.); achilleos.mariliz@ucy.ac.cy (M.A.);
polydorou.christiana@ucy.ac.cy (C.P.)
Department of Chemistry, University of Cyprus, 1678 Nicosia, Cyprus; psleon@ucy.ac.cy
Correspondence: krasia@ucy.ac.cy; Tel.: +357-2289-2288

Academic Editor: Alexander Böker
Received: 26 September 2016; Accepted: 23 November 2016; Published: 29 November 2016

Abstract: The development of new synthetic approaches for the preparation of efficient 3D luminescent
chemosensors for transition metal ions receives considerable attention nowadays, owing to the key role
of the latter as elements in biological systems and their harmful environmental effects when present in
aquatic media. In this work, we describe an easy and versatile synthetic methodology that leads to the
generation of nonconjugated 3D luminescent semi-interpenetrating amphiphilic networks (semi-IPN)
with structure-defined characteristics. More precisely, the synthesis involves the encapsulation of
well-defined poly(9-anthrylmethyl methacrylate) (pAnMMA) (hydrophobic, luminescent) linear
polymer chains within a covalent poly(2-(dimethylamino)ethyl methacrylate) (pDMAEMA) hydrophilic
polymer network, derived via the 1,2-bis-(2-iodoethoxy)ethane (BIEE)-induced crosslinking process of
well-defined pDMAEMA linear chains. Characterization of their fluorescence properties demonstrated
that these materials act as strong blue emitters when exposed to UV irradiation. This, combined with
the presence of the metal-binding tertiary amino functionalities of the pDMAEMA segments, allowed
for their applicability as sorbents and fluorescence chemosensors for transition metal ions (Fe3+ , Cu2+ )
in solution via a chelation-enhanced fluorescence-quenching effect promoted within the semi-IPN
network architecture. Ethylenediaminetetraacetic acid (EDTA)-induced metal ion desorption and
thus material recyclability has been also demonstrated.
Keywords: semi-interpenetrating networks; fluorescent networks; anthracene; metal ions;
fluorescence sensors

1. Introduction
The development of luminescent chemosensors for transition metal ions has attracted considerable
attention in the last years given their high importance in biological systems [1], but also their harmful
environmental effects when they are present in high concentrations in aquatic media [2,3]. To ensure
high efficiency in a fluorescent metal-ion sensor, the light-emitting moiety should be covalently linked
to a metal ion-binding group, and its fluorescence properties must be sensitive to the ion–ligand
interaction [4,5].
Most literature examples focusing on 3D luminescent materials that are used as sensor platforms
deal with metal–organic frameworks (MOFs) [6–15] and 3D coordination polymers [16–20]. Only a
few research groups have been working on the synthesis of fluorescent semi-interpenetrating network
(semi-IPN) architectures. The latter are 3D polymer structures consisting of secondary linear polymer

Polymers 2016, 8, 411

52

www.mdpi.com/journal/polymers

Polymers 2016, 8, 411

chains that are interlaced—but not covalently bonded—with a primary polymer network [21–24].
In this limited number of existing reports, the fluorescent component is either a conjugated conductive
polymer [25–28], fluorescent nanoparticles such as carbon nanodots and quantum dots (QDs) [29–31],
or fluorescent dyes dispersed within or covalently linked to the polymer matrix [32–35]. However,
the abovementioned materials have several disadvantages including: (a) low structural stability and
robustness in the presence of chemical and physical impacts [36–40]; (b) possible leaching and relatively
high probability of aggregation-induced fluorescence quenching of the physically entrapped organic
dyes [41]; and (c) agglomeration phenomena of the fluorescent nanoparticles such as carbon nanodots
resulting in inferior fluorescence properties due to self-quenching [42].
The necessity of developing functional organic polymer networks with controllable architectures
have directed polymer chemists to explore new synthetic routes towards this purpose. Such materials
are considered to be highly advantageous in comparison to their analogues prepared by a noncontrolled
chemical crosslinking process, since the latter are usually highly inhomogeneous owing to the
non-precise molar mass and broad molar mass distributions of the polymer segments constituting
the networks. Consequently, polymer networks characterized by ill-defined architectures usually
exhibit inferior mechanical and swelling properties, whereas their structural and compositional
inhomogeneities restrict the structure-to-property correlation [43].
Although numerous examples on new synthetic approaches resulting in the generation of
“model” or “quasi-model” covalent polymer networks have appeared so far, including “quasi-living”
carbocationic polymerization [44,45], anionic polymerization [46,47], group transfer polymerization
(GTP) [48–50], and controlled radical polymerization processes [51–54], there is only a limited number
of publications discussing the synthesis of semi-IPN exhibiting structure-defined characteristics [55,56].
Very recently, our group has reported on the synthesis of such materials [57]. These consisted of
well-defined hydrophilic and pH-responsive (poly(2-dimethylamino) ethyl methacrylate) (pDMAEMA)
linear chains that were interconnected using 1,2-bis-(2-iodoethoxy)-ethane (BIEE) generating the
pDMAEMA network, and well-defined hydrophobic poly(n-butyl methacrylate) (pBuMA) linear chains
that were encapsulated within the network during the crosslinking process. From our studies, it has
been demonstrated that the mechanical properties of these materials can be easily tuned by adjusting
the content of the encapsulated hydrophobic linear chains, whereas their well-defined structural
characteristics allowed for the prediction of their mechanical response via mathematical modeling.
Giving further credence to the BIEE-crosslinking approach as an alternative to synthetically
demanding and multistep controlled polymerization processes, in the present study we report on
the synthesis of 3D structure-defined emissive (fluorescent) amphiphilic semi-IPN, consisting of
BIEE-crosslinked pDMAEMA segments and embedded hydrophobic and nonconjugated/fluorescent
poly(9-anthrylmethyl methacrylate) (pAnMMA) linear chains both prepared by reversible
addition–fragmentation chain transfer (RAFT)-controlled radical polymerization. Although in this
case the coordinating active part (pDMAEMA) [58–63] is not covalently bound to the fluorescent active
component (pAnMMA), the semi-IPN network architecture promotes the reinforcement of interactions
between the pDMAEMA-complexed transition metal ions and the anthracene moieties of the interlaced
pAnMMA chains, thus promoting chelation-enhanced fluorescence quenching. Moreover, interlacing
of the pAnMMA chains within the nonfluorescent pDMAEMA network diminishes phase separation
phenomena (and, consequently, self-quenching effects). Furthermore, the high molar mass and
hydrophobicity of the macromolecular fluorescent pAnMMA prevents the leaching of the active
component in polar and aqueous solvents.
These materials were further evaluated in the chemosensing of Fe3+ and Cu2+ transition metal
ions. The former is a key element in biological and environmental systems, playing an essential
role in oxygen uptake and metabolism and electron transfer processes [64]. Cu2+ ions, which are
known to exhibit high affinity for N– and O–containing ligands, significantly contribute to the metal
environmental pollution owing to their widespread industrial use. Even though the toxicity of Cu2+

53

Polymers 2016, 8, 411

ions is considerably lower compared to other heavy metal ions, very low Cu2+ concentrations are
highly toxic to certain microorganisms [65].
Consequently, the presented versatile synthetic approach creates new prospects in the generation
of nonconjugated 3D luminescent polymer-based sensors in which fluorescent moieties are combined
with metal-chelating elements, characterized by structure-defined characteristics and tunable
properties, with potential use in metal ion chemosensing.
2. Materials and Methods
2.1. Chemical Reagents
Poly(9-Anthrylmethyl methacrylate) (pAnMMA) (Mn = 27,900 g·mol−1 ; where Mn : number
average molar mass determined by size exclusion chromatography using poly(methyl methacrylate)
(PMMA) calibration standards; polydispersity index, PDI: 1.3) and poly(2-(dimethylamino)ethyl
methacrylate) (pDMAEMA) (Mn : 19,000 g·mol−1 , polydispersity index, PDI: 1.17), were in-house
synthesized by RAFT-controlled radical polymerization according to our previous publications [5].
1,2-bis-(2-iodothoxy)ethane (BIEE, Sigma-Aldrich, 96%, St. Louis, MO, USA), 9-anthracenemethanol
(Sigma-Aldrich, 97%, St. Louis, MO, USA), methanol (MeOH, Sharlau, analytical grade, ACS
reagent, Barcelona, Spain), and tetrahydrofuran (THF, Scharlau, HPLC grade, Barcelona, Spain)
were used as received. FeCl3 ·6H2 O (Sigma-Aldrich, ≥99%, St. Louis, MO, USA), Cu(CH3 COO)2 ·H2 O
(Sigma-Aldrich, ≥99%, St. Louis, MO, USA) and ethylenediaminetetraacetic acid (EDTA) (Sharlau,
99%–101%, Barcelona, Spain) were used as received by the supplier.
2.2. Synthesis of Semi-Interpenetrating BIEE-Crosslinked pAnMMA/pDMAEMA Networks
By following a similar synthetic protocol as that described in a recent publication of our group [57],
a semi-interpenetrating fluorescent polymer network of the type p(AnMMA)/pDMAEMA/BIEE was
prepared via the encapsulation of pAnMMA (5 wt % in respect to the total polymer mass) within
the BIEE-crosslinked pDMAEMA network. The experimental procedure is as follows. In a glass vial,
pAnMMA (5 mg) was dissolved in THF (1.25 mL, 8% w/v solution concentration). To the solution,
pDMAEMA (100 mg, 0.0052 mmol of macro-chain transfer agent, 0.64 mmol per DMAEMA unit) was
added and the mixture was stirred rapidly until complete dissolution of pDMAEMA. To the solution,
BIEE (58 μL, 118 mg, 0.32 mmol) was added using a micropipette. The resulting solution was stirred
rapidly and was then left in a sealed vial at room temperature, under air and without stirring, until
gelation was observed (7 days). The resulting network was then placed in excess methanol (100 mL) for
one week to remove the sol fraction. The sol fraction (14%) was determined gravimetrically, and it was
calculated from the ratio of the dried mass of the extractables to the theoretical mass of all components
in the network (i.e., pAnMMA, pDMAEMA, and BIEE).
2.3. Swelling Behavior
The methanol-swollen network was cut into small pieces and their mass was determined
gravimetrically before placing them in a vacuum oven to dry at ca. 25 ◦ C for 24 h. The mass of
the dry network pieces was then determined, and the degrees of swelling (DSs) in MeOH were
calculated as the ratio of the swollen mass divided by the dry mass. Subsequently, the dried pieces
were placed in deionized water for 2 weeks before determining the water-swollen network masses.
2.4. Fluorescent Characterization
The fluorescence emission spectrum of pre-swollen (in methanol) polymer network was recorded
at the solid state by using a Jasco FP-6300 fluorescence spectrophotometer (Jasco Incorporated, Easton,
MD, USA). The excitation wavelength was set at 370 nm, where AnMMA-containing polymers are
known to exhibit maximum absorption [66]. Fluorescence microscopy was further used for visualizing
the anthracene-containing fluorescent network at its swollen state (in methanol). The swollen network

54

Polymers 2016, 8, 411

was placed in methanol-containing Petri dishes and examined under the Olympus fluorescence
microscope (BX53 System Microscope, Olympus Corporation, Tokyo, Japan). The fluorescence intensity
of the sample was determined by using the DAPI filter (U-FUNA, excitation: 358 nm, emission: 461 nm;
Olympus, Tokyo, Japan). Other filters were used with different excitation and emission spectra, but no
fluorescence was detected. Images were taken at 4× magnification and analyzed using the CellSens
software. The same exposure time was used in all cases.
2.5. Fluorescence Monitoring of Cu2+ and Fe3+ Ions
Initially, Fe3+ and Cu2+ metal ion solutions of various concentrations were prepared by dissolving
FeCl3 ·6H2 O and Cu(CH3 COO)2 ·H2 O metal ion salts in methanol. More precisely, Fe3+ and Cu2+
methanol solutions were prepared, with concentrations in the range 5 × 10−6 to 5 × 10−5 M and
5 × 10−5 to 10−3 M, respectively.
The polymer network was cut in small pieces (m ~0.70 g), which were subsequently immersed
in glass vials containing the metal ion (Fe3+ and Cu2+ ) solutions (5 mL) for 24 h. For comparison
purposes, a control sample was also prepared upon immersing the polymer network in a glass vial
containing pure methanol (5 mL). Afterwards, the pieces were washed twice with pure methanol and
placed in new vials containing methanol (5 mL) prior to their characterization.
For comparison purposes, the fluorescence monitoring of Fe3+ and Cu2+ in methanol
(concentration range 0.1–0.25 mM (Fe3+ ) and 0.1–1.0 mM (Cu2+ )) was carried out by using the low
molar mass 9-anthracenemethanol (1.25 mM) as the metal ion fluorescent chemosensor.
2.6. Desorption Studies—Polymer Network Regeneration
Initially, Cu2+ -loaded polymer networks were prepared by immersing the methanol-swollen
p(AnMMA)/pDMAEMA/BIEE networks in Cu2+ ion methanol solutions of two different Cu2+
concentrations (10−3 and 5 × 10−3 M) for 24 h followed by extensive washing to remove any unbound
cations. For the regeneration process, the Cu2+ -loaded polymer network pieces (m ~0.35 g) were
immersed in aqueous EDTA solution (5 mL, 0.25 M) and they were visualized by fluorescence
microscopy in real time.
3. Results and Discussion
3.1. Synthesis of Semi-Interpenetrating 3D Amphiphilic Fluorescent Networks
The synthetic methodology followed for the preparation of structure-defined, BIEE-crosslinked
fluorescent semi-IPN polymer networks was based on our recent publication [57]. Initially, pDMAEMA
and fluorescent poly(AnMMA) linear homopolymers were prepared by RAFT-controlled radical
polymerization. Unimodal polymers with controlled average molar mass (MWs) and relatively low
PDIs were obtained in both cases [5,57]. The well-defined poly(DMAEMA) and poly(AnMMA) linear
precursors were then dissolved in tetrahydrofuran (i.e., a good solvent for both homopolymers to
ensure good intermixing at a molecular level) and employed as precursors for the generation of the
BIEE/pDMAEMA/pAnMMA semi-IPN polymer networks, as schematically demonstrated in Figure 1.
The gelation process that was carried out at room temperature, under air and without mechanical
stirring, was completed within 7 days. As already described in Section 2.2, the [BIEE]/[DMAEMA]
molar ratio was fixed at 1:2, targeting a 100% degree of crosslinking, since 1 BIEE molecule is capable
of linking together 2 tertiary amino functionalities. However, according to Armes and co-workers [67],
intrachain crosslinking or reaction of one iodide group of the BIEE crosslinker may occur as a result of
deviation from quantitative crosslinking. More precisely, quaternization of the PDMAEMA chains
with BIEE may lead to either intermolecular crosslinking (branching) or intramolecular cyclization,
as shown in Figure 1b. The sol fraction percentage (extractables) that was determined gravimetrically
was 14%, which is in line with that reported in our previous study [57].

55

Polymers 2016, 8, 411

The networks’ swelling behavior is considered to be an important influencing parameter on
their performance in metal ion uptake and sensing. The swelling behavior of the BIEE-crosslinked
p(AnMMA)/pDMAEMA networks was investigated in both, water, and methanol. Based on the
experimental data, the degree of swelling determined in water (19.1 ± 1.7) was comparable to that
found in methanol (16.3 ± 0.8), suggesting that shifting from methanol to water would probably not
lead to significant changes on the network’s sensing performance.

Figure 1.
(a) Chemical structures of the linear poly(2-(dimethylamino)ethyl methacrylate)
(pDMAEMA) and poly(9-anthrylmethyl methacrylate) (pAnMMA) homopolymers and of the
1,2-bis-(2-iodoethoxy)ethane (BIEE) crosslinking agent, and schematic of the one-step synthetic
approach followed for the preparation of 3D luminescent semi-interpenetrating amphiphilic network
(APN); (b) crosslinking reaction scheme presenting both, intra- and interchain crosslinking pathways
that may occur between BIEE and –N(Me)3 pendant moieties of the poly(DMAEMA) linear chains.

3.2. Fluorescence Properties
The fluorescence emission of the network pre-swollen in methanol was investigated by means of
fluorescence spectroscopy and microscopy. As previously described in the experimental section,
a BIEE/pDMAEMA/pAnMMA semi-IPN polymer network was synthesized, in which 5 wt %
of the fluorescent pAnMMA component with respect to the total mass (polymers + crosslinking
agent) was incorporated. As seen in Figure 2, the resulting network displayed strong blue emission
(recorded at 461 nm) under 358 nm excitation wavelength. Moreover, the active fluorescent polymeric
component was homogeneously distributed within the pDMAEMA polymer matrix, as verified by
fluorescence microscopy.
Fluorescence spectroscopy was used for recording the fluorescence emission spectrum of the
methanol-swollen network. The vibronic structure of the network provided in Figure 3 resembled that
of the anthracene fluorophore [66,68] and of the previously reported pAnMMA linear homopolymer
analogue [5].

56

Polymers 2016, 8, 411

Figure 2. (a) Photograph of the BIEE/pDMAEMA/pAnMMA semi-interpenetrating network
(semi-IPN) polymer network when exposed to UV irradiation, displaying strong blue emission;
(b) corresponding fluorescence microscopy image. The scale bar in Figure 2b is 100 μm.
600

Fluorescence Intensity

500
400
300
200
100
0
400

450

500

550

600

650

wavelength (nm)

Figure 3. Fluorescence spectrum of the BIEE/pDMAEMA/pAnMMA semi-IPN polymer network
pre-swollen in methanol. Excitation wavelength: 370 nm.

3.3. Fluorescence Monitoring of Cu2+ and Fe3+ Ions
The BIEE/pDMAEMA/pAnMMA semi-IPN polymer network system was examined regarding
its ability to act as a macromolecular sensor for transition metal ions. More precisely, the study
was focused on examining the chemosensing ability of such fluorescent 3D semi-IPN towards Fe3+
and Cu2+ ions dissolved in methanol. Both fluorescence microscopy and spectroscopy were used
to monitor the qualitative and quantitative changes in the network’s fluorescence intensity when
the latter was immersed in metal ion solutions of various concentrations prepared in methanol.
The metal ion fluorescence monitoring studies were purposely performed in methanol, since this
allowed for a direct comparison of the obtained fluorescence data with those acquired when using
9-anthracenemethanol (model compound) as a fluorophore (see Supplementary Materials), which
is insoluble in water. Moreover, based on previous literature reports [69–73], the use of methanol or
methanol/water mixtures of various volume ratios is typical in metal ion fluorescence monitoring
studies, since methanol—being a protic and very polar solvent—exhibits properties similar to water.
In one such example [73], the authors performed selective fluorescence quenching experiments by
using an aziridine-based molecule possessing pendant anthracene units as a chemosensor, in both
methanol and water as solvents, so as to determine the solvent effect on the quenching/enhancement
mechanism. No significant differences were found in regards to the quenching of the fluorescence
intensity in the presence of different metal ions when using water or methanol as solvents.
Figure 4a provides the fluorescence images corresponding to the BIEE/pDMAEMA/pAnMMA
methanol-swollen network immersed in pure methanol (control samples) and methanol solutions
containing different Fe3+ and Cu2+ metal ion concentrations. From the obtained fluorescence images,
57

Polymers 2016, 8, 411

it can be clearly observed that the fluorescence intensity of the network is effectively quenched in the
presence of both Fe3+ and Cu2+ metal ions, whereas quenching is more pronounced upon increasing
the metal ion concentration as expected. During the analysis by means of fluorescence microscopy,
photographs of the networks immersed in metal ion solutions of various concentrations were also
taken in real time. As seen in Figure 4b, fluorescence quenching can be easily visualized, since
upon immersion of the fluorescent network in the metal ion solutions its fluorescence efficiency
reduces significantly.

(a)

(b)

Figure 4. (a) Fluorescence images of the methanol-swollen BIEE/pDMAEMA/pAnMMA network
immersed in pure methanol (control samples) and methanol solutions containing different Fe3+ and
Cu2+ metal ion concentrations; (b) corresponding photographs. The scale bar in Figure 4a is 100 μm.

Figure 5 presents the fluorescence spectra recorded for the methanol-swollen network samples
after being immersed for 24 h in methanol solutions containing various concentrations of Fe3+ or
Cu2+ , followed by extensive washing to remove any unbound metal ions. As seen in the fluorescence
spectra, a quenching effect is observed in both cases. This is attributed to the presence of unpaired d
electrons in transition metal ions that can effectively quench the anthryl chromophore. The quenching
phenomenon is further increased upon increasing the metal ion concentration, which is in agreement
with our previous studies. Buruiana and co-workers reported on the different quenching mechanisms
that may occur in anthracene-containing systems [74]. These include excimer or exciplex formation,
metal–p complex, electron transfer, and energy transfer. In the case of Fe3+ possessing unpaired
d electrons, the quenching mechanism may involve an energy transfer process from the singlet
excited-state anthracene chromophores to the low-lying half-filled 3d orbitals of Fe3+ [75]. Besides the
obvious decrease in the fluorescence intensity upon increasing the concentration of Fe3+ , a blue
shift is clearly observed in the fluorescence spectra (Figure 5a), whereas in the case of the Cu2+ the
decrease in the fluorescence intensity is first preceded and then accompanied by a red shift (Figure 5b).
According to Micheloni et al. [3], the coordination of metal ions may lead to an enhancement of
the fluorescence emission (chelation-enhanced fluorescence effect, CHEF) or, as in the present study,
58

Polymers 2016, 8, 411

to fluorescence quenching (chelation-enhancement quenching effect, CHEQ), whereas both effects may
be coupled with a red or blue shift of the emission band.
600

1000

Control (no Fe3+)
[Fe3+]: 5x10-6 M

800

[Fe3+]: 10-5 M
[Fe3+]: 5x10-5 M

Control (no Cu2+)
[Cu2+]: 5x10-5 M

500

Fluorescence Intensity

Fluorescence Intensity

1200

600
400

[Cu2+]: 10-4 M
[Cu2+]: 10-3 M

400
300
200
100

200

0

0
400

450

500

550

600

650

400

700

450

500

550

600

650

wavelength (nm)

wavelength (nm)

(a)

(b)

Figure 5. Fluorescence spectra of the methanol-swollen networks samples after being exposed to methanol
solutions of various metal ion concentrations: (a) Fe3+ fluorescence monitoring; (b) Cu2+ monitoring.

According to the fluorescence spectroscopy data, in the case of the Cu2+ there is no obvious
decrease in the fluorescence intensity until high Cu2+ concentrations (10−3 M) are reached, in contrast
to the Fe3+ ions that act as effective quenchers for the anthracene fluorophores when present at much
lower concentrations (~10−5 M). These results are in line with the fluorescence microscopy images
provided in Figure 4.
For further validation of the aforementioned results, control experiments were carried out in
which 9-anthracenemethanol was used as the chemosensor in the monitoring of Fe3+ and Cu2+ in
methanol. From the obtained data (Supplementary Materials, Figure S1, Table S1), the Stern–Volmer
quenching plots were constructed (Figure 6), from which the Stern–Volmer quenching constants
(Ksv) were determined from the Stern–Volmer equation, Io /I = 1 + Ksv × (Cion ), to be (0.33 ± 0.04)
and (16.5 ± 2.8) mM−1 for the Cu2+ and Fe3+ quenchers, respectively. Io and I are the fluorescence
intensities at ~410 nm in the absence and presence of the metal ions, respectively, and (Cion ) is the
concentration of the metal ion quencher. The obtained results are in very good agreement with the
fluorescence spectroscopy data obtained in the case where the networks were used as chemosensors,
thus confirming the higher sensitivity of these systems towards the fluorescence detection of Fe3+
over Cu2+ .

6

Io/I

4

2

0
0.0

0.2

0.4

0.6

0.8

1.0

1.2

metal ion concentration (mM)

Figure 6. Stern–Volmer quenching plots of 9-anthracenemethanol in the presence of Fe3+ (red circles)
and Cu2+ (black circles).

59

Polymers 2016, 8, 411

The fact that in the control experiments using 9-anthracenemethanol as a chemosensor, which
lacks metal ion-binding functionalities, no shift of the emission bands of the anthracene-containing
fluorophore is observed in the presence of both metal ions (Supplementary Materials, Figure S1),
strongly suggests that the semi-IPN network structure promotes the metal–ligand interactions via the
metal-binding tertiary amino groups that are present on the pDMAEMA chains [58–63].
3.4. Metal Ion Desorption—Network Regeneration
The desorption of Cu2+ ions from the networks, and thus their regeneration, was accomplished
upon immersing two different samples of the Cu2+ -loaded networks (pre-equilibrated in 10−3 and
5 × 10−3 M Cu2+ methanol solutions) in EDTA aqueous solutions. EDTA is a well-known chelating
ligand that can bind onto various metal ions, including Cu2+ , forming strong complexes. The metal ion
desorption–regeneration process was qualitatively monitored in real time by fluorescence microscopy.
As seen in the fluorescence images presented in Figure 7, an immediate increase in the fluorescence
intensity of the networks was observed in both cases, qualitatively indicating desorption of the Cu2+
from the networks, thus resulting in the recovery of the networks’ fluorescence efficiency.

Figure 7. Fluorescence images analyzed using the CellSens software. Images (a,b) correspond to the
Cu2+ -loaded system equilibrated in 5 × 10−3 M Cu2+ methanol solution prior to and after immersion
in the ethylenediaminetetraacetic acid (EDTA) aqueous solution, respectively. Images (c,d) correspond
to the Cu2+ -loaded system equilibrated in 10−3 M Cu2+ methanol solution prior to and after immersion
in the EDTA aqueous solution, respectively. The scale bar in Figure 7 is 100 μm.

4. Conclusions
We have demonstrated a simple and versatile methodology that leads to the generation of 3D
fluorescent semi-IPN amphiphilic polymer networks with controlled architectures, deriving from
the well-defined molecular characteristics of their linear precursors. Fluorescence spectroscopy and
microscopy demonstrated that these materials behave as strong blue emitters under UV irradiation in
the presence of only 5 wt % of the pAnMMA linear chains that are interlaced between BIEE-crosslinked
pDMAEMA segments. These systems were further evaluated as sorbents for the uptake and
fluorescence monitoring of transition metal ions (Fe3+ , Cu2+ ) in solution. The combination of the
coordinating active part (pDMAEMA) with the fluorescent active component (pAnMMA) within a
semi-IPN network architecture promotes a chelation-enhanced fluorescence quenching effect that is
more pronounced in the case of Fe3+ . Moreover, desorption of the Cu2+ from the networks could be
realized upon immersion of the latter in an EDTA-containing solution, thus allowing their recyclability.

60

Polymers 2016, 8, 411

Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/8/12/411/s1,
Table S1: Concentration of the metal ion quenchers (Cu2+ , Fe3+ ) and Io /I data obtained by fluorescence spectroscopy
when using 9-anthracenemethanol as a fluorophore; Figure S1: Fluorescence spectra of the 9-anthracenemethanol
after being exposed to methanol solutions of various metal ion concentrations: Cu2+ fluorescence monitoring: left
spectrum; Fe3+ monitoring: right spectrum.
Acknowledgments: This work was supported by the University of Cyprus. We are grateful to Maria Demetriou
for the synthesis of the pAnMMA homopolymer used in the present study and to Triantafyllos Stylianopoulos
(Cancer Biophysics Laboratory, Department of Mechanical and Manufacturing Engineering, University of Cyprus)
for providing access to the fluorescence microscope.
Theodora Krasia-Christoforou conceived and designed the experiments.
Author Contributions:
Kyriakos Christodoulou and Mariliz Achilleos performed the synthesis of the networks. Kyriakos Christodoulou,
Theodora Krasia-Christoforou and Epameinondas Leontidis performed the fluorescence spectroscopy studies and
Christiana Polydorou carried out the fluorescence microscopy experiments. Epameinondas Leontidis contributed
significantly to the interpretation of the obtained fluorescence spectroscopy data and designed the fluorescence
quenching control experiments using 9-anthracenemethanol. Theodora Krasia-Christoforou wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.

2.
3.
4.
5.

6.
7.
8.
9.
10.

11.
12.

13.
14.
15.

Aron, A.T.; Ramos-Torres, K.M.; Cotruvo, J.A., Jr.; Chang, C.J. Recognition- and reactivity-based fluorescent
probes for studying transition metal signaling in living systems. Acc. Chem. Res. 2015, 48, 2434–2442. [CrossRef]
[PubMed]
Prodi, L.; Bolletta, F.; Montalti, M.; Zaccheroni, N. Luminescent chemosensors for transition metal ions.
Coord. Chem. Rev. 2000, 205, 59–83. [CrossRef]
Formica, M.; Fusi, V.; Giorgi, L.; Micheloni, M. New fluorescent chemosensors for metal ions in solution.
Coord. Chem. Rev. 2012, 256, 170–192. [CrossRef]
Fabbrizzi, L.; Licchelli, M.; Pallavicini, P.; Perotti, A.; Sacchi, D. An anthracene-based fluorescent sensor for
transition metal ions. Angew. Chem. Int. Ed. 1994, 33, 1975–1977. [CrossRef]
Demetriou, M.; Krasia-Christoforou, T. Well-defined diblock copolymers possessing fluorescent and metal
chelating functionalities as novel macromolecular sensors for amines and metal ions. J. Polym. Sci. Polym. Chem.
2012, 50, 52–60. [CrossRef]
Yang, J.; Wang, Z.; Li, Y.; Zhuang, Q.; Gu, J. Real-time monitoring of dissolved oxygen with inherent
oxygen-sensitive centers in metal-organic frameworks. Chem. Mater. 2016, 28, 2652–2658. [CrossRef]
Zhao, S.-S.; Yang, J.; Liu, Y.-Y.; Ma, J.-F. Fluorescent aromatic tag-functionalized MOFs for highly selective
sensing of metal ions and small organic molecules. Inorg. Chem. 2016, 55, 2261–2273. [CrossRef] [PubMed]
Wang, X.-P.; Han, L.-L.; Wang, Z.; Guo, L.-Y.; Sun, D. Microporous Cd(II) metal-organic framework as
fluorescent sensor for nitroaromatic explosives at the sub-ppm level. J. Mol. Struct. 2016, 1107, 1–6. [CrossRef]
He, J.; Huang, J.; He, Y.; Cao, P.; Zeller, M.; Hunter, A.D.; Xu, Z. A boiling-water-stable, tunable white-emitting
metal-organic framework from soft-imprint synthesis. Chem. Eur. J. 2016, 22, 1597–1601. [CrossRef] [PubMed]
Yang, L.; Li, X.; Qin, C.; Shao, K.-Z.; Su, Z.-M. A fluorescent sensor for highly selective sensing of nitro
explosives and Hg(II) ions based on a 3D porous layer metal-organic framework. CrystEngComm 2016, 18,
4765–4771. [CrossRef]
Zhao, S.; Lv, X.-X.; Shi, L.-L.; Li, B.L.; Wu, B. An unusual (4,4)-connected 3D porous cadmium metal-organic
framework as a luminescent sensor for detection of nitrobenzene. RSC Adv. 2016, 6, 56035–56041. [CrossRef]
Rachuri, Y.; Parmar, B.; Bisht, K.K.; Suresh, E. Mixed ligand two dimensional Cd(II)/Ni(II) metal organic
frameworks containing dicarboxylate and tripodal N-donor ligands: Cd(II) MOF is an efficient luminescent
sensor for detection of picric acid in aqueous media. Dalton Trans. 2016, 45, 7881–7892. [CrossRef] [PubMed]
Gole, B.; Bar, A.K.; Mukherjee, P.S. Multicomponent assembly of fluorescent-tag functionalized ligands in
metal-organic frameworks for sensing explosives. Chem. Eur. J. 2014, 20, 13321–13336. [CrossRef] [PubMed]
Yi, F.-Y.; Yang, W.; Sun, Z.-M. Highly selective acetone fluorescent sensors based on microporous Cd(II)
metal-organic frameworks. J. Mater. Chem. 2012, 22, 23201–23209. [CrossRef]
He, J.; Yee, K.-K.; Xu, Z.; Zeller, M.; Hunter, A.-D.; Chui, S.S.-Y.; Che, C.-M. Thioether side chains improve
the stability, fluorescence, and metal uptake of a metal-organic framework. Chem. Mater. 2011, 23, 2940–2947.
[CrossRef]

61

Polymers 2016, 8, 411

16.
17.

18.

19.

20.

21.
22.

23.

24.
25.
26.
27.
28.
29.
30.

31.
32.
33.

34.
35.
36.
37.

Liu, J.-J.; Guan, Y.-F.; Lin, M.-J.; Huang, C.-C.; Dai, W.-X. Luminescent coordination polymer with conjugated
Lewis acid sites for the detection of organic amines. Cryst. Growth Des. 2015, 15, 5040–5046. [CrossRef]
Deng, J.; Ma, W.; Yu, P.; Mao, L. Colorimetric and fluorescent dual mode sensing of alcoholic strength in
spirit samples with stimuli-responsive infinite coordination polymers. Anal. Chem. 2015, 87, 6958–6965.
[CrossRef] [PubMed]
Lin, H.-Y.; Le, M.; Sui, F.-F.; Liu, G.-C. Two new 3D Zn(II)/Cu(II) coordination polymers based on
N,N  -Di(3-pyridyl)succinamide and two different aromatic dicarboxylates: Syntheses, crystal structures and
properties. J. Chem. Crystallogr. 2015, 45, 94–102. [CrossRef]
Feng, X.; Li, R.; Wang, L.; Ng, S.W.; Qin, G.; Ma, L. A series of homonuclear lanthanide coordination polymers
based on a fluorescent conjugated ligand: Syntheses, luminescence and sensor for pollutant chromate anion.
CrystEngComm 2015, 17, 7878–7887. [CrossRef]
Xue, Z.; Sheng, T.; Wang, Y.; Hu, S.; Wen, Y.; Wang, Y.; Li, H.; Fu, R.; Wu, X. A series of d(10) coordination
polymers constructed with a rigid tripodal imidazole ligand and varied polycarboxylates: Syntheses,
structures and luminescence properties. CrystEngComm 2015, 17, 2004–2012. [CrossRef]
Dragan, E.S. Design and applications of interpenetrating polymer network hydrogels. A review. Chem. Eng. J.
2014, 243, 572–590. [CrossRef]
Kostanski, L.K.; Huang, R.; Filipe, C.D.M.; Ghosh, R. Interpenetrating Polymer Networks as a Route to
Tunable Multi-responsive Biomaterials: Development of Novel Concepts. J. Biomater. Sci. Polym. Ed. 2009,
20, 271–297. [CrossRef] [PubMed]
Aminabhavi, T.M.; Nadagouda, M.N.; More, U.A.; Joshi, S.D.; Kulkarni, V.H.; Noolvi, M.N.; Kulkarni, P.V.
Controlled release of therapeutics using interpenetrating polymeric networks. Expert Opin. Drug Deliv. 2015,
12, 669–688. [CrossRef] [PubMed]
Mespouille, L.; Hedrick, J.L.; Dubois, P. Expanding the role of chemistry to produce new amphiphilic
polymer (co)networks. Soft Matter 2009, 5, 4878–4892. [CrossRef]
Siddhanta, S.K.; Gangopadhyay, R. Conducting polymer gel: Formation of a novel semi-IPN from polyaniline
and crosslinked poly(2-acrylamido-2-methyl propanesulphonicacid). Polymer 2005, 46, 2933–3000. [CrossRef]
Bonillo, B.; Sprick, R.S.; Cooper, A.I. Tuning photophysical properties in conjugated microporous polymers
by comonomer doping strategies. Chem. Mater. 2016, 28, 3469–3480. [CrossRef]
Geng, T.-M.; Zhu, H.; Song, W.; Zhu, F.; Wang, Y. Conjugated microporous polymer-based carbazole
derivatives as fluorescence chemosensors for picronitric acid. J. Mater. Sci. 2016, 51, 4104–4114. [CrossRef]
Ji, X.; Yao, Y.; Li, J.; Yan, X.; Huang, F. A Supramolecular cross-linked conjugated polymer network for
multiple fluorescent sensing. J. Am. Chem. Soc. 2013, 135, 74–77. [CrossRef] [PubMed]
Li, P.; Huang, L.; Lin, Y.; Shen, L.; Chen, Q.; Shi, W. Printable temperature-responsive hybrid hydrogels with
photoluminescent carbon nanodots. Nanotechnology 2014, 25, 055603. [CrossRef] [PubMed]
Xu, L.; Wang, R.; Kelso, L.C.; Ying, Y.; Li, Y. A target-responsive and size-dependent hydrogel aptasensor
embedded with QD fluorescent reporters for rapid detection of avian influenza virus H5N1. Sens. Actuators
B Chem. 2016, 234, 98–108. [CrossRef]
Yan, J.-J.; Wang, H.; Zhou, Q.-H.; You, Y.-Z. Reversible and multisensitive quantum dot gels. Macromolecules
2011, 44, 4306–4312. [CrossRef]
Cellini, F.; Zhou, L.; Khapli, S.; Peterson, S.D.; Porfiri, M. Large deformations and fluorescence response of
mechanochromic polyurethane sensors. Mech. Mater. 2016, 93, 145–162. [CrossRef]
Nguyen, T.H.; Venugopala, T.; Chen, S.; Sun, T.; Grattan, K.T.V.; Taylor, S.E.; Basheer, M.; Long, A.E.
Fluorescence based fibre optic pH sensor for the pH 10–13 range suitable for corrosion monitoring in
concrete structures. Sens. Actuators B Chem. 2014, 191, 498–507. [CrossRef]
Li, Y.; Zhou, S. A simple method to fabricate fluorescent glucose sensor based on dye-complexed microgels.
Sens. Actuators B Chem. 2013, 177, 792–799. [CrossRef]
Cajlakovic, M.; Lobnik, A.; Werner, T. Stability of new optical pH sensing material based on cross-linked
poly(vinyl alcohol) copolymer. Anal. Chim. Acta 2002, 455, 207–213. [CrossRef]
Stock, N.; Biswas, S. Synthesis of metal-organic frameworks (MOFs): Routes to various MOF topologies,
morphologies, and composites. Chem. Rev. 2012, 112, 933–969. [CrossRef] [PubMed]
Tuci, G.; Rossin, A.; Xu, X.; Ranocchiari, M.; van Bokhoven, J.A.; Luconi, L.; Manet, I.; Melucci, M.;
Giambastiani, G. “Click” on MOFs: A versatile tool for the multimodal derivatization of N-3-decorated
metal organic frameworks. Chem. Mater. 2013, 25, 2297–2308. [CrossRef]
62

Polymers 2016, 8, 411

38.
39.

40.
41.
42.
43.

44.
45.

46.

47.
48.

49.

50.
51.
52.

53.

54.

55.

56.
57.

58.

Oezdemir, E.; Thirion, D.; Yavuz, C.T. Covalent organic polymer framework with C–C bonds as a fluorescent
probe for selective iron detection. RSC Adv. 2015, 5, 69010–69015. [CrossRef]
Jiang, H.-L.; Feng, D.; Liu, T.-F.; Li, J.-R.; Zhou, H.-C. Pore surface engineering with controlled loadings of
functional groups via click chemistry in highly stable metal-organic frameworks. J. Am. Chem. Soc. 2012, 134,
14690–14693. [CrossRef] [PubMed]
Wang, Z.; Cohen, S.M. Postsynthetic modification of metal-organic frameworks. Chem. Soc. Rev. 2009, 38,
1315–1329. [CrossRef] [PubMed]
Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T. Quantum dots versus organic
dyes as fluorescent labels. Nat. Methods 2008, 5, 763–777. [CrossRef] [PubMed]
Lou, Q.; Qu, S.; Jing, P.; Ji, W.; Li, D.; Cao, J.; Zhang, H.; Liu, L.; Zhao, J.; Shen, D. Water-triggered luminescent
“nano-bombs” based on supra-(carbon nanodots). Adv. Mater. 2015, 27, 1389–1394. [CrossRef] [PubMed]
Haraguchi, K.; Takehisa, T. Nanocomposite hydrogels: A unique organic–inorganic network structure with
extraordinary mechanical, optical, and swelling/de-swelling properties. Adv. Mater. 2002, 14, 1120–1124.
[CrossRef]
Blezer, R.; Lindhout, T.; Keszler, B.; Kennedy, J.P. Amphiphilic networks. VIII. Reduced in vitro
thromboresistance of amphiphilic networks. Polym. Bull. 1995, 34, 101–107. [CrossRef]
Süvegh, K.; Domján, A.; Vankó, G.; Iván, B.; Vértes, A. Free volume and swelling dynamics of the poly
[(2-dimethylamino) ethyl methacrylate]-l-polyisobutylene amphiphilic network by positron annihilation
investigations. Macromolecules 1998, 31, 7770–7775. [CrossRef]
Hentze, H.-P.; Krämer, E.; Berton, B.; Förster, S.; Antonietti, M.; Dreja, M. Lyotropic mesophases of
poly(ethylene oxide)-b-poly(butadiene) diblock copolymers and their cross-linking to generate ordered
gels. Macromolecules 1999, 32, 5803–5809. [CrossRef]
Bromberg, L.; Temchenko, M.; Hatton, T.A. Dually Responsive microgels from polyether-modified
poly(acrylic acid): Swelling and drug loading. Langmuir 2002, 18, 4944–4952. [CrossRef]
Simmons, M.R.; Yamasaki, E.N.; Patrickios, C.S. Cationic amphiphilic model networks: Synthesis by group
transfer polymerization and characterization of the degree of swelling. Macromolecules 2000, 33, 3176–3179.
[CrossRef]
Triftaridou, A.I.; Hadjiyannakou, S.C.; Vamvakaki, M.; Patrickios, C.S. Synthesis, characterization, and
modeling of cationic amphiphilic model hydrogels: Effects of polymer composition and architecture.
Macromolecules 2002, 35, 2506–2513. [CrossRef]
Vamvakaki, M.; Patrickios, C.S. Synthesis and characterization of electrolytic amphiphilic model networks
based on cross-linked star polymers: Effect of star architecture. Chem. Mater. 2002, 14, 1630–1638. [CrossRef]
Gao, H.; Li, W.; Matyjaszewski, K. Synthesis of polyacrylate networks by ATRP: Parameters influencing
experimental gel points. Macromolecules 2008, 41, 2335–2340. [CrossRef]
Rikkou, M.D.; Kolokasi, M.; Matyjaszewski, K.; Patrickios, C.S. End-linked amphiphilic polymer conetworks:
Synthesis by sequential atom transfer radical polymerization and swelling characterization. J. Polym. Sci.
Polym. Chem. 2010, 48, 1878–1886. [CrossRef]
Krasia-Christoforou, T.; Patrickios, C.S. Amphiphilic polymethacrylate model co-networks: Synthesis
by RAFT radical polymerization and characterization of the swelling behavior. Macromolecules 2006, 39,
2467–2473. [CrossRef]
Achilleos, M.; Krasia-Christoforou, T.; Patrickios, C.S. Amphiphilic model conetworks based on combinations
of methacrylate, acrylate, and styrenic units: Synthesis by RAFT radical polymerization and characterization
of the swelling behavior. Macromolecules 2007, 40, 5575–5581. [CrossRef]
Wu, W.; Liu, J.; Cao, S.; Tan, H.; Li, J.; Xu, F.; Zhang, X. Drug release behaviors of a pH
sensitive semi-interpenetrating polymer network hydrogel composed of poly(vinyl alcohol) and star
poly[2-(dimethylamino)ethyl methacrylate]. Int. J. Pharm. 2011, 416, 104–109. [CrossRef] [PubMed]
Yao, F.; Xu, L.; Fu, G.-D.; Lin, B. Sliding-graft interpenetrating polymer networks from simultaneous “click
chemistry” and atom transfer radical polymerization. Macromolecules 2010, 43, 9761–9770. [CrossRef]
Achilleos, M.; Mpekris, F.; Stylianopoulos, T.; Krasia-Christoforou, T. Structurally-defined semi-interpenetrating
amphiphilic polymer networks with tunable and predictable mechanical response. RSC Adv. 2016, 6,
43278–43283. [CrossRef]
Shenmin Zhu, S.; Yang, N.; Zhang, D. Poly(N,N-dimethylaminoethyl methacrylate) modification of activated
carbon for copper ions removal. Mater. Chem. Phys. 2009, 113, 784–789.
63

Polymers 2016, 8, 411

59.

60.

61.

62.
63.
64.
65.
66.
67.
68.
69.

70.
71.
72.

73.
74.
75.

Zhu, W.; Song, H.; Du, K.; Zeng, H.; Yao, S. Rapid removal of Cr(VI) ions from aqueous solutions by the
macroporous poly(N,N  -dimethylamino ethyl methacrylate) hydrogels. J. Appl. Polym. Sci. 2013, 128,
2729–2735. [CrossRef]
Jiang, M.; Wang, J.; Li, L.; Pan, K.; Cao, B. Poly(N,N-dimethylaminoethyl methacrylate) modification of
a regenerated cellulose membrane using ATRP method for copper(II) ion removal. RSC Adv. 2013, 3,
20625–20632. [CrossRef]
Zhao, L.; Sun, J.; Zhao, Y.; Xu, L.; Zhai, M. Removal of hazardous metal ions from wastewater by radiation
synthesized silica-graft-dimethylaminoethyl methacrylate adsorbent. Chem. Eng. J. 2011, 170, 162–169.
[CrossRef]
Tokuyama, H.; Ishihara, N. Temperature-swing adsorption of precious metal ions onto
poly(2-(dimethylamino)ethyl methacrylate) gel. React. Funct. Polym. 2010, 70, 610–615. [CrossRef]
Kavakh, P.A.; Yilmaz, Z.; Sen, M. Investigation of heavy metal ion adsorption characteristics of
poly(N,N-dimethylamino ethylmethacrylate) hydrogels. Sep. Sci. Technol. 2007, 42, 1245–1254. [CrossRef]
Cheng-Xiong, Y.; Hu-Bo, R.; Xiu-Ping, Y. Fluorescent metal–organic framework MIL-53(Al) for highly
selective and sensitive detection of Fe3+ in aqueous solution. Anal. Chem. 2013, 85, 7441–7446.
Krämer, R. Fluorescent chemosensors for Cu2+ Ions: Fast, selective, and highly sensitive. Angew. Chem. 1998,
37, 772–773. [CrossRef]
Ferreira de Deus, J.; Andrade, M.L.; Atvars, T.D.Z.; Akcelrud, L. Photo and electroluminescence studies of
poly(methyl methacrylate-co-9-anthryl methyl methacrylate). Chem. Phys. 2004, 297, 177–186. [CrossRef]
Li, Y.; Ryan, A.J.; Armes, S.P. Synthesis of well-defined branched copolymers by quaternization of
near-monodisperse homopolymers. Macromolecules 2008, 41, 5577–5581. [CrossRef]
Byron, C.M.; Werner, T.C. Experiments in synchronous fluorescence spectroscopy for the undergraduate
instrumental chemistry course. J. Chem. Educ. 1991, 68, 433–436. [CrossRef]
Visscher, A.; Bachmann, S.; Schnegelsberg, C.; Teuteberg, T.; Mata, R.A.; Stalke, D. Highly selective and
sensitive fluorescence detection of Zn2+ and Cd2+ ions by using an acridine sensor. Dalton Trans. 2016, 45,
5689–5699. [CrossRef] [PubMed]
Gupta, V.K. A highly selective fluorescence and electrochemical sensor for Ni(II) ions based on
4-(4-phenylthiazole-2-ylimino)pent-2-en-2-ol. Int. J. Electrochem. Sci. 2015, 10, 7854–7865.
Dereka, B.; Svechkarev, D.; Doroshenko, A.O. Facile ultrasensitive monitoring of mercury ions in water by
fluorescent ratiometric detection. Cent. Eur. J. Chem. 2013, 11, 584–593. [CrossRef]
Yu, C.; Zhang, J.; Li, J.; Liu, P.; Wei, P.; Chen, L. Fluorescent probe for copper(II) ion based on a rhodamine
spirolactame derivative, and its application to fluorescent imaging in living cells. Microchim. Acta 2011, 174,
247–255. [CrossRef]
Raje, V.P.; Morgado, P.I.; Ribeiro, M.P.; Correia, I.J.; Bonifácio, V.D.B.; Branco, P.S.; Aguiar-Ricardo, A. Dual
on-off and off-on switchable oligoaziridine biosensor. Biosens. Bioelectron. 2013, 39, 64–69. [CrossRef] [PubMed]
Buruiana, E.C.; Olaru, M.; Simionescu, B.C. Photochemical aspects in anthracene-containing cationic
polyurethanes. Eur. Polym. J. 2007, 43, 1359–1371. [CrossRef]
Weizman, H.; Ardon, O.; Mester, B.; Libman, J.; Oren Dwir, O.; Hadar, Y.; Chen, Y.; Shanzer, A.
Fluorescently-labeled ferrichrome analogs as probes for receptor-mediated, microbial iron uptake. J. Am.
Chem. Soc. 1996, 118, 12368–12375. [CrossRef]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

64

polymers
Article

Thixotropic Supramolecular Pectin-Poly(Ethylene
Glycol) Methacrylate (PEGMA) Hydrogels
Siew Yin Chan 1,2 , Wee Sim Choo 1, *, David James Young 1,2,3, * and Xian Jun Loh 2,4,5, *
1
2
3
4
5

*

School of Science, Monash University Malaysia, Subang Jaya 47500, Malaysia; siew.chan@monash.edu
Institute of Materials Research and Engineering (IMRE), A*STAR (Agency for Science,
Technology and Research), Singapore 138634, Singapore
Faculty of Science, Health, Education and Engineering, University of the Sunshine Coast,
Sunshine Coast, QLD 4558, Australia
Department of Materials Science and Engineering, National University of Singapore,
Singapore 117576, Singapore
Singapore Eye Research Institute, Singapore 168751, Singapore
Correspondence: choo.wee.sim@monash.edu (W.S.C.);
dyoung1@usc.edu.au (D.J.Y.); lohxj@imre.a-star.edu.sg (X.J.L.);
Tel.: +60-3-5514-6114 (W.S.C.); +61-7-5456-3448 (D.J.Y.); +65-6416-8932 (X.J.L.)

Academic Editor: Alexander Böker
Received: 2 October 2016; Accepted: 10 November 2016; Published: 18 November 2016

Abstract: Pectin is an anionic, water-soluble polymer predominantly consisting of covalently
1,4-linked α-D-galacturonic acid units. This naturally occurring, renewable and biodegradable
polymer is underutilized in polymer science due to its insolubility in organic solvents, which renders
conventional polymerization methods impractical. To circumvent this problem, cerium-initiated
radical polymerization was utilized to graft methoxy-poly(ethylene glycol) methacrylate (mPEGMA)
onto pectin in water. The copolymers were characterized by 1 H nuclear magnetic resonance (NMR),
Fourier transform infrared (FTIR) spectroscopy and thermogravimetric analysis (TGA), and used
in the formation of supramolecular hydrogels through the addition of α-cyclodextrin (α-CD) to
induce crosslinking. These hydrogels possessed thixotropic properties; shear-thinning to liquid upon
agitation but settling into gels at rest. In contrast to most of the other hydrogels produced through
the use of poly(ethylene glycol) (PEG)-grafted polymers, the pectin-PEGMA/α-CD hydrogels were
unaffected by temperature changes.
Keywords: pectin; poly(ethylene glycol) methacrylate; cerium; α-cyclodextrin; supramolecular hydrogel

1. Introduction
Pectins are complex carbohydrates derived from dicotyledonous and some monocotyledonous
plants [1]. Commercially, pectins are produced from food industry waste [2]. They are mainly extracted
with hot dilute mineral acid from citrus peel, apple pomace and to a smaller extent, sugar beet pulp [3,4].
Pectin is a safe food additive with no limit on acceptable daily intake. It is non-toxic, biodegradable
and biocompatible [5,6]. Its applications are diverse spanning the food, pharmaceutical, cosmetic and
polymer industries [7–9]. Pectins are primarily utilized as emulsifiers, gelling agents, glazing agents,
stabilizers or thickeners [10].
This family of polysaccharides consists of galacturonic acid (GalA) units covalently-linked by
α-(1→4) glycosidic bonds and are classified according to the degree of methylation (DM). DM is
defined as the molar ratio of methyl-esters present relative to the total moles of GalA units. It is the
major parameter affecting gelling [11]. Pectins are classified as either high methoxy pectin (HMP) with
a DM > 50% or low methoxy pectin (LMP) with a DM < 50%. HMP gels in high co-solute concentration

Polymers 2016, 8, 404

65

www.mdpi.com/journal/polymers

Polymers 2016, 8, 404

with acid through a combination of hydrophobic forces and hydrogen bonding [12]. LMP gels in the
presence of divalent metal cations such as calcium over a broad range of pH and the gels are formed
through ionic cross-linking between free carboxylate groups in an arrangement known as the egg-box
model [13].
We have an interest in grafting pectin with synthetic polymers to improve its processability in
organic solvents and to impart biocompatibility and biodegradability to the resulting copolymer for use
in biomedical applications. Herein, we describe pectin-poly(ethylene glycol) methacrylate (PEGMA)
copolymer and subsequently generated supramolecular hydrogels by threading the poly(ethylene
glycol) (PEG) tendrils with α-cyclodextrin (α-CD) units. Harada et al. [14] first reported supramolecular
hydrogel formed by threading α-cyclodextrin toroids onto PEG of high molecular weight. This
PEG-graft forms physical cross-linking points via the aggregation of the inclusion complexes, yielding
supramolecular hydrogels [15–19].
Pectin is a hydrophilic polymer and insoluble in all organic solvents. Most polymerization
methods require organic solvents, making it a challenge to graft pectin with synthetic polymers. We
have for the first time utilized a redox polymerization reaction in water to graft PEGMA onto pectin
and have thereby generated pectin supramolecular gels with unique rheology.
2. Materials and Methods
Ammonium cerium (IV) sulfate dihydrate, apple pectin and sodium nitrate were purchased from
Sigma-Aldrich, Singapore, Singapore. Methoxy-poly(ethylene glycol) methacrylate (mPEGMA) with
Mn of 10,000 g·mol−1 was purchased from Sinopeg, Xiamen, China. α-Cyclodextrin (α-CD) was
purchased from TCI, Kawaguchi, Japan. All reagent and solvents were used as received.
2.1. Synthesis of Pectin-PEGMA Copolymer, P-10K
Ammonium cerium (IV) sulfate dihydrate (0.005 M, based on the final volume of solution) was
added into a pectin solution (1% w/v in water). The solution was stirred at room temperature for 2 h.
mPEGMA (1% w/v) was then added. The reaction mixture was stirred at room temperature for 48 h.
Chloroform was added and the resulting mixture was left to settle. Pectin-PEGMA was obtained in
61.5% yield by precipitating the white emulsion layer in hexane followed by drying under vacuum at
40 ◦ C. The mechanism of cerium-initiated radical polymerization is shown in Figure 1.

Figure 1. Mechanism of cerium-initiated radical polymerization.

66

Polymers 2016, 8, 404

2.2. Polymer Characterisation
1 H Nuclear magnetic resonance (NMR) spectra were recorded on a JEOL 500 MHz NMR
spectrometer (JEOL, Tokyo, Japan) at room temperature. The 1 H NMR measurements were carried out
with an acquisition time of 4.37 s, a pulse repetition time of 9.37 s and a 90◦ pulse width. Measurements
were done with 16 scans and chemical shifts were referred to the solvent peak (δ = 4.79 ppm for
deuterium oxide, D2 O).
Fourier transform infrared (FTIR) spectra of the pellet samples were recorded on a Perkin Elmer
Spectrum 2000 FTIR spectrometer (Perkin Elmer, Waltham, MA, USA); 64 scans were signal-averaged
with a resolution of 4 cm−1 at room temperature. Pellets were prepared by coating the samples with
potassium bromide.

2.3. Thermal Analysis
Thermogravimetric analysis (TGA) was performed on a TA Instruments TGA Q500 (TA Instruments,
New Castle, DE, USA). Samples were heated from room temperature to 900 ◦ C at a rate of 20 ◦ C·min−1
under continuous nitrogen purge with a flow rate of 40 mL·min−1 .
2.4. Preparation of Pectin-PEGMA/α-CD Hydrogels
Solutions of α-CD in water were added into pectin-PEGMA solutions of different compositions.
The resultant mixtures were sonicated and left to stand at room temperature. Gel formation was
observed at 24 h (Figure 2). The sol-gel transition for P-10K was determined by plotting α-CD
concentration versus polymer concentration over the polymer to α-CD solution composition range of
1%–10% w/v, permitting determination of the critical gelation concentration.

Figure 2. The preparation of pectin-PEGMA/α-CD gels.

2.5. Rheology Studies
The rheological behavior of pectin-PEGMA/α-CD gels was investigated using a Discovery DHR-3
hybrid rheometer (TA Instruments, New Castle, DE, USA) with flat plate geometries (diameters of
20 and 40 mm) in steady and dynamic modes. All the tests were performed using a flat plate geometry
with diameter of 40 mm unless otherwise stated. Amplitude sweeps were performed under oscillatory
shear at strain of 0.01%–100% to ensure that subsequent data were collected in the linear viscoelastic
region (LVR). Frequency sweeps were then performed in the range of 0.01–50 Hz under oscillatory
shear at a strain of 0.05%. Reversibility of the hydrogels was determined by amplitude sweeps at
2 points (0.05% and 25% strain), 5 and 2.5 min, respectively, at each strain point, for 3 cycles. All tests

67

Polymers 2016, 8, 404

were performed at 25 ◦ C (room temperature) and 37 ◦ C to mimic body temperature. Temperature
sweeps were performed using a flat plate geometry with diameter of 20 mm at a strain of 0.5% and
frequency of 1 Hz in the temperature range of 10–40 ◦ C, with a ramp rate of 5 ◦ C·min−1 .
3. Results
3.1. Synthesis and Characterization of Pectin-PEGMA Copolymer, P-10K
Pectin-PEGMA copolymer (P-10K) was prepared by radical polymerization in distilled water
using ammonium cerium (IV) sulfate dihydrate as a redox initiator. Chloroform was added to extract
excess PEGMA. Three layers formed in settling; a cloudy brown solution, a white middle layer
emulsion and at the bottom a transparent chloroform containing the unreacted mPEGMA.
1 H NMR spectroscopy in D O was performed on the middle emulsion layer, but proved
2
inconclusive (Figure S1). FTIR, however, contained absorption characteristics of pectin and mPEGMA
(10KPEGMA). Pectin exhibits a broad peak at around 3450 cm−1 , arising from hydroxyl group
stretching (O–H) (Figure 3) [20]. Absorptions at 1750 and 1630 cm−1 were assigned to the carboxylic
acid and ester groups (C=O stretches) of the pectin polymers [21]. 10KPEGMA displays a broad peak
around 3450 cm−1 but not as intensive as for pectin. The FTIR spectrum of 10KPEGMA also contained
distinctive C–H stretches at 2875 cm−1 . CH2 (bending) stretches at 1450 cm−1 , CH3 (bending) stretches
at 1350 cm−1 and C–O stretches at 1110 cm−1 were also observed. The FTIR spectrum obtained for
10KPEGMA matched those reported by Wang et al. [22] and Bagheri et al. [23].

Figure 3. FTIR spectra of the precursors (pectin and 10KPEGMA) and of copolymer P-10K.

3.2. Thermal Analysis
Thermal stabilities of the precursors and copolymer were studied by TGA. The degradation
temperatures were determined from the peak of derivative weight curves and are summarized in
Table 1. There were two degradation temperatures recorded for pectin; 86.33 and 253.33 ◦ C (Figure 4).
The first degradation temperature corresponds to the loss of water. 10KPEGMA exhibited a degradation
temperature of 413.53 ◦ C. TGA of P-10K revealed three degradation temperatures, corresponding
to those for pectin and mPEGMA. The amount of pectin in P-10K was calculated to be around
17% (Table 1).

68

Polymers 2016, 8, 404

Table 1. Decomposition temperatures (Td ) of the precursors (pectin and 10KPEGMA) and copolymer
P-10K.
Sample

T d (◦ C)

Weight change (%)

Pectin

86.33
253.33

8.95
80.82

10KPEGMA

418.07

99.02

P-10K

85.89
240.29
421.19

5.74
16.70
62.91
Pectin
10KPEGMA
P-10K

100
90
80
Weight (%)

70
60
50
40
30
20
10
0

0

100 200 300 400 500 600 700 800 900 1000

Temperature (°C)

Figure 4. TGA curves of the precursors (pectin and 10KPEGMA) and copolymer P-10K.

3.3. Critical Gelation Concentration Determination
Aqueous solutions of α-CD were added to P-10K solutions and the mixtures became gradually
opaque. The mixtures turned white and formed gels after a time period. The effects of the polymer
and α-CD concentrations on gelling behavior were studied by mixing a range of polymer and α-CD
compositions (1%–10% w/v). Hydrogel could be formed at a low polymer concentration of 1% (w/v)
(Figure 5). The copolymer gelled at an optimum ratio of 1% (w/v) polymer and 5% (w/v) α-CD. Ye et al.
have reported that higher cyclodextrin concentrations increase the gel strength [24].
10

α (%)
Concentration of α-CD

9

Gel

8
7
6
5
4
3

Liquid

2
1
0

0

1

2

3

4

5

6

7

8

9

10

Concentration of Polymer (%)

Figure 5. Sol-gel transition graph for different compositions of P-10K/α-CD.

69

Polymers 2016, 8, 404

3.4. Rheology of Pectin-PEGMA/α-CD Hydrogel
The rheology of hydrogels prepared with 10% P-10K and 10% α-CD was investigated. Amplitude
sweeps to measure shear strain were in the range of 0.01%–100% at temperatures of 25 ◦ C (room
temperature) and 37 ◦ C (body temperature). The hydrogels proved to be highly structured, true gels
with storage moduli (G’) greater then loss moduli (G”) at low shear strain (Figure 6). As the shear
strain increased, G’ began to decrease at a rate faster than that of G”. As the oscillation strain increased
from 0.01% to 100%, the hydrogels changed from a gel (G’ > G”) to a liquid (G’ < G”). The critical
strain of the hydrogel was around 7%–11% for P-10K. At a shear strain greater than the critical strain,
G’ dropped below G”, indicating that the gel network structures had been disrupted by shearing.
The linear viscoelastic region (LVR) of the hydrogels was 0.01% to 1% strain [25].
G'

G"

25 °C
37 °C

5

10

G', G" (Pa)

104
103
102
101
100
10-2

10-1

100

101

102

Oscillation strain (%)

Figure 6. Amplitude sweeps performed from 0.01% to 100% of oscillation strain at 25 and 37 ◦ C on
hydrogels prepared with 10% P-10K and 10% α-CD.

Frequency sweeps were performed on the hydrogels from 0.01–50 Hz at 0.05% strain at 25 and
37 ◦ C. G’ were higher than G” and both the moduli were dependent on frequency at both 25 and 37 ◦ C
(Figure 7). Frequency sweeps provide information on the effect of colloidal forces or the interaction
among particles [26]. As frequency increased, there appeared to be no interaction between particles,
presumably because of sufficient separation to eliminate interactions and/or collisions.
G'

G"

25 °C
37 °C

5

10

G', G" (Pa)

104
103
102
101
100
10-2

10-1

100

101

102

Frequency (Hz)

Figure 7. Frequency sweeps performed from 0.01 to 50 Hz of oscillation strain at 25 and 37 ◦ C on
hydrogels prepared with 10% P-10K and 10% α-CD.

70

Polymers 2016, 8, 404

3.5. Thixotropic Properties of Pectin-PEGMA/α-CD Hydrogels
Instantaneously varying strain at 0.05% and 25% in amplitude sweeps for three cycles indicated
that P-10K/α-CD hydrogels recovered their original gel structure (Figure 8) [27]. G’ was higher than
G” during the first 5 min of 0.05% strain and the gel was immediately sheared into a liquid where
G’ < G” during the 2.5 min of 25% strain. The liquid then reverted back to gel (G’ > G”) when the
strain was reduced to 0.05%. However, the hydrogel required some recovery time to revert back to the
original gel form. The internal network structure could be broken down by shearing, and required
time to rebuild, as shown by the coincides of G’ and G” at each 0.05% strain point after being sheared
at 25% strain. This shear–recovery phenomenon was repeatable for all the hydrogels at a minimum of
three cycles, proving the thixotropy of P-10K/α-CD hydrogels.
G'

G"

25 °C
37 °C

105

G', G" (Pa)

104
103
102
101
100

0

2

4

6

8

10 12 14 16 18 20 22 24
Time (min)

Figure 8. Amplitude sweeps at 0.05% and 25% strains performed instantaneously for three cycles at
25 and 37 ◦ C on hydrogels prepared with 10% P-10K and 10% α-CD.

3.6. The Effect of Temperature on Pectin-PEGMA/α-CD Hydrogels
Temperature sweeps were performed to probe the temperature responsiveness of P-10K/α-CD
hydrogels (Figure 9) and demonstrated that these gels were not affected by temperature from 10 to
40 ◦ C. G’ was higher than G” and the moduli remained almost constant over this range.
G'

G"

105

G', G" (Pa)

104
103
102
101
100

0

5

10

15

20

25

30

35

40

Temperature (°C)

Figure 9. Temperature sweep performed from 10 to 40 ◦ C on hydrogel prepared with 10% P-10K and
10% α-CD.

71

Polymers 2016, 8, 404

4. Discussion
Pectin-PEGMA copolymer was successfully synthesized through cerium-initiated radical
polymerization, as evidenced by FTIR spectroscopy (Figure 3) and TGA (Figure 4). The latter, in
particular, was diagnostic, indicating the presence of two components in P-10K and a shift in the
degradation temperature of the copolymer relative to its component precursors. Based on the weight
change determined from these TGA curves (Table 1), it can be concluded there was 17% pectin in P-10K
(Table 1). The FTIR spectra of this block copolymer supported the presence of pectin as a minority
component (Figure 3).
Pectin-PEGMA copolymer was able to form gels with α-CD (Figure 5). The α-CD threaded onto
PEG tendrils to form stacked inclusion complexes (Figure 10) [28,29] that interact with each other
to aggregate via hydrogen bonding [30]. These polypseudorotaxanes can inter- or intra-crosslink
with each other while some chains of stacked inclusion complexes may remain separated (Figure 11).
A control experiment involving mPEGMA alone turned opaque and eventually white when α-CD was
added, but no gel was formed (Table 2). Inclusion complexes still formed between α-CD and PEG
chains. However, the columns of threaded PEG chains remained free with no gelation [31]. P-10K
required 5% α-CD to gel (Figure 5). Upon addition of 10% α-CD into 10% 10KPEGMA and 10% P-10K
solutions, respectively, the co-polymer sample gelled, but not mPEGMA (Figure 12).
The response of pectin-PEGMA/α-CD hydrogel to shear strain and its dependence on frequency
demonstrated shear-thinning. The higher the strain percentage was, the more liquid-like the hydrogel
became. The time oscillation test further demonstrated that the hydrogel was thixotropic. Thixotropic
materials provide a gel consistency when at rest and flow when shear is introduced (Figure 13).
Pectin-PEGMA/α-CD hydrogel was sheared beyond critical strain, but reverted back to gel on resting.
Most PEG-grafted polymers are temperature-sensitive, with the network structure disrupted by an
increase in temperature [15,19]. High temperature may disrupt the non-covalent interactions in
supramolecular hydrogels, causing α-CD to de-thread from PEG chains [15]. The physical cross-linked
network will then collapse. This effect was not observed for pectin-PEGMA/α-CD hydrogel which
maintained its gel consistency from 10 to 40 ◦ C. We hypothesize that pectin may confer thermal stability
to the pectin-PEGMA/α-CD hydrogels. In support of this proposition, it has been reported that heat
treatment of milk does not affect the ability of pectin to stabilize particles such as casein and denatured
whey complex [32].

Figure 10. Threading of α-CD onto PEG chains to form columns of stacked inclusion complexes.

72

Polymers 2016, 8, 404

Figure 11. Polypseudorotaxanes inter- or intra-crosslink with each other via hydrogen bonding.
Table 2. Ten percent 10KPEGMA and 10% P-10K solutions with 10% α-CD.
Sample

Gel

10KPEGMA
P-10K

8


Figure 12. Pectin-PEGMA solution turned white after the addition of α-CD: (a) original 10% P-10K
solution; and (b) 10% P-10K solution after the addition of α-CD (10%).

73

Polymers 2016, 8, 404

Figure 13. Thixotropic behavior of pectin-PEGMA/α-CD hydrogels.

In conclusion, pectin-PEGMA (P-10K) copolymer was successfully synthesized using
cerium-initiated radical polymerization in water. This process is a safer, greener, eco-friendlier way
to graft polymers. α-CD was incorporated into pectin-PEGMA copolymer and supramolecular
gels were formed. The copolymer gelled at low polymer concentration (1%) with 5% of α-CD.
The total concentration of polymer could be lowered to achieve a desired consistency of gel. These
supramolecular hydrogels were thixotropic, i.e., the gel could be sheared to liquid and reverted back
to gel at rest. Industrially, a thixotropic material is easier to process or transfer without disrupting the
consistency. Interestingly, pectin-PEGMA/α-CD hydrogels were not affected by temperature from
10 to 40 ◦ C, unlike the hydrogels of other PEG-grafted polymers [15,19,33,34]. This supramolecular
hydrogel has potential to be developed for cosmetic, body and hair care products, or even as an
injectable pharmaceutical incipient [35,36]. We are currently evaluating the biocompatibility and
biodegradability of pectin-PEGMA and exploring the response of this material to other external stimuli.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/8/11/404/s1.
Figure S1: 500 MHz 1 H NMR spectra of the precursors (pectin and 10KPEGMA) and of copolymer P-10K.
Acknowledgments: Siew Yin Chan gratefully acknowledges the funding and support from Monash University
Malaysia and the Institute of Materials Research and Engineering, A*STAR. The authors thank Benjamin
Qi Yu Chan and Cally Owh for their assistance with illustrations.
Author Contributions: Xian Jun Loh, Wee Sim Choo and David James Young conceived and designed the
experiments. Siew Yin Chan and Xian Jun Loh designed the polymers. Siew Yin Chan synthesized and
characterized the polymers. Siew Yin Chan, Wee Sim Choo, David James Young and Xian Jun Loh co-wrote the
paper. All authors discussed the results and commented on the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.
4.
5.
6.

7.

Braconnot, H. Recherches sur un nouvel acide universellement répandu dans tous les végétaux. Annales de
Chimie et de Physique 1825, 28, 173–178.
Chan, S.-Y.; Choo, W.-S. Effect of extraction conditions on the yield and chemical properties of pectin from
cocoa husks. Food Chem. 2013, 141, 3752–3758. [CrossRef] [PubMed]
Staunstrup, J. Citrus pectin production and world market. In Proceedings of the International Citrus &
Beverage Conference, Clearwater, FL, USA, 15–18 September 2009.
Ciriminna, R.; Chavarría-Hernández, N.; Inés Rodríguez Hernández, A.; Pagliaro, M. Pectin: A new
perspective from the biorefinery standpoint. Biofuels Bioprod. Biorefin. 2015, 9, 368–377. [CrossRef]
Sharma, R.; Ahuja, M. Thiolated pectin: Synthesis, characterization and evaluation as a mucoadhesive
polymer. Carbohydr. Polym. 2011, 85, 658–663. [CrossRef]
Sriamornsak, P.; Thirawong, N.; Nunthanid, J.; Puttipipatkhachorn, S.; Thongborisute, J.; Takeuchi, H. Atomic
force microscopy imaging of novel self-assembling pectin–liposome nanocomplexes. Carbohydr. Polym. 2008,
71, 324–329. [CrossRef]
Günter, E.A.; Popeyko, O.V. Calcium pectinate gel beads obtained from callus cultures pectins as promising
systems for colon-targeted drug delivery. Carbohydr. Polym. 2016, 147, 490–499. [CrossRef] [PubMed]

74

Polymers 2016, 8, 404

8.
9.
10.
11.

12.
13.
14.
15.

16.
17.
18.
19.
20.

21.
22.

23.

24.
25.
26.

27.
28.

29.

Lupi, F.R.; Gabriele, D.; Seta, L.; Baldino, N.; de Cindio, B.; Marino, R. Rheological investigation of
pectin-based emulsion gels for pharmaceutical and cosmetic uses. Rheol. Acta 2014, 54, 41–52. [CrossRef]
Da Costa, M.P.M.; de Mello Ferreira, I.L.; de Macedo Cruz, M.T. New polyelectrolyte complex from
pectin/chitosan and montmorillonite clay. Carbohydr. Polym. 2016, 146, 123–130. [CrossRef] [PubMed]
Codex Alimentarius. General standards for food additives, codex stan 192–1995. In Pectin; Food and
Agriculture Organization of the United Nations: Rome, Italy, 2015.
Müller-Maatsch, J.; Caligiani, A.; Tedeschi, T.; Elst, K.; Sforza, S. Simple and validated quantitative 1 H NMR
method for the determination of methylation, acetylation, and feruloylation degree of pectin. J. Agric.
Food Chem. 2014, 62, 9081–9087. [CrossRef] [PubMed]
Oakenfull, D.; Scott, A. Hydrophobic interaction in the gelation of high methoxyl pectins. J. Food Sci. 1984,
49, 1093–1098. [CrossRef]
Gidley, M.J.; Morris, E.R.; Murray, E.J.; Powell, D.A.; Rees, D.A. Evidence for two mechanisms of interchain
association in calcium pectate gels. Int. J. Biol. Macromol. 1980, 2, 332–334. [CrossRef]
Harada, A.; Li, J.; Kamachi, M. The molecular necklace: A rotaxane containing many threaded
α-cyclodextrins. Nature 1992, 356, 325–327. [CrossRef]
Ren, L.; He, L.; Sun, T.; Dong, X.; Chen, Y.; Huang, J.; Wang, C. Dual-responsive supramolecular hydrogels
from water-soluble PEG-grafted copolymers and cyclodextrin. Macromol. Biosci. 2009, 9, 902–910. [CrossRef]
[PubMed]
Wenz, G.; Han, B.-H.; Müller, A. Cyclodextrin rotaxanes and polyrotaxanes. Chem. Rev. 2006, 106, 782–817.
[CrossRef] [PubMed]
Harada, A.; Li, J.; Kamachi, M. Preparation and characterization of a polyrotaxane consisting of monodisperse
poly(ethylene glycol) and α-cyclodextrins. J. Am. Chem. Soc. 1994, 116, 3192–3196. [CrossRef]
Tonelli, A.E. Soluble PEG-α-CD-rotaxanes: Where on the PEG chains are the permanently threaded α-CDs
located and are they mobile? Macromolecules 2008, 41, 4058–4060. [CrossRef]
Abdul Karim, A.; Loh, X.J. Design of a micellized α-cyclodextrin based supramolecular hydrogel system.
Soft Matter 2015, 11, 5425–5434. [CrossRef] [PubMed]
Chatjigakis, A.K.; Pappas, C.; Proxenia, N.; Kalantzi, O.; Rodis, P.; Polissiou, M. Ft-ir spectroscopic
determination of the degree of esterification of cell wall pectins from stored peaches and correlation to
textural changes. Carbohydr. Polym. 1998, 37, 395–408. [CrossRef]
Stewart, D.; Morrison, I.M. Ft-ir spectroscopy as a tool for the study of biological and chemical treatments of
barley straw. J. Sci. Food Agric. 1992, 60, 431–436. [CrossRef]
Wang, H.; Feng, Y.; An, B.; Zhang, W.; Sun, M.; Fang, Z.; Yuan, W.; Khan, M. Fabrication of PU/PEGMA
crosslinked hybrid scaffolds by in situ UV photopolymerization favoring human endothelial cells growth
for vascular tissue engineering. J. Mater. Sci. 2012, 23, 1499–1510. [CrossRef] [PubMed]
Bagheri, M.; Bigdeli, E.; Pourmoazzen, Z. Self-assembled micellar nanoparticles of a novel
amphiphilic cholesteryl-poly(L-lactic acid)-b-poly(poly(ethylene glycol)methacrylate) block-brush copolymer.
Iran. Polym. J. 2013, 22, 293–302. [CrossRef]
Ye, H.; Owh, C.; Jiang, S.; Ng, C.; Wirawan, D.; Loh, X. A thixotropic polyglycerol sebacate-based
supramolecular hydrogel as an injectable drug delivery matrix. Polymers 2016, 8, 130. [CrossRef]
Sun, Y.; Kaplan, J.A.; Shieh, A.; Sun, H.-L.; Croce, C.M.; Grinstaff, M.W.; Parquette, J.R. Self-assembly of a
5-fluorouracil-dipeptide hydrogel. Chem. Commun. 2016, 52, 5254–5257. [CrossRef] [PubMed]
Raghavan, S.R.; Hou, J.; Baker, G.L.; Khan, S.A. Colloidal interactions between particles with tethered
nonpolar chains dispersed in polar media: Direct correlation between dynamic rheology and interaction
parameters. Langmuir 2000, 16, 1066–1077. [CrossRef]
Yu, X.; Chen, X.; Chai, Q.; Ayres, N. Synthesis of polymer organogelators using hydrogen bonding as physical
cross-links. Colloid Polym. Sci. 2016, 294, 59–68. [CrossRef]
Li, J.; Harada, A.; Kamachi, M. Sol-gel transition during inclusion complex formation between α-cyclodextrin
and high molecular weight poly(ethylene glycol)s in aqueous solution. Polym. J. 1994, 26, 1019–1026.
[CrossRef]
Li, J.; Li, X.; Ni, X.; Wang, X.; Li, H.; Leong, K.W. Self-assembled supramolecular hydrogels formed
by biodegradable PEO–PHB–PEO triblock copolymers and α-cyclodextrin for controlled drug delivery.
Biomaterials 2006, 27, 4132–4140. [CrossRef] [PubMed]

75

Polymers 2016, 8, 404

30.
31.
32.
33.

34.

35.
36.

Li, J. Self-assembled supramolecular hydrogels based on polymer–cyclodextrin inclusion complexes for drug
delivery. NPG Asia Mater. 2010, 2, 112–118. [CrossRef]
Ye, H.; Owh, C.; Loh, X.J. A thixotropic polyglycerol sebacate-based supramolecular hydrogel showing ucst
behavior. RSC Adv. 2015, 5, 48720–48728. [CrossRef]
Lucey, J.A.; Tamehana, M.; Singh, H.; Munro, P.A. Stability of model acid milk beverage: Effect of pectin
concentration, storage temperature and milk heat treatment. J. Texture Stud. 1999, 30, 305–318. [CrossRef]
Huh, K.M.; Ooya, T.; Lee, W.K.; Sasaki, S.; Kwon, I.C.; Jeong, S.Y.; Yui, N. Supramolecular-structured
hydrogels showing a reversible phase transition by inclusion complexation between poly(ethylene glycol)
grafted dextran and α-cyclodextrin. Macromolecules 2001, 34, 8657–8662. [CrossRef]
Huh, K.M.; Cho, Y.W.; Chung, H.; Kwon, I.C.; Jeong, S.Y.; Ooya, T.; Lee, W.K.; Sasaki, S.;
Yui, N. Supramolecular hydrogel formation based on inclusion complexation between poly(ethylene
glycol)-modified chitosan and α-cyclodextrin. Macromol. Biosci. 2004, 4, 92–99. [CrossRef] [PubMed]
Chee, P.L.; Prasad, A.; Fang, X.; Owh, C.; Yeo, V.J.J.; Loh, X.J. Supramolecular cyclodextrin pseudorotaxane
hydrogels: A candidate for sustained release? Mater. Sci. Eng. 2014, 39, 6–12. [CrossRef] [PubMed]
Kai, D.; Chua, Y.K.; Jiang, L.; Owh, C.; Chan, S.Y.; Loh, X.J. Dual functional anti-oxidant and SPF enhancing
lignin-based copolymers as additives for personal and healthcare products. RSC Adv. 2016, 6, 86420–86427.
[CrossRef]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

76

polymers
Article

The Effect of Thermal History on the Fast
Crystallization of Poly(L-Lactide) with Soluble-Type
Nucleators and Shear Flow
Tianfeng Shen, Piming Ma *, Qingqing Yu, Weifu Dong and Mingqing Chen
The Key Laboratory of Food Colloids and Biotechnology, Ministry of Education, School of Chemical and
Material Engineering, Jiangnan University, 1800 Lihu Road, Wuxi 214122, China; stf0710@126.com (T.S.);
yqq1835@163.com (Q.Y.); wfdong@jiangnan.edu.cn (W.D.); mq-chen@jiangnan.edu.cn (M.C.)
* Correspondence: p.ma@jiangnan.edu.cn; Tel.: +86-510-85917090
Academic Editors: Alexander Böker and Frank Wiesbrock
Received: 14 October 2016; Accepted: 1 December 2016; Published: 10 December 2016

Abstract: The N1 ,N1  -(ethane-1,2-diyl)bis(N2 -phenyloxalamide) (OXA) is a soluble-type nucleator
with a dissolving temperature of 230 ◦ C in poly(L-lactic acid) (PLLA) matrix. The effect of thermal
history and shear flow on the crystallization behavior of the PLLA/OXA samples was investigated
by rheometry, polarized optical microscopy (POM), differential scanning calorimetry (DSC), wide
angle X-ray diffraction (WAXD), and scanning electron microscopy (SEM). The crystallization process
of the PLLA/OXA-240 sample (i.e., pre-melted at 240 ◦ C) was significantly promoted by applying
a shear flow, e.g., the onset crystallization time (tonset ) of the PLLA at 155 ◦ C was reduced from
1600 to 200 s after shearing at 0.4 rad/s for even as short as 1.0 s, while the crystallinity (Xc ) was
increased to 40%. Moreover, the tonset of the PLLA/OXA-240 sample is 60%–80% lower than that of
the PLLA/OXA-200 sample (i.e., pre-melted at 200 ◦ C) with a total shear angle of 2 rad, indicating a
much higher crystallization rate of the PLLA/OXA-240 sample. A better organization and uniformity
of OXA fibrils can be obtained due to a complete pre-dissolution in the PLLA matrix followed by shear
and oscillation treatments. The well dispersed OXA fibrils and flow-induced chain orientation are
mainly responsible for the fast crystallization of the PLLA/OXA-240 samples. In addition, the shear
flow created some disordered α -form crystals in the PLLA/OXA samples regardless of the thermal
history (200 or 240 ◦ C).
Keywords: poly (L-lactide); crystallization; soluble-type nucleator; shear flow; melting process

1. Introduction
Biodegradable and biocompatible poly(L-lactic acid) (PLLA) has recently received more and
more attention [1,2]. It is expected to partially solve the environment issues that associated with
petrochemical materials [3–5]. The main drawbacks of PLLA-based materials are brittleness, low
crystallization rate, low heat resistant temperatures due to the glass transition temperature (Tg ) of
about 55 ◦ C, and low crystallinity after conventional processing. Notably, the low crystallization rate
has restricted the application range of PLLA.
One effective approach to speed up the crystallization of PLLA is by applying nucleating agents
to reduce the nucleating activation energy and simultaneously promote the heterogeneous nucleation
effect to achieve higher crystallinity. Many nucleating agents have been investigated including
talc [6], clay [7], carbon nanotubes [8], and organic additives such as poly(vinylidene fluoride), orotic
acid, N,N-ethylene-bis(12-hydroxylstearamide) (EBH), nucleobases, substituted-aryl phosphate salts
(TMP-5) and N,N  ,N”-tricyclohexyl-1,3,5-benzene-tricarboxylamide (TMC-328), N,N  -bis(benzoyl)
hexanedioic acid dihydrazide (TMC-306), and N1 ,N1  -(ethane-1,2-diyl) bis(N2 -phenyloxalamide)
Polymers 2016, 8, 431

77

www.mdpi.com/journal/polymers

Polymers 2016, 8, 431

(OXA) [9–22]. Among these nucleating agents, TMC-328, TMC-306, and OXA were proven to have
high activity and self-assembly ability [18–22]. However, their solubility and self-assembly behavior
are temperature and environment (e.g., static or dynamic conditions) dependent. As a consequence,
it is still a challenge to optimize the nucleation effect of the self-organized nucleators.
Another approach to speed up the crystallization of semi-crystalline polymers is by using shear
flow which can make polymer chains orientate along the flow direction, resulting in plenty of
row nuclei, that thereby enhances crystallization kinetics significantly [23,24]. The flow-induced
crystallization was mainly applied in polyolefin systems such as poly(ethylene) (PE) and
poly(propylene) (PP), and it exists in polymer processing such as injection molding, extrusion, and film
blowing [25–27]. Although the environmentally friendly PLLA has been commercialized for more than
one decade, the flow-induced crystallization of PLLA has just received limited attention compared
with polyolefins [28,29]. Furthermore, the effect of nucleating agents in combination with shear flow
on the crystallization of PLLA is even less understood [30].
Rheometry was used in the present work to investigate the crystallization behavior of PLLA
in the presence of a soluble-type nucleator (OXA) and shear flow. The effect of the thermal history
(i.e., melting process) and shear conditions are emphasized and the mechanism of the enhanced
crystallization kinetics is discussed. Therefore, the present work not only provides a fundamental
research on a complex PLLA/OXA system but also offers a possible approach for high performance
PLLA-based products.
2. Experimental Section
2.1. Materials
Poly(L-lactide) (PLLA, 4032D) was purchased from Nature Works LLC, Minnetonka, MN, USA,
with a Mn = 2.1 × 105 g·mol−1 , PDI = 1.7, and a D-lactide content of 2%. The nucleating agent,
N1 ,N1  -(ethane-1,2-diyl)bis(N2 -phenyloxalamide) (OXA), with a melting temperature of 338 ◦ C and
a purity of 98%, was synthesized in the laboratory with the chemical structure as shown in Figure S1.
2.2. Sample Preparation
The PLLA and nucleating agent (OXA) were dried at 60 ◦ C in a vacuum oven for 12 h before
use. Both of the PLLA/OXA (100/0.5 wt/wt) and neat PLLA samples were prepared in a chamber
of a rheometer (HAAKE Polylab-OS, Thermo Fisher Scientific, Bremen, Germany), at 180 ◦ C and
50 rpm for 5 min. Each sample was then compression molded at 180 ◦ C and 10 MPa for 2 min
using a hot compression molding machine and subsequently cooled down with room-temperature
compression plates at a pressure of 5 MPa to make disk-shaped samples with a diameter of 25 mm and
a thickness of 1.0 mm. The disk-shaped samples were further dried in vacuum at 60 ◦ C for 12 h before
the rheological measurements.
2.3. Characterization
Differential Scanning Calorimetry (DSC): The crystallization and melting behaviour of the
samples were studied by using DSC (DSC 8000, Perkin Elmer, Waltham, MA, USA). Each sample was
heated to 200 ◦ C at 10 ◦ C/min, held for 3 min at 200 ◦ C, then cooled to 0 ◦ C and re-heated to 200 ◦ C


0 ×
at 10 ◦ C/min. The crystallinity of PLLA (Xc ) is calculated via Xc = (ΔHm − ΔHcc )/ ω ∗ ΔHm
100% [31], where ΔHm and ΔHcc are the measured melt and cold crystallization enthalpy of the PLLA,
0 = 93.6 J/g is the melting
respectively, ω is the weight fraction of the PLLA in the blends, and ΔHm
enthalpy of 100% crystalline PLLA [32]. All tests were carried out in a nitrogen atmosphere.
Polarized optical microscopy (POM): The crystal morphology of the PLA/OXA samples upon
cooling from the melt (200 and 240 ◦ C, respectively) were monitored with a POM (Axio Scope 1,
Carl Zeiss, Oberkochen, Germany) in combination with a Linkam THMS600 hot-stage. Each sample

78

Polymers 2016, 8, 431

was sandwiched between two carefully cleaned glass slides and was first held at 200 or 240 ◦ C for
3 min and then cooled to room temperatures at 10 ◦ C/min. Images were taken at varied temperatures.
Rheology: Rheological experiments were carried out on a DHR-2 rheometer (TA Instruments,
New Castle, DE, USA) in a plate-plate configuration (25 mm in diameter and 1 mm in gap) to study the
isothermal crystallization of the PLLA and PLLA/OXA samples with and without pre-shear treatment.
The experimental procedures for the shear-induced crystallization are illustrated in Figure 1 and were
performed as follows: (1) the samples were annealed at 200 or 240 ◦ C (T1 ) for 3 min; (2) subjected to
a dynamic temperature sweep with a ramp of −5 ◦ C/min to the desired crystallization temperature
(T2 = 155 ◦ C); (3) a shear pulse with controlled shear rates and shear time was applied on each sample;
(4) an oscillatory time sweep was performed at T2 to trace the evolution of the storage modulus of
the samples upon the isothermal crystallization. The strain and frequency were set at 1% and 1 Hz,
respectively for the oscillatory time sweep.
The PLLA/OXA samples treated at 200 and 240 ◦ C (T1 ) are abbreviated as PLLA/OXA-200 and
PLLA/OXA-240, respectively.
o

240 C

o

Temperature ( C)

T1
o

200 C

o

T2 =155 C
Oscillation

Shear

Time (s)
Figure 1. Procedure indications of thermal and shear applications for the rheological measurement of
the PLLA/OXA samples.

Wide Angle X-ray Diffraction (WAXD): WAXD measurements were carried out by using an X-ray
diffractometer (Bruker AXS D8, Karlsruhe, Germany) equipped with a Ni-filtered Cu Kα radiation
source and with a wavelength of 1.542 Å. The measurements were operated at 40 kV and 40 mA with
scan angles from 5◦ to 50◦ and a scan rate of 3◦ /min.
Scanning Electron Microscopy (SEM): The micro-morphology of the sheared PLLA/OXA
samples was observed by using a SEM (S-4800, Hitachi, Tokyo, Japan) at an accelerating voltage of 2 kV.
The cross sections of each sample obtained by cryo-fracture were etched with a 1:2 water-methanol
mixture containing 0.025 mol/L NaOH and were subsequently coated with a thin gold layer
before observation.
3. Results and Discussion
3.1. Effect of OXA on the Crystallization of the PLLA under Static Conditions
The effect of OXA on the non-isothermal crystallization and melting behavior of PLLA was
studied with DSC, as shown in Figure 2. The corresponding thermal parameters are provided in
the Supporting Information (Table S1). No crystallization traces of the PLLA were detected upon
cooling (Figure 2a) followed by a pronounced cold crystallization peak in the subsequent heating
79

Polymers 2016, 8, 431

scan (Tcc = 112.4 ◦ C, ΔHcc = −28.4 J/g, Figure 2b). Huneault et al. reported a Tc of 103.2 ◦ C for PLLA
with talc as a nucleating agent (cooling at 10 ◦ C/min) [6], while Nam et al. reported a Tc of 110 ◦ C
for EBH-nucleated PLLA (cooling at 2 ◦ C/min) [10]. These results demonstrate a poor crystalline
capability of PLLA due to chain stiffness and the lack of efficient nucleators [33,34]. A high Tc and a
sharp crystallization peak correspond to a high crystallization rate. Therefore, the DSC data in Figure 2
indicate that the OXA could speed up the crystallization of the PLLA with a narrow crystallization
peak (Tc = 116.2 ◦ C, Xc = 34.3%). A multi-melting peak behavior can be resulted from different
crystalline forms or the same crystalline forms with different perfections. It is reported that PLLA can
crystallize in three different forms depending on the crystallization conditions (α, β, γ forms) [35–39].
The α form is the most common in PLLA, while β and γ forms can occur due to special processing
conditions. Actually, the β and γ forms should not exist in the present samples as indicated by the
XRD results (see below with shear-0). Therefore, a double melting peak of PLLA may be associated
with the different crystallization conditions between the α and α (also noted as δ) crystals. When
PLLA was crystallized at temperatures that were more suitable for α crystal formation (e.g., around
Tcc in this work), metastable α crystals were formed, and parts of them transformed into stable α
crystallites with a higher melting temperature upon heating, leading to the double melting behaviors
in the second run of DSC [35].Thus, the double endothermic peaks of PLLA in this work are assigned
to a melting/re-crystallization/re-melting mechanism. The Tm1 of PLLA/OXA is slightly higher
than that of PLLA (Figure 2b) indicating a better organization and uniformity of PLLA crystals in the
presence of OXA.

(b)
Heat Flow Endo Up

Heat Flow Endo Up

(a)
PLLA

PLLA/OXA
Tc
30

60

90
120 150
o
Temperature ( C)

Tm1

PLLA

Tm1
Tcc

PLLA/OXA
30

180

Tm2

60
90
120 150
o
Temperature ( C)

180

Figure 2. Differential scanning calorimetry (DSC) curves of the PLLA and PLLA/OXA samples:
(a) cooling from the melt and (b) subsequent heating processes. The cooling and heating rates are
10 ◦ C/min.

The effect of OXA on the isothermal crystallization of PLLA was studied in the temperature range
of 130–145 ◦ C. Figure 3 shows the relative crystallinity (Xt ) as a function of crystallization time (t) and
the half-life crystallization time (t1/2 ) as a function of temperature (Tc ). The crystallization time of the
PLLA/OXA sample is shorter in comparison with that of the PLLA. Taking Tc = 135 ◦ C as an example,
the t1/2 of neat PLLA was 31 min in comparison with 4.5 min of the PLLA/OXA sample. These results
indicate that the OXA significantly accelerated the isothermal crystallization process of PLLA. On the
other hand, the t1/2 of PLLA/OXA increased with increasing Tc because of the difficulty in nucleation
at high(er) temperatures. It has to be noted that no crystallization of PLLA/OXA occurred within
90 min at 145 ◦ C (see Figure S2). In literature, crystallization of PLLA was not observed at 146 ◦ C even
in the presence of poly (D-lactic acid) as a nucleating agent [40]. Thus, it would be more difficult for
the crystallization of PLLA/OXA at a temperature higher than 145 ◦ C.

80

Polymers 2016, 8, 431

32

(a)

28

60
40

PLLA/OXA 130ć
PLLA/OXA 135ć

20
0

neat PLLA: 31min

30

80
t1/2 (min)

Relative crystallinity (%)

100

10

20 30 40 50 60
Crystallization time (min)

13.5 min

12
8
4.5 min

PLLA/OXA 140ć
neat PLLA 135ć

0

(b)

4
0

70

1.9 min

140

135
130
o
Temperature ( C)

125

Figure 3. (a) The relative crystallinity (Xt ) of PLLA and PLLA/OXA as a function of crystallization
time (t), and (b) the corresponding half-life crystallization time (t1/2 ) obtained at Xt = 50%.

3.2. Self-Organization of the OXA upon Cooling from Different Temperatures
Since rheological responses are sensitive to microstructural changes [41,42], they are used to
investigate the self-organization behavior of OXA in the PLLA melt and the crystallization behavior of
PLLA. It is known that the dissolution temperature of OXA in the PLLA melt is around 230 ◦ C [22].
Figure 4 shows the variation of the storage modulus (G ) of the PLLA/OXA samples upon cooling from
200 and 240 ◦ C, respectively. For the PLLA/OXA-240 sample, two steep increases in storage modulus
are observed around 195 and 145 ◦ C, respectively. Two polarized optical microscopy (POM) images of
the PLLA/OXA-240 samples taken upon cooling are presented as insets in Figure 4a,b, which clearly
confirmed the self-organized OXA fibrillar superstructures. Therefore, the strong increase at 195 ◦ C is
associated with the self-organization process of the dissolved OXA into a non-soluble fibrillar network,
while the increase at 145 ◦ C corresponds to the crystallization of the PLLA matrix. Similar phenomena
were observed in PLA/TMC-306 systems as well [18]. In the case of the PLLA/OXA-200 sample,
an inconspicuous increase of G occurred at around 190 ◦ C followed also by a strong increase around
125 ◦ C. As OXA could only be partially dissolved in the PLLA matrix at 200 ◦ C (Image c), the former
increase of G associated with the self-organization of some dissolved OXA is not obvious. The increase
at 125 ◦ C also resulted from the crystallization of the PLLA matrix. It is noticed that the crystallization
temperature of the PLLA/OXA-240 sample is 20 ◦ C higher than that of the PLLA/OXA-200 sample.
Apparently, the melting process is an important factor in the crystallization of PLLA/OXA, which is
studied further in the presence of shear flow (see below).
PLLA/OXA-200
PLLA/OXA-240

PLLA Crys.
6

b

5

O

10

XA

a
lfSe

bl
em
ss

4

10

y

Storage modulus (Pa)

10

a

c

3

10

2

10

120

140

160

180
200
o
Temperature ( C)

220

240

Figure 4. Storage modulus versus temperature for the PLLA/OXA samples upon cooling from
200 and 240 ◦ C, respectively. Three POM images performed under static conditions are shown:
(a) PLLA/OXA-240 at 240 ◦ C; (b) PLLA/OXA-240 at 180 ◦ C; and (c) PLLA/OXA-200 at 200 ◦ C.

81

Polymers 2016, 8, 431

3.3. Effect of the Melting Process and Shear Flow on the Crystallization Behaviors of the PLLA/OXA Samples
Effect of shear rate. Figure 5a shows the G evolutions of the PLLA/OXA-240 samples during
the shear-induced isothermal crystallization at 155 ◦ C. A series of shear rates (γ = 0.1, 0.2, 0.3, 0.4,
and 0.5 rad/s) were examined while the overall shear angle was fixed at 2 rad by adjusting the shear
time. The data for the non-sheared sample (γ = 0.0 rad/s) is plotted for comparison in Figure 5a.
For the non-sheared sample, the G slowly rose with time. Impressively, the curve of G ~time of the
PLLA/OXA-240 samples shifted to a shorter time side rapidly when a shear was applied, regardless
of the shear rate. These results indicate a faster overall crystallization rate of the PLLA/OXA sample
after melting at 240 ◦ C and shearing at 155 ◦ C.
The inflection point of G ~time is defined as the onset crystallization time (tonset ), which is plotted
as a function of shear rate for the PLLA/OXA samples that cooled from both temperatures, as shown
in Figure 5b. The tonset of the PLLA/OXA-240 samples is reduced from 1600 s to around 300 s by
increasing the shear rate from 0 to 0.1 rad/s and then leveled off, indicating a remarkable acceleration
of the crystallization kinetics. The acceleration is mainly contributed by the promoted nucleation
process because the crystallization was performed at the same temperature (155 ◦ C) and the crystal
growth rate of a polymeric material is usually the same at a certain temperature. Figure 5b also shows
that the tonset of the PLLA/OXA-240 samples is much shorter than that of the PLLA/OXA-200 samples
at the same shear rate, and the differences between the tonset values of the PLLA/OXA-240 and the
PLLA/OXA-200 samples are larger at lower shear rates. The non-sheared PLLA/OXA-200 sample did
not show inflection point (tonset ) within a couple of hours (data are not shown here). Therefore, it can
be concluded from these results that a high(er) melting temperature (e.g., 240 ◦ C) and a shear flow are
both beneficial to the fast crystallization of PLLA/OXA samples.

(a)
10

1600

1600

4

10

tonset
0

1000

2000
Time (s)

0 rad/s-20s
0.1 rad/s-20s
0.2 rad/s-10s
0.3 rad/s-6.7s
0.4 rad/s-5s
0.5 rad/s-4s

3000

1093
862

800

603
313

400

4000

PLLA/OXA-240
PLLA/OXA-200

1538

1200

5

10

tonset (s)

Storage modulus (Pa)

(b)

PLLA-OXA-240

6

0.0

0.1

485
241

215

194

189

0.2 0.3 0.4
Shear rate (rad/s)

0.5

Figure 5. (a) Storage modulus of the PLLA/OXA-240 samples as a function of the crystallization time at
155 ◦ C and the shear rate; (b) the tonset values of the PLLA/OXA samples as a function of the shear rate.

The crystallized PLLA/OXA-240 samples were collected after the rheological experiments for
differential scanning calorimetry (DSC) characterization, shown in Figure S3 and Table S2. It was
found that the crystallinity of the PLLA increased monotonically with increasing shear rate, while the
melting temperature remained constant.
Effect of shear time. A shear rate of 0.4 rad/s was selected to further study the effect of shear
time on the crystallization of the PLLA/OXA samples. Figure 6a shows the variations of G of the
PLLA/OXA-240 samples as a function of the shearing time and the crystallization time at 155 ◦ C.
The corresponding tonset as a function of shear time is plotted in Figure 6b. The sheared sample
(0.4 rad/s for 0.5 s) shows a much sharper rise of G in comparison with the non-sheared sample,
demonstrating a rapid overall crystallization process after the shear. However, the G ~time curves did
not shift any more when the shear time was longer than 5 s. It implies that the effective orientation of
microstructures might become dynamically balanced after a critical shear time (tc ). The tc value was
82

Polymers 2016, 8, 431

1–5 s for the PLLA/OXA-240 sample at a shear rate of 0.4 rad/s. The G ~time of the PLLA/OXA-200
samples showed a similar trend as a function of shear time (data are not shown here), whereas the tc
value was 15–20 s at the same shear rate.
Similar to the effect of the shear rate, the tonset of the PLLA/OXA-240 samples reduced by
~90% when a shear time of 1 s was applied, and then leveled off (Figure 6b). In comparison, the
PLLA/OXA-200 sample had a much longer tonset value, notably with a short(er) shear time. Taking a
shear time of 5 s as an example, the tonset of the PLLA/OXA-200 sample was 2 times larger than that
of the PLLA/OXA-240 sample. These results further confirm that a higher melting temperature in
combination with a shear flow can more effectively promote the crystallization process of PLLA.

(a)

PLLA/OXA-240

1600

5

10

0 rad/s-20s
0.4 rad/s-0.5s
0.4 rad/s-1s
0.4 rad/s-5s
0.4 rad/s-10s
0.4 rad/s-20s

4

10

0

1000

2000
Time (s)

3000

tonset (s)

Storage modulus (Pa)

6

10

(b)
603

600

405

400
233

200
0

4000

PLLA/OXA-240
PLLA/OXA-200

1600

194

193

230

208

0

221
191

5

10
15
Shear time (s)

20

Figure 6. (a) Storage modulus of the PLLA/OXA-240 samples as a function of shear time and isothermal
crystallization time at 155 ◦ C and (b) the tonset values of the PLLA/OXA samples as a function of shear
time. The shear rate is fixed at 0.4 rad/s for all samples and the PLLA/OXA-200 sample without shear
did not crystallize within the experimental time span (90 min).

3.4. Crystal Structure and Morphology of the PLLA/OXA Samples
In order to gain deeper insight into the effect of the melt process and shear flow on the
crystallization of PLLA, wide angle X-ray diffraction (WAXD) measurements were carried out and the
diffraction patterns are shown in Figure 7. For the non-sheared PLLA/OXA samples, three diffraction
peaks at 2θ = 16.7◦ , 19.2◦ , and 22.6◦ were detected, correlating to the 200/110, 203, and 105 planes
of PLLA α form crystals, respectively [35,43,44]. Meanwhile, a broad diffraction peak was observed
for both of the non-sheared PLLA/OXA-200 and PLLA/OXA-240 samples, indicating an incomplete
crystallization of the PLLA phase. The peak intensity of the PLLA/OXA-240 sample at 2θ = 16.7◦
was larger than that of the PLLA/OXA-200 sample, indicating a relatively higher crystallinity of the
PLLA/OXA-240 sample. Intriguingly, the diffraction peaks were elevated when a shear flow was
applied, accompanied by the disappearance of the broad diffraction peak. Meanwhile, the diffraction
peaks of the PLLA/OXA samples shifted by 0.3◦ to smaller 2θ positions after applying the shear
flow regardless of the melting temperatures (T1 = 200 or 240 ◦ C). A similar result was observed in
nucleator-modified PLLA fibers where the shift was ascribed to the existence of some disordered
α -form crystals [45,46]. It has been proven that α- and α -form crystals share the same 103 helix chain
conformation and orthorhombic unit cell, but the packing of the side groups in the helical chains of the
α -form crystals is less ordered and looser than that of the α-form crystals [36].

83

Polymers 2016, 8, 431

(200)/(110)

(010)

(203) (015)

shear-0.5-240
shear-0.5-200
shear-0-240
shear-0-200

5

10

15

20
2ș (°)

25

30

35

Figure 7. X-ray diffraction patterns of the PLLA/OXA-200 and PLLA/OXA-240 samples with
and without a shear flow (0.4 rad/s for 5 s) at 155 ◦ C. The samples were taken after the
rheology measurements.

The microstructures of the sheared PLLA/OXA samples taken from different crystallization
periods were studied using SEM, as shown in Figure 8. The OXA molecules in the PLLA melt can
self-organize into fibrils via hydrogen bonding, and are capable of serving as nucleating agents [21].
The self-organized fibrils reassembled into larger needle-like superstructures after a shear flow,
as shown in Figure 8a,c. It was observed that the OXA in the PLLA/OXA-240 sample showed
better self-organization and alignment in comparison with that in the PLLA/OXA-200 sample.
These superstructures were subsequently re-dispersed into even smaller fibrils due to the oscillation
effect during the subsequent isothermal crystallization process (Figure 8b,d). The better dispersed
OXA fibrils provided extra nucleating sites for the accumulation of PLLA crystals, leading to a type of
shish-kebab crystal morphology (area A in Figure 8d). Moreover, the PLLA/OXA-240 sample showed
more and thicker PLLA crystals in comparison with the PLLA/OXA-200 sample (Figure 8b,d).

Figure 8.
SEM images of the PLLA/OXA samples: (a) PLLA/OXA-200 just after shear;
(b) PLLA/OXA-200 after shear and crystallization; (c) PLLA/OXA-240 just after shear and
(d) PLLA/OXA-240 after shear and crystallization. The same shear condition, i.e., 0.4 rad/s for
5 s was applied to each sample and the crystallization is performed under oscillation conditions.
A larger magnification of Images (b,d) (3000×) was used compared with that of Images (a,c) (2000×),
for better visualization purposes.

84

Polymers 2016, 8, 431

3.5. Mechanism Discussion
The above results clearly show that the PLLA/OXA-240 samples have better crystallization ability
than the PLLA/OXA-200 sample under the same shear and crystallization conditions, and are also
better than the non-sheared PLLA/OXA-240 sample. Therefore, a schematic illustration is provided
for the mechanism discussion of shear flow induced crystallization of the PLLA/OXA-240 system
(Figure 9). At 240 ◦ C, the OXA can be melted and well dissolved in the PLLA matrix, as confirmed by
Figure 4a and illustrated in Figure 9a. Upon cooling from 240 to 155 ◦ C, the dissolved OXA molecules
can self-organize more homogeneously than the original OXA aggregates (e.g., cooling from 200 ◦ C)
into fibrils/needle-like superstructures that are capable of serving as nucleating sites which, however,
were randomly distributed. At high temperatures (e.g., 155 ◦ C), PLLA without shear is difficult to
crystallize even in the presence of the OXA fibrils [20–22]. However, the fibrils and PLLA molecules
gradually orientate in the shear direction when a shear flow is applied, see Figure 9c,d. The fibrils
superstructures become finer in dimension and are better dispersed in the PLLA matrix due to the
subsequent oscillation of the rheometer, providing more nucleating sites (Figure 9e). It is believed that
the oriented polymer chains could assemble into a parallel array and form the precursors of primary
nuclei for crystallization [23,24]. Therefore, the crystallization kinetics of the PLLA is promoted both
by the evolution of the OXA superstructures and by a certain extent of the PLLA chain orientation.

Figure 9. Schematic illustration of the enhanced crystallization kinetics of the PLLA/OXA-240 systems
showing the evolution of the OXA superstructures and the orientation of the PLLA macromolecules in
the presence of shear flow. (a) PLLA/OXA melt with dissolved OXA molecules; (b) self-organization
of OXA in the PLLA melt; (c,d) the OXA fibrils and PLLA molecules gradually orientate in the shear
direction; (e) crystallization of PLLA.

4. Conclusions
N1 ,N1  -(ethane-1,2-diyl)bis(N2 -phenyloxalamide) (OXA) was identified as a soluble-type
nucleator for poly(L-lactic acid) (PLLA). In the present work, both OXA and the shear flow were applied
to accelerate the crystallization of PLLA, at two different melt annealing temperatures (200 and 240 ◦ C).
The effect of melting temperature and shear flow on the crystallization of the PLLA/OXA samples at
155 ◦ C was investigated by using rheometry, polarized optical microscopy (POM), differential scanning
calorimetry (DSC), wide angle X-ray diffraction (WAXD) and scanning electron microscopy (SEM).
As a result, the crystallization of the PLLA/OXA-240 sample was significantly sped up by even a
gentle shear flow, e.g., the onset crystallization time (tonset ) of the PLLA could be reduced by ~90%
with a shear flow as small as 0.4 rad, while the crystallinity (Xc ) reached 40%.

85

Polymers 2016, 8, 431

Compared with the PLLA/OXA-200 sample, the tonset of the PLLA/OXA-240 sample was reduced
by 60%–80% under the same shear conditions (a total shear angle of 2 rad). Therefore, the higher
melting temperature (240 ◦ C) does accelerate the crystallization of PLLA in the presence of OXA and
shear flow. OXA can be dissolved completely in the PLLA matrix at 240 ◦ C, while only partially
dissolved at 200 ◦ C. A better organization and uniformity of the OXA superstructures can be achieved
due to the complete pre-dissolution in the PLLA matrix and a subsequent shear and oscillation
treatment. The well dispersed OXA fibrils and shear flow induced PLLA chain orientation are
responsible for the fast crystallization of the PLLA/OXA-240 samples. In addition, the X-ray diffraction
patterns showed that the shear flow created some disordered α -form crystals in the PLLA/OXA
samples regardless of the melting temperatures (200 or 240 ◦ C). The new findings in this work may be
applicable to other OXA-nucleated polymeric systems as well, and thus may expand the application
range of OXA and PLLA.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4360/8/12/431/s1.
Figure S1: The chemical structure of the OXA, Figure S2: DSC heat flow as a function of isothermal crystallization
time and temperatures for the PLLA/OXA (100/0.5 wt/wt) samples, Figure S3: First heating DSC curves of the
sheared PLLA/OXA-240 samples after the rheological experiments at 155 ◦ C, Table S1: Thermal parameters of
the PLLA and PLLA/OXA samples obtained from the DSC cooling and 2nd heating scans, Table S2: Thermal
parameters of the PLLA/OXA-240 samples derived from Figure S3.
Acknowledgments: This work was supported by the Natural Science Foundation of Jiangsu Province
(BK20130147), the National Natural Science Foundation of China (51573074, 51303067), and the Fundamental
Research Funds for the Central Universities (JUSRP51624A).
Author Contributions: Tianfeng Shen and Qingqing Yu performed experiments, wrote the manuscript and
prepared the figures and tables. Piming Ma initiated and guided the work and revised the manuscript. Weifu Dong
and Mingqing Chen provided discussion and suggestions on this work.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.
4.
5.
6.
7.
8.

9.
10.

11.

Rasal, R.M.; Janorkar, A.V.; Hirt, D.E. Poly(lactic acid) modifications. Prog. Polym. Sci. 2010, 35, 338–356.
[CrossRef]
Reddy, M.M.; Vivekanandhan, S.; Misra, M.; Bhatia, S.K.; Mohanty, A.K. Biobased plastics and
bionanocomposites: Current status and future opportunities. Prog. Polym. Sci. 2013, 38, 1653–1689. [CrossRef]
Nampoothiri, K.M.; Nair, N.R.; John, R.P. An overview of the recent developments in polylactide (PLA)
research. Bioresour. Technol. 2010, 101, 8493–8501. [CrossRef] [PubMed]
Liu, H.; Zhang, J. Research progress in toughening modification of poly(lactic acid). J. Polym. Sci. Polym. Phys.
2011, 49, 1051–1083. [CrossRef]
Raquez, J.M.; Habibi, Y.; Murariu, M.; Dubois, P. Polylactide (PLA)-based nanocomposites. Prog. Polym. Sci.
2013, 38, 1504–1542. [CrossRef]
Li, H.; Huneault, M.A. Effect of nucleation and plasticization on the crystallization of poly(lactic acid).
Polymer 2007, 48, 6855–6866. [CrossRef]
Krikorian, V.; Pochan, D.J. Unusual crystallization behavior of organoclay reinforced poly(L-lactic acid)
nanocomposites. Macromolecules 2004, 37, 6480–6491. [CrossRef]
Xu, Z.; Zhang, Y.; Wang, Z.; Sun, N.; Li, H. Enhancement of electrical conductivity by changing phase
morphology for composites consisting of polylactide and poly(ε-caprolactone) filled with acid-oxidized
multiwalled carbon nanotubes. ACS Appl. Mater. Interface 2011, 3, 4858–4864. [CrossRef] [PubMed]
Sun, Y.; He, C. Synthesis and stereocomplex crystallization of poly(lactide)-graphene oxide nanocomposites.
ACS Macro. Lett. 2012, 1, 709–713. [CrossRef]
Nam, J.Y.; Okamoto, M.; Okamoto, H.; Nakano, M.; Usuki, A.; Matsuda, M. Morphology and crystallization
kinetics in a mixture of low-molecular weight aliphatic amide and polylactide. Polymer 2006, 47, 1340–1347.
[CrossRef]
Pan, P.; Shan, G.; Bao, Y. Enhanced nucleation and crystallization of poly (L-lactic acid) by immiscible
blending with poly(vinylidene fluoride). Ind. Eng. Chem. Res. 2014, 53, 3148–3156. [CrossRef]

86

Polymers 2016, 8, 431

12.
13.
14.
15.
16.

17.
18.
19.

20.

21.

22.

23.

24.
25.
26.

27.
28.
29.
30.

31.
32.
33.

Brochu, S.; Prud’Homme, R.E.; Barakat, I.; Jerome, R. Stereocomplexation and morphology of polylactides.
Macromolecules 1995, 28, 5230–5239. [CrossRef]
Schmidt, S.C.; Hillmyer, M.A. Polylactide stereocomplex crystallites as nucleating agents for isotactic
polylactide. J. Polym. Sci. Polym. Phys. 2001, 39, 300–313. [CrossRef]
Yamane, H.; Sasai, K. Effect of the addition of poly(D-lactic acid) on the thermal property of poly(L-lactic
acid). Polymer 2003, 44, 2569–2575. [CrossRef]
Qiu, Z.; Li, Z. Effect of orotic acid on the crystallization kinetics and morphology of biodegradable
poly(L-lactide) as an efficient nucleating agent. Ind. Eng. Chem. Res. 2011, 50, 12299–12303. [CrossRef]
Shi, Y.; Shao, L.; Yang, J.; Huang, T.; Wang, Y.; Zhang, N.; Wang, Y. Highly improved crystallization behavior
of poly(L-lactide) induced by a novel nucleating agent: Substituted-aryl phosphate salts. Polym. Adv. Technol.
2013, 24, 42–50. [CrossRef]
Pan, P.; Yang, J.; Shan, G.; Bao, Y.; Weng, Z.; Inoue, Y. Nucleation effects of nucleobases on the crystallization
kinetics of poly(L-lactide). Macromol. Mater. Eng. 2012, 297, 670–679. [CrossRef]
Bai, H.; Zhang, W.; Deng, H.; Zhang, Q.; Fu, Q. Control of crystal morphology in poly(L-lactide) by adding
nucleating agent. Macromolecules 2011, 44, 1233–1237. [CrossRef]
Bai, H.; Huang, C.; Xiu, H.; Zhang, Q.; Fu, Q. Enhancing mechanical performance of polylactide by tailoring
crystal morphology and lamellae orientation with the aid of nucleating agent. Polymer 2014, 55, 6924–6934.
[CrossRef]
Ma, P.; Xu, Y.; Shen, T.; Dong, W.; Chen, M.; Lemstra, P.J. Tailoring the crystallization behavior of
poly(L-lactide) with self-assembly-type oxalamide compounds as nucleators: 1. Effect of terminal
configuration of the nucleators. Eur. Polym. J. 2015, 70, 400–411. [CrossRef]
Ma, P.; Xu, Y.; Wang, D.; Dong, W.; Chen, M. Rapid crystallization of poly(lactic acid) by using tailor-made
oxalamide derivatives as novel soluble-type nucleating agents. Ind. Eng. Chem. Res. 2014, 53, 12888–12892.
[CrossRef]
Shen, T.; Xu, Y.; Cai, X.; Ma, P.; Dong, W.; Chen, M. Enhanced crystallization kinetics of poly(lactide) with
oxalamide compounds as nucleators: Effect of spacer length between the oxalamide moieties. RSC Adv.
2016, 6, 48365–48374. [CrossRef]
Somani, R.H.; Yang, L.; Hsiao, B.S.; Sun, T.; Pogodina, N.V.; Lustiger, A. Shear-induced molecular orientation
and crystallization in isotactic polypropylene: Effects of the deformation rate and strain. Macromolecules
2005, 38, 1244–1255. [CrossRef]
Kumaraswamy, G.; Kornfield, J.A.; Yeh, F.; Hsiao, B.S. Shear-enhanced crystallization in isotactic polypropylene.
3. Evidence for a kinetic pathway to nucleation. Macromolecules 2002, 35, 1762–1769. [CrossRef]
Hsiao, B.S.; Yang, L.; Somani, R.H.; Avila-Orta, C.A.; Zhu, L. Unexpected shish-kebab structure in a sheared
polyethylene melt. Phys. Rev. Lett. 2005, 94, 117802. [CrossRef] [PubMed]
Somani, R.H.; Hsiao, B.S.; Nogales, A.; Srinivas, S.; Tsou, A.H.; Sics, I.; Balta-Calleja, F.J.; Ezquerra, T.A.
Structure development during shear flow-induced crystallization of i-PP: In-situ small-angle X-ray scattering
study. Macromolecules 2000, 33, 9385–9394. [CrossRef]
Peterlin, A. Drawing and extrusion of semi-crystalline polymers. Colloid Polym. Sci. 1987, 265, 357–382.
[CrossRef]
Ghosh, S.; Viana, J.C.; Reis, R.L.; Mano, J.F. Effect of processing conditions on morphology and mechanical
properties of injection-molded poly L-lactic acid. Polym. Eng. Sci. 2007, 47, 1141–1147. [CrossRef]
Yamazaki, S.; Itoh, M.; Oka, T.; Kimura, K. Formation and morphology of “shish-like” fibril crystals of
aliphatic polyesters from the sheared melt. Eur. Polym. J. 2010, 46, 58–68. [CrossRef]
Tang, H.; Chen, J.B.; Wang, Y.; Xu, J.Z.; Hsiao, B.S.; Zhong, G.J.; Li, Z.M. Shear flow and carbon nanotubes
synergistically induced nonisothermal crystallization of poly(lactic acid) and its application in injection
molding. Biomacromolecules 2012, 13, 3858–3867. [CrossRef] [PubMed]
Tsuji, H. In vitro hydrolysis of blends from enantiomeric poly(lactide)s. Part 4: Well-homo-crystallized blend
and nonblended films. Biomaterials 2003, 24, 537–547. [CrossRef]
Södergård, A.; Stolt, M. Properties of lactic acid based polymers and their correlation with composition.
Prog. Polym. Sci. 2002, 27, 1123–1163. [CrossRef]
Saeidlou, S.; Huneault, M.A.; Li, H.; Park, C.B. Poly(lactic acid) crystallization. Prog. Polym. Sci. 2012, 37,
1657–1677. [CrossRef]

87

Polymers 2016, 8, 431

34.

35.
36.

37.
38.
39.
40.

41.

42.

43.
44.

45.

46.

Inkinen, S.; Hakkarainen, M.; Albertsson, A.C.; Södergård, A. From lactic acid to poly(lactic acid) (PLA):
Characterization and analysis of PLA and its precursors. Biomacromolecules 2011, 12, 523–532. [CrossRef]
[PubMed]
Pan, P.; Kai, W.; Zhu, B.; Dong, T.; Inoue, Y. Polymorphous crystallization and multiple melting behavior of
poly(L-lactide): Molecular weight dependence. Macromolecules 2007, 40, 6898–6905. [CrossRef]
Zhang, J.M.; Duan, Y.X.; Sato, H.; Tsuji, H.; Noda, I.; Yan, S.; Ozaki, Y. Crystal modifications and thermal
behavior of poly(L-lactic acid) revealed by infrared spectroscopy. Macromolecules 2005, 38, 8012–8021.
[CrossRef]
Kawai, T.; Rahman, N.; Matsuba, G.; Nishida, K.; Kanaya, T.; Nakano, M.; Nakajima, K. Crystallization and
melting behavior of poly(L-lactic acid). Macromolecules 2007, 40, 9463–9469. [CrossRef]
Puiggali, J.; Ikada, Y.; Tsuji, H.; Cartier, L.; Okihara, T.; Lotz, B. The frustrated structure of poly(L-lactide).
Polymer 2000, 41, 8921–8930. [CrossRef]
Cartier, L.; Okihara, T.; Ikada, Y.; Tsuji, H.; Puiggali, J.; Lotz, B. Epitaxial crystallization and crystalline
polymorphism of polylactides. Polymer 2000, 41, 8909–8919. [CrossRef]
Wei, X.; Bao, R.; Cao, Z.Q.; Yang, W.; Xie, B.; Yang, M. Stereocomplex crystallite network in asymmetric
PLLA/PDLA blends: Formation, structure, and confining effect on the crystallization rate of homocrystallites.
Macromolecules 2014, 47, 1439–1448. [CrossRef]
Luo, F.; Geng, C.; Wang, K.; Deng, H.; Chen, F.; Fu, Q.; Na, B. New understanding in tuning toughness
of β-polypropylene: The role of β-nucleated crystalline morphology. Macromolecules 2009, 42, 9325–9331.
[CrossRef]
Zhong, Y.; Fang, H.; Zhang, Y.; Wang, Z.; Yang, J.; Wang, Z. Rheologically determined critical shear rates for
shear-induced nucleation rate enhancements of poly(lactic acid). ACS Sustain. Chem. Eng. 2013, 1, 663–672.
[CrossRef]
Wasanasuk, K.; Tashiro, K. Crystal structure and disorder in Poly(L-lactic acid) δ form (α form) and the
phase transition mechanism to the ordered α form. Polymer 2011, 52, 6097–6109. [CrossRef]
Tsuji, H.; Tashiro, K.; Bouapao, L.; Hanesaka, M. Separate crystallization and cocrystallization of
poly(L-lactide) in the presence of L-lactide-based copolymers with low crystallizability, poly(L-lactideco-glycolide) and poly(L-lactide-co-D-lactide). Macromol. Chem. Phys. 2012, 213, 2099–2112. [CrossRef]
Zhang, J.; Tashiro, K.; Tsuji, H.; Domb, A.J. Disorder-to-order phase transition and multiple melting behavior
of poly(L-lactide) investigated by simultaneous measurements of WAXD and DSC. Macromolecules 2008, 41,
1352–1357. [CrossRef]
Zhang, H.; Bai, H.; Liu, Z.; Zhang, Q.; Fu, Q. Towards high-performance poly(L-lactide) fibers via tailoring
crystallization with the aid of fibrillar nucleating agent. ACS Sustain. Chem. Eng. 2016, 4, 3939–3947.
[CrossRef]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

88

polymers
Article

Investigation of Thermal and Thermomechanical
Properties of Biodegradable PLA/PBSA Composites
Processed via Supercritical Fluid-Assisted Foam
Injection Molding
Sai Aditya Pradeep 1,2 , Hrishikesh Kharbas 3 , Lih-Sheng Turng 3 , Abraham Avalos 4 ,
Joseph G. Lawrence 4 and Srikanth Pilla 1,2, *
1
2
3
4

*

Department of Automotive Engineering, Clemson University, Clemson, SC 29607, USA;
spradee@g.clemson.edu
Department of Material Science and Engineering, Clemson University, Clemson, SC 29634, USA
Polymer Engineering Center, Department of Mechanical Engineering, University of Wisconsin-Madison,
Madison, WI 53706, USA; kharbas@gmail.com (H.K.); turng@engr.wisc.edu (L.-S.T.)
Polymer Institute, University of Toledo, Toledo, OH 43606, USA; Abraham.Avalos@utoledo.edu (A.A.);
joseph.lawrence@utoledo.edu (J.G.L.)
Correspondence: spilla@clemson.edu; Tel.: +1-864-283-7216

Academic Editors: Alexander Böker and Frank Wiesbrock
Received: 29 November 2016; Accepted: 5 January 2017; Published: 9 January 2017

Abstract: Bio-based polymer foams have been gaining immense attention in recent years due
to their positive contribution towards reducing the global carbon footprint, lightweighting, and
enhancing sustainability. Currently, polylactic acid (PLA) remains the most abundant commercially
consumed biopolymer, but suffers from major drawbacks such as slow crystallization rate and
poor melt processability. However, blending of PLA with a secondary polymer would enhance the
crystallization rate and the thermal properties based on their compatibility. This study investigates the
physical and compatibilized blends of PLA/poly (butylene succinate-co-adipate) (PBSA) processed
via supercritical fluid-assisted (ScF) injection molding technology using nitrogen (N2 ) as a facile
physical blowing agent. Furthermore, this study aims at understanding the effect of blending and
ScF foaming of PLA/PBSA on crystallinity, melting, and viscoelastic behavior. Results show that
compatibilization, upon addition of triphenyl phosphite (TPP), led to an increase in molecular weight
and a shift in melting temperature. Additionally, the glass transition temperature (Tg ) obtained from
the tanδ curve was observed to be in agreement with the Tg value predicted by the Gordon–Taylor
equation, further confirming the compatibility of PLA and PBSA. The compatibilization of ScF-foamed
PLA–PBSA was found to have an increased crystallinity and storage modulus compared to their
physically foamed counterparts.
Keywords: polylactide; poly(butylene succinate-co-adipate); compatibilization; crystallization; foaming

1. Introduction
Thermoplastic foams, as lightweight materials, are extensively used in sectors such as automotive,
packaging, and aerospace due to advantages such as high strength-to-weight ratios, acoustic properties,
low susceptibility to water vapor, superior impact resistance, and low densities [1]. However, a majority
of these foams have precursors that are sourced from crude oil, which is a finite, non-renewable resource
and a major cause of increasing carbon emissions that contribute to anthropogenic climate change.
In the present paradigm, bio-based compostable thermoplastic foams have been gaining ground in
many industries as they help to meet environmental regulations and standards set by international
Polymers 2017, 9, 22

89

www.mdpi.com/journal/polymers

Polymers 2017, 9, 22

and domestic agencies via their application. Polylactic acid (PLA) is an aliphatic polyester that has
emerged as one of the most commercially successful biopolymers due to its transparency, high strength,
and high stiffness, making it superior to many other bio-based polymers [2–5]. However, despite its
several advantages, commercially available PLA has many inherent weaknesses—in particular, its low
toughness, low heat resistance, and brittleness—that prevent it from being widely adopted for durable
applications. More importantly, PLA has poor melt processability due to a narrow processing window
and a very slow crystallization rate. Typically, a higher crystallinity is desirable in finished products
due to its strong influence on mechanical and thermal properties.
Several strategies exist to overcome the slow crystallization rate and melt processability of PLA,
such as the addition of fillers [6,7], copolymerization [8,9], and melt-blending [10–12], all of which offer
an effective medium for enhancing its overall performance. Among these, blending with inherently
toughened (bio)polymers is one of the most effective solutions. However, most of these physical blends
are immiscible in nature and can lead to the overall deterioration of properties [13]. The successful
application of the reactive compatibilization technique has provided enormous opportunities to enhance
the compatibility of blends that are otherwise immiscible and incompatible. Reactive compatibilization
can be achieved via melt-blending of PLA with other suitable polymers, resulting in the formation of
a block or graft copolymer at the interface and reducing the interfacial tension of immiscible polymer
components, thereby promoting interfacial adhesion [13]. While blending PLA with toughened
polymers enables us to overcome some of its aforementioned drawbacks, the foaming of PLA and its
blends is critical to obtaining lightweight, sustainable thermoplastic foams.
In this context, supercritical fluid (ScF)-assisted injection molding, also known as microcellular
injection molding, has been shown to broaden the processing window for biopolymers such as
PLA, as it employs supercritical N2 or CO2 [14,15]. The broadening of the processing window is
because ScF lowers the melt viscosity of the polymer [16,17] due to the formation of a single-phase
polymer/gas solution, enabling the polymer to be processed at lower temperatures [18]. In addition
ScF-assisted injection molding produces foamed components containing micron-sized cells and high
cell densities while consuming a smaller amount of material and energy and having lower cycle times
vs. conventional injection molding [19–21]. Hence, the production of lightweight, sustainable foams
via eco-friendly processing routes necessitates an advanced understanding of the effects of reactive
compatibilization on thermal and viscoelastic response of such foaming systems for enhancing their
commercial application.
Several studies have been conducted on the melt-blending and foaming of PLA with
secondary (bio)polymers, such as poly(ε-caprolactone) (PCL) [22], poly(hydroxybutyrate) (PHB) [23],
polyhydroxybutyrate-valerate (PHBV) [24], poly(butylene succinate) (PBS) [25], poly(butylene
adipate-co-terephthalate) (PBAT) [26], and poly(butylene succinate-co-adipate) (PBSA) [27]. Wu et al. [22]
observed an increase in the crystallinity of compression-molded PLA-PCL blends upon incorporation
of nanofillers. Abdelwahab et al. [23] reported improvement in crystalinity on compatibalized
blends of PLA/PHB processed via compression-molding. Zhao et al. [24,28] prepared ScF-foamed
physical blends of PLA/PHBV blends and composites, and studied their impact on crystallinity
and thermomechanical properties. Yokohara and Yamaguchi [25] found that compression-molded
PLA/PBS blends led to improved crystallinity and enhanced processability. Javadi et al. [26] studied
the miscibility and the thermal and mechanical properties of ScF-foamed physical blends of PLA/PBAT
and found improvements in damping ability. Ojijo et al. [27] studied compression-molded PLA/PBSA
blends compatibilized via triphenyl phosphite (TPP) and observed a significant increase in crystallinity
and thermal stability.
To the best of our knowledge, no studies have been undertaken on analyzing the effect of
compatibilization and foaming on the crystallization and thermomechanical behavior of PLA/PBSA
blends and composites processed via ScF injection molding. Hence, the objectives of this study are to
understand the impact of compatibilization, ScF foaming, and the addition of talc on thermal behavior
and viscoelastic properties. The (70:30) PLA/PBSA ratio was chosen as a model blend to study and

90

Polymers 2017, 9, 22

understand the above-mentioned effects since it exhibited the highest crystallinity and improved
properties, as detailed in Ojijo et al. [29].
2. Materials and Methods
Commercial polylactic acid (PLA) (3001 D) was purchased in pelletized form from Natureworks
LLC (Minnetonka, MN, USA), with specific gravity of 1.24 and melt flow index of ~22 g/10 min.
Commercially available poly (butylene succinate-co-adipate) (PBSA) (Bionolle #3001) pellets were
sourced from Showa Denko (Tokyo, Japan), its specific gravity being 1.23 and melt flow index being
25 g/10 min. Talc used in this study (Mistrocell M90) was supplied by Imerys Talc (San Jose, CA,
USA) with a mean diameter of 18.8 μm. Coupling agent triphenyl phosphite (TPP) was obtained
from Sigma-Aldrich (Milwaukee, WI, USA), and 2 wt % TPP was used to compatibilize blends.
Commercial-grade nitrogen was sourced from Airgas (Greenville, SC, USA) and used as a blowing
agent in ScF-assisted injection molding.
2.1. Methods
A co-rotating twin screw extruder (ZSK 30 from Werner & Pfleiderer, Stuttgart, Baden-Württemberg,
Germany) was used to compound the eight compositions prepared for this study, as listed in Table 1.
Prior to extrusion, as-received PLA and PBSA pellets were dried at 75 ◦ C for 8 h. Subsequently, talc
and/or TPP were manually mixed with PLA/PBSA pellets in weight compositions listed in Table 1.
Except for pure and talc-filled PBSA, extrusion for other compositions was subsequently carried out
at temperature zones of 130/150/165/170/175 ◦ C at a screw rotation speed of 35 rpm. Due to its
low melting point (91 ◦ C), extrusion of pure and talc-filled PBSA was carried out at temperatures of
100/125/135/140/145 ◦ C.
Table 1. Design of experiment (DOE) formulations in this study (ratio) 1 .

1

Sample

Nomenclature

PLA

PBSA

Talc

Pure PLA
Pure PBSA
Physical Blend
TPP Compatibilized blend
PLA + Talc
PBSA + Talc
Physical Blend + Talc
TPP Compatibilized blend +Talc

A
B
P
C
AT
BT
PT
CT

100
70
70
95
70
70

100
30
30
95
30
30

5
5
5
5

Polylactic acid (PLA), poly(butylene succinate-co-adipate) (PBSA), triphenyl phosphite (TPP).

Extruded pellets of all compositions were dried at 80 ◦ C for 8 h prior to injection molding (IM).
Conventional and ScF-assisted IM were carried out using an injection molding machine (Arburg
Allrounder 3205, Lossburg, Baden-Württemberg, Germany), which was equipped with a Trexel Series
II ScF dosing system, Wilmington, MA, USA. Injection molding parameters are listed in Table 2, while
IM melt temperatures were reduced for pure PBSA to 100/140/145/135/125 ◦ C—similar to extrusion
due to its low melting point. However, in the metering zone, temperatures had to be increased to
ensure a consistent pressure drop during gas dosage. Weight % of supercritical N2 was calculated by
Equation (1):
.

.

mtX (27.8)
wt % ScF =
m
.

(1)

where m is the mass flow rate of ScF (kg/h), t is the ScF dosage time (s), m is the shot weight (g), and
27.8 is a conversion factor.
A total of 24 samples (solid IM and ScF IM with two gas dosages, 0.73 and 0.94 wt %)—3 per each
composition listed in Table 1—were prepared. Subsequently, injection-molded specimens were labeled

91

Polymers 2017, 9, 22

as “XX-Y”, where XX corresponds to nomenclature mentioned in Table 1, while “Y” indicates the
nature of the sample as solid or foamed, with “S” referring to the solid injection-molded sample, “1”
referring to the ScF-assisted injection-molded sample obtained at a ScF gas dosage of 0.73 wt %, and
“2” referring to the ScF-assisted injection-molded sample at a gas dosage of 0.94 wt % ScF.
Table 2. Experimental conditions for solid and supercritical fluid assisted injection molding (ScF IM).
Parameter

Solid Molding

Foamed Molding

Back pressure (MPa)
Melt temperatures (◦ C)
Injection pressure (bar)
Injection speed (cm3 /s)
Holding pressure (bar)
Holding time (s)
Cooling time (s)
Gas dosage (wt %)

10
155/165/175/185/195
2500
65
800
3
60
0

80
155/165/175/185/195
2500
65
0
0
60
0.73 and 0.94

The IM samples were characterized using gel permeation chromatography (GPC), differential
scanning calorimetry (DSC) and a dynamic mechanical analyzer (DMA) in order to understand the
effect of physical and chemical compatibilization, the addition of fillers, and the ScF foaming of PLA
and PBSA on thermal and viscoelastic properties.
2.2. Gel Permeation Chromatography
Number-average molecular weight (Mn ) and polydispersity index (PDI) for solid injection-molded
samples were determined via gel permeation chromatography (GPC) on Waters GPC equipped
with a UV–Vis and RI detector. Chloroform was used as an effluent (flow rate of 1.0 mL/min) at
33 ◦ C. All samples were prepared as 0.5% (w/v) solutions in chloroform, with ~50 μL of sample
injected into the GPC. Prior to injection, the dissolved solution was filtered using a 0.2 μm PTFE filter.
Calibration was done using narrow molecular weight polystyrene standards ranging from ~436 to
~990,500 Daltons.
2.3. Differential Scanning Calorimetry
A differential scanning calorimeter (TA Instruments, Q2000, New Castle, DE, USA) was used to
study the crystallization behavior of all 24 samples. About 7–9 mg of sample was taken in hermetically
sealed aluminum pans. Samples were subjected to heating/cooling/heating cycles at 5 ◦ C/min,
beginning with heating from −100 to 200 ◦ C (to remove any thermal history from processing), held
isothermally for 5 min, cooled to −100 ◦ C, and subsequently heated to 200 ◦ C. The temperature of
cold crystallization (Tcc ), the melting temperature (Tm ), the apparent melting enthalpy (ΔHm ), and the
enthalpy of cold crystallization (ΔHcc ) were determined via DSC curves. The crystallinity of PLA and
PBSA were calculated by Equation (2):
χC (% crystallinity) =

ΔHm − ΔHcc
100
×
W
ΔH 0

(2)

where ΔHm (PLA) and ΔHm (PBSA) are the enthalpies of melting per gram of 100% crystal (perfect
crystal) of PLA and PBSA (93.7 and 142 J/g), respectively, and W is the weight fraction of either PLA
or PBSA in the blend [30,31].
2.4. Dynamic Mechanical Analyzer
Dynamic mechanical analysis was carried out using TA Q800 Dynamic Mechanical Analyzer,
New Castle, DE, USA. Rectangular specimen (4 mm × 8 mm × 70 mm) were cut from the gauge length
of injection-molded specimen and tested in dual cantilever mode. Samples were tested at temperatures
between −50 and 100 ◦ C at a heating rate of 3 ◦ C/min at a 1 Hz frequency and a 0.1% strain amplitude
in order to determine glass transition temperature, storage, and loss moduli.
92

Polymers 2017, 9, 22

3. Results
3.1. Gel Permeation Chromatography
Number-average molecular weight (Mn ), weight-average molecular weight (MW ), and polydispersity
index (PDI) were determined for all solid samples and tabulated in Table 3. As can be seen,
(a)
(b)

(c)

(d)

Mn of PLA (A-S) and PBSA (B-S) were obtained as ~90,000 and ~62,000 Daltons, respectively;
While the Mn of the physical blends (P-S) (~64,000 Daltons) was found to be between that of
PLA (A-S) and PBSA (B-S), compatibilized blends (C-S) showed a higher Mn (~101,796 Daltons)
compared to both P-S (by over ~40,000 Daltons) and A-S (by over ~10,000 Daltons);
Addition of talc resulted in a marginal reduction in the Mn of PLA (AT-S) and a marginal
increase in the Mn of all other compositions—namely, BT-S, PT-S, and CT-S—compared to its
non-talc counterparts—within talc-filled compositions, the Mn of PT-S (~79,026 Daltons) was
found to be between those of AT-S (~85,083 Daltons) and BT-S (~66,173 Daltons), while CT-S
(~108,483 Daltons) showed an improvement over all three compositions;
PDI for pure and talc-filled compatibalized blends (C-S and CT-S) was found to be narrower than
that for other compositions.
Table 3. Mn , MW , polydispersity index (PDI), and area for all compositions.
Sample

M n (Daltons)

PDI

A-S
B-S
P-S
C-S
AT-S
BT-S
PT-S
CT-S

90,039
62,175
64,685
101,796
85,083
66,173
79,026
108,483

1.8
2.1
2.1
1.4
1.7
2.0
1.8
1.4

3.2. Differential Scanning Calorimetry
3.2.1. First Heating Thermograms
Temperature of cold crystallization (Tcc ), melting temperature (Tm ), and their respective ethalpies
of cold crystalization (ΔHcc ) and (ΔHm )—as obtained from first heating thermograms—are reported
for pure polymer (Table 4) and for polymer blends (Table 5), respectively.
Table 4. The thermal behavior of injection-molded pure samples obtained from 1st heating thermograms.
Sample

T cc (◦ C)

ΔH cc (J/g)

T m (◦ C)

ΔH m (J/g)

% Crystallinity

A-S
A-1
A-2
AT-S
AT-1
AT-2
B-S
B-1
B-2
BT-S
BT-1
BT-2

97.3
99.6
100.8
90.2
90.9
91.4
-

26.42
26
22.41
19.23
17.63
18.19
-

168.8
168.7
168.7
168.4
168.2
168.2
92.9
92.9
92.9
93.0
92.8
92.9

45.34
46
47.76
43.51
43.04
44.22
47.22
49.99
50.75
49
51
52

20.19
21.34
27.05
25.91
27.11
27.78
33.25
35.20
35.73
34.50
35.91
36.61

Among individual polymer compositions, pure PLA compositions showed Tcc values of ~97.3 ◦ C
(A-S), ~99.6 ◦ C (A-1), and ~100 ◦ C (A-2), respectively. Compared to the physical blends (Tcc ~81 ◦ C),
93

Polymers 2017, 9, 22

chemically compatibilized blends exhibited a significant reduction in Tcc (Tcc ~71 ◦ C). The introduction
of talc in the PLA samples led to a reduction in Tcc compared to pure PLA compositions, such as from
~97.3 ◦ C (A-S) to ~90.2 ◦ C (AT-S) or from ~100 ◦ C (A-2) to ~91.4 ◦ C (AT-2), respectively. In the case of
physical and chemically compatibilized PLA–PBSA blends, the addition of talc did not significantly
alter Tcc compared to non-talc counterparts. All Tcc values observed in the blended samples correspond
to the PLA component, while PBSA samples (both talc-filled and non-talc) did not exhibit any Tcc value.
With regard to melting behavior, while all PLA samples showed a single melting peak at ~168 ◦ C,
all PBSA samples showed a single melting peak at ~92 ◦ C. However, in the case of blended samples,
two melting peaks were observed, one each corresponding to melting temperatures of PBSA and PLA,
respectively. While physically blended samples showed melting peaks at ~92 and ~167 ◦ C, chemically
blended samples showed a shift in both melting peaks to ~88 and ~155 ◦ C. The addition of talc and/or
ScF was not found to result in any significant shift in melting point (Tm ).
With respect to crystallinity, solid PLA (A-S) exhibited a crystallinity of ~20.19%, while its foamed
counterpart (A-2) showed a higher crystallinity of ~27%. A similar increase in crystallinity of PBSA
was observed from ~33% (B-S) to ~35% (B-1 and B-2). While physically blended foamed samples
(P-1, P-2, PT-1, and PT-2) showed a crystallinity of ~15%–18%, their chemically conjugated foamed
counterparts (C-1, C-2, CT-1, and CT-2) exhibited crystallinity levels of ~20%–28%. The addition of
talc was observed to improve crystallinity to varying degrees for all compositions compared to their
non-talc counterparts, both for solid and foamed compositions. For example, while AT-S showed
improvement in crystallinity by ~5% compared to A-S, physical blends showed improvement by ~2%
compared to their non-talc counterparts.
Table 5. Thermal behavior of injection-molded blends obtained from 1st heating thermograms.
Sample

Tcc (◦ C)

ΔH cc (J/g)

TPBSA
(◦ C)
m1

◦
TPLA
m2 ( C)

ΔH m (J/g)

% Crystallinity

P-S
P-1
P-2
PT-S
PT-1
PT-2
C-S
C-1
C-2
CT-S
CT-1
CT-2

81.2
81.1
81.0
81.1
81.0
80.9
71.3
71.4
71.0
72.1
73.8
74.2

21.97
21.07
20.97
18.5
16.29
16.48
25.33
20.868
18.04
22
14.27
16.7

92.8
92.1
92.5
92.8
93.0
93.1
88.0
88.6
88.1
89.0
88.7
89.0

167.8
167.8
167.9
167.8
167.8
167.5
155.6
158.0
155.5
156.8
154.8
155.2

32.14
32.03
31.62
29.85
28.23
29.19
35.83
33.29
34.07
33
33.18
34.27

15.50
16.70
16.23
17.30
18.20
19.37
16.00
18.93
24.40
16.77
28.83
26.78

3.2.2. Second Heating Thermograms
The second heating thermograms for all samples is shown in Figure 1a–d, while the glass transition
temperature (Tg ), the melting temperature (Tm ), and crystallinity levels (%) obtained from these
thermograms are reported for the pure polymers (Table 6) and for the blends (Table 7).
Tg for solid PLA (A-S) was observed to be 63.8 ◦ C with marginal decrease for both foamed
compositions (A-1 and A-2). A similar trend was observed for PBSA, with Tg gradually reducing from
−41.8 ◦ C (solid PBSA or B-S) to lower values for both foamed counterparts (B-1 and B-2). However, Tg
for both physical and chemical blends could not be observed at the ramp rate tested in this study.
Melting point (Tm ) was observed to be 169 ◦ C for all pure PLA compositions (A-S, A-1, and A-2),
albeit with the addition of talc (AT-S, AT-1, and AT-2) resulting in obtainment of bimodal melting
peaks at ~165 and ~171 ◦ C, respectively. However, PBSA showed a consistent single melting peak
of ~92 ◦ C for all PBSA samples (both talc-filled and non-talc). Physically blended solid samples
showed three melting peaks: one at ~94.6 ◦ C (corresponding to PBSA), and bimodal peaks at ~165 and
~170 ◦ C (corresponding to PLA). Chemically compatibilized solid samples exhibited a similar trend,
with melting peaks at ~90.0 ◦ C (corresponding to PBSA), and bimodal peaks at ~158.8 and ~164.1

94

Polymers 2017, 9, 22

◦ C (corresponding to PLA). Interestingly, their foamed counterparts (P-1, P-2, C-1, and C-2) showed
only two peaks at ~93 ◦ C (corresponding to PBSA) and ~169 ◦ C (a single peak corresponding to PLA).
The addition of talc to blends resulted in the obtainment of bimodal peaks (corresponding to PLA) in
physical blends (PT-S, PT-1, and PT-2), in stark contrast to a single melting peak (corresponding to
PLA) in chemical blends (CT-S, CT-1, and CT-2).
With regard to crystallinity, the addition of talc led to an increase in crystallinity of PLA and PBSA
samples by ~4% and ~5%, respectively. The crystallinity of the PLA component in the blends was
observed to enhance by ~4% for both physical and chemically compatibilized blends, with the effect
of talc being more pronounced for compatibilized blends. Chemically compatibilized foamed blends
(C-1, C-2, CT-1, and CT-2) showed higher crystallinity vs. their physically foamed counterparts (P-1,
P-2, PT-1, and PT-2, respectively). However, the enhancement in crystallinity due to the use of ScF
foaming was not as pronounced as that due to the addition of talc, with ~50% crystallinity observed
for talc-filled chemically compatibilized samples (CT-S, CT-1, and CT-2).

(a)

(b)

(c)

(d)

Figure 1. Second heating Differential Scanning Calorimetry (DSC) curves of (a) non-talc pure;
(b) non-talc blend; (c) talc pure; and (d) talc blend compositions.
Table 6. Thermal properties of injection-molded samples obtained in 2nd heating thermograms.
Sample

T g (◦ C)

TPBSA
(◦ C)
m

A-S
A-1
A-2
AT-S
AT-1
AT-2
B-S
B-1
B-2
BT-S
BT-1
BT-2

63.8
63.1
61.9
63.2
63.1
62.7
−41.8
−42.8
−42.1
−42.3
−43.4
−43.0

92.7
92.4
92.7
-

TPLA
m
T m1

(◦ C)

169.4
169.1
168.8
165.6
165.1
165.6
-

T m2 (◦ C)

ΔH m of
PBSA (J/g)

ΔH m of
PLA (J/g)

% Crystallinity
PBSA

% Crystallinity
PLA

171.5
170.6
171.2
-

37.24
43.16
44.37
45.4
53.06
54.29

36.69
39.17
38.78
40.84
43.01
44.35
-

26.22
30.39
31.24
31.97
37.36
38.23

39.15
41.8
41.3
43.58
46.90
47.33
-

95

Polymers 2017, 9, 22

Table 7. Thermal properties of injection-molded samples obtained in 2nd heating thermograms.
Sample

TPBSA
(◦ C)
m

P-S
P-1
P-2
PT-S
PT-1
PT-2
C-S
C-1
C-2
CT-S
CT-1
CT-2

94.65
93.54
93.63
93.45
93.25
93.29
90.03
87.75
88.61
90.67
90.96
90.28

TPLA
m
T m1 (◦ C) T m2 (◦ C)
165.6
169.33
169.23
164.48
164.20
164.40
158.81
160.39
160.97
160.23
163.54
162.71

170.10
170.62
170.06
171.06
164.14
-

ΔH m of
PBSA (J/g)

% Crystallinity
PBSA

ΔH m of
PLA (J/g)

% Crystallinity
PLA

15.46
12.66
13.45
13.17
13.76
14.64
11.13
9.39
10.67
9.916
9.961
9.54

36.6
23.80
24.47
30.91
32.30
34.36
34.51
22.042
25.04
23.27
23.38
22.39

28.53
25.11
25.21
27.74
27.90
26.83
34.51
30.39
30.74
35.76
34.86
34.49

43.49
38.28
38.43
40.29
42.53
42.90
42.90
46.33
46.86
54.52
53.14
52.58

3.3. Dynamic Mechanical Analysis
Viscoelastic behavior of all samples was studied using DMA to track temperature dependence of
storage modulus and tanδ. Figure 2a–d show storage modulus curves as a function of temperature,
while Figure 3a–d show dependence of tanδ on temperature. All samples exhibited a decline in
storage modulus with an increase in temperature. While a plateau region was observed for all PLA
samples up to its Tg of ~63 ◦ C, similar plateau regions were not observed for the blends and PBSA
samples. Solid and foamed physical blends (P-S, P-1, and P-2) exhibited glass transition at ~62 ◦ C
(corresponding to PLA), while compatibilized blends (C-S, C-1, and C-2) exhibited a shift in Tg to
~53 ◦ C (corresponding to PLA). Tg values for all PBSA samples (B-S, B-1, B-2, BT-S, BT-1, and BT-2)
was the same as that of PBSA at ~−40 ◦ C.

(a)

(b)

(c)

(d)

Figure 2. Storage modulus of (a) non-talc pure; (b) non-talc blend; (c) talc pure; and (d) talc
blend compositions.

96

Polymers 2017, 9, 22

(a)

(b)

(c)

(d)

Figure 3. Tanδ of (a) non-talc pure; (b) non-talc blend; (c) talc pure; and (d) talc blend, compositions.

The storage moduli at −50 and 25 ◦ C for all compositions is reported in Table 8. The storage
modulus at −50 ◦ C was observed to reduce upon the use of ScF for both talc-filled and non-talc PLA
and PBSA samples, with non-talc-based PLA compositions showing a higher storage modulus vs.
non-talc-filled PBSA or blend compositions (Figure 2a,b). Figure 2c,d shows that, in sum, talc-filled
samples (excluding those of PLA) exhibited a higher storage modulus compared to their non-talc
counterparts. While solid physical and chemically compatibilized blends exhibited distinct storage
moduli of ~2500 MPa at −50 ◦ C, microcellular physical blends showed lower storage moduli (2315
and 2028 MPa), while chemically foamed blends showed higher storage moduli. Among foamed
compatibilized blends, non-talc blends at a lower ScF gas dosage (C-1) exhibited the highest storage
modulus among all non-talc blends, while CT-2 showed the highest storage modulus among all
24 samples.
Table 8. Storage moduli of all compositions at −50 and 25 ◦ C.

Sample
A
B
P
C
AT
BT
PT
CT

Storage Modulus (MPa) at −50 ◦ C

Storage Modulus (MPa) at 25 ◦ C

Solid

ScF 1

ScF 2

Solid

ScF 1

ScF 2

3069
2500
2510
2415
2812
2697
2746
2685

2731
2365
2315
2653
2485
2507
2332
2685

2555
2650
2028
2184
2304
2306
2250
3429

3063
292
1776
1709
2611
408
1972
1911

2552
336
1646
1867
2245
355
1578
1826

2403
295
1386
1478
2086
321
1535
2300

97

Polymers 2017, 9, 22

The post-glass transition hump observed in Figure 2a–d is analogous to a cold crystallization
temperature (Tcc ) [32]. While solid PLA samples did not show any Tcc , the foamed PLA samples (A-1
and A-2) exhibited a Tcc of ~108 and ~109 ◦ C, respectively. In contrast, no PBSA sample showed any
Tcc . With regard to the non-talc-filled blend samples, the physically blended samples showed a Tcc at
~96 ◦ C, while chemically compatibilized blends exhibited a lower Tcc at ~89 ◦ C. The addition of talc
was observed to lead to a reduction in Tcc for the PLA samples (AT-1 and AT-2) to ~101 and ~102 ◦ C,
respectively, with a reduction in Tcc for the physically blended samples (~95 ◦ C) and chemically
compatibilized samples (~85 ◦ C). Interestingly, CT-2 did not show any cold crystallization temperature.
Tanδ is the ratio of loss modulus to storage modulus. Table 9 tabulates the glass transition
temperature (Tg ) corresponding to tanδ peaks—as this is often analogous to Tg of the polymer—and
the area under the tanδ curve. As shown in Figure 3a, the Tg of the PLA is ~75 ◦ C; however, with
chemical compatibilization, it was observed to undergo a significant shift to lower temperatures
(~65 ◦ C) (Figure 3b). While the blend compositions showed no Tg corresponding to PBSA, the physical
blends exhibited a Tg of ~72 ◦ C, and compatibilized blends exhibited a relatively low Tg (~64 ◦ C), both
corresponding to the Tg of the PLA. The addition of talc was not found to result in any significant
shift in Tg of any composition based on their tanδ peaks, while the area under the tanδ curve was
observed to reduce for both physical and chemically compatibilized blends compared to the pure
PLA-based compositions.
Table 9. Glass transition temperatures and area under tanδ for all compositions.

Sample
A
B
P
C
AT
BT
PT
CT

T g (◦ C)

Area under tanδ

Solid

ScF 1

ScF 2

Solid

ScF 1

ScF 2

75.1
−27.4
72.9
64.5
72
−28.3
71.2
64.2

71.1
−27.55
71.7
64
72.5
−28.7
71.4
64

71.2
−27.35
71.8
63.7
72.3
−28.9
71.8
64

27.3
9
11.6
12
26.4
8.8
12.48
13.7

26.1
9.9
12.17
13.5
23.2
8.3
12.9
8.1

24.9
10.9
12.35
13.4
25.3
8.1
12.9
14.4

4. Discussion
4.1. Compatibilization Mechanism
Most physical blends of PLA with toughened secondary polymers (including PBSA) are
thermodynamically immiscble [29]. It is common practice to add compatibilizers in order to improve
the compatibility of these immiscible blends. An addition of compatibilizer results in a reduction of
interfacial tension due to the formation of either a block or graft copolymer at interfaces within the
blend, depending on the kind of compatibilizer used [13,33]. For example, an addition of compatibilizers
possessing reactive end groups will result in the formation of block copolymers (with a substantial
increase in Mn ) [34], while an addition of compatibilizers with reactive pendant groups (such as TPP)
will generally result in the formation of graft/branched copolymers [35].
Different researchers have undertaken studies on the effect of compatibilizers with reactive
pendant groups (such as TPP) on polyester-based systems and have proposed two reaction
mechanisms—one by Jacues et al. [35] and the other by Aharoni et al. [36]. These reaction mechanisms
have a strong impact on the compatibilization of polymer blends and their properties. Hence, any
understanding of how the addition of TPP influences the compatibilization of PLA and PBSA in this
study needs to be taken into account. In both of the above-mentioned reaction mechanisms, the first
step is the preferential reaction of hydroxyl end-groups of PLA/PBSA with TPP via the displacement
of one of TPP’s phenoxy groups, as shown in Figure 4a. This leads to the formation of an intermediate

98

Polymers 2017, 9, 22

phosphorus-containing compound (intermediate alkyl diphenyl phosphite). The second step can be
either of the two reaction mechanisms depicted in Figure 4b,c.

(a)

(b)

(c)
Figure 4. Reaction schemes of (a) the initiation of the reaction between triphenyl phosphite (TPP)
and polylactic acid (PLA)/poly(butylene succinate-co-adipate) (PBSA); (b) the propagation reaction
inducing a possible branching mechanism between the hydroxyl ends of PLA/PBSA polymeric chains;
and (c) the propagation reaction inducing a possible chain extension mechanism between the hydroxyl
chain ends of PLA/PBSA polymeric chains. Adapted with permission from [27]. Copyright 2013,
American Chemical Society.

In the first reaction mechanism, the second step involves a multi-substitution reaction of
intermediate alkyl diphenyl phosphite whereby phenoxy groups are replaced with alkyl groups
along with the elimination of phenol, as shown in Figure 4b. It is highly likely that this reaction
continues until phosphorus serves as a binding point for the occurrence of grafting/branching [33].
In contrast, the second mechanism involves ester linkages from polymers, with phenoxy groups of
intermediate product reacting with carboxyl groups of PLA/PBSA (instead of hydroxyl end-groups),
leading to a chain extension without P atoms becoming part of the polymeric chain (Figure 1c).
In all of the above-mentioned reaction schemes, chain extension and/or branching may occur.
In our case, compatibilized blends show a marginal increase in molecular weight (Table 3) compared
to PLA, indicating that branching is a major reaction pathway. This has been observed in a previous
study conducted by Jacues et al. [35], where 2 wt % TPP was used to melt-blend PET/PBT in a ratio

99

Polymers 2017, 9, 22

of 70:30 [35]. The authors observed a small increase in Mn , accompanied by branching of both
polyesters, as proven by an increase in torque oscillations. Harada et al. [37] observed a similar trend
for compatibilized PLA–PBSA blends involving the use of lyscine triisocyanate as a coupling agent,
with cross-linking behavior being reported and accompanied by a small increase in the Mn of the
PLA blends. Further studies involving multi-detector gel permeation chromatography GPC using
viscometry and light scattering might be required to ascertain the exact nature of branching.
4.2. Crystallization & Melting Behavior
Typically, semi-crystalline polymers such as PLA and PBSA can exhibit three kinds of
crystallization behaviors—melt crystallization, cold crystallization, and recrystallization—depending
on the heating/cooling rate adopted. Melt crystallization refers to the formation of crystals during
cooling. Cold crystallization is the ability of amorphous domains to crystallize during heating, while
re-crystallization refers to the reorientation of crystals formed during melt/cold crystallization [6,38].
In our study, the observation during the first heating pertains to the behavior of injection-molded
samples, which were typically subjected to high cooling rates (~200 ◦ C/min), leading to insufficient
time available for crystallization. The second heating cycle erases the prior thermal history of the
samples while subjecting them to a low cooling rate in the first cooling cycle (5 ◦ C/min), and is
indicative of the behavior of the nascent material [14]. Hence, the differences observed in the
behavior of all samples between both heating cycles in this study, such as (1) the occurrence of
cold crystallization only in the first heating cycle; (2) the presence of a single melting peak in the
first cycle vs. double melting peaks in the second heating cycle (both corresponding to PLA) in few
samples; and (3) an enhanced crystallinity of samples after the second heating cycle; all of which can
be attributed to the stark difference in cooling rate.
The presence of Tcc (corresponding to PLA) reported in Tables 4 and 5 during the first heating
cycle and its absence in the second heating cycle was because all amorphous molecular domains had
crystallized in the first cooling cycle upon use of a slow cooling rate (5 ◦ C/min). This is in good agreement
with the observed increase in the crystallinity of blends from the first heating cycle to the second one, as
it indicates that amorphous domains crystallized during the first cooling cycle. Pilla et al. [14] observed
similar behavior in the case of PLA/MWCNT (multi-wall carbon nanotubes) composites.
The absence of cold crystallization (corresponding to PBSA) in PBSA samples and blends could
be due to several factors. First, PBSA molecules tend to undergo a faster rate of crystallization during
cooling, leading to an absence of amorphous domains that could crystallize during reheating [39].
Second, in the case of blend samples, the presence of stiff PLA chains hinders the cold crystallization
of PBSA [29], further making its occurrence impossible in blends. With regard to blends, the physical
PLA/PBSA blends showed a reduction in Tcc compared to the pure PLA samples, which could be
attributed to the possible intermingling of chains of both polymers at the interfaces, resulting in the
early onset of crystallization [29]. A further decrease in Tcc was observed for chemically compatibilized
blends to ~71 ◦ C, which could be attributed to the enhanced compatibility between PLA and PBSA
chains [27].
The reduction in melting temperatures in the compatibilized blend of around ~7 ◦ C in both
heating cycles was due to a stronger interaction between PLA and PBSA chain segments upon the
addition of triphenyl phosphite (TPP), as TPP enhances the mobility of PLA chain segments [27].
This finding is in good agreement with Ojijo et al. [27], who observed a similar lowering in the Tm (to
~152 ◦ C) of compatibilized PLA/PBSA blends prepared via use of similar coupling agents.
Furthermore, solid blends (P-S and C-S) exhibited double melting peaks that were due to the
melting of PLA crystals with different morphologies [15]. Ojijo et al. [29] had observed that PBSA in
molten form has a nucleating effect on the crystallization of PLA, forming crystals of different sizes
and morphologies. Hence, the observed double melting peaks was mainly due to the nucleating effect
of PBSA. This is due to the inability of simultaneous crystallization of both polymers occurring due
to the large difference in their melting temperatures. However, their foamed counterparts (P-1, P-2,

100

Polymers 2017, 9, 22

C-1, and C-2) showed only one melting peak corresponding to a melting of PLA. This indicates that
foaming had a strong impact on the reorientation of crystal structures, leading to the formation of
highly ordered crystals, even as TPP induced strong compatibilization between PLA and PBSA.
The addition of talc also resulted in the obtainment of double melting peaks in PLA (AT-S, AT-1,
and AT-2) and physical blends (PT-S, PT-1, and PT-2), which could be due to the heterogeneous
nucleation effect of talc particles resulting in the obtainment of varying crystal sizes, which is in
agreement with other literature [40–42]. Interestingly, compatibilized blends showed only one melting
peak upon the addition of talc—in stark contrast with the above-mentioned observation. This can be
explained by the reinforcing effect of talc, which enhances bulk crystallinity without impacting crystal
size, as observed by Tanniru and Misra et al. [43] for CaCO3 -reinforced PE composites.
The crystallinity of foamed compatibilized blends was higher compared to the physically blended
counterparts, a phenomenon also observed by Yang et al. [44] on PLA–PBSA compatibilized blends,
who attributed this to branching sites acting as nucleation points, leading to a higher probability of
nucleation compared to the physical blends. This is in good agreement with our molecular weight
results, measured by GPC, indicating a possible occurrence of grafting/branching. For both physical
and chemically compatibilized blends, foaming resulted in a higher degree of crystallinity. This could
also be attributed to the biaxial extensional flow of ScF affecting the orientation of polymer molecules
around cell walls due to foaming, leading to strain-induced crystallization, which results in an increase
in the final crystallinity, as observed by Ameli et al. [45]. A similar trend was also observed by
Zhai et al. [46] in using chemical foaming agents to foam polycaprolactone. The addition of talc led to
an increase in crystallinity for most samples, which could be attributed to the nucleating effect of talc.
4.3. Viscoelastic Behavior
The storage modulus is a measure of energy storage and recovery exhibited during cyclic
deformation, reflecting the elastic moduli of a material. In general, the storage modulus of any
given material can be altered via addition of fillers. Generally, an addition of inorganic fillers is known
to enhance the storage modulus of PLA [41,42,47]. However, the opposite trend was observed in the
pure PLA in this study (A-S and AT-S), which could be due to the inability of talc to exhibit a reinforcing
effect. In general, the reinforcing effect of talc is more pronounced in a material exhibiting less stiffness,
as explained by Tanniru and Misra [43], who have observed a similar effect of fillers on polymeric
materials with reduced stiffness. The pure PLA used in this study exhibited a storage modulus of
3050 MPa at 40 ◦ C, which is far higher than the storage modulus of both pure PLA (2450 MPa) and PLA
containing 10 wt % of silane-treated wood fiber (2556 MPa) reported by Pilla et al. [39]. This excessively
high storage modulus of pure PLA used in our study might be a contributing factor towards the lack
of any reinforcing effect of talc in the talc-filled PLA samples. However, the opposite trend was
observed for both PBSA-based and blended samples due to the elastomeric nature and resultant lower
stiffness of PBSA, resulting in an improvement in the storage modulus upon the addition of talc.
Among solid blends, compatibilized blends showed a lower storage modulus vis-à-vis physical blends,
primarily due to the hindrance in chain movement on account of the possible branching that prevented
chain realignment/packing, as observed by Khonakdar et al. in crosslinked HDPE (High-density
polyethylene) [48]. Similar phenomena was observed by Ibrahim et al. [49] for cross-linked PLA/PCL
(poly(ε-caprolactone)) blends compared to physical PLA/PCL blends, and was attributed to the
creation of voids in the system upon the formation of the crosslinking network. The compatibilized
foamed blends showed a higher storage modulus in this study compared to their physically foamed
counterparts, which could be attributed to the higher crystallinity (observed in Tables 5 and 7) due to
the synergistic effect of TPP and ScF on crystallinity.
With regard to glass transition temperature, the absence of the plateau region in the storage
modulus curve was observed for blend compositions, and can be attributed to the extremely low Tg
value of PBSA (~−40 ◦ C). Similar observations have been made in another study by Ibrahim et al. [49],
where no plateau region was observed in the storage modulus curve of PLA/PCL blends on account of

101

Polymers 2017, 9, 22

a low Tg value of PCL (~−60 ◦ C). Ojijo et al. [29], in their study on PLA/PBSA blends, also observed
similar trends, and attributed the absence of a plateau region to an increased mobility of PBSA chains
above its Tg (~−40 ◦ C), leading to a lowering of blend stiffness.
Storage modulus was observed to undergo a sudden increase after a glass transition,
corresponding to PLA, for all PLA-containing compositions. This increase was analogous to the
cold crystallization from the first heating cycle of DSC, which is in accordance with Zhang et al. [32]
where cold crystallization was observed for both individual and blend compositions (PLA–PHBV
(polyhydroxybutyrate-valerate)–PBS (poly(butylene succinate))) after glass transition. The appearance
of Tcc can be explained by the fact that, for both individual and blend samples, the DMA (Dynamic
Mechanical Analyzer) tests were undertaken on injection-molded samples that possessed low
crystallinity levels due to the use of high cooling rates (as explained in Section 4.2). Such low
crystallinity levels indicated a significant presence of amorphous domains available for crystallization
during heating in DMA, allowing them to crystallize post-glass transition, along with an associated
sudden increase in storage moduli. With regard to blend compositions, the presence of molten PBSA
as nucleating agents acted as an additional factor in enhancing the crystallinity and the subsequent
jump in storage moduli [29].
The trends observed for the glass transition temperature (Tg ) in storage moduli curves and tanδ
curves were in good agreement with each other for all samples (Figures 3 and 4). In the tanδ curve,
a peak was observed in the region where, with increases in temperature, the rate of the decrease in
storage modulus was higher than that of the loss modulus. Temperatures corresponding to the tanδ
peak is often considered as Tg . Interestingly, Tg was not observed for the PBSA component in all
blend samples due to the locking of PBSA chains by hard PLA segments, thus preventing their motion.
Additionally, the use of a lower weight fraction of PBSA meant that a higher share of PBSA chains
were restricted by PLA chain segments, ensuring that no Tg corresponding to PBSA was observed for
blend compositions [29].
Glass transition temperature of blend samples gives us insight into the miscibility of pure polymers
constituting the blends. Tg is typically dependent on the polymer composition of blends, and lies
between the Tg values of pure constituents for a completely miscible blend [50]. To obtain clarity on
the miscibility and effect of TPP on PLA–PBSA blends, a simplified version of the Gordon–Taylor (G-T)
equation (Equation (3)) [51] was applied to Tg obtained from tanδ.
Tg =

W1 Tg1 + kW2 Tg2
W1 + kW2

(3)

Here, Tg1 and Tg2 are the glass transition temperatures of pure components PLA and PBSA,
respectively, while W1 and W2 are the wt % of PLA and PBSA, respectively, and k is a curve-fitting
factor representing the miscibility of the system, with k = 1 indicating the complete miscibility of the
polymers and the lower/higher values of k indicating poor miscibility. Figure 5 depicts Tg of different
blend compositions. Observed Tg values for A-S, P-S, and B-S (~75.1, ~72, and ~−27.4 ◦ C) and A-S,
C-S, and B-S (~75.1, ~64, and ~−27.4 ◦ C) were plotted as the Tg of the talc-filled and ScF-foamed
blends, all of which were found to overlap (Table 8). These observed values were closer to the G-T
curve for k = 0.08 and k = 0.25, where the curve-fitting parameter k showed a value of 0.08 for the
physical blends, indicating the poor miscibility of PLA and PBSA, as they are thermodynamically
immiscible [52]. However, an addition of 2 wt % TPP shifted the Tg of PLA–PBSA blends to around
64 ◦ C, with the k value of 0.25 used to curve fit the G-T equation; this higher value of k indicates the
possibility of enhanced compatibilization.

102

Polymers 2017, 9, 22

Figure 5. Comparison of the experimental and theoretical Tg values of PLA–PBSA blends.

5. Conclusions
Compatibilized blends of PLA and PBSA were successfully processed using TPP via reactive
extrusion and foamed via ScF-assisted injection molding technology. The compatibilization was verified
via an improvement in Mn using GPC, a shift in Tg using DSC, and an improved miscibility as shown by
the G-T equation. Thermal properties of solid and foamed samples, studied using DSC, revealed that the
addition of talc/compatibilizer and the use of ScF foaming had a significant impact on crystallinity, melt,
cold crystallization, and glass transition temperatures. Compatibilized ScF-foamed blends showed an
improvement in crystallinity by ~10% over their physical blend unfoamed counterparts. The viscoelastic
properties of the samples revealed further evidence of compatibilization, as verified by the G-T
equation. Furthermore, compatibilized foamed blends showed superior storage moduli compared to
their physically foamed counterparts due to the synergistic effect of TPP and ScF on crystallinity.
Acknowledgments: The authors are grateful to Kimberly Ivey for her assistance in conducting TGA, DSC, and
DMA tests. Sai Aditya Pradeep would like to acknowledge the support of the Sonoco Fellowship.
Author Contributions: Sai Aditya Pradeep carried out all experiments and analyses. Srikanth Pilla conceived the
experiments and directed the research. Hrishikesh Kharbas and Lih-Sheng Turng assisted with ScF experiments.
Abraham Avalos and Joseph G. Lawrence assisted with extrusion experiments. All authors participated in
discussions of the research, and Sai Aditya Pradeep and Srikanth Pilla wrote the manuscript.
Conflicts of Interest: The authors declare no conflict of financial interest. The funding sponsors had no role in the
design of the study; in the collection, analyses or interpretation of data; or in the writing of the manuscript.

References
1.
2.
3.
4.
5.

Landrock, A.H. Handbook of Plastic Foams: Types, Properties, Manufacture and Applications; Elsevier: Amsterdam,
The Netherlands, 1995.
Urayama, H.; Kanamori, T.; Kimura, Y. Properties and biodegradability of polymer blends of poly(L-lactide)s
with different optical purity of the lactate units. Macromol. Mater. Eng. 2002, 287, 116–121. [CrossRef]
Urayama, H.; Ma, C.; Kimura, Y. Mechanical and thermal properties of poly(L-lactide) incorporating various
inorganic fillers with particle and whisker shapes. Macromol. Mater. Eng. 2003, 288, 562–568. [CrossRef]
Suyatma, N.E.; Copinet, A.; Tighzert, L.; Coma, V. Mechanical and barrier properties of biodegradable films
made from chitosan and poly (lactic acid) blends. J. Polym. Environ. 2004, 12, 1–6. [CrossRef]
Auras, R.A.; Harte, B.; Selke, S.; Hernandez, R. Mechanical, physical, and barrier properties of poly(lactide)
films. J. Plast. Film Sheeting 2003, 19, 123–135. [CrossRef]

103

Polymers 2017, 9, 22

6.
7.
8.
9.
10.

11.
12.

13.

14.
15.
16.
17.
18.
19.
20.

21.
22.

23.

24.

25.
26.
27.

Pilla, S.; Gong, S.; O’Neill, E.; Yang, L.; Rowell, R.M. Polylactide-recycled wood fiber composites. J. Appl.
Polym. Sci. 2009, 111, 37–47. [CrossRef]
Huda, M.S.; Drzal, L.T.; Misra, M.; Mohanty, A.K.; Williams, K.; Mielewski, D.F. A study on biocomposites
from recycled newspaper fiber and poly(lactic acid). Ind. Eng. Chem. Res. 2005, 44, 5593–5601. [CrossRef]
Grijpma, D.W.; van Hofslot, R.D.A.; Supèr, H.; Nijenhuis, A.J.; Pennings, A.J. Rubber toughening of
poly(lactide) by blending and block copolymerization. Polym. Eng. Sci. 1994, 34, 1674–1684. [CrossRef]
Odent, J.; Leclère, P. Toughening of polylactide by tailoring phase-morphology with P[CL-co-LA] random
copolyesters as biodegradable impact modifiers. Eur. Polym. J. 2013, 49, 914–922. [CrossRef]
Odent, J.; Raquez, J.-M.; Dubois, P. Highly toughened polylactide-based materials through melt-blending
techniques. In Biodegradable Polyesters; Wiley-VCH Verlag GmbH & Co., KGaA: Weinheim, Germany, 2015;
pp. 235–274.
Zeng, J.-B.; Li, K.-A.; Du, A.-K. Compatibilization strategies in poly(lactic acid)-based blends. RSC Adv. 2015,
5, 32546–32565. [CrossRef]
Kfoury, G.; Raquez, J.-M.; Hassouna, F.; Odent, J.; Toniazzo, V.; Ruch, D.; Dubois, P. Recent advances in high
performance poly(lactide): From “green” plasticization to super-tough materials via (reactive) compounding.
Front. Chem. 2013, 1, 32. [CrossRef] [PubMed]
Nagarajan, V.; Mohanty, A.K.; Misra, M. Perspective on polylactic acid (PLA) based sustainable materials for
durable applications: Focus on toughness and heat resistance. ACS Sustain. Chem. Eng. 2016, 4, 2899–2916.
[CrossRef]
Pilla, S.; Kramschuster, A.; Gong, S.; Chandra, A.; Turng, L.-S. Solid and microcellular polylactide-carbon
nanotube nanocomposites. Int. Polym. Process. 2007, 22, 418–428. [CrossRef]
Pilla, S.; Kramschuster, A.; Yang, L.; Lee, J.; Gong, S.; Turng, L.-S. Microcellular injection-molding of
polylactide with chain-extender. Mater. Sci. Eng. C 2009, 29, 1258–1265. [CrossRef]
Kwag, C.; Manke, C.W.; Gulari, E. Rheology of molten polystyrene with dissolved supercritical and
near-critical gases. J. Polym. Sci. Part B 1999, 37, 2771–2781. [CrossRef]
Kwag, C.; Manke, C.W.; Gulari, E. Effects of dissolved gas on viscoelastic scaling and glass transition
temperature of polystyrene melts. Ind. Eng. Chem. Res. 2001, 40, 3048–3052. [CrossRef]
Gong, S.; Yuan, M.; Chandra, A.; Kharbas, H.; Osorio, A.; Turng, L.S. Microcellular injection molding.
Int. Polym. Process. 2005, 20, 202–214. [CrossRef]
Lee, J.; Turng, L.-S. Improving surface quality of microcellular injection molded parts through mold surface
temperature manipulation with thin film insulation. Polym. Eng. Sci. 2010, 50, 1281–1289. [CrossRef]
Chandra, A.; Gong, S.; Yuan, M.; Turng, L.-S.; Gramann, P.; Cordes, H. Microstructure and crystallography in
microcellular injection-molded polyamide-6 nanocomposite and neat resin. Polym. Eng. Sci. 2005, 45, 52–61.
[CrossRef]
Yuan, M.; Turng, L.-S.; Gong, S.; Caulfield, D.; Hunt, C.; Spindler, R. Study of injection molded microcellular
polyamide-6 nanocomposites. Polym. Eng. Sci. 2004, 44, 673–686. [CrossRef]
Wu, D.; Lin, D.; Zhang, J.; Zhou, W.; Zhang, M.; Zhang, Y.; Wang, D.; Lin, B. Selective localization of
nanofillers: Effect on morphology and crystallization of PLA/PCL blends. Macromol. Chem. Phys. 2011, 212,
613–626. [CrossRef]
Abdelwahab, M.A.; Flynn, A.; Chiou, B.-S.; Imam, S.; Orts, W.; Chiellini, E. Thermal, mechanical and
morphological characterization of plasticized PLA–PHB blends. Polym. Degrad. Stab. 2012, 97, 1822–1828.
[CrossRef]
Zhao, H.; Cui, Z.; Sun, X.; Turng, L.-S.; Peng, X. Morphology and properties of injection molded solid and
microcellular polylactic acid/polyhydroxybutyrate-valerate (PLA/PHBV) blends. Ind. Eng. Chem. Res. 2013,
52, 2569–2581. [CrossRef]
Yokohara, T.; Yamaguchi, M. Structure and properties for biomass-based polyester blends of PLA and PBS.
Eur. Polym. J. 2008, 44, 677–685. [CrossRef]
Javadi, A.; Kramschuster, A.J.; Pilla, S.; Lee, J.; Gong, S.; Turng, L.-S. Processing and characterization of
microcellular PHBV/PBAT blends. Polym. Eng. Sci. 2010, 50, 1440–1448. [CrossRef]
Ojijo, V.; Ray, S.S.; Sadiku, R. Toughening of biodegradable polylactide/poly(butylene succinate-co-adipate)
blends via in situ reactive compatibilization. ACS Appl. Mater. Interfaces 2013, 5, 4266–4276. [CrossRef] [PubMed]

104

Polymers 2017, 9, 22

28.

29.

30.
31.
32.

33.
34.
35.

36.

37.

38.
39.
40.
41.
42.

43.
44.
45.

46.

47.
48.
49.

Zhao, H.; Cui, Z.; Wang, X.; Turng, L.-S.; Peng, X. Processing and characterization of solid and microcellular
poly(lactic acid)/polyhydroxybutyrate-valerate (PLA/PHBV) blends and PLA/PHBV/clay nanocomposites.
Compos. Part B 2013, 51, 79–91. [CrossRef]
Ojijo, V.; Sinha Ray, S.; Sadiku, R. Role of specific interfacial area in controlling properties of immiscible
blends of biodegradable polylactide and poly[(butylene succinate)-co-adipate]. ACS Appl. Mater. Interfaces
2012, 4, 6690–6701. [CrossRef] [PubMed]
Pilla, S.; Kramschuster, A.; Lee, J.; Clemons, C.; Gong, S.; Turng, L.-S. Microcellular processing of
polylactide–hyperbranched polyester–nanoclay composites. J. Mater. Sci. 2010, 45, 2732–2746. [CrossRef]
Bureepukdee, C.; Suttiruengwong, S.; Seadan, M. A study on reactive blending of (poly lactic acid) and poly
(butylene succinate co adipate). IOP Conf. Ser. Mater. Sci. Eng. 2015, 87, 12070. [CrossRef]
Zhang, K.; Mohanty, A.K.; Misra, M. Fully biodegradable and biorenewable ternary blends from
polylactide, poly(3-hydroxybutyrate-co-hydroxyvalerate) and poly(butylene succinate) with balanced
properties. ACS Appl. Mater. Interfaces 2012, 4, 3091–3101. [CrossRef] [PubMed]
Koning, C.; van Duin, M.; Pagnoulle, C.; Jerome, R. Strategies for compatibilization of polymer blends.
Prog. Polym. Sci. 1998, 23, 707–757. [CrossRef]
Aycock, D.F.; Ting, S.P. Polyphenylene Ether-Polyamide Blends. U.S. Patent 460,074,1 A, 15 July 1986.
Jacques, B.; Devaux, J.; Legras, R.; Nield, E. Reactions induced by triphenyl phosphite addition during melt
mixing of PET/PBT blends: Chromatographic evidence of a molecular weight increase due to the creation of
bonds of two different natures. Polymer 1997, 38, 5367–5377. [CrossRef]
Aharoni, S.M.; Forbes, C.E.; Hammond, W.B.; Hindenlang, D.M.; Mares, F.; O’Brien, K.; Sedgwick, R.D.
High-temperature reactions of hydroxyl and carboxyl PET chain end groups in the presence of aromatic
phosphite. J. Polym. Sci. Part A 1986, 24, 1281–1296. [CrossRef]
Harada, M.; Ohya, T.; Iida, K.; Hayashi, H.; Hirano, K.; Fukuda, H. Increased impact strength of
biodegradable poly(lactic acid)/poly(butylene succinate) blend composites by using isocyanate as a reactive
processing agent. J. Appl. Polym. Sci. 2007, 106, 1813–1820. [CrossRef]
Lee, Y.; Porter, R.S.; Lin, J.S. On the double-melting behavior of poly(ether ether ketone). Macromolecules
1989, 22, 1756–1760. [CrossRef]
Pilla, S.; Gong, S.; O’Neill, E.; Rowell, R.M.; Krzysik, A.M. Polylactide-pine wood flour composites.
Polym. Eng. Sci. 2008, 48, 578–587. [CrossRef]
Jain, S.; Misra, M.; Mohanty, A.K.; Ghosh, A.K. Thermal, mechanical and rheological behavior of poly(lactic
acid)/talc composites. J. Polym. Environ. 2012, 20, 1027–1037. [CrossRef]
Fowlks, A.C.; Narayan, R. The effect of maleated polylactic acid (PLA) as an interfacial modifier in PLA-talc
composites. J. Appl. Polym. Sci. 2010, 118, 2810–2820. [CrossRef]
Cipriano, T.F.; da Silva, A.L.N.; da Fonseca Thomé da Silva, A.H.M.; de Sousa, A.M.F.; da Silva, G.M.;
Rocha, M.G. Thermal, rheological and morphological properties of poly (lactic acid) (PLA) and talc composites.
Polímeros 2014, 24, 276–282.
Tanniru, M.; Misra, R.D.K. On enhanced impact strength of calcium carbonate-reinforced high-density
polyethylene composites. Mater. Sci. Eng. A 2005, 405, 178–193. [CrossRef]
Yang, X.; Xu, H.; Odelius, K.; Hakkarainen, M. Poly(lactide)-g-poly(butylene succinate-co-adipate) with high
crystallization capacity and migration resistance. Materials 2016, 9, 313. [CrossRef]
Ameli, A.; Jahani, D.; Nofar, M.; Jung, P.U.; Park, C.B. Development of high void fraction polylactide
composite foams using injection molding: Mechanical and thermal insulation properties. Compos. Sci. Technol.
2014, 90, 88–95. [CrossRef]
Zhai, W.; Kuboki, T.; Wang, L.; Park, C.B.; Lee, E.K.; Naguib, H.E. Cell structure evolution and the
crystallization behavior of polypropylene/clay nanocomposites foams blown in continuous extrusion.
Ind. Eng. Chem. Res. 2010, 49, 9834–9845. [CrossRef]
Liu, X.; Wang, T.; Chow, L.C.; Yang, M.; Mitchell, J.W. Effects of inorganic fillers on the thermal and
mechanical properties of poly(lactic acid). Int. J. Polym. Sci. 2014, 2014, 827028. [CrossRef] [PubMed]
Khonakdar, H.A.; Morshedian, J.; Wagenknecht, U.; Jafari, S.H. An investigation of chemical crosslinking
effect on properties of high-density polyethylene. Polymer 2003, 44, 4301–4309. [CrossRef]
Ibrahim, A.N.; Wahit, M.U.; Yussuf, A.A. Effect of fiber reinforcement on mechanical and thermal properties
of poly(ε-caprolactone)/poly(lactic acid) blend composites. Fibers Polym. 2014, 15, 574–582. [CrossRef]

105

Polymers 2017, 9, 22

50.
51.
52.

Krutphun, P.; Supaphol, P. Thermal and crystallization characteristics of poly(trimethylene terephthalate)/
poly(ethylene naphthalate) blends. Eur. Polym. J. 2005, 41, 1561–1568. [CrossRef]
Gordon, M.; Taylor, J.S. Ideal copolymers and the second-order transitions of synthetic rubbers. I.
non-crystalline copolymers. J. Appl. Chem. 2007, 2, 493–500.
Eslami, H.; Kamal, M.R. Effect of a chain extender on the rheological and mechanical properties of
biodegradable poly(lactic acid)/poly[(butylene succinate)-co-adipate] blends. J. Appl. Polym. Sci. 2013, 129,
2418–2428. [CrossRef]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

106

polymers
Article

All-Inorganic Intumescent Nanocoating Containing
Montmorillonite Nanoplatelets in Ammonium
Polyphosphate Matrix Capable of Preventing
Cotton Ignition
Jenny Alongi 1, * and Federico Carosio 2
1
2

*

Dipartimento di Chimica, Università degli Studi di Milano, Via Golgi 19, 20133 Milano, Italy
Dipartimento di Scienza Applicata e Tecnologia, Politecnico di Torino, Alessandria Site,
Viale Teresa Michel 5, 15121 Alessandria, Italy; federico.carosio@polito.it
Correspondence: jenny.alongi@unimi.it; Tel.: +39-02-50314108

Academic Editor: Alexander Böker
Received: 29 October 2016; Accepted: 8 December 2016; Published: 10 December 2016

Abstract: In the present manuscript a new concept of completely inorganic intumescent flame
retardant nanocoating comprised of sodium montmorillonite nanoplatelets embedded in an
ammonium polyphosphate matrix has been investigated using cotton as model substrate. The coating,
deposited by multistep adsorption from diluted water-based suspensions/solutions, homogenously
cover each cotton fibers with average thicknesses below 50 nm and add-on up to 5% in weight.
Combustion characterization evidences the interesting properties: indeed, the so-treated fabrics
reached self-extinguishing during horizontal flame spread tests. Furthermore, when the coating
add-on reaches 5%, no ignition has been observed during cone calorimetry tests under 35 kW/m2
heat flux. Residue analyses pointed out the formation of an expanded all-inorganic coating capable
of greatly improving char formation by exerting barrier function towards volatile release and
heat transfer.
Keywords: cotton; flame retardancy; combustion; intumescence; sodium cloisite; APP

1. Introduction
In recent years, the demand for new and sustainable materials has grown and spread in several
research fields with the aim of replacing old and inefficient materials concepts with innovative and
efficient solutions. In particular, the design of fire safe materials is an area of great concern since the
safety and efficiency of conventionally adopted chemistry have been questioned due to perceived
human and environment hazards (e.g., halogen-based compounds have been found in the food chain,
dangerously ending in the bodies of animals and humans) [1–3]. Initial countermeasures to this
problem have led to restrain the use of some flame-retardants (FRs) and to start a campaign aiming to
evaluate the benefit to danger ratio for the remaining chemicals [4]. In this context, finding non-toxic
and high-performing fire retardant solutions is of great industrial and scientific interest.
Nanotechnology represents a possible tool for achieving such goal. Indeed, in the field of
materials science, nanomaterials have demonstrated to possess superior properties due to the achieved
nanostructures. One clear example of nanostructured materials is represented by polymer-layered
nanocomposites; this class of materials has demonstrated remarkable improvements in mechanical
strength, oxygen barrier properties and flammability with respect to neat polymers [5]. For instance,
the use of lamellar shaped nanoparticles allowed for obtaining peculiar gas barrier and flame retardant
properties; indeed, in the former case, when homogeneously dispersed, nanoparticles would create a
tortuous path capable of slowing down gas molecule diffusions through polymer matrix while in the
Polymers 2016, 8, 430

107

www.mdpi.com/journal/polymers

Polymers 2016, 8, 430

latter they would allow for the formation of an inorganic barrier by cumulating on top of a burning
polymer and reducing the combustion kinetics [6]. Such results were achieved with low amounts of
inorganic filler (usually below 5 wt %) with respect of conventional polymer micro-composites and
were tightly bonded for obtaining a nanostructure in which almost each nanoparticle is isolated
from the other ones. However, the nanostructuring was not easy to achieve as a lot of efforts
during processing and time-consuming modifications of the nanoparticles had to be made in order to
homogeneously disperse the nanoparticles within polymer matrix [7].
Recent literature clearly demonstrates that it is possible to overcome such problems, specifically in
the field of gas barrier and flame retardancy, by changing the approach towards nanostructuring [8,9].
In this concept, nanoparticles are removed from the bulk in order to be deposited or assemble on
the surface, thus resulting in a nanoscale or nanostructured coating capable of greatly improving
the performances of the coated polymer. For instance, a PLA (poly(lactic acid))-film coated with
efficiently organized nanoplatelets (ideally aligned parallel to the surface and perpendicularly to the
gas flux) may reach a oxygen permeability several orders of magnitude lower than that of neat polymer
with a substantial improvement of barrier properties with respect to bulk nanocomposites [10,11].
On the other hand, the surface approach for flame retardancy has been demonstrated as a facile and
straightforward path to impressive results. This path has been aided by different surface modification
techniques based on the deposition from aqueous based suspensions of nanoparticles such as the
Layer-by-Layer assembly (LbL) or the simpler nanoparticle adsorption [9,12,13]. In both approaches,
the substrate is exposed to a nanoparticle suspension in order to have the adsorption on the surface.
This can be performed one time such as in nanoparticle adsorption or multiple times as in LbL assembly.
By relying on the electrostatic attractions occurring between nanoparticles and polyelectrolytes in
water, the latter approach deposits differently charged species at each adsorption step in order to
growth a coating by stacking negatively and positively charged layers [14,15]. Due to the availability
of different nanoparticles and polyelectrolytes, many FR actions have been targeted through years.
The first and most simple coatings were assembled for obtaining a completely inorganic layer made of
nanoparticles; to this aim, sodium montmorillonite or silica nanoparticles have been used on fabrics
(cotton and polyester) or thick bulk polymers (polycarbonate (PC), polyamide (PA) and polyester
(PET)) [16–19]. The collected results showed the efficiency of this approach; indeed, the coatings were
able to suppress the melt-dripping of PC and PET and to considerably slow down the combustion
kinetics of these substrates, often allowing for an unexpected increase in the time to ignition (TTI) [17].
The latter result can be considered unexpected as it is in evidenced contrast with the behavior of
nanoparticles observed in bulk nanocomposites where their inclusion in a polymer matrix almost
systematically reduced TTI [20].
Through years of research, more complicated coating compositions have been experimented and
proven successful [21–23]. Of particular interest are the coatings based on the concept of intumescence;
their main FR action is due to the formation of an expanded charred structure on the surface of the
burning polymer with consequent reduction of heat transmitted from the flame and combustible
volatile release [24,25]. This mimics the FR of classical intumescent coatings that are macroscopically
bigger reaching several hundreds of microns in thickness [26]. Intumescence in nanostructured coatings
can be achieved by ensuring the presence of an acid source, a carbon source and a blowing agent within
the coating structure; a simple example is represented by chitosan (CH)/ammonium polyphosphate
(APP) coatings where CH acts as carbon source and APP provides phosphoric acid and ammonia as
blowing agent [27]. Intumescent coatings normally provided better results than nanoparticle containing
ones; as an example, cotton can achieve self-extinguishment behavior during flammability tests only
with the deposition of a polyallylamine/polyphosphate or starch/polyphosphate coating [28,29].
The FR action of this kind of coatings has been improved by the inclusion of nanoparticles, thus
resulting in a hybrid organic expanded structure reinforced by the inorganic particles. However, if one
flaw has to be pointed out, the presence of the organic expanded part represents a weakness of the

108

Polymers 2016, 8, 430

protective coating as from one hand it can be easily oxidized and from the other hand it possess weak
mechanical properties making it prone to collapsing as a consequence of convective motes.
In the present paper we are targeting this flaw by trying to remove the organic part while
maintaining the intumescent features of the coating. To this aim, we aim to the deposition of a
coating containing sodium montmorillonite (MTM) nanoparticles embedded within an ammonium
polyphosphate continuous matrix in order to produce an inorganic expandable structure. The coating
concept is derived from the practical observation of what occurs when a mixture of the MTM and APP
powders is exposed to a radiant heat flux typical of developing fires. Indeed, the increased temperature
triggers the APP dissociation with the release of ammonia and the production of polyphosphoric
acid that reacts with MTM nanoplatelets forming of a silicoalluminophosphate expanded structure
(see Figure S1 in Supplementary Materials). This and other possible interaction reactions have been
already reported in the literature for APP/MTM containing polymers and have been ascribed as one
of the reasons for the improved flame retardancy achieved by the combination of the two components
with respect to the single constituents [30,31]. Here, we use an easy multi-step adsorption process
for the deposition of APP/MTM coatings in order to improve cotton FR properties. Cotton has been
selected as model substrate for the evaluation of the coating performances because of its hydrophilic
nature and well known degradation process. The morphology of treated and untreated fabrics has
been investigated by electron microscopies, the changes in thermal stability have been evaluated by
thermogravimetric analyses (in inert and oxidative atmospheres) and the achieved FR characteristics
have been assessed by horizontal flame spread tests and cone calorimetry. Finally, the residues collected
at the end of the FR tests have been analyzed and a mechanism for the observed FR behavior has
been proposed.
2. Materials and Methods
2.1. Materials
Cotton with an area density of 100 g/m2 was purchased from Fratelli Ballesio S.r.l. (Torino, Italy).
Prior to deposition, cotton fabrics were washed with water and Marseille soap, ethanol and then
diethyl ether. After the washing steps, fabrics were dried in an oven at 70 ◦ C for 1 h.
The sodium montmorillonite was purchased from Southern Clays Products Inc. (Gonzales, TX,
USA) and employed in 1 wt % water suspension. The suspension was kept under magnetic stirring
for 24 h and then centrifuged at 4400 rpm for 5 min in order to remove aggregates, resulting in a final
concentration of 0.7 wt %.
Ammonium polyphosphate (PHOS-CHEK® P30, purchased from ICL Performance Products Inc.,
Milano, Italy) was used for preparing 1 wt % water solution that was kept under magnetic stirring for
24 h prior to use. The 18.2 MΩ deionized water supplied by a Q20 Millipore system (Milano, Italy)
was employed for both APP solutions and MTM suspensions.
2.2. Coating Deposition
Cotton fabrics were alternately dipped in the APP solution and then in MTM suspension.
In between each adsorption step, the fabrics were squeezed using a Padder model FL300 produced
by Gavazzi S.r.l (Bergamo, Italy) and dried in a convection oven at 80 ◦ C for 30 min, mimicking an
impregnation pad-dry industrial treatment. The dipping time was set at 5 min for the first couple of
adsorption steps; subsequent steps were achieved after 1 min. The alternate dipping/padding/drying
cycle was repeated 2 or 4 times in order to achieve final coating add-ons of 2.5% and 5%, respectively.
In the following, coated samples are coded as % add-on APP/MTM.
2.3. Characterization
Scanning Electron Microscopy: the change in surface morphology of treated cotton with respect
to untreated one was evaluated using a Field-Emission Scanning Electron Microscopy (FE-SEM) on a

109

Polymers 2016, 8, 430

ZEISS, FEG model MERLIN. A LEO-1450VP Scanning Electron Microscope (Carl Zeiss Microscopy
GmbH, Jena, Germany) equipped with an X-ray probe (INCA Energy Oxford, Oxfordshire, UK, Cu-Kα
X-ray source, k = 1.540562 Å) was used to perform elemental analysis (Energy Dispersive Spectroscopy,
EDS). In both cases untreated and treated cotton fabrics were cut (10 × 10 mm2 ), fixed to conductive
adhesive tapes and either chromium- (FE-SEM) or gold- (SEM) metallized prior to imaging.
X-ray diffraction: X-ray diffraction spectra (XRD) were collected on 30 × 30 × 0.5 mm3 samples
with a Philips X’Pert-MPD diffractometer (PANalytical, Eindhoven, The Netherlands, Cu-Ka radiation,
k = 1.540562 Å; step size: 0.02; step time: 2 s).
Thermal stability: A TAQ500 thermogravimetric balance (TA-Instruments, Milano, Italy) was
used for thermogravimetric analyses. The tests were performed from 50 to 800 ◦ C (heating rate
of 10 ◦ C/min) in both nitrogen and air (60 mL/min) on 10 mg samples placed in open alumina
pans. The following parameters were assessed: Tonset 5% (temperature at which a 5 wt % weight
loss is registered), Tmax (temperature at which the maximum weight loss is registered), residue at
Tmax and 800 ◦ C. The experimental error was 0.5% on the weight and 1 ◦ C on the temperature.
Thermogravimetric (TG) and derivative (dTG) curves were reported.
Horizontal flame spread tests: The reaction to the flame application of the prepared samples was
evaluated in horizontal configuration. During the test the sample (100 × 50 mm2 ) was placed in a
metallic frame and tilted 45◦ with respect to its longer axis, then a 20 mm blue methane flame is applied
to its short side for 3 s in order to ignite it. Parameters such as burning time, afterglow times and final
residue were registered during the test. At least three tests were performed for each formulation.
Cone calorimetry: An oxygen consumption cone calorimeter (Fire Testing Technology, FTT) was
employed to investigate the combustion behavior of square samples (100 × 100 mm2 ) under 35 kW/m2
in horizontal configuration. Tests were performed following the ISO 5660 standard implementing the
optimized procedure for textiles described elsewhere [32]. The following parameters were registered:
Time To Ignition (TTI, (s)), peak of Heat Release Rate (pkHRR, (kW/m2 )), Total Heat Release (THR,
(MJ/m2 )) and final residue. At least three tests were performed and standard deviation (σ) was
calculated as experimental error; for non-igniting samples the test was repeated five times.
Fourier Transformed Infrared Spectroscopy in Attenuated Total Reflectance (FT-IR ATR): Spectra
were collected at room temperature (range 4000–700 cm−1 , 16 scans and 4 cm−1 resolution) using a
Frontier FT-IR/FIR spectroscopy (Perkin Elmer, Milano, Italy) in ATR configuration, equipped with a
diamond crystal (depth of penetration 1.66 μm, as stated by the producer).
Raman spectroscopy: Analyses were performed on an InVia Raman Microscope (argon laser
source 514 nm/50 mW, Renishaw S.p.A., Torino, Italy) coupled with a Leica DM 2500 optical
microscope (Leica Microsystems S.r.l., Milano, Italy).
3. Results
3.1. Coating Morphology on Cotton
Coating morphology on cotton fibers has been carefully investigated by FE-SEM observation
combined with XRD diffraction. Figure 1 reports micrographs of uncoated and APP/MTM-coated
cotton and XRD spectra performed on neat MTM powder and 5% APP/MTM sample.
Neat cotton shows the typical morphology of a natural fiber with a rough and irregular surface.
When cotton is coated by APP/MTM, no immediate change in morphology can be detected. Indeed,
by a direct comparison of the uncoated and coated fibers, it is really difficult to establish whether a
coating has been deposited.
The presence of the coating has been revealed by investigating sites where, as a consequence of
deformations, the coating is partially detached from cotton surface.

110

Polymers 2016, 8, 430

Figure 1. FE-SEM (Field-Emission Scanning Electron Microscopy) micrographs of: neat cotton (a); 2.5%
APP/MTM sample (b); 5% APP/MTM sample (c); details of 5% APP/MTM (d,e); and XRD spectra of
neat MTM and 5% APP/MTM fabric (f).

As reported in Figure 1, high magnification micrographs point out the presence of a very thin
coating that averages 35 nm for 5% APP/MTM samples. Similar investigations are possible on 2.5%
APP/MTM and result in thinner and less homogenous coatings, which thickness is more difficult to
evaluate (see Figure S2 in Supplementary Materials).
The weight gain after the deposition was found to be around 2.5 and 5 wt %; furthermore, the
deposited nanocoating had no impact on fabric hand and color. From the micrographs collected in
Figure 1, different information can be gathered: the coating is really thin and when deposited on
cotton fibers is capable of literally reproducing the surface irregularities of the original fibers and MTM
nanoplatelets are adsorbed on the surface with a strong in-plane orientation.
The last statement is also confirmed by XRD measurements. As reported in the interpherogram,
neat MTM shows the characteristic peak at 7.0◦ related of the basal spacing between each MTM
nanoplatelet, which is consistent with the literature [33,34]. APP/MTM coated fabrics displayed a
diffraction peak in the same region indicating that MTM is adsorbed as stacks consisting of several
nanoplatelets laying parallel on the surface and are held together by APP in a “brick and mortar-like
structure”, as schematized in Figure 1 [35].
EDS analyses performed on so-coated samples revealed the presence of elements characteristic of
APP (phosphorous) and MTM (silicon), further confirming the presence of both reagents within the
coating (see Figure S3 in Supplementary Materials). By evaluating the different Si and P percentages, it
is possible to semi-quantitatively estimate the coating composition for each sample; 2.5% APP/MTM
contains 70% APP and 30% MTM while this proportion is basically inverted for 5% APP/MTM which
contains 30% APP and 70% MTM. Such difference might be related to the interactions between the
negative phosphate groups of previously adsorbed APP and the positive edges of adsorbing MTM
with the subsequent release of ammonium ions eventually employed for the production of stacked
MTM layers. Such interaction would be possible only after the deposition of few (2–3) APP layer and

111

Polymers 2016, 8, 430

thus be more apparent after eight deposition steps rather than four. However, while this explanation
seems in accordance with XRD analyses (Figure 1f), in the absence of more detailed characterization
the proposed adsorption mechanism remains a speculation. Undoubtedly, composition difference,
along with the total coating add-on, could play an important role during cotton thermal degradation
and flame retardancy.
3.2. Thermal and Thermo-Oxidative Stability
The thermal and thermo-oxidative stability of untreated and coated samples was assessed by
thermogravimetric analyses in nitrogen and air, respectively. The aim is to obtain useful information
concerning the effects of the deposited coating on the degradation pathways of cotton fabrics. First,
only thermal degradation is evaluated in nitrogen environment. Figure 2 reports TG and dTG curves
of untreated and treated samples and Table 1 collects temperature and weight data obtained from
these analyses.

Figure 2. Thermogravimetric (TG) and derivative (dTG) curves of untreated and treated cotton fabrics
in nitrogen.
Table 1. Thermogravimetric data of untreated and treated samples in nitrogen.
Sample

T *onset5% (◦ C)

T *max (◦ C)

Residue at 800 ◦ C (%)

Cotton
2.5% APP/MTM
5% APP/MTM

327
266
276

369
301
310

11
31
34

* From derivative TG curves.

The weight loss of cotton as a function of the temperature can be mostly related to the thermal
degradation of cellulose, which is well known as well as its mechanism is already established.
In nitrogen, the thermal decomposition of cellulose normally occurs by one step in between 300
and 400 ◦ C and is the result of two competitive pathways, one involving the depolymerization of
glycosyl units to volatile products (mainly levoglucosan, furan and furan derivatives) and the other
involving the decomposition of the same units into thermally stable aromatic char (final residue
evaluated at 800 ◦ C, 11%) [36].
Treated fabrics still show a one-step thermal degradation. However, as clearly observable
from Figure 2 the presence of the coating induces a strong anticipation in the degradation process,
(see Tonset5% and Tmax values in Table 1). This anticipation, more precisely defined “sensitization”,
112

Polymers 2016, 8, 430

is well known and associated to APP decomposition that producing phosphoric acid at high
temperature favors the cellulose decomposition towards char formation [37,38].
Interestingly there is an inverse proportionality between the coating add-on and the anticipation
observed in TG curves; this can be explained by taking into account the different coating compositions,
as previously observed from elemental analysis. Indeed, the P/Si ratio for 2.5% APP/MTM treated
samples is higher than for 5% sample, thus indicating a higher amount of APP within the coating with
a consequent increase of the anticipating effect.
On the other hand, the effect of MTM, the presence of which is proportionally higher in 5%
APP/MTM coatings, also has to be considered. Due to the lamellar chemical nature of these
nanoparticles and the preferential orientation achieved through the deposition as observed in Figure 1,
MTM can easily exert a barrier function towards the release of volatile products, thus balancing
the anticipation ascribed to APP. This hypothesis is in agreement with the literature [35]. The latter
combination of the two counterparts is beneficial and indeed provides the highest residue at Tmax and
at 800 ◦ C, nearly tripling the amount left by neat cotton (compare residues at 800 ◦ C in Table 1).
The thermo-oxidative stability of untreated and treated cotton fabrics has been assessed in air,
thus evaluating the effects of an increasing temperature in an oxidizing environment. Figure 3 reports
TG and dTG curves in air of untreated and treated samples and Table 2 collects temperature and
weight data calculated from the plots.

Figure 3. Thermogravimetric (TG) and derivative (dTG) curves of untreated and treated cotton fabrics
in air.
Table 2. Thermogravimetric data of untreated and treated samples in air.
Sample

T *onset5% (◦ C)

T *max1 (◦ C)

T *max2 (◦ C)

Residue at 400 ◦ C (%)

Residue at 800 ◦ C (%)

Cotton
2.5%
APP/MTM
5% APP/MTM

324
263

350

506

18

2

296

508

39

280

313

4

607

48

5

* From derivative TG curves.

The thermal oxidation of cotton normally takes place in two definite steps: the first one between
300 and 400 ◦ C is related to the formation of both volatiles and al aliphatic char (18%), while the second
is due to the almost complete oxidation of the char with the release of CO and CO2 [39–41]. Similar to
what observed in nitrogen, treated fabrics showed anticipation inversely proportional to the coating
add-on (compare Tonset5% values in Table 2).

113

Polymers 2016, 8, 430

Of particular interest is the coating shielding effect towards oxygen during the second degradation
step; here, the 5% APP/MTM sample maintains 40% residue up to 550 ◦ C and achieves the highest
delay in the second degradation step, as confirmed from Tmax2 value that is shifted to 600 ◦ C with an
increase of 100 ◦ C with respect to unmodified cotton (compare Tmax2 values in Table 2). This effect can
be ascribed to the formation of a protective coating capable of greatly postponing and slowing down
the oxidation of the residue produced during the first degradation step.
3.3. Horizontal Flame Spread Tests
Flammability test in horizontal configuration has been employed to investigate the reaction of
untreated and treated fabrics to a direct flame application. This test assesses the propensity of a material
to initiate a fire and represents a fundamental test in the field of flame retardancy. Figure 4 reports
snapshots of uncoated and coated fabrics during the test and Table 3 reports the collected parameters.

Figure 4. Snapshot of flame-spread tests of untreated and treated cotton fabrics.
Table 3. Flammability data from horizontal flame tests of untreated and treated cotton fabrics.

r

r

Sample

Combustion Rate ± σ (mm/s)

Afterglow

Residue ± σ (%)

Cotton
2.5% APP/MTM
5% APP/MTM

1.7 ± 0.04
1.6 ± 0.30
1.3 ± 0.08

Yes
No
No

0
78 ± 5
85 ± 2

Upon flame application, untreated cotton immediately ignites and burns with flames that, fed by
the combustible degradation products of cotton, spread towards the opposite site of sample at an almost
constant speed. As the flame reaches the end of the sample, it vanishes and the remaining charred
cotton fibers are further consumed by a solid-state oxidation characterized by red incandescence
known as afterglow. The high temperatures reached during the afterglow are still capable of spreading
the fire to other ignitable materials, thus posing an additional, although smaller than flames, risk
to safety.
The coating can significantly modify the burning behavior of cotton. Indeed, both treatments
were able to stop the propagation of the flame achieving a self-extinguishing behavior, thus resulting in
very high residues. In detail, upon flame propagation the coating action as protective barrier and char
formation enhancer reduces the release of volatile combustible products; by this way, the combustion
cannot be sustained anymore and the flame gradually reduces in size being confined to a smaller

114

Polymers 2016, 8, 430

and smaller region where, eventually, it vanishes (see Figure 4). Furthermore, any subsequent flame
application cannot ignite the sample again.
3.4. Cone Calorimetry
Cone calorimetry has been employed in order to evaluate the reaction of uncoated and coated
samples to the exposure to a heat flux. The latter is controlled in order to have heat flux values typical
of early stage developing fires (i.e., 35 kW/m2 ) [42]. When exposed to the heat flux, samples start to
degrade releasing combustible gases that are ignited by a spark positioned above the sample. The time
required for reaching ignition is normally referred as time to ignition (TTI). Then, flaming combustion
starts and the heat released is calculated by evaluating the oxygen consumed during the process.
Heat Release Rate (HRR) plots of untreated and treated samples are reported in Figure 5 together
with a schematization of sample behavior during the test. Table 4 reports the collected numerical data.

Figure 5. Average Heat release rate (HRR) plots of untreated and treated cotton during cone calorimetry
tests and schematic representation of sample behavior under testing.
Table 4. Cone calorimetry data of untreated and treated samples.
Sample
Cotton
2.5%
APP/MTM
5% APP/MTM

r
TTI ± σ (s)

r
pkHRR ± σ (kW/m2 )

r
THR ± σ (MJ/m2 )

r
TSR ± σ (m2 /m2 )

36 ± 2

61 ± 4

1.0 ± 0.1

25 ± 8

0

22 ± 5

38 ± 7

0.38 ± 0.05

13 ± 4

13

N.A. *

N.A. *

N.A. *

40 ± 6

19

Residue (%)

* Parameters related to combustion are not available, as samples did not ignite during test.

Unmodified cotton rapidly ignites after 36 s with a quick combustion and an average pkHRR of
61 kW/m2 , without leaving any residue at the end. The 2.5% APP/MTM shows an anticipation in
ignition as TTI is reduced to 22 s; this can be ascribed to the presence of APP that, similar to what
observed in TGA, releases phosphoric acid that favors cellulose dehydration, thus resulting in an
early production of volatiles with subsequent early ignition. This phenomenon is not considered
detrimental, as by this way the production of charred residue is favored despite volatile release [29].
115

Polymers 2016, 8, 430

Thus, the total heat release and combustion kinetics are significantly reduced, as demonstrated by
HRR plots and THR and pkHRR values reported in Table 4.
Surprisingly, 5% APP/MTM samples showed no ignition at all during this test. Upon exposure to
the heat flux, samples start releasing volatile products while simultaneously producing char, as clearly
observable by a change in color (from white to black). The latter process, improved by the presence of
APP, is combined with the barrier effect exerted by MTM that slows down the gas release. Therefore,
by the combination of these two flame retardant actions, the released combustible volatile products are
not able to reach the concentration needed for ignition, as schematically depicted in Figure 5. This is
also confirmed by TSR (Total Smoke Release) values that show an increase for non-igniting samples
(Table 4).
All treated fabrics yielded a compact and coherent residue at the end of the test (see Figure S4
in Supplementary Materials) that, at the higher coating add-on, maintained the original shape of
the fabric. Interestingly, it was possible to handle the residues without damaging them (e.g., they
could be bent by 180◦ without breaking), thus indicating that they partially maintained the original
mechanical properties.
3.5. Residue Analysis and Coating Mechanism
The residues collected at the end of cone calorimetry tests have been analyzed using SEM,
FT-IR/ATR and Raman spectroscopies in order to obtain information useful for interpreting the results
obtained during flammability and cone calorimetry tests. Figure 6 reports SEM observations performed
on 5% APP/MTM, and IR and Raman spectra of both 2.5% APP/MTM and 5% APP/MTM residues.

Figure 6. Analyses on residues after cone calorimetry: low magnification SEM (Scanning Electron
Microscopy) micrograph performed on unburned (a) and burned (b) areas of 5% APP/MTM; detail of
5% APP/MTM fibers after combustion (c); and FT-IR/ATR and Raman spectra (d,e). As observable
from SEM micrograph, the residues were able to maintain the original texture of the fabric while the
single fibers appear damaged and shrunk with respect to the unburned sample (compare Figure 6a,b).
In addition, from a closer observation, the single fibers appear to be surrounded by a sort of expanded
structure that completely covers them. This structure is resulting from the exposure of the original
APP/MTM coating to a heat flux; EDS analysis confirms the composition by pointing out the presence
of both P and Si elements with C being the main component (see Figure S5 in Supplementary Materials).

116

Polymers 2016, 8, 430

IR and Raman spectroscopies provide additional information concerning the structure of the
carbonaceous structure. Indeed, IR signals related to aromatic char can be found at 1590 cm−1 while
the signal at 1040 cm−1 can be associated to Si–O–Si bonds in MTM nanoplatelets [43]. Furthermore,
the presence of a weak peak at 1700 cm−1 associated to C=O bonds indicates the partial oxidation of
the residues.
Raman spectroscopy gives complementary details concerning the nature of the produced char.
As reported in Figure 6e, both residues show two characteristic peaks (namely, G and D bands at 1590
and 1350 cm−1 , respectively) normally associated to polyaromatic hydrocarbons, further confirming
the aromatic nature of the structures produced during combustion [44,45].
Basing on the achieved results and the mentioned above characterization, it is possible to devise
the coating mechanism. To this aim, the coating reaction to heat or flame application is reported in
Figure 7 where a FE-SEM micrograph of coating (5% APP/MTM)) before combustion (Figure 7a),
SEM micrograph of expanded coating (namely, 5% APP/MTM) after combustion (Figure 7b) and a
schematization of the coating flame retardant action (Figure 7c) are depicted.

Figure 7. Coating reaction to heat or flame application: FE-SEM micrograph of coating (5% APP/MTM))
before combustion (a); SEM micrograph of expanded coating (5% APP/MTM) after combustion (b);
and schematization of the coating flame retardant action (c).

Upon exposure to a heat flux or flame, APP starts to degrade releasing ammonia and producing
phosphoric and polypohosphoric acid [38]. Ammonia swells the MTM stacks deposited within the
coating while polypohosphoric acid simultaneously promotes the char formation of cotton, as well
known from the literature [37]. In addition, besides reacting with cotton for enhanced char production,
APP can also react with MTM nanoplatelets; indeed, clay can favor the dissociation of APP and
the phosphoric acid generated can react with MTM nanoplatelets joining them together with the
production of a silicoalluminophosphate [30,31]. In the case of the coating, the inorganic expanded
structure can also act as physical barrier slowing down mass, oxygen and heat transfer between the
flame and the substrate.
4. Conclusions
In the present manuscript, a new concept of completely inorganic intumescent flame retardant
coating deposited by a multistep adsorption process from diluted water-based ammonium
polyphosphate solutions and sodium montmorillonite suspensions has been explored, using cotton as
117

Polymers 2016, 8, 430

model substrate. The coating covers the surface of each cotton fiber resulting in very thin coatings
with thickness below 50 nm where montmorillonite stacks are embedded in a continuous matrix of
ammonium polyphosphate. This coating efficiently enhanced the char production from cotton fibers,
as observed by thermogravimetric analyses in both nitrogen and air, with impressive results achieved
during combustion tests. The coating resulting in a total add-on of 5% with respect to the original
fabric mass achieved self-extinguishing during horizontal flame spread tests and prevented ignition of
the fabrics when exposed to 35 kW/m2 during cone calorimetry test. The flame retardant action has
been ascribed to the formation of a silicoalluminophosphate expanded structure capable of on the one
hand enhancing the cellulose char-forming ability and on the other hand providing a physical barrier
to mass, oxygen and heat transfer between the flame and the substrate. These results demonstrate
the potentialities of the concept proposed and developed within this paper and provide a starting
point for the further improvements of the coating performances. For instance, the coating stability and
durability need to be improved by employing different cross-linking strategies if an application as
protective treatment for fabrics has to be foreseen.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4360/8/12/430/s1,
Figure S1: Images of MTM (a), APP (b) and mixture of APP/MTM (c) after irradiation under 35 kW/m2 heat flux
in a cone calorimeter. Figure S2: FE-SEM micrographs of defects in 2.5% APP/MTM samples. Figure S3: EDS
analyses performed on 2.5% APP/MTM and 5% APP/MTM samples. Figure S4: Images of residues collected
at the end of cone calorimetry tests: (a) untreated cotton, (b) 2.5% APP/MTM and (c) 5% APP/MTM samples.
Figure S5: EDS analyses performed on 5% APP/MTM residues collected at the end of cone calorimetry tests.
Acknowledgments: The Authors want to thank Fabio Cuttica (for cone calorimetry tests), Alessandro Di Blasio
(for SEM analyses) and Mauro Raimondo (for FE-SEM analyses).
Author Contributions: The present manuscript was written with the equal contribution of the two Authors.
J.A. and F.C. conceived and designed the experiments; F.C. performed the experiments and analyzed the data;
J.A. and F.C. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.

3.
4.

5.
6.
7.
8.
9.
10.

Birnbaum, L.S.; Staskal, D.F. Brominated flame retardants: Cause for concern? Environ. Health Perspect. 2003,
112, 9–17. [CrossRef]
Xiao, H.; Shen, L.; Su, Y.; Barresi, E.; Dejong, M.; Hung, H.; Lei, Y.D.; Wania, F.; Reiner, E.J.; Sverko, E.; et al.
Atmospheric concentrations of halogenated flame retardants at two remote locations: The Canadian high
arctic and the Tibetan plateau. Environ. Pollut. 2012, 161, 154–161. [CrossRef] [PubMed]
Barcelo, D.; Kostianoy, A.G. Handbook environmental chemistry. In Brominated Flame Retardants;
Heidelberg, S.-V.B., Ed.; Springer: New York, NY, USA, 2011; Volume 16.
Stieger, G.; Scheringer, M.; Ng, C.A.; Hungerbuhler, K. Assessing the persistence, bioaccumulation potential
and toxicity of brominated flame retardants: Data availability and quality for 36 alternative brominated
flame retardants. Chemosphere 2014, 116, 118–123. [CrossRef] [PubMed]
Ray, S.S.; Okamoto, M. Polymer/layered silicate nanocomposites: A review from preparation to processing.
Prog. Polym. Sci. 2003, 28, 1539–1641.
Kiliaris, P.; Papaspyrides, C. Polymer/layered silicate (clay) nanocomposites: An overview of flame
retardancy. Prog. Polym. Sci. 2010, 35, 902–958. [CrossRef]
Fernandes, N.J.; Akbarzadeh, J.; Peterlik, H.; Giannelis, E.P. Synthesis and properties of highly dispersed
ionic silica-poly(ethylene oxide) nanohybrids. ACS Nano 2013, 7, 1265–1271. [CrossRef] [PubMed]
Priolo, M.A.; Holder, K.M.; Guin, T.; Grunlan, J.C. Recent advances in gas barrier thin films via layer-by-layer
assembly of polymers and platelets. Macromol. Rapid. Commun. 2015, 36, 866–879. [CrossRef] [PubMed]
Malucelli, G.; Carosio, F.; Alongi, J.; Fina, A.; Frache, A.; Camino, G. Materials engineering for
surface-confined flame retardancy. Mat. Sci. Eng. R 2014, 84, 1–20. [CrossRef]
Laufer, G.; Kirkland, C.; Cain, A.A.; Grunlan, J.C. Clay—Chitosan nanobrick walls: Completely renewable
gas barrier and flame-retardant nanocoatings. ACS Appl. Mater. Interfaces 2012, 4, 1643–1649. [CrossRef]
[PubMed]

118

Polymers 2016, 8, 430

11.

12.
13.
14.

15.
16.
17.

18.
19.

20.
21.

22.

23.

24.
25.

26.
27.
28.

29.

30.
31.

Carosio, F.; Colonna, S.; Fina, A.; Rydzek, G.; Hemmerle, J.; Jierry, L.; Schaaf, P.; Boulmedais, F. Efficient gas
and water vapor barrier properties of thin poly(lactic acid) packaging films: Functionalization with moisture
resistant nafion and clay multilayers. Chem. Mater. 2014, 26, 5459–5466. [CrossRef]
Alongi, J.; Tata, J.; Carosio, F.; Rosace, G.; Frache, A.; Camino, G. A comparative analysis of nanoparticle
adsorption as fire-protection approach for fabrics. Polymers 2015, 7, 47–68. [CrossRef]
Alongi, J.; Carosio, F.; Kiekens, P. Recent advances in the design of water based-flame retardant coatings for
polyester and polyester-cotton blends. Polymers 2016, 8, 357–380. [CrossRef]
Decher, G.; Hong, J.D.; Schmitt, J. Buildup of ultrathin multilayer films by a self-assembly process:
III. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on charged surfaces.
Thin Solid Films 1992, 210–211, 831–835. [CrossRef]
Decher, G. Fuzzy nanoassemblies: Toward layered polymeric multicomposites. Science 1997, 277, 1232–1237.
[CrossRef]
Laufer, G.; Carosio, F.; Martinez, R.; Camino, G.; Grunlan, J.C. Growth and fire resistance of colloidal
silica-polyelectrolyte thin film assemblies. J. Colloid. Interface Sci. 2011, 356, 69–77. [CrossRef] [PubMed]
Carosio, F.; Cuttica, F.; Di Blasio, A.; Alongi, J.; Malucelli, G. Layer by layer assembly of flame retardant thin
films on closed cell pet foams: Efficiency of ammonium polyphosphate versus DNA. Polym. Degrad. Stab.
2015, 113, 189–196. [CrossRef]
Carosio, F.; Di Blasio, A.; Alongi, J.; Malucelli, G. Layer by layer nanoarchitectures for the surface protection
of polycarbonate. Eur. Polym. J. 2013, 49, 397–404. [CrossRef]
Apaydin, K.; Laachachi, A.; Ball, V.; Jimenez, M.; Bourbigot, S.; Toniazzo, V.; Ruch, D. Polyallylaminemontmorillonite as super flame retardant coating assemblies by layer-by layer deposition on polyamide.
Polym. Degrad. Stab. 2013, 98, 627–634. [CrossRef]
Fina, A.; Cuttica, F.; Camino, G. Ignition of polypropylene/montmorillonite nanocomposites. Polym. Degrad. Stab.
2012, 97, 2619–2626. [CrossRef]
Cain, A.A.; Nolen, C.R.; Li, Y.C.; Davis, R.; Grunlan, J.C. Phosphorous-filled nanobrick wall multilayer thin
film eliminates polyurethane melt dripping and reduces heat release associated with fire. Polym. Degrad. Stab.
2013, 98, 2645–2652. [CrossRef]
Alongi, J.; Carosio, F.; Malucelli, G. Layer by layer complex architectures based on ammonium polyphosphate,
chitosan and silica on polyester-cotton blends: Flammability and combustion behaviour. Cellulose 2012, 19,
1041–1050. [CrossRef]
Holder, K.M.; Huff, M.E.; Cosio, M.N.; Grunlan, J.C. Intumescing multilayer thin film deposited on clay-based
nanobrick wall to produce self-extinguishing flame retardant polyurethane. J. Mater. Sci. 2015, 50, 2451–2458.
[CrossRef]
Carosio, F.; Alongi, J. Ultra-fast layer-by-layer approach for depositing flame retardant coatings on flexible
pu foams within seconds. ACS Appl. Mater. Interfaces 2016, 8, 6315–6319. [CrossRef] [PubMed]
Carosio, F.; Alongi, J.; Malucelli, G. Flammability and combustion properties of ammonium polyphosphate-/
poly(acrylic acid)-based layer by layer architectures deposited on cotton, polyester and their blends. Polym.
Degrad. Stab. 2013, 98, 1626–1637. [CrossRef]
Alongi, J.; Han, Z.; Bourbigot, S. Intumescence: Tradition versus novelty. A comprehensive review.
Prog. Polym. Sci. 2015, 51, 28–73. [CrossRef]
Carosio, F.; Alongi, J.; Malucelli, G. Layer by layer ammonium polyphosphate-based coatings for flame
retardancy of polyester-cotton blends. Carbohydr. Polym. 2012, 88, 1460–1469. [CrossRef]
Li, Y.C.; Mannen, S.; Morgan, A.B.; Chang, S.C.; Yang, Y.H.; Condon, B.; Grunlan, J.C. Intumescent all-polymer
multilayer nanocoating capable of extinguishing flame on fabric. Adv. Mater. 2011, 23, 3926–3931. [CrossRef]
[PubMed]
Carosio, F.; Fontaine, G.; Alongi, J.; Bourbigot, S. Starch-based layer by layer assembly: Efficient and
sustainable approach to cotton fire protection. ACS Appl. Mater. Interfaces 2015, 7, 12158–12167. [CrossRef]
[PubMed]
Yi, D.; Yang, R. Ammonium polyphosphate/montmorillonite nanocompounds in polypropylene. J. Appl.
Polym. Sci. 2010, 118, 834–840. [CrossRef]
Ma, H.; Tong, L.; Xu, Z.; Fang, Z. Intumescent flame retardant-montmorillonite synergism in ABS
nanocomposites. Appl. Clay Sci. 2008, 42, 238–245. [CrossRef]

119

Polymers 2016, 8, 430

32.
33.
34.
35.

36.
37.

38.

39.
40.
41.
42.
43.
44.

45.

Tata, J.; Alongi, J.; Carosio, F.; Frache, A. Optimization of the procedure to burn textile fabrics by cone
calorimeter: Part I. Combustion behavior of polyester. Fire Mater. 2011, 35, 397–409. [CrossRef]
Norrish, K. The swelling of montmorillonite. Discuss. Faraday Soc. 1954, 18, 120–134. [CrossRef]
Zheng, Y.; Zaoui, A. How water and counterions diffuse into the hydrated montmorillonite. Solid State Ion.
2011, 203, 80–85. [CrossRef]
Carosio, F.; Kochumalayil, J.; Cuttica, F.; Camino, G.; Berglund, L. Oriented clay nanopaper from biobased
components–mechanisms for superior fire protection properties. ACS Appl. Mater. Interfaces 2015, 7,
5847–5856. [CrossRef] [PubMed]
Alongi, J.; Camino, G.; Malucelli, G. Heating rate effect on char yield from cotton, poly(ethylene terephthalate)
and blend fabrics. Carbohydr. Polym. 2013, 92, 1327–1334. [CrossRef] [PubMed]
Davies, P.J.; Horrocks, A.R.; Alderson, A. The sensitisation of thermal decomposition of ammonium
polyphosphate by selected metal ions and their potential for improved cotton fabric flame retardancy.
Polym. Degrad. Stab. 2005, 88, 114–122. [CrossRef]
Camino, G.; Costa, L.; Trossarelli, L. Study of the mechanism of intumescence in fire retardant polymers.3.
Effect of urea on the ammonium polyphosphate pentaerythritol system. Polym. Degrad. Stab. 1984, 7, 221–229.
[CrossRef]
Price, D.; Horrocks, A.R.; Akalin, M.; Faroq, A.A. Influence of flame retardants on the mechanism of pyrolysis
of cotton (cellulose) fabrics in air. J. Anal. Appl. Pyrolysis 1997, 40–41, 511–524. [CrossRef]
Mamleev, V.; Bourbigot, S.; Le Bras, M.; Yvon, J. The facts and hypotheses relating to the phenomenological
model of cellulose pyrolysis interdependence of the steps. J. Anal. Appl. Pyrolysis 2009, 84, 1–17. [CrossRef]
Kandola, B.K.; Horrocks, A.R.; Price, D.; Coleman, G.V. Flame-retardant treatments of cellulose and their
influence on the mechanism of cellulose pyrolysis. J. Macromol. Sci. RMC 1996, C36, 721–794. [CrossRef]
Schartel, B.; Hull, T.R. Development of fire-retarded materials—Interpretation of cone calorimeter data.
Fire Mater. 2007, 31, 327–354. [CrossRef]
Soares, S.; Camino, G.; Levchik, S. Comparative-study of the thermal-decomposition of pure cellulose and
pulp paper. Polym. Degrad. Stab. 1995, 49, 275–283. [CrossRef]
Carosio, F.; Alongi, J. Influence of layer by layer coatings containing octapropylammonium polyhedral
oligomeric silsesquioxane and ammonium polyphosphate on the thermal stability and flammability of
acrylic fabrics. J. Anal. Appl. Pyrolysis 2016, 119, 114–123. [CrossRef]
Ferrari, A.C.; Basko, D.M. Raman spectroscopy as a versatile tool for studying the properties of graphene.
Nat. Nanotechnol. 2013, 8, 235–246. [CrossRef] [PubMed]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

120

polymers
Article

Polysarcosine-Based Lipids: From Lipopolypeptoid
Micelles to Stealth-Like Lipids in Langmuir
Blodgett Monolayers
Benjamin Weber 1 , Christine Seidl 1 , David Schwiertz 1 , Martin Scherer 1 , Stefan Bleher 2 ,
Regine Süss 2 and Matthias Barz 1, *
1

2

*

Institute of Organic Chemistry, Johannes Gutenberg University of Mainz, Duesbergweg 10-14, 55128 Mainz,
Germany; b.weber@uni-mainz.de (B.W.); cseidl@students.uni-mainz.de (C.S.);
daschwie@students.uni-mainz.de (D.S.); m.scherer@uni-mainz.de (M.S.)
Department of Pharmaceutical Technology and Biopharmacy, Institute of Pharmaceutical Sciences,
Albert Ludwigs University of Freiburg, Sonnenstraße 5, 79104 Freiburg im Breisgau, Germany;
stefan.bleher@pharmazie.uni-freiburg.de (S.B.); regine.suess@pharmazie.uni-freiburg.de (R.S.)
Correspondence: barz@uni-mainz.de; Tel.: +49-6131-392-5468

Academic Editor: Alexander Böker
Received: 17 October 2016; Accepted: 2 December 2016; Published: 9 December 2016

Abstract: Amphiphiles and, in particular, PEGylated lipids or alkyl ethers represent an important
class of non-ionic surfactants and have become key ingredients for long-circulating (“stealth”)
liposomes. While poly-(ethylene glycol) (PEG) can be considered the gold standard for stealth-like
materials, it is known to be neither a bio-based nor biodegradable material. In contrast to PEG,
polysarcosine (PSar) is based on the endogenous amino acid sarcosine (N-methylated glycine), but
has also demonstrated stealth-like properties in vitro, as well as in vivo. In this respect, we report on
the synthesis and characterization of polysarcosine based lipids with C14 and C18 hydrocarbon chains
and their end group functionalization. Size exclusion chromatography (SEC) and matrix-assisted
laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis reveals
that lipopeptoids with a degree of polymerization between 10 and 100, dispersity indices around
1.1, and the absence of detectable side products are directly accessible by nucleophilic ring opening
polymerization (ROP). The values for the critical micelle concentration for these lipopolymers are
between 27 and 1181 mg/L for the ones with C18 hydrocarbon chain or even higher for the C14
counterparts. The lipopolypeptoid based micelles have hydrodynamic diameters between 10 and
25 nm, in which the size scales with the length of the PSar block. In addition, C18 PSar50 can be
incorporated in 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) monolayers up to a polymer
content of 3%. Cyclic compression and expansion of the monolayer showed no significant loss of
polymer, indicating a stable monolayer. Therefore, lipopolypeptoids can not only be synthesized
under living conditions, but my also provide a platform to substitute PEG-based lipopolymers as
excipients and/or in lipid formulations.
Keywords: polysarcosine; polypeptoids; surfactants; lipids; NCA polymerization; PSarcosinylated lipids

1. Introduction
Amphiphilic molecules and polymers are commonly applied to lower the surface tension
(or interfacial tension) between two liquids or between a liquid and a solid, which enables their
use as detergents, wetting agents, emulsifiers, foaming agents, and dispersants [1,2]. From a structural
point of view, these polymers can be divided into two classes, which are characterized by the relative
distribution of hydrophilic and lipophilic units. Macromolecules based on intrinsically amphiphilic

Polymers 2016, 8, 427

121

www.mdpi.com/journal/polymers

Polymers 2016, 8, 427

repeating units are summarized as “polysoaps”, whereas polymers with strictly separated parts are
called “macrosurfactants” [3–6]. These macrosurfactants are commonly amphiphilic block copolymers
or lipopolymers, in which a hydrocarbon chain of 12–18 units is attached to a hydrophilic polymer.
With respect to sustainability, bio-based amphiphiles, such as lipopeptides, have gained pronounced
attention as they are based on renewable raw materials [7]. This enormous potential was already
recognized half a century ago [8–10] and, thus, several amino acid or peptide based amphiphiles
have been investigated, in which fatty acid chains, as well as amino acids or peptides, can vary in
composition and length [11].
In addition, PEGylated lipids or alkyl ethers represent an important class of non-ionic
lipopolymers and are key ingredients for the preparation of long-circulating liposomes, since a
significant step in the development of long-circulating liposomes came with the incorporation of
the synthetic polymer poly-(ethylene glycol) (PEG) in liposome compositions. The presence of
PEG on the surface of the liposomal carriers has been shown to extend blood-circulation time
while reducing mononuclear phagocyte system uptake (stealth liposomes). Despite the enormous
achievements of PEGylated lipids, several groups have reported immune responses towards PEG and
PEGylated lipids, leading to the accelerated blood clearance (ABC) phenomenon [12–14]. Moreover,
PEG is not degradable in vivo and relies on complete excretion to avoid storage diseases [15].
Consequently, the finding of alternatives to PEG is a growing field of research [16]. Among various
PEG surrogates, polypeptides and polypeptoids are attracting more and more attention [16–18].
The polypeptoid polysarcosine (PSar) seems to be particularly interesting because it is, on one hand,
based on the endogenous N-substituted amino acid, sarcosine (N-methylated glycine), and on the
other hand, sarcosine can be easily synthesized by a simple nucleophilic substitution reaction of
bromo- or chloroacetic acid and methylamine [19]. Furthermore, PSar can be synthesized under
living conditions from the corresponding α-amino acid N-carboxy anhydride (NCA) [19–21] and has
already demonstrated possessing stealth-like properties comparable to PEG [18,22–26]. Interestingly,
lipopeptides, as non-ionic and bio-based systems, have been practically overlooked. So far only
Gallot and coworkers have reported on the synthesis of PSar-based lipopolymers (lipopeptoids).
In 1986 they reported the synthesis of lipopeptoids using aliphatic amines to initiate the ring opening
polymerization (ROP) of the Sar NCA in chloroform [27,28]. Surprisingly, they had to fractionate the
final lipopolypeptoid yielding different fractions with degrees of polymerization from 10 to 60. Due
to the living nature of Sar NCA ROP, one would expect that such degrees of polymerization can be
directly obtained by adjusting the monomer to initiator ratio. To validate our expectation, we carried
out the synthesis of lipopolypeptoids based on either tetradecylamine (C14 ) or stearylamine (C18 ).
In this work, we report the synthesis and end group functionalization of PSar based
lipopolypeptoids with C14 and C18 hydrocarbon tails. The lipopolypeptoids have a PSar block with
chain lengths (Xn ) from 10–100. Furthermore, we introduce a synthetic pathway, which allows
polymerization of such systems on 50–100 g scale. The final lipopolymers are characterized by 1 H NMR,
1 H-DOSY NMR, SEC and MALDI-TOF mass spectrometry to ensure the living nature of the ring
opening polymerization. We also report on the critical micelle concentration (CMC) of the synthesized
systems, characterize the aggregates by dynamic light scattering, investigate cellular toxicities, and
report on the incorporation of 1, 2, and 3 mol % of PSar45-stearylamine into Langmuir-Blodgett
monolayers of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC).
2. Materials and Methods
n-Hexane was distilled from Na/K and ethyl acetate from CaH2 . Dimethylformamide (DMF)
was purchased from Acros Organics (Geel, Belgium) and dried over BaO and molecular sieves (3 Å),
fractionally distilled under vacuum at 40 ◦ C and stored at −80 ◦ C under the exclusion of light. Prior to
use, DMF was degassed in vacuum to remove traces of dimethylamine. Hexafluoroisopropanol (HFIP)
was purchased from Fluorochem (Hadfield Derbyshire, UK). Millipore water was prepared by a MILLI-Q®
Reference A+ System (Darmstadt, Germany). Octadecylamine and was purchased from Fluka

122

Polymers 2016, 8, 427

(St. Gallen, Switzerland) and was dried at 40 ◦ C under vacuum (1 × 10−3 mbar) for 24 h. Diphosgene
was purchased from Alfa Aesar (Ward Hill, MA, USA) and deuterated solvents from Deutero GmbH
(Kastellaun, Germany). Other chemicals were purchased from Sigma-Aldrich (Taufkirchen, Germany)
and used as received unless otherwise stated. 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) was
purchased from Avanti Polar Lipids (Alabaster, Al, USA) and used without purification. Roswell Park
Memorial Institute (RPMI) cell medium and FCS was purchased from Merck Millipore (Darmstadt,
Germany). HeLa cells were obtained from DSMZ (German Collection of Microorganisms and Cell
Cultures, Braunschweig, Germany).
1 H NMR spectra were recorded on a Bruker (Billerica, MA, USA) AC 400 at a frequency of
400 MHz respectively. Two-dimensional NMR spectra as 1 H DOSY were recorded on a Bruker Avance
III HD 400 at 400 MHz. All spectra were recorded at room temperature (25 ◦ C) and calibrated using
the solvent signals. Melting points were measured using a Mettler FP62 melting point apparatus
at a heating rate of 2.5 ◦ C·min−1 . Gel permeation chromatography (GPC) was performed with
hexafluoroisopropanol (HFIP) containing 3 g·L−1 potassium trifluoroacetate (KTFA) as the eluent
at 40 ◦ C and a flow rate of 0.8 mL·min−1 . The columns were packed with modified silica (PFG
column particle size: 7 μm, porosity: 100 and 1000 Å). Polymethylmethacrylate (PMMA) standards
(Polymer Standards Services GmbH (Mainz, Germany)) were used for calibration and toluene was
used as the internal standard. A refractive index detector (G1362A RID) and an UV-VIS detector
(at 230 nm unless otherwise stated; Jasco (Gross-Umstadt, Germany) UV-2075 Plus) were used for
polymer detection. MALDI-TOF mass spectra [29] were recorded using a Bruker Reflex II MALDI-TOF
mass spectrometer equipped with a 337 nm N2 laser. Acceleration of the ions was performed with
pulsed ion extraction (PIE, Bruker) at a voltage of 20 kV. The analyzer was operated in reflection
mode and the ions were detected using a microchannel plate detector. Mass spectra were processed
by the X-TOF 5.1.0 software (Bruker (Billerica, MA, USA)). A solvent-free sample preparation was
performed using trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB) as
the matrix and sodium trifluoroacetate as the cationizing salt. Calibration was carried out using a
C60 /C70 fullerene mixture. Infrared (IR) spectroscopy was performed on a Jasco FT/IR-4100 with an
ATR sampling accessory (MIRacle, Pike Technologies, Madison, WI, USA) and Spectra Manager 2.0
(Jasco, Gross-Umstadt, Germany) was used for integration.
Surface pressure-area (π vs. A) isotherms were obtained using a Nima Langmuir-Blodgett
trough (KSV Nima, (Espoo, Finland), Coventry, type 611) secured inside an acrylic glass box (Bayer,
Leverkusen, Germany) as a dust shield. The total trough surface area was 200 mm × 100 mm, and
the total trough volume was approximately 150 mL. The effective trough area was controlled by
two hydrophobic barriers that compressed the spread film symmetrically and bilaterally at a rate of
5 cm2 /min. Millipore water was used as subphase in all trials. For all experiments, the subphase
temperature was 25 ± 0.1 ◦ C (15-min delay after the water was filled in and the lipid solution was
spread). Prior to each trial, the water surface was cleaned by aspirating off any residue, such that the
measured surface pressure remained <0.1 mN/m over a full compression. The Langmuir-Blodgett
(LB) components were cleaned with absolute ethanol and chloroform before each experiment, and
the deionized water subphase was replaced after each measurement. Surface pressure measurements
were taken from a Wilhelmy plate (perimeter of 20 mm × 10 mm) made out of chromatography paper,
which was washed several times with absolute chloroform prior to each trial to ensure cleanliness.
Dynamic light scattering measurements were performed at 25 ◦ C using a Malvern (Malvern, UK)
Zetasizer NanoZS with a He/Ne laser (633 nm) at a fixed angle of 173◦ .
Synthesis of sarcosine N-carboxyanhydride. The synthesis of sarcosine NCA was adapted
from literature and modified. A total of 14.92 g (167.4 mmol) sarcosine, dried under vacuum for
1 h, was weighed into a pre-dried, three-neck, round-bottom flask. A total of 300 mL of absolute
tetrahydrofurane (THF) was added under a steady flow of nitrogen, 16.2 mL (134 mmol) of diphosgene
was added slowly via syringe, and the nitrogen stream was reduced. The colorless suspension was
mildly refluxed for 3 h, yielding a clear solution. Afterward, a steady flow of dry nitrogen was led

123

Polymers 2016, 8, 427

through the solution for another 3 h while the outlet was connected to two gas washing bottles filled
with aqueous NaOH solution to neutralize phosgene. The solvent was evaporated under reduced
pressure, yielding a brownish oil as a crude reaction product. The oil was dried under reduced pressure
(1 × 10−3 mbar for 2 h) to obtain an amorphous solid, free of phosgene and HCl, confirmed by testing
against a silver nitrate solution. The crude product was redissolved in 40 mL of THF and precipitated
with 300 mL of dry n-hexane. The solution was cooled to −18 ◦ C and stored for 18 h to complete
precipitation. The solid was filtered under dry nitrogen atmosphere and dried in a stream of dry
nitrogen for 60–90 min and afterwards under high vacuum for 2 h in the sublimation apparatus. The
crude product was sublimated at 85 ◦ C and 1 × 10−3 mbar. The product was collected from the
sublimation apparatus in a glovebox on the same day. The purified product (110 mmol, 65% yield,
colorless crystallites; melting point: 102–104 ◦ C (lit: 102–105 ◦ C)) was stored in a Schlenk tube at
−80 ◦ C and only handled in a glovebox.
1 H NMR (300 MHz, CDCl ): δ/ppm = 4.22 (2H, s, –CH –CO–), 2.86 (3H, s, –CH ).
3
2
3
Synthesis of polysarcosine. Under nitrogen counter flow, Sar-NCA was transferred into a pre-dried
Schlenk tube equipped with a stir bar and again dried under high vacuum for 1 h. Then, the NCA
was dissolved in dry DMF to yield a solution of 100 mg/mL with respect to the NCA. 1/n equivalent
of either tetradecyl amine or stearyl amine was dissolved in pre-dried THF and added to the NCA
solution. The solution was stirred at room temperature and kept at a constant pressure of 1.25 bar
of dry nitrogen via the Schlenk line to prevent impurities from entering the reaction vessel while
allowing CO2 to escape. Completion of the reaction was confirmed by Fourier transform infrared
(FTIR) spectroscopy (disappearance of the NCA peaks (1853 and 1786 cm−1 )). After completion of the
reaction, the polymer was precipitated with cold ether and centrifuged (4500 rpm at 4 ◦ C for 15 min).
After discarding the liquid fraction, new ether was added and the polymer was resuspended in a sonic
bath. The suspension was centrifuged again and the procedure was repeated. After complete DMF
removal by the resuspension steps, the polymer was dissolved in water and lyophilized, obtaining a
colorless, stiff and porous solid.
1 H NMR (400 MHz; DMSO-d ): δ/ppm: 4.43–3.83 (14H; br; (2n)–CO–CH –NH–); 3.14–2.65 (23H;
6
2
br; (3n)–N–CH3 –); 1.53–1.12 (32H; br; –CH2 –(CH2 )16 –CH3 ); 0.86 (3H; t; –CH2 –CH3 ).
Synthesis of carboxy functionalized polymers. The polymer was dissolved in dry DMF with
10 eq. (with respect to the polymer end group) of diisopropylethylamine (DIPEA) and stirred for
30 min. To this solution the 5 eq. succinic acid anhydride was added and stirred overnight at room
temperature. The excess of DIPEA and succinic anhydride were removed by dialysis and the product
was lyophilized. Complete removal was verified by DOSY 1 H NMR.
1 H NMR (400 MHz; DMSO-d ): δ/ppm: 4.69–3.72 (97H; br; (2n)–CO–CH –NH–); 3.10–2.66 (152H;
6
2
br; (3n)–N–CH3 –); 2.42–2.25 (4H; br; –CO–(CH2 )2 –COOH); 1.47–1.17 (32H; br; –CH2 –(CH2 )16 –CH3 );
0.86 (3H; t; –CH2 –CH3 ).
Synthesis of acetylated polymers. The polymer was dissolved in dry DMF with 10 eq. of DIPEA
and stirred for 30 min. To this solution the 5 eq. acetic anhydride or the FITC was added and stirred
overnight at room temperature. Excess DIPEA and acetic acid anhydride were removed by dialysis
and the product was lyophilized. Complete removal was verified by 1 H-DOSY NMR.
1 H NMR (400 MHz; DMSO-d ): δ/ppm: 4.55–3.77 (99H; br; (2n)–CO–CH –NH–); 3.22–2.63 (154H;
6
2
br; (3n)–N–CH3 –); 2.06–1.90 (3H; br; –NCH3 –CO–CH3 ); 1.49–1.14 (32H; br; –CH2 –(CH2 )16 –CH3 ); 0.86
(3H; t; –CH2 –CH3 ).
Synthesis of FITC labeled polymers. The polymer was dissolved in dry DMF with 10 eq. of
DIPEA and stirred for 30 min. To this solution 2 eq. FITC was added and stirred overnight at room
temperature. Excess DIPEA and FITC were removed by dialysis and the product was lyophilized.
Complete removal was verified by 1 H-DOSY NMR.
1 H NMR (400 MHz; DMSO-d ): δ/ppm: 10.32–9.87 (1H; br; FITC–COOH); 8.37–7.37 (3H; br;
6
aromatic –CH–C–COOH–; aromatic CH–CH–COH; aromatic CH–CH–CNHR); 6.74–6.24 (6H; br;

124

Polymers 2016, 8, 427

–CH–CO–CH–CH–; –COH–CH–COR–; CH–CH–CNHR); 4.81–3.71 (93H; br; (2n)–CO–CH2 –NH–);
3.14–2.65 (142H; br; (3n)–N–CH3 –); 1.54–1.10 (32H; br; –CH2 –(CH2 )16 –CH3 ); 0.86 (3H; t; –CH2 –CH3 ).
CMC measurements. CMCs have been determined with a Dataphysics (Filderstadt, Germany)
ring tensiometer (DCATIIEC) at 25 ◦ C. It was calibrated against deionized water purified with a
Milli-Q system (Merck Millipore, Darmstadt, Germany) to 18.2-MΩ cm resistivity and TOC <5 ppb.
All samples were aged for 30 min prior to use.
Dynamic light scattering measurements: Lipopeptoids were dissolved in phosphate buffered
saline (PBS) to yield 0.1 mg/mL. Prior to measurement samples were aged for 30 min.
Cellular Toxicity: Toxicity studies were carried out using the CellTiter-Glo® Luminescent Cell
Viability Assay by Promega. The assay was carried out following the manufacturers’ protocol. HeLa
cells were cultured in RPMI medium with 10% heat-inactivated fetal bovine serum (FCS). Cells were
harvested at 60%–70% confluence, incubated at 37 ◦ C, 95% relative humidity (rh) and 5% CO2 for
24 h in a 24-well plate with lipopeptoids. Lipopeptoids were dissolved in PBS. Medium was replaced
1 h prior to experiments. The experiments were performed in triplicate. Data was normalized to the
untreated control.
Langmuir Blodgett layer formation: Solutions of the polymer-lipid mixtures in chloroform were
spread on the subphase by using a microsyringe (Kloehn, Las Vegas, NV, USA). In a typical experiment,
20–30 μL of the solution was spread dropwise onto the water surface so that a constant mass of lipid
was deposited for each trial. The spreading solution was deposited at regularly spaced locations on
the trough. In all trials, a 15-min evaporation period between the last deposited drop of solution and
the beginning of compression was employed to ensure complete solvent evaporation.
3. Results
3.1. Synthesis of Lipopolypeptoids
The synthesis of 100 mg to 1 g of lipopolymer was conducted using freshly sublimated Sar-NCA,
purified solvent (DMF) and initiator (tetradecyl or stearyl amine). The polymerizations were carried out
at room temperature with a monomer concentration of 0.1 g/mL. After complete monomer conversion
was ensured by FTIR-measurements (disappearance of NCA attributed carbonyl vibration band at 1786
and 1850 cm−1 ) lipopeptoids were precipitated in cold diethylether. Afterwards, lipopolymers were
dried by lyophilization from water and analyzed by 1 H NMR, HFIP SEC, MALDI-TOF, and 1 H-DOSY
NMR. 1 H NMR experiments displayed that the deviation of the obtained degrees to those calculated
are below 10% (Table 1). In hexafluoroisopropanole (HFIP) SEC the synthesized lipopolymers indicate
a symmetric narrowly-distributed molecular weight distribution. The PMMA equivalent molecular
weights are in the range of 5 to 25 kg/mol, while dispersities are between 1.05 and 1.13 (see Table 1).
Furthermore, SEC clearly demonstrates that the hydrodynamic volume scales with the degree of
polymerization (Figure 1a). An influence of the initiator on the control over polymerization was not
detectable, as both aliphatic amines lead to a well-controlled polymerization. To ensure the formation
of lipopolymers and the absence of PSar homopolymers 1 H-DOSY NMR experiments have been carried
out. The diffusion ordered NMRs confirm the absence of low molecular weight or high molecular
weight side-products, since only a single diffusing polymer species is detected, which contains all PSar,
as well as lipid attributed proton signals.
In the next step, the methylamine end groups were quenched with fluoresceine isothiocyanate
(FITC), succinic acid anhydride (COOH functionality), or acetic acid anhydride (neutral end group)
in the presence of diisopropylethylamine (DIPEA) (Scheme 1). While acetylated PSar could be used
as a stealth-only material, carboxylic acid-functionalized lipopolypeptoids are accessible to further
modifications since targeting moieties e.g., antibodies or sugars can be attached. Labeling with a dye,
in this case FITC, will allow analysis of these formulations with fluorescent techniques, e.g., fluorescent
correlation spectroscopy, confocal microscopy, and fluorescence activated cell sorting. To monitor the
end group modification efficiency further 1 H-DOSY NMR experiments have been conducted. This

125

Polymers 2016, 8, 427

method cannot only help to ensure that end group modification is complete, but also confirm the
successful removal of the small molecules used for the polymer modification during workup (dialysis).
(Appendix A Figure A1) With respect to the limits of 1 H NMR spectroscopy, these experiments reveal
the absence of impurities as well as the quantitative conversion of PSar end groups.
Table 1. Polymer analysis of amphiphilic PSars.
Polymer

X n (Calculated)

X n (NMR)

C18 PSar12
C18 PSar30
C18 PSar45
C18 PSar64
C18 PSar117
C14 PSar11
C14 PSar34
C14 PSar53
C14 PSar75
C14 PSar103

10
30
50
70
100
10
30
50
70
100

12
30
45
64
117
11
34
53
75
103

X n (MALDI)

47

49

M n (GPC)

Ð

5232
7619
13,933
15,090
24,890
3960
9035
14,718
18,220
23,210

1.06
1.06
1.07
1.08
1.13
1.05
1.07
1.08
1.12
1.11

Figure 1. Exemplary polymer analysis of lipopolypeptoids (a) HFIP SEC of C18 PSar34 , C18 PSar45 ,
and C18 PSar117 ; (b) MALDI TOF MS of C18 PSar47 ; (c) 1 H NMR of C18 PSar47 ; and (d) 1 H DOSY NMR
of C18 PSar47 .

After the synthesis, solution properties of lipopolypeptoids are investigated, which are the
critical micelle concentration (CMC) and the hydrodynamic diameter (Dh ) in aqueous solution by
dynamic light scattering (DLS). First, we calculated the hydrophilic-lipophilic balance (HLB) for the
synthesized lipopolymers.
126

Polymers 2016, 8, 427

Scheme 1. A synthetic pathway to heterotelechelic lipopolypeptoids.

3.2. Characterization, Solution Properties, and Cellular Toxicity of Lipopolypeptoids
According to W.C. Griffin the hydrophilic-lipophilic balance (HLB) is defined as [30]:
HLB = 20 × (1 −

Mlipohilic
)
Mlipopeptoid

(1)

Therefore, the HLB values are between 15 and 20 for the synthesized lipopolymers, which is a HLB
range for solubilizing agents in general (see Table 2). Solubilizing agents are a class of amphiphiles,
which are hardly able to incorporate water soluble substances into micellar structures. This process
is called micellar solubilization, or, briefly, solubilization [31]. In comparison to other amphiphiles,
solubilizing agents have higher HLB values and biocompatible ones are often used to solubilize
hydrophobic drugs [32].
Table 2. Calculated HLB values and measured CMC.
Polymer

HLB value 1

CMC2 /mg·L−1

Diameter of main peak (DLS/nm)

Distribution

C18 PSar12
C18 PSar30
C18 PSar45
C18 PSar65
C18 PSar117
C14 PSar11
C14 PSar34
C14 PSar53
C14 PSar75
C14 PSar103

16.0
17.8
18.4
19.0
19.5
15.7
18.4
18.9
19.1
19.4

27
62
94
181
1181
213
-

10.1
13.7
19.1
25.2
24.9
9.1
12.7
16.4
19.2
24.4

bimodal
bimodal
monomodal
monomodal
monomodal
monomodal
bimodal
monomodal
bimodal
trimodal

1

calculated according to W.C. Griffin: HLB = 20 × (1 −

(DCATIIEC) at 25

◦ C.

Mlipohilic
2
Mlipopeptoid );

measured using ring tensiometry

With respect to their HLB values, all synthesized lipopolymers can be considered as solubilizing
agents, which raises the question of PSar chain length dependency of CMCs and hydrodynamic
diameter of micelles. Therefore, ring tensiometry was used to determine CMC values of the
synthesized lipopolypeptoids. In the applied tensiometer setup the surface tension at about 70 different
127

Polymers 2016, 8, 427

concentrations was measured. With increasing amounts of surfactant the surface tension of the solution
decreases. Once the CMC is reached the surface tension remains constant and micelles start to form.
All additional surfactants added to the system increase the micellar fraction (Figure 2). Within the
series of C18 lipopolypeptoids, the CMC increases with the degree of polymerization of the PSar block
and, thus, with HLB values from 0.027 (C18 PSar12 ) to 1.181 g/L (C18 PSar117 ) (See Table 1). As expected,
lipopolypeptoids with C14 alkyl chain have significantly higher CMCs than those with C18 alkyl chain.
The aggregation concentration is so high that only for C14 PSar11 a CMC could be determined in the
applied concentration range. Thus, CMCs of the other C14 lipopolymers are above 1.500 g/L.
After determining CMC values, the synthesized lipopolypeptoids micelles have been further
investigated. Therefore, a concentration of 1 g/L was chosen, which is well above the CMC of the
corresponding C18 lipopolymers. Only the lipopolymers with a C14 alkyl chain and the C18 PSar117
have been characterized at a concentration of 10 g/L. The series of C18 lipopolypeptoids formed
aggregates, which increase in size with the degree of polymerization. The aggregates have diameters
between 10.1 nm (C18 PSar12 ) and 24.9 nm (C18 PSar117 ) (see Table 2). For the C14 lipopolypeptoids
the trend is also confirmed, since those micelles have diameters between 9.6 nm (C14 PSar11 ) and
28.1 nm (C14 PSar103 ). C14 PSar103 also shows a unimer fraction next to the micellar fractions (Figure 3).
Micelles formed by C18 and C14 lipopolypeptoids have a comparable diameter at comparable degrees
of polymerization. For the PSars with Xn < 35 a fraction of several hundred nm is observed.
The number-weighted distribution does not display these fractions (Figure A3), which is due to
the overestimation of large fractions in intensity weighted plotting. Since these polymers have only a
single alkyl chain, hydrophobic stabilization of the micelles is very low. This leads to relatively high
CMCs and, consequently, to a dynamic system with high exchange rates, which seems to be the reason
for the formation of more complex aggregates.
In the next step and with a view using lipopolypeptoids as excipients in drug formulations,
the cellular toxicity of lipopolypeptoids was investigated in HeLa cells using the CellTiter-Glo® assay.
This assay is a method of determining the number of viable cells based on quantitation of the ATP
values present, an indicator of metabolically active cells. The quantification relies on a proprietary
thermostable luciferase, which generates a stable “glow-type” luminescent signal depending on ATP
levels. In relation to untreated cells as the positive control, relative cell viability can be determined.
In this case lipopolypeptoids showed no toxicity up to a concentration of 50 μM. For a concentration of
500 μM (1.2 to 4.2 mg) the lipopolymers with low degrees of polymerization showed toxicity, which
was not observed for lipopolymers with high PSar content (Figure 4).

Figure 2. Exemplary CMC measurement shown for C18 PSar15 . CMC was determined by ring
tensiometry. Intersection of the two lines represents the CMC value.

128

Polymers 2016, 8, 427

Figure 3. Intensity-weighted diameter distribution of lipopolypeptoids in aqueous solution determined
by dynamic light scattering (DLS). (a) Series of tetradecyl amine (C14 )-based lipopolypeptoids; and
(b) series of stearyl amine (C18 )-based lipopolypeptoids.

Figure 4. CellTiter-Glo® (CTG) assay toxicity studies of unformulated amphiphilic PSars in HeLa
cells in the concentration range from 0.5 up to 500 μM. Data was normalized to the untreated control.
For better comparability molar concentrations were chosen. 1.2 mg/mL (C18 PSar30 ) and 4.2 mg/mL
(C14 PSar103 ) correspond to 500 μmol.

3.3. Formation and Characterization of Lipopolypeptoid Containing Langmuir-Blodgett Monolayers
In the last part of this study, and with regard to the incorporation of lipopolypeptoids in
liposomes to provide stealth-like properties, it was investigated to which extent lipopolypeptoids
can be incorporated into lipid membranes of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC). On
a Langmuir-Blodgett trough solutions of the lipopolymer-lipid mixture in chloroform were spread
on the subphase. After evaporation of the chloroform, the area was compressed with a speed of
5 cm2 /min. Plotting the surface pressure over the area, a similar behavior is observed for a polymer
content up to 3 mol %. When the compression starts pure DSPC lipids are in the gas analogous phase.
At a pressure from 0 to 3.8 mN/m lipids are in the liquid analogous phase having an average area
of 70 Å2 /mol. At 55 Å2 /mol the slope of the curve rises and the solid analogous phase is reached.
For the mixture bearing 1% lipopolypeptoids the phase transitions remain at the same area/molecule,
but the transition from liquid to solid analogous phase takes place at 6.7 mN/m. Having a mixture
with 2% lipopolypeptoids the onset point of the liquid analogous phase starts at 92 Å2 /mol. The
surface pressure rises slowly to 8.15 mN/m and a small plateau is reached at 59 Å2 /mol (coexistence
129

Polymers 2016, 8, 427

of liquid and solid analogous phase). Transition to solid analogous phase remains at 55 Å2 /mol at a
surface pressure of 10.7 mN/m. The phase transition to liquid analogous phase in the mixture with 3%
C18 PSar47 starts at 100 Å2 /mol. The coexistence of solid and liquid analogous phase end at a surface
pressure of 7.8 mN/m and 60 Å2 /mol.
The area per molecule when the membranes collapse remains constant at around 35 Å2 /mol for
0%, 1%, 2%, and 3%. The surface pressure at the collapsing point is between 45 and 55 mN/m and thus
comparable to monolayers formed by DSPC alone (65 mN/m) (Figure 5). To prove that no polymer
was squeezed out during multiple compression and expension processes, cyclic measurements were
carried out up to a surface pressure of 25 mN/m. This pressure was chosen to be well below the
corruption point, but being in the solid-analogous phase. The hysteresis plots overlay with each other,
indicating that none or only a minor loss of polymer occurs even for the highest lipopolypeptoid
content of 3 mol% (Figure A2).

Figure 5. Isothermal surface pressure of a Langmuir-Blodgett monolayer at 25 ◦ C with different
amounts of lipopolypeptoids (pure DSPC, 1%, 2%, and 3% C18 PSar45 ).

4. Discussion
In contrast to the early work on lipopolypeptoids published by Gallot and coworkers [28,33]
lipopolypeptoids displayed narrow molecular weight distributions and low dispersity indices of
approximately 1.1 when synthesized according to the synthetic methods reported in this paper.
According to MALDI-TOF MS data there are no detectable side products. Furthermore, we could
demonstrate quantitative end group modification using anhydrides or isothiocyanates (ITC) by
NMR studies (1 H NMR and 1 H-DOSY-NMR). Therefore, we do not see any reason for lipopolymer
fractionation as reported by Gallot and coworkers [33]. Eventually the lack of control over the
polymerization reported by the authors is related to the suboptimal choice of chloroform as a solvent
for the reaction. As already demonstrated DMF [34–36], NMP or benzonitrile [18] should be preferred
for the synthesis of such systems, since they ensure dissolution and enable the living ring opening
polymerization of Sar-NCAs.
Kolliphor EL, formerly known as Cremophor EL, is a glycerol-based amphiphile, bearing about
35 PEG units in total and three hydrophobic tails. In comparison to the CMCs of Kolliphor EL of
90 mg/L lipopolypeptoids of comparable HLB values, e.g., C18 PSar11 , have a three-fold lower CMC of
27 mg/L [37]. In comparison to other PEGylated castor oils (e.g., PEG44 CO CMC: 958.2 mg/mL [38])
polysarcosine based lipopolypeptoids have an even ten-fold lower CMC at comparable HLB values.
Since the A2 parameters in water between PEG and PSar are practically identical (unpublished data),
this is somehow an unexpected result. But as PSar has an amide in each monomer unit, it has a less

130

Polymers 2016, 8, 427

flexible backbone compared to PEG. This leads to a stretching of the polymer resulting in a closer
packing and, therefore, to a higher micelle stability. Furthermore, the hydrophobic tails differ, so that
the stacking of the hydrophobic domains also deviates.
These findings relate to the performed DLS measurements. Diameters found for the micelles are
in the range of 9–25 nm. Polymers with a higher HLB value and a higher amount of polysarcosine
assemble into larger aggregates compared to those with a smaller degree of polymerization. These
studies show high dispersities on a micellar level ranging from 0.2–0.4. For C14 PSar103 (D = 0.7)
unimers and larger aggregates are detected in addition to the micellar fraction. Moreover, intensity
weighted DLS displays larger structures of several hundred nanometers for some lipopolypeptoids
(C18 PSar12 , C18 PSar30 , C14 PSar34 , and C14 PSar75 ). However, these larger aggregates do not appear in
the number weighted plot. This underlines that only a very small ratio assembles into large aggregates,
being overestimated by intensity-weighted DLS. In comparison with PSar-based block copolymers,
the lipopolypeptoid-based micelles are less uniform than those based on polypept(o)ides, [39] while
they are comparable with aggregates formed by amphiphilic block copolypeptoids [40]. Likely,
dispersities of assemblies can be lowered by methods for more controlled solution self-assembly,
extrusion, or other preparation techniques. A formulation of these lipopolypeptoids with, for example,
a lipid or hydrophobic molecules will lead to a better hydrophobic stabilization and, therefore, to more
uniform structures. These studies, also of great interest, are beyond the scope of the current article.
The reported amphiphiles with a low degree of polymerization, show toxicity at 500 μM, which
corresponds to 1.2 mg/mL. At 50 μmol no toxicity was detected in HeLa cells. Other polysarcosinebased materials have been reported to be non-toxic in HeLa cells up to 3 mg/mL by Birke et al. [34].
A polymeric hydrophobic block stabilizes aggregates more than an alkyl tail. The slower dynamics
in these micelles compared to the reported lipopolypeptoids lead to reduced interference with
cell membranes. Cremophor EL, as an example for a PEG-based system, shows toxicity at low
concentrations (0.1 g/L) in endothelial cells [41].
Langmuir monolayers of pure DSPC have a phase transition to the solid analogous phase at
55 Å2 /mol. This is the same value as reported by Hao et al. [42] It can also be seen, that with
increasing PSar content the isotherms of the lipopolymers are shifted to higher area/molecule and a
pseudo-plateau is reached at a surface pressure between 8 and 11 mN/m. PEG-based systems reported
by Tanwir also showed pseudo-plateaus with an onset of 8 and 9 mN/m for 1% and 3% PEGylated
lipid (degree of polymerization Xn = 45) [43]. The reported collapsing pressure of 59 mN/m is a little
higher than for our systems (45–55 mN/m). Since the degrees of polymerization are comparable,
this finding may be attributed to the differences in lipid parts of the lipopolymer. In comparison to the
PEGylated DPPE-based system reported by Tanwir the reported lipopolypeptoids bear only a single
alkyl chain.
5. Conclusions
In this work, we report synthetic pathways to lipopolypeptoids with adjustable HLB values,
precise control over molecular weights, dispersity indices, and end group integrity. The synthesized
lipopolymers are of amphiphilic nature and self-assemble into micelles or more complex aggregates
above their CMC in aqueous solution. The lipopolymers are non-toxic to cells up to a concentration of
50 μmol, which is a more than 10 times higher value comaqred to Cremophor EL. This finding points to
a potential application as excipients in drug formulations. In addition, C18 PSar45 -based lipopolymers
can be incorporated into Langmuir Blodgett monolayers based on DSPC up to a concentration of
3 mol % without altering its properties, which indicates the use of such lipopolypeptoids in the
preparation of stealth-like liposomes. Therefore, the reported experiments are a first indication that
polysarcosinylated lipids, named lipopolypeptoids, may be applied as bio-based excipients in drug or
lipid formulations.

131

Polymers 2016, 8, 427

Acknowledgments: All sources of funding of the study should be disclosed. Please clearly indicate grants that
you have received in support of your research work. Clearly state if you received funds for covering the costs to
publish in open access.
Author Contributions: Benjamin Weber, Regine Süss and Matthias Barz conceived and designed the experiments;
Benjamin Weber, Christine Seidl, David Schwiertz, Stefan Bleher and Martin Scherer performed the experiments;
Benjamin Weber, and Matthias Barz analyzed the data; Benjamin Weber and Matthias Barz wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

Appendix A

Figure A1. DOSY 1 H NMR of C18 PSar45 COOH (a) before, and (b) after removal of succinic anhydride.

Figure A2. Isothermic compression and expansion hysteresis at 25 ◦ C with 3% polymer content.

Figure A3. Number weighted diameter distribution of lipopolypeptoid determined by dynamic light
scattering (DLS). (a) Series of tetradecyl amine (C14 ) based lipopolypeptoids; and (b) Series of stearyl
amine (C18 ) based lipopolypeptoids.

132

Polymers 2016, 8, 427

References
1.
2.
3.
4.
5.
6.

7.
8.
9.
10.
11.
12.

13.

14.

15.
16.

17.
18.
19.
20.
21.

22.

23.

Kosaric, N. Biosurfactants: Production, Properties, Applications; CRC Press: New York, NY, USA, 1993.
Holmberg, K. Natural surfactants. Curr. Opin. Colloid Interface Sci. 2001, 6, 148–159. [CrossRef]
Raffa, P.; Wever, D.A.Z.; Picchioni, F.; Broekhuis, A.A. Polymeric surfactants: Synthesis, properties, and links
to applications. Chem. Rev. 2015, 115, 8504–8563. [CrossRef] [PubMed]
Lutz, J.F. Solution self-assembly of tailor-made macromolecular building blocks prepared by controlled
radical polymerization techniques. Polym. Int. 2006, 55, 979–993. [CrossRef]
Laschewsky, A. Molecular concepts, self-organisation and properties of polysoaps. In Polysoaps/Stabilizers/
Nitrogen-15 NMR; Springer: Berlin/Heidelberg, Germany, 1995; pp. 1–86.
Charalambopoulou, A.; Bokias, G.; Staikos, G. Template copolymerisation of N-isopropylacrylamide with
a cationic monomer: Influence of the template on the solution properties of the product. Polymer 2002, 43,
2637–2643. [CrossRef]
Clapés, P.; Infante, M.R. Amino acid-based surfactants: Enzymatic synthesis, properties and potential
applications. Biocatal. Biotransform. 2002, 20, 215–233. [CrossRef]
Yokota, H.; Sagawa, K.; Eguchi, C.; Takehara, M. New amphoteric surfactants derived from lysine. I. Preparation
and properties of Nε-acyllysine derivatives-acyllysine derivatives. J. Am. Oil Chem. Soc. 1985, 62, 1716–1719.
Baschang, G.; Hartmann, A.; Wacker, O. Lipopeptides Having Antitumor Activity. U.S. Patent 4666886 A,
19 May 1987.
Braun, D. Developments with lipoamino acids and their salts. Cosmet. Toilet. 1989, 104, 87–96.
Xia, J. Protein-Based Surfactants: Synthesis: Physicochemical Properties, and Applications; CRC Press: Boca Raton,
FL, USA, 2001.
Saadati, R.; Dadashzadeh, S.; Abbasian, Z.; Soleimanjahi, H. Accelerated blood clearance of PEGylated
PLGA nanoparticles following repeated injections: Effects of polymer dose, PEG coating, and encapsulated
anticancer drug. Pharm. Res. 2013, 30, 985–995. [CrossRef] [PubMed]
Ishida, T.; Harada, M.; Wang, X.Y.; Ichihara, M.; Irimura, K.; Kiwada, H. Accelerated blood clearance of
PEGylated liposomes following preceding liposome injection: Effects of lipid dose and PEG surface-density
and chain length of the first-dose liposomes. J. Control. Release 2005, 105, 305–317. [CrossRef] [PubMed]
Hara, E.; Makino, A.; Kurihara, K.; Yamamoto, F.; Ozeki, E.; Kimura, S. Pharmacokinetic change of
nanoparticulate formulation “Lactosome” on multiple administrations. Int. Immunopharmacol. 2012, 14,
261–266. [CrossRef] [PubMed]
Ulbricht, J.; Jordan, R.; Luxenhofer, R. On the biodegradability of polyethylene glycol, polypeptoids and
poly(2-oxazoline)s. Biomaterials 2014, 35, 4848–4861. [CrossRef] [PubMed]
Barz, M.; Luxenhofer, R.; Zentel, R.; Vicent, M.J. Overcoming the PEG-addiction: Well-defined alternatives to
PEG, from structure–property relationships to better defined therapeutics. Polym. Chem. 2011, 2, 1900–1918.
[CrossRef]
Gangloff, N.; Ulbricht, J.; Lorson, T.; Schlaad, H.; Luxenhofer, R. Peptoids and polypeptoids at the frontier of
supra- and macromolecular engineering. Chem. Rev. 2015, 116, 1753–1802. [CrossRef] [PubMed]
Luxenhofer, R.; Fetsch, C.; Grossmann, A. Polypeptoids: A perfect match for molecular definition and
macromolecular engineering? J. Polym. Sci. Part A 2013, 51, 2731–2752. [CrossRef]
Sisido, M.; Imanishi, Y.; Higashimura, T. Nuclear magnetic resonance spectra of poly(N-alkylamino acid)s.
Biopolymers 1972, 11, 399–408. [CrossRef] [PubMed]
Klinker, K.; Barz, M. Polypept(o)ides: Hybrid systems based on polypeptides and polypeptoids.
Macromol. Rapid Commun. 2015, 36, 1943–1957. [CrossRef] [PubMed]
Fetsch, C.; Grossmann, A.; Holz, L.; Nawroth, J.F.; Luxenhofer, R. Polypeptoids from N-substituted glycine
N-carboxyanhydrides: Hydrophilic, hydrophobic, and amphiphilic polymers with poisson distribution.
Macromolecules 2011, 44, 6746–6758. [CrossRef]
Gangloff, N.; Fetsch, C.; Luxenhofer, R. Polypeptoids by living ring-opening polymerization of N-substituted
N-carboxyanhydrides from solid supports. Macromol. Rapid Commun. 2013, 34, 997–1001. [CrossRef]
[PubMed]
Chapman, R.G.; Ostuni, E.; Liang, M.N.; Meluleni, G.; Kim, E.; Yan, L.; Pier, G.; Warren, H.S.; Whitesides, G.M.
Polymeric thin films that resist the adsorption of proteins and the adhesion of bacteria. Langmuir 2001, 17,
1225–1233. [CrossRef]

133

Polymers 2016, 8, 427

24.
25.
26.
27.

28.
29.

30.
31.
32.
33.
34.

35.

36.

37.
38.
39.
40.
41.

42.
43.

Lau, K.H.A.; Ren, C.; Park, S.H.; Szleifer, I.; Messersmith, P.B. An experimental-theoretical analysis of protein
adsorption on peptidomimetic polymer brushes. Langmuir 2012, 28, 2288–2298. [CrossRef] [PubMed]
Secker, C.; Brosnan, S.M.; Luxenhofer, R.; Schlaad, H. Poly(α-peptoid)s revisited: Synthesis, properties, and
use as biomaterial. Macromol. Biosci. 2015, 15, 881–891. [CrossRef] [PubMed]
Kidchob, T.; Kimura, S.; Imanishi, Y. Amphiphilic poly(Ala)-b-poly(Sar) microspheres loaded with
hydrophobic drug. J. Control. Release 1998, 51, 241–248. [CrossRef]
Yamamoto, F.; Yamahara, R.; Makino, A.; Kurihara, K.; Tsukada, H.; Hara, E.; Hara, I.; Kizaka-Kondoh, S.;
Ohkubo, Y.; Ozeki, E.; et al. Radiosynthesis and initial evaluation of (18)F labeled nanocarrier composed of
poly(L-lactic acid)-block-poly(sarcosine) amphiphilic polydepsipeptide. Nucl. Med. Biol. 2013, 40, 387–394.
[CrossRef] [PubMed]
Gallot, B. Liposarcosine-Based Polymerizable and Polymeric Surfactants; American Chemical Society: Washington,
DC, USA, 1991; pp. 103–113.
Trimpin, S.; Rouhanipour, A.; Az, R.; Räder, H.J.; Müllen, K. New aspects in matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry: A universal solvent-free sample preparation.
Rapid Commun. Mass Spectrom. 2001, 15, 1364–1373. [CrossRef] [PubMed]
Griffin, W.C. Classification of surface-active agents by “HLB”. J. Soc. Cosmet. Chem. 1949, 1, 311–326.
Naught, A.D.M.; Wilkinson, A. IUPAC Gold Book, 2nd ed.; Blackwell Scientific Publication: Oxford, UK, 1997.
Strickley, R.G. Solubilizing excipients in oral and injectable formulations. Pharm. Res. 2004, 21, 201–230.
[CrossRef] [PubMed]
Douy, A.; Gallot, B. New amphipathic lipopeptides, 1. Synthesis and mesomorphic structures of lipopeptides
with polysarcosine peptidic chains. Die Macromol. Chem. 1986, 187, 465–483.
Birke, A.; Huesmann, D.; Kelsch, A.; Weilbächer, M.; Xie, J.; Bros, M.; Bopp, T.; Becker, C.; Landfester, K.;
Barz, M. Polypeptoid-block-polypeptide copolymers: Synthesis, characterization, and application of
amphiphilic block Copolypept(o)ides in drug formulations and miniemulsion techniques. Biomacromolecules
2014, 15, 548–557. [CrossRef] [PubMed]
Heller, P.; Weber, B.; Birke, A.; Barz, M. Synthesis and sequential deprotection of triblock copolypept(o)ides
using orthogonal protective group chemistry. Macromol. Rapid Commun. 2015, 36, 38–44. [CrossRef]
[PubMed]
Holm, R.; Klinker, K.; Weber, B.; Barz, M. Synthesis of Amphiphilic Block Copolypept(o)ides by Bifunctional
Initiators: Making PeptoMicelles Redox Sensitive. Macromol. Rapid Commun. 2015, 36, 2083–2091. [CrossRef]
[PubMed]
Kessel, D. Properties of cremophor EL micelles probed by fluorescence. Photochem. Photobiol. 1992, 56,
447–451. [CrossRef] [PubMed]
El-Azab, W.I.M.; Aiad, I.; Azzam, E.M.S.; Gad, E.A.M. Molecular, Surface, and thermodynamic properties of
nonionic surfactants based on castor oil. J. Dispers. Sci. Technol. 2010, 31, 1150–1156. [CrossRef]
Huesmann, D.; Sevenich, A.; Weber, B.; Barz, M. A head-to-head comparison of poly(sarcosine) and
poly(ethylene glycol) in peptidic, amphiphilic block copolymers. Polymer 2015, 67, 240–248. [CrossRef]
Fetsch, C.; Gaitzsch, J.; Messager, L.; Battaglia, G.; Luxenhofer, R. Self-assembly of amphiphilic block
copolypeptoids—Micelles, worms and polymersomes. Sci. Rep. 2016, 6, 33491. [CrossRef] [PubMed]
Kiss, L.; Walter, F.R.; Bocsik, A.; Veszelka, S.; Ózsvári, B.; Puskás, L.G.; Szabó-Révész, P.; Deli, M.A. Kinetic
analysis of the toxicity of pharmaceutical excipients cremophor EL and RH40 on endothelial and epithelial
cells. J. Pharm. Sci. 2013, 102, 1173–1181. [CrossRef] [PubMed]
Hao, C.; Liu, Q.; Li, Q.; Zhang, J.; Sun, R. Thermodynamic and structural studies of DMPC and DSPC with
DOTAP mixed monolayers at the air–water interface 1. Russ. J. Phys. Chem. A 2016, 90, 214–219. [CrossRef]
Tanwir, K.; Tsoukanova, V. Lateral distribution of a poly(ethylene glycol)-grafted phospholipid in
phosphocholine monolayers studied by epifluorescence microscopy. Langmuir 2008, 24, 14078–14087.
[CrossRef] [PubMed]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

134

polymers
Review

Dynamics of Polymer Translocation: A Short Review
with an Introduction of Weakly-Driven Regime
Takahiro Sakaue 1,2
1
2

Department of Physics, Kyushu University, Fukuoka 819-0395, Japan; sakaue@phys.kyushu-u.ac.jp;
Tel.: +81-92-802-4066
Precursory Research for Embryonic Science and Technology (PRESTO),
Japan Science and Technology Agency, Honcho Kawaguchi, Saitama 332-0012, Japan

Academic Editors: Alexander Böker and Frank Wiesbrock
Received: 14 October 2016; Accepted: 1 December 2016; Published: 7 December 2016

Abstract: As emphasized in a recent review (by V.V. Palyulin, T. Ala-Nissila, R. Metzler), theoretical
understanding of the unbiased polymer translocation lags behind that of the (strongly) driven
translocation. Here, we suggest the introduction of a weakly-driven regime, as described by the linear
response theory to the unbiased regime, which is followed by the strongly-driven regime beyond the
onset of nonlinear response. This provides a concise crossover scenario, bridging the unbiased to
strongly-driven regimes.
Keywords: polymer translocation; memory effect; generalized Langevin equation; nonequilibrium
dynamic; tension propagation

1. Introduction
Macromolecules can be transported through a nanoscale pore by threading it. This process of
polymer translocation has been extensively studied for the last two decades [1–8]. The motivation
for the research arises from its relevance to biopolymer transport in living cells, and most notably,
its connection to the nanopore-based new genome sequencing technique [9].
There are various important factors in the problem, including the electrokinetic effect, the pore
properties (size, geometry, and interaction with polymer, etc.), and the capture process of polymer
into the pore. Among others, one can ask how to characterize the very process of the polymer going
through a simple pore, which is a purely a problem of polymer dynamics. The central quantity here is
the translocation time τ as a function of the chain length N and the driving force f . Many efforts have
been devoted to clarifying the scaling formula of τ, which is expected to be universal for long enough
polymers [7,8].
In the literature, the situation is categorized either by unbiased ( f = 0) or driven (finite f )
translocation, depending on the absence or the presence of a driving force. In either case, an essential
aspect in the problem lies in the collective dynamics of polymer associated with the tension propagation,
which is manifested by the anomalous dynamics of the translocation coordinate (see Figure 1). In the
present paper, we shall introduce the weakly-driven translocation regime in between. As will be
shown, this provides a concise crossover scenario connecting three regimes. In Section 2, we first recall
the memory effect approach to the unbiased translocation proposed by Panja, Barkema, and Ball [10].
In Section 3 , we then construct the weakly-driven regime through the linear response analysis to the
unbiased regime. For larger driving force beyond the onset of nonlinear response f > k B T/( aN ν )
(a: monomer size, k B T: thermal energy), we enter the strongly-driven regime—the basic features of
which are, by now, well understood. (For early theoretical and numerical attempts, see References [11–15].
The subsequent revision to the approach in References [11–13] can be found in References [16–22].)
We shall briefly review it in Section 4. Summary and discussions are given in Section 5.
Polymers 2016, 8, 424

135

www.mdpi.com/journal/polymers

Polymers 2016, 8, 424

N

(trans)

(cis)

no+m(t)
no+1

a[m(t)]ν

no-m(t)

0

a[m(t)]ν

Figure 1. Illustration of a translocating polymer and the transport operation (Δn = 1) to measure the
stress relaxation associated with the tension propagation. The translocation coordinate is defined as
the monomer label n(t) at the pore, which counts the number of monomers already in the trans side at
time t in analogy to the reaction coordinate in chemical reaction process [7,8,23,24].

2. Unbiased Translocation
We first summarize self-similar dynamical properties of a flexible polymer, which will be used
in subsequent discussions. Consider a partial section (with m < N bonds) of a long polymer.
Its characteristic spatial size is Rm  amν , with ν  3/5 in 3-dimensional space. Associated with
it is the characteristic time τm  τ0 ( Rm /a)z , where τ0 is a monomer time scale, and the dynamical
exponent is z = 2 + ν−1 in the case of free-draining dynamics (see Appendix A.1).
Suggested from these two relations is
m(t)  (t/τ0 )1/(νz)

(1)

which describes how the tension created by perturbations propagates along the chain. By setting
m(τp )  N, we find the propagation time
τp  τ0 N νz

(2)

at which the tension reaches the chain end. This is nothing but the longest time τN for the
conformational relaxation.
2.1. Memory Function for Stress Relaxation
Panja et al. suggested that—due to the monomer exchange across the pore—there exists an
imbalance in the tension near the pore, which is responsible for the subdiffusion of the translocation
coordinate [10]. The problem can thus be naturally analyzed within the framework of the linear
response theory. Consider a polymer going through a narrow pore from the head (n = 0). Assume the
polymer is in equilibrium with n0 bonds in the trans side of the membrane, while the remaining
N − n0 bonds are on the cis side; i.e., the monomer’s label at the pore is n(t) = n0 (t < 0) (Figure 1).
This may be realized by letting the polymer equilibrate with the immobilization constraint imposed
on the monomer n0 at the pore. Then, at t = 0, we instantly transport Δn monomer from cis to trans.
By this operation, the translocation coordinate n(t) is changed from n0 to n0 + Δn. The polymer in the
trans side is compressed, while it is stretched in the cis side, which produces the restoring force to the
translocation coordinate.
To keep the imposed change in the translocation coordinate n(t) = n0 + Δn at t > 0, the force is
required whose magnitude decreases with time along with the conformational relaxation. This process
can be analyzed by the force balance equation
 t
t0

ds Γ(t − s)v(s) = f (t)

136

(3)

Polymers 2016, 8, 424

where v(t) = dn(t)/dt, and we may set the lower bound of the time integral as t0 → −∞ by assuming
the system is already in the equilibrium state before the operation is made. In the case of step
displacement Δn imposed at t = 0 (i.e., n(t + 0) = n(t − 0) + Δn), we have v(t) = Δnδ(t). The above
equation is simplified to
ΔnΓ(t) = f (t)

(4)

To evaluate the memory kernel Γ(t), it is important to realize that the entire chain cannot respond to the
operation at once. At time t < τp , only a finite section with m(t) bonds given by Equation (1) close to
the pore can respond to the operation. The deformation of such a responding chain section on the trans
side is evaluated as ΔRm = am(t)ν − a {m(t) + Δn}ν  − a[m(t)ν−1 Δn]ν (Figure 1). The responding
section on the trans side is thus compressed to the amount ∼ a[m(t)ν−1 Δn], which exerts a restoring
force to the translocation coordinate. The magnitude of the force can be evaluated by noting that the
responding domain with m(t) bonds acts as an entropic spring with the spring constant  k B T/Rm (t)2 .
We thus find [10]
f (t)





kB T
|ΔRm |
R m ( t )2
kB T
[νΔn]
am(t)1+ν
 −(1+ν)/(νz)
t
kB T
νΔn
a
τ0

(5)

where Equation (1) is used in the last line. For the quantitative discussion, one should note that the
same force arises due to the cis side stretching, so the net result doubles.
Comparison of Equations (5) with (4) yields the memory kernel
Γ(t) 

kB T
ν
a



t
τ0

−α
(6)

with the stress relaxation exponent
α=

1+ν
νz

(7)

In general, we have 0 < α < 1, reflecting the viscoelastic nature of the response.
2.2. Unbiased Translocation Dynamics
To connect the average stress relaxation with the subdiffusion of the translocation coordinate,
we need to look at each realization of the stochastic processes. One is then led to the generalized
Langevin equation by adding the thermal noise term ξ (t) to the right-hand side of Equation (3), which
is zero mean ξ (t) = 0, and related to the memory kernel via the fluctuation-dissipation theorem
ξ (t)ξ (s) = k B TΓ(|t − s|). The equivalent expression of the generalized Langevin equation is
v(t) =

 t
t0

ds μ(t − s) f (s) + η (t)

(8)

with the mobility kernel μ(t) ∼ −tα−2 and the noise η (t), which again is related to the kernel as
η (t)η (s) = k B Tμ(|t − s|). In the unbiased case f (t) = 0, the mean-square displacement (MSD) can
be derived after integration of the velocity correlation function v(t)v(s) = η (t)η (s) twice with

137

Polymers 2016, 8, 424

respect to time, yielding δn(t)2  ∼ tα , where δn(t) = n(t) − n(0); i.e., the stress relaxation exponent
is equal to the MSD exponent. Therefore, the MSD in translocation coordinate space is obtained as

δn(t)2   (t/τ0 )(1+ν)/(νz)

(9)

2.3. Traveled Fraction at t = τp
The memory kernels obtained for the translocating polymer arise from the viscoelastic response
of the polymer due to the tension propagation. Therefore, the memory persists up to the propagation
time τp given in Equation (2). To find how τp is related to the average time τ for the translocation
process, let us calculate the characteristic displacement of the translocation coordinate at the time scale
τp . We find from Equation (9)

δn(τp )2   N 1+ν  N 2

(10)

Now let us define the quantity

QN ≡

δn(τp )2 
N

(11)

which measures a fraction of the total chain length, which is traveled by the time t = τp .
Using Equation (10), we find
Q N ∼ N −(1−ν)/2  1

(12)

is vanishingly small for asymptotically large N. This indicates that the majority of the monomers
do not pass the pore by t = τp ; i.e., τ τp , so there should be another process to characterize the
translocation dynamics. This requires us to think about the post-propagation stage at t > τp .
2.4. Post-Propagation Stage
At t > τp , the tension has already been propagated up to the chain end, so the motion of
all the chain sections is coherent to generate the ordinary diffusion of the translocation coordinate.
Here, the relevant question is what is the diffusion coefficient Dn for it?
The following matching argument at t = τp

δn(τp )2   Dn τp ,

(13)

where the left-hand side is evaluated by Equation (9), suggests
Dn  τ0−1 N 1+ν−νz

(14)

The characteristic time τpp for the post-propagation stage is thus obtained as
τpp  [ N (1 − Q N )]2 /Dn  N 1+(z−1)ν

(15)

where the last near-equality is based on the estimation Q N  1. Since τpp /τp 1, the scaling formula
for the translocation time in the long chain limit becomes τ = τp + τpp ∼ τpp [10].
3. Weakly-Driven Dynamics
The process may be biased by the external force. For translocating polymers, the force is acting at
the pore, which is realized, for instance, by the voltage drop across the pore. As long as the force is
weak enough f < k B T/R N , the tension dynamics summarized in the beginning of Section 2 is intact,
and the generic linear response argument yields the average dynamics of the translocation coordinate.
138

Polymers 2016, 8, 424

Suppose the system is already in equilibrium, and we start to apply a constant external force f at t = 0.
Equation (8) then simplifies to
dn(t)
= f
dt

 t
0

ds μ(s) ∼ f tα−1

(16)

where, as written after Equation (8), μ(t) ∼ −tα−2 , with 0 < α < 1. The average drift is thus
evaluated as

δn(t) ∼ f tα

(17)

3.1. Weakly-Driven Translocation Dynamics
From Equations (7) and (17), the average drift is

δn(t) ∼ f t(1+ν)/(νz)

(18)

Therefore, the drift distance at the propagation time

δn(τp ) ∼ N 1+ν f

(19)

We thus find the fraction of the system explored by the propagation time
(f)

QN ≡

δn(τp )
∼ Nν f < 1
N

(20)

where the last inequality comes from f < k B T/R N ∼ N −ν .
3.2. Post-Propagation Stage
(f)

The result Q N < 1 for a weakly-driven translocating polymer indicates that the translocation
coordinate can explore only a fraction of the system by the propagation time. This conclusion is
intact even with the fluctuation effect superimposed, since n(t) − n(t) is (according to Equation (8))
described by the same dynamical equation as the unbiased dynamics, for which Q N < 1 as well.
Therefore, the post-propagation stage at t > τp may become an essential part to determine the whole
translocation time τ. Since the tension has already reached the chain ends at t > τp , we expect the
normal drift

δn(t) 

f
t
γn

(t > τp )

(21)

where the friction coefficient
γn ∼ N zν−1−ν

(22)

can be determined by matching Equations (19) and (21) at t = τp (note the consistency of Equations (22)
with (14) through Einstein relation). The characteristic time τpp for the post-propagation stage is thus
found from the relation ( f /γn )τpp = N − δn(τp ); therefore,
τpp 

γn
N ( z −1) ν
(f)
N (1 − Q N ) ∼
f
f

(23)

One can check the dominance of the post-propagation stage from the ratio τpp /τp > 1 in the
linear response regime. Comparing the two time scales τpp in unbiased and weakly-driven regimes,
respectively, given by Equations (15) and (23), we find a characteristic force f ∼ N −1 , above which
there exists a scaling regime of weakly-driven translocation where τ  τpp given by Equation (23)
139

Polymers 2016, 8, 424

applies. For weaker force f < 1/N, the fluctuation effect dominates over the average drift, so we
expect a crossover to unbiased scaling, where τ  τpp given by Equation (15) holds.
4. Strongly-Driven Dynamics
As pointed out in Reference [11,12] and verified in subsequent works [14,15,17], the nonequilibrium
dynamical effect becomes relevant for translocating polymer, which is driven by strong force.
Suppose we grab one end of a polymer (with N monomers), and start to pull it by a constant force f .
When the force is weak enough f < k B T/R N , the whole polymer will follow the force without
noticeable conformational distortion. For a larger force, however, the polymer will be elongated along
the direction of the pulling [25]. The first question here is the relation among the force, the moving
velocity, and the elongation in steady state. Such a relation—which we may call a dynamical equation
of state—is discussed in detail in Reference [26]. The second question is how we can characterize the
transient process toward such an elongated steady state from an initial quiescent state. As one can
infer from the scaling form of the threshold force ∼ k B T/R N ∼ N −ν , such a nonequilibrium effect
shows up with rather moderate force; the “strongly-driven” appellation is only adopted to contrast to
the weakly-driven regime discussed in Section 3.
Given a long relaxation time of the whole polymer τN  τ0 ( Rn /a)z  τ0 N νz , it would be
intuitively clear that the polymer as a whole cannot respond to the pulling force all at once. Instead, only
the subchain part close to the pulled site can initially respond, and thus collectively move in the
pulled direction. Such a responding moving domain will grow with time, the dynamics of which is
associated with how the tensile force propagates along the chain backbone [11,12,26] (see also [27]).
To describe this sort of nonequilibrium response, it is useful to picture the whole polymer as composed
of two distinct domains; a moving and a quiescent domain—the latter of which is yet unaware of the
pulling force at a given moment. In this two-phase picture, we are interested in the dynamics of the
tension front (i.e., domain boundary), which dictates the essential physics in the driven translocation.
Here we follow the argument in Reference [20] (see Appendix A.2). Now, in contrast to the
weakly-driven regime (Figure 1), There appears significant dynamical asymmetry between cis and
trans sides [28]. To set the stage, let us look at Figure 2. There is a thin wall at x = 0 with a small pore,
where the driving force with the constant magnitude f > k B T/R N is locally exerted in the x-direction
from the cis to the trans side. One chain end is initially sucked at time t = 0. The translocation process,
then, proceeds with the tension propagation along the chain. Monomers are numbered from the
first sucked end to the other end (Nth monomer). The moving domain at time t is specified by the
monomer m(t) at the end of the moving domain, the monomer n(t) at the pore, the size R(t), and the
representative (or average) velocity V (t). These can be determined by the following set of equations:
V (t) R(t)



f z −2
−(1−ν)/ν

m(t) − n(t)



Rf

m(t)ν



R(t)

(24)

( Rσ0 )

(25)
(26)

As explained in Reference [13,26], Equations (24) and (25) are the dynamical equations
of state describing the steady-state relation among velocity–extension–force (Equation (24)) and
mass–extension–force (Equation (25)) for a dragged polymer. The information on the initial coiled
conformation prior to the translocation process is contained in Equation (26). Note that σ0  g f /ξ f 
f −(1−ν)/ν , where ξ f  agνf  k B T/ f is the size of tensed blob in the close proximity of the pore, is the
monomer line density at the pore, so that the monomer flux at the pore is
dn(t)
= j0 (t) = σ0 v0 (t)
dt
where v0 (t) is the velocity of the monomer at the pore site.
140

(27)

Polymers 2016, 8, 424

RN
V(t)

V(t) f

m(t)

R(t)

R(t)

n(t)

f

n(t)

Figure 2. Sketch of a translocating polymer driven by a strong force f . (left) Propagation stage:
a growing moving domain (with velocity V (t) and the size R(t)) on the cis side is shaded, while the
chain portion already on the trans side is represented by a dashed curve. (right) Post-propagation
stage: the tension has already reached to far end of the polymer; thus, m(t) = N is constant, and the
moving domain is shrinking with time. In addition, most of monomers are already on the trans side,
so this post-propagation stage adds a finite-size correction to the scaling formula of the translocation time.

To solve the above set of equations, we adopt the ansatz v0 (t) = V (t) as in Reference [16,18],
which amounts to correspond to the iso-flux condition [16,20]. Then, combining Equations (25) and (27),
we obtain


dR(t)
dm(t)
(28)
σ0 V (t) +

dt
dt
dR(t)
dR(t)
⇔
f (z−2)−[(1−ν)/ν] R(t)−1  R(t)(1−ν)/ν
− f −(1−ν)/ν
dt
dt
(29)
where Equations (24) and (26), and the expression for the line density at the pore (see Equation (25))
are used. This leads to the tension propagation law

ν/(1+ν)
R(t) ∼ t f (z−2)−[(1−ν)/ν]

(30)

which would be valid asymptotically under the condition f R(t)/k B T 1. The scaling formula for
(f)

propagation time is thus obtained by setting R(τp ) = R N ;
(f)

τp

∼ N01+ν f −(z−2)+[(1−ν)/ν]

(31)

Note that the above propagation time can be written as
(f)

τp



∼ τN

f RN
kB T

1+(1/ν)−z

< τN

(32)

where the last inequality—valid in the present condition f > k B T/R N —represents the fact that the
tension propagation takes place in a time scale shorter than the conformational relaxation time, a clear
indication of the nonequilibrium nature of the strongly-driven translocation process. Compare this
result with the propagation time Equation (2) for equilibrium (unbiased and weakly-driven) regime.
(f)

After t = τp , the post-propagation stage follows, which can be analyzed by Equation (28),
but with dm(t)/dt = 0 on the right-hand side [13]. A qualitative difference in the dynamical feature

141

Polymers 2016, 8, 424

between propagation and post-propagation stages can be found in the growth rate of n(t); while the
slowing-down in the propagation stage is a consequence of the growth of the moving domain,
the speed-up in the post-propagation stage is due to the decrease of the overall friction with
the process advanced. In contrast to the weakly-driven regime (Section 3), it can be shown that
most of the monomers (in a scaling sense) are transported to the trans side in the propagation
stage. Therefore, the scaling formula for the translocation time is given by the propagation time
(f)

(i.e., τ  τp ) [11,12,16,19] (Note, however, that this does not hold for a semiflexible filament [29]).
To elucidate the crossover between weak and strongly-driven regimes, let us introduce the
characteristic size of a tensed blob ξ f  k B T/ f and the corresponding time scale τ f 0  τ0 (ξ f /a)z .
In the length scale smaller than ξ f , the force is considered to be a weak perturbation. Therefore,
under the force with the strength k B T/R N < f < k B T/a, the physics in the weakly-driven regime
applies up to t = τ f 0 , where the tension propagates according to Equation (1). The nonequilibrium
effect and the cis–trans dynamical asymmetry manifests in larger length and time scales. In this
sense, the length ξ f and time τ f 0 play a role of “initial conditions” for the subsequent nonequilibrium
dynamics. In this way, one can show that at f > k B T/R N , the tension propagation stage dominates the
whole translocation process asymptotically; hence, a natural crossover occurs between two regimes at
that force scale.
5. Summary and Discussions
5.1. Summary of the Scaling Formulae
As stated in Introduction, one of the main purposes of the present note is, aside from reviewing
recent progress in the field, to introduce a weakly-driven regime as the linear response domain to the
unbiased regime. To show that this provides a concise crossover scenario bridging the unbiased to
the strongly-driven regime, let us summarize the scaling formulae for the translocation dynamics in
different regimes, according to the classification scheme proposed here.
5.1.1. Unbiased and Weakly-Driven Regimes
These two regimes may be termed collectively as the equilibrium regime. The statistical
dynamics of the translocation coordinate—characterized here by its first and second moments of
the displacement—follows


f t(1+ν)/(νz) (t < τp )
γn−1 f t
(τp < t < τ )

t(1+ν)/(νz) (t < τp )
{δn(t) − δn(t)}2  ∼
Dn t
(τp < t < τ )

δn(t) ∼

(33)
(34)

where γn ∼ N νz−1−ν and Dn ∼ N 1+ν−νz are the effective friction and diffusion coefficients of the
translocation coordinate in the post-propagation stage. For unbiased case ( f = 0), the drift vanishes,
so Equation (34) reduces to the MSD.
The translocation time is dominated by the post-propagation stage τ ∼ τpp , with

τpp ∼

N 1+(z−1)ν
N ( z −1) ν / f

( f < N −1 )
( N −1 < f < N − ν )

(35)

Comparing the respective characteristic times of post-propagation stages, the characteristic force
f  k B T/( Na) is found; for weaker force, the bias is so weak that the process is essentially unbiased.
Note that the same formula for τ was proposed in Equation (4) of Reference [12], based on the
“equilibrium shape assumption”.

142

Polymers 2016, 8, 424

5.1.2. Strongly-Driven Regime
As described in Section 4, for stronger force f > k B T/R N , one has to take the nonequilibrium
conformational deformation dynamics into account [11,12]. In such a situation, the response
becomes generally nonlinear, and the dynamical asymmetry between cis and trans sides appears [28].
This qualitatively changes the dynamics of tension propagation, hence the structure of the memory
effect for the evolution of the translocation coordinate [11–22]. After the initial tensed blob of size
ξ f  k B T/ f in the vicinity of pore is formed (i.e., t > τ f 0 ), its average time evolution is predicted to
follow (see Appendix A.3)


n(t) ∼

−1 ) / (1+ ν )

t1/(1+ν) f (z−1−ν
t1/(1+ν) f 1/(1+ν)

( N −ν < f < N 0 )
( N0 < f )

(36)

where the characteristic force f  k B T/a separates the so-called trumpet and stem-flower regimes [25].
(f)

This process persists up to the propagation time τp
depends on f ;

(f)

τp

∼

(f)

(f)

given by m(τp ) ∼ n(τp ) ∼ N, which now

N 1+ν f 1+(1/ν)−z
N 1+ ν f −1

( N −ν < f < N 0 )
( N0 < f )

(37)

(f)

The post-propagation stage follows at t > τp [13]. However, in contrast to the unbiased and
the weakly-driven regimes, the propagation stage dominates over the post-propagation stage in the
(f)

sense that the asymptotic translocation time is predicted to be τ ∼ τp [16,19]. Note that—despite
the conspicuous difference in their underlying physics—scaling formulae Equations (35) and (37)
are identical in the case of free-draining dynamics z = 2 + ν−1 . If not, they differ in their scaling
structures, but one can check a smooth crossover at f ∼ N −ν . These results on the scaling formulae of
the translocation time are summarized in Figure 3.

τ

τ

N 1+(z-1)ν
N νz
N 1+ν

-1

f

1+ν-1-z
( -1)

f

-1
UB

WD

N -1 N -ν

SD
(T)

SD
(S-F)

1

-z

-1-(z-1)ν

1+ν
(z-1)ν
( 1+ν)

1+(z-1)ν

( 2+ν)

UB

f

SD
(T)

WD

f

-1

f

-1/ν

N

Figure 3. Dependence of the translocation time on f (left) and N (right) shown in double logarithmic
scale. The locations of various regimes, unbiased (UB), weakly-driven (WD) and strongly-driven
(SD) are specified; the SD regime is further divided into the trumpet (T) and the stem–flower (S-F)
regimes. Note that in the right graph, depicted is the case with f < k B T/a; otherwise, we have only S-F
regime with the slope 1 + ν. Note also that in these plots, we set z = 2 + ν−1 (free draining dynamics),
in which case the plots become particularly simple. The triangles and their nearby numbers designate
slopes (exponents), where the numbers after the arrows specify the values for free draining dynamics.
For other choices of the dynamical exponent (i.e., z = 3 for nondraining (Zimm) dynamics), the slope
in the SD (T) regime in the left graph is changed. The same applies to UB and WD regimes in the right
graph. Comparing these plots with Figure 4 in Reference [16], one finds differences in weak force and
short chain length regions.

143

Polymers 2016, 8, 424

5.2. Discussion
Following the memory effect approach suggested by Panja et al. [10], we have argued Q N < 1 for
the translocation dynamics in unbiased and weakly driven regimes; thus, τ ∼ τpp scales differently
from τp . If, on the contrary, one assumes τ ∼ τp as in Chuang, Kantor, and Kardar [30], one ends up
with the different scaling for the anomalous diffusion of the translocation coordinate. At present,
there seems to be no definite conclusion on which argument is more appropriate, but let us note the
following points on this issue from the literature.
In their Langevin dynamics simulation, de Haan and Slater have demonstrated the increasing
impact of the memory effect on the translocation dynamics with the increase in the solution viscosity
and the chain length [31,32]. In such a memory-predominant situation, they have shown that
(i) the post-propagation stage dominates the translocation time, and (ii) the translocation time scaling
is close to τ ∼ N 2+ν , which is Equation (35) with z = 2 + ν−1 , in accordance with Panja et al.
(see Appendix A.4). However, several reports on the numerically-estimated subdiffusion exponent in
the propagation stage do not match the value in Equation (34) [31–33]. The reported exponent looks to
be closer to the value suggested by Chuang et al. [30].
It should be kept in mind that we have only considered the asymptotic scaling, which would
be valid in the long chain limit. For real situations, a significant finite-size effect would come into
play. For strongly-driven translocation, the role of pore friction has been recently elucidated [8,17].
Similar effects would be likely for unbiased and weakly-driven translocations as well, and it might be
a possible source for the puzzling observation mentioned above.
In Reference [34], the scaling for the anomalous drift

n(t) ∼ ( f t)(1+ν)/(1+2ν)

(38)

has been proposed based on the combination of memory effect argument similar to ours and
the numerical simulation results. However, this is different from our linear response prediction
Equation (18), though the exponent on t is the same (note z = 2 + ν−1 for the free draining dynamics,
so the time exponent is (1 + ν)/(νz) = (1 + ν)/(1 + 2ν)). Because of this, a scaling formula for τ
different from Equation (35) was proposed in Reference [34], but this does not seem to provide a clear
crossover scenario both to unbiased and strongly-driven regimes.
In conclusion, we have suggested the introduction of the weakly-driven regime for polymer
translocation dynamics, which is naturally described by the linear response theory applied to the
unbiased regime. A similar discussion on weakly and strongly-driven dynamics for a polymer
pulled by mechanical force has been recently done in Reference [35]. We hope the resultant concise
crossover scenario will be useful to promote the understanding of polymer translocation dynamics in
near-equilibrium situation, which is necessary to unveil a full picture of the phenomenon.
Acknowledgments: I thank D. Panja for discussion and correspondence (which we did in the year 2013),
and T. Saito for fruitful collaboration. This work is supported by KAKENHI (No. 16H00804, “Fluctuation and
Structure”) from MEXT, Japan, and JST, PREST.
Conflicts of Interest: The author declares no conflict of interest.

Appendix A.
Appendix A.1.
These two critical exponents are useful for the development of a general scenario. The static
exponent ν (often called the Flory exponent in polymer literature)—being the inverse of the
fractal dimension—quantifies the polymer’s spatial extension as a function of its molecular weight.
While ν = 1/2 for ideal chains, the swelling due to the excluded-volume effect results in a larger
value, which depends on the space dimension. On the other hand, the dynamical exponent z is
associated with the dissipation mechanism at work. In the simplest model (as adopted in many

144

Polymers 2016, 8, 424

of simulation studies), the motion of monomers caused by the flow of the surrounding medium is
neglected. This free-draining dynamics (sometimes called local friction or Rouse dynamics) amounts
to set z = 2 + ν−1 . In dilute solution, however, the solvent-mediated hydrodynamic interaction is
known to be relevant. The resultant non-draining dynamics can be approximated by setting z = 3
(Zimm dynamics), at least, in free space.
Appendix A.2.
This preprint [20] was prepared for a supplemental to Reference [19]. The latter is an erratum to
Reference [13], where we have modified earlier scaling predictions [11–13] to be consistent with the
appropriate mass conservation relation across the pore, as pointed out in Reference [16].
Appendix A.3.
This relation is deduced from Equation (30) with the relation m(t) ∼ n(t), which is expected to be
valid asymptotically; see Reference [20].
Appendix A.4.
Notice, however, that the scaling τ ∼ N 2 was observed in Reference [31,32] in the opposite limit of
low viscosity, where the polymer conformational relaxation is rapid. The same scaling is also observed
in the limit of high imposed pore friction [36]; see also earlier works [23,24]. For practical purposes,
one may add the solvent viscosity and the pore friction as extra parameters (possibly their combination
with the chain length) to determine the appropriate regime.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.

Meller, A. Dynamics of polynucleotide transport through nanometre-scale pores. J. Phys. Condens. Matter
2003, 15, R581–R607.
Dekker, C. Solid-state nanopores. Nat. Nanotechnol. 2007, 2, 209.
Keyser, U.F. Controlling molecular transport through nanopores. J. R. Soc. Interface 2011, 8, 1369–1378.
Aksimentiev, A. Deciphering ionic current signatures of DNA transport through a nanopore. Nanoscale 2011,
2, 468–483.
Wanunu, M. Nanopores: A journey towards DNA sequencing. Phys. Life Rev. 2012, 9, 125–158.
Muthukumar, M. Polymer Translocation; CRC Press: Boca Raton, FL, USA, 2011.
Panja, D.; Barkema, G.T.; Kolomeisky, A.B. Through the Eye of the Needle: Recent Advances in
Understanding Biopolymer Translocation. J. Phys. Condens. Matter 2013, 25, 413101.
Palyulin, V.V.; Ala-Nissila, T.; Metzler, R. Polymer translocation: the first two decades and the recent
diversification. Soft Matter 2014, 10, 9016–9037.
Branton, D.; Deamer, D.W.; Marziali, A.; Bayley, H.; Benner, S.A.; Butler, T.; Di Ventra, M.; Garaj, S.; Hibbs, A.;
Huang, X.; et al. The potential and challenges of nanopore sequencing. Nat. Biotechnol. 2008, 26, 1146–1153.
Panja, D.; Barkema, G.T.; Ball, R.C. Anomalous dynamics of unbiased polymer translocation through a
narrow pore. J. Phys. Condens. Matter 2007, 19, 432202.
Sakaue, T. Nonequilibrium dynamics of polymer translocation and straightening. Phys. Rev. E 2007,
76, 021803.
Sakaue, T. Sucking genes into pores: Insight into driven translocation. Phys. Rev. E 2010, 81, 041808.
Saito, T.; Sakaue, T. Dynamical diagram and scaling in polymer driven translocation. Eur. Phys. J. E 2011, 34, 135.
Lehtola, V.V.; Linna, R.P.; Kaski, K. Dynamics of forced biopolymer translocation. Europhys. Lett. 2009,
85, 58006.
Bhattacharya, A.; Binder, K. Out-of-equilibrium characteristics of a forced translocating chain through a
nanopore. Phys. Rev. E, 2010, 81, 041804.
Rowghanian, P.; Grosberg, A.Y. Force-Driven Polymer Translocation through a Nanopore: An Old Problem
Revisited. J. Phys. Chem. B 2011, 115, 14127.
Ikonen, T.; Bhattacharya, A.; Ala-Nissila, T.; Sung, W. Influence of non-universal effects on dynamical scaling
in driven polymer translocation. J. Chem. Phys. 2012, 137, 085101.
145

Polymers 2016, 8, 424

18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.

33.
34.
35.
36.

Dubbeldam, J.L.A.; Rostiashvili, V.G.; Milchev, A.; Vilgis, T.A. Forced translocation of a polymer: Dynamical
scaling versus molecular dynamics simulation. Phys. Rev. E 2012, 85, 041801.
Saito, T.; Sakaue, T. Erratum to: Dynamical diagram and scaling in polymer driven translocation. Eur. Phys. J. E
2012, 35, 125.
Saito, T.; Sakaue, T. Two phase picture in driven polymer translocation. arXiv 2012, arXiv:1205.3861v3.
Saito, T.; Sakaue, T. Process time distribution of driven polymer transport. Phys. Rev. E 2012, 85, 061803.
Sarabadani, J.; Ikonen, T.; Ala-Nissila, T. Iso-flux tension propagation theory of driven polymer translocation:
The role of initial configurations. J. Chem. Phys. 2014, 141, 214907.
Sung, W.; Park, P.J. Polymer Translocation through a Pore in a Membrane. Phys. Rev. Lett. 1996, 77, 783.
Muthukumar, M. Polymer translocation through a hole. J. Chem. Phys. 1999, 111, 10371.
Brochard-Wyart, F. Polymer Chains Under Strong Flows: Stems and Flowers. Europhys. Lett. 1995, 30, 387.
Sakaue, T.; Saito, T.; Wada, H. Dragging a polymer in a viscous fluid: Steady state and transient. Phys. Rev. E
2012, 86, 011804.
Grosberg, A.Y.; Nechaev, S.; Tamm, M.; Vasiyev, O. How long does it take to pull an ideal polymer into a
small hole? Phys. Rev. Lett. 2006, 96, 228105.
Saito, T.; Sakaue, T. Cis-trans dynamical asymmetry in driven polymer translocation. Phys. Rev. E 2013,
88, 042606.
Adhikari, R.; Bhattacharya, A. Driven translocation of a semi-flexible chain through a nanopore: A Brownian
dynamics simulation study in two dimensions. J. Chem. Phys. 2013, 138, 204909.
Chuang, J.; Kantor, Y.; Kardar, M. Anomalous dynamics of translocation. Phys. Rev. E 2001, 65, 011802.
De Haan, H.W.; Slater, G.W. Memory effects during the unbiased translocation of a polymer through a
nanopore. J. Chem. Phys. 2012, 136, 154903.
De Haan, H.W.; Slater, G.W. Using an incremental mean first passage approach to explore the viscosity
dependent dynamics of the unbiased translocation of a polymer through a nanopore. J. Chem. Phys. 2012,
136, 204902.
Luo, K.; Ollila, S.T.T.; Huopaniemi, I.; Ala-Nissila, T.; Pomorski, P.; Karttunen, M.; Ying, S.-C.; Bhattacharya, A.
Dynamical scaling exponents for polymer translocation through a nanopore. Phys. Rev. E 2008, 78, 050901.
Vocks, H.; Panja, D.; Barkema, G.T.; Ball, R.C. Pore-blockade times for field-driven polymer translocation.
J. Phys. Condens. Matter 2008, 20, 095224.
Saito, T.; Sakaue, T. Driven anomalous diffusion: An example from polymer stretching. Phys. Rev. E 2015,
92, 012601.
Polson, M.J.; Dunn, T.R. Evaluating the applicability of the Fokker-Planck equation in polymer translocation:
A Brownian dynamics study. J. Chem. Phys. 2014, 140, 184904.
c 2016 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access

article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

146

polymers
Article

Investigating the Synergistic Effects of Combined
Modified Alginates on Macrophage Phenotype
Hannah C. Bygd 1 and Kaitlin M. Bratlie 1,2,3, *
1
2
3

*

Department of Materials Science & Engineering, Iowa State University, Ames, IA 50011, USA;
hcbygd@iastate.edu
Department of Chemical & Biological Engineering, Iowa State University, Ames, IA 50011, USA
Division of Materials Science & Engineering, Ames National Laboratory, Ames, IA 50011, USA
Correspondence: kbratlie@iastate.edu; Tel.: +1-515-294-7304; Fax: +1-515-294-5444

Academic Editor: Alexander Böker
Received: 30 September 2016; Accepted: 1 December 2016; Published: 6 December 2016

Abstract: Understanding macrophage responses to biomaterials is crucial to the success of implanted
medical devices, tissue engineering scaffolds, and drug delivery vehicles. Cellular responses to
materials may depend synergistically on multiple surface chemistries, due to the polyvalent nature
of cell–ligand interactions. Previous work in our lab found that different surface functionalities of
chemically modified alginate could sway macrophage phenotype toward either the pro-inflammatory
or pro-angiogenic phenotype. Using these findings, this research aims to understand the relationship
between combined material surface chemistries and macrophage phenotype. Tumor necrosis factor-α
(TNF-α) secretion, nitrite production, and arginase activity were measured and used to determine the
ability of the materials to alter macrophage phenotype. Cooperative relationships between pairwise
modifications of alginate were determined by calculating synergy values for the aforementioned
molecules. Several materials appeared to improve M1 to M2 macrophage reprogramming capabilities,
giving valuable insight into the complexity of surface chemistries needed for optimal incorporation
and survival of implanted biomaterials.
Keywords: alginate; macrophage phenotype; TNF-α; synergy

1. Introduction
Macrophages are functionally diverse cells with a multitude of roles in immunity [1], disease [2],
and wound healing [1,3–5], making them an appealing target in research areas like drug delivery [6,7],
regenerative medicine [5], and biomaterial implantation [8]. In particular, they are crucial in the initiation,
propagation, and resolution stages of the foreign body response (FBR) to implanted biomaterials [9–11].
Macrophages are important in all stages of this response due, in part, to their plasticity and heterogeneous
phenotypes. The phenotypes in which macrophages exist are best described as a complex scale,
bookended by classical and alternative activations.
Classically activated, M1, macrophages are said to be pro-inflammatory and cytotoxic [12,13].
As cells that mediate immune responses to bacterial, viral, and fungal infections, they can be activated
by microbial stimuli (lipopolysaccharides (LPS)) or interferon (IFN)-γ, which is released by activated
lymphocytes [14–16]. The inflammatory response that these stimulating factors initiate is followed by
the release of important cytotoxic molecules such as reactive nitrogen intermediates (RNIs) and tumor
necrosis factor (TNF)-α that can have an impact on the presence of tumor cells [8,13,16]. The functions
of M1 macrophages also make them important in the inflammation stages of wound healing and
the FBR, where they produce pro-inflammatory cytokines, phagocytose microorganisms near the
injury, and recruit additional inflammatory cells to the site [8,10,17–19]. Alternatively activated, M2,
macrophages are known to be pro-angiogenic and can be activated by interleukin (IL)-4 or IL-13 [20].
Polymers 2016, 8, 422

147

www.mdpi.com/journal/polymers

Polymers 2016, 8, 422

The main functions of these cells include: maintaining homeostasis, tissue repair and remodeling,
as well as promoting wound healing in the resolution stage of the FBR [8]. They fulfill these roles
with the release of IL-10, vascular endothelial growth factor (VEGF), and transforming growth factor
(TGF)-β [21]. The entire spectrum of phenotypes that exists is imperative in one way or another [4,8],
but an increased presence of either M1 or M2 macrophages may be useful for specific applications.
For example, an earlier M2 macrophage presence rather than prolonged M1 presence may allow for
proper wound closure, but in cancer therapies an M1 macrophage presence rather than pro-angiogenic,
tumor associated M2 like macrophage, may lead to the elimination of cancerous cells.
With the application of wound healing and successful implant incorporation, the significance
of macrophage plasticity is made evident by examining the timeline of macrophage presence during
the FBR [8]. Macrophage phenotype is dynamic throughout the entirety of the FBR, and a balance of
phenotypes is essential for timely progression from injury to proper healing [8]. After implantation of
a biomaterial, macrophages are classically activated by proteins that have adsorbed to the material
surface [22–24]. These macrophages release pro-inflammatory cytokines that continue to attract
monocytes to the injury site leading to chronic inflammation [19]. Attempting and failing phagocytosis
of the large implant, causes the M1 macrophages to fuse into foreign body giant cells (FBGC) [25].
Chronic inflammation is only resolved with a presence of alternatively activated macrophages resulting
from phagocytosis of dying cells around the implant, or stimulation by IL-4 or IL-13, typically from
basophils [20,26,27]. M2 macrophages seek to promote wound healing in the area of the implant [11].
However, prolonged M2 macrophage activity can lead to the formation of an extensive fibrous capsule
that may negatively impact the function of the implant [28]. Finding a balance between chronic
inflammation and continued wound healing remains an issue with many biomedical implants. Using
the idea that material surface chemistries have an impact on macrophage phenotype, it may be
hypothesized that a particular material could orchestrate appropriate progression through the FBR
to avoid both chronic wound healing and complete fibrous encapsulation. This may be particularly
applicable for electrospinning coaxial fibers [29] and sensors [30], in addition to scaffolds for tissue
engineering applications [31].
Many techniques for macrophage reprogramming have been developed that focus on either the
chemical or environmental stimuli to which macrophages are exposed [14,15,32,33]. Previous work in
this lab has indicated that some chemically modified alginates may impact macrophage phenotype [9].
For example, some modifications decreased TNF-α production by M1 macrophages, which could make
them useful for decreasing the inflammatory response to a material. Other modifications appeared to
amplify the presence of M2 macrophages by showing an increased arginase/iNOS (inducible nitric
oxide synthase) ratio compared to the control. Some of the most promising materials showed decreased
TNF-α production and increased arginase/iNOS ratios for all activations of macrophages. Finally, some
indicated that even a small amount of modification may be giving the materials some permselectivity
capabilities, which could be important for applications in drug delivery and tissue engineering.
Given the complexity of biological systems and the immune system in particular, the idea that the
most effective material surface might be made up of several chemistries is reasonable [34]. However,
this can make optimal biomaterial design criteria difficult to ascertain [34]. Cellular responses to
materials are typically directed by surface protein adsorption [22], and many biological interactions,
such as protein–ligand or protein–cell interactions, are polyvalent in nature [22,35]. This suggests
a need for a material with multiple surface chemistries to more closely imitate natural interactions,
as multiple simultaneous interactions may have unique collective properties that vary from those
displayed by each component individually [35]. The synergistic effects of material properties have
typically been studied using high-throughput combinatorial approaches [36]. This logic and testing
method have similarly been employed in synergistic drug delivery, where the aim is to achieve
some synergistic therapeutic effect, reduction of dose or toxicity, and to delay or minimize drug
resistance [37–40].

148

Polymers 2016, 8, 422

The focus of this work was to examine the synergistic effects of a few promising material surface
chemistries described above. In order to more accurately describe the phenotype of macrophages
in vitro, a newer nomenclature will be used throughout this paper that is based on the molecule
used to activate the cells, for example M(LPS) and M(IL-4) [12]. By modifying alginate with one
functional group that amplified M(IL-4) activation (ester or oxime) and one that indicated M(LPS) to
M(IL-4) macrophage reprogramming (amide or nitro), the goal was to achieve an even greater M(IL-4)
response from all macrophage phenotypes [9]. This could be useful in the proper integration of various
biomaterials. Other materials examined here combine a surface modifier that had high permeability
but low production of toxic cytokines (sulfonic acid or nitro) with one that had low permeability but
high cytokine production (ketal or epoxide) [9]. An ideal combination of modifications would have
low permeability and also cause low production of toxic cytokines. Understanding the synergistic
effects of these materials may lead to the development of a diverse library of materials that could
influence any desired macrophage phenotype.
2. Materials and Methods
Experiments were performed with a minimum of three replicates. Results were compared to
controls of unmodified alginate as well as singly modified samples in previous work [9]. All materials
were purchased from Sigma (St. Louis, MO, USA) and used as received, unless otherwise indicated.
Fresh deionized (DI) water (Milli-Q Nanopure, Thermo Scientific, Waltham, MA, USA) was used
throughout this study.
2.1. Materials
Surface modifiers [9], coupled to medium viscosity (~600–900 cps, ~250,000 MW ) alginate
(MP Biomedicals, Santa Ana, CA, USA) were chosen based on the results of previous research in
this lab [9]. This included: glycidamide; tert-butyl 4 aminobutanoate (VWR, Radnor, PA, USA);
malonamide (Fisher, Pittsburgh, PA, USA); 1-amino-4-oxocyclohexane carboxylic acid ethylene ketal;
2,4-dinitro-phenyl-hydoxylamine; 3-aminobenzamide oxime; and 3-amino-1-propane sulfonic acid
(Fisher, Pittsburgh, PA, USA).
2.2. Alginate Modification
The seven different surface modifiers were combined and coupled to the medium
viscosity alginate using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC, Oakwood Chemical,
West Colombia, SC, USA) and N-hydroxysuccinimide (NHS, Thermo Scientific, Waltham, MA, USA).
These materials were modified as previously described [9] with 50 molar equivalents of modifier 1 and
50 molar equivalents of modifier 2, rather than 100 molar equivalents of one modifier.
2.3. Elemental Analysis
In order to determine the percent modification of each material, elemental analysis was performed
to obtain %C, %H and %N. Measurements were recorded in triplicate with an acetanilide calibration
standard, combustion and reduction temperatures of 925 and 640 ◦ C, respectively, and a resulting
accuracy of ±0.3% for each element. All standards and reagents are from Perkin Elmer and/or
Elementar America’s Inc. (Mt. Laurel, NJ, USA). The instrument used was a PE 2100 Series II
combustion analyzer (Perkin Elmer Inc., Waltham, MA, USA).
2.4. Water Contact Angle (WCA)
Samples were prepared by creating a positively charged surface on glass microscope slides with
poly-L-lysine (PLL). This allowed for a thin, even coating of each modified alginate to be applied to
slides by crosslinking with SrCl2 (Alfa Aesar, Haverhill, MA, USA). These coated slides could then
be inverted over a container of water. An air bubble (100 μL) was deposited under the slide and

149

Polymers 2016, 8, 422

imaged using a digital camera (Canon EOS Rebel T3i, Canon, Melville, NY, USA). Five replicates were
collected for each sample before the angle between the slide and the bubble was measured using ImageJ
(NIH, Bethesda, MD, USA) software. The final reported WCA is 180◦ minus the measured angle.
2.5. Compression Modulus
The compression modulus was measured for each sample using manual compression testing
techniques. Modified alginate pegs were made to be approximately 10 × 10 × 4 mm3 in size. These pegs
were placed between two microscope slides and imaged after each of various sized weights were set
on the top slide. The changing distance between the slides in each image was measured using ImageJ,
and used to generate stress-strain curves. Linear portions of these curves were used to determine the
reported compression modulus for each sample (n = 5).
2.6. Cell Culture
RAW 264.7 cells (American Type Cell Collection, ATCC, Manassas, VA, USA) were used
as a model cell line for macrophages in these experiments. Cells were cultured in complete
medium (CM, Dulbecco’s modified Eagle’s medium (DMEM, Mediatech, Inc., Manassas, VA, USA))
supplemented with 10% fetal bovine serum (FBS, Mediatech, Inc., Manassas, VA, USA), 100 U/mL
penicillin, and 100 μg/mL streptomycin) at 37 ◦ C in 5% CO2 . Every three to five days, the cells were
passaged using a cell scraper to detach cells and subcultured between ~6.7 × 103 and 2.7 × 104 cells/cm2 .
2.7. Cell Viability
In order to test the viability of RAW 264.7 cells in contact with the combined modified alginates,
plates were seeded for MTT (methyl thiazolyldiphenyl tetrazolium, Research Products International
Corp., Mt Prospect, IL, USA) assays during passaging. For this, 24-well plates (KSE Scientific, Durham,
NC, USA) were first coated with 0.05% PLL solutions using 200 μL/well and then incubated at 37 ◦ C
for 1 h. Before seeding the cells into the plate, each well was washed twice with sterile phosphate
buffered saline (PBS). The plates were seeded with 125,000 cells/cm2 in 500 μL of CM per well and
25 ng/mL IL-4 (M(IL-4)) (eBioscience Inc., San Diego, CA, USA) or 5 ug/mL LPS (M(LPS)) and the
cells were allowed to adhere for 24 h. A control set of experiments using non-activated cells was
included and referred to as naïve or M(0) cells. Alginate coatings were created by adding ~100 μL of
modified alginate to each well, allowing it to coat the bottom, and excess alginate was removed. Next,
500 μL of 0.2 M SrCl2 were used per well to crosslink the coating. This solution was left in the wells
for approximately 5 min before replacing it with clear CM. After 48 h the supernatant was removed
and saved for further testing. To each well 500 μL of clear CM was added along with 50 μL of MTT
(5 mg/mL in DI water). After a 2 h incubation at 37 ◦ C, 425 μL of the solution was removed from each
well and replaced with 500 μL of dimethyl sulfoxide (DMSO). The absorbance of each plate was read
at 540 nm with a reference of 690 nm using a BioTek Synergy HT Multidetection Microplate Reader
(BioTek, Winooski, VT, USA). Positive controls for each plate were cells with no alginate samples, and
negative controls contained media, modified alginates, and activators. Experiments were performed
in quadruplicate and results are given as the mean value for each sample normalized to the positive
control ± standard deviation.
2.8. TNF-α ELISA
Measurement of TNF-α, an M1 phenotype marker, was performed using commercially available
enzyme linked immunosorbent assay (ELISA) kits (eBioscience, Inc., San Diego, CA, USA) and
performed as described by the manufacturer, using the supernatant collected during viability assays.
2.9. Urea Assay
After 48 h of incubation with the combined modified alginates, the supernatant was removed
and saved for further testing. Cells were then washed with 400 μL of PBS, and the plates were
150

Polymers 2016, 8, 422

placed on ice for 10 min with 100 μL of cell lysis buffer (150 μL protease inhibitor cocktail (Amresco,
Solon, OH, USA) and 15 μL Triton X-100 (Acros Technologies, Elgin, IL, USA) diluted to 15 mL with
DI water) per well. The resulting lysate (25 μL) was transferred from each well to a 96-well plate
(Argos Technologies, Elgin, IL, USA) along with 25 μL of a 10 mM MnCl2 (Fisher, Pittsburgh, PA,
USA) and 50 mM Tris solution (Fisher, Pittsburgh, PA, USA). The plate was incubated for 10 min at
55 ◦ C before adding 50 μL of 1 M arginine (pH 9.7) to each well and incubating at 37 ◦ C for 20 h.
Arginase activity, an M2 phenotype indicator, was measured through the conversion of arginine to
urea. This was done by adding 200 μL of a 1:2 ratio of solution 1 (1.2 g o-phthaldialdehyde (Alfa Aesar,
Haverhill, MA, USA), 1 L H2 O, and 500 μL HCl (Fisher, Pittsburgh, PA, USA)) and solution 2 (0.6 g
N-(naphthyl)ethylenediamine dihydrochloride (Acros Technologies, Elgin, IL, USA), 5 g boric acid
(Fisher, Pittsburgh, PA, USA), 800 mL H2 O, 111 mL sulfuric acid (Fisher, Pittsburgh, PA, USA), diluted
to 1 L with H2 O) [41]. The plate was read at 520 nm with a reference at 630 nm.
2.10. Griess Reagent Assay
Nitrite production, which is also indicative of an M1 phenotype, was measured from the
supernatant collected in the urea assay described above. A standard curve was created using serial
dilutions of 100 μM NaNO2 with volumes of 150 μL. To the remaining wells 150 μL of sample were
added. To the entire plate, 130 μL of DI water and 20 μL of Griess reagent (Acros Technologies,
Elgin, IL, USA) were added and allowed to incubated for 20 min. The plate was read at 448 nm with
a 690 nm reference.
2.11. Immunocytochemistry
RAW 264.7 cells were fluorescently labeled using a previously developed protocol [9]. Briefly,
cells were seeded at 100,000 cells/cm2 on clean, PLL coated glass coverslips in Petri dishes, as M(LPS),
M(IL4), or M(0) and were coated with pairwise modified alginates as described above. After 48 h
of incubation, the cells were stained for CD11c and CD206. These coverslips were imaged with
an EVOS® FLoid® Imaging Station (Life Technologies, Grand Island, NY, USA) using the red channel
(excitation/emission 586/646 nm), blue channel (390/446 nm) and the green channel (482/532 nm).
2.12. TNF-α Diffusion
To test the diffusive properties of these pairwise modified alginates, microparticles were made
using an electrostatic droplet generation technique. Using the same equipment and techniques
previously described [9], microparticles with an average diameter of 685.15 ± 14.20 μm were made
with unmodified alginate. Amino polystyrene particles, 10 mg, (Spherotech Inc., Lake Forest, IL,
USA) were also labeled with 2 mg mouse IgG (Thermo Scientific, Waltham, MA, USA) in PBS using
20 mg of EDC. These labeled polystyrene particles were encapsulated in the unmodified alginate
particles during electrostatic droplet generation using procedures developed by Kulseng et al. [42].
Approximately 200 particles were incubated with 2 mL of supernatant known to contain TNF-α.
This supernatant was collected from M(LPS) RAW 264.7 cells cultured at ~182,000 cells/cm2 after two
days of incubation. After four days of incubation with the particles, the supernatant was removed
and compared to the original TNF-α supernatant using the TNF-α ELISA described in Section 2.8 to
determine how much had diffused into the particles.
2.13. Statistics and Data Analysis
Statistical analysis was performed using JMP® statistical software (Cary, NC, USA). Statistical
significance of the mean comparisons was determined by a two-way ANOVA. Pair-wise comparisons
were analyzed with Tukey’s honest significant difference test. Differences were considered statistically
significant for p < 0.05.

151

Polymers 2016, 8, 422

3. Results
3.1. Modification and Characterization
After the combined modification of alginate using the molecules in Figure 1, the synthesized
materials were characterized for their hydrophobicity, compression modulus and percent modification.
Elemental analysis confirmed modification with %N values being greater for the modified alginates
than unmodified (Figure 2A,B). The WCAs of the combined modified alginates fell in a range
of 39.9◦ –51.2◦ (Figure 2C), compared to the range of 49.7◦ –61.0◦ with the single modifications,
and 54.1◦ ± 2.0◦ for unmodified alginate(Figure 2D). This suggests the materials remained relatively
hydrophilic. The compression moduli of the combined modified alginates ranged from 12.5–30.1 kPa
(Figure 2E), which was similar to the range in compression modulus for the single modifications
(18.8–25.0 kPa), and of unmodified alginate (25.3 ± 5.4 kPa, Figure 2F).

Figure 1. Chemical structures of the molecules used in the modification of medium viscosity alginate.
The functional groups listed here are used as labels in the following figures for convenience.

Figure 2. Cont.

152

Polymers 2016, 8, 422

Figure 2. Characterization of combined and single modified alginates: (A,B) percent modifications;
(C,D) WCAs; and (E,F) compression moduli were measured for double (A,C,E) and single (B,D,F)
modifications. n = 5. Data represent the mean value ± standard deviation. Bars with the same letter
(a–f) are not statistically different (p < 0.05).

3.2. Cell Viability
Biomaterials must be cytocompatible for use as implantable materials. Cytotoxicity of the combined
modified alginates was measured using MTT assays. The viability of the cells was expressed as a
percentage of the positive control of cells cultured on PLL coated tissue culture plastic. Cells exposed to
all materials showed greater than 70% viability (Figure 3). This suggests minimal cytotoxicity, especially
for cells cultured under a hydrogel.

Figure 3. Modified alginates are cytocompaible. Cell viability of M(LPS), M(IL-4) and M(0) RAW 264.7
macrophages encapsulated under modified and unmodified alginate layers. n = 4. Data represent the
mean value ± standard deviation.

153

Polymers 2016, 8, 422

3.3. Arginase/iNOS
Arginase activity was determined for encapsulated macrophages by exposing cell lysate to arginine
and quantifying the urea produced. Nitrites are the stable form of NO and were quantified through
a Griess reagent assay. Absolute values for both urea and Griess assays are given in Figure 4A–D,
with the pairwise modifications on the left and the single modifications on the right for comparison.
Results from these assays are given as urea: nitrite in Figure 4E,F, which corresponds to the amount of
urea produced divided by the amount of nitrite in corresponding samples. Ratio values range from
7.12 mg/μmol with sulfonic acid/ketal to 36.2 mg/μmol with ester/nitro for M(IL-4) macrophages.
A value of 6.88 ± 3.05 mg/μmol was previously determined for M(IL-4) macrophages in the presence
of unmodified alginate [9]. M(LPS) macrophages exhibited a narrow range of 6.20 to 14.1 mg/μmol
compared to 7.52 ± 1.76 mg/μmol for unmodified alginate. Finally, values for M(0) macrophages
ranged from 4.24 mg/μmol with the oxime/nitro modification to 33.8 mg/μmol with ester/nitro. M(0)
macrophages exhibited a urea:nitrite ratio of 6.91 ± 2.13 mg/μmol with unmodified alginate. Higher
values for these ratios indicate a stronger M2 phenotype presence, while low values suggest a M1
phenotype. Most values seem to indicate a stronger M2 presence when compared to unmodified
alginate. Exceptions to this statement include oxime/amide and oxime/nitro M(LPS) macrophages.
The controls for M(IL-4), M(LPS), and M(0) cells were higher than the cells exposed to the modified
and unmodified alginates with 124 ± 30.6, 13.1 ± 2.38 and 94.3 ± 29.5 mg/μmol being measured,
respectively. The urea measured for the controls was in line with the values measured for the cells
exposed to alginate samples—20.1 ± 3.25, 16.4 ± 2.97 and 6.31 ± 1.14 mg/dL for M(IL-4), M(LPS),
and M(0) cells, respectively. However, the nitrite levels were much lower for M(IL-4) and M(0) cells,
leading to a large increase in the urea:nitrite ratio. The measured nitrite levels were 1.62 ± 0.30,
12.5 ± 0.25 and 0.67 ± 0.17 μM for M(IL-4), M(LPS), and M(0) cells respectively.

Figure 4. Cont.

154

Polymers 2016, 8, 422

Figure 4. Molecules produced by polarized RAW 264.7 macrophages encapsulated under modified
alginates. Alginates were layered over activated macrophages and: (A,B) nitrite; (C,D) urea;
(E,F) urea:nitrite; and (G,H) TNF-α were measured for dual (A,C,E,G) and single (B,D,F,H)
modifications. Arginase activity was measured by removing the alginate layer and lysing the cells.
n = 4. Data represent the mean value ± standard deviation. Bars with the same letter (a–d) are not
statistically different (p < 0.05) from data points of the same activation.

3.4. TNF-α ELISA
Production of the cytotoxic cytokine TNF-α by cells exposed to modified alginates was measured
and reported in Figure 4G. Values for single modifications of alginate are included in Figure 4H for
comparison. Most modifications of alginate were able to decrease the amount of TNF-α produced in
comparison to unmodified alginate for all macrophage phenotypes. The exception to this statement
is oxime/amide for M(0) cells. TNF-α production by M(IL-4) macrophages ranged from 1.2 to
1.6 ng/mL with 2.8 ± 0.68 ng/mL produced under unmodified alginate. M(0) macrophages produced
a range of TNF-α that spanned 1.8 to 3.5 ng/mL with unmodified alginate eliciting 1.9 ± 0.17 ng/mL
TNF-α. TNF-α produced by M(LPS) cells spanned a higher range of 2.9 to 5.7 ng/mL, and on
a sample-to-sample basis, the amount of cytokine produced was most often highest for M(LPS)
macrophages. In the presence of unmodified alginate, M(LPS) cells produced 5.0 ± 0.53 ng/mL of
TNF-α. All modifications had statistically lower TNF-α production than unmodified alginate for
M(IL-4) macrophages. Similar to nitrite, TNF-α secretion from M(IL-4) and M(0) cells was much
lower for the control than for the cells in the presence of alginate. The values for the controls were
0.58 ± 0.066, 5.0 ± 0.35 and 0.39 ± 0.016 ng/mL for M(IL-4), M(LPS) and M(0) cells, respectively.
3.5. Fluorescent Imaging
Immunocytochemistry (ICC) fluorescent staining was used to identify the macrophage phenotypes
resulting from culturing cells in the presence of modified alginates. CD11c, an LPS receptor, was used

155

Polymers 2016, 8, 422

to identify M1 macrophages and a CD206, a mannose receptor, was used to indicate an M2 macrophage
phenotype. DAPI was also used as a nuclei stain. Representative images are shown in Figure 5A,B.
These images were used to qualitatively support the results found by measuring arginase activity,
TNF-α secretion, and nitrite production. Control staining images of macrophages with activator alone
are also included in Figure 5C.

Figure 5. Immunocytochemistry staining of macrophages suggests that modified alginates alter
phenotype: (A) nitro/epoxide; and (B) ester/amide modified alginates were coated on cells;
and (C) cells in the absence of alginate are shown for comparison. Cells were stained with fluorescently
labeled CD206 (red, M2 marker) and CD11c (green, M1 marker) markers as well as DAPI (blue).
Scale bar is 50 μm.

3.6. Synergy
Normalizing the percent modifications of each combined alginate sample to that of the single
modifications, we were able to determine weight fractions of each modification. Synergy was calculated
using the equation that was developed by Karande et al. [43]:
S=

mA+B
xmA + (1 − x )mB

(1)

where mA , mB , and mA+B are the measured values for the different modifications and x is the weight
fraction of each modification determined by elemental analysis. These values are shown for each
macrophage phenotype in Figure 6A–D. The synergy values for nitrite were nearly one for M(LPS)
cells exposed to all modifications (0.930–1.147). Synergy values were slightly more dynamic for M(IL-4)
cells, ranging from 0.578 to 1.121, and for M(0) cells, ranging from 0.218 to 1.34. For urea, all of the
synergy values for all of the macrophage activation states were near unity. This resulted in nitrite
being entirely responsible for the synergy values for urea: nitrite. Since M(LPS) cells did not have large
156

Polymers 2016, 8, 422

variation in the synergy values for the amount of nitrite measured in the supernatant, the synergy
values were correspondingly nearly unity (0.854–1.159). Both M(IL-4) and M(0) cells yielded more
dynamic synergy values for urea: nitrite of 0.811 to 2.613 for M(IL-4) and 0.533 to 4.31 for M(0). Finally,
the synergy values for TNF-α production were in the range of 0.783–1.778 for M(IL-4) macrophages,
0.511–1.453 for M(LPS) macrophages, and 0.303–1.246 for M(0) macrophages.

Figure 6. Synergistic indices for molecules produced by polarized RAW 264.7 macrophages: (A) nitrite;
(B) urea; (C) urea:nitrite; and (D) TNF-α for M(LPS), M(IL-4), and M(0) cells were calculated based on
previously obtained values for single modifications of alginates [9]. n = 4. Data represent the mean
value ± standard deviation. Bars with the same letter (a–d) are not statistically different (p < 0.05) from
data points of the same activation. All measured values in panel B were statistically similar.

3.7. TNF-α Diffusion
The amount of TNF-α able to diffuse into the particles was measured using an ELISA after
four days of incubation (Figure 7A) [9]. A wide range of values were calculated for percent loading.
Nitro/ketal did not allow any diffusion of TNF-α, while nitro/epoxide allowed 87% loading. Percent
loading values for single modifications of alginate are included in Figure 7B for comparison. Synergy
values for TNF-α diffusion were calculated in the range of 0 to 4.94 (Figure 7C).

157

Polymers 2016, 8, 422

Figure 7. Modified alginates alter mass transport of TNF-α through layered alginate particles.
(A) Combined; and (B) single modified alginates were layered over alginate particles containing mouse
IgG labeled polystyrene particles. n = 4. Data represent the mean value ± standard deviation. Bars
with the same letter (a–d) are not statistically different (p < 0.05) from data points of the same activation.

4. Discussion
Alginate is an appealing biomaterial for numerous wound healing and tissue engineering
applications. It is a naturally occurring, relatively inexpensive polymer, it has low toxicity and good
biocompatibility, and is easily gelled under mild conditions with divalent cations [44]. The structure of
alginate is similar to that of the extracellular matrix in living tissues, and it provides a relatively inert
and moist microenvironment [44,45] However, the biocompatibility of alginate remains insufficiently
understood. For example, in research by Elliot et al., alginate was used to encapsulate porcine islets in
a clinical trial for diabetes treatment [46]. While the trials were mildly successful in that there were
surviving functional islets, the overall function of the implant was severely diminished by the presence
of granulation tissue that developed in response to the alginate [46]. Many aspects of the polymer
have been studied in an attempt to improve upon the biocompatibility [47–60], and some work has
been done to chemically modify alginate and study the impact this has on cell function [9,61–63].
This research focuses on combining surface modifications to study the synergistic effects of chemical
modification on cellular responses. By using materials that have been studied in the past, we have
been able to compare predicted and actual values to examine synergy of the modifications.
In this study, we aimed to find combined modifications that would improve the biocompatibility
of alginate by reprogramming macrophages towards an M2-like phenotype. Based on the equation
used to calculate synergy, values greater than one for urea:nitrite and less than one for both TNF-α and
nitrite production would be desired. These values are most important for M(LPS) macrophages as this
is the phenotype we are aiming to reprogram. M(LPS) macrophages tend to mediate the inflammatory
stages of the foreign body response to biomaterials, which we are attempting to minimize. TNF-α
production synergy values tended to be less than or equal to one for all phenotypes studied. This may

158

Polymers 2016, 8, 422

be due, in part, to the fact the TNF-α, must diffuse through the alginate coating in order to be measured.
As Figure 7 indicates, some of these combined modifications allow little to no TNF-α diffusion through
the alginate. However, combining this information with the urea synergy values, we can determine
which modifications may be favorably shifting macrophage phenotype towards an M2-like state.
From the values presented in Figure 6, it can be determined that almost all the modifications do
not result in cooperative interactions with M(LPS) macrophages. This is particularly interesting and
somewhat surprising for TNF-α production since variable diffusion of TNF-α through the different
modifications was observed (Figure 7).
Several different observations were noticed when examining synergistic values of TNF-α
production in concert with urea:nitrite. With respect to the M(0) macrophage synergy values, sulfonic
acid/epoxide and nitro/epoxide resulted in antagonistic synergy values for TNF-α production.
Perhaps unsurprisingly, the pairwise modifications with epoxide resulted in increases in TNF-α
production synergy values for M(IL-4) cells. Epoxide was chosen for this study because it resulted in
low permeability and high TNF-α production. Sulfonic acid combinations are also able to increase
TNF-α production synergy values for M(IL-4) cells and increased urea:nitrite values. Sulfonic
acid/epoxide was also able to improve urea: nitrite values, indicating the combinations of these
modifications shifted M(0) cells towards an M2 phenotype. Ester combinations resulted in improved
urea:nitrite values and did not have an observable synergistic effect on TNF-α for M(0) cells. Future
work in examining the synergy of these materials in a dose dependent context may elucidate the
relationship of these modifications on shifting macrophage phenotypes.
By including the TNF-α diffusion study, this work also examined the changes in diffusive
properties of alginate due to combined surface modifications. Ideally, combined modifications would
result in a lower mass transport of TNF-α than in the case of single modifications. This would
be indicated by synergy values less than one. Modifications sulfonic acid/ketal, nitro/ketal,
and oxime/nitro all had values below one. Ketal lowering diffusion was expected since it was
previously observed to have low permeability for TNF-α [9]. Interestingly, when ketal was combined
with nitro or sulfonic acid, the synergy values were less than one even though nitro and sulfonic acid
were selected for their poor permselectivity towards TNF-α and low production of TNF-α. The other
modifications had relatively high synergy values, suggesting that they are not permselective towards
TNF-α, which could potentially be harmful for islet encapsulation as in Elliot et al. [46]. In that example,
tissue development surrounding the implant cut off blood supply to the encapsulated islet preventing
them from functioning properly. In this case, with increased TNF-α release, the toxic cytokine would
be able to reach the islets and impact their ability to regulate blood sugar.
5. Conclusions
Pair-wise modifications of alginate resulted in synergistic and antagonistic responses for M(IL-4),
M(LPS) and M(0) cells for both nitrite and TNF-α secretion. There was no observed advantage or
disadvantage of these modifications in terms of arginase activity. In combining the data measuring the
shifts in macrophage phenotype with that examining TNF-α diffusion through the modified alginates,
improved materials for artificial organs can be obtained. Interestingly, certain modifications appear
to influence the synergistic index for TNF-α diffusion, such as ketal, which lowered the index for
both pair-wise combinations in which it was included. This observation held for material mediated
shifts in macrophage phenotype, e.g., ester for TNF-α secretion from M(0) cells. These results further
demonstrate the importance of chemical moieties on biomaterials in achieving appropriate cellular
responses. Although this study is not conclusive, it provides valuable insight into the impact of
multiple surface modifications on macrophage phenotype. This library is also not large enough to
offer much in the way of justified reasoning for the synergistic effects of the combined modifications.
However, it provided evidence that maximum macrophage reprogramming may require a multitude of
stimuli, and further supported the findings that simply altering the surface chemistry of a biomaterial
may have a significant impact on its biocompatibility.

159

Polymers 2016, 8, 422

Acknowledgments: This work was supported by the National Science Foundation under Grant No. CBET-1227867
and the Roy J. Carver Charitable Trust Grant No. 13-4265. The authors also acknowledge support from NSF
ARI-R2 (CMMI-0963224) for funding the renovation of the research laboratories used for these studies.
Author Contributions: Hannah C. Bygd and Kaitlin M. Bratlie designed the study. Hannah C. Bygd performed
the experiments. Hannah C. Bygd and Kaitlin M. Bratlie analyzed data. Hannah C. Bygd and Kaitlin M. Bratlie
wrote the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.

3.
4.

5.
6.
7.
8.
9.
10.
11.
12.

13.
14.

15.
16.
17.
18.
19.

Mantovani, A.; Sica, A. Macrophages, innate immunity and cancer: Balance, tolerance, and diversity.
Curr. Opin. Immunol. 2010, 22, 231–237. [CrossRef] [PubMed]
Jain, S.; Tran, T.-H.; Amiji, M. Macrophage repolarization with targeted alginate nanoparticles containing
IL-10 plasmid DNA for the treatment of experimental arthritis. Biomaterials 2015, 61, 162–177. [CrossRef]
[PubMed]
Quatromoni, J.G.; Eruslanov, E. Tumor-associated macrophages: Function, phenotype, and link to prognosis
in human lung cancer. Am. J. Transl. Res. 2012, 4, 376–389. [PubMed]
Wynn, T.A.; Barron, L.; Thompson, R.W.; Madala, S.K.; Wilson, M.S.; Cheever, A.W.; Ramalingam, T.
Quantitative assessment of macrophage functions in repair and fibrosis. Curr. Protoc. Immunol. 2011, 2011.
[CrossRef]
Mantovani, A.; Biswas, S.K.; Galdiero, M.R.; Sica, A.; Locati, M. Macrophage plasticity and polarization in
tissue repair and remodelling. J. Pathol. 2013, 229, 176–185. [CrossRef] [PubMed]
Peer, D.; Karp, J.M.; Hong, S.; Farokhzad, O.C.; Margalit, R.; Langer, R. Nanocarriers as an emerging platform
for cancer therapy. Nat. Nanotechnol. 2007, 2, 751–760. [CrossRef] [PubMed]
Kelly, C.; Jefferies, C.; Cryan, S. Targeted liposomal drug delivery to monocytes and macrophages.
J. Drug Deliv. 2011, 2011, 727241. [CrossRef] [PubMed]
Bygd, H.C.; Forsmark, K.D.; Bratlie, K.M. The significance of macrophage phenotype in cancer and
biomaterials. Clin. Transl. Med. 2014, 3. [CrossRef] [PubMed]
Bygd, H.C.; Bratlie, K.M. The effect of chemically modified alginates on macrophage phenotype and
biomolecule transport. J. Biomed. Mater. Res. Part A 2016, 104, 1707–1719. [CrossRef] [PubMed]
Xia, Z.; Triffitt, J.T. A review on macrophage responses to biomaterials. Biomed. Mater. 2006, 1, R1–R9.
[CrossRef] [PubMed]
Ward, W.K. A Review of the Foreign-body Response to Subcutaneously-implanted Devices: The Role of
Macrophages and Cytokines in Biofouling and Fibrosis. J. Diabetes Sci. Technol. 2008, 2, 768–777. [CrossRef]
Murray, P.J.; Allen, J.E.; Biswas, S.K.; Fisher, E.A.; Gilroy, D.W.; Goerdt, S.; Gordon, S.; Hamilton, J.A.;
Ivashkiv, L.B.; Lawrence, T.; et al. Macrophage Activation and polarization: Nomenclature and experimental
guidelines. Immunity 2014, 41, 14–20. [CrossRef] [PubMed]
Martinez, F.O.; Gordon, S. The M1 and M2 paradigm of macrophage activation: Time for reassessment.
F1000Prime Rep. 2014, 6, 13. [CrossRef] [PubMed]
Akilbekova, D.; Philiph, R.; Graham, A.; Bratlie, K.M. Macrophage reprogramming: Influence of latex beads
with various functional groups on macrophage phenotype and phagocytic uptake in vitro. J. Biomed. Mater.
Res. Part A 2015, 103, 262–268. [CrossRef] [PubMed]
Bygd, H.C.; Forsmark, K.D.; Bratlie, K.M. Altering in vivo macrophage responses with modified polymer
properties. Biomaterials 2015, 56, 187–197. [CrossRef] [PubMed]
Sica, A.; Mantovani, A. Macrophage plasticity and polarization: In vivo veritas. J. Clin. Investig. 2012, 122,
787–795. [CrossRef] [PubMed]
Rodriguez, A.; Meyerson, H.; Anderson, J.M. Quantitative in vivo cytokine analysis at synthetic biomaterial
implant sites. J. Biomed. Mater. Res. A 2009, 89, 152–159. [PubMed]
Suwandi, J.S.; Toes, R.E.M.; Nikolic, T.; Roep, B.O. Inducing tissue specific tolerance in autoimmune disease
with tolerogenic dendritic cells. Clin. Exp. Rheumatol. 2015, 33, 97–103.
Mesure, L.; De Visscher, G.; Vranken, I.; Lebacq, A.; Flameng, W. Gene expression study of monocytes/
macrophages during early foreign body reaction and identification of potential precursors of myofibroblasts.
PLoS ONE 2010, 5, e12949. [CrossRef] [PubMed]

160

Polymers 2016, 8, 422

20.
21.
22.

23.
24.
25.

26.
27.

28.
29.

30.
31.
32.
33.
34.
35.
36.

37.
38.
39.
40.

41.
42.

Gordon, S. Alternative activation of macrophages. Nat. Rev. Immunol. 2003, 3, 23–35. [CrossRef] [PubMed]
Gordon, S.; Martinez, F.O. Alternative activation of macrophages: Mechanism and functions. Immunity 2010,
32, 593–604. [CrossRef] [PubMed]
Yang, J.; Mei, Y.; Hook, A.L.; Taylor, M.; Urquhart, A.J.; Bogatyrev, S.R.; Langer, R.; Anderson, D.G.;
Davies, M.C.; Alexander, M.R. Polymer surface functionalities that control human embryoid body cell
adhesion revealed by high throughput surface characterization of combinatorial material microarrays.
Biomaterials 2010, 31, 8827–8838. [CrossRef] [PubMed]
McNally, A.K.; Anderson, J.M. Complement C3 participation in monocyte adhesion to different surfaces.
Proc. Natl. Acad. Sci. USA 1994, 91, 10119–10123. [CrossRef] [PubMed]
Yang, D.; Jones, K.S. Effect of alginate on innate immune activation of macrophages. J. Biomed. Mater. Res. A
2009, 90, 411–418. [CrossRef] [PubMed]
MacLauchlan, S.; Skokos, E.A.; Meznarich, N.; Zhu, D.H.; Raoof, S.; Shipley, J.M.; Senior, R.M.; Bornstein, P.;
Kyriakides, T.R. Macrophage fusion, giant cell formation, and the foreign body response require matrix
metalloproteinase 9. J. Leukoc. Biol. 2009, 85, 617–626. [CrossRef] [PubMed]
Mosser, D.M.; Edwards, J.P. Exploring the full spectrum of macrophage activation. Nat. Rev. Immunol. 2008,
8, 958–969. [CrossRef] [PubMed]
Wang, Y.; Wang, Y.P.; Zheng, G.; Lee, V.W.S.; Ouyang, L.; Chang, D.H.H.; Mahajan, D.; Coombs, J.; Wang, Y.M.;
Alexander, S.I.; et al. Ex vivo programmed macrophages ameliorate experimental chronic inflammatory
renal disease. Kidney Int. 2007, 72, 290–299. [CrossRef] [PubMed]
Anderson, J.M. Biological responses to materials. Annu. Rev. Mater. Res. 2001, 31, 81–110. [CrossRef]
Lu, Y.; Huang, J.; Yu, G.; Cardenas, R.; Wei, S.; Wujcik, E.K.; Guo, Z. Coaxial electrospun fibers: Applications
in drug delivery and tissue engineering. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2016, 8, 654–677.
[CrossRef] [PubMed]
Wujcik, E.K.; Monty, C.N. Nanotechnology for implantable sensors: Carbon nanotubes and graphene in
medicine. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2013, 5, 233–249. [CrossRef] [PubMed]
Panilaitis, B.; Altman, G.H.; Chen, J.; Jin, H.J.; Karageorgiou, V.; Kaplan, D.L. Macrophage responses to silk.
Biomaterials 2003, 24, 3079–3085. [CrossRef]
Wang, D.; Bratlie, K.M. Influence of polymer chemistry on cytokine secretion from polarized macrophages.
ACS Biomater. Sci. Eng. 2015, 1, 166–174. [CrossRef]
Wang, D.; Phan, N.; Isely, C.; Bruene, L.; Bratlie, K.M. Effect of surface modification and macrophage
phenotype on particle internalization. Biomacromolecules 2014, 15, 4102–4110. [CrossRef] [PubMed]
Goldberg, M.; Mahon, K.; Anderson, D. Combinatorial and rational approaches to polymer synthesis for
medicine. Adv. Drug Deliv. Rev. 2008, 60, 971–978. [CrossRef] [PubMed]
Brocchini, S. Combinatorial chemistry and biomedical polymer development. Adv. Drug Deliv. Rev. 2001, 53,
123–130. [CrossRef]
Liu, E.; Treiser, M.D.; Patel, H.; Sung, H.J.; Roskov, K.E.; Kohn, J.; Becker, M.L.; Moghe, P.V. High-content
profiling of cell responsiveness to graded substrates based on combinatorially variant polymers. Comb. Chem.
High Throughput Screen. 2009, 12, 645–655. [CrossRef]
Tallarida, R.J. Drug synergism: Its detection and applications. J. Pharmacol. Exp. Ther. 2001, 298, 865–872.
[PubMed]
Mitragotri, S. Synergistic effect of enhancers for transdermal drug delivery. Pharm. Res. 2000, 17, 1354–1359.
[CrossRef] [PubMed]
Chou, T.C. Drug combination studies and their synergy quantification using the chou-talalay method.
Cancer Res. 2010, 70, 440–446. [CrossRef] [PubMed]
Lehár, J.; Krueger, A.S.; Avery, W.; Heilbut, A.M.; Johansen, L.M.; Price, E.R.; Rickles, R.J.; Short, G.F.;
Staunton, J.E.; Jin, X.; et al. Synergistic drug combinations tend to improve therapeutically relevant selectivity.
Nat. Biotechnol. 2009, 27, 659–666. [CrossRef] [PubMed]
Levinson, S.S. Kinetic centrifugal analyzer and manual determination of serum urea nitrogen, with use of
o-phthaldialdehyde reagent. Clin. Chem. 1978, 24, 2199–2202. [PubMed]
Kulseng, B.; Thu, B.; Espevik, T.; Skjåk-Braek, G. Alginate poly-lysine microcapsules as immune barrier:
Permeability of cytokines and immunoglobins over the capsule membrane. Cell Transplant. 1997, 6, 387–394.
[CrossRef]

161

Polymers 2016, 8, 422

43.
44.
45.
46.

47.

48.
49.
50.
51.

52.

53.
54.

55.

56.
57.
58.
59.

60.
61.
62.
63.

Karande, P.; Jain, A.; Mitragotri, S. Insights into synergistic interactions in binary mixtures of chemical
permeation enhancers for transdermal drug delivery. J. Control. Release 2006, 115, 85–93. [CrossRef] [PubMed]
Lee, K.Y.; Mooney, D.J. Alginate: Properties and biomedical applications. Prog. Polym. Sci. 2012, 37, 106–126.
[CrossRef] [PubMed]
Gombotz, W.R.; Wee, S.F. Protein release from alginate matrices. Adv. Drug Deliv. Rev. 2012, 64, 194–205. [CrossRef]
Elliott, R.B.; Escobar, L.; Tan, P.L.; Muzina, M.; Zwain, S.; Buchanan, C. Live encapsulated porcine islets from
a type 1 diabetic patient 9.5 yr after xenotransplantation. Xenotransplantation 2007, 14, 157–161. [CrossRef]
[PubMed]
De Vos, P.; van Hoogmoed, C.G.; van Zanten, J.; Netter, S.; Strubbe, J.H.; Busscher, H.J. Long-term
biocompatibility, chemistry, and function of microencapsulated pancreatic islets. Biomaterials 2003, 24,
305–312. [CrossRef]
Tam, S.K.; Bilodeau, S.; Dusseault, J.; Langlois, G.; Hallé, J.P.; Yahia, L.H. Biocompatibility and physicochemical
characteristics of alginate-polycation microcapsules. Acta Biomater. 2011, 7, 1683–1692. [CrossRef] [PubMed]
De Vos, P.; De Haan, B.; Van Schilfgaarde, R. Effect of the alginate composition on the biocompatibility of
alginate-polylysine microcapsules. Biomaterials 1997, 18, 273–278. [CrossRef]
Mallett, A.G.; Korbutt, G.S. Alginate modification improves long-term survival and function of transplanted
encapsulated islets. Tissue Eng. Part A 2009, 15, 1301–1309. [CrossRef] [PubMed]
Chen, J.-P.; Chu, I.-M.; Shiao, M.-Y.; Hsu, B.R.-S.; Fu, S.-H. Microencapsulation of islets in PEG-amine modified
alginate-poly(L-Lysine )-alginate microcapsules for constructing bioartificial pancreas. J. Ferment. Bioeng. 1998,
86, 185–190. [CrossRef]
Jork, A.; Thürmer, F.; Cramer, H.; Zimmermann, G.; Gessner, P.; Hämel, K.; Hofmann, G.; Kuttler, B.;
Hahn, H.J.; Josimovic-Alasevic, O.; et al. Biocompatible alginate from freshly collected Laminaria pallida for
implantation. Appl. Microbiol. Biotechnol. 2000, 53, 224–229. [CrossRef] [PubMed]
Petruzzo, P.; Cappai, A.; Ruiu, G.; Dessy, E.; Rescigno, A.; Brotzu, G. Development of biocompatible barium
alginate microcapsules. Transplant. Proc. 1997, 1345, 2129–2130. [CrossRef]
Schneider, S.; Feilen, P.J.; Kraus, O.; Haase, T.; Sagban, T.A.; Lehr, H.A.; Beyer, J.; Pommersheim, R.;
Weber, M.M. Biocompatibility of alginates for grafting: Impact of alginate molecular weight. Artif. Cells
Blood Substit. Biotechnol. 2003, 31, 383–394. [CrossRef]
Zimmermann, U.; Thürmer, F.; Jork, A.; Weber, M.; Mimietz, S.; Hillgärtner, M.; Brunnenmeier, F.;
Zimmermann, H.; Westphal, I.; Fuhr, G.; et al. A novel class of amitogenic alginate microcapsules for
long-term immunoisolated transplantation. Ann. N. Y. Acad. Sci. 2006, 944, 199–215. [CrossRef]
Steele, J.A.M.; Hallé, J.-P.; Poncelet, D.; Neufeld, R.J. Therapeutic cell encapsulation techniques and
applications in diabetes. Adv. Drug Deliv. Rev. 2014, 67–68, 74–83. [CrossRef] [PubMed]
Paredes-Juarez, G.; de Haan, B.; Faas, M.; de Vos, P. A technology platform to test the efficacy of purification
of alginate. Materials 2014, 7, 2087–2103. [CrossRef]
Calafiore, R.; Basta, G. Clinical application of microencapsulated islets: Actual prospectives on progress and
challenges. Adv. Drug Deliv. Rev. 2014, 67–68, 84–92. [CrossRef] [PubMed]
Lee, C.; Shin, J.; Lee, J.S.; Byun, E.; Ryu, J.H.; Um, S.H.; Kim, D.I.; Lee, H.; Cho, S.W. Bioinspired, calcium-free
alginate hydrogels with tunable physical and mechanical properties and improved biocompatibility.
Biomacromolecules 2013, 14, 2004–2013. [CrossRef] [PubMed]
Orive, G.; Hernandez, R.M.; Gascon, A.R. Survival of different cell lines in alginate-agarose microcapsules.
Eur. J. Pharm. Sci. 2003, 18, 23–30. [CrossRef]
Yang, J.-S.; Xie, Y.-J.; He, W. Research progress on chemical modification of alginate: A review.
Carbohydr. Polym. 2011, 84, 33–39. [CrossRef]
Pawar, S.N.; Edgar, K.J. Alginate derivatization: A review of chemistry, properties and applications.
Biomaterials 2012, 33, 3279–3305. [CrossRef] [PubMed]
Vegas, A.J.; Veiseh, O.; Doloff, J.C.; Ma, M.; Tam, H.H.; Bratlie, K.; Li, J.; Bader, A.R.; Langan, E.;
Olejnik, K.; et al. Combinatorial hydrogel library enables identification of materials that mitigate the
foreign body response in primates. Nat. Biotechnol. 2016, 34, 345–352. [CrossRef] [PubMed]
© 2016 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

162

MDPI AG
St. Alban-Anlage 66
4052 Basel, Switzerland
Tel. +41 61 683 77 34
Fax +41 61 302 89 18
http://www.mdpi.com
Polymers Editorial Office
E-mail: polymers@mdpi.com
http://www.mdpi.com/journal/polymers

MDPI AG
St. Alban-Anlage 66
4052 Basel
Switzerland
Tel: +41 61 683 77 34
Fax: +41 61 302 89 18
www.mdpi.com

ISBN 978-3-03842-459-8