Review
Nitroxide-mediated polymerization

https://doi.org/10.1016/j.progpolymsci.2012.06.002Get rights and content

Abstract

Nitroxide-mediated polymerization (NMP) is a controlled/living radical polymerization (CLRP) technique that enables the design of well-defined, functional and complex macromolecular architectures. This comprehensive review covers all aspects, features and achievements of NMP, from its discovery to 2012. All topics related to NMP are thoroughly discussed and detailed in-depth: synthetic approaches to nitroxides and alkoxyamines, kinetic aspects and polymerization features, range of controllable monomers, polymer characterization, polymerization processes (ionic liquids, dispersed media, etc.), macromolecular coupling approaches, functionalization strategies, macromolecular architectures, bio-related and hybrid materials, industrial applications as well as environmental constraints.

Introduction

Synthetic polymers are now an essential part of numerous and varied objects and materials in our everyday life. In the last decades, polymer science has developed into a modern and multidisciplinary research field thanks to fundamental discoveries and achievements. Initially devoted to structural applications, polymers are indeed increasingly involved in higher-value-added functional materials from electronic-, optical- and biomedical-related areas. This explains why polymer science is now considered as an essential and innovative research field from both academia and industry.

Among the crucial contributions witnessed in polymer science is the development of living polymerization techniques, which allows tailor-made macromolecules to be synthesized. From the conceptual point of view, a living polymerization can be seen as a chain polymerization that proceeds without the occurrence of chain transfer and termination events, through the establishment of a dynamic equilibrium between active and dormant species. Initially materialized by Szwarc in 1956 [1] this methodology indeed opened the door to well-defined polymers with precise and predetermined molar masses, compositions, topologies and functionalities. However, those living polymerization techniques exhibit two main drawbacks: (i) the impossibility to polymerize a wide range of functionalized vinylic monomers due to the incompatibility of the active center with certain functional groups and (ii) the requirement of stringent reaction conditions; especially the use of chemically ultrapure reagents as well as the absolute removal of air and of traces of water.

Until 15 years ago, living anionic and cationic polymerizations were the only available methods to reach a high degree of structural and compositional homogeneity of polymers before recent developments in macromolecular synthesis provide a new synthetic tool to easily achieve complex macromolecular architectures: controlled/living radical polymerization (CLRP). This general term gathers several novel free-radical polymerization techniques that enable a high degree of control to be reached. Indeed, free-radical polymerization differs from ionic polymerization by: (i) its relative ease-of-use (only dissolved oxygen has to be eliminated); (ii) the broad range of vinylic monomers which can be polymerized by a radical mechanism and (iii) the numerous processes that can be implemented (bulk, solution, emulsion, dispersion, etc.). However, the main limitation of free-radical polymerization is the total lack of control over the molar mass, the molar mass distribution (MMD), the chain-end functionalities and the macromolecular architecture. Therefore, bringing together the ease of use of free-radical polymerization with the high standard of control provided by living ionic polymerization, within a single polymerization process, simply brought about a revolution in the field of macromolecular synthesis.

In this view, various CLRP methods have been developed since the early 1980s, each of them being based on a different mechanistic approach and having encountered more or less success over the years. Basically, whatever the involved mechanism, their joint, key feature is the establishment of a dynamic equilibrium between propagating radicals, [Pradical dot], and various dormant species (i.e., end-capped, thus unable to propagate) throughout the polymerization process in order to decrease the occurrence of irreversible termination reactions to an extremely low level. The so-obtained equilibrium (Fig. 1) is triggered and governed by thermal, photochemical and/or chemical stimuli. For the success of such an approach, a polymer chain should spend most of the polymerization time under its dormant state.

Among the most well-established methods deriving from this concept are nitroxide-mediated polymerization (NMP) [2], [3], [4], [5], atom-transfer radical polymerization (ATRP) [6], [7], [8], [9] and reversible addition-fragmentation chain transfer (RAFT) [10], [11], [12]. NMP is historically the first and represents perhaps the easiest CLRP technology to apply. The aim of this comprehensive review is to offer the readers a global and in-depth overview of NMP from its emergence to the latest advances. It covers achievements in the synthesis of nitroxides and alkoxyamines, and their development for NMP, detailed kinetic aspects of NMP including the range of monomers that can be controlled as well as its application in homogeneous and heterogeneous media. Polymer characterization, functionalization methods, along with the synthesis of block copolymers and complex architectures, hybrid materials together with bioconjugates/biomaterials, potential applications and environmental considerations are also presented.

During an ideal living polymerization process such as living anionic polymerization, all polymer chains are created at the beginning of the polymerization and then grow homogeneously until the monomer is depleted. However, this phenomenon cannot be observed in a radical process due to the propensity of radicals to undergo self-termination. The control/livingness can only be achieved in the presence of reagents able to reversibly deactivate the propagating radicals and to establish a rapid equilibrium between active and dormant species. An ideal living polymerization system should exhibit: (i) a linear evolution of ln[1/(1  conversion)] with time, accounting for a constant propagating radical concentration; (ii) a linear increase of the number-average molar mass, Mn, with monomer conversion; (iii) low polydispersity indexes (PDIs), Mw/Mn, with Mw the weight-average molar mass; (iv) a quantitative α- and ω-functionalization and (v) the possibility for polymer chains to grow again when additional monomer is introduced, allowing block copolymers to be synthesized.

The termination rate is proportional to the square of the total radical concentration, while the propagation rate is directly proportional to the total radical concentration. Thus, a first strategy to suppress termination is to lower the macroradical concentration. This was first achieved by Otsu et al. [13], [14] with dithiocarbamate compounds. Nevertheless, the first real system that led to a successful living and controlled polymerization was developed by Solomon et al. [2]. In their patent, they described the use of nitroxides and alkoxyamines as a route to control the radical polymerization of several monomers, including acrylates and styrenics. This work derived from their previous studies on initiation, where they used nitroxides as radical trapping agents [4]. The readers interested in the history of this discovery are referred to the following review [4]. Since this pioneering work, Georges et al. [3], followed by many others, provided experimental proofs that this system is effective to control the polymerization of several vinyl monomers. The nitroxide-mediated polymerization was born.

NMP is based on a reversible termination mechanism between the growing propagating (macro)radical and the nitroxide, acting as a control agent, to yield a (macro)alkoxyamine as the predominant species. This dormant functionality generates back the propagating radical and the nitroxide by a simple homolytic cleavage upon temperature increase. When the latter is judiciously chosen, an equilibrium between dormant and active species, namely the activation–deactivation equilibrium, is established (Fig. 2). This equilibrium presents the advantage of being a purely thermal process where neither catalyst nor bimolecular exchange is required. The polymerization kinetics is governed by both this activation–deactivation equilibrium (with K = kd/kc, the activation–deactivation equilibrium constant) and the persistent radical effect (PRE) [15] (see Section 3.1.1 for details).

NMP was originally initiated by a bicomponent pathway, comprising a conventional thermal initiator, such as 2,2′-azobisisobutyronitrile (AIBN) or benzoyl peroxide (BPO), in combination with 2,2,6,6-tetramethylpiperidinyl-1-oxy (TEMPO, N1, Table 1) as the stable free nitroxide [3]. This system has the advantage to use conventional radical polymerization processes with the only addition of free nitroxides, which can be also highly desirable from both economic and practical points of view.

Fine-tuning the [nitroxide]0/[initiator]0 ratio is of high importance since the kinetics of the polymerization is governed by the amount of nitroxide in excess present after the initiation step [16]. As a consequence of a high excess of free nitroxide, the activation–deactivation equilibrium is shifted toward dormant species, which decreases the polymerization rate. All thermal initiators suffer from the difficulty to determine precisely the efficiency of the primary radicals produced by thermal decomposition to induce the polymerization (for instance due to cage effect and induced decomposition) and also the nature of the initiating group since the majority of these primary radicals undergo rearrangement or fragmentation reactions [17]. This usually leads to poorly reproducible polymerization kinetics and to ill-defined polymer end-groups. The [nitroxide]0/[initiator]0 ratio originally used was 1.3 [3]. Recently, Dollin et al. [18] revisited the influence of this ratio over the kinetics and showed that if this value is finely optimized (the ratio depends on the targeted molar mass and can be decreased down to 0.95 in certain conditions), the kinetics of the system could be strongly accelerated.

However, to circumvent this issue, the groups of Rizzardo [2] and Hawker [19], [20] developed the concept of unimolecular initiator that decomposes into both the initiating radical and the nitroxide. This compound, originally termed unimer, is called alkoxyamine initiator. Due to its particular structure, it leads, after dissociation, to a 1:1 release of initiating radical:nitroxide. Interestingly, the structure of the initiating end-group can be tuned to perform advanced macromolecular synthesis or post-modification chemistry (see Section 4.1 for details). Experimentally, it was observed that unimolecular initiators led to a better control over molar masses and MMDs than bimolecular initiating systems [19].

From a purely mechanistic point of view, one could wonder whether the nitroxide remains associated with the same polymeric chain during activation–deactivation cycles or if it can freely diffuse in the reaction medium. This question was elegantly tackled by Hawker and co-workers [21] who performed a cross-over experiment using a combination of a hydroxy-functionalized and a non-functionalized alkoxyamines for the polymerization of styrene (S), during which the migration of the nitroxide in the medium was demonstrated by high performance liquid chromatography (HPLC) (Fig. 3).

Lately, the importance of diffusion-controlled (DC) effects during CLRP processes has been a subject of debate. Some authors have considered these effects as negligible for NMP due to the fact that low molar masses (typical for CLRP) are obtained. However, other groups [22] considered only DC-termination whereas some authors [23] took into account the fact that the reactions of propagation, activation and deactivation of polymer radicals may also become diffusion-controlled. Penlidis and co-workers [24] raised this issue by performing the polymerization of styrene in the presence of size exclusion chromatography (SEC) polystyrene standards in order to obtain high viscosities even at the beginning of the polymerization, while avoiding the formation of a polymer network. The results, with and without the added polymer, were rather similar. The low importance of DC effects in NMP is then explained by the fact that moderately short macromolecules (when compared with typical chain lengths obtained by conventional radical polymerization) are obtained by CLRP processes and by the typical operating temperatures being usually above the glass transition temperature (Tg) of most of the resulting polymers [24].

Section snippets

Nitroxides

Numerous nitroxides have been designed and used for nitroxide mediated polymerization. The nitroxides (N1N120) and the alkoxyamines (A1A17) discussed in this section are gathered in Table 1, Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and in Table 8, Table 9, respectively.

Theoretical consideration: the ideal mechanism

The final elucidation of the NMP mechanism was obtained by the groups of Fischer [15] and Fukuda [250], who theoretically explained the minimization of the irreversible terminations by the persistent radical effect (PRE, see below for details). Previously, Johnson et al. highlighted this phenomenon using theoretical modelings but without performing a full rationalization [251].

The PRE can be qualitatively explained as follows (Fig. 26). We consider a compound (RY) that decomposes into a

Chain-end functionalized polymers

Due to the high chain-end fidelity observed in NMP, various end-functionalized (co)polymers have been prepared so far where α- and ω-functionalization pathways can be distinguished.

Performance benchmarks—brief comparisons with other CLRP methods

NMP was the first controlled/living free-radical polymerization technique to give results very close to those of anionic polymerization in terms of MMD control and ability to create block copolymers and more complex architectures. Other techniques using the same principle of reversible termination were later proposed such as the extensively used ATRP based on transition metal complex catalysts and its numerous variants (such as SET for instance). Tellurium-mediated radical polymerization and

Conclusion and perspectives

During the last thirty years, the research devoted to the synthesis of precisely tailored macromolecular architectures was particularly intense. Thanks to all the recent achievements in this field, it is now well-admitted that by using the whole battery of CLRP techniques, combined with other polymerization techniques and/or efficient coupling approaches [1142], it can give access to almost all kinds of macromolecular architectures. Yet, this undeniable scientific success strongly contrasts

Acknowledgments

The authors are grateful to all the students and colleagues who have been involved in NMP for their significant contributions over the last decade. The authors are indebted to J.-F. Pierson, R. Pirri, O. Guerret, S. Magnet, L. Couvreur, J.-L. Couturier and P. Gérard from Arkema for strong support and fruitful discussions. The French Ministry of Research and the CNRS are acknowledged for financial support. Certain sections of this review, included to present a more complete presentation of the

References (1143)

  • D.H. Solomon

    Genesis of the CSIRO polymer group and the discovery and significance of nitroxide-mediated living radical polymerization

    Journal of Polymer Science Part A: Polymer Chemistry

    (2005)
  • C.J. Hawker et al.

    New polymer synthesis by nitroxide mediated living radical polymerizations

    Chemical Reviews

    (2001)
  • M. Kato et al.

    Polymerization of methyl methacrylate with the carbon tetrachloride/dichlorotris-(triphenylphosphine)ruthenium(II)/methylaluminum bis(2,6-di-tert-butylphenoxide) initiating system: possibility of living radical polymerization

    Macromolecules

    (1995)
  • J.S. Wang et al.

    Controlled living radical polymerization—atom-transfer radical polymerization in the presence of transition–metal complexes

    Journal of the American Chemical Society

    (1995)
  • K. Matyjaszewski et al.

    Atom transfer radical polymerization

    Chemical Reviews

    (2001)
  • M. Kamigaito et al.

    Metal-catalyzed living radical polymerization

    Chemical Reviews

    (2001)
  • J. Chiefari et al.

    Living free-radical polymerization by reversible addition-fragmentation chain transfer: the RAFT process

    Macromolecules

    (1998)
  • M. Destarac et al.

    Dithiocarbamates as universal reversible addition-fragmentation chain transfer agents

    Macromolecular Rapid Communications

    (2000)
  • G. Moad et al.

    Living radical polymerization by the RAFT process

    Australian Journal of Chemistry

    (2005)
  • T. Otsu et al.

    Role of initiator-transfer agent-terminator (iniferter) in radical polymerizations: polymer design by organic disulfides as iniferters

    Die Makromolekulare Chemie Rapid Communications

    (1982)
  • T. Otsu

    Iniferter concept and living radical polymerization

    Journal of Polymer Science Part A: Polymer Chemistry

    (2000)
  • H. Fischer

    The persistent radical effect: a principle for selective radical reactions and living radical polymerizations

    Chemical Reviews

    (2001)
  • P.G. Odell et al.

    Rate enhancement of living free-radical polymerizations by an organic-acid salt

    Macromolecules

    (1995)
  • G. Moad et al.

    The Chemistry of Radical Polymerization

    (2006)
  • M. Dollin et al.

    Rapid additive-free TEMPO-mediated stable free radical polymerizations of styrene

    Journal of Polymer Science Part A: Polymer Chemistry

    (2007)
  • C.J. Hawker et al.

    Initiating systems for nitroxide-mediated “living” free radical polymerizations: synthesis and evaluation

    Macromolecules

    (1996)
  • C.J. Hawker

    Molecular-weight control by a living free-radical polymerization process

    Journal of the American Chemical Society

    (1994)
  • C.J. Hawker et al.

    Radical crossover in nitroxide mediated “living” free radical polymerizations

    Journal of the American Chemical Society

    (1996)
  • F. Diaz-Camacho et al.

    Effect of regime of addition of initiator on TEMPO-mediated polymerization of styrene

    Polymer Bulletin

    (2004)
  • M. Roa-Luna et al.

    Effect of the addition of inert or TEMPO-capped prepolymer on polymerization rate and molecular weight development in the nitroxide-mediated radical polymerization of styrene

    Journal of Applied Polymer Science

    (2008)
  • H.B. Hass et al.

    The nitroparaffins

    Chemical Reviews

    (1943)
  • N. Kocherginsky et al.

    Nitroxide spin labels: reactions in biology and chemistry

    (1995)
  • L.R. Mahoney et al.

    Calorimetric and equilibrium studies on some stable nitroxide and iminoxy radicals. Approximate oxygen–hydrogen bond dissociation energies in hydroxylamines and oximes

    Journal of the American Chemical Society

    (1973)
  • I. Novak et al.

    Electronic structure of persistent radicals: nitroxides

    Journal of Organic Chemistry

    (2004)
  • A.R. Forrester et al.

    Stable nitroxide radicals

    Nature

    (1964)
  • S. Jousset et al.

    Peculiar behavior of beta-phosphonylated nitroxides bearing a tert-octyl group during living/controlled radical polymerization of styrene: kinetics and ESR studies

    Macromolecules

    (2000)
  • E.A. Lissi et al.

    Reaction of di-tert-butylnitroxide radicals

    International Journal of Chemical Kinetics

    (1980)
  • K. Adamic et al.

    Kinetic applications of electron paramagnetic resonance spectroscopy. I. Self-reactions of diethyl nitroxide radicals

    Journal of the American Chemical Society

    (1971)
  • D.F. Bowman et al.

    Kinetic applications of electron paramagnetic resonance spectroscopy. III. Self-reactions of dialkyl nitroxide radicals

    Journal of the American Chemical Society

    (1971)
  • D.F. Bowman et al.

    Kinetic applications of electron paramagnetic resonance spectroscopy. II. Self-reactions of n-alkyl nitroxides and n-phenyl nitroxide

    Journal of the American Chemical Society

    (1971)
  • R.M. Dupeyre et al.

    Nitroxides. XIX. Norpseudopelletierine-N-oxyl, a new, stable, unhindered free radical

    Journal of the American Chemical Society

    (1966)
  • D. Benoit et al.

    Development of a universal alkoxyamine for “Living” free radical polymerizations

    Journal of the American Chemical Society

    (1999)
  • S. Grimaldi et al.

    Synthesis and applications to “Living” free radical polymerization of a new class of nitroxyl radicals

    Polymer Preprints (American Chemical Society Division of Polymer Chemistry)

    (1997)
  • S. Grimaldi et al.

    Acyclic β-phosphonylated nitroxides: a new series of counter-radicals for “Living”/controlled free radical polymerization

    Macromolecules

    (2000)
  • M. Iwamura et al.

    Novel formation of nitroxide radicals by radical addition to nitrones

    Bulletin of the Chemical Society of Japan

    (1967)
  • V.A. Reznikov et al.

    Stable nitroxyl radicals with a hydrogen atom at α-carbon atom of nitroxyl group

    Russian Chemical Bulletin

    (1996)
  • A. Nilsen et al.

    Nitroxide decomposition: implications toward nitroxide design for applications in living free-radical polymerization

    Journal of Polymer Science Part A: Polymer Chemistry

    (2006)
  • E. Breuer et al.

    Nitrones, nitronates and nitroxides

    (1989)
  • A.R. Forrester et al.

    Organic chemistry of stable free radicals

    (1968)
  • K. Hideg et al.

    Recent results in chemistry and biology of nitroxides

    Journal of Heterocyclic Chemistry

    (2005)
  • Cited by (1110)

    View all citing articles on Scopus
    View full text