Reversible addition–fragmentation chain transfer (RAFT) radical polymerization and the synthesis of water-soluble (co)polymers under homogeneous conditions in organic and aqueous media
Introduction
The last 15–20 years has been witness to unprecedented advances in controlled polymer synthesis. This has been due, in part, to the discovery and development of controlled free radical polymerization (CRP) techniques [1], [2], [3]. Today, these techniques facilitate the synthesis of macromolecules with a level of control approaching that of the more traditional living techniques, such as anionic or group transfer polymerizations, while possessing the versatility and robustness of conventional free radical polymerization.
There are several distinct CRP techniques [4]. The oldest of these may be generally classified as stable free radical polymerization (SFRP), with nitroxide-mediated (NMP) [5] systems being the most widely examined, although syntheses are not limited to the application of this specific class of mediating agent [6], [7]. SFRP was first reported by workers at CSIRO in Australia [8] and subsequently developed by Georges and coworkers, as well as others [5], [9], [10], [11], [12], [13]. Atom transfer radical polymerization (ATRP) [14], [15], independently developed by Sawamoto et al. [13] and Wang and Matyjaszewski [16] is a transition metal-mediated polymerization in which the transition metal complex (most commonly a Cu species but may also be a Ru or Fe complex for example) participates in a reversible end-capping process with a propagating polymer chain via an oxidation–reduction mechanism. ATRP has so far been the mostly widely studied of the CRP techniques and has proven to be highly versatile [4], [14], [15], [17]. Reversible addition–fragmentation chain transfer (RAFT) polymerization, also discovered by researchers at CSIRO, was reported in the late 1990s [18], [19], [20], [21], [22], [23]. Around the same time researchers in France described a technique they termed MADIX for Macromolecular Design by Interchange of Xanthate [24]. Both MADIX and RAFT operate via an identical addition–fragmentation chain transfer mechanism [25], [26], i.e. they are identical processes, with MADIX referring specifically to those polymerizations mediated by xanthates. The acronym RAFT describes systems employing all other thiocarbonylthio mediating agents. For simplicity sake we will refer to all systems herein as RAFT polymerizations. One of the newer examples of CRP is tellurium-mediated radical polymerization (TERP) [27], [28], [29], [30], [31], [32], [33]. TERP is controlled by reagents of general formula RTeR′. While this particular technique has been little investigated, it has been shown to work effectively for styrene, N,N-dimethylacrylamide, acrylonitrile and a variety of (meth)acrylates including the functional species 2-hydroxyethyl methacrylate and 2-(dimethylamino)ethyl methacrylate. Closely related to TERP is organostibene-mediated radical polymerization (SBRP) which was reported by Yamago et al. [33], [34], [35]. SBRP operates on the same principle as TERP but was demonstrated to be somewhat more versatile and exhibited a higher degree of control especially with respect to the molecular mass. One notable feature of SBRP is its applicability to both conjugated and unconjugated monomers. For example, the authors demonstrated the ability to control the polymerization of N-isopropylacrylamide, N-vinylpyrrolidone, and vinyl acetate—the latter two species are still difficult to control employing more established CRP techniques. The most recent report of a new CRP technique is that described by Caille et al. and was termed quinone transfer radical polymerization or QTRP [36]. This process appears to be based on a catalytic, reversible homolytic cleavage of a C–C bond of a substituted ortho-quinone. In particular, the catalyst employed is cobalt (II) acetylacetonate and the quinone is phenanthrenequinone. To date, QTRP has only been demonstrated to be effective for thermally-initiated styrene polymerizations, and then only if conversions are limited to ⩽∼50%. Each of these CRP methodologies has certain advantages and disadvantages associated with them; however, none of them exhibit all of the desired features of an “ideal” CRP technique.
One of the more environmentally and commercially significant developments in recent years has been the ability to conduct controlled polymerizations directly in aqueous media under homogeneous conditions [37]. Indeed, NMP [12], ATRP [38], [39], [40], [41], [42], [43], [44], and RAFT [45], [46], [47] have all been employed for such syntheses. While NMP is limited with respect to monomer choice and polymerization conditions, the application of ATRP has been shown to be a reasonably successful strategy for preparing well-defined (co)polymers under homogeneous aqueous conditions, although polymerizations are rarely as well controlled as when conducted in mixed water/organic or wholly organic media [39], [40], [41]. On the other hand, RAFT has thus far proven to be an extremely useful technique for the synthesis of hydrophilic (co)polymers directly in water under homogeneous conditions, although it is also not without its problems [45], [46], [48]. Herein we discuss the scope and limitations with respect to the application of RAFT for the preparation of hydrophilic/water-soluble (co)polymers both in organic and aqueous media under homogeneous conditions. Rather than adopting a strictly historical approach we will examine the RAFT synthesis of water-soluble/dispersible materials by monomer functionality, starting with anionic/acidic monomers, followed by cationic/basic species, then zwitterionic substrates, and finally non-ionic monomers. We have adopted this strategy since this is the order in which our group initiated our early studies. Heterogeneous [49] RAFT systems will not be reviewed.
Section snippets
Reversible addition–fragmentation chain transfer (RAFT) radical polymerization: an overview
RAFT [50], [51], [52] operates on the principle of degenerative chain transfer and as such differs fundamentally from both SFRP [5] and ATRP [14], [15]. Key to successful RAFT is appropriate choice of a so-called RAFT agent or RAFT chain transfer agent (CTA). These are thiocarbonylthio species belonging to one of the following general families of compounds: dithioesters [18], xanthates [53], [54], dithiocarbamates [20], [24], [54], [55], and trithiocarbonates [56], [57], [58], [59], [60], Fig. 1
The RAFT mechanism
RAFT can be described using the same sequence of elementary steps used to describe a conventional free radical polymerization, with several amendments/additions taking into account radical reactions involving the RAFT agent, see Scheme 1. Since this is a free radical polymerization process radicals must first be generated (1, step (i) Scheme 1). Since RAFT is no more than a conventional free radical polymerization conducted in the presence of a suitable thiocarbonylthio species, traditional
Experimental criteria for a controlled RAFT polymerization
When evaluating the characteristics of RAFT systems it is common to focus on the kinetic aspects of the polymerization in an effort to determine whether or not a polymerization proceeds in a controlled manner. Traditional living systems, as defined by Szwarc [90], are chain growth polymerizations which proceed in the complete absence of undesirable termination and chain transfer reactions. Such a definition does not however implicitly suggest the ability to tailor molecular masses nor produce
RAFT polymerization of water-soluble/hydrophilic monomers in aqueous and non-aqueous media
One of the remarkable features of RAFT is its general versatility with respect to monomer choice and polymerization conditions. Of these key attributes, the ability to prepare water-soluble polymers with a high degree of functionality either directly in water or in organic media is particularly noteworthy. Few, if any polymerization technique, has proven to be as applicable to the homogeneous aqueous polymerization of hydrophilic monomers in a controlled fashion as has RAFT. This key attribute
Monomer choice
One of, if not the, major redeeming feature of RAFT is the wide range of functional and non-functional monomers which can be polymerized in a controlled fashion via this technique. To date, RAFT has been successfully employed in the polymerization of nonionic [22], [55], [68], [94], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113], [114], [115], [116], [117], [118], [119], [120], [121], [122], [123], [124], [125]
Limitations of homogeneous aqueous RAFT
As with all polymerization techniques RAFT does suffer from certain limitations. In some instances, especially for nonionic monomers with certain dithioester RAFT agents, polymerizations can be significantly retarded, often requiring extremely long (>24 h) polymerization times to achieve even appreciable conversion. In some systems complete inhibition or an induction period is observed. This observed phenomenon is intimately related to the nature of the intermediate radical species and has been
Modification of gold surfaces with water-soluble RAFT polymers
By virtue of the RAFT mechanism, (co)polymers prepared via this route bear thiocarbonylthio end-groups. These functional groups are readily reduced to thiolate species with NaBH4 under very facile conditions directly in aqueous media. If this reducing chemistry is conducted in the presence of a suitable gold substrate, such as a gold sol or a clean gold plate, then it is possible to directly modify the gold surface via covalent attachment of the thiolate-terminated polymer chains, Fig. 42.
This
Summary
In the eight years since its initial disclosure in the open literature by researchers at CSIRO, the reversible addition–fragmentation chain transfer (RAFT) radical polymerization technique has proven itself to be an extremely versatile tool for synthetic polymer chemists. In this article we have described the recent advances in the application of RAFT polymerization for the preparation of hydrophilic/water-soluble (co)polymers either directly in aqueous media under homogeneous conditions or in
Acknowledgements
ABL wishes to thank the Department of Chemistry and Biochemistry, the Dean of the College of Science and Technology and the VP of Research for generous start-up funds, Additionally, ABL would like to acknowledge the Office of the Provost at USM, the US Department of Energy, Oak Ridge Associated Universities, the National Institutes of Health, the National Science Foundation MRSEC center at USM, and Avery-Dennison for financial support of some of the research described herein. CLM would like to
References (295)
Preparation of functionalized polymers using living and controlled polymerizations
React Funct Polym
(2001)- et al.
Controlled/living radical polymerization in aqueous media: homogeneous and heterogeneous systems
Prog Polym Sci
(2001) - et al.
Dihydroxyl-terminated telechelic polymers prepared by RAFT polymerization using functional trithiocarbonates as chain transfer agent
Polymer
(2004) - et al.
A novel synthesis of functional dithioesters, dithiocarbamates, xanthates and trithiocarbonates
Tetrahedron Lett
(1999) Living and controlled polymerization: synthesis, characterization and properties of the respective polymers and copolymers
(2006)- et al.
New polymer synthesis by nitroxide mediated living radical polymerizations
Chem Rev
(2001) - et al.
A novel “stable” radical initiator based on the oxidation adducts of alkyl-9-bbn
J Am Chem Soc
(1996)