Ganna Gryn’ova, Ellen T. Swann and Michelle L. Coote*
In the present work we use accurate quantum chemistry to evaluate several known and novel nitroxides bearing acid- base groups as pH-switchable control agents for room tem- perature NMP.
Controlled (living) radical polymerization (CRP) is an invaluable tech- nique for synthesising functional materials with predetermined molecu- lar weights and architectures.1 Control is achieved by maintaining the
concentration of the propagating polymeric radicals at a sufficiently low level, relative to the total number of polymer chains, by reversibly trapping them as a dormant form that is protected from deactivation via various side-reactions. There are several successful CRP types, differ- ing principally in the chemistry of this dynamic equilibrium, including: atom transfer radical polymerization (ATRP),2 reversible addition-
fragmentation chain transfer (RAFT)3 and nitroxide mediated polymeri-
zation (NMP).4
,-./0/)+1%@.*.)+,%$2".6.%'4%3".%6+#*%.A(#,#B)#(6;%C)'C+D+3#'*%+*E%2'6C.3#*D% $#E.<).+23#'*$%#*%3".%FGH:)
Of these techniques, NMP (Scheme 1) offers a number of attractive advantages including the low toxicity, colourless and odourless of the nitroxide control agent, and the ease of post-functionalisation of the polymer chain. However, the requirement for relatively high polymeri- zation temperatures (e.g., 120 °C for 2,2’,6,6’-tetramethylpiperidine-N- oxyl (TEMPO, 1) mediated styrene polymerization5,6 contributes to the
high-energy costs associated with the process, and increases the inter- ference from unwanted side-reactions within the polymerization.7 Thus,
an on-going challenge is the design of new nitroxides that are capable of operating at lower temperatures. To this end, N-(2-methylpropyl)-N- (1-diethylphosphono-2,2-dimethylpropyl)-N-oxyl (SG1, 3)8 and its
alkoxyamine initiator BlocBuilderTM (4) affords effective polymeriza-
tion of acrylates and styrene based monomers at temperatures of around 100-110 °C.9 The increased alkoxyamine decomposition rate k
d, relative
to TEMPO under the same conditions, is attributed to the increased polarity of the molecule.10 More recently, 2,2,5-trimethyl-4-phenyl-3-
azahexane-3-nitroxide (TIPNO, 5)11 and its derivatives 6 and 7,12,13
have been used effectively at polymerization temperatures as low as 85 °C.14 Despite these important advances, room temperature NMP re-
mains elusive.
,-./0/)2:%1.,.23.E%*#3)'I#E.$%+*E%+,J'I0+6#*.%2'*3)',%+D.*3$%
As one drives the polymerization temperature closer to room tem- perature, this raises a second challenge – that of controlling the activa- tion and deactivation of the nitroxide. For high temperature nitroxides such as TEMPO, this is easily managed in that the alkoxyamines are stable at normal service / storage temperatures and are easily activated for polymerization by heating. However, for a room temperature active NMP agent, some means of (reversibly) deactivating the alkoxyamine
N R' R" O P N R' R" O + P + M alkoxyamine nitroxide polymer
radical monomer > 80 °C
Unwanted side reactions
(disproportionation, recombination, chain transfer, ...)
kc kd k p N O P(O)(OEt)2 N O P(O)(OEt)2 COOH N O NaO3S N O COOH N O N O N N O HOOC 4-carboxyTEMPO N O TEMPO SG1 BlocBuilderTM TIPNO N O HOOC N O HOOC HOOC N O N O HOOC N O N O PROXYL TMIO 3,4-dicarboxyPROXYL 3-carboxy-etPROXYL 5-carboxyTMIO N O HOOC 3-carboxyPROXYL 4-sulfonateTMIO 6 7 8 1 2 3 4 5 9 10 11 12 13 14 15 COOH
!"##$%&!'(&"%! )*+,-./0%.12!
3!"!!"#$%&'#$!%&'%$!44$!'()! *+,-!./01234!,-!5!*+6!7/834!9/:,6;8!/<!=+6>,-;18!%&'%!
is essential. One possible strategy for switching the reactivity of nitrox- ides and their alkoxyamines is via pH. As shown in Figure 1, both the nitroxide and alkoxyamine feature polar bonds, whose stability should in principle be susceptible to the external electric fields. For example, on the nitroxide side, introduction of a positive charge should destabi- lise the radical and stabilise the alkoxyamine leading to reduced kd, while introduction of a negative charge should do the opposite. On the leaving group side, introduction of a positive charge should destabilise the alkoxyamine (whilst of course having no effect on the nitroxide) and increase kd, whilst a negative charge should do the opposite. In practical terms, manipulating the leaving group on the initial alkoxy- amine is only likely to be effective for the first initiation step, as the leaving group will become an increasingly remote end group on the growing polymer radical.
567+,20 89! ?<<6:;-! /<! :+31@6A! @1/0B-! /2! ;+6! -;3C,4,;,6-! /<! ;+6! 34D/E83>,26! 32A! B1/A0:;!2,;1/E,A6!13A,:34#!F2!;+6!2,;1/E,A6!-,A6!3!B/-,;,G6!:+31@6!-+/04A!-;3C,4,-6! ;+6!34D/E83>,26!32A!A6-;3C,4,-6!;+6!2,;1/E,A6!13A,:34$!463A,2@!;/!+,@+!HI?-!32A! 4/J61!G3406-!/<!(A$!J+,46!3!26@3;,G6!:+31@6!-+/04A!A/!;+6!/BB/-,;6#!F2!;+6!463G(
,2@!@1/0B!-,A6!3!B/-,;,G6!:+31@6!-+/04A!A6-;3C,4,-6!;+6!34D/E83>,26!32A!-;3C,4,-6! ;+6! 13A,:34! KB1/G,A6A! ,;! 16>3,26A! 3--/:,3;6A! J,;+! ,;L$! 463A,2@! ;/! 4/J! HI?-! 32A! +,@+61!G3406-!/<!(A$!J+,46!3!26@3;,G6!:+31@6!-+/04A!A/!;+6!/BB/-,;6#!
In recent years, a number of nitroxide control agents containing acid/base functional groups (e.g., 15) have been developed. These were initially designed with a view to increasing the water solubility of the nitroxides and/or alkoxyamine initiators, rather than switching the reac- tivity itself. Thus, in a number of cases, the kinetic behaviour of these reagents in their charged and non-charged forms has not yet been di- rectly compared.15 Nonetheless, in recent years the prospect of exploit- ing polar effects to pH switch reactivity has received increasing attention, and a number of literature examples are shown in Scheme 3. Summarising these results, protonation of basic groups on either the nitroxide or leaving group can lead to changes in kd of around an order of magnitude in both non-polar and polar solvents, consistent with the behaviour expected from Figure 1. Interestingly, in the literature exam- ples thus far tested negative charges have been shown to have a negli- gible effect. For example, Bertin et al.16 have shown experimentally that the ionisation of COOH in the leaving alkyl fragment of BlocBuilderTM (4) has little effect on its decomposition rate constant kd, though it dra- matically increases its solubility in polar solvents, while Edeleva et al.17 found that deprotonation of carboxy groups on 18 also had negligible effect on kd. Significantly, both of these experiments were carried out in polar solvents.
Recently, we have shown that negative charges can exert much larger pH-switches on the stability of nitroxides in low polarity envi-
ronments.18 Using high-level ab initio molecular orbital theory calcula- tions, validated by mass spectrometry experiments, we have shown that, for example, the bond dissociation energies of a range of alkoxyamines of 4-carboxy-TEMPO (2) are reduced by around 20 kJ mol-1 upon deprotonation of the carboxylic acid group. It is worth noting that this increase in radical stability (by more than 2300 fold) is significantly greater than any of the effects so far reported for positive charges, and acts over a much longer range (5.7Å from negative charge to radical centre, compared with 3.5Å in 18). We have shown that the effects are electrostatic and can be replicated when the charged group is replaced with an electric field of the same polarisation. Moreover, we have shown that this long-range radical stabilisation can occur whenever delocalised radicals interact with localised negative charges, and arises in the increased polarisability of resonance-stabilised radicals, which allows them to minimise electron-electron repulsion. Whilst the effects are largely quenched in water and highly polar solvents such as DMSO, our calculations suggest that they stay largely intact in non-polar organ- ic solvents such as toluene and dichloromethane. Thus, provided solu- bility issues can be overcome, one should in principle be able to increase the kd in a nitroxide polymerization by 3 orders of magnitude (or more) through deprotonation of an acid-base group. This raises the exciting prospect of substantially lowering the polymerization tempera- ture of NMP while simultaneously introducing a convenient “on-off switch”. This could potentially enable faster polymerization of a broad- er range of monomers, under milder conditions and correspondingly diminish the occurrence of the undesired side-reactions. Moreover, the polymer chains so formed will be capped with a pH-sensitive nitroxide –potentially useful for post-functionalization19 via nitroxide radical coupling (NRC)20 or binding to various bioactive substrates.21
:;<2120 =9! M,;613;016! 6E3>B46-! /<! BN! 6<<6:;-! /2! 34D/E83>,26!(A#! O/1! <01;+61!
A6;3,4-!-66!16<!'P!</1!>$!16<-!%%$%)!</1!8?$!32A!16<0'Q!</1!8@!32A!8A#!
In the present work we use rigorously benchmarked18,24,25 quantum- chemical calculations‡ to assess the suitability of several known and novel nitroxides with acidic substituents for the bulk NMP of styrene. We aim to identify the structures that (i) form NO–STY bonds that are initially weaker than those in TEMPO–STY and (ii) can be weakened even further by deprotonating the acid-base group. Our design strategy
N O R" R' P N O R" R' P (A) N O R" R' N O R" R' (B) N O R" R' P N O R" R' P N O R" R' N O R" R' N O R" R' P N O R" R' P N O R" R' N O R" R' N O R" R' P N O R" R' P N O R" R' N O R" R' N O R" R' P N O R" R' P N O R" R' N O R" R' H ig h e r BD E L o w e r BD E L o w e r k d H ig h e r k d N N O N N N HN O H N N + 2H - 2H N O P(O)(OEt)2 N N O P(O)(OEt)2 NH + H - H kd increases
~16 fold at 80 °C in org. solvent ~64 fold at 75 °C in D2O/CD3OD kd decreases ~15 fold at 95-100 °C in H2O and chlorobenzene N N O COOH N O P(O)(OEt)2 COOH + H - H N O P(O)(OEt)2 COO kd change negligible in t-butylbenzene + H - H N N O COO kdchange negligible in H2O COOH COOH COOH COOH ~ 4.7 Å ~ 3.5 Å ~ 3.5 Å ~ 5.2 Å 4 16 17 18
!"#$%&'()*+,%#$%-%!".%/'0+,%1'2#.30%'4%5".6#$3)0%7897! "#!$%&':;%7897;%33;%9<=%>%4%
rate a remote negative charge (e.g., in the form of carboxylate or phos- phate group) so as to induce further stabilisation upon its deprotonation. We already know18 that the largest pH-switches§ are observed for more stabilised radicals and relatively destabilised anions, and that multiple charges lead to higher switches. On these grounds we have formed the test set shown in Figure 2, which includes a number of known com- pounds (2, 11, 12, 14)26 and a number of synthetically achievable deriv- atives of known compounds (8, 19).27 None of these compounds have been assessed for their pH switching ability before. Figure 2 shows their calculated logK values for trapping of a styrene dimer radical in toluene solution at both 25 °C and 120 °C. It is well known that TEMPO is successful in mediating styrene polymerization at 120 °C,5 hence we use the homolysis equilibrium constant K for the correspond- ing alkoxyamine TEMPO–STY as a reference (green line in Figure 2): systems with a lower K are expected to succeed, whereas species with logK > 12 would be too stable to release the propagating radicals at a sufficient rate.28
567#$+(89(?'@+)#3"6$%'4%3".%ABC5%D'*E%"'6',0$#$%.F(#,#D)#(6%2'*$3+*3$%G2'6D#< *+3#'*% #$% E.4#*.E% +$% 3".% 4')H+)E% ).+23#'*;% E.2'6I'$#3#'*% C% 3".% ).J.)$.% '*.K;% 'D3+#*.E%4)'6%3".%LMNO$%4')%3".%3.$3.E%*#3)'P#E.$%H#3"%+%$30)0,%E#6.)%I)'I+@+3< #*@%)+E#2+,;7Q%2+,2(,+3.E%#*%3".%D(,R%$30).*.%$',(3#'*%+3%7S%T5%+*E%978%T5:%U)..*% ,#*.%2')).$I'*E$%3'%,'@(%V%97:
From Figure 2 we note that protonated 4-carboxy-TEMPO 2 per- forms almost identically to the non-substituted TEMPO itself – its logK
is ca. 12 at 120 °C, but is too high at room temperature to afford polymerization. Deprotonation lowers these values by several orders of magnitude, however insufficiently to reach the threshold at 25 °C. 5- Carboxy-TMIO 14 in the neutral form forms bonds that are slightly
almost identical to that of 4-COO–-TEMPO. The pH-switch in this system is non-surprisingly smaller, because the carboxylate group in 14 is stabilised via conjugation with the aromatic ring. 3,4- DicarboxyPROXYL 11 behaves similarly to 2, however when doubly deprotonated, its logK drops by several orders of magnitude. While it is still above 12 at 25 °C, it is expected to afford significant lowering of the polymerization temperature to 60 °C and possibly less (see Table S1 of the ESI). Furthermore, the SG1 derivative 19 forms weaker bonds compared to TEMPO even when unswitched, which are weakened even further in its (2–) form to also allow for polymerization temperatures potentially <60 °C (see Table S1 of the ESI). The carboxylated deriva- tive of TIPNO 8 performs even better in that at 25 °C the neutral form is still above the threshold and hence stable, while the deprotonated form now has logK equal to 10.5, and is thus potentially able to control the NMP of styrene at room temperature. Finally, the PROXYL ana- logue 12 reveals a similar behaviour and also has logK < 12 when deprotonated. This nitroxide is characterised with the largest switch, presumably due to the proximity of the charge to the formal radical centre.
,"%:'#;6"%;((
Based on rigorously benchmarked theoretical calculations, we sug- gest that both the !-ethyl analogue of 3-carboxy-PROXYL 12 and a novel derivative of TIPNO 8 are expected to control styrene NMP at a room temperature, provided their carboxylic groups are deprotonated and the polarity of the environment is kept low. We also show that certain doubly-charged nitroxides – a derivative of the commercially available SG1 with the (2–)-charged phosphate 19 and 3,4- dicarboxyPROXYL 11 – also lower the polymerization temperature required to approximately 60 °C. The principal synthetic challenge in implementing these findings experimentally is that of maintaining the solubility of the deprotonated nitroxide in bulk monomer solution, however, we envisage that, if necessary, this could be achieved without altering the reactivity through inclusion of additional lipophilic side chains on the nitroxide. Lower polymerization temperatures are benefi- cial not only economically, but also in that they are likely to help di- minish the occurrence of unwanted side-reactions. Our results thus offer new strategies for improving the NMP conditions, broadening the scope of controlled monomers and forming polymer chains, capped with a pH-sensitive nitroxide end-group, suitable for further functionalization. 1:<%"='+>7+*+%?;((
We gratefully acknowledge financial support from the Australian Re- search Council (ARC) Centre of Excellence for Free Radical Chemistry and Biotechnology, an ARC Future Fellowship (to M. L. C.), and gen- erous allocations of computing time on the National Facility of the Australian National Computational Infrastructure. We also thank Prof. Martin Banwell and Prof. Michael Monteiro for helpful discussions. )"?+;(&%>($+@+$+%:+;%
Australian Research Council Centre of Excellence for Free Radical Chemistry and Biotechnology, Research School of Chemistry, Australian National University, Canberra, Australian Capital Territory 0200, Aus- tralia. Email: [email protected]
!"##$%&!'(&"%! )*+,-./0%.12!
3!"!!"#$%&'#$!%&'%$!44$!'()! *+,-!./01234!,-!5!*+6!7/834!9/:,6;8!/<!=+6>,-;18!%&'%!
† Electronic Supplementary Information (ESI) available. Full details of the calculations and geometries and energies of all species are provided. See DOI: 10.1039/c000000x/
‡ Calculations were carried using an ONIOM approximation to G3(MP2,CC)(+)//M06-2X/6-31+G(d) in conjunction with CPCM- UAKS/B3LYP/6-31G+(d) solvation energies. For further details, see the ESI.
§ We define pH-switch on HA-X-R BDFE as BDFE(HA-X-R) –
BDFE(–A-X-R). In this was, positive switch means that upon deprotona-
tion the bond is weakened (the BDFE is lowered).
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1 K. Matyjaszewski and J. Spanswick, Mater. Today, 2005, 8, 26.
2 For a review see N. V. Tsarevsky and K. Matyjaszewski, Chem. Rev.,
2007, 107, 2270.
3 For a review see G. Moad, E. Rizzardo and S. H. Thang, Aust. J. Chem.,
2009, 62, 1402.
4 For a review see C. J. Hawker, A. W. Bosman and E. Harth, Chemical
Reviews, 2001, 101, 3661.
5 M. K. Georges, R. P. N. Veregin, P. M. Kazmaier and G. K. Hamer, Macromolecules, 1993, 26, 2987.
6 T. Fukuda, T. Terauchi, A. Goto, K. Ohno, Y. Tsujii, T. Miyamoto, S.
Kobatake and B. Yamada, Macromolecules, 1996, 29, 6393.
7 G. Gryn’ova, C. Y. Lin and M. L. Coote, Polym. Chem., 2013, 4, 3744.
8 D. Benoit, S. Grimaldi, S. Robin, J.-P. Finet, P. Tordo and Y. Gnanou, J. Am. Chem. Soc., 2000, 122, 5929.
9 (a) J. Nicolas, B. Charleux, O. Guerret and S. Magnet, Angew. Chem.
Int. Ed., 2004, 43, 6186; (b) K. Bian and M. F. Cunningham, J. Polym. Sci.: Part A: Polym. Chem., 2006, 44, 414; (c) S. Tomoeda, Y. Kitayama, J. Wakamatsu, H. Minami, P. B. Zetterlund and M.
Okubo, Macromolecules, 2011, 44, 5599.
10 B. Charleux and J. Nicolas, Polymer, 2007, 48, 5813.
11 D. Benoit, V. Chaplinski, R. Braslau and C. J. Hawker, J. Am. Chem.
Soc., 1999, 121, 3904.
12 (a) L. Marx and P. Hemery, Polymer, 2009, 50, 2752; (b) Y. Le Du, L.
Binet, P. Hémery and L. Marx, J. Polym. Sci., Part A: Polym. Chem.,
2012, 50, 2871.
13 R. Nicolay, L. Marx, P. Hémery and K. Matyjaszewski, Macromole-
cules, 2007, 40, 6067.
14 J. Ruehl, N. L. Hill, E. D. Walter, G. Millhauser and R. Braslau, Mac-
romolecules, 2008, 41, 1972.
15 R. P. N. Veregin, M. K. Georges, G. K. Hamer and P. M. Kazmaier, Macromolecules, 1996, 28, 4391; (b) C. Viklund, A. Nordström, K.
Igrum, F. Svec and J. M. J. Fréchet, Macromolecules, 2001, 34, 4361;
(c) W. Huang, B. Charleux, R. Chiarelli, L. Marx, A. Rassat and J.-P.
Vairon, Macromol. Chem. Phys., 2002, 203, 1715.
16 D. Bertin, D. Gigmes, S. R. A. Marque, D. Siri, P. Tordo and G. Trap- po, Chem. Phys. Chem. 2008, 9, 272.
17 M. V. Edeleva, I. A. Kirilyuk, I. F. Zhurko, D. A. Parkhomenko, Y. P.
Tsentalovich and E. G. Bagryanskaya, J. Org. Chem., 2011, 76, 5558.
18 (a) G. Gryn’ova, D. L. Marshall, S. J. Blanksby and M. L. Coote,
Nature Chem., 2013, 5, 474; (b) G. Gryn’ova and M. L. Coote, J. Am. Chem. Soc., 2013, submitted.
19 L. Petton, A. E. Ciolino, B. Dervaux and F. E. Du Prez, Polym. Chem.,
2012, 3, 1867, and references therein.
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
20 Z. Jia, C. A. Bell and M. J. Monteiro, Macromolecules, 2011, 44,
1747.
21 J. Ruehl, C. Morimoto, D. J. Stevens and R. Braslau, React. Funct.
Polym., 2008, 68, 1563.
22 (a) P. Brémond and S. R. A. Marque, Chem. Commun., 2011, 47,
4291; (b) P. Brémond, A. Koïta, S. R. A. Marque, V. Pesce, V.
Roubaud and D. Siri, Org. Lett., 2012, 14, 358.
23 E. Bagryanskaya, D. Bertin, D. Gigmes, I. Kirilyuk, S. R. A. Marque,
V. Reznikov, G. Roshchupkina, I. Zhurko and D. Zubenko, Macro-
mol. Chem. Phys., 2008, 209, 1345.
24 M. L. Coote, C. Y. Lin, A. L. J. Beckwith and A. A. Zavitsas, Phys.
Chem. Chem. Phys., 2010, 12, 9597.
25 J. L. Hodgson, C. Y. Lin, M. L. Coote, S. R. A. Marque and K.
Matyaszewski, Macromolecules, 2010, 43, 3728.
26 (a) J. F. W. Keana, K. Hideg, B. Birrell, O. Hankovsky, G. Ferguson
and M. Parvez, Can. J. Chem. 1982, 60, 1439; (b) L. T. L. Wong, R.
Schwenk and J. C. Hsia, Can. J. Chem., 1974, 52, 3381; (c) D. A.
Reid, S. E. Bottle, A. S Micallef, Chem. Comm., 1998, 17, 1907; (d)
J. T. Paletta, M. Pink, B. Foley, S. Rajca and A. Rajca, Org. Lett.
2012, 14, 5322.
27 C. Lagercrantz and L. Eberson, Acta Chem. Scand., 1995, 49, 399.
28 (a) H. Fischer, Chem. Rev., 2001, 101, 3581; (b) D. Bertin, D. Gigmes,
S. R. A. Marque and P. Tordo, Chem. Soc. Rev., 2011, 40, 2189.
29 Styryl propagating radical was modelled as a dimer to account for the penultimate unit effect, see D. Bertin, P.-E. Dufils, I. Durand, D. Gigmes, B. Giovanetti, Y. Guillaneuf, S. R. A Marque, T. Phan and
many important implications and potentially valuable applications (Figure 6.5.1). For example, the fact that the remote negative charges can affect the radical reactivity implies that a common technique of using negative charges as ‘inert’ labels in the mass