The reactivity of nitroxides towards alkenes

Article abstract of DOI:10.1016/S0040-4039(00)00440-8

At elevated temperatures, nitroxides (e.g. 1,1,3,3-tetramethyl-2,3- dihydroisoindol-2-yloxyl) undergo a slow addition reaction with acrylonitrile, methyl acrylate and styrene to give the bis-nitroxide adducts. With alkenes containing an allylic hydrogen such as methyl methacrylate and 6-methylene-1,4-oxathiepan-7-one, the major reaction observed was hydrogen abstraction. The resulting hydroxylamines can be trapped as Michael addition products. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00440-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3673–3676
The reactivity of nitroxides towards alkenes
Fawaz Aldabbagh,a W. Ken Busfield,a Ian D. Jenkins a,∗ and San H. Thang b
aSchool of Science, Griffith University, Nathan, Brisbane, Queensland 4111, Australia
bCSIRO Molecular Science, Bag 10, Clayton South, Victoria 3169, Australia
Received 9 February 2000; accepted 13 March 2000
Abstract
At elevated temperatures, nitroxides (e.g. 1,1,3,3-tetramethyl-2,3-dihydroisoindol-2-yloxyl) undergo a slow
addition reaction with acrylonitrile, methyl acrylate and styrene to give the bis-nitroxide adducts. With alkenes
containing an allylic hydrogen such as methyl methacrylate and 6-methylene-1,4-oxathiepan-7-one, the major
reaction observed was hydrogen abstraction. The resulting hydroxylamines can be trapped as Michael addition
products. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: nitroxide; alkene; radical trapping; hydroxylamine.
1. Introduction
As the trapping of carbon-centred radicals by nitroxides is more efficient than their addition to
monomers, we investigated the mechanism of the AIBN-initiated polymerization of the ring-opening
monomer, 6-methylene-1,4-oxathiepan-7-one 2,1 by carrying out the reaction of 2 (0.6 mmol) with
alkoxyamine 3 (0.1 mmol) under vacuum, in benzene (2 ml) at 105°C for 6 hours. Upon heating, the
alkoxyamine 3 undergoes reversible homolysis to generate cyanoisopropyl radicals and the nitroxide,
1,1,3,3-tetramethyl-2,3-dihydroisoindol-2-yloxyl (TMIO) 1.2 Thus, 3 behaves as an analogue of AIBN,
while at the same time providing a low concentration of nitroxide 1.
To our surprise, there was no evidence of the addition of cyanoisopropyl radicals to 2. The only
products observed by HPLC/MS analysis were the nitroxide 1 (59%), compound 5 (36%) and the
∗
Corresponding author. E-mail: i.jenkins@az.gu.edu.au (I. D. Jenkins)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00440-8
tetl 16720

3674
alkoxyamine 3 (5%). Product 5 is formed as a result of hydrogen abstraction by the nitroxide 1 (confirmed
in a separate experiment employing only 1 and 2) followed by trapping of the resultant allylic radical at
the least hindered position (Scheme 1).
Scheme 1.
It is clear that cyanoisopropyl radical addition to the monomer double bond is not competitive with
reactions such as hydrogen abstraction from 2. We were unable to determine the precise fate of the
cyanisopropyl radicals, which probably either dimerized or underwent hydrogen transfer to nitroxide 1
to give methacrylonitrile. The other product of hydrogen abstraction is hydroxylamine 4, which was not
detected by HPLC/MS presumably as it underwent rapid oxidation back to nitroxide 1 upon exposure
to air during product analysis. In order to provide some supporting evidence for the formation of 4, we
carried out a separate experiment in which 1 (0.3 mmol) was heated with 2 (1.0 mmol) in the presence of a
good Michael acceptor (acrylonitrile, 3 ml). After 18 hours at 105°C under vacuum, HPLC/MS analysis
showed the formation of 5 (28%), 7a (57%) and 8a (15%) (Scheme 2). The formation of 7a provides
good supporting evidence for the presence of hydroxylamine 4 (as described later, such compounds
undergo facile Michael addition reactions to acrylonitrile). The formation of the di-TMIO adduct 8a was
unexpected and prompted an investigation into the addition reactions of nitroxides with some common
monomers.
Scheme 2.
2. Addition of nitroxides to alkenes
Evacuated solutions of TMIO 1 (1.3 mol l−1) in distilled neat monomer were heated at 110°C for
5 days. The results obtained are summarized in Table 1. Nitroxide addition to styrene proved the most
facile, and it is interesting to note that the analogous di-TEMPO adduct of styrene was recently reported
by Connolly and Scaiano,3 when TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl) was heated in styrene at
120°C for 2 hours. This suggests that resonance stabilization of the intermediate radical 6 is important for
the formation of the bis-nitroxide adducts 8 (Scheme 2). The concurrent formation of significant amounts
of adducts 7a and 7b suggests that the intermediate radical 6 also takes part in a hydrogen abstraction

3675
Table 1
Percentage yields of products as a result of heating TMIO 1 in monomer at 110°C for 5 days
or that some hydroxylamine 4 is formed, and that it undergoes a Michael addition to the respective
monomers (as described later). Interestingly, HPLC/MS analysis did show molecular ions equivalent to
putative intermediates 9 and 10, which can be formed as a consequence of hydrogen abstraction from 8a
and 8b. The yields of compounds 9 and 10 appeared to be low (<1%). Hydrogen abstraction (under the
same reaction conditions) was faster in the case of methyl methacrylate (MMA) to predominantly give
the previously described hydrogen abstraction product 11 in 13% yield.4
3. Addition of hydroxylamines to Michael acceptors
The hydroxylamines of TMIO, TEMPO and 4-oxo-TEMPO were prepared in situ via a simple hydro-
genation over Adams catalyst (PtO2) in dichloromethane at ambient temperature and pressure. After 1–3
hours the yellow/red colouration of the nitroxide had disappeared indicating that the hydrogenation had
reached completion. The solution was filtered, evaporated to dryness, evacuated and stirred in monomer
for 3 hours at ambient temperature. The Michael adducts (Scheme 3) were all isolated in quantitative
or near-quantitative yields. When the reactions were carried out under analogous conditions but in the
presence of 1 equivalent of sodium ethoxide, the reaction reached completion in half an hour, and no
reaction was observed with styrene. It follows that these reactions are proceeding via a regular Michael
addition rather than a radical-catalyzed Michael addition.
The addition of N,N-dialkylhydroxylamines to activated double bonds was first described by Zinner
et al.,5 and more recently by Pastor and Hessell.6 The addition of these sterically hindered compounds
is probably facilitated by the α-effect; however, the corresponding Michael adducts with MMA and
methacrylonitrile were not formed. It is interesting to note that compounds 12 (the formal Michael adduct
of TMIOH and MMA) and 13, previously isolated from the reaction of isopropoxyl radicals with MMA,
may be formed by a radical-catalyzed Michael addition process.4

3676
Scheme 3.
4. Significance and conclusions
Nitroxides are finding increasing use as mediators for ‘living’ free radical polymerizations2 so it is
important to be aware of side-reactions involving nitroxides and alkenes. The bis-nitroxide compounds
8, and the hydroxylamine Michael addition products described above, also have potential as unimolecular
initiators for such ‘living’ polymerization processes. It is clear that any hydroxylamine formed during the
polymerization of acylonitrile or acrylates will undergo rapid hydrogen transfer and addition reactions.
This will slow down the polymerization process, and broaden the molecular weight distribution of the
resulting polymers.7 Although the radical addition of nitroxides to various monomers is slow at the
elevated temperatures at which such ‘living’ polymerizations generally take place, addition and hydrogen
abstraction reactions by nitroxides are significant.
References
1. Evans, R. A.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1994, 27, 7935; Evans, R. A.; Moad, G.; Rizzardo,
E.; Thang, S. H. US Patent 1997, 5665839.
2. Moad, G.; Rizzardo, E.; Solomon, D. H. Macromolecules 1982, 15, 909; Solomon, D. H.; Rizzardo, E.; Cacioli, P. US
Patent 1985, 4581429 (Chem. Abstr. 1985, 102, 221335q); Benoit, D.; Chaplinski, V.; Braslau, R.; Hawker, C. J. J. Am.
Chem. Soc. 1999, 121, 3904, and references cited therein.
3. Connolly, T. J.; Scaiano, J. C. Tetrahedron Lett. 1997, 38, 1133.
4. Busfield, W. K.; Jenkins, I. D.; Rizzardo, E.; Solomon, D. H.; Thang, S. H. J. Chem. Soc., Perkin Trans. 1 1991, 1351.
5. Zinner, G. Angew. Chem. 1959, 71, 311; Zinner, G.; Ritter, W.; Kliegel, W. Pharmazie 1965, 20, 291.
6. Pastor, D. S.; Hessell, E. T. J. Org. Chem. 1988, 53, 5776.
7. Gridnev, A. A. Macromolecules 1997, 30, 7651, and references cited therein.