Catalyzed Radical Termination in the Presence of Tellanyl Radicals

Article abstract of DOI:10.1002/chem.201703064

The decomposition of the diazo initiator dimethyl 2,2′-azobis(isobutyrate) (V-601), generating the Me₂C^.(CO₂Me) radical, affords essentially the same fraction of disproportionation and combination in media with a large range of viscosity (C₆D₆, [D₆]DMSO, and PEG 200) in the 25–100 °C range. This is in stark contrast to recent results by Yamago et al. on the same radical generated from Me₂C(TeMe)(CO₂Me) and on other X-TeR systems (X=polymer chain or unimer model; R=Me, Ph). The discrepancy is rationalized on the basis of an unprecedented RTe^.-catalyzed radical disproportionation, with support from DFT calculations and photochemicaL V-601 decomposition in the presence of Te₂Ph₂.

Full text of DOI:10.1002/chem.201703064

A Journal of
Accepted Article
Title: Catalyzed Radical Termination in the Presence of Tellanyl
Radicals
Authors: Thomas G. Ribelli, Wahidur Rahaman, Krzysztof
Matyjaszewski, and Rinaldo Poli
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201703064
Link to VoR: http://dx.doi.org/10.1002/chem.201703064
Supported by

10.1002/chem.201703064
Chemistry - A European Journal
COMMUNICATION
Catalyzed Radical Termination in the Presence of Tellanyl
Radicals
Thomas G. Ribelli,[a] S. M. Wahidur Rahaman,[b] Krzysztof Matyjaszewski*,[a] Rinaldo Poli*[b],[c]
Abstract: The decomposition of the diazo initiator dimethyl 2,2’-
azobis(isobutyrate) (V-601), generating the Me2C•(CO2Me) radical,
affords essentially the same fraction of disproportionation and
combination in media with a large range of viscosity (C6D6, DMSO-d6
and PEG 200) in the 25-100°C range. This is in stark contrast to recent
results by Yamago et al. on the same radical generated from
Me2C(TeMe)(CO2Me) and on other X-TeR systems (X = polymer
chain or unimer model; R = Me, Ph). The discrepancy is rationalized
on the basis of an unprecedented RTe•-catalyzed radical
disproportionation, with support from DFT calculations and
photochemical V-601 decomposition in the presence of Te2Ph2.
carbon-based radicals was investigated by NMR, SEC and MS to
quantify the respective contributions of Disp and Comb, see
Scheme 1 (A and B). Furthermore, the generation of radicals via
photolysis allowed these termination reactions to be investigated
under a wide variety of conditions by varying temperature and
solvent.
1. Tellurium-Based Termination Reactions
A
kdisp
hν
TeR (- RTe )
R2
R1
R2
R1
R2
R1
R2
Quantification
via NMR & GPC
n
n
R1
kcomb
The mechanism of radical termination is of fundamental
importance in polymer chemistry. Bimolecular radical termination
can occur via either disproportionation (Disp) or combination
(Comb). Disp results in two chains, one with an unsaturated
chain-end and one with a saturated one, while Comb gives a
single chain following C-C coupling.[1] The respective amounts of
Disp and Comb highly depend on the nature of the radical species.
For example, styrenes[2] and acrylonitrile[3] radicals undergo
predominately Comb while methacrylates undergo both Disp and
R = Me, Ph
R1 = COOMe, Ph
R2 = H, Me
B
O
O
O
hν
O
+
O
O
MeTe
kdisp
O
(- MeTe )
2. This work
O
∆ or hν
C
O
O
N
O
kcomb
O
(- N2)
O
O
O
O
4]
Comb.[2b, The termination of acrylates has been of intense,
N
ongoing debate with contrasting evidence in favor of either Comb
or Disp.[2a, 5]
Scheme 1. 1 (top): work using organotellurium initiators to generate carbon-
based radicals from (A) TERP macroinitiators and (B) the unimer model of the
pMMA-TeMe dormant species.[6] 2 (bottom): activation of the diazo compound
V-601 to generate methacrylate radicals in the absence of tellurium compounds.
Recently, Yamago et al. have put forth a few contributions
towards the understanding of the radical termination mechanisms
using
a tellurium-mediated radical polymerization (TERP)
system.[6] Various TERP macroinitiators, for instance poly(methyl
methacrylate) (pMMA-TeR), poly(methyl acrylate) (pMA-TeR)
and polystyrene (pSt-TeR) (R = Ph or Me) species with controlled
molecular weight (MW) and narrow MW distribution, as well as
corresponding low molar mass models (unimers), were
synthesized and used as radical species precursors. The C-Te
bonds were photolyzed, resulting in the pair of carbon- (R0·) and
tellurium-based (·TeR) radicals which then underwent
spontaneous termination processes. The ·TeR radicals ultimately
combine to form the RTe-TeR dimer, whereas the fate of the
While the Disp/Comb ratio was as expected on the basis of
previous literature reports for the pMMA• and pSt• radicals,[6a] it
was surprisingly found that acrylate radicals resulting from a pMA-
TePh macroinitiator terminate almost exclusively (99%) via
7]
Disp,[6b] in contrast with several other previous[5b,
and
subsequent[8] reports. In the most recent contribution,[6c] it was
shown that the product distribution depends on the reaction
medium, a greater viscosity resulting in an increase of the
Disp/Comb ratio. For example, 59% Disp was observed for the
methacrylate unimer model compound Me2C(TeMe)CO2Me at
relatively low viscosity (η = 1.1 mPas), whereas this amount
increased significantly to 94% at η = 84 mPas at room
temperature. Even more glaring was the use of the styrene
unimer model complex CH3(TeMe)CHPh in which case 14% Disp
was observed at low viscosity (η = 1.1 mPas) and increased
drastically to 99% at η = 84 mPas at room temperature. In order
to rationalize these surprising results, Yamago and coworkers
have put forth an “advanced collision model”, which rests on the
principle that Comb is more viscosity-sensitive than Disp, hence
a viscosity increase would result in a more significant retardation
of the Comb rate constant (kc) than of the Disp constant (kd).
[a]
[b]
Mr T. G. Ribelli, Prof. K. Matyjaszewski
Department of Chemistry
Carnegie Mellon University
4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213, USA
E-mail: km3b@andrew.cmu.edu
Dr. S. M. W. Rahaman, Prof. R. Poli
CNRS, LCC (Laboratoire de Chimie de Coordination)
Université de Toulouse, UPS, INPT
205 Route de Narbonne, BP 44099, F-31077, Toulouse Cedex 4,
France
E-mail: rinaldo.poli@lcc-touluse.fr
Prof. R. Poli
[C]
Institut Universitaire de France
1, rue Descartes, 75231 Paris Cedex 05, France
We have also been interested in the termination mechanism
of acrylates, particularly in the case when this is catalyzed by a
transition metal complex such as certain Cu-based ATRP
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/chem.201703064
Chemistry - A European Journal
COMMUNICATION
catalysts.[9] Initial studies have given strong indications that the
Cu-catalyzed radical termination (CRT) overwhelmingly leads to
Disp-type products in the ATRP of acrylates.[9b] In addition, a more
recent analysis of the termination products of acrylate radicals led
to the conclusion that pMA radicals spontaneously terminate
predominantly by Comb, in contrast to the recent report by
Yamago et al.[6b] These results, which will be reported
separately,[10] led us to hypothesize that the presence of RTe•
radicals may promote Disp. In order to verify this hypothesis, it
was necessary to first investigate the fate of radicals related to
those of the investigation by Yamago et al., but obtained in the
absence of RTe•, and then to rationalize any observed differences.
ratio, indicating the action of PhTe• as a Disp catalyst, which was
further supported by DFT calculations We propose a simpler
model, based on a viscosity effect on the radical diffusion away
from the solvent cage, to rationalize the previously reported
results.
The Me2C•(CO2Me) radicals, generated from V-601, were
allowed to terminate under conditions as close as possible to
those used by Yamago et al. for the Me2C(TeMe)CO2Me
decomposition (for details, see the SI). The reactions were carried
out in the same media, except for PEG 200 instead of PEG 400
because of better NMR spectral resolution (however, the viscosity
range of these two solvents partially overlap), and in a wider
temperature range (25-100°C). The % Disp and Comb obtained
in all our experiments are collected in the Supporting Information
(Table S1) and shown graphically in Figure 1 & S12. Examination
of these results shows quite clearly that the temperature and the
solvent nature have little effect, if any, on the relative proportions
of Disp and Comb, thus questioning the validity of the previous
proposed “advanced collision model”.[6c]
(A)
100
80
60
40
A possible way to rationalize the different outcome of the two
decomposition experiments is suggested by the analysis of the
termination reaction as a stepwise process: (i) formation of the
Me2C•(CO2Me) radical and a second partner radical within a
solvent cage, (ii) escape from the cage and (iii) termination by
Disp and/or Comb. For V-601, both isobutyryl radicals needed for
the termination events are formed within the same solvent cage,
hence bimolecular termination may occur either within the solvent
cage or after cage escape. For the tellurium-based system, on
the other hand, cage escape is necessary for bimolecular radical
termination between two isobutyryl radicals to occur. Thus, the
observed difference in product distribution may result from the
presence of side reactions between the Me2C•(CO2Me) and
tellanyl radicals within the cage.[12]
MMA-TeMe
Benzene
DMSO
V601
Benzene
DMSO
20
0
PEG 400
Best Fit
PEG 200
Best Fit
20
40
60
80
100
Temperature (C°)
(B)
100
80
60
40
20
0
MMA-TeMe
V601
Benzene
DMSO
Benzene
DMSO
x 2
RTe
+
Te2R2
R1
R2
TeR
R1
(i)
PEG 400
Best Fit
PEG 200
Best Fit
(ii)
X
X
(iii)
Disp
+
RTe
R2
R1
X
1 10
Viscosity (mPas)
100
Comb
R2
(a)
X = H or polymer chain
R = Me, Ph
RTe
+
(b)
R1 = H, Me
R2 = COOMe, Ph
H
RTe
Figure 1. Fraction of disproportionation vs. (A) temperature and (B) viscosity for
the isobutyryl radical in benzene (squares), DMSO (circles) or PEG 200/400
(triangles). Both the results obtained in this work from the V-601 diazo initiator
(filled symbols) and those reported by Yamago et al.[6c] using the MMA-TeMe
initiator (unfilled symbols) are shown for comparison. [V-601]0 = 10 mM (DMSO
& C6D6) or 100 mM (PEG 200) heated or irradiated for 24 hours. Colored figures
can be seen in Figure S12.
R1
X
+
H
RTe
R2
R1
-unsaturated
Disp product
X
saturated
Disp product
R2
Scheme 2. Proposed mechanism for C-based radical disproportionation
catalyzed by RTe•. (i) Radical pair generation; (ii) solvent cage escape; (iii)
bimolecular radical termination
We report here the use of the diazo initiator dimethyl 2,2’-
azobis(isobutyrate) (V-601) as a radical precursor to study radical
termination.[11] This compound, by either thermal or
photochemical activation, yields a radical identical to one of those
studied by Yamago et al., see Scheme 1C. We show that the
viscosity effects reported for the termination using the
Me2C(TeMe)CO2Me initiator are not observed in the absence of
tellurium. Furthermore, introduction of Te2Ph2 in the photo-
chemical V-601 decomposition in C6D6 increases the Disp/Comb
A possible side reaction is β-H abstraction[13] from the C-
based radical by RTe•, resulting in RTe-H and an alkene (R = Me,
Ph), as shown in Scheme 2, step (a). This process has been
previously shown for Cr[14] and Co[15] complexes and can also be
envisaged for any β-H-containing alkyl chains, such as pMMA,
pMA, pSt and their unimolecular models. The RTe-H intermediate
may then transfer the H atom to a second radical, yielding the
This article is protected by copyright. All rights reserved.

10.1002/chem.201703064
Chemistry - A European Journal
COMMUNICATION
saturated Disp product with regeneration of RTe• as shown in
Scheme 2 step (b). Note that the tellanyl radical produced in step
(b) may dimerize but may also induce disproportionation of
additional radicals that have already escaped from their solvent
cages.
The feasibility of this catalytic disproportionation has been
analyzed by DFT calculations on both steps (a) and (b) of the
catalytic cycle proposed in Scheme 2, using the same MeTe• and
Me2C•(CO2Me) radicals of the experimental investigation
presented in Figure 1 (see the computational details in the SI).
The DFT results are summarized in Figure 2A. All optimized
geometries and energies are collected in the SI. The unrestricted
optimization of the MeTe• + Me2C•(CO2Me) radical pair, carried
out under the broken symmetry approach (total spin S = 0) yielded
a loose van der Waals minimum (vdW1), the shortest contact
being 4.68 Å between the Te atom and one Me H atom. This
adduct is slightly stabilized (-0.8 kcal/mol) relative to the sum of
the two separate radicals on the gas-phase electronic energy
scale (∆Egas), but destabilized on the Gibbs energy scale after
Figure 2. Results of the DFT calculations for steps a and b of Scheme 2. (A) R
= Me, R1 = Me and (B) R = Ph, R1 = H. The gas-phase electronic energies
(∆Egas) and the solvation- and dispersion-corrected free energies (∆GD3,MeCN
,
bold characters in parentheses) are in kcal/mol.
corrections for dispersion and solvation in MeCN (∆GD3,MeCN
=
(+3.5 kcal/mol). In this adduct, each partner maintains its full spin
density (1.000 for MeCH•(CO2Me), of which 0.788 on the
unsaturated C atom; -1.000 for MeTe•, all on the Te atom). Step
a is exoergic by 23.0 kcal/mol from the separated MeTe• +
Me2C•(CO2Me) reagents. We failed to locate the transition state
for this step (TS1) but our attempts prove that this must be located
very near the vdW1 minimum and that the barrier is nearly non-
existent (see details in SI). Likewise, step b starts from a very
loose van der Waals minimum (vdW2, shortest contact of 5.5 Å),
slightly stabilized on the ∆Egas scale (-0.7 kcal/mol) but
destabilized on the ∆GD3,MeCN scale (+0.6 kcal/mol) relative to the
sum of the two separate species. The transformation is again
exoergic through a very low energy and relatively “early” transition
state TS2 with the Te-H bond stretched to 1.738 Å (relative to
1.676 Å in vdW2), an H···C distance of 1.852 Å, and a low
imaginary frequency of 177i cm-1. In TS2, the spin density is still
mostly localized on the C atom (0.657), though a significant
amount is already transferred to the Te atom (0.200), whereas the
H atom being transferred has a very low spin density of -0.036.
It is also noteworthy that the PSt• radical, known to lead
preferentially to Comb,[2] was also forced to dramatically shift
preference to yield a Disp/Comb ratio of 97/3 when generated
from PSt-TePh under very high viscosity conditions.[6c] We
propose that this result can more easily be rationalized by the
combination of a viscosity-dependent solvent cage escape and
the PhTe•-catalyzed Disp process, than by the “advanced collision
model” proposed by Yamago et al.[6c]
In order to substantiate the proposed tellanyl radical-
catalyzed disproportionation, the V-601 photochemical decom-
position at room temperature in C6D6 was carried out in the
presence of a tenfold excess of Ph2Te2, known to generate PhTe•
under irradiation (see details in the SI). In this case, the putative
PhTe•-catalyzed disproportionation may only occur after the V-
601-generated radical pair has escaped from the solvent cage.
Indeed, the Disp/Comb ratio increased from 50/50 in the absence
of PhTe• radicals to 68/32.
As a final remark, the Disp/Comb fraction obtained from
organotellurium initiators appears to be much more skewed in
favour of Disp for pMA than for pSt and pMMA compared to
generally accepted Disp/Comb ratios. This could be related to the
trend of RTe-alkyl bonds bond strengths (i.e. the energy
associated to the homolytic C-Te bond cleavage, step (i) in
Scheme 2). Radicals forming a weaker bond presumably have a
greater recombination barrier (Hammond principle) and may be
able to escape faster from the cage and thus be less susceptible
to undergo the catalyzed Disp process. The DFT calculations
support this view (see Table 1). For acrylate radicals, the
occurrence of transfer reactions such as backbiting must also be
carefully considered.[8, 10]
The above computations show the likelihood of the mechanism
proposed in Scheme 2 for the MMA model radical. Thus, the same
process may also take place for the other organotellurium
reagents, in particular for the controversial case of the pMA•
radical. This was further suggested by an analogous investigation
on the PhTe•/MeCH•(CO2Me) radical pair, giving similar results
(Figure 2B and further details in the SI). At this point, a reasonable
interpretation of the results published in the contribution by
Yamago and coworkers[6c] rests on the hypothesis that the solvent
viscosity affects the rate of step (ii) (solvent cage escape). A
viscosity increase slows down the radical diffusion away from the
cage, providing a bias in favour of the RTe•-catalyzed Disp. As the
solvent viscosity is lowered and thus radical escape from the cage
becomes more efficient, a greater contribution of Comb is
observed for all radicals generated from organotellurium
compounds, approaching the “true” uncatalyzed fraction of Disp
to Comb.[6] The presence of “in-cage” radical reactions has
previously been shown using alkoxy radicals.[12]
Table 1. Homolytic bond strength calculated by DFT for
organotellurium systems.
a variety of
Alkyl radical
•TeR
∆Egas, kcal/mol
∆GD3,MeCN, kcal/mol
This article is protected by copyright. All rights reserved.

10.1002/chem.201703064
Chemistry - A European Journal
COMMUNICATION
Me2C•(CO2Me)
Me2C•(CO2Me)
MeCH•(Ph)
•TeMe
•TePh
•TePh
•TePh
41.4
37.4
37.1
43.3
31.7
33.8
34.3
37.3
Keywords: Radical termination • Tellanyl radical • Catalysis • V-
601 • DFT calculations
References
MeCH•(CO2Me)
[1]
[2]
G. Moad, D. H. Solomon, The Chemistry of free Radical
Polymerization, 2nd ed., Elsevier, Amsterdam, 2005.
a) C. H. Bamford, A. D. Jenkins, Nature 1955, 176, 78-78;
b) Y. Nakamura, T. Ogihara, S. Yamago, ACS Macro
Letters 2016, 5, 248-252.
In conclusion, the V-601 diazo initiator was used to thermally
or photochemically generate isobutyryl radicals, which were
allowed to undergo bimolecular termination. The resulting
fractions of disproportionated (Disp) and combined (Comb)
products were found to remain essentially unchanged over a wide
range of temperatures and solvent viscosities, contrary to recent
reports using organotellurium initiators. In-depth DFT studies
demonstrated that a previously undiscovered RTe• catalysed
radical disproportionation reaction is viable, providing a likely
interpretation of the previously published experimental results.[6]
The proposed mechanism occurs via a β-H abstraction from the
carbon-based radical forming an unsaturated chain end and a
RTe-H intermediate. The latter species can then transfer the
hydride to a second radical forming the saturated species and
completing the catalytic cycle. At higher solvent viscosities, the
alkyl/tellanyl radical pair diffuses more slowly out of the solvent
cage and thus more time is available for the catalysed
disproportionation to occur.
[3]
[4]
[5]
C. H. Bamford, R. W. Dyson, G. C. Eastmond, Polymer
1969, 10, 885-899.
M. Buback, F. Gunzler, G. T. Russell, P. Vana,
Macromolecules 2009, 42, 652-662.
a) G. Ayrey, M. J. Humphrey, R. C. Poller, Polymer 1977,
18, 840-844; b) Z. Szablan, T. Junkers, S. P. S. Koo, T. M.
Lovestead, T. P. Davis, M. H. Stenzel, C. Barner-Kowollik,
Macromolecules 2007, 40, 6820-6833.
[6]
a) Y. Nakamura, S. Yamago, Macromolecules 2015, 48,
6450-6456; b) Y. Nakamura, R. Lee, M. L. Coote, S.
Yamago, Macromolecular Rapid Communications 2016,
37, 506-513; c) Y. Nakamura, T. Ogihara, S. Hatano, M.
Abe, S. Yamago, Chemistry-a European Journal 2017, 23,
1299-1305.
[7]
[8]
[9]
a) G. Odian, in Principles of Polymerization, John Wiley &
Sons, Inc., 2004, pp. 198-349; b) G. Ayrey, M. J.
Humphrey, R. C. Poller, Polymer 1977, 18, 840-844.
N. Ballard, S. Hamzehlou, F. Ruiperez, J. M. Asua,
Macromolecular Rapid Communications 2016, 37, 1364-
1368.
a) K. Schröder, D. Konkolewicz, R. Poli, K. Matyjaszewski,
Organometallics 2012, 31, 7994-7999; b) Y. Wang, N.
Soerensen, M. Zhong, H. Schroeder, M. Buback, K.
Matyjaszewski, Macromolecules 2013, 46, 683-691; c) H.
Schroeder, M. Buback, Macromolecules 2014, 47, 6645-
6651.
Acknowledgements
We thank the Centre National de la Recherche Scientifique
(CNRS) for support through the PICS06782 grant and through the
Laboratoire International Associé (LIA) “Laboratory of
Coordination Chemistry for Controlled Radical Polymerization”.
Support from the NSF (CHE 1707490) is also gratefully
acknowledged. This work was granted access to the HPC
resources of CINES and IDRIS under the allocation (2006 through
2017)-086343 made by GENCI (Grand Equipement National de
Calcul Intensif) and to the resources of the CICT (Centre
Interuniversitaire de Calcul de Toulouse, project CALMIP). We
thank one of the referees for a useful suggestion.
[10]
T. G. Ribelli, K. F. Augustine, M. Fantin, P. Krys, R. Poli,
K. Matyjaszewski, submitted.
M. Yoshioka, T. Otsu, Macromolecules 1992, 25, 559-562.
N. Chowdhury, A. Anoop, N. D. P. Singh, Synthesis 2012,
44, 1745-1754.
A. A. Gridnev, S. D. Ittel, Chemical Reviews 2001, 101,
3611-3659.
L. H. Tang, J. R. Norton, Macromolecules 2006, 39, 8236-
8240.
[11]
[12]
[13]
[14]
[15]
A. F. Burczyk, F. O'Driscoll, G. L. Rempel, J. Polym. Sci.,
Polym. Chem. Ed. 1984, 22, 3255-3262.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703064
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
The bimolecular termination of
(CH3)2C•CO2CH3 radicals generated
from V-601 is essentially solvent and
temperature independent, contrary to
that of the same radicals generated
from (CH3)2C(TeMe)CO2CH3; DFT
calculations and V-601 photolysis in
the presence of Te2Ph2 suggest an
unprecedented Te-catalyzed
Thomas G. Ribelli S. M. Wahidur
Rahaman Krzysztof Matyjaszewski*,
Rinaldo Poli*
Page No. – Page No.
Catalyzed Radical Termination in the
presence of Tellanyl Radicals
disproportionation to account for these
different results.
This article is protected by copyright. All rights reserved.