On the synthesis, characterization and reactivity of n-heteroaryl-boryl radicals, a new radical class based on five-membered ring ligands

Article abstract of DOI:10.1002/chem.201400197

The synthesis and physical characterization of a new class of N-heterocycle-boryl radicals is presented, based on five membered ring ligands with a N(sp²) complexation site. These pyrazole-boranes and pyrazaboles exhibit a low bond dissociation energy (BDE; B-H) and accordingly excellent hydrogen transfer properties. Most importantly, a high modulation of the BDE(B-H) by the fine tuning of the N-heterocyclic ligand was obtained in this series and could be correlated with the spin density on the boron atom of the corresponding radical. The reactivity of the latter for small molecule chemistry has been studied through the determination of several reaction rate constants corresponding to addition to alkenes and alkynes, addition to O ₂, oxidation by iodonium salts and halogen abstraction from alkyl halides. Two selected applications of N-heterocycle-boryl radicals are also proposed herein, for radical polymerization and for radical dehalogenation reactions.

Full text of DOI:10.1002/chem.201400197

DOI: 10.1002/chem.201400197
Full Paper
&
Radical Chemistry
On the Synthesis, Characterization and Reactivity of
N-Heteroaryl–Boryl Radicals, a New Radical Class Based on
Five-Membered Ring Ligands
Mohamad-Ali Tehfe,[a] Stꢀphane Schweizer,[b] Anne-Caroline Chany,[b] Cꢀdric Ysacco,[c]
Jean-Louis Clꢀment,[c] Didier Gigmes,[c] Fabrice Morlet-Savary,[a] Jean-Pierre Fouassier,[d]
Markus Neuburger,[e] Thꢀophile Tschamber,[f] Nicolas Blanchard,*[f] and Jacques Lalevꢀe*[a]
Abstract: The synthesis and physical characterization of
boron atom of the corresponding radical. The reactivity of
the latter for small molecule chemistry has been studied
through the determination of several reaction rate constants
corresponding to addition to alkenes and alkynes, addition
to O2, oxidation by iodonium salts and halogen abstraction
from alkyl halides. Two selected applications of N-heterocy-
cle–boryl radicals are also proposed herein, for radical poly-
merization and for radical dehalogenation reactions.
a new class of N-heterocycle–boryl radicals is presented,
based on five membered ring ligands with
a
N(sp2)
complexation site. These pyrazole–boranes and pyrazaboles
exhibit a low bond dissociation energy (BDE; BꢀH) and ac-
cordingly excellent hydrogen transfer properties. Most im-
portantly, a high modulation of the BDE(BꢀH) by the fine
tuning of the N-heterocyclic ligand was obtained in this
series and could be correlated with the spin density on the
Introduction
phine-boranes could be used in organic synthesis although
their high bond dissociation energy (BDE; BꢀH) was a serious
concern for applications in radical chain reactions. It should
also be noted that their behavior as polarity reversal catalysts
has been evidenced, thus clearly expanding their synthetic in-
terest.[3] Interestingly, it was demonstrated through theoretical
approaches that the BDE(BꢀH) can be tuned by an appropriate
design of the Ligand (L)[4a,b] and that synergistic effects be-
tween the ligand and the substituents on the boron atom had
a pronounced effect on the electronic structure of the boryl
and also on its reactivity.[4c]
Boryl radicals are versatile nucleophilic species that have re-
cently triggered intense studies in organic synthesis.[1] These
radicals are classically produced through a hydrogen abstrac-
tion reaction between a borane and a suitable hydrogen ac-
ceptor. Historically,[2] Roberts demonstrated in a series of ele-
gant reports that boryl radicals derived from amine- or phos-
[a] Dr. M.-A. Tehfe, Dr. F. Morlet-Savary, Prof. J. Lalevꢀe
Institut de Science des Matꢀriaux de Mulhouse IS2M, UMR CNRS 7361
UHA, 15, rue Jean Starcky, 68057 Mulhouse Cedex (France)
E-mail: jacques.lalevee@uha.fr
A recent breakthrough has led to the development of a new
class of boryl radicals based on N-heterocyclic carbene–bor-
anes (NHC–boranes) that can efficiently overcome this BDE(Bꢀ
H) limitation, the BDE(BꢀH) being brought down to about 80–
90 kcalmolꢀ1 compared to 92–105 kcalmolꢀ1 for amine- or
phosphine-boranes.[1] NHC–boranes are now used in elegant
chemical reactions such as radical reductions of xanthates and
related derivatives where they serve as radical hydrogen atom
donors in Barton–McCombie reactions.[1] Based on quantum
mechanical calculations, the 80–90 kcalmolꢀ1 range of BDE
values can also be covered with some borane complexes con-
taining five or six membered ring heterocycles with a N(sp2)
complexation site for the borane moiety.[5] Recently, N-hetero-
aryl-boranes such as 1a (Figure 1), characterized by such
a BDE(BꢀH) feature, have been reported by our laboratories
and the formation as well as the reactivity of the associated
boryl radicals were investigated.[6,7] The kinetic data on boryl
radicals remain, however, rather scarce.[8]
[b] Dr. S. Schweizer, Dr. A.-C. Chany
Universitꢀ de Haute-Alsace, ENSCMu
Laboratoire de Chimie Organique et Bioorganique, EA 4566
3 rue Alfred Werner, 68093 Mulhouse Cedex (France)
[c] Dr. C. Ysacco, Dr. J.-L. Clꢀment, Dr. D. Gigmes
Aix-Marseille Universitꢀ, CNRS, Institut de Chimie Radicalaire
UMR CNRS 7273, 13397 Marseille, Cedex 20 (France)
[d] Prof. J.-P. Fouassier
Formerly, ENSCMu, 3 rue Alfred Werner
68093 Mulhouse Cedex (France)
[e] Dr. M. Neuburger
Department of Chemistry, University of Basel
Spitalstrasse 51, 4056 Basel (Switzerland)
[f] Dr. T. Tschamber, Dr. N. Blanchard
Universitꢀ de Strasbourg, ECPM
Laboratoire de Chimie Molꢀculaire UMR CNRS 7509
25 rue Becquerel, 67087 Strasbourg (France)
E-mail: n.blanchard@unistra.fr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201400197.
Chem. Eur. J. 2014, 20, 5054 – 5063
5054
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
merization (Scheme 1, Eq. 2, conditions B), in line with previous
reports.[16] The structures of these four-coordinate boron spe-
cies were confirmed by X-ray diffraction studies (CCDC 981830
for 2a; 981831 for 2b, BꢀH 1.10–1.15 ꢂ). As shown in Figure 2,
pyrazabole 2a possesses a classical B2N4 boat conformation of
C2v symmetry.[17] Finally, heating a toluene solution of 1c at
858C for 72 h delivered 2a as the major product and also the
nine-membered ring polyborazane 3a as a minor component
of the reaction, in
a concentration-dependent fashion
(Scheme 1, Eq. 3). At 0.05m in toluene, only 4% of 3a was ob-
tained whereas at 0.2m, 16% was found and at 0.7m, up to
38% could be isolated. The structure of 3a was determined by
X-ray diffraction studies (CCDC 981833; BꢀH 1.10–1.12 ꢂ,
Figure 2).
Formation and reactivity of the N-heterocycle–boryl radicals
The boryl radicals are generated here in laser flash photolysis
experiments through an approach based on a hydrogen ab-
straction reaction between the selected boranes and either
Figure 1. N-heteroaryl–borane complexes 1a–e, pyrazaboles 2a, b and poly-
borazane 3a.
C
C
1) the tert-butoxyl radical tBu-O (reactions r1 and 2), the tBu-O
radicals being easily generated at 355 nm by the direct cleav-
In the present paper, we report the synthesis, chemical
mechanisms studies, and synthetic applications of 1b–e, 2a, b,
and 3a (Figure 1) that are the precursors of new boryl radical
species based on five membered ring heterocycles such as pyr-
azoles, oxazoles, or triazoles. A clear modulation of
the BDE(BꢀH) was expected and was accordingly
demonstrated through DFT calculations. Kinetic data
for the formation of these boryl radicals and for their
interaction with alkenes, alkynes, O2, iodonium salts,
and alkyl halides are also provided for the first time.
Structure/reactivity relationships of these new radi-
cals, both in hydrogen abstraction process and in
small molecule radical chemistry, are also discussed.
Finally, two selected applications of these N-heterocy-
cle–boranes are presented in thermal and photoin-
duced radical polymerization and also in radical de-
halogenation reactions.
age of di-tert-butylperoxide[9b] or 2) the benzophenone triplet
3
state BP (reactions r3 and 4). In the case 1), the tert-butoxyl
radical does not absorb at l>300 nm (Figure 3), the observed
intense absorption can be ascribed to the generated boryl (see
Results and Discussion
Synthesis of N-heterocycle borane complexes
N-heteroaryl borane complexes 1a,[6,7] b,[14] c,[15] d,
and e were easily synthesized from the correspond-
ing N-heteroarenes 4a–e and borane·THF complex in
THF at room temperature (Scheme 1, Eq. 1). These
white and non-hygroscopic powders proved to be
shelf stable and could be kept for months open to
air. Interestingly, it was found that pyrazabole 2a
could be obtained in good yield (64%) on a large
scale (50 mmol) when the reaction was run with an
excess of borane·THF complex in THF at room tem-
perature for 15 h (Scheme 1, Eq. 2, conditions A). On
the other hand, access to pyrazabole 2b required
Scheme 1. Synthesis of N-heteroaryl–borane complexes 1a–e, pyrazaboles 2a, b and
heating at reflux for an efficient dehydrogenative di- polyborazane 3a.
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org 5055
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
absorption gives a direct access
C
to the tBu-O /L!BH3 hydrogen
abstraction rate constant (kH);
3
the kH of the BP/L!BH3 interac-
tions are also accessible through
a
quenching
(Table 1). Interestingly,
rate constants are
experiment
high
found
(>108 mꢀ1 sꢀ1) in line with the
quite low calculated BDE(BꢀH)
values for the corresponding
boranes (Table 1). The rate con-
C
stants found for the tBuO /
3
borane and BP/borane reactions
Figure 2. X-ray structures of pyrazabole 2a (left; CCDC 981830) and polyborazane 3a (right; CCDC 981833).
are quite similar. This suggests
Figure 4 and the Supporting Information). The boryl radicals
C
C
associated with 1 and 2 will be referred here as 1 and 2 for
the sake of clarity.
The boryl absorption spectra spread over the visible range
with absorption maxima at 430, 650, 640, 400, and 600 nm for
C
C
C
C
C
1b –e and 2b , respectively. For 1b and 2b , the transient ab-
sorptions are relatively weak and the maximum absorption
wavelength is not very well defined. The rise time of the boryl
Figure 3. A) Kinetic traces recorded at 650 nm for the formation and decay
C
C C
Figure 4. Transient absorption spectra of the radicals A) 1c , B) 1e , and
of the boryl radical (1c ); the increase of [1c] is indicated by an arrow. B) Ki-
C
netic traces recorded at 650 nm for the interaction of the boryl radical with
methylacrylate (MA) in acetonitrile/di-tert-butylperoxyde; the increase of
[MA] is indicated by an arrow. Insert: the associated Stern–Volmer treatment.
Experiments performed under an argon atmosphere.
C) 2b in acetonitrile/di-tert-butylperoxyde. The rise of the boryl radical
occurs within 500 ns. Absorption recorded at different times after the laser
excitation from 0.5 ms to 19.8 ms by steps of 1.94 ms (A), 0.5 ms to 8 ms by
steps of 420 ns (B); absorption recorded at t=1.5 ms (C).
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5056
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion from the NꢀH bond can be ruled out as 1) the calculated
BDE(NꢀH) is higher than BDE(BꢀH) by about 16 kcalmolꢀ1 and
2) no nitrogen centered radical is detected.
Table 1. Rate constants (kH) for the formation of the N-heterocycle–
borane radicals in acetonitrile/di-tert-butylperoxide at 208C; oxidation po-
tential Eox and calculated bond dissociation energy (BDE) of the boranes.
[a]
[c]
[d]
Eox
BDE (BꢀH)[b]
kH
kH
C
tBu-O-O-tBu ! 2 tBu-O ðhnÞ
ðr1Þ
ðr2Þ
ðr3Þ
ðr4Þ
1b
1c
1d
1e
2b
>1
0.5
70.8
85.8
84.8
79.7
87.5
15
35
20
20[b]
20[b]
86 (1.0)[e]
C
C
tBu-O þ L ! BH3 ! tBu-OH þ L ! BH2
>1
>1
25 (0.8)[f]
1
1
3
BP ! BP ðhnÞ and BP ! BP
[a] In V (vs. SCE); [b] in kcalmolꢀ1; [c] (tBu-O·, see r2), 107 mꢀ1
see r4), 107 mꢀ1
measured according to the classical procedure from ref. [9a]; it also corre-
sponds to the boryl radical quantum yield; [f] a higher error bar is associ-
ated with these data due to noisy transients.
s
ꢀ1; [d] (3BP,
3
s
ꢀ1; [e] the ketyl radical quantum yield (in acetonitrile)
C
C
BP þ L ! BH3 ! BP-H þ L ! BH2
Through the direct observation of the boryl radicals in laser
flash photolysis experiments, a convenient access to their reac-
tivity became possible. Indeed, the classical analysis of the
time dependence of the boryl decay provided the rate con-
stants for their onward reactions with unsaturated compounds
(alkenes or alkynes), oxygen, an oxidizing agent (diphenyliodo-
nium hexafluorophosphate) and an alkyl halide (Figure 3,
Table 2). Since the kinetics associated with the boryl radicals
are strongly affected by O2 (indeed, the boryl radical/O2 inter-
that the benzophenone triplet state generated after light ab-
sorption behaves like an alkoxyl radical leading to ketyl and
boryl radicals. In this series of compounds, the selected N-het-
erocycle substituent can affect the BDE(BꢀH) by about 17 kcal
molꢀ1 that is, 70.8 kcal for 1b versus 87.5 kcalmolꢀ1 for 2b.
This is in full agreement with a recent theoretical work which
has shown that the complexation of the borane with a large
number of small heterocycles can lead to BDE(BꢀH) in the 80–
90 kcalmolꢀ1 range.[5]
action rate constants are close to the diffusion limit that is, for
1c , k=2.3ꢃ10 mꢀ1 sꢀ1), the different quenching experiments
with other molecules were all carried out under argon.
The boryl/alkene interaction rate constants correspond to an
addition of the boryl to the alkene double bond (kadd), even in
9
C
ESR-spin trapping (ESR–ST) is a worthwhile approach to
characterize the free radicals generated.[18,19] Typical results for
ESR–ST are displayed in Figure 5. For the phenyl-N-tert-butylni-
the case of acrylates. Indeed, the low interaction rate constant
(<5ꢃ105 mꢀ1 s ) between for example, 1c and a carbonyl
ꢀ1
C
C
C
trone (PBN9 adduct of 1c , a third hyperfine splitting constant
group (acetone) compared to that of 1c to methylacrylate
(HFS) aB due to the boron nuclei must be included to repro-
duce the experimental spectra. The aB value (4.1 G) is in excel-
lent agreement with previous results on other boryls: 3.5<
aB <4.7 G.[2d,f] In the case of 1c, the hydrogen abstraction reac-
(MA) evidences that the addition reactions to MA does not
occur onto the carbonyl. The kadds for the addition of the
C
C
C
C
C
boryls to MA decrease in the series 1c ~1d >2b >1b , 1e
and span over at least two orders of magnitude, thereby high-
lighting the effect of the ligand on the boryl reactivity. The kadd
C
C
of 1c and d to MA is higher than that found for the triethyl-
C
C
amine boryl derived radical
108 mꢀ1 ꢀ1),[8a] this latter species being characterized by the
highest reported rate constant together with the 4-pyrrolidino-
6
(Et3N–BH2 ;
kadd =1.3ꢃ
s
C
C
pyridine borane derived radical 7 (4-pyrrolidinopyridine–BH2 ;
k
add =1.4ꢃ108 mꢀ1 sꢀ1).[6] Interestingly, 1c exhibits also a high
C
reactivity towards an alkyne (3.2ꢃ107 mꢀ1 sꢀ1 vs. 3.1ꢃ
106 mꢀ1 sꢀ1 when considering 7 ). These results support a very
C
high reactivity of the new proposed boryl radicals.
The kadd values for the addition of a given boryl both to an
electron rich alkene (ethylvinylether; EVE) and a less electron
rich alkene (vinylacetate; VA) are almost similar and below
5
5
0.5ꢃ10 (1c , d ) or 5ꢃ10 mꢀ1 sꢀ1 (1e , 2b ). As found in previ-
ous works on other N-heteroarylboryls or NHC-boryls,[6,7,8b] the
fast addition to an electron poor alkene (MA) combined with
a slow addition to an electron rich alkene (VA and EVE) sug-
gests that the boryls investigated here exhibit a high nucleo-
philic character. The transition state (TS) calculated for the ad-
C
C
C
C
C
dition of 1c to MA is depicted in the Supporting Information.
The barrier for the addition process is low (2.8 kJmolꢀ1 at
UB3LYP/6-31+G* level) in full agreement with the measured
high rate constants. Interestingly, a high charge transfer from
the radical to the alkene is found in the transition state (TS;
Figure 5. ESR spectrum observed in spin trapping experiments for the
C
adduct of the boryl radical 1c (using PBN) in tert-butylbenzene/di-tert-butyl-
peroxide (fitting parameters aN =16.3 G; aH =3.95 G; aB =4.1 G; aH =0.6 G).
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5057
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[6]
C
radicals is still enhanced compared to that of 7 .
C
C
C
Table 2. Interaction rate constants characterizing the reactivity of 1b –e , and 2b
with various additives in acetonitrile/di-tert-butylperoxide solvent (at 208C).
C
The 1c /C3H7Br interaction is almost 160 times lower
C
that of 1 /C3H7I.
C
C
C
C
C
1c
k[a]
1b
k[a]
1e
k[a]
1d
k[a]
2b
k[a]
Molecular orbital calculations show that these N-
heteroaryl–boryl radicals exhibit a planar p-type elec-
tronic structure (Figure 6). A significant spin delocali-
zation from the boron atom into the ligand is found
for the SOMO (see Figure 6 and the Supporting Infor-
mation). A similar behavior has been recently found
for NHC–boranes as well as N-heteroaryl–boranes
which have a quite similar BDE(BꢀH) range.[1,6,7,8b]
The spin densities at the boron atom (SB) were calcu-
lated as 0.215; 0.39; 0.355; 0.26; 0.5, 0.509 and 0.36
methylacrylate (MA)
methylmethacrylate (MMA)
vinylacetate (VA)
ethylvinylether (EVE)
4-phenyl-1-butyne
Ph2I+
39
26
<0.05
<0.05
3.2
<0.03
<1
40
2.5
<0.5
<0.5
<0.05
<0.05
<0.5
<0.5
250
120
250
49
0.8
C3H7I
40
<0.05
<0.1
1-C6H13Br
0.25
[a] 107 mꢀ1 sꢀ1
.
C
C
for 1b–e and 2a, b , respectively (Figure 7; calcula-
C
tions at UB3LYP/6-31+G* level). For radicals 1e and
C
0.198e¯) which is fully consistent with the high nucleophilic
b , most of the spin density is located over the heterocycle
C
character of 1c . This stems a general property that is, boryl
ring and, for at least one atom of the heterocycle, higher than
radicals derived from amine-boranes, NHC-boranes and N-het-
eroaryl boranes being already characterized by a nucleophilic
behavior.
C
1c being characterized by an enhanced reactivity compared
to other boryls previously proposed, its oxidation by iodonium
ꢀ
salt (Ph2I+PF6 ) as well as its halogen abstraction property
(from RꢀI or RꢀBr) were examined in detail. These elementary
reactions are important in small molecule radical chemistry.
Since they exhibit low reduction potentials (ca. ꢀ0.2 V), iodoni-
um salts are often used to oxidize radicals.[20] Interestingly, the
rate constant close to the diffusion limit (k>109 mꢀ1 sꢀ1
;
Table 2) supports a very favorable oxidation process. This is in
full agreement with the very low calculated ionization poten-
C
tial of 1c (5.08 eV) as well as its nucleophilic character found
Figure 7. BDE(BꢀH) versus the spin density (SB) on the boron atom for radi-
cals 1a–e, 2a, b and 3a.
above for the addition process. The rate constants (kI) for the
C
C
1c (and 1d )/C3H7I interaction are very high (Table 2). As in the
case of the addition to MA (see above), the reactivity of these
for the boron atom. Therefore, it can be expected that the re-
activity of these radicals will be different from that of “true”
boryl radicals which possesses the highest spin density on B
(SB). Such a behavior was previously studied through theoreti-
cal approaches.[4c,5] Qualitatively, for the structures investigated
here, the BDE(BꢀH) is correlated with the B atom spin density
(SB) in the radical (Figure 7), that is, upon an increase of the de-
localization from the boron atom into the ligand, both SB and
BDE(BꢀH) are decreased. The data previously obtained for
other ligated boranes are also added in Figure 7.[7] Roughly, for
BDE(BꢀH)<80 kcalmolꢀ1, the highest spin density will not
likely reside on the boron atom. This can also explain both the
C
C
lowest reactivity of 1b and e in the reactions with MA and
C3H7I and the excellent performance of 2a in radical dehaloge-
nation reactions (vide infra).
N-heterocycle-boryl radicals as new polymerization initiating
structures
Some selected applications of the novel N-heterocycle-borane/
additive couples as sources of boryl radicals were investigated,
especially in thermal or photoinitiated polymerization process-
es. A quite similar efficiency was found for 1c and d (see the
C
Figure 6. Singly occupied molecular orbital (SOMO) of 1c . The boron atom
is indicated by an arrow.
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5058
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Supporting Information); for the sake of clarity the results ob-
tained for 1c will be mainly presented here.
A thermal redox system based on the N-heterocycle–
borane/Ph2I+ combination
The combination of a N-heterocycle–borane with an oxidation
agent (here an iodonium salt Ph2I+) appears as an efficient ini-
tiating system for the free radical polymerization (FRP) of tri-
methylolpropane triacrylate (TMPTA) even at RT under air
(Figure 8) that is, these polymerization reactions are highly
exothermic. Using 3,5-dimethyl-1H-pyrazole/Ph2I+, no polymer-
ization is observed (Figure 8C, curve c) evidencing the funda-
mental role of the BH3 complexation. The efficiency of these
boranes decreases in the 1c@1e>1b, 2b series (Figure S4 in
the Supporting Information). Compound 1c is characterized by
the lowest oxidation potentials (~0.5 V/SCE; see Table 1) sug-
gesting that a redox process can be involved (r5). In ESR–ST ex-
+
C
periments, Ph are clearly observed in the 1c/Ph2I system in
full agreement with a reduction of the iodonium salt (PBN
adduct: aN =14.4 G; aH =2.2 G; reference values in ref. [6]).
Such a behavior has been also found for N-heteroaryl-boranes
and the expected mechanism is probably similar to that re-
cently proposed.[6] The Ph radical can 1) initiate the polymeri-
C
zation as in ref. [21] (r5) or 2) abstract an hydrogen atom from
1c leading to the formation of a boryl radical (r6) which is fur-
ther oxidized by Ph2I+ (r7); then, a chain process with re-for-
C
mation of Ph occurs
Cþ
C
1 c þ Ph2Iþ ! 1 c þ Ph þ PhI
ðr5Þ
ðr6Þ
ðr7Þ
C
C
Ph þ L-BH3 ! PhH þ L-BH2
þ
þ
C
C
L-BH2 þ Ph2I ! Ph þ PhI þ L-BH2
This redox system can be also successfully applied to the
cationic polymerization of an epoxy monomer (here: (3,4-
epoxycyclohexane)methyl
3,4-epoxycyclohexylcarboxylate;
EPOX–Uvacure 1500 from Cytec). Using 1c/Ph2I+ (2%/2%
w/w), a complete polymerization is obtained at RT within 12 h.
Figure 8. A) Photopolymerization profiles of TMPTA in laminate (irradiation
Xe–Hg lamp) using various photoinitiating systems (1%/1% w/w or 1%
w/w): a) BP/EDB; b) BP/1c; c) BP/1e; d) BP/2b; e) BP alone; f) 1c alone.
B) Photopolymerization profiles of TMPTA under air (irradiation Xe–Hg lamp)
with for various photoinitiating systems (1%/1% w/w or 1% w/w): a) BP/
EDB; b) BP/1c; c) BP/1e; d) BP/2b; e) BP alone; f) 1c alone. C) Sample tem-
perature versus time for the thermal polymerization of TMPTA initiated at RT
by a) 1c/Ph2I+ (1%/1% w/w); b) 1c/Ph2I+ (0.5%/0.5% w/w); c) 3,5-dimethyl-
1H-pyrazole 4c/Ph2I+ (1%/1% w/w); t=0 corresponds to the addition of
the borane complex to the reaction medium.
+
+
C
The cationic species 1c (r5) and the borenium L-BH2 (r7)
are the initiating species of the ring opening process.
A similar cationic polymerization can be also obtained using
a different oxidation agent (AgSbF6 instead of Ph2I+) as the re-
duction of a silver salt by a pyridine–borane was already re-
ported.[22] With 1c/AgSbF6 (1%/1% w/w), a complete polymeri-
zation of EPOX is still obtained in 12 h at RT. The formation of
the polyether network is easily observed at 1080 cmꢀ1 (Fig-
ure S5 in the Supporting Information). The formation of Ag0 is
well evidenced by UV–visible spectroscopy (Figure S5 in the
Supporting Information): a new absorption band centered at
about 400 nm appears in the film in the course of the polymer-
ization process. This can be safely ascribed to the surface plas-
mon resonance (SPR) absorption band of the silver nanoparti-
cles (NPs) as it has been previously reported in[23] that unpro-
tected Ag NPs in the <15 nm size domain display a SPR band
at 390 nm. By analogy with the results obtained above for the
1c/Ph2I+ system, this shows the initiating ability of the cationic
species generated. The oxidation of boryl radicals by silver salt
(in r10 by analogy to r7) can also be expected since these spe-
cies are characterized by a low ionization potential. An overall
proposed mechanism can be proposed in (r8–10):
Cþ
1 c þ AgSbF6 ! 1 c =SbF6ꢀ þ Ag0NP
ðr8Þ
ðr9Þ
Cþ
C
ꢀ
1 c =SbF6 ! L-BH2 þ HSbF6
L-BH2 þ AgSbF6 ! L-BH2þ þ SbF6ꢀ þ Ag0
C
ðr10Þ
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5059
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nol, that is certainly formed via the b-elimination of the radical
intermediate. When pyrazole·BH3 1c or pyrazabole 2a were
used, the amount of this side product was dramatically low-
ered and synthetically useful yields were obtained (78–82%).
We next explored the scope of this dehalogenation reaction
using pyrazabole 2a as the most promising boryl radical pre-
cursor (Scheme 3). When the primary 5c, secondary 5a, b or
Boryl radicals as initiating structures in photoinitiating
systems
The photoinduced FRP profiles of TMPTA using benzophenone
(BP) as the photoinitiator in both the absence and the pres-
ence of the proposed boranes as co-initiators are shown in
Figure 8. The addition of the borane complexes 1b–e to BP
drastically reduced the inhibition time and significantly en-
hanced the polymerization rates and the final conversions (Fig-
ure 8A, curves b–d and in the Supporting Information). This
behavior is much more marked for polymerizations conducted
under air (Figure 8B, curves b–d) where, without the borane,
no polymerization is observed. Therefore, the present boryl
C
radicals (1b–e ) formed in (r4) are attractive initiating species,
the ketyl radicals being unable to initiate a polymerization.[21]
This is in agreement with the high addition rate constants of
boryl radicals to an acrylate unit reported above (e.g., MA in
Table 2). The ability of 1b to act as a co-initiator is very poor in
line with its low reactivity in the addition reaction onto acryl-
ate (see Table 2). Compound 1c appears as the most promising
co-initiator, even better than ethyl-dimethylaminobenzoate
(EDB; which is a reference amine co-initiator).[20c] It also exhibits
the best performance for the FRP in laminate in agreement
with a boryl radical quantum yield close to 1.0 and the highest
addition rate constant to MA (Table 2).
N-heteroaryl-boryl radicals in dehalogenation reactions
Besides the physical chemistry and photopolymerization stud-
ies of N-heteroaryl borane complexes, we investigated their
use in synthetic organic chemistry and more precisely in deha-
logenation reactions of a model substrate, a-iodo ester 5a, in
the presence of an initiator (10 mol%) in toluene at 858C
(Scheme 2). A rapid screen of initiators such as oxygen, di-tert-
butylperoxide, triethyl borane, and azobisisobutyronitrile
(AIBN) revealed that the latter was convenient to use and led
to reproducible results. A very rapid conversion of 5a to the
corresponding reduced species was observed in the case
DMAP·BH3 1a and N-phenylbenzotriazole·BH3 1b (less than
10 min). However, only moderate yields were obtained (62–
67%) due to the formation of a side product, p-methoxyphe-
Scheme 3. Scope of the radical dehalogenation reaction of a-haloester 5a–j
using pyrazabole 2a.
tertiary 5d a-iodo esters were used, good yields of the corre-
sponding dehalogenated products were obtained (91, 82, 58
and 74%, respectively). Various esters could also be employed
in this transformation, such as benzyl esters 5e (54%), f (64%),
the bromine atom being untouched by the boryl radical in this
case and p-methoxybenzyl ester 5g (55%). Alkyl esters were
also tolerated in this reaction, as shown by the clean dehaloge-
nation of 5h in 65% yield. Less efficient reaction partners were
a-iodo Weinreb amides (not shown) or a-bromo esters. As an
example, only 25% of the desired reduced product was ob-
tained when bromo dimethylmalonate 5i was used. Eventually,
as could be expected from a polarity point of view, the dehalo-
genation of dodecyl iodide 5j by the electron-rich boryl radical
C
2a was not efficient and only 9% conversion of the corre-
sponding dodecane was obtained (Scheme 3).
Experimentally, a white precipitate was formed almost im-
mediately in the deep red reaction mixture when the dehalo-
genation was conducted with pyrazabole 2a. After filtration
and recrystallization of this stable solid, X-ray crystallographic
studies demonstrated that this precipitate was the boronium
salt[24] derivative 7a (CCDC 981832; BꢀH 1.056 ꢂ), with iodine
as the counter-anion (Scheme 4 and Figure 9). We surmised
that this borocation 7a is formed spontaneously from the
iodo-pyrazabole 2a–I, a hypothesis that was supported by
direct iodolysis studies of 2a.[25] After extensive optimization, it
Scheme 2. Initial screen of the boryl radical precursors for the dehalogena-
tion reaction of a-iodoester 5a.
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5060
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Synthesis of the boronium salt 7a, byproduct of the dehalogena-
tion reaction (Eq. 1) and its de novo synthesis (Eq. 2).
Scheme 5. Control experiments of the dehalogenation reaction using a radi-
cal scavenger (Eq. 1) and a radical clock process (Eq. 2).
and iodolysis of the lithium enolate intermediate. Heating of
5k at 858C in toluene (in the absence of N-heteroaryl borane
complexes and radical initiators) led to the complete recovery
of the starting material. Eventually, combination of pyrazabole
2a (0.5 equivalent; or DMAP·BH3 1a, not shown) and AIBN
(10 mol%) led to a very fast reaction, the d-iodo-a,b-unsaturat-
ed ester 8 being isolated as the only product in 80% yield in
only 10 min. The formation of this compound could be ex-
plained by the ring-opening of an a-cyclopropyl radical fol-
lowed by the trapping of the primary radical by an iodine
atom, either from the starting material 5k or an intermediate
such as 2a–I (Scheme 4). We also checked that the boronium
salt 7a was not an efficient reaction partner, since only traces
of 8 were obtained under these reaction conditions (not
shown). All together, these results clearly point towards a radi-
cal dehalogenation reaction, in full agreement with the physi-
cal studies conducted in parallel (vide supra).
Figure 9. X-ray structure of diboracation 7a (CCDC 981832).
was found that iodolysis of 2a was best conducted in ethyl
acetate at 858C, the boronium salt 7a thus being obtained in
74% yield on a multigram scale. It should also be noted that
the iodine counter-anion could be exchanged for a hexafluoro-
phosphate (7b, 82%), thus further improving its stability. The
reactivity of this diboracation as an initiator in polymerization
reactions is currently under investigations.
Conclusion
In this article, a new class of radicals derived from N-hetero-
aryl–boranes such as pyrazole–boranes and pyrazaboles was
presented. The synthesis, characterization, and the associated
reactivity of these radicals were investigated. Interestingly,
a high modulation of the BDE(BꢀH) through the choice of the
N-heteroaryl moiety was noted (BDE in the 70.8–91.3 kcalmolꢀ1
range) which could be correlated with the spin density on the
boron atom in the corresponding radical. High rate constants
of boryl radical formation through a hydrogen abstraction in
To gain insights into the radical nature of these dehalogena-
tion reactions, several control experiments were conducted
(Scheme 5). In a first set of reactions, TEMPO (1 equivalent)
was used as a radical inhibitor of the dehalogenation reaction
of 5a and c (Eq. 1). No conversion was observed, giving a first
hint of a radical reaction. Next, the radical clock[26] substrate 5k
(Eq. 2) featuring a a-iodo-a-cyclopropyl motif was prepared in
two steps from commercially available cyclopropylacetic acid
by esterification with p-methoxyphenol followed by lithiation
C
tBuO /borane couples were determined, in agreement with
these low BDEs(BꢀH). The implications on the reactivity of
these boryls are important since the addition rate constants
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5061
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a QST2 approach. The TSs were frequency checked. The charge
transfer from the radical to the alkene was evaluated from the Mul-
liken analysis of the TS.
span over at least two orders of magnitude. Relevant applica-
tions in polymer science and synthetic organic chemistry were
also provided, outlining the high lability of these BꢀH bonds
as well as the excellent reactivity of the corresponding boryl
radicals.
Photopolymerization experiments
For film polymerization experiments, a photoinitiating system was
dissolved into a bulk formulation based on trimethylolpropane tri-
acrylate (TMPTA from Cytec).[12] The formulations sandwiched be-
tween two polypropylene films (20 mm thick) were deposited on
a BaF2 pellet and irradiated with polychromatic light (Xe–Hg lamp,
Hamamatsu, L8252, 150 W). The evolution of the double bond con-
tent was followed by real-time FTIR spectroscopy (Nexus 870, Nico-
let) at room temperature. These experiments were carried out
both in laminate and under air.[12d] Compared to the N-heteroaryl–
boranes previously investigated (such as 1a), an excellent solubility
was found for 1b–e and 2a,b in classical organic solvents. It
should be noted that a slightly turbid solution (albeit stable) was
noted in TMPTA for 2b.
Experimental Section
Benzophenone (BP) was used as model photoinitiator (from Al-
drich). Ethyl dimethylaminobenzoate (EDB; Esacure EDB from Lam-
berti) was chosen as a reference amine co-initiator. Methylacrylate
(MA), vinylacetate (VA), ethylvinylether (EVE), diphenyl iodonium
hexafluorophosphate (Ph2I+), iodopropane, 4-phenyl-1-butyne
were obtained from Aldrich with the highest purity available.
CCDC 981830 (2a), 981831 (2b), 981833 (3a) and 981832 (7a) con-
tain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Laser flash photolysis experiments
Thermal polymerization experiments
Nanosecond laser flash photolysis (LFP) experiments were carried
out using a Q-switched nanosecond Nd/YAG laser (lexc =355 nm,
9 ns pulses; energy reduced down to 10 mJ, from Minilite Continu-
um) and an analyzing system consisting of a ceramic xenon lamp,
a monochromator, a fast photomultiplier and a transient digitizer
(Luzchem LFP 212).[9]
The polymerization was carried out in pill-box for a sample of
1 g. The progress of the exothermic polymerization was fol-
lowed by monitoring the sample temperature using a method
presented in detail in.[13] The addition of the organoboranes
into the formulation corresponds to time t=0 s.
ESR spin trapping experiments
Acknowledgements
The radicals were generated under the polychromatic light expo-
sure of a Xe–Hg lamp (Hamamatsu, L8252, 150 W) of N-heterocy-
cle–borane in di-tert-butyl peroxide/tert-butylbenzene (a small
amount of acetonitrile is also added to ensure a good solubility).
The generated radicals were trapped by phenyl-N-tert-butylnitrone
(PBN). The ESR spectra simulations were carried out with the
WINSIM software.[10] The procedure used has been presented in
detail in ref. [6]. The ESR spectrum simulation includes 10B as men-
tioned in ref. [2d] and [2f].
The authors thank the CNRS, theUniversity of Haute-Alsace, the
University of Strasbourg, the Agence Nationale de la Recherche
(ANR BLAN-0802, ANR SILICIUM) and the Institut Universitaire
de France for financial support.
Keywords: boryl radicals · co-initiator · dehalogenation ·
N-heterocycle–boranes · polymerization · radical chemistry
[1] a) S. H. Ueng, M. Makhlouf Brahmi, E. Derat, L. Fensterbank, E. Lacote,
M. Malacria, D. P. Curran, J. Am. Chem. Soc. 2008, 130, 10082–10083;
b) S.-H. Ueng, A. Solovyev, X. Yuan, S. J. Geib, L. Fensterbank, E. Lacꢄte,
M. Malacria, M. Newcomb, J. C. Walton, D. P. Curran, J. Am. Chem. Soc.
2009, 131, 11256–11262; c) J. C. Walton, Angew. Chem. 2009, 121,
1754–1756; Angew. Chem. Int. Ed. 2009, 48, 1726–1728; d) J. C. Walton,
M. Makhlouf Brahmi, L. Fensterbank, E. Lacꢄte, M. Malacria, Q. Chu, S.-H.
Ueng, A. Solovyev, D. P. Curran, J. Am. Chem. Soc. 2010, 132, 2350–
2358; e) A. Solovyev, S.-H. Ueng, J. Monot, L. Fensterbank, M. Malacria,
E. Lacꢄte, D. P. Curran, Org. Lett. 2010, 12, 2998–3001; f) S.-H. Ueng, L.
Fensterbank, E. Lacꢄte, M. Malacria, D. P. Curran, Org. Lett. 2010, 12,
3002–3005; g) X. Pan, E. Lacꢄte, J. Lalevꢀe, D. P. Curran, J. Am. Chem.
Soc. 2012, 134, 5669–5674; h) Q. L. Chu, M. Makhlouf Brahmi, A. Solo-
vyev, S.-H. Ueng, D. P. Curran, M. Malacria, L. Fensterbank, E. Lacꢄte,
Chem. Eur. J. 2009, 15, 12937–12940; i) D. M. Lindsay, D. McArthur,
Chem. Commun. 2010, 46, 2474–2476; j) M. Horn, H. Mayr, E. Lacꢄte, E.
Merling, J. Deaner, S. Wells, T. McFadden, D. P. Curran, Org. Lett. 2012,
14, 82–85; k) J. Lalevꢀe, S. Telitel, M. A. Tehfe, J. P. Fouassier, D. P.
Curran, E. Lacꢄte, Angew. Chem. 2012, 124, 6060–6063; Angew. Chem.
Int. Ed. 2012, 51, 5958–5961.
Redox potentials
The redox potentials (Eox vs. SCE) were measured in acetonitrile by
cyclic voltammetry with tetrabutyl-ammonium hexafluorophos-
phate 0.1m as a supporting electrolyte (Voltalab 06-Radiometer;
the working electrode was a platinium disk and the reference a sa-
turated calomel electrode-SCE). Ferrocene was used as a reference
(Eox = +0.44 V/SCE).
DFT calculations
All the calculations were performed by using the hybrid functional
B3LYP from the Gaussian 03 suite of program.[11] Reactants and
products were fully optimized at the B3LYP/6-31+G* level
(checked for imaginary frequencies) for the calculations of the dif-
ferent reaction enthalpies (DHr). The selected DFT method exhibits
a relatively good performance for the BDE calculations (see refs. [6]
and [7]). The calculated BDEs are also very similar to those extract-
ed from the calculations using isodesmic reactions (see ref. [7]).
Adiabatic ionization potentials (IP) were calculated from the
energy difference between the relaxed neutral molecule and the
relaxed ion (B3LYP/6-31+G* level). The transition states (TS) were
calculated with DFT methods (UB3LYP/6-31+G* level) using
[2] a) J. R. M. Giles, B. P. Roberts, J. Chem. Soc. Perkin Trans. 2 1983, 743–
755; b) J. A. Baban, B. P. Roberts, J. Chem. Soc. Chem. Commun. 1983,
1224–1226; c) J. A. Baban, V. P. J. Marti, B. P. Roberts, Tetrahedron Lett.
1985, 26, 1349–1352; d) J. A. Baban, V. P. J. Marti, B. P. Roberts, J. Chem.
Soc. Perkin Trans. 2 1985, 1723–1733; e) V. P. J. Marti, B. P. Roberts, J.
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5062
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Chem. Soc. Perkin Trans. 2 1986, 1613–1621; f) J. A. Baban, B. P. Roberts,
J. Chem. Soc. Perkin Trans. 2 1986, 1607–1611; g) I. G. Green, B. P. Rob-
erts, J. Chem. Soc. Perkin Trans. 2 1986, 1597–1606; h) J. N. Kirwan, B. P.
Roberts, J. Chem. Soc. Perkin Trans. 2 1989, 539–550; i) K. M. Johnson,
J. N. Kirwan, B. P. Roberts, J. Chem. Soc. Perkin Trans. 2 1990, 1125–
1132; j) M. Lucarini, G. F. Pedulli, L. Valgimigli, J. Org. Chem. 1996, 61,
4309–4313; k) M. Lucarini, G. F. Pedulli, L. Valgimigli, J. Org. Chem. 1996,
61, 1161–1164; l) J. Lalevꢀe, M. A. Tehfe, X. Allonas, J. P. Fouassier, Mac-
romolecules 2008, 41, 9057–9062.
[14] W. Liao, Y. Chen, Y. Liu, H. Duan, J. L. Petersen, X. Shi, Chem. Commun.
2009, 6436–6438.
[15] a) G. Agrifoglio, Inorg. Chim. Acta 1992, 197, 159–162; b) E. Frauendor-
fer, G. Agrifoglio, Inorg. Chem. 1982, 21, 4122–4125.
[16] a) S. Trofimenko, J. Am. Chem. Soc. 1967, 89, 3165–3170; b) S. Trofimen-
ko, J. Am. Chem. Soc. 1967, 89, 4948–4952; c) W. J. Layton, K. Niedenzu,
P. M. Niedenzu, S. Trofimenko, Inorg. Chem. 1985, 24, 1454–1457; For
a specific NMR study, see: d) W. J. Layton, K. Niedenzu, S. L. Smith, Z.
Anorg. Allg. Chem. 1982, 495, 52–64.
[3] a) J. A. Baban, B. P. Roberts, J. Chem. Soc. Perkin Trans. 2 1988, 1195–
1200; b) B. P. Roberts, Chem. Soc. Rev. 1999, 28, 25–35.
[4] a) P. R. Rablen, J. F. Hartwig, J. Am. Chem. Soc. 1996, 118, 4648–4653;
b) P. R. Rablen, J. Am. Chem. Soc. 1997, 119, 8350–8360; c) D. Lu, C. Wu,
P. Li, Chem. Eur. J. 2014, DOI: 10.1002/chem.201303705.
[5] J. Hioe, A. Karton, J. M. L. Martin, H. Zipse, Chem. Eur. J. 2010, 16, 6861–
6865.
[6] J. Lalevꢀe, N. Blanchard, M. A. Tehfe, A.-C. Chany, J. P. Fouassier, Chem.
Eur. J. 2010, 16, 12920–12927.
[7] J. Lalevꢀe, N. Blanchard, A.-C. Chany, M. A. Tehfe, X. Allonas, J. P. Fouassi-
er, J. Phys. Org. Chem. 2009, 22, 986–993.
[8] a) B. Sheeller, K. U. Ingold, J. Chem. Soc. Perkin Trans. 2 2001, 480–486;
b) M. A. Tehfe, M. Makhlouf Brahmi, J. P. Fouassier, D. P. Curran, M. Mal-
acria, L. Fensterbank, E. Lacꢄte, J. Lalevꢀe, Macromolecules 2010, 43,
2261–2267.
[9] a) J. Lalevꢀe, N. Blanchard, M. A. Tehfe, M. Peter, F. Morlet-Savary, D.
Gigmes, J. P. Fouassier, Polym. Chem. 2011, 2, 1986–1991; b) J. Lalevꢀe,
B. Graff, X. Allonas, J. P. Fouassier, J. Phys. Chem. A 2007, 111, 6991–
6998.
[10] D. R. Duling, J. Magn. Reson. Ser. B 1994, 104, 105–112.
[11] a) Gaussian 03, Revision B-2, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Mont-
gomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M.
Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Och-
terski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, P. Salvador, J. J.
Dannenberg, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B.
Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-
Gordon, E. S. Replogle, and J. A. Pople, Gaussian, Inc., Pittsburgh PA,
2003; b) J. B. Foresman, A Frisch, in Exploring Chemistry with Electronic
Structure Methods. Second Edition, Gaussian. Inc. 1996.
[12] a) J. Lalevꢀe, X. Allonas, S. Jradi, J. P. Fouassier, Macromolecules 2006, 39,
1872–1879; b) J. Lalevꢀe, L. Zadoina, X. Allonas, J. P. Fouassier, J. Polym.
Sci. Part A 2007, 45, 2494–2502; c) J. Lalevꢀe, M. El-Roz, X. Allonas, J. P.
Fouassier, J. Polym. Sci. Part A 2008, 46, 2008–2014; d) M. El-Roz, J. Lale-
vee, X. Allonas, J. P. Fouassier, Macromolecules 2009, 42, 8725–8732.
[13] R. Souane, M. A. Tehfe, J. Lalevꢀe, D. Gigmes, J. P. Fouassier, Macromol.
Chem. Phys. 2010, 211, 1441–1445.
[17] For a discussion of the conformations of pyrazaboles, see: C. P. Brock, K.
Niedenzu, E. Hanecker, H. Noth, Acta Crystallogr. Sect. C 1985, 41, 1458–
1463. The X-ray structure of pyrazabole 2b is presented in the Support-
ing Information.
[18] P. Tordo, Electron Spin Resonance, Vol. 16 (Eds.: N. M. Atherton, M. J.
Davies, B. C. Gilbert), RSC, Cambridge, 1998.
[19] A. Criqui, J. Lalevꢀe, X. Allonas, J. P. Fouassier, Macromol. Chem. Phys.
2008, 209, 2223–2231.
[20] a) J. V. Crivello, Photoinitiators for Free Radical, Cationic and Anionic Pho-
topolymerization, 2nd ed. (Ed.: G. Bradley), Wiley, New York, 1998;
b) J. V. Crivello, Ring-Opening Polymerization (Ed.: D. J. Brunelle), Hanser,
Munich, 1993; c) J. P. Fouassier, Photoinitiation Photopolymerization and
Photocuring, Hanser, Munich, 1995; d) M. A. Tehfe, J. Lalevꢀe, D. Gigmes,
J. P. Fouassier, Macromolecules 2010, 43, 1364–1370; e) J. P. Fouassier, J.
Lalevꢀe, Photoinitiators for Polymer Synthesis-Scope, Reactivity, and Effi-
ciency, Wiley-VCH, Weinheim, 2012.
[21] J. Lalevꢀe, X. Allonas, J. P. Fouassier, J. Phys. Chem. A 2004, 108, 4326–
4334.
[22] M. D. Taylor, L. R. Grant, C. A. Sands, J. Am. Chem. Soc. 1955, 77, 1506–
1507.
[23] a) A. Taleb, C. Petit, M. P. Pileni, J. Phys. Chem. B 1998, 102, 2214–2220;
b) S. Link, M. A. El-Sayed, Annu. Rev. Phys. Chem. 2003, 54, 331–366; c) I.
Pastoriza-Santos, L. M. Liz-Marzꢅn, Pure Appl. Chem. 2000, 72, 83–90;
d) D. L. Van Hyning, W. G. Klemperer, C. F. Zukoski, Langmuir 2001, 17,
3120–3127; e) S. He, J. Yao, P. Jiang, D. Shi, H. Zhang, S. Xie, S. Pang, H.
Gao, Langmuir 2001, 17, 1571–1575.
[24] a) O. P. Shitov, S. L. Ioffe, V. A. Tartakovskii, S. S. Novikov, Russ. Chem. Rev.
1970, 39, 905–922; b) W. E. Piers, S. C. Bourke, K. D. Conroy, Angew.
Chem. 2005, 117, 5142–5163; Angew. Chem. Int. Ed. 2005, 44, 5016–
5036; c) For the first report of such a diboracation, see: D. C. Bradley,
M. B. Hursthouse, J. Newton, N. P. C. Walker, J. Chem. Soc. Chem.
Commun. 1984, 188–190.
[25] For related investigations, see: a) G. E. Ryschkewitsch, J. Am. Chem. Soc.
1967, 89, 3145–3148; b) G. E. Ryschkewitsch, W. W. Lochmaier, J. Am.
Chem. Soc. 1968, 90, 6260–6261.
[26] a) M. Newcomb, Tetrahedron 1993, 49, 1151–1176; b) D. Griller, K. U.
Ingold, Acc. Chem. Res. 1980, 13, 317–323.
Received: January 17, 2014
Published online on March 13, 2014
Chem. Eur. J. 2014, 20, 5054 – 5063
www.chemeurj.org
5063
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim