Homolytic, Heterolytic, Mesolytic - As You Like It: Steering the Cleavage of a HC(sp3)?C(sp3)H Bond in Bis(1H-2,1-benzazaborole) Derivatives

Article abstract of DOI:10.1002/chem.201602698

A set of (3,3′)-bis(1-Ph-2-R-1H-2,1-benzazaborole) compounds, in which R=tBu (Bab-tBu)₂, R=Dipp (Bab-Dipp)₂or R=tBu and Dipp (Bab-Dipp)(Bab-tBu), was synthesized and fully characterized using¹H,¹¹B,¹³C, and¹⁵N NMR spectroscopy as well as single-crystal X-ray diffraction analysis. The central HC(sp³)?C(sp³)H bond with restricted rotation at the junction of both 1H-2,1-benzazaborole rings displayed an intriguing reactivity. It was demonstrated that this bond is easily mesolytically cleaved using alkali metals to form the respective aromatic 1Ph-2R-1H-2,1-benzazaborolyl anions M⁺(THF)_n(Bab-tBu)^?(M=Li, Na, K) and K⁺(THF)_n(Bab-Dipp)^?. Furthermore, the central HC(sp³)?C(sp³)H bond of bis(1H-2,1-benzazaborole)s is also homolytically cleaved either by heating or photochemical means, giving corresponding 1Ph-2R-1H-2,1-benzazaborolyl radicals (Bab-tBu)^.and (Bab-Dipp)^., which rapidly self-terminate. Nevertheless, their formation was unambiguously established by NMR analysis of the reaction mixtures containing products of the self-termination of the radicals after heating or irradiation. (Bab-Dipp)^.radical was also characterized using EPR spectroscopy. Importantly, it turned out that the essentially non-polarized HC(sp³)?C(sp³)H bond in (Bab-tBu)₂is also cleaved heterolytically with 2 equiv of MeLi, giving the mixture of Li⁺(SOL)_n(Bab-tBu)^?(SOL=THF or Et₂O) and lithium methyl-substituted borate complex Li⁺(SOL)_n(Bab-tBu-Me)^?in a diastereoselective fashion.

Full text of DOI:10.1002/chem.201602698

DOI: 10.1002/chem.201602698
Full Paper
&
CÀC Bond Cleavage
Homolytic, Heterolytic, Mesolytic - As You Like It: Steering the
Cleavage of a HC(sp3)ÀC(sp3)H Bond in Bis(1H-2,1-benzazaborole)
Derivatives
[a]
[b]
[c]
[a]
[a]
˚
Martin Hejda, Antonín Lycka, Tomµs Mikysek, Roman Jambor, Ales Ruzicka,
Jaromír Vinklµrek,[a] Claudia Wilfer,[d] Alexander Hoffmann,[d] Sonja Herres-Pawlis,*[d] and
Libor Dostµl*[a]
Abstract: A set of (3,3’)-bis(1-Ph-2-R-1H-2,1-benzazaborole)
compounds, in which R=tBu (Bab-tBu)2, R=Dipp (Bab-
Dipp)2 or R=tBu and Dipp (Bab-Dipp)(Bab-tBu), was syn-
thesized and fully characterized using 1H, 11B, 13C, and
15N NMR spectroscopy as well as single-crystal X-ray diffrac-
tion analysis. The central HC(sp3)ÀC(sp3)H bond with restrict-
ed rotation at the junction of both 1H-2,1-benzazaborole
rings displayed an intriguing reactivity. It was demonstrated
that this bond is easily mesolytically cleaved using alkali
metals to form the respective aromatic 1Ph-2R-1H-2,1-benza-
zaborolyl anions M+(THF)n(Bab-tBu)À (M=Li, Na, K) and K+
(THF)n(Bab-Dipp)À. Furthermore, the central HC(sp3)ÀC(sp3)H
bond of bis(1H-2,1-benzazaborole)s is also homolytically
cleaved either by heating or photochemical means, giving
corresponding 1Ph-2R-1H-2,1-benzazaborolyl radicals (Bab-
C
C
tBu) and (Bab-Dipp) , which rapidly self-terminate. Never-
theless, their formation was unambiguously established by
NMR analysis of the reaction mixtures containing products
of the self-termination of the radicals after heating or irradia-
C
tion. (Bab-Dipp) radical was also characterized using EPR
spectroscopy. Importantly, it turned out that the essentially
non-polarized HC(sp3)ÀC(sp3)H bond in (Bab-tBu)2 is also
cleaved heterolytically with 2 equiv of MeLi, giving the mix-
ture of Li+(SOL)n(Bab-tBu)À (SOL=THF or Et2O) and lithium
methyl-substituted borate complex Li+(SOL)n(Bab-tBu-Me)À
in a diastereoselective fashion.
Introduction
the structural and fundamental point of view, they generally
behave as spectator ligands.[1,2] Recently, a new approach ap-
peared, based on the substitution of relatively chemically inert
and redox-innocent amino moieties by corresponding imino
donor groups.[3] It is obvious that this pendant HC=N group
exhibits a pronounced tendency to undergo further chemical
transformations upon coordination to the central element,
a property typical for so-called actor ligands. For instance, they
are prone to the attack of sufficiently strong nucleophiles
(Scheme 1A).[4] On a similar basis, the saturation of the HC=N
moiety is quite easily achieved using various element hydrides
that are subject to a spontaneous hydrogen shift as intermedi-
ates (Scheme 1B).[5] In both cases, various types of five-mem-
bered heterocycles are formed that would hardly be accessible
otherwise.[6,7] The reduction of these main group element
(N,)C,N-chelates also led to the isolation of unprecedented de-
rivatives in many cases.[8] A special class of these reactions con-
stitutes those in which the particular ligand acts in a non-inno-
cent fashion. These processes are then accompanied by CÀC
coupling reactions within the ligand backbone either directly
at the carbon atom of the HC=N moiety (Scheme 1C),[9a–e] or in
the special case of the annulated central ring.[9f] It was also
demonstrated that in few cases further reduction of these CÀC
coupled species resulted in the formation of unprecedented
heterocyclic aromatic anions (Scheme 1C, D).[9a,e,8d,e,f]
The utilization of various types of monoanionic C,N- or N,C,N-
chelating ligands, enabling the closure of a five-membered
chelating ring (built up as (bis)ortho-functionalized aryls) is
ubiquitous in the chemistry of main group elements.[1] These li-
gands contain a pendant amino donor group in most cases.
Despite the fact that they facilitated the isolation of a plethora
of interesting and unprecedented compounds, at least from
˚
[a] M. Hejda, Dr. R. Jambor, Prof. A. Ruzicka, Dr. J. Vinklµrek, Dr. L. Dostµl
Department of General and Inorganic Chemistry
Faculty of Chemical Technology, University of Pardubice
Studentskµ 573, CZ-532 10, Pardubice (Czech Republic)
E-mail: libor.dostal@upce.cz
[b] Prof. A. Lycka
Research Institute for Organic Syntheses
Rybitví 296, CZ-533 54, Pardubice (Czech Republic)
[c] Dr. T. Mikysek
Department of Analytical Chemistry, Faculty of Chemical Technology
University of Pardubice
Studentskµ 573, CZ-532 10, Pardubice (Czech Republic)
[d] Dr. C. Wilfer, Dr. A. Hoffmann, Prof. S. Herres-Pawlis
Institute for Inorganic Chemistry, RWTH Aachen University
Landoltweg 1, 52074 Aachen (Germany)
E-mail: sonja.herresec-pawlis@ac.rwth-aachen.de
Supporting information including all necessary experimental details for this
article can be found under http://dx.doi.org/10.1002/chem.201602698.
Chem. Eur. J. 2016, 22, 15340 –15349
15340
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
compounds were independently characterized by H, 13C, 11B,
and 15N NMR spectroscopy. Especially the signals of mutually
coupled methine HCÀN groups unambiguously confirmed the
proposed structures (see details in Supporting Information;
Table S1). Regarding the formation of rac-RR/SS-(Bab-Dipp)2
and meso-(Bab-Dipp)2, the in situ formation of the potassium
salt of (Bab-Dipp)À, which further reacts with the respective
C,N-chelated chloroborane, is not ruled out as we have previ-
ously discussed for tBu-substituted analogues (Scheme 2B).[9a]
Scheme 1. Discussed transformations of C,N- or N,C,N-chelated compounds.
This work is focused on the closer inspection of the synthe-
sis, structure, and reactivity of one of the above mentioned
CÀC bonded heterocyclic systems, that is, bis(1H-2,1-benzaza-
borole) derivatives. It is demonstrated how the non-polarized
central HC(sp3)ÀC(sp3)H bond is with a surprising ease cleaved
not only chemically (using strong reducing agents such as
alkali metals) in a mesolytic fashion, but also thermally and
photochemically in a homolytic fashion. Importantly, it was es-
tablished that this bond is able to undergo, under certain reac-
tion conditions, a heterolytical scission as well, which makes
this central HC(sp3)ÀC(sp3)H bond quite extraordinary. Al-
though the cleavage of CÀC bonds has been extensively stud-
ied for a long time, it is still a challenging area. Homolytic
cleavage of nonpolar CÀC bonds is a fairly common mecha-
nism, but isolation and/or characterization of unstable radical
species formed during this these reactions are often diffi-
cult.[10a] Similarly, heterolytic cleavage of CÀX bonds (where X is
a heteroatom such as halogen, nitrogen, etc.) is the foundation
for many organic reactions; however, similar cleavage of CÀC
bonds is still very rare.[10b] Maslak et al. described an extraordi-
nary system in which both mechanisms (homolytic and hetero-
lytic) of CÀC bond cleavage are possible. He even introduced
a third mechanism, the so-called mesolytic cleavage.[10c,d] Thus,
this work describes another interesting example of a system
(i.e., bis(1H-2,1-benzazaborole) where the CÀC bond can be
cleaved in all these three ways depending on the reaction con-
ditions.
Scheme 2. Synthesis of coupled (3,3’)-bis(1H-2,1-benzazaborole) derivatives.
This approach (Scheme 2C) was applied for the preparation
of unsymmetrically substituted derivatives rac-RR/SS-(Bab-
Dipp)(Bab-tBu) and rac-RS/SR-(Bab-Dipp)(Bab-tBu). Both
products were again separated and characterized by 1H, 11B,
13C, and 15N NMR spectroscopy (Table S1). Molecular structures
of new (3,3’)-bis(1H-2,1-benzazaborole)s were unambiguously
confirmed using single-crystal X-ray diffraction analysis
(Figure 1). All structures contain two mutually coupled 1H-2,1-
benzazaborole rings. The central HC(sp3)ÀC(sp3)H bond lengths
from 1.559(2) to 1.594(3) Š are longer in comparison to stan-
dard C(sp3)ÀC(sp3) (1.54 Š) bonds.[11] The BÀN bond lengths
within the 1H-2,1-benzazaborole rings [in the range of
1.405(3)–1.421(3) Š] reflect strong p(N)!p(B) interactions (cf.
Srcov(NÀB)=1.56 Š vs. Srcov(N=B)=1.38 Š[11]).
Results and Discussion
Syntheses and structures of (3,3’)-bis(1-Ph-2-R-1H-2,1-benza-
zaborole) compounds
The
reduction
of
the
C,N-chelated
chloroborane
Importantly, all stereoisomers of the respective compounds
form very stable conformers and some of them may even be
regarded as atropisomers due to the greatly restricted rotation
around the central HC(sp3)ÀC(sp3)H bond (Figure 2 and the
Supporting Information for details).[12] This phenomenon was
L(C=NDipp)BPhCl [L(C=NDipp)=o-(CH=N-2,6-iPr2C6H3)C6H4]
with potassium in toluene gave a mixture of (3,3’)-bis(1-Ph-2-
Dip-1H-2,1-benzazaborole) compounds rac-RR/SS-(Bab-Dipp)2
and meso-(Bab-Dipp)2 (Scheme 2A and Figure S1 in the Sup-
porting Information), similar to rac-RR/SS-(Bab-tBu)2 and
meso-(Bab-tBu)2 reported recently.[9a] After separation, both
1
studied by variable-temperature (VT) H NMR spectroscopy of
meso-(Bab-Dipp)2 and meso-(Bab-tBu)2.[13] In the latter case,
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15341
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Newman projection of the greatly hindered rotation (highlighted
by the black arrows) of enantiomeric pairs (+sc and -sc) for meso-(Bab-
tBu)2 and meso-(Bab-Dipp)2. Dashed arrows represent the hypothetic situa-
tion, in which one enantiomer of meso- (e.g. +sc) turns into to the second
enantiomer (-sc) merely by rotation around the central HC(sp3)ÀC(sp3)H
bond and vice versa.
servation, calculated parameters of the rotation around the
HC(sp3)ÀC(sp3)H bond for meso-(Bab-tBu)2 extracted from
DNMR3 simulations (Figure S3, Supporting Information) are as
follows: kr =kc =405 sÀ1
,
DG°361K =17.0 kcalmolÀ1, DH° =
13.6 kcalmolÀ1, DS° =À9.4 calKÀ1 molÀ1. In contrast, no coales-
cence could be achieved in the case of meso-(Bab-Dipp)2 in
the accessible temperature range (up to 373 K), but this is evi-
dently a consequence of the presence of bulky Dipp groups.
These results, in combination with the fact that both meso-
(Bab-tBu)2 and meso-(Bab-Dipp)2 crystallize in the centrosym-
metric space groups [as a racemate of two enantiomeric forms
(+sc and Àsc) of these stereoisomers], indicate that both
mesoforms exist as racemates of two enantiomers [in the case
of meso-(Bab-Dipp)2, as a racemate of two enantiomeric atro-
pisomers] as a consequence of a hindered rotation around the
central HC(sp3)ÀC(sp3)H bond.[9a] To gain similar activation rota-
tion barriers for all six bis(1H-2,1-benzazaborole) derivatives,
plots of potential energy versus the torsion angle on the
HC(sp3)ÀC(sp3)H bond junction were calculated. The according
energy values were obtained by density functional theory cal-
culations using the Gaussian 09 program suite.[14] From these
calculations, it is evident that the value derived for meso-(Bab-
1
tBu)2 by H VT-NMR (17.0 kcalmolÀ1) is in very good agreement
with the theoretical value (19.0 kcalmolÀ1). All remaining
bis(1H-2,1-benzazaborole) derivatives revealed higher values of
activation rotation barriers (ranging from 20–35 kcalmolÀ1; see
Figures S6–S11 in the Supporting Information).
Regarding the photophysical properties, all bis(1H-2,1-benza-
zaborole) compounds showed two strong absorptions be-
tween approximately 240 and 270 nm in UV/Vis absorption
spectra (Figure S14, Table S7). This is an important property of
these compounds; see the further discussion for details.
Figure 1. Molecular structures of C(sp3)ÀC(sp3) coupled bis(1H-2,1-benzaza-
borole) derivatives. Hydrogen atoms were omitted except for those of the
methine carbons in HC(3)ÀC(3’)H.
indeed the spectra between the slow- and fast-exchange limit
could be obtained and mathematically processed (Figures S2
and S3).
Synthesis and characterization of 1H-2,1-benzazaborolyl
anions
[9a]
A particular area of interest represents two singlets corre-
sponding to the bridging CH groups at 5.33 and 5.79 ppm at
295 K, and their coalescence occurs at 361 K. Based on this ob-
Both rac-RR/SS-(Bab-tBu)2 and meso-(Bab-tBu)2 are readily
reduced in THF by two equivalents of an alkali metal leading
to the cleavage of the central HC(sp3)ÀC(sp3)H bond and for-
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15342
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mation of corresponding alkali metal salts of 1Ph-2tBu-1H-2,1-
benzazaborolyl anions, that is, M+(THF)n(Bab-tBu)À (M=Li, Na,
K,[9a] Scheme 3). Attempts to isolate CÀC coupled radical anion
olyl anions, the structure of (Bab-Dipp)À was used for time de-
pendent (TD)-DFT calculations, considering that the influence
of the counter cation is believed to be negligible. TD-DFT cal-
culations for the (Bab-Dipp)À anion show two intensive transi-
tions at 410 and 451 nm, which correspond to the experimen-
tally observed transitions at 394 and 504 nm (Figure S13,
Table S2). These transitions are all related to excitations starting
from the HOMO (Figure S13), which is the p-orbital of the aro-
matic benzazaborolyl ring. For the 410 nm feature, the excita-
tion goes into the LUMO+2, which lies more on the Dipp
group. For the 451 nm feature, LUMO+1 and LUMO serve as
accepting orbitals, with the LUMO being Dipp group-based
and the LUMO+1 being totally delocalized.
·À
M+(THF)n(Bab-tBu)2 by reaction with one equivalent of an
alkali metal led only to the mixture of the corresponding 1H-
2,1-benzazaborolyl alkali salt and starting compounds, indicat-
ing that this species is probably only an elusive intermediate
(see further discussion). All alkali metal salts were isolated as
air- and moisture-sensitive solids that are almost insoluble in
aromatic and aliphatic solvents, but readily soluble in THF
(Scheme 3). They were characterized using 1H, 11B, 13C,
7
15N NMR, and in case of Li+(THF)n(Bab-tBu)À, also by Li NMR
spectroscopy showing closely related structures of all com-
pounds in solution (Table S1, Supporting Information). The sig-
To gain additional insight into the formation of 1H-2,1-ben-
zazaborolyl anions by the reduction of bis(1H-2,1-benzazabor-
ole) compounds, a CV (cyclic voltammogram) study on rac-RR/
SS-(Bab-tBu)2 (or meso-(Bab-tBu)2) was undertaken. Unfortu-
nately, no signal was detected in the case of Dipp-substituted
bis(1H-2,1-benzazaborole) variants, despite very high concen-
trations in the voltammetric cell (see details in the Supporting
Information). These CVs revealed one irreversible and one
quasi-reversible process in each case,[18a] (scan rates of 20–
1000 mVsÀ1) indicating that the reduction goes through two
discrete one-electron transfers. This finding supports the mech-
anism shown in Scheme 3 that probably follows the so-called
ECE[18b] electrochemical mechanism. In the first irreversible pro-
cess of this proposed mechanism [at Ep =À1.65 V vs. saturated
calomel electrode (SCE); all following potentials are also given
1
nals of the CHÀN moiety in H and 13C NMR spectra are signifi-
cantly low-field shifted [d(1H) in the range of 6.13–6.27 ppm
and d(13C) of 95.4–96.6 ppm, Table S1] pointing to an aromatic
character of these ring systems (see below for the molecular
orbital description).[15] The 11B NMR spectra revealed one signal
in each case [d(11B) in the range 23.3–23.4 ppm, Table S1],
which is slightly high-field shifted in comparison to closely re-
lated 1,2-azaborolyl lithium salts [d(11B)=30–35 ppm].[16] The
7Li NMR chemical shift (À1.4 ppm) of Li+(THF)n(Bab-tBu)À indi-
cates the presence of a solvent-separated ion pair (SSIP) in so-
lution.[17] All NMR data prove similarity of M+(THF)n(Bab-tBu)À
in solution regardless of the alkali metal used. Therefore, only
the potassium salt of the sterically more crowded system, that
is, K+(THF)n(Bab-Dipp)À, was synthesized for comparison.
1
·À
Again the H, 11B, 13C, and 15N NMR spectra proved the analo-
vs. SCE], an elusive radical anion (Bab-tBu)2 is formed that is
gous structure of K+(THF)n(Bab-Dipp)À (Table S1).
immediately cleaved to give anion (Bab-tBu)À and radical
The molecular structure of K+(THF)n(Bab-tBu)À with a poly-
meric chain structure, has been described by us recently.[9a] Un-
fortunately, all of our numerous attempts to grow single crys-
tals of any other analogue suitable for X-ray diffraction analysis
failed. UV/Vis absorption spectra of M+(THF)n(Bab-tBu)À
(where M=Li, Na, K[9a]) and K+(THF)n(Bab-Dipp)À were ac-
quired (Figure S13 and Table S2 in the Supporting Information).
To gain further insight into the structure of 1H-2,1-benzazabor-
(Bab-tBu) , the latter being reduced in a second quasi-reversi-
C
ble[18a] step to another equivalent of anion (Bab-tBu)À (at E0F =
À1.93 V, peak separation of 144 mV at 100 mVsÀ1, Figure 3). It
must be noted that this second quasi-reversible step is super-
imposed on to the exponentially growing background because
the electrochemical window was cut-off by the sample itself
(this is an intrinsic characteristic of the studied compound be-
cause the background measurement was allowed to reach
more negative potential in pure THF, up to À2.6 V). CV curves
were also acquired for parent chloroborane L(C=NtBu)BPhCl
[where L(C=NtBu)=o-(CH=NtBu)C6H4]. This measurement
showed even three redox processes. The first irreversible pro-
cess at Ep =À1.44 V in the cathodic direction was tentatively
assigned to a one-electron reduction to the corresponding rad-
C
ical (Bab-tBu) with elimination of a chloride anion (Figure 3).
The second quasi-reversible process (at E0F =À1.93 V) corre-
C
sponds to the reduction of the radical (Bab-tBu) to the anion
(Bab-tBu)À [see the CV of the corresponding bis(1H-2,1-benza-
zaborole)]. The origin of the third irreversible process in the CV
of L(C=NtBu)BPhCl at Ep =À2.54 V (denoted by an asterisk in
Figure 3) is still unclear and importantly, this redox process was
not observed in the CVs of rac-RR/SS-(Bab-tBu)2 or meso-(Bab-
tBu)2, indicating that it is not important for the reduction of
these bis(1H-2,1-benzazaborole) compounds. Importantly, a visi-
ble shoulder at Ep =À1.65 V is also found in the CV curve of
L(C=NtBu)BPhCl, that becomes even slightly more intense at
lower scan rates. This finding could be rationalized by in situ
Scheme 3. Plausible mechanism of formation of related BabÀ alkali metal
salts during the reduction of symmetrically substituted bis(1H-2,1-benzaza-
borole) derivatives (based on cyclic voltammetry, see Figures 3 and S16).
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15343
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
formation of rac-RR/SS-(Bab-tBu)2 and/or meso-(Bab-tBu)2
C
from electrochemically generated (Bab-tBu) radicals directly in
the voltammetric cell (they are indeed prone to self-termina-
tion; see further discussion). that are then reduced as de-
scribed before (Figure 3 and Figure S16 in the Supporting In-
formation). Considering that the CV of L(C=NDipp)BPhCl is
closely related to that of its tBu analogue, showing again three
processes (Figure S15), it seems reasonable to consider a similar
reduction mechanism also in the case of Dipp-substituted de-
rivatives. The first process occurring at Ep =À1.46 V in the
cathodic direction corresponds to the formation of radical
0F
C
(Bab-Dipp) (Figure S15), followed by further reduction at E =
À1.80 V (peak separation of 236 mV at 100 mVsÀ1), leading to
anion (Bab-Dipp)À (the chemical meaning of the third redox
process at exactly the same redox potential as in the tBu-sub-
stituted derivative with Ep =À2.54 V is unclear). Notably, the
C
comparison of values of redox potentials E0F
À
=
(Bab-tBu) / Bab-tBu
C
À1.93 V with EpL(C=NDipp)BPhCl/(BabÀDipp) =À1.46 V indicates that
the former may be viewed as a stronger reducing agent. Con-
sequently, preparation of unsymmetrically substituted bis(1H-
2,1-benzazaborole) derivatives by the reaction of K+
(THF)n(Bab-tBu)À and L(C=NDipp)BPhCl should probably be
considered as a nucleophilic addition accompanied by an inter-
nal redox reaction, rather than a simple nucleophilic addition
of
a
(Bab-tBu)À anion to the imino C=N bond of
L(C=NDipp)BPhCl (Scheme 2). However, we have recently dis-
covered that simple nucleophiles such as anilides and carb-
anions are able to react with this C=N bond in the structure of
the chelating ligand merely through nucleophilic attack.[4a,b]
Conclusions regarding the mechanism of reduction of
bis(1H-2,1-benzazaborole) compounds are also supported by
theoretical data, which can be found in the Supporting Infor-
mation (Tables S3–S5).
Figure 3. Cyclic voltammograms of parent L(C=NtBu)BPhCl (top) and sub-
sidiary rac-RR/SS-(Bab-tBu)2 (bottom) in THF on a GC working electrode vs.
SCE, 0.1m nBu4N+PF6À as electrolyte. Scan rate 100 mVsÀ1. * denotes an un-
known redox process.
(Bab-Dipp)2 again providing a mixture of both symmetric and
non-symmetric
bis(1H-2,1-benzazaborole)
compounds
Thermal HC(sp3)ÀC(sp3)H bond cleavage in bis(1H-2,1-benza-
(Table S7, Figure S17).
zaborole) compounds
It is important to note, that final products rac-RR/SS-(Bab-
Dipp)(Bab-tBu), rac-RR/SS-(Bab-tBu)2, and rac-RR/SS-(Bab-
Dipp)2 are stable, even when heated up to 1608C, thereby
confirming that RR/SS- stereoisomers are thermodynamically
the most stable. These findings are also supported by density
functional energy calculations, which showed an energetic sta-
bilization of these compounds in comparison with rac-RS/SR-
(Bab-Dipp)(Bab-tBu), meso-(Bab-tBu)2, and meso-(Bab-Dipp)2
(Table S9 in the Supporting Information). It is obvious that the
cleavage of the central HC(sp3)ÀC(sp3)H bonds during the con-
version between respective stereoisomers should produce cor-
In this work (vide supra), we have shown that all bis(1H-2,1-
benzazaborole) derivatives suffer from a (partially) restricted ro-
tation around the central CÀC bond. We have also recently
demonstrated that heating of a toluene solution of meso-
(Bab-tBu)2 at 1308C led to quantitative conversion to the ther-
modynamically more stable rac-RR/SS-(Bab-tBu)2 within
48 h.[9a] These findings led us to study this phenomenon in
more detail. The heating of meso-(Bab-Dipp)2 to 1408C in so-
lution resulted in the formation of rac-RR/SS-(Bab-Dipp)2 (see
the Supporting Information).[19] Initially, this finding has been
ascribed to the inversion at stereogenic carbon centers, but
the heating of a solution of rac-RS/SR-(Bab-Dipp)(Bab-tBu)
provided, besides the expected rac-RR/SS-(Bab-Dipp)(Bab-
tBu), the symmetrically substituted systems rac-RR/SS-(Bab-
tBu)2 and rac-RR/SS-(Bab-Dipp)2 (Scheme 4). This proves that
the conversion between particular diastereoisomers must pro-
ceed by a homolytic cleavage of the central C(sp3)ÀC(sp3)
bond with concomitant recombination to more stable diaste-
reoisomers. This fact is further corroborated by the heating of
a 1:1 stoichiometric mixture of meso-(Bab-tBu)2 and meso-
C
C
responding radicals (Bab-Dipp) or (Bab-tBu) , but all of our at-
tempts to identify them using EPR spectroscopy in the temper-
ature range 140–1808C, (both in a decaline solution as well as
a molten neat sample) failed.[20] This fact is most probably
a result of the low rate of CÀC bond cleavage at the given
temperature (full conversion requires at least 48 h at 1408C in
toluene solutions) and/or fast recombination of the respective
short-lived radicals under these conditions, resulting in their
low steady-state concentration.
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Irradiation at ambient temperature leads to a mixture of all six isomers, whereas heating leads to the three thermodynamically more stable iso-
mers.
Photo-induced HC(sp3)ÀC(sp3)H bond cleavage in bis(1H-2,1-
the mixture of all six possible bis(1H-2,1-benzazaborole) deriva-
benzazaborole) compounds
tives (Table S7, Figure S20).
As mentioned above, all isolated bis(1H-2,1-benzazaborole) de-
rivatives showed strong absorption between about 240–
270 nm in the UV/Vis absorption spectra (Figure S14 in the
Supporting Information). Consequently, all six enantiomeric
pairs of bis(1H-2,1-benzazaborole) variants were independently
irradiated at 254 nm for 2, 4, and 24 h in degassed THF and
the samples were subsequently analyzed using 1H and
13C NMR spectroscopy. It turned out that the particular isomer
converts into a mixture of expected products of the cleavage
of the central HC(sp3)ÀC(sp3)H bond with concomitant recom-
bination in all cases (Table S7). Thus, for the case of irradiation
of rac-RS/SR-(Bab-Dipp)(Bab-tBu), even all six possible prod-
1
ucts were detected in the resulting mixture by H and 13C NMR
spectroscopy (Scheme 4 and Figure 4). This finding unambigu-
ously confirmed that the central HC(sp3)ÀC(sp3)H bond may be
cleaved not only by heating, but also by photochemical
means.[21] Considering that a homolytic cleavage of the target
1
Figure 4. Stacked plot of H NMR and 13C{1H} APT NMR spectra in the me-
thine (CHÀN) group region before and after irradiation of rac-RS/SR-(Bab-
Dipp)(Bab-tBu) at 254 nm, measured in C6D6 after evaporation of the reac-
tion mixture. * denotes the residual THF signal.
C
C
bond is expected, the radicals (Bab-Dipp) or (Bab-tBu) should
be formed as elusive intermediates. Indeed, an irradiation of
rac-RR/SS-(Bab-Dipp)2 or meso-(Bab-Dipp)2 resulted in the ob-
servation of an EPR signal with noticeable splitting at g=
C
2.002, probably corresponding to the radical (Bab-Dipp)
(Figure 5).[22] Importantly, the analyzed solutions containing the
C
radical (Bab-Dipp) became EPR silent upon time indicating re-
combination of present radicals, but repeated irradiation of the
same samples again led to the observation of the same EPR
signal (Figure 5). Finally evaporation of these samples and their
1
analysis using H and 13C NMR spectroscopy proved the pres-
ence of expected compounds, that is, rac-RR/SS-(Bab-Dipp)2
and meso-(Bab-Dipp)2, thereby ruling out formation of any
side-product and/or decomposition of the samples. The iden-
C
tification of the radical (Bab-Dipp) contrasts with the failure in
C
attempts to identify its counterpart (Bab-tBu) by EPR spectros-
copy using an analogous procedure.[23,24] Nevertheless, indirect
Figure 5. Room-temperature X-band (9.42711 GHz) continuous wave (CW)
experimental and simulated EPR spectra of a solution of unstable (Bab-
C
evidence for the formation of (Bab-tBu) in these processes
was obtained by irradiation of a stoichiometric 1:1 mixture of
C
Dipp) generated in a THF solution of rac-RR/SS-(Bab-Dipp)2 or meso-(Bab-
rac-RR/SS-(Bab-tBu)2 and rac-RR/SS-(Bab-Dipp)2 that provided
Dipp)2 by irradiation at 254 nm. Interacting nuclei are colored in grey.
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15345
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The room-temperature[25] EPR spectrum of the (Bab-Dipp)
could be easily monitored by 11B NMR spectroscopy as well
(Figure S26).
C
radical agrees well with the simulated spectrum, in which hy-
perfine splitting corresponds to one hydrogen atom of the me-
The Li+(SOL)n(Bab-tBu)À was characterized by a coincidence
of experimental data with those obtained for this compound
above (see Supporting Information). The methylborate Li+
(SOL)n(Bab-tBu-Me)À in the form of two enantiomeric pairs
(1R,2S,3S/1S,2R,3R and 1R,2S,3R/1S,2R,3S; further abbreviated
thine C(3)-carbon (12.98 G, I=1= for 1H), one nitrogen atom
2
(2.14 G, I=1 for 14N), and one boron atom (2.28 G, I=3/2 for
11B). This finding is in excellent agreement with the theoretical-
C
ly obtained SOMO for the radical (Bab-Dipp) (Figure 6, for cal-
1
1
culated Mulliken spin densities, see Tables S10 and S11, Sup-
porting Information), which is mainly localized at the boron
atom and HCÀN moiety. Although a number of radical species
containing boron atoms within a heterocycle are known,[28]
C
only as RSS/SRR and RSR/SRS) was characterized by H, H-1H
NOESY, 11B, 13C, and 15N NMR spectra (Figure 7B, C and the Sup-
1
porting Information). The H and 13C NMR spectra showed typi-
cal signals for the CH(CH3)N groups and the 11B NMR spectra re-
probably a most relevant congener of (Bab-Dipp) is Nozaki’s vealed signals typical for a borate-type boron atom (À3.6 and
persistent N,N-stabilized neutral boron radical. In this case,
however, the spin density localized at the boron atom seems
to be considerably lower.[29] Furthermore the theoretical study
À4.3 ppm). Additional proof stems from 15N NMR chemical
shifts being high-field shifted by approximately 85 ppm in
comparison with starting rac-RR/SS-(Bab-tBu)2 (d 15N NMR=
À234.5 ppm).
C
also showed close similarity of both SOMOs of (Bab-Dipp) and
C
(Bab-tBu) (Figure 6).
As two molar equivalents of MeLi are necessary for clean
conversion, it is highly probable that the first equivalent of
MeLi does not attack the central HC(sp3)ÀC(sp3)H bond, but
adds to one of the BÀN bonds in an antarafacial fashion in the
starting rac-RR/SS-(Bab-tBu)2, forming a highly reactive inter-
mediate INT (Scheme 5). In this intermediate INT, the lithium
cation seems to be able to approach and interact with the p-
system of the five-membered C3BN ring of the second 1H-2,1-
benzazaborole ring (the grey one in the Scheme 5). Further-
more, in the structure of INT, the absolute configurations of N-
tBu and BÀPh groups, which are connected to newly formed
nitrogen and boron stereogenic centers, are exactly the same
as in the final products RSS/SRR- and RSR/SRS-Li+(SOL)n(Bab-
tBu-Me)À. Finally, the INT must be significantly more labile to-
wards the attack of the second molecule of MeLi [the cleavage
step of the HC(sp3)ÀC(sp3)H] from two possible directions
(endo and exo attack), thus inevitably giving final products
RSS/SRR- and RSR/SRS-Li+(SOL)n(Bab-tBu-Me)À and 1H-2,1-
benzazaborole lithium salt Li+(SOL)n(Bab-tBu)À (Scheme 5).
Finally, the identity of the methylborate Li+(SOL)n(Bab-tBu-
Me)À was also independently corroborated by its preparation
from 1-Ph-2-tBu-3-Me-1H-2,1-benzazaborole using MeLi.[4b] This
reaction quantitatively gave the methylborate Li+(SOL)n(Bab-
tBu-Me)À, but in this case as a mixture of all four possible
enantiomeric pairs (Scheme 5D, Figure 7D, and page S55 in
the Supporting Information). This contrasts with the observa-
tion of only two of such enantiomeric pairs in the reaction of
rac-RR/SS-(Bab-tBu)2 with two equivalents of MeLi
(Scheme 5C, Figure 7C, and page S58). This fact further con-
firms that the 1-Ph-2-tBu-3-Me-1H-2,1-benzazaborole is not
formed during the addition of MeLi to bis(1H-2,1-benzazabor-
ole) variants, thereby supporting the proposed mechanism
going through the intermediate INT.
Figure 6. Calculated SOMOs of respective radicals, at M06-2X/def2-SVP
level.[26,27]
Heterolytic HC(sp3)ÀC(sp3)H bond cleavage in bis(1H-2,1-
benzazaborole) derivatives
It was demonstrated above that the central HC(sp3)ÀC(sp3)H
bond in bis(1H-2,1-benzazaborole) derivatives is quite easily
cleaved mesolytically and homolytically (thermally and photo-
chemically). More surprisingly, it turned out that the same non-
polarized HC(sp3)ÀC(sp3)H bond can be cleaved in a heterolytic
fashion using MeLi. The reaction between either rac-RR/SS-
(Bab-tBu)2 or meso-(Bab-tBu)2 (see the Supporting Informa-
tion) in THF with one molar equivalent of MeLi in Et2O lead to
the formation of a familiar intensively red-colored solution, in-
dicative of the formation of the 1H-2,1-benzazaborolyl lithium
salt Li+(SOL)n(Bab-tBu)À (SOL=THF or diethylether). However,
starting materials rac-RR/SS-(Bab-tBu)2 with the methylborate
Li+(SOL)n(Bab-tBu-Me)À were isolated surprisingly as by-prod-
ucts instead of the expected 1-Ph-2-tBu-3-Me-1H-2,1-benzaza-
borole (Scheme 5B and Figure 7B).[4b] Nevertheless, use of two
molar equivalents of MeLi led to the formation of a 1:1 mixture
of Li+(SOL)n(Bab-tBu)À and the methylborate Li+(SOL)n(Bab-
tBu-Me)À (Scheme 5C, Figure 7C). The progress of the reaction
Conclusions
This contribution describes an intriguing and, to the best of
our knowledge, unprecedented reactivity of
a non-polar
HC(sp3)ÀC(sp3)H bond at the junction between two five-mem-
bered C3BN azaborole rings in (3,3’)-bis(1-Ph-2-R-1H-2,1-benza-
zaborole) compounds (Bab-R)2 (R=tBu or Dipp). Such reactivi-
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15346
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 5. Reaction between rac-RR/SS-(Bab-tBu)2 or 1-Ph-2-tBu-3-Me-1H-2,1-benzazaborole with MeLi, forming four enantiomeric pairs of the methylborate
Li+(SOL)n(Bab-tBu-Me)À. The enantiomeric pair with configuration RSS/SRR is the most favorable due to steric reasons (D). Sterically induced formation of
only two enantiomeric pairs during heterolytic HC(sp3)ÀC(sp3)H bond cleavage (C). For the full proposed mechanism, see page S65 in the Supporting Informa-
tion. SOL=THF or Et2O. The # denotes a compound recently prepared by us.[4b]
ty is far from the conventional behavior of “classical” HC(sp3)À
C(sp3)H bonds. The target bond is smoothly cleaved chemically
(in a mesolytic fashion) giving 1H-2,1-benzazaborolyl anions
M+(THF)n(Bab-tBu)À, but is also cleaved by heating or irradia-
tion giving thus short-lived 1H-2,1-benzazaborolyl radicals
C
(Bab-R) . The existence of these radicals has been confirmed
C
by EPR spectroscopy [for (Bab-R) , where R=Dipp] and/or
using NMR analysis of the self-trapping products. Furthermore,
(Bab-tBu)2 shows intriguing reactivity with MeLi. This reaction
sequence (with two equivalents of MeLi) yielding the 1H-2,1-
benzazaborolyl lithium salt Li+(SOL)n(Bab-tBu)À, along with
the corresponding methyl-substituted borate complexes Li+
(SOL)n(Bab-tBu-Me)À (a derivative of the benzazaborole core),
may be regarded as proof for heterolytical scission of the cen-
tral HC(sp3)ÀC(sp3)H bond. We are convinced that the reactivity
and chemical transformation described in this work are of gen-
eral significant interest and will be applicable to a number of
related CÀC bonded heterocyclic systems. This fact may pave
the way to compounds bearing interest from both theoretical
and chemical point of views, representing otherwise hardly ac-
cessible species.
1
Figure 7. Stacked plot of H NMR spectra ([D8]THF at 294 K) in the region of
Acknowledgements
4.0–6.3 ppm of rac-RR/SS-(Bab-tBu)2 (A), products after addition of MeLi
1
(B,1 equiv; C, 2 equiv), and for comparison, the H NMR spectrum of the
methylborates Li+(SOL)n(Bab-tBu-Me)À in all four possible enantiomeric
The Grant Agency of the Czech Republic (project no. P207/12/
0223) is acknowledged for financial support.
pairs (D).
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15347
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2013, 52, 6298–6301; Angew. Chem. 2013, 125, 6418–6421; e) S. P.
Chia, R. Ganguly, Y. X. Li, C. W. So, Organometallics 2012, 31, 6415–
6419; f) S. P. Chia, H. X. Yeong, S. W. So, Inorg. Chem. 2012, 51, 1002–
1010.
Keywords: atropisomerism · boron · CÀC bond cleavage ·
heterocycles · hindered rotation · radicals
˚
[9] a) M. Hejda, R. Jambor, A. Ruzicka, A. Lycka, L. Dostµl, Dalton Trans.
2014, 43, 9012–9015; b) M. Yamashita, Y. Aramaki, K. Nozaki, New J.
Chem. 2010, 34, 1774–1782; c) T. W. Myers, N. Kazem, S. Stoll, R. D. Britt,
M. Shanmugam, L. A. Berben, J. Am. Chem. Soc. 2011, 133, 8662–8672;
d) C. D. Cates, T. W. Myers, L. A. Berben, Inorg. Chem. 2012, 51, 11891–
11897; e) C. Seow, H.-W. Xi, Y. Li, C.-W. So, Organometallics 2016, 35,
1060–1063; f) S. P. Chia, E. Carter, H.-W. Xi, Y. Li, C. W. So, Angew. Chem.
Int. Ed. 2014, 53, 8455–8458; Angew. Chem. 2014, 126, 8595–8598.
[10] a) For example: K. Uchida, S. Ito, M. Nakano, M. Abe, T. Kubo, J. Am.
Chem. Soc. 2016, 138, 2399–2410; b) D. S. Marlin, M. M. Olmstead, P. K.
Mascharak, Angew. Chem. Int. Ed. 2001, 40, 4752–4754; Angew. Chem.
2001, 113, 4888–4890; c) P. Maslak, J. N. Narvaez, Angew. Chem. Int. Ed.
Engl. 1990, 29, 283–285; Angew. Chem. 1990, 102, 302–304; d) P.
Maslak, T. M. Vallombroso, W. H. Chapman, Jr., J. N. Narvaez, Angew.
Chem. Int. Ed. Engl. 1994, 33, 73–75; Angew. Chem. 1994, 106, 110–113.
[11] a) P. Pyykkç, M. Atsumi, Chem. Eur. J. 2009, 15, 186–197; b) P. Pyykkç,
M. Atsumi, Chem. Eur. J. 2009, 15, 12770–12779; c) P. Pyykkç, S. Riedel,
M. Patzschke, Chem. Eur. J. 2005, 11, 3511–3520.
[1] For examples, see: a) C. I. Rat, C. Silvestru, H. J. Breunig, Coord. Chem.
Rev. 2013, 257, 818–879 and references therein; b) The chemistry of
pincer compounds (Eds.: D. Morales-Morales, C. Jensen), Elsevier, Amster-
dam, 2007, Chapter 16, pp. 357–384 and references therein; c) G.
van Koten, D. Milstein, Organometallic Pincer Chemistry, Springer, Berlin,
2013, Chapter 6, pp. 175–202 and references therein; d) I. J. Casely,
J. W. Ziller, M. Fang, F. Furche, W. J. Evans, J. Am. Chem. Soc. 2011, 133,
˚
5244–5247; e) L. Dostµl, R. Jambor, A. Ruzicka, R. Jirµsko, V. Lochar, L.
Bene, F. De Proft, Inorg. Chem. 2009, 48, 10495–10497; f) L. Dostµl, R.
˚
Jambor, A. Ruzicka, A. Lycka, J. Brus, F. de Proft, Organometallics 2008,
˚
27, 6059–6062; g) M. Bouska, L. Strizík, L. Dostµl, A. Ruzicka, A. Lycka, L.
Benes, M. Vlcek, J. Prikryl, P. Knotek, T. Wµgner, R. Jambor, Chem. Eur. J.
2013, 19, 1877–1881; h) M. Bouska, L. Dostµl, Z. Padelkovµ, A. Lycka, S.
Herres-Pawlis, K. Jurkschat, R. Jambor, Angew. Chem. Int. Ed. 2012, 51,
3478–3482; Angew. Chem. 2012, 124, 3535–3540; i) R. Jambor, M.
Kasnµ, K. N. Kirschner, M. Schürmann, K. Jurkschat, Angew. Chem. Int. Ed.
2008, 47, 1650–1653; Angew. Chem. 2008, 120, 1674–1677.
[12] For a detailed description of atropisomerism, see G. Bringmann, A. J. P.
Mortimer, P. A. Keller, M. J. Gresser, J. Garner, M. Breuning, Angew. Chem.
Int. Ed. 2005, 44, 5384–5427; Angew. Chem. 2005, 117, 5518–5563.
[13] In fact, from all six bis(Bab) derivatives, the activation energy parame-
ters are experimentally accessible by VT-NMR study only for meso-(Bab-
Dipp)2 and meso-(Bab-tBu)2, because only in these cases is the coales-
cence of initially non-equivalent signals of connecting methine hydro-
gens of HC(3)ÀC(3’)H moiety theoretically observable.
[2] There are only a few examples where these C,N-chelating ligands take
˚
part in an important chemical transformation: a) L. Dostµl, A. Ruzicka, R.
Jambor, Eur. J. Inorg. Chem. 2014, 5266–5270; b) L. Balazs, H. J. Breunig,
E. Lork, A. Soran, C. Silvestru, Inorg. Chem. 2006, 45, 2341–2346; c) A.
Meller, H. Hoppe, W. Maringgele, A. Haase, M. Noltemeyer, Organome-
tallics 1998, 17, 123–124.
[3] Similar chemistry has been recently developed for related N,N-chelated
systems, but this remains beyond the central scope of this introductory
section. For examples, see: a) J. Flock, A. Slujanovic, A. Torvisco, W.
Schoefberger, B. Gerke, R. Pçttgen, R. C. Fischer, M. Flock, Chem. Eur. J.
2013, 19, 15504–15517; b) T. Chu, L. Belding, A. van der Art, T. Dudding,
I. Korobkov, G. I. Nikonov, Angew. Chem. Int. Ed. 2014, 53, 2711–2715;
Angew. Chem. 2014, 126, 2749–2753; c) Y. Li, K. C. Mondal, P. Stollberg,
H. Zhu, H. W. Roesky, R. Herbst-Imer, D. Stalke, H. Fiegl, Chem. Commun.
2014, 50, 3356–3358; d) R. J. Baker, C. Jones, M. Kloth, D. P. Mills, New J.
Chem. 2004, 28, 207–213.
[14] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith,
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski and D. J. Fox, Gaussian, Inc., Wallingford, CT, 2013.
[15] A. J. Ashe, III, Comprehensive Heterocyclic Chem. III, Vol. 4, 2008 Chap-
ter 4.17, pp. 1190 and references therein.
˚
[4] a) M. Hejda, A. Lycka, R. Jambor, A. Ruzicka, L. Dostµl, Dalton Trans.
˚
2013, 42, 6417–6428; b) M. Hejda, A. Lycka, R. Jambor, A. Ruzicka, L.
Dostµl, Dalton Trans. 2014, 43, 12678–12688; c) I. Vrµnovµ, R. Jambor,
˚
A. Ruzicka, R. Jirµsko, L. Dostµl, Organometallics 2015, 34, 534–541; d) Y.
Yang, N. Zhao, H. Zhu, H. W. Roesky, Organometallics 2012, 31, 1958–
1964.
˚
[5] a) L. Dostµl, R. Jambor, A. Ruzicka, P. Simon, Eur. J. Inorg. Chem. 2011,
˚
2380–2386; b) M. Novµk, L. Dostµl, M. Alonso, F. De Proft, A. Ruzicka, A.
[16] G. Schmid, Comprehensive Heterocyclic Chem. II, Vol. 3, 1996 Chap-
ter 3.17, pp. 739.
Lycka, R. Jambor, Chem. Eur. J. 2014, 20, 2542–2550; c) I. Vrµnovµ, M.
˚
Alonso, R. Jambor, A. Ruzicka, M. Erben, L. Dostµl, Chem. Eur. J. 2016,
[17] I. Aprahamian, D. Eisenberg, R. E. Hoffman, T. Sternfeld, Y. Matsuo, E. A.
Jackson, E. Nakamura, L. T. Scott, T. Sheradsky, M. Rabinovitz, J. Am.
Chem. Soc. 2005, 127, 9581–9587.
[18] a) According to a reviewer suggestion, this process is defined as quasi-
reversible from the electrochemistry point of view, but it is reversible in
the chemical meaning. This point is caused by the ECE mechanism (see
ref. 18b), which takes place in these systems; b) ECE mechanism means
electrochemistry-chemical reaction-electrochemistry mechanism as de-
fined in: J. Heinze, Angew. Chem. Int. Ed. Engl. 1984, 23, 831–847;
Angew. Chem. 1984, 96, 823–840.
22, 7376–7380.
[6] It is noteworthy that a similar cyclization has been described in related
systems with C,O-chelated compounds: a) K. Peveling, M. Schürmann, S.
Herres-Pawlis, C. Silvestru, K. Jurkschat, Organometallics 2011, 30, 5181–
˚
5187; b) L. Dostµl, R. Jambor, A. Ruzicka, R. Jirµsko, J. Holecek, F. De
Proft, Dalton Trans. 2011, 40, 8922–8934; c) M. Chovancovµ, R. Jambor,
˚
A. Ruzicka, R. Jirµsko, I. Císarovµ, L. Dostµl, Organometallics 2009, 28,
1934–1941; d) K. Peveling, M. Henn, C. Low, M. Mehring, M. Schürmann,
B. Costisella, K. Jurkschat, Organometallics 2004, 23, 1501–1508; e) M.
Mehring, I. Vrasidas, D. Horn, M. Schürmann, K. Jurkschat, Organometal-
lics 2001, 20, 4647–4653; f) K. Peveling, M. Schürmann, R. Ludwig, K. Ju-
rkschat, Organometallics 2001, 20, 4654–4663; g) M. Mehring, C. Low,
M. Schürmann, K. Jurkschat, Eur. J. Inorg. Chem. 1999, 887–898.
[7] Similar hydrosilylation in N,N-chelated compounds has been recently
described: L. Witteman, T. Evers, Z. Shu, M. Lutz, R. J. M. K. Gebbink, M.-
E. Moret, Chem. Eur. J. 2016, 22, 6078–6099.
[19] It is worth noting that heating neat samples of these meso- forms led
to an analogous transformation (see the Supporting Information).
[20] We tried to trap them at 1408C in toluene with benzoyl peroxide or di-
phenyldisulphide, however only a complicated mixture of products was
1
detected by H NMR spectroscopy.
[21] Irradiation in the solid state was also performed (1 hour at 254 nm),
1
˚
however only starting compounds were isolated, as judged by H NMR.
[8] For examples, see: a) P. Simon, F. De Proft, R. Jambor, A. Ruzicka, L.
Dostµl, Angew. Chem. Int. Ed. 2010, 49, 5468–5471; Angew. Chem. 2010,
122, 5600–5603; b) I. Vrµnovµ, M. Alonso, R. Lo, R. Sedlµk, R. Jambor, A.
[22] The set-up for the EPR measurement combined with irradiation ena-
bled only irradiation outside the EPR spectrometer, thus not allowing
for in situ irradiation (see details in the Supporting Information).
˚
Ruzicka, F. De Proft, P. Hobza, L. Dostµl, Chem. Eur. J. 2015, 21, 16917–
16928; c) S. Khan, R. Michel, J. M. Dieterich, R. A. Mata, H. W. Roesky, J.-P.
Demers, A. Lange, D. Stalke, J. Am. Chem. Soc. 2011, 133, 17889–17894;
d) S. P. Chia, H. W. Xi, Y. X. Li, K. H. Lim, C. W. So, Angew. Chem. Int. Ed.
[23] Irradiation of THF solutions of rac-RR/SS-(Bab-Dipp)(Bab-tBu) or rac-
RS/SR-(Bab-Dipp)(Bab-tBu) for the sake of possibly observing EPR sig-
C
C
nals of both respective (Bab-Dipp) and (Bab-tBu) radicals was done.
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15348
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
However, no signals were detected despite the fact that after evapora-
tion of the reaction mixtures and dissolving in C6D6, the NMR measure-
ment proved the formation of a mixture of all six possible bis(Bab)
compounds (Table S6 in the Supporting Information). This can only be
ture. Observation of radicals in this sample required repeated irradia-
tion. This led us to the conclusion that the rate of radical termination
rapidly increases as the temperature decreases.
[26] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–3305.
[27] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
[28] For examples, see: a) I. L. Fedushkin, O. V. Markina, A. N. Lukoyanov,
A. G. Morozov, E. V. Baranov, M. O. Maslov, S. Yu. Ketkov, Dalton Trans.
2013, 42, 7952–7961; b) J. F. Araneda, W. E. Piers, M. J. Sgro, M. Parvez,
Chem. Sci. 2014, 5, 3189–3196; c) H. Braunschweig, F. Breher, C.-W.
Chiu, D. Gamon, D. Nied, K. Radacki, Angew. Chem. Int. Ed. 2010, 49,
8975–8978; Angew. Chem. 2010, 122, 9159–9162.
C
C
rationalized by existence of the respective (Bab-Dipp) and (Bab-tBu)
radicals that immediately terminate to products.
[24] We tried to trap these radicals formed during the irradiation of any of
isomers in THF at 254 nm with TEMPO, benzoyl peroxide, diphenyldisul-
phide, diphenyldiselenide, hexaphenylditin, 2,3-dimethyl-1,3-butadiene
or decacarbonyl dimanganese(0). However, only a mixture of a variety
1
of products was detected by H NMR spectroscopy.
C
[25] For the purpose of obtaining EPR signals of (Bab-Dipp) with higher
[29] Y. Aramaki, H. Omiya, M. Yamashita, K. Nakabayashi, S. Ohkoshi, K.
Nozaki, J. Am. Chem. Soc. 2012, 134, 19989–19992.
resolution, the EPR measurement at lower temperature was performed.
Whereas at room temperature, we were able to observe a relatively
strong signal for this radical, after slight cooling (even at 58C), the ini-
tially relatively strong signal began to fade. When the temperature
went down even further, the signal immediately disappeared and re-
mained EPR silent after subsequent heating back to the room tempera-
Received: June 7, 2016
Published online on September 13, 2016
Chem. Eur. J. 2016, 22, 15340 –15349
www.chemeurj.org
15349
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim