Reduction of C,N-chelated chloroborane: Straightforward formation of the unprecedented 1H-2,1-benzazaborolyl potassium salt

Article abstract of DOI:10.1039/c4dt00812j

Reduction of C,N-chelated chloroborane [2-(CH=NtBu)C₆H ₄]BPhCl (1) with the potassium metal afforded (3,3′)-bis(1-Ph- 2-tBu-1H-2,1-benzazaborole) (2). Compound 2 is formed via C-C reductive coupling reaction. Subsequent reduction of

Full text of DOI:10.1039/c4dt00812j

Dalton
Transactions
COMMUNICATION
Reduction of C,N-chelated chloroborane:
straightforward formation of the unprecedented
1H-2,1-benzazaborolyl potassium salt†
Cite this: Dalton Trans., 2014, 43,
9012
Received 18th March 2014,
Accepted 3rd April 2014
Martin Hejda,a Roman Jambor,a Aleš Růžička,a Antonín Lyčkab and Libor Dostál*a
DOI: 10.1039/c4dt00812j
www.rsc.org/dalton
Reduction of C,N-chelated chloroborane [2-(CHvNtBu)C6H4]-
BPhCl (1) with the potassium metal afforded (3,3’)-bis(1-Ph-2-tBu-
1H-2,1-benzazaborole) (2). Compound 2 is formed via C–C reduc-
tive coupling reaction. Subsequent reduction of 2 with two equi-
valents of the potassium metal produced orange crystals of 1Ph-
2tBu-1H-2,1-benzazaborolyl (Bab) potassium salt K(THF)(Bab) (3).
Compound 3 is able to react with simple electrophiles (MeI or
Me3SiCl) resulting in the formation of substituted 1H-2,1-
benzazaboroles.
Fig. 1 Structures of the discussed BN analogues of indenyl anions.
The isoelectronic relationship between B–N and CvC moieties
has been appreciated and exploited for several decades. Thus,
it is not surprising that substitution of one or more B–N frag-
ments for CvC fragments in classical aromatic hydrocarbons
has led to the development of a new generation of materials
with desirable photophysical, electrochemical or biological
activity.1 A similar approach has been used by Schmid in the
1980s for the synthesis of 1,2-azaborolyl anions by deprotona-
tion of 1,2-azaboroles, which were subsequently ligated to
numerous transition metals as BN analogues of the cyclo-
pentadienyl anion.2 These pioneering studies were later on fol-
lowed by the groups of Ashe, Fu and Fang.3 Benzazaborolyl
anions, in which the azaborolyl fragment is fused with an
extra aromatic ring, are known only in the form 1H-1,2-benza-
zaborolyl isomer (Fig. 1A) and were used as indenyl analogues
for the coordination of transition metals (Ti or Zr), and these
results have been reported in the patent literature.4 Neverthe-
less, the molecular structure of none of these complexes has
been reported. In addition, there is no report dealing with the
second possible isomer 1H-2,1-benzazaborolyl, although
several neutral 1H-2,1-benzazaboroles are known (Fig. 1B).5 We
have recently discovered a facile synthetic strategy for the
preparation of 1H-2,1-benzazaboroles via the nucleophilic
addition of lithium reagents to the activated imine CvN
double bond in C,N-chelated chloroboranes.6 As a part of our
research in this field, we report herein the straightforward for-
mation of potassium salt K(THF)(Bab) (3) (Bab = 1Ph-2tBu-1H-
2,1-benzazaborolyl) as a representative of 1H-2,1-benzazaboro-
lyl anions (Fig. 1B). Compound 3 was formed via non-conven-
tional two step reduction of C,N-chelated chloroborane 1.
Reduction of 1 with potassium in toluene afforded com-
pound 2 (Scheme 1) as a mixture of isomers as a result of the
presence of two stereogenic centres in 2. Thus, presence of
RR-2, SS-2 and meso-2 was expected in the reaction mixture
(see details in the ESI†).
This fact was reflected by the observation of two sets of
1
signals in the corresponding H and 13C NMR spectra of the
reaction mixture (Fig. S1†). Nevertheless, meso-2 and the race-
mate rac-RR/SS-2 could be separated by fractional crystalliza-
1
tion and independently characterized by H, 11B, 13C and 15N
aDepartment of General and Inorganic Chemistry, Faculty of Chemical Technology,
University of Pardubice, Studentská 573, CZ – 532 10 Pardubice, Czech Republic.
E-mail: libor.dostal@upce.cz; Fax: +420466037068; Tel: +420466037163
NMR spectroscopy (see details in the ESI†). Especially, the
presence of typical resonances for methine groups CH–N in
1H and 13C NMR spectra [δ(1H) = 5.36 and 5.81 ppm for meso-
2;7a 5.22 ppm for rac-RR/SS-2; δ(13C) = 74.7 and 71.9 ppm for
meso-2;7a 70.0 ppm for rac-RR/SS-2] together with the absence
of any signal attributable to the imino CHvN group approved
the structure of 2. The 11B NMR spectra revealed one singlet at
bResearch Institute for Organic Syntheses, Rybitví 296, CZ –533 54 Pardubice,
Czech Republic
†Electronic supplementary information (ESI) available: Full synthetic, spectro-
scopic and crystallographic details of reported compounds. CCDC
989938–989940. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c4dt00812j
9012 | Dalton Trans., 2014, 43, 9012–9015
This journal is © The Royal Society of Chemistry 2014

Dalton Transactions
Communication
induced by favourable alignment of highly substituted 1H-2,1-
benzazaborole moieties (see details in the ESI†). This is the
reason why the meso-2 also crystallizes in the centrosymmetric
8
ˉ
space group (P1) as a racemate, but now due to the molecular
chirality caused by asymmetry at the C(7)–C(24) bond
(Fig. S5†). The C(7)–C(24) bond lengths of 1.584(2) and 1.581(2)
Å for rac-RR/SS-2 and meso-2, respectively, indicate the pres-
ence of the C–C single bond, which is a bit elongated in com-
parison with the standard C(sp3)–C(sp3) (1.54 Å) bond. The
B–N bond lengths within the benzazaborole rings [1.409(2)–
1.415(2) Å] are apparently shorter than the sum of covalent
radii for the single bond Σcov(N,B) = 1.56 Å9 and correspond to
the value for the respective double bond (1.48 Å)9 reflecting
the strong π(N) → π(B) interaction. Consequently, the coordi-
nation geometry around both nitrogen and boron atoms
remains essentially trigonal planar.
Scheme 1 Synthesis of compound 2 illustrating two possible mecha-
nisms for its formation.
Treatment of 210 with potassium in THF at ambient temp-
erature produced a red solution from which dark orange crys-
tals of 3 were obtained after workup (Scheme 2). Compound 3
44.8 and 42.0 ppm for meso-27b and rac-RR/SS-2, respectively,
thereby proving the presence of sp2 hybridized boron atoms
within the 1H-2,1-benzazaborole core.2b It is noteworthy that it
turned out that heating of a toluene solution of meso-2 led to
quantitative conversion to rac-RR/SS-2 within 48 h (see details
in the ESI†).
Molecular structures of rac-RR/SS-2 and meso-2 were unam-
biguously determined by single-crystal X-ray diffraction analy-
sis (Fig. 2). rac-RR/SS-2 crystallizes in the centrosymmetric
P21/c space group. Based on molecular structures of rac-RR/
SS-2 and meso-2, all stereoisomers form stable atropisomers due
to the restricted rotation around the C(7)–C(24) bond, which is
1
was characterized with the help of H, 11B, 13C and 15N NMR
spectroscopy (see details in the ESI†). The signals due to the
CH–N moiety in 1H and 13C NMR spectra are significantly low-
field shifted [δ(1H) = 6.21 and δ(13C) = 96.4 ppm] in compari-
son with 2 or related 1H-2,1-benzazaboroles6 pointing to the
aromatic character of this ring system.1b The 11B NMR spec-
trum of 3 revealed one signal at 23.4 ppm, which is high-field
shifted in comparison with related 1,2-azaborolyl transition
metal complexes [δ(11B) = 30–35 ppm], thereby indicating
higher electron density on the boron atom in 3 in comparison
with these 1,2-azaborolyl metal complexes.1b,2b The 15N NMR
spectrum of 3 revealed one signal at −209.4 ppm, which is
slightly down-field shifted in comparison with the values
observed for meso-2 [δ(15N) = −232.6 and −233.3 ppm]7a and
for rac-RR/SS-2 [δ(15N) = −234.5 ppm]. The molecular structure
of 3 was unambiguously established by single-crystal X-ray
diffraction analysis (Fig. 3). To the best of our knowledge,
3 represents the first example of a structurally characterized 1H-
2,1-benzazaborolyl anion (in the form of its potassium salt) as
the BN analogue of the indenyl moiety. The potassium atom
K(1) is coordinated by two benzazaborolyl anions and one THF
molecule with the bond length K(1)–O(1) of 2.668(2) Å that
approaches Σcov(K,O) = 2.59 Å.9 This bonding situation leads to
the formation of a chiral supramolecular structure of 3, which
crystallizes in space group P21. This supramolecular chirality is
caused by rotation around a bisector defined by the centroid
Fig. 2 Molecular structures of RR isomer of rac-RR/SS-2 (top) and one
of the atropisomers of meso-2 (bottom) (30% thermal ellipsoids; hydro-
gen atoms omitted except those connecting CH groups). Selected bond
lengths (Å), angles and torsion angles (°) of RR isomer of rac-RR/SS-2
(values for the atropisomer of meso-2 are given in brackets): C(24)–C(7)
1.584(2) [1.581(2)], B(1)–N(1) 1.413(2) [1.409(2)], B(2)–N(2) 1.412(2) [1.415(2)],
N(1)–C(7) 1.497(2) [1.4920(19)], N(2)–C(24) 1.495(2) [1.486(2)], N(1)–
C(7)–C(24) 112.33(12) [118.59(12)], N(2)–C(24)–C(7) 112.44(12) [117.60
(12)], N(1)–C(7)–C(24)–N(2) 179.32(12) [82.65(16)].
Scheme 2 Synthesis of compound 3 and its reactivity with selected
electrophiles.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9012–9015 | 9013

Communication
Dalton Transactions
preparation of 2 (Fig. S2†), which disappeared during the reac-
tion, may be indicative of in situ formation of the potassium
salt 3. Compound 3 can then smoothly react with the present
chloroborane 1 resulting in the formation of 2. To support this
presumption, the reaction of isolated 3 with one molar equi-
valent of 1 was performed and, indeed, it provided a mixture
of rac-RR/SS-2 and meso-2 in nearly quantitative yield as judged
by NMR spectroscopy. This means that 3 may serve as a com-
peting reagent to potassium metal during the formation of 2
(Scheme 1). In the light of these facts, we are not able to dis-
tinguish between these two mechanisms (I or II) at the
moment, but it seems probable that 2 is formed by both of
them simultaneously. It is noteworthy that analogous C–C
coupling reactions have been recently observed for α-iminopyr-
idyl substituted Ge, Al, Ga, and Zn complexes and lanthanide
(Yb, Sm) derivatives of redox active ligands.15 Compound 3
also readily reacts with simple electrophiles such as MeI or
Me3SiCl; in this case formation of substituted 1H-2,1-benzaza-
boroles 4 and 5 (Scheme 2) was observed. Compounds 4 and 5
Fig. 3 View of the polymeric structure of 3 (30% thermal ellipsoids;
hydrogen atoms omitted; symmetry operators: a = 1 − x, −1/2 + y, 1 − z
b = 1 − x, 1/2 + y, 1 − z). Selected bond lengths (Å): K(1)–O(1) 2.668(2),
K(1)–B(1) 3.192(2), K(1)–N(1) 3.1541(18), K(1)–C(1) 3.028(2), K(1)–C(2)
2.882(2), K(1)–C(7) 2.965(2), K(1)–B(1b) 3.479(2), K(1)–N(1b) 3.3388(17),
K(1)–C(1b) 3.123(2), K(1)–C(2b) 2.838(2), K(1)–C(7b) 2.992(2), B(1)–N(1)
1.460(3), C(1)–B(1) 1.517(3), C(1)–C(2) 1.455(3), C(2)–C(7) 1.381(3), C(7)–
N(1) 1.418(3).
1
were characterized by elemental analysis and H, 11B, 13C, 15N
and 29Si NMR spectroscopy (see details in the ESI†).
In summary, compound 3, a potassium salt of the first
structurally characterized BN analogue of the indenyl anion,
has been prepared by unusual two step reduction of the start-
ing chloroborane 1 involving both C–C bond coupling and
subsequent C–C bond cleavage reactions. This approach
seems to be applicable to the preparation of related main
group element aromatic systems. Furthermore, 3 is able to
react not only with simple electrophiles leading to quantitative
formation of benzazaboroles 4 and 5, but also it converts with
chloroborane 1 into C–C coupled compound 2, thereby
opening up a new strategy for the preparation of such C–C
bridged heterocyclic systems. Finally, compound 3 represents
a promising starting material for the preparation of indenyl-
like metal complexes. All these possibilities are now being
examined in our labs.
of the five-membered C3BN ring of the 1H-2,1-benzazaborolyl
moiety and the potassium atom.
One of the C3BN rings coordinates to the K(1) ion in an
approximately η5-fashion. The bond lengths describing this
bonding interaction [K(1)–N(1) 3.1541(18), K(1)–C(1) 3.028(2),
K(1)–C(2) 2.882(2), K(1)–C(7) 2.965(2) and K(1)–B(1) 3.192(2) Å]
are similar to those observed for the only structurally charac-
terized alkali metal analogue, i.e. lithium 1,2-azaborolyl
reported by Schmid11 considering different radii of both alkali
metals. In contrast, the interaction between the K(1) atom and
the second C3BN ring may be considered as η2-type mediated
by C(2b) and C(7b) atoms [K(1)–C(2b) 2.838(2) and K(1)–C(7b)
2.992(2) Å]. The difference in coordination of both C3BN rings
is supported by distances between K(1) and centroids of the
respective C3BN rings being 2.787 (η5 ring) vs. 2.913 Å (η2
ring). The B(1)–N(1) bond length [1.460(3) Å] is slightly shorter
than the corresponding distance observed in the lithium 1,2-
azaborolyl [1.503(6) Å], while other bond distances within the
C3BN ring are comparable to this lithium compound.11
The Grant Agency of the Czech Republic (project no. P207/
12/0223) is acknowledged for financial support.
Notes and references
The mechanism of formation of 2 is also of particular inter-
est. Compound 2 may be formed by fast12 recombination of
two carbon centred radicals, formed by reduction of 1 with
potassium (Mechanism I in Scheme 1). It is worth noting that
analogous C–C reductive coupling has been recently reported
by Nozaki et al. during reduction of base-stabilized difluoro-
boranes.13 Furthermore, the same working group clearly
showed that even their stable N-coordinated heterocyclic
boron radical may be described by several resonance struc-
tures, but the carbon centred radical has the major contri-
bution to the structure.14 All these facts support Mechanism I
(Scheme 1). Despite these facts, the second mechanism (Mech-
anism II in Scheme 1) came into mind. The appearance of
intensive red colour at the interface between the potassium
mirror and the reaction mixture in the Schlenk tube during
1 For reviews see: (a) M. J. D. Bosdet and W. E. Piers,
Can. J. Chem., 2008, 86, 8 and references cited therein;
(b) A. J. Ashe III, Comprehensive Heterocyclic Chem. III, 2008,
ch. 4.17, vol. 4, p. 1190 and references cited therein.
2 For example see: (a) G. Schmid, Comments Inorg. Chem.,
1985, 4, 17 and references cited therein; (b) G. Schmidt,
Comprehensive Heterocyclic Chem. II, 1996, vol. 3, ch. 3.17,
p. 739; (c) G. Schmid and M. Schütz, J. Organomet. Chem.,
1995, 492, 185; (d) G. Schmid, O. Boltsch and R. Boese,
Organometallics, 1987, 6, 435.
3 For example see: (a) A. J. Ashe III, Organometallics, 2009,
28, 4236; (b) S. Y. Liu, M. M. C. Lo and G. C. Fu, Tetra-
hedron, 2006, 62, 11343; (c) S. Y. Liu, M. M. C. Lo and
G. C. Fu, Angew. Chem., Int. Ed., 2002, 41, 174; (d) H. Yang,
9014 | Dalton Trans., 2014, 43, 9012–9015
This journal is © The Royal Society of Chemistry 2014

Dalton Transactions
Communication
X. Fang, J. W. Kampf and A. J. Ashe III, Polyhedron, 2005,
24, 1280; (e) X. Fang and J. Assoud, Organometallics, 2008,
27, 2408.
9 (a) P. Pyykkö and M. Atsumi, Chem. – Eur. J., 2009, 15, 186;
(b) P. Pyykkö and M. Atsumi, Chem. – Eur. J., 2009, 15,
12770; (c) P. Pyykkö, S. Riedel and M. Patzschke, Chem. –
Eur. J., 2005, 11, 3511.
4 (a) S. Nagy, R. Krishnamatti and B. P. Etherton, U.S. Patent
6,228,959, 2001; Chem. Abstr., 1997, 126, 19432j; 10 Regardless of whether meso-2 or rac-RR/SS-2 or even a
(b) Q. Wang, P. Zoricak and X. Gao, Can. Patent Appl.
2,225,014, 1999; Chem. Abstr., 1999, 142, 219701.
5 For example see: (a) V. Lee et al., U.S. Patent 7,465,836, 2008;
mixture of both was used.
11 G. Schmid, D. Zaika, J. Lehr, N. Augart and R. Boese, Chem.
Ber., 1988, 121, 1873.
(b) V. Lee et al., U.S. Patent 8,106,031, 2012; (c) A. M. Genaev, 12 All our attempts to characterize the radical intermediate by
S. M. Nagy, G. E. Salnikov and V. G. Shubin, Chem. EPR spectroscopy failed.
Commun., 2000, 1587; (d) T. S. De Vries, A. Prokofjevs, 13 M. Yamashita, Y. Aramaki and K. Nozaki, New J. Chem.,
J. N. Harvey and E. Vedejs, J. Am. Chem. Soc., 2009, 131, 2010, 34, 1774.
14679; (e) A. Rydzewska, K. Slepokura, T. Lis, P. Kafarski and 14 Y. Aramaki, H. Omiya, M. Yamashita, K. Nakabayashi,
P. Młynarz, Tetrahedron Lett., 2009, 50, 132.
6 M. Hejda, A. Lyčka, R. Jambor, A. Růžička and L. Dostál,
Dalton Trans., 2013, 42, 6417.
S. Ohkoshi and K. Nozaki, J. Am. Chem. Soc., 2012, 134,
19989.
15 (a) Y. Li, K. C. Mondal, P. Stollberg, H. Zhu, H. W. Roesky,
R. Herbst-Imer, D. Stalke and H. Fiegl, Chem. Commun.,
2014, 50, 3356–3358; (b) R. J. Baker, C. Jones, M. Kloth and
D. P. Mills, New J. Chem., 2004, 28, 207; (c) I. L. Fedushkin,
S. Dechert and H. Schumann, Angew. Chem., Int. Ed., 2001,
40, 561; (d) M. Westerhausen, T. Bollwein, N. Makropoulos,
T. M. Rotter, T. Harbereder, M. Suter and H. Nöth,
Eur. J. Inorg. Chem., 2001, 851; (e) R. J. Baker and C. Jones,
Chem. Commun., 2003, 390.
7 (a) Due to the atropisomerism observed in meso-2 two
signals for CH–N groups were observed in the corres-
ponding 1H, 13C and 15N NMR spectra (see details in the
ESI†); (b) Two signals should be also observed in the 11B
NMR spectrum, but they are overlapped due to the similar
chemical shifts and broadening of these signals.
8 As a racemate of two atropisomers see the ESI† for a
detailed discussion.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 9012–9015 | 9015

Copyright of Dalton Transactions: An International Journal of Inorganic Chemistry is the
property of Royal Society of Chemistry and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.