Reactivity of C,N-chelated organoboron compounds with lithium anilides-formation of unexpected 1,2,3-trisubstituted 1H-2,1-benzazaboroles

Article abstract of DOI:10.1039/c3dt32850c

A set of C,N-intramolecularly coordinated boranes containing various C,N-chelating ligands L^1-3 (where L¹ = [o-(CHNtBu)C ₆H₄], L² = [o-(CHN-2,6-iPr₂C ₆H₃)C₆H₄], L³ = [o-(CH₂NMe₂)C₆H₄]); L ^1-3BCl₂ (for 1 L = L¹, for 2 L = L², for 5 L = L³), L¹BPhCl (3) and L¹BCy ₂ (4) (where Cy = cyclohexyl) were synthesized and fully characterized by multinuclear NMR spectroscopy and in cases of 1 and 3-5 by the single crystal X-ray diffraction analysis. The reaction of 1-3 with the anilides ArNHLi (Ar = 2,6-Me₂C₆H₃ or 2,6-iPr ₂C₆H₃) proceeded via unexpected addition of anilide across the CN bond yielding 1,2,3-trisubstituted 1H-2,1-benzazaboroles 6-11, whose structures were unambiguously established by single crystal X-ray diffraction analysis (except for 11) and multinuclear NMR spectroscopy. In contrast, compounds 4 and 5 were inert towards ArNHLi. The investigation dealing with the reaction mechanism between the parent boranes 1-3 and ArNHLi revealed that amidolithiation of the CN double bond involved in the ligand backbones is the crucial step of the whole reaction. The CN double bond in 1-3 is activated by its coordination to the ortho bonded Lewis acidic boron center, which was also proven by the fact that the non-substituted ligand L¹H did not react with ArNHLi under the same reaction conditions in an analogous reaction.

Full text of DOI:10.1039/c3dt32850c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Reactivity of C,N-chelated organoboron compounds
with lithium anilides – formation of unexpected
1,2,3-trisubstituted 1H-2,1-benzazaboroles†
Cite this: Dalton Trans., 2013, 42, 6417
Martin Hejda,a Antonín Lyčka,b Roman Jambor,a Aleš Růžičkaa and Libor Dostál*a
A set of C,N-intramolecularly coordinated boranes containing various C,N-chelating ligands L1–3 (where
L1 = [o-(CHvNtBu)C6H4], L2 = [o-(CHvN-2,6-iPr2C6H3)C6H4], L3 = [o-(CH2NMe2)C6H4]); L1–3BCl2 (for 1
L = L1, for 2 L = L2, for 5 L = L3), L1BPhCl (3) and L1BCy2 (4) (where Cy = cyclohexyl) were synthesized and
fully characterized by multinuclear NMR spectroscopy and in cases of 1 and 3–5 by the single crystal
X-ray diffraction analysis. The reaction of 1–3 with the anilides ArNHLi (Ar = 2,6-Me2C6H3 or 2,6-iPr2C6H3)
proceeded via unexpected addition of anilide across the CvN bond yielding 1,2,3-trisubstituted
1H-2,1-benzazaboroles 6–11, whose structures were unambiguously established by single crystal X-ray
diffraction analysis (except for 11) and multinuclear NMR spectroscopy. In contrast, compounds 4 and 5
were inert towards ArNHLi. The investigation dealing with the reaction mechanism between the parent
boranes 1–3 and ArNHLi revealed that amidolithiation of the CvN double bond involved in the ligand
backbones is the crucial step of the whole reaction. The CvN double bond in 1–3 is activated by its
coordination to the ortho bonded Lewis acidic boron center, which was also proven by the fact that the
non-substituted ligand L1H did not react with ArNHLi under the same reaction conditions in an analo-
gous reaction.
Received 28th November 2012,
Accepted 9th January 2013
DOI: 10.1039/c3dt32850c
www.rsc.org/dalton
synthesized by the reaction of RBCl2 with an excess of lithium
anilides.1,3a,4c By an analogy, we decided to prepare a set of
intramolecularly coordinated chloroboranes of the general
Introduction
The chemistry of dianionic boraamidinate ligands1 [RB(NR′)2]2−
,
analogues of popular amidinate ligands2 [RC(NR′)2]−, has formula LBCl2 (where L is a C,N-chelating ligand) and to treat
gained considerable attention recently (Fig. 1A). These ligands them with sterically demanding lithium anilides ArNHLi. Sur-
have been used with success for stabilization of both main prisingly, these attempts did not lead to formation of desired
group and transition metal compounds.3 They were also bis(arylamino)boranes LB(NHAr)2, as precursors for boraami-
shown to be able to stabilize various radical species.4 The dinate ligands, instead unexpected addition of anilide across
common feature of all boraamidinate ligands is an application the CvN bond took place and a variety of highly substituted
of simple organic groups bonded to the boron atom.1 As we 1H-2,1-benzazaboroles was isolated in good yields (see further
are interested in the chemistry of C,N- or N,C,N-chelated main discussion).
group metal compounds,5 we decided to try to incorporate a
Azaboroles represent a class of compounds, which is known
C,N-chelating ligand (Fig. 2) into the boron atom in the structure in the literature for a long time and can be prepared by a
of classical boraamidinates and use such modified ligands for variety of reaction paths.6 They can also be transformed into
coordination of main group elements (Fig. 1B). The precursors an aromatic system and used as ligands for both transition
of classical boraamidinates bis(arylamino)boranes may be and main group metals as cyclopentadienyl class ligand ana-
logues (Fig. 1C).6,7 Analogous benzazaboroles, in which the
azaborole fragment is fused with an extra aromatic ring, are
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
bResearch Institute for Organic Syntheses, Rybitví 296, CZ-533 54, Pardubice,
also known in the literature. These compounds are able to
exist in two possible isomeric forms i.e. 1H-1,2-benzazaboroles
(Fig. 1D) and 1H-2,1-benzazaboroles (Fig. 1E). The first group
of 1H-1,2-benzazaboroles has been used, after deprotonation,
as ligands for transition metals as analogues of the indenyl
anions (Fig. 1D).8 Reports dealing with substituted 1H-2,1-
benzazaboroles also appeared in the literature and they were
Czech Republic
†Electronic supplementary information (ESI) available: Full assignment of NMR
data for 6–11. CCDC 908864–908873. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c3dt32850c
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6417

View Article Online
Paper
Dalton Transactions
with the lithium anilides ArNHLi (Ar = 2,6-Me2C6H3, or 2,6-
iPr2C6H3). The crucial step of these reactions is an amido-
lithiation (a nucleophilic addition of lithium anilide) of the
CvN double bond in the structure of the ligands L1,2, which is
most probably activated by the strong N→B intramolecular
interaction.
Results and discussion
Syntheses and structures of starting compounds 1–5
Compounds 1–4 (Scheme 1) have been synthesized using a
well documented procedure based on the reaction between
organolithium precursors and (organo)boron halides.15 Com-
pound 5 was prepared according to the literature procedure.16
All compounds were obtained as stable crystalline solids in
good yields (82% – 1, 75% – 2, 85% – 3, 77% – 4), which can
be stored for a long time under an inert atmosphere. 1–5 are
highly soluble in chlorinated solvents and showed only limited
solubility in aromatic solvents, hexane and pentane.
The identity of 1–4 was unambiguously established with the
help of elemental analysis and 1H, 10B,17 13C and 15N NMR
spectroscopy (see the Experimental section). The molecular
structures of 1, 3–5 were determined with the help of single
crystal X-ray diffraction analyses and are depicted in Fig. 3 and
4 together with selected structural parameters given in the
figure captions. The crystallographic data of 1, 3–5 are sum-
marized in the Experimental section.
Fig. 1 Structures of known azaboroles and related anions.
The common feature of molecular structures of 1, 3–5 is the
presence of intramolecular N→B interaction, which are charac-
terized by the N–B bond distances 1.602(4) Å for 1, 1.632(3) Å
for 3, 1.689(2) Å for 4, 1.634(2) Å for 5. The N–B bond distances
become slightly longer going from 1 to 3 and 4 reflecting low-
ering of the Lewis acidity of the central boron atom in this set
of compounds, which results in an attenuation of the N→B
interaction. Nevertheless, all B–N bond distance values still
approach the sum of covalent radii Σcov(N,B) = 1.56 Å18 and
correspond to the values observed for other structurally charac-
terized C,N-chelated boranes.15 Toyota and co-workers19 and
later on Hopfl20 developed a concept to gauge the strength of
donor–acceptor bonds in amine–borane complexes known as
percent tetrahedral character (%THC). This approach was
Fig. 2 Structures of the ligands used in this work.
studied due to their antiviral and antibacterial properties.9 The
first examples of 1,2-disubstituted 1H-2,1-benzazaboroles were
synthesized by Köster et al. by condensation reactions of
aminoborane adducts at elevated temperatures (∼200 °C).10 In
contrast, 1,2,3-trisubstituted 1H-2,1-benzazaborole rings are
still scarce. These compounds were prepared using intramole-
cular condensation of o-(aminomethyl)benzeneboronic
acids.11 Later on, Nagy et al. discovered cyclization of amino-
boranes under milder conditions using AlCl3 in dichloro-
methane at
0
°C.12 In 2009, Młynarz et al. described
unexpected formation of 1,2,3-trisubstituted 1H-2,1-benzaza-
borole via a reaction of (N-benzyl)benzylideneimine-2-boronic
acid with diethylphosphite.13 Recently, several substituted
1H-2,1-benzazaboroles have been obtained in the course of the
investigation of nitrogen-directed aromatic borylation as well.14
Herein, we report on an unexpected procedure for for-
mation of 1,2,3-trisubstituted 1H-2,1-benzazaboroles using the
reactions of the C,N-intramolecularly coordinated boranes,
containing C,N-chelating ligands L1–3 (where L1
= [o-
(CHvNtBu)C6H4], L2 = [o-(CHvN-2,6-iPr2C6H3)C6H4], L3 = [o-
(CH2NMe2)C6H4], Fig. 2); L1–3BCl2 (for 1 L = L1, for 2 L = L2, for
5 L = L3), L1BPhCl (3) and L1BCy2 (4) (where Cy = cyclohexyl)
Scheme 1 Preparation of the starting C,N-chelated boranes 1–5.
6418 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 3 ORTEP plot of a molecule of 1 (left) and 3 (right) showing 40% prob-
ability displacement ellipsoid. Hydrogen atoms are omitted for clarity. Only one
position of the disordered tBu moiety is shown in the case of 1 (see experimen-
tal for details). Selected bond lengths [Å] and bonding angles [°]: For 1 (sym-
metry operator a = x, 1/2 − y, z): B(1)–C(1) 1.583(4), B(1)–N(1) 1.602(4), B(1)–
Cl(1) 1.8582(17), N(1)–B(1)–C(1) 99.7(2), N(1)–B(1)–Cl(1) 109.92(15), C(1)–B(1)–
Cl(1) 112.64(14), Cl(1)–B(1)–Cl(1a) 111.45(14). For 2: B(1)–C(1) 1.602(3),
B(1)–N(1) 1.632(3), B(1)–Cl(1) 1.893(2), B(1)–C(12) 1.598(3), N(1)–B(1)–C(1)
97.93(14), N(1)–B(1)–Cl(1) 107.09(12), N(1)–B(1)–C(12) 114.23(15), C(1)–B(1)–
Cl(1) 107.74(13), C(1)–B(1)–C(12) 115.92(15), C(12)–B(1)–Cl(1) 112.66(14).
Scheme 2 Preparation of 6–9.
2,6-iPr2C6H3) was studied with the aim of formation of bis(aryl-
amino)boranes, which would be substituted at the boron atom
by chelating ligands L1–3. No reaction was observed between 5
and ArNHLi even when an excess of the lithium reagent was
used at elevated temperatures and only the starting material 5
was detected in the reaction mixture or after workup as a
major isolable product. In contrast, the reaction between 1 or
2 and anilides ArNHLi proceeded smoothly (Scheme 2) to give
substituted 1H-2,1-benzazaboroles 6–9 that were observed after
workup instead of targeted bis(arylamino)boranes (68% – 6,
81% – 7, 66% – 8, 55% – 9). All 1H-2,1-benzazaboroles 6–9
were isolated as colourless solids which are stable for a long
time under an inert atmosphere. All compounds were fully
characterized using multinuclear NMR spectroscopy (see the
Experimental section). The molecular structures of 6–9 were
unambiguously established with the help of X-ray diffraction
analyses. The molecular structures are given in Fig. 5 and 6.
The crystallographic data of 6–9 are summarized in the Exper-
imental section.
Fig. 4 ORTEP plot of a molecule of 4 (left) and 5 (right) showing 40% prob-
ability displacement ellipsoid. Hydrogen atoms are omitted for clarity. Selected
bond lengths [Å] and bonding angles [°]: For 4: B(1)–C(1) 1.612(3), B(1)–N(1)
1.689(2), B(1)–C(12) 1.641(2), B(1)–C(18) 1.641(2), N(1)–B(1)–C(1) 95.79(12),
N(1)–B(1)–C(12) 113.58(14), N(1)–B(1)–C(18) 110.69(13), C(1)–B(1)–C(12) 108.64(13),
C(1)–B(1)–C(18) 112.77(15), C(12)–B(1)–C(18) 114.01(13). For 5: B(1)–C(1) 1.578(2),
B(1)–N(1) 1.634(2), B(1)–Cl(1) 1.8664(18), B(1)–Cl(2) 1.8464(19), N(1)–B(1)–C(1)
99.56(12), N(1)–B(1)–Cl(1) 107.82(11), N(1)–B(1)–Cl(2) 111.49(11), C(1)–B(1)–Cl(1)
111.10(12), C(1)–B(1)–Cl(2) 117.30(12), Cl(1)–B(1)–Cl(2) 108.97(9).
We are fully aware about the presence of the chiral center
(at the C(7) atom) in the molecular structures of 6–9, but all
compounds crystallized in the centrosymmetric space group
P21/c as racemates.
applied also to compounds 1, 3–5. The %THC values in the
series 1 (79.0%), 3 (66.6%) and 4 (69.3%) indicate an
attenuation of the present N→B interaction in this set of
compounds. The %THC value for 5 of 73.8% is significantly
The molecular structures of 6–9 proved the formation of
benzene fused azaborole rings. The central ring systems are
essentially planar in all cases. Both boron B(1) and nitrogen
N(3) atoms adopt a planar environment consistent with the sp2
hybridization within the five-membered ring as the sum of the
angles around both atoms is close to the ideal value of 360°.
The B(1)–N(3) bond lengths in 6–9 are apparently shorter than
the sum of covalent radii for single bond Σcov(N,B) = 1.56 Å18
and correspond to the value for the respective double bond
(1.48 Å)18 (B(1)–N(3) bond distances are 1.431(3) Å for 6, 1.429
(4) Å for 7, 1.429(4) Å for 8, 1.422(4) Å for 9). This fact proves a
multiple character of these bonds and reflects strong π–π inter-
lower than for the corresponding dichloro-complex
1
suggesting that the N→B interaction is a bit stronger in the
case of 1. This is most probably caused in part by better donat-
ing properties of imino- (1) than amino- (5) functionality and
also reflects the presence of an sp3 nitrogen atom in 5 placing
both methyl groups right beside the chlorine atoms on the
boron atom (rather than between the boron substituents as in
the case of 1).
Reactivity of compounds 1, 2 and 5
The reaction of dichloroboranes 1, 2 and 5 with two equiva- action between both bonding partners. Similarly, the B(1)–N(2)
lents of lithium anilides ArNHLi (Ar = 2,6-Me2C6H3 or bond distances in 6–9 (1.401(3) Å for 6, 1.407(4) Å for 7,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6419

View Article Online
Paper
Dalton Transactions
and B(1)–N(2) distances are significantly shorter than the value
observed for intramolecular N→B interactions in the precursor
1 (1.602(4) Å). In contrast, the remaining nitrogen atoms N(1)
are bent away from the central plane in 6–9 reflecting the tetra-
hedral array at the C(7) carbon atom.
The solution structures of 6–9 were studied with the help of
multinuclear NMR spectroscopy and all obtained data are
summarized in Table 1. As all flanking NH–Ar or Ar groups
(Ar = 2,6-Me2C6H3 or 2,6-iPr2C6H3) present in the structures of
6–9 are non-equivalent, rather complicated sets of data were
observed (for full assignment see ESI†). Only one set of signals
was observed in all NMR spectra of 6–9, although they contain
a chiral center (at the C(7) atom as labeled in molecular struc-
tures or at the position 3 according to IUPAC nomenclature –
Table 1). All observed NMR data are entirely consistent with
molecular structures as obtained for 6–9 with the help of X-ray
1
diffraction analysis in the solid. The H NMR spectra revealed
Fig. 5 ORTEP plot of a molecule of 6 (up) and 7 (bottom) showing 40% prob-
ability displacement ellipsoid. Hydrogen atoms except NH and C(7)H groups are
omitted for clarity. Selected bond lengths [Å] and bonding angles [°]: For 6:
B(1)–C(1) 1.568(4), B(1)–N(3) 1.431(3), B(1)–N(2) 1.401(3), C(1)–C(2) 1.390(3),
C(2)–C(7) 1.510(3), C(7)–N(3) 1.460(3), C(1)–B(1)–N(2) 126.9(2), C(1)–B(1)–N(3)
105.9(2), N(2)–B(1)–N(3) 127.2(2). For 7: B(1)–C(1) 1.570(4), B(1)–N(3) 1.429(4),
B(1)–N(2) 1.407(4), C(1)–C(2) 1.397(4), C(2)–C(7) 1.513(4), C(7)–N(3) 1.455(4),
C(1)–B(1)–N(2) 127.2(2), C(1)–B(1)–N(3) 106.3(2), N(2)–B(1)–N(3) 126.6(2).
a typical signal for the aliphatic CH group in the position 3 of
the 1H-2,1-benzazaborole core (at 5.74 ppm for 6; 5.70 ppm for
7; 5.77 ppm for 8 and 5.64 ppm for 9).12 These values are sig-
nificantly upfield shifted in comparison with the imine CH
group in the starting compounds 1 and 2 (at 7.55 ppm for 1
and 8.35 for ppm 2). Another proof for the formation of
1H-2,1-benzazaborole cycles is the detection of doublets (at
2.96 ppm for 6; 3.26 ppm for 7; 3.41 ppm for 8 and 3.46 ppm
for 9) assigned to hydrogen atoms from NH groups directly
bonded to the CH groups in position 3 of 1H-2,1-benzaza-
borole cycles and also observation of singlets (at 4.04 ppm for
6; 4.33 ppm for 7; 3.78 ppm for 8; 3.96 ppm for 9) proving the
presence of the second NH group bonded to the boron atom.
Similarly, the 13C NMR spectra established the presence of a
CH group in the position 3 of the 1H-2,1-benzazaborole core
(at 75.9 ppm for 6; 78.7 ppm for 7; 80.7 ppm for 8 and
83.7 ppm for 9).12 We decided to measure 10B NMR spectra
since Király observed recently17 that the background signals
arising from the NMR probe and tube are practically invisible
in solution 10B NMR spectra, contrary to the more sensitive 11
B
NMR spectra. The 10B NMR spectra (see the Experimental
section) of 6–9 showed one signal in all cases (at 29.5 ppm for
6; 29.5 ppm for 7; 29.7 ppm for 8 and 30.0 ppm for 9), which
are shifted upfield (Δδ ∼ 23 ppm) in comparison to the start-
ing compounds 1 and 2 (at 6.5 ppm for 1; 6.8 ppm for 2)
reflecting a change of the hybridization of the boron atom
from sp3 to sp2.6a,d The 10B NMR chemical shifts in 6–9 are
consistent with 11B NMR chemical shifts of already prepared
1H-2,1-benzazaboroles with an N-substituent directly bonded
to the boron atom.8c The 15N NMR spectra together with
Fig. 6 ORTEP plot of a molecule of 9 (up) and 8 (bottom) showing 40% prob-
ability displacement ellipsoid. Hydrogen atoms except NH and C(7)H groups are
omitted for clarity. Selected bond lengths [Å] and bonding angles [°]: For 8:
B(1)–C(1) 1.561(3), B(1)–N(3) 1.429(4), B(1)–N(2) 1.409(3), C(1)–C(2) 1.405(4),
C(2)–C(7) 1.511(3), C(7)–N(3) 1.466(3), C(1)–B(1)–N(2) 130.7(2), C(1)–B(1)–N(3)
105.8(2), N(2)–B(1)–N(3) 123.5(2). For 9: B(1)–C(1) 1.562(4), B(1)–N(3) 1.422(4),
B(1)–N(2) 1.404(4), C(1)–C(2) 1.401(4), C(2)–C(7) 1.513(4), C(7)–N(3) 1.463(3),
C(1)–B(1)–N(2) 130.1(3), C(1)–B(1)–N(3) 105.6(2), N(2)–B(1)–N(3) 124.3(3).
1
1J(15N, H) coupling constants (Table 1) of 6–9 were measured
for completeness. Three signals of different nitrogen atoms
appeared in the 15N NMR spectra. Two of them showed coup-
ling with the hydrogen atom proving the presence of two NH
groups.
Reactivity of compounds 3 and 4
1.409(3) Å for 8, 1.404(4) Å for 9) point to a similar π–π inter-
action and N(2) atoms are coplanar with the ring system again As the reaction between 1 (2) and lithium anilides resulted in
reflecting sp2 hybridization at boron atoms B(1). Both B(1)–N(3) an unexpected addition of anilides across the CvN bond
6420 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 1 1H, 10B, 13C and 15N NMR data (δ = [ppm], J = [Hz]) of the central cores of 6–11 and 13 in C6D6 at 298 K with the numbering scheme of the 1H-2,1-benza-
zaborole cycles
Compound
Position
6
7
8
9
10
11
13
1/δ 10B
29.5
−268.4
5.74 (d)
29.5
−267.2
5.70 (d)
29.7
−280.3
5.77 (d)
30.0
−280.2
5.64 (d)
42.2
−223.7
5.92 (d)
42.2
−223.3
5.90 (d)
29.0
−324.6
5.38 (d)
2/δ 15N
3/δ 1H
(δ 13C)
(75.9)
(78.7)
(80.7)
(83.7)
(78.4)
(81.0)
(79.0)
3a/δ 13C
4/δ 1H
a
a
a
a
a
a
a
6.22 (d)
(121.9)
6.68 (td)
(128.3)
6.76 (td)
(127.6)
6.56 (dt)
(130.2)
152.6
4.04 (s)
(−302.0)
81.0
2.96 (d)
(−293.4)
70.7
6.12 (d)
(122.4)
6.60 (td)
(127.7)
6.70 (td)
(127.2)
6.41 (dt)
(131.1)
151.8
4.33 (s)
(−307.1)
80.4
3.26 (d)
(−292.8)
69.3
6.87 (d)
(122.9)
6.89 (m)
(129.8)
6.91 (m)
(128.5)
6.97 (dt)
(130.9)
154.8
3.78 (s)
(−307.6)
82.0
3.41 (d)
(−303.4)
72.3
6.57 (d)
(123.2)
6.77 (td)
(129.0)
6.81 (td)
(127.9)
6.59 (dt)
(132.0)
153.2
3.96 (s)
(−313.6)
81.2
3.46 (d)
(−300.8)
68.6
6.36 (d)
(121.8)
6.85 (td)
(129.2)
7.04 (td)
(128.1)
7.40 (dt)
(131.8)
152.7
—
—
—
3.09 (d)
(−300.1)
71.3
6.23 (d)
(122.3)
6.78 (td)
(128.8)
7.01 (td)
(128.0)
7.40 (dt)
(131.8)
151.9
—
—
—
3.42 (d)
(−297.6)
70.5
7.38 (d)
(123.5)
7.16 (m)
(129.8)
6.94 (td)
(127.7)
6.65 (dt)
(131.2)
155.9
3.77 (s)
(−315.0)
81.3
1.18 (d)
(−306.5)
75.6
(δ 13C)
5/δ 1H
(δ 13C)
6/δ 1H
(δ 13C)
7/δ 1H
(δ 13C)
7a/δ 13C
B–NH/δ 1H
(δ 15N)
1J (15N,1H)
CH–NH/δ 1H
(δ 15N)
1J (15N,1H)
a Signal of the carbon atom 3a was not observed. s = singlet, d = doublet, td = triplet of doublets, dt = doublet of triplets, m = multiplet.
yielding 1H-2,1-benzazaboroles, it seemed to be reasonable to
investigate similar reactions with compounds containing
different substituents at the central boron atom. Thus, the
reaction of 3 with one equivalent of lithium anilides ArNHLi
(Ar = 2,6-Me2C6H3 or 2,6-iPr2C6H3) was studied and proceeded
also as an addition of anilide giving 1H-2,1-benzazaboroles 10
and 11 (Scheme 3), similarly to analogous reactions of 1 and 2.
No reaction was obtained between 4 and ArNHLi even under
forcing reaction conditions using elevated temperatures and
an excess of ArNHLi (see further discussion). The 1H-2,1-
benzazaboroles 10 and 11 were observed after workup in
moderate yields (62% – 10, 61% – 11). Both compounds were
fully characterized using multinuclear NMR spectroscopy (see
Scheme 3 Preparation of 10 and 11.
Compound 10 crystallized as a racemate in the centrosym-
the Experimental section). The molecular structure of 10 was metric space group P21/c. The molecular structure of 10 is
unambiguously established with the help of X-ray diffraction closely related to those found for 6–9. The central ring system
analyses and the molecular structure is given in Fig. 7. The is again almost ideally planar with the significant π–π inter-
crystallographic data of 10 are given in the Experimental action between boron and nitrogen atoms within the five-
section.
membered azaborole ring as demonstrated by a rather short B
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6421

View Article Online
Paper
Dalton Transactions
Fig. 7 ORTEP plot of a molecule of 10 showing 40% probability displacement
ellipsoid. Hydrogen atoms except NH and C(7)H groups are omitted for clarity.
Selected bond lengths [Å] and bonding angles [°]: B(1)–C(1) 1.557(3), B(1)–N(1)
1.406(3), C(1)–C(2) 1.392(3), C(2)–C(7) 1.516(3), C(7)–N(1) 1.480(2), C(1)–B(1)–
N(1) 106.76(16), C(1)–B(1)–C(20) 123.05(20), C(1)–B(1)–C(20) 130.19(17).
(1)–N(1) bond distance of 1.406(3) Å. The second nitrogen
atom N(2) is again bent out from the central plane being co-
ordinated to the carbon atom C(7), while the ipso-carbon atom
C(20) of the phenyl ring remains in this plane reflecting the
sp2 hybridization of the boron atom.
The structure of 10, as determined in the solid state, is
retained in solution and also the structure of 11 is analogous,
as clearly established with the help of 1H, 10B, 13C and 15N
1
NMR spectroscopy. The situation in H and 13C NMR spectra
Scheme 4 Plausible mechanism for formation of studied 1H-2,1-benzaza-
boroles 7 and 13.
for 10 and 11 corresponds with the description made before
for 6–9. The closure of the 1H-2,1-benzazaborole cycle in 10
and 11 was confirmed by appearance of a doublet for aliphatic
CH groups in the position 3 of the 1H-2,1-benzazaborole cycle
of 1H-2,1-benzazaboroles 6–11. A similar reaction pathway was
discovered by Młynarz et al., who reported an addition of
diethyl phosphite across the CvN double bond in structurally
related boronic acids.13 In the case of the formation of 6–9, the
whole reaction is then terminated by the substitution of the
remaining chlorine atom at the boron center with the second
equivalent of the lithium anilide (see Scheme 4, where the
mechanism is illustrated for the reaction between 1 and 2,6-
iPr2C6H3NHLi). Interestingly, compound L3BCl2 (5), which is
void of any CvN double bond, is inert to ArNHLi under the
same reaction conditions. This fact supports our presumption
that the nucleophilic attack of the CvN double bond is the
first step in the reaction of 1–3 with ArNHLi (Scheme 4).
Although many attempts were performed to isolate the
intermediate 12 especially using the reaction between 1 and
ArNHLi (Ar = 2,6-iPr2C6H3) in a 1 : 1 molar ratio in various sol-
vents and at different temperatures majority (see further dis-
cussion) of them failed and only the cyclized product 7
together with one half of unreacted starting material 1 were
isolated and characterized with the help of multinuclear NMR
spectroscopy. The first step is formation of intermediate 12
1
in the H NMR spectra (at 5.92 ppm for 10 and 5.90 ppm for
11) and furthermore the doublet of NH groups directly bonded
to this CH group was detected for both compounds (at
3.09 ppm for 10 and 3.42 ppm for 11). Similarly, the presence
of a CH group was corroborated by the 13C NMR spectra. The
10B NMR spectra also showed one signal (at 42.2 ppm for 10
and 42.2 ppm for 11). Values of chemical shifts are about
∼10 ppm downfield shifted in comparison with the values for
6–9. The boron atoms in 10 and 11 exhibit lower electron den-
sity6a,d caused by a B-bonded phenyl group in comparison to
6–9 where π–π interaction between boron and nitrogen atoms
leads to an increase of the electron density of the boron atoms.
Nevertheless, these values are comparable to those found in
analogous B-phenyl substituted 1H-2,1-benzazaboroles.12 The
15N NMR spectra showed only 2 signals (Table 1) for each com-
pound, and one of them corresponds to the NH group, which
is consistent with the proposed structure.
Unexpected addition reaction
Recently, we have discovered a similar cyclization procedure in that contains cyclic systems with the B–Cl bond. This chlorine
a diorganoantimony compound containing ligand L2. In this atom is most probably much more reactive than the chlorine
case, the first step was the formation of an antimony hydride, atoms or the CHvN bond in the starting material 1 and
which then rearranged to the cyclized compound via hydrogen smoothly reacts with the second molecule of the lithium
migration.5c In contrast, the attack of the CvN double bond anilide (of the starting one equivalent) to give final product 7
(amidolithiation) in the structure of starting compounds 1–3 leaving half an equivalent of the starting 1 unreacted in the
with subsequent elimination of lithium chloride resulting in reaction mixture (established by NMR spectroscopy). It is note-
the ring closure and formation of an intermediate 12 accord- worthy that the boron atom changes its hybridization from
ing to Scheme 4 is a more probable mechanism for formation formally sp3 in the starting 1 to sp2 after cyclization to 12.
6422 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
The high reactivity of similar B–Cl bonds towards lithium
The presence of an effective N→B intramolecular inter-
reagents in analogous five-membered azaborole systems and action in the structure of starting boranes 1–3 seems to be also
their low stability were documented in the literature7b and a very important point, which should be taken into account,
thus it supports our suggestions of the mechanism illustrated when considering the reactivity of 1–3. This interaction is most
in Scheme 4.
probably crucial for the activation of the CvN double bond
Interestingly, when the reaction between 1 and ArNHLi toward the amidolithiation. To test this idea, the reaction of
(Ar = 2,6-iPr2C6H3) in a 1 : 1 molar ratio was performed at low the non-substituted ligand L1H with one equivalent or an
temperatures (−80 °C) in toluene second product 13 besides 7 excess (2.5 equiv.) of ArNHLi was performed, but no addition
was detected in the reaction mixture and was isolated after of the anilide across the CvN double bond was observed even
workup as a pure compound (Scheme 4). Compound 13 again after prolonged heating of the reaction mixtures and only the
represents the product of the addition reaction between 1 and starting material L1H was isolated.
2,6-iPr2C6H3NHLi (although the starting compounds were
Similarly, compound 4, although containing ligand L1 and
mixed in 1 : 1 ratio, it means that starting 1 was detected as a weaker but still significant N→B interaction as demonstrated
by-product). In this case however, the position of the nitrogen by the %THC value, did not react with ArNHLi. This fact may
substituents is different in comparison to the analogue 7. The be ascribed mainly to the lack of any chlorine atom in the
plausible explanation for formation of 13 is illustrated in structure of 4 that prevents elimination of lithium chloride
Scheme 4.21 This mechanism may involve a lithium and hydro- and closure of the azaborole ring.
gen atom migration after the addition of the anilide across the
CvN bond leading to an intermediate 12′ after elimination of
lithium chloride. 12′ then reacts with the second equivalent of
ArNHLi under formation of 13. Presumably, the lithium con-
Conclusions
In conclusion, we have discovered an unexpected one pot pro-
cedure for preparation of 1,2,3-trisubstituted 1H-2,1-benzaza-
boroles. The nucleophilic addition of the lithium anilides to
the CvN double bond in the structure of the ligand, which is
activated by the intramolecular N→B interaction, is the crucial
step of the whole reaction. Further investigations focusing on
such type of addition reactions in the case of similar intra-
molecularly coordinated organometallic compounds using
various types of nucleophiles as well as central metal atoms
are currently under study in our labs.
taining addition product is stable enough only at low tempera-
tures and, thus, has enough time to rearrange before lithium
chloride elimination and this allows formation of 12′ and sub-
sequently 13.21 This presumption is supported by the fact that
no evidence for formation of 13 was observed, when the reac-
tion was performed at ambient temperature.
Compound 13 was unambiguously characterized with the
help of elemental analysis and 1H, 10B, 13C and 15N NMR spec-
troscopy (see the Experimental section). The molecular struc-
ture of 13 was determined with the help of single crystal X-ray
diffraction analyses and is depicted in Fig. 8 together with
selected structural parameters given in the figure captions. It
is evident that the position (exocyclic) of the tBu group in 13 is
different in comparison to 7, where the tBu moiety is incorpor-
ated within the azaborole ring. The basic structural features of
the molecular structure of 13 are similar to the structurally
related analogues 6–9 and its structure will not be discussed
here in more detail.
Experimental
General procedures
All air and moisture sensitive manipulations were carried out
under an argon atmosphere using standard Schlenk tube tech-
niques. All solvents were dried using a Pure Solv–Innovative
Technology equipment. The starting compounds: BCl3 (1 M
solution in hexane), C6H5BCl2 (97%), Cy2BCl (1 M solution in
hexane), were obtained from the commercial suppliers and
used as delivered. The ligands L1,2 were prepared according to
published procedures22 or by analogous procedures. Com-
pound 5 was synthesized according to the literature.16
NMR spectroscopy
1H, 10B, 13C and 15N NMR spectra were recorded on Bruker 500
Avance or Bruker 400 MHz spectrometers, using a 5 mm
1
tunable broad-band probe. Appropriate chemical shifts in H
and 13C NMR spectra were related to the residual signals of the
solvent (CDCl3: δ(1H) = 7.27 ppm and δ(13C) = 77.23 ppm,
C6D6: δ(1H) = 7.16 ppm and δ(13C) = 128.39 ppm), 10B NMR
spectra were related to external standard B(OMe)3 (δ(10B) =
18.1 ppm), 15N NMR spectra were related to external neat nitro-
methane (δ(15N) = 0.0 ppm). For compounds 1–4, 6–11 and
Fig. 8 ORTEP plot of a molecule of 13 showing 40% probability displacement
ellipsoid. Hydrogen atoms except NH and C(7)H groups are omitted for clarity.
Selected bond lengths [Å] and bonding angles [°]: B(1)–C(1) 1.564(4), B(1)–N(2)
1.409(3), B(1)–N(3) 1.420(3), C(1)–C(2) 1.402(3), C(2)–C(7) 1.521(4), C(7)–N(3)
1.481(3), C(1)–B(1)–N(3) 105.7(2), C(1)–B(1)–N(2) 129.5(2), N(2)–B(1)–N(3)
124.8(2).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6423

View Article Online
Paper
Dalton Transactions
1321 the full assignment of all signals in all measured NMR 2 × 10 mL of hexane. The obtained insoluble material was
spectra was managed with the help of various techniques dried in vacuo and extracted by dichloromethane (3 × 25 mL).
including 1H, 13C{1H} APT, 1H–1H COSY, 1H–13C HMQC and The resulting suspension was filtered and the filtrate was eva-
1H–13C HMBC. We decided to measure the 10B isotope instead porated in vacuo to give 2 as a pale ginger crystalline solid
of conventional measurement of the boron isotope 11B, (isolated yield 1.86 g, 75%), decomp. p. 132 °C.
because of appearance of background-free signals caused by
Anal. calc. for C19H22BCl2N (MW 346.10): C, 65.9; H, 6.4;
an unusual isotopic effect on the signal to background Found: C, 66.1; H, 6.3%. 1H NMR (500 MHz, CDCl3): δ 1.00
ratio.17 Nevertheless, the chemical shift of the nuclei is not (6H, d, (CH3)2CH), 1.25 (6H, d, (CH3)2CH), 3.09 (2H, sep, 2 ×
influenced and is essentially the same regardless of 10B or 11B (CH3)2CH), 7.17 (1H, d, Ar-H), 7.23 (1H, dd, Ar-H), 7.34 (1H,
NMR spectra are acquired. 15N NMR chemical shifts were dd, Ar-H), 7.40 (1H, td, Ar-H), 7.67 (2H, m, Ar-H), 7.80 (1H, dd,
obtained from 1H–15N HMBC spectra. 1J(15N, 1H) coupling Ar-H), 8.35 (1H, s, CHvN). 13C{1H} NMR (125.76 MHz, CDCl3):
constants of present NH groups were measured by 1D δ 22.9 ((CH3)2CH), 26.8 ((CH3)2CH), 29.1 (2 × (CH3)2CH), 124.4,
1H–15N gs HMQC. Table 1 summarizes all NMR data of the 124.5, 127.7, 128.5, 129.6, 129.8, 133.9, 136.1, 142.8, 144.5
central cores for 6–11 including the multiplicities of the (Ar-C), 173.4 (CHvN), (Ar-C1-B) not observed. 10B NMR
1
signals in the H NMR spectra. Thus, these NMR data are not (42.99 MHz, CDCl3): δ 6.8.
again repeated in the syntheses description for respective com-
Synthesis of [o-(CHvNtBu)C6H4]B(C6H5)Cl (3). nBuLi
pounds. The full assignment including assignment of all (22.0 mL of 1.6 M solution in hexane, 35.2 mmol) was added
bonded aryl/alkyl groups bonded to central cores are available to a solution of L1Br (8.46 g, 35.2 mmol) in hexane (80 mL) at
in the ESI.†
0 °C and stirred for 1 h at this temperature. Additional hexane
(80 mL) was added to the resulting yellow-orange suspension
of the lithium compound. A solution of C6H5BCl2 (5.60 g,
Syntheses
Synthesis of [o-(CHvNtBu)C6H4]BCl2 (1). nBuLi (6.3 mL of 35.2 mmol) in hexane (10 mL) was added under intensive stir-
1.6 M solution in hexane, 10.1 mmol) was added to a solution ring to the suspension of lithium compound at 0 °C. The
of L1Br (2.43 g, 10.1 mmol) in hexane (20 mL) at 0 °C and obtained ivory mixture was allowed to reach r.t. and stirred
stirred for 1 h at this temperature. Additional hexane (50 mL) overnight. The mixture was filtered, the filtrate was discarded
was added to the resulting yellow-orange suspension of the and the remaining insoluble material was washed with 2 ×
lithium compound L1Li. A 1 M solution of BCl3 in hexane 10 mL hexane. The obtained insoluble material was dried
(10.1 mL, 10.1 mmol) was added under intensive stirring to in vacuo and extracted by dichloromethane (3 × 25 mL). The
the suspension of the lithium compound at 0 °C. The obtained resulting suspension was filtered and the filtrate was evapor-
cream mixture was allowed to reach r.t. and stirred overnight. ated in vacuo to give 3 as an ivory crystalline solid (isolated
The mixture was filtered, the filtrate was discarded and the yield 8.46 g, 85%), decomp. p. 165 °C.
remaining insoluble material was washed by 2 × 10 mL of
Anal. calc. for C17H19BClN (MW 283.60): C, 72.0; H, 6.8;
hexane. The obtained insoluble material was dried in vacuo Found: C, 72.2; H, 6.9%. 1H NMR (500 MHz, C6D6): δ 1.09 (9H,
and extracted by dichloromethane (3 × 25 mL). The resulting s, (CH3)3C), 6.94 (1H, td, Ar-H), 7.10 (1H, td, Ar-H), 7.16 (2H,
suspension was filtered and the filtrate was evaporated m, Ar-H), 7.25 (2H, td, Ar-H), 7.50 (1H, d, Ar-H), 6.67 (2H, d,
in vacuo to give 1 as a pale ginger crystalline solid (isolated Ar-H), 7.84 (1H, s, CHvN). 13C{1H} NMR (125.76 MHz, C6D6):
yield 1.99 g, 82%), decomp. p. 178 °C.
δ 31.1 ((CH3)3C), 61.3 ((CH3)3C), 126.2, 127.0, 127.2, 128.2,
Anal. calc. for C11H14BCl2N (MW 241.95): C, 54.6; H, 5.8; 130.3, 132.7, 134.4, 136.4 (Ar-C), 143.6 (Ph-C1-B), 164.7
Found: C, 54.4; H, 5.6%. 1H NMR (500 MHz, C6D6): δ 1.32 (9H, (Ar-C1-B), 169.2 (CHvN). 10B NMR (42.99 MHz, C6D6): δ 6.1.
s, (CH3)3C), 6.92 (2H, m, Ar-H), 7.16 (1H, td, Ar-H), 7.55 (1H, s, 15N NMR (40.55 MHz, C6D6): δ −115.4 (CHvN).
CHvN), 7.83 (1H, d, Ar-H). 13C{1H} NMR (125.76 MHz, C6D6):
Synthesis of [o-(CHvNtBu)C6H4]B(C6H11)2 (4). nBuLi
δ 30.9 ((CH3)3C), 62.0 ((CH3)3C), 126.5, 128.0, 129.7, 134.6, (7.8 mL of 1.6 M solution in hexane, 12.5 mmol) was added to
135.3 (Ar-C), 169.3 (CHvN), ((Ar-C1-B) not observed. 10B NMR a solution of L1Br (3.00 g, 12.5 mmol) in hexane (40 mL) at
(42.99 MHz, C6D6): δ 6.5. 15N NMR (40.55 MHz, C6D6): 0 °C and stirred for 1 h at this temperature. Additional hexane
δ −123.9 (CHvN).
(50 mL) was added to the resulting yellow-orange suspension
Synthesis of [o-(CHvN-2,6-iPr2C6H3)C6H4]BCl2 (2). nBuLi of the lithium compound. A 1 M solution of (C6H11)2BCl in
(4.5 mL of 1.6 M solution in hexane, 7.2 mmol) was added to a hexane (12.5 mL, 12.5 mmol) was added under intensive stir-
solution of L2Br (2.47 g, 7.2 mmol) in hexane (20 mL) at 0 °C ring to the suspension of the lithium compound at 0 °C. The
and stirred for 1 h at this temperature. Additional hexane obtained ivory mixture was allowed to reach r.t. and stirred
(30 mL) was added to the resulting yellow-orange suspension overnight. The mixture was filtered and the filtrate was stored
of the lithium compound L2Li. A 1 M solution of BCl3 in at −30 °C to give the first crop of 4. The remaining insoluble
hexane (7.2 mL, 7.2 mmol) was added under intensive stirring material after reaction was dried in vacuo and extracted by
to the suspension of the lithium compound at 0 °C. The dichloromethane (3 × 30 mL). The resulting suspension was
obtained ginger mixture was allowed to reach r.t. and stirred filtered and the filtrate was evaporated in vacuo to give the
overnight. The mixture was filtered, the filtrate was discarded second crop of 4 as a white crystalline solid (isolated yield
and the remaining insoluble material was washed with 3.25 g, 77%), m.p. 157 °C.
6424 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Anal. calc. for C23H36BN (MW 337.35): C, 81.9; H, 10.8; the minimal amount of hot hexane to give 8 as colorless single
1
Found: C, 82.1; H, 10.6%. H NMR (500 MHz, C6D6): 0.78 (2H, crystals (isolated yield 499 mg, 66%), m.p. 213 °C.
qd, Cy-H), δ 1.17 (9H, s, (CH3)3C), 1.23–1.47 (12H, m, Cy-H),
Anal. calc. for C35H42BN3 (MW 515.54): C, 81.5; H, 8.2;
1.77 (2H, dd, Cy-H), 1.83 (2H, dd, Cy-H), 1.93 (2H, dd, Cy-H), Found: C, 81.3; H, 8.0%. For NMR data of the core see Table 1
2.04 (2H, d, Cy-H), 7.10 (1H, td, Ar-H), 7.31 (2H, m, Ar-H), 7.74 and for complete assignment see the ESI.†
(1H, d, Ar-H), 7.91 (1H, s, CHvN). 13C{1H} NMR (125.76 MHz,
Synthesis of racemic 1,3-(NH-2,6-iPr2C6H3)2-2-(2,6-iPr2C6H3)-
C6D6): δ 28.7, 29.9, 30.4 (Cy-C), 31.1 ((CH3)3C), 32.0, 32.2 1H-2,1-benzazaborole (9). Compound 9 was prepared analo-
(Cy-C), 33.0 (Cy-C1-B), 59.7 ((CH3)3C), 125.0, 125.7, 130.5, gously to the procedure described for 8. nBuLi (2.2 mL of
131.1, 138.3 (Ar-C), 165.7 (CHvN), 170.5 (Ar-C1-B). 10B NMR 1.6 M solution in hexane, 3.6 mmol) and 2,6-diisopropyl-
(42.99 MHz, C6D6): δ 8.0. 15N NMR (40.55 MHz, C6D6): aniline (636 mg, 3.6 mmol) gave corresponding lithium
δ −109.6 (CHvN).
Synthesis of racemic 1,3-(NH-2,6-Me2C6H3)2-2-tBu-1H-2,1- less single crystals (isolated yield 623 mg, 55%), m.p. 221 °C.
benzazaborole (6). nBuLi (4.1 mL of 1.6 M solution in hexane, Anal. calc. for C43H58BN3 (MW 627.75): C, 82.3; H, 9.3;
anilide that was reacted with 2 (620 mg, 1.8 mmol). 9 are color-
6.6 mmol) was added to a solution of 2,6-dimethylaniline Found: C, 82.4; H, 9.5%. For NMR data of the core see Table 1
(803 mg, 6.6 mmol) in hexane (20 mL) at 0 °C. The obtained and for complete assignment see the ESI.†
white suspension of lithium anilide was stirred for 30 min at
Synthesis of racemic 1-C6H5-2-tBu-3-(NH-2,6-Me2C6H3)-1H-
0 °C. A finely divided suspension of 1 (802 mg, 3.3 mmol) in 2,1-benzazaborole (10). nBuLi (1.8 mL of 1.6 M solution in
hexane (40 mL) (1 was partially dissolved under intensive stir- hexane, 2.9 mmol) was added to a solution of 2,6-dimethyl-
ring in boiling hexane, then cooled to 0 °C) was added to the aniline (356 mg, 2.9 mmol) in hexane (20 mL) at 0 °C. The
suspension of the lithium anilide at 0 °C. After 15 min the obtained white suspension of lithium anilide was stirred for
reaction mixture was allowed to reach r.t. and stirred over- 30 min at 0 °C and after that it was added to a suspension of 3
night. The resulting ivory suspension was filtered to give a pale (833 mg, 2.9 mmol) in hexane (40 mL) at 0 °C. The resulting
yellow filtrate and the insoluble material was extracted by mixture was stirred for 15 min at 0 °C and then was allowed to
additional hexane (10 mL). The filtrate volume was concen- reach r.t. and stirred overnight. The resulting pale yellow sus-
trated to incipient crystallization and slow cooling to 5 °C gave pension was filtered to give a pale yellow filtrate and the in-
colorless crystals of
m.p. 134 °C.
6
(isolated yield 932 mg, 68%), soluble material was extracted by additional hexane (10 mL).
The filtrate volume was concentrated to incipient crystalliza-
Anal. calc. for C27H34BN3 (MW 411.39): C, 78.8; H, 8.3; tion and slow cooling to 5 °C gave colorless crystals of 10
Found: C, 80.0; H, 8.5%. For NMR data of the core see Table 1 (isolated yield 667 mg, 62%), m.p. 106 °C.
and for complete assignment see the ESI.†
Anal. calc. for C25H29BN2 (MW 368.32): C, 81.5; H, 7.9;
Synthesis of racemic 1,3-(NH-2,6-iPr2C6H3)2-2-tBu-1H-2,1- Found: C, 81.2; H, 7.7%. For NMR data of the core see Table 1
benzazaborole (7). Compound 7 was prepared analogously to and for complete assignment see the ESI.†
the procedure described for 6. nBuLi (3.0 mL of 1.6 M solution
Synthesis of racemic 1-C6H5-2-tBu-3-(NH-2,6-iPr2C6H3)-1H-
in hexane, 4.8 mmol) and 2,6-diisopropylaniline (851 mg, 2,1-benzazaborole (11). Compound 11 was prepared analo-
4.8 mmol) gave corresponding lithium anilide that was reacted gously to the procedure described for 10. nBuLi (1.2 mL of 1.6
with 1 (581 mg, 2.4 mmol). 7 was obtained in the form of col- M solution in hexane, 1.9 mmol) and 2,6-diisopropylaniline
orless single crystals (isolated yield 1.01 g, 81%), m.p. 118 °C.
(343 mg, 1.9 mmol) gave corresponding lithium anilide that
Anal. calc. for C35H50BN3 (MW 523.60): C, 80.3; H, 9.6; was reacted with 3 (549 mg, 1.9 mmol). Isolated yield 501 mg,
Found: C, 80.5; H, 9.2%. For NMR data of the core see Table 1 61%. Neither X-ray diffraction analysis nor melting point could
and for complete assignment see the ESI.†
Synthesis of racemic 1,3-(NH-2,6-Me2C6H3)2-2-(2,6- observed just as honey-like material that decomposes slowly
iPr2C6H3)-1H-2,1-benzazaborole (8). nBuLi (1.8 mL of 1.6 M over time.
be obtained, because any attempt to crystallize 11 failed. It was
solution in hexane, 2.9 mmol) was added to a solution of 2,6-
Anal. calc. for C29H37BN2 (MW 424.43): C, 82.1; H, 8.8;
dimethylaniline (357 mg, 2.9 mmol) in hexane (20 mL) at 0 °C. Found: C, 79.9; H, 8.9%. For NMR data of the core see Table 1
The obtained white suspension of lithium anilide was stirred and for complete assignment see the ESI.†
for 30 min at 0 °C. A finely divided suspension of 2 (510 mg,
Synthesis of racemic 1-(NH-2,6-iPr2C6H3)-2-(2,6-iPr2C6H3)-3-
1.5 mmol) in hexane (40 mL) (2 was partially dissolved under (NH-tBu)-1H-2,1-benzazaborole (13). nBuLi (2.9 mL of 1.6 M
intensive stirring in boiling hexane, then cooled to 0 °C) was solution in hexane, 4.7 mmol) was added to a solution of 2,6-
added to the suspension of lithium anilide at 0 °C. After diisopropylaniline (835 mg, 4.7 mmol) in hexane (30 mL) at
15 min the reaction mixture was allowed to reach r.t. and 0 °C. The obtained white suspension of lithium anilide was
stirred overnight. The resulting ivory suspension was decanted stirred for 30 min at 0 °C and after that was added to a pre-
and the solution was discarded. The white insoluble residue cooled suspension of 1 (1.14 g, 4.7 mmol) in toluene (40 mL)
was washed with hexane (5 mL) and then extracted with at −80 °C. The resulting mixture was stirred for 30 min at
benzene (30 mL). The obtained suspension was filtered to give −80 °C and then was allowed to reach r.t. and stirred over-
a pale yellow filtrate. The filtrate was evaporated in vacuo to night. The resulting ginger suspension was filtered to give a
dryness. The obtained crude product was recrystallized from ginger filtrate, which was evaporated in vacuo to give pale
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6425

View Article Online
Paper
Dalton Transactions
ginger-green honey-like matter. This matter was treated with
Crystallographic data for 5. C9H12BCl2N, M = 215.91, ortho-
hexane (10 mL) leading to formation of a white precipitate that rhombic, P212121, a = 9.4390(4), b = 9.5611(5), c = 11.6760(3) Å,
was isolated. The obtained white solid (containing both 1 and V = 1053.73(8) Å3, Z = 4, T = 150(1) K, 9643 total reflections,
13) was extracted with the mixture of hexane (30 mL) and 2404 independent (Rint = 0.023, R1 (obs. data) = 0.027, wR2 (all
toluene (10 mL). The obtained pale green suspension was fil- data) 0.062), S = 1.122, Δρ, max., min. [e Å−3] 0.181, −0.216,
tered, the residue was discarded and the filtrate was evapor- CCDC 908867.
ated in vacuo to give a pale yellow powder. The obtained
Crystallographic data for 6. C27H34BN3, M = 411.38, mono-
powder was recrystallized from boiling hexane (35 mL) to give clinic, P21/c, a = 9.8030(5), b = 17.1271(11), c = 15.1920(12) Å,
13 as colorless crystals (isolated yield 602 mg, 49% according β = 108.461(6), V = 2419.4(3) Å3, Z = 4, T = 150(1) K, 17 290 total
to 2,6-diisopropylaniline), m.p. 187 °C. 13 is stable for a long reflections, 5423 independent (Rint = 0.047, R1 (obs. data) =
time (several months) in an argon atmosphere.
0.067, wR2 (all data) 0.130), S = 1.185, Δρ, max., min. [e Å−3
]
Anal. calc. for C35H50BN3 (MW 523.60): C, 80.3; H, 9.6; 0.163, −0.209, CCDC 908866.
Found: C, 80.4; H, 9.5%. For NMR data of the core see Table 1
and for complete assignment see the ESI.†
Crystallographic data for 7. C35H50BN3, M = 523.59, mono-
clinic, P21/c, a = 13.4110(10), b = 10.3260(9), c = 26.2331(18) Å,
β = 115.566(6), V = 3277.1(5) Å3, Z = 4, T = 150(1) K, 19 620 total
reflections, 6170 independent (Rint = 0.070, R1 (obs. data) =
X-ray crystallography
The suitable single crystals were mounted on a glass fibre with 0.078, wR2 (all data) 0.159), S = 1.109, Δρ, max., min. [e Å−3
an oil and measured on a four-circle diffractometer KappaCCD 0.254, −0.332, CCDC 908868.
]
with a CCD area detector by monochromatized MoKα radiation
Crystallographic data for 8. C35H42BN3, M = 515.53, mono-
(λ = 0.71073 Å) at 150(1) K. The numerical23 absorption correc- clinic, P21/c, a = 12.5801(13), b = 15.2930(14), c = 19.0710(19) Å,
tions from the crystal shape were applied for all crystals. The β = 125.552(7), V = 2985.1(6) Å3, Z = 4, T = 150(1) K, 23 236 total
structures were solved by the direct method (SIR92)24 and reflections, 6729 independent (Rint = 0.054, R1 (obs. data) =
refined by a full matrix least squares procedure based on F2 0.064, wR2 (all data) 0.118), S = 1.161, Δρ, max., min. [e Å−3
]
(SHELXL97).25 Hydrogen atoms were fixed into idealized posi- 0.303, −0.242, CCDC 908865.
tions (riding model) and assigned temperature factors Hiso
Crystallographic data for 9. C43H58BN3, M = 627.73, mono-
(H) = 1.2 Ueq (pivot atom) or of 1.5 Ueq for the methyl moiety clinic, P21/c, a = 10.2110(13), b = 21.1459(17), c = 18.6160(13) Å,
with C–H = 0.96, 0.97, and 0.93 Å for methyl, methylene, and β = 106.805(8), V = 3847.9(7) Å3, Z = 4, T = 150(1) K, 30 030 total
hydrogen atoms in the aromatic ring, respectively. The tBu reflections, 8550 independent (Rint = 0.067, R1 (obs. data) =
groups in 1 are disordered along the special position where a 0.077, wR2 (all data) 0.139), S = 1.175, Δρ, max., min. [e Å−3
C9 methyl atom remains on the same place. This disorder has 0.507, −0.557, CCDC 908864.
]
been treated by splitting of C8 ipso carbon – the parent carbon
Crystallographic data for 10. C25H29BN2, M = 368.31, mono-
atom of the tBu group and a C10 methyl group into two equi- clinic, P21/c, a = 10.3891(9), b = 10.2250(4), c = 20.5820(15) Å,
valent positions while C10a is symmetry related. All the crystal β = 102.468(7), V = 2134.8(3) Å3, Z = 4, T = 150(1) K, 19 153 total
structures of studied compounds revealed a bit higher wR2 reflections, 4751 independent (Rint = 0.039, R1 (obs. data) =
parameters, which is caused by very limiting diffraction limits 0.058, wR2 (all data) 0.115), S = 1.168, Δρ, max., min. [e Å−3
of even very large crystals. Crystallographic data for structural 0.172, −0.206, CCDC 908870.
]
analysis has been deposited with the Cambridge Crystallo-
Crystallographic data for 13. C35H50BN3, M = 523.59, mono-
graphic Data Centre, CCDC nos. 801796–801806.
clinic, P21/c, a = 8.9470(6), b = 20.765(2), c = 17.1239(15) Å, β =
Crystallographic data for 1. C11H14BCl2N, M = 241.94, 98.560(7), V = 3145.9(5) Å3, Z = 4, T = 150(1) K, 28 078 total
orthorhombic, Pnma, a = 13.5840(3), b = 6.9169(7), c = 12.8092(7) reflections, 6984 independent (Rint = 0.045, R1 (obs. data) =
Å, V = 1203.54(14) Å3, Z = 4, T = 150(2) K, 7481 total reflections, 0.076, wR2 (all data) 0.159), S = 1.097, Δρ, max., min. [e Å−3
1316 independent (Rint = 0.023, R1 (obs. data) = 0.044, wR2 (all 1.112, −0.335, CCDC 908871.
data) 0.104), S = 1.285, Δρ, max., min. [e Å−3] 0.247, −0.392,
]
CCDC 908869.
Crystallographic data for 3. C17H19BClN, M = 283.59, ortho-
rhombic, P212121, a = 8.8400(7), b = 11.2160(7), c = 15.6841(12)
Acknowledgements
Å, V = 1555.1(2) Å3, Z = 4, T = 150(1) K, 13 296 total reflections, The authors thank the Grant agency of the Czech Republic
3504 independent (Rint = 0.037, R1 (obs. data) = 0.039, wR2 (all project no. P207/11/0223.
data) 0.095), S = 1.110, Δρ, max., min. [e Å−3] 0.411, −0.256,
CCDC 908872.
Crystallographic data for 4. C23H36BN, M = 337.34, ortho-
rhombic, Pbca, a = 8.3280(6), b = 15.7381(16), c = 30.885(2) Å,
Notes and references
V = 4048.0(6) Å3, Z = 8, T = 150(1) K, 25 505 total reflections,
4398 independent (Rint = 0.041, R1 (obs. data) = 0.059, wR2 (all
data) 0.114), S = 1.166, Δρ, max., min. [e Å−3] 0.254, −0.214,
CCDC 908873.
1 C. Fedorchuk, M. Copsey and T. Chivers, Coord. Chem. Rev.,
2007, 251, 897.
2 For recent reviews see: (a) F. Edelmann, Adv. Organomet.
Chem., 2008, 57, 183; (b) M. Asay, C. Jones and M. Driess,
6426 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Chem. Rev., 2011, 221, 354; (c) S. Collins, Coord. Chem. Rev.,
2011, 255, 118; (d) A. A. Mohamed, Coord. Chem. Rev.,
2010, 254, 1918.
(f) G. Schmidt and M. Schütz, J. Organomet. Chem., 1995,
492, 185; (g) S. Liu, M.-C. Lo and G. C. Fu, Tetrahedron,
2006, 62, 11343; (h) X. Fang, Y. Deng, Q. Xie and
F. Moingeon, Organometallics, 2008, 27, 2892.
8 (a) S. Nagy, R. Krishnamatti and B. P. Etherton, U.S. Patent
6,228,959, 2001; S. Nagy, R. Krishnamatti and
B. P. Etherton, Chem. Abstr., 1997, 126, 19432j; (b) Q. Wang,
P. Zoricak and X. Gao, Can. Patent Appl. 2,225,014, 1999;
Q. Wang, P. Zoricak and X. Gao, Chem. Abstr., 1999, 142,
219701; (c) M. Yamashita, Y. Aramaki and K. Nozaki, New
J. Chem., 2010, 34, 1774.
3 For several examples see: (a) T. Chivers, C. Fedorchuk
and M. Parvez, Inorg. Chem., 2004, 43, 2643;
(b) T. Chivers, C. Fedorchuk, G. Schatte and M. Parvez,
Inorg. Chem., 2003, 42, 2084; (c) D. Fest, C. D. Habben,
A. Meller, G. M. Sheldrick, D. Stalke and F. Pauer, Chem.
Ber., 1990, 123, 703; (d) T. Albrecht, G. Elter,
M. Noltemeyer and A. Meller, Z. Anorg. Allg. Chem., 1998,
624, 1514; (e) A. Heine, D. Fest, D. Stalke, C. D. Habben,
A. Meller and G. M. Sheldrick, J. Chem. Soc., Chem.
Commun., 1990, 742; (f) T. Chivers, D. J. Eisler,
9 (a) V. Lee, et al., U.S. Patent 7,465,836, 2008; (b) V. Lee,
et al., U.S. Patent 8,106,031, 2012.
C. Fedorchuk, G. Schatte, H. M. Tuononen and 10 (a) R. Köster, K. Iwasaki, S. Hattori and Y. Morita,
R. T. Boer’e, Inorg. Chem., 2006, 45, 2119; (g) T. Chivers,
X. Gao and M. Parvez, Angew. Chem., Int. Ed. Engl., 1995,
34, 2549; (h) H.-J. Koch, H.W. Roesky, S. Besser and
Liebigs Ann. Chem., 1968, 720, 23; (b) R. Köster and
K. Iwasaki, Advances in Chemistry, 1964, Chapter 16,
vol. 42, p. 148.
R. Herbst-Irmer, Chem. Ber., 1993, 126, 571; 11 (a) P. T. Hawkins and A. U. Blackham, J. Org. Chem., 1967,
(i) D. R. Manke and D. G. Nocera, Inorg. Chem., 2003, 42,
4431; ( j) D. R. Manke and D. G. Nocera, Inorg. Chim.
Acta, 2003, 345, 235; (k) D. R. Manke, Z.-H. Loh and
D. G. Nocera, Inorg. Chem., 2004, 43, 3618.
32, 597; (b) H. E. Dunn, J. C. Catlin and H. R. Snyder,
J. Org. Chem., 1968, 33, 4483; (c) M. Lauer and G. Wulff,
J. Organomet. Chem., 1983, 256, 1.
12 A. M. Genaev, S. M. Nagy, G. E. Salnikov and V. G. Shubin,
Chem. Commun., 2000, 1587.
4 (a) T. Chivers, D. J. Eisler, C. Fedorchuk, G. Schatte,
H. M. Tuononen and R. T. Boeré, Inorg. Chem., 2006, 45, 13 A. Rydzewska, K. Slepokura, T. Lis, P. Kafarski and
2119; (b) J. Konu, H. M. Tuononen, T. Chivers, P. Młynarz, Tetrahedron Lett., 2009, 50, 132.
A. M. Corrente, R. T. Boeré and T. L. Roemmele, Inorg. 14 T. S. De Vries, A. Prokofjevs, J. N. Harvey and E. Vedejs,
Chem., 2008, 47, 3823; (c) T. Chivers, C. Fedorchuk, J. Am. Chem. Soc., 2009, 131, 14679.
G. Schatte and J. K. Brask, Can. J. Chem., 2002, 80, 821; 15 For example see: (a) L. Horner, U. Kaps and G. Simons,
(d) T. Chivers, D. J. Eisler, C. Fedorchuk, G. Schatte,
H. M. Tuononen and R. T. Boeré, Chem. Commun., 2005,
3930.
5 For recent examples see: (a) P. Šimon, F. de Proft,
R. Jambor, A. Růžička and L. Dostál, Angew. Chem., Int. Ed.,
2010, 49, 5468; (b) L. Dostál, R. Jambor, A. Růžička and
J. Holeček, Organometallics, 2008, 27, 2169; (c) L. Dostál,
R. Jambor, A. Růžička and P. Šimon, Eur. J. Inorg. Chem.,
2011, 2380; (d) M. Bouška, L. Dostál, Z. Padělková,
J. Organomet. Chem., 1985, 287, 1; (b) R. Schlengermann,
J. Sieler and E. Hey-Hawkins, Main Group Met. Chem., 1997,
2, 141; (c) Z. García-Hernández and F. P. Gabbaï, Z. Natur-
forsch., B: Chem. Sci., 2009, 64, 1381; (d) Z. M. Heiden,
M. Schedler and D. W. Stephan, Inorg. Chem., 2011, 50,
1470; (e) M. Yamashita, K. Kamura, Y. Yamamoto and
K. Akiba, Chem.–Eur. J., 2002, 8, 2976; (f) M. Yamashita,
Y. Yamamoto, K. Akiba, D. Hashizume, F. Iwasaki,
N. Takagi and S. Nagase, J. Am. Chem. Soc., 2005, 127, 4354.
S. Herres-Pawlis, K. Jurkschat, A. Lyčka and R. Jambor, 16 A. Meller, H. Hoppe, W. Maringgele, A. Haase and
Angew. Chem., Int. Ed., 2012, 51, 3478; (e) P. Šimon, M. Noltemeyer, Organometallics, 1998, 17, 123.
R. Jambor, A. Růžička, A. Lyčka, F. De Proft and L. Dostál, 17 P. Király, Magn. Reson. Chem., 2012, 50, 620.
Dalton Trans., 2012, 41, 5140; (f) M. Bouška, L. Dostál, 18 (a) P. Pyykkö and M. Atsumi, Chem.–Eur. J., 2009, 15, 186;
F. De Proft, A. Růžička, A. Lyčka and R. Jambor, Chem.–Eur.
J., 2011, 17, 455.
6 (a) G. Schmidt, Comprehensive Heterocyclic Chemistry II,
(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.
1996, Chapter 3.17, vol. 3, p. 739; (b) G. Schmidt, Comments 19 (a) S. Toyota and M. Oki, Bull. Chem. Soc. Jpn., 1992, 65,
Inorg. Chem., 1985, 4, 17; (c) A. J. Ashe III and X. Fang, Org.
Lett., 2000, 2, 2089; (d) S.-Y. Liu, M.-C. Lo and G. C. Fu,
1832; (b) S. Toyota and M. Oki, Bull. Chem. Soc. Jpn., 1992,
65, 2215.
Angew. Chem., Int. Ed., 2002, 41, 174; (e) A. J. Ashe III, 20 H. Hopfl, J. Organomet. Chem., 1999, 581, 129.
Comprehensive Heterocyclic Chemistry III, 2008, Chapter 21 We are grateful to one of the reviewers of this manuscript,
4.17, vol. 4, p. 1189.
who suggested that the lithium atom in the structure of the
addition product can be coordinated either to anilido-
nitrogen or imino-nitrogen or may even bridge these two
atoms. Such migration of the lithium atom precisely
explains formation of both compounds 7 and 13 in one
reaction depending on the conditions, where especially the
formation of 13 was obscure to us.
7 (a) A. J. Ashe III, Organometallics, 2009, 28, 4236;
(b) X. Fang and J. Assoud, Organometallics, 2008, 27, 2408;
(c) G. Schmidt, O. Boltsch and R. Boese, Organometallics,
1987, 6, 435; (d) G. Schmidt and M. Schütz, Organo-
metallics, 1992, 11, 1789; (e) X. Fang, X. Li, Z. Hou,
J. Assoud and R. Zhao, Organometallics, 2009, 28, 517;
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6417–6428 | 6427

View Article Online
Paper
Dalton Transactions
22 D. Zhao, W. Gao, Y. Mu and L. Ye, Chem.–Eur. J., 2010, 16, 24 A. Altomare, G. Cascarone, C. Giacovazzo, A. Guagliardi,
4394.
M. C. Burla, G. Polidori and M. Camalli, J. Appl. Crystal-
logr., 1994, 27, 1045.
23 P. Coppens, in Crystallographic Computing, ed. F. R. Ahmed,
S. R. Hall, and C. P. Huber, Copenhagen, Munksgaard, 25 G. M. Sheldrick, SHELXL-97, A Program for Crystal Structure
1970, pp. 255–270.
Refinement, University of Göttingen, Germany, 1997.
6428 | Dalton Trans., 2013, 42, 6417–6428
This journal is © The Royal Society of Chemistry 2013