A variable-temperature study of 1,2-bis-(dimethylamino)-1,2-bis-(2,6- dimethylanilino)diborane

Article abstract of DOI:10.1107/S0108270111039230

The title compound, C₂₀H₃₂B₂N₄, is monoclinic at ambient temperature but triclinic (pseudo-monoclinic) below 150 K. The structures of the two phases, determined at 200 and 120 K, respectively, are very similar, the mol-ecular symmetry being crystallographic C 2 and approximate (local) C 2, respectively. There is significant π conjugation within each N - B - N moiety, but none between them or between the N - B - N and arene moieties.

Full text of DOI:10.1107/S0108270111039230

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
After several unsuccessful attempts, we obtained just one
crystal of the low-temperature phase, ꢂ-(I), by annealing a
crystal with three cycles of cooling–heating above the phase
transition (230 to 160 K) before finally cooling it to 120 K. On
warming to 230 K, the crystal remained triclinic, but cracked
on warming to room temperature. We tried the annealing
procedure on other samples but without success.
The structure of ꢂ-(I), determined from the data set
collected at 120 K, is triclinic, although the actual difference
from ꢁ-(I) is small. To simplify the comparison, ꢂ-(I) is
presented here in the (nonstandard) pseudo-monoclinic lattice
setting analogous to that of ꢁ-(I). Interestingly, this lattice of
ꢂ-(I) can be transformed by the operation (101/101/010) to a
C-centred pseudo-monoclinic lattice with a = 15.844, b =
ISSN 0108-2701
A variable-temperature study of
1,2-bis(dimethylamino)-1,2-bis(2,6-
dimethylanilino)diborane
Andrei S. Batsanov,a* George Bramham,b Nicholas C.
Normanb and Christopher A. Russellb
aDepartment of Chemistry, University of Durham, South Road, Durham DH1 3LE,
England, and bSchool of Chemistry, University of Bristol, Bristol BS8 1TS, England
Correspondence e-mail: a.s.batsanov@durham.ac.uk
ꢀ
˚
13.702 and c = 9.645 A, and ꢁ = 89.91, ꢂ = 93.46 and ꢃ = 89.66 ,
3
and V = 2090 A . However, we found no approximate struc-
˚
tural or Laue symmetry (Rint = 0.51) associated with this cell.
Received 17 August 2011
Accepted 24 September 2011
Online 28 September 2011
The title compound, C20H32B2N4, is monoclinic at ambient
temperature but triclinic (pseudo-monoclinic) below 150 K.
The structures of the two phases, determined at 200 and 120 K,
respectively, are very similar, the molecular symmetry being
crystallographic C2 and approximate (local) C2, respectively.
There is significant ꢀ conjugation within each N—B—N
moiety, but none between them or between the N—B—N and
arene moieties.
Comment
Tetra(amido)diborane(4) derivatives are important as pre-
cursors in diborane(4) chemistry (Brotherton, 1964), espe-
cially since the Suzuki reaction became widely used in the
synthesis of complex molecular species (Miyaura & Suzuki,
1995). The title compound, (I), was first reported by Patton et
al. (2001). In the course of our studies of substitution reactions
in diboranes (Baber et al., 2005), we reproduced the synthesis
of (I) and structurally characterized it by X-ray diffraction,
which revealed a phase transition.
Under ambient conditions, (I) is monoclinic. On slow
cooling, this ꢁ form (fully studied at 200 K) persisted to 160 K,
as was indicated by express lattice parameter determinations
at 292, 274, 250, 230, 200, 180 and 160 K. The phase transition
was observed between 160 and 150 K, at which point the peaks
of most reflections became split, but the separations between
the maxima were not wide enough to index the components
separately. If flash-cooled below 150 K, the crystals shattered.
Figure 1
The molecular structure of (I) in the ꢁ (top) and ꢂ (bottom) forms.
Displacement ellipsoids are drawn at the 50% probability level and
dashed lines indicate the minor disordered components of the methyl
groups. Primed atoms are generated by the crystallographic twofold axis
(symmetry code: ꢁx + , y, ꢁz + ).
1
2
1
2
o394 # 2011 International Union of Crystallography
doi:10.1107/S0108270111039230
Acta Cryst. (2011). C67, o394–o396

organic compounds
Form a-(I)
The molecular structure of (I) (Fig. 1) is very similar to that
of its mesityl analogue, 1,2-bis(2,4,6-trimethylanilido)-1,2-bis-
(dimethylamido)diborane, (II) (Fırıncı et al., 2010). In the
ꢁ-(I) form, the molecule possesses a crystallographic twofold
axis passing through the mid-point of the B—B bond. In the ꢂ
form, this twofold symmetry was noncrystallographic yet
almost perfect. In fact, the structure of ꢂ-(I) can be solved and
refined to R1 = 0.125 (Rint = 0.255) in the space group P2/n of
the ꢁ phase, ignoring the deviation of the ꢁ and ꢃ angles from
90ꢀ.
Crystal data
3
˚
V = 1056.63 (19) A
Z = 2
C20H32B2N4
Mr = 350.12
Monoclinic, P2=n
a = 10.4957 (11) A
Mo Kꢁ radiation
ꢄ = 0.07 mmꢁ1
T = 200 K
0.53 ꢃ 0.35 ꢃ 0.05 mm
˚
˚
b = 9.6327 (10) A
˚
c = 10.5554 (11) A
ꢂ = 98.06 (2)ꢀ
Data collection
Siemens SMART 1000 CCD area-
detector diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2006)
Tmin = 0.924, Tmax = 1.000
10983 measured reflections
2436 independent reflections
1779 reflections with I > 2ꢅ(I)
Rint = 0.028
All B and N atoms have planar–trigonal geometry. In each
case, the planes of a B atom and its two adjacent N atoms
coincide within experimental error, and the B—N distances
(Tables 1 and 3) indicate a degree of multiple bonding (Weber
et al., 2001). On the other hand, strong twists around the B—B
bond [67.6 (2)ꢀ in ꢁ-(I) and 65.9 (2)ꢀ in ꢂ-(I)] and around the
N—C(Ar) bonds [69.2 (2)ꢀ in ꢁ-(I), and 67.1 (2) and 68.8 (2)ꢀ
in ꢂ-(I)] preclude any ꢀ-conjugation. Thus, the B—B bond is
essentially single; its length falls within the usual range of
Refinement
R[F2 > 2ꢅ(F2)] = 0.043
wR(F2) = 0.128
S = 1.05
2436 reflections
124 parameters
1 restraint
H atoms treated by a mixture of
independent and constrained
refinement
ꢁ3
˚
Áꢆmax = 0.28 e A
ꢁ3
˚
Áꢆmin = ꢁ0.16 e A
˚
1.69–1.75 A for tetramido-diborane derivatives (Weber et al.,
2001; Baber et al., 2005) and is practically the same as in (II)
˚
[1.735 (6) A].
The two dimethylaryl groups in molecule (I) are nearly
parallel, the interplanar angle being 3.5 (1)ꢀ in both forms.
These groups stack closely together in an offset face-to-face
Form b-(I)
Crystal data
C20H32B2N4
Mr = 350.12
Triclinic, P1
ꢃ = 87.45 (3)ꢀ
V = 1045.0 (2) A
Z = 2
Mo Kꢁ radiation
ꢄ = 0.07 mmꢁ1
T = 120 K
˚
manner, with mean interplanar separations of 3.48 and 3.46 A
3
˚
in the ꢁ and ꢂ forms, respectively. Similar stacking occurs
between the dimethylaryl groups of adjacent molecules
related by an inversion centre. In this case, the arene planes
are rigorously parallel, with an interplanar separation of
˚
a = 10.5043 (13) A
˚
b = 9.6449 (11) A
˚
c = 10.4428 (14) A
ꢁ = 92.69 (3)ꢀ
ꢂ = 98.29 (3)ꢀ
0.52 ꢃ 0.21 ꢃ 0.02 mm
˚
˚
3.4270 (5) A in ꢁ-(I) and 3.2018 (7) A in ꢂ-(I). Thus, the
structure contains an infinite stacking motif, running parallel
to the [011] direction.
Data collection
Bruker SMART 6000 CCD area-
detector diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2006)
Tmin = 0.715, Tmax = 0.862
9870 measured reflections
3693 independent reflections
2774 reflections with I > 2ꢅ(I)
Rint = 0.027
Since the lone electron pairs of the N atoms are involved in
ꢀ conjugation, the molecule of (I) contains no electronegative
acceptors for strong hydrogen bonds. The N—H bonds point
roughly towards the ꢀ orbitals of arene C atoms of adjacent
molecules within the stack. In ꢁ-(I) there is one symmetrically-
independent contact of this type, viz. N1—H1Nꢂ ꢂ ꢂC3(ꢁx, ꢁy,
ꢁz) (Table 2), whereas in ꢂ-(I) there are two, one with the
same notation and the other N3—H3Nꢂ ꢂ ꢂC13(ꢁx + 1, ꢁy,
ꢁz + 1) (Table 4). These Hꢂ ꢂ ꢂC distances are 2.77 (1), and
Refinement
R[F2 > 2ꢅ(F2)] = 0.053
wR(F2) = 0.150
S = 1.05
3693 reflections
249 parameters
2 restraints
H atoms treated by a mixture of
independent and constrained
refinement
ꢁ3
˚
Áꢆmax = 0.35 e A
ꢁ3
˚
Áꢆmin = ꢁ0.24 e A
˚
2.76 (2) and 2.76 (2) A, respectively. Rowland & Taylor (1996)
estimated the standard van der Waals Hꢂ ꢂ ꢂC separation as
˚
3.02 A, using H-atom positions normalized by moving the H
atom along the observed X—H bond until this bond length
˚
matched the neutron diffraction value (0.983 A for N—H).
All H atoms were located in difference maps. Methyl groups were
refined as rigid bodies rotating around C—C or N—C bonds, with
˚
C—H = 0.98 A. The C10H3 group in ꢁ-(I) was treated as ideally
Applying such normalization to (I) and (II) reduces the
above-mentioned Hꢂ ꢂ ꢂC distances to 2.67, and 2.68 and
disordered over two conformations rotated from one another by 60ꢀ.
The arene H atoms were treated as riding on their C atoms, with
˚
2.66 A, respectively, hence these contacts can be regarded as
weak hydrogen bonds (Desiraju & Steiner, 1999).
˚
C—H = 0.95 A. The amino H atoms were refined with N—H
˚
distances restrained to 0.88 (2) A. The Uiso values of the methyl H
atoms were constrained to 1.5 times, and those of other H atoms to
1.2 times, the Ueq value of the attached C or N atom.
Experimental
Data collection: SMART (Bruker, 1998) for ꢁ-(I); SMART
(Bruker, 2001) for ꢂ-(I). For both forms, cell refinement: SAINT
(Bruker, 2007); data reduction: SAINT (Bruker, 2007); program(s)
used to solve structure: SHELXTL (Version 6.12; Sheldrick, 2008);
Compound (I) was prepared from the reaction between B2(NMe2)4
and two equivalents of NH2-2,6-Me2C6H3 in toluene and crystals
were isolated by slow cooling.
ꢄ
Acta Cryst. (2011). C67, o394–o396
Batsanov et al.
Two forms of C20H32B2N4 o395

organic compounds
Table 1
Table 3
Selected bond lengths (A) for ꢂ-(I).
˚
˚
Selected bond lengths (A) for ꢁ-(I).
N1—C1
N1—B1
1.4237 (16)
1.4290 (19)
N2—B1
B1—B1i
1.4197 (19)
1.727 (3)
N1—C1
N1—B1
N2—B1
N3—C11
1.422 (2)
1.433 (3)
1.423 (3)
1.424 (2)
N3—B2
N4—B2
B1—B2
1.428 (3)
1.424 (3)
1.726 (3)
1
2
1
2
Symmetry code: (i) ꢁx þ ; y; ꢁz þ .
Table 2
Hydrogen-bond geometry (A, ) for ꢁ-(I).
Table 4
ꢀ
ꢀ
˚
˚
Hydrogen-bond geometry (A, ) for ꢂ-(I).
D—Hꢂ ꢂ ꢂA
D—H
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D—Hꢂ ꢂ ꢂA
D—Hꢂ ꢂ ꢂA
D—H
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D—Hꢂ ꢂ ꢂA
N1—H1Nꢂ ꢂ ꢂC3ii
Symmetry code: (ii) ꢁx; ꢁy; ꢁz.
0.87 (1)
2.77 (1)
3.5631 (18)
152 (1)
N1—H1Nꢂ ꢂ ꢂC3i
0.88 (2)
0.87 (2)
2.76 (2)
2.76 (2)
3.544 (2)
3.528 (3)
148 (2)
148 (2)
N3—H3Nꢂ ꢂ ꢂC13ii
Symmetry codes: (i) ꢁx; ꢁy; ꢁz; (ii) ꢁx þ 1; ꢁy; ꢁz þ 1.
program(s) used to refine structure: SHELXTL; molecular graphics:
OLEX2 (Version 1.1-ꢂ+++; Dolomanov et al., 2009); software used to
prepare material for publication: OLEX2.
Bruker (1998). SMART. Version 5.049. Bruker AXS Inc., Madison, Wisconsin,
USA.
Bruker (2001). SMART. Version 5.625. Bruker AXS Inc., Madison, Wisconsin,
USA.
Bruker (2007). SAINT. Version 7.46A. Bruker AXS Inc., Madison, Wisconsin,
USA.
Desiraju, G. & Steiner, T. (1999). In The Weak Hydrogen Bond. Oxford
University Press.
The authors thank Professor J. A. K. Howard for helpful
discussions.
Dolomanov, O. V., Bourhis, L. J., Gildea, R. J., Howard, J. A. K. & Puschmann,
H. (2009). J. Appl. Cryst. 42, 339–341.
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: MX3055). Services for accessing these data are
described at the back of the journal.
¨
˘
¨ ¨ ¨ ¨
Fırıncı, E., Soyleyici, H. C., Gziroglu, E., Temel, E., Buyukgungor, O. & S¸ahin,
Y. (2010). Polyhedron, 29, 1465–1468.
Miyaura, N. & Suzuki, A. (1995). Chem. Rev. 95, 2457–2483.
Patton, J. T., Feng, S. G. & Abboud, K. A. (2001). Organometallics, 20, 3399–
3405.
Rowland, R. S. & Taylor, R. (1996). J. Phys. Chem. 100, 7384–7391.
Sheldrick, G. M. (2006). SADABS. Version 2006/1. University of Go¨ttingen,
Germany.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122.
Weber, L., Schneider, M. & Lo¨nnecke, P. (2001). J. Chem. Soc. Dalton Trans.
pp. 3459–3464.
References
Baber, R. A., Charmant, J. P. H., Cook, A. J. R., Farthing, N. E., Haddow, M. F.,
Norman, N. C., Orpen, A. G., Russell, C. A. & Slattery, J. M. (2005). Dalton
Trans. pp. 3137–3139.
Brotherton, R. J. (1964). Progress in Boron Chemistry, edited by H. Steinberg
& A. L. McCloskey, ch. 1. New York: Macmillan.
ꢄ
o396 Batsanov et al. Two forms of C20H32B2N4
Acta Cryst. (2011). C67, o394–o396