Tuning the electronic properties of cyclopentadienyl analogs with CB 2N2 frameworks: 1,2-diphenyl-1,2-diaza-3,5-diborolyl ligands and their alkali metal salts

Article abstract of DOI:10.1039/b802930j

Two heterocyclic cyclopentadienyl analogs with a CB₂N ₂ skeleton, 4-methyl-1,2,3,5-tetraphenyl-1,2-diaza-3,5-diborolidine and 4-methyl-3,5-dimethylamino-1,2-diphenyl-1,2-diaza-3,5-diborolidine were prepared through cyclocondensation of the corresponding 1,1- bis(organochloroboryl)ethane with 1,2-diphenylhydrazine. The former diazadiborolidine featured a cyclopentadiene-like structure with short B-N bonds and a planar ring framework, while in the latter the B-N bonds were noticeably longer and the ring framework was considerably folded as a result of the interaction between boron and the electron donating NMe₂ groups. The dimethylamino substituted diazadiborolidine could not be deprotonated due to the reduced acidity of the ring proton, however, the B-phenylated analog was easily deprotonated and the lithium, sodium and potassium 1,2-diaza-3,5-diborolyls were isolated and structurally characterized. The solid state structures of the lithium and sodium salts were similar, with an η¹-coordinated π ligand and three THF molecules completing the coordination sphere of the metal. The potassium salt featured a highly unusual mono-dimensional polymeric structure with the metal π-coordinated by the CB₂N₂ ligand and two of the phenyl groups on boron and nitrogen, and σ-coordinated by one THF molecule.

Full text of DOI:10.1039/b802930j

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions theme issue on:
Main group chemistry: from molecules to materials
Guest Editor Simon Aldridge
University of Oxford, UK
Published in issue 26, 2008 of Dalton Transactions
Images reproduced with permission of Daniel M. German (left) and Vadim Kessler (right)
Papers published in this issue include:
Heteroleptic metal alkoxide “oxoclusters” as molecular models for the sol–gel synthesis of
perovskite nanoparticles for bio-imaging applications
Gulaim A. Seisenbaeva, Vadim G. Kessler, Robert Pazik and Wieslaw Strek, Dalton Trans., 2008,
DOI: 10.1039/b801351a
Hydride encapsulation by molecular alkali-metal clusters
Joanna Haywood and Andrew E. H. Wheatley, Dalton Trans., 2008, DOI: 10.1039/b717563a
Manipulation of molecular aggregation and supramolecular structure using self-assembled lithium
mixed-anion complexes
J. Jacob Morris, Dugald J. MacDougall, Bruce C. Noll and Kenneth W. Henderson, Dalton Trans.,
2008, DOI: 10.1039/b719565f
Toward selective functionalisation of oligosilanes: borane-catalysed dehydrogenative coupling of
silanes with thiols
Daniel J. Harrison, David R. Edwards, Robert McDonald and Lisa Rosenberg, Dalton Trans., 2008,
DOI: 10.1039/b806270f
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Tuning the electronic properties of cyclopentadienyl analogs with CB2N2
frameworks: 1,2-diphenyl-1,2-diaza-3,5-diborolyl ligands and their alkali
metal salts†
Hanh V. Ly, Jari Konu, Masood Parvez and Roland Roesler*
Received 22nd February 2008, Accepted 22nd April 2008
First published as an Advance Article on the web 20th May 2008
DOI: 10.1039/b802930j
Two heterocyclic cyclopentadienyl analogs with a CB2N2 skeleton, 4-methyl-1,2,3,5-
tetraphenyl-1,2-diaza-3,5-diborolidine and 4-methyl-3,5-dimethylamino-1,2-diphenyl-1,2-
diaza-3,5-diborolidine were prepared through cyclocondensation of the corresponding
1,1-bis(organochloroboryl)ethane with 1,2-diphenylhydrazine. The former diazadiborolidine featured a
cyclopentadiene-like structure with short B–N bonds and a planar ring framework, while in the latter
the B–N bonds were noticeably longer and the ring framework was considerably folded as a result of
the interaction between boron and the electron donating NMe2 groups. The dimethylamino substituted
diazadiborolidine could not be deprotonated due to the reduced acidity of the ring proton, however, the
B-phenylated analog was easily deprotonated and the lithium, sodium and potassium
1,2-diaza-3,5-diborolyls were isolated and structurally characterized. The solid state structures of the
1
lithium and sodium salts were similar, with an g -coordinated p ligand and three THF molecules
completing the coordination sphere of the metal. The potassium salt featured a highly unusual
mono-dimensional polymeric structure with the metal p-coordinated by the CB2N2 ligand and two of
the phenyl groups on boron and nitrogen, and r-coordinated by one THF molecule.
we characterized ligands of type C, and reported on sandwich
complexes of these ligands with metals from Groups 1 and 12 of the
Introduction
periodic table.3a–c These studies showed that the metal complexes of
the electron rich, pentaalkyl substituted ligands C1 were difficult to
purify and crystallize because of their extreme solubility in organic
solvents. The less electron rich trialkyl diphenyl ligands C2 proved
to be easier to prepare through deprotonation because of the
increased acidity of the ring proton in the neutral precursor, and
had superior crystallization properties. Bicyclic ligands derived
from C2 demonstrated excellent p-bridging properties in dinuclear
ruthenium complexes, and the potential to generate multidecker
sandwich compounds.3d
We report herein on further studies concerning the tuning of the
coordination properties of ligands of type C. The synthesis and
characterization of the ligands C3 and C4, which were expected
to have good crystallization properties and differ considerably
in their electron donating abilities, was enticing because of the
stability and commercial availability of 1,2-diphenylhydrazine.
The tremendous importance of cyclopentadienyl as a ligand
in coordination chemistry and catalysis1 has driven the quest
for the development of heterocyclic analogs displaying similar
coordination behavior and different electronic properties. Anionic
heterocycles featuring, aside from carbon, boron alone or in
combination with nitrogen or chalcogens and displaying excellent
p-coordinating properties have been extensively investigated in
this regard. Pioneered by Schmid and co-workers, the 1,2-
azaborolidines A (E = N)2a–c have been further investigated by
Ashe2f ,j and subsequently extended to 1,2-oxa2k and thiaboro-
lidines A2g (E = O, S) and 1,3-thiaborolidines B.2d,e The synthesis
and properties of 1,2-azaborolidines A (E = N) have also been
investigated by Fu,2h,i,l and the suitability of some of these five-
membered heterocycles as ancillary ligands in catalytic systems
has been examined.2e,l
Results and discussion
The ligand precursors 1 and 2 were prepared through the
condensation of previously reported geminal bis(organochlo-
roboryl)ethanes4 with symmetric diphenylhydrazine in the pres-
ence of triethylamine (Scheme 1). The products were isolated as
colorless crystalline solids showing good solubility in ethers and
aromatic hydrocarbons and poor solubility in hexane and pentane.
The 1H and 13C NMR spectra of 1 and 2 featured the expected
signals. The distinctive doublet-quartet pattern of the Me–CHB2
moiety was observed in the 1H NMR spectra of both substances,
and only one signal was observed for the methyl protons of the
Aiming to explore the p-coordinating properties of cyclopen-
tadienyl analogs with low carbon content in the ring framework,
Department of Chemistry, University of Calgary, 2500 University Drive,
Calgary, Alberta, T2N 1N4, Canada
† Electronic supplementary information (ESI) available: NMR spectra
of all new compounds. CCDC reference numbers 678560–678564. For
ESI or crystallographic data in CIF or other electronic format see DOI:
10.1039/b802930j
3454 | Dalton Trans., 2008, 3454–3460
This journal is
The Royal Society of Chemistry 2008
©

Scheme 1 Synthesis of ligand precursors 1 and 2.
dimethylamino groups in 2, indicative of a low rotational barrier
around the exocyclic B–N bond. The 13C NMR spectrum of 2
featured two independent, yet very close, broad signals for the
dimethylamino group, suggesting that the spectrum was collected
in the proximity of the coalescence temperature. The signals
corresponding to the ring carbon and proton in 2 are considerably
high field shifted with respect to 1 (0.65 vs. 1.65 ppm for 1H
and 8.9 vs. 23.0 ppm for 13C), as a consequence of the increased
electron density ascribed to the electron donating ability of the
dimethylamino groups on boron. The EI mass spectra of 1 and 2
displayed intense signals for the molecular ions and products of
their fragmentation.
Fig. 1 Molecular structure of 1, with thermal ellipsoids drawn at 50%
probability. For clarity, all hydrogen atoms except for the one on the ring
carbon have been omitted.
The structural determinations for 1 and 2 (Table 1) confirmed
the identity of the compounds and, since the structures of the
neutral precursors to ligands C1 and C2 had not been determined,
offered a first insight into the molecular structures of such
derivatives. To our knowledge, no crystal structure for either of
the protonated precursors to the much investigated ligands A
(E = N) have been reported. The ring framework of 1 showed
excellent planarity (Fig. 1), with a sum of the pentagon angles
of 539.8◦ (Table 2), a folding along the B · · · B axis of only
2.8◦ and a propeller-like arrangement of the phenyl substituents
that is typical for tetra- and penta-phenylcyclopentadienyl. The
ring skeleton in 2 however, featured an envelope conformation
with a considerable folding angle along the B(1) · · · N(2) axis
(28.7◦), likely in order to alleviate the steric strain between
the ring substituents (Fig. 2). The strong p-interaction between
the extraannular nitrogen centers and the boron atoms imposes
very small dihedral angles between the planes defined by the
substituents on boron and the exocyclic nitrogen (5.8 and 14.8◦).
Fig. 2 Molecular structure of 2, with thermal ellipsoids drawn at 50%
probability. For clarity, all hydrogen atoms except for the one on the ring
carbon have been omitted.
By comparison, in 1 the phenyl substituents on boron form
dihedral angles of 43.8 and 46.8◦ with the planes defined by
the substituents on boron. The phenyl groups in 2 need to
accommodate the nearly planar arrangement of the (Me2N–
B)2CHMe unit and especially the steric bulk of the dimethylamino
Table 1 Selected data and structure refinement details for 1, 2, 3a(thf)3, 3b(thf)3, and 3c(thf)
1
2
3a(thf)3
3b(thf)3
3c(thf)
Empirical formula
Formula weight
Crystal system
Space group
C26H24B2N2
386.09
C18H26B2N4
320.05
C38H47B2LiN2O3
608.34
Monoclinic
C38H47B2N2NaO3
624.39
Monoclinic
C30H31B2KN2O
496.29
Monoclinic
Triclinic
Triclinic
¯
¯
P 1
P 1
C2/c
P21
P21/n
˚
a/A
9.274(5)
10.117(7)
12.837(8)
106.43(3)
91.33(3)
110.43(3)
1072.6(12)
2
9.751(3)
10.426(3)
10.457(3)
117.055(15)
90.450(18)
101.589(17)
921.4(5)
2
22.732(7)
8.931(2)
35.3500(11)
90
104.183(11)
90
6958(3)
8
1.161
9.154(5)
14.153(5)
13.912(7)
90
100.47(2)
90
1772.4(15)
2
1.170
11.393(2)
9.782(2)
24.168(5)
90
99.73(3)
90
2654.8(9)
4
1.242
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
Z
d
calc/g◦cm−3
1.195
50.5
0.068
1.154
50.6
0.068
2hmax
/
54.9
0.0712
54.9
0.083
50.0
0.226
l(Mo Ka)/mm−1
Independent reflections
Data/restraints/parameters
GOF on F2
3854 (Rint = 0.027)
2854/0/272
1.021
0.0605
0.1801
3333 (Rint = 0.023)
3333/0/217
1.05
0.0436
0.1148
7828 (Rint = 0.037)
7828/0/415
1.006
0.0557
0.1540
4189 (Rint = 0.047)
4189/1/416
1.020
0.0555
0.1440
4623 (Rint = 0.036)
4623/0/325
1.028
0.0479
0.1323
R1(F) [I > 2r(I)]
wR2(F2) [all data]
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3454–3460 | 3455
©

◦
˚
Table 2 Selected bond lengths (A) and angles ( ) for 1, 2, 3a(thf)3, 3b(thf)3, and 3c(thf)
1
2a
3a(thf)3 (M = Li)
3b(thf)3 (M = Na)
3c(thf) (M = K)
Cintra–Bb
extra–B
B–N
N–N
1.572(4), 1.581(4)
1.565(4), 1.568(4)
1.412(3), 1.416(3)
1.452(3)
1.578(2), 1.592(2)
1.389(2), 1.399(2)
1.467(2), 1.497(2)
1.4572(17)
1.490(3), 1.501(3)
1.579(3), 1.586(3)
1.464(2), 1.481(3)
1.4374(19)
1.482(6), 1.499(6)
1.583(5), 1.584(5)
1.466(5), 1.486(5)
1.438(4)
1.479(3), 1.499(3)
1.583(3), 1.585(3)
1.458(3), 1.494(3)
1.443(2)
C
C
intra–Cextra
1.462(4)
1.429(3), 1.438(3)
—
—
—
1.550(2)
1.4087(19), 1.4146(19)
—
—
—
1.527(3)
1.407(2), 1.431(2)
2.351(4)
2.908(4), 2.983(4)
1.947(4)–2.009(3)
104.62(16)
1.522(5)
1.415(4), 1.418(5)
2.680(4)
2.939(5), 3.189(5)
2.325(3)–2.360(4)
105.7(3)
1.524(3)
1.404(3), 1.427(3)
3.021(2)
3.249(3), 3.466(3)
2.6443(19)
105.01(19)
N–Cextra
M–Cintra
M · · · B
M–O
B–Cintra–B
102.3(2)
98.87(12)
C
extra–Cintra–B
121.0(2), 125.3(2)
107.6(2), 108.4(2)
126.8(2), 127.9(2)
124.2(2), 124.7(2)
110.44(18), 111.11(18)
128.48(19), 129.66(19)
117.56(17), 118.05(17)
30.1
111.18(13), 111.33(13)
108.95(13), 109.52(13)
126.75(14), 128.92(14)
122.07(14), 123.51(14)
104.46(11), 109.05(11)
122.49(12), 129.60(12)
114.43(12), 114.45(11)
97.2
126.50(17), 128.55(16)
108.63(15), 109.73(15)
129.51(17), 131.02(17)
119.23(16), 121.54(17)
107.11(13), 109.52(14)
120.51(14), 131.56(14)
113.54(13), 115.25(14)
58.1
125.6(3), 128.2(3)
108.3(3), 108.7(3)
128.1(4), 129.4(3)
121.8(3), 123.3(3)
107.6(3), 109.2(3)
125.0(3), 128.5(3)
115.5(3), 116.0(3)
52.8
126.8(2), 128.2(2)
108.92(19), 109.51(19)
127.4(2), 129.7(2)
120.68(19), 123.2(2)
106.72(16), 109.33(17)
119.92(17), 131.20(18)
114.15(17), 118.25(16)
49.0
Cintra–B–N
Cintra–B–Cextra
C
B–N–N
B–N–Cextra
N–N–Cextra
extra–B–N
C
extra–N–N–Cextra
pentagon angles
ꢀ
539.8
530.8
539.6
539.5
539.5
a For 2, (B)Cextra is (B)Nextra
.
b Cintra and Cextra indicate carbon atoms that are part of the ring (intraannular) and directly connected to the ring (extraannular),
respectively.
groups, and are pushed considerably out of planarity, forming
a CNNC torsion angle of 97.2◦ (vs. 30.1 in 1). This deviation
from planarity would be expected to have a negative impact on
the electron delocalization along the ring framework of C4 and
consequently on its ligand properties.
˚
The N–N bond lengths in 1 and 2 (1.452(3) and 1.4572(17) A,
respectively) are practically identical with the value measured in
hydrazine4 and in the alkali metal salts of ligands C1 and C2.2a,b In 1,
˚
the B–N bonds measure 1.412(3) and 1.416(3) A and are situated
ꢀ
=
in the range observed for acyclic R2B NR 2 compounds (R =
ꢀ
5a
ꢀ
fluorenyl, R = iPr, 1.390(14) A; R = R = Me, 1.403(1) A;5b R =
˚
˚
ꢀ
5c
˚
fluorenyl, R = Me, 1.400(12) A ). Very similar B–N bond lengths,
˚
ranging between 1.380(6) and 1.430(6) A have been measured in
Scheme 2 Deprotonation of ligand precursors 1 and 2.
other neutral five-membered rings featuring a BNC3 skeleton and
tricoordinated boron and nitrogen centers.6 The intraannular B–N
˚
bonds in 2 are, with 1.467(2) and 1.497(2) A, considerably longer
than those in 1, as a result of the competitive electron donation
to boron from the endo and exocyclic nitrogen atoms. The
electron donation from the more electron rich exocyclic nitrogen
is facilitated by the coplanar arrangement of the substituents (vide
supra) and results in shorter extraannular B–N bonds (1.4087(19)
and 1.4146(19)). The intra and extraannular B–C bonds in 1 are
to the methyl group were easily noticeable in the 1H NMR
spectra, which displayed all the expected resonances. The most
remarkable change in the 13C NMR spectra upon deprotonation
was the shift of the broad signal corresponding to the ring carbon
from 23.0 ppm in the neutral precursor to 93.6–98.1 ppm in the
alkali metal salts, indicative of the change in hybridization and
electron density of this atom. Such a dramatic shift is typical for
cyclopentadienyl derivatives with highly ionic character and has
been observed for C1 and C2.3b All other signals in the 13C NMR
spectra of 3a–c could be assigned. The 11B NMR spectrum of
1 underwent little change upon deprotonation, the broad singlet
shifting from 44 ppm in 1 to 41–42 ppm in 3a–c. Deprotonation
of 2 was unsuccessfully attempted using several strong bases
(Scheme 2). While the nitrogen bases showed no reaction with
2 in THF, nBuLi resulted in decomposition to a complex mixture
of products. This is likely a consequence of the excellent p donating
ability of the dimethylamino groups, resulting in increased electron
density on the CB2 moiety and ultimately a decreased acidity of
the ring proton. The lack of reactivity is in good agreement with
the results of the NMR studies of 2 (vide supra).
˚
similar in length (1.565(4)–1.592(2) A), and comparable to the
intraannular B–C bonds observed in 2 (1.578(2), 1.592(2) A), as
˚
well as in simple, unassociated organoboranes, R3B (R = Mes,
7a,c
7b
˚
˚
1.579(2), 1.580(3) A;
R = Ph, 1.571(3), 1.589(5) A; R =
7d
7e
˚
˚
tBu, 1.618(3) A; R = Me3SiCH2, 1.563(3)–1.573(3) A ). It
can be concluded that the structure of 1 resembles strongly the
structure of cyclopentadienes.8 The steric and electronic effects of
the dimethylamino substituents have a considerable impact on the
structure of 2, which by comparison to 1 appears substantially
distorted.
The deprotonation of 1 was accomplished using a variety of
reagents (Scheme 2) and the alkali metal salts 3a–c were isolated
as white crystalline solids. The disappearance of the ring proton
and the resulting collapse of the doublet signal corresponding
3456 | Dalton Trans., 2008, 3454–3460
This journal is
The Royal Society of Chemistry 2008
©

The structures of the alkali metal salts 3a(thf)3, 3b(thf)3 and
3c(thf) feature the ligand C3 and one or three additional THF
molecules coordinated to the metal (Table 1). These structures
offer the first opportunity to directly assess the transformations
suffered by a cyclopentadienyl analog with a CB2N2 framework
upon deprotonation and coordination to a metal (Table 2). The
metric parameters of the ligand C3 in its alkali metal salts, 3a(thf)3,
3b(thf)3 and 3c(thf), are very similar if not identical to the metric
parameters of the related ligands C1 and C2 in their alkali metal
salts. A detailed comparison of the bond lengths and angles of the
cyclopentadienyl analogs with CB2N2 skeletons in these latter salts
to the metric parameters of other boron compounds was already
presented,3b and hence the current discussion will focus on the
changes suffered by the ligands upon deprotonation, and on the
changes in the structure of the ligands as a result of the change of
organic substituents on boron and nitrogen.
Fig. 3 Molecular structure of 3a(thf)3, with thermal ellipsoids drawn at
50% probability. For clarity, all hydrogen atoms as well as the methylene
groups in THF have been omitted.
The N–N bonds in the alkali metal salts described herein are
practically equal in length with the N–N bonds observed in the
neutral precursors 1 and 2. The B–N bonds in 1, however, are
˚
lengthened by ca. 0.05 A upon deprotonation. Consequently,
the B–N bonds in the alkali metal salts of ligand C3 (1.458(3)–
˚
1.494(3) A) are slightly longer than the typical bonds in borazines
9
˚
(1.42–1.44 A) and similar to the intraannular B–N bonds in
2. The deprotonation of 1 has its most dramatic effect on
the intraannular B–C bonds, which are shortened by 0.07–
˚
0.10 A in the alkali metal salts 3a–c and lie within the range
observed for other boron heterocycles featuring a delocalized p
system, such as 1,2-azaborolinyl (1.46–1.50 A).2a–c,i The immediate
˚
conclusion of the comparison between the structures of 1 and
3a–c supports previous observations on the electron distribution
in anionic cyclopentadienyl analogs with CB2N2 framework: The
deprotonation of the neutral precursor results in the shortening
of the B–C bonds and lengthening of the B–N bonds, very
similar to the effect of the deprotonation of cyclopentadienes to
cyclopentadienyls. However, the resulting electron delocalization
appears to be limited to the N–B–C–B–N system and does not
extend over the N–N bond, which retains a single-bond character.
The CB2N2 ring skeletons in the salts 3a–c are slightly folded
along the B(1) · · · N(2) axis, with dihedral angles of 6.0–7.8◦. The
CNNC torsion angles are wider than in 1 by 20 to 30◦ and this
appears counterintuitive at first, as deprotonation results in the
formation of extended p systems. The difference is however, in
good agreement with the reduction in the bond order of the B–
N bonds, which is not complemented by an increase in the bond
order of the N–N bond. The addition of the M(thf)n (M = Li, Na,
K; n = 1, 3) moiety to the coordination sphere of the ligand also
results in an increased steric strain, which could contribute to the
widening of the CNNC torsion angles.
Fig. 4 Molecular structure of 3b(thf)3, with thermal ellipsoids drawn at
50% probability. For clarity, all hydrogen atoms as well as the methylene
groups in THF have been omitted.
1
Fig. 5 Perpendicular projection onto the CB2N2 plane, revealing the g
hapticity of the ligand in 3a(thf)3 (left), 3b(thf)3 (center) and 3c(thf) (right).
1
the g coordination mode is highly uncommon in these salts.10 In
the structures available for comparison, the distance between the
5
The salts 3a(thf)3 (Fig. 3) and 3b(thf)3 (Fig. 4) feature piano stool
structures with distorted tetrahedral coordination geometries of
metal ion and the plane of the g coordinated cyclopentadienyl
11b
11c,d
˚
˚
ligand measures 2.04 A for Li, 2.44–2.48 A for Na
and 2.74–
1
11a,e
˚
the metals and g C coordination modes of the ligand (Fig. 5).
2.78 A for K.
Although the bonding between the alkali metals
and the cyclopentadienyl ligands is mainly ionic in nature,10a the
A very similar structure was determined for the sodium salt of
3b
˚
the ligand C2, with a very short C–Na distance of 2.616(2) A.
considerably larger distances between the lithium and sodium ions
˚
˚
With 2.680(4) A, the slightly longer C–Na distance in 3b(thf)3 is
and the plane of the CB2N2 ligands in 3a(thf)3 (2.25 A) and 3b(thf)3
˚
in agreement with the expected lower donor ability of the ligand
C3 as compared to C2. Somewhat more common in mixed cy-
clopentadienyl salts of alkali metals and lanthanoids, three-legged
piano stool geometries are surprisingly rare in the coordination
chemistry of ether containing alkali metal cyclopentadienyls, and
(2.68 A) are good indicators for the reduced coordinating ability
of the ligand C3. Because the p ligand clearly coordinates in an
1
g fashion through the carbon atom (Fig. 5) the larger covalent
12
˚
˚
radius of boron (0.822 A vs. 0.772 A for carbon) is not likely to
be a considerable contributing factor to the increased distance.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3454–3460 | 3457
©

In the solid state, 3c(thf) features a highly unusual monodi-
mensional polymeric structure (Fig. 6). The potassium ion is
skeleton, the ligand precursors 1 and 2 have been synthesized
and structurally characterized. Diazadiborolidine 1 featured a
cyclopentadiene-like structure with short B–N bonds and could be
easily deprotonated, yielding the lithium, sodium and potassium
salts 3. A comparison between the metric parameters of the
protonated ligand 1 and the salts 3a–c confirmed that the p
electron delocalization over the N–B–C–B–N skeleton in the
anionic CB2N2 ligand does not extend over the N–N bond, which
displays a single bond character. The combined effects of the
four electron-withdrawing phenyl groups imparted poor electron
donating properties to the ring skeleton of the ligand, which proved
to have inferior coordinating ability in comparison to ligands C1
and C2. This was reflected by the long metal to ligand distances
in the lithium and sodium salts, and by the p coordination of
the potassium ion by the phenyl groups of the ligand instead
of the CB2N2 ring. Diazadiborolidine 2 represented the other
extreme, with the excellent p donating ability of the dimethylamino
substituents increasing the electron density at the ring carbon to
an extent sufficient to prevent its deprotonation and preclude the
synthesis of its alkali metal salts.
1
g C coordinated by the CB2N2 ligand, as observed in the lithium
6
3
and sodium analogs, and g (NPh) and g (BPh) coordinated by
two phenyl groups of another ligand. The coordination sphere
of the metal ion is completed by a THF molecule, leading to a
distorted tetrahedral geometry. Polydecker sandwich architectures
are common solid state arrangements in alkali metal cyclopentadi-
enyls, especially in the absence of coordinating solvents. However,
structures where the metal coordinates to one or more phenyl
ligands, despite the availability of one face of the cyclopentadienyl
ligand, have to our knowledge not been reported. Such preference
has only been observed for a cyclopentadienyl analog featuring
a carbon-free B2N3 framework.13 Two solid state structures
of potassium fluorenyls are known, where the potassium ion
coordinates to the phenyl instead of the cyclopentadienyl ring of
the ligand.14 This type of solid state p-interaction between phenyl
groups and potassium ions are nevertheless common in potassium
tetraphenylborates,15 and enable the selective precipitation of
potassium ions from aqueous solutions using tetraphenylborate as
a counterion.16 The distances between potassium and the planes of
As expected, the two or four phenyl substituents imparted
excellent crystallization properties to the neutral ligands and
their salts, an important asset for the purification and structural
characterization of these derivatives. It was concluded however,
that for electronic reasons attributed to the substitution pattern,
1 and 2 are not optimal precursors for the design and synthesis
of transition metal sandwich compounds featuring “inorganic”
ligands.
˚
the phenyl ligands in 3c(thf) measure 2.99 A, and are very similar
to the values observed in tetraphenylborates (2.92–3.02 A).14,15 The
˚
distance between the potassium ion and the plane of the CB2N2
˚
ligand in 3c(thf) measures 3.01 A, but this value is difficult to
use for comparison because of the lack of similar substitution
patterns for the metal in the chemistry of cyclopentadienyls. The
preference of the potassium ion for p-coordination by the phenyl
groups instead of the CB2N2 ring is another indication of the poor
coordinating ability of ligand C3.
Experimental section
General considerations
All operations were performed under an argon atmosphere using
standard Schlenk and glove box techniques. Solvents were dried
and deoxygenated prior to use. Starting materials were prepared
according to reported procedures or purchased from commercial
suppliers. The 1,2-diphenylhydrazine was purchased from Sigma
Aldrich and the small amount of diazobenzene it contained was
removed prior to use by washing with hexane. NMR spectra
were collected on a Bruker Advance DRX-400 spectrometer and
calibrated with respect to C6D5H (1H, 7.15 ppm), THF-d7 (1H,
3.58 ppm), toluene-d7 (1H, 2.09 ppm), C6D6 (13C, 128.39 ppm),
THF-d8 (13C, 67.57 ppm), toluene-d8 (13C, 20.4 ppm), BF3·Et2O
(11B, 0 ppm), and LiCl (7Li, 0 ppm). Mass spectra were performed
by the Analytical Instrumentation Laboratory, Department of
Chemistry, University of Calgary.
Fig. 6 Fragment from the polymeric chain of 3c(thf) in the solid state,
with thermal ellipsoids drawn at 50% probability. For clarity, only the
ipso carbon atom in the phenyl substituent on N(1) is represented and
all hydrogen atoms, as well as the methylene groups in THF, have been
omitted.
Synthesis of 4-methyl-1,2,3,5-tetraphenyl-1,2-diaza-
3,5-diborolidine, 1
A mixture of 1,2-diphenylhydrazine (737 mg, 4.00 mmol) and
triethylamine (810 mg, 8.00 mmol) in benzene (50 mL) was slowly
added to a stirring solution of 1,1-bis(chlorophenylboryl)ethane
(1.10 g, 4.00 mmol) in hexane (20 mL), producing a pale yellow sus-
pension. The mixture was stirred at ambient temperature for 12 h
and the white precipitate was subsequently filtered off. Volatiles
were removed under full vacuum, leaving behind a pale yellow solid
that was washed twice with hexane. Yield: 63% (0.98 g) colorless
Conclusion
In an attempt to assess the limits in the tuning of the co-
ordinating ability of cyclopentadienyl analogs with a CB2N2
3458 | Dalton Trans., 2008, 3454–3460
This journal is
The Royal Society of Chemistry 2008
©

1
3
1
3JHH = 7.2 Hz, o-BC6H5). 13C{ H} NMR (100 MHz, THF-d8, 298
solid. H NMR (400 MHz, C6D6, 298 K): d 1.45 (3H, d, JHH
=
3
8.2 Hz, HCCH3), 1.65 (1H, q, JHH = 8.2 Hz, HCCH3), 6.65
K): d 14.8 (s, CCH3), 26.5 (s, O(CH2CH2)2), 68.4 (s, O(CH2CH2)2),
98.1 (s, br, B2CCH3), 118.1 (s, p-NC6H5), 122.5 (s, m-NC6H5),
124.5 (s, p-BC6H5), 126.7 (s, o-NC6H5), 127.5 (s, m-BC6H5), 135.1
3
3
(2H, t, JHH = 7.3 Hz, p-NC6H5), 6.75 (4H, t, JHH = 7.5 Hz,
3
m-NC6H5), 6.87 (4H, d, JHH = 7.5 Hz, o-NC6H5), 7.11 (6H, m,
1
(s, o-BC6H5), 148.8 (s, br, i-BC6H5), 149.6 (s, i-NC6H5). 11B{ H}
1
m- and p-BC6H5), 7.44 (4H, d, o-BC6H5). H NMR (400 MHz,
3
NMR (128 MHz, THF-d8, 298 K): d 40.5 (s, br). 7Li{ H} NMR
1
THF-d8, 289 K): d 1.28 (3H, d, JHH = 8.2 Hz, HCCH3), 1.67
3
(1H, q, JHH = 8.2 Hz, HCCH3), 6.97 (2H, m, p-NC6H5), 7.04
(155 MHz, THF-d8, 298 K): d −0.82 (s). Colorless block crystals
of 3a(thf)3 were obtained by slow diffusion of liquid hexane into
a THF solution of 3a.
(8H, m, o- and m-NC6H5), 7.15 (6H, m, m- and p-BC6H5), 7.29
(4H, d, o-BC6H5). 13C{ H} NMR (100 MHz, THF-d8, 298 K): d
1
11.0 (s, HCCH3), 23.0 (s, br, HCCH3), 127.1 (s, p-NC6H5), 128.2
(s, m-NC6H5), 129.1 (s, o-NC6H5), 129.3 (s, p-BC6H5), 130.2 (s,
m-BC6H5), 134.9 (s, o-BC6H5), 137.5 (s, br, i-BC6H5), 142.1 (s,
Synthesis of 4-methyl-1,2,3,5-tetraphenyl-1,2-diaza-3,5-diborolyl
sodium, 3b
i-NC6H5). 11B{ H} NMR (128 MHz, THF-d8, 298 K): d 44.0 (s,
1
br). EI-MS (70 eV): m/z (%): 386 (76) [M]+, 371 (21) [M − Me]+.
Colorless thin plate-like crystals of 1 were obtained by cooling a
concentrated solution of 1 in a mixture of C6H6 and hexane to
−35 ◦C.
A solution of 1 (0.358 g, 0.927 mmol) in THF (10 mL) was added to
a stirring solution of sodium bis(trimethylsilyl)amide, NaHMDS
(0.170 g, 0.927 mmol) in THF (15 mL) and the yellow mixture
was subsequently stirred at ambient temperature for 3 h. The
volatiles were removed under vacuum leaving behind a colorless
solid that was washed twice with hexane (30 mL) and dried under
vacuum. The product was obtained as a colorless powder. Yield:
97% (0.40 g). 1H NMR (400 MHz, THF-d8, 298 K): d 1.77 (2H, m,
O(CH2CH2)2), 1.82 (3H, s, CCH3), 3.62 (2H, m, O(CH2CH2)2),
6.45 (2H, t, 3JHH = 7.1 Hz, p-NC6H5), 6.76 (4H, t, 3JHH = 7.5 Hz,
Synthesis of 4-methyl-1,2-diphenyl-3,5-dimethylamino-1,2-diaza-
3,5-diborolidine, 2
A mixture of 1,2-diphenylhydrazine (175 mg, 0.950 mmol) and
triethylamine (192 mg, 1.90 mmol) in benzene (20 mL) was
slowly added to a stirring solution of 1,1-bis(dimethylamino-
chloroboryl)ethane (200 g, 0.950 mmol) in hexane (20 mL).
The resulting orange-yellow suspension was stirred at ambient
temperature for 3 h and the white precipitate was filtered off.
Volatiles were removed under full vacuum, leaving behind the
product as a colorless solid. Yield: 51% (154 mg). 1H NMR
3
m-NC6H5), 6.87 (4H, d, JHH = 8.4 Hz, o-NC6H5), 7.02 (2H, t,
3JHH = 7.4 Hz, p-BC6H5), 7.12 (4H, t, 3JHH = 7.8 Hz, m-BC6H5),
7.27 (4H, d, 3JHH = 7.8 Hz, o-BC6H5). 13C{ H} NMR (100 MHz,
1
THF-d8, 298 K): d 13.8 (s, CCH3), 26.5 (s, O(CH2CH2)2), 68.4
(s, O(CH2CH2)2), 93.6 (s, br, B2CCH3), 119.5 (s, p-NC6H5), 123.0
(s, m-NC6H5), 125.4 (s, p-BC6H5), 127.1 (s, o-NC6H5), 127.8 (s,
m-BC6H5), 134.9 (s, o-BC6H5), 146.7 (s, br, i-BC6H5), 148.5 (s,
3
(400 MHz, toluene-d8, 298 K): d = 0.65 (q, 1H, JHH = 8.1 Hz,
3
1
HCCH3), 1.01 (d, 3H, JHH = 8.1 Hz, HCCH3), 2.57 (s, 12H,
i-NC6H5). 11B{ H} NMR (128 MHz, THF-d8, 298 K): d = 40.1
N(CH3)2), 6.82 (t, 2H, p-C6H5), 6.96 (m, 4H, m-C6H5), 7.19 (m,
4H, o-C6H5); 13C NMR (100 MHz, toluene-d8, 298 K): d = 8.9 (s,
br, HCCH3), 10.2 (s, HCCH3), 39.3 (s, br, N(CH3)3), 40.3 (s, br,
N(CH3)3), 116.6 (s, m-C6H5), 119.8 (s, p-C6H5), 129.0 (s, o-C6H5),
149.5 (s, i-C6H5); 11B NMR (128 MHz, toluene-d8, 298 K): d =
37.4 (s, br); EI-MS (70 eV): m/z (%): 320 (26) [M]+; HRMS for
H26C18N411 B2 calcd 320.2344, found 320.2356. Colorless thin plate
crystals of 2 were obtained by cooling a concentrated solution of
2 in a toluene–hexane mixture to −35 ◦C.
(s, br). Colorless block crystals of 3b(thf)3 were obtained after
several days by cooling a concentrated solution of 3b in a mixture
of THF and pentane to −35 ◦C.
Synthesis of 4-methyl-1,2,3,5-tetraphenyl-1,2-diaza-3,5-diborolyl
potassium, 3c
A solution of 1 (0.200 g, 0.518 mmol) in THF (10 mL) was added to
a stirring solution of potassium bis(trimethylsilyl)amide, KHMDS
(0.116 g, 0.518 mmol) in THF (15 mL) and the yellow mixture was
subsequently stirred at ambient temperature for 2 h. The volatiles
were removed under vacuum leaving behind a pale yellow solid,
which was washed twice with hexane (30 mL) and dried under
vacuum. The product was obtained as a light brown powder. Yield:
89% (0.20 g). 1H NMR (400 MHz, THF-d8, 298 K): d = 1.12 (3 H,
Synthesis of 4-methyl-1,2,3,5-tetraphenyl-1,2-diaza-3,5-diborolyl
lithium, 3a
A
yellow solution of lithium 2,2,6,6-tetramethylpiperidide,
LiTMP, was prepared from N-butyllithium in hexane (1.6 M,
0.33 mL, 0.53 mmol) and tetramethylpiperidine, TMP (75 mg,
0.53 mmol) in THF (3 mL). A solution o◦f 1 (0.20 g, 0.53 mmol)
in THF (10 mL) was pre-cooled to −35 C and mixed with the
pre-cooled solution of LiTMP and the reaction was allowed to
take place for 2 h at −35 ◦C. The solvent was removed under
vacuum leaving behind a pale yellow solid that was washed twice
with hexane (30 mL) and dried under vacuum. The product was
3
t, JHH = 7.0 Hz, O(CH2CH3)2), 1.78 (3H, s, CCH3), 3.39 (2 H,
3
3
q, JHH = 7.0 Hz, O(CH2CH3)2), 6.46 (2H, t, JHH = 7.2 Hz, p-
3
NC6H5), 6.77 (4H, t, JHH = 7.4 Hz, m-NC6H5), 6.84 (4H, d,
3JHH = 8.0 Hz, o-NC6H5), 7.01 (2H, t, 3JHH = 7.4 Hz, p-BC6H5),
7.12 (4H, t, 3JHH = 7.3 Hz, m-BC6H5), 7.34 (4H, d, 3JHH = 6.9 Hz,
o-BC6H5). 13C{ H} NMR (100 MHz, THF-d8, 298 K): d 14.4
1
(s, CCH3), 15.9 (s, O(CH2CH3)2), 66.5 (s, O(CH2CH3)2), 97.7 (s,
br, B2CCH3), 119.4 (s, p-NC6H5), 122.8 (s, m-NC6H5), 125.3 (s,
p-BC6H5), 127.2 (s, o-NC6H5), 128.0 (s, m-BC6H5), 134.8 (s, o-
1
obtained as an off-white powder. Yield: 90% (0.26 g). H NMR
(400 MHz, THF-d8, 298 K): d 1.73 (3H, s, CCH3), 1.77 (7H, m,
O(CH2CH2)2), 3.62 (7H, m, O(CH2CH2)2), 6.35 (2H, t, JHH
3
1
=
BC6H5), 147.2 (s, br, i-BC6H5), 148.5 (s, i-NC6H5). 11B{ H} NMR
3
7.2 Hz, p-NC6H5), 6.70 (4H, t, JHH = 7.3 Hz, m-NC6H5), 6.84
(128 MHz, THF-d8, 298 K): d 40.1 (s, br). Colorless block crystals
of 3c(thf) were obtained by slow diffusion of liquid hexane into a
THF solution of 3c.
3
3
(4H, d, JHH = 7.5 Hz, o-NC6H5), 6.94 (2H, t, JHH = 6.1 Hz,
3
p-BC6H5), 7.06 (4H, t, JHH = 7.2 Hz, m-BC6H5), 7.34 (4H, d,
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3454–3460 | 3459
©

Crystal structure determinations
Chem., 1992, 3, 399–406; (c) H. Braunschweig, C. von Koblinski, M.
Neugebauer, U. Englert and X. Zheng, J. Organomet. Chem., 2001, 619,
305–312.
Data were collected at a temperature of 173(2) K using the
6 (a) G. Schmid, D. Zaika, J. Lehr, N. Augart and R. Boese, Chem. Ber.,
1988, 121, 1873–1880; (b) E. P. Mayer and H. No¨th, Chem. Ber., 1993,
126, 1551–1557; (c) G. Schmid, S. Haske, D. Zaika, R. Boese and D.
Bla¨ser, Chem. Ber., 1994, 127, 73–80; (d) G. Schmid, J. Reschke and R.
Boese, Chem. Ber., 1994, 127, 1891–1894.
7 (a) J. F. Blount, P. Finocchiaro, D. Gust and K. Mislow, J. Am. Chem.
Soc., 1973, 95, 7019–7029; (b) F. Zettler, H. D. Hausen and H. Hess,
J. Organomet. Chem., 1974, 72, 157–162; (c) M. M. Olmstead and P. P.
Power, J. Am. Chem. Soc., 1986, 108, 4235–4236; (d) A. R. Cowley,
A. J. Downs, S. Marchant, V. A. Macrae, R. A. Taylor and S. Parsons,
Organometallics, 2005, 24, 5702–5709; (e) M. U. Kramer, D. Robert, Y.
Nakajima, U. Englert, T. P. Spaniol and J. Okuda, Eur. J. Inorg. Chem.,
2007, 665–674.
x and u scans on a Nonius Kappa CCD diffractometer with
˚
graphite monochromated Mo-Ka radiation with k = 0.71073 A
(Table 1), and corrected for Lorentz and polarization effects
and for absorption using the multi-scan method.17 The structures
were solved by the direct methods18 and expanded using Fourier
techniques.19 Non-hydrogen atoms were refined anisotropically
using SHELXL97.20 Hydrogen atoms were included at geomet-
rically calculated positions during the refinement using the riding
model. For 3b, in the absence of significant anomalous scattering
effects, the Flack parameter did not give a refinable value and an
absolute structure could not be established and therefore, Friedel
pairs were merged21 (CCDC reference numbers 678560 (1), 678561
(2), 678562 (3a(thf)3), 678563 (3b(thf)3), and 678564 (3c(thf))).
8 (a) L. D. Field, T. W. Hambley, C. M. Lindall and A. F. Masters, Inorg.
Chem., 1992, 31, 2366–2370; (b) T. Haumann, J. Benet-Buchholz and
R. Boese, J. Mol. Struct., 1996, 374, 299–304; (c) H. Schumann, K.
Su¨hring, R. Weimann and M. Hummert, Z. Naturforsch., B: Chem.
Sci., 2005, 60, 527–532.
9 (a) R. Boese, A. H. Maulitz and P. Stellberg, Chem. Ber., 1994, 127,
1887–1889; (b) C. A. Jaska, K. Temple, A. J. Lough and I. Manners,
J. Am. Chem. Soc., 2003, 125, 9424–9434; (c) T. Ja¨schke and M. Jansen,
Z. Anorg. Allg. Chem., 2004, 630, 239–243.
Acknowledgements
This work was supported by the Natural Sciences and Engineering
Research Council of Canada, the Canada Foundation for Innova-
tion, and the Alberta Science and Research Investments Program.
10 (a) S. Harder, Coord. Chem. Rev., 1998, 176, 17–66. and erratum: S.
Harder, Coord. Chem. Rev., 2000, 199, 331–334; (b) P. Jutzi and G.
Reumann, J. Chem. Soc., Dalton Trans., 2000, 2237–2244.
11 (a) J. Lorberth, S.-H. Shin, S. Wocadlo and W. Massa, Angew. Chem.,
Int. Ed. Engl., 1989, 28, 735–736; (b) R. Haag, R. Fleischer, D. Stalke
and A. de Meijere, Angew. Chem., Int. Ed. Engl., 1995, 34, 1492–1495;
(c) H. Bock, T. Hauck, C. Nather and Z. Havlas, Z. Naturforsch., B:
Chem. Sci., 1997, 52, 524–534; (d) T. Nishinaga, D. Yamazaki, H. Stahr,
A. Wakamiya and K. Komatsu, J. Am. Chem. Soc., 2003, 125, 7324–
7335; (e) T. Kitagawa, K. Ogawa and K. Komatsu, J. Am. Chem. Soc.,
2004, 126, 9930–9931.
12 R. T. Sanderson, J. Am. Chem. Soc., 1983, 105, 2259–2261.
13 H. V. Ly, J. H. Chow, M. Parvez, R. McDonald and R. Roesler, Inorg.
Chem., 2007, 46, 9303–9311.
14 (a) Y. Liu, D. Ballweg, T. Mu¨ller, I. A. Guzei, R. W. Clark and R. West,
J. Am. Chem. Soc., 2002, 124, 12174–12181; (b) S. Neander, J. Ko¨rnich
and F. Olbrich, J. Organomet. Chem., 2002, 656, 89–96.
15 (a) K. Hoffmann and E. Weiss, J. Organomet. Chem., 1974, 67, 221–228;
(b) T. Alaviuhkola, J. Bobacka, M. Nissinen, K. Rissanen, A. Ivaska
and J. Pursiainen, Chem.–Eur. J., 2005, 11, 2071–2080.
16 (a) G. Wittig, G. Keicher, A. Ru¨ckert and P. Raff, Justus Liebigs Ann.
Chem., 1949, 563, 110–126; (b) G. Wittig, Angew. Chem., 1950, 62, 231–
236; (c) F. P. Cassaretto, J. J. McLafferty and C. E. Moore, Anal. Chim.
Acta, 1965, 32, 376–380.
17 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276, 307–326.
18 A. Altomare, M. Cascarano, C. Giacovazzo and A. Guagliardi, J. Appl.
Crystallogr., 1993, 26, 343–350.
19 P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, R. de
Gelder, R. Israel and J. M. M. Smits, The DIRDIF-94 program system,
Technical Report of the Crystallography Laboratory, University of
Nijmegen, The Netherlands, 1994.
References
1 (a) A. Togni and R. L. Halterman, Metallocenes, Wiley-VCH,
Weinheim, 1998, vol. I, II; (b) N. J. Long, Metallocenes: An Introduction
to Sandwich Complexes, Blackwell Science Ltd., London, 1998; (c) P.
Jutzi and N. Burford, Chem. Rev., 1999, 99, 969–990.
2 (a) J. Schulze and G. Schmid, Angew. Chem., Int. Ed. Engl., 1980,
19, 54–55; (b) J. Schulze, R. Boese and G. Schmid, Chem. Ber., 1980,
113, 2348–2357; (c) G. Schmid, Comments Inorg. Chem., 1985, 4, 17–
32; (d) A. J. Ashe III, X. Fang and J. W. Kampf, Organometallics,
1998, 17, 2379–2381; (e) A. J. Ashe III, X. Fang and J. W. Kampf,
Organometallics, 1999, 18, 1821–1823; (f) A. J. Ashe III and X. Fang,
Org. Lett., 2000, 2, 2089–2091; (g) A. J. Ashe III, J. W. Kampf and M.
Schiesher, Organometallics, 2000, 19, 4681–4683; (h) S.-Y. Liu, I. D.
Hills and G. C. Fu, Organometallics, 2002, 21, 4323–4325; (i) S.-Y. Liu,
M. M.-C. Lo and G. C. Fu, Angew. Chem., Int. Ed., 2002, 41, 174–176;
(j) A. J. Ashe III, H. Yang, X. Fang and J. W. Kampf, Organometallics,
2002, 21, 4578–4580; (k) J. Chen, X. Fang, Z. Bajko, J. W. Kampf and
A. J. Ashe III, Organometallics, 2004, 23, 5088–5091; (l) S.-Y. Liu, I. D.
Hills and G. C. Fu, J. Am. Chem. Soc., 2005, 127, 15352–15353.
3 (a) H. V. Ly, T. D. Forster, D. Maley, M. Parvez and R. Roesler, Chem.
Commun., 2005, 4468–4470; (b) H. V. Ly, T. D. Forster, A. M. Corrente,
D. J. Eisler, J. Konu, M. Parvez and R. Roesler, Organometallics, 2007,
26, 1750–1756; (c) H. V. Ly, T. D. Forster, M. Parvez, R. McDonald
and R. Roesler, Organometallics, 2007, 26, 3516–3523; (d) H. V. Ly,
H. M. Tuononen, M. Parvez and R. Roesler, Angew. Chem., Int. Ed.,
2008, 47, 361–364.
4 K. Kohata, T. Fukuyama and K. Kuchitsu, J. Phys. Chem., 1982, 86,
602–606.
5 (a) W. Maringgele, M. Noltemeyer and A. Meller, J. Organomet. Chem.,
1989, 365, 187–199; (b) R. Boese, N. Niederpru¨m and D. Bla¨ser, Struct.
20 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112–122.
21 H. D. Flack, Acta Crystallogr., Sect. A: Found. Crystallogr., 1983, 39,
876–881.
3460 | Dalton Trans., 2008, 3454–3460
This journal is
The Royal Society of Chemistry 2008
©