An experimental solution to the missing hydrogens question surrounding the macropolyhedral 19-vertex boron hydride monoanion [B 19H22]-, a simplification of its synthesis, and its use as an intermediate in the first example of syn -B18H 22 to anti -B18H22 isomer conversion

Article abstract of DOI:10.1021/ic901976y

The macropolyhedral [B₁₉H₂₂]^- monoanion 1 and the dianion [B₁₉H₂₁]^2- 2 are synthesized in consistent 86-92% yields by the reaction of [PSH]⁺[syn-B ₁₈H₂₁]^- with BH₃(SMe₂) in 1,2-Cl₂C₂H₄ at 72 °C. [PS is an abbreviation for Proton Sponge, 1,8-bis-(dimethylamino)naphthalene. PSH is its protonated derivative.] The molecular structures of 1 and 2 were elucidated as their [PS{BH₂}]⁺ and [PS{BH₂}]₂ ⁺ salts 1a and 2a by single-crystal X-ray diffraction studies, in which all atoms were located, and supported by mass spectrometric analyses together with calculations of the cluster molecular geometries (ab ignitio and/or DFT) and of ¹¹B chemical shifts based on GIAO-DFT shielding tensors. Acidification of dianion 2 with CF₃COOH in acetonitrile, H₂SO₄ in dichloromethane, or aqueous HCl results in the clean formation of the monoanion [B₁₉H₂₂]^- 1. Conversely, shaking a concentrated acetonitrile solution of 1 in 0.5 M aqueous NaOH cleanly yields the [B₁₉H₂₁]^2- dianion 2. Reaction of a dichloromethane solution of 1 with a 36% aqueous solution of HCHO in the presence of H₂SO₄ quantitatively converts 1 at room temperature to a 1:1 mixture of the syn- and anti-isomers of B ₁₈H₂₂. This cluster dismantling process is the first example of a syn- to anti-B₁₈H₂₂ isomer conversion.

Full text of DOI:10.1021/ic901976y

4092 Inorg. Chem. 2010, 49, 4092–4098
DOI: 10.1021/ic901976y
An Experimental Solution to the “Missing Hydrogens” Question Surrounding the
Macropolyhedral 19-Vertex Boron Hydride Monoanion [B19H22]-, a Simplification
of Its Synthesis, and Its Use As an Intermediate in the First Example of syn-B18H22
to anti-B18H22 Isomer Conversion
†
†
†
Michael G. S. Londesborough,*,† Jonathan Bould,*,†,‡ Tomas Base, Drahomır Hnyk, Mario Bakardjiev,
ꢀꢁ
ꢁ
´
†
‡
ꢁ
ꢀ §
´
Josef Holub, Ivana Cısarova, and John D. Kennedy
†
ꢁ
Institute of Inorganic Chemistry v.v.i., Academy of Sciences of the Czech Republic, Rezꢁ by Prague, 250 68,
Czech Republic, ‡Department of Chemistry, University of Leeds, Leeds, LS6 9JT, United Kingdom, and
§Faculty of Natural Science, Charles University, Hlavova 2030, 128 43, Prague 2, Czech Republic
Received October 6, 2009
The macropolyhedral [B19H22]- monoanion 1 and the dianion [B19H21]2- 2 are synthesized in consistent 86-92%
yields by the reaction of [PSH]þ[syn-B18H21]- with BH3(SMe2) in 1,2-Cl2C2H4 at 72 °C. [‘PS’ is an abbreviation for
‘Proton Sponge’, 1,8-bis-(dimethylamino)naphthalene. ‘PSH’ is its protonated derivative.] The molecular structures of
1 and 2 were elucidated as their [PS{BH2}]þ and [PS{BH2}]2þ salts 1a and 2a by single-crystal X-ray diffraction
studies, in which all atoms were located, and supported by mass spectrometric analyses together with calculations of
the cluster molecular geometries (ab ignitio and/or DFT) and of 11B chemical shifts based on GIAO-DFT shielding
tensors. Acidification of dianion 2 with CF3COOH in acetonitrile, H2SO4 in dichloromethane, or aqueous HCl results in
the clean formation of the monoanion [B19H22]- 1. Conversely, shaking a concentrated acetonitrile solution of 1 in
0.5 M aqueous NaOH cleanly yields the [B19H21]2- dianion 2. Reaction of a dichloromethane solution of 1 with a 36%
aqueous solution of HCHO in the presence of H2SO4 quantitatively converts 1 at room temperature to a 1:1 mixture of
the syn- and anti-isomers of B18H22. This cluster dismantling process is the first example of a syn- to anti-B18H22 isomer
conversion.
Introduction
1960s and early 1970s resulted in the description of B13H19,9
B14H18,10 B14H20,11 B15H23,12 B16H20,13 and the syn and anti
The development of the full potential of boron-containing
cluster chemistry requires the development of cluster archi-
tectures larger than the 12-boron icosahedron.1 The high
molecular boron content and stability of such species make
them commercially interesting compounds for several appli-
cations in, for example, medicine2-4 and materials.2,5-8 The
rapid expansion of boron-containing cluster chemistry in the
14
isomers of B18H22 and B20H16,15,16 all resulting from the
intimate cluster fusion of smaller borane clusters. Further
binary boron hydride cluster structural motifs have subse-
quently been rare and essentially limited to B12H16 and its
conjugate [B12H15]- anion,17 the [B22H22]2- anion,18,19 the
(9) Huffman, J. C.; Moody, D. C.; Schaeffer, R. Inorg. Chem. 1976, 15, 227.
ꢁ
ꢀ
ꢁ
(10) Hermanek, S.; Fetter, K.; Plesek, J.; Todd, L. J.; Garber, A. R. Inorg.
*To whom correspondence should be addressed. E-mail: michaell@
iic.cas.cz (M.G.S.L.), jbould@gmail.com (J.B.).
(1) Shea, S. L.; Bould, J.; Londesborough, M. G. S.; Perera, S. D.;
Chem. 1975, 14, 2250.
(11) Huffman, J. C.; Moody, D. C.; Schaeffer, R. J. Am. Chem. Soc. 1975,
97, 1621.
Franken, A.; Ormsby, D. L.; Jelı
´
nek, T.; Stı
´
br, B.; Holub, J.; Kilner,
(12) Rathke, J.; Schaeffer, R. Inorg. Chem. 1974, 13, 3008.
ꢁ
ꢁ
ꢀ
ꢁ
(13) Plesek, J.; Hermanek, S.; Hanousek, F. Collect. Czech. Chem.
C. A.; Thornton-Pett, M.; Kennedy, J. D. Pure Appl. Chem. 2003, 75, 1239.
ꢁ
(2) Plesek, J. Chem. Rev. 1992, 92, 269.
Commun. 1968, 33, 699.
(3) Hawthorne, M. F. Angew. Chem., Int. Ed. 1993, 32, 950.
(4) Soloway, A. H.; Tjarks, W.; Barnum, B. A.; Rong, F. G.; Barth, R. F.;
Codogni, I. M.; Wilson, J. G. Chem. Rev. 1998, 98, 1515.
(5) Lu, X.; Shao, L.; Wang, X.; Chen, Q.; Liu, J.; Chu, W. K.; Bennet, J.;
Ling, P. Proc. Electrochem. Soc. 2001, 2001, 337–344.
(14) Olsen, F. P.; Vasavada, R. C.; Hawthorne, M. F. J. Am. Chem. Soc.
1968, 90, 3946.
(15) Friedman, L. B.; Dobrott, R. D.; Lipscomb, W. N. J. Am. Chem. Soc.
1963, 85, 3505.
(16) Miller, N. E.; Muetterties, E. L. J. Am. Chem. Soc. 1963, 85, 3506.
(17) Brewer, C. T.; Swisher, R. G.; Sinn, E.; Grimes, R. N. J. Am. Chem.
Soc. 1985, 107, 3558.
(18) Volkov, O.; Dirk, W.; Englert, U.; Paetzold, P. Z. Anorg. Allg. Chem.
1999, 625, 1193.
(6) Kawasaki, Y.; Kuroi, T.; Yamashita, T.; Horita, K. H., T.; Ishibashi,
M.; Togawa, M.; Ohno, Y.; Yoneda, M.; Horsky, T.; Jacobson, D.; Krull,
W. A. Nucl. Instrum. Methods Phys. Res., Sect. B 2005, 237, 25.
(7) Heo, S.; Hwang, H.; Cho, H. T.; Krull, W. A. Appl. Phys. Lett. 2006,
89.
(19) Hosmane, N. S.; Franken, A.; Zhang, G. M.; Srivastava, R. R.;
Smith, R. Y.; Spielvogel, B. F. Main Group Met. Chem. 1998, 21, 319.
(8) Harris, M. A.; Huynh, C. Solid State Technol. 2007, 50, 33.
r
pubs.acs.org/IC
Published on Web 03/29/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 9, 2010 4093
[fac-B20H18]2- and [B21H18]- anions,20and two species pre-
viously identified as the [B19H20]- monoanion and its con-
jugate [B19H19]2- dianion, reported in 2000.21 The [B19H20]-
empiricalformulagiven21 to thefirst of these 19-vertexcluster
species by Dopke et al. was subsequently questioned22 by
Jemmis et al., who proposed, on the basis of DFT calcula-
tions, that the [B19H20]- formulation was missing two hydro-
gen atoms and did therefore not fit into their mno rules for
fused borane clusters.23 Although Jemmis et al. did not
suggest the location of these hydrogen atoms from the
DFT calculations, Kiani and Hofmann subsequently pro-
posed24 a molecular structure for the monoanion based on
calculations. Considering the limited number of cluster types
available in the boron hydride structural continuum, parti-
cularly in the larger “macropolyhedral” category, it is im-
portant to experimentally certify those postulated by theory
and calculation. In this contribution, we present clear experi-
mental evidence confirming the macropolyhedral 19-vertex
boron hydride anions in question as being [B19H22]- and
[B19H21]2-, and our X-ray diffraction, NMR spectroscopic,
and calculation studies reasonably locate the positions of all
hydrogen atoms.
Scheme 1
cluster open-face hydrogen atoms for the originally published
{B19H20}- as well as for anions 1 and 2.)
DFT and subsequent RMP2(fc) calculations [the terms R
(restricted) and fc (frozen core) are in subsequent discussion
omitted for simplicity] were carried out contemporaneously
with the synthetic and crystallographic work in order to
support the full structural characterization of the anions 1
and 2. Figures 1 and 2 show both the crystallographically
determined and the DFT-calculated molecular structures of1
and 2. Their shared overall macropolyhedral fused-cluster
framework is notionally derived by the replacement of two
bridging hydrogen atoms on the open face of one of the two
10-vertex nido subclusters in the 18-vertex syn-B18H22 by a
{BH2} moiety to form one nido 10-vertex subcluster and one
nido 11-vertex subcluster with a shared two-atom B-B edge.
For the monoanion, 1, the single-crystal X-ray diffraction
study reveals 19 cluster boron atoms and 22 cluster hydrogen
atoms. All hydrogen atoms connected to boron were located
from the difference Fourier synthesis and were refined with
isotropic thermal parameters. The structural calculations for
1 minimized energetically with five bridging hydrogen atoms,
three on the 10-vertex subcluster and two on the 11-vertex
subcluster, mirroring the configuration found in the crystal
structure (Figure 1). Table 1 lists the calculated 11B nuclear
shieldings, expressed as 11B NMR chemical shifts, together
with the measured 11B and 1H-{11B} chemical shift data for 1.
Table 2 lists selected calculated and measured interatomic
dimensions for both 1 and 2. The calculated and measured
interatomic separations for 1 compare very well, with the
majority of the calculated distances being within 0.015 A of
the experimental values. Exceptions are in the long edge of
the 11-vertex subcluster, B(7)-B(11), where the calculated
distance is longer by þ0.025 A, and the long edge in the 10-
vertex subcluster, namely, B(7)-B(100), þ0.024 A. More
refined MP2/6-31G(d) calculations of the molecular geome-
try showed a slight improvement in these geometrical para-
meters to -0.011 and 0.020 A respectively but with no
decisive changes in the subsequent 11B chemical shift values.
That the calculated structures at the B3LYP/6-31þG(d,p)
and MP2/6-31G(d) levels and methodologies represent good
models of 1 may be taken from a consideration of the
positions of the cluster bridging hydrogen atoms. Thus, the
bridging hydrogen atoms spanning B(9)-B(10) and B-
(10)-B(11) are quite unsymmetrical in both the calculated
and the measured data with the distances to B(10) for
μH(910) and μH(101) being 1.453 and 1.475 A (calcd), which
correspond well to the measured distances of 1.429(17) and
1.470(19) A, respectively. A similar asymmetry is seen for the
bridging hydrogen atom adjacent to the B(70)-B(8) inter-
cluster “hinge” atoms: μH(708) to B(8) is 1.383 (calcd),
1.470(19) A (measd). The shorter arms for all three hydrogen
atoms are in the range 1.239-1.282 (calcd) and 1.172-
(18)-1.195(19) A (measd). These asymmetries, tending to-
ward an endo-terminal nature for the bridging hydrogen
Results and Discussion
The ready synthesis14,25-27 and stability of both syn and
anti isomers of B18H22 make them a convenient starting point
for advanced investigation of large-molecule boron hydrides.
In this context, Dopke et al. showed that the reaction of [anti-
B18H20]2- with BH2Cl(SMe2) or BHCl2(SMe2) produces a
19-boron anion, which they proposed to be of formulation
[B19H20]-, in ca. 70% yield. However, their reported proce-
dure requires the use of vacuum line techniques, rigorously
dried solvents, and difficult-to-handle potassium hydride.
Moreover, in our hands, this method has proved to be
unreliable with inconsistent yields. We have now found that
a simple stirring of warm 1,2-dichloroethane solutions of syn-
B18H22 and 1,8-bis-(dimethylamino)naphthalene together
with the nonhalogenated BH3(SMe2) followed by simple
filtration affords the [B19H22]- (1) and [B19H21]2- (2) anions
in consistent yields of 86-92%. The selective formation of 1
or 2 depends on the precise experimental conditions discussed
below. Each is characterized by single-crystal X-ray diffrac-
tion analysis. A comparison of the 11B NMR properties of
anions 1 and 2 with the species synthesized by Dopke et al.21
show them to be identical. However, our full structural
characterizations now reveal the empirical formulas [B19H22]-
and [B19H21]2-, each with two more hydrogen atoms than
proposed by those authors and now consistent with the
suggestions by Jemmis et al.22 and Kiani and Hofmann.24
More specifically, the previously “missing” hydrogen atoms
are now located experimentally. (See Scheme 1 for positions of
(20) Bernhardt, E.; Brauer, D. J.; Finze, M.; Willner, H. Angew. Chem.,
Int. Ed. 2007, 46, 2927.
(21) Dopke, J. A.; Powell, D. R.; Gaines, D. F. Inorg. Chem. 2000, 39, 463.
(22) Jemmis, E. D.; Balakrishnarajan, M. M.; Pancharatna, P. D. Inorg.
Chem. 2001, 40, 1730.
(23) Jemmis, E. D.; Balakrishnarajan, M. M.; Pancharatna, P. D. Chem.
Rev. 2002, 102, 93.
(24) Kiani, F. A.; Hofmann, M. Inorg. Chem. 2006, 45, 6996.
(25) Adams, L.; Hosmane, S. N.; Eklund, J. E.; Wang, J.; Hosmane, N. S.
J. Am. Chem. Soc. 2002, 124, 7292.
ꢁ
ꢁ
´
br, B.; Hanousek, F. Collect. Czech.
ꢁ
ꢁ
ꢀ
(26) Plesek, J.; Hermanek, S.; Stı
Chem. Commun. 1967, 32, 1095.
(27) Li, Y.; Sneddon, L. G. Inorg. Chem. 2006, 45, 470.

4094 Inorganic Chemistry, Vol. 49, No. 9, 2010
Londesborough et al.
Figure 2. Lower diagram: Representation of the crystallographically
Figure 1. Lower diagram: Representation of the crystallographically
2-
determined molecular structure of the [B19H21
]
anion 2 (50% prob-
determined molecularstructure ofthe [B19H22]
- anion1 (50%probability
ability thermal ellipsoids), as determined in its [PS{BH2}]2þ salt 2a, with
the cations omitted to aid clarity. Upper diagram: The B3LYP/6-31þG(d,p)
DFT calculated structure of 2. Selected measured and calculated distances
are shown in Table 2.
thermal ellipsoids), as determined in its [PS{BH2}]þ salt 1a, with the
cation omitted to aid clarity. Upper diagram: The B3LYP/6-31þG(d,p)
DFT calculated structure of1. Selected measuredandcalculated distances
are shown in Table 2.
atoms. The proton removed by the base comes from the
10-vertex subcluster on the bridging position between B(8)
and B(70), confirming the proposal of Dopke et al. The
11-vertex subcluster features a bridging hydrogen atom
spanning the B(10)-B(11) edge similar to that in the mono-
anion, but the asymmetry, which is apparent in the mono-
anion, is not evident here, either in the crystal structure or by
calculation. The second open-face hydrogen atom, H(9B),
exhibits a long distance of 1.92(4) A measured (2.130/2.073 A
DFT/MP2 calcd) to the neighboring B(10) vertex and is thus
tending toward an endo-terminal disposition, as may be
found in the [B11H14]- anion.28a Overall, the experimentally
observed bridging hydrogen atom characteristics are mir-
rored in the results of the structural calculations (Figure 2,
upper diagram), with, in particular, the more symmetrical
bridging nature of μH(10,11) and the substantial endo char-
acter of H(9B) being apparent in both cases.
The other measured and calculated interboron distances
around the open face of the 11-vertex subcluster in 2 are
reasonably in agreement, and the higher level MP2/6-31G(d)
calculation does not substantially improve the match in most
cases. Table 3 shows the calculated and measured 11B NMR
spectra for 2. The tentative assignments for the boron
resonances are based on a combination of GIAO-B3LYP/
6-31þG(d,p) calculated chemical shifts and (11B-11B
COSY), [1H-1H] COSY, and 1H-{11B selective} experi-
ments. A high confidence in assignments suggested by COSY
experiments is not warranted in this compound, which has a
large proportion of its resonances very close together and
overlapping. Additionally, the correspondence between the
calculated and measured boron spectra is not as close as is the
atoms, are also apparent from B3LYP, MP2, and single-
crystal X-ray studies of the single-cluster nido-[B11H14]-
anion itself, in which the facial bridging and endo-terminal
hydrogen atoms undergo rapid and unresolved fluxional
exchange.28a By contrast, no fluxional behavior was observed
in the proton spectrum for 1 between þ80 and -39 °C,
possibly due to a “fixing effect” of the replacement of an
endo-terminal/bridging hydrogen atom on the open face of
the [B11H14]- anion by the B(7)B(8) “intercluster hinge” in 1.
The majority of the 11B NMR assignments for the calculated
boron chemical shifts compare favorably with the assign-
ments made by Dopke et al. from [11B-11B] COSY correla-
tions.21 Some assignments are switched, for example, B(5)
and B(3) in Table 1, but these are only separated by 1 ppm,
which is well below the precision expected for the GIAO-
B3LYP method in these systems.
The [B19H21]2- dianion 2 was also isolated and initially
identified by NMR spectroscopy by comparison with the 11B
NMR data reported by Dopke et al., who had proposed a
[B19H19]2- formulation but who did not report a single-
crystal X-ray diffraction study. In their work, dianion 2
was obtained by deprotonation of monoanion 1, which we
have confirmed to be [B19H22]-; this suggests that the
[B19H19]2- formulation given by Dopke et al. is also short
by two hydrogen atoms. Our single-crystal X-ray diffraction
study of 2 (Figure 2, lower diagram) confirms the 19 cluster
boron atoms and now indeed shows 21 cluster hydrogen
(28) (a) Volkov, O.; Radacki, K.; Thomas, R. L.; Rath, N. P.; Barton, L.
ꢂ
J. Organomet. Chem. 2005, 690, 2736. (b) Farras, P.; Teixidor, F.; Branchadell,
V. Inorg. Chem. 2006, 45, 7947.

Article
Inorganic Chemistry, Vol. 49, No. 9, 2010 4095
Table 1. Measured 11B-{1H} Chemical Shifts Together with Calculated 11B
Chemical Shifts for the [B19H22][PS{BH2}] 1a
þ24.0/þ21.2 and þ6.5/þ4.9 ppm. Similarly, at the high-field
end, the resonances appearing at -28.7 ppm (assignment
B(6)) and -33.5 ppm (assignment B(40)) differ considerably
from the calculated values of -38.9/-15.7 and -41.9/-46.3
ppm. The attempts to model the system at the GIAO-
B3LYP/II//MP2/6-31G(d) level do not decisively improve
the comparison even though the overall calculated spectra, as
seen in Figure 3, are similar. However, the calculations are,
nevertheless, consistent with the experimentally determined
structure, and they reveal the locations of all the open-face
hydrogen atoms in reasonable positions as expected.
calculated B3LYP/
6-31þG(d,p)
tentative
assignmentsa
measuredc
lit.b
δ(11B) [δ(1H)]
δ(11B) [difference]
B(100)
B(30)
B(8)
10
30
7
þ12.8d [þ3.76]
þ12.5 [-0.3]
þ12.0 [-0.8]
þ3.8 [þ7.0]
þ12.8e [þ3.59]
-5.2 [singlet,
-3.24 bridging H]
þ5.5 [singlet]
-3.2 [þ2.86]
B(7)
8
-4.3 [-9.8]
-6.1 [-2.9]
-7.2 [-2.0]
-7.2 [þ0.3]
-8.8 [-1.3]
-9.8 [-2.3]
-10.4 [-2.9]
-10.4 [-2.0]
-20.3 [-1.7]
-23.2 [-2.1]
-22.7 [-0.9]
-23.8 [þ0.6]
-25.2 [þ0.2]
-26.0 [þ0.5]
-28.6 [-0.9]
-42.3 [0.0]
B(90)
B(4)
90
1
-5.2 [þ2.29]
B(2)
4
-7.5 [þ2.12]
These latter considerations perhaps suggest that the meth-
ods of calculation may not constitute the origin of the
discrepancy in the NMR data. A possible interpretation
may be a subtle isomerization upon dissolution or perhaps
fluxionality associated with the 11-vertex open face. In
accord withthefluxionality ofthe bridging and endo-terminal
hydrogen atoms of the [B11H14]- anion28a and other 11-
vertex systems,28b a conversion in 2 of the endo hydrogen
atom on B(9) to B(9)B(10) bridging character could be
visualized as the B(10)B(11) bridging hydrogen atom con-
comitantly converted to BH(11) endo character. To test this,
we carried out structural minimizations and calculations on
species in which the BH(9) open-face hydrogen atom was
fixed as endo, and then in which the other open-face hydrogen
atom was fixed as endo on BH(11) instead. The first approxi-
mated closely to the global minimum for 2 and to the solid-
state structure in 2a, with shieldings close to the calculated
values in Table 3. The second, with the open-face hydrogen
atom fixed as endo on B(11), was found to be very close in
energy, ca. 4.5 kJ mol-1 higher, which suggests that fluxion-
ality could occur and that the observed population of the two
forms would be essentially equal, and that mean shielding
values would be observed if the barrier was low, which is
reasonable. However, equal-weight averaging of the calcu-
lated boron nuclear shieldings for the two cases brought
some, but not all, of the extreme chemical shifts more into line
with those observed, but it also produced a bigger spread in
the central region, indicating that, if such a fluxionality is
involved, it is only part of the phenomenon (see Table 4 in the
Supporting Information for a full list of calculated 11B NMR
chemical shifts).
B(10)
B(80)
B(10)
B(1)
10
80
100
2
-7.5 [þ2.21]
-7.5 [þ2.39]
-7.5 [þ2.59]
-8.4 [þ2.59]
B(9)
9
-18.6f [þ2.10, -1.88(2)
]
]
B(11)
B(20)
B(6)
B(5)
B(3)
B(70)
B(40)
H2B.PS cation
11
20
6
-21.2g [þ0.65, -2.41(2)
-21.9 [þ1.46]
-24.4 [þ0.99]
-25.4 [-0.87]
-26.5 [þ0.64]
-27.7 [þ1.69]
-42.3 [-0.36]
þ1.3h [þ2.51]
3
5
70
40
þ1.8 [-0.5]
a Assignments based on a combination of GIAO-B3LYP/6-31þ
G(d,p) calculated chemical shifts, [11B-11B] COSY, and 1H-{11
B
selective} experiments. b Assignments taken from ref 21. c Proton de-
coupled δ(11B) chemical shifts measured in CD3CN at 295 K. d δ(11B)
þ13.1 ppm at 343 K. e δ(11B) þ12.2 ppm at 343 K. f Apparent doublet of
doublets 1J(11B-1H) 143 Hz, xJ(11B-1H) 49 Hz. g Unresolved coupling.
h Triplet, 1J(11B-1H) 120 Hz.
Table 2. Selected DFT Calculated and X-Ray Measured Interatomic Distances
for [B19H22][PS{BH2}] 1a and [B19H21][PS{BH2}]2 2aa
monoanion, 1
dianion, 2
DFT
Meas.
DFT
MP2
Meas.
B(1) -B(2)
1.784
1.773
1.786
1.772
1.764
1.851
1.951
1.885
1.891
1.961
1.863
1.930
2.066
1.779
1.816
1.240
1.453
1.475
1.239
1.333
1.300
1.337
1.322
1.282
1.383
1.776(2)
1.768(2)
1.784(2)
1.768(2)
1.757(2)
1.841(2)
1.936(2)
1.876(3)
1.907(3)
1.936(2)
1.851(2)
1.934(2)
2.042(2)
1.774(2)
1.795(3)
1.170(17)
1.429(17)
1.470(19)
1.172(18)
1.268(17)
1.245(17)
1.280(18)
1.223(18)
1.195(19)
1.470(19)
1.790
1.748
1.806
1.769
1.784
1.817
2.190
1.987
1.836
1.981
1.705
1.839
2.169
1.824
1.772
1.191
2.130
1.342
1.286
1.340
1.302
1.327
1.329
1.782
1.748
1.796
1.764
1.776
1.828
2.114
1.961
1.826
1.943
1.710
1.826
2.085
1.766
1.806
1.197
2.073
1.345
1.284
1.335
1.300
1.329
1.335
1.787(5)
1.763(5)
1.805(5)
1.795(5)
1.813(6)
1.848(4)
2.090(5)
1.933(5)
1.938(5)
1.942(4)
1.750(5)
1.830(5)
2.146(5)
1.823(5)
1.819(6)
1.13(4)
1.92(4)
1.41(4)
1.26(4)
1.29(2)
1.30(3)
1.26(3)
1.49(3)
B(1) -B(3)
B(1) -B(4)
B(1) -B(5)
B(1) -B(6)
B(7) -B(8)
B(8) -B(9)
B(9) -B(10)
B(10) -B(11)
B(11) -B(7)
B(8)-B(70)
Electrospray high resolution mass spectrometry gave the
same spectrum for both 1a and 2a, consistent with [B19H22]-.
The spectrum shows, with the exception of a small amount of
[B18H21]-, no fragmentation of the nonadecaborate anion.
Reversing the polarity of measurement provided the expected
spectrum for the [PS{BH2}]þ cation in both cases.
B(70)-B(80)
B(7)-B(100)
B(90)-B(100)
B(80)-B(90)
B(9)-H(910/9B)
H(910/9B)-B(10)
B(10)-H(101)
H(101)-B(11)
B(100)-H(109)
H(109)-B(90)
B(90)-H(890)
H(890)-B(80)
B(70)-H(708)
H(708)-B(8)
In the synthesis of 1a and 2a, the choice of solvent
was selected on the basis of polarity and boiling point: 1,2-
dichloroethane holds all starting materials in solution, but
the products conveniently precipitate, thereby facilitating
their quick and simple isolation. Importantly, the boiling
point of 1,2-Cl2C2H4 is high enough to permit sufficient
heating of the reaction mixture to boil off SMe2 from its
{BH3} adduct, thus liberating the reactive borane moiety.
Through repeated experimentation, it became clear that the
reaction temperature has an influence on the outcome of the
synthesis. If the reaction mixture in 1,2-dichloroethane is
brought to reflux temperature (83-84 °C), then, despite the
fact that the course of reaction appears the same to the eye,
the precipitated product isnot compound1a or2abut instead
[B12H12][PSH]2. However, if the temperature of reaction is
a For a fuller listing of distances, see the Supporting Information.
case for the monoanion 1 (see Figure 3). Thus, it may be
noted that the GIAO chemical shift values for the pair of
resonances at the low-field end of the spectrum, δ(11B) þ12.8
ppm (assignment B(8)) and þ12.8 ppm (assignment B(100)),
do not compare well with the DFT/MP2 calculated values of

4096 Inorganic Chemistry, Vol. 49, No. 9, 2010
Londesborough et al.
Table 3. Measured 11B-{1H} Chemical Shifts Together with Calculated 11B Chemical Shifts for the [B19H21][PS{BH2}]2 2a
tentative assignmentsa
lit.b
measured δ(11B)[δ(1H)]a,c
measuredb
DFT [difference]d
MP2
B(8)
8
þ12.8 [singlet]
þ12.8 [þ3.78, -1.84]
-2.2 [þ2.73]
þ13.6 (singlet)
þ11.8
-1.2
þ24.0 [þ11.2]
þ6.5 [-6.3]
þ0.5 [þ2.7]
-1.9 [þ1.3]
-5.0 [þ0.7]
-5.2 [þ1.5]
-9.6 [-1.8]
-10.3 [-1.5]
-12.3 [-2.5]
-16.4 [-4.8]
-16.5 [þ1.5]
-17.2 [þ2.6]
-17.4 [þ3.3]
-18.2 [þ3.3]
-19.1 [þ3.1]
-20.0 [þ2.8]
-22.8 [þ0.7]
-38.9 [-10.0]
-41.9 [-8.4]
þ1.9 [þ0.1]
þ21.2
þ4.9
B(100)
B(30)
B(1)
10
30
10
1
-0.6
-3.2 [þ2.35]
-3.2
-4.3
-5.6
B(10)
B(7)
B(2)
B(11)
B(5)
-5.7 [þ2.38]
-5.8
2
-7.3 [singlet]
-7.5
-16.9
-20.6
-14.9
-41.4
-22.3
-18.9
-21.3
-19.8
-17.7
-6.6
-22.2
-22.9
-15.7
-46.3
90
7
-7.8 [þ2.38]
-8.3
-8.8 [þ2.55]
-9.3(singlet)
-10.9
-12.1
-18.2
-19.5
-20.1
-20.7
-21.8
-22.4
-22.7
-29.7
-35.2
80
4
-9.8 [þ1.20]
B(9)
-11.6 [þ1.95]
-18.0 [þ1.94, -1.45]
-19.8 [þ1.46, -2.67]
-20.7 [þ0.97, -3.62]
-21.5 [-1.33]
-22.2 [þ1.78]
-22.8 [þ1.19]
-23.5 [þ1.55]
-28.9 [þ0.97]
-33.5 [þ0.66]
þ1.8 [þ2.57]
B(70)
B(80)
B(90)
B(20)
B(4)
9
11
100
70
3
B(3)
5
B(10)
B(6)
20
6
B(40)
H2B.PS cation
40
a Assignments based on a combination of GIAO-B3LYP/6-31þG[d,p] calculated chemical shifts and 11B-11B COSY, [1H-1H] COSY, and 1H-{11
DMF-d7. d Difference in the 11B NMR chemical shifts between the calculated and measured data for this work.
B
selective} experiments b Data from ref 21. Recorded at 160.4 MHz, measured in CH3CN, and referenced externally to BF3[OEt2] in C6D6. c Measured in
no 19-vertex products and OEt2 was the preferred solvent
for 19-boron product formation. Finally, in the system we
report here, it is of interest to note that only the syn-B18H22
isomer undergoes reaction with BH3(SMe2). Under identical
conditions, anti-B18H22 affords its conjugate dianion [anti-
B18H20]2-
.
The monoanion and dianion may be easily intercon-
verted. Thus, treatment of monoanion 1 with a base results
in dianion 2, and conversely, with acid, 2 reverts to the
monoanion 1. In the presence of H2SO4, the addition of a
36% aqueous HCHO to a solution of 1 in CHCl3 leads to a
fast and exothermic reaction affording an almost quanti-
tative conversion to a 1:1 mixture of syn and anti isomers of
B18H22 as determined by integrated 11B NMR spectrosco-
py. This suggests that, at room temperature, the oxidative
cluster dismantling procedure can occur with equal prob-
ability at either the B(9) or B(11) atom on the 11-vertex
subcluster of anion 1. However, the product isomer ratio is
sensitive to the temperature. Integrated 11B NMR spec-
troscopy showed that, at lower temperatures, 0 °C, a
mixture of 60% syn-B18H22 and 40% anti-B18H22 is
formed, and at higher temperatures, ca. 50 °C, the ratio
is reversed.
The synthesis reported here of the 19-boron monoanion 1
begins with syn-B18H22, and therefore its deboronation to a
1:1 mixture of syn- and anti-isomers of B18H22 constitutes the
first example of syn-B18H22 to anti-B18H22 isomer conversion.
Published syntheses only exist for mixtures14 of anti-B18H22 and
syn-B18H22 or for anti-B18H22 only,25-27 but no scheme has yet
been found for the formation of isomerically pure syn-B18H22.
However, synthesis of 1 using the procedure of Dopke et al.,
from anti-B18H22, followed by a subsequent deboronation of 1,
as described here, would effect a useful route for anti-B18H22 to
syn-B18H22 conversion, and it thus represents a potentially
useful method for syn isomer enrichment of B18H22.
A further noteworthy feature in compounds 1a and 2a is in
the formation of the boronium cation [PS{BH2}]þ in the
reaction. Such boronium cations, stabilized by monodentate
or polydentate amine ligands and paired with a borane
cluster anion such as in [H2B[N(CH3)2C2H5]2[B12H12] or
Figure 3. A stick diagram comparing the measured and the GIAO-
B3LYP/6-31þG(d,p) calculated 11B chemical shifts for the monoanion
[B19H22][PS{BH2}] 1 (upper) and the measured GIAO-B3LYP/6-31þ
G(d,p) and GIAO-B3LYP/II//MP2/6-31G* calculated shifts for the
dianion [B19H21][PS{BH2}]2 2 (lower).
kept lower than 62 °C, then insertion of a boron vertex into
the {B18} skeleton does not occur, and the [B18H21][PSH]
remains unchanged. In THF solvent, formation of
[B12H12][PSH]2 is favored over the production of 1a and
2a. A similar solvent sensitivity was noted by Dopke et al. in
their route, where reactions carried out in THF produced

Article
Inorganic Chemistry, Vol. 49, No. 9, 2010 4097
in a 50 mL 2-necked round-bottomed flask containing a magnetic
stir bar. Under a flow of N2 gas, 1,2-dichloroethane (15 mL) was
added, and the resultant yellow solution was allowed to stir at
room temperature under a stream of N2(g) for 15 min. 11B and
1H NMR spectroscopic measurements of a sample of the yellow
solution confirmed the complete formation of [syn-B18H21][PSH].
BH3(SMe2) (0.77 g, 0.95 mL, 10 mmol) was then injected into the
mixture, and the stirring continued at room temperature for
a further 20 min. The temperature was then raised to 72 °C
(( 1 °C), and the mixture stirred at this temperature for
approximately 16 h. During this time, [B19H21][PS{BH2}]2 2a
precipitates (0.59 g, 0.87 mmol, 87%) as a yellow semicrystalline
solid. The precipitate was collected by filtration, washed with
1,2-Cl2C2H4, and dried in the air at room temperature. Multi-
nuclear NMR studies revealed at this stage excellent levels of
purity suitable for further use. For analytical purposes, crystals
were obtained from CH3CN/Et2O solvent diffusion. The
HRMS spectrum of 1a shows a typical boron isotopomer
envelope with a cutoff peak at m/z 231.3508 [231.3485 calcd],
which is representative of the [11B19H22]- anion.
Figure 4. Representation of the crystallographically determined molecular
structure of the cation [C10H6(NMe2)2BH2]þ in [C10H6(NMe2)2BH2][B19-
H22] 1a (50% probability thermal ellipsoids). Selected distances (in A) to
N(1): B(20) 1.5964(18), C(1) 1.4789(16), C(11) 1.5064(18), C(12) 1.5094(18);
to N(2): B(20) 1.595(2), C(8) 1.4795(17), C(13) 1.5086(17), C(14) 1.5083(16).
Angles (in deg): N(1)B(20)N(2) 112.12(10), B(20)N(1)C(11) 113.65, (11)-
B(20)N(1)C(12) 103.13(11), B(20)N(2)C(13) 114.98(11), B(20)N(2)C(14)
103.08(11), B(20)N(1)C(1) 111.89(10), B(20)N(2)C(8) 110.99(10).
Synthesis of [B19H22][PS{BH2}] (1a). Concentrated H2SO4
was slowly dripped into a stirred CH2Cl2 suspension (15 mL) of
[B19H21][PS{BH2}]2 (0.59 g, 0.87 mmol) until complete dissolu-
tion occurred. At this point, stirring was discontinued and the
top CH2Cl2 layer separated, reduced in volume by half, and
layered with hexane to form light yellow crystals of
[B19H22][PS{BH2}] (0.38 g, 0.84 mmol, 97%) by slow diffusion
of the hexane into the dichloromethane solvent.
the [PSBH2][B2H7] salt, have long been recognized.29-31
However, this is the first structural characterization of the
[1,8-bis-(dimethylamino)naphthalene(BH2)]þ boronium ca-
tion (Figure 4). These types of compounds may be formed
from the reflux of diborane, B2H6, with amine ligand pre-
cursors; in the present work, it may be noted that diborane is
formed in refluxing solutions of BH3(SMe2).
Synthesis of Mixtures of [B19H21][PS{BH2}]2 (2a) and
[B19H22][PS{BH2}] (1a). Using the above procedure for the
synthesis of 2a, but with a 2-molar excess of Proton Sponge
rather than a 10-molar excess as above and a 5-molar excess of
BH3(SMe2) rather than a 10-molar excess, led to the formation
of a mixture of compounds 1a and 2a (approximately 60% 2a
and 40% 1a by integrated 11B NMR spectroscopy, combined
yield 92%). Isolation of this mixture, dissolution into CH3CN,
and either (a) shaking with 0.5 M aqueous NaOH or (b) addition
of CF3COOH led to the formation of pure 2a or 1a, respectively.
Formation of anti-B18H22 and syn-B18H22 from [B19H21][PS{BH2}]2
(2a). To a suspension of [B19H21][PS{BH2}]2 (0.22 g, 0.32 mmol) in
CHCl3 was added concentrated H2SO4 (1 mL) to effect a complete
dissolution of the solids (the 11B NMR spectrum revealed a
complete conversion of 2a to 1a), at which point a 36% water
solution of CH2O (0.5 mL) was added, producing discoloration of
the reaction mixture and effervescence. After the effervescence
stopped, the top CHCl3 layer was separated, the solvent removed,
and the B18H22 product extracted into boiling hexanes. Removal of
the hexane gave white, semicrystalline B18H22 (0.066 g, 0.31 mmol,
96%). Addition of the formaldehyde at 0 °C results in an approxi-
mately 60% syn-B18H22 and 40% anti-B18H22 mixture of isomers,
room temperature gives 50% syn-B18H22 and 50% anti-B18H22,
and þ50 °C gives 40% syn-B18H22 and 60% anti-B18H22, all as
determined by integrated 11B NMR spectroscopy.
Conclusion
A new, convenient, high-yield route to the nonadecaborate
system is described. Both monoanionic [B19H22]- and dia-
nionic [B19H21]2-, compounds 1 and 2, were experimentally
characterized by single-crystal X-ray diffraction analysis and
thereby provide a definitive solution to the question proposed
by Jemmis et al. regarding the missing hydrogen atoms in the
nonadecaborate anions first described by Dopke et al.
Additionally, a method for the controlled cluster dismantling
of 1 leading to mixtures of syn- and anti-B18H22 isomers, is
described. As 1 was made from syn-B18H22, this is the first
example of syn-B18H22 to anti-B18H22 isomer conversion. If
used in conjunction with the synthesis of1 from anti-B18H2 as
described by Dopke et al., then the dismantling to give a syn-
and anti-B18H22 mixture constitutes a useful new method for
the formation of the syn isomer, or for syn enrichment of a
syn- and anti-B18H22 mixture.
Experimental Section
Single-Crystal X-Ray Crystallographic Analyses. Crystal data
for [B19H22][PS{BH2}] 1a: C14H42B20N2, M = 454.70, yellow
prism, 0.4 ꢀ 0.3 ꢀ 0.25 mm3, monoclinic, space group P21/c
(No. 14), a = 12.8900(2), b = 17.7510(3), c = 11.9970(3) A,
β=91.1980(12)°, V=2744.44(9) A3, Z=4, Dc=1.100 g/cm3,
All experiments were carriedout in anambient atmosphere
unless otherwise stated. syn-B18H22 was synthesized by the
literature method.14 Proton Sponge (1,8-bis-(dimethyl-
amino)naphthalene), borane-dimethyl sulfide complex, and
1,2-dichloroethane solvent were purchased from Aldrich
Chemical Co. and used as received. NMR spectroscopy was
performed at 9.4 T on a Varian MERCURY 400 High
Resolution System. High resolution mass spectrometry,
HRMS, was carried out on a Finnigan Fleet instrument using
electrospray ionization using a CH3CN solvent.
F
000=960; Nonius KappaCCD area detector, Mo KR radiation,
λ=0.71073 A, T=150(2) K, 2θmax=55.0°, 50 670 reflections
collected, 6295 unique (Rint =0.037). The molecular structure
was solved with SIR9732 and refined with SHELXL-97.33 Final
GoF=1.027, R1=0.0502, wR2=0.1302, R indices based on
Synthesis of [B19H21][PS{BH2}]2 (2a). syn-B18H22 (0.22 g,
1 mmol) and Proton Sponge (2.2 g, 10 mmol) were placed together
(32) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo,
C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R.
J. Appl. Crystallogr. 1999, 32, 115.
(29) Shore, S. G.; Parry, R. W. J. Am. Chem. Soc. 1955, 77, 6084.
(30) Miller, N. E.; Muetterties, E. L. J. Am. Chem. Soc. 1964, 86, 1033.
(31) Keller, P. C.; Rund, J. V. Inorg. Chem. 1979, 18, 3197.
€
€
(33) Sheldrick, G. M. In SHELX-97; University of Gottingen: Gottingen,
Germany, 1997.

4098 Inorganic Chemistry, Vol. 49, No. 9, 2010
Londesborough et al.
5166 reflections with I >2σ(I) (refinement on F2), 425 para-
meters, 0 restraints. Lp and absorption corrections applied, μ =
0.052 mm-1. CCDC 747585.
sets. The final level of optimizations was carried out at a
correlated level of theory (MP2) with the 6-31G(d) basis set.
The GIAO NMR nuclear shielding predictions were performed
on the final optimized geometries. GIAO NMR nuclear shield-
ing predictions were performed both on the B3LYP and MP2
optimized geometries, and the computed 11B shielding values
were related to the BF3.OEt2 scale using the experimental δ(11B)
value of B2H6, þ16.6 ppm, as a primary reference.39
Crystal data for [B19H21][PS{BH2}]2 2a: C28H61B21N4, M =
680.82, colorless prism, 0.35 ꢀ 0.27 ꢀ 0.18 mm3, triclinic, space
group P1 (No. 1), a = 8.8508(18), b = 10.575(2), c = 11.461(2)
A, R = 81.11(3), β = 81.85(3), γ = 75.56(3)°, V = 1020.4(4) A3,
Z = 1, Dc = 1.108 g/cm3, F000 = 362; Mach3 Kappa-type 4-
circle goniostat, Mo KR radiation, λ = 0.71073 A, T = 150(2)
K, 2θmax = 59.4°, 28 061 reflections collected, 10 214 unique
(Rint = 0.0520). The molecular structure was solved and refined
with SHELXS-97 and SHELXL-97.34 Final GoF = 1.050, R1
= 0.0601, wR2 = 0.1494, R indices based on 7722 reflections
with I >2σ(I) (refinement on F2), 542 parameters, 3 restraints.
Acknowledgment. We thank the Grant Agency of the
Academy of Sciences of the Czech Republic (M200320904,
KAN400480701 and IAA400320901) and the Ministry of Edu-
cation, Youth and Sports (LC523) for financial support and
ꢁ ꢀ
´
Mgr. M Kvıcalova for assistance in mass spectrometry measure-
ments.
Lp and absorption corrections applied, μ = 0.057 mm-1
.
Absolute structure parameter = 4(2). CCDC 747586.
Computational Method. Calculations carried out in this study
were performed using the Gaussian 98 and Gaussian 03
packages. Structures were initially optimized with the STO-
3G* basis sets using standard ab initio methods.35,36 The next
level of optimizations, including frequency analyses to confirm
the true minima, together with GIAO NMR nuclear-shielding
predictions, were performed using the B3LYP functional,37,38
with the 6-31þG(d,p) and Huzinaga TZP (denoted as II) basis
Supporting Information Available: Listings of DFT calculated
atomic coordinates; mass spectra for compounds 1a and 2a;
Complete Table 2. DFT calculated and X-ray measured intera-
tomic distances for the monoanion [B19H22][PSBH2] 1a and the
dianion [B19H21][PSBH2]2 2a; Table 4. Measured 11B-{1H}
chemical shifts together with calculated 11B chemical shifts for
[B19H21][PSBH2]2 2a with the open-face endo hydrogen atom
fixed in (a) the BH(9) and (b) the BH(11) positions as well as the
equal-weight average of these calculations.This material is
available free of charge via the Internet at http://pubs.acs.org.
(34) Sheldrick, G. Acta Crystallogr., Sect. A 2008, 64, 112.
(35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;
Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,
J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 98; Gaussian, Inc:
Wallingford, CT, 2004.
(36) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R., Jr.; Zakrzewski, V. G.; Stratmann, R. E.; Burant,
J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M.
C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala,
P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov,
B. B.; Liu, A. L.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 03; Pittsburgh, 1998.
(37) Becke, A. D. J. Chem. Phys. 1993, 5648.
(38) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(39) Onak, T.; Tseng, J.; Diaz, M.; Tran, D.; Arias, J.; Herrera, S.; Brown,
D. Inorg. Chem. 1993, 32, 487.