Synthesis and characterization of μ,μ′-M(B5H8)2 (M=Cd, Hg and Zn): A reassignment of the NMR spectra for 2,3-μ-metalloderivatives of pentaborane(9)

Article abstract of DOI:10.1016/S0022-328X(00)00588-X

The series of compounds μ,μ′-M(B₅H₈)₂, where M=Cd (I), Zn (II), or Hg (III), has been prepared from the reaction between K(B₅H₈) and the metal chloride in THF at low temperatures. The species were characterized by multinuclear NMR spectroscopy, elemental analysis and mass spectrometry. The NMR spectral assignments were confirmed using ¹¹B{¹H} selective decoupling experiments and heteronuclear ¹¹B-¹H chemical shift correlation spectroscopy. The results appeared to be in conflict with the earlier assignment of the ¹¹B-NMR spectra for 2,3-μ-metalloderivatives of pentaborane(9) which appear in the literature. For systems with an electrophilic group replacing a bridging H atom in B₅H₉, the B atoms most shifted in the ¹¹B-NMR spectrum, which appear at the lower field, were assigned to the ones closest to the metal group. Our results for I-III suggested that the assignment was reversed, that is the higher field resonance is the one closest to the metal group. Thus, we reexamined the spectra of a series of related compounds including 2,3-μ-SnPPh₃(B₅H₈), μ,μ′-SnPh₂(B₅H₈), μ,1′-SnPh₂(B₅H₈), other Sn species and 2,3-μ-Cu(dppe)B₅H₈. In all cases our assignments were in accord with those for I-III and we suggest that this is general for all such B₅H₉ derivatives.

Full text of DOI:10.1016/S0022-328X(00)00588-X

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 614–615 (2000) 223–230
Synthesis and characterization of m,m%-M(B5H8)2 (M=Cd, Hg and
Zn): a reassignment of the NMR spectra for
2,3-m-metalloderivatives of pentaborane(9)
Hong Fang, Jonathan Bould, Janet Braddock-Wilking, Lawrence Barton *
Department of Chemistry and the Center for Molecular Electronics, Uni6ersity of Missouri-St. Louis, St. Louis, MO 63121, USA
Received 24 April 2000
Dedicated to Professor Sheldon G. Shore in celebration of his 70th birthday and a career of major contributions to inorganic chemistry.
Abstract
The series of compounds m,m%-M(B5H8)2, where M=Cd (I), Zn (II), or Hg (III), has been prepared from the reaction between
K(B5H8) and the metal chloride in THF at low temperatures. The species were characterized by multinuclear NMR spectroscopy,
elemental analysis and mass spectrometry. The NMR spectral assignments were confirmed using 11B{1H} selective decoupling
experiments and heteronuclear 11B–1H chemical shift correlation spectroscopy. The results appeared to be in conflict with the
earlier assignment of the 11B-NMR spectra for 2,3-m-metalloderivatives of pentaborane(9) which appear in the literature. For
systems with an electrophilic group replacing a bridging H atom in B5H9, the B atoms most shifted in the 11B-NMR spectrum,
which appear at the lower field, were assigned to the ones closest to the metal group. Our results for I–III suggested that the
assignment was reversed, that is the higher field resonance is the one closest to the metal group. Thus, we reexamined the spectra
of a series of related compounds including 2,3-m-SnPPh3(B5H8), m,m%-SnPh2(B5H8), m,1%-SnPh2(B5H8), other Sn species and
2,3-m-Cu(dppe)B5H8. In all cases our assignments were in accord with those for I–III and we suggest that this is general for all
such B5H9 derivatives. © 2000 Elsevier Science B.V. All rights reserved.
Keywords: Boranes; Metalloboranes; NMR spectroscopy; Group 12 metals
1. Introduction
The main group units thus introduced into the cluster
contain elements from p-block Groups 2 [6], 13 [7], 14
[8], and 15 [9]. The first pentaborane(9) derivative with
a metal group in a bridging position was prepared by
Brice and Shore [10] and characterized fully [11]. Now,
examples of bridged substituted metalloboranes based
on pentaborane(9) are known for nickel [12,13], palla-
dium [13,14], platinum [13,14], copper [10,11,15,16],
silver [15], gold [15,17], beryllium [6], zinc [6], cadmium
[18], silicon [8a–f,19], germanium [8b,c], tin [8b,8g,20]
and lead [8b]. The generic structure of these species is
given in Fig. 1. Pentaboranes(9) bridged by mercury
[21] are also known and we had prepared some related
SnPPh2-bridged pentaboranes(9) [22]. Herein we extend
this latter work to some Zn and Cd bridged derivatives
and also clarify some aspects of the 11B-NMR spectra
for the 2,3-m-metallopentaboranes [23].
Nido-pyramidal boranes and carboranes such as
B5H9, B6H10 and C2B4H8 contain bridging hydrogen
atoms that are acidic and which may be removed with
strong bases. This was demonstrated independently by
the research groups of Onak [1a], Gaines [1b] and
Shore [1c] in 1967. The resulting anions were shown to
be susceptible to electrophilic attack by Lewis acids in
studies, reported soon after initial discovery by the
three groups, in which the original bridging hydrogen is
replaced by the Lewis acid [2]. For B5H9, the Lewis
acids used range from simple species such as a proton
[1b,c] and borane(3) [3] through more complex main
group [2a,4] species and transition metal moieties [5].
* Corresponding author. Fax: +1-314-5165342.
E-mail address: lbarton@jinx.umsl.edu (L. Barton).
0022-328X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00588-X

224
H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
neck and a glass stopper on the other. After evacuation
on the vacuum line, an excess of 0.8 ml of B5H9 (7.7
mmol) and 15 ml of THF were condensed in the vessel
at −198°C. Deprotonation of the B5H9 was carried out
by stirring at −78°C for about 2 h until H2 evolution
ceased. The mixture was cooled again to −198°C, and
evacuated, flushed with N2 and a tipper tube containing
550 mg anhydrous CdCl2 (3.0 mmol) was attached to
the vessel under positive pressure of N2. After evacua-
tion, the CdCl2 was tipped in the flask and the reaction
mixture was stirred at −78°C overnight and at −35°C
for 3 h. The reaction mixture was then warmed to 0°C
over a period of 2 h and stirred at that temperature for
an additional 1.5 h. Traces of a greenish metallic-like
deposit, assumed to be Cd, were observed. THF was
removed under vacuum at room temperature (r.t.) to
give a viscous residue, which was extracted with 10 ml
of CH2Cl2. The contents of the flask were filtered at
−78°C by suction through a frit on the vacuum line to
remove KCl. The product crystallized from CH2Cl2
during filtration. Element analysis for Cd before the
crystals were dried gave 28.56% and 29.45% obs. (calc.
Fig. 1. Generic structure for 2,3-m-(MLn)B5H8 species.
2. Experimental
2.1. General
Reactions were carried using a vacuum line, Schlenk
tubes and an inert atmosphere box employing standard
methods [24]. All solvents were reagent grade and were
dried and distilled prior to use and stored in Pyrex
vessels with Teflon stopcocks. B5H9 was obtained from
laboratory stock, and distilled on the vacuum line
before use. Anhydrous ZnCl2 and CdCl2 were dried
from ZnCl2 or CdCl2·2.5H2O (Fisher Scientific) in a
vacuum oven at 120°C for 48 h. HgCl2 (Fisher Scien-
tific) was purified by sublimation above 100°C under
vacuum. KH (Alfa) obtained as a 20–25% mineral oil
suspension, was washed repeatedly with pentane to
remove the oil until it was a free-flowing white powder.
The activity of the powder, in reactions with methanol,
was 85–95%. m,m%-Hg(B5H8)2 [21], 2,3-m-SnPh3(B5H8)
1
for Cd(B5H8)2·2THF: 29.53%). H-, 11B-NMR and IR
spectra
confirmed
the
composition
to
be
Cd(B5H8)2·(THF)x after evacuation at ambient temper-
ature for 15 min on the vacuum line. This species is
soluble in THF, Et2O, CH2Cl2 and CHCl3 and spar-
ingly soluble in pentane, hexane. The product was dried
further at r.t. affording 410 mg of white solid
Cd(B5H8)2 (65% yield) which melts at 95°C with decom-
position. Analysis for Cd gave 49.06% and 48.28%
(calc. for Cd(B5H8)2: 47.55%). Commercial elemental
analyses on Cd(B5H8)2 were unsuccessful due to its high
sensitivity to air and moisture, and also to incomplete
combustion since the compound does not contain car-
bon. Schwarzkopf Microanalytical Laboratory reported
that the species exploded when treated with nitric acid.
Cd(B5H8)2 is slightly soluble in CH2Cl2 and CHCl3. If
dry isolated Cd(B5H8)2 is dissolved THF at ambient
temperature, it completely decomposes to [B9H14]−.
Attempts to grow crystals resulted in the formation of
good crystals of K[B9H14] if traces of KCl were present
[25]. The IR spectrum of Cd(B5H8)2·(THF)x run as a
KBr pellet, showed weak absorbencies of the coordi-
nated THF at 2965(w), 2878(w), 1024(w), 884(w). Ab-
sorbencies were observed for the B5H8 cage at 2522(s),
1797(w). Unassigned bands are 1376(m, br), 1263(m),
1089(w), 959(s), 806(m), 708(w), 605.8(w), 565(w).
NMR data are given in Table 1. The mass spectrum
exhibited the expected Cd(B5H8)2 envelope with a cut-
off at m/q 242 attributed to [B10H16Cd]+. The observed
m/q values (relative intensity) for the Cd(B5H8)2 molec-
ular ion cluster were 231(0), 232(6.17), 233(27.66),
234(48.83), 235(60.04), 236(85.90), 237(100), 238(87.12),
239(73.24), 240(50.45), 241(17.70), 242(7.00). The calcu-
lated m/e data were 231(5.25), 232(11.25), 233(26.15),
[20],
m,m%-Sn(B5H8)2Ph2,
m,m%-Sn(B5H8)2Ph2,
m1%-
Sn(B5H8)2Ph2 [22], and 2,3-m-Cu(dppe)(B5H8) [16] were
prepared as described in the literature. NMR spectra
were obtained on a 500 MHz Bruker ARX-500 NMR
spectrometer operating at 500 and 160.4 MHz to ob-
1
serve H and 11B resonances, respectively. Assignments
were confirmed by selective decoupling experiments and
heteronuclear 2D 11B–1H correlation spectra. 11B
chemical shifts are reported in ppm, positive signs
denoting a shift at a lower field with respect to
Et2O·BF3 reference (0.0 ppm). 1H-NMR shifts was
measured relative to SiMe4. Mass spectra were run as
solids at 70 eV on a Varian/Mat 311A spectrometer
equipped with a Technivent data system. IR spectra
were run as KBr pellets on a Perkin–Elmer 1604 FTIR
spectrometer. Elemental analysis for Cd were obtained
on a GBC Model 904BT double-beam AAS instrument
in the flame atomization mode by comparison with
standard Cd2+ solutions.
2.2. Preparation of Cd(B5H8)2
Potassium hydride (250 mg, 5.3 mmol calc. from 85%
activity) was weighed in a glove box into a 50 ml two
neck reaction flask, equipped with a extractor on one

H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
225
Table 1
1H-NMR data for Group 12 metal-bis[pentaboranyl(9)] complexes a
Compound
1H
1H{11B)
m,m%-Zn(B5H8)2·2THF
−3.61[s, br, 4H, Hm(3,4), Hm(2,5),
Hm(3%,4%), Hm(2%,5%)]
−2.19 [s, br, 2H, Hm(4,5), Hm(4%,5%)]
0.44 [q, 2H, H(1), H(1%), J(11B–1H)=170 Hz]
1.3–3.0 [m, br, 8H, Ht(2–5), Ht(2%-5%), Junres]
1.96 [s, 6H, Hb THF], 3.89 [s, 6H, Ha THF]
0.44 [s, 2H, H(1), H(1%)]
1.82 [s, 4H, Ht(2,3), Ht(2%,3%)]
2.39 [s, 4H, Ht(4,5), Ht(4%,5%)]
m,m%-Cd(B5H8)2-(THF)x
−3.61 [s, br, 4H, Hm(3,4), Hm(2,5)
Hm(3%,4%), Hm(2%,5%)]
−2.03 [s, br, 2H, Hm(4,5), Hm(4%,5%)]
0.68 [q, 2H, H(l), H(1%), J (11B–1H)=172 Hz]
1.3–3.1 [m, br, 8H, Ht(2–5), Ht(2%–5%), Junres]
1.93 [s, 2H, Hp THF], 3.81 [s, 2H, Ha THF]
0.68 [s, 2H, H(1), H(1%)]
1.88 [s, 4H, Ht(2,3), Ht(2%,3)]
2.49 [s, 4H, Ht(4,5), Ht(4%,5%)]
b,c
m,m%-Cd(B5H8)2·PPh3
−3.41 [s, br, 4H, Hm(3,4), Hm(2,5),
Hm(3%,4%), Hm(2,5%)]
−2.38 [s, br, 2H, Hm(4,5), Hm(4%,5%)]
0.49 [q, 2H, H(l), H(1%), J(11B–1H)=165 Hz]
1.3–3.1 [m, br, 8H, Ht(2–5), Ht(2%–5%), Junres]
2.26 [s, 4H, Ht(4,5), Ht(4%,5%)]
−2.51 [s, 4, Ht(4,5), Ht(4%,5%)]
0.49 [s, 2H, H(1), H(1%)]
1.84 [s, 4H, Ht(2,3), Ht(2%,3%)]
m,m%-Hg(B5H8)2
−2.48 [s, br, 4H, Hm(3,4), Hm(2,5)
Hm(3%,4%), Hm (2%,5%)]
−2.48 [s, sh, 4H, Hm(3,4), Hm(2,5),
Hm (3%,4%), Hm (2%,5%), J(199Hg–1Hm=100 Hz]
−1.44 [s-sh, 2H, Hm(4,5), Hm(4%,5%), J
−1.44 [s, br, 2H, Hm(4,5), Hm(4%,5%)]
(
199Hg–1Hmm=152 Hz]
−1.58 [q, 2H, H(1), H(1%), J (11B–1H)=176 Hz]
2.47 [q, 4H, Ht(2,3), Ht(2%,3%), Junres]
2.51 [q, 4H, Ht(4,5), Ht(4%,5%), Junres]
−1.58 [s, 2H, H(1), H(1%)]
2.47 [s, 4H, Ht(2,3), Ht(2%,3%),
J(199Hg–1Ht)=160 Hz]
2.51 [s, 4, Ht(4,5), Ht(4%,5%)]
a All spectra were recorded at 500 MHz at room temperature in CDCl3 and assigned by 1H-{11Bselective} and 2D-HETCOR experiments.
Abbreviations: s, singlet; d, doublet; q, quartet; m, multiplet; br, broad; sh, sharp; unres, unresolved coupling.
b Resonances for the phenyl groups on PPh3 and omitted.
c 31P data: l(31P) (CDCl3, 223 K) +20.83, singlet {1H}.
234(51.38), 235(79.09), 236(97.34), 237(100), 238(87.00),
239(60.43), 240(30.81), 241(11.68), 242(4.67). The fit is
not perfect, and this may be ascribed to the low inten-
sity of the molecular ion cluster, which is affected by
baseline noise. The [M–B5H8]+ cluster, has a base peak
at m/e 175, and cut off at m/e 179 corresponding to
[B5H8Cd]+. The observed m/e data were 170(12.83),
171(27.58), 172(57.86), 173(76.26), 174(98.93), 175(100),
176(98.36), 177(59.28), 178(8.44), 179(19.92), and the
calculated m/e data were 170(5.80), 171(19.33),
172(47.81), 173(79.73), 174(95.59), 175(100), 176(86.05),
177(58.36), 178(16.566), 179(13.22). This fit is very
good. A qualitative test 0.2 M AgNO3 confirmed the
absence of Cl−.
which time deprotonation was complete. The H2 was
pumped away at 198°C and under positive N2 flow, a
side arm tip tube containing ZnCl2 (410 mg, 3 mmol)
was attached to the second neck of the reaction flask.
The reaction mixture was stirred at −78°C overnight.
The ZnCl2 appeared to dissolve in the THF. The mix-
ture was warmed to −35°C and stirred for 3 h and
then at 0°C for an additional 0.5 h. The solvent was
removed in vacuum to give a viscous compound, which
was evacuated for 15 min at ambient temperature. The
residue was extracted with 15 ml of CH2Cl2 and the
resulting solid KCl was filtered. Some crystals formed
under the filter frit. The resultant clear filtrate was
reduced to 2 ml under vacuum and 10 ml pentane
resulted in a trace of precipitate so the solvent was
removed and the resulting compound was dried under
vacuum for 1 h at r.t. to afford a white solid 390 mg,
42% yield of Zn(B5H8)2·2THF. The compound melts
with decomposition above 98°C and is soluble in
CHCl3, CH2Cl2, THF, Et2O and C6H6. Some decompo-
sition to [B9H14]− can be observed from solutions in
THF, Et2O and C6H6. Mass spectra data were support-
2.3. Preparation of Zn(B5H8)2
KH 260 mg (5.52 mmol, calc. as 85% activity) was
placed in a 50 ml two-necked flask with one neck fitted
with an extractor and the other neck stoppered. B5H9
(0.8 ml, 0.77 mmol) was condensed at −198°C. The
mixture was warmed to −78°C and stirred for 3 h after

226
H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
ive of the formulation. The observed m/e values for the
molecular ion [Zn(B5H8)2]+ were: (relative intensity)
183(0), 184(0.73), 185(5.65), 186(22.68), 187(56.29),
188(93.01), 189(100), 190(84.92), 191(67.48), 192(50.30),
193(25.95), 194(8.56) and the calculated data are
183(0.20), 184(1.38), 185(6.68), 186(22.71), 187(54.03),
188(88.80), 189(100), 190(82.68), 191(64.24), 192(50.38),
193(28.78), 194(11.31). The fit is excellent; details of the
molecular ion envelopes, along with the calculated ones
are available as supplemental information. The IR spec-
trum clearly indicated THF coordination with absorben-
cies (cm−1) at 2982(m), 2889(m), 1458(w) 1346(m),
1177(w), 1022(s), 883(s). Two bands for wB–Ht at 2579(s),
and 2517(s) and one bond for wB–Hm at 1803(w, br) and
unassigned bonds at 1376(m), 1296(w), 1262(w), 1095(w),
1049(m), 960(m), 943(m), 922(m), 859(m), 808(w),
749(w), 713(w), 651(w), 605(s) and 578(m) were observed.
Cl− was shown to be absent using the AgNO3 test.
CHCl3. Elemental analysis and mass spectral data sup-
port the proposed formulation. I is quite different from
[m,m%-Cd(B5H8)2·2THF], which is very soluble in CHCl3,
CH2Cl2, THF, Et2O and pentane. Elemental analysis for
Cd were obtained for both m,m%-Cd(B5H8)2·2THF, before
removal of solvent, and m,m%-Cd(B5H8)2 after drying.
Invariably THF solvates the complex in amounts ranging
from 0 to 2 mol. The 1H-NMR spectrum in CDCl3 gives
broad resonances at −3.46 ppm and −2.03 ppm in 1:2
area ratio, respectively, in the upfield region assigned to
bridging H atoms, together with a quartet at 0.68
(J=172 Hz) ppm assigned to the apical H atom and
overlapped broad quartets between 1 and 3 ppm. On
11B-decoupling, the two upfield resonances become
sharper, the apical H resonance becomes a singlet and the
low field overlapped resonances are seen as two singlets
at 1.88 and 2.49 ppm, in area ratio 2:2. These latter are
assigned to the basal boron atoms suggesting that the two
B5H8 cages are each bonded to Cd at positions bridging
basal H atoms. The 160.4 MHz 11B-NMR spectrum of
I in CDCl3 affords a high-field doublet, of relative area
1, at −49.0 ppm, assigned to the apical boron atoms,
and two well-separated, equally intense doublets at
−18.3 ppm, (J=130 Hz) and −8.5 ppm (J=152 Hz)
assigned to the basal boron atoms. These observations
are in contrast to the spectra of the only known cadmium
complexes, Cd(m-B5H8)Cl(PPh3) and Cd(m-1-Br–
B5H7)Cl(PPh3), for which 11B-NMR spectra were run at
28.87 MHz affording broad overlapped basal boron
resonances with ca. 4 and 5 ppm separation, respectively
[18]. The 11B resonances for I and m,m%-Cd(B5H8)2·2THF
collapse to singlets on proton-decoupling. Unfortunately
we were unable to grow crystals suitable for X-ray
diffraction analysis; however, elemental analysis and
mass spectral data support the formulation derived from
the NMR data. Similar data were obtained for [m,m%-
Zn(B5H8)2] (II), also a new compound, and for [m,m%-
Hg(B5H8)2] (III), which was prepared for comparisons
purposes. NMR data for I–III are given in Table 1 and
a proposed structure for I, typical of the class of
compounds, is given in Fig. 2.
2.4. Tests for possible rearrangement
Samples of about 20 mg [m,m%-Cd(B5H8)2·2THF] was
dissolved in THF, Et2O and CDCl3 separately in NMR
tube at r.t., and NMR sample was sealed under vacuum.
The 11B spectra were monitored periodically over several
months. The only change detected was the appearance
of a small amount of [B9H14]−.
2.5. Reaction of v,v%-Cd(B5H8)2·2THF with PPh3
[m,m%-Cd(B5H8)2·2THF], 30 mg (0.08 mmol), was
stirred with 30 mg (0.12 mmol) PPh3 at −78°C in
CH2Cl2 for 8 h, solvent was evaporated at −35°C until
the solid was dry. An NMR sample was prepared in
CDCl3 and 11B and 1H are given in Table 1, implying the
existence of m,m%-Cd(B5H8)2PPh3.
3. Results and discussion
The assignments of the 11B-NMR spectra for I–III
differ from those previously published for pentabo-
rane(9) derivatives with a substituent in a bridging
position replacing a bridging hydrogen atom. The orig-
inal assignments of the 11B-NMR spectra were based on
the assumption that the basal boron resonances which
were shifted the least from those in pentaborane(9) would
be the B(4,5) atoms [8a–c,8f,18] (see Fig. 1). Thus the
lowest field resonances in the 11B-NMR spectra of
2,3-m-bridge-substituted nido-pentaboranes were as-
signed to the boron atoms bonded to the metal sub-
stituent. This was based on the observation that these
resonances were shifted from those in B5H9 itself whereas
the upfield resonance was not and on the assumption that
the boron atoms B(2,3), to which the metal substituent
is bonded, are most likely to be shifted and broadened.
The reaction between K(B5H8) and anhydrous CdCl2
in 2:1 ratio in THF affords [m,m%-Cd(B5H8)2] (I) in 65%
yield. The product is slightly soluble in CH2Cl2 and
Fig. 2. Proposed structure for species I.

H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
227
Thus in the original series of Group 14 derivatives
discovered and several others since then, assignments
were based on this premise. For example, in 2,3-m-
(Me3Si)B5H8, the lower field basal boron resonance
which is seen at l=8.7 ppm was assigned to B(2,3)
and the higher field resonance, observed at 13.2 ppm
(versus 12.5 ppm for B5H9 [26]), was assigned to
B(4,5) [8b]. Analogous assignments persisted in more
recent work [8c,18,13,20,27] but we noticed some
anomalies with these assignments, first when examin-
ing the spectra of [m,2-SnPh2(B5H8)2] [22,28]. We
found for this system that B(4), which is the boron
atom furthest from the basal boron atom to which a
tin atom is sigma-bonded, falls at the lowest field
strength. This conflicts with previous assignments for
pentaboranes(9) substituted at the 2-position [8c,20].
Thus, we reexamined the spectra of some related
bridged-substituted pentaborane(9) species we had
cies m,m%-SnPh2(B5H8)2, m,1%-SnPh2(B5H8)2, 2,3-m-Cu-
(dppe)B5H8 afforded similar results. In each case the
low field resonance in the 11B-NMR spectrum was
assigned to B(4,5).
Our 11B-NMR assignments for I–III were also
made on the basis of selective 11B decoupling of the
proton spectra and the data we obtained were in ac-
cord with our new results for [2,3-m-(SnPh3)B5H8],
which had been obtained simultaneously, and the
other species listed above. That is the low field reso-
nance in the 11B spectra of all these species corre-
sponds to the atoms B(4,5). In order to confirm these
results, we conducted heteronuclear 11B–1H chemical
shift correlation spectroscopy experiments. Fig. 3
shows the 11B–1H correlated 2D-NMR spectra for
2,3-m-(SnPh3)B5H8, m,m%-Zn(B5H8)2, m,m%-Cd(B5H8)2,
m,m%-Hg(B5H8)2. All give similar results. The low field
basal 11B resonance correlates with all three bridging
H atoms and thus is assigned to B(4,5) and the up-
field basal boron resonance correlates only with the
bridging H resonance of area 2, and is assigned to
B(2,3). We conducted the same experiments for m,m%-
SnPh2(B5H8)2, m,1%-SnPh2(B5H8)2, and 2,3-m-Cu(d-
ppe)B5H8, obtaining similar results in each case thus
confirming our reassignments. Since we ran spectra
on a 500 MHz spectrometer which had been previ-
ously obtained at lower field strength, we report
herein the newly recorded spectra for 2,3-m-
(SnPh3)B5H8, and 2,3-m-Cu(dppe)B5H8. The data are
tabulated in Table 2 and they provide much more
details than was available from spectra run at lower
field strength. Of note is our observation of the
119Sn–1H coupling constant of 67 Hz for the terminal
H atoms on the B atoms which are bridged by the Sn
1
prepared previously by conducting 500 MHz H{11B}-
NMR selective decoupling experiments. The bridging
H atoms in 2,3-m-substituted pentaboranes(9) exhibit
two broad resonances in area ratio 1:2, corresponding
to Hm(4,5) and Hm(3,4)/Hm(2,5), respectively (see Fig.
1). We observed that when the frequency correspond-
ing to the higher field basal boron resonance in [2,3-
m-(SnPh3)B5H8] was irradiated, the bridging H atom
resonance of relative area 2 sharpened. Also, irradia-
tion at the other (lower field) basal boron atom field
position resulted in the sharpening of both bridging
H atom resonances, the one of area 1 being the most
affected. This implies that the lower field resonance,
which couples to all the Hm resonances, is B(4,5) and
the higher field resonance which only couples to the
Hm resonance of area 2, is B(2,3), the B atoms adja-
cent to the metal. Analogous experiments for the spe-
Table 2
a
1H- and 11B-NMR data for 2,3-m-(SnPh3)(B5H8) and 2,3-m-Cu(dppe)B5H8
Compound
11B
H
1H{11B}
2,3-m-(SnPh3)(B5H8)
−9.7 [d, 2B, B(4,5),
J(11Bꢀ1H)=151 Hz]
−13.6 [d, 2B, B(2,3),
J(11Bꢀ1H)=148 Hz]
−49.4 [d, 1B, B(1),
J(11Bꢀ1H)=176 Hz]
−2.58 [s, br, 2H, Hm(3,4), Hm(2,5)
−1.97 [s, br, 1H, Hm(4,5)]
0.91 [q, 1H, H(1),
J(11Bꢀ1H)=175 Hz]
2.0–3.5 [m, br, 4H,
0.91 [s, 1H, H(1)]
2.58 [s, 2H, Ht(4,5)]
2.39 [s, 2H, Ht(2,3)
J(119Snꢀ1H)=67 Hz]
Ht(2–5), Junres.
]
2,3-m-Cu(dppe)B5H8
−13.1 [d, 2B, B(4,5),
J(11Bꢀ1H)=149 Hz]
−16.37 [d, 2B, B(2,3),
J(11Bꢀ1H)=124 Hz]
−49.1 [d, 1B, B(1),
J(11Bꢀ1H)=161 Hz]
−3.00 [s, br, 2H, Hm(3,4),
Hm(2,5)]
−2.67 [s, br, 1H, Hm(4,5)]
0.12 [q, 1H, H(1),
0.12 [s, 1H, H(1)]
1.96 [s, 2H, Ht(2,3)]
2.09 [s, 2H, Ht(4,5)]
J(11Bꢀ1H)=161 Hz]
1.3–3.1 [q, br, 4H, Ht(2–5), Junres.
]
2.47 [s, 4H, from CH2ꢀCH2]
a All spectra are recorded at 500 MHz at room temperature in CDCl3 and assigned by 1H-{11Bselective} and 2D HETCOR experiments.
Abbreviations: s=singlet, d=doublet, q=quartet, m=multiplet, br=broad, sh=sharp, unres=unresolved coupling.

228
H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
Fig. 3. 11B–1H correlated 2D-NMR spectra for (a) m,m%-Zn(B5H8)2, (b) m,m%-Hg(B5H8)2, (c) 2,3-m-(SnPh3)B5H8, and (d) m,m%-Cd(B5H8)2.
adjacent 11B nuclei is known to be quite small when
moiety. Such values have not been reported previously.
they are H-bridged [30], but coupling through the
bridging H atoms is not [31]. Although we were able to
discern the presence of 113Cd satellites in the proton
spectra of I, they were too much overlapped to allow
the reporting of J(113Cd–1H) values.
Thus, we conclude that for all 2,3-m derivatives of
B5H9, the basal boron resonance at low-field arises
from B(4,5) and the upfield one from B(2,3). The only
other system for which the assignment of the basal
boron resonances is the same as ours was for 2,3-m-
(C6H5)2P-B5H8, although this system is not quite the
same in that the bridging moiety, (C6H5)2P is bonded to
the boron atoms through two normal two-center, two
electron bonds, rather than by a three-center two-elec-
tron bond [32]. Using hindsight, the assignments we
For m,m%-Hg(B5H8)2, which has been previously de-
4
scribed [18], we note from our results a large J(199Hg–
1H) coupling of 152 Hz to the bridging proton H(4,5)
comparable in magnitude to the two-bond coupling to
the terminal protons H(2) and H(3) [2J(199Hg–1H)] 160
Hz and previously unseen in the lower dispersion 32
4
MHz spectrum. The coupling is large although J and
5J(199Hg–1H) couplings of 39 Hz have been observed to
the pendant methyl groups in m,m%-[(CH3)2C2B4H5H8]2-
Hg [18]. Also of note is our observation that 199Hg
coupling to the distal Hm atoms in m,m%-Hg(B5H8)2 is
larger than that to the proximal Hm atoms {J(199Hg–
1Hm45)=152 Hz versus J(199Hg–1Hm34,25)=100 Hz}.
This is presumably due to enhanced long-range cou-
pling due to the W effect [29]. Direct coupling between

H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
229
Organometallic Chemistry II, Pergamon Press, Oxford, 1995,
Vol. 1, pp. 275–373.
[6] (a) D.F. Gaines, K.M. Coleson, J.C. Calabrese, J. Am. Chem.
Soc. 101 (1979) 3979. (b) D.F. Gaines, K.M. Coleson, J.C.
Calabrese, Inorg. Chem. 20 (1981) 2185.
[7] (a) D.F. Gaines, T.V. Iorns, J. Am. Chem. Soc. 92 (1970)
4571. (b) M.A. Nelson, M. Kameda, S.A. Snow, G. Kodama,
Inorg. Chem. 21 (1982) 2898.
[8] (a) D.F. Gaines, T.V. Iorns, J. Am. Chem. Soc. 89 (1967)
4249. (b) D.F. Gaines, T.V. Iorns, J. Am. Chem. Soc. 90
(1968) 6617. (c) D.F. Gaines, J. Ulman, Inorg. Chem. 13
(1974) 2792. (d) T.C.Geisler, A.D. Norman, Inorg. Chem. 9
(1970) 2167. (e) A.B. Burg, Inorg. Chem. 13 (1974) 1010. (f)
J.B. Leach, G. Oates, S. Tang, T. Onak, J. Chem. Soc. Dalton
Trans (1975) 1018. (g) L. Barton, H. Fang, D.K. Srivastava,
T.A. Schweitzer, N.P. Rath, Appl. Organomet. Chem. 10
(1996) 183.
[9] (a) A.B. Burg, H. Heinen, Inorg. Chem. 7 (1968) 1021. (b)
D.F. Gaines, D.E. Coons, Inorg. Chem. 25 (1986) 364. (c)
D.E. Coons, D.F. Gaines, Inorg. Chem. 27 (1987) 1985. (d)
R.W. Miller, K.J. Donaghy, J.T. Spencer, Organometallics 10
(1991) 1161. (e) B.H. Goodreau, R.L. Ostrander, J.T. Spencer,
Inorg. Chem. 30 (1991) 2066. (f) R.W. Miller, K.T. Donaghy,
J.T. Spencer, Phosphorus Sulfur Silicon 57 (1991) 287.
[10] (a) V.T. Brice, S.G. Shore, J. Chem. Soc. Chem. Commun.
(1970) 1312, (b) V.T. Brice, S.G. Shore, J. Chem. Soc. Dalton
Trans. (1975) 334.
describe herein are obvious. The downfield resonance is
usually broader and this broadening was ascribed to the
influence of the metal atom. Now we can see that it
really is due to coupling to the bridging hydrogen
atoms rather than to the substituent. The assignment of
the 11B spectra from comparisons with that of the
parent B5H9 is not reasonable, because these 2,3-m-
M(B5H8) species are formed from the reaction of
[B5H8]− salts with complex metal halides. The chemical
shifts of the 2,3-m-M(B5H8) systems are more appropri-
ately compared to those of the anion [B5H8]−, for
which the basal boron resonance is observed at 17 ppm
[33]. Actually, when the anion [B5H8]− bonds to bridg-
ing metal or metalloid substituents, the electron distri-
bution is changed. One may expect that the boron atom
bonded to the metal have higher charge density than
the basal boron atoms further away from metal. Thus
one would expect the B atoms closest to the metal to be
more shielded. This pertains in organolithium com-
pounds when for example in n-butyllithium [34] and
n-propyllithium [35], the resonance for the carbon atom
is which closest to the metal is at the highest field,
because it is strongly shielded.
[11] N.N. Greenwood, J.A. Howard, W.S. McDonald, J. Chem.
Soc. Dalton Trans. (1975) 334.
[12] N.N. Greenwood, J. Staves, J. Chem. Soc. Dalton Trans.
(1977) 1788.
Acknowledgements
[13] J.D. Kennedy, J. Staves, Z. Naturforsch. Teil B 34B (1979)
808.
[14] N.N. Greenwood, J. Staves, J. Chem. Soc. Dalton Trans.
(1978) 1146.
[15] N.N. Greenwood, J. Staves, J. Chem. Soc. Dalton Trans.
(1978) 1144.
[16] P.K. Rush, L. Barton, Polyhedron 4 (1985) 1741.
[17] N.W. Alcock, L. Parkhill, M.G.H. Wallbridge, Acta Crystal-
logr. Sect. C 41 (1985) 716.
We acknowledge the NSF (grant nos. CHE-9311557,
CHE-9727570), the Missouri Research Board and UM-
St. Louis for research grants to L.B., NSF (grant no
CHE-9318696) and the UM-St. Louis Center for
Molecular Electronics for funds which purchased the
NMR facilities, Dr. J.J. O’Brien for assistance with the
atomic absorption measurements and Mr. J. Kramer
for the mass spectral measurements.
[18] N.N. Greenwood, J. Staves, J. Chem. Soc. Dalton Trans.
(1977) 1786.
[19] J.C. Calabrese, L.F. Dahl, J. Am. Chem. Soc. 93 (1971) 6042.
[20] D.K. Srivastava, N.P. Rath, L. Barton, Organometallics. 11
(1992) 2263.
References
[21] N.S. Hosmane, R.N. Grimes, Inorg. Chem. 18 (1979) 2886.
[22] H. Fang, D. Zhao, N.P. Rath, L. Brammer, L. Barton,
Organometallics. 14 (1995) 1700.
[23] Presented at BUSA-V-MEX, Guanajuato, Mexico (1996) Paper
No. 1, 1996.
[24] D.F. Shriver, M.A. Drezdon, The Manipulation of Air-Sensi-
tive Compounds, John Wiley, New York, 1986.
[25] J. Bould, J.D. Kennedy, M.Thornton-Pett, H. Fang, N.P.
Rath and L. Barton (in preparation).
[26] (a) T. Onak, H. Landesman, R.E. Williams, I. Shapiro, J.
Phys. Chem. 63 (1959) 1533. (b) J.D. Odom, P.D. Ellis, H.C.
Walsh, J. Am. Chem. Soc. 93 (1971) 3529. (c) H.D. Johnson
II, S.G. Shore, in: E.L. Muetterties (Ed.), Boron Hydrides,
Academic Press, New York, 1975, p. 101.
[1] (a) T.P. Onak, G.B. Dunks, I.W. Searcy, J. Spielman, Inorg.
Chem. 6 (1967) 1465. (b) D.F. Gaines T.V. Iorns, J. Am.
Chem. Soc. 89 (1967) 3375. (c) R.A. Geanangel, S.G. Shore, J.
Am. Chem. Soc.89 (1967) 6771.
[2] (a) D.F. Gaines, Acc. Chem. Res. 6 (1973) 416. (b) S.G.
Shore, Pure Appl. Chem. 69 (1977) 717. (c) N.N. Greenwood,
Pure Appl. Chem. 69 (1977) 791.
[3] H.D. Johnson, S.G. Shore, J. Am. Chem. Soc. 93 (1971) 3798.
[4] (a) D.F. Gaines, in: R.W. Parry, G.J. Kodama (Eds.) Boron
Chemistry 4, Plenary Lectures at the 4th International Meeting
on Boron Chemistry, Salt Lake City, Pergamon Press, Oxford,
1979, pp. 73–82. (b) L.J. Todd, Chapter 4, in: R.N. Grimes
(Ed.), Metal Interactions with Boron Clusters, Plenum Press,
New York, 1982.
[27] H. Fang, D. Zhao, L. Brammer and L. Barton, J. Chem. Soc.
Chem. Commun. (1994) 1531.
[28] H. Fang, PhD Dissertation, University of Missouri-St. Louis,
1985.
[29] P. Hore, Nuclear Magnetic Resonance, Oxford Science Publi-
cations, Oxford University Press, Oxford, 1995.
[5] (a) K.B. Gilbert, S.K. Boocock, S.G. Shore, Part 6, Chapter
41, in: G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Com-
prehensive Organometallic Chemistry, Oxford, Pergamon Press,
1982, pp. 879–945. (b) J.D. Kennedy, Prog. Inorg. Chem. 32
(1984) 651. (c) L. Barton, D.K. Srivastava, Chapter 8, in: G.
Wilkinson, F.G.A. Stone, E.W. Abel (Eds.) Comprehensive
.

230
H. Fang et al. / Journal of Organometallic Chemistry 614–615 (2000) 223–230
[30] D.F. Gaines, G.M. Edvenson, B.R. Adams, Inorg. Chem. 26
(1987) 1813.
[33] H.D. Johnson, II, R.A. Geanangel, S.G. Shore, Inorg. Chem. 9
(1970) 908.
[31] H. No¨th, B. Wrackmeyer, Chapter 8, in: Nuclear Magnetic
Resonance Spectroscopy of Boron Compounds, Springer,
Heidelberg, 1978.
[32] B.H. Goodreau, R.L. Ostrander, J.T. Spencer, Inorg. Chem. 30
(1991) 2066.
[34] G. Fraenkel, A.M. Fraenkel, M.J. Geckle, F. Schloss, J. Am.
Chem. Soc. 101 (1979) 4745.
[35] S. Bywater, P. Lachance, D.J. Worsfold, J. Phys. Chem. 79
(1975) 2148.
.