σ-Borane complexes of manganese and rhenium

Article abstract of DOI:10.1021/ja001546o

The late transition metal borane complexes (MeCp)Mn(CO)₂(HBR₂) (R = alkyl, alkoxy) and Cp*Re(CO)₂(HBpin) (HBpin = pinacolborane, Cp* = pentamethylcyclopentadienyl) have been synthesized and isolated. All complexes were characterized spectroscopically, and X-ray crystal structure analyses were performed for (MeCp)Mn(CO)₂(HBpin), (MeCp)Mn(CO)₂(HBcat) (HBcat = catecholborane) and (MeCp)Mn(CO)₂(HBCy₂) (Cy = cyclohexyl). These data show that the complexes contain a borane ligand with a weakened but intact B-H bond. The borane ligand in (MeCP)Mn(CO)₂(HBcat) is replaced by PhCCPh, Ph₃SnH, Ph₂MeSiH, CO, and excess HBpin in ligand substitution reactions. The mechanism of substitution by PhCCPh was investigated. Kinetic studies showed that the reaction is first-order in complex and that borane and PhCCPh react competitively with the 16-electron intermediate, (MeCp)Mn(CO)₂. The ΔH((+)) value for borane dissociation was 24 ± 3 kcal/mol for (MeCp)Mn(CO)₂(HBcat) and 21 ± 1 kcal/mol for (MeCp)Mn(CO)₂(HBCy₂), which provided an upper limit on the metal-borane binding energies. Dynamic NMR spectroscopy of (MeCp)Mn(CO)₂(HBcat) revealed two types of rotations of the borane ligand.

Full text of DOI:10.1021/ja001546o

J. Am. Chem. Soc. 2000, 122, 9435-9443
9435
σ-Borane Complexes of Manganese and Rhenium
Sabine Schlecht and John F. Hartwig*
Contribution from the Department of Chemistry, Yale UniVersity, P.O. Box 208107,
New HaVen, Connecticut 06520-8107
ReceiVed May 4, 2000
Abstract: The late transition metal borane complexes (MeCp)Mn(CO)2(HBR2) (R ) alkyl, alkoxy) and Cp*Re-
(CO)2(HBpin) (HBpin ) pinacolborane, Cp* ) pentamethylcyclopentadienyl) have been synthesized and
isolated. All complexes were characterized spectroscopically, and X-ray crystal structure analyses were performed
for (MeCp)Mn(CO)2(HBpin), (MeCp)Mn(CO)2(HBcat) (HBcat ) catecholborane) and (MeCp)Mn(CO)2(HBCy2)
(Cy ) cyclohexyl). These data show that the complexes contain a borane ligand with a weakened but intact
B-H bond. The borane ligand in (MeCp)Mn(CO)2(HBcat) is replaced by PhCCPh, Ph3SnH, Ph2MeSiH, CO,
and excess HBpin in ligand substitution reactions. The mechanism of substitution by PhCCPh was investigated.
Kinetic studies showed that the reaction is first-order in complex and that borane and PhCCPh react competitively
with the 16-electron intermediate, (MeCp)Mn(CO)2. The ∆Hq value for borane dissociation was 24 ( 3 kcal/
mol for (MeCp)Mn(CO)2(HBcat) and 21 ( 1 kcal/mol for (MeCp)Mn(CO)2(HBCy2), which provided an upper
limit on the metal-borane binding energies. Dynamic NMR spectroscopy of (MeCp)Mn(CO)2(HBcat) revealed
two types of rotations of the borane ligand.
Introduction
hydrocarbon that is held in place, perhaps principally, by
hydrophobic interactions with a double-A-frame porphyrin
ligand.18
σ-Complexes1-4 are reactive intermediates that typically
precede oxidative addition of substrates with an X-H bond.1
σ-Complexes are, therefore, intermediates in catalytic hydro-
genation,5 hydrosilylation,6,7 and hydroboration8 reactions. In
contrast to the large number of dihydrogen complexes9 and
σ-silane compounds,6,7 monoborane σ-complexes are limited in
number and have been isolated only recently.10-13 They are the
only simple, isolable H-X σ-complex containing a single first-
row element X. Alkane complexes have been detected spec-
troscopically in solution,14 in the gas phase,15 and by matrix
isolation,16,17 but the only isolated alkane complex contains a
The isolation and characterization of σ-complexes contributes
to our understanding of the bonding interactions between
substrates with X-H bonds and unsaturated transition-metal
compounds. Silane, dihydrogen, and now borane complexes are
stabilized by exceptionally strong interactions between the metal
center and the H-X bond. Thus, a precise electronic description
of stable σ-compounds can help us to understand what controls
the strength of these complexes when they act as reaction
intermediates and cannot be isolated.
Puddephatl19a and, more recently, Shimoi19b reported phos-
phine-ligated boryl complexes (R3PH2B-M). The hydrogen
atoms of these complexes, however, point away from the metal
center. These complexes are, therefore, more akin to Lewis
acid-base adducts of boryl complexes than to the σ-borane
complexes reported here. Shimoi has also reported complexes
of phosphinoboranes R3PBH3.13 In this case no metal-boron
interaction was detected, again distinguishing them from the
σ-complexes of three-coordinate boranes which are reported
here.
* To whom correspondence should be addressed.
(1) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.
(2) Schubert, U. AdV. Organomet. Chem. 1990, 30, 151.
(3) Schubert, U.; Scholz, G.; Mu¨ller, J.; Ackermann, K.; Wo¨rle, B. J.
Organomet. Chem. 1986, 306, 303.
(4) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem. Soc.
1986, 108, 1537.
(5) Halpern, J. Science 1982, 217, 401.
(6) Chalk, A. J.; Harrod, J. F. J. Am. Chem. Soc. 1965, 87, 16.
(7) Sommer, L. H.; Lyons, J. E.; Fujimoto, H. J. Am. Chem. Soc. 1969,
91, 2670.
(8) Beletskaya, I.; Pelter, A. Tetrahedron 1997, 53, 4957.
(9) Kubas, G. J. Acc. Chem. Res. 1988, 21, 120.
(10) Hartwig, J. F.; Muhoro, C. N.; He, X. J. Am. Chem. Soc. 1996,
118, 10936.
(11) Muhoro, C. N.; Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1997,
36, 1510.
(12) Muhoro, C. N.; He, X.; Hartwig, J. F. J. Am. Chem. Soc. 1999,
121, 5033.
(13) Shimoi, M.; Nagai, S.; Ichikawa, M.; Kawano, Y.; Katoh, K.;
Uruichi, M.; Ogino, H. J. Am. Chem. Soc. 1999, 121, 11704.
(14) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120, 9953.
(15) Brown, C. E.; Ishikawa, Y.-I.; Hackett, P. A.; Rayner, D. M. J.
Am. Chem. Soc. 1990, 112, 2530.
We felt that borane complexes of manganese, such as
(MeCp)Mn(CO)2(HBR2), were important synthetic targets be-
cause they would allow one to make comparisons to early
transition-metal borane complexes which have been prepared
recently and to silane, (MeCp)Mn(CO)2(HSiR3), and stannane,
(MeCp)Mn(CO)2(HSnR3), complexes, which are the original
σ-complexes.3,20 To allow for such comparisons, our investiga-
tions addressed the influence of the borane substituents on the
three-center bonding interaction, the strength of the metal-
borane interactions, and the mechanism of the displacement of
borane by more strongly coordinating dative ligands.
(16) Rest, A. J.; Whitwell, I.; Graham, W. A. G.; Hoyano, J. K.;
McMaster, A. D. J. Chem. Soc. 1987, 1181.
(17) Turner, J. J.; Burdett, J. K.; Perutz, R. N.; Poliakoff, M. Pure Appl.
Chem. 1977, 49, 271.
(18) Evans, D. R.; Drovetskaya, T.; Reed, C. A.; Boyd, P. D. W. J. Am.
Chem. Soc. 1997, 119, 3633.
(19) (a) Elliot, D. J.; Levy, C. J.; Puddephatl, R. J.; Holah, D. G.; Hughes,
A. N.; Magnesun, V. R.; Moser, I. M. Inorg. Chem. 1990, 29, 5015. (b)
Kawano, Y.; Yasue, T.; Shimoi, M. J. Am. Chem. Soc. 1999, 121, 11744.
(20) Graham, W. A. G.; Jetz, W. Inorg. Chem. 1971, 10, 4.
10.1021/ja001546o CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/13/2000

9436 J. Am. Chem. Soc., Vol. 122, No. 39, 2000
Schlecht and Hartwig
Scheme 1
Scheme 2
Figure 1. 1H NMR resonance of the hydride in 1 at 25 °C (a) with
and (b) without 11B decoupling.
free boranes. The deuterated compound (MeCp)Mn(CO)2-
(DBcat) (1-d) showed a 2H NMR chemical shift that was
indistinguishable from that of the hydride resonance in protiated
1; no isotopic perturbation on the chemical shift was observed.22
The IR spectrum of 1 contained a broad band at 1606 cm-1
,
most likely corresponding to the H-Mn stretching frequency,
with the B-H stretch at low frequency, as seen for silane
complexes.23 Strong bands at 1995 cm-1 (νsCO) and 1937 cm-1
(νasCO) were observed for the carbonyl stretches. Similar bands
at 1603 cm-1 (νHB), 1983 cm-1 (νsCO), and 1921 cm-1 (νas-
CO) were observed for pinacolborane complex 2.
Complexes 1 and 2 were also characterized by X-ray
crystallography. The two complexes showed similar connectivity
and structural properties. Acquisition parameters are given in
Table 1, selected bond distances and angles are given in Tables
2 and 3, and ORTEP diagrams of 1 and 2 are provided in Figures
2 and 3. The metal-bound hydrogen atom was found in the
difference Fourier map of both complexes and was refined. Both
complexes contain extremely small H-Mn-B angles of 37.2-
(8)° and 38.2(6)°. The Mn-B distances of 2.083(2) Å in 1 and
2.149(2) Å in 2 were longer than those in the related boryl
complexes CpFe(CO)2(Bcat) [d(Fe-B) ) 1.96 Å] or CpFe-
(CO)2(BPh2) [d(Fe-B) ) 2.03 Å].24 The B-H bond lengths
of 1.29(2) Å in the HBcat complex 1 and 1.31(2) Å in the HBpin
complex 2 are between the value of 1.26 Å found in Cp2Ti-
(HBcat)2 and that of 1.35 Å in Cp2Ti(HBcat-3-F)(PMe3), the
only other crystallographically characterized σ-borane com-
plexes of three-coordinate boron.12 These values for 1 and 2
are significantly longer than the 1.17 Å distances calculated for
catecholborane and the closely related cyclic alkoxyboranes.25
Synthesis and Structure of (MeCp)Mn(CO)2(HBR2) (R )
Me, Cy) (3, 4). Reaction of photochemically generated (MeCp)-
Mn(CO)2 with dialkylboranes did not occur, presumably due
to the dimeric structures of these boranes. However, the desired
dialkylborane complexes were formed by the reaction of the
manganese anion with halodialkylboranes as shown in Scheme
2. (MeCp)Mn(CO)2(HBMe2) (3) was obtained in 82% yield by
reacting K[(MeCp)Mn(CO)2H] with BrBMe2 in pentane at room
temperature. The dicyclohexylborane complex 4 was prepared
in a similar manner from the hydridomanganate and ClBCy2 in
85% yield. Complex 3 was a red oil that was judged pure by
1H, 13C, and 11B NMR spectroscopy. Cyclohexylborane complex
4 was as a red solid. Compound 3 decomposed quickly at room
temperature. Compound 4 was stable at room temperature for
Results
Synthesis and Structures of (MeCp)Mn(CO)2(HBcat) (1)
and (MeCp)Mn(CO)2(HBpin) (2). Borane complexes of the
(MeCp)Mn(CO)2 fragment were generated by photolysis of
(MeCp)Mn(CO)3 and by nucleophilic substitution using
[(MeCp)Mn(CO)2H]-, as shown in Schemes 1 and 2. Photolysis
of (MeCp)Mn(CO)3 in THF for 30 min in the presence of a
large excess (14-16 equiv) of the corresponding borane
produced (MeCp)Mn(CO)2(HBcat) (1) in 53% yield and (MeCp)-
Mn(CO)2(HBpin) (2) in 18% yield as yellow solids after
recrystallization from pentane. The use of THF as a solvent
was essential, presumably because of its ability to stabilize the
photolytically generated (MeCp)Mn(CO)2. The reaction of either
ClBcat or ClBpin generated in situ with the known anionic
manganese hydride K[(MeCp)Mn(CO)2H]21 in pentane at room
temperature (Scheme 2) provided an alternative route that
formed 1 and 2 in higher yields. The reaction was complete
after 10 min and gave 92% yield of (MeCp)Mn(CO)2(HBcat)
and 65% yield of (MeCp)Mn(CO)2(HBpin).
(MeCp)Mn(CO)2(HBcat) was stable at room temperature as
a solid, but decomposed slowly in solution. The pinacolborane
derivative decomposed at room temperature over hours, even
as a solid, but was stored at -30 °C for an extended period of
time without decomposition.
The 11B NMR shifts of both 1 and 2, 46 ppm for 1 and 45
ppm for 2, were far downfield from those of the the free boranes.
High-field signals at -14.46 ppm for 1 and -15.66 ppm for 2
were observed in the 1H NMR spectrum. These hydride
resonances were broad, due to the proximity of the boron atom,
and sharpened upon 11B decoupling (Figure 1). The scalar
(22) Hartwig, J. F.; DeGala, S. R. J. Am. Chem. Soc. 1994, 116, 3661.
(23) Delpech, F.; Sabo-Etienne, S.; Daran, J. C.; Chandret, B.; Hussein,
K.; Marsden, C. J.; Barthelat, T. C. J. Am. Chem. Soc. 1999, 121, 6668.
(24) Hartwig, J. F.; Huber, S. J. Am. Chem. Soc. 1993, 115, 4908.
(25) The distances were calculated from z-matrices in Roblen, P. R.;
Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 4648.
coupling constants, J(1H,11B), were approximately 95 Hz in
both complexes, which is roughly half the value found in the
1
(21) Braunschweig, H.; Ganter, B. J. Organomet. Chem. 1997, 545, 163.

σ-Borane Complexes
J. Am. Chem. Soc., Vol. 122, No. 39, 2000 9437
Table 1. Acquisition Parameters for 1, 2, and 4
empirical formula
formula weight
color of crystal
crystal system
space group
C14H12BO4Mn (1)
309.99
yellow
triclinic
C14H20BO4Mn (2)
318.06
yellow
monoclinic
P21/a
a ) 12.6474(3) Å
b ) 10.0519(3) Å
c ) 12.8851(4) Å
C20H30BO2Mn (4)
368.20
red
triclinic
P1h
a ) 6.7421(3) Å
b ) 9.3350(3) Å
c ) 16.1767(5) Å
R ) 95.464(3)°
â ) 90.115(2)°
γ ) 107.928(2)°
2
P1h
unit cell dimension
a ) 7.7712(4) Å
b ) 9.5508(4) Å
c ) 9.7226(4) Å
R ) 101.005(3)°
â ) 108.881(2)°
γ ) 97.727(2)°
2
â ) 115.635(2)°
Z value
4
goodness-of-fit
final R indices
1.64
R ) 0.030
Rw ) 0.044
2.92
R ) 0.031
Rw ) 0.049
2.27
R ) 0.037
Rw ) 0.050
Table 2. Selected Bond Distances in 1, 2, and 4
distance (Å)
bond
1
2
4
Mn-B
2.083(2)
1.57(2)
1.29(2)
1.413(2)
1.404(2)
2.149(2)
1.53(2)
1.31(2)
1.376(2)
1.376(2)
2.187(3)
1.49(2)
1.24(2)
1.590(3)
1.590(3)
Mn-H(1)
B-H(1)
B-O(3) or B-C(9)
B-O(4) or B-C(15)
Table 3. Selected Bond Angles in 1, 2, and 4
angle (deg)
atom
H(1)
Mn
atom
atom
H(1)
1
2
4
Mn
B
B
38.2(6)
48.6(7)
37.2(8)
44.7(9)
33.2(7)
41.1(9)
C(1)
O(3) or C(9)
Mn C(2)
90.66(7) 90.32(7) 93.4(1)
B O(4) or C(15) 109.8(1) 112.7(1) 117.9(2)
Figure 3. ORTEP diagram of 2.
bands at 1975 and 1910 cm-1. Similar values for 4 at 1597
cm-1, 1967 cm-1 (νsCO), and 1901 cm-1 (νasCO) were
observed.
Complex 4 was also characterized by X-ray crystallography.
Acquisition parameters are given in Table 1, selected bond
distances and angles are given in Tables 2 and 3, and an ORTEP
diagram of 4 is provided in Figure 4. As for 1 and 2, the metal-
bound hydrogen atom was located in the difference Fourier map
and was refined. The Mn-B distance of 2.19 Å is longer than
that in the alkoxyborane complexes 1 and 2, and the Mn-H
bond is shortened to 1.49 Å. The B-H distance of 1.24 Å in 4
appeared shorter than that in the two alkoxyborane complexes,
but the difference in bond lengths is close to two standard
deviations (Table 2). If significant, the shorter bond distance
would be caused by differences in metal-ligand interaction,
because a free monomeric dialkylborane is computed to have a
slightly longer, 1.20 Å, B-H distance than the 1.17 Å distance
in a free dialkoxyborane.25 The overall structural features clearly
indicate a weaker metal-boron interaction and a more pro-
nounced hydride character for the borane hydrogen.
Figure 2. ORTEP diagram of 1.
several hours, but decomposed within 1 day when in solution
and as a solid.
The 11B NMR resonances for 3 and 4 were broad singlets at
101 and 104 ppm. Like 1 and 2, the dialkylborane complexes
Synthesis of Cp*Re(CO)2(HBpin) (5). A synthetic pathway
different from those for the manganese complexes was used to
prepare Cp*Re(CO)2(HBpin) (5). The reaction of cis-Cp*Re-
(CO)2(Bpin)226 with methanol or neopentyl alcohol in benzene
at room temperature gave 5 in quantitative yield after 2 h. The
compound is a colorless solid that was stable at room temper-
1
showed high-field signals in the H NMR spectrum for the
hydride. A broad singlet at -17.06 ppm was observed for
dimethyl 3, and a broad singlet at -16.96 ppm was detected
for dicyclohexyl 4. In both cases, the hydride signal sharpened
upon 11B-decoupling, indicating the presence of a scalar coupling
between the boron and the hydrogen atoms. The IR spectrum
of 3 in hexanes showed an IR band at 1592 cm-1 and carbonyl
(26) Chen, H.; Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1999, 38,
3391.

9438 J. Am. Chem. Soc., Vol. 122, No. 39, 2000
Schlecht and Hartwig
Figure 5. Formation of diphenylacetylene complex, monitored at λ
) 420 nm.
Table 4. Observed Rate Constants at Different Concentrations of
Catecholborane and Diphenylacetylene; T ) 70 °C
[1],
mmol/L
[HBcat],
mmol/L
[PhCCPh],
mmol/L
kobs
(·104, s-1
)
0.645
0.645
0.645
0.645
0.645
3.188
12.750
25.500
51.000
76.500
9.719
9.719
9.719
9.719
9.719
14.0
12.0
10.1
7.9
Figure 4. ORTEP diagram of 4.
Scheme 3
6.7
0.242
0.242
0.242
0.242
0.242
0.242
4.688
4.688
4.688
4.688
4.688
4.688
1.873
3.745
7.491
14.981
59.925
119.850
11.3
13.1
16.3
17.2
18.9
18.9
HBpin and Ph2MeSiH were conducted photochemically because
the products are not stable at 60 °C. Irradiation of 1 for 30 min
using a mercury-arc lamp in the presence of 10 equiv of HBpin
gave complex 2 in 69% yield. Reaction of 1 with Ph2MeSiH
under similar conditions gave (MeCp)Mn(CO)2(SiPh2Me)(H)21
in 94% yield. For all ligand-substitution reactions that we
conducted, the conversion of (MeCp)Mn(CO)2(HBcat) was
higher than 95%. The amount of side products never exceeded
10%.
Kinetic Studies of the Reactions of (MeCp)Mn(CO)2-
(HBcat) (1) and (MeCp)Mn(CO)2(HBCy2) (4) with PhCCPh.
Kinetic studies of the reaction of 1 with diphenylacetylene were
conducted to determine the mechanism of this process and to
obtain an upper limit on the Mn-(HBcat) dative bond energy.
The kinetic measurements were made using UV/vis spectros-
copy, monitoring the band at 420 nm, due to the absorption of
the orange diphenylacetylene complex, which was formed in
the reaction. The reactions were run in toluene at 70 °C. As
indicated by the excellent fit to a single exponential for the
formation of the product (Figure 5), the substitution of cat-
echolborane by diphenylacetylene was first-order in the borane
complex. As shown by the rate constants in Table 4, the
observed rate constant depended on the concentration of both
the added borane and the incoming ligand. The reaction rate
increased with increasing concentrations of diphenylacetylene
when the ratio of alkyne to HBcat was low. However, saturation
of the reaction rate was observed at high concentrations of
[PhCCPh] (Figure 6). Added catecholborane inhibited the ligand
exchange at low alkyne/borane ratios. A plot of 1/kobs versus
the concentration of catecholborane at constant alkyne concen-
tration is provided in Figure 7.
ature. A broad singlet was observed for the hydrogen of the
1
coordinated borane in the H NMR spectrum at -11.06 ppm.
Compound 5 resonated at 46 ppm in the 11B NMR spectrum.
The IR bands were similar to those of the manganese compounds
and were found at 1603 cm-1, 1981 cm-1 (νsCO), and 1924
cm-1 (νasCO). Crystals suitable for X-ray diffraction were not
obtained, despite repeated attempts.
Ligand-Substitution Reactions of (MeCp)Mn(CO)2(HBcat)
(1). The catecholborane in 1 was replaced by a variety of ligands
under both photochemical and thermal conditions. Scheme 3
summarizes the ligand-substitution reactions. All of these
reactions were conducted in the presence of at least 5 equiv of
incoming ligand to minimize dimerization of the potential
intermediate, (MeCp)Mn(CO)2, which forms the insoluble
[(MeCp)Mn(CO)2]2. Except for the conversion of 1 to 2, which
was conducted in THF, all reactions were carried out in benzene
solvent.
Thermal dissociation of catecholborane required a temperature
of 60 °C. Diphenylacetylene displaced the catecholborane ligand
and formed the previously described27 [(MeCp)Mn(CO)2(η2-
PhCCPh)] in 94% yield. Reaction with 2 atm of CO at 60 °C
for 2 h gave (MeCp)Mn(CO)3 in 92% yield. Reactions with
The temperature dependence of the reaction rate for displace-
ment of HBcat and HBCy2 was evaluated, and activation
parameters were determined to obtain an upper limit on the
binding energies of HBcat and HBCy2 to the metal fragment.
(27) Strohmeier, W.; Laporte, H.; v. Hobe, D. Chem. Ber. 1962, 95, 455.

σ-Borane Complexes
J. Am. Chem. Soc., Vol. 122, No. 39, 2000 9439
Figure 6. kobs versus the concentration of diphenylacetylene.
Figure 9. Eyring plot for the reaction of (MeCp)Mn(CO)2(HBCy2)
with PhCCPh.
Discussion
Spectroscopic and Structural Studies. X-ray diffraction, in
combination with NMR and IR spectroscopy, was used to
understand how free neutral boranes bind to the manganese and
rhenium fragments. In short, the crystallographic and spectro-
scopic data show direct bonding between the metal (Mn or Re)
and boron, between the metal and hydrogen, and between boron
and hydrogen. This section of the discussion is organized
according to the different bonding interactions and the spec-
troscopy that support their existence.
Figure 7. 1/kobs versus the concentration of catecholborane.
1. Evidence for B-H Bonding. Several lines of data support
B-H bonding. First, the crystal structures of 1, 2, and 4
supported the presence of this interaction. All three complexes
showed a lateral configuration in a four-legged piano-stool
geometry with OC-Mn-CO angles which were near 90°. In
diagonal dicarbonyl complexes, angles of 100-110° are typi-
cally found.28,29 The H-Mn-B angles were small (33-38°)
and are inconsistent with a structure that would be described
by a Mn(III) center containing a boryl group and a hydrogen in
two separate coordination sites. Consistent with this small angle,
B-H bond lengths were between 1.24(2) and 1.31(2) Å. These
values are clearly within B-H bonding distance but are
elongated relative to the 1.17 and 1.20 Å distances calculated
for related free boranes.25
Figure 8. Eyring plot for the reaction of (MeCp)Mn(CO)2(HBcat) with
PhCCPh.
A direct bonding interaction between boron and hydrogen
was also shown clearly by 1H{11B} and by 11B NMR spectros-
copy. Scalar coupling between the boron and the hydride was
observed directly by 11B NMR spectroscopy for compounds 1
and 2, and the coupling constants of 95 Hz were too large for
a scalar coupling solely through the metal. Known complexes
with two discrete cis-boryl and hydride ligands do not show
any evidence for scalar coupling; sharp hydride signals are
observed for those compounds.30 In addition, the 1H NMR
signals for the hydrides of 1 and 2 were broad and sharpened
upon boron decoupling. The half-width and the shape of the
borane hydride resonances showed a strong temperature de-
pendence for all complexes. This phenomenon is well-known
For the reaction of the catecholborane complex, rate constants
were measured from 55 to 85 °C for the following concentra-
tions of reactants: [1] ) 0.45 mM, [HBcat] ) 8.35 mM, and
[PhCCPh] ) 29.50 mM. These concentrations were chosen such
that kobs ) k1, the rate for borane dissociation (see discussion).
The Eyring plot (Figure 8), provided the activation parameters
∆Hq ) 24.0 ( 2.6 kcal/mol and ∆Sq ) -1.1 ( 0.9 eu. For the
reaction of the dicyclohexylborane complex 4 with dipheny-
lacetylene, rate constants were measured from 25 to 55 °C, with
[4] ) 0.68 mM, [PhCCPh] ) 50.95 mM, and no added
dicyclohexylborane, because the free borane is a dimer. As in
the case of the catecholborane complex, these conditions ensure
that kobs ) k1. The reaction was first-order in borane complex.
The Eyring plot (Figure 9) provided the activation parameters
∆Hq ) 21.0 ( 1.1 kcal/mol and ∆Sq ) -5.9 ( 3.3 eu. The
displaced dicyclohexylborane reacted immediately with the
excess diphenylacetylene to form the hydroboration product
dicyclohexyl-cis-1,2-diphenyl(vinyl)borane. The identity of this
side product was confirmed by comparing the 1H and 11B NMR
spectra to those of material formed from the addition of
diphenylacetylene and dimeric dicyclohexylborane in a separate
reaction.
for BH4 complexes31 and other compounds that contain a
-
proton bound to a quadrupolar nucleus.32 This “virtual self-
decoupling”31 gave additional, indirect evidence for scalar
coupling between the boron and the hydrogen atoms.
(28) Dong, D. F.; Hoyano, J. K.; Graham, W. A. G. Can. J. Chem. 1981,
59, 1455.
(29) Smith, R. A.; Bennett, M. J. Acta Crystallogr. Sect. B 1977, 33,
1113.
(30) Irvine, G. J.; Lesley, G.; Marder, T. B.; Norman, N. C.; Rice, C.
R.; Robins, E. G.; Roper, W. R.; Whitell, G. R.; Wright, L. J. Chem. ReV.
1998, 98, 2685.
(31) Marks, T. J.; Shimp, L. A. J. Am. Chem. Soc. 1972, 94, 1542.
(32) Pople, J. A. Mol. Phys. 1958, 1, 168.

9440 J. Am. Chem. Soc., Vol. 122, No. 39, 2000
Schlecht and Hartwig
Figure 11. Intramolecular motions in complex 1.
distance is lengthened and the M-H distance is shortened in 2
and 4 relative to the distances in 1. Catecholborane complex 1
showed the most typical η2, side-on geometry. The borane ligand
in pinacolborane complex 2 is pivoted relative to that in 1 to
create a Mn-B bond that is elongated by 0.07 Å and an Mn-H
distance that is shortened by 0.04 Å relative to that in 1. The
borane in cyclohexyl complex 4 is further pivoted to create an
Mn-B distance that is 0.04 Å longer and an Mn-H distance
that is 0.04 Å shorter than those in 2.
The IR-stretching frequencies for the CO groups were
sensitive to the identity of the borane. The νCO values were
higher for the alkoxyborane complexes 1 and 2 than for the
dialkylborane complexes 3 and 4. The values of 1995 cm-1
(νsCO) and 1937 cm-1 (νasCO) for the catecholborane complex
1 indicated that HBcat is a weaker donor and a stronger acceptor
in complex 1 than is HBpin in complex 2, for which stretching
frequencies of 1983 cm-1 (νsCO) and 1921 cm-1 (νasCO) were
found. The carbonyl stretching frequencies of 3 and 4 suggested
that dimethylborane and dicyclohexylborane were even stronger
σ-donating and weaker π-accepting ligands than pinacolborane.
A more electron-rich B-H bond that acts as a stronger donor
to the metal may override increases in Lewis acidity that would
be expected for the dialkylborane ligands.
Both the core structure and the IR-stretching frequencies
follow the same ordering of the sigma-donating character of
the borane. The electron density on the borane hydrogen should
increase with increasing sigma donation from the substituents.
This increase in hydridic character should shorten the M-H
distance, which apparently leads to the pivoting of the borane
about the B-H bond. Steric effects may also contribute to the
lengthening of the M-B distance. However, space-filling
models of the structures of the complexes suggest that steric
effects between the borane substituents and the small MeCpMn-
(CO)2 fragment are not strong.
Figure 10. Core geometries in borane complexes 1, 2, and 4.
IR spectroscopy provided information about the B-H bond
order. IR bands between 1606 cm-1 (1) and 1592 cm-1 (3)
corresponded to a stretching mode involving the hydride, which
is likely to be predominantly M-H in character. The B-H
stretches are even lower in frequency and much lower in energy
than the B-H-stretching frequency in free monomeric boranes
(2660 cm-1 for catecholborane). These data indicated a reduced
B-H bond order. Rhenium HBpin complex 5 showed the same
IR pattern as the manganese analogue 2, indicating a similar
reduction in B-H bond order. Thus, borane complex 5 contains
a partial H-X bond, in contrast to the silicon analogue CpRe-
(CO)2(H)(SiPh3),28,29 which contains H and SiPh3 as two
individual ligands.
2. Evidence for M-B Bonding. The X-ray structures of 1,
2, and 4 showed evidence for a partial M-B bond order. The
bond lengths of 2.083(2) Å in 1 and 2.149(2) Å in 2 are certainly
within M-B bonding distance but are longer than those in the
related boryl complexes CpFe(CO)2(Bcat) [d(Fe-B) ) 1.96 Å]
or CpFe(CO)2(BPh2) [d(Fe-B) ) 2.03 Å].24 The presence of a
metal-boron bond was also established by the 11B NMR
chemical shifts of the complexes (45-46 ppm for dialkoxy-
boranes and 101-104 ppm for dialkylboranes). The chemical
shifts are located far downfield from the 11B NMR signals of
monomeric, free boranes or Lewis base adducts of dialkyl-
boranes.33 The 11B NMR chemical shifts lie near those of the
resonances for boryl complexes such as [Mn(CO)5(Bcat)] (43
ppm), [CpFe(CO)2(Bcat)] (52 ppm), and [CpFe(CO)2(BPh2)]
(121 ppm).24
3. Evidence for M-H Bonding. Direct M-H bonding was
1
confirmed by the high-field H NMR shifts of the hydrogen
5. Dynamic Processes. Rapid site exchange processes were
atoms of the coordinated boranes. These resonances are at a
higher field than those of four-coordinate hydridoborates or
Lewis acid-base adducts of boranes.33 The assignment of the
extreme high-field 1H NMR signals to the borane hydrogen was
revealed by 13C NMR spectroscopy. The room temperature 13
C
NMR spectrum of 1 showed only one carbonyl resonance (226.4
ppm) and two methine carbons in the catechol backbone. These
data showed that the intramolecular motion A shown in Figure
11, and perhaps motion B, were occurring in 1 at room
temperature. Cooling of 1 in THF-d8 to -115 °C led to
decoalescence of the carbonyl resonance into two signals of
equal intensity (225.1 and 230.5 ppm, Tc ) 173 K). The catechol
backbone, however, showed two methine resonances at this
temperature, as it did at room temperature. Assuming the
catecholate resonances could be resolved in a frozen structure,
motion B appears to occur on the NMR time scale, even at
-115 °C. The two methine carbon atoms in the MeCp ligand,
which are diastereotopic, were not distinct in solution at even
2
confirmed by the use of DBcat in the synthesis of 1-d1. A H
NMR signal was observed at the same chemical shift as its
protiated analogue 1. The hydride ligands were also identified
by X-ray diffraction, and the observation of M-H bond
distances between 1.49(2) and 1.57(2) Å indicated a direct M-H
bond.
4. Comparison of Overall Structural Features for the
Three Complexes. Figure 10 shows changes in core geometry,
determined by X-ray diffraction, for the three complexes. In
the series of compounds 1, 2, and 4, the center of the B-H
bond is roughly equidistant from the metal center, but the M-B
(34) Mann, B. E.; Taylor, B. F. Organometallic Chemistry: 13C NMR
Data for Organometallic Compounds; Academic Press: London, 1981; pp
225-227.
(33) No¨th, H.; Wrackmeyer, B. Nuclear Magnetic Resonance Spectros-
copy of Boron Compounds; Springer: Berlin, 1978; p 461.

σ-Borane Complexes
J. Am. Chem. Soc., Vol. 122, No. 39, 2000 9441
-115 °C, most likely because of their small differences in
chemical shift.34 From the data for the carbonyl signals, an
activation barrier of ∆Gq ) 7.4 ( 1.2 kcal/mol can be estimated
for motion A in Figure 11. This value is in good agreement
with the rotational barriers of other σ-donor/π-acceptor ligands
on CpM(CO)2 fragments such as substituted and unsubstituted
olefins and acetylenes; in these cases the rotational barriers are
8-10 kcal/mol.35-37
Scheme 4
Ligand-Substitution Reactions and Mechanistic Studies.
Several ligand-substitution reactions were performed to probe
potential reversibility of borane coordination and to assess the
relative stabilities of σ-borane, silane, and stannane complexes
of the (MeCp)Mn(CO)2 16-electron fragment. Borane complex
1 reacted with excess HSiPh2Me or HSnPh3 to form the
corresponding silyl hydride21 and stannane38,39 complexes in
good yields. The stannane derivative is thermally much more
stable than the borane complex, which is similar in stability to
the silyl hydride. Although evaluated only qualitatively, the
equilibria between borane and silane or stannane complexes
favored the silyl hydride and stannane complexes and free
borane.
Detailed kinetic studies were undertaken of the reaction of
catecholborane complex 1 with diphenylacetylene. The reaction
showed a first-order dependence on the concentration of
complex 1 (Figure 5), and the observed reaction rates depended
on the concentrations of both the catecholborane and the
diphenylacetylene. Added catecholborane was found to retard
the reaction. Reactions conducted with increasing concentrations
of PhCCPh revealed saturation behavior (Figure 6). The reaction
was first-order in [PhCCPh] at low [PhCCPh], but became zero-
order in this reagent when its concentration was much higher
than the added borane concentration. Similar kinetic behavior
was observed for the reaction of the silane complex CpMn-
(CO)2(HSiPh3) with PPh3.40 This rate behavior is characteristic
of dissociative ligand-substitution reactions. It is consistent with
the mechanism in Scheme 4, which is described by the rate
expression in eq 1. At high concentrations of PhCCPh, this
expression reduces to a first-order expression where kobs ) k1.
The observed rate constant can also be expressed as in eq 2.
intermediate for formation of an alkyne complex and a borane
complex, are provided in Figure 7 and are derived from the
plot of 1/kobs versus [HBcat] at a constant [PhCCPh]. The
selectivity for the reaction of the unsaturated fragment with
PhCCPh versus borane (k2/k-1 ) 6.8) was low, considering the
greater stability of the alkyne complex. Presumably, the high
reactivity of the 16-electron intermediate makes it relatively
unselective. The electrophilic nature of the borane may also
accelerate its reaction with the electron-rich Mn(I) intermedi-
ate.40
In addition to reaction orders, Eyring activation parameters
were obtained for the ligand-displacement reaction. We obtained
the activation parameters with a high concentration of PhCCPh
so that the value of kobs ) k1. The enthalpy of activation for k1,
∆Hq ) 24 kcal/mol, provides an upper limit for the Mn-
(HBcat) bond-dissociation energy. The ∆Hq value is slightly
lower than that for the dissociation of triphenylsilane from
(MeCp)Mn(CO)2(HSiPh3) (25.5 kcal/mol) and significantly
higher than the enthalpy of binding in most η2-H2 complexes
(13-17 kcal/mol).41 A very similar dissociation enthalpy of 25
( 3 kcal/mol was found for the dissociation of catecholborane
from the titanocene fragment.12 Eyring activation parameters
were also obtained for the reaction of dicyclohexylborane
complex 4, and ∆Hq for the k1 step was 21 kcal/mol, indicating
a metal-borane binding energy that is roughly 4 kcal/mol
weaker than that in the catecholborane complex. As mentioned
above, space-filling models do not indicate strong differences
in steric properties of 1, 2, and 4. This information suggests
that electronic differences account for the relative stabilities of
1 and 4. Although complex 3 may decompose by pathways other
than simple borane dissociation, its pronounced thermal instabil-
ity is consistent with this assertion.
d[(MeCp)Mn(CO)2(HBcat)]
)
dt
k1k2[PhCCPh]
[(MeCp)Mn(CO)2(HBcat)] )
k2[PhCCPh] + k-1[HBcat]
Conclusion
-kobs[(MeCp)Mn(CO)2(HBcat)]
k1k2[PhCCPh]
The synthesis and isolation of σ-borane complexes of
manganese and rhenium show that σ-borane complexes are not
restricted to early transition metals such as titanium and can
encompass many or all of the transition metals. These complexes
can be constructed in three different ways: coordination of a
metal fragment to the free borane, alcoholysis of a bis-boryl
complex, or borylation of an anionic metal-hydride complex.
The thermal stability of the borane complexes of manganese is
comparable to that of other σ-complexes of Mn(I) and is
significantly higher than that for titanocene-borane complexes.
The most important difference between σ-borane compounds
and X-H σ-complexes of ligands with saturated elements X is
the availability of a free p-orbital at boron. This orbital allows
for back-donation from the HOMO of the metal fragment into
the boron.12 In alkane, silane, or stannane complexes, back-
donation occurs into the H-X σ*-orbital, which is much higher
in energy than the borane p-orbital.
with kobs
)
(1)
(2)
k2[PhCCPh] + k-1[HBcat]
k-1
[HBcat]
1
1
)
+
kobs k1 k k
2[PhCCPh]
1
Values for k1, the rate constant for dissociation of the borane
ligand, and for k2/k-1, the partitioning of the 16-electron
(35) Alt, H.; Herberhold, M.; Kreiter, C. G.; Strack, H. J. Organomet.
Chem. 1974, 77, 353.
(36) Herberhold, M.; Kreiter, C. G.; Stu¨ber, S.; Wiedersatz, G. O. J.
Organomet. Chem. 1975, 96, 89.
(37) Faller, J. W.; Johnson, B. V. J. Organomet. Chem. 1975, 88, 101.
(38) Khaleel, A.; Klabunde, K. J.; Johnson, A. J. Organomet. Chem.
1999, 572, 11.
(39) Lee, S. W.; Yang, K.; Martin, J. A.; Bott, S. G.; Richmond, M. G.
Inorg. Chim. Acta 1995, 232, 57.
(40) Hart-Davis, A. J.; Graham, W. A. G. J. Am. Chem. Soc. 1971, 93,
4388.
(41) Heinekey, D. M.; Oldham, W. J. Chem. ReV. 1993, 93, 913.

9442 J. Am. Chem. Soc., Vol. 122, No. 39, 2000
Schlecht and Hartwig
(C6D6): δ 45 [d, 1J(1H,11B) ) 88 Hz]. IR (hexanes, cm-1): 1983 (νsCO,
s), 1921 (νasCO, s), 1603 (m).
Experimental Section
General considerations. Unless otherwise noted, all manipulations
Preparation of (MeCp)Mn(CO)2(HBMe2) (3). K[(MeCp)Mn-
(CO)2H] (0.050 g, 0.2 mmol) was suspended in 1.5 mL of pentane,
and 0.021 mL (0.2 mmol) of BrBMe2 were added with vigorous stirring.
After 10 min, the red mixture was filtered, and the solvent was
evaporated under reduced pressure. Compound 4 was obtained as a
red oil, which was judged to be pure by NMR spectroscopy. Yield:
0.042 g (82%). 1H NMR (C6D6): δ 4.10 (s, 2H), 3.99 (s, 2H), 1.42 (s,
3H), 1.21 (s, 6H), -17.06 (br, 1H). 13C NMR (C6D6): δ 226.9, 103.5,
85.0, 84.4, 24.3(br), 13.2. 11B NMR (C6D6): δ 101 (s). IR (hexanes,
cm-1): 1975 (νsCO, s), 1910 (νasCO, s), 1592 (m).
were conducted using standard Schlenk techniques or in an inert
1
atmosphere glovebox. H NMR spectra were obtained on either a GE
QE 300 MHz or a GE Ω 300 MHz Fourier transform spectrometer.
13C NMR spectra were recorded on either a Bruker AM 500 MHz or
a GE QE 300 MHz spectrometer. 11B and 2H NMR spectra were
obtained on the GE Ω 300 spectrometer operating at 96.38 and 46.13
MHz, respectively. 1H NMR spectra were recorded relative to residual
protiated solvent. 11B NMR spectra were recorded in units of parts per
million relative to BF3‚Et2O as an external standard.
Unless specified otherwise, all reagents were purchased from
commercial suppliers and used without further purification. K[(MeCp)-
Mn(CO)2H],21 B-chloro-pinacolborane,42 B-deuterio-catecholborane,43
and cis-Cp*Re(CO)2(Bpin)226 were prepared using literature procedures.
B-Chloro-pinacolborane was used in situ as prepared and was not
isolated. Protiated solvents were distilled from purple solutions contain-
ing sodium and benzophenone. Deuterated solvents were dried similarly
but were collected by vacuum transfer. Reaction yields that were
Preparation of (MeCp)Mn(CO)2(HBCy2) (4). K[(MeCp)Mn-
(CO)2H] (0.050 g, 0.2 mmol) was suspended in 1.5 mL of pentane,
and 0.220 mL (0.2 mmol) of a 1.0 M solution of ClBCy2 in hexanes
were added with vigorous stirring. The mixture turned red immediately.
After 10 min, the reaction was filtered, and the solvent was evaporated
under reduced pressure. The borane complex was obtained as a red
microcrystalline solid. Yield: 0.069 g (85%). 1H NMR (C6D6): δ 4.28
(s, 2H), 4.08 (s, 2H), 1.1-1.9 (m, 25H), -16.96 (br, 1H). 13C NMR
(C6D6): δ 226.6, 102.7, 85.3, 84.3, 48.3(br), 30.4, 28.2, 27.3, 13.3.
11B NMR (C6D6): δ 104 (s). IR (hexanes, cm-1): 1967 (νsCO, s), 1901
(νasCO, s), 1597 (m).
1
obtained by H NMR spectroscopy were determined by using C6Me6
as the internal standard.
Photochemical reactions were conducted using a Hanovia 450 W
medium-pressure mercury-arc lamp placed in a Pyrex immersion well.
Preparation of (MeCp)Mn(CO)2(HBcat) (1). (a) Photochemical
Synthesis. (MeCp)Mn(CO)3 (500 mg, 2.3 mmol) was dissolved in 80
mL of THF, and 3.30 g (27.5 mmol) of catecholborane were added.
The solution was irradiated for 30 min at room temperature. The solvent
was evaporated from the deep-yellow solution, and a dark-yellow oil
was obtained. Pentane (50 mL) was added, the solution was filtered,
and its volume was reduced to 4 mL. This concentrated solution was
stored at -30 °C for 14 h, and the resulting yellow solid was
recrystallized from pentane to obtain yellow crystals of 1 in 53% yield
(0.378 g).
Preparation of Cp*Re(CO)2(HBpin) (5). cis-Cp*Re(CO)2(Bpin)2
(0.046 g, 0.07 mmol) was dissolved in 1.0 mL of benzene, and 2.9 µL
(0.07 mmol) of methanol were added slowly. After stirring for 2 h the
volatile materials were removed, and a colorless solid remained. This
solid was recrystallized from 0.5 mL of pentane at -30 °C. Cp*Re-
(CO)2(HBpin) was obtained in quantitative yield (0.037 g). The reaction
also proceeds with other alcohols but the use of methanol allowed for
1
evaporation of the MeOBpin byproduct under reduced pressure. H
NMR (C6D6): δ 1.88 (s, 15H), 1.15 (s, 12H), -11.06 (br, 1H). 11B
NMR (C6D6): δ 46 (s). IR (hexanes, cm-1): 1981 (νsCO, s), 1924
(νasCO, s), 1603 (m).
(b) Salt extrusion. K[(MeCp)Mn(CO)2H] (1.00 g, 4.3 mmol) was
suspended in 30 mL of pentane, and 0.67 g (4.3 mmol) of B-
chlorocatecholborane were added with vigorous stirring. Immediate
formation of KCl was observed, and the solution turned yellow. After
10 min, the reaction mixture was filtered, reduced in volume, and stored
at -30 °C for 14 h. The product was recrystallized from pentane to
Ligand-Substitution Reactions. (a) Thermal Reactions. (MeCp)-
Mn(CO)2(HBcat) (15 mg, 0.048 mmol) and 3.9 mg (0.024 mmol) of
C6Me6 as an internal standard were dissolved in 0.6 mL of C6D6. Five
equiv of the appropriate incoming ligand were added. In the case of
the reaction with CO, 2 atm of CO were added. The mixtures were
reacted at 60 °C for 2 h in an NMR sample tube. Yields of product
1
provide 1.286 g of 1 as a microcrystalline powder (92% yield). H
1
were determined by comparison of the reactant and product H NMR
NMR (C6D6): δ 7.04 (m, 2H), 6.77 (m, 2H), 4.15 (s, 2H), 4.07 (s,
2H), 1.50 (s, 3H), -14.46 (br, 1H). 13C NMR (THF-d8): δ 226.1, 151.2,
121.4, 110.5, 104.1, 83.8, 83.4, 12.7. 11B NMR (C6D6): δ 46 (s, br).
integrals versus those for the internal standard.
(b) Photochemical Reactions. Samples were prepared as described
for the thermal reactions. For the reaction with HSiPh2Me, C6D6 was
used as a solvent; for reaction with pinacolborane, THF-d8 was used.
The solutions were irradiated in NMR sample tubes for 30 min at room
temperature. The following compounds were obtained in the yields
provided in parentheses: (MeCp)Mn(CO)2(HBpin) (2) (69%), (MeCp)-
Mn(CO)2(HSiPh2Me) (94%), (MeCp)Mn(CO)2(PhCCPh) (94%), (MeCp)-
Mn(CO)2(HSnPh3) (91%) and (MeCp)Mn(CO)3 (92%).
Kinetic Studies of the Reaction of 1 with PhCCPh. All experi-
ments were monitored at 70 °C by UV/vis spectroscopy at a wavelength
of λ ) 420 nm, which is the absorption maximum of the diphenyl-
acetylene complex formed in the reaction. Data were fit to a single
exponential for the disappearance of starting material.
(a) Dependence of the Reaction Rate on the Concentration of
Catecholborane. A stock solution was prepared containing 4.0 mg
(0.013 mmol) of 1 and 34.6 mg (0.194 mmol) of PhCCPh in 20.0 mL
of toluene. For each individual experiment, 3.0 mL of the stock solution
were used, and 1.02 µL (0.009 mmol), 4.08 µL (0.038 mmol), 8.16 µL
(0.077 mmol), 16.32 µL (0.153 mmol), or 24.48 µL (0.230 mmol) of
catecholborane were added to the aliquots.
(b) Dependence of the Reaction Rate on the Concentration of
PhCCPh. A stock solution was prepared containing 3.0 mg (0.009
mmol) of 1 and 20.0 µL (0.167 mmol) of catecholborane in 40.0 mL
of toluene. For each individual experiment, 3.0 mL of the stock solution
were used, and 1.0 mg (0.006 mmol), 2.0 mg (0.011 mmol), 4.0 mg
(0.022 mmol), 8.0 mg (0.045 mmol), 32.0 mg (0.180 mmol), or 64.0
mg (0.360 mmol) of PhCCPh were added to the aliquots.
A value 1J(1H,11B) of 98 Hz was obtained from a high temperature 11
B
NMR spectrum (T ) 373 K). IR (hexanes, cm-1): 1995 (νsCO, s),
1937 (νasCO, s), 1606 (m, br). Anal. Calcd for C14H12BMnO4: C, 54.25;
H, 3.91. Found: C, 54.02; H, 3.86.
Preparation of (MeCp)Mn(CO)2(DBcat) (1-d). The deuterio
analogue of 1 was prepared in 48% yield in a manner similar to the
2
photochemical preparation of 1 but using DBcat. H NMR (C6H6): δ
-14.4 (s, br, 1D). 11B NMR (C6H6): δ 46 (s, br). IR (hexanes, cm-1):
1146 (m, br).
Preparation of (MeCp)Mn(CO)2(HBpin) (2). (a) Photochemical
Synthesis. Pinacolborane complex 2 was synthesized as described for
the photochemical preparation of 1, starting with 0.500 g (2.3 mmol)
of (MeCp)Mn(CO)3. Yield: 0.132 g (18%).
(b) Salt Extrusion. A solution of B-chloropinacolborane in pentane
was generated by the addition of 1.0 mL (1.0 mmol) of a 1.0 M solution
of BCl3 in heptane to a stirred solution of 0.118 g (1.0 mmol) of pinacol
in 5 mL of pentane. A white precipitate that formed during the addition
was removed by filtration, and the filtrate was used as obtained. This
filtrate was added to a suspension of 0.175 g (0.8 mmol) of K[(MeCp)-
Mn(CO)2H] in 8 mL of pentane. After 30 min, the solution was filtered,
and the solvent was evaporated under reduced pressure. The resulting
yellow solid was recrystallized twice from pentane, and yellow crystals
1
were obtained. Yield: 0.157 g (65%). H NMR (C6D6): δ 4.22 (s,
2H), 4.13 (s, 2H), 1.61 (s, 3H), 1.14 (s, 12H), -15.66 (br, 1H). 13C
NMR (THF-d8): δ 221.1, 104.9, 83.8, 82.8, 79.9, 29.9, 13.7. 11B NMR
(c) Measurement of Eyring Parameters. A stock solution was
prepared containing 3.0 mg (0.009 mmol) of 1, 20 µL (0.167 mmol)
(42) Herberich, G. E.; Fischer, A. Organometallics 1996, 15, 58.
(43) Ma¨nnig, D.; No¨th, H. J. Chem. Soc., Dalton Trans. 1985, 1689.

σ-Borane Complexes
J. Am. Chem. Soc., Vol. 122, No. 39, 2000 9443
of catecholborane and 105 mg (0.590 mmol) of PhCCPh in 20.0 mL
of toluene. Individual experiments were run at 55, 60, 65, 75, and 85
°C with 3.0 mL of the stock solution
radiation. Crystal data are provided in Table 1. A total of 4421
reflections was collected. The space group was determined to be P21/
a. For Z ) 4 and FW ) 318.06, the calculated density was 1.43 g/cm3.
The structure was solved by heavy-atom Patterson methods and
expanded using Fourier techniques. H1 was located on the difference
Fourier map. The non-hydrogen atoms were refined anisotropically.
Some hydrogen atoms, including H1, were refined isotropically, while
the rest were included in fixed positions. The final cycle of full-matrix
least-squares refinement was based on 3551 observed reflections [I >
5.00σ(I)] and 185 variable parameters and converged with unweighted
and weighted agreement factors of R ) 3.1% and Rw ) 4.9%. The
crystallographic data of 2 were deposited Cambridge Crystallographic
Data Center under reference code 147134.
Crystallographic Analysis of 4. Red single crystals of 4 that were
suitable for X-ray diffraction studies were obtained by cooling a solution
of 4 in pentane:toluene 9:1 at -30 °C for 3 days. X-ray diffraction
data for 4 were taken at 183 K with graphite monochromatic Mo KR
radiation. Crystal data are provided in Table 1. A total of 4323
reflections was collected. The space group was determined to be P1h.
For Z ) 2 and FW ) 368.20, the calculated density was 1.27 g/cm3.
The structure was solved by direct methods and expanded using Fourier
techniques. H1 was located on the difference Fourier map. The non-
hydrogen atoms were refined anisotropically. Some hydrogen atoms,
including H1, were refined isotropically, and the rest were included in
fixed positions. In the case of the methyl-group hydrogen atoms, one
hydrogen was located in the difference map and included at an idealized
distance to set the orientation of the other two hydrogen atoms. The
final cycle of full-matrix least-squares refinement was based on 3097
observed reflections [I > 5.00σ(I)] and 221 variable parameters and
converged with unweighted and weighted agreement factors of R )
3.7% and Rw ) 5.0%. The crystallographic data of 4 were deposited
in the Cambridge Crystallographic Data Center under reference code
147135.
Kinetic Studies of the Reaction of 4 with PhCCPh. All experi-
ments were monitored by UV/vis spectroscopy at a wavelength of λ )
440 nm. Data were fit to a single exponential for the disappearance of
starting material. Eyring measurement: A stock solution was prepared
containing 5.0 mg (0.014 mmol) of 1 and 181 mg (1.017 mmol) of
PhCCPh in 20.0 mL of toluene. Individual experiments were run at
25, 35, 40, 45, 50, and 55 °C with 3.0 mL of the stock solution.
Preparation of Dicyclohexyl-cis-1,2-diphenyl(vinyl)borane. To a
solution of 50.0 mg (0.28 mmol) of diphenylacetylene in 1.0 mL of
toluene, 50.0 mg (0.28 mmol) of dicyclohexylborane44 were added,
and the solution was stirred at room temperature for 15 min. The solvent
was evaporated under reduced pressure, and the remaining solid was
recrystallized from diethyl ether at -35 °C. The dicyclohexyl-cis-1,2-
diphenyl(vinyl)borane was obtained as a colorless, microcrystalline solid
1
in 92% yield (92 mg). H NMR (C6D6): δ 1.2-1.5 (m, 11 H), 1.6-
1.8 (m, 11H), 6.78 (s, 2H), 6.95-7.10 (m, 4H), 7.15-7.25 (m, 4H),
7.41 (s, 1H). 11B NMR (C7D8): δ 76 (s).
Crystallographic Analysis of 1. Yellow single crystals of 1 that
were suitable for X-ray diffraction studies were obtained by cooling a
solution of 1 in pentane at -30 °C for 2 days. X-ray diffraction data
for 1 were obtained at 183 K with graphite monochromatic Mo KR
radiation. Crystal data are provided in Table 1. A total of 4797
reflections were collected, 3465 of which were unique (Rint ) 0.030).
The space group was determined to be P1h. For Z ) 2 and FW ) 309.99,
the calculated density was 1.57 g/cm3. A SORTAV absorption
correction was applied. The structure was solved by heavy-atom
Patterson methods and was expanded using Fourier techniques. H1 was
located on the difference Fourier map. The non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were refined isotropically. The
final cycle of full-matrix least-squares refinement was based on 3091
observed reflections (I > 3.00σ(I)) and 229 variable parameters and
converged with unweighted and weighted agreement factors of R )
3.0% and Rw ) 4.4%. The crystallographic data of 1 were deposited
in the Cambridge Crystallographic Data Center under reference code
147133.
Acknowledgment. We thank Susan DeGala for the crystal
structure analyses of compounds 1, 2, and 4. The NSF is
gratefully acknowledged for supporting this work. S.S. thanks
the Deutsche Forschungsgemeinschaft for a postdoctoral fel-
lowship.
Crystallographic Analysis of 2. Yellow single crystals of 2 that
were suitable for X-ray diffraction studies were obtained by cooling a
solution of 2 in pentane at -30 °C for 4 days. X-ray diffraction data
for 2 were taken at 183 K with graphite monochromatic Mo KR
Supporting Information Available: Data for compounds
1, 2, and 4. This material is available free of charge via the
Internet at http://pubs.acs.org.
(44) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents; Academic
Press: London, 1988; p 426.
JA001546O