Dimeric rhodium ui-silylene and u-n2-silyl complexes: Catalytic silicon-silicon bond formation and X-ray structures of {PrPCCIPPrRhL-SiRR' (R = R' = Ph and R = Me, R' = Ph) and [{Pri2PCH2CH2PPri2}Rh(H)]2(-i72-H-SiMe2)2t

Article abstract of DOI:10.1021/om980651s

The stoichiometric and catalytic reactions of secondary silanes with the rhodium hydride bridged dimer [(dippe)Rh]2(/*-H)2 (1; dippe = l,2-bis(diisopropylphosphino)ethane) are described. The reaction of 1 with Pli2SiH2 results in the formation of the bis(

Full text of DOI:10.1021/om980651s

958
Organometallics 1999, 18, 958-969
Articles
Dim er ic Rh od iu m µ-Silylen e a n d µ-η2-Silyl Com p lexes:
Ca ta lytic Silicon -Silicon Bon d F or m a tion a n d X-r a y
Str u ctu r es of [{P r i P CH2CH2P P r i }Rh ]2(µ-SiRR′)2 (R ) R′
2
2
) P h a n d R ) Me, R′ ) P h ) a n d
[{P r i P CH2CH2P P r i }Rh (H)]2(µ-η2-H-SiMe2)2
†
2
2
Lisa Rosenberg,*,‡ Michael D. Fryzuk, and Steven J . Rettig§
Department of Chemistry, University of British Columbia, 2036 Main Mall,
Vancouver, British Columbia, Canada V6T 1Y6
Received J uly 29, 1998
The stoichiometric and catalytic reactions of secondary silanes with the rhodium hydride
bridged dimer [(dippe)Rh]2(µ-H)2 (1; dippe ) 1,2-bis(diisopropylphosphino)ethane) are
described. The reaction of 1 with Ph2SiH2 results in the formation of the bis(µ-silylene)
complex [(dippe)Rh]2(µ-SiPh2)2 (2a ); a similar reaction ensues upon addition of MePhSiH2
or MepTolSiH2 (pTol ) p-tolyl) to 1 except that the bis(silylene) complexes exist as a mixture
of cis and trans stereoisomers. Dimethylsilane reacts with 1 to generate the dinuclear complex
[(dippe)Rh(H)]2(µ-η2-H-SiMe2)2 (4d ). The hydride dimer acts as a catalyst precursor for the
dimerization of excess Ph2SiH2 to tetraphenyldisilane (Ph2SiHSiHPh2). A catalytic cycle is
proposed that consists of dinuclear intermediates.
In tr od u ction
become generally accepted (Scheme 1),11,12 but while
individual steps in this cycle have been observed in
stoichiometric reactions of various mononuclear, late
transition-metal systems,4 for no one system have all
the steps been observed to occur stoichiometrically.
Studies of silane dehydropolymerization using late-
metal catalysts have not ruled out the possibility of
dinuclear intermediates.
The activation of Si-H bonds by transition-metal
complexes, originally the focus of research into homo-
geneously catalyzed hydrosilation,1-3 has come under
renewed scrutiny to determine its role in dehydrogena-
tive silicon coupling reactions.4-6 Silicon-silicon bond
formation catalyzed by d0 metallocene catalysts is now
known to occur via σ-bond metathesis reactions at
mononuclear centers,7 but despite long-standing prece-
dence for the activity of late-transition-metal catalysts
toward this reaction,8-10 no mechanism of dehydroge-
native silicon-silicon coupling as catalyzed by the late
transition metals has been conclusively established. A
catalytic cycle based on oxidative-addition and reduc-
tive-elimination steps at a metal hydride center has
We report here the catalytic activity of the dinuclear
rhodium hydride complex [(dippe)Rh]2(µ-H)2 (1; dippe
) 1,2-bis(diisopropylphosphino)ethane) toward the de-
hydrogenative dimerization of diphenylsilane.13,14
A
series of dinuclear rhodium µ-silylene and µ-η2-silyl
complexes, whose synthesis and characterization are
described, are implicated as intermediates in this
process: their reactivity provides evidence for all steps
of a catalytic cycle responsible for silicon-silicon cou-
pling, occurring at a dinuclear center.
† This article is dedicated to the memory of Steven J . Rettig.
‡ Present address: Department of Chemistry, University of Mani-
toba, Winnipeg, Manitoba, Canada R3T 2N2.
§ Professional Officer: UBC X-ray Crystal Structural Laboratory.
Deceased October 27, 1998.
Resu lts a n d Discu ssion
Syn t h esis, Ch a r a ct er iza t ion a n d R ea ct ivit y of
Dim er ic, µ-Silylen e Com p lexes. Addition of 2 equiv
(1) Chalk, A. J .; Harrod, J . F. J . Am. Chem. Soc. 1965, 87, 16.
(2) Harrod, J . F.; Chalk, A. J . In Organic Syntheses Via Metal
Carbonyls; Wender, I., Pino, P., Eds.; Wiley: New York, 1977; p 690.
(3) Ojima, I. In The Chemistry of Organic Silicon Compounds; Patai,
S., Rappoport, Z., Eds.; Wiley: New York, 1989; p 1479.
(4) Tilley, T. D. Comments Inorg. Chem. 1990, 10, 37.
(5) Yamashita, H.; Tanaka, M. Bull. Chem. Soc. J pn. 1995, 68, 403.
(6) Sharma, H. K.; Pannell, K. H. Chem. Rev. 1995, 95, 1351.
(7) Woo, H.; Walzer, J . F.; Tilley, T. D. J . Am. Chem. Soc. 1992,
114, 7047.
(11) Curtis, M. D.; Epstein, P. S. Adv. Organomet. Chem. 1981, 19,
213.
(12) Other mechanisms for this reaction have been advanced that
are based on the insertion of terminal silylene ligands into M-Si
bonds: Corey, J . Y. In Advances in Silicon Chemistry; Larson, G. L.,
Ed.; J AI Press: Greenwich, CT, 1991; Vol. 1, p 327. Hengge, E.;
Weinberger, M. J . Organomet. Chem. 1993, 443, 167.
(13) The activity of 1 for this catalytic dimerization was reported
previously in a communication; mechanistic details were unavailable
at that time.14
(8) Brown-Wensley, K. A. Organometallics 1987, 6, 1590.
(9) Lappert, M. F.; Maskell, R. K. J . Organomet. Chem. 1984, 264,
217.
(10) Ojima, I.; Inaba, S.; Kogure, T. J . Organomet. Chem. 1973, 55,
C7.
(14) Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J . Organometallics
1991, 10, 2537.
10.1021/om980651s CCC: $18.00 © 1999 American Chemical Society
Publication on Web 02/13/1999

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 959
Sch em e 1
chelate rings does not affect the C2 axis running through
both Si atoms. The observed diamagnetism of 2a
requires a formal single bond between the two rhodium
atoms, and this is borne out by the Rh-Rh separation
of 2.921(2) Å and the acute silylene bridge angles of
77.25(8) and 76.57(7)°. The Si-Si distance across the
dinuclear rhodium center is approximately 3.71 Å,19
which precludes any interaction between the two sili-
cons. The dimensions of this structure are consistent
with those generally observed for dimeric M2Si2 com-
plexes.20
Complex 2a is relatively inert, being unreactive
toward CO and C2H4, but it does react with hydrogen.
Solutions of 2a in hexanes or toluene placed under an
atmosphere of hydrogen react to give [(dippe)Rh]2(µ-H)-
(µ-η2-H-SiPh2) (3a ) plus diphenylsilane, as determined
of diphenylsilane to [(dippe)Rh]2(µ-H)2 (1) results in the
formation of the bis(µ-silylene) complex [(dippe)Rh]2(µ-
SiPh2)2 (2a ) with simultaneous loss of 3 equiv of
hydrogen (eq 1).15,16 This complex, soluble in aromatic
1
by H and 31P{1H} NMR spectroscopy. The dihydride
dimer 1 is not observed in these reactions; this is in
keeping with the observation that addition of hydrogen
to 3a does not cause the elimination of diphenylsilane.21
Thus, the addition of 1 equiv of diphenylsilane to 1 is
an irreversible reaction. The reaction of complex 2a with
excess diphenylsilane will be discussed in the section
describing catalytic results.
The methylphenylsilylene analogue of 2a , [(dippe)-
Rh]2(µ-SiMePh)2 (2b), was isolated as an orange powder.
The 31P{1H} spectrum of 2b indicates the formation of
two products in this reaction; these are geometric
isomers arising from the potentially cis or trans disposi-
tion of the Si-Me (or Si-Ph) groups across the Rh2Si2
plane (eq 2).
(1)
solvents and sparingly soluble in aliphatic hydrocarbon
and halogenated solvents, is air- and moisture-sensitive
in solution but is air-stable in the solid state for short
periods of time (hours). The 31P{1H} NMR spectrum of
2a shows a center-symmetric doublet of multiplets
typical of an AA′A′′A′′′XX′ system; the pattern is similar
to those observed for other dinuclear rhodium complexes
in this system, including that of 1.17,18 The 1H NMR
spectrum of 2a is consistent with a solution structure
in which the rhodium atoms have a square-planar
arrangement of ligands, with substituents on the tet-
rahedral silicons projecting into and out of the plane.
X-ray diffraction analysis of a single crystal of 2a
confirmed the dimeric structure of this molecule with
bridging silylene groups. Two views of the molecular
structure of 2a are shown in Figure 1, crystallographic
data are given in Table 1, and pertinent bond distances
and angles are shown in Table 2 . Each rhodium center
in the dimer has a distorted-square-planar geometry (if
one ignores the Rh-Rh bond); when the dimer is viewed
down the Rh-Rh axis, the two square planes are
skewed relative to each other, with an angle between
the two mean square planes of 46.95°. Loss of planarity
due to puckering of the two five-membered Rh(dippe)
(2)
The H and 31P{1H} NMR spectra of 2b did not allow
1
peak assignments for the cis and trans isomers, and the
two could not be separated by fractional crystallization.
The extremely low solubility of these compounds has
hampered study of this isomerism, preventing determi-
(19) The coordinates for the molecular structure of 3a were entered
into the Chem 3D Plus program through a Molecule Editor using the
Cache system by Techtronix, and the Si-Si distance was calculated
using the Molecule Editor.
(20) Anderson, A. B.; Shiller, P.; Zarate, E. A.; Tessier-Youngs, C.
A.; Youngs, W. J . Organometallics 1989, 8, 2320.
(21) Rosenberg, L. Ph.D. Thesis, University of British Columbia,
1993.
(15) 2a can also be prepared by addition of 1 equiv of diphenylsilane
to [(dippe)Rh]2(µ-H)(µ-η2-H-SiPh2) (3a ).16
(16) Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J . Organometallics
1996, 15, 2871.
(17) Fryzuk, M. D.; J ones, T.; Einstein, F. W. B. Organometallics
1984, 3, 185.
(18) Piers, W. E. Thesis, University of British Columbia, 1988.

960 Organometallics, Vol. 18, No. 6, 1999
Ta ble 1. Cr ysta llogr a p h ic Da ta a
2a trans-2b
52H84P4Rh2Si2
Rosenberg et al.
4d
formula
fw
C
C42H80P4Rh2Si2
970.97
C32H80P4Rh2Si2
1095.11
brown, prism
monoclinic
C2/c (No. 15)
12.933(6)
22.538(6)
18.875(8)
90
99.99(4)
90
5418(4)
4
850.86
color, habit
cryst syst
space group
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
yellow-orange, prism
triclinic
P1h (No. 2)
11.307(3)
12.036(3)
10.798(2)
104.26(2)
117.38(2)
67.77(2)
1203.2(5)
1
yellow, prism
orthorhombic
P212121 (No. 19)
16.869(2)
19.741(3)
12.862(2)
90
90
90
4283(2)
Z
F
4
calcd, g cm-3
1.342
2296
1.340
510
1.319
1800
9.80
F(000)
µ(Mo KR), cm-1
cryst size, mm
transmissn factors
scan type
7.91
8.81
0.05 × 0.15 × 0.30
0.63-1.00
ω-2θ
0.10 × 0.15 × 0.40
0.90-1.00
ω-2θ
0.20 × 0.25 × 0.45
0.93-1.00
ω-2θ
scan range, ω, deg
scan speed, deg min-1
data collected
2θmax, deg
cryst decay, %
total no. of rflns
no. of unique rflns
Rmerge
1.21 + 0.35 tan θ
16 (up to 9 scans)
+h, +k, (l
55
6.4
6675
1.00 + 0.35 tan θ
16 (up to 9 scans)
+h, (k, (l
65
negligible
9114
1.31 + 0.35 tan θ
32 (up to 9 scans)
+h, +k, +l
60
negligible
6883
6392
0.048
8726
0.028
6883
no. of rflns with I g 3σ(I)
no. of variables
R
2756
272
0.036
6177
227
0.030
4936
388
0.025
Rw
0.028
0.029
0.026
GOF
1.45
1.68
1.18
max ∆/σ (final cycle)
0.16
-0.41 to +0.39
0.08
-0.35 to +0.51
0.04
-0.34 to +0.28
residual density, e Å-3
a
Conditions and definitions: temperature, 294 K; Rigaku AFC6S diffractometer; Mo KR (λ ) 0.710 69 Å) radiation; graphite
monochromator; takeoff angle 6.0°, aperture 6.0 × 6.0 mm at a distance of 285 mm from the crystal; stationary background counts at
each end of the scan (scan/background time ratio 2:1); σ2(F2) ) [S2(C + 4B)]/Lp2 (S ) scan rate, C ) scan count, B ) normalized background
count), function minimized ∑w(|Fo| - |Fc|)2, where w ) 4Fo2/σ2(Fo2); R ) ∑||Fo| - |Fc||/∑|Fo|; Rw ) (∑w(|Fo| - |Fc|)2/∑w|Fo| )1/2; GOF )
2
[∑w(|Fo| - |Fc|)2/(m - n)]1/2 (where m ) number of observations, n ) number of variables). Values given for R, Rw, and GOF are based on
those reflections with I g 3σ(I).
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 2a a
Rh(1)-Rh(1)*
Rh(1)-P(2)
Rh(1)-Si(2)
P(1)-C(3)
2.921(2)
2.288(2)
2.054(5)
1.852(6)
1.860(6)
1.864(7)
1.898(6)
Rh(1)-P(1)
Rh(1)-Si(1)
P(1)-C(1)
P(1)-C(4)
P(2)-C(5)
Si(1)-C(15)
2.295(2)
2.357(2)
1.831(6)
1.886(6)
1.848(6)
1.911(5)
P(2)-C(2)
P(2)-C(6)
Si(2)-C(21)
Rh(1)*-Rh(1)-P(1) 135.34(4) Rh(1)*-Rh(1)-P(2) 138.26(5)
Rh(1)*-Rh(1)-Si(1) 51.71(4) Rh(1)*-Rh(1)-Si(2) 51.37(4)
P(1)-Rh(1)-P(2)
P(1)-Rh(1)-Si(2)
P(2)-Rh(1)-Si(2)
Rh(1)-P(1)-C(1)
Rh(1)-P(1)-C(4)
C(1)-P(1)-C(4)
Rh(1)-P(2)-C(2)
Rh(1)-P(2)-C(6)
C(2)-P(2)-C(6)
86.34(6) P(1)-Rh(1)-Si(1)
86.21(6) P(2)-Rh(1)-Si(1)
159.30(5) Si(1)-Rh(1)-Si(2)
108.3(2) Rh(1)-P(1)-C(3)
122.2(2) C(1)-P(1)-C(3)
102.9(3) C(3)-P(1)-C(4)
107.7(2) Rh(1)-P(2)-C(5)
123.1(3) C(2)-P(2)-C(5)
102.8(4) C(5)-P(2)-C(6)
162.41(5)
89.45(6)
103.09(5)
116.0(2)
101.9(3)
103.0(3)
116.4(2)
100.2(3)
103.5(3)
Rh(1)-Si(1)-Rh(1)* 76.57(7) Rh(1)-Si(1)-C(15) 121.7(2)
Rh(1)-Si(1)-C(15)* 115.5(2) C(15)-Si(1)-C(15)* 105.0(4)
Rh(1)-Si(2)-Rh(1)* 77.25(8) Rh(1)-Si(2)-C(21) 119.8(2)
Rh(1)-Si(2)-C(21)* 114.4(2) C(21)-Si(2)-C(21)* 108.7(4)
F igu r e 1. ORTEP view of 2a (33% probability thermal
ellipsoids are shown for the non-hydrogen atoms).
a
1
Asterisks denote the symmetry operation 1 - x, y,
/ - z.
2
diagram of trans-2b and Table 3 for relevant bond
distances and bond angles.) The dimensions of the
molecular structure of trans-2b are analogous to those
observed for 2a , except that far less twisting of the two
chelate rings relative to each other renders the geom-
etries at rhodium closer to square planar (if one ignores
nation of the ratio of isomers formed. However, crystals
of the complex suitable for an X-ray diffraction study
were obtained by slow evaporation of solvent from a very
dilute solution of 2b in hexanes. The structure obtained
was of the trans isomer. (See Figure 2 for an ORTEP

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 961
F igu r e 2. ORTEP view of trans-2b (33% probability
thermal ellipsoids are shown for the non-hydrogen atoms).
Ta ble 3. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for tr a n s-2ba
F igu r e 3. ORTEP view of 4d (33% probability thermal
ellipsoids are shown for the non-hydrogen atoms).
Rh(1)-Rh(1)*
Rh(1)-P(2)
Rh(1)-Si(1)*
P(1)-C(3)
2.8925(9)
2.2760(8)
2.3531(8)
1.866(3)
1.848(3)
1.865(3)
1.902(2)
Rh(1)-P(1)
Rh(1)-Si(1)
P(1)-C(1)
P(1)-C(4)
P(2)-C(5)
Si(1)-C(15)
2.273(1)
2.345(1)
1.856(3)
1.856(3)
1.868(3)
1.908(2)
with NMR spectra for trans-2b) is less soluble than the
cis isomer and has been isolated pure by fractional
crystallization. This isomer is stable in solution at room
temperature but when heated to 70 °C or higher begins
to convert to the cis isomer. At these high temperatures
an equilibrium between the isomers is established, with
64% trans to 36% cis being the highest conversion
observed. The approach to a cis/trans equilibrium from
pure trans-2c at various temperatures was monitored
P(2)-C(2)
P(2)-C(6)
Si(1)-C(16)
Rh(1)*-Rh(1)-P(1) 136.27(2) Rh(1)*-Rh(1)-P(2)
136.89(2)
Rh(1)*-Rh(1)-Si(1) 52.12(2) Rh(1)*-Rh(1)-Si(1)* 51.88(2)
P(1)-Rh(1)-P(2)
P(1)-Rh(1)-Si(1)*
86.83(3) P(1)-Rh(1)-Si(1)
86.93(3) P(2)-Rh(1)-Si(1)
160.92(3)
87.07(3)
104.00(3)
117.3(1)
101.9(1)
103.8(1)
117.98(9)
101.7(1)
104.1(1)
126.37(8)
P(2)-Rh(1)-Si(1)* 161.52(2) Si(1)-Rh(1)-Si(1)*
1
Rh(1)-P(1)-C(1)
Rh(1)-P(1)-C(4)
C(1)-P(1)-C(4)
Rh(1)-P(2)-C(2)
Rh(1)-P(2)-C(6)
C(2)-P(2)-C(6)
107.93(8) Rh(1)-P(1)-C(3)
119.8(1) C(1)-P(1)-C(3)
103.9(1) C(3)-P(1)-C(4)
107.85(8) Rh(1)-P(2)-C(5)
119.07(9) C(2)-P(2)-C(5)
103.9(1) C(5)-P(2)-C(6)
by H NMR spectroscopy. However, kinetic analysis of
the resulting reaction profiles was prevented by poor
reproducibility of the results at specific temperatures
and by inconsistent temperature dependence overall.
Syn th esis, Ch a r a cter iza tion a n d Rea ctivity of a
Dim er ic, µ-η2-Silyl Com p lex. Addition of 2 equiv of
dimethylsilane to 1 (or addition of 1 equiv of dimethyl-
silane to [(dippe)Rh]2(µ-H)(µ-η2-H-SiMe2) (3d )16) in
hexanes gives a pale, golden yellow solution from which
yellow crystals of the formula [(dippe)Rh]2(Me2SiH2)2
(4d ) can be obtained (eq 3).
Rh(1)-Si(1)-Rh(1)* 76.00(3) Rh(1)-Si(1)-C(15)
Rh(1)-Si(1)-C(16) 112.81(8) Rh(1)*-Si(1)-C(15) 127.01(8)
Rh(1)*-Si(1)-C(16) 111.02(8) C(15)-Si(1)-C(16)
102.4(1)
a
Asterisks denote the symmetry operation 1 - x, 1 - y, 1 - z.
the Rh-Rh bond). The single crystal used for the
diffraction experiment was dissolved in d8-toluene and
1
its H NMR spectrum recorded, allowing assignment of
the peaks due to the trans isomer in spectra of mixtures
of the two isomers. The other peaks were assigned to
the cis isomer.
(3)
A more soluble cis/trans isomer pair of the bis(µ-
silylene) complex 2 is isolated from the reaction of 1 with
2 equiv of methyl-p-tolylsilane (eq 2). Alternatively, the
intermediate
complex
[(dippe)Rh]2(µ-H)(µ-η2-H-
SiMeTolp) (3c), which results from the addition of 1
equiv of methyl-p-tolylsilane to 1 and shows details in
An X-ray crystallographic study carried out on a
single crystal of 4d confirms the dinuclear structure of
this complex. (See Figure 3 for an ORTEP diagram and
Table 4 for pertinent bond lengths and bond angles.)
The structural relationships between phosphine, silyl,
and hydride ligands are more clearly presented in the
side and end views of 4d shown in Figure 4. The
1
its H and 31P{1H} NMR spectra similar to those seen
for [(dippe)Rh]2(µ-H)(µ-η2-H-SiMePh) (3b),16 reacts with
another 1 equiv of methyl-p-tolylsilane to produce a
mixture of cis- and trans-2c. These isomers are initially
formed in a trans/cis ratio of roughly 2:1. The trans
isomer (designated trans on the basis of comparison

962 Organometallics, Vol. 18, No. 6, 1999
Rosenberg et al.
hydride. The geometry around the rhodium centers is
roughly octahedral. Two edge-sharing planes, each of
which contains the two rhodium atoms, a silicon atom,
and the “agostic” hydride associated with that silicon,
are almost orthogonal to each other. A dihedral angle
of 74.5° gives the dinuclear core a “butterfly” structure
(Figure 4).
The observed diamagnetism of [(dippe)Rh(H)]2(µ-η2-
H-SiMe2)2 (4d ) indicates the presence of a formal,
single Rh-Rh bond, consistent with a Rh-Rh separa-
tion of 2.8575(5) Å. While the Si-Si separation in 4d of
3.135 Å22 is sufficiently long to preclude any bonding
interaction (the normal range for a Si-Si single bond
is 2.33-2.70 Å23), this distance is much shorter than
those normally observed for dimeric complexes contain-
ing bridging silylene ligands (3.852-4.225 Å24). Obvi-
ously the shorter distance is due largely to steric
considerations in the butterfly-shaped structure, but it
does give rise to the possibility of elimination of a
silicon-silicon-bonded entity from the dinuclear center.
The relevance of this structure to dehydrogenative
coupling of silanes catalyzed by 1 is discussed further
in the section concerned with the catalytic dimerization
of diphenylsilane (vide infra).
F igu r e 4. (a) Side and (b) end schematic views of the solid-
state structure of 4d . In the end view, the phosphine
ligands have been omitted for clarity.
Ta ble 4. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 4d
Rh(1)-Rh(2)
Rh(1)-P(2)
Rh(1)-Si(2)
Rh(1)-H(3)
Rh(2)-P(4)
Rh(2)-Si(2)
Rh(2)-H(4)
Si(1)-C(30)
Si(2)-C(31)
Si(2)-H(2)
2.8575(5)
2.269(1)
2.337(1)
1.52(5)
2.364(1)
2.474(1)
1.52(4)
1.889(5)
1.896(5)
1.72(4)
Rh(1)-P(1)
Rh(1)-Si(1)
Rh(1)-H(1)
Rh(2)-P(3)
Rh(2)-Si(1)
Rh(2)-H(2)
Si(1)-C(29)
Si(1)-H(1)
Si(2)-C(32)
2.391(1)
2.526(1)
1.69(4)
2.279(1)
2.324(1)
1.70(4)
1.910(5)
1.67(5)
1.892(5)
Features related to the two agostic Si-H bonds in the
structure of 4d are very similar to those observed for
the silyl hydride complexes 3a and 3d .16 The Rh-Si
Rh(2)-Rh(1)-P(1) 116.71(3) Rh(2)-Rh(1)-P(2) 156.18(3)
Rh(2)-Rh(1)-Si(1)
Rh(2)-Rh(1)-H(1)
P(1)-Rh(1)-P(2)
P(1)-Rh(1)-Si(2)
P(1)-Rh(1)-H(3)
P(2)-Rh(1)-Si(2)
P(2)-Rh(1)-H(3)
Si(1)-Rh(1)-H(1)
Si(2)-Rh(1)-H(1)
H(1)-Rh(1)-H(3)
50.69(3) Rh(2)-Rh(1)-Si(2)
92(2) Rh(2)-Rh(1)-H(2)
55.81(3)
83(2)
108.10(4)
89(2)
128.92(4)
93(1)
80.17(4)
133(2)
86.72(4) P(1)-Rh(1)-Si(1)
162.05(4) P(1)-Rh(1)-H(1)
87(2)
P(2)-Rh(1)-Si(1)
100.56(4) P(2)-Rh(1)-H(1)
95(2)
41(2)
107(2)
170(2)
Si(1)-Rh(1)-Si(2)
Si(1)-Rh(1)-H(3)
Si(2)-Rh(1)-H(3)
distances within the three-center, two-electron interac-
tions are 0.20 and 0.13 Å longer than the corresponding
unbridged Rh-Si bonds. In comparison to other di-
nuclear complexes with agostic Si-H bonds, these bond
“lengthenings” represent oxidative additions of the
Si-H bond to rhodium which have been arrested at an
early and an intermediate stage, respectively.25
76(2)
Rh(1)-Rh(2)-P(3) 157.52(3)
Rh(1)-Rh(2)-P(4) 113.99(3) Rh(1)-Rh(2)-H(2)
95(1)
86.88(4)
133.73(4)
91(2)
105.12(4)
86(2)
101(1)
44(1)
177(2)
Rh(1)-Rh(2)-H(4)
P(3)-Rh(2)-Si(1)
P(3)-Rh(2)-H(2)
P(4)-Rh(2)-Si(1)
P(4)-Rh(2)-H(2)
Si(1)-Rh(2)-Si(2)
Si(1)-Rh(2)-H(4)
Si(2)-Rh(2)-H(4)
Rh(1)-Si(1)-Rh(2)
Rh(1)-Si(1)-C(30) 125.5(2)
Rh(2)-Si(1)-C(29) 125.5(2)
Rh(2)-Si(1)-H(1)
C(29)-Si(1)-H(1)
Rh(1)-Si(2)-Rh(2)
82(2)
100.37(4) P(3)-Rh(2)-Si(2)
92(1) P(3)-Rh(2)-H(4)
162.41(4) P(4)-Rh(2)-Si(2)
94(1) P(4)-Rh(2)-H(4)
P(3)-Rh(2)-P(4)
Complex 4d is fluxional with an average structure in
solution that is symmetric: all four phosphines are
chemically equivalent, as are the four hydrides. When
the temperature of a solution of 4d is lowered and
81.52(4) Si(1)-Rh(2)-H(2)
78(2)
133(2)
Si(2)-Rh(2)-H(2)
H(2)-Rh(2)-H(4)
72.06(3) Rh(1)-Si(1)-C(29) 118.5(2)
1
monitored by both 31P{1H} and H NMR spectroscopy,
Rh(1)-Si(1)-H(1)
42(2)
Rh(2)-Si(1)-C(30) 118.6(2)
decoalescence and spectra are observed that can be
interpreted in terms of a slow exchange regime. A
simple doublet observed at room temperature in the 31P-
{1H} NMR spectrum attests to the equivalence of the
four phosphines (with coupling to 103Rh) in the averaged
structure. At approximately -29 °C a coalescence point
is reached and at -54 °C two signals of equal intensity
can be resolved. Thus, the process that exchanges two
114(2)
95(2)
C(29)-Si(1)-C(30)
C(30)-Si(1)-H(1)
98.2(2)
101(2)
72.82(3) Rh(1)-Si(2)-C(31) 125.1(2)
Rh(1)-Si(2)-C(32) 119.6(2)
Rh(2)-Si(2)-C(31) 122.3(2)
Rh(2)-Si(2)-H(2)
C(31)-Si(2)-H(2)
Rh(1)-H(1)-Si(1)
Rh(1)-Si(2)-H(2)
Rh(2)-Si(2)-C(32) 120.8(2)
116(1)
43(1)
91(1)
97(2)
C(31)-Si(2)-C(32)
C(32)-Si(2)-H(2)
Rh(2)-H(2)-Si(2)
97.7(3)
102(1)
93(2)
hydrides attached to rhodium and silicon were located
and refined isotropically.
(22) The coordinates for the molecular structure of 4d were entered
into the Chem 3D Plus program through a Molecule Editor using the
Cache system by Techtronix, and the Si-Si distance was calculated
using the Molecule Editor.
(23) Pham, E. K.; West, R. Organometallics 1990, 9, 1517.
(24) Zarate, E. A.; Tessier-Youngs, C. A.; Youngs, W. J . J . Am. Chem.
Soc. 1988, 110, 4068.
Complex 4d contains two unsymmetrically bridging
silyl groups, each bound to one rhodium center through
a covalent Si-Rh bond and to the other rhodium center
through a Rh-H-Si three-center, two-electron bond.
Each rhodium atom is also bound to one terminal
(25) Schubert, U. Adv. Organomet. Chem. 1990, 30, 151.

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 963
Sch em e 2
pairs of phosphines between inequivalent sites in a less
symmetric structure has been slowed on the NMR time
scale at this temperature. A value for ∆Gq(244 K) of 10.2
( 0.2 kcal/mol was calculated for this process on the
basis of the coalescence of the signals observed at -54
°C. Further cooling of 4d to -94 °C causes four new
signals to emerge in the 31P{1H} NMR spectrum, all
broad doublets. The changes that occur in the spectra
between -54 and -94 °C do not resemble a normal
decoalescence and may be attributable to conformational
effects of the various chelate rings in concert with the
complicated core of the dinuclear complex. In the room-
temperature 1H NMR spectrum, a single hydride signal,
a multiplet at -11.7 ppm, splits into two broad singlets
as the sample is cooled to -64 °C, with coalescence
occurring at -39 °C. This indicates that the two types
of hydrides in the complex are being exchanged between
inequivalent sites; the process responsible for this
exchange has ∆Gq(234 K) ) 10.2 ( 0.2 kcal/mol, as
calculated from the observed coalescence. Similar to
what was seen in the 31P{1H} NMR spectra, when the
sample is cooled to lower temperatures new signals
emerge which appear to be due to four inequivalent
hydrides in a less symmetric structure. Again, the
changes in these spectra at low temperature are not
typical of decoalescences but are more likely to be due
to conformation effects.26
their J Rh-P values tend to be significantly reduced.29
Thus, the signal at 84 ppm (1J Rh-P ) 150 Hz) is due to
the phosphines trans to Rh (P2, P3), whereas the signal
at 72 ppm (1J Rh-P ) 97 Hz) is due to the phosphines
trans to Si (P1, P4).
The coalescences observed in the above variabe-
temperature NMR spectra are consistent with a flux-
ional process occurring for a solution structure that is
approximately the same as the solid-state structure of
4d . Two distinct phosphorus environments arise from
The hydride signals observed in the 1H NMR spec-
trum at -64 °C are also consistent with the structure
shown, where there are two types of hydride ligands:
two are involved in the agostic Si-H bonds to rhodium
(H1 and H2), and the others are terminal rhodium
hydrides (H3 and H4).30 Thus, the signal at -10.6 ppm
is assigned to the agostic hydrides (H1, H2) and the
signal at -12.7 ppm to the terminal hydrides (H3, H4).
A mechanism for exchange of phosphines and hy-
drides between inequivalent sites in 4d is based on one
previously established by variable-temperature NMR
studies for a series of analogous bis(silyl) complexes
derived from primary silanes (Scheme 2).31 The mech-
anism involves simultaneous twisting of the two rhod-
ium coordination spheres. At each rhodium center the
H-Rh-H axis rotates through 90°, with concerted
breaking of one agostic Si-H bond (e.g. Si-H1) and the
formation of another (e.g. Si-H3): this process ex-
changes the terminal hydrides on rhodium and the
agostic silicon hydrides. A concurrent twisting of the
phosphine chelate rings through 90° exchanges the two
types of phosphorus on each ligand (e.g. P1, originally
trans to silicon, becomes trans to Rh2, while P2 under-
goes the opposite change).
an orthogonal arrangement of the two chelate rings: one
phosphine from each bis(phosphine) ligand (P2, P3) is
trans to the Rh-Rh bond, and the two remaining
phosphines (P1, P4) are each trans to an unbridged Rh-
Si bond. P1 and P4 are chemically equivalent, as are P2
and P3, resulting in the two doublets observed in 31P-
{1H} NMR spectrum of 4d at -54 °C. The strong trans
influence of the silyl ligands in 4d 27,28 allows us to assign
the signals in this spectrum: empirical studies show
that 31P signals for phosphorus nuclei trans to strong
trans-influence ligands are shifted to higher field, and
(26) The structure of this unsymmetrical species forming at low
temperature in solutions of 4d is not obvious. It may be a conforma-
tional isomer of the more symmetric structure corresponding to the
low-temperature limit of the observed decoalescences, where the
phosphine ligands, and hence the hydrides, are less symmetrically
disposed.
The molecular structure of 4d suggests that the
complex should be extremely labile toward loss of either
H2 or Me2SiH2. The complex does decompose, particu-
larly in solution at room temperature or above, to give
(27) The strong trans influence of silyl ligands is manifested in the
solid-state structure of 4d in the lengthening of the Rh-P bond lengths
by 0.1 Å for the phosphines (P1, P4) trans to the silyl ligands relative
to the Rh-P bonds for the phosphines (P2, P3) trans to the Rh-Rh
bond.
(28) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987; pp 242-243.
(29) Meek, D. W.; Mazanec, T. J . Acc. Chem. Res. 1981, 14, 266.
(30) It is possible that in solution the agostic Si-H bonds in 4d have
completely oxidatively added to give four terminal hydride ligands
while the geometry of the complex remains approximately the same;
this situation would still give rise to two types of hydride ligands.
(31) Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J . Inorg. Chim. Acta
1994, 222, 345.

964 Organometallics, Vol. 18, No. 6, 1999
Rosenberg et al.
both [(dippe)Rh]2(µ-SiMe2)2 (2d ) and 3d (eq 4). The
(4)
decomposition of 4d can be directed toward one or the
other of these products by careful control of the reaction
conditions. When sealed NMR samples of 4d are heated,
the 31P{1H} NMR spectra show signals appearing for
both 2d and the hydrogen adduct of 3d , {[(dippe)Rh]2-
(µ-H)(µ-η2-H-SiMe2)}‚H2.21 However, when an NMR
sample is heated at 45 °C while open to a nitrogen
manifold and bubbler to allow any volatile products to
escape the NMR tube, monitoring by 31P{1H} NMR
shows the slow formation of a single product, 3d . The
positive pressure of nitrogen apparently discourages
formation of the bis(µ-silylene) complex 2d , and it may
be that a slight vacuum, as is present in sealed NMR
samples, is required to initiate the loss of hydrogen from
4d . Alternatively, addition of ethylene to a solution of
4d causes the quantitative formation of 2d (eq 5). It is
species have not been isolated or detected spectroscopi-
cally. While for 4d no elimination of a Si-Si-bonded
species was observed under any of these reaction
conditions, the increased crowding at proposed inter-
mediate 4a might facilitate Si-Si reductive coupling.32
Ca ta lytic Dim er iza tion of Dip h en ylsila n e. We
previously reported the catalytic activity of 1 in the
isotopic exchange of hydrides for deuterides on diphen-
ylsilane, with deuterium gas as a source of deuterides
(eq 6).14 Higher substrate concentrations and the use
(6)
of aromatic solvents appear to encourage a side reaction,
the disproportionation of diphenylsilane to triphenyl-
silane and phenylsilane (eq 7), though monitoring the
(5)
(7)
1
reaction in d6-benzene by H NMR spectroscopy indi-
cates that the deuterium exchange reaction occurs more
quickly than the disproportionation reaction.11,21,33 Very
high concentrations of diphenylsilane, or the addition
of the catalyst to the diphenylsilane prior to the
introduction of deuterium, tends to give small amounts
of 1,1,2,2-tetraphenyldisilane, the product of dehydro-
genative dimerization of diphenylsilane (eq 8). When the
not clear whether the ethylene in this reaction is being
hydrogenated by 4d , or if the reversible coordination of
ethylene simply prompts the elimination of hydrogen
from the bis(silane) complex. A similar reaction was
performed using styrene (PhCHdCH2); however, this
reaction did not proceed at room temperature, and
heating the mixture caused decomposition of 4d to 3d
and 2d . No ethylbenzene was detected. Complex 4d may
be regenerated by placing solutions of 2d under an
atmosphere of hydrogen.
(8)
Placing a solution of 4d under an atmosphere of
dihydrogen leaves the molecule apparently unchanged,
as shown by 31P{1H} NMR spectroscopy. However, there
is evidence for a slow exchange: NMR spectroscopic
monitoring of a sample of 4d sealed under deuterium
deuterium atmosphere is replaced by nitrogen, dimer-
ization of diphenylsilane becomes the major process
occurring in the presence of catalytic amounts of 1.14
The active species in the dimerization of diphenyl-
silane as catalyzed by 1 may be dinuclear rather than
1
gas by H NMR spectroscopy showed a slow decrease
in intensity of the hydride resonance relative to the
ligand resonances in the spectra.
(32) The recently reported structure of a dicobalt complex, con-
strained to a butterfly-shaped core by the bridging disilyl ligand -Si-
(Me)CH2CH2Si(Me)-, shows an extremely short Si‚‚‚Si separation of
2.691(2) Å. It is suggested that optimization of the Co-Si bonding in
this type of structure forces the reduction of the Si-Co-Si angles, due
to overlap of filled Co d orbitals with empty σ* orbitals located mainly
on the Si’s: Bourg, S.; Boury, B.; Carre´, F.; Corriu, R. J . P. Organo-
metallics 1997, 16, 3097. Similarly, a short Si‚‚‚Si separation of 2.85
Å was reported for a dinuclear rhodium complex with bridging silicons
where a “cradle” structure approaches the same geometry as observed
for 4d : Wang, W.-D.; Eisenberg, R. J . Am. Chem. Soc. 1990, 112, 2,
1833.
Formation of the bis(µ-silylene) complexes 2a -c prob-
ably proceeds via the initial bis(silane) adduct II, which
is analogous to 4d . As in the decomposition of 4d to 2d
described above, elimination of 2 equiv of H2 from an
intermediate such as II would generate a bis(µ-silylene)
complex; the formation of such an intermediate has
already been directly observed by monitoring the reac-
tions of primary silanes with 1.31 The butterfly-shaped
core of the proposed intermediates 4a -c would be
sterically hindered relative to 4d ; these higher energy
(33) The disproportionation of organosilicon compounds in the
presence of transition metals, requiring activation of both Si-H and
Si-C bonds, is quite common, though not well-understood.

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 965
Sch em e 3
Sch em e 4
mononuclear. Scheme 3 shows a possible catalytic cycle,
based on the same progression of oxidative-addition and
reductive-elimination steps as depicted in Scheme 1: all
of these reactions occur at dinuclear centers and gener-
ate dinuclear products, and all of the complexes shown
have been isolated or identified spectroscopically.
To establish the activity of 2a in this dehydrogenative
coupling reaction, we examined its reactivity with
diphenylsilane, in the absence of complexes 1 and 3a .
In the presence of excess diphenylsilane (4-5 equiv) an
orange solution of 2a in benzene reacts to give a yellow
solution. The 31P{1H} NMR spectrum of this solution is
complex but may be attributed to a single complex in
solution, compound 5. The 1H NMR spectrum of the
reaction mixture indicates the presence of both rhodium
hydrides and terminal silicon hydrides in 5 and also the
presence of triphenylsilane and unreacted diphenyl-
silane. Phenylsilane, the byproduct of the dispropor-
tionation reaction that produces triphenylsilane, is not
observed. It is possible that 5 results from the reaction
of 2a with PhSiH3 and is a “mixed silane” analogue of
tris(silyl) complexes previously isolated from the reac-
tions of n-butylsilane and p-tolylsilane with 1.31 Com-
plex 5 has not been isolated pure in the solid state. No
tetraphenyldisilane was observed in the reaction mix-
ture; therefore 2a is not responsible for the catalytic
silicon coupling reaction observed in the presence of 1
and diphenylsilane. This is consistent with our earlier
observation that 2a does not react with hydrogen to give
1.
Scheme 4 shows a more likely catalytic cycle, invoking
the activity of proposed intermediate 4a in the silicon-
silicon coupling reaction. As discussed above, 4a is
strongly implicated as an intermediate in the formation
of 2a . This complex should be extremely reactive (as
evidenced by the solution behavior of its dimethylsilane
analogue 4d ) and prone to reductive elimination of a
silicon-silicon bond (vide supra). Thus, in the stoichio-
metric chemistry of this system described so far, there
is evidence of (i) oxidative addition of silane to a metal
hydride complex with (ii) elimination of hydrogen to
generate a silyl complex16 and (iii) oxidative addition
of silane to a metal silyl complex.31
directly observed for the dinuclear rhodium center of
this system. However, the reverse reaction, oxidative
addition of a Si-Si bond to 1, does occur stoichiomet-
rically. When a mixture of 1,1,2,2-tetraphenyldisilane
and 1 in toluene is heated to 70 °C, the major product
observed by 31P{1H} NMR is 3a , the formation of which
requires the cleavage of the Si-Si bond of the disilane
(eq 9). Thus the transition state or intermediate re-
(9)
quired for a Si-Si bond-breaking step is thermodynami-
cally accessible for this dinuclear system; presumably
that for the Si-Si bond formation step must also be
accessible. The principal silicon-containing compound
in this reaction mixture is unreacted disilane. Scheme
5 outlines the various equilibria involved in this reac-
tion, assuming that the initial adduct of the disilane
with 1 is intermediate 4a , with a structure analogous
to 4d (step A). This intermediate has already been
identified as a likely precursor for the reductive elimi-
nation of a silicon-silicon bond from the dinuclear
rhodium center (vide supra) and, like 4d , would also be
capable of decomposing through elimination of either
hydrogen (step B) or silane (step C). However, in the
presence of equilibrium amounts of 1, any diphenylsi-
lane formed reacts instantly and irreversibly to give 3a
plus hydrogen (step D). This reaction consumes 1,
shifting the initial equilibrium (A) to the left. The
irreversibility of step D in this system is ultimately
responsible for the final product mixture of 3a and
1,1,2,2-tetraphenyldisilane. Hydrogen is also being elimi-
nated from 4a to give 2a (B), but in the presence of
hydrogen (generated by the reaction of 1 with diphen-
ylsilane; step D) this reaction is easily reversed, par-
ticularly at high temperatures.
Step iv in the catalytic cycle shown in Scheme 4, the
reductive elimination of a Si-Si bond, has not been

966 Organometallics, Vol. 18, No. 6, 1999
Rosenberg et al.
1H NMR spectra were recorded on Varian XL-300, Bruker
WP-200, Bruker WH-400, and Bruker AMX-500 spectrometers.
With d6-benzene as solvent the spectra were referenced to
C6D5H at 7.15 ppm and with d8-toluene as solvent the spectra
were referenced to the CD2H residual proton at 2.09 ppm. 31P-
{1H} NMR spectra were recorded at 121.4 MHz on the Varian
XL-300 or at 202.3 MHz on the Bruker AMX-500 and were
referenced to external P(OMe)3 at +141.0 ppm relative to 85%
H3PO4. 29Si NMR spectra were run at 59.6 MHz on the Varian
XL-300 and were referenced to external TMS at 0.0 ppm.
Sch em e 5
Elemental analyses were carried out by Mr. P. Borda of this
department. Gas chromatography/mass spectrometry was
carried out by Ms. L. Madilao of this department.
Syn t h eses of Com p lexes a n d R ea ct ivit y St u d ies.
[(d ip p e)Rh ]2(µ-H)(µ-η2-H-SiMeTolp ) (3c). This complex
was prepared by the same method as for 3a ,b,d (113 mg (0.154
mmol) of 1; 21 mg (0.154 mmol) of methyl-p-tolylsilane).16
Reddish brown crystals were obtained in 76% yield (102 mg).
1H NMR (C6D6, ppm): Hortho 7.93 (d, 2H, 3J H
) 7.3 Hz);
-H
m
o
Hmeta 7.13 (d, 2H); SiC6H5CH3 2.17 (s, 3H); CH(CH3)2 1.98
(mult, 8H); PCH2CH2P, SiCH3 1.26-1.36 (overlapping d and
s, 11H); CH(CH3)2 1.21 (dd, 12H, 3J P-H ) 14.5 Hz, 3J H-H ) 7.0
Hz); CH(CH3)2 1.12-0.94 (mult, 36H); Rh-H -6.00 (pt, 2H,
J P-H ) 20.6 Hz, 1J Rh-H ) 15.7 Hz). Note: For 3c in d8-toluene
the singlet resonance due to SiCH3 shifts to 1.24 ppm and the
PCH2CH2P resonance is seen as a doublet at 1.33 ppm (2J P-H
It is important to note that, for the reaction shown
in eq 9, the observed results require the formation of
an intermediate 4a from 1 and 1,1,2,2-tetraphenyldi-
silane and also require that its formation be reversible.
This implied reversibility is the key to the catalytic
activity of 4a : we propose that complex 4a represents
the final intermediate in the catalytic dehydrogenative
coupling of diphenylsilane to give 1,1,2,2-tetraphenyl-
disilane.
) 12.6 Hz). 31P{1H} NMR (C6D6, ppm): 94.3 (d mult, J Rh-P
)
159 Hz). 29Si NMR (C7D8, ppm): 157-166 (br mult). Anal. Calcd
for C36H76P4Rh2Si: C, 49.88; H, 8.84. Found: C, 50.20; H, 8.86.
[(d ip p e)Rh ]2(µ-SiP h 2)2 (2a ). This compound can be pre-
pared by two methods.
(1) To a red solution of [(dippe)Rh]2(µ-H)(µ-η2-H-SiPh2)
(3a ;16 275 mg, 0.300 mmol) in hexanes (5 mL) was added
dropwise a solution of diphenylsilane (55 mg, 0.300 mmol) in
hexanes (3 mL). The supernatant was decanted from the bright
yellow-orange precipitate that formed, and the precipitate was
washed with cold hexanes to remove any unreacted silyl
hydride complex. The yellow solid was then stirred in a large
volume of toluene (25-30 mL) for several days until most of
the powder had dissolved to give a bright orange solution. After
filtration of the solution through a Celite pad, orange crystals
were obtained from a minimum volume of toluene by cooling
to -40 °C. Yield: 58% (191 mg).
(2) A reaction identical with that described above was
carried out in toluene instead of hexanes. The dark red solution
of 3a changed to bright orange upon addition of the diphenyl-
silane. After filtration of the solution through a Celite pad,
orange crystals were obtained from a minimum volume of
toluene by cooling to -40 °C.
Exp er im en ta l Section
Gen er a l P r oced u r es a n d Rea gen t Syn th eses. All ma-
nipulations were performed under prepurified nitrogen in a
Vacuum Atmospheres HE-553-2 workstation equipped with
an MO-40-2H purification system or using Schlenk-type
glassware. The term “reactor bomb” refers to a cylindrical,
thick-walled Pyrex vessel equipped with a 5 mm (or 10 mm,
for larger bombs) Kontes Teflon needle valve and a ground-
glass joint for attachment to a vacuum line. NMR tubes used
for preparing sealed samples are 8 or 9 in. Wilmad 507 PP
NMR tubes with an outer b14 joint attached by glass-blowing.
Toluene and hexanes were predried over CaH2 and then
distilled from molten sodium and sodium-benzophenone ketyl,
respectively, under argon. Deuterated benzene (C6D6, 99.6
atom % D) and deuterated toluene (C7D8, 99.6 atom % D) were
purchased from MSD Isotopes and dried over 4 Å molecular
sieves. The dried deuterated solvents were then degassed using
three “freeze-pump-thaw” cycles and were vacuum-trans-
ferred before use. Ethylene and deuterium gas were purchased
from Matheson Gas Products, the deuterium being passed
through a glass coil immersed in liquid nitrogen to remove
any traces of water and oxygen. Hydrogen gas was purified
by being passed through a column packed with activated 4 Å
molecular sieves and MnO.
While the second method described is faster and more
straightforward, it tends to give lower yields and second crops
can be contaminated with any excess 3a left in solution. The
identity of the yellow powder formed from the first method
has not yet been established.
1H NMR (C6D6, ppm): Hortho 8.31 (dd, 4H, 3J H
o ) 7.2 Hz,
-H
m
4J H
) 1.3 Hz); Hmeta, Hpara 7.38-7.20 (overlapping mults,
-H
p
o
2
6H); CH(CH3)2 1.96 (mult, 8H); PCH2CH2P 1.30 (d, 8H, J P-H
) 15.2 Hz), CH(CH3)2, 0.97-0.74 (mult, 48H). 31P{1H} NMR
(C6D6, ppm): 79.3 (d mult, J Rh-P ) 134 Hz). Anal. Calcd for
Literature methods were used to prepare [(dippe)Rh]2(µ-H)2
(1).17,34 Ph2SiH2 and MePhSiH2 were purchased from the
Aldrich Chemical Co., dried by refluxing over calcium hydride
overnight, and distilled. The silanes Me2SiH2 and MeTolpSiH2
were prepared by LiAlH4 reduction of Me2SiCl2 (purchased
from Aldrich) and MeTolpSiCl2 (purchased from Petrarch
Systems) using slightly modified literature procedures.21,35,36
C
52H84P4Rh2Si2: C, 57.03; H, 7.73. Found: C, 56.79; H, 7.72.
Rea ction of [(d ip p e)Rh ]2(µ-SiP h 2)2 (2a ) w ith H2. This
reaction was carried out on a larger scale, as described
previously for the preparation of d2-3a from 2a and deuterium
gas,16 with diphenylsilane being identified by 1H NMR spec-
troscopy as the principal organosilicon product. The reaction
was also carried out in a sealed NMR tube and monitored over
several days: [(dippe)Rh]2(µ-SiPh2)2 (2a ; 25 mg, 0.023 mmol)
was dissolved in d8-toluene (0.6 mL) in a sealable NMR tube.
The solution was degassed by one freeze-pump-thaw cycle.
One atmosphere of hydrogen gas was introduced, and the
(34) Fryzuk, M. D.; McConville, D. H.; Rettig, S. J . J . Organomet.
Chem. 1993, 445, 245.
(35) Doyle, M. P.; DeBruyn, D. J .; Donnelly, S. J .; Kooistra, D. A.;
Odubela, A. A.; West, C. T.; Zonnebelt, S. M. J . Org. Chem. 1974, 39,
2740.
(36) Benkeser, R. A.; Landesman, H.; Foster, D. J . J . Am. Chem.
Soc. 1952, 74, 648.

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 967
sample was sealed. The sample was then frozen in liquid
nitrogen while being transported to the NMR spectrometer and
then thawed (the orange solution did not change in color) and
placed in the spectrometer. The initial 1H NMR spectrum
showed signals due to the starting bis(µ-silylene) species 2a ,
[[(dippe)Rh]2(µ-H)(µ-η2-H-SiPh2)]‚H2 (3a ‚H2),21 Ph2SiH2, and
an unknown species (Hortho at 8.15 ppm). The initial 31P{1H}
NMR spectrum showed signals due to 2a and [[(dippe)Rh]2-
(µ-H)(µ-η2-H-SiPh2)]‚H2 and also peaks due to a minor product
at 90 ppm (br d) and at 78.8 ppm (d). The spectra were run
again after the sample had sat at room temperature for 2 days.
The 31P{1H} NMR spectrum showed the presence of only two
products, 3a and an as-yet unidentified complex, 5, which is
thought to result from the reaction of 2a with the product of
disproportionation of diphenylsilane, PhSiH3 (see the following
reaction). This is further supported by the 1H NMR spectrum,
which shows the presence of both Ph2SiH2 and Ph3SiH, along
with a signal (overlapping with that due to dissolved hydrogen)
that might be due to [PhSiH2]2.
[(d ip p e)Rh ]2(µ-SiMeTolp )2 (2c). To a stirred, dark green
solution of [(dippe)Rh]2(µ-H)2 (1; 107 mg, 0.146 mmol) in
hexanes (5 mL) was added dropwise a solution of methyl-p-
tolylsilane (40 mg, 0.292 mmol) in hexanes (10 mL) to give a
bright orange solution. After filtration of the solution through
a Celite pad, orange crystals were obtained from a minimum
volume of hexanes by cooling to -40 °C. Yield: 71% (104 mg).
While both cis and trans isomers are formed from this reaction,
the less soluble trans isomer can be isolated by fractional
crystallization. This complex can also be prepared from the
addition of a single equivalent of methyl-p-tolylsilane to 3c.
1
3
trans-2c. H NMR (C6D6, ppm): Hortho 7.64 (d, 4H, J H
-H
o
m
) 7.5 Hz); Hmeta 7.03 (d, 4H), CH(CH3)2 2.50-2.29 (overlapping
mult, 8H); SiC6H5CH3 2.13 (s, 6H); SiCH3 1.73 (br s, 6H);
PCH2CH2P 1.63-1.45 (mult, 4H); CH(CH3)2 1.22 (dd, 12H,
3J H-P ) 18.3 Hz, 3J H-H ) 7.5 Hz); PCH2CH2P, CH(CH3)2 1.13-
3
0.94 (mult, 28H); CH(CH3)2 0.57 (dd, 12H, J H-P ) 15.3 Hz,
3J H-H ) 7.2 Hz). 31P{1H} NMR (C6D6, ppm): 82.7 (d mult, J P-Rh
) 161 Hz).
1
3
R ea ct ion of [(d ip p e)R h ]2(µ-SiP h 2)2 (2a ) w it h E xcess
P h 2SiH2. A solution of [(dippe)Rh]2(µ-SiPh2)2 (2a ; 10 mg,
0.0091 mmol) in d6-benzene was added to diphenylsilane (7
mg, 0.037 mmol). Over 1 day the solution color changed from
bright orange to yellow. 31P{1H} NMR spectroscopy showed
the presence of a single product, complex 5. Signals due to 5
cis-2c. H NMR (C7D8, ppm): Hortho 7.86 (d, 4H, J H
)
-H
o
m
7.8 Hz); Hmeta 7.11 (d, 4H); CH(CH3)2 2.37-1.92 (overlapping
mult, 8H); SiC6H5CH3 2.19 (s, 6H); PCH2CH2P, SiCH3 1.38-
1.27 (overlapping s and d, 14H); CH(CH3)2 1.15-0.85 (mult,
48H). 31P{1H} NMR (C6D6, ppm): 80.8 (d mult, J P-Rh ) 163
Hz).
1
as well as Ph2SiH2 and Ph3SiH were observed in the H NMR
Anal. Calcd for C44H84P4Rh2Si2: C, 52.90; H, 8.47. Found:
C, 52.66; H, 8.40.
spectrum, along with an unassigned signal at 5.93 ppm.
Complex 5 was never isolated free of silicon-containing
byproducts; therefore some of the peaks in the aromatic region
were buried under peaks due to the other products, and
relative integrals for the aromatic protons were unobtainable.
Relative integrals and peak assignments are tentative. 1H
NMR (C6D6, ppm): Si-H 8.56 (br s, 2H, w1/2 ) 30 Hz); Hortho
Con ver sion of P u r e tr a n s-2c to cis/tr a n s Mixtu r es. A
solution of trans-[(dippe)Rh]2(µ-SiMeTolp)2 (2c; 23 mg, 0.023
mmol) in d8-toluene (0.6 mL) was placed in an NMR tube with
a Teflon screw cap. The tube was closed and heated to the
appropriate temperature (70, 90, 110, or 130 °C) in an oil bath.
The tube was removed from the oil bath once every 30 min
and was placed in an ice bath to quench the equilibration
8.27 (d, 3J H
) 6.9 Hz); Hortho 8.04 (dd, 3J H
) 7.6 Hz,
-H
-H
m
o
m
o
3J H o ) 1.6 Hz); Hmeta 7.44 (mult); CH(CH3)2 2.54 (mult, 4H);
1
-H
reaction. A H NMR spectrum of the sample was run before
p
3
CH(CH3)2 2.11 (mult, 4H); CH(CH3)2, 1.65 (dd, 12H, J P-H
)
returning the sample to the oil bath. Appearance of the cis
isomer in solution was monitored by integration of the Hortho
signals at 7.64 ppm (trans) and 7.86 ppm (cis). The samples
were heated over 8-10 h periods.
3
13.5 Hz, J H-H ) 7.2 Hz); PCH2CH2P, CH(CH3)2 1.52-0.49
(overlapping mult, 32H); CH(CH3)2, 0.25 (dd, 12H, 3J P-H ) 15.6
3
Hz, J H-H ) 7.2 Hz); Rh-H -4.46 (br s, 2H, w1/2 ) 55 Hz);
[(d ip p e)R h (H )]2(µ-η2-H -SiMe2)2 (4d ). Dimethylsilane
(0.840 mmol, 2.1 equiv, 271 mmHg in a 57.5 mL constant-
volume bulb) was vacuum-transferred to a dark green solution
of [(dippe)Rh]2(µ-H)2 (1; 293 mg, 0.400 mmol) in frozen hexanes
(5 mL) at -196 °C. The solution was warmed slightly below
room temperature (ice-water bath), by which time the solution
had changed to a pale, golden yellow color. Yellow crystals
were obtained from a minimum volume of hexanes by cooling
to -40 °C. Yield: 62% (211 mg). The product is obtained in
purest form if a minimum of hexanes is used from the start of
the reaction and if the solution is kept cool throughout. If
solvent has to be removed from the product solution to induce
crystallization, there is normally some decomposition of the
product to [(dippe)Rh]2(µ-H)(µ-η2-H-SiMe2) (3d )16 and [(dippe)-
Rh]2(µ-SiMe2)2 (2d ); i.e.. loss of 1 equiv of silane or 1 equiv of
hydrogen. This same decomposition occurs slowly if solutions
of 4d are allowed to stand at room temperature for periods of
time longer than 5-10 min. Complex 4d can also be prepared
by the addition by vacuum transfer of 1 equiv of dimethylsilane
to a solution of 3d in hexanes, with similar yields. 1H NMR
Rh-H -6.79 (br s, 1H, w1/2 ) 60 Hz); Rh-H -13.05 (br s, 1H,
w1/2 ) 25 Hz); Rh-H -13.48 (br s, 1H, w1/2 ) 25 Hz). 31P{1H}
NMR (C6D6, ppm): 98.1 (d mult, 1P); 85.1 (d mult, 1P); 82.9
(d mult, 1P); 72.4 (mult, 1P).
[(d ip p e)Rh ]2(µ-SiMeP h )2 (2b). This complex can be pre-
pared by the same methods as for 2a (96 mg (0.113 mmol) of
3d ; 14 mg (0.113 mmol) of MePhSiH2; yield 65% (72 mg)).
Unfortunately, even when the reaction is carried out in toluene
the product tends to precipitate quickly, giving a bright orange
powder which has low solubility in most common solvents.
However, slow evaporation of a very dilute hexanes solution
did eventually yield crystals suitable for an X-ray crystal-
lographic study. This complex has two geometric isomers, cis
and trans, which could not be completely separated by crystal-
lization, despite the fact that the trans isomer is less soluble
than the cis isomer.
trans-2b. 1H NMR (C6D6, ppm): Hortho 7.71 (dd, 4H, 3J H
-H
m
o
4
) 6.6 Hz, J H
) 1.5 Hz); Hmeta 7.17 (mult, 4H); Hpara 7.08
-H
p
o
(mult, 2H); CH(CH3)2 2.38 (mult, 8H); SiCH3 1.71 (s, 6H);
PCH2CH2P 1.64-1.47 (mult, 4H); CH(CH3)2 1.19 (dd, 12H,
3J P-H ) 17.1 Hz, 3J H-H ) 7.2 Hz); PCH2CH2P, CH(CH3)2 1.14-
3
(C7D8, ppm): CH(CH3)2 1.91 (mult, 8H, J H-H ) 7.0 Hz);
3
PCH2CH2P 1.25 (d, 8H, 2J P-H ) 10.4 Hz); CH(CH3)2 1.16-1.02
(two dd); Si(CH3)2 0.88 (s, 12H); Rh-H -11.59 (mult (second-
order pattern), 4H). Note: In d6-benzene the Si(CH3)2 reso-
nance is seen as two singlets at 0.94 and 0.98 ppm; it is not
known why the two inequivalent signals arise. 31P {1H} NMR
(C6D6, ppm): 77.3 (br d, J Rh-P ) 118 Hz). Anal. Calcd for
0.92 (mult, 28H); CH(CH3)2 0.53 (dd, 12H, J P-H ) 15.6 Hz,
3J H-H ) 7.3 Hz). 31P{1H} NMR (C6D6, ppm): 82.8 (d mult,
1J Rh-P ) 162 Hz).
1
3
cis-2b. H NMR (C6D6, ppm): Hortho 8.03 (dd, 4H, J H
)
-H
m
o
7.8 Hz, 4J H
) 1.2 Hz); Hmeta 7.33 (t, 4H, J avg ) 7.2 Hz);
-H
p
o
Hpara 7.19 (mult, 2H); CH(CH3)2 2.27 (br s, 4H); CH(CH3)2 2.06
(br s, 4H); SiCH3, PCH2CH2P 1.43-1.30 (overlapping s and d,
14H); CH(CH3)2 1.17-0.91 (mult, 48H). 31P{1H} NMR (C6D6,
C
32H80P4Rh2Si2: C, 45.17; H, 9.48. Found: C, 45.30; H, 9.60.
Th er m a l Decom p osition of [(d ip p e)Rh (H)]2(µ-η2-H-
1
ppm): 80.9 (d mult, J Rh-P ) 163 Hz).
SiMe2)2 (4d ). Approximately 25 mg (0.029 mmol) of [(dippe)-
Rh(H)]2(µ-η2-H-SiMe2)2 (4d ) was dissolved in d8-toluene in an
NMR tube equipped with a Teflon screw cap/gas inlet adapter.
Anal. Calcd for C42H80P4Rh2Si2‚0.3C7H8: C, 52.85; H, 8.29.
Found: C, 52.84; H, 8.30.

968 Organometallics, Vol. 18, No. 6, 1999
Rosenberg et al.
An initial 31P{1H} NMR spectrum was run to establish the
purity of the sample; then the sample was attached to a
nitrogen manifold. The sample was heated in an oil bath at
45 °C while open to the nitrogen manifold and bubbler, to allow
any volatile products to escape the NMR tube. The progress
of the reaction was checked periodically by 31P{1H} NMR
spectroscopy. After 4 h a substantial amount of [(dippe)Rh]2-
(µ-H)(µ-η2-H-SiMe2) (3d ) had formed, with 23% conversion of
4d to 3d . After 24 h all of 4d had disappeared, giving 3d as
the principal product, along with some decomposition products
due to traces of air and water in the system.
using 1 as catalyst; vide infra) (19 mg, 0.051 mmol) in toluene
(3 mL). Initially there was no color change. The mixture was
heated overnight at 70 °C, during which time the solution color
changed to red. The toluene was removed under vacuum, and
d6-benzene was added to the residues to make up an NMR
sample. The 31P{1H} NMR spectrum showed the principal
product to be [(dippe)Rh]2(µ-H)(µ-η2-H-SiPh2) (3a ; 76%).
Signals due to small amounts of [(dippe)Rh]2(µ-SiPh2)2 (2a ;
13%) and the unknown complex 5 (11%) were also observed.
The 1H NMR spectrum showed the principal silicon-containing
product to be 1,1,2,2-tetraphenyldisilane.
Rea ction of [(d ip p e)Rh (H)]2(µ-η2-H-SiMe2)2 (4d ) w ith
H2. Approximately 25 mg (0.029 mmol) of [(dippe)Rh(H)]2(µ-
η2-H-SiMe2)2 (4d ) was dissolved in d8-toluene in a sealable
NMR tube. The tube was attached to a vacuum line by a needle
valve adapter, and slightly less than 1 atm of hydrogen was
introduced. No color change was observed. The 1H and 31P-
{1H} NMR spectra of the sealed sample revealed that no
reaction with the hydrogen had occurred. In the 1H NMR
spectrum the hydride resonance at -11.59 ppm had lost
resolution, which might indicate that hydride exchange with
hydrogen in solution was occurring.
Note: Complex 2a , [(dippe)Rh]2(µ-SiPh2)2, does not react with
Ph2SiH-HSiPh2.
Ca ta lytic Dim er iza tion of P h 2SiH2. A typical reaction
procedure is as follows: A solution of diphenylsilane (254 mg,
1.38 mmol) in toluene (1.25 mL) was placed in a 50 mL round-
bottom flask equipped with a stirbar and condenser. The
catalyst (1; 0.25 mL of a 0.055 M solution in toluene, 0.014
mmol) was added to the substrate solution by syringe under
a strong flow of nitrogen. The solution was degassed by
evacuation of the headspace and then heated to 60 °C under
nitrogen, open to a bubbler. After 18-24 h the mixture was
taken into a glovebox, where the catalyst was removed on a
Florisil column. The product mixture was analyzed by 1H NMR
spectroscopy and by GC-MS. 1H NMR spectra showed the
disilane [Ph2SiH]2 to be the principal product, with yields as
high as 78%. Also seen in the spectra were small signals due
to unreacted Ph2SiH2 and to Ph3SiH. A small singlet at 4.85
ppm was due to a minor, unidentified product. The GC-MS
analysis confirmed the presence of [Ph2SiH]2, Ph2SiH2, Ph3-
SiH, and PhSiH3 in these mixtures, but no other peak was
observed which might be due to the unidentified product
observed in the 1H NMR spectra. Parent ions (M+ (m/e))
observed for these products in the mass spectra are 107
(PhSiH3), 184 (Ph2SiH2), 260 (Ph3SiH), and 366 ([PhSiH2]2).
Because the solvent (toluene) was removed from the reaction
mixture under vacuum, no PhSiH3 (bp 120 °C) was detected
in the 1H NMR spectra. The disilane [Ph2SiH]2 could be
separated from these reaction mixtures by recrystallization
from hexanes.
Rea ction of [(d ip p e)Rh (H)]2(µ-η2-H-SiMe2)2 (4d ) w ith
D2. This reaction was carried out in the same manner as the
preceding reaction, except that the sealable NMR tube con-
taining solid 4d was cooled in an ice bath before the addition
of the d8-toluene and the solution was kept cool while the
deuterium gas was added and the tube was sealed. The initial
31P{1H} and 1H NMR spectra showed no reaction with the
deuterium. The sample was monitored by NMR spectroscopy
over a 2-week period, after which time approximately 50% of
the hydride in the sample had been exchanged for deuteride,
as determined from integration of the 1H NMR spectrum.
[(d ip p e)Rh ]2(µ-SiMe2)2 (2d ). This complex is formed by
dissociation of 2 equiv of H2 from 4d if it is left in solution at
room temperature or is heated. However, it is difficult to get
complete conversion to 2d by this method; the product is often
mixed with leftover 4d and with 3d formed by a concurrent
dissociation of dimethylsilane. A reaction which gives pure 2d
as product is that of 4d with an excess of ethylene. Solid
[(dippe)Rh(H)]2(µ-η2-H-SiMe2)2 (4d ; 58 mg, 0.068 mmol) was
weighed into a thick-walled reactor bomb. The bomb was
cooled to 0 °C in an ice bath, and under a strong flow of
nitrogen, toluene (5 mL) was added, giving a pale, golden
yellow solution. The headspace above the solution was evacu-
ated, and 1 atm of ethylene was introduced. The solution was
stirred at 0 °C for 15 min, during which time the solution color
changed to orange-red. Orange-red crystals were obtained from
a minimum volume of toluene by cooling to -40 °C. Yield: 68%
(39 mg). 1H NMR (C6D6, ppm): CH(CH3)2 2.18 (mult, 8H);
PCH2CH2P 1.36 (d, 8H, 2J P-H ) 15.3 Hz); Si(CH3)2 1.14 (s, 12H)
overlapping with CH(CH3)2 1.18-1.02 (overlapping mult, 48H).
Note: For 2d in d8-toluene the Si(CH3)2 resonance is seen at
1.04 ppm, a solvent-related, upfield shift of 0.10 ppm. 31P{1H}
NMR (C6D6, ppm): 83.3 (d mult, 1J Rh-P ) 167 Hz). Anal. Calcd
for C32H80P4Rh2Si2: C, 45.17; H, 9.48. Found: C, 45.30; H, 9.60.
Rea ction of [(d ip p e)Rh ]2(µ-SiMe2)2 (2d ) w ith H2. Ap-
proximately 25 mg (0.030 mmol) of [(dippe)Rh]2(µ-SiMe2)2 (2d )
was dissolved in d8-toluene in a sealable NMR tube. The tube
was attached to a vacuum line by a needle valve adapter, and
slightly less than 1 atm of hydrogen was introduced. The tube
was tapped for 5-10 min to encourage diffusion of the gas into
the orange-red solution, which changed to a pale yellow color.
The tube was sealed, and NMR spectra (1H and 31P{1H}) of
the sample were run. The product of the reaction was [(dippe)-
Rh(H)]2(µ-η2-H-SiMe2)2 (4d ), as determined by comparison
with NMR spectra of an authentic sample.
The dimerization reaction does proceed at room tempera-
ture, but with lower yields of disilane.
P r ogr ess of Rea ction Mon itor ed by 1H NMR. In the
glovebox, a solution of diphenylsilane (126 mg, 0.68 mmol) in
d6-benzene (3 mL) was placed in a 250 mL thick-walled reactor
bomb, with a small stirbar. The bomb was taken out of the
glovebox and attached to a vacuum line. The catalyst solution
was also prepared in the glovebox: [(dippe)Rh]2(µ-H)2 (1; 10
mg, 0.014 mmol) in d6-benzene (1 mL; a 50:1 substrate-to-
catalyst ratio). The dark green catalyst solution was added to
the substrate by syringe under a flow of nitrogen. The mixture
immediately changed from clear to orange and then within
30 s lightened to yellow. The bomb was left open to the
nitrogen manifold and bubbler throughout the experiment.
Samples were withdrawn periodically by syringe and analyzed
by 1H NMR spectroscopy, the first being taken 5 min after
addition of the catalyst to the substrate and the final being
taken 3 h later. The final 1H NMR spectrum showed a 19%
yield of [Ph2SiH]2.
Ca lcu la tion of ∆Gq for th e P h osp h in e a n d Hyd r id e
Exch a n ge in 4d . ∆Gq was calculated using the value for the
rate constant37 kC (where kC ) π∆υC/(2)1/2) in the Eyring
equation
∆Gq ) -RTC ln[(kCh)/(kBTC)]
where R ) gas constant, TC ) temperature of coalescence, ∆υC
) peak separation at the low-T limit, h ) Planck’s constant,
and kB ) Boltzmann constant. For the phosphorus resonances
R ea ct ion of [(d ip p e)R h ]2(µ-H )2 (1) w it h P h 2SiH -
HSiP h 2. To a dark green solution of [(dippe)Rh]2(µ-H)2 (1; 37
mg, 0.051 mmol) in toluene (5 mL) in a small reactor bomb
was added 1,1,2,2-tetraphenyldisilane (prepared from Ph2SiH2
(37) Thomas, W. A. Annu. Rev. NMR Spectrosc. 1968, 1, 43.

Rh µ-Silylene and µ-η2-Silyl Complexes
Organometallics, Vol. 18, No. 6, 1999 969
in the variable-temperature 31P{1H} NMR spectra of 4d , Tc )
244 K (-29 °C) and ∆υc ) 1590 Hz. For the hydride resonances
parameters, and all remaining hydrogen atoms were fixed in
calculated positions with C-H ) 0.98 Å and BH ) 1.2Bbonded
atom. Secondary extinction corrections were applied for trans-
2b and 4d (Zachariasen type, isotropic), the final values of the
1
in the variable-temperature H NMR spectra of 4d , Tc ) 234
K (-39 °C) and ∆υc ) 705 Hz. The coalescence temperatures
were estimated visually from the spectra and have an error
of approximately (5 K.
extinction coefficients being [9.2(2)] × 10-7 and [4.4(1)] × 10-8
,
respectively. No extinction correction was necessary for 2a .
Neutral atom scattering factors and anomalous dispersion
corrections were taken from ref 39. A parallel refinement of
the mirror image of 4d gave substantially higher residuals (R
) 0.028 and Rw ) 0.030), thus establishing the absolute
configuration for the particular crystal used for data collection.
Selected bond lengths and bond angles for 2a , trans-2b, and
4d appear in Tables 2-4, respectively. Tables of final atomic
coordinates and equivalent isotropic thermal parameters,
anisotropic thermal parameters, bond lengths and angles,
torsion angles, intermolecular contacts, and least-squares
planes are included as Supporting Information.
X-r a y Cr ysta llogr a p h ic An a lyses of [(d ip p e)Rh ]2(µ-
SiRR′)2 (R ) R′ ) P h , 2a ; R ) P h , R′ ) Me, 2b) a n d
[(d ip p e)Rh (H)]2(µ-η2-H-SiMe2)2 (4d ). Crystallographic data
appear in Table 1. The final unit-cell parameters were obtained
by least squares on the setting angles for 25 reflections with
2θ ) 10.9-24.6, 26.5-29.4, and 35.2-40.6°, respectively, for
2a , trans-2b, and 4d . The intensities of 3 standard reflections,
measured every 200 reflections throughout the data collection,
decayed linearly by 6.4% for 2a and showed only small random
fluctuations for trans-2b and 4d . The data were processed38
and corrected for Lorentz and polarization effects, decay (for
2a ), and absorption (empirical, based on azimuthal scans).
The structure analysis of 2a was initiated in the centrosym-
metric space group C2/c and that of trans-2b in the centrosym-
metric space group P1h on the basis of the E statistics, these
choices being confirmed by subsequent calculations. The
structures were all solved by the Patterson method. Molecules
of 2a and trans-2b have exact (crystallographic) C2 and Ci
symmetry, respectively. The C(6) isopropyl group of 2a dis-
plays high thermal motion and is probably disordered, but no
attempts were made to refine a disordered model. Geometrical
parameters for this moiety consequently deviate from expected
values. One isopropyl carbon atom of 4d (C(16)) was modeled
as (68:32) 2-fold disordered. All non-hydrogen atoms were
refined with anisotropic thermal parameters. The metal-bound
hydrogen atoms in 4d were refined with isotropic thermal
Ack n ow led gm en t. Financial support for this work
was provided by the NSERC (operating grants to M.D.F.
and a postgraduate scholarship to L.R.). We thank
J ohnson-Matthey for a generous loan of RhCl3‚3H2O.
Su p p or tin g In for m a tion Ava ila ble: Complete tables of
non-hydrogen and rhodium hydride parameters, bond lengths
and angles, calculated hydrogen atom parameters, anisotropic
thermal parameters, torsion angles, and intermolecular con-
tacts and least-squares planes for 2a ,b and 4d . This material
is available free of charge via the Internet at http://pubs.acs.org.
OM980651S
(39) (a) International Tables for X-ray Crystallography; Kynoch
Press: Birmingham, U.K. (present distributor Kluwer Academic:
Boston, MA), 1974; Vol. IV, pp 99-102. (b) International Tables for
Crystallography; Kluwer Academic: Boston, MA, 1992; Vol. C, pp 200-
206.
(38) TEXSAN: Structure Analysis Package; Molecular Structure
Corp., The Woodlands, TX, 1992.