Conversion of perfluorobenzyl complexes of rhodium to fluorinated oxarhodacycles

Article abstract of DOI:10.1021/om0008171

The perfluorobenzylrhodium complex [Cp*Rh(CO)(CF₂C₆F₅)(I)], 1 (Cp* = C₅Me₅), reacts with N-methylmorpholine-N-oxide (NMO) with loss of CO₂ and formation of the iodo-bridged dimer [Cp*Rh(μ₂-I)(CF₂C₆F₅)] ₂, 5, which has been characterized by X-ray crystallography. While attempts to prepare hydroxo complexes by reaction of [Cp*Rh(PMe₃)(CF₂C₆F₅)(I)], 3, with sources of hydroxide were unsuccessful, treatment of 3 with moist silver oxide affords the oxametallacycle [Cp*(PMe₃)RhO(C₆F₄)CF₂, 9, which undergoes rapid hydrolysis of the α-CF₂ group by adventitious moisture to afford the crystallographically characterized analogue [Cp*(PMe₃)RhO(C₆F₄)C=O], 10. A brief discussion of possible mechanisms is presented.

Full text of DOI:10.1021/om0008171

Organometallics 2001, 20, 363-366
363
Con ver sion of P er flu or oben zyl Com p lexes of Rh od iu m to
F lu or in a ted Oxa r h od a cycles
Russell P. Hughes,*,† Danielle C. Lindner,† Louise M. Liable-Sands,‡ and
Arnold L. Rheingold‡
Departments of Chemistry, 6128 Burke Laboratory, Dartmouth College,
Hanover, New Hampshire 03755, and University of Delaware, Newark, Delaware 19716
Received September 21, 2000
Ta ble 1. Cr ysta l Da ta a n d Su m m a r y of X-r a y Da ta
Collection
Summary: The perfluorobenzylrhodium complex [Cp*Rh-
(CO)(CF2C6F5)(I)], 1 (Cp* ) C5Me5), reacts with N-
methylmorpholine-N-oxide (NMO) with loss of CO2 and
formation of the iodo-bridged dimer [Cp*Rh(µ2-I)-
(CF2C6F5)]2, 5, which has been characterized by X-ray
crystallography. While attempts to prepare hydroxo
complexes by reaction of [Cp*Rh(PMe3)(CF2C6F5)(I)], 3,
with sources of hydroxide were unsuccessful, treatment
of 3 with moist silver oxide affords the oxametallacycle
[Cp*(PMe3)RhO(C6F4)CF2], 9, which undergoes rapid
hydrolysis of the R-CF2 group by adventitious moisture
to afford the crystallographically characterized analogue
[Cp*(PMe3)RhO(C6F4)CdO], 10. A brief discussion of
possible mechanisms is presented.
5
10
formula
fw
space group
a, Å
b, Å
c, Å
C
34H30F14I2Rh2
C20H24F4O2PRh
506.27
1164.20
P2(1)/c
15.966(2)
11.330(2)
20.031(3)
90
90.19(2)
90
3623.7(7)
4
P2(1)/n
8.6399(6)
21.296(2)
12.1241(9)
90
108.286(9)
90
2118.1(2)
4
R, deg
â, deg
γ, deg
V, Å3
Z
D(calcd), g/cm3
abs coeff, mm-1
temp, K
diffractometer
radiation
R(F), %a
R(wF2), %a
2.134
2.712
1.588
0.928
298(2)
We have previously described the synthesis of the
perfluorobenzyl rhodium complexes 1-31 and the reac-
tion of 2 to afford the cationic water complex 4.2 The
R-CF2 group in 4 proved to be remarkably sensitive
toward hydrolysis by the coordinated water ligand, and
the role of the counterion in these hydrolysis reactions
was shown to affect the rate of hydrolysis.2 Here we
report the structure of a dimer prepared from decarbo-
nylation reaction of 1 and attempts to prepare hydroxo
analogues of 4, which lead to the unexpected formation
of oxametallacycles.
Siemens P4
Mo KR 0.71073 Å
3.29
4.23
8.39
7.79
Quantity minimized ) R(wF2) ) ∑[w(Fo2 - Fc2)2]/∑[(wFo2)2]1/2
R ) ∑∆/∑(Fo), ∆ ) |(Fo - Fc)|.
;
a
Treatment of a methylene chloride solution of 1 with
N-methylmorpholine-N-oxide (NMO) resulted in the
expected loss of the CO ligand, presumably as CO2, and
clean formation (as monitored by NMR spectroscopy)
of the stable iodide-bridged dimer 5. The dimer was
1
characterized in solution by H and 19F NMR spectros-
copy and in the solid state by X-ray diffraction and
elemental analysis. Crystallographic data can be found
in Table 1. The ORTEP plot with atom-labeling scheme
is depicted in Figure 1. The pairs of pentamethylcyclo-
pentadienyl ligands and perfluorobenzyl ligands are
mutually trans, as might be expected on the basis of
steric effects. A major difference between the two sides
of the dimer is the conformation of each of the perfluo-
robenzyl ligands, with one oriented with its perfluo-
rophenyl ring tilted in toward the bridging ligands and
the other tilted away from the core. The interligand
bond angles show a constricted angle between I(1)-
Rh(1)-I(2) [82.27(4)°] and I(1)-Rh(2)-I(2) [81.55(4)°]
F igu r e 1. ORTEP drawing and atom-labeling scheme for
5. Ellipsoids are drawn at 35% probability except for those
of the phenyl rings (see Experimental Section). Hydrogens
are omitted for clarity. Selected distances (Å) and angles
(deg): Rh(1)-C(11) 2.079(13), Rh(2)-C(28) 2.214(10), Rh-
(1)-I(1) 2.7186(14), Rh(1)-I(2) 2.7322(13), Rh(2)-I(1)
2.7536(13), Rh(2)-I(2) 2.7368(14), C(11)-F(1) 1.43(2), C(11)-
F(2) 1.349(15), C(28)-F(8) 1.355(12), C(28)-F(9) 1.189-
(11): F(1)-C(11)-F(2) 101.9(11), F(8)-C(28)-F(9) 107.6-
(10), Rh(1)-C(11)-C(12) 125.6(9), Rh(2)-C(28)-C(29)
119.3(7).
† Dartmouth College.
‡ University of Delaware.
as a result of the four-membered ring core. As expected,
slightly more obtuse angles are observed between the
nonbridging ligands, C(11)-Rh(1)-I(1) [95.5(4)°], C(11)-
Rh(1)-I(2) [94.2(4)°], C(28)-Rh(2)-I(1) [92.7(2)°], and
(1) Hughes, R. P.; Lindner, D. C.; Rheingold, A. L.; Yap, G. P. A.
Organometallics 1996, 15, 5678-5686.
(2) Hughes, R. P.; Lindner, D. C.; Rheingold, A. L.; Liable-Sands,
L. M. J . Am. Chem. Soc. 1997, 119, 11544-11545.
10.1021/om0008171 CCC: $20.00 © 2001 American Chemical Society
Publication on Web 12/14/2000

364 Organometallics, Vol. 20, No. 2, 2001
Notes
C(28)-Rh(2)-I(2) [89.0(3)°]. Interestingly, the bond
lengths associated with the two perfluorobenzyl groups
are considerably different. For example, the distance
between Rh(1)-C(11) [2.079(13) Å] is significantly
shorter than that for Rh(2)-C(28) [2.214(10) Å]. Com-
parison values are 2.190(11) Å in 3,1 2.152(14) Å in the
iridium analogue of 1,3 and 2.113(9) Å in the cation 4.2
The R-C-F distances in 5 are also different. One set
exhibits a rather long C-F bond [C(11)-F(1) ) 1.43(2)
Å] and one normal [C(11)-F(2) ) 1.349(15) Å], whereas
the other set has one unusually short distance [C(28)-
F(9) ) 1.189(11) Å] and one normal [C(28)-F(8) )
1.355(12) Å]. The F-C-F bond angle which contains
the longest C-F bond is consistent with greater p-
character in that bond, as the angle is significantly more
acute than that containing the unusually short C-F
bond [F(1)-C(11)-F(2) ) 101.9(11)° vs F(8)-C(28)-F(9)
) 107.6(10)°]. Comparison values are 1.257(14) and
1.327(12) Å in 3 with an F-C-F angle of 105.3(9)°,1
1.375(16) and 1.319(17) Å with F-C-F of 105.6(12)° in
the iridium analogue of 1,3 and 1.388(10) and 1.412(10)
Å with F-C-F of 101.8(6)° in 4.2 The correspondingly
obtuse Rh-CF2-C(ipso) angles [Rh(1)-C(11)-C(12) )
125.6(9)° and Rh(2)-C(28)-C(29) ) 119.3(7)°] appro-
priately reflect the degree of narrowing in the F-C-F
angle; the smaller the F-C-F bond angle, the greater
its corresponding Rh-CF2-C(ipso) angle. On the other
hand, the CF2-C(ipso) bond lengths are the same
[1.508(15) Å for C(11)-C(12) and 1.499(11) Å for C(28)-
C(29)]. Once again comparison values can be found in
the corresponding bond angles [and C-C distances] of
117.2(7)° [1.486(14) Å] in 3,1 116.4(10)° [1.528(17) Å] in
the iridium analogue of 1,3 and 116.0(5)° [1.542(9) Å]
in 4.2 Acute F-C-F angles and obtuse M-CF2RF angles
are not unusual,3 but the perfluorobenzyl ligand appears
to be capable of accommodating quite a wide range of
distances and angles internally and between itself and
the metal.
The data obtained from solution are consistent with
the solid-state structure of 5, but the presence of
symmetry equivalent benzylic fluorine substituents
requires that the perfluorobenzyl groups rotate freely
about their respective Rh-CF2 bonds, each sampling an
effective plane of symmetry, which bisects the two
geminal fluorine atoms. In addition, the triplet coupling
observed for the CF2 resonance (J FF ) 29 Hz) suggests
that the pentafluorophenyl ring rotates freely about the
CF2-C6F5 bond, affording an equal coupling with both
ortho fluorine atoms. The 2:1:2 ratio of peaks upfield of
the benzylic fluorines is also consistent with free rota-
tion about the CF2-C6F5 bond. Similar observations
have been made for other perfluorobenzyl complexes.1-3
A similar complex was reported by Stone and co-
workers.4 They found that reaction of [Rh(η5-C5H5)-
(C2H4)2] and n-C3F7I affords the oxidative addition
product [Rh(η5-C5H5)(C2H4)(C3F7)I], which if heated to
50 °C, loses ethylene to give a halide dimer 6, analogous
to 5. Our attempts to mimic this reaction by using [Rh-
(η5-C5Me5)(C2H4)2] and perfluorobenzyl iodide were
unsuccessful. The lack of reactivity in the pentameth-
ylcyclopentadienyl case may be attributable to steric
blocking of the metal center by the Cp* ring, which has
a greater cone angle than Cp (188° for Cp* vs 148° for
Cp).5,6
We were interested in the preparation of hydroxo
complex 7 because, in theory, it could be treated with a
variety of proton sources to generate the aqua complex
4, thereby allowing a more controlled entry to studying
the rate of subsequent hydrolysis (vide supra) in the
presence of various counterions.2 Complexes containing
hydroxo ligands are not common among late transition
metals, presumably because of their tendency to donate
a proton to another M-OH complex, resulting in the
formation of bridging oxo complexes, which are often
very stable.7 Bergman and co-workers have been suc-
cessful in obtaining and characterizing some late transi-
tion metal hydroxo complexes,8 including the iridium
complex 8,9 which allowed us hope that synthesis of 7
might be an achievable goal.
Attempts to effect the substitution of iodide in 2 using
NaOH or CsOH in THF and water returned only
starting material. On reflection, moist silver oxide was
then considered as a potentially good reagent, combining
a source of hydroxide with Ag+ as a well-known10 iodide
abstracting agent in these systems. However, when 2
was treated with freshly prepared moist silver oxide,
the desired hydroxo complex 7 was not observed, and a
different product 9 was formed in quantitative yield.
(5) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(6) Coville, N. J .; Loonat, M. S.; White, D.; Carlton, L. Organome-
tallics 1992, 11, 1082-1090.
(7) Sharp, P. R. J . Chem. Soc., Dalton Trans. 2000, 2647-2657.
(8) Bergman, R. G. Polyhedron 1995, 14, 3227-3237.
(9) Woerpel, K. A.; Bergman, R. G. J . Am. Chem. Soc. 1993, 115,
7888-7889.
(10) Hughes, R. P.; Husebo, T. L.; Maddock, S. M.; Rheingold, A.
L.; Guzei, I. A. J . Am. Chem. Soc. 1997, 119, 10231-10232.
(3) Hughes, R. P.; Smith, J . M.; Liable-Sands, L. M.; Concolino, T.
E.; Lam, K.-C.; Incarvito, C.; Rheingold, A. L. J . Chem. Soc., Dalton
Trans. 2000, 873-879.
(4) Stone, F. G. A.; Mukhedkar, A. J .; Mukhedkar, V. A.; Green, M.
J . Chem. Soc. A 1970, 3158-3161.

Notes
Organometallics, Vol. 20, No. 2, 2001 365
Sch em e 1
F igu r e 2. ORTEP drawing and atom-labeling scheme for
5. Ellipsoids are drawn at 35% probability. Hydrogens are
omitted for clarity. Selected distances (Å) and angles
(deg): Rh-C(14) 2.010(4), Rh-O(1) 2.103(3), Rh-P 2.2728-
(13), C(14)-O(2) 1.203(5), C(14)-C(15) 1.472(6), C(15)-
C(16) 1.412(6), C(16)-O(1) 1.326(5): C(14)-Rh-O(1) 82.61-
(15), Rh-O(1)-C(16) 110.4(3), Rh-C(14)-O(2) 124.8(3),
O(2)-C(14)-C(15) 124.2(4), P-Rh-O(1) 90.05(9), P-Rh-
C(14) 86.81(14).
Complex 9 was characterized in solution by H, 19F,
and 31P{1H} NMR spectroscopy. The 1H NMR spectrum
showed the expected phosphorus coupled doublets for
Cp* and PMe3. The 31P{1H} NMR spectrum appeared
as a doublet of doublets of doublets, with coupling to
rhodium (J PRh ) 162 Hz) and each of the R-CF2 fluorines
(J PF ) 40, 27 Hz). In C6D6 the 19F NMR spectrum of 9
was initially puzzling, as the CF2 resonances are ac-
cidentally isochronous, but on changing the NMR
solvent to acetone-d6, the chemical shifts of the two
geminal fluorine atoms appeared as the expected strongly
coupled AB pattern with J FF ) 234 Hz. One CF2 fluorine
exhibited additional splitting to phosphorus and rhod-
ium (J FP ) 40, J FRh ) 12) with an additional coupling
to the ortho fluorine on the aromatic ring (J FF ) 12),
while the second CF2 fluorine exhibited a smaller coup-
ling to phosphorus (J PF ) 27 Hz) and similar coupling
(8 Hz) to rhodium and to the ortho fluorine. Only four
fluorine peaks of equal intensity were observed corre-
sponding to the substituents on the aryl ring.
1
Isolation of an adequate sample of 9 for elemental
analysis was hampered by an unusually facile hydroly-
sis during recrystallization to give 10. Formation of 10
results in the expected disappearance of the CF2 reso-
nances in the 19F NMR spectrum and appearance in its
IR spectrum of an additional band at ν 1658 cm-1
corresponding to the CdO stretch of the newly formed
acyl. X-ray quality crystals were obtained of 10 for
confirmation of the overall structure. Crystallographic
data are presented in Table 1. An ORTEP plot with
atom-labeling scheme is found in Figure 2. As expected,
the distance from the metal to the sp2-acyl carbon (Rh-
C(14) ) 2.011(4) Å) is shorter than that to the sp3 carbon
in 4 [2.113(9) Å] and is typical of a Rh-C(sp2) distance
[1.982 Å].11 The Rh-O(1) [2.103(3) Å] distance is in the
range for a typical rhodium-oxygen bond,11 but is
significantly shorter than that in 4 [2.164(7)].
Two possible mechanisms for the formation of the
oxygen-carbon bond in 9 are presented in Scheme 1.
Path A involves initial formation of the desired hydroxo
complex 7, followed by intramolecular attack of coordi-
nated OH on the ortho position of the perfluorobenzyl
ring, eliminating HF in an addition-elimination reac-
tion, to give the observed product. The other possibility
considers the formation of an initial silver-iodide
complex (path B) which then directs the attack of
hydroxide to the ortho position of the fluoroaryl ring
without prior involvement of 7. Several studies have
shown that such silver-halide interactions are favor-
able and may be the first step in the mechanism of
halide abstraction or substitution.12,13 Clearly, if 7 could
be prepared by an alternative route, its potential role
in path A could be evaluated. Alas, attempts to produce
7 by deprotonation reactions of 5 have been uniformly
unsuccessful; most bases either displace water or react
by other pathways to produce hydrido complexes.14
(11) Orpen, A. G.; Brammer, L.; Allen, F. H.; Kennard, O.; Watson,
D. G.; Taylor, R. J . Chem. Soc., Dalton Trans. 1989, S1-S83.
(12) Liston, D. J .; Lee, Y. J .; Scheidt, W. R.; Reed, C. A. J . Am. Chem.
Soc. 1989, 111, 6643-6648.
(13) Strauss, S. H. Chem. Rev. 1993, 93, 927-942.

366 Organometallics, Vol. 20, No. 2, 2001
Notes
the filtrate showed a single product (5), which was isolated
after removal of the volatiles. Crystallization from methylene
chloride and hexanes afforded deep red crystals (39%). 1H
Finally, the unusual facility with which hydrolysis of
the CF2 group in 9 occurs is worthy of comment. Similar
facile hydrolyses have been observed in the rhodium
perfluorobenzyl complex 4 (vide supra),2 in the trans-
formation of metallacycle 11 to 12 under all but the most
rigorous conditions of water exclusion,15 and in the ring-
linked complex 13, which is rapidly transformed to 14
by even traces of adventitious moisture.1 While the
R-CF2 hydrolysis in 4 has been shown unambiguously
to involve the coordinated water molecule,2 and that of
11 has been suggested to involve coordinated water,15
the corresponding hydrolyses of 9 and 11 appear not to
involve ligated water and may be a consequence of
catalysis by traces of exogenous acid present, as ob-
served for other systems.16-18 The CF2 groups in all
these cases are unusually activated, with additional
benzylic or allylic stabilization of the initially formed
carbocation resulting from fluoride loss.
NMR (C6D6): δ 1.28 (s). 19F NMR (C6D6): δ -40.7 (t, J FF
)
29, 2F, CRF2), -134.2 (m, 2F, ortho), -157.4 (t, J FF ) 22, 1F,
para), -164.0 (m, 2F, meta). Anal. Calcd for C34H30F14I2Rh2:
C, 35.08; H, 2.60. Found: C, 34.59; H, 2.54.
Rh (η5-C5Me5)(CF 2C6F 4O)(P Me3) (9). Rh(η5-C5Me5)(CF2-
C6F5)(I)(PMe3) (2; 83 mg, 0.126 mmol) was dissolved in CH2-
Cl2 (5 mL) to give a orange-yellow solution. Freshly prepared
Ag2O‚xH2O (40 mg) was then added as a solid, and the solution
turned a yellow color. The reaction mixture was stirred for 18
h and then filtered. The volatiles were removed under vacuum
to give 9 as a yellow solid (52 mg, 78%). 1H NMR (acetone-d6):
δ 1.73 (d, J HRh ) 3, 15H, C5Me5), 1.37 (d, J HP ) 11, 9H, PMe3);
(C6D6) δ 1.36 (d, J HRh ) 3, 15H, C5Me5), 0.84 (d, J HP ) 11, 9H,
PMe3); (CD2Cl2) δ 1.70 (d, J HRh ) 3, 15H, C5Me5), 1.33 (d, J HP
) 11, 9H, PMe3). 19F NMR (acetone-d6): δ -47.3 (dm, J AB
234, J FP ) 27, J FF ) J FRh ) 8, 1F, CRFA), -48.1 (ddt, J AB
)
)
234, J FP ) 40, J FF ) J FRh ) 12, 1F, CRFB), -147.3 (m, 1F),
-161.5 (t, J FF ) 20, 1F), -168.7 (m, 1F), -181.1 (m, 1F); (C6D6)
δ -47.0 (m, 2F, CF2), -146.5 (m, 1F), -159.9 (t, J FF ) 20, 1F),
-168.4 (m, 1F), -179.3 (m, 1F); (CD2Cl2) δ -48.4 (m, 2F, CF2),
-148.2 (m, 1F), -161.1 (t, J FF ) 20, 1F), -169.1 (m, 1F),
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All reactions were performed in
oven-dried glassware, using standard Schlenk techniques,
under an atmosphere of dinitrogen which had been deoxygen-
ated over BASF catalyst and dried over Aquasorb, or in a
Braun drybox. 1H NMR (300 MHz), 19F NMR (282.2 MHz), and
31P NMR (121.4 MHz) spectra were recorded on a Varian Unity
Plus 300 FT spectrometer at 23 °C; coupling constants were
-180.3 (m, 1F). 19F{31P} NMR (acetone-d6) δ -47.3 (dt, J AB
)
234, J FF and J FRh ) 8, 1F, CRFA), -48.1 (dt, J AB ) 234, J FF
and J FRh ) 12, 1F, CRFB). 31P{1H} NMR (acetone-d6): δ 6.4
(ddd, J PRh ) 160, J PF ) 40, J PF ) 27, PMe3); (C6D6) δ 4.3 (ddd,
J PRh ) 162, J PF ) 40, J PF ) 27, PMe3); (CD2Cl2) δ 4.9 (ddd,
J PRh ) 160, J PF ) 39, J PF ) 29, PMe3). Accurate microanalysis
was precluded by the facility of hydrolysis to 9 to give 10.
Rh (η5-C5Me5)(C(O)C6F 4O)(P Me3) (10). On standing in
solution, under dinitrogen, compound 9 hydrolyzes quantita-
tively to the acyl 10, which was isolated as yellow crystals from
1
recorded in hertz. H NMR chemical shifts were recorded as
ppm downfield from tetramethylsilane and referenced to the
solvent. 19F NMR chemical shifts were recorded as ppm and
internally referenced to CFCl3. Chemical shifts for 31P{1H}
NMR were recorded as ppm and externally referenced to 85%
H3PO4. The infrared spectra were recorded on a Perkin-Elmer
FTIR 1600 Series spectrometer. Microanalyses were performed
by Schwarzkopf Microanalytical Laboratory, Woodside, NY.
Rea gen ts. Hydrocarbon and ethereal solvents were distilled
under dinitrogen from sodium or sodium/potassium alloy and
benzophenone ketyl, and halogenated solvents from calcium
hydride. Deuterated solvents were purchased from ISOTEC
Inc. or Cambridge Isotope Laboratories. NMO (4-methylmor-
pholine-N-oxide) was purchased from Aldrich. Silver(I) oxide
hydrate was prepared according to the literature method,19 as
were complexes 1 and 2.1
[Rh (η5-C5Me5)(CF 2C6F 5)I]2 (5). Rh(η5-C5Me5)(CF2C6F5)I-
(CO) (1; 250 mg, 0.410 mmol) was dissolved in methylene
chloride (10 mL), and NMO (50 mg, 0.410 mmol) was added
as a solid. The reaction mixture was stirred overnight. During
the reaction a precipitate formed that was filtered via cannula.
Washing the precipitate with benzene redissolved it, and it
was added to the original solution. The 19F NMR spectrum of
1
methylene chloride and hexanes. H NMR (C6D6): δ 1.36 (d,
J HRh ) 2.9, 15H, C5Me5), 0.70 (d, J HP ) 10.3, 9H, PMe3). 19F
NMR (C6D6): δ -146.1 (m, 1F), -156.1 (t, J FF ) 21, 1F),
-167.3 (m, 1F), -179.8 (m, 1F). 31P{1H} NMR (C6D6): δ 2.9
(d, J PRh ) 166, PMe3). The compound was also characterized
by a single-crystal X-ray diffraction experiment.
Cr ysta llogr a p h ic Str u ctu r a l Deter m in a tion s. Crystal,
data collection, and refinement parameters for 5 and 10 are
given in Table 1. The systematic absences in the diffraction
data are uniquely consistent for the reported space groups.
The structures were solved using direct methods, completed
by subsequent difference Fourier syntheses, and refined by
full-matrix least-squares procedures. Semiempirical absorption
corrections were applied to the data sets. To maintain a better
than 10:1 data-to-parameter ratio, the carbon atoms of the
phenyl rings in 5 were isotropically refined. All other non-
hydrogen atoms in 5 and all non-hydrogen atoms in 10 were
refined with anisotropic displacement coefficients.
Ack n ow led gm en t. R.P.H. is grateful to the Na-
tional Science Foundation and to the Petroleum Re-
search Fund, administered by the American Chemical
Society, for generous financial support.
(14) Hughes, R. P.; Lindner, D. C.; Smith, J . M.; Kovacik, I.; Zhang,
D. Unpublished observations
(15) Hughes, R. P.; Rose, P. R.; Rheingold, A. L. Organometallics
1993, 12, 3109-3117.
(16) Clark, G. R.; Hoskins, S. V.; Roper, W. R. J . Organomet. Chem.
1982, 234, C9-C12.
Su p p or tin g In for m a tion Ava ila ble: Atomic fractional
coordinates, bond distances and angles, and anisotropic ther-
mal parameters for 5 and 10 are available free of charge via
the Internet at http://pubs.acs.org.
(17) Appleton, T. G.; Berry, R. D.; Hall, J . R.; Neale, D. W. J .
Organomet. Chem. 1989, 364, 249-273.
(18) Burrell, A. K.; Clark, G. R.; Rickard, C. E. F.; Roper, W. R. J .
Organomet. Chem. 1994, 482, 261-269.
(19) Brauer, G. Handbuch Pra¨p. Anorg. Chem., 3rd ed.; Ferdinand
Enke Verlag: 1975-1981; Vol. 2.
OM0008171