Synthesis and reactions of fac-Re(bpy)(CO)3(CH2OR) (R = H, acetyl; bpy = 2,2′-bipyridyl)

Article abstract of DOI:10.1021/om980102m

Photoassisted reactions of fac-Re(bpy)(CO)₃-(CH₂OR) (1, R = H; 2, R = OAc) and fac-Re(bpy)(CO)₃-CH₂Ph (5) in MeOH provide fac-Re(bpy)(CO)₃(OCH₃) (3); 2 and 3 have been structurally characterized. Radical paths are indicated for 2 and 5; a nonradical path, involving photoassisted β-hydride elimination from the hydroxymethyl group, is proposed for 1.

Full text of DOI:10.1021/om980102m

Organometallics 1998, 17, 2689-2691
2689
Syn th esis a n d Rea ction s of fa c-Re(bp y)(CO)3(CH2OR)
(R ) H, Acetyl; bp y ) 2,2′-Bip yr id yl)
Dorothy H. Gibson,* Bradley A. Sleadd, and Xiaolong Yin
Department of Chemistry and Center for Chemical Catalysis, University of Louisville,
Louisville, Kentucky 40292
Ashwani Vij
Department of Chemistry, University of Idaho, Moscow, Idaho 83844
Received February 18, 1998
Summary: Photoassisted reactions of fac-Re(bpy)(CO)3-
(CH2OR) (1, R ) H; 2, R ) OAc) and fac-Re(bpy)(CO)3-
CH2Ph (5) in MeOH provide fac-Re(bpy)(CO)3(OCH3) (3);
2 and 3 have been structurally characterized. Radical
paths are indicated for 2 and 5; a nonradical path, in-
volving photoassisted â-hydride elimination from the
hydroxymethyl group, is proposed for 1.
Reaction of Re(bpy)(CO)4+OTf- (OTf ) trifluorometh-
anesulfonate)5 with NaBH4 in MeOH afforded a red
product which has been identified as fac-Re(bpy)(CO)3-
(CH2OH) (1); the product, obtained in 69% yield, has
been characterized by elemental analysis and spectral
data.6 The acetoxymethyl derivative of 1, fac-Re(bpy)-
(CO)3(CH2OAc) (2; Ac ) acetyl), was prepared by
treating 1 with acetic acid (or acetic anhydride). Com-
pound 2 has been characterized by elemental analysis,
spectral data7, and an X-ray structure determination.8,9
In dry MeOH, under N2, irradiation of 1 through
Pyrex with a 450 W UV lamp (Ace-Hanovia) over 20 min
afforded a yellow solution which yielded a red solid, fac-
Re(bpy)(CO)3(OCH3) (3), in 97% yield. The yield of 3
was not affected by the exposure of samples to O2.
Compound 3 has been characterized by elemental
analysis, spectral data,10 and X-ray crystallography11
(see Figure 1). Photolyses of 1 in dry methanol-d4,
conducted in NMR tubes with and without O2, showed
that CH2O (present as the hemiacetal13,14 in 68% yield
as established by an NMR internal standard) was the
Ruthenium and rhenium polypyridyl complexes bear-
ing C1 ligands are of interest because of the possible
intermediacy of such compounds in photochemically and
electrochemically promoted reductions of carbon diox-
ide.1 Some of the C1 ligands are expected to be the same
as those suggested in catalytic reductions of CO.2
Furthermore, although many coordination complexes of
ruthenium and rhenium with polypyridyl ligands are
known, few organometallic compounds (those with
carbon ligands other than CO) have been characterized
and little is known of their chemical reactions.3 We
have begun to study compounds of these types, particu-
larly those with metallocarboxylate and formyl ligands.4
(1) (a) Hawecker, J .; Lehn, J . M.; Ziessel, R. J . Chem. Soc., Chem.
Commun. 1983, 536. (b) Sullivan, B. P.; Bolinger, C. M.; Conrad, D.;
Vining, T. J .; Meyer, T. J . J . Chem. Soc., Chem. Commun. 1985, 1414.
(c) Sullivan, B. P.; Conrad, D.; Meyer, T. J . Inorg. Chem. 1985, 24,
3640. (d) Hawecker, J .; Lehn, J .-M.; Ziessel, R. Helv. Chim. Acta 1986,
69, 1990. (e) Ishida, H.; Tanaka, K.; Tanaka, T. Organometallics 1987,
6, 181. (f) Bruce, M. R. M.; Megehee, E.; Sullivan, B. P.; Thorp, H.;
O’Toole, T. R.; Downard, A.; Meyer, T. J . Organometallics 1988, 7, 238.
(g) Sullivan, B. P.; Bruce, M. R. M.; O’Toole, T. R.; Bolinger, C. M.;
Megehee, E.; Thorp, H.; Meyer, T. J . In Catalytic Activation of Carbon
Dioxide; Ayers, W. M., Ed.; ACS Symp. Ser. 363; American Chemical
Society: Washington, DC, 1988; Chapter 6. (h) Ziessel, R. In Catalysis
by Metal Complexes: Photosensitization and Photocatalysis Using
Inorganic and Organometallic Compounds; Kalyanasundaram, K.,
Gra¨tzel, M., Eds.; Kluwer Academic: Dordrecht, The Netherlands, 1993;
p 217. (i) Christensen, P.; Hammett, A.; Muir, A. V. G.; Timney, J . H.
J . Chem. Soc., Dalton Trans. 1992, 1455. (j) Yoshida, T.; Tsutsumida,
K.; Teratani, S.; Yasufuku, K.; Kaneko, M. J . Chem. Soc., Chem.
Commun. 1993, 631. (k) Stor, G. J .; Hartl, F.; van Outersterp, J . W.
M.; Stufkens, D. J . Organometallics 1995, 14, 115. (l) J ohnson, F. P.
A.; George, M. W.; Hartl, F.; Turner, J . J . Organometallics 1996, 15,
3374. (m) Klein, A.; Vogler, C.; Kaim, W. Organometallics 1996, 15,
236.
(5) Shaver, R. J .; Rillema, D. P. Inorg. Chem. 1992, 31, 4101.
(6) To a solution of Re(bpy)(CO)4OTf 5 (0.20 g, 0.35 mmol) in 25 mL
of MeOH at 0 °C was added NaBH4 (0.058 g, 1.5 mmol) in several
portions. The reaction flask was wrapped in Al foil and stirred for 1 h
at 0 °C. The solvent was evaporated, and the resultant red solid was
collected and washed with water (100 mL). A red solid was obtained
(1; 0.11 g, 69%) and dried under vacuum; mp 158-160 °C dec. Anal.
Calcd for C14H11N2O4Re: C, 36.76; H, 2.42. Found: C, 36.86; H, 2.47.
IR (DRIFTS, KCl): νCO 1987 (s), 1865 (vs) cm-1 1H NMR (CD3OD): δ
.
9.06 (2H, d, J ) 5.7 Hz), 8.52 (2H, d, J ) 8.4 Hz), 8.12 (2H, t, J ) 8.1
Hz), 7.55 (2H, t, J ) 7.8 Hz), 3.37 (2H, s). 13C NMR (CD3OD): δ 203.98,
194.44, 156.59, 153.74, 139.24, 127.80, 124.68, 70.05.
(7) Compound 1 (0.30 g, 0.65 mmol) was added to 5 mL of acetic
acid. The mixture was stirred for 2 min; during this time, the solution
became yellow as the starting material dissolved. The resulting solution
was added to ca. 50 mL of water to effect precipitation of 2 (0.28 g;
86% yield); mp >210 °C. Anal. Calcd for C16H13N2O5Re: C, 38.47; H,
2.62. Found: C, 38.28; H, 2.63. IR (DRIFTS, KCl): νCO 2017, 1997 and
1885 cm-1; νester
1696 cm-1 1H NMR (CD3CN): δ 8.98 (2H, d, J )
.
CO
5.3 Hz), 8.37 (2H, d, J ) 8.0 Hz), 8.11 (2H, t, J ) 7.8 Hz), 7.53 (2H, t,
J ) 6.2 Hz), 3.94 (2H, s), 1.41 (3H, s). 13C NMR (CD3CN): δ 203.06,
194.37, 171.70, 156.29, 153.89, 139.52, 127.96, 124.40, 69.33, 20.98.
(8) Crystal data9 for fac-Re(bpy)(CO)3(H2COC(O)Me) (2): C16H13N2O5-
Re, Mr ) 499.48; monoclinic, space group P21/n; a ) 9.7402(2) Å, b )
15.8826(4) Å, c ) 11.2409(2) Å, R ) 90°, â ) 107.905(1)°, γ ) 90°. Final
R indices (I > 2σ(I)): R1 ) 0.0390, wR2 ) 0.0768.
(2) Dombek, B. D. Adv. Catal. 1993, 32, 325.
(3) See, for example: (a) Ishida, H.; Tanaka, K.; Morimoto, M.;
Tanaka, T. Organometallics 1986, 5, 724. (b) Tanaka, H.; Nagao, H.;
Peng, S.-M.; Tanaka, K. Organometallics 1992, 11, 1450. (c) Tanaka,
H.; Tzeng, B.-C.; Nagao, H.; Peng, S.-M.; Tanaka, K. Inorg. Chem. 1993,
32, 1508. (d) Toyohara, K.; Nagao, H.; Adachi, T.; Yoshida, T.; Tanaka,
K. Chem. Lett. 1996, 27. (e) Hankka, M.; Kiviaho, J .; Ahlgren, M.;
Pakkanen, T. A. Organometallics 1995, 14, 825.
(4) (a) Gibson, D. H.; Ding, Y.; Sleadd, B. A.; Franco, J . O.;
Richardson, J . F.; Mashuta, M. S. J . Am. Chem. Soc. 1996, 118, 11984.
(b) Gibson, D. H.; Sleadd, B. A.; Mashuta, M. S.; Richardson, J . F.
Organometallics 1997, 16, 4421.
(9) Additional commentary about the structure and ORTEP packing
diagrams are available in the Supporting Information.
(10) The crude product was recrystallized from CH2Cl2/hexane to
give a red solid, mp >210 °C. Anal. Calcd for C14H11N2O4Re: C, 36.67;
H, 2.42. Found: C, 36.51; H, 2.47. IR (DRIFTS, KCl): νCO 2003 and
1877 cm-1 1H NMR (CD3OD): δ 9.06 (2H, d, J ) 5.5 Hz), 8.56 (2H, d,
.
J ) 8.0 Hz), 8.25 (2H, t, J ) 8.0 Hz), 7.69 (2H, t, J ) 5.5 Hz), 3.60
(3H, s). 13C NMR (CD3OD): δ 199.97, 194.26, 157.50, 154.07, 141.24,
128.60, 125.07, 61.60.
S0276-7333(98)00102-2 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 05/23/1998

2690 Organometallics, Vol. 17, No. 13, 1998
Communications
ligands) which appear to involve Re-R bond homolysis,
we decided to probe the behavior of 1 and 2 further. It
seemed possible that homolysis of the Re-CH2OH bond
in 1 might result in fac-Re(bpy)(CO)3H (4)19 and CH2O
being formed. However, the stability, and solubility, of
1 prevented the use of solvents other than methanol to
examine this possibility. Since hydride 4 is known to
insert CO2 readily,19 it seemed possible that insertion
of CH2O into the Re-H bond of 4 could account for some
of 3 formed from 1. Reaction of 4 with CH2O (generated,
in situ, by sonication of paraformaldehyde dispersed in
acetone) occurs slowly in the dark and provides 3 as one
of the products (the yield of 3 reaches a maximum of
25% after 1 h). This represents the first example of
CH2O insertion into a rhenium hydride bond; however,
Berke et al.20 have shown that some rhenium-hydrides
are capable of inserting aromatic aldehydes to give
phenoxy derivatives. Under photolysis conditions, mul-
tiple products were generated from the reaction of 4
with CH2O in acetone and 3 was not present. The slow
thermal reaction, and the complexity of the product
mixture after a photolysis reaction, suggested that the
formation of 3 from photolysis of 1 in MeOH does not
occur by reaction of CH2O with hydride 4. Reaction of
hydride 4 with MeOH occurs very slowly; however, the
reaction is rapid (5 min) upon photolysis and provides
3 in 90% yield. Exact parallels with other hydrides
are not known, although some hydrides have been re-
ported21 to react with more acidic hydroxylic reagents
(e.g., phenol, hexafluoroisopropyl alcohol) to give phen-
oxide/alkoxide products. The observation that hydride
4 reacts rapidly with MeOH upon photolysis made the
Re-CH2OH bond homolysis pathway, followed by H-
atom abstraction by the rhenium fragment, a viable
possibility for the conversion of 1 to 3. However, the
absence of ethylene glycol as a product made it seem
less likely.
F igu r e 1. ORTEP drawing of 3 with thermal ellipsoids
shown at the 30% probability level. Selected bond distances
(Å) and bond angles (deg) are as follows: Re-O(3), 2.081-
(5); Re-C(14), 1.908(7); Re-C(11), 1.912(7); Re-C(12),
1.900(7); O(3)-C(13), 1.397(8); C(13)-O(3)-Re, 121.6(4);
O(3)-Re-C(14), 172.7(2).
other major product in addition to 3-OCD3. As expected
from previous work on related rhenium complexes,15
samples of 3 are converted (within 5 min) to 3-OCD3
upon dissolution in methanol-d4. Neither ethylene
glycol nor formic acid was produced from 1 under either
conditions. Solutions of 1 in MeOH, in the dark and
without O2, slowly converted to 3 over a period of several
days. Photoinduced conversion of a hydroxymethyl
complex to a metal hydride and CH2O has been observed
previously, but the reaction pathway was not detailed.16
In view of the recent observations of Schanze et al.17
on the behavior of fac-Re(bpy)(CO)3(R) (R ) alkyl,
benzyl) under photolysis and the related observations
of Stufkens et al.18 (on complexes with other R-diimine
(11) Crystal data9 for fac-Re(bpy)(CO)3(OCH3) (3): C14H11N2O4Re‚1/2
H2O, Mr ) 466.46; triclinic, space group P1h; a ) 10.7936(2) Å, b )
11.2033(2) Å, c ) 14.1032(10) Å, R ) 113.0600(10)°, â ) 106.4020-
(10)°, γ ) 94.4960(10)°, V ) 1470.37(4) Å3, Z ) 4, Dcalcd ) 2.107 Mg/
m3, T ) 213(2) K, F(000) ) 884, λ ) 0.710 73 Å (Mo KR). Intensity
data were collected on a crystal of dimensions 0.10 × 0.10 × 0.05 mm
mounted on a Siemens SMART CCD area detector diffractometer. Of
The previous study of the photolysis of fac-Re(bpy)-
(CO)3(CH2Ph) (5) in MeOH identified toluene as a
reaction product but did not establish the fate of the
rhenium fragment.17 A sample of 522 was prepared; its
photolysis in MeOH provided 3 in 73% yield (isolated).
Reactions in methanol-d4 showed that the benzyl frag-
ment from 5 is converted primarily to bibenzyl (identi-
fied by 1H NMR spectral comparison with an authentic
sample) together with other organic compounds, which
were not identified; toluene was a very minor product,
â-phenylethanol was absent, and most importantly, the
formaldehyde hemiacetal was not present. Photolysis
of 5 in methanol-d4 which had been purged with O2
7495 reflections measured (θmax ) 25.38°), 5146 were unique (Rint
)
0.0282). Semiempirical absorption corrections were applied (transmis-
sion 0.43416-0.68708). The structure was solved by direct methods
and refined by full-matrix least squares on F2. Final R indices (I >
2σ(I)): R1 ) 0.0298, wR2 ) 0.0596. Programs used were standard
control software and SHELXTL 5.03.12
(12) SHELXTL 5.03 (PC-version), Program Library for Structure
Solution and Molecular Graphics; Siemens Analytical Instruments
Division: Madison, WI, 1995.
(13) The 1H NMR spectrum showed a singlet at δ 4.64 assigned to
the methylene protons of the hemiacetal. This resonance also results
after sonication of paraformaldehyde in methanol-d4 for several hours
(a small resonance at δ 9.50, for the CH2O monomer, is also present).
Addition of a drop of commercial formalin (which contains mainly
formaldehyde hydrate) to methanol-d4 generates the same single
resonance at δ 4.64 after several hours of equilibration. Under our
conditions, the hemiacetal is expected to be the predominant species
present (see: Hurd, C. D. J . Chem. Educ. 1966, 43, 527). After removal
of solvent from the product mixture resulting from photolysis of 1 in
MeOH and dissolution in methanol-d4, the formaldehyde hemiacetal
is present in low yield (diminished by partial removal with solvent).
(14) Addition of dimedone to the mixture results in slow disappear-
ance of the resonance at δ 4.64 and formation of the dimedone-
formaldehyde derivative (compared with an authentic sample: Kemp,
W. Qualitative Organic Analysis; McGraw-Hill: Maidenhead, England,
1986; p 89).
(18) Rossenaar, B. D.; Kleverlaan, C. J .; van de Ven, M. C. E.;
Stufkens, D. J .; Vlcek, A., J r. Chem. Eur. J . 1996, 2, 228.
(19) (a) Sullivan, B. P.; Meyer, T. J . J . Chem. Soc., Chem. Commun.
1984, 1244. (b) Sullivan, B. P.; Meyer, T. J . Organometallics 1986, 5,
1500.
(20) Nietlispach, D.; Yeghini, D.; Berke, H. Helv. Chim. Acta 1994,
77, 2197.
(21) Ayllo´n, J . A.; Gervaux, C.; Sabo-Etienne, S.; Chaudret, B.
Organometallics 1997, 16, 2000.
(22) Compound 5 has been prepared previously,17 utilizing THF as
solvent. We obtained better results as follows: 0.40 g (0.79 mmol) of
Re(bpy)(CO)3Br was added to a Schlenk flask under N2; 50 mL of
anhydrous ether was then added, followed by 4 mL of PhCH2MgCl
(2.2 M in ether), and the mixture was stirred for 2.5 h. Water was
then added to destroy the remaining Grignard reagent, and solvent
was removed under vacuum. The residue was extracted with 3 × 20
mL of CH2Cl2. These extracts were combined and dried over MgSO4;
after filtration, the filtrate was concentrated to 20 mL. Hexane was
then added to effect precipitation of 0.34 g (83% yield) of 5.
(15) Simpson, R. D.; Bergman, R. G. Organometallics 1993, 12, 781.
(16) Van Voorhees, S. L.; Wayland, B. B. Organometallics 1985, 4,
1887.
(17) Lucia, L. A.; Burton, R. D.; Schanze, K. S. Inorg. Chim. Acta
1993, 208, 103.

Communications
Organometallics, Vol. 17, No. 13, 1998 2691
Sch em e 1. P ossible P a th w a y for th e Con ver sion
of 1 to 4
afforded benzaldehyde (18%) and benzyl alcohol (51%)
at the expense of bibenzyl; the yield of 3-OCD3 was 85-
90% (NMR internal standard) and was not affected by
O2. Benzylic radicals, photogenerated in MeOH/O2, are
known to yield aromatic aldehydes as final products.23
We investigated reactions of complex 2 to determine
whether Re-CH2OR bond homolysis would occur in a
compound more closely related to 1. Photolysis of 2 in
MeOH, without O2, results in a 70% yield of the methoxy
complex 3 together with ethylene glycol diacetate
(AcOCH2CH2OAc, the product expected from radical
coupling; identified by 1H NMR spectral comparison
with an authentic sample). After photolysis of 2 in
methanol-d4, the reaction mixture showed, in addition
to 3-OCD3, the same diacetate together with a small
amount of formaldehyde hemiacetal; these two products
account for most of the methylene groups which were
lost from 2. Interestingly, when the reaction was
conducted in the presence of O2, compound 3 was not
in the product mixture; the major rhenium product
(80%) was fac-Re(bpy)(CO)3(OAc) (6)24 together with 7%
of fac-Re(bpy)(CO)3(O2CH)19 (7; identified by comparison
with an authentic sample). Solvolysis of 2 by MeOH
does not occur, and there is no direct source of acetic
acid which could react with the initially formed 3 to give
6. However, if the acetoxymethylene radical (•CH2OAc)
follows the path of the benzyl radical in MeOH/O2, a
possible rationale is available. Acetic formic anhydride
could be formed by oxidation of the radical. Solvolysis
of this anhydride by methanol-d4 could give both methyl
formate and methyl acetate (both deuterated at the
methoxy group) and the corresponding acids (deuterated
only at the carboxyl carbon atom). The acids could react
with initially formed 3 to give 6 and 7. Alternatively,
reaction of the mixed anhydride with 3 is possible; acetic
anhydride cleaves 3 rapidly to give acetate 6 as the final
product. Thus, both 6 and 7 could result from the mixed
anhydride. We have not investigated these possibilities
further.
in Scheme 1. Photolysis of 4 in MeOH, as discussed
above, will provide 3. The proposed dynamic behavior
of the bipyridyl ligand, with formation of monodentate
bpy intermediates, has been demonstrated for bipyridyl-
like ligands in other metal complexes.25 Alternatively,
intermediate hydride A might react with methanol
directly and provide 3 after loss of CH2O. Base-
promoted conversion of hydroxymethyl complexes to
hydrides, with loss of CH2O, has been observed previ-
ously.26 Thus, a pendant pyridyl ligand might assist
the conversion of 1 to 4; however, added pyridine does
not promote the conversion thermally. Thus, we favor
a â-elimination pathway for the conversion of 1 to a
hydride (A or 4).
Reactions of 1 in MeOH differ from those of 2 and 5.
The presence of the hydroxyl group in 1 allows an
alternate pathway to be dominant which does not
involve Re-CH2R bond homolysis. As with our results
on photoassisted reactions of a ruthenium formyl com-
plex with bipyridyl ligands,4a the determinative factors
are believed to be (a) photoassisted dechelation of one
bipyridyl nitrogen ligand and (b) the availability of a
group (from a cis location) which can quickly occupy the
newly created coordination vacancy. Further work is
required to support these proposals and to define their
implications in catalytic chemistry.
The lack of sensitivity of solutions of 1 in MeOH
toward O2, the absence of ethylene glycol in product
mixtures, and absence of either formic acid or formate
7 as products when O2 was present, support a nonradi-
cal pathway for conversion of 1 to 3 in MeOH. Also,
CH2O is liberated only in conversions of compound 1 to
3. Hydride 4 could be generated from 1, as suggested
Ack n ow led gm en t . Support by the United States
Department of Energy, Division of Chemical Sciences
(Office of Basic Energy Sciences), Office of Energy
Research, is gratefully acknowledged. The single-crystal
CCD X-ray facility at the University of Idaho was
established with the assistance of the NSF-Idaho EP-
SCoR program under NSF Grant No. OSR-9350539 and
the M. J . Murdock Charitable Trust, Vancouver, WA.
(23) Slocum, G. H.; Schuster, G. B. J . Org. Chem. 1984, 49, 2177.
(24) (a) Photolysis of 2 in methanol-d4/O2 for 10 min resulted in two
major products, 6 and another compound believed to be an isomer (the
1H NMR showed an acetoxy methyl resonance at δ 1.91 and resonances
for bipyridyl protons which are distinct from 6; the ratio of the two
compounds was 2:1 of the unknown to 6). Continued photolysis, or
evaporation of solvent and redissolution, resulted in conversion of the
second compound to 6. Samples were too dilute to provide meaningful
IR spectral data. (b) Alternate preparation of 6: hydride 4 (0.050 g,
0.12 mmol) was added to AcOH (7 mL) at room temperature. The
mixture was stirred for 5 min. Solvent was removed, and the residue
was washed with 3 × 10 mL of ether. The resultant yellow solid was
dried to give 6 (0.056 g, 98%), mp >210 °C. Anal. Calcd for C15H11N2O5-
Re: C, 37.11; H, 2.28. Found: C, 36.96; H, 2.37. IR (DRIFTS, KCl):
Su p p or tin g In for m a tion Ava ila ble: Tables of data col-
lection and refinement parameters, anisotropic thermal pa-
rameters, atomic positional parameters, bond distances, and
bond angles, ORTEP and packing diagrams, and text giving
additional commentary about the structures of 2 and 3 (18
pages). Ordering information is given on any current mast-
head page.
OM980102M
νCO 2018, 1904, and 1888 cm-1; νCOO 1624 and 1316 cm-1 1H NMR
.
(CD3CN): δ 9.04 (2H, d, J ) 4.9 Hz), 8.38 (2H, d, J ) 8.2 Hz), 8.18
(2H, t, J ) 7.9 Hz), 7.60 (2H, t, J ) 11.5 Hz), 1.45 (3H, s). 13C NMR
(CD3CN): δ 207.86, 198.11, 175.89, 156.70, 154.57, 140.52, 128.01,
124.65, 23.13. The product obtained from photolysis of 2 in MeOH/O2
had identical spectral properties.
(25) Gogoll, A.; O¨ rnebro, J .; Grennberg, H.; Ba¨ckvall, J .-E. J . Am.
Chem. Soc. 1994, 116, 6, 3631.
(26) (a) Thorn, D. L. Organometallics 1982, 1, 197. (b) Thorn, D. L.;
Tulip, J . H. Organometallics 1982, 1, 1580. (c) Lin, Y. C.; Milstein, D.;
Wreford, S. S. Organometallics 1983, 2, 1461.