Speciation and mechanistic studies of chiral copper(I) Schiff base precursors mediating asymmetric carbenoid insertion reactions of diazoacetates into the Si-H bond of silanes

Article abstract of DOI:10.1021/om0003786

The speciation and mechanistic activities of chiral copper(I) Schiff base precursors, mediating asymmetric carbenoid insertion reactions of diazoacetates into the Si-H bond of silanes, are studied. Mechanistic analysis of the [(R,R-1)Cu^I(CH₃CN)]⁺-mediated asymmetric insertion of the carbenoid precursor PhC(N₂)CO₂Me into the Si-H bonds of silanes (R,R-1 is a chiral diimine ligand) indicates, that despite structural complexity, the reaction is mediated by a mononuclear and apparently highly preorganized copper-carbene moiety, which performs concerted insertion of the carbenoid carbon into a weakly polarized Si-H bond, the latter entering hydrogen-first from the open side of the C₂-symmetric activity.

Full text of DOI:10.1021/om0003786

2896
Organometallics 2000, 19, 2896-2908
Sp ecia tion a n d Mech a n istic Stu d ies of Ch ir a l Cop p er (I)
Sch iff Ba se P r ecu r sor s Med ia tin g Asym m etr ic Ca r ben oid
In ser tion Rea ction s of Dia zoa ceta tes in to th e Si-H Bon d
of Sila n es
Les A. Dakin,† Patricia C. Ong,† J ames S. Panek,*,† Richard J . Staples,‡ and
Pericles Stavropoulos*,†
Department of Chemistry, Boston University, Boston, Massachusetts 02215, and Department of
Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138
Received May 3, 2000
[Cu(CH3CN)4](PF6) and the chiral C2-symmetric diimine ligand (1R,2R)-bis((2,6-dichlo-
robenzylidene)diamino)cyclohexane (R,R-1) (1.2 equiv) mediate asymmetric carbenoid inser-
tion of aryl diazoesters into the Si-H bond of silanes in good to high yields and levels of
enantiocontrol. Dichloromethane solutions of [Cu(CH3CN)4](PF6)/R,R-1 afford yellow crystals
of [CuI(R,R-1)(CH3CN)‚CuI(R,R-1)(CH3CN)2‚CuI (R,R-1)3](PF6)4‚2CH2Cl2‚3Et2O, which fea-
2
ture three distinct copper complexes in the crystal lattice. 1H NMR and electrospray ionization
MS (ESI-MS) studies establish that only [CuI(R,R-1)(CH3CN)]+ is present in solution in high
yields. Upon addition of stoichiometric PhC(N2)CO2Me, cations [CuI(R,R-1)(C(CO2Me)Ph)]+
and [CuI(R,R-1)(C(CO2Me)Ph)(CH3CN)]+ are detected by ESI-MS, consistent with the
presence of a copper-carbenoid moiety. The catalytically active precursor is most likely a
mononuclear unit of the type [CuI(R,R-1)(CH3CN)n]+, as suggested by the linearity of plots
relating the enantiomeric excess (ee) of product to that of the ligand (Kagan’s analysis).
Hammett plots correlate enhanced catalytic reactivities with stabilization of a sizable positive
charge on the carbenoid carbon and a smaller positive charge on the silicon atom, but the
corresponding enantioselectivities are insensitive to these electronic properties. The kinetic
isotope effect for carbenoid insertion into PhMe2Si-H(D) varies from 2.12 (-40 °C) to 1.08
(25 °C), in agreement with other small KIE values observed for processes in which Si-H
activation is involved in the turnover-limiting step. A linear Eyring plot of ln(k(major
enantiomer)/k(minor enantiomer)) over a range of 80 K supports the notion of a single step
controlling the levels of enantioselection. No H/D scrambling is observed in competitive
carbenoid insertions into PhMe2Si-D/Ph2MeSi-H, indicating that the insertion reaction takes
place in a concerted fashion. These results are discussed in light of an early transition state,
characterized by hydrogen-first penetration of the Si-H bond into the copper-carbenoid
cavity, which is assumed to impart high levels of enantioselectivity due to intrinsic
preorganization under the influence of the specific ligand and aryl diazoesters employed.
In tr od u ction
trophilic metal-carbene species,4 generated in situ via
decomposition of R-diazocarbonyl compounds,5 have
been known to mediate both intra- and intermolecular6
carbenoid C-H and Si-H insertion reactions, as well
as ylide type “insertions” into N-H, O-H, and S-H
bonds7 (eq 1).
The selective activation of unfunctionalized C-H1 and
Si-H2 bonds is a transformation of great synthetic
potential and utility. Of particular interest are meth-
odologies3 that involve insertion of a metal-activated
electrophilic moiety into a C-H or Si-H bond under
high levels of regio-, diastereo-, or enantiocontrol. Elec-
† Boston University.
‡ Harvard University.
(1) (a) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T.
H. Acc. Chem. Res. 1995, 28, 154-162. (b) Olah, G. A.; Prakash, G. K.
S. In The Chemistry of Alkanes and Cycloalkanes; Patai, S., Rappoport,
Z., Eds.; J ohn Wiley & Sons: New York, 1992; pp 609-652. (c) Shilov,
A. E.; Shul’pin, G. Chem. Rev. 1997, 97, 2879-2932.
(2) (a) Ojima, I.; Li, Z.; Zhu, J . In The Chemistry of Organic Silicon
Compounds; Rappoport, Z., Apeloig, Y., Eds.; J ohn Wiley & Sons: New
York, 1998; Vol. 2, Part 2, pp 1687-1792. (b) Tilley, T. D. In The
Silicon-Heteroatom Bond; Patai, S., Rappoport, Z., Eds.; J ohn Wiley
& Sons: New York, 1991; pp 245-364. (c) Corey, J . Y.; Braddock-
Wilking, J . Chem. Rev. 1999, 99, 175-292.
The most prominent metal-centered reagents in this
context are dirhodium(II) carboxylate and carboxami-
(3) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds; Wiley-Interscience: New
York, 1998.
(4) (a) Doyle, M. P.; Forbes, D. C. Chem. Rev. 1998, 98, 911-935.
(b) Doyle, M. P. Chem. Rev. 1986, 86, 919-939. (c) Padwa, A.; Austin,
D. J . Angew. Chem., Int. Ed. Engl. 1994, 33, 1797-1815.
10.1021/om0003786 CCC: $19.00 © 2000 American Chemical Society
Publication on Web 06/29/2000

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2897
date species, chiral versions of which have been shown4
to induce remarkable levels of enantioselection in C-H
and Si-H insertion reactions. Intermolecular insertion
into C-H bonds8 has proven to be particularly chal-
lenging, as evidenced by the only recently disclosed
examples of highly selective, Rh(II)-catalyzed, asym-
metric carbenoid insertion into the C-H bonds of
cycloalkanes9 (eq 2) or the R-C-H bonds of tetrahydro-
furan and cyclic amines.10
by chiral Rh(II) carbenoid species,7a,13 generated via
decomposition of phenyl or vinyl diazoacetates (eq 4).
Currently emerging as more economical alternatives
to rhodium-based reagents are chiral C2-symmetric
copper(I) diimine compounds, whose utility in mediating
carbenoid (CudCRR′) or nitrenoid (CudNR) addition to
unsaturated substrates has been amply demonstrated
by J acobsen’s group in enantioselective cyclopropana-
tion14 (eq 5) and aziridination14,15 (eq 6) reactions,
In contrast, asymmetric intramolecular carbenoid
C-H insertion reactions have been widely demon-
strated,11 especially in synthetic methodology that
requires formation of five-member heteroatom-contain-
ing rings12 (eq 3).
The more reactive Si-H bonds can readily undergo
asymmetric intermolecular insertion reactions catalyzed
(5) (a) Ye, T.; McKervey, M. A. Chem. Rev. 1994, 94, 1091-1160.
(b) Brunner, H. Angew. Chem., Int. Ed. Engl. 1992, 31, 1183-1185.
(6) Davies, H. M. L. Aldrichim. Acta 1997, 30, 107-114.
(7) (a) Bulugahapitiya, P.; Landais, Y.; Parra-Rapado, L.; Planche-
nault, D.; Weber, V. J . Org. Chem. 1997, 62, 1630-1641, and references
therein. (b) Simal, F.; Demonceau, A.; Noels, A. F. Tetrahedron Lett.
1999, 40, 63-66.
(8) (a) Adams, J .; Poupart, M.-A.; Greainer, L.; Schaller, C.; Quimet,
N.; Frenette, R. Tetrahedron Lett. 1989, 30, 1749-1752. (b) Demon-
ceau, A.; Noels, A. F.; Hubert, A. J .; Teyssie, P. J . Mol. Catal. 1988,
49, L13-L17. (c) Callott, H. J .; Metz, F. Nouv. J . Chim. 1985, 9, 167-
171. (d) Demonceau, A.; Noels, A. F.; Hubert, A. J .; Teyssie, P. Bull.
Soc. Chim. Belg. 1984, 93, 945-948. (e) Callott, H. J .; Metz, F.
Tetrahedron Lett. 1982, 23, 4321-4324. (f) Demonceau, A.; Noels, A.
F.; Hubert, A. J .; Teyssie, P. J . Chem. Soc., Chem. Commun. 1981,
688-689. (g) Scott, L. T.; DeCicco, G. J . J . Am. Chem. Soc. 1974, 96,
322-323. (h) Ambramovitch, R. A.; Roy, J . J . Chem. Soc., Chem.
Commun. 1965, 542-543.
(9) (a) Davies, H. M. L.; Hansen, T. J . Am. Chem. Soc. 1997, 119,
9075-9076. (b) Davies, H. M. L.; Hodges, L. M.; Matasi, J . J .; Hansen,
T.; Stafford, D. G. Tetrahedron Lett. 1998, 39, 4417-4420.
(10) Davies, H. M. L.; Hansen, T.; Hopper, D. W.; Panaro, S. A. J .
Am. Chem. Soc. 1999, 121, 6509-6510.
(11) Doyle, M. P. In Comprehensive Organometallic Chemistry II;
Hegedus, L. S., Ed.; Pergamon Press: New York, 1995; Vol. 12,
Chapters 5.1 and 5.2.
(12) (a) Doyle, M. P.; Kalinin, A. V.; Ene, D. G. J . Am. Chem. Soc.
1996, 118, 8837-8846. (b) Bode, J . W.; Doyle, M. P.; Protopopova, M.
N.; Zhou, Q.-L. J . Org. Chem. 1996, 61, 9146-9155. (c) Doyle, M. P.;
Dyatkin, A. B.; Roos, G. H. P.; Can˜as, F.; Pierson, D. A.; van Basten,
A.; Mu¨ller, P.; Polleux, P. J . Am. Chem. Soc. 1994, 116, 4507-4508.
(d) Doyle, M. P.; Westrum, L. J .; Wolthuis, W. N. E.; See, M. M.; Boone,
W. P.; Bagheri, V.; Pearson, M. M. J . Am. Chem. Soc. 1993, 115, 958-
964. (e) Taber, D. F.; You, K. K.; Rheingold, A. L. J . Am. Chem. Soc.
1996, 118, 547-556. (f) Taber, D. F.; Song, Y. J . Org. Chem. 1996, 61,
6706-6712. (g) Wang, P.; Adams, J . J . Am. Chem. Soc. 1994, 116,
3296-3305.
respectively. Another major class of widely applicable
C2-symmetric chiral reagents constitutes the bis(ox-
azolinyl) (box) and bis(oxazolinyl)pyridyl (pybox) Cu(I)
and Cu(II) systems, which have been developed by
Masamune,16 Evans,17 and Pfaltz18 to catalyze, inter
alia, enantioselective cyclopropanation, Diels-Alder,
and aldol reactions. These copper reagents have a close
(13) (a) Davies, H. M. L.; Hansen, T.; Rutberg, J .; Bruzinski, P. R.
Tetrahedron Lett. 1997, 38, 1741-1744. (b) Buck, R. T.; Doyle, M. P.;
Drysdale, M. J .; Ferris, L.; Forbes, D. C.; Haigh, D.; Moody, C. J .;
Pearson, N. D.; Zhou Q.-L. Tetrahedron Lett. 1996, 37, 7631-7634. (c)
Landais, Y.; Parra-Rapado, L.; Planchenault, D.; Weber, V. Tetrahe-
dron Lett. 1997, 38, 229-232. (d) Landais, Y.; Planchenault, D.
Tetrahedron Lett. 1994, 35, 4565-4568.
(14) Li, Z.; Quan, R. W.; J acobsen, E. N. J . Am. Chem. Soc. 1995,
117, 5889-5890.
(15) Li, Z.; Conser, K. R.; J acobsen, E. N. J . Am. Chem. Soc. 1993,
115, 5326-5327.
(16) (a) Lowenthal, R. E.; Abiko, A.; Masamune, S. Tetrahedron Lett.
1990, 31, 6005-6008. (b) Lowenthal, R. E.; Masamune, S. Tetrahedron
Lett. 1991, 32, 7373-7376.
(17) (a) Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M.
J . Am. Chem. Soc. 1991, 113, 726-728. (b) Evans, D. A.; Miller, S. J .;
Lectka, T. J . Am. Chem. Soc. 1993, 115, 6460-6461. (c) Evans, D. A.;
Faul, M. M.; Bilodeau, M. T. J . Am. Chem. Soc. 1994, 116, 2742-2753.
(d) Evans, D. A.; Kozlowski, M. C.; Burgey, C. S.; MacMillan, D. W. C.
J . Am. Chem. Soc. 1997, 119, 7893-7894. (e) Evans, D. A.; J ohnson,
J . S. J . Am. Chem. Soc. 1998, 120, 4895-4896. (f) Evans, D. A.; Burgey,
C. S.; Paras, N. A.; Vojkovsky, T.; Tregay, S. W. J . Am. Chem. Soc.
1998, 120, 5824-5825.
(18) (a) Mu¨ller, D.; Umbricht, G.; Weber, B.; Pfaltz, A. Helv. Chim.
Acta 1991, 74, 232-240. (b) Pfaltz, A. Adv. Catal. Proc. 1995, 1, 61-
94.

2898 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
precedent in Pfaltz’s chiral Cu(II)-semicorrin systems.19
Notable is also the class of Cu(I) reagents with C2-
symmetric bisazaferrocene ligands introduced by Fu.20
The presumed generation of a distinct copper carbene
or copper nitrene intermediate, situated preferentially
at a mononuclear copper site, has provided a unifying
mechanistic hypothesis14 to account for comparable
levels of stereocontrol between cyclopropanation and
aziridination reactions, catalyzed by any given member
of the chiral copper diimine class of compounds. Assum-
ing that incipient formation of a copper carbene inter-
mediate is also responsible for insertion reactions, as
generally accepted12d for similar dirhodium(II)-based
carbenoid insertions into C-H and Si-H bonds, a
mechanistic scenario of wide applicability is then emerg-
ing that can be conveniently tested on the same class
of chiral C2-symmetric copper diimine compounds.
A preliminary report21 from our laboratories in col-
laboration with J acobsen’s group demonstrated that
these copper catalysts can induce good to high yields of
enantioselection in carbenoid Si-H insertion reactions,
albeit restricted to the aryldiazoacetate class of R-dia-
zocarbonyl precursors (eq 7). In the present article, we
degassed by three freeze-pump-thaw cycles. Methanol and
water were degassed by bubbling nitrogen or argon for 0.5 h.
Starting materials were purchased from Aldrich and are of
the highest available purities. [Cu(CH3CN)4](PF6) was freshly
prepared prior to use from Cu2O and HPF6 according to a
procedure by Kubas.22 Analytical thin-layer chromatography
was performed on Whatman Reagent 0.25 mm silica gel 60-A
plates. Flash chromatography was performed on E. Merck
silica gel 230-400 mesh. Enantiomeric excess (ee) values were
determined by chiral HPLC (Chiralcel OD) with hexanes/
isopropyl alcohol (98:2) as the mobile phase with a flow rate
of 1 mL/min at 254 nm.
[Cu I(R,R-1)(CH3CN)‚Cu I(R,R-1)(CH3CN)2‚Cu I2(R,R-1)3]-
(P F 6)4‚2CH2Cl2‚3Et2O (2). [Cu(CH3CN)4](PF6) (200 mg, 0.54
mmol) and R,R-123 (274 mg, 0.64 mmol) were dissolved in CH2-
Cl2 (30.0 mL) and stirred for 10 min. The yellow solution was
filtered, and the filtrare was reduced under vacuum to 5 mL.
Diffusion of diethyl ether into this solution at -10 °C provides
initially a small amount (10 mg, 4.5%) of colorless crystals of
[Cu(CH3CN)4](PF6)‚CH3CN, which were filtered off. Diffusion
of excess diethyl ether into the filtrate affords yellow crystals
1
of 2 (370 mg, 78.5%). H NMR (CD2Cl2): δ 8.473 (s, 2H, HCd
N), 7.412 (q, 6H, Ph), 3.540 (dd, 2H, N-CH (cy), 2.437 (d, 2H,
cy), 2.044 (d, 2H, cy), 1.938 (s, 3H, CH3CN), 1.608 (m, 2H, cy),
1.466 (m, 2H, cy). Anal. Calcd for C120H133N13Cl24Cu4F24O3P4:
C, 41.29; H, 3.84; N, 5.22; Cl, 24.38; F, 13.06; P, 3.55. Found:
C, 41.82; H, 3.79; N, 5.18; Cl, 24.89; F, 13.33; P, 3.43.
Rep r esen ta tive P r oced u r e for Asym m etr ic Ca r ben oid
In ser tion in to th e Si-H Bon d . (2S)-Meth yl (1-p h en yl-2-
d im eth ylp h en ylsilyl)p r op a n oa te (3). To a round-bottom
examine in detail structural and mechanistic aspects
of the copper-mediated Si-H insertion reaction to reveal
unanticipated level of complexity in the speciation of
the chiral copper diimine species in solution. Despite
this complexity, several mechanistic indicators are
consistent with the notion that a mononuclear site is
responsible for the generation of a metal carbenoid
precursor which performs genuine insertion into the
Si-H bond of aryl and alkyl silanes, most likely featur-
ing an early transition state that retains elements of
hydride transfer character.
flask charged with a stir bar is added [Cu(CH3CN)4](PF6) (53
mg, 0.142 mmol) and the dichlorodiimine ligand R,R-1 (73 mg,
0.170 mmol). The mixture was dissolved in dry CH2Cl2 (4.5
mL), and the resulting homogeneous yellow solution was
allowed to react at room temperature for 30 min. To this
solution was added via syringe phenyldimethylsilane (388 mg,
2.85 mmol). The resulting solution was allowed to stir for 10
min at room temperature before being cooled to -40 °C. To
the cooled reaction mixture was added dropwise a solution of
PhC(N2)CO2Me (251 mg, 1.42 mmol) in CH2Cl2 (1.4 mL). The
resulting mixture was allowed to stir at -40 °C for 48-72 h
before being diluted with 10% ethyl acetate/hexanes and
flushed through a thin pad of SiO2. The resulting solution was
then concentrated under vacuum to yield crude insertion
product (3). The crude product was purified by SiO2 chroma-
tography using 2.5% ethyl acetate/hexanes as eluant, yielding
3 as a colorless oil (347 mg, 1.22 mmol, 86% yield). HPLC: tR
minor 4.33 min, tR major 10.14 min. 1H NMR (400 MHz,
CDCl3): δ 7.42-7.25 (m, 9H, Ar-H), 7.18 (m, 1H, Ar-H), 3.64
(s, 1H, CH), 3.58 (s, 3H, CO2CH3), 0.380 (s, 3H, Si-CH3), 0.360
(S, 3H, Si-CH3). 13C NMR (67.5 MHz, CDCl3): δ 173.05,
133.94, 133.89, 129.56, 128.28, 127.98, 127.62, 127.56, 125.61,
51.23, 45.95, -4.18, -4.60. IR (neat): νmax 3528, 3026, 2952,
1955, 1600 cm-1. CIMS (NH3 gas): 284.1, 207.1, 135.1, 118.0,
89.0. CIHRMS: calcd for [C17H20O2Si + H]+ 284.1232, found
Exp er im en ta l Section
Gen er a l Con sid er a tion s. Unless otherwise stated, all
operations involving synthesis and characterization of copper
reagents were performed under a pure dinitrogen or argon
atmosphere, using Schlenk techniques on an inert gas/vacuum
manifold or in a drybox (O2, H2O < 2 ppm). Hexane, petroleum
ether, and toluene were distilled over Na; THF and diethyl
ether, over Na/Ph2CO. Acetonitrile and methylene chloride
were distilled over CaH2. Ethanol and methanol were distilled
over the corresponding magnesium alkoxide, and acetone was
distilled over drierite. Deuterated solvents for NMR experi-
ments were purchased from Cambridge Isotope Laboratory.
All solvents, with the exception of methanol and water, were
284.1298. [R]23 ) -19.09° (c ) 1.21, CHCl3).
(19) (a) Fritschi, H.; Leutenegger, U.; Pfaltz, A. Angew. Chem., Int.
Ed. Engl. 1986, 25, 1005-1006. (b) Fritschi, H.; Leutenegger, U.;
Siegmann, K.; Pfaltz, A.; Keller, W.; Kratky, C. Helv. Chim. Acta 1988,
71, 1541-1552. (c) Fritschi, H.; Leutenegger, U.; Pfaltz, A. Helv. Chim.
Acta 1988, 71, 1553-1565. (d) Leutenegger, U.; Umbricht, G.; Fahrni,
C.; von Matt, P.; Pfaltz, A. Tetrahedron 1992, 48, 2143-2156.
(20) Lo, M. M.-C.; Fu, G. C. J . Am. Chem. Soc. 1998, 120, 10270-
10271.
D
Analytical data for all other products of carbenoid insertion
into Si-H bonds discussed in the text have been deposited as
Supporting Information.
Gen er a l P r oced u r e for t h e Gen er a t ion of Ha m m et t
P lots. (a ) P a r a -Su bstitu ted P h en yl Dia zoester s. The
(21) Dakin, L. A.; Schaus, S. E.; J acobsen, E. N.; Panek, J . S.
Tetrahedron Lett. 1998, 39, 8947-8950.
(22) Kubas, G. J . Inorg. Synth. 1979, 19, 90-92.

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2899
Ta ble 1. Cr ysta llogr a p h ic Da ta a for
[Cu I(R,R-1)(CH3CN)‚Cu I(R,R-1)(CH3CN)2‚
Cu I2(R,R-1)3](P F 6)4‚2CH2Cl2‚3Et2O (2)
following is a typical procedure for determining the relative
reactivity of p-X-PhC(N2)CO2Me (X ) MeO, Me, F, CF3) versus
PhC(N2)CO2Me in carbenoid insertion reactions into the Si-H
bond of PhMe2Si-H. To a round-bottom flask charged with a
stir bar is added [Cu(CH3CN)4](PF6) (18 mg, 0.049 mmol) and
the dichlorodiimine ligand R,R-1 (25 mg, 0.059 mmol). The
mixture was dissolved in dry CH2Cl2 (4.5 mL), and the
resulting homogeneous yellow solution was allowed to react
at room temperature for 30 min. To this solution was added
via syringe phenyldimethylsilane (67 mg, 0.495 mmol). The
resulting solution was allowed to stir for 10 min at room
temperature before being cooled to 0 °C. To the cooled reaction
mixture was added dropwise a solution of p-MeO-PhC(N2)CO2-
Me (100 mg, 0.495 mmol) and PhC(N2)CO2Me (87 mg, 0.495
mmol) in CH2Cl2 (1.4 mL). The reaction was allowed to run
until completion (TLC) and was diluted with cold 10% ethyl
acetate/hexanes. This crude reaction mixture was subjected
to a short silica gel plug to remove the copper reagent and
was concentrated under vacuum. The ratio of the products (p-
MeO/H) was determined through analysis of the crude reaction
formula
fw
cryst syst
space group
Z
C120H133Cl24Cu4F24N13O3P4
3490.24
orthorhombic
P2(1)2(1)2(1)
4
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
T, K
18.0299(3)
18.7372(5)
44.781(1)
90
90
90
15128.5(7)
213(2)
color
yellow
1.532
1.102
1.031
d
calc, g/cm3
µ, mm-1
GOF on F2
R1b (wR2c), %
6.31 (7.90)
a
1
Obtained with graphite-monochromated Mo KR (λ ) 0.71073
b 2
mixture using H NMR spectroscopy.
Å) radiation. R1 ) ∑||Fo| - |Fc||/∑|Fo|. c wR2 ) {∑[w(Fo - Fc2)2/
(b) P a r a -Su bstitu ted Ar yl Sila n es. The following is a
typical procedure for determining the relative reactivity of p-X-
PhMe2SiH (X ) MeO, Me, F, CF3) versus PhMe2SiH in
carbenoid insertion reactions of PhC(N2)CO2Me into the Si-H
bond of the aryl silanes. To a round-bottom flask charged with
a stir bar is added [Cu(CH3CN)4](PF6) (28 mg, 0.074 mmol)
and the dichlorodiimine ligand R,R-1 (38 mg, 0.089 mmol). The
mixture was dissolved in dry CH2Cl2 (4.5 mL), and the
resulting homogeneous yellow solution was allowed to react
at room temperature for 30 min. To this solution was added
via syringe a mixture of p-MeO-PhMe2SiH (123 mg, 0.744
mmol) and PhMe2SiH (101 mg, 0.744 mmol). The resulting
solution was allowed to stir for 10 min at room temperature
before being cooled to 0 °C. To the cooled reaction mixture was
added dropwise a solution of PhC(N2)CO2Me (131 mg, 0.744
mmol) in CH2Cl2 (1.4 mL). The reaction was allowed to run
until completion (TLC) and was diluted with cold 10% ethyl
acetate/hexanes. This crude reaction mixture was subjected
to a short silica gel plug to remove the copper reagent and
was concentrated under vacuum. The ratio of the products (p-
MeO/H) was determined through analysis of the crude reaction
∑[w(Fo2)2]}1/2
.
LT-2 low-temperature apparatus operating at 213 K. Data
were measured using omega scans of 0.3° per frame for 30 s,
such that a hemisphere was collected. A total of 1271 frames
were collected with a final resolution of 0.75 Å for [CuI(CH3-
CN)4](PF6)‚CH3CN and [CuI (R,R-1)3](PF6)2‚PhSCO2Et‚0.75CH2-
2
Cl2‚0.5CH3CN, and 0.80 Å for 2. The first 50 frames were
recollected at the end of data collection to monitor for decay.
Cell parameters were retrieved using SMART software24 and
refined using the SAINT software,25 which corrects for Lp and
decay. Absorption corrections were applied using SADABS26
supplied by G. Sheldrick. The structures are solved by the
direct method using the SHELXS-9727 program and refined
by least-squares methods on F2, SHELXL-97,28 incorporated
in SHELXTL-PC V 5.10.29
The structures were solved in the space groups specified in
Table 1 and in the Supporting Information by analysis of sys-
tematic absences. All non-hydrogen atoms are refined aniso-
tropically. Hydrogens were calculated by geometrical methods
and refined as a riding model. The crystals used for the diffrac-
tion studies showed no decomposition during data collection.
All drawings (Figures 1 and 2 and ORTEP diagrams in the
Supporting Information) are done at 50% probability ellipsoids.
Oth er P h ysica l Mea su r em en ts. 1H and 13C NMR spectra
were recorded on J EOL GSX-270 and Varian XL-400 NMR
spectrometers. FT-IR spectra were obtained on a Perkin-Elmer
1800 spectrometer. UV-vis spectra were obtained on a Hewlett-
Packard 8452A diode array spectrometer, equipped with an
Oxford DN1704 cryostat and ITC4 temperature controller for
low-temperature measurements. Electrospray ionization (ESI)
mass spectra were recorded using a Platform II MS (Micro-
mass Instruments, Danvers, MA). Samples were introduced
from solutions specified in the text at a flow rate of 5 µL/min
from a syringe pump (Harvard Apparatus). The electrospray
probe capillary was maintained at a potential of 3.0 kV, and
1
mixture using H NMR spectroscopy.
Gen er a l P r oced u r e for th e Deter m in a ton of th e Kin e-
tic Isotop e Effect. Kinetic isotope effects (KIEs) were deter-
mined by dissolving equimolar amounts of H-SiMe2Ph (283
mg, 2.08 mmol) and D-SiMe2Ph (284 mg, 2.08 mmol) along
with the copper complex [Cu(CH3CN)4](PF6) (52 mg, 0.140
mmol) and the chiral diimine ligand R,R-1 (72 mg, 0.168 mmol)
in CH2Cl2 (4.5 mL) upon stirring. The reaction was then
brought to the desired temperature using a cryogenic cooling
bath, and PhC(N2)CO2Me (245 mg, 1.4 mmol) was introduced
dropwise via syringe in CH2Cl2 (1.4 mL). The reaction was
allowed to run until consumption of the phenyl diazoester
(monitored by TLC) and was diluted with cold 10% ethyl
acetate/hexanes. This crude reaction mixture was subjected
to a short silica gel plug to remove the copper reagent and
was concentrated under vacuum. The ratio of the products (H/
1
D) was determined by H NMR analysis of the crude reaction
(23) Larrow, J . F.; J acobsen, E. N.; Gao, Y.; Hong, Y.; Nie, X.; Zepp,
C. M. J . Org. Chem. 1994, 59, 1939-1942.
mixture with the assistance of authentic samples of the
expected products.
(24) SMART V 5.050 (NT) Software for the CCD Detector System;
Bruker Analytical X-ray Systems: Madison, WI, 1998.
(25) SAINT V 5.01 (NT) Software for the CCD Detector System;
Bruker Analytical X-ray Systems: Madison, WI, 1998.
(26) SADABS Program for Absorption Corrections Using Siemens
CCD Based on the Method of Robert Blessing: Blessing, R. H. Acta
Crystallogr. 1995, A51, 33-38.
X-r a y Str u ctu r e Deter m in a tion s. Crystallographic data
were determined for [CuI(CH3CN)4](PF6)‚CH3CN (Supporting
Information), [CuI(R,R-1)(CH3CN)‚CuI(R,R-1)(CH3CN)2‚CuI -
2
(R,R-1)3](PF6)4‚2CH2Cl2‚3Et2O (2) (Table 1), and [CuI (R,R-1)3]-
2
(PF6)2‚PhSCO2Et‚0.75CH2Cl2‚0.5CH3CN (Supporting Informa-
tion). Single crystals were picked from the crystallization
vessel (coated with Paratone-N oil if necessary due to air-
sensitivity or desolvation), mounted on a glass fiber using
grease, and transferred to a Siemens (Bruker) SMART CCD
(charge coupled device) based diffractometer equipped with an
(27) Sheldrick, G. M. SHELXS-97 Program for the Solution of
Crystal Structure; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(28) Sheldrick, G. M. SHELXL-97 Program for the Refinement of
Crystal Structure; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(29) SHELXTL 5.10 (PC-Version) Program Library for Structure
Solution and Molecular Graphics; Bruker Analytical X-ray Systems:
Madison, WI, 1998.

2900 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
F igu r e 1. Solid-state structures of the mononuclear components [CuI(R,R-1)(CH3CN)]+ (top left) and [CuI(R,R-1)(CH3-
CN)2]+ (top right) in 2, showing 50% probability ellipsoids and the atom-labeling scheme. Space-filling models of [CuI-
(R,R-1)(CH3CN)]+ (bottom left) and [CuI(R,R-1)(CH3CN)2]+ (bottom right) are shown below each structure. Selected
interatomic distances (Å) and angles (deg) for [CuI(R,R-1)(CH3CN)]+: Cu1A-N1A 2.077(7), Cu1A-N2A 1.996(6), Cu1A-
N3A 1.836(8), N3A-C21A 1.115(10), N3A-Cu1A-N2A 139.6(3), N3A-Cu1A-N1A 136.9(3), N2A-Cu1A-N1A 83.1(3),
C1A-N1A-Cu1A 124.1(6), C8A-N2A-Cu1A 126.7(6), N2A-C8A-C9A 121.8(9), N1A-C1A-C2A 122.1(9), N3A-C21A-
C22A 176.9(12). For [CuI(R,R-1)(CH3CN)2]+: Cu1B-N1B 2.111(7), Cu1B-N2B 2.099(8), Cu1B-N3B 1.969(9), Cu1B-
N4B 1.914(8), N3B-C21B 1.117(11), N4B-C23B 1.056(10), N4B-Cu1B-N3B 117.8(4), N4B-Cu1B-N2B 111.2(4), N3B-
Cu1B-N2B 113.3(3), N4B-Cu1B-N1B 114.6(3), N3B-Cu1B-N1B 114.5(3), N2B-Cu1B-N1B 79.4(3), Cu1B-N2B-C8B
124.6(8), Cu1B-N1B-C1B 133.3(7), N3B-C21B-C22B 177.0(13), N4B-C23B-C24B 179.6(17).
the orifice to skimmer potential (“cone voltage”) was varied
from 15 to 30 V. Spectra were collected in the multichannel
acquisition mode. EI and CI mass spectra were obtained on a
Finnigan MAT-90 mass spectrometer. Optical rotations were
recorded on a Rudolph Research Autopol III digital polarimeter
at 589 nm. Microanalyses were done by H. Kolbe, Mikroana-
lytisches Laboratorium, Mu¨lheim an der Ruhr, Germany, and
by Quantitative Technologies Inc., Whitehouse, NJ .
infra). To assess the type of copper compound(s) present
in CH2Cl2 prior to the addition of any substrate, the
yellow solution was concentrated under vacuum and
diethyl ether was allowed to diffuse into this solution
at -20 °C. Two types of crystalline material are
obtained from these solutions. Minor amounts of color-
less crystals have been identified by X-ray analysis (see
Supporting Information) to be the starting material
[CuI(CH3CN)4](PF6)‚CH3CN. The major component of
the precipitated material is a yellow crystalline species
which has been shown by single-crystal X-ray analysis
to consist of three distinct copper-containing sites in a
single-crystal lattice, with overall stoichiometry [CuI-
Resu lts a n d Discu ssion
Syn th esis a n d Ch a r a cter iza tion of Cop p er (I)
P r ecu r sor Com p ou n d s. Reaction of [CuI(MeCN)4]-
(PF6) (10 mol %) with the chiral Schiff base (1R,2R)-
bis((2,6-dichlorobenzylidene)diamino)cyclohexane (R,R-
1, 12 mol %) in CH2Cl2 generates a yellow solution
which mediates the carbenoid insertion of methyl aryl
diazoacetates into the Si-H bond of alkyl and aryl
silanes with good to high levels of enantioselection (vide
(R,R-1)(CH3CN)‚CuI(R,R-1)(CH3CN)2‚CuI (R,R-1)3](PF6)4‚
2
2CH2Cl2‚3Et2O (2).
Solid -Sta te Str u ctu r es. The structure of the cationic
[CuI(R,R-1)(CH3CN)]+ component in 2 (Figure 1 (top

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2901
left)) reveals the presence of a chelating diimino ligand
and a single acetonitrile coordinated to a copper ion by
virtue of a slightly distorted C2-symmetric trigonal
planar geometry. The deviation from ideal C2 symmetry
is demonstrated by small variations of relevant bond
distances (Cu1A-N1A ) 2.077(7), Cu1A-N2A ) 1.996-
(6) Å) and angles (N3A-Cu1A-N1A ) 136.9(3)°, N3A-
Cu1A-N2A ) 139.6(3)°, C21A-N3A-Cu1A ) 172.5-
(10)°). As better indicated by the space-filling diagram
(Figure 1 (bottom left)), the mutual disposition of the
phenyl groups30 (torsion interplanar angle ) 129.9°) is
substantially influenced by the presence of the 2,6-
dichloro substituents, placing atoms Cl2A and Cl4A in
inward positions with respect to the cavity. The copper
center lies almost exactly on the imaginary line con-
necting carbon atoms C2A and C9A of the two phenyl
rings (C2A‚‚‚C9A ) 6.5 Å). The cleft generated by the
phenyl substituents is anticipated to accommodate the
catalytically active carbenoid unit upon displacement
of acetonitrile.
The structure of the second mononuclear copper
cation in 2, [CuI(R,R-1)(CH3CN)2]+ (Figure 1 (top right)),
reveals a distorted tetrahedral metal center (N1B-
Cu1B-N2B ) 79.3(3)°, N3B-Cu1B-N4B ) 117.8(4)°,
torsion angle N1B-Cu1B-N2B N3B-Cu1B-N4B )
89°) which possesses approximate C2 local symmetry,
featuring substantial deviation from ideal C2 geometry.
As the space-filling model indicates (Figure 1 (bottom
right)), the two flanking phenyl groups are positioned
in a manner similar to that observed in the case of the
trigonal-planar copper complex. The cavity so generated
is occupied by two acetonitrile moieties oriented at
positions of minimum interaction in relation to the
phenyl groups. In contrast to the trigonal-planar copper
site, the tetrahedrally coordinated copper(I) is coordi-
natively and electronically saturated and would thus
require dissociation of at least one acetonitrile prior to
coordination of any substrate moiety.
F igu r e 2. Solid-state structure of the dinuclear species
[CuI (R,R-1)3]2+ (top) in 2, showing 50% probability ellips-
2
oids and the atom-labeling scheme, and space-filling model
(bottom). Selected interatomic distances (Å) and angles
(deg): Cu1-N1 1.911(6), Cu1-N3 2.032(6), Cu1-N4
2.048(6), Cu2-N2 1.941(6), Cu2-N6 2.025(6), Cu2-N5
2.073(6), N1-Cu1-N3 141.9(2), N1-Cu1-N4 135.1(2),
N3-Cu1-N4 80.8(2), N2-Cu2-N5 131.3(2), N2-Cu2-N6
144.1(2), N6-Cu2-N5 82.5(2), Cu1-N3-C21 138.1(6),
Cu1-N4-C28 128.1(6), Cu1-N1-C1 124.0(6), Cu2-N2-
C8 121.9(6), Cu2-N5-C41 126.9(6), Cu2-N6-C48 135.2(6).
per complex (Figure 2 (bottom)) indicate that each
copper ion is tightly protected from substrate binding
by the surrounding dichloro-substituted phenyl groups
and central cyclohexane ring. The position of the
pendant phenyl and cyclohexane units associated with
the bridging diimine ligand is indicative of the orienta-
tion expected of the docking carbene moiety upon
catalysis.
The structure of the unique dicopper species [CuI -
2
(R,R-1)3]2+ in 2 (Figure 2 (top)) reveals two trigonal-
planar copper centers, each coordinated by a chelating
diimine ligand and connected to the other copper site
via a third diimine ligand that acts as bridging rather
than chelating moiety. An approximate local C2 axis
passes through the bridging diimino ligand, bisecting
internuclear distances C15-C16 and C18-C19. The
mutual orientation of the flanking phenyl groups of the
chelating diimino ligands defines a substantially wider
cavity than that observed in the case of the mononuclear
sites, reflecting the flexibility of the chelating diimino
ligand to accommodate a moiety in the cleft that is more
sterically demanding than acetonitrile. In the cavity, the
pendant phenyl and cyclohexane units associated with
the bridging diimine ligand assume a configuration of
minimum interaction vis-a`-vis the phenyl groups of the
chelating diimine ligands. The mode of coordination
displayed by the bridging diimino unit exemplifies the
potential of the present ligand to effect undesirable
oligomerization of chiral copper centers31 at a ligand-
to-copper ratio of g1. Space-filling models of the dicop-
Solu tion Str u ctu r es. To examine which of the above
species, and possibly others, are present in solution at
concentrations pertaining to catalytic conditions, a
spectroscopic examination of the system [CuI(MeCN)4]-
(PF6) (3.20 × 10-5 M)/R,R-1 (3.84 × 10-5 M) was
1
undertaken in CH2Cl2. The H NMR spectrum of this
catalytic combination in CD2Cl2 at 25 °C shows the
presence of only one set of well-resolved ligand-based
protons and a sharp resonance due to CH3CN. No
appreciable broadening or further resolution of these
resonances is observed at lower temperatures (from 25
to -80 °C), indicating the presence of a single major
copper species in solution. However, careful inspection
of the 1H NMR spectrum’s baseline suggests that
residual weak resonances may be due to other ligated
copper species present in minor amounts (e5%).
(30) For selectivity imposed by aromatic interactions in substrate
binding see: (a) Quan, R. W.; J acobsen, E. N. J . Am. Chem. Soc. 1996,
118, 8156-8157. (b) Evans, D. A.; Kozlowski, M. C.; Murry, J . A.;
Burgey, C. S.; Campos, K. R.; Connell, B. T.; Staples, R. J . J . Am. Chem.
Soc. 1999, 121, 669-685.
(31) For similar formation of chiral Cu(I) polymers see: Evans, D.
A.; Woerpel, K. A.; Scott, M. J . Angew. Chem., Int. Ed. Engl. 1992, 31,
430-432.

2902 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
To further investigate the identity of the 1H NMR
active species and other low-concentration copper-
containing components, positive-ion electrospray ioniza-
tion mass spectrometry (ESI-MS) was applied to yellow
solutions of [CuI(CH3CN)4](PF6)/R,R-1 in CH2Cl2 at
concentrations noted above. At low cone voltages, the
major molecular ion revealed by ESI-MS possesses the
correct mass (m/z ) 532 amu) and isotopic distribution
for [CuI(R,R-1)(CH3CN)]+ and is accompanied by its
acetonitrile-free fragment [CuI(R,R-1)]+ (m/z ) 491
amu), the latter increasing in abundance with increas-
ing cone voltage at the expense of [CuI(R,R-1)(CH3-
CN)]+. A second low-abundance ion evident in the ESI-
MS spectrum, but surprisingly not encountered among
the solid-state counterparts, corresponds in mass (m/z
) 919 amu) and isotopic distribution to [CuI(R,R-1)2]+.
Varying the ratio of [CuI(CH3CN)4](PF6)/(R,R-1) from
0.8 to 1.2 results in similar ESI-MS spectra, with the
exception that at the higher end of the range [CuI(R,R-
1)2]+ is almost completely suppressed in favor of the ion
[CuI(CH3CN)2]+ (m/z ) 145 amu), a fragment of [CuI(CH3-
CN)4]+. Otherwise, the crystallographically documented
ions [CuI(R,R-1)(CH3CN)2]+ and [CuI (R,R-1)3]+ are
2
present only in trace amounts. It is conceivable that the
representation of these ions in isolable amounts in the
solid state is favored by equilibria influenced by the
crystallization procedure.
We conclude then that [CuI(R,R-1)(CH3CN)]+ is the
major (>95%) copper-containing species in solutions of
[CuI(CH3CN)4](PF6) and R,R-1 in CH2Cl2 at concentra-
tions pertinent to catalysis. Other detectable copper
complexes, with most prominent the [CuI(R,R-1)2]+ and
[CuI(CH3CN)4]+ ions, are present in minor or trace
F igu r e 3. (Top) Positive-ion electrospray ionization MS
spectrum of solutions of [Cu(CH3CN)4](PF6)/R,R-1 and
PhC(N2)CO2Me in CH2Cl2. (Bottom) Theoretical (above)
and experimental (below) ESI-MS spectra of [CuI(R,R-1)-
(C(CO2Me)Ph)]+ (left) and [CuI(R,R-1)(C(CO2Me)Ph)(CH3-
CN)]+ (right).
amounts ([CuI(R,R-1)(CH3CN)2]+, [CuI (R,R-1)3]+).
2
Stoich iom etr ic Rea ction s of th e [Cu I(CH3CN)4]-
(P F 6)/R,R-1 System . (a ) Ar yl Dia zoester s. To assess
the order by which the active copper site(s) may activate
the two substrates of the carbenoid Si-H insertion
reaction, and possibly gain insight into the intermedi-
ates generated, we performed stoichiometric reactions
of [CuI(CH3CN)4](PF6)/R,R-1 with aryl diazoesters and
silanes, conducted separately for each substrate.
1
could be made by H NMR, probably owing to its low
concentration in solution. As evidenced by 1H NMR, the
major copper-containing species in these solutions is still
the [CuI(R,R-1)(CH3CN)]+ ion.
The reaction of [CuI(CH3CN)4](PF6) (10%) and R,R-1
(12%) with 1 equiv of PhC(N2)CO2Me in CH2Cl2 was
followed by variable-temperature 1H NMR, UV-vis, and
ESI-MS spectroscopies. Apparently, no reaction takes
place in the temperature range from -80 to -15 °C for
a period of 1 h, but within the temperature window from
-10 to 0 °C, a rapid change of color occurs from an
initial yellow to a deep red-orange solution (λmax ) 277,
298 (br), 429 (br) nm). 1H NMR spectra indicate that in
the temperature region above -10 °C PhC(N2)CO2Me
(conveniently followed by observing the proton reso-
nance of CO2CH3 at δ ) 3.83) is progressively consumed,
while three thermally stable products emerge which
incorporate the carbenoid moiety (δ ) 3.99, 3.79, and
3.53 (CO2CH3)). Following chromatographic purification,
these products have been identified by 1H NMR and
HRMS, and by comparison with authentic samples, to
be PhCOCO2Me (δ ) 3.99 (minor product)) and cis/
trans-PhC(CO2Me)dCPh(CO2Me) (major products). De-
spite the appearance of several minor resonances that
can be partly due to the colored species, no positive
identification of this thermally unstable intermediate
Positive-ion electrospray ionization MS (ESI-MS) was
then employed to identify molecular ions at low con-
centrations. Addition of 0.3 equiv of PhC(N2)CO2Me to
[CuI(CH3CN)4](PF6)/R,R-1 (1:1.2) in ice-cold CH2Cl2
generates a red-orange solution, which upon examina-
tion by ESI-MS (cone voltage 5-30 V) gives rise to four
novel low-yield molecular ions (Figure 3 (top)), in
addition to those observed by ESI-MS in solutions of
[CuI(CH3CN)4](PF6)/R,R-1 alone, as noted above. Two
of these ions contain copper and have been assigned by
mass and isotopic distribution to [CuI(R,R-1)(C(CO2Me)-
Ph)]+ (m/z ) 639 amu) and [CuI(R,R-1)(C(CO2Me)Ph)-
(CH3CN)]+ (m/z ) 680 amu) (Figure 3 (bottom)). The
other two ions are devoid of copper and correspond to
[(R,R-1)(C(CO2Me)Ph) + H]+ (m/z ) 577 amu) and
[(R,R-1)(C(CO2Me)(N2)Ph + H]+ (m/z ) 605 amu). The
observed ions [CuI(R,R-1)(C(CO2Me)Ph]+ and [CuI(R,R-
1)(C(CO2Me)Ph)(CH3CN)]+ are consistent with the pres-
ence of the carbenoid unit CudC(CO2Me)Ph, although
bond connectivities cannot be established. Dinitrogen
adducts of these copper-containing species are not
detected even at very low cone voltages; thus it is

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2903
Ta ble 2. Ca ta lytic In ser tion of Meth yl
P h en yld ia zoester in to Si-H Bon d s of Sila n es
unlikely that the observed ions are fragments of their
diazo precursors. Apparently, acetonitrile can coordinate
to copper simultaneously with the carbenoid unit. It is
unlikely that MeCN is solvated rather than metal
coordinated, as the ion persists undiminished even at
high cone voltages. It is presently unclear whether any
of these copper-containing ions are also responsible for
the observed transient red-orange color. We note how-
ever that the absorption shifts to purple in the case of
the para-substituted phenyl diazoester p-MeO-PhC-
(N2)CO2Me.
cone anglea
(deg)
yield
T (°C) (%)
ee
(%)
silane
øb
S/R
(b) Sila n es. An important manifestation of the
catalytic reaction is that a small amount of an orange-
brownish flocculent material is generated at room
temperature upon stirring the solution of the organosi-
lane and [CuI(MeCN)4](PF6)/R,R-1 for 5 min prior to
adding the diazoester. To gain better understanding of
this unanticipated reaction, [CuI(MeCN)4](PF6)/R,R-1
was mixed with 1 equiv of PhMe2Si-H in CD2Cl2, stirred
for an hour, and centrifuged, and the supernatant was
examined by 1H NMR, using hexamethylbenzene as
PhMe2SiH
PhMe2SiH
Ph2MeSiH
Ph2MeSiH
Ph3SiH
Ph3SiH
Et3SiH
Et3SiH
(n-Bu)3SiH
(n-Bu)3SiH
(n-Hex)3SiH
(n-Hex)3SiH
(i-Pr)3SiH
(i-Pr)3SiH
(i-Bu)3SiH
(i-Bu)3SiH
(Me3Si)3SiH
(Me3Si)3SiH
(t-Bu)2MeSiH
(t-Bu)2MeSiH
122
122
136
136
145
145
132
132
136
136
136
136
160
160
143
143
182
182
161
161
10.60
10.60 -40
0
88
85
79
77
87
47
90
87
83
7.61 76
10.77 83
6.70 74
11.00 84
2.89 49
6.07 72
6.99 75
10.51 83
9.87 82
15.44 88
10.77 83
12.70 87
3.64 57
12.10
12.10
13.25
13.25
6.30
6.30
5.25
5.25
5.00
5.00
3.45
3.45
5.70
5.70
0.80
0.80
2.85
2.85
0
-40
0
-40
0
-40
0
-40
84
87
73
0
-40
0
1
71
internal standard. These H NMR spectra indicate that
-40
NRc
94
N/A
N/A
all starting components of the mixture are retained
almost at the initial levels, apparently without any
detectable new features, suggesting that the observed
interaction is a low-yield event. Surprisingly, the floc-
culent precipitant can be resuspended in CH2Cl2 to
generate a mixture which can still mediate the insertion
of PhC(N2)CO2Me into the Si-H bond of PhMe2Si-H,
albeit at low yield (<10%) and selectivity (10% ee). The
supernatant is also catalytically active, providing com-
parable levels of stereocontrol versus the corresponding
catalytic reaction in which the precipitant has not been
centrifuged off. As aryl silanes are known to be good
hydride donors,32 these results raise suspicions whether
any copper hydride species formed in minute amounts
are participating in the catalytic cycle (vide infra).
(c) Oth er Sm a ll Molecu les. To assess the electron
donor ability of the copper site toward stabilizing the
carbenoid moiety, the electronically similar CO was
allowed to react with yellow solutions of [CuI(MeCN)4]-
(PF6)/R,R-1 in CH2Cl2. The solution turned pale yellow,
and colorless crystals of [CuI(R,R-1)(CO)](PF6), not
suitable for X-ray analysis, were deposited upon diffu-
sion of diethyl ether. The diagnostic νCO stretching
frequency at 2092 cm-1 indicates that the degree of Cu-
CO stabilization due to π-back-bonding is limited, as
expected for late transition elements. Apparently, the
Cu-CO bond is loosely held together owing to its
σ-bonding component or, as recently suggested,33 due
to a predominant electrostatic interaction operating in
this type of late transition metal cationic species. The
weakness of the Cu-CO interaction is also indicated
by the reversibility of the CO addition reaction under
vacuum or a stream of argon. The modest π-back-
donation is expected to be beneficial to the catalytic
properties of the copper carbenoid moiety, inasmuch as
the electrophilic nature of the metal bound carbene will
be preserved, and the electron density associated with
the Cu-C bond will be readily available for the Si-H
insertion process.
0
9.95 82
-40
0
-40
-10
-40
NR
45
NR
80
N/A
N/A
9.98 84
N/A
N/A
9.00 80
NR
N/A
N/A
a
b
Tolman’s cone angles.41 Electronic parameter ø.42 c NR )
no reaction.
Attempts to gain insight into the geometric features
and interactions of the Cu-C(CO2Me)Ph unit within the
ligand cavity, by means of installing a mimicking Cu-
S(CO2Et)Ph moiety, were not successful. Crystals ob-
tained by diffusion of diethyl ether into a solution of
[CuI(MeCN)4](PF6)/R,R-1 and PhSCO2Et in CH2Cl2 were
shown by X-ray analysis (see Supporting Information)
to be [CuI (R,R-1)3](PF6)2‚PhSCO2Et‚0.75CH2Cl2‚0.5CH3-
2
CN. The molecule features the same dinuclear copper
unit found in 2, and PhSCO2Et appears to be solvated
in the crystal lattice rather than coordinated to the
copper site.
Ca ta lytic Ca r ben oid Si-H In ser tion Rea ction s.
Yellow solutions of [CuI(MeCN)4](PF6) (10 mol %) and
R,R-1 (12 mol %) in CH2Cl2 mediate carbenoid insertion
of methyl phenyl diazoacetate to the Si-H bond of
organosilanes with good to high yields and levels of
enantiocontrol34 (Table 2). The absolute stereochemistry
of the major enantiomer obtained from the R,R-1 ligand
has been determined to be (S) by virtue of its transfor-
mation to the corresponding methyl-(S)-(+)-mande-
late.35 The catalytic reactions are performed in the
temperature range between 0 and -40 °C, since car-
benoid Si-H inserion reactions conducted below -40
°C proved to be impractically slow, whereas at temper-
atures above 0 °C the level of enantiocontrol is seriously
compromised. Moreover, use of [CuI (C6H6)(OTf)2]36
2
rather than [CuI(MeCN)4](PF6) provides comparable
(34) For reviews on the synthetic scope of chiral silane reagents
see: (a) Fleming, I.; Barbero, A.; Walter, D. Chem. Rev. 1997, 97, 2063-
2192. (b) Masse, C. E.; Panek, J . S. Chem. Rev. 1995, 95, 1293-1316.
(35) Fleming, I. Chemtracts Org. Chem. 1996, 9, 1-64.
(36) Salomon, R. G.; Kochi, J . K. J . Am. Chem. Soc. 1973, 95, 3300-
3310.
(32) Hays, D. S.; Fu, G. C. J . Org. Chem. 1997, 62, 7070-7071.
(33) Goldman, A. S.; Krogh-J espersen, K. J . Am. Chem. Soc. 1996,
118, 12159-12166.

2904 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
results in terms of yield and enantioselectivity, but the
air-sensitivity of the reagent makes its use less conve-
nient. In contrast, [CuII(OTf)2] markedly diminishes the
levels of enantiocontrol.37 Importantly, [CuI(MeCN)4]-
(PF6) (10 mol %) alone mediates carbenoid Si-H inser-
tion, as evidenced by the formation of the racemic
mixture of products derived from the carbenoid insertion
of PhC(N2)CO2Me into PhMe2Si-H. The reaction is
faster than the corresponding reaction in the presence
of ligand, as indicated by monitoring the product yields
of the ligand-free reaction (83% at 0 °C, 2 h) versus those
of the reaction in the presence of ligand (69%). However,
the final yield drops to less than 5% if the concentration
of [CuI(MeCN)4](PF6) is diminished to 1 mol %. Appar-
ently, the frequently observed “ligand acceleration”
phenomenon38 is not operative in the present instance.
It is also notable that the ligand R,R-1 has been selected
among a wide range of other aryl-substituted diimine
ligands (aryl ) phenyl, 1-naphthyl, 2,6-F2C6H3, 4-ClC6H4,
2,4,6-Me3C6H2, 3,5-(CF3)2C6H3, 2,5,6-Cl3C6H2) and is
remarkably the same as that which induces the highest
levels of enantioselectivity in cyclopropanation and
aziridination reactions.14,15 Moreover, the presence of
the aryl substituent on the carbenoid precursor has
proven to be essential in imparting high levels of
enantiocontrol. For instance, the vinyl diazoesters RCHd
CH(N2)CO2Me (R ) Me, Et, Ph) insert into the Si-H
bond of PhMe2Si-H with very low selectivities (ee ) 17-
36%).
Mech a n istic In vestiga tion s. (a ) Lin ea r ity of Cor -
r ela tion betw een ee of P r od u ct a n d ee of Liga n d .
The presence of potentially active copper sites in which
more than one exchangeable chiral ligand is coordinated
per copper ion or cluster (as for instance in [CuL2]+ or
[Cu2L3]2+) can lead, according to Kagan’s analysis,39 to
nonlinear correlations40 between the ee of the product
and the ee of the ligand (or auxiliary). Taking active
[CuL2]+ ions as an example and assuming that enan-
tiomerically impure ligand is employed (for instance if
L is a mixture of R,R-1 and S,S-1), nonlinear effects
would arise due to the presence of active or inactive
meso species [Cu(R,R-1)(S,S-1)]+. The sole exception
leading to linearity is the remote case of equal reactivity
between the meso species and its homochiral congeners,
yielding racemic and enantiomeric products, respec-
tively. Otherwise, “positive” deviation from linear cor-
relation between ee of the product and ee of the ligand
is expected if the reactivity of the meso species is inferior
to that of the homochiral catalysts (maximum deviation
is achieved in the event of an inactive meso species) and
“negative” deviation in the reverse scenario. Presum-
ably, “negative” nonlinear behavior can also be obtained
due to the presence of metal species (in the present case
[CuI(MeCN)4]+) which remain uncomplexed by the
F igu r e 4. Correlation of the enantioselectivity of car-
benoid PhC(N2)CO2Me insertion into the Si-H bond of
PhMe2Si-H as a function of the enantioselectivity of ligand
1 varied as noted in the figure. The catalyst is [Cu(CH3-
CN)4](PF6) (10%)/1 (12%) in CH2Cl2 at 0 °C.
chiral ligand and mediate the formation of racemic
product at rates superior to those associated with the
production of the enantiomerically enriched product. An
additional source of nonlinear behavior discussed by
Kagan39 consists of accumulation of catalytically inac-
tive metal reagents (usually products of polymerization)
complexed by chiral ligands that feature an overall ee
value that deviates from the nominal ee value of the
initially added ligand. Nonlinear correlation of ee of the
product versus the ee of the ligand would then arise
because the remaining pool of catalytically active metal
sites retains ligands with an “effective” ee that also
differs from the nominal ee value.
The linearity test is used in the present study as a
mechanistic criterion of the importance of copper-
containing species other than the mononuclear [(R,R-
1)Cu(CH3CN)n]+ ions in interfering with the outcome
of enantioselection. Those include ions that have been
structurally identified in the solid state and/or in
solution ([Cu(CH3CN)4]+, [Cu2(R,R-1)3]2+, [Cu(R,R-
1)2]+). Figure 4 indicates that, for a series of [CuI-
(MeCN)4](PF6)/1 reagents for which the ee of 1 was
intentionally varied from 0 to 100% by mixing appropri-
ate amounts of R,R-1 and S,S-1, the ee of the product
of PhC(N2)CO2Me insertion into PhMe2Si-H correlates
in a perfectly linear fashion (R2 ) 0.998) with the ee of
the ligand. This result is consistent with the exclusive
operation of mononuclear [Cu(1)]+-type units in solution
and diminishes the importance of any other copper-
containing species in conferring mechanistic complexity.
Presumably, these peripheral copper sites are present
in minute amounts and/or are unproductive and in any
event do not alter the enantiomeric integrity of the
ligand or product associated with or generated by the
catalytically active copper centers. A linear correlation
between ee of the product and ee of the ligand has been
previously reported14 for asymmetric cyclopropanation
and aziridination reactions catalyzed by the present Cu-
(I) diimine reagents.
(37) Carbenoid Si-H insertion reactions mediated by achiral Cu-
(II) reagents have been previously demonstrated: (a) Bagheri, V.;
Doyle, M. P.; Taunton, J .; Claxton, E. E. J . Org. Chem. 1988, 53, 6158-
6160. (b) Andrey, O.; Landais, Y.; Planchenault, D.; Weber, V.
Tetrahedron 1995, 51, 12083-12096.
(38) (a) J acobsen, E. N.; Marko´, I.; Mungall, W. S.; Schro¨der, G.;
Sharpless, K. B. J . Am. Chem. Soc. 1988, 110, 1968-1970. (b) J acobsen,
E. N.; Marko´, I.; France, M. B.; Svendsen, J . S.; Sharpless, K. B. J .
Am. Chem. Soc. 1989, 111, 737-739.
(39) Guillaneux, D.; Zhao, S.-H.; Samuel, O.; Rainford, D.; Kagan,
H. B. J . Am. Chem. Soc. 1994, 116, 9430-9439.
(40) Puchot, C.; Samuel, O.; Dun˜ach, E.; Zhao, S.; Agami, C.; Kagan,
H. B. J . Am. Chem. Soc. 1986, 108, 2353-2357.
(b) Ster eoelectr on ic Effects. Lin ea r F r ee-En er gy
Rela tion sh ip s. Inspection of the ee values for a series
of aryl and alkyl silanes (Table 2) reveals that the levels

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2905
F igu r e 5. Correlation of the reactivity of carbenoid p-X-
PhC(N2)CO2Me insertion into the Si-H bond of PhMe2Si-H
as a function of σpara values of the substituent X. The re-
activity is expressed as log of the ratio of carbenoid inser-
tion products of p-X-PhC(N2)CO2Me over PhC(N2)CO2Me.
F igu r e 6. Correlation of the enantioselectivity of car-
benoid p-X-PhC(N2)CO2Me insertion into the Si-H bond
of PhMe2Si-H as a function of σpara values of the substi-
tuent X. The enantioselectivity is expressed as log of the
ratio of the carbenoid insertion enantiomers (major over
minor).
of enantiocontrol are largely insensitive to the steric and
electronic properties of the silanes, inasmuch as the ee
values vary in a narrow range (72-88%) for silanes
exhibiting wide distribution of Tolman’s cone angles41
and electronic ø parameters.42 Two notable exceptions
are encountered with the silanes Ph3Si-H and (i-Pr)3Si-
H, for which the corresponding ee values (49 and 57%,
respectively) at 0 °C are markedly low. The low yields
obtained in the case of the good hydride donor (Me3-
Si)3SiH43 are associated with high yields of isolable
PhCH2CO2Me.
To closely investigate the electronic properties of the
transition state for the carbenoid Si-H insertion reac-
tion, a series of para-substituted aryl methyl diazoac-
etates (p-X-PhC(N2)CO2Me; X ) MeO, Me, H, F, CF3)
has been employed. A Hammett plot (Figure 5) of log-
(kX/kH) as a function of the characteristic σpara value for
each substituent affords a satisfactory linear correlation
(F ) -1.12, R2 ) 0.970), suggesting the presence of a
sizable positive charge on the carbenoid carbon in the
transition state. The ratio kX/kH is determined by
measuring relative product yields, assuming that the
reaction is under kinetic control. Surprisingly, the
corresponding plot (Figure 6) of log of the ratio of
enantiomers (S(major)/R(minor)) as a function of σpara
indicates that the levels of enantiocontrol are insensitive
to the electronic properties of the carbenoid intermedi-
ate.
The electronic properties of the transition state, as
influenced by the silane component, have also been
evaluated with the aid of a series of para-substituted
aryl silanes (p-X-PhMe2Si-H; X ) MeO, Me, H, F, CF3).
A Hammett plot (Figure 7) of log(kX/kH) (derived by
competition kinetics) as a function of σpara features a
reasonably linear relationship (F ) -0.54, R2 ) 0.960),
suggesting that a small partial positive charge resides
on the silicon atom as part of the transition state.
However, the corresponding diagram (Figure 8) of the
F igu r e 7. Correlation of the reactivity of carbenoid PhC-
(N2)CO2Me insertion into the Si-H bond of p-X-PhMe2Si-H
as a function of σpara values of the substituent X. The
reactivity is expressed as log of the ratio of carbenoid
insertion products of p-X-PhMe2Si-H over PhMe2Si-H.
log of the ratio of enantiomers (S(major)/R(minor)) as a
function of σpara reveals no correlation between the
electronic properties of the silane and the enantioselec-
tivity of the carbenoid insertion reaction into the Si-H
bond. Taken together, these Hammett correlations
suggest that stabilization of the positive charge devel-
oped on both the carbon atom of the carbenoid and the
silicon atom of the silane enhances the reactivity but
apparently does not influence the selectivity of the
process.
(c) Tem p er a tu r e Dep en d en ce of En a n tioselec-
tivity. To investigate whether the observed enantiose-
lectivity may be determined in a stepwise mechanism
(for instance, in two selection steps44 which may become
(44) (a) Haag, D.; Runsink, J .; Scharf, H.-D. Organometallics 1998,
17, 398-409. (b) Enders, D.; Gielen, H.; Breuer, K. Tetrahedron:
Asymmetry 1997, 8, 3571-3574. (c) Heller, D.; Buschmann, H.; Scharf,
H.-D. Angew. Chem., Int. Ed. Engl. 1996, 35, 1852-1854. (d) Go¨bel,
T.; Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1993, 32, 1329-
1331.
(41) Tolman, C. A. Chem. Rev. 1977, 77, 313-348.
(42) Bartik, T.; Himmler, T.; Schulte, H.-G.; Seevogel, K. J . Orga-
nomet. Chem. 1984, 272, 29-41.
(43) (a) Chatgilialoglu, C. Chem. Rev. 1995, 95, 1229-1251. (b)
Chatgilialoglu, C. Acc. Chem. Res. 1992, 25, 188-194.

2906 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
rates two linear regions, as evidenced in numerous other
cases of chemoselective systems,44 argues that deter-
mination of enantioselectivity in a single step cannot
be excluded with the present system. Ridd’s analysis46b
of certain combinations of activation parameters that
would afford linear plots of ln P versus 1/T in two-step
selection processes exemplifies the limitations of taking
the linearity of these plots as a strict mechanistic
criterion. However, if a one-step selection level is truly
operative, then the activation parameters ∆∆Hq and
∆∆Sq derived from the modified Eyring plot are mean-
ingful. The values calculated from the slope and inter-
cept of Figure 9 (∆∆Hq ) -1.58 ( 0.13 kcal mol-1; ∆∆Sq
) -1.82 ( 0.49 cal mol-1 K-1) indicate that the ener-
getic difference in the diastereotopic transition states
is subject to not only enthalpic but also significant entro-
pic discrimination. The latter may be better understood
in terms of an early, less structured, transition state.47
(d ) Deter m in a tion of th e P r im a r y Deu ter iu m
Kin etic Isotop e Effect (KIE) a n d Its Tem p er a tu r e
Dep en d en ce. To evaluate the involvement of carbenoid
Si-H insertion in the rate-limiting step and possibly
assess the relative geometry and position of the transi-
tion state with respect to Si-H bond activation, a pri-
mary KIE has been determined by allowing intermo-
lecular competition for carbenoid Si-H(D) insertion into
PhMe2Si-H/PhMe2Si-D (1:1) at 0 °C. The value so ob-
tained (kH/kD ) 1.29) is relatively small but significant.
In addition, the kH/kD value varies significantly with
temperature (from 2.12 (-40 °C) to 1.08 (25 °C) (vide
infra)), suggesting that Si-H bond activation is par-
ticipating in the “turnover-determining step”.48 These
values are consistent with other small KIE values re-
ported12g,13c,49 for rate-limiting Si-H or C-H insertion
reactions catalyzed by rhodium and copper reagents.
The closely related hydride transfer from silanes to car-
bocations also provides modest experimental50 or theo-
retically calculated KIE values51 (from 1.1 to 2.7) for
transition states varying widely in terms of Si-H (1.5-
2.5 Å) and C-H (3.0-1.1 Å) distances as well as C-H-
Si angles (129-180°). The higher end of the KIE value
range (2.0-2.7) has been computationally achieved51 for
acute C-H-Si angles and late transition states.
F igu r e 8. Correlation of the enantioselectivity of car-
benoid PhC(N2)CO2Me insertion into the Si-H bond of p-X-
PhMe2Si-H as a function of σpara values of the substituent
X. The enantioselectivity is expressed as log of the ratio of
the carbenoid insertion enantiomers (major over minor).
F igu r e 9. Eyring plot for the carbenoid PhC(N2)CO2Me
insertion into the Si-H bond of PhMe2Si-H mediated by
[Cu(CH3CN)4](PF6) (10%)/R,R-1 (12%) in CH2Cl2.
rate limiting in different temperature regions (Scarf
behavior45)), we examined the temperature dependence
of enantiopurity according to the modified Eyring equa-
tion:46
The geometry of the transition state in H-transfer
processes and possible elements of tunneling52 have
been occasionally assessed53 by examining the temper-
ature dependence of the KIE.54 A plot of ln(kH/kD) as a
function of 1/T based on the Arrhenius equation (eq 9)
for an elementary reaction step provides ratios of
Arrhenius preexponential factors AH/AD and differences
of activation energies (EH - ED), which, according to
ln P ) -∆∆Hq/RT + (∆∆Sq/R)
(8)
where P is the ratio of overall rate constants k(major
enantiomer)/k(minor enantiomer). For the [CuI(MeCN)4]-
(PF6)/R,R-1-mediated insertion of the carbenoid precur-
sor PhC(N2)CO2Me into the Si-H bond of PhMe2Si-H,
an Eyring plot (Figure 9) of ln(k(S)/k(R)) (calculated
from the ratio of amounts of major (S) over minor (R)
enantiomer, assuming kinetic control in the formation
of products) as a function of 1/T over the temperature
range from -40 to +40 °C reveals a satisfactory linear
relationship (R2 ) 0.960), although excellent nonlinear
fits to polynomial expressions can also be obtained. The
lack of an apparent inversion temperature which sepa-
(47) Palucki, M.; Finney, N. S.; Pospisil, P. J .; Gu¨ler, M. L.; Ishida,
T.; J acobsen, E. N. J . Am. Chem. Soc. 1998, 120, 948-954.
(48) Bosnich, B. Acc. Chem. Res. 1998, 31, 667-674.
(49) (a) Sulikowski, G. A.; Lee, S. Tetrahedron Lett. 1999, 40, 8035-
8038. (b) Demonceau, A.; Noels, A. F.; Costa, J .-L. J . Mol. Catal. 1990,
58, 21-26. (c) Baer, T. A.; Gutsche, C. D. J . Am. Chem. Soc. 1971, 93,
5180-5186.
(50) (a) Mayr, H.; Basso, N.; Hagen, G. J . Am. Chem. Soc. 1992,
114, 3060-3066. (b) Chojnowski, J .; Fortuniak, W.; Stanczyk, W. J .
Am. Chem. Soc. 1987, 109, 7776-7781.
(51) Apeloig, Y.; Merin-Aharoni, O.; Danovich, D.; Ioffe, A.; Shaik,
S. Isr. J . Chem. 1993, 33, 387-402.
(45) Buschmann, H.; Scharf, H.-D.; Hoffmann, N.; Esser, P. Angew.
Chem., Int. Ed. Engl. 1991, 30, 477-515.
(46) (a) Gypser, A.; Norrby, P.-O. J . Chem. Soc., Perkin Trans. 2
1997, 939-943. (b) Hale, K. J .; Ridd, J . H. J . Chem. Soc., Perkin Trans.
2 1995, 1601-1605.
(52) Bell, R. P. The Tunnel Effect in Chemistry; Chapman and
Hall: New York, 1981.
(53) Kwart, H. Acc. Chem. Res. 1982, 15, 401-408.
(54) Melander, L.; Saunders, W. H., J r. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; pp 44-45.

Carbenoid Insertion Reactions of Diazoacetates
Organometallics, Vol. 19, No. 15, 2000 2907
Sch em e 1
Unfortunately, all attempts toward further inter-
rogating the timing of the transition state and the mode
of Si-H activation (concerted insertion, radical H-atom
abstraction, hydride transfer) by determining primary
and most importantly secondary KIE values50a,58 for Ph2-
SiH2, Ph2SiHD, and Ph2SiD2 failed due to reduction of
the catalyst to copper metal in the presence of dihy-
drosilanes.
F igu r e 10. Arrhenius plot for the competitive carbenoid
PhC(N2)CO2Me insertion into the Si-H(D) bond of PhMe2-
Si-H/PhMe2Si-D (1:1) mediated by [Cu(CH3CN)4](PF6) (10%)/
R,R-1 (12%) in CH2Cl2.
(e) Con tr ol Exp er im en ts To Assess P ossible
Si-H Bon d Clea va ge P r ior to Ca r ben oid Ad d ition .
The visible interaction of silanes with solutions of the
copper catalyst in the absence of phenyl diazoacetate
noted above raised the question whether copper hy-
drides,59 formed via oxidative addition of silane48 or by
electrophilic hydride transfer to the copper site,49a,60
may be involved in the catalytic reaction. Generation
of a copper hydride reagent from the reaction of PhMe2-
SiH and CuCl in DMI (1,3-dimethylimidazolidinone) has
been recently reported61 to mediate conjugate reduction
of R,â-unsaturated carbonyl compounds.
Kwart,53 can be used as criteria to distinguish between
linear and nonlinear (bent) H-transfer transition states.
For instance, a bent transition state53 for H-transfer is
characterized by near temperature independence of KIE
(EH - ED ≈ 0) and an unusually high ratio of AH/AD
(g1.4). It has to be noted that the validity of these
correlations has been questioned.55 Especially high
values of AH/AD have been suspected55 to conceal
mechanistic complexity rather than characterize an
elementary step.
ln(kH/kD) ) ln(AH/AD) - (EH - ED)/(RT)
(9)
To evaluate the copper-hydride hypothesis as well
as any other mechanistic scenario that requires cleavage
of the Si-H bond prior to addition of the Si/H elements
to the carbenoid carbon, we performed a number of
control experiments. First, we note that the present
catalytic system failed to mediate hydrosilation (PhMe2-
SiH) of acetophenone, a reaction that most likely
depends on oxidative addition of the silane to the metal
center.2a,48,62 Most importantly, a catalytic reaction
(Scheme 1) mediated by [CuI(MeCN)4](PF6)/R,R-1, in-
volving PhC(N2)CO2Me and a mixture of PhMe2Si-D
and Ph2MeSi-H, showed no evidence of H/D scrambling
(1H NMR) in the insertion products PhCD(SiMe2Ph)-
CO2Me and PhCH(SiMePh2)CO2Me or in the remaining
starting silanes. Furthermore, a carbenoid insertion
reaction of PhC(N2)CO2Me into PhMe2Si-D in the pres-
ence of H2 takes place without H atom enrichment of
the C-D site in the product PhCD(SiMe2Ph)CO2Me.
These results strongly support the case for a concerted
carbenoid Si-H insertion reaction. However, the pos-
sibility still exists that an initial hydride transfer to the
KIE values have been determined for the intermo-
lecular carbenoid Si-H(D) insertion into PhMe2Si-H/
PhMe2Si-D (1:1) at five different temperatures (-40,
-30, -10, 0, +25 °C). A plot of ln(kH/kD) as a function
of 1/T affords a reasonably linear correlation (R2
)
0.978), from which values of AH/AD (0.11 ( 0.03) and
ED - EH (1.36 ( 0.12 kcal mol-1) have been extracted.
The value of AH/AD is not in the classical region for an
elementary H/D-transfer step (0.7 e AH/AD e 21/2),56
while the value of ED - EH is substantially larger by
comparison to the initial zero-point energy difference
(∆ZPE ) 0.87 kcal/mol) between PhMe2Si-H (νSi-H
)
2170 cm-1) and PhMe2Si-D (νSi-D ) 1560 cm-1), as
determined by IR. Both AH/AD and ED - EH values fall
in a region that has been discussed in the literature57
as indicative of tunneling effects (albeit also associated
with high KIE values) in H-transfer mechanisms.
However, due to the extrapolation involved (especially
in the determination of AH/AD), the validity of the
computed parameters needs to be further substantiated
for a wider temperature region and a variety of silanes.
Obviously, the possible interference of this quantum
mechanical effect would not permit precise definition
of the transition state.
(58) (a) Merrigan, S. R.; Le Gloahec, V. N.; Smith, J . A.; Barton, D.
H. R.; Singleton, D. A. Tetrahedron Lett. 1999, 40, 3847-3850. (b)
Singleton, D. A.; Thomas, A. A. J . Am. Chem. Soc. 1995, 117, 9357-
9358.
(59) (a) Mahoney, W. S.; Brestensky, D. M.; Stryker, J . M. J . Am.
Chem. Soc. 1988, 110, 291-293. (b) Lipshutz, B. H.; Ung, C. S.;
Sengupta, S. Synlett. 1989, 1, 64-66. (c) Tsuda, T.; Satomi, H.;
Hayashi, T.; Saegusa, T. J . Org. Chem. 1987, 52, 439-443.
(60) (a) Sulikowski, G. A.; Cha, K. L.; Sulikowski, M. M. Tetrahe-
dron: Asymmetry 1998, 9, 3145-3169. (b) Lee, S.; Lim, H.-J .; Cha, K.
L.; Sulikowski, G. A. Tetrahedron 1997, 53, 16521-16532.
(61) Ito, H.; Ishizuka, T.; Arimoto, K.; Miura, K.; Hosomi, A.
Tetrahedron Lett. 1997, 38, 8887-8890.
(55) Anhede, B.; Bergman, N.-Å. J . Am. Chem. Soc. 1984, 106,
7634-7636.
(56) Stern, M. J .; Weston, R. E., J r. J . Chem. Phys. 1974, 60, 2808-
2814.
(57) (a) Shaffer, M. W.; Leyva, E.; Soudararajan, N.; Chang, E.;
Chang, D. H. S.; Capuano, V.; Platz, M. S. J . Phys. Chem. 1991, 95,
7273-7277. (b) Kim, Y.; Kreevoy, M. M. J . Am. Chem. Soc. 1992, 114,
7116-7123. (c) Bosch, E.; Moreno, M.; Lluch, J . M. J . Am. Chem. Soc.
1992, 114, 2072-2076. (d) Sorokin, A.; Robert, A.; Meunier, B. J . Am.
Chem. Soc. 1993, 115, 7293-7299.
(62) (a) Noyori, R. Asymmetric Catalysis in Organic Syntheses; J .
Wiley & Sons: New York, 1994. (b) Ojima, I. Catalytic Asymmetric
Syntheses; VCH Publishers: Weinheim, Germany, 1993.

2908 Organometallics, Vol. 19, No. 15, 2000
Dakin et al.
Sch em e 2
a rather wide C-H-Si angle, with the bulky silyl group
positioned at the cavity’s wide opening opposite the
metal center. The early transition state better accom-
modates the insensitivity of ee values with respect to
the steric requirements of the silanes and also accounts
for the rather modest charge transfer from silicon to
carbon in conjuction with H transfer and the significant
entropic contribution to the energy difference of the
diastereomeric transition states.
The good to high levels of enantiocontrol observed
(usually associated with a late transition state) are most
likely due to the characteristics of the copper carbenoid
cavity, as dictated by the specific interactions of ligand
moieties and the carbenoid. The structure of the cavity
is thus considered to be sufficiently preorganized so that
it could largely direct the trajectory of the Si-H ap-
proach toward the most favorable diastereotopic region.
With the present combination of ligand and carbenoid
precursor, this implies effective blockage of one side of
the carbenoid moiety. In other combinations for which
ee values have been shown21 to drop precipitously, this
steric restriction is lifted so that the approaching silane
has the opportunity to effectively sample both sides of
the carbenoid. Consistent with this notion of high
specificity is the observation that the present catalyst
system is also the most effective in asymmetric cyclo-
propanation (and aziridination) reactions. Further defi-
nition of the copper carbenoid structure is thus neces-
sary to elucidate the origin of enantiocontrol. An
additional point to be addressed in future studies is the
stereochemistry of the C-Si bond formation in relation
to Cu-C bond cleavage, which is expected66 to proceed
via inversion of the conformation of the incipient ste-
reocenter developed upon hydride transfer to the car-
benoid carbon.
carbenoid carbon would generate a silicenium cation63
either as part of the transition state (nonsynchronous
concerted mechanism) or even as a discrete intermediate
(two-step mechanism) which could then recombine with
the carbon site prior to escaping. All attempts to capture
a potential silicenium cation by virtue of F- or [BF4]-
have been unsuccessful.64
Con clu sion s
Taken together, the mechanistic studies provide
evidence to support a reaction pathway65 characterized
by (i) rate-limiting carbenoid insertion into the Si-H
bond as suggested by the temperature-dependent KIE
values; (ii) a concerted insertion of the carbenoid into
the Si-H bond as invoked by control H/D scrambling
experiments and suggested by the temperature depen-
dence of the enantioselectivity (single-step enantiose-
lection level); (iii) a polar transition state featuring
positive charges on the carbon and to a lesser extent
silicon atom (Hammett plots on reactivity); (iv) insen-
sitivity of the levels of enantiocontrol with respect to
the electronic properties of the carbenoid and silane
moieties (Hammett plots on enantioselectivity), as well
as in regards to the steric properties of the silanes; and
(v) substantial dependence of ee values on the ligand
characteristics and the presence of the phenyl substitu-
ent on the carbenoid moiety.
Although the evidence presented above cannot unique-
ly reconstrust the geometric features of the transition
state, an attractive working hypothesis (Scheme 2)
includes a hydrogen-first approach of the modestly
polarized Siδ+-Hδ- vector toward the carbenoid carbon
atom from the open side of the cavity and in a direction
orthogonal to the electrophilic p-orbital residing on the
carbon. An early transition state is preferred, featuring
Ack n ow led gm en t. We thank Profs. W. P. Giering
and A. Prock for fruitful discussions, and Clementina
Reyes for valuable experimental assistance. Financial
support of this work by the U.S. Environmental Protec-
tion Agency (R823377-01-1 to P.S.) and the NIH/NCI
(RO1 CA56304 to J .S.P.) is gratefully acknowledged. A
J ames Flack Norris and Theodore William Richards
Undergraduate Summer Scholarship (P.C.O) from the
Northeastern Section of the American Chemical Society
is also acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Analytical data for
products of carbenoid insertions into Si-H bonds of silanes.
ORTEP diagrams and Tables S1-S15 containing listings of
crystal data and structure refinement, atomic coordinates and
equivalent isotropic displacement parameters, interatomic
distances and bond angles, anisotropic displacement param-
eters, and hydrogen coordinates and isotropic displacement
parameters for compounds [CuI(CH3CN)4](PF6)‚CH3CN, 2, and
(63) (a) Reed, C. A. Acc. Chem. Res. 1998, 31, 325-332. (b) Maerker,
C.; Schleyer, P. v. R. In The Chemistry of Organic Silicon Compounds;
Rappoport, Z., Apeloig, Y., Eds.; J ohn Wiley & Sons: New York, 1998;
Vol. 2, Part 1, pp 513-555. (c) Lickiss, P. D. Ibid.; pp 557-594. (d)
Lambert, J . B.; Zhao, Y. Angew. Chem., Int. Ed. Engl. 1997, 36, 400-
401. (d) Olah, G. A.; Rasul, G.; Li, X.-Y.; Buchholz, H. A.; Sanford, G.;
Prakash, G. K. S. Science 1994, 263, 983-984. (e) Pauling, L. Science
1994, 263, 983.
[CuI (R,R-1)3](PF6)2‚PhSCO2Et‚0.75CH2Cl2‚0.5CH3CN. This ma-
2
terial is available free of charge via the Internet at http://
pubs.acs.org.
OM0003786
(64) Prakash, G. K. S.; Wang, Q.; Li, X.-y.; Olah, G. A. New J . Chem.
1990, 14, 791-792.
(65) For an early mechanistic analysis of dichlorocarbene insertion
into Si-H bonds see: Seyferth, D.; Damrauer, R.; Mui, J . Y.-P.; J ula,
T. F. J . Am. Chem. Soc. 1968, 90, 2944-2948.
(66) (a) Brookhart, M.; Liu, Y.; Goldman, E. W.; Timmers, D. A.;
Williams, G. D. J . Am. Chem. Soc. 1991, 113, 927-939. (b) Brookhart,
M.; Liu, Y. J . Am. Chem. Soc. 1991, 113, 939-944. (c) Casey, C. P.;
Smith, L. J . Organometallics 1989, 8, 2288-2290.