Single-step hydride transfer from CpMo(CO)2(PPh3)H to protonated ketones

Article abstract of DOI:10.1021/om950984e

The only alcohol complex formed during the ionic hydrogenation of acetone by CpMo-(CO)₂(PPh₃)H (1b) and CF₃SO₃H in acetonitrile is trans-CpMo(CO)₂(PPh₃)(i-PrOH)⁺ (4); on standing the coordinated isopropyl alcohol of 4 is replaced by solvent, forming trans-CpMo-(CO)₂(PPh₃)(CH₃CN)⁺ (2b). Treatment of 1b in the same solvent with CF₃SO₃H alone yields the more stable cis acetonitrile complex 3b. The final products from treating an acetonitrile solution of acetone and 1b with CF₃SO₃H thus include both 2b and 3b. The trans stereochemistry of 4 implies that the ionic hydrogenation of acetone by 1b involves a single-step hydride transfer rather than separate e^- and H^? transfers. The equilibrium constant for the protonation of acetone by CF₃SO₃H in CH₃CN at 25 °C is 2.4 × 10^-2, and the rate constant for hydride transfer from 1b to Me₂COH⁺ is 12 300 M^-1 s^-1 under the same conditions. Other ketones undergo ionic hydrogenation more slowly.

Full text of DOI:10.1021/om950984e

Organometallics 1996, 15, 4515-4520
4515
Sin gle-Step Hyd r id e Tr a n sfer fr om Cp Mo(CO)2(P P h 3)H to
P r oton a ted Keton es
Kjell-Tore Smith,†,‡ J ack R. Norton,*,† and Mats Tilset‡
Departments of Chemistry, Colorado State University, Fort Collins, Colorado 80523-1872, and
University of Oslo, P.O. Box 1033 Blindern, N-0315 Oslo, Norway
Received December 27, 1995X
The only alcohol complex formed during the ionic hydrogenation of acetone by CpMo-
(CO)2(PPh3)H (1b) and CF3SO3H in acetonitrile is trans-CpMo(CO)2(PPh3)(i-PrOH)+ (4); on
standing the coordinated isopropyl alcohol of 4 is replaced by solvent, forming trans-CpMo-
(CO)2(PPh3)(CH3CN)+ (2b). Treatment of 1b in the same solvent with CF3SO3H alone yields
the more stable cis acetonitrile complex 3b. The final products from treating an acetonitrile
solution of acetone and 1b with CF3SO3H thus include both 2b and 3b. The trans
stereochemistry of 4 implies that the ionic hydrogenation of acetone by 1b involves a single-
step hydride transfer rather than separate e- and H• transfers. The equilibrium constant
for the protonation of acetone by CF3SO3H in CH3CN at 25 °C is 2.4 × 10-2, and the rate
constant for hydride transfer from 1b to Me2COH+ is 12 300 M-1 s-1 under the same
conditions. Other ketones undergo ionic hydrogenation more slowly.
In tr od u ction
Sch em e 1
M-H + E+ f M+ + E-H
Transfer of hydrogen from transition metals, as a
proton (H+), a hydrogen atom (H•), or a hydride (H-), is
an important step in numerous catalytic and stoichio-
metric processes.1 Apparent “hydride” transfer to pro-
tonated ketones, aldehydes, alkynes, and olefins is a
step in the “ionic hydrogenation” of these substrates;2
such transfers also appear to be a key step in the
catalytic reduction of nicotinamide cofactors3 and may
be involved in the hydrogenase-catalyzed transfer of
reducing equivalents from H2 to such species.4 How-
ever, it has been unclear whether these reactions occur
by single-step hydride transfer (Scheme 1) or by step-
wise transfer of an electron and a hydrogen atom
(Scheme 2). The same alternatives (removal of H- in a
single step or removal of H• after initial one-electron
oxidation) exist for the removal of “hydride” from organic
ligands.
Sch em e 2
M-H + E+ f M-H•+ + E•
M-H•+ + E• f M+ + E-H
reactions that occur when organometallic complexes are
treated with trityl reagents. In 1982 Hayes and Cooper
isolated a radical cation from the reaction of Cp2W-
(CH3)2 with Ph3C+.5a In 1985 Gladysz, Cooper, and co-
workers found that electron transfer from CpRe(NO)-
(PPh3)R to Ph3C+ was uphill;5b in 1987 Gladysz, Parker,
and co-workers trapped with O2 the trityl radicals
resulting from electron transfer from CpRe(NO)(PPh3)R
to Ph3C+ and established a stepwise (Scheme 2) mech-
anism for these “hydride” transfer reactions.5c More
recently, Cheng and Bullock have examined the reac-
tions of various hydride complexes with Ph3C+ and have
argued for a single-step (Scheme 1) mechanism from the
unfavorable equilibrium constants for electron transfer
and from the values (1.7-1.8) of kH/kD.5d
In 1994 Hembre and McQueen showed that “hydride”
transfer from Cp*(dppm)RuH to N-methylacridinium
occurred without reduction of methyl viologen dication
([MV]2+), despite the less negative potential of the latter;
they concluded that the transfer did not involve an
initial one-electron reduction and that its mechanism
was thus that given in Scheme 1.4b Recently Hembre,
McQueen, and Day have found that Cp*(dppf)RuH
undergoes facile one-electron oxidationsthe first step
of Scheme 2.4c
There have been several efforts at distinguishing
between these alternatives in the “hydride” abstraction
† Colorado State University.
‡ University of Oslo.
X Abstract published in Advance ACS Abstracts, September 1, 1996.
(1) (a) Transition Metal Hydrides; Dedieu, A., Ed.; VCH: New York,
1991. (b) Eisenberg, D. C.; Norton, J . R. Isr. J . Chem. 1991, 31, 55-
66. (c) Bullock, R. M. Comments Inorg. Chem. 1991, 12, 1. (d) Collman,
J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G. Principles and
Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987; pp 80-95, 381-389, and Chapters 10
and 13. (e) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis, 2nd
ed.; Wiley: New York, 1992; Chapters 2, 3, 4, and 8.
(2) (a) Song, J .-S.; Szalda, D. J .; Bullock, R. M.; Lawrie, C. J . C.;
Rodkin, M. A.; Norton, J . R. Angew. Chem., Int. Ed. Engl. 1992, 31,
1233. (b) Bullock, R. M.; Song, J .-S. J . Am. Chem. Soc. 1994, 116, 8602.
(c) Bullock, R. M.; Luan, L.; Song, J .-S. J . Org. Chem. 1995, 60, 7170.
(3) (a) Steckhan, E.; Herrmann, S.; Ruppert, R.; Dietz, E.; Frede,
M.; Spilka, E. Organometallics 1991, 10, 1568. (b) Ryabov, A. D.;
Menglet, D. L.; Levi, M. D. J . Organomet. Chem. 1991, 421, C16. (c)
Westerhausen, D.; Herrmann, S.; Hummel, W.; Steckhan, E. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1529.
Some of us have previously reported that “hydride”
transfer from CpM(CO)2(L)H (M ) W, L ) PMe3, 1a ;
(4) (a) Collman, J . P.; Wagenknecht, P. S.; Hembre, R. T.; Lewis,
N. S. J . Am. Chem. Soc. 1990, 112, 1294. (b) Hembre, R. T.; McQueen,
S. J . Am. Chem. Soc. 1994, 116, 2141. “dppm” ) Ph2PCH2PPh2. (c)
Hembre, R. T.; McQueen, J . S.; Day, V. W. J . Am. Chem. Soc. 1996,
118, 798. “dppf” ) 1,1′-bis(diphenylphosphino)ferrocene.
(5) (a) Hayes, J . C.; Cooper, N. J . J . Am. Chem. Soc. 1982, 104, 5570.
(b) Asaro, M. F.; Bodner, G. S.; Gladysz, J . A.; Cooper, S. R.; Cooper,
N. J . Organometallics 1985, 4, 1020. (c) Bodner, G. S.; Gladysz, J . A.;
Nielsen, M. F.; Parker, V. D. J . Am. Chem. Soc. 1987, 109, 1757. (d)
Cheng, T.-Y.; Bullock, R. M. Organometallics 1995, 14, 4031.
S0276-7333(95)00984-8 CCC: $12.00 © 1996 American Chemical Society

4516 Organometallics, Vol. 15, No. 21, 1996
Smith et al.
Sch em e 3
tion 1 is a true hydride transfer (Scheme 1). It is
reasonable to suggest that the intermediate in reaction
2 is the 17-electron CpM(CO)2L•, leading to the mech-
anism in eqs 6-8.
M ) Mo, L ) PPh3, 1b) to Ph2ArC+ (Ar ) p-MeOC6H4)
in MeCN gives mainly trans-CpM(CO)2L(NCMe)+ (2)
(reaction 1), while the oxidation of these hydrides by
single-electron reagents (e.g., Cp2Fe+) in the presence
of base gives mainly cis-CpM(CO)2L(NCMe)+ (3) (reac-
tion 2).6
We now suggest that this difference in stereochemical
outcome (2 vs 3) can be used to distinguish between
Scheme 1 and Scheme 2 mechanisms for other hydride
transfer reactions. The mechanism in Scheme 1 should
give 2, whereas the electron transfer mechanism (Scheme
2) should give 3.
Resu lts a n d Discu ssion
The cis cation 3 is more stable than the trans cation
2. The latter is converted to 3 by catalytic amounts of
a reducing agent (Cp2Co),6 implying the mechanism
shown in Scheme 3: the radical formed by reduction
(eq 3) undergoes facile isomerization to the cis stereo-
isomer (eq 4), which can reduce another 1 equiv of 2
(eq 5).
Rea ction of Cp Mo(CO)2(P P h 3)H (1b) a n d Ac-
eton e in th e P r esen ce of Acid . When the hydride
1b and acetone were dissolved in acetonitrile-d3 and
1
excess acid (either CF3SO3H or HBF4) was added, H
NMR showed the formation of isopropyl alcohol as well
as trans-CpMo(CO)2(PPh3)(NCMe)+ (2b), cis-CpMo(CO)2-
(PPh3)(NCMe)+ (3b), and another organometallic prod-
uct 4. When the solution was stored at ambient
temperature, the resonances of 4 disappeared at the
same rate that the resonances for isopropyl alcohol and
2b grew inswith a half-life of approximately 45 min.
Its 1H NMR spectrum showed that 4 contained
coordinated isopropyl alcohol; the formation of 4 is
parallel to the previously reported formation of CpW-
(CO)3(i-PrOH)+ from CpW(CO)3H and protonated
acetone.2a The 31P splitting (2.3 Hz) of the Cp 1H
resonance of 4 was characteristic of the trans isomer of
the four-legged piano-stool complex CpM(CO)2(PR3)X;7
the single methyl 1H doublet of the coordinated i-PrOH
showed that a plane of symmetry passed through it.
Compound 4 was thus trans-CpMo(CO)2(PPh3)(i-PrOH)+.
The subsequent conversion of 4 to 2b is merely the
displacement of isopropyl alcohol by acetonitrile, a
better donor ligand (Scheme 4); 2b and i-PrOH are
formed in a 1:1 ratio.
When 1b was dissolved in acetonitrile-d3 and treated
with excess acid, only 3b and H2 were observed. This
observation suggested that some of the 3b formed in
the 1b/acetone/acid reaction arose from direct reaction
of the acid with 1b. Two competing reaction pathways
have thus been drawn in Scheme 4: pathway A, forming
3b from 1b and acid, and pathway B, forming 4 (and
eventually 2b) from 1b, acetone, and acid.
As reaction 2 gives the thermodynamic product 3
while reaction 1 gives the kinetic product 2, we have
suggested6 that reaction 2 involves an intermediate
permitting equilibration of stereochemistry, while reac-
(6) (a) Ryan, O. B.; Tilset, M.; Parker, V. D. J . Am. Chem. Soc. 1990,
112, 2618. (b) Smith, K.-T.; Tilset, M. J . Organomet. Chem. 1992, 431,
55.
(7) Manning, A. R. J . Chem. Soc. A 1967, 1984.

Hydride Transfer from CpMo(CO)2(PPh3)H to Ketones
Organometallics, Vol. 15, No. 21, 1996 4517
Sch em e 4
F igu r e 1. Determination of Keq, the equilibrium constant
for protonation of acetone by triflic acid in CH3CN (see eq
14).
Keq
-
Me2CO + CF3SO3H y z Me2COH+ + CF3SO3 (9)
Dep en d en ce of P r od u ct Ra tio on Me2COH+ Con -
cen tr a tion . The [Me2COH+] involved in pathway B
should (because of the speed with which such protona-
tion equilibria become established) reflect the equilib-
rium extent of protonation of acetone. If Scheme 4 is
correct, and if the cis and trans isomers present8 in a
solution of 1b equilibrate rapidly, the product ratio (4
+ 2b)/3b should depend on the ratio of pathway B to
pathway A, i.e., on [H+] vs [Me2COH+]. At constant
acidity and excess acetone, the initial product ratio (4
+ 2b)/3b should be linear in [Me2COH+].
-
[Me2COH+][CF3SO3
]
Keq
)
(10)
[Me2CO][CF3SO3H]
A2
[Me2COH+] ≈ [CF3SO3-] ) 1 -
[Me2CO]0 (11)
(12)
(
)
A1
A
[Me2CO] ) 2[Me2CO]0
A1
Equ ilibr iu m for th e P r oton a tion of Aceton e by
Tr iflic Acid . The absorption maximum that acetone
shows at 274 nm disappears when it is protonated. This
fact was used to estimate the equilibrium constant Keq
for eq 9, the protonation of acetone by CF3SO3H in CH3-
CN. (A pKa value of 2.6 has been reported9 for CF3-
SO3H in that solvent.) If eqs 11-13 are substituted into
eq 10, the latter becomes eq 14. (A1 is the absorbance
when only acetone is present, and A2 is the absorbance
of the same solution after acid is added.)
The absorbances A1 and A2 were measured at 274,
290, and 300 nm for four different ratios of [Me2CO]0
to [CF3SO3H]0. A plot of the left vs right functions in
eq 14 (Figure 1) gave a straight line as predicted. The
slope implied Keq ) [2.4(4)] × 10-2, in reasonable
agreement with an early report from conductivity data
that the pKa of Me2COH+ in CH3CN is ∼-0.1.10 The
concentrations of acetone calculated from that value of
Keq implied an extinction coefficient of 16.0(3) M-1 cm-1
for acetone at 274 nm, a value close to that reported for
the n f π* transition of acetone in other solvents.11
Effect of [Me2COH+] on th e P r od u ct Ra tio fr om
Aceton e/1b/Acid . We were therefore able to calculate
[Me2COH+] from the amounts of acetone and CF3SO3H
A2
[CF3SO3H] ) [CF3SO3H]0 - 1 -
[Me2CO]0 (13)
(
)
A1
2
A2
A1
A2 [CF3SO3H]0
A2
1 -
) Keq
- 1 -
(14)
(
)
{ (
(
))}
A1
A1
[Me2CO]0
present in any reaction with 1b. At low [Me2COH+],
1
the H NMR measured ratios of 4 + 2b to 3b showed
(Figure 2) the expected linear dependence on [Me2-
COH+]; however, at sufficient high [Me2COH+] the ratio
of 4 + 2b to 3b leveled off to a constant value, 17(2).
The fact that the line for low Me2COH+ concentrations
went through the origin implied that, as assumed in
Scheme 4, path A gave only the cis 3b and none of the
trans 2b.
Th er m od yn a m ic Sta bility of 4 vs Its Cis An a log
5. Scheme 4 also assumes that only the trans alcohol
complex 4 is formed by hydride transfer from 1 to
protonated acetone. An alternative pathway B would
begin with the formation of 5, the cis analog of 4,
followed by rapid isomerization of 5 to 4. To check out
this possibility, 5 was prepared independently (eq 15)
by dissolving 1b in a 1:1 mixture of CH2Cl2 and
isopropyl alcohol and adding HBF4. The 1H NMR of the
product in CD2Cl2 showed that only cis 5 (and no trans
4) was formed. (Two signals were observed for the
coordinated i-PrOH methyl groups and a singlet for the
Cp resonance, both indicating the unsymmetric cis
structure.) However, when 5 was dissolved in CD3CN
(eq 16), the color immediately changed from red to
brown, and 1H NMR showed that 3b was the only
complex present.
(8) Faller, J . W.; Anderson, A. S. J . Am. Chem. Soc. 1970, 92, 5852.
(9) Fujinaga, T.; Sakamoto, I. J . Electroanal. Chem. Interfacial
Electrochem. 1977, 85, 185. As quoted by: Izutsu, K. Acid-Base
Dissocation Constants in Dipolar Aprotic Solvents; Blackwell: Oxford,
1990; p 28.
(10) Kolthoff, I. M.; Chantooni, M. K., J r. J . Am. Chem. Soc. 1973,
95, 8539.
(11) For example, in heptane the extinction coefficient is 15 M-1
cm-1 for the acetone absorption at 279 nm. Silverstein, R. M.; Bassler,
G. C.; Morrill, T. C. Spectrometric Identification of Organic Compounds,
5th ed.; Wiley: New York, 1991; p 296.

4518 Organometallics, Vol. 15, No. 21, 1996
Smith et al.
pseudo-first-order rate constants less than the first-
order rate constant for cis f trans isomerization of 1b.
For example, when [Me2COH+] ) 0.05 M, the first-order
rate constant for the disappearance of 1b by pathway
B would be 615 s-1 at 25 °C in CH3CN. The rate
constant for the isomerization of cis-1b to trans-1b is
1560 s-1 at the same temperature in CDCl3.8
Of course, the first-order rate constants for cis-1b h
trans-1b no longer exceed those for hydride transfer to
Me2COH+ at higher concentrations of that electrophile.
It seems likely that the trans isomer of 1b is more
reactive toward Me2COH+. In earlier papers6 we have
suggested on steric grounds (suppression of the reactiv-
ity of a cis hydride ligand by the large PR3) that only
the trans stereoisomer of a substituted hydride such as
1b can transfer hydride to Ph2ArC+. Cheng and Bullock
have shown that the trans isomer of CpMo(CO)2(PCy3)H
reacts with Ph3C+ at least 300 times more rapidly than
the cis isomer.5d The exclusive formation of the trans
alcohol complex 4 in pathway B is consistent with our
proposal that only the trans isomer of 1b reacts with
Me2COH+.
F igu r e 2. Dependence of (4 + 2b)/3b on [Me2COH+] when
1b was treated with excess acetone in the presence of CF3-
SO3H (CD3CN, 25 °C).
The fact that the plot in Figure 2 becomes level at
[Me2COH+] g 0.10 M confirms that cis f trans isomer-
ization of 1b can become rate-limiting (eq 17). Cheng
and Bullock have also identified this situation in the
reaction of CpMo(CO)2(PCy3)H with Ph3C+.5d
+
[Me2COH
cis-1b slow8 trans-1b
]8 4
(17)
fast
Wh y Ar e th e Cis Ca tion s 3b a n d 5 th e On ly
P r od u cts fr om th e Rea ction of 1b w ith Acid ? Much
evidence suggests that hydride complexes are proto-
nated more readily on the hydride ligand than on the
metal.14 The kinetic products of such protonations are
frequently dihydrogen complexes. There is, however,
no obvious reason why cis-1 should be protonated more
readily than trans-1. It is possible that trans-CpMo-
(CO)2(PPh3)(H2)+ isomerizes to cis-CpMo(CO)2(PPh3)-
(H2)+ faster than H2 is replaced by MeCN (Scheme 4)
or i-PrOH (eq 15). There is reason to believe the cis
dihydrogen complex would be more stable: an analogous
tungsten dihydride complex has the structure shown in
Figure 3.6a,15
Ion ic Hyd r ogen a tion of Oth er CdO Bon d s. Ex-
periments similar to those described for acetone were
also performed for other ketones and aldehydes. An
immediate reaction was observed in all cases when the
hydride 1b and a ketone or aldehyde 6 was dissolved
in CD3CN and triflic acid added (eq 18). 1H NMR
spectra showed that both 2b and 3b were formed, along
Thus, ligand exchange (replacement of isopropyl
alcohol by acetonitrile) is much faster (seconds instead
of hours) for 5 than for its trans analog 4. It is thus
unlikely that any 5 is formed during the reaction of 1b
with Me2COH+ (pathway B), because free isopropyl
alcohol is formed only in conjunction with 2b (1:1 ratio
observed); the formation of 5 would have led to ad-
ditional alcohol.
Why is ligand exchange for 5 so much faster than for
4? The likely explanation is steric. The alcohol ligand
in 5 is cis to the sterically demanding PPh3 ligand and
thus dissociates faster than the alcohol ligand in 4.
Kin etic Con sid er a tion s for th e F or m a tion of 4.
We have assumed that the interconversion of the cis and
trans isomers of 1b is fast relative to their reaction with
H+ (pathway A) or with Me2COH+ (pathway B). The
first-order rate constants for cis-1b h trans-1b are
known in CDCl3,8 and they should not change ap-
preciably with solvent.
Comparison requires knowing the absolute rate of
pathway A or B. We have therefore used stopped-flow
methods to examine the reaction of 1b with Me2COH+.
The UV-vis spectrum of the product 4 showed an
absorbance maximum at 486 nm that was easily moni-
tored. When 0.807 mM 1b was treated with 170 mM
acetone and 28.5 mM CF3SO3H in CH3CN at 25 °C, the
absorbance grew from zero to 0.21; the reaction showed
only a single exponential, with an observed first-order
(12) The observation of a single rate constant implies that “Boundary
Condition II” (conformational interconversion faster than conversion
of either stereoisomer to product) holds and that the situation can be
described by the Winstein-Holness equation kB(obs) ) (mole fraction
of 1 that is cis)[kB(cis)] + (mole fraction of 1 that is trans)[kB(trans)].
See section D1 in: Seeman, J . I. Chem. Rev. 1983, 83, 83.
(13) The reaction of i-PrCHO with CpMo(CO)3H in CD2Cl2 in the
presence of excess CF3CO2H has been shown to be first-order in
i-PrCHO and first-order in CpMo(CO)3H.2a
(14) (a) Chinn, M. S.; Heinekey, D. M. J . Am. Chem. Soc. 1987, 109,
5865. (b) Chinn, M. S.; Heinekey, D. M. J . Am. Chem. Soc. 1990, 112,
5166. (c) J ia, G.; Morris, R. H. J . Am. Chem. Soc. 1991, 113, 875. (d)
Kristja´nsdo´ttir, S. S.; Norton, J . R. In ref 1a, Chapter 9. (e) Parkin,
G.; Bercaw, J . E. J . Chem. Soc., Chem. Commun. 1989, 255. (f) Hamon,
P.; Toupet, L.; Hamon, J .-R.; Lapinte, C. Organometallics 1992, 11,
1429.
rate constant of 108 s-1 12
When the reaction was
.
assumed to be first-order in Me2COH+,13 a second-order
rate constant of 12 300 M-1 s-1 was deduced from the
calculated [Me2COH+] of 8.76 mM.
In the linear portion of Figure 2, [Me2COH+] < 0.10
M, this second-order rate constant would have produced
(15) Bullock, R. M.; Song, J .-S.; Szalda, D. J . Organometallics 1996,
15, 2504-2516.

Hydride Transfer from CpMo(CO)2(PPh3)H to Ketones
Organometallics, Vol. 15, No. 21, 1996 4519
Rea ction of 1b w ith P r oton a ted Aceton e. In a typical
experiment 1b (6.0 mg, 0.0125 mmol) was dissolved in CD3-
CN (0.5 mL) in an NMR tube. The tube was sealed by a
septum, and acetone (8 µL, 0.1088 mmol) and triflic acid (30
µL, 0.341 mmol) were added by syringe. The solution im-
mediately turned red. The 1H NMR spectrum was recorded
immediately (within 10 min) and showed 3b (9%), 2b (13%),
and 4 (78%) as products. 1H NMR (CD3CN): 3b-d3, δ 5.65 (s,
F igu r e 3. Structure (from ref 15) of the dihydride CpW-
+
5H);6b 2b-d3, δ 5.35 (d, J PH ) 1.8 Hz, 5H);6b 4, δ 5.37 (d, J PH
)
(CO)2(PMe3)H2
.
2.3 Hz, 5H), 4.25 (d, J HH ) 7.8 Hz, 1H), 3.56 (doublet of septet,
J HH ) 7.8 Hz, J HH ) 6.4 Hz, 1H), 1.16 (d, J HH ) 6.4 Hz, 6H).
Free isopropyl alcohol was also observed at δ 4.62 (septet, J HH
) 6.1 Hz, 1H) and 1.39 (d, J HH ) 6.1 Hz, 6H).
P r oton a tion of 1b. The hydride 1b (5.8 mg, 0.012 mmol)
was dissolved in CD3CN (0.5 mL) in an NMR tube, and triflic
acid (1.2 µL, 0.014 mmol) was added. Gas evolution (H2) was
observed. 1H NMR (CD3CN) showed 3b-d3 as the only
product: δ 7.2-7.7 (m, 15 H), 5.65 (s, 5 H).6b A signal for
dissolved H2 was also observed at δ 4.56 (s).
Equ ilibr iu m Con sta n t Deter m in a tion for th e Equ ilib-
r iu m betw een Aceton e a n d Tr iflic Acid . Four different
solutions of acetone (3, 3, 4, and 5 µL, respectively) in 3 mL of
acetonitrile were prepared, and the UV spectra were recorded
for each solution. Triflic acid (20, 30, 20, and 20 µL, respec-
tively) was added, and the UV spectra were recorded again.
The absorbances A1 (without acid) and A2 (with acid) were
measured at three different wavelengths (274, 290, and 300
nm, respectively) for each of the four solutions; the decrease
in absorbance was typically between 25 and 35%. The results
from this experiment are shown in Figure 1 and give the
equilibrium constant as 2.4(4) × 10-2; the intercept is zero
within experimental error.
Dep en d en ce of P r od u ct Ra tio fr om Aceton e/1b/Acid
on [Acet on e]. Samples of 1b (6.0 mg, 0.0125 mmol each)
were dissolved in CD3CN (0.5 mL each) in NMR tubes.
Acetone (varying from 1 to 150 µL) was added to the tubes;
then triflic acid (30 µL, 0.341 mmol) was added. The tubes
were stored in liquid nitrogen until their NMR spectra were
recorded. The ratios (4 + 2b)/3b were determined by integra-
tion of the peaks. The results are given in Figure 2. When
[Me2COH+] was <0.075 M, the ratio (4 + 2b)/3b showed a
linear dependence on [Me2COH+]; linear regression gave a
slope of 213(13) and an intercept of -0.9(0.6).
Syn th esis of cis-5. The hydride 1b (50.0 mg, 0.104 mmol)
was dissolved in a mixture of dichloromethane (3 mL) and
isopropyl alcohol (3 mL). HBF4‚Et2O (250 µL, 1.83 mmol) was
added, and the solution was stirred for 20 min. Some of the
solvent was removed under vacuum, and diethyl ether was
added until the red product precipitated (34.2 mg, 0.055 mmol,
53%). 1H NMR (CD2Cl2): δ 5.70 (s, 5 H), 3.00 (m, 1 H), 2.68
(d, J HH ) 3.2 Hz, 1 H), 1.06 (d, J HH ) 6.3 Hz, 3H), 0.36 (d, J HH
) 6.2 Hz, 3H). When this compound was dissolved in CD3-
CN, it reacted rapidly with that solvent to form 3b.
with trans alcohol complexes 7 analogous to 4. How-
ever, the cis CD3CN complex 3b was the main product,
formed in yields between 55 and 65% in all cases. The
coordinated alcohols in 7a and 7b were replaced by
acetonitrile more rapidly (at ambient temperature t1/2
) 6 and 17 min, respectively) than the isopropyl alcohol
in 4, while the coordinated alcohol in 7c was replaced
more slowly (at ambient temperature t1/2 ) 68 min) than
the isopropyl alcohol of 4. The exchange rates probably
reflect the steric bulk of the alcohol ligands and its effect
on the ease with which they dissociate.
The formation of more 3b than with acetone implies
that for these ketones pathway A is substantially faster
than pathway B. Presumably hydride transfer from 1b
is slower to the cations from 6 than to protonated
acetone.
Con clu sion
Stop p ed -F low Obser va tion of th e Rea ction of 1b w ith
P r oton a ted Aceton e. One solution contained 1b (15.5 mg,
0.0323 mmol) and acetone (500 µL, 6.80 mmol) in acetonitrile
(20.0 mL); the other contained triflic acid (100 µL, 1.14 mmol)
in acetonitrile (20.0 mL). These solutions were placed in two
reservoirs in a High-Tech SF-41 Canterbury stopped-flow
apparatus (equipped with an RSM-1000 scanning system from
OnLine Instruments). An increase in absorbance of up to 0.21
was observed when these solutions were mixed. The pseudo-
first-order rate constant (kobs ) kB[Me2COH+]) was determined
for eight different “shots” of 500 µL (250 µL from each solution);
The fact that only trans alcohol complexes are formed
by pathway B in Scheme 4 implies that these reactions
are true hydride transfers (Scheme 1), rather than
electron transfers followed by hydrogen atom transfers
(Scheme 2).
Exp er im en ta l Section
Gen er a l P r oced u r es. All manipulations involving inor-
ganic and organometallic compounds were carried out by
vacuum line, Schlenk, syringe, or inert atmosphere box
techniques. Acetonitrile was distilled from P4O10; CD3CN and
CD2Cl2 were vacuum-transferred from P4O10; ether was dis-
tilled from sodium benzophenone ketyl; acetone was stored
over MgSO4.
the average was 108 s-1
.
The calculated [Me2COH+] of 8.7
mM gave a second-order rate constant (kB) of 12 300 M-1 s-1
.
Rea ction betw een 1b a n d P h 2COH+. An NMR tube was
charged with 1b (6.2 mg, 0.013 mmol) and benzophenone (6a ;
1
Residual H shifts in the deuterated solvents were used as
(16) (a) King, R. B.; Stone, F. G. A. Inorg. Synth. 1963, 7, 99. (b)
Asdar, A.; Tudoret, M. J .; Lapinte, C. J . Organomet. Chem. 1988, 349,
353. (c) Kalck, P.; Poilblanc, R. C. R. Seances Acad. Sci., Ser. C 1968,
267, 536.
internal standards in the 1H NMR spectra. CpMo(CO)2-
(PPh3)H (1b) was prepared according to published proce-
dures.16

4520 Organometallics, Vol. 15, No. 21, 1996
Smith et al.
4.3 mg, 0.024 mmol) and dissolved in CD3CN (0.5 mL). Triflic
acid was added (2.0 µL, 0.023 mmol), and the color im-
mediately changed from light yellow to dark red. The 1H NMR
spectrum (after 10 min) showed that 3b (63%) was the main
product. Also observed were 2b (29%) and 7a (8%). 1H NMR
for 7a (CD3CN): δ 5.39 (d, J HP ) 2.3 Hz, 5 H). The signals
for coordinated diphenylmethanol were hidden under the
phenyl signals between 7.2 and 7.9 ppm. Free diphenyl-
methanol was observed with one signal at δ 6.30 (d, J HH ) 7.8
Hz, 1 H). The coordinated diphenylmethanol in 7a exchanged
with CD3CN (t1/2 ) 6 min) to form 2b.
Rea ction betw een 1b a n d (p-MeO-C6H4)(H)COH+. An
NMR tube was charged with 1b (6.6 mg, 0.014 mmol) and (p-
MeO-C6H4)(H)CO (6c; 5.5 µL, 0.045 mmol) dissolved in CD3-
CN (0.5 mL). Triflic acid was added (4.0 µL, 0.046 mmol), and
the color immediately changed from light yellow to dark red.
The 1H NMR spectrum (after 10 min) showed that 3b (56%)
was the main product. Also observed were 2b (14%) and 7c
(30%). 1H NMR for 7c (CD3CN): δ 5.34 (d, J HP ) 2.4 Hz, 5
H), 4.52 (d, J HP ) 5.9 Hz, 1 H), 3.79 (s, 3 H). The signal for
the methoxy group of free alcohol was observed at δ 3.73 (s, 3
H). The other signals for coordinated and free alcohol were
hidden under the phenyl signals between δ 7.0 and 7.6.
Exchange of the coordinated alcohol in 7c with CD3CN formed
2b-d3 with t1/2 ) 68 min.
Rea ction betw een 1b a n d (p-MeO-C6H4)(Me)COH+. An
NMR tube was charged with 1 (7.2 mg, 0.015 mmol) and (p-
MeO-C6H4)(Me)CO (6b; 3.5 mg, 0.023 mmol) and dissolved in
CD3CN (0.5 mL). Triflic acid was added (2.0 µL, 0.023 mmol),
and the color immediately changed from light yellow to dark
red. After 10 min the 1H NMR spectrum showed that 3b (66%)
was the main product. Also observed were 2b (14%) and 7b
(20%). 1H NMR for 7b (CD3CN): δ 5.30 (d, J HP ) 2.0 Hz, 5
H), 4.8 (d, 1 H), 3.82 (s, 3 H). The signal for the methoxy group
for free alcohol was observed at δ 3.74 (s, 3 H). The other
signals for coordinated and free alcohol were hidden under the
phenyl signals between δ 7.0 and 7.6. Exchange of the
coordinated alcohol in 7b with CD3CN formed 2b-d3 with t1/2
) 17 min.
Ack n ow led gm en t. We gratefully acknowledge sup-
port from Statoil under the VISTA program, adminis-
tered by the Norwegian Academy of Science and Letters
(stipend to K.-T.S.), and from the U.S. National Science
Foundation (Grant CHE-9531751). We are grateful to
Dr. Morris Bullock for helpful discussions.
OM950984E