Aqueous organometallic chemistry. 3. Catalytic hydride transfer reactions with ketones and aldehydes using [Cp*Rh(bpy)(H2O)](OTf)2 as the precatalyst and sodium formate as the hydride source: Kinetic and activation parameters, and the significance of steric and electronic effects

Article abstract of DOI:10.1016/j.jorganchem.2009.10.015

The reaction of a precatalyst, [Cp*Rh(bpy)(H₂O)](OTf)₂ (1), with sodium formate provided the hydride complex, [Cp*Rh(bpy)(H)]⁺ (2), in situ, at pH 7.0, which was then evaluated in an aqueous, catalytic hydride transfer process with water soluble substrates that encompass 2-pentanone (3), cyclohexanone (4), acetophenone (5), propionaldehyde (6), benzaldehyde (7), and p-methoxybenzaldehyde (8). The initial rates, r_i, of appearance of the reduction product alcohols at 23 °C provided a relative rate scale: 8 > 7 ≈ 6 > 5 > 4 > 3, while the effect of concentration of substrate, precatalyst, and sodium formate on r_i, using 7 as an example, implicates [Cp*Rh(bpy)(H)]⁺ formation as the rate-limiting step. The experimental kinetic rate expression was found to be: d[alcohol]/dt = k_cat[1][HCO₂Na]; substrate being pseudo zero order in water. The steric effects were also analyzed and appeared to be of less importance intra both the ketone and aldehyde series, but an inter series comparison appeared to show that the aldehydes had less of a steric effect on the initial rate, i.e., 7 > 4 by a factor of 3.6, while the aldehyde series appeared to have some moderate electronic influence on rates, presumably via electron donation to increase binding to the Cp*Rh metal ion center, in accordance with these proposed concerted binding/hydride transfer reactions. A proposed catalytic cycle will also be presented.

Full text of DOI:10.1016/j.jorganchem.2009.10.015

Journal of Organometallic Chemistry 695 (2010) 145–150
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Aqueous organometallic chemistry. 3. Catalytic hydride transfer reactions
with ketones and aldehydes using [Cp*Rh(bpy)(H O)](OTf) as the precatalyst
2
2
and sodium formate as the hydride source: Kinetic and activation parameters,
and the significance of steric and electronic effects
,
1
Carmen Leiva, H. Christine Lo, Richard H. Fish *
Lawrence Berkeley National Laboratory, University of California, Berkeley, CA 94720, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The reaction of a precatalyst, [Cp*Rh(bpy)(H O)](OTf) (1), with sodium formate provided the hydride
complex, [Cp*Rh(bpy)(H)] (2), in situ, at pH 7.0, which was then evaluated in an aqueous, catalytic
hydride transfer process with water soluble substrates that encompass 2-pentanone (3), cyclohexanone
2
2
Received 17 September 2009
Received in revised form 13 October 2009
Accepted 14 October 2009
Available online 25 October 2009
+
(
4), acetophenone (5), propionaldehyde (6), benzaldehyde (7), and p-methoxybenzaldehyde (8). The ini-
tial rates, r , of appearance of the reduction product alcohols at 23 °C provided a relative rate scale:
> 7 ꢀ 6 > 5 > 4 > 3, while the effect of concentration of substrate, precatalyst, and sodium formate on
i
8
r
Keywords:
+
i
, using 7 as an example, implicates [Cp*Rh(bpy)(H)] formation as the rate-limiting step. The experimen-
Aqueous organometallic chemistry
Kinetics and activation parameters
Mechanisms
tal kinetic rate expression was found to be: d[alcohol]/dt = kcat[1][HCO Na]; substrate being pseudo zero
2
order in water. The steric effects were also analyzed and appeared to be of less importance intra both the
ketone and aldehyde series, but an inter series comparison appeared to show that the aldehydes had less
of a steric effect on the initial rate, i.e., 7 > 4 by a factor of 3.6, while the aldehyde series appeared to have
some moderate electronic influence on rates, presumably via electron donation to increase binding to the
Cp*Rh metal ion center, in accordance with these proposed concerted binding/hydride transfer reactions.
A proposed catalytic cycle will also be presented.
Ó 2009 Elsevier B.V. All rights reserved.
1
. Introduction
alytic hydride transfer reactions, and furthermore, to test our re-
cently published mechanism that demonstrated the importance
5-pentam-
-(N,N)-2,2 -bipyridine)(aqua)]
O)](OTf) (1), as a precatalyst for
of coordination of the substrate to the
g
5
Recently, we demonstrated the usefulness of the [
g
-Cp*Rh metal center dur-
2
0
ethylcyclopentadienyl)rhodium(
g
ing the concerted hydride transfer process [2].
(
OTf)
2
complex, [Cp*Rh(bpy)(H
2
2
the regioselective reduction of N-benzylnicotinamide triflate, an
NAD model, in the presence of sodium formate as the hydride
2+
+
+
source, using 1:1 H
aqueous NAD model, b-nicotinamide ribose-5 -methyl phosphate,
2
O/THF as one solvent system, as well as an
+
0
Rh
Rh
N
OH2
N
H
at pH 6.5, to exclusively provide the kinetic 1,4-dihydro product
N
N
[
1,2]. An extensive kinetic and mechanistic study also showed that
the 1,4-dihydro regioselectivity was a consequence of binding of
the 3-amide carbonyl group to the Cp*Rh metal ion center [2]. In
an effort to extend the scope of aqua complex 1 in catalytic reac-
tions, we have studied its reactivity to ascertain whether the
in situ generated hydride, [Cp*Rh(bpy)(H)] (2), formed by reaction
of 1 with sodium formate, would have utility in other aqueous, cat-
1
2
+
Previous studies of organometallic hydride complexes in aqueous,
and/or biphasic media reduction reactions with water soluble/insol-
uble substrates, such as substituted olefins, aldehydes, ketones, and
halogen substituted carboxylic acids, have been reported by various
groups, with a focus on structure-activity relationships, while the
evaluation of kinetic, thermodynamic, steric, and electronic param-
eters were still aspects that needed further scrutiny [3]. We also
preface our remarks about the previous studies with the fact that
*
Corresponding author. Fax: +1 510 486 7303.
E-mail address: rhfish@lbl.gov (R.H. Fish).
RHF would like to dedicate this contribution on aqueous organometallic
1
chemistry to Ferenc Joó, friend and colleague, in honor of his 60th birthday.
0
022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.10.015

1
46
C. Leiva et al. / Journal of Organometallic Chemistry 695 (2010) 145–150
precatalyst 1 had been tested in the direct reduction of pheneth-
ylmethyl ketone in water, in the presence of sodium formate as
the hydride source, by Steckhan and coworkers [4a]. However, in
those reported studies, the ketone reduction reaction was actually
a ‘‘control experiment” for the main enzymatic reactions using
Therefore, we present our recent results on the use of in situ
generated 2 in catalytic hydride transfer reactions that provided
a structure-activity relationship with regard to ketone and alde-
hyde substrates, and delved into factors such as steric and elec-
tronic parameters to observe any conceivable effects on the
initial rates of reduction. We also studied the kinetics and activa-
tion parameters of this reaction with regards to concentrations of
substrate, precatalyst 1, and sodium formate, on the initial rates
of reaction, to provide insights into the mechanism of reduction,
and to present a plausible catalytic cycle for the reduction of the
ketone and aldehyde substrates.
+
in situ generated 1,4-NADH, from the reduction of natural NAD
with 2 [4b], to affect ketone reduction reactions in the presence
of the enzyme, horse liver alcohol dehydrogenase (HLADH), and
provide chiral alcohols. This latter study [4a] showed that the direct
ketone reduction with 2 was not favored in the presence of the bet-
+
ter Cp*Rh binding substrate, NAD , and that HLADH enzymatic ke-
tone reductions to chiral alcohols prevailed (>99% ee); significant
direct reduction of the ketone substrate by 2 would have compro-
mised the % enantioselectivity, but these authors found no signifi-
cant direct hydride reduction of their ketone substrate (<7% with
phenethylmethyl ketone) in their control experiments, and thus,
no further experiments on any direct hydride reductions with
in situ generated 2 and aldehydes/ketones in water were reported
by those authors.
2
. Results and discussion
2.1. Structure-activity relationships: steric and electronic effects
The reactivity of in situ generated 2 in catalytic hydride transfer
reactions was evaluated in aqueous solution at pH 7.0 (buffered) at
2
3 °C with various water soluble ketone and aldehyde substrates
More importantly, in our own independent studies using biomi-
+
that encompassed: 2-pentanone (3), cyclohexanone (4), acetophe-
none (5), propionaldehyde (6), benzaldehyde (7), and p-methoxy-
benzaldehyde (8) (Chart 1), using GC and GC/MS techniques to
follow the initial rates/turnover frequencies, and establish a struc-
ture-reactivity relationship (Table 1).
metic model NAD compounds with in situ generated 2 in water, in
similar chiral reductions of ketones with HLADH [5], we found that
the direct reduction ‘‘control experiments” of a variety of ketones
+
were in fact favored in the absence of NAD or their biomimetic mod-
els. Therefore, as stated, we wanted to expand the scope of aqueous
reduction chemistry of 2 by evaluating structure-activity relation-
ships, mechanistic details, and determine kinetic, thermodynamic,
steric, and electronic parameters, in this example, by studying a
variety of aldehydes and ketones in water at pH 7.0, and at RT.
The initial rates of appearance of the reduction product alco-
ꢁ
1
hols, r
i
(M s ), provided a relative rate scale: 8 > 7 ꢀ 6 > 5 > 4 > 3,
and thus, it appeared that steric effects were more pronounced
with the ketone series in comparison to the aldehyde series, while
the electronic effects with aldehydes also appeared to influence
their reactivity over the ketones; aldehyde 8, 6 times faster than
ketone 4, and 1.5 times faster than 7, presumably due to the elec-
tron-donating p-methoxy group on the benzene ring that appar-
ently moderately increased carbonyl group binding to the
Cp*Rh metal ion center. This binding phenomena of substrate to
the
cial for the regioselective reduction of various NAD models, N-
Moreover, several earlier reports, for example, with in situ
+
formed [(Cp*Ir)
(l-H)(
l
-OH)(
l
-HCOO)] showed catalytic ketone
2
and aldehyde substrate reduction ability [3f–h]. These initial
studies by Ogo et al. were conducted at a low pH (3.2) [3f,g],
5
1
g
-
where protonation of the in situ formed Cp*Ir-
l
or
g
-H to pro-
duce H
2
could possibly compete with hydride transfer to sub-
5
g
-Cp*Rh metal ion center was also found recently to be cru-
strate, thereby mitigating the carbonyl reduction reaction, as
well as, from further pH mediated equilibria with, for example,
+
2+
benzyl-3-substituted pyridium triflate salts in 1:1 THF/H O with
2
bose-5 -methyl phosphate [2].
a catalyst precursor, such as [Cp*Ir(H
O)
]
. Moreover, those ini-
2
2
3
+
[1,2], as well as, an aqueous NAD model, b-nicotinamide ri-
tial studies were not conducted at high concentrations of formate
ion (saturation kinetics), which further limited catalytic activity,
and inhibited mechanistic interpretations, including understand-
ing the importance of binding of the substrate to the Cp*Ir metal
ion center [1,2]. More recently, Ogo et al. [3i,j] did study various
0
Furthermore, as stated, aryl aldehyde, 8, was slightly faster
than aryl aldehyde, 7, as well as alkyl aldehyde, 6; therefore, alde-
hydes, 6–8, were ꢂ2 to ꢂ20 times more reactive than the one aryl
and two alkyl ketones studied. Moreover, in the ketone series, the
relative rate sequence, 5 > 4 > 3, also defined that steric effects
were not significant and that electronic effects were relatively
more important, although the sterically less demanding alkyl cyc-
lic ketone, 4, was somewhat faster in the reduction reaction over
alkyl ketone, 3, by a relative rate factor of 3.6. The electronic effect
of the aryl group directly bonded to the carbonyl appeared to
moderately affect the relative rates for the series of ketones,
which, in our opinion, implies phenyl group donation to increase
electron density at the carbonyl oxygen. Thus, this electronic ef-
fect was less effective for the aldehyde series in going from 6 to
6
+
parameters mentioned above with both [
6 6
g -C Me )Ru(bpy)(H)]
and [Cp*Ir(bpy)(H)]+ hydride complexes, and various ketones at
pH 4.0 and 2.0, respectively, from 25 to 70 °C, where the higher
temperatures caused turnover numbers to be greater, and steric
and electronic effects were not dominant. These authors preferred
a mechanism where the substrate was not directly binding to the
Cp*Ir metal ion center; but rather, acid promoted transfer hydro-
genation with no metal–substrate involvement. Furthermore, pre-
vious studies by Darensbourg et al. with an aqueous phase
4 2
soluble (PTA) RuCl precatalyst, for reductions (sodium formate)
of aromatic and aliphatic aldehydes (biphasic reaction conditions)
also showed no apparent steric or electronic effects, including
other data on inhibition by CO of the reduction reaction, which
to our thinking, may have a potential mechanism that included
binding of the carbonyl substrate to the Ru(ll) metal ion center
O
CH3CH2CH2CCH3
CCH3
[
2
3k]. Kuo et al. have studied [Cp MoH(OTf)] in aqueous reductions
O
3
O
4
5
of acetone and substituted derivatives [3n]. In their postulated
mechanism, binding of the ketone to the Mo metal ion center oc-
curred with hydride transfer, while electron-withdrawing groups
on the acetone nucleus structure facilitated hydride transfer to
the carbonyl carbon atom. An excellent overview on transition
metal carbonyl hydride mechanisms and applications was pub-
lished by Bullock and coworkers [6].
H3CO
C-H
C-H
CH3CH2C-H
O
O
O
6
7
8
Chart 1.

C. Leiva et al. / Journal of Organometallic Chemistry 695 (2010) 145–150
147
Table 1
a
Ketone and aldehyde reduction products, initial and relative rates, and turnover frequencies .
, M s 1
ꢁ
)
Relative rate
1.0
TOF (h
1.0
ꢁ1
)
Substrate
Product
Initial rate (r
6.9 ꢅ 10ꢁ
i
6
3
H
CH CH CH CCH
3
2
2
3
OH
4
5
H
2.5 ꢅ 10ꢁ
3.8 ꢅ 10ꢁ
5
3.6
5.6
3.0
5.0
OH
5
H
CCH3
OH
5
6
7
3
CH CH
2 2
CH OH
8.3 ꢅ 10ꢁ
9.1 ꢅ 10ꢁ
1.3 ꢅ 10ꢁ
12
13
10.0
11
5
CH OH
2
4
8
20
15
H3CO
CH OH
2
a
A general procedure for all the kinetic experiments was as follows: a 10 mL aliquot of the sodium formate solution (deoxygenated, 0.31 M in buffer NaH
.1 M, pH 7.01–7.04)) was transferred with a gas-tight syringe into a volumetric flask (25 ml) with a stir bar under N . Then, for example, benzaldehyde, 7, (31.5
.29 mmol) and 10 L of ethyl acetate (internal standard) were added to the reaction mixture, followed by 100 L of a 0.050 M stock solution of [Cp*Rh(bpy)(H
2
PO
l
4
/Na
L, 32.9 mg,
2
4
HPO ,
0
0
2
l
l
2
O)](OTf) , 1,
2
via syringe, and the total solution rapidly mixed at 23 °C in a controlled temperature bath. The reaction progress was followed by GC and GC–MS.
8
with a relative rate ratio of 6/7 = 1.08, while the steric effect was
also apparently not critical in the aldehyde series. To reiterate, any
increase in electron density on the carbonyl group; i.e., a better
hyde substrates, Eqs. (1)–(4), in analogy with the biomimetic
NAD reduction with 2 [2]:
+
r
-
5
donating carbonyl oxygen, facilitates binding to the
g
-Cp*Rh me-
k1
2
þ
þ
½
Cp ꢃ RhðbpyÞðH
2
OÞꢄ þHCOꢁ2
½Cp ꢃ RhðbpyÞðO
2
CHÞꢄ þH
2
O
tal ion center, presumably in concert with hydride transfer. There-
fore, the overall picture from the results in Table 1 showed that
the steric parameters have a moderate effect on rate, while the
electronic effects were evident, but were rather subtle, with a rel-
ative rate scale of ketone and aldehyde reactivity, spanning a
small range of 1–20.
1
k2
A
ð1Þ
ð2Þ
þ
þ
k3
k4
½Cp ꢃ RhðbpyÞðO CHÞꢄ
2
½Cp ꢃ RhðbpyÞHꢄ þCO
2
A
B
O
þ
½Cp ꢃ RhðbpyÞðg C1 -O-OCH
RÞꢄ2þÞꢄ
½
Cp ꢃ RhðbpyÞHꢄ þ
2
B
RCH
k5
2
.2. Mechanistic aspects: kinetic and activation parameters
ð3Þ
ð4Þ
½Cp ꢃ RhðbpyÞðg C1 -O-OCH
RÞꢄ2þÞꢄ þH
O
þ
k6
1 þ RCH OH
The mechanisms for these aqueous, catalytic hydride transfer
2
3
2
reactions with ketone and aldehyde substrates, utilizing 2 as the
catalyst, would be important to elucidate to adequately describe
the reaction pathways for this process in water [3]. Therefore,
the effect of concentration of one of the substrates, 7, precatalyst,
Since the concentration of the intermediate A was not zero, but
had very little change with time, and that of intermediates B and C
never in appreciable quantities, it was appropriate to apply the
steady-state approximation to them. Moreover, a Michaelis–Men-
ten saturation kinetic effect, indicative of a pre-equilibrium, was
observed experimentally for formate ion (Eq. (1)). In order to ob-
tain a simplified final rate-law equation, the concept of a rate-
determining step involved in the mechanism was studied (see sup-
porting information). Thus, two possibilities were proposed for the
putative rate-limiting step: (1) the formation of the
1
, and sodium formate, on the initial rates of reduction, r
low conversions to the alcohol product (ꢂ10%), was studied, and
indeed, r was found to be dependent on the concentrations of pre-
catalyst 1, and formate ion from the plots of r
i
, at
i
ꢁ
1
i
(M s ) vs. [precat-
alyst 1, and formate ion (M)], but not on the concentration of the
substrate, 7. We surmised from the results with substrate 7 that
the concentrations of all the ketone or aldehyde substrates studied
were all pseudo zero order. Further verification of the reaction or-
der with one representative substrate, 7, catalyst 2, and formate
+
[Cp*Rh(bpy)H] , intermediate B in Eq. (2); or less likely, (2) the for-
mation of the substrate-B complex, intermediate C in Eq. (4),
ion as determined from plots of ln r
i
vs. ln [7], ln [1], and ln [HCO2-
where we have combined the binding and hydride transfer steps
Na], provided slopes of 0.0, 1.0, and 1.0, respectively.
to simplify the kinetic interpretation. If indeed the formation of
+
The experimental rate-law being: d(CHOH)/dt = kcat[1][HCO2-
[Cp*Rh(bpy)H] was the rate-determining step, then k
3
<< k
4
.
ꢁ1
ꢁ1
+
Na] with kcat = 6.81 M
s
at 295.6 K.
Therefore, the formation of [Cp*Rh(bpy)H] was proposed as the
rate-limiting step in the reduction of the aldehydes and ketone sub-
strates, and this postulate was in agreement with the experimental
rate orders (ln, ln plots, Supporting Information) found in aqueous
The lack of significant steric parameters, but subtle electronic
effects, on the initial and relative rates of carbonyl reduction
shown in Table 1 appeared to implicate both coordination of
5
the substrate ketone or aldehyde to the
g
-Cp*Rh metal ion
2
media: [1], 1.0; [HCO Na], 1.0 and [7] 0.0, respectively. As a result,
center with concerted hydride transfer, as we discovered previ-
the initial rate of formation of the alcohol product was independent
of the substrate concentration (3–8) under these aqueous reaction
conditions and provided a zero order for substrate, which was con-
sistent with our previous regioselective reduction results with an
ously for the NAD+ models [1,2], while the kinetic results
+
showed that [Cp*Rh(bpy)(H)] , 2, formation was presumably
the rate-limiting step; even though the substrate does not
appreciably effect the rate, we surmised that binding of sub-
strate to the Cp*Rh metal ion center was a critical step for hy-
dride transfer to the carbonyl carbon. Therefore, the following
equations and equilibria could be established for our mechanis-
tic interpretations of the observed reductions of the ketone/alde-
+
+
aqueous NAD model and NAD itself with 2 at pH 6.5 [1,2]. Although
+
the rate-determining step was thought to be [Cp*Rh(bpy)H] forma-
tion, binding to the Cp*Rh metal centerwas assumed to be critical for
+
alcohol formation; in the reduction of the biomimetic NAD model,
2
N-benzylnicotinamide triflate, with 2, in 1:1 THF/H O, the rate-

1
48
C. Leiva et al. / Journal of Organometallic Chemistry 695 (2010) 145–150
ꢁ
4
4
.5
3
complex 2 at 4.95 ꢅ 10 M, and HCO
2
Na at 0.3069 M. An Arrhe-
3
2
1
0
y = -12143x + 42.963
nius plot of ln kcat vs. 1/T for substrate 7 (Fig. 1) allowed us to
determine the activation energy, E
2
a
, while the other parameters,
R = 0.9979
such as the enthalpy, entropy, and free energy of activation were
.5
2
à
à
obtained as follows:
D
H ;
DS from Eyring plots of ln kcat/T vs. 1/
à
à
T;
D
G =
D
H ꢁ T
DS. (Table 2). All the thermodynamic and activa-
.5
1
tion parameters in Table 2 were consistent with those found for
+
+
the reduction of NAD and the biomimetic models of NAD with
in H O [2].
Thus, our above-mentioned kinetic analysis, and the data in Ta-
.5
2
2
0
3.25E-03 3.30E-03 3.35E-03 3.40E-03 3.45E-03 3.50E-03
ble 1, allowed us to propose a plausible catalytic cycle for the
reduction of the ketone/aldehyde substrates, 3–8, with 1 as the
precatalyst, and sodium formate as the hydride source (Scheme
1
/T (K)
Fig. 1. Arrhenius plot of ln kcat vs. 1/T for substrate 7.
1
). The reaction of 1 with HCO
2
Na provided the [Cp*Rh(bpy)(O-
C(O)H)] complex, A, which decomposed to the hydride complex,
Cp*Rh(bpy)(H)] , (B), via a b-hydrogen elimination reaction to
+
+
[
determining step was found to be the binding of the amide carbonyl
of the NAD model to the Cp*Rh hydride complex, 2 [1,2].
produce CO [1,2]. The important role of the carbonyl functionality
of substrates 3–8 was to coordinate to the Cp*Rh metal ion center.
2
+
Thus, the reaction rate (
max) at high concentrations of formate ion:
m
) would approach a maximum value
It should be emphasized that even though the rate-limiting step
was found to be [Cp*Rh(bpy)(H)] formation, we proposed, from
+
(
m
m
max = k
3
[1]
0
. There-
ꢁ
fore, a plot of 1/
m
vs. 1/[HCO
2
], at a given [1]
0
concentration
the results in Table 1, that substrate binding to the Cp*Rh metal
ion center was still crucial for product formation. This was further
substantiated by reaction of a 1:1 mixture of substrate 4 and its
reduction product, cyclohexanol, which showed no effect on the
initial rate of reduction, and therefore, no effect on hydride forma-
tion, nor apparently, the binding of substrate 4 to 2.
ꢁ
2
(
3.2 ꢅ 10 M), is shown to be a straight line (Fig. 7, Supporting
ꢁ4 ꢁ1
Information), which provided the values of k
3
= 1.4 ꢅ 10 (s
)
ꢁ
2
and K
m
= 3.68 ꢅ 10 (M) for the reactions conducted in water.
The temperature effect on the rate constant, kcat, was studied in
the range of 289–306 K (higher temperatures were not feasible) for
substrate 7 in water (pH 7.0) at concentrations of 3.1 ꢅ 10 M,
ꢁ2
We further suggest that this binding process of carbonyl sub-
strate to the Cp*Rh metal ion center occurred in concert with hy-
dride transfer (Scheme 1, C I, transition state). It should also be
noted that hydride transfer, I to II in C, could possibly be a revers-
ible process as was recently found in the regioselective reduction
Table 2
Thermodynamic and activation parameters for substrate 7.
à
+
D
D
D
H
S
G
(kcal/mol)
(eu)
(kcal/mol) @ 25°C
23.6 ± 1.0
24.9 ± 1.4
16.1
of NAD models [2].
à
Finally, displacement of the alcohol product and recycling 1, via
à
+
another hydride [Cp*Rh(bpy)(H)] molecule, completes the plausi-
E
a
(kcal/mol)
24.1 ± 1.01
ble catalytic cycle, with substrate 7 as an example.
O
+
CH
CO2
7
k4
k5
k3
Rh
2
N
N
H
Hydride Formation
+
+
B
O
Rh
A
Rh
C
N
OCH
O
N
N
H
H
N
I
k1
k2
HCO2-
2+
+
Rh
O
Rh
N
C
N
H
OH
2
k6
N
N
H
N
N
N
N =
II
C
H
3
O+
Precatalyst
CH OH
2
1
Scheme 1. Plausible catalytic cycle for the reduction of all the substrates, 3–8, with 7 as an example.

C. Leiva et al. / Journal of Organometallic Chemistry 695 (2010) 145–150
149
3
. Conclusions
The significant results in the aqueous reductions of the car-
4. Experimental
4.1. Materials
[Cp*RhCl
+
bonyl substrates, 3–8, with [Cp*Rh(bpy)H] (2), to exclusively
form the kinetic product alcohols, were the rate-limiting
formation of [Cp*Rh(bpy)H] , and the proposed binding of the
carbonyl substrates to the
2
]
2
was purchased from Colonial Metals and was used
+
as received. [Cp*Rh(bpy)(H
the literature method [2], using [Cp*Rh(H
O)](OTf)
was prepared according to
O) ](OTf) as the precur-
2
2
5
g
-Cp*Rh metal ion center, in concert
2
3
2
with hydride transfer. This crucial binding step (Scheme 1, C1)
appeared to be moderately dominated by steric effects in an in-
ter comparison to both the ketone and aldehyde series, while
electronic effects, although subtle, were epitomized by the re-
sults with substrates 4 and 8. The ketone series, 3–5, showed
a slight steric effect during the reduction reaction for the alkyl
sor [7]. Water (HPLC grade) was purchased from Aldrich. All the ke-
tones, aldehydes, and alcohols were purchased from Aldrich and
used as received after GC/MS analysis.
4.2. Equipment
ketone substrates, 5 > 3, while the aryl ketone showed
a
The pH values were obtained with an Orion 601A pH meter
equipped with an Orion semimicro combination electrode. GC or
GC/MS analyses were performed on a Hewlett–Packard instrument
(HP 5890) with a J&W, DB-Wax column (30 m, film thickness
moderate electronic effect, with 5 > 4 > 3, where the electron-
donating effect of the benzene ring directly bonded to the car-
bonyl group appeared to increase binding to the Cp*Rh metal
ion center.
0.25 lm, I.D. 0.25 mm). A circulating water bath (VWR 1160) was
The aldehyde series, 6–8, which appeared to also minimize ste-
ric effects during the reduction reaction, emphasized a subtle elec-
tronic effect, where placing further electron density on the
carbonyl oxygen; i.e., comparing substrate 7 to 8, increased the ini-
tial rate of reduction by a moderate factor of ꢂ1.5. It was also
important to note that the rate-limiting step in these aqueous car-
used to control the temperature. For experiments conducted at dif-
ferent temperatures, the circulating water bath was warmed up for
at least 1 h before the kinetic studies were initiated.
4.3. Kinetic procedures for the reductions of ketones or aldehydes, 3–8,
with precatalyst [Cp*Rh(bpy)(H
2 2
O)](OTf) and sodium formate
+
bonyl reduction reactions, the formation of [Cp*Rh(bpy)H] , was
the same in aqueous solution for the regioselective reduction of
The kinetic experiments at 23 °C, which were controlled by a
circulating water bath (VWR 1160), were followed using GC and
GC–MS techniques. For experiments conducted at different tem-
peratures, the circulating water bath was warmed up for at least
1 h before the kinetic studies were initiated. Stock solutions of
+
+
natural NAD and the aqueous NAD model, b-nicotinamide ri-
0
bose-5 -methyl phosphate, at pH 6.5 [1,2].
These reported results can be best compared with the previous
+
findings by Ogo et al. [3i,j], which utilized [Cp*Ir(bpy)(H)] , an
6
+
analog of 2, and [
g
-C
6
Me
6
)Ru(bpy)(H)] in an acidic water med-
[Cp*Rh(bpy)(H
buffer NaH PO
the kinetic experiments in H
(N ) prior to being used. The induced volume change was taken
2
O)]OTf
2
(0.050 M) and sodium formate (0.31 M in
0.1 M, pH = 7.01–7.04) were made for
O. All the solutions were degassed
ium, with several similar ketone substrates. Firstly, the results
2
4
/Na HPO
2
4
6
+
with the hydride complex, [
g
-C
6
Me
6
)Ru(bpy)(H)] , at pH 4.0 un-
2
der saturation kinetics with sodium formate, were compared for
steric and electronic effects [3i]. From our perspective, the steric
effects have a minimal consequence on the reported TOF, in
agreement with the similar ketone substrates in Table 1. The elec-
tronic effect results were more difficult to compare, but the fact
that e-withdrawing groups on the carbonyl group should possibly
2
into account for the calculation of the turnover frequency (TOF).
A general procedure for all the kinetic experiments was as fol-
lows: A 10 mL aliquot of the sodium formate solution (deoxygen-
ated) was transferred with a gas-tight syringe into a volumetric
flask (25 ml) with a stir bar under N
hyde (0.029 mmol, 32.93 mg), 7, and 10
nal standard) were added to the reaction, followed by 100
0.050 M stock solution of [Cp*Rh(bpy)(H
O)](OTf) via syringe,
2
. Then 31.5
L of ethyl acetate (inter-
L of a
lL of benzalde-
0
limit the protonation step on the carbonyl oxygen, Ogo et al s.
l
mechanism of choice, does bode better, in our opinion, for direct
l
6
bonding of the carbonyl to
g
-(C
6
Me
6
)Ru metal ion center. For
2
2
+
comparison to the Ogo et al. [Cp*Ir(bpy)(H)] results [3j], again
with several similar ketones, the steric effects seemed to
moderately affect the TOF, while the electronic effects appeared
to be less important, but binding of the substrate ketone to the
Cp*Ir center can not be overlooked. Moreover, to reiterate, the
and the total solution rapidly mixed at 23 °C in a controlled tem-
perature bath. A GC–MS sample was run every 10 min (or 1 h
depending on the substrate) until the starting material disap-
peared. Calibration curves for all substrates were recorded to ob-
tain the GC response factors, and calculate concentrations of
substrate reacted and products formed with time. All the kinetic
experiments were carried out with constant stirring, while some
reactions were followed for more than 24 h. The values of the ini-
(
4 2
PTA) RuCl results of Darensbourg et al. [3m], while difficult to
compare to the presented results because the aldehyde substrate
reactions were conducted in a biphasic mode, did have an impor-
tant inhibition result with benzaldehyde (7), in the presence of
CO, which quenched the reduction reaction, and intimated that
binding of the carbonyl group to the Ru metal ion center was pos-
sibly important for the reduction to proceed. Kuo et al. [3n] also
i
tial rate (r ) were obtained from at least 3 runs, using plots of ln C
(product) vs. time.
4.4. Activation parameters for benzaldehyde, 7
2
proposed binding of the substrate ketone to the [Cp MoH(OTf)]
metal ion center during transfer of the hydride to the carbonyl
of the ketone substrate.
An analogous procedure with the same amount of 7 (0.031 M)
as described above was used, except that the solutions were run
+
à
à
Finally, we wish to categorize [Cp*Rh(bpy)H] as an example of
at several temperatures. The values of
D
H and
DS were obtained
a hydride that can bind to the substrate functional group, and in
that process, has constricted the transition state for hydride trans-
fer, in this example, to the carbonyl carbon atom [1,2]. Thus, there
is the potential of forming chiral alcohols from the reduction of
achiral ketones with the development of sterically limited chiral li-
gands on the Cp*Rh metal ion center, and will be pursued in the
future.
from plots of ln kcat/T vs. 1/T and the E
a
was obtained from the plot
à
à
à
ln kcat vs. 1/T.
D
G =
D
H ꢁ TDS
Acknowledgments
We gratefully acknowledge Department of Energy funding from
the Advanced Energy Projects and Technology Research Division,

1
50
C. Leiva et al. / Journal of Organometallic Chemistry 695 (2010) 145–150
Office of Computational and Technology Research under DOE Con-
tract No. DE AC02-05CH11231.
(g) N. Makihara, S. Ogo, Y. Watanabe, Organometallics 20 (2001) 497;
h) S. Ogo, N. Makihara, Y. Kaneko, Y. Watanabe, Organometallics 20 (2001)
903;
(
4
(
(
i) S. Ogo, T. Abura, Y. Watanabe, Organometallics 21 (2002) 2964;
j) T. Abura, S. Ogo, Y. Watanabe, S. Fukuzumi, J. Am. Chem. Soc. 125 (2003)
Appendix A. Supplementary material
4
(
149;
k) D. Darensbourg, N.W. Stafford, F. Joó, J. Reibenspies, J. Organomet. Chem.
488 (1995) 99;
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2009.10.015.
(
(
(
l) F. Joó, L. Nádasdi, A.Cs. Bényei, D. Darensbourg, J. Organomet. Chem. 512
1996) 45;
m) D. Darensbourg, F. Joó, M. Kannisto, Á. Kathó, J.H. Relbenspies,
References
Organometallics 11 (1992) 1990;
n) L.Y. Kuo, T.J.R. Weakley, K. Awana, C. Hsia, Organometallics 20 (2001) 4969;
(
[
[
1] H.C. Lo, O. Buriez, J.B. Kerr, R.H. Fish, Angew. Chem., Int. Ed. 38 (1999) 1429.
2] H.C. Lo, C. Leiva, O. Buriez, J.B. Kerr, M.M. Olmstead, R.H. Fish, Inorg. Chem. 40
(o) L.Y. Kuo, D.M. Finigan, N.N. Tadros, Organometallics 22 (2003) 2422;
(p) F. Joó, Aqueous Organometallic Catalysis, Kluwer Academic Publishers,
Dordrecht, 2001. and references therein;
(
2001) 6705.
[
3] (a) I.T. Horváth, F. Joó, Aqueous Organometallic Chemistry and Catalysis,
Kluwer Academic Publishers, Dordrecht, 1995. and references therein;
(q) B. Cornils, W.A. Herrmann (Eds.), Aqueous-Phase Organometallic Catalysis:
Concepts and Applications, Wiley-VCH, Weinheim, 2004. and chapters therein.
[4] (a) D. Westerhausen, S. Herrmann, W. Hummel, E. Steckhan, Angew. Chem., Int.
Ed. Engl. 31 (1992) 1529. and references therein;
(b) E. Steckhan, S. Herrmann, R. Ruppert, E. Dietz, M. Frede, E. Spika,
Organometallics 10 (1991) 1568. and references therein.
[5] H.C. Lo, R.H. Fish, Angew. Chem., Int. Ed. 41 (2002) 478.
[6] T.-Y. Cheng, B.S. Brunschwig, R.M. Bullock, J. Am. Chem. Soc. 120 (1998) 13121.
[7] M.S. Eisen, A. Haskel, H. Chen, M.M. Olmstead, D.P. Smith, M.F. Maestre, R.H.
Fish, Organometallics 14 (1995) 2806.
(
b) P.A. Chaloner, M.A. Esteruelas, F. Joó, L.A. Oro, Homogeneous
Hydrogenation, Kluwer Academic Publishers, Dordrecht, 1994. p. 287, and
references therein;
(
(
(
c) W.A. Herrmann, C.W. Kohlpainter, Angew. Chem., Int. Ed. 32 (1993) 1524;
d) F. Joó, Á. Kathó, J. Mol. Catal. A 116 (1997) 3. and references therein;
e) F. Joó, J. Kovács, A.Cs. Bényei, L. Nádasdi, G. Laurenczy, Chem. Eur. J. 7 (2001)
1
93. and references therein;
(
f) S. Ogo, N. Makihara, Y. Watanabe, Organometallics 18 (1999) 5470;