Mechanistic investigations of an enantioselective hydrogenation catalyzed by a ruthenium-BINAP complex. 1. Stoichiometric and catalytic labeling studies

Article abstract of DOI:10.1021/om980113f

The enantioselective hydrogenation of (Z)-methyl α-acetamidocinnamate ((Z)-MAC) is catalyzed by [Ru((R)-BINAP)(H)(MeCN)_n(sol)_3-n](BF₄) (2, n = 0-3, sol = acetone or methanol, (R)-BINAP = (R)-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl) in up to 94% ee (R). The mechanism of this catalytic hydrogenation was investigated using in situ NMR spectroscopy, and by comparing the stereochemistry and regiochemistry of deuterium-labeling studies carried out on the catalytic reaction to those carried out on an isolated, possible catalytic intermediate. The isolated species is the olefin-hydride insertion product, [Ru((R)-BINAP)-((S)-MACH)(MeCN)](BF₄) ((S_Cα)-1). Compound (S_Cα)-1 is the only species present in detectable amounts, by ¹H and ³¹P NMR spectroscopy, during the catalytic hydrogenation at room temperature. The absolute configuration at the stereogenic α-carbon of MACH is the same (assuming stereospecific replacement of ruthenium with hydrogen) as that of the major enantiomer of the catalytic hydrogenation ((R)-N-acetylphenylalanine methyl ester ((R)-MACH₂)). Results obtained from the stoichiometric hydrogenolysis and deuteriolysis of (S_Cα)-1, from the catalytic deuteration of (E)-MAC and (Z)-MAC, and from the reaction of (S_Cα)-1 with excess (Z)-MAC-Me-d₃ all indicate that formation of (S_Cα)-1 is rapid and reversible prior to the irreversible hydrogenolysis of the ruthenium-carbon bond. The sum of the stereoselectivities and regioselectivities of the formation and hydrogenolysis of (S_Cα)-1 equals the stereoselectivity and regioselectivity of the catalytic hydrogenation. Solvolysis of the ruthenium-carbon bond occurs to less than 4% during the catalytic hydrogenation carried out in methanol. Removal of MeCN from the catalyst system has no effect on the enantioselection of the catalytic hydrogenation.

Full text of DOI:10.1021/om980113f

2228
Organometallics 1998, 17, 2228-2240
Mech a n istic In vestiga tion s of a n En a n tioselective
Hyd r ogen a tion Ca ta lyzed by a Ru th en iu m -BINAP
Com p lex. 1. Stoich iom etr ic a n d Ca ta lytic La belin g
Stu d ies
J ason A. Wiles and Steven H. Bergens*
Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2
Received February 19, 1998
The enantioselective hydrogenation of (Z)-methyl R-acetamidocinnamate ((Z)-MAC) is
catalyzed by [Ru((R)-BINAP)(H)(MeCN)n(sol)3-n](BF4) (2, n ) 0-3, sol ) acetone or methanol,
(R)-BINAP ) (R)-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl) in up to 94% ee (R). The
mechanism of this catalytic hydrogenation was investigated using in situ NMR spectroscopy,
and by comparing the stereochemistry and regiochemistry of deuterium-labeling studies
carried out on the catalytic reaction to those carried out on an isolated, possible catalytic
intermediate. The isolated species is the olefin-hydride insertion product, [Ru((R)-BINAP)-
((S)-MACH)(MeCN)](BF4) ((SC )-1). Compound (SC )-1 is the only species present in
R
R
1
detectable amounts, by H and 31P NMR spectroscopy, during the catalytic hydrogenation
at room temperature. The absolute configuration at the stereogenic R-carbon of MACH is
the same (assuming stereospecific replacement of ruthenium with hydrogen) as that of the
major enantiomer of the catalytic hydrogenation ((R)-N-acetylphenylalanine methyl ester
((R)-MACH2)). Results obtained from the stoichiometric hydrogenolysis and deuteriolysis
of (SC )-1, from the catalytic deuteration of (E)-MAC and (Z)-MAC, and from the reaction of
R
(SC )-1 with excess (Z)-MAC-Me-d3 all indicate that formation of (SC )-1 is rapid and reversible
R
R
prior to the irreversible hydrogenolysis of the ruthenium-carbon bond. The sum of the
stereoselectivities and regioselectivities of the formation and hydrogenolysis of (SC )-1 equals
R
the stereoselectivity and regioselectivity of the catalytic hydrogenation. Solvolysis of the
ruthenium-carbon bond occurs to less than 4% during the catalytic hydrogenation carried
out in methanol. Removal of MeCN from the catalyst system has no effect on the enan-
tioselection of the catalytic hydrogenation.
In tr od u ction
are unknown.2,3 Among the reasons that enantioselec-
tive interactions have not been elucidated for these
hydrogenations is that all reported observations of
putative catalytic intermediates are either incomplete,4
providing little information about how the substrate is
bonded to ruthenium, or they are of species in which the
substrate is not bonded to ruthenium through a prochiral
functionality.5 In a recent communication,6 we reported
We report and compare results obtained from deute-
rium-labeling studies on a putative catalytic intermedi-
ate to those obtained from the overall catalytic cycle
for the enantioselective hydrogenation of (Z)-methyl
R-acetamidocinnamate ((Z)-MAC) using a ruthenium-
((R)-BINAP) system as catalyst ((R)-BINAP ) (R)-2,2′-
bis(diphenylphosphino)-1,1′-binaphthyl). Complexes of
ruthenium(II) and BINAP are among the most success-
ful catalyst systems for the enantioselective hydrogena-
tion of prochiral olefins and ketones.1 Despite the high
enantiomeric excesses (ee’s) and turnover numbers that
often distinguish these systems, the asymmetric inter-
actions that are responsible for their enantioselection
(2) For representative examples of mechanistic investigations of
hydrogenations catalyzed by ruthenium-BINAP complexes, see: (a)
Chan, A. S. C.; Chen, C. C.; Yang, T. K.; Huang, J . H.; Lin, Y. C. Inorg.
Chim. Acta 1995, 234, 95-100. (b) Brown, J . M.; Rose, M.; Knight, F.
I.; Wienand, A. Recl. Trav. Chim. Pays-Bas 1995, 114, 242-251. (c)
Brown, J . M. Chem. Soc. Rev. 1993, 25-41. (d) Kawano, H.; Ikariya,
T.; Ishii, Y.; Saburi, M.; Yoshikawa, S.; Uchida, Y.; Kumobayashi, H.
J . Chem. Soc., Perkin Trans. 1 1989, 1571-1575.
(3) The mechanism of the enantioselective hydrogenation of (Z)-MAC
catalyzed by rhodium-bis(phosphine) complexes is well documented:
(a) Giovannetti, J . S.; Kelly, C. M.; Landis, C. R. J . Am. Chem. Soc.
1993, 115, 4040-4057. (b) Landis, C. R.; Halpern, J . J . Am. Chem.
Soc. 1987, 109, 1746-1754. (c) Brown, J . M.; Chaloner, P. A.; Morris,
G. A. J . Chem. Soc., Perkin Trans. 2 1987, 1583-1588. (d) Halpern, J .
Science 1982, 217, 401-407. (e) Brown, J . M.; Chaloner, P. A. J . Am.
Chem. Soc. 1980, 102, 3040-3048. For enantioselective reductions of
R-aminoacrylic acid derivatives by dideuterium gas catalyzed by
rhodium-bis(phosphine) complexes, see: (f) Burk, M. J .; Feaster, J .
E.; Nugent, W. A.; Harlow, R. L. J . Am. Chem. Soc. 1993, 115, 10125-
10138. (g) Scott, J . W.; Keith, D. D.; Nix, G., J r.; Parrish, D. R.;
Remington, S.; Roth, G. P.; Townsend, J . M.; Valentine, D., J r.; Yang,
R. J . Org. Chem. 1981, 46, 5086-5093. (h) Detellier, C.; Gelbard, G.;
Kagan, H. B. J . Am. Chem. Soc. 1978, 100, 7556-7561. (i) Koenig, K.
E.; Knowles, W. S. J . Am. Chem. Soc. 1978, 100, 7561-7564.
* To whom correspondence should be addressed. E-mail:
Steve.Bergens@ualberta.ca.
(1) (a) Ager, D. J .; Laneman, S. A. Tetrahedron: Asymmetry 1997,
8, 3327-3355. (b) Noyori, R. Acta Chem. Scand. 1996, 50, 380-390.
(c) Kumobayashi, H. Recl. Trav. Chim. Pays-Bas 1996, 115, 201-210.
(d) Akutagawa, S. Appl. Catal., A 1995, 128, 171-207. (e) Noyori, R.
Asymmetric Catalysis in Organic Synthesis; Wiley: New York, 1994.
(f) Noyori, R. Tetrahedron 1994, 50, 4259-4292. (g) Takaya, H.; Ohta,
T.; Noyori, R. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; VCH:
New York, 1993; Chapter 1. (h) Noyori, R. CHEMTECH 1992, 22, 360-
367. (i) Noyori, R. Science 1990, 248, 1194-1199. (j) Noyori, R.; Takaya,
H. Acc. Chem. Res. 1990, 23, 345-350. (k) Noyori, R.; Kitamura, M.
In Modern Synthetic Methods 1989; Scheffold, R., Ed.; Springer-
Verlag: Berlin, 1989; pp 115-198. See also ref 6 and references
therein.
S0276-7333(98)00113-7 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 04/28/1998

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2229
Ta ble 1. Hyd r ogen a tion of (Z)-MAC Ca ta lyzed
the first observation and structural characterization of
a ruthenium-BINAP complex with a substrate bonded
to ruthenium through a stereogenic carbon center
derived from the prochiral functionality of the substrate.
The complex [Ru((R)-BINAP)((S)-MACH)(MeCN)](BF4)
by 2a
((SC )-1) results from the stoichiometric or catalytic
R
entry
P(H2), atm
solvent
T, °C
ee, % (R)
94
92
83
75
92
93
87
(excess substrate) reaction between (Z)-MAC and the
catalyst system [Ru((R)-BINAP)(H)(MeCN)n(sol)3-n]-
(BF4) (2, n ) 0-3, sol ) acetone or methanol)7 at room
1
2
3
4
5
6
7
8
9
2
4
20
50
4
4
4
4
4
acetone
acetone
acetone
acetone
acetone
acetone
MeOH
THF
30
30
30
30
50
10
30
30
30
30
temperature. Compound (SC )-1 is the product of ole-
R
fin-hydride insertion with placement of ruthenium at
the R-olefin carbon of (Z)-MAC (eq 1). To identify with
85
64
CH2Cl2
MeCN
10
50
no reaction
a
Reaction conditions: 2 mol % 2; [2] ) 2.6 mM; stir rate ) 1100
rpm; 100% conversion except where noted otherwise.
diate are relevant to the origins of enantioselection by
the catalytic reaction. As part of our efforts to deter-
mine if (SC )-1 is an authentic catalytic intermediate,
R
we report the stereoselectivity and regioselectivity of the
olefin-hydride insertion forming (SC )-1, the stereose-
R
lectivity and regioselectivity of the reverse â-hydride
elimination, and the extent to which formation of (SC )-1
R
is reversible prior to hydrogenolysis of the ruthenium-
carbon bond. These results will be compared to the
stereoselectivity and regioselectivity of the catalytic
hydrogenation. We also report the extent of solvolysis
of the ruthenium-carbon bond during catalytic hydro-
genations carried out in methanol solution and the effect
of removing MeCN from the system on the catalytic
enantioselectivity.
reasonable certainty that (SC )-1 or any species observed
R
during an enantioselective catalytic reaction is a cata-
lytic intermediate leading to the major enantiomer of
the product, the following criteria must be satisfied: (1)
the absolute configuration of the putative intermediate
must directly extrapolate to that of the major enanti-
omer of the product. Although satisfaction of this
criterion does not constitute proof of the authenticity
of the putative interemediate, failure to satisfy this
criterion is proof that the species is not an intermediate
leading to the major enantiomer. (2) The stereoselec-
tivities and regioselectivities (as determined by isotopic
labeling or other probes) of the stoichiometric reactions
forming the putative intermediate and of the reactions
converting it to product (and free catalyst) must sum-
mate to give the overall stereoselectivity and regiose-
lectivity of the catalytic reaction. (3) The kinetics of the
stoichiometric reactions forming the putative intermedi-
ate and those of the reactions converting it to product
(and free catalyst) must comprise at least a component
of the overall kinetics of the catalytic cycle. In addition
to this information, the identities and kinetics of the
enantioselective step(s) must be obtained to determine
if the asymmetric interactions in the putative interme-
Resu lts a n d Discu ssion
Op tim iza tion of th e Ca ta lytic Hyd r ogen a tion .
The pressure of dihydrogen gas, the solvent, and tem-
perature were varied to optimize the hydrogenation of
(Z)-MAC using 2 as the catalyst. Table 1 summarizes
our results. Among the solvents surveyed, the highest
enantiomeric excess (ee) was obtained in acetone, being
even higher than that in methanol, a common solvent
for hydrogenations using ruthenium-BINAP catalyst
systems.1 The turnover frequency8 was higher in
acetone (∼0.9 min-1) than in methanol (∼2.7 × 10-2
min-1)7 as well. Although we have not determined the
maximum turnover number in acetone, we achieved 980
turnovers in 91% ee (R) after 4 days at 30 °C under 4
atm of dihydrogen gas.
The ee in acetone solution decreased as the dihydro-
gen pressure was increased (Table 1). Inverse relation-
ships between ee and dihydrogen pressure have been
reported for other catalyst-substrate systems.1e Bar-
ring mass transfer effects,9 an inverse relationship
between ee and pressure of dihydrogen gas often implies
(4) (a) King, S. A.; DiMichele, L. In Catalysis of Organic Reactions;
Scaros, M. G., Prunier, M. L., Eds.; Marcel Dekker: New York, 1995;
Vol. 62, pp 157-166. (b) Saburi, M.; Takeuchi, H.; Ogasawara, M.;
Tsukahara, T.; Ishii, Y.; Ikariya, T.; Takahashi, T.; Uchida, Y. J .
Organomet. Chem. 1992, 428, 155-167. (c) Ohta, T.; Takaya, H.;
Noyori, R. Tetrahedron Lett. 1990, 31, 7189-7192.
(5) (a) Ashby, M. T.; Khan, M. A.; Halpern, J . Organometallics 1991,
10, 2011-2015. (b) Ashby, M. T.; Halpern, J . J . Am. Chem. Soc. 1991,
113, 589-594.
(6) Wiles, J . A.; Bergens, S. H.; Young, V. G. J . Am. Chem. Soc. 1997,
119, 2940-2941.
(8) These turnover frequencies were determined by setting up and
taking down the pressure reactor as quickly as possible during a
hydrogenation. Their values are, therefore, to be taken as approximate.
Reaction conditions: 2 mol % 2; [2] ) 2.6 mM; stir rate ) 1100 rpm;
T ) 30 °C; P(H2) ) 4 atm.
(9) All reactions were carried out under reaction-rate-limiting
conditions to eliminate gas-liquid mass transfer effects. For a study
describing gas-liquid mass transfer effects on enantioselectivity, see:
Sun, Y.; Landau, R. N.; Wang, J .; LeBlond, C.; Blackmond, D. G. J .
Am. Chem. Soc. 1996, 118, 1348-1353.
(7) Complex 2 is a well-defined catalyst system for a variety of
transformations, including the hydrogenation of olefins. Wiles, J . A.;
Lee, C. E.; McDonald, R.; Bergens, S. H. Organometallics 1996, 15,
3782-3784.

2230 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens
ylalanine methyl ester ((R)-MACH2), the major enan-
tiomer formed by the catalytic hydrogenation. As we
reported in the communication, the stoichiometric reac-
tion of (SC )-1 with dihydrogen gas in methanol solution
R
generated MACH2 at a similar rate and ee as the cata-
lytic hydrogenation in methanol.6 We now report that
although the stoichiometric hydrogenolysis of (SC )-1 in
R
acetone solution also produces MACH2 in similar ee to
the catalytic hydrogenation (stoichiometric, 89% (R) (eq
2); catalytic 92% ee (R)), the rate of the stoichiometric
F igu r e 1. 31P{1H} NMR spectrum of an operating catalytic
hydrogenation of (Z)-MAC using 2 as catalyst in acetone-
d6. The scale of the expansions is 5 times that of the
original plot.
that the identity of the enantioselective step changed
with the increase in dihydrogen pressure.
The ee in acetone increased slightly as the tempera-
ture was decreased from 50 to 10 °C; the turnover
frequency, however, was too low to be of practical use
below 10 °C. We conclude from these results that the
optimum conditions for hydrogenation of (Z)-MAC using
2 as the catalyst are acetone solution, 30 °C, and 4 atm
of dihydrogen gas.
Mech a n istic In vestiga tion s. Figure 1 shows a 31P-
{1H} NMR spectrum recorded of an operating catalytic
hydrogenation of (Z)-MAC after 8 turnovers under
conditions (acetone, pressure of dihydrogen gas ) 1 atm,
25 °C, 4 mol % 2) similar to the optimum conditions
described in the previous section.10 Compound (SC )-1
R
was the only detectable ruthenium species in solution
at room temperature.11 Similar results were obtained
using methanol, tetrahydrofuran, and methylene chlo-
ride as the solvent.12 Deuterium labeling and other
experiments will show (vide infra) that it is extremely
hydrogenolysis slows over the course of the reaction in
this solvent. The product distribution after 4 h was 31%
(SC )-1, 58% [Ru((R)-BINAP)(H)(η6-MACH2)](BF4) (3,
R
MACH2 is bonded to ruthenium as an η6-ligand through
the arene ring),6 and 11% MACH2.13 The hydrogenoly-
sis was 97% complete after 24 h. MACH2 was liberated
from 3 by refluxing the mixture in MeCN solution before
the ee was determined. The conditions for the stoichio-
metric hydrogenolysis in acetone were the same as the
catalytic hydrogenation in acetone except that excess
unlikely that the formation of (SC )-1 is an irreversible
R
kinetic trap under these conditions. Further, the cata-
lytic hydrogenation of (Z)-MAC using isolated (SC )-1
R
as the catalyst occurs at the same rate and ee as when
2 is used as the catalyst. Its relatively high concentra-
tion during catalysis is good evidence that (SC )-1,
R
(Z)-MAC was absent and the concentration of (SC )-1
R
whether a catalytic intermediate or not, is the most
stable ruthenium compound in solution under these
conditions.
was approximately 4-fold higher for the stoichiometric
hydrogenolysis than for the catalytic hydrogenation.
Hydrogenolysis of ruthenium-carbon bonds in coordi-
natively saturated 18-electron ruthenium(II) compounds
has been shown to require prior ligand dissociation to
generate a 16-electron intermediate. For example, the
hydrogenolysis of the ruthenium-acyl bond in RuCl-
(COC7H9)(CO)2(PPh3)2 requires prior dissociation of a
triphenylphosphine ligand.14 It is quite likely that the
Hyd r ogen olysis of (SC )-1 in Aceton e. The abso-
r
lute configuration at the carbon center bonded to
ruthenium in (SC )-1 is S. Stereospecific replacement
R
of ruthenium by hydrogen will form (R)-N-acetylphen-
(10) The differences between the conditions under which the spec-
trum was recorded and the optimum conditions for the catalytic
hydrogenation resulted from the technical difficulties in carrying out
such a reaction in our NMR tubes.
hydrogenolysis of (SC )-1 proceeds via prior dissociation
R
(11) A small amount (∼8%) of one other species was detected in the
of the MeCN ligand to generate the 16-electron inter-
mediate [Ru((R)-BINAP)((S)-MACH)](BF4), which may
contain a weakly coordinating acetone ligand. Since
MeCN binds strongly to ruthenium(II) complexes such
31P NMR spectrum of the catalytic mixture recorded after cooling to
2
-40 °C (31P{1H} ((CD3)2CO, 161.9 MHz) δ 39.4 (d, J P-P ) 23.6 Hz,
2
1P), 55.4 (d, J P-P ) 23.6 Hz, 1P). This species is present in similar
amounts in the 31P NMR spectrum recorded at -40 °C of isolated and
purified (SC )-1. We suspect it is either the species obtained from
as (SC )-1,15 the decrease in rate as the hydro-
R
R
dissociation of acetonitrile from (SC )-1 or it is the opposite diastere-
R
omer in equilibrium with (SC )-1. We have been unable to assign other
signals to this species.
R
(13) These values were determined by 1H NMR spectroscopy.
Analysis of this mixture by 31P NMR spectroscopy confirmed the
(12) Compound (SC )-1 typically comprised 80-90% of the ruthenium-
containing species in Rthese solvents under these conditions. The other
compounds present were mainly a mixture of [Ru((R)-BINAP)(H)(η6-
MACH2)](BF4) and a species which we tentatively propose as [Ru((R)-
BINAP)(H)(η6-(Z)-MAC)](BF4). In these complexes, MACH2 and (Z)-
MAC are bound to ruthenium as η6-arene ligands.
presence of (SC )-1 (35%), 3 (65%), and a number of unidentified species
R
(quantities too low to accurately integrate). We believe that these
unidentified species are decomposition products of 3 that generate
uncoordinated MACH2.
(14) J oshi, A. M.; J ames, B. R. Organometallics 1990, 9, 199-205.

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2231
a
Sch em e 1
F igu r e 2. The distribution of MACH2-dn isotopomers from
the deuteriolysis of (ScR)-1 in acetone.
a
Species shown in quotation marks are approximate rep-
resentations of the true species in solution.
Rever sibility of F or m a tion of (SC )-1. Reaction of
r
genolysis proceeded in acetone solution most likely arose
from the increase in the concentration of MeCN which
occurred as compound 3 accumulated in solution (eq 2).
Kinetic investigations of the stoichiometric hydrogenol-
ysis and of the catalytic hydrogenation of (Z)-MAC are
underway, and their results will be compared in due
course. We note presently that although added MeCN
is a strong inhibitor of the catalytic hydrogenation of
(Z)-MAC using 2 as the catalyst, free MeCN does not
build up to detectable levels during the catalytic reaction
because the excess (Z)-MAC in solution reacts with 2
diastereomerically pure (SC )-1 with dihydrogen gas
R
generated (R)-MACH2 in ee’s similar to those of the
catalytic reactions in methanol and in acetone. The
similar ee’s for the catalytic and stoichiometric reactions
indicate that formation of (SC )-1 was to some extent
R
reversible and that the other diastereomer(s) of (SC )-1
R
reacted to form (S)-MACH2. To confirm that its forma-
tion is reversible, compound (SC )-1 was reacted with
R
2.8 equiv of (Z)-MAC-Me-d3 in acetone solution at 35
°C (eq 4). The system achieved equilibrium (assuming
to generate (SC )-1, thereby preventing formation of 3
R
(see Figure 1). We note further that the dissociation of
MeCN prior to the hydrogenolysis of (SC )-1 or of a
R
related species does not contribute to the enantioselec-
tion of the catalytic reaction; the ee for the catalytic
hydrogenation of (Z)-MAC using [Ru((R)-BINAP)(H)(η6-
(rac)-MACH2)](BF4) as the catalyst in methanol was
86% (R) (eq 3), in accord with that obtained using (SC )-1
R
or 2 as catalysts (each giving 87% ee (R)).
(15) No dissociation of MeCN from (SC )-1 was observed at room
temperature.
R
Keq ) 1), demonstrating that formation of (SC )-1 was
R

2232 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens
reversible. Compounds (SC )-1 and (SC )-1-Me-d3 were
temperatures well below 0 °C (eq 6).18 Under dihydro-
R
R
the only ruthenium species detected in solution. Scheme
1 shows our proposed sequence of steps for the reversible
formation of (SC )-1 and (SC )-1-Me-d3 from 2 reacting
R
R
with (Z)-MAC or with (Z)-MAC-Me-d3, respectively,
assuming the reverse of formation of (SC )-1 proceeds
R
via a â-hydride elimination. â-Hydride elimination
within (SC )-1 followed by dissociation of (Z)-MAC would
R
generate 2, which can either react with (Z)-MAC or with
(Z)-MAC-Me-d3 to generate the corresponding isoto-
pomers of (SC )-1. A consequence of the formation of
R
(SC )-1 being to some extent reversible via a â-hydride
R
elimination on the time scale of the hydrogenolysis is
gen gas, compound 2 is, therefore, in rapid equilibrium
with small concentrations of the ruthenium(IV) trihy-
dride [Ru((R)-BINAP)(H)3(MeCN)x(acetone)y](BF4) (4)
(or the corresponding ruthenium(II) hydride-η2-dihy-
drogen species), resulting in the observed H-D ex-
change with dideuterium gas. An equilibrium between
2 and 4 would, in principle, favor 2 because 4 must
contain at least one mutually trans-disposition of a
phosphine and a hydride ligand, an unfavorable situa-
tion as both of these ligands have strong trans influ-
ences.19 Any free 2 generated in solution during the
stoichiometric deuteriolysis of 1 is likely to quickly react
with dideuterium gas to generate 2-d.
that stoichiometric deuteriolysis of (SC )-1 must form
R
some MACH2 with deuterium at the â-position. This
condition on the proposed mechanism of the reverse
process is investigated in the next section.
Deu ter iu m Stu d ies in Aceton e. The stoichiometric
deuteriolysis of (SC )-1 carried out in acetone solution
R
slowed after ∼76% conversion (eq 5). Figure 2 shows
The MACH2-R-d1 (12%) produced by the stoichiomet-
ric deuteriolysis (Figure 2) almost certainly resulted
from the deuteriolysis of the ruthenium-carbon bond
in (SC )-1. The product which formed in large excess
R
(80%) was (R,R and S,S)-MACH2-R,â-d2, with deuterium
at both the R- and â-positions. This product is that of
a net cis-addition of dideuterium gas to the carbon-
carbon double bond of (Z)-MAC, and its presence is
strong evidence for the formation of (SC )-1 being
R
reversible via a â-hydride elimination. Another pos-
sibility, a stereoselective C-H activation at the â-posi-
tion of MACH2-R-d1, is ruled out by our finding that
reaction of MACH2 with 2-d under dideuterium gas does
not result in H-D exchange at the R- or â-positions of
MACH2. Scheme 2 shows the sequence of steps involv-
ing olefin-hydride insertion and â-hydride elimination
that accounts for the isotopomers of MACH2-dn produced
the composition of the reduced MAC after the deuteri-
olysis had slowed ((R,R and S,S)-MACH2-R,â-d2,16,17
80%; MACH2-R-d1, 12%; (R,R and S,S)-MACH2-â-d1, 6%;
and MACH2, ∼2%). The combined ee was 90% (R,R and
R), which is similar to that of the catalytic hydrogena-
tion of (Z)-MAC in acetone (92% (R)). MACH2 labeled
by deuterium at the â-position formed as predicted by
the â-hydride elimination mechanism for the reverse of
by the stoichiometric deuteriolysis of (SC )-1 in acetone.
R
Elimination of the pro-R-â-hydride in (SC )-1 followed
R
formation of (SC )-1 (Scheme 1). It is necessary to
R
by olefin dissociation generates (Z)-MAC and 2. The
stereochemistry of the â-hydride elimination will be
discussed in more detail below where it will be shown
determine the reactivity of 2 toward dideuterium gas
to more fully interpret these results.
Further reaction between 2 and dihydrogen gas (1
atm) was not detected by NMR spectroscopy in acetone
solution over temperatures ranging from -80 to 25 °C.
Reaction with dideuterium gas (1 atm), however, re-
sulted in the immediate formation of 2-d, even at
that elimination of the pro-S-â-hydride within (SC )-1
R
is highly unlikely. Compound 2, generated via the
â-hydride elimination, will quickly react with the excess
dideuterium gas in the reactor to produce 2-d (eq 6).
Coordination of (Z)-MAC to 2-d through the si-olefin face
followed by insertion into the ruthenium-deuteride
bond will generate (SC ,RC )-â-d1-1 (of R absolute con-
(16) All isotopic substitution patterns of MACH2-dn were assigned
by comparison of their 1H and 2H NMR spectra to those of N-
acetylphenylalanine. See ref 3h and the Experimental Section for
details.
R
â
figuration at the deuterated â-carbon center). In strong
support of this sequence of steps is that placement of
(17) For all reductions with dideuterium gas in acetone, a portion
of the product was deuterated at nitrogen. In general, there was an
approximate correlation between the extent of deuteration at nitro-
gen and the length of time the reaction was exposed to dideuterium
gas in the presence of 2. Only an approximate correlation could be
obtained because H-D exchange was rapid upon exposure to even trace
amounts of water. Control experiments in which MACH2 was exposed
to 2 mol % 2 under deuterium gas in acetone showed that 2 catalyzes
the H-D exchange between the N-H bond of MACH2 and dideuterium
gas. We believe that deuteration at nitrogen is a side reaction which
is not part of the mechanism of hydrogenation of (Z)-MAC catalyzed
by 2.
acetone solutions of (SC )-1 under lower pressures of
R
(18) An NMR tube containing a -80 °C argon-saturated acetone
solution of 2 was quickly ejected from a cooled (-80 °C) NMR probe,
injected with an excess of dideuterium gas through a rubber septum,
shaken for ∼15 s, and quickly returned to the probe. NMR spectra
recorded at -80 °C indicated 2-d was the only ruthenium species
formed in solution, with concomitant formation of HD gas.
(19) 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.

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2233
Sch em e 2
dideuterium gas (1 atm) results in H-D exchange with
the pro-R-â-hydride. Further, reaction of 2-d (made by
reaction of 2 with dideuterium gas (eq 6)) with (Z)-MAC
produced (SC ,RC )-â-d1-1 as the only detectable ruthe-
is unknown, the 85:15 ratio of (R,R and S,S)-MACH2-
R,â-d2 to MACH2-R-d1 represents the lower limit to the
extent of reversibility of formation of (SC )-1 during the
R
deuteriolysis. We were unable to isolate MACH2-R-d1
in order to determine its ee, which would be 100% (R)
R
â
nium species in solution. The stereochemistry of the
olefin-hydride insertion results in a net cis addition of
ruthenium and hydrogen across the carbon-carbon
double bond of (Z)-MAC. Assuming retention of con-
figuration at the R-carbon, deuteriolysis of the ruthe-
nium-carbon bond in (SC ,RC )-â-d1-1 will produce
if it formed from direct deuteriolysis of (SC )-1 without
R
prior formation of 2 and (Z)-MAC. Since the lower limit
to the extent to which formation of (SC )-1 reversed is
R
∼85%, equilibration between (SC )-1, 2, and (Z)-MAC
R
must be relatively rapid and nearly complete before
deuterolysis of the ruthenium-carbon bond occurs
under these conditions.
R
â
(R,R)-MACH2-R,â-d2, the product of net cis addition of
dideuterium to the carbon-carbon double bond of (Z)-
MAC. (S,S)-MACH2-R,â-d2 would result from coordina-
tion of (Z)-MAC to 2-d via the re-olefin face followed by
deuteriolysis of the ruthenium-carbon bond.
As discussed above, â-hydride elimination and dis-
sociation of (Z)-MAC from 1 during the stoichiometric
deuteriolysis will produce (Z)-MAC and 2 in the pres-
ence of excess dideuterium gas (Scheme 2). Compound
2 in this mixture can either react with (Z)-MAC to
Only the (R,R) and (S,S) diastereomers of MACH2-
R,â-d2 were produced by the deuteriolysis of (SC )-1. This
R
stereoselectivity is a consequence of the direct relation-
ship between whether â-hydride or â-deuteride elimina-
tion occurred within (SC ,RC )-â-d1-1 and the stereo-
R
â
chemistry of the resulting MAC. This relationship also
exists in the opposite diastereomer (RC ,SC )-â-d1-1. As
R
â
shown in Scheme 3, (Z)-MAC will result from â-deu-
teride elimination and the less stable (E)-MAC-â-d1 will
result from â-hydride elimination. Reaction of (E)-MAC-
â-d1 with 2-d followed by deuteriolysis of the ruthenium-
carbon bond must generate MACH2-R,â,â-d3.20 Since
MACH2-R,â,â-d3 was not detected among the products,
regenerate (SC )-1 (leading to MACH2-R-d1 after deute-
R
riolysis of the ruthenium-carbon bond) or it can react
with dideuterium gas to produce 2-d (leading eventually
to (R,R and S,S)-MACH2-R,â-d2). Since the relative rate
of formation of (SC )-1 versus that of 2-d in this mixture
R

2234 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens
Sch em e 3
we conclude that â-hydride elimination within (SC )-1
is highly stereoselective, favoring the pro-R-â-hydride
deuteration of (E)-MAC and (Z)-MAC were carried out
to investigate the stereoselectivity of the catalytic
hydrogenation.
R
in (SC )-1 (this position is occupied by deuterium in
R
(SC ,RC )-â-d1-1) and the pro-S-â-hydride in (RC )-1 (this
Ster eoselectivity a n d Regioselectivity of th e
Ca ta lytic Hyd r ogen a tion . The catalytic deuteration
of (Z)-MAC in acetone generated (R,R and S,S)-MACH2-
R,â-d2 as the only detectable diastereomers in 92% ee
(R,R) (eq 7). The net cis addition of dideuterium gas
R
â
R
position is occupied by deuterium in (RC ,SC )-â-d1-1).
R
â
The origins of this stereoselectivity likely derive from
the unfavorability of formation of (E)-MAC over forma-
tion of (Z)-MAC. Similar control over the stereoselec-
tivity of â-hydride elimination was observed during the
intramolecular hydrosilylation of olefins catalyzed by
[Rh((S)-BINAP)(acetone)2](ClO4).21 We note that it is
possible that elimination of the pro-S-â-hydride does
occur within (SC )-1, but the resulting (E)-MAC does not
R
dissociate from ruthenium but rather reinserts into the
ruthenium-hydride bond to regenerate (SC )-1. The
R
stoichiometric deuteriolysis of (SC )-1 is mute on this
R
point because â-hydride elimination within (SC ,RC )-â-
R
â
across (Z)-MAC, although consistent with the stereo-
d1-1 will be of no net consequence if the resulting (E)-
MAC-â-d1 reinserts into the ruthenium-hydride bond
without prior to olefin dissociation. We find it unlikely,
however, that (Z)-MAC will dissociate from ruthenium
but (E)-MAC will not.
chemistry of the stoichiometric deuteriolysis of (SC )-1,
R
it is also consistent with any mechanism involving
olefin-hydride insertion followed by reductive elimina-
tion of a carbon-hydrogen bond. The net cis addition
implies, however, that if the catalytic hydrogenation
The most likely explanation for formation of the other
minor products of the stoichiometric deuteriolysis, (R,R)-
MACH2-â-d1 (6%) and MACH2 (∼2%), is solvolysis of the
ruthenium-carbon bonds in (SC ,RC )-â-d1-1 and (SC )-
proceeds through (SC )-1 or a similar species, any
R
â-hydride elimination which occurred was highly ste-
reoselective, favoring formation of (Z)-MAC.
The catalytic hydrogenation and deuteration of (E)-
MAC were carried out to investigate if formation of
R
â
R
1, respectively, by adventitious water or by acetone.22
We conclude from the predominant formation of (R,R
and S,S)-MACH2-R,â-d2 by the stoichiometric deuteri-
(SC )-1 (Figure 1) or of a similar species is reversible
R
during the catalytic hydrogenation (eq 4). The catalytic
hydrogenation of (E)-MAC in acetone produced MACH2
in 91% ee (R), which is similar to the ee for hydrogena-
tion of (Z)-MAC (92% (R)). The similar ee’s indicate that
both hydrogenations proceed through the same olefin
geometry, implying that the hydrogenation of (E)-MAC
proceeds via a rapid prior isomerization to (Z)-MAC.23
The catalytic deuteration of (E)-MAC was carried out
to determine if olefin isomerization occurs prior to
formation of MACH2. Insertion of (E)-MAC into a
ruthenium-deuterium bond with placement of ruthe-
nium at the R-carbon followed by â-elimination favoring
the (Z) isomer would form (Z)-MAC-â-d1 with deuterium
at the â-position. Deuteration of (Z)-MAC-â-d1 would
then lead to MACH2-R,â,â-d3. Figure 3 shows the
distribution of isotopomers produced by the catalytic
deuteration of (E)-MAC in acetone (MACH2-R,â,â-d3,
69%; (R,R and S,S)-MACH2-R,â-d2, 20%; MACH2-â,â-
olysis of (SC )-1 that olefin coordination, olefin-hydride
R
insertion, and their reverse transformations are rapid
relative to the hydrogenolysis of the ruthenium-carbon
bond under these reaction conditions (which include a
build up of MeCN in solution as the reaction proceeds).
We also conclude that â-hydride elimination within
(SC )-1 is highly stereoselective, favoring formation of
R
the more stable olefin isomer, (Z)-MAC. The catalytic
(20) Assuming, for the sake of argument that â-hydride elimination
did not occur during the deuteration of (E)-MAC, the absolute config-
uration of MACH2-R,â,â-d3 resulting from deuteration of (E)-MAC-â-
d1 would depend on the enantioselectivity of the catalyst toward (E)-
MAC.
(21) (a) Bergens, S. H.; Noheda, P.; Whelan, J .; Bosnich, B. J . Am.
Chem. Soc. 1992, 114, 2121-2128. (b) Bergens, S. H.; Noheda, P.;
Whelan, J .; Bosnich, B. J . Am. Chem. Soc. 1992, 114, 2128-2135.
(22) Other explanations include reaction via the small amount of
HD gas generated by reaction of 2 with D2 and olefin-hydride insertion
of (Z)-MAC into 2 via the opposite regiochemistry (with ruthenium at
the â-olefinic carbon and hydrogen at the R-olefinic carbon) followed
by deuteriolysis of the ruthenium-carbon bond. Reaction with HD gas
cannot be ruled out, but it is unlikely because the HD gas generated
by reaction of 2 with D2 amounts to ∼2% of the D2 in the reactor and
because the concentration of 2 relative to 2-d is probably quite low.
(23) Another possibility, which we do not rule out, is that the
intrinsic face selectivity of the hydrogenation toward (E)-MAC and (Z)-
MAC are the same.

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2235
detectable product from the catalytic deuteration of (Z)-
MAC (eq 7).
The amount of MACH2-R,â,â-d3 in the product mix-
ture (69%) is the lower limit to the extent of isomeriza-
tion of (E)-MAC to (Z)-MAC prior to deuteration because
we propose that (R,R and S,S)-MACH2-R,â-d2 (20%) also
arose from such a process. As discussed previously,
(R,R and S,S)-MACH2-R,â-d2 (20%) results from a net
cis addition of dideuterium to (Z)-MAC (eq 7). We
believe deuterium-free (Z)-MAC was produced during
the catalytic deuteration of (E)-MAC through reaction
of (E)-MAC with 2, which resulted from the stereose-
lective â-hydride elimination within (RC ,RC )-â-d1-1 or
R
â
(SC ,SC )-â-d1-1 (Scheme 4, top, and as discussed in the
R
â
previous paragraph). It is reasonable to propose that
rather than reacting with dideuterium gas to generate
2-d, some of 2 reacted with (E)-MAC (which is in up to
50-fold excess under catalytic conditions) to generate a
deuterium-free diastereomer of (SC )-1 or structurally
R
related species (Scheme 4, middle). (Z)-Stereoselective
â-hydride elimination followed by substrate dissociation
will generate 2 and deuterium-free (Z)-MAC. Deutera-
tion of (Z)-MAC then generates (R,R and S,S)-MACH2-
R,â-d2 (20%).
Scheme 5 shows another conceivable route to forma-
tion of deuterium-free (Z)-MAC during the catalytic
deuteration of (E)-MAC.3h,i In this sequence, 2-d acti-
vates the N-H bond of (E)-MAC to generate the cor-
responding amido-allyl complex 5. π-σ-π rearrange-
ment through formation of a ruthenium-carbon σ bond
at the â-position would form the amido-allyl complex
5′, which could eliminate (Z)-MAC-N-d. Deuteration of
(Z)-MAC-N-d will generate (R,R and S,S)-MACH2-
R,â,N-d3. As mentioned previously,17 varying amounts
of product deuterated at nitrogen were observed in all
reactions carried out under dideuterium gas in acetone
solution. Control experiments show that 2 catalyzes the
H-D exchange between dideuterium and the N-H
group of MACH2 and (by extension) presumably also of
MAC. The formation of N-D products by the catalytic
reaction under dideuterium gas, therefore, cannot be
used as direct evidence for involvement of amido-allyl
(5, 5′) or related species in the catalytic hydrogena-
tion. We have no experimental evidence that conclu-
sively distinguishes between reaction of 2 with (E)-MAC
F igu r e 3. The distribution of MACH2-dn isotopomers from
the catalytic deuteration of (E)-MAC using 2 as catalyst
in acetone. The isotopomers (R,R and S,S)-MACH2-â-d1 and
MACH2-R-d1 (combined 3% of total mixture) are not shown.
d2, 4%; (R,S and S,R)-MACH2-R,â-d2, 4%; (R,R and S,S)-
MACH2-â-d1, ∼1%; and MACH2-R-d1, ∼2%). The com-
bined ee of the products was 92% (R,R and R). MACH2-
R,â,â-d3 amounted to 69% of the product mixture, which
is strong evidence that prior isomerization of (E)-MAC
to (Z)-MAC occurred during the catalytic deuteration.
Scheme 4 shows the sequence of steps that accounts for
this distribution of isotopomers via a rapid formation
of (SC )-1 or a related species which is reversible via a
R
stereoselective â-hydride elimination, as shown for the
stoichiometric deuteriolysis of (SC )-1. Coordination of
R
(E)-MAC to 2-d through the re-face or the si-face
followed by olefin-hydride insertion generates (RC ,RC )-
R
â
or (SC ,SC )-â-d1-1, respectively (Scheme 4, upper left).
R
â
to form (SC )-1 (Scheme 4, middle) or formation of an
R
Stereoselective â-hydride elimination within (RC ,RC )-
R
â
amido-allyl species (Scheme 5, 5 and 5′) to account for
formation of (Z)-MAC during the catalytic deuteration.
Evidence against an amido-allyl species, however, is
that little, if any, H-D exchange between nitrogen and
the R- or â-carbon centers occurs during the catalytic
deuteration of (Z)-MAC in methanol solvent (vide infra,
eq 8). It is reasonable to expect amido-allyl species
such as 5 and 5′ to some extent undergo deuteride
addition to the â-carbon-forming species such as 6
and 6′ (Scheme 5). Among other things, these sequences
of steps would allow H-D exchange between nitro-
gen and the â-carbon. The lack of an observable amount
of such exchange during the catalytic deuteration of
(Z)-MAC in methanol favors formation of (Z)-MAC
via reaction of 2 with (E)-MAC (Scheme 4, middle) over
an amido-allyl route.
â-d1-1 or (SC ,SC )-â-d1-1 to favor formation of (Z)-MAC
R
â
followed by olefin dissociation generates (Z)-MAC-â-d1,
2, and subsequently 2-d (by reaction of 2 with dideu-
terium (eq 6)). Deuteration of (Z)-MAC-â-d1 using 2-d
as the catalyst generates MACH2-R,â,â-d3, the predomi-
nant product of the catalytic deuteration (69%, Scheme
4, upper right). An initial isomerization of (E)-MAC to
(Z)-MAC via the same sequence of steps likely accounts
for the catalytic hydrogenations of (E)-MAC and (Z)-
MAC occurring with similar ee’s. The formation of
MACH2-R,â,â-d3 is strong evidence that at least 69% of
(E)-MAC underwent isomerization to (Z)-MAC-â-d1 via
the reversible formation of (SC )-1 or of a structurally
R
related species during the catalytic reaction (although
not necessarily as part of the catalytic cycle). Further,
a rapid and reversible formation of (SC )-1 via stereo-
R
selective â-hydride elimination is also consistent with
formation of (R,R and S,S)-MACH2-R,â-d2 as the only
The major product of the catalytic deuteration of (E)-
MAC, MACH2-R,â,â-d3 (69%), resulted from formation

2236 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2237
Sch em e 5
Meth a n ol a s th e Solven t. The catalytic deuteration
of (Z)-MAC was carried out in methanol to determine
the ratio of solvolysis to hydrogenolysis during catalysis
in this protic solvent. As shown in eq 8, the product
ratio was (R,R and S,S)-MACH2-R,â-d2, 96%; and (R,R
and S,S)-MACH2-â-d1, 4% (combined ee ) 89% (R,R)).
The low percentage of MACH2 with hydrogen at the
R-position shows that solvolysis of a ruthenium-carbon
bond is not a major pathway of the catalytic hydrogena-
tion in methanol. The 96% relative abundance of (R,R
and S,S)-MACH2-R,â-d2 also shows that little H-D
exchange occurs between nitrogen and the R- and
â-carbon centers during the catalytic hydrogenation.
This is evidence against involvement of an amido-allyl
(Scheme 5, 5 or 5′) or related species in the catalytic
cycle. We note that although the catalyst does effect
H-D exchange between dideuterium and the N-H
group of MACH2 and (by extension) likely of (Z)-MAC,
any deuterium substitution at nitrogen that arose from
such a process would immediately H-D exchange with
the methanol solvent, effectively maintaining a proton
at nitrogen throughout the catalytic hydrogenation.
The stoichiometric hydrogenolysis of (SC )-1 slowed
R
as the reaction proceeded in acetone (eq 2) but went to
completion in methanol. In methanol solvent, the rate
of the stoichiometric hydrogenolysis was similar to the
turnover frequency of the catalytic hydrogenation.6 The
of (SC )-1 or a related species via a rapid net cis addition
R
of ruthenium and deuterium across the carbon-carbon
double bond, followed by a rapid stereoselective â-hy-
dride elimination. We propose that it is likely that a
further 20% of the product, (R,R and S,S)-MACH2-R,â-
d2, also resulted from such a sequence of steps. We
therefore conclude from the product distributions of the
catalytic deuteration of (E)-MAC and (Z)-MAC in ac-
etone that the sum of the stereoselectivities and regio-
stoichiometric deuteration of (SC )-1 was carried out in
R
methanol to investigate the origins of this discrepancy
in behavior between acetone and methanol solvents.
Figure 4 shows the distribution of organic products from
the stoichiometric deuteriolysis of (SC )-1 in methanol
R
(eq 9; (R,R and S,S)-MACH2-R,â-d2, 58%; MACH2-R-d1,
5%; (R,R and S,S)-MACH2-â-d1, 16%; and MACH2,
21%). The combined ee was 85% (R,R and R), similar
selectivities of the reversible formation of (SC )-1 and
R
its hydrogenolysis equals the overall stereoselectivity
and regioselectivity of the catalytic hydrogenation.
One of the minor products of the catalytic deuteration
of (E)-MAC, (R,S and S,R)-MACH2-R,â-d2 (4%), likely
arose from the direct deuteration of (E)-MAC without
prior isomerization to (Z)-MAC (Scheme 4, left, deute-
riolysis of (SC ,SC )-â-d1-1)). Of the remaining minor
R
â
products of the catalytic deuteration of (E)-MAC, we
speculate that MACH2-R-d1 (∼2%) resulted from deu-
teriolysis of 1 or from reaction of HD gas with (E)-MAC
catalyzed by 2. (R,R and S,S)-MACH2-â-d2 (1%) and
MACH2-â,â-d2 (4%) likely arose from solvolysis of the
ruthenium-carbon bond in (RC ,SC and SC ,RC )-â-d1-1
R
â
R
â
and â,â-d2-1, respectively by adventitious water.

2238 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens
influence the catalytic hydrogenation as the turnover
frequency is higher in acetone than in methanol, at least
980 turnovers can be achieved in acetone, and the
catalytic hydrogenation in methanol solution does not
proceed via solvolysis of a ruthenium-carbon bond.
Con clu sion s
This work is the first mechanistic study of a ruthe-
nium-BINAP-catalyzed hydrogenation to be compli-
mented by a solid-state structure of a possible interme-
diate with a prochiral substrate group bonded to
ruthenium. We conclude from the results of this study
that the stereoselectivities and the regioselectivities of
the rapid reversible formation of (SC )-1 and its conver-
R
sion to MACH2 (via irreversible hydrogenolysis of the
ruthenium-carbon bond) summate to give the stereo-
selectivity and regioselectivity of the catalytic hydroge-
nation. We also conclude that solvolysis of the ruthe-
nium-carbon bond is a minor (<4%) pathway of the
catalytic hydrogenation in methanol solvent and that
the presence of MeCN in the system does not contribute
to the enantioselection of the catalytic reaction. Al-
though (SC )-1 is the only detectable species in solution
R
during the catalytic reaction at room temperature, the
data from this study do not distinguish between whether
(SC )-1 is a catalytic intermediate or a species in
R
F igu r e 4. The distribution of MACH2-dn isotopomers from
the deuteriolysis of (ScR)-1 in methanol.
equilibrium with the true catalytic intermediate. Fur-
ther, a direct comparison between the stoichiometric
and catalytic rate data are complicated by both the build
up of MeCN, which occurs in solution during the
stoichiometric hydrogenolysis, and by the ability of
to the catalytic hydrogenation in methanol (87% (R)).
Only the (R,R)- and (S,S)-diastereomers of MACH2-R,â-
d2 were detected, showing that â-elimination within
(SC )-1 to undergo solvolysis in methanol. Both of these
R
(SC )-1 proceeds by the same stereoselectivity as that
R
processes do not occur during the catalytic hydrogena-
tion. A kinetic study to investigate these issues is
underway in our laboratories.
observed in acetone. The ratio of (R,R and S,S)-MACH2-
R,â-d2 to MACH2-R-d1 (∼12:1) also indicates that the
reversible formation of (SC )-1 was faster than deute-
riolysis of the ruthenium-Rcarbon bond. The MACH2
(21%) and (R,R and S,S)-MACH2-â-d1 (16%) resulted
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All glassware was sequentially
treated with ethanolic ammonia solution and acetone and dried
overnight in an oven at 120 °C before use. Reactions involving
organometallic compounds were conducted under an atmo-
sphere of dry argon gas (Praxair, 99.998%) using standard
Schlenk techniques. Organometallic products were isolated
in a drybox filled with dinitrogen gas (Praxair, 99.998%) and
stored at -30 °C for prolonged periods.
from solvolysis of the ruthenium-carbon bond in (SC )-1
R
and in (SC ,RC )-â-d1-1, respectively. The total amount
R
â
of MACH2 and (R,R and S,S)-MACH2-â-d1 (37%) is
likely a direct indication of the amount of solvolysis
that occurred under these conditions, because control
experiments show that 2 does not catalyze the H-D
exchange between dideuterium gas and methanol at an
appreciable rate. At this stage, we conclude that the
similarity of rates for the stoichiometric hydrogenolysis
in methanol and for the catalytic hydrogenation in
methanol is coincidental and is not of mechanistic
significance.
P h ysica l a n d Sp ectr a l Mea su r em en ts. NMR spectra
2
were recorded using Bruker AM-400 (1H at 400.1 MHz, H at
61.4 MHz, 13C at 100.6 MHz, and 31P at 161.9 MHz) and
Bruker AM-R-300 (13C CP/MAS at 75.5 MHz) spectrometers.
1
The chemical shifts for H, 2H, and 13C were measured relative
to external TMS and referenced to residual solvent signals;
those signals for 31P were measured relative to an 85% H3PO4
external reference. Routine electron-impact high-resolution
mass spectrometry (HRMS (EI)) of organic compounds were
performed on a Kratos MS50 spectrometer. Mass spectromet-
ric analyses of organometallic compounds were performed by
positive-mode electrospray ionization (ESI-MS (pos)) on a
Micromass ZabSpec Hybrid Sector-TOF spectrometer. Cal-
Unlike the reaction in acetone solution, the stoichio-
metric hydrogenolysis of (SC )-1 in methanol solution
R
was not significantly inhibited, even by the addition of
a 20-fold excess of MeCN. The reasons why the sto-
ichiometric hydrogenation of (SC )-1 is not appreciably
R
inhibited by MeCN in methanol solution likely relate
to the ability of (SC )-1 to undergo solvolysis in methanol
R
1
culated m/z values refer to the isotopes 12C, H, 14N, 16O, 31P,
to generate MACH2. As solvolysis likely proceeds by
and 102Ru. Optical rotations were measured with a Perkin-
Elmer 241 polarimeter at 589 nm using 1.0 dm cells. Mi-
protonation of the ruthenium center in (SC )-1, we
R
suspect that solvolysis is not significantly inhibited by
MeCN because protonation of a metal center does not
require a vacant coordination site.24 Whatever their
origins, the relative susceptibilities of the stoichiometric
(24) (a) Crabtree, R. H.; Quirk, J . M.; Fillebeen-Khan, T.; Morris,
G. E. J . Organomet. Chem. 1979, 181, 203-212. (b) Ashworth, T. V.;
Singleton, J . E.; de Waal, D. J . A.; Louw, W. J .; Singleton, E.; van der
Stok, E. J . Chem. Soc., Dalton Trans. 1978, 340-347. See also,
electrophilic attack of HgBr2 on Cp*Os(CO)(PMe2Ph)Me: Sanderson,
L. J .; Baird, M. C. J . Organomet. Chem. 1986, 307, C1-C4.
hydrogenolysis of (SC )-1 to inhibition by MeCN in
R
acetone and in methanol solution do not appreciably

Hydrogenation Catalyzed by a Ru-BINAP Complex
Organometallics, Vol. 17, No. 11, 1998 2239
croanalyses were performed at the University of Alberta
Microanalysis Laboratory.
resulting dark amber solution was shaken, and the solvent
was removed under reduced pressure to give a yellow solid
discolored by a brown residue. Dropwise addition of diethyl
ether (80 mL) to a methylene chloride solution (2.5 mL) of the
product afforded a yellow powder. The product was filtered
off, washed with diethyl ether (5 × 10 mL), and dried under
dynamic vacuum (16 h) to yield 534.1 mg (87%) of (SCR)-1. NMR
Ma t er ia ls. All solvents (reagent grade) were purchased
from Caledon and Fisher Scientific, except deuterated solvents
(99.8-99.9% D) which were purchased from Cambridge Isotope
Laboratories. Acetone (3 Å molecular sieves), acetone-d6 (3 Å
molecular sieves), acetonitrile (phosphorus pentoxide), diethyl
ether (lithium aluminum hydride), methanol (magnesium
methoxide), methanol-d4 (magnesium methoxide-d6), methyl-
ene chloride (calcium hydride), methylene chloride-d2 (calcium
hydride), tetrahydrofuran (potassium benzophenone ketyl),
and tetrahydrofuran-d8 (potassium) were distilled from the
appropriate drying agents under an atmosphere of argon gas
prior to use. All other solvents were used as received. Dihy-
drogen gas (Praxair, 99.99%) was passed through an Alltech
Oxy-Trap to remove trace amounts of oxygen, whereas dideu-
terium gas (Aldrich, 99.8%; Praxair, 99.7%) was used as re-
ceived. R-Acetamidocinnamic acid was purchased from Aldrich
and recrystallized 5 times from ethanol/hexanes before use.
(Z)-Methyl R-acetamidocinnamate was prepared by the esteri-
fication of R-acetamidocinnamic acid using diazomethane.25,26
(Z)-Methyl-4-benzaloxazolone,27 (E)-methyl R-acetamidocin-
namate,26 and [Ru((R)-BINAP)(1-3;5-6-η-C8H11)(MeCN)]-
(BF4)7 were prepared using established procedures. (rac)-N-
Acetylphenylalanine methyl ester was obtained by hydrog-
enating (Z)-methyl R-acetamidocinnamate using [Rh(η4-nor-
bornadiene)(1,2-bis(diphenylphosphino)ethane)](ClO4)28 as the
catalyst. All other materials were purchased from Aldrich
with the exception of (S)-N-acetylphenylalanine methyl ester
(Serva) and Florisil (60-100 mesh, Fisher Scientific).
(Z)-Meth yl-d 3 r-Aceta m id ocin n a m a te. To a stirred sus-
pension of (Z)-methyl-4-benzaloxazolone (459.7 mg, 2.46 mmol)
in benzene (5 mL) at room temperature, a ∼1 M sodium
methoxide solution (prepared from sodium (73.8 mg, 3.21
mmol) and methanol (3 mL)) was added. The resulting pale
yellow solution was allowed to stir for 5 min, after which the
solution was acidified with aqueous 0.1 M HCl. The slightly
opaque benzene layer was extracted with methylene chloride
(3 × 10 mL) and washed with water (3 × 20 mL). The organic
layer was dried over anhydrous magnesium sulfate, filtered,
and evaporated to yield 226.4 mg (41%) of (Z)-Methyl-d3
R-acetamidocinnamate. NMR spectroscopic data (50 °C): 1H
(400.1 MHz, CDCl3) δ 2.00 (br s, 3H, NHCOCH3), 7.30 (m, 5H),
7.45 (br s, 2H); 2H{1H} (61.4 MHz, CHCl3) δ 3.80 (s, CO2CD3);
13C{1H} (100.6 MHz, CDCl3) δ 22.8 (br s, NHCOCH3), 124.9
(quaternary), 128.5 (methine), 129.4 (methine), 129.7 (meth-
ine), 132.5 (br s, methine), 133.7 (quaternary), 165.9 (CO2CH3),
169.4 (br s, NHCOCH3). The methoxy 13C resonance appears
at δ 52.4 ppm for (Z)-methyl R-acetamidocinnamate. HRMS
(EI, direct insert) m/z 222.1083 (M+, exact mass calcd for
C12H10D3NO3 222.1084). Anal. Calcd for C12H13NO3: C, 65.74;
H, 5.98; N, 6.39. Found: C, 65.39; H, 5.93, N, 6.19.
spectroscopic data for (SC )-1 (CD2Cl2, 25 °C): 1H (400.1 MHz)
R
δ 1.71 (s, 3H, CH3CN), 1.95 (s, 3H, NHCOCH3), 2.70 (dd, 2J H-H
4
) 14.0 Hz, J P-H ) 4.5 Hz, 1H, pro-R-CH2Ph), 3.98 (s, 3H,
2
CO2CH3), 4.15 (d, J H-H ) 14.0 Hz, 1H, pro-S-CH2Ph), 5.87
5
(br d, J P-H ) 2.7 Hz, 1H, NHCOCH3), 6.40-8.00 (aromatic);
2
31P{1H} (161.9 MHz) δ 33.6 (d, J P-P ) 23.8 Hz, 1P), 59.7 (d,
2J P-P ) 23.8 Hz, 1P); 13C{1H} (100.6 MHz) δ 4.7 (s, CH3CN),
20.9 (s, NHCOCH3), 38.5 (s, CH2Ph), 53.3 (s, CO2CH3), 67.3
2
2
(dd, J C-P
) 42.2 Hz, J C-P ) 3.9 Hz, Ru-C), 126-143
trans
cis
3
(aromatic and CH3CN), 160.8* (d, J C-P ) 3.3 Hz), 179.7* (d,
3J C-P ) 7.0 Hz); 13C CP/MAS (75.5 MHz) δ 6, 21, 39, 56, 69,
128, 134, 161, 181. ESI-MS (pos) m/z 985.2 ((M - BF4)+, exact
mass calcd for C58H49N2O3P2Ru 985.2). Anal. Calcd for
C58H49BF4N2O3P2Ru: C, 64.99; H, 4.61; N, 2.61. Found: C,
63.85; H, 4.65; N, 2.86. The signals marked with an asterisk
are those of the amide carbonyl and ester carbonyl carbons.
Isotopic-labeling experiments have not been conducted to
unambiguously assign these signals.
[R u ((R)-BINAP )(H )(η6-(r a c)-MACH 2)](BF 4). (rac)-N-
Acetylphenylalanine methyl ester (15.1 mg, 0.068 mmol) and
[Ru((R)-BINAP)(1-3;5-6-η-C8H11)(MeCN)](BF4) (65.5 mg, 0.068
mmol) were dissolved in acetone (8.0 mL) under an atmosphere
of argon gas and subjected to 3 freeze-pump-thaw cycles. The
reactor was backfilled with dihydrogen gas (20 psig) at room
temperature and shaken 10 min. The resulting solution was
concentrated (∼1 mL) under reduced pressure. Addition of
diethyl ether (∼100 mL) followed by filtration afforded a yellow
powder. The product was washed with diethyl ether (5 × 10
mL) and dried under dynamic vacuum (24 h) to yield 39.4 mg
(56%) of [Ru((R)-BINAP)(H)(η6-(rac)-MACH2)](BF4). A reaction
mixture monitored by 1H and 31P NMR spectroscopy confirmed
that the two diastereomeric complexes were formed in quan-
titative yield. The individual diastereomer, [Ru((R)-BINAP)-
(H)(η6-(S)-MACH2)](BF4), was prepared using (S)-N-acetylphe-
nylalanine methyl ester in
a manner analogous to that
described above. NMR spectroscopic data (CD2Cl2, 25 °C) for
(RBINAP,SMACH2) diastereomer: 1H (400.1 MHz) δ -9.14 (dd,
2
2J P-H ) 40.3 Hz, J P-H ) 29.3 Hz, 1H, Ru-H), 1.83 (s, 3H,
NHCOCH3), 2.51 (dd, 2J H
) 13.8 Hz, 3J H â ) 5.0 Hz, 1H,
-H
R
-H
â
â′
2
3
C6H5CH2CH (Hâ)), 2.82 (dd, J H
) 13.8 Hz, J H
) 7.8
-H
â
-H
R
â′
â′
Hz, 1H, C6H5CH2CH (Hâ′)), 3.62 (s, 3H, CO2CH3), 4.40 (d, 3J H-H
) 6.1 Hz, 1H, C6H5CH2CH), 4.47 (d of app t, 3J H
)
-H
R
â′
3
3J H
) 7.8 Hz, J H
) 5.0 Hz, 1H, C6H5CH2CH (HR)),
-H
R
-H
R
[Ru ((R)-BINAP )((S)-MACH)(MeCN)](BF4) ((SCr)-1). [Ru-
((R)-BINAP)(1-3;5-6-η-C8H11)(MeCN)](BF4) (547.5 mg, 0.571
mmol) was partially dissolved in acetone (20.0 mL) under an
atmosphere of dry argon gas and subjected to 3 freeze-pump-
thaw cycles. The reactor was backfilled with dihydrogen gas
(20 psig) at room temperature and vigorously shaken for 10
min to generate a clear, dark orange solution. This extremely
air-sensitive solution was subjected to 2 freeze-pump-thaw
cycles and backfilled with argon gas. To this solution at room
temperature, an acetone solution (5.0 mL) of (Z)-methyl
R-acetamidocinnamate (125.2 mg, 0.571 mmol) was added. The
N-H
â
3
3
4.55 (app t, J H-H ) 6.1 Hz, 1H, C6H5CH2CH), 5.45 (d, J H-H
3
) 6.2 Hz, 1H, C6H5CH2CH), 5.78 (app t, J H-H ) 6.1 Hz, 1H,
3
C6H5CH2CH), 6.03 (app t, J H-H ) 5.9 Hz, 1H, C6H5CH2CH),
6.15-8.20 (BINAP); 31P{1H} (161.9 MHz) δ 50.4 (d, J P-P
)
2
45.5 Hz, 1P), 51.5 (d, 2J P-P ) 45.5 Hz, 1P). NMR spectroscopic
data (CD2Cl2, 25 °C) for (RBINAP,RMACH2): 1H (400.1 MHz) δ
-9.06 (dd, 2J P-H ) 36.0 Hz, 2J P-H ) 33.6 Hz, 1H, Ru-H), 1.88
2
3
(s, 3H, NHCOCH3), 2.56 (dd, J H
Hz, 1H, C6H5CH2CH (Hâ)), 2.98 (dd, 2J H â′ ) 14.0 Hz, 3J H
) 14.0 Hz, J H â ) 8.0
-H
R
-H
â
â′
-H
â
-H
R â′
) 5.0 Hz, 1H, C6H5CH2CH (Hâ′)), 3.62* (s, CO2CH3), 4.43*
(C6H5CH2CH (HR)), 4.89 (d, 3J H-H ) 6.0 Hz, 1H, C6H5CH2CH),
(25) Casey, M.; Leonard, J .; Lygo, B.; Procter, G. Advanced Practical
Organic Chemistry; Chapman & Hall: London, 1990; pp 70-73.
(26) Vineyard, B. D.; Knowles, W. S.; Sabacky, M. J .; Bachman, G.
L.; Weinkauff, D. J . J . Am. Chem. Soc. 1977, 99, 5946-5952.
(27) Herbst, R. M.; Shemin, D. In Organic Syntheses; Blatt, A. H.,
Ed.; Wiley: New York, 1943; Collect. Vol. II, pp 1-3.
(28) Schrock, R. R.; Osborn, J . A. J . Am. Chem. Soc. 1971, 93, 2397-
2407.
3
3
4.95 (app t, J H-H ) 6.0 Hz, 1H, C6H5CH2CH), 5.06 (d, J H-H
3
) 6.0 Hz, 1H, C6H5CH2CH), 5.60 (app t, J H-H ) 6.0 Hz, 1H,
3
C6H5CH2CH), 6.03 (app t, J H-H ) 6.0 Hz, 1H, C6H5CH2CH),
6.00-8.20 (BINAP); 31P{1H} (161.9 MHz) δ 51.1 (d, J P-P
)
2
2
45.0 Hz, 1P), 51.4 (d, J P-P ) 45.0 Hz, 1P). ESI-MS (pos) m/z
946.2 ((M - BF4)+, exact mass calcd for C56H48NO3P2Ru 946.2).

2240 Organometallics, Vol. 17, No. 11, 1998
Wiles and Bergens
R,â-d2, and MACH2-R,â,â-d3). Of these seven isotopomers, six
were identified and quantified by their 1H{2H} NMR spectra.
The seventh isotopomer, MACH2-â,â,â-d3 was identified by
HRMS (EI) and its relative amount was determined from the
methoxy internal standard signal after subtracting the con-
tributions from the other isotopomers. The 1H{2H} NMR
spectrum of each isotopomer was in accord with line patterns
predicted by substitution of the appropriate proton(s) by
deuteron(s) in the assigned ABX system of MACH2. For
example, (R,R and S,S)-MACH2-R,â-d2 was idenified by its
F igu r e 5. Designations used for the ABX spin system in
the 1H NMR spectrum of N-acetylphenylalanine methyl
ester (MACH2).
Anal. Calcd for C56H48BF4NO3P2Ru: C, 65.12; H, 4.68; N, 1.36.
Found: C, 64.39; H, 4.67; N, 1.68. Signals marked with an
asterisk are overlapped by resonances of the (RBINAP,SMACH
)
proton signal (δ 3.03 (s, 1H, Hâ )) which showed the absence
2
2
diastereomer.
of coupling to other protons and was shifted slightly upfield
from the Hâ signal for nondeuterated MACH2.30 Similarly,
Gen er a l P r oced u r e for Ca ta lytic Hyd r ogen a tion s. A
glass pressure reactor (Lab Glass) was employed for hydro-
genation reactions where the pressure of dihydrogen gas was
e4 atm. For reactions requiring elevated pressures, a Parr
cell-disruption bomb was used. In general, the appropriate
pressure reactor was charged with substrate and catalyst (2
mol %) under an atmosphere of argon gas. The reaction vessel
was flushed with dihydrogen gas and its contents dissolved
in the appropriate deoxygenated solvent ([catalyst] ) 2.6 mM).
The reactor was pressurized (dynamic pressure)/heated to the
desired levels and allowed to react for, typically, 48 h while
stirring at 1100 rpm. Pressures were quoted as gauge pres-
sure plus 1 atm. After complete conversion of reactant to
product (confirmed by 1H NMR spectroscopy and HRMS (EI)),
the mixtures were evaporated under reduced pressure and the
residues analyzed without further purification. Enantiomeric
excesses were spectroscopically determined (1H NMR) via
chiral lanthanide shift reagent29 (europium tris[3-(trifluorom-
ethylhydroxymethylene)-(+)-camphorate] ((+)-Eu(tfc)3)) in chlo-
roform-d. The ratio of the methoxy signals (ca. 4 ppm; 1:1 for
(rac)-N-acetylphenylalanine methyl ester) was used to quantify
all enantiomeric excesses. In all cases, addition of (S)-N-
acetylphenylalanine methyl ester caused a decrease in enan-
tiomeric excess, indicating the absolute configuration of the
major enantiomer was R. The catalyst residue was removed
before polarimetry by passing an ethyl acetate solution of the
reaction mixture through a column of Florisil. The enantio-
meric excess and absolute configuration of the major enanti-
omer for this mixture were determined as described above and
confirmed by the magnitude and the sign of the optical
rotation,3f respectively, from polarimetry measurements.
Gen er a l P r oced u r e for Ca ta lytic Deu ter a tion s. Reac-
tions were performed in a manner similar to that described
above with the following exceptions: (1) catalyst and substrate
were dissolved in the appropriate solvent under an atmosphere
of argon gas and subjected to 3 freeze-pump-thaw cycles prior
to backfilling with dideuterium gas to the desired levels; and
(2) the reactions were initially stirred under dynamic pressure
(15 min), followed by static pressure for the remainder of the
reaction. The relative proportions of isotopomers in the
product mixture were determined via NMR spectroscopy and
2
(R,R and S,S)-MACH2-â-d1 was identified by its proton signals
3
3
(δ 3.03 (d, J H
) 6.1 Hz, 1H, Hâ2), 4.85 (dd, J H
) 7.8
-H
R
-H
R NH
â2
Hz, 3J H
) 6.1 Hz, 1H, HR). The other isotopomers were
-H
R
â2
identified in the same manner. Identifications were confirmed
by 1H{2H, BB; 1H, sel} NMR experiments. The combined
enantiomeric excess and absolute configuration of the products
for each mixture of isotopomers was determined using the
aforementioned procedure. NMR spectroscopic data (CDCl3,
25 °C) for N-acetylphenylalanine methyl ester (MACH2): 1H
(400.1 MHz) δ 1.93 (s, 3H, NHCOCH3), 3.03 (dd, 2J H
)
-H
â1
â2
3
2
13.9 Hz, J H
) 6.1 Hz, 1H, Hâ ), 3.11 (dd, J H
) 13.9
-H
-H
â2
R
â2
2
â1
3
Hz, J H
) 6.1 Hz, 1H, Hâ ), 3.68 (s, 3H, CO2CH3), 4.85 (d
-H
R
â1
1
3
3
3
of app t, J H
) 7.8 Hz, J H
) J H
) 6.1 Hz, 1H,
-H
R
-H
R
-H
R
NH
â1
â2
3
HR), 6.27 (br d, NHCOCH3), 7.08 (d, J H-H ) 6.9 Hz, 2H,
o-C6H5), 7.24 (m, 3H, m- and p-C6H5); 13C{1H} (100.6 MHz) δ
22.8 (NHCOCH3), 37.6 (CH2CH, Câ), 52.1 (CO2CH3), 53.1
(CH2CH, CR), 126.9 (C6H5), 128.4 (C6H5), 129.0 (C6H5), 135.9
(ipso-C6H5), 169.7 (CO2CH3), 172.1 (NHCOCH3).
Gen er a l P r oced u r e for Hyd r ogen olysis/Deu ter iolysis
of (SCr)-1. Reactions were performed in a manner analogous
to that described above for catalytic hydrogenations/deutera-
tions with the exception that [(SCR)-1] ) 10 mM. Typical
reaction times were 4 (in methanol) and 24 h (in acetone), after
which all volatiles were removed under reduced pressure. The
remaining residue was refluxed in acetonitrile under argon
gas overnight (∼15 h), followed by evaporation under reduced
pressure. The solid residue was washed with ethyl acetate,
and the solution was passed through a column of Florisil to
remove residual ruthenium-BINAP species. Removal of the
solvent on a rotary evaporator yielded pure N-acetylphenyl-
alanine methyl ester, which was analyzed without further
purification. Enantiomeric excesses and the distributions of
deuterium were determined using the aforementioned proce-
dures.
Su bstr a te Equ ilibr iu m w ith (SCr)-1. In a 5-mm NMR
tube, (Z)-methyl-d3 R-acetamidocinnamate (5.8 mg, 2.6 × 10-5
mol) and (SCR)-1 (9.9 mg, 9.2 × 10-6 mol) were dissolved in a
mixture of CD2Cl2 (∼0.1 mL, used to ensure complete dissolu-
tion of (SC )-1) and acetone-d6 (∼0.5 mL). The sample was
R
immediately placed in a thermostated NMR probe (T ) 35 °C)
and monitored by 1H NMR spectroscopy. The extent of the
reaction was determined by the ratio of the methoxy signals
of (SCR)-1 and (Z)-MAC (singlets at 4.0 and 3.8 ppm, respec-
tively) using a benzylic proton of (SCR)-1 as the internal
standard (doublet at 4.2 ppm). 31P NMR spectroscopy indi-
cated the only ruthenium species present in the reaction
1
HRMS (EI). The ABX spin system in the H NMR spectrum
for the R- and â-protons of MACH2 were assigned by compari-
son to the reported assignments established for N-acetyl-
phenylalanine, the acid analogue of MACH2 (Figure 5).3h Each
isotopomer was quantified by integrating the signals of the
protons and the methoxy signal as an internal reference. The
amount of deuterium substitution at each position was ob-
tained from 2H{1H} NMR spectra or deduced from the proton
signals in the 1H{2H} NMR spectrum by subtraction if the
signals in the 2H{1H} NMR spectrum overlapped and could
not be deconvoluted. The presence of MACH2-dn (n ) 0-3)
were confirmed by the parent ion peaks in the mass spectrum.
Of the eight possible isotopomers of MACH2-dn (n ) 0-3), only
seven were observed during the course of this investigation
(MACH2, (R,R and S,S)-MACH2-â-d1, MACH2-R-d1, MACH2-
â,â-d2, (R,S and S,R)-MACH2-R,â-d2, (R,R and S,S)-MACH2-
mixture was (SC )-1 and (SC )-1-Me-d3.
R
R
Ack n ow led gm en t. This work was financially sup-
ported by the Natural Sciences and Engineering Re-
search Council of Canada, by the University of Alberta,
and by Firmenich SA.
OM980113F
(30) The deuterium-induced isotope effects on 1H nuclear shieldings
were typically 8 ppb (â-upfield shift) and 10 ppb (R-upfield shift) per
deuterium. See: Hansen, P. E. In Annual Reports on NMR Spectros-
copy; Webb, G. A., Ed.; Academic Press: London, 1983; Vol. 15, pp
105-234.
(29) Parker, D. Chem. Rev. 1991, 91, 1441-1457.