CO dissociation mechanism in racemization of alcohols by a cyclopentadienyl ruthenium dicarbonyl catalyst

Article abstract of DOI:10.1021/ja1098066

¹³CO exchange studies of racemization catalyst (η⁵-Ph₅C₅)Ru(CO)₂Cl and (η⁵-Ph₅C₅)Ru(CO)₂(Ot-Bu) by ¹³C NMR spectroscopy are reported. CO exchange

Full text of DOI:10.1021/ja1098066

COMMUNICATION
pubs.acs.org/JACS
CO Dissociation Mechanism in Racemization of Alcohols by
a Cyclopentadienyl Ruthenium Dicarbonyl Catalyst
Madeleine C. Warner, Oscar Verho, and Jan-E. B€ackvall
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, SE-106 91 Stockholm, Sweden
S Supporting Information
b
Scheme 1. Schematic DKR of 1-Phenylethanol (CALB =
ABSTRACT: 13CO exchange studies of racemization cat-
alyst (η5-Ph5C5)Ru(CO)2Cl and (η5-Ph5C5)Ru(CO)2(Ot-
Bu) by 13C NMR spectroscopy are reported. CO exchange
for the active catalyst form, (η5-Ph5C5)Ru(CO)2(Ot-Bu) is
approximately 20 times faster than that for the precatalyst
(η5-Ph5C5)Ru(CO)2Cl. An inhibition on the rate of race-
mization of (S)-1-phenylethanol was observed on addition
of CO. These results support the hypothesis that CO dis-
sociation is a key step in the racemization of sec-alcohols by
(η5-Ph5C5)Ru(CO)2Cl, as also predicted by DFT cal-
culations.
Candida antarctica Lipase B)3b
Scheme 2. Simplified Mechanism of Racemization of
sec-Alcohols by Ruthenium Catalyst 1
echanistic studies of hydrogen transfer reactions using
M
ruthenium catalysts have recently attracted considerable
attention,1 due to the extensive applications of these catalysts in
racemizations and (transfer) hydrogenations.1a,2 For example,
ruthenium catalyst 1 has successfully been employed as a
racemization catalyst in dynamic kinetic resolution (DKR) of
sec-alcohols. Ruthenium catalyst 1 is used in combination with a
lipase and an acyl donor to afford the corresponding enantio-
merically pure acetates in high yields and ee’s (Scheme 1).3
Ruthenium catalyst 1 is highly efficient in racemizing sec-
alcohols, and we have previously studied the racemization
mechanism of this catalyst in detail.4 Ruthenium complex 1 is
activated by t-BuOK to produce ruthenium tert-butoxide com-
plex 2. We recently showed that this ligand exchange takes place
via assistance of coordinated CO, with the formation of an acyl
intermediate A (Scheme 2, step i).5 Rapid alkoxide migration
from carbon to ruthenium converts intermediate A to complex 2
within a few minutes at room temperature. Subsequent alcohol-
alkoxide exchange produces a ruthenium sec-alkoxide complex 3
(Scheme 2, step ii). Racemization proceeds via an inner-sphere
mechanism which makes the participation of a free ruthenium
hydride species, (η5-Ph5C5)Ru(CO)2H, improbable.3b Since sec-
alkoxide complex 3 is an 18-electron complex, racemization via
β-hydride elimination requires creation of a free coordination
site on ruthenium, which can be formed by either η5 f η3 ring
slippage4 to give intermediate 4 or by loss of a CO ligand6
forming intermediate 5 (Scheme 2, step iii or iv). On the basis of
the observation1c that CO exchange does not occur in the related
Shvo hydride (η5-Ph4C4COH)Ru(CO)2H, CO exchange in 3
was considered to be unfavored, and therefore, the η5 f η3 ring
slip pathway was previously proposed.4,7 However, recent den-
sity functional theory (DFT) calculations showed that the
potential energy barrier for η5 f η3 ring slip in 3 was higher
than expected, 42 kcal/mol.6 This is ascribed to steric factors
between the phenyl groups, since the potential energy barrier
drops 11 kcal/mol when the phenyl groups are exchanged for
hydrogen atoms. The calculated potential energy barrier is lower
for the mechanism involving CO dissociation. The energy required
for creating an active 16-electron species by dissociation of a CO
ligand was calculated to be 22.6 kcal/mol.6 These results suggest
that loss of a CO ligand is more favored compared to the η5 f η3
ring slippage pathway. Recently we reported the coordination of
an olefin to ruthenium during the racemization which strongly
suggests the creation of a vacant site on ruthenium during
racemization.8 Also, Nolan and co-workers have reported the
use of well-defined 16-electron N-heterocyclic carbene- (NHC)
and phosphine-containing ruthenium(II) complexes as effective
Received: November 1, 2010
Published: February 11, 2011
r
2011 American Chemical Society
2820
dx.doi.org/10.1021/ja1098066 J. Am. Chem. Soc. 2011, 133, 2820–2823
|

Journal of the American Chemical Society
COMMUNICATION
Figure 2. 13CO incorporation (%) versus time (min) for ruthenium
tert-butoxide complex 2.
Figure 1. 13CO incorporation (%) versus time (h) for ruthenium
chloride 1.
Scheme 3. Proposed Mechanism for 13CO Exchange on
Complex 2 (steps i and ii) and for the Formation of Complex 7
(steps iii and iv)
racemization catalysts of aromatic and aliphatic sec-alcohols.9 We
now report results, involving CO exchange and CO inhibition,
which strongly support a CO dissociation pathway in the
racemization of secondary alcohols by the activated form of
ruthenium catalyst 1.
An interesting question to answer is whether CO dissociation
or exchange can be observed from ruthenium catalyst 1 or
ruthenium tert-butoxide complex 2. In preliminary CO exchange
experiments, 13CO-labeled ruthenium catalyst 1 (44% 13CO)
was treated with t-BuOK (1.2 equiv) in toluene-d8 in an NMR
tube. Analysis by 13C NMR after addition of 12CO (g) (0.3 equiv
and 0.9 equiv) to the preformed 13CO-labeled tert-butoxide
complex 2 showed no free 13CO at 184.7 ppm10 in either
reaction after 12 h. This could, however, be attributed to
problems with detection of small amounts of dissociated 13CO.
Instead, attempts of studying the incorporation of 13CO in
unlabeled ruthenium catalyst 1 and tert-butoxide complex 2 were
made. The rate of incorporation was studied by addition of 13CO
(g) (0.0124 mmol, 0.32 equiv) to ruthenium chloride 1 (0.039
mmol, 1 equiv) in dry toluene-d8 at room temperature in four
separate reactions. The reactions were analyzed by 13C NMR after 3,
6, 24, and 40 h, respectively. Under the specified reaction conditions
the maximum amount of 13CO incorporation that can theoretically
be achieved is 14% (see Supporting Information). After 24 h of
reaction time the 13CO incorporation was 8.6%. The 13CO content
was determined by comparison of the integrals to a reference
spectrum for unlabeled ruthenium catalyst 1 (1.1% 13C content).
The plotted results show that 13CO incorporation slows down
considerably after prolonged reaction times, as expected for this type
of equilibrium reaction (Figure 1).
A similar study was performed with ruthenium tert-butoxide
complex 2. 13CO (g) (0.0124 mmol, 0.32 equiv) was added to
preformed complex 2 (0.039 mmol, 1 equiv) dissolved in dry
toluene-d8 at room temperature in three separate reactions.
Analyses by 13C NMR were made after 30, 60, and 90 min,
respectively. The rate of 13CO incorporation for ruthenium tert-
butoxide complex 2 was found to be approximately 20 times
faster than that for ruthenium chloride 1. After only 30 min of
reaction time the 13CO incorporation was 5.0% (Figure 2). For
ruthenium chloride complex 1, it took approximately 10 h to
reach 5.0% 13CO incorporation. The 13CO content was deter-
mined by comparison of the integrals to a reference spectrum for
unlabeled ruthenium tert-butoxide complex 2 (1.1% 13C content).11
This dramatic increase in rate is attributed to the back-donation
from the t-BuO group, which stabilizes the intermediate 6 and most
likely also the transition state for the CO dissociation. Thus,
dissociation of CO is expected to be more facile in tert-butoxide
complex 2 compared to chloride complex 1.6,12
During the CO exchange studies of ruthenium tert-butoxide
complex 2, the peaks for complex 2 disappeared gradually, and
formation of a new complex was observed. The new complex was
observed only after ∼50 min when the concentration of carbon
monoxide was low (0.3 equiv of 13CO). At higher carbon monoxide
concentrations (>1 equiv of 13CO) significant amounts of this
new complex were observed in less than 20 min. The new
complex was therefore first thought to be a decomposition
product due to the sensitivity of ruthenium tert-butoxide com-
plex 2 toward moisture and air. At closer inspection of the 13C
NMR the new complex was assigned as carboalkoxydicarbonyl-
ruthenium complex 713 (Scheme 3, See Supporting Information
for characterization details).
Dissociation of CO from ruthenium tert-butoxide complex 2
produces intermediate 6 with a free coordination site at the metal
center (Scheme 3; step i). Intermediate 6 is the 16-electron
complex responsible for the racemization. Coordination of a
13CO molecule to ruthenium in 6 is now possible producing
13CO-enriched ruthenium tert-butoxide complex (Scheme 3;
step ii). In a competing pathway, alkoxide migration from
ruthenium to CO produces an alkoxycarbonyl complex B with
a free coordination site at ruthenium (Scheme 3; step iii).
Complex 7 is then formed irreversibly after coordination of
CO (Scheme 3; step iv). It is interesting to note that the latter
pathway does not lead to any CO exchange of 2 since step iv is
irreversible. Nolan has recently shown that a 16-electron cyclo-
pentadienylruthenium complex with an NHC ligand, analogous
to 6, is an active racemization catalyst for sec-alcohols.9b
The irreversible formation of complex 7 (Scheme 3; step iv) is
based on the observation that 7 cannot be converted back into
complex 2, even under reduced pressure. Furthermore, it was
2821
dx.doi.org/10.1021/ja1098066 |J. Am. Chem. Soc. 2011, 133, 2820–2823

Journal of the American Chemical Society
COMMUNICATION
of the racemization rate was observed, providing additional
support for reversible CO dissociation as a key step in the
racemization mechanism of sec-alcohols.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
characterization data for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
Figure 3. Effect of added CO on the racemization of (S)-1-phenyletha-
nol (120 μL, 0.99 mmol) by 1 (6.4 mg, 0.01 mmol, 1 mol %). 250, 300,
350, and 400 μL of CO corresponds to 0.0104, 0.0125, 0.0146, and
0.0166 mmol.
’ AUTHOR INFORMATION
Corresponding Author
jeb@organ.su.se
found that complex 7 is not active as a racemization catalyst for
sec-alcohols which is in accordance with the irreversibility.14 The
equilibrium between 2 and B (step iii) was confirmed by the
observed first-order dependence of [CO] in the formation of
complex 7 from complex 2 (see Supporting Information).
The effect of the carbon monoxide concentration on the rate
of racemization of sec-alcohols was also studied. t-BuOK (0.03 mmol,
3 mol %) was added to ruthenium chloride 1 (0.01 mmol, 1 mol
%) in dry toluene (2.0 mL). 12CO (250-400 μL, 0.011-
0.017 mmol) was added via syringe after formation of ruthenium
tert-butoxide complex 2. (S)-1-phenylethanol (0.99 mmol, 1 equiv,
>99% ee) was added to the reaction. Aliquots for GC analysis
were withdrawn after 1, 3, and 10 min. Each experiment was run
twice, and the mean values were plotted in a graph. A control
racemization experiment without any added CO was also per-
formed as a reference.
A catalyst loading of 1 mol % produces an efficient system,
capable of racemizing the substrate within 3 min at room tem-
perature (Figure 3; purple). The results show that addition of
250 μL of CO had a negligible effect on the rate of racemization
(Figure 3; dark blue). When 300 μL of CO was added, the rate of
racemization decreased to less than half. This shows some
inhibition of the racemization process by CO (Figure 3; pink).
Addition of 350 μL of CO led to a substantial decrease in rate.
(Figure 3; green). A further decrease in racemization rate was
observed when 400 μL of CO was added. The sample still
exhibited 80% ee after 10 min of reaction time (Figure 3; light
blue). At higher concentrations of added CO the racemization of
(S)-1-phenylethanol was completely inhibited, and no racemiza-
tion could be detected within 10 min. This suggests that
formation of 7 or corresponding 70 is fast at high concentrations
of CO.
’ ACKNOWLEDGMENT
The Swedish Research Council, the European Research
Council (ERC AdG 247014), the Berzelius Center EXSELENT,
and the Knut and Alice Wallenberg foundation are gratefully
acknowledged for financial support.
’ REFERENCES
(1) (a) Samec, J. S. M.; B€ackvall, J.-E.; Andersson, P. G.; Brandt, P.
Chem. Soc. Rev. 2006, 35, 237. (b) Conley, B. L.; Pennington-Boggio,
M. K.; Boz, E.; Williams, T. J. Chem. Rev. 2010, 110, 2294–2312. (c) Casey,
C. P.; Singer, S. W.; Powell, D. R.; Hayashi, R. K.; Kavana, M. J. Am. Chem.
Soc. 2001, 123, 1090. (d) Casey, C. P.; Johnson, J. B. J. Org. Chem. 2003, 68,
1998. (e) Casey, C. P.; Clark, T. B.; Guzei, I. A. J. Am. Chem. Soc. 2007, 129,
11821. (f) Guan, H.; Iimura, M.; Magee, M. P.; Norton, J. R.; Zhu, G. J. Am.
Chem. Soc. 2005, 127, 7805. (g) Åberg, J. B.; B€ackvall, J.-E. Chem.—Eur. J.
2008, 14, 9169. (h) Hamilton, R. J.; Bergens, S. H. J. Am. Chem. Soc. 2008,
130, 11979.
(2) (a) Gladiali, S.; Alberico, E. Chem. Soc. Rev. 2006, 35, 226.
(b) Ikariya, T.; Blacker, A. J. Acc. Chem. Res. 2007, 40, 1300. (c) Abbel,
R.; Abdur-Rashid, K.; Faatz, M.; Hadzovic, A.; Lough, A. J.; Morris, R. H.
J. Am. Chem. Soc. 2005, 127, 1870–1882. (d) Zimmer-De Iuliis, M.;
Morris, R. H. J. Am. Chem. Soc. 2009, 131, 11263–11269. (e) Takebaya-
shi, S.; John, J. M.; Bergens, S. H. J. Am. Chem. Soc. 2010, 132, 12832–
12834. (f) Baratta, W.; Chelucci, G.; Magnolia, S.; Siega, K.; Rigo, P.
Chem.—Eur. J. 2009, 15, 726. (g) Samec, J. S. M.; B€ackvall, J.-E.
Encyclopedia of Reagents for Organic Synthesis, 2nd ed.; Fuchs, P. L.,
Ed.; Wiley Interscience: New York, 2009; Vol. 7, pp 5557-5564.
(3) (a) Martín-Matute, B.; Edin, M.; Bogꢀar, K.; B€ackvall, J.-E. Angew.
Chem., Int. Ed. 2004, 43, 6535. (b) Martín-Matute, B.; Edin, M.; Bogꢀar,
K.; Kaynak, F. B.; B€ackvall, J.-E. J. Am. Chem. Soc. 2005, 127, 8817.
(c) Tr€aff, A.; Bogꢀar, K.; Warner, M.; B€ackvall, J.-E. Org. Lett. 2008, 10,
4807.
(4) Martín-Matute, B.; Åberg, J. B.; Edin, M.; B€ackvall, J.-E. Chem.—
Eur. J. 2007, 13, 6063.
(5) Åberg, J. B.; Nyhlꢀen, J.; Martín-Matute, B.; Privalov, T.; B€ackvall,
J.-E. J. Am. Chem. Soc. 2009, 131, 9500.
The inhibition by carbon monoxide on the racemization of sec-
alcohols strengthens the hypothesis that reversible CO dissocia-
tion is a key step in the racemization mechanism.
In conclusion we have found that CO exchange, monitored by
13C NMR and the use of 13CO, occurs with ruthenium chloride 1
and ruthenium tert-butoxide complex 2 in the racemization
mechanism of sec-alcohols. The CO exchange is approximately
20 times faster for ruthenium tert-butoxide complex 2 compared
to that for ruthenium chloride complex 1. During the CO ex-
change experiments on ruthenium tert-butoxide complex 2, for-
mation of carboalkoxydicarbonylruthenium complex 7 was ob-
served. The effect of different concentrations of CO on the rate of
racemization of (S)-1-phenylethanol was also investigated, and it
was found that small amounts of added CO have a negligible
effect on the racemization rate. At higher concentrations, inhibition
(6) Nyhlꢀen, J.; Privalov, T.; B€ackvall, J.-E. Chem.—Eur. J. 2009, 15,
5220.
(7) On the basis of a 13CO study Casey has recently considered CO
dissociation in a Shvo hydride derivative (with two tolyls and two
phenyls on the Cp ring) in a chain mechanism for exchange of D2 with
the hydride; however, in the dark the CO exchange was very slow: Casey,
C. P.; Johnson, J. B.; Jiao, X.; Beetner, S. E.; Singer, S. W. Chem. Commun.
2010, 46, 7915–7917.
(8) Åberg, J. B.; Warner, M. C.; B€ackvall, J.-E. J. Am. Chem. Soc. 2009,
131, 13622.
(9) (a) Bosson, J.; Nolan, S. P. J. Org. Chem. 2010, 75, 2039. (b) Bosson,
J.; Poater, A.; Cavallo, L.; Nolan, S. P. J. Am. Chem. Soc. 2010, 132, 13146–
13149.
2822
dx.doi.org/10.1021/ja1098066 |J. Am. Chem. Soc. 2011, 133, 2820–2823

Journal of the American Chemical Society
COMMUNICATION
(10) Selg, P.; Brintzinger, H. H.; Andersen, R. A.; Horvꢀath, I. T.
Angew. Chem., Int. Ed. 1995, 34, 791.
(11) Although the 13CO exchange with chloride 1 within experi-
mental error fits with an equilibrium of a theoretical maximum of 14%
(Figure 1), the corresponding exchange with tert-butoxide complex 2
does not. This is because some 13CO is irreversibly tied up in complex 7.
Also, there is some decomposition of complex 2 due to traces of air and
humidity. In addition, the theoretical maximum of 14% was based on the
assumption that the commercial 13CO is 99% 13CO. The purity seems to
be lower.
(12) (a) Poulton, J. T.; Sigalas, M. P.; Folting, K.; Streib, W. E.;
Eisenstein, O.; Caulton, K. G. Inorg. Chem. 1994, 33, 1476–1485.
(b) Lunder, D. M.; Lobkovsky, E. B.; Streib, W. E.; Caulton, K. G.
J. Am. Chem. Soc. 1991, 113, 1837–1838.
(13) (a) For the related (η5-Me5C5)Ru(CO)2(COOt-Bu) complex
see: Suzuki, H.; Omori, H.; Moro-oka, Y. J. Organomet. Chem. 1987, 327,
C47. (b) For a related formyl complex see: Summer, C. E.; Nelson, G. O.
J. Am. Chem. Soc. 1984, 106, 432–433.
(14) It was also found that complex 7 (12CO) does not undergo
exchange with 13CO when left for 24 h.
2823
dx.doi.org/10.1021/ja1098066 |J. Am. Chem. Soc. 2011, 133, 2820–2823