Nucleophilic catalysis with π-bound nitrogen heterocycles: Synthesis of the first ruthenium catalysts and comparison of the reactivity and the enantioselectivity of ruthenium and iron complexes

Article abstract of DOI:10.1021/ja981061o

Three ruthenium complexes that bear π-bound nitrogen heterocycles have been synthesized. It is established that these complexes serve as effective nucleophilic catalysts for a range of processes, including the acylation of alcohols with diketene, the ring opening of azlactones, and the addition of alcohols to ketenes; their activity is comparable to or somewhat greater than the corresponding iron catalysts. The relative efficiency of die ruthenium complexes as asymmetric catalysts is also evaluated in the kinetic resolution of secondary alcohols, ruthenium is markedly less effective than iron, but in the deracemization/ring opening of azlactones, ruthenium is slightly more enantioselective. This study documents or the first time the impact of the metal on the reactivity and on the enantioselectivity of nucleophilic catalysts based on π-bound nitrogen heterocycles.

Full text of DOI:10.1021/ja981061o

J. Am. Chem. Soc. 1998, 120, 7479-7483
7479
Nucleophilic Catalysis with π-Bound Nitrogen Heterocycles:
Synthesis of the First Ruthenium Catalysts and Comparison of the
Reactivity and the Enantioselectivity of Ruthenium and Iron
Complexes
Christine E. Garrett and Gregory C. Fu*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed March 30, 1998
Abstract: Three ruthenium complexes that bear π-bound nitrogen heterocycles have been synthesized. It is
established that these complexes serve as effective nucleophilic catalysts for a range of processes, including
the acylation of alcohols with diketene, the ring opening of azlactones, and the addition of alcohols to ketenes;
their activity is comparable to or somewhat greater than the corresponding iron catalysts. The relative efficiency
of the ruthenium complexes as asymmetric catalysts is also evaluated: in the kinetic resolution of secondary
alcohols, ruthenium is markedly less effective than iron, but in the deracemization/ring opening of azlactones,
ruthenium is slightly more enantioselective. This study documents for the first time the impact of the metal
on the reactivity and on the enantioselectivity of nucleophilic catalysts based on π-bound nitrogen heterocycles.
Introduction
ary alcohols (Fe-C5Ph5-DMAP),2b,c providing the best enanti-
oselectivity reported to date for either process with a nonenzymatic
catalyst.
In view of the wide array of reactions that are subject to
catalysis by nucleophiles, there have been surprisingly few
reports of effective asymmetric variants of these processes.1
Consequently, we have recently initiated a program directed at
the design and development of enantioselective nucleophilic
catalysts.2,3 We have focused our efforts on chiral derivatives
of planar nitrogen heterocycles (e.g., DMAP4 and imidazole),
in part because these catalysts exhibit both high activity and
broad applicability.
In our initial studies, we established that nitrogen heterocycles
that are π-complexed to iron comprise a versatile new family
of nucleophilic catalysts.2a In subsequent work, we demon-
strated that enantiopure planar-chiral π-bound heterocycles
catalyze both the deracemization/ring opening of azlactones (Fe-
DMAP*)3 and the asymmetric acylation of unsaturated second-
An appealing design feature of these planar-chiral nucleo-
philic catalysts is the potential to tune their reactivity and their
enantioselectivity through appropriate choice of metal fragment
(MCpx above, where Cpx ) a cyclopentadienyl-derived ligand).
In earlier work, we examined the effect of a change in Cpx,
and we determined that in the case of the asymmetric acylation
of secondary alcohols, the proper choice of Cpx is critical to
achieving high selectivity (eq 1; selectivity factor ) (rate of
fast-reacting enantiomer)/(rate of slow-reacting enantiomer)).5
(1) For example, see: (a) Acylation of alcohols: Vedejs, E.; Daugulis,
O.; Diver, S. T. J. Org. Chem. 1996, 61, 430-431. Kawabata, T.; Nagato,
M.; Takasu, K.; Fuji, K. J. Am. Chem. Soc. 1997, 119, 3169-3170. (b)
Aldol: Denmark, S. E.; Winter, S. B. D.; Su, X.; Wong, K.-T. J. Am. Chem.
Soc. 1996, 118, 7404-7405. (c) Allylation of aldehydes: Iseki, K.; Kuroki,
Y.; Takahashi, M.; Kobayashi, Y. Tetrahedron Lett. 1996, 37, 5149-5150.
Denmark, S. E.; Coe, D. M.; Pratt, N. E.; Griedel, B. D. J. Org. Chem.
1994, 59, 6161-6163. (d) Reduction of ketones: Schiffers, R.; Kagan, H.
B. Synlett 1997, 1175-1178. (e) [2+2] Cycloaddition: Wynberg, H.;
Staring, E. G. J. J. Am. Chem. Soc. 1982, 104, 166-168. Wynberg, H.
Top. Stereochem. 1986, 16, 87-129. Calter, M. A. J. Org. Chem. 1996,
61, 8006-8007. (f) [3+2] Cycloaddition: Zhu, G.; Chen, Z.; Jiang, Q.;
Xiao, D.; Cao, P.; Zhang, X. J. Am. Chem. Soc. 1997, 119, 3836-3837.
(2) Kinetic resolution of secondary alcohols: (a) Ruble, J. C.; Fu, G. C.
J. Org. Chem. 1996, 61, 7230-7231. (b) Ruble, J. C.; Latham, H. A.; Fu,
G. C. J. Am. Chem. Soc. 1997, 119, 1492-1493. (c) Ruble, J. C.; Tweddell,
J.; Fu, G. C. J. Org. Chem. 1998, 63, 2794-2795.
(1)
(3) Dynamic kinetic resolution of azlactones: Liang, J.; Ruble, J. C.;
Fu, G. C. J. Org. Chem. 1998, 63, 3154-3155.
(4) (a) Scriven, E. F. V. Chem. Soc. ReV. 1983, 12, 129-161. (b) Hassner,
A.; Krepski, L. R.; Alexanian, V. Tetrahedron 1978, 34, 2069-2076. (c)
Ho¨fle, G.; Steglich, W.; Vorbru¨ggen, H. Angew. Chem., Int. Ed. Engl. 1978,
17, 569-583.
We have now begun to explore the effect that a change in M
has on the reactivity and the enantioselectivity of these π-bound
(5) Reference 2c and unpublished results (J. C. Ruble).
S0002-7863(98)01061-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/18/1998

7480 J. Am. Chem. Soc., Vol. 120, No. 30, 1998
Garrett and Fu
heterocycles. In our initial studies, we have chosen to evaluate
the impact of substituting Fe with Ru.6 A qualitative analysis
based on steric effects suggests that, due to longer metal-ligand
bond distances, the Ru complexes might exhibit lower enanti-
oselectivity (“less chiral”) and greater reactivity (less sterically
hindered) than the corresponding Fe complexes. Consideration
of electronic effects also points to the possibility of enhanced
reactivity in the case of ruthenium catalystssfor example, it is
established that a ruthenocenyl group better stabilizes an adjacent
cation than does a ferrocenyl group (A),7 which suggests that
the nitrogen of pyrindinylruthenium complexes might be more
nucleophilic than that of pyrindinyliron complexes (cf. A and
B). In this report, we synthesize the first Ru-based nucleophilic
catalysts, and we compare their reactivity and their enantiose-
lectivity with the corresponding Fe catalysts.
(3)
The synthesis of the η5-C5Ph5 analogue, Ru-η5-C5Ph5-DMAP,
obviously requires a different approach. The only method that
has been described for the generation of mixed-sandwich
ruthenocenes that bear an η5-C5Ph5 group involves treatment
of (η5-C5Ph5)Ru(CO)2X with a Cpx anion.10 We have found
that this strategy can be extended to the synthesis of Ru-C5-
Ph5-DMAP (17%; eq 4).11
(4)
Worth noting from a practical standpoint is that these ruthenium
complexes (Ru-Pyrrole*, Ru-DMAP*, and Ru-C5Ph5-DMAP)
are reasonably stable to air and to moisture, even in solution
(e.g., they may be purified by flash chromatography).
Results and Discussion
Synthesis of Catalysts. To evaluate the consequences of
substitution of Fe with Ru, we chose to investigate the ruthenium
analogues of three of the azaferrocene and pyrindinyliron
complexes that we had explored earlier, specifically, Ru-
Pyrrole*, Ru-DMAP*, and Ru-C5Ph5-DMAP. None of these
complexes had previously been synthesized.
In 1988, Gassman reported that an array of mixed-sandwich
ruthenocenes Cp*RuCpx can be generated through treatment of
[Cp*RuCl2]x with Cpx anion.8 Kelly subsequently applied this
strategy to the first (and only) synthesis of an azaruthenocene,
Cp*Ru(η5-C4Me4N).9 We have established that azaruthenocene
Ru-Pyrrole* can also be produced through this procedure. Thus,
reaction of commercially available [Cp*RuCl2]x with lithium
pyrrolide in THF at room temperature affords Ru-Pyrrole* in
good yield after purification by flash chromatography (51%;
eq 2).
Resolution of Ru-DMAP* and Ru-C5Ph5-DMAP; Struc-
tural Characterization and Determination of the Absolute
Configuration of (+)-Ru-DMAP*. As is the case for their
iron analogues, the enantiomers of Ru-DMAP* and Ru-C5Ph5-
DMAP can be separated through chiral HPLC (semipreparative
Daicel Chiralcel columns). In the case of Ru-DMAP*, the (+)
enantiomer elutes first; in the case of Ru-C5Ph5-DMAP, the (-)
enantiomer elutes first. This elution behavior is identical to
that observed for the corresponding iron complexes.2b
We were able to obtain X-ray quality crystals of enantiopure
(+)-Ru-DMAP*, thereby confirming our structural assignment
and allowing determination of absolute configuration (Figure
1). Interestingly, the (+) enantiomers of Ru-DMAP* and Fe-
DMAP* have the same stereochemistry.2b Structurally, the
primary difference between Ru-DMAP* and Fe-DMAP*12 is
the distance between the cyclopentadienyl planes (3.64 and 3.32
Å, respectively).13
Relative Reactivity of Fe and Ru Complexes. Our initial
reactivity studies focused on Ru-Pyrrole*. We had established
earlier that Fe-Pyrrole* serves as a nucleophilic catalyst for an
array of processes, including the addition of benzyl alcohol to
phenylethylketene and the acylation of 1-phenylethanol with
diketene.2a,14 We have determined that Ru-Pyrrole* also
catalyzes each of these reactions. In the case of the acylation
of 1-phenylethanol with diketene, Ru-Pyrrole* is somewhat less
(2)
Pyrindinylruthenium complex Ru-DMAP* can be synthesized
analogously, albeit in more modest yield (24%; eq 3).
(6) For studies comparing ferrocene- and ruthenocene-derived ligands,
see: (a) Hayashi, T.; Ohno, A.; Lu, S.-j.; Matsumoto, Y.; Fukuyo, E.;
Yanagi, K. J. Am. Chem. Soc. 1994, 116, 4221-4226. (b) Abbenhuis, H.
C. L.; Burckhardt, U.; Gramlich, V.; Martelletti, A.; Spencer, J.; Steiner,
I.; Togni, A. Organometallics 1996, 15, 1614-1621. (c) Li, S.; Wei, B.;
Low, P. M. N.; Lee, H. K.; Hor, T. S. A.; Xue, F.; Mak, T. C. W. J. Chem.
Soc., Dalton Trans. 1997, 1289-1293.
(10) Slocum, D. W.; Duraj, S.; Matusz, M.; Cmarik, J. L.; Simpson, K.
M.; Owen, D. A. In Metal-Containing Polymeric Systems; Plenum: New
York, 1985.
(11) We have confirmed the structural assignment for Ru-C5Ph5-DMAP
through a low-resolution X-ray crystal structure (M. M.-C. Lo).
(12) Ruble, J. C., Hoic, D. A. Unpublished results.
(13) The corresponding distances for ruthenocene and ferrocene are 3.68
and 3.32 Å, respectively (Hardgrove, G. L.; Templeton, D. H. Acta.
Crystallogr. 1959, 12, 28-32; Dunitz, J. D.; Orgel, L. E.; Rich, A. Acta.
Crystallogr. 1956, 9, 373-375).
(14) For other catalysts, see: (a) Addition of alcohols to ketenes: Tidwell,
T. T. Ketenes; Wiley: New York, 1995. (b) Acylation of alcohols with
diketene: Wilson, S. R.; Price, M. F. J. Org. Chem. 1984, 49, 722-725.
(7) (a) Turbitt, T. D.; Watts, W. E. J. Chem. Soc., Perkin Trans. 2 1974,
185-189. (b) Hill, E. A.; Richards, J. H. J. Am. Chem. Soc. 1961, 83, 3840-
3846.
(8) Gassman, P. G.; Winter, C. H. J. Am. Chem. Soc. 1988, 110, 6130-
6135.
(9) Kelly, W. J.; Parthun, W. E. Organometallics 1992, 11, 4348-4350.

Catalysis with π-Bound N Heterocycles
J. Am. Chem. Soc., Vol. 120, No. 30, 1998 7481
(8)
In conclusion, a study of a cross-section of reactions reveals
that the choice of metal has a measurable impact on the activity
of nucleophilic catalysts based on π-bound heterocycles, with
ruthenium complexes generally displaying greater reactivity than
the corresponding iron complexes.
Figure 1. ORTEP illustration, with thermal ellipsoids drawn at the
35% probability level, of (+)-Ru-DMAP*.
effective than Fe-Pyrrole* (eq 5).
Relative Enantioselectivity of Fe and Ru Complexes.
Planar-chiral Fe-C5Ph5-DMAP is the most enantioselective
nonenzymatic catalyst reported to date for the acylation (kinetic
resolution) of aryl-alkyl carbinols.1a,2,17 We had demonstrated
earlier that the stereoselectivity of this process is sensitive to
the choice of Cpx (eq 1), and comparison of the selectivity factor
for the reaction catalyzed by Fe-C5Ph5-DMAP with that for Ru-
C5Ph5-DMAP establishes that it is also sensitive to the choice
of M (eq 9).18,19
(5)
On the other hand, Ru-Pyrrole* is significantly more effective
than its Fe analogue in catalyzing the addition of benzyl alcohol
to phenylethylketene. In side-by-side reactions, under otherwise
identical conditions, the ester forms more than twice as quickly
in the presence of the Ru complex (eq 6).
(9)
(6)
Correspondingly, Fe-DMAP* is the most enantioselective
nonenzymatic catalyst reported to date for the deracemization/
ring opening (dynamic kinetic resolution) of azlactones.3,20
Under otherwise identical conditions, Ru-DMAP* displays
slightly enhanced stereoselectivity, thereby establishing a new
benchmark for this process (eq 10).21
Turning to the chiral DMAP derivatives, we chose to explore
the reactivity of Ru-C5Ph5-DMAP and Ru-DMAP* in the two
reactions for which their Fe analogues are particularly effective
asymmetric catalystssthe acylation of secondary alcohols2 and
the ring opening of azlactones by alcohols,3 respectively. In
the case of the ring opening of azlactones, we have determined
that Ru-DMAP* is a somewhat more active catalyst than is Fe-
DMAP* (eq 7).15
(10)
Summary and Conclusions
(7)
The study of nucleophilic catalysis by π-complexed hetero-
cycles is still in its infancy. While one investigation has
described a dramatic difference in enantioselectivity due to a
change in the structure of a remote cyclopentadienyl group
(15) The half-life for the corresponding reaction in the presence of
4-dimethylaminoquinoline (5%) is 70 h.
(16) The half-life for the corresponding reaction in the presence of
4-dimethylaminoquinoline (1%) is 80 h.
(17) Oriyama, T.; Hori, Y.; Imai, K.; Sasaki, R. Tetrahedron Lett. 1996,
37, 8543-8546.
For the acylation of secondary alcohols, Ru-C5Ph5-DMAP
is more than twice as effective as Fe-C5Ph5-DMAP (eq 8). Thus,
in the presence of 1% of the Ru complex, formation of the ester
requires 1.3 h to proceed to 50% completion, whereas with the
Fe complex, 3.5 h are necessary.16
(18) The (-) enantiomers of Fe-C5Ph5-DMAP and Ru-C5Ph5-DMAP
preferentially acylate the same enantiomer of 1-phenylethanol (R), and based
on this and other data, we have tentatiVely assigned the absolute configu-
ration of (-)-Ru-C5Ph5-DMAP. Unfortunately, we have not yet been able
to obtain an X-ray crystal structure of enantiopure Ru-C5Ph5-DMAP that
is of sufficient quality to permit a definitive assignment.

7482 J. Am. Chem. Soc., Vol. 120, No. 30, 1998
Garrett and Fu
Benzoic acid was recrystallized prior to use. Phenylethylketene22 and
2-phenyl-4-methyloxazalone23 were prepared according to literature
methods.
All reactions were carried out under an atmosphere of nitrogen or
argon in oven-dried glassware with magnetic stirring.
(Cpx), no report has yet explored the effect of a change in the
transition metal (M) on either the reactivity or the enantiose-
lectivity of this new family of catalysts. We therefore undertook
a study of the Ru analogues of three Fe complexes known to
be active (and in two cases enantioselective) nucleophilic
catalysts. Based on steric and electronic considerations, we
anticipated that the Ru systems might exhibit higher reactivity
and lower enantioselectivity than the corresponding Fe systems.
In this paper, we have described the synthesis and resolution
of the target Ru complexes, which are the first non-Fe-based
π-bound heterocycles to be explored as nucleophilic catalysts.
With respect to reactivity, the Ru analogues serve as effective
catalysts for an array of processes, providing acceleration
comparable to or somewhat greater than the corresponding Fe
complexes.
With respect to enantioselectivity, we have evaluated the Ru
catalysts in the two reactions for which the planar-chiral Fe
compounds define the nonenzymatic state of the art. In the
case of the kinetic resolution of aryl-alkyl carbinols, a process
known to be sensitive to the choice of Cpx, we have found that
the selectivity is also sensitive to the choice of M: Ru-C5Ph5-
DMAP is significantly less enantioselective than is Fe-C5Ph5-
DMAP. In contrast, in the case of the deracemization/ring-
opening of azlactones, Ru-DMAP* is superior to Fe-DMAP*,
providing a modestly improved benchmark for this reaction.
We have thus documented for the first time the impact that
the choice of metal has on the reactivity and on the enantiose-
lectivity of nucleophilic catalysts based on π-bound heterocycles.
Other approaches to tuning these systems are under investiga-
tion, and these studies will be reported in due course.
Ru-Pyrrole* (eq 2).8,9 n-BuLi (2.56 mL, 4.10 mmol) was added
to a solution of pyrrole (274 µL, 4.08 mmol) in THF (10 mL), providing
a light-yellow solution. After 1 h of stirring at room temperature,
[Cp*RuCl2]x (504 mg, 1.64 mmol) was added, resulting in a red-brown
solution. After stirring for 3 h at room temperature, the solution had
turned bluish. After stirring for a total of 24 h, the reaction mixture
was passed through a short plug of alumina, and a yellow band was
collected (EtOAc as eluent), which provided a brown crystalline solid
after evaporation of the solvent. Flash chromatography (silica; hexane
f EtOAc) furnished a golden-yellow solid (255 mg, 51% yield;
unoptimized).
1H NMR (500 MHz, C6D6) δ 1.84 (s, 15H), 4.39 (s, 2H), 5.45 (s,
2H); 13C NMR (125 MHz, CDCl3) δ 12.5, 76.5, 85.5, 94.9; IR (neat)
2969, 2901, 2854, 1472, 1378, 1348, 1269, 1191, 1105, 1067, 1034,
1004, 855, 844, 804, 740, 703, 637, 457 cm-1; HRMS m/z 303.0561
[M+], calcd for C14H19NRu: 303.0561. Anal. Calcd for C14H19NRu:
C, 55.61; H, 6.33; N, 4.63. Found: C, 55.89; H, 6.42; N, 4.56; mp
(under N2): 150-152 °C; TLC (PMA positive) Rf ) 0.60 (EtOAc).
Ru-DMAP* (eq 3).8,9 n-BuLi (0.50 mL, 0.80 mmol) was added
dropwise to a flask containing 4-dimethylaminopyrindine2a (120 mg,
0.746 mmol) in THF (5 mL). The resulting reddish solution was stirred
at room temperature for 1 h, and then [Cp*RuCl2]x (14.4 mg, 0.340
mmol) was added, providing a dark-brown solution. After stirring for
18 h at room temperature, the reaction mixture was filtered through
silica (10% NEt3/EtOAc as eluent), and a yellow-brown solution was
collected and then concentrated. The resulting brown-green solid was
chromatographed several times (hexane f EtOAc f 10% NEt3/
EtOAc), affording a green-yellow solid (32.2 mg, 24% yield; unopti-
mized).
1H NMR (500 MHz, C6D6) δ 1.66 (s, 15H), 2.60 (s, 6H), 4.34 (t, J
) 2.5, 1H), 4.59 (dd, J ) 1.3, 2.8 Hz, 1H), 5.28 (dd, J ) 1.3, 2.8 Hz,
1H), 5.43 (d, J ) 5.0, 1H), 8.42 (d, J ) 5.5, 1H); 13C NMR (125 MHz,
CDCl3) δ 10.7, 41.4, 67.1, 69.9, 75.8, 77.3, 83.2, 93.7, 113.8, 151.0,
157.3; IR (neat) 2901, 1559, 1538, 1442, 1380, 1350, 1334, 1033, 1020,
903, 815, 787 cm-1; HRMS m/z 396.1142 [M+] calcd for C20H26N2-
Ru: 396.1140. Anal. Calcd for C20H26N2Ru: C, 60.74; H, 6.63; N,
7.08. Found: C, 60.94; H, 6.90; N, 6.89; mp (under N2): 140-142
°C; TLC Rf ) 0.55 (10% NEt3/EtOAc).
The enantiomers of Ru-DMAP* were separated through semi-
preparative chiral HPLC (Daicel Chiralcel OD, 1 cm × 25 cm;
2-propanol/hexane/diethylamine 22/78/0.2; 3.0 mL/min). One enan-
tiomer was collected from 8.25 to 11.00 min, and the other enantiomer
was collected from 15.25 to 20.00 min.
Experimental Section
General. 1H and 13C nuclear magnetic resonance spectra were
recorded on a Varian XL-300 or a VXR-500 NMR spectrometer at
ambient temperature. 1H data are reported as follows: chemical shift
in parts per million downfield from tetramethylsilane (δ scale),
multiplicity (br ) broad, s ) singlet, d ) doublet, t ) triplet, q )
quartet, and m ) multiplet), coupling constant (Hz), and integration.
13C chemical shifts are reported in ppm downfield from tetramethyl-
silane (δ scale). All 13C spectra were determined with complete proton
decoupling.
Infrared spectra were obtained on a Perkin-Elmer Series 1600 FT-
IR spectrophotometer. High-resolution mass spectra were recorded on
a Finnegan MAT System 8200 spectrometer. Microanalyses were
performed by E + R Microanalytical Laboratory, Inc. Gas chroma-
tography analyses were accomplished on a Hewlett-Packard model 5890
Series 2 Plus gas chromatograph equipped with a flame ionization
detector and a model 3392A integrator.
A crystal suitable for X-ray analysis was grown of the fast-eluting
enantiomer (evaporation of an Et2O/pentane solution at 4 °C). [R]20
) +969.5° (c ) 0.13, CHCl3).
D
Ru-C5Ph5-DMAP (eq 4). n-BuLi (0.250 mL, 0.400 mmol) was
added dropwise to a solution of 4-dimethylaminopyrindine2a (64.1 mg,
0.400 mmol) in toluene (2 mL), resulting in a cloudy, tan reaction
mixture. After stirring for 1 h at room temperature, a purple solution
of (C5Ph5)Ru(CO2)Br24 (275.4 mg, 0.404 mmol) in toluene (3 mL) was
added, providing a brown reaction mixture. The solution was
transferred to a two-neck flask fitted with a reflux condenser, and it
was then refluxed for 22 h. After cooling to room temperature, the
reaction mixture was concentrated, and the resulting brown residue was
extracted (Et2O/H2O). The organic layer was passed through alumina
and then concentrated. Flash chromatography (silica; 50% EtOAc/
hexane f 10% NEt3/EtOAc) provided a yellow solid (46.8 mg, 17%
yield; unoptimized).
Analytical thin-layer chromatography was performed using EM
Reagents 0.25 mm silica gel 60 plates. Flash chromatography was
performed on EM Reagents silica gel 60 (230-400 mesh).
Solvents were distilled from the indicated drying agents: benzene
(Na/benzophenone); pentane (Na/benzophenone); hexane (Na/ben-
zophenone); THF (Na/benzophenone); Et2O (Na/benzophenone); tolu-
ene (Na).
Pyrrole was distilled from CaH2 and stored at -34 °C under nitrogen.
Benzyl alcohol, 1-phenylethanol, diketene, MeOH, NEt3 (from CaH2),
Ac2O (from quinoline), and tert-amyl alcohol were distilled prior to
use. C6D6 and toluene-d8 were dried over alumina before use. n-BuLi
(1.6 M in hexane) and [Cp*RuCl2]x (Strem) were used as received.
(19) (-)-Ru-C5Ph5-DMAP can be recovered in essentially quantitative
yield at the end of the reaction.
(20) Belokin, Y. N.; Bachurina, I. B.; Tararov, V. I.; Saporovskaya, M.
B. Bull. Acad. Sci. USSR, DiV. Chem. Sci. 1992, 41, 422-429.
(21) (-)-Ru-DMAP* can be recovered in essentially quantitative yield
at the end of the reaction.
(22) Baigrie, L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc.
1985, 107, 5391-5396.
(23) (a) Chen, F. M. F.; Kuroda, K.; Bentoiton, N. L. Synthesis 1979,
230. (b) Mohr, F.; Stroschein, F. Chem. Ber. 1909, 42, 2521.
(24) Slocum, D. W.; Matusz, M.; Clearfield, A.; Peascoe, R.; Duraj, S.
A. J. Macromol. Sci., Chem. 1990, A27, 1405-1414.

Catalysis with π-Bound N Heterocycles
J. Am. Chem. Soc., Vol. 120, No. 30, 1998 7483
1H NMR (500 MHz, CDCl3) δ 3.02 (s, 6H), 4.73 (t, J ) 2.8, 1H),
5.27 (dd, J ) 1.0, 1.5 Hz, 1H), 5.40 (d, J ) 2.0, 1H), 5.74 (d, J ) 5.0,
1H), 6.85 (d, J ) 7.5, 10H), 7.00 (t, J ) 8.0, 10H), 7.07 (m, 5H), 8.11
(d, J ) 5.0 Hz, 1H); 13C NMR (125 MHz, CDCl3) δ 41.7, 69.5, 72.9,
79.1, 81.5, 91.8, 97.3, 116.8, 126.2, 127.0, 132.3, 134.9, 153.0, 156.9;
IR (neat) 3055, 2925, 1600, 1564, 1540, 1502, 1443, 1397, 1349, 1028,
784, 740, 699, 572, 556 cm-1; HRMS m/z 706.1921 [M+], calcd for
C45H36N2Ru: 706.1922. Anal. Calcd for C45H36N2Ru: C, 76.57; H,
5.14; N, 3.97. Found: C, 76.36; H, 4.99; N, 4.13; mp: >250 °C;
TLC Rf ) 0.60 (10% NEt3/EtOAc).
Acylation of 1-Phenylethanol with Ac2O (eq 8). A stock solution
was prepared of (S)-1-phenylethanol (109 mg, 0.903 mmol) and NEt3
(141 µL, 1.01 mmol) in tert-amyl alcohol (1.5 mL). (+)-Fe-C5Ph5-
DMAP and (+)-Ru-C5Ph5-DMAP (0.0028 mmol) were weighed into
each of two vials, and stock solution (580 µL) was added to each vial.
The solutions were warmed slightly in order to completely dissolve
the catalysts, which are otherwise slow to dissolve. The solutions were
then cooled to 4 °C, and Ac2O (32 µL, 0.34 mmol) was added to each
vial.
Aliquots were removed periodically from each reaction. The alcohol
and acetate were separated from the catalyst by chromatography, and
the conversion was assayed by GC.
The enantiomers of Ru-C5Ph5-DMAP were separated through
preparative HPLC (Daicel Chiralcel AD, 5 cm × 50 cm; ethanol/
hexane/diethylamine 5/95/0.3; 50 mL/min). One enantiomer was
Kinetic Resolution of 1-Phenylethanol (eq 9). A stock solution
was prepared of (()-1-phenylethanol (90 µL, 0.75 mmol) and NEt3
(78 µL, 0.56 mmol) in tert-amyl alcohol (1.5 mL). (-)-Ru-C5Ph5-
DMAP (1.9 mg, 0.0027 mmol) was weighed into a vial, and stock
solution (0.56 mL) was added to the vial. The vial was warmed slightly
in order to completely dissolve the catalyst. The solution was then
cooled to 4 °C, and Ac2O (18 µL, 0.19 mmol) was added.
After 4.5 h at 4 °C, an aliquot was removed. The alcohol and acetate
were separated from the catalyst by flash chromatography (25% f
75% EtOAc/hexane), and they were then analyzed by chiral GC
(Chiraldex BPH), which revealed a 71% ee of R acetate and a 54% ee
of S alcohol at 43% conversion (w s ) 10).
collected starting at 62 min ([R]20 ) +552.9° (c ) 0.14, CHCl3)),
D
and the other enantiomer was collected starting at 103 min.
A crystal suitable for a low-resolution X-ray crystal structure was
grown by slow diffusion at room temperature of hexane into an Et2O
solution of Ru-C5Ph5-DMAP, followed by cooling to -11 °C.
Acylation of (+)-1-Phenylethanol with Diketene (eq 5). A stock
solution was prepared of (()-1-phenylethanol (32 µL, 0.26 mmol) and
diketene (24 µL, 0.31 mmol) in CD2Cl2 (4.0 mL). Fe-Pyrrole* and
Ru-Pyrrole* (0.0050 mmol) were weighed into each of two vials, and
stock solution (1.6 mL) was added to each vial. Each reaction solution
was transferred to a screw-cap NMR tube, and the remaining stock
solution was transferred to a third screw-cap NMR tube (control
reaction). The reactions were monitored by 1H NMR in order to
determine the half-lives for reaction.
Acknowledgment. We thank Dr. Gregory A. Reichard
(Schering-Plough), Diego A. Hoic, Jack Liang, Michael M.-C.
Lo, J. Craig Ruble, and Jennifer Tweddell for assistance. Support
has been provided by the Alfred P. Sloan Foundation, the
American Cancer Society, the Camille and Henry Dreyfus
Foundation, Eli Lilly, Firmenich, Glaxo Wellcome, the National
Institutes of Health (National Institute of General Medical
Sciences, R01-GM57034), the National Science Foundation
(predoctoral fellowship to C.E.G.; Young Investigator Award,
with funding from Merck, Pharmacia & Upjohn, Bristol-Myers
Squibb, DuPont, Bayer, Rohm & Haas, and Novartis), Pfizer,
Procter & Gamble, and the Research Corporation.
Addition of Benzyl Alcohol to Phenylethylketene (eq 6). A stock
solution was prepared of benzyl alcohol (31 µL, 0.30 mmol) and
phenylethylketene (40 µL, 0.27 mmol) in C6D6 (2.8 mL). A portion
of this solution (0.7 mL) was added to each of three sealable NMR
tubes. A stock solution of Fe-Pyrrole* and of Ru-Pyrrole* (0.0070
mmol catalyst in 1.0 mL of C6D6) was prepared. One of the catalyst
stock solutions (0.1 mL) or C6D6 (0.1 mL; control reaction) was added
to each of the three NMR tubes. The reactions were then monitored
1
by H NMR to determine the half-lives for reaction.
Ring Opening of an Azlactone with MeOH (eqs 7 and 10). A
stock solution was prepared of azlactone (52.6 mg, 0.300 mmol),
benzoic acid (3.8 mg, 0.031 mmol), and MeOH (18 µL, 0.44 mmol) in
toluene-d8 (3.0 mL). (-)-Fe-DMAP* and (-)-Ru-DMAP* (0.0050
mmol) were weighed into each of two vials, and stock solution (1.0
mL) was added to each vial. Each reaction solution was transferred to
Supporting Information Available: Crystal structure data
for Ru-DMAP* and Fe-DMAP* (11 pages, print/PDF). See
any current masthead page for ordering information and Web
access instructions.
1
a sealable NMR tube, and the reactions were monitored by H NMR
in order to determine the half-lives for reaction. A separate control
experiment (no catalyst) was also run. After the reactions were
complete, the R-amino acid derivatives were isolated by flash chro-
matography and analyzed by chiral GC (Chiraldex GTA).
JA981061O