Gas-phase stereoselective binding to mn/salen catalysts

Article abstract of DOI:10.1021/ol8004563

Gas-phase equilibrium measurements have been used to determine the stereoselectivity of binding the enantiomers of 1-phenylethanol to manganese/salen asymmetric epoxidation catalysts. There is significant selectivity in the gas-phase binding, and the results are compared to data from condensed-phase epoxidations. The study demonstrates the utility of a novel internal standard approach that allows for rapid, accurate measures of the stereoselectivity of gas-phase ligand binding. Moreover, the data suggest that gas-phase binding stereoselectivity could be a potential predictor of condensed-phase enantioselectivity.

Full text of DOI:10.1021/ol8004563

ORGANIC
LETTERS
2008
Vol. 10, No. 9
1771-1773
Gas-Phase Stereoselective Binding to
Mn/Salen Catalysts
Adelaide E. Fagin,‡ Guoping Wang,† Michel C. Lau,‡ and Scott Gronert*,‡,†
Department of Chemistry and Biochemistry, San Francisco State UniVersity, 1600
Holloway AVenue, San Francisco, California 94132, and Department of Chemistry,
Virginia Commonwealth UniVersity, 1001 West Main Street, Richmond, Virginia 23284
sgronert@Vcu.edu
Received February 29, 2008
ABSTRACT
Gas-phase equilibrium measurements have been used to determine the stereoselectivity of binding the enantiomers of 1-phenylethanol to
manganese/salen asymmetric epoxidation catalysts. There is significant selectivity in the gas-phase binding, and the results are compared to
data from condensed-phase epoxidations. The study demonstrates the utility of a novel internal standard approach that allows for rapid,
accurate measures of the stereoselectivity of gas-phase ligand binding. Moreover, the data suggest that gas-phase binding stereoselectivity
could be a potential predictor of condensed-phase enantioselectivity.
Since their discovery in the mid-1990s by Jacobsen1 and
Katsuki,2 chiral manganese salen complexes, such as I and
II (Scheme 1), have proven to be exceptionally useful
under various reaction conditions.4–12 A commonly invoked
mechanism involving a radical intermediate, A, is presented
in Scheme 2. Although intermediates in metal/salen epoxi-
dations have been probed in the past by mass spectrom-
etry,13–20 here we provide the first report of the gas-phase
stereoselectivity of binding manganese/salen catalysts to
chiral ligands.
Scheme 1. Metal Salen Complexes
Previous gas-phase work with these catalysts has focused
on the reactivity of intermediates in the catalytic cycle.
Plattner and co-workers13–17 have shown that a variety of
ligated and oxidized Mn(salen)+ species can be formed by
(4) Palucki, M.; Finney, N. S.; Pospisil, P. J.; Guler, M. L.; Ishida, T.;
Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 948
.
(5) Finney, N. S.; Pospisil, P. J.; Chang, S.; Palucki, M.; Konsler, R. G.;
catalysts for asymmetric epoxidations.3 Multiple mechanisms
have been suggested, and it is possible that some compete
Hansen, K. B.; Jacobsen, E. N. Angew. Chem., Int. Ed. Engl. 1997, 36,
1720
.
(6) Jacobsen, H.; Cavallo, L. Organometallics 2006, 25, 177
(7) Cavallo, L.; Jacobsen, H. Inorg. Chem. 2004, 43, 2175
(8) Collman, J. P.; Zeng, L.; Brauman, J. I. Inorg. Chem. 2004, 43, 2672
(9) Adam, W.; Roschmann, K. J.; Saha-Moller, C. R.; Seebach, D. J. Am.
Chem. Soc. 2002, 124, 5068
(10) Abashkin, Y. G.; Burt, S. K. Org. Lett. 2004, 6, 59
(11) Miura, K.; Katsuki, T. Synlett 1999, 783
(12) Spin issues are critical on the reaction surface. Linde, C.; Akermark,
B.; Norrby, P. O.; Svensson, M. J. Am. Chem. Soc. 1999, 121, 5083
.
.
‡ San Francisco State University.
.
† Virginia Commonwealth University.
(1) Zhang, W.; Loebach, J. L.; Wilson, S. R.; Jacobsen, E. N. J. Am.
Chem. Soc. 1990, 112, 2801.
.
.
(2) Irie, R.; Noda, K.; Ito, Y.; Katsuki, T. Tetrahedron Lett. 1991, 32,
1055.
.
(3) McGarrigle, E. M.; Gilheany, D. G. Chem. ReV. 2005, 105, 1563.
.
10.1021/ol8004563 CCC: $40.75
Published on Web 04/11/2008
2008 American Chemical Society

model also comes from experimental work by Xia,22,23 who
has shown that these catalysts give stereoselective alcohol
oxidations. It should be emphasized that we are not attempt-
ing to model the epoxidation process itself (which is
mechanistically complex), but rather the general chiral
recognition features of the salen catalysts.
Scheme 2. Epoxidation Catalytic Cycle
The experiments were completed on a modified Finnigan
LCQ quadrupole ion trap,24 and nanospray ionization was
used to generate Mn(salen)+ ions from Mn(salen)+Cl-
precursors. The Mn(salen)+ ions were isolated in the ion trap
and then allowed to react with alcohol vapor mixtures (partial
pressure ∼10-6 torr) in the helium buffer gas. Cations I and
II readily add one alcohol ligand to give a five-coordinate
species. Addition of a second alcohol ligand is not observed
to a significant extent. After 200-500 ms, increasing the
reaction time did not alter the spectra, an indication that
equilibrium had been attained (Figure 2). Data are given in
electrospray ionization. The bare metal complexes, Mn-
(salen)+, can capture ligands in gas-phase reactions, and there
is evidence that the oxo species can transfer oxygen to
electron-rich alkenes and sulfides. Vairamani and co-work-
ers20 have observed related behavior with chromium salen
species. Finally, Chen and Chisholm18,19 have examined the
gas-phase reactivity of propylene oxide/M(salen)+ com-
plexes.
Intermediate A is not a practical target for gas-phase
experiments, so instead, we have focused on a very simple
model system to explore the asymmetric environment created
by the catalyst, namely the binding of a chiral alcohol to the
catalyst. Although the nature of the Mn-O bond is very
different and the axial ligand is missing in this model, it can
probe the overall steric interactions involved in binding to
an asymmetric salen complex. Support for this choice comes
from DFT calculations on the parent Mn(salen)+ species.21
Comparison of the geometries of complexes with •CH2CH2O•
and CH3CH2OH as ligands (Figure 1) shows that there are
Figure 2. (a) Reaction of (RR) I (Z ) C(CH3)3) with a 2:1
2-propanol/1-butanol mixture, 200 ms reaction time. (b) Reaction
of (RR) I (Z ) C(CH3)3) with a 2:1 2-propanol/1-butanol mixture,
500 ms reaction time. MnS+ is the reactant ion.
Table 1. For simple alcohols, it appears that polarizability
drives the equilibrium and larger alcohols are favored. Little
Table 1. Gas-Phase Equilibrium Constants for Binding Ligands
to Manganese Salen Catalysts (Scheme 1):a
Figure 1.
Optimized geometries of Mn(salen)+ complexed to (a)
•CH2CH2O• and (b) CH3CH2OH at the BP86 level in the quintet
state.
MnS(L1)+ + L2 f MnS(L2)+ + L1
L1
i-PrOH
BuOH
i-PrOH
L2
BuOH
catalyst
K
I (Z ) C(CH3)3)
I (Z ) C(CH3)3)
I (Z ) C(CH3)3)
I (Z ) C(CH3)3)
I (Z ) C(CH3)3)
I (Z ) CH3)
1.7b
1.8b
2.1b
2.0b
1.3c
1.3c
1.4c, d
1.5c
general similarities in the structures, with the major difference
being a longer Mn-O distance with the alcohol. This
suggests that alcohol models can provide a rough measure
of the asymmetric environment created by the salen ligand
but will exhibit weaker steric interactions. Support for the
PentOH
(R)-2-BuOH
(S)-2-BuOH
(S)-1-PhEtOH
(S)-1-PhEtOH
(S)-1-PhEtOH
(S)-1-PhEtOH
i-PrOH
(R)-1-PhEtOH
(R)-1-PhEtOH
(R)-1-PhEtOH
(R)-1-PhEtOH
II (Z ) H)
II (Z ) OCH3)
(13) Feichtinger, D.; Plattner, D. A. Angew. Chem., Int. Ed. Engl. 1997,
36, 1718.
a (R,R) enantiomer of catalyst. Averages of at least six measurements
on three different days. b Direct equilibrium measurements with expected
uncertainties of (10%. c From experiments employing internal standard
and corrected for isotope effects (Scheme 3, eqs 1-4). Expected uncertain-
ties are ( 4%. d SS salen was used. 1/K is listed for consistency.
(14) El-Bahraoui, J.; Wiest, O.; Feichtinger, D.; Plattner, D. A. Angew.
Chem., Int. Ed. 2001, 40, 2073.
(15) Plattner, D. A. Int. J. Mass Spectrom. 2001, 207, 125.
(16) Plattner, D. A.; Feichtinger, D.; El-Bahraoui, J.; Wiest, O. Int. J.
Mass Spectrom. 2000, 196, 351.
(17) Feichtinger, D.; Plattner, D. A. Chem. Eur. J. 2001, 7, 591.
(18) Chen, P.; Chisholm, M. H.; Gallucci, J. C.; Zhang, X. Y.; Zhou,
Z. P. Inorg. Chem. 2005, 44, 2588.
stereoselectivity is seen with the enantiomers of 2-butanol,
but this is not surprising given its small size.
To gain more insight into the stereoselectivity of the
binding, we turned to 1-phenylethanol (1-PhEtOH) as a probe
(19) Schon, E.; Zhang, X. Y.; Zhou, Z. P.; Chisholm, M. H.; Chen, P.
Inorg. Chem. 2004, 43, 7278.
(20) Kumar, M. K.; Prabhakar, S.; Kumar, M. R.; Reddy, T. J.;
Premsingh, S.; Rajagopal, S.; Vairamani, M. Rapid Commun. Mass
Spectrom. 2004, 18, 1103.
1772
Org. Lett., Vol. 10, No. 9, 2008

alcohol.25 For greater accuracy in the measurements, an
internal standard approach was adopted (Scheme 3). The
in the complexation process. Although small in comparison
to the ee’s typically seen in epoxidations, the value is
significant for a room temperature gas-phase process, par-
ticularly given the weak binding in the alcohol complex. The
data also indicate a modest isotope effect correction, KH/KD
) 0.97. It is interesting to note that a preference for reaction
with the S enantiomer of the alcohol previously was seen in
the condensed-phase oxidation of 1-phenylethanol with
diacetoxyiodobenzene and the R,R enantiomer of I (Z )
C(CH3)3).22 However, the mechanism of that reaction is
complicated by the need for KBr as a cocatalyst.
Scheme 3. Gas-Phase Reaction System
Experiments using the (R,R) enantiomer of catalyst I (Z
) CH3) give nearly the same preference and have a KS/R
value of 1.29. However, changing the overall structure of
the catalyst has a significant effect on the stereoselectivity
of the ligand binding. Similar experiments with catalysts II
(Z ) H) and II (Z ) OCH3) lead to considerably larger KS/R
values, 1.45 and 1.48, respectively. These values correspond
to de’s of approximately 20%. The effect of catalyst structure
on binding stereoselectivity is noteworthy because condensed-
phase epoxidations of styrene26 give higher ee’s for catalysts
of structure II than I.27 These outcomes suggest that the gas-
phase binding experiments might offer insight into the
inherent stereoselectivity of the epoxidation catalyst.
In summary, we have reported a sensitive, gas-phase
technique for assessing the stereoselectivity of model ligand
binding to manganese/salen catalysts and observed systematic
trends in binding selectivity. In addition, there are some
intriguing correlations with condensed-phase enantioselec-
tivities. The approach can be generalized to other systems
and potentially provide crude, but rapid measures of cata-
lysts’ asymmetric induction capabilities. Moreover, the data
provide excellent target values for calibrating the accuracy
of computational models of binding stereoselectivity in these
model systems.
catalyst ion was allowed to react with a 50/50 mixture of
enantiomerically pure C6H5CHOHCH3 (R or S) and racemic
C6D5CHOHCH3. This approach eliminates the need for
absolute pressure measurements and cancels out any mass
discrimination in ion detection. Assuming that all the
diastereomeric complexes have similar isotope effects, the
stereochemical preference, KS/R, and generic isotope effect,
KH/KD, can be extracted from the data. The method is general
and greatly reduces experimental uncertainties.
Sample spectra are shown in Figure 3 for the R,R
enantiomer of I (Z ) C(CH3)3). The data indicate an absolute
Figure 3. (a) Reaction of (R,R)-I (Z ) C(CH3)3) with a 1:1 mix of
Acknowledgments. We thank Professor Eric Jacobsen for
the generous donation of the catalysts used in the study.
Funding from the Petroleum Research Fund (46857-AC4)
and the NSF (CHE-0348809) is gratefully acknowledged.
(R)-C6H5CH(OH)CH3 and racemic C6D5CH(OH)CH3. (b) Reaction
of (R,R)-I (Z ) C(CH3)3) with a 1:1 mix of (S)-C6H5CH(OH)CH3
and racemic C6D5CH(OH)CH3. MnS+ is the reactant ion.
Supporting Information Available: Tables containing the
computed geometries and the measured isotope effects. This
material is available free of charge via the Internet at
http://pubs.acs.org.
binding free energy of about -12.5 kcal/mol for the alcohols.
Applying the equations in Scheme 3 to this data, one obtains
an S preference of 1.29, which is equivalent to a de of 12%
OL8004563
(21) All calculations were completed with Gaussian03. The theoretical
approach was based on earlier work by Cavallo and Jacobsen (refs 6 and
7). A quintet state was used with the BP86 functional. A triple-ꢀ basis set
was used on Mn (6-311+G*), and double-ꢀ basis sets were used for the
other atoms (6-31+G* for O,N and 6-31G* for C,H). Frisch, M. J. Gaussian
03, Revision B04; Gaussian, Inc.: Pittsburgh, PA, 2003. See the Supporting
Information for the full reference.
(24) Gronert, S. Mass Spectrom. ReV. 2005, 24, 100.
(25) Stereoselectivity has been observed in the gas-phase fragmentation
of 1-phenylethanol/vanadium/binaphthol complexes. Schroder, D.; Schwarz,
H. Int. J. Mass. Spectrom. 2004, 231, 139.
(26) Styrene is the best analogy to the 1-phenylethanol system, but it is
imperfect because one expects initial C-O bond formation at the ꢁ carbon
in the epoxidation.
(22) Li, Z.; Tang, Z. H.; Hu, X. X.; Xia, C. G. Chem. Eur. J. 2005, 11,
1210.
(23) Sun, W.; Wang, H. W.; Xia, C. G.; Li, J. W.; Zhao, P. Q. Angew.
Chem., Int. Ed. 2003, 42, 1042.
(27) Palucki, M.; Pospisil, P; J, W., Z.; Jacobsen, E. N. J. Am. Chem.
Soc. 1994, 116, 9333.
Org. Lett., Vol. 10, No. 9, 2008
1773