Theoretical prediction of selectivity in kinetic resolution of secondary alcohols catalyzed by chiral DMAP derivatives

Article abstract of DOI:10.1021/ja302420g

The mechanism of esterification of the secondary alcohol 1-(1-naphthyl)ethanol 9 by isobutyric anhydride catalyzed by 4-pyrrolidinopyridine (PPY, 11) and a series of single enantiomer atropisomeric 4-dialkylaminopyridines 8a-g has been studied computationally at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level. Comparison of the levels of enantioselectivity predicted computationally with the results obtained experimentally allowed the method to be validated. The value of the approach is demonstrated by the successful prediction that a structural modification of an aryl group within the catalyst from phenyl to 3,5-dimethylphenyl would lead to improved levels of selectivity in this type of kinetic resolution (KR) reaction, as was subsequently verified following synthesis and evaluation of this catalyst (8d). Experimentally, the selectivity of this type of KR is found to exhibit a significant deuterium isotope effect (for 9 vs d₁-9).

Full text of DOI:10.1021/ja302420g

Article
pubs.acs.org/JACS
Theoretical Prediction of Selectivity in Kinetic Resolution of
Secondary Alcohols Catalyzed by Chiral DMAP Derivatives
Evgeny Larionov,†,∥ Mohan Mahesh,‡,§ Alan C. Spivey,*,‡ Yin Wei,†,⊥ and Hendrik Zipse*,†
†Department of Chemistry, LMU Munchen, Butenandtstrasse 5-13, D-81377 Munchen, Germany
̈
̈
‡Department of Chemistry, South Kensington Campus, Imperial College, London, SW7 2AZ, United Kingdom
S
* Supporting Information
ABSTRACT: The mechanism of esterification of the
secondary alcohol 1-(1-naphthyl)ethanol 9 by isobutyric
anhydride catalyzed by 4-pyrrolidinopyridine (PPY, 11) and
a series of single enantiomer atropisomeric 4-dialkylaminopyr-
idines 8a−g has been studied computationally at the B3LYP/
6-311+G(d,p)//B3LYP/6-31G(d) level. Comparison of the
levels of enantioselectivity predicted computationally with the
results obtained experimentally allowed the method to be
validated. The value of the approach is demonstrated by the successful prediction that a structural modification of an aryl group
within the catalyst from phenyl to 3,5-dimethylphenyl would lead to improved levels of selectivity in this type of kinetic
resolution (KR) reaction, as was subsequently verified following synthesis and evaluation of this catalyst (8d). Experimentally, the
selectivity of this type of KR is found to exhibit a significant deuterium isotope effect (for 9 vs d1-9).
Spivey et al. have also developed a series of axially chiral,
atropisomeric derivatives of 4-dialkylaminopyridines 8 as
catalysts for the KR of racemic secondary alcohols (Scheme 1).9
INTRODUCTION
■
Organocatalysis has been at the forefront of research in organic
chemistry in recent years, and one of the most studied fields
concerns acyl group transfer reactions mediated by nucleophilic
chiral catalysts.1 Structurally diverse amine, phosphine, and
alcohol derivatives have been designed and synthesized for the
kinetic resolution (KR) of alcohols and related stereoselective
transformations.2 In particular, chiral 4-dimethylamino-pyridine
(DMAP) derivatives 1−5 have been demonstrated to be
effective catalysts for enantioselective acyl-transfer reactions to
alcohols by Vedejs,3 Kawabata,4 Fu,5 and others.6 Additionally,
Birman7 and others8 have demonstrated that various bicyclic
amidines (e.g., 6 and 7) are highly effective catalysts for this
type of transformation (Chart 1).
Scheme 1. KR of Alcohol 9 by Isobutyrylation, Catalyzed by
Atropisomeric DMAP Derivatives 89c
Chart 1. Representative Structures of 4-DMAP and Amidine-
Based Catalysts Used for the KR of Alcohols
These experiments were performed using racemic 1-(1-
naphthyl)ethanol 9 as the substrate and isobutyric anhydride (1
equiv) as acyl donor in the presence of 1 mol % of the
enantiomerically pure biaryl catalysts 8. Under these conditions,
the alcohol (R)-9 reacts faster than the alcohol (S)-9 to
produce ester (R)-10, and a selectivity factor s = 16 was
obtained using the 4-diethylamino catalyst (−)-(Sa)-8a at −78
°C.9c,10 In a similar manner, KR experiments performed using
Received: March 12, 2012
Published: May 8, 2012
© 2012 American Chemical Society
9390
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
atropisomeric derivatives having various 4-dialkylamino groups
revealed that the selectivities (and activities) of these catalysts
depend on the nature of this group with the 4-di-n-butyl
derivative 8e showing the optimal combination of selectivitiy
and activity.9c These findings motivated us to perform a
combined theoretical and experimental investigation to better
understand the acylation reaction of racemic alcohols and the
factors influencing the selectivities of these catalysts.
Currently, only a small number of theoretical studies have
been reported in which not only conformational properties of
the acylpyridinium intermediates have been studied4a,11 but also
direct prediction of the outcome of KR experiments with
alcohols has been attempted. These studies deal, however, with
chiral imidazole and amidine derivatives. Sunoj et al.12a have
studied computationally the enantioselective acetylation of
trans-cyclohexane-1,2-diol, catalyzed by a N-methylimidazole-
based peptide, which was designed by Schreiner et al.12b These
authors have shown that hydrogen bonding between the diol
substrate and the peptide backbone plays an important role in
determining the enantioselectivity. Analysis of the transition-
state structures for the acylation of 1-phenylethanol catalyzed
by amidine 6 (Chart 1), carried out by Houk et al., supports the
decisive role of π−π interactions in mediating chiral recognition
in the KR of secondary benzylic alcohols.7h
With the aim to improve our understanding of the
relationship between catalyst structure and the level of
enantioselectivity imparted during chiral DMAP catalysis of
acyl-transfer reactions to the point where predictions can be
made, we have investigated computationally the acylation of the
racemic secondary alcohol 1-(1-naphthyl)ethanol 9 catalyzed
by the series of catalysts 8 in detail. The key question here is
whether the enantioselectivities of chiral DMAP-catalyzed acyl-
transfer reactions can be rationalized by examination of the
transition state (TS) of the rate-determining step that is also
considered as the selectivity-determining step.
Single point calculations were again performed at the B3LYP/
6-311+G(d,p) level of theory as well as at MP2(FC)/6-
311+G(d,p) and MP2(FC)/6-31+G(2d,p) levels with Gaussian
03.15 Dispersion corrections to the DFT (termed DFT-D)
proposed by S. Grimme16 were used to calculate the accurate
dispersion interaction at the B3LYP-D/6-311+G(d,p) level
using the ORCA 2.6.4 program package.17 Thermochemical
corrections to free energies and enthalpies at 298.15 K (G298
and H298) as well as at 195.15 K (H195 and G195) were
calculated at the same level as that used for geometry
optimization.
RESULTS AND DISCUSSION
■
Catalysis by PPY. Catalyst 8a has been shown to be very
effective for the KR of secondary alcohols.9 However, the
selectivity values can vary slightly as a function of conversion: s
= 16.7 at 12% conversion (2 h) using 1 equiv anhydride and s =
15.9 at 27% conversion (8 h) with 1 equiv anhydride (for the
reaction shown in Scheme 1, see Supporting Information).10
There are also several communications of the conversion-
dependent selectivity in the literature.18 Additionally, it was
noted, that the reaction mixture at −78 °C is not
homogeneous, while at room temperature no precipitation
was observed.9d This phenomenon may be explained by the
formation of insoluble triethylammonium salts (in toluene),
which can in principle affect the reaction (rate or/and
selectivity). Should this be the case, we must expect that the
Arrhenius plot for the reaction rate or even selectivity will not
necessarily remain linear. In order to understand the role of
precipitation, we decided to study the reaction kinetics at
different temperatures. As a model system, the isobutyrylation
of racemic 1-(1-naphthyl)ethanol 9 catalyzed by PPY (11) was
chosen (Scheme 2).
Scheme 2. Isobutyrylation of 1-(1-Naphthyl)ethanol 9,
Catalyzed by PPY (11)
COMPUTATIONAL DETAILS
■
All conformers of reactants and products were searched
carefully and optimized at the B3LYP/6-31G(d) level, and
single point calculations were added at the B3LYP/6-311+G-
(d,p) level of theory in order to obtain relative enthalpies at
B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level. The systems
studied here are very flexible and have a large conformational
space. The conformational spaces of transition states TS1 and
TS2 for the 4-pyrrolidinopyridine (PPY) catalyst (11) and 4-
diethylaminopyridine catalyst 8a were initially studied with the
optimized potentials for liquid simulations (OPLS)-AA force
field using the Monte Carlo conformational search facility
implemented in BOSS 4.6.12 The conformational space of
transition state TS2 for catalysts 8b and 8c was also initially
studied with the OPLS-AA force field.13 The conformers
identified by the force field within an energy window of 40 kJ
mol−1 were then reoptimized at the B3LYP/6-31G(d) level of
theory, and single point calculations were carried out at the
B3LYP/6-311+G(d,p) level of theory. Intrinsic reaction
coordinate (IRC) calculations were executed using the
lowest-energy conformers of these TSs to obtain structures of
the reactant complex, the intermediate acylpyridinium salt
(pyac) on the nucleophilic catalysis pathway, and the product
complex. The transition states TS3 and TS4 for catalysts PPY
(11) and 8a on the base-catalyzed pathway were located based
on the previously suggested “four-” and “six-membered” ring
structures14 and optimized at the B3LYP/6-31G(d) level.
This reaction follows second-order kinetics, and rate
constants were measured at eight different temperatures from
−25 °C to −80 °C. In order to determine the activation
parameters, the obtained data were fitted using the Eyring
equation (Figure 1).
Figure 1. Eyring plot for the reaction shown in Scheme 2.
9391
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
The value obtained for the activation enthalpy ΔH‡ of 12.8
reactions feature a similar combination of minimal activation
enthalpies ΔH‡ and large, negative activation entropies ΔS‡.20
Computational Study of the Catalytic Cycle with PPY
(11). The DMAP-catalyzed acylation of alcohols by anhydrides
and acyl chlorides is generally believed to proceed via a
nucleophilic catalysis mechanism.1b This mechanism is
supported by the computational study of the DMAP-catalyzed
acetylation of tert-butanol with acetic anhydride.14 It has
however been noted that in the case of primary alcohols a
mechanism involving base catalysis may become competitive.21
In order to investigate, which mechanism of catalysis
(nucleophilic or general base) is operative for the isobutyr-
ylation of secondary alcohol 9 catalyzed by PPY (11), we
studied the catalytic cycle computationally. The free energy
profile as calculated at the B3LYP/6-311+G(d,p)//B3LYP/6-
31G(d) level at the experimental temperature of 195.15 K is
shown in Figure 2 (enthalpies ΔH‡ and free energies ΔG‡ at
298 K are shown in the Supporting Information).
First, the reaction profile was calculated assuming initial
formation of a ternary complex of reactants and catalyst PPY
(11) for both the nucleophilic and general base catalysis
pathways.14 Along the nucleophilic catalysis pathway, the
reactant complex passes through the first transition state
TS1·9 to yield intermediate pyac·9, which then passes through
the second TS2 with concomitant proton transfer to give the
product complex. The alternative base catalysis pathway
proceeds through the concerted “six-membered” ring TS3 or
kJ mol−1 (Table 1) is quite small for a reaction in solution,
Table 1. Comparison of Activation Parameters for the
Acylation of Alcohol 9 Catalyzed by PPY (11)
a
b
b
c
activation parameter
exp
TS1·9
TS2
TS2
ΔH‡, kJ mol−1
ΔG‡, kJ mol−1
ΔS‡, J mol−1 K−1
+12.8
+23.8
+177.5
−788
+11.0
+168.3
−806
+12.6
+118.8
−356
d
+84.3
−240
a
Experimental values from the Eyring plot from 193 to 248 K.
b
Activation parameters of TS1·9 and TS2 calculated at B3LYP/6-
c
311+G(d,p)//B3LYP/6-31G(d) level at 195 K. Calculated at 298 K.
d
At 298 K.
whereas the negative value obtained for the activation entropy
ΔS‡ of −240 J mol−1 K−1 is typical for a bimolecular
reaction.19,20 By comparison, Vedejs reported that the
isobutyrylation of 1-(1-naphthyl)ethanol 9 catalyzed by a chiral
phospholane in toluene also had a small activation enthalpy
(ΔH‡ = 5.8 or 12.5 kJ mol−1, depending on the reacting
enantiomer of 9)19 and that the activation entropies were
similarly large and negative irrespective of the reacting
enantiomer of 9 (ΔS‡ = −311 or 306 J mol−1 K−1). Vedejs’
data imply that the ΔG‡ term, which determines enantiomer
discrimination, is dominated by differences in activation
enthalpies ΔH‡. The kinetic parameters of other acyl-transfer
Figure 2. Free energy profile (ΔG195) of the PPY-catalyzed acylation of alcohol 9 as calculated at B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level.
9392
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
Figure 3. Gas-phase free energy profile (ΔG298) of the acylation of alcohol 9 catalyzed by (−)-(Sa)-8a as calculated at the B3LYP/6-311+G(d,p)//
B3LYP/6-31G(d) level of theory.
“four-membered” ring TS4 to the product complex in a single
step. The six-membered transition state TS3 of the base-
catalyzed route is slightly more favorable than the four-
membered TS4. The activation parameters (relative to separate
reactants) for transition states TS1·9 and TS2 calculated at the
B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level of theory are
collected in Table 1.
kJ mol−1) is close to the theoretically predicted value for the
rate-limiting transition state TS2 (+12.6 kJ mol−1 at 298 K,
Table 1). The activation entropy of TS2 (−356 J mol−1 K−1 at
298 K) deviates significantly from the experimental value of
−240 J mol−1 K−1. These latter deviations (cf., ΔG‡
84.3
exp 298
vs ΔG‡
118.8 kJ mol−1) may be due to the presently
calc 298
employed harmonic oscillator/rigid rotor model for thermal
corrections.
Both enthalpy and free energy values suggest the nucleophilic
route to be more favorable and the first step of this route to be
rate-limiting. This was surprising, because the second step is
generally considered to be rate determining.1b,14 However,
inclusion of the alcohol molecule in the first step appeared to
be entropically unfavorable. Therefore an alternative path
through TS1 and intermediate pyac, which does not include
alcohol 9, was calculated (marked in red in Figure 2). The new
transition state TS1 without alcohol 9 has a higher enthalpy
than TS1·9 but is more stable in terms of free energy. The
subsequent intermediate pyac can then react with alcohol 9
through TS2 to give products. Interestingly, the covalently
bound dihydropyridine intermediate pyac_cov is slightly more
stable than the ion pair pyac and may be present in equilibrium
with pyac. Transition state TS2 (marked in green in Figure 2)
has a higher free energy than TS1, and therefore via this
nucleophilic catalysis pathway, the second step is rate limiting.
The experimental value of the activation enthalpy ΔH‡ (+12.8
In order to test whether these observations persist at other
theoretical levels, additional single point calculations were
performed using the B3LYP/6-31G(d) structures. The MP2
method as well as the DFT method with dispersion correction
(DFT-D)16 were chosen in combination with the 6-311+G-
(d,p) basis set due to their superior performance in describing
dispersion interactions. The results show that the use of MP2
and B3LYP-D levels for single point calculations stabilizes all
the TSs relative to the reactants and reactant complex (see
Supporting Information). All these different theoretical
methods predict the base-catalyzed route to be less favorable
than the nucleophilic route (by ca. 30−40 kJ mol−1). Transition
states of the base-catalyzed route, TS3 and TS4, have similar
energies at different levels of theory and are therefore equally
feasible for the base-catalyzed pathway.
Computational Study of the Catalytic Cycle with
Catalyst (−)-(Sa)-8a. In order to check whether the
9393
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
mechanism of the acylation of secondary alcohols catalyzed by
PPY (11) also persists for the acylation catalyzed by the chiral
catalyst (−)-(Sa)-8a, we next investigated the nucleophilic and
general base catalysis pathways for the reaction of racemic 1-(1-
naphthyl)ethanol (9) with isobutyric anhydride catalyzed by
(−)-(Sa)-8a also at the B3LYP/6-311+G(d,p)//B3LYP/6-
31G(d) level of theory. The nucleophilic and general base
catalysis pathways are plotted in Figure 3 using the lowest-
energy conformers. The diastereoisomeric TSs and intermedi-
ates are denoted as (R)-* and (S)-*, depending on the
configuration of the involved alcohol.
with experimental observation that PPY (11) is more
catalytically active than compound 8a. Furthermore, the free
energy difference of 5.1 kJ mol−1 between (R)- and (S)-TS2 is
accompanied by an enthalpy difference of 6.0 kJ mol−1,
confirming that the origin of stereoselection is essentially of
enthalpic nature.
Analysis of the optimized geometries of transition states TS2
reveals that all conformers can be classified into one of the four
structural types shown in Figure 4, which also shows a pictorial
The diastereoisomers including (R) alcohol (9) are always a
few kJ mol−1 lower than those including the (S) alcohol. The
most energetically favorable transition state (R)-TS3 along the
base catalysis pathway is located 60 kJ mol−1 above transition
state TS1 and 51 kJ mol−1 above transition state (R)-TS2 of the
nucleophilic catalysis pathway (Table 2). Single point
Table 2. Relative Enthalpies ΔH298 and Free Energies ΔG298
(in kJ mol−1) for Stationary Points Located on the Potential
Energy Surface at the B3LYP/6-311+G(d,p)//B3LYP/6-
31G(d) Level in the Gas Phase for the Acylation of Alcohol 9
Promoted by Catalyst (−)-(Sa)-8a
ΔH298
ΔG298
nucleophilic catalysis
(R)-
(S)-
(R)-
(S)-
reactants
0.0
0.0
reactant complex
−23.0
26.8
7.6
−22.5
34.3
59.7
133.3
106.3
64.2
134.2
111.1
TS1·9
pyac·9
11.7
TS1
58.7
57.8
114.6
110.4
pyac
TS2
14.1
20.1
124.0
129.1
product complex
products
−87.4
−86.9
−0.7
−4.7
−21.6
−19.4
base catalysis (concerted)
TS3
(R)-
(S)-
(R)-
(S)-
67.2
77.5
175.3
185.7
calculations have also been performed at the MP2/6-31G-
(d)//B3LYP/6-31G(d) level of theory for the best conformers
of the (R)-TSs (see Supporting Information). The energy of
(R)-TS3 is also higher than that of (R)-TS1 and (R)-TS2 by
more than 30 kJ mol−1 at the MP2/6-31G(d)//B3LYP/6-
31G(d) level. This indicates that the nucleophilic catalysis
pathway is more favorable than the general base catalysis
pathway, which is in line with the results for the acylation
catalyzed by PPY (11) discussed above.
The free energy difference between the diastereoisomeric
TSs of the rate-determining step is the key value for predicting
the enantioselectivity. In accordance with the results for PPY-
catalyzed acylation, the first transition state TS1 has a lower
free energy when alcohol 9 is not included in the structure
(Table 2). Hence, in the nucleophilic route the reactants pass
through the first transition state TS1 to yield intermediate pyac,
which is common for both alcohol enantiomers. After
complexation with either the (R) or (S) alcohol 9, the (R)-
or (S)-pyac·9 intermediates pass through their respective
second TS2s to the product complex. Transition state TS2 is
rate limiting for both alcohols and should therefore also be
selectivity determining. Free activation energies ΔG298 for the
acylation catalyzed by (−)-(Sa)-8a (124.0 kJ mol−1 for (R)-TS2
and 129.1 kJ mol−1 for (S)-TS2) are higher than the ΔG298
value calculated for PPY (118.8 kJ mol−1). This is in accordance
Figure 4. Relative enthalpies (kJ mol−1) of conformers of TS2 with
catalyst (−)-(Sa)-8a (top) and structures of the most stable
conformers of (R)-TS2 (left bottom) and (S)-TS2 (right bottom)
as calculated at the B3LYP/6-31G(d) level of theory. Each low-energy
conformer in classes I−IV is represented by an individual symbol (see
Supporting Information for further details). Distances are given in Å.
representation of the relative energies of the conformers of (R)-
and (S)-TS2. Generally speaking, the carboxyl carbonyl group
is bound to the left or right side of the pyridine ring by weak
hydrogen bonding, and the alcohol approaches the reaction
center from either the front or the back face of the pyridine
ring. In class I the carboxylate group is bound to the right side
of the pyridine ring, and the alcohol approaches the reaction
center from the back. For this class, the conformers with the
(R) alcohol are more stable than the conformers with the (S)
alcohol by more than 20 kJ mol−1. In classes II and III the
conformers with the (S) alcohol are more stable than the
conformers with the (R) alcohol. All the conformers in class IV
have rather poor stabilities with both the (R) and the (S)
alcohols. The most stable conformer with the (R) alcohol
belongs to class I, which is more stable than the most stable
conformer with the (S) alcohol (belonging to class III) by 5.1
kJ mol−1. Thus, the calculations predict that when employing
9394
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
Table 3. Comparison of Experimental9d,10 and Calculated Energy Differences ΔH(S − R) and ΔG(S − R) (in kJ mol−1) of the
a
Diastereoisomers of TS2 for Catalysts 8a−g
b
c
c
c
experiment
B3LYP/6-31G(d)
B3LYP/6-311+G(d,p)
B3LYP-D/6-311+G(d,p)
catalyst
s
ln s
ΔG195
ΔG298
ΔH298
ΔG195
ΔH195
ΔG298
ΔH298
ΔG298
ΔH298
8a
8b
8c
8e
16
10
3.5
31
2.77
2.30
1.25
3.43
4.49
3.73
2.03
5.56
4.19
3.67
5.35
4.62
6.77
5.33
5.48
4.80
5.65
5.60
6.13
6.12
6.09
1.93
1.42
1.45
8.23
3.99
8.32
4.31
9.90
5.82
4.18
0.17
10.27
0.0224
8.23
5.70
8.29
5.75
11.99
0.0205
14.06
7.01
6.01
15.30
0.5509
2.13
9.49
d
e
correlation coefficient R2
0.9641
6.72
0.0011
6.26
0.9485
6.70
0.3987
9.29
0.6496
4.50
8d
8f
27
11
9
3.30
2.40
2.20
5.35
3.89
3.57
4.78
4.41
4.76
4.51
6.14
5.85
4.22
8g
3.98
2.12
2.74
2.11
7.37
5.04
0.91
−0.98
0.5666
correlation coefficient R2
0.0967
0.4702
0.0103
0.3975
0.1812
0.2667
0.2307
a
b
Positive numbers imply a preference for (R) alcohol. Selectivities s are taken from the experimental results from ref 9c for catalysts 8b, 8c, and 8e
(all using 2 equiv anhydride, t = 9 h) or were recorded in this work for catalysts 8a,10 8d, 8f, and 8g (all using 1 equiv anhydride, t = 8 h, see
Supporting Information). Level of theory used for single point calculations based on B3LYP/6-31G(d) geometries and thermal corrections. Energy
c
values are Boltzmann-averaged over the maximum available number of conformers (the actual numbers of conformers used for averaging are shown
d
in the Supporting Information). Correlation of the calculated energy differences ΔH(S − R) and ΔG(S − R) with experimental values ln s for
e
catalysts 8a−c and 8e. Correlation of the calculated energy differences ΔH(S − R) and ΔG(S − R) with experimental values ln s for all studied
catalysts 8a−g.
(−)-(Sa)-8a as the KR catalyst, the (R) alcohol 9 should react
faster than the corresponding (S) alcohol, which is in line with
experimental observation.9
The B3LYP/6-31G(d) optimized structures of the most
stable conformers of (R)- and (S)-TS2 as shown in Figure 4
illustrate that alcohol 9 (shown in light green) approaches the
reaction center from the back face of the pyridine ring in (R)-
TS2 and from the front face of the pyridine ring in (S)-TS2.
There is no significant steric hindrance when the alcohol
approaches the reaction center from the back face of the
pyridine in (R)-TS2. In contrast, alcohol approaching from the
front face of the pyridine in (S)-TS2 encounters some steric
repulsion between the tilted phenyl ring of the catalyst 8a and
the naphthyl ring of alcohol 9, thus raising the energy of (S)-
TS2 relative to that of (R)-TS2 by ca. 5 kJ mol−1.
Figure 5. Correlation between experimental enantioselectivities and
calculated enthalpy differences ΔH298 between (R)- and (S)-TS2, as
calculated at the B3LYP/6-31G(d) level.
The Origin of Selectivity: Variations of the Dialkyla-
mino Group. Spivey et al. have shown experimentally that
varying the 4-dialkylamino substituent significantly influences
the selectivities of catalysts.9c The selectivities decreases in the
level predict the (R) alcohol to be more reactive than the (S)
alcohol, which is in full agreement with experimental results.
order 8e, 8a, 8b, 8c, with pyrrolidino-substituted catalyst 8c
being the least selective (see Scheme 1). We therefore chose to
investigate theoretically the selectivities of the catalyst series
8a−8e using the same alcohol [1-(1-naphthyl)ethanol, 9] as
substrate in order to compare to the experimental results. The
enthalpy and free energy differences between the diaster-
eoisomers of TS2, which were considered to be the selectivity-
determining TSs, were calculated for catalysts 8a−c and 8e by
DFT methods and are listed in Table 3.
The calculated free energy differences ΔG298 for TS2 for
catalysts 8a−c and 8e calculated at B3LYP/6-31G(d) level did
not reproduce experimental values of ΔG195, calculated using eq
1; indeed, they predict that the pyrrolidino-substituted catalyst
8c should have higher selectivity than both 8a and 8b (Table
3). However, the calculated enthalpy differences ΔH298 do
correlate with experimental selectivities quite closely (Figure 5,
blue line, correlation coefficient R2 = 0.9641 for catalysts 8a−c
and 8e). The thermal corrections recalculated at 195 K do not
improve the correlation between experimental results and
calculated free energies or enthalpies (Table 3). In general,
enthalpy and free energy differences between (R)- and (S)-TS2
for catalysts 8a−c and 8e calculated at the B3LYP/6-31G(d)
ΔΔG1≠95
RT
kR
kS
ΔGSTS − ΔGRTS
ΔGTS
RT
ln s = ln
= −
=
=
RT
(1)
Employing the combined DFT method B3LYP/6-311+G-
(d,p)//B3LYP/6-31G(d) also does not yield a better
correlation of experimental selectivities with calculated free
energies or enthalpies (Table 3). The inclusion of dispersion
corrections (at B3LYP-D/6-311+G(d,p) level), however,
improves the correlation with calculated free energies. More-
over, after addition of the dispersion corrections, the enthalpy
differences ΔH298 are significantly smaller. Noticeably, the
enthalpy differences ΔH298 between (R)- and (S)-TS2 for
catalysts 8a−c and 8e calculated at the B3LYP-D/6-311+G-
(d,p)//B3LYP/6-31G(d) level can be correlated with exper-
imental enantioselectivities (Table 3, correlation coefficient R2
= 0.6496 for catalysts 8a−c and 8e). Hence it is apparent that
enthalpy differences ΔH298 between (R)- and (S)-TS2 for
catalysts 8a−c and 8e calculated at the B3LYP/6-31G(d) or the
combined B3LYP-D/6-311+G(d,p)//B3LYP/6-31G(d) levels
9395
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
can be used for the rationalization of experimentally measured
enantioselectivities in KR experiments.
This method was therefore used to calculate enthalpy
differences ΔH298(S − R) for catalysts 8d, 8f, and 8g, shown
in Chart 2. The results obtained for ΔH298(S − R) at the
B3LYP/6-31G(d) level show that the catalysts 8f and 8g are
expected to display moderate levels of stereoselectivity, while
the derivative 8d should display greater levels of selectivity
relative to the parent catalyst 8a (see Table 3).
The higher calculated enthalpy difference ΔH298(S − R) for
catalyst 8d relative to 8a can be rationalized by unfavorable
interactions between the naphthyl ring system of the (S)-
alcohol and one of the methyl groups of the 3,5-dimethylphenyl
substituent in the catalyst in (S)-TS2 of type III. This leads to a
higher relative stability of the type II conformers (see
Supporting Information for further details). For catalysts 8f
and 8g the para substituents exert no corresponding steric
repulsion, and electronic effects do not appear to play an
important role.
Synthesis of Catalysts 8d, 8f, and 8g and Determi-
nation of Their Experimental Selectivities for KR of
Alcohol 9 with Isobutyric Anhydride. Synthesis of these
new analogues followed a synthetic route adapted from that
used to make the parent catalyst 8a starting from commercially
available 4-pyridone and 1-bromo-2-naphthol and proceeding
via triflate 12 (Scheme 3, for details, see Supporting
Information).9a
Comparison of TS structures for 8a vs 8e, i.e., containing
NEt2 vs NBu2 groups, respectively, reveals what appear to be
only small differences, of which the slightly more advanced
proton transfer between alcohol and carboxylate counterion in
the (R)-TS2(8e) is the most notable stabilizing feature (see
Supporting Information). That the butyl substituents in catalyst
8e interact differently with the surrounding biaryl π-systems in
TSs of the (R)- and (S)-alcohols can be seen in the calculations
exploring the effects of dispersion corrections. Only for catalyst
8e can we observe an increase in the predicted selectivity on
inclusion of dispersion corrections (Table 3). The proposed
dihedral angles of the biaryl bonds in the TSs for 8a and 8e
differ by less than 2°, suggesting that this is not the origin of the
observed selectivity differences in this series of catalysts.
Selectivity Predictions for KR of Alcohol 9 with
Isobutyric Anhydride Catalyzed by Catalysts with
Alternative ‘Blocking Groups’ (8d, 8f, and 8g). Having
established a correlation between computed and experimentally
determined selectivities for KR of alcohol 9 by known catalysts
8a−c and 8e, we were motivated to predict the expected
selectivities for the KR of alcohol 9 by a set of unknown
derivatives of catalyst 8a containing different substituents in the
phenyl ring, i.e., 3,5-dimethyl- (8d), 4-methoxy- (8f), and 4-
trifluoromethyl substitution (8g) (Chart 2). These derivatives
Scheme 3. Synthesis of Catalysts 8d, 8f, and 8g
Chart 2. New Chiral Pyridine Derivatives 8d, 8f, and 8g
Used To Model KR of Secondary Alcohols
a
a
All catalysts have (Sa)-configuration.
were selected to assess the roles of steric bulk, electron
donation and electron withdrawal, respectively, on the
effectiveness of this ‘blocking group’ within the catalyst
structure. The plan was to calculate the relevant enthalpy
differences ΔH(S − R) between diastereomeric TS2s for each
catalyst 8d, 8f, and 8g and then to prepare and test the new
catalysts experimentally.
In order to reduce the computational effort, it was considered
desirable not to carry out the full conformational search for
each new TS2 but rather to use the conformations obtained for
TS2 with the parent catalyst 8a as a basis in each case. The
envisaged variations of the catalyst structure were not
anticipated to dramatically change the conformational space
of TS2. Thus, taking the optimal conformations of TS2 for
catalyst 8a, introducing the modified aryl substituents, and then
reoptimizing the TSs at the B3LYP/6-31G(d) level would
avoid a lengthy conformational search. Detailed analysis of the
confomational space of TS2 (see Supporting Information)
shows that taking into account type I conformations for (R)-
TS2 and both types II and III for (S)-TS2 is necessary to find
the most stable conformations of TS2 (these are marked by
green boxes in Figure 4). Moreover, averaging over just the
three most stable conformations gives sufficiently accurate
enthalpy differences (see Supporting Information for details).
The initially prepared racemates were separated into their
atropisomeric enantiomers by semi-preparative chiral stationary
phase HPLC (CSP-HPLC) and the (Sa) enantiomers obtained
in >99.9% ee in all cases (for details see Supporting
Information). The absolute configuration of these atropisomers
was assigned by correlation of the Cotton effects in their
circular dichroism (CD) spectra to that of the parent catalyst
8a.9b Using these catalysts, a series of KR experiments using 1-
(1-naphthyl)ethanol (9) under conditions comparable to those
shown in Scheme 1 but using 1 equiv of isobutyric anhydride
were performed, and the levels of conversion and selectivity
determined by analytical CSP-HPLC (see Supporting In-
formation). The catalyst containing the 3,5-dimethyl sub-
stituted phenyl group, compound 8d, displayed a significantly
higher level of selectivity (s = 27) than the parent phenyl-
substituted catalyst 8a (s = 16), whereas both the 4-methoxy-
and 4-trifluoromethyl substituted catalysts 8f and 8g displayed
lower levels of selectivity (s = 11 and 9, respectively, Table 3).
Comparison between Experimental Selectivities and
Theoretical Predictions. With the experimentally measured
selectivities s for the new catalysts 8d, 8f, and 8g in hand, it is
possible to quantify the predictive value of the theoretically
calculated enthalpy differences ΔH298(S − R) (Table 3). The
9396
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
enthalpy differences ΔH298(S − R) calculated at the B3LYP/6-
31G(d) level can be correlated with experimental enantiose-
lectivities (Figure 5, red line, correlation coefficient R2 = 0.8792
for catalysts 8a, 8d, 8f, and 8g). Figure 5 shows that the
derivatives of catalyst 8a, substituted in the phenyl ring, have a
larger slope of the correlation line as compared to the catalysts
8a−c and 8e, which have different 4-dialkylamino substituents
indicating that the ‘blocking group’ has a more decisive role in
determining the selectivity of a catalyst than the 4-dialkylamino
substituents. The thermal corrections recalculated at 195 K do
not improve the correlation between experimental results and
calculated enthalpies or free energies. Inclusion of dispersion
corrections (at the B3LYP-D/6-311+G(d,p) level) slightly
improves the overall correlation of experimental selectivities
with calculated enthalpy differences for all studied catalysts 8a−
g (Table 3).
parent catalyst 8a (sH/sD = 1.92, ΔGD‑H(S − R) = 1.05 kJ
mol−1). On changing the phenyl substituent to the 3,5-
(CH3)2C6H3 substituent in the catalyst 8d the drop in
selectivity becomes smaller (sH/sD = 1.57, ΔGD‑H(S − R) =
0.74 kJ mol−1). The effect on the level of conversion (i.e., rate
of reaction) is marginal for both catalysts 8a and 8d (ΔC(H‑D)
<3%).
These experimental data imply that in all cases the rate-
limiting TSs become slightly less stabilized (i.e., higher in
energy and the overall rate is reduced) with deuterium
replacing hydrogen but that the effect is more pronounced
for the favored (R)-TS relative to the disfavored (S)-TS.
Moreover, the differential effect of this on the two competing
TSs is greatest for the parent catalyst 8a.
Reliable computational modeling of these intriguing KIEs
awaits experimental determination of detailed rate laws for the
individual proton transfer events so as to allow appropriate
combination of calculated KIEs for the corresponding TSs
depicted in Figures 2 and 3.
The experimental results are therefore consistent with the
expectation from theory that 3,5-disubstitution enhances the
enantioselectivity induced by the phenyl ‘blocking group’ in this
class of catalyst, whereas 4-substitution does not.
Experimental Deuterium Isotope Effects. Mechanistic
insight can often be provided by the study of deuterium isotope
effects both by experiment and computation.24 To this end, we
have performed a series of experiments using the achiral PPY
catalyst 11 and the enantiomerically pure catalysts 8a and 8d to
compare the isobutyrylation of alcohols 9 and d1-9 under
otherwise identical conditions. Since all the key TSs [i.e.,
TS1·9, TS2, TS3, and TS4 for isobutyrylation using PPY (11,
Figure 2) and (R)/(S)-TS1·9, (R)/(S)-TS2, (R)/(S)-TS3 for
isobutyrylation using 8a (Figure 3)] involve transfer of the
hydrogen atom which is initially part of the alcohol hydroxyl
group to the isobutyryl carboxylate, the introduction of a
deuterium atom as in d1-1-(1-naphthyl)ethanol (d1-9) would be
expected to significantly influence these reactions. The kinetic
isotope effect (KIE) for isobutyrylation of alcohols 9 and d1-9
catalyzed by PPY (11) was experimentally measured at −78 °C
(see Supporting Information), and kH/kD found to be 2.64,
which corresponds to a free energy difference of 1.58 kJ mol−1.
Similarly, the KIEs for isobutyrylation of alcohols 9 and d1-9
catalyzed by catalysts 8a and 8d were measured (Table 4).
The data reveal that changing from alcohol 9 to deuterated
alcohol d1-9 affects both the level of conversion and the level of
enantioselectivity achieved in these catalyzed KR processes.
The effect on enantioselectivity is most pronounced on the
CONCLUSIONS
■
The rate accelerations provided by 4-DMAP-type catalysts in
the acylation of alcohols by anhydrides are generally believed to
be the result of a nucleophilic catalysis mechanistic reaction
manifold. In this work, we have confirmed that the nucleophilic
mechanism is more favorable than the general base mechanism
for the reaction of 1-(1-naphthyl)ethanol (9) with isobutyric
anhydride catalyzed by the achiral PPY catalyst 11 using
computational methods (DFT at B3LYP/6-311+G(d,p)//
B3LYP/6-31G(d) level). We have also examined catalysis of
the same reaction when performed as a KR using the chiral 4-
DMAP catalyst (−)-(Sa)-8a. The TSs of the enantioselectivity-
determining step have been identified for each enantiomer of
the alcohol 9 and have been found to constitute four types in
which the two alcohol enantiomers prefer different directions of
approach to the key chiral acylpyridinium intermediate. The
key TS models have also been used to predict the levels of
enantioselectivity for two series of related catalysts differing in
the dialkylamino substituent (catalysts 8a−c and 8e) and the
aryl substitution pattern (catalysts 8a, 8d, 8f, and 8g),
respectively. The successful validation of the predictions for
highly selective catalyst 8d further supports the mechanistic
model chosen here; this and the trends with respect to the
influence of the dialkyl amino and aryl ‘blocking substituents’
on enantioselectivity are anticipated to provide a platform for
the future development of even more selective catalysts.23
The experimentally determined KIEs obtained in this work
demonstrate for the first time that alcohol deuteration has a
significant effect on the selectivity of these catalyzed alcohol KR
reactions, and understanding these effects computationally is
likely to provide additional important insight into the detailed
energetics of proton transfer in the rate/selectivity-determining
TSs of these processes. The current computational model
cannot explain these results for this complex multi-TS system,24
possibly due to inaccuracies in accounting for entropy25 and/or
handling tunneling effects26 and asymmetry in proton-transfer27
events. Work to overcome these limitations is planned and will
be reported in due course.
Table 4. Influence of H-Bonding on KR of Alcohol 9 by
Isobutyrylation Using Catalysts 8a and 8d
a
b
c
d
entry catalyst alcohol C, %
s
ΔG(S − R) ΔGD‑H (S − R)
1
2
3
4
(−)-8a
(−)-8a
(−)-8d
(+)-8d
9
27
24
39
36
16
8.3
27
17
4.48
3.43
5.32
1.05
d1-9
9
0.74
e
d1-9
4.59
a
b
Conversion C was calculated from HPLC. Selectivity factor s was
determined using Kagan’s equation.22 ΔG(S − R) (kJ mol−1) was
c
EXPERIMENTAL DETAILS
General Procedure for the Isobutyrylation of 1-(1-
naphthyl)ethanol (9) Catalyzed by PPY (11) (Scheme 2, Figure
1). A solution of ( )-1-(1-naphthyl)ethanol 9 (2 mmol), NEt3 (6
■
d
calculated using eq 1 at T = 195.15 K. ΔGD‑H(S − R) = ΔGD(S − R)
e
− ΔGH(S − R) (kJ mol−1). As the antipodal catalyst was used in this
experiment, the values are for ΔG(R − S) rather than ΔG(S − R).
9397
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
mmol), 1,3,5-trimethoxybenzene (0.33 mmol), and PPY (11, 0.5 mol
%) in toluene (8 mL) was cooled to −78 °C. (i-PrCO)2O (4 mmol)
was added dropwise with vigorous stirring. Every 10−30 min, an
aliquot of the reaction mixture (50 μL) was carefully removed and
quenched with MeOH (1 mL). The solvents were distilled off under
reduced pressure, and 1H NMR spectra were measured. The signals of
the ester 10 at δ 6.64 ppm and the alcohol 9 at δ 5.70 ppm were
integrated, and the conversion y is given by eq 2:
(quantitative yield). The enantiomeric excess for the unreacted alcohol
9 and the alcohol obtained by the ester saponification (10 → 9) was
established by analytical CSP HPLC using Chiralcel OD-H (0.46 cm
I.D. × 25 cm L; 5 μm silica). Mobile phase: n-hexane/2-propanol, 90/
10; flow rate: 1 mL min−1; temperature: 35 °C; sample conc. = 2 mg/
mL in 2-propanol; and injection: 1 or 2 μL). Each sample was analyzed
at least twice to obtain concordant ee values (see Supporting
Information).
Iester
Iester + IROH
ASSOCIATED CONTENT
* Supporting Information
Synthetic procedures, characterization data of new compounds,
procedures of the KR, computational data, and complete ref 15.
This material is available free of charge via the Internet at
http://pubs.acs.org.
y =
·100%
■
(2)
S
Dependence of the conversion y vs time t was fitted by eq 3 for
second-order reaction kinetics:
⎛
⎞
⎟
⎠
1
0
y = y 1 −
⎜
0
2ek(t−t ) − 1
⎝
(3)
(4)
AUTHOR INFORMATION
Corresponding Author
zipse@cup.uni-muenchen.de; a.c.spivey@imperial.ac.uk
■
k = k2[ROH]0
where k2 is a rate-constant of the second-order reaction and t0 has a
meaning of time axis offset. With this parameter in the fitting process,
it is not necessary to measure the starting point of the reaction exactly.
The variable y0 allows for rescaling of the conversion axis. The rate
constants were measured two times and then averaged (see Supporting
Information for full information).
Present Addresses
§MRC Laboratory of Molecular Biology, Hills Road, Cam-
bridge, CB2 0QH, United Kingdom.
∥Department of Organic Chemistry, University of Geneva,
Quai Ernest Ansermet 30, Geneva, Switzerland.
⊥State Key Laboratory of Organometallic Chemistry, Shanghai
Institute of Organic Chemistry, Chinese Academy of Sciences,
345 Lingling Road, Shanghai 200032 China.
General Procedure for the KR of 1-(1-naphthyl)ethanol (9)
Using Isobutyric Anhydride Catalyzed by Chiral-DMAP
Catalysts 8a, 8d, 8f, and 8g (Table 3). All KR experiments were
run in duplicate. An oven-dried microwave glass vial (0.5−2.0 mL
capacity, 1.25 cm I.D. × 8.0 cm L, Biotage Ltd.) equipped with a
Teflon magnetic stirring bar (Biotage Ltd.) was charged with (−)-{3-
[2-(3,5-dimethylphenyl)naphthalen-1-yl]pyridin-4-yl}diethylamine 8d
(0.0038 g, 0.01 mmol, 0.01 equiv) and 1-(1-naphthyl)ethanol 9
(0.1722 g, 1.0 mmol, 1.0 equiv). The vial was sealed with an airtight
aluminum/rubber septum (Reseal design, Biotage Ltd.) using a
crimper. The contents in the vial were dried in vacuo and purged with
argon gas ( ×3). Dry toluene (2 mL) and dry NEt3 (0.105 mL, 0.75
mmol, 0.75 equiv) were added to the vial under an argon atmosphere,
and the mixture was stirred on an isopropanol bath which was
maintained at a constant −78 °C using a cryostat. After 30 min,
isobutyric anhydride (0.166 mL, 1.0 mmol, 1.0 equiv) was added to
the reaction mixture under an argon atmosphere, and the mixture was
stirred (∼1000 rpm) at −78 °C. After 2 h, 1 mL of the reaction
mixture was drawn out using a syringe and rapidly transferred into a
sealed vial (Biotage as described above) containing a magnetic stirring
bar and methanol (2 mL/0.5 mmol of substrate) that has been
precooled to −78 °C for at least 1 h under an argon atmosphere. The
remainder of the reaction mixture was continued to be stirred at −78
°C for an additional 6 h (total time = 8 h), after which it was quenched
with cold methanol (2 mL) that was precooled at −78 °C as described
above. Each of the reaction mixtures were then stirred and allowed to
warm to r.t. overnight [Note: Abrupt warming of the reaction mixture
during the quenching process must be avoided]. After the quenching
process was complete, each of these reaction mixtures was quickly
evaporated to dryness under reduced pressure, and the crude material
was subjected to flash chromatography on SiO2 [eluent: CH2Cl2/n-
hexane (1:1) then CH2Cl2] to give ester 10 (0.018 g, 15% for 2 h
reaction; 0.046 g, 38% for 8 h reaction) as a colorless oil and unreacted
alcohol 9 [0.072 g, 84% (recovered isolated yield) for the 2 h reaction;
0.052 g, 60% (recovered isolated yield) for the 8 h reaction] as a white
solid. The spectroscopic and analytical data of the unreacted alcohol 9
and ester 10 were consistent with those reported in the literature.9a
Analysis of the enantiomeric purity of the alcohol 9 was performed
directly, whereas that of the ester 10 was performed after hydrolysis to
the alcohol 9, as described below.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Synthetic aspects of the work described in this manuscript were
supported by a grant from the EPSRC (GR/S04284/01).
Theoretical and mechanistic aspects have been supported by a
grant of the Deutsche Forschungsgemeinschaft (DFG) in the
focus program SPP 1179 “Organocatalysis” (ZI 439/10).
REFERENCES
■
(1) (a) Hofle, G.; Steglich, W.; Vorbruggen, H. Angew. Chem. 1978,
̈
̈
90, 602; Angew. Chem., Int. Ed. Engl. 1978, 17, 569. (b) Spivey, A. C.;
Arseniyadis, S. Angew. Chem. 2004, 116, 5552; Angew. Chem., Int. Ed.
Engl. 2004, 43, 5436.
(2) (a) Vedejs, E.; Jure, M. Angew. Chem., Int. Ed. 2005, 44, 3974.
(b) Wurz, R. Chem. Rev. 2007, 107, 5570. (c) Spivey, A. C.; McDaid,
P. Asymmetric Acyl Transfer Reactions. In Handbook of Asymmetric
Organocatalysis; Dalko, P., Ed.; Wiley-VCH, Weinheim, Germany,
2007, pp 287−329. (d) Spivey, A. C.; Arseniyadis, S. Top. Curr. Chem.
2010, 291, 233. (e) Muller, C. E; Schreiner, P. R. Angew. Chem., Int.
̈
Ed. 2011, 50, 6012.
(3) (a) Vedejs, E.; Chen, X. J. Am. Chem. Soc. 1996, 118, 1809.
(b) Vedejs, E.; Chen, X. J. Am. Chem. Soc. 1997, 119, 2584. (c) Duffey,
T. A.; MacKay, J. A.; Vedejs, E. J. Org. Chem. 2010, 75, 4674.
(4) (a) Kawabata, T.; Nagato, M.; Takasu, K.; Fuji, K. J. Am. Chem.
Soc. 1997, 119, 3169. (b) Kawabata, T.; Yamamoto, K.; Momose, Y.;
Yoshida, H.; Nagaoka, Y.; Fuji, K. Chem. Commun. 2001, 2700.
(c) Kawabata, T.; Stragies, R.; Fukaya, T.; Nagaoka, Y.; Schedel, H.;
Fuji, K. Tetrahedron Lett. 2003, 44, 1545. (d) Kawabata, T.; Stragies,
R.; Fukaya, T.; Fuji, K. Chirality 2003, 15, 71−76. (e) Muramatsu, W.;
Kawabata, T. Tetrahedron Lett. 2007, 48, 5031. (f) Kawabata, T.;
Muramatsu, W.; Nishio, T.; Shibata, T.; Uruno, Y.; Stragies, R.
Synthesis 2008, 747.
Saponification of Ester 10. The ester 10 obtained from each KR
experiment was hydrolyzed by heating to reflux in 5% NaOH/MeOH
(2 mL for 0.5 mmol substrate) for 5 min. After evaporation of the
solvent, the residue was subject to rapid flash chromatography (SiO2)
eluting with CH2Cl2 to give the hydrolyzed alcohol 9 as a white solid
(5) (a) Ruble, J. C.; Fu, G. C. J. Am. Chem. Soc. 1998, 120, 11532.
(b) Tao, B.; Ruble, J. C.; Hoic, D. A.; Fu, G. C. J. Am. Chem. Soc. 1999,
121, 5091. (c) Fu, G. C. Acc. Chem. Res. 2000, 33, 412. (d) Fu, G. C.;
Ie, Y. Chem. Commun. 2000, 119. (e) Arai, S.; Bellemin-Laponnaz, S.;
9398
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399

Journal of the American Chemical Society
Article
Fu, G. C. Angew. Chem., Int. Ed. 2001, 40, 234. (f) Fu, G. C. Acc. Chem.
Res. 2004, 37, 542. (g) Arp, F. O.; Fu, G. C. J. Am. Chem. Soc. 2006,
128, 14264. (h) Wurz, R. P.; Lee, E. C.; Ruble, J. C.; Fu, G. C. Adv.
Synth. Cat. 2007, 349, 2345.
entries in Scheme 1) and so is not directly comparable to the others
listed.
(11) Wei, Y.; Held, I.; Zipse, H. Org. Biomol. Chem. 2006, 4, 4223.
(12) (a) Shinisha, C. B.; Sunoj, R. B. Org. Lett. 2009, 11, 3242.
(b) Muller, C. E.; Wanka, L.; Jewell, K.; Schreiner, P. R. Angew. Chem.,
(6) (a) Naraku, G.; Shimomoto, N.; Hanamoto, T.; Inanaga, J.
Enantiomer 2000, 5, 135. (b) Kotsuki, H.; Sakai, H.; Shinohara, T.
Synlett 2000, 116. (c) Priem, G.; Anson, M. S.; Macdonald, S. J. F.;
Pelotier, B.; Campbell, I. B. Tetrahedron Lett. 2002, 43, 6001.
(d) Jeong, K.-S.; Kim, S.-H.; Park, H.-J.; Chang, K.-J.; Kim, K. S. Chem.
Lett. 2002, 31, 1114. (e) Priem, G.; Pelotier, B.; Macdonald, S. J. F.;
Anson, M. S.; Campbell, I. B. J. Org. Chem. 2003, 68, 3844.
(f) Pelotier, B.; Priem, G.; Campbell, I. B.; Macdonald, S. J. F.; Anson,
M. S. Synlett 2003, 679. (g) Pelotier, B.; Priem, G.; Macdonald, S. J. F.;
Anson, M. S.; Upton, R. J.; Campbell, I. B. Tetrahedron Lett. 2005, 46,
9005. (h) Yamada, S.; Misono, T.; Iwai, Y. Tetrahedron Lett. 2005, 46,
2239. (i) Dalaigh, C. O.; Hynes, S. J.; Maher, D. J.; Connon, S. J. Org.
Biomol. Chem. 2005, 3, 981. (j) Diez, D.; Gil, M. J.; Moro, R. F.;
Garrido, N. M.; Marcos, I. S.; Basabe, P.; Sanz, F.; Broughton, H. B.;
Urones, J. G. Tetrahedron: Asymm. 2005, 16, 2980. (k) Poisson, T.;
̈
Int. Ed. 2008, 47, 6180.
(13) (a) Jorgensen, W. L.; Tirado-Rives, J. J. Comput. Chem. 2005,
26, 1689. (b) Jorgensen, W. L.; Maxwell, D. A.; Tirado-Rives, J. J. Am.
Chem. Soc. 1996, 118, 11225.
(14) Xu, S.; Held, I.; Kempf, B.; Mayr, H.; Steglich, W.; Zipse, H.
Chem.Eur. J. 2005, 11, 4751.
(15) Frisch, M. J., et al. Gaussian 03, revision D.01; Gaussian, Inc.:
Pittsburgh, PA, 2004.
(16) (a) Grimme, S. J. Comput. Chem. 2006, 27, 1787. (b) Grimme,
S. J. Comput. Chem. 2004, 25, 1463−1473.
(17) Neese, F. ORCA 2.6.4, an ab initio density functional and
semiempirical program package, 2007.
(18) (a) Satyanarayana, T.; Kagan, H. B. Tetrahedron 2007, 63, 6415.
(b) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. Adv. Synth. Catal. 2001,
343, 5.
Penhoat, M.; Papamicael, C.; Dupas, G.; Dalla, V.; Marsais, F.;
̈
(19) Vedejs, E.; Daugulis, O.; Harper, L. A.; MacKay, J. A.; Powell, D.
R. J. Org. Chem. 2003, 68, 5020.
Levacher, V. Synlett 2005, 2285. (l) Yamada, S.; Misono, T.; Iwai, Y.;
Masumizu, A.; Akiyama, Y. J. Org. Chem. 2006, 71, 6872. (m) Dalaigh,
C. O.; Hynes, S. J.; O’Brien, J. E.; McCabe, T.; Maher, D. J.; Watson,
G. W.; Connon, S. J. Org. Biomol. Chem. 2006, 4, 2785. (n) Yamada,
(20) For discussion of activation parameters for nucleophilic catalysis
in acyl-transfer reactions, see: (a) Oakenfull, D. G.; Riley, T.; Gold, V.
Chem. Commun. 1966, 385. (b) Milstien, J. B.; Fife, T. H. J. Am. Chem.
Soc. 1968, 90, 2164. (c) Schowen, R. L.; Behn, C. G. J. Am. Chem. Soc.
1968, 90, 5839. (d) Butler, A. R.; Robertson, I. H. J. Chem. Soc., Perkin
Trans. 2 1975, 660.
̌
S.; Yamashita, K. Tetrahedron Lett. 2008, 49, 32. (o) Sam
́ ̌
al, M.; Mísek,
J.; Stara, I. G.; Stary, I. Collect. Czech. Chem. Commun. 2009, 74, 1151.
(p) Crittall, M. R.; Rzepa, H. S.; Carbery, D. R. Org. Lett. 2011, 13,
1250.
́
́
(21) Bonduelle, C.; Martín-Vaca, B.; Cossío, F. P.; Bourissou, D.
(7) (a) Birman, V. B.; Uffman, E. W.; Jiang, H.; Li, X.; Kilbane, C. J. J.
Am. Chem. Soc. 2004, 126, 12226. (b) Birman, V. B.; Jiang, H. Org.
Lett. 2005, 7, 3445. (c) Birman, V. B.; Li, X.; Jiang, H.; Uffman, E. W.
Tetrahedron 2006, 62, 285. (d) Birman, V. B.; Li, X. Org. Lett. 2006, 8,
1351. (e) Birman, V. B.; Guo, L. Org. Lett. 2006, 8, 4859. (f) Birman,
V. B.; Jiang, H.; Li, X. Org. Lett. 2007, 9, 3237. (g) Birman, V. B.; Li, X.
Org. Lett. 2008, 10, 1115. (h) Li, X.; Liu, P.; Houk, K. N.; Birman, V.
B. J. Am. Chem. Soc. 2008, 130, 13836. (i) Li, X.; Jiang, H.; Uffman, E.
W.; Guo, L.; Zhang, Y.; Yang, X.; Birman, V. B. J. Org. Chem. 2012, 77,
1722.
(8) (a) Shiina, I.; Nakata, K. Tetrahedron Lett. 2007, 48, 8314.
(b) Shiina, I.; Nakata, K.; Onda, Y.-s. Eur. J. Org. Chem. 2008, 5887.
(c) Xu, Q.; Zhou, H.; Geng, X.; Chen, P. Tetrahedron 2009, 65, 2232.
(d) Shiina, I.; Nakata, K.; Sugimoto, M.; Onda, Y.-s.; Iizumi, T.; Ono,
K. Heterocycles 2009, 77, 801. (e) Shiina, I.; Nakata, K.; Ono, K.;
Sugimoto, M.; Sekiguchi, A. Chem.Eur. J. 2010, 16, 167. (f) Shiina,
I.; Nakata, K.; Ono, K.; Onda, Y.-s.; Itagaki, M. J. Am. Chem. Soc. 2010,
132, 11629. (g) Hu, B.; Meng, M.; Wang, Z.; Du, W.; Fossey, J. S.; Hu,
X.; Deng, W.-P. J. Am. Chem. Soc. 2010, 132, 17041. (h) Wagner, A. J.;
David, J. G.; Rychnovsky, S. D. Org. Lett. 2011, 13, 4470.
(i) Belmessieri, D.; Joannesse, C.; Woods, P. A.; MacGregor, C.;
Jones, C.; Campbell, C. D.; Johnston, C. P.; Duguet, N.; Concellon,
C.; Bragg, R. A.; Smith, A. D. Org. Biomol. Chem. 2011, 9, 559.
(j) Viswambharan, B.; Okimura, T.; Suzuki, S.; Okamoto, S. J. Org.
Chem. 2011, 76, 6678.
Chem.Eur. J. 2008, 14, 5304.
(22) Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249.
(23) From the trends established herein with respect to the influence
of the 4-dialkylamino and aryl blocking groups on catalyst selectivity, it
would be anticipated that the catalyst combining both 4-NBu2 and 3,5-
(CH3)2C6H3 groups would exhibit high levels of selectivity. This is
indeed the case. For the KR of alcohol 9 under conditions identical to
those used in Table 3 for catalysts 8d, 8f, and 8g [i.e., 8 h, −78 °C,
toluene, 1 equiv (i-PrCO)2O, 0.75 equiv Et3N, 1 mol% cat], this
catalyst displays a s value of 44 (C = 48.2%). The synthesis and scope
of this catalyst for the KR of various alcohols will be the subject of a
forthcoming communication.
(24) Giagou, T.; Meyer, M. P. J. Org. Chem. 2010, 75, 8088.
(25) O’Leary, D. J.; Hickstein, D. D.; Hansen, B. K. V.; Hansen, P. E.
J. Org. Chem. 2010, 75, 1331.
(26) Nagel, Z. D.; Klinman, J. P. Chem. Rev. 2010, 110, P45.
(27) Stoyanov, E. S.; Stoyanova, I. V.; Reed, C. A. J. Am. Chem. Soc.
2010, 132, 1484.
(9) (a) Spivey, A.; Fekner, T.; Spey, S. E. J. Org. Chem. 2000, 65,
3154. (b) Spivey, A. C.; Zhu, F.; Mitchell, M. B.; Davey, S. G.; Jarvest,
R. L. J. Org. Chem. 2003, 68, 7379. (c) Spivey, A. C.; Leese, D. P.; Zhu,
F.; Davey, S. G.; Jarvest, R. L. Tetrahedron 2004, 60, 4513. (d) Spivey,
A. C.; Arseniyadis, S.; Fekner, T.; Maddaford, A.; Leese, D. P.
Tetrahedron 2006, 62, 295.
(10) Upon reinvestigation in this work, the selectivity factor s = 24
reported in refs 9a and 9c for catalyst (−)-(Sa)-8a in the KR
corresponding to Scheme 1 (and the value s = 29 for the analogous
reaction using just 1 equiv of anhydride, see ref 9a) was found to be
incorrect as the result of HPLC integration errors arising from an
impurity co-running with one enantiomer of the alcohol 9. The correct
values should be s = 16 for both cases. Additionally, the selectivity
factor s = 9 reported in ref 9c for catalyst (−)-(Sa)-8i was a typo and
should have read s = 39; this value was however obtained in a KR
experiment run at −93 °C over 14.3 h (cf. −78 °C and 9 h for all other
9399
dx.doi.org/10.1021/ja302420g | J. Am. Chem. Soc. 2012, 134, 9390−9399