A mechanistic investigation of the kinetic resolution of secondary aromatic alcohols using a ferrocene-based planar chiral 4-(dimethylamino)pyridine catalyst

Article abstract of DOI:10.1002/chem.201405793

A detailed computational and kinetic analysis of the acetylation of 1-phenylethanol with acetic anhydride catalyzed by planar chiral 4-(dimethylamino)pyridine (DMAP) catalyst (-)-1 is presented. The study includes a computational investigation of the pote

Full text of DOI:10.1002/chem.201405793

DOI: 10.1002/chem.201405793
Full Paper
&
Kinetic Resolution
A Mechanistic Investigation of the Kinetic Resolution of
Secondary Aromatic Alcohols Using a Ferrocene-Based Planar
Chiral 4-(Dimethylamino)pyridine Catalyst
[b]
[a, b]
Laura Mesas-Sꢀnchez and Peter Dinꢁr*
Abstract: A detailed computational and kinetic analysis of
the acetylation of 1-phenylethanol with acetic anhydride
catalyzed by planar chiral 4-(dimethylamino)pyridine (DMAP)
catalyst (À)-1 is presented. The study includes a computa-
tional investigation of the potential-energy surface including
the acylation and stereoselective transition states at the DFT
level of theory. Experimentally, the kinetic study shows that
the reaction proceeds in a first-order manner in catalyst,
whereas both substrates, acetic anhydride and 1-phenyletha-
nol, show fractional order, which is in accordance with
steady-state conditions. The fractional order depends on an
equilibrium between the free catalyst and the acetylated
catalyst.
Introduction
Since the discovery of the nucleophilic catalyst 4-(dimethyl-
[
1]
amino)pyridine (DMAP) by Steglich and Hçfle and Litvinenko
[
2]
and Kirichenko, DMAP has been used as a versatile catalyst in
a variety of group-transfer reactions, for example, the acylation
[
3–5]
of alcohols and amines.
The currently accepted mecha-
[
3,6–8]
nism
for acylation reactions of alcohols involves an attack
of the electrophile from the nucleophilic catalyst to form an
acylium cation with the corresponding counteranion (acetate
in the case of acetic anhydride) in a pre-equilibrium. The
alcohol then reacts with the acylated catalyst in the rate-
determining second step and transfers the acyl group onto the
alcohol to form the ester product together with the deactivat-
ed (protonated) catalyst. The DMAP catalyst is presumably re-
generated by means of deprotonation of the auxiliary base
triethylamine, which is not involved in the rate-limiting step
Scheme 1. Schematic representation of the DMAP-catalyzed acetylation of
alcohol.
(
Scheme 1).
[
11]
Since the early use of DMAP as an efficient catalyst in the
resolutions of 1-arylethanols. The use of this derivative in
stoichiometric amounts in the presence of two equivalents of
a Lewis acid led to enantiomerically enriched products with
acylation of alcohols, several chiral DMAP catalysts have been
developed for the synthesis of enantiomerically pure alcohols
by means of kinetic resolution, which has become an impor-
[11,12]
a high enantiomeric excess (ee).
oped the synthesis of planar chiral ferrocenyl pyrrole/DMAP
Fu and co-workers devel-
[
9,10]
tant field of organic synthesis.
Early attempts by Vedejs and
[13]
Chen introduced a chiral DMAP derivative into the kinetic
analogues. Planar chiral DMAP derivative (À)-1 catalyzed the
kinetic resolution of secondary aryl and allylic alcohols with
excellent selectivity to yield the remaining alcohol in high
[a] Dr. P. Dinꢀr
Department of Chemistry
KTH-Royal Institute of Technology
Teknikringen 30, 10044 Stockholm (Sweden)
Fax: (+)46-8-7912333
[14]
enantiopurity. In addition, Fu and co-workers also showed
that the catalyst was efficient in the kinetic resolution of a
[14]
[15]
racemic trans-diol, the desymmetrization of meso-diols
and propargylic alcohols, and secondary 1-arylamines and
[16]
[17]
E-mail: diner@kth.se
[18]
[
b] L. Mesas-Sꢁnchez, Dr. P. Dinꢀr
Department of Chemistry-BMC
Uppsala University, Box 576
indolines. Moreover, Fu et al. expanded the use of planar
chiral DMAP catalyst (À)-1 by the development of the first
non-enzymatic acetylation of secondary alcohols in combina-
7
5123, Uppsala (Sweden)
[19]
tion with a ruthenium-based racemization catalyst.
Since
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405793.
those early developments, the kinetic resolution of secondary
Chem. Eur. J. 2015, 21, 1 – 10
1
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
of alcohols between the planar chiral DMAP catalyst (À)-1 with
previous computational investigations of DMAP by Zipse and
co-workers, initially several different transition states and inter-
mediates for the acetylation of tert-butanol were optimized at
the B3LYP level of theory using the 6-31G(d,p) basis set for C,
H, O, and N and an ECP basis set (LANL2DZ) for Fe, followed
by a single-point calculation at the B3LYP/6-311+G(d,p) level
of theory (for C, H, O, and N). The nucleophilic reaction path-
way involves the formation of the weakly coordinated complex
À1
I1 (À2.9 kcalmol ) between catalyst (À)-1, tert-butanol, and
acetic anhydride, which then react to form the acetylated cata-
lyst and the acetate ion through transition state TS1 (activation
À1
enthalpy: 10.8 kcalmol ). In the transition state, the hydroxy
group of the tert-butanol hydrogen bonds to the acetic anhy-
dride. Following the intrinsic reaction coordinate (IRC) in the
forward direction of TS1 leads to an intermediate that consists
of an acylium cation with the acetate as the corresponding
Figure 1. Examples of nucleophilic catalysts for the acylative kinetic
resolution of secondary aryl alkyl alcohols.
À1
counteranion (I2; 3.6 kcalmol ). A comparison of transition
state TS1 with the previously calculated transition state for
[7]
DMAP reveals that TS1 is an earlier transition state than that
in the DMAP case, which is manifested in the longer CÀN
bond-forming distance (1.795 compared to 1.719 ꢃ) between
the nitrogen and the carbonyl carbon and a shorter CÀO
alcohols has been achieved by using different chiral nucleo-
[
20]
philic catalysts (Figure 1).
Currently, only a few theoretical studies on rate-limiting tran-
sition states have been reported for these kinetic resolutions.
These cases include the computational study of the enantio-
selective acetylation of trans-cyclohexane-1,2-diol catalyzed by
[24]
bond-breaking distance (1.613 compared to 1.717 ꢃ).
The
reaction proceeds without the formation of a tetrahedral
intermediate, which is similar to previous computational inves-
[
21]
[7]
a N-methylimidazole-based peptide by Sunoj et al.; the in-
vestigation of the transition-state structures for the acylation
of 1-phenylethanol catalyzed by amidine by Houk and Birman
tigations of the DMAP-catalyzed reaction by Zipse et al. In
contrast to DMAP, two different isomers of the acylium cation
of catalyst (À)-1 can be formed with the methyl group either
pointing towards or away from the cyclopentadiene ring of
the ferrocene (Figure 2). Fu and co-workers have determined
[
22]
et al.; and the investigation of the rate-determining states of
[
23]
a chiral DMAP catalyst by Spivey and Zipse et al.
Owing to these recent reports on the theoretical studies of
the rate-limiting transitions states of the acylation reaction, we
were interested in improving the understanding of the rate-
determining transition states of acyl-transfer reactions using
planar chiral DMAP catalyst (À)-1. Therefore, we have
performed a computational study of the acetylation of the
(
(
R)- and (S)-1-phenylethanol catalyzed by the DMAP derivative
À)-1 using DFT. The key question here was to determine the
rate-determining states of the reaction and explore if the
enantioselectivities of chiral DMAP-catalyzed acyl-transfer
reactions could be rationalized by examination of the transi-
tion state (TS) of the reaction. In addition, we also analyzed
the kinetics of the reaction by using reaction progress kinetic
analysis (RPKA) and performed an NMR spectroscopic study of
the equilibrium between the catalyst and the acetylating
agent.
Figure 2. A) X-ray structure of Fu and co-workers’ acetylated catalyst with
À
SbF
6
as the counter ion (two molecules of tetrahydrofuran have been omit-
[16]
ted for clarity). B) Optimized structure of the most stable conformer of the
acylium cation at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p) level of theory.
C) Optimized structure of the less stable 1808-flipped conformer of the
acylium cation at the B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) level of theory.
Results and Discussion
the structure of a complex between the acylium cation of cata-
À
lyst (À)-1 together with SbF as a counterion by using X-ray
6
Computational studies
[16]
crystallography. In the isolated compound, the acetyl methyl
group points towards the cyclopentadiene ring in the most
stable acylium cation (Figure 2A). This was also confirmed by
NMR spectroscopy in which nuclear Overhauser effect (NOE)
was observed between the methyl protons and the C2 proton
In the computational study, we wanted to investigate both the
transition states for the nucleophilic reaction between the
acetic anhydride and catalyst (À)-1 (TS1) and the transition
states involved in the acetylation of the alcohol (TS2). To com-
pare the potential-energy surface of the nucleophilic acylation
[16]
of the cyclopentadiene ring.
At the B3LYP/6-311+G(d,p)//
&
&
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
2
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
B3LYP/6-31G(d,p) level of theory, the energy difference be-
tween the two positively charged acylium cations (Figure 2B,C)
formation (TS1). A comparison of the free energies of the two
transition states also suggests that the acetylation of tert-buta-
nol is the highest-energy transition state of the acetylation re-
À1
is 2.0 kcalmol , which is in agreement with the experimentally
À1
observed X-ray structure.
action (34.6 and 39.8 kcalmol for TS1 and TS2b, respective-
Following the intrinsic reaction coordinates from either TS1
which leads to I2) or from the acetylation transition states
TS2a–d (which lead to intermediates I2a–d) generate inter-
mediates in which the counteranion and the tert-butanol are
also included.
ly). In previous computational studies, the nucleophilic path-
way for acetylation of alcohols using pyridine-based catalysts,
such as DMAP, 4-pyrrolidinopyridine (PPY), and a chiral DMAP
catalyst was found to be favored over the base-catalyzed path-
(
[7,23,25]
way.
The transition state (TS3) of the concerted base-cat-
For all these intermediates, the conformers with a carbonyl
oxygen pointing towards the cyclopentadiene ring (I2, I2a,
and I2b) are more stable than the conformers with a 1808 flip
of the acetyl group (I2c and I2d; see the Supporting
Information), which is in agreement with the experimentally
observed acylium cation.
alyzed pathway for the acetylation of tert-butanol using
À1
(À)-1 was higher in enthalpy (20.2 kcalmol ) than TS2b (12.2
À1
kcalmol ), which suggests that the nucleophilic pathway is
favored over the base-catalyzed pathway (see the Supporting
Information).
Next we wanted to investigate the transition states of the ki-
netic resolution of the (R)- and the (S)-1-phenylethanol using
(À)-1 and acetic anhydride. Initially, tert-butanol was ex-
changed for 1-phenylethanol and the transition states that
lead to R and S products were optimized at the B3LYP/6-
31G(d,p) level of theory, followed by a single-point calculation
at the B3LYP/6-311+G(d,p) level of theory. The enthalpy of acti-
vation for the lowest transition state for the acetylation of the
The next step was to determine which of the acylium
cations that are kinetically favored in the acetylation of the
tert-butanol. Several different transition states (TS2a–d) were
calculated for the addition of tert-butanol to the acylium
cation structures. The transition states for the nucleophilic
pathway proceed by means of a concerted transition state in
which the acetate ion deprotonates the hydroxy group of the
alcohol simultaneously as the alcohol oxygen attacks the
acetyl moiety and breaks the CÀN bond. The most stable tran-
sition state (TS2b) for the acetylation of tert-butanol is similar
to the energetically favored transition state for the DMAP-cata-
lyzed reaction previously investigated by Zipse et al. (see
À1
(R)-1-phenylethanol decreases (5.9 kcalmol ) relative to tert-
butanol, and the activation enthalpy is in fact lower than the
transition state for the acetylation of the catalyst (TS1;
À1
10.8 kcalmol ; Figure 4). This suggests that the formation of
the acylium cation intermediate is the rate-determining state
of the reaction.
[
7]
Figure 3). The activation enthalpy of the most stable transi-
À1
tion state (TS2b) is 12.2 kcalmol and suggests that the transi-
The same relationship is seen for the Gibbs free energies of
the two different transition states with a free-energy difference
tion state for the acetylation of the alcohol (TS2a) is slightly
higher in energy than the transition state for the acetyl cation
À1
of 3.5 kcalmol in favor of (R)-TS2-1 relative to TS1 (see
Figure 4). An investigation of
the transition states for the
base-catalyzed reaction (TS3a
and TS3b) of (R)-1-phenyletha-
nol using (À)-1 show that the
nucleophilic pathway is favored
À1
with 7.2 kcalmol
over the
base-catalyzed pathway when
considering the relative Gibbs
free energy (see the Supporting
Information). The different se-
lectivity-determining transition
states that lead to the R and
the S product were investigat-
ed. Manual rotation of the CÀC
single bond and CÀO bond of
1-phenylethanol and rotation
around the O-H-O-C bonds be-
tween the 1-phenylethanol and
the acetate ion were generated
from several different starting
structures. In total, 320 starting
structures (160 for each enan-
tiomer) were generated and op-
timized at the B3LYP/6-31G(d,p)
level of theory with a constraint
À1
Figure 3. Gas-phase enthalpy profile (DH298K) [kcalmol ] for the nucleophilic mechanisms in the DMAP-catalyzed
reaction of acetic anhydride with tert-butanol at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p) level of theory. The
Gibbs free energies in (DDG298K) [kcalmol ] are given in parentheses.
À1
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
3
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 4. Gas-phase Gibbs free energy profile (DG298K) for the nucleophilic mechanisms in the planar DMAP-catalyzed reaction of acetic anhydride with
À1
(
R)- and (S)-1-phenylethanol at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p) level of theory. The enthalpies in (DDH298K) [kcalmol ] are given in parentheses.
on the CÀO bond that forms between the alcohol and the
acetyl group. The unique structures from the constrained opti-
mization were then optimized at the B3LYP/6-31G(d,p) level of
theory, thus leading to several unique transition states for the
formation of R and S product (18 and 20 transition states, re-
spectively). Since the catalyst (À)-1 includes several aromatic
rings, and in addition p–p interactions between the catalyst
and the substrate could influence the selectivity of the kinetic
resolution, dispersion interactions might play an important
role in describing the potential-energy surface correctly. There-
fore, the transition states of the acetylation of (R)- and (S)-1-
phenylethanol were also optimized using the Grimme correc-
tion for dispersion (GD3) at the B3LYP/6-31G(d,p)-GD3 level of
theory. Lastly, we also wanted to investigate the effect of solva-
tion on these transition states.To mimic the effects of tert-amyl
alcohol, we optimized the most stable transition states using
the polarizable conductor calculation model (CPCM, UFF, 2-
methyl-2-propanol) and tert-butanol as the solvent in combina-
tion with the Grimme correction for dispersion (B3LYP/6-
ly reported selectivity factors for this reaction (s=27 (1.95 kcal
À1
À1
mol ) at room temperature; s=44 (2.05 kcalmol ) at
[
15,26]
08C).
The energy difference increases slightly upon inclu-
°
À1
sion of dispersion (DDG =3.8 kcalmol ), and the same trend
is seen if the polarizable conductor calculation model (CPCM,
UFF, 2-methyl-2-propanol) is included in the optimization
°
À1
(DDG =4.1 kcalmol ). This increase is probably due to an ad-
ditional parallel offset p–p interaction between the phenyl ring
of the substrate and the acetylated pyridine ring of the cata-
À1
Table 1. Relative enthalpies and free energies [kcalmol ] for the lowest
transition states leading to the R and S product at different levels of
theory.
[a]
Method
DE [0 K]
DH [298 K]
DG [298 K]
Gas phase
B3LYP/6-31G(d,p)
B3LYP/6-311+G(d,p)
1.63
2.08
1.75
2.20
2.56
3.01
[a]
Gas phase/dispersion (GD3)
B3LYP/6-31G(d,p)-GD3
B3LYP/6-311+G(d,p)-GD3
31G(d,p)-GD3-UFF). The most stable transition state at all levels
4.47
3.77
4.51
3.82
3.80
3.11
of theory, (R)-TS2-1, leads to the R product given the absolute
energy, enthalpy, and the Gibbs free energy. The gas-phase ab-
solute energy and Gibbs free-energy differences between the
[
b]
Dispersion (GD3)/CPCM (UFF)
B3LYP/6-31G(d,p)-GD3-UFF
B3LYP/6-311+G(d,p)-GD3-UFF
6.29
5.94
5.93
5.59
4.42
4.07
most stable transition states for the kinetic resolution of the
[
c]
À1
(
R)- and (S)-alcohols varies from 1.63 kcalmol at the B3LYP/6-
À1
[
[
[
a] B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p).
b] B3LYP/6-311+G(d,p)-GD3//B3LYP/6-31G(d,p)-GD3.
c] B3LYP/6-311+G(d,p)-GD3-UFF//B3LYP/6-31G(d,p)-GD3-UFF.
3
1G(d,p) level of theory to 3.01 kcalmol at the B3LYP/6-
311+G(d,p) level of theory, respectively (Table 1). The energy
differences are in qualitative agreement with the experimental-
&
&
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
4
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
stant “excess” between the two reactants during the progress
of the reaction (here the “excess” refers to the difference in ini-
tial concentration of the reactants). The rates of the reaction at
different and the “same excess” of the substrates were then
plotted versus the substrate concentration to obtain mechanis-
tic information, such as reaction orders of the catalyst and the
substrates in the reaction. The benefit of using this technique
is that the reaction is monitored for the “complete” reaction
(
typically 15–85%) and allows the detection of changes of the
mechanism, which is not the case for the typical initial rate ki-
netic measurements. In RPKA, the reaction can be monitored
under the “true” reaction conditions, which is often not the
case for pseudo-first-order kinetic measurements in which one
of the reactants is present in a large excess amount. Addition-
ally, RPKA also allows for the detection of catalyst decomposi-
tion and product inhibition during the course of the reaction.
Since the kinetic resolution catalyzed by (À)-1 preferably
acetylates the (R)-alcohol, we investigated the reaction
between (R)-1-phenylethanol and acetic anhydride at 178C
(
Scheme 2).
Scheme 2. Schematic representation of the (À)-1-catalyzed acetylation of
(
R)-1-phenylethanol.
Figure 5. Optimized geometries and relative Gibbs free energies (DDG298K
)
The reaction progress was monitored using ReactIR, and the
À1
[
kcalmol ] for the acetylation reaction between (À)-1 and acetic anhydride
measured data were calibrated with the conversion of the
at different levels of theory. A) B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p).
1
reaction as measured by means of H NMR spectroscopy to
B) B3LYP/6-311+G(d,p)-GD3//B3LYP/6-31G(d,p)-GD3. C) B3LYP/6-311+G(d,p)-
GD3-UFF//B3LYP/6-31G(d,p)-GD3-UFF.
obtain reliable kinetic data (Figure 6A). If the mismatched
enantiomer, (S)-1-phenylethanol, was used in the acetylation
reaction, the conversion into the acetylated product proceed-
ed, as expected, more slowly than the (R)-alcohol. The rate of
the reaction is derived from the acquired data set (concentra-
tion versus time) by fitting the data to an arbitrary function
(sixth-order polynomial) and differentiating this function to
obtain the data (rate versus time) in Figure 6B.
lyst, which leads to a shorter distance between the two aro-
matic rings (3.2–3.5 ꢃ) in (R)-TS2-1 (Figure 5B,C). As seen in the
transition states for the acetylation of tert-butanol, the acetate
deprotonates the alcohol as it is approaching the carbonyl
center. Relative to the case of tert-butanol, the alcohol proton
is more shifted to the acetate, the O···C distance that forms is
longer (1.956 ꢃ), and the CÀN bond is shorter (1.521 ꢃ; Fig-
ure 5A).
First, we investigated how the rate of the reaction was influ-
enced by the catalyst concentration by plotting rate/[(À)-1]
total
(i.e., turnover frequency (TOF)) versus the concentration of
[
Ac O] according to the “same excess” protocol. The reaction
2
in Scheme 2 was monitored at two different catalyst
Kinetic studies
concentrations (0.0025 and 0.0050m) and plotted against the
To gain more insight into the mechanism of the acetylation of
concentration of [Ac O]. The two graphs of TOF versus
2
1-phenylethanol, we wanted to determine the reaction orders
[Ac O] from the runs with different catalyst concentrations
2
of the reactants and the catalyst. The kinetic analysis was
made on the basis of reaction progress kinetic analysis (RPKA)
overlap, which suggests that the reaction is of first order with
respect to the catalyst (À)-1 (Figure 7).
[
27]
developed by Blackmond. In this method, the assumption is
made that one molecule of alcohol consumes one molecule of
anhydride as it reacts to form the ester, which results in a con-
To investigate the robustness of the reaction, product inhibi-
tion of the catalyst, and catalyst degradation, an additional
experiment was performed under the same conditions, except
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
5
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
shown in Figure 7. The main difference in the two experiments
is that in the standard trial (Figure 7, squares) the catalyst has
been utilized for half the reaction when it reaches the
half-concentration of the alcohol, which is the same initial
concentration for the other trial (Figure 7, circles). Since the
two graphs overlap, the conclusion can be drawn that no
significant decomposition or product inhibition of the catalyst
takes place.
Next, we wanted to explore the reaction orders of (R)-1-
phenylethanol and acetic anhydride according to the “different
excess” protocol of the RPKA. According to this protocol, the
reaction is performed under several “different excess”
conditions to determine the reaction orders of the reactants. A
plot of [rate]/[Ac O] versus [(R)-1-phenylethanol] should give
2
straight, overlapping lines if the reaction order is 1 for both
the alcohol and the anhydride.
As seen in Figure 8A, such a plot produces slightly curved,
non-overlapping graphs, which suggests a non-integer order
of the reactants and a more complex rate equation. The de-
sired overlap of the curves is best achieved if the exponent for
the acetic anhydride concentration is adjusted to 0.7, and the
most linear curves are obtained if 0.8 is used for the (R)-1-phe-
nylethanol (Figure 8B). If we consider a mechanism similar to
the case of DMAP in Scheme 1 with a pre-equilibrium between
the free catalyst and acetic anhydride with the acetylated cata-
Figure 6. A) Fraction conversion versus time for the reaction in Scheme 2
1
monitored with ReactIR and H NMR spectroscopy. The graph shows data for
different experiments with (R)-1-phenylethanol (run 1: ReactIR= ;
&
1
1
H NMR=*; run 2: ReactIR=~, H NMR=X) relative to (S)-1-phenylethanol
ReactIR=^). B) Rate versus time derived from polynomial fitting from the
(
reaction progress.
Figure 7. Rate/[(À)-1] versus [Ac
2
O] for the reaction in Scheme 2 at two dif-
: [(À)-1]=0.0050m, [ROH]=0.2485m; ~:
O] using differ-
ent starting concentrations of (R)-1-phenylethanol. *: [(À)-1]=0.0050m,
ROH]=0.1270m.
ferent concentrations of (À)-1.
&
[(À)-1]=0.0025m, [ROH]=0.2485m. Rate/[(À)-1] versus [Ac
2
[
Figure 8. A) Rate/[Ac
2
O] versus [(R)-1-phenylethanol] for the reaction in
0
.7
Scheme 2 at different [“excess”] amounts. B) Rate/[Ac
2
O] versus [(R)-1-phe-
nylethanol] for the reaction in Scheme 2 at different [“excess”].
Ac O]=0.3179m, [“ex-
O]=0.3876m, [“excess” amount]=0.14m; ~: [Ac
cess”]=0.07m; *: [Ac
0
.8
&
:
for starting from half the concentration of (R)-1-phenylethanol.
[
2
2
The plots of the TOF versus [Ac O] for these experiments are
2
O]=0.5299m, [“excess” amount]=0.28m.
2
&
&
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
6
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[
19]
lyst, the rate expression for the formation of acetylated alcohol
can be derived. The rate expression depends on the concentra-
tion of acetic anhydride both in the numerator and the
denominator, which is consistent with the observed fractional-
order dependence in acetic anhydride concentration
mixture (1:1) at À208C. Kinetic studies of the DKR of 1-phe-
nylethanol using acetyl isopropyl carbonate and (À)-1 in com-
bination with a ruthenium racemization catalyst suggested
that the reaction order was fractional in the acetyl isopropyl
carbonate.
(Scheme 3).
The equilibrium between the catalyst and acetic anhydride
1
was studied using H NMR spectroscopy. Addition of acetic an-
hydride (20 mL, 27 equiv) to a solution of (À)-1 (0.0076m) in
CDCl (1.0 mL) yielded an equilibrium that was strongly shifted
3
towards the free catalyst and the acetic anhydride (about 1%
of the acetylated catalyst). To better mimic the tert-amyl alco-
hol solvent frequently used in the kinetic resolution of (À)-1,
the same experiment was performed in tert-[D ]butanol.
10
Acetic anhydride was sequentially added to a solution of
(
À)-1 (0.004m) in tert-[D ]butanol (0.5 mL) at 208C. (Figure 9).
10
Upon addition of acetic anhydride (0.21m), the ratio between
the free and acetylated catalyst becomes shifted towards the
acetylated catalyst, which is present in about 30% on the basis
of the NMR spectroscopic signals that correspond to the N-
methyl groups (Figure 9). Further addition of acetic anhydride
(
0.41m in total) gives conditions similar to those under the
kinetic resolutions and shifts the ratio further towards the ace-
Scheme 3. Schematic representation of the (À)-1-catalyzed acetylation of
(
R)1-phenylethanol.
A special case of the pre-equilibrium is the saturation
kinetics according to Michaelis–Menten in which the catalyst is
saturated by the substrate (anhydride) to form the catalyst–
substrate complex. This complex then reacts with the second
substrate (alcohol), and the rate of the reaction only depends
on the concentration of the catalyst and the second substrate.
A scenario with a mixed resting state between free catalyst
and catalyst–substrate complex would lead to a fractional
order in the acetic anhydride. Alternatively, if we consider the
steady-state approximation, the rate can be expressed accord-
ing to Equation (2) in Scheme 3, in which both substrate
concentration terms are present in the denominator. This is in
agreement with the observed non-integer reaction orders for
acetic anhydride (0.7) and (R)-1-phenylethanol (0.8) and would
suggest that the catalyst is distributed between the free state
and its acetylated state.
To determine the resting state of the catalyst, we wanted to
explore the equilibrium between the catalyst and acetic anhy-
dride. In previous studies, (À)-1 was reported to be present
mostly as the free catalyst in a 1:1 mixture of catalyst and
[
16]
acetic anhydride in deuterated CH Cl .
In a more recent
2
2
1
Figure 9. H NMR spectra in the spectral region of the methyl groups at-
tached to the nitrogen obtained at room temperature in tert-[D10]butanol:
study, Fu and co-workers showed that an excess amount of
acetyl isopropyl carbonate (20 equiv) almost completely
formed the acetylated complex in a tert-amyl alcohol/toluene
A) (À)-1 (0.004m); B) 0.21m Ac
E) 1.22m Ac O.
2 2 2
O; C) 0.41m Ac O; D) 0.78m Ac O; and
2
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
7
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
tylated catalyst (38%). In this spectrum (see the Supporting In-
formation), we also observed the formation of acetylated tert-
butanol (d=1.92 ppm), which corresponds to about 1.4% con-
version on the basis of the acetic anhydride. Further additions
led to more formation of the acetylated catalyst, which was
now the major species in solution. From two different experi-
ments, the averaged equilibrium constant was determined to
Transition states with one imaginary frequency were confirmed to
describe the correct movement on the PES by mode analysis.
Representative procedure for ReactIR experiments under
standard conditions
An oven-dried reaction vessel equipped with a stirrer bar (À)-1
(10 mg, 0.0151 mmol, 0.02 equiv) and t-amyl alcohol (anhydrous,
1.74. The equilibrium between the free and the acetylated cat-
2
.64 mL; total volume of 3.0 mL) was added to the vessel, which
alyst is in agreement with the fractional order in acetic
anhydride concentration that is observed from the RPKA.
was capped. The mixture was sonicated at 508C until the catalyst
dissolved. The reaction was allowed to cool to room temperature
and was then placed in a water bath at 178C (a thermostat-
equipped glass bath connected to fresh running water for control
the temperature). The flask was attached to the ReactIR probe that
Conclusion
had been previously purged with N . The ReactIR data recording
2
In this work, we have investigated the nucleophilic mechanism
of the kinetic resolution of 1-phenylethanol using the chiral
planar DMAP catalyst (À)-1 at the B3LYP/6-311+G(d,p)//B3LYP/
was then started. (R)-1-Phenylethanol (90 mL, 0.74 mmol,
1.00 equiv) and triethylamine (160 mL, 1.15 mmol, 1.54 equiv) were
added sequentially and the reaction mixture was stirred for 5–
1
0 min. After stirring for an additional 10 min, a background IR
6
-31G(d,p) level of theory. The transition state for the acetyla-
spectrum was collected and used as a reference. Finally, acetic
tion of the catalyst is slightly higher in energy relative to the
transition state for the acetylation of 1-phenylethanol. The cal-
culations also predict the correct stereochemistry of the kinetic
resolution since the lowest transition state ((R)-TS2-1) leads to
the R product. The same result is also observed when the
Grimme dispersion correction (GD3) and the polarizable con-
ductor calculation model (CPCM, UFF) is included. In the most
stable transition state, a slipped p–p interaction is observed
between the phenyl ring of the substrate and the pyridine ring
of the catalyst. A kinetic analysis using the reaction progress
kinetic analysis suggests that the kinetic resolution proceeds
with a reaction order of 1 for catalyst (À)-1 and with a fraction-
al order for the acetic anhydride (0.7) and 1-phenylethanol
anhydride (110 mL, 1.16 mmol, 1.56 equiv) was added (t=0).
Acknowledgements
P.D. thanks Vetenskapsrꢄdet, Uppsala University, and KTH for
financial support as well as the Swedish Supercomputer
Centre (NSC) for computational resources. The Knut and Alice
Wallenberg Foundation is gratefully acknowledged for
financing of the ReactIR instrument.
Keywords: acylation
·
chirality
·
density functional
(0.8). An NMR spectroscopic investigation of the equilibrium
calculations · kinetics · nucleophilic addition
between the catalyst and acetic anhydride reveals that the cat-
alyst exists in two different states, the free and the acetylated
catalyst, which also explains the fractional-order dependence
on the acetic anhydride concentration under steady-state
conditions.
[
[
[
1] a) W. Steglich, G. Hçfle, Angew. Chem. Int. Ed. Engl. 1969, 8, 981; Angew.
Chem. 1969, 81, 1001; b) G. Hçfle, W. Steglich, Synthesis 1972, 619–621.
2] L. M. Litvinenko, A. I. Kirichenko, Dokl. Akad. Nauk SSSR 1967, 176, 97–
100.
3] For reviews, see: a) G. Hçfle, W. Steglich, H. Vorbrꢅggen, Angew. Chem.
Int. Ed. Engl. 1978, 17, 569–583; Angew. Chem. 1978, 90, 602–615;
b) E. F. V. Scriven, Chem. Soc. Rev. 1983, 12, 129–161; c) A. Hassner in En-
cyclopedia of Reagents for Organic Synthesis (Eds.: L. A. Paquette, D.
Crich, P. L. Fuchs, G. A. Molander), Wiley, Chichester, 1995, pp. 2022–
Experimental Section
2
024; d) U. Ragnarsson, L. Grehn, Acc. Chem. Res. 1998, 31, 494–501;
e) D. J. Berry, C. V. Digiovanna, S. S. Metrick, R. Murugan, ARKIVOC 2001,
01–226.
Computational details
2
All stationary points were initially optimized at the B3LYP level of
[28]
[4] A. C. Spivey, S. Arseniyadis, Angew. Chem. Int. Ed. 2004, 43, 5436–5441;
Angew. Chem. 2004, 116, 5552–5557.
theory as implemented in Gaussian 09 revision D.01 using the 6-
1G(d,p) basis set for C, H, O, and N and an ECP basis set
3
[
[
5] A. Hassner, L. R. Krepski, V. Alexanian, Tetrahedron 1978, 34, 2069–2076.
6] M. R. Heinrich, H. S. Klisa, H. Mayr, W. Steglich, H. Zipse, Angew. Chem.
Int. Ed. 2003, 42, 4826–4828; Angew. Chem. 2003, 115, 4975–4977.
7] S. Xu, I. Held, B. Kempf, H. Mayr, W. Steglich, H. Zipse, Chem. Eur. J.
2005, 11, 4751–4757.
[8] V. Lutz, J. Glatthaar, C. Wꢅrtele, M. Serafin, H. Hausmann, P. R. Schreiner,
Chem. Eur. J. 2009, 15, 8548–8557.
9] A. C. Spivey, S. Arseniyadis, Top. Curr. Chem. 2010, 291, 233–280.
[29]
(
LANL2DZ) for Fe. On the basis of the geometries at this level of
theory, single-point calculations at the B3LYP/6-311+G(d,p) level of
[30]
theory were performed. The transition states for the acetylation
[
of (R)- and (S)-1-phenylethanol were also optimized using the
[31]
Grimme correction for dispersion (D3) at the B3LYP/6-31G(d,p)-
D3 level of theory followed by single-point calculation at the
B3LYP/6-311+G(d,p) level of theory. Lastly, the most stable transi-
tion states were also optimized using the Grimme correction for
dispersion (D3) in combination with the polarizable continuum
[
[
10] a) R. P. Wurz, Chem. Rev. 2007, 107, 5570–5595; b) H. Pellissier, Adv.
Synth. Catal. 2011, 353, 1613–1666; c) C. E. Mꢅller, P. R. Schreiner,
Angew. Chem. Int. Ed. 2011, 50, 6012–6042; Angew. Chem. 2011, 123,
[30]
model (CPCM, UFF, 2-methyl-2-propanol) with the default radii
UFF) in Gaussian 09 revision D.01 (B3LYP/6-31G(d,p)-D3-UFF). All
6136–6167.
(
[
[
11] E. Vedejs, X. Chen, J. Am. Chem. Soc. 1996, 118, 1809–1810.
12] E. Vedejs, X. Chen, J. Am. Chem. Soc. 1997, 119, 2584–2585.
[13] a) G. C. Fu, Acc. Chem. Res. 2004, 37, 542–547; b) J. C. Ruble, G. C. Fu, J.
Org. Chem. 1996, 61, 7230–7231; c) J. C. Ruble, G. C. Fu, J. Am. Chem.
geometries were characterized as minima or saddle points on the
potential-energy surface (PES) by using the sign of the eigenvalues
of the force-constant matrix obtained from a frequency calculation.
&
&
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
8
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Soc. 1998, 120, 11532–11533; d) R. P. Wurz, E. C. Lee, J. C. Ruble, G. C.
Fu, Adv. Synth. Catal. 2007, 349, 2345–2352.
[22] X. Li, P. Liu, K. N. Houk, V. B. Birman, J. Am. Chem. Soc. 2008, 130,
13836–13836.
[
14] a) J. C. Ruble, H. A. Latham, G. C. Fu, J. Am. Chem. Soc. 1997, 119, 1492–
[23] E. Larionov, M. Mahesh, A. C. Spivey, Y. Wei, H. Zipse, J. Am. Chem. Soc.
2012, 134, 9390–9399.
1493; b) S. Bellemin-Laponnaz, J. Tweddell, J. C. Ruble, F. M. Breitling,
G. C. Fu, Chem. Commun. 2000, 1009–1010.
15] J. C. Ruble, J. Tweddell, G. C. Fu, J. Org. Chem. 1998, 63, 2794–2795.
16] B. Tao, J. C. Ruble, D. A. Hoic, G. C. Fu, J. Am. Chem. Soc. 1999, 121,
[24] Zipse and co-workers recently reported the unsolvated transition states
for the acylation of the PPY catalyst and a chiral DMAP catalyst to be
the lowest-energy transition states. Despite considerable effort, we
were not able to find the unsolvated transition state for the acetylation
of (À)-1.
[
[
5091–5092.
[
17] a) Y. Ie, G. C. Fu, Chem. Commun. 2000, 119–120; b) S. Arai, S. Bellemin-
Laponnaz, G. C. Fu, Angew. Chem. Int. Ed. 2001, 40, 234–236; Angew.
Chem. 2001, 113, 240–242.
[25] C. Bonduelle, B. Martꢇn-Vaca, F. P. Cossꢇo, D. Bourissou, Chem. Eur. J.
2008, 14, 5304–5312.
[
[
18] F. O. Arp, G. C. Fu, J. Am. Chem. Soc. 2006, 128, 14264–14265.
19] S. Y. Lee, J. M. Murphy, A. Ukai, G. C. Fu, J. Am. Chem. Soc. 2012, 134,
[26] DDG was calculated from the selectivity factor using the following
°
equation: lns=ln(k
R
/k
S
)=À(DDG /RT).
[
27] a) D. G. Blackmond, Angew. Chem. Int. Ed. 2005, 44, 4302–4320; Angew.
Chem. 2005, 117, 4374–4393; b) J. S. Mathew, M. Klussmann, H. Iwa-
mura, F. Valera, A. Futran, E. A. C. Emanuelsson, D. G. Blackmond, J. Org.
Chem. 2006, 71, 4711–4722.
1
5149–15153.
20] a) T. Kawabata, M. Nagato, K. Takasu, K. Fuji, J. Am. Chem. Soc. 1997,
19, 3169–3170; b) A. C. Spivey, T. Fekner, S. E. Spey, H. Adams, J. Org.
[
1
Chem. 1999, 64, 9430–9443; c) G. Naraku, N. Shimomoto, T. Hanamoto,
J. Inanaga, Enantiomer 2000, 5, 135–138; d) A. C. Spivey, A. Maddaford,
T. Fekner, A. J. Redgrave, C. S. Frampton, J. Chem. Soc. Perkin Trans.
[
28] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Na-
katsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Ha-
segawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A., Jr., Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Ra-
ghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakr-
zewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Dan-
iels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian
1
2000, 3460–3468; e) K.-S. Jeong, S.-H. Kim, H.-J. Park, K.-J. Chang, K. S.
Kim, Chem. Lett. 2002, 31, 1114–1115; f) T. Kawabata, R. Stragies, T.
Fukaya, Y. Nagaoka, H. Schedel, K. Fuji, Tetrahedron Lett. 2003, 44,
1545–1548; g) G. Priem, B. a. Pelotier, S. J. F. Macdonald, M. S. Anson,
I. B. Campbell, J. Org. Chem. 2003, 68, 3844–3848; h) S. A. Shaw, P.
Aleman, E. Vedejs, J. Am. Chem. Soc. 2003, 125, 13368–13369; i) C. ꢆ.
Dꢀlaigh, S. J. Hynes, D. J. Maher, S. J. Connon, Org. Biomol. Chem. 2005,
3, 981–984; j) T. Poisson, M. Penhoat, C. Papamicael, G. Dupas, V. Dalla,
F. Marsais, V. Levacher, Synlett 2005, 2285–2288; k) S. Yamada, T.
Misono, Y. Iwai, Tetrahedron Lett. 2005, 46, 2239–2242; l) E. Busto, V.
Gotor-Fernꢀndez, V. Gotor, Tetrahedron: Asymmetry 2006, 17, 1007–
09, revision D.01, Gaussian, Inc., Wallingford CT, 2009.
1
7
016; m) H. V. Nguyen, D. C. D. Butler, C. J. Richards, Org. Lett. 2006, 8,
69–772; n) S. Yamada, T. Misono, Y. Iwai, A. Masumizu, Y. Akiyama, J.
[
29] a) R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971, 54, 724–
728; b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56,
2257–2261; c) P. C. Hariharan, J. A. Pople, Theor. Chem. Acc. 1973, 28,
213–222; d) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270–283.
Org. Chem. 2006, 71, 6872–6880; o) T. A. Duffey, S. A. Shaw, E. Vedejs, J.
Am. Chem. Soc. 2009, 131, 14–15; p) V. B. Birman, E. W. Uffman, H.
Jiang, X. Li, C. J. Kilbane, J. Am. Chem. Soc. 2004, 126, 12226–12227;
q) V. B. Birman, H. Jiang, Org. Lett. 2005, 7, 3445–3447; r) V. B. Birman, X.
Li, Org. Lett. 2006, 8, 1351–1354; s) X. Li, H. Jiang, E. W. Uffman, L. Guo,
Y. Zhang, X. Yang, V. B. Birman, J. Org. Chem. 2012, 77, 1722–1737; t) B.
Hu, M. Meng, Z. Wang, W. Du, J. S. Fossey, X. Hu, W.-P. Deng, J. Am.
Chem. Soc. 2010, 132, 17041–17044; u) B. Hu, M. Meng, J. S. Fossey, W.
Mo, X. Hu, W.-P. Deng, Chem. Commun. 2011, 47, 10632–10634.
[
[
30] J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980, 72, 650–654.
31] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[
32] a) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995–2001; b) M.
Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24, 669–
681.
[21] a) C. B. Shinisha, R. B. Sunoj, Org. Lett. 2009, 11, 3242–3245; b) C. E.
Mꢅller, L. Wanka, K. Jewell, P. R. Schreiner, Angew. Chem. Int. Ed. 2008,
Received: October 23, 2014
47, 6180–6183; Angew. Chem. 2008, 120, 6275–6278.
Published online on && &&, 0000
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
9
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
DMAP-ing it out: A detailed kinetic
analysis and computational study of the
acetylation of 1-phenylethanol with
acetic anhydride catalyzed by planar
chiral 4-(dimethylamino)pyridine
&
Kinetic Resolution
L. Mesas-Sꢁnchez, P. Dinꢀr*
& – &&
&
A Mechanistic Investigation of the
Kinetic Resolution of Secondary
Aromatic Alcohols Using a Ferrocene-
Based Planar Chiral 4-
(DMAP) catalyst (À)-1 is presented (see
figure). The kinetic study suggests that
the reaction proceeds in a first-order
manner in catalyst, while both
(Dimethylamino)pyridine Catalyst
substrates acetic anhydride and 1-
phenylethanol show fractional order.
&
&
Chem. Eur. J. 2015, 21, 1 – 10
www.chemeurj.org
10
ꢂ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!