Biphasic enantioselective partitioning studies using small-molecule chiral selectors

Article abstract of DOI:10.1016/j.tet.2005.05.060

Enantioselective partitioning of racemic N-3,5-(dinitrobenzoyl)leucine or racemic naproxen was studied using a two-component chiral phase transfer approach. A combination of an achiral ion-pairing reagent and a chiral complexing agent (selector) is necessary to effect enantioselective partitioning between an aqueous bicarbonate solution and a nonpolar organic solvent. In these biphasic resolutions, the interplay between the ion-pairing reagent and the selector is essential for maximizing enantioselectivities. Furthermore, the lipophilicity of the ion-pairing reagent, the concentration of the ion-pairing reagent and selector, and the polarity of the organic solvent all exert a considerable influence on the biphasic process. In this manuscript, we conduct optimization studies through analysis of solvent, concentration and ion-pairing effects. Conclusions concerning the mechanistic rationale behind enantioselective partitioning are given.

Full text of DOI:10.1016/j.tet.2005.05.060

Tetrahedron 61 (2005) 7562–7567
Biphasic enantioselective partitioning studies using
small-molecule chiral selectors
*
Seth E. Snyder, James R. Carey and William H. Pirkle
Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA
Received 3 March 2005; revised 6 May 2005; accepted 16 May 2005
Available online 20 June 2005
Abstract—Enantioselective partitioning of racemic N-3,5-(dinitrobenzoyl)leucine or racemic naproxen was studied using a two-component
chiral phase transfer approach. A combination of an achiral ion-pairing reagent and a chiral complexing agent (selector) is necessary to effect
enantioselective partitioning between an aqueous bicarbonate solution and a nonpolar organic solvent. In these biphasic resolutions, the
interplay between the ion-pairing reagent and the selector is essential for maximizing enantioselectivities. Furthermore, the lipophilicity of
the ion-pairing reagent, the concentration of the ion-pairing reagent and selector, and the polarity of the organic solvent all exert a
considerable influence on the biphasic process. In this manuscript, we conduct optimization studies through analysis of solvent, concentration
and ion-pairing effects. Conclusions concerning the mechanistic rationale behind enantioselective partitioning are given.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
solvent conservation can be circumvented with enantio-
selective membrane-based separations. Several examples
have been demonstrated using chiral selectors that are
soluble in one of two liquid phases separated by a semi-
permeable membrane. The chiral selector mediates transfer
of a racemic compound across the membrane by associating
with each enantiomer of the racemate, generating dia-
stereomeric complexes. Owing to the nature and direction-
ality of the intermolecular interactions involved, these
complexes may be energetically nondegenerate. As a result,
the selector is capable of influencing the position of
equilibrium of each enantiomer between the respective
phases. In general, the more highly complexed enantiomer
is preferentially extracted into the phase containing the
selector. A variety of chiral selectors have been employed in
enantioselective membrane separations including tartaric
acid derivatives,7 chiral amine hydrochlorides,8 chiral
ionophores,9 chiral crown ethers,10 polyamino acid deriva-
tives,11 cyclodextrins,12 quinine-based compounds13 and
hydroxyproline derived compounds.14
Owing to the dramatic increase in sales of single enantiomer
drugs over the past decade, the development of new chiral
technologies continues to be a very active area of research in
the pharmaceutical and specialty chemicals industries.
Strategies aimed at the preparative resolution of racemic
compounds are amongst the most important means of
producing drugs in single enantiomer form.1 Traditional
methods employ crystallization2 and preparative chromato-
graphic separations.3 Despite their widespread usage in
industrial scale-up processes, these techniques are limited to
batch or sequential processes.4 From an economic
standpoint, a continuous separation process is ideal. Indeed,
simulated counter-current processes have found growing
application in enantioselective separation processes. In
particular, simulated moving-bed (SMB) technology has
become an important means of performing preparative
chiral separations.5 Nonetheless, SMB often requires
extremely thorough optimization procedures and can be
quite costly.
Chiral selectors synthesized from small organic molecules
developed in our laboratories are capable of separating a
variety of enantiomers when used as chromatographic chiral
stationary phases (CSPs).15 These selectors have been
designed using first principles through consideration of the
minimum requirements that necessitate strong chiral
recognition. The majority of selectors incorporate a
combination of hydrogen bonding and p–p interactions.
Acting in conjunction, these intermolecular forces can exert
a significant level of stereochemical control. Many designed
An alternative approach to continuous enantioselective
separation processes involves using membrane devices.6
Many of the drawbacks associated with conventional
separation strategies such as low substrate throughput and
Keywords: Extraction; Biphasic; Kinetic resolution; Chiral selector;
Naproxen; Phase transfer catalyst.
*
Corresponding author. Tel.: C1 217 3332819; fax: C1 217 3332685;
e-mail: sesnyder@mail.med.upenn.edu
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.05.060

S. E. Snyder et al. / Tetrahedron 61 (2005) 7562–7567
7563
Scheme 1.
chiral selectors show high levels of enantiodiscrimination
for pharmaceutically important compounds and are thus
amenable to preparative chromatographic and membrane-
based separation processes. For instance, a chiral selector
has been used in a hollow fiber membrane system (Sepracor)
to separate the enantiomers of N-protected amino acids.16
The approach relies on using the chiral selector in
combination with an achiral phase transfer catalyst (PTC)
to achieve appreciable rates of extraction across the
membrane. Recently, we have extended the methodology
to performing biphasic kinetic resolutions by coupling the
enantioselective transport to a subsequent chemical
reaction, a process referred to as two-component chiral
phase transfer catalysis (Scheme 1).17 Although these
kinetic resolutions are performed as batch processes, they
are potentially extendable to membrane type reactors.
Herein, we report on the various factors that contribute to
enantiodiscrimination in two-component chiral phase
transfer catalysis, the ultimate goal being optimization for
potential enantioselective membrane extraction systems,
membrane reactors, and process-scale chiral organocatalytic
reactions.
extraction step. To this end, a chiral selector ((S)-1, (S)-2a or
(3S,4S)-2b) was dissolved in an organic solvent and mixed
with an aqueous sodium bicarbonate solution containing
either racemic N-(3,5-dinitrobenzoyl)leucine, (G)-3, or
racemic naproxen, (G)-4, in the presence of a suitable
achiral ion-pairing reagent. Three symmetrical ion-pairing
reagents were employed in this study, namely tetra-
hexylammonium chloride (THAC) tetrahexylammonium
bromide (THAB) and tetrabutylammonium chloride
(TBAC). The simple partitioning studies were designed to
study the effects of selector concentration, racemate
2. Results and discussion
Success of the kinetic resolutions in two-component chiral
phase transfer catalysis stems primarily from the enantio-
selective partitioning step, although secondary effects such
as differences in reaction rates between the complexed and
uncomplexed enantiomer contribute slightly.17 Therefore,
our primary focus involved optimizing the enantioselective
Table 1. Biphasic Enantioselective Partitioning of (G)-3 in the Presence of (S)-1a
EntrySolvent
Solvent
(S)-1 (equiv)
[(S)-1] (M)
NC(alkyl)4XK
NC(alkyl)4XK
(mol%)
% Extractedb
%ee (S)c
1
2
3
4
5
6
7
8
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CCl4
CCl4
CCl4
CCl4
CCl4
CCl4
CCl4
CCl4
CCl4
Hexane
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.25
0.25
0.25
1.0
1.0
1.0
1.0
0.008
0.025
0.10
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4ClK
NC(hexyl)4BrK
NC(butyl)4ClK
20
20
20
20
20
20
20
20
20
20
40
40
40
40
20.2
19.6
19.5
21.0
19.5
19.7
21.0
19.4
20.0
19.5
38.0
37.5
17.0
19.5
27
50
70
84
76
93
96
50
57
67
73
74
91
90
0.30
0.008
0.025
0.10
0.002
0.006
0.025
0.025
0.025
0.025
0.025
9
10
11
12
13
14
N
C(hexyl)4ClK
a Standard conditions entailed vigorously mixing 0.10 mmol (1.0 M equiv) of (G)-3 in 4.0 mL of saturated NaHCO3 with an organic solution containing
selector (S)-1 and the indicated ion-pairing reagent at room temperature. Layers were separated and 3 was derivatized with excess phenacyl bromide.
b Determined by HPLC internal standard analysis (see Section 4).
c Refers to %ee of the compound extracted into the organic layer. Determined by chiral HPLC (see Section 4).

7564
S. E. Snyder et al. / Tetrahedron 61 (2005) 7562–7567
concentration, solvent and ion-pairing reagent on enantio-
selective partitioning.
used as the extraction solvent. In each case, 1 M equiv of the
chiral selector (S)-1 and 20 mol% THAC are present in the
organic layer. Changing the effective concentration of
selector (and THAC) in the organic layer was accomplished
by diluting with methylene chloride. The results from these
partitioning studies reveal that the quantity of 3 extracted
into CH2Cl2 is generally equivalent to the mole% of THAC
in the solution. The selector by itself is incapable of
effecting transfer between layers. The enantiomeric excess
of the extracted 3 is highly dependent on the concentration
of selector in the CH2Cl2. A threefold increase in
enantioselectivity is observed as selector concentration is
increased from 0.008 to 0.30 M.
Results of enantioselective partitioning of (G)-3 in the
presence of chiral selector (S)-1 are displayed in Table 1. A
general schematic of the equilibrium process is shown in
Figure 1. In all cases, 1 M equiv of (G)-3 was dissolved in
an aqueous bicarbonate solution and mixed with an organic
solvent containing the selector and a tetraalkylammonium
halide. Mixtures were shaken vigorously and allowed to
equilibrate. The layers were separated and analyzed.
Equilibration occurs rapidly once the layers are mixed and
results are independent of the contact time between layers.
The concentration of (G)-3 in the aqueous layer was kept
constant at 0.025 M for all enantioselective extractions
shown in Table 1. The natures of the organic solvent and
tetraalkylammonium salt as well as the concentration of
selector (S)-1 were varied from one run to the next.
The same general trends are observed with carbon
tetrachloride (CCl4) as the extraction solvent. However,
the polarity of the organic solvent also has a considerable
influence on enantioselective bias. Association constants for
selector/substrate complexes are maximum in nonpolar
solvents where there is less competitive solvation from the
In entries 1–4 of Table 1, methylene chloride (CH2Cl2) was
Figure 1. Schematic of the two-component enantioselective partitioning process.

S. E. Snyder et al. / Tetrahedron 61 (2005) 7562–7567
7565
medium. Hence, extractions performed in CH2Cl2 are far
less selective than those performed in CCl4. For instance,
comparing entry 2 with entry 6 in Table 1, an increase from
50%ee (19.6% extracted) to 93%ee (19.7% extracted) is
observed when the extraction solvent is changed from
CH2Cl2 to CCl4 under otherwise identical conditions. The
ratio of racemate/selector in the biphasic solution also has a
significant influence on enantioselective partitioning. As
seen in entries 6 and 10 in Table 1, decreasing the number of
molar equivalents of selector (S)-1 from 1.0 to 0.25 while
keeping the selector concentration constant (by adjusting
the amount of CCl4) gives significantly lower enantio-
selectivities of extraction. Presumably, in the later case, an
increased amount of achiral extraction ensues as the selector
is effectively ‘tied-up’ through its strong association with a
single enantiomer of the racemate. Changing the counterion
of the achiral ion-pairing reagent from chloride to bromide
has little influence on the enantioselective partitioning
process, as seen in entries 11 and 12 of Table 1.
enantiomeric ion-pairs formed between TBAC and (G)-3
partition almost exclusively into the aqueous layer. On the
other hand, in the presence of selector (S)-1, a significant
quantity of 3 is extracted into the CCl4 (e.g., entry 13 in
Table 1). This result indicates that the association constant
between selector (S)-1 and (S)-3 is sufficiently large as to
perturb the partitioning equilibrium that exists without
selector present. The diastereomeric associates likely form
at the biphasic interface as two polar groups of (S)-3 lose
their hydrophilicity owing to strong hydrogen bonding
interactions with the selector. Furthermore, a multipoint
p–p interaction provides the essential third point of contact
and ensures strong complexation necessary for extraction
into the organic layer.
Background extraction is also reduced significantly by using
hydrocarbon solvents such as hexane (entry 14 of Table 1)
or decane, although symmetrical ion-pairing reagents used
in this study have limited solubility in both hydrocarbon and
aqueous solutions. From an environmental standpoint,
hydrocarbon solvents are far more ideal than halogenated
solvents such as CCl4.As a result, we explored the use of
decane as a solvent in the two-component chiral phase
transfer catalysis (Scheme 1). With phenacyl halides as
alkylating reagents, reactions proceed very slowly unless
significant quantities of ion-pairing reagent are added.
Simple methyl or ethyl esters of 3 can be prepared by the
two-component chiral phase transfer methodology.
Although methyl halides are not reactive in the two-
component systems, dimethyl sulfate ((CH3O)2SO2)) is a
highly reactive methylating reagent capable of converting 3
into its corresponding ester quite rapidly with significant
enantioselectivity (Scheme 2). Notably, the reaction is
accomplished in a hydrocarbon solvent (decane) using a
catalytic amount of THAB. However, competitive hydroly-
sis of dimethyl sulfate restricts one to use greater than
stoichiometric quantities of this electrophile. As shown in
Scheme 2, four equivalents of dimethyl sulfate are required
for 50% of 3 to be esterified.
To analyze the influence of changing the mol% of achiral
ion-pairing reagent, it is necessary to define the stereo-
selectivity factor (s) of enantioselective partitioning. In a
kinetic resolution, the s factor represents the ratio of rate
constants between the enantiomers and is given by the
following equation:
s Z ln½1 KCð1 Cee0Þꢀ=ln½1 KCð1 Kee0Þꢀ
where ee0 is the enantiomeric excess of product at
conversion C.18 The s factor is expected to remain constant
throughout the course of a kinetic resolution unless
secondary effects (e.g., product inhibition) contribute to
the selectivity of the process. In the present case, the s factor
for enantioselective partitioning may be defined as the
relative ratio of partitioning coefficients of the enantiomers
between the respective layers. The above equation still
applies, accept that the term C corresponds to the total
amount extracted into the organic layer. As displayed in
Table 1, increasing the mol% of THAC from 20% (entry 6)
to 40% (entry 11), the total quantity of 3 extracted increases
from 19.7 to 38% and the ee decreases from 93 to 73%,
corresponding to an s factor of 35 and 10, respectively. The
decrease in s factor most likely results from a greater degree
of achiral partitioning with increased amounts of THAC,
although the influence of ion-pair aggregation may also be a
factor.
Chromatographic separation of underivatized naproxen,
(G)-4, may be accomplished through using a CSP derived
from (S)-2a or (3S,4S)-2b. Separation factors are modest,
ranging from 2.25 to 2.95 depending on the means by which
the selector is tethered to the silica support.19 Enantiomers
of naproxen may also be separated by the selector mediated
partitioning process described above. Data for the biphasic
enantioselective partitioning of 0.10 mmol (1.0 M equiv) of
(G)-4 in the presence of 0.10 mmol of selector (S)-2a or
(3S,4S)-2b and 20-mol% THAC is given in Table 2. Unlike
biphasic extractions of 3 discussed above, the quantity of 4
extracted into the organic layer not only depends on the
To mitigate the effects of achiral background partitioning,
one may adjust the lipophilicity of the ion-pairing reagent
such that the selector is necessary to effect transfer between
the aqueous and organic layers. Under standard partitioning
conditions (Table 1) but in the absence of a chiral selector,
Scheme 2.

7566
S. E. Snyder et al. / Tetrahedron 61 (2005) 7562–7567
Table 2. Biphasic enantioselective partitioning of (G)-4 with 20-mol% THACa
Entry
Solvent
Selector
[Selector] (M)
[(G)-4] (M)
% Extractedb
%ee (S)c
sd
1
2
3
4
5
6
7
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CCl2/CH2Cl2
CH2Cl2
CCl2/CH2Cl2
(S)-2a
(S)-2a
0.025
0.10
0.025
0.10
0.025
0.10
0.025
0.025
0.025
0.025
0.025
0.025
0.10
20.8
11.6
20.1
13.5
16.8
18.0
17.0
18
33
33
51
50
45
50
1.5
2.2
2.1
3.4
3.3
2.9
3.3
(3S,4S)-2b
(3S,4S)-2b
(3S,4S)-2b
(3S,4S)-2b
(3S,4S)-2b
e
e
0.10
a Standard conditions entailed vigorously mixing 0.10 mmol (1.0 M equiv) of (G)-4 in 4 mL of saturated NaHCO3 with an organic solvent containing the
indicated chiral selector and 20 mol% THAC at room temperature. Layers were separated and 4 was derivatized with excess dimethyl sulfate.
b Determined by HPLC internal standard analysis (see Section 4).
c Determined by chiral HPLC (see Section 4).
d Stereoselectivity factor.
e Used a 3:1 ratio of CCl4/CH2Cl2.
mol% of THAC in solution but also the amount of solvent in
the respective layers. Generally, the quantity extracted is
equal to or less than 20%. Dilution of the organic layer or
concentration of the aqueous layer results in a larger
quantity of naproxen ion-pairs extracted. For comparative
purposes, the stereoselectivity factor (s) for each experiment
was calculated. Comparing entry 1 to entry 2 or entry 3 to
entry 4 of Table 2, one observes that more concentrated
solutions of the selector result in less total naproxen
extracted but with higher s factors. The cis-3 substituted
selector (3S,4S)-2b gives higher s factors than (S)-2a for
these partitioning studies, a result consistent with chromato-
graphic studies (table 2; entry 1,2 vs 3,4).20 Ideally, one
should develop a more soluble analogue of (3S,4S)-2b by
increasing the bulk on the second stereocenter of the
cyclohexane ring, hence making it possible to perform these
enantioselective partitionings in more nonpolar or concen-
trated solvent systems.
facilitate the production of future CSPs and chiral
organocatalysts.
4. Experimental
The racemates and chiral selectors used in this study were
prepared in a manner described by the original workers.19,21
Selector (3S,4S)-2b was obtained from Merck. All ion-
pairing reagents, phenacyl bromide and dimethyl sulfate
were purchased from Aldrich. Carbon tetrachloride was
purchased from Fisher. HPLC was conducted using the
following equipment: injectors (Beckman, Rheodyne),
pumps (Rainin Rabbit-HPX, Beckman 100B, Alcott 760),
variable wavelength (l) UV-detector (Linear UVIS 200),
recorder/integrator (HP 3390A). Chromatographic runs
were recorded at ambient temperature with the flow rate
2 mL/min. Dimensions of all analytical HPLC columns
were 24 cm!4.6 mm.
4.1. Typical procedure for enantioselective partitioning
3. Conclusions
A solution containing (G)-3 (0.10 mmol, 32.5 mg) in 4 mL
saturated NaHCO3 was added to 4 mL CH2Cl2 containing
(S)-1 (0.10 mmol, 28.8 mg) and THAC (0.02 mmol,
7.8 mg). The solution was added to a screw-capped
scintillation vial, shaken vigorously for 1 min and allowed
to settle for an additional 3 min. The contents were placed in
a separatory funnel and the layers were separated. The
organic layer was dried over magnesium sulfate. The
extracted ion-pairs of 3 were converted to their phenacyl
ester derivatives by the addition of excess phenacyl bromide
(0.12 mmol, 23.9 mg). Enantioselective partitioning of
(G)-4 were carried out in similar fashion except that
enantioenriched extracted ion-pairs of 4 were converted to
their methyl ester derivatives by the addition of excess
dimethyl sulfate. Residual dimethyl sulfate was hydrolyzed
with a saturated solution of NaHCO3.
The biphasic enantioselective partitioning studies and
kinetic resolutions described in this paper may have a
significant practical value. The method provides a simple
and convenient way to resolve a variety of enantiomers,
provided that a suitable chiral complexing agent is
available. Two such examples were demonstrated,
resolution of N-(3,5-dinitrobenzoyl)leucine and resolution
of naproxen. In the former case, high levels of enantio-
discrimination were achieved both for enantioselective
partitioning and kinetic resolutions, even for the simple
batch experiments described. Furthermore, results for
separation of naproxen by a single-stage batch extraction
process are encouraging although enantioselectivities are far
below that needed for practical separation of enantiomers.
The same will likely be true of many racemate/selector
combinations. Hence, a continuous process involving
multiple membrane units (staging) will be needed for such
cases. Indeed, Ding and co-workers have developed an
enantioselective hollow-fiber membrane approach using
countercurrent flow of the two phases leading to complete
separation of enantiomers, even for a selector with an
undesirably low enantioselectivity.14 This type of enantio-
selective membrane approach can clearly be applied to two-
component chiral phase transfer separations and reactions.
Furthermore, this methodology provides a simple means of
assessing models of chiral recognition. This should greatly
4.2. HPLC analysis
Enantiomeric excess was determined by chiral HPLC by
analyzing the ester derivatives of the extracted ion-pairs of 3
and 4. In the former case the enantiomers of the phenacyl
ester of 3 were separated using (D)-Leucine CSP (mobile
phase 15% 2-propanol in hexane, lZ270 nm) available
from Regis Technologies. Enantiomers of methyl esters of 4
were separated using an (R,R)-Whelk-O1 CSP (mobile

S. E. Snyder et al. / Tetrahedron 61 (2005) 7562–7567
7567
in Industry: the Commercial Manufacture and Applications of
Optically Active Compounds; Collins, A. N., Sheldrake, G. N.,
Crosby, J., Eds.; Wiley: Chichester, 1992; Chapter 2.
(c) Newman, P.; Optical Resolution Procedures for
Chemical Compounds; Optical Resolution Information
Center, Manhattan College: Riverdale, NY, 1981; Vol.
1–3.
phase 10% 2-propanol in hexane, lZ283 nm) available
from Regis Technologies. In all cases, absolute configur-
ations were assigned by comparison with authentic samples.
Quantities extracted were determined by HPLC using
internal standard analysis. The chiral selector was used as
an internal standard. Calibration curves for each study were
generated by preparing stock solutions of the chiral selector
with the appropriate racemic ester derivative of the
compound being studied (i.e., the phenacyl ester derivative
of 3 or the methyl ester derivative of 4) at various
concentrations. Linear plots were obtained in all cases. To
determine the amounts extracted, aliquots from the worked-
up organic layer were injected and compared to the
calibration data.
3. (a) Allenmark, S. Chromatographic Enantioseparation:
Methods and Applications 2nd ed.; Ellis Horwood: Chichester,
1991; Chapter 9. (b) Chiral Separations by HPLC: Appli-
´
cations to Pharmaceutical Compounds; Krstulovic, A. M.,
Ed.; Halsted: New York, 1989. (c) Pirkle, W. H.; Hamper,
B. C. In Preparative Liquid Chromatography; Biddlingmeyer,
B., Ed.; Journal of chromatography library; Elsevier:
Amsterdam, 1987; Vol. 38; Chapter 7.
4. White, C. A. In Preparative Process-scale Liquid Chroma-
tography; Subramanian, G., Ed.; Ellis Horwood: New York,
1991.
4.3. Procedure for biphasic reaction between (G)-3 and
dimethyl sulfate in the presence of selector (S)-1
(Scheme 2)
5. For reviews see, (a) Beste, Y. A.; Lisso, M.; Wonzy, G.; Arlt,
A. J. Chromatogr. A 2000, 868, 169–188. (b) Ruthven, D. M.;
Ching, C. B. Chem. Eng. Sci. 1989, 11, 1011–1038.
6. For a review of membranes in chiral separations, see: Chiral
Separation Techniques: a Practical Approach; Subramanian,
G., Ed. 2nd ed.; 2001.
To a 10 mL round bottom flask equipped with magnetic stir
bar was added 1.0 mol equiv of (G)-3 (0.10 mmol,
32.5 mg) dissolved in saturated NaHCO3. To this solution
was added 3.0 mL of decane containing 1.0 mol equiv (S)-1
(0.10 mmol, 28.8 mg) and 0.1 mL of CH2Cl2 containing
2-mol% THAB (0.002 mmol, 0.87 mg). The biphasic
mixture was stirred magnetically and 0.1 mL of CH2Cl2
containing 4.0 mol equiv dimethyl sulfate (0.40 mmol
50.5 mg) was added drop wise. Aliquots were assayed
periodically by chiral HPLC using a (R, R)-Whelk-O1
column (12% 2-propanol in hexane, lZ283 nm) available
from Regis Technologies. Conversion was measured by
internal standard analysis (see above). Production of ester
stopped after approximately 2 h. At this point, close to 50%
conversion had been reached. Addition of more dimethyl
sulfate leads to additional conversion. Workup was carried
out as follows: the organic layer was diluted with CH2Cl2,
and the aqueous layer was diluted by adding H2O. The
layers were separated and the organic layer was washed
sequentially with 1 M HCl, saturated NaCl and water. The
solution was dried over sodium sulfate and concentrated
under reduced pressure to yield a mixture of product ester,
THAB and (S)-3. Separation of the mixture was accom-
plished by flash column chromatography (SiO2, hexane/
ethyl acetate).
7. (a) Prelog, V.; Dumic, M. Helv. Chim. Acta 1986, 69, 5–11.
(b) Prelog, V.; Mutak, S.; Kovacevic, K. Helv. Chim. Acta
1983, 66, 2279–2284. (c) Prelog, V.; Stojanac, Z.; Kovacevic,
K. Helv. Chim. Acta 1982, 65, 377–384.
8. Lehn, J. M.; Moradpour, A.; Behr, J. P. J. Am. Chem. Soc.
1975, 97, 2532–2534.
9. (a) Prelog, V. Chimia 1983, 37, 12. (b) Thoma, A. P.; Viviani-
Nauer, A.; Schellenberg, K. H.; Bedekovic, D.; Pretsch, E.;
Prelog, V.; Simon, W. Helv. Chim. Acta 1979, 62, 2303–2316.
(c) Prelog, V. Pure Appl. Chem. 1978, 50, 893–904.
(d) Armstrong, A.; Still, W. C. J. Org. Chem. 1992, 57,
4580–4582.
10. For reviews, see: (a) Stoddart, J. F. In Chiral Crown Ethers;
Eliel, E. L., Wilen, S. H., Eds.; Topics in Stereochemistry;
Wiley: New York, 1988; Vol. 17, p 207. (b) Potvin, P. G.;
Lehn, J. M. In The Design of Selective Complexing Agents;
Izatt, R. M., Christensen, J. J., Eds.; Wiley: New York, 1987;
p 167.
11. Maruyama, A.; Adachi, N.; Takatsuki, T.; Torii, M.; Sanui, K.;
Ogata, N. Macromolecules 1990, 23, 2748–2752.
12. Armstrong, D. W.; Jin, H. L. Anal. Chem. 1987, 59,
2237–2241.
¨
13. Kellner, K.-H.; Blasch, A.; Chmiel, H.; Lammerhofer, M.;
Acknowledgements
Lindner, W. Chirality 1997, 9, 268–273.
14. Ding, H. B.; Carr, P. W.; Cussler, E. L. AIChE J. 1992, 38,
1493–1498.
This work was financially supported by gifts from the estate
of Louise V. Leonard, from AstraZeneca and a grant from
the University of Illinois Research board to S.E.S.
15. For a review, see: Welch, C. J. J. Chromatogr. A 1994, 666,
3–26.
16. Pirkle, W. H.; Bowen, W. E. Tetrahedron: Asymmetry 1994, 5,
773–776.
17. Pirkle, W. H.; Snyder, S. E. Org. Lett. 2001, 3, 1821–1823.
18. Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18,
249–330.
References and notes
19. Pirkle, W. H.; Welch, C. J. J. Liq. Chromatogr. 1992, 15,
1947–1955.
1. For example, see: Crosby, J. J. Tetrahedron 1991, 47,
4789–4846.
20. Wolf, C.; Pirkle, W. H. Tetrahedron 2002, 58, 3597–3603.
21. Pirkle, W. H.; Koscho, M. E. J. Chromatogr. A. 1999, 840,
151–158.
2. (a) Sheldon, R. A. Chirotechnology: Industrial Synthesis of
Optically Active Compounds; Marcel Dekker: New York,
1993; Chapter 3. (b) Bayley, C. R.; Vaidya, N. A. In Chirality