one has a two-component CPTC. At present, few one-
component chiral phase transfer catalysts are known which
selector is decreased, one observes a corresponding decrease
in the enantiomeric purity of the transported ion pair.
In biphasic reactions containing all the aforementioned
components, with BPB present initially, the enantiomeric
purity of the ester again increases as the concentration of
selector (S)-2 is increased and as the concentration of THAC
is decreased (Table 2). Ideally, one should adjust the
lipophilicity of the achiral PTC so that, by itself, it cannot
effect transport of the carboxylate anions unless the lipophilic
chiral selector is also involved. This would suppress the
background production of racemic ester, arising from trans-
port occurring independent of the chiral selector. However,
this background reaction can be lessened through use of an
excess of the chiral selector even if the achiral PTC is capable
of unassisted transport.
are effective in converting racemic acids into enantioenriched
esters, something normally considered to be the province of
selected enzymes.
Dilution reduces reaction rate and enantioselectivity. As
expected for a kinetic resolution, the enantiomeric purity of
ester 1b decreases somewhat as the extent of conversion
increases, owing to depletion of the bicarbonate solution of
the more rapidly esterified (S)-enantiomer of DNBleu.
Using a standard reaction protocol, it was found that in a
stirred biphasic solvent system of aqueous sodium bicarbon-
ate and carbon tetrachloride containing 1.0 mol equiv each
of racemic DNBleu, 1a, and 0.5 mol equiv of BPB, little or
no esterification occurs at room temperature. Addition of 1.0
mol equiv of the L-proline-derived selector, (S)-2, does not
alter this situation. However, addition of a small amount of
the achiral THAC leads to production of ester 1b. Enantio-
merically enriched ester is produced when selector (S)-2 is
present; racemic ester is produced when it is not. This latter
observation demonstrates that THAC is sufficiently lipophilic
to transport the carboxylate anion into the carbon tetrachlo-
ride in the absence of selector (S)-2, which, by itself however,
is incapable of effecting transport of the strongly solvated
sodium carboxylate to any significant extent.
Simple partitioning experiments demonstrate that the
enantiomeric purity and the amount of transported DNBleu
depends on both the amount of THAC and selector (S)-2
present. When the carbon tetrachloride layer was isolated
after equilibration with the sodium bicarbonate, racemic 1a,
selector (S)-2, and THAC mixture, and BPB is then added,
subsequent HPLC assay on an (R,R)-Whelk-O column shows
that the amount of ester produced corresponds to the amount
of THAC added (Table 1). Furthermore, as the amount of
Conversely, the enantiomeric purity of the residual (R)-
DNBleu increases as the extent of esterification increases.
Reaction at 4 °C increases the enantioselectivity of the
esterification but slows the process. Because ester can be
produced by an achiral pathway as well as by the process
involving selector (S)-2, the extent to which each process
contributes affects the enantiomeric composition of the ester
produced. Consequently, the stereoselectivity factor (s) is
influenced by the concentrations of the various species. In
one of the examples shown in Table 2, the s factor was 25.8,
a minimum value for the stereoselectivity of the unadulterated
chiral process.
The success of this particular kinetic resolution process
stems from the enantioselective transport process, which
largely masks secondary processes. For example, while the
more stable homochiral complex of (S)-2 and ion pair formed
from (S)-DNBleu and THAC is present in the carbon
tetrachloride to a far greater extent than its heterochiral
counterpart, they may, and do, have different reactivities
toward BPB. Addition of BPB to a carbon tetrachloride
solution containing (S)-2 and the racemic ion pair shows that,
initially, the (R)-ester is formed somewhat (ca. 1.5 times)
more rapidly. Thus, either the less stable diastereomeric ion
pair complex is more reactive than its more stable homochiral
counterpart or it is relatively more dissociated and the
noncomplexed ion pair is more reactive toward BPB than
either of the diastereomeric complexes.
Table 1. Extraction of (() 1a from Saturated Sodium
Bicarbonate into Carbon Tetrachloride with THAC (0.19 equiv)
and Varying Amounts of Selector (S)-2a
(
() 1a (equiv) (S)-2 (equiv) THAC (equiv) %extractedb %eec
1
1
1
1
.0
.0
.0
.0
2.00
1.00
0.50
0.25
0.19
0.19
0.19
0.19
19
19
19
19
86
77
59
46
Although good kinetic data are hard to obtain in biphasic
reactions owing to the dependence of reaction rates on mixing
rates, one can certainly envision the occurrence of a type of
a
Standard conditions entailed use of 0.037 mmol (1 mol equiv) of (()-1
“
product inhibition” arising from the ability of the major
and the indicated number of mol equiv of the other reagents in 2.6 mL of
saturated sodium bicarbonate and 2.6 mL of CCl4. The layers were
equilibrated by rapid stirring at room temperature for 15 min. Layers were
separated, and the CCl4 was washed with water and dried over magnesium
sulfate. Excess BPB was added to the CCl4, and aliquots were assayed
periodically on racemic (%extracted) and (R,R)-Whelko (%ee) HPLC
product, the (S)-ester, to associate with selector (S)-2, thus
lowering its effective concentration, something which could
affect both reaction rates and enantioselectivities. Regardless
of these complexities, the details of which remain to be
unraveled, it is clear that two-component CPTCs can offer
a practical means of kinetic resolution. The system described
here is not optimized and is considered more a “proof of
b
columns. The selector was utilized as the internal standard. Represents
percentage of DNBleu carboxylate ion extracted into organic layer.
c
Enantiomeric excess of the p-bromophenacyl bromide ester.
1822
Org. Lett., Vol. 3, No. 12, 2001
Pirkle
