5222
K. Tateishi et al. / Tetrahedron Letters 54 (2013) 5220–5223
2b (>99% ee) in the less polar fraction. In terms of practicality and
isolated yield, these data constitute a meaningful 21% and 26%,
respectively, of the enantiomerically pure product, obtained from
the starting racemic mixture. Determination of the absolute config-
uration of the enantiomerically pure product 2b, by comparison
with the authentic sample, has revealed its (S)-configuration.7
As one can see from chart C, the area of 2b has several shoulders
prompting us to investigate their composition in more detail. To
this end we isolated six fractions, as shown in Figure 4, and studied
their enantiomeric composition as well as the absolute configura-
tion of the excess enantiomer. The first fraction was found to be
noticeably enriched (39% ee) in (S)-2b enantiomer, while rest of
the fractions (from the second to the sixth) contained excess of
(R)-2b with gradually decreasing enantiomeric purity from 23 to
10% ee. One may assume that a similar enantiomeric composition
might be also observed in the experiments conducted with 1 and 3
equiv of reagent (S)-1b, charts (A) and (B), respectively. However,
in these cases the separation of the enantiomeric species is less
efficient and the corresponding shoulders are not pronounced.
Considering these preliminary results we may conclude that our
design was quite successful as the inducing reagent (S)-1b led to
a separation of (S)-2b enantiomer transforming the initial race-
mate 2b into an enantiomerically enriched state which further
underwent the SDE previously described for this type of
compounds.
Finally, we decided to briefly assess the generality of this meth-
od for resolution of other racemic N-formyl (phenyl)ethylamine
derivatives. For example, racemate 3b [N-formyl 1-(4-methoxy-
phenyl)ethylamine] (60 mg) was mixed with (S)-1b (275 mg), in
a mole ratio of 1:5.5 and subjected to MPLC under the standard
conditions (Fig. 5). Similar to the results obtained for resolution
of 2b, the chiral reagent 1b was eluted first followed by the area
of compound 3b, clearly showing several shoulders. The first, less
polar fraction, contained enantiomerically pure (99% ee) (S)-3b,7
which was obtained in 14% recovery yield (4.3 mg). The next frac-
tion was enantiomerically enriched (31% ee) in the same (S)-enan-
tiomer, followed by the fractions containing (R)-enantiomer in
excess of various enantiomeric purity.
NHCHO
NHCHO
Me
+
Me
rac-3b
60 mg
(S)-1b
MeO
:
eluent:
275 mg
hexane/AcOEt = 1
mol ratio 5.5
1
(S)-3b
99%ee
(S)-1b
(4.3 mg)
(S)-3b 31%ee
(R)-3b 5%ee
(R)-3b 8%ee
(R)-3b 5%ee
[min]
60
20
40
retention time
Figure 5. MPLC chart of the mixtures of (S)-1b and rac-3b with a mole ratio of
5.5:1.
NHCHO
Me
NHCHO
Me
+
rac-4b
eluent:
hexane/AcOEt = 2
(S)-1b
Me
:
60 mg
164.5 mg
mol ratio
3
1
first fraction
(S)-4b (>99%ee, 8.5 mg)
second fraction
(S)-4b (15%ee, 15.6 mg)
third fraction
(S)-1b and 4b [(R)-rich]
In another experiment, using racemate 4b [N-formyl 1-(4-
methylphenyl)ethylamine] (60 mg), mixed with the inducing re-
agent (S)-1b (164.5 mg) in a mole ratio of 1:3, we obtained a bit
different results as compared with the derivatives 2b and 3b. In
this case (Fig. 6) the first collected fraction contained enantiomeri-
cally pure (S)-4b (>99% ee, 8.5 mg, recovery yield 28%) followed by
another (S)-enantiomerically enriched fraction (15% ee, 15.6 mg).
The last, third collected fraction, contained the inducing reagent
(S)-1b and (R)-enantiomerically enriched 4b. Thus, the elution or-
der of enantiomers of 4b and the reagent (S)-1b were different
form the 2b and 3b cases, but the stereochemical outcome, result-
ing in the separation of the (S)-enantiomer 4b,7 was virtually the
same. Accordingly, one may agree that this new method might
have some degree of generality, at least in a series of structurally
[min]
90
70
100
80
retention time
Figure 6. MPLC chart of the mixtures of (S)-1b and rac-4b with a mole ratio of 3:1.
similar compounds, using the same chiral inducing reagent (S)-
1b. Obviously, different classes of compounds may require another
inducing reagent or modified conditions, but clearly, the concept
discovered in this study constitutes a new, non-conventional ap-
proach for racemate resolution.
One may agree, that elucidation of the detailed mechanism of
the observed in this study chromatographic profiles will require a
separate investigation. However, what seems to be obvious is
that the chiral inducing reagent interacts selectively with one
of the enantiomers of the starting racemate producing high-order
species having different retention times. As we have suggested in
the previous work,3 compounds of this class, tend to form syndio-
tactic hetero-chiral, hydrogen bonded associations which are
preferred in a solution over isotactic homo-chiral associations.5
Thus, we may assume that (S)-chiral inducing reagent, selectively
forms high-order species with (R)-enantiomer of the racemate
(Scheme 1, m > n), which have longer retention times as
compared to monomeric species of the (S)-enantiomer, being
eluted first.
(R)-2b (23%ee)
(S)-2b (39%ee)
(R)-2b (22%ee)
(R)-2b (20%ee)
(R)-2b (16%ee)
(S)-2b
(>99%ee)
less polar
fraction
(R)-2b (10%ee)
[min]
50
60
retention time
Figure 4. Detailed ee change of 2b in chart C (Fig. 3).