Separation of diol enantiomers - A remarkable influence of water

Article abstract of DOI:10.1016/S0957-4166(02)00035-6

The water content of ⁱPrOH is shown to have a profound influence on the separation of two enantiomers of 2,2′-dibromohydrobenzoin on a Daicel OJ column. ⁱPrOH containing 0.47% water gives no separation, whereas increasing the amount of water in the eluent to 3.4% effects complete separation. The influence is not general.

Full text of DOI:10.1016/S0957-4166(02)00035-6

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 3329–3331
Separation of diol enantiomers—a remarkable influence of water
Paul Wyatt
School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK
Received 17 December 2001; accepted 24 January 2002
i
Abstract—The water content of PrOH is shown to have a profound influence on the separation of two enantiomers of
i
2,2%-dibromohydrobenzoin on a Daicel OJ column. PrOH containing 0.47% water gives no separation, whereas increasing the
amount of water in the eluent to 3.4% effects complete separation. The influence is not general. © 2002 Elsevier Science Ltd. All
rights reserved.
The separation of chiral diols by chiral HPLC to
determine enantiomeric excess is a routine and well
used technique and hydrobenzoin 2 is readily separated
using a Daicel OJ column.1,2 We had recently been
separating enantiomers of diol 1 using a 9:1 mixture of
heptane:iPrOH. Fig. 1 illustrates a typical run. How-
was completely lost (Fig. 2), something that was reme-
died on returning to the old batch of PrOH. The
i
magnitude of the difference of separation between the
two batches is profound. In one case there is no
detectable separation and in the other there is a base-
line silence of over 3 min between the enantiomers.
i
i
ever, on changing to a new batch of PrOH, separation
Switching between the two batches of PrOH showed
Figure 1. Separation of enantiomers of ( )-1. 9:1 hep-
Figure 2. 9:1 Heptane:iPrOH (‘new’).
tane:iPrOH (‘old’).
E-mail: paul.wyatt@bristol.ac.uk
0957-4166/01/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00035-6

3330
P. Wyatt / Tetrahedron: Asymmetry 12 (2001) 3329–3331
that the effect was entirely reproducible. Separations
were performed on an OJ column with an eluent flow
rate of 0.8 mL min−1 and with detection at 225 nm.3
After investigating several potential reasons for this
change, the difference between the two batches was
found to be water content. The water content of the
i
new batch of PrOH was found to be 0.47% v/v,
whereas the old batch had acquired seven times this
amount at 3.4% v/v. The water content was readily
1
determined by H NMR analysis.
Figure 3. 9:1 Heptane:iPrOH (‘new’+H2O).
i
Addition of water to this ‘dry’ PrOH then gave the
separation shown in Fig. 3. The effect of water is to
reduce the retention time of both enantiomers but one
much more than the other. Once again, the effect was
entirely reproducible and we switched back and forth
between the two systems several times. Addition of
more water did not seem to improve the separation
Nevertheless, the influence of water on two similar diols
was also investigated. Ordinary hydrobenzoin 2 and its
3,3%-difluorinated derivative 3 were also subjected to
i
HPLC using PrOH with differing water content. At
i
and, in fact, full separation was achieved using PrOH
with just 2% water.
one end of the scale, a solvent system using fresh iPrOH
was used, and at the other extreme the solvent was
saturated with water. Solvent systems with intermediate
water content were also examined.
Although modifiers like diethylamine or trifluoroacetic
acid are well known and mentioned by Daicel for the
OJ column, we had not come across water as a
modifier. Indeed, Daicel do not recommend solvents
other than hexane, ethanol or isopropanol with the OJ
column and specifically warn against the use of water,
and other polar solvents.4
However, water had a negligible effect on the enan-
tioseparation of both diols. Figs. 4 and 5 show the
chromatograms for 2 and 3, respectively, using both
iPrOH with a water content of 0.5% (‘dry’) and 3.5%
(‘wet’).5
i
Figure 4. Separation of ( )-2 using ‘dry’ (left) and ‘wet’ (right) PrOH.

P. Wyatt / Tetrahedron: Asymmetry 12 (2001) 3329–3331
3331
i
Figure 5. Separation of ( )-3 using ‘dry’ (left) and ‘wet’ (right) PrOH.
Conclusion
References
1. Okamoto, Y.; Aburutani, R.; Hatada, K. J. Chromatogr.
1987, 389, 95–102.
2. Matsubara, S.; Hashimoto, Y.; Okano, T.; Utimoto, K.
The enantiomers of 2,2%-dibromohydrobenzoin are
completely resolved on an OJ column with a 1865:200:7
heptane:iPrOH:H2O solvent system and not resolved at
all with a 1810:200:1 solvent system. Isopropanol and
ethanol are very commonly used eluents for HPLC and
although the water content is likely to vary greatly with
these solvents it is scarcely considered as a factor. These
results show that although water may well not be
important, its influence may also be profound. We
speculate that the ortho positioning of the substituent
may well be more important than the nature of the
substituent itself.
Synlett 1999, 1411–1412.
3. The chromophore is very weak at 254 nm.
4. http://www.daicel.co.jp/chiral/e/product/instruction/
oj.htmL
5. Water content of 3.5% is near the saturation point of the
solvent system. Although 500 mL of water may be mixed
i
with PrOH to a total volume of 10 mL (5% v/v), an
emulsion is formed when heptane is then added to form a
i
9:1 mix of heptane:wet PrOH.