94 Li et al.
EXPERIMENTAL
Preparation of Derivatizing Agent 1
According to the method in reference [10], the chi-
ral derivatizing agent 1 was obtained as a light yel-
low solid when PCl3 and BINOL were refluxed un-
der dry N2 for several hours (monitored by TLC).
Yield 98.7%, 31P = 178.55. The crude product was
used directly in the following reaction without fur-
ther purification because of its high sensitivity to-
wards moisture.
SCHEME 1
which is useful for the ee determination of the chi-
ral substrate [8]. However the H H or H X cou-
pling pattern often makes the analysis difficult be-
cause of the overlapping of signals. With respect to
the use of 31P nuclei, the chemical shift dispersions
are larger and the nuclei are very sensitive to small
structural changes in diastereomeric adducts. Espe-
cially when broad band-proton decoupling is used,
most of the spectra are very simple. Thus, 31P NMR
methods have become very popular because of these
attractive features. Several phosphorus derivatizing
agents (e.g. I–IV) have been developed in recent
years [9].
Synthesis of Diastereomeric Adducts 3a and 3b
In a 5 ml reaction flask, 0.2 mmol of CDA 1, 3 ml
of dry toluene, and 0.18 mmol of alcohol or amine
2 were mixed under N2. Then a mixture of 1 ml of
toluene and 0.2 mmol of Et3N was added drop wise
with vigorous stirring of the mixture. After 30 min,
the reaction mixture was filtered. Removal of the sol-
vent under reduced pressure gave the crude product,
which was directly examined by 31P spectroscopy
without further purification.
RESULTS AND DISCUSSION
Associated with the use of CDAs, two of the major
problems are the observation and resolution of ap-
propriate signals in complex NMR spectra and the
potential for asymmetric induction during the for-
mation of diastereomers. In order to test the po-
tential of asymmetric induction we chose several
racemic (entries 1–11, Table 1) and enantiomerically
enriched (entries 12–14, Table 1) alcohols or amines
as the reaction substrates. The findings are summa-
rized in Table 1.
As detailed in Table 1, we found that no ap-
parent kinetic discrimination was observed for the
aromatic alcohols (entries 1–7). Moreover, the value
of 1 (5.10–8.38 ppm) is larger than the ones ob-
tained from the other chiral phosphorus derivatizing
agents reported previously. For example when rac-
␣-methylphenylmethanol was used as the substrate,
the 1 (6.88 ppm) was much larger than the results
determined by use of other chiral phosphorus deriva-
tizing agents [11–14], which ranged from 0.10 ppm
to 1.40 ppm.
In this paper, we reported a simple and highly
efficient 31P NMR method for the ee determination
of chiral alcohols and amines (Scheme 1) based on
the use of the chiral phosphorus derivatizing agent 1,
which is easily prepared from (S)-1,1-bis-2-naphthol
(BINOL) and PCl3. Derivatizing agent 1 reacted with
chiral alcohols or chiral amines 2 in the presence
of triethylamine to form the diastereomeric adducts
3a and 3b. The adducts 3a and 3b were not further
purified and were directly recorded by 31P NMR spec-
troscopy (Scheme 2).
However, for the chiral aliphatic alcohols, great
asymmetric induction was observed (entries 8–10).
Furthermore, we found that a group substitued on
the chiral center had a significant effect on the 1 .
For instance, the values of 1 decreased from 0.44
ppm to 0.10 ppm with an increase of steric hindrance
(entries 8–10). On the other hand, there was also no
good base-line separation (entry 10).
SCHEME 2
Li, Kang Ying
