60 Dimukhametov et al.
group would have to adopt an unfavorable axial po-
sition in the transition state. It should be noted that,
in the case of the late transition state realization for
this reaction, the fact of the diequatorial position of
substituents Ph and Et in the final product suggests
the intramolecular re-face attack of the P(III) atom
on the C N bond. The attack on the si face is less
favorable, because it results in a less favorable final
product with the axial–equatorial positions of sub-
stituents Et and Ph.
The preferable formation of stereoisomers A and
C with the P O group in the equatorial position is
most probably explained by the fact that in the salts
9, the Arbuzov reaction is realized by chloride anion
attack at the carbon atom of the sterically less loaded
equatorial P O C fragment.
Though, as it was indicated above, the share
of the reaction direction starting from the proto-
nation of the phosphorus atom in phosphites 7 in
the course of the reaction of ꢀ-aldiminoalcohols 2
with chlorophosphites 3 and 4 in the overall reaction
scheme is small, in this case, similar stereochemical
considerations are also valid.
General Procedure for Reaction of P(III)
Chlorides 3,4 with ꢀ-Aldiminoalcohols 2a,b
Into a stirred solution of chlorophosphite 3 or 4
(10 mmol) in dry chloroform (10 ml), a solution
of 2a or 2b (10 mmol) in chloroform (5 ml) was
added dropwise at room temperature. Stirring was
continued for 1 h. Chloroform was evaporated and
the oily residue was submitted to a column chro-
matographic separation to give stereoisomeric 1,4,2-
oxazaphosphorines 5 or 6.
2-(ꢀ-Chloroethyl)-2-oxo-3-phenyl-1,4,2-oxazaphos-
phorine (5). Eluant, toluene–acetone 2:1; overall
yield 83%. Diastereoisomer A: colorless solid; mp
78–79◦C (from benzene–hexane, 1:1); 31P NMR
(CD3CN) δ 13.7. Anal. Calcd. for C11H15NO3PCl: C,
47.91; H, 5.44; N, 5.08; P, 11.25; Cl, 12.88. Found:
C, 47.88; H, 5.40; N, 4.91; P, 11.15; Cl, 12.79. Di-
astereoisomer B: colorless solid; mp 107–107.5◦C
(from acetone); 31P NMR (CD3CN) δ 16.9. Anal.
Found: C, 48.18; H, 5.25; N, 5.17; P, 11.41; Cl, 12.65.
2-(ꢀ-Chloroethyl)-2-oxo-3-phenyl–5-ethyl-1,4,2-
oxazaphosphorine (6) from Chlorophosphite 3.
Eluant, toluene–acetone 4:1; overall yield 88%.
Epimer C: colorless solid; mp 104–105◦C (from
EXPERIMENTAL
benzene–dichloromethane, 4:1); [αa]D = +115.9◦
20
Materials and Spectroscopy
(C, 2.71, CH3OH); 31P NMR (CD3CN) δ 13.6. Anal.
Calcd. for C13H19NO3PCl: C, 51.40; H, 6.26; N, 4.61;
P, 10.21; Cl, 11.70. Found: C, 51.49; H, 6.10; N, 4.53;
P, 10.02; Cl, 11.58. Mixture of epimers (C:D = 1:10):
All syntheses were performed under an atmosphere
of dry argon. All solvents and starting reagents
were distilled immediately prior to use. Commercial
R-(−)-2-aminobutanol-1 (1b) has [α]D = −7.01◦
20
20
20
colorless oil; nd = 1.5302; [α]D = +16.2◦ (C, 1.69,
CH3OH); 31P NMR (CD3CN) δ 17.4 (major), 13.5
(minor). Anal. Found: C, 51.55; H, 6.38; N, 4.73; P,
10.01; Cl, 11.89.
(neat). R-(+)-N-benzylidene-2-aminobutanol-1 (2b)
20
derived from benzaldehyde and 1b has [α]D
=
+28.03◦ (C, 13.25, CH3OH), lit. [13] [α]D = +39.30◦.
The optical purity of 2b is 71.3%. Column chro-
matography and TLC were performed on silica gel
L 100/160 ꢃ and on Silufol 254 plates, respectively.
31P NMR spectra were recorded with a Bruker MSL-
400 spectrometer at 162 MHz (external standard 85%
H3PO4). 1H NMR spectra were obtained in CD3CN on
a Bruker WM-250 spectrometer at 250 MHz. Optical
activity measurements were performed with a Pola-
mat A Carl–Zeiss polarimeter.
20
2-(ꢀ-Chloroethyl)-2-oxo-3-phenyl–5-ethyl-1,4,2-
oxazaphosphorine (6) from Chlorophosphite 4.
Eluant, toluene:acetone 4:1; overall yield 62%.
Physico-chemical constants, and spectral data of
the pure epimer C are identical with those derived
from chlorophosphite 3.
The X-ray diffraction data for the crystals were
collected on a CAD4 Enraf–Nonius automatic diff-
ractometer. Further details consisting of the molecu-
lar and crystal structures of the studied compounds,
as well as the crystal data, details of structure deter-
mination, atomic coordinates, and bond lengths and
angles will be published elsewhere. Crystallographic
data (excluding structure factors) for the structures
reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre. All figures
were made using the program PLATON [14].
CONCLUSIONS
Thus, we have found a new access to stereoselective
synthesis of ꢁ-aminophosphonates based on the re-
action of chiral ꢀ-aldiminoalcohols with chlorophos-
phites. The reactions take place under mild condi-
tions with good yields, and this makes it possible
to introduce a wide range of substituents at the ꢁ-
position to the phosphorus atom. The intramole-
cular pathway of the formation of the P C bond in
the P C N fragment allows maximal control of the