Camphor-based oxazaphospholanes as chiral templates for the enantioselective synthesis of α-chlorophosphonic acids

Article abstract of DOI:10.1016/S0957-4166(99)00455-3

Constrained camphor-derived oxazaphospholanes have provided an efficient entry for the preparation of enantiomerically enriched α-chlorophosphonic acids. (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(99)00455-3

Pergamon
Tetrahedron: Asymmetry 10 (1999) 4277–4280
Camphor-based oxazaphospholanes as chiral templates for the
enantioselective synthesis of α-chlorophosphonic acids
Giovanni Battista Giovenzana,a Roberto Pagliarin,a Giovanni Palmisano,b Tullio Pilati c and
Massimo Sisti d,∗
aDipartimento di Chimica Organica e Industriale, Via Venezian 21, 20133 Milano, Italy
bDipartimento di Scienze Mediche, Viale Ferrucci 33, 28100 Novara, Italy
cCentro di Studio delle Relazioni tra Struttura e Reattività Chimica-CNR, Via Golgi 19, 20133 Milano, Italy
dDipartimento di Scienze Chimiche Fisiche Matematiche, Via Lucini 3, 22100 Como, Italy
Received 1 October 1999; accepted 14 October 1999
Abstract
Constrained camphor-derived oxazaphospholanes have provided an efficient entry for the preparation of enan-
tiomerically enriched α-chlorophosphonic acids. © 1999 Elsevier Science Ltd. All rights reserved.
Phosphonic acids and their derivatives play an important role in medicinal chemistry as mimics of the
tetrahedral transition state of sp2 SN reactions and as non-hydrolyzable surrogates for the corresponding
carboxylic acids and phosphates. These compounds exhibit interesting and unique properties as antiviral
agents, haptens to elicit catalytic antibodies, peptide analogues, enzyme inhibitors and antibiotics.1
In contrast to the extensively investigated amino phosphoryl compounds, α-halophosphonic acids
have received less attention even though it has been reported that they may function as inhibitors of
some viral and human DNA polymerases.2 A synthetic approach to α-halophosphonic acids with high
enantiomeric excesses would be of significant value considering that the absolute configuration at the
α-position in substituted phosphonic acids has been shown to be important for the biological activity.2,3
To the best of our knowledge, few methods have been reported to date for the asymmetric synthesis of
α-chlorophosphonic acids.4
We describe here a straightforward entry to enantiomerically enriched α-chlorophosphonic acids 1
based on diastereoselective alkylation of a chiral non-racemic chloromethyl phosphonic acid equivalent
(Scheme 1).
We elected to use the camphor derivative 2 as a highly defined template considering its ready
availability in either enantiomeric form. Furthermore, related rigid aminoalcohol backbones have been
widely employed in diastereoselective syntheses.5
∗
Corresponding author. Tel: 39-02-2363469; fax: 39-02-2364369; e-mail: sistimax@icil64.cilea.it
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00455-3

4278
G. B. Giovenzana et al. / Tetrahedron: Asymmetry 10 (1999) 4277–4280
Scheme 1.
The β-aminoalcohol 2 was obtained according to literature methods5b,c starting from (1R)-(−)-
camphorquinone. By reacting compound 2 (24 mmol) with chlorophosphonic acid dichloride (28
mmol) in anhydrous toluene and in the presence of triethylamine (56 mmol) at 60°C for 6 h, the two
diastereoisomeric 2-chloromethyl-1,3,2-oxazaphospholanes 36 and 4 were obtained in the ratio 10:1 as
judged from the intensity of the 31P NMR signals (δP=35.99 and 37.79 ppm, respectively). The two
diastereoisomers could be cleanly separated by flash silica gel chromatography with light petroleum:ethyl
acetate 9:1→3:7 as eluant (75% overall yields). It has been reported7 that the absolute stereochemistry of
1,3,2-oxazaphospholanes derived from ephedrine could be assigned by the deshielding of the proton
1
HC-O-P_O cis to the phosphoryl oxygen in the H NMR spectrum. In our case, the proton at C2
(monoterpene numbering) resonates at 4.37 and 4.47 ppm for the diastereoisomers 3 and 4, respectively.
The major 2-chloromethyl-1,3,2-oxazaphospholane 3 was treated with LDA (1.1 equiv.) at −90°C in
THF to generate the Li+-3−, followed, after 2 h, by addition of benzyl bromide (1.0 equiv.). After standing
at −90°C for 5 h, the reaction mixture was quenched with methanol.
A
31P NMR spectrum of the crude reaction mixture indicated the formation of a 24:1 mixture of the
monoalkylation products of 3. The major diastereoisomer 5d6 was obtained by a two-fold crystallization
step from n-hexane–diethyl ether affording a single product (δP=38.18 ppm) in 60% yield. This was
submitted to X-ray single-crystal structural analysis in order to ascertain the stereochemical outcome of
the alkylation step and to further confirm the absolute stereochemistry at the phosphorus stereocentre
(Fig. 1).
Figure 1. ORTEP II8 plot of 5d with arbitrary numbering. Thermal ellipsoid at 50% probability level; H atoms of methyl groups
are omitted for clarity

G. B. Giovenzana et al. / Tetrahedron: Asymmetry 10 (1999) 4277–4280
4279
In light of the highly diastereoselective alkylation of Li+-3−, we decided to test the generality of this
behaviour. The results obtained in the alkylation of Li+-3− with a variety of alkyl halides are reported in
Table 1.
Table 1
Diastereoselective alkylation of Li+-3−
The diastereofacial selectivity to give the major R-isomer at the newly created stereogenic centre
is probably due to efficient steric hindrance by the N-isopropyl moiety in 3, thereby directing the
electrophile to the less hindered side of Li+-3−.
Treatment of compounds 5a–d in THF with 4N HCl at 60°C promoted removal of the chiral auxiliary9
and the reaction mixture was purified by Dowex® 50W-X8 (H+-form) chromatography to afford the
required α-chlorophosphonic acids 1a–d in good to excellent yields. Comparison of specific rotations of
1a–d with those of the literature4a gave ee values in very good agreement with the de values found for
5a–d.
In summary, we have shown that alkylation of the anion derived from 2-chloromethyl-1,3,2-
oxazaphospholane 3 gives rise to α-chlorophosphonic acids in a straightforward and highly stereoselec-
tive fashion. Finally, the chlorine atom present in 3 represents a useful handle for further manipulation.
Work is in progress in our laboratories to establish the synthetic potential of this novel chiral non-racemic
chloromethyl phosphonic acid equivalent.
X-Ray crystallography of 5d: formula C21H31ClNO2P, Fw=540.30, four-circle Syntex P4 diffracto-
meter, graphite monochromator, Mo-Kα radiation, λ=0.71073 Å, crystal dimensions=0.58×0.48×0.44
mm3, orthorhombic, space group P212121, a=7.2301(6), b=10.3394(8), c=28.739(3) Å, V=2148.4(3) Å3,
Z=4, Dc=1.224 g cm−3, µ(Mo-Kα)=0.267 mm−1, F(000)=848 electrons. Intensity data (5584) collected
in the range 4<2θ<55° with limits h=0, 9; k=0, 13; l=−37, 37; merging R=0.0134, total number of
reflections=4945, observed [I>2σ(I)] 3851. Data reduction with Lorentz and polarization corrections
was carried out by XSCANS;10 no absorption correction was applied. The structure was solved by
direct methods (SIR9211) and refined (SHELX9712) by full-matrix least-squares based on 4945 Fo ;
2
2
2
2
all reflections used with weights w=1/[σ(Fo )+(0.0583P)2] were P=(Fo +2Fc )/3. Non-H atoms were
refined anisotropically; H atoms of the isopropyl moiety were in calculated positions, the others were
refined isotropically. The isopropyl group may be affected by partial rotational disorder; in fact it presents
large displacement parameters (mainly for methyl C13, see Fig. 1), the distance C11–C13 [1.413(5) Å]
is very short and it was impossible to refine H atoms of the group. Neutral atomic scattering factors
and anomalous dispersion effects were those included in SHELX97. The absolute configuration of the
molecule was determined without ambiguity by Flack’s13 method whose value converged to 0.03(6).
Final agreement indices (based on F2) were R=0.0581, Rw=0.1031 (R=0.0381, Rw=0.0951 on observed
reflections) and goodness-of-fit 1.028.

4280
G. B. Giovenzana et al. / Tetrahedron: Asymmetry 10 (1999) 4277–4280
Acknowledgements
We would like to thank Dr. R. Allegrini for his contribution to the work during his time in the
Department. Financial support was provided by MURST (ex 40%).
References
1. (a) The Role of Phosphonates in Living Systems; Hilderbrand, R. L., Ed.; CRC Press: Boca Raton, FL, 1983; (b) Engel, R.
Chem. Rev. 1977, 77, 349–367; (c) Kukhar’, V. P.; Solodenko, V. A. Russ. Chem. Rev. 1987, 56, 859–874.
2. McKenna, C. E.; Khawli, L. A.; Bapat, A.; Harutunian, V.; Cheng, Y.-C. Biochem. Pharmacol. 1987, 36, 3103–3106.
3. Kametani, T.; Kigasawa, K.; Hiiragi, M.; Wakisaka, K.; Haga, S.; Sugi, H.; Tanigawa, K.; Suzuki, Y.; Fukawa, K.; Irino,
O.; Saita, O.; Yamabe, S. Heterocycles 1981, 16, 1205–1242.
4. (a) Hanessian, S.; Bennani, Y. L.; Delorme, D. Tetrahedron Lett. 1990, 31, 6461–6464; (b) Jacquier, R.; Lhassani, M.;
Petrus, C.; Petrus, F. Phosphorus and Sulfur 1993, 81, 83–87; (c) Yuan, C. Y.; Li, S. S.; Wang, G. Q.; Wu, H. Chin. Chem.
Lett. 1993, 4, 753–756.
5. (a) Colombo, L.; DiGiacomo, M.; Brusotti, G.; Milano, E. Tetrahedron Lett. 1995, 36, 2863–2866; (b) Denmark, S. E.;
Rivera, I. J. Org. Chem. 1994, 59, 6887–6889; (c) Bonner, M. P.; Thornton, E. R. J. Am. Chem. Soc. 1991, 113, 1299–1308;
(d) Kouklovsky, C.; Pouilhes, A.; Langlois, Y. J. Am. Chem. Soc. 1990, 112, 6672–6679; (e) Oppolzer, W. Tetrahedron
1987, 43, 1969–2004.
6. Selected data. Compound 3: m.p. 115°C (hexane–Et2O); [α]D25=−29.5 (c 1, CHCl3); 1H NMR (CDCl3, 200 MHz) δ 4.37
dd [1H] (3J=7.6 Hz, 3JH–P=5.2 Hz), 3.89 dd [1H] (2JAB=14.5 Hz, 2JH–P=11.8 Hz), 3.70 dd [1H] (2JAB=14.5 Hz, 2JH–P=6.4
3
Hz), 3.59 dd [1H] (3J=7.6 Hz, JH–P=4.9 Hz), 3.55 m [1H], 2.02 d [1H] (3J=4.5 Hz), 1.85–1.45 m [3H], 1.40 d [3H]
(3J=6.5 Hz), 1.36 d [3H] (3J=6.5 Hz), 1.34 s [3H], 1.05 s [3H], 0.97 m [1H], 0.88 s [3H]; 31P NMR (CDCl3, 81.0 M Hz)
δ 35.99; Elem. anal.: C, 54.85; H, 8.20; N, 4.50; calcd for C14H25ClNO2P: C, 54.99; H, 8.24; N, 4.58. Compound 5d:
m.p. 108°C (hexane–Et2O); [α]D25=+8.45 (c 1.6, CHCl3); H NMR (CDCl3) δ 7.32 m [5H], 4.39 dd [1H] (3J=7.5 Hz,
1
3JH–P=5.3 Hz), 4.23 ddd [1H] (2JH–P=11.9 Hz, 3J=6.2 Hz, 3J=2.4 Hz), 3.74 ddd [1H] (2JAB=14.7 Hz, 3JH–P=5.3 Hz, 3J=2.4
3
3
3
Hz), 3.57 dd [1H] (3J=7.5 Hz, JH–P=4.6 Hz), 3.56 m [1H], 3.05 ddd [1H] (2JAB=14.7 Hz, JH–P=5.7 Hz, J=11.9 Hz),
2.04 d [1H] (3J=5.0 Hz), 1.77–1.55 m [3H], 1.39 d [3H] (3J=6.7 Hz), 1.37 s [3H], 1.34 d [3H] (3J=6.6 Hz), 1.08 s [3H],
0.97 m [1H], 0.90 s [3H]; 31P NMR (CDCl3) δ 38.18; Elem. anal.: C, 63.65; H, 7.97; N, 3.50; calcd for C21H31ClNO2P:
C, 63.71; H, 7.89; N, 3.54.
7. Jommi, G.; Miglierini, G.; Pagliarin, R.; Sello, G.; Sisti, M. Tetrahedron 1992, 35, 7275–7288.
8. Johnson, C. K. ORTEP II, 1976 Report ORNL-5138. Oak Ridge National Laboratory, Tenn., USA.
9. The chiral auxiliary was recovered almost quantitatively with unchanged enantiomeric excess.
10. Fait, J. XSCANS User Manual, 1991. Siemens Analytical X-Ray Instruments, Inc., Madison, Wisconsin, USA.
11. Altomare, A.; Cascarano, G.; Giacovazzo, G.; Guagliardi, A.; Burla, M. C.; Polidori G.; Camalli, M. J. Appl. Cryst. 1994,
27, 435.
12. Sheldrick, G. M. SHELX97. Program for the Refinement of Crystal Structures, 1997. Univ. of Göttingen, Germany.
13. Flack, H. D. Acta Cryst. 1983, A39, 876–881.