P(RNCH2CH2)3N-catalyzed diastereoselective synthesis of oxazolines

Article abstract of DOI:10.1016/S0040-4039(01)01270-9

We report herein that the strong nonionic bases P(RNCH₂CH₂)₃N(R = Me, 1a; i-Pr, 1b) catalyze the diastereoselective synthesis of thirteen oxazolines in the presence of 5-30% of these catalysts. The formation of the oxazolines proceeds under very mild conditions in good to excellent yields with high diastereoselectivity (>95:5) for the trans isomer.

Full text of DOI:10.1016/S0040-4039(01)01270-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6263–6266
P(RNCH2CH2)3N-catalyzed diastereoselective synthesis
of oxazolines
Philip Kisanga,a,b Palanichamy Ilankumarana,c and John G. Verkadea,*
aDepartment of Chemistry, Iowa State University, Ames, IA 50011, USA
bAldrich Chemical Company, 940 W. St. Paul Ave., Dept. 271, Milwaukee, WI 53233, USA
c122B, Cedar Lane, Highland Park, NJ 08904, USA
Received 27 March 2001; revised 10 July 2001; accepted 13 July 2001
Abstract—We report herein that the strong nonionic bases P(RNCH2CH2)3N(R=Me, 1a; i-Pr, 1b) catalyze the diastereoselective
synthesis of thirteen oxazolines in the presence of 5–30% of these catalysts. The formation of the oxazolines proceeds under very
mild conditions in good to excellent yields with high diastereoselectivity (>95:5) for the trans isomer. © 2001 Elsevier Science Ltd.
All rights reserved.
Oxazolines are versatile intermediates in the synthesis
of b-substituted serines1 which are of significant impor-
tance because of their utility in the synthesis of various
antibiotics.2 Thus, the serine moiety constitutes the
primary core structure of antibiotics, such as hypeptin2b
and leucinostatin.3 Ethyl isocyanoacetate, a synthon for
the formation of oxazolines, is relatively acidic and can
be deprotonated by a variety of bases for coupling with
aldehydes to afford oxazolines. However, the lack of
diastereoselectivity of such reactions has rendered this
synthetic route of limited use. Among catalysts that
have been reported to effect the conversion of alde-
hyde/ether isocyanoacetate mixtures to oxazolines are
ZnCl2,4 a ZnCl2/CuCl system,5 NaCN/EtOH,6 and
Cu2O.7 The copper(I) oxide-catalyzed process leads to
the formation of varying ratios (1.5:1.0–0.4:1.0) of
diastereomers7 and this catalyst also induces migration
of the imine double bond with the resultant formation
of two tautomers, giving a complex mixture of prod-
ucts.7 Furthermore, the presence of a,b-unsaturation
leads to Michael addition, rendering this method of
marginally practical utility. Although the NaCN/EtOH
system affords oxazoline in high yields,6 its inability to
induce excellent diastereoselectivity with aldehydes
other than phenyl acetaldehyde restricts its use.8 The
ZnCl2/CuCl catalyzed process also leads to the forma-
tion of diastereomeric mixtures (7:1–1:1, trans:cis).5 As
a result of these poor selectivities, alternative routes to
the b-hydroxy a-amino acids have been developed,9
among which are the condensation of glycine with
aldehydes on Ni(II) complexes9a and an electrophilic
amination reaction.9b Since nickel is highly toxic, this
methodology is not attractive in an industrial setting.
The electrophilic amination process required 2.5–4.2
equiv. of LDA and also 1.5 equiv. of di-t-butylazodi-
carboxylate.9b Moreover, the diastereoselectivity in this
reaction ranges from 94:6–75:25 (trans:cis).9b The use
of excess LDA as a base has been reported to effect the
reaction of ethyl isocyanoacetate with o-anisaldehyde
to give the corresponding oxazoline in 80% yield at
−75°C.9c Enzyme systems10 have also been investigated
for the synthesis of the broad spectrum antibiotic thi-
amphenicol 1a and florfenicol 2a.10c
R
OH
R
P
N
N
R
N
R
NHCOCHCl2
MeO2S
N
2a R = Me
2b R - i-Pr
1a R = OH
1b R = F
The proazaphosphatranes 2a and 2b11 first synthesized
in our laboratories have recently attracted interest as
versatile highly basic12 catalysts and reagents for variety
of useful transformations.13–24 We therefore decided to
explore additional examples of reactions in which
stereoselectivity is an issue. Since oxazolines are ver-
satile intermediates for the synthesis of serine deriva-
tives, which in turn could open a route to interesting
unnatural amino acids, we decided to investigate the
Keywords: oxazolines; proazaphosphatranes; catalysts; diastereoselec-
tivity; nonionic base.
* Corresponding author. Tel.: (515) 294-5023; fax: (515) 294-0105;
e-mail: jverkade@iastate.edu
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01270-9

6264
P. Kisanga et al. / Tetrahedron Letters 42 (2001) 6263–6266
O
isolated in 95% yield.25 Both 2a and 2b afforded similar
yields within experimental error (data for 2a not shown
in Table 1).
1. P(RNCH2CH2)3N
R = Me, i-Pr
2. R'CHO
N
O
EtO
CNCH2CO2Et
R'
1
Comparison of the H and 13C NMR data for product
R' = alkyl, Ar
3a to literature data5 revealed that the trans-oxazoline
ethyl carboxylate was the only product of the reaction.
Attempted reaction of p-fluorobenzaldehyde in
isobutylnitrile afforded complex reaction mixtures.
However, by carrying out the reaction with ethyl
isocyanoacetate at a lower temperature (see Table 1)
for 15 min followed by further reaction at room tem-
perature for 1–1.5 h, p-fluorobenzaldehyde, p-chloro-
3
Scheme 1.
possibility of inducing a diastereoselective reaction
between aldehydes and ethyl isocyanoacetate (Scheme
1).
benzaldehyde,
p-methylsulfonylbenzaldehyde
and
We report here the use of catalytic amounts of the
proazaphosphatranes 2 as catalysts in the synthesis of
oxazoline alkyl carboxylates with high diastereoselectiv-
ity. The reaction of benzaldehyde with ethyl iso-
cyanoacetate in THF in the presence of 20% of 2b at
room temperature for 1 h afforded a product mixture
that contained the trans-oxazoline 3a as the major
pivalaldehyde reacted to afford the corresponding
trans-oxazoline ethyl carboxylates, in addition to small
amounts of the corresponding Knoevenagel products in
the cases of p-methylsulfonylbenzaldehyde and
pivalaldehyde as shown by 1H NMR spectroscopic
analysis. On the other hand aromatic aldehydes bearing
electron donating groups (2,5-dimethylbenzaldehyde,
2,5-dimethoxybenzaldehyde, o-anisaldehyde, 2-naph-
thaldehyde and 2-furaldehyde required room tempera-
ture to afford similar yields (Table 1) with the exception
of 2,5-dimethylbenzaldehyde which consistently formed
large amounts of the Knoevenagel product. However,
by reducing the temperature to −63°C and carrying out
the reaction for 16 h followed by quenching with
MeOH at −40°C, clean formation of the desired oxazo-
line ethyl carboxylate was realized with 2,5-dimethyl-
benzaldehyde. Nitrobenzaldehyde consistently formed a
mixture of the oxazoline and a Knoevenagel product
that could not be separated. The oxazoline ethyl car-
boxylate 3g thus obtained in 97% yield is of significance
because an analogous oxazoline has previously been
used as an intermediate in the synthesis of thiampheni-
col (1a) and florfenicol (2b).10b,c The aliphatic aldehydes
isobutyraldehyde, n-heptaldehyde, and n-butyraldehyde
gave complex mixtures consisting of desired com-
pounds, the Knoevenagel products and other uniden-
tified species.
1
product according to the H NMR spectroscopy. How-
ever, the reaction was not clean and led to the forma-
tion of other uncharacterized side products. No
significant improvement was observed upon reducing
the temperature to −78°C or raising the temperature to
40°C. However, upon changing the solvent to isobuty-
ronitrile, the desired oxazoline ethyl carboxylate 3a was
Table 1. Reaction of aldehydes with ethyl isocyanoacetate
in isobutyronitrile to give 3
Aldehyde
Mol% 2b T, °C (t, min) Yield of 3, %
Benzaldehyde
0.20
25 (60)
3a, 95
3b, 98
3c, 94b
3d, 99
3e, –d
3f, 78
3, 97
p-Fluorobenzaldehyde 0.20
p-Chlorobenzaldehyde 0.20
p-Cyanobenzaldehyde 0.30
p-Nitrobenzaldehyde
p-Anisaldehyde
−5 (15)a
−20 (15)a
−63 (75)
−63 (75)
25 (120)
25 (75)
0.30
0.20
p-Methylsulfonylbenz- 0.30
aldehyde
2,5-Dimethylbenz-
aldehyde
2,5-Dimethoxybenz-
aldehyde
Isobutyraldehyde
Pivalaldehyde
n-Heptaldehyde
(E)-Cinnamaldehyde
n-Butyraldehyde
o-Anisaldehyde
2-Naphthaldehyde
2-Furaldehyde
0.30
0.30
−63 (960)
3h, 93b
3, 91
25 (90)
Acknowledgements
0.05
0.05
0.05
0.20
0.05
0.20
0.20
0.20
25 (60)c
3o, –e
3k, 87
3l, –e
3m, 68
3a, –e
3o, 75
3p, 97
3q, 93
−25 (60)c
−25 (60)c
−25 (75)c
−25 (60)c
25 (60)
We thank the National Science Foundation for the
grant support of this research.
25 (60)
25 (60)
References
a This was followed by stirring at room temperature for 60 min.
b Very small amounts (2–3%) of the cis-isomer were observed by 1H
NMR spectroscopy of the crude reaction mixture.
c The reactants were mixed at −63°C, stirred for 15 min and then at
rt for time periods specified.
1. (a) Hining, H.; Seufer-Waserthal, P.; Weber, H. Tetra-
hedron 1990, 46, 3841; (b) Gaparski, C. M.; Miller, M. J.
Tetrahedron 1991, 47, 5367; (c) Togni, A.; Pastor, S. D.;
Ribs, G. J. Organomet. Chem. 1990, 381, C21.
2. (a) Saeed, A.; Young, D. W. Tetrahedron 1992, 48, 2507;
(b) Shoji, J.; Hinoo, H.; Hattori, T.; Kirooka, K.;
Kimura, Y.; Yoshida, T. J. Antibiot. 1988, 42, 1460.
3. Hukushima, K.; Arai, T.; Mori, Y.; Tsuboi, M.; Suzuki,
M. J. Antibiot. 1983, 36, 1613.
d An inseparable mixture containing the Knoevenagel product was
observed even at −63°C overnight.
e An inseparable complex mixture containing the oxazoline, the Kno-
evenagel product and unidentified materials was observed by 1H
NMR spectroscopy.

P. Kisanga et al. / Tetrahedron Letters 42 (2001) 6263–6266
6265
indicated in Table 1 and stirred for 5 min. To this
solution was added 2.0 mmol of the aldehyde and then
the solution was stirred at this temperature for a fur-
ther 15 min after which it was allowed to stir at room
temperature for 1 h. The reaction mixture was loaded
onto a small silica gel column and eluted with ethyl
ether. Removal of the solvent in vacuo afforded a
crude mixture that was purified by eluting it on a silica
gel column with ether/hexane. The ratio of ether was
increased in 5% increments. The oxazolines eluted with
40% ether in hexane. The more polar oxazolines 3b–e
and 3g were purified by eluting them with incrementally
increasing amounts (5%) of ethyl acetate in hexane.
They eluted 40% ethyl acetate in hexane. Compound
3a: see discussion; 3b: 1H NMR (CDCl3): l 1.34 (t,
3H), 4.30 (q, 2H), 4.575 (dd, 1H), 5.66 (d, 2H), 7.090
(AB q, 2H), 7.31 (AB q, 2H). 13C NMR (CDCl3): l
170.3, 164.1, 161.6, 156.2, 134.8 (d, J=3 Hz), 127.5 (d,
J=8 Hz), 115.9 (d, J=22 Hz), 81.6, 75.5, 62.1, 14.1.
HRMS: Calcd for C12H12FNO5 237.0801, found m/e
(M+) 237.0800; 3c: Isolated with an inseparable Knoev-
enagel by-product. 1H NMR (CDCl3): l 1.35 (t, 3H),
4.32 (q, 2H), 4.55 (dd, 1H), 5.65 (d, 1H), 7.09 (d, 1H),
7.31–7.25 (m, 2H), 7.39–7.36 (m, 2H). 13C NMR
(CDCl3): l 170.3, 156.2, 137.5, 134.7, 130.8, 129.2,
127.0, 81.5, 75.5, 62.2, 14.2. HRMS: Calcd for
C12H12ClNO5 253.0506, found m/e (M+) 253.0509; 3d:
1H NMR (CDCl3): l 1.35 (m, 3H), 4.32 (dq, 2H), 4.56
(dd, 1H), 5.75 (d, 1H), 7.14 (d, 1H), 7.46 (m, 2H), 7.70
(m, 2H). 13C NMR (CDCl3): l 169.9, 169.9, 156.1,
144.2, 132.8, 129.9, 126.1, 118.3, 113.5, 81.0, 75.5, 62.3,
14.1. HRMS: Calcd for C13H12N2O5 244.0848, found
m/e (M+) 244.0848; 3f: The 1H NMR compared favor-
ably with that reported in Liebigs Ann. Chem. 1972,
763, 1; 3g: 1H NMR (CDCl3): l 1.355 (t, 3H), 3.09 (s,
3H), 4.32 (dq, 2H), 4.58 (dd, 1H), 5.79 (d, 1H), 7.18
(d, 1H), 7.57 (d, 2H), 7.99 (m, 2H). 13C NMR
(CDCl3): l 169.9, 156.1, 145.1, 140.7, 128.1, 126.3, 80.9,
75.5, 62.2, 44.4, 14.1. HRMS: Calcd for C13H15NO5S
297.0671, found m/e (M+) 297.0671; 3h: 1H NMR
(CDCl3): l 1.30 (m, 6H), 2.29 (s, 3H), 4.26 (m, 2H),
4.54 (d, 1H), 5.90 (d, 1H), 7.03 (overlapping area, 3H),
7.12 (d, 1H). 13C NMR (CDCl3): l 170.6, 156.6, 136.7,
136.2, 131.9, 130.8, 129.3, 125.9, 79.9, 75.0, 61.9, 21.0,
18.7, 14.2. HRMS: Calcd for C14H17NO3 247.1208,
4. (a) Badr, M. Z. A. B.; Aly, M. M.; Fahmy, A. M.
Bull. Chem. Soc. Jpn. 1981, 54, 1844; (b) Kirihata, M.;
Kawahara, S.; Ichimoto, I.; Ueda, H. Agric. Biol.
Chem. 1990, 54, 753.
5. Ito, Y.; Matsura, T.; Saegusa, T. Tetrahedron Lett.
1985, 26, 5781.
6. Hoppe, D.; Schollkopf, U. Angew. Chem., Int. Ed.
Engl. 1970, 9, 300.
7. (a) Saegusa, T.; Ito, Y.; Konoshita, H.; Tomita, S. J.
Org. Chem. 1971, 36, 3316; (b) Rich, D. H.; Dhaon,
M. K.; Dunlap, B.; Millaer, S. P. F. J. Med. Chem.
1986, 29, 978.
8. Rao, M. N.; Holkar, A. G.; Ayyanger, N. R. J. Chem.
Soc., Chem. Commun. 1991, 1007.
9. (a) Soloshonok, V. A.; Avilov, D. V.; Kukhar, N. S.;
Kochetkov, K. A.; Orlova, S. A.; Pysarevsky, A. P.;
Struchkov, Y. T.; Raevsky, N. I.; Belokon, Y. N. Tetra-
hedron: Asymmetry 1995, 6, 1741; (b) Gunati, G.; Banfi,
L.; Narisano, E. Tetrahedron 1988, 44, 5562; (c)
Colonna, S.; Manfredi, A.; Solladie´-Cavallo, A.; Quaz-
zoti, S. Tetrahedron Lett. 1990, 31, 6185.
10. (a) Clark, J. E.; Fischer, P. A.; Schumacher, D. P.
Synthesis 1991, 891; (b) Wu, G.; Schumacher, D. P.;
Tormos, W.; Clark, J. E.; Murphy, B. L. J. Org. Chem.
1997, 62, 2996; (c) Clark, J. E.; Fischer, P. A.; Schu-
macher, D. P. Synthesis 1991, 891.
11. (a) Schmidt, H.; Lensink, C.; Xi, S.-K.; Verkade, J. G.
Z. Anorg. Allg. Chem. 1989, 578, 75; (b) Wroblewski,
A.; Pinkas, J.; Verkade, J. G. Main Group Chem. 1995,
1, 69; (c) Kisanga, P.; Verkade, J. G., manuscript in
preparation.
12. D’Sa, B.; Kisanga, P.; Verkade, J. G.; Schwesinger, R.
J. Org. Chem. 2000, 65, 5431.
13. D’Sa, B.; Verkade, J. G. J. Org. Chem. 1996, 61,
2963.
14. Ilankumaran, P.; Verkade, J. G. J. Org. Chem. 1999,
64, 3086.
15. Kisanga, P.; McLeod, D.; D’Sa, B.; Verkade, J. G., in
preparation.
16. D’Sa, B.; Kisanga, P.; Verkade, J. G. J. Org. Chem.
1998, 63, 3691.
17. Wang, Z.; Kisanga, P.; Verkade, J. G. J. Org. Chem.
1999, 64, 6559.
18. Kisanga, P.; Verkade, J. G. J. Org. Chem. 1999, 64,
4298.
19. Kisanga, P.; D’Sa, B.; Verkade, J. G. J. Org. Chem.
1998, 63, 10057.
20. Arumugam, S.; McLeod, D.; Verkade, J. G. J. Org.
Chem. 1997, 62, 4927.
21. Tang, J.-S.; Verkade, J. G. J. Org. Chem. 1994, 59,
7793.
22. Yu, Z.; Verkade, J. G., in preparation.
23. Wang, Z.; Verkade, J. G. Heteroatom. Chem. 1998, 9,
687.
24. Wang, Z.; Verkade, J. G. Tetrahedron Lett. 1998, 39,
9331.
1
found m/e (M+) 247.1211; 3i: H NMR (CDCl3): l 1.34
(t, 3H), 3.77 and 3.76 (two s, 6H), 4.29 (dq, 2H), 4.48
(dd, 1H), 5.88 (d, 1H), 6.83 (d, 2H), 6.87 (s, 1H), 7.07
(d, 1H). 13C NMR (CDCl3): l 170.9, 156.1, 153.7,
150.2, 128.2, 113.7, 112.3, 111.6, 79.1, 74.8, 61.7, 55.8,
14.2. HRMS: Calcd for C14H17NO5 279.1107, found m/
1
e (M+) 279.1107; 3k: H NMR (CDCl3): l 0.93 (s, 9H),
1.31 (t, 3H), 4.24 (q, 2H), 4.39 (s, 2H), 6.95 (s, 1H).
13C NMR (CDCl3): l 171.3, 156.7, 89.1, 68.2, 61.6,
33.7, 24.5, 14.1. HRMS: Calcd for C10H17NO5
199.1208, found m/e (M+) 199.1209; 3o: On the basis of
the 1H NMR spectrum, this known compound con-
tained an inseparable 8% of the corresponding Knoeve-
nagel by-product. Because no NMR spectra have been
reported for 3o, we report them here. 1H NMR of 3o
(CDCl3): l 1.33 (t, 3H), 3.79 (s, 3H), 4.29 (q, 2H), 4.51
25. The desired amount of 2b (Table 1) was weighed into a
flask under nitrogen. To this was added 3.0 mL of
isobutyronitrile, followed by 2.0 mmol of ethyl iso-
cyanoacetate. The reaction mixture was placed in a
constant temperature bath adjusted to the temperature

6266
P. Kisanga et al. / Tetrahedron Letters 42 (2001) 6263–6266
(dd, 1H), 5.91 (d, 1H), 6.89 (d, 1H), 6.96 (s, 1H), 7.08
13C NMR (CDCl3): l 170.5, 156.4, 136.1, 133.3, 129.2,
128.1, 127.8, 126.7, 126.6, 125.1, 122.8, 82.4, 75.4, 62.0,
14.1. HRMS: Calcd for C16H15NO3 269.1052, found m/
e (M+) 269.1052; 2q: The 1H NMR spectrum compared
favorably with that reported in J. Org. Chem. 1990, 55,
1649.
(d, 1H), 7.32–7.27 (m, 2H). 13C NMR (CDCl3): l
171.0, 156.3, 156.2, 129.7, 127.1, 126.2, 120.7, 110.6,
79.4, 74.7, 61.7, 55.3, 14.3; 3p: 1H NMR (CDCl3): l
1.33 (t, 3H), 4.30 (q, 2H), 4.69 (dd, 1H), 5.85 (d, 1H),
7.16 (d, 1H), 7.37 (dd, 1H), 7.49 (m, 2H), 7.83 (m, 4H).
.