Synthesis of substituted cyclopropylphosphonates by Michael Induced Ring Closure (MIRC) reactions

Article abstract of DOI:10.1055/s-2002-32587

A variety of cyclopropylphosphonates were prepared in moderate to good yields by Michael Induced Ring Closure of trialkyl phosphites with the corresponding β-bromoalkylidene cyanoacetates and malonates.

Full text of DOI:10.1055/s-2002-32587

LETTER
1089
Synthesis of Substituted Cyclopropylphosphonates by Michael Induced Ring
Closure (MIRC) Reactions
S
ynthesis of Subs
h
titute
d
Cyclo
r
propylp
i
hospho
s
nates tian V. Stevens,* Gino Van Heecke, Carmen Barbero, Krystyna Patora, Norbert De Kimpe, Roland Verhé
Department of Organic Chemistry, Faculty of Agricultural and Applied Biological, Sciences, Ghent University, Coupure links 653,
9000 Gent, Belgium
Fax +32(9)2646243; E-mail: Chris.Stevens@rug.ac.be
Received 4 April 2002
Here, we wish to report a straightforward synthesis of sub-
Abstract: A variety of cyclopropylphosphonates were prepared
stituted cyclopropylphosphonates using the MIRC meth-
odology. Despite the multitude of methods known in the
in moderate to good yields by Michael Induced Ring Closure of
trialkyl phosphites with the corresponding -bromoalkylidene
literature, the substitution pattern that can be obtained,
and even further elaborated, using the MIRC methodolo-
gy is unprecedented, to the best of our knowledge.
cyanoacetates and malonates.
Key words: Michael additions, ring closure, phosphorylations,
cyclizations, carbocycles
Although the reaction of electron poor allyl halides bear-
ing one electron withdrawing group with phosphites has
been studied extensively during the study of the Arbuzov
reaction, the reaction on doubly activated allyl halides has
not been investigated. Therefore, it was decided to study
the reaction of phosphites with double activated alkyl-
idene acetates since the reaction could either give rise to
the corresponding phosphonates by an Arbuzov reaction
or to the corresponding cyclopropylphosphonates by a
MIRC reaction.
For many years cyclopropylphosphonates and cyclopro-
pylphosphonic acids have been the subject of interest for
a variety of reasons. They have been prepared as mimics
of 1-aminocyclopropane carboxylic acid (ACC) with a
high inhibitory activity for the ACC-deaminase and ala-
nine racemase,1,2 as analogues of (–)-allonorcoronamic
acid,3 as structural moieties of nucleotides,4 as potential
herbicides or plant growth regulators,5 as potential insec-
ticides,6 as phosphonic analogues of the antidepressant
minalcipran7 and as constrained analogue of the GABA
antagonist phaclophen.8 Therefore, a variety of methods
have been developed to prepare cyclopropylphosphonates
such as the phosphonylation of halogenated cyclopro-
panes,9 the decomposition of diazophosphonates,10,11 the
cyclopropanation of vinylphosphonates,12–15 electrochem-
ical methods,16–18 the addition of phosphites to iminium
salts,19,20 fragmentation and rearrangement of epoxy-
ethylphosphonates,21 alkylation of benzylphosphonate
carbanions with 1,2-dibromoethane5,22 and the addition of
fumarates to phosphonylated sulphonium ylides.23
The alkylidene malonates, alkylidene malonitriles, and
alkylidene cyano acetates 3 were prepared by Knoevena-
gel condensation of an aldehyde and a doubly activated
methylene compound 2 (malonate or malonitrile)
(Scheme 1). When volatile aldehydes were used, a three-
fold excess was utilized. The Michael acceptors 3 were
obtained in high yield (56-93%) after azeotropic removal
of water during the reaction.26 These were then subse-
quently treated with NBS in carbon tetrachloride under
UV irradiation in the presence of benzoyl peroxide in or-
der to introduce the bromo substituent at the allylic posi-
tion.27 The bromination reaction needed a careful
monitoring by TLC or NMR since prolonged heating of
the reaction mixtures often lead to decomposition of the
brominated compounds with a considerable loss of de-
sired end product 4.28 Although the reaction gives good
results in most of the cases, sometimes it occurred that the
On the other hand, the Michael Induced Ring Closure
(MIRC) reaction is a well established methodology for the
construction
of
cyclopropanes
using
several
nucleophiles24 and for cycloalkanes in general.25
Scheme 1
Synlett 2002, No. 7, 01 07 2002. Article Identifier:
1437-2096,E;2002,0,07,1089,1092,ftx,en;G08502ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

1090
C. V. Stevens et al.
LETTER
reaction was difficult to reproduce and that more NBS Table 1 Reaction Conditions, Yields and Physical Constants of
Cyclopropylphosphonates 7.
needed to be added with additional irradiation in order to
obtain the required product (cf. monitoring). Purification
R1
R2
R3 R4
Ya
Z
Condi- Bp
Yield
[%]
of the brominated products was not performed and some
impurities were removed during purification after the ring
closure.
tions
[°C/mm
[°C/h] Hg] or Rf
(flash)
The -bromoalkylidene malonate derivatives 4 were then
treated with trialkyl phosphites under nitrogen atmo-
sphere, which lead to the isolation of cyclopropylphos-
phonates in moderate to good yields in the case of tertiary
allyl halides (Scheme 2).29,30
a
Me
Me
H
Me
E
E
E
20/8
128/0.2
0.43b
68
72
b
c
d
e
Me
Me
Me
Et
Me
Me
Et
H
H
H
H
Me
Me
Me
Me
CN 20/4
148/0.9
117/0.1
55e
20
CN CN 20/2
However, the formation of the cyclopropylphosphonates
turned out to be strongly dependent upon the primary, sec-
ondary, or tertiary character of the reacting allyl halides.
The reaction of trialkyl phosphites with primary allyl ha-
lides led to the formation of 2-phosphonoalkylidenes via
an Arbuzov reaction. It is obvious that in the case of pri-
mary electrophilic allyl halides the Arbuzov reaction is fa-
vored while with secondary and tertiary substrates the
Michael addition was the predominant reaction. Also,
steric effects at the alkene seemed to play an important
role so that the presence of an additional substituent (R3 =
not H) sterically hindered the Michael addition. On the
other hand, treatment with less nucleophilic tris(ethylth-
io)phosphites resulted in cyclopropanation if strongly ac-
tivated substrates such as methyl 4-bromo-2-cyano-4-
methyl-2-pentenoate were used (Table 1, entry b).
E
E
E
E
20/:10 126/0.05 58e
Et
50/9
128/0.1
0.35c
51
40
f
Penta meth- H Me
ylene
E
E
20/12 147/0.05 30
50/9 76
0.93c
g
h
i
Me
Me
Me
Me
Me
Me
H
H
H
Et
E
E
E
E
E
E
20/12 140/0.05 68
i-Pr
20/12 123/0.1
58
-
d
ClCH2
CH2
20/24
j
i-Pr
H
H
Me
E
E
20/24 135/0.3
0.4c
76
76
k
l
Et
Et
Et
Et
H
H
H
Me
Et
E
E
E
E
CN 110/5
50/9
0.32c
44e
38
E
0.49
The different behavior of primary, secondary or tertiary
allyl halides can be explained by the proposed reaction
pathway leading to the cyclopropane derivatives
(Scheme 2). The mechanism involves a Michael addition
of the phosphorus nucleophile followed by an intramolec-
ular nucleophilic substitution with formation of an inter-
mediate phosphonium cyclopropane 6, which upon
expulsion of an alkyl halide afforded the cyclopropyl
phosphonate 7. In the case of 7k for example, the two di-
astereomers were formed and could be observed easily in
m Me
Me
Me Me
Et Me
CN 120/1
CN 120/1
127/0.1
138/0.1
72e
67e
n
H
a E = COOMe.
b Eluent: Petroleum ether–EtOAc, 20:80.
c Eluent: Petroleum ether–EtOAc, 40:60.
d Decomposed upon distillation.
e Mixture of diastereomers (approximately 1:1 ratio).
Scheme 2
Synlett 2002, No. 7, 1089–1092 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Synthesis of Substituted Cyclopropylphosphonates
1091
the 31P NMR-spectrum showing two signals at 21.50 and
21.27 ppm in a 1:1 ratio (CDCl3). No efforts were made to
separate the diastereomers.
(8) Hanessian, S.; Cantin, L.-D.; Roy, S.; Andreotti, D.;
Gomtsyan, A. Tetrahedron Lett. 1997, 38, 1103.
(9) Hirao, T.; Hagihara, M.; Agawa, T. Bull. Chem. Soc. Jpn.
1985, 58, 3104.
The occurrence of competitive pathways involves the Ar-
buzov reaction versus a MIRC reaction. This difference
can be explained by the general observation that primary
and allyl halides furnish phosphonates upon reaction with
phosphites, while tertiary compounds are unreactive.
Only a few secondary halides reacted satisfactorily. These
results are also in agreement with the SN2 nature of the Ar-
buzov reaction.
(10) Seyferth, D.; Marmor, R. S.; Hilbert, P. J. Org. Chem. 1971,
36, 1379.
(11) Regitz, M.; Scherer, H.; Anschütz, W. Tetrahedron Lett.
1970, 10, 753.
(12) Minami, T.; Yamanouchi, T.; Tokumasu, S.; Hirao, I. Bull.
Chem. Soc. Jpn. 1984, 57, 2127.
(13) Midura, W. H.; Krysiak, J. A.; Wieczorek, M. W.; Majzner,
W. R.; Mikolajczyk, M. Chem. Commun. 1998, 1109.
(14) Midura, W. H.; Krysiak, J. A.; Mikolajczyk, M. Tetrahedron
1999, 55, 14791.
(15) Yamazaki, S.; Takada, T.; Imanishi, T.; Moriguchi, Y.;
Yamabe, S. J. Org. Chem. 1998, 63, 5919.
(16) Jubault, P.; Goumain, S.; Feasson, C.; Collignon, N.
Tetrahedron 1998, 54, 14767.
(17) Goumain, S.; Jubault, P.; Feasson, C.; Collignon, N.
Synthesis 1999, 1903.
(18) Goumain, S.; Jubault, P.; Feasson, C.; Collignon, N.
Tetrahedron Lett. 1999, 40, 8099.
(19) Fadel, A.; Tesson, N. Eur. J. Org. Chem. 2000, 2153.
(20) Fadel, A.; Tesson, N. Tetrahedron: Asymmetry 2000, 11,
2023.
(21) Griffin, C. E.; Kraas, E.; Terasawa, H.; Griffin, G. W.;
Lankin, D. C. J. Heterocycl. Chem. 1978, 15, 523.
(22) Nasser, J.; About-Jaudet, E.; Collignon, N. Phosphorus
Sulfur Silicon 1990, 54, 171.
(23) Kondo, K.; Liu, Y.; Tunemoto, D. J. Chem. Soc., Perkin
Trans. 1 1974, 1279.
(24) Verhé, R.; De Kimpe, N.; De Buyck, L.; Courtheyn, D.; Van
Caenegem, L.; Schamp, N. Bull. Soc. Chim. Belg. 1983, 93,
371.
Primary tetrasubstituted allyl halides gave rise to aliphatic
phosphonates while cyclopropylphosphonates were
formed with trisubstituted allyl halides. In addition, sub-
stitution of an ester by a stronger electron withdrawing
group, for example a nitrile group, favored Michael addi-
tion leading to cyclopropanation. The presence of two
electron withdrawing groups seems to be essential in or-
der to obtain cyclopropanes.
Treatment of primary and secondary monoactivated allyl
bromides with trialkyl phosphites did not lead to cyclo-
propanation, but gave rise to the corresponding aliphatic
phosphonates, while tertiary allyl bromides did not react.
The proposed reaction mechanism has been supported by
kinetic studies where the influence of the nature of the
halogen and the electron withdrawing groups, the substi-
tution pattern of the allyl halide and the nucleophilicity of
the trivalent phosphorus compounds have been studied
(Table 2).31 The kinetic results show that the reaction is of
second order; first order in allyl halide and first order in
nucleophile. The rate-determining step turned out to be
the addition of the phosphorus nucleophile at the double
bond as proven by the enhancement of the reaction rate
with stronger electrophilicity of the allyl halides and the
remarkable decrease in reaction rate when the addition is
sterically more hindered by the alkyl substitution. In addi-
tion, the reaction rate is not very much dependent on the
nature of the leaving group.
(25) Little, R. D.; Verhé, R.; Monte, W. T.; Nugent, S.; Dawson,
J. R. J. Org. Chem. 1982, 47, 362.
(26) Verhé, R.; De Kimpe, N.; De Buyck, L.; Courtheyn, D.;
Schamp, N. Bull. Soc. Chim. Belg. 1978, 87, 215.
(27) General Procedure for the Bromination of
Alkylidenemalonates or Cyanides. A solution of 10 mmol
of Knoevenagel adduct 2, 13 mmol of NBS and 10 mg of
BPO in 5 mL of CCl4 was refluxed under UV irradiation for
approximately 2 hours. The end of the reaction was
monitored carefully by 1H NMR. Evaporation of the solvent
gave almost pure bromoalkylidene malonates or cyanides 4
in good yields.
In summary, a straightforward entry is described into a
range of functionalised cyclopropylphosphonates by a
MIRC reaction of phosphites with doubly activated tertia-
ry allyl halides, which can be valuable for Wadsworth–
Emmons reactions (R3 = H) for the introduction of cyclo-
propylidene units.
(28) Spectral data of 4k: 1H NMR [CDCl3, (ppm)]: 1.06 (6 H, t,
J = 7.3 Hz, Me2), 2.30 (4 H, m, CH2), 3.91 (3 H, s, OMe),
7.74 (1 H, s, CH=). 13C NMR [CDCl3, (ppm)]: 10.31 (Me),
34.06 (C 2); 42.64 (CBr), 53.71 (OMe); 106.37 (=C);
113.31 (CN); 161.83 (CH=); 162.74 (COOMe). IR (cm–1):
2231 (CN), 1740 (C=O), 1621 (C=C). MS (m/z) (%): 234
(0.5); 182(12); 181(100); 148(60); 122(12); 121(29);
107(14); 94(14); 84(13); 49(12). Anal. Calcd. for
C10H14BrNO2: C 46.17; H 5.42; N 5.38. Found: C 46.35; H
5.53; N 5.02.
References
(1) Erion, M. D.; Walsh, C. T. Biochemistry 1987, 26, 3417.
(2) Groth, U.; Lehmann, L.; Richter, L.; Schöllkopf, U. Liebigs
Ann. Chem. 1993, 427.
(3) Hercouet, A.; Le Corre, M.; Carboni, B. Tetrahedron Lett.
2000, 41, 197.
(4) Hah, J. H.; Gil, J. M.; Oh, D. Y. Tetrahedron Lett. 1999, 40,
8235.
(5) Diel, P. J.; Maier, L. Phosphorus and Sulfur 1984, 20, 313.
(6) Reid, J. R.; Marmor, R. S. J. Org. Chem. 1978, 43, 999.
(7) Duquenne, C.; Goumain, S.; Jubault, P.; Feasson, C.;
Quirion, J.-C. Org. Lett. 2000, 2, 453.
(29) General Procedure for the Preparation of the
Cyclopropylphosphonates. A mixture of 2.5 mmol
bromoalkylidenemalonate or -cyanide 4 and 3 mmol of
trialkyl phosphite was stirred under nitrogen atmosphere for
the appropriate time (Table 1). The volatile compounds
formed and the excess of phosphite were then removed by
evaporation under vacuum and the mixture was purified by
flash chromatography or by distillation.
(30) Spectral data of 7a: 1H NMR [CDCl3, (ppm)]: 1.24 (3 H, d,
J= 1.3 Hz, Me); 1.58 (3 H, s, Me); 1.86 (1 H, d, J = 1.9 Hz,
CH-P); 3.76 (3 H, s, OMe); 3.77 (3 H, d, J =1.9 Hz, OMe);
Synlett 2002, No. 7, 1089–1092 ISSN 0936-5214 © Thieme Stuttgart · New York

1092
C. V. Stevens et al.
LETTER
3.78 (3 H, d, J = 1.6 Hz, OMe); 3.81 (3 H, s, OMe). 13C NMR
[CDCl3, (ppm), JP-C(Hz)]: 19.53 (Me, 5); 22.21 (Me, 5.1);
27.65 (CHP, 188); 30.38 (CMe2, 3.6); 43.56 [C(COOMe]2,
3.6); 52.15 (OMe, 6.1); 52.65 (OMe); 52.99 (OMe, 4.9);
53.03 (OMe); 166.64 (COOMe, 123.3); 167.92 (COOMe).
31P NMR referenced to H3PO4 [CDCl3, (ppm)]: 25.00. IR
(cm–1): 1256 (P=O), 1733 (C=O). MS (m/z) (%): 294 (M+,
0.5); 235(12); 231(20); 229(30); 220(8); 203(21); 186(9);
185(86); 153(100); 125(12); 110(14); 109(18); 93(11);
79(12). Anal. Calcd. for C11H19O7P: C 44.90; H 6.51.
Found: C 44.65; H 6.63.
(31) For details of Reaction rates see Table 2.
Table 2 Kinetic Data of the Reactions of the electrophilic Allyl
Halides 4 with Trivalent Nucleophiles to 7 at 35 °C.
Entry
X
K 1/mol.s
Rel. K
1.00
a
a
b
c
d
e
f
Br
Cla
Br
Br
Br
Br
Br
Br
Br
Br
95.8 10–6
57.6 10–6
6.6 10–3
0.60
68.89
346.56
0.30
33.2 10–3
28.9 10–6
8.2 10–6
0.19
14.6 10–6
176.1 10–6
139.0 10–6
16.1 10–6
0.15
g
h
I
2.88
1.45
0.17
a The chloro derivative was prepared according to the literature pro-
cedure by condensation of the chloroaldehyde and the malonate us-
ing titanium (IV)chloride and pyridine.24
Synlett 2002, No. 7, 1089–1092 ISSN 0936-5214 © Thieme Stuttgart · New York