Mild and efficient Cs2CO3-promoted synthesis of phosphonates

Article abstract of DOI:10.1016/j.tetlet.2003.09.045

A mild and convenient synthesis for phosphonates using cesium carbonate (Cs₂CO₃), tetrabutylammonium iodide (TBAI) and DMF was developed at room temperature. Numerous dialkyl phosphites were screened using a diverse array of alkyl halides and these reaction conditions were found to be highly efficient producing various phosphonates exclusively in moderate to high yields.

Full text of DOI:10.1016/j.tetlet.2003.09.045

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 8617–8621
Mild and efficient Cs2CO3-promoted synthesis of phosphonates
Richard J. Cohen, Daniel L. Fox, Jarrod F. Eubank and Ralph Nicholas Salvatore*
Department of Chemistry, Western Kentucky University, 1 Big Red Way, Bowling Green, KY 42101-3576, USA
Received 7 August 2003; revised 5 September 2003; accepted 5 September 2003
Abstract—A mild and convenient synthesis for phosphonates using cesium carbonate (Cs2CO3), tetrabutylammonium iodide
(TBAI) and DMF was developed at room temperature. Numerous dialkyl phosphites were screened using a diverse array of alkyl
halides and these reaction conditions were found to be highly efficient producing various phosphonates exclusively in moderate
to high yields.
© 2003 Elsevier Ltd. All rights reserved.
New or improved methods for phosphonate synthesis
continue to attract much attention because phospho-
nates display a multitude of robust biologically impor-
tant properties serving as natural products,1 analogues
of phosphates,2 phosphonopeptides,3 amino acid ana-
logues,4 and as pro-drugs.5 Given their ubiquitous use
as pharamacological agents6 and as crucial synthetic
intermediates7 (e.g. Wadsworth–Emmons and related
reactions), construction of the PꢀC bond still remains a
formidable challenge. Traditionally, organophospho-
nates are prepared via a Michaelis–Arbuzov8 or a
Michaelis–Becker reaction9 utilizing the nucleophilic
properties of tervalent P(III) compounds (e.g. trialkyl
phosphites or alkali metal salts of dialkyl phosphites) in
the presence of alkyl halides. Depending on the method
of choice, these conventional reaction conditions are
often not convenient requiring elevated temperature,10
the use of strong anhydrous base,11 and very long
reaction times.12 Moreover, these procedures often lead
to a complicated mixture of side products or result in
poor yield of the phosphonate.13 Therefore, to circum-
vent these aforementioned chronic problems, numerous
other methods have been investigated.14 Nonetheless,
each of these methods lack in generality prompting us
to embark on a milder and improved procedure better
suited toward the synthesis of phosphonopeptides and
phosphonotyrosine (pTyr)-containing peptidomimetics.
Such mimetics are widely used for designing biological
probes to modulate aberrant cellular signaling for the
treatment of cancer and other diseases.
In the presence of cesium carbonate and tetra-
butylammonium iodide (TBAI) at room temperature
H-phosphonate anion 2, was smoothly generated from
dialkyl phosphite (1) in-situ. Subsequent alkylation of
anion 2 with an alkyl halide produced the correspond-
ing phosphonate 3 exclusively using anhydrous N,N-
dimethylformamide (DMF) as the solvent of choice
(Scheme 1).
Initially, we undertook an optimization study using
dimethyl phosphite (4) and benzyl chloride as model
coupling partners. At the outset of this work, we began
our approach by screening a variety of different bases
for the efficient synthesis of phosphonate 5. Numerous
alkali metal carbonates were probed including lithium,
sodium, potassium, rubidium and cesium carbonate. Of
the various alkali metal carbonates examined, cesium
carbonate was the most successful base (Table 1, entry
5) delivering dimethyl benzylphosphonate (5) in highest
yield.15 Other carbonate bases were not as effective for
the same transformation (Table 1, entries 1–4). We
attribute this high selectivity to the unprecedented
‘cesium effect’.16 Cesium bases have enjoyed much suc-
cess in facilitating numerous alkylations efficiently.17
This methodology therefore, we believe, further con-
tributes to existing cesium base promoted transforma-
tions.17 As further indication, we found that cesium
hydroxide in the presence of activated molecular sieves,
Keywords: phosphonate; dialkyl phosphite; cesium carbonate; tetra-
butylammonium iodide.
* Corresponding author. Tel.: +1-270-745-3271; fax: +1-270-745-5361;
e-mail: ralph.salvatore@wku.edu
Scheme 1.
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.09.045

8618
R. J. Cohen et al. / Tetrahedron Letters 44 (2003) 8617–8621
Table 1. Use of various bases in Cs2CO3-promoted alkyla-
tion of 4 with benzyl chloride
Table 2. Phosphonate formation of 6 utilizing various
alkyl halides
also proved to be suitable, however, a lower product
yield was noticed (entry 6). Although not shown in the
table, the use of organic bases such as Et3N and DBU
displayed poor conversions mimicking product yields
very similar to the starting entry.
Interestingly, the inclusion of tetrabutylammonium
iodide (TBAI) as a phase transfer catalyst was found to
have a marked effect on producing high product yields.
Therefore, to gain further insight the crucial role TBAI
plays on the course of the reaction, we decided to
pursue an important comparative study. As shown in
Scheme 2, in the presence of TBAI, the desired phos-
phonate 5 was acquired in nearly quantitative yield
after 24 h.18 However, using the same reaction condi-
tions in the absence of TBAI, the reaction appeared
sluggish to consume 4, resulting in a poor 31% yield of
5 after the same time period.
72 h (entry 6). In addition, we found introduction of
the isobutyl moiety also proved successful (entry 7).
Furthermore, other sterically demanding secondary
bromides including isopropyl bromide, 2-bromobutane
and cyclohexyl chloride also followed similar suite
delivering the corresponding phosphonate products 7
exclusively (entries 8–10). Tertiary substrates, as
expected, were resistant to alkylation employing these
procedures. It is noteworthy to highlight no side prod-
ucts stemming from elimination or hydrolysis were
detected in any case.19
Table 3. Phosphonate formation using dialkyl phosphites,
and alkyl halides in the presence of Cs2CO3 and TBAI
As delineated in Table 2, various primary and sec-
ondary bromides were examined and found to be gener-
ally applicable. As depicted in entry 1, allyl bromide
coupled with diphenyl phosphite (6) efficiently to
provide the unsaturated phosphonate 7 in 77% yield
(entry 1). Likewise, an unactivated halide, 1-bromo-
propane was equally prosperous (entry 2). In addition,
extension of the alkyl chain on the halide-coupling
partner also proved fruitful (entries 3–5). Therefore, we
next directed our efforts toward the synthesis of a
complex phosphonate analog of biological importance,
geranyl phosphate. Using the aforementioned tech-
niques, 6 underwent alkylation with geranyl bromide
preparing the phosphonate analog 7 in 52% yield after
Scheme 2.

R. J. Cohen et al. / Tetrahedron Letters 44 (2003) 8617–8621
8619
Using the conditions described, we next screened various
dialkyl phosphites. As shown in Table 3, numerous
primary aliphatic bromides were united with dibenzyl
phosphite in moderate to high yields (entries 1–3). As
represented in entry 4, a sterically hindered secondary
bromide, 2-bromobutane, also underwent consolidation
offering 3 in 50% yield using the disclosed protocol. In
a continuing study, we explored the effect of alkyl chain
length on the dialkyl phosphite. Diethyl phosphite
reacted smoothly with both benzyl bromide and 2-(bro-
moethyl)benzene after approximately 29 h (entries 5 and
6). Also, dipropyl phosphite was found to behave simi-
larly using benzyl bromide (entry 7). In contrast, an
electron-rich halide (PMBCl) furnished a higher yield of
Scheme 4.
alkylations. Furthermore, the use of protecting groups
was completely avoided providing shortened synthetic
sequences. With this important result in hand, as well as
keeping the synthesis of phosphonopeptides3,21 in mind,
we envisioned the use of a bare b-amino bromide, as a
synthon might also prove facile. Therefore, after con-
struction of the b-amino bromide derived from isoleu-
cinol using a similar reported bromination protocol,22 we
subjected this chiral template to our conditions. Phos-
phite 6 successfully conjugated with the aforementioned
b-amino bromide to generate rearranged product 13
cleanly, and in satisfactory yield. Moreover, racemiza-
tion was not detected during the alkylation of this chiral
substrate while complications stemming from hydrolysis
or N-alkylation were not observed to a noticeable
extent.23
the desired phosphonate
3
(entry 8). Whereas,
(BuO)2P(O)H was also successfully transformed into the
coupled product in 84% isolated yield after 45 h (entry
9). In addition, a lipophilic phosphite, dilauryl phosphite,
was converted to 3, in excellent yield (entry 10). Similarly,
(iPrO)2P(O)H, a sterically congested phosphite, which
fails to undergo reaction using conventional methods,
proved particularly satisfactory with both BnBr and
1-bromo-3-phenylpropane (entries 11 and 12).
As demonstrated in Scheme 3, this improved phosphoryl-
ation method was also effective in the synthesis of unique
substituted phosphonates that contain sensitive func-
tional groups. Employing our mild Cs2CO3-promoted
conditions, dibenzyl phosphite (8) united with ethyl
bromoacetate after 66 h affording the substituted phos-
phonate 9 in high yield. Since fluoroalkyl phosphonates
have attracted attention,20 we next endeavored toward
the construction of the PꢀC bond which contain the
electron-withdrawing trifluoro-moiety. Employing simi-
lar reaction conditions, commercially available bis(2,2,2-
trifluoroethyl)phosphite (10) smoothly coupled with
benzyl bromide. Neither ester or phosphonate hydrolysis
were detected in either example.
Finally, we applied this methodology to the synthesis of
a phosphonate which is a crucial intermediate in the
synthesis phosphonomethylphenylalanine (L-Pmp), a
mimetic that exhibits similar properties to phosphoty-
rosine.24 As shown in Scheme 5, the synthesis of (4-
methylbenzyl)phosphonic acid di-tert-butyl ester (15)
was carried out by reaction of 4-methylbenzylbromide
with cesium di-tert-butyl phosphite generated in-situ
from (tBuO)2P(O)H (14) in DMF at room temperature
in high yield (90%). Subsequent reactions on building
block 15 can lead to the synthesis of protected Fmoc-L-
Pmp(But)2-OH 16, which is a very useful synthon for the
synthesis of Pmp-containing peptides and pepti-
domimetics.24
In summary, we report a mild and efficient method
for the preparation of phosphonates using cesium car-
bonate and tetrabutylammonium iodide offering a gen-
eral synthetic protocol. These improved procedures
augmented by the use of mild base and ambient reaction
temperature proved suitable for use with a wide variety
of dialkyl phosphites and alkyl halides. Moreover, the
synthesis of extremely hindered phosphonates otherwise
difficult to prepare using conventional methods was
particularly noteworthy. Furthermore, our procedures
discussed herein were superior to known methods, since
chiral templates encompassing amino acid derivatives
and labile functionalities proved compatible. Applica-
tions of this methodology toward the preparation of
After screening various classes of dialkyl phosphites with
numerous alkyl halides establishing the scope and limita-
tion of our procedure, we next embarked on the synthesis
of phosphonates that hold important applications in
synthesis. As shown in Scheme 4, diphenyl phosphite (6)
underwent alkylation with unprotected 2-bromoethyl-
amine·HBr to afford a derivative of the biologically
important 2-aminoethylphosphonic acid (AEPA), 12 in
67% isolated yield. It is crucial to note, no direct
N-alkylation products were detected, which demon-
strates our methods are highly chemoselective in tuning
Scheme 3.
Scheme 5.

8620
R. J. Cohen et al. / Tetrahedron Letters 44 (2003) 8617–8621
higher order phosphonopeptides and phoshonoty-
rosine-containing peptidomimetics are currently in pro-
gress and will be reported in due course.
Fleming, I., Eds.; Pergamon: London, 1979; Vol. 2, p.
1189 and references cited therein; (g) Shi, D.-E.; Wang,
Y.-M.; Chen, R.-Y. Heteroatomic Chem. 2000, 11, 261–
266.
15. Representative experimental procedure: To a solution of
dimethyl phosphite (4) (0.50 g, 0.42 mL, 4.54 mmol, 1
equiv.) in anhydrous DMF (23 mL) was added cesium
carbonate (4.44 g, 13.6 mmol, 3 equiv.) and tetra-
butylammonium iodide (5.03 g, 13.6 mmol, 3 equiv.)
with vigorous stirring for 1 hour at room temperature
under a nitrogen atmosphere. After this time period,
benzyl chloride (1.73 g, 1.6 mL, 13.6 mmol, 3 equiv.)
was added and stirred for an additional 24 h. The
resulting milky white suspension was then poured into
water (30 mL) and extracted with EtOAc (3×30 mL).
The combined organic layers were washed with water
(2×30 mL), brine (30 mL), and dried over anhydrous
sodium sulfate. The filtrate was concentrated in vacuo,
and subjected to flash chromatography (hexanes–
EtOAc, 9:1 v/v) affording dimethyl benzylphosphonate
(5) as a clear yellow oil (0.88 g, 97%). 1H NMR (270
MHz, CDCl3) lppm 3.09 (d, 2JPH=21.7 Hz, 2H), 3.59
(d, 3JPOCH=11 Hz, 6H), 7.23 (br s, 5H). 13C NMR (70
MHz, CDCl3) lppm 32.76 (1JPC=138.0 Hz), 52.83
Acknowledgements
Financial support from the NSF-KY EPSCoR (596166)
is gratefully acknowledged, as is support from Western
Kentucky University. Also, we wish to thank Chemetall
for their generous supply of cesium bases.
References
1. For a recent review, see: Fields, S. C. Tetrahedron
1999, 55, 12237–12273.
2. For a review, see: Blackburn, M. G. Chem. Ind. (Lon-
don) 1981, 134–144.
3. (a) For an overview, see: Failla, S.; Finocchiaro, P.;
Consiglio, G. A. Heteroatomic Chem. 2000, 11, 193–
504; (b) For a recent report, see: Rushing, S. D.; Ham-
mer, R. P. J. Am. Chem. Soc. 2001, 123, 4861–4862.
4. Kafarski, P.; Lejczak, B. Phosphorus Sulfur Silicon
Relat. Elem. 1991, 63, 193–215.
(2JPOC=7.1 Hz), 126.94 (5JPC=3.5 Hz), 128.59 (4JPC
=
3.0 Hz), 129.65 (d, 3JPC=6.6 Hz), 131.16 (d, 2JPC=9.6
Hz).
5. For a review, see: Krise, J. P.; Stella, V. J. Adv. Drug
Delivery Rev. 1996, 19, 287–310.
16. For reviews on the ‘cesium effect’, see: (a) Ostrowicki,
A.; Vogtle, F. In Topics In Current Chemistry; Weber,
E.; Vogtle, F., Eds.; Springer-Verlag: Heidelberg, 1992;
Vol. 161, p. 37; (b) Galli, C. Org. Prep. Proced. Int.
1992, 24, 287; (c) Blum, Z. Acta Chem. Scand. 1989,
43, 248; (d) Dijstra, G.; Kruizinga, W. H.; Kellogg, R.
M. J. Org. Chem. 1987, 52, 4230.
6. For representative examples of new biologically active
phosphonates, see: (a) Hildebrand, R. The Role of
Phosphonates in Living Systems; CRC Press: Boca
Raton, 1983; (b) Engel, R. Chem. Rev. 1977, 77, 349;
(c) Kafarski, P.; Leczak, B. Phosphorus Sulfur Silicon
Relat. Elem. 1991, 63, 193.
7. Engel, R. Chem. Rev. 1977, 77, 349–367.
17. For examples of efficient cesium-promoted alkylations,
see: (a) Parrish, J. P.; Dueno, E. E.; Kim, S.-I.; Jung,
K. W. Synth. Commun. 2000, 30, 2687; (b) Salvatore,
R. N.; Flanders, V. L.; Ha, D.; Jung, K. W. Org. Lett.
2000, 2, 2797; (c) Dueno, E. E.; Chu, F.; Kim, S.-I.;
Jung, K. W. Tetrahedron Lett. 1999, 40, 1843; (d) Sal-
vatore, R. N.; Nagle, A. S.; Schmidt, S. E.; Jung, K.
W. Org. Lett. 1999, 1, 1893; (e) Salvatore, R. N.; Shin,
S. I.; Nagle, A. S.; Jung, K. W. J. Org. Chem. 2001,
66, 1035; (f) Salvatore, R. N.; Sahab, S.; Jung, K. W.
Tetrahedron Lett. 2001, 42, 2055; (g) Salvatore, R. N.;
Schmidt, S. E.; Shin, S. I.; Nagle, A. S.; Worrell, J. H.;
Jung, K. W. Tetrahedron Lett. 2000, 41, 9705; (h) Sal-
vatore, R. N.; Ledger, J. A.; Jung, K. W. Tetrahedron
Lett. 2001, 42, 6023; (i) Kim, S.-I.; Chu, F.; Dueno, E.
E.; Jung, K. W. J. Org. Chem. 1999, 64, 4578; (j) Sal-
vatore, R. N.; Chu, F.; Nagle, A. S.; Kapxhiu, E. A.;
Cross, R. M.; Jung, K. W. Tetrahedron 2002, 58, 3329;
(k) Salvatore, R. N.; Shin, S. I.; Flanders, V. L.; Jung,
K. W. Tetrahedron Lett. 2001, 42, 1799; (l) Salvatore,
R. N.; Nagle, A. S.; Jung, K. W. J. Org. Chem. 2002,
67, 674.
8. (a) Bhattacharya, A. K.; Thyagarajan, G. Chem Rev.
1981, 81, 415–430 and references cited therein; (b)
Michaelis, A.; Kaehne, R. Ber. Dtsch. Chem. Ges. 1898,
31, 1048–1055; (c) Arbusov, B. A. Pure Appl. Chem.
1964, 9, 315–370.
9. (a) Methoden der organischen Chemie (Houben-Weyl);
Muller, E., Ed.; George Thieme Verlag: Stuttgart, 1964;
Vol. XII/1, p. 433; (b) Michaelis, A.; Becker, T. Chem.
Ber. 1897, 30, 1003.
10. Kosolapoff, G. M. J. Am. Chem. Soc. 1945, 67, 1180.
11. (a) Reactions and Methods of Organic Compound Inves-
tigation; Kabachnik, M., Ed.; Goskhimizdat: Moscow,
1953; Vol. 13, p. 427; (b) Grapov, A. F. Reakts.
Metody Issled. Org. Soedin. 1966, 15, 41.
12. Kosolapoff, G. M. Organophosphorus Compounds;
Wiley: New York, 1950; Chapter 7.
13. (a) Kosolapoff, G. J. Am. Chem. Soc. 1945, 65, 2959;
(b) Weizhen, Y.; Xiugao, L. Synthesis 1985, 10, 986.
14. For other synthetic methods, see: (a) Tomilov, A. P.;
Martynov, B. I.; Pavlova, N. A. J. Electroanal. Chem.
2001, 507, 46–48; (b) Kem, K. M.; Nguyen, N. V.;
Cross, D. J. J. Org. Chem. 1981, 46, 5188–5192; (c)
Kosolapoff, G. M. Org. React. 1951, 6, 273–338; (d)
Quin, L. D. A Guide To Organophosphorus Chemistry;
John Wiley: New York, 2000; pp. 133–165; (e) Hand-
book of Organophosphorus Chemistry; Engel, R., Ed.;
Marcel Dekker: New York, 1992; (f) Edmundson, R. S.
In Comprehensive Organic Synthesis; Trost, B. M.;
18. TBAI is strongly believed to act as a phase-transfer
catalyst facilitating alkylations and higher product
yields. For other phase-transfer catalyzed P-alkylations,
see: (a) Kem, K. M.; Nguyen, N. V.; Cross, D. J. J.
Org. Chem. 1981, 46, 5188; (b) Weber, W. P.; Gokel,
G. W. Phase Transfer Catalysts in Organic Synthesis;
Springer-Verlag: New York, 1977; (c) Starks, C. M.;

R. J. Cohen et al. / Tetrahedron Letters 44 (2003) 8617–8621
8621
Liotta, C. Phase Transfer Catalysis: Principles and Tech-
niques; Academic Press: New York, 1978; (d) Fedorynski,
M.; Wojciechowski, K.; Matacz, Z.; Makosza, M. J. Org.
Chem. 1978, 43, 4682; (e) Kluba, M.; Zwierzak, A. Syn-
thesis 1978, 2, 134–137.
21. For the synthesis of phosphonopeptides, see: Kafarski,
P.; Lejczak, B. In Aminophosphonic and Aminophosphinic
Acids, Chemistry and Biological Activity; Kukhar, V. P.;
Hudson, H. R., Eds.; John Wiley: Chichester, UK, 2000;
pp. 173–203 and references cited therein.
22. Nagle, A. S.; Salvatore, R. N.; Chong, B. D.; Jung, K.
W. Tetrahedron Lett. 2000, 41, 3011.
19. In most examples, the starting dialkyl phosphites were
not completely consumed during the reaction, accounting
for additional mass balance.
20. (a) Burton, D. J.; Ishihara, T.; Flynn, R. M. J. Fluorine
Chem. 1982, 20, 121–126; (b) Xu, Y.; Prestwich, G. D.
Org. Lett. 2002, 4, 4021–4024.
23. 1H, 13C, and 2D NMR analyses indicate the product as a
single diastereomer.
24. Li, P.; Zhang, M.; Peach, M. L.; Liu, H.; Yang, D.;
Roller, P. P. Org. Lett. 2003, 5, 3095.