Palladium(0)-catalyzed benzylation of H-phosphonate diesters: An efficient entry to benzylphosphonates

Article abstract of DOI:10.1055/s-0028-1087522

A new, efficient method for the synthesis of benzylphosphonate diesters via a palladium(0)-catalyzed cross-coupling reaction between benzyl halides and H-phosphonate diesters, using Pd(OAc)₂ as a palladium source and Xantphos as a supporting ligand, has been developed. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0028-1087522

LETTER
225
Palladium(0)-Catalyzed Benzylation of H-Phosphonate Diesters: An Efficient
Entry to Benzylphosphonates
S
ynthesis of Benzy
a
lphosphon
s
ates ton Lavén,a Jacek Stawinski*a,b
a
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, 10691 Stockholm, Sweden
E-mail: js@organ.su.se
b
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61704 Poznan, Poland
Received 24 September 2008
The above drawbacks of the Michaelis–Arbuzov and the
Michaelis–Becker reactions usually become apparent dur-
ing synthesis of more complex structures e.g. those that
can be interesting from biological point of view.2 To alle-
viate these problems, in such instances the formation of
the P–C bond has to be carried out at the early stages of
the synthesis,16 or dedicated reagents with the preformed
P–C linkages,17 have to be used. For example, in the syn-
thesis of antisense benzylphosphonate-modified oligonu-
cleotides, that show potent in vivo and in vitro inhibitory
effects against hepatitis C virus,18 benzylphosphonous
dichloride was used as a starting material.19 This lessened
some inconveniences of the Michaelis–Arbuzov and the
Michaelis–Becker reactions, but introduced new prob-
lems, e.g. scrambling of the substituents at phosphorus
due to fast ligand exchange in the tervalent P(III) precur-
sors used. In addition, since such methods are based on
dedicated phosphonylating reagents, the whole synthetic
Abstract: A new, efficient method for the synthesis of benzylphos-
phonate diesters via a palladium(0)-catalyzed cross-coupling reac-
tion between benzyl halides and H-phosphonate diesters, using
Pd(OAc)2 as a palladium source and Xantphos as a supporting
ligand, has been developed.
Key words: benzylphosphonates, palladium cross-coupling, H-phos-
phonates, C-phosphonates
Although phosphorus compounds containing the P–C
bond are not particularly abundant in nature,1 they have
attracted considerable interest as isosteric analogues of
phosphate esters.2–4 Due to the presence of the P–C bond,
these compounds are usually resistant to enzymatic hy-
drolysis and exhibit a vast array of biological activity.2,3,5
This constitutes a primary rationale of using C-phospho-
nates analogues as target-specific modulators of variety of
biological processes, e.g. as pesticides and therapeutics.2,5
For the formation of the P–C(sp3) bond the most common sequence has to be repeated for each new benzylphospho-
approaches are probably those involving the Michaelis– nate analogue synthesized.19
Arbuzov6,7 and Michaelis–Becker8 reactions. Although
Inspired by potent biological activity of various ben-
quite general in scope, these methods suffer from impor-
zylphosphonate derivatives, on one hand, and mild reac-
tant drawbacks.
tion conditions provided by transition metal chemistry for
For the Michaelis–Arbuzov reaction6,7 (a reaction of tri- the synthesis of nucleoside aryl-20,21 and vinylphospho-
alkyl phosphites with alkyl halides), the formation of nates,22 on the other one, we set out to develop a new, gen-
alkylphosphonates is exothermic,9 but the activation ener- eral method for a palladium-catalyzed formation of
gy is usually high, and this necessitates a prolonged heat- benzylphosphonate diesters. Although benzylpalladi-
ing at high temperature.7,10 Since during the course of the um(II) complexes were prepared for the first time some 40
reaction a new alkyl halide is formed, this has to be less years ago,23 activation and functionalization of benzylic
reactive than the one used as a substrate, or it has to be re- derivatives by palladium catalysts is by far less common24
moved from the reaction media.11 Some problems may than that of aromatic and olefinic compounds.25 This is
arise for P-chiral phosphites due to regioselectivity of the particularly true for cross-coupling reactions with phos-
dealkylation process,7 and in addition, a stereochemical phorus nucleophiles, for which only a few examples of a
outcome of the reaction (retention of configuration or P–C(sp3) bond formation using benzyl halides have been
epimerization) may depend on the reaction conditions and reported so far.26
the kind of a nucleophile used.12 As to the Michaelis–
As a model reaction for our studies, a cross-coupling of
Becker reaction8 (a reaction of alkali metal salts of dialkyl
diethyl H-phosphonate with benzyl bromide, catalyzed by
phosphites with alkyl halides), this usually occurs under
palladium(0), was investigated using different palladium
milder reaction conditions, but it requires strong bases13
sources and monodentate or bidentate ligands (Table 1).
and, due to high reactivity of the generated phosphite an-
The experiments were carried out in THF, at 66 ºC, in the
presence of N,N-diisopropylethylamine (DIPEA) as a
ions, complex reaction mixture can be formed (SET
mechanism),13,14 especially with substrates having a
base, and the extent of conversion into diethyl ben-
pseudohalide character.15
zylphosphonate was estimated after 20 hours by 31P NMR
spectroscopy.
SYNLETT 2009, No. 2, pp 0225–0228
x
x.
x
x.
2
0
0
9
Attempted synthesis of diethyl benzylphosphonate using
10 mol% of Pd(dba)2 as a catalyst was unsuccessful, as no
Advanced online publication: 15.01.2009
DOI: 10.1055/s-0028-1087522; Art ID: G31708ST
© Georg Thieme Verlag Stuttgart · New York

226
G. Lavén, J. Stawinski
LETTER
the product formation could be detected after 20 hours by
the 31P NMR spectroscopy. With 10 mol% of Pd(PPh3)4
[the prevalent catalyst for the P–C(sp2) bond formation27]
ca. 22% of the desired diethyl benzylphosphonate was
formed, but the yield was not improved upon prolonged
heating.
PPh2
PPh2
PPh2
PPh2
n
Ph2P
PPh2
Fe
n = 2, dppe
n = 3, dppp
n = 4, dppb
BINAP
dppf
With Pd(OAc)2 and monodentate ligands, namely, with an
electron-deficient tris(p-chlorophenyl)phosphine [P(4-
C6H4Cl)3], the conversion into benzylphosphonate in-
creased to 71%, but with a trialkylphosphine, tris(cyclo-
hexyl)phosphine (PCy3), it was lower than that for PPh3
(Table 1). We also screened several bisphosphines as pos-
sible supporting ligands (Figure 1). Unfortunately, most
of them, e.g. 1,2-bis(diphenylphosphino)ethane (dppe),
1,3-bis(diphenylphosphino)propane (dppp), and 1,4-
bis(diphenylphosphino)butane (dppb), and also those fre-
quently used in cross-coupling reactions with heteroatom
nucleophiles (including H-phosphonates),25,28 [e.g. 2,2¢-
bis(diphenylphosphino)-1,1¢-binaphtyl (BINAP, racemic)
and 1,1-bis(diphenylphosphino)ferrocene (dppf)], turned
out to be rather inefficient (Table 1, entries 4–8).
Since Stockland et al.30 reported that reductive elimina-
tion from metalphosphonate complexes is significantly
accelerated in the presence of large bite angle phosphine
ligands, we investigated DPEphos and its more rigid
counterpart, Xanthpos31 (Figure 1) as possible bidentate
ligands for our reaction. To our delight, there was a sharp
increase in the yield of benzylphosphonate diesters forma-
O
O
PPh2
DPEphos
PPh2
PPh2
Xantphos
PPh2
Figure 1 Structures of the bidentate ligands used in experiments
summarized in Table 1
mide was virtually quantitative (Table 1, entries 9 and
10). Monitoring progress of the reaction using 31P NMR
spectroscopy revealed that the cross-coupling in the pres-
ence of Xantphos as a supporting ligand went to comple-
tion within three hours at 50 °C.
Having found an efficient catalytic system for the forma-
tion of diethyl benzylphosphonate, we carried out similar
experiments with different H-phosphonate diesters and
various benzyl halide derivatives (Table 2).
Table 2 Reaction of Different H-Phosphonate Diesters with
tion in the reaction with DPEphos, and for Xantphos, the Various Benzyl Halide Derivatives
cross-coupling of diethyl H-phosphonate with benzyl bro-
O
X
P(OR)2
O
conditions
Table 1 Efficiency of Different Ligands in the Cross-Coupling
Reaction between Diethyl H-Phosphonate and Benzyl Bromide
+
HP(OR)2
O
O
Y
Y
conditions
HP(OEt)2
BnP(OEt)2
Entry
R
Y
X
Time
Isolated
yield
Entry
Ligand
Conversion (%)a
b
1
2
PPh3
22
71
15
0
1
Et
H
Br
Cl
Br
Br
Br
Cl
Br
Cl
Br
Cl
Br
3 h
3 h
2 h
2 h
4 h
4 h
4 h
4 h
2 h
2 h
2 h
95%
94%
91%
94%
98%
86%
94%
93%
74%
71%
89%
P(4-C6H4Cl)3
PCy3
2
Et
H
3
3
Et
Me
MeO
F
4
dppe
4
Et
5
dppp
5
5
Et
6
dppb
21
14
19
86
99
6
Et
Cl
H
7
BINAP
dppf
7
i-Pr
i-Pr
Ph
Ph
Bn
8
8
H
9
DPEphos
Xantphos
9
H
10
10
11
H
Reaction conditions: (EtO)2P(O)H (0.1 mmol), Pd(OAc)2 (0.1 equiv),
ligand [monodentate (0.4 equiv) or bidentate (0.2 equiv)], N,N-diiso-
propylethylamine (1.2 equiv), benzyl bromide (1.5 equiv), THF [1
mL; containing H2O (0.05 equiv)],29 66 °C in a sealed tube, 20 h.
a Conversion into the product was determined by 31P NMR.
b Pd(PPh3)4 was used.
F
Reaction conditions: (RO)2P(O)H (1 mmol), Pd(OAc)2 (0.1 equiv),
Xantphos (0.2 equiv), benzyl halide (1.5 equiv), N,N-diisopropyl-
ethylamine (1.3 equiv), THF [10 mL; containing H2O (0.05 equiv)],29
reflux under inert atmosphere.
Synlett 2009, No. 2, 225–228 © Thieme Stuttgart · New York

LETTER
Synthesis of Benzylphosphonates
227
As apparent from data in Table 2, the reactions investigat- Pd (OAc)2, N,N-diisopropylethylamine, and Xantphos
ed were only slightly sensitive to the kind of benzyl ha- (Scheme 2).
lides and the H-phosphonate diesters used. They were
somewhat faster for the electron-rich benzyl halides
Br(Cl)
(Table 2, entries 3 and 4), while the presence of electron-
withdrawing substituents in the aromatic ring, slightly
slowed down the reaction (Table 2, entries 5 and 6).
R1O
R1O
Thy
Thy
O
O
O
O
O
P
O
P
On the H-phosphonate part, there was a small decrease in
the reaction rates when going from diethyl to diisopropyl
H-phosphonates (Table 2, entries 1 and 2 vs. 7 and 8), and
a small shortening of the reaction time was observed for
diphenyl H-phosphonate (Table 2, entries 9 and 10).32 It is
worth noticing here that there were no significant differ-
ences in the reaction times when the corresponding benzyl
chlorides vs. bromides were used as substrates.
Pd(OAc)2
H
Xantphos
base
THF
O
Thy
O
Thy
O
O
R2O
R2O
2a, (RP) ('fast', δP = 27.4 ppm)
2b, (SP) ('slow', δP = 28.4 ppm)
1a, (RP) ('fast', δP = 6.9 ppm)
1b, (SP) ('slow', δP = 8.6 ppm)
Thy = thymin-1-yl; R1 = 4,4'-dimethoxytrityl; R2 = tert-butyldimethylsilyl
Assuming that the reactions investigated followed a stan-
dard three-step Pd-catalytic cycle33 (Scheme 1), the ob-
served trends in reactivity can be explained in the
following way.
Scheme 2 Synthesis of dinucleoside benzylphosphonates
The reaction was complete within three hours and pro-
duced the expected dinucleoside benzylphosphonate 2
(ca. 1:1 mixture of the diastereomers) quantitatively (31P
NMR spectroscopy). As for simple dialkyl H-phospho-
nates, also dinucleoside derivative 1 reacted readily with
benzyl chloride to afford the desired product 2 in a com-
parable yield.
HPO(OR)
2 + Base
BnPdX
BnX
A
B
H-Base+ X–
Pd(II)
Pd(0)
BnPdPO(OR)2
C
To have some insight into the stereochemistry of this pal-
ladium(0)-catalyzed cross-coupling, the reaction with
benzyl bromide was repeated using separate diastereo-
mers of dinucleoside H-phosphonate 1 (1a and 1b).37 It
was found that the P–C bond formation was completely
stereospecific (most likely retention of configuration) as
the RP diastereomer 1a (dP = 6.9 ppm) afforded exclusive-
ly the diastereomer 2a, (dP = 27.4 ppm), while the SP dia-
stereomer 1b (dP = 8.6 ppm) produced the other
diastereomer of benzylphosphonate 2b (dP = 28.4 ppm).38
A: oxidative addition
B: ligand exchange
C: reductive elimination
BnPO(OR)2
Scheme 1 A catalytic cycle for a palladium(0)-mediated benzyl-
phosphonates formation
The fact that benzyl chlorides and benzyl bromides dis-
played similar reactivity indicated that the oxidative addi-
tion was probably not the rate-determining step. This
conclusion was additionally supported by the finding that
electron-deficient benzyl halides reacted slower than the
electron-rich ones, since the opposite order of reactivity
would be expected when the oxidative addition were ki-
netically significant.34,35 Thus, it seemed likely that the
slowest step of the catalytic cycle was probably the reduc-
tive elimination (step C, Scheme 1). Such an assumption
was justified by two reaction features: (i) a higher reactiv-
ity of the electron-rich35,36 benzyl derivatives in the cross-
coupling reaction, and (ii) a significant increase in the re-
action rate when large bite angle phosphine ligands30 were
used. Of course, one cannot exclude a kinetic participation
of some additional factors in the mechanistic pathway.
Since more acidic diphenyl H-phosphonate reacted faster
than diethyl H-phosphonate, it is possible that the ligand
substitution process may, at least to some extent, affect
the reaction rate.
In conclusion, we have developed a new, general method
for efficient preparation of benzylphosphonate diesters
via the Pd(0)-catalyzed cross-coupling reaction of benzyl
chlorides or bromides with H-phosphonate diesters. Sev-
eral ligands have been evaluated for their ability to cata-
lyze this P–C(sp3) bond formation, and the most efficient
catalytic system was found to consist of Pd(OAc)2 as a
palladium source and Xantphos as a supporting ligand.
The reaction is stereospecific (most likely retention of
configuration at the phosphorus centre20,39), proceeds un-
der mild conditions, and permits preparation of various
benzylphosphonate derivatives from a common precur-
sor. Since the underlying chemistry of the method is dif-
ferent from that of the Michaelis–Arbuzov and the
Michaelis–Becker reactions, this new protocol expands
the range of synthetic methods available for the prepara-
tion of biologically important C-phosphate analogues.
As a final stage of these investigations, we checked the ef-
ficacy of the developed synthetic protocol by reacting di-
nucleoside H-phosphonates 1 (a diastereomeric mixture,
ca. 1:1) with benzyl bromide in THF in the presence of
Acknowledgment
Financial support from the Swedish Research Council is gratefully
acknowledged.
Synlett 2009, No. 2, 225–228 © Thieme Stuttgart · New York

228
G. Lavén, J. Stawinski
LETTER
(29) Presence of water in the reaction mixture facilitated
reduction of palladium(II) acetate and resulted in improved
reproducibility of the reactions.
(30) Stockland, R. A.; Levine, A. M.; Giovine, M. T.; Guzei,
I. A.; Cannistra, J. C. Organometallics 2004, 23, 647.
(31) Klingensmith, L. M.; Strieter, E. R.; Barder, T. E.;
Buchwald, S. L. Organometallics 2006, 25, 82.
(32) Lower yields of diphenyl benzylphosphonate for the
reactions of diphenyl H-phosphonate (Table 2, entries 9 and
10) were due to partial hydrolysis of the starting H-
phosphonate under the reaction conditions.
(33) Kalek, M.; Stawinski, J. Organometallics 2007, 26, 5840.
(34) Stille, J. K.; Lau, K. S. Y. Acc. Chem. Res. 1977, 10, 434.
(35) Amatore, C.; Jutand, A. J. Organomet. Chem. 1999, 576,
254.
References and Notes
(1) (a) Kittredge, J. S.; Roberts, E. Science 1969, 164, 37.
(b) White, A. K.; Metcalf, W. W. Annu. Rev. Microbiol.
2007, 61, 379.
(2) Engel, R. Chem. Rev. 1977, 77, 349.
(3) Kafarski, P.; Lejczak, B. Phosphorus, Sulfur Silicon 1991,
63, 193.
(4) Huang, J. M.; Chen, R. Y. Heteroatom. Chem. 2000, 11, 480.
(5) Engel, R. In Handbook of Organophosphorus Chemistry;
Engel, R., Ed.; Marcel Dekker: New York, 1992, 559.
(6) (a) Michaelis, A.; Kaehne, R. Chem. Ber. 1898, 31, 1048.
(b) Methoden der organischen Chemie (Houben-Weyl), Vol.
XII/1; Müller, E., Ed.; George Thieme Verlag: Stuttgart,
1964, 433.
(7) Bhattacharya, A. K.; Thyagarajan, G. Chem. Rev. 1981, 81,
(36) Hartwig, J. F. Acc. Chem. Res. 1998, 31, 852.
(37) Stawinski, J.; Strömberg, R.; Zain, R. Tetrahedron Lett.
1992, 33, 3185.
415.
(8) (a) Michaelis, A.; Becker, T. Chem. Ber. 1897, 30, 1003.
(b) Methoden der organischen Chemie (Houben-Weyl), Vol.
XII/1; Müller, E., Ed.; George Thieme Verlag: Stuttgart,
1964, 446. (c) Waschbüsch, R.; Carran, J.; Marinetti, A.;
Savignac, P. Synthesis 1997, 727.
(9) Brill, T. B.; Landon, S. J. Chem. Rev. 1984, 84, 577.
(10) Harizi, A.; Zantour, H. Phosphorus, Sulfur Silicon 2004,
179, 1883.
(11) Saady, M.; Lebeau, L.; Mioskowski, C. Helv. Chim. Acta
1995, 78, 670.
(12) DeBruin, K. E.; Chandrasekaran, S. J. Am. Chem. Soc. 1973,
95, 974.
(13) Kers, A.; Stawinski, J.; Dembkowski, L.; Kraszewski, A.
Tetrahedron 1997, 53, 12691.
(14) (a) Witt, D.; Rachon, J. Phosphorus, Sulfur Silicon 1995,
107, 33. (b) Witt, D.; Rachon, J. Heteroatom. Chem. 1996,
7, 359.
(15) (a) Birckenbach, L.; Kellermann, K. Chem. Ber. 1925, 58,
786. (b) Michalski, J.; Skowronska, A.; Lopusinski, A.
Phosphorus, Sulfur Silicon 1991, 58, 61.
(16) (a) Yao, Z.-J.; Gao, Y.; Burke, T. R. Tetrahedron:
Asymmetry 1997, 10, 3727. (b) Li, P.; Zhang, M.; Peach,
M. L.; Liu, H.; Yang, D.; Roller, P. P. Org. Lett. 2003, 5,
3095.
(17) Löschner, T.; Engels, J. W. Nucleic Acids Res. 1990, 18,
5083.
(18) (a) Alt, M.; Eisenhardt, S.; Serwe, M.; Renz, R.; Engels,
J. W.; Caselmann, W. H. Eur. J. Clin. Invest. 1999, 29, 868.
(b) Lehman, T. J.; Engels, J. W. Bioorg. Med. Chem. 2001,
9, 1827.
(19) Amberg, S.; Engels, J. W. Helv. Chim. Acta 2002, 85, 2503.
(20) Johansson, T.; Stawinski, J. Chem. Commun. 2001, 2564.
(21) Lavén, G.; Stawinski, J. Coll. Symposium Series 2005, 7,
195.
(22) Abbas, S.; Hayes, C. J. Synlett 1999, 1124.
(23) Fitton, P.; McKeon, J. E.; Ream, B. C. J. Chem. Soc., Chem.
Commun. 1969, 370.
(24) Liegault, B.; Renaud, J.-L.; Bruneau, C. Chem. Soc. Rev.
2008, 36, 290.
(25) Prim, D.; Campagne, J.-M.; Joseph, D.; Andrioletti, B.
Tetrahedron 2002, 58, 2041.
(26) Bravo-Altamirano, K.; Huang, Z. H.; Montchamp, J. L.
Tetrahedron 2005, 61, 6315.
(27) Schwan, A. L. Chem. Soc. Rev. 2004, 33, 218.
(28) Abbas, S.; Hayes, C. J.; Worden, S. Tetrahedron Lett. 2000,
41, 3215.
(38) Typical Procedure for the Preparation of Dinucleoside
Benzylphosphonates 2: Pd(OAc)2 (0.05 mmol), Xantphos
(0.1 mmol), and N,N-diisopropylethylamine (mmol), were
refluxed for ca. 3 h in degassed THF (5 mL) containing H2O
(0.025 mmol). To this, separate diastereomers of
dinucleoside H-phosphonate 1 (1a or 1b; 0.5 mmol),37 and
benzyl bromide (0.75 mmol), dissolved in THF (2 mL), were
added and the mixture was heated under reflux for 3 h. After
concentration and partition of the reaction mixture between
sat. aq NaHCO3 and CH2Cl2, the product was purified by
silica gel column chromatography using a stepwise gradient
of ethanol (0–5%) in CH2Cl2 containing triethylamine
(0.02%). Compounds 2 were obtained as off-white solids
(purity >98%, 1H NMR spectroscopy). Compound 2a: 83%
yield from 1a (probably RP diastereomer). HRMS: m/z [M +
Na]+ calcd for C54H65N4NaO13PSi+: 1059.3947; found:
1059.3908. Compound 2b: 84% yield from 1b (probably SP
diastereomer). HRMS: m/z [M + Na]+ calcd for
C54H65N4NaO13PSi+: 1059.3947; found: 1059.3941.
Benzylphosphonates (Table 2) prepared from benzyl
chlorides vs. benzyl bromides were spectrally
indistinguishable, and were obtained as yellowish oils
(purity >98%, 1H NMR spectroscopy). Diethyl
benzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C11H17NaO3P+: 251.0808; found: 251.0818. Diethyl 4-methyl-
benzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C11H17NaO3P+: 265.0964; found: 265.0975. Diethyl
4-methoxybenzylphosphonate: HRMS: m/z [M + Na]+ calcd
for C12H19NaO4P+: 281.0913; found: 281.0907. Diethyl 4-
fluorobenzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C11H16FNaO3P+: 269.0713; found: 269.0727. Diethyl 4-
chlorobenzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C11H16ClNaO3P+: 285.0418; found: 285.0394. Diisopropyl
benzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C13H21NaO3P+: 279.1121; found: 279.1127. Diphenyl
benzylphosphonate: HRMS: m/z [M + Na]+ calcd for
C19H17NaO3P+: 347.0808; found: 347.0798.
The benzylphosphonate diesters synthesized were
characterized by 1H NMR, 13C NMR, and 31P NMR
spectroscopy.
(39) (a) Xu, Y.; Zhang, J. J. Chem. Soc., Chem. Commun. 1986,
1606. (b) Zhang, J.; Xu, Y.; Huang, G.; Guo, H. Tetrahedron
Lett. 1988, 29, 1955.
Synlett 2009, No. 2, 225–228 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.