A concise method for synthesis of diaryl aryl- or alkylphosphonates

Article abstract of DOI:10.1016/j.tetlet.2005.11.028

A convenient method has been developed to synthesize diaryl arylphosphonates from triaryl phosphites, triethyl phosphite, and aromatic halides. The new method relies on the triethyl phosphite assisted nickel catalyzed Arbuzov reaction and can be applied to synthesize certain diaryl alkylphosphonates without catalysts.

Full text of DOI:10.1016/j.tetlet.2005.11.028

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 277–281
A concise method for synthesis of diaryl aryl- or alkylphosphonates
Qiang Yao* and Sergei Levchik
Supresta US LLC, 430 Saw Mill River Road, Ardsley, NY 10502, USA
Received 19 October 2005; accepted 4 November 2005
Available online 22 November 2005
Abstract—A convenient method has been developed to synthesize diaryl arylphosphonates from triaryl phosphites, triethyl phos-
phite, and aromatic halides. The new method relies on the triethyl phosphite assisted nickel catalyzed Arbuzov reaction and can
be applied to synthesize certain diaryl alkylphosphonates without catalysts.
Ó 2005 Elsevier Ltd. All rights reserved.
Phosphonates have important applications in flame
retardancy,1,2 organic synthesis,3 and biological applica-
tions.4,5 The Arbuzov reaction is regarded as the
premier synthetic method in the preparation of phos-
phonates from phosphites and halides.6 The mechanism
of the Arbuzov reaction has been shown to involve two
nucleophilic attacks, that is, the formation of a quasi-
phosphonium salt and the dealkylation of this salt,
accompanied by rearrangement to a phosphonate
(Scheme 1).7
phosphonates.8 In spite of these advances, the transi-
tion-metal catalyzed Arbuzov reaction faces the same
challenge when directly applied to the synthesis of diaryl
arylphosphonates (R, R0 = aryl in Scheme 1) from tri-
aryl phosphites, because it would require the nucleophilic
displacement on an aryl group in the rearrangement step
of the Arbuzov reaction.13
In our project to evaluate the performance of organo-
phosphorus additives in polymers, we set a goal to
develop industrially feasible synthetic methods for diaryl
arylphosphonates, which have been only accessible
through tedious and expensive routes, that is, Frie-
del–Crafts reaction of aromatic compounds by PCl3
followed by the oxidation of aromatic dichlorophos-
phines and the subsequent esterification of arylphos-
phonic dichlorides with aromatic alcohols.14,15 Since
the Arbuzov reaction requires the presence of at least
one alkylating group in a quasiphosphonium structure
to satisfy the subsequent dealkylation to phosphonates,
we envisioned that triaryl arylphosphonium salts may
transform to diaryl arylphosphonates in the presence
of an alkylating agent, which is able to first transeste-
rify the quasiphosphonium salt and subsequently allow
the rearrangement step of the Arbuzov reaction. We
now find that diaryl arylphosphonates can indeed
be formed from triaryl phosphites, aromatic halides,
and alkylating agents, and the new method can be
easily extended to certain diaryl alkylphosphonates as
well.
However, due to the aryl groupsÕ resistance to nucleo-
philic attack, the classic Arbuzov reaction was essen-
tially limited to the preparation of alkylphosphonates
(R0 = alkyl in Scheme 1). The synthesis of arylphospho-
nates (R0 = aryl in Scheme 1) represents one early chal-
lenge. In this regard, the transition-metal catalyzed
Arbuzov reaction has gained remarkable success.8–12
TavsÕs seminal work on nickel catalyzed phosphonyl-
ation of aryl halides by trialkyl phosphites extends the
classic Arbuzov reaction to the synthesis of dialkyl aryl-
(RO)3P+R'
+
X-
(RO)3P:
+
R' - X
(RO)2(R')P+- O - R
+
X-
(RO)2P(O)R' + RX
X = Cl, Br, I
Scheme 1. Mechanism of the Arbuzov reaction.
In the first attempt, triphenyl phosphite was mixed with
iodobenzene in the presence of nickel chloride. Metha-
nol was then used to assist the Arbuzov rearrangement.
Entry 1 of Table 1 shows that diphenyl phenylphospho-
Keywords: Triaryl phosphite; Triethyl phosphite; Diaryl arylphospho-
nate; Diaryl alkylphosphonate; Arbuzov reaction.
*
Corresponding author. Tel.: +1 914 245 5920; fax: +1 914 674
9438; e-mail: john.yao@supresta.com
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.11.028

278
Q. Yao, S. Levchik / Tetrahedron Letters 47 (2006) 277–281
Table 1. Diphenyl phenylphosphonate from triphenyl phosphite
This synthetic method can be easily applied to the acti-
vated alkyl halides where the catalyst is not necessary.
For example, diphenyl benzylphosphonate was readily
obtained quantitatively (Table 2, entry 13). However,
the attempt to make diphenyl pentylphosphonate only
met limited success (Table 2, entry 14).
(TPPi) and iodobenzene in the presence of an alkylating agenta
Entry
Alkylating agent
Yield (%)b
1c
2d
CH3OH
(EtO)3P
77
98
a 0.06 mol TPPi, 160 °C/4 h, the amount of NiCl2 (5 mol %) was based
on iodobenzene.
The NiCl2-catalyzed Arbuzov reaction of trialkyl phos-
phites with aryl halides involves in situ reduction of
NiCl2 to tetrakis(trialkyl phosphite) nickel(0).18 We pro-
pose a reaction mechanism based on the oxidative addi-
tion and reductive elimination on nickel(0) with an
additional transesterification step, which generates the
mixed phosphites (Scheme 2).19–22
b 31P NMR results.
c TPPi/idobenzene/CH3OH = 1/1/1.3, phenol was formed.
d TPPi/idobenzene/triethyl phosphite (TEPi) = 2/3/1.1, no phenol was
generated.
nate was obtained in a good yield (77%). However,
diphenyl methylphosphonate (23%) and a theoretical
quantity of phenol together with methyl iodide were also
simultaneously produced. From the point of atom econ-
omy,16 this route has a disadvantage. Since we specu-
lated that the rearrangement may be preceded by a
transesterification, we then resorted to triethyl phos-
phite, which undergoes ester exchange with triphenyl
phosphite.17 To our delight, diphenyl phenylphospho-
nate was obtained in a high yield (98%) with ethyl iodide
as the major by-product.
When triphenyl phosphite was mixed with triethyl phos-
phite, an equilibrium was established between starting
materials and ethyl diphenyl and diethyl phenyl phosph-
ites, which were formed via ester exchange. Obviously,
those phosphites with alkylating ability induce the for-
mation of arylphosphonates I, II, and III in the presence
of aryl halides under the influence of a nickel(0) catalyst
(Scheme 2, left circle). Indeed, in the course of the reac-
tion, not in the final products, we found the formation
of both diethyl arylphosphonates and ethyl aryl aryl-
phosphonates in the 31P NMR. However, the absence
of di/mono-ethyl arylphosphonates (II and III in
Scheme 2) in the final products suggests they act as
alkylating agents (Scheme 2, right circle).
This triethyl phosphite assisted method was extended to
a variety of aryl halides and triaryl phosphites. Most of
the diphenyl arylphosphonates were obtained in excel-
lent yields (Table 2, entries 1–6, 8, 10, and 12). While
a detailed kinetic study was not performed, it was
observed that the electron donating groups on the
aromatic ring of the aryl halide usually facilitated the
reaction while electron withdrawing groups suppressed
the reaction.18 Thus, while diphenyl 4-methoxy-
benzenephosphonate (Table 2, entry 5) was obtained
in 97% yield at 163 °C in 2.5 h, ethyl diphenyl-
phosphonobenzoate (Table 2, entry 6) required 6.5 h
to reach a similar conversion at even higher tempera-
ture, and diphenyl 4-tolylphosphonate and diphenyl
phenylphosphonate (Table 2, entries 4 and 3) were syn-
thesized under intermediate conditions.
That II and III act as alkylating agents is supported by
the fact that the change of the addition order did not
alter the reaction products. For example, immediately
after diethyl phenylphosphonate (III in Scheme 2,
R1 = H) was synthesized from triethyl phosphite/iodo-
benzene/NiCl2, adding triphenyl phosphite and iodo-
benzene to the reaction solution yielded the same final
products, diphenyl phenylphosphonate (I in Scheme 2,
R1 = H and Ar = Ph) and ethyl iodide, as those
obtained by adding triethyl phosphite to the solution
of triphenyl phosphite/iodobenzene/NiCl2. The interme-
diate ethyl phenyl phenylphosphonate was observed in
the 31P NMR in both cases. The mechanism leading to
the formation of diaryl alkylphosphonates is likely to
be similar to the above mechanism except that the
P–C bond is directly formed without the metal being
involved.
Aryl iodides generally reacted faster than aryl bromides,
while non-activated aryl chlorides were inert under simi-
lar reaction conditions (Table 2, entries 1, 3, 7, and 9).
The functional groups in aryl iodides may interfere with
the nickel halide catalyst. Formation of a few percent
of diphenyl methylphosphonate was noted when
4-methoxyiodobenene was used instead of 4-methoxy-
bromobenzene (Table 2, entry 2). Apparently, meth-
ylphosphonate was formed via methyl iodide, which
stemmed from the cleavage of the 4-methoxy group on
the benzene ring.
In contrast to the transition-metal-catalyzed carbon–
carbon bond formation of aromatic halides where elec-
tron-withdrawing groups usually facilitate the reactions
by activating the benzene ring and thus the oxidative
addition step,23 the retardation of electron-withdrawing
groups on phosphonylation may suggest that the reduc-
tive elimination of arylphosphonates from the presumed
nickel complex is the rate-determining step.
Steric hindrance significantly affects the course of the
reaction. It has been observed in the classic Arbuzov
reaction that the presence of a b branch in the alkyl
halide retards the reaction. Thus, it was not too surpris-
ing that the reaction of triphenyl phosphite and methyl
o-iodobenzoate went very slowly (Table 2, entry 11). A
sterically demanding phosphite also significantly low-
ered the yield (Table 2, entry 15).
In conclusion, we have developed a triethyl phosphite
assisted method to synthesize diaryl aryl- or
alkylphosphonates from triaryl phosphites and aryl or
activated alkyl halides. This new method easily extends
the classic Arbuzov reaction to cover the otherwise dif-
ficultly synthesized diaryl aryl- or alkylphosphonates. Its

Q. Yao, S. Levchik / Tetrahedron Letters 47 (2006) 277–281
Table 2. Diaryl phosphonates from triaryl phosphites and aromatic or alkyl Halidesa
279
X
O
P
O
O
5 mol% NiCl2
O P O
O
3EtX
+
(EtO)3P
+ 3
+
3
2
R1
R1
Entry
Halide
Product
Yield (%)b
R1
R2
Temperature (°C)
Time (h)
O
R1
R2
O
P
O
R2
1
2
3c
4c
5c
6
I
I
H
OMe
H
Me
OMe
C(O)OEt
H
160
155
173
170
163
173
180
4
2
6
6
2.5
6.5
8
93 (98)
(95)
Br
Br
Br
Br
Cl
(98)
94 (98)
87 (97)
87 (97)
0
7
O
O
P
O
P
O
O
O
8
9
Br
Cl
Br
Cl
178
180
2.5
8
97 (99)
0
O
O
P
Br
O
P
O
O
O
10
85 (98)
Br
163
5
O
P
C(O)OEt
O
O
I
11
12
(12)
C(O)OEt
180
160
8
O
P
S
O
O
O
S
89 (97)
I
4
O
P
CH2Br
O
13d
92 (97)
CH2
175
5
(continued on next page)

280
Q. Yao, S. Levchik / Tetrahedron Letters 47 (2006) 277–281
Table 2 (continued)
Entry
Halide
Product
Yield (%)b
(35)
R1
R2
Temperature (°C)
Time (h)
O
P
14d
I
O
O
O
C5H10
180
8
I
O
15e
(32)
P
O
180
8
a 0.06 mol TPPi, TPPi/halide/TEPi/NiCl2 = 2/3/1.1/0.15, NiCl2 was used, unless otherwise specified.
b Isolated yield with 31P NMR yield in parentheses.
c NiCl2 and NiBr2 were compared, no difference in the yield and rate was observed.
d No catalyst used.
e Tris(2,4-di-tert-butylphenyl)phosphite used.
(ArO)3P + (EtO)3P
NiCl2
(ArO)2(EtO)P + (ArO)(EtO)2P
O
P
O
P
O
P
ArO
OEt
R1
EtO
OEt
R1
ArO
OAr
R1
+
OAr
+
+
ArO
R1
OAr
P
Ni
L
X-
(I)
(II)
(III)
X
EtX
OAr
OAr
ArO
R1
O
P
+
R1
OAr
ArO
NiL2
P
Ni
L
-O
OEt
OAr
P
Ni
L
X
OR
R1
OR
RO
OEt
RO
R1
+
P
Ni
L
P
Ni
L
O
X
X
L
R1
R1
R1
R1
OAr
ArO-
EtX
ArO
O
P
OEt
P
+
OR
O
Ni
L
RO
L
NiL2
ArO
OEt
P
Ni
L
X-
R1
OAr
ArO
L
P O
R1
O
O
Ni
L
L
P
OAr
R1
ArO P OEt
(IV)
R1
R1
(I)
(II)
L= triaryl phosphite, triethyl phosphite, and mixed phosphites, R=Ar or Et
Scheme 2. Plausible mechanism of the formation of diaryl arylphosphonates.
simplicity and general utility should find wider applica-
tion in the synthesis of diaryl arylphosphonates or acti-
vated alkylphosphonates. In the future, we will study the
possibility of using non-activated aryl chlorides and
alkyl halides for this type of reaction.
Acknowledgements
We thank Supresta for permission to publish this work.
We also thank Professor Edward Weil at Polytechnic
University for helpful discussions.

Q. Yao, S. Levchik / Tetrahedron Letters 47 (2006) 277–281
281
10. Balthazor, T. M.; Miles, J. A.; Stults, B. R. J. Org. Chem.
1978, 43, 4538.
Supplementary data
General experimental procedures, 1H, 31P, and 13C
NMR data and spectra of compound 1, 3, 4, 6, 8–10,
and 13. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.tetlet.2005.11.028.
11. Balthazor, T. M. J. Org. Chem. 1980, 45, 2519.
12. For a recent example on the transition-metal catalyzed
Michaelis–Becker reaction to construct P-aryl bond please
see: Gelman, D.; Jiang, L.; Buchwald, S. L. Org. Lett.
2003, 5, 2315.
13. A recent book dedicated to phosphonate chemistry does
not discuss diaryl arylphosphonates, probably because of
the lack of their synthetic methods by the Arbuzov
reactions. Savignac, P.; Iorga, B. Modern Phosphonate
Chemistry; CRC Press: Boca Raton, 2003.
14. Buchner, B.; Lockhart, L. J. Am. Chem. Soc. 1951, 73,
755.
15. Shim, K. S. U.S. Patent 4 252 740, 1981.
16. Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259.
17. Trimethyl phosphite has an extreme odor, long chain alkyl
phosphites generate alkyl halides of high boiling points,
which are difficult to remove.
References and notes
1. Papazoglou, E. S. Flame Retardants for Plastics. In
Handbook of Building Materials for Fire Protection;
Harper, C. A., Ed.; McGraw-Hill: New York, 2004; pp
4.1–4.88.
2. Weil, E. D. Phosphorus Flame Retardants. In Kirk-
Othmer Encyclopedia of Chemical Technology; 4th ed.;
John Wiley: New York, 1993; Vol. 10, pp 976–998.
3. Maryanoff, B.; Reitz, A. Chem. Rev. 1989, 89, 863.
4. Engel, R. Chem. Rev. 1977, 3, 349.
18. Balthazor, T. M.; Grabiak, R. C. J. Org. Chem. 1980, 45,
5425.
19. Oxidative addition of aryl halides to a phosphine Ni(0)
complex is known. Fahey, D. R. J. Am. Chem. Soc. 1970,
92, 402. Also see Ref. 20.
5. Freeman, G. A.; Rideout, J. L.; Miller, W. H.; Reardon,
J. E. J. Med. Chem. 1992, 35, 3192.
6. Bhattachary, A. K.; Thyarajan, G. Chem. Rev. 1981, 81,
415.
7. Kirby, A. J.; Warren, S. G. The Organic Chemistry of
Phosphorus. In Reaction Mechanisms in Organic Chemis-
try; Eaborn, C., Chapman, N. B., Eds.; Elsevier Pub.
Com.: New York, 1967; pp 37–44.
20. Fahey, D. R.; Mahan, J. E. J. Am. Chem. Soc. 1977, 99,
2501.
21. The cobalt and palladium analogs of the Ni+ ionic
complex (IV) in Scheme 2 have been demonstrated.
Landon, S. J.; Brill, T. B. Inorg. Chem. 1984, 23, 4177.
Also see Ref. 22.
8. Tavs, P. Chem. Ber. 1970, 103, 2428.
9. Axelrad, G.; Laosooksathit, S.; Engel, R. J. Org. Chem.
1981, 46, 5200.
22. Bao, Q. B.; Geib, S. J.; Rheingold, A. L.; Brill, T. B. Inorg.
Chem. 1987, 26, 3443.
23. Littke, A.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 6989.