Chemistry Letters Vol.38, No.7 (2009)
739
ArP(O)(OEt)2
P(O)(OEt)2
NH2
P(O)(OEt)2
NH2
PhP(O)(OEt)2 + ArPPh2
PPh3
NEt3
F3C
F3C
PPh3
Pd(0)(PPh3)3
ArH
P(O)(OEt)2
P(O)(OEt)(O)(NEt4)
23
ArX
PPh3
PPh3
24
Ar PPh2
PPh3
Scheme 3. Dealkylation of arylenediphosphonate 23.
Ar Pd H
PPh3
Pd
Ph
P(O)(OEt)2
Ph PPh2
Ar Pd P(O)(OEt)2
PPh3
PPh3
Ar Pd
X
PPh3
phite the reaction selectivity increases and the diphosphonate 19
was isolated in 53% yield among reduced products i.e diethyl
2-aminophosphonate (20) (4%), diethyl 4-aminophosphonate
(21) (10%), and diethyl phenylphosphonate (7) (12%) (Entry 8).
Dibromide 22 gave the desirable diphosphonate 23 in 40%
yield (Entry 9). Surprisingly, the yield of the reaction leading
to phosphonate 23 decreased when the reaction time was pro-
longed up to 3 days. Applying a more polar eluting solvent
mixture during a chromatographic purification (10% of MeOH
in CH2Cl2), the mono ammonium salt 24 has been separated
(Scheme 3). Precedent observations of a phosphonate dealkyla-
tion in the presence of tertiary amines11 or hydrazine12 have been
described. Scale-up experiments using 5–10 g bromides 5, 11,
13, and 15 gave the desired phosphonates according to the yields
given in Table 1.
HOEt
PPh3
(NR3H)X
Ar PPh2
Pd
Ph
X
HP(O)(OEt)2
+ NR3
HP(O)(OEt)2
+ NR3
(NR3H)X
PPh3
Scheme 2. Simplified mechanism for the Pd-catalyzed phos-
phorylation, the aryl–aryl interchange and the hydrodebromina-
tion reactions.
us to suggest that compound 7 results from the aryl–aryl inter-
change reaction8 in the intermediate Pd(PPh3)2ArX complex
formed during the phosphorylation catalytic cycle (Scheme 2).9
In fact, even this side-reaction has never been described in
the palladium-catalyzed phosphorylation reaction, it was sug-
gested as a possible pathway of the formation of phosphine-de-
rived by-products in the palladium-catalyzed cross-coupling re-
actions.10 Our additional experiments aiming to optimize the
yield of compound 6 were unsuccessful. When dicyclohexyl-
methylamine (NCy2Me) was employed as a base a similar prod-
uct distribution was observed. The reaction in the presence of
2 mol % Pd(OAc)2/6 mol % PPh3 resulted in lower yields of
by-product 7 without observing better yields of the product 6
due to the incomplete conversion of the starting bromide.
When bromide 8 bearing an electron-withdrawing group
was allowed to react under standard conditions incomplete reac-
tion was observed (Entry 4). Interestingly, increasing of catalyst
loading or ratio of [Pd]/PPh3 did not afford a satisfactory con-
version of the starting bromide. However, the reaction was com-
pleted with twice the amount of diethyl phosphite and the desir-
able product was obtained in 71% yield.
Naphthalene derivatives 11 and 13 were more reactive as
compared to anilines (Entries 5 and 6). Compound 11 reacted
with diethyl phosphite even in the presence of 2 mol % of the
palladium precursor. Both NEt3 or NCy2Me can be used as a
base, leading to comparable yield (88–91%) of the phosphonate
12 (Entry 5). The N-methyl derivative 13 was less reactive but
the reaction was still selective. A good yield (85%) for the syn-
thesis of phosphonate 14 was obtained in the presence of
5 mol % of the catalyst precursor (Entry 6). Further increase of
the steric bulk on the aromatic bromide results in a significant
decrease of the product yield.
In conclusion, we have developed an efficient catalytic route
to diethyl arylphosphonates bearing amino and alkylamino
groups on a benzene, naphthalene, or anthracene ring. These
conditions can be applied for the synthesis of amino-substituted
arylenediphosphonates if corresponding arylene dibromides are
used as starting compounds.
References and Notes
1
2
b) T. Hirao, T. Masunaga, N. Yamada, Y. Ohshiro, T. Agawa, Bull.
Soc. Chim. Fr. 1995, 132, 290. d) N. Defacqz, B. de Bueger, R.
b) G. P. Basmadjian, S. Singh, B. Sastrodjojo, B. T. Smith, K. Avor,
F. Chang, S. L. Mills, T. W. Seale, Chem. Pharm. Bull. 1995, 43,
3
´
1902. c) V. Penicaud, C. Maillet, P. Janvier, M. Pipelier, B. Bujoli,
4
5
6
7
8
9
B. R. Aluri, M. K. Kindermann, P. G. Jones, I. Dix, J. Heinicke, Inorg.
A. Bessmertnykh, C. Morkos Douaihy, S. Muniappan, R. Guilard, Syn-
thesis 2008, 1575.
Supporting Information is available electronically on the CSJ-Journal
With bromoanthracene 15 the reaction was incomplete even
when 10 mol % of the catalytic system was employed and the
product 16 was isolated in 37% yield (Entry 7). Under optimal
conditions, i.e. using 2 equiv of diethyl phosphite, 10 mol % of
the catalyst system and NCy2Me, the starting bromide was fully
consumed but the phosphonate 16 was only obtained in 49%
yield among 2-aminoanthracene (17).
M. C. Kohler, T. V. Grimes, X. Wang, T. R. Cundari, R. A. Stockland,
We then studied the possibility of diphosphorylation of aryl-
ene dibromides in a one-step procedure. When 2,4-dibromoani-
line (18) reacted with 2.4 equiv of diethyl phosphite in the pres-
ence of 10 mol % of the catalyst system several products were
formed in the reaction mixture. With an excess of diethyl phos-
11 K. Gorog, L. Bodnar, E. Dudar, M. Kocsis, S. Gaal, M. Tasnadi, E.
Egyhazi, V. M. Varga, I. Kajati, G. Kis, J. Molnar, B. Toth, I. Cserhati,
T. Kaptas, S. Csete, U.S. Patent 4 675 431, 1987.
12 E. N. Dolzhnikova, E. N. Tzvetkov, G. S. Petrova, Zh. Obshch. Khim.
1976, 46, 1903.