New feature of Friedel-Crafts phosphonation of anisoles: Unexpected in situ methylphosphorylation reaction

Article abstract of DOI:10.1055/s-1999-2706

Anisoles 1, reacting with AlCl₃ and PCl₃ with appropriate reagent ratios, give, in good yields, the corresponding diaryl methylphosphonates 2 or the methylphoshinates 3b,c and the methylphosphine oxides 4b,c. This unexpected in situ methylphosphorylation explains the reported limited and conflicting results to obtain methoxy-substituted arylphosphonous dichloride with the same reagents. A suggested mechanism is also reported.

Full text of DOI:10.1055/s-1999-2706

822
LETTER
New Feature of Friedel-Crafts Phosphonation of Anisoles: Unexpected In Situ
Methylphosphorylation Reaction
Graziano Baccolini,* Carla Boga
Dipartimento di Chimica Organica “A. Mangini”, Università di Bologna, Viale Risorgimento 4, I-40136 Bologna, Italy
Fax +39 (51) 6443654; E-mail: baccolin@ms.fci.unibo.it
Received 8 February 1999
with various molar proportions of reagents and different
Abstract: Anisoles 1, reacting with AlCl3 and PCl3 with appropri-
reaction conditions. The reaction of 1b with an appropri-
ate reagent ratios, give, in good yields, the corresponding diaryl
ate ratio of the reagents, afforded a mixture containing as
major products the corresponding diaryl methylphospho-
methylphosphonates 2 or the methylphoshinates 3b,c and the
methylphosphine oxides 4b,c. This unexpected in situ methylphos-
phorylation explains the reported limited and conflicting results to ryl derivatives 2b, 3b and 4b arising from an unexpected
obtain methoxy-substituted arylphosphonous dichloride with the
same reagents. A suggested mechanism is also reported.
in situ methylphosphorylation (Scheme1).
Key words: anisoles, Friedel-Crafts reactions, phosphorylation,
phosphorus compounds, phosphorus trichloride
Since the last century, the Friedel-Crafts-type reaction1
using PCl3 and AlCl3 has been of interest for the direct
phosphonation of an aromatic ring. However, this reaction
is very sensitive to the type of substituents present on the
aromatic ring.2 In particular, very limited success and con-
flicting results have been reported when the reaction is
carried out with anisoles. For example, anisole,3,4 upon
treatment with PCl3 and AlCl3 gave complex reaction
mixtures depending on the quality of AlCl3 with poor
yields of methoxy-substituted arylphosphonous dichlo-
ride or phenylphosphorodichloridite and large amounts of
Scheme 1
undistillable residues:
After several repeated attempts with small variations of
reagent ratios and reaction conditions we could control the
regiochemistry of this methylphosphorylation reaction in
order to obtain mainly the diaryl methylphosphonates 2
(see Scheme 1). The best results were obtained with a
1: AlCl3: PCl3 ratio of 1:0.6:1, at 70-80 °C in atmosphere
of dry nitrogen and without solvent (Table 1). In the case
of 1c the reaction was complete in 7 h while in the other
cases in about 10-15 h. A stoichiometric excess of PCl3
was useful also to ease the stirring and the homogeneity of
the reaction mixture. Compounds 2 have been fully
To increase the yield and the selectivity of this phospho-
nation reaction the use4a of SnCl4 and very recently the
use4c of BiCl3 have been reported. In the past years we
discovered5 an unusual reaction of PCl3 and AlCl3 with
thioanisoles which gave a new heterocyclic system (fused
1,2,3-benzothiadiphosphole) containing the P-PS2 unit.
Recently, the course of this unexpected diphosphonation
reaction has been studied6 and we have found that the re-
action follows a complex multistep pathway which was a
priori unpredictable and that the outcome of the reaction
is very dependent on the ratio of the reagents and on the
good quality of AlCl3 (sublimated prior to use) which in
this reaction is a true reagent and not a catalyst.
1
characterised7 by H and 31P NMR and spectral data of
2a,b,c,d were in accord with existing literature data.8
When the reaction was run with 1b,c and a reagent ratio
of 1:0.3:1, with the same sublimated AlCl3, methylphos-
phinates 3b,c and methyl phosphine oxides 4b,c, respec-
tively, were also formed as major products in a total yield
of about 60% besides small amounts of the expected 2b,c
(Table 1). In contrast, when the reaction was run on 1a,d,e
the major product observed is always the corresponding
methyl phosphonate 2a,d,e but in lower yield, and no ap-
preciable amount of the related ring substituted products
3 and 4 is detected.
Then, we thought to study again the reaction of anisoles 1
with AlCl3 and PCl3, in order to find results which might
explain the reported3,4 formation of large amounts of un-
distillable residues. Several reactions were carried out
Synlett 1999, No. 6, 822–824 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Unexpected In Situ Methylphosphorylation of Anisoles
823
Table 1.
signal of 2c. Thus, the two signals might be assigned to in-
termediates A and B (Scheme 2), although more unequiv-
ocal data are necessary for their exact structural
determination. However, their shifts are consistent with
their ionic representation.9 Evidence for the initial equilib-
rium:
1
comes from the H NMR spectrum of the AlCl3-anisole
CDCl3 solution at 60 °C: the methoxy and aromatic pro-
tons are shifted selectively downfield, relative to anisole
alone.
Diaryl methylphosphonates 2 are useful as both enzyme
model substrates,10 fungicides11 and reactive intermedi-
ates.12 The literature describes two basic routes for their
preparation. One involves the condensation of meth-
ylphosphonic dichloride with the appropriately substitut-
ed phenol.13 The second procedure consists of reacting a
triarylphosphite with methyl iodide alone14 or in methan-
olic solution8 for 3 h at 200-250 °C. It should be noted that
methylphosphonic dichloride is not readily available
while in the second procedure the reaction temperature
does not permit to obtain the corresponding p-halogen di-
aryl methylphosphonates in good yields. In our procedure,
lower temperatures (70-80 °C) allow to obtain these halo-
gen derivatives 2d,e in good yields.
This could be due to the fact that the presence of the me-
thyl or methoxy substituent activates the aromatic ring to-
wards electrophilic substitution. The products 3b,c and
4b,c have been separated by flash chromatography on sil-
ica gel and have been fully characterised.7
The major issue was to determine whether products 3b
and 4b are ortho or meta substituted with respect to the
Me substituent. As reported in the literature,4a a methyl
group ortho to a P atom should show a significant cou-
pling. In our case no coupling was observed, in agreement
with the structures we are reporting. NOE experiments
obtained by irradiating the methyl group, further con-
firmed our assignment.
Probably, compounds 2,3,4 might be a part (or all) of the
so-called undistillable residues reported previuosly.
In conclusion, we have found that the reaction of anisoles
1 with AlCl3 and PCl3, with adequate reagents ratio, lead
to the diaryl methylphosphonates 2a-e or the methylphos-
phinates 3b,c and the methylphosphine oxides 4b,c in
good yields. Formation of these compounds is due to a
new unexpected in situ methylphosphorylation which
could be useful for obtaining other new P-methyl substi-
tuted compounds and gives also an explanation about the
limited success in the phosphonation of anisoles obtained
so far with the same reagents.
Acknowledgement
Scheme 2
Investigation supported by University of Bologna (funds for selec-
ted research topics A.A 1997-1999). Thanks are also due to Con-
siglio Nazionale delle Ricerche (CNR), Italy, for financial support.
In order to obtain supporting evidences for a possible
pathway of this unexpected methylphosphorylation, ali-
quots of a reaction mixture between 1c, AlCl3 and PCl3
References and Notes
31
were analyzed by P NMR spectroscopy. After severals
(1) Michaelis, A. Chem.Ber., 1879, 12, 1009.
(2) Fild, M.; Schmutzler, R. In Organic Phosphorus Compounds,
Vol.4, G. Kosolapoff and L.Maier Eds., Wiley-Interscience,
1972, p 79.
minutes the 31P NMR spectrum showed only the signal of
PCl3 (δ = 219.2 ppm).9 After 3h we noted two prevalent
strong signals at δ = 178 and 104 ppm as broad quartets.
During the reaction no signal corresponding to MePCl2
(δ = 192.1 ppm) or its complexes (δ = 131.9 and 97.5
ppm)9 was observed indicating a possible concerted
mechanism involving the migration of the methyl group
on the phosphorus atom, as depicted in Scheme 2. At the
end of the reaction we noted the prevalence of the second
signal and after addition of water we noted the disappear-
ance of the above two signals and the appearance of the
(3)(a) Michaelis, A. Justus Liebigs Ann.Chem., 1896, 294, 1.
(b) Kunz, P. Ber., 1894, 27, 2559
(4) (a) Miles, J. A.; Beeny, M. T.; Ratts. K.W. J.Org.Chem.,
1975, 40, 343. (b) Engel, R. In Synthesis of Carbon-
Phosphorus Bonds, CRC Press, Inc. Boca Raton, Florida,
1988, chap.5 and 6. (c) Siméon, F.; Jaffrès, P.; Villemin, D.
Tetrahedron 1998, 54, 10111.
(5) (a) Baccolini, G.; Mezzina, E.; Todesco, P.E.; Foresti, E.
J.Chem.Soc.,Chem.Commun., 1988, 304. (b) Baccolini, G.;
Synlett 1999, No. 6, 822–824 ISSN 0936-5214 © Thieme Stuttgart · New York

824
G. Baccolini, C. Boga
LETTER
Mezzina, E.; Todesco, P.E. J. Chem.Soc.,Perkin Trans.1,
1988, 3281.
(6) Baccolini,G.; Beghelli, M.; Boga, C. Heteroatom Chem.
1997, 551.
3c as viscous oil,1H NMR (300MHz, CDCl3): δ(ppm) 7. 46
(dd, 1H, J = 14.4, J = 3.2 Hz), 7.05(dd, 1H, J = 9.1, J = 3.2
Hz), 7.01-6.8 (m, 4H), 6.72 (dd, 1H, J = 9.1 Hz, J = 6.8 Hz),
3.88 (s, 3H), 3.75 (s, 3H), 3.67 (s, 3H), 1.89 (d, 3H, J = 15.4
Hz); 31P NMR(121.47 MHz, CDCl3): δ(ppm) 40.9 (broad
quintet, J = 15.4 Hz); HRMS: Found, 322.0967, Calc. for
C16H19O5P, 322.0970.
(7) General procedure for preparation of 2a-e, 3b,c and 4b,c
A solution of 0.1 mol of anisole 1, 0.1 mol (8.7 mL) of
phosphorus trichloride, and 0.06 mol of aluminum trichloride
(sublimated prior to use) was refluxed (70-80 °C) and stirred
under dry nitrogen for specified lenght of time (see Table 1).
At the end, the reaction mixture appeared as a viscous oil,
which was treated with a mixture of ice and dichloromethane;
the organic layer was washed with a 5% aq. NaOH solution
and two times with water. After removal of the solvent,
compounds 2 were separated by distillation under reduced
pressure or by flash chromatography in 50-70% yield.
In similar manner compounds 3b,c, and 4b,c were obtained
using a 1b,c : AlCl3: PCl3 ratio of 1:0.3:1.
4b as greasy solid, 1H NMR(300MHz, CDCl3): δ(ppm) 7.44
(dd, 2H, J = 14.0 Hz, J = 3.7 Hz), 7.27 (m, 2H), 6.81 (dd, 2H,
J = 8.4 Hz, J = 5.5 Hz), 3.70 (s, 6H), 2.28 (s, 6H), 2.10 (d, 3H,
J = 14.6 Hz); 31P NMR(121.47 MHz, CDCl3): δ(ppm) 30.48
(m); HRMS: Found, 304.1222, Calc. for C17H21O3P,
304.1228.
4c as greasy solid, 1H NMR(300MHz, CDCl3): δ(ppm) 7.23
(dd, 2H, J = 14.8 Hz, J = 2.8 Hz), 7.03 (dd, 2H, J = 8.9Hz,
J = 2.8Hz), 6.81 (dd, 2H, J = 8.9 Hz, J = 7.4 Hz), 3.76 (s, 6H),
3.70 (s, 6H), 2.15 (d, 3H, J = 14.6 Hz); 31P NMR(121.47
MHz, CDCl3): δ(ppm) 30.30 (m); HRMS: Found, 336.1122,
Calc. for C17H21O5P, 336.1127.
Selected spectral data for new compounds 2e, 3b,c and 4b,c
All gave satisfactory elemental analyses. Yields are not
optimized. The 31P signals are given in ppm downfield from
85% H3PO4.
(8) Honig, M.L.; Weil, E.D. J.Org.Chem., 1977, 42, 379.
(9) Symmes, C., Jr.; Quin, L.D. J.Org.Chem., 1978, 43, 1250.
(10) Brass, H. J.; Bender, M. L., J Am. Chem. Soc., 1973, 95, 5391.
(11) Roy, N.K.; Nidiry, E.S.; Vasu, K.; Bedi, S.; Lalljee, B.; Singh,
B. J.Agric.Food Chem., 1996, 44, 3971.
2e as viscous oil, 1H NMR(300MHz, CDCl3): δ(ppm) 7.44(dd,
4H, J = 9.1, J = 0.7 Hz), 7.07(dd, 4H, J = 9.1, J = 1.4 Hz),
1.81 (d, 3H, J = 17.7 Hz); 31P NMR(121.47 MHz, CDCl3):
δ(ppm) 25.3 (bq, J = 17.7 Hz), HRMS: Found, 403.8810,
Calc. for C13H11Br2O3P, 403.8812.
(12) Coover Jr. H.W.; McConnell, L.R.; McCall, M.A.
Ind.Eng.Chem. 1960, 52, 409.
3b as viscous oil, 1H NMR(300MHz, CDCl3): δ(ppm) 7. 75
(dd, 1H, J = 13.7, J = 2.4 Hz), 7.29(dd, 1H, J = 8.2, J = 2.4
Hz), 7.15 (bs, 4H), 6.81 (dd, 1H, J = 8.2 Hz, J = 7.2 Hz), 3.90
(s, 3H), 2.27 (s, 3H), 2.24 (s, 3H), 1.88 (d, 3H, J = 15.4 Hz);
31P NMR(121.47 MHz, CDCl3): δ(ppm) 40.8 (quintet of
doublets, J = 15.4 Hz, J = 7.2 Hz); HRMS: Found, 290.1071,
Calc. for C16H19O3P, 290.1072.
(13) Hoskin, F.C. Can. J. Chem., 1957, 35, 581.
(a) Michaelis, A.; Kaehne, R. Ber., 1898, 31, 1048. (b)
Behrman, E.J.; Biallas, M.J.; Brass, H.J.; O'Edwards, J.; Isaks,
M. J.Org.Chem. 1970, 35, 3063.
Article Identifier:
1437-2096,E;1999,0,06,0822,0824,ftx,en;G06999ST.pdf
Synlett 1999, No. 6, 822–824 ISSN 0936-5214 © Thieme Stuttgart · New York