Reconsideration of the base-free batch-wise esterification of phosphorus trichloride with alcohols

Article abstract of DOI:10.1021/op049958v

Batch-wise esterification of phosphorus trichloride with different alcohols in the absence of base and cleavage of the reaction products by the HCl released in course of the reaction were reinvestigated. The role of the kind of alcohol, mixing order of reagents, temperature, time of reaction, and excursion of gaseous HCl in the proportional composition of the reaction products were studied. Considering the mechanism of esterification and cleavage of the products, the optimized conditions to retain the cleavage process and high-yield production of dialkyl hydrogen phosphonates were determined.

Full text of DOI:10.1021/op049958v

Organic Process Research & Development 2004, 8, 401−404
Technical Notes
Reconsideration of the Base-Free Batch-Wise Esterification of Phosphorus
Trichloride with Alcohols
H. Fakhraian* and A. Mirzaei
Department of Chemistry, Imam Hossein UniVersity, Tehran, Iran
Scheme 1. Mechanism pathway proposed for the
esterification of PCl3 and formation of DAHP
Abstract:
Batch-wise esterification of phosphorus trichloride with differ-
ent alcohols in the absence of base and cleavage of the reaction
products by the HCl released in course of the reaction were
reinvestigated. The role of the kind of alcohol, mixing order of
reagents, temperature, time of reaction, and excursion of
gaseous HCl in the proportional composition of the reaction
products were studied. Considering the mechanism of esteri-
fication and cleavage of the products, the optimized conditions
to retain the cleavage process and high-yield production of
dialkyl hydrogen phosphonates were determined.
alcohols (such as EtOH and i-PrOH).4,7 The reaction yield
for MeOH was increased to 70-80% in continuous pro-
cess.9,10
Introduction
Dialkyl hydrogen phosphonates (DAHP) are considered
as the essential intermediates to prepare key functionalized
phosphonic acid derivatives, i.e., herbicides, fungicides,
insecticides, and antibiotics.1 Dimethyl hydrogen phospho-
nate is used as a fungicide2 and an intermediate in the
production of trichlorfon (insecticide Dipterex).3,4
The reaction of the appropriate alcohol with phosphorus
trichloride (known as the McCombie process5) is the most
common method to prepare DAHP. This reaction can be
performed in batch4-8 or continuous process.9-12
The reaction yield in batch process, reported as 35-40%
for MeOH, was increased to 80-90% for more bulky
Dimethyl and diethyl hydrogen phosphonates were pre-
pared by this method. Diisopropyl hydrogen phosphonate
and the higher homologues could also be prepared by another
method in which one mole of phosphorus trichloride was
reacted with two moles of alcohols and one mole of water
to avoid conversion of one mole of alcohol to alkyl halide.13
It is said that usually the trialkyl ester is first formed and
then cleaved by the hydrogen chloride to give an alkyl
chloride and a diester (DAHP).14 This reaction proceeded
by way of quasi-phosphonium ion as an intermediate which
upon alkylation, decomposes to a DAHP and an alkyl halide
(Scheme 1).15
However, this hypothetical mechanism has been ques-
tioned. The over-all reaction consists of two steps: (1) the
replacement of chlorine by an alkoxy group (esterification),
through which hydrogen chloride is liberated and (2) the
reaction of hydrogen chloride with the product (HCl cleav-
age) and a Michaelis-Arbusov rearrangement which results
in elimination of an alkyl chloride. This causes the conversion
of triple-connected to quadruple-connected phosphorus. A
number of different mechanisms may be postulated that differ
only with respect to the extent in which step 1 precedes or
follows step 2 (Scheme 2).
* Corresponding author. Fax: +98-7313938. E-mail: fakhraian@yahoo.com.
(1) Edmundson R. S. The Chemistry of Organophosphorus Compounds;
Hartley, F. R., Ed.; John Wiley & Sons: New York, 1996; Vol. 4, p 147.
(2) Philagro Ger. Offen. Patent 2,510,034; Chem. Abstr. 1976, 84, 26856t.
(3) Haraszti, J.; Dudas, J.; Marosvolgyi, S.; Papp, A.; Dudas, J.; Kerekes, F.;
Molnar, I.; Nagy, L. Hungarian Teljes Patent 6635; Chem. Abstr. 1974,
80, 70963s.
(4) Barthel, W. F.; Giang, P. A.; Hall, S. A. J. Am. Chem. Soc. 1954, 76,
4186.
(5) Cook, H. G.; McCombie, H.; Saunders: B. C. J. Chem. Soc. 1945, 873.
(6) (a) Vilceanu, R.; Schulz, P.; Kurunczi, L. Rom. Patent 60,534; Chem. Abstr.
1978, 89, 42431j. (b) Vilceanu, R.; Schulz, P.; Kurunczi, L. Rom. Patent
63,090; Chem. Abstr. 1980, 92, 22038v. (c) Craiu, C.; Dumitrescu, G.;
Kurunczi, L.; Schulz, P.; Vilceanu, R. Rom. Patent 59,706; Chem. Abstr.
1978, 89, 5921y.
(7) Misra, A. K.; Lal, J. B. Indian J. Appl. Chem. 1965, 28, 187.
(8) Gabor, S.; Karoly, J.; Gyula, S. Fr. Demande Patent 2,392,029; Chem. Abstr.
1979, 91, 140333q.
The formed hydrogen chloride can cause cleavage of
DAHP and reduction of product yield in high extent if the
(9) Cambell, C. H.; Chadwick, D. H.; Kaufman, S. Ind. Eng. Chem. 1957, 49,
1871.
(10) Baanauckas, C. F.; Hodan, J. J. U.S. Patent 3,331,895; Chem. Abstr. 1967,
67, 63764f.
(13) Gann, P. W.; Heider R. L. U.S. Patent 2,692,890; Chem. Abstr. 1955, 49,
12529i.
(11) Stauffer Chemical Co. Neth. Appl. Patent 6,612,966; Chem. Abstr. 1967,
67, 53665k.
(12) Sauli, V. Chem. Prum. 1973, 23, 554; Chem. Abstr. 1974, 80, 70233d.
(14) Cambell, C. H.; Chadwick, D. H. J. Am. Chem. Soc. 1955, 77, 3379.
(15) Chadwick, D. H.; Watt, R. S. Phosphorus and its Compounds; Van Wazer,
J. R., Ed.; Interscence Publishers Inc.: New York, 1966; Vol. 1, p 1267.
10.1021/op049958v CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/04/2004
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
401

Scheme 2. Mechanism pathway for the esterification of PCl3 and cleavage of the products by HCl
Table 1. Proportional composition (%)a of the phosphonic acid derivatives after the esterification of PCl3 with appropriate
alcoholb
PCl3
ROH
(mol, mL)
solvent
(mL)
time temp (RO)2P(O)H 4 (RO)(HO)P(O)H 5 (HO)2P(O)H 6 (RO)ClP(O)H 8
exptb (mol, mL)
(h)
(°C)
(%)
(%)
(%)
(%)
1
2
3
4
5c
6
7
8
0.05, 4.4
0.05, 4.4
0.05, 4.4
0.05, 4.4
0.05, 4.4
0.5, 44
MeOH (0.15, 6)
MeOH (0.15, 6)
MeOH (0.15, 6)
MeOH (0.15, 6)
MeOH (0.15, 6)
i-PrOH (1.5, 115)
n-BuOH (0.3, 27.5)
-
1
25
25
25
0
0
0
73
58
83
84
91
97
98
98
26
36
14
-
9
3
2
2
1
6
-
-
3
16
-
-
-
-
CHCl3 (8)
CHCl3 (8)
CHCl3 (8)
CH2Cl2 (8)
CH2Cl2 (48)
CH2 Cl2 (16)
0.5
0.5
0.5
1
0.5
0.5
0.5
-
-
-
-
-
-
0.1, 8.8
0.5, 44
0
0
i-BuOH (1.5, 138.5) CH2Cl2 (48)
a Based on 31P NMR spectroscopy. b Addition order of the reagents in experiments 1 and 2 (dropwise addition of PCl3 to alcohol) was reversed in experiments 3-8.
c Vacuum stripping of HCl was performed in experiments 5-8.
reaction conditions are not controlled. The hydrogen chloride
can be eliminated by saturating the reaction mixture with
dry ammonia and filtering the resulting solid ammonium
chloride prior to fractional vacuum distillation of the alkyl
halide and DAHP.
Another method for obviating such a side reaction is to
react the alcohol and phosphorus trichloride in a continuous
system under conditions in which the hydrogen chloride is
lost at once as a gas and the reaction heat is removed by the
reaction products. Here, the presence of a low-boiling
temperature solvent reduces the amount of heat in reaction
medium as well.9-12
Attempts have been made to ameliorate the yield of the
above reaction in batch process, but some parameters of the
reaction influencing the yield have not been deeply consid-
ered yet. High-yield production of DAHP (4), depends on
the relative rate of esterification and HCl cleavage reactions
and on the reagent mixing order. The intensive and high
exothermic reaction of phosphorus trichloride with alcohol
prohibits mixing equal amounts of the two reagents together,
and one should be added dropwise to the other. Thus, the
order of each reagent in the mixing procedure can influence
the relative amount of esterification and HCl cleavage
products. This fact has not been properly investigated, and
in most cases, phosphorus trichloride was added dropwise
to alcohol.
phosphorus trichloride), temperature, time of reaction, and
immediate removal of the formed hydrogen chloride.
Result and Discussion
The esterifications of phosphorus trichloride by the
appropriate alcohol (ROH; R ) Me, i-Pr, n-Bu, i-Bu) have
been performed under different conditions, and the product
compositions of reaction mixture were characterized by 31
P
NMR spectroscopy (Table 1). The esterification of phos-
phorus trichloride by MeOH was done in 0.05 M scale, and
after obtaining the optimized conditions, the same reactions
were performed in higher molar scale (0.5-1 M) to prepare
diisopropyl, di-n-butyl, and diisobutyl hydrogen phospho-
nates.
All possible reactions relating to esteification (vertical
reactions) and HCl cleavage of the products (horizontal
reactions) are outlined in Scheme 2.
Dropwise addition of phosphorus trichloride to MeOH at
25 °C during 1 h in dry and solvent-less conditions, afforded
a mixture of 4, 5, and 6 (R ) Me) with proportional
composition of 73, 26, and 1% respectively (Table 1,
experiment 1; Figure 1a). Allowing this mixture to stand at
current conditions for 3 h, modified the proportional com-
position of 4, 5, and 6 to 40, 50, and 10%, respectively
(Figure 1b). It can be deduced that some part of 4 was
cleaved by the HCl present in the reaction medium to form
5 and 6.
In this contribution, batch process esterification of phos-
phorus trichloride by different alcohols have been reconsid-
ered. Special attention has been paid to the parameters such
as mixing order of reagents (precedent of alcohol or
31P NMR spectra of the reaction mixture displayed a
concentration-dependent upfield chemical shift of phosphorus
in 5 (0.3 ppm) and 6 (1.3 ppm). This should be attributed to
402
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development

Performing the same reaction in 0 °C inhibited the
formation of 5 and 6 (the first and second cleavage products)
and led to some extent of 8 (16%) (Table 1, experiment 4
and Figure 1e). Lowering the reaction temperature (from 25
to 0 °C), seemed to influence the esterification rate of
phosphorus trichloride (Table 1, experiments 3 and 4).
However, it had little effect on the HCl cleavage of the
principal intended product (4) in the time required for
completion of the esterification procedure (Table 1, experi-
ments 3 and 5).
Vacuum stripping of the HCl released in course of the
reaction during 1 h afforded the best yield of dimethyl
hydrogen phosphonate (91% based on 31P NMR and 84%
based on the product weighted after evaporation of solvent)
(Table 1, experiment 5). In this experimentsperformed in
CH2Cl2 as solvent during 1 hsdespite vacuum stripping,
some amount of cleavage product (5) was present in the
reaction mixture (9%), but the esterification procedure was
completed, and 8 was absent in the reaction mixture. CH2-
Cl2 has been used as solvent regarding industrial consider-
ation and also for maintaining the reaction temperature, not
exceeding 40 °C in an uncontrolled situation.
The alcohols such as i-PrOH, n-BuOH, and i-BuOH have
been successfully applied to the phosphorus trichloride
esterification beyond the optimized conditions of MeOH
(Table 1, experiments 6, 7, and 8).
In our experimental conditions, we never observed 1 and
2 or 7 and 9. Observation of 8 in certain conditions indicated
the formation of 2 before HCl cleavage. The esterification
of phosphorus trichloride proceeding by three successive
steps, mono-, di-, and triesterification and the influence of
mixing order of the reagents in the inhibition of HCl cleavage
of the product seemed to indicate the order of esterification
rate as PCl3 > Cl2P(OR) > ClP(OR)2. Effectively, if this
was not the case, the mixing order should not have affected
the mixture composition, and at once trialkyl hydrogen
phosphonate would have been formed and cleaved by HCl.
Esterification of 7, 8, and 9 seemed to be more difficult
than that of 1 and 2 (such as for their homologues in which
H is substituted by an alkyl group) and in our experimental
conditions should not occur. Thus, HCl cleavage of 3
occurred more readily than that of 1 and 2 and increased
with the degree of esterification. However, the minor amount
of production of 4, via esterification of 8, could not be
excluded.
It is worthy to note that freshly distilled dimethyl
hydrogen phosphonate was unstable and strongly hydrolyzed
when exposed to air moisture; therefore, it should be used
immediately after preparation. The otherssmore bulky
dialkyl hydrogen phosphonatessare more stable and were
not readily hydrolyzed. Stirring the reaction mixture after
the esterification of phosphorus trichloride by i-PrOH during
2 h at 25 °C did not alter the composition of the reaction
mixture. The reverse effect is observed concerning the
correspondents trialkyl phosphite. Indeed, trimethyl phosphite
is more resistant to the hydrolysis reaction than other trialkyl
phosphites with more bulky alkyl groups (R ) i-Pr, n-Bu,
and i-Bu).
Figure 1. 31P NMR spectrum (in CDCl3) of the reaction
mixture (a) after dropwise solvent-less addition of PCl3 to
MeOH at 25 °C (experiment 1, Table 1), (b) after stirring the
precedent mixture during 3 h at 25 °C, (c) after dropwise
addition of PCl3 to MeOH at 25 °C (experiment 2, Table 1);
(d) after dropwise addition of MeOH to PCl3 at 25 °C
(experiment 3, Table 1); (e) after dropwise addition of MeOH
to PCl3 at 0 °C (experiment 4, Table 1) (experiments 2-4 were
performed using CHCl3 as solvent).
the possibility of hydrogen bonding in concentrated solutions
of 5 and 6.
Performing the same experiment with CHCl3 as solvent
afforded a mixture in which the cleavage of the intended
principal product (4) and formation of 5 and 6 have been
increased (Table 1, experiment 2; Figure 1c). This could be
assigned to the augmentation of reaction medium volume
(because of the solvent presence) and dissolution of HCl
therein.
In another attempt to prepare dimethyl hydrogen phos-
phonate, the amount of reactants were tripled, but the quantity
of solvent remained constant as in experiment 5 (Table 1).
As a result, the yield of product (weighted) was raised to
about 98%, confirming the influence of the reaction medium
volume on the HCl cleavage of the product.
Inversion of the reagent mixing order (addition of MeOH
to phosphorus trichloride) caused the augmentation of the
proportional amount of 4 and formation of a minor quantity
of 8 (3%) (Table 1, experiment 3; Figure 1d).
Vol. 8, No. 3, 2004 / Organic Process Research & Development
•
403

1
Table 2. 31P and H NMR spectra of dialkyl hydrogen phosphonate DAHP (alkyl ) Me, i-Pr, n-Bu, i-Bu)
DAHP
31P NMR δ, J (ppm, Hz)
1H NMR δ, J (ppm, Hz)
(MeO)2P(O)H
(i-PrO)2P(O)H
(i-BuO)2P(O)H
(n-BuO)2P(O)H
8.5, 15.5 (dm, 700, 12)
1.9, 8.7 (dt, 680, 8)
5.1, 12.0 (dq, 690, 8)
5.1, 12.0 (dq, 690, 8)
3.5 (d, CH3, 7.5), 5.1, 7.9 (ds, PH)
1.3 (d, CH3, 7.5), 4.7 (m, CHO), 5.4, 8.2 (ds, PH)
0.6 (d, CH3, 7), 1.5 (m, CH), 3.4 (t, CH2O, 10), 5, 7.8 (ds, PH)
0.95 (t, CH3, 7.5), 1.4 (m, CH2), 1.7 (m, CH2), 4.0 (m, CH2O), 5.4, 8.1 (ds, PH)
Summary
Esterification of Phosphorus Trichloride with ROH
Determining the mechanism of the esterification and
cleavage reaction have permitted us to accomplish the
optimized conditions in batch process following which
DAHPs (alkyl ) Me, i-Pr, n-Bu, i-Bu) were prepared with
yield and purity of over 95%. The mixing order of reagents
(dropwise addition of ROH to phosphorus trichloride) and
the use of a minimum amount of solvent seem to influence
considerably the product yield. The esterification of phos-
phorus trichloride occurring in three steps (mono-, di-, and
triesterification) and the influence of mixing order of the
reagents on the yield of the products (DAHP) illustrate the
order of esterification as PCl3 > ROPCl2 > (RO)2PCl. Each
of the esterification products ((RO)PCl2, (RO)2PCl, and
(RO)3P) can also be subjected to the cleavage process by
hydrogen chloride released in the course of the reaction.
However, the influence of mixing order of the reagents on
product yield showed that the cleavage rate increased with
the degree of the phosphorus trichloride esterification and
was in the order of (RO)3P > (RO)2PCl > ROPCl2. It was
observed that dimethyl hydrogen phosphonate was unstable
and strongly sensitive toward the cleavage by HCl and
hydrolysis by the environmental moisture. Depending on the
number of carbons in the alkyl groups, the more bulky
DAHPs were more stable toward the above reaction.
(R ) Me, i-Pr, n-Bu, and i-Bu). Phosphorus trichloride (11
mL, 0.125 mol) and CH2Cl2 (4 mL) were placed in a 100-
mL flask immersed in an ice bath and equipped with a
magnetic stirrer and a condenser (the head of which was
connected to a water vacuum pump). Methanol (12.75 mL,
0.375 mol) diluted with CH2Cl2 (4 mL) was dropwise added
(0.5 mL/min) to the precedent mixture. The mixture was
stirred for another 10 min. Evaporation of CH2Cl2 afforded
13.5 g (0.122 mol, 98%) of dimethyl hydrogen phosphonate
of 91% purity (1H and 31P NMR characteristics in Table 2).
The above procedure was performed for esterification of
phosphorus trichloride by other alcohols, and after evapora-
tion of CH2Cl2 and alkyl chloridesformed in the course of
reactionsthe following results were obtained (1H and 31P
NMR characteristics in Table 2).
Esterification of phosphorus trichloride (44 mL, 0.5 mol)
by 2-propanol (115 mL, 1.5 mol) yielded 69.6 g (0.419 mol,
84%) of diisopropyl hydrogen phosphonate of 97% purity.
Esterification of phosphorus trichloride (8.8 mL, 0.1 mol)
by n-butyl alcohol (27.5 mL, 0.3 mol) yielded 18.3 g (0.094
mol, 94%) of di-n-butyl hydrogen phosphonate of 98%
purity.
Esterification of phosphorus trichloride (44 mL, 0.5 mol)
by isobutyl alcohol (138.5 mL, 1.5 mol) yielded 91 g (0.468
mol, 94%) of diisobutyl hydrogen phosphonate of 98%
purity.
Experimental Section
General Methods. NMR spectra were recorded on a
Bruker DPX-250 instrument (250 MHz for 1H and 100 MHz
for 31P). CDCl3 was used as solvent; chemical shifts were
reported in δ (ppm) from TMS (1H) and 85% H3PO4 (31P),
with downfield shifts positive.
Received for review February 21, 2004.
OP049958V
404
•
Vol. 8, No. 3, 2004 / Organic Process Research & Development