Phosphorylation of heptafluorobutanol in the presence of metal chloride - Ether catalytic systems

Article abstract of DOI:10.1023/A:1012701510769

Catalytic activity of complex catalysts based on Li, Na, K, and Cs chlorides and polydentate ligands, viz., dibenzo-18-crown-6, mono-, di-, and tetraglymes, and poly(ethylene glycols) PEG-600 and PEG-1000 was examined in the phosphorylation of heptafluorobutanol with phosphorus oxychloride. The introduction of an organic ligand into a catalytic system leads to improvement in solubility of the salt used as a catalyst and to an increase in the reaction rate by a factor of 1.3-2.8. The catalytic systems based on LiCl and poly(ethylene glycols) proved to be most efficient.

Full text of DOI:10.1023/A:1012701510769

Russian Chemical Bulletin, International Edition, Vol. 50, No. 8, pp. 1457—1460, August, 2001
1457
Phosphorylation of heptafluorobutanol in the presence
of metal chloride—ether catalytic systems
I. Yu. Kudryavtsev« and L. S. Zakharov
Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences,
ul. Vavilova 28, 119991 Moscow, Russian Federation.
Fax: +7 (095) 135 5085. E-mail: zaq@ineos.ac.ru
Catalytic activity of complex catalysts based on Li, Na, K, and Cs chlorides and
polydentate ligands, viz., dibenzo-18-crown-6, mono-, di-, and tetraglymes, and poly(ethylene
glycols) PEG-600 and PEG-1000 was examined in the phosphorylation of heptafluorobutanol
with phosphorus oxychloride. The introduction of an organic ligand into a catalytic system
leads to improvement in solubility of the salt used as a catalyst and to an increase in the
reaction rate by a factor of 1.3—2.8. The catalytic systems based on LiCl and poly(ethylene
glycols) proved to be most efficient.
Key words: catalytic phosphorylation, polyfluoroalkanols, homogeneous electrophilic
catalysis, complex catalysts, Lewis acids, polydentate ligands, ethers.
The simplest and most convenient procedure for the
synthesis of polyfluoroalkyl esters of oxygen-containing
pentavalent phosphorus acids involves catalytic phos-
phorylation of polyfluoroalkanols with the correspond-
ing phosphorus acid chlorides. This method was used
for the preparation of various esters and chlorides of
phosphoric,1—3 phosphonic,3—5 and phosphinic6 acids
based on primary,1—6 secondary,7—12 and tertiary9,13
polyfluoroalkanols. Recently,14—16 we have demonstrated
that the catalytic activity of metal salts in these reac-
tions depends substantially on the acceptor properties of
the electrophilic catalysts used. Thus, weak Lewis acids
exhibit higher catalytic activity than strong acids. At the
same time, the weakest Lewis acids, viz., chlorides of
Group I metals, are less readily soluble in organic
reaction media3 because they occur as ionic salts char-
acterized by high lattice energies, the energies of their
solvation in these media being low.
and HCl) and react with metal chloride to form com-
plexes sufficiently stable for their solubility to be in-
creased. At the same time, molecules of phosphorus
acid chloride in the coordination sphere of the metal
atom should not be completely replaced by the ligand
molecules; otherwise it would lead to inhibition of the
catalytic reaction. We decided to use ethers. Although
there is evidence17—19 that the C—O—C bond in ethers
is cleaved under the action of Lewis acids (BCl3, AlCl3,
SnCl4, or ZnCl2) as well as under the action of hydro-
gen chloride or POCl3 in the presence of catalytic
amounts of Lewis acids (ZnCl2 or HgCl2), we believed
that the electron-withdrawing properties of chlorides of
Group I metals are too weak to cause cleavage of the
ether bond. However, this question called for experi-
mental verification. To ensure the formation of rather
stable complexes and taking into account that the oxy-
gen atom in ethers possesses rather weak donor proper-
ties, we chose ligands containing several donor atoms
capable of forming chelate complexes, viz., dibenzo-18-
crown-6 (DB18C6), mono-, di-, and tetraglymes, and
poly(ethylene glycols) PEG-600 and PEG-1000.
The aim of the present study was to examine the
catalytic activity of complex catalysts in the phosphory-
lation of heptafluorobutanol with phosphorus oxychloride.
We examined catalytic systems of this type based on
LiCl, NaCl, KCl, and CsCl in the model reaction of
2,2,3,3,4,4,4-heptafluorobutanol (1) with POCl3.
Results and Discussion
To increase the solubility of catalysts utilized in the
phosphorylation of polyfluoroalkanols with phosphorus
acid chlorides, we used catalytic systems based on com-
plexes of metal chlorides with organic ligands. One
would expect that the addition of organic ligands will
lead to an increase in solubility of the salt used as the
catalyst due to improvement in conditions of cation
solvation. When choosing ligands, we proceeded from
the requirement that they must be stable under condi-
tions of catalytic phosphorylation (high temperature,
the presence of phosphorus acid chloride, Lewis acid,
Catalyst, 160 °C
–HCl
3 C3F7CH2OH + POCl3
ꢀ
(C3F7CH2O)3PO
Metal chloride and the ligand were taken in a molar
ratio such that there were six donor oxygen atoms
per metal atom (five oxygen atoms in the case of
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 8, pp. 1386—1388, August, 2001.
1066-5285/01/5008-1457 $25.00 ©ꢀ2001 Plenum Publishing Corporation

1458
Russ.Chem.Bull., Int.Ed., Vol. 50, No. 8, August, 2001
Kudryavtsev and Zakharov
tetra(ethylene glycol) dimethyl ether). The activity of
the catalysts was estimated from the time it took for the
reaction to be completed, which was judged visually
from cessation of evolution of hydrogen chloride. The
effect of the addition of the ligand on the efficiency of
the catalyst was evaluated by comparing the duration of
the reaction in the presence of the initial metal chloride
(control) with that in the presence of a catalytic system
containing the organic ligand. The reproducibilities of
the reaction time and the yield of the product were ±5%
and ±0.5%, respectively.
We examined the catalytic systems involving chlo-
rides of Group I metals and dibenzo-18-crown-6. It
appeared that the addition of this ligand to the reaction
system led to acceleration of the reaction by a factor of
1.3—2.4 compared to the reactions proceeding in the
presence of the corresponding pure metal chloride. Ap-
parently, this is due to the gain in solubility of the
catalyst and an increase in its effective concentration in
the solution (Table 1). Thus, unlike LiCl, which was
only partially soluble in the reaction mixture, the
LiCl + DB18C6 catalytic system formed a homoge-
neous solution already at the initial stage of the reac-
tion. Although the catalyst gradually precipitated, the
overall reaction time decreased by a factor of 1.3.
The reaction was even more accelerated (1.5 times)
in the presence of the NaCl + DB18C6 system in spite
of the fact that the catalyst was dissolved incompletely.
In this series of catalysts, the maximum acceleration of
the reaction (2.4 times) was observed in the case of the
KCl + DB18C6 system, i.e., in the case of the cation
whose diameter (2.66 Å) corresponds most closely to
the size of the cavity of this crown ether (2.90 Å).20 This
catalytic system, like the lithium salt, was completely
soluble only at the initial stage of the reaction. A further
increase in the size of the cation (the catalytic system
based on CsCl; the diameter of the cation was 3.34 Å)20
led to an insignificant decrease in the coefficient of
acceleration (to 2.3), the solubility of the catalyst being
also somewhat improved compared to that of pure CsCl.
It should be noted that all catalytic systems in
question are sufficiently stable under the reaction con-
ditions, which made it possible to obtain phosphate 2 in
good yield.
We also examined the possibility of the use of cata-
lytic systems involving acyclic polydentate ethers with a
flexible chain, which are capable of adapting themselves
to the coordination polyhedron of the metal atom, and
studied the effect of the structure of these ligands on the
efficiency of catalysis. Taking into account that the use
of LiCl in the systems with dibenzo-18-crown-6 pro-
vided the maximum rate of phosphorylation, we carried
out comparison using catalytic systems based on this salt.
All linear ligands under study also improved the
solubility of LiCl in the reaction mixture and the cata-
lytic systems involving these ligands ensure acceleration
of the reactions compared to those performed in the
presence of pure LiCl (the initial stage of the reac-
tion always proceeded under homogeneous con-
ditions). In the LiCl + 3MeOCH2CH2OMe and
LiCl + 2MeO(CH2CH2O)2Me catalytic systems, there
are six donor oxygen atoms per metal atom. However,
the reaction in the presence of the catalytic system
based on monoglyme proceeded somewhat more slowly
(was accelerated by a factor of 1.5—1.6) than in the
presence of the system based on diglyme (was acceler-
ated by a factor of 1.9). It should be noted that phos-
phate 2 was obtained in equal yields in both cases. The
observed deceleration of the reaction is, apparently,
associated with the lower boiling point of monoglyme
Table 1. Catalytic phosphorylation of heptafluorobutanol with phosphorus oxy-
chloride
Catalyst
Reaction
time/h
Yield of
phosphate
2 (%)
Solubility of
the catalyst*
LiCl
LiCl + DB18C6
NaCl
NaCl + DB18C6
KCl
KCl + DB18C6
CsCl
CsCl + DB18C6
LiCl + 3MeOCH2CH2OMe
LiCl + 2MeO(CH2CH2O)2Me
LiCl + MeO(CH2CH2O)4Me
LiCl + PEG-600
LiCl + PEG-1000
3.6
2.7
93.8
86.9
91—92
86.4
91—92
83.8
85.6
81.4
91.3
91.3
92.5**
82.0
p
c → p
i
p
i
c → p
c → i
c → p
c → p
c → p
c → p
c
19—20
13.0
32—34
13.6
15.0
6.5
2.3
1.9
2.7
1.3
1.3
83.8
c
* The visual solubility of the catalysts: c, completely soluble; p, partially soluble;
i, insoluble to a noticeable extent. The direction of the change of the catalyst
solubility in the course of the reaction is indicated by an arrow.
** The product contained a tetraglyme impurity.

Catalytic phosphorylation
Russ.Chem.Bull., Int.Ed., Vol. 50, No. 8, August, 2001
1459
compared to that of diglyme resulting in a decrease in
the temperature of the boiling reaction mixture.
Apparently, a decrease in the number of the do-
nor oxygen atoms per metal atom to five in the
LiCl + MeO(CH2CH2O)4Me catalytic system is re-
sponsible for a slight increase in the reaction time due
to a decrease in the solubility of the catalyst compared
to the catalytic systems based on other oligo(ethylene
glycol) ethers. Moreover, the essential drawback of
tetraglyme as a component of a catalytic system is that
its boiling point is close to the boiling point of the target
phosphate 2 resulting in reduction in the purity of the
product due to which additional purification is required.
It should be noted that all ethers of ethylene glycols
used also proved to be stable under conditions of cata-
lytic phosphorylation as evidenced, in particular, by the
high yields of phosphate 2.
Homogeneous catalytic systems providing the maxi-
mum reaction rate were prepared with the use of
poly(ethylene glycols) with the molecular weights of 600
and 1000 (PEG-600 and PEG-1000) as ligands. The
amounts of the ligands in these systems corresponded to
the following ratio: six O atoms of ether per Li atom.
The use of these catalysts led to a decrease in the
reaction time by a factor of 2.8 compared to the reac-
tion in the presence of pure LiCl. A substantial decrease
in the yield of the phosphate in the case of the catalytic
systems based on poly(ethylene glycol) is attrubutable to
consumption of POCl3 in chlorination of the terminal
hydroxyl groups of the ligands.
The experimental data demonstrated that the solu-
bility of the above-mentioned catalysts decreased as the
content of polyfluoroalkanol in the reaction mixture
decreased in the course of the reaction, which indicates
that heptafluorobutanol 1 solvates salts of the catalyst
more efficiently than the final phosphate 2 and less
basic intermediate phosphorochloridates and POCl3.
Unlike phosphoryl ligands, which solvate only cations,
polyfluoroalkanol can solvate both cations (coordina-
tion at the oxygen atom of the hydroxyl group) and the
chloride ion (hydrogen bonding). Acceleration of the
reaction due to an increase in the solubility of salts used
as the catalyst in the presence of small additives of
polydentate ethers indicates that these ethers rather
efficiently solvate chlorides of Group I metals and are
superior to POCl3 and phosphorylation products in this
respect.
The inner coordination spheres of these complexes
involve the starting phosphorus acid chloride, polydentate
ether, and, probably, polyfluoroalkanol. Coordination
of acid chloride to the metal cation through the oxygen
atom of the phosphoryl group (see Ref. 16 and refer-
ences cited therein) ensures activation of this ligand due
to an increase in the effective positive charge on the
phosphorus atom and promotion of the nucleo-
philic substitution of the chlorine atom at the phospho-
rus atom.
Due to the low dielectric constant of the medium
and rather weak donor properties of phosphoryl com-
pounds and ethers present in solutions, the counterion
(chloride) is, apparently, also involved in the inner
coordination sphere of the metal atom. The formation
of an active complex containing the chlorine atom in
the inner coordination sphere of the metal atom (a
contact ion pair) is also indicated by the high catalytic
activity of the LiCl + polydentate ether systems compa-
rable with that of the best homogeneous catalysts based
on lithium salts.14,15 This is associated with the fact that
the displacement of the anion from the inner coordina-
tion sphere of the catalytic complex giving rise to
solvent-separated ion pairs (for example, in the case of
LiClO4) leads to a sharp decrease in the catalytic ac-
tivity.14
The next stage involves the nucleophilic substitution
of the chlorine atom at the phosphorus atom in coordi-
nated phosphoryl halide:
+ROH
(>POCl)•MCl•L
(>PO2R)•MCl•L
,
n
n
–HCl
where R is polyfluoroalkyl.
Then the active complex is regenerated through the
displacement of the phosphoryl ligand in the coordina-
tion sphere of metal by the molecule of the unreacted
initial acid chloride
+>POCl
(>PO2R)•MCl•L
(>POCl)•MCl•L ,
n
n
–>PO2R
and the catalytic cycle is closed.
The resulting phosphate 2 can be readily isolated
from the reaction mixture by vacuum distillation. The
yields of 2 in the reactions with the use of some
catalytic systems, though somewhat lower than those
obtained in the reactions involving pure metal chloride,
exceed 90% and approach the maximum yield ob-
tained in the reaction with the use of pure LiCl as the
catalyst.
On the whole, the results of the present study pro-
vide evidence that the approach proposed by us has
promise in the design of efficient homogeneous catalytic
systems for the phosphorylation of polyfluoroalkanols
with phosphorus acid chlorides based on complexes of
weak Lewis acids with organic ligands of the polyether
type. The approach allows one to substantially acceler-
ate the reaction with retention of a good yield of
polyfluoroalkyl phosphate.
The probable reaction mechanism of catalytic phos-
phorylation can be represented as follows. At the first
stage, dissolution of the catalyst in the reaction mixture
gives rise to mixed complexes with an ether ligand and
phosphoryl halide involved in the inner coordination
sphere of the metal atom
MCl + nL + >POCl
(>POCl)•MCl•L
,
n
where M is the Group I cation and L is alcohol, ether,
or a phosphoryl ligand, which does not undergo chemi-
cal conversions in this catalytic cycle.

1460
Russ.Chem.Bull., Int.Ed., Vol. 50, No. 8, August, 2001
Kudryavtsev and Zakharov
6. L. S. Zakharov, S. V. Dobrokhotova, V. V. Pisarenko,
N. N. Godovikov, and M. I. Kabachnik, Izv. Akad. Nauk
SSSR, Ser. Khim., 1975, 478 [Bull. Acad. Sci. USSR, Div.
Chem. Sci., 1975, 24 (Engl. Transl.)].
7. L. S. Zakharov, E. I. Goryunov, S. T. Ioffe, L. L. Morozov,
T. M. Shcherbina, and M. I. Kabachnik, Izv. Akad. Nauk
SSSR, Ser. Khim., 1976, 1834 [Bull. Acad. Sci. USSR, Div.
Chem. Sci., 1976, 25 (Engl. Transl.)].
8. M. I. Kabachnik, L. S. Zakharov, P. V. Petrovskii, and E. I.
Goryunov, Izv. Akad. Nauk SSSR, Ser. Khim., 1987, 2879
[Bull. Acad. Sci. USSR, Div. Chem. Sci., 1987, 36, 2679
(Engl. Transl.)].
9. M. I. Kabachnik, L. S. Zakharov, E. I. Goryunov, I. Yu.
Kudryavtsev, G. N. Molchanova, M. A. Kurykin, P. V.
Petrovskii, T. M. Shcherbina, and A. P. Laretina, Zh. Org.
Khim., 1994, 64, 902 [Russ. J. Org. Chem., 1994, 64 (Engl.
Transl.)].
10. M. I. Kabachnik, L. S. Zakharov, I. Yu. Kudryavtsev, P. V.
Petrovskii, T. M. Shcherbina, and A. P. Laretina, Izv. Akad.
Nauk, Ser. Khim., 1994, 1840 [Russ. Chem. Bull., 1994, 43,
1743 (Engl. Transl.)].
11. M. I. Kabachnik, L. S. Zakharov, E. I. Goryunov, Yu. V.
Zeifman, P. V. Petrovskii, T. M. Shcherbina, and A. P.
Laretina, Dokl. Akad. Nauk, 1995, 341, 346 [Dokl. Chem.,
1995 (Engl. Transl.)].
12. E. I. Goryunov, G. N. Molchanova, P. V. Petrovskii, L. S.
Zakharov, and M. I. Kabachnik, Izv. Akad. Nauk, Ser.
Khim., 1998, 1776 [Russ. Chem. Bull., 1998, 47, 1728 (Engl.
Transl.)].
13. M. I. Kabachnik, L. S. Zakharov, E. I. Goryunov, V. S.
Shaidurov, A. V. Kashkin, Yu. L. Bakhmutov, and V. F.
Zabolotskikh, Izv. Akad. Nauk SSSR, Ser. Khim., 1985, 957
[Bull. Acad. Sci. USSR, Div. Chem. Sci., 1985, 34, 875
(Engl. Transl.)].
Experimental
We used the following compounds of reagent grade: POCl3
purified by double distillation (b.p. 107 °C), 2,2,3,3,4,4,4-hepta-
fluorobutanol (1) distilled successively over concentrated H2SO4
and P2O5 (b.p. 96 °C), and dimethyl ethers of mono-, di-, and
tetra(ethylene glycols) purified by distillation over metallic
sodium. Their physicochemical properties are identical with the
published data. Dibenzo-18-crown-6 of chemically pure grade
was used without additional purification. Poly(ethylene glycols)
PEG-600 and PEG-1000 (Loba Chemie) were dried before use
in a desiccator over concentrated H2SO4 at ∼20 °C.
Catalytic phosphorylation of heptafluorobutanol (1) with
POCl3 (general procedure). A mixture of heptafluorobutanol
(1) (17.0 g, 0.085 mol), POCl3 (3.8 g, 0.025 mol), metal
chloride (2 mmol), and the corresponding amount of an or-
ganic ligand was refluxed on a bath at 160 °C using a reflux
condenser equipped with a bubble counter (filled with concen-
trated H2SO4) until evolution of HCl ceased. Excess polyfluoro-
alkanol was removed in vacuo using a water-aspirator pump at
the bath temperature <100 °C. The residue was distilled in
vacuo and phosphate 2 was obtained, b.p. 78—80 °C (0.5 Torr),
20
nD
1.3105 (cf. lit. data1). The catalysts, the metal chlo-
ride : polyether ratios, the reaction duration, and the yields of
phosphate 2 are given in Table 1.
This work was financially supported by the Russian
Foundation for Basic Research (Project Nos. 96-15-
97298 and 00-15-97386).
References
14. I. Yu. Kudryavtsev, L. S. Zakharov, and M. I. Kabachnik,
Zh. Prikl. Khim., 1995, 68, 1506 [Russ. J. Appl. Chem.,
1995, 68 (Engl. Transl.)].
15. M. I. Kabachnik, L. S. Zakharov, I. Yu. Kudryavtsev, and
T. M. Shcherbina, Zh. Prikl. Khim., 1995, 68, 2003 [Russ. J.
Appl. Chem., 1995, 68 (Engl. Transl.)].
16. I. Yu. Kudryavtsev, L. S. Zakharov, and M. I. Kabachnik,
Izv. Akad. Nauk, Ser. Khim., 1998, 2070 [Russ. Chem. Bull.,
1998, 47, 2015 (Engl. Transl.)].
17. R. L. Burwell, Chem. Rev., 1954, 54, 615.
18. S. Fried and R. D. Kleene, J. Am. Chem. Soc., 1941,
63, 2691.
19. P. Peringer and P. P. Winkler, Inorg. Chim. Acta,
1986, 118, L1.
20. Host Guest Complex Chemistry: Macrocycles, Synthesis, Struc-
tures, Applications, Eds. F. Vögtle and E. Weber, Springer,
Berlin—Heidelberg—New York—Tokyo, 1985.
1. L. S. Zakharov, V. V. Pisarenko, N. N. Godovikov, and
M. I. Kabachnik, Izv. Akad. Nauk SSSR, Ser. Khim., 1971,
2503 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1971, 20,
2373 (Engl. Transl.)].
2. L. S. Zakharov, V. V. Pisarenko, N. N. Godovikov, and
M. I. Kabachnik, Izv. Akad. Nauk SSSR, Ser. Khim., 1971,
2671 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1971, 20,
2536 (Engl. Transl.)].
3. I. Yu. Kudryavtsev, L. S. Zakharov, and M. I. Kabachnik,
Izv. Akad. Nauk SSSR, Ser. Khim., 1982, 2535 [Bull. Acad.
Sci. USSR, Div. Chem. Sci., 1982, 31 (Engl. Transl.)].
4. M. I. Kabachnik, N. N. Godovikov, V. V. Pisarenko, and
L. S. Zakharov, Izv. Akad. Nauk SSSR, Ser. Khim., 1972,
1667 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1972, 21,
1617 (Engl. Transl.)].
5. M. I. Kabachnik, L. S. Zakharov, E. I. Goryunov, A. P.
Kharchenko, V. F. Zabolotskikh, G. D. Novikova, and
V. A. Svoren´, Izv. Akad. Nauk SSSR, Ser. Khim., 1979,
656 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1979, 28 (Engl.
Transl.)].
Received November 20, 2000;
in revised form March 23, 2001