Facile and practical synthesis of 2,6-dichloropurine

Article abstract of DOI:10.1021/op049878r

A facile and industrially viable process for preparation of 2,6-dichloropurine is reported. The process involves direct chlorination of xanthine with phosphorus oxychloride and a weak nucleophilic organic base, such as amidine, guanidine base, or Proton-Sponge.

Full text of DOI:10.1021/op049878r

Organic Process Research & Development 2004, 8, 962−963
Facile and Practical Synthesis of 2,6-Dichloropurine
Qi Zeng,* Bangzhou Huang, Knut Danielsen, Rajesh Shukla, and Thomas Nagy
Department of Process Research and DeVelopment, Borregaard Synthesis, Inc. Newburyport,
Massachusetts 01950, U.S.A.
Abstract:
A facile and industrially viable process for preparation of 2,6-
dichloropurine is reported. The process involves direct chlori-
nation of xanthine with phosphorus oxychloride and a weak
nucleophilic organic base, such as amidine, guanidine base, or
Proton-Sponge.
Figure 1. Reaction of xanthine with POCl3.
Introduction
2,6-Dichloropurine is an important pharmaceutical inter-
mediate.1 It has been widely used in the preparation of purine
nucleosides and purine nucleotides.2 There are mainly two
routes for the preparation of 2,6-dichloropurine. The first
route is by chlorinating the purine ring structure, for example
chlorinating xanthine (2,6-dihydroxypurine) with pyrophos-
phoryl chloride in a sealed tube at high temperature3 or with
phosphorus oxychloride at reflux in the presence of a phase
transfer catalyst,4 chlorinating 6-chloropurine, hypoxanthine
or its N-oxide with phosphorus oxychloride,5 chlorinating
2,6-dithiopurine with chlorine gas at low temperature.6 The
second route to prepare 2,6-dichloropurine is by building the
purine ring with barbituric acid derivative7 or 2,4-dichloro-
5,6-diaminopyridmidine8 as starting material. Unfortunately,
neither of these routes is very practical in terms of industrial
scale-up operations. They require either long preparation
steps or complicated preparation procedures. Most of all, they
all suffer from very low yields.
Results and Discussion
A general method for the synthesis of chloropurines is
by chlorination of the corresponding hydroxypurines with
phosphorus oxychloride usually in the presence of a tertiary
amine, such as triethylamine, N,N-dimethyl- or N,N-diethy-
laniline. 6-Chloropurine,9 6,8-dichloropurine,10 and 2,6,8-
trichloropurine11 have been prepared in this manner. How-
ever, we could not succeed in the synthesis of 2,6-
dichloropurine from xanthine by using this method. The
addition of water into phosphorus oxychloride,3 resulting in
the formation of pyrophosphoryl chloride, does convert
xanthine to 2,6-dichloropurine in a sealed tube at high
temperature, but it is not viable on an industrial scale.
To ultilize the xanthine skeleton for preparing 2,6-
dichloropurine, chlorination of xanthine with POCl3 was
examined in various solvents (N,N-dimethylformamide, di-
methyl sulfoxide, or N-methylpyridinone), but none of them
had promising results. When hexamethylphosphoric triamide
(HMPA) or hexamethylphosphorus triamine (HMPT) was
used, a moderate yield (ca. 20%) of 2,6-dichloropurine was
obtained. However, HMPA and HMPT are strong carcino-
genic reagents, which prevents application of this process
in industrial scale.
Since xanthine is not very soluble in most organic solvents
or even in POCl3 under reflux conditions, facilitating
dissolution of xanthine in POCl3 became our focus. We
figured that a strong base perhaps can remove the proton(s)
in xanthine or form an ion pair, thus facilitating the
dissolution and chlorination of xanthine.
We have found that xanthine can be successfully con-
verted into 2,6-dichloropurine with phosphorus oxychloride
in the presence of some amidine, guanidine bases, or Proton-
Sponge. See Table 1.
In this contribution, we present a new facile and practical
procedure for the preparation of 2,6-dichloropurine by chlor-
inating xanthine with phosphorus oxychloride (Figure 1).
(1) For example, the first synthesis of Acyclovir was achieved using 2,6-
dichloropurine as starting material: (a) Schaeffer, H. J. U.S. Patents 4,-
199,574, 1976; 4,287,188, 1978; 4,294,831, 1978; 4,323,573, 1978;
4,360,522, 1978. (b) Schaefer, H. J.; Beauchamp, L.; de Miranda, P.; Elion,
G. B.; Bauer, D. J.; Collins, P. Nature 1978, 272, 583.
(2) SciFinder database: for example, (a) Verdine, G. L.; Li, D. WO Pat.
9839334, 1998. (b) Nair, V.; Pal, S. WO Pat. 9817781, 1998
(3) (a) Elion, G. B.; Hitchings, G. H. J. Am. Chem. Soc. 1956, 78, 3508. (b)
Saxena, N. K.; Gupta, P. K.; Bhakuni, D. S. Indian J. Chem. 1980, 19B,
332.
(4) Hayashi, T.; Kumasawa, Y.; Nishikawa, J. (Sumika Fine Chemicals Co.,
Ltd.). Jpn. Pat. No, 2002088082, 2002.
(5) (a) Kawashima, H.; Kumashiro, I. (Ajinomoto Co., Ltd.). Jpn. Pat.
45011508, 1970. (b) Kawashima, H.; Kumashiro, I.; Takenishi, T. U.S.
Patent 3,314,938, 1967. (c) Kawashima, H.; Kumashiro, I. Bull. Chem.
Soc. Jpn. 1967, 40, 639.
Table 2 shows the effect of different molar ratios of DBU
to xanthine on the yield of 2.6-dichloropurine (POCl3 was
used both as reagent and as solvent). It was observed that,
(6) (a) Singh, P. K.; Saluja, S.; Pratap, R.; George, C. X.; Bhakuni, D. D.
Indian J. Chem. 1986, 25B, 823. (b) Beaman, A. G.; Robins, R. K. J. Appl.
Chem. 1962, 12, 432.
(7) (a) Montgomery, J. A.; Holum, L. B. J. Am. Chem. Soc. 1958, 80, 404. (b)
Robins, R. K.; Dille, K. L.; Christensen, B. E. J. Org. Chem. 1954, 19,
930.
(9) Bendich, A.; Russell, P. J.; Fox, J. J. J. Am. Chem. Soc. 1954, 76, 6073.
(10) Robins, R. K. J. Am. Chem. Soc. 1958, 80, 6671.
(11) Robins, R. K.; Christensen, B. E. J. Am. Chem. Soc. 1952, 74, 3624.
(8) (a) Montgomery, J. A. J. Am. Chem. Soc. 1956, 56, 1928. (b) Legraverend,
M.; Boumchita, H.; Bisagni, E. Synthesis 1990, 587.
962
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development
10.1021/op049878r CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/16/2004

Figure 2. Mechanism for the formation of 2,6-dichloropurine from xanthine.
Table 1. Base-catalyzed chlorination of xanthine
preparation process is more economical (single-step reaction
and relatively higher yield) compared to the literature
methods mentioned above. This phosphorus oxychloride base
catalysis method may be applicable to chlorination of some
heterocyclic compounds which bear hydroxyl groups and are
hard to dissolve in organic solvents or in phosphorus
oxychloride. General applicability of this procedure to
various heterocyclic compounds is under investigation.
entry
basea
yield (%)
1
2
3
4
5
DBU
DBN
47
40
30
20
0
Proton-Sponge
TMG
DABCO
a Note: DBU ) 1,8-diazabicyclo[5,4,0] undec-7-ene; DBN ) 1,5-diaza-
bicyclo[4,3,0]non-5-ene; Proton-Sponge ) 1,8-bis(dimethylamino)-naphthalene;
TMG ) 1,1,3,3-tetramethylguanidine; DABCO ) 1,4-diaza-bicyclo[2,2,2]octane.
Experimental Section
General Procedures. All melting points are uncorrected.
NMR spectra were run at 400 MHz. All reactions were
conducted under nitrogen.
Table 2. Effect of molar ratio of xanthine/POCl3/DBU on
the yield of 2,6-dichloropurinea
molar ratio of
yield (%) of
2,6-dichloropurine
2,6-Dichloropurine. Xanthine (6.08 g, 0.04 mol) and
POCl3 (61.5 g, 0.40 mol) were mixed at room temperauture
and then slowly heated to 50 °C under nitrogen. DBU (36.5
g, 0.24 mol) was added dropwise under vigorous stirring.
The mixture was heated to reflux (ca. 105 °C) for 6 h (around
100 min xanthine dissolved, and the reaction mixture formed
a brown solution). The reaction mixture was cooled to 40
°C and slowly transferred to ice-water (300 g) under
vigorous stirring. The brown solution obtained was neutral-
ized to pH ) 4 with 50% aqueous NaOH (80 mL) and then
filtered through a pad of Celite. The yellow aqueous solution
was extracted with ethyl acetate (2 × 150 mL). The organic
extracts were combined and concentrated under vacuum on
a rotavapor. A deep-yellow residue (6.0 g) was obtained
which was dissolved in 100 mL of methanol at reflux,
decolorized with 0.15 g of charcoal, filtered through a pad
of Celite, and cooled to 5 °C for recrystallization. The off-
white crystals were filtered and washed with 10 mL of
methanol and then dried in a vacuum oven at 70 °C
overnight. The obtained product weighed 3.56 g (47%). The
product was characterized by comparison (HPLC, NMR, MS,
etc.) with an authentic sample of 2,6-dichloropurine.
entry
xanthine/POCl3/DBU
1
2
3
4
1/10/1.5
1/10/3.0
1/10/6.0
1/10/8.0
8b
36c
47
52
a Reaction conditions: temperature, 105 °C (reflux); time, 6 h. b Reaction
time was prolonged to 14 h. c 2,6-Dichloropurine obtained was yellow in color.
when the ratio of xanthine to DBU is 1 to 1.5, the reaction
time was prolonged to 14 h, and the reaction still did not go
to completion. Most of xanthine remained intact as original
fine particles. When the ratio of xanthine to DBU is 1 to 3,
the reaction took 8 h to complete. The crude product was
brown powder. After recrystallization, the product was
yellow in color. Although the yield increased from 47% to
52% when the ratio of xanthine to DBU was 1 to 8, the
reaction mixture was very viscous and difficult for workup.
The condition is also not as cost-effective as when the ratio
of xanthine to DBU was 1 to 6.
The proposed mechanism for the first chlorination step
is the following (Figure 2): With the proton being removed
by the base, xanthine became anionic, dissolved in POCl3,
and then formed a phosphonate intermediate. It facilitated
the chloride to attack on the ring.
Acknowledgment
The second chlorination may occur with the assistance
of the base, which is similar to the above procedure, or
without the help of the base, being directly chlorinated by
POCl3, since 2-chloro-6-hydroxypurine is much more soluble
than xanthine in POCl3.
We thank the management of Borregaard Synthesis, Inc.
for permission to publish this work. Cooperation extended
by all the colleagues of the Department of Process Research
& Development and the Department of Quality Control is
highly appreciated.
Conclusions
Received for review June 18, 2004.
OP049878R
With the utilization of the xanthine skeleton and inex-
pensive phosphorus oxychloride, this 2,6-dichloropurine
Vol. 8, No. 6, 2004 / Organic Process Research & Development
•
963