Practical synthesis of 5-Fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine, a key intermediate in the preparation of potent deoxycytidine kinase inhibitors

Article abstract of DOI:10.1021/op900060u

A practical synthesis of 5-fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine, a key intermediate in the preparation of a new class of potent deoxycytidine kinase (dCK) inhibitors, is described. The commercially available 2,4-dichloro-5-fluoropyrimidine (12) i

Full text of DOI:10.1021/op900060u

Organic Process Research & Development 2009, 13, 807–811
Practical Synthesis of 5-Fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine, a Key
Intermediate in the Preparation of Potent Deoxycytidine Kinase Inhibitors
Haiming Zhang, Jie Yan,† Ramanaiah C. Kanamarlapudi,‡ Wenxue Wu,* and Philip Keyes
Chemical DeVelopment, Lexicon Pharmaceuticals, 350 Carter Road, Princeton, New Jersey 08540, U.S.A.
Scheme 1. Preparation of dCK inhibitors from 1
Abstract:
A practical synthesis of 5-fluoro-2-(piperidin-4-yloxy)pyrimidin-
4-amine, a key intermediate in the preparation of a new class of
potent deoxycytidine kinase (dCK) inhibitors, is described. The
commercially available 2,4-dichloro-5-fluoropyrimidine (12) is
converted in four telescoped steps to tert-butyl 4-(4-amino-5-
fluoropyrimidin-2-yloxy)piperidine-1-carboxylate (6a) which upon
deprotection gives 5-fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine
dihydrochloride (1a) in about 68% overall yield. This process
proved to be an economical alternative to a Mitsunobu-based
synthesis.
Scheme 2. Medicinal chemistry synthesis of 1a
Introduction
Deoxycytidine kinase (dCK) is an enzyme that catalyzes
phosphorylation of pyrimidine and purine deoxynucleosides,
one of the initial biochemical steps in the deoxynucleoside
salvage pathway that supplies precursors for DNA synthesis.1
Human dCK is known to be involved in the activation of several
chemotherapeutically important nucleoside analogues.2 Through
phenotypical analysis of knockout (KO) mice, Lexicon identi-
fied dCK as a potential drug target in multiple therapeutic areas
within cancer, immunological disorders, and infectious diseases.3
Internal medicinal chemistry research led to a class of O-linked
pyrimidin-4-amine-based compounds as potent inhibitors of
dCK.4 We report herein a scalable synthesis of a key intermedi-
ate for the preparation of this class of compounds, 5-fluoro-2-
(piperidin-4-yloxy)pyrimidin-4-amine (1, Scheme 1).
Results and Discussion
The medicinal chemistry group used a synthesis based on
the Mitsunobu reaction5 of commercially available 5-fluorocy-
tosine (2)6 and N-Boc-4-piperidinol (3) (Scheme 2).4 The
resulting iminophosphorane7 4 is then treated with HCl to give
the 5-fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine dihydro-
chloride (1a) in about 60% yield. While this synthesis worked
well on small scale and provided rapid access to gram quantities
of 1 for early investigations of structure activity relationships
(SAR), it gave inconsistent yields on scale-up. As a process, it
also suffers several major shortcomings in addition to the usual
issues associated with Mitsunobu reactions. Up to 2 equiv of 2
are required to fully convert the piperidinol 3 at a reasonable
reaction rate. The low solubility of 2 as well as that of 4
necessitates a very large volume of reaction solvent (100 mL
THF/g of 3). Additionally, the unreacted 2 remaining in the
reaction mixture poses significant purification challenges. As a
result, the iminophosphorane 4 had to be purified by chroma-
tography to remove large amounts of 2 and side products prior
to deprotection in order to produce 1a of acceptable purity.
Although a vendor was able to use this chemistry to produce
500 g of 1a in about 40% overall yield to support lead
* To whom correspondence should be addressed. E-mail: wwu@
lexpharma.com.
† Current address: Amgen, Inc., One Kendall Square, Building 1000,
Cambridge, MA 02139. E-mail: yanj@amgen.com.
‡ Current address: Concert Pharmaceuticals, Inc., 99 Hayden Ave, Suite 500,
Lexington, MA 02421. E-mail: rkanamarlapudi@concertpharma.com.
(1) (a) Arner, E. S. J.; Eriksson, S. Pharmacol. Ther. 1995, 67, 155–186,
and references cited therein. (b) Chottiner, E. G.; Shewach, D. S.;
Datta, N. S.; Ashcraft, E.; Gribbin, D.; Ginsburg, D.; Fox, I. H.;
Mitchell, B. S. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 1531–1535.
(2) (a) Van Rompay, A. R.; Johansson, M.; Karlsson, A. Pharmacol. Ther.
2003, 100, 119–139. (b) Sabini, E.; Ort, S.; Monnerjahn, C.; Konrad,
M.; Lavie, A. Nat. Struct. Mol. Biol. 2003, 10, 513–519. (c) Sabini,
E.; Hazra, S.; Konrad, M.; Burley, S. K.; Lavie, A. Nucleic Acids
Res. 2007, 35, 186–92. (d) Lee, H.; Hanes, J.; Johnson, K. A.
Biochemistry 2003, 42, 14711–14719.
(5) For a review of the Mitsunobu reaction, see: Hughes, D. L. Org. React.
1992, 42, 335–656.
(3) Some deoxycytidine kinase inhibitors have been reported. For
examples, see: (a) Krenitsky, T. A.; Tuttle, J. V.; Koszalka, G. W.;
Chen, I. S.; Beacham, L. M., III.; Rideout, J. L.; Elion, G. B. J. Biol.
Chem. 1976, 251, 4055–4061. (b) Ward, A. D.; Baker, B. R. J. Med.
Chem. 1977, 20, 88–92.
(6) (a) 5-Fluorocytosine is an antifungal with the generic drug name of
flucytosine. For a recent review on this prodrug, see: (b) Vermes, A.;
Guchelaar, H.-J.; Dankert, J. J. Antimicrob. Chemother. 2000, 46, 171–
179.
(4) Augeri, D. J.; Carlsen, M.; Carson, K. G.; Fu, Q.; Healy, J. P.; Heim-
Riether, A.; Jessop, T. C.; Keyes, P. E.; Shen, M.; Tarver, J. E.; Taylor,
J. A.; Xu, X. U.S. Pat. Appl. Publ 20080146571 A1, 2008.
(7) For recent reviews on iminophosphorane, see: (a) Wamhoff, H.;
Richardt, G.; Stoelben, S. AdV. Heterocl. Chem. 1995, 64, 159–249.
(b) Molina, P.; Vilaplana, M. J. Synthesis 1994, 1197, 218.
10.1021/op900060u CCC: $40.75  2009 American Chemical Society
Published on Web 04/30/2009
Vol. 13, No. 4, 2009 / Organic Process Research & Development
•
807

Scheme 3. Alkylation of 5-fluorocytosine (2)
Scheme 5. Alkoxylation of protected
aminochloropyrimidines 9
Scheme 4. Alkoxylation of aminochloropyrimidine 9a
The desired alkoxylation product 6a was isolated in about 58%
yield after column chromatography.12 Various other conditions
were screened without significant improvement.13
We reasoned that the reactivity of 9a is likely attenuated by
partial deprotonation of the amino group under the strongly basic
reaction conditions. Indeed, the aqueous pKa of 2-chloropyri-
midin-4-amine, the des-fluoro analogue of 9a, was reported to
be 16.4,14 which is comparable to that of tert-butanol.15
Moreover, both the deprotonated starting material 9a and
deprotonated product 6a can compete with the alkoxide of 3 in
reactions with 9a, leading to the amination side products 10
and 11.16 Therefore, protection of both hydrogens on the amino
group of 9a should not only prevent the amination side reactions
but also improve its reactivity as an electrophile.
A number of protected chloropyrimidines (9b-e) were
evaluated (Scheme 5). Among these, 9b and 9c, both prepared
from 9a, underwent deprotection under the reaction conditions
even at a lower temperature (50 °C), giving similar alkoxylation
results as reaction of 9a. The dibenzyl-protected chloropyrimi-
dine 9d, a crystalline compound readily available from reaction
of the inexpensive 2,4-dichloro-5-fluoropyrimidine (12) with
dibenzylamine,17 reacted with 3 in the presence of sodium tert-
butoxide at 50 °C to give the crystalline product 6d in high
yield (90%). However, complete debenzylation of 6d under
various conditions was not successful. For example, under
hydrogenation or transfer hydrogenation conditions, 6d was
converted cleanly to the des-fluoro compound 13, probably
through the ring-hydrogenated intermediate 14 followed by
dehydrofluorination (Scheme 6). It is surprising to note that the
optimization work, a significant amount of experimental effort
resulted in little improvement of this process. Clearly, a different
approach was needed to support a potentially very aggressive
development timeline for the eventual drug candidate.
Initial process development effort on using a different
alkylating agent, such as the mesylate 5, was not fruitful
(Scheme 3). Contrary to the milder Mitsunobu conditions (room
temperature), reaction of 5 required prolonged heating at higher
temperature (80 °C) and resulted in a mixture of the O- and
N-substituted products (6a and 7, respectively)3b,8 in addition
to a significant amount of the elimination product 8. Even under
conditions that favor O-alkylation (cesium carbonate as the base
and DMSO as the solvent) mesylate 5 reacted with 5-fluoro-
cytosine (2) to give about 35% yield of 6a, 14% yield of 7,
and 32% yield of 8 after column chromatography.9 A number
of other leaving groups were screened under various conditions
without significant yield improvement.10
To eliminate the selectivity problem associated with the
alkylation approach, an alternative bond connection based on
alkoxylation of commercially available 2-chloro-5-fluoropyri-
midin-4-amine (9a) was investigated (Scheme 4). However, in
addition to the desired product 6a, this reaction produces a
significant amount of amination side products 10 and 11,11
which were difficult to remove without chromatography. Under
conditions that discourage these side reactions (2 equiv of 3,
sodium tert-butoxide, diglyme, 120 °C), a mixture of 6a, 10,
and 11 were produced in about 83/15/2 ratio by HPLC area.
(11) Structure of 11 was based on LC-MS analysis and stereochemical
considerations. The isomeric structure shown below cannot be
completely ruled out.
(8) Brown, D. J. The Pyrimidines; Wiley: New York, 1962. Supplements:
1970, 1985, and 1994.
(9) (a) It should be noted that isolation of the N- and O-alkylation products
6a and 7 enabled structural confirmation of the original Mitsunobu
product by NMR techniques. (b) For automated structure verification
by NMR, see: Keyes, P.; Hernandez, G.; Cianchetta, G.; Robinson,
J.; Lefebvre, B. Magn. Reson. Chem. 2009, 47, 38–52.
(10) Leaving groups screened: tosylate, 4-chlorobenzenesulfonate, 4-ni-
trobenzenesulfonate, 2-napthalenesulfonate, triflate, and iodide; bases
screened: sodium hydride, lithium tert-butoxide, potassium carbonate,
and DBU; polar solvents screened: 1,4-dioxane, DMF, NMP, and
HMPA.
808
•
Vol. 13, No. 4, 2009 / Organic Process Research & Development

Scheme 6. Hydrogenation of 6d
substances may be due to further reactions of the allylic
anions, and therefore adding a proton source in the reaction
medium could help control the color. Although the
isomerization using lithium tert-butoxide in tert-butanol
was slow, it proceeded at a reasonable rate in a mixture
of DMSO and tert-butanol, resulting in adequate control
of the color. The optimum solvent ratio was determined
to be 1:1 DMSO/tert-butanol.
Among the intermediates, 6a is a solid, while 9e, 6e, and
15 are oils, and thus isolation is not convenient. Although 15
can be directly converted to 1 under strongly acidic conditions,
isolation and purification of 1 proved difficult due to the
presence of oily impurities and byproduct. We decided to
telescope the steps from the amination through the hydrolysis
(see Experimental Section) and effect purification on 6a.
Hydrolysis of 15 was carried out at pH 2.0-3.0 without
affecting the Boc group.19 The propionaldehyde side product
was removed by extraction with aqueous sodium bisulfite, and
residual color was further reduced with a charcoal treatment.
Compound 6a was isolated in about 75% overall yield by
crystallization from ethyl acetate/n-heptane. The purity of 6a
was typically greater than 98%.
Due to the very high water solubility of 1, Boc-
deprotection and isolation of 1a were carried out under
anhydrous conditions. Screening experiments led to iso-
propyl alcohol as the optimum solvent for the reaction
and isolation. Upon treatment of 6a with excess anhydrous
HCl in isopropyl alcohol, the product 1a was isolated as
a white to pale-yellow solid in about 90% yield directly
from the reaction mixture by a simple filtration.
This process was successfully demonstrated at 100 g
scale (Scheme 8). The overall yield of 1a was 68%, and
the purity was greater than 99% compared to 40% yield
and similar purity for the Mitsunobu process practiced at
the aforementioned contract manufacturer. In fact, the
same contract manufacturer gave a quotation for 10 kg
of 1a using this process that was more than 50% lower
than that using the Mitsunobu process.
In summary, we have developed a practical and economical
synthesis for 5-fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine
hydrochloride (1a) from inexpensive 2,4-dichloro-5-fluoropy-
rimidine (12). This synthesis avoids numerous process issues
arising from practicing the alternative chemistry based on the
Mitsunobu reaction.
Scheme 7. Isomerization of the allyl groups in 6e
benzyl groups survived these conditions. Alternative deprotec-
tion conditions were attempted without success.18
Gratifyingly, double-protection with the allyl group
gave satisfactory results. The alkoxylation product 6e was
prepared by means similar to that for 6d.17 Although
rhodium- or palladium-catalyzed deallylation is well-
known,19 limited screening experiments using transition-
metal-catalyzed deallylation did not give any promising
leads. However, isomerization under standard base-
catalyzed conditions (potassium tert-butoxide in DMSO,
room temperature) gave a mixture of bis-enamine geo-
metric isomers 15 (Scheme 7). Interestingly, a significant
amount of DMSO-substituted side product 16 (up to 30%
by HPLC area) was observed under these conditions. Upon
a limited screen of reaction conditions, it was discovered
that formation of side product 16 was readily controlled
by using lithium tert-butoxide as the base. These reaction
conditions gave an intensely dark color which was not
removed by charcoal treatment and resulted in difficult
workup and isolation of 6a. We speculated that the dark
(12) Due to the higher solubility of 9a, the volume efficiency of this
alkoxylation reaction was much better than the Mitsunobu reaction.
Additionally, purification of 6a from this reaction mixture was simpler
than that from the Mitsunobu reaction. As a result, the medicinal
chemistry group utilized this approach to prepare larger quantities of
1a (see ref 4).
(13) Bases screened: DBU, cesium carbonate, lithium tert-butoxide, potas-
sium tert-butoxide, sodium hydride, methyl magnesium bromide,
diethylzinc, and methyllithium; solvents screened: dioxane, DMF,
DMSO, triglyme; and temperatures screened: 80, 100, and 120 °C.
(14) Harris, M. G.; Stewart, R. Can. J. Chem. 1977, 55, 3800–3806.
(15) (a) Smith, M. B.; March, J.; Eds. March’s AdVanced Organic
Chemistry: Reactions, Mechanisms, and Structure, 5th ed.; Wiley: New
York, 2001; p 331. (b) Reeve, W.; Erikson, C. M.; Aluotto, P. F. Can.
J. Chem. 1979, 57, 2747–2754.
Experimental Section
HPLC Conditions. Method 1: Column: Sunfire C18, 5 µm,
4.6 mm × 50 mm. Column temperature: 40 °C. Flow: 3 mL/
min. Wavelength: 220 nm. Solvent A: 0.1% trifluoroacetic acid
(17) Only product from amination of the 4-chloro position was observed. Durr,
G. J. J. Med. Chem. 1965, 8, 253–255. See also ref 8.
(16) The self-condensation product of 9a shown below was detected by
LC-MS during the reaction.
(18) (a) Boc-deprotection was observed without debenzylation under
strongly acidic conditions such as hydrobromic acid and methane-
sulfonic acid. Product 1 was found to be unstable in the presence of
excess triflic acid. No debenzylation was observed under oxidative
conditions such as DDQ and cerium ammonium nitrate (CAN). (b)
Similarly difficult debenzylation was reported previously for a
4-dibenzylaminopyrimidine, see: Boger, D. L.; Honda, T.; Dang, Q.
J. Am. Chem. Soc. 1994, 116, 5619–5630.
(19) Greene, T. W., Wuts, P. G. M., Eds. ProtectiVe Groups in Organic
Synthesis, 3rd ed.; Wiley: New York, 1999.
Vol. 13, No. 4, 2009 / Organic Process Research & Development
•
809

Scheme 8. Final synthesis of 1a
in water. Solvent B: acetonitrile. Linear gradient: 10% B to 90%
B in 5 min. Method 2 (for analysis of final product 1a only):
Column: Luna C18(2), 5 µm, 4.6 mm × 250 mm. Flow: 0.7
mL/min. Wavelength: 220 nm. Solvent A: 10 mM ammonium
acetate in water. Solvent B: acetonitrile. Linear gradient: 10%
B to 90% B in 8 min and hold for 2 min.
Retention times: 1a, 0.25-0.32 min (method 1), 6.66 min
(method 2); 6a, 1.61 min; 6e, 2.85 min; 9e, 3.44 min; 12, 2.16
min; 15, 3.97 and 4.06 min.
(additional THF was added during the distillation to ensure
dryness of the concentrate) and diluted with 400 mL THF.
Residual solids were removed by filtration and washed with
THF (100 mL). The combined filtrates were concentrated and
diluted with DMSO (400 mL) to give a solution of compound
6e (96.3% purity by HPLC area) which is used directly in the
next step. An analytically pure sample of 6e was obtained by
column chromatography and characterized. MS: MH+ ) 393.1;
1H NMR (CDCl3) δ 7.85 (d, J ) 14.0 Hz, 1H), 5.87 (m, 2H),
5.20 (m, 4H), 4.98 (m, 1H), 4.16 (d, J ) 5.3 Hz, 4H), 3.80 (m,
2H), 3.24 (m, 2H), 1.90 (m, 2H), 1.78 (m, 2H), 1.48 (s, 9H);
13C NMR (CDCl3) δ 159.8, 155.2, 152.8 (d, J ) 6.6 Hz), 144.1
(d, J ) 27.8 Hz), 143.0 (d, J ) 257 Hz), 133.5, 117.5, 79.8,
72.7, 51.2, 51.1, 31.7, 28.8.
To a solution of lithium tert-butoxide (71.92 g, 0.90
mol) in a mixture of tert-butanol (400 mL) and DMSO
(400 mL), was added the above compound 6e in DMSO.
This mixture was heated to 60 °C and stirred for 1 h
(HPLC showed complete isomerization of both allyl
protecting groups) to give compound 15 (mixture of two
isomers in about 40/60 ratio by HPLC). The reaction
mixture was cooled to 50 °C, and 6 N HCl was added
slowly until the pH of the mixture reached about 2.5 (about
174 mL 6 N HCl). This mixture was stirred at 50 °C for
12 h (HPLC shows complete deprotection of diallyl with
the Boc protecting group intact). It was cooled to room
temperature and diluted with ethyl acetate (1000 mL) and
then carefully neutralized with a mixture of saturated
aqueous sodium bicarbonate (500 mL) and brine (300 mL).
The organic layer was washed with 30% sodium bisulfite
(300 mL) and then with water (300 mL × 2). The organic
layer was then heated with charcoal (Darco G-60, 20 g)
and Celite (Celpure P100, 20 g) at 40 °C for 1 h. This
mixture was filtered through a Celite pad (Celpure P100),
and the filtrate was concentrated under vacuum at a
temperature below 40 °C to afford a yellow slurry. The
slurry was diluted with ethyl acetate (500 mL), and the
mixture was heated at about 80 °C until all solids were
dissolved. The solution was concentrated atmospherically
to 300 mL, and heptane was added slowly until the mixture
tert-Butyl 4-(4-amino-5-fluoropyrimidin-2-yloxy)piperi-
dine-1-carboxylate (6a). To a solution of 2,4-dichloro-5-
fluoropyrimidine (12, 100.0 g, 0.60 mol) in THF (1000 mL)
was added diallylamine (64.00 g, 81.1 mL, 1.1 equiv) followed
by triethylamine (166.7 mL, 1.20 mol) over 15 min, and the
resulting mixture was stirred at room temperature for 2 h (HPLC
showed >99% conversion). The mixture was filtered, and the
solids were washed with THF (100 mL × 2). The pale-yellow
filtrate was washed with a mixture of brine (150 mL) and 10%
aqueous citric acid (250 mL). The organic layer was concen-
trated under vacuum at a temperature below 40 °C to about
200 mL (additional THF was added during the distillation to
ensure a KF of the concentrate of below 0.4%) and diluted with
200 mL of THF. Residual solid was removed by filtration and
washed with THF (100 mL). The crude amination product 9e
(98.7% purity by HPLC area) in the combined filtrates were
used directly in the next chemical step. An analytically pure
sample of 9e was obtained by column chromatography and
1
characterized. MS: MH+ ) 228.0; H NMR (CDCl3) δ 7.92
(d, J ) 6.0 Hz, 1H), 5.85 (m, 2H), 5.22 (m, 4H), 4.18 (dd, J )
5.6 Hz, 1.0 Hz, 4H); 13C NMR (CDCl3) δ 154.4, 152.9 (d, J )
6.6 Hz), 145.7 (d, J ) 257 Hz), 144.0 (d, J ) 26.3 Hz), 132.7,
118.2, 51.3, 51.2.
To a THF solution (500 mL) of sodium tert-butoxide (86.50
g, 0.90 mol) and Boc-piperidinol 3 (126.8 g, 0.63 mol) was
added the above THF solution of 9e over 15 min. This mixture
was stirred at 40 °C for 2 h (HPLC showed >98% conversion).
Without cooling, water (500 mL) and brine (300 mL) were
added. The layers were spli,t and the organic layer was washed
with brine (500 mL). The organic layer was then concentrated
under vacuum at a temperature below 40 °C to about 200 mL
810
•
Vol. 13, No. 4, 2009 / Organic Process Research & Development

started to turn cloudy (about 500 mL). This mixture was
then cooled gradually to 0 °C and stirred for 1 h at 0 °C.
The solids were collected by filtration, washed at 0 °C
with 1:3 ethyl acetate/heptane (100 mL × 2), and dried
under vacuum at 50 °C to give 6a as a yellow solid (138.0
g, 74% yield, 98.6% purity by HPLC area). Mp by DSC
(differential scanning calorimetry): 166 °C (peak temper-
ature). HRMS: Calcd for C14H22FN4O3 (MH+): 313.1676.
0 °C, and stirred for 1 h at 0 °C. The solid was filtered, washed
with isopropyl alcohol (138 mL), and dried under vacuum at
50 °C to constant weight to afford 1a as a pale-yellow solid
(116.0 g, 92% yield, 99.2% purity by HPLC area). Mp by DSC:
224-246 °C dec. Anal. Calcd for C9H15Cl2FN4O: C, 37.91;
H, 5.30; N, 19.65; Cl, 24.87. Found: C, 37.65; H, 5.47; N, 19.21;
Cl, 24.67. HRMS: calcd for C9H14FN4O (MH+): 213.1152.
Found: 213.1145. 1H NMR (D2O) δ 7.90 (d, J ) 2.4 Hz, 1H),
5.30 (m, 1H), 3.30 (m, 2H), 3.20 (m, 2H), 2.10 (m, 4H). 13C
NMR (D2O) δ 158.8 (d, J ) 15 Hz), 154.7, 140.2 (d, J ) 249
Hz), 128.4 (d, J ) 31 Hz), 72.6, 40.7, 26.6.
1
Found: 313.1662. H NMR (CDCl3) δ 7.88 (d, J ) 2.4
Hz, 1H), 5.18 (s, 2H), 5.04 (m, 1H), 3.78 (m, 2H), 3.30
(m, 2H), 1.95 (m, 2H), 1.78 (m, 2H), 1.47 (s, 9H). 13C
NMR (CDCl3) δ 160.0, 155.2, 155.1 (d, J ) 13 Hz), 142.7
(d, J ) 247 Hz), 141.1 (d, J ) 20 Hz), 80.0, 72.4, 40.9
(br s), 31.0, 28.8.
Acknowledgment
We thank Leonard Hargiss for his assistance in LC-MS.
Marianne Carlsen and Mark Bednarz are acknowledged for their
work on the Mitsunobu-based approach.
5-Fluoro-2-(piperidin-4-yloxy)pyrimidin-4-amine Dihy-
drochloride (1a). To a solution of 6a (138.0 g, 0.44 mol) in
isopropyl alcohol (552 mL) was added slowly at 60 °C a
solution of HCl in isopropyl alcohol which was preprepared
by slow addition of acetyl chloride (126 mL, 1.77 mol) to
isopropyl alcohol (276 mL) at 0 °C. The mixture was heated at
60 °C for 1 h (HPLC shows complete deprotection), cooled to
Received for review March 13, 2009.
OP900060U
Vol. 13, No. 4, 2009 / Organic Process Research & Development
•
811