An efficient synthesis of 3,3-difluoro-pyrrolidine hydrochloride starting with 2-chloro-2,2-difluoroacetic acid

Article abstract of DOI:10.1016/j.jfluchem.2011.08.009

A facile and fluorination-free synthesis of 3,3-difluoropyrrolidine hydrochloride (3), an important synthon in the synthesis of biologically active compounds, is reported. The seven-step synthesis starts from the commercially available 2-chloro-2,2-difluoroacetic acid (1) in a three-step telescoped process that produces crystalline N,N-diethyl-2,2-difluoro-3-hydroxy-4- nitrobutanamide (2). A convenient and high-yielding reductive nitromethylation of 2 followed by a catalytic hydrogenation/cyclization sequence and borane reduction affords 3 in good yield and purity.

Full text of DOI:10.1016/j.jfluchem.2011.08.009

Journal of Fluorine Chemistry 135 (2012) 354–357
Contents lists available at SciVerse ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Short communication
An efficient synthesis of 3,3-difluoro-pyrrolidine hydrochloride starting with
2-chloro-2,2-difluoroacetic acid
a
b
Lulin Wei a, , Teresa M. Makowski , Jennifer L. Rutherford
*
a Chemical Research and Development, Pfizer Worldwide Research and Development, Eastern Point Road, PO Box 8013, Groton, CT 06340-8013, USA
b Lafayette College, Department of Chemistry, Hugel Science Center, Easton, PA 18042, USA
A R T I C L E I N F O
A B S T R A C T
Article history:
A facile and fluorination-free synthesis of 3,3-difluoropyrrolidine hydrochloride (3), an important
synthon in the synthesis of biologically active compounds, is reported. The seven-step synthesis starts
from the commercially available 2-chloro-2,2-difluoroacetic acid (1) in a three-step telescoped process
that produces crystalline N,N-diethyl-2,2-difluoro-3-hydroxy-4-nitrobutanamide (2). A convenient and
high-yielding reductive nitromethylation of 2 followed by a catalytic hydrogenation/cyclization
sequence and borane reduction affords 3 in good yield and purity.
Received 13 July 2011
Received in revised form 10 August 2011
Accepted 16 August 2011
Available online 22 August 2011
Keywords:
ß 2011 Elsevier B.V. All rights reserved.
3,3-Difluoropyrrolidine hydrochloride
2-Chloro-2,2-difluoroacetic acid
Reductive nitromethylation
Catalytic hydrogenation/cyclization
1. Introduction
pyrrolidione derived from (2R)-4-hydroxypyrrolidine-2-carboxyl-
ic acid in a multi-step synthesis that included hazardous and
The introduction of a gem-difluoro moiety in a pharmaceutical
agent can modify its biological activity, pharmacodynamism and
metabolism and has gained popularity over the years [1–3].
Indeed, a gem-difluorocyclohexane was incorporated in maraviroc,
the active ingredient of SelzentryTM, a CCR5 inhibitor for the
treatment of HIV [4]. The fluorination of ketones with (diethyla-
mido)sulfur trifluoride (DAST) pioneered by Middleton [5] has
been the leading method used for the synthesis of this type of
expensive reagents [17]. To address these issues, Xu developed an
alternative process which did not rely on the fluorination of a
pyrrolidone and featured a Claisen rearrangement to establish the
difluoro quarternary center [18]. This approach also provided a
route to
3 from commercially available starting material,
difluorochloroacetic acid (1).
During the course of development of a Pfizer candidate [19], we
were presented with the challenge to investigate an efficient/cost
effective process for the synthesis of 3, a key building block for this
candidate. We also selected 1 as the starting point of our synthesis
but sought an alternative synthetic sequence.
fluorinated building blocks. Over the years,
a number of
alternatives to DAST have been reported to address safety concerns
[6–8] and several new synthetic methods have attempted to
address these issues [9,10].
2. Results and discussion
The alternative to the preparation of fluorinated compound is to
elaborate starting materials already containing the fluorine
moiety. Due to the ozone depleting effect of many fluorocarbons,
increasingly tougher regulations have been imposed on fluorine-
substituted compounds around the world [11]. Traditionally used
building blocks for 3, such as geminally difluoro-substituted
butenes, are at risk to disappear.
As shown in Scheme 1, our retrosynthesis aimed at achieving the
goals of employing the inexpensive and readily available starting
material 1 and avoiding the use of expensive and unstable
fluorinating reagents such as DAST. It was hoped that conversion
of 1 to a suitable protected g-aminocarboxylic acid derivative (4)
wouldleadtothedesiredlactam(5)upondeprotection,releaseofthe
freeamineand intramolecularcyclization. Reduction ofthecarbonyl
of 5 would then provide the requisite fluorinated pyrolidine (3).
An ester or an amide was considered as a suitable carboxylate
derivative. From a literature survey [20,21] we decided to explore
ethyl 2,2-difluoro-3-hydroxy-4-nitrobutyrate and N,N-diethyl-
The 3,3-difluoropyrrolidine 3 is a structural motif which has
been used in the synthesis of a large number of important
biologically active compounds including several dipeptidyl pepti-
dase IV inhibitors [12–16]. The original approach reported for the
synthesis of
3 employed DAST-mediated fluorination of 3-
2,2-difluoro-3-hydroxy-4-nitrobutanamide
2 as intermediates
which would contain the carbon framework and the masked
primary amine of our target. Upon preparation of both compounds,
* Corresponding author. Tel.: +1 860 441 8701; fax: +1 860 441 5540.
E-mail address: lulin.wei@pfizer.com (L. Wei).
0022-1139/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2011.08.009

L. Wei et al. / Journal of Fluorine Chemistry 135 (2012) 354–357
355
F
F
generated by reacting 2 with acetic anhydride with catalytic 4-
dimethylaminopyridine. As expected, it was found that the
acylation of 2 proceeded more rapidly with acetic anhydride
when DMAP was used as a catalyst. The resulting acetate was
telescoped to the next step and reacted with sodium borohydride
to afford nitroamide 6 [23].
F
F
F
F
Cl
X
F
F
O
CO2H
N
H
5
N
H
3
O
•HCl
R2N
1
4
The conversion of 6 to lactam 5 required a more extensive
study to identify conditions which provided high yielding and
high quality material. Initial attempts at a transfer hydrogenation
protocol produced only moderate yield (40%) [24]. Disappointed
by the yield for this transformation, we engaged in a broad screen
of catalysts known to be effective for the reduction of nitro
groups. Aside from starting material (6) and product (5), two
other compounds were observed; the uncyclized amide (7) and
over-reduced alkane (8). As shown in Table 1, several variables
Scheme 1. Retrosynthesis of 3,3-difluoropyrrolidine (3).
amide 2 was selected for future work due to its crystallinity, ease of
handling and facile purification.
As shown in Scheme 2, compound 2 was readily prepared in a
three-step process from chlorodifluoroacetic acid 1 using similar
procedures as previously reported by Tsukamoto and Kitazume
[20,21] and Mulliez et al. [22]. With 2 in hand, nitroalkane 6 was
Scheme 2. Synthesis of 3,3-difluoropyrrolidine (3). (a) N,N-Diethylcarbamoyl chloride, Et3N, room temp., 2 h; (b) ethyl toluene-p-sulfonate, Zn, DMF, 90 8C, 6 h followed by
H2SO4, EtOH, 10 min; (c) MeNO2, K2CO3, THF, room temp., 2 h; 55% for a–c; (d) Ac2O/cat. DMAP; (e) 1 M NaBH4 in EtOH, rt. 97% for d–e; (f) 5% Pt/C, JM 5R117, 70 psig, toluene;
90%; (g) BH3ÁTHF, 50 8C, 82%.
Table 1
Catalyst evaluation and reaction conditions for conversion of 6 to 5.
.
Cat.
Composition
T 8C
P psig
Sol.
% 6/5/7/8
1
A
B
C
D
A
B
C
E
5% Pd/C, 0.5% V/C
1% Pt/C, 2% V/C
5% Pt/C
50
50
50
50
70
70
70
70
70
70
70
70
70
70
60
60
60
60
60
60
60
60
70
70
70
50
50
50
50
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
MeOH
MeOH
THF
93/7/0/0
2
92/8/0/0
3
14/42/0/44
30/16/0/54
10/78/12/0
20/60/18/0
12/88/0/0
22/78/0/0
11/89/0/0
14/86/0/0
5/95/0/0
4
5% Pt(S)/C
5% Pd/C, 0.5% V/C
1% Pt/C, 2% V/C
5% Pt/C
5
6
7
8
5% Pt(S)/C
5% Pt/C
9
F
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
G
C
H
D
E
5% Pt/C
5% Pt/C
5% Pt/C
23/77/0/0
5/95/0/0
5% Pt(S)/C
5% Pt/AP
4/96/0/0
C
D
C
D
C
D
C
D
C
C
C
5% Pt/C
0/35/15/49
0/21/17/62
10/63/24/3
9/59/32/0
100/0/0/0
100/0/0/0
93/6/0/0
5% Pt(S)/C
5% Pt/C
5% Pt(S)/C
5% Pt/C
THF
DME
5% Pt(S)/C
5% Pt/C
DME
EtOAc
EtOAc
PhMe
PhMe
PhMe
5% Pt(S)/C
5% Pt/C, 0.75 AcOH
5% Pt/C, 0.5 AcOH
5% Pt/C*
90/8/0/0
100/0/0/0
82/18/0/0
1/99/0/0
All reductions unless noted were on the Biotage EndeavorTM for 24 h.
Catalyst source: A = Degussa CE105, B = Degussa CF105, C = JM 5R117, D = JM B106032-5, E = JM 5R94, F = JM 5R18, G = JM 5R103, H = JM 5R128 M.
*
5 g scale on Biotage AtlantisTM
.

356
L. Wei et al. / Journal of Fluorine Chemistry 135 (2012) 354–357
were explored during our initial screen. While lower pressure did
not appear effective (entries 1–4), the reaction seem to improve
when conducted at 70 psig. Toluene was selected as the solvent of
choice since the evaluation of alternative solvents did not lead to
improvement (entries 15–22). Addition of acetic acid (entries 23–
24) led to inhibition of the reaction. While several Pt/C sources
proved to be suitable for this reaction in toluene at 70 8C and
70 psig, we narrowed our catalyst selection to the use of JM
5R117. While the screen afforded a 95:5 ratio of 5:6 (entry 11) it
was further improved to 99:1 at a 5 g scale (entry 25). Better
mixing of the gas, liquid, and solid phases in the larger reactor
explains the difference between screening ratio of 5:6 and the
isolated scaled ratio. The use of these conditions during the
hydrogenation/cyclization produced the desired product 5 in a
90% isolated yield, which was then crystallized as a yellow solid
from toluene.
combined organic layers were dried over MgSO4 and filtered.
The solvent was evaporated to afford a crude mixture of 6 (6.5 g,
97%) which was used in the next step without further
purification. The analytically pure sample was prepared by
chromatography on silica gel (Merck silica gel 60, 40–63
(heptanes/EtOAc, 5:1 as eluent) to give a light yellow oil 1H NMR
(CDCl3) 1.07 (t, 3H, J = 7.0 Hz, NCH2CH3), 1.13 (t, 3H, J = 7.0 Hz,
mm)
d
NCH2CH3), 2.80–2.92 (tt, 2H, J = 16.6 Hz, 7.03 Hz, CH2CF2), 3.26–
3.30 (q, 2H, J = 7.0 Hz, 7.0 Hz, NCH2CH3), 3.41–3.45 (q, 2H,
J = 7.0 Hz, 7.0 Hz, NCH2CH3), 4.55–4.58 (t, J = 7.0 Hz, 2H,
O2NCH2); 13C NMR (CDCl3)
d 12.1 (CH3), 14.0 (CH3), 32.5 (t,
J = 24.5 Hz, CH2CF2), 41.4 (NCH2), 41.7 (NCH2), 69.0 (O2NCH2),
118.0 (t, J = 257.8 Hz, CF2), 161.3 (t, J = 28.3 Hz, CO). Elemental
analysis: calcd for C8H14F2N2O3: C, 42.86; H, 6.29; N, 12.49.
Found: C, 42.90; H, 6.44; N. 12.32.
The final part of the synthesis was accomplished by the
reduction of lactam 5 with borane in THF at 50 8C. After the
reduction, the reaction was quenched with dry MeOH followed by
4 M HCl in dioxane to hydrolyze the six-membered borazine
derivative as well as to convert the volatile free base 3,3-
difluoropyrrolidine to its HCl salt. The title compound 3 was
isolated in good yield (82%) and excellent quality [25].
4.3. Preparation of 3,3-difluoropyrrolidin-2-one (5) using transfer
hydrogenation
To a solution of crude 6 (5 g, 22 mmol) in MeOH (200 ml) was
added 10% Pd on carbon (2.32 g, 2.18 mmol), ammonium formate
(6.19 g, 98 mmol) under N2 at room temperature. The mixture was
heated at 45–50 8C overnight and cooled to room temperature. The
catalyst was removed by filtration. The filtrate was concentrated
and re-dissolved in THF (50 ml). The resulting mixture was heated
at 50 8C overnight. The solution was concentrated and residue was
purified by column chromatography (3/5 EtOAc/heptane followed
by EtOAc) to afford 3,3-difluoropyrrolidin-2-one (5) (1.0 g, 40%) as
light-yellow crystalline solid (40%). Mp 68–71 8C; 1H NMR (CDCl3)
3. Conclusion
In conclusion, an efficient synthesis of 3,3-difluoro-pyrrolidine
hydrochloride starting from the commercially available chlorodi-
fluoroacetic acid (1) in about 40% overall yield was developed. The
route involves 7 synthetic steps which upon telescoping led to a
process containing only four isolations. An extensive catalyst
screen identified preferred conditions for the reduction of the
nitroalkyl moiety followed by subsequent cyclization to the 5-
membered ring lactam (5) which was further reduced to the
pyrrolidine (3) using borane-THF.
d
2.49 (tt, 2H, J = 15.2, 6.4 Hz), 3.43 (tq, 2H, J = 6.4 Hz, 1.2 Hz), 8.18
(br, s, 1H); 13C NMR (CDCl3)
d
30.4 (t, J = 23.0 Hz), 36.7 (t,
J = 3.8 Hz), 117.8 (t, J = 249.4 Hz), 167.0 (t, J = 30.7 Hz); the data
was in accordance with the reported data from an alternative
synthesis [24].
4.4. Preparation of 3,3-difluoropyrrolidin-2-one (5)
4. Experimental
To a solution of crude 6 (5 g, 22 mmol) in toluene (50 ml) was
added 5% Pt/C (Johnson Matthey type 5R117). The reaction mixture
was purged with nitrogen 4 times (pressurize to 50 psig and vent)
and with hydrogen 4 times. The reaction was then heated to 70 8C,
pressurized to 70 psig and hydrogenated for 16 h. The reaction was
filtered through Celite to remove the catalyst, washed with toluene
(10 ml), and concentrated to about 5 ml to precipitate 5 as a yellow
solid (2.4 g, 90%).
4.1. General information
Melting points were determined on a capillary apparatus and
were uncorrected. Solvents and reagents were reagent grade,
obtained from commercial sources and used without further
purification. Zinc powder purchased from Aldrich was freshly
activated according to literature method [22]. All reactions were
carried out in oven dried glassware under nitrogen. NMR spectra
were run in CDCl3 on a Bruker 400 MHz instrument and recorded at
the following frequencies: proton (1H, 400 MHz), carbon (13C,
4.5. Preparation of 3,3-difluoropyrrolidine HCl (3)
100 MHz). Chemical shifts (
constant (J) in Hz. Compound 2 was prepared by formylation of
-difluorinated Reformatsky reagent and reaction with nitro-
methane, using known literature procedures [20–22]. Elemental
analyses were performed by Intertek USA, Inc. using a Perkin-
Elmer 2400 Elemental Analyzer.
d
) are reported in ppm and coupling
To a solution of 3,3-difluoropyrrolidin-2-one (5) (0.36 g,
30 mmol) in dry THF (4 ml) under N2 was added borane in THF
(10 ml of 1 M solution) dropwise. The reaction mixture was stirred
at 50 8C for 4 h, cooled to 0 8C, quenched with dry MeOH (1.5 ml)
followed by a 4 M solution of hydrogen chloride in 1,4-dioxane
(5 ml, 20 mmol). The mixture was warmed to 50 8C for 1 h,
concentrated and then treated with 1 ml of toluene to afford 3,3-
difluoropyrrolidine hydrochloride (3) (0.35 g, 82%): 1H NMR
a,a
4.2. Preparation of N,N-diethyl-2,2-difluoro-4-nitrobutanamide (6)
(CD3OD)
d 3.72 (t, 2H, J = 11.9 Hz), 3.60 (t, 2H, J = 7.8 Hz), 2.60
A solution of N,N-diethyl-2,2-difluoro-3-hydroxy-4-nitrobu-
tanamide 2 (7.2 g, 30 mmol), DMAP (183 mg, 1.5 mmol), Ac2O
(3.33 g, 3.1 ml, 32.7 mmol), and ether (75 ml) was stirred at
room temperature for 4 h and concentrated. A solution of NaBH4
(2.25 g, 59.5 mmol) in EtOH (60 ml) was added dropwise to
the above nitroacetate at 0 8C and the solution was stirred at
room temperature overnight. The reaction was quenched
with 1 N HCl (100 ml). Ethanol was evaporated and the aqueous
layer was extracted with ethyl acetate (3Â 100 ml). The
(m, 2H). This NMR data was in accordance with the reported data
from an alternative synthesis [18].
Acknowledgements
´
The authors thank Dr. Stephane Caron for support, guidance and
review of the manuscript. The authors also thank Dr. Pascal Dube´
and Dr. Juan Colberg for their encouragements made during the
course of this work.

L. Wei et al. / Journal of Fluorine Chemistry 135 (2012) 354–357
357
R.A. Patel, S.B. Patel, A. Petrov, G. Scapin, J. Shang, R.S. Roy, A. Smith, J.K. Wu, S. Xu,
B. Zhu, N.A. Thornberry, A.E. Weber, J. Med. Chem. 49 (2006) 3614–3627.
[13] T. -YTsai, M.S. Coumar, T. Hsu, H.-P. Hsieh, C.-H. Chien, C.-T. Chen, C.-N. Chang, Y.-
K. Lo, S.-H. Wu, C.-Y. Huang, Y.-W. Huang, M.-H. Wang, H.-Y. Wu, H.-J. Lee, X. Chen,
Y.-S. Chao, W.-T. Jiaang, Bioorg. Med. Chem. Lett. 16 (2006) 3268–3272.
[14] Z. Pei, X. Li, T.W. Von Geldern, K. Longenecker, D. Pireh, K.D. Stewart, B.J. Backes, C.
Lai, T.H. Lubben, S.J. Ballaron, D.W.A. Beno, A.J. Kempf-Grote, H.L. Sham, J.M.
Trevillyan, J. Med. Chem. 50 (2007) 1983–1987.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jfluchem.2011.08.009.
References
[15] C.A. Smith, L.E. Cross, K. Hughes, R.E. Davis, D.B. Judd, A.T. Merritt, Tetrahedron
Lett. 50 (2009) 4906–4911.
[16] Y. Xiong, B.R. Teegarden, J.-S.K. Choi, S. Strah-Pleynet, M. Decaire, H. Jayakumar,
P.I. Dosa, M.D. Casper, L. Pham, K. Feichtinger, B. Ullman, J. Adams, D. Yuskin, J.
Frazer, M. Morgan, A. Sadeque, W. Chen, R.R. Webb, D.T. Connolly, G. Semple, H.
Al-Shamma, J. Med. Chem. 53 (2010) 4412–4421.
[1] J.M. Dener, C. O’Bryan, R. Yee, E.J. Shelton, D. Sperandio, T. Mahajan, J.T. Palmer, J.R.
Spencer, Z. Tong, Tetrahedron Lett. 47 (2006) 4591–4595.
[2] F. Rossi, F. Corcella, F.S. Caldarelli, F. Heidempergher, C. Marchionni, M. Auguadro,
M. Cattaneo, L. Ceriani, G. Visentin, G. Ventrella, V. Pinciroli, G. Ramella, I.
Candiani, A. Bedeschi, A. Tomasi, B.J. Kline, C.A. Martinez, D. Yazbeck, D.J. Kucera,
Org. Process Res. Dev. 12 (2008) 322–338.
[3] J. Tadanier, W. Cole, J. Org. Chem. 26 (1961) 2436–2440.
[4] D.A. Price, S. Gayton, M.D. Selby, J. Ahman, S. Haycock-Lewandowski, B.L. Stam-
men, A. Warren, Tetrahedron Lett. 46 (2005) 5005–5007.
[17] G. Giardina, G. Dondio, M. Grugni, Synlett (1995) 55–57.
[18] F. Xu, B. Simmons III, J. Armstrong, J. Murry, J. Org. Chem. 70 (2005) 6105–6107.
[19] M.J. Ammirati, K.M. Andrews, D.D. Boyer, A.M. Brodeur, D.E. Danley, S.D. Doran, B.
Hulin, S. Liu, R.K. McPherson, S.J. Orena, J.C. Parker, J. Polivkova, X. Qiu, C.B. Soglia,
J.L. Treadway, M.A. VanVolkenburg, D.C. Wilder, D.W. Piotrowski, Bioorg. Med.
Chem. Lett. 19 (2009) 1991–1995.
[5] W.J. Middleton, J. Org. Chem. 40 (1975) 574–578.
[20] T. Tsukamoto, T. Kitazume, J. Chem. Soc., Chem. Commun. (1992) 540–541.
[21] T. Tsukamoto, T. Kitazume, J. Chem. Soc., Perkin Trans. 1 (1993) 1177–1181.
[22] L. Bijeire, C. Denier, M. Mulliez, W. Tavares De Lima, Phosphorus, Sulfur Silicon
Relat. Elem. 182 (2007) 881–891.
[6] J. Cochran, Chem. Eng. News 57 (1979) 4.
[7] P.A. Messina, K.C. Mange, W.J. Middleton, J. Fluorine Chem. 42 (1989) 137–143.
[8] W.J. Middleton, Chem. Eng. News 57 (1979) 43.
[9] R.P. Singh, J.n.M. Shreeve, Synthesis (2002) 2561–2578.
[23] R.H. Wollenberg, S.J. Miller, Tetrahedron Lett. 19 (1978) 3219–3222.
[24] T. Okano, N. Takakrua, Y. Nakano, A. Okajima, S. Eguchi, Tetrahedron 51 (1995)
1903–1920.
[10] A. L’Heureux, F. Beaulieu, C. Bennett, D.R. Bill, S. Clayton, F. La Flamme, M. Mir-
mehrabi, S. Tadayon, D. Tovell, M. Couturier, J. Org. Chem. 75 (2010) 3401–3411.
[11] A. McCulloch, J. Fluorine Chem. 123 (2003) 21–29.
[25] R.D. Roberts, H.E. Ferran, M.J. Gula, T.A. Spencer, J. Am. Chem. Soc. 23 (1980)
7054–7058.
[12] S.D. Edmondson, A. Mastracchio, R.J. Mathvink, J. He, B. Harper, Y.J. Park, M.
Beconi, J. Di Salvo, G.J. Eiermann, H. He, B. Leiting, J.F. Leone, D.A. Levorse, K. Lyons,