Synthesis and applications of the first polyfluorous proline derivative

Article abstract of DOI:10.1016/S0957-4166(02)00796-6

Starting from trans-4-hydroxy-L-proline, a polyfluorinated proline derivative having a fluorine content of 54% has been prepared. It has been tested as a catalyst in the aldol reaction between p-nitrobenzaldehyde and acetone with 73% ee in BTF and in copper(I)-catalyzed allylic oxidation of cyclohexene with 20% ee in HFIP.

Full text of DOI:10.1016/S0957-4166(02)00796-6

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 139–143
Synthesis and applications of the first polyfluorous proline
derivative
Fabienne Fache* and Olivier Piva
Laboratoire de Chimie Organique, Photochimie et Synthe`se, Universite´ Claude Bernard Lyon I, CNRS UMR 5622,
Baˆt. J. Raulin, 43 bd du 11 novembre 1918, 69622 Villeurbanne cedex, France
Received 22 October 2002; accepted 20 November 2002
Abstract—Starting from trans-4-hydroxy-L-proline, a polyfluorinated proline derivative having a fluorine content of 54% has been
prepared. It has been tested as a catalyst in the aldol reaction between p-nitrobenzaldehyde and acetone with 73% ee in BTF and
in copper(I)-catalyzed allylic oxidation of cyclohexene with 20% ee in HFIP. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
Proline is a well known a-amino acid used in numerous
reactions.1 It can act as a ligand in asymmetric transi-
tion-metal catalysis such as allylic oxidation of olefins2
or reduction by transfer hydrogenation of ketones.3 It
can also be an effective organocatalyst4,5 for aldol and
Michael reactions.6 Even if the first reported use of
proline in asymmetric catalysis was made by Hajos and
Parrish7 in the early 1970s for intramolecular aldol
reactions, proline is still a chiral compound of current
interest as is proven by the large number of articles,
from letters to reviews, which have been published on
the subject this year.1,4,5,8,9
We first had to choose an anchoring site for the
perfluorinated alkyl chain we wished to graft on the
proline skeleton. As both the nitrogen atom and the
carboxylic function have to remain untouched we
decided to start from 4-hydroxy-L-proline and to intro-
duce a perfluorinated chain via radical addition of a
perfluoroalkyl group to an allyl ether funtionality. The
overall synthesis is depicted in Scheme 1.
Before the introduction of the allyl group, the amino
group has to be protected. The t-butoxycarbonyl group
(t-Boc) was first tried but at the end of the synthesis,
after deprotection with trifluoroacetic acid, it was
difficult to recover the free polyfluoroamino acid. We
finally chose the benzyloxycarbonyl group (Cbz), which
was cleaved at the end of the synthesis by hydrogenoly-
sis during the reduction of the double bond by catalytic
hydrogenation.
At the same time, there is growing interest for non-
standard solvents such as fluorinated ones in asymmet-
ric catalysis.10 These media were selected to ensure
good separation between the catalyst and the products
and to allow recycling of the catalytic system. The
common strategy was to prepare polyfluorous ligands
and therefore numerous well known chiral ligands such
as BINOL,11 BINAP,12 quinine13 or MOP14 derivatives
have been synthesized and used in fluorous biphase
media (FBS).
Thus, 4-hydroxy-L-proline was derivatized using benzyl
chloroformate in the presence of aq. NaHCO3. Com-
pound 1 was isolated quantitatively and could be used
in the following step without further purification. Allyl-
ation of 1 gave 2 using NaH and allyl bromide and
after saponification the free acid 3 was isolated with
51% isolated yield. The double bond thus introduced
allowed now the introduction of the fluorinated chain.
Several methods were tested. With sodium dithionite,15
only traces amounts of the desired product were iso-
We want to report here the first synthesis of a proline
derivative bearing a polyfluorinated ether chain and the
results of its first tests in asymmetric catalysis.
* Corresponding author. Fax: 00 33 (0)4 72 44 81 36; e-mail: fache@
univ-lyon1.fr
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00796-6

140
F. Fache, O. Pi6a / Tetrahedron: Asymmetry 14 (2003) 139–143
Scheme 1. Reagents and conditions: (i) Cl(CO)OCH2Ph, NaHCO3; (ii) NaH, DMF, BrCH2CH:CH2; (iii) NaOH, H2O; (iv) C8F17I,
CuCl, NH2CH2CH2OH; (v) H2, Pd/C.
16
lated and BEt3/O2 was totally inefficient. Finally, the
method described by Riess et al.17 using CuCl, C8F17I
and ethanolamine gave 4 in 60% isolated yield. Only
the elimination product 4 was detected, with no traces
of the iodinated product. The E isomer was the most
abundant (E/Z ratio: 80/20). We did not try to separate
them as the double bond has to be reduced under H2
pressure using Pd/C to give 5 in 85% yield, with an
overall fluorine content of 54%.
where allylic oxidation was performed with success in
benzenetrifluoride (BTF). In this solvent, the reactivity
of our system was high but again the enantioselectivity
was very poor (Table 1, entry 5). Finally, we tested
HFIP (hexafluoroisopropanol) as solvent, and with 5
77% isolated yield and 20% ee were obtained on cyclo-
hexene. We have very recently shown21 that Cu2O
combined with a perfluoroacid C11F23COOH allowed
the allylic oxidation of olefins in HFIP and that our
system was recyclable. In the absence of acid no reac-
tion occurred in this solvent and we proposed that the
acid is necessary to ensure good solubility of the cata-
lytic species in HFIP. In this asymmetric version of our
system, the acid was useless, probably because the
polyfluoroproline derivative 5 plays the same role and
ensures the solubilization of copper in HFIP. More-
over, the catalytic system could be recovered very easily
(see experimental part) and reused without additional
copper and ligand (Table 1, entry 7). Unfortunately,
both activity and selectivity diminished (54% isolated
yield and 13% ee).
Proline has been successfully used in allylic
oxidation2,18 of olefins with copper derivatives. We
decided therefore to investigate the performance of our
new polyfluorinated proline derivative 5 in this reaction
(Scheme 2).
Numerous chiral ligands have been described for such a
reaction and ees of up to 80%19 have been achieved
with cyclohexene. More specifically, with proline in
benzene, ee of 45% was obtained by Muzart et al.2 with
59% isolated yield. Under the same conditions with
Cu2O (10% Cu) and 5 similar activity was obtained
(53% yield, Table 1, entry 1) but the enantioselectivity
was much lower (10%). We therefore searched for a
more suitable solvent and tested different systems. Pre-
liminary results are summarized in Table 1.
Our polyfluorous proline derivative 5 was also used in
the aldol reaction between p-nitrobenzaldehyde and
acetone (Scheme 3).
It has been shown that the intermolecular asymmetric
reaction between a ketone and an aldehyde was possi-
ble in the presence of proline if a large excess of the
ketone donor was used. Thus, in DMSO, 76% ee and
68% isolated yield were obtained in 5 h for the reaction
depicted in Scheme 3, with 30% proline.5 Extraction of
the crude product was necessary to get rid of the excess
DMSO. We decided to take advantage of the fluorous
tail of 5 and to use benzenetrifluoride (BTF) as solvent
(Table 2).
To take advantage of the fluorotail of 5, we decided to
work in a biphasic system. We chose the ClCH2CH2Cl/
FC-72 mixture for our first test and were surprised to
see no reaction (Table 1, entry 3) whereas ClCH2CH2Cl
or FC-72 alone led respectively to 37% and 58% iso-
lated yield with 3% and 17% ee. In the course of our
study, Muzart et al.20 published an interesting study
The results with 5 were similar to those obtained with
proline in DMSO under standard conditions (Table 2,
entry 2), that is to say, with excess acetone and 25% of
organocatalyst. Nevertheless, there was no need for
extraction and the crude mixture was purified directly
by chromatography without evaporation of the reaction
solvent (BTF). In the same solvent, proline itself was
Scheme 2. Allylic oxidation of cyclohexene.

F. Fache, O. Pi6a / Tetrahedron: Asymmetry 14 (2003) 139–143
141
Table 1. Influence of the solvent on the rate and enantioselectivity of the allylic oxidation of cyclohexene in presence of 5
Entry
Solvent
Time (h)
Isolated yield (%)
ee* (%) (abs. conf.)
1
2
3
4
5
6
7
Benzene
5
5
5
5
5
2
4
53
37
0
58
61
77
54
10 (S)
3 (S)
–
17 (S)
4 (S)
20 (S)
13 (S)
ClCH2CH2Cl
ClCH2CH2Cl+FC-72
FC-72
BTF
HFIPa
Standard conditions: see experimental part; a: without benzoic acid; * ee determined by polarimetry.
Scheme 3. Aldolization reaction.
Table 2. Influence of the reaction conditions on the rate and enantioselectivity of the reaction between p-nitrobenzaldehyde
and acetone in benzenetrifluoride
slightly less efficient in terms of enantioselectivity, but
the reaction took 2 days as compared to only 1 day
for 5. This may be due to the solubility of proline,
which may not be high enough in BTF at the concen-
tration used. More interestingly, when we worked with
a ketone/aldehyde ratio of 1:1, proline was inactive,
whereas 5 was still efficient. As proline is insoluble in
BTF, we can assume that it is the proline solubilized
in acetone that is active and with only an equimolar
quantity of acetone and aldehyde a very small quan-
tity of proline is available for reaction. When we used
only 7% of the organocatalyst, both the activities and
selectivities were similar for proline and its
polyfluorous derivative 5.
3. Conclusion
We have synthesized the first polyfluorous analogue of
proline. The same activity and selectivity were
observed in the aldol reaction between acetone and
p-nitrobenzaldehyde with our compound in benzen-
etrifluoride than with proline itself in DMSO but in a
more volatile solvent and with no need for extraction
before chromatography. For allylic oxidation, the
reaction occurred but with low enantioselectivity. Con-
sidering the numerous reactions using proline, work is
in progress to widen the scope of the applications of
our polyfluoroproline derivative in asymmetric reac-
tions.

142
F. Fache, O. Pi6a / Tetrahedron: Asymmetry 14 (2003) 139–143
4. Experimental
4.4. N-Benzyloxy-trans-4-allyloxy-L-proline, 3
A solution of 2 (2.3 g, 6.67 mmol) in MeOH (8 mL)
was treated with a solution of NaOH (8 g, 20 mmol) in
water (6 mL) were added and the reaction mixture was
allowed to stir at room temperature for 24 h. After
addition of water (5 mL), the pH was adjusted at 10 by
addition of 5N aq. HCl and the reaction mixture was
extracted with Et2O (2×10 mL). The pH was then
adjusted to 4 and the solution extracted with AcOEt
(3×40 mL). These organic phases were dried over
MgSO4 and after evaporation of the solvent and purifi-
cation by flash-chromatography, pure 3 was obtained
(1.83 g, 90%). Oil; Rf=0.65 (CH2Cl2/MeOH: 90/10); 1H
NMR (CDCl3): l 2.13–2.45 (m, 2H), 3.64 (m, 2H), 3.98
(m, 2H), 4.15 (m, 1H), 4.51 (m, 1H), 5.22, 4H, CH2ꢀ
+CH2-Ar), 5.87 (m, 1H, CHꢀ), 7.31 (m, 5H, Ar-H),
8.25 (m, 1H, COOH); 13C NMR (CDCl3): l 35.0 and
36.8, 51.9, 60.5, 67.3 and 67.8 CH2-Ar, 70.2 CH2ꢀ, 75.9
and 76.4, 117.5, 127.6, 128.0, 128.2, 128.4, 128.6 C Ar,
134.2 CHꢀ, 136.1 and 136.4 Cq Ar, 154.5, 156.2.
4.1. General
All commercially available reagents were used as
received. Cyclohexene and all the solvents except HFIP
have been distillated from CaH2 before use. All reac-
tions were monitored by TLC (GF254 Merck). Column
chromatography was performed on silica gel 60 (230–
240 mesh, Merck). Optical rotations were recorded on a
Perkin–Elmer 343 polarimeter. NMR spectra were
recorded with a Bruker AMX 300 spectrometer and
1
referenced as following: H (300 MHz), internal SiMe4
at l 0.00 ppm, 13C (75 MHz), internal CDCl3 at l 77.23
ppm and 19F (282 MHz) external CFCl3 at l 0.00 ppm.
4.2. N-Benzyloxy-trans-4-hydroxy-
L
-proline 1
A solution of trans-4-hydroxy-
L
-proline (4 g, 30.4
mmol) in THF (46 mL) and satd aq. NaHCO3 solution
(80 mL) at 0°C was treated by dropwise addition of
benzoyl chloride (8.56 mL, 60 mmol). The solution was
stirred overnight at room temperature. The pH was
maintained at 1 by addition of 2N aq. HCl and the
reaction mixture was extracted with AcOEt and dried
over MgSO4. After evaporation of the solvent and
flash-chromatography on silica gel, pure 1 was isolated
(7.8 g, 29.4 mmol, 97%). Oil; Rf=0.5 (CH2Cl2/MeOH:
95/5); [h]2D5=−92.2 (c 0.8, CHCl3), lit.22 [h]2D4=−94.1 (c
4.5. N-Benzyloxy-trans-4-(perfluorooctyl)allyloxy-L-pro-
line, 4
A mixture of ethanolamine (1.33 mL, 22 mmol), 3 (1.35
g, 4.43 mmol), t-BuOH (6.6 mL), CuCl (133 mg, 1.34
mmol) and C8F17I (12 g, 22.15 mmol) was stirred at
80°C for 3 days. After extraction with AcOEt, and
flash-chromatography on silica gel, pure 4 was isolated
(1.92 g, 60%). Oil; Rf=0.65 (CH2Cl2/MeOH: 90/10); 1H
NMR (CDCl3): l 2.1–2.6 (m, 2H), 3.68 (m, 2H), 4.14–
4.55 (m, 2H), 5.2 (m, 2H, CH2-Ar), 5.6 (m, 1H, CHꢀ
Z), 5.89 (m, 1H, CHꢀ E), 6.23 (m, 1H, CF2-CH Z),
6.44 (m, 1H, CF2-CH E), Z/E ratio: 20/80, 7.35 (m, 5H
Ar); 19F NMR (CDCl3): l −126.5 (s, 2F, CF2), −124.2
(s, 2F, CF2), −123.6 (s, 2F, CF2), −123.1 (s, 4F, CF2),
−121.9 (s, 2F, CF2), −112.2 (s, 2F, CF2CH E), −108.4
(s, 2F, CF2CH Z), −81.2 (s, 3F, CF3).
1
0.76, CHCl3) H NMR (CDCl3): l 2.2 (m, 2H, CH2),
3.60 (m, 3H, CH2+CH-OH), 4.22 (m, 1H, CH), 5.09
(m, 2H, CH2), 6.38 (OH), 7.29 (m, 5H, H Ar); 13C
NMR (CDCl3) (2 conformational isomers): l 38.0 and
38.9, 54.5 and 54.9, 57.7 and 58.0, 67.6 and 67.7, 69.2
and 69.8, 127.6–128.6 C Ar, 136.0 and 136.1 Cq Ar,
155.1 and 155.9 N-CꢀO, 175.1 and 176.2 CꢀO.
4.3. N-Benzyloxy-trans-4-(allyloxy)allyl-L-prolinate, 2
4.6. trans-4-(Perfluorooctyl)propyloxy-L-proline, 5
A solution of 1 (1.98 g, 7.5 mmol) in dry DMF (18 mL)
at 0°C under nitrogen was treated with NaH (60%
dispersion in oil, 470 mg, 19.5 mmol). After stirring for
30 min, allyl bromide (1.6 mL, 18.7 mmol) was added
with a syringe and the solution was stirred at room
temperature overnight. After addition of water (13
mL), 5N HCl was added until the pH value reached 2.
The solution was then extracted with Et2O (3×40 mL)
and dried over MgSO4. After evaporation of the solvent
and flash-chromatography on silica gel, pure 2 were
isolated (1.47 g, 4.3 mmol, 57%). Oil; Rf=0.57
(petroleum ether/AcOEt: 3/1); [h]2D5=−41.1 (c 1,
A suspension of 4 (1.8 g, 2.49 mmol) and Pd/C (500
mg, 10%) in EtOH (35 mL) was stirred at rt under 12
bar H2 pressure for 24 h. After filtration on celite and
evaporation of the solvent, pure 5 (1.25 g, 2.1 mmol,
1
85%) was isolated. Oil; [h]2D5=−11.7 (c 0.47, EtOH); H
NMR (CDCl3): l 1.85 (m, 1H), 2.12 (m, 2H), 2.48 (m,
1H), 2.92 (m, 4H), 3.50 (m, 5H), 4.16 (m, 1H); 19F
NMR (CDCl3): l −126.7 (s, 2F, CF2), −123.9 (s, 2F,
CF2), −123.3 (s, 2F, CF2), −122.5 (s, 6F, CF2), −114.9
(s, 2F, CF2-CH2), −81.4 (s, 3F, CF3); MH+ calculated
for C16H15F17NO3: 592.07804, found 592.07887.
1
CHCl3); H NMR (CDCl3): l 2.15 (m, 1H, CH2), 2.39
(m, 2H, CH2), 3.74 (m; 2H, CH2), 3.99 (m, 2H, CH2
6
-O-
4.7. Allylic oxidation reaction
CH), 4.16 (m, 1H, CH-CꢀO), 4.46 (m, 1H, CH2-O-
CꢀO), 4.55 (m, 1H, CH-O), 4.67 (d, 1H, J=5.4 Hz,
CH2-O-CꢀO), 5.22 (m, 6H, CH2ꢀCH+CH-Ar), 5.75 (m,
1H, CHꢀ), 5.90 (m, 1H, CHꢀ), 7.33 (m, 5H, Ar-H); 13C
NMR (CDCl3): l 36.0, 37.2, 52.4 and 52.1, 58.5, 66.1,
67.6, 70.6, 75.1, 75.2, 117.8, 119.0 and 118.9, 128.2,
128.3, 128.4, 128.7, 128.8, 132.2 and 131.9, 134.6, 136.9
and 136.7 Cq Ar, 155.3 and 154.8 N-CꢀO, 172.8 CꢀO.
Cyclohexene (0.5 mL, 5 mmol), Cu2O (7 mg, 0.049
mmol), benzoic acid (183 mg, 1.5 mmol), 5 (80 mg, 0.13
mmol) and t-butyl peroxybenzoate (190 mL, 1 mmol)
were mixed in a solvent (1.5 mL) under reflux. Evapo-
ration of the liquid phase and flash-chromatography on
silica gel led to the pure product. For recycling experi-
ment, after evaporation of the solvent, the crude mix-

F. Fache, O. Pi6a / Tetrahedron: Asymmetry 14 (2003) 139–143
143
ture was rinsed with petroleum ether and the catalyst
(blue precipitate) was recovered by simple decantation
of the supernatant liquid.21 The solid catalyst could be
re-used without further addition of copper or ligand. Ee
were measured by polarimetry [h]2D5=−71 (c 2.74,
CHCl3).2
8. (a) Northrup, A. B.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2002, 124, 6798–6799; (b) Kumaragu-
rubaran, N.; Juhl, K.; Zhuang, W.; Bogevig, A.; Jor-
gensen, K. A. J. Am. Chem. Soc. 2002, 124, 6254–6255;
(c) Bogevig, A.; Juhl, K.; Kumaragurubaran, N.;
Zhuang, W.; Jorgensen, K. A. Angew. Chem., Int. Ed.
2002, 41, 1790–1793.
9. (a) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed.
2001, 40, 3726–3748; (b) Jarvo, E. R.; Miller, S. J.
Tetrahedron 2002, 58, 2481–2495.
10. Horva`th, I. T. Acc. Chem. Res. 1998, 31, 641–650.
11. Nakamura, Y.; Takeuchi, S.; Okumura, K.; Ohgo, Y.;
Curran, D. P. Tetrahedron 2002, 58, 3963–3969.
12. Birdsall, D. J.; Hope, E. G.; Stuart, A. M.; Chen, W.;
Hu, Y.; Xiao, J. Tetrahedron Lett. 2001, 42, 8551–8553.
13. Fache, F.; Piva, O. Tetrahedron Lett. 2001, 42, 5655–
5657.
14. (a) Cavazzini, M.; Pozzi, G.; Quici, S.; Maillard, D.;
Sinou, D. Chem. Commun. 2001, 1220–1221; (b) Mail-
lard, D.; Bayardon, J.; Kurichiparambil, J. D.; Ngue-
fack-Fournier, C.; Sinou, D. Tetrahedron: Asymmetry
2002, 13, 1449–1456.
4.8. Aldol reaction
p-Nitrobenzaldehyde (75.5 mg, 0.5 mmol) was dis-
solved in benzenetrifluoride (4 mL) under argon. Excess
acetone (1 mL) were added and then the catalyst (pro-
line or 5). After 48 h stirring the product was purified
by flash-chromatography on silica gel without extrac-
tion and without evaporation of the solvent of reaction
(BTF and acetone). The aldolization product was
esterified with pure (S)-(+)-2-phenylpropionic acid and
the enantiomeric excess of the reaction was measured
1
by H NMR (300 MHz). In the aldolization product, l
CH3: 2.19 ppm. After esterification: l: 2.05 ppm (major
product), 2.25 (minor product).
15. Wakselman, C.; Tordeux, M.; Clavel, J.-L.; Langlois, B.
J. Chem. Soc., Chem. Commun. 1991, 993–994.
16. Takeyama, Y.; Ichinose, Y.; Oshima, K.; Utimoto, K.
Tetrahedron Lett. 1989, 30, 3159–3162.
17. Manfredi, A.; Abouhilale, S.; Greiner, J.; Riess, J. E.
Bull. Soc. Chim. Fr. 1989, 872–878.
18. Andrus, M. B.; Lashley, J. S. Tetrahedron 2002, 58,
845–866.
19. DattaGupta, A.; Singh, V. K. Tetrahedron Lett. 1996,
37, 2633–2636.
20. Le Bras, J.; Muzart, J. J. Mol. Catal. A: Chem. 2002,
185, 113–117.
21. Fache, F.; Piva, O. Synlett 2002, 12, 2035–2036.
22. Robinson, J. K.; Lee, V.; Claridge, T. D. W.; Baldwin,
S. E.; Schofield, C. J. Tetrahedron 1998, 54, 981–996.
References
1. List, B. Tetrahedron 2002, 58, 5573–5590.
2. Levina, A.; Muzart, J. Tetrahedron: Asymmetry 1995, 6,
147–156.
3. Ohta, T.; Nakahara, S.; Shigemura, Y.; Hattori, K.;
Furukawa, I. Chem. Lett. 1998, 6, 491–492.
4. List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem.
Soc. 2000, 122, 2395–2396.
5. List, B.; Pojarliev, P.; Castello, C. Org. Lett. 2001, 3,
573–575.
6. Yamaguchi, M.; Yokota, N.; Minami, T. J. Chem. Soc.,
Chem. Commun. 1991, 1088–1089.
7. Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39,
1615–1621.