Catalyzed reformatsky reactions with ethyl bromofluoroacetate for the synthesis of α-fluoro-β-hydroxy acids

Article abstract of DOI:10.1021/jo015778x

The presence of catalytic amounts of CeCl₃ improves yields and simplifies procedure in the Reformatsky reactions of ethyl bromofluoroacetate with aldehydes and ketones to generate diastereomeric mixtures of α-fluoro-β-hydroxy esters, some of which can be separated by crystallization or column flash chromatography. Diastereomerically pure α-fluoro-β-hydroxy acids are obtained by mild alkaline hydrolysis of the resolved α-fluoro-β-hydroxy esters. Detailed NMR data of new α-fluoro-β-hydroxy esters and α-fluoro-β-hydroxy acids are also presented.

Full text of DOI:10.1021/jo015778x

72
J . Org. Chem. 2002, 67, 72-78
Ca ta lyzed Refor m a tsk y Rea ction s w ith Eth yl Br om oflu or oa ceta te
for th e Syn th esis of r-F lu or o-â-h yd r oxy Acid s
R. Ocampo,† W. R. Dolbier, J r.,*,‡ K. A. Abboud,‡ and F. Zuluaga§
Depto. de Quı´mica, Universidad de Caldas, AA.275, Manizales, Colombia, Department of Chemistry,
University of Florida, Gainesville, Florida 32611, and Depto. de Qu´ımica, University del Valle,
AA 25360 Cali, Colombia
wrd@chem.ufl.edu
Received May 21, 2001
The presence of catalytic amounts of CeCl3 improves yields and simplifies procedure in the
Reformatsky reactions of ethyl bromofluoroacetate with aldehydes and ketones to generate
diastereomeric mixtures of R-fluoro-â-hydroxy esters, some of which can be separated by crystal-
lization or column flash chromatography. Diastereomerically pure R-fluoro-â-hydroxy acids are
obtained by mild alkaline hydrolysis of the resolved R-fluoro-â-hydroxy esters. Detailed NMR data
of new R-fluoro-â-hydroxy esters and R-fluoro-â-hydroxy acids are also presented.
Sch em e 1. Cla ssica l Refor m a tsk y Rea ction
In tr od u ction
The Reformatsky reactions of ethyl bromoacetate 1c
or its fluorinated analogues 1a and b with carbonyl
compounds, to generate â-hydroxyesters 4 or their fluori-
nated analogues 2 and 3 (Scheme 1), is a well-known
synthetic methodology, with widespread use in a number
of multistep syntheses. The version of the Reformatsky
that uses ethyl bromofluoroacetate 1a to obtain R-fluoro-
â-hydroxy esters 2 was first attempted 45 years ago1 with
zinc and benzene, toluene, or xylene as the solvent. At
about the same time, classical Aldol-type condensation
reactions of ethyl fluoroacetate with carbonyl compounds
were reported to give the same products, using the
sodium enolate of this ester, which was generated by
sodium hydride2 or sodium ethoxide.3 Poor yields (∼20%)
were generally obtained in these studies, but over the
last two decades such condensation reactions, using the
lithium enolate of the monofluoroester, have given better
results.4,5 This direct condensation process is reported to
give good yields of aldol-type products and also to provide
some significant diastereoselectivity, but it has the
serious disadvantage of using the very toxic ethyl fluoro-
acetate.6 Some improvements of the original Reformatsky
methodology of halomonofluoro esters have appeared in
the more recent literature. One of these involves the use
of ethyl chlorofluoroacetate with Mg and I2 in benzene,7
and recently, it was reported that the Reformatsky
reaction of aldehydes with ethyl chlorofluoroacetate in
the presence of lanthanide catalysts, using DMA as the
solvent,8 provides high isolated yields with aldehydes but
only 40% with ketones. Machleidt and Wessendorf re-
ported the use of bromofluoro ester 1a and Zn, in the
presence of I2, in a binary mixture of ether-THF,9
whereas Branda¨nge and co-workers were able to carry
out the condensation by refluxing the starting bromo-
fluoroester 1a and a ketone with a THF slurry of Zn to
give acceptable yields.10 Linderman and co-workers mim-
icked Hallinan and Fried’s difluoro-Reformatsky meth-
odology11 to synthesize a number of R-fluoro-â-hydroxy
esters.12
In almost all of the above cases, the reported Refor-
matsky reactions are reasonably efficient. However,
improvements in yields, shorter reaction times, and
procedure simplifications are possible, and some such
improvements are reported in this paper.
On the other hand, until the present work, no good
method had been reported for the hydrolysis of the
R-fluoro-â-hydroxy esters to their respective carboxylic
acids. Indeed, reports indicated that such esters are
largely destroyed when subjected to alkaline hydrolysis,
with very poor yields of acids (14%) being obtained.2
Apparently, use of aqueous NaOH, even at low concen-
trations, gives rise to destructive fluoride elimination. If
one protects the â-hydroxyl group, for example by ben-
† Universidad de Caldas.
‡ University of Florida.
§ University del Valle.
(1) McBee, E. T.; Pierce, O. R.; Christman, D. L. J . Am. Chem. Soc.
1955, 77, 1581.
(2) Bergmann, E. D.; Cohen, S. J . Chem. Soc. 1961, 3537.
(3) Martynov, V. F.; Titov, M. I. J . Gen. Chem. USSR (Eng. Transl.)
1962, 32, 716.
(4) Welch, J . T.; Seper, K.; Eswarakrishnan, S.; Samartino, J . J .
Org. Chem. 1984, 49, 4720.
(5) Pelter, A.; Kvicala, J . Tetrahedron Lett. 1997, 38, 4877.
(6) Welch, J . T.; Eswarakrishnan, S. J . Chem. Soc., Chem. Commun.
1985, 186.
(8) Shen, Y.; Qi, M. J . Chem. Res., Synop. 1993, 222.
(9) Machleidt, H.; Wessendorf, R. J ustus Liebigs Ann. Chem. 1964,
674, 1.
(10) Branda¨nge, S.; Dahlman, O.; Mo¨rch, L. J . Am. Chem. Soc. 1981,
103, 4452.
(7) Bergmann, E. D.; Cohen, S.; Hoffman, Rand-Meir J . Chem. Soc.
1961, 3457.
(11) Hallinan, E. A.; Fried, J . Tetrahedron Lett. 1984, 25, 2301.
(12) Linderman, R. J .; Graves, D. M. J . Org. Chem. 1989, 54, 661.
10.1021/jo015778x CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/14/2001

Synthesis of R-Fluoro-â-hydroxy Acids
J . Org. Chem., Vol. 67, No. 1, 2002 73
Sch em e 2. CeCl3-Ca ta lyzed Refor m a tsk y
Rea ction of Eth yl Br om oflu or oa ceta te
The procedure involves simply mixing the reactants
and stirring them vigorously at room temperature for
0.5-2 h under a blanket of dry N2. Completion of the
reaction is observed by a change of the typical color of
Zn to a brownish slurry. The reaction is very clean, with
few side products, and when starting with symmetrical
ketones a simple flash filtration of the crude product
through a short silica gel pad is sufficient to purify the
product. Yields of isolated products range from 60 to 95%,
and the results are summarized in Table 1.
zylation,7,13 the problems with hydrolysis disappear, but
protection and eventual deprotection will make any
synthesis longer. Acid-catalyzed hydrolysis of R-fluoro-
â-hydroxy esters is also a low-yield procedure, with
typical yields of 10-14%. Under these conditions, elimi-
nation competes with the desired hydrolysis,14 making
this approach also not useful for the synthesis of R-fluoro-
â-hydroxy acids. R-Fluoro-â-hydroxyacids have been pre-
viously prepared in 43-54% yield by a procedure of very
limited general applicability, via the diazotization reac-
tions of â-hydroxy-R-amino acids such as threonine or
serine using polyhydrogen fluoride/pyridine.15
In this paper, we report that alkaline hydrolysis can
be accomplished under very mild conditions and using
short reaction times with excellent yields being obtained
if water is not used as the solvent. In this way, we have
been able to synthesize a variety of R-fluoro-â-hydroxy
acids 5, some of them as pure racemic diastereomers.
We wish to present our experiences on the Reformatsky
reaction of ethyl bromofluroacetate 1a , catalyzed by
CeCl3, by using formally the same methodology which
Shen and Qi reported with ethyl bromodifluoroacetate
and the mild alkaline hydrolysis of R-fluoro-â-hydroxy
esters 2 to obtain R-fluoro-â-hydroxy acids 5. A detailed
information on NMR data of 2 and 5 is also presented.
As seen in Table 1, even relatively unreactive ketones,
such as benzophenone (entry 9) and pinacolone (entry 7),
react very efficiently and give products in acceptable
yield.
It has been reported that organocerium reagents retain
strong nucleophilicity but show a very reduced tendency
to effect deprotonation.19 This is probably the reason
CeCl3 enhances the outcome of the Reformatsky process.
In the absence of carbonyl compounds, the bromofluo-
roesters do not appear to react with the Zn, as evidenced
by 19F NMR analyses of the crude reaction mixtures.
However, addition of pinacolone to the mixture leads to
rapid reaction. It seems that the presence of carbonyl
compounds stimulates the formation of the organo-
metallic intermediate, creating an environment that
favors formation of condensation product in preference
to the reduction pathway. Although severe steric hin-
drance around the carbonyl group causes the reduction
product to increase somewhat, the condensation pathway
is still favored.
Ster eoch em istr y. As expected, unsymmetrical ke-
tones and aldehydes gave rise to the diastereoisomeric
erythro and threo mixtures, separable in some cases by
conventional flash column chromatography or by frac-
tional crystallization (Table 1). In all cases, both diaster-
eomers are readily distinguishable from each other by
their NMR spectra. When starting from aldehydes, the
correlation found by Elkik and Francesch20 works out
consistently in such way that the erythro isomer always
exhibits a 19F NMR signal at lower field compared to that
exhibited by the respective threo isomer. The threo
isomer is also recognized to have a typical coupling
constant J Hâ-F of 16-20 Hz, significantly larger than that
of the erythro isomer.
After the cerium chloride-catalyzed Reformatsky reac-
tion with acetophenone, it was possible to crystallize the
(RR/SS) diastereomer of ethyl 2-fluoro-3-hydroxy-3-phen-
ylbutanoate 2h (Table 1, entry 8) directly from the crude
product mixture giving 92.5% purity. Further purification
by recrystallization from hot hexanes could produce 95%
diastereomerically pure (RR/SS) 2h , the relative stereo-
chemistry of which was confirmed by X-ray crystal-
lographic analysis. In a similar fashion, both diastereo-
mers of ethyl 2-fluoro-3-hydroxy-3,4-diphenylbutanoate
(2k and 2l) (Table 1, entry 10) could be separated by flash
column chromatography, with the stereochemistry of 2k
also being confirmed by X-ray analysis.
No attempts were made to separate the diasteromeric
mixtures derived from aldehydes. However, because of
adequate signal resolution, spectroscopic NMR assign-
ments were able to be made. Thus, on the basis of 19F
NMR integration, it was possible to estimate that, within
experimental error, the ratio of isomers was in each case
Resu lts a n d Discu ssion
Recent years have seen the development of synthetic
procedures involving lanthanide metals, especially ce-
rium, in organometallic reactions. One of the most
important examples is the conversion of organollithium
compounds to organocerium derivatives by CeCl3.16 It is
well-known that this salt exerts a strong activation of
carbonyl compounds toward addition of organometallics.17
Inspired in this fact and by Shen and Qi’s use of cerium
methodology for the difluoro-Reformatsky reaction,9,18 we
carried out a systematic examination of the Reformatsky
reaction of ethyl bromofluoroacetate 1a with ketones and
aldehydes, in the presence of 4 molar % CeCl3‚7H2O
(Scheme 2).
(13) (a) Shiozaki, M.; Kobayashi, Y.; Arai, M. Tetrahedron 1991, 47,
7021. (b) Shiozaki, M.; Arai, M. J . Org. Chem. 1989, 54, 3754.
(14) Elkik, E.; Francesch, C. Bull. Soc. Chim. Fr. 1973, 4, 1277.
(15) (a) Olah, G. A.; Surya Prakash, G. K.; Chao, Y.-L. Helv. Chim.
Acta 1981, 64, 250, 2528. (b) Olah, G. A.; Welch, J . Synthesis 1974,
652.
(16) (a) Imamoto, T. In Comprenhensive Organic Synthesis; Trost,
B. M., Fleming, I., Schreiber, S. L., Eds.; Organocerium reagents;
Pergamon Press: Oxford, 1991; Vol. 1,
p 231. (b) Imamoto, T.
Lanthanides in Organic Synthesis; Academic Press: London, 1994; p
80. (c) Imamoto, T. Pure Appl. Chem. 1990, 62, 747. (d) Imamoto, T.;
Takiyama, N.; Nakamurra, K.; Hatajima, T.; Kamiya, Y. J . Am. Chem.
Soc. 1989, 111, 4392. (e) Imamoto, T.; Kusumoto, T.; Tawarayama,
Y.; Sugiura, Y.; Mita, T.; Hatanaka, Y.; Yokoyama, M. J . Org. Chem.
1987, 52, 281. (f) Imamoto, T.; Kusumoto, T.; Tawarayama, Y.; Mita,
T.; Hatanaka, Y.; Yokoyama, M. J . Org. Chem. 1984, 49, 3904. (g)
Imamoto, T.; Takiyama, N.; Nakamurra, K. Tetrahedron Lett. 1985,
26, 4763.
(17) Panev, S.; Dimitrov, V. Tetrahedron: Asymmetry 2000, 11, 1517.
(18) Shen, Y.; Qi, M. J . Fluorine Chem. 1994, 67, 229.
(19) J ohnson, C. R.; Tait, B. D. J . Org. Chem. 1987, 52, 281.
(20) Elkik, E.; Francesch, C. Bull. Soc. Chim. Fr. 1986, 3, 423.

74 J . Org. Chem., Vol. 67, No. 1, 2002
Ta ble 1. Ca ta lyzed Refor m a tsk y Rea ction s of Eth yl Br om oflu or oa ceta te
Ocampo et al.
reported
yield (%)
erythro/threo
ratio
entry
product
R1
R2
yielda (%)
From Aldehydes
n-C4H9
1
2
3
2a
2b
2c
H
H
H
83
90
92
48:52b
48:52b
50:50b
n- C5H11
Ph
77
65c
From Symmetrical Ketones
4
5
6
2d
2e
2f
Me
Et
Me
Et
89
90
94
14d
44e
CH2CH2CH2CH2CH2
reported
yield (%)
(RR/SS)/(RS/SR)
entry
product
R1
R2
yielda (%)
ratio
From Unsymmetrical Ketones
7
8
9
2g
t-Bu
Ph
Ph
Me
Me
Ph
Bn
Bn
61
90
85
92
86
59:41b
42:58f
2h and 2i
2j
29e
14e
10
11
2k and 2l
2m
Ph
Bn
42:58f
a
b
Yield of isolated products. Relative composition estimated by 19F NMR integration and the diastereomers were not separated.
c Reference 10. Reference 2. e Reference 1. f The diastereomeric mixture could be separated (95% pure) by crystallization and column
d
chromatography.
Sch em e 3. Mild Alk a lin e Hyd r olysis of
R-fluoro-â-hydroxy acids that proved useful for conversion
to the â-lactones), there is no reason to believe that the
esters derived from aldehydes will not undergo equally
efficient hydrolysis.
r-F lu or o-â-h yd r oxy Ester s 1
Ster eoch em istr y. Typically, alkaline hydrolysis of
esters occurs by nucleophilic addition to the carbonyl
group without disturbing an R-chiral center, unless
especially acidic protons are present in the R-position. It
is known that fluorine atoms decrease acidity of R-hy-
drogens, because of the resultant Iπ repulsion in such
planar carbanions.23 R-Fluoro esters should therefore be
less acidic than their nonfluorinated counterparts, and
they should thus be hydrolyzed by the normal BAC
2
1:1. Such lack of significant diastereoselectivity would
certainly constitute a disadvantage of this method, if
stereocontrol is needed. A somewhat higher diastereose-
lectivity is reported for the condensations of ethyl fluo-
roacetate enolate (promoted with HMDS)21 but, as men-
tioned before, such starting material is a very toxic
compound.
pathway. As such, it would be expected that the alkaline
hydrolysis of R-fluoro-â-hydroxy esters 2 should occur
with retention of the configuration at the R-chiral centers.
Consistent with this assumption, 19F NMR analyses of
the crude reaction mixtures revealed a relative composi-
tion of diastereomers nearly identical to that of the
starting materials. For example, a crude product with a
composition of diastereomers 95:5 was obtained when
starting from a mixture 95:5 of (RR/SS)/(RS/SR) of 2h ;
the logical conclusion is that the 95% isomeric acid is also
the RR/SS diasteromer. There is also a logical resem-
blance of the NMR data of the R-fluoro-â-hydroxy acids
5 to those of their starting R-fluoro-â-hydroxy esters 2,
particularly the chemical shift of the fluorine atom, the
signal of the proton in the R-position, the signals of
protons or carbons of the methyl or methylene groups
attached to the â-carbon and the 13C NMR signals of the
â-carbon and of the carbonyl group (Table 2).
Additionally, as it can be seen in Table 2, the NMR
data of the diasteromeric R-fluoro-â-hydroxy acids are
consistent with the earlier reported empirical correlations
reported by Welch for R-fluoro-â-hydroxy esters, that is
that the 13C NMR signal of the aliphatic carbon attached
to the â-position of the (RR/SS) diastereomer usually
appears somewhat upfield with respect to the analogous
signal of the (RS/SR) diastereomer. Therefore, our as-
signment of the relative configuration of the acids is
Mild Alk a lin e Hyd r olysis of r-F lu or o-â-h yd r oxy
Ester s, 2. Use of dilute aqueous sodium hydroxide led
to no success, even when at low concentration. Using
“anhydrous hydroxide” (generated using potassium tert-
butoxide in dry ether)22 was equally unsuccessful, ap-
parently because the strong basicity of the medium leads
to decomposition of the R-fluoro-â-hydroxy esters.
Use of a cold solution of sodium hydroxide in absolute
ethanol in concentrations of ∼0.2 M provided a very
effective condition for the hydrolysis, resulting in the
synthesis of R-fluoro-â-hydroxy acids, 5, in yields >90%
(Scheme 3). In a typical experiment, the starting R-fluoro-
â-hydroxy ester is dissolved in absolute ethanol to a
concentration of ∼0.2 M and then cooled. A cold solution
of 0.2 M sodium hydroxide in absolute ethanol was then
added dropwise and the mixture stirred for 1-2 h.
Purification of the products can be accomplished by
crystallization from a binary mixture of benzene/hexanes.
Although the hydrolyses reported here are only of those
esters prepared from ketones (because these were the only
(21) Luo, L. Doctoral Dissertation, Department of Chemistry,
University of Florida, 1998, p 96.
(22) Gassman, P. G.; Schenk, W. N. J . Org. Chem. 1977, 52, 918.
(23) Smart, B. E. In Organofluorine Chemistry, Principles and
Applications; Banks, R. E., Smart, B. E., Tatlow, J . C., Eds.; Plenum
Press: New York, 1994; Chapter 3, p 76 and references therein.

Synthesis of R-Fluoro-â-hydroxy Acids
J . Org. Chem., Vol. 67, No. 1, 2002 75
Ta ble 2. Com p a r a tive NMR Da ta of r-F lu or o-â-h yd r oxy Ester s 2 a n d r-F lu or o-â-h yd r oxy Acid s 5: Rep r esen ta tive
Sign a ls
1H NMR
Hγ or Hγ′
1.70
2.4
1.74
2.4
1.67
2.4
1.70
2.3
13C NMR
entry
1
substituents
compd
HR 19F NMR
4.92 -195.12 25.59
47.7 46.9 2.6
4.98 -193.56 25.32
47.4 47.4 2.7
5.00 -192.53 26.19
47.7 46.9 4.0
5.08 -191.32 26.34c
47.6 47.4 3.9
Cγ or Cγ′
Câ
CR
93.03 168.22
193.8 24.7
92.91 171.68
194.8 25.1
93.59 168.66
197.8 23.7
-COO-
R1 ) Ph; R2 ) Me (R*,R*)-2h X ) Et δ (ppm)a
74.94
20.7
75.05
20.8
75.10
19.7
J (Hz)b
2
3
R1 ) Ph; R2 ) Me (R*,R*)-5h X ) H δ (ppm)a
J (Hz)b
R1 ) Me; R2 ) Ph (R*,S*)-2i X ) Et δ (ppm)a
J (Hz)b
4
R1 ) Me; R2 ) Ph (R*,S*)-5i X ) H δ (ppm)a
74.44c 93.89c 169.10c
J (Hz)b
20.6
78.64
20.1
78.44
21.7
77.45
19.7
77.62
19.8
76.73
18.6
77.02
19.5
188.7
91.98 168.83
198.9 23.6
91.96 168.73
188.9 25.1
91.84 167.84
192.9 24.7
92.06 171.36
193.5 25.3
90.88 169.45
202.0 22.7
90.61 173.57
199.51 23.9
88.51 169.87
196.4 22.1
88.46 173.41
194.9 23.7
25.1
5
R1 ) R2 ) Ph
R1 ) R2 ) Ph
2j
5j
X ) Et δ (ppm)a
J (Hz)b
5.48 -189.58
47.1
46.9
6
X ) H δ (ppm)c
J (Hz)b
6.10 -189.95
46.2
44.9
7
R1 ) Ph; R2 ) Bn (R*,R*)-2k X ) Et δ (ppm)a
3.32, 3.38
1.5, 1.8
3.34, 3.39
1
3.13, 3.34
2.7 ea
3.13, 3.34
2.7, 2.2
5.12 -197.12 44.19
47.7 49.1 3.5
5.15 -195.94 44.06
47.7 47.4 4.2
4.88 -191.16 45.26
47.1 47.1 2.6
4.92 -190.04 45.28
46.5 47.1
J (Hz)b
8
R1 ) Ph; R2 ) Bn (R*,R*)-5k X ) H δ (ppm)a
J (Hz)b
R1 ) Bn; R2 ) Ph (R*,S*)-2l X ) Et δ (ppm)a
J (Hz)b
9
10
11
12
R1 ) Bn; R2 ) Ph (R*,S*)-5l X ) H δ (ppm)a
J (Hz)b
R1 ) R2 ) Bn
R1 ) R2 ) Bn
2m
5m
X ) Et δ (ppm)a
J (Hz)b
2.74, 2.94, 2.97
2.3, 3.3, 2.1
2.70, 2.94
4.52 -195.79 42.38, 42.75 75.06
45.6
4.57 -194.71 41.98
45.9 46.9
44.9
20.6
75.22
20.6
X ) H δ (ppm)a
J (Hz)b
2, 3
a
b
In CDCl3; reference: TMS for 1H; CFCl3 for 19F NMR; CDCl3 at 77.00 ppm for 13C NMR. J HF values in 1H and 19F NMR; or J CF
values in 13C NMR. c In DMSO-d6.
dependent upon our assumption of retention of configu-
ration during hydrolysis of the esters whose relative
configurations were determined by X-ray analysis, also
by extending the application of the Welch’s configura-
tional correlations.4
used in our syntheses of R-fluoro-â-lactones and dias-
teromerically pure monofluoroalkenes, work that is un-
derway.
Exp er im en ta l Section
Pure diastereomers of R-fluoro-â-hydroxy acids 5h ,i,k ,l
could be obtained by conventional crystallization using
binary mixtures of benzene/hexanes.
Rea gen ts. Ethyl bromofluoroacetate of 98% purity was
purchased from SynQuest and used without further purifica-
tion. Aldehydes and ketones were purchased from Aldrich
(purities ranging from 97% to 99.9%) and used without any
additional treatment. THF from Fisher was dried with Na and
K (with benzophenone as indicator). CeCl3‚7H2O stored under
argon was purchased from Aldrich; after each use, it was
purged with argon.
In str u m en ts a n d Ma ter ia ls. NMR spectra were recorded
on a VXR300 instrument or a Gemini300 or a Mercury 300 at
299.95 MHz for 1H, 75.42 MHz for 13C, and 282.23 MHz for
19F NMR spectra and in CDCl3 as solvent, unless otherwise
mentioned. TMS, CDCl3, and CFCl3 were used as internal
standards for 1H, 13C, and 19F NMR spectra, respectively.
Melting points were measured in a Thomas-Hoover capillary
apparatus and are not corrected. Distillation of starting
materials and flame drying the reaction apparatus under a
blanket of N2 or argon are strongly encouraged to obtain better
yields and shorter reaction times.
Syn t h esis of r-F lu or o-â-h yd r oxy E st er s 2. Gen er a l
P r oced u r e. A round-bottomed flask was dried, purged with
argon, and charged with 0.60 g of freshly etched zinc, 48 mg
of CeCl3‚7H2O, 5 mmol of the respective carbonyl starting
material, and 8 mL of dry THF and capped with a septum.
The slurry was stirred vigorously, and 7.5 mmol of ethyl
bromofluoroacetate was added dropwise. The mixture was
allowed to stir until the gray color of zinc was changed to a
brownish slurry; typically, it takes 2-6 h, and depending on
the use of some excess of zinc the change of color is not fully
evident. Then 10 mL of ethyl acetate was added to quench the
Con clu sion
The Reformatsky reaction of ethyl bromofluoroacetate
in the presence of CeCl3‚7H2O is the most efficient
procedure thus far reported for preparing R-fluoro-â-
hydroxy esters from aldehydes or ketones. With a 4%
molar concentration of CeCl3‚7H2O, isolated yields ranged
from 60 to 95%. The reaction procedure is very simple,
and fast flash column chromatography can be readily
used to obtain pure products. As expected, aldehydes and
unsymmetrical ketones give rise to mixtures of erythro
and threo diastereomers24 with low diastereoselectivity.
However, in some cases the mixtures can be separated
by conventional chromatography. Mild alkaline hydroly-
sis with a cold 0.2 M ethanolic solutions of sodium
hydroxide gives rise to R-fluoro-â-hydroxy acids 5 in
excellent yields. In this manner, the synthesis of R-fluoro-
â-hydroxy acids 5 has become a straightforward process.
Pure acid diastereomers were able to be obtained and
(24) The terms erythro and threo make sense if aldehydes are the
starting materials; for those products bearing two R groups in the â
position, Welch used the terms anti and syn (ref 4).

76 J . Org. Chem., Vol. 67, No. 1, 2002
Ocampo et al.
ppm, m, 10H; 2.50 ppm, broad singlet, 1H; 4.25 ppm, q, J HH
reaction, followed by addition of 10 mL of a saturated aqueous
solution of NH4Cl and 10 mL of brine. The mixture was filtered
by suction over a bed of silica, and the residue was washed
with ethyl acetate. The filtrate was transferred to a separatory
funnel, and the organic layer was separated, the aqueous
phase was extracted with ethyl acetate (3 × 20 mL), and the
organic phases were combined and dried with magnesium
sulfate, filtered over a bed of silica, and evaporated. The
products were then further purified by flash column chroma-
tography, eluting with some appropriate solvent carefully
chosen by TLC. For an improved resolution of erythro/threo
mixtures, chromatography with a bigger column may be
needed.
)
7.2 Hz, 2H; 4.63 ppm, d, J HR-F ) 48.3 Hz, 1H. 19F NMR: δ
-199.80 ppm, d, J HR-F ) 46.9 Hz. 13C NMR: δ 14.22 ppm;
21.23 ppm; 21.16 ppm; 25.48 ppm; 32.69 ppm, d, J CF ) 3.0
Hz; 32.82 ppm, d, J CF ) 3.5 Hz; 61.75 ppm; 72.43 ppm, d, J CF
) 19.7 Hz; 93.54 ppm, d, J CF ) 190.8 Hz; 168.51 ppm, d, J CF
) 24.2 Hz.
Eth yl 2-F lu or o-3-h yd r oxy-3,4,4-tr im eth ylp en ta n oa te,
2g. Sta r tin g fr om 3,3-Dim eth yl-2-bu ta n on e (P in a colon e).
The product was purified by flash column chromatography
eluting with ethyl acetate, giving rise to 61% yield of yellow
oil identified as a diastereomeric (RR/SS)/(RS/SR) mixture, in
a 41:59 ratio. Spectral properties: 1H NMR: δ 1.01 ppm, two
doublets overlapped, J HF ) 1.2 Hz, 9H; 1.19 ppm, d, J HF ) 2.1
Hz overlapped with 1.22 ppm, d, J HF ) 2.4 Hz, 3H; 1.32 ppm,
t J HH ) 7.2 Hz, overlapped with 1.33 ppm J HH ) 7.2 Hz, 2H;
4.12 ppm to 4.42 ppm, a cluster of signal which includes the
CH2 portion of the ethyl group of both isomers and the
respective OH groups; 3H total; 4.94 ppm, d, J HR-F ) 48 Hz,
42% of 2H (probably RR/SS isomer); 4.99 ppm, d, J HR-F ) 48
Hz, 58% of 2H (probably RS/SR isomer). 19F NMR: δ -190.58
ppm, d, J HR-F ) 49.1 Hz, (59%, probably RS/SR isomer);
-192.22 ppm, d, J HR-F ) 49.1 Hz, (41%, probably RR/SS
isomer). 13C NMR: major isomer: δ 14.27 ppm, 18.12 ppm and
19.80 ppm (three methyl groups from the tert-butyl portion);
25.60 ppm (methyl group from the ethyl ester portion); 26.09
ppm (methyl group adjacent to the carbinol portion); 38.24 ppm
(quaternary carbon from the tert-butyl portion); 62.13 ppm
(methylene group from the ethyl ester portion); 76.86 ppm, d,
J CF ) 19.6 Hz, (quaternary carbon from the carbinol portion);
Eth yl 2-Flu or o-3-h ydr oxyh eptan oate, 2a. Star tin g fr om
Va ler a ld eh yd e. The diastereomeric product mixture (48:52)
was purified by flash column chromatography, eluted with a
binary mixture 85:15 hexanes/ethyl acetate. The stereoisomers
were not separated by the column. Yield: 83%. Spectral
properties: 1H NMR of the diastereomeric mixture: δ 0.89 and
0.90 ppm, two doublets, J HH ) 7.2 Hz, 3H; 1.30 ppm, t, J HH
)
7.2 Hz, 3H; 1.36 and 1.60 ppm, m, 6H; 2.27 ppm, d, J HF ) 8.7
Hz (52% of 1H, threo isomer); 2.58 ppm, d, J HF ) 6.0 Hz (48%
of 1H, erythro isomer); 3.96 ppm, m, overlapped with 4.02 ppm,
m, 1H; 4.27 ppm, q, J HH ) 7.2 Hz, overlapped with 4.28 ppm,
q, J HH ) 7.2 Hz, 2H; 4.81 ppm, dd, J HR-F ) 48.4 Hz, J HH ) 2.5
Hz, 52% de 1H; 4.85 ppm, dd, J HR-F ) 48.4 Hz; J HH ) 3.7 Hz,
48% de 1H. 19F NMR: δ -200.11 ppm, dd, J HR-F ) 49.1 Hz,
J HF ) 17.2 Hz (erythro isomer, 47.7% of the mixture); -208.73
ppm, dd, J HR-F ) 49.1 Hz, J Hâ-F ) 24.4 Hz (threo isomer, 52.3%
of the mixture).
Eth yl 2-F lu or o-3-h yd r oxyocta n oa te, 2b. Sta r tin g fr om
Hexa n a l. A diastereomeric product mixture (48:52) was
obtained after aqueous workup and solvent evaporation. Yield
of crude product mixture: 90%. Spectral properties: 19F
NMR: δ -199.8 ppm, dd, J HR-F ) 49.1 Hz, J Hâ-F ) 18 Hz (48%
of the mixture); -207.6 ppm, dd, J HR-F ) 49 Hz, J Hâ-F ) 24
Hz (52% of the mixture), signals which match those reported
previously.13
E t h yl 2-F lu or o-3-h yd r oxy-3-p h en ylp r op a n oa t e, 2c.
Sta r tin g fr om Ben za ld eh yd e. A diastereomeric product
mixture (50:50) was obtained after aqueous workup and
solvent evaporation. Yield of crude product mixture: 92%.
Spectral properties: 19F NMR: δ -199.0 ppm, dd, J HR-F ) 49
Hz, J Hâ-F ) 18 Hz (49.8% of the mixture, erythro isomer);
-204.4 ppm, dd, J Hâ-F ) 49 Hz, J Hâ-F ) 24 Hz (50.2% of the
mixture, threo isomer), signals which match those reported
previously.13,21
91.12 ppm, d, J CF ) 197.4 Hz (CH-F); 170.49 ppm, d, J CF )
23.7 Hz. Minor isomer: 14.20 ppm, 18.14 ppm and 19,83 ppm
(three methyl groups from the tert-butyl portion); 25.64 ppm
(methyl group from the ethyl ester portion); 26.12 ppm (methyl
group adjacent to the carbinol portion); 37.94 ppm (quaternary
carbon from the tert-butyl portion); 61.85 ppm (methylene
group from the ethyl ester portion); 76.01 ppm, d, J CF ) 19.6
Hz, (quaternary carbon from the carbinol portion); 92.39 ppm,
d, J CF ) 188.3 Hz (CH-F); 168.87 ppm, d, J CF ) 25.2 Hz.
E t h yl 2-F lu or o-3-h yd r oxy-3-p h en ylb u t a n oa t e: Iso-
m er s (RR/SS)-2h a n d (RS/SR)-2i. Sta r tin g fr om Aceto-
p h en on e. The (RR/SS) isomer was isolated as crystals (mp
76.5-78.0 °C) directly from the crude mixture, contaminated
with 5% of the (RS/SR) isomer. The remaining oil was resolved
by flash column chromatography, giving rise to one crystalline
isomer, identical to the first one; the other isomer remained
as an oil. Combined yield: 90%, in a ratio 58:42. Further
purification of the (RR/SS) isomer may be accomplished by
fractional crystallization with hexane. Spectral properties of
2h , [(RR/SS)-isomer]: 1H NMR: δ 1.07 ppm, t J HH ) 7.2 Hz;
3H; 1.70 ppm, d, J HF ) 2.4 Hz, 3H; 3.23 ppm, broad singlet,
1H; 4.10 ppm, q J HH ) 7.2 Hz, 2H; 4.92 ppm, d, J HR-F ) 47.7
Hz, 1H; 7.25 ppm to 7.50 ppm, m, 10H. 19F NMR: δ -195.12
ppm, d, J HR-F ) 46.9 Hz. 13C NMR: δ 13.95 ppm; 25.59 ppm,
d, J CF ) 2.6 Hz; 61.85 ppm; 74.94 ppm, d, J CF ) 20.7 Hz; 93.03
ppm, d, J CF ) 193.8 Hz; 125.57 ppm; 127.92 ppm; 128.42 ppm;
142.26 ppm, d, J CF ) 3.0 Hz; 168.22 ppm, d, J CF ) 24.7 Hz.
The remainder oil partially resolved by flash column chroma-
tography, giving rise to another crop of the crystalline isomer
(76% pure, contaminated with the second isomer) and an oil
(the second isomer, 87% pure, contaminated with the first
isomer). Spectral properties of 2i, [(RS/SR) isomer]: 1H
NMR: δ 1.06 ppm, t J HH ) 7.2 Hz; 3H; 1.67 ppm, d, J HF ) 2.4
Eth yl 2-Flu or o-3-h ydr oxy-3-m eth ylbu tan oate, 2d. Star t-
in g fr om Aceton e. The product was flash eluted with ethyl
acetate, giving rise to 89% yield of an oil. Spectral properties:
1H NMR: δ 1.28 ppm, s, overlapped with 1.31 ppm, t, J HH
7.2 Hz, 9H; 2.82 ppm, broad singlet, 1H; 4.28 ppm, q, J HH
)
)
7.2 Hz, 2H; 4.64 ppm, d, J HR-F ) 48.0 Hz, 1H. 19F NMR: δ
-196.06 ppm, d, J HR-F) 49.1 Hz. 13C NMR: δ 14.24 ppm; 25.07
ppm, d, J CF ) 8.6 Hz, overlapped with 25.12 ppm, d J CF ) 8.1
Hz; 61.91 ppm; 71.55 ppm, d, J CF ) 21.2 Hz; 93.71 ppm, d,
J CF ) 191.4 Hz; 168.58 ppm, d, J CF ) 23.7 Hz.
Eth yl 3-Eth yl-2-flu or o-3-h yd r oxyp en ta n oa te, 2e. Sta r t-
in g fr om 3-P en ta n on e. The product was flash eluted with
ethyl acetate, giving rise to 90% yield of an oil. Spectral
properties: 1H NMR: δ 0.94 ppm, t, J HH ) 7.5 Hz, overlapped
with 0.96 ppm, t, J HH ) 7.5 Hz, 6H; 1.34 ppm, t, J HH ) 7.2
Hz, 3H; 1.62 ppm, m, J HH ) 7.5 Hz, overlapped with 1.64 ppm,
m, J HH ) 7.5 Hz, 4H total; 2.62 ppm, broad singlet, 1H; 4.31
ppm, q, J HH ) 7.2 Hz, 2H; 4.83 ppm, d, J HR-F ) 47.7 Hz, 1H.
19F NMR: δ -199.66 ppm, d, J HR-F ) 49.1 Hz. 13C NMR: δ
7.22 ppm; 7.50 ppm; 14.16 ppm; 26.92 ppm, d, J CF ) 3.0 Hz;
27.97 ppm, d, J CF ) 3.0 Hz; 61.81 ppm; 75.37 ppm, d, J CF
19.2 Hz; 91.52 ppm, d, J CF ) 191.9 Hz; 169.13 ppm, d, J CF
23.7 Hz.
E t h yl 2-F lu or o-2-(1′-h yd r oxycycloh exyl)a cet a t e, 2f.
Sta r tin g fr om Cycloh exa n on e. The product was purified
by flash column chromatography, eluting with 85:15 hexanes/
ethyl acetate to give a 94% yield of a yellow oil. Spectral
properties: 1H NMR: δ 1.29 ppm, t J HH ) 7.2 Hz, 3H; 1.54
Hz, 3H; 3.57 ppm, d, J HF ) 0.6 Hz, 1H; 4.09 ppm, dq, J HH
)
7.2 Hz, J HF ) 3.6 Hz, 2H; 5.00 ppm, d, J HR-F ) 47.7 Hz, 1H;
7.24 ppm to 7.36 ppm, and 7.44 to 7.50 ppm, 10H. 19F NMR:
-192.53 ppm, d, J HR-F ) 46.9 Hz. 13C NMR: 13.92 ppm; 26.19
ppm, d, J CF ) 4.0 Hz; 61.94 ppm; 75.10 ppm, d, J CF ) 19.7
Hz; 93.59 ppm, d, J CF ) 197.8 Hz; 125.22 ppm, d, J CF ) 2.5
Hz; 127.83 ppm; 128.38 ppm; 142.88 ppm; 168.66 ppm, d, J CF
) 23.7 Hz.
)
)
Eth yl 2-F lu or o-3-h yd r oxy-3,3-d ip h en ylp r op a n oa te, 2j.
Sta r tin g fr om Ben zop h en on e. The product was purified by
flash column chromatography, eluting with a binary mixture
of 90:10 hexanes/ethyl acetate, giving rise to 85% yield of a

Synthesis of R-Fluoro-â-hydroxy Acids
J . Org. Chem., Vol. 67, No. 1, 2002 77
white solid (mp 112-113 °C). Spectral properties: 1H NMR:
δ 1.01 ppm, t J HH ) 7.2 Hz, 3H; 4.07 ppm, m J HH ) 7.2 Hz,
overlapped with a broad singlet, 4H; 5.48 ppm, d, J HR-F ) 47.1
Hz; aromatic cluster, 7.20 ppm to 7.84 ppm, 10H. 19F NMR: δ
-189.58 ppm, d, J HR-F ) 46.9 Hz. 13C NMR: δ -13.84 ppm;
materials. The aqueous phase was cooled, and a cold 1 M HCl
aqueous solution was added dropwise until pH 3 (checked with
litmus paper). The resultant cloudy mixture was extracted
with ether three times, each one with a volume of ether equal
to the volume of aqueous extract. The organic extract was dried
with MgSO4, filtered, and evaporated, and the resulting
R-fluoro-â-hydroxy acids were crystallized in an appropriate
solvent.
62.11 ppm; 78.64 ppm, d, J CF ) 20.1 Hz; 91.98 ppm, d, J CF
)
198.9 Hz; aromatic signals from 126 to 142 ppm; 168.83 ppm,
d, J CF ) 23.6 Hz.
Eth yl 2-F lu or o-3-h yd r oxy-3,4-d ip h en ylbu ta n oa te: Iso-
m er s (RR/SS)-2k a n d (RS/SR)-2l. Sta r tin g fr om 1,2-
Dip h en yleta n on e (Deoxyben zoin ). The stereoisomeric mix-
ture was resolved by flash column chromatography eluting
with a binary mixture 85:15 hexanes/ethyl acetate, giving rise
to 47.8% yield of a white solid (mp 99.5-101.0 °C) and 44.2%
yield of an oil, which crystallizes later upon standing, mp 49-
51 °C. Combined yield: 92%. A 5-g scale gave rise to similar
results; a crop of the more polar isomer was obtained by
crystallization from ethanol; the mother liquors (containing a
75:25 mixture of the less polar/more polar isomers), with
solvent evaporated off and resolved by flash column chroma-
tography, finally afforded isolated diastereomeric racemic
mixtures. Spectral properties of 2k , [(RR/SS) isomer]: 1H
NMR: δ 0.97 ppm, t J HH ) 7.2 Hz; 3H; 3.23 ppm, s, 1H; 3.32
Syn th esis of (RR/SS)-2-F lu or o-3-h yd r oxy-3-p h en ylbu -
tyr ic Acid , 5h . Starting from 1.120 g (5.0 mmol) of ethyl (RR/
SS)-2-fluoro-3-hydroxy-3-phenylbutyrate, 95% diastereomer-
ically pure dissolved in 25.0 mL of ethanol, and 0.225 g of 98%
NaOH (5.0 mmol) dissolved in 25 mL of ethanol. The product
was crystallized from a binary mixture 20:80 benzene/hexanes.
0.883 g of a white solid was formed (90% yield of isolated
product); mp 100-106 °C dec. 19F NMR analysis of the
composition of the product demonstrated the presence of 4.6%
of the RS/SR isomer (δ -189.39 ppm, d, J HR-F ) 49.1 Hz) and
95.4% of the RR/SS isomer (δ -191.81 ppm, d, J HR-F ) Hz).
3
1H NMR (DMSO-d6): δ 7.48 ppm, d, J HH ) 7.2 Hz, 2H; 7.31
3
3
ppm, t, J HH ) 7.2 Hz, overlapped with 7.23 ppm, t, J HH
)
7.2 Hz, 3H; 5.01 ppm, d, J HR-F ) 48.0 Hz, 1H; 1.55 ppm, d,
4J HF ) 1.8 Hz, 3H. Recrystallization from the same binary
solvent system, under more diluted conditions and collecting
an early crop, gave rise to 0.50 g of 100% diastereomerically
pure RR/SS isomer. Spectral properties: 1H NMR: δ 7.46 to
7.51 ppm, m, 2H; 7.42 to 7.29 ppm, m, 3H; 4.98 ppm, d, J HR-F
ppm, dd, J gem
) 13.8 Hz; J HF ) 1.5 Hz, 1H; 3.38 ppm, dd,
HH
J gem ) 13.8 Hz; J HF ) 1.8 Hz, 1H; 3.98 ppm, qd, J HH ) 7.2
HH
Hz; J HF ) 1.8 Hz, 2H; 5.12 ppm, d, J HR-F ) 47.7 Hz, 1H;
aromatic clusters at 6.9 ppm to 7.4 ppm, 10H. 19F NMR: δ
-197.12 ppm, d, J HR-F ) 49.1 Hz. Contaminated with 8.1% of
the RS/SR isomer. 13C NMR: δ 13.78 ppm; 44.19 ppm, d, J CF
) 3.5 Hz; 61.74 ppm; 77.45 ppm, d, J CF ) 19.7 Hz; 91.84 ppm,
4
) 47.4 Hz, 1H; 1.74 ppm, d, J HF ) 2.4 Hz, 3H. 19F NMR: δ
-193.56 ppm, d, J HR-F ) 47.4 Hz. 13C NMR: δ 25.32 ppm, d,
J CF ) 2.7 Hz; 75.05 ppm, d, J CF ) 20.8 Hz; 92.91 ppm, d, J CF
) 194.76 Hz; 125.54 ppm; 128.26 ppm; 128.62 ppm; 141.75
ppm; 171.68 ppm, d, J CF ) 25.1 Hz. Anal. Calcd: C, 60.602;
H, 5.594; F, 9.586. Found: C, 60.438; H, 5.631.
d, J gem ) 192.9 Hz; 126.15 ppm; 127.01 ppm; 127.83 ppm;
CF
128.12 ppm; 128.15 ppm; 131.04 ppm; 135.01 ppm; 140.38
ppm, d, J CF ) 3.5 Hz; 167.84 ppm, d, J CF ) 24.7 Hz. Spectral
properties of 2l, [(RS/SR) isomer]: 1H NMR: δ 0.98 ppm, t
Syn th esis of (RS/SR)-2-F lu or o-3-h yd r oxy-3-p h en ylbu -
tyr ic Acid , 5i. Starting from 3.69 g (15.6 mmol) of 95% pure
ethyl 2-fluoro-3-hydroxy-3-phenylbutyrate (a mixture 68% RS/
SR and 32% RR/SS) dissolved in 80 mL of ethanol and 0.76 g
of 98% NaOH (18.0 mmol) dissolved in 80 mL of ethanol, 2.75
g of a white product was obtained (89% yield). This product
was recrystallized from a binary mixture 40:60 benzene:
hexanes giving rise to a product 99.7% diastereomerically pure
J HH ) 7.2 Hz; 3H; 3.13 ppm, dd, J gem ) 14.1 Hz; J HF ) 2.7
HH
Hz, 1H; 3.34 ppm, dd, J gem ) 14.1 Hz; J HF ) 2.70 Hz, 1H;
HH
3.66 ppm, broad singlet, 1H; 4.02 ppm, m, J HH ) 7.2 and 10.2
Hz, 2H; 4.88 ppm, d, J HR-F ) 47.1 Hz, 1H; aromatic clusters
at 7.07 ppm to 7.60 ppm, 10H. 19F NMR: δ -191.16 ppm, d,
J HR-F ) 47.1 Hz, contaminated with traces of the RR/SS
isomer. 13C NMR: δ 13.85 ppm; 45.26 ppm, d, J CF ) 2.6 Hz;
62.01 ppm; 76.73 ppm, d, J CF ) 18.6 Hz; 90.88 ppm, d, J gem
RS/SR (19F NMR, (DMSO-d6) δ ) -189.46 ppm, d, J HR-F
)
CF
) 202.0 Hz; 125.55 ppm; 125.60 ppm; 126.94 ppm; 128.11 ppm;
128.25 ppm; 133.29 ppm; 135.55 ppm, 142.20 ppm, 169.45
ppm, d, J CF ) 22.7 Hz.
47.7 Hz) and 0.3% RR/SS isomer (δ ) -191.89 ppm, d, J HR-F
) 49.6 Hz); mp 114-118 °C dec. Spectral properties: 1H
NMR: δ 7.47 ppm, d, J HH ) 7.2 Hz, 2H; 7.27 to 7.40 ppm, m,
3H; 5.08 ppm, d, J HR-F ) 47.6 Hz, 1H; 4 to 6 ppm, br. s, 2 H;
Eth yl 3-Ben zyl-2-flu or o-3-h yd r oxy-4-p h en ylbu ta n oa te,
2m . Sta r tin g fr om 1,3-Dip h en yla ceton e. The product was
purified by flash column chromatography, eluting with 60:40
CH2Cl2/hexanes, giving rise to 86% yield of a yellow solid (mp
70.5-71.5 °C). A 10-g scale reaction gave rise to 90% yield of
crude product, which was recrystallized from hot ethanol,
giving rise to a total 80% isolated yield of pure product.
Spectral properties: 1H NMR: δ 1.24 ppm, t, J HH ) 7.2 Hz,
4
1
1.70 ppm, d, J HF ) 2.3 Hz, 3H. H NMR (DMSO-d6): δ 7.48
ppm, d, J HH ) 7.3 Hz, 2H; 7.31 ppm, t, J HH ) 7.1 Hz, 2H; 7.22
ppm, tt, J HH ) 7.2, 2.1 Hz, 1H; 5.08 ppm, d, J HR-F ) 47.6 Hz,
4
1H; 4.1 ppm, br s; 1.55 ppm, d, J HF ) 1.5 Hz, 3H. 19F NMR:
δ -191.316 ppm only peak, d, J HR-F ) 47.4 Hz. 13C NMR
(DMSO-d6): δ 26.34 ppm, d, J CF ) 3.9 Hz; 74.44 ppm, d, J CF
) 20.6 Hz; 93.89 ppm, d, J CF ) 188.7 Hz; 125.75 ppm; 126.90
ppm; 127.89 ppm; 144.99 ppm, d, J CF ) 3.0 Hz; 169.10 ppm,
d, J CF ) 25.1 Hz. Anal. Clacd: C, 60.602; H, 5.594; F, 9.586.
Found: C, 60.991; H, 5.409.
3H; 2.74 ppm, dd, J gem
) 13.9 Hz, J HF ) 2.3 Hz, 1H; 2.94
HH
ppm, dd, J gem ) 14.1 Hz, J HF ) 3.3 Hz, 1H; 2.97 ppm, d, J HF
HH
) 2.1 Hz, 2H; 3.27 ppm, s, 1H; 4.05 ppm, q, J HH ) 7.2 Hz, 2H;
4.52 ppm, d, J HR-F ) 45.6 Hz; 7.12 ppm to 7.36 ppm, m, 10H.
19F NMR: δ -195.79 ppm, d, J HR-F ) 44.9 Hz. 13C NMR: δ
14.13 ppm; 42.38 ppm; 42.75 ppm; 62.21 ppm; 75.06 ppm, d,
J CF ) 20.6 Hz; 88.51 ppm, d, J CF ) 196.4 Hz; 127.00 ppm,
127.04 ppm, 128.33 ppm, 128.36 ppm, 131.18 ppm, 131.24,
135.77, and 135.93 ppm (aromatic signals); 169.87 ppm, d, J CF
) 22.1 Hz.
Syn th esis of r-F lu or o-â-h yd r oxya cid s 5. Gen er a l P r o-
ced u r e. The starting ethyl R-fluoro-â-hydroxy ester was mixed
with ethanol (to an approximate concentration of 0.2 M) in a
round-bottomed flask and cooled in an ice-water bath. Then,
cold NaOH, dissolved in ethanol (approximate concentration
0.2 M), was added dropwise to the starting material solution
with vigorous agitation and cooling. The reactant mixture was
stirred until room temperature was reached (it took about 1-2
h in each case). Completion of the reaction was monitored by
19F NMR analysis. The solvent was evaporated, and the
organic salt was formed as a white solid, which was dissolved
in water and washed with ether to remove nonreacted organic
Syn th esis of 2-F lu or o-3-h yd r oxy-3,3-d ip h en ylp r op i-
on ic Acid 5j. Starting from 0.580 g (2.0 mmol) of ethyl
2-fluoro-3-hydroxy-3,3-diphenylpropionate in 25 mL of ethanol
and 100 mg (2.4 mmol) of 98% NaOH in 10 mL of ethanol;
0.45 g of a white solid were isolated after crystallization in a
binary mixture 20:80 benzene/hexanes; 88% yield isolated
product; mp 108 °C dec. Spectral properties: 1H NMR (DMSO-
3
d6): δ 7.4 to 7.6 ppm, two doublets, J HH ) 7.8 Hz, 4H; 7.1 to
3
7.3 ppm, m, J HH ) 7.2 Hz, 6H; 6.104 ppm, d, J HR-F ) 46.2
Hz, 1H. 19F NMR (DMSO-d6): δ -189.954 ppm, d, J HR-F
)
44.88 Hz. 13C NMR (DMSO-d6): δ 78.44 ppm, d, J CF ) 21.7
Hz; 91.96 ppm, d, J CF ) 188.9 Hz; 126.10 ppm; 126.39 ppm;
126.59 ppm; 126.77 ppm; 127.62 ppm; 127.84 ppm; 143.87
ppm, d, J CF ) 4.1 Hz; 144.91 ppm; 168.73 ppm, d, J CF ) 25.1
Hz.
Syn th esis of (RR/SS)-2-F lu or o-3-h yd r oxy-3,4-d ip h en -
ylbu tyr ic Acid , 5k . Starting from 3.024 g (10.0 mmol) of ethyl
2-fluoro-3-hydroxy-3,4-diphenylbutyrate (95% RR/SS and 5%
RS/SR) dissolved in 50 mL of ethanol and 0.48 g of 98% NaOH

78 J . Org. Chem., Vol. 67, No. 1, 2002
Ocampo et al.
(10.0 mmol) dissolved in 50 mL of ethanol. A white solid (2.578
g, 94% yield) was obtained, mp 108-110 °C dec. The product
was crystallized from a binary mixture 20:80 benzene/hexanes,
giving rise to 100% diastereomerically pure RR/SS product.
Spectral properties: 1H NMR: δ 7.27-7.36 ppm, m, 5H; 7.10-
ppm, d, J CF ) 23.68 Hz. Anal. Calcd: C, 70.82; H, 5.94; F,
6.59. Found: C, 70.59; H, 6.13.
X-r a y Exp er im en ta l Da ta . 2h data: a ) 7.8371(6) Å, b )
9.0799(7) Å, c ) 0.1826(7) Å, R ) 66.681(1)°, â ) 72.712(1)°, γ
) 84.537(1)°, V ) 572.76(8) Å3, Z ) 2, triclinic P-1, T ) 173
K. 2k data: a ) 9.9547(5) Å, b ) 10.4166(5) Å, c ) 15.0323(7)
Å, â ) 94.239(1)°, V ) 1554.5(1) Å3, Z ) 4, monoclinic P2(1)/
n, T ) 173 K.
3
4
7.20 ppm, m, 3H; 6.90 ppm, dd, J HH ) 7.0 Hz, J HH ) 1.9 Hz,
2H; 5.7 ppm, broad singlet, 2H; 5.15 ppm, d, J HR-F ) 47.7 Hz,
2
1H; 3.39 ppm, d, J HH ) 14.7 Hz; overlapped with 3.34 ppm,
2
3
dd, J HH ) 14.7 Hz, J HF ) 1 Hz approximately, 2H total. 19F
NMR: δ -195.94 ppm, d, J HR-F ) 47.4 Hz. 13C NMR: δ 44.06
ppm, d, J CF ) 4.2 Hz; 77.62 ppm, d, J CF ) 19.8 Hz; 92.06 ppm,
d, J CF ) 193.48 Hz; 126.13 ppm; 127.29 ppm; 128.10 ppm;
128.31 ppm; 128.33 ppm; 131.02 ppm; 134.44 ppm; 139.78 ppm
d, J CF ) 2.8 Hz; 171.36 ppm, d, J CF ) 25.3 Hz. Anal. Calcd:
C, 70.062; H, 5.512; F, 6.926. Found: C, 70.014; H, 5.661.
Syn th esis of RS/SR-2-F lu or o-3-h yd r oxy-3,4-d ip h en yl-
bu tyr ic Acid , 5l. Starting from 2.57 g (8.5 mmol) of ethyl
2-fluoro-3-hydroxy-3,4-diphenylbutyrate (98% diastereomeri-
cally pure RS/SR and 2% RR/SS) dissolved in 40 mL of ethanol
and 0.40 g of 98% NaOH (8.5 mmol) dissolved in 45 mL of
ethanol. A white solid (2.16 g, 93% yield) was obtained, mp
91-92 °C. The product was recrystallized from a binary
mixture 15:75 benzene/hexanes, giving rise to 100% diaste-
reomerically pure RS/SR isomer. Spectral properties: 1H
The structures were solved by the direct methods in
SHELXTL5 and refined using full-matrix least-squares meth-
ods.25 The non-H atoms were treated anisotropically, whereas
the hydrogen atoms were calculated in ideal positions and were
riding on their respective carbon atoms, except the hydroxy
proton in 2k which was obtained from a difference Fourier
map and refined freely. The H2 and F1 on C2 were found to
be disordered and were refined in two parts, thus allowing for
a 95:5 ratio of the diastereomers. Their site occupation factors
were dependently refined to 0.95(1) for the major part and,
consequently, 0.05(1) for the minor part. The minor group was
constrained to maintain a geometry similar to the major part.
For 2h , a total of 156 parameters were refined in the final
cycle of refinement using 2335 reflections with I > 2σ(I) to
yield R1 and wR2 of 3.70% and 9.55%, respectively. For 2k , a
total of 205 parameters were refined in the final cycle of
refinement using 2999 reflections with I > 2σ(I) to yield R1
and wR2 of 4.45% and 12.00%, respectively. Refinement was
done using F2.
3
4
NMR: δ 7.44 to 7.48 ppm, dt, J HH ) 8.5 Hz, J HH ) 1.4 Hz,
2H; 7.23 to 7.36 ppm, m, 6H; 7.15 to 7.20 ppm, m, 2H; 4.92
2
ppm, d, J HR-F ) 46.5 Hz, 1H; 3.34 ppm, dd, J HH ) 13.8 Hz,
2
3
3J HF ) 2.7 Hz, 1H; 3.13 ppm, dd, J HH ) 13.9 Hz, J HF ) 2.2
Hz, 1H. 19F NMR: δ -190.04 ppm, d, J HR-F ) 47.1 Hz. 13C
NMR: δ 45.28 ppm; 77.02 ppm, d, J CF ) 19.5 Hz (overlapped
with the CDCl3 triplet); 90.61 ppm, d, J CF ) 199.51 Hz; 125.43
ppm; 125.47 ppm; 127.25 ppm; 128.18 ppm; 128.35 ppm;
Ack n ow led gm en t. We gratefully acknowledge the
Instituto Colombiano para el Fomento de la Ciencia y
la Tecnolog´ıa “Francisco J ose´ de Caldas” COLCIEN-
CIAS (Grant No. 1106-05-038-99), Vice-Rector´ıa de
Investigaciones y Postgrados from Universidad de Cal-
das, Vice-Rector´ıa de Investigaciones from Universidad
del Valle. Support of this research in part by the
National Science Foundation (W.R.D.) is acknowledged
with thanks, and K.A.A. also acknowledges the National
Science Foundation and the University of Florida for
funding of the purchase of the X-ray equipment.
131.21 ppm; 134.90 ppm; 141.23 ppm; 173.57 ppm, d, J CF
23.9 Hz.
)
Syn th esis of 3-Ben zyl-2-flu or o-3-h yd r oxy-4-p h en ylbu -
tyr ic Acid , 5m . Starting from 3.16 g (10.0 mmol) of ethyl
3-benzyl-2-fluoro-3-hydroxy-4-phenylbutyrate in 50 mL of
ethanol and 0.4 g of 98% NaOH (10.0 mmol) in 50 mL of
ethanol. The product was crystallized from a mixture of 10:90
benzene/hexanes, 96% yield of isolated product, mp 117-119
°C dec. Spectral properties: 1H NMR: δ 7.19-7.32 ppm, m,
10H; 6.6 ppm, broad singlet, 2H; 4.569 ppm, d, J HR-F ) 45.9
Su p p or tin g In for m a tion Ava ila ble: ORTEP drawings
and X-ray crystal reports for compounds 2h and 2k and
pertinent ROE data. This material is available free of charge
via the Internet at http://pubs.acs.org.
2
3
Hz, 1H; 2.98 ppm, s, 2H; 2.94 ppm, dd, J HH ) 14.1 Hz, J HF
2
3
) 3.0 Hz, 1H; 2.70 ppm, dd, J HH ) 13.80 Hz, J HF ) 2 Hz,
1H. 19F NMR: δ -194.707 ppm, d, J HR-F ) 46.85 Hz. 13C
NMR: δ 41.98 ppm, 75.22 ppm, d, J CF ) 20.6 Hz; 88.46 ppm,
d, J CF ) 194.9 Hz; 127.18 ppm; 128.42 ppm; 128.52 ppm;
131.10 ppm; 131.15 ppm; 135.30 ppm; 135.38 ppm; 173.41
J O015778X
(25) Sheldrick, G. M. SHELXTL5; Bruker-AXS: Madison, WI, 1998.