Application of (S)-trifluorolactic acid for chiral derivatizing agent

Article abstract of DOI:10.1016/S0022-1139(99)00051-2

Optical purities of alcohols could be conveniently estimated by at least one of such analytical methods as GC, LC or ¹⁹F-NMR after derivatization into the corresponding esters by reaction with chiral trifluorolactic acid. O-Alkylation of triflu

Full text of DOI:10.1016/S0022-1139(99)00051-2

Journal of Fluorine Chemistry 97 (1999) 213±221
Application of (S)-tri¯uorolactic acid for
chiral derivatizing agent
Toshio Kubotaa,*, Jun Kanegaa, Toshimasa Katagirib
aDepartment of Materials Science, Ibaraki University, Nakanarusawa, Hitachi, Ibraki 316-8511, Japan
bDepartment of Applied Chemistry, Okayama University, Tsushimanaka, Okayama 700-8530, Japan
Received 12 November 1998; received in revised form 9 December 1998; accepted 9 December 1998
Dedicated to Dr. Yoshiro Kobayashi on the occasion of his 75th birthday
Abstract
Optical purities of alcohols could be conveniently estimated by at least one of such analytical methods as GC, LC or 19F-NMR after
derivatization into the corresponding esters by reaction with chiral tri¯uorolactic acid. O-Alkylation of tri¯uorolactates was found to
enhance the separation of isomers by GC and 19F-NMR analyses. # 1999 Elsevier Science S.A. All rights reserved.
Keywords: Tri¯uorolactic acid; Chiral derivatizing agent; Chiral alcohol; Optical purity; Diastereomer
1. Introduction
possible to estimate optical purities of diastereomeric sub-
strates by judicious modi®cation of this speci®c site, even
when good separation is not attained. Moreover, TFLA
seems to be a good derivatizing agent because (i)
19F-NMR, sensitive to subtle circumstantial changes, can
be used, (ii) the resultant derivatives can be applied to GC
analysis due to increased volatility by three ¯uorines, and
(iii) this compound is readily accessible by several synthetic
methods [11±17], among which resolution of racemic TFLA
[13] and nitric acid oxidation of chiral tri¯uoropropene
oxide [14±16] are excellent routes of choice. Chiral tri¯uor-
opropene oxide is also prepared through asymmetric reduc-
tion of 1-bromo-3,3,3-tri¯uro-2-propanone [17]. Further-
more, optically pure (S)-TFLA can be obtained through
recrystallization twice from the ether±chloroform (1:50)
system [14±16].
Recent increasing demands of chiral compounds in the
®elds of medicinals, electronic devices, and so on has led to
the rapid development of a number of novel as well as
versatile asymmetric synthetic methods (see, for example,
[1]), which also strongly require appropriate analytical
means for determination of optical purities of chiral products
thus obtained. One of the easiest methods for this purpose
would be GC and LC equipped with chiral ®xed phase
columns, but the setting of analytical conditions is intricate.
A new type of chiral derivatizing agents (CDA), a-cyano-
a-¯uorophenylacetic acid and a-cyano-a-¯uoro-p-tolylace-
tic acid for 19F-NMR analysis were reported by Takeuchi
et al. [2±5], and the utility of a-(tri¯uoromethyl)-a-metho-
xypropionic acid was clari®ed by Yasuhara et al. for GC
analysis [6]. Although many other CDAs are also commer-
cially available and provide optical purity information by
way of chromatographic or spectroscopic analyses [7±9],
they are generally quite expensive and sometimes unstable
for long storage. Additionally, some of them, for example
MTPA, must be transformed to reactive forms before reac-
tion with substrates [10].
This article describes the investigation for the potential
ability of TFLA as a chiral derivatizing agent for optically
active chiral secondary alcohols.
2. Results and discussion
2.1. Esterification of (S)-TFLA with chiral secondary
alcohols
On the other hand, as tri¯uorolactic acid (TFLA) has a
hydroxyl group a to the carbonyl moiety, it might be
When (S)-TFLA is used as CDA or resolving reagent of
alcohols (1), esteri®cation is desirable to readily proceed
under mild conditions in quantitative yield. TFLA could not
*Corresponding author. Tel.: +81-294-38-5060; fax: +81-294-34-4668;
e-mail: kubotacc@hit.ipc.ibaraki.ac.jp
0022-1139/99/$ ± see front matter # 1999 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 0 5 1 - 2

214
T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
complex mixture. Fortunately, as TFLA has relatively high
acidity, TFLA is considered to behave as an acid catalyst for
esteri®cation. Actually, azeotropic dehydration from a mix-
ture of (S)-TFLA and alcohols (1) readily proceeded to form
the corresponding esters (2) using benzene or toluene as a
solvent even without any additional acid catalysts such as
p-toluenesulfonic acid or sulfuric acid (Scheme 1). In spite
of the hydroxyl group at the chiral center, the self-conden-
sation of TFLA affording the corresponding oligomers did
not proceed at all under the esteri®cation condition and even
re¯uxing of benzene solution of TFLA in the presence of
p-toluenesulfonic acid. No racemization was observed
during the esteri®cation procedure by GC analysis using
a column modi®ed with chiral stationary phase (CP-Cyclo-
dex-b236M; GL Science).
On the other hand, since 2-acyloxy-3,3,3-tri¯uoropropio-
nyl chloride derived from TFLA readily reacted with amines
to form the corresponding amide with racemization, TFLA
is unfavorable for CDAs for chiral amines.
2.2. Chromatographic separation of diastereomers of
esters 2
As esters 2 have a tri¯uoromethyl group at a-position to
carbonyl, their volatility is considered to be suitable for GC
analysis. Actually, the retention times of 2 were no more
than 20 min under general analytical conditions with TC-
WAX (PEG-20M equivalent, strong-polar capillary col-
umn). In Table 1 are summarized such analytical results
Scheme 1.
be derived in acid chloride form with thionyl chloride or
phthaloyl dichloride, because the reaction procduces a
Table 1
Gas chromatographic separation of diastereomeric ester 2
Ester 2
Separation index
GC conditionsa
c
1
)
ꢀb
k0
Rsd
1.637
Initial temp. (8C)e
Program rate (8C min
2a
2b
2c
2d
2e
2f
1.063
1.025
1.000
1.022
1.033
1.080
1.152
1.056
1.056
1.038
1.671
1.688
1.728
1.670
1.506
1.027
1.000
1.000
1.000
1.083
9.337
2.067
3.246
2.262
7.892
5.459
12.03
10.57
9.912
7.690
14.70
2.163
1.728
9.592
8.667
12.86
6.902
3.233
15.26
3.715
70
120
120
120
140
150
140
120
120
100
80
0
4
4
0
0
0
1
0
0
0
1
4
4
4
1
1
4
4
4
0
1.420
0.000
1.482
1.568
3.286
1.921
1.718
1.761
2.215
10.440
6.905
3.724
6.304
8.110
1.903
0.000
0.000
0.000
3.083
2g
2h
2i
2j
2k
2l
120
120
120
140
110
120
120
120
160
2m
2n
2o
2p
2q
2r
2s
2t
a TC-WAX (0.25 mm i.d.Â30 m, GL Science, PEG-20M compatible capillary column) was used. Carrier: 70 ml/min (He); sprit ratio: 100:1.
b Separation factor.
c Capacity factor of latter fraction.
d Resolution.
e Each initial temperature was held for 10 min, and all final temperatures were controlled at 2008C except the cases where program rate was zero.

T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
215
The relatively large resolutions of allylic esters 2h and 2i
would be stemmed from the structural difference between
alkyl and vinyl groups. Similarly, the large diastereomeric
separation of 2g might result from the polar terminal
ethoxycarbonyl. In contrast, as the steric difference around
the chiral center in the alkyl part between ester diastereo-
mers is relatively small in the case of 2q and 2r, the
diastereomeric separation did not occur.
Moreover, similar GC analysis of 2 was performed under
normalized conditions. The plots between the carbon num-
ber in the alkyl part of esters (2a±e) and separation factors ꢀ
had a minimum at nˆ6 (2c). On the contrary, the curve of ꢀ
in similar plots for the esters bearing a tri¯uoromethyl group
at the alkyl part (2k±o) had a maximum at nˆ6 (2m)
(Fig. 1). This quite sharp contrast was considered not only
due to the steric factor difference around chiral centers of
alkyl parts, but also due to the polarity difference of the
whole molecular structure.
In the LC analysis, the diastereomeric separation of 2
could not be observed using the reversed phase type column
(ODS), but in half of the samples examined here, base-line
separation (Rs>1.500) was observed by a silica gel column
eluted with hexane±2-propanol (95:5). The analytical
results are summarized in Table 2.
Fig. 1. Plots of separation factor (ꢀ) vs. alkyl carbon chain length of 2a±e
and 2k±o. Normalized GC conditions: column, TC-WAX (0.25 mm
i.d.Â30 m, GL Science); carrier, 70 ml (He, split ratio, 100:1); initial
temperature, 1208C was held for 10 min; program rate, 48C/min; final
temperature, 2008C.
2.3. NMR spectroscopic separation of 2
as separation factor (ꢀ), capacity factor (k0) of isomer with
longer retention time and resolution (Rs>1.500 :base-line
separation) as the analytical indication functions.
19F-NMR spectrometry is one of the most useful instru-
mental techniques for the analysis of optical purity when
Table 2
Diastereomeric separation of 2 using HPLCa
1
Ester 2
Separation index
Flow rate (ml min
)
ꢀb
RTc (min)
Rsd
2a
2b
2c
2d
2e
2f
1.000
1.108
1.037
1.056
1.059
1.073
1.058
1.000
1.062
1.033
1.000
1.000
1.090
1.000
1.133
1.000
1.076
1.270
1.069
1.000
4.180
29.74
4.730
26.30
4.560
5.193
6.353
3.170
9.700
31.86
27.19
7.892
4.520
13.42
14.21
9.010
3.666
3.580
27.35
3.940
0.000
1.995
1.029
1.548
1.617
1.941
1.544
0.000
1.038
1.532
0.000
0.000
1.905
0.000
2.001
0.000
1.709
3.167
1.564
0.000
9.0
1.0
6.0
1.0
6.0
6.0
9.0
9.0
3.0
1.0
3.0
3.0
9.0
3.0
3.0
3.0
9.0
9.0
3.0
9.0
2g
2h
2i
2j
2k
2l
2m
2n
2o
2p
2q
2r
2s
2t
a Column: Inertsil SIL 100A (4.0 mm i.d.Â250 mm, GL Science); eluent: n-hexane-2-propanol (95/5 [v/v]); charged sample: 5 ml (concentrationˆ10 mg/
1.0 ml).
b Separation factor.
c Retention time.
d Resolution.

216
T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
Table 3
19F-NMR shift difference of 2a
Ester 2
Áꢁ (Hz)
2a
2b
2c
2d
2e
2f
1.69
3.39
2.26
1.69
2.26
10.05
3.67
2g
2h
2i
3.90
2.15
Scheme 2.
2j
2.71
1.81 (1.69)b
1.75 (1.86)b
2k
2l
resolution of diastereomeric peaks or signals. Methoxy-
methyl (MOM) and 2-methoxyethoxymethyl (MEM) chlor-
ide were selected as the alkylating reagents because of their
high reactivity towards the hydroxyl group.
Derivatization to MOM and MEM ethers (3, 4) was
carried out through a Williamson type method in quantita-
tive yield (Scheme 2). As esters 2 have an a-methine proton
in TFLA unit, epimerization might occur. However, any
racemization did not occur in these cases con®rmed by GC
and/or LC analyses using columns with chiral stationary
phase.
2m
2n
2o
2p
2q
2r
2s
2.03 (2.43)b
1.98 (2.26)b
2.15 (1.81)b
2.37
1.24
3.67
1.58
2t
6.27
a Hitachi R1200F (56.46 MHz) spectrometer was used. Solvent: CDCl3.
b Áꢁ of CF3 group at alcohol part of 2.
¯uorine-containing CDAs are used. In particular, even a
reaction mixture of derivatization step in a variety of
common solvents can be directly used for the determination
of the optical purity, since most substrates whose optical
purities have to be determined do not have ¯uorines in the
molecules. Although the CF3 signal of the TFLA unit
appears as a doublet in 19F-NMR, the spectra of 2 showed
clear base-line separated diastereomeric signals and gave
highly precise quanti®cation of optical purities even by a
56.46 MHz 19F-NMR spectrometer (Table 3).
In the GC analyses of 3 and 4, most of the samples
showed the base-line separated peaks. Especially, diaster-
eomeric esters (3b±e, 4b±e) from 2-alkanols showed base-
line separation patterns with larger resolution values than
those of non-O-alkylated esters 2. On the contrary, the
diastereomeric peaks were observed as a completely sepa-
rated pattern in the cases of unsubstituted 2f, but O-alkyla-
tion (3f, 4f) resulted in a decrease in Rs values (Rs<1.500;
overlapping pattern) (Table 4).
The plotting pattern of ꢀ values vs. carbon number of
chiral alcohols in both O-MOM esters and O-MEM ester
(3a±e and 4a±e) showed a maximum value at 2-hexylesters,
quite different from that of non-O-alkylated ester (2a±e)
(see Figs. 2 and 1). These results suggested that O-alkyla-
tion of the original esters 2a±e can change the separation
mode in the GC column by masking the polar hydroxyl
group. Thus, it was found that O-alkylation of 2 by MOM or
MEM groups is effective for the enhancement of resolution
of diastereomeric peak separation on the gas chromatogram
in many cases.
On the other hand, isolation of chiral derivatized samples
1
is necessary for H-NMR measurement. Additionally, as
many signals originated from samples appear in the
1H-NMR spectra, quanti®cation of optical purities is sig-
ni®cantly more complicated than in the case of 19F-NMR.
Although doublet methyl signals at 1-position of secondary
alkyl moiety of 2a±h appeared as diastereomerically sepa-
rated patterns (Áꢁˆ8.00±14.40 Hz), quanti®cation using
1H-NMR was concluded to be available only for 2-alkanols
but unfavorable for the other alcohols.
On the 19F-NMR analysis, the results for esters 3a±e and
4a±e showed the shift difference enhancement, while the
resolutions became poorer by O-alkylation in the case of
3f±s and 4f±s. All of the separation patterns of 3 and 4 were
clearly base-line separation and shift differences were large
enough to determine optical purities of original alcohols
(Table 5).
As acetal methylene signals of MOM and MEM groups
are observed as singlet sometimes at relatively isolated shift
position (typically, ꢁ 4.2±5.0 ppm) in 1H-NMR, it is
expected that the methylene signals are available for quan-
ti®cation of optical purity. Although the Áꢁs of terminal
2.4. Improvement of diastereomeric separation by
O-alkylation of 2
The results described above showed that TFLA, easy to
be stored and to be used, can be utilized as a CDA for chiral
secondary alcohols. However, in some cases diastereomeri-
cally separated signals or peaks of TFLA-based esters are
1
not observed by GC, LC, H-NMR or 19F-NMR.
Fortunately, as esters 2 have a hydroxyl group at
a-position to carbonyl of the acid part, it is thought that
O-alkylation of hydroxyl group may favorably affect the

T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
217
Table 4
Gas chromatographic separation of O-alkylated diastereomeric esters 3 and 4
Ester 3 or 4
Separation index
GC conditionsa
c
1
ꢀb
k0
Rsd
Initial temp. (8C)c
Program rate (8C min
)
3a
3b
3c
3d
3e
3f
1.062
1.041
1.052
1.039
1.019
1.002
1.040
1.000
10.57
9.662
6.519
5.991
7.701
49.40
7.592
8.982
1.621
1.635
1.591
1.684
1.792
1.448
1.972
0.000
70
80
0
1
0
4
4
1
0
0
100
120
120
60
3j
70
3s
120
4a
4b
4c
4d
4e
4f
1.049
1.026
1.042
1.062
1.081
1.003
1.042
1.000
7.420
8.258
8.494
6.275
12.04
16.34
8.494
17.06
1.588
1.566
1.590
1.880
2.125
1.471
1.981
0.000
120
120
80
0
4
0
0
0
5
0
4
130
160
60
4j
80
120
4s
a TC-WAX (0.25 mm i.d.Â30 m, GL Science, PEG-20M compatible capillary column) was used. Carrier: 70 ml/min (He); sprit ratio: 100:1.
b Separation factor.
c Capacity factor of latter fraction.
d Resolution.
e Each initial temperature was held for 10 min, and all final temperatures were controlled at 2008C except for the cases when program rate was zero.
methyl moieties in MOM and MEM groups 3a±f and 4a±f
were 0.80±22.0 Hz and 0.00±11.6 Hz, respectively, separa-
tion of acetal methylene signals were not detected except in
the case of 3f.
3. Conclusion
As described above, it was found that enantiomerically
pure TFLA is a chiral derivatizing agent for chiral secondary
alcohols and the protocol with tandem steps including O-
alkylation provides a high performance way for determina-
tion of optical purity of chiral alcohols using GC and
19F-NMR.
4. Experimental
TFLA was prepared by nitric acid oxidation of (S)-
tri¯uoropropene oxide, which was supplied by Japan
Energy. 1,1,1-Tri¯uoroalkan-2-ols were supplied by Morita
1
Chemical. H-NMR were recorded with JEOL GSX-400
(400 MHz) spectrometer in CDCl3. Chemical shifts were
recorded in ppm, down®eld from internal tetramethylsilane.
Table 5
19F-NMR shift difference of 3 and 4a
O-MOM form
Ester 3
O-MEM form
Ester 4
Áꢁ (Hz)
Áꢁ (Hz)
3a
3b
3c
3d
3e
3f
4.52
3.95
2.49
5.08
4.38
2.59
2.03
2.35
4a
4b
4c
4d
4e
4f
5.08
3.95
4.23
2.26
6.21
1.65
1.88
2.78
Fig. 2. Plots of separation factor (ꢀ) vs. alkyl carbon chain length of 3a±e
and 4a±e. Normalized GC conditions: column, TC-WAX (0.25 mm
i.d.Â30 m, GL Science); carrier, 70 ml (He, split ratio, 100:1); initial
temperature, 1208C was held for 10 min; program rate, 48C/min; final
temperature, 2008C.
3j
4j
3s
4s
a Hitachi R-1200F (56.46 MHz) spectrometer was used. Solvent: CDCl3.

218
T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
19F-NMR spectra were recorded with a Hitachi R-1200F
(56.46 MHz) spectrometer in CDCl3 for isolated samples,
and chemical shifts were reported in ppm down®eld from
external CF3CO2H. Infrared (IR) spectra were obtained on a
JASCO A-120 spectrometer and all spectra were reported in
wave numbers (cm 1), with the reference being at the
1601.4 cm 1 absorption of a polystyrene ®lm. High resolu-
tion mass spectra (HRMS) were taken on a JEOL DX-300
spectrometer operating with EI mode.
2-Decyl 3,3,3-tri¯uoro-2-hydroxypropionate (2e): Yield
99%. 1H-NMR ꢁ 0.880 and 0.883 (3H, t, Jˆ6.2 Hz), 1.281
and 1.314 (3H, d, Jˆ6.0 Hz), 1.38±1.70 (14H, m), 3.46 (1H,
s), 4.439 and 4.449 (1H, q, Jˆ6.8 Hz), 5.11 (1H, m).
19F-NMR ꢁ 1.240 and 1.280 (d, Jˆ6.8 Hz). IR (KBr) v
3530, 1740 cm 1. HRMS (EI): Calcd. for C13H23O3F3
284.1599, found 284.1673.
1-Phenylethyl 3,3,3-tri¯uoro-2-hydroxypropionate (2f):
Yield 99%. 1H-NMR ꢁ 1.621 and 1.641 (3H, d, Jˆ6.4 Hz),
3.42 (1H, br), 4.458 and 4.512 (1H, q, Jˆ6.8 Hz). 6.029 and
6.051 (1H, q, Jˆ6.8 Hz). 19F-NMR ꢁ 0.592 and 0.770 (d,
Jˆ6.8 Hz). IR (KBr) v 3300, 1720 cm 1. HRMS (EI):
Calcd. for C11H11O3F3 248.0660, found 248.0735.
Ethyl (3,3,3-tri¯uoro-2-hydroxypropanoyl)butanonate
(2g): Yield 95%. 1H-NMR ꢁ 1.249 and 1.267 (3H, t,
Jˆ7.2 Hz) 1.371 and 1.407 (1H, d, Jˆ6.0 Hz), 2.55±2.76
(2H, m), 3.40 (1H, br), 4.14 (2H, m), 4.442 and 4.448 (1H, q,
Jˆ6.8 Hz), 5.51 (1H, m). 19F-NMR ꢁ 0.432 and 0.497 (d,
Jˆ6.8 Hz). IR (KBr) v 3560, 1760 cm 1. HRMS (EI):
Calcd. for C9H13O5F3 258.0715, found 258.0632.
3-But-1-enyl 3,3,3-tri¯uoro-2-hydroxypropionate (2h):
Yield 95%. 1H-NMR ꢁ 1.398 and 1.423 (3H, d, Jˆ6.4 Hz),
1.73 (1H, m), 3.40±3.44 (2H, m), 3.40 (1H, br), 4.442 and
4.459 (1H, q, Jˆ6.8 Hz). 19F-NMR ꢁ 0.621 and 0.690 (d,
Jˆ6.8 Hz). IR (KBr) v 3540, 1770 cm 1. HRMS (EI):
Calcd. for C7H9O3F3 198.0504, found 198.0573.
4.1. General procedure for TFLA esters 2
A solution of alcohol 1 (1 mmol) and TFLA (1.2 mmol)
in benzene (40 ml) was re¯uxed using Dean-Stark trap. The
completion of the reaction was con®rmed by monitoring
using 19F-NMR and/or GC. Typically, the reaction com-
pleted within 4 h. For 19F-NMR and GC analysis for
diastereomeric separation, the reaction mixture could be
directly used as the samples. For 1H-NMR and LC analysis
to separate diastereomeric signals or peaks, the reaction
mixture was treated as follows: Benzene was evaporated
under reduced pressure and hexane (40 ml) was added into
the residual syrup. After trituration, the hexane layer was
separated from unreacted TFLA. The hexane solution was
concentrated and dried-up to produce the corresponding
ester 2. If further puri®cation was necessary, distillation or
dry column ¯ash chromatography (silica gel, hexane) was
adopted.
3-Oct-1-enyl 3,3,3-tri¯uoro-2-hydroxypropionate (2i):
Although the isolated yield 90% was lower than other esters
2, the yield estimated from 19F-NMR spectrum of the
2-Butyl 3,3,3-tri¯uoro-2-hydroxypropionate (2a): Yield
1
1
98%. H-NMR ꢁ 0.918 and 0.937 (3H, t, Jˆ7.2, 7.0 Hz),
reaction mixture was quantitative. H-NMR ꢁ 0.883 and
0.892 (3H, t, Jˆ6.4 Hz) 1.20±1.80 (8H, m), 3.45 (1H, br),
1.285 and 1.317 (3H, d, Jˆ6.0 Hz), 1.60±1.72 (2H, m), 3.49
(1H, br), 4.433 and 4.444 (1H, q, Jˆ6.4 Hz), 5.06 (1H, tq,
Jˆ3.2, 6.4 Hz). 19F-NMR ꢁ 1.640 and 1.670 (d, Jˆ6.4 Hz).
IR (KBr) v 3550, 1750 cm 1. HRMS (EI): Calcd. for
C7H11O3F3 200.0660, found 200.0710.
4.461 and 4.478 (1H, q, Jˆ6.8 Hz), 5.38 (1H, q, Jˆ6.0 Hz),
5.24±5.35 (2H, m), 5.73±5.86 (1H, m). 19F-NMR ꢁ 1.708
1
and 1.746 (d, Jˆ6.8 Hz). IR (KBr) v 3315, 1770 cm
.
HRMS (EI): Calcd. for C11H17O3F3 254.1130, found
254.1056.
2-Pentyl 3,3,3-tri¯uoro-2-hydroxypropionate (2b): Yield
95%. 1H-NMR ꢁ 1.232 (3H, t, Jˆ6.0 Hz), 1.284 and 1.317
(3H, d, Jˆ6.0 Hz), 1.34±1.73 (4H, m), 4.449 and 4.457 (1H,
q, Jˆ6.8 Hz), 5.12 (1H, q, Jˆ6.8 Hz), 6.02 (1H, br).
19F-NMR ꢁ 1.082 and 1.142 (d, Jˆ6.8 Hz). IR (KBr) v
3620, 1700 cm 1. HRMS (EI): Calcd. for C8H13O3F3
214.0817, found 214.0883.
2-Hexyl 3,3,3-tri¯uoro-2-hydroxypropionate (2c): Yield
95%. 1H-NMR ꢁ 0.894 and 0.909 (3H, t, Jˆ6.8 Hz), 1.285
and 1.318 (3H, d, Jˆ6.4 Hz), 1.57±1.68 (6H, m), 3.49 (1H,
s), 4.434 and 4.441 (1H, q, Jˆ6.8 Hz), 5.11 (1H, m).
19F-NMR ꢁ 1.130 and 1.20 (d, Jˆ6.8 Hz). IR (KBr) v
3530, 1740 cm 1. HRMS (EI): Calcd. for C9H15O3F3
228.0973, found 228.0905.
3-Hexyl 3,3,3-tri¯uoro-2-hydroxypropionate (2j): Yield
93%. 1H-NMR ꢁ 0.90 (3H, t, Jˆ6.8 Hz), 0.93 (3H, t,
Jˆ7.2 Hz), 1.24±1.70 (6H, m), 4.444 and 4.462 (1H, q,
Jˆ6.8 Hz), 5.00±5.07 (1H, m). 19F-NMR ꢁ 0.550 and 0.698
(d, Jˆ6.8 Hz). IR (KBr) v 3325, 1720 cm 1. HRMS (EI):
Calcd. for C9H15O3F3 228.0973, found 228.0900.
1,1,1-Tri¯uoro-2-butyl 3,3,3-tri¯uoro-2-hydroxypropio-
1
nate (2k): Yield 98%. H-NMR ꢁ 1.027 and 1.032 (3H, t,
Jˆ7.6 Hz), 1.80±1.90 (4H, m), 4.44 (1H, q, Jˆ6.8 Hz), 5.35
(1H, m), 5.80 (1H, br). 19F-NMR ꢁ 0.589 and 0.621 (3F,
d, Jˆ7.2 Hz), 1.350 and 1.380 (3F, d, Jˆ6.8 Hz). IR (KBr) v
3345, 1710 cm 1. HRMS (EI): Calcd. for C7H8O3F6
254.0377, found 254.0302.
2-Octyl 3,3,3-tri¯uoro-2-hydroxypropionate (2d): Yield
96%. 1H-NMR ꢁ 0.881 and 0.887 (3H, d, Jˆ6.8 Hz), 1.282
and 1.315 (3H, d, Jˆ6.4 Hz), 1.40±1.70 (10H, m), 3.46 (1H,
s), 4.431 and 4.440 (1H, q, Jˆ6.8 Hz), 5.11 (1H, m).
19F-NMR ꢁ 2.140 and 2.170 (d, Jˆ6.8 Hz). IR (KBr) v
3400, 1720 cm 1. HRMS (EI): Calcd. for C11H19O3F3
256.1286, found 256.1342.
1,1,1-Tri-¯uoro-2-pentyl 3,3,3-tri¯uoro-2-hydroxypro-
pionate (2l): Yield 95%. 1H-NMR ꢁ 0.963 and 0.967
(3H, t, Jˆ7.6 Hz), 1.80±1.90 (4H, m), 3.62 (1H, q,
Jˆ6.8 Hz), 4.62 (1H, m), 4.46 (1H, br). 19F-NMR ꢁ
0.590 and 0.621 (3F, d, Jˆ7.2), 1.320 and 1.353 (3F,
d, Jˆ6.8 Hz), IR (KBr) v 3340, 1760 cm 1. HRMS (EI):
Calcd. for C8H10O3F6 268.0534, found 268.0452.

T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
219
1,1,1-Tri¯uoro-2-hexyl l 3,3,3-tri¯uoro-2-hydroxypropio-
1
Jˆ7.2 Hz). 19F-NMR ꢁ 0.500 and 0.611 (d, Jˆ6.8 Hz).
IR (KBr) v 3520, 1720 cm 1. HRMS (EI): Calcd. for
C13H21O3F3 282.14428, found 282.1513.
nate (2m): Yield 93%. H-NMR ꢁ 0.950 and 0.955 (3H, t,
Jˆ7.6 Hz), 1.37±1.50 (6H, m), 3.32 (1H, br), 4.60 (1H, q,
Jˆ6.8 Hz), 5.40 (1H, m). 19F-NMR ꢁ 0.540 and 0.575
(3F, d, Jˆ7.2), 1.250 and 1.293 (3F, d, Jˆ6.8 Hz). IR (KBr)
v 3355, 1760 cm 1. HRMS (EI): Calcd. for C9H12O3F6
282.0691, found 282.0602.
4.2. General procedure for O-alkylated TFLA esters 3
and 4
1,1,1-Tri¯uoro-2-octyl l 3,3,3-tri¯uoro-2-hydroxypropio-
1
A solution of ester 2 (0.5 mmol) in THF (5.0 ml) was
added into a suspension of sodium hydride (0.5 mmol) in
THF (5.0 ml) at 08C, and then, a solution of MOM or MEM
chloride (0.6 mmol) in THF (5.0 ml) was dropped into the
mixture at 08C. After stirring for 2 h at room temperature,
the reaction mixture was concentrated in vacuo and the
residue was extracted with ether. The ethereal solution was
dried over anhydrous sodium sulfate. The syrup obtained by
evaporation of the ethereal solution was puri®ed by dry
column ¯ash chromatography (silica gel, hexane). Although
the isolated yields were not high, the yields determined by
19F-NMR of the reaction mixture using benzotri¯uoride as
an internal standard were quantitative values almost in every
case.
nate (2n): Yield 92%. H-NMR ꢁ 0.876 and 0.879 (3H, t,
Jˆ7.6 Hz), 1.20±1.31 (10H, m), 3.50 (1H, br), 3.90 (1H, q,
Jˆ6.8 Hz), 5.40 (1H, m). 19F-NMR ꢁ 0.486 and 0.522
(3F, d, Jˆ7.2), 1.245 and 1.285 (3F, d, Jˆ6.8 Hz). IR (KBr)
v 3360, 1770 cm 1. HRMS (EI): Calcd. for C11H16O3F6
310.1004, found 310.0904.
1,1,1-Tri¯uoro-2-decyl l 3,3,3-tri¯uoro-2-hydroxypropio-
1
nate (2o): Yield 98%. H-NMR ꢁ 0.885 and 0.888 (3H, t,
Jˆ7.6 Hz), 1.27±1.57 (10H, m), 3.32 (1H, br), 3.60 (1H, q,
Jˆ6.8 Hz), 5.40 (1H, m). 19F-NMR ꢁ 0.440 and 0.478
(3F, d, Jˆ7.2), 1.255 and 1.287 (3F, d, Jˆ6.8 Hz). IR (KBr)
v 3380, 1740 cm 1. HRMS (EI): Calcd. for C13H20O3F6
338.1317, found 338.1202.
3-Octyl 3,3,3-tri¯uoro-2-hydroxypropionate (2p): Yield
92%. 1H-NMR ꢁ 0.878 and 0.915 (3H, t, Jˆ7.2 Hz), 1.28±
1.32 (9H, m), 1.56±1.70 (4H, m), 3.475 and 3.479 (1H, d,
Jˆ2.6 Hz), 4.446 and 4.464 (1H, q, Jˆ6.8 Hz), 5.11 (1H,
m). 19F-NMR ꢁ 0.545 and 0.588 (d, Jˆ6.8 Hz). IR (KBr) v
3590, 1770 cm 1. HRMS (EI): Calcd. for C11H19O3F3
256.1286, found 256.1212.
2-Butyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3a): Yield 69%. 1H-NMR ꢁ 0.915 and 0.925 (3H, t,
Jˆ7.6 Hz), 1.266 and 1.281 (3H, d, Jˆ5.6 Hz), 1.58±1.69
(2H, m) 3.436 and 3.439 (3H, s), 4.52 (1H, q, Jˆ6.8 Hz),
4.79 (2H, s), 5.01 (1H, m). 19F-NMR ꢁ 3.000 and 3.080 (d,
Jˆ7.6 Hz). HRMS (EI): Calcd. for C9H15O4F3 244.0922,
found 244.0835.
3-Decyl 3,3,3-tri¯uoro-2-hydroxypropionate (2q): Yield
99%. 1H-NMR ꢁ 0.878 and 0.888 (3H, t, Jˆ6.8 Hz), 0.909
and 0.927 (3H, t, Jˆ6.0 Hz), 1.27±1.66 (16H, m), 3.48 (1H,
br), 4.434 and 4.453 (1H, q, Jˆ6.8 Hz), 5.08 and 5.10 (1H,
m). 19F-NMR ꢁ 0.567 and 0.589 (d, Jˆ6.8 Hz). IR (KBr) v
3540, 1710 cm 1. HRMS (EI): Calcd. for C13H23O3F3
284.1599, found 284.1696.
2-Pentyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3b): Yield 93%. 1H-NMR ꢁ 0.915 and 0.925 (3H, t,
Jˆ7.6 Hz), 1.265 and 1.281 (3H, d, Jˆ6.0 Hz), 1.30±1.70
(4H, m), 3.434 and 3.436 (3H, s), 4.51 (1H, q, Jˆ6.8 Hz),
4.78 (2H, s), 5.08 (1H, m). 19F-NMR ꢁ 3.308 and 3.378 (d,
Jˆ6.8 Hz). HRMS (EI): Calcd. for C10H17O4F3 258.1079,
found 258.1004.
2-Methylbutyl 3,3,3-tri¯uoro-2-hydroxypropionate (2r):
Yield 97%. 1H-NMR ꢁ 1.249 and 1.267 (3H, t, Jˆ6.8 Hz),
1.371 and 1.407 (3H, d, Jˆ6.0 Hz), 2.55±2.76 (3H, m), 3.40
(1H, br), 4.14 (2H, m), 4.434 and 4.453 (1H, q, Jˆ6.8 Hz),
5.08 (1H, m). 19F-NMR ꢁ 0.432 and 0.479 (d, Jˆ6.8 Hz). IR
(KBr) v 3520, 1710 cm 1. HRMS (EI): Calcd. for
C8H13O3F3 214.0817, found 214.0887.
2-Hexyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3c): Yield 77%. 1H-NMR ꢁ 0.892 and 0.900 (3H, t,
Jˆ6.4 Hz), 1.266 and 1.282 (3H, d, Jˆ6.0 Hz), 1.31±1.72
(6H, m), 3.436 and 3.444 (3H, s), 4.51 (1H, q, Jˆ7.2 Hz),
4.78 (2H, s), 5.06 (1H, m). 19F-NMR ꢁ 3.150 and 3.194 (d,
Jˆ7.2 Hz). HRMS (EI): Calcd. for C11H19O4F3 272.1235,
found 272.1178.
2-(3,3,3-Tri¯uoro-2-hydroxypropionyl)-3,3-dimethylbul-
tanolide (2s): Although the isolated yield 88% was lower
than other esters 2, the yield estimated from 19F-NMR
spectrum of the reaction mixture was quantitative.
1H-NMR ꢁ 1.147 and 1.160 (3H, s), 1.232 and 1.258
(3H, s), 3.95 (1H, d, Jˆ8.8 Hz), 4.03 (1H, d, Jˆ8.8 Hz),
4.14 (1H, s), 4.710 and 4.767 (1H, q, Jˆ6.8 Hz), 5.458 and
5.524 (1H, s). 19F-NMR ꢁ 0.767 and 0.795 (d, Jˆ6.8 Hz). IR
(KBr) v 3530, 1750 cm 1. HRMS (EI): Calcd. for
C9H11O5F3 256.0559, found 256.0610.
2-Octyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3d): Yield 88%. 1H-NMR ꢁ 0.892 and 0.900 (3H, t,
Jˆ6.8 Hz), 1.266 and 1.282 (3H, d, Jˆ6.0 Hz), 1.29±1.75
(10H, m), 3.436 and 3.438 (3H, s), 4.51 (1H, q, Jˆ6.8 Hz),
4.78 (2H, s), 5.01 (1H, m). 19F-NMR ꢁ 3.225 and 3.315 (d,
Jˆ6.8 Hz). HRMS (EI): Calcd. for C13H23O4F3 300.1548,
found 300.1626.
2-Decyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3e): Yield 84%. 1H-NMR ꢁ 0.881 and 0.884 (3H, t,
Jˆ6.0 Hz), 1.263 and 1.278 (3H, d, Jˆ6.0 Hz), 1.19±1.70
(14H, m), 3.434 and 3.436 (3H, s), 4.51 (1H, q, Jˆ6.8 Hz),
4.78 (2H, s), 5.05 (1H, m). 19F-NMR ꢁ 3.338 and 3.16 (d,
Jˆ7.6 Hz). HRMS (EI): Calcd. for C15H27O4F3 328.1861,
found 328.1806.
(1S, 2R, 5S)-(‡)- and (1R, 2S, 5R)-( )mentyl (S)-3,3,3-
1
tri¯uoro-2-hydroxypropionate (2t): Yield 91%. H-NMR ꢁ
0.727 and 0.780 (3H, d, Jˆ6.8 Hz), 0.82±2.45 (15H, m),
3.42 (1H, s), 3.94 (1H, m), 4.820 and 4.895 (1H, q,

220
T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
1-Phenylethyl 3,3,3-tri¯uoro-2-(methoxymethoxy)pro-
1.68 (14H, m), 3.382 and 3.384 (3H, s), 3.54 (2H, t,
Jˆ4.6 Hz), 3.79 (2H, t, Jˆ4.6 Hz) 4.584 (1H, q, Jˆ6.8 Hz)
4.879 and 4.880 (2H, s), 5.05 (1H, m). 19F-NMR ꢁ 3.558 and
3.668 (d, Jˆ7.2 Hz). HRMS (EI): Calcd. for C17H31O5F3
372.2124, found 372.1997.
pionate (3f): Yield 89%. 1H-NMR ꢁ 1.597 and 1.612
(3H, d, Jˆ6.8 Hz), 3.337 and 3.392 (3H, s), 4.544 and
4.557 (1H, q, Jˆ6.4 Hz), 4.741 and 4.749 (2H, s) 6.01
(1H, q, 6.4 Hz), 7.35 (5H, m). 19F-NMR ꢁ 3.208 and 3.254
(d, Jˆ6.4 Hz). HRMS (EI): Calcd. for C13H15O4F3
292.0922, found 292.0861.
1-Phenylethyl
3,3,3-tri¯uoro-2-[(2-methoxyethoxy)-
methoxy]propionate (4f): Yield 93%. 1H-NMR ꢁ 1.590
and 1.609 (3H, d, Jˆ6.8 Hz), 3.366 and 3.395 (3H, s),
3.47±3.49 (2H, m), 3.56±3.58 (2H, m), 4.614 and 4.633
(1H, q, Jˆ7.2 Hz), 4.843 and 4.848 (2H, s), 6.000 and 6.005
(1H, q, Jˆ6.4 Hz) 7.28±7.38 (5H, br). 19F-NMR ꢁ 3.390 and
3.419 (d, Jˆ7.2 Hz). HRMS (EI): Calcd. for C15H19O5F3
336.1185, found 336.1306.
3-Hexyl 3,3,3-tri¯uoro-2-(methoxymethoxy)propionate
(3j): Yield 81%. 1H-NMR ꢁ 0.906 and 0.914 (3H, t,
Jˆ5.8 Hz), 0.922 and 0.932 (3H, t, Jˆ6.0 Hz) 1.18±1.79
(6H, m), 3.390 and 3.393 (3H, s), 4.60 (1H, q, Jˆ6.8 Hz),
4.81 (2H, s), 5.00 (1H, m). 19F-NMR ꢁ 4.120 and 4.126 (d,
Jˆ6.8 Hz). HRMS (EI): Calcd. for C11H19O4F3 272.1235,
found 272.1287.
3-Hexyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
2-[3,3,3-Tri¯uoro-2-(methoxymethoxy)propionyloxy]-
1
propionate (4j): Yield 83%. H-NMR ꢁ 0.904 and 0.907
3,3-dimethylbutanolide (3s): Yield 61%. H-NMR ꢁ 1.121
(3H, t, Jˆ5.4 Hz), 0.920 and 0.926 (3H, t, Jˆ6.0 Hz) 1.19±
1.78 (6H, m), 3.382 and 3.384 (3H, s), 3.52±3.58 (2H, m),
3.77-3.81 (2H, m), 4.610 (1H, q, Jˆ6.8 Hz) 4.88 (2H, s),
4.99 (1H, m). 19F-NMR ꢁ 4.000 and 4.033 (d, Jˆ6.8 Hz).
HRMS (EI): Calcd. for C13H23O5F3 316.1498, found
316.1406.
and 1.148 (3H, s), 1.216 and 1.243 (3H, s) 3.452 and 3.504
(3H, s), 3.96 (1H, d, Jˆ6.8 Hz) 4.04 (1H, d, Jˆ8.8 Hz), 4.15
(2H, s), 4.56 (1H, q, Jˆ7.2 Hz), 5.012 and 5.029 (1H, m).
19F-NMR ꢁ 3.000 and 3.042 (d, Jˆ7.2 Hz). HRMS (EI):
Calcd. for C11H15O6F3 300.0821, found 300.0713.
2-Butyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
2-[3,3,3-Tri¯uoro-2-[(2-methoxyethoxy)methoxy]propio-
1
propionate (4a): Yield 70%. H-NMR ꢁ 0.909 and 0.923
nyloxy]-3,3-dimethylbutanolide (4s): Yield 67%. H-NMR
(3H, t, Jˆ7.6 Hz), 1.258 and 1.277 (3H, d, Jˆ6.4 Hz) 1.59±
1.67 (2H, m), 3.38 (3H, s), 3.53±3.56 (2H, m), 3.78±3.80
(2H, m), 4.592 and 4.595 (1H, q, Jˆ6.8 Hz) 4.88 (2H, s),
5.00 (1H, m). 19F-NMR ꢁ 3.270 and 3.360 (d, Jˆ6.8 Hz).
HRMS (EI): Calcd. for C11H15O5F3 288.1185, found
288.1104.
ꢁ 1.105 and 1.162 (3H, s), 1.209 and 1.259 (3H, s), 3.413 and
3.427 (3H, s), 3.57±3.60 (2H, m), 3.71±3.73 (2H, m), 3.97
(1H, d, Jˆ7.8 Hz), 4.05 (1H, d, Jˆ7.8 Hz), 4.55 (1H, q,
Jˆ6.8 Hz), 5.46 and 5.53 (1H, s). 19F-NMR ꢁ 3.666 and
3.715 (d, Jˆ6.8 Hz). HRMS (EI): Calcd. for C13H19O7F3
344.1083, found 344.1159.
2-Pentyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
propionate (4b): Yield 83%. H-NMR ꢁ 0.916 and 0.935
(3H, t, Jˆ6.0 Hz), 1.282 and 1.316 (3H, d, Jˆ6.0 Hz) 1.36±
1.73 (4H, m), 3.406 and 3.414 (3H, s), 3.53±3.60 (2H, m),
3.71±3.80 (2H, m), 4.440 and 4.448 (1H, q, Jˆ6.8 Hz) 4.819
and 4.828 (2H, s), 5.11 (1H, m). 19F-NMR ꢁ 2.900 and 2.970
(d, Jˆ6.8 Hz). HRMS (EI): Calcd. for C11H15O5F3
302.1341, found 302.1248.
Acknowledgements
The authors are grateful to Japan Energy for the gift of the
(S)-tri¯uoropropene oxide, and Morita Chemical for the gift
of 1,1,1-tri¯uoro-2-alkanols.
2-Hexyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
propionate (4c): Yield 85%. H-NMR ꢁ 0.880 and 0.884
References
(3H, t, Jˆ6.8 Hz), 1.144 and 1.204 (3H, d, Jˆ6.0 Hz) 1.28±
1.31 (6H, m), 3.401 and 3.409 (3H, s), 3.56±3.60 (2H, m),
3.71±3.77 (2H, m), 4.429 and 4.42 (1H, q, Jˆ7.2 Hz) 4.749
and 4.758 (2H, s), 4.96 (1H, m). 19F-NMR ꢁ 3.076 and 3.151
(d, Jˆ7.2 Hz). HRMS (EI): Calcd. for C13H23O5F3
316.1498, found 316.1402.
[1] M. Nogradi, Stereoselective Synthesis, VCH, Weinheim, 1994.
[2] Y. Takeuchi, N. Itoh, H. Note, T. Koizumi, K. Yamaguchi, J. Am.
Chem. Soc. 113 (1991) 6318.
[3] Y. Takeuchi, N. Itoh, T. Koizumi, J. Chem. Soc., Chem. Commun.
(1992) 1514.
[4] Y. Takeuchi, N. Itoh, T. Satoh, T. Koizumi, K. Yamaguchi, J. Org.
Chem. 58 (1993) 1812.
2-Octyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
[5] Y. Takeuchi, K. Konishi, H. Hori, T. Takahashi, T. Kometani, K.L.
Kirk, J. Chem. Soc., Chem. Commun. (1998) 365.
[6] F. Yasuhara, M. Takeda, Y. Ochiai, S. Miyano, S. Yamaguchi, Chem.
Lett. (1992) 251.
propionate (4d): Yield 70%. H-NMR ꢁ 0.882 and 0.886
(3H, t, Jˆ6.8 Hz), 1.152 and 1.165 (3H, d, Jˆ6.0 Hz) 1.27±
1.32 (10H, m), 3.404 and 3.412 (3H, s), 3.57-3.59 (2H, m),
3.71±3.77 (2H, m), 4.37 and 4.445 (1H, q, Jˆ7.2 Hz) 4.75
(2H, s), 5.10 (1H, m). 19F-NMR ꢁ 3.108 and 3.220 (d,
Jˆ7.2 Hz). HRMS (EI): Calcd. for C15H27O5F3 344.1811,
found 344.1704.
[7] H. Kawa, N. Ishikawa, Chem. Lett. (1980) 843.
[8] H. Kawa, F. Yamaguchi, N. Ishikawa, Chem. Lett. (1982) 745.
[9] R.E. Doolittle, P.R. Heath, J. Org. Chem. 49 (1984) 5041.
[10] J.A. Dale, H.S. Mosher, J. Am. Chem. Soc. 95 (1973) 512.
[11] T. Kubota, Y. Kondoh, T. Ohyama, T. Tanaka, Nippon Kagaku Kaishi
(1989) 1576.
2-Decyl 3,3,3-tri¯uoro-2-[(2-methoxyethoxy)methoxy]-
1
propionate (4e): Yield 70%. H-NMR ꢁ 0.879 and 0.882
(3H, t, Jˆ5.8 Hz), 1.254 and 1.274 (3H, d, Jˆ6.4 Hz) 1.28±
[12] T. Kubota, M. Iijima, T. Tanaka, Tetrahedron Lett. 33 (1992)
1351.

T. Kubota et al. / Journal of Fluorine Chemistry 97 (1999) 213±221
221
[13] C. von dem Bussche-Hunnefeld, C. Cescato, D. Seebach, Chem. Ber.
125 (1992) 2795.
[16] T. Katagiri, C. Yoda, K. Furuhashi, K. Ueki, T. Kubota, Chem. Lett.
(1996) 115.
[17] P.V. Ramachandran, B.Q. Gong, H.C. Brown, J. Org. Chem. 60
(1995) 41.
[14] K. Furuhashi, in: A.N. Colons, G.N. Sheldreke, J. Crosby (Eds.),
Chirality in Industry, Wiley, New York, 1992, p. 167.
[15] T. Katagiri, F. Obara, S. Toda, K. Furuhashi, Synlett. (1994) 507.