Practical asymmetric synthesis of trifluoromethyl-containing aminoester using a modified davis protocol

Article abstract of DOI:10.1021/op700259d

Practical synthesis of aminoester 1 starting from 1,1,1-trifluoro-3- iodopropane is presented. Use of Ti(Oi-Pr)4 as a Lewis acid for condensation of intermediate aldehyde 8 with (S)-(+)-p-toluenesulfinamide was found to be critical. Conditions for a reproducible and high-yielding Wittig reaction of aldehyde hydrate with phosphorus ylide 4, that appear to have general applicability, are described.

Full text of DOI:10.1021/op700259d

Organic Process Research & Development 2008, 12, 424–428
Practical Asymmetric Synthesis of Trifluoromethyl-Containing Aminoester Using a
Modified Davis Protocol
Asaf Alimardanov,* Jean Schmid, Jay Afragola, and Gulnaz Khafizova
Wyeth Research, 401 North Middletown Road, Pearl RiVer, New York 10965, U.S.A.
Abstract:
Results and Discussion
The presence of trifluoroethyl substituents in 1 imposes
limitations on synthetic approaches to the target. The strong
electron-withdrawing nature of a trifluoromethyl group signifi-
cantly deactivates trifluoroethyl halides or triflates towards
nucleophilic substitution by corresponding enolates. At the same
time, the instability of trifluoroethyl metal species hinders
possible cross-coupling approaches with dibromodehydroamino
acid derivatives followed by asymmetric hydrogenation. An
approach relying on condensation or Wittig olefination of
1,1,1,5,5,5-hexafluoropentan-3-one was also considered and
ruled out. Although the ketone is a known compound,3 its
reported synthesis relies on the isolation of a minor component
of a mixture for a key intermediate, requires specialized
equipment, and poses potential safety concerns. Therefore, the
decision was made to use the reported sequence,1b relying on
conversion of ester 6 to 2-(2,2,2-trifluoroethyl)-4,4,4-trifluo-
robutyl alcohol and further to aldehyde 8 as a starting point for
development of a scaleable synthesis.
The low-yielding sequence of steps in the reported
approach,1b reduction of ester 6 to alcohol, oxidation of alcohol
to aldehyde 8, and formation of sulfinimine, was perceived to
be due to volatility of the intermediate alcohol and the aldehyde.
The synthetic scheme was therefore modified to avoid the
alcohol intermediate and use the higher-boiling Weinreb amide
7 instead. The synthesis used to access ester 1 is shown in
Scheme 1.
r-Phosphanilydene Ester Preparation. Interestingly, on
larger scale, the conversion of iodide 2 to phosphonium salt 3
did not proceed in the reported high yield under the described
conditions (toluene, 110 °C),1b giving only 40–45% yield of 3.
Such an outcome was presumably a result of substrate loss over
the reaction time (28 h) due to its low boiling point (55 °C)
relative to the reaction temperature. This issue could be resolved
either by utilizing a sealed vessel, or by accelerating the reaction
by increasing the concentration of the less expensive reagent,
Ph3P. Not surprisingly, linear dependence between number of
equivalents of Ph3P and the yield of phosphonium salt was
observed, with 3 forming in 92–96% yield when 3 equiv of
Ph3P were used.
Practical synthesis of aminoester 1 starting from 1,1,1-trifluoro-
3-iodopropane is presented. Use of Ti(Oi-Pr)4 as a Lewis acid for
condensation of intermediate aldehyde 8 with (S)-(+)-p-toluene-
sulfinamide was found to be critical. Conditions for a reproducible
and high-yielding Wittig reaction of aldehyde hydrate with
phosphorus ylide 4, that appear to have general applicability, are
described.
Introduction
L-Bis(trifluoromethyl)valine methyl ester (1) (Scheme 1), an
unnatural amino acid derivative, is an intermediate of interest
for Medicinal Chemistry programs,1 as it offers the possibility
to introduce a biologically unique and metabolically stable
pharmacophore. Figure 1
The synthesis of 1 has been described.1b The reported
approach provides for the preparation of 1 in 2.5% overall yield,2
allowing access to milligram quantities of the substrate. As quick
access to multigram quantities of ester 1 was needed, rapid
development of a more practical and scaleable protocol allowing
larger-scale synthesis was required.
During our studies of the synthesis of this intermediate we
have observed a lack of critical information in the synthetic
literature on parameters influencing several key steps of our
selected approach, which relies on seemingly well-established
transformations. Among the issues encountered were the
following: developing a protocol for efficient condensation of
an aldehyde with a sulfinamide chiral auxiliary under milder-
than-reported conditions, developing a practical protocol for
the selective Wittig reaction of a phosphorus ylide with an
aldehyde hydrate, and determining critical parameters affecting
the conversion of a phosphonium halide to an R-phosphanily-
dene ester reproducibly.
This communication describes our findings in these areas
that led to the development of a practical, industrially applicable
synthesis of L-bis(trifluoromethyl)valine methyl ester. A discus-
sion of critical parameters affecting key transformations, and
reproducibility and scaleability of the preparation of 1 is
provided.
Conversion of phosphonium salt 3 to ylide 4 was reported
to proceed with 80% yield. Surprisingly, under controlled
conditions using the reported protocol (sequential addition of
LiHMDS and ethyl chloroformate at -78 °C) alkoxycarbony-
lation proceeded with variable yields on larger scale, typically
* Author to whom correspondence may be sent. E-mail: alimara@wyeth.com.
(1) (a) Kreft, A. F.; Resnick, L.; Mayer, S. C., Diamantidis, G.; Cole, D.
C.; Harrison, B. L.; Zhang, M.; Hoke, M.; Ting, W.; Galante, R. C.
U.S. Patent 20040198778, 2004; Chem. Abstr. 2004, 141, 314147.
(b) Zhang, M.; Porte, A.; Diamantidis, G.; Sogi, K.; Kubrak, D.;
Resnick, L.; Mayer, S. C.; Wang, Z.; Kreft, A. F.; Harrison, B. L.
Bioorg. Med. Chem. Lett. 2007, 17, 2401.
(3) (a) Van Der Puy, M. J. Org. Chem. 1997, 62, 6466. (b) Van Der Puy,
M. U.S. Patent 6,072,088; Chem. Abstr. 2000, 130, 15138. (c)) Van
Der Puy, M.; Madhavan, G. V. B.; Bemmin, T. R. U.S. Patent
5,395,997; Chem. Abstr. 1995, 122, 239173.
(2) Estimated 80% ee.
424
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development
10.1021/op700259d CCC: $40.75
2008 American Chemical Society
Published on Web 05/01/2008

Scheme 1
giving 4 in 32–47% yield.4 A significant amount of urethane
was formed as a byproduct.
monitored by complete dissolution).5 Subsequent addition of
ethyl chloroformate at -78 °C afforded ester 4, now reproduc-
ibly in up to 89% yield on ∼200 g scale.
Although conversion of iodide 3 to ylide 4 is a one-pot
process, there are several successive individual processes that
have to occur in sequence for the transformation to proceed to
completion (Scheme 2), including: deprotonation to form
intermediate ylide 11, nucleophilic substitution of ethyl chlo-
roformate to form intermediate salt 12, and deprotonation of
12 with 11 (transylidation) or LiHMDS. Interestingly, incom-
plete dissolution of 3 even after prolonged time was observed
in low-yielding cases, which pointed to solubility as a potential
cause of yield irreproducibility. Poor solubility of salt 3 would
cause inefficient deprotonation and incomplete conversion to
the intermediate ylide 11, which would translate into incomplete
conversion to ylide 4. Unreacted LiHMDS remaining in the
system would then slowly react with ethyl chloroformate,
forming the urethane byproduct after hydrolysis. The efficiency
of intermediate 11 formation was found to be dependent on
the temperature of initial deprotonation.
The reproducibility issue described above may possibly be
unnoticed during smaller-scale experiments, when the rate of
LiHMDS addition (exothermic process) and, consequently,
internal temperature, are not carefully controlled (uncontrolled
exotherm during initial deprotonation may inadvertently drive
the reaction to completion). However, on larger-scale experi-
ments, when strict control over reaction parameters is always
exercised, exact determination of the parameters is crucial for
successful conversion.
Wittig Olefination of Trifluoroacetaldehyde Hydrate. The
following Wittig olefination of ylide 4 with commercially
available trifluoroacetaldehyde hydrate in a sealed vessel in THF
at 100 °C indeed proceeded to yield the desired olefin 5.1b
However, 25–30% of a byproduct, ethyl trifluorobutyrate 13,
forms along with 5 under these conditions, as determined by
NMR.
Simply raising the temperature of deprotonation to -5 to
5 °C range ensured complete consumption of 3 (can be
(5) (a) Ullmannn, J.; Hanack, M. Synthesis 1989, 685. (b) Morimoto, Y.;
Shirahama, H. Tetrahedron 1996, 52, 10631. (c) Bestman, H. J.;
Schulz, H. Liebigs Ann. Chem. 1964, 674, 11. (d) Isler, O.; Gutmann,
H.; Montavon, M.; Ruegg, R.; Ryser, G.; Zeller, P. HelV. Chim. Acta
1957, 40, 1242.
(4) In one instance 81% was observed on a smaller, 5-g scale.
Vol. 12, No. 3, 2008 / Organic Process Research & Development
•
425

Scheme 2
Figure 1. Byproducts in the formation of 5, 9.
Phosphorus ylides are known to be susceptible to hydrolysis,6
which is catalyzed by acids. Such a hydrolysis would result in
formation of impurity 13. It should be noted that trifluoroac-
etaldehyde, being a gas in its pure form at room temperature,
is commercially available as its acetal (hydrate). Dissociation
of the acetal to form free aldehyde is needed for the Wittig
reaction to proceed, hence necessitating high temperature and
use of a sealed vessel. Water produced from the acetal is
responsible for ylide hydrolysis. Interestingly, it was found that
commercial trifluoroacetaldehyde hydrate (acetal) contains not
only water of the hydrate but also unbound water as well as
residual HCl and HF, which would catalyze formation of
impurity 13. Therefore, critical for this transformation is efficient
removal of water and any acidic impurities from the reaction
medium.
It was found that by incorporating pretreatment of trifluo-
roacetaldehyde hydrate with NaHCO3 and MgSO4, running the
Wittig reaction in the presence of a water-consuming agent
(MgSO4), and switching solvents to the less water-miscible
MTBE (to increase MgSO4 efficiency), one could generate acid-
free anhydrous trifluoroacetaldehyde in situ.7 This, in turn,
suppressed formation of impurity 13 to a level below 2%,
yielding 5 reproducibly in 87% isolated yield.
Hydrogenation of olefin 5 afforded saturated ester 6, which
was used further without isolation as a THF solution. As the
direct reduction of ester 6 to the aldehyde was not selective, 6
was converted to Weinreb amide 7 using the standard protocol.8
DIBAL reduction to aldehyde 8 proceeded cleanly at -70 °C.
The dichloromethane solution of 8, after phase split and aqueous
wash, was dried and used further as such due to volatility
concern. The yield of aldehyde 8 was determined by NMR after
adding a known amount of mesitylene as an internal standard
to the stock solution and integrating corresponding peaks.
Sulfinimine Formation. With aldehyde 8 available, the
introduction of chiral auxiliary was studied. Conversion of
aldehydes to p-toluenesulfinimines is well documented.9 How-
ever, using standard conditions described in literature ((S)-p-
toluenesulfinamide and Ti(OEt)4 in CH2Cl2 or Et2O), sulfinimine
9 could only be obtained in 13–34% yield, along with numerous
byproducts, regardless of temperature or concentration, and
corresponding to results reported for this substrate.1b Alternative
reported methods of preparation of sulfinimines (4 Å molecular
sieves, Py-TsOH)7 did not offer any advantages. It should be
noted that such a low yield is not due to aldehyde volatility,
instability, or incomplete consumption. Aldehyde 8 was found
to be stable as a CH2Cl2 solution at room temperature for weeks,
and was fully consumed under condensation conditions.
Examination of the product mixture by LC/MS suggested
formation of 14 and 15 as major byproducts of condensation,
present in up to 31% and 6%, respectively. The amount of both
of these byproducts increased with reaction temperature and
time.
The presence of impurity 14 pointed to involvement of
Ti(OEt)4 in the byproduct formation. The Ti(OEt)4 additive
plays a dual role in aldehyde condensation with sulfinamide,
acting as a water-consuming agent and as a Lewis acid to
activate the aldehyde towards nucleophilic addition. However,
Ti(OEt)4 activates condensation product 9 towards nucleophilic
addition as well, while also acting as a source of ethoxide
nucleophile to generate byproduct 14. It was reasoned that
pivotal for the success of this condensation would be an additive
possessing all following characteristics: being a water-consum-
ing agent; being a Lewis acid, but a less active one than
Ti(OEt)4; and, if such an additive is a metal alkoxide, having a
bulkier and less nucleophilic alkoxide group. Indeed, the use
of Ti(Oi-Pr)4 offered significant advantage, allowing to obtain
sulfinimine 9 in 90% isolated yield as a single product. Other
critical parameters for this reaction include temperature (40 °C)
and concentration (∼0.1 M). At high concentrations alkoxide
addition byproduct would be formed even with Ti(Oi-Pr)4.
Impurity 15 was presumably formed by tautomerization of
9 under prolonged reaction conditions, and its formation was
suppressed using the newly developed protocol.
(6) (a) Bestmann, H. J. Angew. Chem., Int. Ed. 1965, 4, 583. (b) Bestmann,
H. J. Angew. Chem., Int. Ed. 1965, 4, 645.
Strecker reaction using Et2AlCN-i-PrOH proceeded as
expected,10 giving 10 in 90% yield as a 9:1 mixture of
diastereomers, which was improved to a 98:2 ratio by recrys-
(7) For peparation and low-temperature trapping of anhydrous trifluoro-
acetaldehyde, see: (a) Shechter, H.; Conrad, F. J. Am. Chem. Soc.
1950, 72, 3371. (b) Husted, D. R.; Achlbrecht, A. H. J. Am. Chem.
Soc. 1952, 74, 5422. (c) Negishi, J.; Kaneda, S.; Yamamoto, Y.;
Sugimori, Y.; Hags, Y.; Morino, Y. GB 2260322, 1992.
(10) (a) Davis, F. A.; Reddy, R. E.; Portonovo, P. S. Tetrahedron Lett.
1994, 35, 9351. (b) Davis, F. A.; Portonovo, P. S.; Reddy, R. E.; Chiu,
Y. J. Org. Chem. 1996, 61, 440. (c) Davis, F. A.; Fanelli, D. L. J.
Org. Chem. 1998, 63, 1981.
(8) Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini, G.; Dolling,
U.-H.; Grabowski, E. J. J. Tetrahedron Lett. 1995, 36, 5461.
(9) (a) Davis, F. A.; Zhang, Y.; Andemichael, Y.; Fang, T.; Fanelli, D. L.;
Zhang., H. J. Org. Chem. 1999, 64, 1403. (b) Davis, F. A.; Reddy,
R. E.; Szewszyk, J. M.; Portonovo, P. S. Tetrahedron Lett. 1993, 34,
6229. (c) Davis, F. A.; Reddy, R. E.; Szewszyk, J. M.; Reddy, G. V.;
Portonovo, P. S.; Zhang, H.; Fanelli, D.; Reddy, R. T.; Zhou, P.;
Carroll, P. J. J. Org. Chem. 1997, 62, 2555.
(11) For an explanation of the sense of stereoinduction, see refs 10a and b.
(12) Ester 1 was converted to several pharmacologically active derivatives
of interest in two steps. Enantiomeric purity was determined for one
of the derivatives purified by recrystallization and was found to be
>99% ee by chiral HPLC.
426
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development

tallization (73% yield).11 Subsequent hydrolysis and esterifica-
tion in MeOH/HCl afforded 1 in 10 steps and 17% overall
yield.12
4 , 4 , 4 - T r i fl u o r o - 2 - ( 2 , 2 , 2 - t r i fl u o r o e t h y l ) -
but-2-enoic Acid Ethyl Ester (5). Trifluoroacetaldehyde
hydrate (150 g, technical grade, pH 1) was stirred with
anhydrous sodium bicarbonate (15 g, powder) to result in a
mildly foaming suspension. Anhydrous magnesium sulfate
powder (60 g) was added, followed by addition of MTBE (300
mL) to result in a mildly exothermic reaction. The suspension
was kept in a water bath at 10 °C for 10 min, filtered through
a fluted filter funnel, and washed with MTBE (2 × 250 mL).
The filtrate (pH 7.2) was charged into a 2 L “Parr” pressure
reactor containing the starting ylide 4 (204 g, 0.474 mol). To
the mixture was added anhydrous magnesium sulfate powder
(60 g). The reaction vessel was heated to 70-75 °C with stirring
for 15 h. The pressure in the “Parr” reactor rose to 18–21 psi.
The reaction was cooled to room temperature, and the mixture
was filtered. The filter cake was washed with MTBE. The
filtrate was distilled at 60–70 mm/Hg to remove most of the
MTBE in the first fraction, and the remainder was collected in
the second fraction. The pressure for the second fraction was
reduced to 20 mm/Hg to yield 121.7 g. The second fraction
(121.7 g) was redistilled at 20 mm/Hg with a bath temperature
at 80 °C to yield a main fraction of a low viscosity liquid (103.5
g, yield 87%, bp 53–55 °C at 20 mmHg). 1H NMR (300 MHz,
CDCl3): δ 6.95 (q, 1 H, J ) 8 Hz), 4.33 (q, 2 H, J ) 7.3 Hz),
3.50 (q, 2 H, J ) 9.9 Hz), 1.35 (t, 3 H, J ) 7.3 Hz). MS (M
+ H): 251.
4,4,4-Trifluoro-2-(2,2,2-trifluoroethyl)butyric Acid Ethyl
Ester (6). 4,4,4-Trifluoro-2-(2,2,2-trifluoroethyl)but-2-enoic acid
ethyl ester 5 (225 g, 0.9 mol) was dissolved in tetrahydrofuran
(700 mL) and treated with 5% Pd/C (17 g). The mixture was
reduced by hydrogenation in a “Parr” shaker in a 2.5 L pressure
bottle at 50 psi. The reaction is exothermic to 45 °C and is
controlled by interrupting the shaking motion of the “Parr”
shaker. The reaction was completed in approximately 2 h (as
judged by NMR). The reaction mixture was filtered through a
2-in. bed of “Solka Floc”/magnesium sulfate to give a clear,
colorless solution of the title compound (225 g in 1182 g of
tetrahydrofuran, quantitative yield). 1H NMR (300 MHz,
CDCl3): δ 4.14 (q, 2 H, J ) 7.2 Hz), 2.99–2.89 (m, 1 H),
2.68–2.48 (m, 2 H), 2.4–2.21 (m, 2 H), 1.22 (t, 3H, J ) 7.2
Hz). MS (M + H): 253.
4,4,4-Trifluoro-N-methoxy-N-methyl-2-(2,2,2-trifluoro-
ethyl)butyramide (7). N,O-Dimethylhydroxylamine hydro-
chloride (90 g, 0.92 mol) was added to a solution of 4,4,4-
trifluoro-2-(2,2,2-trifluoroethyl)butyric acid ethyl ester 6 (116.28
g, 0.46 mol) in tetrahydrofuran (610.8 g weight of the solution).
The mixture was cooled to -15 to -20 °C with a dry ice/
acetone bath. To the reaction mixture was added dropwise a
solution of i-propylmagnesium chloride (924 mL, 2 M in
tetrahydrofuran, 1.848 mol) over a period of 1 h, keeping the
temperature at -15 to -20 °C. After the addition, the reaction
was stirred at that temperature for 30 min. The reaction was
quenched by adding dropwise hydrochloric acid (2 N, 600 mL,
1.2 mol). The reaction proceeds at a very rapid rate, ac-
companied by a large exothermic temperature excursion during
addition of the first 50 mL. The temperature did not exceed
3 °C resulting in a thick suspension first, subsequently becoming
a clear solution with two layers. The mixture was extracted with
Conclusions
In summary, milder and more practical protocols were
established for condensation of aldehyde 8 with a sulfinamide
chiral auxiliary, and for selective Wittig reaction of phosphorus
ylide 4 with an aldehyde hydrate. Both protocols appear to have
general applicability. Critical parameters affecting key steps of
described synthesis were determined. As a result, a scaleable
and industrially applicable selective synthesis of ester 1 was
developed, allowing easy access to multigram quantities of the
substrate.
Experimental Section
(3,3,3-Trifluoropropyl)triphenylphosphonium Iodide (3).
A solution of 1,1,1-trifluoro-3-iodopropane 2 (223.96 g, 1 mol)
and triphenylphosphine (786.87 g, 3 mol) was prepared in
toluene (800 mL). This solution was stirred at reflux for 12 h.
The solid product precipitated from the reaction mixture
throughout the course of the reaction. The reaction was allowed
to cool to ambient temperature and then cooled to ∼5 °C in an
ice bath. The solid precipitate was isolated by filtration and dried
in vacuo at 25 °C to give a white powder (461.94 g, 95% yield).
1H NMR (300 MHz, CDCl3) δ 7.90–7.85 (m, 9H), 7.78–7.74
(m, 6H), 4.09–4.01 (m, 2H), 2.71–2.57 (m, 2H). Anal. Calcd
for C21H19F3IP: C, 51.87; H, 3.94. Found C, 51.99; H, 3.90.
4,4,4-Trifluoro-2-(triphenyl-λ5-phosphanylidene)bu-
tyric Acid Ethyl Ester (4). A suspension of (3,3,3-trifluoro-
propyl)triphenylphosphonium iodide 3 (194.5 g, 0.4 mol) in
tetrahydrofuran (anhydrous, 800 mL) was cooled to -5 °C in
an ice/brine bath under nitrogen. To this suspension, lithium
bis(trimethylsilyl)amide (1.0 M in THF, 800 mL, 0.8 mol) was
added dropwise over 2 h. The temperature was maintained
below 5 °C throughout the addition. The reaction mixture was
then cooled to -75 °C in a dry ice/acetone bath. To this
solution, ethylchloroformate (76.5 mL, 0.8 mol) was added
dropwise over 30 min. The reaction was stirred at -75 °C for
an additional hour and allowed to warm to room temperature
overnight. The reaction mixture was poured onto brine (1.5 L)
and stirred for 30 min. The layers were separated, and the
organic layer was washed with brine (200 mL). The aqueous
layer was washed with methylene chloride (2 × 200 mL), and
the combined organics were concentrated to a residue. This
residue was redissolved in methylene chloride (500 mL), dried
over MgSO4, and filtered through a plug of magnesol. The
solvent was reduced to a minimum (∼100 mL) in vacuo, and
the product was precipitated with hexanes (250 mL). The
solvent was completely removed in vacuo, and the solid product
was triturated in hexanes (500 mL). The solid was isolated by
filtration and dried overnight in vacuo at 25 °C to give a beige
powder (152.8 g, 89% yield). 1H NMR (300 MHz, CDCl3, for
two rotamers) δ 7.66–7.45 (m, 15H), 4.05 and 3.70 (q, 2H, J
) 7.2 Hz), 2.81–2.64 (m, 2H), 1.21 and 0.41 (t, 3H, J ) 7.2
Hz); MS (M + H): 431. Anal. Calcd: C, 66.97; H, 5.15. Found:
C, 66.37; H, 5.28.
Vol. 12, No. 3, 2008 / Organic Process Research & Development
•
427

MTBE (1.5 L). The aqueous phase was re-extracted with MTBE
(0.5 L). The combined organic extracts were washed with brine
(2 × 0.5 L). The organic phase was dried over anhydrous
magnesium sulfate powder and filtered, and the filtrate was
concentrated in vacuo at max. 35 °C to provide 121 g of an
oil. The oil was distilled at 15 mmHg/bp 64–68 °C to give the
title compound as oil (120 g, yield 87%). 1H NMR (300 MHz,
CDCl3): δ 3.735 (s, 3 H), 3.6–3.5 (m, 1 H), 3.22 (s, 3 H),
2.76–2.60 (m, 2 H), 2.35–2.2 (m, 2 H). MS (M + H): 268.
4,4,4-Trifluoro-2-(2,2,2-trifluoroethyl)butyraldehyde (8).
To a solution of 4,4,4-trifluoro-N-methoxy-N-methyl-2-(2,2,2-
trifluoroethyl)-butyramide (53.4 g, 0.2 mol) in dichloromethane
(300 mL) was added diisobutylaluminum hydride (58 mL, 0.32
mol) over 20 min at -68 to -62 °C. The reaction mixture was
stirred at -70 °C for 1 h and then poured into 1.5 L of ice
containing 500 mL of 6 N HCl. The phases were split, and the
aqueous phase was extracted with 300 mL of dichloromethane.
The combined organic phase was washed with 1 N HCl and
dried over MgSO4. The solution was filtered through a 2.5-cm
pad of silica gel. The silica gel pad was washed with dichlo-
romethane. The total volume of the solution: 2 L. NMR analysis
of the solution using an internal standard (mesitylene) indicated
formation of the title product in 62.5% yield. The solution was
stored over 4 Å molecular sieves and used as such for further
filtered through Celite. The Celite pad was washed with MTBE.
The phases were separated, and the aqueous phase was extracted
with MTBE. The combined organic fraction was dried over
MgSO4 and filtered through 2.5-cm pad of silica gel. The silica
gel pad was washed with MTBE. The combined solution was
concentrated to an oil, which was treated with heptane. The
resultant white solid was filtered and washed with heptane to
afford 17.2 g of the title product as a 9:1 mixture of diastere-
omers (90% yield). The crude material was recrystallized from
25 mL of hot MTBE and 70 mL of heptane, filtered, washed
with heptane to afford 14 g of the title product (95% de, 73%
yield). 1H NMR (300 MHz, CDCl3; major diastereomer): δ 7.56
(d, 2 H, J ) 8.2 Hz), 7.37 (d, 2 H, J ) 8.2 Hz), 5.6 (d, 1 H, J
) 8.6 Hz), 4.33 (dd, 1 H, J ) 8.6, 4.3 Hz), 2.77–2.55 (m, 2
H), 2.5–2.2 (m, 6 H). MS (M + H): 373.
(2S)-2-Amino-5,5,5-trifluoro-3-(2,2,2-trifluoroethyl)pen-
tanoic Acid Methyl Ester (1). 4-Methyl-N-[(1S)-1-cyano-4,4,4-
trifluoro-2-(2,2,2-trifluoroethyl)butyl]-(S)-benzenesulfinimide 10
(10 g, 29 mmol) was dissolved in concentrated hydrochloric
acid (200 mL), and the mixture was heated under reflux for
15 h. The reaction was cooled to room temperature. A
byproduct, toluene-4-thiosulfonic acid S-p-tolyl ester, separated
from the aqueous solution as a white crystalline solid and was
filtered off. The aqueous filtrate was concentrated in vacuo to
a sticky white solid. The crude amino acid was taken up in
concentrated hydrochloric acid (200 mL) and extracted with
toluene (2 × 50 mL). The aqueous phase was concentrated in
vacuo, coevaporating with toluene (4 × 70 mL) to give a solid
compound. The aminoacid was dissolved in methanol (400 mL),
treated with anhydrous hydrochloric acid (4 N, 100 mL), and
refluxed for 72 h. The reaction was evaporated in vacuo to a
foam (60% ester conversion by NMR). The reaction mixture
was dissolved in methanol (300 mL) and treated with ethereal
hydrochloric acid (2 N, 100 mL) and refluxed for 24 h. The
solution was concentrated to a solid (80% ester conversion by
NMR). The crude mixture was dissolved in water and extracted
with MTBE. The aqueous phase was basified with solid sodium
bicarbonate and extracted with MTBE (2 × 100 mL). The
organic layer was dried over magnesium sulfate, filtered, and
concentrated in vacuo to give the title compound as a solid.
(4.6 g, 62% yield). 1H NMR (300 MHz, CDCl3): δ 3.78 (s, 3
H), 3.75 (d, 1 H, J ) 2.2 Hz), 2.66–2.52 (m, 2 H), 2.35–2.20
(m, 1 H), 2.18–2.06 (m, 2 H). 13C NMR (75 MHz, CDCl3): δ
174.28, 126.67 (q, J ) 278 Hz), 126.45 (q, J ) 278 Hz), 54.91,
52.58, 34.58 (q, J ) 29 Hz), 33.41 (q, J ) 29 Hz), 31.40. Anal.
Calcd for C8H11F6NO2: C 35.96, H 4.15, N 5.24; found: C
36.18, H 4.20, N 5.27. HRMS (for M + H) calcd: 268.07668;
found: 268.07582. [R]D +11.2 (c ) 1, MeOH).
1
transformations. H NMR (300 MHz, CDCl3; solvent and
mesitylene peaks excluded): δ 9.74 (d, 2 H, J ) 1.2 Hz),
3.02–2.98 (m, 1 H), 2.80–2.62 (m, 2 H), 2.46–2.36 (m, 2 H).
(S)-4-Methyl-N-[(1Z)-4,4,4-trifluoro-2-(2,2,2-trifluoro-
ethyl)butylidene]benzenesulfinamide (9). To a dichlorometh-
ane solution of 4,4,4-trifluoro-2-(2,2,2-trifluoroethyl)butyral-
dehyde 8 (1000 mL; contains 62.5 mmol of the aldehyde;
prepared as described above; dried over 4 Å molecular sieves)
were added titanium isopropoxide (95 mL, 314.4 mmol, 97%
pure) and (S)-(+)-p-toluenesulfinamide (11.86 g, 76.4 mmol,
1.2 equiv). The reaction mixture was stirred at 40 °C for 5.5 h,
cooled to rt, and poured into ice–water (400 mL) at 0 °C. The
mixture was stirred at room temperature for 1 h, then filtered
through Celite. Phases were separated, the aqueous phase was
extracted with dichloromethane. The combined organic fraction
was washed with brine and dried over MgSO4. The resultant
mixture was filtered through a 5-cm pad of silica gel and con-
centrated to afford 19.2 g (90%) of the title product. 1H NMR
(300 MHz, CDCl3): δ 8.24 (d, 1 H, J ) 3.8 Hz), 7.53 (d, 2 H,
J ) 8.2 Hz), 7.31 (d, 2 H, J ) 8.2 Hz), 3.28–3.15 (m, 1 H),
2.7–2.53 (m, 2 H), 2.52–2.32 (m, 5 H). MS (M + H): 346.
4-Methyl-N-[(1S)-1-cyano-4,4,4-trifluoro-2-(2,2,2-trifluoro-
ethyl)butyl]-(S)-benzenesulfinamide (10). To a mixture of
THF (110 mL) and diethylaluminum cyanide (112 mL of 1 M
toluene solution, 112 mmol) was added isopropanol (8.7 mL,
113 mmol) over 5 min at 2-4 °C. The mixture was stirred at
that temperature for 1 h and then cooled to -65 °C. A solution
of (S)-4-methyl-N-[(1Z)-4,4,4-trifluoro-2-(2,2,2-trifluoroethyl-
)butylidene]benzenesulfinamide 9 (18.5 g, 56.2 mmol) in THF
(250 mL) was added over 30 min at -65 to -60 °C. The
reaction mixture was stirred at that temperature for 15 min and
then allowed to slowly warm up to 0 °C. The mixture was
stirred at 0 °C for 3 h and then poured onto ice–water (1200
mL) containing NH4Cl (150 g). The resultant suspension was
Acknowledgment
We thank Dr. K. Tabei for help with LC/MS analysis.
Supporting Information Available
1H and 13C NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
Received for review November 12, 2007.
OP700259D
428
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development