The Mitsunobu reaction in the synthesis of α,α-difluoro-β-amino acids

Article abstract of DOI:10.1016/S0022-1139(01)00430-4

A three-stage strategy is proposed to prepare α,α-difluoro-β-amino acids starting from aldehydes and with ethyl bromodifluoroacetate as a fluorine source. The Mitsunobu reaction as a key step was studied by ³¹P and ¹⁹F NMR for alkyl-

Full text of DOI:10.1016/S0022-1139(01)00430-4

Journal of Fluorine Chemistry 111 >2001) 69±76
The Mitsunobu reaction in the synthesis of a,a-di¯uoro-b-amino acids
Natalie A. Fokina*, Andrei M. Kornilov, Valery P. Kukhar
Institute of Bioorganic Chemistry and Petrochemistry NAS of Ukraine, Murmanskaya 1, Kyiv 02094, Ukraine
Received 4 October 2000; accepted 15 June 2001
Abstract
A three-stage strategy is proposed to prepare a,a-di¯uoro-b-amino acids starting from aldehydes and with ethyl bromodi¯uoroacetate as
a ¯uorine source. The Mitsunobu reaction as a key step was studied by 31Pand 19F NMR for alkyl- and aryl-substituted a,a-di¯uoro-b-
hydroxyesters and performed under optimized conditions giving the target compounds. # 2001 Elsevier Science B.V. All rights reserved.
Keywords: Ethyl bromodi¯uoroacetate; Aldehydes; The Mitsunobu reaction; a,a-Di¯uoro-b-amino acids; NMR spectroscopy
1. Introduction
starting from aliphatic and aromatic aldehydes 1a±e
>Scheme 1). In order to obtain the desired esters of a,a-
di¯uoro-b-hydroxy acids, the Reformatsky reaction of ethyl
bromodi¯uoroacetate with aldehydes 1a±e was accom-
plished affording the corresponding products 2a±e in high
preparative yields >81±92%) [9]. The key step of the present
methodology is the conversion of hydroxyesters 2a±e to
nitrogen-containing derivatives 3a±e.
An increasing interest in b-amino acids arises from their
ability to exhibit interesting pharmacological effects and
their utility as components of peptides with enhanced enzy-
matic stability compared with a-amino acids [1]. The repla-
cement of hydrogen by ¯uorine in amino acids, modifying
theirphysical, chemicalandsteric properties, causesimprove-
ment or alteration of their biological activity [2]. The present
work was undertaken to investigate a new route to b-amino
acids with a di¯uoromethylene group in the a-position.
The Mitsunobu reaction is known to be a useful tool for
the replacing a hydroxyl group by different nucleophiles [3].
A number of nitrogen nucleophiles have been utilized in this
procedure to afford amino derivatives [3±6]. Being inter-
ested in an ef®cient approach to biologically important a,a-
di¯uoro-b-amino acids from easily accessible materials, we
turned our attention to this method. We also take into
account that the Mitsunobu reaction mostly occurs under
mild conditions and is often employed in the synthesis and
further transformations of various classes of natural pro-
ducts. The Mitsunobu methodology have been applied for
amino acids synthesis [7,8], but there were no examples of b-
amino acids among them.
Despite that the Mitsunobu reaction is widely used in
synthetic organic chemistry, the mechanistic details of this
reaction are still a subject of debates and intensive studies.
This caused by the fact that depending on the order of
addition of reactants and type of substrates, different reac-
tion pathways may take place. In the current study several
factors were examined in order to improve the overall
ef®ciency of the method for our ¯uorine-containing sub-
strates 2a±e. The reaction was conducted using a standard
redox system, i.e. diethyl azodicarboxylate >DEAD)/triphe-
nylphosphine. Phthalimide was chosen as a most commonly
used nucleophilic component for the amino function intro-
1
duction through the Mitsunobu procedure [7,10]. 31Pand H
NMR spectroscopy is usually used for mechanistic studies of
the Mitsunobu reaction; the presence of the ¯uorine atom in
the initial hydroxy esters 2 allowed us to gain additional
information from the 19F NMR data.
Initially, we checked whether THF, almost always used as
a solvent [7,11±13] in the Mitsunobu reaction, is relevant in
our case for the amination step >Scheme 1), and compared
the reaction yields with those of other solvents. In these
experiments a 20% solution of DEAD 5 was added dropwise
2. Results and discussion
Herein, we describe the three-stage strategy, which allows
an ef®cient preparation of a,a-di¯uoro-b-amino acids
at 08C to the mixture of the hydroxyester 2a, P hP 6, and
phthalimide 7 under argon, and the reaction was left to stir at
room temperature. The results are summarised in Table 1.
3
* Corresponding author. Fax: ‡38-44573-2552.
E-mail address: natfokina@yahoo.com >N.A. Fokina).
0022-1139/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved.
P II: S 0 0 2 2 - 1 1 3 9 > 0 1 ) 0 0 4 3 0 - 4

70
N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
Scheme 1.
Table 1
Yields of the Mitsunobu reaction for hydroxyester 2a in different solvents
Solvent
Yielda 3a >%)
Unconverted starting
materiala 2a >%)
Molar ratio,
ROH:DEADb:Ph3P:PhthNHb
CH2Cl2
THF
26
66
79
25
30
16
1:1:1:1
1:1:1:1
1:1:1:1
Toluene
a Estimated from 19F NMR data; measured after 20 h.
b DEAD: diethyl azodicarboxylate; PhthNH: phthalimide.
It is seen that toluene was the best solvent tested, and
therefore it was employed in all further experiments.
Having a moderate conversion in all experiments, we
decided to study the effect of stoichiometry and order of
addition of reactants. The results of these experiments
performed with hydroxyester 2a are presented in Table 2.
It is apparent from Table 2 that method D, when reaction
was carried out in the presence of excess of reagents, and
when DEAD solution was added as a last component to the
reaction mixture, is a most preparatively useful. The Mitsu-
nobu reaction with hydroxyester 2a was complete within an
hour in that case, as monitored by 19F NMR. When phtha-
limide 7 is added last >method F) the conversion is also
found to be high, but yield of product 3a is decreased by By-
products.
obtained the desired product 3d in 32% yield. The same
experiments performed at the temperatures À20 and ‡228C
were less successful.
3->4-Chloro-phenyl)-2,2-di¯uoro-3->ethoxycarbonylazo)-
propanoic acid ethyl ester >8d), formed as a result of compe-
titive DEAD N-alkylation, were found as one of the main
By-products. It should be noted that carrying out this process
with aliphatic hydroxyester 2a by non-optimal methods
A±C, E, F >Table 2) gave the corresponding product 8a in
yield up to 5% according to 19F NMR. In order to understand
the differences in the Mitsunobu reaction proceeding for
alkyl- and aryl-substituted a,a-di¯uoro-b-hydroxyesters we
examined these reactions by 31PNMR and 19F NMR.
Using the best addition order >method D) with the appro-
priate molar ratio of the reagents, we started with aromatic
hydroxyester 2d and obtained quite a low yield and low
conversion to the desired product 3d. After several experi-
ments with different molar excess of the reagents >mixing at
08C, Table 3), we found the optimal reaction conditions, and
The Mitsunobu reaction is generally envisaged as pro-
ceeding through oxyphosphonium salts and phosphoranes.
Table 2
Yields of the Mitsunobu reaction for hydroxyester 2a depending on
addition order and ratio of reactants [n-C5H11CH>OH)CF2CO2Et >2a),
DEAD >5), Ph3P >6) and phthalimide >7)]
Table 3
Yields of the Mitsunobu reaction for hydroxyester 2d
Method Reactant addition
order
Molar ratio
Yielda
Unconvertedb
2a:5:6:7
3a >%) 2a >%)
Experimenta
Molar excess of
the reagentsb
Yieldc
Unconvertedd
3d >%)
2d >%)
A
B
C
D
E
F
2a, 6, 7 ‡ 5
1:1:1:1
79
67
74
95
77
88
16
21
11
3
6 ‡ 7 ‡ 5 ‡ 2a
6 ‡ 2a ‡ 5 ‡ 7
2a, 6, 7 ‡ 5
1:1:1:1
1:1:1:1
1
2
3
1.2
3
10
25
32
46
8
1:1.2:1.2:1.4
1:1.2:1.2:1.4
1:1.2:1.2:1.4
7
0
6 ‡ 7 ‡ 5 ‡ 2a
15
3
a For temperature and solvent see Section 4.12.
b Ph3P, DEAD, PhthNH.
c Isolated yields after 20 h.
d Estimated from 19F NMR data.
6 ‡ 2a ‡ 5 ‡ 7
a Isolated yields after 20 h.
b Estimated from 19F NMR data.

N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
71
Scheme 2.
Scheme 2 shows that on adding the alcohol to the preformed
betaine 9, intermediate O,N-phosphoranes 10, O,O-phos-
phoranes 12 and >or) corresponding alkoxyphosphonium
hydrazide salts 11 and alkoxyphosphonium alkoxide salts
13 can be formed. Phosphoranes 10 have been postulated
[14] to participate in the Mitsunobu reaction, although they
have never been directly observed. Formation of dialkoxy-
triphenylphosphoranes 12 was suggested by Itzstein and
Jenkins [15]. Walker and coworkers suggested that this
compound was an intermediate in the special case when
the acid was added last [11], but Hughes et al. was unable to
observe any phosphoranes even in the absence of an acid
[16]. Addition of the acidic component HX results in the
transformation of these intermediates to alkoxyphospho-
nium salts 14 which can exist in equilibrium with dialkox-
yphosphoranes 12 as has been previously reported [17]. As is
evident from the literature, the possibility to detect by NMR
any of intermediates mentioned above depends on the nature
of the reagents, solvent polarity and the order of addition of
the components. In some cases, when the Mitsunobu reac-
tion was carried out with acidic components with low pKa
>>11), the alkylated hydrazine derivatives 15 were detected
as the main By-products [18].
9 peak being still present together with Ph3PO peak. We
attributed the signal at À59.37 ppm to compound 12 pro-
ceeding from the stoichiometry of the reaction and relative
peaks intensities. As mentioned above, phosphorane 10 was
never observed previously in contrast to compound 12 [15].
Addition of phthalimide >1.5 equivalent) as an acidic com-
ponent resulted in the appearance of an alkoxyphosphonium
salt 14 signal at ‡65.29 ppm [12,17] together with peaks of
betaine 9 >minor), Ph3PO >major) and dialkoxytriphenylpho-
sphorane 12 >À59.37 ppm, minor).
The same experiment was performed for the aryl-substi-
tuted hydroxyester 2d. The addition of compound 2d >1
equivalent) to the preformed betaine >1.5 equivalent) gave
a
31PNMR spectrum with betaine peak, Ph 3PO peak and
signals at À49.68 and À47.11 ppm corresponding to the
phosphoranes 10 and 12. After the addition of phthalimide,
peaks of Ph3PO >major), betaine >minor) and phosphoranes
>minor) were detected in the 31PNMR spectrum, similarly to
thespectraldataobtainedforcompound2a. Therefore, wecan
conclude that there were no qualitative differences in the 31
P
NMR spectra for hydroxyesters 2a and 2d at the point before
the addition of the acidic component >HX) to the reaction
mixture. However, the relative intensities of the observed
peaks were not identical, the ratio of 9:Ph3PO:12:10 signals
for compound 2a was approximately 7:2:4:2 and 4:2:1.5:1 for
compound 2d. Thus, before the addition of phthalimide to the
mixture of Ph3P, DEAD and ROH much more betaine relative
to Ph3PO was observed in the case of aliphatic hydroxyester
2a than for the aromatic one. According to the 19F NMR
spectra >measured in parallel with 31PNMR), aliphatic hydro-
xyester 2a generated intermediate phosphoranes at this
stage, which were transformed to the product after adding
the acidic component. Reaction of Ph3Pand DEAD with
aromatic hydroxyester 2d led to the formation of the DEAD
In our experiments with hydroxyesters 2a and 2d we
changed the order of addition of the reagents from the
optimal to detect possible intermediates. Treatment of
Ph3P>1.5 equivalent) with DEAD >1.5 equivalent) in toluene
under nitrogen resulted in appearance of two peaks in the 31
P
NMR spectrum. The major peak was located at ‡43.48 ppm
and the minor at ‡24.77 ppm corresponding to the betaine 9
and triphenylphosphine oxide, respectively. Addition of the
hydroxyester 2a >1 equivalent) to this mixture caused the
appearance of two new peaks at À59.37 and À60.11 ppm
corresponding to the phosphoranes >12, 10) with the betaine

72
N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
N-alkylation product 8d and hydrazone 16d together with the
intermediate phosphoranes. Irreversible formation of com-
pounds 8d and 16d caused lowering of intensities of the peaks
of betaine and phosphoranes relative to the Ph3PO signal
in the 31PNMR spectrum. In the 19F NMR spectrum these
By-products were detected together with the target compound
3d after addition of phthalimide.
product compositions for alkyl- and aryl-substituted hydro-
xyesters. Following the Mitsunobu protocol for the condi-
tions optimized for each type of the substrate, b-phthalimido
derivatives were obtained starting from compounds 2a±e
>Table 4). The molar ratios of ROH:HX:Ph3P:DEAD were
1:1.4:1.2:1.2 for the aliphatic hydroxyesters 2a±c and
1:7:7:7 for the aromatic hydroxyesters 2d,e. Some ¯uor-
ine-containing By-products 8a±e, 16d,e and 17a±c were
separated by column chromatography and fully character-
ized. The alkylated hydrazine derivative 15, which was
described in the literature as the major By-product in the
Mitsunobu reaction, was not detected in our experiments;
compound 8 was found instead [18]. As was mentioned
above, adducts 8a,b as well as ole®ns 17a,b were formed
from hydroxyesters 2a,b in the experiments under non-
optimal conditions. Hydrazonoesters 16d,e were detected
as the main By-products from the reaction of aromatic
hydroxyesters 2d,e. It was noted above, that performing
the Mitsunobu reaction with phthalimide being added last to
the mixture of Ph3P/DEAD/ROH, gave the hydrazone 16d
before the addition of the acidic component.
We also note that the equilibrium between phosphoranes
10, 12 and alkoxyphosphonium salt 14 on adding phthali-
mide was shifted in favor of the salt 14 in the case of
hydroxyester 2a whereas for hydroxyester 2d, phosphoranes
10 and 12 but not alkoxyphosphonium salt 14 were
observed.
Whereas 31PNMR was helpful in monitoring the process
and in de®ning some of the phosphorus containing inter-
mediates, 19F NMR allowed us to analyze the components of
the ®nal reaction mixtures and therefore to compare the
Table 4
Products of the Mitsunobu reaction for b-hydroxyesters 2a±e
Hydroxyester,
Product 3 >yield, %)
By-product 8 >yield, %)
By-product 16 >yield, %)
By-product 17 >yield, %)
a: R ˆ n-C5H11
b: R ˆ n-C4H9
±
±
c: R ˆ i-C3H7
d: R ˆ 4±Cl±Ph
e: R ˆ 4±F±Ph
±

N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
73
As it is seen from the Table 4, decreasing yields of phtha-
4.2. Materials
limido adducts 3c±e were found relative to adducts 3a,b.
Presumablythisisduetothesterichindranceofhydroxyesters
2c±e compared to compounds 2a,b. The facilitation the
process and therefore improvement in the yields of the desired
products possibly lies in using the more acidic HX [19].
The ®nal step of our protocol consists in the transforma-
tion of the N-protected esters 3 to free amino acids 4.
Hydrazinolysis of the phthalimido derivatives, which is
traditionally accomplished [10] for their transformation to
amines, did not lead to the desired amino acids. We removed
the phthalimido moiety together with the ester function
under the acidic conditions [20]. Phthalimidoesters were
heated under re¯ux in a HBr/HOAc system and, after
subsequent propylene oxide treatment, the desired a,a-
di¯uoro-b-amino acids 4a,b,d were obtained. The transfor-
mation of the N-protected esters 3 to free amino acids 4 was
performed in 53±57% yields.
THF was puri®ed by distillation from sodium ketyl.
DEAD and deuterochloroform were dried by standing over
molecular sieves. Benzene and toluene were dried by dis-
tillation from phosphorus pentoxide. Column chromatogra-
phy was performed using silica gel 60 >0.040±0.063 mm)
Merck, and for TLC silica gel 60 F254 Merck plates were
used.
4.3. 2,2-Difluoro-3-hydroxy-octanoic acid ethyl ester ,2a)
Obtained by procedure 4.5 from hexanal with yield
82%. Spectral data are identical to those in the literature
[21].
4.4. 2,2-Difluoro-3-hydroxy-heptanoic acid ethyl ester ,2b)
Obtained by procedure 4.5 from valeraldehyde with yield
79%. Spectral data are identical to those given in the
literature [22].
3. Conclusions
4.5. 2,2-Difluoro-3-hydroxy-4-methylpentanoic
acid ethyl ester ,2c)
a,a-Di¯uoro-b-amino acids are now available in three
steps from aldehydes and ethyl bromodi¯uoroacetate. The
key Mitsunobu amination step was studied for a,a-di¯uori-
nated b-aryl- and b-alkyl-b-hydroxyesters. It was observed
by 31PNMR that phosphoranes of the type Ph 3P>OR)2,
Ph3P>OR)N>CO2R0)NHCO2R0 and oxyphosphonium salt
To a re¯uxing suspension of 1.9 g >25.6 mg atom) of
freshly activated Zn dust in 22 ml of dry THF was added a
solution of 5 g >24.6 mmol) of ethyl bromodi¯uoroacetate
and 1.42 g >19.7 mmol) of isobutyraldehyde in 22 ml THF at
a rate so as to maintain vigorous re¯ux. After 30 min the
reaction mixture was cooled, quenched with saturated aq.
NaHCO3, 1 M aq. KHSO4 and ether and left to stir for
15 min. After this time the layers were separated and the
aqueous phase was extracted with ether. The combined
organic layers were washed with brine, dried over MgSO4,
evaporated, and distilled to yield 2.51 g >65%) 2c as a
‡
[Ph3P>OR)] XÀ were involved as intermediates. The reaction
conditions were optimised for all substrates. Solvent was
shown to play a signi®cant role in reaction ef®cacy; the best
yieldswereobtainedintoluenesolution. Thecorrect sequence
for adding the reagents was found to be important, the DEAD
solution has to be added to the reaction mixture last in order
to reduce or eliminate conversion and regioselectivity pro-
blems. For aromatic hydroxyesters a large excess of DEAD,
Ph3Pand phthalimide should be used to achieve satisfactory
preparative yields of desired products. The detailed 19F NMR
studies allow observation of azoesters 8 and hydrazonoesters
16 as new By-products in the Mitsunobu reaction.
1
colourless liquid, bp 60±638C/0.06 mm Hg. H NMR >d,
CDCl3): 1.04 >t, JHH ˆ 6:6 Hz, 6H); 1.36 >t, JHH ˆ 7:2 Hz,
3H); 1.93±2.1 >m, 1H); 2.77 >bs, 1H); 3.76±3.88 >m, 1H);
4.35 >q, JHH ˆ 7:2 Hz, 2H); 19F NMR >d, CDCl3): À112.17,
À121.36 >ABX, JFF ˆ 266 Hz, JHF ˆ 17:5, 7.6 Hz, CF2).
IR: n ˆ 1760, 2840±3000 cmÀ1. Anal. calc. for C8H14F2O3:
C, 48.98; H, 7.19. Found: C, 49.09; H, 7.23%.
4. Experimental details
4.1. General
4.6. 3-,4-Chloro-phenyl)-2,2-difluoro-3-hydroxy-
propionic acid ethyl ester ,2d)
1H NMR >299.9 MHz), 13C NMR >75 MHz) and 19F
NMR >282.2 MHz) spectra were recorded on a Varian
VXR-300 spectrometer. 31PNMR >36.2 MHz) spectra were
recorded on a Jeol FX-90Q spectrometer. 19F NMR spectra
were recorded with external CFCl3 and 1H NMR with
external tetramethylsilane as references. GC/MS were mea-
sured on Hewlett-Packard 5890/5870. IR spectra were
obtained with a Specord IR-75 spectrometer in CHCl3
solution. Melting points were determined in open capillaries
and are uncorrected.
Obtained by procedure 4.5 from 4-chlorobenzaldehyde
with yield 93%. Spectral data are identical to those in the
literature [22].
4.7. 3-,4-Fluoro-phenyl)-2,2-difluoro-3-hydroxy-propionic
acid ethyl ester ,2e)
Obtained by procedure 4.5 with yield 87% as a colourless
1
liquid, bp 122±1268C/0.06 mm Hg. H NMR >d, CDCl3):

74
N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
1.29 >t, JHH ˆ 7:2 Hz, 3H); 3.31 >bs, 1H); 4.3 >q, JHH
ˆ
Anal. calc. for C16H17F2NO4: C, 59.07; H, 5.27; N, 4.31.
Found: C, 59.21; H, 5.20; N, 4.25%.
7:2 Hz, 2H); 5.15 >dd, JHF ˆ 15:3, 7.5 Hz, 1H); 7.0±7.1
>m, 2H); 7.3±7.4 >m, 2H); 19F NMR >d, CDCl3): À121.2,
À114.14 >ABX, JFF ˆ 264 Hz, JHF ˆ 15:3, 7.5 Hz, CF2);
À112.8 . . . À112.6 >m, CF). IR: n ˆ 1605, 1760, 2950±
3000 cmÀ1. Anal. calc. for C11H11F3O3: C, 53.23; H, 4.47.
Found: C, 53.39; H, 4.52%.
4.12. General procedure for b-aryl-N-protected esters 3d,
3e preparation under the Mitsunobu reaction
The solution of 0.92 g >5.31 mmol) of DEAD in 6 ml of
dry toluene was added dropwise to a stirred and cooled >08C)
mixture of 0.75 mmol of hydroxyester 2, 0.78 g >5.31 mmol)
of phthalimide and 1.39 g >5.31 mmol) of triphenylphos-
phine in 15 ml of toluene under an argon atmosphere. The
reaction mixture was allowed to warm to room temperature
and after 20 h the solvent was partially evaporated and the
crude mixture was chromatographed on silica gel >eluent
benzene).
4.8. General procedure for b-alkyl-N-protected
esters 3a, 3b, 3c preparation by the Mitsunobu
reaction
The solution of 0.19 g >1.07 mmol) of DEAD in 2 ml of
dry toluene was added dropwise to a stirred and cooled >08C)
mixture of 0.89 mmol of hydroxyester 2, 0.18 g >1.25 mmol)
of phthalimide and 0.28 g >1.07 mmol) of triphenylpho-
sphine in 3.2 ml of toluene under an argon atmosphere.
The reaction mixture was allowed to warm to room tem-
perature and after 1 h the solvent was partially evaporated
and the crude mixture was chromatographed on a silica gel
>eluent benzene).
4.13. 2-Chloro-3-,4-chloro-phenyl)-3-,1,3-dioxo-1,3-
dihydro-isoindol-2-yl)-2-fluoro-propionic acid ethyl ester
,3d)
Obtained by procedure 4.12 with yield 0.09 g >32%) as a
colourless oil. 1H NMR >d, CDCl3): 1.18 >t, JHH ˆ 7:2 Hz,
3H); 4.17±4.31 >m, 2H); 6.01 >dd, JHF ˆ 18:0, 13.5 Hz, 1H);
7.34 >d, JHH ˆ 8:5 Hz, 2H); 7.64 >d, JHH ˆ 8:5 Hz, 2H);
7.73±7.79 >m, 2H); 7.83±7.9 >m, 2H); 19F NMR >d, CDCl3):
À109.02, À105.22 >ABX, JFF ˆ 261:0 Hz, JHF ˆ 18:0,
13.5 Hz, CF2). IR: n ˆ 1605, 1760, 2950±3000 cmÀ1. Anal.
calc. for C19H14ClF2NO4: C, 57.95; H, 3.58; N, 3.56. Found:
C, 58.12; H, 3.62; N, 3.60%.
4.9. 3-,1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-2,2-difluoro-
octanoic acid ethyl ester ,3a)
Obtained by procedure 4.8 with yield 0.29 g >95%) as a
colourless oil. 1H NMR >d, CDCl3): 0.84 >t, JHH ˆ 7:5 Hz,
3H); 1.17±1.4 >m, 9H); 1.8±1.93 >m, 1H); 2.53±2.7 >m, 1H);
4.22±4.39 >m, 2H); 4.7±4.86 >m, 1H); 7.72±7.8 >m, 2H);
7.84±7.91 >m, 2H); 19F NMR >d, CDCl3): À111.8 . . .
À109.65 >m, CF2). IR: n ˆ 1750, 2840±3005 cmÀ1. Anal.
calc. for C18H21F2NO4: C, 61.18; H, 5.99; N, 3.96. Found: C,
61.28; H, 6.05; N, 3.90%.
4.14. 3-,1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-2,2-difluoro-
3-,4-fluoro-phenyl)-propionic acid ethyl ester ,3e)
Obtained by procedure 4.12 with yield 0.15 g >53%) as a
colourless oil. 1H NMR >d, CDCl3): 1.16 >t, JHH ˆ 7:2 Hz,
3H); 4.16±4.3 >m, 2H); 6.03 >dd, JHF ˆ 18:0, 13.5 Hz, 1H);
7.01±7.09 >m, 2H); 7.67±7.72 >m, 2H); 7.73±7.78 >m, 2H);
7.84±7.89 >m, 2H); 19F NMR >d, CDCl3): À112.12 . . .
4.10. 3-,1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-2,2-difluoro-
heptanoic acid ethyl ester ,3b)
Obtained by procedure 4.8 with yield 0.27 g >90%) as a
colourless oil. 1H NMR >d, CDCl3): 0.82 >t, JHH ˆ 7:5 Hz,
3H); 1.14±1.4 >m, 7H); 1.7±1.83 >m, 1H); 2.5±2.67 >m, 1H);
4.2±4.37 >m, 2H); 4.69±4.85 >m, 1H); 7.71±7.79 >m, 2H);
7.84±7.92 >m, 2H); 19F NMR >d, CDCl3): À111.78 . . .
À109.64 >m, CF2). IR: n ˆ 1760, 2840±3000 cmÀ1. Anal.
calc. for C17H19F2NO4: C, 60.17; H, 5.64; N, 4.13. Found: C,
60.13; H, 5.60; N, 4.05%.
À111.92 >m, CF); À109.34, À105.24 >ABX, JFF
ˆ
260:0 Hz, JHF ˆ 18:0, 13.5 Hz, CF2). IR: n ˆ 1600, 1760,
2950±3000 cmÀ1. Anal. calc. for C19H14F3NO4: C, 60.48;
H, 3.74; N, 3.71. Found: C, 60.56; H, 3.76; N, 3.66%.
4.15. General procedure for preparation of amino acids 4
A mixture of 6 ml of 48% hydrobromic acid, 6 ml of
glacial acetic acid and 6.34 mmol of ester 3 was heated
under re¯ux for 15 h. On cooling most of the phthalic acid
crystallized. After ®ltration, the ®ltrate and water washes
were concentrated under reduced pressure practically to
dryness. The residue was dissolved in 6 ml of water, ®ltered
and concentrated. Diluting with water and subsequent con-
centration were repeated several times. To the crude amino
acid hydrobromide, dissolved in methanol, 12.7 mmol of
propylene oxide was added in one portion. The reaction
mixture was evaporated to dryness and crystallized.
4.11. 3-,1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-2,2-difluoro-
4-methyl-pentanoic acid ethyl ester ,3c)
Obtained by procedure 4.8 with yield 0.09 g >44%) as a
colourless oil. 1H NMR >d, CDCl3): 0.87 >d, JHH ˆ 6:3 Hz,
3H); 1.2 >d, JHH ˆ 6:3 Hz, 3H); 1.25 >t, JHH ˆ 7:2 Hz, 3H);
2.9±3.17 >m, 1H); 4.16±4.38 >m, 2H); 4.43±4.57 >m, 1H);
7.71±7.81 >m, 2H); 7.83±7.93 >m, 2H); 19F NMR >d,
CDCl3): À106.36, À104.66 >ABX, JFF ˆ 260 Hz,
JHF ˆ 15, 10.4 Hz, 2F); IR: n ˆ 1760, 2850±3000 cmÀ1
.

N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
75
4.16. 3-Amino-2,2-difluoro-octanoic acid ,4a)
colourless oil. 1H NMR >d, CDCl3): 0.89 >t, JHH ˆ 7:5 Hz,
3H); 1.25±1.47 >m, 10H); 1.7±1.79 >m, 2H); 4.23 >q,
JHH ˆ 7:2 Hz, 2H); 4.28±4.39 >m, 2H); 5.13±5.27 >m,
1H); 19F NMR >d, CDCl3): À118.85, À114.96 >ABX,
JFF ˆ 266 Hz, JHF ˆ 12:7, 8.7 Hz, CF2). IR: n ˆ 1760,
1780, 2840±3000 cmÀ1. Anal. calc. for C12H20F2N2O4: C,
48.97; H, 6.85; N, 9.52. Found: C, 48.92; H, 6.88; N, 9.45%.
Crystallized from 2-propanol/methanol >2/1). Yield 0.7 g
>57%), mp 232±2368C. 1H NMR >d, CD3OD): 0.92 >t,
JHH ˆ 7:5 Hz, 3H); 1.27±1.88 >m, 8H); 3.61±3.75 >m,
1H); 19F NMR >d, CD3OD): À127.6, À122.6 >ABX,
JFF ˆ 259 Hz, JHF ˆ 14:4, 10.7 Hz, CF2). Anal. calc. for
C8H15F2NO2: C, 49.22; H, 7.75; N, 7.18. Found: C, 49.15;
H, 7.79; N, 7.15%.
4.21. 2,2-Difluoro-3-,ethoxycarbonylazo)-3-
methylpentanoic acid ethyl ester ,8c)
4.17. 3-Amino-2,2-difluoro-heptanoic acid ,4b)
Obtained by procedure 4.8. Isolated after column chro-
matography >eluent benzene) with yield 0.04 g >15%) as a
colourless oil. 1H NMR >d, CDCl3): 1.03 >d, JHH ˆ 6:9 Hz,
3H); 1.05 >d, JHH ˆ 7:2 Hz, 3H); 1.32 >t, JHH ˆ 7:2 Hz,
3H); 1.35 >t, JHH ˆ 7:2 Hz, 3H); 2.12±2.28 >m, 1H); 4.23 >q,
JHH ˆ 7:2 Hz, 2H); 4.33 >q, JHH ˆ 7:2 Hz, 2H); 5.05 >ddd,
JHF ˆ 15:2, 9.6 Hz, JHH ˆ 5:4 Hz, 1H); 19F NMR >d,
Crystallized from 2-propanol/methanol >2/1). Yield 0.65 g
1
>57%), mp ˆ 219±2218C. H NMR >d, CD3OD): 0.94 >t,
JHH ˆ 7:5 Hz, 3H); 1.3±1.9 >m, 6H); 3.63±3.77 >m, 1H); 19
F
NMR >d, CD3OD): À127.5, À122.5 >ABX, JFF ˆ 259 Hz,
JHF ˆ 14:4, 10.7 Hz, CF2). Anal. calc. for C7H13F2NO2: C,
46.40; H, 7.23; N, 7.73. Found: C, 46.33; H, 7.26; N, 7.66%.
CDCl3): À112.49, À116.35 >ABX, JFF ˆ 264 Hz, JHF
ˆ
4.18. 3-Amino-2,2-difluoro-3-,4-chloro-phenyl)-propionic
acid ,4d)
15:2, 9.6 Hz, CF2). IR: n ˆ 1760, 1780, 2850±3000 cmÀ1
.
Anal. calc. for C11H18F2N2O4: C, 47.14; H, 6.47; N, 10.00.
Found: C, 47.25; H, 6.45; N, 9.95%; MS m/z >%): 223 >1),
176 >2), 167 >3), 158 >7), 145 >18), 134 >8), 124 >32), 113 >4),
105 >34), 96 >32), 91 >43), 77 >8), 73 >100), 63 >12), 55 >11),
47 >1).
Crystallized from methanol. Yield 0.79 g >53%), mp ˆ
222±2258C. 1H NMR >d, CD3OD): 5.1 >dd, JHF ˆ 16:2,
8.4 Hz, 1H); 7.46±7.56 >m, 4H); 19F NMR >d, CD3OD):
À114.97, À107.84 >ABX, JFF ˆ 259 Hz, JHF ˆ 16:2,
8.4 Hz, CF2). Anal. calc. for C9H8ClF2NO2: C, 45.88; H,
3.42; N, 5.94. Found: C, 45.75; H, 3.46; N, 5.87%.
4.22. 3-,4-Chloro-phenyl)-2,2-difluoro-3-
,ethoxycarbonylazo)-propanoic acid ethyl ester ,8d)
4.19. 2,2-difluoro-3-,ethoxycarbonylazo)-octanoic acid
ethyl ester ,8a)
Obtained by procedure 4.12. Isolated after column chro-
matography >eluent benzene) with yield 0.08 g >30%) as a
yellow oil. 1H NMR >d, CDCl3): 1.3 >t, JHH ˆ 7:2 Hz, 3H);
1.31 >t, JHH ˆ 7:2 Hz, 3H); 4.15±4.26 >m, 2H); 4.27±4.39
>m, 2H); 6.06 >dd, JHF ˆ 15:0, 8.7 Hz, 1H); 7.3±7.46 >m,
4H); 19F NMR >d, CDCl3): À118.22, À113.87 >ABX,
JFF ˆ 268 Hz, JHF ˆ 15:0, 8.7 Hz, CF2); 13C NMR >d,
CDCl3): 13.81 >CH3), 14.06 >CH3), 63.41 >CH2), 65.18
>CH2), 75.87 >dd, J1 ˆ 28:3 Hz, J2 ˆ 36 Hz, CH), 112.14
>dd, J1 ˆ 306 Hz, J2 ˆ 313 Hz, CF2), 128.85 >CH), 129.54
>d, J1 ˆ 2:4 Hz, C), 129.74 >CH), 135.99 >CCl), 153.42
>CO), 162.16 >dd, J1 ˆ 35:6 Hz, J2 ˆ 38:6 Hz, CO). IR:
n ˆ 1605, 1760, 1780, 2950±3000 cmÀ1. Anal. calc. for
C14H15ClF2N2O4: C, 48.22; H, 4.34; N, 8.03. Found: C,
48.36; H, 4.36; N, 8.07%; MS m/z >%): 336 >1), 316>1), 263
>2), 246 >7), 244 >23), 216 >7), 198 >4), 174 >23), 141 >100),
113 >17), 96 >10), 77 >18), 63 >2), 51 >3).
Obtained by procedure 4.8. Isolated after column chro-
matography >eluent benzene) with yield 0.005 g >2%) as a
1
colourless oil. H NMR >d, CDCl3): 0.89 >t, JHH ˆ 7:5 Hz,
3H); 1.25±1.48 >m, 12H); 1.69±1.79 >m, 2H); 4.23 >q,
JHH ˆ 7:2 Hz, 2H); 4.28±4.39 >m, 2H); 5.13±5.27 >m, 1H);
19F NMR >d, CDCl3): À118.84, À114.95 >ABX,
JFF ˆ 266 Hz, JHF ˆ 12:7, 8.7 Hz, CF2); 13C NMR >d,
CDCl3): 13.83 >CH3), 13.87 >CH3), 14.09 >CH3), 22.29
>CH2), 24.38 >CH2), 27.30 >CH2), 31.25 >CH2), 63.20
>CH2), 64.79 >CH2), 75.11 >dd, J1 ˆ 29:4 Hz, J2 ˆ
34:4 Hz, CH), 113.02 >dd, J1 ˆ 298:5 Hz, J2 ˆ 309:8 Hz,
CF2), 154.33 >CO), 162.59 >dd, J1 ˆ 33:7 Hz, J2 ˆ 38:5 Hz,
CO). IR: n ˆ 1760, 1780, 2840±3000 cmÀ1. Anal. calc. for
C13H22F2N2O4: C, 50.64; H, 7.19; N, 9.09. Found: C, 50.75;
H, 7.17; N, 9.13%; MS m/z >%): 287 >1), 263 >1), 244 >3), 223
>2), 195 >6), 186 >9), 173 >14), 164 >4), 152 >10), 144 >11), 137
>4), 124 >39), 116 >14), 113 >18), 101 >100), 91 >28), 83 >44),
69 >5), 63 >9), 55 >16).
4.23. 2,2-Difluoro-3-,ethoxycarbonylazo)-3-,4-fluoro-
phenyl)-propanoic acid ethyl ester ,8e)
Obtained by procedure 4.12. Isolated after column chro-
matography >eluent benzene) with yield 0.08 g >32%) as a
4.20. 2,2-Difluoro-3-,ethoxycarbonylazo)-heptanoic
acid ethyl ester ,8b)
1
colourless oil. H NMR >d, CDCl3): 1.3 >t, JHH ˆ 7:2 Hz,
6H); 4.14±4.26 >m, 2H); 4.27±4.37 >m, 2H); 6.07 >dd,
JHF ˆ 15:0, 9.0 Hz, 1H); 7.06±7.13 >m, 2H); 7.43±7.49
>m, 2H); 19F NMR >d, CDCl3): À118.34, À114.51 >ABX,
Obtained by procedure 4.8. Isolated after column chro-
matography >eluent benzene) with yield 0.008 g >3%) as a

76
N.A. Fokina et al. / Journal of Fluorine Chemistry 111 ,2001) 69±76
JFF ˆ 264 Hz, JHF ˆ 15:0, 9.0 Hz, CF2); À111.93 . . .
À111.73 >m, CF). IR: n ˆ 1605, 1760, 1780, 2850±
3000 cmÀ1. Anal. calc. for C14H15F3N2O4: C, 50.61; H,
4.55; N, 8.43. Found: C, 50.49; H, 4.52; N, 8.39%; MS m/z
>%): 320 >0.5), 300 >0.5), 256 >1), 247 >1), 228 >15), 200 >5),
197 >6), 182 >4), 158 >19), 137 >2), 125 >100), 107 >5), 97
>20), 75 >2), 57 >1), 51 >1).
>m, 1H); 6.21±6.37 >m, 1H); 19F NMR >d, CDCl3): À103.55
>d, JHF ˆ 8:6 Hz, CF2). Anal. calc. for C9H14F2O2: C, 56.24;
H, 7.34. Found: C, 56.15; H, 7.38%.
4.28. 2,2-Difluoro-4-methyl-pent-3-enoic acid
ethyl ester ,17c)
Obtained by the Mitsunobu procedure for hydroxyester
2c. Isolated after column chromatography >eluent benzene)
with yield 0.05 g >34%) as a colourless oil. H NMR >d,
4.24. 3-,4-Chloro-phenyl)-3-,ethoxycarbonyl-hydrazono)-
2,2-difluoro-propionic acid ethyl ester ,16d)
1
CDCl3): 1.34 >t, JHH ˆ 7:2 Hz, 3H); 1.8±1.9 >m, 6H); 4.32
>q, JHH ˆ 7:2 Hz, 2H); 5.38±5.51 >m, 1H); 19F NMR >d,
CDCl3): À98.31 >d, JHF ˆ 15 Hz, CF2). Anal. calc. for
C8H12F2O2: C, 53.93; H, 6.79. Found: C, 53.85; H, 6.84%.
Obtained by procedure 4.12. Isolated after column chro-
matography >eluent benzene) with yield 0.1 g >37%) as a
colourless oil. 1H NMR >d, CDCl3): 1.28 >t, JHH ˆ 7:2 Hz,
3H); 1.42 >t, JHH ˆ 7:2 Hz, 3H); 4.23 >q, JHH ˆ 7:2 Hz,
2H); 4.45 >q, JHH ˆ 7:2 Hz, 2H); 7.33 >d, JHH ˆ 8:4 Hz,
2H); 7.54 >d, JHH ˆ 8:4 Hz, 2H); 19F NMR >d, CDCl3):
À102.94 >s, CF2)). IR: n ˆ 1600, 1740, 1770, 2950±
3000 cmÀ1. Anal. calc. for C14H15ClF2N2O4: C, 48.22; H,
4.34; N, 8.03. Found: C, 48.41; H, 4.38; N, 8.09%.
Acknowledgements
We are grateful to the Ukrainian Fundamental Research
Fund for ®nancial assistance for this work.
4.25. 3-,Ethoxycarbonyl-hydrazono)-2,2-difluoro-3-,4-
fluoro-phenyl)-propionic acid ethyl ester ,16e)
References
[1] E. Juaristi >Ed.), Enantioselective Synthesis of b-Amino Acids,
Wiley, New York, 1997, p. 127.
Obtained by procedure 4.12. Isolated after column chro-
matography >eluent benzene) with yield 0.04 g >14%) as a
colourless oil. 1H NMR >d, CDCl3): 1.28 >t, JHH ˆ 7:2 Hz,
3H); 1.42 >t, JHH ˆ 7:2 Hz, 3H); 4.24 >q, JHH ˆ 7:2 Hz,
2H); 4.44 >q, JHH ˆ 7:2 Hz, 2H); 7.02±7.12 >m, 2H); 7.4±
7.5 >m, 2H); 19F NMR >d, CDCl3): À108.65, >s, CF);
À103.39 >s, CF2)). IR: n ˆ 1600, 1740, 1770, 2950±
3000 cmÀ1. Anal. calc. for C14H15F3N2O4: C, 50.61; H,
4.55; N, 8.43. Found: C, 50.76; H, 4.59; N, 8.50%.
[2] V.P. Kukhar, V.A. Soloshonok >Eds.), Fluorine-Containing Amino
Acids, Wiley, New York, 1994, p. 140.
[3] O. Mitsunobu, Synthesis >1981) 1±28.
[4] P.G. Sammes, D.J. Thetfort, Chem. Soc., Perkin Trans. 1 >1989) 655.
[5] H.-G. Brunker, W. Adam, J. Am. Chem. Soc. 117 >1995) 3976±3982.
[6] J.R. Henry, L.R. Marcin, M.C. McIntosh, P.M. Scola, G.D. Harris Jr.,
S.M. Weinreb, Tetrahedron Lett. 30 >42) >1989) 5709±5712.
[7] M. Wada, T. Sano, O. Mitsunobu, Bull. Chem. Soc. Jpn. 46 >1973)
2833±2835.
[8] P. Bravo, G. Caviccio, M. Crucianelli, A. Poggiali, M. Zanda,
Tetrahedron: Asymm. 8 >16) >1997) 2811±2815.
4.26. 2,2-Difluoro-oct-3-enoic acid ethyl ester ,17a)
[9] H. Fukuda, M. Tetsu, T. Kitazume, Tetrahedron 52 >1) >1996) 157±
164.
Obtained by the Mitsunobu procedure for hydroxyester
2a. Isolated after column chromatography >eluent benzene)
with yield 0.006 g >3%) as a colourless oil. H NMR >d,
[10] D.L. Hughes, The Mitsunobu Reaction in Organic Reactions, Wiley,
New York, 1992, p. 42.
[11] M. Varasi, K.A.M. Walker, M.L. Maddox, J. Org. Chem. 52 >1987)
4235±4238.
1
CDCl3): 0.9 >t, JHH ˆ 7:5 Hz, 3H); 1.2±1.48 >m, 7H); 2.12±
2.33 >m, 2H); 4.32 >q, JHH ˆ 7:2 Hz, 2H); 5.58±5.74 >m,
1H); 6.2±6.36 >m, 1H); 19F NMR >d, CDCl3): À103.53 >d,
JHF ˆ 8:7 Hz, CF2). Anal. calc. for C10H16F2O2: C, 58.24;
H, 7.82. Found: C, 58.15; H, 7.85%.
[12] A. Pautard-Cooper, S.A. Evans Jr., Heterocycles 35 >1) >1993) 349±
365.
[13] J.E. Macor, J.M. Wehner, Heterocycles 35 >1) >1993) 349±365.
[14] R.D. Guthrie, I.D. Jenkins, Aust. J. Chem. 36 >1983) 767±774.
[15] M. Itzstein, I.D. Jenkins, Aust. J. Chem. 36 >1983) 557±563.
[16] D.L. Hughes, R.A. Reamer, J.J. Bergan, E.J.J. Grabovski, J. Am.
Chem. Soc. 110 >1988) 6487±6491.
4.27. 2,2-Difluoro-hept-3-enoic acid ethyl ester ,17b)
[17] D. Camp, I.D. Jenkins, J. Org. Chem. 54 >1989) 3045±3049.
[18] T. Tsunoda, Y. Yamamiya, Sh. Ito, Tetrahedron Lett. 34 >10) >1993)
1639±1642.
Obtained by the Mitsunobu procedure for hydroxyester
2b. Isolated after column chromatography >eluent benzene)
[19] S.F. Martin, J.A. Dodge, Tetrahedron Lett. 32 >26) >1991) 3017±
3020.
1
with yield 0.01 g >7%) as a colourless oil. H NMR >d,
[20] F. Weygand, W. Oettmeier, Rus. Chem. Rev. 4 >1970) 290±300.
[21] R.J. Linderman, D.M. Graves, J. Org. Chem. 54 >3) >1989) 661±668.
[22] Y. Shen, M. Qi, J. Fluorine Chem. 67 >3) >1994) 229±232.
CDCl3): 0.9 >t, JHH ˆ 7:5 Hz, 3H); 1.19±1.47 >m, 5H);
2.13±2.35 >m, 2H); 4.32 >q, JHH ˆ 7:2 Hz, 2H); 5.59±5.75