Preparation of (-)-(R)-2-(2,3,4,5,6-pentafluorophenoxy)-2-(phenyl-d5)acetic acid: An efficient 1H NMR chiral solvating agent for direct enantiomeric purity evaluation of quinoline-containing antimalarial drugs

Article abstract of DOI:10.1016/j.tetasy.2014.11.007

The title compound was prepared as a racemate from (±)-mandelic acid-d₅ in one step. The corresponding (-)-(R)-enantiomer (98% ee) was obtained by resolution with (-)-(R)-1-phenylethylamine and evaluated as a chiral solvating agent (CSA) for direct ¹H NMR enantiomeric excess determination of mefloquine (Lariam), chloroquine (Chloroquine Bayer), and hydroxychloroquine (Plaquenil) enantiomers. The displayed non-equivalence was high for signals in the aromatic region of all three antimalarials. Thus, the mandelic acid derivative described herein may be considered as the first efficient CSA 'invisible' in the aromatic region, useful for direct ¹H NMR ee value determination of chiral quinoline-containing antimalarial drugs.

Full text of DOI:10.1016/j.tetasy.2014.11.007

Tetrahedron: Asymmetry 25 (2014) 1605–1611
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Preparation of (À)-(R)-2-(2,3,4,5,6-pentafluorophenoxy)-2-
1
(phenyl-d5)acetic acid: an efficient H NMR chiral solvating agent
for direct enantiomeric purity evaluation of quinoline-containing
antimalarial drugs
⇑
Maria Maddalena Cavalluzzi , Angelo Lovece, Claudio Bruno, Carlo Franchini, Giovanni Lentini
Dipartimento di Farmacia-Scienze del Farmaco, Università degli Studi di Bari ‘Aldo Moro’, via E. Orabona n. 4, 70126 Bari, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 September 2014
Accepted 12 November 2014
The title compound was prepared as a racemate from ( )-mandelic acid-d5 in one step. The corresponding
(À)-(R)-enantiomer (98% ee) was obtained by resolution with (À)-(R)-1-phenylethylamine and evaluated
as a chiral solvating agent (CSA) for direct 1H NMR enantiomeric excess determination of mefloquine
(LariamÒ), chloroquine (Chloroquine BayerÒ), and hydroxychloroquine (PlaquenilÒ) enantiomers. The
displayed non-equivalence was high for signals in the aromatic region of all three antimalarials. Thus,
the mandelic acid derivative described herein may be considered as the first efficient CSA ‘invisible’ in
the aromatic region, useful for direct 1H NMR ee value determination of chiral quinoline-containing
antimalarial drugs.
Ó 2014 Elsevier Ltd. All rights reserved.
F
1. Introduction
Cl
N
F
F
F
F
N
Malaria is one of the most prevalent and deadly infectious dis-
eases all over the world.1–3 The last few decades have witnessed
the relentless emergence of widespread resistance to old first-line
drugs.1–4 Thus, the need for efficacious antimalarial agents is
important. The ideal antimalarial drug should be endowed with
efficiency (i.e., fast-action at low single-dosage) and tolerability
(absence of severe side-effects, mainly in children and pregnant
women).1 Recent efforts to improve physicians’ panoply against
malaria have focused on combined therapy and innovative drug
development. Unfortunately, partial efficacy,5 drug resistance,
and side-effects1–4 have seriously flawed these attempts. Several
traditional, quinoline-containing antimalarial agents such as chlo-
roquine 1, hydroxychloroquine 2, and mefloquine 3 are dispensed
as racemates, while evidence indicates that the corresponding
isolated enantiomers may have different pharmacodynamic,6,7,11
toxicological,8,10,13 and pharmacokinetic properties.9,10,12,14 Thus,
the chiral switch15 approach (the use of single enantiomers in lieu
of the corresponding racemate) might offer a straightforward way
to safer and more efficacious drugs.
F
HN
R
N
HO
HN
chloroquine 1 (R = H)
hydroxychloroquine 2 (R = OH)
mefloquine 3 (R*,S*)
Given this goal, facile methods to ascertain the enantiomeric
excess of the isolated 1–3 enantiomers should be available. A
plethora of HPLC7,16,17 and CE16–20 methods to evaluate the ee
values of enantiomerically enriched forms of 1–3 have been devel-
oped. No reliable 1H NMR spectroscopy methods devoted to the ee
determination of these antimalarials have been reported to date.
Carroll and Blackwell21 reported on the use of tris[3-(trifluoro-
methylhydroxymethylene)-d-camphorato]europium(III) to assess
the enantiomeric purity of (À)- and (+)-3. This chiral lanthanide
shift reagent, however, required acetylation of the analytes, caused
distortion of the baseline, and did not induce baseline resolution of
the acetyl singlets of (À)- and (+)-3 acetyl derivatives. Aslam and
Craig reported22 that the enantiomeric purity of the diastereomeric
⇑
Corresponding author. Tel.: +39 080 5442232; fax: +39 080 5442724.
E-mail address: mariamaddalena.cavalluzzi@uniba.it (M.M. Cavalluzzi).
http://dx.doi.org/10.1016/j.tetasy.2014.11.007
0957-4166/Ó 2014 Elsevier Ltd. All rights reserved.

1606
M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
salts of (À)- and (+)-2 with (+)-(S)-binaphthylphosphoric acid can
be determined by monitoring the ratio of the two methyl doublets
at 1.22 and 1.26 ppm. However, the aliphatic region of the 1H NMR
spectrum of ( )-2Á(+)-(S)-binaphthylphosphate reported therein
demonstrates that the above doublets were far from being baseline
resolved. Recently, a method based on NMR under anisotropic
conditions was applied to the enantiodiscrimination of (À)- and
(+)-3.23 The polyacrylamide-based gel used as a chiral alignment
medium would be applicable for enantiomeric assignment but
does not allow ee determination. Over the last decade, we have
developed a series of O-aryl mandelic acids as chiral solvating
agents (CSAs) for the direct 1H NMR ee value determination
of several biologically important amines.24,25 In particular, the
homochiral amines (9–21) as CSAs. By recording the spectra in
CDCl3, 9 and 12–19 were completely ineffective, while unsatisfac-
tory chemical shift non-equivalences (
observed in the presence of 10 and 11. With the aim to increase the
d value observed using 11 as the CSA, experiments were per-
Dd <0.02 ppm, Table 1) were
D
formed using higher analyte/CSA ratios (entries 1–3, Table 2) but
even under these conditions, no suitable non-equivalence was
observed. Several deuterated solvents were then used and an
encouraging
Dd was finally obtained in toluene-d8 (entries 4–6).
Further increases of the analyte/CSA molar ratio in the latter sol-
vent gave poor improvements (entries 7 and 8). By recording the
spectrum at low temperature (–10 °C, entry 9), the highest
Dd
value was finally obtained. Unfortunately, under these conditions
it was not possible to determine the ee values for (À)-5 and (+)-5
since precipitation of the corresponding diastereomeric salts
occurred. The best results were obtained when two N-benzyl
O-pentafluorophenyl mandelic acid derivative
4 was able to
baseline resolve two aromatic signals of a mexiletine analogue pre-
senting no singlets or doublets in a free high field region of its 1H
NMR spectrum.25 Thus, we hypothesized that 4 might form
sufficiently stable complexes with 1–3 to induce chemical shift
non-equivalence on the quinoline ring proton signals. In order to
avoid crowding of CSA signals in the aromatic region of the 1H
NMR spectrum where splitting of signals should occur, we focused
our attention on 2-(2,3,4,5,6-pentafluorophenoxy)-2-(phenyl-d5)-
acetic 5.
analogues, (+)-(R)-N-benzyl-a-methylbenzylamine 20 and (+)-
bis[(R)-1-phenylethyl]amine 21, were used as CSAs. In fact, the
non-equivalences produced on the methine signal in the presence
of 20 and 21 were 0.065 and 0.060, respectively, at room temper-
ature. Therefore, 20 was used as the CSA and the ee values of (À)-5
and (+)-5 were 98% and 96%, respectively. The CSAs bearing an
N-benzyl moieties 20, 21 caused the inversion of the enantiomer
CH singlet positions in the corresponding spectra, with the (S)-
enantiomer signal falling to lower fields than the (R)-enantiomer,
which was opposite to what was observed in the presence
of 11.
F
F
F
F
O
R
R
F
R
R
R2
CO2H
NH2
N
H
R
R1
(−)-(S)-1-phenylethylamine 9 (R1 = R2 = H)
12
(+)-(R)-β-methylphenethylamine
4: R = H
5:
(+)-(R)-N, -dimethylbenzylamine
(R1 = H, R2 = Me)
α
10
R = D
(+)-(R)-α,4-dimethylbenzylamine 11 (R1 = Me, R2 = H)
R
2. Results and discussion
NH2
N
H
2.1. Synthesis and resolution of O-pentafluorophenyl mandelic
acid-d5
9
8
N
OH
(8R,9S)-cinchonine 14 (R = H)
15
13
(−)-(S)-1-(1-naphthyl)ethylamine
Racemic O-pentafluorophenyl mandelic acid-d5 (RS)-5
[2-(2,3,4,5,6-pentafluorophenoxy)-2-(phenyl-d5)acetic acid] was
prepared via the synthetic route shown in Scheme 1. (RS)-2-
Hydroxy-2-(D5)phenylacetic acid (RS)-8 was prepared as
previously described,26 by submitting hexadeuterobenzene 6 to
Friedel–Crafts acylation with 2,2-dichloroacetyl chloride to give
dichloroacetophenone-d5 7. Treatment of 7 with NaOH gave (RS)-
8, which underwent a copper-catalyzed coupling with hexafluoro-
benzene25,27 to afford the desired compound (RS)-5.
(8S,9R)-cinchonidine
(R = H)
(8R,9S)-quinidine 16 (R = OMe)
HO
N
H
H
H
O
OH
NH2
H
O
N
H
H
HO
O
O
OH
O
OH
O
O
H
The resolution of O-pentafluorophenyl mandelic acid-d5 (RS)-5
by using (R)-1-phenylethylamine [(R)-PEA] as the resolving agent
was then performed. Thus, a stoichiometric amount of the optically
active amine was added to an ethanol solution of (RS)-5 (Scheme 2)
and the diastereomeric salt was recrystallized twice from absolute
ethanol. In order to liberate the enriched enantiomer of O-penta-
fluorophenyl mandelic acid-d5, the salt was dissolved in water
and, after acidification with 2 M HCl and extraction with EtOAc,
white crystals of (À)-5 (14% yield) were obtained. The partially
resolved (+)-O-pentafluorophenyl mandelic acid-d5 was easily
recovered from the mother liquors of the first recrystallization step
and was then treated with (S)-PEA as the resolving agent for fur-
ther resolution. After two recrystallization steps, (+)-O-pentafluor-
ophenyl mandelic acid-d5 was recovered from (+)-5Á(S)-PEA in 27%
yield. The enantiomeric purities of the enantiomers of 5 were ten-
tatively assessed by direct 1H NMR analysis by using a number of
tetracycline 17
brucine 18
R
O
P
Ph
Ph
N
H
N
H
(2S)-2-[(diphenylphosphoryl)methyl]
pyrrolidine 19
20
(+)-(R)-N-benzyl-α−methylbenzylamine
21
(R = H)
(+)-bis[(R)-1-phenylethyl]amine
(R = Me)
Finally, the absolute configurations of O-pentafluorophenyl man-
delic acid-d5 enantiomers were determined on the basis of CD
analysis (Fig. 1). The (R)-configuration was assigned to the levoro-
tatory isomer of 5 by spectroscopic correlation with (À)-(R)-O-pen-
tafluorophenyl mandelic acid (À)-(R)-4, synthesized as previously
reported,25 both their corresponding CD curves showing negative
Cotton effects (Fig. 1a).

M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
1607
F
F
F
F
D
D
O
D
D
D
O
D
D
OH
D
F
D
D
D
D
D
D
Cl
D
D
D
D
OH
i
ii
COOH
iii
Cl
O
D
D
D
6
7
(RS)-5
(RS)-8
Scheme 1. Reagents and conditions: (i) 2,2-dichloroacetyl chloride, AlCl3, CH2Cl2, microwave irradiation, 70 °C, 8 min; (ii) NaOH, H2O, microwave irradiation, 110 °C, 11 min;
(iii) hexafluorobenzene, Cs2CO3, CuI, butyronitrile, 95 °C, 24 h.
F
F
F
F
D
D
O
D
D
F
NH2
COOH
+
D
(R)-PEA
(RS)-5
EtOH
(−)-5.(R)-PEA Ixx [α]D20 = −75 (c 2, MeOH)
(+)-5.(R)-PEA [α]D20 = +29 (c 2, MeOH)
2 M HCl-EtOAc
EtOH
5.
)- (R)-PEA IIxx [
α
]
−
20
(+)-5
−
(
=
97 (c 1.5, MeOH)
D
(S)-PEA/EtOH
2 M HCl-EtOAc
(98% ee)
20
(+)-5.(S)-PEA Ixx [α]D = +78 (c 2, MeOH)
−
)-
5
(
EtOH
5.
20
(+)- (S)-PEA IIxx [α]D = +86 (c 2, MeOH)
2 M HCl-EtOAc
5
(+)- (96% ee)
Scheme 2. Resolution of (RS)-O-pentafluorophenyl mandelic acid-d5 (RS)-5.
Table 1
CSAs screened for 1H NMR enantiodiscrimination with (RS)-5
Table 2
Variation of temperature, solvent, and analyte/CSA ratio to improve the efficiency of
11 as a CSA for 1H NMR enantiodiscrimination with (RS)-5
CSA
Ratio
D
da
Entry
Temp (°C)
Solvent
Ratio
Dd
9
1:4
0
10
11
12
13
14
15
16
17
18
19
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5c
0.009
0.017
1
2
3
4
5
6
7
8
9
rt
rt
rt
rt
rt
rt
rt
rt
À10
CDCl3
CDCl3
CDCl3
C5D5N
1:2
1:3
1:4
1:2
1:2
1:2
1:3
1:4
1:4
0.017
0.018
0.019
0
0.036
0.040
0.041
0.042
0.066
0
0
0b
0
0
0
0
0
C6D6
Toluene-d8
Toluene-d8
Toluene-d8
Toluene-d8
a
b
c
Spectra recorded in CDCl3.
Spectrum recorded in CDCl3/toluene-d8/MeOH.
1:2 and 1:0.67 analyte/CSA ratios were also used (Dd = 0).

1608
M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
Figure 1. (a) Observed (À)-(R)-O-pentafluorophenyl mandelic acid [(À)-(R)-4] and (À)-(R)-5 CD spectra (dashed and solid lines, resp.) in water; (b) observed (À)-(R)-4 and
(+)-(S)-5 CD spectra (dashed and solid lines, resp.) in water; (c) observed (À)-(R)-4 and (À)-(R)-5 UV absorption spectra (dashed and solid lines, respectively) in water.
2.2. Application of (À)-(R)-O-pentafluorophenyl mandelic acid-
d5 as a chiral 1H NMR discriminating agent
blets shifting upfield with a lower Dd (0.20, Fig. 4b, and Table 3)
and partially overlapping the HC-3 singlet. Conversely, by reducing
the CSA to 1 equiv the enantioseparation was optimal but salt pre-
The chiral discrimination ability of (À)-(R)-O-pentafluorophenyl
mandelic acid-d5 (À)-(R)-5 was evaluated toward the clinically rel-
evant antimalarial agents chloroquine 1, hydroxychloroquine 2,
and mefloquine 3. When recording the 1H NMR spectra of both 1
and 2 in the presence of 1.5 equiv of (À)-(R)-5 in CDCl3, good enan-
tiodiscrimination was observed for Ar HC-5 of both compounds.
Baseline separation of the proton doublet signal was obtained for
cipitation occurred at room temperature (Dd 0.20, Fig. 4c, and
Table 3). Finally, satisfactory results were obtained when CD2Cl2
was added as a co-solvent, with the doublets falling in a free
spectrum region with good non-equivalence (0.22, Fig. 4d, and
Table 3) and no precipitation occurring. In order to verify the appli-
cability of (À)-(R)-5 as a CSA to ee value determination of quino-
line-containing antimalarials characterized by a primary amine
group, the CDCl3 spectrum of ( )-primaquine 22 in the presence
of 1.5 equiv of (À)-(R)-5 was recorded. Unfortunately, no enantio-
discrimination was observed.
both compounds, with
(Fig. 2a,b, and Table 3). Compound 2 showed a slightly lower
D
d values of 0.113 and 0.078, respectively
d,
D
probably due to the lower basicity and lipophilicity of its tertiary
amine group. Higher non-equivalences were obtained for some
aromatic protons of mefloquine 3 possibly as a consequence of
the higher rigidity of the analyte structure and the participation
of the OH group in the transient complex formation and stabiliza-
tion. As shown in Figure 3a and Table 3 (CDCl3, analyte/CSA molar
ratio 1:1.5), enantiodifferentiation with peak baseline resolution
O
N
HN
NH2
primaquine 22
was observed for a doublets signal (HC-5 or HC-7,
Dd 0.27) and
for the HC-6 triplet ( d 0.28). Unfortunately, the upfield doublet
D
Thus, the degree and type of substitution on the most basic
nitrogen seems to play a pivotal role in transient complex forma-
tion. However, several other structure determinants, such as elec-
tron density of the quinoline ring and basicity of its nitrogen atom,
as well as the distance between the quinoline and aniline-type
nitrogen atoms, might contribute.
and the upfield triplet overlapped with other signals. With the
aim of shifting the overlapped signals into a free region of the spec-
trum, the influence of the analyte/CSA ratio was evaluated. When
2 equiv of CSA were used, no improvement was obtained
(Fig. 3b). Conversely, by lowering the mefloquine/CSA molar ratio
to 1:0.7, the HC-6 triplet shifted downfield, away from the solvent
signal (Dd 0.17, Fig 3d and Table 3). When the spectra were
recorded in toluene-d8 with a mefloquine/CSA molar ratio of
3. Conclusions
1:1.5, the downfield doublets were almost free from overlapping
In conclusion, a simple procedure to obtain O-pentafluorophe-
nyl mandelic acid-d5 enantiomers in highly enriched optically
(HC-5 or HC-7,
Dd 0.25, Fig. 4a and Table 3). A further increase of
CSA resulted in a worsening of the signal separation, with the dou-

M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
1609
Figure 2. Enantiodiscrimination for HC-5 doublets of chloroquine 1 and hydroxychloroquine 2 with (À)-(R)-5 (a and b, respectively), by recording the spectrum in CDCl3 with
an analyte/CSA molar ratio of 1:1.5.
microwave reactor, with continuous stirring. The temperature
was measured and controlled by a built-in infrared detector. The
structures of the compounds were confirmed by routine spectro-
metric analyses. Only the spectra for compounds not previously
described are given. Melting points were determined on a Gallenk-
amp melting point apparatus in open glass capillary tubes and are
uncorrected. 1H NMR and 13C NMR spectra were recorded on a Var-
ian Mercury-VX spectrometer operating at 300 and 75 MHz for 1H
and 13C, respectively, using CDCl3 as a solvent, unless otherwise
indicated. Chemical shifts are reported in parts per million (ppm)
relative to the residual non-deuterated solvent resonance: CDCl3,
d 7.26 (1H NMR) and 77.2 (13C NMR). J values are given in Hz. Ele-
mental analyses were performed on a Eurovector Euro EA 3000
analyzer. Optical rotations were measured on a Perkin Elmer (Nor-
Table 3
Magnitude of non-equivalences determined in the presence of (À)-(R)-5
Amine
Affected protons
Analyte/CSA molar ratio
D
da
1
2
3
Ar HC-5
Ar HC-5
Ar HC-6
1:1.5
1:1.5
1:1.5
1:0.7
1:1.5
1:1.5
1:2
0.113
0.078
0.28
0.17
0.27
Ar HC-5 or HC-7
0.25b
0.20b
0.20b
0.22c
1:1
1:1
a
b
c
Spectra were registered in CDCl3, unless otherwise noted.
Solvent: toluene-d8.
Solvent: toluene-d8/CD2Cl2 = 4:1.
walk, CT) Mod 341 spectropolarimeter; concentrations are
20
expressed in g/100 mL, and the cell length was 1 dm, thus [a]
D
active forms was reported. Unambiguous attributions of (R)- and
(S)-absolute configurations to (À)- and (+)-O-pentafluorophenyl
mandelic acid-d5, respectively, have been given. The (R)-enantio-
mer was an efficient CSA for the direct 1H NMR ee determination
of mefloquine (LariamÒ), chloroquine (Chloroquine BayerÒ), and
hydroxychloroquine (PlaquenilÒ), inducing remarkable non-equi-
valences in the aromatic region of the spectra of these analytes.
The signals of the quinoline-containing antimalarials could be used
for ee evaluation since they were obscured by those of the CSA,
with the latter being devoid of any signal in the aromatic region
of its 1H NMR spectrum.
values are given in units of 10À1 deg cm2 gÀ1. TLC analyses were
performed on precoated silica gel on aluminum sheets (Kieselgel
60F254, Merck). CD and UV curves were registered on a J-810 model
JASCO spectropolarimeter and the concentration of the solution
was 10À5 M. CD and UV measurements were performed at room
temperature and are baseline corrected and smoothed.
4.1.1. (RS)-2-Hydroxy-2-(D5)phenylacetic acid (RS)-8
Prepared as reported in the literature.26 Yield: 82%. Spectromet-
ric and spectroscopic data were in agreement with the literature.26
4.1.2. (RS)-2-(2,3,4,5,6-Pentafluorophenoxy)-2-(phenyl-d5)acetic
acid (RS)-5
4. Experimental
4.1. General
A mixture of (RS)-mandelic acid-d5 (RS)-8, (1.0 g, 6.37 mmol),
hexafluorobenzene (2.8 mL, 24.2 mmol), CuI (7%), and Cs2CO3
(5.2 g, 16.2 mmol) in butyronitrile (10 mL) was heated to 95 °C
under an N2 atmosphere for 24 h. The solvent was then removed
under vacuum; the residue was taken up with water and washed
twice with EtOAc. The aqueous phase was acidified with citric acid
and extracted three times with EtOAc. The combined organic
phases were dried over anhydrous Na2SO4 and concentrated under
vacuum to give 1.18 g of the desired product as a brownish solid
All chemicals were purchased from Sigma–Aldrich or Lancaster.
Solvents were RP grade unless otherwise indicated. Chloroquine 1,
hydroxychloroquine 2, and mefloquine 3 were recovered by
extraction from tablets of Chloroquine BayerÒ, PlaquenilÒ, and
LariamÒ, respectively. Compound 19 was prepared as previously
reported.28 The reactions under microwave irradiation were car-
ried out at constant temperature in a CEM Discover BenchMate

1610
M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
Figure 4. Enantiodiscrimination for aromatic protons of mefloquine 3 in toluene-d8
with an analyte/CSA molar ratio of 1:1.5 (a), 1:2 (b), 1:1 (c), 1:1 by adding 0.2 mL of
CD2Cl2 (d).
Figure 3. Enantiodiscrimination for aromatic protons of mefloquine 3 in CDCl3 with
0.044 mol) was added and a crude solid formed (Scheme 2). The
mixture was heated to 70 °C to dissolve the solid and then slowly
cooled to room temperature. The obtained crystalline salt was col-
an analyte/CSA molar ratio of 1:1.5 (a), 1:2 (b), 1:1 (c), 1:0.7 (d).
(57%): 1H NMR (CDCl3): d 5.79 (s, 1H); 13C NMR (irradiated at
lected by filtration and washed with cold ethanol (30 mL) to give
282.32 MHz for 19F-decoupling): d 82.4 (d, J
= 151.9 Hz, 1C),
= 24.7 Hz, 2C),
C
20
1H—13
C
(À)-5Á(R)-PEA diastereomeric salt {3.5 g, mp: 178–180 °C; [
a]
=
D
127.5 (t, J
= 24.4 Hz, 2C), 128.8 (t, J
2H—13
2H—13
C
À75 (c 2, MeOH)}. This diastereomeric salt was recrystallized from
129.9 (t, J
= 24.6 Hz, 1C), 131.3 (1C), 133.3 (1C), 136.4 (2C),
2H—13
C
ethanol (35 mL) to afford 1.4 g of (À)-5Á(R)-PEA as white crystals
20 = À97 (c 1.5, MeOH)}. This salt was dissolved
140.2 (2C), 143.4 (1C), 173.9 (1C); GC–MS (70 eV) (methyl ester)
{mp: 184–185 °C; [
a
]
D
m/z (%) 278 (M+À59, 17), 126 (100).
in H2O and the resulting solution was acidified with 2 M HCl. The
liberated acid was extracted by ethyl acetate and the combined
organic phases were dried over Na2SO4 and removed under reduced
pressure. (À)-O-Pentafluorophenyl mandelic acid-d5 (À)-5 was
obtained as a beige solid {1.02 g, 14% yield based on (À)-O-pentaflu-
orophenyl mandelic acid-d5 in the starting racemic mixture; mp:
4.2. Resolution of (RS)-2-(2,3,4,5,6-pentafluorophenoxy)-2-
(phenyl-d5)acetic acid (RS)-5
To a solution of racemic O-pentafluorophenyl mandelic acid-d5
(14.2 g, 0.044 mol) in 50 mL of ethanol, (R)-PEA (5.66 mL,
86–88 °C; [
a]
20 = À151 (c 2, CHCl3), [
a]
20 = À135 (c 1, MeOH); 98%
D
D

M.M. Cavalluzzi et al. / Tetrahedron: Asymmetry 25 (2014) 1605–1611
1611
ee, 1H NMR. Anal. Calcd for (C14H2D5F5O3): C, 52.02; H, 0.62; D, 3.12.
Found: C, 52.21; H + D, 2.84}.
3. Shetty, P. Nature 2012, 484, S14–S15.
4. Dondorp, A. M.; Yeung, S.; White, L.; Nguon, C.; Day, N. P.; Socheat, D.; von
Seidlein, L. Nat. Rev. Microbiol. 2010, 8, 272–280.
5. Fairhurst, R. M.; Nayyar, G. M.; Breman, J. G.; Hallett, R.; Vennerstrom, J. L.;
Duong, S.; Ringwald, P.; Wellems, T. E.; Plowe, C. V.; Dondorp, A. M. Am. J. Trop.
Med. Hyg. 2012, 87, 231–241.
6. Karle, J. M.; Olmeda, R.; Gerena, L.; Milhous, W. K. Exp. Parasitol. 1993, 76, 345–
351.
7. Ding, J.; Hall, D. G. Angew. Chem., Int. Ed. 2013, 52, 8069–8073.
8. Toovey, S. Travel Med. Infect. Dis. 2009, 7, 2–6.
9. Schlagenhauf, P. J. Travel Med. 1999, 6, 122–133.
10. Choong, E.; Dobrinas, M.; Carrupt, P.-A.; Eap, C. B. Expert Opin. Drug Metab.
Toxicol. 2010, 6, 953–965.
11. Fink, E.; Minet, G.; Nickel, P. Arzneimittelforschung-Drug Res. 1979, 29, 163–164.
12. Augustijns, P.; Peeters, L.; Croes, K.; Verbeeck, R. K. Eur. J. Drug Metabol.
Pharmacokinet. 1999, 24, 105–108.
13. Tagoe, C. N. B.; Ofori-Adjei, D. Teratology 1995, 52, 137–142.
14. Ducharme, J.; Fieger, H.; Ducharme, M. P.; Khalil, S. K. W.; Wainer, I. W. Br. J.
Clin. Pharmacol. 1995, 40, 127–133.
15. Tucker, G. T. Lancet 2000, 355, 1085–1087.
16. Amin, N. C.; Blanchin, M.-D.; Aké, M.; Fabre, H. J. Chromatogr., A 2012, 1264, 1–
12.
17. Magalhães, I. R. S.; Bonato, P. S. Curr. Pharm. Anal. 2010, 6, 15–29.
18. Wongwan, S.; Scriba, G. K. E. Electrophoresis 2011, 32, 2669–2672.
19. de Oliveira, A. R. M.; Cardoso, C. D.; Bonato, P. S. Electrophoresis 2007, 28, 1081–
1091.
The partially resolved (+)-O-pentafluorophenyl mandelic acid-
d5 was easily recovered from the mother liquor (+)-5Á(R)-PEA
{mp: 118–119 °C; [a]
20 = +29 (c 2, MeOH)} and used for further
D
resolution. Thus, 8.55 g of (+)-5 were treated with (S)-PEA as the
resolving agent. After the first recrystallization from ethanol
(30 mL), 5.0 g of diastereomeric salt was obtained {mp: 177–178
°C; [
ethanol (24 mL) to give 2.66 g of (+)-5Á(S)-PEA {mp: 176–177 °C;
20 = +86 (c 2, MeOH)}. Finally, applying the above extraction
a]
20 = +78 (c 2, MeOH)}. This salt was recrystallized from
D
[a]
D
procedure to the so-obtained salt, (+)-O-pentafluorophenyl
mandelic acid-d5 [(+)-5]was obtained as a beige solid {1.93 g, 27%
yield based on (+)-O-pentafluorophenyl mandelic acid-d5 in the
starting racemic mixture, mp: 87–88 °C; [a]
20 = +119 (c 2, CHCl3);
D
96% ee, 1H NMR. Anal. Calcd for (C14H2D5F5O3): C, 52.02; H, 0.62;
D, 3.12. Found: C, 52.06; H + D, 2.60}.
Acknowledgments
20. Németh, K.; Tárkányi, G.; Varga, E.; Imre, T.; Mizsei, R.; Iványi, R.; Visy, J.;
Szemán, J.; Jicsinszky, L.; Szente, L.; Simonyi, M. J. Pharm. Biomed. Anal. 2011,
54, 475–481.
21. Carroll, F. I.; Blackwell, J. T. J. Med. Chem. 1974, 17, 210–219.
22. Ansari, A. M.; Craig, J. C. J. Chem. Soc. Perkin Trans. 2 1994, 2, 1731–1734.
23. Schmidt, M.; Sun, H.; Leonov, A.; Griesinger, C.; Reinscheid, U. M. Magn. Reson.
Chem. 2012, 50, S38–S44.
This work was accomplished thanks to the financial support of
the Ministero dell’Istruzione, dell’Università e della Ricerca (MIUR
2010FPTBSH _002) and Assessorato alla Qualità dell’Ambiente
(Regione Puglia). The authors are indebted to Prof. Renzo Luisi
and Dr. Laura Carroccia who donated compound 19. The authors
thank Antonio Palermo for technical help and Dr. Ottavio Ungaro
for his support in several steps of the work.
24. Carbonara, G.; Carocci, A.; Fracchiolla, G.; Franchini, C.; Lentini, G.; Loiodice, F.;
Tortorella, P. ARKIVOC 2004, part v, 5–25.
25. Cavalluzzi, M. M.; Bruno, C.; Lentini, G.; Lovece, A.; Catalano, A.; Carocci, A.;
Franchini, C. Tetrahedron: Asymmetry 2009, 20, 1984–1991.
26. Bruno, C.; Lentini, G.; Lovece, A.; Cavalluzzi, M. M.; Carocci, A.; Catalano, A.;
Franchini, C. J. Chem. 2013. http://dx.doi.org/10.1155/2013/386238. article
number 386238.
References
1. Biamonte, M. A.; Wanner, J.; Le Roch, K. G. Bioorg. Med. Chem. Lett. 2013, 23,
2829–2843.
2. WHO, World Malaria Report 2012, http://www.who.int/malaria/publications/
world_malaria_report_2012/en/index.html; accessed july 2014.
27. Job, G. E.; Buchwald, S. L. Org. Lett. 2002, 4, 3703–3706.
28. Tan, B.; Zeng, X.; Lu, Y.; Chua, J.; Zhong, G. Org. Lett. 2009, 11, 1927–1930.