Preparation, conformational analysis and behaviour as chiral solvating agents of 9-anthrylpentafluorophenylmethanol enantiomers: Study of the diastereomeric association

Article abstract of DOI:10.1016/j.tetasy.2004.03.045

The synthesis, structure and behaviour as chiral solvating agents of the enantiomers of 9-anthrylpentafluorophenylmethanol is reported.

Full text of DOI:10.1016/j.tetasy.2004.03.045

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 1615–1621
Preparation, conformational analysis and behaviour as chiral
solvating agents of 9-anthrylpentafluorophenylmethanol
enantiomers: study of the diastereomeric association
a
Miriam Perez-Trujillo, Albert Virgili and Elies Molins
a,
b
*
ꢀ
a
ꢀ
ꢁ
Departament de Quımica Universitat Autonoma de Barcelona, 08193 Bellaterra, Barcelona, Spain
bInstitut de Ciꢁencia dels Materials de Barcelona (CSIC), Campus de la UAB, 08193 Cerdanyola, Spain
Received 11 March 2004; accepted 30 March 2004
Available online
Abstract—The synthesis, structure and behaviour as chiral solvating agents of the enantiomers of 9-anthrylpentafluorophenyl-
methanol is reported.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
Several alcohols containing anthracene and trifluoro-
methyl groups, such as Pirkle’s alcohol1 or a,a0-bis(tri-
fluoromethyl)-9,10-anthracenedimethanol,2 which are
commercially available, are used as chiral solvating
agents (CSA). The presence of a bulky group, such as
tert-butyl3 or adamantyl, transmits structural rigidity
that compensates for the lack of acidity in the methinic
proton. The presence of the aromatic group (anthra-
cene) in the structure is one of the factors that allows the
CSA to differentiate between enantiomers. Its proximity
and relative position to the substrate affords an impor-
tant and differential magnetic influence. The per-fluoro-
phenyl group has a strong electron withdrawing char-
acter and its presence in the CSA could increase the
zones around the enantiomers where the magnetic field
of the NMR becomes modified by the presence of the
anisotropic groups. The preparation of the enantiomers
of 9-anthrylpentafluorophenylmethanol 1 gives us a
novel CSA with new interactions and new associations
that offer new possibilities for enantiodistinction.
2.1. Preparation and characterization of 9-anthrylpenta-
fluorophenylmethanol
The preparation of the enantiomers of 9-anthrylpenta-
fluorophenylmethanol was proposed in two ways
(Scheme 1): the preparation and separation of the
racemic compound and the use of enantioselective
reactions to prepare each of the enantiomers directly. In
both cases the starting point would be the same inter-
mediate, the lithium derivative of anthracene that is
obtained via reaction of 9-bromoanthracene with
F
F
F
H
HO
F
F
F
F
F
Li
Br
11
F
CHO
F
1
4
8
5
nBu-Li
ether
ether
1
F
F
F
F
COCl
Swern oxidation
F
F
F
F
F
F
F
F
F
H
HO
O
F
CBS reduction
F
1
2
* Corresponding author. Tel.: +34-935812924; fax: +34-935811265;
e-mail: albert.virgili@uab.es
Scheme 1. Synthesis of 9-anthrylpentafluorophenylmethanol.
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.03.045

1616
M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
n-BuLi. Racemic 1 was obtained by a subsequent reac-
tion with pentafluorobenzaldehyde in a global yield of
60%. The enantioselective pathway was based on the
enantioreduction of prochiral ketones with boranes
catalyzed by chiral oxazaborazolidines (CBS reduc-
tion).4 We prepared the 9-anthrylperfluorophenyl
ketone 2 by reaction of anthryl-lithium with perflu-
orobenzoyl chloride in 55% yield. We also prepared
compound 2 in an almost quantitative yield (95%), by
the oxidation of alcohol 1 with dimethyl sulfoxide (Swern
oxidation).5 Although the CBS enantio- reduction of 9-
anthryltrifluoromethyl ketone has already been
described6 with a very good yield and enantiomeric ex-
cess, which we have repeated with similar results, an
analogous reaction with ketone 2 was not possible after
testing different conditions. Moreover, the reduction of
this compound with NaBH4 or LiAlH4 was not viable,
thus showing the low electrophilic character of this ke-
tone.
parative Whelk-O2 (250 · 25 mm) column using hexane/
isopropyl alcohol (97/3, 3 mL/min) and (98/2, 12 mL/
min, respectively), as the elution solvent. The first eluted
compound was dextro and was further assigned to (+)-
(S)-3. The second one was laevo and corresponded to
())-(R)-3. Treatment of each isolated ester with KOH
(0.05 M) gave the enantiomers (+)-(S)-1 and ())-(R)-1,
respectively.
The X-ray structure (Fig. 2)8 of the first eluted isomer
was obtained and provisionally assigned as the (S)-
enantiomer. The asymmetric unit was composed by two
molecules that show
a parallelism between each
anthracene and one of the perfluorophenyl rings. Two
kinds of attractive p–p stacking interactions were seen:9
one being a face to face parallel displaced between one
anthracene ring and one perfluorophenyl ring, while the
other being the edge to face showed by the two
anthracene rings and between the two perfluorophenyl
rings.
9-Anthrylpentafluorophenylmethanol 1 was character-
ized by NMR. At room temperature, the 1H NMR
spectrum showed well-resolved signals for all the pro-
tons of the molecule. Nevertheless, a significant shift of
several peaks was observed when the concentration of
the sample changed (Fig. 1). For example, methinic
proton H11 shifts 0.13 ppm when the concentration of
the sample decreases from 0.075 to 0.007 M possibly due
to autoassociation processes.
The absolute configuration of each enantiomer was
1
deduced by the different H NMR spectra of the dia-
stereoisomeric (S)-a-methoxyphenylacetic acid esters as
reported elsewhere.10 The structural study11 of these
esters showed a conformational equilibrium between the
syn-periplanar (sp) and anti-periplanar (ap) conforma-
tions of the a-carboxylic bond of the acid part. The first
(sp) was more stable and was favoured by lowering the
temperature. The absolute configuration of each isolated
isomer was determined by considering the differences in
the chemical shifts of the protons and the distribution of
the groups around of the sp conformation of each dia-
stereoisomer.
The restricted rotation about the C9–C11 (sp2–sp3) bond
in (9-anthryl)carbinols derivatives7 reflected in separated
signals for absorption of H1 and H8 is well known. We
have reported 13 kcal/mol for tert-butyl and 9.8 kcal/mol
for the phenylderivatives. Nevertheless, in contrast to
most of these molecules, the 1H NMR spectrum of
compound 1 only shows one doublet corresponding to
the proton pair H1 and H8. After reducing the temper-
ature to 190 K, the two protons were still indistin-
guishable, thus confirming that the rotational barrier
about the sp2–sp3 bond is lower than 9 kcal/mol.
(S)-a-Methoxyphenylacetates of the enantiomers of 9-
anthrylpentafluorophenylmethanol 1 were prepared by
reaction of each enantiomeric alcohol (R)-1 and (S)-1
with the (S)-a-methoxyphenylacetyl chloride, obtaining
the stereoisomers (SR)-4 and (SS)-4. The same com-
pounds were obtained by a similar preparation of the
mixture from racemic 1 and subsequent separation by
HPLC with the Whelk-O2 column using hexane/iso-
propyl alcohol (95/5, 10 mL/min).
Enantiomers of 1 were separated by Chiral HPLC.
Compound 3, the acetate derivative of 1, was quanti-
tatively prepared by treatment with acetyl chloride. The
pure enantiomers (R)-3 and (S)-3 were separated on a
semi-preparative Whelk-O1 (250 · 10 mm) and a pre-
C x10 -2
M
H 2,7
H 4,5
H 10
H 11
H 3,6
H 1,8
0.69
0.80
0.96
1.20
1.60
2.40
3.00
4.10
6.10
7.50
8.50 8.40 8.30 8.20 8.10 8.00 7.90 7.80 7.70 7.60 7.50 7.40 7.30
(ppm)
Figure 2. Ortep drawing of the X-ray diffraction analysis of acetate
(R)-3.
Figure 1. 1H NMR spectra of 1 at several concentrations (CDCl3).

M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
1617
The anisotropy of the anthryl group determines the local
magnetic fields nearby and the position of the peaks in
the NMR spectra. Considering only the syn conforma-
tions, in the first case, the (S,S)-4, methoxy group and
H12 are on the same side of the anthracene group being
strongly shielded, corresponding to the NMR spectrum
of ester of dextro alcohol (+)-(S)-1. In the (S,R)-4 is the
phenyl group that is influenced by the anisotropy of the
anthracene ring being shielded to low frequencies and
corresponding to the spectrum of the ester of laevo
alcohol ())-(R)-1.
H11
H10
CHCl3
H1/8
H4/5
B
A
8.70
8.60
8.50
8.40
8.30
8.20
8.10
8.00
(ppm)
2.2. Enantiorecognition activity
H3/6 and H2/7
The activity as a chiral solvating agent of enantiomers of
1 was studied against racemic compounds 1-(1-naph-
thyl)ethylamine 5 and 1-aminoindane 6 (Fig. 5). In all
cases, the presence of a basic group and the aromatic
rings allowed the formation of an intermolecular
hydrogen bond and a p-stacking stabilization interac-
tion. The experiments were carried out in CDCl3, adding
portions of the CSA to a solution (0.05 M) of racemic
substrate, until maximum differentiation of the corre-
sponding enantiomer’s signals occurred.
Ph
Ph
B
A
7.60
7.52
7.44
7.36
(ppm)
7.28
7.20
7.12
H12
CH3
B
A
3.50 3.46 3.42 3.38 3.34 3.30 3.26
(ppm)
4.98
4.94
4.90
(ppm)
4.86
4.82
Figure 3. Portions of the 1H NMR spectrum (at 300 K) of stereoi-
somers of (S)-a-methoxyphenylacetate of 9-anthrylpentafluor-
ophenylmethanol 4. (A) Ester of ())-1, (S,R)-4 and (B) ester of (+)-1,
(S,S)-4.
1
Figure 3 shows the H NMR spectra of both esters of 4
where very different chemical shifts for the same proton
of each diastereoisomer are shown. The ester of laevo
isomer ())-1 shows the H12 and CH3 deshielded while
the phenyl protons, H11 and the most anthracene pro-
tons are shielded in comparison with those of the dextro
alcohol (+)-1. Low temperature 1H NMR spectra
showed greater differences.
Figure 5. Substrates tested for the enantiodiscrimination.
In the case of substrate (RS)-5, the use of alcohol (S)-1
as a chiral solvating agent afforded good results
obtaining a good resolution of the signals corresponding
to the H9 and methyl group. This methyl group is only
occasionally12 separated.
The equilibrium between the syn-periplanar and anti-
periplanar conformations for both stereoisomers is
shown in Figure 4. The syn conformations are the most
stable ones11 and are favoured by the low temperatures.
1
Figure 6 shows the evolution of the H NMR spectrum
of the 1-(1-naphthyl)ethylamine 5 when the addition of
CSA (S)-1 was incremented. The doublet corresponding
to the methyl group in the racemic compound appears
as two completely separated doublets of the two
enantiomers when 1.8 equiv (or more) of (S)-1 are ad-
ded. After the addition of an excess of one of the
enantiomers of 5, (R)-5, and considering the integration
of the two signals, we confirmed that the methyl of the
(R)-5 enantiomer shifted to higher field than the methyl
of the (S)-5. Similarly, at the same ratio, the quartet of
the H9 results in two quartets, one of each enantiomer,
and in a similar arrangement than before.
H11
H11
Ph
O
H3CO
O
H12
C6F5
Antr
C6F5
Antr
S
S
S
S
O
O
Ph
H3CO
H12
ap-conformation of (S,S)-4
sp-conformation of (S,S)-4
H11
H11
Ph
O
H3CO
O
H12
Antr
C6F5
Antr
C6F5
S
R
S
O
R
O
Ph
H3CO
H12
ap-conformation of (S,R)-4
sp-conformation of (S,R)-4
The maximum separation obtained corresponded to the
addition of 3.8 equiv of (S)-1. Further additions of
Figure 4. Conformational equilibrium of stereoisomers of (S)-a-
methoxyphenylacetates of 9-anthrylpentafluorophenylmethanol 4.

1618
M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
R
S
H
9
2.4
1.8
1.4
1.0
0.6
0.0
[(S )-1]
5.10 5.00 4.90 4.80 4.70 4.60 4.50 4.40
(ppm)
0
Figure 8. Evolution of the enantiodifferentiation of H2, H2 , H3 and
0
H3 of ( )-aminoindane 6 when several quantities of (S)-1 were added.
R
S
CH
3
2.4
when all components are completely dissolved, Job’s
method13 can be applied to calculate the stoichiometry
of the complex in solution. Complex (S)-1Æ(S)-5 was
analyzed: 10 samples of a constant total concentration
were prepared containing variable ratios of the two
1.8
1.0
0.6
0.2
1
components. H NMR spectra of these samples were
recorded at 300 K and the chemical shift variations were
observed for several protons of compound (S)-5. The
Job plot of the variation of the chemical shift of some
amine protons H2, H8 and CH3 versus the ratio between
the concentration of amine (S)-5 and the total concen-
tration (0.01 M) is shown in Figure 9. The maximum
value of the parabolic curve in the X coordinate gives us
the stoichiometry. In the three plots, a value of 0.5 was
found, meaning that the stoichiometry of the complex is
1:1.
0.0
1.60 1.56 1.52 1.48 1.44 1.40 1.36 1.32
(ppm )
Figure 6. Aliphatic part of 1H NMR spectrum of the enantiodis-
crimination process of racemic 5 by the addition of several portions of
(S)-9-anthrylpentafluorophenylmethanol (S)-1.
solvating agent precipitate of the components. Figure 7
shows the evolution of the difference obtained when
portions of the (S)-1 were added to a solution of racemic
1-(1-naphthyl)ethylamine 5.
H
8
10.0
8.0
6.0
4.0
2.0
0.0
H
2
CH
3
[(S)-1]
Figure 7. Evolution of the enantiodifferentiation of H9 and CH3 of
rac-1-(1-naphthyl)ethylamine 5 when several quantities of (S)-1 were
added.
0.0
0.5
[5]/0.01
1.0
The distinction of the enantiomers o0f 6 could be fol-
Figure 9. Job plot of the association of (S)-9-anthrylpentafluorophen-
ylmethanol (S)-1 with (S)-1-(1-naphthyl)ethylamine (S)-5.
0
lowed in protons H2, H2 , H3 and H3 . Figure 8 shows
the enantiodiscrimination of these protons when (S)-1
was added to a racemic mixture. Although the signals of
aromatic protons of 6 also undergo a differentiating
shift, it was impossible to follow because of its com-
plexity.
As is well known, the chemical shift differences observed
when an association complex is formed is a consequence
of two factors, the difference in the chemical shift of the
complexes (i.e., the different geometry of the associa-
tion) and the different populations of the complex (i.e.,
the difference in the binding constants K.
Due to the quality of the results obtained in the enan-
tiodiscrimination of (RS)-1-(1-naphthyl)ethylamine 5,
the complexes responsible for the phenomena were
studied. Stoichiometry of the complex was the first
parameter studied. As the chemical shift of a nucleus is a
mean value of the several species present in solution,
Thus, the binding constant of each diastereomeric
complex should be taken into account. The binding
constants of the complexes formed between the enantio-

M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
1619
mers of amine 5 and alcohol (S)-1 were determined
using the equimolar method, which according to Bou-
quant and Chuche,14 can be applied if the stoichiometry
of the complex is 1:1.
The equimolar method also allows us to measure the
chemical shift of several protons in the isolated associ-
ated state. The intersection of the straight lines of each
proton and each enantiomer with the axis gives the
corresponding dc. Figure 11 affords the plot corre-
sponding to the proton H9 of both enantiomers of 5 at
several temperatures. The various lines (corresponding
to one enantiomer at several temperatures) converge on
a mean value of the chemical shift of the H9 in the
association complex. The values for various protons,
together the experimental chemical shift for compound
( )-5, are presented in Table 2 meaning a standard and
comparative value.
A sample formed by a solution (0.05 M) of one of the
enantiomers of 5 and (R)-1 (0.05 M) was prepared in
CDCl3. This sample was diluted (four times) by addi-
tions of 0.1 mL of CDCl3 and the 1H NMR spectra
registered at four temperatures. Figure 10 shows the plot
of the variation of the chemical shifts (Dd) of the protons
1=2
of amine 5 (H2, H9 and CH3) versus (Dd=S0Þ at 285 K,
where S0 is the initial concentration of the solute and dc
is the chemical shift of each proton in the pure complex.
The binding constant for every proton studied was
determined from the gradient [ðdc=KÞ1=2] of the plotted
straight lines. The average values and the free enthalpies
obtained at several temperatures are expressed in
Table 1.
∆δ
)
(H9
0.50
0.40
0.30
0.20
0.10
0.00
S 255
R 255
S 270
R 270
S 285
R 285
R 300
S 300
0.80
1.30
1.80
2.30
2.80
3.30
(∆δ/S)1/2
Figure 11. Measurement of the binding constant of the association
between each enantiomer of 1-(naphthyl)ethylamine 5 and (R)-9-
anthrylpentafluorophenylmethanol (S)-1 measuring the variation of
H9 at 300, 285, 270 and 255 K.
Table 2. Extrapolated chemical shift of several protons of (R)-5 and
(S)-5 of the complex formed with (S)-1
H
dH
(S)-1Æ(S)-5
(S)-1Æ(R)-5
Figure 10. Straight lines obtained when an equimolar method is
applied to the determination of the binding constant between each of
the enantiomers of amine 5 with (S)-1 at 285 K.
H9
H4
4.960
7.750
7.652
1.548
4.513 0.013
7.672 0.003
7.278 0.005
1.277 0.006
4.399 0.012
7.683 0.007
7.356 0.021
1.265 0.008
H2
CH3
Table 1. Binding constants of the complex formation between each
enantiomer of 1-(1-naphthyl)ethylamine 5 and (S)-9-anthrylpentaflu-
orophenylmethanol (S)-1
3. Experimental
T/K
(R)-5Æ(S)-1
(S)-5Æ(S)-1
K/Mꢀ1
DG/kJ/mol
K/Mꢀ1
DG/kJ/mol
NMR spectra were recorded at 400.13 and 500.13 MHz
1
300
285
270
255
7.4 1.4
12.3 1.1
30.9 2.9
72.1 5.4
)5.0 0.2
)6.0 0.1
)7.7 0.1
)9.1 0.1
7.2 0.7
14.2 0.9
32.0 2.3
73.1 3.9
)4.9 0.4
)6.3 0.2
)7.8 0.2
)9.1 0.2
for H. The temperature was controlled to 0.1 ꢁC. The
complete identification of the NMR signals was carried
out with the aid of several 1D (NOE) and 2D (COSY,
HMQC and HMBC) spectra.
The NMR titration method was carried out with 0.5 mL
of a solution 0.05 M of the tested compounds 5 and 6.
After addition (at constant volume) of several portions
of 0.2–0.5 equiv of CSA 1, NMR spectra were measured
and the variations of the chemical shifts calculated for
each addition. The measurements were continued until
maximum enantiodiscrimination (2.5–4.4 equiv).
The similarity of the constant values means that the
enantiodiscrimination takes place because of the differ-
ences between the structures of the complexes of asso-
ciation; the factor of the concentration being a minor
cause. As the process is exothermic, the binding constant
increases when the temperature decreases. While with
other chiral solvating agents,2 when decreasing the
temperature the difference between constants of the
enantiomers increases considerably, in the present case
the uniformity is maintained with low temperatures not
enhancing the enantiodiscrimination.
Chiral semi-preparative HPLC was carried out using a
(R,R)-Whelk-O1 (250 · 10 mm) column and a pre-
parative HPLC using a (R,R)-Whelk-O2 (250 · 25 mm)
column.

1620
M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
1
3.1. 9-Anthrylpentafluorophenyl ketone 2
kmax (nm) (ethanol): 214, 248, 368. H NMR (CDCl3,
300 K, 400.13 MHz): d (ppm) 8.46 (s, 1H, H10), 8.43 (d,
J1;2 ¼ J8;7 ¼ 8:8 Hz, 2H, H1 and H8), 8.00 (d,
J4;3 ¼ J5;6 ¼ 7:6 Hz, 2H, H4 and H5), 7.51 (s, 1H, H11),
7.49 (m, 2H, H2 and H7), 7.45 (m, 2H, H3 and H6), 3.28
(s, 1H, OH). 13C NMR (CDCl3, 300 K, 100.62 MHz): d
(ppm) 131.5 (C9), 129.6 (C10), 129.5 (C4 and C5), 126.5
(C2 and C7), 124.9 (C3 and C6), 123.9 (C1 and C8), 66.2
(C11).
Butyllithium (1.63 mL, 2.5 M in hexane) was slowly
added to a solution of 9-bromoanthracene (0.700 g,
2.7 mmol) in anhydrous diethyl ether (30 mL) and in a
stream of N2. The mixture was stirred for 0.5 h and then
added to a solution of pentafluorobenzoyl chloride
(1.18 mL, 8.2 mmol in 15 mL of dried diethylether) at
)78 ꢁC. The resulting mixture was stirred at )78 ꢁC for
5 h. The reaction was treated with a saturated solution
of NH4Cl (2 · 15 mL), a 10% NaOH solution (2 · 15 mL)
and water (2 · 15 mL). The organic layer was dried over
anhydrous MgSO4, evaporated and then purified by
flash chromatography (hexane/CH2Cl2, 8:2). Ketone 2
(0.55 g, 55%) was obtained: mp: 118–122 ꢁC. IR (KBr)
cmꢀ1: 2921, 1671–1650, 1518, 1490. EM m=z (%): 372
(100), 205 (58), 177 (72), 167 (8). 1H NMR (CDCl3,
300 K): d (ppm) 8.65 (s, 1H, H10), 8.10 (m, 2H, H4 and
H5), 7.90 (m, 2H, H1 and H8), 7.54 (m, 4H, H2, H7, H3
and H6). 13C NMR (CDCl3, 300 K): d (ppm) 133.4 (C8a
and C1a), 131.0 (C4a and C5a), 130.9 (C10), 129.1 (C4 and
C5), 128.3 (C9), 127.7 (C2 and C7), 125.6 (C3 and C6),
123.7 (C1 and C8). EM m=z (%): 372 (100), 205 (58), 177
(72), 167 (8). Anal. Calcd for C21H9F5O: C, 67.75; H,
2.44. Found: C, 67.67; H, 2.51.
The pure enantiomers were obtained after hydrolysis of
the isolated acetate derivatives 3.
25
(+)-1: ½aꢁ ¼ þ32 (c 2, ethyl acetate)
D
25
())-1: ½aꢁ ¼ ꢀ32 (c 2, ethyl acetate)
D
3.3. (RS)-9-Anthrylpentafluorophenyl acetate 3
Dimethylaminopyridine (DMAP) (107 mg, 0.87 mmol),
triethylamine (4.26 mL, 30.7 mmol) and acetyl chloride
(0.80 mL, 11.2 mmol) were added to a solution of 1
(822 mg, 2.20 mmol in 35 mL of anhydrous dichloro-
methane). After 3 h, the reaction had finished and the
mixture washed with water (2 · 25 mL), HCl 1 M
(2 · 25 mL) and a solution of 10% NaHCO3 (2 · 25 mL).
The organic layer was separated, dried and evaporated.
The solid residue was purified by chromatography on
silica gel (hexane/CH2Cl2 6:4). Acetate 3 (0.79 g, 86%)
was obtained: mp: 154–158 ꢁC. Anal. Calcd for
C23H13F5O2: C, 66.35%; H, 3.15%. Found: C, 65.96%;
H, 3.01%. IR (KBr) cmꢀ1: 2963, 2927, 2855, 2963, 1745,
1524, 1496, 1219. EM m=z (%): 416 (379),373 (117), 357
(206), 178 (365). UV kmax (nm) (ethanol): 214, 250, 366.
1H NMR (CDCl3, 300 K, 400.13 MHz): d (ppm) 8.61 (s,
1H, H11), 8.52 (s, 1H, H10), 8.45 (d, J1;2 ¼ J8;7 ¼ 8:8 Hz,
2H, H1 and H8), 8.03 (d, J4;3 ¼ J5;6 ¼ 8:2 Hz, 2H, H4 and
H5), 7.54 (dd, J2;1 ¼ J7;8 ¼ 8:8 Hz, 2H, H2 and H7), 7.47
(dd, J3;4 ¼ J6;5 ¼ 8:2 Hz, 2H, H3 and H6), 2.18 (s, 3H,
CH3). 13C NMR (CDCl3, 300 K, 100.62 MHz): d (ppm)
170.0 (C@O), 131.4 (C9), 130.3 (C10), 130.0 (C4a and
C5a), 129.5 (C4 and C5), 127.0 (C1a and C8a), 126.9 (C2
and C7), 125.0 (C3 and C6), 123.7 (C1 and C8), 65.8(C11),
20.8 (CH3).
Synthesis of 2 from alcohol 1 was carried out by a Swern
oxidation. To
a
solution of DMSO (0.50 mL,
7.05 mmol) in anhydrous dichloromethane (7 mL) at
)78 ꢁC and kept under N2, trifluoroacetic anhydride
(0.97 mL, 6.85 mmol) was added. After 15 min a solution
of racemic 9-anthrylpentafluorophenylmethanol
1
(0.256 g, 0.68 mmol) and DMSO (1 mL, 14.09 mmol) in
4 mL of anhydrous dichloromethane was slowly added
over the trifluoroacetic anhydride solution. The reaction
mixture was stirred for 1 h and 25 min and then trieth-
ylamine (2.20 mL, 15.76 mmol) was added. The reaction
mixture was still stirred for 1 h at )78 ꢁC, washed with
water (2 · 10 mL) and with a saturated solution of NaCl
(2 · 10 mL) and the organic layer dried over anhydrous
MgSO4 and then concentrated. The solid residue was
purified by column chromatography on silica gel (hex-
ane/dichloromethane, 9:1) to give compound 2 (0.240 g,
95% yield).
3.4. (S)-a-Methoxyphenylacetate of anthrylpentafluoro-
phenylmethyl 4
3.2. (RS)-9-Anthrylpentafluorophenylmethanol 1
DMAP (0.004 g, 0.033 mmol in 2 mL of dried dichlo-
romethane) was added all at once to a solution of
0.118 g (0.316 mmol) of the enantiopure alcohol (R)-1,
0.052 g (0.313 mmol) of (S)-a-methoxyphenylacetic acid
and 0.065 g (0.315 mmol) of DCC in 8 mL of dried di-
chloromethane. After 24 h, the dicyclohexylurea was
removed by filtration, the filter cake washed with three
3 mL portions of hexane and the combined filtrates
washed with 2 · 2 mL cold 1 M aqueous hydrochloric
acid, 2 · 2 mL saturated sodium bicarbonate and
2 · 2 mL saturated brine. The organic phase was then
dried over MgSO4 and filtered and the solvent removed
to afford a yellow solid. The crude product was purified
by flash chromatography (CH2Cl2/hexane, 9:1) to give
compound (S,R)-4 (0.11 g, 69%). (S,R)-4: 1H NMR
To a solution of 9-bromoanthracene (0.44 g, 1.7 mmol)
in anhydrous diethyl ether (15 mL) and in a stream of
N2, butyllithium (1.60 mL, 1.6 M in hexane) was slowly
added. The mixture was stirred for 0.5 h and then added
to a solution of pentafluorobenzaldehyde (0.31 mL,
2.5 mmol in 15 mL of dried diethylether) at )78 ꢁC. The
resulting mixture was stirred at )78 ꢁC for 2 h. The
reaction was washed with a saturated solution of NH4Cl
(2 · 25 mL), a 10% NaOH solution (2 · 25 mL) and
water (2 · 25 mL). The organic layer was dried over
anhydrous MgSO4, evaporated and then purified by
flash chromatography (hexane/CH2Cl2 7:3). Alcohol
(RS)-1 (0.380 g, 60%) is obtained: mp: 168–170 ꢁC. IR
(KBr) cmꢀ1: 3308 (O–H, broad), 2963, 1524, 1498, 1112.
EM m=z (%): 374 (26), 207 (8), 179 (100), 168 (5). UV

M. Pꢀerez-Trujillo et al. / Tetrahedron: Asymmetry 15 (2004) 1615–1621
1621
(CDCl3, T ¼ 300 K): d (ppm) 8.66 (s, 1H, H10), 8.50 (s,
1H, H11), 8.33 (m, 2H, H4 and H5), 8.01 (m, 2H, H1 and
H8), 7.44 (m, 4H, H2, H7, H3 and H6), 7.16 (m, 5H,
phenyl group), 4.92 (s, 1H, H12), 3.41 (s, 3H, CH3).
difference Fourier peak and hole were 0.237 and
)0.221 e Aꢀ3. Crystallographic data have been deposited
ꢂ
with the been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication
number CCDC 234752.
Starting from the (S)-enantiomer of alcohol 1, the (S,S)-
4 isomer was obtained. (S,S)-4: 1H NMR (CDCl3,
T ¼ 300 K): d (ppm) 8.63 (s, 1H, H10), 8.55 (s, 1H, H11),
8.40 (d, J4;3 ¼ J5;6 ¼ 8:7 Hz, 2H, H4 and H5), 8.06 (d,
J1;2 ¼ J8;7 ¼ 8:2 Hz, 2H, H1 and H8), 7.52 (m, 4H, H2,
H7, H3 and H6), 7.40 (m, 5H, phenyl group), 4.84 (s, 1H,
H12), 3.32 (s, 3H, CH3).
References and notes
1. Pirkle, W. H.; Hoover, D. J. Top. Stereochem. 1982, 13,
263–331; Gil, J.; Virgili, A. J. Org. Chem. 1999, 64, 7277–
7280.
Both esters, (S,R)-4 and (S,S)-4, could also be achieved
starting from the racemic form of alcohol 1. The reac-
tion described above gives a mixture of the diastero-
meric esters 4, which are separated by HPLC (Welk-O1
column, 95% hexane-5% isopropanol). The first eluted
ester was the (S,R)-4 isomer while the second one was
the (S,S)-4 isomer, both of which were obtained with
100% ee.
ꢀ
2. Pomares, M.; Sanchez-Ferrando, F.; Virgili, A.; Alvarez-
Larena, A.; Piniella, J. F. J. Org. Chem. 2002, 67, 753–758.
ꢀ
3. Moragas, M.; Cervello, E.; Port, A.; Jaime, C.; Virgili, A.;
Ancian, B. J. Org. Chem. 1998, 63, 8689–8695.
4. Itsuno, S. Org. React. 1998, 52, 395–576.
5. Mancuso, A. J.; Huang, S.; Swern, D. J. Org. Chem. 1978,
43, 2480–2482.
6. Corey, E. J.; Bakshi, R. K. Tetrahedron Lett. 1990, 31,
611–614.
7. de Riggi, I.; Virgili, A.; de Moragas, M.; Jaime, C. J. Org.
Chem. 1995, 60, 27–31.
3.5. Crystallography
8. Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
9. Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew.
Chem., Int. Ed. 2003, 42, 1211–1250.
A suitable crystal (0.46 · 0.42 · 0.14 mm3) of (R)-3 was
selected for X-ray single crystal diffraction experiments
and mounted at the tip of a glass fibre on an Enraf–
Nonius CAD4 diffractometer producing graphite
monochromated Mo Ka radiation. After the random
search of 25 reflections, the indexation procedure gave
~ ꢁ
10. Seco, J. M.; Quinoa, E.; Riguera, R. Chem. Rev. 2004, 104,
ꢀ
ꢀ
17–117; Garcıa, R.; Seco, J. M.; Vazquez, S. A.; Quinoa,
E.; Riguera, R. J. Org. Chem. 2002, 67, 4579–4589;
~ ꢁ
~ ꢁ
Latypov, S. K.; Seco, J. M.; Quinoa, E.; Riguera, R.
J. Org. Chem. 1996, 61, 8569–8577.
11. Latypov, S. K.; Seco, J. M.; Quinoa, E.; Riguera, R.
rise to the cell parameters a ¼ 13:700ð3Þ, b ¼ 15:472ð3Þ
~ ꢁ
3
ꢂ
ꢂ
and c ¼ 17:466ð3Þ A. The cell volume was 3702.2(13) A ,
the calculated density 1.49 g/cm3 and the absorption
coefficient 0.13 mmꢀ1. The assigned space group was
P21 21 21 with Z ¼ 4. Data were collected up to 2h ¼ 60ꢁ
in the x–20 scan mode resulting in 5922 reflections.
Absorption correction was performed following the PSI-
scan semi-empirical method. The structural resolution
procedure was made using the WinGX package.15
Solving for structure factor phases was performed by
SIR200216 and the full-matrix refinement, by
SHELXL97.17 Non-H atoms were refined anisotropically
and H-atoms were introduced in calculated positions
and refined riding on their parent atoms. The final R
indices were for I > 2rðIÞ: R1 ¼ 0:0633, wR2 ¼ 0:1218
and for all data: R1 ¼ 0:2023, wR2 ¼ 0:1535. The largest
J. Org. Chem. 1995, 60, 504–515.
~
12. Munoz, A.; Virgili, A. Tetrahedron: Asymmetry 2002, 13,
1529–1534.
13. Job, P. Ann. Chim. 1928, 9, 113–203.
14. Bouquant, J.; Chuche, J. Tetrahedron Lett. 1972, 13, 2337–
2340.
15. Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837–838.
16. Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano, G.
L.; Giacovazzo, C.; Polidori, G.; Spagna, R. Sir2002: a
new Direct Methods program for automatic solution and
refinement of crystal structures. J. Appl. Cryst. 2003, 36,
1103.
17. Programs for Crystal Structure Analysis (Release 97-2).
€
€
Sheldrick, G. M. Institut fur Anorganische Chemie der
€
€
Universitat, Tammanstrasse 4, D-3400 Gottingen, Ger-
many, 1998.