Effect of the presence of a free hydroxyl group on the enantiodiscrimination properties of cholic acid based CSPs bearing 2-naphthylcarbamate and 3,5-dinitrophenylcarbamate moieties in the HPLC resolution of racemic compounds

Article abstract of DOI:10.1016/j.tetasy.2005.10.036

Two new chiral selectors, obtained by derivatizing two of the three hydroxyl groups of cholic acid with 2-naphthylisocyanate and 3,5-dinitrophenylisocyanate, have been prepared and linked to silica gel to obtain chiral stationary phases (CSPs) for the HPL

Full text of DOI:10.1016/j.tetasy.2005.10.036

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 3820–3828
Effect of the presence of a free hydroxyl group on the enantio-
discrimination properties of cholic acid based CSPs bearing
2-naphthylcarbamate and 3,5-dinitrophenylcarbamate
moieties in the HPLC resolution of racemic compounds
Anna Iuliano* and Andrea Ruffini
`
Dipartimento di Chimica e Chimica Industriale, Universita di Pisa, Via Risorgimento 35, 56126 Pisa, Italy
Received 14 September 2005; revised 18 October 2005; accepted 24 October 2005
Abstract—Two new chiral selectors, obtained by derivatizing two of the three hydroxyl groups of cholic acid with 2-naphthyliso-
cyanate and 3,5-dinitrophenylisocyanate, have been prepared and linked to silica gel to obtain chiral stationary phases (CSPs)
for the HPLC separation of enantiomers. The enantiodiscriminating capability of the two CSPs has been compared to that of
the analogous CSP obtained from an exhaustively derivatized cholic acid based selector, in order to establish the effect of the pres-
ence of a free hydroxyl group on the enantiodiscrimination properties of this kind of selector. The chromatographic results dem-
onstrate that the enantioselectivity of these selectors strongly depends on the position of the hydroxyl group on the cholestanic
backbone.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Even better results were obtained by linking to silica gel,
a cholic acid derivative possessing 2-naphthylcarbamoyl
groups as p-basic moiety and a 3,5-dinitrophenylcarba-
moyl unit as p-acid moiety.5 Both the efficiency and ver-
satility of these CSPs were dependent not only on the
nature of the moieties introduced but also on their rela-
tive position on the cholestanic skeleton.3 Indeed the
CSP, which afforded the best results was CSP 1
(Fig. 1) obtained by linking to silica gel the cholic acid
derivative possessing 2-naphthylcarbamoyl moieties at
the 3- and 7-position and 3,5-dinitrophenyl group at
the 12-position of the cholestanic backbone.5
The preparation of broad spectrum independent chiral
stationary phases (CSPs) is an attractive goal for scien-
tists interested in enantioselective HPLC because in
these phases the characteristics of the independent
CSPs1 (known interaction mode with the substrates, pre-
dictability of resolution) are coupled to wide applicabil-
ity. For this reason efforts were made to find structural
features that would guarantee efficiency and versatility
to a chromatographic selector.2
Over the last few years, we have demonstrated that the
use of bile acid derivatives, which possess aromatic units
having different electronic characters, allows us to
obtain CSPs showing both efficiency and versatility.3
In fact, by linking to silica gel a deoxycholic acid deriva-
tive, which had a 3,5-dimethylphenylcarbamoyl moiety
(p-basic) and a 3,5-dichlorophenylcarbamoyl group (p-
acid), a CSP able to separate by HPLC both p-acid
and p-basic racemic compounds with good enantioselec-
tivity factors, was obtained.4
This CSP was able to resolve by HPLC p-acid and p-
basic racemic compounds and also some underivatized
racemates, such as binaphthols and alkylarylcarbinols.5
These encouraging results prompted us to improve fur-
ther the enantiodiscriminating properties of this kind
of CSP by changing some structural characteristics of
the cholic acid derived selector and, hence, we became
interested in checking the effect of the presence of a free
hydroxyl group. Taking into account that the versatility
of CSP 1 is attributable to the presence of electronically
different aromatic groups and that the best relative dis-
position of these groups on the cholic acid skeleton is
that one where 3,5-dinitrophenylcarbamoyl moiety is
*
Corresponding author. Tel.: +39 0502219232; fax: +39
0502219260; e-mail: iuliano@dcci.unipi.it
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.10.036

A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
3821
the cholic acid derived CSP 1, but also to establish what
is the best arrangement of the functional groups to
achieve good enantioselectivity.
NO2
H
O
O2N
N
O
CH3
O
N
2. Results and discussion
2.1. Synthesis of the CSPs
S
O
N
O
The cholic acid based selector of CSP 2 was synthe-
sized as previously described,5 whereas the synthetic
route to selector 11, from which CSP 3 can be prepared,
is summarized in Scheme 1. Due to the introduction of
two different arylcarbamoyl moieties at the 7- and 12-
position of the cholestanic backbone, selective protec-
tion of the 3 and 7 hydroxy groups of the cholic acid
was mandatory. Since the hydroxy group at the 3-posi-
tion of the cholestanic skeleton, which is equatorial is
more reactive than the other two, which are axial, the
selective protection of this group can be obtained easily
because it will react preferentially with respect to the
other hydroxyl groups with a great number of reactants.
Selective protection of the 7-hydroxy group can be
accomplished by means of its regioselective oxidation
with NBS in alkaline solution,6 because it is well known
that the reduction, with NaBH4, of the carbonyl func-
tion takes place with complete stereoselectivity,7 the ste-
reogenic centre being restored to the same absolute
configuration as in cholic acid. Therefore, as shown in
Scheme 1, 4 was reacted with NBS in a bicarbonate solu-
tion and ketoacid 5 was converted into the correspond-
ing N-methylallylamide 6, using the mixed anhydride
method.8
H
H
O
O
O
N
Si
O
H3CO
Figure 1. Structure of CSP 1.
located at the 12-position, two parent CSPs having a
free hydroxyl group can be obtained (Fig. 2).
NO2
H
O2N
N
O
CH3
S
O
O
N
O
O
OH
H
O
N
Si
O
H3CO
The 3-hydroxyl group was protected as an acetate by
reacting 6 with an excess of acetic anhydride at 80 ꢁC.9
The 3-acetyl derivative was obtained in 55% yield, after
chromatographic purification, because conversion of the
substrate in the acetylation reaction was not complete,
even after prolonging the reaction time. The reaction
of 7 with a 1.5-fold excess of 3,5-dinitrophenyliso-
cyanate in refluxing toluene followed by reduction with
NaBH4 afforded 9 with complete stereoselectivity. In
order to derivatize the hydroxyl group as a 2-naphthyl-
carbamate, 9 was reacted with 2 equiv of 2-naphthyliso-
cyanate in refluxing toluene for 24 h. These reaction
conditions allowed us to obtain compound 10 in 75%
yield, after chromatographic purification. Selector 11
was eventually obtained by hydrolysis of the acetyl
group by means of HCl in refluxing methanol.
CSP 2
NO2
H
O
O2N
N
O
CH3
S
O
N
HO
O
O
O
N
H
Si
O
H3CO
CSP 3
Selectors 11 and 12 were covalently linked to silica gel,
as described in Scheme 2, by means of reaction with a
5-fold excess of 3-mercaptopropyltrimethoxysilane in
the presence of AIBN in refluxing CHCl3.10 Under these
reaction conditions, the selectors were fully converted
into the corresponding silane derivatives, which were
separated from the excess of 3-mercaptopropyltrimeth-
oxysilane simply by washing with pentane. The grafting
to silica gel was carried out in toluene at reflux for 24 h
and the derivatized silica gels, after being thoroughly
washed and dried, were employed for packing stainless
steel columns of 15 cm (internal diameter 4.6 mm).
Figure 2. Structure of CSPs 2 and 3.
Due to the different arrangement of the substituent
groups on the cholestanic backbone, these two CSPs
can behave differently as far as both efficiency and ver-
satility are concerned.
Therefore, the synthesis of these CSPs and their use in
the HPLC resolution of racemic compounds will not
only allow us to shed light on the effect of a free hydr-
oxyl group on the enantiodiscrimination properties of

3822
A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
O
O
O
OH
OH
OH
b
OH
OH
N
CH3
a
c
HO
O
HO
OH
HO
O
5
4
6
NO2
H
O
N
O2N
O
O
OH
O
N
N
e
d
CH3
CH3
O
O
O
O
O
O
CH3
CH3
7
8
NO2
NO2
H
O
H
O
N
N
O2N
O2N
O
O
O
O
N
N
CH3
CH3
f
g
O
OH
O
O
NO2
O
CH3
O
CH3
O
N
H
9
10
H
O
O2N
N
O
O
N
CH3
HO
O
O
N
H
11
Scheme 1. Reagents and conditions: (a) NBS, 0.37 M NaHCO3, rt to 85 ꢁC; (b) (1) Bu3N, dioxane, 10 ꢁC, (2) EtOCOCl, (3) N-methylallylamine,
10 ꢁC to rt; (c) acetic anhydride, toluene 80 ꢁC, 5 h; (d) 3,5-dinitrophenylisocyanate, refluxing toluene; (e) NaBH4, THF/MeOH, 0 ꢁC to rt; (f) 2-
naphthylisocyanate, refluxing toluene; (g) MeOH/HCl, reflux, 1 h.
The amount of selector bonded to silica gel was deter-
mined by means of elemental analysis and results were
comparable for both the CSPs.
results obtained using CSPs 2 and 3 in the HPLC sepa-
ration of these racemic compounds compared with those
obtained using CSP 1.
2.2. Use of CSPs 2 and 3 in the chromatographic
resolution of racemic compounds
CSP 2, which possesses a free OH group at the 7-posi-
tion of the cholestanic backbone, resolved, among the
p-acid racemates (runs 1–5), only 4-nitrobenzamides
15a and 15b (runs 4 and 5). In contrast, CSP 3 was able
to resolve the aminoacid derivative 14a (run 1) not enan-
tiodiscriminated by CSP 1 and compounds 14c and 15a
with a values comparable to those obtained upon CSP 1.
These results suggest that the replacement of a 2-naph-
thylcarbamate moiety with a hydroxyl group results in
worsening the enantiodiscriminating capabilities of this
kind of selector towards p-acid racemic compounds, in
The enantiodiscriminating capability of CSPs 2 and 3
was tested using the racemic compounds reported in
Chart 1.
Some of these racemates were resolved by CSP 1, so that
a comparison of the enantioresolution properties of
CSPs 2 and 3 with respect to those of CSP 1 could be
easily made. Table 1 reports on the chromatographic

A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
3823
NO2
NO2
H
H
O
O2N
N
O2N
N
O
O
O
O
O
N
S
Si(OCH3)3
N
a
CH3
CH3
R1O
OR2
R1O
OR2
13a: R1 = H, R2 = 2-naphthylcarbamoyl
13b: R1 = 2-naphthylcarbamoyl, R2 = H
11, 12
NO2
H
O
O2N
N
O
O
O
N
S
Si
O
H3CO
CH3
b
R1O
OR2
CSPs 2-3
Scheme 2. Reagents and conditions: (a) 3-mercaptopropyltrimethoxysilane, AIBN, refluxing CHCl3; (b) silica gel, refluxing toluene.
R
O
Me
O
R
O
NO2
R'
N
H
Ar
N
R'
N
H
H
NO2
NO2
16a: R = iPr, R' = Ph
16b: R = tBu, R' = Ph
16c: R = iPr, R' = Ph
16d: R = Me, R' = CN
16e: R = iPr, R' = CN
16f: R = tBu, R' = Me
14a: R = CH2Ph, R' = COOMe
14b: R = CH3, R' = COOMe
14c: R = Ph, R' = CH2OH
15a: Ar = Ph
15b: Ar = 1-naphthyl
Me
H
N
R'
R
R
OH
OH
O
R
R
O
N
18a: R = H, R' = Ph
OH
17a: R = H
17b: R = OCH2-CH=CH2
18b: R = COPh, R' = Ph
18c: R = COPh, R' = 4-OMePh
18d: R = COPh, R' = 2-naph
18e: R = iBu, R' = Ph
18f: R = iBu, R' = 4-OMePh
18g: R = iBu, R' = iBu
N
Cl
R'
Me
O
N
20a: R = Me, R' = Cl
20b: R = R' = H
20c: R = H, R' = Cl
H
19
Chart 1.
particular when the OH is located at the 7-position. In
fact in this case, 2-naphthylcarbamoyl moiety, responsi-
ble for the interaction with p-acid racemates, is linked to
the less stereochemically demanding11 3-position and
this most likely prevents an efficient enantioselective
interaction with these racemic compounds. In contrast
when the p-basic arylcarbamoyl group is linked at the
7-position, as in CSP 3, p-acid racemic compounds bear-
ing a phenyl group (runs 1, 3 and 4) were resolved, prob-
ably because the enantiodiscriminating capability of this
selector takes advantage of the spatial proximity of the
two different arylcarbamoyl moieties, which could allow
the p-acid racemic compounds having these structural
features to experience two contemporary p–p interac-
tions (phenyl-3,5-dinitrophenylcarbamoyl and 3,5-
dinitrophenyl-2-naphthylcarbamoyl).
Better results were obtained in the chromatographic reso-
lution of p-basic racemic compounds. CSP 2 is able to
resolve the racemates enantiodiscriminated by CSP 1
(runs 6–11) except for 16b (run 7). However, the enantio-
selectivity factors are in general lower than those
obtained with CSP 1 (except for 15b, 18b and 20b),
suggesting that the replacement of the naphthylcarba-

3824
A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
Table 1. Chromatographic resolutiona of racemic compounds on CSPs 1–3
Run
Compound
CSP 2
CSP 3
CSP 1
ac (e.o.)d
1f
Eluente
b
b
b
k0
ac (e.o.)d
Rsj
k0
ac (e.o.)d
Rsj
k0
1
2
3
4
5
6
7
8
14a
14b
14c
15a
15b
16a
16b
17a
17b
18a
18b
20a
20b
20c
3.75
4.85
15.01
9.26
2.99
5.72
4.72
3.57
3.72
2.28
1
1
1
2.22
3.80
13.21
1.76
1.93
2.68
2.85
3.5
3.13
1.58
10.50
2.51
9.19
7.95
1.14
1
1.15
1.11
0.88
4.87
7.96
10.18
3.04
3.62
3.06
1.98
1.36
1.06
3.69
6.59
5.74
7.71
7.35
A
A
A
A
A
B
C
D
D
C
B
1.18
0.78
1.05
1.20g
1.05
1.14
1.02
1i
1.15h
1.06h
1.13
1.13
1
0.54
1.02
1.07f
1
1.06f
1.17
1.12
1.23
1.38
1.51
0.83
1.07
1.15
1.21
1.32
1.09h
1.16i
1.44
1.06
1.18
0.55
0.82
0.92
1.18f
9
1.46
1.19
1.15
1.16 (+)
1.20 (+)
1.19 (+)
1.40f
10
11
12
13
14
1.15
1.04
1.08 (+)g
1.06 (+)g
1.16 (+)f
17.12
3.91
32.71
10.16
A
A
A
1.26 (+)h
1.11 (+)
0.89
1.20
a Chromatographic conditions: UV detection (k 254 nm), flow 1 mL/min, T = 25 ꢁC.
b Retention factor of the first eluted enantiomer.
c Enantioselectivity factor.
d Sign of the circular dichroism (at 254 nm) of the first eluted enantiomer.
e A = hexane–dichloromethane–2-propanol 70:30:3; B = hexane–dichloromethane–2-propanol 90:10:1; C = hexane–dichloromethane–2-propanol
80:20:1; D = hexane–dichloromethane–2-propanol 90:10:2; E = hexane–dichloromethane–2-propanol 70:30:5, F = hexane–dichloromethane–2-
propanol 70:30:7.
f Eluent E.
g Eluent F.
h Eluent C.
i Eluent = hexane–dichloromethane–2-propanol 95:5:1.
j Resolution factor.
moyl moiety at the 7-position with a hydroxyl group
weakens the enantioresolution capability of the selector
towards this class of racemic compounds. This probably
happens because of enlargement of the cavity of the
selector of CSP 2 with respect to that of CSP 1 selec-
tor,12 due to the absence of the hindering 2-naphthyl-
carbamoyl moiety at position 7. It is known that the
racemic compounds interact with similar selectors, by
approaching them from the cleft formed by the concave
part of the cholestanic backbone and the appended
groups,13 therefore its enlargement can make it possible
that the enantiomers of a racemic substrate fit to the
same extent in the cleft and hence their enantiodiscrim-
ination is prevented. A different situation was found in
the case of CSP 3, which was able to resolve all the p-
basic racemic compounds enantiodiscriminated by CSP 1
with higher values of the enantioselectivity factors (runs
6–11), in particular in the case of the two binaphthols
17a and 17b (runs 8 and 9). These results can be
explained by taking into account that the chiral cleft
of CSP 3 selector has an intermediate size between those
of CSP 1 and CSP 2 selectors, since the 3-position lacks
2-naphthylcarbamoyl moiety and the two arylcarba-
moyl groups are closer when located at the 7- and 12-
positions.12 This makes it possible for easier access of
the substrates into the cleft of CSP 3 than into the cavity
of CSP 1 and, due to the spatial proximity of the two
arylcarbamoyl groups, we can suppose that these race-
mates can establish p–p face to face interaction with
3,5-dinitrophenylcarbamoyl moiety together with a p–
p face to edge interaction with 2-naphthylcarbamoyl
group,14 which gives rise to the better enantiodiscrimi-
nating capability of CSP 3. These interactions are not
possible in the case of CSP 1 probably because the nar-
rower cavity prevents a deep access of the substrates.15
The trend observed with the p-basic racemates was also
confirmed in the chromatographic separation of the
benzodiazepines 20 (runs 12–14) that CSP 3 resolved
not only better than CSP 2 but also than CSP 1. In
order to check if the differences in enantiodiscriminat-
ing capability of the three CSPs can be attributed to
a change of the enantiorecognition mechanism, the elu-
tion order of compounds 20 was determined. Knowl-
edge of the elution order allows to establish with
which enantiomer of the substrate the CSP forms the
more stable diastereoisomeric adsorbate. If the elution
order of the same compounds is the same upon differ-
ent CSPs, then the more stable diastereoisomeric adsor-
bate is formed with the same enantiomer of the
racemate and this suggests that the enantiorecognition
mechanism must be similar on different CSPs. CSPs
1–3 showed for compounds 20 the same elution order,
so suggesting that the differences in enantiodiscriminat-
ing capability are not attributable to a change of the
chiral recognition mechanism in passing from one to
another CSP.
The good chromatographic results obtained with CSPs 2
and 3 towards the resolution of p-basic racemic com-
pounds enantiodiscriminated by CSP 1, prompted us
to verify the enantioresolution properties of these CSPs
towards other p-basic racemates, not resolved by CSP 1.
Table 2 reports on the chromatographic results obtained

A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
Table 2. Chromatographic resolutiona of p-basic racemic compounds on CSPs 2 and 3
3825
Run
Compound
CSP 2
CSP 3
b
b
k0
ac
Rsd
k0
ac
Rsd
1
2
3
4
5
6
7
8
9
16c
16d
16e
16f
18c
18d
18e
18f
18g
19
3.55
20.61
5.83
1
1
1
1
1.34
16.53
4.41
1.34
10.26
6.72
0.80
1.75
1.25
2.99
1.19
1.13
1.17
1.14
1.14
1.12
1.38
1.20
1.23
1.17
1.26
0.88
1.23
0.95
1.22
1.16
1.15
0.91
1.03
1.18
2.24
15.24
12.06
2.97
7.14
0.85
1.07
1
1.17e
1.12e
1
0.61
0.86
1.15
10
6.07
1.10
1.09
a Chromatographic conditions: UV detection (k 254 nm), flow 1 mL/min, T = 25 ꢁC, eluent hexane–dichloromethane–2-propanol 80:20:1.
b Retention factor of the first eluted enantiomer.
c Enantioselectivity factor.
d Resolution factor.
e Eluent = hexane–dichloromethane–2-propanol 90:10:1.
using CSPs 2 and 3 in the resolution of various aryl-
amides 16c–f and 19 and N-aryl or N-alkyl amides of
arylpropionic acids. Perusal of Table 2 shows that
CSP 3 was not only more efficient but also more versa-
tile than CSP 2, which separated the enantiomers of only
four racemates (runs 5, 7, 8 and 10), with lower enantio-
selectivity factors than those observed on CSP 3.
Various benzamides were well resolved on CSP 3 (runs
1–4) suggesting that the presence of an aryl group linked
to the stereogenic centre of these compounds is not
mandatory.
selection, is prevented. The medium sized chiral cleft of
CSP 3 allows the enantiomers to enter and to experience
enantioselective interactions with both the arylcarba-
moyl moieties, which give rise to their chromatographic
separation.15
4. Experimental
4.1. General
1H and 13C NMR spectra were recorded on an NMR
Varian Gemini 200 or on a Varian VXR-300 in CDCl3
or in DMSO-d6, using TMS as internal standard. The
following abbreviations were used: s = singlet; d = dou-
blet; dd = double doublet; t = triplet; ddt = double dou-
ble triplet; m = multiplet.
3. Conclusions
Chromatographic data concerning the HPLC resolution
of both p-acid and p-basic compounds on CSPs 2 and 3
compared to those related on the use of CSP 1 allows us
to reach some conclusions about the effect of the pres-
ence of a free OH group on the enantiodiscrimination
capability of cholic acid based mixed selectors. The
effect of the free hydroxyl group depends on its position
on the cholestanic backbone. The presence of a hydroxyl
group at the 7-position, as in CSP 2, does not signifi-
cantly improve the enantiodiscrimination ability of this
kind of selector either towards p-acid or p-basic racemic
compounds. On the contrary, when the OH group is at
the 3-position of cholic acid system, as in CSP 3, a more
efficient and versatile chiral selector is obtained, in par-
ticular towards p-basic racemic compounds. In fact CSP
3 is able not only to resolve the p-basic racemates enan-
tiodiscriminated by CSP 1 with higher a values, but also
to separate the enantiomers of p-basic compounds not
resolved by CSP 1. On the basis of these results we
can suppose that the different enantiodiscriminating
capability is due to different sizes of the chiral cleft
formed by the concave part of the cholic acid with the
appended arylcarbamoyl groups. When this cleft is too
large, as in CSP 2, both enantiomers of the racemic com-
pounds interact to the same extent. In contrast, when
the cleft is too narrow, as in CSP 1, a deep access of
the enantiomers into the cavity, and hence their enantio-
TLC analyses were performed on silica gel plates
Macherey-Nagel 60 F254. Chromatographic purifica-
tions were performed using silica gel Macherey-Nagel,
70–230 or 230–400 for flash chromatography. Optical
rotations were measured at the sodium D-line on a dig-
ital polarimeter JASCO-DIP 360: concentration, solvent
and temperature are those specified for every measure.
Melting points were taken using a Kofler Reichert-Jung
apparatus and are uncorrected. IR spectra were
recorded on a Perkin–Elmer 1710 spectrophotometer.
HPLC analyses were performed on a Jasco PU-980
chromatograph, equipped with a JASCO UV-975 detec-
tor and a circular dichroism JASCO J-710 detector. Ele-
mental analyses were carried out at Dipartimento di
`
Scienze e Tecnologie Chimiche dellÕUnivesita degli Studi
di Udine.
Toluene, THF and dioxane were refluxed over sodium-
benzophenone and distilled before use. N-Methylallyl-
amine, pyridine and tributylamine were distilled over
CaH2. Chloroform was distilled before use. 2-Naph-
thoylchloride was distilled under reduced pressure and
3,5-dinitrobenzoylchloride was recrystallized from
petroleum ether. Ethyl chloroformate was distilled

3826
A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
under nitrogen atmosphere just before use. N-Bromo-
succinimide was recrystallized from water. 3-Mercapto-
propyltrimethoxysilane was distilled under reduced
pressure before use. Unless otherwise specified, the
reagents were used without any purification. Compounds
6 and 12 were synthesized as previously reported and
matched the reported characteristics.5
4.4. N-Allyl-N0-methyl-3-acetyl-7-hydroxy-12-(3,5-di-
nitrophenyl)carbamoyloxycholan-24-amide, 9
Compound 8 (1.6 g, 2.2 mmol) was dissolved in a mix-
ture of dry THF (4 mL) and methanol (20 mL) cooled
to 0 ꢁC, and NaBH4 (0.14 g, 3.7 mmol) added slowly.
When gas evolution ceased, the mixture was allowed
to warm to room temperature and stirred overnight.
The solvent was removed at reduced pressure, and the
remaining solid dissolved in ethyl acetate. The solution
was washed with 10% KHCO3 (2 · 30 mL) and brine
(2 · 30 mL), then dried over anhydrous Na2SO4. After
removing the solvent under reduced pressure, the crude
product was purified by flash chromatography (SiO2,
4.2. N-Allyl-N0-methyl-3-acetyl-7-oxo-12-hydroxy-
cholan-24-amide, 7
Acetic anhydride (2.05 mL, 21.7 mmol) in dry toluene
(3.15 mL) was added to a solution of amide 6 (3 g,
6.5 mmol) in dry toluene (3.15 mL) and the resulting
mixture was stirred for 5 h at 80 ꢁC, then cooled to
room temperature and diluted in CH2Cl2. The solvent
was removed under reduced pressure and the crude
product purified by flash chromatography (SiO2,
CH2Cl2–acetone = 80:20) affording 9 (1.3 g, 1.8 mmol,
22
81%): mp = 135–142 ꢁC; ½aꢀD ¼ þ56:4 (c 0.99, CH2Cl2);
1H NMR (200 MHz, CDCl3, d): 0.76 (s, 3H, CH3); 0.89
(s, 3H, CH3); 0.96–2.40 (m, 28H, CH and CH2 steroidal
and OH 7); 2.16 (s, 3H, OCOCH3); 2.88 and 2.92 (s,
3H, N–CH3); 3.80–4.01 (m, 3H, CH2–CH@CH2 and
CH 7); 4.51 (m, 1H, CH 3); 5.05–5.20 (m, 3H,
CH2@CH and CH 12); 5.67 (ddt, 1H, CH@CH2); 8.66
(t, 1H, aromatic); 8.76 (d, 2H, aromatics); 9.08 (d,
1H, NH carbamate); 13C NMR (50 MHz, CDCl3, d):
12.1, 17.8, 21.3, 22.4, 22.9, 25.6, 26.6, 27.4, 27.6, 30.8,
34.6, 35.1, 38.4, 39.0, 41.0, 43.4, 45.2, 47.7, 50.1, 52.2,
68.0 (C 7), 74.3 (C 12), 78.5 (C 3), 112.2, 116.7
(@CH2), 118.0, 132.7 (ACH@), 141.4, 148.8, 152.8
(C@O carbamate), 170.6 (C@O acetyl). IR (KBr,
cmꢁ1): 3854, 3821, 3751, 3736, 3712, 3689, 3676, 3650,
3447, 3115, 2941, 2872, 2346, 1734, 1624, 1548, 1466,
1424, 1344, 1246, 1224, 1193, 1075, 1025, 1009, 897,
806, 769, 732, 656, 611, 400.
CH2Cl2–acetone = 75:25) affording 7 (1.7 g, 3.4 mmol,
20
52%): mp = 74–78 ꢁC; ½aꢀD ¼ þ17:3 (c 0.99, CH2Cl2);
1H NMR (200 MHz, CDCl3, d): 0.67 (s, 3H, CH3);
0.96 (d, 3H, CH3 21); 1.17 (s, 3H, CH3); 0.90–2.50
(m, 28H, CH and CH2 steroidal, OH 12); 2.16 (s, 3H,
OCOCH3); 2.89 and 2.93 (s, 3H, N–CH3); 3.87–3.98
(m, 3H, CH2ACH@CH2 and CH 12); 4.66 (m, 1H,
CH 3); 5.06–5.28 (m, 2H, CH2@CH); 5.75 (ddt, 1H,
CH@CH2); 13C NMR (50 MHz, CDCl3, d): 12.8,
17.6, 21.2, 22.7, 24.1, 25.8, 27.5, 29.1, 33.0, 33.6, 34.6,
34.9, 35.9, 40.7, 45.1, 45.7, 46.5, 49.4, 72.0 (C 12),
72.8 (C 3), 170.5 (C@O acetyl), 211.0 (C@O carbony-
lic). IR (KBr, cmꢁ1): 3813, 3440, 2948, 1732, 1709,
1629, 1468, 1392, 1364, 1244, 1080, 1049, 924, 802,
597, 400.
4.3. N-Allyl-N0-methyl-3-acetyl-7-oxo-12-(3,5-dinitro-
phenyl)carbamoyloxycholan-24-amide, 8
4.5. N-Allyl-N0-methyl-3-acetyl-7-(2-naphthyl)carba-
moyloxy-12-(3,5-dinitrophenyl)carbamoyloxycholan-24-
amide, 10
3,5-Dinitrophenylisocyanate (1.21 g, 5.8 mmol) was
added to a solution of 7 (1.5 g, 3.0 mmol) in dry toluene
(45 mL) and the resulting mixture heated at reflux and
stirred overnight, then allowed to cool to room temper-
ature. The solvent was removed under reduced pressure
and the crude product purified by flash chromatography
2-Naphthylisocyanate (0.65 g, 3.6 mmol) and DMAP
(0.14 g, 1.1 mmol) were added to a solution of 9 (1.2 g,
1.7 mmol) in dry toluene (40 mL) and the resulting mix-
ture heated at reflux and stirred for 24 h, then filtered to
remove 2-naphthylurea; the solvent was evaporated
under reduced pressure and the obtained solid purified
by flash chromatography (SiO2, CH2Cl2–ethyl acetate =
(SiO2, CH2Cl2–acetone = 92:8) affording
8
(1.7 g,
21
2.4 mmol, 80%): mp = 138–145 ꢁC; ½aꢀD ¼ þ63:1 (c 1,
1
CH2Cl2); H NMR (200 MHz, CDCl3, d): 0.67 (s, 3H,
85:15) to afford 10 (1.1 g, 1.2 mmol, 74%): mp = 144–
23
CH3); 0.96 (d, 3H, CH3 21); 1.17 (s, 3H, CH3); 0.90–
2.50 (m, 27H, CH and CH2 steroidal); 2.16 (s, 3H,
OCOCH3); 2.89 and 2.96 (s, 3H, N–CH3); 3.85–4.01
(m, 2H, CH2–CH@CH2); 4.63 (m, 1H, CH 3); 5.04–
5.28 (m, 3H, CH2@CH and CH 12); 5.71 (ddt, 1H,
CH@CH2); 8.65 (t, 1H, aromatic); 8.76 (d, 2H, aromat-
ics); 9.06 (d, 1H, NH carbamate); 13C NMR (50 MHz,
CDCl3, d): 12.4, 17.8, 21.2, 22.7, 23.9, 25.9, 26.4, 27.6,
30.2, 30.9, 31.3, 33.1, 33.5, 33.8, 34.6, 35.1, 36.8, 38.4,
42.0, 45.1, 45.2, 45.7, 46.6, 46.9, 48.9, 50.3, 52.3, 72.9
(C 3), 112.1, 116.9 (@CH2), 117.2, 117.9, 132.4
(ACH@), 141.5, 146.5, 148.7, 152.9 (C@O carbamate),
170.4 (C@O acetyl), 173.5, 174.0 (C@O amide), 211.0
(C@O carbonylic). IR (KBr, cmꢁ1): 3854, 3838, 3751,
3244, 3117, 2956, 2364, 2345, 1734, 1710, 1654, 1617,
1546, 1466, 1424, 1345, 1245, 1222, 1066, 1048, 1022,
919, 897, 806, 765, 732, 669, 656, 400.
148 ꢁC; ½aꢀD ¼ þ87:1 (c 0.97, CH2Cl2); 1H NMR
(200 MHz, CDCl3, d): 0.76 (s, 3H, CH3); 0.86 (d,
3H, CH3 21), 0.93 (s, 3H, CH3); 0.96–2.40 (m, 27H,
CH and CH2 steroidal); 2.03 (s, 3H, OCOCH3);
2.83 and 2.87 (s, 3H, N–CH3); 3.73–3.95 (m, 2H,
CH2–CH@CH2); 4.52 (m, 1H, CH 3); 4.90–5.14 (m,
4H, CH2@CH and CH 12 and CH 7); 5.63 (ddt, 1H,
CH@CH2); 7.25–7.80 (m, 9H, H naphthylics and NH
carbamates), 8.01–8.70 (m, 3H, aromatics); 13C NMR
(50 MHz, CDCl3, d): 12.1, 14.1, 17.6, 20.9, 21.1, 22.4,
22.8, 25.7, 26.8, 27.3, 29.1, 30.1, 30.6, 31.3, 33.5, 34.3,
34.9, 35.0, 37.8, 40.6, 43.8, 45.4, 47.4, 50.1, 52.2, 72.0
(C 7), 73.9 (C 12), 78.4 (C 3), 112.0, 114.7, 116.9
(@CH2), 117.6, 117.9, 119.2, 124.6, 126.3, 127.2, 127.4,
128.7, 130.0, 132.1, 133.8 (ACH@), 135.4, 141.3, 141.4,
148.6, 152.9 (C@O carbamate), 153.1 (C@O carbamate),
167.7, 170.3 (C@O acetyl), 173.7, 174.0 (C@O amide).

A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
3827
IR (KBr, cmꢁ1): 3853, 3838, 3750, 3735, 3690, 3676,
3649, 3385, 3108, 2941, 2871, 2361, 2343, 1733, 1634,
1605, 1547, 1506, 1472, 1432, 1344, 1245, 1224, 1127,
1071, 956, 920, 893, 854, 807, 731, 668, 655, 473, 400.
36H, CH and CH2 steroidal, CH2 chain and OH 3);
2.81 and 2.84 (s, 3H, N–CH3); 3.45 (m, 1H, CH 3);
3.54 (s, 9H, Si(OCH3)3); 4.99–5.14 (m, 2H, CH 7 and
CH 12); 7.25–7.80 (m, 9H, H naphthylics and NH car-
bamates), 8.09–8.75 (m, 3H, aromatics); 13C NMR
(50 MHz, CDCl3, d): 8.7, 9.2, 12.6, 18.2, 18.8, 19.0,
23.1, 23.5, 26.0, 27.5, 27.8, 28.7, 29.2, 31.3, 32.3, 33.5,
34.8, 35.2, 38.2, 39.0, 39.4, 39.8, 40.2, 40.5, 40.7, 40.9,
41.1, 41.5, 43.6, 46.0, 47.0, 47.9, 49.5, 50.6, 50.8, 58.4,
58.7, 66.1, 72.0 (C 7), 74.2 (C 12), 78.4 (C 3), 81.8,
112.2, 114.7, 118.1, 120.3, 125.1, 127.1, 127.9, 128.2,
129.1, 129.9, 130.1, 134.4, 137.8, 142.4, 149.3, 154.2
(C@O carbamate), 173.1 (C@O amide). IR (KBr,
cmꢁ1): 3750, 3381, 2938, 2363, 2344, 1729, 1634, 1606,
1547, 1506, 1466, 1432, 1344, 1244, 1222, 1079, 1001,
955, 897, 808, 768, 731, 655, 474, 400.
4.6. N-Allyl-N0-methyl-3-hydroxy-7-(2-naphthyl)carba-
moyloxy-12-(3,5-dinitrophenyl)carbamoyloxycholan-24-
amide, 11
Compound 10 (1.06 g, 1.2 mmol) was dissolved in a mix-
ture of MeOH (30 mL) and HCl (0.80 mL) and the
resulting mixture heated at reflux and stirred for 1 h.
After cooling to room temperature, the mixture was
poured into water and extracted with CH2Cl2. The com-
bined organic extracts were washed with 10% HCl
(2 · 20 mL), 10% NaHCO3 (2 · 20 mL) and water then
dried over anhydrous Na2SO4. The solvent was evapo-
rated under reduced pressure and the crude product
purified by chromatography (SiO2, CH2Cl2–ethyl
4.7.2. N-Methyl-N0-[(trimethoxysylylpropylthio)propyl]-
3-(2-naphthyl)carbamoyloxy-7-hydroxy-12-(3,5-dinitro-
phenyl)carbamoyloxycholan-24-amide, 13b. The mix-
acetate = 70:30) affording 11 (0.85 g, 1.0 mmol, 84%):
18
mp = 159–163 ꢁC; ½aꢀD ¼ þ93:4 (c 0.98, CH2Cl2);
ture was heated at reflux for 30 h and 1.01 g of
1H NMR (200 MHz, CDCl3, d): 0.70 (s, 3H, CH3);
0.83 (d, 3H, CH3 21), 0.87 (s, 3H, CH3); 0.96–2.40 (m,
28H, CH and CH2 steroidal and OH 3); 2.81 and 2.84
(s, 3H, N–CH3); 3.48 (m, 1H, CH 3); 3.75–3.95 (m,
2H, CH2–CH@CH2); 4.94–5.10 (m, 4H, CH2@CH,
CH 7 and CH 12); 5.63 (ddt, 1H, CH@CH2); 7.25–
7.93 (m, 9H, H naphthylics and NH carbamates),
8.40–8.70 (m, 3H, aromatics); 13C NMR (50 MHz,
CDCl3, d): 12.1, 14.1, 17.6, 20.9, 21.1, 22.4, 22.8, 25.7,
26.8, 27.3, 29.1, 30.1, 30.6, 31.3, 33.5, 34.9, 35.0, 37.8,
40.6, 43.8, 45.4, 47.4, 50.1, 52.2, 71.8 (C 7), 72.2 (C
12), 77.4 (C 3), 112.0, 114.7, 116.2 (@CH2), 117.6,
117.9, 119.2, 124.6, 126.5, 127.2, 128.7, 130.0, 132.1,
133.8 (ACH@), 135.4, 141.3, 141.4, 148.6, 152.8 (C@O
carbamate), 153.3 (C@O carbamate), 167.7, 173.7,
174.0 (C@O amide). IR (KBr, cmꢁ1): 3854, 3838,
3751, 3690, 3676, 3643, 3629, 3384, 2930, 2362, 2344,
1734, 1607, 1547, 1507, 1432, 1344, 1245, 1224, 1119,
1073, 1000, 956, 898, 855, 808, 732, 669, 655, 610, 474,
400.
22
product 13b was obtained: mp = 89–95 ꢁC, ½aꢀD
¼
1
þ56:9 (c 1.02, CH2Cl2); H NMR (200 MHz, CDCl3,
d): 0.70 (d, 2H, CH2), 0.75 (s, 3H, CH3); 0.90 (s, 3H,
CH3); 0.95 (d, 3H, CH3 21); 0.92–2.45 (m, 35H, CH
and CH2 steroidal, and chain CH2); 2.95 (s, 3H, N–
CH3); 3.55 (s, 9H, Si(OCH3)3); 4.60 (m, 1H, CH 3)
5.05–5.25 (m, 2H, CH 7, CH 12); 6.91 (s, 1H, NH carba-
mate); 7.27–7.50 (m, 3H, H naphthylics); 7.65–7.80 (m,
3H, H naphthylics); 7.90 (s, 1H, H naphthylic); 8.60 (t,
1H, H p-3,5-dinitrophenylic); 8.65 (d, 2H, H o-3,5-dini-
trophenylics); 8.80 (s, 1H, NH carbamate); IR (KBr,
cmꢁ1): 3444, 2944, 2855, 2366, 2344, 1733, 1716, 1694,
1633, 1600, 1544, 1505, 1466, 1433, 1344, 1505, 1466,
1433, 1344, 1244, 1222, 1188, 1077, 994, 900, 822, 805,
766, 733, 477.
4.8. General procedure for the preparation of CSPs
A solution of 3-mercaptopropyltrimethoxysylanic deriv-
ative in dry toluene (20 mL) was added dropwise to a
suspension of spherical silica gel (2.5 g) [previously dried
under reduced pressure (p = 0.01 mmHg) at 180 ꢁC for
15 h] in dry toluene (15 mL). The resulting mixture
was heated at reflux, with gentle stirring, for 24 h. After
cooling to room temperature, the silica was filtered and
washed sequentially with toluene (3 · 30 mL), methanol
(3 · 30 mL), acetone (3 · 30 mL), CH2Cl2 (3 · 30 mL)
and pentane (2 · 30 mL), in that order, then dried under
reduced pressure (p = 0.01 mmHg) at 45 ꢁC for 8 h. The
amount of selector linked to silica gel was then deter-
mined by elemental analysis.
4.7. Preparation of silane derivatives 13: representative
procedure
3-Mercaptopropyltrimethoxysilane (0.8 mL, 4.2 mmol)
was added to a solution of the selector (0.8 mmol) in
CHCl3 (10 mL) and the resulting mixture heated at
reflux until TLC analysis showed complete conversion
of the substrate. After cooling, the solvent was removed
at reduced pressure, and the remaining oil was dispersed
in pentane (30 mL); the precipitate was filtered and
washed again with pentane (5 · 30 mL) to afford the
pure product.
CSP 2: C, 9.26; H, 1.50; N, 0.98% corresponding to
0.233 mmol/g.
4.7.1. N-Methyl-N0-[(trimethoxysylylpropylthio)propyl]-
3-hydroxy-7-(2-naphthyl)carbamoyloxy-12-(3,5-dinitro-
phenyl)carbamoyloxycholan-24-amide, 13a. The mix-
ture was heated at reflux for 48 h and 0.93 g of
CSP 3: C, 8.95; H, 1.56; N, 0.87% corresponding to
0.207 mmol/g.
product
13a
was
obtained:
mp = 97–102 ꢁC;
16
½aꢀD ¼ þ82:6 (c 1.01, CH2Cl2); 1H NMR (200 MHz,
CDCl3, d): 0.70 (d, 2H, CH2), 0.79 (s, 3H, CH3); 0.91
(d, 3H, CH3 21), 0.94 (s, 3H, CH3); 0.96–2.40 (m,
Two 15 cm stainless steel columns were slurry packed
with each of these materials, using conventional high
pressure packing techniques.

3828
A. Iuliano, A. Ruffini / Tetrahedron: Asymmetry 16 (2005) 3820–3828
6. Fieser, L. F.; Rajagopalan, S. J. Am. Chem. Soc. 1949, 71,
Acknowledgements
3935.
7. DÕSouza, L.; Maitra, U. J. Org. Chem. 1996, 61, 9494–
9502.
´
We are indebted to Dr. G. Felix for the packing of the
columns. This work was supported by MIUR (PRIN
2003) and University of Pisa.
8. Bellini, A. M.; Quaglio, M. P.; Guarneri, M.; Cavazzini,
G. Eur. J. Med. Chem.-Chim. Ther. 1983, 18, 185.
9. Grand, R.; Reichstein, T. Helv. Chim. Acta 1945, 28, 344.
10. Rosini, C.; Bertucci, C.; Pini, D.; Altemura, P.; Salvadori,
P. Tetrahedron Lett. 1985, 26, 3361–3364.
References
11. Iuliano, A.; Scafato, P. Tetrahedron: Asymmetry 2003, 14,
611.
12. Alagona, G.; Ghio, C.; Iuliano, A.; Monti, S.; Pieraccini,
I.; Salvadori, P. J. Org. Chem. 2003, 68, 3145.
13. Vaton-Chanvrier, L.; Oulyadi, H.; Coquerel, G.; Combret,
Y.; Combret, J. C. Chirality 2001, 13, 668.
1. Pirkle, W. H.; Pochapsky, T. C. Chem. Rev. 1989, 89, 347–
362; Gasparrini, F.; Misiti, D.; Villani, C. J. Chromatogr.
A 2001, 906, 35–50.
2. (a) Welch, C. J. J. Chromatogr. A 1994, 666, 3; (b) Iuliano,
A.; Attolino, E.; Salvadori, P. Eur. J. Org. Chem. 2001,
3523–3529; (c) Iuliano, A.; Attolino, E.; Salvadori, P.
Tetrahedron: Asymmetry 2002, 13, 1805–1815; (d) Iuliano,
A.; Lecci, C. Tetrahedron: Asymmetry 2003, 14, 1345.
14. Pirkle, W. H.; Welch, C. J. Tetrahedron: Asymmetry 1994,
5, 777–780.
15. This is only an interpretation of the chromatographic
data, based on the knowledge of the conformations of
these kinds of selectors12 and on a known mode of
interaction,13 and not a hypothesis of an enantiorecogni-
tion mechanism, which deserves further studies to be
elucidated: work is currently in progress on this topic.
´
3. Iuliano, A.; Felix, G. J. Chromatogr. A 2004, 1031, 187.
´
4. Iuliano, A.; Salvadori, P.; Felix, G. Tetrahedron: Asym-
metry 1999, 10, 3353–3364.
´
5. Iuliano, A.; Pieraccini, I.; Felix, G.; Salvadori, P. Tetra-
hedron: Asymmetry 2002, 13, 1265–1275.