Photochemically immobilized 4-methylbenzoyl cellulose as a powerful chiral stationary phase for enantioselective chromatography

Article abstract of DOI:10.3390/molecules21121740

A process to immobilize para-methylbenzoyl cellulose (PMBC) on silica gel has been developed and applied to prepare chiral stationary phases (CSPs) for enantioselective chromatography. The immobilization was achieved by simple irradiation of the polysaccharide derivative with ultraviolet light after coating on a silica gel support. The influence of parameters such as irradiation time and solvent on immobilization effectiveness were investigated. The performance of the prepared immobilized phases were evaluated by injection of a series of racemic compounds onto the packed columns and determination of their chiral recognition ability. By contrast to the classical coated phase, the immobilized CSP can be used under various chromatographic conditions without limitation of organic solvent types as the mobile phase. This extended applicability permits to improve selectivity and to resolve chiral compounds which are not or only poorly soluble in the mobile phases which are compatible with the non-immobilized PMBC stationary phase.

Full text of DOI:10.3390/molecules21121740

molecules
Article
Photochemically Immobilized 4-Methylbenzoyl
Cellulose as a Powerful Chiral Stationary Phase for
Enantioselective Chromatography
Eric Francotte *, Dan Huynh and Tong Zhang †
Novartis Pharma AG, Global Discovery Chemistry, 4002-Basel, Switzerland; dan.huynh@novartis.com
* Correspondence: eric.francotte@francotteconsulting.ch; Tel.: +41-78-405-9924
† Present address: Chiral Technologies Europe, Parc d’Innovation, Bd. Gonthierd’Andernach, B.P. 80140,
67404 Illkirch, France; tzhang@chiral.fr.
Academic Editor: Yoshio Okamoto
Received: 7 November 2016; Accepted: 9 December 2016; Published: 17 December 2016
Abstract: A process to immobilize para-methylbenzoyl cellulose (PMBC) on silica gel has
been developed and applied to prepare chiral stationary phases (CSPs) for enantioselective
chromatography. The immobilization was achieved by simple irradiation of the polysaccharide
derivative with ultraviolet light after coating on a silica gel support. The influence of parameters such
as irradiation time and solvent on immobilization effectiveness were investigated. The performance
of the prepared immobilized phases were evaluated by injection of a series of racemic compounds
onto the packed columns and determination of their chiral recognition ability. By contrast to the
classical coated phase, the immobilized CSP can be used under various chromatographic conditions
without limitation of organic solvent types as the mobile phase. This extended applicability permits
to improve selectivity and to resolve chiral compounds which are not or only poorly soluble in the
mobile phases which are compatible with the non-immobilized PMBC stationary phase.
Keywords: enantioselective chromatography; chiral stationary phase; immobilized methylbenzoyl
cellulose; photochemistry
1. Introduction
Polysaccharide-based chiral stationary phases (CSPs), especially those derived from cellulose
and amylose derivatives, have proved to be versatile and effective means for enantioseparations by
high performance liquid chromatography (HPLC) and supercritical fluid chromatography (SFC) [1,2].
They cover a large scope of applications on either analytic or preparative scale. Originally these
CSPs were prepared as a coating by physical adsorption of the polysaccharide derivatives on a
silica gel support [3–6] or as a pure polymer in the form of beads [7,8]. Over the last decade,
a second generation of polysaccharide-based CSPs has appeared, consisting of the immobilized
versions of the previously coated phases. This second generation of phase shows the unique
advantage of being more stable and tolerating almost all kind of mobile phases by contrast to the
‘classical’ non-immobilized polysaccharide-based CSPs which readily dissolve in organic solvents,
such as methylene chloride, chloroform, tetrahydrofuran, etc. This feature of the non-immobilized
polysaccharide CSPs was a considerable limitation for analytes which exhibit low solubility, in
particular for preparative applications.
Several techniques have been developed to immobilize polysaccharide derivatives on silica gel
and these different techniques have been reviewed in recent papers [9–11]. Okamoto and his group
was particularly active in this field and carried out already in 1987 the immobilization of cellulose
tris(3,5-dimethylphenylcarbamate) and tris(3,5-dichlorophenylcarbamate) on 3-aminopropyl silica
Molecules 2016, 21, 1740; doi:10.3390/molecules21121740
www.mdpi.com/journal/molecules

Molecules 2016, 21, 1740
2 of 12
gel using a di-isocyanate moiety as the linking component [12]. Other groups developed processes
to immobilize polysaccharide derivatives on silica gel. For example, Oliveros at al. performed
the immobilization of polysaccharide derivatives through a radically initiated polymerization of a
10-undecenoyl group which was incorporated on the polysaccharide backbone [13]. We also developed
a technique of immobilization consisting in a photochemical or thermal treatment of the polysaccharide
derivatives after coating on silica gel [14–17]. In this report we show that this approach is very
simple and does not need the relatively tedious operation of selective protection and deprotection
chemistry used in most of the other immobilization approaches, as no attaching or crosslinking group
is introduced onto the polysaccharide chain. The simplicity of the method as well as the influence
of various parameters on the immobilization effectiveness is exemplified with the application to
para-methylbenzoyl cellulose (PMBC).
2. Results and Discussion
2.1. Preparation of the Immobilized PMBC-Im Stationary Phases (CSP)
Immobilization of PMBC after coating on silica gel was achieved by irradiation of a suspension of
the coated material in an inert solvent for a few hours, using an immerged mercury lamp surrounded
by a Pyrex cooling jacket (Figure 1). Acetonitrile, or mixtures of methanol-water were used as the
photochemical suspension solvent. After photochemical treatment, each sample was extracted with
a mixture of chloroform, dichloromethane and tetrahydrofuran in order to remove the non-bonded
parts of PMBC (see experimental part in Section 3.3 for more details).
Figure 1. Schematic photochemical setup for immobilization of PMBC on silica gel.
Table 1 summarizes the preparation conditions and the characterization of the prepared
immobilized PMBC-Im CSPs. The table indicates the CSP name, the amount of coated PMBC,
the solvent mixture for irradiation, the irradiation time, the amount of immobilized material after
irradiation and extraction treatment, and the immobilization rate.
Table 1. Preparation conditions and characterization of the immobilized PMBC-Im CSPs.
Amount of
Coated
PMBC (Weight %)
Amount of
Immobilized
PMBC (Weight %)
Irradiation Medium
(Volume Ratio)
Irradiation
Time (h)
Immobilization
Rate (%)
CSP Name
PMBC-S
28
28
28
28
28
28
28
25
na
na
15
18
23
24
30
20
22
na
na
PMBC-Im A
PMBC-Im B
PMBC-Im C
PMBC-Im D
PMBC-Im E
PMBC-Im F
PMBC-Im G
25% MeOH in water
25% MeOH in water
25% MeOH in water
25% MeOH in water
25% MeOH in water
50% MeOH in water
100% acetonitrile
23.8
23.8
24.3
24.3
24.3
22
71.2
85.1
86.7
86.9
86.6
78.6
83.3
21.7
na: not applicable.
In addition, a column packed with the coated material with no further treatment
(no immobilization) has also been prepared as a reference (PMBC-S). This column has obviously
not been used with mobile phases likely to dissolve the coated polysaccharide derivative PMBC.

Molecules 2016, 21, 1740
3 of 12
Immobilization rate refers to the final amount of the immobilized polymer compared to the initial
amount of coated PMBC on silica gel. From the results in Table 1 it can be noticed that while an
irradiation time of 15 h in 25% methanol in water affords about 70% immobilization (PMBC-Im A),
increasing the irradiation above 20 h does not further increase notably the immobilization degree which
remains almost unchanged around 87% in the same suspension medium (PMBC-Im B-E). Irradiation in
50% methanol (PMBC-Im F) or pure acetonitrile (PMBC-Im G) give slightly lower immobilization rates.
Interestingly, the immobilization occurs without any addition of radical initiator or photosensitizer.
Physical and chemical methods have been applied to get some insight into the immobilization
mechanism. However, the insolubility of the material considerably limits the choice of physical
means to investigate the structure of the obtained stationary phase at the molecular level. Solid state
NMR, IR, and Raman spectroscopy were applied but no significant differences were found between
the material obtained before and after immobilization. UV-irradiation of para-methylbenzoyl cellobiose
was used as a model for PMBC and was carried out in solution but no trace of oligomers could
be detected, suggesting that the immobilization process implicates a polymer effect. A very low
percentage of crosslinking is actually sufficient to lead to a tridimensional insoluble network. It is not
yet clear whether the polysaccharide derivative is immobilized by cross-linking between the polymer
chains or by bonding to the silica gel carrier. It is known that cellulose and cellulose derivatives can
degrade under UV treatment, producing radicals which might further react intra- or intermolecularly,
but the exact degradation pathway is usually not easy to evidence [18–20]. In these reports, it was also
shown that the formation of radicals does not necessarily need the presence of a photoinitiator.
The chiral recognition ability as well as the retention properties of the CSPs listed in Table 1 have
been assessed using a series of racemic compounds. Figure 2 shows the structures of the selected
racemic substances.
Figure 2. Structures of the tested racemic compounds.
2.2. Influence of Irradiation Time on the Chromatographic Properties of the PMBC-Im CSPs
Selectivity and retention factors obtained for selected racemic solutes on the PMBC-Im CSPs
prepared by irradiation in a mixture of methanol–water (25/75 in volume) after 18 h (PMBC-Im B), 24 h
(PMBC-Im D), and 30 h (PMBC-Im E) are shown in Table 2. Results obtained on the non-immobilized
phase PMBC-S are also shown as a reference. The classical mobile phase hexane/2-propanol (9/1, v/v)
was used for all chromatographies.

Molecules 2016, 21, 1740
4 of 12
Table 2. Chromatographic results of the enantioselective separations of selected racemic compounds on photochemically immobilized PMBC-Im CSPs: influence of
irradiation time in the mixture of methanol–water 25/75 (v/v). Mobile phase: hexane/2-propanol (9/1, v/v).
PMBC-S
PMBC-Im B
PMBC-Im D
PMBC-Im E
Compound
α
k1
k2
Rs
α
k1
k2
Rs
α
k1
k2
Rs
α
k1
k2
Rs
1
2
4
5
7
9.15
1.47
1.00
1.30
1.22
2.53
1.81
1.39
1.41
1.64
1.85
5.31
14.06 14.06
9.98
7.63
3.99
7.85
1.72
2.93
3.91
16.89
7.82
3.71
1.21
0
0.65
nd
2.25
1.60
0.59
0.58
1.61
7.06
1.31
1.09
1.25
1.20
2.13
1.72
1.40
1.38
1.61
1.74
4.16
9.28
12.31
5.45
10.11
4.55
1.27
nd
0.89
0.87
2.41
2.31
1.02
1.12
3.81
6.33
1.28
1.00
1.24
1.19
2.13
1.70
1.39
1.37
1.59
1.94
5.03
10.44 10.44
11.4
7.66
4.89
7.78
1.87
3.02
3.72
12.28
6.45
4.67
1.15
0
0.75
0.62
2.44
1.82
0.80
0.85
2.25
6.89
1.30
1.12
1.22
1.16
2.16
1.70
1.36
1.33
1.57
1.98
4.86
9.38
13.62
6.30
10.50
4.37
1.04
0.86
0.66
0.54
2.37
1.82
0.86
0.85
2.10
13.01
9.35
10.11
14.23
2.39
10.12 12.63
14.13
9.08
10.41
13.2
2.60
4.14
5.91
11.29 13.80
6.85
4.49
7.15
1.74
2.77
3.36
8.23
9.56
12.31
2.42
3.81
5.39
7.72
4.77
8.03
1.98
3.15
3.69
8.95
10.29
13.67
2.69
4.20
5.80
9
13
16
17
19
4.14
6.41
nd: not determined.

Molecules 2016, 21, 1740
5 of 12
Although there are some variations between the three immobilized CSPs in terms of selectivity and
retention, most differences are relatively small. Selectivity values of the phases obtained after 18 h and
30 h irradiation respectively differ only in the range of 1%–4%. The difference for the retention factor
values is more significant when comparing the same two CSPs, 4% to 13% depending on the compound,
and might reflect the slightly higher amount of immobilized PMBC on the PMBC-Im E phase.
Compared to the non-immobilized phase PMBC-S, the negative impact of the immobilization
on selectivity values is somewhat more marked especially for compounds 1, 2, and 9 but it remains
minor for the other compounds. The small drop in selectivity indicates that the immobilization process
partially disturbs the original structure of the cellulose derivative or restricts the access to the chiral
interaction sites. However, for most substances the disturbance remains minor. This is probably
due to the fact that, unlike other immobilization approaches, the designed process does not involve
the introduction of specific functional immobilization groups onto the cellulose derivative, which
alter the regularity of the molecular and supramolecular structure of the polysaccharide chains. The
extent of selectivity decrease is not identical for each compound. It varies from 3% to 40% showing
that each individual substance responses to the structural change of the polysaccharide differently.
This observation is in accordance with previous investigations on cellulose triacetate which support
the concept of the presence of multiple interaction sites in polysaccharide-based CSPs [21]. For the
substances which exhibit a decrease of selectivity after immobilization using the conventional mobile
phase hexane-isopropanol, the effect can in most instances be largely compensate by the utilization of
mobile phases which are excluded with the non-immobilized stationary phase due to its solubility in
these mobile phases (see Section 2.4). Interestingly, for the racemic substance
on the immobilized phase.
4 the selectivity is better
2.3. Influence of Irradiation Solvent on the Chromatographic Properties of the PMBC-Im CSPs
It is known that photochemical reactions are dependent on solvent conditions and this effect
has recently been reviewed by Klán et al. [22]. In order to investigate this parameter, we examined
the effect of the suspension solvent on both immobilization rate and chiral recognition ability of the
resulting immobilized CSPs PMBC-Im C, F, and G. For PMBC-Im C and F, 25% and 50% methanol in
water was used respectively. For PMBC-Im G, pure acetonitrile was used as the solvent for irradiation
of the suspended coated silica bead.
As indicated in Section 2.1, the immobilization rate is somewhat affected by the type of used
solvent. Increasing too much the methanol content seems to be less effective (PMBC-Im F) in terms
of immobilization. This was confirmed by a test using pure methanol which afforded only 49%
immobilization after 20 h. It has been reported that the stabilization of radicals can be considerably
influenced by the polarity of the solvent, affecting the speed and the course of photochemical processes
for small molecules [23–25]. However, this influence can be hardly predicted. Due to the insolubility of
the immobilized composite prepared in this study and the unknown structure after immobilization of
the cellulose derivative on silica gel, it is difficult to hypothesize about the exact influence of the solvent.
In terms of enantioselectivity, the three CSPs shows comparable results (Table 3) when
hexane–isopropanol 9/1 (v/v) is used as the mobile phase, indicating that the intrinsic chiral
recognition ability is identical for all phases and not much affected by the solvent conditions.
There is no general trend for a CSP to be more performing, and again it depends on the racemic
substance. For example, compounds
1, 5, and 13 are slightly better resolved on PMBC-Im F, while
compound , and are slightly better resolved on PMBC-Im G. CSP PMBC-Im C shows generally
2,
7
9
higher retention properties (higher retention factors) but this probably due to the higher content of
immobilized chiral material.

Molecules 2016, 21, 1740
6 of 12
Table 3. Chromatographic results of the enantioselective separations of selected racemic compounds
on photochemically immobilized PMBC-Im CSPs: influence of applied solvent for irradiation. Mobile
phase: hexane/2-propanol (9/1, v/v).
PMBC-Im G
PMBC-Im F
PMBC-Im C
Compound
α
k1
k2
Rs
α
k1
k2
Rs
α
k1
k2
Rs
1
2
4
5
7
7.22
1.40
1.00
1.24
1.22
2.21
1.70
1.38
1.36
1.59
1.59
3.63
8.07
8.66
5.23
3.53
6.13
1.36
2.18
2.78
11.50
5.06
8.07
10.80
6.37
7.80
10.50
1.87
2.98
4.43
5.04
1.28
0
0.64
0.82
2.76
2.06
0.80
0.86
1.74
7.52
1.34
1.08
1.27
1.17
2.13
1.76
1.34
1.32
1.58
1.40
3.59
6.93
7.40
5.32
3.25
5.62
1.44
2.26
2.62
10.51
4.81
7.50
9.38
6.21
6.93
9.87
1.94
2.98
4.15
6.32
1.51
0.51
0.95
0.76
3.45
2.66
1.04
1.11
2.27
7.05
1.32
1.05
1.23
1.18
2.17
1.72
1.39
1.37
1.61
1.85
4.87
9.85
10.86
7.42
4.80
7.62
1.85
2.91
3.65
13.09
6.41
10.40
13.40
8.76
10.40
13.10
2.56
4.53
1.36
nd
0.73
0.63
3.07
1.96
1.07
0.84
2.44
9
13
16
17
19
3.98
5.86
nd: not determined.
2.4. Modulating Enantioselectivity on Immobilized PMBC-Im CSPs with Mobile Phase Types and Composition
The unique feature of the immobilized polysaccharide-based phases is their ability to combine the
extended chiral recognition ability of polysaccharide derivatives with the utilization of an unlimited
range of mobile phases. This combination offers at least three major benefits, (i) the opportunity to
achieve or improve enantioselectivity; (ii) the option to modulate retention time according to specific
needs; (iii) the possibility to increase solubility of the racemic solute permitting to optimize productivity
for preparative separations.
These benefits are demonstrated by the numbers shown in Table 4 which summarizes the
chromatographic data of PMBC-Im E for a variety of racemic compounds using mixtures of heptane
and chloroform varying between heptane/chloroform 90/10 (v/v) and 60/40 (v/v) as the mobile
phase. Chloroform was chosen because it is considered as a highly solvating agent for many poorly
soluble compounds and possesses distinct properties in terms of selectivity. Values obtained on the
same material before and after immobilization with hexane/2-propanol 90/10 as the mobile phase
are also shown as a reference in Table 4. This mobile phase mixture is the usual mobile phase for the
non-immobilized polysaccharide-based CSPs which do not tolerate chlorinated solvents.
Table 4 shows the selectivity values, retention factors (k2) and resolution obtained for 19 racemic
compounds. Comparing the selectivity values of the CSPs PMBC-S and PMBC-Im E using
hexane/2propanol 90/10 (v/v) as the mobile phase clearly shows that the selectivities systematically
decrease after immobilization. The drop ranges between 2% and 15%, with one exception for compound
1
which shows a more significant decrease (25%) under the applied mobile phase conditions. However,
for most of the racemic compounds, the selectivity could be recovered by using heptane-chloroform
mixture as the mobile phase. For ten substances out of 15, the enantioselectivity was better on the
immobilized CSP PMBC-Im E and for 4 compounds (6, 11, 14, 15) the separation which could not be
achieved on the non-immobilized phase due to the poor solubility in the mobile phase or excessive
retention time (no elution) were easily resolved with heptane–chloroform mixtures as the mobile
phases. An example of improved separation and selectivity is shown in Figure 3 for compound 16
when changing the mobile phase from hexane/2-propanol to heptane/chloroform.

Molecules 2016, 21, 1740
7 of 12
Table 4. Chromatographic results of the enantioselective separations of racemates 1–19 on photochemically immobilized PMBC-Im E: influence of mobile phase.
PMBC-S
PMBC-Im E
PMBC-Im E
PMBC-Im E
PMBC-Im E
PMBC-Im E
Compound
Hexane/2-propanol 9/1
Hexane/2-propanol 9/1
Heptane/Chloroform 90/10
Heptane/Chloroform 75/25
Heptane/Chloroform 67.5/32.5
Heptane/Chloroform 60/40
α
k2
Rs
α
k2
Rs
α
k2
Rs
α
k2
Rs
α
k2
Rs
α
k2
Rs
1
2
9.15
1.47
1.37
1.00
1.30
16.89
7.82
3.71
1.21
0.60
0.00
0.65
6.89
1.30
1.30
1.12
1.22
13.62
6.30
4.37
1.04
2.44
0.86
0.66
3.29
1.42
1.31
1.25
1.30
1.11
2.99
1.00
2.11
1.30
3.69
4.07
6.36
2.76
3.08
2.00
1.90
nd
3.70
0.00
5.04
2.16
1.85
1.17
1.23
1.20
1.34
1.32
1.29
1.26
1.57
1.22
3.09
1.00
1.68
1.58
1.23
1.65
1.59
1.09
1.76
0.72
0.66
2.56
0.72
2.14
1.82
2.06
2.60
2.62
2.26
2.44
1.52
9.62
0.00
4.64
1.74
2.30
4.50
4.62
0.50
5.64
1.57
1.00
1.21
1.18
1.36
1.43
1.23
1.25
1.38
1.19
2.45
1.00
1.60
1.43
1.24
1.54
1.47
1.00
1.65
0.49
0.40
3.17
3.17
1.41
6.97
11.68
8.05
0.82
1.20
19.90
0.49
2.62
22.05
27.53
4.18
6.34
0.67
3.22
1.56
0.00
1.80
1.54
1.96
3.16
2.18
2.08
1.50
1.10
7.52
0.00
3.72
0.92
2.12
3.70
3.70
0.00
4.54
1.36
1.00
1.19
1.16
1.38
1.50
1.16
1.23
1.27
1.18
2.10
1.00
1.48
nd
0.37
0.30
2.24
2.09
1.01
4.46
7.58
5.27
0.62
0.88
8.77
0.36
1.69
12.04
14.27
2.95
4.31
0.50
2.22
0.80
0.00
1.46
1.16
1.76
3.30
nd
3
8.40
5.46
21.22
30.58
13.89
127.82
49.42
1.00
4.85
4
14.06
13.01
10.50
13.82
5.18
2.16
5
6
not soluble
9.35
not soluble
8.95
13.84
20.70
14.72
1.32
7
1.22
1.19
2.53
1.30
nd
nd
2.25
0.50
1.16
1.17
2.16
1.28
0.54
0.98
2.37
1.68
8
14.46
13.00
1.86
0.88
0.74
5.62
0.00
2.62
nd
9
10.11
10.29
7.32
10
11
12
13
14
15
16
17
18
19
8.28
8.03
10.49
not eluted
5.77
1.90
not soluble
13.75
not soluble
8.42
58.45
0.82
1.00
1.81
0.00
1.60
1.00
1.70
0.00
1.82
1.18
1.89
nd
4.96
14.23
13.67
35.00
not eluted
not eluted
29.16
52.23
4.91
4.99
not eluted
not eluted
2.39
not eluted
not eluted
2.69
59.68
66.00
6.55
1.25
1.43
1.37
1.00
1.55
nd
1.39
1.41
1.13
1.64
0.59
0.58
nd
1.36
1.33
1.09
1.57
0.86
0.85
nd
1.98
1.83
1.15
2.02
5.36
4.92
1.18
7.98
2.86
2.78
0.00
3.52
4.14
4.20
10.83
1.04
5.27
5.84
5.20
6.41
1.61
5.80
2.10
20.07
nd: not determined.

Molecules 2016, 21, 1740
8 of 12
Figure 3. Chromatographic resolution of compound 16. (a) on PMBC-S, mobile phase hexane/2-propanol
90/10 (v/v); ( ) on PMBC-Im E, mobile phase hexane/2-propanol 90/10 (v/v); ( ) on PMBC-Im E,
mobile phase heptane/chloroform 75/25 (v/v).
b
c
For compounds 8, 11, 14, and 15 a higher content of chloroform is necessary to achieve the
enantiomeric separation as retention is too long even with 10% chloroform in the mobile phase. Almost
all compounds show a selectivity and retention decrease when the proportion of chloroform in heptane
is increasing (Figures 4 and 5).
For compounds 12 and 18 the separation is even lost above 30% of chloroform, indicating that
for each enantiomeric pair the mobile phase composition has to be adjusted properly in order to keep
selectivity while reaching a reasonable analysis time. They are three exceptions (compounds 5, 6,
and 15) for which selectivity is increasing with augmenting the proportion of chloroform (Figure 4).
However, like for the other substances, the retention values are decreasing when the proportion of
chloroform is increased (Figure 5). This effect might be extremely valuable for reducing analysis time
as shown for compound 19 in Figure 6.
Figure 4. Chromatographic resolution of compound
1–19 on PMBC-Im E with heptane/chloroform as
the mobile phase. Enantioselectivity vs. chloroform content in heptane.

Molecules 2016, 21, 1740
9 of 12
Figure 5. Chromatographic resolution of compounds
1–19 on PMBC-Im E with heptane/chloroform as
the mobile phase. Retention factor k2 vs. chloroform content in heptane.
Figure 6. Chromatographic resolution of compound 19 on PMBC-Im E using heptane /chloroform
mixtures as a mobile phase: (a) 60/40 (v/v); (b) —– 75/25 (v/v); and (c) . . . . . . 90/10 (v/v).
Compound
6 exemplifies another benefit of the immobilized phase, namely the possibility
to separate the enantiomers of substance which are not soluble in the conventional mobile phase
(alkane–alcohol mixtures) for analysis or preparative scale resolutions. The compound is completely
insoluble in hexane/2-propanol (9/1 v/v), but well soluble in mixture of heptane and chloroform. This
solubility problem is often encountered in the practice.
For compounds
15 which are not eluted with heptane-chloroform 90/10 (v/v), increasing the chloroform content
permits to considerably reduce the retention times, up to almost 30 times for substance . This is of
4, 6, 7, 13 and 17 which exhibit very high retention factors and compounds 11,
14
,
6
course of high value for analytical separations which are preferably performed at short retention times
while maintaining a high enantioselectivity. This effect is illustrated for compound 6 on Figure 7.

Molecules 2016, 21, 1740
10 of 12
Figure 7. Chromatographic resolution of compound
6
on PMBC-Im E using heptane /chloroform
. . . . . . 75/25 (v/v).
mixtures as a mobile phase: ( 60/40 (v/v); ( ) —– 67.5/32.5 (v/v); and (c)
a
)
b
For a few compounds (2, 12, 18) higher contents of chloroform cause a rapid elution of the two
enantiomers which practically coelute at the dead time, leading to a loss of the separation.
3. Materials and Methods
3.1. Preparation of para-Methylbenzoyl Cellulose (PMBC)
para-Methylbenzoyl cellulose (PMBC) was prepared by reaction of microcrystalline cellulose
(Avicel, Merck, Darmstadt, Germany) with an excess of 4-methylbenzoyl chloride in a mixture of
pyridine and triethylamine in the presence of a catalytic quantity of 4-(dimethylamino)pyridine, as
previously described [8].
3.2. Preparation Silica Gel Coated PMBC (PMBC-S)
Typically, a solution of 3.9 g PMBC dissolved in 100 mL dichloromethane is added to silica
gel (10 g) (Nucleosil Si 4000-10, from Macherey-Nagel, Düren, Germany), which was preliminarily
modified with 3-aminopropyltriethoxysilane. The mixture is stirred moderately for 30 min before
evaporation of the solvent at 35 ◦C under reduced pressure (400 mbar) by means of a rotavapor.
3.3. Preparation of Immobilized PMBC CSP (PMBC-Im)
After coating, 5.1 g of the obtained solid were suspended in 400 mL of a mixture of methanol and
water or in pure acetonitrile. A high-pressure mercury-vapor lamp (125 W, HPK, Philips, Brussels,
Belgium) with a Pyrex lamp housing was dipped into this suspension and turned on. The irradiation
time was varied between 18 and 30 h. After photochemical treatment, the solution is filtered off and the
obtained powder is washed with pure ethanol. The material is suspended for 24 h at room temperature
in a mixture of 10 mL ethanol, 50 mL chloroform and 100 mL tetrahydrofuran in order to dissolve
the non-bonded parts of the polymer. The solid phase is filtrated off and again washed with 100 mL
chloroform and 100 mL THF. The dried material is finally suspended in a mixture of dichloromethane
and chloroform (each 20 mL). An 8-fold volume of hexane was added dropwise to the suspension
(1.6 mL/min) under moderate magnetic stirring. After completion of the addition of hexane, the
◦
suspension is filtrated off and the isolated solid is dried for 2 h at 120 C. The final coating ratio, which
is defined as the ratio of weight of the immobilized polymer on the total phase weight, was estimated
by the recovered amount of methylene chloride-soluble PMBC and by elemental analysis.

Molecules 2016, 21, 1740
11 of 12
3.4. Column Packing of Immobilized PMBC CSPs
The material was packed into a stainless-steel column (250
the slurry method using a mixture of hexane/ethanol (70/30 v/v) as a carrying liquid.
×
4.6 mm I.D., Macherey-Nagel) by
3.5. Equipment and Chromatographic Conditions
The HPLC system consisted of a LC-6A or LC-10 unit (Shimadzu, Kyoto, Japan) equipped with
UV-visible detector and a 20-microliter sample loop. All chromatographic runs were performed at
room temperature at a flow rate of 0.7 mL/min. The eluates were monitored at 254 nm. The dead time
was determined by injecting 1,3,5-tri-tertbutylbenzene onto the column.
3.6. Racemic Samples
Compounds 2, 5, 6, 9–13, 15, and 19 were from Novartis. The synthesis of compounds 16 and
17 is described elsewhere [26]. Compounds 1, 3, 4, 7, 8, and 14 were purshased from Sigma-Aldrich
(Buchs, Switzerland).
4. Conclusions
Irradiation of coated PMBC phases with UV light constitutes a simple and efficient technique
to immobilize para-methylbenzoyl cellulose on silica gel. The immobilized phase combines the
advantages of the high chiral recognition ability of the cellulose derivative and the option to apply
a wide range of mobile phase types, especially those which are not tolerated by the corresponding
non-immobilized phase because of its high solubility in most organic solvents. It was shown that the
extended range of applicable mobile phases can be exploited not only to separate the enantiomers of
substances which are insoluble in the classical mobile phases but in many instances also to increase
selectivity and to considerably reduce excessive retention times for strongly retained compounds.
With the immobilized phase almost any kind of mobile phase is applicable, giving a highly extendable
means to improve and adjust the chromatographic parameters such as retentiveness, enantioselectivity,
and solubility of the solutes according to the intended purpose.
Author Contributions: E.F. conceived the project. D.H. and T.Z. designed and performed the experiments, and
analyzed the data.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
Okamoto, Y.; Yashima, E. Polysaccharide derivatives for chromatographic separation of enantiomers.
Angew. Chem. Int. Ed. 1998, 37, 1020–1043. [CrossRef]
Ikai, T.; Okamoto, Y. Structure Control of Polysaccharide Derivatives for Efficient Separation of Enantiomers
by Chromatography. Chem. Rev. 2009, 109, 6077–6101. [CrossRef] [PubMed]
Okamoto, Y.; Kawashima, M.; Hatada, K. Useful chiral packing materials for high-performance liquid
chromatographic resolution of enantiomers: Phenylcarbamates of polysaccharides coated on silica gel. J. Am.
Chem. Soc. 1984, 106, 5357–5359. [CrossRef]
4.
5.
6.
7.
Okamoto, Y.; Kawashima, M.; Hatada, K. Chromatographic resolution: XI. Controlled chiral recognition of
cellulose triphenylcarbamate derivatives supported on silica gel. J. Chromatogr. 1986, 363, 173–186. [CrossRef]
Okamoto, Y.; Kaida, Y. Polysaccharide derivatives as chiral stationary phases in HPLC. J. High Res. Chromatogr.
1990, 13, 708–712. [CrossRef]
Okamoto, Y.; Kaida, Y. Resolution by High-performance Liquid Chromatography Using Polysaccharide
Carbamates and Benzoates as Chiral Stationary Phases. J. Chromatogr. A 1994, 666, 403–419. [CrossRef]
Francotte, E.; Wolf, R.M. Benzoyl cellulose beads in the pure polymeric form as a new powerful sorbent for
the chromatographic resolution of racemates. Chirality 1991, 3, 43–55. [CrossRef]

Molecules 2016, 21, 1740
12 of 12
8.
9.
Francotte, E.; Wolf, R.M. Chromatographic Resolution on Methylbenzoyl Cellulose Beads. Modulation of the
Chiral Recognition by Variation of the Position of Methyl Group on the Aromatic Ring. J. Chromatogr. 1992
595, 63–75. [CrossRef]
,
Ikai, T.; Yamamoto, C.; Kamigaito, M.; Okamoto, Y. Immobilized Polysaccharide-Based Chiral Stationary
Phases for HPLC. Polym. J. 2006, 38, 91–108. [CrossRef]
10. Shen, J.; Ikai, T.; Okamoto, Y. Synthesis and application of immobilized polysaccharide-based chiral stationary
phases for enantioseparation by high-performance liquid chromatography. J. Chromatogr. A 2014, 1363, 51–61.
[CrossRef] [PubMed]
11. Shen, J.; Okamoto, Y. Efficient Separation of Enantiomers Using Stereoregular Chiral Polymers. Chem. Rev.
2016, 116, 1094–1138. [CrossRef] [PubMed]
12. Okamoto, Y.; Aburatani, R.; Miura, S.; Hatada, K. Chiral Stationary Phases for HPLC: Cellulose
Tris(3,5-dimethylphenylcarbamate) and Tris(3,5-dichlorophenylcarbamate) Chemically Bonded to Silica Gel.
J. Liq. Chromatogr. 1987, 10, 1613–1628. [CrossRef]
13. Oliveros, L.; Lopez, P.; Minguillon, C.; Franco, P. Chiral Chromatographic Discrimination Ability
of a Cellulose 3,5-Dimethylphenylcarbamate/10-Undecenoate Mixed Derivative Fixed on Several
Chromatographic Matrices. J. Liq. Chromatogr. 1995, 18, 1521–1532. [CrossRef]
14. Francotte, E.; Huynh, D. Immobilized halogeno-phenylcarbamate derivatives of cellulose as novel chiral
stationary phases for enantioselective drug analysis. J. Pharm. Biomed. Anal. 2002, 27, 421–429. [CrossRef]
15. Wetli, H.; Francotte, E. Automated screening platform with isochronal-parallel analysis and conditioning for
rapid method development of preparative chiral separations. J. Sep. Sci. 2007, 30, 1255–1261. [CrossRef]
[PubMed]
16. Francotte, E. Photochemically Cross-Linked Polysaccharide Derivatives as Supports for the Chromatographic
Separation of Enantiomers. Patent WO 9627615 A1, 16 September 1996.
17. Francotte, E.; Zhang, T. Photochemically Cross-Linked Polysaccharide Derivatives Having no
Photopolymerisable Functional Groups. Patent WO 9704011 A1, 6 February 1997.
18. Rabek, J.F. Photodegradation and photo-oxidative degradation of heterochain polymers. In Polymer
Photodegradation: Mechanisms and Experimental Methods; Chapman & Hall: London, UK, 1995; p. 342.
19. Fan, H.; Li, G.; Yang, F.; Yang, L.; Zhang, S. Photodegradation of cellulose under UV light catalysed by TiO2.
J. Chem. Technol. Biotechnol. 2011, 86, 1107–1112. [CrossRef]
20. Yatagai, M.; Zeronian, S.H. Effect of ultraviolet light and heat on the properties of cotton cellulose. Cellulose
1994, 1, 205–214. [CrossRef]
21. Francotte, E.; Wolf, R.M. Preparation of chiral building blocks and auxiliaries by chromatography on cellulose
triacetate (CTA I). Indications for the presence of multiple interaction sites in CTA I. Chirality 1990, 2, 16–31.
[CrossRef]
22. Klán, P.; Šolomek, T.; Bochet, C.G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.; Kostikov, A.; Wirz, J.
Photoremovable Protecting Groups in Chemistry and Biology: Reaction Mechanisms and Efficacy. Chem. Rev.
2013, 113, 119–191. [CrossRef] [PubMed]
23. Stewart, G.; Smith, K.; Chornes, A.; Harris, T.; Honeysucker, T.; Dasary, R.S.; Yu, H. Photochemical reaction
of nitro-polycyclic aromatic hydrocarbons: Effect by solvent and structure. Environ. Chem. Lett. 2010, 8,
301–306. [CrossRef] [PubMed]
24. Reusch, W. Photochemistry. Website Michigan State University. Available online: https://www2.chemistry.
msu.edu/faculty/reusch/virttxtjml/photchem.htm (accessed on 6 December 2016).
25. Ninomiya, I.; Naito, T. Photochemical Synthesis; Academic Press: San Diego, CA, USA, 1989;
ISBN: 0-12-519490-0.
26. Francotte, E.; Lang, R.W.; Winkler, T. New chiral fluoroanthryl derivatives: Resolution of the enantiomers by
chromatography on cellulose esters and their evaluation as chiral solvating agents in NMR spectroscopy.
Chirality 1991, 3, 177–182. [CrossRef]
©
2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).