J

ISSN 1615-9306 · JSSCCJ 42 (17) 2743– 2882 (2019) · Vol. 42 · No. 17 · September 2019 · D 10609

JOURNAL OF

S SEPARATION
S SCIENCE 17 19

Methods www.jss-journal.com
Chromatography · Electroseparation

Applications
Biomedicine · Foods · Environment
Received: 26 March 2019 Revised: 1 June 2019 Accepted: 19 June 2019

DOI: 10.1002/jssc.201900307

RESEARCH ARTICLE

Epoxide-derived mixed-mode chromatographic stationary

phases for separation of active substances in fixed-dose
combination drugs

Shuanghong Zhang Qian-Hong Wan Yan Li

School of Pharmaceutical Science &

Technology, Tianjin University, Tianjin,
P. R. China
Correspondence
Shuanghong Zhang, School of Pharmaceutical
Science & Technology, Tianjin University, 92
Weijin Road, Tianjin, 300072, P. R. China.
Email: zsh2016ky@sina.com
A method for the preparation of novel mixed-mode reversed-phase/strong cation
exchange stationary phase for the separation of fixed-dose combination drugs has
been developed. An epoxysilane bonded silica prepared by vapor phase deposition was
used as a starting material to produce diol, octadecyl, sulfonate, and mixed octade-
cyl/sulfonate groups bonded silica phases. The chemical structure and surface cover-
age of the functional groups on these synthesized phases were confirmed by fourier-
transform infrared and solid-state 13 C NMR spectroscopy and elemental analysis.
Alkylbenzene homologs, basic drugs, nucleobases and alkylaniline homologs were
used as probes to demonstrate the reversed-phase, ion exchange, hydrophilic inter-
action and mixed-mode retention behaviors of these stationary phases. The octade-
cyl/sulfonate bonded silica exhibits pronounced mixed-mode retention behavior and
superior retentivity and selectivity for alkylaniline homologs. The mixed-mode reten-
tion is affected by either ionic or solvent strength in the mobile phase, permiting opti-
mization of a separation by fine tuning these parameters. The mixed-mode stationary
phase was applied to separate two fixed-dose combination drugs: compound reserpine
tablets and compound methoxyphenamine capsules. The results show that simultane-
ous separation of multiple substances in the compound dosage can be achieved on the
mixed-mode phase, which makes multi-cycles of analysis for multiple components
obsolete.

KEYWORDS
basic compounds, fixed-dose combinations, mixed-mode chromatography, retention mechanisms, station-
ary phases

1 I N T RO D U C T I O N
A fixed-dose combination (FDC) drug is defined as a drug
product in which two or more drug components are combined
in a single dosage form, such as a capsule or tablet. By reduc-
Article Related Abbreviations: FDC, fixed-dose combination; SE,
epoxysilane bonded silica; SE–C18 , octadecyl bonded silica; SE–C18 /SO3 H,
octadecyl/sulfate bonded silica; SE–Diol, diol bonded silica; SE–SO3 H,
sulfonate bonded silica.
ing the number of pills a person must take each day, FDC
drugs can help improve adherence to a treatment regimen [1].
In addition, the combined use of the active substances may
improve the therapeutic efficacy in comparison to the use
of the single active substance [2]. However, the diverse
properties of the active substances in the FDC drugs pose
significant analytical challenges.
The development of new rapid and efficienct analytical
techniques for the determination of active substances in FDC
drugs is part of the current effort to ensure their quality [3].

2796 © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com J Sep Sci 2019;42:2796–2804.
ZHANG ET AL.
Reversed-phase chromatography remains the preferred and
most efficient methods for analysis of FDC drugs [4]. How-
ever, the application of this method is limited as ionic species
are hardly retained in this mode of separation. A number of
measures have been taken to alleviate this problem, which
include ion suppression or ion-pair chromatography [5]. In
ion suppression, base or acid is added into the mobile phase,
which may exceed the tolerance pH range of silica based sta-
tionary phases [6]. In ion-pair chromatography, the use of sur-
factant in the mobile phase requires long equilibrium time.
The majority of ion-pairing agents are not volatile, making it
difficult to work with online MS. The use of stationary phases
with multiple functional groups has been proposed as an alter-
native approach to mitigate those problems.
The stationary phases with multiple functionalities are
commonly known as mixed-mode stationary phases. Accord-
ingly, mixed-mode chromatography is coined to refer to a
chromatography process in which two or more retention
mechanisms are operating [7–9]. Owing to synergistic action
between multiple functionalitities, mixed-mode chromatog-
raphy offers several advantages over conventional single-
mode chromatography. For example, the retention of ionic
compounds can be enhanced significantly in mixed-mode
chromatography due to cooperative binding. It is not neces-
sary to adjust the pH value of the mobile phase to exceed
the usable range, and to add surfactants to the mobile
phase to form ion pairs. Recently, Calinescu and colleagues
demonstrated successful separation of acetaminophen and its
impurities by using reversed-phase/cation exchange station-
ary phase [10]. The majority of the reversed-phase/cation
exchange stationary phases reported so far are based on the
use of benzenesulfonic acid as ion exchange ligand. For
solutes with aromatic rings, the retention mechanism is not
purely ionic interactions owing to the fact that π-π inter-
actions are super imposed on the ionic interactions, lead-
ing to peak broadening [11,12]. For this consideration, it
is of great interest to develop mixed-mode chromatographic
stationary phases containing no aromatic rings and apply
them to the separation of basic aromatic compounds in FDC
drugs.
In this study, we present a report on the synthesis, char-
acterization of a new mixed-mode stationary phase and its
application to separation of FDC drugs. Epoxysilane bonded
silica was used as an intermediate phase for preparation of
stationary phases with multiple functionalities [13,14]. Diol,
octadecyl, sulfonate and octadecyl/sulfonate functionalized
silica samples were prepared by following ring opening, acy-
lation and bisulfite addition reaction, respectively. The resul-
tant phases were characterized by FTIR spectroscopy, solid-
state 13 C NMR spectroscopy, elemental analysis. Their reten-
tion properties were evaluated in different chromatographic
operation modes and the effects of ion and solvent strength in
the mobile phase on retention were investigated. Finally, the
2797
potential of the mixed-mode stationary phase in pharmaceu-
tical analysis was explored by separation of two FDC drugs.
2 M AT E R I A L S A N D M E T H O D S
2.1 Chemicals and reagents
Porous silica spherical particles, BaseLine Sil-8 (8 μm,
300 m2 /g, pore size 10 nm) and a BaseLine C18 column
(150 × 4.6 mm ID, 5 μm) were supplied by BaseLine
Chromtech Research Centre (Tianjin, China). Epoxysilane, γ-
(2,3-epoxypropoxy)- propytrimethoxysilane, was purchased
from Wuda Silicone Materials (Wuhan, China). Sodium
bisulfite, hydrochloric acid, acetonitrile, methanol, potassium
phosphate monobasic, and ammonium formate were pur-
chased from Guangfu Chemical Reagents (Tianjin, China).
Stearoyl chloride was purchased from Xiensi (Tianjin, China).
Test compounds were obtained from various sources. FDC
drugs were sourced locally: compound reserpine tablets by
Yabao Pharmaceuticals, the batch number 180631; compound
methoxyphenamine capsules by Changxing Pharmaceuticals,
the batch number 20180321.
2.2 Preparation of stationary phases
The HCl activated and dried silica particles (10 g) and
expoxysilane (24 mmol) were placed in a Teflon-lined auto-
clave and heated at 150◦ C for 8 h. The resultant particles,
denoted as SE, were washed with ethanol and dried in vac-
uum at 80◦ C overnight.
The SE particles (5 g) were heated in 50 mL of 0.5% H2 SO4
at 80◦ C for 8 h. The silica particles were filtered and washed
with water until neutral. Then the silica particles, denoted as
diol bonded silica (SE–Diol), were dried in vacuum at 80◦ C
overnight.
The SE–Diol particles (5 g) were dispersed into 50 mL of
N,N-dimethylformide. The suspension was cooled in ice bath
and stearoyl chloride (3 mL, 12 mmol) was dropwise added
with continuous stirring. Upon completion of the addition,
the suspention was brought to room temperature and the reac-
tion was continued with shaking for 9 h. The functionalied sil-
ica particles were filtered and washed with toluene and alco-
hol successively. Then, the silica particles, denoted as octade-
cyl bonded silica (SE–C18 ), were dried in vaccum at 80◦ C
overnight [15].
The SE particles (5 g) were dispersed into 50 mL of 0.5 M
sodium bisulfite and the pH of the suspension was adjusted to
pH 7 with triethylamine. The mixture was heated at 80◦ C for
8 h. The silica particles were filtered and washed with dis-
tilled water. Then the silica particles, denoted as sulfonate
bonded silica (SE–SO3 H), were dried in vaccum at 80◦ C
overnight [16].
2798
FIGURE 1 Schematic showing the synthesis of the stationary phases
The preparation of octadecyl/sulfonate group bonded sil-
ica, denoted as octadecyl/sulfate bonded silica modified
(SE–C18 /SO3 H), was the same as for SE–C18 except that
SE–SO3 H (5 g) was used instead of SE–Diol.
2.3 Characterization
IR and solid-state 13 C NMR spectra of the particle sam-
ples were obtained on a Tensor 27 spectroscopy (Bruker,
Germany) and a solid-state Varian Infinity Plus-300 MHz
spectrometer (Varian, American), respectively. The elemental
analysis was performed on a VarioEL (Elementar, Germany).
2.4 Sample preparation
The sample solutions of alkylbenzene and p-alkylaniline
homologs were prepared in methanol with each substance at
5 μL/mL. The sample solutions of basic drug probes (dicyan-
diamide, melamine, and metformin) and nucleobases (ade-
nine, cytosine, guanine, thymine, and uracil) were prepared
in water at 1 mg/mL for each substance.
The sample solutions of compound reserpine tablets were
prepared as follows: 10 tablets were grinded into fine pow-
ders. The powders were transferred to a 25-mL volumetric
flask and diluted by the mixture solution of methanol water
(50:50, v/v) to the marker.
The sample solutions of compound methoxyphenamine
capsules were prepared as follows: the contents taken from
five capsules were grinded finely. The powders were trans-
ferred into a 50-mL volumetric flask. 20 mL ACN was added
and the mixture was ultrasonicated for 20 min. Then, 10 mL
ZHANG ET AL.
of water was added and ultrasonicated for 5 min and ACN was
added to the marker.
2.5 Chromatographic evaluation
Stainless steel columns (150 × 4.6 mm i.d.) were packed with
the synthesized phases using a conventional slurry packing
technique. Chromatographic evaluation was conducted on an
Agilent 1200 series HPLC system equipped with a vacuum
degasser, quaternary pump, autosampler, column oven, vari-
able wavelength UV-Vis detector and ChemStation version B
03.01 from Agilent Technologies (Palo Alto, CA, USA).
3 RESULTS AND DISCUSSION
3.1 Synthesis and characterization
The procedures for synthesis of functionalized silica parti-
cles are illustrated in Figure 1. A vapor-phase silanization
method was employed to produce epoxide modified silica SE,
from which diol, octadecyl ester (C18 ), sulfonate (SO3 H), and
C18 /SO3 H functionalized silica phases were obtained by fol-
lowing ring opening, acylation and bisulfate addition reaction,
respectively.
The synthesized phases were characterized by infrared
spectroscopy. As shown in Figure 2A, the bands at 3464 and
1050 cm−1 assigned to the stretching vibration of O–H and
Si–O bonds respectively, can be observed for all the samples
under study. The bands at 2863 and 2929 cm−1 attributed to
the stretching vibration of C–H indicate the bonding of the
ZHANG ET AL. 2799

FIGURE 2 (A) The infrared and (B) solid-state 13 C NMR spectra of the silica samples

epoxysilane. The band around 1700 cm−1 due to the stretch- TABLE 1 Elemental analysis and surface coverage of stationary
ing vibration of C=O suggests the formation of octadecyl phases
ester [17]. The bands around 649 cm−1 attributed to sul- Elemental
fonate can be oberseved in the spectra of SE–SO3 H and analysis Surface coverage (𝛍mol/m2 )
SE–C18 /SO3 H [18]. These results confirm that the functional Phase C% S% Epoxya Diola C18 a SO3 Hb
groups have been successfully bonded to the silica substrate. SE 8.02 0 4.62 N/A N/A N/A
The chemical structures of the functional groups on the SE–Diol 5.52 0 N/A 3 N/A N/A
synthesized phases were further studied by solid-state 13 C SE–C18 11.14 0 N/A 1.89 1.11 N/A
NMR spectroscopy. As shown in Figure 2B, peaks around SE–SO3 H 7.70 0.73 N/A 3.53 N/A 0.97
90 ppm attributed to the oxygen linked carbon can be observed SE–C18 /SO3 H 10.93 0.81 N/A 2.87 0.61 1.14
for all the phases. The large peaks around 60 ppm attributed a
The surface coverage calculate based on carbon loading.
to the carbon of the long alkyl chain observed for SE–C18 and b
The surface coverage calculate based on sulfur loading.
–C18 /SO3 H verify the formation of octadecyl ester [19]. Small
peaks around 50 ppm attributed to sulfonate linked carbon on the surface. A study is underway to investigate the factors
appear in the spectra of SE–SO3 Na and –C18 /SO3 H phases. that influence the convertion yields during synthesis.
Based on these observations, it can be concluded that prepara-
tion of the multiple stationary phases with intended functional 3.2 Chromatographic evaluation
groups has been successfully achieved by the synthesis proce-
dures developed in this work. A comparative study was carried out to reveal the retention
Surface coverages of the functional groups in the syn- properties of the synthesized phases. For this purpose, a vari-
thesized stationary phases were calculated from carbon and ety of compounds were chosen as probes and their reten-
sulfur percentages determined by elemental analysis [20]. tion behaviors were evaluated under typical reversed-phase,
As shown in Table 1, the surface density of the epoxy group ion exchange, hydrophilic interaction, and mixed-mode con-
for the intermediate phase SE is 4.62 μmol/m2 . When it was ditions, respectively.
converted to diol bonded phase under acidic condtions, the
surface coverage was reduced to 3 μmol/m2 , probably due to 3.2.1 Reversed-phase mode
chemical leaching. The surface concentration of the octadecyl A homolog series of alkylbenzene were used as probes to
ester group on SE–C18 is 1.11 μmol/m2 , about one third of evaluate the reversed-phase properties of the synthesized sta-
that for SE–Diol. The surface density of the sulfate group on tionary phases. As shown in Figure 3A, the alkylbenzenes
SE–SO3 H is 0.97 μmol/m2 . While the sulfonate concentra- are eluted at dead time on SE–Diol and –SO3 H, but increas-
tion on SE–C18 /SO3 H is comparable to that on SE–SO3 H, ingly retained on SE–C18 /SO3 H and baseline separated on
the octadecyl ester concentration is about half of that for SE–C18 with an elution order commensurate with the increas-
SE–C18 , even though they were prepared with the same ratio ing side chain length. Evidently, nonpolar or hydropho-
of stearoyl chloride to the silica substrate. The decrease in bic interactions are the principal driving force for solute
the surface coverage for SE–C18 /SO3 H may be attributed to retention in this mode, and therefore these phases can be
the steric hindrance arising from the bonded sulfonate groups ranked according to their hydrophobicity in the order of
2800 ZHANG ET AL.

FIGURE 3 Chromatographic evaluation. (A) Reversed-phase mode: mobile phase, methanol-water (60: 40, v/v); injection volumn, 5 μL;
temperature; 30◦ C; wavelength, 254 nm. Solute: 1, toluene; 2, ethylbenzene; 3, n-propylbenzene; 4, n-butylbenzene; 5, amylbenzene. (B) Ion
exchange mode: mobile phase, ACN- NH4 H2 PO4 (1.7%, m/m, pH 3.0) (30: 70, v/v); injection volumn, 5 μL; temperature, 30◦ C; wavelength,
220 nm. Solute:1, dicyandiamide; 2, melamine; 3, metformin. (C) Hydrophilic interaction mode: mobile phase, water- HCOONH4 (20 mM, pH 6.8)
(10: 90, v/v); injection volumn, 5 μL; temperature, 30◦ C; wavelength, 254 nm. Solute: 1, thymine; 2, uracil; 3, adenine; 4, cytosine; 5, guanine. (D)
Mixed interaction mode: mobile phase, ACN-KH2 PO4 (30 mM, pH 2.5) (40: 60, v/v); UV wavelength, 254 nm; injection volumn, 5 μL; temperature,
30◦ C; wavelength, 254 nm. Solute: 1, p-toluidine; 2, p-ethylaniline; 3, p-propylaniline; 4, p-butylaniline; 5, p-pentylaniline

SE–Diol=SO3 H < –C18 /SO3 H < –C18 . The higher selectiv- the order of SE–C18 < –Diol < –SO3 H < –C18 /SO3 H.
ity observed with the SE–C18 phase may be attributed to its Obviously, the sulfonate group has a pronounced effect on the
greater octadecyl ester surface coverage when compared with hydrophilicity of the SE–C18 /SO3 H phase as it increases the
the mixed-mode phase SE–C18 /SO3 H. wettability of otherwise the essentially hydrophobic phase.

3.2.2 Cation exchange mode 3.2.4 Mixed interaction mode

A basic drug substance metformin hydrochloride and its
related impurities dicyandiamide and melamine were used as
probes to investigate the cation exchange properties of the
synthesized stationary phases. As shown in Figure 3B, they
are eluted at dead time on SE–Diol and -C18 , but baselined
separated on SE–SO3 H and –C18 /SO3 H with an elution order
of dicyandiamide < melamine < metformin. Ionic interac-
tions are expected to be the principal driving force in this
mode of chromatography and therefore the phases can be
ranked according to their ion exchange strength in the order of
SE–Diol=–C18 < –SO3 H < –C18 /SO3 H. The slightly higher
selectivity for basic compounds on SE–C18 /SO3 H may be
attributed to its greater surface coverage of sulfonate groups.
3.2.3 Hydrophilic interaction mode
A set of five nucleobases including adenine, cytosine, gua-
nine,thymine, and uracil were used as probes to characterize
the hydrophilic interaction properties of the synthesized
stationary phases. As shown in Figure 3C, they are weakly
retained and poorly separated on SE–C18 and -Diol. How-
ever, a significant improvement in retention and separation is
observed on SE–SO3 H and –C18 /SO3 H with an elution order
of thymine < uracil < adenine < cytosine < guanine, which
is typical in this mode of separation [8,21]. These stationary
phases can be ranked according to their hydrophilicity in
A homolog series of alkylanilines were used as probes to
gauge the mixed-mode properties of the synthesized phases.
As shown in Figure 3D, the alkylanilines are not retained on
SE–Diol and only weakly retained on SE–C18 and –SO3 H.
However, a significant improvement in retention and selectiv-
ity occurs with SE–C18 /SO3 H, leading to baseline separation.
The alkylanilines are eluted in the order consistent with their
increasing side-chain length.
To understand these behavior, it is necessary to consider
that the alkylanilines are completely ionized in the mobile
phase at pH 2.5 and therefore act as organic ions with
hydrophobic tails. They are highly solvated under the separa-
tion conditions, making them preferable to stay in the mobile
phase when the hydrophobicity or ionic interaction strength of
the stationary phase is weak. This is obviously the case with
SE–Diol. They are slightly retained and separated on SE–C18
due to increasing hydrophobicity of the stationary phase. In
this case the principal driving for retention is nonpolar inter-
actions between the hydrophobic tail of the solute and octade-
cyl group of the stationary phase. This is manifested by the
elution order of increasing side chain length. In the case of
SE–SO3 H, the alkylanilines are weakly retained and eluted in
a reversed order. The change in elution order indicates that
the principal driving force for retention is ionic interactions
of the ionized amino group with the sulfonate group on the
ZHANG ET AL.
FIGURE 4 Effects of Ionic strength. (A) Plots of log k vs. log [K+ ] and (B) plots of k vs. 1/[K+ ] for p-alkylaniline homologs. Mobile phase,
50% ACN-KH2 PO4 (potassium concentration varying from 10 mM to 50 mM, pH 2.5); others same as for Figure 3D. Solute: (■) p-toluidine; (●)
p-ethylaniline; (▲) p-propylaniline; (▼) p-butylaniline; (◆) p-pentylaniline
stationary phase. The alkylanilines with longer side chains
tend to be less retained because they are increasingly sol-
vated by organic solvent molecules in the mobile phase. It can
be concluded that the retention of the alkylanilines on both
phases SE–C18 and –SO3 H is governed by monovalent inter-
actions or single-mode retention mechanism.
In contrast, alkylanilines are capable of engaging bivalent
interactions with the bifunctional groups on the stationary
phase. The ionized amine head and hydrophobic tail of the
alkylaniline molecule can interact simultaneously with the
complementary groups, i.e., sulfonate and octadecyl ester, on
the stationary phase to form transient bivalent molecular com-
plexes. Thus, the retention of the alkylanilines is markedly
increased on the mixed-mode phase than its single-mode
counterparts.
3.3 Effect of mobile phase composition
on retention
In view of the different retention behaviors observed for the
alkylanilines on single-mode and mixed-mode phases, it is of
interest to learn more about the mechanisms underlying these
observations. For this purpose, experiments were performed
to study the effects of the mobile phase composition on the
retention of alkylanilines. SE–Diol was not included in this
part of study because of its poor retentivity for alkylanilines.
3.3.1 Ionic strength
To study the effects of ionic strength, the concentration of
potassium dihydrogen phosphate in the mobile phase was var-
ied from 10 to 30 mM while the volume ratio of the organic
solvent acetonitrile was fixed at 50%. According to classical
ion exchange theory [22], a plot of logarithm of retention fac-
tor (logk) vs. logarithm of concentration of counterions (logC)
should produce a straight line with a slope value equal to -1 for
2801
singly charged ions. Any deviations from -1 suggest non-ionic
contributions such as hydrophobic interactions to the reten-
tion.
With alkylanilines as analytes, the plots of logk versus log
[K+ ] are shown in Figure 4A and regression analysis data
are summarized in Supporting Information Table S1. The
slope values are in the range of −0.97 to −1.02, close to
-1on SE–SO3 H. By contrast, significant deviation from -1
are noted on SE–C18 and –C18 /SO3 H phases, indicating that
hydrophobic interactions are superimposed on the ionic inter-
actions as previously reported in the literature [22].
The hydrophobic contribution to the retention can be esti-
mated by extrapolating the plot of k vs. 1/C to its origin as
previously reported. According to the displacement model
for ion exchange chromatography, a plot of k vs. recipro-
cal concentration of counterions, i.e., 1/C, should produce a
straight line with the intercept value assigned to the reversed-
phase retention component. The plots of k versus 1/[K+ ] are
shown in Figure 4B and the regression analysis data are sum-
marized in Supporting Information Table S2. The reversed-
phase retention components are in the range of −0.069 −
0.79, 0.042 − 0.16, and 0.79 − 2.51, respectively, on SE–C18 ,
–SO3 H and –C18 /SO3 H. The mixed-mode phase displays sig-
inicantly greater reversed-phase retention components, allow-
ing improved separations of basic compounds with varying
hydrophobicities to be achieved.
3.3.2 Solvent strength
As both the reversed-phase and ion exchange retention mech-
anisms can be affected by the variation of the solvent strength
in the mobile phase, it is important to learn how the mixed-
mode retention varies with the change in the solvent strength.
To study this effect, the volume ratio of the organic solvent
in the mobile phase was varied from 40 to 100% whereas the
concentration of ammonium formate (pH 2.5) was maintained
2802 ZHANG ET AL.

exchange (SE–C18 /SO3 H) mechanism where the retention

is all reduced by the solvation effect. At the other end of
solvent strength range, the retention is governed by mixed
hydrophilic/ionic interactions where the retention is increased
with a reduction in electric permittivity of the mobile
phase.

3.4 Applications
Comparative studies on the synthesized phases show that
SE–C18 /SO3 H is unique for its high retentivity and selectiv-
ity for basic compounds. Thus, experiments were performed
to explore the potential of this mixed-mode phase in analysis
of FDC drugs or compound dosages.

FIGURE 5 Effects of solvent strength. Mobile phase, ACN
(varying from 40 to 100%)- HCOONH4 (ammonium concentration
fixed at 20 mM, pH 2.5); others same as for Figure 3D. Solute: (■)
p-toluidine; (●) p-ethylaniline; (▲) p-propylaniline; (▼)
p-butylaniline; (◆) p-pentylaniline
fixed at 20 mM. As shown in Figure 5, the retention of alky-
lanilines decrases continuously until a minimum at about 80–
90% is reached. And then it increases with further increase
in the organic solvent valume ratio. These “U” shaped curves
suggest a change in retention mechanism occurs with varia-
tion of the solvent strength. Over the range from 40 to 80%,
the retention is governed by either reversed-phase (SE–C18 )
or ion exchange (SE–SO3 H) or mixed reversed-phase/ion
3.4.1 Separation of compound
reserpine tablets
Compound reserpine tablets are often prescribed to treat high
blood pressure. The drug contains nine components, in which
six organic compounds are of particular interest as active
ingredients. At present, the determination of these six com-
ponents has been carried out in batches under different con-
ditions because they differ widely in property and concen-
tration in the dosage [23]. Comparative chromatograms of
compound reserpine tablets obtained on SE–C18 /SO3 H and
commercial BaseLine C18 phases are shown in Figure 6A.
Six components of compound reserpine tablets are simulta-
neously separated on the SE–C18 /SO3 H phase. By contrast,
only four peaks are observed on the conventional C18 bonded

FIGURE 6 Separation of fixed-dose combination drugs. (A) Compound reserpine tablets. Columns, SE–C18 /SO3 H (up) and Baseline C18
(bottom); mobile phase, ACN as mobile phase A, HCOONH4 (100 mM, pH 3.0) as mobile phase B, water as mobile phase C, gradient elution: from
0 to 3 min, A: 5% B: 50% C: 45%; from 3 to 20 min, A 40% B 50% C 10%; from 20 to 25 min, A: 5% B: 50% C 45%; flow rate, 1.0 mL/min;
injection volumn, 20 μL; temperature, 30◦ C; wavelength, 268 nm. Peak: 1, hydrochlorothiazide; 2, vitamin B6; 3, bismuthrazine; 4, reserpine; 5,
vitamin B1; 6, promethazine. (B) Compound methoxyphenamine capsules. Column, SE–C18 /SO3 H (up); mobile phase, 50% ACN- 50% KH2 PO4
(50 mM, pH 3.0); injection volumn, 5 μL; temperature, 30◦ C; wavelength, 230 nm. Column, BaseLine C18 (bottom); mobile phase, 35% ACN- 65%
KH2 PO4 (50 mM, pH 3.0); others same as on SE–C18 /SO3 H. Peaks: 1, aminophylline; 2, noscapine; 3, methoxyphenamine; 4, chlorphenamine
ZHANG ET AL.
silica phase where vitamin B1, vitamin B6 and bismuthrazine
sulfate are co-eluted at dead time. The poor retentivity and
selectivity for the polar and ionic compounds make the
conventional C18 bonded silica phases less fit for analysis of
complex samples such as compound dosages.
3.4.2 Separation of compound
methoxyphenamine capsules
Compound methoxyphenamine capsules are widely used in
treatments of asthma and chronic obstructive pulmonary dis-
ease. The drug contains four active ingredients with three of
them being strong bases. Currently, methods for analysis are
largely based on ion-suppression reversed-phase chromatog-
raphy [6], which unfortunately is not compatible with online
MS detection. Comparative chromatograms of the compound
methoxyphenamine capsules separated on SE–C18 /SO3 H and
commercial BaseLine C18 phases are shown in Figure 6B.
Four components are baseline separated within 12 min with
symmetric peaks on SE–C18 /SO3 H. Baseline separation is
also achieved on BaseLine C18 though, but with much poorer
peak shape and severe peak tailing. More interestingly, the
elution order of compounds 2 and 3 is reversed when the
column used is changed from SE–C18 /SO3 H to BaseLine
C18. This is because the dissociation of the basic substance
increases its polarity, reducing its affinity for the station-
ary phase in single-mode reversed-phase chromatography. By
contrast, the increasing polarity of the solute is compensated
by stronger ionic interactions on the mixed-mode phase, thus
offering superior selectivity which is otherwise not available
on conventional reversed-phase columns.
4 CONC LU D I NG R E M A R K S
A mixed-mode reversed-phase/strong cation exchange sta-
tionary phase containing no aromatic rings has been success-
fully prepared from epoxysilane bonded silica. The phase con-
sisting of octadecyl and sulfonate groups bonded to silica, des-
ignated as SE–C18 /SO3 H, exhibits superior separation per-
formance for basic drugs, nucleobases, and basic alkylani-
lines in comparison with single-mode phases, SE–C18 , and
–SO3 H. This mixed-mode phase is ideally suited for simulta-
neous analysis of multiple components in FDC drugs owing
to its greater selectivity for compounds of diverse properties.
The retention on the SE–C18 /SO3 H phase can be regulated
by fine-tuning the composition of the mobile phase, i.e., ionic
and solvent strength, allowing optimization to be achieved in
two dimentianal space. In future study, we plan to investi-
gate the effects of the mole ratios of the functional groups
on the SE–C18 /SO3 H phase on retention. They are expected
to play an important role in causing cooperative binding but
have been poorly understood up to date.
2803
ACK NOW L E D G E M E N T
The authors would like to thank Dr. Lei Chen and Xiaohuan
Wang from Tianjin University for stimulating discussions.
CO N F L I C T O F I N T E R E ST
The authors have declared no conflict of interest.
O RC I D
Shuanghong Zhang
https://orcid.org/0000-0002-5224-7932
Qian-Hong Wan https://orcid.org/0000-0003-0359-6447
REFERENCES
1. Lai, J. M. L., Yang, S. L., Avoi, R., Treating more with less: effec-
tiveness and event outcomes of antituberculosis fixed-dose com-
bination drug versus separate-drug formulation (ethambutol, Iso-
niazid, rifampicin and pyrazinamide) for pulmonary tuberculosis
patients in real-world clinical practice. J. Glob. Insect. Dis. 2019,
11, 2–6.
2. Gupta, A. K., Arshad, S., Poulter, N. R., Compliance, safety, and
effectiveness of fixed-dose combinations of antihypertensive agents
a meta-analysis. Hypertension 2010, 55, 399–407.
3. Duarte, L. M., Amorim, T. L., Chellini, P. R., Adriano, L. H. C.,
de Oliveira, M. A. L., Sub-minute determination of rifampicin and
isoniazid in fixed dose combination tablets by capillary zone elec-
trophoresis with ultraviolet absorption detection. J. Sep. Sci. 2018,
41, 4533–4543.
4. Oliveira, M. A. L., Chellini, P. R., Amorim, T. L., Simultaneous
determination of rifampicin, isoniazid, pyrazinamide and ethambu-
tol in fixed dose combination antituberculosis pharmaceutical for-
mulations: a review. Anal. Meth. 2018, 10, 1103–1116.
5. Kozlov, V. V., Danilov, L. L., Reversed-phase ion-pair high-
performance liquid chromatography assay of polyprenyl diphos-
phate oligomer homologues. J. Sep. Sci. 2016, 39, 525–527.
6. Yin, C. H., Tang, C., Wu, X. Y., HPLC determination of amino-
phylline, methoxyphenamine hydrochloride, noscapine and chlor-
phenamine maleate in compound dosage forms with an aqueous-
organic mobile phase. J. Pharm. Biomed. Anal. 2003, 33, 39–43.
7. Zhang, Y. J., Li, Q. Q., Dai, B. L., Zhou, S. Y., Zhang, M. L., Han,
H. F., Qiu, H. D., A versatile polar-embedded polyphenyl phase for
multimodal separation in liquid chromatography. J. Chromatogr. A
2018, 1553, 81–89.
8. Wu, Q., Sun, Y. M., Zhang, X. L., Zhang, X., Dong, S. Q., Qiu,
H. D., Wang, L. T., Zhao, L., Multi-mode application of graphene
quantum dots bonded silica stationary phase for high performance
liquid chromatography. J. Chromatogr. A 2017, 1492, 61–69.
9. Yang, B. B., Liu, H. M., Chen, J., Guan, M., Qiu, H. D., Preparation
and evaluation of 2-methylimidazolium-functionalized silica as a
mixed-mode stationary phase for hydrophilic interaction and anion-
exchange chromatography. J. Chromatogr. A 2016, 1468, 79–85.
10. Călinescu, O., Badea, I. A., Vlădescu, L., Meltzer, V., Pincu, E.,
HPLC separation of acetaminophen and its impurities using a
mixed-mode reversed-phase/cation exchange stationary phase. J.
Chromatogr. Sci. 2012, 50, 335–342.
2804
11. Li, X. L., Guo, Z. M., Sheng, Q. Y., Xue, X. Y., Liang, X. M.,
Sequential elution of multiply and singly phosphorylated peptides
with polar-copolymerized mixed-mode RP18/SCX material. Ana-
lyst 2012, 137, 2774–2776.
12. Long, Z., Guo, Z. M., Xue, X. Y., Zhang, X. L., Nordahl, L., Liang,
X. M., Selective separation and purification of highly polar basic
compounds using a silica-based strong cation exchange stationary
phase. Anal. Chim. Acta 2013, 804, 304–312.
13. Song, Q. G., Zhao, K. L., Xia, H. J., Liu, J. W., Bai, Q., A novel
reversed-phase and weak anion-exchange mixed-mode stationary
phase based on horizontal polar-copolymerized approach for sep-
aration of small organic molecules and inorganic anions. Talanta
2019, 197, 592–598.
14. Xia, H. F., Don, B. B., Zheng, M. J., Mixed-mode chromatogra-
phy integrated with high-performance liquid chromatography for
protein analysis and separation: Using bovine serum albumin and
lysozyme as the model target. J. Sep. Sci. 2016, 39, 1900–1907.
15. Wu, Q. C., Zhang, Q. L., Gao, D., Nie, L., Wang, H. B., Yao, S. J.,
Lin, D. Q., Mixed-mode expanded-bed adsorption for human serum
albumin separation. Ind. Eng. Chem. Res. 2018, 57, 1039–1047.
16. Liu, H. F., Sheng, M. S., Pan, Z. J., Qiu, D. J., Guo, Z. H., Gan, F. X.,
Synthesis of P(St-BA-MAA-SPAA) nano-latex and its application
in high anti-corrisive coating. J. Cent. South Univ. Technol. 2008,
15, 54–58.
17. Tang, X. Y., Li, X. X., Sun, Y. W., Xiao, Y., Wang, Y.,
Thiol-ene click derived structurally well-defined per(3,5-
dimethyl)phenylcarbamoylated cationic cyclodextrin separation
material for achiral and chiral chromatography. J. Sep. Sci. 2018,
41, 2710–2718.
18. Ren, X. J., Zhang, K. L., Gao, D., Fu, Q. F., Zeng, J., Zhou, D.,
Wang, L. J., Xia, Z. N., Mixed-mode liquid chromatography with a
stationary phase co-functionalized with ionic liquid embedded C18
and an aryl sulfonate group. J. Chromatogr. A 2018, 1564, 137–144.
ZHANG ET AL.
19. Bocian, S., Krzemińska, K., Buszewski, B., A study of separation
selectivity using embedded ester-bonded stationary phases for liq-
uid chromatography. Analyst 2016, 141, 4340–4348.
20. Berendsen, G. E., Galan, L. D., Preparation and chromatographic
properties of some chemically bonded phases for reversed-phase
liquid chromatography. J. Liq. Chromatogr. 1978, 1, 561–586.
21. Kalíková, K., Kozlík, P., Gilar, M., Tesařová, E., Properties of
two amide-based hydrophilic interaction liquid chromatography
columns. J. Sep. Sci. 2013, 36, 2421–2429.
22. Yang, X. Q., Dai, J., Carr, P. W., Analysis and critical comparison of
the reversed-phase and ion-exchange contributions to retention on
polybutadiene coated zirconia and octadecyl silane bonded silica
phases. J. Chromatogr. A 2003, 996, 13–31.
23. Tian, J., Che, B. Q., Tian, J., HPLC determination for the content
uniformity of the six components in compound reserpine tablets.
Chin. J. Pharm. Anal. 2018, 38, 1634–1640.
S U P P O RT I NG IN FO R M AT I O N
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
How to cite this article: Zhang S, Wan Q-H, Li
Y. Epoxide-derived mixed-mode chromatographic
stationary phases for separation of active substances
in fixed-dose combination drugs. J Sep Sci 2019;
42:2796–2804. https://doi.org/10.1002/jssc.201900307