Determination of the enantiomeric purity of S-ropivacaine by capillary electrophoresis with methyl-β-cyclodextrin as chiral selector using conventional and complete filling techniques

Article abstract of DOI:10.1016/S0928-0987(99)00035-4

Capillary electrophoresis (CE) methods based on the conventional and complete filling techniques for determination of the enantiomeric purity of S-ropivacaine are described. The complete filling technique is a separation method which can be used instead of the partial filling technique in order to reduce the total analysis time, when the chiral selector solution does not absorb UV light. In the complete filling technique the total length of the capillary is filled with the chiral selector solution, prior to application of the analyte. During the run both ends of the capillary are connected to the background electrolyte, i.e. without chiral agent. An interlaboratory study was performed to validate the method. The limit of detection and quantification for R-ropivacaine were found to be about 0.6 and 1.6 μg/ml, respectively, corresponding to 0.1 and 0.25% enantiomeric purity of S-ropivacaine. Good performances were demonstrated for the repeatability and linearity. The consumption of the chiral selector was about 160 times lower with the complete filling technique compared with the conventional CE technique. Copyright (C) 1999 Elsevier Science B.V.

Full text of DOI:10.1016/S0928-0987(99)00035-4

European Journal of Pharmaceutical Sciences 9 (1999) 17–24
www.elsevier.nl/locate/ejps
Determination of the enantiomeric purity of S-ropivacaine by capillary
electrophoresis with methyl-b-cyclodextrin as chiral selector using
conventional and complete filling techniques
a,b ,
a
b
a
*
Ahmad Amini
, Britta Wiersma , Douglas Westerlund , Ulla Paulsen-S o¨ rman
a
Medical Products Agency, Division of Chemistry and Pharmacy, Box 26, S-751 03 Uppsala, Sweden
Analytical Pharmaceutical Chemistry, Uppsala University Biomedical Centre, Box 574, S-751 23 Uppsala, Sweden
b
Received 29 January 1999; accepted 10 May 1999
Abstract
Capillary electrophoresis (CE) methods based on the conventional and complete filling techniques for determination of the
enantiomeric purity of S-ropivacaine are described. The complete filling technique is a separation method which can be used instead of the
partial filling technique in order to reduce the total analysis time, when the chiral selector solution does not absorb UV light. In the
complete filling technique the total length of the capillary is filled with the chiral selector solution, prior to application of the analyte.
During the run both ends of the capillary are connected to the background electrolyte, i.e. without chiral agent. An interlaboratory study
was performed to validate the method. The limit of detection and quantification for R-ropivacaine were found to be about 0.6 and 1.6
mg/ml, respectively, corresponding to 0.1 and 0.25% enantiomeric purity of S-ropivacaine. Good performances were demonstrated for the
repeatability and linearity. The consumption of the chiral selector was about 160 times lower with the complete filling technique
compared with the conventional CE technique.
 1999 Elsevier Science B .V . All rights reserved.
Keywords: Quantitative analysis; Enantioseparation; Capillary zone electrophoresis; Local anaesthetic drugs; Conventional technique; Complete filling
technique
1
. Introduction
background electrolyte. There are a wide variety of chiral
selectors including cyclodextrins (Fanali, 1989; Chan-
kvetadze, 1997) which can be employed for enantioresolu-
tion in CE. CE can be the preferred or complementary
separation method of analysis in routine application areas
such as chiral purity determination and quantitative analy-
sis, where CE may offer considerable benefits in term of
reduced time and costs (Altria, 1993). The partial filling
technique (PFT) is a new separation technique which has
been employed for different analytical tasks (Valtcheva et
al., 1993; Tanaka and Terabe, 1995; Amini and Wester-
lund, 1997). In PFT the chiral selector solution is intro-
duced, up to the detection window, in the capillary prior to
application of the analyte. During the run both ends of the
capillary are dipped into the background electrolyte
(BGE), i.e. without chiral selector. Originally this tech-
nique was developed to avoid UV detection interference
between the analyte and the UV absorbing chiral selector
(Valtcheva et al., 1993). PFT can also be used when the
chiral selector does not absorb UV (Amini and Paulsen-
S o¨ rman, 1997), in order to decrease the consumption of
the chiral selector, which is important when using an
Enantioseparation for the quantitative analysis of drugs
and trace impurities is essential in pharmaceutical analysis
and quality control and required for registration of new
drug products by the regulatory authorities. There are
many examples where enantiomers of a racemic drug give
different pharmacological and toxicological effects in
biological systems (Ariens, 1989).
High demands on the documentation of pharmaceutical
quality have led to an increased need of developing
accurate, reliable and convenient analytical methods for
measuring enantiomeric purity. Most chiral separation
methods are based on chromatography. Capillary electro-
phoresis (CE) offers greater separation efficiency than that
of gas chromatography (GC) and high performance liquid
chromatography (HPLC) (Allenmark, 1992). Direct en-
antioseparation with CE can easily be performed by adding
an optically active agent, i.e. a chiral selector, to the
*
Corresponding author. Present address: Department of Chemistry,
Purdue University, Lafayette, IN 47907, USA.
0
928-0987/99/$ – see front matter
 1999 Elsevier Science B .V . All rights reserved.
PII: S0928-0987(99)00035-4

1
8
A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
expensive chiral additive, as well as to avoid contamina-
tion of the ion-source employing capillary electrophoresis
coupled to mass spectrometry (Nelson et al., 1996; J a¨ ver-
falk et al., 1998). When a chiral selector has a low UV
absorptivity, the capillary can be filled with the selector
solution prior to application of the solute, the complete
filling technique (CFT), in order to decrease the capillary
preconditioning time compared to that for PFT. In PFT the
chiral selector solution is introduced into the capillary by
pressure (3.4 kPa) which takes about 15 min to fill a
capillary with 40 cm effective length and 50 mm inner
diameter. In CFT the capillary is flushed (pressure ¯100
kPa) with the selector solution, taking about 25 s to fill the
total length of the capillary. Generally, the filling time is
less than 2 min including rinsing and conditioning with the
selector solution. The most important property of CFT,
compared to the conventional technique, for routine ana-
lyse is reduced expenses.
long-acting local anaesthetic drug of the amide-type main-
ly used for surgery and post-operative pain relief.
CFT has also been compared, from the point of view of
chiral selector solution consumption and precision, with a
conventional CE technique under the same conditions. The
three techniques, PFT, CFT and conventional, are de-
scribed in Fig. 1. The reproducibility of the method was
determined by an interlaboratory study with the parame-
ters: analysts, instruments, capillaries from different bat-
ches, coated and uncoated capillaries, and days.
2. Materials and methods
2.1. Equipment
The capillary zone electrophoresis experiments were
conducted by use of a Beckman P/ACE System 2100
(Beckman Instruments Inc., Palo Alto, CA, USA) equipped
with a 47 cm (40 cm effective length)350 mm I.D. (363
O.D.) fused silica capillary (Polymicro Technologies,
This paper is aimed at giving an insight into the
possibility of employing the CFT for quantitative analysis.
A CE method based on CFT with methyl-b-CD as chiral
selector for quantitative analysis of the enantiomeric purity
of S-ropivacaine was developed and validated.
Ropivacaine, on the market as the S-enantiomer, is a new
3
D
Phoenix, AZ, USA) and a Hewlett Packard Capillary
Electrophoresis system (Waldbronn, Germany) comprising
a diode-array detector and a Chem Station for system
control, data collection and data analysis. It was equipped
with a 48.5 cm (40 cm effective length)350 mm I.D. either
uncoated fused silica (Polymicro Technologies) or poly-
acrylamide coated capillary (Hjert e´ n, 1985). Detections
were carried out by UV at 214 nm. The capillaries were
thermostated at 208C. Separations were performed at 25
3
D
kV (Beckman) and 30 kV (HP ) towards the cathodic
electrode.
2
.2. Chemicals
Methyl-b-CD (met-b-CD), degree of substitution 10.14–
1
4.17, was purchased from KEBO-Lab (Stockholm,
Sweden). R- and S-ropivacaine, S-prilocaine and R-
mepivacaine were kindly supplied by Astra Pain Control
AB (S o¨ dert a¨ lje, Sweden). The solute structures are given
in Fig. 2.
Fig. 1. Comparison between the separation techniques. (a) PFT (partial
filling technique) where less than the effective length of the capillary is
filled with the chiral selector solution. (b) CFT (complete filling tech-
nique) where the whole length of the capillary is filled with the chiral
selector solution. In PFT and CFT both ends of the capillary are
immersed in plain buffer solution during the run. (c) Conventional CE,
the capillary is filled with the chiral selector solution, and prior to the run
and during the run both ends of the capillary are dipped in the chiral
selector containing BGE, i.e. the same buffer as the capillary is flushed
with.
Fig. 2. Structures of the solutes.

A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
19
Background electrolyte (BGE) was made by mixing
9.8 ml of 1.0 M NaOH, 32.2 ml of 1.0 M phosphoric acid
and deionized water to 500 ml; the ionic strength (I) was
2.5. Calculation of separation parameters
1
The peak resolution of a racemate is given by the
0
.04 and the pH was 3.0.
resolution factor (R ) calculated from the equation
s
In purity determinations, R-ropivacaine was spiked at
different concentrations, in the range of 0.6 to 62.5 mg/ml,
into a BGE, pH 3.0 and I 5 0.004, containing S-ropivacine
R 5 1.18(t
2 tmig1)/(w(1)b1 / 2 1 w(2)b1 / 2
mig2
)
(1)
s
where tmig2 and tmig1 are the migration times of the
enantiomers and w(1)b1 / 2 and w(2)b1 / 2 are the widths of the
peaks at half height. The selectivity factor (a) is calculated
from
6
25 mg/ml. The ratio [R]/[S]% was used to give the
degree of enantiopurity. S-Prilocaine or R-mepivacaine,
when used as internal standard, was spiked into the sample
solution at a concentration of 10 or 20 mg/ml. The racemic
solution of ropivacaine was prepared by mixing equal
amounts of S-ropivacaine and R-ropivacaine to a final
concentration of 10 mg/ml in the BGE diluted 10 times.
N,N-Dimethyl formamide (DMF) for UV spectroscopy
a 5 tmig2 /tmig1 5 mobs1 /mobs2
(2)
The number of theoretical plates was evaluated from the
expression
(
Fluka BioCmica, Buchs, Switzerland) was used for
2
N 5 5.54(tmig1 /wb1 / 2
)
(3)
determination of the electroosmotic mobility.
All other chemicals utilized in the investigation were of
analytical grade. All solutions were passed through a
membrane filter, Minisart N, 0.45 mm (Sartorius, Goetting-
en, Germany) prior to use. Deionized water was used in all
experiments.
RMD, i.e. the relative mobility difference of the enantio-
mers, as a measure of the enantioselectivity, was evaluated
from (Jorgenson and Lukacs, 1981)
RMD 5 Dmobs /mav(obs)
(4)
where Dmobs and mav(obs) are the differences in the ob-
served mobilities of the enantiomers (mobs(1) 2 mobs(2)) and
the average of the mobilities (mobs(1) 1 mobs(2))/2, respec-
tively.
2
.3. Capillary conditioning
A new fused silica capillary was rinsed with 0.1 M HCl
for 5 min, with water for 5 min, then with 0.1 M NaOH for
min and finally again with water for 10 min, before the
5
2
.6. Quantitative analysis
BGE was introduced into the capillary. The uncoated
capillaries were flushed with 0.1 M NaOH for 2 min before
every run. The capillaries were stored in water over night
and the uncoated capillaries were rinsed with 0.1 M NaOH
and water every morning. The polyacrylamide coated
capillary was only washed with water and BGE.
R-Ropivacaine was spiked at different concentrations
1.6–62.5 mg/ml) with the S-enantiomer (625 mg/ml).
Quantification of R-ropivacaine as impurity was carried out
by peak area measurements.
In order to compensate for the difference in residence
time in the detector, the peak area was corrected by its
migration time (Area/tmig) (Altria et al., 1994), and to
construct the calibration curve the ratio between the
corrected areas of the solute and internal standard was
plotted versus the ([R-ropivacaine]/[S-ropivacaine]%)
ratio.
(
2
.4. Procedure
Separation solutions were prepared by dissolving met-b-
CD (76 mM) as the chiral selector in the BGE. The
uncoated capillaries were conditioned according to the
following procedure before every run: water for 2 min; 0.1
M NaOH for 2 min; water for 2 min; BGE for 2 min; and
finally with the chiral selector solution for 2 min prior to
application of the analyte.
The solute concentrations giving signal-to-noise ratios of
and 10 were selected to be the limit of detection (LOD)
and the limit of quantification (LOQ), respectively.
3
The coated capillary was flushed between runs first by
water, 3 min, and then by BGE, 2 min, before application
of the selector solution. Both ends of the capillary were
conducted to the BGE and a constant voltage of either 25
kV (Beckman) or 30 kV (HP ) was applied over the
capillary.
3. Results and discussion
3.1. Enantioseparation of racemic ropivacaine by
different CE techniques
3
D
In conventional CE, 76 mM of the chiral selector
dissolved in the BGE was used as running buffer, i.e. both
ends of the capillary were immersed in the BGE containing
the chiral selector. The analyte sample was introduced into
the capillary by pressure (3.4 kPa) for 5 s.
Conventional CE methods for the enantioresolution of
ropivacaine have been reported (S a¨ nger-van de Griend et
al., 1996; Amini et al., 1997).
The enantioseparation of racemic ropivacaine was in-
vestigated using the BGE, pH 3.0 and I 5 0.04, containing

2
0
A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
capillary are dipped into the running buffer containing the
chiral selector. In the CFT mode the total length of the
capillary was filled with the selector prior to application of
the solutes; separation of the ropivacaine enantiomers is
illustrated in Fig. 3. During the run both ends of the
capillary were immersed in the buffer solution without
chiral selector (BGE). Table 1 compares the separation
systems with respect to migration time, efficiency, res-
olution, selectivity factor and selector consumption. First it
can be stated that CFT consumes about 160 times less
chiral selector solution than the conventional technique.
Flushing and filling the capillary with the chiral selector
solution for 2.0 min consumed about 5 ml of the solution.
The resolution factors obtained using conventional CE
and CFT are larger than 3.8, which is adequate for
quantitative analysis of impurities (Doyle, 1991). The
enantioresolution is higher in the coated capillary, owing
2
5
2
to the lower electroosmotic flow (m ¯ 0.7 3 10 cm /
eo
V s). The enantioresolution increases when the electro-
osmotic flow decreases (Jorgenson and Lukacs, 1981):
Fig. 3. Enantioseparation of racemic ropivacaine using the complete
filling technique with methyl-b-CD as chiral selector. BGE: phosphate
buffer, pH 3.0 and I 5 0.04. Met-b-CD as chiral selector was dissolved at
a concentration of 76 mM in the BGE. Prior to application of the analyte
sample the capillary was flushed and completely filled with the chiral
selector solution for 2.0 min. Enantioseparation was conducted using a
Hewlett-Packard system equipped with a 48.5 cm (40 cm effective
length)350 mm I.D. capillary thermostated at 208C; detection, 214 nm;
applied voltage, 30 kV; current, ¯43 mA. Injection technique: pressure
1
/ 2
R 5 N Dm(obs) /[4(m(obs)av)]
s
1
/ 2
5
N
Dv(obs) /[4(v(obs)av)]
(5)
For definition of m-terms, see Eq. (4); N is the number of
theoretical plates; Dv(obs) and v(obs)av are the differences in
the observed migration velocity of the enantiomers and
average observed velocity of the enantiomers, respectively.
The higher resolution obtained with the Beckman instru-
(
3.4 kPa for 5 s). [racemic ropivacaine]510 mg/ml. Peak 15R-
ropivacaine; peak 25S-ropivacaine.
3
D
ment compared to HP when using CFT depends on the
3
D
applied voltage, which was higher in HP , 30 kV (25 kV
in Beckman). A higher electrical field produces a higher
migration velocity, i.e. a higher v(obs)av, which results in
lower enantioresolution, see Eq. (5). It should be noted
that the electroosmotic mobility in both systems is nearly
7
6 mM met-b-CD, employing different separation tech-
niques, i.e. the conventional technique and CFT, in order
to compare resolution, efficiencies, reproducibility and
consumption of the chiral selector solution.
2
4
2
In the conventional CE technique both ends of the
the same, i.e. ¯0.42310
cm /V s. When the heat
Table 1
Comparison of separation parameters obtained from enantioseparation of racemic ropivacaine using different separation techniques and instruments (n 5 6).
Conditions: as those in Figs. 2–4
Separation system
N(R) /am
CI
N(S) /m
6CI
Rs
6CI
a
tmig(S)
6CI
tmig(R)
6CI
ttot /run
(min)
Selector
consumption
per run (ml)
6
RSD%
b
CZE
(
452 309
641 287
482 098
625 795
470 932
658 304
531 014
628 971
606 022
631 156
4.01
60.33
12.83
60.32
13.43
61.04
9.11
60.13
11.57
60.18
11
21
22
18
22
[met-b-CD]50)c
Conventional tech.
[met-b-CD]576 mM)
Complete filling tech.
[met-b-CD]576 mM)
Complete filling tech.
[met-b-CD]576 mM)
Complete filling tech.
474 824
628 397
414 028
643 005
468 336
629 483
590 493
639 651
1 / 2
4.25
60.10
4.20
60.19
3.86
60.11
5.15
1.04
0.1
1.04
0.2
1.03
0.1
1.04
0.1
13.35
60.35
13.99
61.04
9.42
0.32
12.06
60.20
800
5
(
c
(
b
5
(
d
5
(
[met-b-CD]576 mM)
60.14
a
The 95% confidence interval (5mean62.57(SD/n ).
Hewlett-Packard (HP ).
Beckman.
b
c
d
3D
3D
3D
24
2
25
2
HP coated capillary (electroosmotic flow in both the Beckman and HP was 0.42310 cm /V s using fused silica capillaries and 0.7310 cm /V s
in the coated capillary).

A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
21
dissipation is insufficient, application of a higher voltage
can produce Joule heating and thereby impair peak res-
olution. However, in our case the lower enantioresolution
areas. The deviation in the peak area of R-ropivacaine is
smaller when it is corrected with the area of the IS.
The effective mobility (m ) of the enantiomers is the
eff
3
D
with the HP
instrument does not depend on zone
sum of two different mobilities (Guttman et al., 1988):
broadening (Table 1), since higher efficiencies were ob-
tained with this instrument.
meff 5 h[rop]/([rop] 1 [rop–CD])jmrop
The differences in migration times of the enantiomers
1
h[rop–CD]/([rop] 1 [rop–CD])jmCD–rop
(6)
3
D
obtained with the Beckman and HP
instruments are
significant at the 95% confidence interval due to the
difference in the applied voltages. The differences in the
migration times and resolution factors between CFT and
the conventional technique under the same conditions were
not significant, however. The variation in separation
performance was greater for the Beckman instrument,
employing CFT for separation of racemic ropivacaine; the
reason for this is not known, but may be due to the
condition of the capillary. There was a tendency towards
higher efficiencies with conventional CE and CFT with the
where [rop] and [rop–CD] are equilibrium concentrations
of the free and complexed ropivacaine, respectively, and
mrop and mCD–rop are the corresponding electrophoretic
mobilities. The selector phase (76 mM) is overloaded by
S-ropivacaine at very high concentration (625 mg/ml ¯2
mM), resulting in a higher free fraction of the enantiomer
and higher effective mobility (shorter migration time),
since mrop . mCD–rop, compare Tables 1 and 2. The availa-
bility of the selector for the other solutes, i.e. R-
ropivacaine and IS, will decrease due to competition with
S-ropivacaine, and their effective mobilities will also
increase. The RMD of the ropivacaine enantiomers de-
creased from 0.033 to 0.008 with increasing concentration
of S-ropivacaine.
3
D
HP
instrument (higher applied voltage) compared to
CFT with the Beckman, but the difference was not
significant. The efficiency was higher in the coated than in
the uncoated capillary owing to the longer migration time
in addition to the smaller electrostatic interaction between
the solute and capillary wall in the coated capillary,
resulting in less zone broadening.
Quantitative determination of the enantiomeric purity of
S-ropivacaine was evaluated by an interlaboratory study
with CFT and conventional CE. R-Ropivacaine was quan-
tified with met-b-CD, as chiral selector, dissolved at a
concentration of 76 mM in the BGE.
3
.2. Enantiopurity analysis of S-ropivacaine
In order to construct calibration curves, R-ropivacaine
(
1.6 to 62.5 mg/ml) was spiked with S-ropivacaine (625
To investigate the repeatability of the migration times
and peak areas the analyte sample, consisting of R-
ropivacaine, S-ropivacaine (625 mg/ml), S-prilocaine (IS,
0 mg/ml), was injected 45 times into a system based on
CFT (Table 2). The concentration of R-ropivacaine was
.1 mg/ml corresponding to 0.5% of the S-enantiomer. The
mg/ml), using S-prilocaine (10 mg/ml) as the internal
standard (IS). The corrected peak areas of R-ropivacaine
and the IS were calculated and the ratio between the areas
was analysed by linear regression. All calibration curves
showed good linearity in the concentration range 1.6–62.5
mg/ml; the numerical data for the linear regression analy-
sis are given in Table 3. There are no significant differ-
ences in the slope of the plots determined by different
methods and instruments and the intercepts were not
significantly different from zero. The lower slope of the
plot obtained with the coated capillary is due to the use of
a different IS: R-mepivacaine (20 mg/ml). The analysis of
S-prilocaine as IS in the coated capillary produced peaks
with areas difficult to determine precisely, resulting in
unreliable calibration curves with low regression factors.
The reason for the fluctuating peak area is not known.
R-Mepivacaine, owing to its migration time (shorter than
that for R-ropivacaine), was shown to be a better IS with
this capillary (Fig. 4). The tailing peak shape of S-
ropivacaine gave a high resolution from the faster migrat-
ing R-enantiomer. The inter- and intra-day precision was
shown to be adequate (Table 4); at the 0.3% impurity level
the reproducibility was .10%.
2
3
results show good precision for migration times and peak
Table 2
Precision data obtained from 45 replicate injections of a solution of
R-ropivacaine (3.1 mg/ml), S-ropivacaine (625 mg/ml) and S-prilocaine
20 mg/ml) as internal standard, employing CFT. Separation conditions:
(
met-b-CD (76 mM) in the BGE, i.e. phosphate buffer pH 3.0 and
3
D
I 5 0.04. Enantioseparation was conducted using a Hewlett-Packard
system equipped with a 48.5 (40 cm effective length)350 mm I.D. fused
silica capillary thermostated at 208C; detection, 214 nm; applied voltage,
0 kV. The capillary was flushed and filled with the chiral solution prior
to application of the analyte sample. During the run both ends of the
capillary were dipped in the BGE. Injection technique: pressure (3.4 kPa
for 5 s)
3
Analysis parameter
Mean value 6RSD%
tmig (R-ropivacaine) 3.1 mg/ml
tmig (S-ropivacaine) 625.0 mg/ml
tmig (S-prilocaine) as IS, 20 mg/ml
Area (R-ropivacaine)
Corrected area (R-ropivacaine) (A1)
Area (S-prilocaine)
8.2261.2 min
8.2961.2 min
9.1061.3 min
3.762611.0
0.45369.9
The LOD was found to be 0.6 mg/ml, corresponding to
a 0.1% impurity level. The lowest concentration of R-
ropivacaine which could be determined with acceptable
precision was 1.6 mg/ml (0.25%) (Fig. 5).
32.27265.0
3.55564.7
Corrected area (S-prilocaine) (A2)
Corrected area ratio A /A
0.12767.6
1
2

2
2
A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
Table 3
Summary of calibration curve data for ropivacaine, using S-prilocaine as internal standard (n 5 6–8). Conditions: as those given in Figs. 2–4
Concentration range
Separation
technique
Slope
6SD
Intercept
6SD
Regression
factor (r )
2
([R-ropivacaine]/
S-ropivacaine])%
[
a
0
0
0
0
0
0
.3–10.0%
.3–10.0%
.3–10.0%
.3–10.0%
.3–10.0%
.3–10.0%
a
Intra-day
Inter-day
Intra-day
Intra-day
Inter-day
Intra-day
CFT
CFT
Conv.
CFT
CFT
CFT
0.32160.010
0.31360.013
0.32060.008
0.31660.009
0.31460.012
0.21760.006
20.01660.010
20.02760.012
20.001560.03
0.01860.011
0.02260.030
20.00360.010
0.9988
0.9980
0.9986
0.9990
0.9985
0.9962
a
b
d
b
b
c
3D
3D
Hewlett-Packard (HP ).
b
c
d
Beckman.
3D
HP coated capillary (IS: R-mepivacaine).
The conventional CE technique.
As demonstrated, the complete filling technique could
Consumption of the chiral selector solution was only 5
ml (¯500 mg met-b-CD) per run compared to 800 ml
(¯80 mg met-b-CD) with the conventional technique.
CFT is the preferred technique compared with the partial
successfully be employed for quantitative analysis. A great
advantage is that CFT is less expensive compared with the
conventional CE technique.
Fig. 4. Analysis of a S-ropavaine solution (625 mg/ml) containing 0.3% R-ropivacaine (1.9 mg/ml) and R-mepivacaine (20 mg/ml) as internal standard.
BGE: phosphate buffer, pH 3.0 and I 5 0.04. Met-b-CD as chiral selector was dissolved at a concentration of 76 mM in the BGE. Prior to application of
the analyte sample the capillary was flushed and completely filled with the chiral selector solution for 2.0 min. The determination was conducted using a
Hewlett-Packard system equipped with a 48.5 cm (40 cm effective length)350 mm I.D. polyacrylamide coated capillary thermostated at 208C; detection,
2
14 nm; applied voltage, 30 kV; current, ¯36 mA. Injection technique: pressure (3.4 kPa for 5 s). Peak 15R-mepivacaine (IS); peak 25R-ropivacaine;
peak 35S-ropivacaine.

A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
23
Table 4
Reproducibility of the peak area of R-ropivacaine at different concentrations in the presence of the other enantiomer at a concentration of 625 mg/ml
n 5 6–8). Conditions: as those given in Table 2
(
a,b
c
Concentration
[R-ropivacaine]/
[S-ropivacaine]
(%)
Area(ratio) 6RSD%
Area(ratio) 6RSD%
Intra-day
av
av
[R-ropivacaine]
mg/ml)
(
Intra-day
Inter-day
1
.9
0.3
1.6
3.2
6.4
10.2
0.12168.5
0.45865.6
0.96364.3
1.85462.8
3.10661.7
0.12269.6
0.45466.1
0.99964.9
1.85963.2
3.21961.9
0.04768.6
0.22468.2
0.55064.0
1.26863.1
2.15861.8
1
2
4
6
0.0
0.0
0.0
4.0
a
Area5[Area(solute) /tmig(solute)]/[Area(IS) /tmig(IS)].
b
c
The separations were performed in fused silica capillaries with S-prilocaine (10 mg/ml) as IS.
The separations were performed in a polyacrylamide coated capillary with R-mepivacaine (20 mg/ml) as IS.
filling technique, giving a shorter preconditioning time
which reduces the total analysis time. The prerequisite
condition is the use of a transparent chiral selector.
References
Allenmark, S.G., 1992. Chromatographic Enantioseparation: Methods and
Applications, 2nd ed., Ellis Horwood, Chichester.
Altria, K.D., 1993. Quantitative aspects of the application of capillary
electrophoresis to the analysis of pharmaceuticals and drug related
impurities. J. Chromatogr. 646, 245–257.
Altria, K.D., Goodall, D.M., Rogan, M.M., 1994. Quantitative applica-
tions and validation of the resolution of enantiomers by capillary
electrophoresis. Electrophoresis 15, 824–827.
Amini, A., Paulsen-S o¨ rman, U., 1997. Enantioseparation of local anaes-
thetic drugs by capillary zone electrophoresis with cyclodextrins as
chiral selectors using a partial filling technique. Electrophoresis 18,
1
019–1025.
Amini, A., Paulsen-S o¨ rman, U., Lindgren, B.H., Westerlund, D., 1997.
Enantioseparation of anaesthetic drugs by capillary zone electro-
phoresis using cyclodextrin-containing background electrolytes. Elec-
trophoresis 19, 131–137.
Amini, A., Westerlund, D., 1997. Evaluation of association constants
between drug enantiomers and human a -acid glycoprotein by apply-
1
ing a partial-filling technique in affinity capillary electrophoresis.
Anal. Chem. 70, 1425–1430.
Ariens, E.J., 1989. In: Krstulovic, A.M. (Ed.), Chiral Separations by
HPLC, Ellis Horwood, Chichester, p. 31.
Chankvetadze, B., 1997. Capillary Electrophoresis in Chiral Analysis,
Wiley, New York.
Doyle, T.D., 1991. In: Ahuja, S. (Ed.), Chiral Separations by Liquid
Chromatography, American Chemical Society, Washington, DC, p. 27.
Fanali, S., 1989. Separation of optical isomers by capillary zone electro-
phoresis based on host–guest complexation with cyclodextrins. J.
Chromatogr. 474, 441–446.
Fig. 5. Detection of 0.25% R-ropivacaine in the presence of S-
ropivacaine. The pure S-enantiomer, 625 mg/ml, was spiked with 1.6
mg/ml R-enantiomer (0.25%). BGE: phosphate buffer, pH 3.0 and I 5
Guttman, A., Paulus, A., Cohen, A.S., Grinberg, N., Krager, B.L., 1988.
Use of complexing agents for selective separation in high-performance
capillary electrophoresis. Chiral resolution via cyclodextrins incorpo-
rated within polyacrylamide gel columns. J. Chromatogr. A 448,
41–53.
Hjert e´ n, S., 1985. High performance capillary electrophoresis, elimination
of electroosmosis and solute adsorption. J. Chromatogr. 347, 191–198.
Jorgenson, J.W., Lukacs, K.D., 1981. Zone electrophoresis in open-tubular
glass capillaries. Anal. Chem. 53, 1298–1302.
0
.04. Met-b-CD as chiral selector was dissolved at a concentration of 76
mM in the BGE. Prior to application of the analyte sample the capillary
was flushed and completely filled with the chiral selector solution for 2.0
min. The determination was conducted using a Beckman system equipped
with a 47 cm (40 cm effective length)350 mm I.D. capillary thermostated
at 208C; detection, 214 nm; applied voltage, 25 kV; current, ¯31 mA.
Injection technique: pressure (3.4 kPa for 5 s). Peak 15R-ropivacaine;
peak 25S-ropivacaine; peak 35S-prilocaine (IS).
J a¨ verfalk, E., Amini, A., Westerlund, D., Andr e´ n, P.E., 1998. Chiral
separation of local anaesthetics by a capillary electrophoresis/partial

2
4
A. Amini et al. / European Journal of Pharmaceutical Sciences 9 (1999) 17–24
filling technique coupled on-line to micro-electrospray mass spec-
trometry. J. Mass Spectrom. 33, 183–186.
Nelson, W.M., Tang, Q., Harrata, A.K., Lee, C.S., 1996. On-line partial
filling micellar electrokinetic chromatography–electrospray ionization
mass spectroscopy. J. Chromatogr. 749, 219–226.
S a¨ nger-van de Griend, C.E., Gr o¨ ningsson, K., Westerlund, D., 1996.
Chiral separation of anaesthetics with capillary electrophoresis.
Evaluation of the inclusion complex of the enantiomers with heptakis
Tanaka, Y., Terabe, S., 1995. Partial separation zone technique for the
separation of enantiomers by affinity electrokinetic chromatography
with proteins as chiral pseudo-stationary phases. J. Chromatogr. 694,
277–284.
Valtcheva, L., Mohammad, J., Pettersson, G., Hjert e´ n, S., 1993. Chiral
separation of b-blockers by high-performance capillary electrophoresis
based on non-immobilized cellulase as enantioselective protein. J.
Chromatogr. 638, 263–267.
(
2,6-di-O-methyl)-b-cyclodextrin. Chromatographia 42, 263–268.