Separation and determination of ephedrine and pseudoephedrine in Ephedrae Herba by CZE modified with a Cu(II)–L-lysine complex

Fei LiZhongtao DingQiu-E Cao

Key Laboratory ofMedicinal Chemistryfor Nature Resource,Ministry of Education,School of Chemical Scienceand Technology,Yunnan University,Yunnan, P. R. China

Received July 15, 2007Revised September 3, 2007Accepted September 17, 2007

Research Article

Separation and determination of ephedrineand pseudoephedrine in Ephedrae Herba byCZE modified with a Cu(II)–L-lysine complex

A CZE method using a complex of 2.5 mM Cu(II)–L-lysine (molar ratio is 1:2) as additive ina run buffer solution composed of Tris-H3PO4 (pH 4.5) was developed for the simultaneousdetermination of ephedrine and pseudoephedrine within 4 min. The effects of pH, com-position, and concentration of run buffer as well as the composition and concentration ofthe Cu(II)–L-lysine complex on the separation were investigated. The linear ranges for thedetermination of ephedrine and pseudoephedrine were 15.0–225.0 and 20.0–250.0 mg/Lwith LODs both of 5.0 mg/L. Satisfactory result for the determination of ephedrine andpseudoephedrine in Ephedrae Herba from different producing area was obtained by theproposed method. Ephedrine and pseudoephedrine were separated effectively with eachother and with the other compounds in the sample. The RSD for the determination of thetwo constituents in the samples varied from 1.82 to 2.76%, and the recovery ranged between95.0 and 104.0%.

Keywords:

Cu(II)–L-lysine complex / CZE / Ephedrae Herba / Ephedrine / PseudoephedrineDOI 10.1002/elps.200700334

658 Electrophoresis 2008, 29, 658–664

1 Introduction

Ephedrae Herba (Ma Huang) has long been used in tradi-tional Chinese medicine for its diaphoretic, antipyretic,respiratory, antitussive, and antiasthmatic effects. It isknown to contain alkaloids (the so-called Ephedrae alkaloids)as its major bioactive compounds. Among these alkaloids, apair of diastereoisomeric compounds, ephedrine and pseu-doephedrine can typically account for more than 80% of thealkaloid content of the dried herb [1, 2]. Therefore, theseparation and determination of them are of pharmaceuticalimportance.

The separation of ephedrine and pseudoephedrine isdifficult because of their similarity in chemical structure andpKa (the pKa values of ephedrine and pseudoephedrine are9.58 and 9.74, respectively [3]). Several methods includingGC [4, 5], HPLC [6], and TLC [7] have been reported for thedetermination of ephedrine and pseudoephedrine. However,these methods require a derivatization step or exotic mobile

phase to allow for separation of the diastereoisomeric sub-stances. In addition, some methods of CE such as nonaque-ous CE (NACE) [8], micellar EKC (MEKC) [9], and CZEmodified by CD and its derivations [10–15] have also beenreported. The separation time of these CE methods [8–15] arealmost longer than 10 min, and the reproducibility of NACE[8] is not always sufficient.

Since Davankov and Rogozhin [16] firstly reported theirwork in 1971, ligand-exchange chromatography has becomean important method for resolving the enantiomers ofamino acid by HPLC. Based on this principle, the firstapplication of ligand-exchange CE (LECE) to the resolutionof dansylated amino acids, using a Cu(II)–L-histidine com-plex as a selector has been reported in 1985 by Gassman et al.[17]. Since then, other CE methods for the separation ofdansylated amino acids by using this type of complex asadditive have been introduced in succession [18–20]. Never-theless, no CZE method has been reported for the separationof pharmaceutical enantiomers, including ephedrine andpseudoephedrine, by using a Cu(II)–L-amino acid complex asadditive.

In this study, a Cu(II)–L-lysine complex was firstly usedas additive in a run buffer solution composed of Tris-H3PO4

(pH 4.5) for the separation of ephedrine and pseudoephe-drine by CZE. Ephedrine and pseudoephedrine were sepa-rated effectively with each other and with the other com-pounds in the sample in 4 min. Compared with some of the

Correspondence: Professor Qiu-E Cao, School of ChemicalScience and Technology, Yunnan University, Kunming 650091,P. R. ChinaE-mail: [email protected]: 186-871-5033726

Abbreviation: LECE, ligand-exchange CE

© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2008, 29, 658–664 CE and CEC 659

reported methods [4–15], the CZE method presented in thispaper is simpler, rapid, and possesses higher sensitivity andefficiency.

2 Materials and methods

2.1 Regents and materials

Authentic ephedrine and pseudoephedrine were obtainedfrom the National Institute for the Control of Pharmaceuticaland Biological Products (Beijing, China). Authentic L-lysineand benzyltriethylammonium were obtained from Merck(Darmstadt, Germany). CuSO4?5H2O was of analytical grade(Shanghai Chemical Company, Shanghai, China).

Stored solutions (0.5 mg/mL) of ephedrine, pseudoe-phedrine, and internal standard (benzyltriethylammonium)were prepared by diluting in ethanol (stored at 47C). Thestandard solution for condition testing, which contained100 mg/mL ephedrine, 100 mg/mL pseudoephedrine, and50 mg/mL benzyltriethylammonium, was diluted from storesolution by ethanol.

All other chemicals were of analytical grade and redis-tilled water was used for the preparation of buffer and relatedaqueous solutions.

2.2 Apparatus and CE conditions

Experiments were carried out on a Beckman P/ACE™ MDQCE system (Beckman Coulter, USA) equipped with a photo-DAD operating at 254 nm and a 57 cm (50 cm effectivelength)675 mm id uncoated fused-silica capillary (YongnianOptical Fiber Factory, Hebei, China). A pHS-2C pH meterwith a combination electrode (Rex Instrument Factory,Shanghai, China) was applied for pH measurements.

The sample was injected by pressure (0.5 psi, 3447.38 Pa)for 5 s at the cathode of the capillary. All separations wereachieved at 257C by 20 kV and computer-controlled using 32Karat software (Beckman Coulter). The electrolyte, whichwas a buffer solution of 20 mM Tris (the pH was adjusted byH3PO4) containing a mixture of Cu(II)–L-lysine and internalstandard (benzyltriethylammonium), was filtered through a0.45 mm membrane filter before use. The capillary was con-ditioned with 0.1 mol/L NaOH (3 min), distilled water(2 min), and separation buffer (4 min) previous to start-up.

2.3 Sample preparation and analysis

Pulverized Ephedrae Herba (1000 g) was refluxed with30 mL of ethanol for 60 min. After being filtered, the residuewas refluxed with 30 mL of ethanol for 30 min again. Thefiltrate was collected and distilled to nearly dry at first, andthen an aliquot of the internal standard solution (benzyl-triethylammonium) was added. The mixture was finally dis-solved to 50 mL with ethanol and passed through a 0.45 mmfilter before injection.

After the sample solution was analyzed directly under theselected conditions, an appropriate concentration of thestandard substances (ephedrine or pseudoephedrine) wasadded accurately to the sample solution, then filtered througha 0.45 mm membrane and followed by separating and analyz-ing under the same conditions as those for the sample solu-tion. Peaks of the two analytes in electropherograms could beidentified by comparing the peak height in the electro-pherograms of the sample solution before and after thestandard substances of ephedrine or pseudoephedrine wereadded. The recovery for the determination of ephedrine andpseudoephedrine then could be calculated as percent of theconcentration of the standard substances determined to thatreally added in the sample solution.

2.4 Calculation

The resolution (Rs) was calculated according to the followingformula:

Rs ¼ 1:18t2 � t1

W11=2 þW2

1=2(1)

where t2 and t1 are the migration times of pseudephedrineand ephedrine, W1

1=2 and W21=2 are the peak widths at half

height of ephedrine and pseudoephedrine, respectively.The relative migration time (tr) of the analyte is defined

as follows:

tr ¼ t� tin (2)

where t and tin are the migration times of the analyte and theinternal standard (benzyltriethylammonium), respectively.

The relative peak area (Ar) of the analyte is defined asfollows:

Ar ¼A

Ain(3)

where A and Ain are the peak areas of the analyte and theinternal standard (benzyltriethylammonium), respectively.

3 Results and discussion

3.1 Optimization of separation and analytical

conditions

3.1.1 Effect of pH

In order to separate ephedrine and pseudoephedrine undersimplest condition by CZE, modification of pH was con-sidered. In the absence of Cu(II) and L-lysine, buffers of20 mM Tris with various pH values ranging from 3.0 to 9.0adjusted by phosphoric acid were initially used to separatethe two analytes. Figure 1 summarizes the effects of bufferpH on the relative migration times. It was observed that the


660 F. Li et al. Electrophoresis 2008, 29, 658–664

Figure 1. Effect of pH on the migration time. CE conditions:20 mM Tris-H3PO4 buffer, 75 mm id657 cm (50.0 cm effectivelength) uncoated fused-silica capillary, applied voltage of 20 kV,capillary temperature of 257C, injection at 0.5 psi for 5 s, 254 nm.The standard solution for condition testing contained 100 mg/mLephedrine, 100 mg/mL pseudoephedrine, and 50 mg/mL benzyl-triethylammonium.

two analytes could not be separated effectively in the test pHrange. The relative migration times of the two analytes wereboth shortened firstly, and then prolonged while the resolu-tions decreased always with increasing the pH of buffer. AtpH .7.0, the peaks of ephedrine, pseudoephedrine, and theneutral compound (ethanol) crowded together with seriousoverlapping. The possible explanation could be expressed asfollows: as pH increased from 3.0 to 9.0, ephedrine or pseu-doephedrine changed from positive ions to nearly neutralforms (their pKa all over 9.0), while the wall of capillaryaltered from neutral form to the form with negative charge.The opposite variation of the charge in the analyte moleculeand on the wall of the capillary resulted in the minimummigration time appearing at pH 5.0. pH 4.5 was adopted toyield better resolution and shorter separation time.

3.1.2 Effect of composition and concentration of

buffer

Different buffers such as Tris-H3PO4, Na2B4O7-HCl,NaH2PO4-H3PO4, and NaAc-HAc with pH fixed at 4.5 weretested to separate ephedrine and pseudoephedrine. Theresults pointed out that different buffers lead to similarseparation of ephedrine and pseudoephedrine, but the opti-mal baseline was obtained by Tris-H3PO4 buffer.

A little improvement of the resolution was observedwhen the concentration of Tris-H3PO4 increased from 5 to30 mM, but the baseline aberration as well as the migrationtime of the solutes both manifold slightly. To shorten analy-sis time and reduce baseline noise, 20 mM Tris-H3PO4

(pH 4.5) was selected for the separation, although ephedrineand pseudoephedrine were still overlapping under suchconditions (see Fig. 2A).

Figure 2. (A) Electropherogram of standard mixture withoutCu(II)–L-lysine complex in buffer. Peaks 1, 2, and 3 representbenzyltriethylammonium, ephedrine, and pseudoephedrine.Buffer composition was 20 mM Tris-H3PO4 (pH 4.5). The otherseparation conditions are the same as those in Fig. 1. (B) Elec-tropherogram of standard mixture with 3 mmol/L Cu(II) in buffer.Buffer composition was 20 mM Tris-H3PO4 (pH 4.5) with 3.0 mMCu(II). The other separation conditions and peaks are the same asthose in Fig. 2A. (C) Electropherogram of the standard mixture ofephedrine and pseudoephedrine. Buffer composition was 20 mMTris-H3PO4 (pH 4.5) containing 2.5 mM Cu(II) and 5.0 mM L-lysine.The other separation conditions and the peaks are the same asthose in Fig. 2A.



3.1.3 Effect of the composition and concentration of

the Cu(II)–L-lysine complex

Based on the consideration that ephedrine and pseudoephe-drine both possess a hydroxyl group and a nitrogen atom,which can yield a complex with Cu(II), and an electron-fullaromatic ring, which can interact with L-lysine, a basic aminoacid with a pI at 9.4 and owing a positively charged side chain(an amino butyl group) at pH 4.5, the effect of Cu(II) orL-lysine on the separation was firstly studied by adding themseparately as modifier to the electrolyte containing 20 mMTris-H3PO4 (pH 4.5). The results pointed out that the additionof L-lysine alone produced almost no influence in the separa-tion, but the addition of Cu(II) in the absence of L-lysineexerted an obvious influence in the resolution. The resolutionof ephedrine and pseudoephedrine increased with increasingthe concentration of Cu(II) from 0.5 to 3.0 mM at first, andthen decreased when the Cu(II) concentration was above3.0 mM. The optimum separation was obtained by adding3.0 mM Cu(II) in buffer. The electropherogram shown inFig. 2B pointed out that ephedrine and pseudoephedrine inthe standard mixture were separated under such conditions,but the results from the application of the method for thedetermination of ephedrine and pseudoephedrine in theextract of Herba Ephedrae suggested that the coexistingchemicals in the samples interfered seriously in the determi-nation. The recovery for the determination of the two analytesespecially of pseudoephedrine was too high owing to the dis-turbance of other components in sample.

To obtain effective separation of ephedrine and pseudo-ephedrine with the other components in real sample, LECEwas then taken into account according to the principle repor-ted in references [16–20], and the Cu(II)–L-amino acid com-plex was selected as additive. Although L-lysine has seldombeen used in the LECE [18] and produced almost no influencein the separation by adding it alone in buffer, it was also cho-sen for the formation of a complex with Cu(II) as an additivein this study by consideration of its interaction with ephedrineor pseudoephedrine, which might increase the stability of theternary complex formed among Cu(II), L-lysine and ephe-drine or pseudoephedrine, and subsequently result in theimprovement of the resolution of the method.

The composition of the Cu(II)–L-lysine complex wasoptimized by adding 3.0 mM Cu(II) and various amount ofL-lysine in the buffer of Tris-H3PO4 (pH 4.5). The resultsshown in Table 1 indicate that the resolution of ephedrineand pseudoephedrine was firstly increased and thendecreased with increase in the molar ratio of Cu(II) toL-lysine in the complex. The maximum resolution wasachieved at a 1:2 ratio of Cu(II) to L-lysine.

The concentration of the Cu(II)–L-lysine complex withthe molar ratio of Cu(II) to L-lysine as 1:2 also showed a con-siderable influence on the separation of ephedrine andpseudoephedrine. With increasing the complex concentra-tion from 1.0 to 5.0 mM, the resolution of ephedrine andpseudoephedrine was improved while the migration time

Table 1. Influence of mol ratio of Cu(II) and L-lysine on the reso-lution of ephedrine and pseudoephedrine

Cu(II) and L-lysine (molar ratio) 1:0 1:1 1:2 1:3 1:4

Resolution (Rs) 2.03 3.09 4.64 3.87 2.97

and the baseline noise manifested together with the sensi-tivity for the detection reduced. This might result from thefact that the complex of Cu(II)–L-lysine possessed positivecharge at pH 4.5 and had UV absorption at the detectionwavelength of 254 nm. When the complex charged oppositecharge of the capillary, which charged negative charge atpH 4.5, so it could adsorb in the capillary, and then resultedin the decline of EOF and extension of the analysis time.Moreover, because the complex has absorption at the detec-tion wavelength, the increase in the complex concentrationproduced the strengthening of the absorption of background,magnifying the baseline noise and reducing the sensitivitycertainly. Taking sensitivity, resolution, analysis time, andbaseline into account together, 2.5 mM Cu(II)–L-lysine com-plex with the molar ratio of Cu(II) to L-lysine as 1:2 wasadded as additive in 20 mM Tris-H3PO4 (pH 4.5) for theseparation of ephedrine and pseudoephedrine.

3.1.4 Effect of apparatus conditions

With this selected electrolyte system, the influence of appliedvoltage, separation temperature, and injection time were alsostudied in detail. The results indicated that increase in theapplied voltage from 15 to 30 kV produced a decrease in themigration time and improvement of resolution, while thebaseline noise was magnified due to the excessive Joule heatcaused by the too much higher applied voltage. Moreover,increasing the cartridge temperature from 15 to 357C wouldresult in shortening the migration time and reducing theresolution. The effects of the injection time on separationpointed out that the sensitivity for the determinationincreased with the increase in injection time varying in therange of 3–10 s, while the shape of peaks broadened. There-fore, 20 kV of applied voltage, 257C of separation tempera-ture, and 5 s of injection time were selected as the apparatusconditions.

Under the conditions selected above, ephedrine, pseud-oephedrine, and the internal standard (benzyltriethyl-ammonium) were separated effectively. A typical electro-pherogram for the separation of the standard mixture ofthese three constituents is shown in Fig. 2C.

3.2 Mechanism for complex reaction among Cu(II), L-

lysine, and analyte

To investigate the interaction of analyte with the complex ofCu(II)–L-lysine in the capillary, the UV absorption spectra ofCu(II), L-lysine, pseudoephedrine, Cu(II)–L-lysine, Cu(II)–



Figure 3. UV spectrum using 20 mM Tris-H3PO4 (pH 4.5) as refer-ence solution. Curves 1–6 represent 1, 50 mM L-lysine; 2, 10 mMpseudoephedrine; 3, 1 mM Cu(II); 4, 1 mM Cu(II) with 1 mMpseudoephedrine; 5, 1 mM Cu(II) with 3 mM L-lysine; and 6, 5with 1 mM pseudoephedrine.

pseudoephedrine, and Cu(II)–L-lysine-pseudoephedrine in20 mM Tris-H3PO4 buffer (pH 4.5) was obtained as shown inFig. 3. It could be seen that the shape of spectra including themaximum absorption wavelength and the absorbance ofCu(II)–L-lysine system (line 5 in Fig. 3) was different from thatof Cu(II) (line 3) and L-lysine (line 1), which suggested thatCu(II) could bind with L-lysine to form a complex. The molarratio of Cu(II) to L-lysine in the absence of pseudoephedrinewas determined as 1:3 by Job’s method of continuous varia-tion and by the molar ratio method. Figure 3 also shows thatCu(II) could bind with pseudoephedrine in the absence ofL-lysine, which produced the difference among the spectra ofCu(II)–pseudoephedrine (line 4), Cu(II) (line 3), and pseu-doephedrine (line 2). When pseudoephedrine was added tothe complex system of Cu(II)–L-lysine with the molar ratio ofCu(II) to L-lysine as 1:3, the spectra (line 6) changed further incomparison with the spectra of Cu(II)–L-lysine (line 5) andCu(II)–pseudoephedrine (line 4), a slight bathochromic shiftof the maximum absorption wavelength together with theincrease in the absorbance was observed. This result certifiedthat a reaction between pseudoephedrine and the complex ofCu(II)–L-lysine occurred. Regarding the fact that the opti-mum molar ratio of Cu(II) to L-lysine was 1:2 in the CEoperation and the results shown in ref. [19, 20], we conclude

that a ternary complex might be formed among Cu(II),L-lysine, and pseudoephedrine by the partly replacing pro-cedure of pseudoephedrine to L-lysine in the complex ofCu(II)–L-lysine.

The absorption curves of ephedrine, Cu(II)–ephedrine,and Cu(II)–L-lysine-ephedrine systems were basically over-lapped by the curves of pseudoephedrine, Cu(II)–pseudo-ephedrine, and Cu(II)–L-lysine–pseudoephedrine, respec-tively. So the replacing procedure of ephedrine to L-lysine inthe Cu(II)–L-lysine complex also happened.

As a result of DL-isomerism of ephedrine and pseudoe-phedrine, the stabilities of their ternary complexes were dis-tinct. Ephedrine is an L-isomerism, so it is easier to replaceL-lysine in the Cu(II)–L-lysine complex and form a morestable ternary complex with Cu(II) and L-lysine comparedwith pseudoephedrine, which possesses D-steric configura-tion, according the ref. [19, 20]. The stable binding of theanalyte to the constituent in the run buffer can accelerate themigration of the analyte in the CE [19, 20]. Therefore, themigration time of the peak of ephedrine is shorter than thatof pseudoephedrine in the electropherogram.

3.3 Characteristics of quantitative analysis

Calibration graphs (relative peak area, Ar, vs. concentration,C, in mg/L) for the determination of ephedrine and pseu-doephedrine were constructed by using 50 mg/L benzyl-triethylammonium as internal standard. The linear ranges,regression equations, and correlation coefficients of thestandard curves were summarized in Table 2. Furthermore,the LODs (calculated by treble of S/N) and the RSD of therelative migration time (tr) as well as the relative peak-area(Ar) for seven replicate determinations of the standard mix-ture of ephedrine and pseudoephedrine with the concentra-tion both of 100 mg/L are also summarized in Table 2. Theresults show that the method was of wide linear range, goodreproducibility, and high sensitivity.

3.4 Determination of ephedrine and

pseudoephedrine in the sample

Ethanol solution of sample extracts was firstly filteredthrough a 0.45 mm membrane, then injected and separatedunder the selected conditions. Figure 4 shows the electro-

Table 2. Analytical characteristics for the determination of the ephedrine and pseudoephedrine

Constituents Linear range(mg/L)

Regression equationa) r LOD (mg/L) RSD (n = 7,%)b)

tr Ar

Ephedrine 15.0–225.0 A = 0.00851 1 0.0136c 0.9987 5.0 1.98 2.15Pseudoephedrine 20.0–250.0 A = 0.0102 1 0.0119c 0.9915 5.0 2.50 2.30

a) A is the relative peak-area and c is the concentration expressed as mg/L.b) tr and Ar represent the relative migration time (defined as the migration time of analyte relative to that of the internal standard (benzyl-

triethylammonium)) and relative peak area (defined as the peak area of analyte relative to that of the internal standard).



Figure 4. Electropherograms of the extract of Ma Huang (Hebei2). The separation conditions and the peaks are the same asthose in Fig. 2C.

pherogram for one Ma Huang sample (sample 4, obtainedfrom Hebei) and Table 3 summarizes the results for thequantitative analysis of ephedrine and pseudoephedrine in allthe samples. Moreover, to verify the accuracy of the resultsobtained in the determination of Ma Huang samples by thismethod, the recovery was also determined. From the resultslisted in Table 4, it could be seen that the recovery for thedetermination of ephedrine and pseudoephedrine in the foursamples were satisfactory, which indicated that the coexistingchemicals in the extract of the samples did not interfere in thedetermination by this method, and the quantitative resultsobtained by this method are accurate and reliable.

Table 3. Results for the determination of ephedrine and pseudo-ephedrine in sample extracts (n = 5)

Samplesa) Ephedrine (%) Pseudoephedrine (%)

Found value RSD Found value RSD

1 0.67 1.98 0.22 2.252 0.60 2.04 0.19 2.763 0.79 2.15 1.824 0.84 2.00 2.54

a) Samples 1 and 2 are Ma Huang obtained from Gansu (China),and 3 and 4 are Ma Huang obtained from Hebei (China).

4 Concluding remarks

CE has been shown to provide selectivity and high effi-ciency in the separation and determination of ephedrineand pseudoephedrine in a herbal drug by optimizing thepH, composition, and concentration of run buffer as wellas the composition and concentration of the Cu(II)–L-lysine complex within 4 min. There was no interferencefrom the other constituents of the extracts in varioussources of samples. This method is simple, economic,stable, and rapid. It is especially suitable for analyzingbulky samples and for quality control in pharmaceuticalplants.

The authors have declared no conflict of interest.

Table 4. Recovery for the determination of ephedrine and pseudoephedrine in sample extracts (n = 5)

Samplesa) Ephedrine Pseudoephedrine

Foundb)

(mg/L)Addedc)

(mg/L)Recovery(%)

RSD(%)

Foundb)

(mg/L)Addedc)

(mg/L)Recovery(%)

RSD(%)

1 135.3 50.0 104.0 2.15 44.44 50.0 95.0 1.982 119.8 50.0 98.0 1.87 37.93 50.0 103.0 2.173 154.8 40.0 96.0 1.96 54.88 50.0 101.0 2.544 168.0 40.0 99.0 2.01 50.00 50.0 97.0 1.65

a) Samples are as same as that in Table 2.b) Concentration of ephedrine or pseudoephedrine found in the extract of samples.c) Concentration of ephedrine or pseudoephedrine added in the extract of sample.



5 References

[1] Yang, Y., Feng, W. S., The Handbook of Distillation andSeparation of the Chemicals in Chinese Traditional Medicine,China Press of Traditional Chinese Medicine, Beijing 2001, p.311.

[2] Ren, R. A., Identification of Chinese Traditional Medicine,Shanghai Scientific & Techincal Publishers, Shanghai 1998, p.454.

[3] Kuang, X. H., Zhaongyao Huaxue, China Press of TraditionalChinese Medicine, Beijing 2003, p. 343.

[4] Marchei, E., Pellegrini, M., Pacifici, R., Zuccaro, P., Pichini, S.,J. Pharm. Biomed. Anal. 2006, 41, 1633–1641.

[5] Wang, M., Marriott, P. J., Chan, W. H., Lee, A. W. M., Huie, C.W., J. Chromatogr. A 2006, 1112, 361–368.

[6] Imaz, C., Carreras, D., Navajas, R., Rodriguez, C. et al., J.Chromatogr. 1993, 631, 201–205.

[7] Cheng, Y., Feng, Q., Wang, H. R., Chin. Tradit. Pat. Med. 1990,12, 15–18.

[8] Liu, R. M., He, F. Y., Chin. J. Anal. Lab. 2003, 22, 38–40.

[9] Wang, W. P., Li, C. H., Li, Y., Hu, Z. D., Chen, X. G., J. Chroma-togr. A 2006, 1102, 273–279.

[10] Liu, Y. M., Sheu, S. J., J. Chromatogr. 1992, 600, 370–372.

[11] Flurer, C. L., Lin, L. A., Satzger, R. D., Wolnik, K. A., J. Chro-matogr. B 1995, 669, 133–139.

[12] Szökö, É., Gyimesi, J., Barcza, L., Magyar, K., J. Chromatogr.A 1996, 745, 181–187.

[13] Hellriegel, C., Händel, H., Wedig, M., Steinhauer, S. et al., J.Chromatogr. A 2001, 914, 315–324.

[14] Mateus-Avois, L., Mangin, P., Saugy, M., J. Chromatogr. B2003, 791, 203–216.

[15] Phinney, K. W., Ihara, T., Sander, L. C., J. Chromatogr. A2005, 791, 90–97.

[16] Davankov, V. A., Rogozhin, S. V., J. Chromatogr. 1971, 60,280–283.

[17] Gassmann, E., Kuo, J. E., Zare, R. N., Science 1985, 230, 813–818.

[18] Lu, X. N., Chen, Y., Guo, L., Yang, Y. F., J. Chromatogr. A2002, 945, 249–255.

[19] Schmid, M. G., Grobuschek, N., Lecnik, O., J. Biochem. Bio-phys. Methods 2001, 48, 143–148.

[20] Blancao, M., Valverde, I., Trends Anal. Chem. 2003, 22, 428–433.


Documents

Separation and determination of ephedrine and pseudoephedrine in Ephedrae Herba by CZE modified with a Cu(II)–L-lysine complex