高效液相層析(HPLC)及毛細管電泳(CE)分析方法，兩者同屬分離技術，它們遵循不同的分離機制，各有其優缺點，是目前最常用來定量中藥指標成分的分析方法。本研究在開發CE方法，分析川芎及茵陳蒿成分；併用HPLC及CE方法，分析類化合物及柑橘類藥材成分，並比較CE及HPLC方法之優劣，最後利用已開發柑橘類藥材的HPLC分析方法，進行其基原的化學辨識研究。
結合逆向毛細管區帶電泳層析(reversed EOF CZE)及膠束電動力學毛細管層析(MEKC)兩分析技術，可成功的分離川芎藥材的有機酸及精油成分。於reversed EOF CZE方法，使用70%含8 mM Na2B4O7－3mM NaH2PO4－8 mM LTAC(pH=9.02)的溶液及30% 的CH3CN，可於12分鐘內定量分析九個有機酸(phthalic acid, caffeic acid, protocatchuic acid, folic acid, nicotinic acid, vanillic acid, p-hydroxybenzoic acid, ferulic acid及folinic acid)；在MEKC方法中，使用60%含20 mM Na2B4O7－10 mM NaH2PO4－40 mM SDS(pH=9.02)的溶液及20% CH3CN和20 % MeOH，可於30分鐘內定量分析butylphthalide, senkunolide A, ligustilide及butylidenephthalide等四個精油成分。
使用MEKC分析技術，藉著添加SDS於硼酸鹽緩衝溶液中，以0.05 M NaOH 調pH值至9.82時，可於42分鐘內成功地分離茵陳蒿中的十二個指標成分(phenol, o-cresol, m-cresol, p-cresol, 4-ethyl phenol, 6-ethyl phenol, eugenol, capillarisin, scopletin, chlorogenic acid和caffeic acid)，尤其是HPLC中不易分離的cresols及ethyl phenol位置異構物，可在此方法中分離。
為比較CE與HPLC兩方法之優劣，並了解化合物結構與分析方法的關聯性，本研究分別對十二種較為常見的類化合物(anthraquinone, chrysphanol, aloe-emodin, alizarin, purpurin, emodin, sennoside B, sennoside A, anthraquinone 2-carboxylic acid, quinarizalin, rhein及anthraflavic acid)，進行開發CE與HPLC分析方法。在CE部分，採CZE分離模式，使用90%含30 mM Na2B4O7 (以0.05 M NaOH調pH值至10.56)與10%的CH3CN (v/v)為緩衝溶液，於39分鐘內，可分離11個類化合物(anthraquinone除外)。在HPLC部分，利用磷酸鹽及氰甲烷沖提系統，可在63分鐘內，分離十二個類化合物；本分析方法可直接應用於大黃藥材的萃取液，並可定量其中六個成分(chrysphanol, aloe-emodin, emodin, sennoside B, sennoside A及rhein)。實驗結果顯示，兩分析方法在系統的適宜性、分析時間、化合物移動順序及各化合物之理論板數等方面各有明顯差異。
使用HPLC與CE分析技術，分析柑橘類藥材十八個指標成分(umbelliferone, citropten, imperatorin, narirutin, naringin, naringenin-7-glucoside, hesperidin, neohesperidin, quercetin, naringenin, hesperitin, chrysine, sinensetin, nobiletin, acacetin, tangeretin, 5-demethyl-nobiletin and synephrine)。在HPLC部分，利用磷酸鹽及甲醇/氰甲烷(1/1)沖提系統，可在60分鐘內，分離此十八個化合物；在CE 部分，採MEKC分離模式，使用40 mM Na2B4O7 及40 mM SDS (以0.05 M NaOH調pH值至10.01)與氰甲烷(v/v)為緩衝溶液，於40分鐘內，可分離此十八個化合物中的十三個化合物。本分析方法可直接應用於枳實、枳殼及柑橘藥材的萃取液。
此外，本研究以HPLC方法，分析市售柑橘類藥材飲片共五十批，進行其基原之化學辨識研究。發現市售枳實只有綠衣枳實一種，而枳殼藥材則有酸橙枳殼及香圓枳殼。酸橙枳殼中HE/NG含量比值>0.49，香圓枳殼含量比值<0.40；在酸橙枳殼中HE/NE之含量比值>2.30，香圓枳殼含量比值<1.31；在酸橙枳殼中NGC/NE之含量比值>0.21，香圓枳殼之含量比值<0.02。根據以上數據，可做為辨識枳實、枳殼基原的參考依據。

The methods to determine Chinese herbs are usually performed by capillary electrophoresis (CE) and high-performance liquid chromatography (HPLC). Both of these methods follow different separation mechanisms. In this study, CE method was used to assay the contents of constituents of Ligustici Rhizoma and Artemisiae Capillaris Herba. Two techniques including CZE and HPLC have been developed to separate anthraquinones and citrus herbs. The goals of this study are to compare the differences and to establish the relationship of analyses and analytical condition between CE and HPLC. Finally, we used the HPLC method to postulate the origin and quality of the citrus herbs.
A reversed electroosmotic flow capillary zone electrophoresis (reversed EOF CZE) and a micellar electrokinetic chromatography (MEKC) were developed to analyze the organic acids and essential oil constituents of Ligustici Rhizoma. In reversed EOF CZE method, a buffer solution containing 8 mM sodium borate, 3 mM sodium dihydrogenphosphate and 9 mM lauryltrimethylammonium chloride and acetonitrile (7: 3) were found to be the most suitable approach to determine the contents of phthalic acid, caffeic acid, protocatchuic acid, folic acid, nicotinic acid, vanillic acid, p-hydroxybenzoic acid, ferulic acid and folinic acid within 12 minutes. In MEKC technique based on sodium dodecyl sulfate was applied to analyze butylphthalide, senkunolide A, ligustilide and butylidenephthalide within 30 minutes.
Using the techniques of MEKC by adding sodium dodecyl sulfate to sodium borate solution and adjusting to pH 9.82 with 0.05 M NaOH, we can separate phenol, o-cresol, m-cresol, p-cresol, 4-ethyl phenol, 6-ethyl phenol, eugenol, capillarisin, scopletin, chlorogenic acid and caffeic acid within 42 minutes. Especially, the isomers of cresol and ethylphenol that can’t be separated in HPLC have been analyzed in this method.
In order to compare the superiority and shortcoming of both methods and to understand the relationship between chemical structures and analysis methods, this study has developed both CE and HPLC methods for the analysis of twelve anthraquinones (anthraquinone, chrysphanol, aloe-emodin, alizarin, purpurin, emodin, sennoside B, sennoside A, anthraquinone 2-carboxylic acid, quinarizalin, rhein and anthraflavic acid). A buffer solution containing 30 mM sodium borate (adjusted to pH=10.56 with 0.05 M NaOH) and CH3CN (9: 1) in CZE or with a linear gradient elution containing 20 mM KH2PO4 (adjusted to pH=2.91with 0.05% H3PO4) and MeOH in HPLC was found to be the best. Contents of six components (chrysphanol, aloe-emodin, emodin, sennoside B, sennoside A, rhein) in Rhei Rhizoma could easily be determined within 39 minutes by CE or 63 minuts by HPLC. The effect of buffers on this separation and the validation of two methods were studies.
The HPLC and CE methods for separating citrus herbs were established. The major components of citrus herbs contained umbelliferone, citropten, imperatorin, narirutin, naringin, naringenin-7-glucoside, hesperidin, neohesperidin, quercetin, naringenin, hesperitin, chrysine, sinensetin, nobiletin, acacetin, tangeretin, 5-demethyl-nobiletin and synephrine. The HPLC analysis was carried out within 60 minutes by using a gradient solvent system of phosphate salt buffer-methanol-acetonitrile. Using the techniques of MEKC by adding sodium dodecyl sulfate to sodium borate solution and adjusting to pH 10.01 with 0.05 M NaOH, we could separate eighteen of the thirteen components within 40 minutes. The effect of buffers on this separation and the validation of two methods were studies.
In addition, we used HPLC method to analyze the citrus herbs collected from market. In this study, we found that the Aurantii Fructus Immaturus belonged to Poncitrus trifoliata RAF, and the Aurnatii Fructus Maturus was derived from Citrus aurantium L. and C. Wilsonii TANAKA. The two kinds of Aurnatii Fructus Maturus can be distinguished by the ratios of HE/NG, HE/NE and NGC/NE. The ratio HE/NG in C. aurantium was higher than 0.49, but less than 0.40 in C. Wilsonii. In addition, the ratio HE/NE was higher than 2.30 for the farmer, but less than 1.31 for the latter; the ratio NGC/NE was higher than 0.21 for the former, and less than 0.02 for the latter. From the data of chemical analysis of an herb’s constituents, we can postulate the origin and quality of the herb.

1. M. Twett, Proc. Warsaw Soc. Nat. Sci. Biol., 1930, 14, 6.
2. A. J. P. Martin and R. L. M Synge, J. Biochem., 1941, 35, 1358.
3. T. James and A. J. P. Martin, Analyst, 1952, 77, 915.
4. E. Stahl, Chemiker-Ztg., 1958, 82, 323.
5. J. F. K. Huber and J. A. R. J. Hulsman, Anal. Chem. Acta, 1967, 38, 305.
6. S. Hjerten, Chromatogr. Rev., 1967, 9, 122.
7. A. Tiselius, Tran. Faraday Soc., 1937, 33, 524.
8. J. W. Jorgenson and K. D. Lukacs, Anal. Chem., 1981, 53, 1298.
9. S. Hjerten and M. O. Zhu, J. Chromatogr., 1985, 346, 265.
10. S. Hjerten, J. L. Liao and K. Yao, J. Chromatogr., 1987, 387, 127.
11. J. R. Mazzeo and I. S. Krull, BioTechniques, 1991, 10, 638.
12. S. Hjerten and M. D. Zhu, J. Chromatogr., 1985, 327, 157.
13. S. Hjerten and M. D. Zhu, Protides Biol. Fluids, 1985, 33, 537.
14. S. Hjerten, K. Elenbring F. Kita, J. L. Liao, A. J. C. Chen, C. J. Siebert and M. D. Zhu, J. Chromatogr., 1957, 403, 1987.
15. A. S. Cohen and B. L. Karger, J. Chromatogr., 1987, 397, 409.
16. A. S. Cohen, A. Paulus and B. L Karger, Chromatographia, 1987, 24, 14.
17. E. M. Everaerts and P. E. M. Verheggen, ed., New Directions in Electrophoretic Methods, J. W. Jorgenson and M. Phillips, Amer. Chem. Soc. Symp. Vol. 335, Chap. 4, Washington DC, 1997.
18. J. W. Jorgenson and K. D. Lukacs, J. Chromatogr., 1981, 218, 209.
19. J. W. Jorgenson and K. D. Lukacs, J. High Resoln Chromatogr. Chromatogr. Comm., 1951, 4, 230.
20. J. W. Jorgenson and K. D. Lukacs, Clin. Chem., 1981, 27, 1551.
21. J. W. Jorgenson and K. D. Lukacs, Science, 1983, 222, 266.
22. J. W. Jorgenson, Trends Anal. Chem., 1984, 3, 51.
23. K. Altria and C. Simpson, Anal. Proc., 1986, 23, 453.
24. T. Tsuda, J. High Resoln Chromatogr. Chromtogr. Comm., 1987, 10, 622.
25. S. Terabe, K. Otsuka and T. Ando, Anal. Chem., 1989, 61, 25l.
26. K. Otsuka and S. Terabe, J. Microcol. Sep., 1989, l, 150.
27. A. T. Tsuda, K. Normura and G. Nakagawa, J. Chromatogr., 1982, 248, 241.
28. M. J. Sepaniak and R. O. Cole, Anal. Chem., 1987, 59, 472.
29. T. Balchunas and M. J. Sepania, Anal. Chem., 1988, 60, 1466.
30. A. M. Martin, G. Guiochon, Y. Warbroehol and J. W. Jorgenson, Anal. Chem., 1985, 57, 559.
31. M. M. Bushey and J. W. Jorgenson, J. Microcol. Sep., 1989, 1, 125.
32. A. Dobashi, T. Ono, S. Hara and J. Yamaguchi, J. Chromatogr., 1989, 480, 413.
33. S. Terabe, H. Utsumi, K. Otsuka, T. Ando, T. Inomata, S. Kuze and Y. Hanaoka, J. High Resoln Chromatogr. Chromatogr. Commun., 1986, 9, 666.
34. D. E. Burton, M. J. Sepaniak and M. P. Maskarinec, J. Chromatogr. Sci., 1987, 25, 514.
35. S. Terabe, K. Otsuka, K. Ichikawa, A. Tsuchiya and T. Ando, Anal. Chem., 1984, 56, 111.