The effect of structure modification of chondroitin sulfate C on its enantioselectivity to several representative basic drugs in capillary electrophoresis was investigated. Chemical desulfation showed no remarkable decrease in selectivity, whereas depolymerization with chondroitinase ABC resulted in complete loss of selectivity. Comparison with chondroitin sulfate A indicated considerable decrease in selectivity with this isomer. The great retention of enantioselectivity in the desulfated derivative suggests that the selectivity comes from the difference of the magnitude of an interaction in the multipoint mechanism between a part of the drug molecule and a functional group in chondroitin sulfate C other than the sulfate group. The sulfate group is not considered to play a major role for chiral separation. The complete loss of selectivity by depolymerization is consistent with a general tendency of lower selectivity in smaller saccharides, and the priority of chondroitin sulfate C to chondroitin sulfate A suggests the importance of the hydroxyl group at C in the galactosamine 4 residue. During the course of this work we observed heavy tailing of the peaks of basic drugs in some batches of uncoated fused-silica capillaries under acidic conditions and solved this problem by oubly coating capillaries with Polybrene followed by chondroitin sulfate C. On the other hand, we demonstrated the usefulness of a special technique which uses a short, wider bore PTFE tube-attached capillary for the study of the effect of depolymerization, in order to minimize sample amount.

We introduced colominic acid as a new chiral selector for capillary electrophoresis of basic drugs. Use of a low concentration phosphate buffer containing this polysaccharide and a Polybrene/colominic cid double coated capillary allowed excellent separation of the enantiomers of primaquine, chloroquine and tryptophan. Other drugs giving partial enantioseparation include laudanosine and salbutamol. Capillary coating with Polybrene followed by colominic acid eliminated the problems of peak tailing and low reproducibility of migration time in uncoated capillaries. The optimum pH was in the acidic region but varied among drugs. A low capillary temperature of 16℃ and a colominic acid oncentration of 9w/v% are recommended for practical analysis of these drugs. Colominic acid preparations having higher molecular masses gave better enantioseparation, and N-acetylneuraminic acid, the component monosaccharide, did not give any enantioseparation.

我们合成了五种新的不对称偶氟羧生物，并研究了它们与钙离子之间的一种特殊反应-B型反应，考察了试剂分子结构与反应性能的关系，着重探讨了讨羧基偶氮羧与钙离子的B型反应行为，在柠檬酸升质中。钙离子与该试剂形成组成比为1∶2的灵敏的稳定配合物，最大吸收波长为718nm（△λ=157nm），表观摩尔吸光系数ε=1.51×luL.Mol-1.cm-1，测毫条件下，1-13μgCa/35mL符合Beer定律，采用小体积显色，反应可在15分钟内进行完全配合物可稳定4小时不变。动力学研究结果表明，对羧基偶氯羧与钙离子只发生B型反应，反应级数为2。

本文建立了对羧基偶氮羧分光光度法测定微量钙的方法，研究了试剂与钙的B型显色反应的条件、影响因素及形成机理。在柠檬酸介质中，钙与该试剂只发生灵敏的13型显色反应，络合物最大吸收波长为718nm（△λ=157nm），表观摩尔吸光系数ε=1.51×105L·mor-1·cm-1，反应完全后，吸光度可保持4h不变。方法用于直接测定复方氯化钠注射液中的钙，回收率在99.20%～103.5%之间，结果与标准方法一致，相对标准偏差小于1.5%。

基于间抓偶氮安替比林可与钙在碱性介质中形成灵敏的兰色络合物这一显色反应，本文在自行组装的带微型计算机的流动注射分析仪上，以分光光度计作检测器，建立了测定微量钙的流动注射分光光度法。最佳显色反应条件：0.08mol/L NaOH-0.012%显色剂一三乙醇胺溶液，测定波长为630nm。以三乙醇胺为掩蔽剂，方法选择性较高。钙的线性范围为1.0～15.0μg/ml。检测限0.5μg/ml。进样频率可达55样次/h。方法可不经分离直接测定药物和水中的钙，回收率及情密度均良好