Jh. Jumppanen et al., OPTIMIZATION OF RESOLUTION OF 13 DIURETICS IN CZE BY CONTROLLING CAPILLARY LENGTH AND ELECTROOSMOTIC FLOW VELOCITY, The Journal of microcolumn separations, 6(6), 1994, pp. 595-604
The separation of diuretics in CZE can be further improved by controll
ing the capillary length and the analysis times independently, while o
ther parameters are kept constant. This can be done by using a modifie
r in the electrolyte solution which has a strong effect on the electro
osmosis, but negligible effect on the selectivity of the electrolyte s
ystem. In this work, [Cu(NH3)(4)](2+) was used at low concentration (O
- 140 mu M) to control the electroosmotic flow velocity (v(eo)) and h
ence the analysis time, determined as the migration time of the last m
igrating compound (t(fin)) With other parameters, especially V, kept c
onstant, the use of shorter capillaries enables the use of high electr
ic field strength for the separation. The use of quadratic regression
models proved to be an effective way to study the effect of L(tot) and
t(fin) on the R(s) of each analyte pair. The theoretical values for t
he separation of peak maxima correlated well with the values predicted
by the empirical regression models. The actual band broadening was mu
ch stronger, however, than what was predicted by the overly simplified
theoretical calculations based on the electrophoretic mobilities as w
ell as the self diffusion coefficients of the diuretics, determined by
the two marker technique. Further study is needed on theoretical mode
lling of the band broadening, while the use of regression models and d
esirability functions provided a convenient way to determine the optim
al conditions. The results showed that independent control of L(tot) a
nd t(fin) offers an effective way to further optimize capillary electr
ophoretic separation. This method should prove important in cases wher
e sufficient separation is difficult to achieve.