Extensive retention for non-suppressed ion chromatography of anions have been acquired for 17 analytes (halides, oxohalides, nitrite, nitrate, sulfite, sulfate, bisulfite, thiosulfate, phosphate, thiocyanate, carbonate, acetate and oxalate) on three stationary phases (Waters IC Pak A, Hamilton PRP-X100 and Vydac 302.IC 4.6) using 7 eluent types (benzoate, phthalate, hydroxide, carbonate/bicarbonate, gluconate/borate, p-toluenesulfonate and phosphate). These retention data are used to assess the validity of retention models which predict a linear relationship between the logarithm of solute capacity factor and the logarithm of the activity of the eluent competing anion. The linearity of these plots is uniformly good, but the slopes differ markedly from those predicted from theory. When the eluent contains two competing anions neither the dominant equilibrium approach nor the effective charge approach give reliable prediction of the slopes. Optimization of one eluent parameter at a time ( e.g. the concentration of the competing anion in the eluent) can be successful if the slope of the retention plot is determined by measurement of analyte retention times at two eluent compositions falling at the extremes of the range of eluent compositions under consideration. An example of this “end points” method is provided, in which the concentration of a phthalate eluent is optimized.
Read full abstract