Ultra-high performance liquid chromatography: Accessible surface area, packing uniformity and broadening from hardware

Ao Zeng, Purdue University


Ultra-high performance liquid chromatography is a vital tool for proteomics and pharmaceutical analysis. With the success of protein drugs, particularly therapeutical monoclonal antibodies, there is wide interest to improve the performance of protein separation. The objective of this research is to improve protein separation by studying the accessible surface area of common column material, improving packing uniformity of columns and characterizing band broadening from hardware. Surface area of columns affects analyte retention. Conventional column material, including totally porous particles and core-shell particles, have high surface area due to their porous structure. However, monoclonal antibodies drugs are 150 kDa in size and much larger than the traditional small molecule drugs. Larger analytes have less access to the free volume inside pores, leading to disproportionate loss of accessible surface area. Accessible surface area of common columns were studied by continuing loading analyte of various size onto the column under retaining conditions until all the accessible surface area were occupied by the analytes and a breakthrough signal was observed. It was found that totally porous particles provide one magnitude higher accessible surface area for small molecules than core-shell particles and non-porous particles. Core-shell particles provide more accessible surface area for small molecules than non-porous particles do. These findings agree with previous studies where the surface area of the materials were measured with nitrogen adsorption isotherm. For larger analyte, monoclonal antibodies, however, the three column material provided similar accessible surface area. This demonstrated that larger analytes have less access to the pores of conventional column material. The second study was to improve the packing uniformity of columns packed with non-porous particles. A gap was observed in the front of the column, which decreased the packing uniformity. The problem was due to that the most pressure drop during packing was inside a narrow tubing with the old packing procedure. The problem was solved by packing with higher effective pressure on the column bed. Also, the new packing technology was extended to packing bare silica, which allows in-column modification. The success of capillaries made with in-column modification provide a strong incentive to make stainless steel columns by in-column modification. The columns packed with bare silica and modified in-column showed even better packing uniformity. The third study was on the band broadening from hardware. Band broadenings lead to loss of resolution. Labeled monoclonal antibodies was stacked onto the head of the column with non-porous particles. The injected plug was less than 0.5 mm in length. Typical peak width at the detector was 30 s (2 σ), which converts to 36 mm in physical width before the peak eluted out of the packed bed. The current results indicate that the injection band width is narrow. The majority of the tubing are from the packed bed or the exit frit of the column. Overall, the research demonstrated that non-porous particles provided adequate accessible surface area for the analysis of monoclonal antibodies. The packing uniformity was improved with the new packing method.




Wirth, Purdue University.

Subject Area

Analytical chemistry

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server