Monoliths from poly(ethylene glycol) diacrylate and dimethacrylate for capillary hydrophobic interaction chromatography of proteins

Rigid monoliths were synthesized solely from poly(ethylene glycol) diacrylates (PEGDA) or poly(ethylene glycol) dimethacrylates (PEGDMA) containing different ethylene glycol chain lengths by one-step UV-nitiated polymerization. Methanol/ethyl ether and cyclohexanol/decanol were used as bi-porogen mixtures for the PEGDA and PEGDMA monoliths, respectively. Effects of PEG chain length, bi-porogen ratio and reaction temperature on monolith morphology and back pressure were investigated. For tri- and tetra-ethylene glycol diacrylates (i.e., PEGDA 258 and PEGDA 302), most combinations of methanol and ethyl ether were effective in forming monoliths, while for diacrylates containing longer chain lengths (i.e.. PEGDA 575 and PEGDA 700), polymerization became more sensitive to the bi-porogen ratio. A similar tendency was also observed for PEGDMA monomers. Polymerization of monoliths was conducted at approximately 0 C and room temperature, which produced significant differences in monolith morphology and permeability. Monoliths prepared from PEGDA 258 were found to provide the best chromatographic performance with respect to peak capacity and resolution in hydrophobic interaction chromatography (HIC). Detailed study of these monoliths demonstrated that chromatographic performance was not affected by changing the ratios of the two porogens, but resulted in almost identical retention times and comparable peak capacities. An optimized PEGDA 258 monolithic column was able to separate proteins using a 20-min elution gradient with a peak capacity of 62. Mass recoveries for test proteins were found to be greater than 90, indicating its excellent biocompatibility. All monoliths demonstrated nearly no swelling or shrinking in different polarity solvents, and most of them could be stored dry, indicating excellent stability due to their highly crosslinked networks. The preparation of these in situ polymerized single-monomer monolithic columns was highly reproducible. The relative standard deviation (RSD) values based on retention times of retained proteins were all within 2.2%, and in most cases, less than 1.2%. The RSD values based on peak areas were within 9.5%, and in most cases, less than 7.0%. The single-monomer synthesis approach clearly improves column-to-column reproducibility. (C) 2010 Elsevier B.V. All rights reserved.