Evaluation and optimization of two complementary cross-flow ultrafiltration systems toward isolation of coastal surface water colloids

While colloidal phases in natural waters are important to the speciation and bioavailability of trace metals and organic compounds, accurate isolation of these submicron entities from their lower molecular-weight counterparts has proved challenging. Here, both laboratory and field experiments have been conducted to constrain integrity aspects of two different cross-flow filtration (CFF) systems. We tested both a commonly applied 5000 cm(2) CFF system (Pellicon2, Millipore; manufacturer-specified 1 and 3 kDa cutoff) and a less studied mini-CFF system with a 50 cm2 membrane area (Pellicon Lab Scale XL, Millipore; with 5, 10, 100, and 1000 kDa cutoff). A natural water matrix amended with a series of fluorescently tagged colloidal probes was employed to demonstrate that a cross-flow ratio (CFR; retentate-to-permeate flux) > 15 was required for both of these systems to obtain high recoveries of colloids. Contrary to previous calibrations of CFF, also at colloid concentrations characteristic of natural waters, the established retention profiles for the colloid probes on these systems, when operated under CFR > 15, affirmed the manufacturer-specified cutoff values (in parentheses): 2.4 kDa (1 kDa), 3.3 kDa (3 kDa), 6.1 kDa (5 kDa) and 8.7 kD (10 kDa). A concentration factor (cf) > 10 was found necessary for the reliable determination of the colloidal pool of several elements in the surface waters of the open Baltic Sea. Application of CFR > 15 and cf > 10 returned recoveries around 100% for the studied organic carbon, Ca, Mo, Fe, Cu, and Ni on both systems. Furthermore, the trend of colloid association for the transition elements in the offshore Baltic surface waters followed expectations from their Irving-Williams series of coordination chemistry.