Marie Wahlgren

Protein adsorption to solid surfaces

The phenomenon of protein adsorption to solid surfaces affects the performance of many materials and processes, in areas ranging from medicine to biochemical engineering. Controlling protein adsorption, from solutions of single proteins as well as from more complex mixtures, requires an understanding of the mechanism(s) by which it occurs. This, in turn, entails detailed characterization of both the protein and the solid surface and identification of those factors controlling the adsorption process.

Membrane Characterization by the Contact Angle Technique: II. Characterization of UF-Membranes and Comparison between the Captive Bubble and Sessile Drop as Methods to obtain Water Contact Angles

Several commercial UF-membranes are characterized by water contact angles. Two different methods of contact angle measurements, the sessile drop and the captive bubble methods, are compared. Differences in results between the two methods occur for the DDS GR 61 membrane; otherwise the methods give rather consistent results. The advancing and the receding contact angles are measured for the UF-membranes. All membranes show considerable contact angle hysteresis. The cellulose acetate and the polyacrylonitrile membranes show less hysteresis than the polysulfone and the polyolefine membranes. The order of hydrophobicity according to advancing contact angle is polyolefine > polysulfone > cellulose acetate > polyacrylonitrile. The contact angle data are compared with Hansens solubility data. Some correlation between θr and δtotal can be seen. θa correlates to the hydrogen and the polar part of the solubility parameter.

Contact angles of ultrafiltration membranes and their possible correlation to membrane performance

The captive bubble method was used to describe the wetting characteristics of a number of commercial ultrafiltration (UF) membranes. The membranes belonged to one of two homologous (same material, different cut-off) series made by the same manufacturer. One series was made of polysulphone and the other one of cellulose triacetate. The porosimetric characteristics of the membranes have also been measured. The combined data were used to explain the fouling behaviour of the membranes upon ultrafiltration of solutions containing dextran, whey protein concentrate and silicate sols. The cellulose triacetate series is characterized by lower receding contact angle and smaller contact angle hysteresis and shows better flux behaviour (permeate flux during UF and pure water flux recovery at the end of UF) than the polysulphone series. Within the same series the mean permeability pore size shows a better correlation with membrane flux behaviour than with contact angle hysteresis.

Adsorption of globular model proteins to silica and methylated silica surfaces and their elutability by dodecyltrimethylammonium bromide

The interaction between a cationic surfactant (dodecyltrimethylammonium bromide) and six model proteins adsorbed on to methylated silica and silica surfaces was investigated. The proteins were bovine serum albumin, cytochrome c, β-lactoglobulin, α-lactalbumin, lysozyme and ovalbumin. The adsorption of the proteins at pH 7 and their subsequent removal by surfactant were studied by in situ ellipsometry. The degree of desorption upon dilution and the degree of elutability were compared and no relationship between these parameters could be found, which indicates that the mechanisms behind the two ways of protein removal are quite different. Further, the degree of elutability by surfactant was related to the physicochemical properties of the proteins. It was found that the size, charge, temperature of denaturation and adiabatic compressibility influenced the degree of elutability at the hydrophilic negatively charged silica surfaces for those of the model proteins that were still adsorbed after buffer rinsing. Negatively charged proteins with high denaturation temperatures, indicating high structural stability, did not adsorb on to this surface (ovalbumin) or adsorbed to a very low degree and were desorbed upon rinsing with buffer (β-lactoglobulin). All proteins adsorbed on to the hydrophobic methylated silica and the parameters that seemed to influence the degree of elutability were size and shell hydrophobicity of the proteins.

β-Lactoglobulin fouling and its removal upon rinsing and by SDS as influenced by surface characteristics, temperature and adsorption time

The extensive fouling common in the food industry puts high demands on equipment cleaning. The adsorption of β-lactoglobulin and its removal by the anionic surfactant sodium dodecyl sulphate (SDS) were followed at pH 6.0 using in situ ellipsometry. Hydrophilic chromium oxide and stainless steel together with hydrophobic methylated silica were studied at different temperatures. Differences between chromium oxide and steel were small, while hydrophobic silica showed significantly different initial adsorption kinetics and adsorbed amounts. Also, the temperature-dependence of the amount desorbed upon rinsing as well as of the overall cleanability differed greatly. At around the β-lactoglobulin denaturation temperature, multilayer build-up at the surface was seen, and the cleanability was very low. Of two protein adsorption times employed, the longer resulted, for metal oxide surfaces, in less desorption during rinsing.

The concentration dependence of adsorption from a mixture of β-lactoglobulin and sodium dodecyl sulfate onto methylated silica surfaces

The adsorption from a mixture of SDS and β-Lactoglobulin, 1:5 (), onto a methylated silica surface was studied in situ by ellipsometry. The amounts adsorbed from different concentrations of the mixture, at pH 7, were compared with those adsorbed from the corresponding pure SDS and β-lactoglobulin solutions. At high concentrations of the mixture, where the CMC of SDS is approached or exceeded, the adsorbate was probably dominated by SDS, indicated by similar kinetics and amounts adsorbed as for SDS solution alone. The amount adsorbed increased, in this concentration range, when the system was rinsed with buffer solution. This was probably due to an exchange between SDS and β-lactoglobulin when the system was diluted. At intermediate concentrations of SDS and β-lactoglobulin, the amounts adsorbed from the mixtures increased and reached a maximum. This maximum was observed both before and after rinsing. Before rinsing the adsorbate might have been a mixture of SDS and β-lactoglobulin while after rinsing probably only β-lactoglobulin remained. At low concentrations, larger amounts were adsorbed from the mixture than from β-lactoglobulin solution alone. In this concentration range rinsing caused minor desorption, indicating that SDS is co-adsorbed with the protein, even in those cases were no adsorption from a pure SDS solution was seen. This indicated that the binding of SDS to β-lactoglobulin at low concentrations is stronger than to the silica surface and that this binding facilitated the adsorption of protein. (Less)

Adsorption from lipase-surfactant solutions onto methylated silica surfaces

Abstract In situ ellipsometry was used to study the adsorption/desorption of highly purified lipase from Humicola lanuginosa in mixtures with surfactants, at the solid/liquid interface. The effect of the surfactant was studied both when it was allowed to adsorb in mixture with lipase and when added after lipase adsorption. Silica surfaces, totally or partially methylated, were used and the surfactants were SDS (anionic), C12E5 and a commercial alcohol ethoxylate (AE) (both nonionic). The experiments were carried out so as to simulate a laundry process and the pH throughout the study was kept at 9 by means of Tris-HCl buffer. From the results it was shown that the lipase did not adsorb until the diluted, that is during the rinsing period. This was found for all the lipase-surfactant mixtures in the study. However, the amount of lipase adsorbed was larger after rinsing in mixture with SDS, compared with C12E5. The results from addition of surfactant after lipase adsorption indicated that the lipase was replaced by surfactant. This was found for SDS and C12E5 at both the hydrophobic surface and the surface with intermediate hydrophobicity.

Time and temperature aspects of β-lactoglobulin removal from methylated silica surfaces by sodium dodecyl sulphate

The adsorption of β-lactoglobulin onto methylated silica surfaces and the subsequent protein removal by the anionic surfactant sodium dodecyl sulphate (SDS) were followed using in-situ ellipsometry. Experiments were performed at pH 6.0 in phosphate-buffered saline solution. Parameters varied include temperature, length of time for protein adsorption from solution and surface residence time of β-lactoglobulin. The temperature was kept constant throughout a trial, and the majority of experiments were carried out at a few degrees below the protein denaturation temperature as reported from differential scanning calorimetry studies. β-Lactoglobulin adsorption at high temperatures resulted in aggregation at the surface after a lag phase of several minutes. Varying the protein adsorption time and thus the amount adsorbed while keeping the protein surface residence time fixed did not seem to affect the amount desorbed upon rinsing or the amount eluted by surfactant. For short β-lactoglobulin adsorption times, the adsorbed amounts were comparable at all temperatures studied. The temperature hardly affected the amount desorbed during rinsing, but did however have a pronounced influence on the protein removed by surfactant. Up to around 60°C practically all β-lactoglobulin was eluted by the SDS. The fraction removed then decreased with temperature, with a sharp drop between 70 and 73°C, and a further decline at higher levels. SDS was seen to be highly inefficient at removing β-lactoglobulin adsorbed at temperatures above 70°C. The trend observed is attributed to temperature-dependent changes in the protein resident on the surface. The β-lactoglobulin surface residence time was seen to significantly affect the elutability. At short residence times the removal efficiency was comparably high, but decreased with time. However, no significant difference could be detected between two sufficiently long residence times. The behaviour is consistent with the assumption of multiple states of adsorbed proteins, together with slow conformational changes in the adsorbed protein layer.