Langdon, Alan G.Nisbet, Trevor John2025-09-262025-09-261979https://hdl.handle.net/10289/17670The literature on the properties of stainless steel surfaces, the physico-chemical properties of milk and the surface chemistry of proteins was reviewed to gain an insight into the formation, of protein-mineral films (“milkstone”) on the contact surfaces of dairy milk lines. Protein adsorption processes were considered fundamental to the formation of such deposits. Standard methods of protein separation and purification were employed to obtain purified αₛ₁, β, and K-casein. Radiochemical methods provided a convenient means of studying protein adsorption on to low surface area systems. Of the various methods for labelling proteins, radioiodination by a slightly modified version of Hunter and Greenwood Chloramine T method proved to be the most satisfactory. By this method, stable labelled milk proteins of high specific activity containing not more than one iodine atom per protein molecule and having properties that could not be distinguished from those of unlabelled proteins were prepared. In the course of the work several further aspects of the iodination reaction and the reactions of iodine species with milk components were investigated. Iodination efficiencies were found to be sensitive to the storage time of the radioactive iodine stock solutions (supplied free of reducing agent at pH 8-11). The drop in iodination efficiency was correlated with an increase in the concentration of oxidised iodine species in the radioactive stock solutions. Studies of the iodination of whole milk showed that with the exception of K-casein, the tyrosyl residues of all proteins in milk were iodinated with approximately the same iodination efficiency. The iodination efficiencies of K-casein tyrosyls were significantly lower. A theoretical examination of the effect of the hydrophobicity of the molecular environment on tyrosyl iodination efficiency, indicated the likelihood that the affect observed was due to the incorporation of the K-casein in inaccessible regions of the micelle rather than a molecular environment effect. Iodinated caseins, when added to whole skim milk became bound to the micelle fraction. In the case of β and K-casein, the binding was temperature dependent and at least partially reversible. αₛ₁-casein binding to micelle was largely independent of temperature and irreversible. The stabilising effect of calcium ions on diluted micelle suspensions was demonstrated. The variation in the K-casein content of casein micelles with micelle size was also demonstrated. When inorganic iodine was added to milk, part of this became encapsulated by milk micelles. However, there was virtually no chemical binding of the iodine to the milk proteins of cooled milk. Techniques were developed to study the surface chemistry of radioactively labelled milk proteins at stainless steel-water interfaces. The amount adsorbed was calculated according to the principle of isotope dilution. Autoradiography and high resolution autoradiography, using nuclear track emulsion and scanning electron microscopy confirmed the uniform distribution of the adsorbed protein. Studies of adsorption from whole skim milk, whey and milk serum, indicated that of the various milk proteins serum caseins had the greatest affinity for stainless steel surfaces. Adsorption data for purified caseins and mixtures of casein indicated that β-casein was the species preferentially adsorbed. Generally it was observed that adsorption was largely irreversible and providing adequate time was allowed, proceeded to form a relatively constant close packed film of adsorbed protein molecules. The amount of protein in the films ranged from 2.0 mg m⁻² to 3.0 mg m⁻² irrespective of protein concentration, except at high concentrations of β-casein and to some extent, αₛ₁-casein, where a second adsorption process seemed to be present. This effect also seemed to be present in studies of adsorption from whole milk. The kinetics of adsorption and desorption were monitored by tracer techniques. The data for adsorption from unstirred solutions indicated the process was diffusion controlled. A diffusion model was developed which accounted for the rate of adsorption and the amount adsorbed in terms of the size, shape and hydration of the protein molecule, the viscosity of the medium and the effective surface area of the stainless steel surface. The model was used to estimate hydration and axial ratio values for β-casein. Values of 1.9 grams of water per gram of protein for the hydration, and 11 for the axial ratio (a/b) were obtained. Desorption of adsorbed protein is a complex process that does not obey simple rate laws. Desorption was facilitated by hydroxyl ions and anionic surfactants but no simple relationship between rate of desorption and species concentration was obtained. Simple adsorption process cannot account for the formation of macroscopic milk protein films on milk contact surfaces. After the surface has been covered by the initial adsorbed layer, no further protein could be adsorbed until the surface of this layer was modified in some way, for example, by the adsorption of positively charged colloids of iron oxide, chromium oxide or calcium phosphate. Deposition by evaporation of solution films, or precipitation from solution are other mechanisms by which macroscopic films may be formed. The film thicknesses of the drained milk films formed after immersions of clean stainless steel surfaces into dilute milk solutions remained independent of dilution at 3.8 ± 0.6 x 10⁻⁴ cm. The properties of protein films formed at surfaces can be altered by thermal denaturation, chemical modification and also presumably by bacterial action. Analytical and microscopic examination of actual milkstone deposits generally supported a previous milkstone classification based upon chemical analysis, i.e. fat deposits, high protein, deposits, calcium phosphate milkstone and iron milkstone. As a result of the present investigation the further classification of high protein deposits into chlorine-induced (from sodium hypochlorite) and phosphate-induced (from phosphoric acid) was suggested. It does not seem likely that milkstone deposits are formed by a single mechanism. This conclusion was supported by an analysis of data for deposits formed in simulator trials where evaporation of milk films (in one simulator) and precipitation induced by hypochlorite (in two simulators) were major factors. The variety of possible mechanisms for deposition is due to the range of conditions that can occur in milk lines. The results of the present work will assist in determining the mechanism operative in a particular case and will allow the development of better procedures for the removal and prevention of such deposits.enAll items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.Thesis