The fractionation and concentration of whey protein and casein streams from skimmed milk

Abstract

Food processing systems face the challenge to feed a human population of 7.7 billion (2019) people that is expected to increase to 9.7 billion by 2050 (UN, 2015). Food production carries lots of environmental impacts, from the water used to grow crops and feed animals to the impacts of fertilisers on the water itself. Dairy products are no exception to that since they are the second main source of environmental impacts for the food sector, due to their total amount of product rather than the intrinsic impacts per unit of product, when compared to meat.
Dairy products include a wide range of products from milk and cheese to proteins and immunoglobulins. In order to separate the milk into its constituents (proteins, water, immunoglobulins, etc) several methods have been developed over the years, ranging from chemical precipitation to membrane filtration. Membrane filtration has gained special attention among the key industrial players because of membranes long lifespan, their versatility and their low impact on the food product due to the physical method employed for the separation.
However, milk fractionation to separate the different components, specially to separate the different protein groups (casein and whey proteins) still has some challenges that need to be overcome, such as the full separation of casein and whey protein groups using polymeric membranes. In this regard, this project studied the main operational factors affecting such separation, in order to achieve the full separation of casein and whey protein groups using polymeric membranes. There is great interest in industry in having pure streams of whey and casein proteins because they find applications in routinely used products such as infant formula or for the production of imitation cheeses.
This PhD thesis starts with a general introduction (Chapter 1) covering the key aspects of dairy technologies, proteins production and main uses. Furthermore, it also overviews the current challenges that need to be overcome, leading to the establishment of the project aims and objectives.
An extensive literature review (Chapter 2) was included to cover the properties of milk and milk proteins, with an special focus on casein micelles due to their major role in milk protein fractionation associated to micelle composition variations with temperature. The literature review also analyses the key analytical methods for milk proteins detection and quantification to determine the best analytical tool for the project. It can be highlighted that the High-Performance Liquid Chromatography (HPLC) methods have made great advances in the last 10 years making them one of the best approaches for milk protein detection nowadays.
Chapter 3 explains the main experimental work performed in this thesis, particularly on the filtration protocol, including the cleaning steps and the filtration equipment used for milk protein separation.
Chapter 4, 5 and 6 cover the results of this thesis. In particular, Chapter 4 covers the development of the selected analytical method using HPLC, Chapter 5 the study of the interactions between skimmed milk and the Polyvinylidene difluoride (PVDF) membrane and Chapter 6 the study of protein fractionation under a selection of industrial conditions.
The development of the analytical method described in Chapter 4 was carried out to have a tool to identify the main protein groups within milk. Interestingly, this method can identify and quantify in a single HPLC injection the alpha, beta and kappa casein proteins and the alpha-Lactalbumin, beta-Lactoglobulin and Bovine Serum Albumin of the whey proteins. Furthermore, the advancements of the project allowed us to use this method even for measuring fouling layer samples which allowed us to determine protein fouling layer compositions, a key factor for understanding membrane filtration processes.
The study of the milk and PVDF membrane interactions described in Chapter 5, focused on understanding the operational conditions that defined the protein separation. It must be highlighted that when fixing the crossflow velocity and transmembrane pressure casein rejection was controlled by two factors, outlet pressure and temperature, having higher casein rejection at higher pressure and having higher protein rejection at low temperatures (10 °C). Under the same conditions the whey proteins rejection behaviour was less pronounced which indicated that their rejection was controlled by temperature but not by outlet pressure.
Finally, Chapter 6 highlights the final experiments focused on studying the behaviour of proteins and the fouling layer composition under industrial conditions followed by slight modifications to test how pressure, crossflow velocity and temperature could improve the current industrial conditions. The results showed that the current low-pressure conditions in industry favours the fractionation of proteins. Transmembrane pressure increases from 1 bar to 2.8 and 4.7 bar only favoured casein proteins transmission reducing the final fractionation of proteins. The membrane fouling composition study revealed that at any operational condition the main components of the fouling layer were the casein proteins (>94% of all proteins) with a residual share of whey proteins (<6%).
The final chapter of this PhD summarises the main conclusions and compiles an overview of future research pathways to be carried out both in industry and academia.

Date of Award	17 Nov 2021
Original language	English
Awarding Institution	University of Bath
Supervisor	Michael Bird (Supervisor), Barbara Kasprzyk-Hordern (Supervisor), John Chew (Supervisor) & Jannis Wenk (Supervisor)