Ultra-high temperature (UHT) treated milks have a typical shelf life of 9 – 12 months. Sedimentation in UHT milks is a storage stability problem which reduces customer acceptance. The aim of this research was to understand the mechanisms of sedimentation and finds ways to reduce it.

Two pilot scale UHT plants were used for producing UHT milks. Both plants were capable of direct heating with steam or indirect heat by heat exchange. Sedimentation in both direct and indirect UHT milks was studied at different storage intervals for up to 25 weeks. The UHT milks were stored at 20 °C in a dark temperature controlled room. Sediment weight, pH, ionic calcium, micelle size and zeta potential were measured at different intervals. Sediment composition was analysed using chemical and PAGE techniques.

The UHT sediment was mostly protein and minerals. Sediment contained mostly κ-depleted caseins and some whey proteins. Indirect sediment had more whey protein that direct. Based on literature and obtained results a four step mechanism was hypothesized. During UHT treatment casein micelles are sterically destabilized due to κ-casein dissociation (Step 1 destabilization). During UHT treatment, β-lactoglobulin denature and associate with casein micelles and restore some of the lost steric stability (Step 2 stabilization). These modified casein micelles aggregate together by ionic calcium bridging (Step 3 aggregation) during storage and then settle to the bottom of the container (Step 4 settling).

To validate the mechanism multiple trials were conducted using the two different UHT plants. Sediment weight was measured after 4 weeks of dark storage at 20 °C. Different physicochemical properties of the milk post UHT were measured - pH, ionic calcium, ionic conductivity, whey protein denaturation using HPLC, κ-casein dissociation and β-lactoglobulin association using PAGE, particle size distribution using a Malvern Mastersizer, micelle size and zeta potential using a Malvern Zetasizer, and water of hydration of micelles.

To isolate the effect of temperature from other phenomenon during direct steam injection (60 °C s-1 heating rate, steam shear, liquid-vapour interface and steam bubble cavitation) destabilization and stabilization was studied by heating milk from 4 °C to 70 °C.

Destabilization was studied by attempting to create UHT milks with different amount of κ-casein on micelle surface post UHT. Two techniques were used: 1) heat treatment at different pH values, 2) transglutamination at different temperatures.

Stabilization was studied by increasing β-lactoglobulin association with casein micelles by 1) changing casein to whey protein ratio (70:30, 75:25, 80:20) by adding whey protein isolate, 2) modifying direct UHT process to have either a 5 minute pre heating step or an additional 5 minute post-flash heating step at temperatures above which whey protein denaturation occurs (80, 90 and 100 °C). A mixed UHT (different combinations of direct and indirect steps) trial was done to find which part of indirect UHT process results in increased stability of the casein micelles against sedimentation.

Aggregation was studied by changing ionic strength of milk using monovalent and divalent cation salts to test if the ionic strength or ionic calcium bridging is more important in sedimentation.

It was concluded that destabilization in the UHT process cannot be reduced. Due to the given heating profile of a direct and indirect UHT, destabilization adversely affects stabilization in direct UHT process but not in indirect.

It was also concluded that stabilization does not depend on just the amount of β-lactoglobulin associating with the micelles but the conformation of the β-lactoglobulin attaching to the micelles is also important. It was concluded that the conformations of β-lactoglobulin attaching to micelles during indirect heating imparts better stabilization of the micelles than indirect cooling.

It was concluded that ionic calcium bridging and not ionic strength leads to aggregation of the casein micelles.

Overall the mechanism holds true, but is more nuanced than originally hypothesized. Destabilization affects stabilization. Stabilization also depends on the conformation of β-lactoglobulin polymers associating with micelles.

Direct UHT process can not be modified to reduce sedimentation sufficiently. Slow heating in indirect UHT is found to impart stability to the casein micelles against sedimentation.