Heat-stable whey protein isolate made using isoelectric precipitation and clarification

Residual lipids (RL) in whey protein isolate (WPI) are detrimental to optimal functional applications (e.g., foaming and low turbidity) and contribute to off-flavor development during powder storage. The objective of this research was to prepare an experimental WPI by removing RL without using the traditional microfiltration process and compare its properties with commercially available WPI made using microfiltration and some other whey powders. We hypothesize that by adjusting the pH of whey to <5.0, we would be close to the isoelectric point of any remaining denatured proteins (DP) and phospholipoproteins (PLP), and therefore reduce electrostatic repulsion between these molecules. Furthermore, demineralization of the acidified whey protein solution by UF combined with diafiltration (DF) should reduce ionic hindrance to aggregation and thereby help with the aggregation of these DP as well as most RL; centrifugation or clarification could be used to remove these materials. Calcium should also be more extensively removed by this approach, which should improve the heat stability of the experimental WPI. Demineralization was achieved on a pilot scale by acidifying liquid (cheese) whey protein concentrate containing 34% protein (WPC-34) to pH 4.5 using HCl, and UF of the whey protein solution along with extensive DF using acidified (pH ∼3.5) reverse osmosis filtered water. Demineralized whey protein solution was adjusted to various combinations of pH (4.1-4.9), conductivities (500-2,000 μS/cm), and protein concentrations (1%-7%) and then centrifuged at 10,000 × g for 10 min. The effective sedimentation (precipitation) of RL in these treatments was estimated by measuring the turbidity of the supernatants. Maximum precipitation was observed at pH 4.5 to 4.7. Reducing conductivity via UF/DF increased the precipitation of RL due to reduced ionic hindrance to aggregation. Maximum sedimentation of RL was observed at protein concentrations ≤3% because of a higher density difference between the precipitate and serum phase. Analysis by sodium dodecyl sulfate polyacrylamide gel electrophoresis analysis confirmed the sedimentation of phospholipoproteins, caseins, and DP upon isoelectric precipitation at pH ∼4.5, while native whey proteins or undenatured whey proteins remained soluble in the supernatant, unaffected by the pretreatment. To scale up the process, 750 L of fluid WPC-34 was acidified and demineralized by UF (volume concentration factor = 1.35) and DF until the permeate solids reached 0.1% (when desired demineralization was achieved), clarified using a pilot-scale desludging clarifier to remove RL, neutralized, ultrafiltered to concentrate the protein, and then spray-dried to produce an experimental WPI (91% protein and 1.8% fat on a dry basis [db]). In another trial, demineralized UF concentrate was clarified by gravity sedimentation and the supernatant was neutralized, ultrafiltered, and spray-dried to produce a second experimental WPI (91% protein and <1% fat db). These experimental WPI powders were compared with several commercially available WPI powders to assess functional properties such as solubility, heat stability, foamability and foam strength, gelation, and sensory attributes over accelerated storage. Experimental WPI had excellent functional properties, had low turbidity, were highly heat stable, and only developed very slight-to-slight off-flavors upon accelerated storage, and their properties were comparable to the WPI manufactured commercially using microfiltration even after accelerated storage.