Lactose-free milk is growing, too. Many processors are trying to capitalize on this trend, as it may help keep lactose-sensitive consumers drinking milk, rather than gravitating to the alternative sector. Turner Dairy Farms, Pittsburgh, for example, is rolling out Turner’s Fresh Lactose-Free Milk in whole and 2 percent fat varieties.
Organic, grass-fed and probiotic also are ways milk processors are trying to differentiate and keep consumers drinking milk. Minneapolis-based Kemps, a wholly owned subsidiary of Dairy Farmers of America, recently introduced milks designed to support healthy digestion. The half-gallon cartons of fat-free and 2 percent reduced-fat milk contain Bifidobacteria and Lactobacillus acidophilus.
Filtration makes innovation possible
Dairy processes account for the vast majority of membrane filtration applications in the food industry, and concentration of the protein in milk and cheese whey typically involves ultrafiltration, microfiltration and reverse osmosis (RO). By concentrating and fractionating milk solids, membrane filtration has made available a number of industrial ingredients and new products, such as Fairlife, the premium-priced milk beverage from Coca-Cola and Select Milk Producers, a group headed by Michael McCloskey of Fair Oaks Farms.
Beginning in 2004, Select Milk has produced milk beverages with high levels of protein and calcium and reduced lactose by using ultrafiltration, RO and an enzymatic process. But the recovery of whey protein in concentrations of 80 percent (WPC80) marks membrane filtration’s greatest impact, according to the American Dairy Products Institute’s Veronique Lagrange, who calls WPC80 “the workhorse for the industry.”
The Elmhurst, Ill.-based institute (www.adpi.org) represents the interests of producers of dry milk, whey products and evaporated milk. Lagrange, the institute’s strategic development consultant, points out that filtration technology has been used in dairy production since the 1970s, but continuous technological improvement has enabled processors to better apply it to boost throughput and yield.
The larger pores in microfiltration systems for milk help retain larger proteins as well as casein and fat. The resulting retentate can increase yield and reduce whey generation in cheese production, she says. “You recover the whey from the milk instead of afterward.” Taste improvement also has been noted, though current federal regulations limit the retentate’s use to cheeses without a standard of identity, such as pizza cheese.
Unlike plant proteins, dairy proteins are well suited for separation by filtration, a relatively benign process that fits nicely with the clean label movement. Conversely, extraction of plant proteins may require thermal or chemical separation. The drawback with filtration is the amount of energy required. The possibility of reducing energy inputs is attracting interest in forward osmosis (FO).
FO requires operating pressure of up to 65 bar, or approximately 943 psi. FO, on the other hand, relies on the higher osmotic pressure of a draw solution to pull a feed stream across a membrane. In trials conducted by Shanti Bhushan, process development engineer at GEA Systems North America (www.gea.com), Hudson, Wis., a lactose draw solution drew a whey feed at an operating pressure of 0.5 bar, about 7.2 psi, a slight vacuum.
Membrane fouling had relegated forward osmosis to lab and experimental status. However, GEA is using a membrane developed by Porifera Inc. (www.porifera.com), a Hayward, Calif., firm established in 2009 to meet the U.S. Dept. of Defense’s DARPA desalination challenge. The team developed a FO membrane that could desalinate seawater mixed with algae and sand without clogging.
GEA engineers have invested two years of R&D work on developing a sanitary FO filtration system. The culmination of that work is a skid-mounted unit that is being transferred from Europe to the Hudson pilot plant for customer trials.
A comparable RO system would require three pumps, whereas the FO system uses a single, smaller feed pump, according to Bhusham. As for fouling, “it won’t be worse,” he says, adding fouling is more a function of the precision of pretreatment.
In the WPC80 process, RO filtration may be followed by freeze concentration, an expensive process that isn’t needed with FO. Proponents maintain operating costs for FO are lower than RO, though Bhushan cautions energy reductions are contingent on the availability of a draw solution that does not require a separate regeneration step.
For some applications, operating temperatures are less than 15C/59F, resulting in improved product quality.
Of the potential applications identified by GEA, a draw solution of cheese brine provides the greatest differential in osmotic pressure to a feed solution of whey protein isolate 90 and WPC80. Mother liquor from lactose production also provides a large osmotic pressure differential. As a practical matter, the highest concentration of total solids from FO is limited to 40 percent.
“If there is waste energy or heat energy that is not being used in the factory, FO makes a case,” Bhushan concludes. “It will be more economical only if the right conditions exist.”
Over-engineering is a rabbit hole every manufacturer struggles to avoid, dairy processors included. As quality demands increase, more equipment often is added until the process sputters instead of purrs. Milk powder production provides an example.
Thermophilic and thermoduric spore-formers inevitably survive the milk pasteurization process, and removing the water only serves to concentrate them. Until a few years ago, 1,000 colony forming units (CFU) per milligram was the standard for high-grade milk powder, but customers now are demanding levels of 500 CFU or lower. Attacking the problem at the farm level is the usual approach, but engineers at Caloris Engineering (caloris.com) in Easton, Md., decided there was room for improvement in the process itself.
“Every solution that was tried seemed to make the process more complicated,” says Bruce Skinner, Caloris’ sales manager. “We took a step back and said, ‘Let’s make this as simple and bullet-proof as we can.”
Twenty hours of production followed by four hours of CIP is the conventional daily routine for evaporators in milk-powder plants. “The evaporator runs good the first half of the day, but by the end of production, it’s fouling more and more” as biofilm develops, Skinner notes. Caloris’s solution: 10-hour runs followed by two-hour CIP cycles.
The process was first used three years ago at an undisclosed facility and again at Canyon, Texas-based Lone Star Dairy Products’ plant, which was commissioned in December 2016. The facility processes 2.5 million lbs. of raw milk per day. Spore counts under 100 CFU are consistently being recorded, according to Skinner.
To avoid starving the downstream spray dryer, Caloris engineers incorporated two small evaporators and a holding tank for the feed stream while CIP is being done. A pre-evaporator builds milk solids to about 36 percent, with a second unit increasing the feed to 50 percent. “We completely decouple upstream processes from the spray drier so we can clean,” Skinner explains.
CIP chemicals tend to run cleaner, he adds, which should reduce the frequency of replacement, although dairies have not reported chemical cost reductions compared to standard production cycles.