UC Davis

Plastic products are important commodities on a global scale and have been used extensively in almost every aspect of human life. The massive production and use of conventional plastics have created major concerns, including the unsustainable use of fossil carbon feedstocks, the emissions of greenhouse gases in the manufacturing processes, and the disposal of most of the plastic waste and consequent impacts on the natural environment at the end of the lifecycle. As a response to these problems, there has been a global trend towards alternative bioplastic products that are more environmentally friendly than traditional ones. As one popular alternative, polyhydrolxyalkanoates (PHA) are a family of naturally occurring biodegradable plastic materials used as a replacement for the conventional plastics in various applications. Since PHA has similar thermal and mechanical properties to a broad spectrum of different types of conventional plastics, it has been widely used in many applications including packaging, food service, agriculture, medical supplies, consumer products, chemicals, and environmental services. PHA has an established market in recent decades, however, the current bottleneck for the market growth is its high production costs, which are commonly three to four times higher than other bioplastic and traditional plastic products of similar properties and use in the market. There is an urgent need to develop a bioprocessing system to produce PHA with high quality at low cost. Therefore, the goal of this study was to develop an integrated PHA production system utilizing inexpensive feedstocks including food waste and cheese byproducts, which provides high production efficiency and product quality, and low production cost.

Previous studies showed that Haloferax mediterranei, a wild-type extreme halophilic archaeon, can be a superior candidate as a microbial PHA producer than freshwater microbes, with regards to the utilization of waste or byproduct feedstocks for PHA production. It can provide robust production of Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV), a type of high-value PHA, from various renewable feedstocks. Meanwhile, its high salt tolerance enables natural resistance to contamination introduced by microbes in waste-derived streams, which can reduce the costs from sterilization or pasteurization normally incurred in commercial scale productions. In addition, it offers an easier approach for polymer extraction by salt-free (fresh) water, which consumes less energy and chemicals than current PHA production systems. Food waste and food processing byproducts (such as lactose) have good potentials as feedstocks for PHBV production by H. mediterranei because they are rich in sugars and short-chain carboxylates that can be metabolized to PHBV. Pre-treatment processes including anaerobic fermentation, enzymatic hydrolysis, and membrane filtration are required prior to release and to recover those nutrients from the complex macronutrients contained in the feedstocks. The culture of this strain essentially consists of saline medium, carbon source, nitrogen source, phosphorous source, and trace elements. Besides nutrient conditions, a successful cell cultivation process also requires suitable growth conditions including temperature, salinity, pH, and aeration rate.

With the aim of utilizing food waste as low-cost feedstock to produce poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) by Haloferax mediterranei, the effects of acetate (Ac), propionate (Pr), butyrate (Bu), and the short-chain carboxylates derived from food waste were examined on the microbial growth and PHBV production. Results showed that a mixture of carboxylates yielded a 55% more PHBV than glucose. The food-waste-derived nutrients achieved yields of 0.41 to 0.54 g PHBV/g Ac from initial loadings of 450 mg/L to 1800 mg/L Ac of total carboxylates. The consumption of individual carboxylate varied among the different compositions of the carbon source.

Based on previous findings, this study further tested the feasibility of using real-world feedstocks which contain different sugars and VFAs as the main carbon sources for PHBV production by the microbes. Four different types of feedstocks including food waste hydrolysate, food waste fermentate, whey sugar (lactose), and delactosed permeate were collected, pretreated with different processes, and utilized as the sole substrate to culture H. mediterranei. Before microbial cultivation, food waste hydrolysate and fermentate were processed through microfiltration for the recovery of nutrients and removal of suspended particles. The two cheese processing byproduct streams were subjected to enzymatic hydrolysis to convert lactose into monosaccharides. After pretreatment, each substrate was loaded with two sCOD loadings of 22.5 g/L and 45 g/L to culture H. mediterranei and the effects of substrate type and loading on PHBV production were studied. The results show that with the same substrate loading, the substrate derived from food waste fermentate gave higher final cell mass and PHBV concentrations than food waste hydrolysate; and 45 g/L sCOD resulted in the highest PHBV production (3.92 ± 0.08 g/L) within the same culturing time. Comparing substrates derived from whey sugar and delactosed permeate, with 45 g/L sCOD loading, cells grew much faster in whey sugar hydrolysate. Food waste fermentate resulted in higher HV contents than food waste hydrolysate; and substrate loading influenced the HV content of PHBV derived from delactosed permeate, which may be due to the prolonged growth time.

Since food waste fermentate was proven to be one of the better candidates among the four real-world inexpensive feedstocks, more research was conducted to develop an integrated PHBV production system utilizing fermented food waste from commercial anaerobic digesters as feedstock. The effects of feedstock type, substrate loading, and aeration rate were studied and the optimum levels (FWP 1, 40 g/L sCOD, and 2.5 vvm) was determined to yield the highest productions of cell mass and PHBV. A 6-L benchtop bioreactor system has been designed, constructed, and operated to demonstrate the integrated PHA production system from food waste. Through dynamic monitoring of this bioreactor system, the dissolved oxygen level of cell broth was found to drop dramatically to below 20% sat. during the exponential growth phase with a fixed aeration rate. A strategy of maintaining a higher DO level above 50% sat. resulted in faster cell growth but led to lower final cell densities, final PHBV concentration and yield. A novel approach was developed to recycle and reuse the spent salts within the system, should reduce the costs of purchasing raw materials and wastewater treatment and disposal. The spent saline medium (SSM), after the treatment by H2O2 and concentrated to a brine solution through rotary evaporation, was successfully reused through four consecutive batches, where both 80% and 90% SSM recycling strategies yielded comparable cell growth and PHA production to the original batch. The PHA production system developed in this study has the potential for practical applications in the current bioplastic industry. The additional revenue from PHA production can help promote the circular economy for the current food system, increase the market share of bioplastics, and create incentives for better practice of organic waste management in general.

In addition to food waste, the study has also developed an integrated PHA production system with utilizing whey sugar from commercial cheese making facilities as feedstock. Through the dynamic monitoring in a 6-L benchtop bioreactor system, different aeration strategies led to similar results as food waste that with a higher DO level above 50% saturation resulting in faster cell growth, and higher consumption of galactose, but lower final cell densities and PHBV production. The direct reuse of spent saline medium (SSM) was also successfully achieved through four consecutive batches with various substrate loadings, where both 80% and 90% SSM recycling strategies yielded comparable cell growth, PHBV production and PHBV profile to the original batch. This novel PHA production system has the potential of practical applications in the current bioplastic industry; meanwhile, through utilizing whey byproducts and waste streams, an additional revenue can be provided to the dairy industry.

Based on the system design developed here, economic feasibility assessments were conducted to project integrated industrial scale processing. Byproduct streams from a local cheese plant, with an input of 168.7 metric ton/day (MT/day) lactose (186 US ton/day), were used as the feedstock. SuperPro® Designer (Intelligen Inc., USA) was used as the software for the process design and economic modeling. Three scenarios with different processes for the treatments of used enzyme and spent medium were investigated and the major factors that influence the overall economics were identified. The simulated system produces 9700 MT/year PHBV with an assumed PHA yield of 0.2 g PHBV/g lactose and an overall process efficiency of 87%. The breakeven price was found to be more sensitive to the lactose price than enzyme price. The scenario with enzyme reuse and spent medium recycling achieved the lowest breakeven price among the scenarios, which can be less than 4 $/kg PHA based on a delactosed permeate (DLP) unit price of 0.1 $/kg. The study suggests utilizing dairy derived feedstocks has the potential to make PHA competitive in the bioplastic market, which could be beneficial to both the dairy and bioplastic industries.