Optimal Design of PHAs Plants with Alternative Substrates

FERNANDO D. RAMOS; CLAUDIO A. DELPINO; MARCELO A. VILLAR; M. SOLEDAD DIAZ

Pittsburgh

Congreso; AICHE ANNUAL MEETING 2018; 2018

In this work we proposed a mixed integer nonlinear programming (MINLP) model for the optimal design of apoly(hydroxyalkanoate)s (PHAs) biorefinery. PHAs are a family of biodegradable polyesters synthesized by certainmicroorganisms as an energy reserve. PHAs mechanical and thermal properties together with its biodegradable abilities makethem promising candidates for sustainable polymer production (Dietrich et al., 2017). The proposed PHAs production processpresent different technologies embedded within a superstructure for the three main stages: raw materials pretreatment,biosynthesis and biopolymers extraction and purification.Different potential raw materials are considered as carbon source for pretreatment stage, namely, glycerol, starch andsugarcane. Crude glycerol (glycerol 60.05 wt %, methanol 22.59 wt %, water 10.00 wt %, ashes 2.80 wt %, sodiummethoxide 2.62 wt %, soaps 1.94 wt %) can be fed to the biosynthesis stage or it can be purified to improve PHAproductivity (Ramos et al., 2017). If glycerol purification step is selected, methanol can be sold as a co-product afterdistillation. Also selling purified glycerol is set as a possible alternative (García Prieto et al., 2017). Three potentialprocesses are considered for starch obtention. The first one includes starch production process from corn, the secondoption involves the acquisition of corn starch from market and the third one contemplates the possibility of using cassavastarch. In this point, the superstructure considers the possibility of obtaining glucose through the liquefaction andsaccharification of starch, to be used as a substrate for the microbial growth in the PHAs production step. Furthermore, asugarcane-based process for sucrose production is included in the propose superstructure. Sugarcane bagasse, animportant residue from sugar industries, can be processed for thermal and electrical energy co-generation. Sugarcane juice(mainly sucrose) can be used as a carbon source for PHAs bioproduction. As alternatives to the mentioned process, thepossibility of buying processed sucrose o sugarcane molasses are included in the model. Concerning the biosynthesis, thetechnologies involved in this stage are highly dependent of the selected substrate. They include a sterilization step and twobioreactors, one for biomass growth and the other one for biopolymer accumulation. Finally, four extraction and purificationalternatives are considered in the PHA production process: use of enzymes, solvent, surfactant and NaOCl, or surfactantand chelate.The proposed superstructure is formulated as an MINLP problem and implemented in GAMS (Brooke et al., 2013) for netpresent value (NPV) maximization. Model equality constraints include mass and energy balances, yield equations anddetailed capital cost for process equipment, while inequality constraints include process and product specifications andoperating bounds on process units. The resulting MINLP model for the production of 10,000 t/y of PHA has 8,249continuous variables, 25 discrete variables, 7,456 constraints and it is solved using DICOPT (CONOPT and CPEX)(Grossmann et al., 2003) in a CPU time of 14,625 s. The optimal configuration includes the use of sugarcane as rawmaterial for the production of PHA employing the enzymatic extraction and purification method. The value of the objectivefunction results NPV= 75.01 MM$. The biopolymer production cost is 3.02 $/kg, which is in concordance to worldwide PHA production cost (nearly 3 $/kg) reported by Koller et al. (2012). The energy consumption for the PHA production is 22,56MJ/kg, which results similar to other PHAs plants presented by Lopez-Arenas et al. (2017) and Akiyama et al. (2003). Also,other profitability indexes are calculated. In this sense, considering a discount rate of 10 % and a project lifetime of 15years, the return on investment (ROI) of 22.36 %, the payback period (PP) of 3 years, and the internal rate of return (IRR) of52.53 % indicate the economic profitability of the project. Finally, we performed a sensitivity analysis in order to point out thetechnological aspects that could be improved to achieve a higher profit on the biorefinery.