Integrated catalysis for upgrading microbial derived carboxylic acids to renewable fuels and value-added chemicals

Vardon, Derek R

Permalink

https://hdl.handle.net/2142/89166 Copy

Description

Title

Integrated catalysis for upgrading microbial derived carboxylic acids to renewable fuels and value-added chemicals

Author(s)

Vardon, Derek R

Issue Date

2015-09-09

Director of Research (if dissertation) or Advisor (if thesis)

Strathmann, Timothy J

Doctoral Committee Chair(s)

Strathmann, Timothy J

Committee Member(s)

• Beckham, Gregg T
• Guest, Jeremy S
• Singh, Vijay

Department of Study

Civil & Environmental Engineering

Discipline

Environmental Engineering in Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• renewable chemicals
• renewable fuels

Abstract

In order to transition to a renewable carbon society, economically and environmentally sustainable technologies are needed to displace our dependence on petroleum. Carboxylic acids are a diverse class of biological metabolites that can be converted to renewable fuels and chemicals to offset our consumption of petroleum. However, significant challenges occur when integrating catalysis with biological processes, which include: (1) biological conversion produces carboxylic acids at relatively dilute levels (<20 wt%) in broth that can contain residual impurities, creating separation and downstream process challenges, (2) microbial acids can contain unique chemical moieties (e.g., polyunsaturated bonds, hydroxyl groups, ester linkages) compared to aliphatic petroleum, requiring tailored catalytic upgrading strategies to produce fuels and chemicals, and (3) carboxylic acid valorization can occur through a multitude of unit process schemes, necessitating early-stage techno-economic analysis to identify key bottlenecks for further development. To address these challenges, this thesis investigates integrated catalysis for upgrading microbial derived acids to renewable fuels and value-added chemicals. To target both renewable fuels and value-added chemicals from microbial acids, the following research objectives were pursued: (1) hydrothermal catalysis was investigated for deoxygenating monocarboxylic acids to diesel-grade hydrocarbons with in situ hydrogen production from renewable organic donors, (2) separation and catalysis was examined for recovering and cis,cis-muconic acid from culture broth and converting it to adipic acid, the latter compound being a high-value polymer precursor for nylon-6,6 production, and (3) key economic drivers and technical targets were identified for the downstream processing of muconic acid to adipic acid using preliminary techno-economic analysis. Initially, hydrothermal catalysis was investigated for converting long chain saturated and unsaturated carboxylic acids to hydrocarbon fuels using a Pt-Re catalyst supported on activated carbon (AC). The addition of Re as a secondary metal was shown to enhance the rate of carboxylic acid deoxygenation and modify the chemisorption behavior of Pt, suggesting alloy formation. Decarboxylation/decarbonylation of the carboxylate group was observed as the primary reaction pathway, and characterization of the Pt-Re/AC catalyst by x-ray photoelectron spectroscopy determined that hydrogen in the headspace resulted in a reduced oxidation state of the metals after exposure to hydrothermal conditions. Lastly, the addition of glycerol as an in situ hydrogen donor proved effective at meeting process hydrogen demands through aqueous phase reforming reactions. The application of the Pt-Re/AC catalyst system was then evaluated using a complex monocarboxylic acid feedstock derived biologically from lignin. The microorganism Pseudomonas putida KT2440 was initially used to biologically “funnel” lignin derived monomers to intracellular medium chain length polyhydroxyalkanoates (mcl-PHAs). Shake flask studies demonstrated that P. putida produces mcl-PHAs from both mixed model compounds and complex lignin derived monomers derived from corn stover. Extraction and characterization of lignin derived mcl-PHAs showed similar physicochemical properties compared to mcl-PHAs produced from clean glucose, and thermal depolymerization readily converted mcl-PHAs to alkenoic acid monomers. Subsequent catalytic processing of alkenoic acids in hydrothermal media with Pt-Re/AC produced linear hydrocarbons, similar to the model compound fatty acid study, demonstrating the integrated biological and catalytic conversion of lignin to hydrocarbon fuel. In order to target value-added chemicals from microbial acids, the downstream separation and catalytic upgrading of cis,cis-muconic was evaluated for the production of adipic acid, the latter molecule being a high-value polymer precursor for nylon-6,6 production. Expanding on previous work, a metabolically engineered strain of P. putida KT2440 was used to produce muconic acid extracellularly from both model and lignin derived monomers. Following fed-batch biological conversion of p-coumaric acid, activated carbon purification was shown to effectively remove broth non-target upstream metabolites, color compounds, and unconverted substrate. Muconic acid was then recovered from culture broth by pH/temperature shift crystallization in high purity (>97%) and yield. Catalyst batch screening studies of commercial Pd, Pt, Ru identified Pd as a highly active metal for muconic acid hydrogenation, although leaching was observed. As a follow-up, the downstream separation and catalysis of muconic acid was further examined to improve the separation purity, evaluate catalyst stability, and demonstrate bio-adipic acid polymerization to nylon-6,6. Following crystallization, dissolution of muconic acid crystals in ethanol with membrane filtration removed insoluble inorganic salts, producing muconic acid at 99.8% purity. In house catalysts were synthesized on both carbon and silica supports and tested in batch hydrogenation screening reactions. Pd and Rh were identified as highly active on both carbon and silica supports when compared to Ru and Pt, although Pd leached significantly, with a greater extent on silica. To further evaluate the stability of Rh/AC, continuous trickle bed hydrogenation demonstrated steady state partial conversion for 48 h, followed by complete conversion until 96 h, with a return to partial steady state conversion for 120 h of time on stream. Characterization of the post reaction Rh/AC catalyst showed a modest increase in support surface area and pore volume, moderate loss in active metal surface area, and minor increase in metal crystallite size. Bio-adipic acid derived catalytically from muconic acid was then polymerized to nylon-6,6, and characterization of the polymer confirmed properties comparable to nylon produced from adipic acid of petrochemical origin. Lastly, preliminary techno-economic analysis was conducted to evaluate key economic drivers and technical targets for the downstream processing of muconic acid to adipic acid. An nth-generation downstream plant was modeled to produce 75 million kg of adipic acid per year. For the base-case process model, the following technical parameters were employed: cell free culture broth containing 50 g/L muconate and 2 g/L of non-target aromatic compounds was purified continuously with activated carbon. Muconic acid was then recovered by pH/temperature shift crystallization, dissolved in ethanol, and filtered to provide a condensed phase for catalytic processing. Muconic acid in ethanol was catalytically converted to adipic acid over a packed bed reactor containing 2%Rh/AC, and a second train of evaporative crystallization with rotary filtration and drying recovered adipic acid as the final product. The largest capital costs for the base case model were activated carbon regeneration kilns and the packed bed hydrogenation reactor. Variable operating costs were comparable throughout, excluding the cost of incoming muconate broth which was the largest variable expense by far. Economic analysis of the base case model determined a minimum selling price of bio-based adipic acid of \$1.90/kg, within the 5-year historical range (\$1.75-2.50/kg) for petroleum derived adipic acid. Lastly, single point sensitivity analysis determined that the broth ratio of muconate to non-target aromatic compounds was a major non-linear cost driver, as well as the required reactor throughput for the 2%Rh/AC catalyst. Overall, this thesis demonstrated that integrated catalysis can convert both model and complex microbial acids derived from lignocellulosic feedstocks to renewable fuels and value-added chemicals. Upstream biological funneling is uniquely suited to address the heterogeneity of complex biomass monomer streams, while tailored separation schemes have potential to produce carboxylate feedstocks of suitable purity for value-added chemical production. The unique functional moieties of microbial acids will require tuned reaction conditions and catalytic formulations depending when targeting renewable fuels and chemicals, while the challenges of substrate acidity, residual impurities, and potentially harsh reactions conditions will require robust catalyst development.

Graduation Semester

2015-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/89166

Copyright and License Information

Copyright 2015 Derek R. Vardon

Owning Collections