MANUFACTURE OF 2,5-FURANDICARBOXYLIC ACID FROM FURFURAL PRODUCED FROM A LIGNOCELLULOSIC WASTE STREAM

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: SMALL BUSINESS GRANT
• Reporting Frequency: Annual
• Accession No.: 1028539
• Grant No.: 2022-33530-37203
• Cumulative Award Amt.: $175,000.00
• Proposal No.: 2022-00839
• Project Start Date: Jul 1, 2022
• Project End Date: Feb 28, 2023
• Grant Year: 2022
• Program Code: [8.1]- Forests & Related Resources

Recipient Organization
K S E INC
665 AMHERST RD
SUNDERLAND,MA 013759420

Performing Department
(N/A)

Non Technical Summary
The overall goal of this Phase I SBIR program is to evaluate, optimize, and economically scale novel technology to convert a lignocellulose waste stream into a value-added intermediate chemical, specifically 2,5-Furandicarboxylic acid (FDCA), at high selectivity and without pre-purification of the biomass compounds. FDCA is targeted to replace petroleum-based acids used extensively in the textiles, polyesters, and plastics industries. The global market as a drop-in replacement for terephthalic acid (PTA) is estimated at 79M metric tons in 2020 with a CAGR of 3.3%.Common sources of lignocellulose include renewable, woody substrates, agri-residues, dedicated energy crops, and municipal solid wastes. One identified feedstock, from forest related resources, being a hydrolysate stream from pulp & paper processing.Project tasks include improving reactive distillation & catalytic conversion of pentose carbohydrates to furfural, converting batch scale to continuous oxidation of furfural to furoic acid, review alternative alkali metal catalyst combinations to economically maximize FDCA yield, implement a closed-loop catalyst regeneration process to control costs, and develop an economic model based on the R&D program to demonstrate overall economic value proposition. A subsequent Phase II program would further enhance, optimize and scale up these technologies to develop an integrated and highy intensified process targeting a pulp mill hydrolysate waste stream for production of a value-added FDCA product.Societal benefits of utilizing FDCA from this novel process include health improvements associated with replacing endocrine disrupting phthalate-based plastics, improved barrier/thermo-mechanical properties of consumer plastics, originating from renewable resources, with process technology demonstrating a more sustainable carbon footprint.

Animal Health Component

50%

Research Effort Categories

Basic

25%

Applied

50%

Developmental

25%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
511	0660	1000	100%

Knowledge Area
511 - New and Improved Non-Food Products and Processes;

Subject Of Investigation
0660 - Paper and pulp derived products;

Field Of Science
1000 - Biochemistry and biophysics;

Keywords

fdca

biobased chemicals

biobased plastic

biomass

lignocellulose

value-added products

woody

Goals / Objectives
The overall goal of this Phase I SBIR program is to evaluate, optimize, and economically scale novel technology to convert a lignocellulose waste stream into a value-added intermediate chemical, specifically 2,5-Furandicarboxylic acid (FDCA), at high selectivity and without pre-purification of the biomass compounds. FDCA is targeted to replace petroleum-based acids used extensively in the textiles, polyesters, and plastics industries. The global market as a drop-in replacement for terephthalic acid (PTA) is estimated at 79M metric tons in 2020 with a CAGR of 3.3%.Common sources of lignocellulose include renewable, woody substrates, agri-residues, dedicated energy crops, and municipal solid wastes. One identified feedstock, from forest related resources, being a hydrolysate stream from pulp & paper processing.Project tasks include improving reactive distillation & catalytic conversion of pentose carbohydrates to furfural, converting batch scale to continuous oxidation of furfural to furoic acid, review alternative alkali metal catalyst combinations to economically maximize FDCA yield, implement a closed-loop catalyst regeneration process to control costs, and develop an economic model based on the R&D program to demonstrate overall economic value proposition. A subsequent Phase II program would further enhance, optimize and scale up these technologies to develop an integrated and highy intensified process targeting a pulp mill hydrolysate waste stream for production of a value-added FDCA product.Societal benefits of utilizing FDCA from this novel process include health improvements associated with replacing endocrine disrupting phthalate-based plastics, improved barrier/thermo-mechanical properties of consumer plastics, originating from renewable resources, with process technology demonstrating a more sustainable carbon footprint.The overall goal of this Phase I SBIR program is to evaluate, optimize, and economically scale novel technology to convert a lignocellulose waste stream into a value-added intermediate chemical, specifically 2,5-Furandicarboxylic acid (FDCA), at high selectivity and without pre-purification of the biomass compounds. FDCA is targeted to replace petroleum-based acids used extensively in the textiles, polyesters, and plastics industries. The global market as a drop-in replacement for terephthalic acid (PTA) is estimated at 79M metric tons in 2020 with a CAGR of 3.3%.Common sources of lignocellulose include renewable, woody substrates, agri-residues, dedicated energy crops, and municipal solid wastes. One identified feedstock, from forest related resources, being a hydrolysate stream from pulp & paper processing.Project tasks include improving reactive distillation & catalytic conversion of pentose carbohydrates to furfural, converting batch scale to continuous oxidation of furfural to furoic acid, review alternative alkali metal catalyst combinations to economically maximize FDCA yield, implement a closed-loop catalyst regeneration process to control costs, and develop an economic model based on the R&D program to demonstrate overall economic value proposition. A subsequent Phase II program would further enhance, optimize and scale up these technologies to develop an integrated and highy intensified process targeting a pulp mill hydrolysate waste stream for production of a value-added FDCA product.Societal benefits of utilizing FDCA from this novel process include health improvements associated with replacing endocrine disrupting phthalate-based plastics, improved barrier/thermo-mechanical properties of consumer plastics, originating from renewable resources, with process technology demonstrating a more sustainable carbon footprint.?

Project Methods
The overall goal of this Phase I Small Business Innovation Research program is to evaluate, optimize, and economically scale novel technology to convert a lignocellulose waste stream into a value-added intermediate chemical, specifically 2,5-Furandicarboxylic acid (FDCA), at high selectivity and without pre-purification of the biomass compounds. Specific objectives of the proposed program include:Enhance catalyst selectivity for the hydrolysis and dehydration of pentose compounds to improve furfural yield from a representative lignocellulosic streamConvert furfural oxidation from bench-scale to continuous operation critically reviewing catalyst substrate and recovery process, optimal caustic species to maximize cost-efficiency of the process, including an effective drying process critical to successful carboxylationImprove furoate carboxylation targeting reduced costs in the form of less expensive alkali metal catalysts and/or reaction conditions which enable reduced residence timeImplement an alkali metal catalyst recovery and regeneration process to enable a continuous, cost-effective processingDevelop a techno-economic model for demonstration of the value-proposition relative to competing technologies

Progress 07/01/22 to 02/28/23

Outputs
Target Audience:Our objectives thru this program are to manufacture a renewable, carbon-negative, healthier alternative to petroleum-based terephthalic acid (TPA). Our manufacture of furandicarboxylic acid (FDCA) as that replacement offers a healthier alternative to TPA manufactured consumer goods. This FDCA replacement replaces phthalate laced TPA, offering a significantly decarbonized pathway to meet explosive global plastics demand while drastically reducing GHG emissions in the process. Our FDCA will provide a cost-effective and environmentally sustainable process serving the global good. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed us to hire two summer student interns, training them valuable lab, research and communication skills, better preparing them for evaluating and entering the work force after graduation. It has also allowed us valuable analytics evaluation training, utilizing the lab analytics space at the University of Massachussetts to fine tune and better our analysis of finished FDCA product. How have the results been disseminated to communities of interest?Our public reporting has been sent the USDA-NIFA REEporting system for public sharing and evaluation. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Task 1:Significant progress has been made adapting the reactive distillation process from an aqueous-based system toward a bi-phasic system with resultant improvement in yields from ~50% to upwards of 80% for xylose-to-furfural. However, opportunity also exists for the solvent-based reactive distillation system to extend to capture hexose feeds, at a minimum to target HMF production for a dual FDCA pathway, but also potentially conversion to furfural which could triple potential furfural production through the proposed furanic carboxylation pathway. A prospective run expanding the xylose reaction phase/extractive phase setup was completed for fructose which demonstrated reasonable performance at 100% conversion and 57% selectivity (AR28Y1, Table 5). This can be compared to the prior ionic liquid system where selectivity of 75-80% to HMF was achieved. Although not initially as effective it leaves the prospect open that hexose dehydration will proceed in the THF system and further study could be performed in Phase II to drive both glucose and xylose to furfural with a common dehydration reaction setup. Figure 4 represents the new process configuration which will be deployed based on the experimental program. A notable implication being the THF reactive phase significantly reduces the energy intensity of the furfural recovery process. The initial design conducted the xylose dehydration reaction in aqueous phase and, with the heat capacity of 4.187 kJ/kg K for water, significant energy is required to vaporize both water and furfural overhead as in Figure 1. With a 2-phase system, a simple decantation step can be inserted to remove waste water and with the heat capacity of THF being 60% lower at 1.77 kJ/kgK, significant energy savings can be achieved by THF recovery overhead with furfural recovered in the bottoms for processing in the subsequent oxidation reaction block. Task 2: There are several factors that ideally control the overall selectivity for furoic acid in this reaction sequence, and all of them have to be balanced in order to achieve high conversion and selectivity. The first need to be met is the molar flow ratios of aldehyde to hydroxide. The proper stoichiometric ratio must be maintained during the operation of the reaction system, if the concentration of furfural is too high then the conversion suffers, if the MOH flow rate is too high then the disproportionation reaction is favored over the oxidation reaction. Beyond this, maintaining the proper diffusion of furfural in the system is of high priority, followed by maintaining the proper reaction temperature range of 50-700C. The result of using 50wt% CsOH feed solutions is that the nominal flow rates for the feed need to be adjusted. Without adjustment, the molar flow ratios become out of alignment, favoring furfural which has an appreciable effect on the steady state conversion and selectivity. Moreover, because of the lack of diffusion in the system, the aforementioned disproportionation reaction is favored with an emphasis on the production of furfuryl alcohol and other byproducts i.e. resins and the like. This eliminated the diffusion issues associated with higher concentration solutions, allowing for increased hydrate formation and ultimately a reduction in total side products formed netting an 88% selectivity for target furoic acid. Improvements to the system can be explored in a Phase II program, specifically targeting the reduction in furfuryl alcohol side products and increasing the total furoic acid yields obtained. Scaling of the reaction system to larger reactors and more concentrated reagent streams will be fruitful in making a commercially viable process. Task 3: The process was scaled by a factor of 10 to 5L utilizing a Parr Series 4848 reactor. Select runs in Table 9 demonstrate the progression where higher moisture contents of the feed while scaling the drying processes were initially detrimental to reaction yields, particularly as moisture content rose above 5wt% (ChemAP66Y). However, successful application of pre-feed drying processes achieved multiple runs demonstrating furoic acid conversion ~80% with selectivity for FDCA 80-90%. Overall, successful scale-up was achieved with FDCA production ramping from ~10g scale to 100g scale with minimal impact on performance demonstrating the efficacy of the novel process. Regenerated carbonate from Task 4 was utilized as well as furoate from the oxidation reaction in Task 2 for carboxylation, Table 10. Regenerated carbonate conversion was on par with reagent-grade carbonate, but selectivity eroded ~20-30pps. This batch of carbonate was associated with ion-exchange runs taking Cl- ion concentration down only to 15 g/L. Subsequent carbonate batches were taken to < 5 g/L but results are comparable (AQ90W, AQ93Y). Oxidation production yields an effluent stream with unconverted furfural, some furfuryl alcohol, and unreacted CsOH. Ether extraction was performed to remove soluble impurities and the stream was run through a CO2 gas absorption column (see Task 4) which converted any remaining CsOH to cesium carbonate, a catalyst in the carboxylation reaction. The results showed effective performance as portrayed in ChemAQ94Y with furoic acid conversion of 81% and FDCA selectivity of 78%. Task 4:With pilot-scale ion bed in place, cesium chloride filtrate from the pilot acid precipitation reactor was processed with results summarized in Table 12. Initial pilot-plant chloride concentrations were higher than modeled (>125 g Cl-/L) as the acid precipitation was being conducted in excess to ensure complete FDCA precipitation. Additionally, higher feed rates were attempted and with higher feed ion concentration, effluent concentration was initially higher than desired (~50 g Cl-/L) with more frequent bed regeneration required. The process was tuned back to lower feed rates and careful attention to ion breakthrough (reflecting spent resin) and effluent concentrations were steadily reduced to low single-digit chloride ion concentration in the effluent. Significant milestones achieved were not only tuning operating parameters but demonstrating highly acidic, high ion concentration, pilot-plant precipitation product could effectively be converted to CsOH. Additionally, the same ion resin batch was utilized demonstrating reliable regeneration capability used for economic modelling. Task 5:The intensive, catalytic processing of select reactive intermediates enables KSE to achieve significant capital cost savings versus competitors. With estimated total capital investment (TCI) of $43M for a 60M lb/yr plant, this would be 25% of the cost published by Avantium for a plant 5 times smaller (5Ktpa or 11M lb/yr). Closed-loop recycle of expensive catalyst and extraction solvents enable these costs to be capitalized in working capital included in the TCI. Furthermore, by utilizing waste streams as the source for xylose, considerable improvement in cost of goods sold are demonstrated. Use of conservative FDCA sales value at $0.85/lb provide sufficient cushion for potential capital or feedstock cost escalation while still yielding attractive IRR, ROIC, and operating margin metrics.

Publications

Progress 07/01/22 to 02/28/23

Outputs
Target Audience:Our objectives as a result of this program are to manufature a renewable, carbon-negative, healthier alternative to petroleum-based terephthalic acid (TPA). Our manufacture of furandicarboxylic acid (FDCA) as that replacement offers a healthier alternative to TPA manufactured consumer goods. This FDCA replacement replaces phthalate laced TPA, offering a significantly decarbonized pathway to meet explosive global plastics demand while drastically reducing GHG emissions in the process. Our FDCA will provide a cost-effective and environmentally sustainable process serving the global good. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed us to hire two summer student interns, training them valuable lab, research and communication skills, better preparing them for evaluating and entering the work force after graduation. It has also allowed us valuable analytics evaluation training, utilizing the lab analytics space at the University of Massachussetts to fine tune and better our analysis of finished FDCA product. How have the results been disseminated to communities of interest?Our public reporting has been sent the USDA-NIFA REEporting system for public sharing and evaluation. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Task 1: Significant progress has been made adapting the reactive distillation process from an aqueous-based system toward a bi-phasic system with resultant improvement in yields from ~50% to upwards of 80% for xylose-to-furfural. However, opportunity also exists for the solvent-based reactive distillation system to extend to capture hexose feeds, at a minimum to target HMF production for a dual FDCA pathway, but also potentially conversion to furfural which could triple potential furfural production through the proposed furanic carboxylation pathway. A prospective run expanding the xylose reaction phase/extractive phase setup was completed for fructose which demonstrated reasonable performance at 100% conversion and 57% selectivity (AR28Y1, Table 5). This can be compared to the prior ionic liquid system where selectivity of 75-80% to HMF was achieved. Although not initially as effective it leaves the prospect open that hexose dehydration will proceed in the THF system and further study could be performed in Phase II to drive both glucose and xylose to furfural with a common dehydration reaction setup. Figure 4 represents the new process configuration which will be deployed based on the experimental program. A notable implication being the THF reactive phase significantly reduces the energy intensity of the furfural recovery process. The initial design conducted the xylose dehydration reaction in aqueous phase and, with the heat capacity of 4.187 kJ/kg K for water, significant energy is required to vaporize both water and furfural overhead as in Figure 1. With a 2-phase system, a simple decantation step can be inserted to remove waste water and with the heat capacity of THF being 60% lower at 1.77 kJ/kgK, significant energy savings can be achieved by THF recovery overhead with furfural recovered in the bottoms for processing in the subsequent oxidation reaction block. Task 2: There are several factors that ideally control the overall selectivity for furoic acid in this reaction sequence, and all of them have to be balanced in order to achieve high conversion and selectivity. The first need to be met is the molar flow ratios of aldehyde to hydroxide. The proper stoichiometric ratio must be maintained during the operation of the reaction system, if the concentration of furfural is too high then the conversion suffers, if the MOH flow rate is too high then the disproportionation reaction is favored over the oxidation reaction. Beyond this, maintaining the proper diffusion of furfural in the system is of high priority, followed by maintaining the proper reaction temperature range of 50-700C. The result of using 50wt% CsOH feed solutions is that the nominal flow rates for the feed need to be adjusted. Without adjustment, the molar flow ratios become out of alignment, favoring furfural which has an appreciable effect on the steady state conversion and selectivity. Moreover, because of the lack of diffusion in the system, the aforementioned disproportionation reaction is favored with an emphasis on the production of furfuryl alcohol and other byproducts i.e. resinsand the like. This eliminated the diffusion issues associated with higher concentration solutions, allowing for increased hydrate formation and ultimately a reduction in total side products formed netting an 88% selectivity for target furoic acid. Improvements to the system can be explored in a Phase II program, specifically targeting the reduction in furfuryl alcohol side products and increasing the total furoic acid yields obtained. Scaling of the reaction system to larger reactors and more concentrated reagent streams will be fruitful in making a commercially viable process. Task 3: The process was scaled by a factor of 10 to 5L utilizing a Parr Series 4848 reactor. Select runs in Table 9 demonstrate the progression where higher moisture contents of the feed while scaling the drying processes were initially detrimental to reaction yields, particularly as moisture content rose above 5wt% (ChemAP66Y). However, successful application of pre-feed drying processes achieved multiple runs demonstrating furoic acid conversion ~80% with selectivity for FDCA 80-90%. Overall, successful scale-up was achieved with FDCA production ramping from ~10g scale to 100g scale with minimal impact on performance demonstrating the efficacy of the novel process. Regenerated carbonate from Task 4 was utilized as well as furoate from the oxidation reaction in Task 2 for carboxylation, Table 10. Regenerated carbonate conversion was on par with reagent-grade carbonate, but selectivity eroded ~20-30pps. This batch of carbonate was associated with ion-exchange runs taking Cl- ion concentration down only to 15 g/L. Subsequent carbonate batches were taken to < 5 g/L but results are comparable (AQ90W, AQ93Y). Oxidation production yields an effluent stream with unconverted furfural, some furfuryl alcohol, and unreacted CsOH. Ether extraction was performed to remove soluble impurities and the stream was run through a CO2 gas absorption column (see Task 4) which converted any remaining CsOH to cesium carbonate, a catalyst in the carboxylation reaction. The results showed effective performance as portrayed in ChemAQ94Y with furoic acid conversion of 81% and FDCA selectivity of 78%. Task 4: With pilot-scale ion bed in place, cesium chloride filtrate from the pilot acid precipitation reactor was processed with results summarized in Table 12. Initial pilot-plant chloride concentrations were higher than modeled (>125 g Cl- /L) as the acid precipitation was being conducted in excess to ensure complete FDCA precipitation. Additionally, higher feed rates were attempted and with higher feed ion concentration, effluent concentration was initially higher than desired (~50 g Cl- /L) with more frequent bed regeneration required. The process was tuned back to lower feed rates and careful attention to ion breakthrough (reflecting spent resin) and effluent concentrations were steadily reduced to low single-digit chloride ion concentration in the effluent. Significant milestones achieved were not only tuning operating parameters but demonstrating highly acidic, high ion concentration, pilot-plant precipitation product could effectively be converted to CsOH. Additionally, the same ion resin batch was utilized demonstrating reliable regeneration capability used for economic modelling. Task 5: The intensive, catalytic processing of select reactive intermediates enables KSE to achieve significant capital cost savings versus competitors. With estimated total capital investment (TCI) of $43M for a 60M lb/yr plant, this would be 25% of the cost published by Avantium for a plant 5 times smaller (5Ktpa or 11M lb/yr). Closed-loop recycle of expensive catalyst and extraction solvents enable these costs to be capitalized in working capital included in the TCI. Furthermore, by utilizing waste streams as the source for xylose, considerable improvement in cost of goods sold are demonstrated. Use of conservative FDCA sales value at $0.85/lb provide sufficient cushion for potential capital or feedstock cost escalation while still yielding attractive IRR, ROIC, and operating margin metrics.

Publications