From beech wood to itaconic acid: case study on biorefinery process integration
- 1.2k Downloads
Renewable raw materials in sustainable biorefinery processes pose new challenges to the manufacturing routes of platform chemicals. Beside the investigations of individual unit operations, the research on process chains, leading from plant biomass to the final products like lactic acid, succinic acid, and itaconic acid is increasing. This article presents a complete process chain from wooden biomass to the platform chemical itaconic acid. The process starts with the mechanical pretreatment of beech wood, which subsequently is subjected to chemo-catalytic biomass fractionation (OrganoCat) into three phases, which comprise cellulose pulp, aqueous hydrolyzed hemicellulose, and organic lignin solutions. Lignin is transferred to further chemical valorization. The aqueous phase containing oxalic acid as well as hemi-cellulosic sugars is treated by nanofiltration to recycle the acid catalyst back to the chemo-catalytic pretreatment and to concentrate the sugar hydrolysate. In a parallel step, the cellulose pulp is enzymatically hydrolyzed to yield glucose, which—together with the pentose-rich stream—can be used as a carbon source in the fermentation. The fermentation of the sugar fraction into itaconic acid can either be performed with the established fungi Aspergillus terreus or with Ustilago maydis. Both fermentation concepts were realized and evaluated. For purification, (in situ) filtration, (in situ) extraction, and crystallization were investigated. The presented comprehensive examination and discussion of the itaconate synthesis process—as a case study—demonstrates the impact of realistic process conditions on product yield, choice of whole cell catalyst, chemocatalysts and organic solvent system, operation mode, and, finally, the selection of a downstream concept.
KeywordsBiorefinery process Bio-chemical conversion Bio-based platform chemical Itaconic acid Bioeconomy
The global change to renewable feedstocks drives the development of new processes for a wide range of functional platform chemicals and fuels [1, 2, 3, 4]. An industrial production via biotechnological processes has to be adapted to new challenges such as recalcitrant raw materials, oxygenized molecules, and chemically highly demanding starting materials, such as lignocellulose . Within the large spectrum of possible products, organic acids are particularly interesting as building blocks. Several studies have defined criteria for a competitive biotechnological process: Low price for bio-catalysts, flexible utilization of different (low price) feedstocks (straw, grass, wood, etc.), and low environmental impact [6, 7, 8, 9, 10, 11]. The fact that optimized bioprocesses can already economically compete with their petrol-based counterparts is demonstrated by the strong market positions of citric and lactic acid, organic acids that are produced by fermentation in megaton scale . For new processes, space–time yield and product titer have often to be drastically improved, especially for the bulk production of biofuels.
Beside the investigation of individual unit operations, there have been several research projects on process chains leading from plant biomass to the final products like lactic acid, succinic acid, and itaconic acid [13, 14, 15, 16]. The development and optimization of (bio-)catalysts based on pure or idealized model substrates are absolutely necessary, but in case of biorefinery processes, it is doubtful whether the results can be transferred to an integrated process, where substrates contain impurities, or are only available in a diluted form.
Itaconic acid is considered as one of the most promising building blocks for the synthesis of a variety of product classes [2, 17]. It can be converted to the potential second-generation biofuel 3-methyltetrahydrofuran (3-MTHF), which has superior combustion and emission properties compared to gasoline . Itaconic acid is currently produced at industrial scale by the filamentous fungus Aspergillus terreus from renewable substrates like molasses [19, 20]. However, A. terreus is sensitive to substrate impurities, which is reflected by the strong effect of several medium components on itaconic acid formation [20, 21, 22, 23, 24, 25].
This study aims at analyzing the potential and pitfalls of the complete process chain for the conversion of plant biomass (beech wood) into the platform chemical itaconic acid, thereby identifying areas of future research needs.
Conceptual overview of the itaconic acid process
A first general process scheme for production of itaconic acid from wooden biomass was already proposed by Kobayashi in 1978 . After itaconic acid was mentioned as a very promising platform chemical , it became the focus of several larger research projects in the last decade. The “German Lignocellulose Feedstock Biorefinery Project”, investigated itaconic acid production from wood hydrolysates . The project “Development of integrated production of polyitaconic acid from northeast hardwood biomass” investigated the utilization of hardwood, softwood, and corn gluten feed for the production of polyitaconic acid . In both projects, inhibitory compounds negatively influenced the growth and itaconic acid production of A. terreus.
Since 2007, the German Cluster of Excellence “Tailor-Made Fuels from Biomass (TMFB)” aims at establishing innovative and sustainable processes for the conversion of plant biomass into fuels, for novel low-temperature combustion engines with high efficiency and low pollutant emission. In this project, itaconic acid is an intermediate platform chemical and can be converted into several potential fuel candidates, such as 3-methyltetrahydrofuran [29, 30]. The process can be classified as a “Lignocellulose Feedstock Biorefinery” according to the definition of Kamm and Kamm  and the German VDI , since it uses biomasses with low water content as raw materials, e.g., wood, straw, corn stover as well as cellulose-containing biomass and waste.
OrganoCat pretreatment and enzymatic hydrolysis
Since the pretreatment significantly affects saccharification , enzymatic hydrolysis of the solid, cellulose-rich beech wood pulp was investigated (Fig. 2b). The hydrolysis of 20 g L−1 cellulose-rich beech wood pulp originating from OrganoCat pretreatment was compared to the hydrolysis of 20 g L−1 pure α-cellulose using a commercial cellulose cocktail, Celluclast®. After 144 h, 60% and 74% conversion into glucose was determined for cellulose-rich beech wood pulp and α-cellulose, respectively. The lower glucose conversion for the beech wood pulp can be explained by a different accessibility of the cellulose fibers for cellulases, caused by residual cellulose crystallinity and lignin content. However, adaption of the enzyme mixture of cellobiohydrolases, endoglucanases, and beta-glucosidase in the Celluclast® enzyme cocktail to the specific cellulose structure of beech wood would further improve performance [55, 60, 61, 62, 63].
The utilization of lignin is essential for any biorefinery concept based on lignocellulose. A critical parameter for lignin processing is its high heterogeneity. Therefore, Stiefel et al. described a method based on statistical design of experiments to quantify the effect of temperature, alkalinity, catalyst, lignin concentration and current density on the molecular weight, monomer production, UV absorbance as well as acid solubility of the treated lignin . For further lignin processing, different strategies were investigated on the engineering and on the bio-chemical level. A promising option is the usage of electrochemical membrane reactors to degrade lignin into low-molecular-weight compounds [38, 39, 40, 41]. Alternative ways include chemical procedures like the ruthenium-catalyzed C–C bond cleavage , alcohol oxidation and subsequent cleavage into aromatics , copper- and vanadium-catalyzed oxidative cleavage , base-catalyzed depolymerisation  as well as mechanochemical degradation .
Separation and recycling of oxalic acid
The results demonstrate that the proposed separation process is technically feasible. Recoveries for oxalic acid of 80% can be reached, depending on the initial sugar concentration. To process high monosaccharide concentrations, nanofiltration could be used in diafiltration mode. As a drawback, the oxalic acid concentration in the permeate will decrease. Meanwhile, it was also realized that part of the oxalic acid catalyst is decomposed upon extended operation times. Therefore, currently, experiments are conducted to find an acid which can easily be recycled and which is not decomposed under the fractionation conditions.
Itaconic acid production with A. terreus and U. maydis
Industrial production of itaconic acid is commonly achieved with the filamentous fungus A. terreus, which exhibits high product yields and concentrations under optimal conditions . Product formation in A. terreus is highly dependent on diverse factors including pH, oxygen supply, power input, phosphate concentration, and the presence of metal ions. In turn, product formation is poorly reproducible, if these factors are not precisely controlled [20, 25, 65]. Therefore, a second fungal producer, U. maydis, has been investigated by the TMFB consortium. Its wild type is a less efficient producer of itaconic acid than A. terreus, but its unicellular, non-filamentous morphology provides advantages for large-scale fermentation. Importantly, U. maydis is more tolerant to medium impurities. This is of lesser importance when working with pure glucose as feedstock, but crucial when using chemo-catalytically pretreated and hydrolyzed biomass. Besides itaconic acid, wild-type U. maydis produces several potentially interesting products like malic acid, succinic acid, and 2-hydroxyparaconic acid [56, 66, 67, 68]. The characterization of the metabolic pathway for itaconic acid  enabled rational metabolic engineering, resulting in the genetically modified U. maydis Δcyp3Petefria1 in which itaconic acid production is increased, while by-product formation is reduced .
In Fig. 4c, a genetically modified U. maydis strain is shown. The maximum itaconic acid concentration is approx. 55 g L−1 after 163 h and a maximum volumetric productivity of 0.4 g L−1 h−1 after 90 h is achieved. The product yield is 0.48 gIA ginitial glucose −1 . U. maydis is known for converting hemicellulose into fungal biomass and itaconic acid, as shown in Fig. 4d. As represented by the oxygen transfer rate (OTR) and itaconic acid concentration, U. maydis is not only capable of growing and producing itaconic acid on pure glucose/xylose mixtures, but also on pure xylose. It can also degrade xylan and can be engineered to secrete xylanases and cellulases, making it potentially applicable in a consolidated bioprocess [75, 76]. In general, organisms which are capable to convert C5 and C6 sugars are preferable for biorefinery processes. To investigate the robustness of U. maydis and its capability to convert also hemi-cellulosic sugars from pretreated beech wood, investigations were performed by Klement et al. . It could clearly be shown that U. maydis tolerates oxalic acid concentrations up to 0.1 M, which is the applied concentration in the OrganoCat process, presented above (Fig. 2a). The oxalic acid concentration transferred into the fermentation step depends on the operation conditions of the nanofiltration, as presented in Fig. 4. Further experiments investigating the robustness of U. maydis are presented in Klement et al. .
To increase the volumetric productivity, a continuous process mode is an option. However, product concentrations are often lower in continuous processes compared to repeated-batch and fed-batch processes. In addition, the biocatalyst is also continuously washed out. For this reason, Carstensen et al.  used an in situ membrane module in a continuously operated stirred tank reactor to increase the volumetric activity and to generate a cell free, itaconic acid containing permeate stream. A volumetric productivity of 0.8 g L−1 h−1 was obtained in that work, even though only an unoptimized wild-type strain U. maydis MB215 was used in these experiments. Subsequently, an advanced reversed flow diafiltration technique was developed [79, 80]. A combination of genetically enhanced strains of U. maydis and the membrane bioreactor promises even higher volumetric productivities.
Extraction and back extraction of itaconic acid
The comparison of different downstream opportunities (crystallization, reactive extraction, precipitation, electrodialysis, diafiltration, and adsorption) for itaconic acid was published by Magalhães et al.  and reflects also the experience of this study that reactive extraction is superior in respect to energy demand and scalability. Therefore, an in situ removal by reactive extraction has been realized. In Fig. 4b, results of experiments with isopropyl myristate as organic carrier solvent and trioctylamine as reactant for an in situ itaconic acid reactive extraction are depicted. For both initial concentrations, 215 g L−1 as well as 268 g L−1 glucose is completely converted into itaconic acid. These results reflect the biocompatibility of the solvent fraction to A. terreus, which is also important if process water is recycled. Remarkably, no glucose remains in the medium at all, even though higher initial glucose concentrations were added than in the experiment, as shown in Fig. 4a. The maximum itaconic acid concentrations are 87 g L−1 and 105 g L−1 with a yield for both experimental conditions of approx. 0.41 gIA g initial glucose −1 . Therefore, in situ extraction of itaconic acid is a powerful tool to increase the efficiency of the process [82, 83]. Further details are described in Kreyenschulte et al. .
For the back extraction of itaconic acid into the aqueous phase (Fig. 1), a pH shift by the addition of alkaline solution of, e.g., NaOH, can be realized. With pH values in the aqueous phase higher than 5.55 (pKa2), back extraction with an efficiency of 100% is feasible (Fig. 5b, c). The consequence is the formation of a salt. By adding approx. 35 mL NaOH per liter isopropyl myristate containing 300 g L−1 pure itaconic acid, the product can be completely back extracted into the water phase. The efficiency of this back extraction can be described by the molar ratio of 0.69 molIA mol NaOH −1 .
Crystallization of itaconic acid
The published industrial benchmark for itaconic acid production is a batch process based on molasses. By beginning of 2018, the average price for molasses in Europa was approx. 128 € t−1 (http://www.proplanta.de). Due to changes in sugar market regulation by the European government, the price of sugar dropped to approx. 110 € t−1 in August 2018 (New York Mercantile Exchange). Taking into account that molasses contains only 43–45% of convertible sugars, the often-mentioned industrial process loses its relevance as benchmark under the current situation on the sugar market—at least in Europa. Nieder-Heitmann et al. published different economic scenarios for itaconic acid producing biorefinery concepts . Among others, the authors have compared itaconic acid production based on glucose with production based on lignocellulose, which should theoretically be more price competitive. In case of a lignocellulosic feedstock, most economically relevant parameters of the process are the itaconic acid yields based on glucose as well as on xylose, followed by the volumetric productivity and the initial glucose concentration (which reflects the efficiency of pretreatment and enzymatic hydrolysis). With respect to the itaconic acid yield on glucose, both compared organisms U. maydis and A. terreus are in the same range of approx. 0.5 gIA ginitial glucose −1 (theoretical yield 0.72 gIA g glucose −1 , ) depending on experimental conditions and cited publications. A. terreus as well as U. maydis can (co-)metabolize xylose into biomass and itaconic acid in low yields and rates. The published volumetric productivity of A. terreus is higher, but U. maydis seems to have a lot of potential for further improvement on the genetic level as well as on the process engineering level, due to its higher growth rate, its resistance towards impurities, and its single-cell morphology. Since the impact of the initial glucose concentration is significantly lower in comparison to the itaconic acid yields , it might be tolerable if some glucose remains in the fermentation broth as long as high product yields can be achieved and the downstream processing is not severely affected. However, to quantify the impact of all process relevant and economic parameters in detail, a technoeconomic analysis would be essential.
A comprehensive examination of the itaconic acid process and the interfaces between the unit operations demonstrates the impact of realistic process conditions on the feasibility, product yields, choice of microorganisms, and operation mode. The decision for a suitable chemo-catalytic fractionation method is probably the most essential choice to make. As shown by the results of the cellulose hydrolysis as well as fermentation with U. maydis or A. terreus, all the following steps will be influenced by the digestibility of the pretreated biomass by the cellulolytic enzymes and inhibitors originating from the pretreated biomass. For the microbial conversion of pretreated biomass, evaluation parameters like robustness against impurities and inhibitors and flexibility of feedstock are at least as important as classical parameters like yield and volumetric productivity. In this context, the genetically enhanced U. maydis is ultimately superior to A. terreus. In case of itaconic acid and the achieved concentrations after fermentation, reactive extraction by trioctylamine and isopropyl myristate is a suitable way to significantly increase the itaconic acid concentration before crystallization. Experiments with in situ extraction demonstrate the biocompatibility of trioctylamine and isopropyl myristate. Therefore, a recycling of the aqueous phase containing dissolved extractant would be harmless for the microbial system. The presented results towards an integrated itaconic acid process clearly demonstrates that the shift to lignocellulosic substrates challenges existing processes and shows the importance of including the interfaces between unit operations into the research efforts.
LR, TK, PG, DK, BH, TM, AE, RS, YW, NW, LB, AS, WL, CB, MW, AJ, MR, and JB designed the research. LR, TK, PG, DK, EA, BH, TM, AE, YW, and NW performed experiments and analyzed the data. LR and JB wrote the paper. All authors read and approved the final manuscript.
We thank the Deutsche Forschungsgemeinschaft (DFG) for financial support.
The authors declare that they have no competing interests.
Availability of data and materials
The data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Ethics approval and consent to participate
This work was performed as part of the DFG Cluster of Excellence “Tailor-Made Fuels from Biomass” EXC 236, which is funded by the Excellence Initiative by the German federal and state governments to promote science and research at German universities.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 2.Werpy T, Petersen G. Top value added chemicals from biomass: volume I—results of screening for potential candidates from sugars and synthesis gas. Washington DC: US Department of Energy (US); 2004.Google Scholar
- 26.Kobayashi T. Production of itaconic acid from wood waste. Process Biochem. 1978;13(5):15–22.Google Scholar
- 28.Itaconix. Development of integrated production of polyitaconic acid from northeast hardwood biomass. Grant from National Institute of Food and Agriculture (NIFA) 2009–2012.Google Scholar
- 30.Marquardt W, Harwardt A, Hechinger M, Kraemer K, Viell J, Voll A. The biorenewables opportunity-toward next generation process and product systems. AIChE J. 2010;56(9):2228–35.Google Scholar
- 32.Biotechnologie V-F. Classification and quality criteria of biorefineries. Duesseldorf: VDI-Gesellschaft Technologies of Life Sciences; 2016.Google Scholar
- 36.vom Stein T, Grande PM, Kayser H, Sibilla F, Leitner W, de Dominguez Maria P. From biomass to feedstock: one-step fractionation of lignocellulose components by the selective organic acid-catalyzed depolymerization of hemicellulose in a biphasic system. Green Chem. 2011;13(7):1772–7.CrossRefGoogle Scholar
- 57.N KT. Organosolv pulping and recovery process. In.: Google Patents; 1971.Google Scholar
- 66.Guevarra ED, Tabuchi T. Production of 2-Hydroxyparaconic and Itatartaric Acids by Ustilago cynodontis and Simple Recovery Process of the Acids. Agric Biol Chem. 1990;54(9):2359–65.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.