Overview of biomass conversion pathway
The overall processing strategy for biomass conversion to fuels and chemicals assessed in this work is presented in Fig. 2. The two-stage alkaline pre-extraction followed by alkaline-oxidative pretreatment method is used to fractionate lignocellulose biomass into various lignin and sugar streams for downstream conversion. As shown in Fig. 2, this approach provides the flexibility to accommodate shifting market conditions. It does this by yielding several lignin products that can target multiple markets, altering the properties of the lignin, and varying the partitioning of lignin into the three intermediate product pools, or target molecules.
One key set of target molecules includes phenolic acid and aldehyde monomers (vanillin, vanillic acid, syringaldehyde, syringic acid, and others) that can be directed towards high-value, low-volume markets (e.g., flavor and fragrance compounds). Specifically, the flavor and fragrance industry has a total global market size of $28 billion with strong continued growth forecasted in developing countries that tracks GDP . As one example, synthetic vanillin is produced at the scale 37,000 tonnes/year primarily from petroleum-derived aromatics with a market value on the order $10/kg . A small fraction of this market is bio-based vanillin derived from the alkaline oxidation of wood-derived spent sulfite black liquor and from the biological conversion of plant-derived ferulic acid . This market offers a potential high-value niche for lignin-derived bio-based aromatics and a single biorefinery processing 2000 tonne/day of woody biomass would only displace 0.2% of this market. Another potential market for lignin-derived monomers includes medium-value, high-volume functionalized aromatic chemical markets [38–40], with applications that could include bio-based polymers such as a replacement for Bisphenol A in thermoset resins  and as novel bio-based monomer in polyesters . A second class of target molecules is highly functionalized oligomeric lignins that have potential in resin formulations for bio-based polyurethane coatings. Well-established challenges for utilizing polymeric lignins as a feedstock for polyurethane resin production include its dark color, low solubility in reaction solvents, low reactivity, high polydispersity, and brittleness . If these challenges can be overcome, process-derived lignins represent an opportunity as a renewable source of polyols in the production of polyurethane resin for coating applications. This unique and innovative approach for lignin depolymerization to yield aromatic monomers will yield a subset of lignins that are well-suited for application as polyols in polyurethane resin formulations with properties that include low molecular weights, low polydispersities, low glass transition temperatures, and high reactivities.
Unlike hydrogenation/reductive approaches to lignin depolymerization or conversion whereby alcohols, aldehydes, carboxylates, and aromatics are reduced and deoxygenated, oxidations can preserve and generate oxygen-containing functional groups (i.e., vanillin, vanillic acid, syringaldehyde, syringic acid, acetosyringone, acetovanillone). While the oxygen content of biomass-derived compounds is a negative for fuels applications, oxygen-containing groups are useful for providing chemical functionality and reactivity for use as platform chemicals or as reactive aromatic polymers that can be incorporated into polyurethane resin formulations to increase their bio-based content.
Capital and operating costs
Table 4 shows the material balance for the eight pretreatment conditions assessed in this study based on our prior experimental study . As shown, varying the pretreatment conditions impacted both the monomeric sugar (glucose and xylose) yields and the extent of delignification, thereby affecting the yields of biofuels and lignin-based products (polyols and aromatic monomers). With increasing pretreatment severity (i.e., temperature and oxidant loading), the yields of products (monomeric sugars and lignin) were increased (Additional file 1: Table S3). However, increasing the pretreatment severity also resulted in increased capital and operating costs. Therefore, an optimum balance between the process costs and the product yields needed to be identified with the technoeconomic model.
Figure 3a shows the total capital costs for all eight pretreatment conditions modeled in this study. The use of both H2O2 and O2 as co-oxidants during the alkaline-oxidative pretreatment stage increased the capital cost compared to the alkaline-oxidative pretreatment with H2O2 only. Moreover, the case with alkaline pre-extraction performed at 90 °C and alkaline-oxidative pretreatment performed with only 8% H2O2 had the lowest total capital cost ($20.1 million), while the case with alkaline pre-extraction performed at 120 °C and the alkaline-oxidative pretreatment with 8% H2O2 and 50 psig O2 had the highest total capital cost ($42.2 million). This could be attributed to the higher cost reactor; the addition of 50 psig O2 requires a much thicker vessel than the case without using O2.
Figure 3b displays the operating costs for the eight pretreatment conditions. As shown, under the same alkaline pre-extraction temperature (120 °C), using O2 in addition to H2O2 during the alkaline-oxidative pretreatment stage only slightly increases operating costs. O2 was assumed to be recovered from air on site, during which only the electricity was used as a contributor to the operating cost. In contrast, reducing H2O2 utilization from 8 to 2% reduced operating costs by $42 million/year due to the relatively high cost of purchasing H2O2 ($1/kg); this could lead to a considerable decrease in MFSP. To probe further the operating cost, the individual contributors to the operating cost were also investigated (Additional file 1: Table S4). Moreover, when using the solubilized lignin for high-value products instead of burning for energy, the required electricity increased for the cases that solubilized more lignin during the pretreatment process; this also increased the operating cost.
Minimum fuel selling price (MFSP)
Figure 4 shows the estimated MFSP ($/L) for the eight pretreatment conditions considered in this study. Two scenarios are presented for the MFSP. In the first scenario, the soluble lignin that is not precipitated is burned for energy, while in the second scenario, the soluble lignin in the Cu-AHP extract that is not precipitated is assumed to be recoverable and sold at the same price as the precipitated lignin ($0.80/kg). The cost of pretreatment chemicals had a large influence on the MFSP, accounting for 40% of the total operating costs for the base case of a 120 °C alkaline pre-extraction followed by an alkaline-oxidative Cu-AHP pretreatment with 8% H2O2 (120 °C—Cu-AHP 8% H2O2). If we assumed that the acid-soluble lignin was not recoverable, the MFSP using H2O2 as the only oxidant [(120 °C—Cu-AHP 8% H2O2) and (90 °C—Cu-AHP 8% H2O2)] was between $1.32/L and $1.08/L depending on the temperature of the alkaline pre-extraction stage. Conversely, when O2 was used as a co-oxidant and the H2O2 loading was reduced from 8 to 2%, the MFSP decreased to between $0.94/L and $0.85/L. This is because this sizable reduction in pretreatment chemical usage did not result in a corresponding large reduction in sugar yields (Additional file 1: Table S2; ). Eliminating the H2O2 entirely led to slight increase in MFSP due to an appreciable reduction in both the sugar and lignin yields (Additional file 1: Table S2; ). Importantly, if the acid-soluble lignin can be recovered for value-added products, then the MFSP can be reduced by an additional $0.10/L (down to $0.77/L) if O2 is employed as a co-oxidant during the Cu-AHP process (120 °C—Cu-AHP 2% H2O2 + O2). The use of O2 as a co-oxidant increased the amount of lignin solubilized during pretreatment, but a larger proportion of this lignin was acid soluble. Thus, the difference in MFSP between the two assumptions (all solubilized lignin is recoverable versus only precipitated lignin) was greater when O2 was employed as a co-oxidant.
The TEA indicates that the overall MFSP can be reduced by nearly 40% by using O2 as a co-oxidant in the Cu-AHP process relative to the Cu-AHP pretreatment using H2O2 only. This is due both to a decrease in pretreatment operating cost (due to a reduction in H2O2 loading) and to an increase in both glucose and lignin yield. The primary tradeoff for oxygen utilization is a modest increase in electricity usage to generate the oxygen as well as an increase in capital costs (the oxygen production unit is assumed to cost $9.7 million, while the cost of increasing the pressure rating of the pretreatment vessel is $12.4 million). Despite these costs, the added capital cost only increased ~ 6% (Fig. 2) and therefore did not greatly impact the MFSP. From the results in Fig. 3, pretreatment conditions of alkaline pre-extraction (120 °C) and alkaline-oxidative pretreatment (2% H2O2 and O2) were selected as the base case for further analysis. Moreover, a detailed list of contributors to the MFSP of the selected base case (2% H2O2 and O2) was also provided (Additional file 1: Table S4).
Effect of lignin valorization on MFSP
The above analysis assumed only lignin that was solubilized and recoverable by acid precipitation could be utilized as a polyol substitute. Multiple other scenarios were also considered: (1) no lignin was recovered for additional value as a worst-case scenario; (2) 16% of the recovered lignin (based on results obtained from lignin depolymerization following the method of sequential Bobbitt’s salt oxidation followed by formic acid-catalyzed depolymerization process ) could be sold as monomers, increasing its value to $2.00/kg, while the remainder of the precipitated lignin was burned for fuel; (3) the same 16% of recovered lignin is sold as monomers, but the remaining recovered lignin was sold as a polyol substitute; (4) the solubilized but not precipitated lignin could also be recovered and sold as a polyol substitute ($0.80/kg); (5) 16% of all solubilized lignin (including the non-precipitated portion) was sold as monomers (with the remainder as a polyol substitute), and (6) the precipitated lignin was sold as a polyol substitute, while 48% of the non-precipitated lignin was sold as monomers (Fig. 5).
Importantly, if the acid-soluble lignin can be recovered for value-added products, then the MFSP can be reduced by an additional $0.07/L (down to $0.78/L) if O2 is employed as a co-oxidant during the Cu-AHP process (120 °C—Cu-AHP 2% H2O2 + O2). As noted above, the use of O2 as a co-oxidant increased the amount of lignin solubilized during pretreatment, but a larger proportion of this lignin was acid soluble. Thus, a strategy to recover this soluble lignin will be important to further optimize this process due to the presence of oxygen. Likewise, if the value of the lignin can be increased by conversion to aromatic monomers, the MFSP can be reduced further to $0.73/L. This is due solely to increased value of lignin, as it increases from 12 to 26% of the total revenue of the biorefinery. An intermediate approach, in which 48% of the soluble lignin can be recovered and sold as high-value monomers, also significantly reduces the cost to $0.74/L. If lignin is not recovered as a co-product, the MFSP is $1.03/L, indicating the importance of lignin recovery during Cu-AHP pretreatment. In the case that the precipitated lignin is converted to monomers at a 16% yield but the remaining precipitated lignin can only be burned as fuel, the MFSP is only $0.88/L, less than the base case in which the precipitated lignin is used as a polyol substitute. Thus, while developing this technology, it is imperative that either yields for monomers increases or the process allows for the remaining lignin to be used as a polyol substitute.
Significant advances have been made to reduce the input costs of copper-catalyzed alkaline hydrogen peroxide pretreatment while simultaneously maintaining high sugar yields (95% glucose and ~ 100% xylose of initial sugar composition) [23, 43, 44]. Despite this, the operating costs for pretreatment were still high at approximately $71 million/year for a 2000 dry tonne/day facility (Fig. 4) or $97/tonne biomass, resulting in a $1.03/L MFSP if no lignin was recovered as a value-added product. This decreased to $0.85/L if precipitated lignin was recovered as a polyol substitute and $0.78/L if all soluble lignin could be recovered as a polyol substitute. While the technology to produce polyurethane products from lignin is relatively well understood, the possibility of producing monomers can reduce the selling price further down to $0.73/L. While challenges currently remain to commercializing this technology, it demonstrates that further selling price reductions are possible as improvements in lignin valorization continue. Thus, the combination of reduced pretreatment inputs while maintaining high sugar and lignin solubilization and improved usage of recovered lignin is instrumental in obtaining economically competitive biofuels.
Understanding the impact of key parameters on the MFSP is of great importance to developing this technology further. Sensitivity of the MFSP with the sequential two-stage alkaline pre-extraction and alkaline-oxidative pretreatment of hybrid poplar (the selected base case) is summarized in Fig. 6, in which the capital and operating costs were also included. Yield of both sugar and lignin had the highest impact on the final biofuel selling price, indicating the importance of recovering all of the solubilized material. Likewise, the value of the lignin, used either in polyurethane applications or as lignin monomers, also resulted in large changes in the biofuel selling price. This indicates that revenue, rather than the individual costs of the refinery, drives the economics of the process. Each of the cost drivers selected, namely hydrogen peroxide cost, pretreatment capital cost, oxidation pressure, and total oxygen usage had relatively minimal impact on the final selling price of the fuel. This analysis provides evidence that, if the high yields and potentially high value for lignin can be maintained as the process is scaled to more industrially relevant conditions, the potential for economic value will remain even if costs are greater than initially anticipated.
The results further indicate that it is of great importance to include the lignin properties and valorization strategies when establishing TEA models for biorefinery. This is because the lignin value has a significant impact on the MFSP (based on the scenarios we studied). To include the lignin value in the model, there are several methods to be considered: (1) purifying lignin for the various applications in order to design downstream separations; (2) recycling or treating any waste streams generated from these separations, and (3) drying and packaging of the final lignin product. Synergies may be found if the final product (such as polyurethane) is produced at the same location. Further laboratory optimization of lignin separations and purification would also be required.
Cu-AHP is promising technology to improve the production of biofuels from lignocellulose, and this economic assessment illustrates the importance of considering high-value co-products when assessing these technologies. The pretreatment process described herein demonstrated high sugar and lignin yields while reducing the raw material input in the pretreatment, thereby yielding biofuel at a cost as low as $0.85/L. In addition to the high yields, diversification of the lignin products into higher value products has potential to reduce the fuel costs even further to $0.73/L, compared to a value of $1.03/L if the lignin is only used as a fuel source. Given the promising results that both high-value lignin and high sugar yields can be obtained while significantly reducing the pretreatment costs, further research is thus warranted on improving and integrating the pretreatment and lignin valorization technologies, moving both of them to a more commercially ready state. Modeling this integration as the technology continues to progress will also be instrumental in optimizing the conditions and ensuring the process is economically viable.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 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 in a credit line to the data.