Utilization of lignocellulosic biomass as sustainable feedstock for the bioproduction of n-butanol as an advanced biofuel: current progress and future perspectives

1 Introduction

n-Butanol (1-butanol) has attracted substantial attention as a next-generation biofuel (Emmanuel and Ezeji, 2025; Liu et al., 2025). Compared with ethanol, n-butanol exhibits a high energy density, low hygroscopicity, superior blending compatibility with gasoline, enabling its use in existing combustion engines and fuel infrastructure (Lee S. Y. et al., 2008). In addition to its application as an advanced biofuel, n-butanol serves as an important intermediate in the chemical industry for producing acrylates, acetates, and plasticizers, thereby bridging energy and material sectors in a low-carbon economy (Lee et al., 2016). These advantages drive n-butanol as a promising candidate for sustainable energy systems and circular bioeconomy development.

Traditionally, microbial production of butanol is primarily achieved via acetone–butanol–ethanol (ABE) fermentation by solventogenic clostridia, particularly Clostridium acetobutylicum, Clostridium beijerinckii. These anaerobic Gram-positive bacteria convert carbohydrates into acids during the acidogenic phase, followed by a metabolic switch to solventogenesis, during which butanol, acetone, and ethanol are produced (Liao et al., 2015). Despite its historical significance and industrial relevance, ABE fermentation still suffers from several intrinsic limitations. First, butanol exerts strong product toxicity on microbial cells, typically inhibiting growth and metabolism at relatively low concentrations (∼10 g/L), which severely constrains final titers (Arsov et al., 2021). Second, the metabolic shift between acidogenesis and solventogenesis is highly sensitive to environmental conditions, leading to process instability and batch-to-batch variability (Li et al., 2012; Luo et al., 2019). Third, low carbon conversion efficiency, coupled with by-product formation and complex redox regulation, results in suboptimal yields (Zheng et al., 2015). Finally, the reliance on refined sugars or starch-based feedstocks further increases production costs, limiting economic competitiveness at industrial scale (de Moraes Altafini et al., 2026; Zhang et al., 2026a).

To overcome the challenges, the selection of sustainable and low-cost feedstocks has become a critical research focus (Heidari et al., 2025; Joshikumar et al., 2026). Lignocellulosic biomass, which is the most abundant renewable carbon resource, offers an alternative to conventional sugar-based substrates. Agricultural residues and forestry wastes represent an enormous, non-food-competitive carbon reservoir that can support large-scale biofuel production. The utilization of lignocellulosic feedstocks can reduce dependence on edible biomass and contribute to waste valorization and carbon neutrality objectives. Therefore, integrating lignocellulosic biomass into ABE fermentation systems is widely recognized as a key strategy toward economically viable and environmentally sustainable n-butanol production. However, the efficient bioconversion of lignocellulosic biomass to biobutanol faces multiple interconnected barriers, mainly including structural complexity and recalcitrance of lignocellulose, lack of effective pretreatments and unclear understanding of the fermentable inhibitors to clostridial metabolism, high cost of enzyme for hydrolysis and limited efficiency, high energy requirement of >14 MJ/kg-butanol, etc. (Himmel et al., 2007; Xue et al., 2017; Zheng et al., 2015). Of these barriers, the present review focuses on the pretreatment–fermentable inhibitor–strain engineering structure, because this framework represents the recent process improvement in biobutanol production field and it was not systematically summarized in recent reviews. The hierarchical structure largely limits sugar release efficiency and necessitates pretreatment processes to disrupt biomass recalcitrance (Wang et al., 2026; Zhang et al., 2026b). Pretreatment of lignocellulosic biomass is therefore an essential step to enhance enzymatic digestibility and improve fermentable sugar yield. However, most pretreatment strategies inevitably generate a variety of microbial inhibitors, including furan derivatives (furfural, 5-hydroxymethylfurfural), weak acids (acetic acid, formic acid), and phenolic compounds derived from lignin depolymerization (Jönsson and Martín, 2016; Luo et al., 2021b). These compounds exert severe inhibitory effects on solventogenic clostridia by disrupting membrane integrity, inducing oxidative stress, and interfering with central metabolic pathways and redox balance (Liu et al., 2019; Luo et al., 2020). Consequently, inhibitor formation has become a major bottleneck in lignocellulose-based ABE fermentation systems, often necessitating additional detoxification steps that further increase process complexity and cost.

Considerable progress has been made in recent years toward developing efficient lignocellulose-to-n-butanol conversion systems. The development of improved pretreatments has shifted from conventional acid and alkaline methods to emerging green solvents such as ionic liquid (IL) and deep eutectic solvent (DES). DESs, characterized by their green and biocompatible nature, have attracted considerable interest as promising media for lignocellulose pretreatment (Luo et al., 2024; Mqoni et al., 2026). Their straightforward synthesis from hydrogen bond donors (HBDs) and hydrogen bond acceptors (HBAs) under moderate conditions has led to their recognition as “designer solvents”. Meanwhile, some advances span multiple levels, including engineering of microbial strains with enhanced tolerance and metabolic efficiency, and optimization of fermentation processes through in situ product recovery and integrated bioprocessing strategies (de Moraes Altafini et al., 2026; Su et al., 2024). These efforts collectively aim to overcome the trade-offs between lignocellulose deconstruction efficiency, inhibitor formation, and microbial productivity. Despite these advancements, a fully integrated and economically competitive lignocellulose-based n-butanol production platform has not yet been achieved. Key challenges remain in balancing pretreatment severity with inhibitor generation, improving microbial robustness under industrially relevant conditions, and reducing downstream separation costs associated with low butanol titers. Therefore, a systematic understanding of the entire value chain, from biomass structure and pretreatment chemistry to microbial metabolism and process engineering, is essential for guiding future development.

In this review, we systematically summarize the current advances in lignocellulosic biomass utilization for n-butanol production. We first analyze the structural features of lignocellulose and the impact of pretreatment-derived inhibitors on butanol fermentation performance. Subsequently, we provide a comprehensive overview of conventional and emerging pretreatment strategies, microbial and metabolic engineering approaches, and fermentation process optimization techniques. Finally, we highlight the remaining challenges and propose future perspectives toward the development of efficient, sustainable, and industrially viable lignocellulose-based n-butanol bioproduction systems.

2 Fermentation inhibitors derived from pretreatments and their effects on butanol biosynthesis

2.1 Formation features of inhibitors during pretreatment process

Lignocellulosic biomass is a heterogeneous polymeric material mainly composed of cellulose (40%–50%), hemicellulose (25%–30%), and lignin (15%–25%), with minor fractions of ash, proteins, other extractives. The relative composition varies significantly depending on biomass species, growth conditions, geographic origin, and environmental factors. As shown in Figure 1A, cellulose is a linear polysaccharide consisting of β-1,4-linked glucose units (Baruah et al., 2018), which can be hydrolyzed into glucose under acidic or enzymatic hydrolysis. Hemicellulose is a heterogeneous branched polymer composed of both pentose and hexose sugars, while lignin is an amorphous three-dimensional aromatic polymer formed by the random coupling of phenylpropane units via ether and carbon–carbon linkages. The tight cross-linked architecture among cellulose, hemicellulose, and lignin confers strong structural recalcitrance, severely limiting enzymatic accessibility and microbial conversion efficiency (Ren et al., 2024; Tarasov et al., 2018). Pretreatment processes are therefore essential to disrupt this compact matrix, partially remove lignin, reduce cellulose crystallinity, and enhance enzymatic digestibility (Ren et al., 2024; Zhao X. et al., 2022). However, the severity and chemistry of different pretreatment methods inevitably lead to the formation of various microbial inhibitors with distinct compositions and concentrations.

These pretreatment-derived inhibitors are generally categorized into three major groups, including weak organic acids, furan derivatives, and phenolic compounds (Baral and Shah, 2014; Jönsson and Martín, 2016). As shown in Figure 1B, weak acids, including acetic acid, formic acid, and levulinic acid, are mainly generated from hemicellulose deacetylation and subsequent sugar degradation reactions. Acetic acid is released from acetyl groups in hemicellulose. Furan derivatives, including furfural and 5-hydroxymethylfurfural (5-HMF), are produced through dehydration of pentoses and hexoses during severe thermal or acidic pretreatment conditions. Over-degradation of hemicellulose is a primary source of furfural, while hexose dehydration contributes to 5-HMF formation. Phenolic compounds, derived from lignin depolymerization, include vanillin, syringaldehyde, 4-hydroxybenzaldehyde, and ferulic acid. In general, the concentrations of furfural and 5-HMF vary considerably across different hydrolysates, yet generally remain below 3.5 g/L. In contrast, formic acid and levulinic acid, originating from singular pathways and being less readily formed, are typically present at low levels (0.1–0.6 g/L), whereas acetic acid can reach concentrations as high as 8.0 g/L (Zha et al., 2012). In a microwave-assisted alkalic salt pretreatment, the resulting corncob slurry contained three major fermentation inhibitors of 1.16 g/L acetic acid, 0.15 g/L dihydrobenzofuran, and 0.14 g/L 2-methoxy-4-vinylphenol, and furfural and 5-HMF were not detected (Sewsynker-Sukai and Gueguim Kana, 2018). When pretreatment of corn stover by steam explosion pretreatment at 205 °C for 5.4 min with a total solids content of 40% (w/w), the unwashed hydrolysate exhibited inhibitor concentrations of formic acid (6.80 g/L), acetic acid (7.81 g/L), total phenolics (4.5 g/L), benzoic acid (0.90 g/L), HMF (0.56 g/L), and furfural (0.71 g/L) (Huang et al., 2011). These compounds exhibit strong inhibitory effects on clostridial metabolism, thereby posing a major challenge for efficient lignocellulose-based fermentation systems (Liu et al., 2020; Liu et al., 2018).

2.2 The effects of weak acids on butanol fermentation

It was reported that weak acid inhibitors exert a dual and concentration-dependent impact on ABE fermentation (Baral and Shah, 2014; Cho et al., 2012). In C. acetobutylicum, the overall ABE titer dropped to 77% of the control upon supplementation with 0.4 g/L formic acid, and to 25% when formic acid was increased to 1.0 g/L. Furthermore, the co-existence of 8.7 g/L acetic acid and 0.4 g/L formic acid reduced the total ABE production from 15.1 to 8.6 g/L (Cho et al., 2012). This inhibition is largely attributed to the intracellular pH drop caused by weak acid accumulation, which creates an excessively acidic environment. Since clostridia generates limited ATP through anaerobic environment, they must expend considerable ATP to expel excess protons from the cytoplasm, thereby impairing cell growth and metabolism. Moreover, even low intracellular formic acid levels can provoke overproduction of acetic and butyric acids while suppressing solventogenesis, leading to the “acid crash” phenomenon during the typical biphasic fermentation process, where the metabolic shift from acid production to solvent synthesis fails to proceed properly (Wang et al., 2011). Acid crash arises from the inability of Spo0A-dependent phosphorelay to activate solventogenesis genes, coupled with NADH depletion and precursor shortage due to pH-inhibited enzymes, as well as disrupted inositol phosphate signaling. This locks fermentation in acidogenesis, causing acetic/butyric acid accumulation and negligible butanol production (Karstens et al., 2021). In contrast, it was reported that C. beijerinckii NCIMB 8052 exhibited no inhibition response in solvent production to any tested formic acid level (0.0–1.0 g/L), even when acetic acid was added together with formic acid (Cho et al., 2012). Therefore, C. beijerinckii exhibits higher tolerance to weak acids, attributed to the presence of formate dehydrogenase (FDH) in its cell membrane, which enables sustained solventogenic capacity without significant inhibition under comparable conditions. These observations highlight that the susceptibility to weak acid stress is strain-dependent and critically influences process stability and productivity in ABE fermentation.

2.3 The effects of furan derivatives on butanol fermentation

Furan derivatives are typical inhibitory compounds generated during lignocellulosic pretreatment. When present individually at low concentrations, furfural or 5-HMF can enhance solventogenic capacity in C. beijerinckii BA101 (Ezeji et al., 2007); however, the co-occurrence of mixed furans drastically suppresses xylose utilization and reduces overall solvent production in C. beijerinckii DSM 6422 (Bellido et al., 2014). Under anaerobic conditions, clostridia cells can partially detoxify furfural and 5-HMF by reducing them to less cytotoxic derivatives, furfuryl alcohol and 2,5-bis(hydroxymethyl)furan, respectively. Nevertheless, in the real biomass hydrolysates, the actual concentrations of these furan aldehydes are often high, severely impairing microbial metabolism. They can decrease the activities of key enzymes in both glycolysis and TCA cycle, and consequently lower glucose consumption rates. Importantly, the enzymatic reduction of furan compounds consumes intracellular reducing power (NADH, NADPH), which are critically required for solvent biosynthesis, thereby creating a metabolic competition that further compromises butanol formation. In other words, the depletion of NADH and NADPH caused by furan detoxification has direct mechanistic consequences for the butanol biosynthetic pathway, which requires four NADH-dependent reduction steps (i.e., acetoacetyl-CoA to 3-hydroxybutyryl-CoA, crotonyl-CoA to butyryl-CoA, butyryl-CoA to butyraldehyde, and butyraldehyde to butanol) to convert two molecules of acetyl-CoA into one molecule of butanol (Lee S. Y. et al., 2008). Moreover, furan derivatives can undergo hydration under acidic conditions to generate additional weak acids such as formic acid and levulinic acid, leading to a multi-inhibitor scenario that strongly represses cell growth (Hu et al., 2019). In C. acetobutylicum ATCC 824, the inhibition of furfuryl alcohol/furoic acid at high concentrations of >2 g/L was less than that of furfural, which mainly attributed to the detoxification mechanism of furfural in C. acetobutylicum cells (Liu et al., 2019). Systematic studies on C. acetobutylicum revealed that co-stress from furfural, 5-HMF, formic acid, and acetic acid not only reduces the activities of glycolytic and TCA cycle enzymes but also suppresses metabolites and key enzymes involved in solvents synthesis and redox metabolism (Liu et al., 2019). In response, the cells upregulate the activity of peroxidases and proteins associated with gluconeogenesis, oxidative TCA cycling, and oxidative stress, indicating an active stress-induced adaptive response. However, this metabolic regulation comes at a significant cost, as it redirects cellular resources toward survival and maintenance, ultimately diminishing the efficiency of butanol synthesis (Liu et al., 2019). In summary, furan derivatives, especially in mixtures and at elevated concentrations, impose multifaceted inhibitory effects on ABE fermentation through direct enzyme inhibition, depletion of reducing power, and synergistic interactions with weak acids.

2.4 The effects of phenolic compounds on butanol fermentation

In lignocellulose-butanol biorefinery process, lignin-derived phenolics, generated from the degradation of the lignin during pretreatment, represent a class of potent inhibitors for butanol fermentation (Figure 2). The inhibitory mechanisms of phenolics on clostridia mainly include membrane damage, oxidative stress because of the generation of reactive oxygen species (ROS), reduction of substrate transport efficiency, and redox imbalance (Liu et al., 2020; Liu et al., 2019; Liu et al., 2018; Luo et al., 2021b). The fermentation performance of C. acetobutylicum ATCC 824 with the addition of five typical lignin-derived phenolics (vanillin, vanillic acid, p-hydroxybenzaldehyde, p-hydroxybenzoic acid, and ferulic acid) in defined medium reveals that co-existence of phenolics exerts synergistic inhibitory effects on butanol biosynthesis (Luo et al., 2019). The results were consistent with the independent observation that 0.5 g/L p-coumaric acid alone reduces butanol yield by 90%, while 2 g/L p-coumaric or ferulic acid completely abolishes solventogenesis (Ezeji and Blaschek, 2008). The inhibition severity of aromatic aldehydes on C. acetobutylicum metabolism is affected by the ortho substituents (OH > OCH₃ > CHO) and is strongly related to the position of the hydroxyl group (Li et al., 2017). Although their concentrations of phenolic compounds are generally lower than those of furfural and 5-HMF, they exhibit considerably higher toxicity, primarily owing to their lipophilic character (logP ∼1.2–1.5) (Singh and Vermaas, 2026), which facilitates rapid penetration through the cell membrane. This membrane disruption leads to intracellular accumulation of ROS and induces cell death (Luo et al., 2021b). Transcriptomic analyses have revealed distinct inhibitory mechanisms among different phenolic compound. For instance, RNA-seq profiling of C. beijerinckii under 0.5 g/L ferulic acid stress revealed differential expression of genes involved in redox-related cofactors, riboflavin metabolism, two-component systems, and DNA replication (Liu et al., 2017). Further gene co-expression network analysis based on RNA-seq data demonstrated that p-coumaric acid primarily affects ribosomal protein gene expression, flagellar assembly, and DNA recombination, whereas syringaldehyde and vanillin significantly perturb purine, starch, and sucrose metabolism as well as sporulation and signal transduction functions; ferulic acid additionally induces metabolic disorders in glycosyltransferase-related gene expression (Liu et al., 2020).

Comparative studies on phenolic acids versus phenolic aldehydes have shown that phenolic aldehydes exert stronger inhibition on both cell growth and solvent synthesis in C. acetobutylicum, which is attributed to their differential modulation of gene expression related to membrane transporters, the phosphotransferase system (PTS), and glycolytic pathways (Luo et al., 2021b; Luo et al., 2020). In the real lignocellulosic hydrolysates, phenolic acids and phenolic aldehydes co-exist, and thus researchers further investigated the stress mechanisms under combined phenolic compounds. The results indicate that after 12 h of combined phenolic stress, glucose utilization efficiency decreased by 21% (Luo et al., 2021b). RNA-seq analysis revealed that the combined stress conditions induced differential expression of genes associated with critical physiological processes, including sporulation, cell division, antioxidant activity, membrane transporter proteins, and heat shock proteins; a global inhibition mechanism map of solventogenic clostridia in response to phenolic compounds was subsequently constructed (Luo et al., 2021b). Notably, comparative inhibitor studies have demonstrated that low concentrations of formic acid and phenol (0.5 g/L) exert stronger inhibitory effects on butanol fermentation than high-concentration furfural (2 g/L), while low-concentration furans (1 g/L) may enhance fermentation performance; specifically, phenol stress primarily affects fatty acid and nucleic acid metabolism, whereas furfural disrupts intracellular reductive compound accumulation, and formic acid leads to retarded intracellular metabolism and slow cell growth (Liu et al., 2019). The results indicate that phenolic compounds, along with weak acids and furan derivatives, constitute a complex multi-inhibitor system in the real lignocellulosic hydrolysates, with their concentration profiles varying significantly depending on the pretreatment method, and that synergistic inhibitory effects among these inhibitors further complicate the fermentation process. Therefore, a comprehensive understanding of phenolic stress mechanisms is essential for guiding rational strain engineering to enhance inhibitor tolerance in solventogenic clostridia.

3 Engineering strategies for enhancing butanol fermentation performance using lignocellulosic biomass as substrates

3.1 Metabolic engineering strategies for enhancing butanol tolerance and production under stress environments

Microbial detoxification of furan and phenolic inhibitors is primarily achieved through oxidation–reduction reactions, which convert highly toxic aldehydes into less toxic acids and alcohols (Liu et al., 2024; Yi et al., 2015). These resultant products subsequently enter the central carbon metabolism for complete degradation. In contrast, the metabolic disposition of weak acids follows a distinct mechanism, predominantly relying on acetate assimilation and reuptake. Acetate is converted into acetyl-CoA via acetyl-CoA synthetase and channeled into the TCA cycle. Notably, the toxicity and metabolic fates of these inhibitor classes differ markedly. Ezeji et al. demonstrated that C. beijerinckii BA101 exhibits robust tolerance to low concentrations of acetic acid, furfural, and 5-HMF; in fact, furfural and 5-HMF at 3 g/L can be metabolized and even enhance fermentation performance (Ezeji et al., 2007). However, phenolic compounds present a far more formidable challenge. Consequently, improving strain tolerance to phenolic compounds is a critical bottleneck that must be addressed for viable butanol production. To overcome this limitation, several effective strategies have been explored. Adaptive evolution via continuous cultivation has proven successful for isolating inhibitor-tolerant strains. Guo et al. isolated a high-tolerance C. beijerinckii strain that, when using non-detoxified pretreated hydrolysate containing 2.1 g/L phenolics and 40.3 g/L reducing sugars, produced 2.6 g/L acetone, 7.5 g/L butanol, and 0.5 g/L ethanol (Guo et al., 2013). Metabolic engineering has been applied to enhance solventogenesis and acid reassimilation. Lu et al. constructed the recombinant strain C. beijerinckii CC101-SV6 by overexpressing the aldehyde/alcohol dehydrogenase gene (adhE2) alongside the coenzyme A transferase gene (ctfAB) (Lu et al., 2017). Compared to strains expressing only adhE2, C. beijerinckii CC101-SV6 exhibited improved solvent synthesis and organic acid uptake, producing approximately 12 g/L butanol while efficiently reassimilating organic acids and effectively preventing the “acid crash” phenomenon. Additionally, this strain displayed enhanced co-utilization of glucose and xylose, reflecting superior inhibitor tolerance (Lu et al., 2017). However, sole overexpression of adhE2 imposed a metabolic burden during early fermentation, leading to insufficient ATP supply and NADH/NAD⁺ imbalance, which partially prevented acetyl-CoA from being properly metabolized into organic acids. Recently, efflux pump engineering has been investigated to enhance inhibitors’ tolerance. Jiménez-Bonilla et al. overexpressed efflux pump genes from Pseudomonas putida in Clostridium saccharoperbutylacetonicum (Jiménez-Bonilla et al., 2020). Interestingly, while overexpression of the entire gene cassette reduced tolerance, singular overexpression of the subunit gene (srpB) markedly bolstered strain robustness. This engineered strain exhibited enhanced growth in media containing 17% more furfural and 50% more ferulic acid, ultimately achieving butanol titers comparable to those of inhibitor-free control batches (Jiménez-Bonilla et al., 2020). Furthermore, regulatory modulation of PTS has shown pleiotropic benefits. Wu et al. constructed C. acetobutylicum mutants by either overexpressing or knocking out the glucose-specific PTS IICBA-encoding gene (Wu et al., 2019). Given that this gene exerts pleiotropic regulatory functions in glucose utilization and solventogenesis, the engineered strains exhibited improved energy/cofactor regeneration, enhanced butanol production to 12.5 g/L with an increment of 316.7%, and elevated inhibitor tolerance during the exponential growth phase. Above findings highlight that while weak acids and furans can be metabolically assimilated or even exploited to boost performance, phenolic compounds represent the predominant inhibitory barrier. In addition, CRISPR/Cas system has emerged as a highly versatile and powerful platform for genetic manipulation in diverse species including solventogenic clostridia. For instance, improved CRISPR/Cas9 tools for C. acetobutylicum were developed by integrating a xylose-inducible Cas9 cassette from Streptococcus pyogenes into the genome, thereby eliminating background expression. The system achieved 93%–100% editing efficiency across multiple gene deletions (upp, ldhA), including the first clean deletion of the ptb-buk operon, thus enabling rapid and systematic metabolic engineering of this solvent-producing bacterium (Wilding-Steele et al., 2021). Another CRISPR/Cas9 editing tool was also developed based on a two-plasmid system in C. acetobutylicum that was used to investigate the contribution of chromosomal butanol dehydrogenase genes during solventogenesis (Wasels et al., 2020). In addition, genome-scale metabolic models (GSMMs) of C. acetobutylicum have been successfully constructed by semi-automated reverse engineering algorithm, and flux balance analysis and nonlinear programming (Lee J. et al., 2008; Senger and Papoutsakis, 2008). The diverse engineering approaches, ranging from adaptive evolution and metabolic pathway reinforcement to efflux pump modulation and PTS regulatory engineering, provide a multifaceted theoretical and practical framework for developing robust solventogenic clostridial strains with enhanced inhibitor tolerance.

3.2 Consolidated bioprocessing for enhancing butanol production

Consolidated bioprocessing (CBP) integrates feedstock hydrolysis, enzymatic saccharification, and subsequent solvent fermentation into a single operational step (Zhao M. et al., 2022; Zhao et al., 2019), thereby enabling solventogenic clostridia to directly convert lignocellulosic substrates into butanol and other solvents. This strategy might reduce the estimated minimum butanol selling price (MBSP) (Wen et al., 2020). A prominent implementation of this concept involves the co-culture of two or more microorganisms that jointly utilize the substrate, simultaneously alleviating inhibitor toxicity while sustaining solvent synthesis. Technically, co-culture systems capitalize on functional complementarity among different microbial partners, broadening the substrate spectrum, improving tolerance to both products and inhibitors, and achieving efficient solvent production (Fu et al., 2025; Xin et al., 2017).

Several representative studies illustrate the potential of this approach. Wen et al. developed a dual-clostridia CBP system comprising Clostridium cellulovorans and C. beijerinckii (Wen et al., 2020). By knocking out the acetate kinase gene (Clocel_1892) and lactate dehydrogenase gene (Clocel_1553) while overexpressing the butyrate kinase gene (Clocel_3674), they redirected carbon flux toward butyrate formation. Concurrent overexpression of the organic acid reassimilation genes (ctfAB) further accelerated acetate conversion. This dual-strain system successfully accomplished cellulase production, lignocellulose hydrolysis, and co-fermentation of hexoses and pentoses in one step, generating butanol directly (Wen et al., 2020). In another study, Cui et al. co-cultivated C. beijerinckii G117 with a recombinant Bacillus subtilis 1A1 (Cui et al., 2020). The Bacillus subtilis consumed dissolved oxygen to establish an anaerobic environment for C. beijerinckii, and the lactate produced by B. subtilis was re-assimilated by C. beijerinckii. Furthermore, B. subtilis 1A1, engineered to express a secondary alcohol dehydrogenase, converted acetone from C. beijerinckii G117 into isopropanol, thereby maintaining intracellular redox balance. Recently, the co-culture of engineered Clostridium tyrobutyricum and Clostridium ljungdahlii was also developed for enhancing butanol production and carbon recovery from glucose (Li et al., 2024). Similarly, improvement of the interspecies electron transfer in C. acetobutylicum and C. ljungdahlii co-culture system can be achieved by a hybrid biochar-methyl viologen (BM) composite (Li et al., 2026). The BM-assisted co-culture system produced >17 g/L butanol with higher yield of 0.39 g/g in glucose-based medium. By exploiting the distinct physiological, biochemical, and metabolic traits of different strains, these co-culture systems not only achieved in situ degradation of fermentation inhibitors but also enabled simultaneous production of target solvents.

Nevertheless, several limitations persist. In many microbial co-culture systems, the molecular-level interactions between partner strains remain poorly understood. Industrial application faces additional challenges, including the acquisition of suitable CBP-compatible strains, efficient co-expression of multiple hydrolytic enzymes, enhancement of overall fermentation synergy, and precise regulation of enzyme composition ratios. To improve system performance and foster more effective microbial interactions, future research should focus on controlling oxygen transfer rates, modulating population ratios, and elucidating the molecular mechanisms governing co-culture dynamics. Addressing these key issues will be essential for translating CBP-based co-culture strategies into robust, economically viable industrial processes.

3.3 Development of efficient pretreatments and fermentation strategies for enhancing butanol bioproduction from lignocellulosic biomass

Pretreatment has become a determining factor for both sugar recovery and subsequent solventogenic fermentation. The development of pretreatment technologies for lignocellulosic butanol production has followed a clear progression from conventional acid and alkaline pretreatments to physicochemical fractionation methods and, more recently, to green solvent systems designed to improve biomass deconstruction while reducing environmental burdens (Dharmaraja et al., 2023; Ding et al., 2024; Long et al., 2025). To illustrate this trend, Table 1 summarizes recent clostridia-based fermentation results obtained across a wide spectrum of pretreatment methodologies, from dilute acid and alkali to steam explosion, ILs, and DESs.

TABLE 1. Butanol fermentation performance by solventogenic clostridia using pretreated lignocellulosic biomass as substrates.

T _BuOH: Titer of butanol (g/L); P _ABE: Productivity of total ABE (g/L/h); Y _ABE: Yield of total ABE, from fermentable substrates (g/g); N/A: not available.

: Biphasic pretreatment is p-toluenesulfonic acid/pentanol co-solvent with AlCl₃ addition.

: [TEA][HSO₄] is a protic ionic liquid (IL) consisting of triethylammonium hydrogen sulfate with a molar ratio of 1:1.

: TEBAC/LA, is a DES, prepared by triethylbenzyl ammonium chloride and lactic acid with a molar ratio of 1:7.

: CC-EG-Ni is a ternary deep eutectic solvent (DES) prepared by choline chloride: ethylene glycol: nickel chloride hexahydrate with a molar ratio of 1:2:0.016.

: ChCl-SA-Gly is a DES, prepared with choline chloride, succinic acid, and glycerol at a molar ratio of 1:0.5:0.5.

Dilute sulfuric acid pretreatment excels at hemicellulose solubilization, yet the generation of furans and phenolics often depresses microbial metabolism. Jawad et al. (2024) applied 1% H₂SO₄ at 121 °C for 15 min to pretreat CS and obtained a butanol titer of only 2.4 g/L, despite a remarkably high ABE productivity (0.55 g/L/h) and yield (0.42 g/g) in a co-culture of Saccharomyces cerevisiae and C. beijerinckii (Jawad et al., 2024). In this co-culture system, S. cerevisiae SG could metabolize acetic acid, HMF, furfural and vanillin from 0.2–1.3 g/L to 0.05–0.3 g/L. The low butanol titer points to incomplete sugar release, while the high ABE productivity suggests that the co-culture rapidly consumes the available substrates for producing ethanol as the main product. In contrast, a two-stage H₂SO₄/aqueous ammonia pretreatment was used for CS fractionation, achieving 10.8 g/L butanol by a thermotolerant C. acetobutylicum L7 (GlcG) at 42 °C with SSF strategy (Wu et al., 2021). In this case, ABE productivity of 0.30 g/L/h and a yield of 0.31 g/g were obtained. Alkaline post-treatment likely neutralizes acids and precipitates inhibitory compounds, effectively decoupling the solubilization step from detoxification process. Alkaline, mainly NaOH, pretreatment selectively removes lignin by cleaving β-O-4 ether bonds in lignin via nucleophilic attack by hydroxide ions (Mensah et al., 2022; Shakeel et al., 2022), but its performance varies drastically with biomass and strain types. Pretreatment of triticale straw by 1% NaOH at 180 °C for 1 h retained 72% cellulose, and 65% lignin and 91% hemicellulose were removed. In this case, the hydrolysates from the pretreated straw can be used by C. acetobutylicum NRRL B-591 to produce 8.4 g/L butanol (Rezaei et al., 2021), while 8.95 g/L butanol was obtained by C. acetobutylicum ATCC 824 from the pretreated giant Juncao grass under milder conditions (2.5% NaOH, 82 °C, 60 min) with semi-SSF mode (Sonwai et al., 2025). These titers are competitive, but the severe alkaline conditions generate large volumes of saline wastewater. Recently, water hyacinth and Japanese bamboo were also pretreated by 2%–5% NaOH at 100 °C–120 °C, the butanol productions were only 0.5–2.09 g/L with lower productivities of ∼0.01 g/L/h (Chinwatpaiboon et al., 2023; Tri et al., 2024).

To reduce chemical consumption and improve process economics, physicochemical pretreatments such as hydrothermal processing and steam explosion have received increasing attention. Hydrothermal pretreatment uses high-temperature water as the reaction medium and therefore avoids the addition of strong acids or bases (Ruiz et al., 2021; Singh et al., 2023). Under hydrothermal conditions of 200 °C for 10 min, pretreated switchgrass hydrolysate containing ∼100 mg/L inhibitors (furfural, HMF, vanillin, syringaldehyde) was metabolized by C. beijerinckii to produce butanol of 11.14–11.42 g/L under fed-batch fermentation, with ABE productivities of 0.29 g/L/h and yields ranging from 0.31 to 0.32 g/g (Adesanya et al., 2022). Similar results were obtained in batch culture, where hydrothermally pretreated switchgrass produced 11.21 g/L butanol and an ABE yield of 0.35 g/g (Olorunsogbon et al., 2022). In contrast, sugarcane bagasse subjected to hydrothermal pretreatment generated only 6.33 g/L butanol (Zetty-Arenas et al., 2021). These differences suggest that pretreatment efficiency cannot be separated from feedstock characteristics. Nevertheless, hydrothermal processing demonstrated that competitive butanol production could be achieved without introducing additional chemicals into the system.

The limitations of individual pretreatment methods subsequently stimulated the development of hybrid fractionation strategies. A biphasic pretreatment (p-toluenesulfonic acid/pentanol with AlCl₃) of poplar, with >80% delignification, >95% xylan removal, and minimal cellulose loss of 7.8%, achieved 10.5 g/L butanol with an ABE yield of 0.33 g/g, while hybrid pretreatment of Pleurotus ostreatus spent substrate generated 9.84 g/L butanol and an ABE yield of 0.40 g/g (Madadi et al., 2025), one of the highest values reported from pure cultures (Table 1). Such results illustrate that preserving carbohydrates while selectively removing lignin is more beneficial for butanol fermentation than maximizing biomass disruption alone. Steam explosion further shifted attention from laboratory-scale fractionation toward process scalability. CS subjected to steam explosion at 170 °C–190 °C which finally produced 10.88 g/L butanol by C. acetobutylicum ABE P-1201 in a 1 m³ fed-batch fermentor, with an ABE productivity of 0.14 g/L/h and a yield of 0.32 g/g (Su et al., 2024). The significance of this study lies not only in the achieved titer but also in the demonstration that lignocellulosic butanol fermentation can be maintained at pilot scale. Consequently, the objectives of pretreatment research have gradually expanded from maximizing sugar release to improving process robustness and facilitating industrial implementation.

Although conventional pretreatments have enabled substantial progress in lignocellulosic butanol production, several challenges remain unresolved. High energy input, intensive chemical usage, wastewater generation, and incomplete lignin valorization continue to compromise process sustainability. These issues have driven the emergence of solvent-based fractionation technologies. ILs were among the green solvents explored for lignocellulose deconstruction because of their exceptional ability to disrupt intermolecular hydrogen bonding. Microwave-assisted pretreatment of corn stover using the protic ionic liquid [TEA][HSO₄] removed 93.2% of xylan and 72.9% of lignin (Wang et al., 2024). The [TEA][HSO₄]-pretreated corn stover hydrolysate produced 8.65 g/L butanol with an ABE productivity of 0.13 g/L/h and a yield of 0.33 g/g. Although the resulting butanol titer was comparable to those obtained by conventional alkaline pretreatment (Table 1), effective fractionation was achieved within only 3 min of microwave irradiation. This considerable reduction in processing time highlighted the potential of ILs for process intensification. Nevertheless, solvent cost, recyclability, and microbial compatibility remain major constraints that have prevented their widespread application. These considerations have accelerated the transition toward DESs, which retain many advantages of ILs while employing inexpensive and often biodegradable components (Novia et al., 2026; Xie et al., 2026). Recent studies have already demonstrated their effectiveness. CS pretreated with TEBAC-LA could be used by C. beijerinckii P260 for producing 8.53 g/L butanol together with an exceptionally high ABE productivity of 0.68 g/L/h and an ABE yield of 0.38 g/g, indicating rapid sugar assimilation and efficient solventogenesis (Lin et al., 2021). Further improvements were achieved with choline chloride-based DESs. Rice straw pretreated with ChCl-LA-EG produced 9.62 g/L butanol with an ABE productivity of 0.33 g/L/h and a yield of 0.35 g/g (Poy et al., 2023). More recently, pretreatment of corn stover with ChCl-SA-Gly (1:0.5:0.5) at 120 °C for 3 h achieved 88.7% glucan retention and 65.4% xylan removal (Luo et al., 2024). The obtained hydrolysate increased butanol production to 11.9 g/L in fed-batch fermentation by C. acetobutylicum ATCC 824, and integration with a yeast-Clostridium co-culture further elevated the butanol concentration to 12.4 g/L with a productivity of 0.36 g/L/h, representing the highest butanol titer reported from the real lignocellulosic hydrolysates (Table 1). In addition, even water-containing ChCl-SA-Gly DES systems maintained reasonable performance, producing 6.5 g/L butanol from reed straw (Shao et al., 2025).

In summary, the pretreatment of lignocellulose for butanol fermentation has progressed from intensive chemical and thermal methods towards solvent-based fractionation with lower energy and reagent footprints. Acid and alkaline routes, though capable of achieving butanol titers of 8–11 g/L, are constrained by inhibitor generation, high salt loads, and inconsistent performance across feedstocks. Hydrothermal and steam explosion reduce chemical inputs but remain energy-intensive and substrate-dependent, as illustrated by the 11.4 g/L from switchgrass versus 6.3 g/L from bagasse under similar hydrothermal conditions. The introduction of ILs shortened pretreatment time to minutes, yet solvent cost and recovery limit their practicality. DESs have since emerged as a more viable alternative, offering comparable delignification at 100 °C–120 °C with biodegradable components. The fermentation data substantiate this advantage. DES-pretreated CS yielded 12.4 g/L butanol in fed-batch co-culture and a productivity of 0.68 g/L/h in a separate study (Table 1). These performance gains stem not from higher sugar release alone, but from reduced inhibitor carryover, which allows more complete substrate conversion and faster solventogenesis.

4 Discussion and future perspectives

n-Butanol offers distinct fuel properties that ethanol cannot match. However, biobutanol production remains economically uncompetitive when based on food-derived substrates such as corn, sugarcane, and starch-based substrates (Luo et al., 2017; Xue et al., 2013). Lignocellulose circumvents the food-fuel conflict, but its conversion suffers from inhibitor formation during pretreatment, a bottleneck that affects every downstream step. The formation of inhibitors is inherent to pretreatment chemistry and varies with biomass composition and process severity. Weak acids, furan derivatives, and phenolics are the three major classes, each exerting distinct but overlapping toxic effects. Weak acids lower intracellular pH and force cells to expend ATP for proton extrusion, creating an energy drain that impairs growth. Furans, including furfural and 5-HMF, consume reducing equivalents such as NADH and NADPH, which are precisely the cofactors required for butanol biosynthesis. Phenolic compounds, though present at lower concentrations than acids and furans, are the most potent because of their lipophilic character, which allows rapid membrane penetration and provokes oxidative stress. More critically, these inhibitor classes act synergistically; their combined effect in the real hydrolysates routinely exceeds the sum of individual toxicities. The specific inhibitor profile is not fixed but depends on feedstock type, temperature, residence time, catalyst loading, and the presence of stabilizers (Luo et al., 2021a). This complexity makes empirical optimization laborious and often inconclusive. Data-driven approaches, particularly machine learning models trained on pretreatment parameters, offer a systematic means to forecast inhibitor yields and to identify operating units that balance sugar recovery against microbial compatibility. This strategy would reduce reliance on post-treatment detoxification, which adds cost and loses sugar.

Parallel to these predictive efforts, pretreatment technology itself has undergone a clear evolution. The acid and alkaline treatments, while capable of delivering titers of 8–11 g/L for certain feedstocks, demand high energy and chemical inputs and produce substantial inhibitor loads. Hydrothermal processing and steam explosion reduce chemical consumption but remain energy-intensive and feedstock-sensitive. ILs shortened pretreatment time to minutes but are constrained by high solvent cost and recycling difficulties (Verdía Barbará et al., 2025). DESs have emerged as a more practical alternative, offering competitive delignification at moderate temperatures with biodegradable, inexpensive components (Mqoni et al., 2026). The butanol fermentation performances confirm this advantage, that DES-pretreated corn stover produced the highest reported butanol titer of 12.4 g/L based on a ternary DES pretreatment (Luo et al., 2024). This outcome reinforces that pretreatment selectivity, meaning the preservation of carbohydrates while minimizing lignin degradation products, is as important as the absolute sugar yield. Meanwhile, pilot-scale demonstration (>1 m³) and long-term solvent recycling (>50 cycles) remain to be demonstrated before DES-based processes can be considered commercially ready.

A systems perspective is essential because pretreatment, inhibitor generation, hydrolysis, fermentation, and product recovery are tightly interlinked. For example, increasing pretreatment severity may boost sugar yield but simultaneously elevates inhibitors that depress fermentation and complicate downstream purification (Jönsson and Martín, 2016). Future process designs should incorporate integrated monitoring of inhibitor levels, adaptive fermentation control, and in situ product removal to maintain balanced performance across the entire chain. Beyond these engineering considerations, the utilization of biomass components demands rethinking. Current practice focuses almost exclusively on cellulose and hemicellulose, while lignin, which accounts for 15%–25% of the lignocellulosic biomass, is typically burned for heat or discarded (Liu and Cheng, 2024; Schutyser et al., 2018). This approach undervalues a significant fraction of the feedstock and undermines overall process economics. Emerging technologies for lignin depolymerization into aromatic monomers, resins, or specialty chemicals should be incorporated into the biorefinery. Doing so requires fractionation conditions that preserve lignin’s native structure for subsequent upgrading, a constraint that must guide the design of next-generation solvents and pretreatment protocols.

In summary, several issues should be focused on lignocellulose-butanol biorefinery pathway. First, the development of recyclable, low-cost DES systems with tunable selectivity for carbohydrate release and lignin preservation. Second, engineering of Clostridium strains with broad tolerance to weak acids, furans, and phenolic mixtures and efficient co-utilization of glucose and xylose, moving beyond single-gene modifications toward systems-level rewiring of redox and stress responses. A comparative analysis of the impact of adaptive evolution, metabolic engineering, efflux pump engineering, and co-culture indicates that no single strategy is universally optimal. The greatest near-term impact is likely to come from the integration of metabolic engineering for targeting redox balance and solventogenesis with co-culture for in situ detoxification. Third, the establishment of consolidated or semicontinuous processes that combine pretreatment, enzymatic hydrolysis, and fermentation into fewer unit operations, thereby reducing capital and operating costs. Fourth, the active modulation of inhibitor formation through optimized pretreatment parameters guided by predictive models, supplemented by biological detoxification rather than post-treatment washing. Fifth, the rigorous techno-economic analysis (TEA) and life-cycle assessment (LCA) that covers the entire value chain, including solvent recycling and lignin valorization. Recent TEA reported a wide range of the minimum butanol selling price (MBSP) for lignocellulosic biobutanol, typically ranging from approximately US$ 0.5–2.2/kg, depending on feedstock cost, conversion pathway, fermentation performance, and downstream recovery strategy (Mailaram and Maity, 2022; Patel et al., 2025; Qureshi et al., 2020). The MBSP is quite competitive with the current butanol selling price of approximately US$ 1.5/kg (Patel et al., 2025), but large-scale commercialization of biobutanol still faces several challenges. TEA studies identify feedstock cost as the primary economic bottleneck for biobutanol production. Low product titer arising from butanol cytotoxicity in ABE fermentation necessitates in situ product recovery via integrated techniques such as adsorption, membrane separation, liquid-liquid extraction, or gas stripping, to alleviate inhibition and reduce downstream separation energy (Xue et al., 2017; Zhang et al., 2026a). In addition, field-to-wheel life cycle GHG emission for the n-butanol bioconversion is similar at 4.5 kg CO₂-eq/gasoline gallon equivalent (GGE) (Tao et al., 2014). The transition from conventional methods to green solvents is not an endpoint, but a necessary element of an integrated biorefinery where each component of the plant cell wall contributes to the final product. Achieving this vision requires coordinated advances in chemistry, biology, and process engineering, guided by whole-system evaluation rather than isolated optimization.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Qiang Yang, Michigan State University, United States

Reviewed by: Kgodiso Judith Mabaso, Vaal University of Technology, South Africa

Karla D. González-Gloria, Universidad Autónoma de Coahuila, Mexico