Cutting-edge technological advancements in biomass-derived hydrogen production

Abstract

Production of hydrogen as carbon-free energy from renewable organic waste biomasses has been adopted for the long-term sustainability of a circular economy through various chemical and biological conversion processes. Conversion of waste biomasses to hydrogen provides dual benefits of low-cost energy-dense biofuel production and simultaneous waste reduction in eco-friendly valorization. Advancements in existing chemical and biological processes through light-induced photoreformation and microbial syntrophy-mediated metabolic induction in fermentation, respectively, facilitated holistic conversion of biowaste for maximum recovery of hydrogen by minimizing by-product generation. This review focuses on various thermochemical, photocatalytic reformation, and biological processes involving direct or indirect conversion of solid organic biomasses to hydrogen and their possible technological advancements to generate waste-to-value-added products. The techno-economic assessment describes the feasibility of waste biomass-derived hydrogen production over other technologies for industrial implementation.

Graphical abstract

1 Introduction

Energy is a key part of sustainable societal advancement in the twenty-first century (Zhu et al. 2020). The worldwide energy consumption is expected to increase by 1.5% per year from ~ 1.8 × 10⁵ terawatts (TW) in 2020 to approximately 2.6 × 10⁵ TW by 2050 (Nalley and LaRose 2021). Intensive utilization of fossil fuel resources in industrial, municipal, and agricultural purposes continues to emit greenhouse gases, producing global warming and other negative consequences (Bajracharya et al. 2016). An ever-increasing worldwide demand for energy and the concomitant environmental pollution have led to the search for clean and sustainable energy systems (Saha et al. 2016). Renewable energy sources (e.g., wind, solar, and hydraulic power) are explored as feasible substitutions to conventional fossil fuels; however, their application is restricted due to spatial and temporal barriers (Zhu et al. 2020). Despite the increase in renewable energy usage, fossil energy resources remain a leading source of energy production, generating ~ 80% of overall energy (Tekade et al. 2020; Yuksel and Ozturk 2017). The CO₂ emissions are mainly produced from burning of fossil fuels, which also generates several other noxious gases, such as NOx and SOx. Such environmental issues motivated the scientific community in constant search for new alternative sources of energy, specifically renewable ones to reduce dependency on fossil fuels (Gao et al. 2018).

Hydrogen (H₂), a carbon-free energy carrier with the highest known energy density (142 kJ g⁻¹), is considered a cutting-edge source of clean energy (Chang et al. 2018). Producing hydrogen through adaptable and eco-friendly processes is important for the sustainable hydrogen economy and future of clean energy (Tu et al. 2017; Zhu et al. 2020). However, conventional H₂ production from nonrenewable resources, i.e., coal gasification, hydrocarbon reforming, plasma reforming, steam reforming of natural gas, water electrolysis, etc., is unsustainable for the circular economy due to their high carbon-footprint (~ 830 million tons per year) and energy-intensive process (Megía et al. 2021; Singh et al. 2015). Photocatalytic or photoelectrocatalytic green hydrogen production has gained attention for its long-term sustainability because it facilitates water splitting under visible light for solar-to-chemical energy; however, it provides low productivity of 5% (Hendi et al. 2020; Lu et al. 2017). Nonetheless, photocatalytic reforming or photoreforming through the combination of light-induced H₂ generation from water and oxygenated organic biomasses such as alcohols and carboxylic acids produced from biomass fermentation; saccharides; biopolymers has appeared to be a highly important intermediary route between photocatalytic water splitting and photo-oxidation (Fig. 1). The conversion of organic biomass-derived oxygenates into H₂ gas through photoreforming involves energy input in the form of light radiation (Puga 2016).

Fig. 1

Generation of H₂ (i) and O₂ (iii) through water splitting under anaerobic conditions. Oxidation of organic biomass-derived oxygenates (iv) occurs in aerobic conditions, followed by reduction to water (ii). All reactions require light absorption onto irradiated M/TiO₂ (M = noble metal) photocatalyst to generate separate charges, i.e., e⁻ in the conduction band (CB) and h⁺ in the valence band (VB). Thus, biomass photoreformation can occur under anaerobic conditions in combination with water splitting and biomass oxidation

Utilization of renewable resources including organic waste biomasses (e.g., lignocellulosic agricultural wastes, fruit and food wastes, fat, oil and grease) and municipal/industrial wastewater for the production of H₂ appears to be the most convenient alternative to conventional processes for the reduction of production cost (Chang et al. 2018; Godvin Sharmila et al. 2022; Saha et al. 2019). Nearly 181.5 billion tons of lignocellulosic agricultural biomass is generated around the globe annually, although only a small percentage of it is refined and reused, leaving enormous amounts of organic waste (40–50% of their original mass) in the environment (Granone et al. 2018; Orozco et al. 2014; Talavera-Caro et al. 2020). The Food and Agricultural Organization of the United Nations estimated that the disposal of food waste has reached ~ 1.3 billion tons globally (Karmee 2016). Utilization of renewable wastes in hydrogen production diminishes their natural uncontrolled degradation and reduces the environmental threat of global warming (Saha et al. 2018). Adopting suitable approaches compatible with the circular economy concept to recycle renewable waste resources to generate hydrogen energy are necessary to limit uncontrolled degradation of these waste materials and their by-products (Kim et al. 2016). Small-scale, inexpensive, electron-driven chemical processes are crucial for decentralized transformation of biomass to produce high-value specialty chemicals or lower-value commodity chemicals or fuels (Akhade et al. 2020). Electrolytic hydrogen yields of 0.1 − 0.2 mg mg⁻¹ of raw biomass have been achieved at a cathode utilizing cellulose, starch, lignin, protein, and lipids of different biomasses (Hibino et al. 2018). Hydrothermal carbonization of food waste followed by steam gasification generated 28.08 mmol H₂ g⁻¹ dry waste (Duman et al. 2018). The theoretical potential of hydrogen yield through hydrogenogenic dark fermentation of various lignocellulosic biomasses could reach 14–21 g H₂ kg⁻¹ substrate (Sołowski et al. 2020). Therefore, waste organic biomasses could be an alternative feedstock for sustainable energy sources such as H₂.

In this relevance, this review discusses the fundamental and practical features of various biomass-derived hydrogen production technologies, including thermochemical, photocatalytic, and biological processes and their possible technological advancements to generate waste-to-value-added products. The assessment of techno-economic features describes the feasibility of biomass-derived hydrogen production for industrial implementation of advanced technologies.

2 Hydrogen generation through chemical pathways

Production of hydrogen is feasible from biomass feedstock by numerous chemical and biological processes, including (1) direct gasification, where raw lignocellulose biomass is converted to syngas containing H₂ and CO as major constituents and (2) indirect conversion of biomasses to chemical intermediates and subsequent reformation, where oligomeric oxygenated compounds are formed by pyrolysis/hydrolysis followed by reformation into gaseous H₂ and CO₂ (Fig. 2) (Alonso et al. 2010; Navarro et al. 2009).

Fig. 2

Conversion of lignocellulosic biomass to produce H₂ through direct and indirect pathways

Biomass gasification is one of the oldest technologies to produce electricity and heat, having been employed since the 1940s (Huber et al. 2006a). In this process, liquid or solid carbonaceous substrates, such as lignocellulosic biomasses react with oxygen, air, and/or steam at comparatively lower temperatures (700–1000 °C) than coal gasification to form syngas, which consists of H₂, CH₄, CO, CO₂, and N₂. Combined reactions of steam gasification, pyrolysis, and partial oxidation result in solid, liquid, and gas phases in biomass gasification (Table 1) (Puga 2016). In steam reforming, the feedstock is reacted with water to yield CO₂, CO, and H₂. Pyrolysis involves thermal disintegration of the biomass feedstock into solid, liquid, and gaseous products in absence of steam or oxygen. Partial oxidation methods involve less O₂ than required by stoichiometry for combustion. The water–gas shift reaction (WGSR, where generation of H₂ and CO₂ occurs through a reaction between CO and water) and water–gas shift methanation (where the interaction of H₂ and CO produces H₂O and CH₄) also take place in gasification. To promote the gasification reactions, heat is produced through direct gasification (exothermic reaction and partial combustion within the gasifier generate heat) or indirect gasification (heat is formed at the outer part of the gasifier and moves inside) (Sikarwar et al. 2016).

Table 1 Reaction mechanisms and enthalpy of cellulosic biomass gasification reactions

2.1 Direct gasification

The direct gasification method involves primary, secondary, and tertiary processes (Fig. 3). In the first gasification stage, gaseous CO₂, H₂O, oxygenated vapors, and primarily oxygenated liquids are produced as main products from solid biomass along with a few by-products including cellulosic molecules, lignin-based methoxyphenols, hydroxyacetaldehyde, levoglucosan, and their corresponding hemicellulosic derivatives (Usino et al. 2021). Initial pyrolysis of the organic substances generates nonreactive lower-molecular weight vapor containing monomers and their fragments. This vapor is free of cracking derivatives of subsequent gas-phase reactions. Charcoal is produced from slow pyrolysis, which preserves the characteristics of lignocellulose. In the second step, the primary liquids and vapors produce gaseous aromatics, olefins, CO₂, CO, H₂, H₂O, and secondary compressed oils like aromatics and phenols. The primary vapors create cracking (regimes of secondary interaction) in the biomass upon heating above 500 °C, while the temperature regime of the secondary reaction is within 700–850 °C. The products from the secondary reaction enter the tertiary reaction as temperature increases to 850–1000 °C, generating CO₂, CO, H₂O, H₂, and polynuclear aromatics (PNA) substances containing methyl-derived aromatics like methyl naphthalene, methyl acenaphthylene, indene, and toluene (Sikarwar et al. 2016). A liquefied tertiary phase is generated from condensation of some tertiary substances including naphthalene, benzene, anthracene/phenanthrene, acenaphthylene, and pyrene. Coke and soot are generated during the secondary and tertiary reactions. Coke is produced through thermolysis of organic vapors and liquids, while the decomposed yields of hydrocarbons after homogeneous nucleation at high temperature in the gaseous state produce soot (Alonso et al. 2010). Usually, the inorganic constituent of the gasification feedstock is transformed into bottom ash (further eliminated from reactor) or fly ash (leaves along with product gas). The constituents of ash include K₂O, CaO, P₂O₅, SiO₂, MgO, Na₂O, and SO₃. The operating temperature should be below 1000 °C (at which ash melts) to restrict slagging and sintering of ash (Herman et al. 2016).

Fig. 3

Process elucidating the biomass gasification. Pyrolysis and gasification severity increase with temperature and time in direct gasification of biomass, producing H₂, CO₂, CO, and H₂O as major products

The composition of gas released from the gasification vessel is dependent upon the composition of biomass feedstock, the gasifying mediator, and the gasification method (Navarro et al. 2009). Tars containing heavier hydrocarbons cause clogging and blocked filters as they agglomerate on certain filters and in downstream pipes (Devi et al. 2003). Tars can also produce other downstream complications and choke fuel outlines and injectors in interior combustion engines. Choosing a suitable gasification reactor and conditions can reduce the quantity of tars. Addition of solid catalysts including Ni, Pd, Pt, and Ru, reinforced on dolomite, Rh/SiO₂/CeO₂ and CeO₂/SiO₂ within the gasification vessel can reduce the concentration of tar (Tomishige et al. 2004). Wet impregnation or dry mixing of alkali metal catalysts including K₂CO₃, Na₃H (CO₃)₂, Na₂CO₃, CsCO₃, Na₂B₄O₇·10H₂O, NaCl, ZnCl₂, KCl, and AlCl₃·6H₂O with biomass derivatives can decrease tar generation (Sutton et al. 2001). However, alkali metals are not profitable due to char formation, increased ash content, reduced carbon transformation, and the complicated recycling process. Nevertheless, a mixture of CO, H₂, CO₂, and some CH₄ is produced along with other gaseous ingredients and solid offshoots like char through a one-step transformation of raw or mechanically pretreated biomass feedstock in a direct gasification process under oxygen and steam environments at high temperature (750–1000 °C) (Alonso et al. 2010; Navarro et al. 2007). Tar generation can be reduced through movement of the gaseous products of the secondary reaction over the char bed as a catalyst (Sikarwar et al. 2016). Despite its limitations of harsh conditions, this adaptable gasification process can be altered for a varied array of biomass derivatives to achieve hydrogen yields of 50–80 g H₂ kg⁻¹ biomass (Parthasarathy and Narayanan 2014).

2.2 Indirect processes based on reforming and photoreforming of oxygenates

2.2.1 Reforming of oxygenates

The indirect process is a combination of primary liquification by pyrolysis to produce bio-oils and mixed oxygenated elements and by pretreatment/hydrolysis and fermentation of cross-linked, heavier biopolymers (lignin, hemicellulose, and cellulose) into mixed oxygenated chemical ingredients having lower molecular weight, followed by catalytic reformation (Demirbaş 2001; Zinoviev et al. 2010). Producing H₂ via reformation of pyrolysis oils is a consistent route for reclamation of complex and unstable biomasses such as lignocellulosic agricultural wastes, food waste, biopolymers, etc. H₂-enriched gaseous streams are produced by reforming bio-oils in water (either liquid or steam) at high temperatures (~ 850 °C) with low retention time on reinforced Ni or noble metal catalysts (Rioche et al. 2005). Another procedure involves reformation in aqueous phase under milder temperatures (250–350 °C) (Vispute and Huber 2009). Nevertheless, the required temperatures for the reforming processes are high (250–850 °C) in both cases. Compounds with functional groups like carboxylic acids or carbonyls can be degraded (through condensation, cyclisation, etc.), while saccharides can be decomposed by dehydration, polymerization, and so on. Therefore, thermo-catalytic reforming suffers due to formation of unwanted carbonaceous byproducts, which can deactivate the catalyst and initiate pyrolysis of saccharides even before their interactions with catalyst media to form CO in large amount (Fu et al. 2005).

Reforming of various oxygenates derived from hydrolysis of lignocellulose in an aqueous phase followed by sequential transformations is another indirect pathway for biomass-to-hydrogen production (Cortright et al. 2002; Huber et al. 2003). As stated by Dumesic et al., usage of appropriate catalysts such as Pd, Pt, or Ni–Sn catalysts (desirably sustained on Al₂O₃) can regulate the selectivity of this process by minimizing alkane production (Davda et al. 2005). These catalysts are highly active for water–gas shift reactions (WGSR), C–C scission, and dehydrogenation to produce fewer side reactions, such as methanation or C–O scission. The formation of light oxygenates from lignocellulosic biomass and its subsequent reformation in the aqueous phase to yield H₂ selectively at milder conditions in comparison to the one-step processes (gasification) indicates the indirect process as a candidate for biomass-to-hydrogen production. However, the thermo-catalytic aspect is energy exhaustive and might require high pressures (~ 35 bar) and temperatures (550–900 °C) (Lewis et al. 2003; Naikoo et al. 2021). This causes side reactions or degradation, which have unfavorable impacts on H₂ selectivity (50–60% of the theoretic maximum) and produce excessive amounts of alkanes (Huber and Dumesic 2006). In this scenario, methods of augmenting the reformation (Eq. 1) while limiting other conversions are desired and would be advantageous for biomass-derived H₂ production technologies.

$${\text{C}}_{x} {\text{H}}_{y} {\text{O}}_{z } + \left( {2x - z} \right){\text{H}}_{2} {\text{O}} \to \left( {2x + y/2 - z} \right){\text{H}}_{2} + x{\text{CO}}_{2}$$

(1)

There are four substantial reformation technologies for transforming biomass to hydrogen: catalytic steam reforming (CSR), catalytic partial oxidation (CPO), auto-thermal reforming (ATR), and aqueous phase reforming (APR) (Zinoviev et al. 2010). CSR, the conventional reforming process, is adopted to produce H₂ from fossil hydrocarbons, specifically methane. This technique is used to treat oxygenated substrates (C_nH_mO_k, Eqs. 2 and 3) along with stoichiometric WGSR process for augmenting H₂ production.

$${\text{C}}_{n } {\text{H}}_{m} {\text{O}}_{k} + \left( {n - k} \right){\text{H}}_{2} {\text{O}} \leftrightarrow n{\text{CO}} + \left( {n + m/2 - k} \right){\text{H}}_{2}$$

(2)

$${\text{C}}_{n } {\text{H}}_{m} {\text{O}}_{k} + \left( {2n - k} \right){\text{H}}_{2} {\text{O}} \leftrightarrow n{\text{CO}}_{2} + \left( {2n + m/2 - k} \right){\text{H}}_{2}$$

(3)

Reforming of biomass derivatives such as methanol, glycerol, ethylene glycol, and sorbitol is thermodynamically more promising at low temperatures (~ 277 °C) with respect to hydrocarbons having similar numbers of C atoms (Davda et al. 2005). Therefore, the reforming procedure is more suitable for use with the WGSR than is hydrocarbon reforming as there is a possibility of further reactions of CO₂ and CO with H₂ at lower temperatures (27–377 °C), producing alkanes by Fischer–Tropsch synthesis (FTS) or methanation (Shabaker et al. 2004; Vaidya and Lopez-Sanchez 2017). In addition, gas phase homogeneous thermal disintegration of oxygenated hydrocarbons and cracking reactions (Eq. 4) on the acidic parts of the catalytic surface decrease the selectivity and disable the catalyst.

$${\text{C}}_{n } {\text{H}}_{m} {\text{O}}_{k} \to {\text{C}}_{x} {\text{H}}_{y} {\text{O}}_{z} + \left( {{\text{H}}_{2} , {\text{CO}}, {\text{CO}}_{2} ,{\text{CH}}_{4} ,{\text{Hydrocarbons}}} \right) + {\text{coke}}$$

(4)

Coke formation over the catalyst support is facilitated during the reformation of oxygenated organic substances because of higher molecular weights, more instauration, and aromaticity. Hence, reforming requires specific biomass characteristics such as less molecular weight, saturation, and non-aromaticity to curtail the generation of or eliminate coke byproducts from the catalyst by steam gasification.

CPO (Eq. 5) is a noteworthy alternative to CSR technology, where the fuel interacts with a lower amount of oxidizer (O₂) than that required by stoichiometry for complete combustion.

$${\text{C}}_{n } {\text{H}}_{m} {\text{O}}_{k} + \left( {n - k/2} \right){\text{O}}_{2} \to n{\text{CO}}_{2} + \left( \frac{m}{2} \right){\text{H}}_{2}$$

(5)

The CPO process has several advantages over the CSR method, such as lack of coking, smaller reactor size, and facile CO₂ recovery. Nevertheless, this process has risks associated with explosions, specifically in the region of premixing, while maintaining the stability of the catalyst is also critical. The process has been well-explored for H₂ production from lightweight alcohols compared to complex oxygenated substances (Liguras et al. 2004; Miyazawa et al. 2006; Navarro et al. 2007). The autothermal reforming (ATR) process is a combined CPO and CSR approach providing a thermally self-sustaining faster method, where oxygen, hydrocarbons/oxygenated hydrocarbons, and steam are reactants (Eq. 6).

$$2{\text{C}}_{n } {\text{H}}_{m} {\text{O}}_{k} + \left( {2n - k/2} \right){\text{H}}_{2} {\text{O}} + \left( {n - k/2} \right){\text{O}}_{2} \to 2n{\text{CO}}_{2} + \left( {2n + m - k} \right){\text{H}}_{2}$$

(6)

In contrast, aqueous phase reforming (APR) is an adaptable single-step technology for biomass derivatives (oxygenated alcohols, sugars, glycerol, ethylene glycol) to produce H₂, using reinforced metals and metal amalgams in the form of heterogeneous catalysts (Chheda et al. 2007; Cortright et al. 2002). This process involves the breakage of C–C and C–H and/or O–H bonds to produce adsorbed species over the catalyst support. There are several advantages of the APR process in comparison to CSR, including (1) no need to vaporize the feed, (2) use of relatively high pressures (15–50 bar) and low temperatures (150–270 °C) that promote the WGSR and yield a small amount of CO (< 1000 ppm), (3) use of low temperatures to reduce unwanted disintegration reactions like cracking, and (4) use of high pressures to support the refinement of H₂ (e.g., by membranes or PSA) by sequestering CO₂ (Ghasemzadeh et al. 2016; Huber et al. 2006b). However, use of expensive noble-metal-assisted catalysts at high pressure for an elevated reaction time increases the unit cost of H₂ production from dilute feedstock solutions in this process. The use of strong mineral acids in the aqueous phase leads to the leaching of the active noble metal in the solution, which limits the recyclability of the catalysts and causes corrosion in the reactor. Aqueous discharges of this process also pose serious environmental concerns due to the leaching of metal catalysts (Vaidya and Lopez-Sanchez 2017).

Separation of hydrogen from the mixture of gaseous contaminants requires separate purification/refinement processes. The selection of a suitable H₂ purification method is dependent on the concentrations and downstream impacts of contaminants like N₂ and CO (Santhanam et al. 2017). The biomass-derived H₂ contains several gaseous contaminants including O₂, CO₂, CO, CH₄, and moisture. The foremost H₂ refining technologies can be divided into four categories of (1) physical adsorption (Liu et al. 2009), (2) chemical absorption, (3) cryogenic processes (Lu et al. 2007), and (4) membranes (Bernardo et al. 2010, 2009; Drioli and Giorno 2009). In this context, WGSR plays pivotal role in regulation of the CO/H₂ levels as CO forms CO₂ and H₂ upon interaction with water (Table 1). Industrial H₂ generation by the WGSR is performed in a set of two reactors: (1) a high-temperature WGS reactor at 350–500 °C having a Fe-oxide-mediated catalyst and (2) a low-temperature WGS reactor at 200 °C possessing Cu-functionalized catalyst (Bartholomew and Farrauto 2010). In the first reactor, the concentration of CO decreases to ~ 2–3%, which reduces further to ~ 0.2% in the second reactor under low-temperature. Pressure swing sorption, preferential air oxidation (PROX), and Pd membranes can be used to produce highly purified H₂ (Rostrup-Nielsen 2001). Nanometer-sized Au-amended catalysts are highly effective for oxidation of CO in WGSR (Carrettin et al. 2004; Fu et al. 2005, 2003). Kim et al. developed an alternative to the WGS called the PROX method, where CO is transformed to CO₂ and electrical energy at higher rates by consuming aqueous polyoxometalates (POM) in comparison to the WGSR (Kim et al. 2005a, 2004). The complete reaction involves oxidation of CO and H₂O to produce CO₂ and protons with POM, e.g., H₃PMo₁₂O₄₀, in the presence of Au catalyst. The aqueous and reduced mixture of POMs and H⁺ can be utilized to generate electricity at the anode of a proton exchange membrane (PEM) containing fuel cell (Kim et al. 2005b). In this method, the POM solution is re-oxidized, and the consumption rate of CO (as turnover frequency of 0.75–5 s⁻¹) is greater than that of WGS at room temperature.

2.2.2 Photoreforming of oxygenates

Light-driven oxygenate reformation to produce H₂ and CO₂ upon suitable photocatalytic substances has excellent potential to drive next generation advancements. This technique involves effective reformation of various oxygenates (that can be biomass derived) including alcohols, carboxylic acids, saccharides, polymers in an aqueous medium using semiconductor-supported photocatalysts. Operation at 20–60 °C during the reaction minimizes the scope of degradation and creates high H₂ selectivity (Chen et al. 2010). Kondarides and colleagues demonstrated transformation of typical biomass-based oxygenates such as alcohols, glycerol, saccharides, biopolymers to H₂ on Pt/TiO₂ photocatalysts simulated by sunlight without deactivation (Daskalaki and Kondarides 2009; Kondarides et al. 2008). In some cases, the reformation process is endergonic (ΔG⁰ = 237 kJ mol⁻¹), which signifies the storage of solar energy into the produced products as chemical energy and increases the total heat content of the produced fuel (Cargnello et al. 2011). The photocatalytic reformation of biomass oxygenates broadens the selection routes for direct H₂ generation because it involves inexpensive feedstocks and solar light as the energy source, and the process is proficient, adaptable, and energy efficient. Solid materials consisting of light-absorbing semiconductors such as Au/TiO₂, Pt/Ti-MOF-NH₂, Ni/Au/TiO₂ along with metallic co-catalysts act as photocatalysts during H₂ evolution from aqueous phase. The semiconductors allow formation of charge transporters (electrons and holes) through light absorption, and they transfer to the substance surfaces, where the redox reaction occurs. Therefore, the semiconductor band gap must be matched with the incident photon energy. That is, the electrons are promoted from their valence to conduction band at a shorter wavelength compared to the material absorption as dictated by its band gap. The electron transfers from the conduction band to acceptor, and donor to valence band–hole annihilation occurs only when the energy points are respectively higher and lower than the consistent redox pairs. Co-catalysts are usually integrated on the semiconductor surface and can improve the efficiency by delivering active parts for H₂ evolution by reducing protons or associated substances. At the oxidation half reactions, the semiconductor surface itself catalyzes the transfer of holes to oxygenates or electron from oxygenates to the semiconductor with associated extinction of the photogenerated holes (Fang and Shangguan 2019; Hu et al. 2014; Sun et al. 2019b).

Oxide semiconductors having d⁰ or d¹⁰ metals are considered the most efficient photocatalysts. TiO₂ is generally the ideal semiconductor to design heterogeneous photo-catalysts and anodes (Martha et al. 2015; Naldoni et al. 2019). The conduction band energy is slightly higher than the redox potential of H⁺/H₂ in TiO₂, and band flexibility imparts additional advantages in the H₂ formation reaction in solution (Fig. 4a). Light-triggered electrochemical water splitting can be achieved using platinum black as a cathode and rutile TiO₂ as the photoanode (Fujishima-Honda process). The high photoactivity of TiO₂ in oxidation half-reactions increased its popularity for light-stimulated H₂ production from various oxygenated biomass feedstocks (Schneider et al. 2014). Nevertheless, enhanced surface area with availability of more surface sites, superior crystallinity, and lesser propensity toward electron–hole rearrangement enhances the efficiency of TiO₂ (Moretti et al. 2014). Similarly, WO₃ amended with Pt/Au or SnO₂/self-doped with Sn (II)/co-doped with Ce and Sb can catalyze photoreformation of biomass-derived glycerol (Liu et al. 2015a; Tanaka et al. 2014). ZnO has also demonstrated noteworthy photoactivity to produce H₂ from methanol or formaldehyde (Guo et al. 2015). ZnO nanoroads formed on graphene showed high efficiency for glycerol photoreformation under Xe light, while generation of ZnS in the presence of thiosulfate anions mediated hydrothermal treatment and improved the efficiency (Bao et al. 2015; Lv et al. 2015). NiO, ZnO, CuO, Cu₂O, Fe₂O₃, Co₃O₄, and VO₂ also showed high efficiency in the photoreformation process. Mixed oxides including SrTiO₃ (Peng et al. 2015a), BiVO₄ (Xue and Wang 2015), and Bi₂WO₆ (Panmand et al. 2015) have recently emerged as excellent photocatalysts to produce H₂ by photoreformation.

Fig. 4

Generation of H₂ through photoreformation using semiconducting photocatalysts of TiO₂ (a), CdS (b), Cd-MOF and Cu-MOF (c), and TpDTz COF (d) (Biswal et al. 2019; Naldoni et al. 2019; Song et al. 2017; Zhao et al. 2020)

Metal chalcogenides appeared as efficient photocatalysts because of their small band gaps, which produce visible light activity. The higher electronic energy states in comparison to oxides facilitate the reduction of conduction band photoexcited electrons (Peng et al. 2016). CdS is used to facilitate H₂ generation due to its relatively negative conduction band energy in comparison to proton reduction (Fig. 4b) (Zhao et al. 2020). H₂ generation from glycerol and ascorbic acid using Pt/CdS as a photocatalyst composite has recently been explored (Bastos et al. 2014; Zhou et al. 2015), while surface layer formation on ZnS has proven advantageous for continuous formic acid reforming (Wang et al. 2014). Nevertheless, use of porous metal organic frameworks (MOFs) like UiO-66 could resist Cd²⁺ leaching (Shen et al. 2015; Zhou et al. 2015). Photoreforming of lactic acid has been studied using metal sulfide composites such as MoS₂-CdS, CdS, and NiS-CdS microstructures embedded with Te microspheres (Hu et al. 2015). Solid mixtures of Zn and Cd sulfide composites with Zn/Cd ratio 0.4:0.6 (Cd_xZn_1−xS) have also shown adjustable band gaps for H₂ production by photoreforming (Kozlova et al. 2014; Lopes et al. 2015). Similarly, ZnS-Bi₂S₃ nanoparticle composites accumulated on ZnO nanotubes on graphene have shown efficiency in glycerol photoreforming (Xitao et al. 2014). CuInS₂ has recently been used in the photoreforming of ethanol (Li et al. 2015a), while quaternary chalcogenides like AgInZn₇S₉ are gaining significance due to the possibility of controlled band gaps, facilitating photocatalytic H₂ production (Peng et al. 2015b). Surface amendments of CdSe nanoparticles with suitable pendant groups having attached carboxylate moieties reduces their water-solubility, while addition of Ni or Co to the resulting colloid make the systems highly efficient to produce H₂ from ascorbic acid (Das et al. 2013; Gimbert-Suriñach et al. 2014; Han et al. 2015, 2012; Wang et al. 2015b).

Highly porous crystalline metal and organic frameworks (MOFs and COFs) have been extensively studied in the last decade due to their high semiconducting properties (Fig. 4c, d) (Bavykina et al. 2020; Stegbauer et al. 2014). MOFs offer a potential platform for photocatalytic formation of H₂ due to their structural tunability and regularity, where H₂ productivity is equivalent to the transmission efficiency of photogenerated electrons. Cu-MOF shows more facile reduction to produce H₂ at a photocatalytic rate of 18.96 μmol h⁻¹ due to its more negative conduction band compared to Cd-MOF (Song et al. 2017). A series of bifunctional FeX@Zr₆-Cu (X = Cl⁻, Br⁻, BF₄⁻, and AcO⁻)-functionalized MOFs exhibited high efficiencies to generate H₂ with a turnover number (TON) up to 33,700 h⁻¹ due to the increase in H₂ productivity (Pi et al. 2020). Stegbauer et al. explored photoactive COFs through the synthesis of 1,3,5-tris-(4-formyl-phenyl)triazine (TFPT)-COF for photocatalytic H₂ production (Stegbauer et al. 2014). Thiazolo[5,4-d]thiazole (TpDTz)-COF enabled long-term (70 h) H₂ production with a productivity of 941 μmol h⁻¹ g⁻¹ (Biswal et al. 2019).

Nanocarbon materials and N- or P-doped semiconducting graphene (g-C₃N₄) are considered efficient, metal-free catalysts for the photoreforming process (Latorre-Sánchez et al. 2013; Liu et al. 2015b; Xiang et al. 2015; Ye et al. 2015). Light-harvesting poly(3-hexylthiophene) amended electron mediator g-C₃N₄ composites and Pt co-catalyst have shown high H₂ production efficiency from aqueous ascorbic acid (Zhang et al. 2015b). The catalytic Pt/g-C₃N₄ platform resulted in hydrogen production from wastewater under visible light through consumption of H₂O₂ by sacrificial macrolide antibiotics (Xu et al. 2017). Incorporation of co-catalysts, such as metal oxide nanoparticles and gold nanoparticles, on a semiconductor can enhance the efficiency of photoreforming (Serra et al. 2015; Xu and Xu 2015). Photocatalysis is attributed to various factors, including light-blocking, controlled substrate oxidation over semiconductor surfaces, and electron–hole rearrangement on disproportionate co-catalyst. The rate of light-stimulated H₂ generation on Pd/TiO₂ from aqueous MeOH is controlled by the availability of the perimeter across the interface of co-catalyst and semiconductor surface. According to Bowker et al. charge transfer processes including the adsorbed substrate preferentially occur at the interface of metal-support, which becomes more abundant upon increased metal loading. At an optimum concentration, the particles merge with reduced-perimeter (Al-Mazroai et al. 2007). The elements Cu (Petala et al. 2015), Ni (Chen et al. 2015b; Fujita et al. 2016), Fe (Cao et al. 2015d), and Co (Mahoney et al. 2015) have recently been explored to develop economical co-catalysts. Nanoparticles based on transition metal phosphide cocatalysts have gained attention for H₂ production in light-triggered systems (Yue et al. 2015). Co₂P- and Ni₂P-loaded CdS nanorods are highly efficient at producing H₂ from mandelic and lactic acids (Cao et al. 2015a, 2014, 2015c). Co_0.85Se and Ni₂B supported on Se were recently found to be promising co-catalysts for inexpensive H₂ evolution by a photoreforming approach (Cao et al. 2015b; Wang et al. 2015c). RGO/CdS and RGO/TiO₂ composites have also been found to be effective for H₂ production by photocatalysis from lactic acid/water or water/ethanol solutions (Babu et al. 2015; Nagaraju et al. 2015).

Biomass feedstocks have enormous potential to provide useful oxygenates for strategic conversion of bio-based wastes to H₂ energy through biomass valorization (Taipabu et al. 2022). Lignocellulosic materials are most abundant as plant mass, which can be obtained from forest products (wood or shrubs), agricultural offshoots (wheat straw, rice husks, or corn hobs), energy-based plants (water hyacinth or sorghum), and various urban and industrial waste materials (Dahmen et al. 2019). Crops supplying starch/sugar and vegetable oil are also advantageous for generating H₂ due to the presence of hydrogen-rich polysaccharides and long chain fatty acids (LCFAs), respectively (Salama et al. 2019; Sołowski et al. 2018). Biomass-derivatives are composed of various oxygen-enriched functional groups including acetals, carbonyls (aldehydes and ketones), carboxylic groups, ethers, and hydroxyl (alcohols, phenols, and polyols) (Wu et al. 2016). Various biomass-derived potential substances like alcohols, aldehydes, polyols, ketones, saccharides, and carboxylic acids play a key role in producing H₂ by photoreforming of biomass feedstocks (Table 2). Thus, various renewable waste-derived feedstocks can produce H₂ at high rates, opening an opportunity for waste valorization.

Table 2 Photocatalytic H₂ production from various biomass-derived oxygenates via photoreformation

3 Hydrogenogenic fermentation for evolution of hydrogen

Fermentative biohydrogen is considered a potential contender for many conventional energy sources due to its environmentally friendly nature, which lessens fossil fuel consumption and pollution control measures (Ahmed et al. 2021). There are two fermentative processes that can yield H₂: dark fermentation (anaerobic hydrogenogenic acidogenic fermentation) and photo-fermentation. These technologies have the potential to become cost competitive as they can use low-value waste biomasses as feedstock (Jiao 2021). The dark fermentation pathway is light-independent and performs heterotrophic fermentation using facultative or obligate anaerobic microbes (Basak et al. 2020; Chang et al. 2018). In this process, hydrogenogenic microorganisms anaerobically utilize solid/soluble organic matter from low-cost organic wastes and municipal/industrial wastewater producing gas mixture (i.e., H₂, CO₂) and various byproducts including volatile fatty acids (VFAs), ethanol, acetone, propanol, and butanol (Eqs. 7–12) (Wang and Yin 2021; Zhou et al. 2018). The yields of H₂ vary depending on the fatty acid pathway and the type of sugar present in the waste biomasses (e.g., glucose, sucrose, xylose) (Sarangi and Nanda 2020).

$${\text{C}}_{5} {\text{H}}_{10} {\text{O}}_{5} + 3.33{\text{H}}_{2} {\text{O}} \to 1.67{\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} + 1.67{\text{CO}}_{2} + 3.33{\text{H}}_{2}$$

(7)

$${\text{C}}_{5} {\text{H}}_{10} {\text{O}}_{5} + 1.67{\text{H}}_{2} {\text{O}} \to 0.83{\text{C}}_{4} {\text{H}}_{8} {\text{O}}_{2} + 1.67{\text{CO}}_{2} + 1.67{\text{H}}_{2}$$

(8)

$${\text{C}}_{6} {\text{H}}_{12} {\text{O}}_{6} + 2{\text{H}}_{2} {\text{O}} \to 2{\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} + 2{\text{CO}}_{2} + 4{\text{H}}_{2}$$

(9)

$${\text{C}}_{6} {\text{H}}_{12} {\text{O}}_{6} \to {\text{C}}_{4} {\text{H}}_{8} {\text{O}}_{2} + 2{\text{CO}}_{2} + 2{\text{H}}_{2}$$

(10)

$${\text{C}}_{6} {\text{H}}_{12} {\text{O}}_{6} + {\text{H}}_{2} {\text{O}} \to {\text{C}}_{2} {\text{H}}_{6} {\text{O}} + {\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} + 2{\text{CO}}_{{2}} + 2{\text{H}}_{2}$$

(11)

$${\text{C}}_{12} {\text{H}}_{22} {\text{O}}_{11} + {\text{H}}_{2} {\text{O}} \to 2{\text{C}}_{4} {\text{H}}_{8} {\text{O}}_{2} + 4{\text{CO}}_{2} + 4{\text{H}}_{2}$$

(12)

Substrates such as polysaccharidic waste biomasses, which are rich in glucose, sucrose, starch, cellulose, and lignocellulose, are considered suitable for dark fermentation due to their conversion efficiency to hydrogen and low cost to enhance hydrogen productivity by minimizing the process cost (Chen et al. 2006; Nasr et al. 2015; Reaume 2009). However, the complex structure of waste biomasses requires an efficient pretreatment process before fermentation to create microbial accessibility to fermentable sugars (Saha et al. 2016). Dark fermentation involves converting complex organic matter into simpler products through hydrolysis and acidogenesis by producing short- and medium-chain fatty acids (SCFAs-MCFAs) and gaseous hydrogen (Fig. 5) (Chang et al. 2018). Several microorganisms harbor various hydrogenase enzymes (such as [FeFe]-hydrogenase, [NiFe]-hydrogenase, and [NiFeSe]-hydrogenase), which stimulate production and recycling of hydrogen under an anaerobic environment (de Sá et al. 2011). Among these enzymes, only [FeFe]-hydrogenase catalyzes hydrogen production, while the other two utilize produced hydrogen and are often found in hydrogen-consuming microorganisms. Comparatively higher expression of [FeFe]-hydrogenase (around 100-fold) than [NiFe]-hydrogenase in strict and facultative bacteria allows production of enough hydrogen during dark fermentation (Sołowski et al. 2018). Clostridia, Escherichia coli, Enterobacter, Citrobacter, Alcaligenes, and Bacillus strains of Gram-positive or Gram-negative bacteria are dominant in fermenters or digesters under strict or facultative anaerobic environments (Ferraren-De Cagalitan and Abundo 2021). Metabolic pathways and substrate diversity studies revealed the dominance of Clostridium species in dark fermentation utilizing a wide variety of substrates for the production of H₂ (3.94 mol H₂ mol⁻¹ hexose − 5.42 mol H₂ mol⁻¹ carboxymethylcellulose) through acetate, butyrate, and propionate pathways (Wang and Yin 2021).

Fig. 5

Metabolic pathways involved in dark fermentation of complex organic substrates such as microalgae (a), waste activated sludge (b), food waste (c), and fat, oil, grease (d) to produce hydrogen, short- and medium-chain fatty acids, ethanol, and butanol (Akhlaghi et al. 2019; Elsamadony et al. 2021; Saha et al. 2019; Wan et al. 2016; Wang and Yin 2021; Yun et al. 2016)

Production of hydrogen reached the maximum of 1.5 mol H₂ mol⁻¹ of fructose and 1.3 mol H₂ mol⁻¹ of sucrose when Firmicutes dominant anaerobic consortium from brewery wastewater was used as inoculum (Pachiega et al. 2019). Here, the yield of hydrogen depends on specific parameters such as substrate type, loading rate, system pH and temperature, microbial adaptability to specific substrate, inoculum-to-substrate ratio, and concentrations of various additives in the fermentation broth (Akhlaghi et al. 2019; Li et al. 2021; Ren et al. 2022; Saha et al. 2020). An increase of microalgal loading as fermentation substrate to 40 g dry cell weight L⁻¹ in mesophilic dark fermenters achieved a maximum hydrogen yield of 36 mL H₂ g⁻¹ through the acetate pathway, which decreased upon further loading (Fig. 5a) (Yun et al. 2016). Around a 13% decrease in the hydrogen yield occurred when sucrose was the primary substrate compared to fructose (1.5 mol H₂ mol⁻¹) (Pachiega et al. 2019). Clostridium acetobutylicum produced 63 ml H₂ g⁻¹ starch in solid state (total solid: 20%) dark fermentation of ground wheat (Ozmihci 2017). Use of waste activated sludge in thermophilic (55 ℃) reactors produced 317% higher hydrogen yield at pH 10 over mesophilic (37 ℃) reactors (Fig. 5b). However, VFA production (especially, acetate and propionate) was much higher in the mesophilic reactors (Wan et al. 2016). Production of hydrogen (3.8 mol H₂ mol⁻¹ of sugar) reached its theoretical yield (4 mol H₂ mol⁻¹ of glucose) during hyperthermophilic fermentation of fruit and vegetable waste using the halophilic bacterium Thermotoga maritima (Saidi et al. 2018). Acidogenic dark fermentation of food waste (FW) at an inoculum-to-substrate ratio of 50:50 produced 160 L H₂ kg⁻¹ TOC_FW during mesophilic operation at pH 5.5 (Fig. 5c) (Akhlaghi et al. 2019). Incorporation of ryegrass into sewage sludge fermentation increased the final hydrogen yield (60 mL g⁻¹ VS) by five times at a ratio of 30:70 compared to the yield in sludge alone (Yang and Wang 2017). Thus, the hydrogen productivity varies depending on substrate variations and fermentation conditions.

Acclimatization of the microbial inoculum to enrich the hydrogenogenic bacterial population improved the conversion of specific substrates to hydrogen. Hydrogen yield was boosted to 19.5 L H₂ L⁻¹ in thin stillage using acclimatized anaerobic digester sludge due to enrichment of acidogenic Clostridium acetobutyricum, Klebsiella pneumonia, Clostridium butyricum, and Clostridium pasteurianum, which improved the specific hydrogen production rate (3.5 times higher than with unacclimatized sludge) (Nasr et al. 2011). Bioaugmentation with Clostridium thermocellum improved hydrogen yield by 96.80% in thermophilic fermentation of paper sludge through increase of holocellulose degradation rate (32.95%) (An et al. 2018). The presence of active acidogenic bacteria belonging to the phyla Firmicutes and Bacteroidetes in the acclimatized inoculum improved the hydrogen yield by 48% compared to its unacclimatized counterpart (Chang et al. 2018). Cellobiose-acclimatized sewage sludge enriched the population of genus Clostridium as a main hydrogen producer and produced 2.08 mmol H₂ g⁻¹ filter paper (Ho et al. 2012). Enrichment of iron-dependent hydrogenases expressing Anaerofilum sp. in the bioreactors enhanced the hydrogen yield to 13.64 mmol in five months of operation (Venkata Mohan et al. 2011). A combination of various additives such as zero-valent iron, nano zero-valent iron, Ni⁰ nano particles, and biochar stimulated the acidogenesis process through the activation of various hydrogenases to achieve hydrogen yields near the theoretical potential of the substrate (Mohd Jamaludin et al. 2023; Ren et al. 2022). Metallic additives improved enzymatic activities through oxidation/reduction to promote hydrogen production (Saha et al. 2020). Upon addition to fermentation reactors, zero-valent metals oxidize in the aqueous phase and promote the synthesis and activity of major enzymes such as hydrogenase and ferredoxin (Taherdanak et al. 2015). Shifting of a microbial population toward Clostridium occurred upon addition of Fe²⁺ ions in grass-fermenting reactors, which enhanced the hydrogen yield (72.8 mL g⁻¹ dry grass) by 49.6% (Yang and Wang 2018a). Fe⁰ nanoparticles at a concentration of 400 mg L⁻¹ increase the hydrogen production rate and yield by 128% and 73%, respectively, by shortening the lag period and facilitating substrate hydrolysis and utilization through the enrichment of Clostridium sp. (Yang and Wang 2018b). Fe⁰ nanoparticle supplementation in dark fermentation improved the hydrogen yield to 20 mL H₂ g⁻¹ VS of microalgae (6.5 times higher than the control) by enriching the populations of Clostridium and Terrisporobacter sp. (Yin and Wang 2019).

Addition of magnetite at a concentration of 100 mg L⁻¹ induced the growth of acidogenic bacteria such as Sporolactobacillus, Clostridium, and Coprothermobacter, which increased the hydrogen production by 46% (Gökçek et al. 2023). Magnetite nanoparticles embedded in granular activated carbon (GAC) improved the hydrogen productivity rate in dark fermentation by 63.99% compared to non-magnetite GAC (Mohd Jamaludin et al. 2023). The high specific surface area and porosity of biochar act as a support matrix to microbial attachment, growth, and cellular vitality and improve syntrophic interspecies interactions, further expediting the holistic conversion of substrate to fermentative products. Use of granular activated carbon as a support matrix for biofilm development enhanced the hydrogen yield of 1.77 mol H₂ mol⁻¹ substrate_consumed during thermophilic dark fermentation (Jamali et al. 2016). Biofilm formation on the surface of biochar during co-culture of Enterobacter aerogenes and E. coli in fermentation of municipal solid waste maximized the hydrogen yield to 96.63 ml H₂ g⁻¹ carbohydrate_initial by accelerating COD removal (53%) and shortening the lag phase from 12.5 h to 8.1 h (Sharma and Melkania 2017). Supplementation of sugarcane bagasse-derived biochar improved hydrogen production (84.58 mL) by enhancing the hydrogen productivity (> 74%) of Ethanoligenens harbinense (Li et al. 2021). The synergistic effect of combining biochar and Fe⁰ nanoparticles at concentrations of 600 mg L⁻¹ and 400 mg L⁻¹, respectively, enhanced the hydrogen yield to 50.6 mL g⁻¹ dry grass and shortened the lag phase by accelerating enriched Clostridium-mediated hydrolysis and fermentation (Yang and Wang 2019). Addition of zero-valent iron-activated carbon to wastewater sludge fermentative reactors improved the hydrogen yield (1.33 mol H₂ mol⁻¹ glucose) by 50% through enrichment of the Clostridium spp. population (Zhang et al. 2015a). The synergistic impact of biochar (621 mg L⁻¹) and Ni⁰ nanoparticle (17.2 mg L⁻¹) in dark fermentation increased the hydrogen production rate (558 mL h⁻¹) and hydrogen yield (237 mL g⁻¹) by decreasing the lag phase to 4.82 h (Sun et al. 2019a). Thus, a combination of metallic additives as co-factor for hydrogenases and biochar improved the microbial enzymatic activities through induction of electron transport among syntrophic partners and enriched the hydrogenogenic bacterial population in acidogenic fermenters for improved hydrogen production from organic biomass.

Utilization of lipidic waste such as FOG (fat, oil, grease) in dark fermentation produced a large amount of hydrogen from saturated and unsaturated LCFAs along with C4–C7 carboxylates through carboxylic chain elongation of SCFAs (Fig. 5d) (Saha et al. 2019). β-Oxidation of saturated (e.g., stearic acid, palmitic acid, myristic acid) and unsaturated (linoleic acid, linoelaidic acid, oleic acid) LCFAs during anaerobic fermentation generated C2 carboxylate and H₂ as an electron acceptor (Eq. 13) (Cavaleiro et al. 2016; Elsamadony et al. 2021). These C2 carboxylate and H₂ play a major role in the thermodynamic feasibility of the β-oxidation pathway, which demands a low hydrogen partial pressure of 1 Pa and acetate concentration of 8 or 9 mmol l⁻¹ in the reactors (Saha et al. 2020). Failing to retain these conditions led to accumulation of saturated LCFA (palmitic acid, ∼90%) in the fermenters as a major intermediate of the degradation pathways of the longer chain unsaturated LCFAs (Saha et al. 2019). Palmitic acid and myristic acid emerged as the main intermediates of the monounsaturated LCFAs degradation in the absence of methanogenic activity, leading to their accumulation in the reactors (Cavaleiro et al. 2016). Syntrophic coupling among acidogenic LCFA-degrading bacteria with acetate-utilizing and hydrogenotrophic microorganisms help sustained an optimum environment to facilitate continuous β-oxidation (Cavalcante et al. 2017). The occurrence of reverse metabolic traits is feasible in the presence of energy-rich, reduced compounds such as ethanol and lactate to provide energy and reduce equivalents and acetyl-CoA during the anaerobic reactor microbiome-mediated β-oxidation (Fig. 5) (Spirito et al. 2014). Various MCFAs such as C4–C6 carboxylates are the products of secondary anaerobic fermentation of SCFAs after acidogenic fermentation, where ethanol acts as an electron donor for carboxylic acid chain elongation/ reverse β-oxidation (Eqs. 14 and 15) (Angenent et al. 2016; Cavalcante et al. 2017).

$$n{\text{CH}}_{3} ({\text{CH}}_{2} )_{n} {\text{COOH}} + n{\text{H}}_{2} {\text{O}} \to {\text{CH}}_{3} ({\text{CH}}_{2} )_{n - 2} {\text{COOH}} + n{\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} + n{\text{H}}_{2}$$

(13)

$$5{\text{C}}_{2} {\text{H}}_{6} {\text{O}} + 3{\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} \to 4{\text{C}}_{4} {\text{H}}_{8} {\text{O}}_{2} + 3{\text{H}}_{2} {\text{O}} + 2{\text{H}}_{2}$$

(14)

$$12{\text{C}}_{2} {\text{H}}_{6} {\text{O}} + 3{\text{C}}_{2} {\text{H}}_{4} {\text{O}}_{2} \to 5{\text{C}}_{6} {\text{H}}_{12} {\text{O}}_{2} + 8{\text{H}}_{2} {\text{O}} + 4{\text{H}}_{2}$$

(15)

Syntrophic cooperation among ruminal mixed microbiome and cellulose-converting Clostridium kluyveri produced valeric and caproic acids in the presence of ethanol (Weimer et al. 2015). Anaerobic fermentation of municipal solid waste in a two-stage system consisting of an acidification and a chain elongation reactors achieved caproate production of 12.6 g l⁻¹, the highest among MCFAs (Grootscholten et al. 2014). A MCFA yield of 67% was obtained during anaerobic fermentation of wine lees (settled yeast cells and ethanol) using acclimatized inoculum along with hydrogen (45% of produced biogas), which primarily consists of caprylate and caproate at 36% each (Kucek et al. 2016b). Production of hydrogen (3000 ppm) occurred in an anaerobic up-flow bioreactor producing caprylate (0.33 g L⁻¹ h⁻¹) as a major MCFA when ethanol and acetate were used as feedstocks (Kucek et al. 2016a). The thermodynamic feasibility of MCFA production in dark fermentation is highly supportive due to presence of a secondary by-product (ethanol), and it promoted the growth and syntrophic relationship among Clostridium kluyveri and ethanol-producing bacteria to facilitate the production of 10–20% more hydrogen with MCFA (especially, caproate) (Ding et al. 2010). Thus, conversion of organic biomass in dark fermentation could lead to the production of various value-added chemicals and increase hydrogen yield through the induction of microbial metabolic syntrophy.

4 Techno-economic assessment of hydrogen technologies

The sustainable production of hydrogen is of prime importance for its commercialization as a fuel. A competitive and reasonable price of hydrogen fuel can be achieved through increased rate of production and reliability of processing (Godvin Sharmila et al. 2022). There are several key drivers that mainly influence the green hydrogen economy, i.e., (a) cost of substrate/feedstock and pretreatment, (b) cost of the hydrogen production system, (c) cost of downstream processing for purification of hydrogen, (d) cost of transportation and storage, and (e) cost of distribution (Lane et al. 2021; Prabakar et al. 2018). The physicochemical processes for production of hydrogen are highly efficient with respect to productivity and purity; however, the economic viability of these processes is low due to their high energy requirements (Godvin Sharmila et al. 2022). Various technologies have been developed and tested for commercial hydrogen production. The chemical processes with broad industrial applications include water splitting by electrolysis, gasification of coal, and steam reforming, while biological processes include photobiological hydrogen production and dark fermentation (Brar et al. 2022; Mahmod et al. 2021; Qyyum et al. 2022). Among chemical pathways, electrolysis is widely employed, whereas steam reforming of methane is the most cost-effective method for hydrogen production, with a cost of US$ 7 GJ⁻¹ (Kalamaras and Efstathiou 2013). The combination of electrolysis with a geothermal process for energy could cost around US$ 1.09 kg⁻¹ (Yilmaz et al. 2019). The pyrolysis and gasification of biomass could generate hydrogen fuel at a cost of around US$ 8.9–15.5 GJ⁻¹ and US$ 10–14 GJ⁻¹, respectively, which varies with the cost of raw materials (Balat and Balat 2009).

Biological processes are comparatively expensive because of applicability issues. The major bottlenecks include the high operational costs of bioprocessing, process instability, and low H₂ yields (Das and Basak 2021). For instance, hydrogen production via a photofermentation process costs around US$ 502.10 GJ⁻¹, which amounted to 90% of the total cost (Bhatia et al. 2021). One of the major hurdles in commercialization of hydrogen production through dark fermentation is the excessive cost of substrates (Yukesh Kannah et al. 2021). The cost of dark fermentative hydrogen production from glucose increased 5–10 times higher than that of photofermentation (US$ 2 kg⁻¹ H₂) and indirect biophotolysis (US$ 1.42 kg⁻¹ H₂) due to high cost of substrate, which consumed 88% of the production cost (Ahmed et al. 2021; Godvin Sharmila et al. 2022). The cost for bioprocessing can be significantly reduced if waste biomasses from industrial or agricultural sectors are utilized, since complex carbohydrates are available in massive quantities at comparatively cheaper prices. The waste-based hydrogen economy can overcome the largest electricity/energy barrier and encourage sustainable fuel production with greater energy security. The biomass substrate cost is highly subject to principal, operational, and maintenance costs, for levelled energy cost (LCOE) is less responsive to biohydrogen, which costs around US$ 7 kg⁻¹ (Ahmed et al. 2021; Lee et al. 2008). Palm oil milling effluent is a by-product of the process and must be treated to meet environmental regulations. It can be utilized as a cost-effective feedstock for hydrogen production considering its low cost (Nurul-Adela et al. 2016). Hydrogen production through dark fermentation of food waste is inexpensive, US$ 0.814 kWh⁻¹, which provided an investment return of 29.8% within 7.2 years of the payback period (Cudjoe et al. 2022). The two-stage anaerobic technology is thought to be cost effective as its payback time is substantially shorter than its expected life (Mahmod et al. 2021).

Industrial effluents from dairy; slaughterhouse; and food and starch processing industries; and oil refineries also can be an effective feedstock for hydrogen production as it contains ample organic matter. A techno-economic study of hydrogen production using black liquor as a substrate via dark fermentation showed H₂ productivity of 12,450 m³ year⁻¹ and annual revenue of US$ 35,481, with an initial investment payback period of 5.92 years (Tawfik et al. 2021b). Another study of H₂ and CH₄ generation from biscuit wastewater treatment via anaerobic digestion indicated payback periods of 5.7 and 7.1 years, respectively, although the initial investment for H₂ was 22% higher than that of CH₄ production due to more sophisticated processing (Tawfik et al. 2021a). These technologies achieved green energy production and industrialization with pollution prevention to address sustainable development goals (SDGs). The investment profits and payback period for hydrogenation of petrochemical industry wastewater under psychrophilic conditions were US$ 24,295 year⁻¹ and 7.13 years, respectively (Elreedy et al. 2019).

The world’s biggest economies including USA, China, Japan, and India have encouraged investment in the development of hydrogen fuel (Lee et al. 2011). Among these four nations, China created the largest market for biohydrogen and has expected to reach the industry’s output value to US$ 157.44 billion in 2025 (Koty 2022; Nakano 2022). Japan, USA, and India have started investment of around US$ 3.4, 8, and 25 billion, respectively to bring down the cost of green hydrogen to US$ 1 kg⁻¹ by 2030 (Chaudhary 2022; DOE 2022; Nakano 2021). Such a large investment by these big economies will ensure state-of-the-art developments in biohydrogen infrastructures and technologies that will eventually produce more benefits than investment with better cost feasibility. The hydrogen economy significantly prevents pollution. It has been estimated that utilization of hydrogen fuel instead of gasoline for more than 15 years can save ~ 29.9 million tonnes of CO₂-eq (Wulf and Kaltschmitt 2012). Moreover, the efficiency of diesel and hydrogen engines appears to be comparable (Stichnothe and Azapagic 2009). Therefore, employing hydrogen as a substitute fuel for conventional fossil fuels might significantly reduce GHG emissions. The future of hydrogen industry is bright as this green economy can play a key role in achieving SDGs by supporting eco-friendly technologies, instigating sustainable production methods, and conserving natural resources (Qyyum et al. 2022). The hydrogen economy fosters societal equity, promotes sustainable growth, and reduces environmental risk and resource requirements. Regulations supporting low carbon emissions or minimized CO₂ emission to overcome economic challenges should be employed (Prabakar et al. 2018). Development of novel technologies such as single-pot processes and integrated bioprocessing are necessary to reduce the costs of processing.

5 Concluding remarks and outlook

The feasibility of hydrogen production from various organic waste biomasses has been assessed through numerous thermochemical, photocatalytic, and biological processes over the past decades. Conversion of oxygenated biomass derivatives to green hydrogen through selective light-triggered reformation (photoreformation) at ambient conditions in the presence of active metallic photocatalysts has emerged as a sustainable method preferable to thermocatalytic degradation or gasification. The use of graphene-based carbonaceous photocatalysts and earth-abundant low-cost metals, such as Cu and Ni, could be advantageous to resist the photocorrosion of metal sulfides. Alternatively, dark fermentation has appeared as a potential contender due to its ability to produce multiple value-added chemicals along with hydrogen. Optimization of operational parameters such as substrate type, loading rate, system pH, temperature, microbial acclimatization to specific substrate, inoculum-to-substrate ratio, and concentrations of various additives (especially, Fe and Ni ions and biochar) in dark fermentation could improve the holistic conversion of organic wastes and simultaneous maximization of product yields. Upgrading produced hydrogen and downstream processing of value-added chemicals in the fermentation broth require substantial research to facilitate maximum energy recovery from organic wastes to ensure long-term sustainability. Thus, fostering eco-friendly technologies such as photoreformation and dark fermentation for hydrogen production from waste bioresources and promoting green energy utilization for prevention of environmental pollution could brighten the future of hydrogen industry.