Syngas conversion to biofuels and biochemicals: a review of process engineering and mechanisms

Abstract

Syngas is generated by thermochemical conversion of a wide range of organic wastes, or it is directly produced as an industry off-gas. The syngas components are converted into alcohols and other high-value-added bioproducts by acetogenic bacteria primarily via the Wood–Ljungdahl pathway. The feasibility of the syngas fermentation process to produce ethanol and other biochemicals is affected by many factors, such as microorganisms, fermentation strategies, gas–liquid mass transfer, and reactor types and design. This study offers new perspectives on the fermentation of syngas by focusing on all these factors to achieve commercialization of these value-added products. Moreover, it includes concepts regarding industrial applications by focusing on metabolic engineering and life cycle assessment for evaluating alternative sustainability dimensions and optimizing the production of ethanol and other biochemicals. This review paper lays a foundation for comparative studies that can be carried out to improve the technological, environmental, and socioeconomic aspects of bioethanol production.

---

1. Introduction

Syngas is composed of different gases, such as carbon monoxide (CO), carbon dioxide (CO₂) and hydrogen (H₂) and is produced by thermal gasification of a wide range of organic wastes. The gasification process of these various types of feedstocks is executed in a gasifier at high pressure and temperature in the presence of gasifying agents, including oxygen or steam.¹ According to one study, the estimated production of syngas is approximately 598 million tons per year.² Apart from the thermal gasification process, syngas is generated as a waste gas from the chemical and steel industry, and its composition is influenced by a number of parameters, such as the type of gasifier and operational conditions.³ Syngas fermentation is meaningful for carbon emission reduction as well biofuel production.⁴

Since the demand for energy worldwide is predicted to increase by 47% over the next 30 years and the consumption of liquid fuel is predicted to climb by 64% relative to 2020, hence, in order to reduce the demand for carbon-based fuels, increasing the production of biofuels might be a more sustainable approach.^5,6

The market for biofuels will reach USD 201.21 billion by 2030, expanding at a compound annual growth rate (CAGR) of 8.3% between 2021 and 2030. Utilizing cutting-edge techniques to achieve better titers, rates, and yields is essential for the environmentally responsible, cost-effective, and sustainable production of biofuels.^7,8 To efficiently ferment syngas, various studies have reported useful acetogens that have been isolated and are capable of utilizing syngas as the sole carbon source for their growth.

Acetogens are a phylogenetically and physiologically varied group of anaerobic bacteria, with over 100 species dispersed over 23 bacterial genera. The common characteristic of all acetogens is that they utilize the Wood–Ljungdahl pathway (WLP) to fix C1 gases. During gas fermentation, CO or CO₂ is employed as a carbon source, while CO or H₂ offers reducing equivalents. The condensation of the methyl group, a carbonyl group by CO or CO₂, and coenzyme A at the expense of one adenosine triphosphate (ATP) molecule occurs in the WLP over the course of many steps to produce one molecule of acetyl-CoA, which eventually leads to the generation of biomass, and other metabolites are produced from the generated acetyl-CoA.⁹

Among these, several species of Clostridium have been widely reported for syngas fermentation, such as Clostridium carboxidivorans, which is capable of producing hexanol, butanol, and ethanol from syngas using the WLP. Moreover, Clostridium species have carbon monoxide dehydrogenase (CODH), which participates in catabolic CO oxidation. Since acetogens utilize H₂/CO₂ as the main substrates for growth, therefore, CO toxicity remains a major issue.¹⁰ Carbon monoxide binds metal clusters in enzymes and prevents the binding of the enzymes' natural substrates. Therefore, to avoid fully depleting the ferredoxin pool or inhibiting important enzymes, carboxydotrophic organisms should effectively limit the toxicity of carbon monoxide. CODH catalyzes the oxidation of CO to CO₂ upon reduction of ferredoxin at a high specific activity and detoxifies carbon monoxide. This is why most Clostridium species are used as biocatalysts for syngas fermentation; nonetheless, some studies have also identified other species that can potentially utilize C1 gases and convert them directly into acetate and butyrate. The most common products of syngas fermentation are ethanol, acetate, n-butanol, 2,3-butanediol and hexanol, which have vast industrial applications, including their usage as solvents and fuels. They are also used in the textile industry for dyeing, rubber production and polyurethane production; the detailed applications of the major syngas fermentation products are presented in Table 1.¹¹

Table 1 Major syngas fermentation products and applications

Name	Applications	Ref.
Acetate (C2H4O2)	Textile industry (dyeing), rubber production, concrete sealant, food, and flavoring	12
Ethanol (C2H6O)	Precursor of acetic acids, diethyl ether, and ethyl halides	13
n-Butanol (C4H10O)	Precursor of butyl esters, acetates, pharmaceuticals, polymers, plastics, and herbicides	14 and 15
2,3-Butanediol (C4H10O)	Polyurethane production, solvent, glycerin substitute, and sanitary products	16
Hexanol (C6H14O)	Solvent, shellac, resins, and hormones	17
Single-cell protein	Used as animal feed	LanzaTech Inc., (http://www.lanzatech.com)

---

The production of biofuels and biochemicals from syngas has gained increasing attention over the last few years, but understanding and efficient execution of the process is still far from being achieved. The latest reviews on biofuel production studied syngas fermentation; however, process engineering has seldom been reported. Moreover, the feasibility of the syngas fermentation process to produce ethanol is influenced by many factors, including microorganisms, fermentation strategies, gas–liquid mass transfer, and reactor types and design. These limitations are an impediment to syngas fermentation technology commercialization. This review offers new perspectives on the fermentation of syngas by focusing on all these factors to achieve commercialization of ethanol and other biochemicals. This paper sheds light on metabolic engineering, which can be applied to improve process engineering and eventually lead to improvements in reaction conditions, thus providing smart product recovery strategies. In addition, this paper also discusses scale-up strategies, commercialization, and future research directions while providing a technological road map.

2. Microorganisms and metabolic pathways

Biofuel and biochemical production by syngas fermentation has several benefits compared to metal catalytic conversion, for instance, higher biocatalyst specificity, lower cost of energy, independence of a fixed H₂ : CO ratio, and greater resistance to catalyst poisoning.¹⁸ Research over the past several years has reported anaerobes that utilize CO and H₂ for acetate, butyrate, butanol, ethanol, and formate production, as shown in Table 2. Most syngas-fermenting microorganisms develop autotrophically by utilizing the components of syngas as the sole source of carbon and energy. This involves the reductive acetyl-CoA pathway, in which carbon from CO and CO₂ is assimilated into biomass and CO and H₂ are utilized as energy sources.^3,19 This pathway is adopted by several microbes including Moorella thermoacetica that utilizes CO₂.²⁰ Several other anaerobic microbes that adopt the Wood–Ljungdahl pathway (Fig. 1) to produce acetic acid as a major product have been displayed in Table 2.

Table 2 Microbial species, culture strategies, growth conditions, doubling times and substrates for the production of ethanol and other value added bioproducts

	Species	Temp. (°C)	pH	Doubling time (h)	Substrates	Products	Ref.
Pure cultures	Acetobacterium woodii	30	6.8	13	CO/H2/CO2	Acetate	27
	Butyribacterium methylotrophicum	37	5.5–6.0	12–20	CO	Acetate, butyrate, butanol & ethanol	28
	C. autoethanogenum	20–44	4.5–6.5	NA	CO/H2/CO2	Acetate, ethanol, and 2,3-butanediol	16
	C. carboxidivorans	24–42	4.4–7.6	6.25	CO/CO2/H2	Acetate, ethanol, butyrate, butanol, caproate, and hexanol	29
	C. drakei	18–42	4.6–7.8	8.3	CO/CO2/H2	Acetate, ethanol, butyrate, and butanol	30
	C. ljungdahli	30–40	4.0–6.0	3.8	CO/H2/CO2/N2	Acetate, ethanol, 2,3-butanediol, and formic acid	31
	C. ragsdalei	25–40	5.0–7.5	5.7	CO/H2/CO2	Acetate, ethanol, and 2,3-butanediol	17
	C. aceticum	30–37	7.7	NA	CO/CO2	Acetate	32
	Eubacterium limosum	38–39	7.0–7.2	7.0–18.2	CO2/H2	Acetate and butyrate	27
	M. thermoacetica	55	7	18	CO	Acetate	33
	Sporomusa ovata	15–45	5.0–8.1	13	CO2/H2	Acetate and ethanol	34
Co-cultures	Alkalibaculum bacchi and C. propionicum	37	6.0–8.0	NA	Syngas	Ethanol, n-propanol, and n-butanol	14
	C. ljungdahlii and C. kluyveri	35	5.7–6.4	NA	Syngas	Acetate, butyrate, caproate, ethanol, butanol, hexanol, 2,3-butanediol, and octanol	35
	C. kluyveri and C. autoethanogenum	37	5.5–6.5	NA	CO/CO2/H2	Acetate, butyrate, caproate, ethanol, butanol, and hexanol	36
Multi-stage culture	C. ljungdahlii (reactor 1)	37	5.9	NA	CO/H2/CO2/N2	Ethanol and acetate	37
	Aspergillus oryzae (reactor 2)	35	6.5	NA	Fermentation effluent	Malic acid	7
	C. autoethanogenum (stage 1)	30	5.75	NA	CO/H2/CO2/N2	Acetate, ethanol, and 2,3-butanediol
	Mixed microbial consortia (MMC) (stage 2)	30	NA	NA	Fermentation effluent	PHA
Genetically engineered	C. coskatii [p83_tcb]	37		NA		3-Hydroxybutyrate (3-HB)	38
	C. autoethanogenum PHB	37	5	NA		Poly-3-hydroxybutyrate (PHB)	39

---

Fig. 1 Wood–Ljungdahl pathway, “Adapted from ref. 21”.

The initial discovery of the metabolic acetogenesis process, or the reduction of two molecules of CO₂ with hydrogen into one molecule of acetate, in a microbial community from sewage was made in 1932.²² In 1936, Clostridium aceticum was successfully isolated from soil for the first time. The following stoichiometry was used to determine how much acetate would be produced from CO₂ and H₂.²³

2CO₂ + 4H₂ → CH₃COOH + 2H₂O

Later, Moorella thermoacetica was isolated which converted one molecule of glucose to nearly three molecules of acetate.²⁴

C₆H₁₂O₆ → 3CH₃COOH

In the 1980s, further research was carried out to determine how CO₂ turned into acetate. Thus, the WLP gave insights into the metabolism of acetic acid production by Moorella thermoacetica while utilizing CO₂.^20,25 The ideal temperature for growth is between 55 and 60 °C. It can grow on a variety of other substrates, including methanol and ethanol. Sugars and methoxylated aromatic compounds are other substrates employed in addition to CO, CO₂, and hydrogen. M. thermoacetica is not dependent on the Na⁺ ion.²⁶ CO dehydrogenase/acetyl-CoA synthase (CODH/ACS), which forms a C–C bond between CO and a methyl group, is the key enzyme in this route. The active site of this core enzyme contains cobalt and nickel. Acetogenic bacteria, which can convert two moles of CO₂ with four moles of hydrogen into acetate, were the primary organisms used to study WLPs. With −94 kJ mole⁻¹ acetate, this reaction is exergonic.

One of the most thoroughly researched acetogens in recent years is Acetobacterium woodii. It is a Gram-positive, mesophilic (30 °C), quinone- and cytochrome-deficient bacterium. The presence of Na⁺ ions is necessary for A. woodii development.

The Gram-positive, chemolithotrophic, motile anaerobe C. autoethanogenum is regarded as another model acetogen with established industrial applications. Its main natural products when grown autotrophically on syngas are acetate, ethanol, 2,3-butandiol, and lactate. The WLP, which dates back thousands of years, is thought to be the main mechanism that promotes autotrophic production of fermentative products and biomass.²⁶

Microorganisms show a variety of metabolic pathways that can be engineered to overcome the obstacles in the biofuel production. These pathways have revolutionized the norms for producing both conventional and advanced biofuels (bioethanol, biobutanol, etc.). drop-in fuel, biohydrogen, biodiesel, etc. By changing the molecular mechanisms connected to the metabolic pathways that produce fuel, this method tries to improve the metabolic performance of microorganisms. The availability of cutting-edge research tools such as CRISPR/Cas9, MAGE, and other ‘omics’ platforms and databases, such as genomics, transcriptomics, proteomics, and metabolomics, has accelerated metabolic engineering-based approaches and made biofuel production efficient and labor intensive.⁴⁰ Several metabolic engineering techniques have been used to modify metabolic pathways for better biofuel generation, including improvement of carbon fluxes toward the target pathway, incorporation of heterologous pathways, introduction of consolidated pathways, engineering redox balance for adequate supply of NADH/NAD(P)H (cofactor engineering), and directed enzyme evolution.^41–45

3. Gas–liquid mass transfer mechanism

The gas–liquid mass transfer of CO and H₂ in the aqueous phase is one of the main bottlenecks of syngas fermentation.^46,47 Various approaches, including increased pressure, different reactor configurations, high flow rates, large specific gas–liquid interfacial areas, modified fluid flow patterns, innovative impeller designs, use of microbubble dispersers and varying mixing times and speeds, have been investigated for enhancing the solubility of gas in the aqueous phase of the fermentation medium.^48–51 Most of these methods enhance the power-to-volume ratio that increases the breakup of the bubbles, thereby enhancing the surface area for mass transfer.⁵²

Generally, mass transfer takes place either via diffusion or convection. The diffusion of mass occurs at a different concentration gradient of a component that moves through a mixture from a high concentration gradient to a low concentration gradient. On the other hand, convection is mass transfer due to the bulk motion of a fluid, where the component movement requires mixing. There are three phases of mass transfer: gas segregation within the bulk phase, gas diffusion from the gaseous to the aqueous phase, and gas diffusion into the cells of microbes. The bulk segregation is instantaneous, and the microbial cell specific surface area causes the liquid film resistance to be negligible. Therefore, gas diffusion into the aqueous medium results in limited mass transfer.^53,54 Diffusion is described in two ways. The first is called Fick's law, which helps study diffusion with respect to position.

where j_i denotes the flux/unit area, D indicates the coefficient of diffusion, c₁ is the diluted component concentration and z denotes the distance of diffusion.

The second mathematical representation comprises a mass transfer coefficient.

where N₁ is the flux/unit area, kl is the coefficient of mass transfer, c_1i is the interface concentration, and c₁ is the bulk concentration. N₁ is not the same as j_i, as the second model is not completely reliant on the phenomenon of diffusion; rather, it also includes convection. It is difficult to determine which model to use in a given scenario. If the research is very fundamental, then using Fick's law would be preferable. Nonetheless, if the experimental data describe a more general representation of mass transfer, then the latter is more suitable. Another formula that describes the gas/liquid interface is stated below:

Here, N denotes the mass flux/unit area, kl is the coefficient of mass transfer, c_1i is the interface concentration and c₁ is the bulk concentration. This is a steady-state model; however, to evaluate diffusion with time, a time-dependent model would be more appropriate. One such example is Fick's second law:

The concentration is dependent on time and the flux is dependent on the distance of diffusion. Nevertheless, the limitations of the film model are that it overlooks undefined film turbulence. Another model that describes mass transfer is the penetration model, where the mass transfer is related to the contact time.

where D denotes the molecular diffusivity of the gas in liquid and τ represents the average residence time.

This model is also called “Higbie's model”, which assumes that each liquid element at the gas–liquid interface is exposed to the gas for a short time. The three main assumptions of this model are (1) mass transfer from the gas into a liquid element occurs under unsteady-state conditions, (2) each liquid element stays in contact with the gas for the same duration of time, and (3) equilibrium occurs at the gas–liquid interface. This model is an improvement on the previously mentioned film model, as mass transfer takes place under unsteady-state conditions in different processes at the industrial scale. Various models have been reported in the published works, however, the selection of the most suitable model depends on different factors including bioreactor design and fermentation medium.

Alternative bioreactor configurations have also been employed to achieve efficient mass transfer.^48,52 The most consistent factor to investigate the efficient mass transfer rate is the gas–liquid volumetric mass transfer coefficient, which indicates the hydrodynamic reactor conditions. According to a study, a maximum kl of 190 and 75 L h⁻¹ was achieved for H₂ gas in a STR at 300 rpm using a mixed culture of sulfate-reducing bacteria as a biocatalyst.⁵⁵ Another study stated kl values of 0.4 to 91 L h⁻¹ for CO for seven types of bioreactors.⁴⁸ It has been reported that multilayered composite hollow fiber, nonporous, ultrathin membranes made from polyethylene potentially increase the gas–liquid mass transfer in the liquid phase by preventing permeation of liquid and only allowing gases to pass through. If the inlet gas pressure is increased, higher levels of gas saturation might be achieved.⁵⁶ Many studies have focused on using composite hollow fiber (CHF) membranes to enhance CO mass transfer by investigating the impacts of recirculation flow rates and inlet gas pressure on the solubility of gas.^48,57 These studies indicated that employing CHF membranes is advantageous and increases CO transfer into the liquid phase. Another way to solve this problem is to employ a multipronged strategy for an efficient syngas fermentation process that relies on three basic approaches: pathway and strain optimization and development of the process. Nonetheless, efficient strain development has been negatively impacted by the lack of high-throughput strain engineering workflows, and there is not a well-defined pathway for industrial scale-up. Therefore, more work can be done to integrate high-throughput strain engineering workflows, omics, cell-free systems, kinetic modeling, fermentation scale-up and life-cycle assessments. Different heterologous pathway enzymes can be optimized to achieve the desired molecular transformations; later the strains can be optimized to increase the production of the targeted products.⁵⁸

The above-mentioned strategies may lead to enhancement in the gas–liquid mass transfer rates of poorly soluble gases into the liquid phase. Thus, it is clear that further advancements will be made in the near future.

4. Factors affecting syngas fermentation and optimization

4.1. Medium composition

A syngas fermentation medium must contain essential nutrients for supporting the metabolic growth of syngas-fermenting acetogens, which include carbon, organic and inorganic nitrogen sources, amino acids, and trace metals. The most essential medium component is carbon, which plays a key role in the growth and production of primary and secondary metabolites. Carbon sources are metabolized at different rates, which also affects biomass formation. According to a previous study, it was reported that the gradual assimilation of carbon sources usually enhances secondary metabolite production.^59,60 Different carbon sources along with their roles are summarized in Table 3. Notably, in fermentation processes where medium components contribute significantly to the product cost, the choice of carbon sources in the medium is very important. Apart from carbon source assimilation, the nature of the carbon source also influences the amount and type of the product. Therefore, it is important to study carbon source dynamics to determine their role as substrates in the process of fermentation. Inorganic nitrogen sources such as ammonium ions (NH₄⁺) have been reported to significantly influence cell mass and ethanol production by C. ragsdalei.⁶¹ Organic nitrogen sources such as tryptone and peptone help in the synthesis of core proteins during metabolite production and enhancement of carbon uptake.⁶² Amino acids are building blocks of different metalloenzymes and proteins in the Wood–Ljungdahl pathway. According to research that studied the effects of trace elements on syngas fermentation, it was reported that if nickel (Ni⁺²) was removed from the C. ragsdalei medium, no cell growth or acetic acid and ethanol production was observed.⁶³ Other trace elements, such as WO₄⁻², Fe⁺², Co⁺², and Mo⁺⁶, were also removed from the medium to study the effects on ethanol formation, and it was seen that the ethanol concentration was reduced by 97%, 82%, 24% and 38%, respectively. In the same research study, the effect of adding Ni⁺², Zn⁺², and SeO₄⁻ was also observed. Ethanol production increased with the addition of these trace elements along with an increase in FDH activity. The activity of CODH and hydrogenase was also enhanced by the addition of Ni; nonetheless, Fe removal inhibited these enzymes. Another study explored the impacts of W on S. ovata grown with a mixture of CO₂ and H₂ (80 : 20 by volume). It was observed that there was a 19% enhancement in cell growth, whereas the conversion of propionate and butyrate increased by 60% and butyrate by 32% along with the production of alcohols.⁶⁴ The effect of Zn concentration (7–280 μM) on the growth of C. carbooxidivorans was also studied, and it was observed that Zn concentration not only doubled cell growth but also increased ethanol, butanol, and hexanol production.⁶⁵ In conclusion, an optimized medium that has a balanced concentration of nutrients is very important to support the growth of microbes and the production of high value-added bioproducts.

Table 3 Different nutrient sources (carbon, nitrogen, amino acids, and trace elements) and their impacts on syngas fermentation

	Composition/source	Biological catalyst	Effects on the production	Ref.
Syngas (CO : H2 : CO2 : N2 : CH4)	20 : 10 : 20 : 50 : 0	C. carboxidivorans	1.9 g L^−1 butanol; 2.7 g L^−1 ethanol; 0.85 g L^−1 hexanol	66 and 67
	50 : 15 : 35 : 0 : 0	C. carboxidivorans	1 g L^−1 butanol; 2 g L^−1 ethanol; 0.5 g L^−1 hexanol	66 and 67
	60 : 0 : 40 : 0 : 0	C. carboxidivorans	1.8 g L^−1 ethanol, 0.66 g L^−1 butanol, 3.35 g L^−1 acetic acid, and 0.38 g L^−1 hexanol	67 and 68
	25 : 44 : 10 : 10 : 11	C. carboxidivorans	1.76 g L^−1 ethanol; 1.32 g L^−1 acetic acid; 0.43 g L^−1 butanol	67 and 69
	40 : 30 : 30 : 0 : 0	C. ragsdalei	Significant reduction in hydrogenase activity with NH3 in the syngas	67 and 70
	32.5 : 32.5 : 16 : 19 : 0 (without impurities)	C. ljungdahlii	2.47 g L^−1 ethanol; 16.75 g L^−1 acetic acid	67 and 71
	80 : 0:20 : 0 : 0 (without impurities)	C. carboxidivorans	1.17 g L^−1 ethanol; 0.96 g L^−1 acetic acid; 0.56 g L^−1 butanol	67 and 68
	80 : 0:20 : 0 : 0 (0.1 g L^−1 H2S)	C. carboxidivorans	3.2 g L^−1 ethanol, 0.8 g L^−1 acetic acid, and 0.38 g L^−1 hexanoic acid	67 and 68
	80 : 0 : 20 : 0 : 0 (0.1 g L^−1 NaNO3)	C. carboxidivorans	1.1 g L^−1 ethanol, 0.38 g L^−1 acetic acid, and 2.04 g L^−1 butyric acid	67 and 68
Inorganic nitrogen	NH4^+	C. ragsdalei	Cell mass and ethanol production reduced by 33% and 41%, respectively, if removed	61
Organic nitrogen	Tryptone	C. autoethanogenum	45.9% enhancement in ethanol production with 0.5 g L^−1 tryptone	62
	Peptone	C. autoethanogenum	44.3% enhancement in ethanol with 0.5 g L^−1 peptone	62
Amino acids	L-Cysteine	C. autoethanogenum	Cysteine addition did not impact ethanol production. Conc. higher than 0.5 g L^−1 negatively impacts the biomass	72
Trace elements	Fe	C. ragsdalei	Ethanol reduced by 82% on eliminating Fe	63 and 73
	Co	C. ragsdalei	Ethanol reduced by 24% on eliminating Co	63 and 73
	Mn	C. ragsdalei	No effect on cell growth, ethanol, and acetate production	63 and 73
	Ni	C. ragsdalei	No effect on cell growth, ethanol, and acetate production	63 and 73
	Mo	C. ragsdalei	Ethanol decreased by 51% with 20% less Mo	17, 63 and 73
	Se	C. ragsdalei	Ethanol production increased by 52% on adding 10.6 μM Se	63 and 73
	W	C. ragsdalei	Ethanol increased by 102% on adding 6.8 μM W	63, 64 and 73
		S. ovata	Ethanol increased by 206% on adding 0.1 μM W	63 and 73
	Zn	C. ragsdalei	Ethanol increased 4.2-fold with 66.9 μM Zn	63 and 73
		C. carboxidivorans	3.0-Fold, 7.6-fold, and 44-fold increase in ethanol, butanol and hexanol on adding 280 μM Zn	73 and 74

---

Different medium optimization techniques can be employed to enhance the productivity of ethanol and other value-added bioproducts. These techniques include single factor studies, artificial neural networks (ANNs), Plackett–Burman design, response surface methodology and advanced synthetic biology techniques. Employing them either as a single technique or in combination may give desirable results.

4.2. Fermentation conditions

Syngas fermentation is greatly influenced by the process operating conditions, including pH, temperature, and pressure. The pH of the medium leads to a shift between the acetogenesis and solventogenesis phases of fermentation, as volatile fatty acid (VFA) accumulation during the former causes a reduction in pH, which is then followed by the latter, resulting in the conversion of acid to alcohols.³⁵ Many studies have reported this phenomenon while studying the impacts of pH on the syngas fermentation process using different biocatalysts, such as C. carboxidivorans, anaerobic mixed cultures, C. ljungdahlii and C. ragsdalei.^75–79 These studies have indicated that a pH drop ranging between 4.5 and 5.0 leads to solventogenesis, whereas operating at a relatively higher pH (5.0–6.0) favors acetogenesis, which is related to cell growth and VFA production. A study was carried out using mixed cultures of C. ljungdahlii and C. kluyveri, and a narrow pH range was reported for the production of elongated chain alcohols.⁸⁰ A study developed a novel method to control syngas supply based on pH that shows the activity of the cell. This method has resulted in the stability of the process and enhanced ethanol production during the continuous fermentation of syngas.⁷⁹ Temperature is another important factor for the growth and metabolism of different acetogens, as shown in Table 2. However, it should be noted that the optimal temperature needed for growth and VFA accumulation might not be favorable for the production of alcohol. For instance, a study reported that C. carboxidivorans exhibits enhanced production of alcohol and chain elongation at 25 °C.⁸¹ Another study reported higher production of acetate, butanol, butyrate, ethanol and hexanol with C. carboxidivorans when the temperature was decreased from 37 °C to 25 °C.⁸² The pressure of the system is another important parameter that affects the fermentation process. The main obstacle of syngas fermentation is the rate of mass transfer, which can be overcome by increasing the pressure. A study specifically focused on studying the impacts of moderately elevated pressures (5–10 bar) on the process.⁸³ Previous researchers have laid a strong foundation for syngas fermentations carried out at higher pressures to increase mass transfer rates; however, moderate elevation of pressures using substrate gases, including CO₂, CO, and H₂, has not been fully explored. Although the growth rate of acetogens and formation of the product are improved at higher pressures, the titers must be increased to facilitate industrialization in the future. Therefore, further investigations into the optimization of operating parameters are needed. A summary of the operating conditions is given in Table 4.

Table 4 Effects of operating conditions (pH, temperature, and pressure) on syngas fermentation

Substrate gas	Biocatalyst	Operating conditions				Conversion efficiency of gas	Impact on product formation	Ref.
		pH	Temp. °C	Pressure	Reactor type
Syngas CO/CO2/N2(25 : 15 : 60)	C. carboxidivorans	5.2	37	NR	BCR	CO: 60%	0.33 mol ethanol/mol CO; 0.04 mol acetate/mol CO	75
CO	C. carboxidivorans	5.75(*)	33	NR	Continuous gas-fed	CO: 50%	1.04 g L^−1 acetic acid; ethanol: 7.52 g L^−1	76
CO	C. carboxidivorans	4.75(**)	33	NR	Continuous gas-fed	CO: 50%	0.06 g L^−1 acetic acid; ethanol: 4.21 g L^−1	76
CO/H2/CO2 (60 : 30 : 10)	Anaerobic mixed cultures	6.9	37	1.0 atm	This experiment was carried out in vials	NR	0.92 mol VFA/mol syngas	77
CO and H2	C. ljungdahlii	5.9	37	1,4, and 7 bar	STR (batch)	NR	At 7 bar 4 g L^−1 of VFAs produced comprising 15.6% acetic acid, 1.7% ethanol, and 82.7% formic acid	78

---

4.3. Reactors

One of the most crucial factors that should be considered for the optimal growth of microorganisms during syngas fermentation is the bioreactor configuration that will significantly influence the entire process, whether it is batch, continuous, semicontinuous, or a combination of processes. Therefore, at an industrial scale, different factors influence the final choice of the bioreactor and the operation mode. Different types of reactors, including stirred vessels, bubble columns, trickle bed or membrane reactors, and gas lift and loop reactors, have been studied and compared by scientists for syngas fermentation.^54,84 The most suitable reactors for syngas fermentation are those that achieve high cell concentrations and mass transfer in an energy saving manner.^54,55 The reactor types presented in Fig. 2 have been widely studied in different laboratory settings.^4,28,85,86 A schematic diagram of different types of reactors is illustrated in Fig. 2.

Fig. 2 Main types of syngas fermentation reactors. (A) continuous stirred tank reactor (CSTR), (B) bubble column reactor (BCR), (C) trickle bed reactor (TBR), (D) hollow fiber membrane bioreactor (HFMBR), and (E) gas lift bioreactor (GLR); “adapted from ref. 4”.

A CSTR is shown in Fig. 2A, where mechanical agitation is performed for mixing and the desired reaction conditions are adjusted easily. However, in high volume (>500 m³) fermenters, an increased energy input is needed for sufficient mixing. Many reaction-engineering studies have been reported on CSTRs, both in continuous and batch modes. Nonetheless, most of the published research work is basic and only investigated different parameters affecting the syngas fermentation process, but the development of an optimized reactor has not been reported. Although the CSTR is commonly used for syngas fermentation, it is not preferable for industrial-scale usage because it is energy intensive.^48,55,87,88 As a high energy input is an impediment to economic scale-up for obtaining value-added products, the simplest and most attractive reactors with regards to industrial application are bubble columns (Fig. 2B) due to their low cost and energy requirements.⁸⁹ There is a possibility of large volumes in bubble columns, but liquid phase mixing is limited. Bubble column reactors (BCRs) have complex flow systems and axial gradients in gas pressure, biomass concentrations, pH of the fermentation medium and height of the reactor. Hence, to enhance efficiency, BCRs have been modified as gas lift reactors (GLRs), in which the gas enters via the lower end, thus causing upward flow of the gas bubbles and liquid phase.^90–92 One of the most common problems with reactors is the availability of biocatalysts. To ensure a higher cell concentration, biofilm reactors, such as trickle bed reactors (TBRs) (Fig. 2C), have been developed to prevent a loss of cells while achieving high gas–liquid transport rates. However, the stability of the biofilm is a gray area, and very little research has been performed on it. Therefore, biofilm reactors with immobilized cells are of interest to scientists in further research.⁹³ Another type of reactor is the HFMBR (Fig. 2D), which is a unique reactor that comprises microporous hollow fiber membranes as gas distributors. The diffusion of gas molecules to the liquid takes place via the membrane, where the biocatalyst sticks to the surface as a biofilm; therefore, the membrane works as a gas input as well as a carrier for useful microorganisms.^88,94 HFMBRs are gaining increasing attention because of their low energy input requirements and high mass transfer rates.⁸⁸ Nonetheless, the selection of a suitable membrane material is an important factor because the membrane surface strongly impacts mass transfer. Furthermore, special HFMBRs for gas fermentation are not easily available in the market. Moreover, membrane stability under real fermentation conditions is also a matter of concern.^88,95 Although the membrane module offers a great advantage with regard to mass transfer, this configuration is relatively intricate.^88,96 As shown in Fig. 2E, GLRs result in improved mixing and a more defined flow profile.⁹⁰ Nonetheless, in GLRs, there is a problem of substrate limitation due to gas voids in the downcomer area. To solve these bottlenecks in conventional BCRs and GLRs, an external pump is attached to the BCR, which improves liquid phase mixing and achieves higher k_La values.⁹¹ In the literature, this reactor configuration is known as a forced circulation loop reactor.⁹²

Data on syngas fermentation carried out at the pilot scale using different types of reactors have rarely been reported. However, the information extracted from published patents and articles indicates that the most famous companies, including BRI/Ineos Bio and LanzaTech, use the CSTR and loop reactor, respectively, for fermentation. Coskata has published a report stating the use of an HFM module reactor, but no information on reactors with a volume >10 L has been published.⁹⁷ Additionally, Ineos Bio and Coskata went out of business, and LanzaTech is solely operating at a larger scale.⁸⁹

To overcome the challenges of using different reactor types at the industrial scale, a bubble column with liquid circulation is expected to be a sound method for the successful commercialization of syngas fermentation. A standard column with circulation is not capital intensive, and the forced circulation facilitates sufficient mixing of the aqueous phase and gas–liquid mass transfer (GLMT) rates. Two main factors that determine the power input are the pump flow rate and the volume of substrate gas. For energy efficiency, GLMT can be optimized by installing internal packings, improving gas delivery systems, or using other supporting technologies.^46,98 Additionally, moderate pressurization is advisable to enhance mass transfer, as high solubility at higher pressure makes the substrate gas available for the biocatalyst.^46,89 Generally, simple design of the reactor and continuous fermentation have been proven to be better strategies to achieve higher kl values.^{31,55,84,89,99} A thorough comparison of different reactor types with regard to their performance and scale-up cost is given in Table 5. Furthermore, a two- or multistage system is preferable because it has improved process control: if acidogenesis and solventogenesis phases are separated, then process parameters can be optimized to enhance reactor productivity.^80,84,100 A summary of different types of reactors, such as a CSTR (160 L), TBR (144 L), HFMBR (10 L), HFMBR (7.5 L), loop reactor (71 L), GLR (50 L), forced-circulation loop reactor with a secondary loop (390 L and 9800 L), MBBR (36 m3) with biocatalysts such as Clostridium ljungdahlii ERI2 ATCC 55380, Clostridium ljungdahlii ERI2 ATCC 55380, Clostridium ragsdalei ATCC BAA-622, Clostridium ragsdalei ATCC BAA-622, Clostridium autoethanogenum DSMZ 19630, and Clostridium autoethanogenum DSMZ 10061, with different specifications for each reactor is given in a previously published work.⁴ However, the most important parameters to choose a suitable reactor include achievable gas–liquid mass transfer rates, investment, operational costs, application perspective, scale-up and commercialization.

Table 5 A comparison of the performance and scale-up costs of different types of reactors for syngas fermentation ¹¹²

---

4.4. Culture strategies

Microbes interact with one another and have complementary roles in ecological networks where they coexist.¹⁰¹ Two-phase systems and cocultures have gained increasing attention for use in syngas fermentation, as they resemble the microflora of a real fermenter, which can promote the accumulation of value-added bioproducts. Cross-feeding processes create this synergy; for instance, the metabolic products formed by one or many organisms serve as substrates for the others.¹⁰² Mixed culture fermentation between Alkalibaculum bacchi and C. propionicum has been reported. The use of Alkalibaculum bacchi in a monoculture produced only ethanol from syngas, whereas the mixed culture produced additional bioproducts, including n-propanol and n-butanol, in addition to ethanol.¹⁴ According to another study, C. ljungdahlii and C. kluyveri were cross-fed to yield butanol and hexanol.³⁵ Cocultures of C. kluyveri and C. autoethanogenum were employed in the chain elongation of organic acids and alcohols.¹⁰³ For C. ljungdahlii alone, CO was lethal, but when employed in a coculture with C. autoethanogenum, the two microbes exhibited sound growth and produced butanol, butyric acid, hexanol and hexanoic acid.³⁶ Syngas was used as the primary substrate to develop a two-stage fermentation system to produce malic acid.³⁷ In the first reactor, syngas was converted into acetic acid using C. ljungdahlii as a biocatalyst that was later used as a carbon source for the growth of Aspergillus oryzae in the second reactor. This implies that syngas can be converted into a variety of value-added bioproducts and can be used as an alternative substrate to produce malic acid, which generally grows on sugars and glycerol. Another study proposed a two-stage fermentation in which syngas (CO : CO₂ : H₂ : N₂, 30 : 10 : 20 : 40) fixation occurred in the first stage by utilizing C. autoethanogenum under anaerobic conditions to obtain 2.66 g L⁻¹ acetic acid, ethanol and 2,3 butanediol.⁷ In the second stage, mixed microbial consortia (MMCs) were employed to produce PHA in a fed-batch culture utilizing first-stage products as substrates. Recently, the genetic editing tool CRISPR/Cas9 has been employed for deleting the target genes in C. coskatii, C. ljungdahlii and C. autoethanogenum to modify the genome of these acetogens to make them more efficient in increasing the production of desired products.^38,39,104 Among all strategies mentioned above, cocultures of microorganisms are a promising approach to expand syngas utilization to obtain a broad spectrum of high value-added bioproducts. This is because they are potentially more resilient than monoculture approaches. Furthermore, apart from in vivo studies, many researchers are making efforts to model synthetic cocultures in silico on the basis of kinetic or genomic data.¹⁰⁵ Microbial coculture can be combined with experimental work and in silico simulations to improve and accelerate the ecology of consortia, leading to advanced microbial cultivation. While cross-feeding seems to be a promising approach for designing synthetic communities, microbial interactions, including metabolite secretion and competition, can most likely lead to lower efficiency of the coculture. Sometimes ‘unwanted interactions’ can be helpful in creating ‘artificial’ interdependencies to make the cultures more robust. For instance, some strains that generate toxins to kill an undesired contaminant or have a metabolic activity that triggers a specific metabolism can be exploited. However, microbial interactions and the effects of prolonged coevolution during this process still need to be explored. These topics require further research to gain deep insights into the applicability of the most suitable culture strategies in the biotechnology industry.

5. Fermentation system modeling

5.1. Kinetic modeling

Thermodynamics govern the course and potential size of the events that take place along the production route, whereas kinetics characterizes the speed of those reactions as well as the overall rates of CO and H₂ consumption, acetic acid and ethanol build-up, and cell growth. Overall rates are anticipated to be proportional to the number of cells in the fermenter (XV_L), with the specific growth rate (μ) for growth and the specific absorption rate (qCO for CO and qH₂ for H₂) serving as the coefficients of proportionality. A kinetic model, such as Michaelis–Menten for enzyme-mediated processes, is used to connect individual reaction rates to the concentrations of the reactants and products. In a kinetic model such as the Monod equation, the specific growth and uptake rates are also associated with the concentration of substrates, such as CO and H_2, inside the cell.¹⁰¹

Basically, kinetic models aim to describe certain features of cellular function or production. Biotechnology is interested in any procedure that could affect how well a cell factory functions. The scope and concentration of potentially pertinent kinetic models are broad due to the variety of cell factories and the abundance of interesting products that can be created by them. Depending on the issue, they may focus on internal or environmental factors such as temperature, pH, osmolality, product and byproduct toxicity, and the type and mode of the fermenter. These include cellular processes such as metabolism, protein maturation and secretion, signaling, gene regulation, stress responses, cell cycle progression, and apoptosis.¹⁰⁶

Various kinetic models, such as spatiotemporal metabolic models, have also been employed to solve the bottlenecks of syngas fermentation. A study developed a model for a BCR using C. ljungdahlii as a model organism for syngas fermentation. This modeling approach comprised combining a genome-scale reconstruction of C. ljungdahlii metabolism with multiphase transport equations that consider both the process of convection and dispersion occurring in the column. The reactor model equations and programs were solved using the MATLAB-based code DFBA lab. Simulations were carried out to analyze the significant parameters, including the ethanol-to-acetate ratio, ethanol titer, and conversion of CO and H₂. The model predictions obtained from this study might be helpful in cellular and process engineering to overcome the main bottlenecks for ethanol production in a BCR.¹⁰⁷ Another study developed a model for predicting the performance of a syngas-fermenting BCR for the recovery of anhydrous ethanol. An optimization framework was developed that employed artificial neural networks and genetic algorithms to enhance various process conditions and design variables, including the MESP, cost, energy efficiency and reactor productivity.¹⁰⁸ Another study employed a spatiotemporal metabolic model by using CO fermentation data assembled from a laboratory-scale BCR using MATLAB. This modeling approach comprised combining a genome-scale reconstruction of C. autoethanogenum metabolism with multiphase transport equations that consider both the convective and dispersive processes that impact CO transport, biomass, and secreted byproducts. The simulated values were in conformity with the measured values for acetate, ethanol, and biomass concentrations at a single gas flow rate. Later, the predictions were carried out at three different gas flow rates. Spatiotemporal modeling has the potential to assist in the commercial-scale design of processes involved in syngas fermentation to produce high value-added bioproducts.¹⁰⁹ Other kinetic models have been developed to study ethanol and acetate production from syngas fermentation. For example, a study employed kinetic modeling to investigate the parameters affecting syngas fermentation in different pressurized batch bioreactors, such as substrate uptake, intrinsic growth, and product recovery, for anaerobic bacteria using C. ljungdahlii as the model organism. A dual substrate such as CO and H₂ growth kinetic model using Luong and Monod was developed for describing the growth of C. ljingdahlii. A maximum specific growth rate of 0.195 h⁻¹ and Monod constants of 0.855 atm and 0.412 atm for CO and H₂ were achieved. Furthermore, the impacts of CO on the growth of the cell at high pressures were studied, and it was found that if the CO pressure exceeded 0.743 atm, no significant growth was observed. Other kinetic models that were employed to study the above-mentioned parameters include the Volterra model, Andrews model, and modified Gompertz model.¹¹⁰

Conclusively, a model that describes the aforementioned characteristics of cell factories may be created specifically for a given application, such as a particular pathway for the manufacture of a particular metabolite, or it may describe more generic characteristics. The vast range of timeframes of often modeled processes reflects the diversity of kinetic model objectives and domains in biotechnology.

5.2. Reactor modeling

Computational fluid dynamics (CFD) is a computer-assisted technique that is generally employed to simulate complex fluid flow using numerical approaches to predict fluid mechanics and heat transfer. Its calculation is based on model division of a real piece in various cells, which is followed by applying differential equations to each individual cell to improve the flow models by obtaining the evolution of physical properties and flow variables, including temperature, pressure, velocity, and turbulence.¹¹¹ A recent study employed the CFD technique to investigate syngas fermentation in a 125 m³ BCR using C. ljungdahlii DSM 13528 as a model organism. The factors that were investigated included short- and long-term responses, circulation time and bacterial motion. A pseudostationary gas gradient was conducted with CFD in a Euler–Euler approach, whereas the microorganism movement was simulated as Lagrangian massless particles. Lagrangian trajectory analysis of the gas gradients of a BCR is helpful in predicting performance and therefore minimizing the risk in scale-up. Although CFD has been employed in different studies for reactor modeling, there are some disadvantages to CFD simulations, including accurate modeling of the composition. Therefore, scientists have developed alternative models to study the desired thermodynamic properties in the virtual setting and to better address the complexity of the process.¹¹²

5.3. Process modeling

One of the most popular chemical engineering tools is the commercially available software called Aspen Plus®, designed by AspenTech®. It is commonly used for simulating real processes by investigating any compound and mixture under various operational conditions. Many studies have employed Aspen modeling to optimize reactor performance. A study developed a simulation model with ASPEN Plus for the gasification of lignocellulosic biomass combined with syngas fermentation to produce ethanol.¹¹³ The simulation was carried out in three modules: biomass gasification, syngas fermentation, and ethanol recovery. The model predicted that a gasification temperature of 700–1000 °C and an equivalence ratio from 0.2–0.4 were optimal for high ethanol production.¹¹³ Another study also proposed a novel simulation model using Aspen Plus® based on the biorefinery concept for the conversion of 1200 tons per day of switchgrass to anhydrous ethanol.¹¹⁴ The simulation model in this study was based on the same three modules as mentioned above and predicted an annual production of approximately 36.5 million gallons of anhydrous ethanol. Another study modeled an HFMBR to solve the problem of gas–liquid mass transfer during syngas fermentation by optimizing three operational factors: membrane surface area per working volume (A/v), water velocity (VL), and specific gas flow rate (Vg). The maximum kl of 385.0 h⁻¹ in water was observed for the mass transfer of CO at an A/v of 0.56 cm⁻¹, a VL of 2.20 cm s⁻¹, and a Vg of 1.02 min⁻¹, which is relatively higher than the kl values given by other studies using different models. A three-factor quadratic model and a dimensionless model were developed from experimental data with high correlation coefficients to facilitate the scale-up process. The most significant factor with regard to kl is the membrane surface area, which positively impacts the process.⁹⁴ The membrane surface area is an important parameter considered in process modeling because CO and H₂ are less soluble in water than O₂ (104 g g⁻¹) at 37 °C, and therefore, gas–liquid mass transfer is difficult. To improve gas–liquid mass transfer, numerous studies have examined various reactor designs and configurations including hollow fiber membrane reactors (HFMRs), bubble column reactors, trickling bed reactors, CSTRs, and gas lift reactors.

According to Yasin et al., the HFMR volumetric mass transfer coefficient k_La increases as the membrane surface area per unit working volume and pressure increases. In microbial syngas systems, researchers have suggested utilizing a submerged HFMR, which results in high CO mass transfer rates. As the fluctuations in the pressure through the hollow fiber lumen and the membrane's surface area per unit working volume of the liquid influence the gas–liquid k_La,¹¹⁵ a study has reported that with a lower pressure of 37.23 kPa and a larger membrane surface area per unit working volume AS/VL of 62.5 m⁻¹, a higher volumetric mass transfer coefficient k_La of 155.16 h⁻¹ can be achieved.¹¹⁶ In other types of reactors, such as BCRs, the most important parameters that are considered include gas and liquid phase concentrations, whereas in CSTRs, the specific gas uptake, cell retention time, ethanol and acetic acid concentration, cell concentration, CO₂ conversion efficiency, and H₂ conversion efficiency are important parameters.

Different studies have been carried out to use different modeling approaches to study these parameters to solve the main bottlenecks of syngas fermentation. A summary of these is given in Table 6.

Table 6 Summary of CFD, Aspen, and MATLAB used in different studies for the optimization of different types of bioreactors for syngas fermentation

Reactor type	Parameters and modeling approaches	Main findings	Microbial catalyst	Ref.
Bubble column reactors	Gas and liquid phase concentrations	Mass transfer coefficient of 500 h^−1 for CO; predicted ethanol (130 g L^−1, predicted conversion of H2 and CO of 89% and 34%)	C. ljungdahlii	107
	Modeling approach: MATLAB based code DFBA lab
Down-draft reactors	Low heating value (LHV), temperature, equivalence ratio (ER), cold gas efficiency (CGE), and quantity of bioethanol	Optimal gasification temperature (700–1000 °C). Ethanol 0.114 kg h^−1 per 1 kg h^−1 of garden waste	C. ljungdahlii	113
	Modeling approach: ASPEN plus V.10; the non-random-two-liquid (NRTL) model; HCOALGEN and DCOALIGT models; MCINCPSD stream, three sub streams of the MIXED, CIPSD, and NCPSD classes
CSTR	Specific gas uptake, cell retention time, ethanol & acetic acid concentration, cell concentration, CO2 conversion efficiency, H2 conversion efficiency	Predicted ethanol production of 36.5 million gallons per year	C. ragsdalei	114
	Modeling approach: Aspen Plus V. 8.2; NRTL model
External microporous hollow fiber membrane diffuser	CO transfer, membrane surface area per working volume, water velocity, and specific gas flow rate	Higher kl of CO using an HFM diffuser, the membrane area positively impacts CO kl	NR	94
	Modeling approach: empirical (quadratic and dimensionless) regressions
Pressurized batch bioreactors	Growth kinetics of bacteria with CO and H2 at high partial pressures, substrate uptake rate & product formation	Maximum CO uptake rate is 34.364 mmol g^−1 cell h^−1, maximum production rate of ethanol and acetate is 0.172 and 0.096 mmol L^−1 h^−1, respectively, at a P(CO) of 0.598 atm and 0.539 atm, respectively. For H2 the maximum growth rate was achieved at 0.412 atm	C. ljungdahlii	110
	Modeling approach: luong + monod; double-experimental + monod; luong + tessier; double-experimental + moser; luong + moser; double-experimental + tessier; andrew + tessier; edwards + monod; edwards + tessier; andrew + monod; haldane + moser; haldane + monod
CSTR	Gas composition, dilution rate, gas flow rates, and cell recycling	The predicted ethanol production was 2 g L^−1 h^−1 with a substrate gas composition of 54% CO and 46% H2 and a dilution rate of 0.06 h^−1	C. ljungdahlii	108
	Modeling approach: MATLAB
Bench-scale bubble column reactor	CO and CO2 gas–liquid mass transfer, cellular consumption of CO and CO2, and production of ethanol, acetate, 2,3-butanediol, lactate and CO2	Ethanol production of 18.7 g L^−1 d^−1 and acetate production of 10.2 g L^−1 d^−1 at a gas flow rate of 700 ml min^−1 at 53.5% conversion of inlet CO	C. autoethanogenum	109
	Modeling approach: MATLAB code DFBA lab
Bubble column reactor	Bacterial motion patterns, circulation time, qnd short- and long-term responses	Substrate limitations (97% of all cells) while 84% react to CO limitations when exposed to the stress zone for more than 70 s	C. ljungdahlii	112
	Modeling approach: computational fluid dynamics

---

5.4. Life cycle assessment

The environmental performance of biofuels requires evaluation that helps in technology and policy development for their sustainable production.¹¹⁷ Recently, the assessment of environmental performance has been carried out through life cycle assessment (LCA), which is a standard tool given by the International Organization for Standardization (ISO) for quantifying the environmental impacts of natural resource utilization and emissions.¹¹⁸ This is done by collecting relevant information about the material and energy requirements and outputs of each step of the product life cycle, as shown in Fig. 3.

Fig. 3 Life cycle assessment; reproduced from¹¹⁹, with permission from Elsevier, copyright 2023.

According to preliminary life-cycle assessment (LCA) research on the environmental effects of China's steel mill waste petrol conversion to ethanol, there is a possibility for a low-emission fuel product. This article provides preliminary LCA results for ethanol made from biogenic (corn stover, forestry residues, and switch grass) feedstock and updates data for syngas conversion to ethanol based on design enhancements from LanzaTech's scale-up work and application of the technology to a domestic United States market. GHG emissions are calculated for each phase of the ethanol life cycle and contrasted with those of other ethanol-producing methods and conventional transportation fuels.¹²⁰

Various studies have carried out life cycle assessments of bioethanol and improvements have been observed in environmental performance with regard to greenhouse gas (GHG) emissions and consumption of energy compared to gasoline, which are summarized in Table 7. Considering industries, a study has used the LCA method to quantify the global warming potential of different scenarios for bioethanol with the LanzaTech process. Different scenarios were considered in which ethanol was produced from industry off gases as well as biomass such as corn stover, forest residue, or switchgrass, extracting input data from published reports, databases, and estimates of LanzaTech engineering processes. According to the standard LCA method, ethanol production via LanzaTech fermentation has reported a 60% reduction in GHG emissions compared with that of conventional gasoline, and biomass-based ethanol achieves approximately 90% reductions in emissions. All the input parameters have been reported in the published work.¹²⁰ It was reported that the fermentation technology reported by LanzaTech is a viable option to produce next-generation biofuels that meet the standard policy requirements of United States Renewable Fuels regarding the fuels with the lowest footprint. Conclusively, LCA proposes a unique biorefinery design that is energy efficient and mitigates the environmental effects of biomass-derived ethanol production.

Table 7 Life cycle assessment

Object of the assessment		Experimental inputs	Output products	System boundaries	Ref.
Biomasses	Processing technologies
Switchgrass	Biorefinery	Pretreatment, enzymatic hydrolysis, fermentation, recovery, and purification	Ethanol, electricity, furfural, acetic, and formic acid	Switchgrass cultivation, switchgrass transportation to the production facility, ethanol and co-product production, and ethanol distribution	118
Rice straw	Ethanol plant	Extraction process including water and varying alkali concentrations, dilute acid pretreatment, enzymatic hydrolysis, and ethanol production	Ethanol	Feedstock acquisition and ethanol production and transport	121
Wheat straw	Biochemical conversion	An advanced second-generation technology	Ethanol	WS cultivation, transportation, ethanol production, ethanol blending and distribution, and use in E10 and E85 passenger cars	122
Miscanthus	Biorefinery	Pretreatment, enzymatic hydrolysis, fermentation, and distillation/separation	Ethanol	Miscanthus cultivation, production, and transport of needed inputs such as fertilizers, pesticides, propagation material, management and harvesting operations	123
Cassava straw	Biorefinery	Improved hydrothermal pretreatment, enzymatic hydrolysis, fermentation, distillation, and adsorption dehydration	Bioethanol	Cultivation, transportation, preparation, and bioethanol conversion	124
Biomasses of first- and second-generation type	Biochemical conversion	Pretreatment, fermentation and ethanol recovery, wastewater treatment plant (WWTP), anaerobic digestion (AD) and combined heat and power (CHP) system	Ethanol; electricity surplus	(Well-to-wheel) bioethanol production from biomass at the plant gate to the ethanol end use	125
Corn	Biorefinery	Cleaning/milling, liquefaction/fiber separation, fermentation, and distillation	Ethanol	(Cradle-to-gate) analysis including infrastructure and fertilizer, pesticide and diesel use and the production of anhydrous ethanol at the plant	126

---

6. Scale up and commodification: a global perspective

Generally, the most efficient syngas bioreactor achieves enhanced mass transfer rates and high cell concentrations and is economical at the same time.⁵⁴ For the evaluation of syngas reactors, a ranking system was proposed in previously published studies, which mainly focused on reactor performance and scale-up costs.¹¹² The main parameters for scale-up include distribution, bubble breakage, bubble retention times, mass transfer rates, low production costs, low energy costs, high pressure, low contamination risks, low heat dissipation, microorganism yields, easy product recovery, and kl values. Due to the evaluation system, it was established by a previous study that the best performing reactors for gas fermentation include the microbubble STR combination and MBBR. The scale-up approximation factors include land acquisition and improvements, building area, laboratory instruments, large-scale bioreactor system, medium supply station, starter culture cultivation tank, immobilization material, product purification, product storage tank, centrifuge (foam) pumps, valves, piping and control, electrical engineering and supervision and construction expenses. It was reported that the most cost-efficient reactors in scale-up are the BCR, ALR and TBR. Considering the design and scale-up parameters, the most appropriate reactor configuration for syngas fermentation includes the BCR, ALR, MBR and TBR.

Until 2012, only three companies named INEOS Bio, Coskata and LanzaTech were dealing with the commercialization of syngas fermentation.²¹ However, the expansion of this industry has taken place over the last decade, and this section introduces all current industrial initiatives along with the substrates and products summarized in Table 8. INEOS Bio (2013) is a subsidiary of INEOS that utilizes propriety isolates of C. ljungdahlii. Their pilot-scale demonstration reported a production of 380 L ethanol per ton of dry feedstock. First, the gasification of the feedstock is carried out, and later, it is fermented to bioethanol and other high value-added bioproducts. Another company is Coskata, which uses C. ragsdalei and C. carboxidivorans as licensed biocatalysts from Oklahoma. Other strains, such as C. coskatii, have also been reported to be used for ethanol production. Coskata has reportedly produced 380 L of ethanol per ton of dry softwood in Madison and Pennsylvania. In 2015, Coskata ran out of business, and Synata Bio acquired the technology in 2016. LanzaTech (2005) has a diverse product range primarily focusing on ethanol and 2,3-butanediol production. Apart from syngas, LanzaTech also utilizes CO-rich industrial off-gases for producing chemicals, including 2,3-butanediol, butanol and ethanol. A demonstration plant has been set up in Auckland, New Zealand, that generates 380 L ethanol per year utilizing off-gases from steel mills. Industrial scale facilities have been established in China since 2017 with an annual production of 16 M gallons and in Gent (Belgium) with ArcelorMittal since 2018 with an annual production of 21 M gallons (LanzaTech, 2018). More commercial scale projects have been established in South Africa (Swayana), India (Indian Oil) and California (Aemetis) utilizing ferroalloy off-gases, refinery off-gases and gasified orchard biomass, respectively.¹²⁷

Table 8 Industries operating globally for the production of ethanol and other value-added products

Country	Company	Feedstock for syngas production	Products
USA	Coskata Inc., (http://www.coskata.com)	Non-renewable sources (natural gas, and coal), other gases include industrial gases; renewable sources such as lignocellulosic biomass	Butanol, butanediol, ethanol, propanol, hexanol, and fatty acids
	OPX Biotechnologies Inc., (http://www.opxbio.com)	CO2 and H2	Fatty acids
	LanzaTech Inc., (http://www.lanzatech.com)	Steel mill off gas and coal producers; lignocellulosic biomass such as wood residues	Ethanol and 2,3-butanediol
	Syngas Biofuels Energy Inc., (http://www.syngasbiofuelsenergy.com)	CO2 and H2	Fuel n-butanol
	BRI Energy Inc., USA (http://www.brienergy.com)	Organic waste such as municipal solid waste (MSW), agricultural residues and coal	Ethanol and electrical energy
	Aemetis Inc (https://www.aemetis.com/)	Orchard wood and nutshells	Ethanol
Switzerland	INEOS Bio, (http://www.ineos.com/businesses/ineos-bio/company)	Organic waste	Ethanol
Netherland	BioMethanol Chemie Nederland B.V., (http://www.biomcn.eu/)	Bio-feedstock like biogas & feedstock such as glycerine	Methanol
EU	SYNPOL project platform, EU (http://www.synpol.org)	MSW, sewage sludge and agricultural residues	Polyhydroxyalkanoates, hydroxybutyrate, butanediol, and succinate
China	Beijing Shougang Co., Ltd. (http://www.shougang.com.cn/)	Syngas produced as steel mill-off gas	Ethanol
	Jupeng Biotech Ltd. China (http://www.jupengbio.com/)	Organic waste	Ethanol
Luxembourg	ArcelorMittal (http://corporate.arcelormittal.com)	Steel mill-off gas (syngas)	Ethanol
South Africa	Swayana http://www.swayana.co.za	Ferroalloy off-gases	Ethanol
India	IndianOil https://iocl.com	Refinery off-gases	Ethanol

---

The White Dog Lab was established to work on mixotrophic fermentation by utilizing sugars and syngas to produce acetone and isopropanol, which was also exhibited by LanzaTech (2017). Although many companies that ferment syngas are currently operating, their efficiency and profitability are not clear, and more research is required to study the commercialization aspect. To make the business more profitable, new fermentation products can be introduced with higher market prices while proposing cost-effective scale-up technologies. Conclusively, syngas fermentation has high potential with a steadily growing market.

7. Innovative strategies to optimize the syngas fermentation process

With the above-mentioned detailed review of the recent advancements in syngas fermentation technologies for biofuel and biochemical production, it is important to indicate that with the main bottlenecks of syngas fermentation, such as mass transfer and cell factories, multiple advanced synthetic biology techniques can be employed, rather than individual applications, to determine the solutions to the current problems for the production of biofuels and biochemicals, as shown in Fig. 4. The behavior of the cell in the reactor flow field can be analyzed by employing CFD modeling. Multiomics analysis can be used to enhance the mechanism of substrate utilization, product metabolism and metabolic migration in the gas fermentation process. It is expected that if genomic, metagenomic, and metabolic information is widely available, it can allow the development of precise computational algorithms that are useful in predicting the most efficient biosynthetic pathways. Moreover, optimized pathways when combined with the latest experimental tools reduce efforts and facilitate the construction and optimization of metabolic pathways. Reactor optimization should be based on the combination of metabolism and reactor simulation. It is not a traditional optimization method based on trial and error. Hence, reactor optimization, CFD modeling, and other related technologies can be integrated to establish efficient metabolic flow control methods for efficient conversion to ethanol. As these technologies advance, it is expected that the goal of enhancing syngas fermentation will be achieved soon while overcoming the shortcomings of the process.

Fig. 4 Innovative strategies to enhance biofuel and biochemical production from syngas fermentation.

8. Conclusions

Syngas fermentation is a convenient method to convert CO₂, H₂ and CO gaseous streams into fuels and chemicals. However, the main factors that contribute to enhance the fermentation process are the modifications of the bioreactor design for increasing mass transfer rates, enhancing the concentration of microbial cells in the bioreactor, optimization of economical media, genetic engineering of syngas-fermenting microbes with relatively higher rates of gas uptake and product yield, and process control and modeling. Although significant advancements in syngas fermentation have been reported for making it a commercially feasible technology, more research is needed for the advanced development of syngas fermentation technology to improve the productivity of high value-added bioproducts at low costs. If genetically modified microorganisms are developed by employing advanced synthetic biology techniques, they will lead to enhancements in process efficiency and control. Conclusively, syngas fermentation technology has the potential for utilizing abundantly available organic waste to produce biofuels and other value-added chemicals with no adverse environmental impacts; therefore, the development of this technology can significantly contribute to sustainable development.

Abbreviations

CO	Carbon monoxide
CO2	Carbon dioxide
H2	Hydrogen
CAGR	Compound annual growth rate
WLP	Wood–Ljungdahl pathway
ATP	Adenosine triphosphate
CODH	Carbon monoxide dehydrogenase
ACS	Acetyl-CoA synthase
CSTR	Continuous stirred-tank reactor
LCA	Life cycle assessment
CHF	Composite hollow fiber
ANN	Artificial neural network
VFA	Volatile fatty acid
BCR	Bubble column reactor
TBR	Trickle bed reactor
HFMBR	Hollow fiber membrane bioreactor
GLR	Gas lift bioreactor
FBR	Fiber bioreactor
GLMT	Gas–liquid mass transfer
MMC	Mixed microbial consortia
ISO	International Organization for Standardization
CFD	Computational fluid dynamics

Conflicts of interest

The authors declare that they have no conflicts of interest.

Acknowledgements

This project was financially supported by the National Key Research and Development Program of China (2018YFA0901500), Science and Technology Partnership Program, Ministry of Science and Technology of China (KY202001017), and Tianjin Synthetic Biotechnology Innovation Capacity Improvement Project (TSBICIP-IJCP-001-04 and TSBICIP-BRFI-002).

References

1. N. L. Slatter, B. Vichanpol, J. Natakaranakul, K. Wattanavichien, P. Suchamalawong, K. Hashimoto, N. Tsubaki, T. Vitidsant and W. Charusiri, ACS Omega, 2022, 7, 44951–44961.
2. A. Schreiber, A. Peschel, B. Hentschel and P. Zapp, Front. Energy Res., 2020, 8, 1–17.
3. Y. He, C. Kennes and P. N. L. Lens, Chemosphere, 2022, 299, 134425.
4. I. K. Stoll, N. Boukis and J. Sauer, Chem.-Ing.-Tech., 2020, 92, 125–136.
5. F. Regis, A. H. A. Monteverde and D. Fino, Energy, 2023, 274, 127318.
6. V. Ahuja, A. K. Bhatt, B. Ravindran, Y. H. Yang and S. K. Bhatia, Sustain, 2023, 15, 1–21.
7. B. Lagoa-Costa, H. N. Abubackar, M. Fernández-Romasanta, C. Kennes and M. C. Veiga, Bioresour. Technol., 2017, 239, 244–249.
8. T. Janković, A. J. J. Straathof and A. A. Kiss, Sep. Purif. Technol., 2023, 322, 124320.
9. J. Bae, Y. Song, H. Lee, J. Shin, S. Jin, S. Kang and B. K. Cho, Chem. Eng. J., 2022, 428, 131325.
10. S. Shen, G. Wang, M. Zhang, Y. Tang, Y. Gu, W. Jiang, Y. Wang and Y. Zhuang, Bioresour. Bioprocess., 2020, 7, 1–13.
11. M. T. Allaart, M. Diender, D. Z. Sousa and R. Kleerebezem, Microb. Biotechnol., 2023, 16, 697–705.
12. P. Hu, S. Chakraborty, A. Kumar, B. Woolston, H. Liu, D. Emerson and G. Stephanopoulos, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 3773–3778.
13. K. Liu, H. K. Atiyeh, B. S. Stevenson, R. S. Tanner, M. R. Wilkins and R. L. Huhnke, Bioresour. Technol., 2014, 151, 69–77.
14. K. Liu, H. K. Atiyeh, B. S. Stevenson, R. S. Tanner, M. R. Wilkins and R. L. Huhnke, Bioresour. Technol., 2014, 152, 337–346.
15. P. Dürre, FEMS Microbiol. Lett., 2016, 363, 1–7.
16. M. Köpke, C. Mihalcea, F. M. Liew, J. H. Tizard, M. S. Ali, J. J. Conolly, B. Al-Sinawi and S. D. Simpson, Appl. Environ. Microbiol., 2011, 77, 5467–5475.
17. J. R. Phillips, H. K. Atiyeh, R. S. Tanner, J. R. Torres, J. Saxena, M. R. Wilkins and R. L. Huhnke, Bioresour. Technol., 2015, 190, 114–121.
18. A. M. Henstra, J. Sipma, A. Rinzema and A. J. Stams, Curr. Opin. Biotechnol., 2007, 18, 200–206.
19. A. Katsyv and V. Müller, Front. Bioeng. Biotechnol., 2020, 8, 1–23.
20. S. W. Ragsdale and E. Pierce, Biochim. Biophys. Acta - Proteins Proteom., 2008, 1784, 1873–1898.
21. J. Daniell, M. Köpke and S. D. Simpson, Energies, 2012, 5, 5372–5417.
22. F. Fischer, R. Lieske and K. Winzer, Biochem. Z., 1932, 245, 1–12.
23. K. T. Wieringa, Antonie Van Leeuwenhoek, 1940, 6, 251–262.
24. F. E. Fontaine, W. H. Peterson, E. McCOY, M. J. Johnson and G. J. Ritter, J. Bacteriol., 1941, 43, 701–715.
25. G. Pregnon, Characterization of the metabolism of Eubacterium limosum by a quantitative systems biology approach, Université de Toulouse, 2023.
26. A. Mickan, Central dissimilatory pathways of the thermophilic acetogen Thermacetogenium phaeum, Universität Konstanz, 2019.
27. B. R. S. Genthner and M. P. Bryant, Appl. Environ. Microbiol., 1987, 53, 471–476.
28. A. J. Grethlein, R. M. Worden, M. K. Jain and R. Datta, Appl. Biochem. Biotechnol., 1990, 24–25, 875–884.
29. J. S. C. Liou, D. L. Balkwill, G. R. Drake and R. S. Tanner, Int. J. Syst. Evol. Microbiol., 2005, 55, 2085–2091.
30. A. S. Gößner, F. Picardal, R. S. Tanner and H. L. Drake, FEMS Microbiol. Lett., 2008, 287, 236–242.
31. J. L. Gaddy, US Pat., US006136577A, 1996, vol. 136, pp. 423–464.
32. K. Arslan, B. Bayar, H. Nalakath Abubackar, M. C. Veiga and C. Kennes, Bioresour. Technol., 2019, 292, 121941.
33. R. Kerby and J. G. Zeikus, Curr. Microbiol., 1983, 8, 27–30.
34. P. L. Tremblay, D. Höglund, A. Koza, I. Bonde and T. Zhang, Sci. Rep., 2015, 5, 1–11.
35. H. Richter, B. Molitor, M. Diender, D. Z. Sousa and L. T. Angenent, Front. Microbiol., 2016, 7, 1–13.
36. S. Benito-Vaquerizo, M. Diender, I. Parera Olm, V. A. P. Martins dos Santos, P. J. Schaap, D. Z. Sousa and M. Suarez-Diez, Comput. Struct. Biotechnol. J., 2020, 18, 3255–3266.
37. F. Oswald, S. Dörsam, N. Veith, M. Zwick, A. Neumann, K. Ochsenreither and C. Syldatk, Front. Microbiol., 2016, 7, 1–12.
38. S. Flüchter, S. Follonier, B. Schiel-Bengelsdorf, F. R. Bengelsdorf, M. Zinn and P. Dürre, Biomacromolecules, 2019, 20, 3271–3282.
39. R. S. P. de Lemgruber, K. Valgepea, R. Tappel, J. B. Behrendorff, R. W. Palfreyman, M. Plan, M. P. Hodson, S. D. Simpson, L. K. Nielsen, M. Köpke and E. Marcellin, Metab. Eng., 2019, 14–23.
40. S. Joshi and S. D. Mishra, Bioresour. Technol., 2022, 352, 127037.
41. F. Liew, A. M. Henstra, M. Kӧpke, K. Winzer, S. D. Simpson and N. P. Minton, Metab. Eng., 2017, 40, 104–114.
42. S. Dusséaux, W. T. Wajn, Y. Liu, C. Ignea and S. C. Kampranis, Proc. Natl. Acad. Sci. U.S.A., 2020, 117, 31789–31799.
43. G. Bokinsky, P. P. Peralta-Yahya, A. George, B. M. Holmes, E. J. Steen, J. Dietrich, T. S. Lee, D. Tullman-Ercek, C. A. Voigt, B. A. Simmons and J. D. Keasling, Proc. Natl. Acad. Sci. U.S.A., 2011, 108, 19949–19954.
44. J. Zhou, C. Wang, L. Yang, E. S. Choi and S. W. Kim, Enzyme Microb. Technol., 2015, 68, 50–55.
45. J. Abatemarco, A. Hill and H. S. Alper, Biotechnol. J., 2013, 8, 1397–1410.
46. M. Yasin, N. Jang, M. Lee, H. Kang, M. Aslam, A. A. Bazmi and I. S. Chang, Bioresour. Technol. Rep., 2019, 7, 100207.
47. A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney, C. A. Eckert, W. J. Frederick, J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and T. Tschaplinski, Science, 2006, 311, 484–489.
48. P. C. Munasinghe and S. K. Khanal, Bioresour. Technol., 2010, 101, 5013–5022.
49. A. J. Ungerman and T. J. Heindel, Biotechnol. Prog., 2007, 23, 613–620.
50. D. Birch and N. Ahmed, Powder Technol., 1996, 88, 33–38.
51. J. A. Kaster, D. L. Michelsen and W. H. Velander, Appl. Biochem. Biotechnol., 1990, 24–25, 469–484.
52. M. D. Bredwell and R. M. Worden, Biotechnol. Prog., 1998, 14, 31–38.
53. S. Achinas, J. Mulder and G. J. W. Euverink, in Handbook of Biofuels, INC, 2022, pp. 511–527.
54. K. T. Klasson, M. D. Ackerson, E. C. Clausen and J. L. Gaddy, Resour., Conserv. Recycl., 1991, 5, 145–165.
55. M. D. Bredwell, P. Srivastava and R. M. Worden, Biotechnol. Prog., 1999, 15, 834–844.
56. K. C. Lee and B. E. Rittmann, Water Res., 2002, 36, 2040–2052.
57. P. C. Munasinghe and S. K. Khanal, Bioresour. Technol., 2012, 122, 130–136.
58. F. E. Liew, R. Nogle, T. Abdalla, B. J. Rasor, C. Canter, R. O. Jensen, L. Wang, J. Strutz, P. Chirania, S. De Tissera, A. P. Mueller, Z. Ruan, A. Gao, L. Tran, N. L. Engle, J. C. Bromley, J. Daniell, R. Conrado, T. J. Tschaplinski, R. J. Giannone, R. L. Hettich, A. S. Karim, S. D. Simpson, S. D. Brown, C. Leang, M. C. Jewett and M. Köpke, Nat. Biotechnol., 2022, 40, 335–344.
59. V. Singh, S. Haque, R. Niwas, A. Srivastava, M. Pasupuleti and C. K. M. Tripathi, Front. Microbiol., 2017, 7, 1–16.
60. C. F.-B. R. Robles-, M. C. Veiga, C. Kennes and C. Naveira-, Microb. Biotechnol., 2023, 1, 1–16.
61. J. Saxena and R. S. Tanner, World J. Microbiol. Biotechnol., 2012, 28, 1553–1561.
62. H. Im, T. An, R. Kwon, S. Park and Y. K. Kim, Biotechnol. Bioprocess Eng., 2021, 26, 476–482.
63. J. Saxena and R. S. Tanner, J. Ind. Microbiol. Biotechnol., 2011, 38, 513–521.
64. F. Ammam, P. L. Tremblay, D. M. Lizak and T. Zhang, Biotechnol. Biofuels, 2016, 9, 1–10.
65. D. Li, C. Meng, G. Wu, B. Xie, Y. Han, Y. Guo, C. Song, Z. Gao and Z. Huang, J. Ind. Microbiol. Biotechnol., 2018, 45, 61–69.
66. Á. Fernández-Naveira, M. C. Veiga and C. Kennes, J. Chem. Technol. Biotechnol., 2017, 92, 1178–1185.
67. C. Benevenuti, P. Amaral, T. Ferreira and P. Seidl, Reactions, 2021, 2, 391–407.
68. K. Doll, A. Rückel, P. Kämpf, M. Wende and D. Weuster-Botz, Bioprocess Biosyst. Eng., 2018, 41, 1403–1416.
69. C. Benevenuti, M. Branco, M. Do Nascimento-Correa, A. Botelho, T. Ferreira and P. Amaral, Fermentation, 2021, 7, 1–13.
70. D. Xu and R. S. Lewis, Biomass Bioenergy, 2012, 45, 303–310.
71. A. Infantes, M. Kugel, K. Raffelt and A. Neumann, Fermentation, 2020, 6, 1–26.
72. H. N. Abubackar, M. C. Veiga and C. Kennes, Bioresour. Technol., 2012, 114, 518–522.
73. X. Sun, H. K. Atiyeh, R. L. Huhnke and R. S. Tanner, Bioresour. Technol. Rep., 2019, 7, 100279.
74. D. Li, C. Meng, G. Wu, B. Xie, Y. Han, Y. Guo, C. Song, Z. Gao and Z. Huang, J. Ind. Microbiol. Biotechnol., 2018, 45, 61–69.
75. US Pat., US 200702275447A1, 2007, pp. 1–8.
76. Á. Fernández-Naveira, H. N. Abubackar, M. C. Veiga and C. Kennes, Appl. Microbiol. Biotechnol., 2016, 100, 3361–3370.
77. F. M. Pereira, M. M. Alves and D. Z. Sousa, in 13th World Congress on Anaerobic Digestion, 2013, vol. 36, pp. 1–4.
78. F. Oswald, I. K. Stoll, M. Zwick, S. Herbig, J. Sauer, N. Boukis and A. Neumann, Front. Bioeng. Biotechnol., 2018, 6, 1–9.
79. US Pat., US 10,017,789 B2, 2018.
80. H. Richter, B. Molitor, H. Wei, W. Chen, L. Aristilde and L. T. Angenent, Energy Environ. Sci., 2016, 9, 2392–2399.
81. S. Ramió-Pujol, R. Ganigué, L. Bañeras and J. Colprim, Bioresour. Technol., 2015, 192, 296–303.
82. S. Shen, Y. Gu, C. Chai, W. Jiang, Y. Zhuang and Y. Wang, Bioresour. Technol., 2017, 239, 236–243.
83. W. Van Hecke, R. Bockrath and H. De Wever, Bioresour. Technol., 2019, 293, 1–10.
84. US Pat., US009834792B2, 2017, pp. 1–13.
85. M. Farcasiu, M. K. Jain, R. M. Worden and H. E. Grethlein, Bioconversion of Coal-Derived Synthesis Gas to Liquid Fuels, USA, 1995.
86. Y. Shen, R. C. Brown and Z. Wen, Appl. Energy, 2017, 187, 585–594.
87. B. Acharya, P. Roy and A. Dutta, Biofuels, 2014, 5, 551–564.
88. M. Yasin, Y. Jeong, S. Park, J. Jeong, E. Y. Lee, R. W. Lovitt, B. H. Kim, J. Lee and I. S. Chang, Bioresour. Technol., 2015, 177, 361–374.
89. R. Takors, M. Kopf, J. Mampel, W. Bluemke, B. Blombach, B. Eikmanns, F. R. Bengelsdorf, D. Weuster-Botz and P. Dürre, Microb. Biotechnol., 2018, 11, 606–625.
90. V. M. Y. Chisti, Chem.-Ing.-Tech., 1989, 61, 931–932.
91. K. Schügerl, Chem. Ing. Tech., 1980, 52, 951–965.
92. A. Fadavi and Y. Chisti, Chem. Eng. J., 2005, 112, 73–80.
93. K. Asimakopoulos, H. N. Gavala and I. V. Skiadas, Chem. Eng. J., 2018, 348, 732–744.
94. P. H. Lee, S. Q. Ni, S. Y. Chang, S. Sung and S. H. Kim, Chem. Eng. J., 2012, 184, 268–277.
95. US Pat., US008198055B2, 2012, vol. 2, pp. 1–10.
96. Y. Shen, R. Brown and Z. Wen, Biochem. Eng. J., 2014, 85, 21–29.
97. K. Bullis, How Coskata Makes Biofuels A versatile new process for making biofuels could slash their cost, https://www.technologyreview.com/2008/02/19/221678/how-coskata-makes-biofuels/.
98. US Pat., US 8,178,330 B2, 2012, vol. 2, pp. 1–5.
99. A. J. Grethlein, R. M. Worden, M. K. Jain and R. Datta, J. Ferment. Bioeng., 1991, 72, 58–60.
100. H. Richter, M. E. Martin and L. T. Angenent, Energies, 2013, 6, 3987–4000.
101. Y. Cui, K. L. Yang and K. Zhou, Trends Biotechnol., 2021, 39, 914–926.
102. L. Goers, P. Freemont and K. M. Polizzi, J. R. Soc., Interface, 2014, 11, 1–13.
103. M. Diender, A. J. M. Stams and D. Z. Sousa, Biotechnol. Biofuels, 2016, 9, 1–11.
104. H. Huang, C. Chai, N. Li, P. Rowe, N. P. Minton, S. Yang, W. Jiang and Y. Gu, ACS Synth. Biol., 2016, 5, 1355–1361.
105. M. Diender, I. Parera Olm and D. Z. Sousa, Curr. Opin. Biotechnol., 2021, 67, 72–79.
106. J. Almquist, M. Cvijovic, V. Hatzimanikatis, J. Nielsen and M. Jirstrand, Metab. Eng., 2014, 24, 38–60.
107. J. Chen, J. A. Gomez, K. Höffner, P. I. Barton and M. A. Henson, Biotechnol. Biofuels, 2015, 8, 1–12.
108. E. M. De Medeiros, J. A. Posada, H. Noorman and R. M. Filho, Biotechnol. Bioeng., 2019, 116, 2473–2487.
109. J. Chen, J. Daniell, D. Griffin, X. Li and M. A. Henson, Biochem. Eng. J., 2018, 129, 64–73.
110. M. Mohammadi, A. R. Mohamed, G. D. Najafpour, H. Younesi and M. H. Uzir, Sci. World J., 2014, 2014, 1–8.
111. L. Vaquerizo and M. J. Cocero, Comput. Chem. Eng., 2018, 113, 152–161.
112. F. Siebler, Scale-up of gas fermentations - Modelling tools for risk minimisation, Doctor of Engineering thesis, University of Stuttgart, 2020.
113. S. Safarian, R. Unnthorsson and C. Richter, Fermentation, 2020, 6(3), 68.
114. O. Pardo-planas, H. K. Atiyeh, J. R. Phillips, C. P. Aichele and S. Fermentation, Bioresour. Technol., 2017, 245, 925–932.
115. M. Yasin, M. Cha, I. S. Chang, H. K. Atiyeh, P. Munasinghe and S. K. Khanal, in Biomass, Biofuels, Biochemicals: Biofuels: Alternative Feedstocks and Conversion Processes for the Production of Liquid and Gaseous Biofuels, ed. A. Pandey, C. Larroche and S. Ricke, Elsevier, 2nd edn, 2019, pp. 301–327.
116. A. Owoade, A. S. Alshami, D. Levin, S. Onaizi and Z. O. Malaibari, Biofuels, Bioprod. Biorefin., 2023, 10, 25–32.
117. H. L. MacLean and S. Spatari, Environ. Res. Lett., 2009, 4, 1–10.
118. V. Larnaudie, M. D. Ferrari and C. Lareo, Renewable Energy, 2021, 176, 606–616.
119. S. González-García, M. T. Moreira, G. Feijoo and R. J. Murphy, Biomass Bioenergy, 2012, 39, 378–388.
120. R. M. Handler, D. R. Shonnard, E. M. Griffing, A. Lai and I. Palou-Rivera, Ind. Eng. Chem. Res., 2016, 55, 3253–3261.
121. S. Soam, M. Kapoor, R. Kumar, R. P. Gupta, S. K. Puri and S. S. V. Ramakumar, J. Cleaner Prod., 2018, 197, 732–741.
122. A. Zucaro, A. Forte and A. Fierro, J. Cleaner Prod., 2018, 194, 138–149.
123. J. Lask, M. Wagner, L. M. Trindade and I. Lewandowski, GCB Bioenergy, 2019, 11, 269–288.
124. H. Lyu, S. Yang, J. Zhang, Y. Feng and Z. Geng, Bioresour. Technol., 2021, 323, 124586.
125. J. Sadhukhan, E. Martinez-Hernandez, M. A. Amezcua-Allieri, J. Aburto and J. A. Honorato S, Bioresour. Technol. Rep., 2019, 7, 100230.
126. D. Gerrior, K. Delsoz Bahri, A. Kermanshahi-pour, M. J. Eckelman and S. K. Brar, Sustain. Prod. Consum., 2022, 30, 359–376.
127. L. Axelsson, M. Franzén, M. Ostwald, G. Berndes, L. Gopakumar and N. H. Ravindranath, Biofuels, Bioprod. Biorefin., 2012, 6, 246–256.

---