Value-Added Products from Natural Gas Using Fermentation Processes: Products from Natural Gas Using Fermentation Processes, Part 2

Abstract

Methanotrophic bacteria can use methane as their only energy and carbon source, and they can be deployed to manufacture a broad range of value-added materials, from single-cell protein (SCP) for feed and food applications over biopolymers, such as polyhydroxybutyrate (PHB), to value-added building blocks and chemicals. SCP can replace fish meal and soy for fish (aquacultures), chicken, and other feed applications, and also become a replacement for meat after suitable treatment, as a sustainable alternative protein. Polyhydroxyalkanoates (PHA) like PHB are a possible alternative to fossil-based thermoplastics. With ongoing and increasing pressure toward decarbonization in many industries, one can assume that natural gas consumption for combustion will decline. Methanotrophic upgrading of natural gas to valuable products is poised to become a very attractive option for owners of natural gas resources, regardless of whether they are connected to the gas grids. If all required protein, (bio) plastics, and chemicals were made from natural gas, only 7, 12, 16–32%, and in total only 35–51%, respectively, of the annual production volume would be required. Also, that volume of methane could be sourced from renewable resources. Scalability will be the decisive factor in the circular and biobased economy transition, and it is methanotrophic fermentation that can close that gap.

1. Introduction

1.1 Methanol

Today, methanol is mainly produced from synthesis gas. It is a very versatile basic chemical. Processes to further process methanol include the following ones:

• methanol to gasoline (MtG)

• methanol-to-hydrocarbons (MtH)

• methanol to olefins (MtO), and

• methanol to propylene (MtP).

The most used strain to obtain methanol from methane by biochemical conversion is Methylosinus trichosporium. M. trichosporium was found to produce up to 4101 mg methanol/L/day [1]. Methanol is of interest, for example, for fuel cells and combustion, apart from being a basic chemical for synthesis.

In Ref. [2] immobilized Methylocystis bryophila were used to produce methanol by repeated batch fermentation from a simulated biogas mixture.

1.2 Biopolymers

Biopolymers are defined as either being biobased and/or biodegradable according to certain standards. They are used as thermoplastics, elastomers, and thermosets, to replace conventional polymers. Methanotrophs have been described to produce, for instance, PHB [1].

1.2.1 Polyhydroxyalkanoates (PHA) including PHB

The common energy storage compound in microorganisms is glycogen. When a shortage of essential nutrients occurs, particularly nitrogen or phosphorus, several microorganisms are able to store their energic carbon in a different compound, which is polyhydroxybutyrate (PHB). That PHB, which is accumulated intracellularly in granules up to 50% dry cell weight and above, can be extracted and used as a thermoplastic material. PHB is a biopolymer that is both biobased and biodegradable. The biodegradability can occur aerobically or anaerobically, in different environments. This property, together with the property set that is comparable to the commodity polyolefin PP (polypropylene), makes PHB an interesting bioplastics material. PHB belongs to the class of PHA (polyhydroxyalkanoates), which are polyesters and as such are naturally occurring compounds [3]. In plants, only PHB can be found [4], whereas microorganisms can produce a wide variety of PHA copolymers depending on available comonomers.

PHB formation was seen with type II methanotrophs. For instance, Methylocystis parvus OBBP was found to accumulate 60% of PHB on nitrogen limitation (ammonium), as opposed to only 36% accumulation with nitrate limitation. The strain Methylosinus trichosporium IMV3011 could accumulate 47% PHB with both nitrogen sources, nitrate and ammonium. Methylocystis hirsute could accumulate up to 51% PHB on ammonium (again after N-limitation). A total of 51% of PHB accumulation were found with a mixed culture dominated by Methylocystis GB25 on ammonium [1], see Table 1.

Strain	PHB yield [g PHB/g CH4]	Accumulation condition
Methylocystis GB25	0.4	Sulfur deficiency
	0.45	Potassium deficiency
	0.22	Iron deficiency
	0.52	Ammonium deficiency
	0.55	Phosphorus deficiency
	0.37	Magnesium deficiency
Methylocystis parvus OBBP	0.34	Nitrogen limitation
Methylobacterium Organophilum CZ-2	0.43	Nitrogen limitation
Mixed culture dominated by type II	0.4	Nitrogen limitation
Methylocystis sp. WRRC1	0.67	Nitrogen and copper limitation

Table 1.

Yield of PHB found for selected strains of methanotrophs (type II).

Source: [1].

The yield that was obtained varied between 0.22 and 0.67 g of PHB per gram of methane.

EPS production can limit the yield. There were no extracellular products (EPS) under methane-limited growth conditions [5].

In Table 2, several bioreactor configurations to produce PHB were compared.

Bioreactor configuration	pH	Temp. [°C]	CH4:O2 or air	Nitrogen source
5 L batch fermenter	6.8–7.2	30	1:3	Nitrate
2 L batch fermenter	7	30	1:3	Ammonium and nitrate
5 L fed-batch fermenter	7	34	1:1	Ammonium and nitrate
70 L pressure bioreactors	5.7	38	pCH4 = 30%, pO2 = 15%	Ammonium
1 L bubble column bioreactor	7	30	1:1	Ammonium
2.5 L bubble column bioreactor with internal gas recirculation	7.3	25	Polluted air emission (4% CH4); EBRT = 30 min; GRR = 0.5 m^3/(m^3 min)	Nitrate
1.4 L vertical loop bioreactor	7	30	1:1	Ammonium
0.5 L jacketed stirred tank reactor	7.2	25	Gas flow 0.4 L/min; CH4 conc. 2 g/m^3	Nitrate
4 L completely mixed batch reactor	–	–	1:1	Cyclic between ammonium and nitrate
4 L sequencing batch reactor	–	30–32	1:4 (8 h); 1:1 (16h)	Nitrate
15.2 L fluidized bed reactor	6.5–6.9	20–23	1:2.3	Nitrogen gas

Table 2.

How different fermenters fare in PHB accumulation studies.

EBRT = empty bed residence time; GRR = gas recycling rate; Source: [1].

While PHB generally resembles PP in its property set, the material has a low elongation at break and due to its high crystallinity is brittle. Adding a few percent of valeric acid as a comonomer to PHB, yielding PHBV, makes the material softer and thereby more versatile. PHBV can be synthesized by methanotrophs, too. Another important PHA copolymer is PHBH, with a certain content of hexanoic acid being incorporated in the polymer.

Biopolymers can be an environmentally benign material class. For a life cycle assessment (LCA) for biopolymers made from biogas, see Ref. [6].

In Ref. [7], the yields were found to be 1.13 g of PHB per gram of methane for Methylosinus trichosporium OB3b and 0.88 g of PHB per gram of methane for Methylocystis parvus OBBP [7], which is significantly above the figures reported in Table 2, see Table 3.

Parameter		Units	Model value in LCA study [6]	Observed for strain OB3b	Observed for strain OBBP
(1)	Percent PHB	% [g PHB/g dry weight]	50	29	60
(2)	Non-PHB biomass = [100/(1)] – 1	g biomass	1.00	2.45	0.66
(3)	Methanotrophic growth yield	g biomass/g methane	0.345	0.63	0.73
(4)	Methane requirement for the production of non-PHB biomass = (2)/(3)	g methane	2.90	3.89	0.92
(5)	Yield of PHB on methane	g PHB/g methane	0.55	1.13	0.88
(6)	Methane requirement for the production of 1 g of PHB during PHB production phase = 1/(5)	g methane	1.82	0.88	1.13
(7)	Total methane requirement for both phases = (4) + (6)	g methane	4.72	4.77	2.05

Table 3.

Production of PHB from methane.

The numbers are based on 1.0 g of PHB. Source: [8].

Methanotrophic strains producing PHB are listed in Table 4.

Strain	Carbon source	PHA structure	PHA yield	Molecular weight
Methylocystis parvus OBBP	Methane	PHB	0.34 g/g methane	–
Methylocystis parvus OBBP	Methane	PHB	0.88 g/g methane	–
Methylosinus trichosporium OB3b	Methane	PHB	18.1 mg/mg dry cell weight	–
Methylosinus trichosporium OB3b	Methane	PHB	901.8 mg/L	–
Methylosinus trichosporium OB3b	Methane	PHB	1.13 g/g methane	–
Methylosinus trichosporium IMV3011	Methane	PHB	0.6 g/L	1500 kDa
Methylocystis sp. GB 25 DSMZ 7674	Methane	PHB	0.55 g/g methane	Up to 2500 kDa
Methylobacterium organophilum CZ-2	Methane	PHB	233.3 mg/L	–
Methylocystis hirsuta	Methane	PHB and PHV	0.63 g/g methane	–

Table 4.

PHB production by methanotrophs.

Source: [1].

Yield figures have a broad spread. PHB will form copolymers when comonomers, such as valeric acid, are supplied in the medium.

1.2.2 PHBV and other PHA

As stated above, polyhydroxyalkanoates (PHAs) are (bio)polyesters of hydroxy acids. They can be naturally synthesized by bacteria with the purpose of hoarding carbon under N- and P-limitation [9]. PHA consists of 3-, 4-, 5-, and 6-hydroxycarboxylic acids [10]. Lactic acid, citric acid, glycolic acid, malic acid, mandelic acid, and tartaric acid, by contrast, are alpha hydroxy acids (1-hydroxycarboxylic acids). Thereby, their polymers are no PHA. An example of a 2-hydroxycarboxylic acid (beta hydroxy acids, where the acid and hydroxy functional groups are separated by two carbon atoms) is salicylic acid. For hydroxybutyric acid, there are three isomers—alpha, beta, and gamma. When we talk about PHB, we typically mean poly(3-hydroxybutyric acid). Valeric acid (petanoic acid) has four isomers. It is found in some foods.

Polyhydroxyalkanoates (PHA) are established biopolymers. Some well-known brands and producers are given below in alphabetic order:

• AirCarbon®/Newlight Technologies

• EnMat/TianAn

• Minerv-PHA™/Bio-On

• Nodax™/Danimer Scientific

• PHBH™/Kaneka

• TephaFlex®/Tepha [11].

The organization Go!PHA [12] is promoting PHA.

To synthesize the copolymer PHBV, valeric acid is supplied as a comonomer to the fermentation broth.

It was found that above valerate concentrations of 0.7% (by volume), the PHBV accumulation in Methylocystis sp. WRRC1 was inhibited [1]. While the standard content of PHB, without any valerate present, was found to be 30%, it reached 15% only with high valeric acid concentrations. At 0.34% (by volume) of valerate, the PHBV content in the cells reached 60%, with 50% comonomer content [1]. Methylocystis parvus OBBP could be used to make different PHA from various supplied comonomers. 3-hydroxy-butyrate (3HB), butyrate, valerate, hexanoate, and octanoate were added and nitrogen limitation was applied [13]. The products included P(3HB-co-4HB), P(3HB-co-5HV-co-3HV), and P(3HB-co-6HHx-co-4HB). Table 5 shows PHBV production by methanotrophs.

Methanotrophs	Substrate	Co-substrate	PHBV content [%]	HV content [%]	Biomass density [g/L]
Mixed culture dominated by Methylocystis sp.	Methane	Valerate 100 mg/L	44	20	1.2
	Methane	Valerate 400 mg/L	30	39	1
Methylocystis parvus OBBP	Methane	Propionate 100 mg/L	32	8	1.25
	Methane	Valerate 100 mg/L	54	22	1.82
	Methanol	Valerate 100 mg/L	52	22	–
	Formate	Valerate 100 mg/L	58	15	–
Methylosinus trichosporium OB3b	Methane	Valerate 100 mg/L	50	20	1.72
Methylocystis hirsuta	Biogas	Valeric acid 130 mg/L	54	25	1.7
Methylocystis sp. WRRC1	Methane	Valerate 0.34%	60	50	3
	Methane	Valerate 0.34%, and biopolymer accumulation in a copper-free medium	78	58	3

Table 5.

PHBV accumulation from various substrates.

Source: [1].

Figure 1 shows how various PHA can be obtained by methanotrophs through suitable comonomer addition.

Figure 1.

Production of different PHA by comonomer choice. (a) Shows co-substrates without the hydroxy group and (b) With the hydroxyl group. Source: [13].

In Table 6, details on PHA yields from M. parvus OBBP with different comonomers are given.

Co-substrates	wt% PHA polymer	PHA monomer ratio [mol%]					TSS [mg/L]
		3HB	3 HV	4HB	5HB	6HHx
None	42 ± 3	100	0	0	0	0	1600 ± 180
Butyrate (1.2 mM)	55 ± 3	100	0	0	0	0	1660 ± 200
3-Hydroxybutyrate (1.2 mM)	59 ± 5	100	0	0	0	0	1820 ± 220
4-Hydroxybutyrate (1.2 mM)	50 ± 4	91.5	0	9.5	0	0	1720 ± 240
Valerate (1.2 mM)	54 ± 4	75.0	25.0	0	0	0	1760 ± 160
5-Hydroxyvalerate (1.2 mM)	48 ± 4	95.0	1.4	0	3.6	0	1640 ± 180
Hexanoate (1.2 mM)	56 ± 4	100	0	0	0	0	1740 ± 200
6-Hydroxyhexanoate (1.2 mM)	48 ± 3	97.6	0	1.0	0	1.4	1680 ± 220
Octanoate (1.2 mM)	54 ± 3	100	0	0	0	0	1720 ± 180

Table 6.

Synthesis of different PHA through a variation of fatty acid co-substrates with the strain M. parvus OBBP.

TSS = total suspended solids. Source: [13].

Table 7 lists the physical properties of the obtained PHA.

PHA products	Molecular weights		Thermal properties			Mechanical properties
	Mn	Mw/Mn	Tm [°C]	ΔHm [J/g]	Tg [°C]	E [GPa]	σt [MPa]	εb [%]
P3HB	1.48E+06	1.82	178	83	3	3.0	43.2	5.2
P(3HB-co-24 mol% 3HV)	1.32E+06	2.24	147	45	−1	1.0	22.0	50.5
P(3HB-co-3.0 mol% 4HB)	1.33E+06	2.12	148	65	−2	1.2	35.6	176
P(3HB-co-9.5 mol% 4HB)	1.22E+06	2.01	135	47	−5	0.8	31.2	284
P(3HB-co-3.6 mol% 5HV-co-1.4 mol% 3HV)	1.26E+06	2.17	144	44	−2	0.8	29.9	106
P(3HB-co-1.4 mol% 6HHx-co-1.0 mol% 4HB)	1.27E+06	2.11	150	40	−1	0.7	27.6	134
Commercial P3HB	7.38E+05	2.02
Commercial PHBV	4.48E+05	2.18

Table 7.

Properties of the PHA materials derived by methanotrophic fermentation.

M_n = number average molecular weight; M_w = weight average molecular weight; T_m = melting temperature; ΔH_m = apparent heat of fusion; T_g = glass transition temperature; E = Young’s modulus; σ_t = tensile strength; ε_b = elongation at break; Source: [13].

As Table 7 shows, the elongation at break is strongly increased by the comonomers, improving the biopolymer properties compared to heat PHB. Applications of PHA are described in Ref. [14].

1.2.3 Other biopolymers

Several microorganisms, also methanotrophs, produce EPS (extracellular polymeric substances, exopolysaccharides) [15, 16], which might be used potentially for biopolymer applications. A higher oxygen concentration increases the excretion of EPS by methanotrophs [8]. Methylobacter luteus, Methylomonas rubra, Methylococcus thermophilus, and Methylobacter ucrainicus were found to produce EPS from methane in the range of 0.5–0.8 g/L. In general, it is extreme conditions that promote EPS production. M. alcaliphilum 20Z, under moderately saline conditions, gave 1.8 g EPS/g of biomass when grown in a bubble column bioreactor for the treatment of dilute methane emissions [8].

A well-known biopolymer is PLA (polylactic acid). Its monomer, lactic acid, has been obtained from methanotrophs, too, see Table 8.

Product	Strain	Methane	Temperature	Cultivation type	Process details	Titer (productivity)
Lactic acid	Methylomicrobium buryatense 5GB1S pLhldh mutant strain	20% CH4	30°C	Batch	Increased nitrate, phosphate, and trace elements in the medium	0.808 g/L
Lactic acid	Methylomicrobium alcaliphilum 20Z mutant strain	20% CH4 (33% mock biogas in air)	30°C	Continuous	Bubble column bioreactor	0.027 g/g DCW/h
Lactic acid	M. buryatense 5GB1 mutant strain pAMR4	21% CH4	30°C	Batch	Ammonium as N source	0.50 g/L
Crotonic acid	M. buryatense 5GB1C mutant strain pCA09	25% CH4	30°C	Batch	–	0.06 g/L
Butyric acid						0.08 g/L
D-Lactic acid	Methylomonas sp. DH-1, LA-tolerant strain JHM80	20% CH4	30°C	Batch	Increased nitrate	1.19 g/L
Acetic acid	Mixed culture dominated by Candidatus “Methanoperedens nitroreducens”	90% CH4	Room temperature	Batch	Nitrogen limitation	0.097 g/L (1620 μmol/L)
Muconic acid	M. buryatense 5GB1 mutant strain pMUC	20% CH4	30°C	Semi-continuous	CSTR continuous gas supply	0.012 g/L
Succinic acid	Methylomonas sp. DH-1 mutant DS-GL strain	30% CH4	30°C	Batch	pH 6.9	0.195 g/L
3-HP acid (hydroxypropionic)	Methylosinus trichosporium OB3b mutant MCRMP strain	30% CH4	30°C	Batch	pH 7.0	0.061 g/L
4-HB acid (hydroxybutyric)	M. trichosporium OB3b 4HB-SY4 mutant strain	40% CH4	30°C	Batch	–	0.011 g/L (10.5 mg/L)

Table 8.

Organic acids from methanotrophs.

The table gives the titer (g/L). Source: [17].

PLA is biobased and degradable (however, it requires 70°C for disintegration, so “home compost” standards are not met by PLA-based bioplastics products; The material is only “industrially compostable,” see EN 13432 standard). PLA is suitable for food packaging (FDA rating as gras = generally recognized as safe). Today, PLA is made from sugar-derived LA, which requires agricultural starch or sugar production with the associated food/feed competition over land, fertilizer and water consumption, etc. Significant efforts are undertaken to make the enzymatic hydrolysis of lignocellulosic biomass economically viable, however, no breakthrough has been achieved in that field yet (compare 2nd generation biofuel production attempts). Methane-derived PLA can offer a lower environmental footprint and be produced with less price volatility than agriculture-based material. Of PLA, several copolymers and blends exist, e.g. with glycolic acid (PLA + GLA, PLGA), as well as compounds. For LA production by methanotrophs, see [17, 18].

1.3 Other value-added products accessible through methane fermentation

Using natural and engineered strains, not only protein and biopolymers can be obtained from methane, but also several other compounds, for example

• Isoprene [19]

• Squalene [20]

• Succinic acid [21]

• Methanobactin [22]

• Ectoine [23, 24]

• Muconic acid [25]

• Astaxanthin [26]

For a recent review on bioproducts from methane using methanotrophs, see Ref. [17]. A methanotroph-based 2nd generation biorefinery, that is, single-input, multiple-output configuration is treated in Refs. [23, 27]. In general, multi-product biorefineries are better from an economical point of view for methane-based industrial biomanufacturing. For biorefinery downstream processes, see Refs. [28, 29, 30].

1.4 Protein from methane: Single-cell protein (SCP)

Protein is a vital part of our diet. It can be found in plants and meat. For several decades, there has been interest in single-cell protein (SCP), that is, protein derived from yeasts, algae, and bacteria [31, 32]. Yeast cells have been used to make beverages and bread for over 4500 years [33], and also algae have a long history of food. Bacterial SCP is of particularly high interest for commercial production as will be outlined below.

There exit proteinaceous methanotrophs, which contain high amounts of protein made from methane. Bacterial SCP, also called BPM (bacterial protein meal), can be obtained from methane with methanotrophs, in pure and mixed cultures, see Refs. [34]. Continuous aerobic fermentation [35, 36] is generally preferred for higher yield.

A mixed culture with Methylococcus capsulatus (Bath), complemented by Ralstonia sp, Brevibacillus agri, and Aneurinibacillus sp. [37] was described to yield SCP. Also, SCP production from methanol has been reported [38], using methylotrophs.

Applications of SCP are mainly in the feed and food sector. In 1997, the United Nations “Protein-Calorie Advisory Group” discussed the safety of SCP for Animal and Human Feeding [39], apart from other commissions and panels [40, 41, 42].

Twenty-three years earlier, in November 1974, the European Association of Single-Cell Protein Producers (Association européenne des producteurs de protéines unicellulaires) UNICELPE had been founded [43].

SCP produced on a commercial scale has been reported to be deployed in these areas:

• As animal feed, for example, for poultry, calves, mink, and pigs.

• As food additives, for example, vitamin carrier, aroma carrier, and emulsifying agent.

• In industrial processes, for example, as foam-stabilizing agents and in the processing of leather and paper [33, 44, 45].

For food, another application is to boost the nutritional value, for example, of baked food items, ready-made meals, and soups. Single cells have also been used in the food industry as starter cultures (e.g., in bread, beer, and wine making by various yeasts) [33, 44, 45].

SCP has been described as antiobesity food [46]. A patent [47], EP 79.1641475, discloses the use of lipids from methanotrophs for cholesterol reduction. The use of biogas as feedstock for methanotrophs is detailed in Refs. [48, 49]. In Ref. [50], the simultaneous production and use of SCP and PHB bioplastics are discussed.

1.4.1 History of SCP

In Germany during World War I and II, yeast was used to produce food for soldiers, prisoners, and the civilian population [51, 52]. Also, Russia and Japan used similar single-cell proteins for food [51]. Different feedstocks had been tested and used, also fossil-based ones, which gave a negative touch to the products in some peoples’ eyes. Hence, in the year 1966, the term “single-cell protein” (SCP) was introduced by Carl L. Wilson, building upon prior successful productions of protein from fossil sources, to provide a more positive term than “petroprotein.” In fact, in the Soviet Union, early attempts were made to obtain cost-effective protein from oil. “BVK” (belkovo-vitaminny kontsentrat = protein-vitamin concentrate) plants close to oil refineries in Kstovo (year 1973) and Kirishi (year 1974) were built [33]. Till 1989, eight such factories had been constructed by the Soviet Ministry of Microbiological Industry [33].

In Russia, the single-cell protein variants were called “Gaprin” (SCP from methane) [53], “Paprin” (SCP from paraffins), [54, 55] “Meprin” (SCP from methanol), and Eprin (SCP from ethanol). The total capacity was estimated at 1.5 million tons of SCP per year [56].

Work was pioneered by Alfred Champagnat [57, 58].

For petroleum-derived SCP, see [59, 60, 61, 62].

Also in Western Europe, similar attempts were made. British ICI (Imperial Chemical Industries Ltd) succeeded in commercializing Pruteen™ single-cell protein. As feedstock, methanol was used, deploying Methylophilus methylotrophus in a 1500 m³ airlift fermenter [63, 64]. That production was a major milestone in biotechnology [65]. It can be considered the foundation for mycoprotein Quorn™ [66, 67], which is made from a fungus (Fusarium venenatum) from sugar and sold on the order of several 10,000 tons/year today as a meat replacement.

The historic Pruteen™ plant was designed for an output of 50,000–75,000 metric tons/year. The investment was $70 million (1979), and development costs had been $20 million [68].

Scientific American [69] wrote in 1981: “Beer, Wine, bread, and cheese have been made by microorganisms since Neolithic times. To them have been added spirits, yogurt, pickles, sauerkraut, Oriental fermented foods, and today single-cell protein.”

The BP (British Petrol) SCP process for “Toprina” is described in Ref. [70].

French activities from SCP from methane are reviewed in Ref. [71].

German Hoechst/Uhde had developed Probion™ as a single-cell protein from Methylomonas clarae with 70% protein content [56]. A purified version, with 90% protein content with the nucleic acids and fats being removed, was intended for human consumption under the name Probion-S [56].

Liquipron™ (by Liquichimica from Italy) production is detailed in Ref. [72].

Another early, large single-cell protein project was Bioprotein™, which company Norferm (owned 50–50 by Statoil and Anglo-Norwegian pharmaceuticals group Nycomed Amersham [73], later DuPont) developed. Out of Norferm, the companies Calysta and UniBio developed. The Norferm SCP plant can be seen in Figure 2. Bioprotein™ was made from methane.

Figure 2.

The historic Norferm single-cell protein production facility at Tjeldbergodden, Norway. Source: [74].

In the Norferm process, methane (from natural gas), oxygen, ammonia, and minerals were fermented. The concentration of the biomass was 20 g/L, and the temperature was 45°C in a continuous process. By centrifugation, the biomass was concentrated to 80–90 g/L in centrifuges and then further to 220 g/L by ultra-filtration. By a “short intensive heat process, the cells are partially opened to improve the digestibility of the protein.” Spray drying is used to dry the biomass. The specific growth rate was 0.2/h, corresponding to biomass productivity of 4 kg/h/m³ [74].

These numbers are confirmed from additional sources; industrial production of SCP is feasible at an output of 4.0 kg biomass/m³ reactor/h with an expected yield of 0.8 g biomass/g CH₄ [75]. The process is shown in Figure 3.

Figure 3.

Process diagram of the single-cell protein process at Norferm. Source: [74].

The cooling requirement amounted to 62.3 MJ/kg of biomass, and the total volume of the loop reactor measured 300 m³. As a strain, M. capsulatus was used to yield the following product composition:

• Crude protein: 70%

• Fat: 10%

• Fiber: 1%

• Carbohydrates: 12%

• Minerals: 1%

Norferm’s BioProtein^® (BP) SCP had been approved in the EU “as a protein source in animal feeds since 1995, for fattening pigs (8%), calves (8%) and salmon (19-33%)” [76].

Table 9 provides an overview of several other commercially implemented SCP processes from the “golden age” of the industry around the year 1977 when also licensing options were available [78, 79]. An interesting retrospective view is provided by [80].

Organization	Location	Substrate	Organism	Product name	Capacity [tons/year]	Status
British Petroleum (BP)	Lavéra, France	Gas oil	Candida tropicalis	Toprina	16,000	Shut down
British Petroleum (BP)	Grangemouth, Scotland	n-Paraffins	Candida lipolytica	Toprina	4000	Operating
BP-ANIC	Sarroch, Sardinia	n-Paraffins	Candida lipolytica	Toprina	100,000	Constructed, awaiting government approvals
Liquichimica (with Kanegafuchi)	Saline di Montebello, Italy	n-Paraffins	Candida maltosa	Liquipron	100,000	Constructed, awaiting government approvals
USSR	Several sites	n-Paraffins	Candida spp.	BVK	300,000 or more	Operating
Imperial Chemical Industries (ICI)	Teesside, UK	Methanol	Methylophilus methylotropha	Pruteen	50–75,000	Construction approved, start-up late 1979
Amoco	Hutchinson, Minnesota, USA	Ethanol	Tortula yeast (Candida utilis)	Torutein	5–7000	Operating

Table 9.

Commercial-scale SCP processes with hydrocarbons and alcohols and feedstocks.

Source: [77].

Another well-known SCP derived from n-alkanes (paraffins) was Fermosin™ [61].

Table 10 gives additional projects (status 1977).

Organization	Location	Substrate	Organism	Status and capacity [tons/year]
Societé Nationale Sempac: Engineering by Process Engineering Co. (PEC; Member of Chemap, Switzerland)	Algeria	Molasses	Yeast	20,000	Project start December 1977 (Algiers)
	Bulgaria	Methane, methanol, n-paraffins	Depends upon the process chosen	100,000	Discussions with ICI, BP, Shell, USSR, location will depend on the process chosen
Alberta Gas Chemicals with Alberta Research Council, Celanese Canada, with Mitsubishi Gas Chemical	Canada	Methanol, n-paraffins	Yeast	100,000	Study, for a plant in Medicine Hat (Alberta/Canada)
	Czechoslovakia	Ethanol	Yeast	4000 (pilot plant); in planning: 100,000 t/a in Kojetin/Northern Moravia; pilot plant in operation
Joint East Germany/Polish project using USSR/East German-developed process	East Germany	Gas oil	Yeast	60,000 t/a in Schwedt
Groupement Francais des Proteines (50% owned by Institute Francais du Petrole; other 50% shared by Elf-Erap and Cie-Francaise des Petroles)	France	n-Paraffins	Yeast	Pilot plant (Soleize); earlier plans for 100,000 tons/year plant canceled
Institute of Petroleum	India	“crude oil”	Yeast	Pilot plant (Gujarat, near Baroda)
Hebrew University (Yissium Research Devel. Co.) with Dor Chemicals (Haifa)	Israel	Methanol	Bacteria	50 tons/year pilot plant to be scaled up to 1000 and then to 25–100,000 tons/year
Montedison	Italy	Ethanol; carbohydrate	Depends upon the process chosen	Two small pilot plants. Ethanol jointly with Czechoslovakia, carbohydrate jointly with PEC, Switzerland; joint marketing study with Amoco
Societa Italiana Resine S.p.A., Milan (SIR)		Methanol	Candida boidini	Pilot plant (Milan)
Dainippon Ink and Chemicals (with Koa Oil)	Japan	n-Paraffins	Yeast	Foundation laid for 60–120,000 tons/year plant; canceled
Kanegafuchi (with Maruzen)		n-Paraffins	Yeast	60,000 tons/year plant; canceled
Mitsubishi Ga Chemical Co.		Methanol	Yeast, bacteria	4000–5000 semicommercial (Niigata)
Mitsubishi Petrochemical		Ethanol	Yeast
Kyowa Hakko Kogyo		n-Paraffins	Yeast	100,000 tons/year; canceled
Mitsui Toatsu Chemical		n-Paraffins	Yeast	50–60,000 tons/year; canceled
Asahi Chemical		n-Paraffins	Yeast	50–60,000 tons/year; canceled
Sosa Texcoco SA	Mexico	CO2	Spirulina maxima	1 ton/day (Mexico City)
Roniprot L.L.A.; Technology from Japan (reports of negotiations with Ron, Dainippan, Sumito Shoji Kaisha)	Romania	n-Paraffins	Yeast	60,000 tons/year (Arges; Jassyon) said to have started construction in 1973
Petromin (state-owned Saudi Arabian Oil Co.) with British Petroleum	Saudi Arabia	n-Paraffins	Yeast	Study for 100,000 tons/year (Al Jubail)
Instituto de Fermentaciones Industriales, Centro Superior de Investigaciones Cientificas (government sponsored, but autonomous)	Spain	Ethanol	Hansenula anomala	Pilot plant, 100 kg/day (Madrid)
Norsk Hydro and AB Marabou formed “Norprotein” (financial support from Nordic Industrial Fund and Swedish Council for Technical Development)	Sweden	Methanol		Pilot plant planned (Sundbyberg)
Chi Yee Solvent Works, Chinese Petroleum Corp.	Taiwan	Kerosene, fuel oil, gas oil	Pseudomonas
Phillips Petroleum	United States	Methanol	Bacteria	Pilot plant, Bartlesville, Oklahoma
Bio Proteinas de Venezuela (Venproteinas) (with BP), Stone and Webster, contractor	Venezuela	n-Paraffins	Yeast	100,000 tons/year (early 1979) (Puerto la Cruz)
Schick-Chemie-Technik GmbH (Cologne), with technology licensed from the Institute of Industrial Fermentation, Spain (see above)	West Germany	Ethanol	Hansenula anomala	100,000 tons/year (several 50 m^3 fermenters) planned for 1978
Society for Biotechnological Research (Braunschweig Stoeckheim) with Wintershall AG (subsidiary of BASF)		“Oil”	Aerobic mycobacteria	Effluent to be used for secondary oil recovery in a field test in Lower Saxony
Friedrick Uhde GmbH; Hoechst AG (Gelsenberg has withdrawn), with government support		Methanol (also n-paraffins)	Methylomonas sp., Candida lipolytica	Pilot plant planned (earlier plans for demonstration plant (100 tons/year) at Gelsenkirchen-Horst canceled), ultimate plans for 100,000 tons/year
Kohlenstoffbiologische Forschungsstation, E.V., Dortmund		CO2	Scenedesmus acutus	200 tons/year

Table 10.

Additional SCP projects as of 1977.

Source: [77].

The overview above gives testimony of a fairly developed industry, which then disappeared again from the market. Fifty years ago, SCP succeeded technologically, but then failed economically, mainly due to cheap soy prices and increasing fossil fuel costs, as this quotation for the pertinent literature exemplifies:

“By 1974, Shell announced plans for enlarged pilot plant facilities at Sittingbourne, and a development program in Amsterdam, with the ultimate goal of constructing a 100,000 ton/year plant. In the spring of 1976, however, Shell announced that it had stopped work on commercialization, and the expansion and development plans were “indefinitely postponed.” The decision was said to be based on three factors—the low price of soybeans and maize; the potential in large areas of the world for expanding existing sources of protein; and the difficulty of applying Shell’s sophisticated process in “underdeveloped countries” [77].

So Shell bailed out of SCP within a short period of time, after prior heavy investments. That step cannot be considered unusual, compare to the press release from Dow in 2005 [81], when Dow stepped out of the bioplastics business “due to slow sector maturation” after having invested an estimated 750 million USD.

Four seminal books on SCP are [51, 82, 83, 84].

1.4.2 Use of SCP: Feed and food

SCP has multiple possible applications, both in the feed sector and in the food sector [33, 48, 85], see Table 11:

Potential application for SCP in animal diets (feed)	Potential application for SCP in foodstuff
• “Calves having fattening ability.; • Poultry having fattening ability.; • Pigs having fattening ability.; • Fish breeding.; •Laying hens feed.; • Feeding of household animal” [65, 86]	• “As carriers of vitamin.; • As emulsifying agent.; • As carriers of scent.; • In soups.; • Improving the nutritional worth of baked items.; • In readymade meals.; • Within food recipes” [65, 86]

Table 11.

Where SCP could be used.

Source: [33, 48].

Kuzniar et al. have suggested the use of methanotrophic bacterial biomass as a mineral feed ingredient for animals [87].

Also, applications in paper processing, leather processing, and foam stabilization are mentioned in the literature [33, 48].

In Ref. [88], SCP is discussed as a basis for microbial growth media, and in Ref. [89] for wood adhesives.

For (animal) feed applications, the dried bacterial biomass, which typically contains 70% of protein, can be fed directly. In SCP for (human) food applications, the content of nucleic acids (NA) has to be reduced from 10–15 to approximately 1–2%, which can be done thermally or enzymatically. Nucleic acid reduction by heat shocks is described in Ref. [90].

1.4.3 Properties of bacterial SCP

Table 12 summarizes the properties of different SCPs:

Feature	Bacteria	Yeast	Filamentous fungi	Algae
Growth rate	Highest	Quite high	Lower than bacteria and yeast	Low
Substrate	A wide range of substrates	Most substrates except hydrocarbons and CO2	Limited substrates (mostly starchy and cellulosic materials)	Light, inorganic carbon sources, for example, CO2 (preferably)
pH range	5–7	5–7	3–8	up to 2
Cultivation system	Bioreactors	Bioreactors	Bioreactors	Open ponds, tanks in sunlight
Risk of contamination	High; precautions are necessary	Low	Low if grown below pH 5	High
Biomass recovery	Sometimes problematic; new improved methods are needed	Easy by centrifugation	Easy for filamentous or pellet forms	Difficult and costly with unicellular algae
Protein content (crude)	80% or more	55–60%	50–55%	Up to 60%
Amino acid profile	Generally good, a small deficit in S-containing acids	Generally good, deficit in S-containing amino-acids	Low in S-containing amino acids	Generally good; low in S-containing amino acids
Nucleic acid content	High (8–14%)	High (5–12%)	High (3–10%)	Low (4–6%)
Removal of nucleic acids	Necessary
Toxins	Gram-negative bacteria may produce endotoxins	–	Many species produce mycotoxins	Three types of toxins: endotoxin, neurotoxins, heptotoxins
Other features	–	High vitamin B content	Chitin may contain a significant proportion of N content, which is unavailable	Low yield (1–2 g dry wt/L). High chlorophyll content is unsuitable for humans

Table 12.

Comparison of some cultural and biochemical characteristics of various microbe groups to make SCP.

Source: [91].

Bacteria are very well suited to make SCP, see Table 5. They exhibit the highest growth rates. On the other hand, the nucleic acid content is elevated compared to SCP by other microorganisms. The extraordinary growth performance of bacteria is also visible in Table 13.

Organism	Mass doubling time
Bacteria	10–120 min
Algae/molds	2–6 h
Yeasts	10–120 min
Plants	1–2 weeks
Chickens	2–4 weeks
Pigs	4–6 weeks

Table 13.

Mass doubling time of different organisms.

Source: [91].

With doubling times of the mass on the order of 10–100 minutes, bacteria and yeasts grow incomparably faster than plants or animals. This translates into unsurpassed productivity, as Table 14 illustrates.

Organism (1000 kg)	Amount of protein produced in 24 h [kg]
Beef cattle	10^−1
Soybeans	10^1
Yeast	10^2
Bacteria	10^12

Table 14.

Comparison of protein production efficiency of selected organisms over 24 h.

Source: [91].

Within 24 h, a starting mass of bacteria of 1000 kg can yield, theoretically, 10¹² kg of protein, whereas beef would only produce 0.1 kg and soy in the order of 10 kg. An analysis of bacterial meal (BM) derived from methane is given in Table 15.

Feedstock	Bacterial culture	Crude protein	Lipids	Ash	Nucleic acids
Methanol	Methylophilus methylotrophus	81.3	7.2	9.1	15.9
Methane	Methylococcus capsulatus, Ralstonia sp., Brevibacillus agri, Aneurinibacillus sp.	73.2	10.7	8.5	9.9
		68.1	10.4	8.0	Not determined
		68.7	8.0	8.0	Not determined
		73.4	8.4	7.7	Not determined
		71.9	8.3	6.7	Not determined
		69.5	8.1	6.2	11.1
		67.0	9.9	6.4	Not determined

Table 15.

Properties of BM (bacterial meal) made from methane and methanol, based on g/100g of dry mass.

Source: [65].

As one can infer from Table 15, the protein content approaches and exceeds 70%. Table 16 shows the protein content of SCP from selected microorganisms.

Bacteria	Substrate	SCP (%)	Reference
Haloarcula sp. IRU1	Petrochemical wastewater	76	[92]
Methylococcus capsulatus, Ralstonia sp., Brevibacillus agri	Methane (natural gas)	67–73	[65]
Aneurinibacillus sp., Methylomonas and Methylophilus spp., Metilococ capsulatus	Gas and liquid products of sewage	<41	[93]
Methylococcus capsulatus (bath)	Methane	53	[94]
Methylophilus sp	Supernatant and biogas	24	[93]
Methylocapsa acidiphila	Methane	59	[95]
Methilomonas sp.	Natural gas	69.3	[96]
Methilomonas sp.	Biogas and supernatant of sewage sludge	56	[93]

Table 16.

Protein content of bacteria expressed as SCP (single-cell protein) grown on different substrates.

Source: Excerpt from [44].

Another important aspect of SCP is its quality, which can be expressed by the amino acid profile. In Table 17, the composition of amino acids of bacterial SCP compared to other proteins is shown.

Amino acids		BM	Soybean meal	Fishmeal	Methanol-grown bacterial protein (Pruteen™)
Indispensible amino acids	Arginine	6.3	7.4	6.2	4.6
	Histidine	2.2	2.7	2.5	1.9
	Isoleucine	4.4	4.7	4.7	4.3
	Leucine	7.5	7.5	7.9	7.0
	Lysine	5.6	6.1	8.2	6.0
	Methionine	2.6	1.3	3.0	2.4
	Phenylalanine	4.2	5.0	4.1	4.1
	Threonine	4.3	3.9	4.0	4.6
	Tryptophan	2.2	1.4	0.9	0.9
	Valine	5.8	4.8	5.3	5.6
Dispensible amino acids	Alanine	7.1	4.2	6.1	7.1
	Apartic acid	8.5	11.2	9.9	8.8
	Cysteine + cystine	0.7	1.5	0.9	0.7
	Glutamic acid	10.6	18.2	12.6	10.6
	Glycine	4.9	4.2	6.0	5.7
	Proline	3.8	5.0	4.3	2.9
	Serine	3.6	5.2	4.1	3.3
	Tyrosine	3.6	3.8	3.2	3.4

Table 17.

Amino acid profile of SCP made from natural gas (BM = bacterial meal).

For reference, the data for soybean meal and fishmeal are provided. The rightmost SCP was obtained from methanol, Pruteen™, by Methylophilus methylotrophus [g/16 g N ± standard deviation of free, hydrated amino acids]. Source: [65].

Bacterial SCP is approved as a feed ingredient in the EU feed catalog (EU 68/2013) [63]. SCP for feed has been tested extensively, see Table 18 for an overview.

SCP-protein evaluations by animal tests
Organism	Digestability	Protein efficiency ratio	Biological value
Algae	65–86	0.7–2.6	48–81
Bacteria	80–90		73–82
Fungi			10–75
Yeast	81–96	0.9–1.7	32–69
Yeast and methionine	96	2.0–2.3	91–96
FAO/WHO reference protein	100		100

Table 18.

As early as 1981, SCP was tested for feed.

Source: [97].

SCP feed trials were made with Drosophila [87] several monogastric species [98], including rats [99, 100], pigs [61, 100, 101, 102], dogs [45, 103], (lactating) cows [104], veal calves [105], chickens (broilers) [72], mink (Mustela vison), fox (Alopex lagopus), Atlantic salmon (Salmo salar), rainbow trout (Oncorhynchus mykiss) [106], Pacific White Shrimp (Penaeus vannamei) [107, 108], Atlantic halibut (Hippoglossus hippoglossus), tilapia (Oreochromis niloticus, Oreochromis mossambicus) [109, 110], Japanese yellowtail (Seriola quinqueradiata) [111], zebrafish (Danio rerio) [112] and other aquaculture species [113, 114]. In feed, SCP can replace fish meal and soy. It was found in Ref. [65] that “bacterial meal (BM) derived from natural gas fermentation, utilizing a bacteria culture containing mainly the methanotroph Methylococcus capsulatus (Bath), is a promising source of protein based on criteria, such as amino acid composition, digestibility, and animal performance and health.” [65].

The oral immunogenicity of bacterial SCP was tested by [115].

The use of SCP for feed and food was also proposed by John H. Litchfield in 1979 [116].

An example from Germany (1943–1949) is the “wood sausage,” a spread made in Wildshausen from paper production waste (sugars) using the fungus Oidium laktis. The process was licensed from Biosyn GmbH [117]. Other yeast spreads on the market are Marmite, Cenovis, and Vegemite: “Spent brewer’s yeast (Saccharomyces cerevisiae) have been sold for more than a century in yeast extracts such as Marmite® (Unilever and Sanitarium Health Food), Vegemite® (Bega Cheese Ltd.), Cenovis® (Gustav Gerig AG), and Vitam-R® (VITAM Hefe-Produkt GmbH).” [117]

“Another commercially available yeast, Torula (Candida utilis, renamed as Pichia jadinii), a widely used flavoring agent, is also high in protein. Torula was used in Provesteen^® T, produced by the Provesta Corporation in the 1980s, along with similar products using Pichia and Kluyveromyces yeast)” [118].

1.4.4 Aquatic species

Approximately half of the global fish and aquatic species production comes from aquaculture operations, with an increasing share. Fish meal is volatile in price and, due to overfishing, is not a sustainable feed material. The same holds true for soy; while soy is cheap, its immense monoculture production has resulted in rainforest destruction on a global level. It was shown that soybean meal-induced enteritis in Atlantic salmon (Salmo salar) can be prevented by cell wall fractions obtained from Methylococcus capsulatus [119].

1.4.5 Chicken

Various tests were carried out to feed chickens with bacterial SCP. SCP could maintain chicken growth performance and increase the feed to gain (feed conversion) and reduce the fat content in chicken [120]. “Replacing soybean meal and small amounts of rendered fat with BPM caused less intensity of odor and less rancid flavor of chicken meat” [120].

Processed animal protein (PAP) from poultry/pig for poultry/pig [121] is currently being investigated, too, but brings back memories about BSE [122]. Also, potential quantities are limited.

1.4.6 Mink

Another species where SCP was tested successfully is mink (Mustela vison). SCP digestibility tests were carried out in 2009 by the Department of Animal and Aquacultural Sciences at The Norwegian University of Life Sciences, Ås [123]. Mink are of importance for the fur industry.

1.5 Status of single-cell protein as food

Oral delivery of bacteria is not novel [124]; for an overview of beneficial microorganisms in food and nutraceuticals, see Ref. [125]. Microorganisms in food are covered in Refs. [126, 127, 128].

In the literature, there is ample coverage of food trials of SCP in the last 50 years.

One example is a trial from 1979 to include SCP “Pekilo” in sausages and meat balls [129].

SCP has a peculiar taste. Some variants, which have a high content of glutamic acid, have been proposed and used instead of monosodium glutamate (MSG) [118] as the flavor enhancer in food.

Pekilo is a microfungus-derived SCP from Paecilomyces varioti based on pulp and paper waste streams. It is now being revived by Enifer [130]. Another work with SCP from waste is discussed in Ref. [131], SCP with the fungus P. chrysosporium in [132].

As early as 1971, single-cell protein had been subjected to clinical testing in tolerance trials in adults [133]. For immunogenicity testing of food proteins, see Ref. [134].

For humans, it is recommended that the daily nucleic acid (NA) intake does not exceed 2 g, as higher quantities (particularly RNA and, to a lesser extent, DNA) have been reported to increase the content of uric acid in blood plasma to an unhealthy level [135], leading to gout [136]. As SCP contains a comparatively large amount of nucleic acids, it needs special processing for food applications. For feed, by contrast, the nucleic acid species do not harm. In Table 19, approaches for RNA reduction in SCP are summarized.

Method	Protocol	Organism	Initial and final % RNA (of dry wt)
Base-catalyzed hydrolysis	0.12 N NaOH for 30 min at 50°C	P. variotii	9%→3%
Chemical extraction	5% NaCl at 120°C	Baker’s yeast	Complete removal
Cell disruption; physical separation:	Disintegration-glass beads	Different microorganisms	Final % RNA = 2
(a) Enzymatic treatment	Malt sprout nuclease	Yeast	Final % RNA < 3
(b) Chemical	Succinic anhydride, pH 4.2–4.4	S. carlsbergensis	6–8%→1.8%
(c) Chromatography	Cellex E-Ecteola cellulose	C. utilis, S. cerevisiae, S. carlsbergensis, Z. mobilis	Less than 3% of initial content
Exogenous RNAase	Bovine pancreatic ribonuclease, 55–56°C, pH 6.7–8.0	S. utilis, C. intermedia	Final % RNA = 1.5
Endogenous RNAase	90°C, pH 2, 20 min	R. glutinis	6.5%→1.2%
-heat shock	45°C, pH 5.8–6.9, 2 h	C. utilis	6–8%→2%
	68°C for a few sec, 50°C, 1 h, 55°C, 1 h	C. utilis	7–8%→1–2%
	50–60°C + NaCl	S. cerevisiae	7–8%→1.4%

Table 19.

RNA reduction in SCP.

Source: [51].

Tables 20–23 compare different treatments of SCP to reduce nucleic acids [137].

Amino acids	Yeast cells [mg/g]	Protein extracted with NaOH [mg/g]
Cystine + cysteine	8.15	11.88
Lysine + histidine	10.81	16.55
Aspargagine	7.58	15.5
Aspartic acid	1.89	3.54
Serine + glycine	7.82	8.86
Glutamic acid	4.11	5.72
Threonine	26.11	33.1
Alanine	29.94	32.79
Tyrosine	11.89	14.77
Methionine + tryptophan	6.94	11.96
Phenylalanine	7.14	9.45
Leucine + isoleucine	22.85	37.28

Table 20.

Change of the amino acid composition of yeast cells by NaOH extraction.

Source: [137].

NaCl [%]	RNA	Reduction [%]
1	4.54	6.4
2	4.05	16.5
3	3.68	24.1
4	3.19	34.2
5	4.05	16.5
Control	4.85	0.0

Table 21.

Reduction of RNA in yeast cells (Candida lipolytica YB-423) by NaCl treatment, incubation at 50°C for 20 min.

Source: [137].

Type of protein	Total nitrogen [%]	Total protein [%]	RNA [%]	DNA [%]
Yeast cells	6.72	42.00	4.90	2.13
Protein extract	10.80	67.50	1.22	0.399

Table 22.

Nucleic acid and protein content of yeast cells by extraction with NaOH.

Source: [137].

Temperature [°C]	Time [s]	RNA [%]	DNA [%]	Reduction of nucleic acids [%]
60	5	3.325	1.460	31.93
	10	3-150	1.375	35.63
	20	2.712	1.350	42.22
	40	2.625	1.320	43.88
	80	2.450	1.300	50.92
70	5	2.187	1.250	51.11
	10	1.662	1.240	58.72
	20	1.575	1.230	60.10
	40	1.400	1.200	63.02
	80	1.000	1.150	69.42
Untreated cells		4.900	2.130	0.0

Table 23.

RNA and DNA in yeast cells (C. lipolytica YB-423) treated by heat shock.

Source: [137].

There is currently a strong movement toward alternative protein (alt protein), as concerns over the healthiness and sustainability of meat, first and foremost imported beef, are increasing all over the world. Hence there is good potential for SCP for food applications. By the end of 2021, Solein™ submitted its dossier for SCP food approval in the EU [138].

Occupational health aspects of SCP are discussed in Ref. [139], where it was found that inhalable dust needs to be avoided.

2. Operational aspects of methane fermentation

Methane fermentation occurs naturally and is carried out in bioreactors. Medium optimization and mineral requirement are treated in Refs. [140, 141, 142].

The reactions can be described as follows for methane limitation 4.9 g is for bacteria and products together [5]:

$$\underset{16\mathrm{g}}{{\mathrm{CH}}_4}+\underset{56\mathrm{g}}{1.8{\mathrm{O}}_2}\to \underset{16.1\mathrm{g}\mathrm{dry}\mathrm{wt}.}{\text{Bacteria}}+\underset{15.8\mathrm{g}}{0.34{\mathrm{CO}}_2}$$E1

Under oxygen limitation, it was found [5]:

$$\underset{16\mathrm{g}}{{\mathrm{CH}}_4}+\underset{9.6\mathrm{g}}{0.3{\mathrm{O}}_2}\to \underset{4.9\mathrm{g}\mathrm{dry}\mathrm{wt}.}{\text{Bacteria}}+\text{Products}+\underset{2.2\mathrm{g}}{0.05{\mathrm{CO}}_2}$$E2

By contrast, [96] found for methane limitation:

$$4.25\mathrm{g}{\mathrm{O}}_2+1.245\mathrm{g}{\mathrm{CH}}_4\to 1\mathrm{g}\text{cells}+2.0\mathrm{g}{\mathrm{CO}}_2$$E3

For oxygen limitation, this formula could be determined [72]:

$$2.5\mathrm{g}{\mathrm{O}}_2+1.01\mathrm{g}{\mathrm{CH}}_4\to 1\mathrm{g}\text{cells}+1.216\mathrm{g}{\mathrm{CO}}_2$$E4

For fermentation technology in general, see Refs. [143, 144, 145, 146], for its economics [147].

2.1 SCP production process

As SCP has already been produced at a large scale some 50 years ago, see above, there is ample experience with gas fermentation. SCP production is summarized in Table 24.

Cultivation operation	Growth modality	Capital and operational considerations	Emerging commercial examples
Aerobic bioreactor	Heterotrophs; Mixotrophs	High cell mass yield; High capital costs; High energy consumption; Sterile operation; Significant installed industrial capacity	Methanol, glycerol, or ethanol—KnipBio; Glucose—Veramis; Cellulose—Arbiom, Menon, EniferBio
Anaerobic bioreactor	Heterotrophs; Mixotrophs	Low cell mass yield; Low capital costs; Low energy consumption; Requires metabolite production and valorization; Non-sterile operation; Most installed industrial capacity	Glucose or glycerol—White Dog Labs; Yeast separation—Fluid Quip Technologies & ICM
Gas bioreactor	Methylotrophs; Chemoautotrophs; Mixotrophs	Variable cell mass yield; High capital costs; High energy consumption; Could require metabolite production and valorization; Sterile & non-sterile operation; Limited installed industrial capacity	Methane—Calysta, Unibio, String Bio & Circe Biotechnologie; CO2, H2, and O2—Kiverdi, Novo Nutrients, Deep Branch Biotechnology, Solar Foods, Avecom & LanzaTech; Glucose & syngas—White Dog Labs
Photosynthetic bioreactor	Photoautotrophs; Mixotrophs	High cell mass yield; Low capital costs; High energy consumption; Sterile operation; No known installed capacity	CO2 and light—Bioprocess Algae & Pond Technologies
Open cultivation systems	Photoautotrophs; Heterotrophs; Mixotrophs	Variable cell mass yield; Low capital costs; Low energy consumption; Non-sterile operation; Limited installed industrial capacity	Brewing by-products—iCell Sustainable Nutrition; Open photosynthetic system—Cellana

Table 24.

Production of SCP.

Source: [148], extended from Refs. [149, 150].

As Table 24 shows, different approaches are being followed. For a review on SCP, see Refs. [85, 151].

A major difficulty with methane is its low solubility in water, resulting in low productivity. Paraffin oil as a “methane vector” for improved mass transfer and higher cell densities was suggested in the literature [64]. Koutinas et al. have proposed to use of γ-alumina pellets to improve methane fermentation [152]. Emulsion-based fermentation to enhance the mass transfer of methane is presented in Ref. [153].

For the bioprocess, Table 25 contrasts two reactor configurations, the “classic” CSTR (continuously stirred tank reactor) and the loop reactor.

Conventional stirred vessels	Modern loop reactors
High energy costs	Low energy costs
10 kW/m^3 (approx.)	2 kW/m^3 (approx.)
Less defined fluid flow pattern	Quantified fluid flow pattern
Intrafermentor cooling (jacket)	Extrafermentor cooling (plate exchanger)
Complicated heat removal	Easy removal of the fermentation heat load
Limited scale-up	Less limited scale-up
Indefinite mixing—irregular residence times	More definite mixing—regular residence times
Cell yield on oxygen of 10–20%, hence requiring a high aeration rate	Cell yield on oxygen of 40–50%, hence requiring less aeration
Batch and continuous operation	Continuous operation
Low yield and productivity	High yield and productivity
Limited control	Efficient control
Moving parts (easily contaminated)	No moving parts
Expensive capital and running costs	Cheaper to install and run

Table 25.

Fermenter comparison for SCP production.

Source: [51].

For methane fermentation in a sequencing batch bioreactor, see Ref. [154], for batch fermentation, see Ref. [155], and for a cascade of fermenters, see Ref. [156].

The historic Norferm process (see above) used a loop reactor, which can be seen as a PFR (plug flow reactor) comparable to an airlift fermenter only with a different agitation mode of the fluid, using a pump instead of the injected gasses to circulate the medium. Table 26 assesses the relative production costs of different SCP processes.

Capital and manufacturing cost [%] for several SCP processes
	Relative costs (units) for:
	Methanol	Ethanol	Methane	n-Paraffins
Total capital cost	100	97.6	150.8	107.7
Total manufacturing cost	100	229.6	75.2	129.7
Comparative production costs of SCP
Process	Relative production costs [%] for:
	Raw materials
	Substrate	Other chemicals	Utilities	Labor	Miscellaneous	Total
Yeast—paraffin	29.4	33.9	23.8	8.4	14.5	100
Bacterium—methanol	47.4	26.4	14.2	6.2	5.8	100
Yeast—ethanol	63.9	9.9	12.0	5.1	9.1	100
Fungus—sulfite liquor	17.0	33.7	24.8	11.0	13.5	100
Bacterium—bagasse	25.7	13.5	36.6	8.3	15.9	100
Algae—CO2	?	16.5	13.8	15.3	54.4	-
Fungus—lignocellulosic wastes	?	14.5	11.3	22.9	51.3	-

Table 26.

Comparative costs of various SCP processes.

Source: [51].

It is estimated that methane fermentation comes with the highest investment costs (CAPEX), but will allow the lowest operational costs (OPEX), making that feedstock attractive for large-scale operations. When we look at the relative cots for substrate and utilities, there are also marked differences in the various processes. A waste stream has the advantage of low costs, but stable quality and quantity have to be ensured. Utility costs will depend on the geographic location of the site, apart from the unit operations chosen. Cooling costs are determined by the process temperature, and labor costs can be controlled by the degree of plant automation and the complexity of the process. Costs in SCP production are further discussed in Ref. [157].

For engineering factors in the production of SCP, see Refs. [158, 159]. For process development, see Refs. [160].

Downstream processing depends on the exact target product(s). The “classic” separation of cells from the fermentation medium is centrifugation, followed by spray drying. Imasaka et al. have suggested cross-flow filtration of the fermentation broth with ceramic membranes [161], and Yang et al. the continuous methane fermentation in a fixed-bed reactor packed with loofah [69]. Heat treatment or enzymatic treatment [132, 162] can be used for nucleic acid reduction and digestability improvements.

2.2 PHB production process

In Ref. [163], the authors carried out a techno-economic assessment of methanotrophic PHB manufacturing, compare Figure 4.

Figure 4.

PFD (process flow diagram) to convert methane into PHB, with mass balance. Source: [163].

The key findings from Ref. [163] for a PHB production on the order of 100,000 t/a by methanotrophic fermentation were:

• Downstream process: acetone–water solvent extraction

• Production costs: $4.1–$6.8/kg PHA

• Raw material costs reduction compared to sugar feedstock: 22% instead of 30–50%

• Cooling costs: 28% of the operational costs

• Strong advantages of using thermophilic methanotrophs—production costs go down to $3.2–5.4/kg PHA [163], because of cheaper cooling (cooling water instead of a refrigerant)

These figures of 3.2–6.8$/kg of PHB are very promising, as today PHA suffers from high production costs and hence a limited market.

3. Discussion

Today, we are in an “oil economy,” as crude oil is the basis for a major share of energy and materials production. With an accelerating shift toward a circular economy and renewable resources, a paradigm change is about to happen. Biorefineries and biobased products are seeing strong interest from various stakeholders, as do renewable energies, such as wind and solar.

However, when a realistic view is applied, one will quickly see that decarbonization is not that simple. Biomass and hydrogen will play an important role, for sure; the major issue with stepping out of oil is the sheer size of the industry. When one wants to replace the feedstock for 400 million tons of polymers per year, and for hundreds of millions of tons of base chemicals, agricultural resources are simply not sufficiently available, at least not without creating serious disruptions in feed and food production. There is not only a distribution problem of feed and food but a more fundamental land scarcity issue. We are simply not able to convert all “unmanaged” land into fields and pastures to cater to the world population not only for food but also for materials. Neither can be the productivity of the existing land be pushed upwards indefinitely; for fast, secure, and reliable scale-up of SCP, biopolymer, and other materials produced from non-agricultural and non-oil sources, methane from natural gas seems to be the one option.

Based on the circular economy concept, deriving nutrients from (bio)waste is a sustainable approach. For an analysis of SCP made from biowaste as a feed additive using, see Refs. [164, 165].

3.1 SCP

In Table 27, the market of alternative protein sources is summarized.

Protein source		Production volume [Mton DM/y]	Farm gate price [$/kg DM)	Average protein content [% DW]	Price per unit protein [$/kg protein DM]
Animal	Fish	66.7	2.07	15–20	10–14
	Pork	108.5	1.54	20	7.7
	Chicken	92.7	1.43	31	4.6
	Beef	62.7	2.70	25	10.8
Vegetable	Soybean	320.2	0.37	35	1.1
	Wheat	712.7	0.19	12	1.6

Table 27.

Different animal and vegetable protein sources compared by their production volumes and prices.

DM = dry matter. Reproduced from [166].

As Table 27 exemplifies, the “farm gate” price of vegetable protein is rather low.

Production volumes of alternative protein from microbes are tabulated (Table 28).

Organisms	Production volume [ton DM/y]	Production costs [€/kg DM]	Global market value [Billion €]	Yearly growth [% per year]	Remarks
Yeast	3,000,000	–	9.2	7.9	Mostly commercialized as baker’s yeast and for ethanol fermentation. Global market value projected to 2019
Algae (microalgae)	9000	4–25	2.4	10	Besides feed and food, derivatives are also used
Mycoprotein (Quorn™)	25,000	–	0.214	–	Investments for a plant of 22,000 tons per year were done in 2015
Bacteria (Profloc™)	5000	1–1.1	–	–	No commercial production yet
Bacteria (FeedKind™)	80,000	–	–	–	No commercial production yet
Valpromic	5000	–	–	–	No commercial production yet

Table 28.

Current status of different microbial proteins based on their market size and production volumes.

Source: [166].

Today, yeast is undoubtedly the largest volume SCP source, also for food applications.

In Ref. [167], SCP production by the yeast Kluyveromyces marxianus var marxianus is described.

The FeedKind™ plant is currently under construction [168], Profloc™ went out of business.

When we assume a protein demand per person of 70 g per day of SCP, with a world population of 7.9 billion people, the theoretical market potential for bacterial SCP would be 0.5 million tons per day or 200 million tons per year. At a conversion ratio of 1 g CH₄ to 1 g of SCP, we arrive at ∼288 million m³ of methane, which is approximately 7% of today’s natural gas consumption. So if we were to provide all protein for humanity by bacterial SCP, only a fraction of the natural gas stream would be required.

Bacterial single-cell protein has also been envisioned as a possible protein source in a global food catastrophe, where agricultural protein production is suddenly impaired, as elaborated by Juan B. Garcia Martinez [169, 170].

3.2 Bioplastics

Today, bioplastics have a market share of 1–2% of conventional plastics materials. It is estimated that bioplastics could replace 90% of petrochemical plastics, particularly in standard application like packaging (only for high-performance materials, such as PEEK or PFTE, no suitable bioplastics counterparts is yet known to exist). Table 29 takes a look at which bioplastics could replace the most common petrochemical plastics. For instance, LDPE could be replaced to some extent by a “drop in” material of similar property set (bio-PE), and by biopolymers with different characteristics, such as PBAT, PBS, and PHA, to the other part.

Non-biodegradable				Biodegradable
Petrochemical Plastics	Bioplastics (drop-in, partly biobased)			Petrochemical Plastics		Bioplastics
			Bio-PTT	PBAT (can be partly biobased)	PBS	PHA	PLA	TPS	Cellulose-based
LDPE	bio-PE	55		10		15	10	5	5
PP	bio-PP	10	5		10	20	20	15	20
HDPE	bio-PE	50		10		15	10	10	5
PET	bio-PET	60	10			5	20		5
PS						20	30	25	25
PVC	bio-PVC	50				20			30
EPS								70	30
PA	bio-PA	80
PUR	bio-PUR	80	20			10		10
Other thermo-plastics				10	10	20	20	20	20
Other plastics				10	10	20	20	20	20

Table 29.

Overview of the technical substitution potential of regular plastics (left columns) by bioplastics.

Numbers are in %. Source: [86].

As Table 29 shows, a handful of “drop-in” and degradable bioplastics can replace the most common petrochemical plastics. Overall, it is estimated that up to 90% of conventional polymers can be replaced by biopolymers. The benefits of such a replacement are depicted in Figure 5.

Figure 5.

GWP (global warming potential), land use, and water use of petrochemical and bioplastics packaging materials. Source: [86].

For simplification, the land use of petroplastics is set to zero, as it is negligible. Also, as Figure 7 shows, the water use of petroplastics is low. The error bars indicate the possible range of the figures. According to IfBB [171], the footprint of bioplastics is considerable. For instance, 1 ton of PHB requires 2.86 tons of sugar (glucose) or 3.24 tons of starch for its production. One ton of PLA requires 1.47 tons of sugar or 1.67 tons of starch. Yields of crops differ, for example, 10.03 tons of sugar/ha for sugar beet and 0.83 tons of starch/ha for wheat, resulting in specific land requirements for the materials’ feedstocks, see Figure 6.

Figure 6.

The footprint of the two bioplastics PHB and PLA. Source: [171].

PLA today is the most important bioplastics material, and there is a shortage of supply in the market, leading to a surge in prices. The strong growth is expected to continue. Major producers of PLA are Total/Corbion (Purac™) and Cargill/Natureworks (Ingeo™). The PHB market cannot be considered mature, as the volume is still minute, but there are several established players, see Table 30.

Current industrial production of polyhydroxyalkanoates (PHA)
Company name	Carbon substrate	Product name	Production [t/a]
Danimer Scientific (formerly Meredian Holdings Group Inc./MHG)	Canola oil	Seluma™	15,000
Metabolix/Antibioticos	Witchgrass, camelina, sugar cane	Mirel, Mvera™	10,000
TianAn Biologic Material Co	Corn/cassava starch	ENMAT™	10,000
Tianjin GreenBio	Corn starch	SoGreen™	10,000
Bio-on	Beet or sugar cane	Bio-on™	10,000
Shenzhen Ecomann Biotech. Co	Corn starch		5000
PHB Industrial	Sugar cane	Biocycle™	2000
Kaneka	Vegetable oil	Aonilex™	1000
Biomer	Sugar (sucrose)	Biomer P™	–
Newlight Technologies	Waste methane	AirCarbon™	>500

Table 30.

Players in the PHB market today (status 2016).

Source: [163].

Taking an existing biopolymer and devising a more cost-effective production technology is more likely to bring success than trying to synthesize and/or isolate a totally novel bioplastics material. Hence PLA and PHB are promising materials for methanotrophic fermentation.

Taking 90% of 400 million tons of polymers and a conversion ratio of 1 g of polymer per g of methane, we see that roughly 13% of the global natural gas production would be required to provide the feedstock for the plastics. When we further assume that in the future, a significant share of polymer materials will be recycled, the demand for virgin polymers will be lower, also reducing the fraction of natural gas needed to cater to it.

Scale-up of PHB production by methanotrophic fermentation is discussed in Refs. [172, 173, 174, 175].

3.3 Value-added materials

Likewise for the above-mentioned and further chemicals that can be obtained from methane by fermentation, there is a sound market. Like SCP and biopolymers, they can be produced virtually without requiring (agricultural) land. Decoupling manufacturing from both crude oil and farming activities is attractive as it relieves pressure on existing feed and food value chains. When we estimate that between 500 and 1000 million tons of chemicals are to be made from other sources than oil, at again a rough value of 1 g methane per g of product, we end up with 16–32% of natural gas consumption per year.

3.4 Scenario of methane as sole feedstock

So if we were to produce “all” global demand for protein, plastics, and chemicals from natural gas, which for sure is an exaggerated assumption, we would require approximately 1/3 to 1/2 of the annual natural gas production of today. Let us revisit briefly Figure 7 and add the above estimations to it:

Figure 7.

Graphical depiction of annual natural gas demand for theoretically meeting the entire feedstock demand for SCP, plastics, and chemicals, compared to the natural gas production figures in IEA’s scenarios STEPS, APS, and NZE, which span 3300–4500 billion m³/year annual demand between 2010 and 2030 (compare Figure 7).

Also, biogas could cater to that raw material need. Production volumes are not yet close to 2000 billion m³ per year, but several studies suggest that that potential is within reach.

The authors are firmly convinced that methanotrophic gas fermentation is the key enabling technology to succeed in abandoning crude oil as universal feedstock, by allowing cost-effective and scaleable production of protein, plastics, and chemicals, in a sustainable way, by avoiding the need for immense agricultural and associated production factors, such as water, fertilizer, and pesticides. As Liew et al. have started in their review article [49]—gas fermentation offers a flexible platform for large, industrial-scale production of low-carbon-fuels and chemicals from various feedstocks. This view is shared by other researchers, see Ref. [176].

4. Summary and outlook

The world faces serious challenges from climate change, growing population, and increasing industrialization. The demand for food skyrockets, as does the desire for energy and various products. The capacity of the world’s oceans to supply fish and of the world’s fields to provide feed and food is limited, and measures to boost productivity have partly been exhausted. Prominent footprint calculations show clearly that the rate with which resources are consumed surpasses the regeneration by a factor of more than 2. Earth exhaustion day is advancing from year to year.

It is well-understood that the current meat production is not sustainable. Cultured meat obviously still needs significant development time to become cost-competitive [177]. Aquaculture can provide fish to the world’s plate despite overfishing, but it needs fodder, which today is often taken from oceans—fish meal and fish oil, or soy, which has its own sustainability issues. Plant-based protein for food is not necessarily sustainable either, considering the large land areas that are required, besides fertilizers, pesticides, etc.

There is an urgent need for large-scale and cheap protein sources that are independent of land use. This has re-sparked interest in microbial protein production, which does not need but very little land and water—feeding microbes sugar or starch virtually perverts any attempts for sustainability: Methane can be a very valid option here. Øverland et al. argued: “In recent years, the increasing global demand for sustainable protein sources, independent of marine origin, agricultural land use, and climatic changes, has led to renewed attention on the potential of microbial protein for use in animal production. Focus has been on methane, the main component of natural gas, which is found widely in nature [28] as an attractive substrate for bacterial protein production. The abundant supply, cheap transportation, and reasonable cost of natural gas indicate that protein production from natural gas could be realistic on a large scale. Using methane-oxidising bacteria as an amino acid source in animal nutrition may spare over-exploited sources of protein suitable for direct human consumption” [65].

Protein from natural gas might, in the unfortunate event of a global food catastrophe, be a vital protein source for several years. As Allfed states: “Human civilization’s food production system is unprepared for global catastrophic risks (GCRs). Catastrophes capable of abruptly transforming global climate, such as supervolcanic eruption, asteroid/comet impact, or nuclear winter, which could completely collapse the agricultural system. Responding by producing resilient foods requiring little to no sunlight is more cost-effective than increasing food stockpiles, given the long duration of these scenarios (6–10 years)” [169]. The preliminary techno-economic analysis revealed that bacterial SCP can be ramped up fast for global food production when needed [169]. Also, bacterial SCP might play a pivotal role in the energy transition, by providing suitable storage capacity for excess renewable energy in a Power2protein setup [178].

We see strong market dynamics in natural gas fermentation these days. This can be inferred from trends in scientific publications, patent applications, and commercial activities, such as pilot or manufacturing plant establishments by several market incumbents. It can be expected that the interest in this set of technologies will be even more in the near and medium-term future, as the key enabling technology for scaling up bioplastics production, protein manufacturing, and chemicals synthesis in a sustainable way, decoupled from agricultural operations.

Conflict of interest

The authors declare that they have no conflict of interest.