Duckweed Biogas: What the Digestion Data Supports

Search for duckweed biogas and the answer arrives as a single number wrapped in the phrase “great potential”. It is usually a biomethane potential in millilitres of methane per gram of volatile solids, quoted the way you would quote a fuel’s calorific value, as though it were a property of the plant.

It is not. Across the published record that number moves by nearly a factor of three, and the thing that moves it is rarely the species. This article gives the range with the conditions attached, and then makes the point that the literature supports but almost nobody states: duckweed performs as a co-substrate, not as a feedstock.

The range, not the number

Start with the most recent primary measurement. Three duckweed varieties grown on swine lagoon wastewater were digested under thermophilic conditions and returned 205, 217 and 262 mL CH4 per gram of volatile solids, with first-order hydrolysis constants of 0.205 to 0.285 per day. That paper also does the field a service by tabulating what came before it: a mesophilic review range of 190 to 340, alkaline and fermentation pretreatments at 190 and 230, co-digestion with food waste at 232, co-digestion with concentrated municipal and swine wastewaters at 340 and 361, and thermally pretreated waste activated sludge at 468.

The top of the published range comes from the study everyone benchmarks against, which crossed four duckweed species with three wastewater effluents. Spirodela polyrhiza returned 340 to 413 NL CH4 per kg VS and Landoltia punctata 343 to 408, both higher than the two Lemna species in the same design.

The bottom comes from the case closest to a real application. A mixed duckweed harvest taken off domestic wastewater phytoremediation, Lemna gibba with L. minuta and Wolffia columbiana, gave 153.4 mL CH4 per gram VS at 40 percent biodegradability, against 55 percent for the bulrush it was compared with.

So: 153 to 413. All three results are correct. They are answers to different questions, and a page that quotes one of them as the duckweed biogas number has thrown away the part that would have let you use it.

Duckweed is a co-substrate, not a feedstock

This is the finding that reorganises everything else, and it is hiding in a study about beer.

Researchers needed a nitrogen-rich material to rescue a lignin-stripped brewers spent grain pulp, and used Lemna minor for the job. In the process they state plainly what the duckweed literature tends not to: Lemna minor by itself is not a good biogas producer. Digesting it alone took 99 days to complete, against 23 days for the pulp. Co-digested, the pair reached a higher cumulative biomethane yield than the untreated spent grain had managed on its own.

The mechanism is in the companion characterisation. Lemna minor came in at 37.05 percent lignin in its fibre composition and a carbon to nitrogen ratio of 6.90, with 4.78 percent total Kjeldahl nitrogen in dry matter. That combination is the whole story. Very high nitrogen relative to carbon is exactly wrong for a digester on its own and exactly right as a corrective for a carbon-rich, nitrogen-poor substrate. On specific energy, duckweed alone gave 19.3 MJ per kg VS while the co-digestion gave 64.9.

What co-digestion adds, measured

Three independent pairings tell the same story from different industries.

On swine wastewater, adding Lemna minor biomass to the feed raised the gas production rate by 40 percent and specific methane production by 41 percent, from 93.0 to 131.0 mL CH4 per gram of COD. The same trial removed 58.9 percent of COD, 66.1 percent of phosphorus and 74.0 percent of ammonium nitrogen from the water on the way.

In a continuous stirred tank reactor on tapioca starch wastewater, specific methane production went from 0.28 to 0.35 NL CH4 per gram of COD removed, a 1.3-fold increase, with methane content holding at 64 to 65 percent. The authors attribute it to nutrient balance: the COD to nitrogen to phosphorus ratio moved from 300:4.4:0.8 toward the 300:5:1 target.

With cattle manure the result is blunter. In 0.6 litre batch digesters at 38 degrees, 100 percent duckweed produced more cumulative biogas than 100 percent cattle manure, 1015.5 against 862 millilitres, but the peak was 1206 at a 25:75 duckweed to manure mix. Neither pure feedstock won. Treat those absolute millilitres as internal to that study, since cumulative volume from bottle digesters does not compare across papers, but the shape of the curve is the point.

The common thread is nitrogen. In every one of these, duckweed is being used to fix somebody else’s carbon to nitrogen ratio. That is a real and valuable role. It is not the same claim as “duckweed is a biogas crop”, and if you are sizing a project the difference decides whether you need a partner feedstock contract.

Four variables that move the yield more than the species does

The water the biomass grew on

In the four-species comparison, methane potential varied with the effluent the duckweed was cultured in as much as with the species, and biomass grown on anaerobic digestion effluent scored highest. Growth water is not a background detail in the method section. It is a variable in the answer, which is one reason duckweed cultivation is designed rather than simply started.

Temperature, and how long you are willing to wait

A preprint running the same Lemnaceae digestion design across two seasons found 184 NmL CH4 per gram VS under winter conditions at 18 degrees and 323 under summer conditions at 26.6 degrees. It has not completed peer review, so weigh it accordingly, but the direction matches the thermophilic result above, where higher temperature completed the batch test in 28 days rather than the 55 a comparable mesophilic test needed.

Inoculum to substrate ratio

In that same preprint the best-performing ratio was not constant: 2.0 in the winter runs, 1.0 in the summer runs. A yield quoted without its inoculum to substrate ratio is missing an operating parameter that changed the result by more than the species did.

Pretreatment, and what it costs you

The comparison table above is worth re-reading as a decision. Alkaline and fermentation pretreatments delivered 190 and 230, while untreated co-digestion with concentrated wastewaters delivered 340 and 361. On that evidence the cheaper lever is choosing a co-substrate, not adding a process step.

Extraction ahead of digestion is a different question and the answer is more encouraging than expected. When valuable components were chemically extracted from Lemna minor first, the residual biomass still reached 0.16 litres of biogas per gram of organic carbon at 49.8 percent methane, only about 9 percent below the primary biomass. The cost showed up in time instead: stable digestion of the residual material took at least 30 days to start.

Where biogas actually belongs: the stream that can never be feed

Almost every result above used biomass grown on wastewater, manure or spent process water. That is not incidental, and it is the reason this topic matters to us.

At SERAPH the rule is published rather than implied: biomass grown on wastewater, manure or contaminated water is never sold or blended as food or feed, and the two streams are separated from the first sensor reading. Which leaves an obvious question for anyone running a treatment pond, and biogas is the honest answer to it. Anaerobic digestion is a route that does not require the biomass to be clean enough to eat.

That makes the energy question and the animal feed question genuinely separate decisions about genuinely separate material, taken at different points in the process. Anyone answering both at once with the same harvest has already made an error that no yield figure will correct.

Metals do not disappear, they move into the digestate

One study follows a contaminant all the way through, and it is the most useful thing published on this. Duckweed was used to take zinc and ammonium out of contaminated water, then the harvested biomass was digested.

Zinc suppressed methanogenesis. The clean control’s liquid hydrolysate reached 201.67 mL per gram of COD by day 30, while the zinc-loaded fractions reached 99.88 to 129.82. Solid fractions performed poorly for a separate reason, the medium acidified to roughly pH 5. And the authors flag the part that decides a project: residual metals end up in the digestate.

Recovering energy from contaminated biomass does not dispose of the contaminant. It relocates it into a smaller volume that now has its own handling route to plan and pay for. Anyone treating digestate as a free fertiliser by-product on the strength of a clean-stream study is planning for a different material than the one they will have, which is the same question duckweed phytoremediation raises about the biomass in the first place.

What this evidence base still does not cover

Look back at the sources. Almost all of it is batch bottle testing over 28 to 99 days, on biomass grown indoors or in bench-scale effluent, at a held temperature. The one continuous reactor was fed a fixed daily dose of dried duckweed, not a variable harvest off a real pond in November.

Which means the published record answers “can this biomass be digested” convincingly and “what will this site produce across a year” not at all. Between those two sits everything a project needs: seasonal biomass availability, how the duckweed on manure effluent composition drifts through a season, how much biomass survives harvesting and dewatering, and what the digester does when the feed rate is not a constant.

We should be equally clear about our own position. SERAPH is at TRL 3, proof of concept, with outdoor field validation still ahead, and we publish no yield figures of our own. What we do have is the map: a knowledge base of 256 indexed duckweed papers across 31 research areas in five domains, each carrying a coverage weight. Energy sits in the shallow band, and this article is what a shallow band looks like when you read all of it.

How to read any duckweed biogas number

Seven questions for the next figure you are handed:

1. Mono-digestion or co-digestion, and with what.
2. Per gram of volatile solids, per gram of COD, or cumulative millilitres. These do not convert into each other.
3. What water the biomass was grown on.
4. Mesophilic or thermophilic, and over how many days.
5. What inoculum to substrate ratio.
6. Pretreated or not, and at what cost in time.
7. Whether that biomass carried metals, and where the digestate goes.

A number that survives all seven can be built on. One that does not is still a real result. It just belongs to somebody else’s reactor, and the distance between their reactor and your pond is the whole engineering problem, the same distance that separates a lab growth curve from a season of duckweed wastewater treatment.

FAQ

How much biogas does duckweed produce?

Between 153 and 413 mL CH4 per gram of volatile solids in the published record, depending on species, growth water and digestion conditions. A mixed harvest off domestic wastewater phytoremediation sits at the bottom of that range and Spirodela polyrhiza grown on anaerobic digestion effluent at the top.

Is duckweed better than manure for biogas?

In the one study that compared them directly, 100 percent duckweed out-produced 100 percent cattle manure, but the highest yield came from a 25 percent duckweed and 75 percent manure blend. The useful answer is a ratio, not a winner.

Does duckweed need pretreatment before anaerobic digestion?

Not necessarily. Reported yields after alkaline and fermentation pretreatment, 190 and 230 mL CH4 per gram VS, sit below untreated co-digestion with concentrated wastewaters at 340 and 361. On that evidence, selecting a co-substrate is the cheaper lever.

What methane content does duckweed biogas have?

Roughly 50 percent in batch bioreactor work on Lemna minor mixtures, and a stable 64 to 65 percent in a continuous reactor co-digesting duckweed with tapioca starch wastewater. Methane content tracks the co-substrate and the reactor regime more than the plant.

Can biogas be produced from duckweed grown on contaminated water?

Yes, and that is the point of a remediation stream, subject to two constraints. Metals inhibit methanogenesis, and the metal ends up concentrated in the digestate rather than destroyed. That biomass is never food or feed under any circumstances.