Evaporation or reverse osmosis for the liquid fraction of digestate?
Updated: 7 days ago
Short answer: Reverse osmosis (RO) comes from drinking and process water treatment. On the liquid fraction of digestate it needs ultrafiltration (UF) in front, frequent membrane cleaning, and it recovers around half of the volume as permeate. The rest is a concentrate that still has to be stored, hauled or treated. An evaporator takes out up to 90 % of the volume with mechanical vapour recompression (MVR) on thin liquid fractions up to 2.2 % DS, or around two thirds with a thermal multi-stage evaporator on waste heat, and binds the ammonia in a scrubber as ammonium sulphate solution (ASL). Which route is cheaper depends on the whole system, compared per m³ of recovered water.
What each process does with digestate
UF and RO push the liquid through membranes. UF holds back solids and colloids, RO holds back dissolved salts and nutrients. You get a permeate (the treated water) and a concentrate that carries the nutrients and salts.
An evaporator boils the water off (the thermal DV series under vacuum). You get a distillate, a nutrient concentrate and, with a scrubber, ASL. MKR builds two types:
MVR evaporator: runs on electricity, no waste heat needed. Around 22.5 litres of distillate per kWh of electricity for the whole plant, which is around 44 kWh per m³ of distillate. Feed up to 2.2 % dry solids, volume reduction up to 90 % on thin liquid fractions up to 2.2 % DS.
Thermal multi-stage evaporator (DV series): runs on CHP waste heat. Around 4.3 to 4.7 litres per kWh of heat. Feed up to 4.5 % dry solids, volume reduction around two thirds.
How much water do you recover?
Published data for membrane plants on agricultural digestate show a permeate share of up to around 50 % (Bolzonella et al. 2018). A pilot plant on a 2.5 MW biogas plant recovered 38 % of the volume as permeate, and around 54 % when the UF retentate was recirculated (Gienau et al. 2018). In other words: out of every 100 m³ that go into the membrane line, around half comes back as concentrate.
An MVR evaporator on a suitable feed takes out up to 90 m³ of every 100 m³. That difference matters more than any single energy figure, because every cubic metre left over has to be stored and hauled.
What happens to the RO concentrate?
It stays a liquid with the nutrients and salts of the digestate, and with around half of the original volume it still needs storage, transport and spreading.
RO concentrate can be evaporated further. In a recent laboratory trial with RO concentrate from a digestate plant, we concentrated it by a factor of 5.9 to 1. Chloride in the concentrate reached around 7,000 mg/l, and for continuous operation we recommended a factor of 4 to 1. So evaporation is not only an alternative to RO, it can also be the next step after it.
Cleaning and availability
Membranes foul on digestate. In the pilot plant mentioned above, the membranes were cleaned chemically every one to two weeks, with acid and alkaline steps of around two to three hours (Gienau et al. 2018).
On our reference evaporator on a food-waste AD plant, a full acid and alkaline cleaning runs once a week in the day shift, and the heat transfer stayed stable from August to October. Measured operating data are in our article "How does an evaporator perform in daily operation?".
How to compare energy figures fairly
Energy figures for digestate treatment are often compared on different terms. A fair comparison needs five things:
The same system boundary. Count UF, RO and the treatment of the RO concentrate, not only the RO stage.
Electricity and heat shown separately. A kWh of CHP waste heat does not cost the same as a kWh of electricity.
The same feed. Same medium, same dry solids, same pre-treatment.
The same reference. Per m³ of recovered water (permeate or distillate), not per m³ of feed. With 50 % yield, every m³ of permeate needs 2 m³ of feed.
Measured values, not design values. For the MVR evaporator we state the whole plant, measured at the switch cabinet.
For membrane treatment of digestate, published values for the complete process lie between 16 and 25 kWh per m³ treated (as cited by Gienau et al. 2018). At around 50 % permeate, that is around 32 to 50 kWh per m³ of permeate. The treatment of the concentrate is not included.
So on energy per m³ of recovered water, the two routes can be closer than often assumed. The bigger differences are the volume left over and what happens to it.
What happens to the nitrogen?
With RO, most of the nitrogen ends up in the concentrate and goes to the field with it.
With an evaporator, ammonia moves into the vapour. A scrubber binds it with sulphuric acid as ASL, a liquid nitrogen-sulphur fertiliser. The digestate itself is not acidified. Whether ASL can be used under the EU RENURE criteria depends on the national implementation and on the product meeting the criteria.
How ASL and the nutrient concentrate go back to the field from a food-waste AD plant is shown in our article "From plate to field".
When does which route fit?
Your situation | Route to look at |
|---|---|
CHP waste heat available | Thermal multi-stage evaporator (DV series) |
No waste heat, liquid fraction up to 2.2 % dry solids | MVR evaporator |
Low-salt feed, a use for the concentrate nearby | UF and RO |
RO already in place, concentrate volume is the problem | Evaporation of the RO concentrate |
From our experience: reverse osmosis becomes the more economical route at larger volumes, roughly above 200,000 m³ of liquid fraction per year. Below that, an evaporator is usually the more reliable choice, because RO needs ultrafiltration upstream and still leaves a concentrate that often has to be evaporated as well.
Every project starts with your own digestate. Before a binding offer we run a trial evaporation with your sample in our laboratory.
Where we replaced reverse osmosis
German plant, washing liquid from separated digestate fibres: an ultrafiltration and RO concept did not work on this medium. Our MVR evaporator replaced it and shortened the process chain. The washing liquid is not digestate, but it looks similar to a thin liquid fraction.
Food-waste AD plant in Switzerland: RO did not work with this medium. It was replaced by our thermal multi-stage evaporator (DV series), which has been running on CHP waste heat for more than 5 years.
See it for yourself
Reference visits and a talk with the operator of our food-waste reference plant are possible on request. The plant was shown at a Biomasse Suisse field day in 2025, the slides are public on the Biomasse Suisse website.
FAQ
How much water does RO recover from digestate?
Published data show around 38 to 54 % as permeate, typically around half of the volume (Bolzonella et al. 2018, Gienau et al. 2018).
Above what volume does reverse osmosis make sense for digestate?
In our experience, reverse osmosis becomes the more economical route at larger volumes, roughly above 200,000 m³ of liquid fraction per year. Below that, an evaporator is usually the more reliable choice, because RO needs ultrafiltration upstream and still leaves a concentrate that often has to be evaporated as well.
How much volume does an evaporator take out?
Up to 90 % with an MVR evaporator on thin liquid fractions up to 2.2 % DS, and around two thirds with a thermal multi-stage evaporator, depending on the feed.
How much electricity does an MVR evaporator need?
Around 22.5 litres of distillate per kWh for the whole plant, which is around 44 kWh per m³ of distillate. Operating costs start at around €8 per m³ of feed.
Can RO concentrate be evaporated?
Yes. In a laboratory trial we concentrated RO concentrate from a digestate plant by 5.9 to 1. For continuous operation we recommended 4 to 1 because of the chloride content.
Can the distillate be discharged?
That depends on local limits and on the finishing steps after the evaporator. On our reference plant it is reused as process water.
Sources
Bolzonella, D., Fatone, F., Gottardo, M., Frison, N. (2018): Nutrients recovery from anaerobic digestate of agro-waste: techno-economic assessment of full scale applications. Journal of Environmental Management. https://doi.org/10.1016/j.jenvman.2017.08.026
Gienau, T., Brüß, U., Kraume, M., Rosenberger, S. (2018): Nutrient recovery from anaerobic sludge by membrane filtration: pilot tests at a 2.5 MWe biogas plant. International Journal of Recycling of Organic Waste in Agriculture. https://doi.org/10.1007/s40093-018-0218-6
Comments