Sustainable Protein and Bioenergy from Green-crop Refining

Biofuels could be produced without loss of food production potential from the land, and with a substantial reduction in net biomass feedstock cost, by harvesting lush green perennial crops like grass or lucerne (alfalfa) and extracting large quantities of high quality protein as a coproduct. 

The costs of the harvesting and processing would be offset by sale of the protein, which is an established poultry feed, and more recently has been shown to be suitable for aquaculture.

The straightforward mechanical process can extract over two tonnes of protein per hectare per year from perennial green crops, and that protein, as chicken or aquaculture feed, could be worth around US$2500 per tonne of protein.

A basic form of the extraction process[1] was developed in the UK, France, New Zealand and the US in the 1960s through to the mid-1980s, as a means of creating value from the 30-40% of the typical crop protein content that is regarded as surplus to livestock requirements[2] [3] [4]. Commercialization attempts proved to be technically successful but economically marginal under the conditions of the time. However, the surviving group of companies in France has been operating on a large scale (~50,000 hectares) for over 30 years.

My research group piloted new techniques that can extract up to 80% of the crude protein in the crop. Our high-extraction "biorefinery" process[5] was developed for the New Zealand Liquid Fuels Trust Board as part of a national response to the 1980s fuels crisis. The process was extensively tested and modeled[6], but scale-up beyond 1000kg of crop per hour was prevented by the ending of the crisis.

Today there is renewed uncertainty of fuel supply, exacerbated by climate change, nitrogen pollution, and decline of soils and fisheries. These factors are changing economic drivers dramatically, and I would like to see the green-crop refining process reconsidered.

The Green-crop refining process

After intensive protein extraction the fibre and carbohydrate crop components would be converted by fermentation or gasification to gaseous or liquid fuels. The process costs would be heavily offset by the value of the protein by-product.

The main difference between the well-established moderate-extraction process and our high-extraction process is in the raw-material grinding stage. In the former the "pulping" is done with a simple (though unusual) hammer mill, and the pressing of the protein-rich juice from the pulp is done using screw presses similar to those used in the wine industry. In the intensive extraction process there is further fine-milling in a disc mill of the type used for the "mechanical refining" of wood-chips. Recycling loops are used to keep the properties of the pulp in a workable range. No water or other chemical inputs need to be added.

Most lignocellulosic biofuel processes could be adapted to a protein-extracted feedstock without adding much equipment. Most crops harvested for conversion to biofuels will require fine grinding anyway to achieve high throughput. In addition, protein in biofuel feedstocks will have to be dealt with somehow anyway, or it will form a noxious nitrogenous waste stream that will be expensive to treat. The operations required to recover the protein are few: steam-heating of the plant juices, centrifugal separation of the protein precipitate (a simple decanter centrifuge operation), and optional drying.

The energy cost of intense grinding is moderate. In our pilot trials 70-80% protein extraction was achieved using an energy input of 200kWh per tonne of crop dry-matter.

The residue after extraction of protein from grass or alfalfa is a finely-divided fibre almost free of cell contents, low in moisture content, and naturally low in lignin. If non-protein extracted components are recycled back into the fibre fraction, the material available for biofuels manufacture is roughly 65% fibre, 30% soluble sugars and minerals, and 5% non-protein nitrogen compounds, with a yield of around 10 tonnes DM/ha under common NZ conditions.

Carbon emissions from the soil should be low because the crops will typically be perennial and need little or no cultivation.

Co-product Markets

The protein concentrate has a ready market as poultry feed, where it provides yolk and skin pigments as well as a protein rich in the economically-important amino acids tryptophan, isoleucine and threonine.  The protein concentrate is rich in omega-3 fatty acids[7].

The potential value as a human food of protein concentrates made from lucerne has been extensively examined[8], and they have recently been assessed as safe for human consumption by the European Food Safety Authority[9].

The protein concentrate can be further refined to produce higher-value products, such as a "white" undenatured protein with food-processing functionality, omega-3 fatty acids, carotene, xanthophyll (the principal pigment for poultry), vitamin E, and other nutraceuticals. Variations on the refining process were developed in the 1980s, and are being further developed in France in an EU research programme[10] [11]. 

The crude protein concentrate may have a particular value in aquaculture, which is projected to treble in size in New Zealand (to $1B/y) over the next 15 years[12], and to increase worldwide [13]. High-value carnivorous fish prefer diets rich in fishmeal, but the international commodity price of fishmeal has been rising: the February 2010 FAO fishmeal commodity price was US$1627/tonne (see chart below) [14]. 

The amino acid profile of leaf protein is quite similar to that of fishmeal[15]. Leaf protein concentrate is already being offered commercially as an aquaculture feed, with claims of improved growth and pigmentation benefits[16].

The annual yield of protein concentrate from a hectare of regularly-harvested high-productivity New Zealand pastoral land, using known processes, should be about 3.6 tonnes fishmeal-equivalent[17], which at current fishmeal prices could be worth more than US$6000 (NZ$8000) per hectare per year.

Synergies with other processes

Green-crop harvesting is quite seasonal, with peak availability in spring. There may be opportunities to share much of the process plant and services infrastructure with other biomass crop-waste sources that tend to have peak availability in the autumn. There are other potential synergies with livestock abattoirs, where the excellent ensiling properties and high digestibility of the extracted fibre have potential to provide high-weight-gain lairage feed for the autumn peak intake of animals.


Trends in protein-source prices

Commodity price trends

Chart data source:




Rod McDonald

Hamilton. New Zealand


Phone: +64 212268441



[1] Desialis, Concentrated Alfalfa Extract / Process. (accessed 2010)

[2] European Commission RTD Info magazine, New opportunities for Lucerne, (accessed 2010)

[3] NorthDakota State University, Study to Focus on Benefits of Value-Added Alfalfa Processing to Region, 1999. (accessed 2010)

[4] Pirrit N, Dunstall M. Drying of fibrous crops using geothermal steam and hot water at the Taupo Lucerne Company.

[5] Vaughan SR, McDonald RM, Donnelly PE, Hendy NA, Mills RA (1984). The Biomass refinery as a route to fuel alcohol from green crops. Proc. 6th International Symposium on Fuel Alcohol Technology, Ottawa, May 1984.

[6] Vaughan SR, McDonald RM,  A feasibility study of the production of ethanol by hydrolysis and fermentation of protein extracted lucerne fibre, MAFTech Liquid Fuels Trust Board contract 310/13/1, October 1987 (850 pages). Filed at as LF1134, though incorrectly titled. (inaccessible 2011)

[7] Desialis, Concentrated Alfalfa Extract / Composition, (accessed 2010)

[8] Zanin V, A new nutritional idea for man :lucerne leaf concentrate, Association for the Promotion of Leaf Concentrate in Nutrition, 1998. (accessed 2010)

[9] European Food Safety Authority, Opinion on the safety of Alfalfa protein concentrate as food. The EFSA Journal (2009) 997, 1-19. (accessed 2010)

[10] European Commission,  EU research leads to a major innovation for agro-food industry, Press release 28 June 2000. (accessed 2010)

[11] European Commission Research Infocentre, The hidden virtues of lucerne, November 2001. (accessed 2010)

[12] Heatley P, Speech to Aquaculture New Zealand annual meeting, November 2009. (accessed 2010)

[13] World fish production. (accessed 2011)

[14] FAO International commodity prices. (accessed 2010)

[15] McDonald RM, Donnelly PE, Mills RA, Vaughan SR (1985). High value products from Lucerne: a New Zealand perspective. Proc. 15th International Grasslands Conference, Kyoto, pp910-911

[16] Vitalfa LLC, Alfalfa Nutrient Concentrate for Aquaculture. (accessed 2011)

[17] Assumes crop harvest of 16tDM/ha/y at average 25% crude protein content (Nx6.25), extracting 80% of the crude protein, and recovering 75% of the extracted crude protein as heat-precipitable true protein. This assumption yields 2.4 tonnes of protein, or 4.8 tonnes of concentrate at 50% protein content. Equivalence with fishmeal is adjusted for 65% protein content in fishmeal, with no allowance for extra value due to potential pigmentation or growth enhancement.





Site updated 1 January 2012