Aqueous Enzymatic Extraction of Oil from Rapeseeds Denmark 1991-1994 Pilot Scale



Vegetable oils and animal fats are important ingredients for several different foodstuffs and non-food products. About 80% of the total market of oils and fats is of vegetable origin [1]. Rapeseed contains oil (40-46%) and protein (20-30%) of high quality, but also glucosinolates which can be critical for both oil and protein quality. Especially if too high concentrations of these compounds or their degradation products are present in the oil and protein products. Novel processing methods of rapeseed have therefore attracted appreciable attention [2].

Oil from rape seed is traditionally produced by a combined process using pressing followed by extraction with organic solvents, like hexane. These organic solvents are recycled. This process creates in principle two products, the oil and the low-valued meal, that is used as feed ingredient or as fertilizer. The content of glucosinolates, the too high content of cellulose and tannins, aromatic cholinesters, and phytic acid in the conventional rapeseed meal still present some nutritional disadvantages. It is only usable for cows. The use of solvents presents a constant danger of explosion or fire risks and pollution [3,4]. In some countries (United States) the building of new factories using hexane to extract the oil is prohibited [5].

Cleaner Production Principle:

Process modification

Cleaner Production Application:

In contrast to traditional processing the enzyme process is based on the use of water as solvent and cell wall degrading enzymes to facilitate an easy and mild fractionation of oil, protein and hulls. The oil is found inside plant cells, linked with proteins and a wide range of carbohydrates like starch, cellulose, hemi-cellulose and pectins. The cell content is surrounded by a rather thick wall which has to be opened so the protein and oil can be released. Thus, when opened by enzymatic degradation, down-stream processing makes fractionation of the components possible to a degree which cannot be reached when using the conventional technique like pressing and hexane extraction.

Novo Nordisk developed a cell wall degrading enzyme, the multi-activity preparation SP 311, which is based on one selected strain of Aspergilus niger [3].The enzyme process provides oil and protein of a higher quality. The processing of oil seeds can be divided into the following steps: pre-treatment, release of oil and purification of the products.

In the pilot plant Bornholm the following method is developed:

Upstream process [6]:

Time(h) Drymilling (Hammer mill) 400 kg Rapeseed, riddle 3.0 mm. Heating of 800 kg water to 95oC.
0.00 "Cooking"- 15 min at 95oC (to inactivate the myrosinases and lipases). After adding the rapeseed the temperature drops to about 85oC and heating is necessary to reach 95oC, life steam.
0.45 Cold water is added to total 1750 kg.
1.00 The rapeseed slurry is Fryma milled in-line cooling to 50oC, heat exchanger or tank cooling.
2.00 pH adjustment to 4.5 approximately 7 kg 75 % H3PO4.
2.10 Addition of enzyme, 4 kg, SP-311.
2.15 The rapeseed slurry is Fryma milled, in-line.
3.15 The rapeseed slurry is Fryma milled, in-line.

This takes between 3 and 4 hours enzymatic treatment. Separation and purification of the products comprised decanting to remove hulls. Oil, protein-rich meal (Novo rapeseed meal), and syrup were separated in three washing and centrifuging steps. The fraction with Novo rapeseed meal was finally spray-dried, and the syrup was evaporated. Compared to the traditional process, the rapeseed products are thus divided into three fractions (oil, rapeseed meal, syrup) and the hulls. The consumption of energy, water etc. for the conventional and the enzymatic process are given in table 1.

Table 1: Annual consumption processing 100,000 tons of seeds a year [7].

  Enzyme process Conv. Process
Steam (4 bar) 46080 ton 27454 ton
electricity 12168 Mwh 9519 MWh
phosphoric acid 1498 ton 0
NaOH 90 ton 90 ton
enzyme(SP 331) 504 ton 0
process water 288000 m3 101000 m3
cooling water 115200 m3 123000 m3
Hexane 0 200000-400000 L


plant (day)

plant (shift)







These figures are based on calculations for a fictional factory (upscale of a pilot plant). The water used in this process is recycled as far as possible. The products from the enzymatic extraction, especially the protein meal, is improved compared to the normal hexane extracted rape meal according to tests and feeding trials of Novo Nordisk. The quality of the enzymatic extracted rape meal is such that it can be used as substitute for soy bean concentrate, fish meal or skim milk powder for feeding of piglets, minks and calves. The glucosinolate content in the protein meal is much smaller than in the conventional rapeseed meal.

Table 2 : Product specifications for the enzyme and conventional process [7]

Components in % w/w Conv Oil Enzyme Oil Conv metal Enzyme metal Enzyme fibres Enzyme Syrup
Water 0 0 11,5 10,0 10,0 50,0
protein 0 0 34,0 49,3 15,5 13,4
fat 100 100 1,7 23,3 5,1 0,2
fibers 0 0 appr.35 10 47,0 0,0
'sugar' 0 0 ? ? ? 20,6

The sum of the %w/w components will not always give 100% because some other small components are present in the products which are not measured.

The fat contains some lecithins (phospholipids), which gives an indication about the quality of the oil. The enzymatic produced oil contains 0.03 % w/w lecithins against 1.8 % w/w in the conventional produced oil. When the lecithin content is too high it has to be reduced by mixing the oil with water. For the other products the % w/w lecithins are not known.

The wastes from both processes are small quantities of solid wastes, condensed water and steam. Some of the water used still remains in the products (almost 50% of the syrup is water). In the traditional process a small loss of hexane is inevitable.

Scale of operation

The separation process in the pilot plant Bornholm is a batch process and is based on a capacity of 800 kg rapeseed/h.100,000 ton rapeseeds a year as raw material will provide the following product quantities.

Table 3 : Annual capacities [7].

Enzyme Process Production tons/year
Oil 35000
protein meal 32400
fibers 16301
syrup 28303
Conventional Process  
Oil 38000
Rape meal 62000

Stage of development

After one year of operation the pilot plant is up to good standards and fully operational. The Whole crop Biorefinery Project is organised by The Bioraf Foundation in co-operation with 10 research institutes and industrial companies from 5 European countries. The project receives financial support from the EC's ECLAIR-Program and different national research and development Programs. The main objective of the project is to develop and test the integrated Biorefinery Concept with the objective of finding new utilisation and markets for agricultural crops. The project spans over a period of 42 months from June 1991 to December 1st 1994 [5].

Level of Commercialization

The enzyme is commercially available at Novo Nordisk since 1982 and is also used for other purposes. For the purpose of enzymatic oil extraction described in this case study a totally new factory has to be build. The data in this case study resulted from pilot-plant operations at Novo Nordisk facilities in Denmark. Pilot plant trials have been carried out in other countries too. At this moment there is no other commercial application yet.

Table 4 : Material/Energy Balances per 100,000 tons of seeds

Material category

waste generation

Quantity before

organic solvents

Quantity after

no organic solvents

hexane 200,000-400,000 0
phosphoric acid 0 1498 ton
water use 224000 m3 403200 m3
electricity 9519 MWH 12168 MWh
steam (4 bar) 27454 ton 46080 ton

The quantity of hexane used in the conventional process is not known but is estimated at a few liters per ton of seeds.

Environmental and Economic Benefits:

Environmental benefits

No hexane or other organic solvents are used, less environmental risks.
Less risk for an explosion, each year 1 oil factory explodes somewhere in the world.
The content of glucosinolates, the content of cellulose and tannins, aromatic cholinesters and phytic acid in the rapeseed meal is lower. In the conventional process these components still present some nutritional disadvantages.
Higher quality products, so better and increased use of raw materials.

Investment Costs

To implement the process a totally new plant has to be build. The investment costs for a plant processing 100,000 tons of rapeseed per year, are estimated as follows in table 5.

One opportunity Novo Nordisk is presently working on is the synergism of an oil factory and a potato factory. During the potato season the factory is used to produce potato meal and during the rest of the year the factory is used for the extraction of oil out of rape seed. According to Novo Nordisk this construction is economically feasible because only small adjustments are necessary [8].

Table 5 : Investment costs in DKK [7].

  Enzyme process Conventional process
process equipment 24,97 46,68
drying equipment 27,63  
Utilities 11,60  
Engineering and installations 10,85 11,46
buildings(silos, etc) 25,21 24,37
unexpected costs 13,03 46,68
Total 113.29 105.81

Operational and Maintenance Costs

Table 6 : production costs, data from Denmark 1989 in mil. DKK [7].

  Enzyme process Conventional process
Manpower 4,8 5.8
energy 15.0 5.1
Subsidiary mat. enz. 38.0 0.0
Subsidiary mat. chem. 5.0 1.0
prod.costs per year 62,8 11,9

The enzymatic process for oil extraction from rape seeds offers several advantages, but it is considerably more expensive than the traditional solvent based process. For the biggest part, the variable costs per ton of processed rapeseed are constant, but for the capital costs, the size-related economic effects are considerable for a capacity which is 20 times larger, the fixed costs are only 4 to 4.5 times bigger.

It is estimated that enzymatic rapeseed processing can be profitable at an annual capacity of approximately 60,000 tons of processed rapeseeds per year [5]. For these estimations the product prices used are the average market prices for the familiar products (from the conventional process) and calculated substitution prices for the specific biorefined products.

Payback Time

The payback time is not known yet. The enzymatic process is 5 - 6 times more expensive than the conventional hexane process. These costs have to be covered mainly by additional income from the by-products, since the quality of the oil is only slightly improved by the lack of phospholipid content. If the enzymatic extracted rape meal is compared to similar quality feed ingredients (soy bean concentrate, fish meal or skim milk powder) and not to lower quality ingredients( conventional rapeseed meal) the costs of the process can be covered. At this moment the higher quality of the products is not valued in the market and therefore the prices are too low.


No commercial experience in continuous production.
The new process only prevents the use of an unknown quantity of hexane. On the opposite a lot more water, energy and phosphoric acid is used. This makes it very complicated to say which one is "cleanest production". When we take into account that once a year a factory with hexane explodes the balance is positive for the enzymatic process.
The farmers, who should participate in the biorefinery, are very traditional.
The recovery of oil is in the enzymatic process about 8% lower than in the conventional process (see table 3; complementary, the enzyme meal has a higher fat content, Table 2). It must be assumed that this lower recovery is accounted for in estimates on the economic feasibility, but nevertheless it will be an optical drawback as the oil is the main product of the processes.


Type of Source Material : Midterm Reports of the project, technical information, internal reports.


  1. Guido Ruivenkamp,"De invoering van Biotechnologie in de agro-industriale produktie keten", uitgever Jan v. Arkel Utrecht, 1989.
  2. S ren Krogh Jensen, "Biochemical and Physiological Investigations of the meal and syrup fractions from aqueous enzymatic rapeseed processing", Novo Nordisk A/S, Bagsvaerd, 1990.
  3. Hans Sejr Olsen, Novo Industri A/S,"Aqueous enzymatic extraction of oil from seeds", paper presented at the Asean food conference '88, 24-26 October 1988, Bangkok, Thailand.
  4. Hans Sejr Olsen, Novo Industri A/S,"Aqueous enzymatic extraction of rape seed oil"0, lecture given at workshop on Agricultural refineries, a Bridge from Farm to industry, 16-18 September, 1987.
  5. Morten Gylling, "The whole crop biorefinery project", midterm Assessment-concept, Denmark, October 1993.
  6. Midterm Report," Aqueous enzymatic extraction of oil, protein, starch, syrups etc." The whole crop biorefinery project, met o.a. Novo Nordisk, 1993.
  7. Technical information, Novo Nordisk, NJ-930126 KiJa, 1993.
  8. Interview Hans Sejr Olsen, Novo Nordisk.

Level of detail of the Source Material

Some more detail about the enzymatic process is available.

Industry/Program Contact and Address
Hans Sejr Olsen
Bioindustrial Group
Novo Nordisk A/S
Novo all
2880 Bagsvaerd
Tel:    +45-4442 2045;      Fax:    +45 4468 5096
Novo Nordisk
Abstractor Name and Address
Annelies den Braber
TME, Institute for Applied Environmental Economics
Grote Marktstraat 24
2511 BJ The Hague
Tel: +3170-3464422; Fax: +3170-3623469
Name and Location of Company
Pilot plant Bornholm
Bioraf Denmark Foundation
Lykkesvej 11 B
DK-3720 Aakirkeby
Tel: +45 56975556

Review Status:

This case study was submitted to UNEP IE by the Institute for Applied Environmental Economics (TME) of the Netherlands on the part of the UNEP IE working group on Biotechnology. The case study was edited for the ICPIC diskette in August 1995.

Subsequently, in September 1998 the case study underwent a technical review by Dr. Prasad Modak, Environmental Management Centre, Mumbai, India.