Livestock Research for Rural Development 20 (7) 2008 Guide for preparation of papers LRRD News

Citation of this paper

Palm kernel cake extraction and utilisation in pig and poultry diets in Ghana

M Boateng*, D B Okai, J Baah* and A Donkoh

Department of Animal Science, Faculty of Agriculture, College of Agriculture and Natural Resources. Kwame Nkrumah University of Science and Technology, Kumasi – Ghana

*Lethbridge Research Centre, Agriculture and Agri-Food Canada, Lethbridge, AB T1J 4B1




The President’s Special Initiative on oil palm has spurred oil palm cultivation and processing and would subsequently lead to an increase in the production of palm kernel cake  which is the by-product obtained after the extraction of the oil. A review was therefore carried out to investigate the various methods of extracting palm kernel oil and palm kernel cake production and the extent of palm kernel cake utilization in Ghana. Two industrial-scale and two other cottage-scale palm kernel oil extraction set-ups were visited to ascertain the processes employed in the production of the palm kernel oil and palm kernel cake. The available literature on palm kernel cake was obtained in order to undertake a review of palm kernel cake utilization in animal nutrition in Ghana and elsewhere.


The “Traditional” and Expeller processes were identified as the two methods of palm kernel oil extraction in Ghana and these produce the “Cottage-type” and “Factory-type” palm kernel cakes respectively. The literature surveyed showed that in one study, a 30% palm kernel cake-diet with high level of residual fat led to a higher average daily gain and better feed conversion efficiency compared to a low-fat palm kernel cake-based diet; but increased carcass fat with a consequent reduction in leanness in pigs. A 20% palm kernel cake-diet also yielded positive responses in broilers and layers but beyond that, reduced egg numbers and quality were recorded.


The high fibre content has been identified as the main limitation to its optimum utilization in the feeding of non-ruminant livestock. Some authors have observed that palm kernel cake may also be quite gritty and may require grinding before feed compounding. The cottage-type palm kernel cake has been found to have a rather strong smell which may lead to reductions in feed intake and growth performance if high levels are included in diets. Some studies have shown that supplementation of palm kernel cake-diets with exogenous non-starch polysaccharide degrading enzymes such as Mannanases may help to ensure its maximum utilization by pigs and poultry and help to reduce feed cost as well as promoting a more efficient use of a material that has a potential of becoming a major environmental pollutant.

Key words: African oil palm, expeller, factory-type, indigenous


The African oil palm, Elaeis guineensis (Jacq.) is believed to have originated from the tropical rain forest region of West and Central Africa (Hartley 1988). It produces fruits which are characterized by a thick fibrous and oil-rich mesocarp and a white opaque kernel embedded in a brown casing referred to as the endocarp. The African oil palm offers major prospects for development in that every part of the plant can be put into profitable use. It produces two main commercial products: palm oil, which is approximately 22% of the weight of the fresh fruit bunch , and the palm kernel oil which represent 4-6% of the fresh fruit bunch. When the nut is processed, it yields palm kernel oil and palm kernel cake or meal depending on the method of extraction. According to Okeudo et al (2005), the by-product from the mechanical expeller procedure is referred to as palm kernel cake, whilst that from the solvent extraction technique is called palm kernel meal. But for ease of referencing in this review, palm kernel cake and palm kernel meal are used interchangeably. Palm kernel processing in Ghana is by either one of two processes viz: expeller press which yields the “Factory-type” palm kernel cake and an indigenous local technique that produces the “Cottage-type” palm kernel cake.


Local/traditional palm kernel cake production 

Local/traditional oil extraction in Ghana is primarily carried out by women in the countryside. Their activities commence with the purchase of the nuts mainly at the peak of the rains when there is abundant supply of the palm kernels and prices are lowest. These women move from house to house in the oil palm processing towns buying and bulking the kernels for the palm kernel oil production. The nut processing and oil extraction is usually undertaken during the dry season and consists of the following:


Drying of kernels


The palm kernel oil extraction process begins with moisture reduction in the kernels. This is achieved in the traditional setup by storing the kernels in sacks for two to four weeks. In some cases, the kernels are spread on the bare ground in the open to enhance the drying process and reduce drying time. Moisture reduction in the kernel has been identified as a pre-condition for excellent cracking. Good cracking results in kernels which are whole and completely detached from the shells.




This is mainly done by the women and their household with the help of solicited labour from other children in the vicinity. It is done by crushing the shell in between two stones. This tends to be very slow and laborious and does not prove very helpful where large quantities of nuts are to be processed. One advantage associated with hand cracking is that there is no need for shell and kernel separation as the kernels are separated from their nuts as they are cracked.


It is common nowadays to find mechanized cracking units/mills at the various processing centres. This has become important because the throughput of these processing outfits has increased far above what can be manually cracked. The only limitation to the mechanized cracking is that, the product comes out as kernel-shell mixture which must be separated in a different process.


Kernel and shell separation


There are two processes involved in this step. First, there is the separation by winnowing. The kernel-shell mixture is put in a container and tipped down and as they fall, the lighter weight shell and some fibre that might have accompanied the nuts are blown away by the air current. The second step is by the use of a clay bath which operates on the principle of floatation. The clay bath is prepared to a viscosity that allows the kernels to float whiles the shells sink to the bottom and the floating kernels are scooped into baskets, washed with clean water and then dried. A clay bath of specific gravity 1.17 is reported to be the ideal for effective kernel and shell separation (Hartley 1988; Asiedu 1989).


Oil extraction


After drying, the kernels are roasted to release some amount of the oil and then ground into a paste in a motor-operated mill or grinder. The paste is mixed with water and boiled to release the oil which is from time to time skimmed from the surface. In some localities, the kernel paste is mixed with boiling water and allowed to stay overnight after which the oil settles on top and is then skimmed off. The solid residue which is the wet palm kernel cake is usually heaped in the processing centres until there is a demand for it. If there is none, it is virtually abandoned to become a substrate for mould growth and also pollute the environment.


Factory type palm kernel cake 

The expeller/mechanical process has been identified as the most common method of oil extraction this part of the world. This method generally has four steps:

1.                  Depericarping

2.                  Nut cracking

3.                  Kernel and shell separation

4.                  Oil extraction




Depericarping commonly refers to the process of removing fibre from the nuts. This is achieved through the use of a rotating drum fitted with baffles. The fibre-nut mixture is fed into the drum rotating at 15rpm. The baffles elevate the fibre-nut mixture and allow them to drop. As they fall, a current of air is passed through them which blows the partially dried fibre to the exit.


The clean nuts may then be conveyed to a nut silo for drying. At the Ghana Oil Palm Development Company (GOPDC) for instance, nuts are dried to a moisture content of about 14% before they are fed into the cracker. According to Hartley 1988, the moisture content of the nut must be less than 16% for the kernels to be sufficiently shrunk from the shells in order to achieve good cracking.


Nut cracking


Cracking is achieved when the kernel passes through two rollers rotating in opposite directions. Movement of the rollers imposes pressure on the kernel and causes the shell to break, releasing the kernel. In other mills, nuts are cracked by a centrifugal cracker.  Nuts fed into the cracker are thrown out of slots and hurled against a cracking ring. The nutcase breaks upon impact and releases the kernel. One disadvantage associated with the centrifugal cracker is that nuts with long fibres at the tail may not crack if the fibre side comes against the cracking ring.


Kernel and shell separation


Generally, there are two ways of achieving this and both of these mimic that of the traditional set up. These methods are known as dry/pneumatic and wet separation.

In the dry separation, which is likened to winnowing in the local set up, fragments of light shells and those with long fibres that may have densities a little lower than that of the kernels are blown out of the separator.


The second (wet) separation is achieved through the use of water in a hydrocyclone which replaces the clay bath. Through the activity of centrifugal force, fast-moving water current is created within the cylinder-like hydrocyclone. The heavy particles, particularly the shells, are drawn towards the bottom of the hydrocyclone whilst the kernels remain afloat and are washed into a collecting chamber. After separation, the kernels are dried before being pressed for the oil. At one of the factories visited, it is a common practice to dry the kernels to a moisture content of 6% as against the 5 – 7% reported by Asiedu (1989) before they are considered suitable for oil extraction.


Oil extraction


First, the kernels are fed into a grinder which breaks them into small fragments. The kernel fragments are put through a roller mill which presses the kernels into flakes. The flaked material is then subjected to steam conditioning, before it is finally conveyed into the expeller for the oil to be pressed out.  On the other hand, the kernel flakes can be put through the press directly after flaking. This is normally followed by a second press to achieve maximum extraction. It must however, be appreciated that a great deal of heat evolves during the direct press and this goes to compensate for the steam conditioning in the earlier process.


Palm kernel cake utilization in Ghana 

In view of the abundance of palm kernel cake in many West African countries, the potential of palm kernel cake as a cheap material for feeding livestock must be fully exploited. The inability to do so thus far can be attributed to either the lack or the flow of information on chemical compositions, nutritive values, improvement methods and feeding responses of animals fed palm kernel cake-based diets.


Nutritive value of palm kernel cake


Although palm kernel cake supplies both protein and energy, the proximate composition of palm kernel cake suggests that it can be classified as an energy source. The solvent-extracted and expeller pressed palm kernel cakes are of different quality depending upon the extent and efficiency of oil extraction (Sundu et al 2006). Expeller-pressed palm kernel cake has higher oil content than solvent-extracted meals. O’Mara et al (1999) reported values of 5 – 12 % for expeller pressed and 0.5 – 3 % for solvent-extracted palm kernel meal. These two processes thus make the percentage of other nutrients, such as crude protein and minerals, lower in expeller-pressed palm kernel cake. According to O’Mara et al (1999), the expeller pressed palm kernel cake has lower crude protein content than the solvent-extracted palm kernel cake (Table 1).

Table 1.  Proximate analysis of palm kernel cake (g/kg DM) after different oil extraction procedures1

Method of extraction


Ether Extract

Crude fibre







Solvent extraction





1O’Mara et al 1999

With a crude protein content of 14-21% (Table 2), palm kernel cake is precluded as a protein source as in most of the cases, the crude protein level is less than 20%.  However, the crude protein content is enough to meet the requirement of most ruminants despite other negative attributes reported by other researchers (Hair-Bejo and Alimon 1995; Hassan and Yeong 1999)


The resistance of dietary fibre in palm kernel cake to monogastric digestive enzymes has also been reported by Sundu and Dingle 2003. The fibre content of palm kernel cake (21-23%) has been associated with a decline in nutrient digestibility of palm kernel cake, especially when fed to monogastrics. This notwithstanding, palm kernel cake can still be a major source of energy and protein to farm animals. Analyses have shown that about 49% of the dry matter in palm kernel cake is in the form of nitrogen free extractives (Sundu et al 2006).

Table 2.  Nutrient Composition of palm kernel cake




Dry matter, %


Sundu et al 2005

Crude protein, %



Crude fibre, %



Lipid, %



Ash, %



Gross energy, Kcal/kg



Metabolisable energy, MJ/kg




Alimon 2004







Sundu and Dingle 2003 had earlier reported that during processing palm kernel cake may also undergo Maillard reaction (the reaction of mannose with amino groups leading to the formation of a brown complex) due to heat applied in the process before oil extraction or due to heat evolved during oil extraction, and this will adversely affect the digestibility of the feedstuff.


The fibre content of palm kernel cake may not make it a first choice feedstuff for monogastrics but ruminants with their rumen microbial population can make very good use of the nutrients in them. palm kernel cake has largely been used in Sub-Sahara Africa for cattle feeding, especially dairy animals where it can serve as the main protein source (Bedingar and Degefa 1990).


Reports by Yeong 1983 and Hutagalung et al (1982) suggest that the amino acid composition of palm kernel cake is not very good (Table 3). However, very high figures have been recorded for the availability of its amino acids. Nwokolo et al (1976) had estimates for all the essential amino acids and indicated their availabilities to be in excess of 85% except for valine which is only 68.4% available.

Table 3.  Amino acid composition of palm kernel cake, %


Yeong et al  1983

Hutagalung et al 1982











































Palm kernel cake utilization in poultry


Due to its high fibre content, the use of palm kernel cake in poultry rations is very limited. Osei and Amo (1987) evaluated palm kernel cake as a broiler feed ingredient. palm kernel cake partially replaced maize at levels of 0, 5, 7.5, 10, 12.5 and 15%. The addition of palm kernel cake to the diet had no significant effect on feed consumption and body weight up to 8 weeks of age. However, feed conversion efficiency significantly declined as palm kernel cake levels reached 12.5% and above.


Palm kernel cake utilization in swine


Palm kernel cake has been found to reduce the cost of swine diets. In a study on growing pigs, Okai and Opoku-Mensah (1988) assessed the performance of the pigs on palm kernel cake at 0, 10 and 20% inclusion levels. At the end of the 28-day trial period, all the growth performance criteria were found to be higher for the palm kernel cake diets except for feed intake which was similar for all the three treatments. Back fat thickness was however higher (P< 0.05) for the palm kernel cake-based diets. There was a decrease in the cost of a kg feed with increasing levels of dietary palm kernel cake.


Rhule 1996 also fed the palm kernel cake at 0, 20, 30 and 40% inclusion levels to grower pigs from 25 to 90 kg liveweight without any adverse effect on performance. The average daily weight gains (ADG) by the pigs for the grower stage were 0.46, 0.46, 0.43 and 0.41kg respectively on the 0, 20, 30 and 40% palm kernel cake diets. These values 0 %palm kernel cake and 20% palm kernel cake were similar (P < 0.05) and significantly higher (P < 0.05) than with the 30% palm kernel cake diet, which were in turn similar to those with the 40% palm kernel cake diet. But at the finisher phase, there was not significant difference between the ADGs though that of the 40% palm kernel cake was lower.


In another study, Rhule 1998 evaluated the effect of two diets containing palm kernel cake from two oil mills (A and B), incorporated into the diets at 300 g/kg level. The average daily gains of the pigs were 0.57, 0.46, and 0.49 kg/day for those on control, palm kernel cake A and palm kernel cake B diets respectively, during the grower period. The corresponding values were 0.60, 0.63, and 0.65 kg/day during the finisher period, and 0.60, 0.54, and 0.55 kg/day during the entire grower-finisher period. It was also reported that the palm kernel cake with high level of residual fat led to a higher average daily gain and better feed conversion efficiency but there was increased carcass fatness and reduced leanness in pigs.


Work by Okai et al (2006) under the AgSSIP Non-ruminant Programme indicated that when a 34.5% palm kernel cake diet was fed to pigs, there was no deleterious effect on carcass characteristics. However, pigs on a control, maize-based diet reached market weight earlier than those on the palm kernel meal diet.


Current research on palm kernel cake utilization


It can be inferred from the literature reviewed that palm kernel cake has been accepted as a viable feed ingredient. The only hindrance to its application has been the high level of non-starch polysaccharide which poses a limitation to the rate of utilization. The current trend in modern animal nutrition is to find means of making the non-starch polysaccharides more available to farm animals, especially monogastrics. One approach that has proved useful in recent years is the addition of enzymes to the feed. The ability of exogenous enzymes to break down high fibre feedststuffs and improve their nutritive value has been reported by Ofuya and Nwajiuba (1990, Chen et al (1997), Okai et al (2000) and Iyayi and Davies (2005).


Sundu and Dingle (2003) have there are basically three enzymes that are needed to improve the nutritive value of palm kernel meal, namely: mannanase, α galactosidase and celullase to digest the mannan, cellulose and the galactosidic side chains of the palm kernel cake. Laboratory studies have proven this theory (Balasubramaniam 1976). It is for this reason that exogenous enzymes are now being tested to determine the extent of improvement in the digestion of palm kernel meal.




The authors wish to acknowledge the help and support obtained from the Ghana Oil Palm Development Company Limited, the Agriculture Services Sub-Sector Improvement Programme (AgSSIP), Basic Environmental Systems & Technology Inc., Canada and the Department of Animal Science, KNUST- Ghana.


Alimon A R 2004 The Nutritive Value of Palm Kernel Cake for Animal Feed. Palm Oil Development. No. 40, June 2004. Malaysian Palm Oil Board


Asiedu J J 1989 Processing Tropical Crops. ELBS/Macmillan. Hong Kong. pp. 182-183


Balasubramaniam K 1976 Polysaccharides of the kernel of maturing and mature coconuts. Journal of Food Science 41: 1370-1373


Bedingar T and Degefa G 1990 Trends in agro-byproducts and their feeding potential in sub-Saharan Africa.  Working document No. 14, ILCA, Addis-Ababa


Chen J, Lu T and Han Z 1997 Poultry production in China and the potential for using enzyme preparations. In: Enzymes in Poultry and Swine Nutrition. IDRC Publication, pp 154-155


Hair-Bejo M and Alimon AR 1995 The protective role of zinc in palm kernel cake  toxicity in sheep. Malaysian Journal of Nutrition 1: 75-82, 1995


Hartley C W S 1988 The Oil Palm. 3rd Edition. Tropical Agriculture Series. Longman, Singapore. Pp 724-729


Hassan O A and Yeong S W 1999 By-products as animal feedstuffs. In: Oil Palm and The Environment: A Malaysian Perspective (Editors: S Gurmit, K H Lim, L Teo and D K Lee), pp. 225-239. Malaysian Oil Palm Growers’ Council, Malaysia


Hutagalung R I, Mahyuddin  M and Jalaludin S C 1982 Feeds for farm animals from the oil palm. In: The oil Palm in Agriculture in the Eighties, Volume 11. Editors: Pushprajah and C Poh-Soon, Incorporated Society of Planters, Kuala Lumpur, pp 609 – 622


Iyayi E A and Davies B I 2005 Effect of enzyme supplementation of palm kernel meal and brewer’s dried grains on the performance of broilers. International Journal of Poultry Science 4 (2): 76-80


Nwokolo E N, Bragg D B and Saben H S 1976 The availability of amino acids from palm kernel, soybean, cotton seed and rape seed meal for the growing chick. Poultry Science 55: 2300-2304


Ofuya C O and Nwajiuba C J 1990 Microbial degradation and utilization of cassava peel. World Journal of Microbiology and Biotechnology 6: 144-146


Okai D B and Opoku-Mensah D 1988 The effects of varying levels of palm kernel meal on growth performance and carcass characteristics of pigs. Proc. 18-19th Ghana Animal Science Association Symposium, UST, Kumasi. pp. 54-57


Okai D B, Olympio O S and Anim P K 2000 Responses of grower-finisher pigs to diets containing varying levels of wheat bran with or without Optizyme – an exogenous enzyme complex. Journal of the University of Science and Technology 20 (1-3):55-61


Okai D B, Rhule S W A, Nyannor E K D and Tandoh I 2006 Feed Packages Based on Local Agro-industrial By-products for Small-scale Pig Farmers in the Ashanti Region. Paper presented at the Project Leaders Meeting, AgSSIP Non-Ruminant Programme held at Animal Research Institute. Frafraha, Accra


Okeudo N J Eboh K V and Izugboekwe N V 2005 Growth rate, carcass characteristics and organoleptic quality of broilers fed graded levels of palm kernel cake. International Journal of Poultry Science 4 (5): 330-333


O’Mara F P, Muligan F J, Cronin E J, Rath M and Caffrey P J 1999 The nutritive value of palm kernel meal measured in vivo and using rumen fluid and enzymatic techniques. Livestock Production Science 60:305-316


Osei S A and Amo J 1987 Palm kernel cake as a broiler feed ingredient. Poultry Science 66:1870-1873


Rhule S W A 1996 Growth rate and carcass characteristics of pigs fed on diets containing palm kernel cake. Animal Feed Science and Technology 61:167-172


Rhule S W A 1998 The influence of type of palm kernel cake on the growth rate and carcass characteristics of pigs. Ghana Journal of Agricultural Science 31 (2): Abstract


Sundu B and Dingle J 2003 Use of enzymes to improve the nutritional value of palm kernel meal and copra meal. Proceedings of Quensland Poultry Science Symposium. Australia 1(14):1-15


Sundu B, Kumar A and Dingle J 2006 Palm kernel meal in broiler diets: effect on chicken performance and health. World Poultry Science Journal 62:316-325


Yeong S W 1983 Amino acids availability of palm kernel cake, palm oil sludge and sludge fermented product (Prolima) in studies with chickens. MARDI Research Bulletin 11:84-88

Received 29 February 2008; Accepted 10 April 2008; Published 3 July 2008

Go to top