Sources of copra, palm kernel, and cashew nut co-products from feed mills in South-East Asia, and a source of almond hulls from Australia were delivered to DSM Nutritional Products, Singapore. The suppliers provided between 100 and 300 grams of each ingredient. Samples were labeled, cataloged, and then shipped to the University of Illinois, Urbana, IL, USA, where most of the chemical analyses were conducted. Oilseed coproducts included two sources of copra expellers from the Philippines, two sources of copra meal from the Philippines, and one source of copra meal from Vietnam. Twelve sources of palm kernel expellers from Indonesia, the Philippines, and Vietnam were also used. Nut co-products included one source of almond hulls from Australia and one cashew nut meal from Indonesia.

Samples of all feed ingredients were finely ground and analyzed for dry matter (method 930.15) [11] and ash (method 942.05) [11]. Gross energy was analyzed using an isoperibol bomb calorimeter (model 6400; Parr Instruments, Moline, IL, USA). Samples were analyzed for amino acids (AA; method 982.30 E a, b, and c) [11] on a Hitachi AA Analyzer (Model L8800; Hitachi High Technologies America Inc., Pleasanton, CA, USA) and nitrogen was analyzed by combustion (method 990.03) [11] using a LECO FP628 Nitrogen Analyzer (LECO Corp., Saint Joseph, MI, USA). Crude protein was calculated as nitrogen×6.25. Acid-hydrolyzed ether extract (AEE) was analyzed using 3N HCl (Ankom^HCl; Ankom Technology, Macedon, NY, USA) followed by crude fat extraction using petroleum ether (Ankom^XT15; Ankom Technology, USA). Insoluble and soluble dietary fiber were quantified according to method 991.43 [11] using the Ankom^TDF Dietary Fiber Analyzer (Ankom Technology, USA). Total dietary fiber was calculated as the sum of insoluble and soluble dietary fiber. Minerals were analyzed (method 985.01 a, b, and c) [11] using inductively coupled plasma-optical emission spectrometry (ICP-OES; Avio 200; PerkinElmer, Waltham, MA, USA). Sample preparation included dry ashing at 600°C for 4 h (method 942.05; 10) [11] and wet digestion with nitric acid (method 3050 B) [12]. Total starch was analyzed in the copra and nut co-products using the glucoamylase procedure (method 979.10) [11]. Sugars including glucose, fructose, maltose, sucrose, stachyose, and raffinose were also analyzed in copra and nut co-products using high-performance liquid chromatography (Dionex App Notes 21 and 92).

For each analysis of copra and nut co-products, analyzed proximate components were added and subtracted from the concentration of dry matter in each ingredient to calculate the un-analyzed rest fraction according to the following equation:

The rest fraction for palm kernel expellers was calculated using the same equation, with the exception that total starch and sugars were not analyzed in palm kernel expellers and, therefore, not included in the equation.

The coefficient of variation and average concentration of nutrients in samples within each group of feed ingredients were calculated if two or more samples from the same country were available. Normality of residues and homogeneity of variances were verified using the UNIVARIATE procedure (SAS 9.4 Institute Inc. Cary, NC, USA). Data were analyzed by analysis of variance using the PROC MIXED procedure in SAS to test statistical differences between copra co-products and between country of origin for palm kernel expellers. The replicate sample was the experimental unit for all analyses. The feed ingredient or country was the fixed effect, and the replicate sample was the random effect. Means were calculated using the LSMEANS statement in SAS, and when significant, means were separated using the PDIFF option in the MIXED procedure. Results were considered significant at p<0.05.

The chemical composition of copra co-products (Tables 1 and 2) and palm kernel expellers (Tables 3 and 4) were within the range of published values for these ingredients [3,13–15]. The analyzed components in the chemical composition of copra co-products and palm kernel expellers were close to 100%, indicating that all nutrients in these ingredients were accounted for [13]. This is indicated by the fact that the calculated rest fraction was close to or less than 3% for all samples. Except for dry matter, all nutrient analysis results were adjusted to 88% dry matter because this is a typical value for oilseeds and allows for a direct comparison without the influence of moisture.

Copra expellers had greater (p <0.05) concentrations of dry matter and AEE compared with copra meal. However, copra meal had greater (p<0.05) concentrations of total dietary fiber (soluble and insoluble) and copper than copra expellers. The differences in nutrient composition between copra expellers and copra meal are primarily due to differences in residual oil remaining in the product [3], and the nutritional quality of oilseed meals is dependent on the process of oil extraction [1]. Copra meal samples from Vietnam and the Philippines appeared to be very similar in composition.

No differences were observed between the palm kernel expellers from Indonesia and Vietnam, with the exception of histidine and tyrosine, which were greater (p<0.05) in palm expellers from Indonesia compared with palm kernel expellers from Vietnam. The observation that there were few differences in AA concentration between countries for palm kernel expellers indicates that the growing conditions in Indonesia and Vietnam did not have major impacts on composition. It also appears that the processing was not different between the two countries.

Both copra and palm kernel co-products are used as protein sources in animal diets, but their protein content is relatively low when compared with other oilseed meals, with approximately 21.50% in copra co-products and 15.50% in palm kernel expellers. Calculated lysine/crude protein ratio for samples in this study was on average 2.30% and 2.70% for copra co-products and palm kernel expellers, respectively. However, arginine is the indispensable AA present in the greatest concentration in both co-products with calculated arginine/crude protein ratios of 9.80% and 10.70% for copra co-products and palm kernel expellers, respectively. The high concentration of arginine (2% for copra co-products and 1.70% for palm kernel expellers) may result in antagonism because arginine competes for the same transporter in the enterocytes as lysine [16]. As a result, adequate levels of digestible lysine in diets are required to mitigate the negative effects of the high arginine content of these co-products [3].

Total dietary fiber concentrations in copra expellers aver aged 42%, 47% in copra meal, and approximately 60% in palm kernel expellers, and these values are consistent with published data [3]. Because of the high fiber content, palm kernel expellers may not be suitable for feeding of young pigs and poultry [5]. However, with adequate levels of standardized ileal digestible AA, inclusion of up to 10% palm kernel expellers may be acceptable for weanling pigs [3]. Inclusion of more than 10% of copra co-products in diets for weanling pigs reduced average daily feed intake, most likely due to the slow rate of passage of the fiber through the digestive tract, which results in increased gut fill and therefore reduced feed intake [17].

When compared with other oilseed meals, except soybean meal, copra co-products has a high concentration of sucrose, with around 7%, but unlike soybean meal and other legumes, copra co-products do not contain oligosaccharides that may increase diarrhea in young animals [18,19]. The sugar profile of copra co-products determined in this study is in accordance with data for copra expellers [13]. Because of the high sucrose content in copra co-products, digestible and metabolizable energy values are greater in copra co-products than in palm kernel expellers [3]. The sucrose concentration in copra meal is most likely due to the sucrose in the coconut sap, which transports nutrients to the coconut [20].

The mineral composition demonstrated that potassium was present in high concentration in both co-products, with approximately 2.10% in copra co-products and 0.64% in palm kernel expellers, but mineral concentration in feed ingredients may be influenced by soil mineral composition [21]. Copra and palm kernel co-products were also high in phosphorus and iron, with an average of 0.62% phosphorus and 600 mg/kg iron, with the level of phosphorus being close to published values for soybean meal. These values are also in agreement with previously reported values for copra meal and palm kernel expellers [22]. Less than one-third of the phosphorus in copra co-products is bound to phytate, whereas the majority of the phosphorus in palm kernel expellers is bound to phytate [22]. Therefore, standardized total tract digestibility of phosphorus in diets containing palm kernel expellers can be improved if phytase is added [22].

The chemical composition of almond hulls and cashew nut meal (Tables 5 and 6) were in agreement with published values for these ingredients [7,8,23,24]. Except for dry matter, all analyzed nutrients were adjusted to 88% dry matter to allow for a direct comparison of chemical composition among ingredients. The analyzed components in almond hulls added up to close to 100%, indicating that all nutrients for this ingredient were accounted for [13]. In contrast, despite extensive nutrient analysis of cashew nut meal, analyzed nutrients accounted for only 94.50% of this ingredient, and it is not clear what the remaining 5.50% consist of. It is, however, possible that additional carbohydrate fractions that were not captured in the analyzed carbohydrate components, account for the remaining nutrients. Nevertheless, AEE represented a significant part of the analyzed nutrients in cashew nut meal, although at a lower concentration in this study compared with previous data [8]. The cashew nut meal contained approximately 35% AEE, which is due to the fact that fresh raw cashew kernels contain approximately 48% AEE, however, the nutritional composition of processed cashew nut meal is influenced to some extent by the region where the cashew trees are grown [25].

The nutrient composition of cashew nut meal analyzed in this study also indicated a high concentration of AA, with approximately 8% being indispensable AA and a lysine/crude protein ratio of 4.50%, which is very favorable for pigs and poultry. The 24% total dietary fiber in the cashew nut meal will likely not be a problem in diets for pigs and poultry, and the AEE concentration will increase energy in the diets. Cashew nut meal may partially replace corn in swine diets [26], and can also be included in broiler diets at an inclusion rate of 20% to 25% without affecting growth performance [10].

Almond hulls is a co-product with some variation in com position among different sources, and the nutritional quality is determined by cultivation, variety, processing, and harvesting conditions [7]. Analyzed composition of almond hulls in this study demonstrated that sugar, primarily glucose, fructose, and sucrose, accounted for nearly 23% of total nutrient concentration. This is generally in agreement with reported values for almond hulls, although sucrose was slightly lower in this study than reported by Sequeira and Lew [27] and DePeters et al [9]. Rain-damage may reduce sugar concentration, and thus energy values in almond hulls [7]. Very high concentration of iron was also observed in both nut co-products, with approximately 500 mg/kg in almond hulls and 750 mg/kg in cashew nut meal.

The analyzed chemical composition demonstrated that almond hulls contained around 44% total dietary fiber, and almond hulls may, therefore, be more suitable for ruminants than for pigs and poultry. Due to its fiber and energy contents, almond hulls can be used as a forage or as a concentrate ingredient for ruminants [28], with inclusion levels limited to less than 30% in diets for growing steers [29]. Diets containing almond hulls also need to be supplemented with extra protein due to the low protein concentration in almond hulls (approximately 5%). Calvert and Parker [30] demonstrated that 10% to 15% conventional grain may be replaced with almond hulls in diets for growing pigs if protein levels are balanced.