Livestock Research for Rural Development 28 (10) 2016 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Effect of fermentation system on protein enrichment of cassava (Manihot esculenta) root

Vanhnasin Phoneyaphon, Nouphone Manivanh and T R Preston1

Faculty of Agriculture and Forest Resource, Souphanouvong University, Luang Prabang, Lao PDR
1 Centro para la Investigación en Sistemas Sostenibles de Producción Agropecuaria (CIPAV),
Carrera 25 No 6-62 Cali, Colombia


The substrate for a solid state fermentation was composed of fresh cassava roots as the carbohydrate source, urea to provide ammonia, diammonium phosphate as a combined source of ammonia and phosphorus, and yeast (Saccharomyces cerevisiae) as the fermenting organism. The treatments were variants on the traditional system used by farmers in the process of making rice “wine”; prior steaming of the cassava root; no steaming; anaerobic system (no air access); aerobic, spreading the substrate in thin layers (about 10 cm to facilitate air exposure.

The levels of true protein increased from 2 to 7% in DM after 7 days of incubation, representing a conversion of crude to true protein of approximately 70%. There appeared to be no benefits from: (i) steaming the cassava root prior to fermentation; and (ii) an aerobic system, providing access to air during the fermentation as opposed to anaerobic conditions with the substrates held in a sealed plastifor c bag. Oven–drying at 100°C for 24h of samples of cassava root containing urea, DAP and yeast, prior to or after fermentation, led to varying losses of N which decreased with the increase in duration of the fermentation.

Key words: aerobic, anaerobic, DAP, Manihot esculenta, steaming, true protein, urea, yeast


In most tropical countries there is an imbalance in availability of feeds rich in protein compared with those composed predominantly of carbohydrates. As a result, livestock production systems depend to a considerable extent on imports of protein -rich supplements, especially soybean meal.

An approach that would provide a partial solution to this problem is through protein-enrichment of feeds rich in carbohydrates by solid-state fermentation with micro-organisms. Khempaka et al (2014) showed that aerobic fermentation of steamed cassava pulp (crude protein content 2.0% in DM) with Aspergillus oryzae and urea (3.5% in DM of the pulp) resulted in a product with 8.1% true protein in DM, which could be incorporated at up to 16% of an intensive broiler diet without affecting growth performance and feed conversion. Improved growth rates in Moo Laat pigs were reported by Manivanh and Prreston (2016) when protein-enriched cassava root replaced ensiled Taro (Colocacia ensiformis) in a diet based on ensiled banana psuedo stem>

The aim of the research described in this paper was to study diffeent ways of protein enrichment of cassava roots based on the procedure traditionally used by farmers to produce rice wine.

Materials and methods

Experiment 1

The substrate for the solid state fermentation was composed of fresh cassava root as the carbohydrate source, urea to provide ammonia, diammonium phosphate as a combined source of ammonia and phosphorus, and yeast (Saccharomyces cerevisiae) as the fermenting organism. There were four treatments that were variants on the traditional system used by farmers in the process of making rice “wine”: Steaming prior to fermentation (ST) or not steamed (NST); aair access during fermentation (AA) or no air access (NAA).

The treatments were arranged as a 2*2 factorial with 4 replications. For each treatment/replicate, the basal substrate was 1 kg of freshly ground cassava root (CR).

For ST, the cassava root was steamed for 30 minutes, allowed to cool for 15 minutes and then mixed with urea (2% DM basis), diammonium phosphate (0.8% DM  basis) and yeast (2% DM basis).
NST: The same as ST but not steamed. 
AN: After mixing the substrates were spread on a plastic sheet at a depth of about 2cm to facilitate access to air.
NAN: After mixing, the substrates were put in a plastic bag which was closed to prevent air entry.


On days 0, 3 and 7, samples were taken from each treatment/replicate and dried in an oven at 100°C for 24h to determine the DM content. The dried samples were then analyzed for crude and true protein (AOAC 1990). For estimation of true protein, 0.5 g of the “oven-dried” sample was put in a 125ml Erlenmeyer flask with 50 ml of distilled water, allowed to stand for 30 minutes, after which 10ml of 10% TCA (trichloracetic acid ) were added and allowed to stand for a further 20-30 minutes. The suspension was then filtered through Whatman #54 paper by gravity. The filtrate was then discarded and the remaining filter paper and suspended substrate were transferred to a kjeldahl flask for standard estimation of total N.

Statistical analysis

The data were analyzed using the general linear option in the ANOVA program of the Minitab (2000) software. Sources of variation were: steamed (yes or no), air (yes or no),  interaction steam*air and error.

Experiment 2

This experiment was done to demonstrate the losses in nitrogen from the fermentation substrates in experiment 1 when they were subjected to 100°C oven drying prior to analysis for CP and TP.

The procedure for preparation of substrates was similar to that used for the NAN treatment in experiment 1.  Prior to fermentation, and 3 and 7 days after fermentation started, samples were analysed for crude and true protein using: (a) the fresh material; and (b) the dried material after heating at 100°C for 24h.

Results and discussion

Experiment 1

Data on the compostion of the substrates are in Table 1.

Table 1. Composition of substrates




Cassava root

DM, %





Nitrogen, % in DM





Crude protein, % in DM





#Phosphorus 20%

 Crude protein 

After adding and mixing the urea, DAP and yeast with the cassava root, and prior to fermentation,  the crude protein values were only slightly higher than in the cassava root (2% in DM) (Figure 1). After 3 days the crude protein had increased to 7.5% and by 7 days it was 16% in DM.

Figure 1. Mean values for crude and true protein according to days fermented (there were no differences among fermentation systems so the data were averaged for all systems)
True protein

There were no benefits  in true protein content from steaming the cassava root prior to fermentation and no differences between the aerobic and anaerobic treatments (Table 2)

Table 2. Mean values for true protein (% in DM) of substrate before and after 3 and 7 days of fermentation


Aeration Steaming




















It is assumed that during the process of fermentation increasing amounts of nitrogen (from the urea and DAP) would be converted to yeast, but that the overall level of “crude” protein would not change.  However, it was observed that at time zero (prior to fermentation), the crude protein  was only 3% increasing to 7.5% after 3 days and to 16% after 7 days,. These results indicated that nitrogen was being lost by drying the samples (at 100°C for 24h) prior to kjeldahl analysis,  but that the degree of loss decreased with time of fermentation reaching"zero" loss after 7 days. The proposed explanation is that in the process of fermentation, the action of the yeast led to increasing proportions of urea-N and DAP-N being converted to precursors of protein (ie: amino acids and peptides) and to protein per se, and that these compounds were not destroyed (evaporated?) by 100°C drying of the samples prior to determination of kjeldahl-N. The final value of 16% crude protein in DM aftet 7 days fermentation is higher than the theoretical content of N*6.25 which is about 13% in DM.  The difference could be accounted for by the loss of carbohydrate substrate which provided the energy for the growth of the yeast. Losses of  the order of 20% of substrate DM during a similar 10-day fermentation were observed by Sengxayalath Phoutnapha (2016, personal communication).

The results from experiment 2 (Figure 2) in which kjeldahl-N was determined on fresh samples, or after drying the samples at 100°C for 24h ,connfirmed that nitrogen is lost when a mixed substrate of fresh cassava root, urea, DAP and yeast is oven-dried for 24h at 100°C. At time zero, after addition of urea, DAP and yeast, the mean CP in the fermentation substrates that were not oven-dried was 8.5% in DM, reflecting the N contained in the cassava root, urea, DAP and yeast. After 24h of fermentation the CP remained the same indicating no loss of N over this period. By contrast, the CP in the DM of samples oven-dried at 100°C prior to fermentation (3% in DM) reflected only the protein provided by the cassava root (2%) and the yeast (1%). After 24h fermentation, the CP in oven-dried samples had risen to 4% in DM, indicating a similar tendeny to that observed in Experiment 1 (ie: reduced loss of nitrogen after commencement of fermentation).

Figure 2. Crude protein in the substrates determined in dried (100°C for 24h) or undried (fresh) samples prior to, or after, 24h fermentation with yeast, urea and DAP.
Nitrogenous compounds arising from yeast fermentation od urea and DAP

The unknown feature of the research concerns the nature of the non-protein nitrogenous compounds remaining at the end of the fermentation. This is not likely to be an issue in the feeding of ruminants, but could be of consequence when the objective is to produce “protein-enriched” cassava root for feeding to monogastric animals (pigs, chickens and fish) for which excessive levels of elemental or ionized ammonia could be toxic. This requires: (i) more detailed analysis to identify the nature of the non-protein nitrogenous compounds present at the end of the fermentation and not converted to yeast; and (ii) improving the efficiency of the fermentation to ensure that all the nitrogenous compounds are converted to true protein.



This research was done by the senior author as part of the requirements for the MSc degree in Animal Production "Improving Livelihood and Food Security of the people in Lower Mekong Basin through Climate Change Mitigation" in Cantho University, Vietnam. The authors acknowledge financial support for this research from the MEKARN II project financed by Sida.


AOAC 1990 Official methods of analysis. Association of Official Analytical Chemists, Arlington, Virginia, 15th edition, 1298 pp.

Khempaka S, Thongkratok R, Okrathok S and Molee W 2014 An evaluation of cassava pulp feedstuff fermented with A. oryzae, on growth performance, nutrient digestibility and carcass quality of broilers. Journal of Poultry Science, 51, 71-79.

Manivanh N and Preston T R 2016 Replacing taro (Colocasia esculenta) silage with protein-enriched cassava root improved the nutritive value of a banana stem (Musa spp) based diet and supported better growth in local pigs (Moo Laat breed). Livestock Research for Rural Development. Volume 28, Article #97.

Received 9 July 2016; Accepted 14 August 2016; Published 1 October 2016

Go to top