Cassava is one of the most important staple foods grown in tropical Africa.
It has gained popularity in sub-Saharan Africa because of its appreciated role
as a food security crop (Oduro and Ellis, 2000). Since,
it is potentially able to provide more food cal ha-1 than most crops,
it is being looked upon as the main hope for combating human starvation in the
sub-region. Unfortunately, the food security potential of cassava is negated
by the vulnerability of the crop to deterioration shortly after harvest, limiting
its contribution to incomes. The roots and leaves also contain various amounts
of cyanide, which at high levels, are toxic to both humans and animals. Therefore
after harvest, cassava has to be quickly converted into suitable forms of low
cyanide levels with longer and stable shelf life (Asiedu,
1989; Opara, 1999). The processing of cassava into
various forms that combine the advantages of diversity, nutritional value and
convenience of use is further means of promoting its consumption among different
strata of the society (Oduro and Ellis, 2000).
The various derivatives into which fresh cassava roots can be processed are
unlimited. By far its processing into a fermented, dried, granular food product
called gari is more popular in sub-Saharan Africa than other derivatives (Asiedu,
1989; Opara, 1999). In Ghana for example, gari processing
is becoming a fast expanding enterprise, providing employment and income generation
opportunities for many farmers and commercially oriented individuals in the
Traditional cassava processing into gari is however, very labour intensive
and productivity is often too low to justify the investment of labour, time
and money. Also, the yields from many indigenous processing methods are often
of poor quality, further reducing their market value and the profitability of
the industry. A wide array of indigenous practices and technologies exist in
traditional gari processing with only a few of the unit operations mechanized
to some extent, mainly to reduce the enormous drudgery involved. Improvement
of cassava processing techniques would therefore greatly increase labour efficiency,
incomes and living standards of cassava farmers and the urban poor, as well
as enhance the shelf-life of the product (Hahn, 2006).
Engineering interventions to improve the industrial process is often by way
of reducing drudgery and maximizing gari yield through the mechanization of
the unit operations with appropriate machinery and by the use of improved stoves
to maximize energy efficiency and improve the working conditions of the roasting
In an effort to maximize gari yield in a given processing system, there is the need for detailed process analysis in order to identify the various factors accounting for material losses and to subsequently re-design the process to reduce losses. As a first step, the mass balance for each unit operation and the entire process must be determined. Subsequently, the garification rate (i.e., the rate of cassava conversion into gari) associated with the processing system has to be determined to provide a numerical index that can be used to estimate the gari yield from any given quantity of fresh tubers. With this numerical index, alternative processes or the effect of process variations can be compared. This knowledge is invaluable for development and evaluation of new processing technologies.
Therefore, the purpose of this study was to set out a standard measure to quantify the losses occurring at the various processing stages and then determine the garification rate associated with a common traditional gari processing method using a set of standard equipment.
MATERIALS AND METHODS
Background of Traditional Gari Processing Equipment and Method
Traditional gari processing is basically a manual operation involving 10
basic steps with some variations in the processing details in different resource
areas. The processes are peeling, washing of peeled tubers, grating, sacking
of grated pulp, dewatering, fermentation, sifting, roasting, sieving and cooling
in that order (Asiedu, 1989). The range of equipment for
cassava processing are presented by Foodnet (2002).
Peeling and Washing
Peeling of the harvested tubers may be done in the farm to reduce the bulk
of material transported home. On the other hand, the fresh tubers may be transported
to a peeling bay at home. Peeling involves the removal of the outer cortex,
the neck and the small tapering end. The sizes and shapes of cassava tubers
differ tremendously, making the peeling process difficult to mechanize and therefore,
it is usually done manually with a knife (Asiedu, 1989).
Manual peeling requires little technical knowledge and is less risky. It is
however, a laborious process, requiring about 1 h for one person to peel 25-37
kg (Opara, 1999). The peeling rate however, depends on
the skill of the individual. After peeling, the roots are thoroughly washed
in a basin of clear water.
Washing is followed by grating or rasping of the peeled tubers into fine
dough. Grating raptures the tissues and brings together the differentially compartmentalized
substrate (cynogenic glucoside) and enzyme (Linamarase) to react and release
toxic hydrogen cyanide, which can subsequently be removed (Asiedu, 1989). The
greater the extent of cellular damage, the greater the interaction of the two
and the greater the subsequent cyanogens removal.
Grating is an energy intensive process and may be done manually or by use of an engine or motor-driven mechanical grater. The manual grater consists of a perforated rectangular grating sheet fixed upon two wooden frames on the opposite longitudinal edges and bowed in the middle so that, the jagged ends of the perforations are outwards. The operator rubs the peeled cassava tuber briskly over the jagged ends, which rasps and reduces the tuber into grates. Recent developments have also produced pedal types of manual rolling graters. In this version, the grating sheet is wrapped over a cylindrical drum, which can be revolved by pedaling or cranking.
Mechanical engine or motor-driven graters offer more intensive advantages than manual graters in terms of capacity and drudgery removal; and have superseded the use of manual graters where resources are available. A number of different kinds of designs have been formulated. They are the disc type and the cylindrical drum type.
The main working component of the cylindrical drum type is a grating drum (rotor or roller) in the form of a hollow cylinder normally made of hard wood or cold-drawn mild steel, having a number of grooves milled longitudinally into rasping blades or the steel sheet may be perforated. The jagged ends of the perforations are responsible for the grating (Asiedu, 1989). The drum is mounted onto a central shaft, which is supported by ball bearings on either side. A pulley is keyed to the shaft at one side and this receives rotary power from the prime mover through a flat belt.
The cylindrical drum is housed in a wooden or metal framework, the top part of which consists of a hopper, with the base slanted towards one side of the drum to form the discharge chute. The hopper may be trapezoidal or oval-shaped to reduce spillage. The lower portion of the housing (just below the drum) extends into the chute through which the grated mass leaves the machine. Between the cylindrical drum and the chute is a clearance adjustment board, which can be moved in or out to vary the clearance between the drum and the lower end of the hopper that leads into the chute. Reducing the clearance results in fine mash whilst increasing the clearance produces medium or coarse mash.
In the disc type, the grating surface consists of a perforated circular galvanized steel sheet mounted on to a circular wooden plate. Together, they sit on a metal plate which in turn, is keyed to a power-driven vertical shaft. A hopper which also serves as a cover for the grating chamber sits on top and is designed to prevent the cassava pieces from spilling over the hopper.
Mechanical graters may be designed into mobile types or may be stationary in fixed installations. A 1.0-5.0 h.p. petrol engine may be used for mobile graters and 5.5-10.0 h.p. diesel engine for stationery graters. Where electricity is available, it is more economical to use an electric motor instead. Power from the prime mover is transmitted to the grating machine through a belt drive. Man-operated graters have grating capacities ranging from 15-25 kg harvest-1 while, the capacities of the mechanical power driven types range from 0.5-1.2 t harvest-1 depending on the type of machine.
Fermentation and Dewatering
After grating, the wet dough is collected into jute sacks or large baskets
and left in a fermentation rack or in the open to ferment for 1 to 5 days depending
on the local practices and consumer preferences (Asiedu, 1989). Fermentation
allows glucosides (the bound cyanide) to be broken down into soluble hydrogen
cyanide (HCN), which dissolves in the water and is subsequently expressed off
(Asiedu, 1989). The time it is allowed to ferment affects the garis color,
taste, texture as well as the level of removal of the hydrogenic cyanide content.
Generally, a period of 24 h is enough to reduce the cyanide content to a safe
level for human consumption.
Dewatering involves the removal of the HCN-dissolved water and some quantity of starch from the fermented dough. Depending on the resources available, a pile of stones or a mechanical screw press may be used. The mechanical screw press is more efficient with less drudgery and is available in two versions-the single screw press and the double screw press. The single screw press consists of a perforated metal cage or basket with a vertical screw and a ram assembly. The fermented dough is loaded into jute sacks and placed in the cage. The ram is screwed down against the sacks squeezing off the water, which exit through the perforations.
The double-screw press works on the same principle but the ram consist of a heavy horizontal press bar, which is screwed down between two vertical screws. Wooden boards are placed over the loads to distribute the pressure over a large contact surface and to avoid the bursting of the jute sack. De-watering is complete in about a day and since only about 24 h is also required for fermentation, most traditional practices perform both fermentation and de-watering together.
Dewatering leaves a damp cake of cassava dough, which must be pulverised
and the ungrated particles sieved off. A woven mesh made of woven splinters
of cane is used for the sifting. It is placed over a wooden box and screened
so that the fibre and ungrated lumps are trapped above as the loose, damp flour
is collected in the box.
Roasting, (also called frying) is a most important step and requires a lot
of skill to produce gari of the desired quality. It is mainly a dehydration
process during which the sifted flour is subjected to heat treatment on a hot
surface to dry the flour. The hot surface is provided by heating an open earthenware
or iron pan gently from below. A small quantity of the wet flour is spread thinly
in the pan and a portion of calabash is used to stir continuously to distribute
the heat fairly within the mass and to avoid the burning of the flour.
In the process, the starch within the flour undergoes partial gelatinization, a process during which the starch granules absorb water, swell and undergo structural changes. Gelatinization of starch begins at a temperature of about 60°C and is complete at about 73-81°C and is followed by drying from a moisture content of about 45 to 10-15%. The finished product is crispy with a yellowish-brown color and a characteristic taste.
Estimation of the Material Losses and Gari Recovery
Gari processing, as described above, involves both mass and heat transfer.
Material losses during gari processing may be defined as the mass transfer from
the original product during processing. It occurs at all stages of processing
in the form of dry matter losses (peels, dough, fibre, etc.) the removal of
liquid starch, water and moisture evaporation at the different stages. The loss
at each stage of processing was found by finding the difference in the weight
of material recovered before and after the stage. This was then calculated as
a percentage of the initial weight of the fresh cassava. On this basis, the
individual losses were calculated.
The test crops consisted of a local cassava variety and different improved
varieties harvested at different ages of maturity (10, 12 and 14 months) from
the Asuansi Research Station in the Central Region of Ghana. The experiments
were conducted in January, 2004 and March, 2005 in the Agro-Processing laboratory
of the School of Agriculture, University of Cape Coast. The main limitation
to this experiment was our inability to obtain all the varieties at all the
different ages for testing since, the crops were not directly under our control
and were also cultivated for other research interests.
The following equipment were used to estimate the material losses and gari
yield from the various cassava varieties.
||A 3.5 h.p. engine-driven cassava grater (MK1 model) with a capacity of
up to 1 t harvest-1. Designed by the Postharvest Engineering
Unit of the International Institute of Tropical Agriculture (IITA), Ibadan,
Nigeria, this model has been widely promoted in West Africa among IITA partners
as a better alternative to the local models used throughout West Africa.
The grating surface is a perforated metal sheet wrapped on a cylindrical
drum. A unique feature of this design is the oval-shaped hopper to reduce
||A double screw press
||A rectangular wooden box sifter
||An insulated-walled chimney stove with an open iron pan on the fire box
Peeling Losses (Lp)
Let, W1 is the initial weight of fresh cassava tubers, W2
is weight of peeled tubers. Then, W1-W2 represents the
Grating Losses (LG)
Let, W3 is weight of the collected wet dough from the grater.
Then, W2-W3 represents the grating losses:
Fermentation/Dewatering Losses (LD)
Let W4 is weight of dough after fermentation and dewatering.
Then, W3-W4 represents the fermentation/dewatering losses:
Sifting Losses (LS)
Material losses during sifting are mainly due to spillage, the residual
fibre and the ungrated masses that are retained over the sifter. These may be
given out as animal feed. Let, W5 is weight of the sifted flour.
Then, W4-W5 represents the sifting losses.
Roasting Losses (RL)
Material losses encountered at the roasting stage include evaporation of
moisture into the atmosphere as well as spillage of particles as the operator
stirs through with a portion of calabash. Let, W6 is weight of roasted
flour (gari). Then, W5-W6 represents the roasting losses.
Determination of the Percentage Gari Yield or the Garification Rate
The total loss for each variety was obtained by adding all the losses in
Eq. 1-5. The total amount of gari obtained,
expressed as a percentage of the fresh cassava roots is thus equal to 100 minus
the total percentage losses for each cassava variety. The mean values of the
losses were also determined.
From the results obtained a relationship between the initial weight of fresh
cassava tubers and the yield of gari was established from the relation:
|| Weight of gari yield
||Initial weight of cassava tubers
||Constant (garification rate)
From knowledge of the garification rate associated with a given processing system, an entrepreneur can estimate gari yield from a quantity of fresh cassava roots purchased.
RESULTS AND DISCUSSION
The percentage material losses at the different stages of processing and the final gari yield from the different varieties harvested at different ages are shown in Table 1. Average peeling loss was 27.87%, grating loss 3.95%; fermentation/dewatering loss 24.42%; sifting loss 2.37% and roasting loss 18.29%. The total loss was recorded to be 76.90% with average gari yield of 23.10% of the initial weight of the fresh tubers.
The general trend also showed that older cassava roots gave higher gari yields than younger roots. The different varieties also showed difference in the gari yields.
The results from the above experiment give an indication that age and variety
have influences on the yield of gari. Comparing those same varieties harvested
at different ages, it was noted that UCC 504 harvested after 12 months yielded
a higher percentage (20.60%) than the same variety harvested after 10 months,
which yielded 18.24% of the fresh tuber weight.
material losses for the different unit operations for different cassava
varieties harvested at different ages|
|Source: Boahen (2004) and
A similar trend was observed for UCC 505 harvested at 12 and 10 months, respectively
and for varieties UCC 007, UCC 11 and UCC BB harvested at 14 and 12 months,
respectively. This general trend may be accounted for by the higher dry matter
accumulation and lower moisture contents in older crops. The only variety which
deviated from this trend was UCC 506, where the 12 month old crop yielded higher
gari than the 14 month old crop. This could be because UCC 506 may have over
matured by 14 months. According to Opara (1999), during
growth season, both root and starch production increase rapidly to their maximum
value and they decline afterwards.
The material losses also varied across the stages of operation as indicated in Table 1. On the average the greatest losses were recorded at the peeling, dewatering and roasting stages while grating and sifting losses were the lowest recorded. Peeling losses averaged about 27.87%; grating losses averaged 3.95%; dewatering losses, 24.42%; sifting losses, 2.37% and roasting losses 18.29%.
Peeling basically requires the removal of the outer cortex which contains a
high concentration of HCN, the stumps, the tapering ends and very often, rotten
portions. But alongside, some desirable dry matter is lost. The amount of these
materials lost therefore depends largely on the cultivar, the quality of the
fresh cassava tubers and the peeling efficiency. The peels alone make up about
10-15% of the fresh weight but according to Opara (1999),
hand peeling losses average 25-30%, with higher mechanized peeling losses of
Losses occurring in the mechanical grater are attributable to spillage from the hopper as a result of vibrations. It also includes dough held between the jagged ends of the perforations, those sticking to the surface of the chute and lumps of ungrated cassava locked up in the corners of the machine. Firmly bolting the grater on the foundation floor to reduce vibrations or appropriate design of the hopper to contain the materials would provide a measure of control over greater losses.
Material loss at the dewatering stage is mainly the removal of HCN-dissolved
water and starch from the product. The quantity of water expressed depends on
the moisture content of the cassava, which is also a function of its age and
variety. Varieties of lower moisture contents tend to record lower losses at
this stage. All things being equal, older crops with higher dry matter accumulation
should also have lower dewatering losses. Following this rule, one would expect
dewatering losses for the younger crops to be significantly higher than older
crops. But the results showed consistently lower dewatering losses for varieties
harvested at older ages than their corresponding varieties harvested at younger
ages. This inconsistency may be explained by the different seasons of harvest.
As noted by Hahn (2006) during the early rainy season,
the dry matter content of roots is usually lower than in the dry season.
Sifting losses are more associated with the operator skill to reduce spillage whilst roasting losses also depend on the moisture content as well as operator skill.
The mean gari yield associated with the test processing system ranged from about 21-24% of the initial weight of fresh tubers, averaging at about 23%. This translates into an average garification rate of 0.23. Since, different cassava varieties harvested at different ages were passed through the same system of equipment and the same processing method, it may be assumed that the effects of varietal differences and age of the cassava roots on the garification rate (K) were neutralized by each other so that the average garification index of 0.23 can be said to be representing the efficiency of the processing system. All other factors being equal, this factor K can be said to be an attributable property of the processing equipment and may be used as a numerical index to compare the efficiencies of different processing systems in terms of gari yield; that is, the higher the garification index K, associated with a system, the higher the potential gari yield.
It has been established from this study that three main factors account for
the material losses and consequently, the expected gari yield from any given
set of processing equipment and method. They are the varietal characteristics,
the harvesting age and season and what has been established as the garification
rate or index which is an inherent property of the processing system.
||Different varieties have different dry matter and moisture contents and
hence different amount of peels, starch and moisture that are removed from
the roots during processing into gari. Moisture content also depends on
the age and season of harvest. At high moisture content more water will
have to be expressed reducing the remaining useful weight. Generally, older
cassava roots yield a higher percentage of gari than those harvested too
||Material losses are highest at the peeling, dewatering and roasting stages
and therefore, improvement of the gari yield should target the development
and use of cassava varieties with lower peels and moisture content
||The garification rate associated with a set of processing equipment provides
a bench mark for comparing the efficiencies of different processing equipment.
It may be improved by detailed analysis of the various unit operations for
the material losses in order to reduce the losses at the critical stages
Special thanks go to the Agricultural Engineering Department of the University of Cape Coast for granting financial assistance from its allocation of the 2004/2005 Academic Facility User Fees (AFUF) to purchase raw cassava for the project. We sincerely appreciate the assistance of Mr. J. Dadzie, Mr. K. Conduah during the experiments.