Abstract: This study concerns waste processing by composting in Lomé (Togo). It presents the composition of this waste, the treatment by composting of this waste recovered in the centers of transit and collected within individual households, an evaluation of its use as a fertilizer and finally a market research to determine its economic value. The composition of the Municipal Solid Waste showed that even though mineral components like sand and gravel represent the most important part of the waste arriving at the final dump location, organic material makes up a large part of households wastes. In order to determine the optimal conditions for composting biodegradable materials and for the overall process, a study was launched at a transfer site within a district of Lomé. Four types of compost were investigated, two with organics collected directly in households and two with raw waste stored at transfer sites that had been collected by a non-governmental organization in charge of primary collection. Amendment of the compost with natural fertilizer, natural phosphate and chicken manure was also tested. It was found that the quality of the types of compost was quite similar except for the one in which natural phosphate and manure had been added. Studies on the toxicity of the types of compost on their agricultural effectiveness and their economic value were also carried out. The market research demonstrated that the production costs of the compost were low enough to make it an attractive and potentially profitable alternative to existing fertilizers such as manure.
INTRODUCTION
Twenty years ago, Lomé, the capital of Togo, was a pleasant city that was considered to be an attractive destination for international tourists. Unfortunately, recent political and socio-economic turmoil within the country have affected Lomé negatively. Rural depopulation has occurred at a rate that is among the highest in African countries (Parrot et al., 2009). In addition, over the past twenty years Togo has experienced a reduced economic growth and a decline in the Gross Domestic Product (GDP) per capita (IMF, 2003). There is an interest, therefore, in activities that might contribute to the economic development of newly urbanized areas within Togo. The creation of solid waste is unavoidable consequence of production and consumption activities of society. In developing countries the waste generation ratio of households varies between 0.4-0.7 kg/cap/day, but seasonal variations can influence the amount of waste generation and its physical and chemical characteristics. For instance, high humidity during the rainy season or the consumption of vegetables and fruits can influence the moisture content and organic content of Municipal Solid Waste (MSW) (De Guardia et al., 2010; Lin, 2008). Variations in these characteristics can in turn affect whether treatment processes such as recycling or composting can be effective (Tinoco et al., 2009). At the present time, recycling of materials collected door-by-door by either municipal or non-governmental organizations (NGOs) is not done on an organized basis but is instead done by individual scavengers and waste pickers. In Lomé issues related to Solid Waste Management (SWM) have been neglected for decades and the economical stagnation of the country, with resulting low investment in the SWM sector is seen as a serious problem. As a result the inhabitants of Lomé are faced with the problem of uncontrolled disposal of solid wastes close to their households. Faced with similar problems in other developing nations, NGO members, gardeners and scientists in India (Zurbrugg et al., 2004) or Bangladesh (Zurbrugg et al., 2005) have investigated the possibility that composting of MSW that can improve the functioning of the solid waste management system while producing a marketable product.
Here we investigate the efficacy of composting in Lomé, Togo. The idea to be investigated is whether compost can be produced in Lomé in a sustainable and economical manner and whether the compost produced could be of a quality that allows it to be used as an alternative fertilizer to animal manure. If so, then this compost produced in the urban zone could be transferred to agricultural areas adjoining Lomé. It is expected, based on local conditions, that composting would best be carried out in decentralized systems. If done properly, the use of compost could result in a variety of environmental benefits. Composting organic materials that have been diverted from landfills avoids the eventual production of methane (Domingo and Nadal, 2009; Rudrum et al., 2002) and leachate (Kjeldsen et al., 2002) in the landfills while producing a product with economic value. Furthermore, using compost can reduce the agricultural demands for water, fertilizers and pesticides (Hoitink et al., 1997; Weltzien, 1989; Sahin and Bolukbasi, 2009). Composting also extends the life of the municipal landfill by diverting organic materials from landfills to agricultural lands.
To gather information on different parameters of the composting process, this study has been set up to control and analyse two boxes and two windrows of different compositions of compostable materials. Chemical analyses of the compost were followed during the time of the aerobic decomposition of the waste. The finished compost was then evaluated as a fertilizer by performing growth studies on carrots. In addition, the finished compost was evaluated to estimate its economic value and the potential that its production could be done profitably.
MATERIALS AND METHODS
Origin of waste and sampling: Domestic waste samples were collected from 33 houses randomly identified in the middle District of Lomé after an investigation conducted by the NGO in charge of waste collection in the District. The 33 houses were selected in order to sample wastes from households having a range of economic standing. Two collection methods were utilized. In the first, only compostable wastes were collected from the household. In the second, raw waste was collected and then sorted to eliminate materials not suitable for composting (papers-cardboards, textiles, plastics, glasses, metals, miscellaneous, hazardous, etc.). The samples for physical characterization were obtained in both the dry season and in the wet season; For the selective collection, a mass of 250 kg waste was characterized; for the raw waste, a mass of 500 kg was collected and characterized. One of the samples obtained by selective collection was amended with chicken manure and organic phosphate. Sorted wastes were obtained from two different districts within Lomé. The four types of composts to be evaluated were prepared according to the following combinations of raw materials (Table 1).
Goal of this study was to compare the effect of manure, compost made from waste, manure and compost made from household waste alone. Very often, farmers use this kind of manure as organic fertilizer in agriculture without any preliminary treatment. Consequently, a considerable emission of harmful gases is released into atmosphere (greenhouse gases), nitrogen losses from manure are large and contamination of soil with pathogens is possible. Manure handling, storage and disposal continue to present major problems for poultry producers throughout the world (Petric et al., 2009).
Compost C and D are drawn because of the waste composition of various neighborhoods in terms of quality of fermentable matter.
Waste of compost A and B are from the collection of neighborhood Agbalépédogan. The compost activation process is performed by micro natural organisms.
Windrow and box preparation: After the sorting of the waste to remove unsuitable materials (metals, glass, miscellaneous and hazardous), the remaining organic waste was poured into buckets to build the compost pile. Two composting methods (box and windrow) were utilized. These two methods are described below.
Box composting method (or reactor composting method): In this method all the operations were conducted under a roof to protect the compost from excessive rain and sun (Zurbrugg, 2003). Compostable materials were placed in boxes that were 1.2 m long, 1.2 m wide and 0.5 m high. The boxes were made of cement and were constructed to allow air penetration on their sides. The bottom of the reactor was designed with a slope to facilitate leachage drainage.
Windrow composting method: The organic wastes were piled on the ground after sorting and were protected from excessive sun and rain by a composting fleece, permeable to air but not to rain water (Zurbrugg et al., 2005). A drainage system collected leachate and rainwater, which was used for watering of the windrows.
| Table 1: | Composition of different biodegradable materials |
| Compost A = Organic fraction of the waste collected directly from houses, Compost B = Organic fraction+Chicken manure+natural phosphate, in the proportion of 15/3/1, Compost C = Domestic raw waste after sorting undesirable (unwanted waste) from Bè Avéto District, Compost D = Domestic raw waste after sorting undesirable from Agbalépédogan District | |
The dimensions of the windrow were 1 m long, 2.5 m wide and 1.7 m high. These methods are less costly than other composting technologies, such as in-vessel composting (Van Haaren, 2009).
Turning frequency: The boxes and windrows were turned every 2 weeks to allow the aeration of the pile, the homogenization and also the cooling of the waste during aerobic degradation. In normal windrow composting practice, oxygen may not penetrate throughout the body of the windrow. Therefore, some anaerobic reaction may take place, resulting in methane formation. However, with adequate turning, the amount of methane generated in windrows is very small (Van Haaren et al., 2010).
Compost control parameters
Temperature: The temperature was determined every day with an alcohol thermometer
at different points of the pile (deep, middle and bottom) (Waste
Concern, 2006; Unmar and Mohee, 2008).
Moisture content (H%): The moisture content of sample (m0 = 100 g) was determined at 105°C in an oven (Yobouet et al., 2010). An approximate method was also used in the field in which a handful of compost was firmly squeezed. If no drops of water emerged, then the moisture content of the waste was considered to be too small. Moisture content in the field was adjusted by watering:
| (1) |
where, M0 = weight of sample (100 g), M1 = weight of sample after drying at 105°C
Organic matter content: To measure organic matter content, 25 g of compost were burned at 550°C in an oven for 2 h (Unmar and Mohee, 2008; Sutherland, 1998). The content in organic matter or in volatile solid was obtained by the difference of weight between the mass of the dry waste and the mass of the burned waste.
pH measurement: Composts pH was measured in a 1:5 (w/v) composts ratio to distilled water. To measure pH, 20 g of dry matter were mixed with 100 mL of distilled water. The suspension was homogenized by magnetic stirring during 15 min. The pH was measured directly by reading using a pH-meter with a combined glass electrode (Yu and Huang, 2009; Belyaeva and Haynes, 2009).
Other compost quality criteria: When collecting a sample from a decomposing pile, one has to take into account that the material is highly heterogeneous. Recognizing this, samples were collected by taking several handfuls of material from different places within the pile. These samples were then mixed again and one kilogram of material was collected from the mixture.
Nutrient and contaminant analysis: The preparation of the sample for the analysis of all parameters except nitrogen and carbon contents consisted of a wet digestion in acidic conditions (HCl/HNO3).
| • | Total phosphorus was determined by color spectrometry using ammonium molybdate and ascorbic acid |
| • | Total Kjeldahl Nitrogen (TKN) was determined by the Kjeldahl method (Barrena et al., 2010). |
| • | Cationic species (Na, K, Mg and Ca) and heavy metals (Pb, Ni, Cd) were determined by Atomic Adsorption Spectrometry AAS (Adouby et al., 2007; Bustamante et al., 2008; Belyaeva and Haynes, 2009; Koffi et al., 2010) |
Germination Index (GI) test: The biomaturity test was conducted with a fresh water extract from the compost that was dropped into a plastic petri dish with a filter paper. Ten milliliter of diluted compost extract was put on each petri dish. Twenty corn (Zea mays) seeds and twenty bean (Virgna unguiculata) seeds as basis cultures in Togo and twenty cress (Lepidium sativum) L. seeds were distributed on the filter paper and incubated at ambient temperature (28°C) in the dark (Pourhadian and Khajehpour, 2010) for 48 h under cover (Chikae et al., 2006). The numbers of germinating seeds were counted and the lengths of the roots were measured. For the control, 10 mL of distilled water was used rather than compost extract. The GI was calculated by the formula of Zucconi (Bustamante et al., 2008; Zucconi et al., 1981):
Two treatment levels (GI50 and GI75) were used in which the compost extract made up either either 50 or 75% of the sample. The same water used for the control was used for these dilutions. The Germination Index (GI) was defined as the arithmetic average of the 50% (GI50) and 75% (GI75) treatment levels:
| (2) |
Agricultural study
Field test: The composts A, B and C were tested in comparison with
chicken manure and artificial fertilizer (NPK). All the trials were conducted
in the same field. For each test a plot size of 2.8x1 m was used. Within the
plot an area of 25x1 cm was sown with 3 or 4 carrot seeds. The following manure
treatment levels were used:
| T0 | : | Natural soil with no application |
| T1 | : | Natural soil with chicken manure at a dose of 20 t ha-1 |
| T2 | : | Natural soil with inorganic fertilizer (N15P15K15) |
| T3 | : | Natural soil with inorganic fertilizer (N30P30K30) |
| T4 | : | Natural soil with inorganic fertilizer (N60P45K45) |
A, B and C are Natural soil with composts A, B, C at a dose of 20 t ha-1.
The carrot seeds were allowed to germinate and grow for a period of 90 days in the field. The field was then harvested and the total mass of carrot tubers for each treatment level was calculated.
Statistical analysis: Standard errors for the field tests were calculated using general statistical methods like the ones described by Salant and Dillmann (1994) and Rea and Parker (1997).
RESULTS AND DISCUSSION
Physical characterization of collected waste: Not surprisingly, the composition of the wastes collected directly from the households was found to be significantly different from that of the non-sorted raw waste (Fig. 1a, b). There were also some differences in composition observed between the dry and rainy seasons. For the waste collected directly from the homes, 66-75% was compostable, 20-30% was non-compostable and less than 10% were fine grain (<20 mm) material (Fig. 1a). For the waste collected directly from the households, a high organic matter of 70 to 80% was found. Relative moistures of 50 to 70% were also observed. For the raw waste collected at the final discharge site, the results of the characterization of two seasons (wet and dry) revealed a rate of 15-22% of compostable fraction, 30-32% of non-compostable waste and a high proportion of fine fraction 46-56% (Fig. 1b). The average humidity ranged from 15% in the dry season to 44% in the wet season. Organic matter represented an average of 24-25% of municipal waste with a rate of 8-9% organic matter in the fine fraction. This is not advantageous for composting of raw waste as it indicates a high percentage of non-compostable mineral materials (e.g., sand and gravel).
The results revealed that the waste composting closer to the households or in the neighborhoods could be more effective than on the centralized final disposal as a higher percentage of the material was compostable.
Follow-up of process parameters
Moisture content (H%): All the substrates in boxes or in windrows
were always controlled and the moisture content was maintained at 40-50% during
the process. Less watering was needed during the rainy season to maintain the
optimal water content. The Table 2 show water requirement
in different seasons.
| Fig. 1: | Composition of waste after (a) collection directly from household and (b) final discharge |
Temperature: Daily temperature recorded during the composting process clearly showed the two commonly-seen composting phases: the thermophilic phase (T>50°C) for the boxes (Fig. 2a, b) and T>60°C for the windrows (Fig. 2c, d)) and the mesophilic phase (T<40°C) for the boxes (Fig. 2a, b) and T = 40-50°C for the windrows (Fig. 2c, d). As expected, turning the compost generally had the effect of beginning a phase of temperature increase as fresh organic material and oxygen was mixed into the pile the pile temperature with occasional turning (Hassen et al., 2001). The maximum temperature seen in the boxes was 50-60°C (Fig. 2a, b). A maximal temperature of 60-70°C was observed in the windrows (Fig. 2c, d). Present results are similar to those reported by Chikae et al. (2006). Composts C and D reached the temperature of the environment after 65 days (Fig. 2c, d).
Evolution of pH according to the organic matter content: The pH values ranged from 8 to 9.6, waste collected from the households (A, B Table 3) contained a quantity of ashes from basic wood or coal representing almost all the sources of energy in households. Three of the four composts recorded a pH above 9 which can induce the volatilization loss of ammonia as the acid/base equilibrium shifts from NH4+ to NH3 (pKa = 9.2) (Fadiran and Dube, 2009). Compost B was the only one in which the pH was below the ammonia pKa, with a pH at 8.7 (Table 3). Changes in the pH through time are a function of the fluctuating alkalinity during the composting process (Sawyer and McCarty, 1978; Komilis and Ham, 2006). A possible reason for this effect may be that a quantity of ashes from basic wood or coal increased the pH level in the composting, thus partly counteracting the toxic effect of low pH. Organic acids have been shown to be more toxic at an initial lower pH value (Yu and Huang, 2009; Beck-Friis et al., 2003).
The organic matter decreases with increasing composting time (Table 3), as expected. The decrease is related to the aerobic decomposition of organic matter into CO2 and H2O.
| Table 2: | Water requirement in different seasons |
| Table 3: | Evolution of pH and Organic Matter (OM) in the boxes and in the windrows every two weeks |
| nd: not determined | |
| Fig. 2: | Temperature versus time in boxes and windrows (arrows indicate the turning of waste). (a) Compost A, (b) Compost B, (c) Compost C and (d) Compost D |
Mass balance: Mass balances for the waste management processes were different for composts A and B, where the compostable waste was taken from the households and samples C and D, in which waste was first sorted and then composted (shown schematically in Fig. 3). Composts A and B used the box compost method while samples C and D were composted in windrows. In samples C and D, the collected material included recyclable materials like metals and plastics which could be further sorted. After what the remaining materials could be checked for some unwanted components and could be reintroduced in the process. In general the compost mass decreased in time as the moisture and organic matter contents decreased as a result of the composting process.
Chemical compost quality: Chemical characteristics of the four compost samples were found to be similar to values available in the literature. The four compost samples (A, B, C, D Table 4) were compared to the chemical characteristics of a compost from Nan-Tzu District in Taiwan (Lin, 2008) and Bangladesh (Waste Concern, 2001). pH levels were generally higher than the pH values measured in the Guinea and Bangladesh composts (Table 4). The nitrogen content was slightly lower than the other composts, which as mentioned earlier could be the result of ammonia volatilization associated with pH values above the NH+4/NH3 pKa value (Fadiran and Dube, 2009; Chang and Chen, 2010). Carbon and organic matter contents were similar to the corresponding values taken from the literature. The carbon to nitrogen ratio (C/N) because of the relatively low nitrogen content, was therefore high in comparison with other domestic waste composts (Lin, 2008; Waste Concern, 2001; Chang and Hsu, 2008; Chang and Chen, 2010). The elevated phosphate rate in compost B is due to the addition of natural phosphate to the organic fraction at the beginning of the process. All the composts were found to have a high content of inorganic nutrient like phosphate, Na2O, MgO, K2O or CaO (Table 4) which is an advantage for the use in agriculture.
| Fig. 3: | Solid waste mass balance for the four compost samples. T0 = natural soil with no application, T1 = natural soil with chicken manure at a dose of 20 t ha-1, T2 = natural soil with inorganic fertilizer (N15P15K15), T3 = natural soil with inorganic fertilizer (N30P30K30) and T4 = natural soil with inorganic fertilizer (N60P45K45), A, B, C = natural soil with composts A, B, C at a dose of 20 t ha-1 |
| Table 4: | Chemical composition and pH of dry matter of 4 composts A, B, C, D in comparison with others MSW composts in equatorial areas |
| *Lin. (2008), **Waste Concern (2001), nd = not determined | |
The content of organic matter, which was approximately 30%, would be very helpful for water retention in amended soils.
Heavy metal (Pb, Ni, Cd) concentrations were measured in all four compost samples and compared to standard values for composts (Table 5). It can be observed that the lead contents are above the standards fixed by the French Norm NFT 44051. It is troubling that lead in these concentrations was found in these composts. The heavy metal content in food products is low, especially for those heavy metals that are not essential to plant nutrients, such as Pb and Cd. The heavy metal content in the organic fraction most probably originates from the layer of humus that is found in the topsoil in the household gardens (Veeken and Bert, 2002; Nilsson, 1972; Coughtrey at al., 1979). The organic fraction contains 11% of metals on dry weight of household waste with a concentration going to 154 mg kg-1 of Pb.
Germination test: The mean values of ten replicate germination tests were calculated for each of the four, composts for three different plants (Table 6). The results of the germination tests indicate the maturity of the composts and the lack of any compounds that would be toxic to the plants or interfere with seed germination.
A Germination Index (GI) of 50% has been used as indicator of phytotoxin-free composts (Zucconi et al., 1981). For the four composts and the three plants tested, the GI varied from 75 to 87%. The lowest GI occurred for compost B, which was the household waste sample amended with chicken manure. The relatively high observed GI values indicate a very low content of the toxic substances such as the acetic acid. The acetic acid can completely inhibit the germination of cress seeds starting from a concentration of 300 mg kg-1 (De Vleeschauwer et al., 1981). The reduction of the germination of seeds by the acetic acid was reported for cress and corn (Keeling et al., 1994). The high observed GI values for these composts indicate that the composts produced are stable and mature.
Agricultural value of the compost: The produced compost brought a significant improvement to the growth of carrot tubers (Fig. 4). For instance, the field test without fertilizer produced 1500 kg ha-1 of carrot tubers, while composts A, B, C had carrot productions of 2699, 2998 and 2685 kg ha-1, respectively.
| Table 5: | Concentration of some metallic pollutants in the different composts |
| *AFNOR 2005 | |
| Table 6: | Mean values of germination index, GI |
| Fig. 4: | Yield of carrot tubers |
| Table 7: | Labor cost estimates for compost production |
No statistically significant difference was found between carrot tuber productions for fields treated with compost as compared with the field treated with chicken manure (Fig. 4). Fields treated with artificial fertilizer (treatments T2, T3 and T4, Fig. 4) produced higher carrot tuber production.
Compost market and its selling price : Studies conducted by Kessler (2004) and a market study showed that gardeners in Lomé and farmers in towns surrounding Lomé need fertilizer for their fields but find the cost of artificial fertilizers or natural fertilizers such as chicken manure to be too high. Given the agricultural value observed for compost, these farmers and gardeners represent a large potential market for the compost. To investigate whether compost could be produced profitably we assume that a 20 kg bag of compost could be sold for 700 CFA (1.1 euro) francs (the local currency abbreviated FCFA), which is approximately 40% less than the typical cost for 20 kg bag of chicken manure (1200 CFA (1.8 euro) francs per 20 kg bag). At this price for compost a ton of compost would fill 50 bags and produce a revenue of 35,000 FCFA/ton (53 euros/ton). We estimate the potential compost production rate and associated costs. Four composters working 6 h per day and 6 days per week can sort 52 tons of raw waste and produce 11 tons of compost per month. These 11 tons of compost would give a montly revenue of of 385,000 FCFA (588 euros). Assuming that production can occur year-round, this gives a yearly revenue for the sale of compost of 4,620,000 FCFA. Estimated labor costs (Table 7) for the composters and the associated support staff are 3,050,000 FCFA (4,656 euros) per year.
Based upon these cost and revenue estimates, four full-time composters with their support staff could produc an annual profit of 1,570,000 FCFA (2,397 euros) through the sale of compost. Other possible revenues that might also result include the sale of recyclable materials, household participation in the collection, the grants of private vs. public, the environmental service to the community and potential carbon credits. On the other hand costs not considered in this analysis include maintenance of the site and facilities, purchase of consumables, expenses related to water and energy, wages and overhead management, taxation, marketing, sales, etc.) and costs investment, often involving the depreciation for depreciable property (site development, purchase of mobile equipment and transfer fees for feasibility studies and engineering). If these costs can be kept below 34% (1,570/4,620) of total costs, or if the additional revenues can offset some or all of these additional costs, then the compost could be produced profitably.
CONCLUSION
The physical characterization showed that a majority of wastes (70% on average) collected directly from the households was biodegradable. The composts produced using two different candidate collection and processing methods do not present risks, aside from lead concentrations which exceeded the French standard AFNOR for compost. The results of an agricultural study indicated a good opportunity for the composting of urban waste directly collected in the households in different districts of Lomé. Previous studies demonstrated that a demand exists for an economical fertilizer in the area surrounding Lomé. The compost had a benefit that was equivalent to an alternative fertilizer (chicken manure) and could be produced profitably at an assumed price that would be significantly less than the manure.
ACKNOWLEDGMENTS
The work presented in this study was financially supported by the Service of the French Cooperation, with the CORUS research programme.