- Email: [email protected]
Performance of a domestic cooking wick stove using fatty acid methyl esters (FAME) from oil plants in Kenya
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
Available at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Performance of a domestic cooking wick stove using fatty acid methyl esters (FAME) from oil plants in Kenya Agatha W. Wagutu a, Thomas F.N. Thoruwa b,*, Sumesh C. Chhabra a, Caroline C. Lang’at-Thoruwa a, R.L.A. Mahunnah c a
Department of Chemistry, Kenyatta University, P.O. Box 43844-0100, Nairobi, Kenya Department of Energy Engineering, Kenyatta University, P.O. Box 43844, Nairobi, Kenya c University of Dar-es Salaam, Muhimbili College of Medicine, P.O. Box 53486, Dar-es Salaam, Tanzania b
article info
abstract
Article history:
With depletion of solid biomass fuels and their rising costs in recent years, there has been
Received 7 March 2008
a shift towards using kerosene and liquefied petroleum gas (LPG) for domestic cooking in
Received in revised form
Kenya. However, the use of kerosene is associated with health and safety problems.
21 March 2010
Therefore, it is necessary to develop a clean, safe and sustainable liquid bio-fuel. Plant oil
Accepted 22 March 2010
derivatives fatty acid methyl esters (FAME) present such a promising solution. This paper
Available online 7 May 2010
presents the performance of a wick stove using FAME fuels derived from oil plants: Jatropha curcus L. (Physic nut), Croton megalocarpus Hutch, Calodendrum capense (L.f.) Thunb., Cocos
Keywords:
nucifera L. (coconut), soyabeans and sunflower. The FAME performance tests were based on
Kenya
the standard water-boiling tests (WBT) and compared with kerosene. Unlike kerosene all
Jatropha carcus L. (Physic nut)
FAME fuels burned with odorless and non-pungent smell generating an average firepower
Croton megalocarpus Hutch
of 1095 W with specific fuel consumption of 44.6 g L1 (55% higher than kerosene). The flash
Calodendrum Capense (L.f.) Thunb.
points of the FAME fuels obtained were typically much higher (2.3e3.3 times) than kero-
Cocos nucifera L. (coconut)
sene implying that they are much safer to use than kerosene. From the results obtained, it
Soya bean
was concluded that the FAME fuels have potential to provide safe and sustainable cooking liquid fuel in developing countries. ª 2010 Elsevier Ltd. All rights reserved.
1.
Introduction
With depletion of solid biomass fuels (wood and charcoal fuels), their rising costs in recent years [1e4] and their effects on indoor pollution [3e8], there has been shift to use kerosene and liquefied petroleum gas (LPG) for domestic cooking and heating applications [9]. In fact, more than 94% of Kenya’s population uses kerosene for cooking and/or lighting applications [10]. The kerosene wick stove and lamps are widely used in peri-urban and urban areas while LPG is widely used in upper and middle class income homes. However, the use of
kerosene is associated with health and safety problems leading to respiratory tract infections and pulmonary diseases [11]. Fires resulting from kerosene stove incidents in urban slum dwellings have been rising claiming lives and leaving hundreds of thousands homeless. In South Africa, it is claimed that 46 000 domestic kerosene stove fires occurred in 2000 leaving 50 000 people severely burned and 100 000 homes destroyed [12,13]. Therefore, it is necessary to develop clean, safe, affordable and sustainable alternative liquid bio-fuels. Fatty acid methyl esters (FAME) derived from the following Kenyan oil plants present such a promising solution: Jatropha
* Corresponding author. Tel.: þ254 28710901/12; Cell: 254 725 986 248; fax: þ254 20 8711575. E-mail addresses: [email protected], [email protected] (T.F.N. Thoruwa). 0961-9534/$ e see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2010.03.016
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
carcus L. (Physic nut); Croton megalocarpus Hutch; Calodendrum capense (L.f.) Thunb., Cocos nucifera L. (coconut), sunflower and soybeans. In this paper, the performance of the Chinese brand wick stove model 62, loaded with the FAME fuels was investigated and compared with petroleum kerosene in terms of thermal transfer efficiency, specific fuel consumption and firepower. Experiments were based on standard water-boiling test (WBT) Version 3.0 [14]. The test mimicked rough simulation of cooking process consisting of three phases: a high-power phase from a cold start, a high-power phase from a hot start, and a low-power (simmering) phase.
1.1.
Liquid fuel cooking stoves
Liquid fuels can be burned in either wick stoves or pressure stoves. In this study, a Chinese brand model 62 wick stove (Fig. 1) was chosen for the experiments because it is the simplest and cheapest wick stove in the Kenyan market, costing only USD5. The stove model is widely used in East Africa to meet basic household thermal needs including cooking and heating applications [15,16]. The basic design of the wick stove is comprised of two perforated bare carbon steel cylinders positioned so that the wicks are between the two perforated cylinders to hold flames. The distance between the inner and outer flame holder is a little more than the thickness of the wicks, usually around 12 mm. The height of the cylindrical flame holders is about 8e10 cm. Wicks are fixed in a holder such that they can move up and down causing them to emerge into an annular space formed by the two perforated steel shells. The wick brings fuel to the base of the cylinders through capillary action and the radiant heat from the flame holders serves to evaporate the fuel from the wicks. The draft created by the flames draws air through the perforations in the flame holders into the annular space. The fuel vapor is then combined with oxygen flowing through the perforations; the mixture burns with a stable blue flame. Every wick stove has raised guide ribs on the inside and outside of the wicks. These guide ribs indicate the maximum height of the wicks when fully raised. Raising the wicks too high may result in the upper part of the blue flame becoming yellow, which is an indication of cracking of the kerosene in the flame. The cracking of kerosene is mostly accompanied by the production of soot, which reduces combustion quality [17].
1251
1.2. Liquid fuels for domestic cooking and heating applications Kerosene fuel is the commonly used liquid cooking fuel in most developing countries [15e19]. The liquid fuel cooking stoves are generally efficient (over 50% thermal efficiency) and cook quickly and conveniently compared to traditional wood and charcoal stoves. They are very popular in the urban areas in comparison with other biomass fuel cooking stoves. The implications of utilizing FAME fuels to replace kerosene would result in reduced health risks associated with the use of kerosene and wood fuels and possibilities of creating oil crop plantations utilizing marginal lands in Kenya (estimated to be about 80% of total land mass) and thereby reduce heavy dependency on use of solid biomass energy.
1.3.
Objective
The objective of this paper was to investigate the wick stove performance indicators of the Kenyan FAME fuels using the standard water-boiling test (WBT) and compare them with kerosene fuel.
2.
Methodology
2.1.
Extraction of oil from Kenyan oil seeds
The seeds of C. megalocarpus and C. capense harvested during 2007 were obtained from Kenya Forestry Research Institute (KEFRI) in central Kenya (00 230 S 36 560 E) while Jatropha curcus seeds harvested during 2007 were supplied by the Nairobi based Vanilla Development Foundation (VDF) from their plantations in Eastern Kenya (01 300 S 37 150 E). Coconut harvested during 2007 was purchased from a local market in Mombasa town, coastal region (04 020 S 39 430 E). Soybean and sunflower oils were bought during 2007 from the local supermarkets. The seeds were then sun-dried for 72 h and then dehulled manually and ground finely using a mortar. The samples were then soaked in n-hexane each for three days in order to extract maximum amounts of oils. The oilehexane mixture was then filtered under pressure and then separated using a rotary evaporator with the temperature of the bath set at 40 C. Recovered hexane was then returned to the cake mixture for further extraction. The oil samples were then
Fig. 1 e Chinese wick stove, wheel brand model 62.
1252
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
treated with anhydrous Na2SO4 to remove any moisture. The resulting oil was used for transesterification process [20].
2.2.
Preparation of fatty acid methyl esters (FAME) fuels
Samples of fatty acid methyl esters (FAME) were prepared using the standard method by base catalyzed transesterification process using the normal 6:1 M ratio of methanol to oil [20e22].
2.3.
Tb ¼ ð100 h=300Þ C
FAME fuel properties
The FAME thermal performance tests, the fuel properties of each FAME and kerosene fuel were analyzed using ASTM (American Society of Testing and Material) procedures for petroleum products [23,24] at the Kenya Pipeline Company Limited, based in Nairobi. Calorific values of the fuels analyzed were determined using the Gallenkamp Auto-bomb model calorimeter in the university laboratory [25]. The FAME fuel property results are summarized in Table 1 in order to explain the variability of stove thermal performance.
2.4.
Wick stove performance experiments
The wick stove performance experiments were conducted using the standard Water-Boiling Test (WBT) version 3.0 to determine stove performance indicators. The tests were in principle used to compare FAME thermal performance against kerosene fuel using the standard WBT protocol viz. highpower phase cold and warm start conditions and low-power simmering phase. The method was chosen because of its simplicity and replicability [14]. The test compared the thermal performance of 6 different FAME cooking fuels against kerosene. The WBT measured time to boil, thermal transfer efficiency, specific fuel consumption (SFC), and maximum and minimum firepower [14,26e28].
2.4.1.
thermometer at 5 min time intervals until the water was brought to boil. The thermometer was positioned in the center of the pot, roughly 3 cm above the pot bottom. The average boiling temperature was taken as the local boiling point [14]. An average of 95 C was recorded. This result was checked against the estimated local boiling temperature based on local altitude h (m) using equation (1). The two were found to coincide.
Determination of local boiling point
The tests were carried out at Kenyatta University in a Chemistry laboratory room located 25 km East of Nairobi, 1 N, 2 E, with elevation 1500 m above the sea level. The laboratory had 4 standard windows and one standard door. Only one window and the door remained open during the tests. The local boiling point temperature was determined using kerosene fuel loaded onto a wick stove. Approximately 1000 g of cold clean water was weighed and poured into the aluminum pot (Fig. 2). The ambient and water temperatures were monitored via the
(1)
For Kenyatta University, Nairobi, Kenya, the altitude h used was 1500 m above sea level.
2.4.2.
High-power cold start
The high-power cold start experiments began with loading each 500 ml test liquid fuel onto the stove at room temperature and the total weight of stove plus fuel was determined. A time duration of 30 min was allowed for the new wicks to absorb the liquid fuels. The stove was then lit with a matchstick and the water brought to boil rapidly. Fire was put off immediately after the local boiling point was reached for each test fuel. Times taken for lighting and boiling were recorded. Three minutes duration was allowed for loading fresh water onto the pot. The evaporated water and the remaining fuel were weighed. A wooden ply board was used during weighing to protect the weighing balance from heat. The weighing lasted 3e4 min. For every test done, the stove was fitted with new wicks of length 28 cm, diameter 10 mm. For every FAME fuel, triplicate experiments were performed using the standard WBT high-power cold start protocol [14] and the results are presented in Table 2.
2.4.3.
High-power phase hot start
The high-power phase hot start experiments were conducted immediately after the high-power cold start tests while the stove was still hot. The water was brought to boil as in the first phase and time taken was recorded. The fire was then put off and mass of fuel consumed and the water evaporated during this phase were determined and recorded and the results are presented in Table 3 [14].
2.4.4.
Low-power phase hot start (simmering)
The low-power phase hot start experiments were conducted immediately after the high-power phase hot start experiments. The stove was lit again and the temperature of the boiled water was determined. The remaining boiled water from the second
Table 1 e Thermophysical properties of FAME fuels and kerosene. Property Density at 20 C (kg m3) Flash point ( C) Color ASTM Viscosity at 40 C (mm2 s1) Water content (%v v1) Calorific value (MJ kg1)
Test method
Croton FAME
Coconut FAME
Capense FAME
ASTM D1298 ASTM D93 ASTM D1500 ASTM D445 ASTM D95 ASTM D 4809
880 104.5 1.4 1.0 4.2 0.3 0.05 0.2 36.28 0.14
871 106.5 0.7 0.1 2.7 0.04 <0.05 34.84 0.02
875 123.5 12 1.0 4.3 1.6 0.05 0.4 39.93 0.12
FAME: fatty acid methyl ester.
Jatropha FAME 877 147.5 0.4 4.4 0.06 40.75
7.4 0.3 0.8 0.2
Sunflower FAME
Soya bean FAME
Kerosene
883 146.5 4.6 0.4 4.6 0.3 0.05 0.14 39.29 0.2
881 136.5 3.3 0.5 4.2 0.07 0.2 0.11 37.76 0.4
788 45.5 5.4 0.1 2.4 0.01 <0.05 43.61 0.7
1253
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
Burning rate (Rc): This is a measure of the rate of fuel consumption while bringing water to a boil. It is calculated by dividing the equivalent fuel consumed by the time of the test (t) in minutes. Rc ¼
mf t
(3)
Specific fuel consumption (SCc): Specific consumption can be defined for any number of cooking tasks and should be considered “the fuel required to produce a unit output” whether the output is boiled water, cooked beans, or loaves of bread [26e28]. In the case of the cold-start high-power WBT, it is a measure of the amount of fuel required to produce 1 l of boiling water starting with a cold stove. It is calculated in this way: Fig. 2 e Chinese wick stove loaded with aluminum pot.
phase was returned to the stove to simmer for 45 min. The fire was controlled by maintaining the water temperature 3 C below the local boiling point by adjusting the wick height. The fire was put out after the 45 min. Water vaporized and the fuel consumed after the 45 min were determined and recorded, and the results are presented in Table 4 [14].
2.5.
4:186 M1 DT þ 2260 M2 mf CV
(2)
In this calculation, the work done by heating water is determined by adding two quantities: (1) the product of the mass of water in the pot (M1), the specific heat of water (4.186 J g1 C), and the change in water temperature (DT ) and (2) the product of the amount of water evaporated from the pot (M2) and the latent heat of evaporation of water (2260 J g1). The denominator (bottom of the ratio) is determined by taking the product of the mass of fuel consumed (mf) during this phase of the test and the calorific value of the fuel (CV).
mf M1
(4)
Firepower (FPc): This is a ratio of the fuel energy consumed by the stove per unit time. It tells the average power output of the stove (in Watts) during the test phase. FPc ¼
mf CV 60 t
(5)
Evaporation rate (Wc): This is a measure of the rate of water loss through evaporation during the test. Wc ¼
Stove performance indicators
Stove performance indicators were determined through calculations based on the modified version of the well-known water-boiling test (WBT) revised calculation procedure [14]. Thermal transfer efficiency (hc): This is the ratio of the work done by heating and evaporating water to the energy consumed by burning the fuel. It is calculated using equation (1) as follows: hc ¼
SCc ¼
M2 t
(6)
3.
Results and discussion
3.1.
FAME properties
Fuel properties obtained from the FAME experiments along with kerosene are presented in Table 1. Kinematic viscosity for coconut methyl ester was found to be 2.7 mm2 s1, which is 13% above that of kerosene (2.4 mm2 s1). Viscosity of the other FAME fuels was found to be 70e90% above that of kerosene. The flash points of the FAME fuels were typically much higher (2.3e3.3 times) than kerosene. This indicates that they are less flammable and therefore can be handled more safely and have the potential to reduce fire incidents such as those reported in South African slum dwellings [11e13]. The FAME fuels also recorded higher densities than kerosene. The gross calorific value of FAME fuels was found to be 12% (on average) lower than that of kerosene. The heating value depends on the composition of the fuel. Kerosene, which
Table 2 e Mean stove performance indicators, high power (cold start) phase. High power test (cold start) Time to boil Fuel consumed Burning rate Thermal transfer efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME
Coconut FAME
Capense FAME
min g g min1 % g l1 g min1 W
21.4 0.6 40.3 0.6 1.9 39% 0.6 45.8 0.6 5.4 0.08 1101
17.5 0.6 36.3 0.6 2.1 47% 41.0 0.6 6.8 0.23 1157 14
22.9 0.6 39.3 0.6 1.7 41% 1.2 45.3 1 6.1 0.19 1073 17
Jatropha FAME 21.2 36.7 1.7 42% 42.7 6.4 1148
0.6 0.6 1 0.6 0.1 5.1
Sunflower FAME
Soya bean FAME
Kerosene
24.5 0.06 40.3 0.6 1.6 38% 0.6 46.2 0.6 5.1 0.12 1042 5.0
23.4 0.6 40.3 0.6 1.7 40% 1 46.6 1 5.6 0.13 1047 13
11.6 22.7 1.9 50% 24.6 6.9 1373
0.6 0.6 0.6 0.6 0.3 15
1254
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
Table 3 e Mean stove performance indicators high power (hot start) phase. High power test (hot start) Time to boil Fuel consumed Burning rate Thermal efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME 19.5 38.0 1.9 40% 42.4 5.4 1131
min g g min1 % g l1 g min1 W
0.6
0.6 0.6 0.17 13
Coconut FAME
Capense FAME
Jatropha FAME
Sunflower FAME
Soya bean FAME
Kerosene
16.1 34.0 2.1 47% 0.6 38.2 0.6 6.6 0.18 1190 13
19.6 0.6 36.3 0.6 1.8 43% 0.6 41.4 0.6 6.5 0.18 1157 18
19.8 0.6 35.0 1.8 43% 1 39.8 0.6 6.4 0.02 1152 23
22.4 38.7 0.6 1.7 38% 43.5 0.6 5.1 0.07 1088 12
21.0 37.7 0.6 1.8 41% 0.6 42.8 1 5.5 0.03 1092 23
10.0 20.7 0.6 2.1 53% 0.6 22.1 0.6 7.0 0.06 1450 5
constitutes paraffins, olefins and aromatics tends to have much higher-energy contents per volume compared to FAME fuels [29]. It was also observed that coconut FAME fuel had the lowest heating value despite being highly saturated. This could be attributed to the short chain (C6eC14) fatty acid composition [23,30].
3.2.
Water-boiling tests performance
Generally, water-boiling tests (WBT) were conducted to demonstrate the suitability of FAME fuels as potential cooking liquid fuels targeting liquid fuel stove markets. The results show that the FAME fuels tested burnt with transparent blue flame and produced no smell apart from the sweet smell of esters and no black matter was deposited at the bottom of the cooking pot. On the other hand, kerosene fuel produced a strong smell leaving an agglomeration of black carbon deposits at the bottom of the cooking pot. The mean values of the stove parameters obtained at the end of the three WBT phases are shown in Tables 3e4.
3.2.1.
High power (cold start) phase
Various fuel properties were found to affect the stove performance in different ways: The viscosity of the fuel affected its availability to the combustion chamber. The capillary action was faster for the less viscous fuels. The coconut methyl ester had viscosity very close to that of kerosene (2.7 and 2.4 mm2 s1 respectively). The time to boil was close, 17.5 and 11.6 min respectively. The rates of burning were also very close: 2.1 and 1.9 g min1 respectively. Other FAME fuels had viscosities ranging from 4.1 to 4.6 mm2 s1 and hence took a longer time (21e25 min) than that required to boil an equal volume of water (Table 2). The specific fuel consumption and firepower seemed to be affected by the combined effects of fuel heating values and the viscosities. Low viscosity combined with high-energy content for kerosene resulted in
less fuel consumption and much shorter time to boil. The heat transfer efficiency of all the FAME fuels was found to be in the same range (38e42%) except for coconut with 47%, very close to kerosene (50%). Kerosene fuel took the least time to boil the water viz. 11.6 min consuming almost 50% less mass compared to FAME fuels to bring equal volumes of water to boil and giving the highest power output (1.4 kW). The firepowers obtained for all the FAME fuels were within the same range (1.0e1.2 kW). The results obtained show relatively consistent thermal efficiencies in the three water-boiling tests and are in good agreement with reported efficiencies [18e20].
3.2.2.
Stove performance during high power (hot start) phase
Table 3 shows the mean values of wick-stove parameters obtained at the end of high-power phase (hot start) for FAME and kerosene fuels. Slightly lesser time (2e3 min) was required to boil the water during high power hot start phase than in cold start for each fuel giving a small increase in the power output (4e6%). This can be explained by the fact that the fuel in the stove tank would be at slightly higher temperature than room temperature. This increased the flash point and lowered the viscosity and hence increased the fuel mobility by capillarity. Heat transfer efficiency remained more or less the same as in cold start, possibly because the metallic stove body conducted heat fast such that within 5 min the body would be at room temperature. The burning rate remained relatively the same as in phase 1.
3.2.3. phase
Stove performance during low-power (simmering)
The goal of low-power simmering phase was to maintain water at a high temperature using minimal stove output power for 45 min as per the WBT protocol. The power required to keep the water simmering (3e4 C) below the boiling temperature seemed to be the same for all the liquid fuels (900e910 W) possibly due to liquid fuel capillary flow
Table 4 e Mean stove performance indicators low-power (hot start) phase. Low-power test (hot start) Time to boil Fuel consumed Burning rate Thermal efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME
Coconut FAME
Capense FAME
Jatropha FAME
Sunflower FAME
Soya bean FAME
Kerosene
min g g min1 % g l1 g min1 W
45.0 67.7 0.6 1.5 42% 0.6 137.2 0.6 8.9 0.06 900 5.2
45.0 71.0 1.6 42% 0.6 147.6 2.9 9.2 0.16 907 4.6
45.0 63.0 1.4 42% 0.6 135.5 2.1 9.1 0.13 902 1.2
45.0 60.7 0.6 1.4 43% 0.6 132.8 1.5 9.3 0.02 905 6.8
45.0 63.0 1.4 41% 128.4 1.5 8.8 0.03 901 8.9
45.0 65.0 1.5 41% 132.7 0.6 8.8 0.04 901 1.2
45.0 56.3 0.6 1.3 43% 110.6 1.2 9.3 0.06 910 7.4
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
100
Temperature (degree, Celcius)
90 80 Kerosene CoconutME CapenseME SunflowerME SoyabeanME CrotonME JatrophaME Ambient
70 60 50
may be attributed to low energy content in FAME fuels (see Table 1) as compared to that of kerosene. However, slightly less time was taken to reach boiling point during high power (hot start) phase due to decreased FAME viscosity and density that enhanced capillary action since the stove was still hot during the starting period.
4.
Conclusions
40 30 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Time (minutes)
Fig. 3 e Typical temperature profiles generated during high power boiling phase (cold start).
limitations. The thermal efficiency for all the liquid fuels remained relatively the same (41e43%). In addition, the wick stove demonstrated 304% higher specific FAME fuel consumption during the low-power phase as compared to the high-power cold start phase. Interestingly, kerosene fuel demonstrated 449.5% higher specific fuel consumption during the low simmering phase compared to the high power phase. These results demonstrate that there is increased liquid fuel capillary action due to pre-heating effects resulting from combined effects of reduced FAME fuel viscosities and densities. This also implies that for conventional slow cooking tasks, the use of FAME fuels in wick stoves is comparable to slow cooking using kerosene fuel.
3.3.
(a) Fatty acid methyl ester (FAME) fuels have successfully been developed from Kenyan oil plants namely Jatropha carcus L. (Physic nut), Croton megalocarpus Hutch, Calodendrum capense (L.f.) Thunb. (Cape Chestnut), Cocos nucifera L. (coconut), Sunflower and soybeans using the standard method by base catalyzed transesterification process of the normal 6:1 M ratio of methanol to plant oils; (b) The FAME fuels tested using the standard water-boiling tests show that FAME fuels have the capacity to generate an average of firepower of 1095 W, which is 20% lower than that of kerosene. (c) Generally all FAME fuels showed an average specific fuel consumption of 44.6 g per liter of water, which was higher by 55%, compared to that of kerosene. (d) It was found that the FAME fuels had much higher flash points (104.5 Ce147.5 C) compared to kerosene whose flash point was only 45.5 C implying that FAME fuels are much safer to use than kerosene with little risk of fire hazards in domestic cooking applications. (e) On the basis of these findings, the FAME fuels have great potential to provide an alternative source of domestic cooking and heating energy for people living in developing countries.
Temperature profiles
Figs. 3 and 4 represent the water temperature profiles during high-power phase and low-power phase using various types of FAME fuels. It was observed that all FAME fuels took almost double the time compared to kerosene to attain the same boiling point temperatures during high and low phases. This 100 90
Temperature (degree, Celcius)
1255
80 Kerosene CoconutME CapenseME SunflowerME SoyabeanME CrotonME JatrophaME Ambient
70 60 50
Acknowledgement We wish to acknowledge the financial support from SIDA sponsorship through the lake Victoria Initiative (VicRes) program. We also wish to thank Kenyatta University for supporting this work through provision of laboratory space and equipment. We also wish to thank Mr. Festus K. Muchena, Chief Chemist Kenya Pipeline Company Ltd who facilitated the analysis of fuel properties at the company’s laboratory, and Vanilla Development Foundation (VDF) which provided Jatropha seeds for free. All those who supported this work directly and indirectly are greatly acknowledged.
references
40 30 20 0
2
4
6
8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Time (minutes)
Fig. 4 e Typical temperature profiles during low-power simmering phase.
[1] Wamukonya L. Energy consumption in three rural Kenya households: a survey. Biomass and Bioenergy 1995;8(6): 445e51. [2] Senelwa KA, Simis REH. Opportunities for small-scale biomasseelectricity systems in Kenya. Biomass and Bioenergy 1999;17:239e55. [3] Owino GA. Changes in household fuel and stove choices and their implications for indoor air pollution quality, climate
1256
[4]
[5] [6] [7] [8]
[9]
[10] [11]
[12]
[13] [14]
[15]
[16]
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
change and loca tropospheric chemistry. MSc thesis, Department of Environmental Science, Kenyatta University, Nairobi, Kenya; 2004. Ndiema CK. Combustion characteristics of the Kenyan biomass stoves. Unpublished PhD thesis dissertation, Department of Fuel and Energy, University of Leeds, England, UK; 1992. Kammen DM. Cook stoves for the developing world. Scientific American 1995;273(1):64e7. Charron D. The ecostove-getting rid of nearly 90% of kitchen wood smoke. Boiling Point 2005;50. ITDG, UK. Smith KR. Update on the health and climate impacts of household solid fuels. Boiling Point 2006;52:3e5. Smith KR, Mehta S. The burden of disease from indoor air pollution in developing countries: comparison of estimates. International Journal of Environmental Health 2003;206(5): 279e89. Bates E, Nigel B, Doing A, Gitonga S. Participatory approaches for alleviating indoor air pollution in rural Kenyan kitchens. Boiling Point 2002;48:12e5. Hannah I. Livelihoods in the urban biomass sector-realities and threats. Boiling Point 2003;49:21e3. Lloyd P, Heinrich G, Mulaudzi F. A comparison of the thermal efficiency of different stoves used nationally in township homes. Domestic use of energy conference. South Africa: Energy Research Center, University of Cape Town; 2004. Llyod P. The safety of paraffin and LPG appliances for domestic use. Journal of Energy in South Africa 2002;13: 56e61. Llyod P. Developing safe paraffin appliances in South Africa. Boiling Point 2009;56:7e9. Bailis R, Damon O, Dean S, Smith KR, Rufus E. The water boiling test (WBT). Household energy and health program. Shell Foundation, www.ehs.berkeley.edu/hem/protocols/ wbt-version3.0Jan2007a.pdf; 2007 [Last accessed 24.06.09]. Floor W, van der Plas R. Kerosene stoves: their performance, use and constraints, joint UNDP/World Bank Energy Sector management assistance program; 1991. Andrews G, Mkapi MC. Vegetable oil as alternative household fuel to imported kerosene in Africa. Leeds African Studies Bulletin 1996;61:48e52.
[17] Verhoeven N. Kerosene wick stoves. Master’s thesis, Faculty of Physics, Eindhoven University of Technology, Eindhoven, Netherlands; 1989. [18] Kruger GT. Firebomb kerosene stoves without social bounds. Addressing energy poverty in South Africa. In: Domestic use of energy conference. Cape Town, South Africa: The Paraffin Safety Association of Southern Africa; 2006. p. 1e7. [19] Truran G. Household energy poverty and paraffin consumption in South Africa. Boiling Point 2009;5:2e6. [20] Fungrai A, Milford A. Biodiesel production: a review. Bioresource Technology 1999;70:1e15. [21] Lang X, Dalai A, Bakhash N, Reaney M, Hertz P. Preparation and characterization of biodiesel from various bio-oil. Bioresource Technology 2001;80:53e62. [22] Schwab A, Bayby M, Freedman B. Preparation and properties of diesel fuels from vegetable oils. Fuel 1987;66: 1372e8. [23] Van Gerpen J, Shanks B, Pruszko R. Biodiesel analytical methods. IowaState University Renewable Products Development Laboratory; 2004. [24] ASTM (American Society for Testing and Materials). ASTM designation D975-92a, standard specification for diesel fuel oils. Philadelphia: ASTM; 1992. p. 310e1. [25] ASTM (American Society for Testing of Materials). Standard method for heat of combustion of liquid hydrocarbon fuels by bomb calorimeter. Philadelphia, United States: ASTM; 1998. p. 133e40. [26] Visser P. The testing of cookstoves: data of water-boiling test as a basis to calculate fuel consumption. Energy Sustain Development 2005;9(1):16e24. [27] Berrueta Vıctor M, Edwards Rufus D, Masera Omar R. Energy performance of wood-burning cookstoves in Michoacan, Mexico. Renewable Energy 2008;33:859e70. [28] Baldwin SF. Biomass stoves: engineering design, development, and dissemination. Princeton, NJ: Center for Energy and Environmental Studies; 1986. [29] Knothe G. Analytical methods used in the production and fuel quality assessment of biodiesel. Journal of American Society of Agricultural Engineers 2000;42:193e200. [30] Pryde E. Fatty acids. Champaign, Illinois: The American Oil Chemists’ Society; 1979. p. 19e23.
Available at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Performance of a domestic cooking wick stove using fatty acid methyl esters (FAME) from oil plants in Kenya Agatha W. Wagutu a, Thomas F.N. Thoruwa b,*, Sumesh C. Chhabra a, Caroline C. Lang’at-Thoruwa a, R.L.A. Mahunnah c a
Department of Chemistry, Kenyatta University, P.O. Box 43844-0100, Nairobi, Kenya Department of Energy Engineering, Kenyatta University, P.O. Box 43844, Nairobi, Kenya c University of Dar-es Salaam, Muhimbili College of Medicine, P.O. Box 53486, Dar-es Salaam, Tanzania b
article info
abstract
Article history:
With depletion of solid biomass fuels and their rising costs in recent years, there has been
Received 7 March 2008
a shift towards using kerosene and liquefied petroleum gas (LPG) for domestic cooking in
Received in revised form
Kenya. However, the use of kerosene is associated with health and safety problems.
21 March 2010
Therefore, it is necessary to develop a clean, safe and sustainable liquid bio-fuel. Plant oil
Accepted 22 March 2010
derivatives fatty acid methyl esters (FAME) present such a promising solution. This paper
Available online 7 May 2010
presents the performance of a wick stove using FAME fuels derived from oil plants: Jatropha curcus L. (Physic nut), Croton megalocarpus Hutch, Calodendrum capense (L.f.) Thunb., Cocos
Keywords:
nucifera L. (coconut), soyabeans and sunflower. The FAME performance tests were based on
Kenya
the standard water-boiling tests (WBT) and compared with kerosene. Unlike kerosene all
Jatropha carcus L. (Physic nut)
FAME fuels burned with odorless and non-pungent smell generating an average firepower
Croton megalocarpus Hutch
of 1095 W with specific fuel consumption of 44.6 g L1 (55% higher than kerosene). The flash
Calodendrum Capense (L.f.) Thunb.
points of the FAME fuels obtained were typically much higher (2.3e3.3 times) than kero-
Cocos nucifera L. (coconut)
sene implying that they are much safer to use than kerosene. From the results obtained, it
Soya bean
was concluded that the FAME fuels have potential to provide safe and sustainable cooking liquid fuel in developing countries. ª 2010 Elsevier Ltd. All rights reserved.
1.
Introduction
With depletion of solid biomass fuels (wood and charcoal fuels), their rising costs in recent years [1e4] and their effects on indoor pollution [3e8], there has been shift to use kerosene and liquefied petroleum gas (LPG) for domestic cooking and heating applications [9]. In fact, more than 94% of Kenya’s population uses kerosene for cooking and/or lighting applications [10]. The kerosene wick stove and lamps are widely used in peri-urban and urban areas while LPG is widely used in upper and middle class income homes. However, the use of
kerosene is associated with health and safety problems leading to respiratory tract infections and pulmonary diseases [11]. Fires resulting from kerosene stove incidents in urban slum dwellings have been rising claiming lives and leaving hundreds of thousands homeless. In South Africa, it is claimed that 46 000 domestic kerosene stove fires occurred in 2000 leaving 50 000 people severely burned and 100 000 homes destroyed [12,13]. Therefore, it is necessary to develop clean, safe, affordable and sustainable alternative liquid bio-fuels. Fatty acid methyl esters (FAME) derived from the following Kenyan oil plants present such a promising solution: Jatropha
* Corresponding author. Tel.: þ254 28710901/12; Cell: 254 725 986 248; fax: þ254 20 8711575. E-mail addresses: [email protected], [email protected] (T.F.N. Thoruwa). 0961-9534/$ e see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2010.03.016
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
carcus L. (Physic nut); Croton megalocarpus Hutch; Calodendrum capense (L.f.) Thunb., Cocos nucifera L. (coconut), sunflower and soybeans. In this paper, the performance of the Chinese brand wick stove model 62, loaded with the FAME fuels was investigated and compared with petroleum kerosene in terms of thermal transfer efficiency, specific fuel consumption and firepower. Experiments were based on standard water-boiling test (WBT) Version 3.0 [14]. The test mimicked rough simulation of cooking process consisting of three phases: a high-power phase from a cold start, a high-power phase from a hot start, and a low-power (simmering) phase.
1.1.
Liquid fuel cooking stoves
Liquid fuels can be burned in either wick stoves or pressure stoves. In this study, a Chinese brand model 62 wick stove (Fig. 1) was chosen for the experiments because it is the simplest and cheapest wick stove in the Kenyan market, costing only USD5. The stove model is widely used in East Africa to meet basic household thermal needs including cooking and heating applications [15,16]. The basic design of the wick stove is comprised of two perforated bare carbon steel cylinders positioned so that the wicks are between the two perforated cylinders to hold flames. The distance between the inner and outer flame holder is a little more than the thickness of the wicks, usually around 12 mm. The height of the cylindrical flame holders is about 8e10 cm. Wicks are fixed in a holder such that they can move up and down causing them to emerge into an annular space formed by the two perforated steel shells. The wick brings fuel to the base of the cylinders through capillary action and the radiant heat from the flame holders serves to evaporate the fuel from the wicks. The draft created by the flames draws air through the perforations in the flame holders into the annular space. The fuel vapor is then combined with oxygen flowing through the perforations; the mixture burns with a stable blue flame. Every wick stove has raised guide ribs on the inside and outside of the wicks. These guide ribs indicate the maximum height of the wicks when fully raised. Raising the wicks too high may result in the upper part of the blue flame becoming yellow, which is an indication of cracking of the kerosene in the flame. The cracking of kerosene is mostly accompanied by the production of soot, which reduces combustion quality [17].
1251
1.2. Liquid fuels for domestic cooking and heating applications Kerosene fuel is the commonly used liquid cooking fuel in most developing countries [15e19]. The liquid fuel cooking stoves are generally efficient (over 50% thermal efficiency) and cook quickly and conveniently compared to traditional wood and charcoal stoves. They are very popular in the urban areas in comparison with other biomass fuel cooking stoves. The implications of utilizing FAME fuels to replace kerosene would result in reduced health risks associated with the use of kerosene and wood fuels and possibilities of creating oil crop plantations utilizing marginal lands in Kenya (estimated to be about 80% of total land mass) and thereby reduce heavy dependency on use of solid biomass energy.
1.3.
Objective
The objective of this paper was to investigate the wick stove performance indicators of the Kenyan FAME fuels using the standard water-boiling test (WBT) and compare them with kerosene fuel.
2.
Methodology
2.1.
Extraction of oil from Kenyan oil seeds
The seeds of C. megalocarpus and C. capense harvested during 2007 were obtained from Kenya Forestry Research Institute (KEFRI) in central Kenya (00 230 S 36 560 E) while Jatropha curcus seeds harvested during 2007 were supplied by the Nairobi based Vanilla Development Foundation (VDF) from their plantations in Eastern Kenya (01 300 S 37 150 E). Coconut harvested during 2007 was purchased from a local market in Mombasa town, coastal region (04 020 S 39 430 E). Soybean and sunflower oils were bought during 2007 from the local supermarkets. The seeds were then sun-dried for 72 h and then dehulled manually and ground finely using a mortar. The samples were then soaked in n-hexane each for three days in order to extract maximum amounts of oils. The oilehexane mixture was then filtered under pressure and then separated using a rotary evaporator with the temperature of the bath set at 40 C. Recovered hexane was then returned to the cake mixture for further extraction. The oil samples were then
Fig. 1 e Chinese wick stove, wheel brand model 62.
1252
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
treated with anhydrous Na2SO4 to remove any moisture. The resulting oil was used for transesterification process [20].
2.2.
Preparation of fatty acid methyl esters (FAME) fuels
Samples of fatty acid methyl esters (FAME) were prepared using the standard method by base catalyzed transesterification process using the normal 6:1 M ratio of methanol to oil [20e22].
2.3.
Tb ¼ ð100 h=300Þ C
FAME fuel properties
The FAME thermal performance tests, the fuel properties of each FAME and kerosene fuel were analyzed using ASTM (American Society of Testing and Material) procedures for petroleum products [23,24] at the Kenya Pipeline Company Limited, based in Nairobi. Calorific values of the fuels analyzed were determined using the Gallenkamp Auto-bomb model calorimeter in the university laboratory [25]. The FAME fuel property results are summarized in Table 1 in order to explain the variability of stove thermal performance.
2.4.
Wick stove performance experiments
The wick stove performance experiments were conducted using the standard Water-Boiling Test (WBT) version 3.0 to determine stove performance indicators. The tests were in principle used to compare FAME thermal performance against kerosene fuel using the standard WBT protocol viz. highpower phase cold and warm start conditions and low-power simmering phase. The method was chosen because of its simplicity and replicability [14]. The test compared the thermal performance of 6 different FAME cooking fuels against kerosene. The WBT measured time to boil, thermal transfer efficiency, specific fuel consumption (SFC), and maximum and minimum firepower [14,26e28].
2.4.1.
thermometer at 5 min time intervals until the water was brought to boil. The thermometer was positioned in the center of the pot, roughly 3 cm above the pot bottom. The average boiling temperature was taken as the local boiling point [14]. An average of 95 C was recorded. This result was checked against the estimated local boiling temperature based on local altitude h (m) using equation (1). The two were found to coincide.
Determination of local boiling point
The tests were carried out at Kenyatta University in a Chemistry laboratory room located 25 km East of Nairobi, 1 N, 2 E, with elevation 1500 m above the sea level. The laboratory had 4 standard windows and one standard door. Only one window and the door remained open during the tests. The local boiling point temperature was determined using kerosene fuel loaded onto a wick stove. Approximately 1000 g of cold clean water was weighed and poured into the aluminum pot (Fig. 2). The ambient and water temperatures were monitored via the
(1)
For Kenyatta University, Nairobi, Kenya, the altitude h used was 1500 m above sea level.
2.4.2.
High-power cold start
The high-power cold start experiments began with loading each 500 ml test liquid fuel onto the stove at room temperature and the total weight of stove plus fuel was determined. A time duration of 30 min was allowed for the new wicks to absorb the liquid fuels. The stove was then lit with a matchstick and the water brought to boil rapidly. Fire was put off immediately after the local boiling point was reached for each test fuel. Times taken for lighting and boiling were recorded. Three minutes duration was allowed for loading fresh water onto the pot. The evaporated water and the remaining fuel were weighed. A wooden ply board was used during weighing to protect the weighing balance from heat. The weighing lasted 3e4 min. For every test done, the stove was fitted with new wicks of length 28 cm, diameter 10 mm. For every FAME fuel, triplicate experiments were performed using the standard WBT high-power cold start protocol [14] and the results are presented in Table 2.
2.4.3.
High-power phase hot start
The high-power phase hot start experiments were conducted immediately after the high-power cold start tests while the stove was still hot. The water was brought to boil as in the first phase and time taken was recorded. The fire was then put off and mass of fuel consumed and the water evaporated during this phase were determined and recorded and the results are presented in Table 3 [14].
2.4.4.
Low-power phase hot start (simmering)
The low-power phase hot start experiments were conducted immediately after the high-power phase hot start experiments. The stove was lit again and the temperature of the boiled water was determined. The remaining boiled water from the second
Table 1 e Thermophysical properties of FAME fuels and kerosene. Property Density at 20 C (kg m3) Flash point ( C) Color ASTM Viscosity at 40 C (mm2 s1) Water content (%v v1) Calorific value (MJ kg1)
Test method
Croton FAME
Coconut FAME
Capense FAME
ASTM D1298 ASTM D93 ASTM D1500 ASTM D445 ASTM D95 ASTM D 4809
880 104.5 1.4 1.0 4.2 0.3 0.05 0.2 36.28 0.14
871 106.5 0.7 0.1 2.7 0.04 <0.05 34.84 0.02
875 123.5 12 1.0 4.3 1.6 0.05 0.4 39.93 0.12
FAME: fatty acid methyl ester.
Jatropha FAME 877 147.5 0.4 4.4 0.06 40.75
7.4 0.3 0.8 0.2
Sunflower FAME
Soya bean FAME
Kerosene
883 146.5 4.6 0.4 4.6 0.3 0.05 0.14 39.29 0.2
881 136.5 3.3 0.5 4.2 0.07 0.2 0.11 37.76 0.4
788 45.5 5.4 0.1 2.4 0.01 <0.05 43.61 0.7
1253
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
Burning rate (Rc): This is a measure of the rate of fuel consumption while bringing water to a boil. It is calculated by dividing the equivalent fuel consumed by the time of the test (t) in minutes. Rc ¼
mf t
(3)
Specific fuel consumption (SCc): Specific consumption can be defined for any number of cooking tasks and should be considered “the fuel required to produce a unit output” whether the output is boiled water, cooked beans, or loaves of bread [26e28]. In the case of the cold-start high-power WBT, it is a measure of the amount of fuel required to produce 1 l of boiling water starting with a cold stove. It is calculated in this way: Fig. 2 e Chinese wick stove loaded with aluminum pot.
phase was returned to the stove to simmer for 45 min. The fire was controlled by maintaining the water temperature 3 C below the local boiling point by adjusting the wick height. The fire was put out after the 45 min. Water vaporized and the fuel consumed after the 45 min were determined and recorded, and the results are presented in Table 4 [14].
2.5.
4:186 M1 DT þ 2260 M2 mf CV
(2)
In this calculation, the work done by heating water is determined by adding two quantities: (1) the product of the mass of water in the pot (M1), the specific heat of water (4.186 J g1 C), and the change in water temperature (DT ) and (2) the product of the amount of water evaporated from the pot (M2) and the latent heat of evaporation of water (2260 J g1). The denominator (bottom of the ratio) is determined by taking the product of the mass of fuel consumed (mf) during this phase of the test and the calorific value of the fuel (CV).
mf M1
(4)
Firepower (FPc): This is a ratio of the fuel energy consumed by the stove per unit time. It tells the average power output of the stove (in Watts) during the test phase. FPc ¼
mf CV 60 t
(5)
Evaporation rate (Wc): This is a measure of the rate of water loss through evaporation during the test. Wc ¼
Stove performance indicators
Stove performance indicators were determined through calculations based on the modified version of the well-known water-boiling test (WBT) revised calculation procedure [14]. Thermal transfer efficiency (hc): This is the ratio of the work done by heating and evaporating water to the energy consumed by burning the fuel. It is calculated using equation (1) as follows: hc ¼
SCc ¼
M2 t
(6)
3.
Results and discussion
3.1.
FAME properties
Fuel properties obtained from the FAME experiments along with kerosene are presented in Table 1. Kinematic viscosity for coconut methyl ester was found to be 2.7 mm2 s1, which is 13% above that of kerosene (2.4 mm2 s1). Viscosity of the other FAME fuels was found to be 70e90% above that of kerosene. The flash points of the FAME fuels were typically much higher (2.3e3.3 times) than kerosene. This indicates that they are less flammable and therefore can be handled more safely and have the potential to reduce fire incidents such as those reported in South African slum dwellings [11e13]. The FAME fuels also recorded higher densities than kerosene. The gross calorific value of FAME fuels was found to be 12% (on average) lower than that of kerosene. The heating value depends on the composition of the fuel. Kerosene, which
Table 2 e Mean stove performance indicators, high power (cold start) phase. High power test (cold start) Time to boil Fuel consumed Burning rate Thermal transfer efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME
Coconut FAME
Capense FAME
min g g min1 % g l1 g min1 W
21.4 0.6 40.3 0.6 1.9 39% 0.6 45.8 0.6 5.4 0.08 1101
17.5 0.6 36.3 0.6 2.1 47% 41.0 0.6 6.8 0.23 1157 14
22.9 0.6 39.3 0.6 1.7 41% 1.2 45.3 1 6.1 0.19 1073 17
Jatropha FAME 21.2 36.7 1.7 42% 42.7 6.4 1148
0.6 0.6 1 0.6 0.1 5.1
Sunflower FAME
Soya bean FAME
Kerosene
24.5 0.06 40.3 0.6 1.6 38% 0.6 46.2 0.6 5.1 0.12 1042 5.0
23.4 0.6 40.3 0.6 1.7 40% 1 46.6 1 5.6 0.13 1047 13
11.6 22.7 1.9 50% 24.6 6.9 1373
0.6 0.6 0.6 0.6 0.3 15
1254
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
Table 3 e Mean stove performance indicators high power (hot start) phase. High power test (hot start) Time to boil Fuel consumed Burning rate Thermal efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME 19.5 38.0 1.9 40% 42.4 5.4 1131
min g g min1 % g l1 g min1 W
0.6
0.6 0.6 0.17 13
Coconut FAME
Capense FAME
Jatropha FAME
Sunflower FAME
Soya bean FAME
Kerosene
16.1 34.0 2.1 47% 0.6 38.2 0.6 6.6 0.18 1190 13
19.6 0.6 36.3 0.6 1.8 43% 0.6 41.4 0.6 6.5 0.18 1157 18
19.8 0.6 35.0 1.8 43% 1 39.8 0.6 6.4 0.02 1152 23
22.4 38.7 0.6 1.7 38% 43.5 0.6 5.1 0.07 1088 12
21.0 37.7 0.6 1.8 41% 0.6 42.8 1 5.5 0.03 1092 23
10.0 20.7 0.6 2.1 53% 0.6 22.1 0.6 7.0 0.06 1450 5
constitutes paraffins, olefins and aromatics tends to have much higher-energy contents per volume compared to FAME fuels [29]. It was also observed that coconut FAME fuel had the lowest heating value despite being highly saturated. This could be attributed to the short chain (C6eC14) fatty acid composition [23,30].
3.2.
Water-boiling tests performance
Generally, water-boiling tests (WBT) were conducted to demonstrate the suitability of FAME fuels as potential cooking liquid fuels targeting liquid fuel stove markets. The results show that the FAME fuels tested burnt with transparent blue flame and produced no smell apart from the sweet smell of esters and no black matter was deposited at the bottom of the cooking pot. On the other hand, kerosene fuel produced a strong smell leaving an agglomeration of black carbon deposits at the bottom of the cooking pot. The mean values of the stove parameters obtained at the end of the three WBT phases are shown in Tables 3e4.
3.2.1.
High power (cold start) phase
Various fuel properties were found to affect the stove performance in different ways: The viscosity of the fuel affected its availability to the combustion chamber. The capillary action was faster for the less viscous fuels. The coconut methyl ester had viscosity very close to that of kerosene (2.7 and 2.4 mm2 s1 respectively). The time to boil was close, 17.5 and 11.6 min respectively. The rates of burning were also very close: 2.1 and 1.9 g min1 respectively. Other FAME fuels had viscosities ranging from 4.1 to 4.6 mm2 s1 and hence took a longer time (21e25 min) than that required to boil an equal volume of water (Table 2). The specific fuel consumption and firepower seemed to be affected by the combined effects of fuel heating values and the viscosities. Low viscosity combined with high-energy content for kerosene resulted in
less fuel consumption and much shorter time to boil. The heat transfer efficiency of all the FAME fuels was found to be in the same range (38e42%) except for coconut with 47%, very close to kerosene (50%). Kerosene fuel took the least time to boil the water viz. 11.6 min consuming almost 50% less mass compared to FAME fuels to bring equal volumes of water to boil and giving the highest power output (1.4 kW). The firepowers obtained for all the FAME fuels were within the same range (1.0e1.2 kW). The results obtained show relatively consistent thermal efficiencies in the three water-boiling tests and are in good agreement with reported efficiencies [18e20].
3.2.2.
Stove performance during high power (hot start) phase
Table 3 shows the mean values of wick-stove parameters obtained at the end of high-power phase (hot start) for FAME and kerosene fuels. Slightly lesser time (2e3 min) was required to boil the water during high power hot start phase than in cold start for each fuel giving a small increase in the power output (4e6%). This can be explained by the fact that the fuel in the stove tank would be at slightly higher temperature than room temperature. This increased the flash point and lowered the viscosity and hence increased the fuel mobility by capillarity. Heat transfer efficiency remained more or less the same as in cold start, possibly because the metallic stove body conducted heat fast such that within 5 min the body would be at room temperature. The burning rate remained relatively the same as in phase 1.
3.2.3. phase
Stove performance during low-power (simmering)
The goal of low-power simmering phase was to maintain water at a high temperature using minimal stove output power for 45 min as per the WBT protocol. The power required to keep the water simmering (3e4 C) below the boiling temperature seemed to be the same for all the liquid fuels (900e910 W) possibly due to liquid fuel capillary flow
Table 4 e Mean stove performance indicators low-power (hot start) phase. Low-power test (hot start) Time to boil Fuel consumed Burning rate Thermal efficiency Specific fuel consumption Evaporation rate Firepower
Units
Croton FAME
Coconut FAME
Capense FAME
Jatropha FAME
Sunflower FAME
Soya bean FAME
Kerosene
min g g min1 % g l1 g min1 W
45.0 67.7 0.6 1.5 42% 0.6 137.2 0.6 8.9 0.06 900 5.2
45.0 71.0 1.6 42% 0.6 147.6 2.9 9.2 0.16 907 4.6
45.0 63.0 1.4 42% 0.6 135.5 2.1 9.1 0.13 902 1.2
45.0 60.7 0.6 1.4 43% 0.6 132.8 1.5 9.3 0.02 905 6.8
45.0 63.0 1.4 41% 128.4 1.5 8.8 0.03 901 8.9
45.0 65.0 1.5 41% 132.7 0.6 8.8 0.04 901 1.2
45.0 56.3 0.6 1.3 43% 110.6 1.2 9.3 0.06 910 7.4
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
100
Temperature (degree, Celcius)
90 80 Kerosene CoconutME CapenseME SunflowerME SoyabeanME CrotonME JatrophaME Ambient
70 60 50
may be attributed to low energy content in FAME fuels (see Table 1) as compared to that of kerosene. However, slightly less time was taken to reach boiling point during high power (hot start) phase due to decreased FAME viscosity and density that enhanced capillary action since the stove was still hot during the starting period.
4.
Conclusions
40 30 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Time (minutes)
Fig. 3 e Typical temperature profiles generated during high power boiling phase (cold start).
limitations. The thermal efficiency for all the liquid fuels remained relatively the same (41e43%). In addition, the wick stove demonstrated 304% higher specific FAME fuel consumption during the low-power phase as compared to the high-power cold start phase. Interestingly, kerosene fuel demonstrated 449.5% higher specific fuel consumption during the low simmering phase compared to the high power phase. These results demonstrate that there is increased liquid fuel capillary action due to pre-heating effects resulting from combined effects of reduced FAME fuel viscosities and densities. This also implies that for conventional slow cooking tasks, the use of FAME fuels in wick stoves is comparable to slow cooking using kerosene fuel.
3.3.
(a) Fatty acid methyl ester (FAME) fuels have successfully been developed from Kenyan oil plants namely Jatropha carcus L. (Physic nut), Croton megalocarpus Hutch, Calodendrum capense (L.f.) Thunb. (Cape Chestnut), Cocos nucifera L. (coconut), Sunflower and soybeans using the standard method by base catalyzed transesterification process of the normal 6:1 M ratio of methanol to plant oils; (b) The FAME fuels tested using the standard water-boiling tests show that FAME fuels have the capacity to generate an average of firepower of 1095 W, which is 20% lower than that of kerosene. (c) Generally all FAME fuels showed an average specific fuel consumption of 44.6 g per liter of water, which was higher by 55%, compared to that of kerosene. (d) It was found that the FAME fuels had much higher flash points (104.5 Ce147.5 C) compared to kerosene whose flash point was only 45.5 C implying that FAME fuels are much safer to use than kerosene with little risk of fire hazards in domestic cooking applications. (e) On the basis of these findings, the FAME fuels have great potential to provide an alternative source of domestic cooking and heating energy for people living in developing countries.
Temperature profiles
Figs. 3 and 4 represent the water temperature profiles during high-power phase and low-power phase using various types of FAME fuels. It was observed that all FAME fuels took almost double the time compared to kerosene to attain the same boiling point temperatures during high and low phases. This 100 90
Temperature (degree, Celcius)
1255
80 Kerosene CoconutME CapenseME SunflowerME SoyabeanME CrotonME JatrophaME Ambient
70 60 50
Acknowledgement We wish to acknowledge the financial support from SIDA sponsorship through the lake Victoria Initiative (VicRes) program. We also wish to thank Kenyatta University for supporting this work through provision of laboratory space and equipment. We also wish to thank Mr. Festus K. Muchena, Chief Chemist Kenya Pipeline Company Ltd who facilitated the analysis of fuel properties at the company’s laboratory, and Vanilla Development Foundation (VDF) which provided Jatropha seeds for free. All those who supported this work directly and indirectly are greatly acknowledged.
references
40 30 20 0
2
4
6
8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Time (minutes)
Fig. 4 e Typical temperature profiles during low-power simmering phase.
[1] Wamukonya L. Energy consumption in three rural Kenya households: a survey. Biomass and Bioenergy 1995;8(6): 445e51. [2] Senelwa KA, Simis REH. Opportunities for small-scale biomasseelectricity systems in Kenya. Biomass and Bioenergy 1999;17:239e55. [3] Owino GA. Changes in household fuel and stove choices and their implications for indoor air pollution quality, climate
1256
[4]
[5] [6] [7] [8]
[9]
[10] [11]
[12]
[13] [14]
[15]
[16]
b i o m a s s a n d b i o e n e r g y 3 4 ( 2 0 1 0 ) 1 2 5 0 e1 2 5 6
change and loca tropospheric chemistry. MSc thesis, Department of Environmental Science, Kenyatta University, Nairobi, Kenya; 2004. Ndiema CK. Combustion characteristics of the Kenyan biomass stoves. Unpublished PhD thesis dissertation, Department of Fuel and Energy, University of Leeds, England, UK; 1992. Kammen DM. Cook stoves for the developing world. Scientific American 1995;273(1):64e7. Charron D. The ecostove-getting rid of nearly 90% of kitchen wood smoke. Boiling Point 2005;50. ITDG, UK. Smith KR. Update on the health and climate impacts of household solid fuels. Boiling Point 2006;52:3e5. Smith KR, Mehta S. The burden of disease from indoor air pollution in developing countries: comparison of estimates. International Journal of Environmental Health 2003;206(5): 279e89. Bates E, Nigel B, Doing A, Gitonga S. Participatory approaches for alleviating indoor air pollution in rural Kenyan kitchens. Boiling Point 2002;48:12e5. Hannah I. Livelihoods in the urban biomass sector-realities and threats. Boiling Point 2003;49:21e3. Lloyd P, Heinrich G, Mulaudzi F. A comparison of the thermal efficiency of different stoves used nationally in township homes. Domestic use of energy conference. South Africa: Energy Research Center, University of Cape Town; 2004. Llyod P. The safety of paraffin and LPG appliances for domestic use. Journal of Energy in South Africa 2002;13: 56e61. Llyod P. Developing safe paraffin appliances in South Africa. Boiling Point 2009;56:7e9. Bailis R, Damon O, Dean S, Smith KR, Rufus E. The water boiling test (WBT). Household energy and health program. Shell Foundation, www.ehs.berkeley.edu/hem/protocols/ wbt-version3.0Jan2007a.pdf; 2007 [Last accessed 24.06.09]. Floor W, van der Plas R. Kerosene stoves: their performance, use and constraints, joint UNDP/World Bank Energy Sector management assistance program; 1991. Andrews G, Mkapi MC. Vegetable oil as alternative household fuel to imported kerosene in Africa. Leeds African Studies Bulletin 1996;61:48e52.
[17] Verhoeven N. Kerosene wick stoves. Master’s thesis, Faculty of Physics, Eindhoven University of Technology, Eindhoven, Netherlands; 1989. [18] Kruger GT. Firebomb kerosene stoves without social bounds. Addressing energy poverty in South Africa. In: Domestic use of energy conference. Cape Town, South Africa: The Paraffin Safety Association of Southern Africa; 2006. p. 1e7. [19] Truran G. Household energy poverty and paraffin consumption in South Africa. Boiling Point 2009;5:2e6. [20] Fungrai A, Milford A. Biodiesel production: a review. Bioresource Technology 1999;70:1e15. [21] Lang X, Dalai A, Bakhash N, Reaney M, Hertz P. Preparation and characterization of biodiesel from various bio-oil. Bioresource Technology 2001;80:53e62. [22] Schwab A, Bayby M, Freedman B. Preparation and properties of diesel fuels from vegetable oils. Fuel 1987;66: 1372e8. [23] Van Gerpen J, Shanks B, Pruszko R. Biodiesel analytical methods. IowaState University Renewable Products Development Laboratory; 2004. [24] ASTM (American Society for Testing and Materials). ASTM designation D975-92a, standard specification for diesel fuel oils. Philadelphia: ASTM; 1992. p. 310e1. [25] ASTM (American Society for Testing of Materials). Standard method for heat of combustion of liquid hydrocarbon fuels by bomb calorimeter. Philadelphia, United States: ASTM; 1998. p. 133e40. [26] Visser P. The testing of cookstoves: data of water-boiling test as a basis to calculate fuel consumption. Energy Sustain Development 2005;9(1):16e24. [27] Berrueta Vıctor M, Edwards Rufus D, Masera Omar R. Energy performance of wood-burning cookstoves in Michoacan, Mexico. Renewable Energy 2008;33:859e70. [28] Baldwin SF. Biomass stoves: engineering design, development, and dissemination. Princeton, NJ: Center for Energy and Environmental Studies; 1986. [29] Knothe G. Analytical methods used in the production and fuel quality assessment of biodiesel. Journal of American Society of Agricultural Engineers 2000;42:193e200. [30] Pryde E. Fatty acids. Champaign, Illinois: The American Oil Chemists’ Society; 1979. p. 19e23.