Charcoal is used for cooking in many parts of the developing world. Charcoal from agricultural materials can also be used as a soil amendment to enhance agricultural production, and is often termed biochar in this case. Charcoal may be produced in various types and scales of systems, but for rural tropical areas, traditional kiln technologies without treatment of the pyrolysis gases dominate. Traditional charcoal industry is considered to be both inefficient and polluting, emitting harmful off-gases containing methane, carbon monoxide and particles. Retort kilns, which recirculate and combust the pyrolysis gases internally have been claimed to overcome this problem. Even though retort technology has frequently been discussed, this paper is the first study determining gas emission factors for operational retort kilns in rural tropical areas. The mean emission factors for the retort ldlns found in this study using identical feedstock were (in g kg(-1) charcoal); carbon dioxide (CO2) = 1950 +/- 209, carbon monoxide (CO) = 157 +/- 64, nonmethane volatile organic components (NMVOC) = 6.1 +/- 3.4, methane (CH4) = 24 +/- 17, total solid particles (TSP) = 12 +/- 18, products of incomplete combustion (plc) = 200 +/- 97 and nitric oxides (NOx) = 1.8 +/- 1.0. The corresponding value for the non-retort ldlns tested was in general higher; CO2 = 2380 +/- 973, CO = 480 +/- 141, NMVOC = 13 +/- 3.8, CH4 = 54 +/- 29, TSP = 7.9 +/- 2.6, PIC = 554 +/- 138 and NOx = 4.3 +/- 1.6. The difference between the kiln types was statistically significant (p < 0.05) for CO, NMVOC, PIC and NOx. However, the retort kilns consumed ignitable fuel such as wood in the start-up phase to reach pyrolysis temperature resulting in insignificant difference in yield between retort and non-retort kilns. In addition, retort kilns are more costly than traditional kilns, which may be challenging for implementation. It is therefore important to continue design development, especially with regard to wood consumption during the start-up phase.
2015. Vol. 72, 65-73 p.