P. CIAMBELLI, V. PALMA, P. RUSSO, S. VACCARO, V. VAIANO
Dipartimento di Ingegneria Chimica e Alimentare, Universita di Salerno, Fisciano (SA),
A bituminous coal, two biomasses and their blends were devolatilised at high heating rates (103-104 K/s). The combustion of the chars was studied by air flow TPO tests while the dependence of the rate of combustion from the temperature was obtained through constant temperature tests in the range 673-823 K. Specific surface area (SSA) and Active Surface Area (ASA) measurements were also performed on fresh and partially burned-off chars. TPO tests evidenced that in some cases the minerals contained in different parent materials may interact each other enhancing or hindering the combustion process. Constant temperature combustion tests allowed the evaluation of the frequency factor and of the apparent activation energy for the char combustion. As carbon conversion increases, coal and blend chars SSA’s markedly increase, the ASA, from C02 desorption, increases and the profile’s maximum moves towards higher temperatures while the ASA, from CO desorption, decreases and the profile’s maximum shifts at lower temperatures. The number of C-O surface complexes strongly increases but this does not correspond to a sample reactivity increase.
The use of biomass as fuel in power plant technology attracts increasing interest because of its potentiality to lower C02 emissions being, essentially, a nearly carbon dioxide neutral fuel. Indeed, investigations showed that the provision (including harvesting, transport and milling) of biomass for power generation produces only about 6 kg C02/GJ against 96.6 kg C02/GJ emitted by hard coal combustion in power generation . However, the complete replacement of fossil fuels with biomasses is prohibitive at least for retrofitting existing plants because of problems occurring both during the preparation (transport, drying, milling, storage) and the combustion (ignition, slagging, burnout, emission problems, low calorific value) of such fuels . Nonetheless, a promising shortterm option for the use of renewable fuels may be the coal-biomass co-combustion. This process allows the reduction of the consumption of fossil fuels and may be advantageously implemented in the existing coal-fuelled power plants with minor modifications . In addition, it has been suggested that the co-firing of biomasses may help also in reducing NOx and SOx emissions at the exhausts of coal fired boilers [3-5].
Keys parameters for the use of coal-biomass blends in existing plants are the devolatilisation behaviour of such blends, the quantitative assessment of the contributions of each component to volatile and char yield, and the role of possible interactions (of the organic and inorganic fractions) during the devolatilisation and/or the combustion stages. However, discrepancies may be found among results obtained by various authors . For coal-biomass mixtures, some studies [7, 8] have reported that the pyrolytic behaviour of blends, in any proportion, consisted of the additive behaviour of the two individual samples, while others  have reported some interaction between the two fuels during copyrolysis. Many authors have analyzed dynamic data and their apparent kinetics [10-13]. Substantial differences in the values of reported kinetic parameters can be due to several factors related to the experimental methods, operating conditions and data analysis, but also to the chemical composition of the raw materials investigated .
A further crucial point is to carry out experiments under operating conditions comparable to those found in industrial furnaces. Only in this way, indeed, experimental data may result suitable for supporting both devolatilisation and char combustion modelling as well for allowing the quantity of biomass in blends to be properly programmed to supply the suitable volatile matter release and to prevent possible undesirable effects due to the interaction between the fuels . It was shown that laboratory-prepared chars, obtained under conditions milder (TG apparatus) than those encountered by coal in utility boilers, resulted more reactive to oxygen than chars from boilers . A study concluded that the differences in reaction rates between residual utility boiler and laboratory chars could be explained by low intrinsic reactivities and that chars were deactivated in the boiler plant . Further experiments using a high – temperature wire-mesh (HTWM) reactor  were able to show that thermal deactivation alone at realistic combustion particle temperatures of 1600-1800 °С and heating times up to 2 s would give chars with intrinsic reactivities as low as those found in utility boiler residual carbons . Complementary studies on residual carbons showed that, compared with the laboratory-generated chars, the residual carbons presented higher crystallinity  and suggested that a thermal deactivation process had occurred. Therefore, detailed char combustion models incorporating deactivation kinetics based on the HTWM reactor data and statistical distributions of particle properties could account for the persistence of unbumt carbon in the later stages of pulverized coal combustion . In contrast, overall char combustion kinetics can satisfactorily model only the first 90% of burnout but seriously overpredict conversion rates in the later stages of combustion.
This work aimed at studying the combustion behaviour of coal-biomass blends and at comparing it to those of the correspondent parent materials. To this end, 10 wt% biomass blends of a bituminous coal and cocoa shells and wood chips were prepared. The samples were pounded under 45 pm and devolatilised in a HTWM reactor. Their reactivity to oxygen was investigated by constant temperature tests in a flow reactor and by temperature programmed tests in a microbalance apparatus. The specific surface area and the active surface area of fresh and partially burned off samples were also determined.