The thermal behavior of cotton straw briquette (CSB), municipal solid waste (MSW), and their blends was investigated using a thermogravimetric analyzer under N2/O2 and CO2/O2 atmospheres at 20 °C/min from an ambient temperature to 1000 °C. The kinetics and synergistic interaction between MSW and CSB in the co-combustion process were evaluated. The results indicated that MSW blended with CSB improved the ignition and burnout characteristics of the blends, while decreasing the comprehensive combustion characteristics index. The suitable proportions of CSB were less than 60% and 40% separately under 80N2/20O2 and 80CO2/20O2 atmospheres, respectively. The inhibitory effect of CO2 induced burnout temperature and residual mass increased, and the high-temperature stage reaction varied. Kinetic analysis of the blends indicated that blending with CSB could promote MSW combustion in the first reaction stage, while the second and third decomposition stages were complicated because of the synergistic interaction between MSW and CSB in the co-combustion process. The nth order reaction model fit the mass loss of theca-combustion process for the blends very well.
Co-Combustion Characteristics and Kinetic Analyses of Biomass Briquette and Municipal Solid Waste in N2/O2 and CO2/O2 Atmospheres
The thermal behavior of cotton straw briquette (CSB), municipal solid waste (MSW), and their blends was investigated using a thermogravimetric analyzer under N2/O2 and CO2/O2atmospheres at 20 °C/min from an ambient temperature to 1000 °C. The kinetics and synergistic interaction between MSW and CSB in the co-combustion process were evaluated. The results indicated that MSW blended with CSB improved the ignition and burnout characteristics of the blends, while decreasing the comprehensive combustion characteristics index. The suitable proportions of CSB were less than 60% and 40% separately under 80N2/20O2 and 80CO2/20O2atmospheres, respectively. The inhibitory effect of CO2 induced burnout temperature and residual mass increased, and the high-temperature stage reaction varied. Kinetic analysis of the blends indicated that blending with CSB could promote MSW combustion in the first reaction stage, while the second and third decomposition stages were complicated because of the synergistic interaction between MSW and CSB in the co-combustion process. The nth order reaction model fit the mass loss of theca-combustion process for the blends very well.
Contact information: a: School of Mechanical Engineering, Hefei University of Technology, Hefei, Anhui 230009, PR China; b: National Urban Energy Measurement Center (Anhui), Hefei, Anhui 230009, PR China: c: Measurement and Control of Mechanical and Electrical System Key Lab of Beijing, Beijing Information Science and Technology University, Beijing 100085, PR China;
* Corresponding author: firstname.lastname@example.org
The “Junk Fortress Besieged” phenomenon and the burning of the municipal solid waste (MSW) has been an expanding environmental problem for Chinese urban development. Fossil fuels are required during the incineration of MSW because of its high moisture contents and relatively low heating values in most parts of China, especially in the summer. However, it not only increases the demand for energy load but also brings a series of environmental problems (Henry et al.2006). The co-firing of MSW with biomass substituted for fossil fuel is a promising option because agricultural straw biomass is an abundant renewable energy and could reduce the problem of biomass open burning that threatens urban traffic safety in China. In addition, the combustion of MSW blended with straw biomass also reduces environmental pollution emitted from MSW incineration (Skodras et al. 2006; Ren et al. 2009).
Up to the present, most studies have mixed non-pretreated straw biomass with MSW for co-firing. The combustion efficiency of raw straw biomass in an incineration plant is only 10% to 15% because of the bulk volume, low energy density, and the seasonal, high material transportation, and storage costs. Moreover, the high moisture contents of biomass suppress the combustion process and increase emissions of wet flue gas (Grover and Mishra 1996). An alternative way is to dry and pulverize raw straw biomass materials before co-firing. In this pre-treatment process, biomass is converted into hard-packed biomass briquettes or biomass pellets, which are an important renewable source of energy production (Larsson et al. 2008). Densification of these materials improves their behavior as a fuel by increasing their homogeneity and allowing a wider range of lignocellulosic materials to be used as fuel (Tabarés et al. 2000; Li et al. 2016). Pelletization may extend fuel devolatilization times compared with the original loose fuel constituents. Also, the amount of oxygen required for the combustion can be well matched with the amount of oxygen diffused from the outside, which positively affects the effectiveness of lateral fuel spreading and evenness of volatile matter release during combustion. This induces less combustion-wave and tends to be more stable during the burning process (Chirone et al. 2008; Roy et al. 2013; Liu et al. 2014).
Biomass pellets can be used in grate furnaces and fluidized bed combustion while offering advantages including easy storage and transport, lower pollution, lower dust levels, and higher heating values (Gil et al. 2010). Therefore, the co-combustion of MSW and biomass briquettes can solve the aforementioned problems, further improve the calorific value of the blend fuels, and enhance the burning efficiency and stability of MSW (Tian et al. 2013). The combustion performance and kinetic characteristics of briquetted biomass differs from non-pretreated biomass, leading to differences in device structure and combustion products (Houshfar et al.2011; Liu et al. 2011; Roy et al. 2013). However, the combined firing of MSW and biomass briquettes, especially for those concerning the combustion kinetics, have not been studied systematically.
Oxygen combustion technology under a CO2/O2 atmosphere, instead of N2/O2 atmosphere, is a promising combustion technology associated with carbon capture and storage because of its low NOx emission and high sulfation efficiency (Liu and Okazaki 2003; Ge et al. 2015; Hees et al.2016). However, comparison between the two models shows that there are distinctions in combustion characteristics such as burning stability, char burn-out, heat transfer, and gas temperature (Molina and Shaddix 2007; Lai et al. 2012). Therefore, there is a need to study combustion characteristics and reaction kinetics of MSW, biomass briquettes, and their blends in different blending ratios under N2/O2 and CO2/O2 atmospheres.
In this study, the kinetic parameters during combustion were obtained by the Coats-Redfern method under the nth order reaction model separately in the CO2/O2 and N2/O2 atmospheres. The synergistic interaction between MSW and CSB in the co-combustion process was also investigated. The obtained results elucidate the combustion process and offer reference and guidance for the design and operation of co-combustion of MSW and biomass briquettes.
The MSW mixture was collected from Hefei University of Technology according to the typical component of MSW in Hefei, China. The samples were purified from raw MSW by the manual removal of metal, glass, dust, etc. Thus, the constituents were combustible, including food waste and fruit peels (FW&FP), plastic, rubber, textile, paper, and a mixture of bamboo and wood (BB&WD). The selected cotton straw biomass briquettes (CSB) provided by Anhui Haosheng Energy Technology Co., Ltd. (Hefei, Anhui, China) had a formation density of 1.36 g/cm3. The MSW and CSB were oven-dried at 105 °C for 12 h, smashed repeatedly by a FW100 crusher manufactured by Zhengzhou Kefeng Instrument and Equipment Co., Ltd. (Zhengzhou, Henan, China), and passed through a sieve with mesh size of 68 μm. The proximate analysis of MSW and CSB was carried out by a MAC-3000 fully auto-measuring industrial analyzer (Jiangyan Guochuang Analytical Instruments Co., Ltd., Taizhou, Jiangsu, China), of which the percentage of fixed carbon was determined by difference. The elemental compositions were carried out on a Vario El Cube elemental analyzer (Elementar, Hanau, Germany). The proximate and ultimate analyses of MSW and CSB are shown in Table 1, and the ash content of adjusted MSW was lower than that of CSB material.
Table 1. Ultimate and Proximate Analyses of MSW and CSB Samples
All treated components were mixed together one by one in a micro rotary mixer for one hour. In addition to the MSW and CSB individual samples, the mass percentages of CSB added to the MSW were 20%, 40%, 50%, 60%, and 80%, which were named 20C80M, 40C60M, 50C50M, 60C40M, and 80C20M, respectively. These samples were uniformly blended by the mechanical mixing method and stored in desiccators until they were used. Each test sample was taken separately by the coning quartering method before the experiment began. The initial weight of the samples for all runs was formulated as 10 ± 0.5 mg.
The combustion characteristics of MSW, CSB, and their blends were studied using a Setsys Evo thermogravimetric analyzer (SETARAM, Lyon, France) with temperature precision and microbalance sensitivity of ±0.3 °C and ±0.023 μg, respectively. The balance was designed above the furnace to prevent pollution of residual products from volatile decomposition. A counterweight and crucible were suspended by filaments at both ends of the TG to minimize the buoyancy effect. The crucible with the sample was always at the center of the chamber to avoid thermogravimetric error caused by changes of the sample gravity position, and the sample could be contacted fully with the atmosphere purged from top to bottom, which was important during the combustion test.
All the non-isothermal combustion experiments in 80N2/20O2 and 80CO2/20O2 atmospheres were carried out from an ambient temperature to 1000 °C at a heating rate of 20 °C/min. The flow rate of mixed gas was maintained at 60 mL/min. To eliminate the systematic errors caused by the weight of the crucible, temperature, and effect of buoyancy, a blank test without samples was conducted to obtain baseline data before each experiment. All the experiments in each case were performed repeatedly at least twice to ensure the accuracy, and the results of reproducibility were quite good.
Characteristics of Combustion Phenomenon
To evaluate the comprehensive combustion characteristics of MSW, CSB, and their blends, the comprehensive combustion characteristic index S was introduced in Eq. 1 (Hu et al. 2015),
where (dW/dt)max and (dW/dt)mean were the maximum and average mass loss rate, respectively. Ti was the ignition temperature determined based on a comprehensive consideration of thermogravimetric and differential thermogravimetric method (TG-DTG), in which a tangent line was drawn that corresponded to the cross-point between the TG curve and the vertical line to the maximum weight loss rate of DTG curve, and the ignition temperature was defined as the intersection between the above tangent line and the tangent line to the point of the mass loss started. The burnout temperature Tf was obtained when the mass loss of sample reached 98% of the weight loss (Fang et al. 2015). The larger value of the index S, which was representative of the characteristics of MSW, CSB, and their blends, the better vigorous combustibility was (Hu et al. 2015).
The combustion reaction process of MSW or CSB can be simplified as typical solid heterogeneous reaction, which may take place through a series of parallel and competitive reactions. The fundamental rate equation of heterogeneous solid phase reactions could be described with the Arrhenius equation (Eq. 2) (Liu et al. 2009; Lin et al. 2014),
where t is the time, E is the apparent activation energy, A is the pre-exponential Arrhenius factor, R is the universal gas constant, T is the reaction temperature, is a function depended on the assumed reaction mechanism in the combustion process of biomass and MSW (Liu et al. 2009; El-Sayed and Mostafa 2014), and n is the order of the reaction.
The degree of conversion of the weight loss is expressed according to Eq. 3,
where m0 and m∞ are the initial and final masses, respectively, of the samples, and mt is the mass of the samples at time t.
For the non-isothermal thermogravimetric experiments with linear heating rate program of β= dT/dt, Eq. 2 is transformed, as shown in Eq. 4,
The Coats-Redfern method is widely used to evaluate the kinetic parameters from TG-DTG curves (El-Sayed and Mostafa 2014; Oyedun et al. 2014). Starting with Eq. 4 and after certain algebra, the kinetic equation leads to the following integral forms, shown in Eqs. 5 and 6,
For n =1, (5)
For n≠1, (6)
Because the temperature range and activation energy in this work, 2RT/E<< 1 (Lin et al. 2014; Oyedun et al. 2014), then could be almost regarded as a constant c. Assuming that the left side of the above Eqs. 5 and 6 both were Y, b = −E/R, and X = 1/T, thus the Eq. 5 and Eq. 6 can be presented in the forms of Y=c + bX. The plot of Y vs. X will obtain various straight lines, while the appropriate reaction order n is determined by the highest value of the correlation coefficient (El-Sayed and Mostafa 2014). The E and A can be obtained from the slope and intercept of the fitted straight line.
RESULTS AND DISCUSSION
Thermogravimetric Analysis of MSW and CSB
The profiles of DTG curves of MSW in Fig. 1b showed that there were three noteworthy peaks in the combustion process under both N2/O2 and CO2/O2 atmospheres, and correspondingly TG curves characterized by three weight loss stages in Fig. 1a were easily distinguished. To further analyze the combustion characteristics of MSW, thermogravimetric experiments for all single components of MSW were also carried out under an N2/O2 atmosphere.
Fig. 1. Combustion of MSW and CSB under N2/O2 and CO2/O2 atmospheres: (a) TG curves;
(b) DTG curves
Figure 2a displays the first and second mass loss peaks of paper and BB&WD presented similar temperature zone of mass loss because they had similar components of lignocellulose. The first peak occurred at 339.5 °C and 334.6 °C, respectively, and accounted for devolatilization of lignocellulosic fractions, in which the hemicelluloses and cellulose had high reactivity and burned completely around 340 °C, while lignin burned within a broader temperature range (Lai et al.2012; Zhou et al. 2015). The second mass loss stage occurred from about 390 to 560 °C and was mainly due to the further decomposition and burning of lignin and afterwards, the combustion of chars (Tang et al. 2011; Fan et al. 2016). Above 600 °C, the DTG curve of paper had the third mass loss stage until 800 °C, which was attributed to the degradation of calcium carbonate and other minerals (Liao and Ma 2010). The combustion of textiles showed that the maximum rate of mass loss at 345.4 °C was attributable to the devolatilization and burning of cotton (mainly composed of cellulose, the main component of textiles). The second peak at 423.7 °C resulted from the decomposition and combustion of polyester, another usual ingredient of textiles (Alongi et al. 2011). The combustion of residues of aforementioned components resulted in the third mass loss stage.
Fig. 2. DTG curves for single component of MSW in N2/O2 atmosphere: (a) DTG curves of paper, textiles, and BB&WD; (b) DTG curves of FW&FP, plastic, and rubber.
As shown in Fig. 2b, the DTG curve of FW&FP, which is composed of rice and banana peel in a mass ratio of 4:1, presented three peaks. The first tiny peak of banana peel devolatilization occurred at a lower temperature of 205.7 °C, revealing the degradation of pectin, hemicelluloses, and sugars (Branca and Blasi 2015; Zhou et al. 2015). Based on the previous studies, the second peak at 317.8 °C was the primary peak caused by the cellulose and starch decomposition and part of lignin combustion (Tang et al. 2011; Branca and Blasi 2015; Fang et al. 2015). The third peak had a broader temperature zone between 400 and 570 °C, which was attributed to the further combustion of lignin and char burning. The first peak of plastic was at 439.8 °C and resulted from the thermal decomposition of polyethylene terephthalate, which was the main component of the plastic, and the second one was most likely due to the combustion of carbonaceous residue and other additives (Brems et al. 2011). The peaks at 276.9 °C and 444.2 °C revealed the decomposition and burning of rubbers, of which natural rubber degraded at lower temperature and synthetic rubber (polybutadiene rubber or styrene-butadiene rubber) decomposed at higher temperature (Cui et al. 1999; Tang et al. 2015). The third one was at 545.2 °C because of the combustion of carbon black and minerals (Cui et al. 1999).
Therefore, three mass loss stages of MSW were obtained: the first decomposition stages centered on 316 °C under both of the two atmospheres with weight loss of about 43%. This stage represented the devolatilization and combustion of lignocellulosic materials, pectin, starch, sugars, and part of rubber components. The second peaks at around 431 °C with mass loss of about 24% were due to the decomposition and combustion of lignin, polyethylene terephthalate, polyester and synthetic rubber. The third ones were caused by the combustion of remaining char, carbon black, and some inorganic compounds with weight loss of about 27%. The main decomposition process resulted from the release and combustion of volatile component, which accounted for the major proportion of MSW in proximate analysis. Similar conclusions were also obtained by Hu et al. (2015). The third mass loss stage of paper from 600 to 800 °C reflected in the decomposition process of MSW as an inconspicuously tiny change might be because of the influence of interactions among the single components (Zhou et al. 2015), which can be hypothesized that was not a mass loss peak in the thermogravimetric analysis.
As shown in Fig. 1a, the remnant of CSB was higher than MSW, which was in accordance with the higher fixed carbon and ash content in Table 2. Obernberger and Thek (2004) stated that a higher ash content of pellets in comparison with unprocessed raw materials could be acceptable, which were destined for industrial use due to the greater robustness and sophistication of industrial combustion systems. Wang et al. (2007) reported that the fixed carbon of briquetted biomass burned more evenly; the terminated temperature and persistence of combustion were increased compared to uncompressed biomass. The ignition temperature of MSW was 284.3 °C, while CSB was 245.7 °C, which was about 40 °C lower than MSW under the N2/O2 atmosphere. The burnout temperature of CSB (557.0 °C) was almost around 20 °C lower than that of MSW (575.5 °C) under the N2/O2 atmosphere. In the CO2/O2 atmosphere, the ignition and burnout temperatures of the CSB were also about 40 °C and 30 °C below the MSW, respectively. Therefore, the ignition and burnout characteristics of MSW could be improved by the addition of CSB.
The CSB, which is mainly composed of lignocelluloses, was different from MSW composed of lignocelluloses, plastics, and rubber, etc. Two prominent mass loss peaks of CSB were shown in Fig. 1b under two atmospheres, and the primary mass loss stages centered on 327.4 C °C in the N2/O2 atmosphere and 328.6 °C in the CO2/O2 atmosphere were attributed to the devolatilization and combustion of cellulose, hemicelluloses, and partial lignin. The second peaks occurred at higher temperatures of 445.0 °C in the N2/O2 atmosphere and 448.1 °C in the CO2/O2atmosphere were due to the combustion of residual lignin and char. As shown in Fig. 1, the TG and DTG curves below 300 °C of CSB in two atmospheres represented the similar tendency, which indicated that the influence of carbon dioxide on the initial phase of combustion was weak. Also, the replacement of N2 by CO2 caused a small shift of curve to higher temperature without significant change in the shape of TG and DTG cures for MSW. The remnant of MSW in the CO2/O2 atmosphere was higher, and the mass loss rate under the CO2/O2 atmosphere was lower. The results of CSB revealed the same characteristics except that a small hump centered on 500 °C formed under N2/O2 atmosphere, while the rate of mass loss declined continuously until about 580 °C in CO2/O2 atmosphere. This result might be due to the reactivity in the CO2/O2atmosphere being reduced at higher temperature.
The above results suggested that carbon dioxide had negative effects on the combustion reaction, which indicated that the materials burned more fully under the 80N2/20O2 atmosphere than the 80CO2/20O2 atmosphere. Lai et al. (2012) reported that the combustion performances of higher oxygen concentration in CO2/O2 atmosphere could be well matched with that in N2/O2atmosphere.
Co-combustion Behavior of the Blends
Effect of blending of MSW with CSB under N2/O2
Figure 3a represents the DTG curves of MSW, CSB, as well as their blends at different quality ratios under N2/O2 atmospheres. It can be seen that the DTG curves of the blends under N2/O2atmosphere lay between that of the individual components. A similar phenomenon was observed in a study of combustion of MSW and oil shale (Fan et al. 2016). As the mass proportion of CSB increased in the blends, the DTG curves changed slightly below 390 °C, while the maximum weight loss rate gradually decreased. Hence, the diversities in combustion profiles of DTG curves were extended. The characteristic parameters of all samples under N2/O2 atmosphere are displayed in Table 2. The mass loss rate of second peaks was higher than the third peaks when the CSB ratios were 20% and 80% in blends as well as the individual compounds, while weight loss rate of third peaks were greater than the second ones at the CSB ratio of 40%, 50%, and 60%. The reason for this result might be the alkali components of CSB were oxygen carriers to the residue char of MSW, which catalyzed char combustion in the third mass loss stage of MSW.
As shown in Table 2, the initial temperature decreased from 284 to 246 °C with the increase of CSB proportion. This was mainly due to the fact that CSB was mainly composed of lignocelluloses, while the plastic fractions in MSW were decomposed at a higher temperature compared with lignocelluloses. The burnout temperature of blends was also lower than that of pure MSW. This result indicated that the ignition and burnout characteristics of blends could be improved by adding CSB into MSW. The total mass loss declined, and remnants of blends increased with the increased quality ratio of CSB. The value of index S of MSW was higher than that of CSB, and the S decreased from 20.12 to 11.39 (×10-07min-2°C-3) with CSB ratio increased in the blends, which indicated that the comprehensive combustion characteristic of MSW was more vigorous than that of CSB on the basis of noncombustibles removed from MSW under realistic industrial processing conditions. Therefore, the combustion characteristics of the blends were improved compared with individual compounds. In addition, when the ratio of CSB was less than 60%, the index S value of blends was still relatively large, which provided a suitable mixing ratio of MSW and CSB under the 80N2/20O2 atmosphere.
Table 2. Combustion Characteristic Parameters of Samples in N2/O2 Atmosphere
aTi , the ignition temperature; bTf , the burnout temperature; c T1, T2, T3, the temperature according to the first peak, the second peak and the third peak;
dDTG1, DTG 2, DTG 3, the rate of mass loss according to the first peak, the second peak and the third peak; eDTGmean, the average rate of mass loss from the ignition temperature to burnout temperature; f Mf, the combustion residue mass; gS, the comprehensive combustion characteristic index;
* the symbols are shared for Table 3.
Table 3. Combustion Characteristic Parameters of Samples in CO2/O2 Atmosphere
Effect of blending of MSW with CSB under CO2/O2
As shown in Fig. 3b, the DTG curves of blends and individual MSW and CSB under the CO2/O2atmosphere were similar to those under the N2/O2 atmosphere. Table 3 also shows the characteristic parameters of all samples under the CO2/O2 atmosphere. The total mass loss of the blends in the CO2/O2 atmosphere were lower than that in the N2/O2 environment at the same ratio of CSB, and the burnout temperature of the blends were all higher than that in the N2/O2environment. Notably, when the ratio of CSB was 50%, the burnout temperature was the highest. This may be due to the ash content of the burning CSB hindering the combustion of the blend, and CO2 has a higher density and specific heat capacity than N2. All the blends revealed three mass loss peaks that were close to MSW, but the CSB combustion only had two mass loss stages in the CO2/O2 atmosphere.
The mass loss rate of the first two mass loss peaks all decreased with the CSB mass percentage increases. The temperatures according to the second and third peaks at the CSB mass ratio of 80% both were more than 30 °C higher than the other blends because of the contribution of CSB to the blend. This phenomenon was similar to that of 80C20M under N2/O2 atmosphere. As shown in Table 3, the index S of blends declined from 18.23 to 9.84 (×10-07min-2 °C-3) with the increase of CSB ratio. Considering the ignition characteristic, burnout characteristic, and index S, the blending of MSW with CSB could improve the comprehensive combustion characteristics to some extent under CO2/O2 atmosphere when the CSB mass percentage was less than 40% in the blend.
Interaction of MSW with CSB under N2/O2 and CO2/O2 Atmosphere
In order to evaluate the possible synergistic interaction between MSW and CSB, the theoretical TG (TGCAL) and DTG curves (DTGCAL) of the blends were calculated by the arithmetic weighted average of the individuals, which could be stated as follows (Hu et al. 2015; Peng et al. 2015),
where λMSW and λCSB were the mass ratio of MSW and CSB in the blends, and WMSW and WCSBwere the mass loss or mass loss rate of MSW and CSB, respectively.
The parameter ΔTG (ΔTG = TGEXP − TGCAL) was introduced to compare and evaluate the degree of deviation between the calculated and experimental results, where TGEXP was the experimental TG curves. Figures 4a and 4b display the ΔTG curves of blends under the N2/O2 and CO2/O2atmospheres, respectively. All the ΔTG curves of blends presented the similar variation tendency, and the deviations at high-temperature stage were substantially larger than that of low-temperature stage. The experimental TG curves at different CSB mass ratios were all under the calculated TG curves at above 465.3 °C in N2/O2 atmosphere and 311.2 °C in N2/O2 atmosphere, and the mass loss of calculated TG curves were lower than those of experimental ones. This phenomenon could be explained as the synergistic interaction during the co-combustion of MSW and CSB. Two maximum peaks and two minimum peaks were observed in all ΔTG curves of blends, and the corresponding temperature were about 297 °C, 445 °C, 366 °C, and 493 °C under the N2/O2 atmosphere and around 429 °C, 560 °C, 368 °C, and 514 °C under the CO2/O2atmosphere. The lag of peak temperature was attributed to the negative effect on the combustion reaction in the CO2/O2 atmosphere. In conclusion, the above analysis indicated that the interaction effects of MSW and CSB co-combustion at high temperature were relatively intense, and the greater influence of interaction occurred at the proportions of CSB of 40% and 50% under both atmospheres.
Figures 4c and 4d show that the calculated and experimental DTG curves when the mass ratio of CSB were 40% and 50% under both atmospheres, respectively. The profiles of each two DTG curves were consistent throughout the combustion process except for a large gap around the peaks of DTG curves, where the experimental mass loss rates were higher than that of calculated ones and the peaks of experimental DTG curve appeared in advance compared to the calculated DTG curve. As shown in Figs. 4c and 4d, the deviations between the calculated and experimental DTG curves mainly occurred in the range 300 to 600 °C, which corresponded with the temperature range of two maximum peaks and two minimum peaks of ΔTG curves under the two atmospheres. This further demonstrated that there was a synergistic effect between the MSW and CSB co-combustion rather than the weighted average of the two individual compounds. Similar interactions in the study of co-combustion of MSW and paper mill sludge, or MSW and oil shale were also obtained by Hu et al. (2015) and Fan et al. (2016), respectively.
Fig. 4. ΔTG curves of blends at different mass ratios in N2/O2 atmosphere (a) and CO2/O2atmosphere (b); the comparison of calculated and experimental DTG curves at mass ratio of CSB were 40% and 50% in N2/O2 atmosphere (c) and CO2/O2 atmosphere (d)
The mechanism of synergistic interactions between MSW and CSB during co-combustion process needed to be investigated. The content of volatile matter in MSW was higher than that of CSB, as shown in Table 1. The combustion of volatile matter generated enormous heat, which promoted char burning of CSB, whose thermal decomposition occurred at higher temperature. Therefore, the mass loss rate of blends and temperature corresponding to the mass loss peak was most advanced with the increased mass ratio of MSW. In contrast, Huang et al. (2014) found that large quantities of catalytic components (such as KCl, KSO, etc.) were contained in the ash of cotton straw. The addition of CSB to MSW had the similar role of catalysts in the combustion process of MSW, and the catalysts induced the weakening of intermolecular interaction in polymeric chains. Thus, the concentrations of volatile gases around MSW increased, and the ignition temperature of blends was decreased as the proportion of CSB increased. In addition, when oxygen was absorbed in the surface of catalysts, activated oxygen spills would be released to react with MSW char at a lower temperature (Shen and Lei 2006). Therefore, the burnout temperature of blends was lower than that of individual MSW under N2/O2 atmosphere, but this phenomenon was not the same as in CO2/O2 atmosphere. A possible cause for this might be that the interaction was influenced by the atmosphere, and further studies on the mechanism of this synergistic interaction were needed by experimental design. As a result, the synergistic interactions in the co-combustion process of MSW and CSB were thought to result from the combined action of individual components, and the co-combustion of MSW and CSB could implement the co-processing of two solid wastes.
Using only one simple dynamic reaction model is not adequate for interpreting complicated co-combustion process of MSW and CSB, and thus the Coats-Redfern method was adopted to obtain the kinetic parameters and correlation coefficients under nth order reaction model under two atmospheres (Table 4). The coefficients of determination values of R2 were all between 0.9406 and 0.9972, reflecting that the nth order reaction model was appropriate, and the calculated parameters were credible. The combustion of pure MSW and pure CSB both exhibited three steps under N2/O2 atmosphere, for which the activation energies values of E were 82.88, 26.68, and 90.26 kJ/mol and 42.58, 5.17, and 84.25 kJ/mol, respectively. The E values of CSB in three stages were all lower than that of MSW, explaining the reason why the ignition and burnout temperature of CSB were lower than that of MSW. A similar phenomenon was also found under CO2/O2atmosphere.
The values of E for MSW were in agreement with those obtained by Liu et al. (2009) and Lai et al. (2011) of MSW ranged from 65.6to137.87 kJ/mol for the first stage, 55to80.18 kJ/mol for the second step, and 7.4to155.17 kJ/mol for the last reaction, respectively. The reaction orders of MSW were also close to that reported in the above-mentioned studies. However, the activation energies and reaction orders of CSB were all lower than that obtained by El-Sayed and Mostafa (2014). The variability of kinetic parameters was affected by the differences in materials, operating facility, method, sample size, and others in experiments.
Table 4. Kinetic Parameters and Correlation Coefficients for All the Samples
aTR, temperature range; b E, apparent activation energy; c n, reaction order; d A, Arrhenius pre-exponential factor;e R2, coefficients of determination.
For the blends under two atmospheres, all runs were similar to MSW with three reaction stages. The value of activation energy, pre-exponential factor, and reaction order in the first stage all showed a trend of monotonous reduction as the mass ratio of CSB increased in blends. The activation energy in the second stage had an increasing tendency at first, and then decreased gradually as the proportion of CSB increased. The activation energy of blends were up to the maximum of 42.79 kJ/mol at CSB mass ratio of 60% under theN2/O2 atmosphere and 33.83 kJ/mol at the 40C60M blend under theCO2/O2 atmosphere, while the activation energy of 80C20M under the two atmospheres were all close to that of CSB. The variation trends of pre-exponential factor and reaction order of blends in the second stage were close to that of activation energy. Unlike the aforesaid two stages, the activation energy, pre-exponential factor, and reaction order in the third stage all changed complexly under the N2/O2 atmosphere but raised to peaks at 60C40M blend firstly and decreased afterwards under the CO2/O2 atmosphere. These phenomena might be attributed to the complicated synergistic interaction during the co-combustion process of MSW and CSB (Lin et al. 2015). As a whole, the blending of MSW with CSB could decrease the kinetic parameters of blends on some level, and the good correlation coefficients indicated that the nth order reaction model could excellently fit the experimental results by the three stages reaction.
- The ignition and burnout characteristics of blends were improved by adding CSB into MSW, while the comprehensive combustibility index S decreased as the CSB ratio increased in the blends. The suitable mass ratios of CSB were less than 60% under the 80N2/20 O2 atmosphere and less than 40% under the 80CO2/20O2 atmosphere.
- The mass loss percentages of blends in theCO2/O2 atmosphere were lower than that in theN2/O2 environment at the same ratio of CSB, and the burnout temperature was higher than in theN2/O2 environment. The replacement of N2 by CO2 had negative effect on burnout temperature, residual mass and high-temperature region combustion.
- The synergistic interactions between MSW and CSB co-combustion were relatively intense at high temperature, and the greater influence of interaction occurred when the proportion of CSB was 40% and 50% under both atmospheres.
- The combustion process of blends was divided into three stages. The synergistic interaction between MSW and CSB promoted activation energy of blends reduced in the first reaction stage, and it complicated the activation energy in the subsequent two reaction stages. The high correlation coefficients values reflected that nth order reaction model was appropriate for describing the weight loss process of blends.
The authors are grateful to the support given by the Municipal Colleges and Universities Innovation Ability Promotion Projects of Beijing Municipal Education Commission (J2014QTXM0204), the Science and Technology Project of Anhui Province (2013AKKG0398), and the International Science and Technology Cooperation Project of Anhui Province (1403062015).
Alongi, J., Ciobanu, M., Tata, J., Carosio, F., and Malucelli, G. (2011). “Thermal stability and flame retardancy of polyester, cotton, and relative blend textile fabrics subjected to sol–gel treatments,” Journal of Applied Polymer Science 119(4), 1961-1969. DOI 10.1002/app.32954
Branca, C., and Blasi, C. D. (2015). “A lumped kinetic model for banana peel combustion,” Thermochimica Acta 614, 68-75. DOI: 10.1016/j.tca.2015.06.022
Brems, A., Baeyens, J., Vandecasteele, C., and Dewil, R. (2011). “Polymeric cracking of waste polyethylene terephthalate to chemicals and energy,” Journal of the Air and Waste Management Association 61(7), 721-31. DOI: 10.3155/1047-32188.8.131.521
Chirone, R., Salatino, P., Scala, F., Solimene, R., and Urciuolo, M. (2008). “Fluidized bed combustion of pelletized biomass and waste-derived fuels,” Combustion and Flame 155(1), 21-36. DOI: 10.1016/j.combustflame.2008.05.013
Cui, H., Yang, J., and Liu, Z. (1999). “Thermogravimetric analysis of two Chinese used tires,” Thermochimica Acta 333(2), 173-175. DOI: 10.1016/S0040-6031(99)00119-7
El-Sayed, S. A., and Mostafa, M. E. (2014). “Pyrolysis characteristics and kinetic parameters determination of biomass fuel powders by differential thermal gravimetric analysis (TGA/DTG),” Energy Conversion and Management 85(9), 165-172. DOI: 10.1016/j.enconman.2014.05.068
Fang, S., Yu, Z., Lin, Y., Hu, S., Liao, Y., and Ma, X. (2015). “Thermogravimetric analysis of the co-pyrolysis of paper sludge and municipal solid waste,” Energy Conversion and Management101, 626-631.DOI: 10.1016/j.enconman.2015.04.026
Fan, Y. L., Yu, Z. S., Fang, S. W., Lin, Y., Liao, Y., and Ma, X. (2016). “Investigation on the co-combustion of oil shale and municipal solid waste by using thermogravimetric analysis,” Energy Conversion and Management117, 367-374. DOI: 10.1016/j.enconman.2016.03.045
Ge, P., Zhu, W. L., Zhou, H. P., Lei, Q., Zhang, Z. R., and Liu, J. J. (2015). “Co-combustion characteristics of inferior coal and biomass blends in an oxygen-enriched atmosphere,” BioResources 10(1), 1452-1461.DOI: 10.15376/biores.10.1.1452-1461
Gil, M. V., Oulego, P., Casal, M.D., Pevida, C., Pis, J. J., and Rubiera, F. (2010). “Mechanical durability and combustion characteristics of pellets from biomass blends,” Bioresource Technology 101(22), 8859-8867. DOI: 10.1016/j.biortech.2010.06.062
Hees, J., Zabrodiec, D., Massmeyer, A., Habermehl, M., and Kneer, R. (2016). “Experimental investigation and comparison of pulverized coal combustion in CO2/O2 and N2/O2atmospheres,”Flow, Turbulence and Combustion 96(2), 417-431. DOI: 10.1007/s10494-015-9662-9
Henry, R. K., Zhao, Y., and Jun, D. (2006). “Municipal solid waste management challenges in developing countries – Kenyan case study,” Waste Management 26(1), 92-100. DOI: 10.1016/j.wasman.2005.03.007
Houshfar, E., Skreiberg, Ø., Løvås, T., Todorović, D., and Sørum, L. (2011). “Effect of excess air ratio and temperature on NOx emission from grate combustion of biomass in the staged air combustion scenario,” Energy & Fuels 25(10), 4643-4654. DOI: 10.1021/ef200714d
Huang, S., Wu, S., Wu, Y., and Gao, J. (2014). “The physicochemical properties and catalytic characteristics of different biomass ashes,” Energy Sources Part A Recovery Utilization and Environmental Effects 36(36), 402-410(9). DOI: 10.1080/15567036.2012.722746
Hu, S., Ma, X., Lin, Y., Yu, Z., and Fang, S. (2015). “Thermogravimetric analysis of the co-combustion of paper mill sludge and municipal solid waste,” Energy Conversion and Management 99, 112-118. DOI: 10.1016/j.enconman.2015.04.026
Lai, Z. Y., Ma, X. Q., Tang, Y. T., and Lin, H. (2011). “A study on municipal solid waste (MSW) combustion in N2/O2 and CO2/O2 atmosphere from the perspective of TGA,” Energy 36(2), 819-824. DOI: 10.1016/j.energy.2010.12.033
Lai, Z., Ma, X., Tang, Y., Lin, H., and Chen, Y. (2012). “Thermogravimetric analyses of combustion of lignocellulosic materials in N2/O2 and CO2/O2 atmospheres,” Bioresource Technology 107, 444-450. DOI: 10.1016/j.biortech.2011.12.039
Larsson, S. H., Thyrel, M., Geladi, P., and Lestander, T. A. (2008). “High quality biofuel pellet production from pre-compacted low density raw materials,” Bioresource Technology 99, 7176-7182. DOI: 10.1016/j.biortech.2007.12.065
Li, A., Liu, H. L., Wang, H., Xu, H. B., Jin, L. F., Liu, J. L., and Hu, J. H. (2016). “Effects of temperature and heating rate on the characteristics of molded bio-char,” BioResources 11(2), 3259-3274. DOI: 10.15376/biores.11.2.3259-3274
Liao, Y., and Ma, X. (2010). “Thermogravimetric analysis of the co-combustion of coal and paper mill sludge,” Applied Energy 87(11), 3526-3532. DOI: 10.1016/j.apenergy.2010.05.008
Lin, Y., Ma, X., Yu, Z., and Cao, Y. (2014). “Investigation on thermochemical behavior of co-pyrolysis between oil-palm solid wastes and paper sludge,” Bioresource Technology 166(8), 444-450. DOI: 10.1016/j.biortech.2014.05.101
Lin, Y., Ma, X., Ning, X., and Yu, Z. (2015). “TGA-FTIR analysis of co-combustion characteristics of paper sludge and oil-palm solid wastes,” Energy Conversion and Management89, 727-734. DOI: 10.1016/j.enconman.2014.10.042
Liu, G., Ma, X., and Yu, Z. (2009). “Experimental and kinetic modeling of oxygen-enriched air combustion of municipal solid waste,” Waste Management 29(2), 792-6. DOI: 10.1016/j.wasman.2008.06.010
Liu, H., and Okazaki, K. (2003). “Simultaneous easy CO2 recovery and drastic reduction of SOx and NOx in O2/CO2 coal combustion with heat recirculation,” Fuel 82(11), 1427-1436. DOI: 10.1016/S0016-2361(03)00067-X
Liu, S., Wang, Y., Bai, B., Su, C., Yang, G., and Zhang, F. (2011). “Analysis on combustion kinetics of corn stalk briquetting densification fuel,” Transactions of the Chinese Society of Agricultural Engineering 27(9), 287-292. DOI: 10.3969/j.issn.1002-6819.2011.09.050
Liu, Y., Wang, X., Xiong, Y., Tan, H., and Niu, Y. (2014). “Study of briquetted biomass co-firing mode in power plants,” Applied Thermal Engineering 63(1), 266-271. DOI: 10.1016/j.applthermaleng.2013.10.041
Molina, A., and Shaddix, C. R. (2007). “Ignition and devolatilization of pulverized bituminous coal particles during oxygen/carbon dioxide coal combustion,” Proceedings of the Combustion Institute 31(2), 1905-1912. DOI: 10.1016/j.proci.2006.08.102
Obernberger, I., and Thek, G. (2004). “Physical characterisation and chemical composition of densified biomass fuels with regard to their combustion behavior,” Biomass and Bioenergy 27(6), 653-669. DOI: 10.1016/j.biombioe.2003.07.006
Oyedun, A. O., Tee, C. Z., Hanson, S., and Chi, W. (2014). “Thermogravimetric analysis of the pyrolysis characteristics and kinetics of plastics and biomass blends,” Fuel Processing Technology 128, 471-481. DOI: 10.1016/j.fuproc.2014.08.010
Peng, X., Ma, X., and Xu, Z. (2015). “Thermogravimetric analysis of co-combustion between microalgae and textile dyeing sludge,” Bioresource Technology 180, 288-95. DOI: 10.1016/j.biortech.2015.01.023
Ren, Q., Zhao, C., Wu, X., Liang, C., Chen, X., Shen, J., and Wang, Z. (2009). “TG-FTIR study on co-pyrolysis of municipal solid waste with biomass,” Bioresource Technology 100(17), 4054-4057. DOI: 10.1016/j.biortech.2009.03.038
Roy, M. M., Dutta, A., and Corscadden, K. (2013). “An experimental study of combustion and emissions of biomass pellets in a prototype pellet furnace,” Applied Energy108, 298-307. DOI: 10.1016/j.apenergy.2013.03.044
Shen, B., and Lei, Q. (2006). “Study on MSW catalytic combustion by TGA,” Energy Conversion and Management 47(11–12), 1429-1437. DOI: 10.1016/j.enconman.2005.08.016
Skodras, G., Grammelis, P., Basinas, P., Kakaras, E., and Sakellaropoulos, G. (2006). “Pyrolysis and combustion characteristics of biomass and waste-derived feedstock,” Industrial and Engineering Chemistry Research 45(11), 3791-3799. DOI: 10.1021/ie060107g
Tabarés, J. L. M., Ortiz, L., Granada, E., and Viar, F. P. (2000). “Feasibility study of energy use for densificated lignocellulosic material (briquettes),” Fuel 79(10), 1229-1237. DOI: 10.1016/S0016-2361(99)00256-2
Tang, Y. T., Ma, X. Q., and Lai, Z. Y. (2011). “Thermogravimetric analysis of the combustion of microalgae and microalgae blended with waste in N2/O2 and CO2/O2 atmospheres,” Bioresource Technology 102(2), 1879-85.DOI: 10.1016/j.biortech.2010.07.088
Tang, Y. T., Ma, X. Q., Lai, Z. Y., and Fan, Y. (2015). “Thermogravimetric analyses of co-combustion of plastic, rubber, leather in N2/O2 and CO2/O2 atmospheres,” Energy 90, 1066-1074.DOI: 10.1016/j.energy.2015.08.015
Tian, Y., Ma, Z., Xu, Q., and Min, Z. (2013). “Features of biomass-waste co-combustion and environmental analysis,” Chinese Agricultural Science Bulletin 29(35), 193-198. DOI: 10.11924/j.issn.1000-6850.2013-0143
Wang, X., Li, D. K., Ni, W. D., Li, Z., and Zhang, H. D. (2007). “Combustion properties of pelletized biomass,” Journal of Combustion Science and Technology 13(1), 86-90. DOI: 10.3321/j.issn:1006-8740.2007.01.018
Zhou, H., Long, Y., Meng, A., Li, Q., and Zhang, Y. (2015). “Thermogravimetric characteristics of typical municipal solid waste fractions during co-pyrolysis,” Waste Management 38(1), 194-200. DOI: 10.1016/j.wasman.2014.09.027