发布时间：

2025-04-30

论文名称：

Effect of potassium-doping and oxygen concentration on soot oxidation in O2/CO2 atmosphere: A kinetics study by thermogravimetric analysis

发表刊物：

Energy Conversion and Management

摘要：

In oxy-fuel combustion, switching from nitrogen to carbon dioxide atmosphere with higher oxygen concentration will affect the oxidation rate of soot, and it was observed in our recent study on soot emission from biomass pyrolysis that potassium (K) crystals are embedded in soot and its precursor clusters. In this study, the effects of K-doping (KCl and K2SO4) and O2 concentration on soot oxidation in O2/CO2 atmosphere are studied using thermogravimetric analysis (TGA), and the extent of catalysis is compared with that in O2/N2 atmosphere. The delays on start, peak and end temperatures of soot oxidation are observed in O2/CO2 atmosphere. However, increase in O2 concentration which promotes oxidation significantly reduces the delay. All the K-doping cases results in accelerated soot oxidation rate, but catalytic role of the K-doping in O2/CO2 is significantly lower than that in O2/N2 because the CO2-enriched environment inhibits the performance of potassium as oxygen carrier. The accelerating degree from K-doping is also affected by the potassium type, doping mass and oxygen concentration. KCl acts as a better, more efficient doping agent than K2SO4 with the increase in doping mass. The catalytic effect of K2SO4 will not change and even decrease at 375 μmol(K)/g(soot) for K2SO4 while the catalytic role of KCl keeps increasing even at 600 μmol(K)/g(soot) for KCl. In O2 concentration range of 5–30%, the accelerating degree from K-doping presents the minimum value around 15%. This phenomenon strongly approves the hypothesis that potassium as the oxygen carrier and accelerating the oxygen transportation, because in the cases of without K-doping and at a high O2 concentration there is no additional active site for more O2 adsorption thus inducing the slow accelerating degree. The kinetic analysis indicates the first order reaction for soot oxidation and also a good compensation relation between apparent activation energy E and logarithmic frequency factor A. E is generally reduced with the atmosphere changing from O2/N2 to O2/CO2, with K-doping, and with O2 concentration decreasing. This study is beneficial to demonstrate the mechanism of how potassium doping and oxygen concentration affect soot oxidation rate in oxy-fuel combustion environment.  Keywords Soot oxidation; Oxy-fuel combustion; O2 concentration; Potassium; Catalysis effect 1. Introduction The awareness of the increase in greenhouse gas emission in large stationary sources has resulted in the development of new technologies that can accommodate capture and sequestration of carbon dioxide [1]. Oxy-fuel combustion has generated significant interest since it was proposed as a competitive carbon capture technology for new built and retrofitted coal-fired power plants, considering the advantages of a relatively moderate efficiency penalty and the lowest retrofit capital expenditure [2]. The concept of oxy-fuel combustion is removal of nitrogen from the oxidizer and then utilize flue gas recycle (mainly carbon dioxide and moisture) in its replacement for combustion process in oxygen [3]. The coal combustion process in oxy-fuel combustion is expected to be different with that in normal air combustion from the aspects of combustion chemistry, mass transfer, and radiation heat transfer. CO2 has a larger specific molar heat than N2, and the coal may be gasified by the CO2, thus the use of CO2 instead of N2 causes a reduction in the propagation speed, stability of the flame and increase in the unburned carbon content. These problems can be overcome by increasing oxygen concentration in oxy-fuel combustion [4], [5] and [6]. In the past decade, quite a lot of investigations have been made to probe into different challenges associated with oxy-fuel combustion. Recent studies have shown the flame radiation heat transfer under oxy-fuel conditions were strongly promoted, due to the higher levels of CO2 and H2O, as well as in-flame soot compared to air combustion conditions [7] and [8].  Soot formation in solid fuel combustion is regarded as a result of secondary pyrolysis of volatile matter especially tar based on polycyclic aromatic hydrocarbons hypothesis, inception, surface growth and surface oxidation [9]. Soot is one of the fine particle pollutants but also important to combustion facilities because of its strong radiation heat transfer effect, that the near-burner coal flame temperature could be lowered by several hundred degree due to the radiation heat transfer to surrounding wall by soot [10], [11] and [12]. After soot is formed in the flame region of reducing atmosphere at high temperatures, soot oxidation will occur in oxidative post-flame region under a relatively lower temperature as a result of attack by molecular oxygen, O2, and the OH radical [13]. The reaction kinetics of soot oxidation is important to the removal efficiency of soot in combustion facilities, and a number of studies have been performed on soot or carbon oxidation by experimental measure and modeling methods. The effects from oxygen partial pressure, temperature, particle size, catalysis and other gas species have been investigated [14], [15], [16], [17] and [18]. De Soete [19], Marsh and Kuo [20] proposed the reaction scheme of soot oxidation involving free carbon sites, chemisorbed localized molecular and atom oxygen, chemisorbed mobile molecular oxygen, and chemisorbed mobile atoms of oxygen. It indicated that the chemical absorption controls the total process with the reaction order around 1.0. Stanmore et al. [21] has reviewed the oxidation of soot from the aspects of experiments, mechanisms and models. Most of previous studies are under air combustion conditions, but recent preliminary studies on soot formation indicated a trend toward a larger soot cloud size and lower soot could temperature when replacing N2 with CO2. On the other hand, soot oxidation kinetics should be also changed in oxy-fuel combustion, because O2 concentration in oxy-fuel combustion is usually increased up to 30–40% and the high concentration of CO2 in oxy-fuel combustion also might enhance the interaction reaction between carbon and CO2[5], [22] and [23]. However, the studies on soot oxidation under oxy-fuel combustion conditions are limited.  Recently, we sampled soot particles and its precursors from biomass pyrolysis in a drop tube furnace at 1000–1300 °C and analyzed the morphology and elemental composition by Scanning Transmission Electron Microscopy (STEM, TECNAI G2 F30 TWIN, FEI, United States). In Fig. 1, potassium (K) crystals were found to be embedded in soot precursor clusters, with the potassium crystals formed through the condensation of potassium vapor. This indicates potassium in biomass not only catalytically affects the soot formation in devolatilization and tar decomposition as widely reported [24], [25], [26], [27] and [28], but also might play a catalytic role on soot oxidation at lower temperatures when potassium vapor condensed. The catalytic oxidation of soot has been widely studied in diesel engine to lower the ignition temperature and promote the burnout of diesel soot [29], [30] and [31]. The metal oxides (CuO, MoO3, F2O3, PbO, MnO2, Co3O4, La2O3, V2O5, Ag2O, etc.) [21] and some composite catalysts (Cu-V-K, Cu-Mn-K, Cu-Fe-K) have been tested and shown a significant catalysis on accelerating soot oxidation rate at low temperatures [31] and [32]. There are many studies on char combustion and gasification with loaded/doped alkali metals and alkaline earth metals [33], [34], [35] and [36] but the effect of potassium on soot oxidation is not yet fully explored. Considering the catalytic mechanism that metal atoms play in accelerating the transportation of O2 and CO2 on soot surfaces, it is reasonable to make the hypothesis that the catalytic acceleration degree of soot oxidation will be changed in oxy-fuel combustion in which there is a higher O2 concentration and enriched CO2 concentration compared with air combustion [37].   Fig. 1.  TEM image and elemental composition of soot precursors with potassium embedded during biomass pyrolysis. Figure options Based on the hypothesis above, in this study, the effects of K-doping and O2 concentration on model soot oxidation in O2/CO2 atmosphere are studied using thermogravimetric analysis (TGA), and the degree of influence of K-doping is compared with that in O2/N2 atmosphere. The effect of K-doping concentration and type (KCl, K2SO4) is also considered. The reaction order n, apparent activation energy E (kJ.mol−1) and frequency factor A (s−1) are further determined through a classical integral method for all the 29 cases in this work.  2. Material and methods 2.1. Material sample properties  The soot (PRINTEX U, Evonik Degussa, Germany) was used after 2 h drying at 105 °C in the cases without potassium doping. The major properties of soot used in TGA tests are shown in Table 1, with a low volatile content, no ash, large surface area, and nano-scale diameter. The morphology of the model soot used is shown in Fig. 2 analyzed by Transmission Electron Microscopy (TEM, HT7700, Hitachi, Japan), and the original particle size is in the scale of 20–30 nm.  Table 1. The properties of soot used in this study. Name Volatile content Ash content Spherical compacted density BET surface area Average original particle diameter Unit % % g/L m2/g nm Value 6 0.04 400 460 13  C content H content O content N content S content Unit wt.% wt.% wt.% wt.% wt.% Value 91.2 0.65 7.79 0.21 0.15 Table options  Fig. 2.  Microscopic morphology of soot in this study by STEM. Figure options 2.2. TGA equipment and test method  Soot oxidation test was conducted by using a TGA apparatus (STA409PC, NETZSCH, Germany). The fuel sample was heated in an alumina crucible under a desired atmosphere in the furnace. The weight loss was recorded on-line by the data recording system as fully described in [38]. In each test, about 5 mg soot samples were heated from ambient temperature to 900 °C with a heating rate of 20 °C min−1 and a total flow rate of 200 ml min−1. The oxygen concentration of O2/CO2 atmosphere was changed in the range of 5–40%. The measurement accuracies of TGA for temperature and mass were 0.1 °C and 0.2 μg, respectively. Thermogravimetric (TG) curves were obtained by continually recording the mass loss with increasing temperature. Derivative thermogravimetric (DTG) curves were obtained by differentiating TG curves.  2.3. K-doping method  The doping method being able to distribute K on soot surface “tightly” is important for a good catalysis, thus the dipping method is adopted in this study [32]. The original soot samples are immersed into KCl or K2SO4 solutions with stirring for 12 h and then dried in an oven of 80 °C for 24 h. After that the dried K-doping samples are ground for TGA tests. We define the K-doping concentration as the mole amount of K per gram soot sample, μmol(K)/g(soot). Three K-doping concentrations are selected in this study: 150, 375, and 600 μmol(K)/g(soot).  2.4. Characteristic parameters and kinetics analysis method  Three characteristic temperatures have been identified including the starting temperature Ts, the ending temperature Te, and the peak temperature Tp. The identification method is shown in Fig. 3 and has also been used in our previous works [39]. To evaluate the soot oxidation process more reasonably, the comprehensive oxidation index S was defined including the effects from Ts, Te, maximum mass loss rate wmax, and average mass loss rate wmean as: S = (wmax × wmean)/(Ts2 × Te).   Fig. 3.  The definition of characteristic parameters in TG-DTG curves of soot oxidation. Figure options The component of soot is simple and the differential curve of mass loss shows only one peak, therefore, Coats-Redfern integral method was adopted to determine the apparent activation energy E (kJ mol−1) and frequency factor A (s−1) [40]. Moreover, previous works indicated that the reaction order of soot oxidation should be about 1, thus f(a) = (1 − a) was selected as the reaction mechanism function. The reaction rate of soot oxidation at high temperatures could be described as following:  equation(1)  Turn MathJax on  by integrating, Eq. (2) was transformed into Eq. (3): equation(2)  Turn MathJax on  in which, a is the mass loss ratio, 1; β is the heating rate, K min−1; R is the idea gas constant, 8.314 J mol−1 K−1; T is the reaction temperature, K. Because in most cases 2RT/E ≪ 1,  could be regarded as constant, therefore, there should be a linear relation between  and 1/T. Using each pair of data value of a and T   in TG curves, we can plot new curves of  versus 1/T. When fitting these curves linearly, E and A could be determined by the slope and intercept of fitting linear function. 3. Results and discussion 3.1. Soot oxidation without potassium doping  3.1.1. Comparison between O2/N2 and O2/CO2 atmospheres  For the convenience of parameter comparison among different cases, the characteristics parameters of all the 29 cases have been listed in Table 2.  Table 2. The summary of characteristic parameters in all the cases of this study. Case # Doping Cpotassium Atmosphere Coxygen Ti Tpeak Tf wmax wmean S – – Mol/g – % °C °C °C %/min %/min %2 min−2 °C−3 1 – – O2/N2 5 632.9 726.1 778.4 12.633 2.453 9.927E−08 2 K2SO4 3.75 × 10−4 O2/N2 5 573.8 664.3 691.7 15.868 2.748 1.915E−07 3 KCl 3.75 × 10−4 O2/N2 5 573.9 646.5 692.6 15.488 2.731 1.854E−07 4 – – O2/N2 20 585 638.2 678.6 19.82 2.816 2.403E−07 5 K2SO4 1.5 × 10−4 O2/N2 20 533.5 601.1 625.6 20.959 3.085 3.631E−07 6 K2SO4 3.75 × 10−4 O2/N2 20 522.4 575.1 603.3 22.575 3.104 4.256E−07 7 K2SO4 6 × 10−4 O2/N2 20 517.4 574.4 601.1 21.930 3.088 4.208E−07 8 KCl 1.5 × 10−4 O2/N2 20 554.9 606.5 630.1 25.262 3.081 4.012E−07 9 KCl 3.75 × 10−4 O2/N2 20 539.3 580.3 604.3 27.380 3.086 4.807E−07 10 KCl 6 × 10−4 O2/N2 20 525.7 573.1 595.1 25.782 3.135 4.915E−07 11 – – O2/CO2 5 650.8 768.8 838.2 9.828 2.294 6.351E−08 12 K2SO4 3.75 × 10−4 O2/CO2 5 589.9 700.2 755.1 11.526 2.548 1.118E−07 13 KCl 3.75 × 10−4 O2/CO2 5 579.4 671.6 741.8 11.571 2.554 1.187E−07 14 – – O2/CO2 10 631.7 715.6 765.7 13.679 2.476 1.108E−07 15 K2SO4 3.75 × 10−4 O2/CO2 10 584.7 688.7 727.8 13.108 2.684 1.414E−07 16 KCl 3.75 × 10−4 O2/CO2 10 580.0 670.2 721.7 12.967 2.687 1.435E−07 17 – – O2/CO2 15 607.4 673.3 714.3 16.825 2.713 1.730E−07 18 – – O2/CO2 20 590.4 651.6 692.2 18.861 2.501 1.955E−07 19 K2SO4 1.5 × 10−4 O2/CO2 20 582.8 657.5 682.8 19.464 2.841 2.384E−07 20 K2SO4 3.75 × 10−4 O2/CO2 20 549.0 637.9 663.0 17.044 2.966 2.530E−07 21 K2SO4 6 × 10−4 O2/CO2 20 550.4 638.6 662.0 17.212 2.936 2.520E−07 22 KCl 1.5 × 10−4 O2/CO2 20 578.4 652.3 676.0 19.745 2.891 2.524E−07 23 KCl 3.75 × 10−4 O2/CO2 20 556.4 617.7 649.1 19.720 2.932 2.877E−07 24 KCl 6 × 10−4 O2/CO2 20 538.0 606.3 637.4 18.461 2.944 2.946E−07 25 – – O2/CO2 30 576.5 630.1 670.9 19.693 2.864 2.529E−07 26 K2SO4 3.75 × 10−4 O2/CO2 30 519.7 583.5 611.7 20.356 3.191 3.932E−07 27 KCl 3.75 × 10−4 O2/CO2 30 537.6 581.1 606.1 26.237 3.180 4.763E−07 28 – – O2/CO2 35 570.2 617.8 663.8 19.488 3.065 2.775E−07 29 – – O2/CO2 40 560.1 606.3 646.4 20.933 3.049 3.153E−07 Note: Cpotassium-potassium doping concentration (mol K per g soot); Coxygen-oxygen concentration (%).  Table options Without potassium doping, the comparisons of TG-DTG curves between O2/N2 and O2/CO2 atmospheres at 5% and 20% O2 concentrations are shown in Fig. 4. An obvious delay on soot oxidation can be observed in O2/CO2 atmosphere. For example, at 5% O2 concentration, Ts and Te increased from 632.9 °C to 650.8 °C and from 778.4 °C to 838.2 °C, respectively, when the atmosphere was changed from O2/N2 to O2/CO2. Two reasons have been ascribed to the delay: (1) the relative diffusivity of O2 in CO2 is lower than in N2; (2) the specific heat capacity of CO2 is larger than N2 and much higher concentration enhances the endothermic reaction of gasification between soot and CO2. Both of the reasons lead to decrease in the soot surface temperature [2]. When the oxygen concentration increases from 5% to 20%, the difference between O2/N2 and O2/CO2 atmosphere is much reduced. This is because at a higher oxygen concentration, the reactivity of soot oxidation is enhanced which induces the TG mass loss curves generally move to a lower temperature region with a lower starting temperature Ts of soot oxidation. As to the effect of atmosphere changing on oxidation index S, it decreases by 37% and 11% at oxygen concentration of 5% and 20%, respectively, when the atmosphere changes from O2/N2 to O2/CO2. This observation is similar with our previous works on bio-char combustion [41].   Fig. 4.  Comparison of TG-DTG curves between O2/N2 and O2/CO2 atmospheres. Figure options 3.1.2. Effect of oxygen concentration  In O2/CO2 atmosphere, the TG-DTG curves and characteristic parameter profiles at seven O2 concentrations (5%, 10%, 15%, 20%, 30%, 35%, 40%) are shown in Figs. 5 and 6. With the increasing in O2 concentration, the TG-DTG curve moves toward the lower temperatures, Ts, Tp, and Te decrease while wmax and S increases. When oxygen concentration increases from 5% to 40%, Ts, Tp, and Te decrease by 90.7 °C, 162.5 °C, and 191.8 °C, respectively; while S increases from 6.63E−8 to 3.15E−7 by about 4 times. The promotion of O2 concentration increasing is mainly due to the enhancement of O2 adsorption on soot particle surfaces. Moreover, Fig. 6 clearly shows that this promotion degree tends to be slow when O2 concentration increases to above 15%, which indicates the active sites for O2 on soot surfaces tend to become saturated.   Fig. 5.  TG-DTG curves at different oxygen concentrations (O2/CO2, without K-doping). Figure options  Fig. 6.  Effect of oxygen concentration on characteristic parameters (O2/CO2, without K-doping). Figure options 3.2. Soot oxidation with potassium doping  3.2.1. Effect of potassium doping and the catalytic mechanism hypothesis  The TG-DTG curves of soot oxidation with different K-doping types and concentrations under both O2/CO2 and O2/N2 atmospheres (20% O2 concentration) are shown in Fig. 7.   Fig. 7.  TG-DTG curves with different K-doping concentration, (a, b)-O2/CO2, (c, d)-O2/N2. (O2 = 20%; L-150 μmol/g; M-375 μmol/g; H-600 μmol/g). Figure options Under both atmospheres, it can be observed that both KCl and K2SO4 doping accelerates the soot oxidation rate. This catalytic acceleration mechanisms for carbon gasification or combustion with metal loaded have been classified as “electron-transfer theories” or “oxygen-transfer theories”, while more and more recent studies believe the later one “oxygen-transfer theories” can give a better explanation, especially in combustion process with adequate oxygen. In “oxygen-transfer theories”, it has been suggested that alkali metal atoms on carbon surface act as active sites for the chemisorption of oxygen, thereby weakening CC surface bonds and promoting the desorption of products CO and CO2[42] and [43], which has been proven through the temperature-programmed desorption test[44] and [45]. Based on the mechanisms of the alkali metal catalyzed gasification of carbon by McKee [46], the possible catalytic pathway of potassium on soot oxidation is illustrated in Fig. 8.   Fig. 8.  The proposed schematic action pathway of potassium playing the catalytic effect. Figure options As shown in Fig. 7, the doped K ions are easy to bond oxygen molecule and produce K2O2 by R1:  equation(R1) 2K+O2=K2O2 Turn MathJax on  The produced K2O2 on soot surfaces is easier to destroy the bond CC than O2 molecules by R2 or R3, and release the produced CO or CO2:  equation(R2) K2O2+C=K2O+CO Turn MathJax on  equation(R3) K2O2+C=2K+CO2 Turn MathJax on  The produced K2O in R2 could be continuously reduced into K releasing CO by R4, or bonded with another oxygen atom to form K2O2 by R5.  equation(R4) K2O+C=2K+CO Turn MathJax on  equation(R5) 2K2O+O2=2K2O2 Turn MathJax on  The reproduced K from R3 and R4 could be circularly reused to bond oxygen molecules through R1.  3.2.2. Effect of potassium doping type and concentration on catalysis degree (20% oxygen concentration)  Fig. 7 also shows that with further increase in doping concentration from 375 μmol/g to 600 μmol/g, the TG-DTG curves of these two cases (blue and green1 lines in Fig. 7) almost overlap for K2SO4-doping while a continuous catalytic effect is still observed for KCl-doping. Furthermore, plots of the extent of catalysis with the doping concentration are presented in Fig. 9. It is clearly observed that when K2SO4-doping concentration is over 375 μmol/g, the accelerating degree (delta start/peak/end temperatures) will be constant or even reduced (black lines, Fig. 9) which is different from what is observed with KCl-doping, as previously explained (red lines, Fig. 9). This “over-saturation” phenomenon of K2SO4-doping has been also observed in the previous studies on the catalytic effects of metal oxides on coal combustion [47]. In the study [47], it was explained that the over-loaded potassium species on soot surfaces blocked the transportation of gas species and increase mass transfer resistance. Hengel and Walker [48] measured the reactivity of lignite char with O2, CO2, and H2O with various calcium loadings, and also found that Ca loadings >4 wt% (about 1000 μmol/g) resulted in no further increase in char reactivity. The similar result was also discussed by Heek and Mühlen [49], and the dispersion conditions of mineral in carbon matrix were proposed as one of the most important factors.   Fig. 9.  Effect of K-doping concentration on the changes of soot oxidation characteristic parameters: (a) start temperature; (b) peak temperature; (c) end temperature; (d) oxidation index. Figure options The comparison result of Ts/Tp/Te between KCl- and K2SO4-doping indicates KCl-doping is generally more efficient, which is clearer in the comparison of comprehensive soot oxidation index S shown in Fig. 9d. Wu et al. [50] measured the gasification rate of Victorian brown coal with Na loading concentration scale comparable with our study, and also found the gasification rate monotonically increasing with Na loading mass increasing.  3.2.3. Effect of oxygen concentration on catalysis degree (O2/CO2, doping concentration = 375 μmol/g)  In oxy-fuel combustion, the oxygen concentration is commonly as high as 30% and even higher [2], therefore, in this part the soot oxidation measurement is conducted in eight cases: 5%, 10%, 20%, 30% O2 concentration with KCl- or K2SO4-doping. The TG-DTG curves under different oxygen concentration with K-doping is shown in Fig. 10. Similar to the trend in Fig. 5 without K-doping, the increasing in O2 concentration obviously promotes soot oxidation.   Fig. 10.  TG-DTG curves at different oxygen concentrations, (a) KCl-doping; (b) K2SO4-doping. Figure options To observe the detailed differences on the TG-DTG curve trend with O2 concentration changing between with and without K-doping cases, we plot the characteristic parameters of TG-DTG curves from Figs. 5 and 10(a)/(b) in Fig. 11. From Fig. 11 we can see for the cases with KCl- or K2SO4-doping (blue and red curves) these parameters generally change linearly, while for the cases without K-doping there is an obvious inflection point at O2 concentration about 15%. This leads to the minimum promotion degree that is referring to the gap between the black curves without K-doping and the blue/red curves with K-doping in Fig. 11. As the explanation in Section 3.1.2, the promotion degree tending to be slow at higher at O2 concentration above 15% is mainly due to there is no more active site for more O2 on soot surfaces before K-doping. In contrast, in the cases of K-doping, as we explained in Section 3.2.1, the loading of potassium on soot surfaces provides much more active sites for O2. More importantly, according to the hypothesis mechanism shown in Fig. 7, potassium ions and atoms as catalysts will be regenerated after bonding O2 and reacting with carbon to releasing CO or CO2, which seems like providing infinite active sites for O2 bonding. This phenomenon strongly proves the role of potassium as the oxygen carrier and accelerating the oxygen transportation.   Fig. 11.  Effect of oxygen concentration on the characteristic parameters of soot oxidation: (a) start temperature; (b) peak temperature; (c) end temperature; (d) oxidation index. Figure options 3.2.4. Comparison on the catalysis degree between O2/N2 and O2/CO2 (doping concentration = 375 μmol/g)  The comparison on the catalysis degree from K-doping between O2/N2 and O2/CO2 atmospheres can be also observed from Fig. 9, in which the changing degrees of the characteristic parameters under O2/N2 and O2/CO2 are solid lines and dash lines, respectively. It clearly shows the promotion degree in O2/CO2 is much lower than that in O2/N2, e.g. the promotion degree of the oxidation index S in O2/CO2 is only 1/3 of that in O2/N2. This might be ascribed to the much higher CO2 and CO concentrations surrounding soot particle surfaces in O2/CO2 atmosphere. According to the mechanism hypothesis in Section 3.2.1, with a very high CO2 and CO concentration, the produced CO2 and CO through the regeneration reactions R2/R3/R4 will be more difficult to be transported out from the soot surfaces and then inhibits these three regeneration reactions. Therefore, the regenerating rate to producing re-new active sites will be significantly reduced, which induces the obviously lower promotion degrees in O2/CO2 atmosphere.  3.3. Kinetics analysis on soot oxidation  By using Coats-Redfern integral method described in Section 2.4 and TG data, the fittings between  and 1/T for all the 29 cases in this study are compared in Fig. 12. All the linearly fitting curves have a fitting coefficient >0.995 when the first order reaction equation is adopted. Based on the slope and intercept of fitting curves, the reaction kinetics parameters E and A can be determined and listed in Table 3.   Fig. 12.  The Arrhenius plotting of all the tested cases of soot oxidation (fitting coefficient >0.995). Figure options Table 3. The summary of kinetics parameters in all the cases of this study. Case # Doping Cpotassium Atmosphere Coxygen E A ln A – – Mol/g – % kJ/mol 1/s – 1 – – O2/N2 5 124 1.59E+04 9.68E+00 2 K2SO4 3.75 × 10−4 O2/N2 5 120.49 3.86E+04 1.06E+01 3 KCl 3.75 × 10−4 O2/N2 5 129.22 1.20E+05 1.17E+01 4 – – O2/N2 20 175.68 8.32E+07 1.82E+01 5 K2SO4 1.5 × 10−4 O2/N2 20 195.05 5.98E+09 2.25E+01 6 K2SO4 3.75 × 10−4 O2/N2 20 201.77 2.89E+10 2.41E+01 7 K2SO4 6 × 10−4 O2/N2 20 188.6 4.59E+09 2.22E+01 8 KCl 1.5 × 10−4 O2/N2 20 213.75 6.32E+10 2.49E+01 9 KCl 3.75 × 10−4 O2/N2 20 215.91 1.78E+11 2.59E+01 10 KCl 6 × 10−4 O2/N2 20 186.02 3.27E+09 2.19E+01 11 – – O2/CO2 5 102.89 5.74E+02 6.35E+00 12 K2SO4 3.75 × 10−4 O2/CO2 5 96.48 6.63E+02 6.50E+00 13 KCl 3.75 × 10−4 O2/CO2 5 98.56 1.08E+03 6.98E+00 14 – – O2/CO2 10 134.33 6.92E+04 1.11E+01 15 K2SO4 3.75 × 10−4 O2/CO2 10 97.64 1.12E+03 7.02E+00 16 KCl 3.75 × 10−4 O2/CO2 10 104.98 3.21E+03 8.07E+00 17 – – O2/CO2 15 159.48 4.15E+06 1.52E+01 18 – – O2/CO2 20 182.69 1.42E+08 1.88E+01 19 K2SO4 1.5 × 10−4 O2/CO2 20 146.4 1.45E+06 1.42E+01 20 K2SO4 3.75 × 10−4 O2/CO2 20 119.36 5.76E+04 1.10E+01 21 K2SO4 6 × 10−4 O2/CO2 20 142.82 1.57E+06 1.43E+01 22 KCl 1.5 × 10−4 O2/CO2 20 144.81 1.30E+06 1.41E+01 23 KCl 3.75 × 10−4 O2/CO2 20 150.47 5.13E+06 1.54E+01 24 KCl 6 × 10−4 O2/CO2 20 141.4 1.90E+06 1.45E+01 25 – – O2/CO2 30 162.36 1.83E+07 1.67E+01 26 K2SO4 3.75 × 10−4 O2/CO2 30 167.92 1.96E+08 1.91E+01 27 KCl 3.75 × 10−4 O2/CO2 30 175.52 5.25E+08 2.01E+01 28 – – O2/CO2 35 166.73 3.81E+07 17.457 29 – – O2/CO2 40 179.52 3.12E+08 19.559 Note: The fitting coefficients determining E and A are above 0.995 for all the cases.  Table options To simplify the kinetics expression, Eq. (3) has been widely proposed to describe the compensation relation between E and A:  equation(3) lnA=a×E-b Turn MathJax on  where a and b represent the compensation factors. The compensation relations between E and ln A are plotted in Fig. 13, and there is a very good linear fitting between E and ln (A). This compensation relation for soot oxidation in TGA can be fitted as lnA = 0.163E−9.4, which is independent on atmosphere, K-doping type and concentration.   Fig. 13.  The compensation relationship between ln A and E of soot oxidation. Figure options In O2/CO2 atmosphere, the effect of K-doping and oxygen concentration on E is shown in Fig. 14. It shows that when the oxygen concentration is ⩽20%, K-doping significantly decreased E values and K2SO4-doping shows a lower E value than KCl-doping. However, when the oxygen concentration is ⩾30%, E values for cases without K-doping is generally stable around 173 kJ/mol due to the absence of active site on soot surfaces even at a higher O2 concentration. The similar phenomenon has also been observed in previous studies [51] and [52] on the effect of oxygen concentration on the kinetics of carbon material oxidation in CO2 enriched atmospheres. As we mentioned in Sections 3.1.2 and 3.2.3, when oxygen concentration is above a certain high value (15–20% in this study), the effect of further increasing in oxygen concentration on the overall reaction process is weakened a lot because there is no active site for more O2 adsorption on soot surfaces. This results in a more dynamic controlling situation at high O2 concentrations, which further leads to the apparent activation energy stable or fluctuating in a certain range.   Fig. 14.  Effect of the oxygen concentration on E value in O2/CO2 (K-doping concentration, 375 μmol/g). Figure options More interestingly, the E values of K-doping cases keep a continuous increasing even at a higher O2 concentration due to the significant enhancement of oxygen transportation under the catalytic effect of potassium. This leads to the E values of cases with K-doping exceed those cases without K-doping in the cases of O2 concentration above 30%. The values of E between O2/N2 and O2/CO2 are compared in Fig. 15 and each pair of data or cases differ at only using N2 or CO2 as balance gas. It clearly indicates a lower apparent activation energy in O2/CO2 atmosphere, which is mainly because the mass loss rate becomes slower referring to a smaller maximum DTG peak value in CO2-enriched environment as shown in Fig. 4.   Fig. 15.  Comparison on E values between O2/N2 and O2/CO2 atmosphere (1–10, O2/N2; 11–13, 18–24, O2/CO2). Figure options 4. Conclusions In this study, the effects of K-doping (KCl and K2SO4) and O2 concentration on soot oxidation in O2/CO2 atmosphere are studied using thermogravimetric analysis (TGA), and the extent of catalysis in cases with and without catalysis are compared with that in O2/N2 atmosphere. The main conclusions were drawn as follows:  The delays on start, peak and end temperatures of soot oxidation are observed in O2/CO2 atmosphere, while the delay degree decreases with the increasing in O2 concentration when oxidation is promoted. All the K-doping cases showed improved soot oxidation rate due to catalysis by potassium. However, the extent of catalysis by potassium is lower in O2/CO2 atmosphere, suggesting that CO2 environment inhibits the performance of potassium as oxygen carrier.  The accelerating degree from K-doping is also affected by the potassium type, doping mass and oxygen concentration. Generally, KCl presents a more efficient catalysis than K2SO4. Results using KCl as the doping agent showed continuous acceleration of soot oxidation even at 600 μmol(K)/g(soot) while doping based on K2SO4 plateaued and even decreased at 375 μmol(K)/g(soot) for K2SO4 as doping mass increases.  The effect of oxygen concentration on accelerating degree presents a minimum around 15% O2 concentration. This phenomenon strongly proves the hypothesis that potassium acts as oxygen carrier, accelerating oxygen transportation to soot surface, leading to promoted soot oxidation when compared to cases without K-doping. Also, in the cases without K-doping and with high O2 concentration, there is no additional active site for more O2 adsorption which explains less soot oxidation in comparison to K-doping cases.  The kinetic analysis indicates the first order reaction for soot oxidation and a good compensation relation between apparent activation energy E and logarithmic frequency factor A. It is observed that E generally reduced with the atmosphere changing from O2/N2 to O2/CO2, with K-doping, and with decreasing O2 concentration.  Acknowledgements This study was supported by the National Natural Science Foundation of China (Nos. 51676157, 51376147, 91544108), and National Key Research and Development Program of China (No. 2016YFC0801904).  References [1] B. Buhre, L. Elliott, C. Sheng, R. Gupta, T. Wall Oxy-fuel combustion technology for coal-fired power generation Prog Energy Combust Sci, 31 (2005), pp. 283–307   [2] L. Chen, S.Z. Yong, A.F. Ghoniem Oxy-fuel combustion of pulverized coal: characterization, fundamentals, stabilization and CFD modeling Prog Energy Combust Sci, 38 (2012), pp. 156–214   [3] M.B. Toftegaard, J. Brix, P.A. Jensen, P. Glarborg, A.D. Jensen Oxy-fuel combustion of solid fuels Prog Energy Combust Sci, 36 (2010), pp. 581–625   [4] M. Gil, J. Riaza, L. Álvarez, C. Pevida, J. Pis, F. Rubiera Oxy-fuel combustion kinetics and morphology of coal chars obtained in N2 and CO2 atmospheres in an entrained flow reactor Appl Energy, 91 (2012), pp. 67–74   [5] P.A. Bejarano, Y.A. Levendis Single-coal-particle combustion in O2/N2 and O2/CO2 environments Combust Flame, 153 (2008), pp. 270–287   [6] Y. Zhou, X. Jin, Q. Jin Numerical investigation on separate physicochemical effects of carbon dioxide on coal char combustion in O2/CO2 environments Combust Flame, 167 (2016), pp. 52–59   [7] F. Xia, Z. Yang, A. Adeosun, B.M. Kumfer, R.L. Axelbaum Control of radiative heat transfer in high-temperature environments via radiative trapping—Part I: Theoretical analysis applied to pressurized oxy-combustion Fuel, 172 (2016), pp. 81–88   [8] P. Chatterjee, J.L. de Ris, Y. Wang, S.B. Dorofeev A model for soot radiation in buoyant diffusion flames Proc Combust Inst, 33 (2011), pp. 2665–2671   [9] M. Frenklach Reaction mechanism of soot formation in flames Phys Chem Chem Phys, 4 (2002), pp. 2028–2037   [10] T.H. Fletcher, J. Ma, J.R. Rigby, A.L. Brown, B.W. Webb Soot in coal combustion systems Prog Energy Combust Sci, 23 (1997), pp. 283–301   [11] S.C. Cheung, R.K. Yuen, G. Yeoh, G.W. Cheng Contribution of soot particles on global radiative heat transfer in a two-compartment fire Fire Saf J, 39 (2004), pp. 412–428   [12] X. Bi, H. Liu, M. Huo, C. Shen, X. Qiao, C.-f.F. Lee Experimental and numerical study on soot formation and oxidation by using diesel fuel in constant volume chamber with various ambient oxygen concentrations Energy Convers Manage, 84 (2014), pp. 152–163   [13] I.M. Kennedy Models of soot formation and oxidation Prog Energy Combust Sci, 23 (1997), pp. 95–132   [14] X. Ma, C. Zangmeister, M. Zachariah Soot oxidation kinetics: a comparison study of two tandem ion-mobility methods J Phys Chem C, 117 (2013), pp. 10723–10729   [15] Z. Du, A.F. Sarofim, J.P. Longwell Activation energy distribution in temperature-programmed desorption: modeling and application to the soot oxygen system Energy Fuels, 4 (1990), pp. 296–302   [16] P. Gilot, F. Bonnefoy, F. Marcuccilli, G. Prado Determination of kinetic data for soot oxidation. Modeling of competition between oxygen diffusion and reaction during thermogravimetric analysis Combust Flame, 95 (1993), pp. 87–100   [17] P. Ciambelli, M. d'Amore, V. Palma, S. Vaccaro Catalytic oxidation of an amorphous carbon black Combust Flame, 99 (1994), pp. 413–421   [18] J.P. Neeft, T.X. Nijhuis, E. Smakman, M. Makkee, J.A. Moulijn Kinetics of the oxidation of diesel soot Fuel, 76 (1997), pp. 1129–1136   [19] De Soete G. Catalysis of soot combustion by metal oxides. Western States section meeting, Salt Lake City; 1988. p. 21–2.  [20] Marsh H, Kuo K. Kinetics and catalysis of carbon gasification, Introduction to Carbon Science. Butterworths, London; 1989.  [21] B.R. Stanmore, J.-F. Brilhac, P. Gilot The oxidation of soot: a review of experiments, mechanisms and models Carbon, 39 (2001), pp. 2247–2268   [22] K. Andersson, R. Johansson, F. Johnsson, B. Leckner Radiation intensity of propane-fired oxy-fuel flames: implications for soot formation Energy Fuels, 22 (2008), pp. 1535–1541   [23] C.R. Shaddix, A. Molina Particle imaging of ignition and devolatilization of pulverized coal during oxy-fuel combustion Proc Combust Inst, 32 (2009), pp. 2091–2098   [24] D.C. Elliott, E.G. Baker The effect of catalysis on wood-gasification tar composition Biomass, 9 (1986), pp. 195–203   [25] A. Jensen, K. Dam-Johansen, M.A. Wójtowicz, M.A. Serio TG-FTIR study of the influence of potassium chloride on wheat straw pyrolysis Energy Fuels, 12 (1998), pp. 929–938   [26] M. Obaidullah, S. Bram, V. Verma, J. De Ruyck A review on particle emissions from small scale biomass combustion Int J Renew Energy Res (IJRER), 2 (2012), pp. 147–159   [27] R. Zhang, R.C. Brown, A. Suby, K. Cummer Catalytic destruction of tar in biomass derived producer gas Energy Convers Manage, 45 (2004), pp. 995–1014   [28] Q. Ren, C. Zhao, X. Wu, C. Liang, X. Chen, J. Shen, et al. Catalytic effects of Fe, Al and Si on the formation of NOX precursors and HCl during straw pyrolysis J Therm Anal Calorim, 99 (2010), pp. 301–306   [29] J.P. Neeft, M. Makkee, J.A. Moulijn Catalysts for the oxidation of soot from diesel exhaust gases. I. An exploratory study Appl Catal B, 8 (1996), pp. 57–78   [30] E. Miró, F. Ravelli, M. Ulla, L. Cornaglia, C. Querini Catalytic combustion of diesel soot on Co, K supported catalysts Catal Today, 53 (1999), pp. 631–638   [31] W. Shangguan, Y. Teraoka, S. Kagawa Promotion effect of potassium on the catalytic property of CuFe2O4 for the simultaneous removal of NOx and diesel soot particulate Appl Catal B, 16 (1998), pp. 149–154   [32] G. Mul, J.P. Neeft, F. Kapteijn, M. Makkee, J.A. Moulijn Soot oxidation catalyzed by a Cu/K/Mo/Cl catalyst: evaluation of the chemistry and performance of the catalyst Appl Catal B, 6 (1995), pp. 339–352   [33] R.I. Backreedy, J.M. Jones, M. Pourkashanian, A. Williams Burn-out of pulverised coal and biomass chars Fuel, 82 (2003), pp. 2097–2105   [34] A. Nzihou, B. Stanmore, P. Sharrock A review of catalysts for the gasification of biomass char, with some reference to coal Energy, 58 (2013), pp. 305–317   [35] D.M. Quyn, H. Wu, C.-Z. Li Volatilisation and catalytic effects of alkali and alkaline earth metallic species during the pyrolysis and gasification of Victorian brown coal. Part I. Volatilisation of Na and Cl from a set of NaCl-loaded samples Fuel, 81 (2002), pp. 143–149   [36] R. Wagner, H.J. Mühlen Effect of a catalyst on combustion of char and anthracite Fuel, 68 (1989), pp. 251–253   [37] Y. Zhaosheng, M. Xiaoqian, L. Ao Thermogravimetric analysis of rice and wheat straw catalytic combustion in air- and oxygen-enriched atmospheres Energy Convers Manage, 50 (2009), pp. 561–566   [38] X. Wang, Z. Hu, S. Deng, Y. Wang, B. Wei, H. Tan Investigation on the synergetic Effect of biomass Co-firing in the atmosphere of O2/CO2 J Biobased Mater Bioenergy, 8 (2014), pp. 481–488   [39] X. Wang, J. Si, H. Tan, Y. Niu, C. Xu, T. Xu Kinetics investigation on the combustion of waste capsicum stalks in Western China using thermogravimetric analysis J Therm Anal Calorim, 109 (2011), pp. 403–412   [40] S. Niu, K. Han, C. Lu Characteristic of coal combustion in oxygen/carbon dioxide atmosphere and nitric oxide release during this process Energy Convers Manage, 52 (2011), pp. 532–537   [41] X. Wang, Z. Hu, S. Deng, Y. Wang, H. Tan Kinetics investigation on the combustion of biochar in O2/CO2 atmosphere Environ Prog Sustain Energy, 34 (2015), pp. 923–932   [42] R. Wagner, H.-J. Mühlen Effect of a catalyst on combustion of char and anthracite Fuel, 68 (1989), pp. 251–253   [43] B. Shen Qinlei. Study on MSW catalytic combustion by TGA Energy Convers Manage, 47 (2006), pp. 1429–1437   [44] I.L. Freriks, H.M. van Wechem, J.C. Stuiver, R. Bouwman Potassium-catalysed gasification of carbon with steam: a temperature-programmed desorption and Fourier Transform infrared study Fuel, 60 (1981), pp. 463–470   [45] M. Cerfontain, J.A. Moulijn Alkali-catalysed gasification reactions studied by in situ FTIR spectroscopy Fuel, 62 (1983), pp. 256–258   [46] D.W. McKee Mechanisms of the alkali metal catalysed gasification of carbon Fuel, 62 (1983), pp. 170–175   [47] L. Wei, D. Qi, R. Li Effects of alkali metal on combustion of pulverized coal and kinetic analysis J China Coal Soc, 10 (2010), pp. 1706–1711   [48] T.D. Hengel, P.L. Walker Jr. Catalysis of lignite char gasification by exchangeable calcium and magnesium Fuel, 63 (1984), pp. 1214–1220   [49] K.H. van Heek, H.J. Mühlen Aspects of coal properties and constitution important for gasification Fuel, 64 (1985), pp. 1405–1414   [50] H. Wu, J. Hayashi, T. Chiba, T. Takarada, C. Li Volatilisation and catalytic effects of alkali and alkaline earth metallic species during the pyrolysis and gasification of Victorian brown coal. Part V. Combined effects of Na concentration and char structure on char reactivity Fuel, 83 (2004), pp. 23–30   [51] S. Niu, C. Lu, K. Han, et al. Thermogravimetric analysis of combustion characteristics and kinetic parameters of pulverized coals in oxy-fuel atmosphere J Therm Anal Calorim, 98 (2009), pp. 267–274   [52] C. Wang, Y. Du, D. Che Reactivities of coals and synthetic model coal under oxy-fuel conditions Thermochim Acta, 553 (2013), pp. 8–15   Corresponding author. 1 For interpretation of color in ‘Figs. 7 and 9’, the reader is referred to the web version of this article.  © 2017 Elsevier Ltd. All rights reserved.  Note to users: Corrected proofs are Articles in Press that contain the authors' corrections. Final citation details, e.g., volume and/or issue number, publication year and page numbers, still need to be added and the text might change before final publication.  Although corrected proofs do not have all bibliographic details available yet, they can already be cited using the year of online publication and the DOI , as follows: author(s), article title, Publication (year), DOI. Please consult the journal's reference style for the exact appearance of these elements, abbreviation of journal names and use of punctuation.  When the final article is assigned to volumes/issues of the Publication, the Article in Press version will be removed and the final version will appear in the associated published volumes/issues of the Publication. The date the article was first made available online will be carried over.  About ScienceDirectRemote accessShopping cartContact and supportTerms and conditionsPrivacy policy Cookies are used by this site. For more information, visit the cookies page. Copyright © 2017 Elsevier B.V. or its licensors or contributors. ScienceDirect ® is a registered trademark of Elsevier B.V.  Recommended articles Numerical and experimental study of the influence of CO2 and N2 dilution on soot formation in laminar coflow C2H4/air diffusion flames at pressures between 5 and 20atm 2015, Combustion and Flamemore Effects of simultaneous hydrogen enrichment and carbon dioxide dilution of fuel on soot formation in an axisymmetric coflow laminar ethylene/air diffusion flame 2016, Combustion and Flamemore Air oxidation of carbon soot generated by laser ablation 1999, Carbonmore View more articles »  Citing articles (0)  Related book content

合写作者：

Xuebin Wang; Shuaishuai Li; Adewale Adeosun

是否译文：

否

发表时间：

2017-01-31