Effect of CO2 during Coal Pyrolysis and Char Burnout in Oxy-Coal

May 18, 2011 - (7-11) Compared to combustion in air, the reduction in char burnout during ..... At pilot-scale or industrial-scale (2100–2300 K), co...
0 downloads 0 Views 1MB Size
ARTICLE pubs.acs.org/EF

Effect of CO2 during Coal Pyrolysis and Char Burnout in Oxy-Coal Combustion Prabhat Naredi and Sarma Pisupati* John and Willie Leone Department of Energy and Mineral Engineering, The EMS Energy Institute, The Pennsylvania State University, University Park, Pennsylvania 16802, USA ABSTRACT: One of the primary concerns in oxy-coal combustion is whether or not conventional coal pyrolysis and char burnout model can be used because of the presence of high CO2 in gas medium around coal particles. This paper attempts to understand the effect of high concentration of CO2 during coal pyrolysis and oxidation by conducting experiments in a drop-tube reactor and a thermogravimetric analyzer (TGA). Two ranks of bituminous coal particles were pyrolyzed in CO2 and Ar atmosphere during the pyrolysis tests and in 21% O2/79% CO2 and air during combustion tests. The results showed that the differences observed during pyrolysis and combustion tests conducted with and without CO2 can be attributed to the effects of charCO2 reaction. This result implies that charCO2 reaction should be included in model reactions while modeling coal pyrolysis or char oxidation in oxy-coal combustion. Furthermore, a faster increase in the rate of char burnout during combustion in O2/CO2 relative to that with air for an lvb coal was due to lower activation energy of CO2 and higher activation value of CCO2 reaction compared to that of an hvCb coal. This result implies that some coals may be more suitable for oxy-coal combustion which can be identified based on the intrinsic activation energy value of charCO2 and charO2 reactions.

’ INTRODUCTION Oxy-coal combustion has received considerable attention as one of the potential approaches to achieving a sequestration ready CO2 gas stream from power plants. Because costs involved in obtaining a pure oxygen stream and compressing flue gas are a significant barrier for the commercial application of this approach,1,2 a few studies have attempted to maximize the apparent advantages of this approach in terms of NOx levels3 and unburnt carbon (UBC),4 and developed computational tools5,6 to better optimize the approach. In various experimental studies conducted for char burnout, a trend of char burnout was observed to be in the order of 21% O2/79% CO2 < Air < 30% O2/70% CO2.711 Compared to combustion in air, the reduction in char burnout during combustion in a 21% O2/79% CO2 mixture is understandably caused by the reduction in gas temperature due to higher specific heat of CO2. A higher char burnout during combustion in a 30% O2/ 70% CO2 mixture than air can be conceived to be due to higher O2 partial pressure. However, some uncertainty remains regarding the role of charCO2 reaction in the overall char burnout process. For example, a classical study conducted by Mitchell and Madsen12 suggested that particle size (> 130 μm) plays a significant role in determining if CO2 content of the ambient gas would affect the rate of char burnout, whereas some of the recent studies show that the endothermic charCO2 reaction takes place even for finer particles.13,14 If CO2 does participate in rate of char conversion, then the question remains whether or not the lower particle temperature caused by the endothermic charCO2 reaction offsets its effect on overall rate of burnout? Recently, a large increase in char burnout was observed for higher ranks of coal than lower ranks during combustion in a 30% O2/ 70% CO2 mixture compared to that of combustion in air.15 However, observation of lower difference in char burnout r 2011 American Chemical Society

between combustion in a 30% O2/70% CO2 mixture and in air for lower ranks of coal was attributed to higher combustion rate and thus leaving a small margin for improvement during the tests. In experimental studies for NOx emissions, a reduction of up to 70% in NOx concentration has been measured in various pilotscale studies.9,16 This reduction has been mainly attributed to the result of the reaction between recycled NOx and hydrocarbon radicals present in the flame,17,18 and reduction in formation of thermal NOx.8 Moreover, an experimental study suggested that the reduction in NOx emissions also depends upon the coal type.19,20 In modeling studies related to oxy-coal combustion, one of the frequent assumptions has been to consider coal pyrolysis during oxy-coal combustion to be the same as in air.21 While a majority of the studies in the past have studied the effect of inert gas media2224 on weight loss or physical properties of resultant char particle from coal pyrolysis, studies to validate the assumption or to determine whether or not pyrolyzing the coal particles in CO2 is different from pyrolyzing in an inert gas medium are lacking.25 A change from an inert gas to a reactive gas such as CO2 can be expected to alter coal pyrolysis behavior because of reaction of the reactive gas with pyrolyzing coal and/or evolved volatile radical species or due to change in physical properties of the gas media. In the literature, it is well-known that a reactive gas such as H2 enhances the volatile yield during coal pyrolysis by reaction with coal char.25 However, Jenkins and Morgan26 attributed the observed increase in weight loss during wet N2, and CO2 atmosphere in comparison to pyrolysis in N2, to the reaction of released volatile matter with the reactive pyrolysis medium, i.e., a reduction in secondary condensation and redeposition of Received: February 4, 2011 Revised: May 2, 2011 Published: May 18, 2011 2452

dx.doi.org/10.1021/ef200197w | Energy Fuels 2011, 25, 2452–2459

Energy & Fuels

ARTICLE

Table 1. Proximate Analysis of a hvCb Coal (170 þ 200 Mesh) and Char Samples moisture (%)

fixed carbon (% dry)

vol. matter (% dry)

coal

3.36

51.84

Ar_1173

1.37

61.45

CO2_1173

1.37

62.31

ash (% dry)

C (%)

H (%)

N (%)

39.68

8.48

69.93

5.16

1.25

27.89

10.66

72.20

3.57

1.25

26.99

10.70

72.20

3.50

1.28

Ar_1373

1.82

73.57

14.75

11.68

76.85

2.37

1.31

CO2_1373

1.33

70.34

16.47

13.19

76.20

2.32

1.32

Ar_1573

1.59

81.03

6.48

12.49

79.55

1.35

1.19

CO2_1573

0.71

77.10

7.22

14.68

76.50

1.36

1.19

Ar_1673 CO2_1673

1.06 0.90

82.38 79.24

4.09 5.20

13.53 15.49

79.73 73.60

1.00 1.18

1.12 1.13

evolved volatile species. Their study considered the reaction between pyrolyzing gas media with coal particles to be negligible for the following two reasons: First, the rate of CH2O or CCO2 reaction measured in a thermogravimetric analyzer at a temperature of 1173 K was much lower (∼5 g/g-h) than the rate of additional weight loss observed during reactive gas media of Drop Tube Reactor (DTR) (∼860 g/g-h). Second, upon acid-washing the coal samples prior to DTR test, the weight loss values were measured to be the same for the three media employed. On the contrary, Peng et al.27 found that initial reactivity of the chars generated during direct gasification in H2O were higher (∼ six times) than those of corresponding chars prepared in nitrogen which was attributed to differences in microstructural rearrangements caused by steam interaction with the pyrolyzing coal particles. Some other studies have also observed differences in coal pyrolysis when an inert gas was replaced with a reactive gas during coal pyrolysis. For example, Messenbock et al.29 found a negligible increase in total volatile yield during the initial heating period when the gas atmosphere changed from either He to CO2 or from He to steam. But in late stages of heating period, the reactive gases caused weight loss at a higher rate (1020 g/g-h) than the expected rates (510 g/g-h) based on the previous reports of char gasification.30 Hayashi and co-workers3134 extensively studied the role of reactive gas atmosphere (steam/CO2) on the nascent char reactivity, and the interaction of the gas medium with evolved volatile species from a brown coal. In a DTR study conducted at 1173 K, an additional weight loss of ∼17% was measured within 4.3 s during steam pyrolysis of coal as compared to an inert pyrolysis in N2. A similar high level of reactivity of a nascent char was also observed for the case of a pyrolysis medium consisting of a CO2 atmosphere.33 Such a significantly higher conversion rate (∼120 g/g-h) was attributed to the catalytic char gasification reaction caused by inherent alkali species such as Na, Ca in the coal.31 The authors also proposed. based on their results, that the chemisorption of CO2 on the N-sites of nascent char surface may block access to N-sites for the -H radicals to form HCN and NH3.32 This proposition implies that the rate of coal-N evolution during oxy-coal combustion may be different than the combustion in air. In a very recent paper, Borrego and Alvarez35 observed that the chars prepared in O2/CO2 and O2/N2 environments in a DTR operating at 1573 K were similar in terms of subsequent reactivity and micropore surface area. However, a higher mesopore surface area was measured for the char prepared in O2/CO2 media as compared to the O2/N2 char. The difference in mesopore surface area for the char prepared in the two media was suggested to have an important implication for the diffusion controlled rate regime of carbon oxidation. Their study also

observed lower weight loss during pyrolysis in CO2 gas than the pyrolysis in an N2 gas medium which was proposed to be due to participation of CO2 in cross-linking reaction and thus preventing coalescence of aromatic clusters. Their result appears to be in contradiction to several published studies described earlier which report a higher weight loss in a CO2 medium. Also based on the reasoning of Borrego and Alvarez,35 the reactivity of the resultant disordered char sample (prepared in CO2) can be expected to be higher than the corresponding char prepared in N2.36 Contrary to expectation, the authors measured the same reactivity for the two char samples. In summary, the progress made in past studies has clarified that nascent char is more reactive than the thermally quenched char, and has suggested that coal pyrolysis in CO2 will have higher weight loss. However, the magnitude and mechanism (physical property change, chemical reaction in the gas phase, or chemical reaction with solid char or combination thereof) by which CO2 participates in coal pyrolysis or oxidation remains still unclear. Another significant question that remains to be answered is whether the rank of coal plays a role in determining whether or not coal pyrolysis and oxidation will be different in oxy-coal (O2/CO2) condition than air (O2/N2).

’ EXPERIMENTAL SECTION Methodology. To address the above-mentioned issues, coal pyrolysis and combustion experiments were conducted in a drop tube reactor (DTR) at different furnace temperatures (1173 1673 K) for two different ranks of coal. A high volatile (hvCb) coal and a low volatile (lvb) coal were used in this study to represent bituminous coal range. The hvCb coal with particle sizes between 74 and 90 μm and the lvb coal with particle sizes between 74 and 112 μm were obtained by wet sieving. To distinguish whether or not CO2 can affect the coal pyrolysis behavior because of changes in physical properties, the particle temperature profiles during the pyrolysis in the two gas media (CO2 and Ar) are predicted using the computational fluid dynamics (CFD) tool FLUENT. Coal Pyrolysis Experiment. Pyrolysis experiments were conducted in CO2 and Ar gas medium at different furnace temperatures at a residence time of about 0.5 s. The resultant char samples from pyrolysis were collected on filter paper and denoted by gas mediumtemperature. For example, a char sample produced in Ar medium at 1373 K was denoted as Ar_1373. Proximate analysis of the coal and the resultant pyrolyzed char particles is presented in Table 1. Table 1 also includes the ultimate analysis of the coal and the resultant char samples tested in the two gas media. 2453

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels Because the collection efficiency of the probe was less than 100% in the DTR tests, a direct measurement of weight loss during experiments was not possible. Therefore, weight loss was calculated using ash as a tracer technique.37 Surface area was calculated using BET method by N2 adsorption experiments at 77 K. Reactivity measurements were conducted by thermogravimetric analysis using a Perkin-Elmer TGA-7 instrument to compare the relative reactivity of DTR produced char samples from the two pyrolysis media. Char Reactivity Measurement Procedure. Typically, reactivity measurement was done using a sample mass of ∼23 mg and an operating temperature of 673 K. The TGA system was purged with N2 at 100 mL/min at room temperature for about 30 min to remove any air contamination in the system. The reactor was then heated to the reaction temperature at 20 K/min increments. The sample was held at the reaction temperature for ∼5 min to establish thermal equilibrium before switching to an air flow of 100 mL/min. Weight change was monitored as a function of time, and ash yield of the sample was measured for each run by raising the temperature to 1123 K. The measured data were converted to a dry ash free (daf) basis, and the rate, as a function of conversion, was calculated for each sample. Combustion Experiments. Combustion experiments were conducted in the DTR for low volatile (lvb) and high volatile coal (hvCb) samples in air and in a 21% O2/79% CO2 mixture at furnace temperatures between 1173 and 1773 K. Although a 30% O2/70% CO2 mixture, as an oxidizer, represents the oxy-coal combustion condition, a 21% O2/79% CO2 mixture was used in this study so that the role of CO2 could be comprehensively understood.

’ RESULTS AND DISCUSSION To determine how much contribution can be made by charCO2 reaction on char burnout, a theoretical calculation was done using an intrinsic combustion modeling approach where rate of char burnout was determined by both the charCO2 and charO2 reactions. Because a wide range of activation energy is reported for charO2 and charCO2 reaction,38 the activation energy of charO2 reaction is taken as an average value of 123 kJ/mol based on the general consensus.39 To illustrate the effect of charCO2 reaction, two extreme values (189 and 256 kJ/mol) of activation energy for charCO2 reaction were chosen where the values for coal-char may fall based on the coal rank and temperature history.40,41 Rate of combustion in air was calculated assuming an O2 partial pressure of 0.21 atm, and ignoring the contribution from charCO2 reaction. For rate of combustion in an O2/CO2 media, an additional contribution from rate of charCO2 reaction was considered using CO2 partial pressure of 0.79 atm. Because gas diffusivity is inversely proportional to square root of molecular weight,42 a 20% decrease in diffusivity occurs for O2CO2 mixture compared to N2, and is used in the calculations. Figure 1 shows the comparison of the rate of char burnout in the two combustion approaches against temperatures ranging from Zone II to Zone III conditions. The figure shows that at labscale combustion conditions (17001900 K), rate of char burnout during combustion in a 21% O2/79% CO2 mixture will always be lower than combustion in air even if activation energy value for charCO2 reaction is in the higher end of the reported values (g250 KJ/mol). In this theoretical comparison of rate of char burnout in the two media, combustion in a 21% O2/79%

ARTICLE

Figure 1. Theoretical calculation for the effect of surrounding CO2 levels on char combustion rate simulating late stages of combustion: (—) normalized rate for combustion in 20% O2; (---) lower O2 diffusivity in CO2 rich combustion media. Plot with symbols account for charCO2 reaction: (4) E = 189 kJ/mol; (0) E = 256 kJ/mol.

Table 2. Comparison of Char Oxidation and Gasification Rate Parameters between hvCb and lvb Rank Coals hvCb-1673

lvb-1673

A (g/(g.s.atm air)) E (kJ/mol) A (g/(g.s.atm air)) E (kJ/mol) charCO2 charO2

3.50  105 1.33  105

189 123

1.07  109 4.54  102

269 94

CO2 was considered to occur at lower temperature (200 K) than combustion in air due to the change in specific heat of the medium. At pilot-scale or industrial-scale (21002300 K), combustion in a 21% O2/79% CO2 mixture can cause higher rate of char burnout than combustion in air depending upon the activation energy value of gasification reactions of a particular char. Figure 1 can also be used to compare the effects of charCO2 reaction during combustion in air and in a 30% O2/70% CO2 mixture on rate of char burnout by comparing at the same particle temperatures. Such a comparison would indicate that charCO2 reaction will contribute in the rate of char burnout only if activation energy value for the reaction is g189 KJ/mol. These theoretical calculations indicated the importance of determining the activation energy of char reaction with O2 and CO2 for a particular coal to accurately understand the effect of CO2 on coal pyrolysis or oxidation. Also, the rate parameters for char-oxidation and char-gasification reaction were obtained using iso-thermal thermogravimetric analyzer and are shown in Table 2. The rate parameter values are calculated as an average value from the onset of a maximum to 80% conversion value in the reactivity plot. Effect of CO2 during Pyrolysis. Figure 2 shows computationally predicted particle temperature profiles in Ar and CO2 gas media at a DTR furnace wall temperature of 1673 K. These profiles in Figure 2 show that the particle heat-up rate in CO2 is slightly faster compared to that of Ar medium, and the final particle temperature attained by coal particles in CO2 medium is ∼20 K higher than that in Ar gas. The radiative property of CO2 2454

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels apparently has negligible effect on particle temperatures. Therefore, predicted minimal increase in particle heat-up rate and final particle temperature in CO2 medium compared to that of inert Ar medium is expected to cause negligible change in char properties. The weight loss from coal particles during pyrolysis tests is plotted against ASTM volatile matter loss in the two pyrolysis gas media and is shown in Figure 3. In general, a higher weight loss value is reported for samples pyrolyzed in CO2 than the volatile matter loss. As expected, similar weight loss (22.57%) at 1173 K in the two gas media suggests that switching the gas from an inert Ar to a reactive CO2 does not affect the pyrolysis at such low temperatures. However, at higher temperatures (for example at 1673 K), a higher weight loss does occur during pyrolysis in a CO2 medium (49.18%) compared to pyrolysis in Ar medium (40.22%). To investigate whether the additional weight loss is the result of CCO2 reaction or due to changes in the gas phase chemistry, weight loss value was calculated by employing the intrinsic rate parameter of charCO2 reaction. Table 3 shows the measured and model-predicted weight loss values. The data indicate that the observed additional weight loss for hvCb rank coal particles at 1673 K in a CO2 medium compared to that in an Ar medium is marginally higher (9.24%) than the calculated weight loss (2.78%) caused by CCO2 reaction. An increase in charCO2 reaction rate by a factor of 2 in model yields about the same weight loss as measured in the experiment. For the lvb coal, the

ARTICLE

predicted and measured weight loss values at 1673 K were nearly the same. Because various studies have suggested that a nascent char in the DTR is more reactive than a thermally stabilized char,43 the additional weight loss observed during pyrolysis in CO2 can be considered to be caused by charCO2 reaction. This result is in agreement with the observation of Messenbock et al.29 and is in contrast to the proposition in which observed additional weight loss in reactive CO2/H2O medium was attributed to the reduction in secondary gas phase reactions.27 To distinguish the effect of pyrolysis gas media on the coal-N distribution between the volatile and char fraction, the N/C ratio and weight loss of the samples is plotted against furnace temperature and is shown in Figure 4 for both ranks of coal. During pyrolysis in Ar, a reduction of up to 25% in N/C ratio of the char sample can be noted from Figure 4 in comparison to the coal sample. This result indicates that the fraction of coal-N released in the gas phase is higher as compared to the evolution of volatile matter. Also it can be noted for both ranks of coal that about 20% increase in N/C ratio occurred at higher temperature (1673 K) pyrolysis in CO2 compared to pyrolysis in Ar. Higher N/C ratio in char samples produced in CO2 compared to the char samples produced in Ar seems to be in agreement with the observation of Hayashi and co-workers32 that CO2 might be blocking the release of N atoms from char particles. However, the increase in N/C ratio may also occur because of higher carbon loss due to gasification reaction. A partial support for this reasoning is also found from the data of ultimate analysis where the carbon content (C) of char pyrolyzed in CO2 is less (73.6%) than that of the char prepared in Ar (79.55%) at higher a temperature (1673 K) but is similar at a lower temperature (up to 1373 K). To distinguish the effect of CO2 from inert Ar on coal pyrolysis, rate versus conversion and physical properties of char samples produced in the DTR were compared. Figure 5 shows the comparison of reactivity for the char samples produced in the Table 3. Additional Weight Loss during Coal Pyrolysis Due to CharCO2 Reactiona 1573 K hvCb coal

lvb coal

Figure 2. Effect of pyrolysis gas medium on particle temperature profile at 1673 K.

a

1673 K

ΔW (%) measured

4.23 ((2.5)

9.24 ((2.5)

ΔW (%) predicted using R

1.1

2.78

ΔW (%) predicted using R*2

4.59

7.98

ΔW (%) measured

4.56 ((2.5)

8.45 ((2.5)

ΔW (%) predicted using R

3.2

7.45

R is rate measured in TGA for char gasification reaction.

Figure 3. Comparison of the weight loss in Ar and CO2 medium during pyrolysis. 2455

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels

ARTICLE

Figure 6. Effect of prior weight loss on reactivity of char sample generated at 1373 K in Ar: (() weight loss 38%; ()) weight loss 43%.

Table 4. Comparison of Physical Properties of hvCb Char Generated in Ar and CO2 Gas

helium density

Figure 4. Effect of pyrolysis gas medium on split of coal-N into volatile and char fraction: (a) hvCb coal; (b) lvb coal.

Figure 5. Reactivity profiles measured in air at 673 K for DTR pyrolyzed chars generated at different temperatures.

two gas atmospheres. An excellent match of the reactivity profile for the char generated at the same temperatures in two gas media indicates that the chars have similar combustion characteristics. Slight differences observed in the rate profile for the chars generated at 1673 K are likely due to additional weight loss of the CO2_1673 sample. A decrease in char reactivity upon an increase in prior mass loss has been reported in the literature44 and has been verified in the study. For this purpose, the coal particles were pyrolyzed at 1373 K in the DTR for two different residence times (0.55 and 0.4 s) in Ar gas medium which had prior weight loss (daf) of 43% and 38%, respectively. Reactivity profiles of the

(cm3/g)

BET

crystallite

analysis

size

average pore volume

surface area

La

Lc

(cm3/g)

(m2/g)

(Å)

(Å)

8 8

8 9

coal

1.37

Ar_1173 CO2_1173

1.19 1.25

Ar_1373

1.36

0.0043

5

CO2_1373

1.45

0.0102

5

Ar_1573

1.46

0.0137

9

CO2_1573

1.65

0.0627

160

Ar_1673

1.59

0.0196

11

11

16

CO2_1673

1.78

0.125

181

13

18

0.00132 0.77 not detectable not detectable

resultant char samples show that the char sample whose prior weight loss was more had a 15% lower reactivity than the other char sample, as shown in Figure 6. Overall, these results suggest that char reactivity of the sample generated in CO2 is similar to that of char generated in an Ar gas medium. The data in Table 4 show that the BET surface area and pore volume of the char samples generated at lower temperatures (up to 1373 K) in CO2 medium (5 m2/g and 0.01 cm3/g) are similar to those of char samples generated in Ar medium (5 m2/g and 0.004 cm3/g). However, BET surface areas and pore volumes are larger for the chars generated at higher temperatures (1673 K) in CO2 medium (181 m2/g and 0.125 cm3/g) compared to the corresponding chars prepared in Ar medium (11 m2/g and 0.019 cm3/g). These results show a higher surface area for the sample prepared in CO2 which is in agreement with Borrego and Alvarez,35 but the apparent difference is due to the additional CCO2 reaction. Effect of CO2 during Combustion. Figure 7 shows comparison of experimentally measured char burnout during combustion in air with that of combustion in a 21% O2/79% CO2 gas medium for the hvCb and the lvb ranks of coal at furnace wall temperatures ranging from 1173 K to 1673 K. Several trends can 2456

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels

Figure 7. Comparison of char burnout in air and in 21% O2/79% CO2 gas medium. Open symbols (21% O2/79% CO2); closed symbols (air); (2,4) hvC coal; (•,O) lv coal.

ARTICLE

Figure 9. Comparison of NOx emission during combustion in air in 21% O2/79% CO2. Open symbols (21% O2/79% CO2); closed symbol (air); (2,4) hvCb coal; (•,O) lvb coal.

Figure 8. Ratio of char reactivity toward different gases for lvb and hvC rank coal.

be observed from the plots shown in Figure 7. First, char burnouts are lower during combustion in a 21% O2/79% CO2 mixture compared to that of combustion in air for both ranks of coal which is expected due to lower particle temperatures in CO2 rich media. However it can be noted that the difference in char burnouts between the two gas media diminishes as furnace temperature is increased. This observation can be explained by the fact that at higher temperatures, an increase in the contribution from the rate of charCO2 reaction possibly compensates for lower particle temperatures. The second point to be noted from Figure 7 is that the difference in char burnout between combustion in air and in a 21% O2/79% CO2 mixture at higher temperatures is lower for the lvb coal compared to that for the hvCb coal. To explain the possible cause, ratio of char reactivity toward O2 and CO2 is extrapolated using the rate parameter shown in Table 2 to higher temperatures for these two ranks of coal and is shown in Figure 8. The measured rate parameters (shown in Table 2) show higher activation of charCO2 reaction and lower activation of charO2 reaction for the lvb coal compared to that of the hvCb coal. These values of activation energy suggest that the overall

Figure 10. Comparison of CO emissions between combustion in air and in 21% O2/79% CO2. Open symbols (21% O2/79% CO2); closed symbol (air); (2,4) hvCb ; (•,O) lvb coal.

rate will increase more for the lvb coal than the hvCb coal upon an increase in temperature during CO2 rich oxy-combustion medium. Figure 8 shows that above 1050 K (at conditions of DTR tests) the contribution of charCO2 reaction becomes higher for the lvb coal than for the hvCb coal. Figure 9 shows comparison of experimentally measured NOx emissions in these two combustion media for the hvCb and lvb ranks of coal. Several observations can be made from the plots shown in Figure 9. First, as furnace temperature increases, NOx emissions increase for both the combustion media, and for both ranks of coal. In conventional combustion in air, an increase in NOx formation with an increase in temperature is expected because of an increase in contribution of thermal-NOx and an increase in conversion of coal-N into NOx. However, observation of a similar trend for combustion in a 21% O2/79% CO2 mixture indicates that thermal NOx is not a significant contributor in NOx formation during the lab-scale combustion tests, and thus an increase in NOx formation with increase in temperature might be 2457

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels occurring because of conversion of coal-N into NOx. Second, the NOx emissions are always lower during combustion in a 21% O2/ 79% CO2 mixture compared to that of combustion in air. A lower NOx emission from combustion in a 21% O2/79% CO2 mixture is expected due to a decrease in gas/particle temperatures (by ∼200 K) compared to that of combustion in air. Lower gas temperature in a 21% O2/79% CO2 media is expected to both decrease the conversion of intermediate NOx precursor species such as HCN and NH3 into NOx, and lower the char burnout due to an increase in retention of coal-N in the char (as observed in pyrolysis experiments). Figure 10 shows the comparison of measured CO emission from the DTR during the two combustion media. Figure 10 shows that as furnace temperature increases, the CO emissions decrease, and the CO emissions are always higher during combustion in a 21% O2/79% CO2 mixture compared to that of combustion in air. For both ranks of coal, the ratio of CO formation in a 21% O2/79% CO2 medium to that of combustion in air is higher at lower furnace temperatures and the ratio decreases as furnace temperatures increase. This result indicates that despite expected increase in CO emission with an increase in temperature due to contribution from charCO2 reaction for 21% O2/79% CO2 medium, a lower value is observed possibly due to the dominance of gas-phase mixing in overall CO emission.

’ CONCLUSIONS The main objective of this research was to determine whether or not CO2 rich media affects the coal pyrolysis and oxidation, and if possible identify the conditions at which such effect will be observed. For this purpose, two ranks of coal were pyrolyzed in Ar and CO2 gas media, and combusted in air and in a 21%O2/ 79% CO2 mixture in a drop tube reactor (DTR) operating at 11731773 K. A comparison of pyrolysis results in the two pyrolyzing media indicated that the properties of the char and the split of coal-N differ only at higher temperatures which are due to contribution from char gasification by the charCO2 reaction. The results implied that coal pyrolysis process remains the same between oxy-coal combustion and combustion in air, and to accurately predict the weight loss or gas phase composition in oxy-coal combustion, the contribution of charCO2 reaction in the model should be accounted for. A comparison of oxidation results in the two combustion media showed lower char burnouts, lower NOx emission, and higher CO emission in a 21% O2/ 79% CO2 medium compared to that of combustion in air for both the ranks of coal, and the difference was observed to diminish at higher furnace temperatures. The difference in char burnout in the two combustion media was also found to depend upon the rank of coal. Comparison of rate parameters of CO2 and CCO2 reaction for an lvb and bituminous coal indicated that more rapid increase in the rate of char burnout for an lvb coal was due to lower activation energy of CO2 and higher activation value of CCO2 reaction compared to that of an hvCb coal. Overall, the results indicated that to accurately predict combustion behavior for oxy-coal conditions, contribution from charCO2 reaction in the model should be considered. Second, the results implied that some coals may be more suitable for oxycoal combustion which can be identified based on the intrinsic activation energy value of the charCO2 and charO2 reactions.

ARTICLE

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Phone: 814-865-0874.

’ ACKNOWLEDGMENT The authors gratefully acknowledge the support of the EMS Energy Institute staff, particularly David Johnson, Ronald Wincek, and Ronald Wasco in conducting experiments in the droptube reactor. ’ REFERENCES (1) Santos, S.; Haines, M.; Davison, J.; Roberts, P. Challenges in the Development of Oxy-combustion Technology for Coal Fired Power Plants. In The Proceedings of 31st International Technical Conference on Coal Utilization & Fuel Systems, Clearwater, 2006. (2) Buhre, B. J. P.; Elliott, L. K.; Sheng, C. D.; Gupta, R. P.; Wall, T. F. Oxy-fuel combustion technology for coal-fired power generation. Prog. Energy Combust. Sci. 2005, 31, 283–307. (3) Chui, E. H.; Majeski, A. J.; Douglas, M. A.; Tan, Y.; Thambimuthu, K. V. Numerical investigation of oxy-coal combustion to evaluate burner and combustor design concepts. Energy 2004, 29 (910), 1285–1296. (4) Arias, B.; Pevida, C.; Rubiera, F.; Pis, J. J. Effect of biomass blending on coal ignition and burnout during oxy-fuel combustion. Fuel 2008, 87 (12), 2753–2759. (5) Gupta, R.; Khare, S.; Wall, T.; Eriksson, K.; Lundstrom, D.; Eriksson, J.; Spero, C. Adaptation of Gas Emissivity Models for CFD based Radiative Transfer in Large Air-Fired and Oxy- Fired Furnaces. In Proceedings of the 31st Technical Conference on Coal Utilization and Fuel Systems, Clearwater, 2006. (6) Khare, S. P.; Wall, T. F.; Farida, A. Z.; Liu, Y.; Moghtaderi, B.; Gupta, R. P. Factors influencing the ignition of flames from air-fired swirl of burners retrofitted to oxy-fuel. Fuel 2008, 87 (7), 1042–1049. (7) Li, X. C.; Rathnam, R. K.; Yu, J. L.; Wang, Q.; Wall, T.; Meesri, C. Pyrolysis and Combustion Characteristics of an Indonesian Low-Rank Coal under O2/N2 and O-2/CO2 Conditions. Energy Fuels 2010, 24, 160–164. (8) Croiset, E.; Thambimuthu, K.; Palmer, A. Coal combustion in O2/CO2 mixtures compared with air. Can. J. Chem. Eng. 2000, 78 (2), 402–407. (9) Miura, T. Advanced Coal Combustion; Nova Science Publishers, 2001; pp 185241. (10) Hu, Y.; Naito, S.; Kobayashi, N.; Hasatani, M. CO2, NOx and SO2 emissions from the combustion of coal with high oxygen concentration gases. Fuel 2000, 79 (15), 1925–1932. (11) Liu, H.; Zailani, R.; Gibbs, B. M. Comparisons of pulverized coal combustion in air and in mixtures of O2/CO2. Fuel 2005, 84 (78), 833–840. (12) Mitchell, R. E.; Madsen, O. H. Experimentally Determined Overall Burning Rates of Pulverized Coal Chars in Specified O2 and CO2 Environment. In 21st Symposium on Combustion, 1986; pp 173181. (13) Murphy, J. J.; Shaddix, C. R. Combustion kinetics of coal chars in oxygen-enriched environments. Combust. Flame 2006, 144 (4), 710–729. (14) Saastamoinen, J. J.; Aho, M. J.; Hamalainen, J. P.; Hernberg, R.; Joutsenoja, T. Pressurized pulverized fuel combustion in different concentrations of oxygen and carbon dioxide. Energy Fuels 1996, 10 (1), 121–133. (15) Arias, B.; Pevida, C.; Rubiera, F.; Pis, J. J. Effect of biomass blending on coal ignition and burnout during oxy-fuel combustion. Fuel 2008, 87 (12), 2753–2759. (16) Farzan, H.; Vecci, S. J.; Pelage, F. C.; Pranda, P.; Bose, A. C., Pilot-scale evaluation of coal combustion in an oxygen-enriched recylced flue gas. In 22nd International Pittsburgh Coal Conference, Pittsburgh, PA, 2005. 2458

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459

Energy & Fuels (17) Nozaki, T.; Takano, S.; Kiga, T.; Omata, K.; Kimura, N. Analysis of the flame formed during oxidation of pulverized coal by an O2-CO2 mixture. Energy 1997, 22 (23), 199–205. (18) Normann, F.; Andersson, K.; Leckner, B.; Johnsson, F. Emission control of nitrogen oxides in the oxy-fuel process. Prog. Energy Combust. Sci. 2009, 35 (5), 385–397. (19) Liu, H.; Zailani, R.; Gibbs, B. A. Pulverized coal combustion in air and in O2/CO2 mixtures with NOx recycle. Fuel 2005, 84 (16), 2109–2115. (20) Tan, Y. W.; Croiset, E.; Douglas, M. A.; Thambimuthu, K. V. Combustion characteristics of coal in a mixture of oxygen and recycled flue gas. Fuel 2006, 85 (4), 507–512. (21) Chui, E. H.; Douglas, M. A.; Tan, Y. W Modeling of oxy-fuel combustion for a western Canadian sub-bituminous coal. Fuel 2003, 82 (10), 1201–1210. (22) Banerjee, N. N.; Murty, G. S.; Rao, H. S.; Lahiri, A. Flash Pyrolysis of Coal - Effect of Nitrogen, Argon and Other Atmospheres in Increasing Olefin Concentration and Its Significance on Mechanism of Coal Pyrolysis. Fuel 1973, 52 (3), 168–170. (23) Eltawil, M. M.; Brown, L. F. Changes in Pore Structure of a Devolatilized Coal Char Upon Further Heating at Lower Temperature. Carbon 1976, 14 (2), 132–133. (24) Walker, P. L.; Pentz, L.; Biederman, D. L.; Vastola, F. J. Influence of Inert Diluent Gases on Rate of Carbon Gasification. Carbon 1977, 15 (3), 165–168. (25) Duan, L. B.; Zhao, C. S.; Zhou, W. Qu.; Chen, X. P. Investigation on Coal Pyrolysis in CO2 Atmosphere. Energy Fuels 2009, 23 (7), 3826–3830. (26) Anthony, D. B.; Hottel, H. C.; Howard, J. B.; Meissner, H. P. Rapid Devolatilization and Hydrogasification of Bituminous Coal. Abstr. Pap. Am. Chem. Soc. 1975, 170, 31–31. (27) Jenkins, R. G.; Morgan, M. E. Pyrolysis of a Lignite in an Entrained Flow Reactor 0.3. Pyrolysis in Reactive Atmospheres of Air, Carbon-Dioxide and Wet Nitrogen. Fuel 1986, 65 (6), 769–771. (28) Peng, F. F.; Lee, I. C.; Yang, R. Y. K. Reactivities of in-Situ and Ex-Situ Coal Chars During Gasification in Steam at 10001400-Degrees-C. Fuel Process. Technol. 1995, 41 (3), 233–251. (29) Messenbock, R.; Dugwell, D. R.; Kandiyoti, R. CO2 and steamgasification in a high-pressure wire-mesh reactor: the reactivity of Daw Mill coal and combustion reactivity of its chars. Fuel 1999, 78 (7), 781–793. (30) Laurendeau, N. M. Heterogenous Kinetics of Coal Char Gasification and Combustion. Prog. Energy Combust. Sci. 1978, 4, 221–270. (31) Hayashi, J. I.; Iwatsuki, M.; Morishita, K.; Tsutsumi, A.; Li, C. Z.; Chiba, T. Roles of inherent metallic species in secondary reactions of tar and char during rapid pyrolysis of brown coals in a drop-tube reactor. Fuel 2002, 81 (15), 1977–1987. (32) Chang, L. P.; Xie, Z. L.; Xie, K. C.; Pratt, K. C.; Hayashi, J.; Chiba, T.; Li, C. Z. Formation of NOx precursors during the pyrolysis of coal and biomass. Part VI. Effects of gas atmosphere on the formation of NH3 and HCN. Fuel 2003, 82 (10), 1159–1166. (33) Jamil, K.; Hayashi, J. I.; Li, C. Z., Pyrolysis of a victorian brown coal and gasification of nascent char in CO2 atmosphere in a wire-mesh reactor. Fuel 2004, 2004, (83), 833-843. (34) Bayarsaikhan, B.; Hayashi, J. I.; Shimada, T.; Sathe, C.; Li, C. Z.; Tsutsumi, A.; Chiba, T. Kinetics of steam gasification of nascent char from rapid pyrolysis of a Victorian brown coal. Fuel 2005, 84 (1213), 1612–1621. (35) Borrego, A. G.; Alvarez, D. Comparison of chars obtained under oxy-fuel and conventional pulverized coal combustion atmospheres. Energy Fuels 2007, 21 (6), 3171–3179. (36) Tamhankar, S. S.; Sears, J. T.; Chin-Yung, W. Coal pyrolysis at high temperatures and pressures. Fuel 1984, 63 (9), 1230–1235. (37) Badzioch, S.; Hawskley, P. G. W. Kinetics of thermal decomposition of pulverized coal particles. Ind. Eng. Chem. Process Des. Dev. 1970, 9 (4), 521–530.

ARTICLE

(38) Laurendeau, N. M. Heterogenous Kinetics of Coal Char Gasification and Combustion. Prog. Energy Combust. Sci. 1978, 4, 221–27038. (39) Russell, N. V.; Beeley, T. J.; Man, C. K.; Gibbins, J. R.; Williamson, J. Development of TG measurements of intrinsic char combustion reactivity for industrial and research purposes. Fuel Process. Technol. 1998, 57 (2), 113–130. (40) Knight, A. T.; Sergeant, G. D. Reactivity of Australian CoalDerived Chars to Carbon-Dioxide. Fuel 1982, 61 (2), 145–149. (41) Osafune, K.; Marsh, H. Gasification Kinetics of Coal Chars in Carbon-Dioxide. Fuel 1988, 67 (3), 384–388. (42) Smith, I. W.; Tyler, R. J. Internal Burning of Pulverized SemiAnthracite - Relation between Particle Structure and Reactivity. Fuel 1972, 51 (4), 312–321. (43) Megaritis, A.; Messenbock, R. C.; Collot, A. G.; Zhuo, Y.; Dugwell, D. R.; Kandiyoti, R. Internal consistency of coal gasification reactivities determined in bench-scale reactors: effect of pyrolysis conditions on char reactivities under high-pressure CO2. Fuel 1998, 77 (13), 1411–1420. (44) Gale, T. K.; Bartholomew, C. H.; Fletcher, T. H. Effects of pyrolysis heating rate on intrinsic reactivities of coal chars. Energy Fuels 1996, 10 (3), 766–775.

2459

dx.doi.org/10.1021/ef200197w |Energy Fuels 2011, 25, 2452–2459