- Email: [email protected]
The radical and bond cleavage behaviors of 14 coals during pyrolysis with 9,10-dihydrophenanthrene
Fuel 182 (2016) 480–486
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
The radical and bond cleavage behaviors of 14 coals during pyrolysis with 9,10-dihydrophenanthrene Muxin Liu a,b,c, Jianli Yang a,c, Yong Yang a,c, Zhenyu Liu d,⇑, Lei Shi d, Wenjing He d, Qingya Liu d a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan, Shanxi Province 030001, China University of Chinese Academy of Sciences, Beijing 100049, China c National Energy Center for Coal to Liquids, Synfuels China Co., Ltd., Beijing 101400, China d State Key Laboratory of Chemical Recourse Engineering, Beijing University of Chemical Technology, Beijing 100029, China b
h i g h l i g h t s The stable and active radicals generated in pyrolysis of 14 coals are quantified. The coals contain unconvertible radicals that are likely to be in fusinite. The amounts of bonds cleaved in coal pyrolysis decrease with increasing coal rank. The radicals generated in pyrolysis are 3 order of magnitude more than that in coal. The bonds cleavage in coal pyrolysis with DHP follows the 1st order kinetics.
a r t i c l e
i n f o
Article history: Received 24 October 2015 Received in revised form 29 May 2016 Accepted 1 June 2016
Keywords: Coal pyrolysis Covalent bond Free radicals Kinetics Hydrogen donation
a b s t r a c t The pyrolysis of 14 coals with carbon contents (C%) of 67.5–94.9% are studied in the presence of 9,10dihydrophenanthrene (DHP) at 440 °C. The amounts of stable radicals in the coals and that generated in the pyrolysis are quantified by electron spin resonance (ESR). The amounts of active radicals generated in the pyrolysis are quantified by the amounts of hydrogen donated by DHP. The changes in quantity of these radicals during the pyrolysis are correlated with the parameters representing the coal rank (C%, the amounts of aromatic and aliphatic carbon). It is found that the quantity of stable radicals of the coals in the pyrolysis increases with an increase in C%. The lignites and bituminous coals break up significantly in the first 2 min in the pyrolysis. All the coals contain some rigid structures that do not break at 440 °C and the structure can be categorized to fusinite. The quantities of active radicals generated in the pyrolysis are approximately 3 orders of magnitude higher than that of the stable radicals in the coals. The total amounts of cleavable bonds in the coals and the rate constants of the bond cleavage are determined by the first order kinetics. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Pyrolysis is an important step in many coal conversion processes [1]. Although it is well recognized that pyrolysis of coals involves two primary reactions, i.e. the decomposition of coal macromolecules to generate primary volatile products and the reaction of the primary volatile products to yield final products, researches and knowledge on coal pyrolysis focused mainly on the yield and composition of final products [2,3]. This dissatisfaction was attributed by many to the complex nature of coals in
⇑ Corresponding author. E-mail address: [email protected] (Z. Liu). http://dx.doi.org/10.1016/j.fuel.2016.06.006 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
structure and in composition. The understanding of coal pyrolysis mechanism is still a subject that needs to be studied. The coal pyrolysis mechanism can also be described more fundamentally from the point of view of radicals. It is because that all the coals contain radicals [4] and the pyrolysis of coals starts with thermal cleavage of weak covalent bonds to generate radical fragments that is followed by reaction of volatile radical fragments to yield final products, such as gas, tar and coke [5,6]. Because a coal contains various types of covalent bonds of different dissociation energies [5], these two radical reactions occur not only sequentially but also concurrently in coal pyrolysis. Furthermore, the products of the second reaction may undergo further pyrolysis to yield additional radical fragments. In addition, the radical fragments may not be fully coupled through the second reaction and
481
M. Liu et al. / Fuel 182 (2016) 480–486
may remain in the final products, typically in coke and tar. These remained radicals were ascribed, at least partially, to the problems encountered in downstream processing of tar, such as increasing in viscosity, solidification during storage, transport and separation [7], as well as cracking and coking during preheating for refining. Clearly, it is very important to thoroughly understand the mechanism of radical generation and radical reaction, and the relationship between these radical behaviors and the covalent bonds in coals. To study the radical mechanism of coal pyrolysis, it is necessary to determine the radical concentrations in coal and in the pyrolysis products, especially during coal pyrolysis under various conditions. ESR has been used in-situ or ex-situ to measure radicals in coals and in coal pyrolysis products, which yields quantitative and qualitative results. The works that have been carried out include those on behavior of total radicals [8,9] and those on kinetic models [10,11]. However, because most of the radical reactions are too faster to be measured by ESR [12], ESR data show only stable radicals that are either poor in mobility or confined in rigid structures that constitutes steric hindrance to other radicals [7,13]. Therefore, many researchers doubted about the ESR’s capability in monitoring the active radicals. Studies have shown that the radicals generated in coal pyrolysis in the presence of hydrogen donor solvents (typically the partially hydrogenated aromatics, 9,10-dihydrophenanthrene (DHP), for example) can be coupled or stabilized by acquiring hydrogen from the solvents, during which the solvents convert to the corresponding aromatics [14,15]. This indicates that in the presence of a sufficient amount of a hydrogen donor solvent the majority of the active radicals generated in coal pyrolysis can be coupled by hydrogen from the donor solvent, which minimizes the coupling of coal radicals and allows determination of the quantity of active radicals by the amount of aromatics formed from the donor solvent. This method has been applied recently to a study of biomass pyrolysis [16], which showed that more than 99.9% of the radicals generated in the pyrolysis of lignin, cellulose and hemicellulose at temperatures of 350, 400 and 440 °C acquired hydrogen from DHP at a DHP to biomass mass ratio of 8 while the stable radicals measured by ESR accounted only less than 0.1% of the total radicals. To determine the quantities of the active and stable radicals in pyrolysis, and to correlate the changes in these radicals with coal rank, the behaviors of these radicals in pyrolysis of 14 coals in the presence of a sufficient amount of DHP were studied at 440 °C in this work. The relations between behaviors of the radicals and the properties of coals are studied, and a kinetic model is developed to show the relations.
Table 1 The ultimate and proximate analyses of the coal samples. Sample name
Xiaolongtan Shengli Daliuta Yilan Yanzhou Zaozhuang AU coking Qinglongshan Luan Ruqigou Qinshui Jingcheng Taixi Sihe
Sample code
XLT SL DLT YL YZ ZZ AUC QLS LA RQG QS JC TX SH
Proximate analysis (wt%)
Ultimate analysis (daf, wt%)
Mad
Ad
Vdaf
C
H
Oa
N
S
14.0 15.0 4.9 5.4 2.1 0.4 0.2 0.5 0.4 0.5 0.8 1.5 1.1 0.8
16.1 13.4 13.2 2.6 12.0 8.9 8.4 10.1 12.2 15.3 12.0 29.0 2.5 8.2
47.8 47.8 42.9 44.0 42.6 34.5 23.0 17.8 13.9 10.1 7.3 7.5 6.8 6.0
67.5 67.7 72.4 73.8 76.7 83.7 85.9 88.6 88.8 90.3 91.0 91.4 91.9 92.1
4.1 4.5 4.5 5.0 5.2 5.2 4.7 4.4 4.2 3.2 3.1 2.9 2.1 3.1
24.1 25.7 20.6 19.5 12.5 8.7 7.2 5.0 4.9 5.3 4.4 4.4 5 3.3
1.9 1.3 1.5 1.4 1.5 1.6 1.7 1.4 1.7 1.0 1.1 0.9 0.9 1.1
2.4 0.8 1.0 0.3 4.1 0.8 0.5 0.6 0.4 0.2 0.4 0.4 0.1 0.4
M: moisture; A: ash; V: volatile; ad: air dry; d: dry; daf: dry-ash-free. a By difference.
1 min before being sealed by a blast burner. Each of the reactors was then inserted into a quartz tube immersed in a fluidized sand bath maintained at 440 °C. The time required to heat the sample to 440 °C is less than 0.25 min. The quartz tube was removed from the sand bath at the designated time and quenched quickly in a water bath and then stored in liquid nitrogen. 2.3. ESR measurements The free radical concentration of a sample was measured at 18 °C by installing the glass-tube reactor directly in a Bruker EMXplus-10/12 ESR spectrometer operated at 9.85 GHz and 0.1 mW. The central magnetic field was 348 mT, the modulation amplitude was 1.0 G, the sweep width was 5 mT, the sweep time was 50 s and the time constant was 0.01 s. The signals were calibrated by 2,2-diphenyl-1-picrylhydrazyl (DPPH). The ESR signal intensity and radical concentration showed excellent linearity, and the reactor itself showed no ESR signal. No radical was detected in blank experiments (with DHP but without the coals). The amount of free radicals in a sample detected by ESR (RD) is calculated by Eq. (1):
RD ¼ NR =mdaf
ð1Þ
where NR is the amount (mol) of radicals in the sample and mdaf is the mass (g) of coal on the dry-ash-free basis. The relative deviation of RD estimated by parallel experiment is less than 10%.
2. Experimental 2.4. Determination of the amount of hydrogen donated by DHP 2.1. Materials A series of coal samples varying from lignite to anthracite are used, and their ultimate and proximate analyses are shown in Table 1. XLT and SL are lignites. YL, YZ, ZZ, AUC, QLS and LA are bituminous coals. RQG, QS, JC, TX and SH are anthracites. The coals were ground to pass 100 mesh sieve, and dried under a vacuum at 110 °C for 24 h. The hydrogen donor solvent DHP was purchased from Tokyo Chemical Industry Co., Ltd. and was used as received. 2.2. Pyrolysis experiments The pyrolysis experiments were performed in glass-tubereactors with 2 mm inside diameter and 38 mm length under a nitrogen atmosphere. Each of the reactors was charged with 2 mg coal and 0–16 mg DHP and then purged with nitrogen for
The amounts of DHP and its reaction products were determined by a high performance liquid chromatograph (HPLC) as reported previously [16]. DHP can donate hydrogen to radicals from coals by converting itself to phenanthrene (PHE). PHE may also form from disproportion and ring-opening reactions of DHP [16] but its amount is low (less than 4.7%) and can be determined by the amounts of 1,2,3,4-tetrahydrophenanthrene (THP) and 2-ethylbiphenyl (EBP) as reported previously [16]. Therefore, the amount of hydrogen donated to coal radicals, RH, is determined by Eq. (2) on the daf coal basis:
RH ¼ 2ðNP —NTHP —NEPB Þ=mdaf
ð2Þ
where NP, NTHP and NEPB are the amounts (mol) of PHE, THP and EPB determined by the HPLC analysis, respectively. The relative deviation of RH estimated by parallel experiment is less than 4%.
482
M. Liu et al. / Fuel 182 (2016) 480–486
It should note that DHP may dehydrogenate to form H2 and PHE, especially in the presence of a coal, which makes RH determined by Eq. (2) to be higher than it should be. To estimate this effect the amounts of H2 generated from the reaction of DHP with a coal at various conditions were analyzed using a gas chromatograph (SICT GC920 with a TCD detector and a TDX-01 carbon molecular sieve packed column). The results show that the amounts of H2 are very small in comparison to that donated by DHP, less than 5.4% in 30 min at 440 °C and a DHP to coal mass ratio of 8, for example. Considering the pyrolysis of coal may also generate H2, the H2 generated from dehydrogenation of DHP should be smaller than the GC result, indicating that the RH determined by Eq. (2) can be assumed to be fully acquired by coal radicals. The total amounts of DHP and PHE after the pyrolysis were found to account for more than 94.5% of the DHP added, suggesting that DHP and PHE do not react with the coal in the pyrolysis. 2.5.
13
C NMR spectra analysis
Molecular structure of the coals is investigated with solid-state C CP/MAS NMR spectroscopy. The experiments were performed on a Bruker AV 300 NMR spectrometer at a carbon frequency of 75.5 MHz. The spectra were recorded at a spinning speed of 12 kHz. The contact time was 3 ms, with a recycle delay of 5 s and a scan number of 2000. The amounts (mol/g) of aliphatic carbon, Cal, and aromatic carbon, Car, in 1 g coal are determined by the integral areas of aliphatic region (0–90 ppm) and aromatic region (100–165 ppm) in the 13C NMR spectra, respectively
13
3. Results and discussion 3.1. The stable radicals in coal and in coal pyrolysis To better understand behavior of the stable radicals in pyrolysis of the 14 coals, DLT coal (C% = 72.4) is studied first in detail to evaluate the effect of DHP quantity. Fig. 1 shows the RD during pyrolysis of the coal at 440 °C with different DHP to coal mass ratios. It can be seen that the RD of the coal, i.e. the datum at 0 min, is 0.99 105 mol/g. At 440 °C and in the absence of DHP, the RD doubles in 2 min, reaching to 2.0 105 mol/g, and then almost quadruples in 310 min, reaching to 3.8 105 mol/g. This monotone increasing trend in RD indicates that coupling of the active radicals generated from the pyrolysis of DLT coal in the absence of a hydrogen donor solvent results in the formation of largersize products that contain unpaired electrons [7,16]. This process is fast in the first few minutes, gets weak thereafter, but continues in 310 min. In the presence of DHP, the concentration of stable radicals shows a different behavior in comparison to that in the absence of DHP, the RD drops rapidly by about 80% in the first 2 min, to approximately 0.2 105 mol/g. This behavior may indicate that the DLT coal particles break up into pieces in 2 min at 440 °C, which exposes the stable radicals to DHP to acquire hydrogen. In pyrolysis time of longer than 2 min, the RD at a DHP to coal mass ratio of 2 increases again, indicating that the quantity of DHP is insufficient to donate hydrogen to all the radicals generated from the coal. When the DHP to coal ratio is higher than 2, however, the RD changes little beyond the first 2 min and remains in the range of 0.07–0.14 105 mol/g, at least in 30 min. This suggests that the quantities of DHP used are sufficient to donate hydrogen to coal radicals continuously to inhibit the coupling of coal radicals, and little stable radicals are formed from newly generated coal radicals. This also suggests that there is a certain amount of stable radicals in the coal, which are perhaps confined in some rigid structure that cannot break up at 440 °C, and therefore cannot
Fig. 1. The quantity of the stable radicals detected by ESR (RD) in pyrolysis of DLT coal at 440 °C in the presence of various amounts of DHP. (a) 0–310 min; (b) 0– 30 min.
acquire hydrogen even if there are plenty of DHP in the pyrolysis system. This type of rigid structure is likely to be that of fusinite which were reported to be unaltered in coal liquefaction processes [17]. In order to study the variation in quantity of the stable radicals in different coals during pyrolysis, the 14 coals were pyrolyzed at 440 °C under a sufficient hydrogen donation capability, i.e. a DHP to coal mass ratio of 8, and the radical concentrations were detected by ESR. The results in Fig. 2 show that the RD of the coals with carbon content of less than 90% (the lignites and the bituminous coals) decreases rapidly by 60–86% in 2 min, indicating significant breakup of the coal particles in the short time. Afterwards, the values of RD of the coals change slightly in 30 min and stay at low levels, corresponding to approximately 4–11% of the original radicals in the coals. These data indicate that all the coals contain rigid structures which cannot breakup at 440 °C. It is interesting to see that the values of RD of the coals with carbon content of
M. Liu et al. / Fuel 182 (2016) 480–486
483
Fig. 3. The changes of RD with carbon content of coals in pyrolysis at 440 °C and a DHP:coal of 8:1.
peaks and the peak temperatures were correlated with C% of the coals [5]. The discussion presented above for Figs. 2 and 3 may imply that the quantity of stable radicals is an important parameter in coal structure. The quantity of the stable radicals in the coals is relevant to their aromatic structures, because it has been reported that the aromatic nucleus with more condensed aromatic rings are beneficial for stabilization of radicals [19,20]. Furthermore, it was reported that the average number of condensed aromatic rings per aromatic nucleus is less than 4 for coals ranked lower than anthracites, but increases sharply to higher than 12 for anthracites [18]. Fig. 4 correlates the RD data with respect to the quantity of aromatic carbon (Car) in the coals, and shows a trend similar to that in Fig. 3, i.e. the RD increases slightly with an increase in Car when
Fig. 2. The quantity of radicals detected by ESR (RD) during pyrolysis of 14 coals at 440 °C and a DHP:coal of 8:1. (a) 0–30 min; (b) 0–3 min.
greater than 90% (the anthracites) show a different behavior, increasing in 0.5 min and then decreasing slightly reaching to the levels close to that of the coals in 30 min, which are much higher than that of coals with C% of lower than 90 under the same conditions, indicating that the anthracites particles do not break up at 440 °C. To better understand the changes in stable radical of coals of different rank the RD data in Fig. 2 (excluding the data in 0 min) are redisplayed in Fig. 3 with respect to the carbon content of the coals. It can be seen that the RD of the coals with C% of lower than 90 increases slowly with an increase in C%. However, the RD of coals with C% of higher than 90 increases sharply and then decreases. These behaviors are consistent with the changes in many physical properties of coals versus coal rank, such as heat capacity, heat conductivity, heat of wetting, solubility parameter and many others [18], and agree with the finding reported by Shi et al. in which DTG peak of 34 coals were deconvoluted into 6
Fig. 4. The changes of RD with Car of coals in pyrolysis at 440 °C and a DHP:coal of 8:1.
484
M. Liu et al. / Fuel 182 (2016) 480–486
the Car is in the range of 3.0–7.0 102 mol/g (the lignites and the bituminous coals), but increases rapidly with an increase in Car when the Car is higher than 7.0 102 mol/g (the anthracites). 3.2. The active radicals and covalent bonds cleaved in pyrolysis of the coals As discussed earlier, the radicals detected by ESR are the stable ones rather than the active ones, and the hydrogen from DHP couples not only the stable radicals originally in coals but also the active radicals generated in the pyrolysis, and the quantity of active radicals can be estimated by the amount of hydrogen donated by DHP. Fig. 5 shows the quantities of hydrogen donated by DHP, RH, over time in the pyrolysis of DLT coal at 440 °C and different DHP to coal mass ratios. Clearly the RH of DLT coal increases rapidly and reaches to values higher than 1.35 102 mol/g in 2 min at all the DHP to coal mass ratios. This RH value is
Fig. 5. The quantity of hydrogen donated by DHP (RH) to active radicals in pyrolysis of DLT coal at 440 °C. (a) 0–310 min; (b) 0–30 min.
approximately 6750 times of the RD value (less than 0.2 105 mol/g, Fig. 1) at the same time. Although the RD value at a DHP to coal mass ratio of 2 increases beyond 2 min, the RH value is still about 800 times of the RD value. When the DHP to coal mass ratio is higher than 2, the RH/RD ratio increases further over time. These behaviors indicate that the quantity of the active radicals generated in the pyrolysis of DLT coal at 440 °C is much higher than the amount of stable radicals detected by ESR. It can also be seen in Fig. 5 that at the DHP to coal mass ratios of 2, 4, 6 and 8, the increases in RH level off at 30, 130, 190 and 250 min, corresponding to the DHP conversions of 94.9, 95.6, 91.6, and 91.7%, respectively, indicating that the quantities of DHP are insufficient for hydrogen donation after these pyrolysis times. At the DHP to coal mass ratios of 8 the hydrogen donated to DLT coal is about 3.5 wt% at 30 min and 5.0 wt% at 70 min, which are in the hydrogen consumption range reported for direct coal liquefaction processes [21]. The similar values of RH in the first 30 min for DHP to coal mass ratios of 6 and 8 indicate that these DHP quantities do not limit the hydrogenation of active radicals generated in the pyrolysis, and the coupling of active coal radicals is greatly inhibited in 30 min at these DHP to coal mass ratios. Fig. 6 shows RH of the 14 coals at a DHP to coal mass ratio of 8. It can be seen that the RH of all the coals increases rapidly in the first
Fig. 6. The quantity of hydrogen donated by DHP (RH) to active radicals in pyrolysis of 14 coals at 440 °C and DHP:coal of 8:1. (a) 0–30 min; (b) 0–3 min.
M. Liu et al. / Fuel 182 (2016) 480–486
2 min and then slowly thereafter. Because the highest DHP conversion is less than 51% for all the coals in 30 min, the amount of DHP used is sufficient to allow estimation of the quantity of active radicals generated from each coal in the pyrolysis. Because the values of RH in Fig. 6 are much greater than that of RD in Fig. 2, they approximately equal to the quantity of active radicals generated in the pyrolysis. Consequently, the quantity of the covalent bonds cleaved in the pyrolysis for each of the coals can be estimated (termed NB,C), which is half of the RH value and is also shown in Fig. 6. Fig. 7 correlates the values of NB,C with C% of the 14 coals. It can be seen that the values of NB,C are in a level of 102 mol/g, indicating cleavage of 1–18 bonds per 100 carbon atoms in the pyrolysis. The value of NB,C increases over time, decreases with an increase in C% of the coals, and is very small for the coals with C% of higher than 90. These data suggest that the bonds cleaved at 440 °C in the coals are primarily aliphatic CAC and CAO bonds which decrease with an increase in coal rank as reported in the literature [5]. It also can be seen in Fig. 7 that the value of NB,C decreases linearly with an increase in C% of the coals with C% of lower than 90, and the shorter the pyrolysis time, the higher of the slope, but the slopes are almost identical for pyrolysis time of greater than 3 min. The higher slopes in the early pyrolysis stage may be attributed to the high contents of carbonyl and carboxyl groups in the low rank coals, which decompose early and may not follow the radical mechanism, i.e. do not acquire hydrogen from DHP. This is partially supported by the finding of Tschamler and Ruiter [22] in a pyrolysis study of resinite, in which the decomposition of carbonyl and carboxyl groups was reported involving internal hydrogen transfer. The identical slopes for the pyrolysis time of longer than 3 min may suggest that the bonds cleaved in these coals are the same and may be primarily attributed to the aliphatic CAC bond. Fig. 8 shows the relationship between NB,C and the quantity of aliphatic carbon (Cal) in the coals. The NB,C increases with increasing Cal in the coals but the trend is complex. The values of Cal of the anthracites (with C% greater than 90) are the lowest and close to 0.4 102 mol/g, and their corresponding NB,C values are also the lowest and close to each other; the values of Cal for the 4 high rank bituminous coals (ZZ, AUC, QLS and LA coals) vary significantly, from 0.6 102 to 1.9 102 mol/g, but the corresponding NB,C
Fig. 7. The quantity of bond cleaved (NB,C) in 14 coals at various time during pyrolysis at 440 °C and DHP:coal of 8:1.
485
Fig. 8. The quantity of bond cleavage (NB,C) of 14 coals with different aliphatic carbon contents (Cal) during pyrolysis at 440 °C and DHP:coal of 8:1.
values are similar; the values of Cal for the lignites and the 3 low rank bituminous coals (DLT, YL and YZ coals) are the highest (2.1–2.5 102 mol/g), and the corresponding values of NB,C are also relatively high. This behavior, although currently unclear, may suggest that the amount of the bond cleaved depends not only on the quantity but also on the structure of the alkyl chains in coals. 3.3. Kinetics of the covalent bond cleavage in the coals during pyrolysis The data in Fig. 5 and the discussion presented above suggest that the amounts of hydrogen donated by DHP in the first 30 min at a DHP to coal mass ratio of 8 equal to the amounts of active radicals generated from the coal in pyrolysis, which are twice as much as the covalent bonds cleaved in the coal. This understanding is supported by the work of Curran [14] where the rate of hydrogen donation by a hydrogen donor solvent was reported to be faster than that of the bonds cleavage in coals. Therefore, the data in Fig. 6 are used to estimate the rates of bond cleavage in the 14 coals in the first 30 min pyrolysis, using the first-order rate expression of Eq. (3), where NB is the amount of covalent bonds cleavable at 440 °C in a coal at time t, and k is the first-order rate constant. Integration of Eq. (3) yields Eq. (4), where NB,0 is the amount of covalent bonds cleavable in the coal at time zero. Because NB can also be expressed as the difference between NB,0 and NB,C (the amount of covalent bonds cleaved at time t) as shown in Eq. (5), Eq. (4) can be rewritten as Eq. (6). The values of NB,0 and k are then determined by nonlinear regression based on the data in Fig. 6.
dNB =dt ¼ kN B
ð3Þ
NB ¼ NB;0 ½expðktÞ
ð4Þ
NB ¼ NB;0 NB;C
ð5Þ
NB;C ¼ NB;0 ½1 expðktÞ
ð6Þ
Fig. 9 shows the NB,0 of the 14 coals determined from the nonlinear regression with a correlation coefficient (R2) of greater than 0.89. It can be seen that NB,0 decreases with an increase in C% of the coals. The decrease is slow for coals with C% of lower than 90, from
486
M. Liu et al. / Fuel 182 (2016) 480–486
exposed to DHP due to breakup of the coal particles, and those formed from cleavage of covalent bonds. The radicals that acquire hydrogen from DHP are approximately 3 orders of magnitude higher than that of the stable radicals in the coals measured by ESR. All the coals contain some rigid structures that do not break up at 440 °C. The anthracites contain much more rigid structure than the lignites and bituminous coals. The relations of the quantities of stable radicals and bond cleaved with carbon content of the coals in the pyrolysis are consistent with the dependences of many physical properties of coals and agree with the understandings of the coal structure versus carbon content. The first order kinetics of covalent bond cleavage shows that the quantities of the cleavable bonds in the coals decrease with the increase of coal rank, and these bonds become hard to cleave with the increase in coal rank. Acknowledgments The authors gratefully acknowledge the financial support from the National Basic Research Program of China (2011CB201300) and the Natural Science Foundation of China (21276019). Fig. 9. The quantity of bonds cleavable at 440 °C (NB,0) in 14 coals.
Fig. 10. The rate constant k of bond cleavage during pyrolysis of 14 coals at 440 °C.
1.8 102 to 1.4 102 mol/g, but much fast for coals with C% of higher than 90, reaching to approximately 0.5 102 mol/g for TX coal (C% = 92). Fig. 10 shows the rate constant k of the 14 coals estimated from the nonlinear regression. It can be seen that the values of k decrease with an increase in C% approximately, which indicates that the covalent bonds become hard to cleave with the increase of coal rank. 4. Conclusion The quantities of active radicals generated from coal in pyrolysis can be determined by the amounts of hydrogen donated by DHP. At a DHP to coal mass ratio of 8 the radicals generated from the coal pyrolysis are primarily coupled with hydrogen donated by DHP. These radicals include those originally in the coals but
References [1] Solomon PR, Fletcher TH, Pugmire RJ. Progress in coal pyrolysis. Fuel 1993;7:587–97. [2] Arenillas A, Rubiera F, Pis JJ, Cuesta MJ, Iglesias MJ, Jimenez A, et al. Thermal behaviour during the pyrolysis of low rank perhydrous coals. J Anal Appl Pyrol 2003;68:371–85. [3] Zhao Y, Hu H, Jin L, He X, Wu B. Pyrolysis behavior of vitrinite and inertinite from Chinese Pingshuo coal by TG–MS and in a fixed bed reactor. Fuel Process Technol 2011;92:780–6. [4] Petrakis L, Grandy DW. Electron spin resonance spectrometric study of free radicals in coals. Anal Chem 1978;50:303–8. [5] Shi L, Liu Q, Guo X, Wu W, Liu Z. Pyrolysis behavior and bonding information of coal—a TGA study. Fuel Process Technol 2013;108:125–32. [6] Liu Z, Shi S, Li Y. Coal liquefaction technologies-development in China and challenges in chemical reaction engineering. Chem Eng Sci 2010;65:12–7. [7] He W, Liu Z, Liu Q, Ci D, Lievens C, Guo X. Behaviors of radical fragments in tar generated from pyrolysis of 4 coals. Fuel 2014;134:375–80. [8] Petrakis L, Grandy DW. Formation and behaviour of coal free radicals in pyrolysis and liquefaction conditions. Nature 1981;289:476–7. [9] Petrakis L, Grandy DW, Jones GL. Free radicals in coal and coal conversions. 7. An in-depth experimental investigation and statistical correlative model of the effects of residence time, temperature and solvents. Fuel 1982;61:21–8. [10] Seehra MS, Ghosh B. Free radicals, kinetics and phase changes in the pyrolysis of eight american coals. J Anal Appl Pyrol 1988;13:209–20. [11] Petrakis L, Jones GL, Grandy DW, King AB. Free radicals in coal and coal conversions. 10. Kinetics and reaction pathways in hydroliquefaction. Fuel 1983;62:681–9. [12] Schulten Z, Schulten K. The generation, diffusion, spin motion, and recombination of radical pairs in solution in the nanosecond time domain. J Chem Phys 1977;66:4616–34. [13] Fowler TG, Bartle KD. Limitations of electron spin resonance spectroscopy in assessing the role of free radicals in the thermal reactions of coal. Energy Fuels 1989;3:515–22. [14] Curran GP, Struck RT, Gorin E. Mechanism of the hydrogen-transfer process to coal and coal extract. I&EC Process Des Dev 1967;6:166–73. [15] Kamiya Y, Nagae S. Relative reactivity of hydrogen donor solvent in coal liquefaction. Fuel 1985;64:1242–5. [16] Liu M, Yang J, Liu Z, He W, Liu Q, Li Y, et al. Cleavage of covalent bonds in the pyrolysis of lignin, cellulose, and hemicellulose. Energy Fuels 2015;29:5773–80. [17] Petrakis L, Grandy DW. Free radicals in coals and coal conversion. 4. Investigation of the free radicals in selected macerals upon liquefaction. Fuel 1981;60:120–4. [18] Elliott MA. Chemistry of coal utilization. New York: John Wiley & Sons, Inc.; 1981. [19] Retcofsky HL, Thompson GP, Hough M, Friedel RA. Electron spin resonance studies of coals and coal-derived asphaltenes. In: Larsen JW, editor. Organic chemistry of coal. Washington DC: American chemical society; 1978. p. 142–55. [20] Retcofsky HL, Stark JM, Friedel RA. Electron spin resonance in american coals. Anal Chem 1968;40:1699–704. [21] Comolli AG, Lee TLK, Popper GA, Zhou P. The Shenhua coal direct liquefaction plant. Fuel Process Technol 1999;59:207–15. [22] Tschamler H, Ruiter E. Coal science. In: Given P, editor. Adv chem series 55. Washington DC: American Chemical Society; 1966. p. 332–43.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
The radical and bond cleavage behaviors of 14 coals during pyrolysis with 9,10-dihydrophenanthrene Muxin Liu a,b,c, Jianli Yang a,c, Yong Yang a,c, Zhenyu Liu d,⇑, Lei Shi d, Wenjing He d, Qingya Liu d a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan, Shanxi Province 030001, China University of Chinese Academy of Sciences, Beijing 100049, China c National Energy Center for Coal to Liquids, Synfuels China Co., Ltd., Beijing 101400, China d State Key Laboratory of Chemical Recourse Engineering, Beijing University of Chemical Technology, Beijing 100029, China b
h i g h l i g h t s The stable and active radicals generated in pyrolysis of 14 coals are quantified. The coals contain unconvertible radicals that are likely to be in fusinite. The amounts of bonds cleaved in coal pyrolysis decrease with increasing coal rank. The radicals generated in pyrolysis are 3 order of magnitude more than that in coal. The bonds cleavage in coal pyrolysis with DHP follows the 1st order kinetics.
a r t i c l e
i n f o
Article history: Received 24 October 2015 Received in revised form 29 May 2016 Accepted 1 June 2016
Keywords: Coal pyrolysis Covalent bond Free radicals Kinetics Hydrogen donation
a b s t r a c t The pyrolysis of 14 coals with carbon contents (C%) of 67.5–94.9% are studied in the presence of 9,10dihydrophenanthrene (DHP) at 440 °C. The amounts of stable radicals in the coals and that generated in the pyrolysis are quantified by electron spin resonance (ESR). The amounts of active radicals generated in the pyrolysis are quantified by the amounts of hydrogen donated by DHP. The changes in quantity of these radicals during the pyrolysis are correlated with the parameters representing the coal rank (C%, the amounts of aromatic and aliphatic carbon). It is found that the quantity of stable radicals of the coals in the pyrolysis increases with an increase in C%. The lignites and bituminous coals break up significantly in the first 2 min in the pyrolysis. All the coals contain some rigid structures that do not break at 440 °C and the structure can be categorized to fusinite. The quantities of active radicals generated in the pyrolysis are approximately 3 orders of magnitude higher than that of the stable radicals in the coals. The total amounts of cleavable bonds in the coals and the rate constants of the bond cleavage are determined by the first order kinetics. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Pyrolysis is an important step in many coal conversion processes [1]. Although it is well recognized that pyrolysis of coals involves two primary reactions, i.e. the decomposition of coal macromolecules to generate primary volatile products and the reaction of the primary volatile products to yield final products, researches and knowledge on coal pyrolysis focused mainly on the yield and composition of final products [2,3]. This dissatisfaction was attributed by many to the complex nature of coals in
⇑ Corresponding author. E-mail address: [email protected] (Z. Liu). http://dx.doi.org/10.1016/j.fuel.2016.06.006 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.
structure and in composition. The understanding of coal pyrolysis mechanism is still a subject that needs to be studied. The coal pyrolysis mechanism can also be described more fundamentally from the point of view of radicals. It is because that all the coals contain radicals [4] and the pyrolysis of coals starts with thermal cleavage of weak covalent bonds to generate radical fragments that is followed by reaction of volatile radical fragments to yield final products, such as gas, tar and coke [5,6]. Because a coal contains various types of covalent bonds of different dissociation energies [5], these two radical reactions occur not only sequentially but also concurrently in coal pyrolysis. Furthermore, the products of the second reaction may undergo further pyrolysis to yield additional radical fragments. In addition, the radical fragments may not be fully coupled through the second reaction and
481
M. Liu et al. / Fuel 182 (2016) 480–486
may remain in the final products, typically in coke and tar. These remained radicals were ascribed, at least partially, to the problems encountered in downstream processing of tar, such as increasing in viscosity, solidification during storage, transport and separation [7], as well as cracking and coking during preheating for refining. Clearly, it is very important to thoroughly understand the mechanism of radical generation and radical reaction, and the relationship between these radical behaviors and the covalent bonds in coals. To study the radical mechanism of coal pyrolysis, it is necessary to determine the radical concentrations in coal and in the pyrolysis products, especially during coal pyrolysis under various conditions. ESR has been used in-situ or ex-situ to measure radicals in coals and in coal pyrolysis products, which yields quantitative and qualitative results. The works that have been carried out include those on behavior of total radicals [8,9] and those on kinetic models [10,11]. However, because most of the radical reactions are too faster to be measured by ESR [12], ESR data show only stable radicals that are either poor in mobility or confined in rigid structures that constitutes steric hindrance to other radicals [7,13]. Therefore, many researchers doubted about the ESR’s capability in monitoring the active radicals. Studies have shown that the radicals generated in coal pyrolysis in the presence of hydrogen donor solvents (typically the partially hydrogenated aromatics, 9,10-dihydrophenanthrene (DHP), for example) can be coupled or stabilized by acquiring hydrogen from the solvents, during which the solvents convert to the corresponding aromatics [14,15]. This indicates that in the presence of a sufficient amount of a hydrogen donor solvent the majority of the active radicals generated in coal pyrolysis can be coupled by hydrogen from the donor solvent, which minimizes the coupling of coal radicals and allows determination of the quantity of active radicals by the amount of aromatics formed from the donor solvent. This method has been applied recently to a study of biomass pyrolysis [16], which showed that more than 99.9% of the radicals generated in the pyrolysis of lignin, cellulose and hemicellulose at temperatures of 350, 400 and 440 °C acquired hydrogen from DHP at a DHP to biomass mass ratio of 8 while the stable radicals measured by ESR accounted only less than 0.1% of the total radicals. To determine the quantities of the active and stable radicals in pyrolysis, and to correlate the changes in these radicals with coal rank, the behaviors of these radicals in pyrolysis of 14 coals in the presence of a sufficient amount of DHP were studied at 440 °C in this work. The relations between behaviors of the radicals and the properties of coals are studied, and a kinetic model is developed to show the relations.
Table 1 The ultimate and proximate analyses of the coal samples. Sample name
Xiaolongtan Shengli Daliuta Yilan Yanzhou Zaozhuang AU coking Qinglongshan Luan Ruqigou Qinshui Jingcheng Taixi Sihe
Sample code
XLT SL DLT YL YZ ZZ AUC QLS LA RQG QS JC TX SH
Proximate analysis (wt%)
Ultimate analysis (daf, wt%)
Mad
Ad
Vdaf
C
H
Oa
N
S
14.0 15.0 4.9 5.4 2.1 0.4 0.2 0.5 0.4 0.5 0.8 1.5 1.1 0.8
16.1 13.4 13.2 2.6 12.0 8.9 8.4 10.1 12.2 15.3 12.0 29.0 2.5 8.2
47.8 47.8 42.9 44.0 42.6 34.5 23.0 17.8 13.9 10.1 7.3 7.5 6.8 6.0
67.5 67.7 72.4 73.8 76.7 83.7 85.9 88.6 88.8 90.3 91.0 91.4 91.9 92.1
4.1 4.5 4.5 5.0 5.2 5.2 4.7 4.4 4.2 3.2 3.1 2.9 2.1 3.1
24.1 25.7 20.6 19.5 12.5 8.7 7.2 5.0 4.9 5.3 4.4 4.4 5 3.3
1.9 1.3 1.5 1.4 1.5 1.6 1.7 1.4 1.7 1.0 1.1 0.9 0.9 1.1
2.4 0.8 1.0 0.3 4.1 0.8 0.5 0.6 0.4 0.2 0.4 0.4 0.1 0.4
M: moisture; A: ash; V: volatile; ad: air dry; d: dry; daf: dry-ash-free. a By difference.
1 min before being sealed by a blast burner. Each of the reactors was then inserted into a quartz tube immersed in a fluidized sand bath maintained at 440 °C. The time required to heat the sample to 440 °C is less than 0.25 min. The quartz tube was removed from the sand bath at the designated time and quenched quickly in a water bath and then stored in liquid nitrogen. 2.3. ESR measurements The free radical concentration of a sample was measured at 18 °C by installing the glass-tube reactor directly in a Bruker EMXplus-10/12 ESR spectrometer operated at 9.85 GHz and 0.1 mW. The central magnetic field was 348 mT, the modulation amplitude was 1.0 G, the sweep width was 5 mT, the sweep time was 50 s and the time constant was 0.01 s. The signals were calibrated by 2,2-diphenyl-1-picrylhydrazyl (DPPH). The ESR signal intensity and radical concentration showed excellent linearity, and the reactor itself showed no ESR signal. No radical was detected in blank experiments (with DHP but without the coals). The amount of free radicals in a sample detected by ESR (RD) is calculated by Eq. (1):
RD ¼ NR =mdaf
ð1Þ
where NR is the amount (mol) of radicals in the sample and mdaf is the mass (g) of coal on the dry-ash-free basis. The relative deviation of RD estimated by parallel experiment is less than 10%.
2. Experimental 2.4. Determination of the amount of hydrogen donated by DHP 2.1. Materials A series of coal samples varying from lignite to anthracite are used, and their ultimate and proximate analyses are shown in Table 1. XLT and SL are lignites. YL, YZ, ZZ, AUC, QLS and LA are bituminous coals. RQG, QS, JC, TX and SH are anthracites. The coals were ground to pass 100 mesh sieve, and dried under a vacuum at 110 °C for 24 h. The hydrogen donor solvent DHP was purchased from Tokyo Chemical Industry Co., Ltd. and was used as received. 2.2. Pyrolysis experiments The pyrolysis experiments were performed in glass-tubereactors with 2 mm inside diameter and 38 mm length under a nitrogen atmosphere. Each of the reactors was charged with 2 mg coal and 0–16 mg DHP and then purged with nitrogen for
The amounts of DHP and its reaction products were determined by a high performance liquid chromatograph (HPLC) as reported previously [16]. DHP can donate hydrogen to radicals from coals by converting itself to phenanthrene (PHE). PHE may also form from disproportion and ring-opening reactions of DHP [16] but its amount is low (less than 4.7%) and can be determined by the amounts of 1,2,3,4-tetrahydrophenanthrene (THP) and 2-ethylbiphenyl (EBP) as reported previously [16]. Therefore, the amount of hydrogen donated to coal radicals, RH, is determined by Eq. (2) on the daf coal basis:
RH ¼ 2ðNP —NTHP —NEPB Þ=mdaf
ð2Þ
where NP, NTHP and NEPB are the amounts (mol) of PHE, THP and EPB determined by the HPLC analysis, respectively. The relative deviation of RH estimated by parallel experiment is less than 4%.
482
M. Liu et al. / Fuel 182 (2016) 480–486
It should note that DHP may dehydrogenate to form H2 and PHE, especially in the presence of a coal, which makes RH determined by Eq. (2) to be higher than it should be. To estimate this effect the amounts of H2 generated from the reaction of DHP with a coal at various conditions were analyzed using a gas chromatograph (SICT GC920 with a TCD detector and a TDX-01 carbon molecular sieve packed column). The results show that the amounts of H2 are very small in comparison to that donated by DHP, less than 5.4% in 30 min at 440 °C and a DHP to coal mass ratio of 8, for example. Considering the pyrolysis of coal may also generate H2, the H2 generated from dehydrogenation of DHP should be smaller than the GC result, indicating that the RH determined by Eq. (2) can be assumed to be fully acquired by coal radicals. The total amounts of DHP and PHE after the pyrolysis were found to account for more than 94.5% of the DHP added, suggesting that DHP and PHE do not react with the coal in the pyrolysis. 2.5.
13
C NMR spectra analysis
Molecular structure of the coals is investigated with solid-state C CP/MAS NMR spectroscopy. The experiments were performed on a Bruker AV 300 NMR spectrometer at a carbon frequency of 75.5 MHz. The spectra were recorded at a spinning speed of 12 kHz. The contact time was 3 ms, with a recycle delay of 5 s and a scan number of 2000. The amounts (mol/g) of aliphatic carbon, Cal, and aromatic carbon, Car, in 1 g coal are determined by the integral areas of aliphatic region (0–90 ppm) and aromatic region (100–165 ppm) in the 13C NMR spectra, respectively
13
3. Results and discussion 3.1. The stable radicals in coal and in coal pyrolysis To better understand behavior of the stable radicals in pyrolysis of the 14 coals, DLT coal (C% = 72.4) is studied first in detail to evaluate the effect of DHP quantity. Fig. 1 shows the RD during pyrolysis of the coal at 440 °C with different DHP to coal mass ratios. It can be seen that the RD of the coal, i.e. the datum at 0 min, is 0.99 105 mol/g. At 440 °C and in the absence of DHP, the RD doubles in 2 min, reaching to 2.0 105 mol/g, and then almost quadruples in 310 min, reaching to 3.8 105 mol/g. This monotone increasing trend in RD indicates that coupling of the active radicals generated from the pyrolysis of DLT coal in the absence of a hydrogen donor solvent results in the formation of largersize products that contain unpaired electrons [7,16]. This process is fast in the first few minutes, gets weak thereafter, but continues in 310 min. In the presence of DHP, the concentration of stable radicals shows a different behavior in comparison to that in the absence of DHP, the RD drops rapidly by about 80% in the first 2 min, to approximately 0.2 105 mol/g. This behavior may indicate that the DLT coal particles break up into pieces in 2 min at 440 °C, which exposes the stable radicals to DHP to acquire hydrogen. In pyrolysis time of longer than 2 min, the RD at a DHP to coal mass ratio of 2 increases again, indicating that the quantity of DHP is insufficient to donate hydrogen to all the radicals generated from the coal. When the DHP to coal ratio is higher than 2, however, the RD changes little beyond the first 2 min and remains in the range of 0.07–0.14 105 mol/g, at least in 30 min. This suggests that the quantities of DHP used are sufficient to donate hydrogen to coal radicals continuously to inhibit the coupling of coal radicals, and little stable radicals are formed from newly generated coal radicals. This also suggests that there is a certain amount of stable radicals in the coal, which are perhaps confined in some rigid structure that cannot break up at 440 °C, and therefore cannot
Fig. 1. The quantity of the stable radicals detected by ESR (RD) in pyrolysis of DLT coal at 440 °C in the presence of various amounts of DHP. (a) 0–310 min; (b) 0– 30 min.
acquire hydrogen even if there are plenty of DHP in the pyrolysis system. This type of rigid structure is likely to be that of fusinite which were reported to be unaltered in coal liquefaction processes [17]. In order to study the variation in quantity of the stable radicals in different coals during pyrolysis, the 14 coals were pyrolyzed at 440 °C under a sufficient hydrogen donation capability, i.e. a DHP to coal mass ratio of 8, and the radical concentrations were detected by ESR. The results in Fig. 2 show that the RD of the coals with carbon content of less than 90% (the lignites and the bituminous coals) decreases rapidly by 60–86% in 2 min, indicating significant breakup of the coal particles in the short time. Afterwards, the values of RD of the coals change slightly in 30 min and stay at low levels, corresponding to approximately 4–11% of the original radicals in the coals. These data indicate that all the coals contain rigid structures which cannot breakup at 440 °C. It is interesting to see that the values of RD of the coals with carbon content of
M. Liu et al. / Fuel 182 (2016) 480–486
483
Fig. 3. The changes of RD with carbon content of coals in pyrolysis at 440 °C and a DHP:coal of 8:1.
peaks and the peak temperatures were correlated with C% of the coals [5]. The discussion presented above for Figs. 2 and 3 may imply that the quantity of stable radicals is an important parameter in coal structure. The quantity of the stable radicals in the coals is relevant to their aromatic structures, because it has been reported that the aromatic nucleus with more condensed aromatic rings are beneficial for stabilization of radicals [19,20]. Furthermore, it was reported that the average number of condensed aromatic rings per aromatic nucleus is less than 4 for coals ranked lower than anthracites, but increases sharply to higher than 12 for anthracites [18]. Fig. 4 correlates the RD data with respect to the quantity of aromatic carbon (Car) in the coals, and shows a trend similar to that in Fig. 3, i.e. the RD increases slightly with an increase in Car when
Fig. 2. The quantity of radicals detected by ESR (RD) during pyrolysis of 14 coals at 440 °C and a DHP:coal of 8:1. (a) 0–30 min; (b) 0–3 min.
greater than 90% (the anthracites) show a different behavior, increasing in 0.5 min and then decreasing slightly reaching to the levels close to that of the coals in 30 min, which are much higher than that of coals with C% of lower than 90 under the same conditions, indicating that the anthracites particles do not break up at 440 °C. To better understand the changes in stable radical of coals of different rank the RD data in Fig. 2 (excluding the data in 0 min) are redisplayed in Fig. 3 with respect to the carbon content of the coals. It can be seen that the RD of the coals with C% of lower than 90 increases slowly with an increase in C%. However, the RD of coals with C% of higher than 90 increases sharply and then decreases. These behaviors are consistent with the changes in many physical properties of coals versus coal rank, such as heat capacity, heat conductivity, heat of wetting, solubility parameter and many others [18], and agree with the finding reported by Shi et al. in which DTG peak of 34 coals were deconvoluted into 6
Fig. 4. The changes of RD with Car of coals in pyrolysis at 440 °C and a DHP:coal of 8:1.
484
M. Liu et al. / Fuel 182 (2016) 480–486
the Car is in the range of 3.0–7.0 102 mol/g (the lignites and the bituminous coals), but increases rapidly with an increase in Car when the Car is higher than 7.0 102 mol/g (the anthracites). 3.2. The active radicals and covalent bonds cleaved in pyrolysis of the coals As discussed earlier, the radicals detected by ESR are the stable ones rather than the active ones, and the hydrogen from DHP couples not only the stable radicals originally in coals but also the active radicals generated in the pyrolysis, and the quantity of active radicals can be estimated by the amount of hydrogen donated by DHP. Fig. 5 shows the quantities of hydrogen donated by DHP, RH, over time in the pyrolysis of DLT coal at 440 °C and different DHP to coal mass ratios. Clearly the RH of DLT coal increases rapidly and reaches to values higher than 1.35 102 mol/g in 2 min at all the DHP to coal mass ratios. This RH value is
Fig. 5. The quantity of hydrogen donated by DHP (RH) to active radicals in pyrolysis of DLT coal at 440 °C. (a) 0–310 min; (b) 0–30 min.
approximately 6750 times of the RD value (less than 0.2 105 mol/g, Fig. 1) at the same time. Although the RD value at a DHP to coal mass ratio of 2 increases beyond 2 min, the RH value is still about 800 times of the RD value. When the DHP to coal mass ratio is higher than 2, the RH/RD ratio increases further over time. These behaviors indicate that the quantity of the active radicals generated in the pyrolysis of DLT coal at 440 °C is much higher than the amount of stable radicals detected by ESR. It can also be seen in Fig. 5 that at the DHP to coal mass ratios of 2, 4, 6 and 8, the increases in RH level off at 30, 130, 190 and 250 min, corresponding to the DHP conversions of 94.9, 95.6, 91.6, and 91.7%, respectively, indicating that the quantities of DHP are insufficient for hydrogen donation after these pyrolysis times. At the DHP to coal mass ratios of 8 the hydrogen donated to DLT coal is about 3.5 wt% at 30 min and 5.0 wt% at 70 min, which are in the hydrogen consumption range reported for direct coal liquefaction processes [21]. The similar values of RH in the first 30 min for DHP to coal mass ratios of 6 and 8 indicate that these DHP quantities do not limit the hydrogenation of active radicals generated in the pyrolysis, and the coupling of active coal radicals is greatly inhibited in 30 min at these DHP to coal mass ratios. Fig. 6 shows RH of the 14 coals at a DHP to coal mass ratio of 8. It can be seen that the RH of all the coals increases rapidly in the first
Fig. 6. The quantity of hydrogen donated by DHP (RH) to active radicals in pyrolysis of 14 coals at 440 °C and DHP:coal of 8:1. (a) 0–30 min; (b) 0–3 min.
M. Liu et al. / Fuel 182 (2016) 480–486
2 min and then slowly thereafter. Because the highest DHP conversion is less than 51% for all the coals in 30 min, the amount of DHP used is sufficient to allow estimation of the quantity of active radicals generated from each coal in the pyrolysis. Because the values of RH in Fig. 6 are much greater than that of RD in Fig. 2, they approximately equal to the quantity of active radicals generated in the pyrolysis. Consequently, the quantity of the covalent bonds cleaved in the pyrolysis for each of the coals can be estimated (termed NB,C), which is half of the RH value and is also shown in Fig. 6. Fig. 7 correlates the values of NB,C with C% of the 14 coals. It can be seen that the values of NB,C are in a level of 102 mol/g, indicating cleavage of 1–18 bonds per 100 carbon atoms in the pyrolysis. The value of NB,C increases over time, decreases with an increase in C% of the coals, and is very small for the coals with C% of higher than 90. These data suggest that the bonds cleaved at 440 °C in the coals are primarily aliphatic CAC and CAO bonds which decrease with an increase in coal rank as reported in the literature [5]. It also can be seen in Fig. 7 that the value of NB,C decreases linearly with an increase in C% of the coals with C% of lower than 90, and the shorter the pyrolysis time, the higher of the slope, but the slopes are almost identical for pyrolysis time of greater than 3 min. The higher slopes in the early pyrolysis stage may be attributed to the high contents of carbonyl and carboxyl groups in the low rank coals, which decompose early and may not follow the radical mechanism, i.e. do not acquire hydrogen from DHP. This is partially supported by the finding of Tschamler and Ruiter [22] in a pyrolysis study of resinite, in which the decomposition of carbonyl and carboxyl groups was reported involving internal hydrogen transfer. The identical slopes for the pyrolysis time of longer than 3 min may suggest that the bonds cleaved in these coals are the same and may be primarily attributed to the aliphatic CAC bond. Fig. 8 shows the relationship between NB,C and the quantity of aliphatic carbon (Cal) in the coals. The NB,C increases with increasing Cal in the coals but the trend is complex. The values of Cal of the anthracites (with C% greater than 90) are the lowest and close to 0.4 102 mol/g, and their corresponding NB,C values are also the lowest and close to each other; the values of Cal for the 4 high rank bituminous coals (ZZ, AUC, QLS and LA coals) vary significantly, from 0.6 102 to 1.9 102 mol/g, but the corresponding NB,C
Fig. 7. The quantity of bond cleaved (NB,C) in 14 coals at various time during pyrolysis at 440 °C and DHP:coal of 8:1.
485
Fig. 8. The quantity of bond cleavage (NB,C) of 14 coals with different aliphatic carbon contents (Cal) during pyrolysis at 440 °C and DHP:coal of 8:1.
values are similar; the values of Cal for the lignites and the 3 low rank bituminous coals (DLT, YL and YZ coals) are the highest (2.1–2.5 102 mol/g), and the corresponding values of NB,C are also relatively high. This behavior, although currently unclear, may suggest that the amount of the bond cleaved depends not only on the quantity but also on the structure of the alkyl chains in coals. 3.3. Kinetics of the covalent bond cleavage in the coals during pyrolysis The data in Fig. 5 and the discussion presented above suggest that the amounts of hydrogen donated by DHP in the first 30 min at a DHP to coal mass ratio of 8 equal to the amounts of active radicals generated from the coal in pyrolysis, which are twice as much as the covalent bonds cleaved in the coal. This understanding is supported by the work of Curran [14] where the rate of hydrogen donation by a hydrogen donor solvent was reported to be faster than that of the bonds cleavage in coals. Therefore, the data in Fig. 6 are used to estimate the rates of bond cleavage in the 14 coals in the first 30 min pyrolysis, using the first-order rate expression of Eq. (3), where NB is the amount of covalent bonds cleavable at 440 °C in a coal at time t, and k is the first-order rate constant. Integration of Eq. (3) yields Eq. (4), where NB,0 is the amount of covalent bonds cleavable in the coal at time zero. Because NB can also be expressed as the difference between NB,0 and NB,C (the amount of covalent bonds cleaved at time t) as shown in Eq. (5), Eq. (4) can be rewritten as Eq. (6). The values of NB,0 and k are then determined by nonlinear regression based on the data in Fig. 6.
dNB =dt ¼ kN B
ð3Þ
NB ¼ NB;0 ½expðktÞ
ð4Þ
NB ¼ NB;0 NB;C
ð5Þ
NB;C ¼ NB;0 ½1 expðktÞ
ð6Þ
Fig. 9 shows the NB,0 of the 14 coals determined from the nonlinear regression with a correlation coefficient (R2) of greater than 0.89. It can be seen that NB,0 decreases with an increase in C% of the coals. The decrease is slow for coals with C% of lower than 90, from
486
M. Liu et al. / Fuel 182 (2016) 480–486
exposed to DHP due to breakup of the coal particles, and those formed from cleavage of covalent bonds. The radicals that acquire hydrogen from DHP are approximately 3 orders of magnitude higher than that of the stable radicals in the coals measured by ESR. All the coals contain some rigid structures that do not break up at 440 °C. The anthracites contain much more rigid structure than the lignites and bituminous coals. The relations of the quantities of stable radicals and bond cleaved with carbon content of the coals in the pyrolysis are consistent with the dependences of many physical properties of coals and agree with the understandings of the coal structure versus carbon content. The first order kinetics of covalent bond cleavage shows that the quantities of the cleavable bonds in the coals decrease with the increase of coal rank, and these bonds become hard to cleave with the increase in coal rank. Acknowledgments The authors gratefully acknowledge the financial support from the National Basic Research Program of China (2011CB201300) and the Natural Science Foundation of China (21276019). Fig. 9. The quantity of bonds cleavable at 440 °C (NB,0) in 14 coals.
Fig. 10. The rate constant k of bond cleavage during pyrolysis of 14 coals at 440 °C.
1.8 102 to 1.4 102 mol/g, but much fast for coals with C% of higher than 90, reaching to approximately 0.5 102 mol/g for TX coal (C% = 92). Fig. 10 shows the rate constant k of the 14 coals estimated from the nonlinear regression. It can be seen that the values of k decrease with an increase in C% approximately, which indicates that the covalent bonds become hard to cleave with the increase of coal rank. 4. Conclusion The quantities of active radicals generated from coal in pyrolysis can be determined by the amounts of hydrogen donated by DHP. At a DHP to coal mass ratio of 8 the radicals generated from the coal pyrolysis are primarily coupled with hydrogen donated by DHP. These radicals include those originally in the coals but
References [1] Solomon PR, Fletcher TH, Pugmire RJ. Progress in coal pyrolysis. Fuel 1993;7:587–97. [2] Arenillas A, Rubiera F, Pis JJ, Cuesta MJ, Iglesias MJ, Jimenez A, et al. Thermal behaviour during the pyrolysis of low rank perhydrous coals. J Anal Appl Pyrol 2003;68:371–85. [3] Zhao Y, Hu H, Jin L, He X, Wu B. Pyrolysis behavior of vitrinite and inertinite from Chinese Pingshuo coal by TG–MS and in a fixed bed reactor. Fuel Process Technol 2011;92:780–6. [4] Petrakis L, Grandy DW. Electron spin resonance spectrometric study of free radicals in coals. Anal Chem 1978;50:303–8. [5] Shi L, Liu Q, Guo X, Wu W, Liu Z. Pyrolysis behavior and bonding information of coal—a TGA study. Fuel Process Technol 2013;108:125–32. [6] Liu Z, Shi S, Li Y. Coal liquefaction technologies-development in China and challenges in chemical reaction engineering. Chem Eng Sci 2010;65:12–7. [7] He W, Liu Z, Liu Q, Ci D, Lievens C, Guo X. Behaviors of radical fragments in tar generated from pyrolysis of 4 coals. Fuel 2014;134:375–80. [8] Petrakis L, Grandy DW. Formation and behaviour of coal free radicals in pyrolysis and liquefaction conditions. Nature 1981;289:476–7. [9] Petrakis L, Grandy DW, Jones GL. Free radicals in coal and coal conversions. 7. An in-depth experimental investigation and statistical correlative model of the effects of residence time, temperature and solvents. Fuel 1982;61:21–8. [10] Seehra MS, Ghosh B. Free radicals, kinetics and phase changes in the pyrolysis of eight american coals. J Anal Appl Pyrol 1988;13:209–20. [11] Petrakis L, Jones GL, Grandy DW, King AB. Free radicals in coal and coal conversions. 10. Kinetics and reaction pathways in hydroliquefaction. Fuel 1983;62:681–9. [12] Schulten Z, Schulten K. The generation, diffusion, spin motion, and recombination of radical pairs in solution in the nanosecond time domain. J Chem Phys 1977;66:4616–34. [13] Fowler TG, Bartle KD. Limitations of electron spin resonance spectroscopy in assessing the role of free radicals in the thermal reactions of coal. Energy Fuels 1989;3:515–22. [14] Curran GP, Struck RT, Gorin E. Mechanism of the hydrogen-transfer process to coal and coal extract. I&EC Process Des Dev 1967;6:166–73. [15] Kamiya Y, Nagae S. Relative reactivity of hydrogen donor solvent in coal liquefaction. Fuel 1985;64:1242–5. [16] Liu M, Yang J, Liu Z, He W, Liu Q, Li Y, et al. Cleavage of covalent bonds in the pyrolysis of lignin, cellulose, and hemicellulose. Energy Fuels 2015;29:5773–80. [17] Petrakis L, Grandy DW. Free radicals in coals and coal conversion. 4. Investigation of the free radicals in selected macerals upon liquefaction. Fuel 1981;60:120–4. [18] Elliott MA. Chemistry of coal utilization. New York: John Wiley & Sons, Inc.; 1981. [19] Retcofsky HL, Thompson GP, Hough M, Friedel RA. Electron spin resonance studies of coals and coal-derived asphaltenes. In: Larsen JW, editor. Organic chemistry of coal. Washington DC: American chemical society; 1978. p. 142–55. [20] Retcofsky HL, Stark JM, Friedel RA. Electron spin resonance in american coals. Anal Chem 1968;40:1699–704. [21] Comolli AG, Lee TLK, Popper GA, Zhou P. The Shenhua coal direct liquefaction plant. Fuel Process Technol 1999;59:207–15. [22] Tschamler H, Ruiter E. Coal science. In: Given P, editor. Adv chem series 55. Washington DC: American Chemical Society; 1966. p. 332–43.