- Email: [email protected]
Study on the preheating stage of low rank coals liquefaction: Product distribution, chemical structural change of coal and hydrogen transfer
Fuel Processing Technology 159 (2017) 153–159
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Study on the preheating stage of low rank coals liquefaction: Product distribution, chemical structural change of coal and hydrogen transfer Pan Hao a,b, Zong-Qing Bai a,⁎, Zhi-Tong Zhao b,c, Jing-Chong Yan d, Xiao Li a,b, Zhen-Xing Guo a, Jun-Li Xu a,b, Jin Bai a, Wen Li a a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China University of Chinese Academy of Sciences, Beijing 100049, PR China State Key Laboratory of Catalysis, Dalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, PR China d Institute of Advanced Energy, Kyoto University Gokasho, Uji, Kyoto 611-0011, Japan b c
a r t i c l e
i n f o
Article history: Received 21 October 2016 Received in revised form 5 January 2017 Accepted 18 January 2017 Available online xxxx Keywords: Low rank coal Preheating stage Product distribution Hydrogen transfer
a b s t r a c t Preheating stage of direct liquefaction of Yunnan lignite (YN) and Hami sub-bituminous coal (HM) in tetralin at temperature range of 200–350 °C was investigated under nitrogen atmosphere. In order to reveal the product characteristics and hydrogen transfer between coal and solvent during preheating process, thermo-gravimetric analyzer coupled with mass spectrometer (TG–MS), X-ray photoelectron spectroscopy (XPS) and gas chromatography coupled with mass spectrometer (GC–MS) were employed. The results show that yields of light products, i.e., oil, gas, and water, increase with raising temperature during the preheating process. YN and HM achieve 51.69% and 44.19% (daf) light products yields at 350 °C, respectively. Moreover, oxygen-containing functional groups, such as carboxyl, ethers, and alcohols in raw coal are reduced after preheating. In addition, hydrogen transfer achieves a perceptible extent even at 200 °C and the amount of transferred hydrogen increases with raising temperature. A positive dependence of hydrogen transfer on conversion is observed during preheating stage. Comparing the two coal samples, YN obtains higher conversion and hydrogen transfer due to its higher thermal reactivity at this temperature range. However, HM achieves higher oil yield than YN does at 350 °C since high hydrogen transfer amount and H/C ratio of raw coal promote the oil yield. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The utilization of low rank coals is limited by the high content of oxygen and low calorific value, in spite of the abundant reserves in world [1]. However, due to the high reactivity and H/C ratio, low rank coals can efficiently convert into oil and other fine chemicals with the process of direct coal liquefaction [2–6]. The direct liquefaction processes have been developed to moderate the conditions and increase the liquid yield [7]. The commercialized direct liquefaction process in China (Shenhua, Inner Mongolia) is depicted as follows. The feedstock of liquefaction, coal-oil slurry prepared by mixing pulverized coal and hydrogen donor solvent homogeneously, is pumped to a preheater and liquefaction reactor successively [8,9]. The preheating stage is required due to it can achieve less energy consumption and more liquid products [10–12]. In this stage, coal-oil slurry undergoes several changes. As described in previous publications [13,14], extraction of coal has taken place when slurry is pumped to the preheater, then extract yield increases with increasing temperature. Meanwhile, a number of labile ⁎ Corresponding author. E-mail address: [email protected] (Z.-Q. Bai).
http://dx.doi.org/10.1016/j.fuproc.2017.01.028 0378-3820/© 2017 Elsevier B.V. All rights reserved.
functional groups in coal are broken [15,16]. The broken fragments of coal can be converted into preasphaltene, asphaltene, and oil during the preheating stage [9]. Subsequently, the unreacted coal converts into light products at higher temperature [17]. In our previous work, the chemical structure and pyrolysis reactivity of coal in non-polar solvent tetralin have been preliminarily studied [15]. It shows that tetralin treatment was effective in dewatering and upgrading of low rank coal. During this process, the structure of unreacted coal was changing with deoxygenation and the thermal reactivity became lower. The change of coal structure would have impact on the subsequent reactions [18, 19]. However, the hydrogenation effect was not considered due to the neglect of light liquid product. Study on chemical characteristics of unreacted coal in the preheating stage is insufficient. In addition, the effect of chemical structural change on liquefaction conversion is unclear. Hydrogen transfer from solvent or gas phase H2 to coal-derived fragments is a necessary step for coal liquefaction. Previous publications [20–22] have reported that hydrogen transfer reactions occur at the preheating stage, and the hydrogen transfer amount from solvent or gas phase H2 to coal is affected by temperature. Meanwhile, light products are formed under the hydrogen transfer effect. Researchers have demonstrated that hydrogen transfer amount is correlated with oil
154
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
yield and liquefaction conversion in liquefaction reaction [23–25]. However, their relation in the preheating stage, which would reveal the hydrogenation level of coal and offer reference to determine the preheating condition, has been little focused. Besides, the hydrogen transfer amount is related to coal rank. Low rank coals achieve higher hydrogen transfer amount than that of high rank coals at same temperature. In addition, more radicals are generated through bond cleavage in low rank coals at same condition [26], which suggests that the thermally induced fragments promote the hydrogen transfer amount. As temperature impacts on both of fragment formation rate and hydrogen transfer rate, the selection of temperature is important for different ranks of coal preheating process. The objective of this work is to explore the product distribution and hydrogen transfer of low rank coals at different preheating temperature during liquefaction. The experimental operating at preheating stage was simulated in a batch autoclave reactor with tetralin as solvent at 200–350 °C. Two Chinese low-rank coals, one lignite and one sub-bituminous coal, were used in this work. And the characteristics of products after preheating were examined to reveal the change of coal and tetralin at various temperatures. The experiments were conducted with N2 as working atmosphere in order to rule out the effect of hydrogen transferring from gas phase to coal samples and focus on the hydrogen donating role of solvent.
sieve column with FID detector were used to analyze C1–C4 hydrocarbons and CO, CO2, respectively. The solid-liquid mixture was totally transferred to a Soxhlet apparatus and extracted with n-hexane for 24 h. The extraction included tetralin and its derivatives, and some light products from coal conversion. And these light products were named as oil in this work. Then n-hexane was removed from extraction by vacuum rotary evaporation, and the remaining tertralin and its derivatives and oil were analyzed without further separation. The n-hexane insoluble solid was dried in a vacuum oven at 80 °C for 12 h, which was weighted to calculate the solid yield by using Eq. (1). Each run was repeated at least 3 times to ensure the duplicability, and the deviations were b2%. The solid samples obtained at 200–350 °C were named as YN-200 or HM-200; YN-250 or HM-250; etc. respectively. The solid samples were further extracted with tetrahydrofuran (THF) for 24 h to separate preasphaltene and asphaltene (PAA). PAA yield was calculated by using Eq. (2). The THF insoluble fraction was named as residue, and the conversion of coal during preheating process was calculated by using Eq. (3).
2. Experimental section
c ¼ 100 ‐
ws ¼
m1 100 mc
ð1Þ
wp ¼
m1 ‐ m2 100 mc
ð2Þ
m2 100 mc
ð3Þ
2.1. Materials Two low rank coals, Yunnan lignite (YN) and Hami sub-bituminous coal (HM), were used in this work. The coal samples were grounded and sieved to b 150 μm, dried in a vacuum oven at 80 °C for 24 h before using. The proximate and ultimate analyses of the two coals are listed in Table 1. The solvents including tetralin (purity ≥ 99.0%), n-hexane (purity ≥ 98.0%), diphenyl (purity ≥ 99.5%), tetrahydrofuran (purity ≥ 99.0%) were commercially pure chemical reagents and used without further purification. 2.2. Preheating treatment In each run, 5.00 g coal (dry basis) and 10.00 g tetralin were packed into 100 mL autoclave (Parr 4598, USA), which was then charged with N2 to 3 MPa (cold pressure) after leak-checking. The reactor was heated from ambient temperature to the desired temperature (200–350 °C) at 5 °C/min with stirring rate of 400 rpm simultaneously. It was kept at the final temperature for 120 min before stopping heating. The gaseous products were collected in a gas bag and analyzed by a gas chromatograph (GC-950, Haixin, China). A 5A sieve column with TCD detector was used to detect H2, N2, CH4, and CO. An Al2O3 column and carbon Table 1 Proximate and ultimate analyses of solid samples obtained at different temperatures. Sample
HM HM-200 HM-250 HM-300 HM-350 YN YN-200 YN-250 YN-300 YN-350
Proximate analysis (wt.%)
Ultimate analysis (wt.%)
Atomic ratio
Mad
Ad
Vdaf
Cdaf
Hdaf
Odafa
Ndaf
Sdb
H/C
O/C
6.94 4.36 3.55 2.77 1.55 9.29 5.24 4.36 2.94 2.32
6.76 6.80 6.95 7.49 10.23 10.27 10.73 11.85 12.94 17.33
54.19 53.44 52.47 52.03 48.78 51.46 49.85 46.42 45.34 42.93
74.35 76.19 77.42 78.78 81.34 68.35 71.30 75.09 77.60 81.47
5.61 5.71 5.77 5.87 5.89 4.73 4.84 5.04 5.17 5.47
18.68 16.76 15.47 13.87 11.00 24.83 21.72 17.60 14.89 10.33
1.07 1.09 1.11 1.21 1.46 1.28 1.36 1.48 1.57 1.92
0.27 0.23 0.21 0.25 0.28 0.73 0.71 0.69 0.67 0.66
0.91 0.90 0.89 0.89 0.87 0.83 0.81 0.81 0.80 0.81
0.19 0.16 0.15 0.13 0.10 0.27 0.23 0.18 0.14 0.10
ad: air dry basis; d: dry basis; daf: dry ash free basis. a By difference. b Total S.
where, ws is the solid yield (wt.%, daf); wp is PAA yield (wt.%, daf); c is the conversion (wt.%, daf); m1 is the weight of n-hexane insoluble solid fraction, g (daf); m2 is weight of tetrahydrofuran (THF) insoluble fraction, g (daf); mc is the weight of feed coal, g (daf). Yield of water produced in the preheating process was determined by the oxygen mass balance in all products, as following Eq. (4): ww ¼ 18
Oc‐Os‐Og‐Oo 100 16 mc
ð4Þ
where, ww is the water yield (wt.%, daf); Oc, Os, Og, and Oo represent the content of oxygen in raw coal, solid, gas, and oil, respectively. The oxygen content in oil and solid were determined by using an elemental analyzer instead of by difference to obtain more accurate data. Due to the relatively high boiling points of tetralin and naphthalene, some light oil would be also evaporated during the evaporation of solvent. The weight of residual oil would be lower than the actual value. Therefore, oil yield was obtained by difference as shown in Eq. (5): wo ¼ 100‐ws ‐ww ‐wg
ð5Þ
where, wo and wg are the oil yield and gas yield (wt.%, daf), respectively. 2.3. Characterization Thermo-gravimetric analyzer (TG, Setsys Evolution, SETARAM, France) coupled with a mass spectrometer (MS, Omnistar, Pfeiffer Vacuum) was used to study the pyrolysis behavior of raw coal. About 11 mg sample was placed in a corundum crucible and heated from 30 to 1050 °C at 10 °C/min in 100 mL/min argon atmosphere. The chemical component of the solvent-oil mixture was identified and quantified by gas chromatography-mass spectrometer (GC–MS, ISQ, Thermo Scientific, USA). The samples were separated using a 30 m × 0.25 mm TR-5MS capillary column (Thermo Scientific, USA). Helium was employed as the carrier gas. The oven temperature was raised from 40 °C to 250 °C at 5 °C/min. The components were identified based on the attached library. Diphenyl was selected as internal standard to quantify tetralin and naphthalene. The contents of carboxyl group in raw coal were determined by chemical method [27]. The aqueous determinations of acidity were
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
155
Fig. 1. TG and DTG curves of HM and YN.
performed by ion exchanging 250 mg coal with 60 mL of BaCl2/ triethanolamine/HCl buffer under vacuum for 180 min. The mixture was then filtered and the filter cake was rinsed with distilled water. Afterwards, the filter cake was stirred with 10 mL, 0.1 mol/L HCl under vacuum. Then the filtrate was taken and titrated against 0.05 mol/L NaOH. The endpoint was taken as the volume of NaOH used to attain pH 5. X-ray photoelectron spectroscopy (XPS) of raw coal and solid samples were obtained with AXIS ULTRA DLD (Kratos, UK) equipped with a monochromatic Al Kα X-ray source (1486.6 eV). High resolution scanning of the C1s region was performed in a step of 0.05 eV with pass energy of 20 eV. Peak deconvolution of C1s spectra was accomplished using the XPSPEAK41 software. 2.4. Determination of hydrogen transfer (HT) from tetralin to coal The amount of hydrogen transferred from tetralin to coal was calculated by using Eq. (6) [28]. HT ðmg=g; daf Þ ¼
4 1:008 mn 1000 128:17 mc
ð6Þ
where, mn is the corrected amount of naphthalene in the recovered solvent constituent determined by GC–MS analysis. Factors of 1.008 and 128.17 are the weight of hydrogen atom and the molecular weight of naphthalene.
and oxygen in YN. However, a reverse result is observed at higher temperature. The final weight loss of YN and HM are 48% and 44% respectively when temperature rises to 1050 °C. The higher weight loss of HM is attributed to its higher volatile matter and lower ash content. The DTG curves can be divided into three stages according to the decomposition of functional groups and condensation of coal structure. At first, the moisture and adsorbed gases are removed when temperature rises from 30 to 200 °C. The second pyrolysis stage is from 200 to 600 °C, at which large amount of volatile matter are produced and extensive polymerization reactions take place. With the temperature rising to above 600 °C, the secondary condensation reactions occur, as well as the decomposition of some mineral matter such as pyrite and CaCO3. From DTG curves, the weight loss rate of YN is higher than that of HM at first stage, which is attributed to the adsorption of water and gas. The disparity of weight loss rate becomes more obvious with temperature rising to the second stage. Moreover, the maximum mass loss rate of the two coal samples, which represents the volatile release level, suggests the higher thermal reactivity of HM at high temperature. The peak temperature of YN at the maximum mass loss rate is about 397 °C, while it shifts to a higher zone (436 °C) for HM. It is related to the interactions between mobile phase and fixed phase in coal and thermal conductivity of coal [29], indicating cross-linking level becomes enhanced as the rank of coal rises. For the two samples, the thermolysis rate of YN is more rapid initially due to the higher heteroatoms in YN sample. With the temperature increasing higher, more fragmental structure in HM breaks up and a large amount of volatile matter is released. As a result, the final weight loss of HM is higher at 1050 °C.
3. Results and discussion 3.2. Product distribution during preheating stage 3.1. Thermo-gravimetric analysis (TGA) of raw coal In order to elucidate the thermal behavior of raw coal and understand the effect of preheating temperature on coal conversion, the TG and DTG (differential thermal gravity) curves of raw coal have been obtained, as shown in Fig. 1. TG curves depict that YN loses more weight than HM from 30 to 442 °C, attributing to the higher content of moisture
Fig. 2 presents the product distribution at different preheating temperatures of the two coals. Solid here is n-hexane insoluble fraction, which is the residual solid after separating oil and water. The solid yields of the two coals show descending trends with the increase of preheating temperature. The solid loss after preheating treatment is attributed to the decomposition of labile functional groups in low rank coal. In
Fig. 2. Product distribution after preheating treatment. a: YN; b: HM.
156
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
Fig. 3. Gas compositions and yield evolved in the preheating process. a: YN; b: HM.
comparison with HM, YN obtains lower solid yield at same preheating temperature. For example, the yield of YN-300 is 71.94%, lower than 83.10% of HM-300. This is due to the higher thermal reactivity of YN at this temperature, as shown in Fig. 1. The solid yields of HM and YN decrease steadily from 92.98% and 89.55% to 83.10% and 71.94%, respectively, at temperature range of 200 to 300 °C. And with temperature further reaching 350 °C, the solid yields of HM and YN decrease to 55.81% and 48.31%, respectively. The descending trend from 200 to 300 °C is much lower than that from 300 to 350 °C, which is related to the thermolysis rates of raw coal at the temperature range. Fig. 1 suggests that thermolysis rates of the two coals dramatically increase above 300 °C. The fragments formed at high temperature are stabilized by hydrogen from tetralin and convert into light products. Accordingly, the yields of oil, water, and gas represent an increase trend with the raising temperature. Due to the rapid thermolysis rates above 300 °C, the oil yields of the two coals from 300 to 350 °C increase more than twice that below 300 °C. The content of water was calculated by oxygen mass balance during the reaction system. Water yield shows a noticeable amount during the preheating stage. Water yields of YN and HM reach about 15.72% and 9.68% respectively at 350 °C. Source of water differs with preheating temperature rising. Initially, water comes from adsorbed water in pores of coal, which couldn't be removed in drying process. When the temperature increases further, oxygen-containing functional groups in coal, such as ketone and hydroxyl structure, can convert into water by adding hydrogen from coal and solvent [30–32], which increase the consumption of hydrogen. More water yields are observed for YN because of its higher oxygen content during the preheating stage. 3.3. Gas analysis during preheating stage
are n-hexane soluble fraction by separating oil and water, which would be converted into light products subsequently. n-Hexane did not affect the solid structure due to its chain alkane composition. The TG-MS analyses of the solid samples show that the intensity of tetralin was extremely weak, indicating that the residual solvent in solid samples could be ignored. The analyses of solid characteristics would reveal the variation of coal chemical structure in the preheating stage accurately and help to understand its subsequent thermal conversion behavior. 3.4.1. Proximate and ultimate analyses of solid samples Table 1 shows the proximate and ultimate analyses of raw coal and solid samples. Obviously, hydrogen content of solid samples increases with the rising temperature, which is different from dehydrogenation of coal during pyrolysis without hydrogen donor [33]. In addition, a fraction of raw coal can be converted into oil during the preheating process. These results show that hydrogen from tetralin transfers to solid and oil effectively during the preheating process. As a result, H/C values of solid samples are still at a high level with the raising temperature. A large amount of oxygen in raw coal transfers into water, CO2, and CO under thermal decomposition. As a result, oxygen content in solid samples decreases obviously with the increasing temperature. Hydrogenation and deoxygenation occur during the preheating process for the solid samples, and it is beneficial to its subsequent liquefaction [7]. 3.4.2. Carbon functionalities of solid samples analyzed by XPS In order to investigate the carbon functionalities in raw coal and solid samples, the XPS C1s spectra were recorded. The C1s spectra can be deconvoluted to show four types of carbon [34–36]: the 284.8 eV peak represents contributions from both aromatic and aliphatic carbon (C\\C, C _C, and C\\H); the 286.3 eV peak represents hydroxyl/phenol
The composition of gas evolved during the preheating treatment is shown in Fig. 3. The gas yields of the two coals increase with rising temperature, and the dominant gas is CO2, which is evolved by de-carboxylation reaction. The content of carboxyl in YN and HM determined by chemical methods is 0.95 and 0.65 mmol/g, respectively. Thus, CO2 yield from preheating treatment of YN is higher than that of HM. H2 is generated when temperature is above 250 °C. However, its amount is extremely low, indicating the condensation reaction of coal with the presence of tetralin is negligible. In comparison with YN, the aliphatic hydrocarbons amount of HM is higher at 350 °C, which is consistent with H/C ratio of the two coals. 3.4. Effect of preheating treatment on characteristics of solid samples With the process to convert into oil, water, and gas, coal undergoes the breakup of macro-structure and hydrogen transfer reaction. In order to reveal the effect of tetralin on coal hydrogenation and variation of carbon functionalities in the preheating process, characteristics of solid obtained at different temperatures are studied. Here, solid samples
Fig. 4. C1s spectra of XPS for YN.
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159 Table 2 Carbon functionalities in raw coal and solid samples from XPS analysis. Sample
Relative proportion (area percent, %) C\ \C, C _C, C\ \H
C\ \O
C_O, O\ \C\ \O
O\ \C_O
YN YN-200 YN-250 YN-300 YN-350 HM HM-200 HM-250 HM-300 HM-350
74.06 76.12 80.26 80.55 84.81 79.38 81.03 82.36 84.60 83.37
18.83 18.44 16.31 16.91 11.94 13.42 12.86 13.22 11.74 11.99
4.03 3.06 1.34 1.21 0.88 3.39 3.00 2.22 2.16 0.84
3.08 2.38 2.09 1.33 2.37 3.81 3.70 2.20 1.50 3.80
groups or ether (C\\OR); the 287.5 eV peak corresponds to carbonyl carbon (C _O and O\\C\\O); the 289.0 eV peak corresponds mainly to carboxylic group, ether or lactone (O\\C_ O). The relative full width at half maximum values of all peaks were fixed to 1.5 (±0.1) eV. Example of XPS C1s spectra for YN into different organic carbon components is shown in Fig. 4. The XPS C1s spectra of other samples are shown in the supplementary data. The relative proportion (RP) for each functional group from C1s deconvolution of all samples is listed in Table 2. Aliphatic and aromatic carbon are the main groups with RP of 74.40% and 79.38% for YN and HM respectively, and the RPs of aliphatic and aromatic carbon groups increase with raising preheating temperature due to cleavage of side-chains. The C\\O bonds, including hydroxyl, phenol groups and ether, show a slight change during the process. Yan et al. have found ethers and alcohols start decomposing at 150 and 200 °C [15], and phenolic groups remain stable below 350 °C. The result indicates most of C\\O bonds are phenolic groups in the two coals, especially for HM. The RPs of carbonyl and carboxyl decrease obviously with the rising preheating temperature. The\\COO peak positions of YN-350 and HM-350 shift to 289.4 eV, which are attributed to the signal of carbonate. The increasing amount of\\COO in these two samples indicates that carbonate is formed at 350 °C and enriched in the solid samples. Perry et al. have found that the amount of carbonate formed in the liquefaction residue was mainly proportional to the amount of ion-exchangeable calcium in feed coal [37]. This suggests that the ionexchangeable calcium in feed coal converts into stable carbonate under certain conditions. The formed carbonate may cause operational difficulty in liquefaction reactor, which should be prevented in liquefaction. 3.5. Hydrogen transfer effect during preheating stage It has been reported that tetralin is stable below 400 °C [24,38], and our experimental result shows that the tetralin mass remained 98.5% at 350 °C in the absence of coal. However, dehydrogenated products were detected under the same condition in the presence of coal. In this work,
157
naphthalene was the main tetralin dehydrogenated product, and dihydronaphthalene was not detected in the GC–MS results. The variation of HT with different preheating temperatures of HM and YN are shown in Fig. 5a, and the corresponding transfer ratios to maximum HT for the two coals are shown in Fig. 5b. The maximum HT was calculated by dehydrogenation of tetralin. It is assumed that all tetralin would convert into naphthalene, and the maximum HT would be 64.89 mg/g and 67.54 mg/g for HM and YN, respectively. As additional information, the overall tetralin mass balance for the reaction reached level of 91–99 wt.%. Fig. 5 shows that hydrogen transfer from tetralin achieves a perceptible extent at 200 °C, and increases significantly above 300 °C. HT reaches 14.74 mg/g and 15.77 mg/g at 350 °C for HM and YN, accounting for 22.72% and 26.04% of the maximum HT, respectively. Accordingly, coal promotes dehydrogenation of tetralin during the preheating process. Temperature is one of the most important factors for hydrogen transfer from tetralin to coal [25]. Attributing to the large amount of fragmental radicals produced at higher temperature, hydrogen transfer increases rapidly from 300 to 350 °C. It has been reported that the quantity of coal-derived radicals during pyrolysis process increases with the decrease of carbon content in coal or rank of coal [26,39–41]. This is because the higher amount of heteroatoms or labile functional groups and lower carbon in low rank coal result in the higher thermal reactivity at same temperature. The unstable coal-derived radicals capture hydrogen from hydrogen donor solvent. Thus the HT amount of YN is higher than that of HM at same temperature in the preheating stage. In traditional liquefaction process, the conversion of coal is defined as the percentage of THF soluble fractions. By reference to the method, conversion in the preheating stage was also investigated and its correlation with hydrogen transfer amount is shown in Fig. 6. The plot of conversion vs. HT displays good linear fit with R-squared value of 0.95. Conversion increases monotonically as a function of hydrogen transfer amount from tetralin to coal. However, it has been reported that greater amounts of hydrogen transfer do not further enhance conversion significantly when HT reaches N 30 mg/g, due to the participation in hydrocracking of heavier products [25]. A linear relation is also observed between oil and hydrogen transfer from Fig. 6. From the product distribution in Fig. 2, the oil yields show little difference between the two coals. Moreover, higher oil yield and lower HT are achieved for HM at 350 °C because of the difference between H/C ratios of HM and YN, which is an important factor for coal selection in direct liquefaction. Relatively high H/C ratio of coal can reduce hydrogen consumption and is beneficial for production of oil. HM has a higher H/C ratio, thus it achieves higher oil yields at 350 °C with a lower consumption of hydrogen. The object of liquefaction is to achieve good quality and high yield of oil by increasing conversion. Therefore, via selecting the suit feedstock, low HT and high oil yield can be obtained at low liquefaction temperature. PAA is THF soluble fraction except oil and water. The result in Fig. 6 shows that the PAA amount increases with the increase of HT. According to reference [42] about the pathways of liquefaction, PAA, the
Fig. 5. Hydrogen transfer at different preheating temperatures.
158
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
Fig. 6. Effect of HT on conversion and some products yields.
intermediate of coal to oil and gas, are produced in parallel as the conversion increases during the initial coal dissolution stage. Therefore, the hydrogen donor promotes the conversion of coal to light products. 4. Conclusions Preheating treatment of two low rank coals between 200 and 350 °C was performed with tetralin as hydrogen donor solvent. During the preheating process, total yields of light products (oil, gas and water) increase with raising temperature, and they achieve the highest level of 44.19% and 51.69% at 350 °C for HM and YN, respectively. The presence of tetralin could effectively increase the hydrogen content of solid samples during the preheating stage, which is beneficial for subsequent conversion. Moreover, the labile oxygen-containing functional groups in raw coal, such as carboxyl groups, ethers, and alcohols decompose gradually during the preheating stage. Hydrogen transfer from tetralin to coal occurs even at the temperature of 200 °C, and it is enhanced by the large amount of fragmental radicals formed at higher temperature. In addition, the amount of hydrogen transfer for YN is higher than that of HM below 350 °C because of their thermal reactivity difference. Both conversion and oil yield are related to the hydrogen transfer amount at 200–350 °C. A positive correlation is observed between the hydrogen transfer and oil yield, and high H/C ratio of raw coal also promotes oil yield. Therefore, increasing hydrogen transfer amount and selecting feed coal with high H/C are effective for improving the oil yield and conversion during liquefaction. Acknowledgments This work was supported by the National Natural Science Foundation of China (21576274), and Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA07060100). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.fuproc.2017.01.028. References [1] H. Osman, S.V. Jangam, J.D. Lease, A.S. Mujumdar, Drying of Low-Rank Coal (LRC)—a review of recent patents and innovations, Drying Technol. 29 (2011) 1763–1783. [2] P.R. Solomon, M.A. Serio, G.V. Despande, E. Kroo, Cross-linking reactions during coal conversion, Energy Fuel 4 (1990) 42–54. [3] S. Vasireddy, B. Morreale, A. Cugini, C. Song, J.J. Spivey, Clean liquid fuels from direct coal liquefaction: chemistry, catalysis, technological status and challenges, Energ. Environ. Sci. 4 (2011) 311–345. [4] I. Mochida, Y. Moriguchi, Y. Korai, H. Fujitsu, K. Takeshita, Efficient liquefaction of Australian brown coal with a hydrogen-donating solvent under atmospheric pressure, Fuel 60 (1981) 746–747.
[5] I. Mochida, K. Iwamoto, T. Tahara, Y. Korai, H. Fujitsu, K. Takeshita, Liquefaction of subbituminous coals under apparently non-hydrogenative conditions, Fuel 61 (1982) 603–609. [6] Z. Liu, S. Shi, Y. Li, Coal liquefaction technologies-development in China and challenges in chemical reaction engineering, Chem. Eng. Sci. 65 (2010) 12–17. [7] I. Mochida, O. Okuma, S.H. Yoon, Chemicals from direct coal liquefaction, Chem. Rev. 114 (2014) 1637–1672. [8] A. Kidoguchi, H. Itoh, M. Hiraide, E. Kaneda, H. Ishibashi, M. Kobayashi, K. Ikeda, K. Imada, K. Inokuchi, Simulation of initial stage reactions in the direct coal liquefaction of sub-bituminous coals, Fuel 80 (2001) 1325–1331. [9] X. Shan, K. Li, X. Zhang, H. Jiang, H. Weng, Reaction kinetics study on the heating stage of the Shenhua direct coal liquefaction process, Energy Fuel 29 (2015) 2244–2249. [10] L. Artok, A. Davis, G.D. Mitchell, H.H. Schobert, Swelling pretreatment of coals for improved catalytic liquefaction, Fuel 71 (1992) 981–991. [11] H. Hu, G. Sha, G. Chen, Effect of solvent swelling on liquefaction of Xinglong coal at less severe conditions, Fuel Process. Technol. 68 (2000) 33–43. [12] Y. Inukai, S. Arita, H. Hirosue, Effect of preheat treatment on coal-liquefaction, Energy Fuel 9 (1995) 67–70. [13] F. Haghighat, A. de Klerk, Direct coal liquefaction: low temperature dissolution process, Energy Fuel 28 (2014) 1012–1019. [14] H. Shui, J. Liu, Z. Wang, M. Cao, X. Wei, Effect of pre-swelling of coal at mild temperatures on its hydro-liquefaction properties, Fuel Process. Technol. 90 (2009) 1047–1051. [15] J. Yan, Z. Bai, J. Bai, Z. Guo, W. Li, Effects of organic solvent treatment on the chemical structure and pyrolysis reactivity of brown coal, Fuel 128 (2014) 39–45. [16] F. Liu, X. Wei, M. Fan, Z. Zong, Separation and structural characterization of the value-added chemicals from mild degradation of lignites: a review, Appl. Energy 170 (2016) 415–436. [17] L. Huang, H.H. Schobert, Comparison of temperature conditions in direct liquefaction of selected low-rank coals, Energy Fuel 19 (2005) 200–207. [18] A. Tsutsumi, K. Yoshida, Rheological behaviour of coal-solvent slurries, Fuel 65 (1986) 906–909. [19] F.E. Ndaji, I.M. Butterfield, K.M. Thomas, Changes in the macromolecular structure of coals with pyrolysis temperature, Fuel 76 (1997) 169–177. [20] A. Ishihara, D. Nishigori, M. Saito, I.P. Sturisna, W. Qian, T. Kabe, Elucidation of hydrogen mobility in functional groups of coals using tritium tracer methods, Energy Fuel 16 (2002) 32–39. [21] A. Ishihara, H. Takaoka, E. Nakajima, Y. Imai, T. Kabe, Estimation of hydrogen mobility in coal using a tritium trace method. Hydrogen extrange reactions of coals with tritiated-water and molecular hydrogen, Energy Fuel 7 (1993) 362–366. [22] A. Ishihara, I.P. Sutrisna, T. Miura, M. Saito, E.W. Qian, T. Kabe, Elucidation of hydrogen transfer between coal and tritiated organic solvent, Energy Fuel 16 (2002) 1490–1498. [23] K. Ceylan, A. Olcay, Dissolution of two Turkish lignites in tetralin under hydrogen or nitrogen: effects of the extraction parameters on the conversion, Fuel 71 (1992) 539–544. [24] J.J. Devlieger, A.P.G. Kieboom, H. Vanbekkum, Behavior of tetralin in coal-liquefaction - examination in long-run batch-autoclave experiments, Fuel 63 (1984) 334–340. [25] L. Artok, H.H. Schobert, O. Erbatur, Temperature-staged liquefaction of selected Turkish coals, Fuel Process. Technol. 37 (1994) 211–236. [26] M. Liu, J. Yang, Y. Yang, Z. Liu, L. Shi, W. He, Q. Liu, The radical and bond cleavage behaviors of 14 coals during pyrolysis with 9,10-dihydrophenanthrene, Fuel 182 (2016) 480–486. [27] D.J. Allardice, L.M. Clemow, W.R. Jackson, Determination of the acid distribution and total acidity of low-rank coals and coal-derived materials by an improved barium exchange technique, Fuel 82 (2003) 35–40. [28] M. Sugano, S. Ohura, R. Endoh, K. Hirano, K. Mashimo, Effects of hydrogen transfer by exchanged cobalt upon liquefaction of low rank coal, Fuel 101 (2012) 228–233. [29] L. Zou, L. Jin, Y. Li, S. Zhu, H. Hu, Effect of tetrahydrofuran extraction on lignite pyrolysis under nitrogen, J. Anal. Appl. Pyrolysis 112 (2015) 113–120. [30] R. Dabestani, P.F. Britt, A.C. Buchanan, Pyrolysis of aromatic carboxylic acid salts: does decarboxylation play a role in cross-linking reactions? Energy Fuel 19 (2005) 365–373. [31] K. Miura, K. Mae, K. Sakurada, K. Hashimoto, Flash pyrolysis of coal following thermal pretreatment at low temperature, Energy Fuel 6 (1992) 16–21. [32] D.C. Cronauer, D.M. Jewell, Y.T. Shah, R.J. Modi, Mechanism and kinetics of selected hydrogen transfer reactions typical of coal liquefaction, Ind. Eng. Chem. Fundam. 18 (1979) 153–162. [33] J. Zhang, Study on the Oxygen Transfer Laws during coal's Pyrolysis Process and Its Effect on the Coking Property, Anhui university of technology, China, 2013. [34] S.R. Kelemen, M. Afeworki, M.L. Gorbaty, M. Sansone, P.J. Kwiatek, C.C. Walters, H. Freund, M. Siskin, A.E. Bence, D.J. Curry, M. Solum, R.J. Pugmire, M. Vandenbroucke, M. Leblond, F. Behar, Direct characterization of kerogen by X-ray and solid-state C-13 nuclear magnetic resonance methods, Energy Fuel 21 (2007) 1548–1561. [35] S.R. Kelemen, K.D. Rose, P.J. Kwiatek, Carbon aromaticity based on XPS II to II∗ signal intensity, Appl. Surf. Sci. 64 (1993) 167–174. [36] Z. Han, X. Zeng, C. Yao, G. Xu, Oxygen migration in torrefaction of Eupatorium adenophorum Spreng and its improvement on fuel properties, Energy Fuel 29 (2015) 7275–7283. [37] G.J. Perry, A. Gray, G.H. Mackay, Carbonate formation during hydrogenation of Victorian brown coal, Fuel Process. Technol. 10 (1985) 285–297. [38] M.C. Tsai, S.W. Weller, Ctalysis of hydrogen transfer in a tetralin-coal system, Fuel Process. Technol. 2 (1979) 313–316.
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159 [39] A. Ishihara, I.P. Sutrisna, M. Ifuku, E.W. Qian, T. Kabe, Elucidation of hydrogen mobility in coal using a fixed bed flow reactor hydrogen transfer reaction between tritiated hydrogen, coal, and tetralin, Energy Fuel 16 (2002) 1483–1489. [40] L. Petrakis, D.W. Grandy, Electron spin resonance spectrometric study of free radicals in coals, Anal. Chem. 50 (1978) 303–308.
159
[41] L. Petrakis, D.W. Grandy, Free radicals in coals and coal conversion. 3. Investigation of the free radicals of selected macerals upon pyrolysis, Fuel 60 (1981) 115–119. [42] R.A. Keogh, B.H. Davis, Comparison of liquefaction pathways of a bituminous and subbituminous coal, Energy Fuel 8 (1994) 289–293.
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Study on the preheating stage of low rank coals liquefaction: Product distribution, chemical structural change of coal and hydrogen transfer Pan Hao a,b, Zong-Qing Bai a,⁎, Zhi-Tong Zhao b,c, Jing-Chong Yan d, Xiao Li a,b, Zhen-Xing Guo a, Jun-Li Xu a,b, Jin Bai a, Wen Li a a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China University of Chinese Academy of Sciences, Beijing 100049, PR China State Key Laboratory of Catalysis, Dalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, PR China d Institute of Advanced Energy, Kyoto University Gokasho, Uji, Kyoto 611-0011, Japan b c
a r t i c l e
i n f o
Article history: Received 21 October 2016 Received in revised form 5 January 2017 Accepted 18 January 2017 Available online xxxx Keywords: Low rank coal Preheating stage Product distribution Hydrogen transfer
a b s t r a c t Preheating stage of direct liquefaction of Yunnan lignite (YN) and Hami sub-bituminous coal (HM) in tetralin at temperature range of 200–350 °C was investigated under nitrogen atmosphere. In order to reveal the product characteristics and hydrogen transfer between coal and solvent during preheating process, thermo-gravimetric analyzer coupled with mass spectrometer (TG–MS), X-ray photoelectron spectroscopy (XPS) and gas chromatography coupled with mass spectrometer (GC–MS) were employed. The results show that yields of light products, i.e., oil, gas, and water, increase with raising temperature during the preheating process. YN and HM achieve 51.69% and 44.19% (daf) light products yields at 350 °C, respectively. Moreover, oxygen-containing functional groups, such as carboxyl, ethers, and alcohols in raw coal are reduced after preheating. In addition, hydrogen transfer achieves a perceptible extent even at 200 °C and the amount of transferred hydrogen increases with raising temperature. A positive dependence of hydrogen transfer on conversion is observed during preheating stage. Comparing the two coal samples, YN obtains higher conversion and hydrogen transfer due to its higher thermal reactivity at this temperature range. However, HM achieves higher oil yield than YN does at 350 °C since high hydrogen transfer amount and H/C ratio of raw coal promote the oil yield. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The utilization of low rank coals is limited by the high content of oxygen and low calorific value, in spite of the abundant reserves in world [1]. However, due to the high reactivity and H/C ratio, low rank coals can efficiently convert into oil and other fine chemicals with the process of direct coal liquefaction [2–6]. The direct liquefaction processes have been developed to moderate the conditions and increase the liquid yield [7]. The commercialized direct liquefaction process in China (Shenhua, Inner Mongolia) is depicted as follows. The feedstock of liquefaction, coal-oil slurry prepared by mixing pulverized coal and hydrogen donor solvent homogeneously, is pumped to a preheater and liquefaction reactor successively [8,9]. The preheating stage is required due to it can achieve less energy consumption and more liquid products [10–12]. In this stage, coal-oil slurry undergoes several changes. As described in previous publications [13,14], extraction of coal has taken place when slurry is pumped to the preheater, then extract yield increases with increasing temperature. Meanwhile, a number of labile ⁎ Corresponding author. E-mail address: [email protected] (Z.-Q. Bai).
http://dx.doi.org/10.1016/j.fuproc.2017.01.028 0378-3820/© 2017 Elsevier B.V. All rights reserved.
functional groups in coal are broken [15,16]. The broken fragments of coal can be converted into preasphaltene, asphaltene, and oil during the preheating stage [9]. Subsequently, the unreacted coal converts into light products at higher temperature [17]. In our previous work, the chemical structure and pyrolysis reactivity of coal in non-polar solvent tetralin have been preliminarily studied [15]. It shows that tetralin treatment was effective in dewatering and upgrading of low rank coal. During this process, the structure of unreacted coal was changing with deoxygenation and the thermal reactivity became lower. The change of coal structure would have impact on the subsequent reactions [18, 19]. However, the hydrogenation effect was not considered due to the neglect of light liquid product. Study on chemical characteristics of unreacted coal in the preheating stage is insufficient. In addition, the effect of chemical structural change on liquefaction conversion is unclear. Hydrogen transfer from solvent or gas phase H2 to coal-derived fragments is a necessary step for coal liquefaction. Previous publications [20–22] have reported that hydrogen transfer reactions occur at the preheating stage, and the hydrogen transfer amount from solvent or gas phase H2 to coal is affected by temperature. Meanwhile, light products are formed under the hydrogen transfer effect. Researchers have demonstrated that hydrogen transfer amount is correlated with oil
154
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
yield and liquefaction conversion in liquefaction reaction [23–25]. However, their relation in the preheating stage, which would reveal the hydrogenation level of coal and offer reference to determine the preheating condition, has been little focused. Besides, the hydrogen transfer amount is related to coal rank. Low rank coals achieve higher hydrogen transfer amount than that of high rank coals at same temperature. In addition, more radicals are generated through bond cleavage in low rank coals at same condition [26], which suggests that the thermally induced fragments promote the hydrogen transfer amount. As temperature impacts on both of fragment formation rate and hydrogen transfer rate, the selection of temperature is important for different ranks of coal preheating process. The objective of this work is to explore the product distribution and hydrogen transfer of low rank coals at different preheating temperature during liquefaction. The experimental operating at preheating stage was simulated in a batch autoclave reactor with tetralin as solvent at 200–350 °C. Two Chinese low-rank coals, one lignite and one sub-bituminous coal, were used in this work. And the characteristics of products after preheating were examined to reveal the change of coal and tetralin at various temperatures. The experiments were conducted with N2 as working atmosphere in order to rule out the effect of hydrogen transferring from gas phase to coal samples and focus on the hydrogen donating role of solvent.
sieve column with FID detector were used to analyze C1–C4 hydrocarbons and CO, CO2, respectively. The solid-liquid mixture was totally transferred to a Soxhlet apparatus and extracted with n-hexane for 24 h. The extraction included tetralin and its derivatives, and some light products from coal conversion. And these light products were named as oil in this work. Then n-hexane was removed from extraction by vacuum rotary evaporation, and the remaining tertralin and its derivatives and oil were analyzed without further separation. The n-hexane insoluble solid was dried in a vacuum oven at 80 °C for 12 h, which was weighted to calculate the solid yield by using Eq. (1). Each run was repeated at least 3 times to ensure the duplicability, and the deviations were b2%. The solid samples obtained at 200–350 °C were named as YN-200 or HM-200; YN-250 or HM-250; etc. respectively. The solid samples were further extracted with tetrahydrofuran (THF) for 24 h to separate preasphaltene and asphaltene (PAA). PAA yield was calculated by using Eq. (2). The THF insoluble fraction was named as residue, and the conversion of coal during preheating process was calculated by using Eq. (3).
2. Experimental section
c ¼ 100 ‐
ws ¼
m1 100 mc
ð1Þ
wp ¼
m1 ‐ m2 100 mc
ð2Þ
m2 100 mc
ð3Þ
2.1. Materials Two low rank coals, Yunnan lignite (YN) and Hami sub-bituminous coal (HM), were used in this work. The coal samples were grounded and sieved to b 150 μm, dried in a vacuum oven at 80 °C for 24 h before using. The proximate and ultimate analyses of the two coals are listed in Table 1. The solvents including tetralin (purity ≥ 99.0%), n-hexane (purity ≥ 98.0%), diphenyl (purity ≥ 99.5%), tetrahydrofuran (purity ≥ 99.0%) were commercially pure chemical reagents and used without further purification. 2.2. Preheating treatment In each run, 5.00 g coal (dry basis) and 10.00 g tetralin were packed into 100 mL autoclave (Parr 4598, USA), which was then charged with N2 to 3 MPa (cold pressure) after leak-checking. The reactor was heated from ambient temperature to the desired temperature (200–350 °C) at 5 °C/min with stirring rate of 400 rpm simultaneously. It was kept at the final temperature for 120 min before stopping heating. The gaseous products were collected in a gas bag and analyzed by a gas chromatograph (GC-950, Haixin, China). A 5A sieve column with TCD detector was used to detect H2, N2, CH4, and CO. An Al2O3 column and carbon Table 1 Proximate and ultimate analyses of solid samples obtained at different temperatures. Sample
HM HM-200 HM-250 HM-300 HM-350 YN YN-200 YN-250 YN-300 YN-350
Proximate analysis (wt.%)
Ultimate analysis (wt.%)
Atomic ratio
Mad
Ad
Vdaf
Cdaf
Hdaf
Odafa
Ndaf
Sdb
H/C
O/C
6.94 4.36 3.55 2.77 1.55 9.29 5.24 4.36 2.94 2.32
6.76 6.80 6.95 7.49 10.23 10.27 10.73 11.85 12.94 17.33
54.19 53.44 52.47 52.03 48.78 51.46 49.85 46.42 45.34 42.93
74.35 76.19 77.42 78.78 81.34 68.35 71.30 75.09 77.60 81.47
5.61 5.71 5.77 5.87 5.89 4.73 4.84 5.04 5.17 5.47
18.68 16.76 15.47 13.87 11.00 24.83 21.72 17.60 14.89 10.33
1.07 1.09 1.11 1.21 1.46 1.28 1.36 1.48 1.57 1.92
0.27 0.23 0.21 0.25 0.28 0.73 0.71 0.69 0.67 0.66
0.91 0.90 0.89 0.89 0.87 0.83 0.81 0.81 0.80 0.81
0.19 0.16 0.15 0.13 0.10 0.27 0.23 0.18 0.14 0.10
ad: air dry basis; d: dry basis; daf: dry ash free basis. a By difference. b Total S.
where, ws is the solid yield (wt.%, daf); wp is PAA yield (wt.%, daf); c is the conversion (wt.%, daf); m1 is the weight of n-hexane insoluble solid fraction, g (daf); m2 is weight of tetrahydrofuran (THF) insoluble fraction, g (daf); mc is the weight of feed coal, g (daf). Yield of water produced in the preheating process was determined by the oxygen mass balance in all products, as following Eq. (4): ww ¼ 18
Oc‐Os‐Og‐Oo 100 16 mc
ð4Þ
where, ww is the water yield (wt.%, daf); Oc, Os, Og, and Oo represent the content of oxygen in raw coal, solid, gas, and oil, respectively. The oxygen content in oil and solid were determined by using an elemental analyzer instead of by difference to obtain more accurate data. Due to the relatively high boiling points of tetralin and naphthalene, some light oil would be also evaporated during the evaporation of solvent. The weight of residual oil would be lower than the actual value. Therefore, oil yield was obtained by difference as shown in Eq. (5): wo ¼ 100‐ws ‐ww ‐wg
ð5Þ
where, wo and wg are the oil yield and gas yield (wt.%, daf), respectively. 2.3. Characterization Thermo-gravimetric analyzer (TG, Setsys Evolution, SETARAM, France) coupled with a mass spectrometer (MS, Omnistar, Pfeiffer Vacuum) was used to study the pyrolysis behavior of raw coal. About 11 mg sample was placed in a corundum crucible and heated from 30 to 1050 °C at 10 °C/min in 100 mL/min argon atmosphere. The chemical component of the solvent-oil mixture was identified and quantified by gas chromatography-mass spectrometer (GC–MS, ISQ, Thermo Scientific, USA). The samples were separated using a 30 m × 0.25 mm TR-5MS capillary column (Thermo Scientific, USA). Helium was employed as the carrier gas. The oven temperature was raised from 40 °C to 250 °C at 5 °C/min. The components were identified based on the attached library. Diphenyl was selected as internal standard to quantify tetralin and naphthalene. The contents of carboxyl group in raw coal were determined by chemical method [27]. The aqueous determinations of acidity were
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
155
Fig. 1. TG and DTG curves of HM and YN.
performed by ion exchanging 250 mg coal with 60 mL of BaCl2/ triethanolamine/HCl buffer under vacuum for 180 min. The mixture was then filtered and the filter cake was rinsed with distilled water. Afterwards, the filter cake was stirred with 10 mL, 0.1 mol/L HCl under vacuum. Then the filtrate was taken and titrated against 0.05 mol/L NaOH. The endpoint was taken as the volume of NaOH used to attain pH 5. X-ray photoelectron spectroscopy (XPS) of raw coal and solid samples were obtained with AXIS ULTRA DLD (Kratos, UK) equipped with a monochromatic Al Kα X-ray source (1486.6 eV). High resolution scanning of the C1s region was performed in a step of 0.05 eV with pass energy of 20 eV. Peak deconvolution of C1s spectra was accomplished using the XPSPEAK41 software. 2.4. Determination of hydrogen transfer (HT) from tetralin to coal The amount of hydrogen transferred from tetralin to coal was calculated by using Eq. (6) [28]. HT ðmg=g; daf Þ ¼
4 1:008 mn 1000 128:17 mc
ð6Þ
where, mn is the corrected amount of naphthalene in the recovered solvent constituent determined by GC–MS analysis. Factors of 1.008 and 128.17 are the weight of hydrogen atom and the molecular weight of naphthalene.
and oxygen in YN. However, a reverse result is observed at higher temperature. The final weight loss of YN and HM are 48% and 44% respectively when temperature rises to 1050 °C. The higher weight loss of HM is attributed to its higher volatile matter and lower ash content. The DTG curves can be divided into three stages according to the decomposition of functional groups and condensation of coal structure. At first, the moisture and adsorbed gases are removed when temperature rises from 30 to 200 °C. The second pyrolysis stage is from 200 to 600 °C, at which large amount of volatile matter are produced and extensive polymerization reactions take place. With the temperature rising to above 600 °C, the secondary condensation reactions occur, as well as the decomposition of some mineral matter such as pyrite and CaCO3. From DTG curves, the weight loss rate of YN is higher than that of HM at first stage, which is attributed to the adsorption of water and gas. The disparity of weight loss rate becomes more obvious with temperature rising to the second stage. Moreover, the maximum mass loss rate of the two coal samples, which represents the volatile release level, suggests the higher thermal reactivity of HM at high temperature. The peak temperature of YN at the maximum mass loss rate is about 397 °C, while it shifts to a higher zone (436 °C) for HM. It is related to the interactions between mobile phase and fixed phase in coal and thermal conductivity of coal [29], indicating cross-linking level becomes enhanced as the rank of coal rises. For the two samples, the thermolysis rate of YN is more rapid initially due to the higher heteroatoms in YN sample. With the temperature increasing higher, more fragmental structure in HM breaks up and a large amount of volatile matter is released. As a result, the final weight loss of HM is higher at 1050 °C.
3. Results and discussion 3.2. Product distribution during preheating stage 3.1. Thermo-gravimetric analysis (TGA) of raw coal In order to elucidate the thermal behavior of raw coal and understand the effect of preheating temperature on coal conversion, the TG and DTG (differential thermal gravity) curves of raw coal have been obtained, as shown in Fig. 1. TG curves depict that YN loses more weight than HM from 30 to 442 °C, attributing to the higher content of moisture
Fig. 2 presents the product distribution at different preheating temperatures of the two coals. Solid here is n-hexane insoluble fraction, which is the residual solid after separating oil and water. The solid yields of the two coals show descending trends with the increase of preheating temperature. The solid loss after preheating treatment is attributed to the decomposition of labile functional groups in low rank coal. In
Fig. 2. Product distribution after preheating treatment. a: YN; b: HM.
156
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
Fig. 3. Gas compositions and yield evolved in the preheating process. a: YN; b: HM.
comparison with HM, YN obtains lower solid yield at same preheating temperature. For example, the yield of YN-300 is 71.94%, lower than 83.10% of HM-300. This is due to the higher thermal reactivity of YN at this temperature, as shown in Fig. 1. The solid yields of HM and YN decrease steadily from 92.98% and 89.55% to 83.10% and 71.94%, respectively, at temperature range of 200 to 300 °C. And with temperature further reaching 350 °C, the solid yields of HM and YN decrease to 55.81% and 48.31%, respectively. The descending trend from 200 to 300 °C is much lower than that from 300 to 350 °C, which is related to the thermolysis rates of raw coal at the temperature range. Fig. 1 suggests that thermolysis rates of the two coals dramatically increase above 300 °C. The fragments formed at high temperature are stabilized by hydrogen from tetralin and convert into light products. Accordingly, the yields of oil, water, and gas represent an increase trend with the raising temperature. Due to the rapid thermolysis rates above 300 °C, the oil yields of the two coals from 300 to 350 °C increase more than twice that below 300 °C. The content of water was calculated by oxygen mass balance during the reaction system. Water yield shows a noticeable amount during the preheating stage. Water yields of YN and HM reach about 15.72% and 9.68% respectively at 350 °C. Source of water differs with preheating temperature rising. Initially, water comes from adsorbed water in pores of coal, which couldn't be removed in drying process. When the temperature increases further, oxygen-containing functional groups in coal, such as ketone and hydroxyl structure, can convert into water by adding hydrogen from coal and solvent [30–32], which increase the consumption of hydrogen. More water yields are observed for YN because of its higher oxygen content during the preheating stage. 3.3. Gas analysis during preheating stage
are n-hexane soluble fraction by separating oil and water, which would be converted into light products subsequently. n-Hexane did not affect the solid structure due to its chain alkane composition. The TG-MS analyses of the solid samples show that the intensity of tetralin was extremely weak, indicating that the residual solvent in solid samples could be ignored. The analyses of solid characteristics would reveal the variation of coal chemical structure in the preheating stage accurately and help to understand its subsequent thermal conversion behavior. 3.4.1. Proximate and ultimate analyses of solid samples Table 1 shows the proximate and ultimate analyses of raw coal and solid samples. Obviously, hydrogen content of solid samples increases with the rising temperature, which is different from dehydrogenation of coal during pyrolysis without hydrogen donor [33]. In addition, a fraction of raw coal can be converted into oil during the preheating process. These results show that hydrogen from tetralin transfers to solid and oil effectively during the preheating process. As a result, H/C values of solid samples are still at a high level with the raising temperature. A large amount of oxygen in raw coal transfers into water, CO2, and CO under thermal decomposition. As a result, oxygen content in solid samples decreases obviously with the increasing temperature. Hydrogenation and deoxygenation occur during the preheating process for the solid samples, and it is beneficial to its subsequent liquefaction [7]. 3.4.2. Carbon functionalities of solid samples analyzed by XPS In order to investigate the carbon functionalities in raw coal and solid samples, the XPS C1s spectra were recorded. The C1s spectra can be deconvoluted to show four types of carbon [34–36]: the 284.8 eV peak represents contributions from both aromatic and aliphatic carbon (C\\C, C _C, and C\\H); the 286.3 eV peak represents hydroxyl/phenol
The composition of gas evolved during the preheating treatment is shown in Fig. 3. The gas yields of the two coals increase with rising temperature, and the dominant gas is CO2, which is evolved by de-carboxylation reaction. The content of carboxyl in YN and HM determined by chemical methods is 0.95 and 0.65 mmol/g, respectively. Thus, CO2 yield from preheating treatment of YN is higher than that of HM. H2 is generated when temperature is above 250 °C. However, its amount is extremely low, indicating the condensation reaction of coal with the presence of tetralin is negligible. In comparison with YN, the aliphatic hydrocarbons amount of HM is higher at 350 °C, which is consistent with H/C ratio of the two coals. 3.4. Effect of preheating treatment on characteristics of solid samples With the process to convert into oil, water, and gas, coal undergoes the breakup of macro-structure and hydrogen transfer reaction. In order to reveal the effect of tetralin on coal hydrogenation and variation of carbon functionalities in the preheating process, characteristics of solid obtained at different temperatures are studied. Here, solid samples
Fig. 4. C1s spectra of XPS for YN.
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159 Table 2 Carbon functionalities in raw coal and solid samples from XPS analysis. Sample
Relative proportion (area percent, %) C\ \C, C _C, C\ \H
C\ \O
C_O, O\ \C\ \O
O\ \C_O
YN YN-200 YN-250 YN-300 YN-350 HM HM-200 HM-250 HM-300 HM-350
74.06 76.12 80.26 80.55 84.81 79.38 81.03 82.36 84.60 83.37
18.83 18.44 16.31 16.91 11.94 13.42 12.86 13.22 11.74 11.99
4.03 3.06 1.34 1.21 0.88 3.39 3.00 2.22 2.16 0.84
3.08 2.38 2.09 1.33 2.37 3.81 3.70 2.20 1.50 3.80
groups or ether (C\\OR); the 287.5 eV peak corresponds to carbonyl carbon (C _O and O\\C\\O); the 289.0 eV peak corresponds mainly to carboxylic group, ether or lactone (O\\C_ O). The relative full width at half maximum values of all peaks were fixed to 1.5 (±0.1) eV. Example of XPS C1s spectra for YN into different organic carbon components is shown in Fig. 4. The XPS C1s spectra of other samples are shown in the supplementary data. The relative proportion (RP) for each functional group from C1s deconvolution of all samples is listed in Table 2. Aliphatic and aromatic carbon are the main groups with RP of 74.40% and 79.38% for YN and HM respectively, and the RPs of aliphatic and aromatic carbon groups increase with raising preheating temperature due to cleavage of side-chains. The C\\O bonds, including hydroxyl, phenol groups and ether, show a slight change during the process. Yan et al. have found ethers and alcohols start decomposing at 150 and 200 °C [15], and phenolic groups remain stable below 350 °C. The result indicates most of C\\O bonds are phenolic groups in the two coals, especially for HM. The RPs of carbonyl and carboxyl decrease obviously with the rising preheating temperature. The\\COO peak positions of YN-350 and HM-350 shift to 289.4 eV, which are attributed to the signal of carbonate. The increasing amount of\\COO in these two samples indicates that carbonate is formed at 350 °C and enriched in the solid samples. Perry et al. have found that the amount of carbonate formed in the liquefaction residue was mainly proportional to the amount of ion-exchangeable calcium in feed coal [37]. This suggests that the ionexchangeable calcium in feed coal converts into stable carbonate under certain conditions. The formed carbonate may cause operational difficulty in liquefaction reactor, which should be prevented in liquefaction. 3.5. Hydrogen transfer effect during preheating stage It has been reported that tetralin is stable below 400 °C [24,38], and our experimental result shows that the tetralin mass remained 98.5% at 350 °C in the absence of coal. However, dehydrogenated products were detected under the same condition in the presence of coal. In this work,
157
naphthalene was the main tetralin dehydrogenated product, and dihydronaphthalene was not detected in the GC–MS results. The variation of HT with different preheating temperatures of HM and YN are shown in Fig. 5a, and the corresponding transfer ratios to maximum HT for the two coals are shown in Fig. 5b. The maximum HT was calculated by dehydrogenation of tetralin. It is assumed that all tetralin would convert into naphthalene, and the maximum HT would be 64.89 mg/g and 67.54 mg/g for HM and YN, respectively. As additional information, the overall tetralin mass balance for the reaction reached level of 91–99 wt.%. Fig. 5 shows that hydrogen transfer from tetralin achieves a perceptible extent at 200 °C, and increases significantly above 300 °C. HT reaches 14.74 mg/g and 15.77 mg/g at 350 °C for HM and YN, accounting for 22.72% and 26.04% of the maximum HT, respectively. Accordingly, coal promotes dehydrogenation of tetralin during the preheating process. Temperature is one of the most important factors for hydrogen transfer from tetralin to coal [25]. Attributing to the large amount of fragmental radicals produced at higher temperature, hydrogen transfer increases rapidly from 300 to 350 °C. It has been reported that the quantity of coal-derived radicals during pyrolysis process increases with the decrease of carbon content in coal or rank of coal [26,39–41]. This is because the higher amount of heteroatoms or labile functional groups and lower carbon in low rank coal result in the higher thermal reactivity at same temperature. The unstable coal-derived radicals capture hydrogen from hydrogen donor solvent. Thus the HT amount of YN is higher than that of HM at same temperature in the preheating stage. In traditional liquefaction process, the conversion of coal is defined as the percentage of THF soluble fractions. By reference to the method, conversion in the preheating stage was also investigated and its correlation with hydrogen transfer amount is shown in Fig. 6. The plot of conversion vs. HT displays good linear fit with R-squared value of 0.95. Conversion increases monotonically as a function of hydrogen transfer amount from tetralin to coal. However, it has been reported that greater amounts of hydrogen transfer do not further enhance conversion significantly when HT reaches N 30 mg/g, due to the participation in hydrocracking of heavier products [25]. A linear relation is also observed between oil and hydrogen transfer from Fig. 6. From the product distribution in Fig. 2, the oil yields show little difference between the two coals. Moreover, higher oil yield and lower HT are achieved for HM at 350 °C because of the difference between H/C ratios of HM and YN, which is an important factor for coal selection in direct liquefaction. Relatively high H/C ratio of coal can reduce hydrogen consumption and is beneficial for production of oil. HM has a higher H/C ratio, thus it achieves higher oil yields at 350 °C with a lower consumption of hydrogen. The object of liquefaction is to achieve good quality and high yield of oil by increasing conversion. Therefore, via selecting the suit feedstock, low HT and high oil yield can be obtained at low liquefaction temperature. PAA is THF soluble fraction except oil and water. The result in Fig. 6 shows that the PAA amount increases with the increase of HT. According to reference [42] about the pathways of liquefaction, PAA, the
Fig. 5. Hydrogen transfer at different preheating temperatures.
158
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159
Fig. 6. Effect of HT on conversion and some products yields.
intermediate of coal to oil and gas, are produced in parallel as the conversion increases during the initial coal dissolution stage. Therefore, the hydrogen donor promotes the conversion of coal to light products. 4. Conclusions Preheating treatment of two low rank coals between 200 and 350 °C was performed with tetralin as hydrogen donor solvent. During the preheating process, total yields of light products (oil, gas and water) increase with raising temperature, and they achieve the highest level of 44.19% and 51.69% at 350 °C for HM and YN, respectively. The presence of tetralin could effectively increase the hydrogen content of solid samples during the preheating stage, which is beneficial for subsequent conversion. Moreover, the labile oxygen-containing functional groups in raw coal, such as carboxyl groups, ethers, and alcohols decompose gradually during the preheating stage. Hydrogen transfer from tetralin to coal occurs even at the temperature of 200 °C, and it is enhanced by the large amount of fragmental radicals formed at higher temperature. In addition, the amount of hydrogen transfer for YN is higher than that of HM below 350 °C because of their thermal reactivity difference. Both conversion and oil yield are related to the hydrogen transfer amount at 200–350 °C. A positive correlation is observed between the hydrogen transfer and oil yield, and high H/C ratio of raw coal also promotes oil yield. Therefore, increasing hydrogen transfer amount and selecting feed coal with high H/C are effective for improving the oil yield and conversion during liquefaction. Acknowledgments This work was supported by the National Natural Science Foundation of China (21576274), and Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA07060100). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.fuproc.2017.01.028. References [1] H. Osman, S.V. Jangam, J.D. Lease, A.S. Mujumdar, Drying of Low-Rank Coal (LRC)—a review of recent patents and innovations, Drying Technol. 29 (2011) 1763–1783. [2] P.R. Solomon, M.A. Serio, G.V. Despande, E. Kroo, Cross-linking reactions during coal conversion, Energy Fuel 4 (1990) 42–54. [3] S. Vasireddy, B. Morreale, A. Cugini, C. Song, J.J. Spivey, Clean liquid fuels from direct coal liquefaction: chemistry, catalysis, technological status and challenges, Energ. Environ. Sci. 4 (2011) 311–345. [4] I. Mochida, Y. Moriguchi, Y. Korai, H. Fujitsu, K. Takeshita, Efficient liquefaction of Australian brown coal with a hydrogen-donating solvent under atmospheric pressure, Fuel 60 (1981) 746–747.
[5] I. Mochida, K. Iwamoto, T. Tahara, Y. Korai, H. Fujitsu, K. Takeshita, Liquefaction of subbituminous coals under apparently non-hydrogenative conditions, Fuel 61 (1982) 603–609. [6] Z. Liu, S. Shi, Y. Li, Coal liquefaction technologies-development in China and challenges in chemical reaction engineering, Chem. Eng. Sci. 65 (2010) 12–17. [7] I. Mochida, O. Okuma, S.H. Yoon, Chemicals from direct coal liquefaction, Chem. Rev. 114 (2014) 1637–1672. [8] A. Kidoguchi, H. Itoh, M. Hiraide, E. Kaneda, H. Ishibashi, M. Kobayashi, K. Ikeda, K. Imada, K. Inokuchi, Simulation of initial stage reactions in the direct coal liquefaction of sub-bituminous coals, Fuel 80 (2001) 1325–1331. [9] X. Shan, K. Li, X. Zhang, H. Jiang, H. Weng, Reaction kinetics study on the heating stage of the Shenhua direct coal liquefaction process, Energy Fuel 29 (2015) 2244–2249. [10] L. Artok, A. Davis, G.D. Mitchell, H.H. Schobert, Swelling pretreatment of coals for improved catalytic liquefaction, Fuel 71 (1992) 981–991. [11] H. Hu, G. Sha, G. Chen, Effect of solvent swelling on liquefaction of Xinglong coal at less severe conditions, Fuel Process. Technol. 68 (2000) 33–43. [12] Y. Inukai, S. Arita, H. Hirosue, Effect of preheat treatment on coal-liquefaction, Energy Fuel 9 (1995) 67–70. [13] F. Haghighat, A. de Klerk, Direct coal liquefaction: low temperature dissolution process, Energy Fuel 28 (2014) 1012–1019. [14] H. Shui, J. Liu, Z. Wang, M. Cao, X. Wei, Effect of pre-swelling of coal at mild temperatures on its hydro-liquefaction properties, Fuel Process. Technol. 90 (2009) 1047–1051. [15] J. Yan, Z. Bai, J. Bai, Z. Guo, W. Li, Effects of organic solvent treatment on the chemical structure and pyrolysis reactivity of brown coal, Fuel 128 (2014) 39–45. [16] F. Liu, X. Wei, M. Fan, Z. Zong, Separation and structural characterization of the value-added chemicals from mild degradation of lignites: a review, Appl. Energy 170 (2016) 415–436. [17] L. Huang, H.H. Schobert, Comparison of temperature conditions in direct liquefaction of selected low-rank coals, Energy Fuel 19 (2005) 200–207. [18] A. Tsutsumi, K. Yoshida, Rheological behaviour of coal-solvent slurries, Fuel 65 (1986) 906–909. [19] F.E. Ndaji, I.M. Butterfield, K.M. Thomas, Changes in the macromolecular structure of coals with pyrolysis temperature, Fuel 76 (1997) 169–177. [20] A. Ishihara, D. Nishigori, M. Saito, I.P. Sturisna, W. Qian, T. Kabe, Elucidation of hydrogen mobility in functional groups of coals using tritium tracer methods, Energy Fuel 16 (2002) 32–39. [21] A. Ishihara, H. Takaoka, E. Nakajima, Y. Imai, T. Kabe, Estimation of hydrogen mobility in coal using a tritium trace method. Hydrogen extrange reactions of coals with tritiated-water and molecular hydrogen, Energy Fuel 7 (1993) 362–366. [22] A. Ishihara, I.P. Sutrisna, T. Miura, M. Saito, E.W. Qian, T. Kabe, Elucidation of hydrogen transfer between coal and tritiated organic solvent, Energy Fuel 16 (2002) 1490–1498. [23] K. Ceylan, A. Olcay, Dissolution of two Turkish lignites in tetralin under hydrogen or nitrogen: effects of the extraction parameters on the conversion, Fuel 71 (1992) 539–544. [24] J.J. Devlieger, A.P.G. Kieboom, H. Vanbekkum, Behavior of tetralin in coal-liquefaction - examination in long-run batch-autoclave experiments, Fuel 63 (1984) 334–340. [25] L. Artok, H.H. Schobert, O. Erbatur, Temperature-staged liquefaction of selected Turkish coals, Fuel Process. Technol. 37 (1994) 211–236. [26] M. Liu, J. Yang, Y. Yang, Z. Liu, L. Shi, W. He, Q. Liu, The radical and bond cleavage behaviors of 14 coals during pyrolysis with 9,10-dihydrophenanthrene, Fuel 182 (2016) 480–486. [27] D.J. Allardice, L.M. Clemow, W.R. Jackson, Determination of the acid distribution and total acidity of low-rank coals and coal-derived materials by an improved barium exchange technique, Fuel 82 (2003) 35–40. [28] M. Sugano, S. Ohura, R. Endoh, K. Hirano, K. Mashimo, Effects of hydrogen transfer by exchanged cobalt upon liquefaction of low rank coal, Fuel 101 (2012) 228–233. [29] L. Zou, L. Jin, Y. Li, S. Zhu, H. Hu, Effect of tetrahydrofuran extraction on lignite pyrolysis under nitrogen, J. Anal. Appl. Pyrolysis 112 (2015) 113–120. [30] R. Dabestani, P.F. Britt, A.C. Buchanan, Pyrolysis of aromatic carboxylic acid salts: does decarboxylation play a role in cross-linking reactions? Energy Fuel 19 (2005) 365–373. [31] K. Miura, K. Mae, K. Sakurada, K. Hashimoto, Flash pyrolysis of coal following thermal pretreatment at low temperature, Energy Fuel 6 (1992) 16–21. [32] D.C. Cronauer, D.M. Jewell, Y.T. Shah, R.J. Modi, Mechanism and kinetics of selected hydrogen transfer reactions typical of coal liquefaction, Ind. Eng. Chem. Fundam. 18 (1979) 153–162. [33] J. Zhang, Study on the Oxygen Transfer Laws during coal's Pyrolysis Process and Its Effect on the Coking Property, Anhui university of technology, China, 2013. [34] S.R. Kelemen, M. Afeworki, M.L. Gorbaty, M. Sansone, P.J. Kwiatek, C.C. Walters, H. Freund, M. Siskin, A.E. Bence, D.J. Curry, M. Solum, R.J. Pugmire, M. Vandenbroucke, M. Leblond, F. Behar, Direct characterization of kerogen by X-ray and solid-state C-13 nuclear magnetic resonance methods, Energy Fuel 21 (2007) 1548–1561. [35] S.R. Kelemen, K.D. Rose, P.J. Kwiatek, Carbon aromaticity based on XPS II to II∗ signal intensity, Appl. Surf. Sci. 64 (1993) 167–174. [36] Z. Han, X. Zeng, C. Yao, G. Xu, Oxygen migration in torrefaction of Eupatorium adenophorum Spreng and its improvement on fuel properties, Energy Fuel 29 (2015) 7275–7283. [37] G.J. Perry, A. Gray, G.H. Mackay, Carbonate formation during hydrogenation of Victorian brown coal, Fuel Process. Technol. 10 (1985) 285–297. [38] M.C. Tsai, S.W. Weller, Ctalysis of hydrogen transfer in a tetralin-coal system, Fuel Process. Technol. 2 (1979) 313–316.
P. Hao et al. / Fuel Processing Technology 159 (2017) 153–159 [39] A. Ishihara, I.P. Sutrisna, M. Ifuku, E.W. Qian, T. Kabe, Elucidation of hydrogen mobility in coal using a fixed bed flow reactor hydrogen transfer reaction between tritiated hydrogen, coal, and tetralin, Energy Fuel 16 (2002) 1483–1489. [40] L. Petrakis, D.W. Grandy, Electron spin resonance spectrometric study of free radicals in coals, Anal. Chem. 50 (1978) 303–308.
159
[41] L. Petrakis, D.W. Grandy, Free radicals in coals and coal conversion. 3. Investigation of the free radicals of selected macerals upon pyrolysis, Fuel 60 (1981) 115–119. [42] R.A. Keogh, B.H. Davis, Comparison of liquefaction pathways of a bituminous and subbituminous coal, Energy Fuel 8 (1994) 289–293.