Abstract

Depolymerization of coal increases the tar yield in coal pyrolysis and enhances the thermoplasticity of the coal, which makes coal more favorable for producing coke and other value-added products like graphite electrodes, carbon fiber, aromatic chemicals, etc. In this study, the authors have proposed a novel coal depolymerization method that combines the oxidation reactions by molecular oxygen and the following reduction reactions by the coexisting gaseous formic acid to upgrade a bituminous coal at 90–150 °C under atmospheric pressure. The softening and melting performance of the treated coals was enhanced when oxygen and formic acid coexisted in gas phase at 90–130 °C. The amount of low-molecular-weight compounds in the coal treated at 90 °C in air containing formic acid vapor significantly increased by 29.0% of the amount of low-molecular-weight compounds in the raw coal measured by the solvent extraction method. Thus, the authors have succeeded in depolymerizing coal by the treatment under mild conditions, which is expected to contribute to coal’s efficient utilization such as increasing the coal extracts and tar yield in the extraction and pyrolysis process, upgrading coal to be more suitable for the raw materials of coke, etc.
1. Introduction
Coal has widely been applied as the raw material of value-added chemical products. Coke, carbon black, graphite electrode, carbon fiber, aromatic chemicals, etc. can be produced by appropriate processing of coals.1 Depolymerization of coal is the key to utilizing coal efficiently since the complete solubilization of coal in conventional solvents at ambient temperature is the ultimate target of coal utilization.2 When low-molecular-weight compounds (LMWC) increased in coal, the thermoplasticity of coal would be enhanced, which is favorable for coke production, and the yield of tar and pitch would also increase, which means more value-added chemical products could be prepared from coal.
Pyrolysis, involved in liquefaction, gasification, and combustion, is one of the most important coal conversion processes.2 The coal liquefaction process depolymerizes coal by the cleavages of ether and methylene bridges.3 A typical liquefaction process, NEDOL process, applied 728 K, 17 MPa, a metal catalyst, and a hydrogen supplier solvent for the coal depolymerization.4 A significant problem of coal liquefaction is that the conditions are too severe, so the cost of the heat energy and plant device is too high to be commercialized.5 A lot of trials have been done to decrease the cost of coal depolymerization via pyrolysis. Liang et al.6 depolymerized two kinds of low-rank coals and the tar yields increased from 3.6 to 5 to 7.8% for Neimeng coal, and from 8.1 to 11.4 to 10.9% for Xinjiang coal after the addition of Mo- and Fe-based catalysts in a batch reactor at 550 °C. Liu et al.7 increased the organic liquid products of coal from 0.2 to 1.6% on adding 1 wt % iron-based catalyst in a Gray-King assay reactor from 150 to 600 °C. These works didn’t apply high pressure to coal pyrolysis, but the yield of LMWC increased only a little, which is not suitable for coal depolymerization.
Oxidation has also been applied for coal depolymerization. Kapo et al.8 and Bimer et al.9 attempted to oxidize coals by oxygen in a NaOH solution under the mild conditions of 110 to 270 °C, 4.0 to 7.5 MPa. They achieved gathering the water-soluble organics in as much as 50% yield on a carbon basis. However, O2 oxidation under severe conditions converts 50% of the carbon into CO2 and would not be suitable for the coal depolymerization process. Milder conditions are required for depolymerization by molecular oxygen.
The mechanism of coal oxidation at low temperatures has been well studied, which is generally considered as two parallel reaction sequences; that is, the direct burn-off and sorption sequences. In the sorption sequence, firstly, the chemisorbed oxygen on the surfaces of the coal pores is inserted into the C–H bonds of −CH2– or −CH3 to form peroxides.10 Thus, ethylene bridges and hydroaromatic compounds, which are abundant in coal, are oxidizable.11 Secondly, these peroxides are unstable and can immediately decompose to form H2O, accompanied by C–C bond or C–O bond cleavage and radical formation.10 Since the radicals formed are also unstable, they would reconnect with each other or oxygen and get polymerized again, forming ether.12 To suppress the polymerization, the H radical is needed to stabilize the coal radicals before the polymerization.
In this study, the authors have proposed a method that combines the oxidation reaction by molecular oxygen and the radical stabilization reaction by formic acid (FA) to promote the depolymerization of coal under mild conditions. The reaction pathway is shown in Figure 1. Treatment with air containing FA vapor was performed, expecting the oxygen oxidation to cleave coal’s C–C or C–O bonds and FA to stabilize the radicals formed by oxygen oxidation immediately before the radicals reconnect to each other.
Figure 1.
Example of the Reaction Pathways.
The validity of the proposed method was examined by evaluating the depolymerization degree of the treated coals, and the quantification of the amount of LWMC in coal was carried out by a solvent extraction method proposed by the authors that uses a flowing stream of a nonpolar solvent such as tetralin or 1-methylnaphthalene under 10 MPa at temperatures lower than 350 °C.13−16 There have been several solvent extraction methods for coal. Strong polar solvents such as pyridine17 and binary solvents such as CS2-NMP18 were used in these methods to investigate the structure of coal. The strong interaction between coal and solvents, however, makes it difficult to separate the solvent from coal. The proposed solvent extraction method is advantageous because a large amount of thermally released molecules can be dissolved in nonpolar solvents such as 1-methylnaphthalene or tetralin and the coal and the solvent can be easily separated.13 Another advantage of the extraction method is, the extraction has excellent coal properties, such as no ash, no water, and high-calorie density. Okuyama et al.19 developed the process to produce ash-free coal (Hyper-coal) using the extraction. Li et al.20 prepared the carbon fiber by using the extracts obtained by this method as the precursor and obtained carbon fibers were comparable to the commercial carbon fibers. Thus, if the LMWC increased by the depolymerization method proposed in the study, the valuable extracts in the solvent extraction method would also increase, which is beneficial for coal’s efficient utilization.
2. Experimental Section
2.1. Coal Sample Preparation
An Australian bituminous coal was used in this study. The raw coal was mechanically sized into below 150 μm. The ultimate analysis results are given in Table 1.
Table 1. Elemental Analysis of the Raw Coal Used.
| ultimate
analysis [wt %, d.a.f.] |
atomic
ratio [-] |
ash | ||||
|---|---|---|---|---|---|---|
| C | H | N | O (diff.) | H/C | O/C | (wt %, d.b.) |
| 86.9 | 4.5 | 0.8 | 7.8 | 0.62 | 0.067 | 4.7 |
2.2. Experimental Procedure of the FA/Air Treatment
The coal was treated at 90–150 °C for 90 min using the apparatus shown in Figure 2. Nitrogen was supplied at the flow rate of 130 cm3/min while the temperature was increased at 10 °C/min; when the temperature reached the target value, the gas was switched to air at the same flow rate and was bubbled through a 50 wt % formic acid water solution. Treatment using nitrogen instead of air with the same bubbler was also carried out for comparison purposes. The gas at the exit of the quartz tube was collected in the gas bag.
Figure 2.
Schematic diagram of the apparatus adopted in this study.
2.3. Analyses of the Products
The gas collected in the gas bag was analyzed by gas chromatography (GC-12A, Shimadzu, Porapak Q column) to determine the amount of CO2 formed during the treatment.
The yield of the treated coal was calculated as follows
Ms is the weight of as-received coal used.
Mf is the weight of treated coal after the treatment.
The coal treated under each condition was evaluated through various analyses. The thermomechanical analyzer (TMA-50/60, Shimadzu) was used to analyze the thermoplastic behaviors of the treated coals. In the thermomechanical analysis (TMA), the displacement depth of a rod into a solid sample bed was continuously monitored while heating the sample at 10 °C/min in a nitrogen flow. A sample was placed in a pan (5.2 mm I.D. and 6.0 mm height) at a height of 1 mm and a rod of 4.3 mm diameter was loaded with a constant load of 0.098 N. The sample will have better thermoplasticity if it has a lower average molecular weight; thus, the TMA analysis, which measures the thermoplasticity of coal, could be a convenient method to evaluate the degree of depolymerization of coal.21,22
The solvent extraction method proposed by the authors was applied to quantify the amount of LMWC of raw and treated coal. The method extracts coal using a flowing stream of a nonpolar solvent, 1-methylnaphthalene, under 10 MPa at temperatures lower than 350 °C. The coal was separated into several fractions having different molecular weights without decomposition.19 The solvent extraction method was used to separate the coals into unextractable high-molecular-weight compounds “residue” (>1000 g/mol), extractable low-molecular-weight compounds, “deposit” (400–1000 g/mol), and “soluble” (<500 g/mol).22 The total yield of the deposit and soluble was used to evaluate the degree of depolymerization of coal in this study.
Elemental analysis (CHN Corder MT-6M, Yanaco) was used to investigate the change in the elemental composition of the coal through the treatment. Fourier transform infrared spectroscopy (FT-IR) spectroscopy (JIR-WINSPECT 50, JEOL) was employed to analyze the functional groups in the coal before and after the treatment.
3. Results and Discussion
3.1. Product Yield
The yields of the treated coals are shown in Figure 3. Whether the atmosphere was FA/ Air or FA /N2, the treated coal yield stayed almost 100% at 90–130 °C, which suggests that the proposed treatment method would not bring too much change to coal. The gas analyses showed that the formed gas in the treatment was mainly CO2 and H2O, since H2O was also contained in the supplied gas, which came from the formic acid water solution, it was difficult to measure the formed H2O. Therefore, we focused on the CO2 amount that was evolved during the treatment.
Figure 3.

Yield of coal under different treatment conditions.
The CO2 amount evolved in the treatment is shown in Figure 4. At each treatment temperature, the amount of CO2 evolved under FA/Air atmosphere was significantly more than that under FA/N2, suggesting that the presence of O2 promoted a reaction producing CO2. The candidate reactions involving O2 as a reactant and CO2 as a product are listed below.
| 1 |
| 2 |
| 3 |
| 4 |
(Ccoal, R, and Ar indicate the carbon in the coal, an alkyl group, and an aromatic group, respectively.)
Figure 4.

CO2 amount evolved under different treatment conditions.
To figure out the source of CO2 evolved during the treatment, comparison experiments at 130 °C were carried out and the results are shown in Figure 5. When there is only FA/Air without coal or only Coal/Air without FA in the reactor, the amount of CO2 is extremely small, which suggests that the reactions 1 and 2 hardly occurred under these conditions. When there was only Coal/FA without Air (O2) in the reactor, the amount of CO2 reached a higher level (0.089 mol/kg-coal) than the former groups (0.019 and 0.012 mol/kg-coal). It could be judged that the reaction 4 had taken place for the coal-inherent radicals. When there were Coal, FA, and Air (O2) all together in the reactor, the CO2 amount became much higher than that of any other groups and reached 0.43 mol/kg-coal. These results suggest that the CO2 is mainly generated from successive reactions of reaction 3 followed by reaction 4, where FA donated H radicals to coal radicals formed by the reactions with O2 and FA turned to CO2. It was thus suggested that the source of CO2 in the proposed treatment method was not coal but mainly FA, and the reduction of coal by FA was promoted by the presence of O2.
Figure 5.

CO2 amount evolved under each control experiment.
3.2. Thermoplastic Behavior of the Treated Coal
The thermoplasticity of the treated coal can reflect the depolymerization degree of the coal as mentioned. In this study, the thermoplasticity of the treated coals obtained under different conditions was evaluated using TMA. The TMA curves of the treated coals and raw coal are shown in Figure 6. The displacement of the rod was normalized by the initial height of the coal sample bed. As the temperature increased, the coal began to soften and melt at around 400 °C and the rod’s displacement started to fall down. The coal with the less value of minimum normalized displacement can be judged to have better thermoplasticity, which should result from the abundance of LMWC in the coal.
Figure 6.
TMA curves of the coals treated under different conditions.
In FA/N2 atmosphere, the TMA curves of the coals treated at 90–150 °C didn’t show significant change from that of the raw coal. However, when the atmosphere was FA/Air and the treatment temperature was from 90 to 130 °C, the minimum normalized displacement dropped to a smaller value than that for the raw coal. These results show that the treatments under FA/Air atmosphere are effective in enhancing the thermoplasticity of coal, which is indicative of depolymerization of the coal.
Figure 7 shows the effect of the treatment temperature on the minimum normalized displacement value in TMA for the treated coals prepared under FA/Air atmosphere and FA/N2 atmosphere. When the treatment temperature for FA/Air atmosphere was 150 °C, the treated coal’s minimum normalized displacement value was more than that of raw coal. For the treatment temperature from 90 to 130 °C, however, the minimum normalized displacement value of the treated coal was less than that of raw coal and decreased with the decrease of treatment temperature. It was suggested that there was a suitable temperature range for the proposed treatment method. When the treatment temperature is too high, polymerization would become dominant. This is presumably because the rate of the radical recombination reaction is faster than the reducing reaction rate by FA at high temperatures.
Figure 7.

Comparison of the minimum normalized displacement values in TMA.
3.3. Necessity of Coexistence of Oxygen and Formic Acid in the Treatment
To investigate the necessity of the coexistence of oxygen and formic acid in the proposed method, 1-step and 2-step experiments were carried out at 90 °C and 1 atm. In the 1-step treatment, oxygen and formic acid coexisted, while in the 2-step experiment, only air was supplied firstly for 90 min, and then formic acid vapor was supplied with nitrogen for 90 min. Figure 8 shows the TMA curves of the treated coal and the raw coal. In the 2-step treatment, the thermoplasticity of the treated coal got even worse than raw coal, whereas the thermoplastic behavior of the treated coal got better in the 1-step treatment. These results suggest that when oxygen and formic acid were supplied separately in the 2-step treatment, the radicals formed by the oxygen oxidation reaction could not be capped by H radicals but were immediately connected with each other, which resulted in polymerization. No radicals to be stabilized were left any longer when formic acid was supplied. Thus, it can be concluded that the coexistence of oxygen and formic acid is necessary for the proposed method.
Figure 8.

TMA curves of the coals obtained through 1-step and 2-step treatment (90 °C, 1 atm).
3.4. Effect of Water in the Treatment
As FA was supplied as moist vapor in this study, it is necessary to investigate the effect of water in the treatment. The experiments using solutions with different water and FA ratios (0 wt % H2O and 100 wt % FA, 50 wt % H2O and 50 wt % FA, 100 wt % H2O and 0 wt % FA) filled in the bubbler were carried out at 130 °C, 1 atm for 90 min Figure 9 shows the TMA curves of the treated coal and raw coal. The thermoplasticity of the coal treated with 0 wt % H2O and 100 wt % FA or 100 wt % H2O and 0 wt % FA didn’t get enhanced compared to that of raw coal, while the thermoplasticity of the coal treated with 50 wt % H2O and 50 wt % FA got enhanced. It is suggested that both water and FA are necessary for the proposed treatment. This could be considered to be the result of the water acting as a promoting agent during the interaction between coal and O212 to accelerate the cleavage reaction in the treatment.
Figure 9.

TMA curves of the coals treated under different H2O ratios (130 °C, 1 atm).
3.5. Quantification of LMWC in the Treated Coal
The LMWC in the treated coal was quantified by the proposed solvent extraction method. Figure 10 shows the yield of each fraction for the raw coal and the coals treated at 90 and 110 °C under FA/Air atmosphere. It was found that the yield of the total deposit and soluble amounts for both treated coals were higher than those for the raw coal. Especially for the coal treated at 90 °C, which had the lowest minimum normalized displacement value in the TMA, the yield of the total deposit and soluble amount increased by 10.1% on the dry raw coal basis, and by 29.0% on the raw coal’s total deposit and soluble amount basis through the treatment. These results suggest that the treatment method proposed in this study could increase the LMWC amount efficiently under mild conditions (90 °C-, 1 atm).
Figure 10.
Yield of each fraction for the raw coal and treated coal.
3.6. Elemental and Functional Group Analyses of Treated Coal
To investigate the chemical change through the treatment, elemental analysis and FT-IR analysis were conducted for the coal treated at 90 °C under FA/Air atmosphere, whose depolymerization degree was the most significant among the conditions employed in this study. Figure 11 shows the amount of each element in the treated coal and raw coal on the raw coal basis. The carbon amount hardly changed through the treatment, while oxygen and hydrogen amounts decreased by 0.29 and 0.66 mol/kg-raw coal, respectively. Even though molecular oxygen existed in the treatment, the oxygen content of the treated coal was still decreased.
Figure 11.
Amount of each element on the raw coal basis in the raw coal and the coal treated at 90 °C under FA/Air atmosphere.
Figure 12 shows the FT-IR spectra of the raw coal and the coal treated at 90 °C under the FA/Air atmosphere. The intensity of the peaks attributed to carboxylic acid O–H bending (1440–1395 cm–1), phenol O–H bending (1390–1310 cm–1), and aldehyde C–H bending (1390–1380 cm–1) increased through the treatment while the intensity of the peak attributed to ether C–O stretching (1150–1050 cm–1) decreased. One of the reasonable reaction paths in the proposed treatment is shown in Figure 13. The C–H bonds in the R–CH2–O–Ar structure were oxidized by oxygen to form peroxide and then the ether link was cleaved in two ways as shown in the figure. The H radicals from FA then stabilize the formed radicals to give aldehyde, phenol, and carboxylic acid. The path also is in good agreement with the elemental results that the carbon amount in the coal has hardly changed through the treatment.
Figure 12.
FT-IR spectra of the raw coal and the coal treated at 90 °C under FA/Air atmosphere.
Figure 13.
Example of reaction pathways occurring in the proposed treatment.
Additionally, the decrease in peak intensity attributed to O–H stretching (3200–3600 cm–1) was observed. Hydroxy groups, which often form hydrogen bonds, would become the cross-linking site and connect the molecules through a dehydration reaction upon heating, which would inhibit coal depolymerization. Thus, the elimination of hydroxy groups could also be considered an advantage of the proposed treatment although the reaction mechanism should be investigated in our future studies.
4. Conclusions
A novel coal treatment method combining the oxygen oxidation reaction and formic acid reduction reaction has been proposed in this study to increase the low-molecular-weight compound (LMWC) amount in the coal under mild conditions (90 °C-, 1 atm). It was found that bituminous coal’s LMWC was successfully increased by as much as 29.0% as compared to the LMWC of the raw coal by the treatment at 90 °C, 1 atm. It was confirmed that oxygen and formic acid must coexist to depolymerize coal by the proposed method. Elemental and FT-IR analyses suggested that the method can cleave the ether groups in coal and remove part of the hydroxy groups as well.
Acknowledgments
This work was supported by JST SPRING, Grant Number JPMJSP2110.
The authors declare no competing financial interest.
References
- Schobert H. H.; Song C. Chemicals and materials from coal in the 21st century. Fuel 2002, 81, 15–32. 10.1016/S0016-2361(00)00203-9. [DOI] [Google Scholar]
- Li Z. K.; Yan H. L.; Yan J. C.; Wang Z. C.; Lei Z. P.; Ren S. B.; Shui H. F. Drying and depolymerization technologies of Zhaotong lignite: A review. Fuel Process. Technol. 2019, 186, 88–98. 10.1016/j.fuproc.2019.01.002. [DOI] [Google Scholar]
- Yoshida T.; Tokuhashi K.; Maekawa Y. Liquefaction reaction of coal: 1. Depolymerization of coal by cleavages of ether and methylene bridges. Fuel 1985, 64, 890–896. 10.1016/0016-2361(85)90138-3. [DOI] [Google Scholar]
- Hirano K. Outline of NEDOL coal liquefaction process development (pilot plant program). Fuel Process. Technol. 2000, 62, 109–118. 10.1016/S0378-3820(99)00121-6. [DOI] [Google Scholar]
- Mochida I.; Okuma O.; Yoon S. H. Chemicals from direct coal liquefaction. Chem. Rev. 2014, 114, 1637–1672. 10.1021/cr4002885. [DOI] [PubMed] [Google Scholar]
- Liang L.; Huang W.; Gao F.; Hao X.; Zhang Z.; Zhang Q. Mild catalytic depolymerization of low rank coals: a novel way to increase tar yield. RSC Adv. 2015, 5, 2493–2503. 10.1039/C4RA13029D. [DOI] [Google Scholar]
- Liu J.; Zhang Q.; Liang L.; Guan G.; Huang W. Catalytic depolymerization of coal char over iron-based catalyst: Potential method for producing high value-added chemicals. Fuel 2017, 210, 329–333. 10.1016/j.fuel.2017.08.074. [DOI] [Google Scholar]
- Kapo G.; Calvert S. Liquid phase oxidation of coal in alkali. Ind. Eng. Chem. Process Des. Dev. 1966, 5, 97–104. 10.1021/i260017a021. [DOI] [Google Scholar]
- Bimer J.; Sałbut P. D.; Berłożecki S. Effect of chemical pretreatment on coal solubilization by methanol-NaOH. Fuel 1993, 72, 1063–1068. 10.1016/0016-2361(93)90309-P. [DOI] [Google Scholar]
- Squir K. R.; Solomon P. R.; Carangelo R. M.; DiTaranto M. B. Tar evolution from coal and model polymers: 2. The effects of aromatic ring sizes and donatable hydrogens. Fuel 1986, 65, 833–843. 10.1016/0016-2361(86)90078-5. [DOI] [Google Scholar]
- Hayashi J.; Matsuo Y.; Kusakabe K.; Morooka S. Depolymerization of lower rank coals by low-temperature O2 oxidation. Energy Fuels 1997, 11, 227–235. 10.1021/ef960104o. [DOI] [Google Scholar]
- Wang H.; Dlugogorski B. Z.; Kennedy E. M. Coal oxidation at low temperatures: oxygen consumption, oxidation products, reaction mechanism and kinetic modelling. Prog. Energy Combust. Sci. 2003, 29, 487–513. 10.1016/S0360-1285(03)00042-X. [DOI] [Google Scholar]
- Ashida R.; Nakgawa K.; Oga M.; Nakagawa H.; Miura K. Fractionation of coal by use of high temperature solvent extraction technique and characterization of the fractions. Fuel 2008, 87, 576–582. 10.1016/j.fuel.2007.02.035. [DOI] [Google Scholar]
- Miura K.; Shimada M.; Mae K.; Huan Y. S. Extraction of coal below 350 °C in flowing non-polar solvent. Fuel 2001, 80, 1573–1582. 10.1016/S0016-2361(01)00036-9. [DOI] [Google Scholar]
- Miura K.; Mae K.; Shindo H.; Ashida R.; Ihara T. Extraction of low rank coals by coal derived oils at 350 °C for producing clean fuels. J. Chem. Eng. Jpn. 2003, 36, 742–750. 10.1252/jcej.36.742. [DOI] [Google Scholar]
- Miura K.; Nakagawa H.; Ashida R.; Ihara T. Production of clean fuels by solvent skimming of coal at around 350 °C. Fuel 2004, 83, 733–738. 10.1016/j.fuel.2003.09.019. [DOI] [Google Scholar]
- Larsen J. W.; Wei Y-C. Macromolecular chemistry of coalification. Molecular weight distribution of pyridine extracts. Energy Fuels 1988, 2, 344–350. 10.1021/ef00009a021. [DOI] [Google Scholar]
- Iino M.; Takanohashi T.; Ohsuga H.; Toda K. Extraction of coals with CS2-N-methyl-2-pyrrolidinone mixed solvent at room temperature: Effect of coal rank and synergism of the mixed solvent. Fuel 1988, 67, 1639–1647. 10.1016/0016-2361(88)90208-6. [DOI] [Google Scholar]
- Okuyama N.; Komatsu N.; Shigehisa T.; Kaneko T.; Tsuruya S. Hyper-coal process to produce the ash-free coal. Fuel Process. Technol. 2004, 85, 947–967. 10.1016/j.fuproc.2003.10.019. [DOI] [Google Scholar]
- Li X.; Zhu X.; Okuda K.; Zhang Z.; Ashida R.; Yao H. Preparation of carbon fibers from low-molecular-weight compounds obtained from low-rank coal and biomass by solvent extraction. New Carbon Mater. 2017, 32, 41–47. 10.1016/S1872-5805(17)60106-9. [DOI] [Google Scholar]
- Li X.; Ashida R.; Miura K. Preparation of high-grade carbonaceous materials having similar chemical and physical properties from various low-rank coals by degradative solvent extraction. Energy Fuels 2012, 26, 6897–6904. 10.1021/ef301364p. [DOI] [Google Scholar]
- Ashida R.; Takashima K.; Miura K.; Okuyama N.; Hamaguchi M.; Shishido T. Prediction of Coking Behavior of Coking Coals and Binder through Characterization Utilizing High Temperature Solvent Fractionation. Tetsu-to-Hagane 2014, 100, 127–133. 10.2355/tetsutohagane.100.127. [DOI] [Google Scholar]







