Influence of Lignin Addition on Coke Strength

Abstract

In the steel industry, promoting the use of low-grade coal is important because of the depletion of coking coal. To compensate for the lack of caking property of low-quality coals, binders must be added to produce coke from coal blends. Lignin is an aromatic polymer and one of the main components, accounting for 30% of wood. The chemical structure and characteristics of lignin differ depending on the raw woody biomass and the extraction method used. In this study, to find biomass-derived binders, alkali lignin, sulfuric acid lignin, kraft lignin, organic solvent lignin, phenolated lignin (lignophenol), and torrefied lignophenol were blended at a ratio of 3 mass% with coal powder (coking coal: thermal coal = 1:1). The blended powder was carbonized to form coke, and its bulk density and strength were measured. With the addition of lignin, the bulk density of coke was almost the same or decreased, and the strength decreased significantly, regardless of the lignin type. Lignophenol was expected to be a potential binder because it was observed to be thermoplastic at the melting temperature of coking coal by an in situ camera; however, it largely reduced the fluidity of the coal and the coke strength. The lignin samples did not behave like asphalt pitch, and it was concluded that it was difficult to use lignin as a binder for blended coal.

1. Introduction

A significant amount of CO₂ is emitted during coke production and iron ore reduction, accounting for 51.6% of emissions from Japan’s manufacturing and construction industries and 12.7% of total emissions from Japan.¹⁾ The reduction of CO₂ emissions in the iron and steel industry is an important issue not only in Japan but also worldwide, and the reduction by biomass use from various perspectives has already been reviewed and discussed.^2,3,4,5) The addition of biomass to coking coal seems a promising method. There are reports that sawdust^{6,7,8,9,10,11)} and other biomass¹⁰⁾ have been blended into coking coal. However, biomass samples have been reported to reduce the fluidity^{6,7,8,9,10,11)} and strength of the resultant coke.^6,9,10,11) Rapid heating is thought to be effective in preventing fluidity loss,^10,11) but mixing more than 5% biomass seems difficult.¹⁰⁾ The main components of wood are cellulose, hemicellulose, and lignin. Cellulose and hemicellulose are polysaccharides, while lignin is an aromatic polymer formed by radical coupling of monolignols. Since cellulose and hemicellulose are oxygen-rich and decompose at lower temperatures than lignin, it would make sense to recover lignin from wood and blend it into coking coal. Unfortunately, there have been no reports of good coke production by adding lignin.^12,13,14) Most studies show that coking coal fluidity is significantly reduced by biomass blending, indicating the difficulty of decarbonization with this strategy.

The use of low-grade coal and the depletion of coking coal are critical issues in the steel industry. Coke and iron ore are fed alternately from the upper part of the blast furnace, and coke must be mechanically strong to withstand the load. Binders must be added to coal blends to compensate for the lack of caking property of low-quality coals. Coal tar and asphalt pitch have been used as additives for a long time, but in recent years research and development of a new additive called “Hypercoal”¹⁵⁾ has also been conducted. Therefore, in this study, we investigated whether lignin can act as a caking additive in coking and thermal coal mixtures. The chemical structure of lignin varies greatly depending on the tree species (softwood or hardwood) and extraction method. The difference of lignin molecular structure between softwood and hardwood is derived from the difference of monolignols. Softwood lignin is generally formed only from coniferyl alcohol, which has one methoxy group (-OCH₃) in aromatic nuclei. On the other hand, the formation of hardwood lignin uses not only coniferyl alcohol, but also sinapyl alcohol, which has two methoxy groups and thus has fewer sites for radical coupling. As a result, it is known that hardwood lignin has more ether linkages and fewer C–C linkages between monolignols than softwood lignin. Various types of lignin were collected, and coking coal and thermal coal at a ratio of 1:1 were coked by adding 3 mass% lignin. Tensile strength was measured to evaluate whether the obtained coke had sufficient strength.

2. Material and Method

2.1. Samples

Powdered coking coal (Goonyella coal) and thermal coal powder were used as coal samples. The particle size was adjusted to ≤ 250 μm, whose average was 180 μm. The industrial analysis data are listed in Table 1. The lignin samples that were used are listed in Table 2. Three lignin alkali samples were purchased from Nacalai Tesque (No. 1) and Sigma-Aldrich (No. 2–3). Softwood kraft lignin samples (BioPiva^TM, No. 6–8) were purchased from UPM Biochemicals, Finland. Organic solvent lignin (No. 9) was obtained from Guangzhou ESUN Biotechnology, China. Sulfuric acid lignin¹⁶⁾ (SAL, No. 4,5) was isolated from softwood or hardwood powders by 72 mass% sulfuric acid hydrolysis, followed by 3 mass% sulfuric acid hydrolysis at Mie University. Crude lignophenol (No. 10) was also isolated at Mie University from wood powder of softwood (western hemlock) by a unique reaction process:¹⁷⁾ p-cresol impregnation followed by 72% sulfuric acid hydrolysis. Torrefied lignophenol (No. 11) was obtained by heating crude lignophenol in a tubular furnace under nitrogen atmosphere. The temperature was increased to 300°C at a rate of 3°C/min and the temperature was maintained for 3 h. The recovery yield was 56.9%. Thermogravimetric analysis of each lignin sample (No. 1–10) was performed at a rate of 5°C/min from 30 to 100°C followed by 3°C/min from 100 to 700°C and the thermoplastic behavior of lignin under heating was observed in situ using a simultaneous differential thermogravimetric measurement device (Hitachi High-Tech Science, STA7200RV) equipped with a real-view function. To obtain thermogravimetric data for torrefied lignophenol (No. 11), crude lignophenol (No. 10) was torrefied in STA7200RV with the same temperature program as that performed in a tubular furnace, then cooled to 40°C and was heated to 600°C at a rate of 3°C/min. Thermogravimetric characteristics of lignin samples were summarized together in Fig. S1 (Supporting Information).

Table 1. Coking coal and thermal coal used in this study.

	Ash	VM	FC	C	H	N	S	O
	(d.b.%)	(d.b.%)	(d.b.%)	(d.a.f%)	(d.a.f%)	(d.a.f%)	(d.a.f%)	(d.a.f%)
Coking coal (Goonyella)	9.1	23.1	67.8	87.3	4.96	2.01	0.54	5.13
Thermal coal	9.5	14.1	76.4	91.5	3.86	1.48	0.47	2.71

d.b.%: dry base %, d.a.f%: dry ash free %

Table 2. Lignin samples used in this study. Weight loss at 300, 400 and 500°C was calculated relative to the weight at 100°C during thermogravimetric analysis at a rate of 3°C/min (Fig. S1).

No.	Name	Elemental analysis (d.b.%)			Weight loss (%)
		H	C	N	300°C	400°C	500°C
1	Lignin alkali (1), water-soluble	4.69	53.51	0.69	–11.1	–28.6	–34.6
2	Lignin alkali (2), water-soluble (Cat. No. 4710003)	5.05	51.32	0.66	–14.8	–30.7	–36.5
3	Lignin alkalii (3), water-insoluble (Cat. No. 370959)	5.68	63.45	1.00	–15.0	–38.5	–48.3
4	Softwood sulfuric acid lignin, SAL (1)	5.26	65.10	0.66	–6.1	–27.8	–42.5
5	Hardwood sulfuric acid lignin, SAL (2)	4.86	59.42	0.79	–11.5	–32.5	–43.6
6	Softwood kraft lignin, BioPiva100TM	5.46	64.97	0.63	–11.9	–36.4	–46.0
7	Softwood kraft lignin, BioPiva190TM	5.65	65.98	0.61	–14.1	–41.1	–51.5
8	Softwood kraft lignin, BioPiva199TM	5.71	65.82	0.62	–11.5	–35.2	–44.1
9	Hardwood organic solvent lignin	5.41	64.70	0.68	–9.8	–34.7	–45.0
10	Softwood crude lignophenol, LP	6.07	70.63	0.59	–18.3	–52.1	–60.0
11	Torrefied softwood lignoohenol	5.54	76.90	0.59	–0.3	–19.9	–32.8

2.2. Coke Formation from Blend of Coal and Lignin

Figure 1 shows the experimental setup. Lignin was added to coking coal and thermal coal at a ratio of 1:1, so that the weight ratio of lignin was approximately 3%. The blended coal-lignin powder was filled into a quartz test tube with a packing density of 0.8 g/cm³. A quartz rod (30.6 g) was inserted into the test tube to prevent contact with air. The test tube was placed in a small electric furnace, and the temperature was increased from room temperature to 1000°C at a rate of 3°C/min to carbonize the coal. After the maximum temperature was maintained for 30 min, the furnace was turned off and cooled to room temperature. The obtained coke was removed from the test tube and its dimensions and weights were measured to calculate the bulk density and tensile strength. The coke was subjected to a radial crushing test (indirect tensile test) using a strength tester (RTC-1325A, ORIENTEC). The tensile strength was obtained from the breaking load using the following formula:   

$$S_t=\frac{2P}{\pi lx}$$

S_t: Tensile strength (MPa), P: Breaking load (N), l: coke thickness (mm), x: coke diameter (mm).

Fig. 1.

Experimental apparatus to carbonize powder mixture of coking coal, thermal coal and lignin (3 mass%). (Online version in color.)

2.3. Measurement of Gieseler Fluidity

The Gieseler fluidity of a 1:1 mixture of coking coal and thermal coal was too low to be measured. Therefore, the 4:1 blend was measured using a Gieseler plastometer (1081-EXU-2, YOSHIDA SEISAKUSHO Co. Ltd.). Next, the change in fluidity by adding 1 mass% or 3 mass% lignin was observed.

3. Results and Discussion

The recovery method for three types of alkali lignin (No. 1–3) has not been disclosed by the producer. Sulfuric acid lignin (No. 4,5) can be quantitatively recovered as an acid-insoluble material by treatment of wood with 72%–3% sulfuric acid, while cellulose and hemicellulose are completely hydrolyzed to monosaccharides. It is formed by the condensation reaction between lignin molecules and is the most thermally stable with the highest molecular weight among the lignin listed in Table 2. Softwood kraft lignin powders (No. 6–8) were recovered and purified from the black liquor of an industrial kraft pulp mill using the LignoBoost process.¹⁸⁾ The difference between them (No. 6–8) was the water content. In the kraft pulping of wood, the most abundant beta-ether linkages in lignin are cleaved, depolymerized, and solubilized in alkaline cooking liquor. The recovery method for organic solvent lignin (No. 9) has not been disclosed; however, it is presumed that this lignin is obtained by the acetic acid organosolv process. Lignin is depolymerized in acetic acid via acidolysis. Lignophenol (No. 10) was isolated from p-cresol-impregnated wood powder at ambient temperature using 72% sulfuric acid. Lignin molecules were grafted onto p-cresol instead of intermolecular lignin condensation, preventing lignin polymerization. With the exception of lignophenol, lignin is insoluble in tetrahydrofuran, the most common solvent used for size-exclusion chromatography analysis of lignin; therefore, their molecular weight distributions are unknown. However, to the best of our knowledge, the molecular weight of lignin has been estimated as follows: sulfuric acid lignin (formed by intermolecular lignin condensation) >> lignophenol (formed without intermolecular condensation and ether bond cleavage) > kraft lignin/organic solvent lignin (by cleavage of ether bonds in lignin).

Lignophenol has been confirmed to melt by heat already at 250°C, and it was confirmed in situ that it is liquid even around 400–500°C, which is the plasticizing temperature range for coking coal (Fig. 2). Figure 3 show the bulk density and tensile strength of coke with 3% lignin. Error bars show the standard deviation. The density tended to be the same or lower than that of the base case (coking coal: thermal coal = 1:1) without lignin. However, the tensile strength decreased significantly with the addition of lignin, whereas the addition of asphalt pitch effectively improved the strength. Figure 4 shows the relationship between the bulk density and tensile strength of the coke. Although a weak positive correlation was observed, it was difficult to allow them to correspond to the type of lignin.

Fig. 2.

Observation of softwood crude lignophenol (No. 10) at temperatures from 300°C to 500°C with a real view camera. (Online version in color.)

Fig. 3.

(a) Density and (b) tensile strength of cokes made from the powder mixture (1:1) of coking coal and thermal coal with 3 mass% lignin addition shown in Table 2. Blended powder was carbonized in quartz test tube from room temperature to 1000°C with a heating rate of 3°C/min. The base case demonstrates the bulk density of cokes made from coking coal and thermal coal (1:1) without lignin. (Online version in color.)

Fig. 4.

Relationship between the bulk density and tensile strength of cokes made from the powder mixture (1:1) of coking coal and thermal coal with 3 mass% lignin addition. Black circle: cokes from blended coal and lignin, white circle: base case (coking coal: thermal coal = 1:1).

We expected lignophenol to be a potential binder because of the thermoplasticity hardly observed in lignin; however, it did not function like asphalt pitch. Figure 5 shows the results of the Gieseler fluidity tests. The fluidity of the 1:1 blend of coking coal and thermal coal was too low to be measured. Therefore, coking coal and thermal coal were mixed at a ratio of 4:1, and the effect of adding 1% or 3% of four lignins among 11 samples was investigated. It was confirmed that the maximum fluidity (MF) decreased with increasing lignin content. Of the four lignins tested, lignophenol decreased coal fluidity the most significantly. Lignophenol should have the effect of fusing coal particles, but it was speculated that plasticized lignophenol penetrates coal powder below the melting temperature range of coking coal. Thermoplastic lignin is expected to reduce the fluidity and inhibit fusion. Moreover, gas production from lignin pyrolysis (Table 2) can inhibit resolidification. Torrefaction of biomass and lignin before mixing with coal has been reported to be effective in removing volatiles.^11,13,14) It was also confirmed that torrefied lignophenol produced almost no volatiles (Table 2). When torrefied lignophenol was used, the large decrease in coal fluidity and tensile strength was suppressed, but never increased.

Fig. 5.

Effect of lignin addition on Gieseler maximum fluidity; 1 or 3 mass% of lignin was added to the blended coking coal and thermal coal (4:1).

4. Conclusions

Eleven types of lignin powder were blended into coal powder (coking coal: thermal coal = 1:1) and carbonized to form coke, the bulk density and strength of which were measured. With the addition of lignin, the bulk density of coke was almost the same or decreased, and the strength decreased significantly, regardless of the lignin type. Lignophenol was the only lignin found to be liquid at the fusing temperature of coking coal and was expected to be a good binder; however, it tended to decrease coal fluidity and coke strength. Unfortunately, it was difficult to use lignin to improve coke strength, similar to asphalt pitch. We have also started experiments to produce coal briquettes using lignin as a binder.¹⁹⁾

Supporting Information

Thermogravimetric properties of lignin samples are available on the Journal website at https://doi.org/10.2355/isijinternational.ISIJINT-2022-528.

Acknowledgments

This work was conducted by the Research Group of “Cokemaking technology for low CO₂ emission and high quality while extending available resources” in ISIJ.

References

• 1)  Ministry of the Environment, Japan: National Greenhouse Gas Inventory Report of JAPAN 2021, Figure 2-2, https://www.nies.go.jp/gio/archive/nir/jqjm1000000x4g42-att/NIR-JPN-2021-v3.0_GIOweb.pdf, (accessed 2022-11-23).
• 2)   E.  Mousa,  C.  Wang,  J.  Riesbeck and  M.  Larsson: Renew. Sustain. Energy Rev., 65 (2016), 1247. https://doi.org/10.1016/j.rser.2016.07.061
• 3)   H.  Suopajärvi,  A.  Kemppainen,  J.  Haapakangas and  T.  Fabritius: J. Clean. Prod., 148 (2017), 709. https://doi.org/10.1016/j.jclepro.2017.02.029
• 4)   H.  Suopajärvi,  K.  Umeki,  E.  Mousa,  A.  Hedayati,  H.  Romar,  A.  Kemppainen,  C.  Wang,  A.  Phounglamcheik,  S.  Tuomikoski,  N.  Norberg,  A.  Andefors,  M.  Öhman,  U.  Lassi and  T.  Fabritius: Appl. Energy, 213 (2018), 384. https://doi.org/10.1016/j.apenergy.2018.01.060
• 5)   H.  Mandova,  W. F.  Gale,  A.  Williams,  A. L.  Heyes,  P.  Hodgson and  K. H.  Miah: Sustain. Energy Technol. Assess., 27 (2018), 23. https://doi.org/10.1016/j.seta.2018.03.001
• 6)   M. A.  Diez,  R.  Alvarez and  M.  Fernández: Fuel, 96 (2012), 306. https://doi.org/10.1016/j.fuel.2011.12.065
• 7)   M. A.  Diez and  A. G.  Borrego: Fuel, 113 (2013), 59. https://doi.org/10.1016/j.fuel.2013.05.056
• 8)   M. G.  Montiano,  C.  Barriocanal and  R.  Alvarez: Fuel, 106 (2013), 537. https://doi.org/10.1016/j.fuel.2012.10.062
• 9)   M. G.  Montiano,  E.  Díaz-Faes,  C.  Barriocanal and  R.  Alvarez: Fuel, 116 (2014), 175. https://doi.org/10.1016/j.fuel.2013.07.070
• 10)   M.  Castro-Díaz,  H.  Zhao,  S.  Kokonya,  A.  Dufour and  C. E.  Snape: Energy Fuels, 26 (2012), 1767. https://doi.org/10.1021/ef2018463
• 11)   S.  Kokonya,  M.  Castro-Díaz,  C.  Barriocanal and  C. E.  Snape: Biomass Bioenergy, 56 (2013), 295. https://doi.org/10.1016/j.biombioe.2013.05.026
• 12)   H.  Suopajärvi,  E.  Dahl,  A.  Kemppainen,  S.  Gornostayev,  A.  Koskela and  T.  Fabritius: Energies, 10 (2017), 1850. https://doi.org/10.3390/en10111850
• 13)   M.  Castro-Díaz,  M. F.  Vega,  E.  Díaz-Faes,  C.  Barriocanal,  U.  Musa and  C.  Snape: Fuel, 258 (2019), 116125. https://doi.org/10.1016/j.fuel.2019.116125
• 14)   A.  Koskela,  H.  Suopajärvi,  O.  Mattila,  J.  Uusitalo and  T.  Fabritius: Energies, 12 (2019), 1533. https://doi.org/10.3390/en12081533
• 15)   T.  Takanohashi,  T.  Shishido and  I.  Saito: J. Jpn. Inst. Energy, 90 (2011), 853 (in Japanese). https://doi.org/10.3775/jie.90.853
• 16)   S.  Yasuda,  K.  Fukushima and  A.  Kakehi: J. Wood Sci., 47 (2001), 69. https://doi.org/10.1007/BF00776648
• 17)   M.  Funaoka: Polym. Int., 47 (1998), 277. https://doi.org/10.1002/(SICI)1097-0126(199811)47:3<277::AID-PI10>3.0.CO;2-P
• 18)   P.  Tomani: Cellul. Chem. Technol., 44 (2010), 53. https://www.cellulosechemtechnol.ro/pdf/CCT44,1-3%20(2010)/P.53-58.pdf, (accessed 2023-05-02).
• 19)   H.  Nonaka,  T.  Tomono,  Y.  Nagashima and  M.  Kitao: ISIJ Int., 63 (2023), 1539. https://doi.org/10.2355/isijinternational.ISIJINT-2023-046