Blast furnace coke substitutes from Victorian brown coal

Iron is usually produced from its ores using coke in a blast furnace (BF). Coke, a hard and macroporous carbon material, is produced from special coals (coking coals) and acts as fuel, smelting agent, and the permeable support for the charge to the BF. No material can completely replace coke in a BF. Coking coals are becoming harder (and more expensive) to obtain. Victorian brown coal (VBC) is accessible, cheap, with low mineral concentrations, which is favourable for iron production in a BF. However, as-mined, it does not form coke, but a char which is too reactive to be used in a BF. The objective of this project is to produce a substitute for BF coke from VBC by physical and chemical treatments and to investigate the use of cementing agents to reduce the reactivity and strengthen the product finally formed. VBC from Loy Yang open cut, and its commercial products, briquettes and char, which were obtained from Australian Char Pty Ltd in lump form, were used as starting materials. VBC tar, coking coal tar pitch, and asphaltene (hexane insolubles from VBC tar) were used as binders. Some of the starting material was pre-treated by acid washing (0.5 M H₂SO₄), hydrothermal dewatering (HTD; 320 °C-35 min) or alkali treatment (KOH (aq), 185 °C-10 h). The elemental analysis and NMR of these materials were determined. Before pelleting, raw VBC, pre-treated VBC, or briquettes were dried at 105 °C under N₂, ground to <0.15 mm, then mixed with the binder in tetrahydrofuran (THF). THF was removed and the mixture was pelleted by a conventional hydraulic press at ambient temperature or using an INSTRON 5569 series Mechanical Tester applying a range of forces, temperatures and times. Some samples were pelleted under N₂ (350 °C-30 min) by “Hot Press Carbonization” (HPC). In some cases, samples were air cured at 200 °C for 2 h. Finally, the samples were carbonised at a range of temperatures and times under N₂ flow, at a low heating rate to minimise cracking of the pellet, then cooled under N₂. The measurements used to evaluate the suitability of the products as substitutes for BF coke were compressive strength and reactivity. The compressive strengths of pellets were measured by using an INSTRON 5569 series Mechanical Tester. Reactivity was measured using a thermogravimetric balance. The sample was heated to 1000 °C at 20 °C /min under N2 and held at 1000 °C for 1 h in a flow (70 ml/min) of 1:1 CO₂:N₂. The reactivity, R60CO₂, was calculated from the weight loss. Physical properties of the products were measured in order to understand what factors controlled the compressive strength and reactivity. Initially, VBC or commercial briquettes were impregnated with tar, pelleted at ambient temperature and carbonized. Products from VBC showed higher compressive strengths (40-60 MPa) and slightly lower reactivity (R60CO₂ 87-89 %) and surface areas (790-800 m²/g) than those from briquettes. The effects of carbonization time, temperature (900 or 950 °C) and tar addition were relatively small. The high reactivity of the samples compared to that of coke (R60CO₂ 13 %) is probably related to their higher surface areas and the smaller extent and greater disorder of their graphitic structure as shown by XRD. The poor results of ambient pelleting and recent literature suggested that hot pelleting of VBC would be advantageous. Therefore, VBC-tar mixture was hot pelleted (150 °C-20 kN for 10 or 30 min), optionally air cured then carbonized (950 °C for 3 h). Products showed higher compressive strength (90-200 MPa) and bulk density (1.17-1.27 g/cm³) than those obtained following ambient pelleting. A high concentration of tar (10-15 wt%) and air curing increased the compressive strength by a further factor of two. The compressive strength was higher than that of a BF coke (20 MPa), but the surface area remained high and the surface was rough (SEM) and the proportion of graphitic structure was small (Raman spectroscopy). These factors probably contributed to the high reactivity of even the strongest products. VBC treated by HTD resembles a higher rank coal (e.g. lower O content), suggesting that HTD coal might carbonize to a less reactive product, like a higher rank coal. HTD treatment reduced the reactivity of the carbonization products, without an unacceptable lowering of the compressive strength. More severe briquetting conditions, acid washing before HTD, air curing and severe carbonization conditions (1200 °C-8 h) all together reduced the reactivity to R60CO₂ 34 %, still much higher than that of a BF coke. The surface area was reduced, but only to 100 m²/g, (cf. 18 m2/g for BF coke) and the proportion of graphitic structure was smaller than in BF coke, so that the higher reactivity may be due to these structural factors. Alkali treated VBC (ATC) appears to melt and fuse upon carbonization, like a coking coal, suggesting that carbonised product might be similar to a BF coke. The ATC with pitch and air curing had a high compressive strength (up to 230 MPa) after carbonization (1200-1300 °C for 2-8 h). The small surface area (as low as 20 m³/g) and smooth surface (SEM) of the products under some conditions suggests that fusion occurred during carbonization. However, the proportion of graphitic structure (Raman and TEM) was lower than for a BF coke and the reactivity of the carbonized products did not fall below R60CO₂ 30 %. Possibly the alkali treatment changed the chemical structure and inhibited graphitisation. Suitable pore structure is necessary for low reactivity, but the chemical structure is also important. Empirical treatments, modifying the structure of brown coal in the direction of higher rank coals, give carbonised products which approach BF coke in reactivity, surface area and the proportion of graphitic structure while maintaining compressive strength.