The effect of breakage top size and mode on coal liberation

Abstract

Coal from the Waterberg coalfield is seen as a potential replacement for the depleting Mpumalanga province coal resources. The Waterberg coalfield contains about 40-50% of South Africa's remaining coal reserves and the coal seams are highly intercalated with carbonaceous mudstones. Coal from this region is used to supply local power stations for electricity generation, semi-soft and metallurgical coal for other industries. However, this coal is associated with high ash and high volatile matter that necessitates the need for beneficiation to improve the coal grade. Comminution is a size reduction method that is used to liberate coal macerals from the gangue. Particles must be reduced to the optimal size that will improve the coal quality. Determining the optimal particle top size makes coal mining and processing economical for coal producers. Crushing to excessively small sizes may lead to the generation of fines that are difficult to handle and may require specialised equipment and wet beneficiation methods, rendering the process uneconomical.

To determine the optimal particle top size for liberation and/or for further retreatment, densimetric analysis techniques are employed. However, traditional densimetric analysis methods are time consuming, laborious and employ hazardous chemicals. Due to evolving technological minerals processing environment and the need for environment friendly and rapid sample processing technology, Debtech (now De Beers Group Ignite) designed and manufactured the RhoVol densimetric analyser. The RhoVol densimetric analyser is a 3D image processing densimetric apparatus that weighs and takes measurements of individual ore samples using a series of cameras placed at different angles to determine their volume, mass and densities. Studies have shown that the RhoVol densimetric analyser may be used for densimetric analysis without the need for float-and-sink analysis (Bothloko, 2023). However, the RhoVol densimetric analyser is unable to capture hidden cavities of porous ore particles and therefore tends to overestimate the volume of particles. As such, a density correction factor is required.

In this study, the effect of breakage mode and top size on coal liberation was investigated using the RhoVol densimetric analyser. Proximate analysis was done to determine the ash, moisture, fixed carbon and volatile matter content of the coal. The breakage modes investigated in this study are compressive and impact crushing. Compressive breakage was performed by applying a compressive force using a custom-built 80 kg rudimentary lawn roller over the coal particles. The coal particles were contained within a steel frame of different heights to obtain the desired top sizes. Impact breakage was performed using a drop weight impact rig. The coal particles were placed at the bottom of the drop weight impact rig with steel stopper rings at different sizes to obtain the required top size. The top sizes investigated in this study were 12 mm, 8 mm, 6 mm, 4 mm and 3 mm. The ideal top size to optimize the yield of low-density clean coal products (RD ˂ 1.4) was the focus of this investigation due to the increase in the demand for high-quality coal (grade A/export coal). A helium gas pycnometer was used to determine the true (skeletal) density of the particles to calibrate the RhoVol data. One of the advantages of using the RhoVol densimetric analyser is its ability to provide additional information such as the particle compactness, elongation and flatness. As such, the effect of the breakage mode and top size on particle shape was also investigated.

Proximate analysis of the 16 mm head sample indicates the low quality of the Waterberg coalfield coal, reflected by its high ash (60%) and mineral matter content. The high ash and mineral matter contents necessitate the need for beneficiation of the coal. The analysed data from the RhoVol and pycnometer was used to measure the washability properties of the coal at different top sizes. Results show that compressive breakage of the head sample to top size of 12 mm did not have any significant effect on the liberation of low RD clean coal products (RD˂ 1.4). However, impact breakage resulted in a small improvement of about 1.7% (by mass) in the proportion low-density coal. These particles are composed of vitrinite-rich coal macerals with little mineral matter. When the particles top sizes were further reduced from 12 to 8 mm, the proportion and of the low-density clean coal products (RD˂ 1.4) for both impact and compression breakage modes was significantly improved. About 3.2% (by mass) of clean coal products were obtained after compressive breakage to top size 8 mm as compared to the 6.4% (by mass) achieved through impact crushing. Further size reduction from 8 to 6 mm resulted in a small increase in the cumulative mass yield of low density (R ˂ 1.4) materials for both modes. Results show that between the RD intervals 1.2-1.4, there is a small difference in the mass yields of clean coal products (RD˂ 1.4) obtained through compressive and impact breakage respectively when the top size was reduced from 8 to 6 mm. Reducing the top size from 6 to 4 mm did not have any significant effect to the mass yield of low-density products (RD ˂ 1.4) after compressive breakage, but a small increase in the mass yield was observed for impact breakage. Crushing down to a top size of 3 mm improved coal liberation for both modes. After every particle top size reduction stage to smaller sizes, the proportion of middlings and high-density coal products decreased.

Coal washability curves show that compressive breakage to top size 12 mm had little effect on the mass yield of clean coal products as compared to the 5.7% low-density clean coal yield obtained after impact crushing. Compressive breakage to top size 8 mm significantly improved the mass yield of low-density clean coal products (at RD cutpoint 1.4) from 0.3% to 14.4%. Impact breakage to top size 8 mm significantly improved the yield of low-density clean coal products from 5.7% to 18%. Compressive breakage to top size 6 mm improved the mass yield of clean coal products to 21.6% as compared to 21.1% obtained after impact breakage. There were progressive but small improvements in the mass yield of clean coal products when the coal was crushed to top sizes 4 mm and 3 mm. Compressive and impact breakage to top size 3 mm resulted in the highest coal yield at 31.3% and 31.4% (at RD cutpoint 1.4) respectively. However, crushing to smaller top sizes (3 mm) can lead to excessive fines generation that can offset the benefits of the improved liberation.

Cumulative mass yield trends show that the yield of the middlings products between RD interval 1.4-1.8 decreases with the reduction in the particles top size. Compared to the head sample (16 mm), impact breakage to top size reduction to 3 mm decreased the middlings fraction from 60-45%. The mass yield of the middlings products was reduced from 60-50%. Compressive breakage of the 16 mm head sample to 3 mm decreased the mass yield of highdensity coal products decreased from 40-18% as compared to 40-20% yield obtained after impact breakage.

Particles top size reduction to smaller sizes liberated more of the lower density clean coal material and improved their mass yield. Crushing 8 mm, 6 mm, 4 mm and 3 mm yielded approximately 25-35% of the low-density clean coal products as compared to 5% attained after impact breakage to 12 mm. As particle top size decreases (6 mm, 4 mm, 3 mm), the difference between compressive and impact crushing is reduced, indicating that the crushing method has less influence on the yield of low-density clean coal yield at finer sizes. Assuming a density cutpoint of 1.4, impact breakage resulted in the highest coal mass yields at top sizes 12 mm, 8 mm and 4 mm. There are negligible differences in the mass yield of compressive and impact crushed coal at top size 6 mm and 3 mm. Significant low-density clean coal yields were reported when the coal was crushed to between 8 mm and 6 mm. To maximize the recovery of low-density clean coal without necessarily generating excessive fines, coal from the Waterberg coalfield should ideally be crushed to between 8- and 6-mm. However, the optimalsize would also rely on plant economics, costs related to fines management, and downstream processing. These findings highlight how important top size and breaking mode are to maximizing yield during coal beneficiation.

When the effect of breakage mode and top size on the particle shape was investigated, results show that head sample is mainly composed of compact, dense and round/spherical particles. Crushing the head sample to top size 12 mm via compressive and impact breakage resulted in a shift in the mass distribution of the particles from the most compact, rounder to midcompact particles. Further size reduction from 12 to 8 mm shows an increased shift in the mass distribution from rounder to irregular and flat particle shapes. The continued shift from spherical to irregular and flat particles was also observed when the top size was further reduced to 6 mm. The frequency of flat particle in the compactness range 0.42 - 0.51 increased from 6.8% to 8.1% after compressive breakage to 6 mm and from 7.5% to 9.1% in the case of impact crushing. Crushing to top size 4 and 3 mm further increased the frequency of particles from spherical to irregular and flat shapes. Crushing to smaller top sizes increased the frequency percentage of flat particles and reduced the amount spherical and irregular particles.

For both breaking modes, crushing to smaller top sizes improved the mass yields of lowdensity clean coal products. The proportion of spherical dense coal particles was reduced with a decrease in the particle top size. Coal liberation was improved when the coal was crushed to smaller, irregular and flat particles. Compared to compressive breakage, impact breakage was the most effective breakage mode and produced more flat, irregularly shaped particles. The findings from this study show that there is a correlation between the breakage mode, top size and particle shape regarding the coal yield. The above findings show that when properly calibrated with pycnometer data, the RhoVol densimetric analyser measurements can be used for rapid densimetric analysis on coal of varying sizes and shape.