Pittsburgh Mining Research Division (PMRD) researchers conducted studies within the BEM to examine the effect of variations in rock dust particle size and their ability to provide an inerting balance between the aerodynamically deposited float coal dust and rock dust applied using pneumatic rock dusters.

The main entry of the BEM was used to perform controlled dust deposition experiments. The single entry is nominally 2.74 m wide by 2.2 m high and 396 m long. A variable-speed fan provided the ventilating air for the dust dispersion studies.

Fifteen dust deposition experiments were conducted in the BEM. Seven experiments were conducted with three rock dusts (RD 1, RD 2, and RD 3) and at two airflows—an average high airflow of about 1.5 m/s and an average low of about 0.6 m/s. Eight experiments were conducted with either 10% or 20% of coal dust mixed with the previously mentioned three rock dusts and an additional rock dust (RD 4). The mixtures were also conducted at high and low ventilation flow rates (Table 1). RD 1 and RD 2 are commercially available and used in coal mines. RD 3 is an anti-caking rock dust developed under a NIOSH BAA agreement (Imerys 2014), and RD 4 consisted of a commercial rock dust having only ~7% of <10 µm particles. The coal dust used in these tests is from the same coal seam and has similar particle size distributions as those used in large-scale inerting studies conducted at the Lake Lynn Experimental Mine (LLEM) (NIOSH 2010).

The differential volume percent and cumulative volume percent size distributions of the four commercial rock dusts and the Pittsburgh seam pulverized coal dust used in these experiments are shown in Figure 1. The particle size distributions of the dispersed coal dust, rock dust, and the deposited mixtures were determined using a Beckman-Coulter LS 13 320 Laser Diffraction Particle Size Analyzer equipped with a Tornado Dry Powder System (B-C). The refractive index (RI) used for all rock dust samples is 1.68+0.0i (i.e., no absorptive component) as it is the value listed in the B-C manual for calcium carbonate. The RI used for all coal dust samples has a significant imaginary component (it is strongly absorbing) and is 1.80+0.3i. The particles were treated as equivalent spherical light scatterers (Mie scattering). Note that all rock dusts, except RD 4, exhibit bimodal distributions.

To minimize the potential formation of dispersed clouds of explosible mixtures in the BEM’s single entry during testing, coal dust and rock dust mixtures of 80% and 90% by mass rock dust were thoroughly mixed by tumbling in 32-gal drums before loading the rock duster for test dispersion. A total of 183 kg of each mixture was dispersed for each test except for tests 14 and 15, where 91 kg were dispersed.

The mid-stream average air velocity for each experiment was measured at all sampling stations using a Davis vane anemometer, and the average volumetric airflow was calculated based upon the entry dimensions. The dust was distributed using an A.L. Lee EL5–200 battery-powered rock duster with the air-dust discharge through a 5-cm diameter nozzle attached to the roof (Figure 2).

Dust was dispersed in two 91-kg) batches, each batch requiring about 20–40 min to disperse. The rock dust discharge rate was not constant. It varied with the size distribution of the rock dust, the depth of dust in the hopper, and dust hang-up on the walls of the hopper. Settled dust was collected on metal dust collection trays measuring 23 cm wide by 33 cm long, and approximately ~800 cm² in area. These were placed on the experimental mine concrete floor, rib shelves, and center post shelves downwind of the rock duster.

Settled dust was also collected on 1.2-m x 1.8-m-1.9-cm thick plywood that was covered with waxed butcher block paper. Samples collected from the tailgate entry of a participating mine were collected by brushing off predetermined marked sections of the ground support cribs.

The experimental procedure used at an active longwall mine was different from that used in the BEM. The longwall operation used pressurized air to entrain and push rock dust through a flexible duct along the ground support shields starting at the headgate and ending at the tailgate, which is a distance of ~305 m in length. At the tailgate, rock dust was injected through a 5.1-cm diameter pipe positioned at mid-entry height near the last shield. Coal dust produced by the shearer was carried along the ventilation path towards the injection pipe. The movement of ground support shields also produced dust that remained airborne, mixing with the rock dust near the last shield before being carried with the ventilating air into the return airway. Before the longwall startup at the beginning of the shift, ground support cribs adjacent to the coal pillar were wiped clean, and metal trays were placed mid-entry on the floor and spaced 6 m apart starting at the longwall panel. After three days, when the longwall shut down for routine system checks and repairs, dust samples were collected. Using a soft bristled brush, the support crib dust samples were gathered at distances 3, 12, and 24 m from the tailgate. Other dust samples were collected from the remaining 8 metal trays that were not lost to the gob as the coal face advanced.

After the dust cleared the entry, the dust samples were collected from each tray or crib surface, sealed in an air-tight plastic bag, and brought to the surface for analysis. After drying, the samples were weighed using a Mettler balance, to the nearest 0.01g, and then underwent a particle size analysis using the B-C. The rock dust content of the deposited coal dust and rock dust mixtures was determined using the low-temperature ashing (LTA) method which burns off the coal component at 515°C (Montgomery 2005, NIOSH 2010).

Shown in Figure 3, the experimental results from Test 9 serve as a representative example of the measured floor mass deposition (g/m²) and mean particle diameter D_m variations with distance downwind of the discharge point. The results indicate that the mass deposited downwind follows a logarithmic decay from x =18.3 to 152 m proportional to ~1/x², while the decay mean diameter is proportional to ~1/√x. The decay of mass deposition and mean diameter is particularly consistent for Tests 1–13 as listed in Table 1, which shows the respective correlations for the 15 experiments conducted in the BEM.

The differential and cumulative size distributions for RD 1 and for floor samples collected afterwards at various distances downwind are shown in Figure 4 for Test 2. This data shows that 38% of particles by mass >74 µm in size dropped out of the air within 18.3 m of the rock duster. This left only about 62% of the released mass of rock dust to inert coal dust at greater distances.

Dispersing RD 2 under the same ventilation conditions (Test 4) yielded the data shown in Figure 5a and Figure 5b. The cumulative size distributions of Test 4 with RD 2 (Figure 5b) are similar to the original RD 2 except that only 4% of particles >74 µm settle out within 18.3 m. Consequently, 96% of the particle mass <74 µm is able to travel downwind to inert coal dust.

Variations in rock dust particle size distributions and mass percentages <74 µm, such as those shown in Figure 1, were expected to produce different inerting effects downwind of the trickle duster. To help illustrate the extent of inerting differences, eight experiments with coal and rock dust mixtures (Table 1) were conducted in the BEM. The 80/20 (RD 1/CD) mixture and the 80/20 (RD 2/CD) mixture were dispersed in separate experiments (Tests 1, 3, 5, and 7) at the two different air velocities listed in Table 1. The deposited samples were collected and analyzed.

Shown in Figures 6 and 7 are the particle size distributions of the coal and rock dust mixtures and of the samples collected from the floor at various distances from the rock duster for Test 5 and Test 7, respectively. A mixture of 80% RD1 and 20% CD was dispersed in Test 5. Approximately 33% of RD 1 is <74 µm in size. As the mixture is discharged from the rock duster, the larger coal and rock dust particles in the range of 74 µm to 250 µm settle on the floor within 5 m to 9 m from the source. Due to the varying particle sizes and densities, the rock dust and coal dust particles settle out of the airstream at differing rates and times. The resulting effects on the total incombustible content are seen in Figure 8. The total incombustible content increases to 97% within 9 m of release. This is substantially higher than the 80% rock dust concentration originally dispersed. Between 9 and 15 m, however, the concentration of rock dust rapidly decreases to 70%, which is less than the 80% incombustible content required in 30 CFR 75.403. At 60 m, the percentage of rock dust returns to the non-explosive range and remains ≥80% for the remainder of the samples.

The rock dust used in Test 7, RD 2, contains only 2% >74 µm particles. The 84.5% incombustible content of the floor samples at 9 m is similar to the concentration in the initial sample (80%). The concentration of rock dust drops to 76% at approximately 12 m but then rapidly returns to 80% at 18 m. The concentration continues to increase, building to about 85% at 40 m, and remains >85% for the remainder of the samples.

These results indicate that a rock dust containing ~33% of >74 µm particles left an area >50 m inadequately rock dusted. This could have a significant impact on the size and variation of explosive zones. Rock dusts with smaller percentages of particles >74 µm produced improved rock dust coverage downwind of application.

Air velocity also has an effect on the location and extent of inert and explosive zones downwind of the rock duster (Figure 9). Comparing downwind inerting results between Test 3 using RD 2 at 0.6 m/s with Test 7 using RD 2 at 1.63 m/s, the zones with <80% rock dust increase from essentially none at the low airflow to 10 m (7 m to 17 m from the rock duster) with the higher airflow. The explosive zone when using RD 1 is even greater, extending from ~ 25 m to 50 m in length with increasing airflows.

Additional Tests 12, 13, 14, and 15 consisted of mixtures created with 90% rock dust uniformly mixed with 10% CD. These tests were conducted to explore the extent to which the combined effect of increasing air velocity and the relative amount of rock dust particles <10 µm would have on the dust mixtures’ abilities to inert over longer distances. Two rock dusts, RD 3 (~42% <10 µm) and RD 4 (~6% <10 µm), were used in these experiments (Figure 10). The rock dust percentages in Tests 12 and 13 continued to increase with distance and decrease with increasing velocity over the same space. These trends are similar to those observed for Tests 1 and 5 where the rock dust contained ~32% <10 µm. Since RD 4 contains only ~6% <10 µm particles, the rock dust percentages measured in Tests 14 and 15 continued to decrease with increasing distance and ventilation velocity.

The PMRD researchers, as part of another study examining mine dust sampling methods, had the opportunity to collect dust samples from ground supports and trays placed in a tailgate entry of a longwall panel. Although these experimental data are limited, the results are presented here for completeness.

At the crib locations ~3 m, ~12 m, and ~24 m inby the tailgate corner, the measured mean diameters were 36, 33, and 7 µm, respectively. Figure 11 shows the changes in particle size distribution. These results cannot be directly compared to the BEM deposition studies since the size distribution of the combined coal dust and rock dust leaving the tailgate changes with the advancement of the face and the movement of the shearer across the ~305-m longwall panel. When the longwall dust samples were collected, the air split at the tailgate entry with more air flowing outby along the coal block and less air flowing towards the gob. Within 3 m of the face, 90% of the coal and rock dust particles collected from cribs were less than 74 µm. At 24 m, 98% of the collected coal and rock dust particles were less than 30 µm. Incombustible results indicate that the total incombustible content increased from 72% at 3 m from the face to 82% at 24 m.