Design and Characterization of Canopy Air Curtain for Protecting Against Diesel Particulate Matter Exposures in Underground Mines

Blasters are one of the highest exposed work groups to diesel particulate matter (DPM) in underground metal/nonmetal mining. These workers can spend a good portion of their day under a canopy in a basket loading blastholes with explosives. Therefore, one way of potentially reducing their exposures to DPM is to place a canopy air curtain (CAC) on the basket of the ANFO loader. In the original design of the CAC on a roof bolting machine and shuttle car, a fan draws in air through a lter to capture the dust and then supplies clean air beneath the canopy where a miner is working. This paper describes the testing of a CAC that was redesigned to t an ANFO loader and prevent exposures to DPM as well as respirable dust. Laboratory measurements demonstrated reductions of submicron particles that relate to the percent reductions of DPM. The CAC provided substantial protection of mine workers to DPM (80% reductions), from within 15.24 cm (6 inches) of the edge of the CAC using a 7.62-cm (3-inch) lip. As the mine worker approaches the edges of the CAC, the percent reduction starts to reduce to the 30–50% range. The mine worker achieves the best results when within 15.24 cm (6 inches) from the edge of the CAC. In addition, the CAC should be located such that the breathing zone of the mine worker is 25.4–50.8 cm (10–20 inches) below the CAC.


Introduction
Exposure to diesel exhaust has been linked to various adverse health outcomes [1,2]. In 2012, the International Agency for Research on Cancer (IARC) categorized diesel engine exhaust as a carcinogen to humans (Group 1) [3]. Exposure to diesel particulate matter (DPM) is especially concerning for underground miners. Since underground miners work alongside diesel equipment in a con ned environment, they can be exposed to some of the highest levels of diesel exhaust in the U.S [4][5][6][7]. Therefore, the Mine Safety and Health Administration (MSHA) promulgated a rule to limit exposures of metal/nonmetal underground miners to DPM to an eight-hour time-weighted average (TWA) of 160 µg/m 3 total carbon (TC) [6,7]. This compliance limit is based upon feasibility of compliance with the goal being to reduce a carcinogen, in this case DPM, to the lowest levels possible.
Since this rule went into effect, DPM exposures in mining have decreased, but they are still above levels of other occupations in other industries, such as agriculture, construction, manufacturing, etc., and in some cases above the personal exposure limit (PEL) for mining [5,8]. A review of MSHA compliance data between [2009][2010][2011][2012][2013][2014] shows that approximately 20-25% of underground miners are out of compliance each year. Further analysis shows that 46% of the out-of-compliance samples are from blasters, load haul dump drivers, and scalers, with blasters being one of the highest exposed professions in mining. Workers in these professions usually work outside of cabs in areas where diesel equipment is operating. The average DPM exposures for blasters are about 156 µg/m 3 TC, with exposures at times above 500 µg/m 3 TC. Mines can have a di cult time ventilating the areas due to the large energy requirements to move air through the large underground openings where these miners commonly work. Some mines use administrative controls to avoid miners' exposure to DPM by having them work on off-schedules or upstream of diesel vehicles [9]. However, these types of solutions are not always feasible or practical.
Blasters can spend a substantial amount of their time at the face loading explosives into blastholes. The face can be quite large, requiring the use of a high lift. The ANFO loader is a machine that incorporates a high lift where a canopy is usually located above the mine worker. Therefore, one potential control technology to help reduce their exposures may be using the canopy air curtain (CAC) attached to the canopy of the ANFO loader.
As demonstrated in Fig. 1, the CAC delivers clean air over the operator's breathing zone. A fan draws in air through a lter to capture the dust and then supplies clean air beneath the canopy where a miner is working. The development of the canopy air curtain (CAC) dates back to the 1970s, starting with the initial development of the CAC by the Donaldson Company, Inc. under contract from the U.S. Bureau of Mines [10]. This CAC was originally developed for continuous miner operators when continuous mining machines had cabs. The need for a CAC on the continuous miner was eliminated when the cab was removed from the machine design. However, CAC development progressed to include designs for a roof bolting machine to protect roof bolters from respirable coal mine dust [11][12][13]. The original design of the CAC by the National Institute for Occupational Safety and Health (NIOSH) provided even air ow across the entire canopy and results from laboratory testing demonstrated reductions of respirable dust concentrations ranging from 67-75% in ventilation air ow ranging from 10-120 fpm [11].
Fletcher Mining Co. incorporated the CAC into their roof bolting machines but revised the design to have the air ow just around the perimeter of the canopy. Laboratory testing showed that this design only reduced the respirable dust concentrations by 17-24% [13]. In order to increase its effectiveness, Fletcher re-designed the air canopy based upon recommendations from NIOSH to provide air ow over the entire canopy, using a staggered perimeter air ow design with a higher ow rate that prevented contaminated air from entering the CAC protection zone. Laboratory testing demonstrated that this change increased the reduction from 17-24% to approximately 50% [14]. Field testing the roof bolter CAC in underground coal mines with this design showed respirable dust reductions ranging from 3-60% [15]. The variations in respirable dust reduction were due to differing ventilation air ows, differing heights the canopy was above the operator, differing times the operator actually worked underneath the CAC, and the low respirable dust concentrations encountered during the study. The next step was to redesign the CAC for shuttle cars. This CAC is square in shape and utilizes uniform air ow across the canopy. Laboratory testing at NIOSH demonstrated reductions in respirable dust with different ventilation rates ranging from 74-83% at a ventilation rate of 0.61 m/s (120 fpm), 39-43% at a ventilation rate of 2.0 m/s (400 fpm), and 6-16% at a ventilation rate of 4.3 m/s (850 fpm). The higher the interference ventilation air ow, the lower the percent reduction in respirable dust [16].

Design of Diesel CAC for ANFO Loader
Since the CAC is successful for reducing respirable dust exposures, it was thought that it could be used to prevent DPM overexposures. The basic design of the shuttle car CAC was modi ed by researchers to use on an ANFO loader. However, there are some extra challenges when trying to reduce DPM instead of dust. The particles of DPM are smaller (submicron and nanometer) than dust particles (greater than 1 micron) and act more like a gas than particles. Therefore, one of the rst adjustments is for the ltration system to capture submicron particles. A MERV 13 lter is currently used in the CAC for shuttle cars, but this lter is only designed to capture 50-75% of submicron particles. This capture e ciency is too low. Therefore, the initial tests were performed with a higher-rated lter: MERV 16. A MERV 16 rated lter was selected based upon past experience [17][18][19]. Higher-rated lter media such a HEPA grade can increase backpressure resulting in decreased air ow and induce leaks around the lter, thereby reducing the amount of protection from DPM to the miner. The MERV 16 rated lter has the ability to capture DPM particles at a high e ciency, while still allowing the required air ows to prevent contaminated air from entering the miner's breathing zone. MERV 16 lters are designed to reduce submicron and nanometer particles by over 95% and have been shown to be effective for use in enclosed cabs [17][18][19].
Since blasters are one of the highest DPM-exposed working groups in mining, the CAC was designed to lter DPM and t onto an ANFO loader. A diesel CAC was developed using the plenum design based on the one developed for the shuttle car CAC as seen in Fig. 2. This CAC was rescaled to t under the canopy of an ANFO loader. This was accomplished by fabricating two 0.91-m by 0.91-m (3-ft by 3-ft) plenums, each with a series of openings at uniform spacing. Each plenum contains its own blower (rated at 1,800 cfm) and operates independently. Both plenums would be attached side by side on the canopy of the ANFO loader to cover the whole area of the ANFO platform/basket. The diesel CAC was designed to produce air ow similar to the shuttle car CAC (1.16 m/s at the borehole). Besides the size, there were a few other modi cations from the shuttle car CAC. The shuttle car CAC was fabricated from metal, but the diesel CAC was made from a durable ame-retardant plastic for lighter weight. As mentioned earlier, the ltration was another difference in the diesel CAC, for a MERV 16 lter was used instead of the MERV 13.
The blower passes air through a V-bank lter setup with a MERV 16 lter and 15.24 cm (6 in) tubing into the CAC.

Evaluation Methods
The ANFO diesel CAC was evaluated in the lab at NIOSH Pittsburgh. The evaluation included the measurement of air ow and particle counts underneath the CAC. For measuring the air ow, this analysis included measuring the air ow across the plenum including at three different heights (25.4, 50.8, and 76.2 cm) below the CAC to determine how rapidly the air ow may decrease. Reed et al. 2019 [14] demonstrated that the air ow and reductions in dust concentrations will start to decrease as one is farther from the CAC. Therefore, three distances below the CAC were measured to determine the location of the CAC in reference to the mine worker which provides protection from DPM. Then, to determine potentially how well the CAC could reduce DPM exposures, the number of particles measured outside of the CAC and the number measured under the CAC were recorded. These measurements were completed at the same distances underneath the CAC used in air ow measurements. These numbers were compared to determine the percent reduction in particles from 0.3-1 µm (size range measured by the ARTI/Met One HHPC-6 particle counters) under the CAC. DPM particles are within this size range; therefore, by measuring the percent reductions in these submicron particles, one may obtain a good idea of the reductions in DPM. Measuring particles in the air at this size range has been used in other studies to estimate the potential reductions in DPM [20,21]. For example, Organiscak et al. [20] used the same instrumentation to measure submicron particles to determine lter e ciencies for enclosed cabs, and Potts and Divers [21] measured submicron particles to evaluate the e ciency of the airstream helmet for reducing DPM concentrations.

Measuring Airstream Flow
To determine the air velocity pro le of the diesel CAC, a test stand was built to support the CAC as seen in 24-cm (6-inch) squares numbering 6 to a side. Each square was uniquely identi ed by letters from front to back from C to H (Y-axis) and then from left to right by numbers ranging from 3 to 8 (X-axis). The velocity using a TSI VelociCalc was measured for one minute and averaged in the center of each square, and triplicate measurements were collected at each square. Then, the grid was moved to be 50.8 cm (20 in) below the canopy, and the velocity measurements were repeated three times.
The measurements were completed again at 76.2 cm (30 in) below the canopy.
The CAC was divided into sections as shown in Fig. 4 with each color representing a quadrant. Since it would be expected for the air ow and protection to be different around the edge compared to the center of the CAC, the edge was divided into six sections and the center was divided into four quadrants. Each section contained three to four 15.24-cm (6-in) squares with all of the central quadrants containing four squares. Measurements for each quadrant were averaged, and then a 95% con dence limit was calculated using the equation below (Eq. 1) [22].

CI-con dence interval, t -t value for degrees of freedom, s -standard deviation, N -number of samples
During the analysis of the initial velocity measurements, the velocity pro le of the initial designed CAC demonstrated dead spots or areas with no air ow, i.e. nonuniform air ow. It is important that the plenum provide uniform air ow across the protection zone. To accomplish this, a honeycomb screen is located inside the plenum across the inlet into the plenum area. The air passes through the honeycomb system into the plenum to minimize turbulence. During investigation of the cause of nonuniform ow, the plenum was opened to examine the interior (Fig. 5). The honeycomb section across the inlet had a ne mesh screen covering the beginning and ending portions of the honeycomb. Most likely, the screen meshes were installed by the plenum builder in an attempt to force air to the center of the 91.44-cm x 5.08-cm (3ft by 2-in) inlet. These screens prevented uniform ow as they provided more resistance at the locations they were installed. NIOSH had recommended the honeycomb design at the inlet as it had been determined that uniform air ow would be provided by the wall behind the inlet, which angles from the 15.24-cm (6-in) inlet opening to almost ush with the other side, as shown in past research [23]. The mesh screens were removed, and the CAC was tested again for velocity and particle counting. All results show the air ows with the mesh screens removed.

Particle Count Measurements
In addition to velocity measurements, in the center of each grid square under the canopy, the number of submicron particles (size range of DPM) was collected to determine potential reductions in DPM. Two ARTI/Met One HHPC-6 particle counters (Hach Ultra Analytics, Grants Pass, OR) were used to simultaneously sample and record the number of particles under the CAC in the center of a square and outside the CAC particle size concentrations for one-minute periods [19]. These instruments count airborne particles in channels from 0.3 to 1.0 µm. The test medium was airborne particles present in the ambient air. The instruments used for measuring under and outside of the CAC were then alternated for another test to average out any instrument sampling biases for each test. Triplicate samples were performed in the center of each square. The number of particles from 0.3 to 1 µm were summed, and the particles under and outside of the CAC were used to determine the amount of reductions in particles. The tests for each section were averaged, and as with the ow measurements, the 95% con dence limits were calculated as described in the previous section.
During the analysis of the particle measurements, it was found that the air ow could be optimized to provide more clean air around the edges. This could be achieved by attaching a brattice cloth (thick tarp material) around the edge of the CAC where a 7.62-cm (3-in) lip would extend below the bottom of the CAC. This prevented interference ventilation air ows from impacting CAC performance and it also directed the plenum air ow downward, improving reductions of submicron particles. Again, the velocity and particle counting measurements were performed. In addition, all measurements were collected for each CAC (right and left -described below).
In order to simplify and obtain another view of the effects of adding the lip and distance from the bottom of the CAC, the percent reductions in the center quadrants was averaged and the 95% con dence limit was calculated as described above.

Air ows and Reductions of Submicron Particles When 10 inches Below ANFO CAC
To provide a perspective for the air ow measurements under the CAC, Fig. 6 shows the orientation of the CAC with respect to the face. Both plenums were evaluated. The front location would be the front of the basket in an ANFO loader if the CAC was attached to the vehicle. The right and left would be from the perspective of an individual facing the ANFO loader. This orientation is used for both velocities and particle reductions.
When reviewing the particle counting data in Fig. 9, the particle reductions in the four central quadrants ranged from 55-78%. The particle reductions in the perimeter sections ranged from 1-59%, which were generally lower than the center. It can be seen that the particle reductions correspond with the air velocities emanating from the plenum. The protection of the miner from DPM will depend upon where the mine worker will be positioned. The position of the CAC should be designed to so that the mine worker will be 15.24 cm (6 in) from the edge, and in the case of this CAC in the front part. For loading a face with the ANFO loader, the mine workers are usually located in the front of the basket, hence the front of the CAC. How much time the blasters are leaning over the edge or towards the edge of the CAC needs to be determined. While the % reductions were good, researchers at NIOSH desired to improve upon these reductions. Therefore, a 7.62-cm (3-in) lip which protruded below the CAC was inserted around the CAC perimeter to improve reductions, especially around the edge.
The 7.62-cm (3-in) lip increased the ow in the front and left edge (Fig. 10). The air ow velocities in the front seemed to equalize, ranging from 0.57 to 1.24 m/s. The four central quadrant air ow velocities ranged from 0.34 to 1.28 m/s, which was less than the previous design without the 7.62-cm (3-in) lip. However, the velocities seem to be more uniform than for the old plenum. There were some sections which had low velocity-1 quadrant at 0.34 m/s and the right and back sections at 0.11 and 0.24 m/s. However, this did not affect the protection as seen in a comparison of Fig. 10 (no lip) and Fig. 11 (with lip).
The 7.62-cm (3-in) lip made a signi cant difference for reductions in particles in the center of the CAC. As seen in Figs. 10 and 12, the reductions in the center quadrants went from 55-78% to 65-86%. The reductions in all but one of the center quadrants were over 74% and closer or over 80% for six out of the eight. This also showed more consistency in the center, for the differences between quadrants in the center was minimized when the 7.62-cm (3-in) lip was added. Around the edge the percent reductions still varied, as the reductions ranged from 7-53% with high variability or con dence limits in some cases. The position of the CAC should be designed for the mine worker to avoid the edges as much as possible when performing his/her duties.

Effect of Distance below CAC
Reed et al. 2019 [14] demonstrated that the air ow and reductions in dust concentrations will start to decrease as one is farther from the CAC. Therefore, the e ciency of the diesel CAC was measured at different distances from the CAC to determine where the CAC needs to be located pertaining to the mine worker to still provide adequate protection from DPM. Figure 12 shows how the air velocity changes the further below the CAC with the 7.62-cm (3-in) lip, but most changes are not more than 10%. There are not too many drastic changes in ow even as one measures from 25.4 cm (10 in) to 76.2 cm (30 in) below the CAC.
However, there are more changes in the CAC with 7.62-cm (3-in) lip percent reductions in submicron particles as seen in Fig. 13. Observing the CAC with the 7.62-cm (3-inch) lip, the reductions in particles decreased as one sampled farther down from the CAC. The average reductions for the center quadrants go from 78 to 69 to 58 as the distance below the CAC is 25.4 cm (10 in), then 50.8 cm (20 in), and nally 76.2 cm (30 in). Mine workers should be as close to the bottom of the CAC plenum as feasible. Being 25.4 to 50.8 cm (10 to 20 in) below the CAC seems to provide reduction in the 70-80% range in the center of the CAC. When constructing the diesel DPM on a vehicle, it should be located so that the mine worker is within 20 inches below the CAC.

Conclusion
The particle count results show that the CAC can provide substantial protection from DPM exposure to mine workers (approximately 80% reduction of submicron particles) when workers are centrally located underneath the CAC which has a 7.62-cm (3-in) lip. As the miner worker approaches within 15.24 cm (6 in) of the edge of the CAC, the percent reduction reduces to 30-50% range. The mine worker should attempt to stay within 15.24 cm (6 in) from the edge of the CAC to achieve the best results. In addition, the CAC should be designed to have the breathing zone of the mine worker to be 25.4 to 50.8 cm (10 to 20 in) below the CAC, for the reductions in submicron particles decreased as the mine worker was farther below the CAC.

Limitations
These reductions are approximations since submicron particles were used to determine them and not an actual atmosphere of DPM.
In addition, this study is laboratory work which is useful for validating control technologies. However, eld work is necessary to establish the feasibility of this control technology and to measure reductions in DPM under actual mining conditions. Since it can matter where the mine worker is located under the CAC, installing the CAC on an ANFO loader at a mine and evaluating the reductions as blasters perform their routine is needed to determine the reductions in DPM from the CAC under actual mining conditions and hence the feasibility of this control technology.

Disclaimer
The ndings and conclusions in this paper are those of the authors and do not necessarily represent the o cial position of the National Institute for Occupational Safety and Health (NIOSH), Centers for Disease Control and Prevention. Mention of any company name or product does not constitute endorsement by NIOSH.