Document 50w94r056Q32E2Lr9wYaLKKvV

R1 9425 REPORT OF INVESTIGATIONS/1992 Relationship of Coal Seam Parameters and Airborne Respirable Dust at Longwalls By J. A. Organiscak, S. J. Page, and R. A. Jankowski Mission: As the Nation's principal conservation agency, the Department of the Interior has respon sibility for most of our nationally-owned public lands and natural and cultural resources. This includes fostering wise use of our land and water resources, protecting our fish and wildlife, pre serving the environmental and cultural values of our national parks and historical places, and pro viding for the enjoyment of life through outdoor recreation. The Department assesses our energy and mineral resources and works to assure that their development is in the best interests of all our people. The Department also promotes the goals of the Take Pride in America campaign by encouraging stewardship and citizen responsibil ity for the public lands and promoting citizen par ticipation in their care. The Department also has a major responsibility for American Indian reser vation communities and for people who live in Island Territories under U.S. Administration. Report of Investigations 9425 Relationship of Coal Seam Parameters and Airborne Respirable Dust at Longwalls By J. A. Organiscak, S. J. Page, and R. A. Jankowski UNITED STATES DEPARTMENT OF THE INTERIOR Manuel Lujan, Jr., Secretary BUREAU OF MINES T S Ary, Director Library of Congress Cataloging in Publication Data: Organiscak, John A. Relationship of coal seam parameters and airborne respirable dust at longwalls / by J.A. Organiscak, S.J. Page, and R.A. Jankowski. p. cm. -- (Report of investigations; 9425) Includes bibliographical references (p. 14). 1. Coal mines and mining--Dust control. 2. Longwall mining--Environmental aspects--United States. 3. Bituminous coal--United States. 1. Page, Steven J. II. Jankowski, Robert A. III. Title. IV. Series: Report of investigations (United States. Bureau of Mines); 9425. TN23.U43 [TN312] 622 s-dc20 [622'.83] 92-10263 CIP CONTENTS Page Abstract............................................................................................................................................................................... Introduction........................................................................................................................................................................ Survey strategy................................................................................................................................................................... Identification of significant coal seam parameters........................................................................................................ Seam types and dust compliance......................................................................................................................................... Discussion of coal seam parameters .................................................................................................................................. Conclusions ............................................................................................................................................................................ References............................................................................................................................................................................... Appendix.--Survey data and analysis..................................................................................................................................... 1 2 2 6 11 13 13 14 15 ILLUSTRATIONS 1. Percentage of U.S. longwalls in compliance with the Federal dust standard ................................................ 2. U.S. coal provinces ................................................................................................................................................ 3. Frequency histogram of coal seam types sampled............................................................................................ 4. Correlation matrix of seam parameters and dust criteria................................................................................. 5. Scatter plot of ash content, volatile matter, and tailgate dust concentration ................................................ 6. Scatter plot of ash and headgate dust concentration ........................................................................................ 7. Scatter plot of ash and headgate specific (normalized)dust............................................................................. 8. Scatter plot of ash and tailgate dust concentration .......................................................................................... 9. Scatter plot of ash and tailgate specific (normalized)dust................................................................................ 10. Scatter plot of volatile matter and tailgate dust concentration ....................................................................... 11. Scatter plot of HGI and tailgate dust concentration ........................................................................................ 12. Average ash and volatile content of provinces sampled................................................................................... 13. Average dust concentrations of provinces sampled .......................................................................................... 14. Average dust compliance concentrations of provinces surveyed ..................................................................... 15. Average ash and volatile content in two seam types sampled in the Eastern Province............................... 16. Average dust concentrations in two seam types sampled in the Eastern Province...................................... 2 4 5 6 7 7 8 8 9 10 10 11 12 12 12 12 TABLES 1. ASTM standards for coal classification............................................................................................................... 2. Chi-square test of sample distribution ............................................................................................................... 3. Differences in dust generation by continuous miners in two different coal seams ...................................... A-l. Dust criteria measurements.................................................................................................................................. A-2. Coal characteristics................................................................................................................................................ A-3. Operation parameters........................................................................................................................................... A-4. General mine characteristics ................................................................................................................................ A-5. Nonlinear model parameter table........................................................................................................................ A-6. Analysis of variance and fit statistics of nonlinear models .............................................................................. 3 5 11 15 16 17 17 18 18 UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT Btu British thermal unit mg milligram Btu/lb British thermal unit per pound mass mg/m3 milligram per cubic meter cfm cubic foot per minute mg/ton milligram per short ton fpm foot per minute pet percent in inch wt pet weight percent in/rev inch per revolution RELATIONSHIP OF COAL SEAM PARAMETERS AND AIRBORNE RESPIRABLE DUST AT LONGWALLS By J. A. Organisesk,1 S. J. Page,2 and R. A. Jankowski3 ABSTRACT The U.S. Bureau of Mines investigated the relationship of bituminous coal seam parameters and the amount of airborne respirable dust generated at longwalls. Dust and coal samples were obtained from 20 longwalls operating in geographically representative coalfields throughout the United States. Statistical analyses of coal seam parameters and airborne respirable dust measurements indicate a likely causal relationship between seam type and respirable dust. Low-ash, high-volatile coal seams were associated with higher airborne respirable dust levels. A negative linear correlation (significant at the 95-pct confidence level) was observed between the seam's ash content and airborne respirable dust, explaining up to 18 pet (R2 = 0.18) of headgate dust level variation and up to 15 pet of tailgate dust level variation. Volatile matter was found to have a positive linear correlation with tailgate dust level, explaining 16 pet of the variation. Further data examination indicates that these relationships with dust are most likely nonlinear, because of improved R2 values over the linear correlations. However, a notable portion of longwall dust production is influenced by other operational parameters, so additional research under more controlled conditions is needed to determine the seam's causative functions. lining engineer. 2Physicist. Supervisory physical scientist. Pittsburgh Research Center, U.S. Bureau of Mines, Pittsburgh, PA. 2 INTRODUCTION Health researchers have strongly suspected a relation ship between dust generation and coal seam type, and nu merous studies have been conducted over the past 25 years to identify this relationship. Several researchers have studied the relationship between coal rank and the preva lence of Coal Workers' Pneumoconiosis (CWP) (1-4):4 Some of these studies found a strong positive correlation and others found a range of weak to very little positive correlation. Thus, there is still some uncertainty regarding the relationship between coal rank and CWP. Other researchers have conducted laboratory studies on the relationships between coal rank, macerals, grindability, and particle size (J-d). These studies have shown conclu sively that there is a significant relationship between coal rank and grindability, and there is a significant relationship between grindability and amount of respirable-sized par ticles found in the product. The relationship between grindability and particle size is not surprising since the Hardgrove grindability index (HGI) is based on particle sizing criteria (7). However, none of these studies actually related airborne respirable dust to grindability. After the enactment of the Federal Coal Mine Health and Safety Act of 1969 in the United States, one field study by the U.S. Bureau of Mines established mean dust exposure differences among various occupations in the in dustry and another field study by the Bureau showed a sig nificant difference in dust exposure for the same occupa tions in different coal mines and/or seams (8-9). The causes of these differences between coal seams could not be quantified, but the differences indicated that coal seam characteristics were one of the likely factors responsible. Under contract to the Bureau, the Southwest Research In stitute extensively reviewed the present knowledge of air borne respirable dust generation and provided several in sightful hypotheses relating coal chemistry and mineralogy (rank, volatility, and ash) to dust generation (10). Fur thermore, Bureau observations on prior longwall dust stud ies indicated notable differences in dust levels in various seams using similar dust control technology. This report describes a recent underground study to determine if type of bituminous coal seam is an influential factor in airborne dust generation. This work is part of the Bureau's pro gram to improve the health of the Nation's miners by reducing their exposure to respirable dust. SURVEY STRATEGY This investigation focused on surveying longwall mining operations. Active longwall mining (shearer) sections were chosen because of the presumed similarities and simplicity in the face ventilation arrangements (head-to-tail ventila tion) for dust sampling. Also, longwall operators' compli ance with the respirable dust standard has recently started to decline (fig. 1) with continued increases in longwall production. Many of the traditional types of dust control technologies are maturing, and further refinement of these traditional technologies may not provide the additional effectiveness needed for more advanced, higher production longwall systems. Therefore, the Bureau is seeking funda mental principles relating seam parameters to airborne dust generation in order to develop and pioneer novel control technology for reducing the amount of respirable dust that becomes airborne. This study was the first phase in identifying any seam parameters associated with air borne respirable dust generation for future causal principle research. Since a randomized sample was economically impracti cal because of the large distances between geographical areas of the country, a "nonprobability purposive" sampling strategy was used. This strategy encompasses drawing a nonprobability sample that conforms to a certain criterion. 4Italic numbers in parentheses refer to items in the list of references preceding the appendix. Figure 1.--Percentage of U.S. longwalls in compliance with the Federal dust standard. 3 The criterion used for this study was to obtain a represent ative sample of longwalls operating in the different bi tuminous seam types (five bituminous classifications of coal of the American Society for Testing and Materials (ASTM)) and in different geographic areas of the country. The minimum number of longwalls needed in a random survey to obtain a representative dust sample of the longwall industry population was 11, based on a targeted pre cision of 0.5 mg/m3 at the 95-pct confidence level with a longwall industry standard deviation of 0.85 mg/m3 (from Mine Safety and Health Administration (MSHA) data). However, 16 longwall sections were surveyed in 15 different coal seams to increase industry coverage since a random sample could not be obtained. Two shifts of sampling were planned throughout the study and were achieved at 13 of the 16 sections surveyed, for a total of 29 shifts of sampling. Secondary data available from 4 other past longwall studies were also used to increase the data base to 20 mines, 17 coal seams, and 33 data files. Data collected for each shift included dust samples, coal samples, ventilation measurements, production during sam pling, and other general mine characteristics. Multiple respirable dust samples were collected at support 10, and roughly 10 supports inby the tailgate of each longwall, with personal sampling instruments. Dust sampling was planned to take place for a minimum of four complete mining passes or face advances. This production goal dur ing sampling was achieved in 70 pet of the shifts sampled. Average airborne dust data collected at the sampling loca tions are shown in table A-l of the appendix. Production during sampling was determined from face advance, seam height, face width, and density. Coal samples were collected from the face conveyor at various times during production. These samples were combined, mixed, coned, and quartered to obtain a small representative sample of the run-of-mine product. These samples were packaged into small airtight containers for transport out of the mine to the laboratory for proximate and Hardgrove grindability analysis. Processing of these coal samples was conducted using ASTM proximate analy sis and HGI classification procedures. Proximate analysis and HGI data are shown in table A-2 of the appendix. The sample set collected approximates a normal dis tribution for bituminous class coals. Figure 2 shows the types of coals in the different coal provinces of the United States; table 1 shows the ASTM coal classifications; and Table 1.--ASTM standards for coal classification Class and group Anthracite: Metaanthracite......... Anthracite ............... Semianthracite......... Bituminous: Low-volatile............. Medium-volatile .... High-volatile A......... High-volatile B......... High-volatile C......... Subbituminous: Subbituminous A ... Subbituminous B .. . Subbituminous C ... Lignitic: Lignite A .................. Lignite B .................. Fixed carbon limits,1 pet >98 92-98 86-92 78-86 69-78 <69 <69 <69 <69 <69 <69 <69 <69 <69 Volatile matter limits,1 pet <2 2- 8 8-14 14-22 22-31 >31 >31 >31 >31 >31 >31 >31 >31 >31 Calorific value limits,2 Btu/lb Agglomerating character >14,000] >14,000 >14,000 >14,000] >14,000 >14,000 13,000-14,000 11,500-13,000 10,500-11,500 10,500-11,500] 9,500-10,500 8,300- 9,500 6,300- 8,300 <6,300, Nonagglomerating.3 Commonly agglomerating.4 Agglomerating.5 Nonagglomerating. 1Dry, mineral-matter-free basis. 2Moist, mineral-matter-free basis. "Moist" refers to the natural inherent moisture of the coal, does not include visible water on the surface. 3lf agglomerating, classify in low-volatile group of the bituminous class. 4There may be nonagglomerating varieties in these groups. 5There are notable exceptions in this group. NOTE.--Coals having 69 pet or more fixed carbon on the dry, mineral-matter-free basis shall be classified according to fixed carbon, regardless of calorific value. 4 Lignite V//A Subbituminous coal Medium and high-volatile bituminous coal Low-volatile bituminous coal Anthracite and semianthracite O 600 IiI Scale, miles Figure 2.--U.S. coal provinces. 5 figure 3 shows a frequency histogram of the set of 33 sam ples divided into the 5 bituminous groups. Medium- and high-volatile A bituminous coal are the most common types of coal in the United States, and the sample data collected correspond to a normal distribution within this bituminous class of coal. Table 2 illustrates the use of the chi-square (x*2) statistic to test the hypothesis that the sample frequency distribution approximates an assumed normal distribution of bituminous coal in the U.S. coal provinces (11). The low-volatile and high-volatile C bi tuminous coal groups were combined with adjacent groups because their frequency was below the recommended group levels for computation of a chi-square test statistic. The hypothesis could not be rejected at the 95-pct con fidence level, so it was believed that a representative sam ple of bituminous coal seams was obtained. Ventilation measurements and other general descriptive data were collected from each operation. Velocity meas urements were made at the dust sampling locations and at 10 support intervals along the face. Headgate, tailgate, and average face velocity and quantity data are shown in table A-3 of the appendix. General information of the op erations sampled, such as mine location, seam, height, face width, and cut sequence used, is shown in table A-4 of the appendix. MIDPOINT Figure 3.--Frequency histogram of coal seam types sampled. Table 2.--Chi-square (*2) test of sample distribution Bituminous coal Midpoint High-volatile C .. . High-volatile B . . . High-volatile A .. . 11,000 Btu' 12,250 BtUj 13,500 Btu Medium-volatile . . Low-volatile .... . 73.5 pet C' 82.0 pet C Heat content: x = 13,561 Btu s = 962 Btu Adjusted midpoint Boundary (XL) z [1 0,500 Btu -3.18 12,000 Btu 11,500 Btu -2.14 13,500 Btu 13,000 Btu [14,000 Btu, 69 pet C -.58 .46 77.5 pet C 78 pet C 1.87 86 pet C 3.04 Carbon content: x = 54.6 pet s = 7.7 pet Area 0-2 0.4993] .4838 .2190, .1772] Net area Expected frequency (B) 0.2803 .3962 9 13 .4693 .4988 .3216 11 z = (XL-x)/s X2 = k (Oi - Ei)2/Ei s i =1 Observed frequency (Oi) 8 (2 + 6) 15 10 (6 + 4) y2 test H0: Oi = Ei Ha: Oi * Ei Let a = 0.05 n = 33 2 = 1^9)2 115-13)2 (10_i.ll)2 = 05in X 9 + 13 11 0,50 Since x2o.os > x^aicuiated cannot reject H0 Test over 3 intervals: df = 2, *20 0S = 5.99 6 IDENTIFICATION OF SIGNIFICANT COAL SEAM PARAMETERS A correlation matrix was used to examine the inter dependence of coal seam parameters and to identify the most important and independent coal seam characteristics with respect to the airborne dust criteria (fig. 4). The measure of association (correlation) used in this matrix is the Pearson product-moment coefficient (r), including sample size and level of significance (p-value) (11). The level of significance is the probability that the variable association is false (Type I error). The dust criterion (de pendent variable) is expressed in both concentrations and airborne respirable mass measured per short ton mined during the sampling period (specific dust) at both the headgate and tailgate sampling locations. The actual airborne dust generated per short ton could not be reliably calculated and used as a criterion variable because of notable variations between the headgate and tailgate air quantities at many of these longwalls (see table A-3 of the appendix). The average percent change in air quantity be tween the headgate and tailgate had a coefficient of varia tion of 1,160 pet (mean = -3.8 pet, standard deviation = 44.1 pet). The wide variation of the percent change in air quantity along the face for all the longwalls makes this parameter unreliable for specific dust determination. Examination of the coal seam parameter correlations show strong associations among several parameters. Some of the strongest and most significant correlations are between ash and sulfur, ash and heat content, ash and vol atility, volatility and HGI, and volatility and moisture content. This is not surprising since various combinations of these parameters affect the carbon content or rank, which is related to the heat content and hardness (HGI) properties of the coal. Moisture content 1.000 (33) 0.000 Sulfur Ash Volatile Heat HGI content content content content Headgate dust cone -0.316 (33) 0.074 -0.365 (33) 0.037 0.566 (33) 0.001 -0.175 (33) 0.330 -0.387 (33) 0.026 0.042 (32) 0.818 1.000 (33) 0.000 0.796 (33) 0.000 -0.436 (33) 0.011 -0.597 (33) 0.000 0.391 (33) 0.024 -0.226 (32) 0.213 1.000 (33) 0.000 -0.531 (33) 0.002 -0.767 (33) 0.000 0.296 (33) 0.944 -0.404 (32) 0.022 1.000 (33) 0.000 0.211 (33) 0.238 -0.857 (33) 0.000 0.199 (32) 0.275 1.000 (33) 0.000 0.031 (33) 0.865 0.248 (32) 0.171 1.000 (33) 0.000 -0.091 (32) 0.622 1.000 (32) 0.000 1.000 (33) 0 .000 KEY Correlation coefficient (r) Sample size Significance level (p) Headgate specific dust -0.032 (32) 0.860 -0.232 (32) 0.202 -0.427 (32) 0.015 0.051 (32) 0.782 0.358 (32) 0.044 0.084 (32) 0.649 0.909 (32) 0.000 1.000 (32) 0.000 T aiIgate dust cone 0.125 (33) 0.487 -0.302 (33) 0.088 -0.329 (33) 0.061 0.405 (33) 0.019 0.149 (33) 0.408 -0.383 (33) 0.028 0.483 (32) 0.005 0.356 (32) 0.046 1.000 (33) 0.000 TaiIgate specific dust 0.004 (33) 0.982 -0.234 (33) 0.190 -0.386 (33) 0.027 0.201 (33) 0.262 0.282 (33) 0.113 -0.099 (33) 0.583 0.631 (32) 0.000 0.709 (32) 0.000 0.757 (33) 0.000 1.000 (33) 0.000 Moisture content Sulfur content Ash content Volatile content Heat content HGI Headgate dust cone Headgate specific dust TaiIgate dust cone TaiIgate specific dust Figure 4.--Correlation matrix of seam parameters and dust criteria. 7 Examination of all the seam parameters and the air borne dust criteria show that ash had the highest correla tion (significant at the 95-pct confidence level) with dust concentration (r = -0.40, p = 0.02) and specific dust (r = -0.43, p = 0.02) at the headgate sampling location, explaining 16 and 18 pet (R2 values) of the variation, re spectively. At the tailgate, volatility had the highest cor relation (r = 0.40, p = 0.02) with the dust concentration, and ash had the highest correlation (r = -0.38, p = 0.03) with specific dust, explaining 16 and 14 pet of the variation, respectively. Ash and volatility are significantly and neg atively correlated with one another (r = -0.53, p = 0.00), so the ash was also fairly correlated with dust concentrations at the tailgate (r = -0.33, p = 0.06). Therefore, it was generally concluded that high-volatile, low-ash coal seams tend to experience higher airborne respirable dust levels. Several operational parameters measured (water application, maximum bit depth, etc.) were found to have insignificant correlations with dust concentrations, and their associations will not be included in any further discussions. Figure 5 shows the three dimensional scatter plot of ash and volatility with respect to the dust concentration at the tailgate. Scatter plots of individual coal seam parameters and dust (concentration and specific) indicate that indeed ash and volatility were the parameters with the strongest re lationships, and these relationships could be accurately characterized as nonlinear in nature. Figures 6 to 9 show Figure 5.--Scatter plot of ash content, volatile matter, and tail gate dust concentration. Figure 6.--Scatter plot of ash and headgate dust concentration. 8 ___________ ASH CONTENT, pet Figure 7.--Scatter plot of ash and headgate specific (normalized) dust. ro E O' E O S cc LlI o z o o CO 3 Q ASH CONTENT, pet Figure 8.--Scatter plot of ash and tailgate dust concentration. 9 ASH CONTENT, pet Figure 9.--Scatter plot of ash and tailgate specific (normalized) dust the scatter plots for the ash and dust data. It appears that the ash parameter could better fit several decay models (y = ae*x and y = ax*) for both the dust concentrations and specific (normalized) dust criteria at both the face lo cations. The parameters and fit statistics of these models are shown in tables A-5 and A-6 of the appendix. Explan ation of the dust variation can be increased to 70 pet (R2 value) at the headgate and 27 pet at the tailgate with the y = ax'b model. Although the y = ax* model is more effi cient than the y = ae*x, it is not a valid model because there is no finite limit to dust generation. Therefore, cau tious judgment should be used with parametric modeling analysis of the survey data because of the fair amount of data scatter not accounted for by the ash parameter and because of the limited amount of data at the extreme ranges of seam types, which can enormously affect the de termination of model parameters. The tailgate data were found to have much more scatter than the headgate data, which was probably because more source contributions were measured at the tailgate (shearer and supports). Since too many uncontrolled and unmeasured parameters exist in the underground mining environment, controlled laboratory experiments are recommended to determine the most reliable and valid model of the ash parameter. The volatility and dust concentration relationship at the tailgate was found to be roughly the opposite of the ash relationship (fig. 10). The volatility data may be better de scribed a polynomial model (y = a - bx + cx2). However, caution again is advised with this type of model determina tion because of the data scatter (more scatter than the ash parameter data) and the low number of points for lowvolatile coals. Since HGI is usually cited as an indicator of coal seam dustiness, it has been examined to observe if this pa rameter is associated with the airborne respirable dust measured. The only strong association found was a signifi cant negative correlation between HGI and tailgate dust concentrations (r = -0.38, p = 0.03; see figure 4, the cor relation matrix), explaining 14 pet of the dust variation. This HGI and tailgate dust concentration association also coincides with the volatile parameter because HGI and volatile matter are also highly and negatively correlated (r = - .86, p = 0.00). Figure 11 shows the scatter plot of HGI versus the tailgate dust concentrations. A decay non linear model would probably better describe the associa tion, but this analysis is cautioned against because of the scatter in the data. Although a negative correlation between HGI and dust was observed (significant at the 95-pct confidence level), DUST CONCENTRATION, mg/m 10 Figure 10.--Scatter plot of volatile matter and tailgate dust concentration. HGI Figure 11.--Scatter plot of HGI and tailgate dust concentration. DUST CONCENTRATION, m g/m 11 this relationship is contrary to the speculated from rela tionship. The theorized relationship is a positive cor relation between HGI and dust generation, derived from laboratory grinding tests. A potential reason for this dis crepancy is that the amount of dust entrained into the airstream may not be directly proportional to the amount generated in the product because some other coal seam parameter affects airborne dust generation or entrainment, such as ash and/or volatility. Some other secondary data that support this hypothesis are the data obtained at two continuous miner sections in two different coal seams (table 3) (12). These data were obtained from several individual cuts. The Pittsburgh Coal Seam had a lower HGI index (61) than the No. 2 gas seam (70) and less respirable dust measured in the mined prod uct. However, the Pittsburgh Seam had a significantly higher amount of airborne dust (concentrations and per cent airborne per amount in product) measured in the re turn. Dust concentrations were 300 pet higher and the portion of airborne dust per respirable dust in run-of-mine product was 318 pet higher in the Pittsburgh Seam than in the No. 2 gas seam. Although no proximate analysis was conducted on these coal samples (not part of the original study), the No. 2 gas seam generally has a higher ash and lower volatile content than does the Pittsburgh Seam. Thus, the authors believe that the amount of airborne res pirable dust generated from mechanized cutting may not be directly related to HGI as theorized or assumed from laboratory studies. A National Coal Board study has also noted,"... stronger coals produce less dust than the weak, but break more explosively, causing a greater proportion of the fine dust formed to be dispersed into the air" (13). Bureau coal-cutting studies in the laboratory have indi cated a direct trend between measured peak cutting forces and airborne respirable dust generation (14). The strong association of airborne respirable dust to ash and vola tility observed from this longwall study indicates that some causal entrainment phenomena linked to these seam parameters may be responsible. Table 3.--Differences In dust generation by continuous miners in two different coal seams Coal seam ... Seam height .. ........... in .. Number of cuts isampled......... HGI.................. Respirable dust igenerated in product ......... Airborne respirable dust generated .... Portion.............. ......... pet.. Air quantity . .. ......... cfm . . Pittsburgh 79 6 61.0 + 0.5 5,458 595 1.32+0.59 0.0230.012 13,467 2,190 No. 2 gas 61 4 70.4 2.0 5,955 879 0.33 0.10 0.0055 0.0017 4,1121,278 SEAM TYPES AND DUST COMPLIANCE From this study it was generally concluded that low-ash, high-volatile coal seams produce more dust. Does this conclusion explain regional differences found in dust com pliance rates throughout the MSHA districts in the coun try? To examine this connection, the seam parameters and dust concentrations (primary data) collected in dif ferent geographic provinces were averaged with 95-pct confidence intervals. MSHA compliance samples (second ary data) were also averaged for all longwall operations in these provinces. Figures 12 and 13 show the average seam types and dust concentrations measured for the Eastern, Interior, and Rocky Mountain Coal Provinces in this study, and figure 14 shows their compliance concentration av erages for 1986-88 from MSHA data. These data show that the average seam parameters and dust levels do vary between provinces. The Rocky Moun tain Province on average has higher volatility and lower ash coal than the other provinces. This province also had the highest average dust concentration measured in this study and the highest average dust concentration from compliance sampling. The Eastern and Interior Provinces on average have lower volatility and higher ash coal seams. These provinces also had lower dust concen trations measured in this study and lower average dust Figure 12.--Average ash and volatile content of provinces sampled. 12 concentrations from compliance sampling. Thus, the rel ative differences among the regional MSHA compliance dust data correspond to average regional seam types and dust levels measured in this study. Similar ash-volatility and dust relationships were also found within the Eastern Province. Figures 15 and 16 show the ash-volatility content and dust concentrations measured within the Eastern Province. Significantly higher dust concentrations were observed at the four longwalls operating in lower ash, higher volatile coal seams. Since the secondary MSHA data concur with the data collected in this study, seam characteristics seem to be an important factor in the amount of dust generated and the ability to maintain compliance. Figure 13.--Average dust concentrations of provinces sam pled. Figure 15.--Average ash and volatile content In two seam types sampled in the Eastern Province. DUST CONCENTRATIONS, mg/m3 Figure 14.--Average dust compliance concentrations of prov inces surveyed. DUST CONCENTRATIONS, mg/m3 Figure 16.--Average dust concentrations in two seam types sampled in the Eastern Province. 13 DISCUSSION OF COAL SEAM PARAMETERS In this longwall study, the Bureau found that low-ash, high-volatile bituminous coals tended to produce more air borne respirable dust. Although these were the chief parameters affiliated with airborne dust generation, their specific causal functions in airborne dust generation are not known. The actual dust generation phenomena may involve several interrelated coal parameters that have chemical and/or mineral causative mechanisms of airborne respirable dust generation. Three possible causative hy potheses are proposed below that could be considered for future research on coal seam airborne dust. The first two hypotheses were proposed by the Southwest Research In stitute based on its basic research on coal fragmentation and dust entrainment conducted for the Bureau (10), and the third hypothesis was proposed by the authors of this report. 1. Coal fragmentation from cutting usually occurs along planes of imperfections formed by ash material (making the coal a more heterogeneous material), which facilitates breakage into a larger size distribution and reduces the amount of respirable dust generated. Note that ash con tent and HGI in this study have an insignificant correlation (fig. 4). This statistic indicates that the ash parameter has very little association with grindability and suggests that coal grinding properties may not be the best indicator for airborne respirable dust formed during cutting or haulage fragmentation. 2. Electrostatic charge on respirable coal dust, which is partially responsible for entrainment, may arise as a consequence of surface-air reactions (volatilization and oxidation). Chemical and mineralogical classification re search on respirable-sized coal dust found an increase in both carbon and hydrogen content and a decrease in in organic chemical content with increasing particle size. This may be due to volatilization of light hydrocarbon from the particle surface. 3. Electrostatic charge properties of ash minerals could possibly be responsible for dust coagulation, reducing the amount of respirable dust that becomes airborne. The major components of ash are typically silica, clay (kaolinite), and sulfur. Sulfur or iron pyrite was highly correlated to ash content, but was not as highly associated with air borne dust generation as was ash. Both sulfur and clay particles can have significantly higher electrostatic charges than coal particles (15-16). These three hypotheses center around chemical or mineralogical causes affiliated with ash and volatility param eters identified in this study. CONCLUSIONS Coal seam parameters are associated with the amount of dust generated at longwalls. Low-ash, high-volatile bituminous coal seams tend to experience higher res pirable dust levels during mining. The ash parameter could be better described by nonlinear decay models for both the dust concentration and specific dust criterion. The volatile matter parameter could better describe the tailgate dust concentrations by a polynomial growth model. HGI was found to be indirectly related to airborne res pirable dust generation, contrary to the hypothesized belief of direct association. The ash-volatility parameters and dust measurements in regional geographic areas (coal provinces) of the United States relatively agreed with MSHA compliance data. The Rocky Mountain Province averaged the highest dust levels for both this survey and MSHA compliance data. This province on average contains low-ash, high-volatile coal seams, coinciding with the ash relationship found in this study. The Eastern and Interior Provinces on average had lower volatile, higher ash coal seams with lower MSHA reported dust levels and higher compliance rates. This ash-volatility and dust relationship was also observed within the Eastern Province, which indicates that low-ash, high-volatile seams are not located only in one province. Additional research studies should strive to verify the ash and volatility association with airborne respirable dust under strictly controlled conditions by reducing the un explained data scatter (most likely caused by numerous op erational parameters). Reducing the data scatter will also improve nonlinear model development of these coal seam parameters. Further research efforts should target verifi cation of the above suggested hypotheses or any other proposed hypotheses for developing fundamental principles of airborne dust generation. 14 REFERENCES 1. Morgan, W. K. C. The Prevalence of Coal Workers' Pneumo coniosis. Am. Rev. Rcspir. Dis., v. 98,1968, pp. 306-310. 2. Bennett, J. G., J. A. Dick, Y. S. Kaplan, P. A. Shand, D. H. Shcnnan, D, J. Thomas, and J. S. Washington. The Relationship Between Coal Rank and the Prevalence of Pneumoconiosis. Br. J. Ind. Med., v. 36,1979, pp. 206-210. 3. Nacyc, R. L., J. K. Mahon, and W. S. Dellinger. Rank of Coal and Coal Workers' Pneumoconiosis. Am. Rev. Rcspir. Dis., v. 103,1971, pp. 350-355. 4. Hurley, J. F., J. Burns, L. Copland, J. Dodgson, and M. Jacobsen. Coal Workers' Simple Pneumoconiosis and Exposure to Dust at 10 Brit ish Coal Mines. Br. J. Ind. Med., v. 39,1982, pp. 120-127. 5. Baafi, E. Y., and R. V. Ramani. Rank and Maceral Effects on Coal Dust. Int. J. Rock Mcch. and Min. Sci. (& Geomcch Abstr.), v. 16, 1979, pp. 107-115. 6. Moore, M. P., and C. J. Bise. The Relationship Between the Hardgrovc Grindability Index and the Potential for Respirable Dust Generation. Paper in Coal Mine Dust Conference (Morgantown, WV, Oct. 8-10,1984). Generic Technology Center for Respirable Dust, 1984, pp. 250-255. 7. American Society forTesting and Materials. Standard Test Meth od for Grindability of Coal by the Hardgrovc Machine Method. D40990 in 1990 Annual Book of ASTM Standards: Volume 05.05.1990, pp. 197-202. 8. Schlick, D. P., R. G. Pcluso, and W. E. Byers. An Evaluation of the Respirable Dust SamplingProgram in United Slates Underground Coal Mines. BuMincs TPR 32,1971, 20 pp. 9. Seibel, R. J., and F. E. McCall. The Dustiness of Different Coal Seams. BuMincs TPR 75, 1974, 9 pp. 10. Stecklein, G. L., R. Branstcttcr, R. Arrowood, D. Davidson, J. Lankford, R. Lyle, and C. Nulton. Basic Research on Coal Fragmenta tion and Dust Entrainment. Volume I---Technical Information (contract J0215009, Southwest Research Inst.). BuMincs OFR 215(l)-83, 1982, 121 pp. 11. Neter, J., W. Wasserman, and G. A. Whitmore. Applied Statistics, Allyn and Bacon, 1988, pp. 329, 526. 12. Ramani, R. V., J. M. Mutmansky, R. Bhaskar, and J. Qin. Funda mental Studies on the Relationship Between Quartz Levels in the Host Material and the Respirable Dust Generated During Mining, Volume I: Experiments, Results and Analysis (contract II0358031, PA State Univ.). BuMincs OFR 36-88,1987, 179 pp.; NTIS PB 88-214325/AS. 13. National Coal Board (United Kingdom). Mining Research and Development Review by Mining Research and Development Establish ment. No. 7, Apr. 1973,12 pp. 14. Strcbig, K. C,, and H. W. Zeller. Effect of Depth of Cut and Bit Type on the Generation of Respirable Dust. BuMincs RI 8042, 1975, 16 pp. 15. Inculct, 1.1., R. M. Quigley, and E. M. J. Beisser. Electrostatic Charges on Clays. IEEE Trans. Ind. Appl., v. IA-21, No. 1, Jan./Fcb. 1985, pp. 23-25. 16. Mukhcrjec, A., D. Gidaspow, and D. T. Wasan. Surface Charge of Illinois Coal and Pyrites for Dry Electrostatic Cleaning. Am. Chem. Soc., v. 32, No. 1,1987, pp. 395-407. APPENDIX.--SURVEY DATA AND ANALYSIS Table A-1.--Dust criteria measurements Mine Shift Coal pro duced, tons A____ B___ C .... D___ E___ F___ G .... H___ 1......... J___ K___ L___ M___ N___ 0 P___ Q .... R___ S .... T___ dumped. 1-4 1-4 1-4 1-5 1 2 1 1 2 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 1 2 1 2 1 2 1,042 1,148 2,377 3,067 2,443 2,443 1,305 2,138 1,426 751 3,240 2,835 3,038 2,126 950 1,425 1,316 1,053 826 826 3,645 3,645 1,786 1,786 2,430 2,126 1,807 1,504 1,604 2,495 2,495 1,382 1,382 Dust concentration, mg/m3 Headgate Tailgate 1.57 .94 .55 1.51 2.11 .78 3.22 1.68 (*) 3.09 3.62 2.50 3.34 3.98 2.82 3.05 .59 .90 9.30 14.30 1.80 1.53 1.54 3.81 .89 1.44 .59 .70 2.46 3.01 .75 1.12 2.40 2.05 .94 1.04 2.36 14.54 11.83 5.24 3.77 3.83 4.21 7.75 9.17 8.68 4.93 5.14 3.62 2.89 5.24 18.12 8.24 1.90 2.53 2.81 3.94 3.57 5.50 .79 3.31 4.87 7.35 6.91 3.72 3.53 Dust mass, mg Headgate Tailgate 0.95 1.18 .52 .49 .28 .53 .89 1.34 1.09 5.47 .25 3.29 2.11 2.89 .73 1.40 t1) .92 1.40 1.68 1.85 3.52 1.17 3.80 1.25 2.92 2.40 2.79 1.21 1.98 1.21 1.34 .35 1.56 .24 1.40 2.03 3.88 3.35 1.88 .90 1.35 .92 1.40 .25 1.14 .71 1.86 .60 2.15 .58 2.07 .38 .40 .36 1.54 1.34 2.53 .68 1.53 .35 2.70 .25 1.04 .51 1.09 NOTE.--Dust measurements not in MRE equivalents. Specific dust, mg/ton Headgate 9.079E-04 4.512E-04 1.195E-04 2.902E-04 4.458E-04 1.040E-04 1.618E-03 3.410E-04 (*) 1.860E-03 5.719E-04 4.109E-04 4.11 IE-04 1.131 E-03 1.271 E-03 8.477E-04 2.675E-04 2.260E-04 2.454E-03 4.051 E-03 2.480E-04 2.510E-04 1.417E-04 3.987E-04 2.449E-04 2.747E-04 2.092E-04 2.420E-04 8.373E-04 2.725E-04 1.395E-04 1.780E-04 3.719E-04 Tailgate 1.131 E-03 4.294E-04 2.234E-04 4.359E-04 2.238E-03 1.346E-03 2.216E-03 6.525E-04 6.445E-04 2.242E-03 1.086E-03 1.339E-03 9.598E-04 1.312E-03 2.088E-03 9.396E-04 1.186E-03 1.333E-03 4.695E-03 2.275E-03 3.701 E-04 3.846E-04 6.355E-04 1.041 E-03 8.844E-04 9.732E-04 2.230E-04 1.021 E-03 1.579E-03 6.128E-04 1.080E-03 7.540E-04 7.916E-04 15 16 Mine Shift A .. B .. C .. D .. E .. F .. G .. H .. 1... J .. K .. L .. M.. N .. 0.. P .. Q.. R .. S .. T .. 1-4 1-4 1-4 1-5 1 2 1 1 2 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 1 2 1 2 1 2 lDry basis. Table A-2.--Coal characteristics Heat content, Btu/lb 14,092 13,651 13,875 13,820 13,980 13,944 14,897 13,189 13,855 15,074 13,899 14,195 14,390 14,687 14,592 14,434 13,265 12,993 13,635 13,875 12,408 12,845 12,926 14,268 10,533 11,079 14,240 13,464 13,635 13,535 13,026 12,694 12,530 Moisture content, pet 3.00 1.00 1.10 1.10 3.03 3.30 .92 .90 .93 .89 1.85 1.99 1.81 1.69 .51 .48 1.60 1.40 3.03 3.21 6.12 6.61 .51 .49 .57 .71 .81 6.19 5.95 4.31 3.54 3.72 3.44 Sulfur content, pet 3.84 .86 2.04 3.68 .57 .43 1.06 1.15 .86 .65 .77 .84 1.17 .77 .50 .51 1.84 2.13 .94 .92 .44 .29 3.68 1.80 5.55 3.87 1.11 .49 .51 .52 .47 3.98 3.07 Ash content,1 pet 9.60 11.90 11.80 20.90 5.36 4.90 2.97 15.00 10.40 2.86 7.47 5.79 5.14 3.38 7.19 7.70 11.36 12.99 5.07 2.00 7.53 4.97 13.47 6.18 29.23 26.53 9.08 4.88 4.89 6.43 10.82 12.23 13.52 Volatile content,1 pet 19.40 29.60 37.60 40.00 47.10 46.10 39.50 32.20 33.10 33.40 38.54 38.87 38.58 39.41 19.05 19.99 38.82 36.96 43.58 41.72 40.80 42.10 37.72 40.63 17.40 16.62 21.63 46.03 47.19 43.09 42.72 36.76 38.65 HGI 125 70 61 70 48 49 59 59 62 62 47 51 51 53 95 97 54 50 56 59 53 56 61 59 79 82 94 49 49 43 38 60 57 17 Table A-3.--Operation parameters Mine A .. B .. C.. D .. E .. F .. G.. H .. 1... J .. K .. L .. M.. N .. O.. P .. Q.. R .. S .. T .. Shift 1-4 1-4 1-4 1-5 1 2 1 1 2 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 1 2 1 2 1 2 Water applied, wt pet NA NA NA NA 0.94 .93 .78 1.94 2.51 .70 1.38 1.68 .94 .94 .81 .88 1.21 1.18 6.37 6.45 8.39 7.70 1.47 2.11 1.45 1.91 3.09 4.98 3.09 2.00 2.73 5.30 3.72 ^rom headgate to tailgate. Max bit depth, in/rev 1.42 3.23 3.11 1.97 5.60 5.07 1.50 4.27 3.60 3.14 4.80 4.53 4.74 5.05 4.00 5.64 2.63 2.49 2.43 1.95 3.10 3.10 3.82 3.27 2.27 3.08 3.71 3.20 3.20 4.13 5.33 4.13 3.73 Air velocity, fpm Head gate Av face Tailgate 250 295 261 241 167 98 418 438 450 329 285 267 230 254 228 250 261 332 256 357 220 242 333 355 403 376 376 102 68 98 249 260 248 222 240 193 284 238 224 541 308 322 816 524 417 257 474 557 340 496 490 373 433 437 408 239 228 278 239 230 258 276 308 258 280 292 419 313 218 416 307 181 425 336 308 476 329 318 253 217 238 327 316 308 335 324 320 657 502 312 294 330 283 213 397 414 210 401 404 Air quantity, cfm Headgate Av face Tailgate 16,875 17,328 30,430 20,859 30,199 32,825 7,424 13,189 21,964 4,580 11,180 9,968 40,016 76,227 62,750 19,763 15,266 16,748 37,903 25,826 31,399 31,399 34,903 34,653 43,563 48,790 21,075 39,796 40,770 67,342 30,135 19,789 19,509 18,393 12,007 25,601 20,948 33,350 34,269 10,353 18,149 20,492 3,053 11,674 10,776 33,320 43,397 40,296 36,451 22,270 19,442 22,203 22,203 33,589 34,076 26,073 25,573 34,440 33,723 18,076 38,457 39,431 51,455 33,825 36,881 37,253 14,929 7,046 19,845 22,321 29,936 43,592 6,380 19,348 20,492 4,400 11,135 8,666 31,562 45,370 32,067 42,833 22,001 19,621 21,181 21,367 37,484 35,536 18,159 15,077 31,570 32,595 19,825 37,484 38,944 31,980 29,008 38,461 37,532 Table A-4.--General mine characteristics Mine Seam Height, in Panel width, ft Cut sequence1 A ... B ... C ... D ... E ... F ... G ... H ... 1 J ... K ... L ... M... N ... O... P ... Q... R ... S . .. T ... Lower Kittanning . .. . Eagle.......................... . Pittsburgh ................. . . Blind Canyon............ . Dorchester................. . No. 2 gas................... . Campbell Creek . . . . . Warfield..................... . Harlan ........................ . Blue Creek................. . Pratt .......................... . Wattis ........................ . Eagle No. 5 (F)......... . Pittsburgh ................. . Upper Freeport......... . . . do............................. . O'Connor................... . Hiawatha................... . Herrin No. 6.............. . 66 72 65 90 114 50 66 60 60 120 80 60 90 108 84 96 84 108 96 90 440 Bi-Di 680 Uni-Di (H-*T cut) 615 Uni-Di (T->H cut) 650 Uni-Di (T-*H cut) 760 Uni-Di (H-*T cut) 580 Uni-Di (T-H cut) 580 Bi-Di 550 Uni-Di (T-H cut) 600 Bi-Di 540 Uni-Di (T-*H cut) 660 Uni-Di (T-H cut) 850 Bi-Di 480 Uni-Di (T-*H cut) 750 Bi-Di 630 Uni-Di (T-H cut) 750 Bi-Di 600 Bi-Di 550 Uni-Di (T-H cut) 700 Uni-Di (T-*H cut) 910 Uni-Di (H-*T cut) 1Bi-Di, bidirectional; Uni-Di, unidirectional; H-*T, headgate to tailgate; T-*H tailgate to headgate. Change in quantity,1 pet -12 -59 -35 7 -1 33 -14 47 -7 -4 0 -13 -21 -40 -49 117 44 17 -44 -17 19 13 -48 -56 -28 -33 -59 -6 -4 -53 -4 94 92 18 Table A-5.--Nonlinear model parameter table Variable and criterion Model Parameter Ash: Headgate dust concentration .. Do.............................................. Headgate specific dust ........... Do.............................................. Tailgate dust concentration . .. Do.............................................. Tailgate specific dust................ Do.............................................. Volatility: Tailgate dust concentration . . . fl> <D $ ae*x ax* ae*x ax* ae*x ax* ax* a - bx + cx2 a b a b a b a b a b a b a b a b a b c H0: Parameter = 0. Hj: Parameter = 0. When Tslatlsllc < Tcnllca^ cannot reject H0, if Tsjaj> Tcraica]( reject H0. Parameter value 15.726 0.306 29.043 1.407 0.010 0.529 0.012 1.738 9.297 0.069 11.959 0.410 2.6 xl O'3 0.102 3.8 x 10'3 0.603 16.350 1.044 0.019 Standard error 4.968 0.079 8.150 0.226 0.003 0.093 0.003 0.212 2.231 0.033 4.273 0.193 6.1 x 10* 0.037 1.2 x 10* 0.177 7.500 0.505 0.008 ^statistic 3.166 3.886 3.564 6.231 3.505 5.709 4.267 8.190 4.167 2.051 2.799 2.119 4.320 2.796 3.296 3.412 2.180 2.065 2.425 T 0.05 1 critical 2.042 2.042 2.042 2.042 2.042 2.042 2.042 2.042 2.040 2.040 2.040 2.040 2.040 2.040 2.040 2.040 2.042 2.042 2.042 Table A-6.--Analysis of variance and fit statistics of nonlinear models Variable and criterion Model Source of variation Sum of squares Ash: Headgate dust concentration. Do................. Headgate specific dust. Do................. Tailgate dust concentration. Do.................. Tailgate specific dust. Do.................. Volatility: Tailgate dust concentration. ae*x ax* ae*x ax* ae*x ax* ae*x ax* a - bx + cx2 Regression Residual .. Regression Residual.. Regression Residual.. Regression Residual. . Regression Residual.. Regression Residual.. Regression Residual.. Regression Residual .. Regression Residual.. 297.89 140.44 330.59 107.74 2.8 x 10* 8 x 10* 2.9 x 10* 1 x 10* 992.66 399.65 987.25 405.07 5.2 x 10* 1.8 x 10* 5.3 x 10* 1.7 x 10* 1,063.35 328.97 H0. Hi' When ^variable " ^criterion. loanable -- Peritenon.' _ < ^critical' cannot reject H0, if F~statistic '> ^critical* i'jct H0. Degrees of freedom 2 30 2 30 2 30 2 30 2 31 2 31 2 31 2 31 3 30 NOTE.--R2 is determined from total corrected source of variation (s (y-y)2). Mean square error 148.95 4.68 165.30 3.59 1.4 x 10* 0.00 1.5 x 10* 0.00 496.33 12.89 493.62 13.07 2.6 x 10* 6 x 10* 2.6 x 10* 1 x 10* 354.45 10.97 R2 0.39 0.53 0.61 0.70 0.15 0.14 0.24 0.27 0.30 ^statistic 31.82 46.02 49.48 69.24 38.50 37.78 45.55 47.18 32.32 ^"critical 3.32 3.32 3.32 3.32 3.31 3.31 3.31 3.31 3.32