This Users Handbook for the Argonne Premium Coal Samples provides the recipients of those samples with information that will enhance the value of the samples, to permit greater opportunities to compare their work with that of others, and aid in correlations that can improve the value to all users. It is hoped that this document will foster a spirit of cooperation and collaboration such that the field of basic coal chemistry may be a more efficient and rewarding endeavor for all who participate.

The different sections are intended to stand alone. For this reason some of the information may be found in several places.

The handbook is also intended to be a dynamic document, constantly subject to change through additions and improvements. Please feel free to write to the editor with your comments and suggestions.

THE EIGHT COALS IN THE ARGONNE PREMIUM COAL SAMPLE PROGRAM*

(adapted from a paper in Energy & Fuels 4, 420-426 (1990))

* Work performed under the auspices of the Office of Basic Energy Sciences, Division of Chemical Sciences, U. S. Department of Energy, under contract number W-31-109-ENG-38.

ABSTRACT

Introduction

The need for a Premium Coal Sample Program was expressed on a number of occasions, culminating in the Coal Sample Bank Workshop held March 27 and 28, 1981 in Atlanta, Georgia. Funding was later made available from the Division of Chemical Sciences of the Office of Basic Energy Sciences of the U. S. Department of Energy for the program. The premium coal samples produced from each coal and distributed through this program are chemically and physically as identical as possible, have well-characterized chemical and physical properties, and will be stable over long periods of time. To achieve these goals, coals have been mined, transported, processed into the desired particle and sample sizes, packaged in humid nitrogen environments as free of oxygen as possible, and carefully characterized by a variety of techniques.

One of the reasons for preparing these samples was to permit workers to determine if the apparent difference in results between two laboratories was due to the samples or the experimental technique.

The premium coal reference splits produced from each coal sample and distributed through this program are chemically and physically as identical as possible, have well-characterized chemical and physical properties, and will be stable over long periods of time. To achieve these goals, coal samples were mined, transported, processed into the desired particle and sample sizes, packaged in humid nitrogen environments as free of oxygen as possible, and carefully characterized by a variety of techniques.

Selection of the Coals

The list was then examined to consider the variety of maceral contents which exist in U. S. coal types. There are three important maceral groups, vitrinite, liptinite and inertinite.

With help from the U. S. Geological Survey, individual coals were identified for collection. The compositional characteristics, primarily carbon content, were used to guide the selection of five of the coals. These provided a rank range from lignite through low volatile bituminous. Anthracites were not included because of their low reactivity and relative scarcity. The other three were selected to give a range of sulfur contents, a larger content of macerals other than vitrinite, and a range of bituminous coals with varied paleobotanical source material. Within the compositional constraints from the cluster analysis one of the samples was selected for its known coking properties. A short description of the source of the samples is given in a later section. Table II gives the identity of these samples and some of their characteristics. Samples #1,2,3,5 and 8 were selected to give a range of compositional parameters, primarily carbon, hydrogen and oxygen, which vary with the degree of coalification of the sample. In addition #3 provides a sample with a relatively high sulfur content. Coal #4 was also selected for its known good coking properties, coal #6 was selected for high liptinite and #7 for high sporinite and inertinite contents.

Collection of Mine Samples

For a typical underground sample, an initial meeting with the mine operator both clarified the plan and minimized the time that the coal sample was exposed to the air. On the morning of the collection, the coal bed face was freshly exposed, by cutting back through about 8' of coal, to provide a block of coal the entire thickness of the coal bed, about 12-18" wide and long enough to provide for the 1 ton sample. The bed was exposed with a continuous miner, and a roof bolter immediately followed to secure the seam roof. The loose coal was scraped away from the sample block, and several layers of plastic were placed on the mine floor. After measuring and calculating the amount needed for the sample, the block was marked for removal by the USGS representative. The block was then removed by the collection crew with hand picks from the roof to the floor to provide a channel-type sample. Particular care was taken to avoid contamination of the sample with nearby material. When the bed thickness exceeded four feet (Illinois #6, Pittsburgh, Pocahontas, Blind Canyon and Lewiston-Stockton), stainless steel drums were taken into the mine for sample collection. For the first sample (Upper Freeport) only, double plastic bags were used to transport the coal to the surface. In all cases, the coal samples were loaded into 55-gallon stainless steel drums from 1/2 to 5 hours after collection and then the drums were purged with argon. An additional supply of chunks of coal in a 15 gallon container was obtained from each coal bed for experimenters who need larger pieces for their work.

The subbituminous sample was collected as a 6" core sample from the Wyodak-Anderson seam that was about 120' thick at the sample site. The drilling contractor also obtained two additional 3" cores, one for the USGS, and the other for long-term storage of the sample. All cores were rinsed with distilled, deaerated water immediately after they were released from the core barrel and loaded into the stainless steel drums for purging and shipment.

The lignite sample was obtained as a series of 3" core-type samples drilled through the 18 foot thick seam at about 20' intervals over a freshly exposed top surface. Cold weather limited reactions at the surface. This approach provided channel-type samples representative of a sample area of about one acre. The samples were also quickly loaded into the drums and purged with argon.

Sample Descriptions

The second sample is from the Wyodak-Anderson seam, a subbituminous coal collected about 6 miles northeast of Gillette, (Campbell County) Wyoming in October 1985. The seam was about 120' thick at the collection site. The sample for processing consisted of a 6" core sample through the entire seam.

The third sample is a high-volatile bituminous coal, from the Illinois #6 or Herrin seam, and was collected about 60 miles southeast of St. Louis in St. Clair County in Illinois in December 1985. The 55-gallon drums were taken into the underground mine for this and all subsequent underground mines because the seam was 7' thick at the Illinois site and at least 6' thick for all other underground samples. The original sample was a block which weighed at least 4 tons. The initial 3 tons were mechanically mined (resulting in a higher mineral matter content) and became the Illinois Basin Coal Sample 105 (fifth in a series of samples). This sample was processed in the nitrogen-filled glove box to provide -20 mesh material which was placed in sealed 5 gallon, one-gallon and one-pint (pound) containers. The final ton was hand-picked in the usual fashion for the premium samples. [1].

The fourth sample (high-volatile bituminous) was obtained underground from the Pittsburgh seam (sometimes referred to as Pittsburgh #8) and was collected about 60 miles south of Pittsburgh in Greene County, Pennsylvania in March 1986. This seam was 6' thick at the collection site. The sample drums were taken into the mine for loading.

The fifth sample is a low-volatile bituminous from the Pocahontas #3 seam, collected underground in Buchanan County, Virginia in June 1986. This seam was 6' thick at the collection site and the drums were loaded in the mine.

The sixth sample (high-volatile bituminous) is an underground Blind Canyon seam sample collected in Emery County, Utah about 150 miles southeast of Salt Lake City in August 1986. The seam was about 7' thick at the collection site. Drums were loaded in the mine.

The seventh sample (high-volatile bituminous) is from the Lewiston-Stockton seam of the Kanawha formation. This was collected underground about 20 miles east of Charleston in Kanawha County, West Virginia in October 1986. The seam was about 6' thick at the collection site and the drums were loaded in the mine.

The eighth sample is a lignite from the Beulah-Zap seam collected in Mercer County, North Dakota, about 8 miles northwest of Beulah in November 1986. The seam was about 18' thick at the collection area. Collection was done by accumulating about 50 3" core samples spaced about 20' apart in each direction. Samples were immediately placed in the stainless steel drums.

Purging and Transportation of Mine Samples

All samples were transported by the same truck and driver. The truck was loaded at ANL with stainless steel drums, argon cylinders and tools and then driven to the mine site. The sealed drums were then transported in a refrigerated semitrailer at temperatures of 40-45 degrees F to prevent freezing and limit chemical reactions on the way to the processing facility at ANL. The samples arrived at the laboratory within one-half to two days of the collection and argon purging, and were immediately unloaded.

Preparation of Research Samples

The processing consisted of several steps starting with the crushing and pulverizing the coal to -20 mesh, and mixing the entire ton-sized batch. This batch was then placed into containers. Half of the batch was placed in sealed 5 gallon pails for regrinding to -100 mesh and subsequent packaging. Each half ton was sealed partly (about 80%) in 5 gallon glass carboys for long term storage and the balance in ampules for distribution to researchers. Details of the processing follow.

After the airlock was purged with nitrogen, the drums were opened, and the contents were transferred to a rotating bar crusher (Jacobsen) using a hydraulic drum dumper (Tubar). The sample was reduced to pieces that would pass between bars with 1/2" openings. The flow of crushed coal to the elevator was regulated with an adjustable gate. The sample was raised with a vibrating lift (FMC Syntron) to the pulverizer (Fitzmill model D) where it was initially ground to pass through a 20-mesh screen. With the exception of the lignite, the entire one-ton batch used for processing was ground and accumulated in a 2000-liter mixer-blender (Littleford). After thorough (established by previous testing program) four-minute mixing the sample was transferred by a tubular conveyer (Dynamet) to a second nitrogen-filled enclosure. This tubular conveyor was used for the last 7 samples. Difficulty with transporting fine material on a second vibrating lift necessitated the equipment substitution. About half the batch was transferred to 5-gallon leverlock pails for transfer back to the initial airlock for later repulverizing to -100 mesh. About 80% of the balance was placed into 5-gallon borosilicate glass carboys for long-term storage. About 5,000 ampoules containing 10 grams of -20 mesh samples were sealed with a hydrogen-oxygen torch. The flame was kept at the stoichiometric composition with a mass flow controller (Linde mass flowmeter/flow controller). The half batch that was transferred to the initial airlock was pulverized to pass a 100-mesh screen and accumulated again in the mixer-blender. After mixing, the -100 mesh material was transferred to the second enclosure for placing into 5-gallon borosilicate carboys and into 10,000 ampoules of 5 grams of -100 mesh material. The ampoule filling-sealing system was built by Kuchar Industrial Service and Supply of Joliet, Illinois. The control system was extensively modified by ANL personnel to provide computer control for ease of changing control parameters. Filling of the ampoules was done with a Mateer-Burt system which uses computer control for the amount of pulverized coal dispensed to each ampoule. The carboys were stoppered with special stopcock assemblies to retain the nitrogen atmosphere after removal from the nitrogen-filled enclosure. The ANL glassblower sealed the carboys for long-term storage. Samples were taken during the processing for homogeneity analyses to establish the uniformity of the samples during the processing and sealing.

The North Dakota Beulah-Zap sample was obtained as a series of channel type samples in each drum. The high moisture content of this coal precluded processing through two successive stages of grinding. The moist material would have accumulated on the pulverizer screen and prevented the final grinding. Instead, one half of the batch was processed for -20 mesh material and the other half batch was ground just once to provide the -100 mesh sample. The high moisture content of the Wyodak sample led to limited production of the -100 mesh material and processing of the material from carboys of -20 mesh with a small pulverizer after the lignite sample. This activity resulted in the production of 4,000 of the -100 mesh ampoules.

Additional processing during the summer of 1991 of material in the carboys provided an additional 8,000 of the -100 mesh Wyodak and 2,000 of the -100 mesh Illinois #6 samples.

Sample Storage

Analysis of Samples

Homogeneity Analyses

As a further test of homogeneity the last three irradiated samples were also analyzed for percent S, gross calorific value, moisture, and ash at the ANL Analytical Chemistry Laboratory. The values are given in Table IV.

The samples for the analyses reported in Table IV were not kept in a vessel at constant humidity, resulting in significant differences in the measured moisture content from fresh samples, but were expected to be consistent within the sample set.

The values given for the standard deviation/average value provide a useful measure of the consistency of the homogeneity of the samples for the given analyses.

Proximate and Elemental Analyses

Additional samples were submitted for a range of analyses including: proximate, ultimate, major and minor elements in the ash, heat content, forms of sulfur, and maceral analysis. A large number of laboratories participated in the analytical studies. The analytical data are given in Table V.

The data from the analyses can be used to calculate empirical formulae based on the organic material in the coal and on the basis of 100 carbon atoms per "formula weight". The formulae for the samples in order of increasing H/C are given below:

Integrity of Seals and Long Term Stability

In order to establish the integrity of the seals, additional work is being done at the Analytical Chemistry Laboratory of the ANL which includes the evaluation of the gas atmosphere inside the ampoules initially at approximately six month intervals (3) and later at annual or longer intervals. Particular attention is given to the oxygen content to ascertain potential oxidation. Ampules were placed in an evacuable chamber, evacuated and the glass containers broken to release the contained gases. Until 1991 these gases were analyzed with a CEC mass spectrometer to indicate the content of different gases. Starting in 1991 data were taken with a VG Gas Analysis mass spectrometer model 30-01. Analytical data were given for nitrogen, oxygen, methane, moisture, hydrogen, carbon dioxide, argon and heavier hydrocarbons. Data for carbon monoxide were not given initially because the CEC mass spectrometer cannot differentiate it from nitrogen. The results indicated that the oxygen is always at or below the limits of detection (30-50 ppm). The carbon monoxide is usually at or below the limits of detection (.05%). The carbon dioxide content has increased in the low rank coals and the Pittsburgh as well as the Pocahontas sample over a multiyear period. The methane content has increased for the higher rank coals. The hydrogen content was substantial in the Pittsburgh and Pocahontas samples.

Reproducibility of Analytical Data

Samples are distributed as 5-gram ampoules of -100 mesh or 10- gram ampoules of -20 mesh material.

The importance of sample mixing must be stressed. All coal samples are subject to segregation on standing. It is necessary to remix the sample before opening the ampoule. The recommended method is to turn the ampoule so that the coal alternately falls into the top and then the body of the ampoule. This should be repeated a number of times (at least 10). Rolling the ampoule on its side is not as effective and should not be relied upon for thorough mixing.

The standard deviations and reproducibilities of the analytical data as reported by Commercial Testing and Engineering Company are given in Table VI. These represent the results from evaluation of the data from as many as 60 different laboratories which are part of the Company.

The samples are intended to be used immediately after opening for all work involving air-sensitive properties. Any unused portion should be discarded, as the original properties may change on further storage.

Long Term Supply

Information Dissemination

Microbes in the Samples

Acknowledgments

Appreciation is expressed to Butterworth-Heinemann and to the American Chemical Society for permission to publish tables and graphs in the Analytical Data section.

References

(2) Harvey,R.D.;Kruse,C.W. J. Coal Qual. 1988 7 109

(3) Vorres,K.S. Proc. 1989 Intl. Conf. Coal Science, 1989 II 1083

Additional references from the "Bibliography on Work with the Argonne Premium Coal Samples" (see Section 5) providing general information on the Argonne Premium Coal Sample Program include numbers: 4, 12, 13, 34 (Illinois Basin Coal Sample Program), 43, 62, 75, 76, 77, 78, 79 (model for biomass), 81.

The coal samples are listed in the order collected. The weight % C, H and O values are given on the moisture and ash-free basis, while S and ash are on the dry basis. The data were provided by Commercial Testing and Engineering, Inc., Lombard, IL from the results of analyses at 55 laboratories in their organization.

The calculated uncertainties (in percent) of the weighted average values for the DPM/g were 1.8% or less for counts with multipliers of up to 10**5 and equal to or less than 0.4% for multipliers of 10**6 or 10**7.

__________________________________________________________________________________________________________

Coal          UF         WY         IL        PITT       POC         UT       WV       ND   

___________________________________________________________________________________________

AR H2O       1.13      28.09       7.97       1.65       0.65       4.63       2.42    32.24
AR Ash      13.03       6.31      14.25       9.10       4.74       4.49      19.36     6.59
AR VM       27.14      32.17      36.86      37.20      18.48      43.72      29.44    30.45
AR S         2.29       0.45       4.45       2.15       0.66       0.59       0.69     0.54
AR Btu      13315       8426      10999      13404      14926      13280      11524     7454
                                                                          
Dry Ash     13.18       8.77      15.48       9.25       4.77       4.71         19.84     9.72
Dry VM      27.45      44.73      40.05      37.82      18.60      45.84        30.17    44.94
Dry S        2.32       0.63       4.83       2.19       0.66       0.62          0.71     0.80
Dry Btu     13467      11717      11951      13629      15024      13925         11810    11001
                                                                          
Dry C       74.23      68.43      65.65      75.50      86.71      76.89        66.20    65.85
Dry H        4.08       4.88       4.23       4.83       4.23       5.49        4.21     4.36
Dry N        1.35       1.02       1.16       1.49       1.27       1.50        1.25     1.04
Dry Cl       0.00       0.03       0.05       0.11       0.19       0.03        0.10     0.04
Dry F        below detection levels for all samples
                                                                          
Pyritic S    1.77       0.17       2.81       1.37       0.15       0.24       0.16     0.14
Sulfate S    0.01       0.03       0.01       0.01       0.03       0.03       0.03     0.03
Organic S    0.54       0.43       2.01       0.81       0.48       0.35        0.52     0.63
                                                                          
MAF C       85.50      75.01      77.67      83.20      91.05      80.69        82.58    72.94
MAF H        4.70       5.35       5.00       5.32       4.44       5.76        5.25     4.83
MAF N        1.55       1.12       1.37       1.64       1.33       1.57        1.56     1.15
MAF Org S    0.74       0.47       2.38       0.89       0.50       0.37         0.65     0.70
MAF Cl       0.00       0.03       0.06       0.12       0.20       0.03         0.12     0.04
MAF O        7.51      18.02      13.51       8.83       2.47      11.58        9.83    20.34
MAF Btu     15511      12843      14140      15018      15777      14613        14733    12185
Dmmf C      88.08      76.04      80.73      84.95      91.81      81.32         85.47    74.05
Dmmf H       4.84       5.42       5.20       5.43       4.48       5.81         5.44     4.90
Dmmf N       1.60       1.13       1.43       1.68       1.34       1.59         1.61     1.17
Dmmf Org     0.76       0.48       2.47       0.91       0.51       0.37        0.67     0.71
Dmmf Cl      0.00       0.03       0.06       0.12       0.20       0.03        0.13     0.04
Dmmf O       4.72      16.90      10.11       6.90       1.66      10.88        6.68    19.13
Dmmf Btu    15980      13020      14696      15336      15908      14728        15247    12370

MM(dry) = 1.08 Ash(dry) + 0.55 S (total dry)

Abbreviations: UF = Upper Freeport seam; WY = Wyodak Anderson seam; IL = Illinois #6 seam; PITT = Pittsburgh seam; POC = Pocahontas #3 seam; UT = Blind Canyon seam; WV = Lewiston-Stockton seam; ND = Beulah-Zap seam; AR = As received; VM = volatile matter; MAF = moisture and ash-free; Dmmf = dry, mineral matter free

2Information on the price and availability of the samples may be obtained from the D. L. Vervack, Chemistry Division, Building 200, Argonne National Laboratory at 630-252-3655 or FAX 630-252-9288..

3Individuals who wish to be placed on the mailing list or receive a copy of the Handbook should write the contact the D.L. Vervack, Chemistry Division, Building 200, Argonne National Laboratory at 630-252-3655 or FAX 630-252-9288..