Search Documents

By G. D. Criswell, F. Palkovits, J. M. Johnson

A paste technology investigation was completed for non-pozzolanic fly ash from Units 3 & 4 of PPL Montana’s Colstrip Steam Electric Station. The purpose was to assess the feasibility of converting from dilute slurry deposition of ash products to a system using thickened or paste ash to reduce excess water storage in the impoundment, reduce seepage losses, and increase storage efficiency. The investigation included laboratory and field testing to characterize the rheological, geotechnical and dewatering properties of the ash products, an assessment of potential seepage reductions and an evaluation of potential for increased ash storage capacity by modifying impoundment operating procedures. The studies concluded that the conversion project was feasible, cost effective, and permittable. Critical facilities for the conversion project are currently under construction.

Jan 1, 2004

By P. D. Roa, O. Noirot, H. Owens, D. E. Walsh, J. R. Mokka

Over a period of eight years (1989 through 1997) the Mineral Industry Research Laboratory (MIRL) at the University of Alaska, Fairbanks, AK, has investigated the effects of temperature, residence time and particle size on the degree of low-rank coal (LRC) upgrading for a nonevaporative, hydrothermal drying process, also known as hot water drying (HWD). The following three Alaskan LRCs were used in this study: Usibelli coal (Seam No. 4) from the Nenana Coal Field, Little Tonzona coal from near McGrath, and Beluga coal from the Beluga-Yentna field on the west side of Cook Inlet. Replicated factorial tests were conducted. Experimental results indicate that most of the low-rank coal upgrading via HWD occurs rapidly within the first 10 to 20 min of residence time. Thereafter, the upgrading process slows down, and more than 100 min of additional residence time is required to produce percentage increases equivalent to what was achieved in the first 10 to 20 min. The properties of the hot water dried products were very sensitive to process temperature. Calorific value, carbon and oxygen contents, equilibrium moisture levels and Hardgrove gindabilities all appear to change linearly with temperature within the 275°C to 325°C (525°F to 61 5 OF) range.

Jan 1, 2000

By Michael E. Salovich

The U. S. Department of Interior, Bureau of Mines (Bureau) has developed a prototype version of a computer application that combines hydrologic analysis with object-oriented graphics. This software is unique in that it presents analytic element analysis within an object-oriented CAD environment. Its user interface employs mouse handling, scrolling windows, dialog boxes, palettes, and pulldown menus. It can import scanned images of maps and overlay these images within the same window. This application was developed using object-oriented programming techniques. Several programming issues concerning graphics had to be resolved during development of this system. These include (1) how to represent hydrologic elements as graphical objects, (2) how to translate screen coordinates to map coordinates, (3) how to implement ?zooming? (changing the magnification level of a window), (4) how to overlay layers of images, and (5) how to import scanned images of maps. An object library was used to implement graphical user interface features such as scrolling windows and pull-down menus. Algorithms for CAD features such as map coordinate translation, zooming, and the layering of images were developed by Bureau researchers. The application has a modular source code structure and can be modified for use in other research applications.

Jan 1, 1992

By Charles N. Alpers, Jane M. Hammarstrom, Robert R. II Seal, Nora K. Foley

Geoenvironmental models are compilations of geologic, geochemical, and hydrologic information that describe both the pre- and post-mining environmental signatures of mineral deposits. The environmental behavior of seafloor massive sulfide deposits varies according to deposit type. This study has divided massive sulfide deposits into six categories: (1) Noranda-type, (2) Kuroko-type, (3) Cyprus- type, (4) Bathurst-type, (5) Besshi-type, and (6) sedimentary-exhalative deposits. In addition to Fe and Al, the heavy metals most commonly associated with drainage from these deposits are Cu, Zn, Cd, Pb, and As. Metal ratios of mine drainage and natural background waters tend to reflect the primary character of the ores. Variations in pyrite and pyrrhotite contents are reflected in lower pH values associated with pyrite-rich deposit types.

Jan 1, 2000

By Cheryl M. McInnes, Graham J. Sparrow, James T. Woodcock

Three composites, assaying about 3 g/t Au and 0.75% Cu, were investigated. Each contained oxide copper minerals, iron oxides, and silicates. One composite also contained copper and iron sulphides. Maximum gold extractions by thiourea leaching were obtained after copper removal by acid pre¬leaching at pH 1.5 and redox potentials of 400 mV (Pt versus SCE). Best thiourea leaching conditions were pH 1.5, redox potentials over 200 mV, and 0.4% thiourea. This gave 90% gold extraction in 2 h with thiourea consumptions of 25 kg/t. Lower potentials gave poorer extractions. Ferrous iron additions before leaching halved thiourea consumptions. Longer leach times and the presence of sulphide minerals sometimes resulted in reprecipitation of gold.

Jan 1, 1989

By Harry W. Barr

Judging by the inquiries that come to Battelle, a number of companies have run into problems in determining the size and character of markets for mineral fillers in general and specialty fillers in particular. Usually these problems arise during a project to develop a new mineral filler when management suddenly realizes that a considerable sum of money will have to be spent before any profits are realized from sales of the new filler. What management needs in this case is the odds for or against the success of the development. And these odds frequently are controlled by the markets for the new filler. Basically, the problems involved in gaging markets can be simplified to how much of this new filler can be used and by whom? But there are some potential booby traps in these simple questions that should be exposed. This paper will discuss a few of these problems such as: 1. Describing the filler so that the prospective customer recognizes its possible applicability. 2. Finding; a prospective customer who can verify the filler's applicability. 3. Expanding successful test results to total industry markets. 4. Appraising competitive products, and 6. Discovering all the possible uses for the new filler.

Jan 1, 1961

By Lynn I. Kelley

Industrial-mineral production contributes on the order of CDN$1.5 billion per year to the economy of Saskatchewan, Canada. Potash is the primary industrial mineral produced in the province, followed by aggregate, sodium sulphate, salt, potassium sulphate, peat, clays, silica sand, calcium-chloride brine, and "clinker." Potentially economic deposits of kaolin, diamond, and building stone are under evaluation. All of the major deposits of commodities discussed here are located south of lat 56°30'N. Most of the locations are presented on Figure 1 POTASH Saskatchewan supplies approximately 30% of the world's demand for potash. Eight conventional underground mines and two solution mines produced an all-time high of more than 13 million metric tonnes of potash (KC1) (8 million tonnes K2O equivalent) in 1997, valued at over CDN$1.4 billion. All of the mines exploit the middle Devonian Prairie Evaporite. The four potash-bearing members of the Prairie Evaporite underlie much of southern Saskatchewan and extend southward into North Dakota and Montana and east- ward into Manitoba (Fig. 2). Halite and anhydrite separate the potassic members. The evaporite beds locally are interrupted by solution structures (anomalous ground). Some of the structures may be related to deep-seated tectonic features.

Jan 1, 1999

By Michel Grenier, David A. Young, Brent Rubeli, Mahe Gangal

This work forms part of a study being conducted into the feasibility of constructing a prototype Load Haul Dump (LHD) vehicle for operation in underground Canadian mines. A series of in-laboratory assessments were performed to determine the relative contributions of a Deutz F4L912W (54 HP- conventional vehicle engine) and a Deutz F2L 2011 (31 HP hybrid vehicle engine) to diesel emissions concentration during mine transient duty cycles and steady-state operating modes respectively. Results indicate significant reductions in the quantity of gaseous and particulate emissions with the application of the hybrid system and a high-efficiency ceramic element filter. This paper describes the engine test protocols and the results of this laboratory study.

Jan 1, 2006

By G. G. O. O. Uwadiale, M. A. Kollere, M. A. U. Nwoke

Birnin Gwari and Alawa graphite deposits were upgraded by flotation. Using 0.5 kg/t kerosene and 0.25 kg/t sodium silicate, flotation concentrates grading 90.7% and 88% Cg with recoveries of 86.2% and 84% were produced from Birnin Gwari and Alawa graphite samples, respectively. The combination of 0.5 kg/t kerosene, 0.2 kg/t pine oil and 0.25 kg/t sodium silicate yielded high-grade graphite (92.3% Cg) with a recovery of 83.9%. Flotation with refined rubber nut oil and kerosene, at a 1:1 ratio, produced a concentrate grade of 86.3% Cg with a recovery of 90.4%. With acid leaching of the flotation concentrate, a high-grade graphite concentrate containing 98% Cg was produced.

Jan 1, 1998

By Richard L. Ranich, Frank F. Aplan, J. B. Hiskey, Robert S. Thompson, B. J. Scheiner

Acidic waters typified by acid mine drainage, copper dump leach solutions and steel pickle liquors will typically contain SO4-2 and cations such as Fe+3, AI+3 and for the pickle liquors, sometimes Cr+3. These acidic waters will often react with limestone, present naturally or introduced for purposes of solution neutralization, to form reaction - inhibiting coatings of gypsum and gelatinous, metal hydroxide precipi¬tates. The coating thickness, as a function of pH, was quantified for both reactive (calcite, limestone) and non-reactive (quartz, barite, fluorite) minerals, and the data were related to the hydroxy species diagrams and to solution turbidity measurements. Coatings were found on the reactive minerals at a lower bulk solution pH than on the non-reactive minerals.

Jan 1, 1993

By W. M. Ma, Daniel W. H. Su, K. Centofanti, Yi Luo, W. L. Zhong, Syd S. Peng

5.1 INTRODUCTION There are several types of surface ground disturbances, each of which is immediately transmitted to surface structures and causes damage of various forms, depending on the location of the structures in the subsidence basin. Many criteria have also been proposed for defining the potential structural damage. All of these topics will be discussed in this chapter. 5.2 TYPES OF GROUND DAMAGES There are two types of ground damage, continuous and dis¬continuous, that could inflict damages to surface structures. Dis¬continuous ground damage is open fractures of various degrees. Four types of fractures are commonly seen (Brauner, 1973b): 1. Fissures or cracks. Cracks are the most common form of ground damage and range from completely closed to several feet wide. Surface cracks are most prominent when the ratio of mining depth to mining height is smaller than 30 to 50 times mining height. Cracks initiate parallel to, but ahead of the moving long¬wall face at some intervals when the depth-to-mining height ratio is less than 25 to 30. As the face continues to move, they open up and then close gradually, and in most cases eventually close up completely when the face moves farther away. Sometimes when the depth-to-mining height ratio is larger, cracks may initiate be¬hind the face and go through the same cycle. Cracks are also more likely to occur in the hill when the face is moving in the uphill direction. Cracks also occur at or near both sides of the panel edges. Those cracks tend to remain open permanently. When the surface is covered with grass or other vegetation, cracks of less than 1 to 2 in. wide usually do not penetrate it and become invisible on the surface. 2. Steps. A step may occur when the longwall face stops abruptly if the mining depth-to-mining height ratio is less than 20. Steps may also occur along the panel edges. Steps also range from a few inches to a few feet in width and depth. If planes of weakness are oriented favorably, ground will slide along these planes and form steps on the surface. 3. Cave-in pits. These result from local collapse of the ground from the mine level to the surface when the mining depth is less than 150 ft and occur mostly above abandoned mined lands. They may be as large as 40 ft wide and deep. 4. Bumps (compression ridges). Bumps result from ground compression and occur mostly near panel center some time after mining. Bumps of up to 4 ft, but mostly 1 to 2 ft, have been identified. Continuous ground damage refers to the formation of the sub¬sidence basin where continuous nondestructive ground deforma¬tions occur. As mentioned earlier, continuous ground deforma¬tions associated with a subsidence basin include subsidence, slope, curvature, displacement, horizontal strain, twisting, and shearing. 5.3 TYPES OF SURFACE STRUCTURAL DAMAGES Both continuous and discontinuous ground disturbance cause damage to surface structures. A. Structural Damages Due to Discontinuous Ground Disturbance 1. Fissures or cracks. Open cracks can cause dislocation of surface structure if the gaps are sufficiently large so that the struc¬tural elements cannot resist but fail. On the other hand, minor cracks affect only the appearance, and not the structural integrity. 2. Steps. When a surface structure is located on a step, some portions of the structure lose support and become overhung. In the case of a residential house, there will likely be many open frac¬tures in the overhung portions, followed in severe cases by sepa¬ration of the superstructure at the edge of the step. 3. Cave-in pits. If the caved-in pit is larger than the surface structure, the structure will drop into the pit and become unstable. If the caved-in pit is smaller than the surface structure, the struc¬ture may either become overhung or lose support in some portions. In either case, it is unsafe for use. 4. Bumps (compression ridges). Bumps will disrupt roadways or house floor. If located under supporting members of the struc¬tures, it may cause damages. B. Structural Damage Due to Continuous Ground Disturbances 1. Subsidence. Pure subsidence or uniform subsidence over an area does not cause any structural damage. However, if it becomes a topographical low area, it may be flooded with water and make the surface structures unusable. It must be noted that the flat bottom area at the center of the final supercritical subsidence basin does not reach that position suddenly. Rather, as shown in Fig. 3.1, it is subjected to the dynamic surface deformation asso¬ciated with the traveling subsidence basin before it is left behind to be part of the flat-bottom area. Therefore any structure located at the center of the final supercritical subsidence basin must be able to resist the dynamic surface deformation in order not to be dam¬aged. 2. Slope. Ground slope will induce tilting of the structure. Tall structures with small base areas, such as water towers, chimneys, and power transmission towers are most sensitive to ground slope. If the ground slope is such that if after tilting the center of the gravity of the tilted structure falls outside the base area, it will topple, but this seldom occurs. Ground slope will change the designed slope of a buried trans¬portation pipeline, either increasing or decreasing the original slope depending on whether the ground slope coincides with or opposes the original slope. There are maximum allowable final slopes for various types of transporting materials. Ground slope also affects highways and railways. There is a maximum allowable final slope for a railway because tractive effort increases 1/1000 of its weight for every increase of 1/1000 in grade. 3. Curvature. There are positive (convex) and negative (con¬cave) curvatures. When a surface structure is located on a positive curvature, both ends become unsupported, followed by bending upward at the midportion. The top portion is subjected to more severe stretching than the lower one. As a result, vertical cracks with a wider gap at the top occur at the window, door frames, on the walls, and at -the wall joints. V-shaped cracks appear at the

Jan 1, 1992

By Vytas Svalbonas

Sophisticated computer analysis methods such as finite elements, finite differences, and numerical integration have become tools in the mineral processing industry for design of large mill shells. While use of such methods allows the analyst-designer to take new freedoms with his designed structure, there are also pitfalls and dangers which must be recognized for the proper use of such sophisticated tools. One of the worst of these is treating any computer answers as "gospel truth". The present paper discusses the knowledge required by the analyst-designer for proper use of computer analysis, as well as explaining to the layman how to identify questionable areas in such structural analyses.

Jan 1, 1979

By Ora H. Rostad

The sequence of events preceeding the discovery of the Mount Emmons molybdenite include some inadvertent near misses by several groups. The events are given in general chronological order to bring out the essentially contemporaneous, but different activities by several groups in two areas on opposite sides of Mount Emmons. The Mount Emmons molybdenite deposit lies in the Elk Mountain mining district. The west edge of the deposit is near the boundary between the Elk Mountain and Ruby mining districts as shown by the USGS map included (see Fig. 1). Considerable history regarding the Keystone silver-lead-zinc mine is given because of its proximity to the Mount Emmons molybdenite deposit which is on the other side of the mountain from Redwell Basin and where other significant activities took place. It should be noted that there was no documentary evidence available to me that would indicate that molybdenite was recognized to be present on, or under, any part of Mount Emmons, including Redwell Basin, before 1970. Veins containing silver-lead-zinc minerals were discovered in the Elk Mountain and Ruby mining districts about 1872. Some discoveries on the south flank of Mount Emmons about 1880 were developed into the Keystone mine. The Keystone has the largest cumulative production of metal ores of any of the mines in these districts. Most of the 413 Mt (455,473 st) of ore produced from the Keystone mine through August 1967 was produced after 1950. The ore averaged 175.4 gt (5.64 oz per ton) Ag, 2.62% Pb, 5.56% Zn, and 0.72% Cu.

Jan 1, 1991

By Mortimer J. Richardson

Dredging can be traced to early times where it may have been derived by the Phoenicians prior to the 4th century B.C. The first tangible design drawings of a bucket ladder (BL) dredge can be found in the collections of Leonardo da Vinci (1452-1519), which he produced when he was attempting to create better methods of channel and port dredging. The perfection of the system did not commence until after the invention of the steam engine in the early 1800's; providing the impetus to progress from man-powered mud mills, horse drawn along polders or dikes in the Netherlands, England, France and Germany, to steam-driven. In the 1860's following the various gold rushes in the U.S., Australia, New Zealand, Siberia and South America, with placers in particular, efforts were begun to apply mechanization to achieve deeper digging. There is a recording of a dredge being installed to mine gold in Montana in the 1860's but did not include a processing system. Then in the 1880's in New Zealand, some practical engineers developed a system that is said to be the first bucket ladder mining (BL/M) dredge for mining gold placers. In studies of the New Zealand gold dredging experience, it is estimated that some 500 dredges were built to mine gold between the 1880's and 1942. A careful analysis of the data, however, shows that very few were profitable. The technology appeared to stagnate in the early part of the century. Two of the BL/M dredge design engineers left New

Jan 1, 1992

By John W. Rushton

THE PLANNED MAINTENANCE PROGRAM All concentrators should start with a planned maintenance program from day one. Every- body thinks they have a program in place before start-up, but in reality, most have only a fraction of what is needed to avoid the "Post Start-up Disaster" (PSD) - two years for start-up is not that uncommon. An effective program has all of the following: (1) Good people. (2) Area Maintenance. (3) Good PM Checklists. (4) Daily Work Schedules. (5) Written Work Orders. (6) Equipment Histories. (7) Backlog Control. The installation of a workable program takes hard work, talent, and commitment. You have to know what you are doing (hard work will not replace talent). Good People Good people are available in any location if you make the effort. Always check references carefully (both superiors, peers and subordinates). Was the last concentrator they worked at successful? Was it clean? Find out! Success or failure follows people around. I know people who are experts on PSD because they have lived through it three or four times. You need some college trained people with hands-on experience. There are no secrets, you have to make the effort. If you make a mistake in hiring (and you probably will), make the tough decision necessary to rectify it quickly. Area Maintenance Area maintenance refers to organization of supervision such that one person is uniquely accountable for every inch of ground and piece of equipment. Avoid overlapping authority if at all possible. You must have a planner as a staff position. A foreman cannot effectively plan and

Jan 1, 1986

By R. C. Bunge

Many minerals float best with fatty acids in the intermediate pH range. For example, fluorite, a semi soluble salt mineral, exhibits a maximum contact angle and a maximum flotation recovery in the pH range of 6-8 after conditioning in a palmitate solution (fig. 1). The maximum in flotation coincides roughly, as many researchers have observed, with the boundary of bulk precipitation of the fatty acid. According to fig. Z this boundary is in the above case of 5xl0-7 M palmitate approximately at pH 6.3. For assessing the phenomena occurring at the bulk precipitation boundary of fatty acids, the surface tensions of carboxylate solutions have been measured under various conditions. Fig. 3 shows the surface tension development of palmitate and myristate solutions of increasing concentration at pH 12. Also shown are the bulk precipitation edges (BPE) of the respective solid species that is stable under these conditions, that is the potassium acid soap, HKAZ. Both curves display a similar behavior. As the concentration of carboxylat,e increases, the surface tension decreases, until a minimum is reached at which the curves level off. The decrease of surface tension with increasing carboxylate concentration is generally taken to indicate the surface excess of the carboxylate ion. The point where the curve suddenly levels off has not yet been explained. While it is well known that the onset of micellization (CMC) may cause such an inflexion of the surface tension curve, this cannot be the case here because the position of this point depends, as shown in fig. 4 for palmitate, on pH (the CMC is generally independent of pH). We believe that this point indicates the precipitation of a colloidal acid soap (colloidal precipitation edge, CPE) before the bulk precipitation of this species sets in (BPE).

Jan 1, 1994

By Stephen G. Hardcastle

3D-CANVENT is a software package that integrates advanced computer aided design (ACRD) true 3D graphics with a mine ventilation simulator. The package runs as a Windows(tm) application to access its printer drivers environment and does not need third party CAD software. It is composed of two primary modules: DMVENT and MINEDESIGNER. DMVENT is a traditional Fortran coded Hardy-Cross iterative ventilation network solver written in 1980 with thermodynamic capabilities. This module is relatively unchanged with the traditional data input options for branch type, specified or calculated resistances, fixed flows, and fixed or variable pressure fans. MINEDESIGNER is the graphics engine that optimizes the ventilation design process. It performs the front-end transformation of input data entered in the graphical interface into the correct format for the solver. At the back-end it reconverts the historically standard tabular data output from the solver into an easily viewed graphical format. ACRD features of MINEDESIGNER are used to generate a 3D wire-frame node and branch network of the mine's ventilation system. The network can be displayed in up to 4 views orientated to XYZ planes or a 3D view. All the views have zoom, pan, slice and rotate options. The graphical interface efficiently permits data entry and editing via a mouse with pick-and-point item selection. Branches can be found or added with "search" and "join" options. Visual interpretation is enhanced by the 16 colour options for branches and numerous graphical attributes. Network locations are readily identified by alpha-numeric names for branches, junctions and fans, and also the logical numbering of junctions. The program is also readily expandable for pollutant simulation and control/monitoring applications.

Jan 1, 1995

By C. E. Tackett

The Greenwich Collieries Co. is a subsidiary of the Pennsylvania Mines Corporation, and a wholly owned subsidiary of the PP&L Co. which serves the northeastern portion of Pa. PP&L annually consumes about 11 million tons of coal per year and is the largest consumer of both bituminous and anthracite coal on the eastern seaboard. Development work for Greenwich started in 1969. The coarse coal plant was constructed in 1970, with the raw fines bypassed. This worked very well for our requirements at that time; however, the Greenwich mining plans were extensively changed, longwall mining units were installed and low coal was encountered, resulting in dirtier coal, a finer coal and an ever increasing ash in the shipped product. All this demanded a more efficient ash removal process for the fine coal fraction. In 1971, the Pennsylvania Department of Environmental Resources and the Federal Environmental Protection Agency were pushing for the now stringent sulphur emission limits at our power plants. The Greenwich Mine coal reserves contained large quantities of 3.5 to 4% sulphur along with some low sulphur reserves. It was, therefore, essential that any new plant have the capability of efficient sulphur removal to meet the Pennsylvania Department of Environmental Resources and the Federal Environmental Protection Agency required limits. To meet these limits our design would have to be capable of producing a product of 2.0% sulphur or less.

Jan 1, 1977

By W. A. Zingg

Impervious zones in copper waste heaps create a problem in contacting the rock with leaching solutions. These zones are normally associated with compacted layers .formed on the surface of each lift by dump truck traffic and by disintegration of the rock exposed to atmospheric conditions. These layers are usually only a few feet in depth--but, nevertheless, act as barriers to vertical flow causing the leach solution to perch in the heap, Consequently, much of the rock remains untouched and the leach solution breaks out the side of the heap instead of continuing downward to bedrock. One solution to this problem is breaking the hard pans with explosive charges. However, to use explosives means adopting new explosive materials and methods. To use this process successfully requires finding answers to operating conditions inherent to waste heaps. Foremost among these are: (1) type and size of casing that can be used because of limitations brought about by drilling problems, caving and, corrosive conditions; (2) perforating the casing at required levels; (3) an explosive capable of being pumped; (4) obtaining detonation of the, main charges through, the wail of the casing; (5) an explosive with the ability to propagate through the voids of the rubble and one which is not affected by water, acidic solutions, or back pressure conditions from hydrostatic heads; (6) to heave and rupture the compacted zones without chambering or recompressing the waste material.

Jan 1, 1970

By R. Woods

INTRODUCTION Gaudin (1), considered that "the mechanism of mineral collection is the central problem of flotation theory". From their work on adsorption phenomena in flotation for more than fifty years (2), Gaudin and his co-workers have made major contributions to the elucidation of this mechanism, a fact acknowledged by their contemporaries in the field (e.g. (3)). Their investigations in a number of areas, e.g. demonstrating that submonolayer quantities of adsorbed collector species can induce flotation, or identifying the influence of substances which can activate or depress (1), have helped to provide the basis for and the development of a theory to account for the interaction between mineral and collector. It was established in the late fifties (4) that the presence of oxygen is a prerequisite for the flotation of metal sulfides with thiol collectors. This fact rendered inadequate the concept of simple ion adsorption being responsible for rendering these minerals hydrophobic. The requirement for oxygen has been explained (1) in terms of this species being a component in chemical reactions between the sulfide surface and the collector. Gaudin has been instrumental in identifying the products of chemical reactions at sulfide minerals (1.5), a line of work which has been continued particularly by Finkelstein and co-workers in South Africa (6). In presenting a detailed mechanism of the mineral-collector- oxygen interaction, it has often been considered that the mineral surface - must first be oxidized to a metal oxy-sulfur species which

Jan 1, 1976