Search Documents

This treatise presents the application of the vertical furnace in a carbon regeneration system. Comparison is made to a rotary kiln. The construction, operation and control of the vertical furnace and system components are discussed. Advantages include high efficiency, fuel economy, minimal mechanical apparatus and maintenance, atmospheric and temperature control, ease of operation and reliability.

Jan 1, 1985

By P. J. Stewart, S. Tracy

This presentation describes the operation and features of the WCC Carbon Reactivation Kiln. Particular emphasis is placed on minimizing carbon losses, corrosion resistance, fuel efficiency and the fate of mercury during reactivation in the multiple-tube vertical retort. Detailed laboratory work indicates that there is complete vaporization of mercury in the kiln, no refluxing in the upper sections and distinct advantages in using this type of kiln when sizing collection equipment.

Jan 1, 1993

By A. H. Reitsema, Hans E. van Dam

In CIP, CIL, and Heap Leach gold recovery operations, a sub-optimal activated carbon performance invariably causes a significant loss of revenues. In this paper it is shown, how dedicated analyses of carbon samples can be used as a guide in plant performance optimiza­tion. Firstly, a brief outline is given of the most important methods of analyses, including the specialized methods not normally used by mining operations, but provided by carbon manufacturers. Then, a number of practical cases are discussed in which carbon sample analyses provided insight which allowed significant improvements.

Jan 1, 1995

By Tom Bruington

This paper discusses the accuracy of determining the quantities of gold and silver on activated carbon in various stages of process for typical precious metals flowsheets. Accurate plant performance evaluation depends upon reliable estimates of carbon and precious metals inventories. The sources of sampling error are identified and methods of reducing them discussed. Several case studies illustrating the sensitivity of flowsheet performance to sampling error are presented.

Jan 1, 1991

By A. F. Raschenko, V. V. Elschin, Y. E. Golodnov

Kinetics and mechanism of gold and silver sorption from thiocarbamide solutions by activated carbon were studies using tne method of equilibrium-kinetic analysis. The intradiffusonal character of the process was determined and the kinetic and thermodynamic constants were calculated. The electrostatic component of the adsorption interaction was estimated by the electro - chemical investigations. The optimal parameters of the desorption of noble metals thiocarbamide complexes were determined. The intensification of gold and silver desorption process is reached through the use of activating additions to the eluent and electro - chemical polarization of activated carbon. Methods for complex and selective extractions of gold and silver from thiocarbamide solutions and pulps by activated carbons were developed.

Jan 1, 1995

By S. Zaman, John Bulter, M. E. Defoe, Franklin T. Davis

This section deals with the carbon minerals and materials: graphite, tar sands, gilsonite, and diamond. as well as sulfur and the sulfides of mercury, arsenic, antimony, and bismuth. Table I contains the production figures for each of these commodities in the United States and in the world. Specifications for concentrates are also summarized as are prices for concentrates at the time the section was written. In each chapter an attempt has teen made to describe the more common commercial mineral involved with the commodity, the processes used to recover and concentrate the minerals, as well as geographical areas where these processes are utilized.

Jan 1, 1985

By Xuehui Zhang, Yun Bai, Qiaosong Li, Huiming Wu

"Construction activities account for a large share of the global Greenhouse Gas emission. In the past 10 years, project decision-makers and researchers have been paying more attention to the carbon emission of tunnel projects. In this paper we study the carbon emission calculation in shield tunneling. The objective of this paper is to: (1) analyze the complex variables which affect the total emission of shield tunneling; (2) develop and illustrate a new data-mining-based method to estimate the carbon emission in the construction phase; (3) assess the emission prediction reliability of a data-mining-based method—the Randomforst model, with R programming language. This study is based on a large-diameter shield tunnel built in Shanghai, China, and shows that a reliable predictive performance can be achieved with this new data-mining-based method. INTRODUCTION Climate change caused by the Greenhouse Gas (GHG) emission has been a hot topic in recent 20 years. Both governments and non-government-organizations are trying its best to control the climate change. Since the 1990s, the general consensus is that carbon emission reduction and carbon sinking (the work of capturing the emitted carbon) are two ways of mitigating the global warming effect, with the former being more exercisable. The challenge posed by global warming also motivates the project decision-makers to take measures to reduce the GHG emission. The 5th Assessment & Report on Climate Change (published by IPCC, the Intergovernmental Panel on Climate Change) suggests that building industry contributes to 5% of the total direct emission. The building industry has shown more awareness of the impact of GHG emissions with construction activities. Shield tunnels act as an important component of the infrastructure system, and have been the dominant way of utilizing the underground space, as well as a competitive alternative to passageway underneath rivers. To study the environmental effects of shield tunneling, there is a need to quantify the carbon emission of shield tunnel construction."

Jan 1, 2016

By J. J. Komadina

Since the advent of resin-in-pulp (RIP) technology for treatment of uranium-bearing ores in the 1950?s to current carbon-in-pulp (CIP) processing of gold values, metallurgies and operators alike have wrestled with the problem of interstage screening. Early gold CIP plants such as at Homestake utilized externally mounted vibrating screens in conjunction with air lifts to effect carbon-pulp disengagement. Other operations world-wide have likewise adopted this technique with varying degrees of success. Subsequent development of in-tank launder screens both in South Africa and the United States have been of the air cleaned variety (EPAC)

Jan 1, 1988

What is the preferred electrolytic cell design and why? Hall: Considerable research remains to be done on electrolytic cell design. Studies are presently being conducted to determine optimum voltage and amperage. Zadra cells, which were used in some of the first carbon plants, are efficient when properly installed. Gold and silver are deposited on large-capacity, cylindrical steel wool cathodes. At times high value sludge falls from the cathode to the bottom of the cell. Rectangular cells, which were developed at some smaller operations in Nevada, require less floor space and have some desirable features in that cathodes can be moved and sludge can be removed from cell bottoms without actually shutting down the circuit. In South Africa two types of cells using graphite and diaphragm configurations are being tested. Duncan: The rectangular cell has one problem it's easy for shortcircuiting to occur. If you don't have a good seal on the side of the tank and on the bottom, the solution is going to want to flow around the bottom and sides of the cathodes and you're going to get poor performance in the cell. That is the advantage the Zadra cell has over the rectangular cell. With good sealing, which is a must, you should be able to get a ratio of the value of the solution going in to the value of the solution coming out of 10 to 1. Ken, can you tell me what you're attaining on the Zadra? Hall: At the Homestake gold operation in Lead, the cathode in the No. 1 electrowinning cell is loaded to about 600 ounces when it is moved to the refinery. At this time the No. 2 cathode is moved to the No. 1 cell. The No. 3 cathode is moved to the No. 2 cell and a cathode with new steel wool is placed in the No. 3 cell. In rectangular cells this sometimes is not possible because the loading on the cathode in the last cell may be as high as that in the No. 1 cell. As Don has pointed out this can probably be prevented by good seal design between cathodes. Potter: At the risk of sounding as though I'm straddling the fence, I really believe that either the round cell, as developed by Jack Zadra, or the rectangular cell will do an acceptable job. The Zadra is a fine unit. Leaks are not unknown but there are not very many. However, the rectangular cell is very saving of floor space. Plastics have been greatly improved during the past year or two and there is a very efficient design for the rectangular cell. It would have six cathodes in it and contain about 25 ft of steel wool. The body of this rectangular cell would be reinforced fiberglass plastic but with an integral lining of a high temperature plastic. I think either cell is so efficient and reliable that I would almost be tempted to flip a coin depending on floor space requirements. Either cell can be used to pull a loaded cathode and to advance the lesser loaded cathodes. Don, what is the configuration of your cell? Duncan: The Pinson rectangular cell has dimensions 2 ft wide, 30 in deep, and 10 ft long-long enough to take a dozen cathodes and anodes. We are currently using six of each but are not up to full capacity as yet. It's a steel tank coated with rubber. We have copper bus bars, which brings to mind that the last cathode in the tank, in one instance, was loaded heavier than the first one. That was probably a function of poor contact between cathodes and the bus bar. You can have erosion of the bus bar and if you don't get current flowing into the anode and out of the cathode, the cells will not function properly. Kappes: Just a few comments on cell design. In an alcohol stripping circuit we are running in Beatty„ NV, we are using PVC pipes for all circuitry piping and a commercial polypropylene electrolytic cell. They seem to be working quite nicely. The cell itself has eight cathodes measuring 18 in2. At six gallons per minute flow into the cell, at an average gold content of about 500 ppm, we seem to get good loading on the first three to four cathodes and not much on the end cathodes. At 12 gallons a minute flow we seem to be seeing gold uniformly along all eight cathodes. We are running at quite a high amperage level, about 400 amps into that cell. Gene McClelland, US Bureau of Mines, Reno Research Center (from the floor): I was associated with a company where the packing density of their steel wool cathodes was much too high and created some short-circuiting problems in their electrowinning circuit. I would like to ask the panel if they can recommend a packing density for cathodes in pounds of steel wool per cubic foot. Hall: The weight of steel wool put on one of our cathodes is 10 lb and the dimensions of the cathode are such that density would be about 0.75 lb/ft3. Duncan: I made a rough calculation and I'm estimating about 2 lb/ft3. Potter: In checking over a number of electrowinning operations, I found apparently very satisfactory operation within the limits that have been mentioned. I have come to a figure of about 0.5-1.0 lb of medium grade steel wool per cubic foot packed as uniformly as possible, of course. One consideration in packing is the fact that there should be good contact between the steel wool and the conductor. Operators tend to compact the steel wool more than would be desirable just to try to maintain good contact.

Jan 10, 1981

A panel discussion on carbon-in-pulp processing of gold and silver ores was one of the highlights of the 1981 AIME Annual Meeting in Chicago. The session generated considerable interest and discussion among panelists and the audience. For those unable to attend the panel, the program was recorded and the discussion appears in this two-part report. The panel was co-chaired by Robert S. Shoemaker, vice president of San Francisco Mining Associates, and Laurence D. Hartzog, principal engineer with Bechtel Civil and Minerals Inc., San Francisco, CA. Panel members included: George Potter, consultant, Tucson, AZ; Kenneth B. Hall, metallurgical superintendent, Homestake Mining Co., Lead, SD; and Donald M. Duncan, resident manager, Pinson Mining Co., Winnemucca, NV. There have been several types of carbon-in-pulp adsorption vessels in use, such as the Dorr-type agitator, the simple propeller agitated tank, the Pachuca tank, and, lately, the draft tube-type of agitated tank. Which one of these is best and why? Potter: On the selection of the most appropriate adsorption vessel, there is a considerable difference of opinion that stems largely from the fact that every ore is different. The mesh of grind, the pulp solids, and the apparent viscosity of the pulp as caused by its clay content and chemical conditions all differ. The traditional Dorr agitator with the slow speed center sweep and either peripheral or center column air lifts has worked quite well on the finer grinds of all minus 65 and probably 70-80% minus 200 mesh. Pachuca tanks have also been successful and they may be capable of handling a coarser feed than the traditional Dorr tank. One uranium mill, for example, handles minus 28 mesh sandstone ore in a Pachuca. The most recent development and one that commands consideration is the draft tube agitator in which there is a turbine which closely fits inside a draft tube. Velocity in the tube is carefully calculated to avoid undue shear and thus abrasion of the carbon. The objective in all cases, of course, is to mix the granular carbon, which is typically 6 x 16 mesh, very gently but thoroughly in a slurry with minimum carbon abrasion. I do not know that there is one outstanding choice just yet and I have been unable so far to get information on C-I-P service for ore grinds as coarse as 35 mesh. Duncan: The major advantage of draft tube-type is the ease of startup. At the Pinson plant we installed draft tube-type agitators but it is too early to quote experience with them. We also haven't operated long enough to determine just what our carbon loss is. The advantage, of course, in the draft tube design is that it requires only about one-third the horsepower input of a conventional agitator. If it has no other advantages, it has that. Hall: The type of vessel most suitable for C-I-P adsorption depends on the type of ore treated and the prevailing operating conditions. In most cases, a deep tank with turbine-type mechanical agitator and low speed tip velocities would be satisfactory. Turbine-type impellers give a positive type of agitation which assures optimum ion contact and reduces short circuiting. Thorough aeration is possible with an air sparge properly located. A mechanical agitator can easily be started up after an outage, but it is usually necessary to drain and wash out a Dorr rake-type or Pachuca before restarting. The turbine-type impeller requires more power, but maintenance costs are negligible if rubber covered impellers are used. Carbon losses are minimal. In smaller plants, Pachuca type agitators provide adequate aeration and agitation. Bob Polak, Occidental Minerals (from the floor): Mr. Hall, you favor the mechanical-type agitator with the preface that the proper design is critical. Could you elaborate on this for us? Hall: The most important thing is low tip speed to prevent carbon attrition. I think the impeller should be sized so that you get a good sweeping action toward the bottom of the tank, across its bottom, up the sides, and back to the center. The advantage of the draft tube is that you get a more positive agitation and you probably get improved aeration too. Polak: Do you know of anyone using an upflow mechanical agitator rather than a downflow unit? Hall: I think that Bob Wilson at Custom Equipment has designed one where he has located the sparge directly beneath the impeller. The air bubbles are broken up by the impeller as the slurry passes through it. The slurry flows upward, outward, down the sides, and, again, back to the center of the tank. I don't know that any tanks of this design are in commercial use. Hans Von Michaelis, Randol International (from the floor): A question to any of the panelists on flat bottom versus conical bottom Pachucas. Hall: I would say the conical bottom would be most suitable in most cases. Some plants have experienced problems with flat bottom Pachucas in that they have a tendency to sand in on the sides of the tank so only the center of the tank is active. Duncan: With our draft tube tank we placed an inverted cone in the center of the tank bottom and blanked off the bottom comer around the circumference of the tank to facilitate movement of the pulp. Other than that, as far as conical bottom tanks are concerned, I'm generally opposed to them because of the cost and height differential. Larry Kramer, Kennecott Minerals Co. (from the floor): Mr. Duncan, you mentioned that the draft tube-type could be agitated with about one-third the horse-power applied to a conventional propellor agitated tank. I have trouble trying to pin down that kind of number. Could you give a bit of rationale as to why that mechanism has a lower horsepower requirement? Duncan: I think it has to do with the fact that below the propeller you have straightening vanes. The pulp, which is flowing vertically downward, is turned and flows upward again without any recirculation. You have an inherently less horse-power requirement in that type of tank. The one-third horsepower requirement is a number I obtained from Lightnin, which supplied us with details of the draft tube. There is probably much more recirculation with a conventional impeller type that is wasted motion. Potter: Last week I saw some agitators in South Africa with short draft tubes and actual turbine-type impellers. This plant was handling 200 kt of ore per month. The tanks were flat bottomed, and the agitation appeared to be quite satisfactory.

Jan 8, 1981

By Peter Bush

The deposition of metastable carbonate minerals in ar. aid, sheltered, coastal environment, leads to the formation of a flat, coastel, sabkha plain. The ground waters of the sabkha plain are modified by evaporation, precipitation of salts and reaction between ions in solution and the host carbonates. The resulting brine is similar to those found in fluid inclusions associated with Mississippi Valley type deposits. The aragonite and high magnesium calcite found in the sediments can concentrate lead and zinc from sea water. On diagenesis the lead and zinc are released into the sabkha brines.

Jan 1, 1979

By G. Robert Ganis

The folded Appalachians are a physiographic province of eastern North America extending from Nova Scotia to Alabama. The province strikes through the state of Pennsylvania (see Fig. 1) comprising roughly 20 percent of the state and it is here that the focus of this paper will attend. The province is a thick sedimentary sequence which was compressed into a folded mountain belt by the Appalachian Orogeny which terminated at the end of the Permian. The stratigraphy contains both carbonate and non-carbonate rocks as a reflection of the depositional environment at any specific time and place during the Paleozoic geosynclinal accumulation. Although a great deal of variety and difference exists in the geology from one end to another, a common thread of structural and stratigraphic style exists. Carbonate rocks, limestones and dolomites, comprise a substantial percentage of the total Appalachian stratigraphy and are interbedded with clastic materials such as sandstones, siltstones, and shales. The carbonate rocks are extensively mined and collectively represent a vast economic resource.

Jan 1, 1983

By Dennis C. Drehmel

Because of their relatively low cost, carbonate rocks are desirable in abatement of air pollution by acid gases. Although many processes and types of equipment have been proposed to control or prevent the emission of acid gases, this paper considers only three: dry-limestone injection, wet-limestone scrubbing, and chemically active fluid beds. For the dry-limestone-injection process, surface area-especially from pores in the 0. 2- to 0.3-micron range-is one of the important parameters which will discriminate between less and more reactive carbonate rocks. Marls, chalks, oolitic calcites, and other fine- grained limestones show larger-than-average surface areas and better- than-average removal efficiencies. Inert chemical impurities such as silica detract from reactivity. A dolomite is not necessarily less re- active than a calcite. For the wet-limestone-scrubbing process, dissolution rate of the carbonate rock controls the differences in stone reactivity. In this process the dolomites are less reactive than the calcites. In addition, more than 90 percent of the dolomitic impurity in Fredonia white lime- stone is inert during limestone scrubbing. Dissolution rate is also a function of particle size for large-grained carbonate rocks. Fine grinding of such rocks will increase the dissolution rate to that of a fine-grained rock. Hence the requirement of high dissolution rates may be met by some grinding of fine-grained material (e. g., marl, chalk, or other limestone) or by fine grinding of coarser grained material. For chemically active fluid beds, chalks have the best capacity to absorb both SO2 and H2S. However, most chalks are probably unsuitable for fluid beds because of high attrition rates. Because dolomites are generally more reactive than calcites with both SO2 and H2S, a dolomite would be the feed material of choice for a chemically active fluid bed.

Jan 1, 1974

By N. A. Abdel-Khalek

The amenability of carbonate separation from a sedimentary phosphate ore using a bioflotation process was studied. Two types of bacteria were chosen for this study. The experiments were performed using oleic acid as collector. The flotation experiments were carried out using statistical designs for optimizing the main operating parameters. A concentrate containing 0.78% MgO and 30.15% P2O5 with a recovery of 92.31% was obtained from a feed containing about 2.45% MgO and 27% P2O5. These specifications could not be obtained using conventional flotation experiments under similar conditions in the absence of bacteria. Measurements of zeta potential, adsorption of bacteria and collector, as well as froth power, were performed to explain the role of microorganisms in the surface modification of both dolomite and phosphate and in their interaction with the collector.

Jan 1, 2009

By S. R. Titley

Carbonate-hosted massive sulfide (60%) ores of the western Cordillera compose a distinct class of deposits although they vary widely in detail and perhaps genesis. They have been important sources of lead, zinc, copper, silver, gold, and more rarely, molybdenum, tin, and tungsten in complex mineral assemblages usually dominated by pyrite. They occur as concordant stratiform bodies, as discordant mantos, and as chimneys crossing tens of meters of section; many are associated with faults and breccia bodies. Environments of formation of carbonate hosts range from craton basin to shelf with ages from Proterozoic to Cretaceous; in many instances, dating has shown mineralization to be significantly younger than hosts. Temperatures of formation determined for many deposits are high (200°-400°) and many districts are centers of igneous activity manifested by dikes, sills, and stocks. However, direct cause and effect relationships have been difficult to establish. Many important problems attend the renewed interest and study of these deposits and include questions of metal-sulfur sources, metallization processes and histories, geological controls at all scales, and relationships to broader patterns of metallogeny.

Jan 1, 1985

By J. Hanna

The University of Alabama Mineral Resources Institute (MRI) has developed a unique process for selective fatty acid flotation of carbon-ate gangue from sedimentary apatites "francolite" in the pH range of 4-6 without collector conditioning of the pulp. The process, which does not require an apatite depressant, was modified to beneficiate a high Mgo siliceous phosphate matrix from south Florida. The modified process includes two flotation stages, one for carbonate removal, as mentioned above, and the other for phosphate recovery from the siliceous gangue. The selective flotation separation of the carbonate and phosphate, from siliceous constituents of the matrix at various pH levels and collector dosages are discussed. In the carbonate flotation stage, about 70% of the Mgo was rejected in the froth and the phosphate flotation stage produced concentrates analyzing about 31% P2O5,0.7% HgO and 4% insol. The overall P2O5 recovery in the two stages is about 85%.

Jan 1, 1989

By Purnima A. Shivdasan

The fluorspar deposits at Okorusu, Namibia are closely related to an alkaline igneous ring complex of late Cretaceous age (125±7 Ma). The fluorite ores consist of relatively fine-grained purple replacement fluorite and subsequent coarser grained purple and green vug-filling fluorite crystals. Homogenization temperature measurements for 38 primary fluid inclusions shows that main stage purple to green fluorite crystals were deposited over a temperature range from 166°C to 144°C, and minor late yellow fluorite was deposited at lower temperatures from 132°C to 128'C. Homogenization and freezing temperatures measured for 18 primary fluid inclusions shows that the fluorite-depositing fluids were less saline than MVT ore fluids and overlap the lower temperature range of fluids that deposited epithermal ore deposits. Mining in two open pits (A and B) at Okorusu has gradually provided exposures which reveal that the fluorite orebodies have been emplaced primarily by the preferential replacement of carbonatites. Evidence for the replacement of carbonatite includes: 1) replacement remnants of carbonatite within the fluorite ores, 2) goethite pseudomorphs after the carbonatite minerals, pyroxene, pyrrhotite, and magnetite that are disseminated in the fluorite ores, 3) partial replacement by fluorite of a titaniferous magnetite rimmed carbonatite dike in the A pit, and 4) partial replacement by fluorite of a large sill-like carbonatite body in the B pit. Electron microprobe analyses of minerals in unreplaced portions of the carbonatites show that their mineralogy consists of titaniferous magnetite, diopside pyroxene, apatite, biotite, and calcite. Magnetite contains 6.97% to 10.46% Ti02, averages 8.44% Ti02, and has an average V203 content of 0.35%. At high magnifications under the SEM, the magnetite shows abundant fine ulvöspinel exsolution lamellae arranged in a cloth texture. Apatite contains moderately high fluorine contents that range 3.21 to 5.0% and average 4.2%, and apatite crystals range in total rare earth oxides from 1.7% to 2.61%. Pyroxenes in the carbonatites are iron-rich diopside, and they contrast to pyroxenes in the fenites that are aegirine-augite. All of the microprobe analyses for mica crystals in the carbonatites give biotite compositions. The fluorite ores also have replaced marble locally in the A pit, especially along the A band in the B pit, and dominantly in the orebody proposed for the new C pit. Carbonatite and marble were preferentially replaced by fluorite because the calcite was readily soluble in the presence of the fluorite-depositing fluids and the dissolution of calcite provided calcium for the formation of fluorite.

Jan 1, 2005

By D. P. Gold

Major sources and reserves of Cb/Nb, Ta, REE, Y, Sc, Zr, Hf, Cu, apatite and vermiculite may be concentrated in primary and secondary minerals associated with different stages and depths of emplacement in carbonatite complexes and kindred alkalic intrusive rocks. Exploration targets include: apatite and pyrochlore (Nb) in the plutonic and hypabyssal core and ring-dikes: bastnaesite (REE), strontianite, barite and fluorite in hydrothermal veins and cone-sheets (subvolcanic setting): thortveitite (Sc,Y) and vermiculite in metasomatized aureoles (fenites): high grade eluvial deposits of apatite and pyrochlore, and secondary REE, Nb, Ti. and V-minerals in supergene and weathered zones.

Jan 1, 1991

By R. Spears, J. B. Vincent, S. Donald, J. Litaker

Previous efforts to remove and concentrate heavy-metal ions from wastewater streams via metalloprotein affinity metal chromatography (MAMC) were limited to the metal transport protein transferrin. Extension of the technique to the enzymes carbonic anhydrase and carboxypeptidase A, which possess different metal affinity properties, is described. Preliminary results indicate that the technique is capable of selectively removing and separating late first-row transition metal ions from aqueous solution. The results suggest that by properly choosing the matrix, i.e., protein or enzyme, MAMC may he applicable to problems involving a wide range of heavy-metal-hearing aqueous solutions.

Jan 1, 1996

By C. A. Pickles, D. Mallette

Silicon nitride (Si3N4 ) is an advanced structural ceramic materia1 which is presently being considered for a number of potential engineering applications, particularly in heat engines, gas turbines, and electronics. With respect to other structural ceramics, silicon nitride has relatively low thermal conductivity and thermal expansion coefficients and a moderate elastic modulus. Thus, during thermal cycling the stresses developed in this material are lower than in other structural ceramics. At the present time silicon nitride is produced via three main processes: thermal decomposition of silicon imide or amide, treatment of silicon tetrachloride or silane vapour with ammonia and nitriding of silicon powder. These three methods of production are discussed with particular attention to the effect of morphology and phase content on the properties of the final product. In this work carbothermal production of silicon nitride has been studied in a high temperature resistance furnace. Compacted briquettes of silica and carbon were produced which were reacted in a nitrogen atmosphere. The optimum reaction temperature and the carbon to silica ratio were found to be 1425OC and ten to one, respectively. The presence of large amounts of ash in the reductant promoted the formation of silicon nitride. The specific surface area of the briquettes did not have a large effect over the range studied in the present work. Briquette seeding with small amounts of Si3N4 was effective in increasing the reaction rates.

Jan 1, 1990