Search Documents

By R. J. Stokes

S. Floreen (international Nickel Co.)— One fairly simple way to differentiate between em brittle me nt due to surface microcracks or due to a dislocation barrier effect might be to load a brittle rock salt crystal in tension to some stress just below the fracture stress and then immerse the loaded crystal in water. If a surface film is retarding plastic flow then one should observe some plastic deformation of the rock salt as

Jan 1, 1962

By L. B. Pankratz

Present-day interest in the rare-earth elements and their compounds, for both metallurgical and ceramic uses, makes it highly desirable to accumulate basic thermodynamic data that can be applied to theoretical considerations of the processing of these substances. Such data still are seriously lacking. This Bureau of Mines paper reports the results of high-temperature heat content measurements of the sesquioxides of four rare-earth elements--europium, gadolinium, neodymium, and samarium--and of the closely related element, yttrium. The temperature ranges covered by the measurements are as follows: Monoclinic Eu2O3, 298°-1,802° K. Cubic Eu203, 298°-1,350° K. Monoclinic Gd203, 298-1,802° K. Cubic Gd203, 298°-1,550° K. Hexagonal Nd203, 298°-1,795° K. Monoclinic Sm203 298°-1,798° K. Cubic Sm203, 298°-1,149° K. Cubic Y203, 298°-1,799° K.

Jan 1, 1962

By Philip E. Niebuhr

A procedure, which can be performed in the field by nonanalytical personnel, is given for sensitive spot tests for detecting tellurium or selenium. The expressed need of a field test for detecting small amounts of tellurium, which may be accompanied by relatively large amounts of selenium, has led to the development of this method. Tellurium search has been impeded by the lack of a satisfactory test from the standpoint of sensitivity and specificity. A quantitative laboratory method exists, but it is expensive and relatively laborious. Several other qualitative tests are known, but they require unstable solutions, which must be freshly prepared for use, and a greater variety of laboratory equipment. The particular advantages of the test described are its simplicity and the stability of the reagents required, there-fore is is well adapted for field use. Also, this test was found to be of practical use in the analytical laboratory when screening samples, thereby reducing analytical costs. The roughly quantitative interpretation of the results can save time by indicating the approximate size of the sample taken for a quantitative determination.

Jan 1, 1962

By A. H. Daane, R. L. Myklebust

The yttrium-manganese system has been investigated by thermal, metallographic, and X-ray diffraction methods. There are three intermetallic compounds present: YMn2 which melts congruently, YMn4, which undergoes syntectic decomposition, and YMn,, which undergoes peritectic decomposition. The compound YMn4 is ferromagnetic at room temperature with a Curie temperature of 214°C. There are eutectics at 25.2, 60.9, and 82.0 wt pct Mn which melt at 878°, 1100°, and 1075°C, respectively. Crys-tallographic data are given for YMn4, and YMn12 . The terminal solid solubilities are low. In a general program of study of yttrium metal in this laboratory some alloy systems of this metal with elements of the first transition period have been examined. This work was originally instigated by experiences in cladding yttrium with jackets of some protective metals such as Inconel for high-temperature service in air.' In some cases a low-melting phase was observed to form between the Inconel and the yttrium resulting in failure of the samples. In characterizing this reaction, a survey was made of the systems of yttrium with chromium, manganese, iron and nickel,, and it was found that a eutectic was formed between these metals and yttrium on the yttrium-rich side of the system. This present study of the Y-Mn system was carried out to examine in more detail the alloying nature of yttrium, and to correlate trends that have been observed in previous related studies. It was observed by Voge13 that in the systems of lanthanum, cerium, and praseodymium with each of the metals in the first transition series, the tendency to form compounds diminished in the order nickel, cobalt, and iron while no compounds were formed with manganese, chromium, or titanium. Beaudry and Daane4 observed a similar behavior in the systems of yttrium with some members of the first transition series except that the tendency for compound formation was greater. Both the Y-Ti5 and Y-Cra systems consist of simple eutectics, while in the Y-Fe,7 Y-CO, and Y-Ni4 systems, there are four, eight, and nine compounds, respectively. In addition to the above trends, the similar atomic radii and electronegativities of yttrium and thorium invite a comparison between their alloying behaviors with a common element such as manganese. In crys- tallographic studies, Florio et al.' have identified three intermediate phases in this system which are ThMn2, Th6Mn23, and ThMn,,. Gschneidner and Waber10 have examined published information on alloy systems of the rare-earth metals and have correlated this information with current alloying theory. From their study, they predicted that the Y-Mn system would contain one intermetallic compound. On the basis of this prediction, the trend in the alloying behavior of yttrium with the elements of the first transition series and the alloying behavior of thorium with manganese, one might expect from one to three intermetallic compounds to form in the Y-Mn system. A consideration of Hume-Rothery's rules of alloying based on size-factor, electronegativity, and valency suggested a small terminal solubility and possible compound formation. The present study was undertaken to confirm these predictions of low terminal solid solubility and compound formation and to establish the general alloying behavior of yttrium with manganese. EXPERIMENTAL Materials. The manganese used in this investigation was obtained from the Foote Mineral Co. as electrolytic plates of 99.9 pct stated purity; the yttrium metal was prepared in this laboratory. Table I gives the analyses of these materials. For the solubility studies at the yttrium-rich end of the alloy system, distilled yttrium, whose major impurity was 200 ppm Ti, was used. Preparation of Alloys. The alloys were formed by comelting the two metals in an are-melting furnace under an atmosphere of argon. The buttons thus formed were inverted and remelted three to five times to promote homogeneity. Due to the high vapor pressure of manganese, it was assumed that the weight lost during are-melting was all manganese. This assumption was based on the very good agreement observed between calculated compositions and chemical analyses of several alloys. The compositions of the dilute alloys used for solid solu-

Jan 1, 1962

By A. R. Taylor

In recent years, thermodynamic data have been used to a greater degree than ever by the chemical and metallurgical engineer as a means of predicting and controlling chemical reactions. To satisfy the spiraling demand for basic physical data needed for the calculations, the Bureau of Mines is studying the heat capacities of a variety of chemical compounds at high and low temperatures. This paper is one of a series presenting such data. The heat capacities of two alkaline earth compounds, strontium bromide and strontium nitrate, were measured in a temperature range of 60° to 300° K. in an adiabatic calorimeter and at temperatures of 300° to about 1,200° K. in a Bunsen ice calorimeter. The construction and operation of the calorimeters are discussed in other reports.3 4 The heat-capacity data presented in this report were extrapolated to 0° K., and smooth values of heat capacity, entropy, enthalpy function, and free-energy function were calculated and tabulated at even 10-degree intervals. The enthalpy data were fitted with algebraic equations by applying the method of least squares; calculations a£ heat-capacity values, entropy, enthalpy function, and free-energy function were made at even 50-degree intervals. The heat and entropy of fusion of strontium bromide were calculated from the enthalpy measurements.

Jan 1, 1962

By Webber. G. R.

Variation in content of some of the major and minor elements in the Lake Dufault granodiorite near Noranda, Quebec has been studied. The element variation and specific gravity data show a pattern consistent with the concept of a concentrically zoned intrusive on the western side of Lake Dufault with a potassium-rich core and a calcium-rich inner zone. The part of the intrusive on the eastern side of Lake Dufault is of more problematical origin, but the suggestion is made that it has resulted from assimilation of volcanic rock by the intrusive. THE purpose of this investigation was to test the application of x-ray fluorescence analysis to the study of variation in chemical com-position of the Lake Dufault granodiorite. The Lake Dufault granodiorite is an intrusive body approximately six miles long by three miles wide located four miles north of Noranda, Quebec. This intrusive is of particular interest because of its -

Jan 1, 1962

By Wesley Holmes

Concentrating, nodulizing, and smelting studies were conducted at the Albany Metallurgy Research Center of the Bureau of Mines to determine whether (1) western iron ores of the type found in the Carter Greek deposit near Dillon, Mont., could be utilized; (2) concentrates of acceptable grade with good iron recovery could be attained by wet-magnetic beneficiation; (3) nodulized concentrates could be smelted with reductants readily available in Montana; (4) iron concentrates could be smelted without prior agglomeration; and (5) high-purity iron could be produced directly from high-grade concentrates, Three small samples and two tonnage samples from the Carter Creek deposit were uniform in chemical composition, and their iron content averaged about 31 percent. They were found to be banded gneisses, and magnetite was the principal iron-bearing mineral.

Jan 1, 1962

By H. Marquis

This paper is concerned with an analysis of a number of failures of large-size, hypervelocity light gas gun chambers and barrels. A brief description of the light gas gun principle and of the various types of guns is first given. The failures and the investigations carried out are then described in detail, with the failures divided into two groups. The first group comprises those failures caused by certain conditions to which the guns may be submitted, and includes, for example, severe local heating of the bore surface causing hardening and cracking of the steel. These conditions were accidental and steps were taken to prevent them. The remaining failures were due to faults in the material or in the heat-treatment, manufacturing or assembling procedures. Specific problems, such as hydrogen embrittlement, and the use of high strength gun barrels above their ductile-brittle transition temperature by preheating prior to firing, are briefly considered. Some interesting observations are also presented on the use of surface markings, and angles of fracture of large fragments, as an

Jan 1, 1962

By Ruth F. Brinkley

This circular describes the activities and publications of the Bureau of Mines, Explosives Research Laboratory4 for the fiscal year 1960 (July 1, 1959 -June 30, 1960). Part 1 describes briefly the principal projects that were active during the report period, and presents some data not scheduled for publication elsewhere in the near future. Certain of the results are only preliminary and should be regarded as tentative, pending the findings of the completed investigation. Publications that appeared during the report period are listed and described briefly in part 2. In general, the research activities reported in these publications antedate fiscal year 1960. The published material relating to activities that are still current is indicated by references in part 1.

Jan 1, 1962

By Clyde L. Brown

The ignition source of fires and explosions in coal mines is sometimes difficult to explain because resulting widespread destruction often erases valuable evidence that might reveal the cause. However, through years of experience, investigators have been able to piece together enough evidence to establish the reasons and, thereby, have been equipped to recommend corrective or preventative measures. The pattern of hazards is not constant. It changes with the development of new practices and new equipment. Important among the new hazards to coal mining are those introduced by electricity. In addition to the dangers of electric shock, fires and explosions from heat and arcs, and premature blasts, certain others have developed as a result of stray electric currents from direct-current traction systems. The conventional direct-current power-distribution system in coal mines is conducive to electrolysis damage to mine tracks, metallic pipelines, grounded power conductors, and structures in a position to become conductors of stray currents. The voltage drop resulting from power demand on intentionally grounded power conductors and accidental grounds on insulated conductors are common sources of stray currents in coal mines. Uncontrolled electric cur-rents flowing through the earth and metallic structures can be a menace, and these contribute to the estimated $6,000,000,000 annual loss caused by corrosion. The purpose of this publication is to discuss some of the electrolysis hazards in the coal-mining industry and to outline corrective measures to alleviate them.

Jan 1, 1962

By Frank A. Peters

This Bureau of Mines report is the second of a series of process evaluations planned to determine the relative processing costs of producing metallurgical-grade alumina from low-grade aluminous materials. Included in these evaluations are recommendations for additional research and development to improve the processes and to lower their costs. The design for each process is based on published work. The processes considered in this report are designed to extract alumina as aluminum chloride from calcined clay with 20 percent hydrochloric acid and to remove the silica residue by filtration. Iron is removed by one of the following processes: Isopropyl ether extraction, in which iron chloride is extracted before decomposing the aluminum chloride; gas precipitation, in which the aluminum chloride is precipitated with hydrogen chloride while leaving the iron chloride in solution; gas precipitation-isopropyl ether extraction, in which the aluminum chloride from precipitation is extracted with isopropyl ether, if necessary; caustic purification, in which a crude alumina is produced and then purified by a modified Bayer operation; or sinter purification, in which a crude alumina is produced and then purified by a lime-soda sinter operation. For plants producing 1,000 tons of alumina a day the estimated operating costs per ton of alumina are: [Isopropyl ether extraction process $ 65.09 Gas precipitation process 89.04 Gas precipitation-isopropyl ether extraction process 119.65 Caustic purification process 73.79 Sinter purification process 80.47] The estimated fixed capital costs per annual ton for these plants on a Marshall and Stevens chemical equipment basis of 240.0 are: [Isopropyl ether extraction process $229.03 Gas precipitation process 287.16 Gas precipitation-isopropyl ether extraction process 356.59 Caustic purification process 266.48 Sinter purification process 279.13]

Jan 1, 1962

By Hsun Hu, R. S. Cline

A detailed studj) of texture formation in 2 pet Al-Fe single crystals with initial orientations of approximately (111) [112], (112) [111], and (112) [111] was made by examining the textures developed on the surface and at various interior sections of the crystal after rolling to various amounts. Depending' upon the initial orientation of the crystal, the surface and interior textures may differ only in sharpness, or may differ essentially in orientation. The orientation of the (111) [112] crystal does not change upon rolling up to 70 pet, but after rolling more than 90 pet a distinct component of (001) (110] orientation is developed. The texture formed in the two (112) (1111 type crystals consists of (111) (1121 and (001) (1101 components. The latter, however, is largely confined to the surface layers. The textures formed on both sides of the crystal are identical, and the texture composition profile is approximately symmetrical with respect to the central section of the strip thickness. The formation of these texture components is analyzed. DURING the past seven or eight years, deformation and recrystallization textures in rolled Si-Fe single crystals have been extensively studied by various investigators.1-7 From these studies, much knowledge on texture formation in bee crystals of various initial orientations was obtained. However, these results also raised many questions, regarding particularly the correlation between deformation and recrystallization textures, as well as the deformation texture itself of some particular crystals. To take a (111) [112] type crystal as an example, it was shown by Dunn and Koh2, 3 that this orientation does not change during deformation by rolling. This finding is consistent with the conclusion reached by Barrett and Leven-son8 with respect to iron crystals that (111) [112] is one of the stable end orientations. However, different recrystallization textures were observed by Dunn and Koh3 in (111) [112] type crystals of different initial thickness, which widened differently during rolling even though their deformation textures were identical. Another interesting example, also from the results of Dunn and Koh2'3 is that of texture formation in a (112) [111] crystal. This crystal developed a two-component deformation texture of (111) [112] plus (001) [110]. Its recrystallization texture, however, was found to be predominantly (110) [001], which is related to the (111) [112] component by a simple [110] rotation. This crystal, therefore, behaved as if the other deformation texture component, (001) [110], were not present. There are also discrepancies among the results of different investigators, as well as numerous unexplained fine features of the deformation texture of crystals of various initial orientations. In order to have a better understanding of all of these points, a detailed study of deformation textures is greatly needed. One of the obvious things that has been completely overlooked in previous texture studies is the possible effect of surface texture. All past work on the texture of Si-Fe single crystals was conducted in a rather simple manner, i.e., the rolled crystals were etched to a very thin sheet; then their textures were examined by transmission X-ray techniques. For recrystallization texture studies, unless precautions are taken by etching off a sufficient amount of the surface layer before annealing, the texture developed in the interior may be affected by the surface layer, which may well have a different initial texture and which may have undergone recrystallization earlier than the interior of the specimen. Because of this effect, it is now planned to reexamine the texture of various rolled and annealed single crystals at the surface, and at various sections below the surface by using reflection techniques. In some cases, both surfaces of the rolled crystal will be examined. In one of Dunn's early papers,9 he noted that at an early stage of rolling the crystal seems to divide roughly through the middle to become a sample which in effect consists of two layers having some what different orientations. This effect has never been further explored. The investigation described in the present paper constitutes the beginning of a series of thorough investigations of texture formation in deformed bee single crystals designed to clear up some of the uncertainties which have been discussed. We have chosen a high-purity 2 pet Al-Fe alloy for this investigation. It is known that the allotropic transformation of iron is eliminated by alloying with approximately 1petAl. Such an alloy, being very similar to silicon ferrite, can be heated to the solidus temperature without phase transformation. One reason for choosing A1-Fe instead of Si-Fe for the present investigation is that we have been able to make single crystals of a vacuum-melted high-purity A1-Fe alloy by the strain-anneal method without much difficulty, but were not successful in doing so with a vacuum-melted high-purity Si-Fe alloy. For many research purposes,

Jan 1, 1962

By R. W. Thompkins

MINE operators are concerned with temperature conditions in their mines, particularly the effects of high temperatures on the l1ealth and efficiency of tl1eir employees. This paper is written to focus attention on the variables of heat transfer between rock and air. It illustrates the difficulties of predicting the manner, and the extent, of heat transfer from theoretical concepts, prior to actual operation of a mine airway. There are ~1any sources of heat in mines. The most important is that of the rock surrounding the airway. This heat, in turn, is a product of the age of the rock, faulting, proximity to recent igneous activity, oxidation of sulphides, hot ground waters, etc. The sources of heat may be constant for any one mine, but they are far from constant when dealing with various mining districts or e1•en with a number of mines in the same district. A second source of heat is that produced by machinery in operation. Practically all the waste

Jan 1, 1962

By M. R. Geer

Efforts to evaluate washery performance are nearly as old as the art of coal cleaning, and in former years there was no uniformity in either the methods used or the basic terminology employed. This situation, and the con- fusion it bred, prevailed until about the time of the First International Coal Preparation Conference in 1950, Largely because of the interest of the French coal research organization Cerchar, the subject of evaluating performance was given prime importance at this conference. As a result, there is now a much greater degree of uniformity in terminology and methods of expressing cleaning results. The International Standards Organization currently is working toward international standards of terminology and methods that should improve the situation even more. The various criteria that are used to evaluate the performance of coal- cleaning equipment fall into three categories, depending upon the extent to which they are dependent on the density composition of the raw coal; some are directly dependent, others are only indirectly dependent, and still others are substantially indenpendent.

Jan 1, 1962

By Bruce I. Thomas

Reconnaissance studies of beach sands along the eastern part of the Gulf of Alaska were conducted by the Bureau of Mines to determine if any of these coastal areas warranted investigations as possible sources of valuable minerals. Because of difficult access and limited transportation facilities only a minimum of sampling, surveying, and camping equipment is utilized to obtain spot samples from key areas as well as information indicating the location, size, and character of the deposits. Such data were obtained from beaches located along approximately 247 miles of the Gulf of Alaska coastline during parts of the 1957 and 1958 field seasons. A total of 201 3-inch-diameter auger holes, spaced roughly at 1 mile intervals, were bored by hand methods to depths ranging from 3 feet to 27 feet; also 33 shovel samples were collected from selected localities. Samples were reduced to a rough concentrate in the field by panning; the rough concentrates were shipped to the Bureau of Mines laboratory at Juneau for further concentration and for petrographic and chemical analyses. Particular attention was given to the heavy-mineral content of the concentrates, although the types and relative amounts of rock-forming minerals were also determined. The reconnaissance indicated that, generally, the valuable mineral content of the beach deposits was too low to be of further interest except in the vicinity of Yakutat and Lituya Bays where erratic but possibly significant concentrations of magnetite and/or ilmenite were found. At Yakutat, about 20 miles of beachline extending southeasterly from Ocean Cape, yielded samplas containing from 1 pound to over 300 pounds of iron and from a trace to over 100 pounds of titanium dioxide (TiO2) per cubic yard of beach material. However, the higher grade samples represented local concentrations of minor importance. The general tenor of the Yakutat beach, as indicated by the airthmetical average of 27 auger-hole samples, is about 35 pounds of iron and 20.5 pounds of titanium dioxide per cubic yard of beach material. Ilmenite is the predominant heavy mineral of the beach deposits in the vicinity of Lituya Bay, but it was found in concentrations only in limited areas near the

Jan 1, 1962

By Joseph M. Singer

This research project had two main objectives: (1) To identify the smoke limits of gaseous fuel-air mixtures and (2) to obtain and evaluate basic chemical information on the formation of smoke in very rich flames near the smoke limit. Such information would do much to establish the circumstances for car-bon deposition in gas appliances and in other types of combustors. Based on tests with ethylene and propane, these studies showed the following: 1. Smoke limits are related to the yellow-tip limits of the fuel. Previous research established sufficient data to permit prediction of yellow-tip limits on gas burners using a wide variety of fuels. The present study of smoke limits deals only with ethylene-air flames, but the concept that was developed is probably applicable to other fuels.

Jan 1, 1962

By William A. Beck

The Construction Aggregates Corp. operates a plant for processing sand and gravel from two deposits in Ottawa County, Mich. The bulk of the sand is obtained from a sand dune on the shore of Lake Michigan, and the gravel is dug from a glacial-drift deposit farther inland. Mining is done with draglines, and the material, both at the digging site and in the plant, is moved mainly by systems of conveyors. The Ferrysburg plant serves customers along the Great Lakes, which is unusual for a business normally local in character. Annual production exceeds 1-1/2 million tons of sand and gravel as concrete aggregate and 200,000 tons of core sand for foundry use.

Jan 1, 1962

By W. A. McKinney

The segregation process consists of heating crushed oxidized or mixed oxide-sulfide copper are with a copper-volatilizing salt and carbonaceous material at about 750° C. to produce metallic copper which is recovered from the furnace calcines by conventional sulfide flotation. The Bureau of Mines made a number of bench-scale laboratory tests on two copper ores to determine the relative merits of using halogen salts, potassium chlorate, or various salt mixtures as volatilizing agents in segregation of the copper. Laboratory investigation of typical siliceous and calcareous oxidized ores revealed that sodium chloride was a better segregating agent than the other salts. Calcium chloride, magnesium chloride, sodium fluoride, or potassium chlorate gave nearly as good segregation as the sodium chloride on the siliceous ore. Segregation of the calcareous ore with calcium chloride was nearly as good as that obtained with sodium chloride, but inferior segregation resulted when magnesium chloride, sodium bromide, or potassium chlorate was employed. The use of sodium bromide or sodium iodide on the siliceous ore, and of sodium fluoride or sodium iodide on the calcareous are, gave appreciably poorer segregation than sodium chloride or the other salts. Other tests of the ores, using mixtures of sodium chloride and the other halogen salts or potassium chlorate, gave segregation approaching that obtained with sodium chloride.

Jan 1, 1962

By John D. Roach

DURING an investigation of the yttrium-hydrogen system aimed at producing solid yttrium hydride specimens containing various amounts of hydrogen, it was observed that yttrium containing approximately 2 wt pct H exhibited a tendency to crack and crumble to a powder on standing in air at room temperature. It was also observed that longer hydriding times, at a given temperature, increased this susceptibility to powdering without an increase in hydrogen content, also that moisture in the air is necessary for the observed effects to occur. Even if cracking did not occur, there was a continual formation of a light gray powder on the as-treated surface of the hydrided yttrium and this reaction continues until the entire hydrided piece has been reduced to a powder. Storing the hydrided specimens in the absence of air or removing the surface either by machining or grinding were effective means of preventing this disintegration of hydrided yttrium. In an attempt to determine the reason for this powdering phenomenon, the surfaces of a number of as-hydrided yttrium specimens were examined by X-ray techniques. In all cases the X-ray pattern obtained showed the major phase to be yttrium hydride (YH2) as would be expected since the specimens contained 2 to 2.2 wt pct H. In some cases a trace of yttrium oxide was observed. There was also a third phase present on the surface of these hydrided specimens which could not immediately be identified. This unknown phase was a face-centered cubic material, NaC1-type structure, with a lattice parameter of 4.855 and a calculated density of 5.914 g per cc. Very slight hand polishing of the surface of the hydrided yttrium specimens completely removed both the unknown phase and the traces of oxide so that only the yttrium hydride pattern remained. Based on the X-ray patterns the quantity of the unknown phase on the surface of the hydrided yttrium appeared to be directly related to the susceptibility of the material to powdering. Work at the Denver Research Institute on the yttrium-nitrogen binary system showed that yttrium nitride (YN) is a face-centered cubic material with a lattice parameter of 4.878A and a density of 5.890 g per cc. They also noted that this compound rapidly disintegrated to a powder on standing in air. The unknown pattern observed in the above specimens corresponds very closely to that of the nitride—fcc structure, 4.885 parameter, and density of 5.914 g per cc. The presence of this thin film of nitride on the surface of the hydrided specimens probably accounts for the observed powder formation and crumbling. The nitride reacts with the water vapor in the air (verified by private communication from Dr. C. Huffine, General Electric Co.) to yield yttrium oxide and ammonia. Ammonia is readily detected when hydrided yttrium specimens were allowed to stand in bottled moist air. The powder formed on the surface of the hydrided specimens was shown by X-ray analysis to be yttrium oxide. This reaction appears to occur primarily at the grain boundaries since discrete particles of yttrium hydride separate from the specimens during this powdering process. The reaction of yttrium nitride with water vapor is believed to be as follows: 2YN + 3H2O - Y203 + 2NH3. Despite the fact that this nitride is present only as an extremely thin surface film, if the above reaction is not prevented by removing this film from the surface either by machining or grinding, the reaction continues until the entire hydrided piece has been reduced to a powder. To account for this continuation of the reaction, it is believed that the following reactionalsooccurs: 2YH2 + 2NH3 -2YN + 5H2. The nitride produced by the latter reaction reacts in turn with water vapor. The reaction therefore becomes autocatalytic and continues until the hydride has been consumed and the entire piece reduced to oxide powder. The amount of nitride required to initiate this reaction is quite small and this nitrogen contamination can occur from a number of sources—hydrogen gas employed, minute leak in the hydriding apparatus or even from degassing of the reaction vessel itself. Longer hydriding times increase the possibility of nitrogen contamination and this is especially true when a dynamic gas system is employed in the hydriding process. The production of stable, solid hydrided yttrium is dependent on the complete absence of nitrogen contamination during processing. If such contamination does occur, powdering of the hydrided product can be prevented by removing the nitride from the surface e.g. grinding or by preventing access of air to the pieces e.g. sealing in wax or plastic. The author wishes to express his appreciation to General Electric Co. for sponsoring this research and for permission to publish the results of work under Subcontract AT-93.

Jan 1, 1962

By Michael G. Zabetakis

Limits of flammability of normal propyl nitrate (NPN) vapor were investigated at ambient and elevated temperatures. Minimum spontaneous ignition and decomposition temperatures were determined with laboratory apparatus over the pressure range from 0 to 4,000 p.s.i.g. Flame speeds and the detonation susceptibility of NPN vapor were measured at elevated temperatures. Stable detonation was obtained with liquid NPN at elevated temperatures, and the detonation velocity was determined to be between 4,700 and 5,100 m./sec.

Jan 1, 1962