Search Documents

By Sabert Oglesby, Albert L. Thomas

Until recently, probably the best means of sampling airborne dusts has been the impinger method. Dust-laden air is drawn into a sampling tube, and the particulate matters separated from the air and collected in the liquid in the impinger. A sample of the suspension is prepared on a slide and counted through a projection microscope. This method is simple and reliable, but it does suffer drawbacks. Since the collected particles must be counted visually, results of the dust count are taken. This seriously impedes studying the effectiveness of dust inhibitors, such as water sprays, or making adjustments in equipment. Greatest disadvantage of all, however, is the tedious task of counting particles in the sample. With the growing desirability of taking more samples, the effort involved in making these counts is a real factor in the cost of dust control. Finally, it is generally recognized that impinger dust counts can be related to health hazard only for specific kinds of dust and under similar conditions. This is because the rather complex mechanism of dust damage depends so much on size distribution of the dust. Counts determined by impinger samples cover a limited size range that does not necessarily correlate with retention by the lungs and upper respiratory tract.

Jan 3, 1959

By E. W. Greig

This paper is an account of exploration for bauxite in the Amazon Basin, Brazil, and particularly of the discovery, exploration and development of the Trombetas deposits by Alcan. A general report of the Trombetas Project has already been published in E/MJ(Anon. 1975) to which the reader is referred for details on ownership, infrastructure and construction as well as the proposal for mining and treating the bauxite. A regional account of bauxite in and around the Guyana Shield and the northern part of the Brazilian Shield is included to help explain the geology of the Trombetas deposits and their relation to other bauxites in northern South America. It is suggested that all these deposits formed during Tertiary time under similar geologic conditions.

Jan 1, 1979

By N. S. Kenny, A. A. Burr

THE purpose of this note is to present evidence that vacancies, which have been produced by quenching from a high temperature, alter the rate of strain-hardening of an hep single crystal. The majority of investigations of the effects of quenched-in vacancies on the yield stress of metals have been made using fee samples in the form of thin wires. However, there is an inherent disadvantage in using such wires to study the effects of vacancies on the rate of strain-hardening of metals because fee wires begin stage II slip either immediately upon yielding or very shortly thereafter. When stage II slip occurs, very rapid strain hardening also occurs, masking the effects of vacancies on the rate of hardening. In this investigation single crystals of magnesium hep were studied since magnesium has only one slip plane at low temperatures and has, therefore, an extensive easy glide region. In addition, the effect of both static and dynamic recovery on the rate of hardening were eliminated by testing in liquid nitrogen. Both forms of recovery occur in hep metals by dislocation climb, which is a diffusion controlled mechanism and is negligible at liquid nitrogen temperatures. Finally, the dependence of the strain-hardening rate on the crystal orientation was eliminated by using the equations of Schmid and Boas.' Variations in the strain-hardening rate could, therefore, be attributed to the presence of vacancies. Single crystals in the form of rods approximately 5 mm in diam and 6 cm long were grown by a modified Bridgman technique from commercial high-purity magnesium (99.97 pet) supplied by the Dow Chemical

Jan 1, 1962

By C. E. Arnold

A BRIEF description of this installation was included in a recent paper by H. Kenyon Burch.1 The purpose of the present paper is to amplify Mr. Burch's description, as it is felt by the writer that the subject merits more than passing attention. In making a comparison between the common method of hoisting men in mine shafts, and elevator practice in high buildings, the former shows to great disadvantage both as regards safety measures and smoothness of operation. It is surprising that in view of the inducements offered, the step of applying the building equipment direct to the mine shaft has not been taken before, as in many respects the conditions of service are similar, and the objects to be attained are identical. Even without doing this, it is certain that safety to life and limb in the usual method of hoisting men underground can be greatly enhanced by the application to the mine hoist, of some of the features of the building elevator. In the case under consideration, the elevator cage runs in a vertical, concrete-lined shaft compartment 5 ft. 11 in. by 5 ft. 6 in. (1.8 by 1.68:m.) in section and 585 ft. (178 m.) deep, while the counterweight runs in the pipe and ladder compartment of a twin shaft 102 ft. (31 m.) distant. The section of the division of this compartment occupied by the counterweight measures 2 ft. 4 ¼ , in. by 3 ft. 10 ½ in. (0.72 by 1.18 m.), and the counterweight is made in the form of a small cage which. suffices to carry two men for the purpose of making inspections or repairs in the pipe and ladder way.

Jan 12, 1917

J. L. McK. YARDLEY,* Pittsburgh, Pa. (written dlscussion ?) .-It is interesting to observe how closely Mr. Hansen agrees with other investigators to the effect that the art of electrolytic zinc has left the realm of mystery. In his introduction, lie has said practically what Thomas French said in his paper "The Future of Electrolytic Zinc. "1 The successful electrolysis of zinc from sulphate solutions has not been an easy matter, and it is only within recent years that the difficult problem of depositing high-grade zinc has been satisfactorily: accomplished on a commercial basis. The principal requisite is. that the electrolyte shall be quite free from certain impurities. The methods by which freedom from these impurities is assured are now very well understood, and with experienced superintendents there is little difficulty in obtaining a high efficiency in that part of the process. When certain underlying principles are recognized, the difficulty encountered in dealing with the filtration of the acid and slimy solutions also largely disappears, although it has been a very serious one to the uninitiated. In the dissolving of the zinc from the roasted ore, there is nothing which any competent chemical' engineer cannot undertake, and any other operations connected with the electrolytic process are either of little or no difficulty, or are common to the retort process. Mr. Hansen also agrees with R. G. Hall's statement in " Some Economic Factors in the Production of Electrolytic Zinc:"2

Jan 10, 1918

By Clark B. Risler, Walter E. Thomas

MILL planning must include electrical drives and a system to supply them. These should be considered at the time metallurgical and mechanical plans are being made. Because it is convenient, flexible, reliable, and economical, electricity is the accepted source for powering mill concentrating equipment. Ore-treating mills have grown to the point where a single mill has been built using 360 motors ranging from 1 to 1000 hp and totaling 50,000 hp. Such a system of electrical drives can be scattered over a square mile and employ 4000 to 220 v. Drive Motor Characteristics: Since ore processing is predominantly a continuous operation, very few of its drives require wide speed variation after initial adjustment. Consequently squirrel cage, wound rotor, and synchronous alternating current motors are most often used.

Jan 5, 1957

By Robert J. Willard

The scanning electron microscope (SEM), became commercially available in 1966. Embodying some unique features not provided by conventional electron microscopes, this new electron-optical instrument offers the scientific community a new way of observing the surfaces of materials. In rock mechanics research, the SEM should prove a useful basic tool in characterizing rock fracture on a very small scale and thus contribute to a better understanding of rock behavior at larger scales. This article presents one example of how the SEM is being applied to such research at the Twin Cities Mining Research Center operated by the U.S. Bureau of Mines in Minneapolis. This research center conducts theoretical and experimental studies on a broad range of mechanisms that induce disintegration by the application of energy to rock.1 Some of these studies are concerned with the relation of microstructure to rock behavior and are carried out mainly with the petrographic microscope. At the same time, however, the center recognizes that this relation must be studied through several orders of magnitude, from lattice-size units of rock-forming minerals to field-size rock units. The SEM fits this scheme by extending the capability of the petrographic microscope into smaller orders of magnitude. However, because its resolution is limited to about 200 A at 10,000x, the SEM is not capable of resolving lattice-size units, and hence a conventional electron microscope with a much greater resolution capability is necessary at this order of magnitude.

Jan 6, 1969

The Library of the above-named Societies is open from 9 A.M. to 10 P.M. on all week-days, except holidays, from September 1 to June 30, and from 9 A.M. to 6 P.M. during July and August. The Library contains about 55,000 volumes, including sets of technical periodicals and the publications of scientific and technical societies. Members of the Institute, with few exceptions, are forced to spend a portion of their time in localities isolated from sources of information. To these the Library can render valuable service through correspondence; letters requesting information will receive special attention. The Library is prepared to furnish references and copies of articles on mining and metallurgical subjects; to determine the existence of mining maps, and to furnish general information as to the geology and mineral resources of all countries. All communications should be made as definite as possible so that the information received. may be what is desired and not include collateral matter which may not be of interest. The time spent in searching for such collateral matter will be saved, and the information will be sent more promptly and in more usable shape. The members of the Institute can be of service to the Library by forwarding copies of mining reports, maps privately issued, and similar material, which will be classified, indexed, and made available to other members. Suggestions for additions to the Library, either by purchase or personal solicitation as gifts, will be welcomed. It is hoped that members while in the city will use the Library freely, and assurance is given that most careful service will be rendered to them.

Jan 3, 1915

By W. H. Hoffman

During the autumn of 1891, several heavy blasts (each throwing down some 3000 tons of ore) were made on what is known as the Theall side of these mines. Immediately after one of these blasts, the writer found that at least one-third of the ore was largely impregnated with pyrites, and besides this, fully 15 per cent. of the entire blast was composed of irregular lumps of magnetic pyrites. This was a revelation wholly unexpected, as previous examinations by many experts had failed to disclose any such quantity of sulphury ore. Over the center of the Theall tunnel there suddenly appeared a perfect keystone, as it were, of pyrites, 3 feet wide at the arch of the tunnel, 13 feet ride at the top, and 20 feet high. This keystone pinched out in February, 1892; but in that month large blocks of pyrites were more plentiful on or near the western wall, and these deposits extended the entire height of the workings, some 43 feet. The second blast in November, in the face and over, and on each side of the tunnel, was far more surprising. The whole cross-section it exposed was thickly interspersed with layers of magnetic pyrites, ranging in thickness from 10 to 24 inches, and quite irregular in form, while all of the ore carried much more than the usual amount of sulphur in the non-magnetic pyrite. One hundred and fifty-five tons from a portion of this breast (western side), showed the following composition upon sampling and analysis: Per cent. Metallic iron,...45.080 Phosphorus,..........304 Sulphur,...........23.510 And the balance of the blast showed: Per cent. Metallic iron,.....• 33.210 Phosphorus,...........349 Sulphur,.......... 6.680 vol. xxI.-33

Jan 1, 1893

By C. K. McArthur

The subject of classification is so broad that this discussion is confined to what the author believes is of prime importance in connection with proper grinding and classification. The years past have witnessed many noteworthy changes in fine grinding and classification, which have been accompanied by a readjustment of ore-dressing flowsheets for the preparation of better ground and/or sized products at an expenditure of less power, steel and maintenance than formerly. Nowhere in the industry is more being done in the way of experimental work to lower costs than in grinding circuits, because, at best, grinding costs are high. Recent years have indicated that, on the average, finer grinding has a real application. Without careful preparation of the process feed, no operation can be highly efficient or successful. In an earlier paper, the findings of an extensive survey at representative concentrators were presented1. Also, in the year of 1935, a technical discussion was given by Dr. J. V. N. Dorr (Some Notes on Classification Trends in America) before the Chemical, Metallurgical and Mining Society of South Africa In both papers mentioned, much has been said regarding flowsheets, grinding and classification. The author is very familiar with these papers, having had considerable to do with the actual operation of the flowsheets, as well as large mills and classifiers in the work described. Some data from the second paper have been included in the following discussion. Circulating Load In this particular paper it seems advisable to deal with the classification phases of the work that to a major extent govern the success of any closed-circuit grinding unit. The question has often been asked in connection with a closed circuit: "What circulating-load ratio should be carried?" This is really an

Jan 1, 1939

By John W. Cobb

The attempts of British investigators to arrive at definite knowledge concerning the nature of the nitrogenous constituents of coal have been mainly made through studies of the behavior of coal on carbonization and gasification. There have been carried out numerous determinitions of the distribution of nitrogen in the products of carbonization under laboratory and large-scale conditions; and by a careful consideration of the extent to which nitrogen is liberated, say as ammonia,. at different temperatures and under different conditions of operation as regards atmosphere, it has been possible to form some general notions as to the nature of the nitrogenous compounds present in the coal; and at the same time to arrive at clearer ideas as to what are the important operative factors in determining the ammonia yield under working conditions. This work has been spread over many years and is summarized below. The difficulties of connecting the results in such a way as to allow of the nitrogenous constituents of the coal being described in terms of specific organic structure are great. They arise, for example, in the carbonization process, because the phenomena of decomposition observed at any one temperature can only refer to compounds decomposing at that temperature, each of them with a previous history of chemical transformations occurring at lower temperatures. To put it in &other way, the compounds occurring in the coke at any stage are not in the least likely to be those present in the original coal. The same difficulty, in a milder form, occurs when the method of differential solution is attempted. There is always the preliminary question, almost impossible to answer, as to how far the separation by the use of the solvent has involved some form of decomposition of the original compounds. In the following summary, an attempt has been made to bring together results obtained by British workers.

Jan 1, 1925

By Michel Daghert, Michel David

In the exploitation of large porphyry type deposits where a cut-off has to be applied and where "internal waste" exists, planning people ask the geologist for the estimated grade of small blocks. Comparison with actual production figures and blast hole assays shows that any kind of weighted average of neighboring DDH (diamond drill hole) samples provides an excessingly smooth picture of reality, resulting most of the time in overevaluation of tons of ore. Solutions based on examination of Past production data exist14 but it would be necessary to have a solution which can be applied before production starts as the preceeding problem also results in an overestimation of the total reserves, often by as much as 20%, thus changing the mine life accordingly and destroying any "optimum planning" which could have been done. The proposed alternative is to forecast the grade distribution of small blocks in a larger block, sufficiently well documented. Several Possible ways to estimate these distributions are reviewed and emphasis is placed on recently developed methods (source G. Matheron) where the coefficients of the orthogonal polynomial series expansion of the distribution of any size of block are directly estimated by kriging. The efficiency of the technique is tested on a level of the Cyprus Pima open pit where the predicted recovery is compared to actual production figures.

Jan 1, 1980

By Erwin Zodrow, De Verle P. Harris

The Smallwood mine, owned and operated by the Iron Ore Co. of Canada, is one of the largest ore producers in Canada, producing about 15 million tons of crude ore per year. Located in the Labrador Trough near, Labrador City, Newfoundland, the open-pit operation has ore consisting chiefly of specular-hematite, with some magnetite. The magnetite content of the ore as a mineral percentage is small, but it is of considerable importance in ore grading because of its effect upon milling. Being harder than specular-hematite, magnetite increases grinding time and milling costs. Specifically, higher grade magnetite (greater than 15%) usually occurs as fine grains in a fine grained siliceous rock matrix (also as larger octahedra in carbonate bands). This tends to increase the overall breaking strength in the crude ore; consequently, the hourly mill rate and the weight recovery decrease. At the same time, because of its relatively high iron content, the magnetite does contribute to the amount of iron in a block of ore. Were the concentration of the magnetite relatively constant, its presence would be of less consequence, for appropriate adjustments could be made in the milling operations. However, the quantity of magnetite present for any run of ore varies considerably; it is the magnetite prediction for a block of ore that constitutes the greatest problem in grading. THE PRESENT ORE GRADING METHOD Ore for the Smallwood mine is graded by a computerized block grading routine (using 50 x 50 x 45-ft blocks) that applies simple weighted (inverse-distance) averaging to ore grades. The results of the block grading program are not reliable with respect to magnetite concentration. Factors responsible for this poor estimation are (1) the structural setting of the deposit, and (2) mineralogical relationships. Neither of these general factors are given adequate consideration by present methods.

Jan 8, 1967

By C. K. McArthur

The subject of classification is so broad that this discussion is confined to what the author believes is of prime importance in connection with proper grinding and classification. The years past have witnessed many noteworthy changes in fine grinding and classification, which have been accompanied by a readjustment of ore-dressing flowsheets for the preparation of better ground and/or sized products at an expenditure of less power, steel and maintenance than formerly. Nowhere in the industry is more being done in the way of experimental work to lower costs than in grinding circuits, because, at best, grinding costs are high. Recent years have indicated that, on the average, finer grinding has a real application. Without careful preparation of the process feed, no operation can be highly efficient or successful. In an earlier paper, the findings of an extensive survey at representative concentrators were presented1. Also, in the year of 1935, a technical discussion was given by Dr. J. V. N. Dorr (Some Notes on Classification Trends in America) before the Chemical, Metallurgical and Mining Society of South Africa In both papers mentioned, much has been said regarding flowsheets, grinding and classification. The author is very familiar with these papers, having had considerable to do with the actual operation of the flowsheets, as well as large mills and classifiers in the work described. Some data from the second paper have been included in the following discussion. Circulating Load In this particular paper it seems advisable to deal with the classification phases of the work that to a major extent govern the success of any closed-circuit grinding unit. The question has often been asked in connection with a closed circuit: "What circulating-load ratio should be carried?" This is really an

Jan 1, 1939

By A. S. Koster, G. D. Rieck

X-ray experiments were carried out to show the occurrence of cylindrical texture in drawn tungsten wire. Only a small amount of (110) (110) cylindrical texture could he found in wires of intermediate thickness, i.e., of about 1.5 mm diameter. DRAWN wires of tungsten and other bcc metals like iron and molybdenum show a well-established (110) fiber texture. Now it might be conceivable that in drawing radial stresses remove the randomness of the crystallites around the [110} axis, thus giving a fiber texture of higher symmetry. In the resulting so-called cylindrical textures a specified plane normal coincides with the radius of the wire. Several authors discuss the occurrence of cylindrical textures in bcc metals. ~ieck' found no cylindrical texture in drawn tungsten wires of 165 n diameter. Leber~ states that in tungsten wires of 3 mm diameter a cylindrical {100)(110) texture is developed, which is gradually converted to a normal fiber texture by subsequent drawing. He also reported a definite cylindrical texture at the surface of iron wires. Peck and Thomas3 observed a nonuniform mode of deformation of the grains in transverse sections of heavily drawn tungsten and iron wires. According to Bhandary and cullity4 a swaged iron wire has neither a perfect (110) fiber texture nor a perfect {100}(110) cylindrical texture. It is clear that the concept of cylindrical texture in bcc metals still lacks a firm experimental basis. Leber's eviden~e,~ which seems to be the strongest, rests upon qualitative interpretation of photographs. The appearance of the 110 and 200 lines on the photograph of 3-mm wire, however, can be very well attributed5 to a normal ( 110) fiber texture superimposed on random background; our Fig. 1, giving a Debye-Scherrer diagram of etched tungsten wire of 180 p, is free of cylindrical texture according to all authors, but is quite similar to Leber's diagram. It was therefore decided to use a different method of X-ray texture determination on tungsten wires of diameters where cylindrical textures may be expected.

Jan 1, 1965

By W. Rostoker, H. Nichols

The embrittlement by mercury of aluminum alloys in various states of anneal, cold work, and aging has been studied. It is shown that single phase and non-precipitation hardened structures irrespective of de,pee of prior cold work demonstrate some small measure of yielding before fracture occurs. Sub-yield-stress failure is confined to the aged state, and embrittlement is accentuated by small plastic prestrains. With large plastic pyestrains, embrittle~nent is progressively reduced. In previous work' it was shown that the propensity of commercial aluminum alloys for embrittlement when wetted with mercury seems to be related primarily to the intrinsic strength in the normal un-wetted condition. The data from a large number of commercial alloys in various states of anneal, cold work, and/or aging gave a generalized relationship. At low strength levels, the wetted fracture strength increases proportionately to the increase in tensile strength; but above 65,000 psi normal ultimate strength, the trend is abruptly reversed, and the wetted fracture stress decreases proportionately to the increase in tensile strength. This paper will demonstrate that the simple character of this relationship between normal tensile strength and wetted fracture strength is fortuitous. It will be shown that structure and thermal-mechanical history are clearly relevant factors. The term brittle fracture has been used to connote the geometry of the fracture surface (transverse vs shear plane) or the negligible extent of preceding plastic strain. At its maximum effectiveness, liquid metal embrittlement can satisfy both criteria—fracture transverse to the axis of tension and without observable plastic strain. In this paper, brittle fracture will signify failure at stresses below the engi-

Jan 1, 1962

By Sukune Takenouchi

Porphyry copper deposits in the southwestern Pacific region differ in some points of geologic features from those of the United States. The deposits distribute along old or active island arcs and are generally younger in age (1-16 m.y. B.P.). In Japan, however, no typical porphyry copper deposits are yet reported, though the geological situation of Japanese islands is similar to that of the southwestern Pacific. Fluid inclusions in some porphyry copper deposits In the southwestern Pacific region and those in igneous rocks genetically related to the ore deposits In Japan have been studied, and the genesis of ore-forming fluids from igneous rocks are discussed. From the results, the possibility of finding porphyry copper-type mineralization in Japan are considered.

Jan 1, 1976

By P. E. McKinney

DEVELOPMENTS in engineering during the past decade, particularly as applied to marine construction, mining machinery and other purposes in which corrosion offers a serious problem, have created a large demand for a non-ferrous metal highly resistant to corrosion and at the same time useful in general construction work as a substitute for steel without materially decreasing the factor of safety or increasing the weight of the various parts over that ordinarily used in the case of mild steel. To obtain the combination of desired properties many compositions have been proposed and used with excellent success, but at the present time probably the most popular and most widely used combination is the non-ferrous alloy commonly called manganese bronze. This is nothing more than a high brass to which have been added, by the proper method of alloying, comparatively small percentages of aluminum, iron, or manganese with the definite purpose in view of strengthening the alloy and rendering it more dense and close-grained than the average yellow-brass casting. In the manufacture of manganese bronze a great deal has been said about the importance of using only the highest grades of raw material and the beneficial or detrimental effects, as the case may be, of various impurities, as well as the importance of adding the ingredients according to various formulas proposed; but in most of these cases the literature on the subject has dealt principally with the manufacture of this alloy from virgin metals And raw materials of the highest purity. It is the purpose of this paper to deal particularly with the possibilities that lie in the proper development of methods for manufacturing such an alloy by more economical methods than those which have usually been discussed in other literature on the subject. Shortages have recently existed in raw materials needed for many products entering into the various branches of engineering work, particularly the non-ferrous metals, and during the next few years of reconstruction the country will be flooded with many

Jan 2, 1919

By Zay Jeffries

IT is well known that many properties of a given metal vary with the size of grain or cell. For most industrial purposes, where high ultimate strength and high elastic limit are desired, the manufacturer tries to produce a. fine-grained structure. For some purposes, for example in transformer iron, a coarse-grained structure is desirable. The terms fine-grained and coarse-grained are used only in a relative sense. For example, a fine-grained cast copper might have grains 100 times larger than coarse-grained high-speed steel. Similarly, a fine-grained steel rail might have grains or cells 100 times larger than coarse-grained wire, which was made of steel from the same heat. In some metals, the change in grain size is more appreciable than the changes in any of the properties determined by the tensile test. Furthermore, in metals which have been annealed or subjected to high temperatures, grain size may prove to be a better indication of the life of the metal in use than the tensile test. One of the authors was able to supplant the tensile test with grain-size determinations in the control of. metal for a specific use, the latter determination being more indicative of. the life of the metal part than the tensile test. A grain-size determination with a tensile test is, of course, more valuable than either singly. At present there is a decided dearth of data on actual grain-size determinations in metals, and consequently a corresponding lack of ability to interpret the determinations after they are made. It would be desirable to have correlation made between grain size and properties of metals, comparing thousands of determinations. In this way, grain-size determinations might be interpreted in much the same manner that tensile tests are now.

Jan 12, 1915

By Roy T. Southin*

The finding of Wojciechowski and Chalmers that columnar crystal growth can be stopped by an already existing network of equiaxed crystals is supported by other recently reported work on both static and mechanically stirred ingots.27,28 Support for the dendrite arm fragmentation mechanism, too, can be obtained by comparing the effects of spinning on ingots cast in a material which has a dendritic solid/liquid interface during solidification with one which has a cellular interface, since no evidence of change at the interface, other than a change in growth direction, can be found in the cellular material. But there are alternative ways of producing an equiaxed zone in materials which characteristically solidify with a cellular interface. For example, interrupted spinning of a normally cast ingot can do it. In this case it has been shown28 that the equiaxed zone is caused by the showing of dendritic particles from the top surface of the ingot to nucleate a network of equiaxed crystals which forms a barrier to the continued growth of the columnar grains. This showering has also been shown to be a major effect in materials solidifying with dendritic interfaces. It has also been proposed28,29 that, for an equiaxed zone to form, not only must nuclei be present ahead of the interface but also these nuclei must grow to sufficient size to be a barrier to the further growth of the columnar grains. A further requirement is that a network of these grains continually be produced ahead of the growing interface; otherwise there is a reversion to columnar growth as found by Wojciechowski and Chalmers. In normally cast ingots it is possible for nuclei ahead of the interface to grow since during a large portion of the solidification time there is no temperature gradient ahead of the interface,30 and, in fact, the liquid may be slightly undercooled.29 This then is the major difference between the work under discussion and normal industrial solidification conditions, since Wojciechowski and Chalmers imposed a considerable temperature gradient in the liquid. The statement in the Wojcieshowski and Chalmers paper that one needs to lower the temperature gradient

Jan 1, 1969