SPALLING AND LOSS IN COMPRESSIVE STRENGTH OF FIRE BRICK1

A preliminary report of the loss of compressive strength when fireclay brick from the Pacific Northwest were subjected to a series of heat treatments to 1350° and 1250°C. It illustrates some of the variations of heat treatment in the manufacturer's kilns and the differences between the high siliceous type of fire brick and the vitrifying clay type with lower free silica content. It is possible that a satisfactory spalling test may be developed in this direction.     

THE BEHAVIOR OF FIRE BRICK IN MALLEABLE-IRON FURNACE BUNGS1

Purpose. —An investigation was conducted to study the requirements of fire clay and bodies used for fire brick in malleable-iron furnace bungs. Tests were made on complete bungs holding forty sample brick in malleable-iron furnace bungs with twenty different fire brick. Laboratory tests were also made in conjunction with them. Results. —The spalling tests bear the closest relation to the service test; those brick losing less than 10% withstand more than fifteen heats. There is also a relation between the porosities and densities of fire brick, which lie between 15 and 28% and 1.5 and 2.6%, respectively, for the best brick. There is no close relation between the load test and softening-points of fire brick and their lifetime in malleable furnace bungs, so these tests are no criterions in judging the serviceability of brick, provided the brick are sufficiently refractory to support the arch at furnace temperatures. Methods for Improving Fire Brick. —The resistance of a brick to spalling may be governed by: (1) the selection of the proper clays, (2) the size of grain and the proportioning of the non-plastic ingredients, (3) the fineness of grain of the bond clay, (4) the manner of molding, and (5) the temperature of firing.     

THE COMMERCIAL DEVELOPMENT OF A KAOLINIC FIRE BRICK1

The development of a kaolinic brick from Georgia clay is described. The high and continued shrinkage of this clay makes it necessary to fire the brick a t a very high temperature. A temperature of over 3000°F was required. The development of a kiln for the firing of the grog and brick was a problem that was satisfactorily solved. A light weight brick for use in marine boilers and a dense refractory for use in glass tanks were developed. The following physical properties of these two refractories are given and compared with other high grade bricks: (1) start of deformation under 25 Ibs. per sq. in. load, (2) 10% deformation under 25 Ibs. per sq. in. load, (3) start of permanent volume change without load, (4) mean coefficient of expansion, (5) cycles in 2900°F air-spalling test, (6) melting point, (7) thermal conductivity a t 1000, 2000 and 2750°F. Various successful applications of this type of brick are described.     

THE FIRE BRICK MATERIALS OF PENNSYLVANIA1

Pennsylvania produces over 40% of all clay fire brick and over 70% of all silica fire brick made in the United States. In 1920 Pennsylvania produced and sold fire clay worth nearly two million, fire clay brick worth nearly twenty-two million, and silica brick worth over ten and one-half million, a total of thirty-four and one-half million dollars worth of brick and clay. Three kinds of material are used for the manufacture of this fire brick. (1) Soft fire clay from one horizon northwest of Pittsburgh and from the “main clay” of the Pittsburgh bed south of Pittsburgh. (2) Hard or flint clay which occurs in many counties of the State. The hard clays are described as occurring at a limited number of horizons in the Allegheny formation or “Lower Productive Coal Measures,” as being irregular in thickness and distribution, ranging from a feather-edge to fifteen or twenty feet in thickness and being of two types, “block” clay in which the structure is homogeneous, and “nodular” clay in which the structure is gnarly or knotty. Brief reference is made to the principal flint clay deposits of the State. (3) The third source of material for fire brick is ganister rock, a pure white quartzite occurring as a massive bedded sandstone at the base of the Silurian series and exposed abundantly in the zig-zag ridges across the central part of the State and as a similar white quartzite at the base of the Cambrian in the southeastern part of the State. A series of tests of flint clays and of soft clays used to mix with the flint clays is appended. These are derived mainly from Clearfield County which is the leading flint clay county.     

ARTIFICIAL SILLIMANITE AS A REFRACTORY1

Alumina-silica mixtures were prepared by fusing quartz, china clay, fire clay, and alumina in the electric furnace. When alumina is less than 68%, crystalline sillimanite (3Al₂O₃.2SiO₂) with glass is produced. This material is not very resistant to loads at high temperatures because of the early fusion and internal lubricating action of the glass surrounding the crystals. Above 68% alumina, crystalline corundum appears am1 the glass is practically absent. This latter composition is very resistant to high-temperature loads when ail interlocking, recrystallized bond is developed. This material is not affected materially by acid slags, but it cannot resist basic slags. However, the dense structure of a brick of material above 68% Al₂O₃ causes less slagging in a laboratory bath test than silica brick. The laboratory made sillimanite-corundum brick withstood higher temperatures than the best silica, magnesia, chrome, fire clay, or zirconia brick even though the cone of fusion of the former is less than that of MgO, Cr₂O₃ or ZrO₂. More and better service tests with a large number of brick fired in large kilns is needed to follow up this laboratory work.     

MISSOURI HARD FLINT CLAY FIRE BRICK1

The hard flint clay obtained in the diaspore pits of Missouri, which has exceedingly low bonding properties and high shrinkage, but a P.C.E. value of cone 34, with the use of as low a percentage of flint clay grog as 20%, can be bonded into a strong brick, free of physical defects and of super-refractory quality. Excessively high forming pressure is not required. A small addition of bentonite adds materially in bonding the clay. The results of spall, hot load, cold compression, porosity, shrinkage, and reheat tests are presented.     

ELASTIC BEHAVIOR AND CREEP OF REFRACTORY BRICK UNDER TENSILE AND COMPRESSIVE LOADS*

Specimens cut from 9-in, brick of nine brands of firebrick, including two high-alumina, four fire-clay, two siliceous fire-clay, and one silica, were subjected to tensile and compressive creep tests at eleven temperatures from 25° to 950°C., inclusive. The duration of each test was approximately 240 days. Small length changes, independent of stress direction (that is, compressive or tensile), occurred at the lower temperatures. The lowest temperatures at which creep was significant were (a) high-alumina brick, 700° to 850°C.; (b) fire-clay brick, 600° to 700°C.; and (c) siliceous and silica brick, 950°C. Creep results under compressive stress could not be correlated with results under tensile stress. Specimens of different brands, at 950° C. showed greatly different capacities to carry load. Repeated heatings caused growth of silica brick of approximately 0.27%. Moduli of elasticity at room temperature were determined before and after the various heat-treatments and resultant changes were recorded. The changes in moduli were 15% or greater for silica and siliceous brick and 4% or less for the fire-clay brick. The moduli of elasticity at room temperature were approximately 2.7–4.3 × 10⁶ for high-alumina brick, 0.6–1.9 × 10⁶ for fire-clay brick, 0.3–1.7 × 10⁶ for siliceous fire-clay brick, and 0.4 × 10⁶ for silica brick.     

SECONDARY EXPANSION OF HIGH-ALUMINA REFRACTORIES*

Several mixtures of raw and calcined diaspore and bauxite with raw and calcined fire clays were prepared and fired at 2700°F. Specimens were refired at higher temperatures, and the linear changes were determined. Bodies composed of high-alumina grog and fire clay expand in the refire, whereas specimens of fire-clay grog that are bonded with ground raw diaspore exhibit shrinkage. The cause of the secondary expansion and shrinkage of these bodies is discussed.     

STUDIES ON THE THERMAL CONDUCTIVITIES OF SOME REFRACTORY MATERIALS1

The method used was a modification of that described by Hepplewhite. The changes made were for the purpose of insuring greater accuracy. An effective means for preventing lateral flow of heat was devised. The coefficients obtained for the several materials studied were (c. g s. units) silica brick .00099, fire clay brick .00169, fire clay slab .00203, light weight fire clay brick .00251, sillimanite brick .00432, electrically sintered magnesia brick .00665, “Alundum” fire brick .00833, “Crystolon” silicon carbide brick .00982. [The work for this paper including the design, erection and operation of the apparatus, most of the calculations and the preparation of the detailed report was done by 0. S. Buckner.]     

THE EFFECT OF RED HEARTS IN FIRE CLAY BRICK1

Laboratory tests made on two brands of fire clay brick indicate that red hearts, which are normally considered as being detrimental, add to both the hot and cold strength and would not be objectionable except in furnaces operated at a low temperature, and in which spalling is an important item. Results of load, reheating and spalling tests are given, together with porosity and iron oxide determinations. A short discussion of the causes of red hearts is also included.     

COLD CRUSHING STRENGTH OF FIRE BRICK1

Data on cold crushing strength in three directions, viz., flat, edge, and endwise of six brands of fire brick are given. Transverse strength data of all these brands are also given. Porosities of all the brick used in these tests were determined by the air-expansion method. The purpose of the investigation was to find whether it is possible to translate the values obtained for crushing strength of fire brick in one direction (flat) into values for the other directions (end and edge); to determine whether transverse strength data can similarly be transformed to crushing strength data and vice versa and whether porosity and crushing strength of fire brick are correlated. The data herein presented do not show the existence of any such simple mathematical relationship between the different properties of the brick. A new capping material, a mixture of sand and molten sulphur, was used for the crushing tests, and was found to be more satisfactory than the other materials commonly used for the purpose. It is recommended that in reporting crushing strength data of fire brick, the brick be tested on end.     

FIRE-CLAY BRICK FOR THE OPEN HEARTH1

Fire-clay brick in the open hearth are normally subjected to medium temperatures and to the action of slag containing high percentages of iron and lime. Temperatures near the maximum of the open hearth itself may exist due to failure to reverse the draft and for that reason brick of high refractoriness are used. Resistance to spalling action is not of primary importance because the temperature change is small. The indications obtained from a laboratory test were that brick having a high alumina content would resist the slag action better than ones higher in silica.     

RELATION OF CRUSHING STRENGTH OF SILICA BRICK AT VARIOUS TEMPERATURES TO OTHER PHYSICAL PROPERTIES*

The failure at elevated temperatures under constant load for silica brick is reported using the Iupuy load test apparatus. The crushing strength at 1500°F, 1800°F. 2100°F, and 2400°F is recorded, as well as the crushing strength at room temperature. The size of test piece utilized normally was 1 by 1 by 2′/2 inches. A definite relationship is shown to exist between the strength at room temperature and that at elevated temperatures. The effect of variation in lime content, bats content, and fluxes is also reported. Data were obtained on brick made from three different quartzites. Additional physical data are reported to give information concerning the properties of the brick tested.     

Reaction Between K2O and Al2O3-SiO2 Refractories as Related to Blast-Furnace Linings

An examination was conducted to determine the mechanism of peeling of fire-clay brick in the low-temperature region of a blast furnace where 3 to 10% K₂O is the principal contaminant. In laboratory tests, as-received high-duty and superduty fire-clay brick and 70% alumina brick treated with KCl-K₂CO₃ mixtures showed no peeling at a temperature of 1600°F. Cracks were found in high-duty brick that were treated with KCN at 1500°F. under partially reducing conditions. X-ray diffraction studies of mixtures of crushed brick and K₂CO₃ indicated the formation of leucite (K₂O.Al₂O₃.4SiO₂) and kaliophilite (K₂O.-Al₂O₃.2SiO₂) at temperatures below 1700°F. These latter data, confirmed by specimens from used blast-furnace linings, showed that silica is the first constituent attacked by alkali. Since the formation of leucite and kaliophilite in fire-clay brick is the probable cause of peeling, the increased reaction of silica, in a dense Al₂O₃.SiO₂ refractory of higher silica content than fire-clay brick, should confine the alkali attack to the surface of the brick in low-temperature applications.     

A STUDY OF THE FACTORS INVOLVED IN THE SPALLING OF FIRE CLAY REFRACTORIES WITH SOME NOTES ON THE LOAD AND REHEATING TESTS AND THE EFFECT OF GRIND ON SHRINKAGE1

Of the three factors, elasticity, coefficient of expansion and rate of temperature change, which affect spalling, the former is by far the most important. Only small differences are found between fire clay mixtures of widely varying structure and composition in the rate at which they change in temperature under like conditions of heating. The coefficient of expansion varies directly with the silica content and differences in this respect of large order were found. However, the spalling on the particular mixtures tested varied almost inversely as the coefficient of expansion. This apparent discrepancy is explained on the basis of greater elastic properties of the brick which had high expansions. The elasticity may be varied between wide limits and is sufficiently important as to overbalance the effect of greater expansion. This property is accordingly the one upon which efforts directed toward the development of non-spalling brick should be centered. It was discovered that a plastic deformation could be obtained at as low a temperature as 635°C. This gives the effect of elasticity and undoubtedly has considerable influence on spalling at the higher temperature ranges. Results are given for a number of load tests which show clearly the importance of hard firing. The secondary expansion of brick made from Pennsylvania flint clay is shown to be influenced by the temperature of reheating, as well as its rate. Detailed results showing the effect of grind and firing on the finished size of the brick included in the investigation are also given.     

PRODUCER GAS IN CONNECTION WITH FIRING FIRE BRICK1

Features in connection with firing fire brick are discussed. In tunnel kiln firing a high temperature at the charging end is essential. Desired features of producer gas firing are discussed and a producer gas fired tunnel kiln installation is described in detail. It is shown that production capacity is dependent on the charging end temperature.     

THERMAL EXPANSION OF SILICA BRICK AND MORTARS1

Thermal expansion from 20 to 950°C and physical data of silica brick from various producing districts in the United States and Europe are presented. The variations in thermal expansion of brick from various parts of kilns are given for plants in the United States. The magnitude of variation of the thermal expansion of silica brick is quite small, the expansion ranging between 1.15% and 1.30% at the highest point of expansion. The expansion of the silica mortars varied between 1.30 to 1.52% depending upon variations in clay, quartzite, and bats. The variations in thermal expansion of silica mortars from various producing plants are also shown. Data on the effect of size of grain, clay content, and P.C.E. on the thermal expansion of mortars are given. An extensive bibliography on thermal expansion of silica brick appears with the paper.     

PRELIMINARY REPORT1 ON THE RESIDUAL KAOLIN AND FELDSPAR IN THE PACIFIC NORTHWEST

Large quantities of residual crystalline kaolin are found in the Inland Empire district of eastern Washington and northwestern Idaho. Preliminary investigation has shown that the properties of these clays, when properly selected and purified, will compare favorably with those of the English china clay upon which the eastern potters and paper manufacturers are largely dependent. With the low cost of the English clay laid down on the Atlantic Coast, it is not believed that the Pacific Northwest clay will be used as a substitute for the English in the eastern markets, but the domestic source will provide the bulk of the raw material for the future whiteware industry in the Pacific Northwest. Commercial quantities of feldspar, quartz, and fire clay are also found in the Pacific Northwest. The work described here was conducted at Seattle under a cooperative arrangement between the College of Mines, University of Washington, and the Northwest Experiment Station of the U. S. Bureau of Mines.     

COMPARISON OF HOT AND COLD MODULUS OF RUPTURE FOR SILICA BRICK

Purpose of the Investigation .—(1) To obtain relative values for the cross-breaking strength of silica brick at temperatures encountered in coke oven practice. (2) To correlate the hot modulus of rupture test, if possible, with the cold modulus of rupture, or cold crushing test, either of which is cheaper and more easily conducted. This report gives the method of making the test, difficulties encountered and results obtained. The report shows a comparison of cold crushing, cold modulus of rupture and hot modulus of rupture on a series of silica brick made from special mixes, commercially burned. Conclusions .—The modulus of rupture of a silica brick at 1350°C is approximately one-third the strength at atmospheric temperature. For this series it averaged from 130 to 189 lbs. per square inch. Too rapid or eccentric heating up to red heat may cause such weakening of the structure or bond that the brick will break under very low pressure. Cross-breaking strength decreases as the temperature increases. Hot modulus of rupture test appears to give results, in most respects, comparable to the cold test, and for routine testing it would seem advisable to use the cold test since it can be made in much shorter time.