Search Documents

By Charles R. Simcoe

The stress-strain curves in the microyield region for various Fe-C solid solutions shown by Davies and Ku are typical examples of parabolic curves defined by the expression a = kEn 111 These data, when replotted on log-log coordinates, as shown in Fig. 9, exhibit three essentially parallel straight lines. It is apparent that carbon increases the flow stresses, but does not affect the exponent n. In other words, the only factor in Eq. [1] affected by in in carbon content is the constant k. As Eq. [1] is written k is the stress at a strain of 1.0. This is unsatisfactory for examining data in the microyield region. Therefore, the equation can be altered to the form: to permit the use of the authors' microyield stress at a strain of 10-6 as k. Extrapolating the curves in Fig. 9 to a strain of 10-6 gives stress values of 2000, 3200, and 5400 psi for 0.008, 0.014, and 0.02 pct C, respectively. These values do not agree exactly with the authors' measured values, but this will not affect the following discussion. The effect of carbon content on the slopes of the stress-strain curves shown in Fig. 2 of the paper by Davies and Ku can be analyzed by differentiating Eq. [2] with respect to strain. Thus: where n is always less than 1.0. Since n is essen- tially constant as a function of carbon content for the data under discussion, the slopes of the stress-strain curves depend only on k and E. If the slopes are examined at various selected Plastic strains. i.e., 10-6, 10-5, and so forth, it follows from Eq. [3] that the dif-ferent slopes for the different carbon contents depend only on the different values of ke=10-6. Thus, the larger ke=10-6, the steeper the linear stress-strain curve will rise with increasing plastic strain. This

Jan 1, 1969

By R. E. Reed-Hill, W. H. Hartt

High-purity magnesium single crystals, oriented with basal plane parallel to stress axis, were deformed in tension at room temperature so as to form second-order (1011)- (1012) twins. Investigation by optical and replica electron microscopy revealed that these twins can support very large internal strains approaching 1000 pct and that fracture follows as a direct consequence of these strains. At low strains the basal plane within (10i1)- (10i2) twins rotates with increasing deformation until it becomes approximately parallel to the twin boundaries. This rotation is not consistent with deformation by basal slip in the double twin, but may be rationalized in terms of nonbasal slip. Observations also indicate that basal slip in the primary (1011) twin, prior to second-order twinning, may contribute to this rotation. Deformation markings that are difficult to interpret in terms of slip on previously observed systems were noted in the twin bands. A possible means of rationalizing these markings is to relate them to grain boundary shear. The fracture associated with these twins is thought to initiate by formation of voids at the twin boundaries. These grow into microcracks, with ultimate fracture occurring by tearing of the interconnecting regions. THE significance of deformation and fracture along second-order (10i1)-(10i2) twins in polycrystalline magnesium1-' and many magnesium alloys4 was discussed in a previous paper.5 It was noted that this double twinning mode effectively controls the room-temperature deformation and fracture of magnesium single crystals, oriented with the basal plane parallel to the tensile axis. Reed-Hill and Robertson6 observed that, although these crystals may fracture at a macroscopic strain of less than 1 pct, shear strains up to 1000 pct could occur within (1011)- (10i2) twin bands. They concluded that the fracture was "ductile" and resulted from concentrating the deformation into a very small volume. The present research was undertaken to obtain a better understanding of the deformation and fracture associated with these twins. EXPERIMENTAL TECHNIQUES The experimental procedure involved straining high-purity magnesium single crystals in tension at room temperature. Specimens were oriented so that the applied stress direction was within 2 deg of the [1010] crystal axis. The specimen cross section was rectangular with faces closely parallel to (1210) and (0002). The specimen preparation and testing procedures have been described in detail elsewhere.5 Second-order (1011)- (1072) twins nucleated during straining were studied using optical and replica electron microscopy techniques. For the latter, cellulose acetate-carbon double replicas were employed. Observations were made on the (1210) crystal surface, which is the plane of shear for twins of this type. In addition to the single crystals, some investigations were carried out on longitudinal polycrystalline specimens. These were obtained from high-purity magnesium plate with a texture in which the basal planes of the grains tended to be nearly parallel to the specimen axis. All. polycrystalline specimens were annealed prior to testing in order to produce a coarse grain structure, permitting Laue back-reflection X-ray photographs to be obtained from grains of interest. EXPERIMENTAL RESULTS a) Deformation Within ( 101 l )-( 1072) Twins. Fig. 1 is an optical micrograph demonstrating the extent of the plastic deformation that can occur in (1011)-(10i2) twin bands. Notice at the lower left corner of the photograph the large displacement of the dark band (a polishing step) where it crosses the twins. The shear strain in the twin band at this position was computed to be 700 pct, using the ratio of step width to twin thickness. It was previously shown5 that (1011)- (1012) twin bands are often composed of small, separate twins aligned close to the macroscopic habit plane. Fig. 2 is an electron micrograph from near the upper end of the twin band in Fig. 1. This illustrates that the twin habit often does not coincide exactly with the macroscopic band habit, but may be inclined at a slightly smaller angle to the matrix basal plane. Figs. 3 through 6 are electron micrographs of the twin band in Fig. 2 illustrating successive steps in the development of deformation inside the band. A twin band such

Jan 1, 1969

By C. A. Stout, S. W. Nicksic

The materials that stubilize emulsions of some Southern California crude oils were isolated from the produced crude. These substances were separated into acidic and nonionic fractions by ion exchange chromatography. Each fraction was further broken into groups of homogeneous molecular weight composition by gel permeation chromatography. Both ionic character and molecular weight of the individual fractions are reflected in the properties of the original crude oil emulsion. Specifically, the more acidic and the higher molecular weight fractions appear to be the most effective stabilizers. INTRODUCTION The treatment of oilfield emulsions is a continuing problem for the oil industry. In almost all cases, these are water-in-oil emulsions. Much more is known about the formation and stabilization of oil-in-water (O/W) emulsions than of the water-in-oil (W/O) type. To understand W/O emulsions, more information is needed on the materials responsible for their formation and stabilization. This is especially true in the case of oilfield emulsions where the stabilizing materials are complex mixtures of large asphaltene-like molecules. To study the chemical and physical properties of these emulsion stabilizers, it is necessary first to separate them from crude oil. This separation was achieved in the case of two Southern California crude oils. The materials that were isolated appear at first to be little different from an asphaltene mixture. However, if a mixture of distilled water and mineral oil containing about 1 percent of the natural emulsion stabilizers is passed through a colloid mill, the W/O emulsion formed may be stored in the laboratory for many months without any sign of breaking. This could not be done using undifferentiated asphaltenes from the same wells, indicating that there is a real difference between these materials and ordinary asphaltenes. The colloid mill used to generate the emulsion was required because tile natural emulsion stabilizers had very little effect on oil-water interfacial tension. The one characteristic — complexity of composition — that emulsion stabilizers share with asphaltenes makes virtually impossible the exact structure determination of the components of the mixture. For this reason, it was decided to investigate first the influence of molecular weight on the ability of these materials to stabilize emulsions. The emulsion stabilizers were first separated into acidic and neutral fractions according to whether or not the material was retained on a weakly basic ion exchange column. The extraction procedure was not designed to remove basic components from tile oils, and no attempt to isolate such a fraction was made. Following this separation, selected fractions were analyzed by gel permeation chromatography. Gel permeation separates according to molecular size, largely independently of chemical or polar characteristics. A heterogeneous mixture is separated into a number of fractions having much more homogeneous molecular weight distribution. A number average molecular weight of each fraction is therefore close to the true molecular weight of the principal component of that fraction. The fraction weights, expressed as weight percent of the sample, and the number average molecular weight of a representative selection of the fractions, will be referred to in this paper as the molecular weight profile of the original sample. DESCRIPTION OF GEL PERMEATION ANALYSIS The first reported gel permeation analyses were for separating partially hydrolyzed protein molecules in an aqueous system using expanded starch granules. Most of the theoretical work on gel permeation has been done by biochemists.1 This theoretical treatment is still incomplete, but the

Jan 1, 1969

By E. H. Boswell, E. M. Cushing

The Mississippi embayment east of the Mississippi River is part of a vast geologic and hydrologic province. The abundance of water available is largely responsible for the agricultural history of the area and is perhaps the key factor in the rapidly accelerating trend toward industrialization. The multiple artesian aquifer systems that underlie the region and the alluvial aquifer along the Mississippi River are virtually undeveloped sources of fresh water in that they are yielding only a fraction of the water available. The potential yield of the 12 aquifer systems is estimated to be about 25,000 mgd (million gpd) of which only about 600 mgd is presently being withdrawn. The amount of water in storage in these systems is about 1600 trillion gal. Ground water having a dissolved-solids content of less than 500 ppm is generally available at depths of less than 1000 ft, and water having a dissolved-solids content of less than 1000 ppm is available in some places to depths of more than 3000 ft. Iron is the most common troublesome chemical constituent in the ground water. Water in varying amounts is also available from streams within the region. The amount of water which originates within the region and which leaves it as streamflow during a year averages about 50 million acre-ft (about 45,000 mgd). An additional 375 million acre-ft (about 340,000 mgd), which originated outside the region, leaves the region as streamflow during an average year. Some streams have considerable flow during periods of no precipitation and the water during these periods generally has a low dissolved-solids content. Large quantities of water from streams and waterbearing sands (aquifers) are available in the Mississippi embayment east of the Mississippi River. Most of the region is underlain by aquifer systems that will yield large quantities of water to wells, making ground water the most readily available source of fresh water. About 90% of the water used in the region is obtained from wells. Streams have been used primarily to dispose of, and dilute, wastes that have been dumped into them. The Mississippi, Ohio and Tennessee rivers, and the lower parts of the Tombigee and Yazoo rivers also serve as a means of transportation. This nominally water-rich region receives from 48 to 56 in. of precipitation in average years and is crossed by several large rivers. The largest, the Mississippi River, is the western boundary of the region discussed in this paper. The economy of the region is basically agricultural, but industry has become increasingly important since World War 11. The diversification and mechanization of agriculture have resulted in a surplus of manpower, most of which is being absorbed by the expansion of industry. Further industrial expansion probably will occur because of the available manpower, natural resources and economical transportation. Increasing demands on available water supplies will be made and the future economic growth of the region will be largely dependent upon wise utilization and management of the water resources.

Jan 1, 1969

By U. Paakkinen, P. Eerola

In connection with theoretical studies of the dynamics and control of the flotation process, flotation cells and circuits at the Pyha- computer control of selective Cu - Zn - Fe sulfide flotation in the latter half of 1969. In this paper, the practical conclusions of these studies, conducted through simulation techniques, are summarized. The application of these results for control purposes implies a knowledge of the dependence of the flotation rate coefficients and the grade of concentrate on several factors, including aeration rate, froth depth, and reagent consumption. The relationships found in laboratory and plant scale tests are described. Off-line optimization studies, based on the relation- ships, are also described. The proposed steps in executing the computer control schedule are presented. In the first stage, a stabilizing control will be carried out using feed-back loops. A feed-forward control system which utilizes simple process models will also be tested. After these elementary stages, a proposed mathematical model of the circuit will be checked out and implemented with computer- run, plant-scale process tests. Off-line simulation will be continued in advance development of the control system. As a final step in the program optimal control of the process will be carried out through this model.

Jan 1, 1969

By C. S. Land

Relative permeability functions are developed for both two- and three-phase systems with the saturation changes in the imbibition direction. An empirical relation between residual nonwetting-phase saturation after water imbibition and initial nonwetting-phase saturations is found from published data. From this empirical relation, expressions are obtained for trapped and mobile nonwetting-phase saturations which are used in connection with established theory relating relative permeability to pore-size distribution. The resulting equations yield relative permeability as a function of saturation having characteristics believed to be representative of real systems. The relative permeability of water - wet rocks for both two- and three - phase systems, with the saturation change in the imbibition direction, may be obtained by this method after properly selecting two rock properties: the residual nonwetting-phase saturation after the complete imbibition cycle, and the capillary pressure curve. INTRODUCTION Relative permeability is a function of saturation history as well as of saturation. This fact was first pointed out for two-phase flow by Geffen et al. 1 and by Osaba et a1.2 Hysteresis in the relative permeability-saturation relation also has been reported for three-phase systems.3 Since saturations may change simultaneously in two directions in a three-phase system, four possible relationships arise between relative permeability and saturation for a water-wet system. The four saturation histories of this system were given by snel14 as 11, ID, DI and DD. I and D refer to the direction of saturation change (imbibition and drainage), with the first letter of the symbol indicating the direction of change of the water phase. As used in this paper, the second letter of the symbol refers to the direction of saturation change of the gas phase, i.e., D and I indicate an increase and decrease, respectively, in gas saturation. Only a few three-phase relative permeability curves have been published. Leverett and Lewis5 published three-phase curves for unconsolidated sand, and Snell4 reported results of several English authors for both drairrage and imbibition three-phase relative permeability of unconsolidated sands. Three-phase relative permeability curves for a consolidated sand were published by Caudle et a1.3 for increasing water and gas saturations (ID). Corey et a1.6 reported drainage (DD) three-phase relative permeability for consolidated sands. Recently, Donaldson and Dean7 and Sarem8 calculated three-phase relative permeability curves from displacement data on consolidated sands, also for saturation changes in the drainage direction. The only published three-phase relative permeability curves for consolidated sands with saturation changes in the imbibition direction (11) are those of Naar and Wygal.9 These curves are based on a theoretical study of the model of Wyllie and Gardner 10 as modified by Naar and Henderson.11 Interest in three-phase relative permeability has increased recently due to the introduction of new recovery methods and refinements in calculation procedures brought about by the use of large-scale digital computers. The scarcity of empirical relations for three-phase flow, and the experimental difficulty encountered in obtaining such data, have made the theoretical approach to this problem attractive. RELATIVE PERMEABILITY AS A FUNCTION OF PORE-SIZE DISTRIBUTION Purcell used pore sizes obtained from mercury-injection capillary pressure data to calculate the permeability of porous solids.12 Burdine extended the theory by developing a relative permeability-pore size distribution relation containing the correct tortuosity term.13 The works of Purcell and Burdine were combined by Corey14 into a form that has

Jan 1, 1969

By I. Javandel, P. A. Witherspoon

The finite element method was originally developed in the aircraft industry to handle problems of stress distribution in complex airframe configurations. This paper describes how the method can be extended to problems of transient flow in porous media. In this approach, the continuum is replaced by a system of finite elements. By employing the variational principle, one can obtain time dependent solutions for the potential at every point in the system by minimizing a potential energy functional. The theory of the method is reviewed. To demonstrate its validity, nonsteady-state results obtained by the finite clement method are compared with those of typical boundary value problems for which rigorous analytical solutions are available. To demonstrate the power of this approach, solutions for the more complex problem of transient flow in layered systems with crossflow are also presented. The generality of this approach with respect to arbitrary boundary conditions and changes in rock properties provides a new method of handling problems of fluid flow in complex systems. INTRODUCTION Problems of transient flow in porous media often can be handled by the methods of analytical mathematics as long as the geometry or properties of the flow system do not become too complex. When the analytical approach becomes intractable, it is customary to resort to numerical methods, and a great variety of problems have been handled in this manner. One such method relies on the finite difference approach wherein the system is divided into a network of elements, and a finite difference equation for the flow into and out of each element is developed. The solution of the resulting set of equations usually requires a high speed computer. When heterogeneous systems of arbitrary geometry must be considered, however, this approach is sometimes difficult to apply and may require large amounts of computer time. The finite element method is a new approach that avoids these difficulties. It was developed originally in the aircraft industry to provide a refined solution for stress distributions in extremely complex airframe configurations. 27 Clough has recently reviewed the application of the finite element method in the field of structural mechanics.6 The technique has been applied successfully in the stress analysis of many complex structures.l, 27, 28 Recognition that this procedure can be interpreted in terms of variational procedures involving minimizing a potential energy functional7 leads naturally to its extension to other boundary value problems. In the field of heat flow, there recently have been introduced several approximate methods of solution that are based on variational principles.2-4, 17 By employing the variational principle in conjunction with the finite element idealization, a powerful solution technique is now available for determining the potential distribution within complex bodies of arbitrary geometry. In the finite element approximation of solids, the continuum is replaced by a system of elements. An approximate solution for the potential field within each element is assumed, and flux equilibrium equations are developed at a discrete number of points within the network of finite elements. For the case of steady-state heat flow, the technique is completely described by Zienkiewicz and Cheung.33 Since the flow of fluids in porous media is analogous to the flow of heat, Zienkiewicz et al. have employed the finite element method in obtaining steady-state solutions to heterogeneous and anisotropic seepage problems. 34 Taylor and Brown have used this method to investigate steady-state flow problems involving a free surface.25 The work of Gurtin has been instrumental in laying the groundwork for the application of finite element methods to linear initial-value problems.12 As a result, Wilson and Nickell have recently

Jan 1, 1969

By N. R. Pillai, T. R. Anantharaman

THE striking alterations in the constitution and structure of cast A1-Si alloys brought about by addition of small quantities of some elements, particularly sodium, have been extensively investigated. Many theories' have also been propounded to explain this so-called lLmodification" process, but their main concern has been more to throw light on the action of sodium than to arrive at general criteria underlying this interesting as well as intriguing phenomenon. Such theories have generally dealt with one or more, but not all, factors that seem to have a direct or indirect influence on the cast structure of A1-Si alloys, e.g.., formation of ternary eutectic, removal of oxides, hydroxides, and so forth, by fluxing, morphological changes in silicon, and inhibition of growth of silicon grains through colloidal action, compound formation, preferential adsorption, or reduction of surface tension. Kim and Heine2 were the first to study the modification of A1-Si alloys in a systematic as well as comprehensive manner and to list the important characteristics of a satisfactory modifier. According to them, an important property of the modifying element is its capacity to form a stable compound with silicon in the appropriate temperature range. Other subsidiary requirements are that the element should get adequately dispersed in the melt at the inoculating temperature and also must not form stable compounds with aluminum. The alkali elements, with the exception of lithium, fulfill these requirements most satisfactorily and have been shown to be successful modifiers by these workers. Sodium is the most effective of these and seems to be in a class by itself for various reasons noted by earlier investigators.37 None of the other elements tried by Kim and Heine, viz., Mg, Ca, Cd, Zn, In, Pb, P, As, Bi, S, and Se, showed any significant modifying action. We wish to report here briefly on substantial modifications in cast eutectic A1-Si alloys brought about by some V group elements, which do not comply with the Kim-Heine requirements. Aluminum and silicon used in the present work were indigenous and had purities of 99.8 and 99.5 pct, re-

Jan 1, 1969

By R. Ward Grosz

Reasonable and timely solutions to the complicated mathematical calculations involved in the mine planning, economic evaluation and slope stability studies related to the Ruth Mine would not have been feasible without the modern digital computers. The Ruth Mine was a prosperous underground copper operation in east central Nevada from 1915 to 1948, when rising mining costs finally exceeded the value of the ore. A substantial reserve of porphyry copper remained in the ground, but until recently all economic evaluations indicated that it was at best a marginal operation and a relatively poor risk of capital. There has been a steady progress in improving all aspects of surface mining since the first steam shovel started stripping overburden; however, developments in just two rather specialized areas proved to be critical in changing the Ruth Mine from a reserve of mineralized rock to an economically attractive orebody. One involved the progress made in recent years in the science of slope stability. Applying the latest techniques in this field to mine planning resulted in reducing the stripping ratio from 6.0:l to 4. 2:l. The other major factor was the efficiency and cost performance of large capacity haulage trucks which made it feasible to go down over 1100 ft. in an open pit and recover ore with a high stripping ratio economically.

Jan 1, 1969

By J. Hudy

An investigation has been made to determine the applicability of the Cendrex X-ray instrument for the measurement of the ash contents of washed bituminous coal products. The results obtained for selected products from five preparation plants in the northern Appalachian region are described. Emphasis of the study was placed on low-ash bituminous coals suit-able for utility or metallurgical markets. There was good agreement between the results obtained by conventional ash determination methods and by the Cendrex for all the products tested. A principal disadvantage of coal in competition with other fuels has been its inconsistent quality, especially with respect to ash content. Mine rnechanization and production from coalbeds of inferior quality have accentuated this failing. Present quality control methods in coal preparation plants are inadequate because (1) control samples taken over a production shift only serve to give the average quality of the product, (2) conventional sampling and analysis data are usually not available at the time a shipment is scheduled, and (3) the plant operator does not have the use of such data when the coal is being processed so that operating conditions may be adjusted for correcting variations in quality. Thus, the development and use of continuous-control monitoring instruments is of prime importance. The Federal Bureau of Mines arranged to investigate an instrument using x-rays for rapid continuous ash determinations. Developed by the Dutch State Mines and first described in 1958,' this instrument, which has found widespread acceptance in Europe, is designed to provide ash analysis of any desired coal feed or product stream with only a five-minute time lag between sample collection and analytical results. With limit controls, this unit could be used to feed back signals to preset the separating gravity of coal washers, control the ratio of various coals needed to produce an acceptable blend, and permit the sorting of a washed product into two qualities, one which is below and the other which is above a prescribed ash content. This unit can also be used discontinuously for rapid determination of coal samples taken in the usual manner. THEORY Di jkstra and Sieswerda of the Dutch State Mines studied the effect of chemical composition and ash content on the absorption coefficient of coal with the transmitted radiation method.' They found that with an X-ray wavelength of 1.79 A and values calculated for the absorption coefficients of ash and combustibles, the absorption coefficient of coal gave a linear relationship when plotted against the ash content. Thus, a method of measuring the absorption coefficient of coal could also be used to measure the ash content. However, practical attempts to measure the ash content of coal by comparing the intensities of the transmitted X-rays failed. The absorption coefficient of coal can also be obtained by irradiating a layer of coal with x-rays and measuring the intensity of the diffused reflection (backscatter method). The relationship between the diffused reflection and the ash content is actually hyperbolic but approximates a straight line over a narrow range. A measuring circuit that balanced the difference in intensity of the reflected radiation from a standard surface against that from a coal sample was used. By plotting this difference of reflected radiation against the ash content, a curve which approximated a straight line over a range of 8 percentage points of ash was obtained. DESCRIPTION OF EQUIPMENT The apparatus performs two distinct functions: (1) the preparation of the sample to a suitable size consist and moisture content for delivery to the x-ray element, and (2) the measurement of the ash content by the x-ray element. In a continuous operation the coal sample is delivered to a conditioner where it is dried to a maximum of 1% surface moisture and pulverized in a high speed mill to -48-mesh. This material is fed, at a rate of 250-300 gpm onto a rotating disk, shown in Fig. 1, where it is mixed with material already present. The mixture is rotated on the disk and

Jan 1, 1969

By C. van Dijk

In Sept., 1960, a steam-drive project was started in the solution-gas drive area of the Schoonebeek field. A part(ern of three five-spots and one four-spot was selected covering an area of 65 acres. The pay in the project area has good lateral continuity and dips slightly to the northeut; it is about SO ft thick and permeability increases from 1,000 and at the bottom to approximately 10,000 md at the top. The oil originally in place was 12.6 X 10' bbl. The oil has an in situ viscosity of about 180 cp. At the start of the steam drive the cumulative primary production due to. solution-ga.7 drive amounted id 4 Percent of the oil originally in place. Reservoir pressure had dropped about 120 psi and no significant primary re-.serves remained. Some 11.3 million bbl of steam (all steam quantities are expressed in barrels of water vaporized) have been injected, resulting in production of an additional 4.1 X I0 9bl of oil, or 33 percent of the oil originally in place. This corresponds to a cumulative oil-stearn rario of 0.37 bbllbbl. It appears that the steam preferentially moves r updip while liquids are produced mainly from downdip wells observations indicate that tile steam flows through only the upper part of the formation. The lateral steam distribution in the pattern is satisfacrory since several prodriction wells hardly reacted and, hence, cori tributcd little to the oil production. Production performance and results from material balance calcutlations agree satisfactorily with the results of large-,scale laboratory experiments. On the basis of these experirmental results the .steam drive, together with a cold water follow-up. is expected to bring ultimate recovery to a value of crt leas: 50 percent of the oil originally in place. No serious production problems have been encountered. However, due to mechanical fuilure, two old prodriction wells and one injection well had to be replaced. An extension of the. steam drive in this area is under connstruction. Introduction The Schoonebeek oil field, discovered in 1943 and developed after World War 11, is situated in the eastern part of the Netherlands. The main oil reservoir in this field is the Valanginian sand. A completely sealing fault divides this reservoir into two areas (Fig. 1): the southwestern part of the sand body where primary production is ob- tained by means of a solution-gas drive, and the remain. der where edge-water drive is the production mechanism. In the greater part of the field the reservoir consists of a single, unconsolidated sand body. The net thickness ranges from 30 to 100 ft and the top is between 2,400 and 2,800 ft below sea level. Formation permeability varies from approximately 10,000 md at the top to values of the order of 1,000 md at the bottom, and porosity is about 30 percent. The reservoir contains a paraffinic oil of 25" API gravity with an in situ viscosity of 160 to 180 cp. Initial oil saturation was high (85 to 90 percent). The relatively large quantity of oil in place (more than 10' bbl), and the low ultimate primary recoveries expected from this field — approximately 15 percent stock-tank oil initially in place (STOIIP) for the water-drive area and 5 percent STOIIP for the solution-gas drive area — clearly indicate ample scope for secondary recovery. Because ies-ervoir and crude characteristics made this field suitable for thermal secondary recovery, a hot-water drive project was started in the water-drive area about 10 years ago. A few years later a steam drive and an in situ combustion project were started in the solution-gas drive area. This paper deals with the performance of the steam-drive project, which was started in Sept., 1960, and which is still in operation. Design of Steam-Drive Project, An experimental investigation of the steam-drive process carried out by schenk in 19561 indicated that under schoonebeek conditions steam injection could be an attractive secondary recovery method. the findings and encouraging results of a pilot test in the Mene Grande field in venezuela,i led to the design of a steam-drive project in the schoonebeek field, Pruject Site and Pattern In 1958 the reservoir pressure in the solution-gas drive area had decreased to about 120 psi, and oil production rates of wells in this area had dropped to 7 to 10 B/D. The cumulative primary production was about 4 percent STOIIP, leaving an oil saturation of approximately 85 percent. In view of the large amount of oil left behind in the reservoir, the solution-gas drive area was selected for the planned steam-drive project. The area in the vicinity of Well S1 3 (Fig. 2) was considered to be suitable since it is at least partly isolated from the rest of the field by faults and the sand is relatively thick (about 80 ft).

Jan 1, 1969

By F. D. George, M. J. Salkind

Al3Ni whisker-reinforced aluminum was found to exhibit good Charpy impact toughness and little notch sensitivity even though its room-temperature tensile elongation parallel to the whiskers is only 2 pct. This impact behavior was maintained d liquid nitrogen temperature (-196"C). It is postulated that this behavior is due primarily to the presence of the continuous aluminum matrix which provides sufficient 10calized ductility in the vicinity of the crack tip to absorb considerable energy from the advancing crack. The impact behavior of Al-Alni was found to be quite anisotropic. Of six orientations studied, the transverse orientation having the notch normal to the whisker axis was found to exhibit the lowest impact energy, whereas the transverse orientation having the notch parallel to the whisker axis was found to exhibit the highest impact energy. A significant differnce was noted between the impact behavior of material containing needlelike whiskers and that containing bladelike whiskers. Only two of the six orientations studied exhibited complete fracture for the material containing needlelike whiskers. On the other had, most of the specimens containing bladelike whiskers exhibited complete fracture. It was postulated that the bladelike whiskers block transverse flow, thus reducing the amount of plastic deformation ahead of the crack tip. One of the more significant advantages of composite materials is the prospect of combining high strength with toughness. In general, toughness is associated with materials which exhibit considerable ductility and can deform plastically in the presence of a stress concentration. Very strong materials which resist plastic deformation generally exhibit low toughness. At first glance, then, it would appear as though strength and toughness are mutually incompatible so that useful engineering materials would have to be a compromise between the two. One approach to the problem of combining the high intrinsic strength of ceramics with the toughness of metals was to mix them together to form a cermet. Unfortunately, the toughness of cermets was found to be rather disappointing. Whisker reinforcement of metals, however, appears to be a more promising approach. It has been demonstrated that whisker-reinforced metals produced by unidirectional solidification exhibit enhanced strength due to the presence of high strength nonmetallic whiskers. The total strain capacity of these composites in the direction of fiber alignment is limited to that of the fibers, the matrix being unable to carry the load once the fibers have failed. A characteristic, then, of whisker composites is low ductility in the direction of whisker alignment, on the order of a few percent elongation. This low elongation, which is usually associated with brittle behavior, should not be taken as an indication of low toughness. Such a material can exhibit significant ductility in directions other than parallel to the fibers7 and can therefore possess significant intrinsic toughness. Toughness in a fiber-reinforced metal is derived from several mechanisms. The first is due to the toughness of the matrix itself. A continuous ductile metal matrix can act as an effective crack arrest medium by undergoing localized plastic deformation. Cracks initiated from the surface of the composite or from a brittle fiber failure must travel through the matrix before reaching another brittle phase particle. A second crack arrest mechanism peculiar to fiber composites is due to the fact that, as a crack travels through the matrix and approaches a fiber, the plastic deformation ahead of the crack tip will result in loading of the fiber. This causes the matrix shear strength in the plastic zone to be apparently higher, thus extracting more energy from the crack and diverting the crack at an angle to the original direction of propagation. A third crack arrest mechanism occurs in fiber composites which exhibit a weak bond between fiber and matrix. The idea was proposed by Cook and Gordons that if a crack propagating transversely in a fiber composite were made to turn and run along the fibers by decohesion of the fiber-matrix bond, then toughness would be imparted by the blunting of the crack tip and the creation of new surfaces. The last mechanism, interfacial decohesion, is commonly noted in naturally occurring fiber composites such as wood, bone, and bamboo, and has been observed in man-made composites such as glass fiber-reinforced resins,g silica fiber-reinforced aluminum," laminated steel," and tungsten and silica fiber-reinforced electroplated copper.'' The first mechanism, crack arrest by plastic deformation in the matrix, has been noted in tungsten wire reinforced cast copper." The purpose of this investigation was to quantitatively assess the toughness of a whisker-reinforced metal as a function of orientation. Previous investigation considered only cracks propagating nominally perpendicular to the reinforcement. In this investigation, crack propagation in three mutually perpendicular directions as well as three intermediate orientations was investigated. The system chosen for study was the unidirectionally solidified A1-A13Ni eu-tectic alloy which has a microstructure consisting of 10 pct by volume of A13Ni whiskers in a matrix of aluminum This material exhibits two different kinds of whisker morphology, depending upon the rate at which it is solidified.' At low rates of solidification (less than 2 cm per hr) the whiskers are bladelike, whereas at higher rates of solidification they are

Jan 1, 1969

By Claude H. P. Lupis, John F. Elliott

The classic Siez~erts' technique has been employed to determine the solubility of oxygen in liquid pure silver and the effect on the oxygen solubility of solute metal additions to liquid silver. The results indicate that oxygen deviates slightly from Sieverts' law in pure silver as = 3.9. Gold, platinum, md palladium decrease the solubility of oxygen in silver. First-order free energy, enthalpy, md entropy interaction coefficients are determined from the data. An important aspect in the establishment of the formalisms to be applied to represent the thermody-namic behavior of components of solutions is the accuracy of the experimentally determined thermody-namic properties. Wagner developed the first-order free energy interaction coefficient.' Recent papers2'3 extend the treatment formally to other thermodynamic properties and define the second- and higher-order interaction coefficients. A current communication4 considers in some depth the requirements and limitations imposed because of experimental error on the use of several formalisms now under active consideration in the field. This communication reports the results of an experimental study of the behavior of oxygen in liquid silver alloys that was carried out concurrently with the development of Refs. 2 and 3 with the hope of providing sufficiently precise information so that enthalpy and entropy interaction coefficients could be determined. THE EXPERIMENTS The Sieverts technique was employed using an improved design of the reaction bulb over that developed originally by Humbert5 and subsequently improved by pehlke6 and Weinstein.7 The Vycor reaction bulb as shown in Fig. 1 was air-cooled and contained a port for the thermocouple angling off to one side and a vertical port for connections to the gas supply and

Jan 1, 1969

By R. H. Tien

The accuracy of the method of steady-state approximation applied to the problems of heat transfer involving phase change (London and Seban's solution) is improved by defining a "modified heat of fusion" which accounts for the effect of specific heat. An analytical expression for this modified heat of fusion is derived in terms of the actual heat of fusion, specific heat, and surface temperature. Numerical solution for the freezing of iron indicates that the ratio of the modified heat of fusion to the actual heat of fusion increases from 1.04 to 1.4 as the surface temperature decreases from 2700° to 1600°F. The heat transfer process accompanying solidification is complicated owing to the moving solid-liquid interface where the latent heat of fusion is released. Exact analytical solution is available only for a few special cases of freezing problems such as Stefan's1 problem. In a commonly used steady-state approximation method, London and seban's2 solution, the effect of specific heat is neglected. This approximation, however, may introduce some degree of error in mathematical analysis when the temperature of the solid surface is much lower than the freezing temperature, as in the case of freezing metals. In a recent work by Robertson and Schenck,3 the problem of freezing a semi-infinite region with finite surface conductance was programed in a digital computer through the finite difference method. A family of curves for the correction of the London-Seban solution was given in a dimensionless plot for different ratios of heat of fusion and specific heat effect; it was shown that the London-Seban solution when applied to the freezing of a mild steel slab under some particular conditions gave results in error by about 10 pct in the solidification rate. Although the Robertson-Schenck computer result has improved the London-Seban solution to a great extent, lack of generality and the tedious computing procedure have restricted its application. In this paper a "modified heat of fusion" which accounts for the effect of specific heat is defined, and is used instead of the actual heat of fusion in the London-Seban solution. An analytical expression of this modified heat of fusion in terms of the specific heat, surface temperature, and actual heat of fusion is derived by comparing the solutions of Stefan and London-Seban for the case of freezing of a semi-infinite region with constant surface temperature. It can also be extended to the case with finite surface conductance, and the results show good agreement with the Robertson-Schenck computer solution. The expression of the modified heat of fusion, the accuracy of the London-Seban solution, and the temperature distribution within the solid are presented graphically in dimensionless form. Numerical solution for the freezing of iron indicates that the ratio of the modified heat of fusion to the actual heat of fusion increases from 1.04 to 1.40 as the surface temperature decreases from 2700° to 1600°F.

Jan 1, 1969

By Wolfgang Pitsch

INVESTIGATIONS on the above topic have recently been published by Shapiro and Kraussl and Jana and wayman in this journal and by Kessler and Pitsch.- Parts of the results in these papers are in good agreement with each other but there are also some results which are substantially different. The main points of these differences shall be briefly discussed in this communication. Shapiro and Krauss' measured single-surface traces of the habit planes of reversed austenite plates formed within the martensite crystals and deduced from these measurements two distinct habit planes A = (0.174; 0.309; 0.935) and B = (0.375; 0.545; 0.749) within 3 deg. Kessler and pitsch4 found instead by two-surface analysis7 one habit plane n = {0.23; 0.62: 0.75) with a real scatter of 6 deg. The latter observation of a single habit plane apart from a certain scatter is in agreement with most of the investigations of martensite transformations. The differences between these two results are so great that they cannot be explained by experimental error. The purpose of this communication is to show that the A- and B-poles of Shapiro and Krauss are due to an erroneous interpretation of a single-surface trace analysis. The difficulties in the interpretation of such single-surface trace analyses are due to the high symmetry of cubic lattices, from which it follows that there are in general twenty-four equivalent possibilities of plotting a single-surface trace in a pole figure. All these possibilities must be taken into account, when putting different traces from different plates together in order to construct the habit plane. Alternatively the poles of all the traces may be plotted in a unit triangle. This is done in Fig. 1 with twenty-nine traces of different reversed austenite plates, all of this were measured by Shapiro and Krauss in the same plane of a single martensite ther traces taken from Fig. 13 in the Shapiro and Krauss paper' which were determined by electron diffraction. The traces in Figs. 1 and 2 have to be compared with the habits A, B, and n which are under discussion. Therefore the great circles of the poles A, B, and n are plotted in Figs. 3, 4, and 5 together with a 3-deg scatter region for poles A, B and a 6-deg scatter region for pole n. Comparing Figs. 1 and 2 with Figs. 3, 4, and 5 one finds the statements which

Jan 1, 1969

By N. A. D. Parlee, I. D. Shah

The solubilities of oxygen in liquid Ag-Au, Ag-Pt, and Ag-Pd alloys have been determined in the range of 940° to 1200°C at 1 atm pressure of oxygen using an improved Sieverts technique. The additions of gold (0 to 57 pct), platinum (0 to 27 pct), and palladium (0 to 22 pct) to liquid silver continuously decrease the solubility of oxygen in liquid silver. Activity coefficients and interaction coefficients of oxygen in these alloys and have been calculated. The interaction coefficientscan be described by the following equations: Approximate second-order interaction coefficients (pM have been calculated. The contribution of the product, pMxNM, to In yo only becomes significant at high alloy concentrations. RECENTLY several investigators1"3 have redeter-mined the solubility of oxygen in liquid silver. In general the recent findings1" for Po, = 1 atm are in good agreement with the equation:4 log %O = 695.4/T + 1.086 [1] It is established1, 2 that the solubility of oxygen follows Sieverts' law reasonably well, at least up to 1 atm partial pressure of oxygen. It should be noted that a very slight deviation from Sieverts' law has been reported by Lupis and Elliott.3 Only recently has the solubility of oxygen in Ag-X binary liquid systems received much attention, and only the Ag-Cu, Ag-Au, Ag-Pt, and Ag-Pd systems have been studied. Par lee, Sacris, and zoellners reported that the solubility of oxygen in liquid silver increases progressively with the small additions of copper (0 to 0.5 pct) up to the point where the copper oxide phase appears. Lupis and ~lliott~ have reported that the solubilities of oxygen in liquid silver decrease progressively with the additions of gold, platinum, and palladium. However, their investigation was limited to 0 to 5 pct of these alloying elements, a range too limited to satisfy the requirements of the research program of this laboratory, where the solubilities of oxygen in the range of 0 to 30 pct were required for the calculation of chemical diffusion coefficients of oxygen in these alloys. The present paper deals only with the solubilities of oxygen in these alloys while another paper will deal with the diffusion of oxygen in these alloys. EXPERIMENTAL METHOD A Sieverts technique was employed which involved an improved design of the reaction tube, Fig. 1, as described briefly in a previous communication.4 Liquid melts (70 to 100 g) were held in Norton Alun-dum crucibles. The induction furnace used had a maximum output of 2.5 kw with very fine control. This induction system had excellent stability and the final equilibrium temperatures measured by calibrated Pt-Pt, 10 pct Rh thermocouples are believed to be within ±2C. These thermocouples were checked against the melting point of pure gold wire, and the results agreed with standard tables (±l°C). Each run involved a careful sequence of operations. The sample was heated to and held at about 900°C in vacuum for 3 to 4 hr with the thermocouple in a retracted position. After melting, the melts were usually exposed to a small amount of oxygen (0.5 to 2.0 cu cm) and then degassed slowly for 1 to 2 min to remove any oxidizable volatile impurities such as carbon and sulfur as CO, CO2, SO2, and SO forth. Argon gas was then allowed to enter the system. The protection tube containing the thermocouple was then

Jan 1, 1969

By P. C. J. Gallagher

A number of recmt determinations for the ratio of extrinsic to intrinsic stacking fault energy in fcc solid solutions are examined. Some of these arise from incomplete analyses which can yield only approxi?nate values for the ratio. Reliable results, on the other hand, obtained using extrinsic-intrinsic fault pairs, show that the extrinsic and intrinsic fault energies are essentially equal in several materials. There is some reason to believe that this finding is of general applicability to fcc elements and alloys. A wide range of values has been obtained for the relative magnitudes of the extrinsic and intrinsic stacking fault energies (yext and yint, respectively) in recently published studies in a variety of materials. In contrast, Hirth and Lothe' using a central force model have shown that out to tenth nearest-neighbor interactions the perturbation in energy caused by both types of fault is the same. Although the model used is not completely valid in metals, there is nevertheless some indication that marked variations of yext/nnt should not be observed from one material to another. In early work in Cu-A1, Cu-Ge, Ni-Co, and stainless steel all the deformation faults observed in the electron microscope were found to be intrinsic in nature, which led to an attitude that the extrinsic fault energy must be considerably greater than the intrinsic. Extrinsic faulting arising from deformation has, however, more recently been directly observed in Au-4.8 at. pct n;~ Ag-6 at. pct Sn and Ag-8 at. pct sn; Ag-7.5 at. pct In and Ag-11.8 at. pct 1n;"' pure silwer and Ag-0.5 at. pct ~n;' and Cu-22 at. pct Zn, Cu-30 at. pct Zn, and Cu-7.5 at. pct ~1.l' Multilayer loops containing extrinsic faulting have also been observed in quenched aluminum." While peak asymmetries in X-ray faulting probability studies were generally attributed to the presence of twins,Lelel2 has recently reinterpreted earlier X-ray data in Ag-Sb alloysU in terms of the presence of extrinsic faulting. The determinations of yeXt/yint made from the above studies are shown in Table I, with a brief description of the techniques employed. A number of the methods utilized are deficient in one or more respects, and the reliability of the values listed will be discussed. ~ele'~ recognizes that his approximate determinations of yext/yint assumes equal numbers of extrin-sically and intrinsically faulted dislocations. It is well-known, however, that such an assumption is not at all justified since extrinsic faulting has but rarely been observed in samples studied in the electron microscope. The only conclusion that should be drawn from the X-ray results at present is that the total intrinsic scattering cross section (i.e., the product of the width of the intrinsically faulted dislocations with their density) is approximately ten times greater than for extrinsic faulting in these particular samples. An important point is that the relative magnitudes of the energies cannot be inferred from results of this type, unless the intrinsic and extrinsic faults form with equal ease. One must recognize that, although a formation barrier may restrict the amount of extrinsic faulting which occurs, this in no way implies that the extrinsic and intrinsic energies should be different. It is unlikely that a worthwhile estimate of the relative densities of extrinsically and intrinsically faulted dislocations can be made at the high deformations present in X-ray samples. ~oretto,'~ from a statistical argument applied to the nonobservation of extrinsically faulted tetrahedra out of a large sample, concluded that yeXt/yint could not be less than -4.5. However, the present author feels that a high-energy formation barrier as just supposed also explains this finding satisfactorily and that no conclusion can possibly be drawn concerning the actual extrinsic stacking fault energy. The same argument also serves to explain the fact that extrinsic faulting has been relatively little observed in the electron microscope. Extrinsic-intrinsic node pairs and isolated extrinsic nodes were first reported by Loretto~ and subsequently by Ives and Ruff,' Gallaher,' and Gallagher and Wash-burn.' Ives and Ruff' found a wide spread in the ratio of extrinsically to intrinsically faulted area in the node pairs they observed, and drew the very tentative conclusion that yeXt/yint 2 2. They recognized that a straightforward comparison of the size of the faulted areas could provide no more than a qualitative result without a theoretical analysis of the dislocation geometry associated with extrinsic faulting. A theoretical

Jan 1, 1969

By Michael Epstein, Daniel E. Rosner

Fume nucleation sufficiently close to vaporizing suvfaces can augment net vaporization rates into cooler environments. Environmental conditions favoring large vaporization rate enhancements are briefly discussed and a previous theoretical treatment of this nucleation phenomenon is generalized to account for the self-regulating effect of condensalion-heat release within the boundary layer. Despite kinetic limitations on homogeneous nucleation, and latent heat release, non-diffusive condensate removal processes appear to make possible large enhancements in steady-state vaporization rates, provided surface temperatures are well below the boiling point. When condensed phases vaporize (or dissolve) into cooler media, the diffusion-limited mass loss rate can be strongly influenced by the process of nucleation/con-densation (or precipitation) within the thermal boundary layer. This condensation process (which typically leads to mists or fumes in the case of evaporation into cooler gases) has the effect of steepening the vapor pressure profiles near the evaporating surface, since the condensation zone acts as a vapor sink. However. the resulting enhancement in the diffusion-limited evaporation rate can be estimated (as first done by Turkdogan1 for the case of molten iron/nickel alloys evaporating into helium) only if one has independent knowledge of the critical supersaturation, sCrit(T), required to homogeneously nucleate the vapor.* In a recent reformulation and generalization of the theoretical model of Ref. 1 it has been shown that, when log sCrit is approximately linear in reciprocal temperature, rather simple expressions can be derived4 for the ratio of the actual rate of vaporization j" to either the minimum (no condensation) rate j"min, or the maximum (equilibrium condensation) rate j"max In the present communication we wish to briefly report on further developments and implications of the formulation of Ref. 4, with emphasis on i) environmental conditions favoring large enhancements in vaporization rate, and ii) the self-regulatory influence of condensation heat release (neglected in Refs. 1 to 4) on predicted vaporization rates. Additionally, we take this opportunity to correct several misprints appear- ing in Ref. 4, and comment on Elenbaas's recent criticism5 of Ref. 1. MAXIMUM POSSIBLE VAPORIZATION RATE IN PRESENCE OF CONDENSATION A nonequilibrium theory is of interest because of the very large difference between the minimum (no condensation) and maximum (equilibrium condensation) vaporization rate. The magnitude of this maximum possible enhancement can be shown quite clearly by combining a result of Refs. 3 and 4 with the fact that for most liquids there is a simple relationship between the molar heat of evaporation, A, and its boiling point, i.e., A/(RTBp) = C, where the constant C, often called the Trouton ratio, takes on values not very different from 11.* More generally, for any substance (including The Trouton ratio (which for water is 13, for methane, 10, and so forth) will be recognized as the ratio of the molar entropy change upon vaporization (at TBP or Ttransf) to the unlversal gas constant R. Its near constancy reflects the fact that the change in atomic order upon vaporization depends only weakly on the kinds of molecules involved. those that sublime under ordinary conditions) we can define a characteristic transformation temperature. Ttransf, by a relation of the form Ttransf =A/(CR), and then examine the maximum possible evaporation rates as a function of how far removed from Ttransf are the surface temperature, Tw, and ambient temperature, T. Subject to the assumptions: 1) equilibrium vapor pressure, pv,eq, everywhere small compared to prevailing total pressure, p, and 2) negligible effect of condensation heat on temperature profile, the maximum enhancement ratio was found (Eq. [17], Ref. 4) to be: where, for most vapors, Nu/NuD (the ratio of heat transfer coefficient to mass transfer coefficient for the same configuration) is a number near unity.* Ex- *An alternative derivation of the Nu = NuD special case of this equation. revealing its validity for arbitrary velocity/temperature profiles in a laminar boundary layer, is given in Ref. 3. amining this result for a "Trouton substance", one obtains the results shown in Fig. 1, constructed for C = 11. Since we are concerned with vaporization enhancements (j'max/J"min > 1) at surface temperatures below Ttransf, this area of interest is shown unshaded. One notes that at a fixed ambient temperature (hence, T/TtranSf) there is a unique surface temperature, 2T , at which j"max/j"min attains its peak value; moreover, the peak enhancement ratio, see dashed locus. Fig. 1, is: (NuA/NuD)(C/4)(Ttransf/T,). Hence, if Nu = NuD, when the ambient temperature is less than 1/4 of TtranSf the peak enhancement exceeds the Trouton

Jan 1, 1969

By J. Szekely, Robert G. Lee

Calculations are presented for the prediction of the combined radiative-convective heat loss from molten steel held in a ladle, covered by initially molten slag. A mathematical formulation is given and numerical solutions are presented of the resultant partial differential equation. It is shown that a slag layer of about 2 to 3 in. thick would prevent any significant heat loss from the top metal surface for Periods of up to 1 hr. For thinner slag layers (or in the absence of the protective slag) appreciable heat losses would occur; plots given in the paper allow the quantitative evaluation of the fall in the metal temperature due to this heat loss. The ironmaking-steel processing operation involves several intermediate steps where the molten metal is held or transfered in ladles. The quantitative assessment of the heat losses occurring under these conditions is of considerable importance in process design considerations since the temperature of the metal may have to meet rigid specifications in any given part of the processing sequence. While the metal is held in a ladle, heat is lost by two mechanisms: i) by conduction into the ladle walls; ii) by radiation and convection from the top surface. Calculations relating to the conductive heat loss, are readily made and information may be found in the literature both on data pertaining to metallurgical situations and on the techniques that are available for performing additional computations.'-3 The evaluation of the combined convective-radiative heat loss is less straightforward, especially when there exists a protective slag layer covering the metal. In this latter case, as the top slag surface loses heat partial solidification may occur and the latent heat thus given up by the slag may represent an effective barrier to the heat loss from the metal. A semiquantitative assessment of the role played by the slag in preventing heat loss from the metal has been made in a previous paper,4 where it was shown that a slag layer 6 in. thick would act as a near-perfect insulator for periods of up to 2 hr. However, this first paper reported on an essentially preliminary investigation that considered only one slag layer thickness; furthermore, the boundary conditions used for the calculations were somewhat restrictive since no allowance was made for changes in the metal temperature. The purpose of this second paper is to extend the scope of the preliminary investigation previously reported by: i) considering several slag layer thicknesses; ii) allowing for variations in the metal temperature; and finally iii) by performing calculations on the net loss from the metal rather than from the slag surface. FORMULATION Consider a slag phase extending from y = 0 to y = L1, covering a metal phase that occupies the region extending from y = Ll to y = L, as illustrated in Fig. 1. Denote the slag and metal temperatures by TS and Tm and assume that initially both slag and metal are molten, having a uniform temperature Ti. At time = zero the surface represented by the y = 0 plane (top slag surface) is brought into contact with cold air, the temperature of which is given as T Thus for time > zero, heat will be transfered from the slag to the air by natural convection and thermal radiation; as a result of this heat loss the slag temperature will fall and after some time a solid phase is formed and a solidification boundary (more realistically described as a zone) will move progressively from y = 0 toward y = L1. Once the region of temperature gradients, moving ahead of the solidification zone, reach the slag-metal interface (y = L1) there will be a net transfer of heat from the metal through the slag to the cold air, across the y = 0 plane. In the formulation of moving boundary problems, involving phase changes that require computer solution, it is convenient to represent both phases by a single equation, making allowance for the latent heat of solidification by assigning an appropriate temperature dependence to the specific heat content. Thus the energy equation for the whole of the slag may be written as follows:

Jan 1, 1969

By R. E. Robbins, H. R. Brager

THIS note describes an unusual result obtained in an electron microscope investigation of cavities produced in commercial type 304 stainless steel irradiated in a thermal reactor (ETR) at 290°C to a high neutron flu-ence, 6 x loz1 n per sq cm. The analysis of the steel used is given in Table I. The unusual result is that the calculated quantity of helium gas produced by transmutation during the irradiation, 12 ppm, is less than one-tenth the amount of gas necessary to balance the surface tension of the observed cavities produced in the steel during ETR irradiation. Therefore, the cavities would be expected to behave like voids; however, they behave like "gas filled" bubbles. The maximum quantity of helium produced in the steel by B" transmutation was 8 ppm. In addition, the amount of helium formed by (n, (1) fast neutron reactions with the major alloy constituents and nitrogen was calculated to be 4 ppm based on the tabulated differential or effective fission-averaged cross sections of elements in type 304 stainless steel irradiated in typical thermal reactors.', In previous investigations concerning commercial austenitic stainless steel irradiated in a thermal reactor, the amount of helium gas calculated to be produced by transmutation has always been significantly greater than the amount calculated on the basis that the bubbles were gas filled, PPs= 2y/R. The balance of the gas was assumed to be in bubbles smaller than could be detected on the electron micrographs. Accordingly, the bubbles were attributed to helium produced from transmutation of impurity boron by thermal neutrons. These bubbles were not uniformly dispersed in the matrix but were located principally at grain boundaries, or were radially positioned around a few precipitates located in the matrix and at the grain bundaries., 4 Voids have also been observed in austenitic stainless steel irradiated at high temperature to high neutron fluences in the EBR-I1 and Dounreay Fast Reactors. The voids were uniformly distributed throughout the matrix. The amounts of helium gas produced by transmutation during these irradiations were less than one-tenth the amounts needed to balance the surface tension of the voids formed by the fast reactor irradiation. The fast reactor results are quite analogous to those reported here for the ETR irradiated steel. Electron microscope examination of the fast reactor irradiated material after a 1 hr postirradiation annealing at 900°C disclosed that essentially all the voids had been eliminated. The authors stated that the voids had developed due to the high concentration of vacancies produced during irradiatin., 6, ' The microstructure of the ETR irradiated material as-tested at 75O°C, Fig. 1, contained about 2 X lot5 cavities per cm3 with an average diameter of 110A. With the Van der Waals b correction and y = 1500 ergs per cm2, 300 ppm He would balance the surface tension of the observed cavities. Stereoscopic examination of the electron micrographs showed that the cavities were neither preferentially located near the grain boundaries nor highly concentrated at specific regions or precipitates. The cavities were nearly uniformly distributed throughout the matrix, on dislocations and at grain boundaries. These observations imply that the cavities were not the sole results of B'O

Jan 1, 1969