Search Documents

By J. R. Goode

Up to one half of neodymium and praseodymium production, key ingredients in permanent magnets, is produced illegally. About 60% of cobalt production, critical in lithium-ion batteries and other applications, comes from central African states, and much of it is produced illegally. Tantalum, important in modern electronics, is also primarily sourced from central Africa and, again, much of it illegally. Illegal mineral production, as exemplified here, is often used to fund armed conflict. Furthermore, it is typically conducted with scant regard for human rights and environmental protection. It also leads to depressed prices which in turn limits the possibility of legitimate producers entering the market. Systems of traceability and provenance, along with cooperative supply chain participants can limit the production of illegitimate material and thereby improve conditions in the countries of origin and reduce downward pressure on prices.

By N. E. Garner

Laboratory experiments were conducted to investigate the cutting of a single diamond on limestone and shale under simulated down-hole conditions. A high-pressure chamber was modified by adding a rock-rotating device so that planing tests could be run. Weight on the diamond, diamond geometry and differential pressure across the face of the rock were varied. It was found that (I) high differential pressure reduced the volume of rock cut by the diamond at a given bit weight by strengthening the rock and changing the mode of failure; (2) a finely powdered rock left in the bottom of the grooves reduced the volume cut by as much as 50 percent; and (3) orientation of a single diamond about the axis of a drill point had considerable influence on diamond cutting efficiency. INTRODUCTION Diamond bits consist of many diamonds held in a matrix provided with water courses for fluid passage to clean, cool and lubricate the diamonds as they drill formation. Each diamond contributes its share to the overall effectiveness of a bit, but a lack of understanding of the performance of a single diamond has hampered efforts to engineer improvements in diamond bits. The need for those improvements prompted this study. This paper describes the findings of a laboratory investigation of the outting action of a single diamond where the differential pressure between the wdlbore and formation, diamond geometry and vertical force on the diamond were independently varied. The tests were conducted in a pressure chamber that provided relative horizontal motion (planing) between a rock and diamond under simulated down-hole pressure conditions. EQUIPMENT HIGH-PRESSURE CHAMBER WITH ROTATING DEVICE The high-pressure chamber used in previous rock mechanics studies' was modified for this study by addition of a rock-rotating device. The modified chamber (Fig. 1) differs from the old chamber in that the rock can be rotated to provide relative horizontal movement between the rock and the penetrator which may be a small element of a diamond or a drag bit. In this study the penetrator was a single diamond. The rock sample holder inside the chamber is mounted on a thrust bearing and is connected to the external drive mechanism by a stem that passes through two seal elements. The seal elements are needed to allow formation pressure to enter the bottom of the rock sample through a small hole in the stem. The Kapseal seal elements (Tefion boots backed by O-rings) provide low friction (hand free) even at differentials of 5,000 psi. The drive mechanism is a variable-speed transmission with a speed range from 0 to 30 rpm and an output torque of 2,800 in.-lb. A right-angle-drive gear box couples the drive to the stem. The new design has two independent pressure systems (formation and borehole) to simulate down-hole rock stress environment. Formation pressure enters from the bottom of the rock through a 6-in. diameter area, and borehole pressure acts over the remaining surface. A 14-in. layer of zero fluid loss oil-base mud poured over the top of the rock and an O-ring seal on bottom allow a differential between the borehole and formation pressures. Bayol 50 was used as the hydraulic fluid in the chamber. The differentials reported here were obtained by elevating the borehole pressure while keeping the formation pressure at zero.

By J. C. Edwards, C. C. Hwang

In the design for a fire or smoke emergency, a main concern is maintaining an evacuation path that is free of smoke and hot gases. In ventilated tunnel fires, smoke and hot gases may form a layer near the ceiling and flow in the direction opposite to the ventilation stream. The existence of the reverse stratified flow has an important bearing on fire fighting and evacuation of underground mine roadways, tunnels, and building corridors. In the present study, a computational fluid dynamics (CFD) program is used as a design tool to model floor-level fires in a ventilated tunnel. The fire is simulated by a diffusion flame of propane issuing from a circular hole on the tunnel floor. The computed profiles of velocity and temperature along channel cross sections are similar to those obtained experimentally. The computed reverse-flow layer length has been correlated with a densimatric Froude number with some success. A non-dimensional parameter involving the source fire intensity is also used for the correlation. These analyses show that an appropriate ventilation air current must be maintained to provide an evacuation path from the source fire clear of smoke and hot gases. It is believed that CFD modeling, employed correctly and interpreted carefully, can be a useful design tool for fire protection.

By Katja Reinharter

Arsenic can be found in different sulfidic copper concentrates and residues, which for several years have commonly been used in copper metallurgy, as the “arsenic-free” resources are getting rare. Due to the high toxicity the removal of arsenic in the copper smelting process is a very important topic. The typical chemical slag and sulfur attack on the refractory material is getting enhanced by the presence of arsenic. This work deals with post-mortem investigations of a magnesia-chromite brick and castable used in copper smelting furnaces showing an additional and increased chemical attack by arsenic. The knowledge on the wear behaviour is not only based on a detailed chemical and mineralogical characterization, but also on thermochemical FactSageTM calculations, which are carried out on provided post-mortem samples. Post-mortem investigations on used refractory materials represent an important prerequisite for the product development, as well as for special engineered lining concepts to support our customers.

By Alper Turan

The use of aggregate pier foundation systems has become widespread as the engineering practice moves towards cost-effective ground improvement solutions that provide accelerated construction schedules. Increased bearing capacity, effective settlement control and potential for liquefaction mitigation are commonly acknowledged benefits of aggregate piers. This paper studies the mechanisms by which aggregate piers interact with each other and with the surrounding ground when excited by seismic loads. A particular emphasis was placed on the response of a raft foundation supported on a site improved using aggregate piers. Three-dimensional nonlinear finite element analyses were conducted using commercial software package, ABAQUS. Non-linear soil behaviour is modelled using a three-dimensional total stress-based bounding surface plasticity model, which is able to simulate stiffness degradation and damping increase in the soil. The soil stratigraphy considered in this study was a 30 m deep soil deposit that comprised a very stiff base layer and a shallow soft surficial clay layer. Ricker Wavelet time histories with different frequency contents are used as the excitation force in the analyses. Various area replacement ratios (ARR) for aggregate piers are studied for a square plan layout. The results presented in this study are considered to be of interest to researchers and designers in evaluating the seismic ground response of the sites improved using aggregate piers.

By I. O. Otunniyi

The natural hydrophobic response flotation method for beneficiation of printed circuit board comminution fines (PCB CF) still requires many studies to further basic understanding of the process and the flotation traits that the PCB CF presents. Varying particle size distribution (PSD) and water recovery under the scheme show that higher mass pull, obtained at the finer PSD, gave lower water recovery but higher value loss. The reverse also holds: the least mass pull at the coarsest PSD used, -106 + 75 µm, gave the highest water recovery per unit mass pull with the highest reverse recovery. This is unlike what obtains in primary resources processing, where higher water recovery is often with increased fines entrainment. The reverse recovery loss here is, therefore, more due to entrapment with the heavy mass pull characteristic of the method than to the entrainment of value with water. Other peculiarities of the system observed are discussed, with the goal of improving the performance of the process, and broadening the general knowledge of flotation systems.

By Zongliang Zhang

Electrowinning is a significant step in metal extraction processes. Because the nature of electrowinning is to reduce metal from its oxidized state, substantial energy consumption accompanies it. Electrowinning processes of different metals have much in common—the deposition of the metal on the cathode and the counter-reaction on the anode. An advanced model was developed that can be applied to a variety of metal electrowinning processes. A model for specific metals can be readily obtained by modifying parameters such as potential, kinetics, and side reactions in the model. The model incorporates mass transfer, gas-liquid flow, electrochemistry, and deposition, making it a comprehensive electrowinning model. The model is applied to copper, nickel, and zinc and verified with the experimental results. In addition, the parameter settings and results of different metals were compared, showing that this model has great versatility and accuracy, thereby making it an efficient and effective research tool.

By D. Berberick

Hecla?s Lucky Friday Mine is a deep, underground producer of silver, lead and zinc in the Coeur d?Alene mining district of northern Idaho. Main access into, and production from, the mine is via the Silver Shaft, a 5.5-m (18-ft) diameter, concrete lined shaft sunk by J.S. Redpath to an original depth of 1,890 m (6,200 ft) in the early 1980s. The shaft is divided into four compartments (Fig. 1), with sets made of 225 mm x 175 mm (9 in. x 7 in.) hollow structural section (HSS) steel buntons and 150-mm x 150-mm (6-in. x 6-in.) HSS dividers spaced every 4.5 m (15 ft) down the length of the shaft. Two skip compartments on the east side of the shaft are equipped with 125-mm x 150-mm (5-in. x 6-in.) HSS steel guides and 9-t (10-st) skips with three deck trailer cages for men and materials. The west side of the shaft is divided into a manway, and a fourth compartment that was originally designed for a service cage. Shaft services consist of a 250-mm (10-in.) a compressed air line and 150-mm (6-in.) sandfill line along the south side of the shaft, and 250-mm (10-in.) discharge (gray) water and 150 mm (6 in.) fresh water lines on the north wall. The mine?s primary 13.8 kV power lines, as well as communication lines are located on the west wall of the shaft. The production hoist is a 3.7-m (12-ft) diameter, double drum, double clutch, 2,238 kW (3,000 hp), direct current unit equipped, at the time, with thyristor drives. The 30-year-old drives would prove to be problematic during the rehabilitation project, leading to an upgrade to modern ABB drives later in the project, after the shaft rehabilitation was complete. The shaft had been subjected to the typical wear and tear of 30 years of operation, including intermittent sandfill and muck spills, and, in the lower areas, some corrosion. During those 30 years, the west compartment, originally designed for the service cage, could only be accessed from a work deck on top of the skips or from the manway.

By Yousef Ghorbani, Abolfazl Danesh, Fatemeh Kazemi, Ataallah Bahrami, Morteza Abdollahi, Mohammad Raouf Hosseini

This study aims to investigate the effects of the operational variables on the concentrate grade, recovery, separation efficiency and kinetic parameters of the copper flotation process. The results indicate that concentrate collecting time and pulp density were the parameters with the greatest effects on the concentrate grade. Considering copper recovery, the concentrate collecting time, collector dosage and pulp density were the most significant variables, in decreasing order of importance. The separation efficiency was found to be mainly influenced by the concentrate collecting time. This study can be helpful in identifying the parameters with the most effects on flotation performance and, consequently, the overall process efficiency, including recovery and separation efficiency. Such an outcome can be used as a decision-based tool to apply the necessary changes in a real-life industrial operation in a more effective way to improve the quality and quantity of the flotation concentration.

By Ho Jung Heo

The proper control of slag composition is a dominant factor affecting yield of iron as well as dephosphorization efficiency in an electric arc furnace (EAF) operation. Moreover, thermophysical properties of slags such as viscosity are strongly affected by the changes in slag composition. Slag foaming, which is also a function of viscosity, is directly connected with the yield of iron and thus optimum slag chemistry are highly required in view of thermodynamics and physical properties. Therefore, we analyzed the slag composition in commercial melt shop (70 ton EAF) with the aim of higher yield of iron as well as higher dephosphorization during EAF process.

By Beate Ina Jenssen

Ferric iron removal from the nickel electrolyte at Glencore Nikkelverk AS in Kristiansand, Norway, is achieved by precipitation of iron hydroxide followed by filtration, and it is desirable to improve filtration properties of the precipitated iron hydroxide in order to decrease the filtration time. The effect of increasing the residence time from the current 45 min to 75 min and temperature from 65 °C to 90 °C on the filtration properties of the precipitated iron hydroxide (akaganéite) was studied by simulating the industrial process conditions in a continuous reactor setup operating at steady state, using ferrous-containing process solution as feed material. It was possible to decrease the filter cake resistance by both increasing residence time and temperature, but temperature gave the most pronounced effect and reduced the filter cake resistance by one order of magnitude.

It is widely accepted that particles stabilize flotation froths and that stable froths result in improved flotation performance. Predicting the effect of particle addition on froth stability is, however, challenging. Furthermore, there is currently no technique that allows the prediction of the effect of different flotation reagents on froth stability, without carrying out froth stability experiments.

The annual General Meeting of the Branch was held in the St. Helena Club on Wednesday, 4th August, 1971 at 8.00 p.m. Mr N. A. Honnet (Immediate Past Chairman) was in the chair. There were also present Mr V. C. Robinson, President of the South African Institute of Mining and Metallurgy. Nine Fellows: Messrs J. R. F. Handley, C. J. Isaac, J. Lorenzon, J. M. Meyer, C. Mostert, D. A. Smith, S. L. S. Swart, I Thomas and R. G. Williams. Five members: Messrs R. Campbell, A. H. Edwards, K. Pearce, G. P. Rundle and R. B. Sutherland. Five graduates: Messrs J. E. Andrews, E. J. Shillaber, M. H. Smith, T. R. Stromsoe and D. J. van Niekerk. Two associates: Messrs A. Paschalides and D. S. Robinson. Eight visitors. Total present: Thirty-one. MINUTES OF THE PREVIOUS GENERAL MEETING As the minutes of the General Meeting held on the 16th June, 1971, had not been published in the Journal, the Chairman moved that the adoption and discussion thereof be deferred to the next General Meeting. This motion was adopted unanimously. The minutes of the General Meeting held on the 10th February, 1971, were taken as read. The adoption, which was proposed by Mr D. A. Smith and seconded by Mr J. M. Meyer, was carried unanimously. At Mr Honnet's request, Mr V. C. Robinson took the chair during the presentation of the Chairman's Annual Report. CHAIRMAN'S ANNUAL REPORT 1970/71 1970/71 Committee-Office Bearers: Chairman - J. N. Saunders Vice Chairman - G. Y. Nisbet Immediate Past Chairman- N. A. Honnet. Due to the sudden retirement of Mr J. N. Saunders, the committee, at a meeting on the 3rd February, 1971 elected Mr G. Y. Nisbet as Chairman with Mr J. M. Meyer as Vice Chairman. After the transfer of Mr Nisbet to C.D.M., the committee, as a meeting on the 12th May, 1971, decided that Mr Meyer hold the Chair in an acting capacity for the remainder of the year. Committee members: Messrs C. J. Isaac, J. G. Kirchener, L. O. W. Leimer, R. S. Pearson, I. G. Thomas and A. von Holdt. Transfers of several office bearers and committee members resulted in Messrs J. M. Meyer, D. N. Smith and G. C. J. Young being co-opted onto the committee.

By Manuel Calderón

Chile is responsible for 30% of the global copper production; the world ́s largest. However, significant challenges are faced in this industry; declining grades (1.13% in 2002 to 0.61% in 2015), high cost of energy, scarcity of water, and environmental concerns. Mining trends suggest the future of the industry will be focused in the underground space, specifically due to the depletion of economically viable surface mineral deposits. For example, the largest block cave mineat Codelco ́s Chuquicamata mine is expected to reduce energy costs by 50% and dust emissions by 97%compared to the open cut mine. To access these underground deposits, critical infrastructure is required; horizontal tunnel developments, extraction levels, vertical raises, and ore passes with the on-time delivery of the highest quality outcomes being a requisite for successful project delivery (Lift 1 of 3 at Chuquicamata Underground has over 1000drawbells).One aspect that makes this scenario even more challenging for deeper underground operations is the increased risk profile (people, equipment, and geotechnical) that forms part of the general environment. Furthermore geotechnical risks should be considered and they include rock burst and windblast. The safe interaction with this environment must be sustainable throughout the full life of the project. This paper presents the Enaex group approach to addressing the blasting challenges and outlines the solutions created. Aspects such as rock mass damage occurrence as a function of blast vibration frequency content has been studied, as well as the development of blasting methodologies for guaranteeing the full 100% extraction of all infrastructure blasts, including the all-important draw bell volumes. The development of such infrastructure volumes in one single event using electronic detonators and the associated achievement of ambitious productivity goals, are described as a critical part of the successful construction of one of the most ambitious UG projects currently present on the world.

By R. C. Davis, R. F. Berry, G. W. White, B. T. O’Connell, L. A. Stacha

The formation density logging tool provides a radioactivity measurement that yields formation densities in situ. The relationship between bulk density and porosity is well understood. With knowledge of grain and fluid densities, porosity may be computed from the indicated formation bulk densities in shale-free formations. The porosities thus determined may be used with a resistivity log for water-saturation determinations. One technique utilizes a density vs resistivity plut. A plot of density vs Sonic-log transit time is used for porosity determinations in shaly sands. This method is particularly suited to wells drilled with oil-base or salt muds. In shale-free formations, comparisons of density values nnd neutron-log readings are used to identify lithologr and, thus, to select appropriate values of grain density for porosity computations. Through the use of Formation Density, Sonic and neutron logs, the interpretation problems caused by complex matrix lithologies are simplified. Other applications of the Formation Density log are found in the identification of minerals in evaporite deposits and in yield determinations of "oil shales". INTRODUCTION Widespread field application has demonstrated that the measurement of formation density is a useful and revealing technique for well logging. Responding to the bulk density variations of the formation, the Formation Density log affords valuable data on the porosity, lithology and fluid content of the formations. Although not without its own limitations, the Formation Density tool may provide effective interpretation under conditions that hamper or preclude the use of other porosity devices. When used in conjunction with other devices, such as the Sonic or neutron, complex matrix conditions may be better understood, defined and evaluated. A previous paper' described the valuable contribution of the Formation Density tool in the logging of empty, or gas-filled, holes. The purpose of the present paper is to describe applications of the device in liquid-filled holes. THE FORMATION DENSITY TOOL The Formation Density tool was devised to measure the bulk densities of formations in situ. Actually, the device responds to electron density, but in the overwhelming majority of cases of interest, the electron density is very closely proportional to the bulk density. In other cases appropriate corrections can be made if necessary. Comparisons of log-derived densities with true densities for the more common matrix minerals are shown in Table 1. The tool consists of a gamma-ray source and detector mounted in a skid which, during the logging operation, is in contact with the borehole wall (see Ref. 1 for schematic). Gamma rays are emitted by the source and diffused through the formation. The number of diffused gamma rays reaching the detector at a fixed spacing from the source are counted. An increase in counting rate by the detector indicates a decrease in bulk density in front of the skid. Conversely, a decreased counting rate signifies an increase in bulk density. A measurement of hole diameter, between the face of the skid and that of the eccentering arm, is recorded simultaneously with the measurement of diffused gamma rays. Since the tool is a wall-contact device, the size of the borehole has little influence, provided good contact is maintained with the wall of the hole. Only in extremely large or small boreholes does the actual diameter become significant. However, for most accurate measurements the borehole size should be taken into account. Fig. 1 presents the departure curves for the conversion of counting rate to bulk density for liquid-filled holes of various sizes. It is of utmost importance that the skid maintain good contact with the wall of the hole. If, due to roughness or caving, there is mud between the skid and the formation, the tool will include in its response the relatively low-density mud, and the apparent density reading will be too low. Normally, permeable beds do not cave appreciably, so that reliable readings are usually available for zones of interest. Nevertheless, the caliper reading over the zone of interest should be examined for any wall roughness. When mud cake exists between the skid and the formation, the Formation Density log sees it as part of the formation. When thick mud cakes exist, a correction is required. For best results in wells having thick mud cakes,

By Wing Ka Ng

Substitution of pulverized coal injection by solid biocarbon fuel has the potential to achieve substantial reduction in GHG emissions associated with blast furnace ironmaking. A systematic evaluation was conducted on the performance of solid bio-carbons produced from a single raw biomass source using different pyrolysis technologies. The potential of solid biocarbon in reducing GHG emission of the blast furnace ironmaking process is strongly influenced by its O/C mass ratio. In order to achieve substantial GHG reduction, it is necessary to replace coal injection by solid biocarbon having a low O/C mass ratio. The pyrolysis technology and conditions employed are key parameters influencing O/C mass ratio of the resultant solid biocarbon. Close collaboration between steelmakers and solid biocarbon producers is critical for producing suitable solid biocarbon fuel to replace pulverized coal injection. A successful partnership should pave the way in achieving substantial reduction in GHG emissions of the blast furnace ironmaking process.

By Charles R. Stewart, William W. Owens

Reservoir performance predictions based on laboratory core test data assume that fluid flow is laminar for the laboratory test. A study has been made to determine the validity of this assumption for laboratory tests on various types of porosity found in producing limestone formation. Data are presented which show that turbulence and slippage can occur during laboratory tests on hetero-geneous-type porosity limestones, thus causing serious errors in measured single-phase permeabilities and two-phase relative permeability characteristics. In single-phase flow tests it is possible to eliminate turbulence and correct for slippage or to eliminate both factors by controlling test conditions. It is not always possible to control test conditions and thereby eliminate turbulence and slippage in two-phase .flow tests. A correction method is presented which can be used to calculate the true two-phase laminar flow relative permeability characteristic even though furbulence and slippage exist. .INTRODUCTION It is customary to make use of Darcy's law and modifications of this law, together with laboratory data on formation core samples to predict the performance of producing reservoirs. Such predictions are based on an assumption that fluid flow is in the laminar or streamline region for the laboratory test. It was the purpose of this inves- tigation to determine the extent to which turbulent flow may occur in laboratory fluid flow tests on hetero-geneous porosity limestones. Considering that turbulent flow conditions might exist in some laboratory fluid flow tests, additional emphasis was placed on the development of a method to correct for turbulence when laminar flow conditions could not be attained. FLUID FLOW CONCEPTS FOR POROUS MEDIA The Influence of Pore Geometry on Fluid Flow One of the more important factors influencing fluid flow in porous media is the geometry of the pore space which includes such characteristics of the pores as size, shape, distribution, roughness, uniformity, etc. In general, oil- and gas-producing formations can be divided into two broad types on the basis of pore geometry. One has been called sandstone-type porosity media, which is characterized by a small range in pore size, uniformity in shape of the pores, smooth pore surfaces and a regular and uniform distribution of pores. The other type has been called heterogeneous porosity media and is usually limited to the dolomites and limestones. This type is characterized by a wide variation in the size, shape, and distribution of the pores and rough, irregular pore surfaces. It is therefore apparent that conditions are much more favorable for turbulent flow* in heterogeneous-type porosity media than in sandstone-type porosity media. Interrelationship Between Turbulence and Gas Slippage In studying the problem of turbulent flow in laboratory tests on porous media, it is necessary to be aware of the interrelationship between slippage and turbulence for gas flow. As a result of slippage or the Klinken-berg effecta, apparent perrneabilities to gas are greater than the true value because there is no stationary layer of gas in contact with the walls of the flow channels. Gas slippage decreases as the mean free path of the gas molecules decreases. Since the mean free path of any gas decreases with increasing density, increases in static pressure result in lower apparent gas permeabilities. However, a reduction in gas permeability can also be due to turbulence. Therefore, in studying only turbulent flow in porous media, it is necessary to hold gas density, and slippage, constant or to reduce slippage to a negligible value by operating at high static pressures. Presentation of Laminar and Turbulent Flow Data A graphical relationship between permeability and a pseudo-Reynolds number, N,, will be used to show the two types of fluid flow, i.e., laminar and turbulent. The usual graphical method for such a description has been the use of friction factor-Reynolds number charts4. On such a logarithmic diagram, the laminar region appears as a straight line having a slope of 45 degrees. As the friction factor decreases and the Reynolds number increases, the turbulent region is reached and appears as a deviation from the 45-degree slope line. However, in petroleum engineering literature resistance of por-

By Sebastiaan Peelman

NdFeB permanent magnets are the best available magnets used in many technology applications. However, at their end-of-life (EoL) most of magnets and the contained REEs are lost during the recycling of the bulk metals. The REEs are classified as the most critical raw materials in the European Union, and recycling of REEs from EoL products will reduce their criticality and contribute to the sustainability of REE. Various technological routes have been reported, but most of the methods are effective for highly concentrated magnets or magnet scrap, which is greatly dependent on expensive pre-dismantling processes. This paper presents various innovative metallurgical solutions to the effective REE recovery from current industrial practice for WEEE recycling, including the ferrous scrap from WEEE shredder products and shredder residues from computer hard disk drives. Both hydrometallurgical and combined hydro- and pyrometallurgical REE recovery routes are developed after demagnetization and physical upgrading.

By K. A. Slagle, D. K. Smith

weight obtained. Additives used in conjunction with salt in these slurries have included silica flour, calcium ligno-sulfonate and cellulose retarders, granular lost-circulation materials, bentonite and selected low-water-loss additives that are not significantly deteriorated by the presence of the chloride ion. In some other areas of South Texas, the salt-saturated slurries have been used quite extensively for improvement of the flow properties of slurries and attainment of better circulation characteristics at lower displacement rates. Concurrently with this property, the protection of shales and shaly sands is also realized, as well as useful retardation of the slurry. Resultant superior cementing jobs have been indicated by both communication tests and acoustic logs for bonding to pipe and formation.' In one section of Louisiana, a major oil company has been using salt-saturated API Class A cement with calcium lignosulfonate retarder for cementing through the Miocene at 9,000 to 10,000 ft. This is another of the situations where interbedding of sands and shales exists, creating difficulty in maintaining formation competency when a fresh-water slurry contacts the clay minerals of the formation. Further work has also been done in the shaly Miocene formation at 13,000 to 15,000 ft where fairly close water or gas contacts are encountered. Indications thus far are that better segregation of these various fluids is obtained by use of the saturated salt slurries because of their improved formation bonding characteristics. In addition to the properties of salt in this situation, attainment of turbulent flow at minimum displacement rates has also been accomplished by use of an additive to help provide exceptional dispersion and viscosity reduction of the slurry. Another oil company was encountering considerable expense in completing wells in Southwestern Louisiana due to extensive block squeeze requirements for effective separation of zones. A very effective mud program was being used to minimize washout in the shale sections and, apparently, a nearly gauge-size hole was being obtained. However, primary cementing results with fresh-water slurries were generally poor. On a few occasions when slurry was actually circulated to the surface, large pieces of shale formation were brought out of the hole with the slurry, indicating a severely water-sensitive, sloughing formation. Inhibition of shale heaving was being accomvlished in the drilling program, and immediately indone ipon circulation of the fresh-water slurry even though it contained a low-fluid-loss additive to reduce filtrate damage in the sands. The subsequent change to salt-saturated slurry yielded 11 successful primary cement jobs out of 12, compared to the previous success ratio of practically zero. Since these were deep, high-temperature wells in the range of 13,000 ft with high-pressure zones necessitating 17.5-lb/gal fluid densities, the slurry used was API Class E cement, silica flour, weight material, retarder, salt saturation (which also reduced the amount of weight material) and maintenance of low fluid loss by use of a salt-compatible additive. Other salt slurries have been used to a limited extent in this same general area for similar problems at depths ranging from 5,000 to 17,000 ft. In the shallower wells, the cement has usually been API Class A where the salt functions as a retarder, and in the deeper wells API Class E cement is used with the additional salt advantage being its increased slurry weight and inability to dissolve salt stringers. A considerable number of squeeze jobs have also been done on older wells using the salt-saturated slurries with very good results. MID-CONTINENT In North Texas, salt-water slurries have been used for cementing the Woodbine sand, Strawn sand, KMA sand and Pettit lime. Shales surrounding these formations have created the same difficulty in obtaining separation of producing zones that has been the problem in other areas. Depths in this area range from 3,400 ft for the Strawn to 7,400 ft for the Woodbine, and concentration of salt has varied accordingly. In the deeper wells, where retardation is desired, saturated salt-water cement is used; for the shallower wells, in order to provide shorter waiting-on-cement times, the amount of salt has been 18 per cent by weight of the mixing water. Results have been excellent with no reported failures on any of the salt cement jobs; where acoustic bond logs have indicated indifferent bonds previously, they are indicating very good bonding for the salt slurries. In Oklahoma various shales of Pennsylvanian age exhibit a high degree of sloughing in the presence of fresh water, causing severe washouts above and below sand formations which it is desired to isolate. This situation exists to some degree in practically all parts of Oklahoma and includes formations of other ages such as the Wood-ford shale. For the past few years, salt-saturated cements and displacement rates as high as practical have been used as a remedy for this problem, with very good results. The Layton and Bartlesville formations are two examples of shaly sands where saturated slurries have been helpful. In one area where five wells were drilled through this type of problem shale without obtaining a satisfactory primary cement job, a change was made to salt-saturated cement preceded by a suitable chemical wash for the drilling mud involved. Acoustic bond logging indicated excellent bonding, and final completion bore out this result by being trouble-free. This type of slurry has also been used extensively on squeeze jobs where shales have been heaving around the producing formations. Predominantly, the basic slurry has been either API Class A cement or a pozzolan cement—although, as deeper wells are being drilled, the use of salt in Class E cement is also increasing. Salt cement in West Texas has been used primarily to help prevent channeling through the shales around the Delaware sand, Queens dolomite and Hope lime. Many of the shaly and dirty sands of this area are sensitive to the filtrate from a fresh-water cement. Salt at 16 to 18 per cent by weight of mixing water has been added to centent, and has been effective in controlling formation damage and communication between zones in these formations. Use of these lower salt concentrations is dictated in this area by the relatively low formation temperatures where retardation of the slurry would create unduly long waiting-on-cement times. Also, quite a bit of cementing has been done in this area using salt concentrations in the accelerating range— that is, 2 to 5 per cent by weight of water. Specifically, these concentrations have been used in coiljunction with high percentages of gel to overcome the retarding effect of the calcium lignosulfonate dispersant, although there are probably several shales where these concentrations could provide some degree of formation stability. On several occasions, the salt has also been used to lower the critical velocity or rate for turbulence with the slurry, particularly in the pozzolan cements. On wells in the Hope lime, it has usually been necessary to squeeze the shale above and below the lime to get a water-free completion. Use of salt-saturated slurries has largely eliminated this

By J. G. Richardson, K. H. Coats

During the initial growth of a gas bubble in an aquifer storage reservoir the injected gas tends to override the water. The resulting low displacement efficiency and high rate of gas travel down-structure make it difficult to maintain water-free production. This paper illustrates the use of two-dimensional, two-phase calculations to simulate this gas-water displacement. Calculations were performed for a stratified aquifer and for several homogeneous sands differing in permeability, porosity and thickness. Results are discussed and compared with performance predictions obtained from the simpler Buckley-Leuerett and Dietz formulas. This comparison indicates the necessity for the two-dimensional calculations in realistically simulating the gravity override of the water by injected gas. INTRODUCTION In recent years natural gas has been stored near markets in aquifers where insufficient storage capacity is available in depleted fields. Operators of these aquifer storage reservoirs have encountered technical problems relating to the gas-water displacement accompanying initial growth of the gas bubble. Injected gas tends to override the water, with a resultant low displacement efficiency and high rate of gas travel down-structure toward spill points. Low displacement efficiency makes if difficult to sustain water-free gas production. Sometimes the fingering of gas down-structure may be so pronounced that the injection rate must be severely curtailed, which excessively lengthens the time required for bubble growth. This paper is concerned with estimating — for given aquifer characteristics and fluid properties — the displacement efficiency, rate of gas movement down-structure and rate of gravity drainage of water behind the gas front. Several published articles relate to simulative capability necessary to handle this problem. The Dietz formula1 and Buckley-Leverett2 method have some applicability to the problem; Woods and comer3 reported one-dimensional, radial calculations which accounted for the two-phase flow of gas and water. Most recent research directed toward understanding aquifer behavior in relation to gas storage has concentrated on single-phase flow in aquifer.4 Douglas, Peaceman and Rachford5 presented a method for calculating multidimensional two-phase flow in reservoirs; their method has been applied in work reported by Nielsen and Tek,6 Blair and peaceman7 and Goddin.8 Refs. 5 and 7 show comparisons between experimental data and calculations similar to those employed here. One purpose of this paper is to illustrate the use of two-dimensional, two-phase flow calculations in simulating the difficult problem of water displacement by gas in a vertical cross-section. The pronounced fluid density and viscosity differences combine with the disparity in horizontal and vertical dimensions of the cross-section to pose one of the more difficult problems in reservoir simulation. The calculations described account for capillary and gravity forces, relative permeability and reservoir heterogeneity. An example reservoir is described and calculated results are presented for a variety of injection rates and values of permeability, reservoir thickness and dip angle. A second purpose is to compare the two-dimensional calculations with results obtained from the Buckley-Leverett and Dietz formulas. For this reason most calculations were performed for a simplified example reservoir to which the latter techniques might reasonably be applied. However, the computer program employed applies equally well to cases involving arbitrary spatial variations of permeability, porosity, reservoir thickness and dip angle.