Search Documents

By F. Liu, H. A. Meier, D. M. Valiveti, P. A. Gordon

Hydraulic fracture is a frequently used technique in the Oil and Gas industry to create high permeability pathways for extraction of hydrocarbons. Numerical simulation of this method poses challenges such as representation of 3D fractures, propagation of randomly oriented fractures in 3D media, enabling fluid leak-off from fracture to poro-elastic medium. We have developed a numerical model using the Generalized Finite Element Method (GFEM) to simulate fluid driven fracture in a poro-elastic medium that addresses these challenges. In the GFEM framework, special enrichment functions are used to augment the solution space in the displacement and pore-pressure fields in order to accurately capture the response in the vicinity of the fracture. Adaptive refinement of the finite element mesh is also employed to facilitate accurate and efficient solution of the model. Our model fully couples fluid flow in the fracture, represented with the lubrication approximation, to the geo-mechanical behavior of the bulk poroelastic rock, comprised of an elastic rock skeleton saturated with fluid. Fracturing fluid exerts traction on the fracture surface and is also allowed to leak-off into the poroelastic medium using a transfer function approach. Stress intensity factors (SIFs) are evaluated based on the Contour Integral Method, accounting for the fluid pressure on the fracture surface. The SIFs are used to govern a full 3-D evolution of the fracture. Verification of the model is performed for known behaviors for the poroelastic response and fluid flow in the fracture.

Jan 1, 2015

By J. Lorengo, C. H. Bucknam

"Standardization of the Inductively Coupled Plasma Mass Spectrometry (ICPMS) method for gold in cyanide solutions was undertaken to provide a more sensitive method to quantify the gold content than the atomic absorption method standardized in 1994 (E1600). The gold concentration range covered is from 0.001 mg/L Au - 0.3 mg/L Au (0.00005 toz/t Au – 0.009 toz/t Au). Interlaboratory testing was carried out with the assistance of the American Society of Testing and Materials (ASTM) International on two committees to determine the within laboratory and between laboratory reproducibility. Three barren solutions from metallurgical operations were measured in triplicate to satisfy statistical requirements for Committee E1 on Analytical Chemistry of Metals, Ores and Related Materials. A synthetically prepared matrix representing the metallurgical processing solutions was used to prepare Youden pairs to satisfy the requirements of Committee D19 on Water. Results of the standardization program are presented. INTRODUCTION Analysis of process solutions for gold concentration requires accurate and sensitive analytical methods in order to manage process control and metallurgical accounting. Flame atomic absorption spectrophotometry (FAAS) has been the work horse of the assay lab for this duty for decades. In 1994 the FAAS standard method was approved by the American Society for Testing and Materials (ASTM) (1). Six process solutions were collected and preserved by measuring the pH and adjusting to greater than pH 10, if necessary. Duplicate solutions were provided to eight laboratories in random order for analysis by FAAS covering the gold concentration range of 0.19 µg/mL (0.006 toz/t.) to 10.7 µg/mL (0.312 toz/t). A high precision method was used, which involved bracketing the sample with the higher and lower standards nearest to its concentration repeating readings three times and averaging the results. The optimum precision for the FAAS gold method was found to be at the 1.00 µg/mL (0.029 toz/t) and lower scope limit 0.3 µg/mL (0.009 toz/t) gold concentration. Since tailings solutions typically have concentrations below the lower scope limit, use of ICPMS was developed at Newmont USA Limited (Newmont) to assay the lower range solutions."

Jan 1, 2016

By P. Hr?enjak

Two groups of methods are usually used in design procedures of underground excavation of dimension stone: the analytical and numerical methods of analysis. In all of these methods the in situ rock mass strength represents the main problem. This is due to massive rocks in dimension stone deposits, consisting of large blocks of rock separated with particular discontinuities, for which the Hoek-Brown empirical criterion, developed for jointed rock masses, cannot be realistically used to estimate rock mass strength, especially for a particular block of rock mass or a particular pillar. On the basis of the regular underground excavation process of dimension stone by separating blocks with the flat jack application, we developed the in situ flexural strength testing method. We carried out numerical analyses and evaluation of uncertainty of measurement in order to validate the in situ flexural strength testing method. On the basis of very good validation results, as well as results of the in situ flexural strength testing of exploitable layers on the ?Kanfanar? quarry, we introduced a new coefficient of reduction, which can be used for correction of other strength parameters determined by laboratory testing for obtaining a realistic value of rock mass strength.

Aug 1, 2013

By D. Tait, M. A. Halim, V. I. Lakshmanan, R. Sridhar

Rare earth elements (REEs), nickel and cobalt find their use in developing strategic materials and their demand is consistently increasing in the world while the extractable ores of these metals are gradually depleting. Process Research ORTECH Inc. (PRO) has developed innovative mixed chloride leaching processes for the recovery of REEs from alumino-silicate ores, and nickel and cobalt from laterite ores. The addition of magnesium chloride in hydrochloric acid enhances the activity of the hydrogen ion by orders of magnitude, which permit rapid leaching rates at moderate temperature and atmospheric pressure, along with high metal recovery. Innovative solvent extraction process steps are used for the separation of these metals from pregnant leach solutions. PRO’s processes also produce value added by-product of high purity iron oxide and recycle chloride lixiviant to the leaching stage. This paper describes the leaching and separation reaction mechanisms and potential process flowsheets.

Mar 1, 2017

By M. Z. Naghadehi, F. F. Nasategay, J. Sattarvand, R. Battulwar, E. E. Gohari

After the recent increase in catastrophic failures of tailings dams such as the Burmadinho, Brazil failure in 2019, the mining industry has actively engaged in the stabilisation of tailings dams. Most of the tailings storage facility (TSF) guidelines suggest a laborious action that can be facilitated through imaging by unmanned aerial vehicles (UAVs). Although UAVs are not unknown to most mining operations, automated rill detection through aerial image processing is a new concept. This study utilises images from sites in the U.S. and employs semantic segmentation with a UNet architecture to detect location and size of these rills. The developed model can be used as a tool for monitoring surface erosion of TSFs.

Sep 1, 2021

By Guillermo J. Román

The precipitation of copper and zinc as cements from a copper alloys' spent pickling solution has been studied at laboratory and pilot scale, with the objective of designing an economic process to recover both metals and render a solution to be either recycled to the pickling process or treated in a standard fashion and produce a non-hazardous sludge. The sulfuric acid spent pickling solution already containing copper and zinc was used first to dissolve another solid residue originated in the copper alloys foundry to neutralize part of the acidity. The resulting enriched solution was treated separately with two reductants,; sodium borohydride and iron powder varying pH and excess of reductant under constant agitation. Under the best conditions, precipitation of over 95 percent of zinc and copper was achieved together with the reduction of lead and cadmium contents respectively. A process for the combined residues treatment is proposed.

Jan 1, 1995

By W. A. Young

Vermiculite deposits on the. Platt ranch, 12 miles southeast of Encampment Carbon County, Wyo., were investigated by the Bureau of Mines early in 1949. Major activity was centered on property held under lease by the Mikolite Mining & Development Co. (figs. 1 and 2). Vermiculite occurs in masses and lenses of varying size and as disseminations in altered hornblende diorite. Alteration of the host rock probably resulted from the proximity of dikes and irregular masses of pegmatite. Investigation of the Mikolite property consisted of surface trenching, diamond drilling, and sampling. Four trenches totaling 5,401 linear feet were excavated with a bulldozer and Diesel power shovel. Three diamond-drill holes totaling 600 feet were drilled from the surface to determine the character of mineralization at depth. Eighty-one bulk samples were obtained from the trenches, and 130 sludge samples were collected from the drill holes. Samples were analyzed, and ore-dressing tests were made in the Bureau of Mines laboratories at Salt Lake City.

Jan 1, 1952

By John R. Howell

"SUMMARYThe process of producing a new cast component requires significant design and development by both the design engineer and the foundry engineer. This effort strives to produce a high quality component at minimum cost. To assist in this development, an analysis system, consisting of several computer programs, has been developed to simulate the metal casting process. This system incorporates analytical geometric solid modeling, finite element analysis, results display, and an innovative technique to model the casting mold with surface boundary conditions.The simulation has produced excellent thermal results and work is in progress to develop techniques to enable use of this data in additional analyses to predict porosity and residual stress. Research is required to produce working parametric formulations of material properties. Software development is required to develop good fluid flow analyses for large. 3D models to simulate filling of and convection in molds.Integrating this system of programs is difficult due to proprietary database structures used by software developers. However, an acceptable means has been accomplished' to move data from one program to another. Further software development is required to facilitate this task. INTRODUCTIONThe process of producing a new cast component requires significant design and development by both the design and foundry engineer. This effort strives to produce a high quality, reliable component at minimum cost in a timely manner. The design engineer needs to know the physical and mechanical properites which can be expected in the actual part at critical areas. This will enable him to design a part with minimum material while fulfilling the performance required. The foundry engineer needs to produce a sound, quality casting with a minimal amount of material In risers, the feeding system and machining stock. This must be done with a minimum· investment in tooling and development time and maximum productivity of labor and capital employed. This task Is continually being made more difficult because of increased competition in the marketplace and the use of castings in more severe service. This has strongly influenced the need for high quality castings.The task of developing a new casting is difficult for both the design and foundry engineer. To assist in this development, Deere & Company has been engaged in the development of an analysis system to simulate material processes. A major portion of this program involves simulation of the metal casting process. This analysis system has evolved into a package of computer programs which simulates casting solidification and subsequent cooling through the finite element method"

Jan 1, 1986

By Frank E. Cash

Data on the accident experience of the pebble-phosphate industry in Florida for 1930 and 1940 are combined in this report with statistics for the 9-year period 1930-38, inclusive, which were published in Information Circulars 69173/, 69684/, and 71005/. The periodic compilation and publication of such data enable the pebble-phosphate industry, as well as other industries, to consider the relative hazards, compare experience, and thereby help to prevent accidents. ACKOWLEDGMENTS Accident data were furnished by the following operating companies: Amalgamated Phosphate Co., Brewster; American Agricultural Chemical Co., Pierce; Coronet Phosphate Co., Plant City; International Agricultural Corporation, Mulberry; Phosphate Mining Co., Nichols; southern Phosphate Corporation, Bartow; and Swift & Co., Fertilizer Works, Bartow.

Jan 1, 1941

By Glasson K. R

The history of Ore Reserve classification and reporting in Australia is largely a history of the Australian mining industry itself. In the late 1800s, when the industry was dominated by small scale, high grade underground mines, Reserves (if estimated) were based on hand sampling and geological inference, and reporting of Reserves was either non-existent or of extremely variable quality. By the mid-20th century, the industry was dominated by large underground base metal and gold mines. Reserve estimation was coming under the control of skilled geologists and mining engineers, however Reserve classification and reporting standards continued to vary in type and quality from mine to mine. Today, large scale, low grade mining operations predominate, and Reserve estimates are generally based on extensive drilling results processed (with geological control) by computer. Through the efforts of a Joint Committee established by The Australasian Institute of Mining and Metallurgy and Australian Mining Industry Council, Reserve and Resource classification and public reporting has been standardised and controlled by adoption of the ""Australasian Code for the Reporting of Identified Mineral Resources and Ore Reserves"" released in 1989. The Code, which will continue to be modified and improved by the Joint Committee, has also been used as a basis for guides to reporting in other countries, and there is now some hope that this difficult and vexing topic may finally become subject to international standards and controls.

Jan 1, 1992

By Vasiliki Founta, Andrew J. Whittle

"This paper describes the development of a 3D finite element model for predicting ground movements caused by construction of 7.1 m diameter twin bored tunnels for the Crossrail project in London using Earth Pressure Balance (EPB) machines. The model uses the commercial FE code, Plaxis 3D, to represent the face pressure, conical shield, grouting process and activation of precast segmental concrete lining systems through a set of boundary conditions that advance through the soil mass along a prescribed trajectory. Soil behavior is described using the Mohr-Coulomb soil model with input parameters calibrated from laboratory element tests. The analyses focus on the performance of the EPB tunneling machines in greenfield conditions beneath Hyde Park, where the tunneling occurs within deep layers of relatively uniform London Clay. The model predictions are in good agreement with surface and sub-surface vertical settlements measured at a well-instrumented section, but tend to overestimate the measured lateral displacements towards the tunnel. Ongoing studies are focused on the role of soil behavior and the approximations in FE model predictions of tunnel-induced ground deformations. Introduction Since their introduction in the 1970’s, closed-face Earth Pressure Balance (EPB) machines have become widely used for soft ground tunnel construction. The EPB design relies on a screw conveyor system to regulate the pressure of the excavated soil/spoil within the head chamber and hence control stability at the face of the machine. One of the key design issues in the application of EPB machines is to establish how control parameters, including the advance rate, face pressure and grouting processes (used to compensate for the tail void at the rear of the shield and to infill gaps between the precast lining rings and surrounding soil) affect and/or control the far-field ground movements. For closed-face systems, short-term deformations are associated with ground losses at the face, overcutting and ploughing of the shield and poor control of grouting in the tail void; while long-term deformations can occur due to consolidation and creep within the surrounding soil (and are closely connected to the seepage boundary conditions;e.g., Laver & Soga, 2012). The control parameters for advancing the EPB machine can have a significant effect on the magnitude and distribution of the far-field deformations. Therefore, analyses of the ground response to tunneling require accurate modeling of the soil stratigraphy and constitutive behavior as well as the construction process. Finite element (FE) methods have been used to simulate tunnel construction since the early 1980s, but only a few studies involved three-dimensional modelling of mechanized tunneling (e.g., Finno and Clough 1985; Bernat and Cambou 1999; Swoboda and Abu-Krisha 1999; Dias et.al. 2000; Melis et.al. 2002; Kasper and Meschke 2004; Möller 2006). In order to get a better understanding of the effect of the construction process and soil properties on the development of the ground deformations more comprehensive models of tunnel excavation and tunnel support are required in order to capture the 3D nature of mechanized tunneling and the effect of construction parameters such as advance rate, face and grout pressure."

Jan 1, 2016

By A. Todd Davis

"Introduction“CAN YQUR Company afford not to train?"" This question was asked by Jack Taylor in the Factory Management Magazine for June, 1950. Mr. Taylor went on to say, ""No, it can't; no Company can"". It's true that many training programmes have not done the job they were designed to do because of a number of factors such as parts of management resenting training, management failing to understand the abstract values of training, and management giving impossible assignments or picking the wrong trainers. However, Mr. Taylor summarizes his article by saying that a properly-designed and •conducted training programme brings a company or a corporation to the point where the question is: ""Can my Company afford not to train?""About two years ago, in July, 1952, Nevada Mines Division of the Kennecott Copper Corporation embarked on a training programme, the same type of programme that had been inaugurated earlier at the •Utah Copper Division at Bingham, Magna, and Garfield, Utah. President Charles Cox of Kennecott, in •a speech made •before the Utah Bankers Association in June of 1951, stated that, ""I think that our programme of employee training is one of the most important phases of our entire operation"". That statement might be said to be one of the guiding principles of training at Nevada Mines Division. Training is supported by top managements and down the line to Division managements. As the years go by, the Company expects to receive many benefits from training."

Jan 1, 1954

By Richard R. Klimpel

INTRODUCTION Previous papers by this author have discussed some of the difficulties involved with optimizing the performance of industrial flotation circuits, e.g. Klimpel (1988, 1989). The two major problems seem to be: 1) the inherent "systems" nature of the froth flotation process which leads to a great deal of self-compensation among the various operating options available and 2) the inherent variations in mineralogy, water chemistry, etc. that always seem to be present in industrial operations. In particular, the influence of reagent dosage has been emphasized in the above referenced work including an industrial system interaction denoted as the R/K trade-off, Klimpel (1984, 1987). Using a simple two parameter model for fitting time - recovery data, Klimpel (1980): [ ] where r is cumulative recovery at time t or at the t th cell, R is the long-term equilibrium recovery and K is the first-order rate of mass removal from the cell (time ), the definition of the R/K trade-off with increasing reagent dosage can be given as follows: at low dosages both R and K increase, at intermediate dosages R continues to increase but K starts to level out, and at high dosages, R may flatten out or still continue to increase but K starts to decrease, sometimes rather quickly. An illustration of this effect in a batch laboratory flotation cell is given in Figure 1. The purpose of this paper is to give a detailed plant illustration of this R/K trade-off phenomena with changing collector dosage which shows just how highly interactive the grindinglflotation system in sulfide mineral processing really is. This is one of about 20 plant level case studies that are being presented and evaluated in an upcoming book, Klimpel (1990). The particular study given here was completed in late 1979 on a copper ore of head grade 0.93% Cu. The frother being used was Dowfroth* 250 at a constant dosage of 0.15 g for each ton of solids feed to the rougher bank and the slurry pH was 10.2. The motivation of the study was to better understand the relationship of changes in feed rate (hence in particle size and state of liberation) with variations in collector dosage (potassium amyl xanthate). The major concern was to determine how much extra economic benefit could be gained by better utilizing the existing grinding mills and rougher bank consisting of eight identical cells. The remainder of the flotation circuit was also studied, including regrinding of rougher tails, but did not prove as critical as the rougher bank operation itself. It appeared from an overall economic viewpoint, the battle was being won or lost in the manner in which the rougher bank was being operated. This was not to minimize the roles of the scavenging and cleaning circuits but those portions of this particular circuit could be rather obviously operated to achieve maximum benefit given the specific output of the rougher bank. A special attempt was made to maintain constant ore type throughout the entire test program. Eight specific test conditions were evaluated including three different feed rates (and hence three different size distributions) to flotation. Tables 1 and 2 summarize the eight test conditions and overall recovery results at the end of the rougher bank for each test. A single test consisted of at least three consecutive eight-hour shifts operating at steady-state with composite cell lip samples taken over the final eight-hour shift. "Trademark of The Dow Chemical Company

Jan 1, 1990

By M. Hardy

"The structural design of mine headframes is challenging due to the significant differences in the geometry, rigidity, and behaviour of these structures compared to conventional buildings. The differences are related to the main purpose of these structures, which is not human occupancy, but rather the support of hoisting equipment. The main objective of this paper is to present the seismic elastic behaviour of headframe structures in comparison with conventional buildings, which is the base case for the definition of seismic loads. This paper highlights the impact of backlegs on the dynamic behaviour of mine headframes, more specifically the resulting torsional sensitivity. INTRODUCTION The design of industrial structures is often governed by industrial processes rather than by architectural constraints. In ground-mounted hoisting plants, the structural configuration of headframes must allow clearances around conveyances and ropes while resisting the lateral loads associated with the hoisting equipment. Typically, these structures use backlegs to withstand the lateral loads, which leads to highly irregular structures with unconventional dynamic behaviour. The codes and standards applicable to the structural design of mine headframes in Canada are often incomplete or not adapted to the reality of practising engineers. This situation creates confusion, and engineers must look at various guidelines and international standards to support the engineering calculations required for the design of these structures. The loads resulting from the support of hoisting equipment, such as ropes and conveyances, are usually simple to calculate and are determined based on the information from manufacturer and industry related guidelines. Moreover, the loads associated with snow accumulation and wind are calculated applying the same methodology used for conventional buildings, without any major issues. On the other hand, the seismic loads are often difficult to calculate based on the methodology from the National Building Code of Canada due to the significant differences in the dynamic behaviour of the structures. The resulting structures may be either over-designed or worse, under-designed. Hence, there is a lack of uniformity in engineers’ designs. The latest edition of CSA S16-14 “Design of steel structures” includes Annex M (Informative) “Seismic design of industrial steel structures” to help designers close the gap between the requirements of traditional regulations and actual engineering practice."

Jan 1, 2019

By Donald E. Hendrickson

A new prinicple of side dumping front end loading for underground mucking machines has completely changed the concept of underground loading and haulage equipment arrangement both in production and development work loading. Versatility and adaptability of this type of machine permits combinations with a broad range of haulage means not possible with -conventional overhead loaders. Steel "armored" conveyors serving this type of loader are replacing conventional rail mounted car haulage to give a continuous loading operation and effect important economics of reduced manpower, faster loading rates and fewer working areas for a required production. First used in Germany this easy to control and highly efficient means of placing rock from the muck pile on or into haulage means has advanced the art of underground rock loading more in the past three years than any other advancement since the first successful small crawler mounted air powered loader was introduced in 1954. Bucket type loaders with a relative new side dump principle are gaining in popularity for underground loading. They are replacing conventional overhead loaders in present applications and becoming first choice in new ones. Primarily for hard rock development and production loading they are also widely used in production loading of coal and other soft materials. This discussion briefly traces the historical development of hard rock underground loaders to the new side dump type and its present population. In more detail various applications, loading techniques and the advantages of operation, production and economy with the side dumper in hard rock loading will be dealt with.

Jan 1, 1966

By A. Sinha, N. K. Babar, H. R. Kalihari, A. K. Chakraborty, M. R. Saharan, C. P. N. Pathak

A case study is presented on the optimization of cable bolting patterns in cut-and-fill stopes at the Chikla Mine, MOIL, India. Rock mass characterization, using Bieniawski’s RMR and Barton’s Q, indicated that the stopes lie in the “no supports required” to “spot bolting required” support categories. In contrast, the current practice of cable bolting in a 2.0 m by 2.0 m grid pattern for stope widths less than 12 m displays different results. Numerical modelling results are in line with the rock mass characterization study. It can be affirmed, as well, that no cable is carrying rock loads greater than 5 t. This fact is derived from observational behaviour of the crown pillar of an experimental stope with open stoping operations and hanging cables in the cut-and-fill stopes of the mine. An increase in the cable bolting grid spacing is, hence, recommended for implementation by mine management, based on the study described in subsequent sections. It is also suggested that the mine adopt the open stoping method, provided results of a rock mechanics instrumentation program in an experimental stope with a cable bolting grid of 2.5 m square are favourable.

Jan 1, 2007

By W. B. Plank

I HAVE been asked by the Chairman of the Engineering Education Committee to outline what the engineering colleges would like the mining companies to do with the young engineer just, out of college. It has been suggested that, because my own personal experience has been almost entirely within the coal industry, I handle this subject from that view- point. Many thoughtful engineering educators are today wondering what effect the present employment policies of many industrial companies are going to have on the young man just starting out in his profession. Since before the World War large industrial concerns, such as those in the electrical and mechanical industries, have gone to the colleges to interview 'likely graduates until today there has developed in most of our engineering colleges, especially in the eastern part of the country, what appears to be a lively competition on the part of these companies in bidding for the young graduate engineer. The employers of mining engineers, with one or two exceptions, have not yet begun to enter this competition. It remains to be seen how serious this situation will become, how it will affect the training of other engineers, and what effect, if any, it will have on the mind of the young man about to enter college who may be inclined to take up mining engineering, but who gains the impression from this well advertised employment service just mentioned that there is no opportunity for him in the mineral industry as com- pared with the opportunities offered in say civil, mechanical, and electrical engineering.

Jan 1, 1929

By N. Coulson, P. F. Stewart

The Mine Health and Safety Act No. 29 of 1996, as amended, embeds an occupational health and safety management system that facilitates communication between representatives of the state, capital, and labour. In underground mines, two communication loops between these role players serve to separate the production chain of command from the political tripartite relations. Worker-elected health and safety representatives are involved in both communication loops, but are severely constrained in escalating their occupational health and safety concerns to the legalpolitical level. This has a direct bearing on the exercise of the Right to Refuse Dangerous Work (RRDW). As previously reported, health and safety representatives were found to primarily consult their production supervisors in preference to their trade-union representatives. This follow-up article presents production supervisor perspectives on the RRDW. The study demonstrates how, for mineworkers, the two distinct communication subsystems constrain, rather than facilitate, implementation of the RRDW.

Apr 14, 2024

By D. G. Hartley, P. C. Hayward, K. R. Weller, U. J. Sterns

The major copper loss at Mount Lyell is in the coarsest particles presented to flotation. An analysis of data collected from the five Primary and six secondary ball mill circuits showed that grinding was limited by low breakage rates in secondary mills, possibly due to oversize balk. For five months, the make-up ball size for one secondary mill was reduced from 50 to 30 mm (2 to 1.1 in.). Frequent steady-state tests indicated that overall average breakage rates could be increased by about 70% using smaller balk. A mathematical simulation of the whole circuit with increased breakage rates showed that the resultant finer grind was likely to improve copper recoveries by about 2%.

Jan 1, 1984

By Stephen J. Navin

The Los Angeles Metro Red Line has installed high density polyethylene (HDPE) as a gas and water barrier around all subsurface structures. The paper reviews the geological setting and resulting need for the barrier. HDPE installation practices are outlined, as are the risks to the barrier, from normal placement methods. Testing and inspection methods used to assure a quality installation are described, and also the field tag method for recording deficiencies and their repair. Specifications and installation changes that will be incorporated into subsequent contracts to improve the quality of the installation are high lighted. The cause of the fire that destroyed part of a tunnel is reviewed as are the revised precautions put in place to prevent a recurrence.

Jan 1, 1991