Search Documents

By G. F. Zhang

Yunnan mineral is a kind of containing Ta-Nb rare earth deposit, the most potential economic valuable element in it is Sc, and raw mineral has 17.0 g/t Sc. Take the character of Sc into account, the key problem in producing scandium chloride molten salt by adaptive technology is how to man-ufacturer anhydrous scandium chloride molten salt. During the course above, the key problem is that preventing Scandium chloride from being hy-drolyzed into ScOCl. The ScOCl content of molten salt not only affect purifying scandium but al-so do harm to the production of Mg-Al alloy. In order to restrain the creation of ScOCl, add ammo-nium chloride into molten salt. As long as ScOCl is created, it would be chloridized. Thus, during the course of dehydrating anhydrous chloride molten salt this method prevents scandium from hydrolyzing. Accordingly, the key phase of this producing course is eliminate ammonium chloride and the surplus crystal water, and the most crucial of it is restrain scandium chloride from being hydrolyzed into ScOCl. This research analyzed the theory that ammonium chloride is trend to be hydrolyzed into ammo-nium chloride, and propose the research scheme of restraining this process from creating ScOCl. The new process of restraining the generating of ScOCl and dehydration for crystal water rare earth chloride is adopted: Inert gas temperature increases rapidly to a certain degree to eliminate ammonium chloride and crystal water. The experimental research were done such as : the temperature of eliminating ammonium chloride and crystal water, amount of ammonium chloride, time, temperature growth speed, influence of inert gas flow toward to scandium percent hydrolysis, whereby, we conclude the optimal process condition in this period: Temperature: 400°C, proportion of oxidized scandium and ammonium chloride is 1:1.5(Proportion of oxidized scandium and additional salt is 1:5.4), time is 120minute, temperature growth speed is 10?/min inert gas flow is 5L/min.

Jan 1, 2014

By K Nguyen

The advances in computerised processors and digital microscopic imaging have enabled more studies that attempt to quantify textural information from rock images. However, a texture classification system designed for geometallurgy has always been challenging, as it combines general image analysis techniques and specific texture classifications in terms of the size, shape and orientation of mineral particles. In this paper, a novel texture analysis technique using grey level radial shell association (GLRSA) and spectral angle mapper (SAM) to discriminate between different geological images is outlined. The technique can be applied in routine analysis and characterisation of mineralogical and textural attributes which may correlate with variability in mineral processing behaviours. Using this texture analysis technique, a set of small texture reference images, a texturelet reference library, has been identified from a suite of drill core images. Rock types are classified according to the proportion of each texturelet type recognised. The library is site specific.CITATION:Nguyen, K, 2013. A new texture analysis technique for geometallurgy, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 187-190 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 29, 2013

By Fred C., Fred C. Bond

Comminution energy is principally energy of deformation before breakage, which appears as heat. An empirical equation is presented which covers the entire comminution range. The new strain-energy theory considers comminution from the known principles of mechanics and the reduction ratio. Energy requirements according to the different theories are compared.

Jan 1, 1950

By H. M. Chance

STRETCHING from New York southwestwardly to Georgia is a great range of hills and mountains consisting of pre-Palaeozic schists, slates, and gneissic and granitoid rocks, known locally by many different names; in Pennsylvaniaa as the South mountains, the Berkshire or Durham and Reading hills, and in Virginia and farther south as the Blue ridge. To the northwest of this range lies the " Great Valley," underlain by Lower Silurian limestones, and known as the Lehigh valley, Cumberland valley, Shenandoah valley, valley of Eastern Tennessee and Georgia and by other local names, but all these are parts of one valley, geologically and geographically continuous from New Jersey to Georgia. Lying upon and along the northern flank of this range of mountains, and dipping northwestwardly under the limestone of the valley, is the Potsdam formation (Primal or No. 1 of the Pennsylvania Geological Survey), which consists of sandstone or conglomerate overlain by slates. The latter are sometimes arenaceous and occasionally are replaced in part by sandstone. This formation varies from a few feet to several hundred feet in thickness. At the outcrop the slates are completely decomposed, oxidized or disintegrated, forming a belt of clay several hundred feet wide occupying the interval between the outcroppings of the sandstone and limestone. This clay-belt includes a more or less continuous series of iron-ore (brown hematite) deposits, which have been, and still are, extensively worked. The belt is more or less continuously mineralized with iron throughout its length, and hundreds of large deposits of ore have been developed and worked, producing many millions of tons of ore.

Sep 1, 1908

By N. T. Cowper

Slurry viscosity ultimately affects the majority of mineral extraction processes. However, to date its role has been disguised because mineral processing equipment has been conservatively designed to operate at minimal slurry viscosity. With most ores this still meant acceptable solids concentration was achieved. However, with the advent of treatment of weathered ores the resulting higher slurry viscosity (rheology) has meant solids concentration has had to be reduced, often to unacceptable levels. To permit operation at higher solids concentration chemical dispersants are employed. These dispersants are expensive and their addition should be controlled. Slurry Systems Pty Limited (SSPL) has developed an On-Line Viscometer to monitor and control slurry viscosity, whether by controlling water addition or by controlling dispersant addition. Using this instrument the slurry rheology can be controlled at the optimum level, maximizing solids throughput, and minimizing water, reagent, and dispersant addition. This paper discusses rheology of typical mineral slurries, the relationship between slurry rheology and percent solids in a slurry, and the need for rheological control in mineral processing. The effects of rheology on various process steps, and the use of dispersants are reviewed. The paper concludes with summarizing actual case histories and cost impact of Theological control.

Jan 1, 1992

By Praterm J. D.

The process parameters for effective utilization of nitric acid as an oxidant for copper-iron sulfides have been developed. Leaching variables found to be important were acid concentration, temperature, time, atmosphere, pressure, mineral type, particle size, and percent solids. Particular attention was devoted to analysis and control of fumes evolved during the oxidation. Responses recorded were metal extraction, elemental sulfur formation and oxidation, product gas composition and evolution rate, reagent consumption, and residue flotation response. The paper concludes by laboratory testing a flow sheet based on the above observations. Ninety-eight percent of the copper contained in a chalcopyrite-bornite concentrate was recovered, requiring 1. 6 pounds of nitric acid to be regenerated externally per pound of copper extracted. Iron remained in the leach residue as hydrogen jarosite and was ultimately rejected from the system by froth flotation. Sixty percent of the sulfur contained, in the heading concentrate reported in the elemental form; the remainder was oxidized to sulfate and eliminated with the hydrogen jarosite in the flotation tails or by neutralization of the solvent extraction raffinate with lime. Nitrogen was evolved primarily as NO and NO2; up to 3 percent N2O was also present. Gases containing sulfur were absent except when pyrrhotite was solubilized in the absence of soluble copper. In this case, H2S was evolved.

Jan 1, 1972

By Paola Campana, Maurizio Lenzi

"In the paper a non linear model for the analysis of the pile response to the axial loads is presented. The study aims to find a general analytical solution concerning the pile-soil interaction modeled by means of the method of the load transfer curves. It makes use of a criterion based on the plastic and on the hardening work locally done by the shear stresses and due to the current elastic shortening of the shaft. Then a specific solution, denoted fractal model, is found applying recurrent relations that link the shaft shear stresses and the tip normal stresses to the pile settlements. These relations are based on that one proposed by Kondner in modeling the soil samples response in the triaxial tests and then used by Chin in the inverse slope method in estimating the pile capacity by the load test results. The comparison between the model predictions and the load tests measures validate the fractal model.INTRODUCTIONThe behavior of the axially loaded piles is known to be characterized by a highly non linear trend caused by the onset of plastic and irreversible deformations in the soil starting just from the first load levels. With the purpose to reach a deeper understanding of the pile-soil interaction, the influence of the soil inelastic properties on the distribution of the axial loads and of the vertical displacements is investigated. The research starts analyzing the behavior of an infinitesimal length of a pile embedded in a non linear homogeneous soil. That initial study leads to a criterion based on the plastic and on the hardening work locally done by the shear stresses and that allows to find the axial loads as a function of the elastic shortenings of the shaft. The relation that links the pile axial load to the elastic strain of the pile material is then used. That allows to find the vertical displacements and the axial loads as function of the depth in a parametric form in which the tip settlement acts as an independent parameter. The method of the load transfer curves is used and the effects of the pile technology are included in the soil constitutive relations. The analysis is then extended to the non-homogenous soils in a very effective way. The solution is applied taking as constitutive relations of the pile-soil interface the one proposed by Kondner to analyze the soil triaxial tests and by Chin to interpolate the pile response measured in the load tests."

Jan 1, 2017

By B Jones, M Chappell

Installing ground support correctly is not an easy task, whether lack of operator training, poor ground conditions or underperforming equipment. Confirming the quality of installation can help with planning and remedial works and help to maintain a safe working environment. A 2?m bolt with 700?mm of encapsulation will pass a 10?t pull test (assuming ground conditions permit); however, it will not provide any data of the unencapsulated length and thus the ability to withstand corrosion. Complimentary testing in the form of non-destructive testing (NDT) can provide factual data that leads to the assessment of the quality of the encapsulation. By improving the quality of the encapsulation, the risk of corrosion is reduced, the ground support durability is improved and hence the asset life is enhanced.This paper presents a case study of a trial using the NDT system RokTel™. Installations of varying quality were installed and tested. The characteristics tested were corrosion, load and free length. These characteristics can be tested and assessed for cable bolts, thread bar bolts and split sets. The aim of this paper is to present NDT testing as a viable method for assessing installation quality and ongoing performance of the ground support.CITATION:Jones, B and Chappell, M, 2014. A non-destructive testing method for measuring bolt condition, in Proceedings AusRock 2014: Third Australasian Ground Control in Mining Conference , pp 307–314 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Nov 5, 2014

By A Melkumyan, F Ramos

This paper addresses the problem of prediction of grade variables over volumes (block estimation) for an autonomous mine. Resource estimation is usually concerned with the prediction of the quality of material over volumes (blocks in resource models). This problem is tackled by developing Bayesian block estimation with the Gaussian processes (GPs) framework. GPs are commonly used for nonparametric regression and classification and offer an elegant solution to deal with incomplete knowledge and information. A GP is defined by its mean and covariance function. The parameters of the covariance function can be learnt by maximising the marginal likelihood of the data. This offers a way of automating the covariance fitting process, while at the same time allowing for an assessment of how well the covariance function fits the data. Within the GP framework the calculation of the average grade over a volume can be performed using Bayesian quadrature, which treats an integral as a random variable. First, it is demonstrated that the block kriging equations in geostatistics are the same as Bayesian quadrature. Then, it is determined when closed form analytical solutions for the integrals are possible and quadrature approaches for approximations are briefly discussed.

Sep 26, 2011

By S. Broda

Waste rock piles (WRP) are made of heterogeneous, coarse grained waste rock, with inclined layers due to segregation along the slopes and haul traffic compacted layers, creating highly contrasted hydrogeological conditions. The internal structure of the pile has a major impact on water and oxygen movement and pore water pressure distribution. In the presence of reactive minerals, this influences the production of acid mine drainage (AMD) or contaminated neutral drainage (CND) with metal leaching from the WRP. Water flow and distribution may also affect slope stability, particularly following major infiltration events (rain/snow melts).The prediction and interpretation of water distribution in WRP is a challenging problem. Numerous numerical investigations of short and long term forecasts have been conducted, and many of these were found of limited reliability. The shortcomings of existing approaches are typically due to difficulties in representing preferential flow processes in macropores. Such phenomena are in fact often neglected in the assessment of water flow in WRP because these are unattainable with the classical porous media approach typically used in this field. This paper focuses on the use of modeling techniques originally designed for massive fractured rock media to simulate flow in WRP. Waste rock properties used in this study are retrieved from large scale laboratory permeability and water retention tests and from in situ infiltration tests. The effect of WRP configurations on water flow into potentially reactive zones is examined with the numerical 3D fully-integrated surface/subsurface, variably saturated flow model HydroGeoSphere. Material properties retrieved from the laboratory tests are represented by means of stochastic facies distribution generated within a probability framework. Modelled soil water contents and hydraulic head distributions, as well as fluxes within a simple waste rock pile, with compacted layers, are compared with results retrieved from simulations based on the classical equivalent porous medium approach. Results indicate significantly increased flow towards the looser waste rock zone by considering heterogeneous material distribution, causing weak or absent capillary barriers and preferential flow paths. Practical implications for the design of waste rock piles are also discussed.

Aug 1, 2013

By J. Scheepers, P. Smith, G. (Yunfei) Qiao

High pressure acid leach (HPAL) and pressure oxidation (POx) or pressure leaching extraction processes occur in autoclave vessels operating at elevated temperature and pressure, acidic conditions with abrasive mineral slurries. The harsh process conditions require specialized severe service metal seated ball valves to support operation of the vessels. Over several decades thermally sprayed titanium dioxide (TiO2) and chrome oxide (Cr2O3) coatings have been applied to extend the service life of the ball valves and other high wear critical components. Variability in the coating application quality control and severe process conditions can result in spalling, coating failure and premature failure of the component resulting is decreased autoclave circuit availability. In 2012 MOGAS, together with BRENCO, initiated research and development for ML-381, metallurgically bonded coating. This paper includes the historical development of severe service ball valve coatings, success of the trials and increasing industry adoption of the ML-381 coating for ball valves and other critical autoclave circuit components.

Mar 2, 2022

By L. Boyd, T. Martin, V. Gall

Water infiltration into subgrade infrastructure can cause major impacts on their performance. In addition to damaging the structure, water intrusion leads to deterioration of installations including electrical and mechanical components and in patron discomfort. To remedy these impacts, leak remediation is often carried out to halt water infiltration. Remediation methods include coatings, drainage, injection/stitch grouting, curtain (backside) grouting, and/or internal umbrella systems. Selection of the rehabilitation method depends on the structure’s use, owners’ priorities, its installations, structural conditions, surrounding ground, and hydrogeologic conditions. Since many factors influence the rehabilitation method chosen, a novel holistic approach is undertaken to understand leakage causes and consequences to develop the most appropriate, efficient, and reliable solution to extend the structure’s life. A reconnaissance phase combines geologic, hydrogeologic, and as-built information with detailed digital scans and visual observations to develop a database of existing conditions. This database, called a tunnelband, is used to develop the rehabilitation solution and made part of the contract documents, allowing for an informed bid by specialty contractors. Tunnelband and the preferred rehabilitation system are portrayed in contract documents, which are procured in various contract types depending on the owner’s preference and project characteristics. The pool of contractors are required to submit their understanding of this holistic approach by developing and supplying the owner with a detailed workplan. The completed rehabilitation is portrayed in detailed as-built drawings which also provide the owner with an operation and maintenance manual outlining for periodic observation of the structure and checking of its performance. Ultimately, this information is implemented into a “BIM Digital Twin” that is used by the operations and maintenance staff for long-term observations. This proposed novel framework for leak rehabilitation is currently being used successfully in a number of projects throughout the United States.

Jun 13, 2023

By Stuart Hardy, Xiangbo Qiu

The South Bank Place project comprised the redevelopment of the iconic Shell Corporation headquarters in Waterloo, located on the Southbank of the River Thames in central London. The site was originally developed in the 1950’s and due to the monitoring of the underlying Bakerloo Line tunnels by the Building Research Station it has become a seminal case history for the long term swelling behaviour of London Clay. The redevelopment, undertaken by a joint venture of Canary Wharf Group and Qatari Diar, included demolition of the existing building (with the exception of the main towers) retention of the existing basement structure and the construction of eight new mixed use buildings, up to 31 storeys in height. The foundations for the new buildings varied according to the building height and the below ground constraints. Two of the buildings are partially founded in reused under reamed piles that were installed as part of the original 1950’s development. The paper describes the novel methodology developed to assess the re-use of the under reamed piles which builds on the guidance published by CIRIA and the BRE for straight shafted piles. The paper also describes the physical investigations and numerical analyses undertaken as part of the assurance process and the procurement of latent defect insurance by the client.

May 1, 2022

By S. Cagirici, Z. Öztürk, A. Hassanzadeh

Many flotation kinetics models have been studied in the literature. Their applicability was extensively investigated and argued in detail. However, model selection criteria were not adequately discussed from the statistical points of view. In this investigation, the kinetic behavior of a complex copper sulfide ore was studied in a mechanical Denver flotation cell focusing on flotation kinetics of chalcopyrite, pyrite and molybdenite. Different flotation kinetics models including nine common empirical models and four mathematical models namely Hill, Chapman (Sigmodial function), single rectangular (Hyperbola equation) and exponential were applied to the experimental data. In addition to assessment of the goodness of fit criterion for each model, a factor of model complexity was considered using information criteria (IC) (i.e. Bayesian information (BIC), low of iterated logarithm (LILC) and Akaike information (AIC) indices). The obtained results showed that the IC indices could simply manifest the best-fitted model to the experimental data. Whereas, the coefficient of determination values (R2) were relatively same for all models. By taking the R2 and model complexity criteria into account, the exponential model was chosen as the best representative mathematical model to demonstrate chalcopyrite kinetic behavior. However, Chapman model was selected as the best one for the flotation of pyrite and molybdenite. In case of the common first-order flotation kinetics models, fast and slow flotation kinetic model (Kelsall) was reasonably fitted the best to the given data of chalcopyrite. However, the gas/solid kinetics adsorption model was chosen as the best-fitted one for pyrite and molybdenite. Furthermore, it was found that mathematical models represent better results in association with flotation kinetic behavior of chalcopyrite, pyrite and molybdenite due to the consideration of more parameters in modeling. Finally, it was concluded that the IC indices must be applied to the process of model selection due to consideration of goodness of fit, complexity of a model and model consistency.

Jan 1, 2018

By Joshua R. Omer

A numerical model is developed for predicting the load transfer and settlement behaviour of vertically loaded, bored cast in-situ piles. Based on analysis of full-scale pile tests, closed-form equations are formulated to uncouple the shaft and base transfer resistance components of the pile/soil system. Appropriate parameters are introduced to take into account: (i) the shaft resistance distribution in unlined and partially lined piles, (ii) the effects of ?soft toe? (voided toe pile) behaviour on base resistance development. For convenience, an interactive computer program has been developed in Matlab? with a user friendly interface system. The program can be applied to predict the complete load-settlement curve and axial force distribution in bored cast in-situ piles if reliable soil test data are available. To demonstrate its validity, the proposed method is applied to 4 test piles, which are also analysed using two previously published methods: (a) elastic solution method and (b) layer discretisation method. Methods (a) and (b) above are based on load transfer (t-z) and finite element models respectively. It is shown that the proposed method yields axial force distributions and load-settlement curves that are in good agreement with the observed pile test data.

Jan 1, 2010

By Gang Mei, Nengxiong Xu, Yanlin Zhang

After open-pit mining finishes, an iron mine with steeply inclined ore-bodies will generally transform to underground mining. Non-pillar sublevel caving method, with horizontal slice mining from up to down, is commonly used for mining deep ore-bodies under the open-pit bottom. As horizontal slice mining proceeds, a gradually enlarging goaf will be formed under the open-pit bottom, the rocks surrounding the handing and heading sides of the goaf will collapse continuously, and then these results in rock-mass failure with large deformation and displacement in the open-pit slope. Based on FLAC3D, a modeling approach to predict pit slope deformation and failure induced by transitioning open-pit to underground mining is proposed in this paper. This method mainly consists of three procedures of modeling, i.e., mining modeling, landslide modeling, and filling modeling, of which landslide modeling is the core of the proposed approach. In the procedures of modeling with FLAC3D, iteration non-convergence is taken as a local failure criterion of pit slope. When iteration calculation cannot converge, the sliding modeling will start by deleting some elements with large displacements, and finish until convergence. Before modeling landslide, several displacement contours of rock-masses in the pit slope are searched out, and each contour is an approximate possible sliding surface of the open-pit slope. From these contours, the most proper one is selected as the slope sliding surface by using the iteration calculation convergence and minimum element deletion as the judgment condition. Landslide modeling is then realized by removing the elements outside the slope sliding surface. Finally, to demonstrate the effectiveness, the proposed approach is used to simulate the process of open-pit to underground mining at Yanqianshan Iron Mine, and to predict the deformation and failure characteristics of open-pit slope during underground mining. The case study results show that: the prediction of the slope stability determined by using our method is consistent with that calculated by using the limit equilibrium methods.

Jan 1, 2015

By M. Ghamgosar, Erarslan N.

Rock cutting includes many unknown parameters which result in different consequences of the different cutting conditions. However, based on the current suggested models for rock cutting, there are some critical parameters that control the fracture initiation and propagation in the rock domain. In general, there are two major cutting tools are used in rock cutting: drag bits and indenters. Different mechanical attacks of cutters result in different fracture models. For example drag bits break the rock parallel to the rock surface, which is called ‘undercutting’ while indenters cut rocks by pressing normal to the rock surface. Both tools cause rock failure by inducing tensile cracks in the rock. However, drag bits induce more tensile cracks compared with the indenters. The full benefits of improvement in rock cutting efficiency will arise from a clear understanding of rock fracture mechanics principles. Therefore, fracture mechanics principles will be applied to the rock cutting as the rock fragmentation process is based on the rock fracturing. The Oscillating Disc Cutting (ODC) has more advantages than drag bits and disc cutters. The forces transmitted back through the cutting head are much lower with ODC. Although ODC attacks rock in much the same way as drag bits, a combination of the oscillating action of the cutter and the undercutting mode produces the lower cutting forces and causes rock fatigue. Fatigue phenomenon in rock has advantages for the rock breaking by applying less effort and energy due to the cyclic loading. The aim of this study is using the Extended Finite Elements Analyses (XEFM) numerical modelling for simulating the ODC cutting with different cyclic loading conditions. An accurate prediction of numerical investigation requires detail laboratory experiments. A series of laboratory tests were performed to obtain the fracture toughness values of our Monsonite rock samples under various cyclic amplitude and frequency values. The numerical results showed different types fatigue damage mechanisms due to the various cyclic loading conditions. Thus, modelling and understanding the crack propagation behaviour depending on different amplitude and frequencies are believed very important for the dynamic rock cutting researches and rock cutting machine manufacturers.

Jan 1, 2015

By Juhyung Lee

Recently, large diameter drilled shafts are often used for foundations of large-scale marine bridges with high capacity and required to resist lateral loads. Steel casings are essential in supporting the hole-diving concrete installation of the drill shafts constructed in the sea. The casings, however, are usually not considered as the structural elements and their load bearing capacity is often overlooked in the design of drilled shafts. As the diameter of drilled shafts becomes larger, there are higher demands for more efficient and economical design of large-scaled drill shafts in the sea. Steel-concrete composite columns are popular for the superstructure of bridges, and the outer steel attached to the shaft increases the shaft resistance through confining the concrete. In this study, the lateral resistance of steel-concrete composite drilled shafts was evaluated quantitatively based on the numerical analysis. When steel casings are used as structural elements such as the composite columns. Ultimate lateral bearing capacity of composite drilled shafts with various diameters was numerically calculated through the 3D finite element analysis. For that, elasto-plastic model with the perfect plasticity is involved to capture the ultimate load. A commercial FEA program, MIDAS-GTS, is used in this study. Real field conditions of the west coast, Korea were examined to set up the ground conditions and pile lengths required for this parametric studies. Detailed characteristics of the stress and displacement distributions are evaluated for the better understanding on the mechanisms of the composite shaft behavior.

Jan 1, 2008

By Morgan M. Sears, Gamal Rashed, Michael M. Murphy, Brent Slaker

In this study, a parametric study was conducted using FLAC3D numerical models to examine the impact of oblique loading, generated from seam dip, on the strength and the failure propagation pattern of a stone pillar using two simplified geometry types. In type 1, the sidewalls of the pillars were assumed to be perpendicular to the roof and the floor, while in type 2, the sidewalls of pillars were assumed to be vertical. The complex pillar geometry in dipping mines was frequently modeled using these two geometries. To capture a complete picture of the effect of seam dip on pillar stability, the modeled width-to-height (W/H) ratio of the pillars, in situ stress field, and pillars roof/floor interfaces were systematically varied to account for the potential distribution of values for these parameters across the underground stone mines in USA. Results from the numerical modeling indicate that dipping pillars have reduced strength compared with horizontal pillars. Also, an asymmetric failure propagation pattern could be obtained depending on an interaction between the W/H ratio, seam dip, in situ stresses, and pillar geometry.

Feb 5, 2021

By Morgan M. Sears, Gamal Rashed, Michael M. Murphy, Brent Slaker

A parametric study was conducted using FLAC3D numerical models to examine the impact of oblique loading on the strength and yield pattern of a stone pillar using two simplified geometry types. In type-1 the side walls of the pillars were assumed to be perpendicular to the roof and the floor, while in type-2 the side walls were assumed to be vertical. These two geometries are frequently used to simplify/model the creation of pillars in dipping mines. Results from the numerical modeling indicate that dipping pillars have reduced strength compared to horizontal pillars, and they are at an elevated risk of failure. Also, an asymmetric failure propagation pattern could be obtained depending on the interaction between the pillar width/height (W/H) ratio, seam dip, in situ stresses and pillar geometry.