Search Documents

By C. Sawmliana

Deep-hole blasting operations in opencast mines are always associated with some annoyance to the surrounding areas in terms of ground vibration, noise, flyrock, etc. These become more severe when underground mines run parallel to them. Blasting in opencast mines may pose danger to the underground mines in terms of roof/side failures, damage of water dams, isolation/ventilation stoppings and other underground installations. The paper discusses a case study in India on the impact of deep-hole blasting in Ramagundam Opencast Project ? III on the stability of various water dams of GDK-6B Incline coal mine operating in close proximity. The stability of underground water dams of GDK-6B Incline mine was endangered due to deep-hole blasting at OCP-III. There were 15 water dams in seam 3 and 16 dams in seam 4 of the underground mine. The physical strength tests using a Schmidt Hammer instrument were carried out in all the 15 water dams of No. 3 seam and 16 water dams of No. 4 seam. The test revealed that there is a decrease in compressive strength values of three dams of seam 3, namely 6A, 6B and 7. Theose three dams were also directly connected to the water source. Water seepage was also observed from the surfaces of the three dams. However, they were found to be stable in accordance with their design parameters and damage point of view while considering a threshold value of ground vibration as 25 mm/s. It was found that the value of tensile stress generated by a vibration of 25 mm/s was much lower (i.e. 0.171N/mm2) than the tensile strength of the two weakest concrete dams (1.85 N/mm2). Ten experimental blasts were conducted at different working benches of the opencast mine and the ground vibration data were recorded in roof and pillars near various underground water dams. The maximum value of vibration data recorded was 5.9 mm/s in roof near dam no. 1 in seam 3 with the dominant peak frequency of 22 Hz. No adverse effect on either of the dams was observed after that blast. During the period of study, no deterioration/adverse impact was found in any dam of the mine. Based on the results of the study and the analyses of the data, optimized blast design parameters and explosives weight per delay were suggested for the opencast mine to maintain the safety of the underground water dams. Keywords: Opencast blasting, damage, water dams, underground mine, ground vibration

Jan 1, 2009

By R. J. Davidson, M. V. Nkosi, V. Veronsee

When activated carbon is used for the recovery of gold and silver from Witwatersrand gold-plant solutions, severe fouling of the carbon by calcium carbonate is experienced. Also, difficulty is encountered in the retention of the adsorbed silver on the carbon because it is displaced by gold. The acidification of plant pregnant solutions before they are contacted with carbon was therefore investigated as a means of overcoming these shortcomings. The results from batch laboratory experiments on the effect of pH on gold adsorption from plant solutions and from synthetic solutions of high ionic strength indicated reaction rate optima in the region of pH 5 and below pH I. The effect of pH on the equilibrium capacity constant for gold adsorption onto activated carbon was also studied, using both a plant solution and a synthetic solution of high ionic strength. In the case of the plant solution, the results suggested only a marginal change in carbon capacity for gold in the pH range 12 to 6, with significant increases in capacity at lower pH values. In the case of the synthetic solution of high ionic strength, the capacity constant was found to increase steadily as the pH was lowered from 12 to I. It was also found that the gold adsorption capacity of the carbon was significantly better for the synthetic solution. No precipitation of gold from the pregnant solution (gold content approximately 6 g/t) was noted as the pH was decreased to values as low as ,. However, about 25 per cent of the 0,6 g/t silver content was found to have precipitated when a pH value of 3,0 was attained. Further substantial precipitation of silver took place as the pH was lowered further. The investigation was extended by the passing of suitably acidified gold-plant solutions through one or more columns containing activated carbon. Preliminary results showed that the particle size of the carbon had a significant effect on the rate of gold adsorption. No displacement of adsorbed silver by gold was observed to take place at an influent pH of 5. In a typical continuous 'merry-go-round' operation involving three columns in which gold was eluted from the lead column every 48 hours, effluent values averaging 0,001 g/t gold could be consistently achieved under simulated plant-operating conditions. The results were obtained at relatively high velocities of influent solution, resulting in very little hold-up of gold in the carbon-recovery circuit. More than 99,9 per cent of the gold and more than 99,8 per cent of the silver was recovered in the adsorption circuit, while the elution data indicated an average elution recovery of 99,8 per cent of the adsorbed gold and 98,2 per cent of the adsorbed silver. Elution of the carbon was carried out at 91 QC. This involved pretreatment of the carbon with a sodium cyanide solution, followed by elution with deionized water. At a loading flowrate of 66 bed volumes per hour, the indicated take-up of metals on the carbon in the lead column was as follows; 53 k g/t gold, 6 k g/t silver, 57 k g/t copper, 14 k g/t nickel, and I k g/t zinc. Only a trace of cobalt was adsorbed. At an elution rate of 1,7 bed volumes per hour, the lead column was found to be completely stripped in under 5 hours, with concentrations of gold in the eluate reaching 6 to 7 g/l. Besides the excellent adsorption and elution results obtained when acidified pregnant solution was used, no fouling of the carbon was indicated, and the pressure drop across the three columns was considered to be well within practical limits. When compared with conventional zinc precipitation, the carbon process would appear to have considerable economic merit, with strong indications that significant savings would be possible both in capital and operating costs. Further pilot-plant experiments, preferably on a gold plant, would be required before a meaningful cost comparison could be made with the existing process.

Jan 1, 1979

By F. H. Deist, R. P. Plewman

The history of development of analytical techniques for the determination of elastic behaviour around tabular excavations, is examined briefly. The design of a completely digital system, which appears to afford considerable improvements in speed, accuracy and versatility compared with existing analogue methods, is described in detail. The practical application of this system is illustrated by means of an actual underground problem. The results are presented, not in the form of specific solutions to the mine's problems, but as indications of the versatility and scope of the new method.

Grains are subject to contamination by mycotoxins which are metabolic by-products of various fungi. The fungal infestations produce local concentrations of the mycotoxins thousands of times higher than the average. These facts make the sampling of the grain for assessment of the mean mycotoxin levels potentially very difficult. In the context of sampling theory, this morphology of the infestations creates a very high distributional and intrinsic heterogeneity of the grain shipment. The distributions of mycotoxin concentrations in seaborne shipments that exist at loading and the number and intensity of infestations that develop under shipping conditions are unknown. All grain shipments carry spores that may give rise to mycotoxin-producing colonies; moisture and temperature determine the final extent of the problem. This paper develops the value of the sampling constant with respect to intrinsic heterogeneity based on data on observed Ochratoxin A (OTA) concentrations in individual wheat kernels. A useful relationship between the distribution of mycotoxin concentrations on a kernel to kernel basis and the distribution of concentration over subsamples is developed.

Jan 1, 2009

By F. R. Deist, R. P. Plewman

Discussion: A. Hazell (Associate Member) and T. J. Kotze (Graduate Member): The authors must be congratulated on the development of a versatile computer programme for determining the stresses and displacements resulting from mining operations. As a design tool it will no doubt take its rightful place together with the analogue computer. Union Corporation made use of this digital computer programme whilst still in its development stage to determine the stress distributions which were causing the unexpected failure of the Marievale sub-incline shaft. Two features arising from this investigation may be of interest to members, namely: A. The solution to the problem of stabilizing a rectangular inclined shaft, which turned out to be contrary to accepted mining practice, and B. A comparison between the results obtained with the pure digital system and the analogue computer. A. The shaft is nine ft high and 20 ft wide, dipping at 9° to the horizontal and crossing the bedding planes. Fig. 1 is a plan and Fig. 2 is a section of the sub-incline shaft pillar area showing the relative positions of the shaft and reef as well as their respective dips. The deepest point in the shaft is 3 300 ft below surface and the requirement of a stress free ground surface may not be satisfied. It also means that the original gravitational stresses are low. A 900 ft wide pillar was left to protect the bottom section of the shaft, the top section being understoped years ago. Extraction of the pillar was started at the beginning of last year, mining operations being carried out in the accepted manner, starting beneath the shaft and mining outwards. The approach did, however, turn out to be the wrong one, because failure of the shaft was noticed soon afterwards. Even small movements were serious in this case due to the low clearance between the skips and the shaft supports. During October, 1968, failure between benchmarks 8 and 10 (see Fig. 1) became so excessive that hoisting speeds had to be reduced. Failure continued to occur and, by the end of March, 1969, failure had propagated up the shaft as far as benchmark 7, and down as far as benchmark 12. As this shaft had a tight hoisting schedule it was imperative that a quick solution be found to arrest the failure.

By R. Pretorius, N. Mautsa

In a cobalt precipitation plant, phosphoric acid is used to eliminate excess uranium from the pregnant leach solution, resulting in a cobalt hydroxide product that has little uranium contamination. However, the amount of phosphoric acid added must be monitored and controlled to avoid financial and sustainability impacts and to prevent negative impacts on the final product quality. One possible solution to this problem is the use of online residual phosphorus measurement techniques to regulate the addition. By continuously monitoring the residual phosphorus levels in the solution, the plant can adjust dosing in real time to maintain the target residual phosphorus level. As a result, the cobalt precipitation plant can improve phosphorus level control, lowering the risk of overdosing, and ensuring consistent product quality. This approach can also improve process efficiency, reduce operational costs, and have a positive environmental impact

Jan 1, 2023

By J. Lux, B. Clermont

Magotteaux, with the co-operation of Anglo Platinum, has erected a fully instrumented pilot plant at Frank concentrator, which is being used to study the influence of a variety of grinding parameters. This paper will discuss only the influence of mill speed and pulp density on secondary stage grinding. A Ø0.82 x1 m open circuit mill was used with reclaimed Merensky tailings as feed. The conclusions apply to this pilot plant but further investigation needs to be done to verify whether the same trends will be applicable to a large industrial-scale mill.

Jan 1, 2004

By K. Moxham, A. Smit, R. Pickering

Since the early 1970s research has been ongoing to develop a cost-effective way of continuously mining the narrow reef, hard rock, typical of South African gold and platinum mines. In the late1980s it was shown that impact ripping was a practical solution for the deep gold mines. In March2001 Lonmin and Sandvik agreed to co-operate in an endeavour to design and manufacture a prototype narrow reef mining machine with the ultimate aim of producing such a machine on a commercial basis that would enable Lonmin to create a safer and more productive environment within its mines and for Sandvik to enhance its leading position as a supplier of a new generation hard rock cutting machine. This paper describes the developed technology, the current progress of the delivery of the required objectives and the optimum mining layout for the roll-out of the mining method.

Jan 1, 2006

By H. Naghavi, M. Karimi

"In self-aerated flotation machines, the gas rate depends on operational variables (e.g. froth depth and impeller speed), pulp properties (e.g. solid content and viscosity), and reagent addition (e.g. type and concentration of frother). The gas rate has a strong correlation with the flotation performance by influencing the gas dispersion properties and froth retention time. A factorial experimental design was used to study how the gas dispersion properties, the froth retention time, and the flotation performance respond to changes in froth depth and impeller speed (as the most common operational variables). An in-depth understanding of the effects of impeller speed and froth depth on the gas dispersion properties, especially the bubble surface area flux and froth retention time, is necessary to improve operating strategies for self-aerated flotation machines. All experiments were carried out in a 50 m3 selfaerated flotation cell in an iron ore processing plant. The results showed that the froth depth affected the metallurgical performance mostly via changing the froth retention time. The impeller speed had two important impacts on the metallurgical performance via varying both the froth retention time and the bubble surface area flux in the froth and pulp zones, respectively. The interaction effects of the froth depth and impeller speed were also established. This allowed us to develop a strategy for operating selfaerated flotation machines based on varying the froth depth and impeller speed with regard to the cell duty. IntroductionA mechanical flotation machine may be divided into two distinct zones. namely the pulp zone and the froth zone (Goodall and O’Connor, 1989; Yianatos, Bergh, and Cortes, 1998; Rahman, Ata, and Jameson, 2015a, 2015b). The overall flotation recovery, Ro, including the true flotation (particles attached to the bubble lamellae) and entrainment (particles recovered in water held in the bubble plateau boundaries), is a function of the recoveries in the two zones. It is calculated as follows (Dobby, 1984):"

Jul 1, 2019

By R. I. Klintworth

The multi-reef area of Cooke 3 Shaft provides a unique mining challenge on account of the very thin middlings between the pay bands and the marginal values of the reef. It is uneconomic or impractical, because of the excessive heights, to mine the middlings using conventional stoping methods. The development of suitable mining methods is discussed with particular emphasis on the control of water inflow and the policy of not jeopardizing or sterilizing other reef horizons by the mining method implemented. No definite solution is currently available, but with the aid of instrumentation, computer analyses and mechanized mining equipment, the mine continues to attempt to maximize extraction and profits.

Jan 1, 1988

By M. E. E. Douglas, A. K. Haines, D. I. Nicol, C. M. Olëm, W. M. Craig, G. D. Louw

A description is given of the design of the continuous ion-exchange demonstration plant at Blyvooruitzicht Gold Mine, including details of the process design, the column construction, and the control system. The operating and process results gathered over a period of seventeen months are summarized, and development work and modifications to the process are discussed. It is concluded that the system comprising continuous loading and continuous elution is technically feasible and can be scaled up with confidence.

Jan 1, 1978

By L. Mabiza

Keywords: Pyrometallurgy, circular furnace, PGM, platinum, Peirce-Smith converter, Zimbabwe The Zimplats Smelter, located on the northern part of the Great Dyke, at the Selous Metallurgical Complex, is the only PGM smelter in Zimbabwe. Zimplats? operations currently consist of the Ngezi Platinum Mine, which is a 1.5 Mt/a open-cast and 0.7 Mt/a underground mine, and a Concentrator and Smelter to which ore is transported 80 km by road. The final converter matte product is transported to Impala Refining Services in South Africa for refining and marketing. The Zimplats Smelter operations consist of a concentrate filtration plant, a Drytech flash dryer, a modified Elkem / Hatch 13.5 MVA circular furnace, two 10? x 15? Peirce-Smith converters, and a matte granulation plant. The Zimplats smelter uses a three-field electrostatic precipitator, supplied by ELB-Brandt, for removal of particulates in the gas streams from both the converter and furnace. Construction of the smelter commenced in 1995, under the BHP / Delta Gold of Australia joint venture. The furnace was lined mid-1996, and smelting operations commenced in 1997. Due to design challenges, the Elkem furnace was modified by Hatch in 1998, and the smelter had not reached full capacity when operations were suspended in 1999. Re-commissioning took place in January 2002.

Jan 1, 2006

By D. G. Krige

The problem addressed is that of estimating grade differences between sub- divisions of a new gold mining prospect when only borehole results are available. Such differential grade estimates are required in choosing the best area for initial production, and in defining the grade profile over the mine's life for purposes of feasibility studies. It is also often necessary to differentiate between the grades and hence the values of subareas, owned by separate companies, for the purpose of fixing relative share participation percentages in the new mining company. It is shown that exaggerated grade differentials will be obtained from grade estimates of the subareas based only on boreholes within each such subarea, respectively, and that it would be more realistic to apply some 'smoothing', preferably by kriging. This is feasible because the lognormal-de Wijsian model can be used to give an indication of the spatial structure for the boreholes, even where they are spaced far apart; also the degree of 'smoothing' indicated is shown to be relatively insensitive to inaccurate estimates of the relevant semivariogram parameters.

Jan 1, 1987

Four engineering products and three consumer products have received awards in the 1978 Shell Design Awards competition organized by the Design Institute of the South African Bureau of Standards. Norman Hanson Visiting Fellowship The Norman Hanson Visiting Fellowship, supported jointly by the National Development Fund for the Building Industry (inaugurated by the Building Industries Federation (8.A.)), the Institute of S.A. Architects, the Association of S.A. Quantity Surveyors, and the National Building Research Institute of the CSIR exists to finance fully or wholly visits by eminent authorities from overseas to South Africa for the purposes of seminars, informed discussions, consultation, and possible active participation in scientific, technological, and development projects of importance to the building industry.

Jan 1, 1979

By M. Keevy, K. Hackney

Centrifugal slurry pump station and deposition sites are a common feature of many mines and mineral processing plants. Yanacocha Mine has completed the design of a pump station consisting of two trains of four pumps each. A number of trade-off studies was conducted to optimize the pump station design. The most significant of these was determining the drive types for the slurry pumps. The pump station duty can vary considerably due to the number of deposition points round the disposal site. The standard solution to this problem on a multi-stage pump train is to install one pump with a variable speed drive to control the pump station output by varying the speed of this pump and stopping and starting other stages. However, there are a number of negative impacts associated with this approach. These include head losses through pumps that are stationary or ?wind-milling?, different wear life on pumps that are running and pumps that are stationary, and slower response to the control system with the possibility of hunting. The use of variable speed drives on all the pumps was considered as an alternative to the standard solution. The factors considered in comparing these options were the control and operational differences, the pump efficiencies and absorbed power, the expected pump wear life, the impact on pump maintenance and the cost of the drive options. For this project the option of using all variable speed drives was selected. The paper presents the following aspects of the study: ?Description of the various options investigated. This includes a discussion on the required duty and the drive options. ?Technical comparison between the options. ?Description of the factors influencing pump performance and life.

Jan 1, 2007

By A. V. Roschin, V. E. Roschin

Composite ores are more often used for production of iron in connection with reserve depletion of traditional ore crude in the regions with developed metallurgy. In particular, the majority of perspective ferrum-ore deposits of the Urals are presented by titanferous magnetite ores which contain a significant amount of titanium and vanadium alongside with ferrum. Thus, the deposits of titanferous magnetites of the Southern Urals on the average contain about 20 % of Fe, 8,7 % TiO2 and 0,24 % V2O5. Due to a high content of titanium oxides it is impossible or it is inexpedient to process such ores in blast furnaces. During work just on titanferous magnetite ore refractory titanium-containing slags are formed in a blast furnace, and in case of its dilution by ores of different types the titanium due to the small content in slag is thrown out. It is often impossible to separate such ores into ferriferous and titanferous concentrates because ferrum and titanium are present in the same minerals. This work attempts to selectively reduce ferrum in solid oxide and to separate metallic iron from titanferousslag with fast fusion. Solid-phase reduction of ferrum has been carried out with carbon at temperature of 1000? 1300°?. In products of solid-phase reduction the extremely disperse mix of metal iron and only partially reduced titanium oxides have been achieved. It is impossible to separate such a mix mechanically. However, the fast fusion of the mix allowed separating it into cast iron and slag with the high content of titanium. Thus the cast iron contained less than 0,1 % of titanium, and the slag did not contain practically any ferrum.

Jan 1, 2007

By A. F. Lee

Dr E. C. Hanekom (Visitor): Two years ago we used the bubble pick-up technique as developed at Anglo American Corporation for screening candidate flotation agents prepared in our laboratories. It was obviously necessary to know to what extent the technique gave reproducible results and a statistical evaluation of the test method was made using a mineral sample provided by Anglo American Corporation. A bubble of 2.0 mm diameter was used and the measuring graticule was divided into squares (units) with sides of 5.0 mm at 25 X magnification. The mineral granules had diameters from 53 to 74 mm. The bubble was loaded by depressing it against a layer of mineral grains to give an arbitrary but constant compression of 0.4-0.5 mm. It was rolled five times through 2 mm horizontal displacement. The bubble was then lifted, its load compacted by gently stirring and the projected area measured in squares (units) on the grid. Each substance was tested at 5 ppm concentration in distilled water at five points in the pH range 3-6. With these standardized conditions of operation the reproducibility of the test was studied. Four materials were tested and at each pH value the procedure of forming a bubble, picking up granules and counting the units was repeated ten times. From these results the standard deviation for a single count at any pH value was calculated for each material (Table 1. The coefficient of variation of approximately 3 per cent was relatively low and it was decided that further refinements in the counting procedure would be of doubtful benefit with the present apparatus. Similar experiments were carried out with 12 sample materials. In this case the bubble formation and pick-up were repeated four times at each pH value, the mean value was calculated for each set of four observations and final pick-up values for the materials were obtained by selecting the highest mean count. Each test was completely repeated a number of times as indicated in Table 2 and the standard deviation calculated. The results allowed the selection of the more promising materials and the rejection of poor performers at the 95 per cent confidence level. All tests were carried out by one operator. Some evidence was obtained that the method is sensitive to operator bias. In conclusion it can be said that from all the results analyzed it would appear that once operation conditions are fixed (viz operator, pH control) the precision of the method is adequate. The author wishes to thank Messrs African Explosives and Chemical Industries Limited for permission to publish these data. G. S. Stander and C. J. Kooij (Visitors): We must congratulate Mr Lee for this most interesting and useful development of the pick-up experimental method. It certainly provides research workers with an instrument which makes flotation more of a science and less of an art. We visited Mr Lee at his laboratory where he gave us a demonstration of the bubble pick-up instrument and technique as developed and improved by him.

Jan 12, 1969

The 107th Annual General Meeting of The South African Institute of Mining and Metallurgy was held at The Country Club, Johannesburg, Napier Road, Auckland Park on Wednesday, 11 August, 2004. Welcome The President extended a special welcome to the guests and representatives of our sister institutes and other associations, also to recipients of awards, senior members of industry, Honorary Life Fellows, past-presidents, our members and other guests, among them the following: Ralph Bagirathi, Executive Member, African Mining Association Rudi Rudolph, President, Association of Mine Managers of South Africa Andrew Whibley, President, Association of Mine Resident Engineers of South Africa Dr Con Fauconnier, President, Chamber of Mines of South Africa Nxolisi Diliza, Chief Executive Officer, Chamber of Mines of South Africa

Jan 1, 2004

By J. W. Klokow, K. L. Riemer, R. I. L. Ferreira

As a result of the increased depth of mining in the South African gold mining industry mining induced seismicity is a severe problem. One of the strategies within the industry to contain the problem is to incorporate a system of regularly spaced stabilising pillars into mine layouts. At the Driefontein Gold Mine a system of closely spaced dip pillars is used to contain seismic energy emissions. Comparisons of seismic data for several mining clusters initially indicated a possible measure of success in the containment and occurrence of larger events. A detailed analysis of data over a 5-year period since the start of mining however, has revealed a delayed seismic response to the increasing strike spans. Significant changes marked by an increase in certain seismic variables have been observed at mining spans of between 100 to 120 meters and could be of significance to pillar spacing in the design of deep level mining layouts. La sismicité occasionnée par l’exploitation minière est un problème grave dans les mines profondes de l’industrie minière aurifère en Afrique du Sud. Une des stratégies utilisées dans l’industrie a été de maîtriser le problème en incorporant un système de piliers de stabilisation régulièrement espacés dans la disposition des mines. A la Driefontein Gold Mine, un système de piliers de pendage proches les uns des autres est utilisé pour contenir les émissions d’énergie sismique. Au début, des comparaisons de donnés sismiques pour plusieurs groupes d’extraction ont indiqué un succès possible dans la limitation et l’occurrence d’évènements sismiques. Cependant, une analyse détaillée des données sur une période de 5 années a révélé un délai entre la progression de l’extraction et une réponse sismique accrue. Des changements considérables dans les variables sismiques ont été observés dans des intervalles d’extraction entre 100 et 120 mètres, ce qui pourrait être significatif pour l’espacement des piliers dans la conception de la disposition des mines de fond. Resultierend aus den zunehmenden Tiefen des Bergbaus in der südafrikanischer Goldbergbauindustrie stellt sich die vom Bergbau hervorgerufene Seismizität als bedrohliches Problem dar. Es war eine der Strategien innerhalb der Industrie, das Problem dadurch einzudämmen, daß regelmäßig plazierte Stabilisierungspfeiler in den Bergbaulayouts inkorporiert wurden. In der Driefontein Gold Mine wird ein System von knapp aneinandergesetzten im Abfallbereich befindliche Pfeiler benutzt, um seismische Energieemissionen einzudämmen. Vergleiche der seismischen Daten für mehrere Bergbaugruppen zeigten anfangs ein mögliches Maß von Erfolg in der Eindämmung und im Auftreten von größeren Fällen an. Eine ausführliche Analyse von Daten über einen 5jährigen Zeitraum hat jedoch ergeben, daß eine Verzögerung zwischen der Progression des Bergbaus und einer erhöhten seismischen Reaktion besteht. Bedeutende Änderungen in den seismischen Variablen wurden an Bergbauspannen von 100 bis 120 Metern beobachtet und könnten für Abstandsbemessung von Pfeilern im Design des Tiefbergbaulayouts von Bedeutung sein.

Jan 1, 2003

By V. M. Lovell, N. P. Finkelstein

The concepts involved in flotation research are discussed and aims are presented. Three different method. of investigation can be used, all of which are applied at the National Institute for Metallurgy (N.I.M.), and these are the empirical, trial and error method, the chemical engineering method, and the physicochemical approachs This paper is devoted to the last of these. Because of the complexity of the physicochemical interactions involved, the work has, to a large extent, been concentrated on only one aspect, viz. the interaction between the mineral surface and reagents in solution, which was considered of major importance in the understanding of the process. The aspects on which work has been done are the concentration of uranium, the reaction between salt minerals and fatty acids, mixed xanthate-cationic collector systems, and the reaction between sulphide minerals and thiol reagents. Selected problems from the first two of these are discussed in some detail and deal with the agglomeration flotation of uranium and the chemical control of the flotation process for phoscorite. A detailed discussion is also presented of the progress made in the understanding of the interaction between sulphide minerals and thiol reagents, which is of more general application. Finally, an indication is given of the future direction of physicochemical research at the N.I.M.