The relative effectiveness of anionic and nonionic polyacrylamide in enhancing the settling rate of flotation tailings was studied by performing batch settling tests. Anionic polyacrylamide was found to be more effective than nonionic polyacrylamide. A liquid flocculant was found to give lower settling rates than dry flocculants due to a lower polyacrylamide content. The aging of tailings resulted in increased settling rates. Residual polyacrylamide was detected in the supernatant water even at low flocculant addition rates and reached significant concentrations before the optimum addition rate was attained. The effect of polyacrylamide on flotation of a copper-molybdenum ore was investigated in a series of batch flotation tests. Flocculant concentrations up to 10 parts per million did not lower either concentrate grade or recovery. The presence of an oil phase from the liquid flocculant also had no noticeable effect on flotation. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

A study has been conducted on the loss of phosphate ions and excessive oleic acid consumption encountered in the selective flotation of dolomite from apatite. An attempt has been made to evaluate the mechanism by which orthophosphate ions depress apatite flotation. Replacement of SO₄⁻² ions on gypsum by HFO₄⁻² ions is found to occur rapidly in solution. The presence of CaHPO₄⁻² 2H₂O on gypsum surfaces is shown using infrared reflectance spectroscopy. CaHPO₄⁻²H₂O is isomorphous with gypsum and has identical lattice parameters. Experiments show that the addition of sulfate ion to the system suppresses the reaction of orthophosphate ions with gypsum by the common ion effect. The proposed reversible reaction in the system is: CaSO₄∙2H₂O + HPO₄⁻²↼⇁ CaHPO₄ ∙2H₂O + SO₄⁻² A proposed mechanism by which orthophosphate ions selectively depress apatite flotation is shown to fit all experimental observations. Orthophosphate ions are known to be potential determining for calcite, dolomite and apatite. It is observed that in the presence of orthophosphate ions calcite and dolomite recovery is higher at pH 6.0 than at pH 8.5. Apatite and gypsum exhibit opposite behavior. It is proposed that the adsorption of strongly hydrogen bonding H₃PO⁴, H₂PO₄⁻² and HPO₄⁻² in the electrical double layer results in collector species being slow to penetrate and adsorb on the minerals. Acid attack on calcite and dolomite results in disruption of the hydrogen bonded layer and allows rapid collector adsorption. Two equally important effects of adsorbed phosphates are to decrease collector adsorption due to the large negative zeta potential generated and to impede fruitful particle-bubble interactions. Brittle froths encountered in flotation at pH 6.0 are related to condensation of surface films formed by oleic acid -sodium oleate complexes. The froth stabilizing effects of hydrophobic dolomite particles is noted. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

The effect of manipulated variables on the flotation response of run-of-mine coal supplied by Bullmoose Coal Mine (North East British Columbia) was studied in a pilot flotation column and a batch mechanical cell using experimental designs of the factorial class after the coal had been stage crushed and ground to about 96 percent minus 600 µm, the feed size to the flotation circuit at Bullmoose Coal Mine. The objective was to optimise statistically the two flotation units usings (1) screening designs, (2) the steepest ascent technique, and (3) central composite designs, and to compare cell performance. The efficiency index was employed as the primary optimisation criterion. The two cells yielded comparable efficiency indices, but the product ash content of clean coal from the flotation column was consistently lower. Evaluation of the efficiency indices of individual size fractions was also carried out at the optimum conditions of each cell. The approach taken was to separate the feed into individual size fractions, retain the fractions separately, and float them separately. The flotation column yielded higher efficiency indices and lower product ash content when a feed with 15 percent solids was floated. Results obtained by floating individual size fractions at 2 percent solids indicated that it is difficult to float sizes above 300 µm in both cells at such a low pulp density. Below 300 µm, the performance of both cells was comparable. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

Electrochemical methods, surface spectroscopy and flotation tests have been used to study the influence of the oxidation of arsenopyrite on its floatability with xahthate. Cyclic voltammetric studies indicated that the oxidation of arsenopyrite at pH greater than 7 results in the formation of ferric hydroxide deposits on the surface of the mineral. Arsenic is oxidized to arsenate and sulphur is oxidized to sulphate. The arsenate is incorporated in the ferric hydroxide deposits while sulphate diffuses into solution. Below pH=7, soluble iron species are formed and the surface becomes increasingly covered with elemental sulphur with decreasing pH. Increasing temperature has no influence on the quantity of hydroxide formed over the range 30° to 45°C but results in thick, porous films at temperature greater than 45°C. The oxidation of arsenopyrite was demonstrated to occur at lower oxidation potentials than for pyrite although this effect decreased with increasing temperature. Mixed potential studies indicated that the potentials required for arsenopyrite oxidation could be achieved with common oxidizing agents. Selective oxidation of arsenopyrite in a bulk pyrite-arsenopyrite concentrate was indicated to be possible. The formation of iron hydroxide deposits on the surface of arsenopyrite resulted in the inhibition of subsequent oxidation of xanthate to dixanthogen at the mineral's surface. ESCA studies confirmed the formation of oxidized iron layers at the surface of arsenopyrite and revealed that essentially all the arsenate which was formed was incorporated in these layers. Sulphur became oxidized at the pH studied and to a large extent went into solution. Flotation studies demonstrated the use of oxidation for arsenopyrite depression. In the presence of oxidation, increasing pH above pH=7 resulted in increased arsenopyrite depression while increasing temperature had little effect until a temperature of 40°C was exceeded. Previously activated arsenopyrite could be depressed through the use of oxidizing agents. Arsenopyrite could be selectively depressed from a bulk pyrite-arsenopyrite concentrate through the use of oxidizing agents. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

Thioglycollic acid (TGA) was found to exhibit strong depressive action on chalcopyrite, galena and sphalerite samples. This depressive action resembled hydrosulphide in that the contact curve of TGA was very similar to that of sodium sulphide. An adsorption model was proposed based on the experimental observations. The model suggests that TGA molecules adsorb onto sulphide mineral surfaces through either their mercapto groups, or their carboxyl groups, with subsequent multilayer adsorption of TGA onto this chemisorbed layer through hydrogen bonding and oxidative dimer bonding. The TGA molecules adsorbed through the carboxyl group were not responsible for the observed depression: these molecules were easily desorbed by potassium ethyl xanthate. Oxygen was found to enhance the adsorption of TGA onto sulphide mineral surfaces and consequently, to enhance the depressive action. The selective separation of synthetic sulphide mineral mixtures with the use of TGA has met limited success. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

Although the technical literature contains abundant references on applications of starch in mineral processing, the majority is not concerned with phosphate mineral systems. Nevertheless, the interaction between starch and apatite surfaces is relevant to both selective flocculation and flotation of phosphate ores. The main objective of this thesis is to investigate in detail such interaction in order to provide a more clear understanding on the behaviour of apatite/starch systems. Considerable research effort was dedicated to a thorough characterization of the starch samples used, especially in those aspects most pertinent to the application of starches as flocculants and depressants. Presence of ionic impurities in the starch samples tested was identified by infrared spectroscopy and microelectrophoresis. These impurities (proteins, carboxylic groups and, possibly, phosphate esters)were found to play an important role in the mechanisms governing the interaction of starch macromolecules and mineral surfaces. In a first stage of this research, the interaction between aqueous solutions of starches (and starch fractions - amylose and amylopectin) with calcium ionic species and surfactants (flotation collectors) was investigated. Depression of solution electrical conductivity, experienced in Ca-starch systems, was indicative of chemical reactions taking place (complexation). For surfactants, evidence for their interaction with starch fractions was obtained by UV-Vis. spectroscopy. The spectra of starch/surfactant solutions in the presence of iodine were altered indicating the substitution of iodine species by surfactant molecules at the helical sites of starch macromolecules. The next step involved the study of the adsorption of starch onto both apatite and silica mineral surfaces. Preliminary tests pointed out that a much stronger interaction took place in the case of apatite. Starch adsorption isotherms obtained for fluorapatite and quartz confirmed the preferential adsorption of starch onto the phosphate mineral surface. Both amylose and amylopectin were strongly adsorbed on fluorapatite but the latter polymer displayed the largest extent of interaction on a weight per area basis. The shape of the adsorption isotherms for the two starch fractions on fluorapatite also corroborates the idea of a stronger interaction by amylopectin. In turn, whole starches displayed adsorption isotherms resembling more closely that obtained for amylopectin. Adsorption of starches on fluorapatite was increased considerably in the presence of Ca ionic species. In the absence of externally added Ca ionic species, the amount of Ca released by the mineral surface was dependent upon the amount of starch adsorbed. These two phenomena can be interpreted as indicating the importance of Ca sites and presence of Ca species for the adsorption of starches, hence justifying the preferential adsorption displayed for apatite. Adsorption of starch on quartz surfaces was also enhanced in the presence of Ca ionic species, once more confirming the important role played by calcium on the adsorption of starches. Flocculation studies were also conducted with fluorapatite, kaolinite and quartz suspensions in the presence of different starches. Under the conditions tested, all starches samples failed to promote aggregation of the two non Ca-bearing minerals. In turn, fluorapatite suspensions were flocculated rather strongly by all starches. Maximum flocculation of fluorapatite was achieved at partial polymer coverage conditions. With one exception (amylose), increasing the concentration of the polymers above an optimum level, generated partial re-stabilization of the suspensions, probably via a steric effect. All starches depressed both anionic and cationic flotation of fluorapatite. Amylose was the least effective depressant among all starches, especially for the cationic flotation system. The depressant action was a function of pH and of the relative amounts of polymer and surfactant. Alkaline pH favoured depression, whereas as the collector level was increased, the depressant action was diminished and eventually eliminated. The experimental evidence gathered in the present research supports a chemical mechanism for the interaction between starch and apatite surfaces. Calcium plays a dominant role, and its importance for the adsorption of starches onto mineral surfaces is most probably related to the formation of complexes between starch impurity-related ionic groups and Ca ionic species. Hydrogen bonding and to lesser extent electrostatic forces are also important for the overall interaction between starch and apatite surfaces. The larger extent of interaction for the amylopectin fraction(highest molecular wt.) as compared to that of amylase (lowest molecular wt. fraction) gives support for the accessory role of hydrogen bonding. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

An electrochemical study was carried out on various electrodes of sphalerite, pyrite, pyrrhotite, chalcopyrite and galena in deoxygenated and air-saturated solutions at different concentrations of potassium ethyl xanthate (KEtX), i.e. 0 M, 10⁻⁵ M, 3 x 10⁻⁵M, 10⁻⁴ M, 3 x 10⁻⁴ M, 10⁻⁴M, 10⁻³ M and pH (1 to 12) at 25°C ± 1°C. The effect of galvanic contact between various sulphides and metallic iron on various electrochemical characteristics of the galvanic couple was also investigated. Following a pseudo Nernst relationship, the rest potentials of various electrode systems became more negative with the higher concentration of KEtX. The electrochemical series of the sulphide minerals investigated, in potassium ethyl xanthate solution, in the order of their nobility, were sphalerite, pyrite, pyrrhotite, chalcopyrite and galena. All sulphide-xanthate-air systems displayed more noble rest potentials than the redox potential of potassium ethyl xanthate at natural pH; however, the rest potentials of most sulphides except for sphalerite became more negative than the redox potential of potassium ethyl xanthate at a pH of 9. Solution purging with air caused the rest potentials of all sulphides investigated to shift in the noble direction because of the oxygen reduction. Various sulphide-iron galvanic couples acquired mixed potentials which lay between the rest potentials of sulphides and iron. The exact value of the mixed potential of the particular electrode system was varied depending on the relative surface area of the two electrodes in galvanic contact. The experimental potential-pH diagrams were constructed for the minerals and metallic iron in the solutions of 0 M and 10~5 M of potassium ethyl xanthate. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

A highly sensitive reflection-absorption infrared spectrophotometric technique was used to investigate the adsorption of sodium oleate from aqueous solutions onto magnesium carbonate (magnesite) thin film substrates. Valuable qualitative results were obtained, regarding the nature of the adsorbed species and their molecular orientation. However, this orientation effect strongly influenced the relative intensities of the infrared absorption bands; this prevented precise quantitative determinations. Under moderately acidic conditions, the original surface charge of magnesite was positive and near zero. The anionic collector was chemisorbed, forming only one type of complex, very similar to the bulk magnesium oleate. The hydrocarbon chains were oriented nearly parallel to the substrate, forming a homogeneous partially polymerized epoxide-polyether layer, strongly laterally bound, at the sites of the former C=C double bonds. The C-0 bonds were formed following a partial oxidation of the oleate in solution, by oxygen from the air. This adsorbed film converted the carbonate surface to a strongly hydrophobic condition, which was responsible for the high flotation recoveries obtained. Under basic conditions, the original magnesite surface was negatively charged. This limited the formation of the chemisorbed collector complex to a lower amount; different complexes, probably physically adsorbed, joined in forming a film. However, this was a more heterogenous film, showing different molecular orientations - - from nearly parallel, to closer to the normal to the adsorbent's surface. These aspects caused a net decrease in the degree of hydrophobicity, in comparison to the acidic conditions. The same oxidation and polymerization effects were observed, although to a lower extent. The film was also very stable. These properties of the adsorbed film were consistent with the lower flotation response observed under basic conditions. Therefore, more concentrated collector solutions were needed, to reach the same high recoveries obtained under moderately acidic conditions. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

This dissertation presents the results of a collector - depressant equilibria study of the flotation systems potassium n-amyl xanthate - sodium cyanide - molybdenite and ethyl zanthate sodium sulfide - molybdenite and the study of conditions necessary for the selective flotation of covellite, tetra-hedrite, bornite, chalcopyrite, molybdenite, marcasite and pyrite in the system ethyl xanthate and sodium cyanide; the conditions necessary for the selective flotation of pyrite, covellite, bornite, galena, chalcopyrite, active sphalerite, and molybdenite in the system potassium ethyl xanthate and sodium sulfide; the conditions necessary for the selective flotation of molybdenite and pyrite in the system of potassium n-amyl xanthate and sodium cyanide; the condition necessary for the flotation of molybdenite and galena in the system potassium n-amyl xanthate and sodium sulfide.

A thesis submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science in Engineering, Johannesburg August 2018 / The flotation process is widely used for upgrading valuable minerals in the field of mining. Many diverse minerals, including most of the world’s base and precious metals are processed by flotation process. Most valuable products produced by flotation pass through the froth phase of the flotation process. The froth phase has attracted more research in recent times because of its significant role in determining the mineral grade and recovery achieved from a flotation operation. The complex processes that occur in the froth phase – detachment, re-attachment, coalescence of bubbles, and competition for attachment sites, mixing and transport all combine to affect the net transfer of mineral particles into the concentrate. Bubbles are formed at different sizes in the pulp phase and coalesce at different rates and as a result the bubble size distribution varies from point to point in the froth phase. Substantial coalescence gives rise to loss of bubble surface area and hence loss of recovery. Competition for attachment sites gives rise to an increase in grade. No method for measuring the variation of froth bubble size distribution (FBSD) was available until Bhondayi and Moys developed one. The method measures the intrabubble impact distance in the froth using a probe dropped at known height through the froth. The average of these intra bubble impacts was considered to be a proxy for froth bubble size distribution; this was calibrated using FBSD. However the measured in the laboratory using photographs taken through the transparent wall of a laboratory cell. A 31 % of error was found and compared to the photographic method, which indicated that the technique over-estimates the actual froth bubble size distributions. This is due to the use of an average IID (proxy) as an estimate of the bubble. In response to the known of actual froth bubble size distribution FBSD in order to quantify the complex processes in the froth phase, an application of a stereological technique/model was developed and tested to obtain estimates of the actual froth bubble size distribution FBSD in lab flotation and Mintek pilot rougher cells as a function of froth height, frother dosage and superficial gas velocity. The model was first validated for a system of flotation with variable froth height in a transparent Wits lab flotation cell. The two-parameter normal distribution model FBSD was considered to fit the model-predicted intrabubble impact distance distribution IIDDs to measured intrabubble impact distance distribution IIDDs. The model was seen to accurately iii predict the FBSD compared to actual FBSD data obtained from above-mentioned conventional photographic method using a calibration scale attached to the transparent flotation cell wall, wherein the experimental IIDDs were accurately fitted by the model-predicted IIDDs. Similar estimation of froth bubble size distribution were also found with the inversion matrix technique. Secondly, the model was then evaluated for flotation condition with variable frother dosage in the Mintek pilot plant rougher cell. The model was seen to estimate the actual FBSD, wherein the IIDDs were precisely predicted compared to experimental IIDDs. Finally, the model validity was then tested for various systems of flotation conditions with variable superficial gas velocity. The model was seen to estimate the actual FBSD for these cases compared to both model-predicted IIDD and experimental IIDDs. The performance of the present model for these systems of flotation was seen to estimate froth bubble size in froth phase from measured IIDD information. Froth bubble size increases with increasing in froth height, and decreases with increasing in frother dosage and superficial gas velocity. Froth height, frother dosage and superficial gas velocity have a strong effect on froth bubble size distribution. / XL2019