Beena Rai

Selective Flocculation of Iron Ore Slimes: Results of Successful Pilot Plant Trials at Tata Steel, Noamundi

A significant portion of currently mined iron ores, that is, 15–20% of run of mine ore, typically ends up as slimes—a slurry of fine particles (<45 micron) with relatively higher alumina contents (7–15% Al2O3). The only viable option currently is to dispose these slimes in tailing ponds for water recovery and future use. Due to increasing demand for steel and rapid consumption of high grade ores, processing and utilization of these slimes has become an urgent necessity. Traditional gravity and magnetic separation is not very effective in the ultrafine size range (less than 45 microns) and hence most of the flowsheets have a desliming step ahead of gravity and magnetic separation. Selective dispersion—flocculation is commercially proven technology to accomplish efficient separation in this size range, provided appropriate selective reagents are available. Our earlier work in TRDDC laboratories has demonstrated that selective flocculation with starch and guar gum is sufficiently selective for the reduction of alumina contents and thus this process should be scaled up for the processing of alumina rich Indian iron ore slimes. TRDDC has also designed a portable set-up which can be used to run pilot plant trials in a continuous mode onsite in the mines. The successful pilot plant trials of a selective flocculation separation process were completed by us at Noamundi mines recently. The results of these plant trials, using a portable pilot plant set up, designed by TRDDC, are summarized in this paper. The slimes sample for these plant trials, was withdrawn from one of the slimes stream at the Tata Steel’s iron ore beneficiation plant at Noamundi mines. Though there was considerable variation in the grade and pulp density of the slimes sample, it was possible to produce a concentrate of consistent quality in a continuous mode. It was for example demonstrated that one can produce a concentrate assaying 65.3% Fe and 2.5% alumina with a yield of 80.4% from a relatively richer grade slime sample containing 60.3% Fe and 4.8% alumina. More importantly, the corresponding tails contained less than 39.7% Fe. For a leaner grade slime sample assaying 53.4% Fe and 7.3% alumina, the corresponding concentrate grade was 63.5% Fe and 3.1% alumina, tailings grade 35.7% Fe and the yield achieved during the plant trials was 63.5%.

Quantitative Estimation of Mineral Phases from Chemical Assays and Powder X-Ray Diffraction Rietveld Analysis: A Case Study on Selective Flocculation of Iron Ore Slimes

Abstract A generic method for quantitative estimation of mineral phases in ores and process streams has been illustrated in this paper with the help of an example taken from our own work on the selective flocculation of alumina rich iron ore slimes. The method used conventional bulk chemical assay data and information on phases present, as determined by powder X-ray diffraction (XRD) pattern analysis. Iron ore slime samples were subjected to selective dispersion–flocculation experiments and the feed, concentrate and tails were analyzed using this method. Chemical assays (iron, alumina, silica and loss on ignition) were determined by wet chemical analysis and mineral phases (hematite, goethite, gibbsite, kaolinite and quartz) were quantified by Rietveld analysis of XRD patterns. Considering the proportion of various elements present in the constituent minerals, the chemical assay data and the reconciliation of assays (as well as the mineral compositions determined by Rietveld analysis, if available), one could estimate the proportion of various minerals present in different process streams. The mineral compositions estimated based on this method matched reasonably well with those obtained by quantitative phase analysis by the Rietveld analysis. In case the Rietveld analysis results were also reconciled with the elemental assays, one could arrive at even a better estimate of the proportion of various minerals in the different process streams. The estimated mineral compositions in feed, concentrate and tails were also used to compute mineral wise recoveries. The approach presented in this paper would provide mineral engineers a simple and relatively easy method to convert elemental assays into mineralogical compositions and thus compute extremely useful and reliable estimates of the mineral-wise recoveries, grades and distributions in the various plant streams, which are necessary for plant optimization and control purposes.