BR112020021665B1 - SOLIDS SEPARATION METHOD - Google Patents

Abstract

The invention relates to a method of separating solids, the method comprising adding the solids to a suspension of particulate matter comprising magnetic particles in a liquid to create a mixture, locating the mixture in a separation vessel such that rotation is imparted mixing around a space enclosed by an outer wall of the container to impart a centrifugal force to the solids; and the mixture is separated into floating and sinking fractions.

Description

Field of Invention

[001] The present invention concerns the separation of solids. The invention particularly concerns the dense medium separation (DMS) of feed fines (less than 1 mm).

Background of the invention

[002] Dense media separation (DMS) is a process widely used in the mining industry to separate valuable minerals from non-valuable rock by differences in density. For example, DMS can be used in the iron ore industry because hematite is denser than quartz, muscovite and kaolinite in general.

[003] The DMS process involves the use of a suspension of particulate matter in a liquid, typically water. The particulate material, or medium, preferably comprises magnetic particles, for example magnetite or ferrosilicon (FeSi) particles, as this facilitates recovery of the particulate material for reuse after the separation process. Particles of particulate matter are fine enough to allow them to be stably suspended in the relevant liquid and normally take the form of powder, whilst being sufficiently dense/heavy to provide the required media density.

[004] The particles of the medium are normally formed by grinding or atomization. The resulting medium suspension is commonly referred to as a dense medium. When the particulate material comprises magnetic or magnetized particles, the suspension of the medium has a greater density than that of the liquid itself. For example, a typical dense medium might have an apparent density of, say, 2.65 specific gravity, while the specific gravity of water is 1. The advantage of using a magnetic particulate material is to facilitate subsequent recovery of the particulate material for reuse. . During use, the medium suspension is contained in a separation vessel, for example, a cyclone vessel (sometimes referred to as a “dense medium cyclone”). The media suspension is generally mixed with the solids to be separated (normally comprising ore) before being transferred to the separation vessel. Where the separation vessel comprises a cyclone, separation is effected by differences in the centrifugal force experienced by particles of different density than the solids to be separated, the less dense material tending to float in the liquid suspension and thereby exiting the cyclone at the top. ("floats"), while the denser material sinks and comes out the bottom ("sinks").

[005] A problem with DMS is that the suspended medium tends to separate from the suspended medium along with the solids to be separated as a result of its relatively high density (typically between 6.7 and 7.1 specific gravity for ferrosilicon). ). Therefore, a stable medium for optimal DMS efficiencies is needed, and optimal efficiencies are a priority more than ever with high commodity prices. Stability is achieved by using a powdered medium that is fine enough to prevent rapid settling of the medium under the centrifugal forces of the cyclone or gravity in the case of a Dense Medium Drum. It is this fineness that gives rise to the majority of media losses for reasons including the following: 1) Fine suspension media adheres to the surface of the ore/solids and becomes difficult to remove from the recovered product at the end of the process. 2) Thin suspension media is more susceptible to corrosion (e.g. oxidation) due to high surface area to volume. 3) Fine suspension media is more difficult to recover in magnetic separators.

[006] The higher hydrodynamic drag forces experienced by fine particles result in low recovery of finer media in magnetic separators.

[007] Commercially available ferrosilicon is manufactured ground or atomized. The atomized version is commonly manufactured in five size fractions: Special Coarse, Coarse, Fine, Cyclone 60 and Cyclone 40 and, as it is spherical, it is easier to wash and more resistant to corrosion, but is more expensive. Milled ferrosilicon is more economical and is commercially available in six different sizes: 100#, 65D, 100D, 150D, 270D, 270F (e.g. from DMS Powders (www.dmspowders.com) or M&M Alloys Limited (www .mandmalloys.com). In conventional DMS plants, where the required specific gravity of the medium is greater than 3.2, as in iron ores, the viscosity of the ground medium is too high for efficient separation, and the atomized ferrosilicon is used.

[008] Generally, the smaller the diameter of the cyclone, the easier it is to operate at higher feed pressures, which results in greater centrifugal forces experienced by the medium and ore particles in the cyclone.

[009] Typically, ferrosilicon losses in cyclone DMS circuits range from 120g of ferrosilicon per ton (g/t) up to 500 g/t. Media losses are known to represent 20% to 40% of the total operating costs of a DMS plant.

[010] It would be desirable to improve separation efficiency for solid feeds with a particle size of -1 mm. It is further desirable to reduce media losses for -1 mm feed DMS systems.

Summary of the invention

[011] The invention provides a method of separating solids, the method comprising: 4) adding the solids to a suspension of particulate matter comprising magnetic particles in a liquid to create a mixture, 11) placing the mixture in a separation vessel of so that rotation is transmitted to the mixture around a space enclosed by an outer wall of the vessel to transmit a centrifugal force to the solids; 111) the mixture is separated into floating and sinking fractions; wherein said particulate material is ferrosilicon having D90 particle size less than 200 μm, the solids are iron ore having a particle size less than 1 mm and greater than 200 microns (-1 mm +200 μm), for example ( -1 mm +212 μm), preferably greater than 400 microns (μm) (-1 +400 μm), the feed density of the mixture is less than 3.8 g/cm3, preferably less than 3.6 g/cm3, the operating pressure of the mixture is 10 to 15D, the cut point differential is between 0.1 and 0.6 of the feed density, and where the mixture includes an iron ore sludge fraction of less than 10 % by weight having a particle size of less than 45 μm.

[012] The suspension of particulate matter comprising magnetic particles in a liquid can be known as the medium.

[013] In one embodiment, the floated and sunken fractions are subjected to magnetic separation before washing to recover particulate material.

[014] Floated and sunk fractions can be passed through separation screens vibrating at a frequency greater than about 50 Hz. The separation screens can have a double oscillation.

Detailed description of the invention

[015] In this specification, the Cyclone Operating Head is a function of feed pressure and feed density (and is reported as a ratio of Total Head (m) to Cyclone Diameter (m), hence the use of “D”.

[016] Operation head (D) = ((Feed Pressure (kPa)/ (Feed Density (t/m3) X Gravitational Acceleration (m/s2)))/(Diameter of a DMS Cyclone (m))

[017] Operating Head and Operating Pressure are often used interchangeably to mean the same thing.

[018] Tests were previously carried out on samples of Kumba iron ore to compare the performance of DMS to that of a spiral circuit in fine iron ore (-1 mm +212 μm), using high and low grade material from the Sishen mine.

[019] It was found that the density distribution of the evaluated samples played an important role in the separation efficiency (Ep) and, consequently, in the general contents and recoveries obtained. High amounts of material of close density have been shown to have a negative effect on both gravity separation methods. It has also been demonstrated that material size distribution is critical in determining the overall Ep obtained in a DMS plant, with significant deterioration of Ep with reducing particle size. An important finding was that, when operating at low DMS feed densities of around 3.4 g/cm3, the FeSi Cyclone 60 was able to achieve breakpoints of >4.0 g/cm3 in fine iron ore (< 1 mm) tested. The use of the more expensive gas atomized FeSi Exxaro was not recommended at this stage, as the required cut-off point in the DMS plant could be achieved with FeSi Ciclone 60. It is worth noting, however, that both batch tests of DMS were carried out without any accumulation of sludge in the circuit, as would be expected in a plant in continuous operation. An accumulation of mud is expected to significantly alter the rheology within the DMS cyclone, along with the performance efficiency of the cyclone.

[020] The current objective of the tests was, therefore, to compare the performance of both types of atomized FeSi in a range of operating conditions in the presence of sludge accumulation, as well as to define the DMS modeling parameters for each of the types of FeSi.

[021] Anglo American Kumba Iron Ore supplied two bulk samples of <1 mm ore for testing at the pilot DMS plant: 1. low grade (BIF sample crushed to <1 mm using HPGR) and 2. high grade iron ore high content (jigging plant feed fines, conventionally crushed to <1 mm).

Experimental procedures: Sample preparation Food preparation

[022] Approximately 1.5 tons of low-grade Banded Iron Stone Formation (BIF) and 4.5 tons of high-grade jig (-1 mm) were provided for the tests. The following sample preparation was performed on each sample received: • Samples were wet sieved at 212μm to produce DMS feed at <1mm >212μm. • The -212 μm fraction was subsequently sieved to 45 μm to produce a <45 μm fraction to be used as slurry addition during DMS passes.

[023] The following feed characterization was performed on -1 mm +212 μm subsamples: • Particle size distribution (PSD) was done on a 2x1 kg subsample, using root two sizing. • Head chemical analyzes were performed on 2x100 g subsamples in duplicate to determine the head content of the sample. • A 1 kg subsample was sized at -1 mm +600 μm and - 600 +212 μm. These fractions were subjected to Mineralogy for Particle Tracking Analysis (PTA) using MLA. • A 500 g subsample was subjected to mineralogy for electron microprobe analysis (EMPA) • Subsamples of 100 kg each (-1 mm +212 μm) of BIF (low content) and Jigue feed fines (high content) were prepared for the pilot DMS trials. Individual 100 kg samples were used for batch DMS passes at relative feed densities of 3.4, 3.6, 3.8 and 4.0 g/cm3 with introduction of -45μm sludge at 5% by mass of quite. Some of the tests were carried out without introducing -45μm sludge for comparison.

DMS assay products

[024] During the DMS operation, the volumetric mass of sunk and float items was obtained. Mass divisions at the DMS plant were determined and subsamples sent for chemical analysis. Subsamples of each product were also subjected to PTA analysis using MLA.

Scope of work Test strategy

[025] Figure 1 and Table 1 below describe the testing strategy for DMS passages. In total, 16 DMS trials were performed - 11 initial trials followed by 5 additional trials. Table 1: Summary of tests carried out

Mineralogy - Particle Tracking Analysis

[026] Mineralogy (Particle Tracking Analysis - PTA) was used to characterize each type of ore in terms of mineral, size distribution and density of the feed and DMS products. The release characteristics of each mineral by size class were also reported.

[027] PTA was performed on the two food subsamples, as well as on all products from the 11 initial trials. Due to the coarse nature of the particles within the samples, each submitted size fraction was assembled into slabs, coated with carbon, and analyzed using autoSEM (MLA). During measurement, MLA generates an X-ray analysis for each region (gray level) within a particle. The measurement mode employed in this study was chosen based on the ore type, passage time, objective of the study, particle size, and the successful ability to delineate mineral grain boundaries within the particles. During the investigation, between 2,500 and 20,000 particles per polished section for each fraction were analyzed and processed via AutoSEM. A greater number of particles assists in acquiring a data set that is statistically representative of the overall sample. Particle characterization data relating to mineral types/compositions, particle size, density, particle population weight percentage, particle area, shape factor, circularity, and perimeter of each particle were verified during off-line processing. For the purposes of this study, calculations based on shape descriptors such as shape factor, circularity, and equivalent circle diameter (ECD) were included in the overall dataset for each size fraction.

Dense Medium Separation (DMS) Pilot Tests

[028] Figure 2 and Figure 3 show the flowchart of the pilot DMS test equipment and the configuration of the pilot DMS plant, respectively, used for the tests. The DMS plant is equipped with a 250 mm diameter Multotec cyclone. The operational parameters of the pilot test were monitored during the test. This includes circulating feed density, cyclone underflow and overflow densities to ensure steady-state conditions. The supply pressure was also monitored to ensure a constant supply pressure to the cyclone. The 600μm DMS sieve bands were covered with 250μm wire mesh to help prevent fines from passing through the sieve bands. DMS tests were performed to achieve values above 12D at different feed densities. A density differential of less than 0.4 RD is recommended for efficient cyclone operation when the proportion of near-gravity feed material is high. The unit has no moving parts, which reduces maintenance requirements.

[029] Once the necessary parameters were achieved and steady-state conditions established, the floated and sunk products from each pass were collected over a period of 30-45 minutes, dried in the sun and weighed. The products were sieved using a 212 μm screen to remove FeSi and a magnet was used to further remove residual FeSi in the +212 μm fractions. Seven representative 10kg subsamples of both float and sink from each passage were subsampled and the products were sent for chemical and mineralogical analysis.

Chemical analysis

[030] All samples were subjected to chemical analysis. Sample preparation was done using a tungsten carbide spray vat, which was cleaned at intervals with soapy water and acetone, to replace conventional cleaning with silica sand, as silica is the main contaminant. An ISO 17025 accredited hematite XRF assay technique was used to determine Fe and the main elements of gangue. Table 2 below shows the description of the method used to analyze the ore products for the tests carried out. Table 2: Analytical method used for ore analysis

Modeling

[031] The PTA results obtained in all DMS tests performed were used to determine the parameters of a standard DMS Weibull model for each of the types of FeSi. The model takes into account the mineral, density and particle size distribution of the feed material to predict DMS performance once the parameters are defined. Correct characterization of the feed material is essential to confidently predict performance. Y = 100(1-exp(-(ln(1/(1-Yp)))*(D/Rhop) ^ (p*d^q))) Equation 1

[032] Also included in the Weibull model are the geometric mean particle size (d) and the mean particle density (D). Sets of values for these parameters were derived for each individual DMS pass performed.

RESULTS Feed characterization Feed size distribution

[033] The samples received were wet sieved separately using a 212 μm Sweco screen. After wet sieving, approximately 64% of BIF (low content) and 55% of the jig feed fines (high content) were in the -1 mm +212 μm fraction. A root 2 size distribution of the feed material was performed in duplicate.

[034] For each sample, a PTA was performed on the feed, as well as on the DMS sinks and floats of each test. The DMS sink and float results were used to construct a reconstituted feed size distribution that could be compared to the measured feed. These results are all reported in Figures 4 and 5 for -1 mm +212 μm, BIF (low grade) and jigue (high grade) feed fines samples, respectively.

[035] In terms of particle size distribution, the duplicate results were reproducible. PTA analysis produced a slightly finer feed size distribution on average (based on reconstructed results). This result is related to the well-known stereological effect created from 2D sectioning of 3D particles for MLA measurements. The total head content of the low-grade sample was 39.7% Fe and that of the jig sample was 48.2% Fe. Detailed results are reported in Appendix A.

Electron Microprobe Analysis (EMPA) of Hematite Particles BIF sample (low content)

[036] The mineralogical evaluation of the BIF sample (low content) indicated that the hematite particles did not contain porosity or microinclusions. Hematite was therefore classified as high-density high-Fe particles and EMPA was therefore not performed on this sample.

Jigue Fines Feed Sample (High Content)

[037] EMPA was performed on the high-grade sample in order to classify the hematite phases correctly and then reliably assign a Fe content and density to each identified phase. Figure 6 shows an SEM image of the high-grade sample. It is worth noting that the number within the grains shows the locations of individual EPMA measurements.

[038] The dark gray grains in the image above were classified as hematite A and consist of high quality clean hematite. The medium gray grains represent micropores containing high quality clean hematite and were classified as hematite B. The presence of micropores resulted in a lower average relative density for this class of hematite than will be realistically experienced in water-based separation. The lower density will not be evident when performing pycnometry, as the particles will be pulverized and the pores removed. The light gray grains have been classified as “hematite” C and are inherently low quality hematites that contain micropores and microinclusions of minerals containing Si, Al and K. These inclusions are so fine (<10 microns) that practical release will not be possible. . It is worth noting that a wide compositional range for this class was observed by EPMA and the Fe content and relative densities attributed to this class are only average. The EMPA data collected is detailed in Appendix B.

Mineral Reference List

[039] After defining the hematite species identified during the EMPA, a reference list of minerals is defined that is used by the MLA to classify the correctly identified particles. Tables 3 and 4 show the mineral reference lists for the BIF (low grade) and jig fines feed (high grade) samples, respectively.

[040] Table 3 shows only one hematite phase (A) being present in the low-grade sample, with a pure hematite content reported as 66.8% Fe at a density of 5.05 g/cm3. Three hematite phases (Table 4) were observed in the high-grade sample with hematite grades ranging from 62.0% to 67.2% and densities between 4.5-5.1 g/cm3. Table 3: Mineral Reference List for Low Grade Sample Table 4: Mineral Reference List for High Grade Sample

Distribution and release of Feed Minerals

[041] The PTA results obtained were used to characterize the feed in terms of size, density and mineral distribution. The release characteristics of hematite and gangue were also defined. Table 5 and Table 6 show the distribution and release of minerals from the low-grade and high-grade sample, respectively. Both the tables and Figure 7 show that 35.7% of hematite A is fully released in the low-grade sample and 60.6% of hematite A is fully released in the high-grade sample. The hematite B (porous) in the high-grade sample is trapped, with only 17.7% being reported as fully released. Hematite C (porous, with microinclusions), although low in content, has a reasonably high proportion of fully released material at 55.2%. Table 5: Distribution and release of minerals from low-grade sample Table 6: Distribution and release of minerals from high-grade sample

[042] Table 6 shows a significant portion of hematite C (26.3%) present in the high-grade sample. This will contribute to a large amount of material of close density to the DMS feed, and will negatively affect the Fe content that can be obtained.

[043] Figure 7 comparatively summarizes the release of hematite for the three identified phases, both for the high-grade and low-grade samples.

Density distribution

[044] Figures 8 to 11 show the density by size distribution of the types of hematite present in the low and high content samples. These figures attempt to visually represent the contribution of each size class of the DMS feed, highlighting the mass percentage and the amount of near-density fractions present.

[045] The density and cumulative grade distributions reported in Table 7 show the best possible theoretical yield grade performance for low and high grade samples. The tables show that under ideal separation conditions: • The low-grade sample, cut at a SG of 4.0, would yield 51.8% product at 64.1% Fe • The high-grade sample, cut at a SG of 4.1 would yield 67.4% product at 64.1% Fe (Note: High Fe content in the 4.9g/ml density class is given as 62.66% Fe. This if due to the fact that the material of this class is dominated by Hematite C (11.6% by mass) trapped by Hematite A (3.5% by mass).

[046] Detailed size and density characterization results are presented in Annex C. Table 7: Feed density distribution for low and high content samples

Dense Media Separation Tests (DMS) Test conditions

[047] DMS tests were carried out in a pilot plant equipped with a 250 mm Multotec DMS cyclone. The -1 mm +212 μm batches prepared from each ore type were passed through the DMS plant, using FeSi Cyclone 60 and Exxaro, as well as a mixture of the two types of FeSi. The mixture was prepared by mixing 30% FeSi Cyclone 60 and 70% Exxaro.

[048] The operational conditions measured for all tests are tabulated in Table 8. The operating pressure values D obtained during the test varied between 11.7 and 13.8. Table 8: DMS operating conditions

Pilot DMS results

[049] The results obtained during pilot DMS tests for low and high content samples are presented.

BIF (low content)

[050] Table 9 below shows the feed for DMS reconstituted from sunk and floated products under different test conditions. The table shows that, under different conditions, the hematite A content in the feed was quite consistent, ranging from 57.2% to 60.7%, with the contribution of bulk gangue being quartz. Table 9: Reconstituted feed for DMS using sunk and float PTA results for low grade sample

[051] Table 10 summarizes the results obtained for testing the low-grade sample. Operating pressure D values varied between 12.8 and 13.8. The table attempts to give an indication of the reliability and reproducibility of the assay results. The Fe content values that are colored compare the head or sink contents tested and chemically recalculated using PTA. Values in parentheses are from directly measured feed samples and are not recalculated. It can be seen that the PTA agreed reasonably well with the measured data. It is worth noting that all reported SiO2 and Al2O3 values are calculated using PTA.

[052] For low and high grade samples, tests done without the presence of sludge are denoted by the suffix N. No clearly relevant trends with the type of FeSi or the relative density of the feed can be easily seen from this view of the results, otherwise: 1. The test without the presence of sludge (Exxaro 3.8N) performed very poorly, showing that the presence of the sludge fraction has a positive (stabilizing) effect on the cyclone's performance. 2. The test at the highest feed density (Mix 4.0) performed very poorly and did not produce high-grade sinkers. Table 10: DMS performance using low-grade sample

Jigue Fines Feed Sample (High Content)

[053] Table 11 below shows the feed for the DMS reconstituted from the sunk and floated products under different conditions for the high-content sample. The results indicate the significant presence of Hematite C, which will affect the quality of the expected product. Compared to the low-grade material, quartz is present as the predominant gangue mineral, while the clay mineral content has increased to about 5%. Table 11: Reconstituted feed for DMS using sunk and float PTA results for high grade sample

[054] Table 12 summarizes the results obtained for testing the high-content sample. The calculated D value was around 11.8 for all tests. The table indicates that the sample's head content was around 48% Fe, measured by chemical assay or via PTA.

[055] The results show that four of the trials calculated higher head contents >50% Fe (PTA and assay), compared to the measured head feed of 48.2% Fe. All quality checks resulted in similar results, with repeated trials reporting elevated calculated head contents. The discrepancy, therefore, cannot be explained.

[056] Similar to the low-grade results, no obviously discernible trends are discernible from this view of the results, other than the clear poor performance of the sludge-free passage again. Table 12: DMS performance using high-grade sample

Modeling DMS Outcomes

[057] A least squares method was used to solve for the model parameters (Yp, Rhop, pq) for each individual run. The following were taken into account for minimizing the sum of squares: • Mass division of sunken and floated DMS • Particle size distribution of sunk and floated DMS • Distribution of Fe content by size class in sunken and floated DMS

[058] The sections below describe how well the model fits the DMS test results for the low and high grade sample.

BIF (low content)

[059] Table 13 shows the model parameters obtained to describe the DMS performance for the low-grade sample. The results show that the actual breakpoints for 1 mm particles obtained during pilot DMS runs on low-grade material for all conditions were much higher than expected for coarse particle DMS operation in all cases. exceeding even the underflow density of DMS.

[060] Table 13: Weibull parameters of DMS and Ep (1 mm particles) for low-grade sample

[061] DMS partitions modeled by size and density are plotted in Figures 12 to 16 for selected assays of the low-grade BIF sample. All modeling was done using the PTA data obtained for the product from each DMS assay. The solid lines in the size and density partition plots represent the PTA data as measured, while the dotted lines represent the best-fit results from the Weibull model. It is worth noting that no stereological corrections were made (size or clearance) during this phase of optimizing the Weibull model parameters.

[062] The results show reasonably good fits between measured and modeled cyclone performance for each of the different operating conditions. In all cases, the DMS sinks showed particle size distributions showing preferential recovery of coarse particles. Much less beneficiation of the finer size classes shown in the grade by size charts, where the fines sunk and float grades approach the fines feed head grade. Jigue Feed Fines Sample (high content)

[063] Table 14 shows the model parameters to describe DMS performance for testing high-grade samples. The same behavior is observed as in the low-grade sample, with the actual cutoffs obtained during pilot DMS runs for all conditions being much higher than the feed density.

[064] Table 14: Weibull parameters of DMS and Ep (1 mm particles) for high-grade sample

[065] DMS partitions modeled by size and density are plotted in Figure 17 to 23 for selected assays of the high-grade sample. Again, reasonably good fits were obtained by the Weibull model, showing preferential particle size distributions of coarse sinkers and preferential beneficiation of coarse hematite.

[066] The density partition plot in Figure 23 for the high-grade sample in Exxaro at 3.8 SG without sludge shows very poor performance with highly flat partition curves for all size classes.

[067] The results obtained showed that each individual DMS test can be very well described using the Weibull model with parameters (Yp, Rho-p, p and q) that apply to the chosen operating conditions. These models can then be used to predictively simulate the response of different samples (mineralogical composition or particle size distribution).

[068] Table 15 and Table 16 show the comparative efficiencies of each of the two types of FeSi, as well as the mixing case. Table 15 uses a 1 mm particle separating at a D50 of 4.50 as the base case and shows how Ep and D50 for finer particle sizes would separate. In all cases, D50 for the finer size classes increases and Ep decreases.

[069] Table 15: Comparative cyclone performance - with 1 mm @D50 of 4.50

[070] Table 16 uses a particle of 1 separating at a D50 of 3.20 as the base case and shows how Ep and D50 for finer sized particles would separate. In comparison to Table 14, it is worth noting that much more efficient separation is evident across all size classes, with very similar Ep values for fine FeSi Exxaro and Cyclone 60. Table 16: Comparative cyclone performance - with 1 mm a D50 from 3.20

Simulations

[071] Three sets of DMS simulations were done keeping high and low feed compositions and size distributions constant, while running a range of DMS feed densities for each type of FeSi. In this way, performance trends can be easily compared: 1. Using an average PTA size and compositions for the low- and high-grade samples, respectively. This represents a finer particle size distribution and the content measured by sieve analysis and XRF melting. 2. Using selected particle size distributions for the low- and high-grade samples. This represents slightly thicker DMS feeds. 3. Higher Hematite A content with lower Hematite C content, resulting in the same Fe content in the DMS feed (high content sample only). This represents the effect of improved feed density distribution for the DMS cyclone.

Low content sample

[072] Figures 24 and 25 show the results comparing the content/recovery profiles for the processing of the low-grade sample using different types of FeSi. Both figures are very similar to the trends produced, resulting in the following findings: 1. At high Fe recoveries in the sunk product, all different types of FeSi produced very similar results. Although Exxaro FeSi is expected to be the most unstable at low FeSi feed densities, it appears that the addition of the sludge fraction helped to stabilize both types of atomized FeSi. 2. Slurry run using FeSi Exxaro resulted in a very low grade product and is not recommended. 3. Unlike traditional grade/recovery profiles, where low recovery is associated with high product grade, all fines DMS curves show a maximum grade. Increasing the DMS feed density beyond this point does not improve the content; on the contrary, both content and recovery are compromised. This is probably due to the rapid increase in viscosity of the FeSi/water mixture with increasing paste density. 4. As expected, Exxaro thin FeSi performs best at high feed densities (low product yield), followed by Blend and finally Cyclone 60. This is likely due to the improved rheology that the gas atomization process imparts to FeSi Exxaro. 5. Simulations using the coarser feed sample, (Figure 24), consistently show that improved grades are obtained for all FeSi types and feed densities. This result is a consequence of better cyclone performance through better Ep values for the coarser size classes.

High content sample

[073] Figures 26 to 28 show the results comparing the content/recovery profiles for the processing of the high-grade sample using different types of FeSi. All three figures are very similar to the trends produced and are in agreement with those obtained for the low-grade sample. Figure 28 shows the predicted DMS performance with increasing Hematite A content and decreasing Hematite C content. As expected, the results indicate further improvement in product content across all operating conditions due to less material of close density with the lowest content of Hematite C. Conclusions • Characterization of the feed ore is essential to understand the quality of the product and the recoveries available, in this case, specifically the effect of porosity and microinclusions. • Results again indicated the sensitivity of DMS to feed size distribution when operating in the <1 mm region. The optimal bottom size fraction for fines DMS must be carefully considered, especially in light of alternative technologies that may be capable of complementing DMS at the finest size fractions. • The tests again showed a deviation between the operational feed density and the cut-off point. Fines DMS achieves a much higher actual cutoff point than expected from the feed density (based on >1mm performance). Much lower feed media densities are therefore required to achieve high density cut-off points. • An increase in feed density does not result in a cleaner product, due to a decrease in separation efficiency at a higher viscosity of the FeSi slurry. • It is recommended to further reduce the feed density (<3.4 g/cm3). Both FeSi Cilone 60 and milled 150D can be evaluated. This can result in even greater product recovery while maintaining Fe content. • At lower feed densities (up to about 3.6 g/cm3), FeSi Ciclone 60 and Exxaro achieve similar results for both tested ores. • The presence of sludge in the FeSi medium proved to be essential for obtaining efficient separation. • DMS Fines model parameters have been successfully established and predictive modeling can be considered to simulate: • The effect of a variety of feed size distributions on performance, • The effect of a variety of mineralogical compositions on performance, • Two stages of DMS can be simulated, such as rougher and cleaner, • Performance of any new ore, characterized by PTA, and • Conditions required for any product requirement. Table 17: Size distribution of low-grade reconstituted feed using PTA Table 18: Low Grade Feed Size Distribution Using Sieving Table 19: Distribution of high content reconstituted feed using PTA Table 20: High Grade Feed Size Distribution Using Sieving Table 21: Chemical analysis for DMS products (low and high content samples) Table 22: Summary of DMS tests for low and high content samples

[074] The method according to the present invention can be further described with reference to Figures 31 and 32, as follows.

PROCESS DESCRIPTION Food Preparation

[075] The feed consists of -1 mm crushed iron ore in one tonne bulk bags fed into a hopper using a jib crane. The material is removed from the hopper onto a conveyor belt that discharges the material into a mixing box from the feed preparation screen, where the material is pulped again before sieving to 400 μm.

Separation in Dense Medium

[076] The oversized feed preparation screen is routed into a mixing box, where it is mixed with the correct medium. The ore and media mixture is pumped into a 150mm dense media cyclone, where the feed pressure is controlled using a variable speed drive installed on the cyclone feed pump.

[077] The underflow from the dense (sunk) medium cyclone (product) gravitates to the magnetic product separator through a mixing box where it is diluted with effluent water. The effluent and non-magnetic from the magnetic product separator reports to the product dewatering/rinsing screen equipped with 212 μm polyurethane panels. The oversized product from the dewatering/rinsing screen is collected in one ton bulk bags for removal while the undersized screen is routed to the effluent tank.

[078] The overflow of the dense (floated) medium cyclone (disposal) gravitates into a residual medium drainage screen equipped with 212 μm polyurethane panels. The undersized waste media drainage screen is sent to the tubular densifier feed tank. The oversized waste media drain screen (2.1m by 1.08m high frequency screen) is repulped with effluent water before feeding to the waste magnetic separator for FeSi recovery. The effluent and non-magnetic (disposal) from the magnetic separator are routed to the waste dewatering/rinsing screen to recover excess waste water before disposal. The undersized waste dewatering screen is sent to the effluent tank.

FeSi recovery circuit

[079] The dense medium cyclone underflow is suitably diluted with effluent water before the magnetic product separator (2000 Gauss CRM 60:60) for FeSi recovery. However, the dense media cyclone overflow is routed to a waste media drain screen before feeding to the magnetic waste separator (2000 Gauss CRM 60:100) for FeSi recovery. The overdense (magnetic) from both the product and residual flow magnetic separators is routed to the circulating media tank via a demagnetizing coil. The underflow from the magnetic separator of product containing effluent and product is routed to a 2.9 m by 1.08 m high frequency dewatering screen.

[080] The magnetic waste separator recovers FeSi from the oversized waste media drainage screen. The effluent and residual material from the magnetic waste separator are sent to the waste dewatering screen to recover excess water before waste disposal. The effluent water is routed to the effluent tank and is used as dilution water for magnetic product separators and disposal. A portion of the effluent water is pumped to the effluent magnetic separator to eliminate any FeSi that may be in the effluent. The underflow of the magnetic effluent separator is used for repulping and diluting the fresh feed prior to the feed preparation screen.

Circulating Media and Densification Circuit

[081] A portion of the correct medium from the collection box is fed to the circulating medium tank (3 m3) together with the overdenses from the magnetic separators, densifying cyclone underflow, tube densification underflow and product medium drainage screen underflow . The tank is equipped with an agitator to keep the FeSi in suspension. The circulating medium tank is equipped with 2 pumps, namely the circulating medium pump and the primary condenser pump. The circulating media pump discharges the correct media into the media collection box. A portion of the media is distributed to the primary feed mixing box, while a portion of the media is recycled back to the circulating media tank. The primary densifying cyclone pump feeds the densifying cyclone (150 mm in diameter). The underflow from the primary densifier gravitates back to the circulating medium while the overflow gravitates to the secondary tubular densifier feed tank. The overflow from the tubular densifier is sent to the magnetic waste separator for FeSi recovery.

Density Control System

[082] Density control of the circulating medium is achieved using a DebTech density controller equipped with a density meter mounted on the correct medium delivery tube. The density controller is connected to an automatic valve for dilution water to the circulating medium tank.

Product Handling/Disposal

[083] The dehydrated and discarded iron ore product from the dehydration screens will be unloaded in one-ton bags and removed using pallet jacks and, occasionally, using a forklift. The filtered sludge will also be removed using a forklift.

Sludge Handling

[084] The undersized feed preparation screen gravitates to a desliming tank from where it is pumped to a geotube for dehydration. Dewatered solids along with a geotube will be removed using a forklift for storage and further processing if necessary. The filtrate from the geotube will gravitate to the well in the geotube area, from where it is pumped to the clarified water tank.

Dense Media Separation (DMS) Pilot Plant Work

[085] In this work product, the following contractions have the following meanings: CM - Correct Medium (ferrosilicon mixed with water in the proportions necessary for Separation into Dense Medium) KGT - Conglomeratic Type Ore PE - Slab Type Ore BIF - Ore of the Banded Iron Formation Type ATS - Anglo Technical Solutions, Minas Crown

-1mm DMS Circuit with 63.5% and 55% Fe

[086] Pilot DMS was carried out using FeSi Exxaro Fino NGA at operating densities ranging from 3.2 to 4.0 t/m3. The ideal operating pressure (cyclic operating pressure) of the DMS cyclone was found to be 15D. During commissioning, different operating pressures and FeSi types were tested until the plant operated stably with optimal fines differentials between 25% and 6%.

Shale Mass Balance Results

[087] One of the objectives of the DMS plant was to produce a fine product with 63.5% Fe in line with Sishen's current fine product specification. For -1 +400 μm, the DMS plant achieved fine product specification at an operating density of 3.4 with 54.5% yield as per Table 23. Table 23: DMS Performance in Shale (-1 +400 μm and -1 +200 μm)

[088] The DMS cyclone was also fed shale samples with a bottom size of 200 μm to determine the impact of size range, since the rupture size for the 150 mm diameter cyclone is about 200μm. Changing from a bottom size of 400 μm to a bottom size of 200 μm, thus feeding the plant with -1 +400 μm, results in 29% more material in the shale feed. As expected, the yield obtained with -1 +200 μm drops for the same operating conditions compared to the feed at -1 +400 μm due to poor fines separation. The impact worsens as the operating density increases, which is effectively the effect of viscosity on the fines.

[089] Pilot work in Shale shows that the -1 +400 μm fraction can produce specification fine product qualities with relatively low operating densities of 3.4 t/m3. DMS separation efficiencies of -1 mm are negatively affected by reducing the feed bottom size from 400 μm to 200 μm and makes it difficult to obtain a fine product of low grade Shale specification.

BIF Mass Balance Results

[090] Similar to shale, the DMS plant should produce, from BIF, a fine product with 63.5% Fe in line with Sishen's current fine product specification, as well as 55% Fe for flotation. For -1 +400 μm, the DMS plant achieved fine product with 63% Fe, while operating at a density of 3.6 t/m3 with 44% yield, as shown in Table 24 Table 24: DMS performance in BIF ( -1 +400 μm and -1 μm +200 μm)

[091] Feeding the DMS plant with BIF with a bottom size of 200 μm effectively brings 27% more material into the feed. The yield impact as a result of the fund size reduction for BIF was not obvious.

[092] Pilot work at BIF shows that the -1 +400 μm fraction can produce specification fine product qualities at relatively low operating densities of 3.2 t/m3 for mixing within the remainder of the fine product.

KGT Mass Balance Results

[093] For KGT, the DMS plant was operated to produce a fine product with 63.5% Fe in line with Sishen's current product specification, as well as 55% Fe for flotation. From -1 +400 μm, the DMS plant achieved fine product with 63.8% Fe, while operating at a density of 3.6 t/m3, which resulted in 47% yield, as shown in Table 25. Table 25: DMS Performance in KGT (-1 +400 μm)

[094] The background size effect was not realized for KGT, as the scope addition was included after the KGT material was processed in the DMS plant.

PE Mass Balance Results

[095] For PE, the DMS plant was operated to produce a fine product with 63.5% Fe in line with Sishen's current product specification, as well as 55% Fe for flotation. From -1 +400 μm, the DMS plant could only produce 60% Fe while operating at a density of 3.6 t/m3, which resulted in 40% conformal yield. Further increasing operating density to 3.8 t/m3 did not help improve product quality, but significantly reduced yield by almost 20% points. In this case, the significant drop in yield is a result of the poor release characteristics of PE as well as the increase in viscosity at higher operating densities. Table 26: DMS performance on PE (-1 +400 μm)

[096] The impact of bottom size on PE has not been further studied, as the plant has already encountered difficulty in making flotation feed with a narrower feed particle size distribution.

Jigue Gravel Mass Balance Results (-1 +400 μm)

[097] As part of the development of the -1 mm low grade beneficiation technology, the current -1 mm jig was processed through the pilot DMS plant to evaluate the potential for producing fine products of specification that the module of Current testing Reflux Classifier at Sishen struggles to catch up for several reasons. The received -1mm jig was delimed to 200μm and processed as -1mm +200μm as well as -1mm +400μm, the results of which are summarized in Table 27. Table 27: DMS Performance on Jigue Gravel (-1mm +400μm and -1 mm +200μm)

[098] The -1 mm pilot DMS plant was able to achieve 64.0% Fe and an average operating density above 3.4 t/m3 for -1 +400 μm gravel, as shown in Table 27. For -1 +200 µm gravel, reducing the DMS cyclone feed bottom size from 400 µm to 200 µm has a negative impact on yield, as would be expected due to poor recovery of material near the breaking size . Also notable is the significant drop in product yield from operating densities of 3.6 t/m3 and, further, to 3.8 t/m3, as illustrated in Figure 33.

[099] The drop in yield and product content when the operating density is increased to 3.8 can be attributed to the increase in average viscosity, which is significant for fines. Figure 34 shows the media rheology for fine FeSi atomized with Exxaro nitrogen gas used in the pilot DMS plant.

[0100] From the operating density of 3.8 t/m3, the viscosity of the medium begins to increase exponentially, which would lead to poor selectivity, since the external centrifugal forces on the dense fine particles are no longer sufficient to overcome the hydrodynamic drag of the viscous medium.

DMS Performance (-1mm +400μm) in Relation to Ore Potential

[0101] Mineralogical analysis on the -1 mm +400 μm feeding the pilot DMS plant was performed to determine the theoretical grade yield potential for each ore type, as illustrated in Figure 35. The pilot DMS plant operated with efficiencies reasonable separation rates, as measured by how closely the pilot grade's yield performance approached the grade's theoretical yield curve. Scott Napier Munn's model is additionally applied to evaluate actual Ep (separation efficiency) and shear densities in the following sections.

[0102] The pilot DMS plant at the relative ore potential performance of -1 mm +400 μm is similar to what was achieved on the Coarse Reflux classifier with the largest average difference between the actual and theoretical grade yield curve at 15% . The metallurgical performance of the -1 mm DMS plant and the -1 mm Reflux classifier were found to be relatively similar. The choice of processing technology to be applied in the -1 mm flowchart will depend on the results of a trade-off study considering Capex, Opex and operability, among other elements.

DMS Separation Efficiency Curves (-1 mm +400 μm)

[0103] The same approach was followed to model the performance of the pilot DMS plant at the -1 mm +400 μm size fraction. DMS cyclone feed, product, and waste stream samples from selected passes were subjected to mineralogical analysis using an MLA to determine density distribution by size.

[0104] The pilot run selected to model the -1 mm +400 μm BIF DMS cyclone separation efficiency, as shown in Figure 36, came from the performance with which the pilot plant achieved 43% mass yield with Fe content of 60%. From the partition curves, it was possible to determine that the effective cutting density achieved was 4.30 t/m3 with an overall Ep of 0.38.

[0105] The pilot run selected to model the -1 mm +400 μm Shale DMS cyclone separation efficiency, as shown in Figure 36, came from the performance with which the pilot plant achieved 55% mass yield with Fe content of 63%. From the partition curves, it was possible to determine that the effective cutting density achieved was 4.52 t/m3 with an overall Ep of 0.40.

[0106] The pilot run selected to model the -1 mm +400 μm KGT DMS cyclone separation efficiency, as shown in Figure 36, came from the performance with which the pilot plant achieved 48% mass yield with Fe content of 63%. From the partition curves, it was possible to determine that the effective cutting density achieved was 4.40 t/m3 with an overall Ep of 0.45.

[0107] For PE, the pilot run selected to model the -1 mm +400 μm DMS cyclone separation efficiency, as shown in Figure 36, came from the performance with which the pilot plant achieved 41% mass yield with Fe content of 60.2%. From the partition curves, it was possible to determine that the effective cutting density achieved was 4.65 t/m3 with an overall Ep of 0.28.

Claims (6)

1. Solids separation method, the method characterized by the fact that it comprises: i) adding the solids to a suspension of particulate matter comprising magnetic particles in a liquid to create a mixture, ii) placing the mixture in a separation vessel of so that rotation is transmitted to the mixture around a space enclosed by an outer wall of the vessel to transmit a centrifugal force to the solids; iii) the mixture is separated into floating and sinking fractions; wherein said particulate material is ferrosilicon having a D90 particle size of less than 200 μm, the solids are iron ore having a particle size of less than 1 mm and greater than 200 microns, the feed density of the mixture is less than 3, 8 g/cm3, the operating pressure of the mixture is 10 to 15D inclusive, the cut-off point differential is between 0.1 and 0.6 of the feed density, and where the mixture includes a fraction of sludge iron ore less than 10% by weight having a particle size of less than 45 μm. 2. Solids separation method, according to claim 1, characterized by the fact that the floating and sinking fractions are subjected to magnetic separation before washing to recover the particulate material. 3. Method, according to claim 1 or 2, characterized by the fact that the floated and sunk fractions are passed through separation screens vibrating at a frequency greater than 50 Hz. 4. Method according to claim 3, characterized by the fact that the separation screens have a double oscillation. 5. Method according to any one of claims 1 to 4, characterized by the fact that the solid iron ore has a particle size of less than 1 mm and greater than 212 microns. 6. Method according to any one of claims 1 to 5, characterized by the fact that the solid iron ore has a particle size of less than 1 mm and greater than 400 microns.