CHARACTERISATION, TESTING AND EVALUATION OF MINERAL RESOURCES

CHARACTERISATION, TESTING AND EVALUATION OF MINERAL RESOURCES


B P Ravi & A.Majumdar, Indian Bureau of Mines, 29, Gorguntepalya, Bangalore 560022, rodlbng@vsnl.net.


INTRODUCTION


The materials that support living beings are derived from earth’s crust. Mineral is defined as natural mostly inorganic compound with definite range of chemical composition with diverse distinct properties for diverse utilitarian purposes of our civilization. The various elements that constitute our crust are unevenly distributed. The geological processes some time lead to concentration of minerals to form a resource in a particular area. The mineral resource is termed as an ore body when it can be economically exploited for utilization. Also ore from such ore body cannot be used directly and needs a preparation and conversion. Mineral processing involves size reduction, liberation and separation based on differential physic-chemical properties yielding different utilitarian products with maximum purity, recovery at minimum cost and environmental disturbance for sustainable development. The aim of this paper is to describe briefly the characterization—processing of minerals and economic evaluation for utilization of mineral resources. The scope of the paper comprises of 1] Brief description of genesis of ore deposits, properties, uses and specifications of minerals. 2] Essential data needed for mineral related enterprise.3]] Diagnostic process characterization-GTM 4] Preliminary testing 5] Bench scale testing for flow sheet development. 6] Confirmatory pilot scale- scaled down simulatory bench scale tests for design. 7] Equipment sizing -PTFR 8] Plant auditing studies.


BRIEF DESCRIPTION OF GENESIS OF ORE DEPOSITS, MINING METHODS AND UTILITARIAN SPECIFICATIONS OF MINERALS


Today no mineral that is exploited from earth’s crust is suitable for direct use. It needs processing based on diverse properties of different phases/minerals associated in the ore. Mineral processing professional acts as a bridge between exploring geologists of a resource, excavating mining professional of a mine from an ore deposit and utilizing value adding enterprising materials engineer. Hence mineral processing professional should have a sound knowledge of characteristics - uses of the minerals for separating them besides some knowledge of geology, mining and subsequent use of the product. The ore forming process [ore genesis] leading to formation of ore deposits with typical examples are enumerated below. A] The magmatic segregations deposits are a result of direct crystallization, segregating and differentiating from igneous melt magma. Typical examples are of chromite, copper, ilemenite, nickel and platinum deposits like the layered chromite deposits of chromite around Jojohatu of Jharkhand and the diamond injection pipe deposits of Panna. B] The residual highly volatile, silicic liquid after magmatic segregation, cools, consolidates results in formation of igneous pegmataties deposits. On the contrary, the metamorphic pegmatite results from result from the concentration of mobile constituents of the melt during the metamorphic differentiation. Typical examples are deposits of mica, beryl, uranium and rare earths like Nellore mica belt of Andhra Pradesh C] an Igneous metamorphic deposit is a result of host rock alteration due to intrusion of igneous bodies. Scorn deposits of Wollastonite, serpentine are typical examples as typified by calcareous wollastonite deposits of Rajasthan. D] Hydrothermal deposits result from hot aqueous fluids of magmatic affiliation migrating long distance from the source and formed either by depositing in open space [Cavity filling] or by chemical interaction with host rock [Replacement]. Typical hydrothermal deposits are gold reef -2- at KGF, Karnataka and lead – zinc ore deposits at Zawar in Rajasthan. D] Sedimentary deposits are formed due to liberation, transportation and deposition due to mechanic-chemical weathering. The beach sand placers deposits of titanium and uranium of Indian peninsula is typical example of mechanical sedimentary deposit. Similarly the BHQ and BMQ deposits Singbhum, Jharkhand and Salem, Tamilnadu and many iron- manganese deposits are examples of mechanic- chemical Sedimentary deposits. E] Deposits may be formed by weathering of rock by concentration and effective leaching- transportation of specific constituents. Most of Indian bauxites are formed due to weathering of rich aluminous rocks like granite-synenite and leaching of silica under tropical weathering with effective drainage. Weathering of sulphide deposists result in oxide-hydroxide [mostly iron] capping called gossans. The extent is limited to water table, weathering rate, period and nature of rock. Gossans aid in exploration of primary Sulphide deposits of economic significance. Typical examples are oxidized capping of Ajjanhalli gold deposit Karnatka, Gossans of Malnjkhand copper deposit and oxide capping of Rampura- Agucha lead- zinc deposit. F] Supergene sulphide deposits are formed by precipitation of solvents from weathering of gossans under low Eh and reducing conditions. Typical examples are covellite-bornite mineralization zone below oxide-carbonate zone in case of Malanjkhand copper deposit. G] Metamorphic deposits are formed either due to metamorphism of existing ore deposits or due to formation of new minerals due to metamorphism of rock types. The manganese deposits of Karnataka, Maharashtra and Madyapradesh and phosphorite deposits of Jhamakotra Rajasthan are examples of metamorphism of earlier sedimentary deposits. The graphite deposit at Sivaganga, Tamilnadu is an example of metamorphic rock deposit. The four types of mining methods are opencast, underground, alluvial and solution. A] Open cast method is the most economical easy mining method for large capacity of hard rocks. This method requires the ore body close to the surface to be economical and is becoming more common with decrease in tenor of ore and increase in high capacity projects B] Underground mining is for very deep hard rock ore bodies It is expensive and is viable for high revenue yielding deposits like precious metal and base metal ore bodies. A combination of opencast followed by underground mining may be used for same ore body to enhance profitability. C]Alluvial mining uses dredges or hydraulic water to mine unconsolidated rocks and is relatively inexpensive D] Solution mining is a method of extracting the mineral from a fractured ore body by pumping in hot lechants. Examples are hot sulphur mining. Though the different properties of minerals, occurrence and uses are important, it is not mentioned here as they are available in mineral processing books like B Wills et al, J Kelley et al, and SME Mineral processing hand book. Fig 1 shows the size limitations of the industrial concentration processes. -3- 3 ESSENTIAL DATA NEEDED FOR A MINERAL PROCESSING VENTURE The essential data needed for a mineral processing enterprise job consists of location, deposit, mining, market specification with trend, literature review and samples. 3.1 Location of deposit The geographical location of deposit with details of topography, climate, water resources and infrastructure facilities and local guidelines on environment is essential. As most of the mineral processing needs water the quality and quantity of water available dictates the capacity of processing plant and nature of process. Paucity of water and availability of natural gas cheaply at Palmeria desert Syria led to development of sustainable dry process comprising of calcination, differential grinding, screening, stoichiometric hydration, air classification yielding fertilizer grade[+30%P2O5]concentrates with 87% P2O5 recovery at 67 wt% yield from calcareous phosphorite dumps of 20%P2O5 as compared to conventional wet differential flotation process. The climatic condition like snowfall-rainy season affects the capacity of plant. Typical examples are high capacity iron ore processing plants in Goa region take care of the difficult mining – transportation period during monsoon period as well as high capacity gold concentration plants in eco-sensitive regions Alaska, Kerala and river placer deposits of Mongolia and CIS. The remoteness and company philosophy affects the choice of equipment. High capacity plants with automated process control are practiced in Western countries and in HZL as compared to under developed countries and small capacity plants [private graphite flotation plants in remote parts of Orissa and AP] The hydraulic transport of concentrate slurry is practiced in remote forest areas to end use plants as in case of KIOCL [Kudremukh to Mangalore port] and ESSAR [ Bailadila to Vizag port] 3.2 Deposit The mineral processing test engineer tends to narrow down the scope of testing for the element the sponsor indicates. A detailed geology of the resource with detailed mineralogy along with possible ore/rock types and possible metallogeny will enable the mineral processing test engineer to look into possible generation of co-products and for achieving a near nil waste technology for overall sustainable development. Typical cases are recovery of by-product scheetlite from gold ores of KGF and HGML. Production of wollastonite and calcite concentrates from calcareous wollostonite deposit of Rajasthan. Similarly efforts are under way to produce marketable low-phos dolomite by-product besides producing SSP grade rock phosphate concentrate from dolomitic phosphorites of Rajasthan. Production of silica sand and different grades of clay from kaolin resources of Kerala. The change in nature of secondary supergene ore to primary ore as in case of sulphide gold- copper deposits necessitate the change in simple leach process to S’ flotation followed by extraction as in case of Malnjkhand copper and Ajjanahalli gold deposits. Similarly change from clayey goethitic hematite ores to coarse grained quartz associated matitized hematite iron ores tend to produce pellet grade concentrates easily with high recoveries and throughput in the existing classification-gravity-WHIMS plant. -4- 3.3 Mining It is essential for test engineers to know the proposed mining rates, methods, sequences, ore types and mine equipment sizes. The data will help in knowing the plant feed characteristics for preparing for suitable typical plant feed or adjusting the plant set points for ore variability. Since media for autogenous/pebble mill comes from mine/coarse crushing plant, care has to be taken for storing ore types separately for subsequent use and for blending. The underground mine may have to be backfilled and backfill plant tails sand properties has to be evaluated for utilization. The S’flotation plant tails treating copper ore associated with quartz-granite rocks could be used for UG mine stope back-fill as compared to ores associated with carbonate-phyllite rocks. 3.4 Market specifications It is axiomatic that buyers are quality conscious leading to encounter of regulatory – quality control restrictions by the mineral processing plants. The test engineer must be aware of current and probable futuristic specifications especially on toxic-undesirable impurities. The change in specifications due to developments in consumer industry shall also be noted while developing the flow-sheet. The market trends shall be noted to make the project flexible by provision for cutting the costs during dull period and for enhanced production during the boom period. The seasonal demand may require for storing the product for selling during the demand period as in case of fertilizer minerals like potash and rock phosphate. The climatic season’s monsoon season and snow period may force concentrates to be stored for long periods. This need to take care in tackling time-degradation of product as in case of pyropheric sulphide – washed coal concentrates. The change pellet grade requirement from export BF grade [+65% Fe] to local sponge grade[+62% Fe] led to enhancement of concentrate yield from 50% to 70% rendering the project sustainable ecologically and economically for siliceous sub-grade iron ores from Bellary. 3.5 Literature review and practice Fortunately mining-metallurgical industries are open minded regarding the exchange of ideas. Time spent on review of literature and good communication – exchange of information on practice is always beneficial. It is a good practice to study industrial practice taking advantage of experience, modernization trends by arranging visits to such facilities along with a team of investor, design test engineer, equipment manufacturer and project leader with his team. Typical examples are use of HPRG ahead of clinker grinding in cement plants was successfully translated in wet iron ore concentrate grinding in KIOCL and utilizing excess concentrator-dewatering capacity of dolomite rock phosphate concentrator by inserting HPRG ahead of ball mill grinding. The above implementation has saved considerable capital and operating costs. This seems to be one of the alternatives for grinding high WI iron ore concentrates to pellet feed sizes to reduce the operating cost for siliceous iron ore concentrators producing pellet grade concentrates. Similarly import of cheap mineral processing equipment from China had opened options for small-medium mine owners to establish plants with low capital and gestation period. -5- 3.6 Samples One will be confident of the results on a large variety of samples than a limited test on a large quantity of limited sample types. It is also recommended to get the more variety of samples are tested by different laboratories [ in house R&D, equipment manufacturers, local, academic, national, international/overseas ]. At various stages of mineral project development cycle, the test engineer may be provided with surface hand samples. Split cores, trench/pit samples, channel samples, stockpile samples. As mine develops bulk lots for confirmative bench scale tests , developmental test work in pilot scale for confirmation, market survey product generation and equipment selection shall be made available. Two essential things regarding the samples are they should be representative [Gy’s formula] and should be fresh. The samples should be sent in air tight plastic containers and may have to be refrigerated in case of samples that degrade. The implementation of good practical sampling and testing technique never diminishes as it helps in scientifically opening and closing a venture on a mineral resource. Establishment of a scheelite processing plant at KGF based on test work on samples drawn from periphery wall of tailing dam and subsequent treatment of slimy feed from tailing pond led to inferior results due to change in granulometry of feed. Similarly the metallurgical recovery of DML grade tungsten concentrates was 40, 50 and 60% for ores from selectively mined veins[0.2%W], mined dumps from selective veins mining [0.1W %] and mechanized bulk mining of granite rock with veins [0.05%W]. The latter reduces the overall operating project cost. Annexure E gives sampling data. Detailed test work on different lead-zinc ores comprising of easily milling carbonate- quartz lode ores, ores associated with quartz, ores associated with mica-schists and ores associated with graphite-mica schists and ores associated with labile pyrite, copper bearing graphite mica schists indicated the increase in tendency of flotation process refractoriness. Employing iron/copper- graphite depressants/ graphite prefloat or use of centrifugal slimy gravity concentrators like MGS for lead and zinc concentrates was developed for refractory labile pyritic copper bearing graphite mica schist Pb-Zn ores. Alternatively such dirty concentrates may have to be processed either in Imperial smelter or Isa smelter. 4 CHARACTERIZATION STUDIES The characterization of minerals for mineral processing consists of 1] Physical characterization 2] Compositional characterization 3] Mineralogical characterization 4] Diagnostic process characterization 4.1 Physical characterization The physical characterization consists of size, shape, density measurements. The size analysis of a mineral sample is normally carried out by wet and dry sieving the sample from screens/ sieves with aperture from 200mm to 0.025mm. The cumulative wt% passing is plotted against log aperture size of passing screen. The size analysis indicate % of lumps, chips, sands, and slime in the sample and distribution of mineral values with size. Size characteristic indicate the amenability of sample to sizing. The sub-sieve analysis is carried out on screen fraction finer than 50 microns and is normally based on calculating time required for settling solids for known height by Stoke’s equation and is carried out on dilute and dispersed pulps. The supernatant water after stipulated settling time is siphoned off and the process is repeated several times till clear supernatant water is obtained. The ratio of mass of residue sediment to the feed sample indicates the cumulative weight % over size fraction at settling particle size in Stoke’s equation. -6- Even cyclosizing or by LASER scattering-extinction based [Malvern] particle size analyzer are used for size analysis The size data may also be obtained by indirect methods by determing specific surface by Blaine air permeability method or by BET adsorption method. The shape of powder particles are determined by microscopes by measuring the aspect ratio. The specific gravity of lumps is determined by Jolly’s walk yard balance and that of powder by Pycnometer [Sp Gravity] bottle method. The bulk density of sample is determined by weighing the loosely packed beds of sample in unit volume containers. The aggregates are subjected to packed bed density measurement to get Proctor’s bulk density. 4.2 Chemical characterization The chemical characterization is also called compositional characterization. A representative portion of sample of appropriate fineness of known weight is decomposed either by acid attack [HCl, aquaregia, 3 HCl + HNO3, HClO3] for digesting carbonates, oxides, sulphides and silicates respectively in PTFE/TEFLON beaker bomb at 1150C , 110 Atm pressure or fusion-extraction with alkali salts [ (Na/K)2 CO3, (NaOH+Na2O2),K2S2O7] at 1000 0 C in nickel crucibles. If the element content is high it may be estimated by gravity method [Silica –volatilization loss by HF, R2O3 group as hydroxide precipitation with ammonium hydroxide, Ca as insoluble oxalate, Mg and Mn as insoluble phosphates and Ba as barium sulphate]. Gravimetric method is normally used to determine LOI, H2O and CO2. Now a days Ca, Mg, Fe and Al are determined by titrating against EDTA sodium salt with suitable indicator and suitable masking agent. Fe titrimetric analysis the solution is titrated against potassium di- chromate with DPA indicator. When a monochromatic light of selected wave length is passed through the sample solution with specific ligand chromospheres forming specific coloured complex, the absorption of intensity of incident light is proportional to concentration c [ Beer-Lambert’s law- A= abc] . Cu, Cr, Ti, Al,Ni, W and V are determined by spectrophotometric methods. When resonance light of intensity Io is passed through cell with ground state atoms, part of light is absorbed by atoms in flame [air- acetylene, nitrous oxide- acetylene]. The N2O-C2H2 flame is used for refractory elements like Al, Si, Cr and Ti. Then Initial intensity Io decreases to intensity I proportionately to concentration of ground state atoms. Abosorbance A = log[Io/I] = abc. The equipment is calibrated and subsequently the elements concentration as low as ppm level is determined by AAS. The inductively coupled plasma [argon flame 100000C] and atomic emission spectroscopy is employed for estimation of more elements up to ppb levels. The decomposition and estimation method depends on mineralogy and its assemblages. Estimation of Fe from simple iron oxide ores with quartz gangue, iron oxide ores with aluminous gangue, iron oxide ores with titaniferrous gangue and iron silicates with refractory spinel silicates are decomposed by simple acid attack to fusion-extraction and strong acid [AR- HClO] in bomb digesters. The chemistry of sample indicate trace impurities to be taken care during test work, trace valuables to be considered for recovery as co-product 4.3 Mineralogical characterization The mineralogical characterization is carried out mostly by microscopic studies. The transparent minerals are studied transmitted light microscope by studying thin sections [0.03mm]. The transparent minerals are identified by colour, shape, form cleavage pleochroism, refractive index, isotropism under polarized light, birefringence, extinction angle under crossed Nicols. The transparent minerals in powder mount slides are determined by Cargille refractive liquids. -7- The opaque minerals are identified under reflected microscope after mounting in epoxy resins and polishing them with emery/diamond by colour, reflectivity, bireflection, form. Cleavage and micro hardness. The quantitative estimation is done either by visual for rough estimation, by grain counting under cross wire while moving sections, by point counting by automatic point counter at fixed set intervals which is realistic and gross counting for minor/trace minerals by counting no of minerals under view and number of trace minerals in such field views. Liberation studies on size fractions indicating shape [anhedral/ euhedral/ prismatic/ flaky], the grain size [fine /medium /coarse, equiangular /unequigranular], locking [intergrowth] (lamellar/simple/motted/ graphic/ disseminated/concentric/vein/network/coated) &inclusions. The grinding requirement due to interlocking, excess slime content and lack of selectivity may have to be carefully attributed and cannot be used as an excuse for poor separation. 4.4 Diagnostic process characterization The DPC determines physical, chemical and mineralogical values in an sample indicating ore types, quality, yield of products, generic process, trace valuables, impurities associated with it based on amenability of sample to size, specific gravity, susceptibility[magnetic], surface properties, sequential selective solubility/smelting. 4,4,1 Amenability to size [SCC] : The as received sample is subjected to sieve and sub-sieve analysis and the fractions are subjected to chemical and mineralogical studies. A low grade clayey ferruginous bauxite sample 40.60%Al2O3, 4.7% reactive SiO2 was scrubbed and subjected to sizing from 50mm to 0.025mm, The above diagnostic amenability sizing yielded a coarse fraction [+0.1mm] assaying 46.5% Al2O3,1.49%SiO2{Reactive}with77% Al2O3 recovery at 67 wt% yield meeting the specifications of reactive SiO2content of the party. 4.4.2 Amenability to specific gravity[GCC]: Sp. Gr. Conc. Criteria = [Dh – l]/[Dl –l] where Dh,Dl and l are specific gravities of heavy, light mineral and medium respectively. The sized fractions of as received sample is subjected to heavy liquid using a mix of TBE [3SpGr] and acetone[0.0 Sp Gr] from 2.59,2.72,2.76,2.81,2,87,and 3] The Specific gravity curve is drawn by plotting *** Wt% Sinks against Specific gravity. Assay curve is obtained by plotting ***. Assay % against *** Wt% Sink. The characteristic curve [Assay against mean CumWt% sink] and tolerance curve [Diffrence *** Wt% sink{0.1 SG} – S G] are also drawn. The reduced specific gravity curve gives tromp curve and value of EPM [{y75-y25}/2] indicate ease of separation and characteristics of sample to gravity concentration. A siliceous manganese ore assaying 41.3% MnO2 when subjected to HLS at 2.6 specific gravity yielded a sink fraction assaying 67% MnO2 at 49.5 wt% yield indicating the sample is amenable to gravity concentration. 4.4.3 Amenability to susceptibility (magnetic) [MCC]: Magnetic susceptibility concentration criteria = [Sm-Sa]/ [Snm-Sa] where Sm, Snm and SA are magnetic susceptibility of magnetic, non-magnetic and medium respectively. The sized fraction of sample assaying 37.2% Fe is subjected to dry magnetic separation from 0.3, 0.69, 1.09 1.66 Tesla and the magnetic fractions are assayed. The cumulative Assay% is plotted against the intensity T. A magnetic fraction at 1.09 T yielded a concentrate assaying 64.3% Fe with 86.2%Fe indicating the amenability of siliceous sample to magnetic separation. -8- 4.4.4 Amenability to selective flotation [FCC]: Micro flotation tests are carried out on uniform sized 1-5 gm mineral sample is reagentized under controlled conditions in a beaker. The reagentized pulp is added to Halimond tube with Teflon coated magnetic stirrer to keep the particles in suspension. The gas is allowed to flow at predetermined rate and pressure and stirring – gas flow rate is controlled by solenoid valves. The hydrophobic particles rise and fall into vertical stem or attached top part of cell while non-float is retained in fritted glass disk bottom. The products are collected weighed and assayed. Amenability of sample to flotation FCC may be obtained this Micro flotation 4.4.5 Amenability to selective chemical extraction [DLT]; DLT is a sequence of leaching/chemical extraction test from less stable phase to intensified chemical extraction for more stable phase. DLT followed by extraction of values is attributed to destroyed phase. It is based on mineralogy and selective destruction of mineral for releasing the values. The following table indicates DLT of gold ores.

Stage Extraction Mineral
NaCN wash
Precipitated free Au
NaCN leach NaCN +CIP Free milling Au
Hot Na2CO3 NaCN +CIP Gypsum
Hot NaOH NaCN +CIP Bauxite and clay
Hot Oxalic acid NaCN +CIP Oxide coating
Hot Acetic acid NaCN +CIP Carbonates
Hot HCl NaCN +CIP Ironoxides,pyrrhotite,unstable galena
Hot H2SO4 NaCN +CIP Labile sulphides
Hot Fe Cl3 NaCN +CIP Base metal sulphides
Hot HNO3 NaCN +CIP Arsenides and sulphides
Acetonitrile NaCN +CIP Organic carbon
Dead Roast NaCN +CIP Carbon
Hot HF NaCN +CIP Silicates
The DLT on GRHalli gold ore indicated 35%,20%, 25%.10% and 10% of extractable Au is attributed to Free milling, oxides-carbonates,sulphides-arsenides, carbon and encapsulated in silicates The DLT indicated that GRHalli ore was carbonaceous silver bearing slulphidic refractory ore and needs mineral lattice breakdown for enhancing Au extraction. 4.4.6 Geological-Mining-Techno-economic mapping of resource [GTM]; a map of the resource indicating the geology, mineralogy, chemistry, ore type, amenability of sample at the point in resource and expected yield/revenue/profit along with mining – processing scheme furnishes the response of the resource with respect to size and time. The GTM data is used to ensure uniform mill feed, generate different ore types for flow sheet parameter modification to obtain the best results. GTM of Rajpura-Dariba deposit categorized ores into easily milling carbonate- quartz lode ores, ores associated with quartz, ores associated with mica-schists and ores associated with graphite-mica schists and ores associated with labile pyrite, copper bearing graphite mica schists indicated the increase in tendency of flotation process refractoriness. Employing iron/copper- graphite depressants/ graphite prefloat or use of centrifugal slimy gravity concentrators like MGS for lead and zinc concentrates was developed for refractory labile pyritic copper bearing graphite mica schist Pb-Zn ores. Alternatively such dirty concentrates may have to be processed either in -9- Imperial smelter or Isa smelter GTM of the deposit helped in maintaining the typical average ore being fed to the mill averting the process fluctuation due to ore variability GTM indicates recoverable concentrates quality quantity and rate based on mining and process efficiency rather than relaying only on assays and reserves. 5 PRELIMINARY BENCH SCALE TESTS The preliminary bench scale tests are carried out after ascertaining ore type and generic process. The preliminary bench scale test is carried out on the sample left over after characterization. The literature survey data on test reports- plant practice on similar ores should be available. The size limitations of processes are given in Annexure C. It is a random exploratory test and may not be true one but furnishes data on left over little quantity sample. The preliminary bench scale tests employ screening as method for particle size refining. Hand jigging for coarse chips, Mozley vanning for powder/slimes is used for studying the amenability to gravity separation. Davis tube testing, Frantz Iso dynamic dry magnetic separation/ Wet high intensity/gradient batch separator and permanent high intensity magnets are employed to study amenability to magnetic separation. Vacumetric dissolved air flotation in graduated cylinders similar to Hallimond tube or in lab flotation machines with small cells are carried out to study the amenability of sample to flotation. The preliminary tests are done under scaled down generic process. The left over carbonaceous silver bearing sulphidic refractory gold ore when subjected to flotation-roast-HCl leach-CIP of concentrate could recover 70% Au as compared to conventional CIP of ore yielding 35% Au recovery. A low grade clayey ferruginous bauxite sample 40.60%Al2O3, 4.7% reactive SiO2 was scrubbed and subjected to sizing from 50mm to 0.025mm, The preliminary amenability sizing yielded a coarse fraction [+0.1mm] assaying 46.5% Al2O3,1.49%SiO2{Reactive}with77% Al2O3 recovery at 67 wt% yield meeting the specifications of reactive SiO2content of the party. A siliceous pyritic fine grained barite assaying 75.02% BaSO4,16.62% SiO2 was subjected to vanning in flat Mozley vanner on -200 + 500 mesh fraction of ground pulp yielded a heavy fraction assaying 96.11 BaSO4 with 53.5% BaSO4 recovery at 42 wt% yield. 6 DETAILED BENCH SCALE TESTING The detailed bench scale tests on different ore types by diverse methods are carried out to evolve simple, sustainable process flow sheet producing stipulated quality concentrate with maximum recovery with a possibility of producing valuable co/by products. The quantity of sample runs from several quintals to tons. It is a vital stage in mineral processing project development and a team of experienced metallurgists are associated in the job. The job is done in several labs and on several samples. The aim of the design testing is to develop a final flow sheet, design criteria and probable equipment. It should contain a material-metallurgical balance, probable energy- water requirements. Any ancillary operation shall be mentioned. The entire as received sample is not subjected to crushing and feed preparation stage as nearly 1/ 2 to 2/3of the as received sample is stored for subsequent bulk tests on developed flow sheet and tests on as received sample. The 1/3 of the as received sample is stage crushed to -10 meshes and refilled to obtain stock sample for detailed bench test program me. The characterizations studies are carried out. The -10 mesh stock sample is subjected to sieve analysis which indicates the amenability of sample to sizing. The samples are ground wet either in standard Denver rod mill with predetermined rod charge, or standard Denver ball mill with predetermined ball charge or Denver pebble mill with -10- pre-determined pebble charge at 67% S density over a stipulated of time in arithmetic sequence with a fixed interval. The ground pulp is subjected to size analysis and studied under microscope for estimating rough degree of liberation. 6.1 Gravity circuit design testing: The aim of the design testing is to develop a final flow sheet, design criteria and probable equipment. Depending on the size gravity test work consists of HMS bucket tests, Jig tests Spiral tests and table tests. The feed preparation stage may involve scrubbing, screening/ classification and needs to be taken care of. The repassing of middlings/ cleaning of rougher concentrate/scavenging of rougher tails may have to be done and has to be recorded. 6.1.1 HMS bucket test: The as received sample is stage crushed to -75 mm and is scrubbed -wet screened over 6mm to reject fines. The screen over size is allowed to drain water completely. 20 lt bucket is filled 2/3 of water and media {Atomised very fine Ferro silicon} so that desired S G is obtained and is kept in suspension with agitator. The HMS media pulp density is measured and is adjusted to stipulate SG. A few lumps are inserted when agitator is stopped and immediately float is removed and sinks are fished out. This process is repeated till all quantum one screen fraction is completed. The procedure is repeated for next fractions. The media SG is changed and either Sink/ Float fractions are subjected to HMS to obtain grade recovery curve at different SG. The respective sink/ float fractions are washed free of media, dried weighed and analyzed. GR curves for each size; overall sample at different SG is calculated. The HMS at different crushed top size/crushing of floats may also be evaluated. The curve giving grade-recovery at different sizes is more helpful. 6.1.2 Jig test: The as received sample is stage crushed to -25mm and screened over10 mesh similar to previous HMS test. The 4” x 6” mineral jig is started as per manufacturer’s instruction. The feed is fed at stipulated rate till sample exhaustion and jig is shut down. The hutch product, jig bed bottom and top bed samples are collected, dried weighed sampled and assayed. The length of stoke, frequency, Hutch water rate, ragging material and different top size are evaluated. The jig is then run under optimized parameters and flow-rates, solid feed and product discharge rates are recorded along with grade- recovery data varying the size, amplitude, frequency, feed rate and Hutch water. 6.1.3 Hydro cyclone/Heavy media cyclone test procedure. : A several quintals of sample are needed for above cyclone/HMC/spiral test rig procedure. The sample is stage ground/crushed and is screened over 1mm, 0.5mm and 0.053mm. The + 1mm fraction is pumped to 150 mm stub HMC along with heavy media pulp[ atomized very fine ferro silicon]and water in an agitated sump. The pump speed is adjusted to get stipulated pressure and media PD is adjusted till the desired density is obtained in HMC U/F. The-6 + 1mm sample is slowly introduced and O/F floats and U/F sink time samples are collected, washed to clean media weighed and dried. The products are subjected to characterization tests. G-R curves are plotted varying size. The hydro-cyclone test is similar to HMC except media is not added and feed PD is varied by adjusting feed [ground] solid to water ratio and varying the cyclone parameters like apex, vortex finder, pressure, feed PD and size. The G-R curves are plotted. The confirmatory test under optimum parameter conditions yield, tromp curve and determine EPM separation character. -11- 6.1.4 Spiral concentrator/table test procedure: As above a several quintals of ground sample is needed. The sample is classified /screened [300/ 500mesh] for spirals/tables to remove slimes. About 60 lts of water is added to the sump and pump rpm is adjusted for 80lpm flow rate. Deslimed material is slowly added to the sump to the stipulated %S. [20/40] pulp density. A known amount of wash water [20lpm] is added. The splitters are adjusted to cut varying from 10 to 90% of feed into concentrate with 10% middling cut. A 5 seconds time samples on products are collected and are weighed wet, dried, sampled and characterized. Grind, pulp-density ,desliming recirculating load and cut position are varied and Grade- recovery curves are plotted against the above parameters. Final test under optimum conditions yields PD, flow-rates. Similarly tabling test procedure is carried out on deslimed/ sized ground pulp samples varying MOG, concentrator cutter position, amplitude, frequency, deck type, cross tilt, dressing water and feed PD and rate. Time samples/ bulk samples are collected, weighed, dried, sampled and characterised. Material and metallurgical balances are prepared and tests on tails with a regrinding step are conducted to verify its significance and discussed mineralogically. 6.2 Magnetic concentration circuit design: Magnetic separation utilizes the force of magnetic field[mH[dH/dx] coacting with other forces producing differential movement of the particle through the field. Davis tube, drum separators and matrix separators are common. The separation may need cleaning and a demagnetizing step may be needed to avert magnetic flocculation. 6.2.1 Wet low intensity magnetic drum separation: The stage ground deslimed magnetite pulp[30%S] is fed to the feed tank by peristaltic pump after adjusting the drum rpm to stipulated level and intensity to 1000Gauss. Denoted amount of wash water is used to wash the rougher magnetic. The rougher tails is repassed in the above drum with slightly low rpm and high intensity of1200 gauss. The scavenger magnetics and scavenger non magnetic product are collected weighed, filtered dried, weighed sampled. The rougher magnetic is fed to the same drum at slightly lower PD[20%S] roataing at higher rpm and at low intensity of 800 gauss. The I cleaner magnetic fraction and nonmagnetic fractions are collected, weighed, filtered, dried weighed and sampled. The products are sent of for characterization studies. It is recommended to conduct Davis tube test at the stipulated intensities to verify the efficiency of LIMS. Mesh of grind, pulp density, drum rotation, feed rate, wash water rate and intensity are some of the variables studied. If any magnetic coagulation is noticed, then magnetic products may have to be passed through demagnetizing coil. 6.2.2 Dry magnetic separation; as in case of WLIS, crushed dry sample is fed to dry RE roll separator via feeder varying the intensity, feed rate, roll speed, feed size and sometimes cutter position. A test under optimum conditions is done. Effect of crushing of non-mag tails before a scavenger pass is also studied. The G-R curves are plotted and suitable flow sheet is prepared. 6.2.1 Wet high intensity magnetic separation; The stage ground pulp is first passed through WLIMS to recover any ferromagnetic fraction. The LIMS tails is the feed to WHIMS. The feed weight is 25 gms/batch for highconcentration of magnetic and 200 gms /batch for low magnetic in the feed. Keeping the matrix constant, the intensity is varied from low to as high as 1.5T. The nonmagnetics at lowinentnisty .0.5 T is then passed at medium intensity of1T. The process is repeated 2 times till maximum intensity of 2T is achieved. The magnetic and 2T nonmagnetic products are collected weighed. The test is repeated varying grind, % loading, -12- matrix, intensity, cleaning-scavenging and desliming [5-10 microns]. The test is repeated in optimum conditions 6.3 Laboratory testing of Electrostatic separator; Carpco 250 mm pinned High Tension separator is normally used. About 2 kg of thoroughly dried hot [2000C] deslimed sand is filled to the hopper. The voltage is raised slowly till electical discharge starts and then lowers slightly to minimize the electrical discharge. Adjust the feeder to feed even stream and collect conducting and insulating products. Weigh and sample and characterize. Repeat the test varying rotor speed, feed rate, electrode voltage, electrode position and product splitter position. The test may consist of scavenging, cleaning operations. 6.4 Laboratory f flotation tests: Typical flotation tests are conducted in cells of 50 to 2000gms. The lab flotation cells are used conditioning the pulp with surfactants. Flotation tests are conducted varying mesh of grind, reagent types, reagent quantities, pulp density, aeration and conditioning time, pH variation. Transfer the ground pulp to cell and adjust the pH, cell RPM to stipulated level. Add the surfactants sequentially depressant, collector and frother and record the dosage and conditioning time to the stipulated level, Open the air **** collect the froth and note the flotation time. As many books and articles have written about flotation testing, it is not discussed in detail here. Normally the collector dosage and condition time for fast floating sulphide minerals with xanthates is fraction [1/20] of medium to slow floating oxide-alkaline earth salt minerals with oleate. The conditioning time for cationic sensitive collectors is similar to that of xanthates but the dosage is relatively high. 6.5 Laboratory leaching tests: The bench scale leach tests of dissolving a metal or a mineral with a suitable recyclable lechant may be carried out. The leaching of sands may done by percolation leach method. The percolation column leach test consists of 200 mm dia glass/PVC column with suitable height. The column bottom is fitted with a filter. The leached solutions are collected in a bottom reservoir. The column is filled with the material to be leached [mostly crushed fines to lumps] and the leachant is fed by gravity from top by reservoir. The liquor samples are drawn periodically and analysed. The leaching time, size, leachant concentration is normally varied. The flow rates are determined to avert channeling. The agitation leach tests are done either in ceramic jar mill rotated by placing on twin parallel rollers, with facility for gas blowing. Alternatively mechanical flotation cell may also be used. The parameters varied are % solids, lechant concentration, size, time. Leaching at elevated temperature, pressure may used for enhanced kinetics. The Eh-pH readings are normally monitored during leaching. Normally gold ores are leached at room temperatures at 50% solids, pH11 using CaO , -200 mesh size using 0.05 to 0.1% free Na CN in solution along with sufficient aeration in ceramic jar mill. The contents after stipulated time [24 hours] leaching is allowed to settle and the supernatant liquor is filtered in a Buckner funnel with Whatman 40 grade paper. The I filtrate is mother liquor. The settled sludge is mixed with 10pH water and filtered with washings. The II filtrate is called as washed liquor. The volumes of ML & WL are noted and sent for assay along with solid residue. -13- 6.6 Laboratory roasting test: Though roasting is normally out of scope of mineral processing, sometimes pre-pyro-processing may assist in concentration. Reduction roasting of paramagnetic iron oxides at 6000 C with reductant carbon mixed ore in limited air supply in a muffle followed by water quenching-grinding-magnetic separation may concentrate ferromagnetic- martitized iron values. 6.7 Final cycle test; The open cycle test may have significant values distributed in the middling which is recovered by recycling with a preparatory stage. The optimum conditions and unit operations in open cycle batch test are maintained in the cycle test. In these tests the intermediate products from one batch test is saved and re introduced at appropriate place/point in the subsequent batch test. Dilution becomes a problem and is controlled by dewatering. The reclaimed water is used at appropriate places in the circuit for maintaining %Solids. When equilibrium is established the mass of fresh input is equal to mass of output. Normally 6-7 cycles are needed to reach equilibrium. The last two cycles represent the equilibrium. The last cycle can give kinetics data and middling streams indicate the extent of circulating load. Different circuits like counter current, co-current and mixed ones are attempted. The material and metallurgical balance of the circuit is prepared. 6.8 Determination of bulk density, angle of repose: These are determined on one cubic meter cube container and measuring inverse cosine of [(44x incline)/ (7 x circumference)] respectively as per the standard methods on desired products of final cycle test. 6.9 Bond’s ball mill work index: It is carried out on stage crushed samples from 6 to 30 mesh [preferably 6 mesh].It is done on 700 cc packed sample in 12” x 12” ball mill,70 rpm, 285 steel balls of 1.5”,1.25”,1”0.75” and 0.63” with ~20,125 gms for producing a constant 250% circulating load, net gm/mill revolution on test sieves from 28 to 325 mesh [ preferably 150 mesh].The product at equilibrium cycle [ producing constant net grams/ mill revolution ( GBP) at 250% circulating load] and fresh feed is subjected to sieve analysis to get 80% passing sizes. The Bond’s ball mill work index WIB is calculated from the following equation. WIB = [44.5] / [{(Pi)0.23} x {(Gbp)0.82} x {(10/ (P80)0.5) - (10/(F80)0.5)}] Where WIB = Bond’s ball mill work index in Kwh /short ton, Pi = Test sieve opening in microns, Gbp =Net grams of test sieve under size produced / mill revolution, P80 =80% cumulative passing size of product, F80 =80% cumulative passing size of feed The metal wear is determined from abrasion index Ai which is loss of SAE 2340,500 Brinnel hardness steel paddle of size 3" x1” x 0.25” rotating at 632 rpm when abraded coaxially in 12” x4” mill counter rotating at 70 rpm with 400 grams of-0.75” + 0.5” of test material for 15 minutes. Metal wear is 0.175[Ai -0.015]0.33 -14- 6.10 Thickening test: Batch settling tests on fresh pulps in ~ 2 liter measuring glass graduated cylinders are carried out with and without addition of flocculants on final fresh pulp products of cycle test. The mud line height with reference to time [minute wise] is marked on cylinder and noted. The nature of pulp surface, floc size, supernatant water quality and qualitative settling rate is recorded. H0 and Hu are marked at 0 and 19 hours of settling. The tangent of h – t curve at compression point intercepting Hu line on t axis gives tu. The slope of the tangent gives the average settling rate. Initially, tests are carried out to select the suitable flocculent. Thereafter tests are done to optimize dosage variation. The initial and final pulp volume is noted. The pulp weight and dewatered- dried solid weight is noted. The unit thickener area UA without safety factor is calculated by Kynch formula.UA = [tu / (C0 x H0)] Where tu ultimate time from curve in days,H0 is initial height of pulp in meters, and C0 is initial pulp concentration in t/m3 6.11 Filtration test: Initially vacuum leaf filter tests using leaf filter of 0.01square meters are carried out on fresh thickened pulps [by thickening the cycle test products under optimum thickening conditions] varying media keeping pulp density, vacuum, cycle time constant noting the filtrate clarity, volume , cake thickness, nature of discharge. Thereafter the filtration parameters like pulp density, vacuum, cycle time are varied to get cake with minimum moisture and maximum productivity. The filter productivity Fv in t/m2 / hour for disc filter without safety factor is as follows. Fv = [(Dry cake weight in tons x 3600)/ (2.5 x dry time in seconds x 0.01m2)] Similarly the pressure filtration test using Larox pressure filter of 0.005 square meter is carried out on thickened pulp varying media. Keeping pulp density, pressure, constant noting the filtrate clarity, volume, cake thickness, nature of discharge. The total cycle time involves pumping, diaphragm pressing, air blowing, and dead time [ 240 seconds]. Thereafter the filtration parameters like pulp density, pressure, cycle time are varied to get cake with minimum moisture and maximum productivity. The filter productivity Fp in t/m2 / hour for pressure filter without safety factor is as follows. Fp = [(Dry cake weight in tons x 3600)/( total cycle time in seconds x 0.005m2)] -15- 7 CONFIRMATORY PILOT SCALE- SIMULATORY BENCH SCALE TESTS The flow sheet in the pilot plant is normally based on final cycle test data evolved at bench level termed as developmental testing. For complex ores, ores with minute quantities of precious values, ores indicating uncertainties and new grinding/ concentrating technology is to be tested, the confirmatory pilot scale tests are recommended. The pilot scale tests are also carried out to determine the scale-up factor from batch/bench operation to continuous operation and hence termed as design testing. The following unit operations are involved. Blending-preparation- sampling, conventional crushing & screening, conventional OC/CC rod mill/ ball mill circuit with screen /classifier/cyclone, scrubbing, screening/classification, batch/continuous gravity concentration by jigs/HMC/spirals/tables/MGS, batch/continuous WLIMS/WHIMS, dry REdrum separator, continuous mechanical/column flotation, thickening- filtration, simulatory tests like Work index, Abrasion index, autogenous WI, Vacuum leaf [cloth/ceramic] – pressure filter test pot pellet test, sintering test, kiln calcinations test. Annexure F gives the importance of pilot scale- simlulatory bench scale tests unit data for process confirmation and for design of plant. A few typical pilot plant flow-sheet schemes are briefly described. The pilot scale test is carried in 3 stages;1] Break in operation tests[Preliminary]- In this period the plant is run for a brief period to fix the MOG and arrange the concentration- ancillary circuit. The pulp is fed and sampling scheme is formulated and any practical snags that arise are solved. 2] Operations for optimization- The parameters of various unit operations and circuit configurations are varied and its effect on results are studied and compared with lab data. The control simulators scaled down lab tests are conducted to verify the pilot scale results. This is the critical stage and the plant is run for a minimum of 12 hours a day. 3] Demonstration; the plant is run continuously and effect of reclaimed water is studied. The bulk products are produced and the productivity metallurgical balance is compared with regular instantaneous sample based metallurgy. Empirical models for effect of variables are prepared. Sensitive factors are indicated. The overall material, energy, metallurgical balances are prepared. All the recommended unit operations [blending, crushing, screening, closed circuit grinding, concentration, dewatering, product handling and recycling the reclaimed water in the configured circuit] in the process flow sheet are used on a continuous basis as far as possible to get a miniature industrial show for increasing the confidence. -16- 8 EQUIPMENT SIZING AND TECHNO-ECONOMIC FEASIBILTY Based on the market, reserves, mining schedule and capital available the schedule indicating maximum and mean throughput is calculated. The material balance is re-prepared for maximum and mean throughput. The utility [ore, water,power,manpower] demand is estimated. Based on the data gathered in pilot plant, catalogues and similar capacity plants treating similar ores, the equipment sizes are calculated as per the standard procedure. The equipment sizes and types/models are chosen and budgetary estimates are obtained. The capital cost is estimated. Similarly the direct operating cost [supplies, utilities, operating& maintenance labor] and the gross operating cost/ton of ore is estimated. The revenue / ton of ore realized are calculated from mean price of finished concentrate. Profit realized/ton of ore is estimated and change in this figure as a function of change in concentrate price, operating cost, productivity is estimated. The net asset value [NAV] for the period of project is estimated and % NAV profit is calculated. A case study of basic engineering and preliminary viability is given for a near waste technology calcareous wollastonite mine waste project. 6 million tons of Wollastonite occurs as skarns in Rajasthan and is mined by open cast method. The wollastonite is hand sorted from calcite. The hand sorted calcareous wollastonite reject is stacked separately and is about 0.3 million tons. The sorted wollastonite product assaying +97% wollastonite and LOI <2% is sold to ceramic industry @ Rs 2000/ ton at mine site. The 50mm mine rejects assaying43.43% CaO, 46.68% SiO2, 5.08% R2O3 and 4.03 % LOI containing 75% wollastonite, 10% calcite, 10%epidote and 5% quartz was received for design test work. The diagnostic characterization subjecting to Is-dynamic magnetic separation – acetic acid leaching of non- mag fraction yielded 97% wollastonite at-150 mesh size, indicating that removal of calcite and epidote will yield stipulated grade. The process comprised of anionic flotation of calcite at -150 mesh d80 74 microns, pH 7.8 using 2.5 kg/t of sulphuric acid as pH moderator and silicate depressant, 1 kg/t of sodium oleate in 4 stages as collector followed by WHIMS of non-float at 2T yielded a wollastonite concentrate assaying 97% wollastonite, 0.41% LOIwith70 wt% yield, meeting the specifications. The lab flotation time was 8’ and the pilot scale flotation time was 25’. The scale up time in plant was estimated to be 4 times that of lab. The rougher calcite float may also used in cement industry. Column cleaning of calcite float yielded low silica calcite concentrate that can be used in filler/sugar industry. A 100tpd-30000tpy project was proposed keeping low reserve of 0.3 million tons and limited market. Simons 600 mm coarse cavity crusher is chosen as top size is 50mm and it has to crush to minus 0.5 inch. The crushing section is designed for 1 shift operation. Accordingly the DDscreen [50 mm, 12mm] for closed circuit crushing with 600 mm crusher is sized for 24tph. The mill kw amounts to 60, indicating a 5’ dia mill. Unfortunately the lowest dia mill and HP available is 150 HP 2m x2.5, which can take 8tph. The mill is under loaded to maintain the grind. A 10” long body cyclone is the order of magnitude choice or two 5” cyclones may be needed. A 1.5 m dia conditioner is chosen for conditioning with sulphuric acid for 10’. Similarly flotation cells of 600 cft is chosen to have a minimum 60’ retention including conditioning time for stage added collector and to take extra throughput The non-float is passed through 1000mm dia WHIMS so that retention time will be sufficient even when the plant is run at double capacity. Though the plant has to be designed for 4.2 the, it has been over designed for 200 tpd as it is minimum commercial size available in market and it can take care of water short fall period with water reclamation. The equipment is chosen based on data available -17- from ‘Basics in mineral processing – Metso minerals Hand book’ The list of equipment is given in Annexure G. The purchased equipment cost is Rs1.2 core. the direct capital cost is Rs2.28crore. The total gross capital cost is Rs3 crore The cost is estimated as per USBM RI 9147 of 1987. The direct operating cost [utilities, supplies maintenance, and labour] is Rs 480/ ton of ore. The fixed cost/ton of ore 190/ton of ore. The gross operating cost amounts to Rs700/ton. The revenue realized/ton of ore is Rs1400/ton @ Rs2000/ton of concentrate at mine. If the calcite product is also salable, revenue, profit is sure to increase substantially with minimum investment and the project is sustainable economically and ecologically, due to low waste- low cost process. 9 PLANT AUDITING STUDIES The viability of a mineral enterprise necessitates the development of specific techno-economic model. The techno-economic model predicts the operating strategy during the economic recession. The co-dependence of process variables, interdependence of different operations in a mineral enterprise, non-quantifiable factors needs continuous monitoring. This continuous monitoring and generation of techno-economic model is difficult. Hence, age old philosophy of improving viability by maximizing recovery and productivity seems unsustainable till significant overall unit cost rate reduction and enhancement of overall unit profit rate is achieved. With stress on cost reduction, project engineers engaged in multiple projects, usually opt profitable projects than looking into performance improvement studies. However, the performance improvement of the plant needs the routine auditing studies, for reducing the overall unit cost and to improve the cash flow. Audit is defined as a formal, thorough and periodic examination – evaluation of a system. The global audit is divided into geological audit, mining audit, marketing audit, energy audit, process audit and environmental-safety audit. It deals with metallurgical process audit and its role in plant overall performance improvement. The aim of process auditing is to understand the effect of the process variables on the profitability. Table 1 denotes the steps and outline of process auditing. The performance improvement studies by process auditing are demanding from time – economic viewpoint. Sometimes it is frustrating due to ill defined objective, improper problem identification and lack of will for implementation. It is a tough job as conceptual ideas have to be sold enumerating the costs, time and risk factors with relation to the benefits obtained. The problem compounds if the historical data is improperly logged and improper location of sampling points in the circuit. However, the total involvement of plant team with proper communications is the key to solve the problems associated with auditing. The data is analysed logically, scientifically, statstically keeping techno-economics in view. Once the problem is identified, test works under simulated conditions, based on evolutionary and revolutionary concepts, are conducted. Conclusions are drawn evaluating the alternatives for solving the problems. The recommendations are made based on sustainable benefits. After on-site implementations, circuit is sampled, the results with techno-economic benefits are evaluated with reference to base line and projected values. Recommendations for improvements are suggested. -15- Table 1: Out line of steps in a metallurgical process auditing

Define project objective and check the objective
Identify the problem
Review the historical data
Designing the sampling campaign
Sample the unit operations/circuit
Logical data analysis and base line data preparation
Test work in simulated conditions-evolutionary and revolutionary concepts
Techno –economic evaluation of alternatives, comparing with similar industrial data base
Recommendations, on site implementation, evaluation and conclusion
A case study of metallurgical process audit of a graphite flotation plant for performance improvement is discussed. ROM graphite is crushed to -10mm in two stages using a primary jaw crusher and an impactor both in closed circuit with 40 and 10mm DD screen.-10mm crushed fines is ground to -30mesh in open circuit rod mill. The RMD is subjected to rougher and scavenger flotation in mechanical cells. The rougher float is cleaned twice using mechanical cells and 900mm column with regrinding before every cleaning stage. The II cleaner float is screened over 72 mesh. -72 mesh fraction is cleaned twice using 600mm columns with one regrinding between cleaning stage. The column tails are scavenged in 1200mm column and the float is fed to 72 mesh screen undersize The +72mesh flaky concentrate and IV cleaner fine concentrate are dewatered. The plant was designed to handle 250TPD graphite, yielding +95% FC concentrates with 90% FC recovery for ROM graphite assaying 17%FC.However, the plant was operating at lower throughput, yielding inferior concentrate grade with low recovery. Systematic sampling followed by lab scale test work simulating plant design conditions indicated the dilution, mismatching size of cells and improper collector-frother regime, was the reason. Flotation circuit was rearranged in CCC mode maintaining the pulp density and cell volume and changing the collector-frother regime from expensive imported reagents to cheap locally available collector frother mixture of kerosene/LDO and MIBC /pine oil. A middling thickener was used ahead of flotation for recirculating cleaner tails and scavenger float. The plant could produce the high revenue yielding +96% FC concentrate with +85% FC recovery nearly maintaining the throughput at stipulated MOG. Table 2 indicates the salient results. Efforts are on to produce finer mill feed by enhancing crushing to improve the throughput to designed levels. Table 2: Results of process auditing – rearrangement of flotation circuit

Particulars Base line Modified
TPH ROM 7.8 7.0
Feed %FC 15.6 15.6
Conc %FC 93.73 96.00
Tails %FC 5.15. 2.60
FC% Recovery 70.6 85.0
TPH Conc 0.92 0.974
Revenue Rs/hr 9200 18000
-16- 10 SUMMARY AND CONCLUSION The paper describes briefly importance of characterization, testing, sizing, economic evaluation and auditing of the established plant of project development on a mineral resource. The critical data, steps to be followed in such a mineral processing venture is briefly described. The necessity to have a thorough knowledge of characteristics - processing of the minerals, some knowledge of geology, mining, product market and its subsequent processing - use, economic- ecological significance by the mineral processing engineer is stressed. 11 ACKNOWLEDGEMENT The author is thankful to the CGIBM and other colleagues of IBM for the guidance and interest. .

More pages