Tailings and Mine Waste Conference

Water recycling at processing plants in water scarce regions : a case study of thickener design for the… Behrouzi, Kourosh; Fard, Majid Vafaei; Raeiszadeh, Anahita; Faeghi nia, Aida Nov 30, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


59368-Behrouzi_K_et_al_TMW_2011.pdf [ 795.25kB ]
JSON: 59368-1.0107704.json
JSON-LD: 59368-1.0107704-ld.json
RDF/XML (Pretty): 59368-1.0107704-rdf.xml
RDF/JSON: 59368-1.0107704-rdf.json
Turtle: 59368-1.0107704-turtle.txt
N-Triples: 59368-1.0107704-rdf-ntriples.txt
Original Record: 59368-1.0107704-source.json
Full Text

Full Text

Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Water Recycling at Processing Plants in Water Scarce Regions – a Case Study of Thickener Design for the Mansour Abad Processing Plant Kourosh Behrouzi Mining Investment Insurance Corporation, Iran Majid Vafaei Fard Mining Investment Insurance Corporation, Iran Anahita Raeiszadeh Azad University of Firuzkooh, Iran Aida Faeghi nia Ceramic Division, Materials and Energy, Iran Abstract Due to the shortage of water sources and its critical consequences, the need for water recovery is of increasing importance. Mineral processing plants are generally water consuming although some wastewater can be recycled. The most common technique to recycle water is through thickening. Thickeners are built to concentrate the solids content of the pulp and clarify the overflow water. In this paper, two approaches are employed for the design of thickeners due to their use in the mining industry. First the Coe and Clevenger method was analysed and then the Fitch and Talmage method was evaluated. In the conditions that settling of solids is for clarifying purposes, counter current decantation may be used. In this paper the method of calculating the optimum thickener in type and size regarding economic factors was evaluated and a case study of thickener design for the Mansour Abad concentration plant is described. The thickener's area was estimated to be 4.7 m2 with depth of 1.80 ±0.20 m. Introduction Mining and industry have brought many positive achievements for mankind but they have not prevented it from unintended negative consequences. Environmental pollution resulting from mining and industrial activities is one of the undesired aspects and many societies are affected. One environmental concern is tailings from flotation plants. The separation of the solid and liquid phases of a suspension is very often a two-stage process. The first stage, known as thickening, is physicochemical and involves the conversion of discrete, unflocculated particles of the suspension into a thickened underflow with a clear overflow, often by the addition of a coagulant or flocculant. This step offers a high degree of control and is efficient irrespective of the scale of the operation. This reliability is, however, of vital importance in terms of establishing the conditions required for a similar effectiveness of the second stage. The second stage is predominantly an physical operation, designed to reduce the remaining water content of the thickened underflow by a suitable solid-liquid separation process, thereby transforming it into a compact solid containing only small quantities of water. The sedimentation concept (batch settling) Particle size, particle density and fluid viscosity are readily recognized factors to be considered in any sedimentation process. Less obvious are particle shape and orientation, the distortion of a deformable particulate, the interference of one particle with another (particularly when concentrations are high), the Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 nearness of the wall of the container, and convection currents. These also have a significant and varying influence on the separation process. Materials with particle diameters of the order of a few microns settle too slowly for most practical operations. Wherever possible, such small particles are agglomerated, or flocculated, into relatively large clumps called 'flocs' that settle out more rapidly. Flocculation can be considered to include both the effect of coagulation and the effect of polymer flocculation. The settling of discrete particulate masses based initially upon modifications of Stokes' law has been included in the research of many different solid-liquid systems. Early research work in this field considered the settling behaviour of spherical particles for a wide range of concentrations, from dilute to concentrated suspensions. A similar approach to that used for spheres was adopted for a study of the sedimentation of small uniform particles by Steinour[1] and later (1950) by Hawskley[2], who both correctly assumed that the up thrust acting on the particles is determined by the density of the suspension rather than that of the fluid alone. For the sedimentation of uniform particles, the increased drag is probably more attributable to a steepening of the velocity gradients than to a change in viscosity as a result of solids concentration. In practical terms, the rate of sedimentation of a suspension of discrete fine particles is extremely difficult to predict because of the large number of factors involved. However, a number of empirical equations have been developed for the rate of sedimentation of suspensions as a result of tests carried out in measuring jars or cylinders. For a given solid and liquid, the chief factors which affect the process are the height of suspension, the diameter of the containing vessel and its shape, and the solids concentration. Wallis [3] attempted to bring together results obtained from a variety of conditions to categorize the behaviour of settling sediments of uniform particles (> 100 µm) sufficiently large for anomalous viscosity effects and flocculation to be regarded as negligible. The settling of flocculated particulate masses is a complicated process involving rearrangements in the sediment long after the flows themselves have settled. Bottom-lying flows are compressed by the weight of the others that settle upon them, since flows usually are bundles of particles held together by weak forces, and which have entrapped within their structure considerable quantities of the liquid medium. This produces sediment with varying degrees of density. A simplified history of the batch sedimentation of a flocculated suspension is shown in Figure1. The upper section is a continuous plot of the four zones that have been observed to develop during sedimentation and the lower sections show the containers at various stages of settling, according to Comings et al. [4] Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011  Figure 1 Sedimentation of a flocculated suspension [4] Cylinder A contains a uniformly mixed but flocculated suspension. Cylinder B shows the situation shortly after when, at the very base of the cylinder, there is a zone (zone 4 in Figure 1), which consists of a mixture of flocs and comparatively large grains resting at the bottom upon one another. This zone has formed from flows initially very close to the container's bottom. Immediately above the lowest zone is a transition layer that is intermediate in density between the deposit and the suspension. This is the zone in which liquid is being squeezed out from among a network of flocs, often calledthe compression zone (zone 3 in Figure 1). Some authors have preferred to break the compression zone into two; notable among these are Coe and Clevenger [5], whose work in the field of sedimentation provided the basis for a considerable amount of work by others who accepted their original assumptions. The work of Coe and Clevenger, however, assumed the settlement of non-colloidal particles as initially discrete units. A deliberately flocculated suspension does not appear to exhibit quite so clearly the characteristics that would allow two sub-zones to be distinguished within the compression zone, and cylinders B and C contain a layer of flows still in suspension. Sometimes this is referred to as the 'free settling zone', but a more appropriate term is that suggested by Anderson and Sparkman [6] - 'the zone of collective subsidence' - because every particulate group, though settling, is retaining its position relative to its neighbours. Zone 2 maintains the concentration of the original suspension. The upper zone (zone 1 in Figure 1) is a zone of liquid which, ideally, is clear if the suspension is well flocculated, and turbid or discoloured if not, thereby still containing fine suspended particles. When well flocculated, the boundary between the liquid and the suspension is normally sharp and easily located almost immediately after mixing. Other zonal boundaries are less sharply defined and are often difficult to detect because of the opacity of the suspension. However, X-ray adsorption techniques have been used to locate them (Gaudin and Fuerstenau [7]). Cylinder C illustrates the continued settling effect when the upper and lower zones increase in volume, the suspension zone decreases, and the zone of sediment undergoing compression remains essentially the same in volume but moves upward. This continues until (cylinder D) the suspension zone disappears and all the solids exist in the form of a sediment. This condition is known as the 'critical sedimentation point'. As the graphical representation shows, the solids-liquid interface follows an approximately linear relationship with time until this point is approached. Following a short transitional phase, the sedimentation continues at an even slower rate, attaining eventually the final condition represented by cylinder E. The compression of the sediment occurs between D and E, and the time involved in this compression phase constitutes the major part of the time involved for the total process. The liquid that accompanies the Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 flows into the deposit is slowly expelled under the weight of the sediment above. This continues until equilibrium is established between the weight of the flows and their mechanical strength. Overall process time often depends very much on the type of settling aid used. Coagulants such as lime may give process times of several hours whereas acrylic-based polymer flocculants can achieve process times of a fraction of this. Process requirements and cost are therefore very important considerations in the selection of a settling aid between these two extremes. Spherical and quite compact flows can often be achieved by using combinations of electrolytes and polyelectrolyte's and mixing in a specialised fashion (Hamza [8]). Such flows are not only denser and faster settling, but also entrain less of the suspending liquid and hence produce a more distinct separation. Thickener design by means of batch settling tests The classic work on the capacity of a continuously operating thickener was provided by Coe and Clevenger [5] in 1916 and was based upon the assumption that the rate of settlement within the zone of settling is a function only of the solids concentration. If so, then it should have the same value in a batch test as in continuous operation. They devised a formula to give the required area of thickener vessel in terms of the various concentrations, using batch settling tests conducted at different initial pulp concentrations to indicate corresponding settling rates at various horizons in the vessel. Once a thickener has achieved steady operating conditions, the total solids flow rate at any level between the feed and underflow points remains the same and the overflow is hopefully clear. Then, by solids volumetric balance: u uu ss f CQQCQ ρρρ −= And by liquid volumetric balance, )1()1( s u u s U CQCQQ ρρ −−−= This, by substitution, gives: )1()1( s u us o C C QCCQQ ρρ −−−= Or )11()1( uu o CC QC C CQQ −=−= where Q0/A represents the superficial velocity of the liquid being displaced by settling solids, C and  ρ are the concentration and density of pulp respectively and lower indexes u, s and f represent underflow, solid and feed ; if the thickener is to discharge a clear overflow this must not exceed the settling rate of solids at concentration C. Thus, by substituting the velocity by U, the equation becomes )11( u ff CCU CQ A −= Application of the Coe and Clevenger method can often lead to values of thickener area which are inadequate, as described by Scott [10], Yoshioka et al.[8], Talmage and Fitch [9], and others. The main source of error suggested for this tendency to underestimate is the behaviour of the suspensions in batch tests, which often leads to overestimation of thickener fluxes. Such behaviour may occur as a result of channelling or short circuiting of fluid through the higher concentrations and may possibly be partly attributable to segregation of the particulate during compression in the case of flocculated suspensions. If the cylinder diameter used is small, wall effects may cause irregularities in settling behaviour. The Coe-Clevenger relationship may still be valid in compression but no adequate Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 theoretical model has been developed which relates interface subsidence rates in batch tests to settling rates in continuous operation. In a batch test, settling commences with a uniform initial concentration of solids as shown in Figure 2. The concentration in zone 3 must range between the original concentration in zone 2 and that of the final condition in zone 4. If the solids handling capacity per unit area (SHC) is lowest at some intermediate concentration a zone of such concentration must start to build up since the rate at which solids enter this zone will be less than the rate at which they will leave. The mathematician Kynch [6] showed how to determine the concentrations and fluxes in these zones by constructions on the transition section of a single batch settling curve. Talmage and Fitch [9] used Kynch's approach to determine the critical zone, or limiting flux, for a continuous thickener directly from a Kynch-type construction.  Figure 2 Batch settling testing [4] Combining Kynch's analysis with the Coe-Clevenger formula gives the graphical construction from which unit area can be obtained (unit area being the area of thickener required to handle unit mass of solids in unit time) as shown on Figure 3a.  Figure 3 (a) Talmage-Fitch construction. (b) Oltman construction [4] Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 The total mass of solids in the batch test is C0h0a. When any capacity-limiting concentration layer reaches the interface, all solids in the column must have passed through it since it was propagated up from the base of the column. ahCwUatC oojjjj =+ )(  If hj, is the height of the interface at time tj, and since it has been shown by Kynch that the upward velocity of any specific layer is constant, j j j t h w =   ahC t h UatC oo j j jjj =+ )( Which simplifies to: jjj oo j tUh hCC + = Referring again to Figure 3a, the value of Uj is the slope of tangent hi to Q at t = tj ot hh u j jj j − − = This simplifies to hi=hj+tjuj Combining equations: oojj hChC = This means that hi is the height of uniform slurry of concentration Cj which contains the same amount of solids as the initial slurry. The settling velocity as a function of concentration may be developed from a single settling test by use of the above relationship. Using arbitrarily chosen values of concentration, Cp the corresponding value of hi can be calculated. Uj can then be determined as the slope of the line drawn through the point hi and the tangent to the settling curve and a complete set of data showing Uj as a function of Cj can therefore be developed from one settling test. These data can be represented graphically as shown in Figure 4  Figure 4 Settling rate as a function of slurry concentration [5] In order to specify the area requirement of a thickener, the concentration layer requiring the maximum area to pass a unit mass of solids must be determined. This may be done by calculating the unit area Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 required for a series of concentrations using the data showing U as the function of concentration (c), developed previously, and substituting in the Coe-Clevenger formula results in:             − = u CC CQ AUA uj ff 11  Whichever concentration layer gives the largest unit area is then used as a design basis. With Figure 3a, a simple geometrical construction may be used to obtain these areas directly. In Kynch's equation: uujjoo hChChC == Where Cu is the desired underflow concentration and hu is the corresponding height: ou uj ujoo u uoo j j hC hh CChC h ChC h C − =−⇒=∧= 1111  and If Uj is represented by slope hj at Q:       −       − =         −⇒ − = u iu oo iu uji iu j t hh hC hh CCt hh u 11  So that: oo u hC tUA = Then the procedure is as follows: 1. Determine hj and hu from the following material balances; point hj corresponding to an arbitrarily selected concentration, Cj, uujjoo hChChC == 2. Draw an 'underflow' line parallel to the time axis at height hu on the settling curve, as shown in Figure 3b; 3. Draw a tangent to the settling curve through point hi; 4. read tu at the intersection of the tangent and the underflow line; 5. Calculate oo u hC tUA = If hc is the required depth of the compression zone for a flat-bottomed tank the depth will become: A Vh cc =  And for a circular tank of diameter d 2 4 d Vh cc pi =  Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Which can also be modified to consider the conical section of the thickener. The total depth of a thickener is usually estimated by allowing for clarification and settling zones as well and these usually lie in the cylindrical section. Between 1m and 1.5m are usually allowed for each to provide for size capacity when this technique is applied. Designing counter current decantation thickeners There are a number of commercially available units often with microprocessor based control for automatic adjustment of flocculant addition. This type of unit is most useful for thickeners handling low density, colloidal or precipitated solids or for high-capacity thickeners discussed earlier. When reviewing overall thickener control systems and their applications in the various metallurgical, chemical and municipal fields, one of the most complex examples is that of the counter current decantation systems. Constant costs can be estimated as follows [10]: nFF V IDFIC cDMSF ×××      × − ×= 48.0)(054.08.15 ς  Variable costs are evaluated via: TCS R RC sonv ×× − − = + 1 1 1  Minimum costs can be estimated by following formula where n is variable: ( ) ( )1 054.080.15 )1( )1( 48.0 21 −×××       × − ××× = − + + RTCS V IDFFFI R nR so cDMS n n ς  Which: CF: constant costs ($/year); CT : total costs($/Y); D :solving ratio; F :solving ratio in inlet;  R :solving ratio over to underflow; FC :constant cost ratio; I :dissolved solid ratio (TPD);  IMS :marshal & swift index;  n :number of thickeners; SO :Feed (TPD);  T :number of working days;  F :correction factor of currency; V :sedimentation velocity (m/hr):  ς  ; pulp Density. Thickener design for water recycling at Mansour Abad processing plant Mansour Abad is a lead and zinc mineral processing plant in Iran. Presently water recycling from the flotation plant is conventional and is using storage cemented ponds and pools for water recovery. There are water shortages in the area and there is a strong need to increase recovery. Accordingly, the design and manufacture of a thickener was proposed to provide a higher density underflow. To design a thickener for tailings from the lead and zinc flotation plant, tests were conducted to determine the velocity of particle sedimentation to assist with the design of the thickener.  The results of these tests are and graphs are shown on Figures 5 and 6. Results shows that without adding flocculant, the velocity rate is low but after adding A100 flocculant with 10-15 g/ton, the sedimentation velocity was considerably increased. The Kynch method was used to determine unit area. U.A. is calculated as flows: Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011  oo u hC t UA × = 0694.0  Which: UA :Unit Area (m2TPD-1) ; tu :time to meet compression point (min) Co :initial pulp concentrates (g/cm3) ho :height of the pulp from cylinder base (cm). The solids content and feed rate was estimated at 27.5% and 10 t/h (240TPD) respectively. The time to meet the compression point with adding 10-15 g/ton flocculant is 4.1 and 2.1min. Substituting this data into the above equation, the unit area will be .0197 and 0.01 m2TPD-1. Considering feed rate equal to 240TPD, the thickener diameter would be 2.4 and 1.7 m. )4.2(7.124001.04 )7.4(4.22400197.04 2 5.0 2 5.0 mmD mmD =      ×× = =      ×× = pi pi  The formula to determine the thickener depth is: )1( )1( − − = ps rs c A th ρρ ρ  Where: hc :required depth for dry solid; tr :time span between compression time and time to meet desired underflow solid percent (hour) ; Referring to Figure 6, tr is estimated to be 14 min or 0.23h. Solid and pulp density was measured at 2.3 and 1.2m respectively and the thickener area was calculated to be 4.7 m2. Substituting data in the associated formula, hc is: 1)(319.0 7.4)12.1(3.2 3.2)13.2( − = ×− ×− = TPHmhc  The feed rate is 2.75 TPH and therefore the required depth is estimated to be 0.8 m. Considering 1±2 m height for retaining clarified water, total depth will be 1.8±2 metre. Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 0 2 4 6 8 10 12 14 16 18 20 22 24 26 Settling Time (min.) he ig ht  fro m  cy lin de r ba se  (cm ) added floc 10g/ton  Figure 5 Sedimentation chart of Pb & Zn waste with adding 10 gr/ton A100 Floc 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 0 2 4 6 8 10 12 14 16 18 20 22 24 Settling Time (min.) he ig ht  fro m  cy lin de r ba se  (cm ) added floc 15 g/ton  Figure 6 Sedimentation chart of Pb & Zn waste with adding 15 gr/ton A100 Floc Economic results After construction of the thickener in accordance with the work described above, water consumption was reduced from 400 m3 per day to 100 m3 per day thus decreasing water provisioning costs.  This work has also improved environmental aspects of the facility.    Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011  Figure 7 Different stages of Manufacturing of Thickener at Mansour Abad Acknowledgements The authors express appreciation to Firuzkooh University, the Young Iranian Scientist Club (bpj) and Mansour Abad mineral processing plant for their kind assistance. References [1] Ralph E Beeman; Joseph H Reitberger; 'An integrated industrial management facility for biological treatment of high nitrate and carbonaceous wastewater', Environmental Progress; New York; Apr 2008 [1] Tzong T. Chen, 'Filter press plugging in zinc plant purification circuits', JOM, New York; Vol. 55, Iss. 4; pg. 28 Apr. 2008 [3] Steinour. HH, 'Rate of sedimentation-non-f1occulated suspensions of uniform spheres-Suspensions of uniform size angular particles'. Ind. Eng Chem., 36,618-624, 804-847 (1944) [4] Hawskley, P.G.W. 'The effect of Concentration on the settling of suspensions and flow through porous media'. Inst. Phys. Symp. 1950. p. 114(1950) [5] Wallis, G.B., 'A Simplified one-dimensional two - Component vertical flow', Inst. Chem. Eng. Symp. on Interaction between Fluids and Particles, 19,9-16(1962) [6] Comings,  E.W.,  Pruiss,  C.E.and  De  Bord,  C.,  'Continuous  settling  and thickening', Ind. Eng Chem., 46, 1164-1172 (1954) [7] Coe, H.S. and Clevenger, G. H., 'Methods for determining the capacities of slime thickening tanks'. Trans. AIME, 55, 356,384(1916) [8] Anderson, A.A. and Sparkman, J.E., 'Review of sedimentation theory', Chem. Eng., 75-80(1959) [9] Gaudin, A.M. and Fuersienau, M.C., 'Experimental and mathematical model of thickening', Trans. Soc. Min. Eng., 223, 122-129 (1962) [10] Hamza, HA., 'Least cost flocculation of clay minerals by polyelectrolyte', Trans. IMM, Sect. C, 87, C212 (1978) [11] Fitch, E.B., 'Current theory and thickener design', Ind. Eng. Chem., 58, 18 (1966) [12] Kynch, G.J.,Trans. Faraday Soc, 48, 166 (1952) Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 [13] Zoma, A., ' Calculation optimum number of stages in continuous countercurrent decantation (CCD), Minerals and metallurgical processing, pp 118-120 (2010) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items