minerals

Article Selective Recovery of Mushistonite from Gravity Tailings of Copper–Tin Minerals in Tajikistan

Lei Sun, Yuehua Hu *, Wei Sun, Zhiyong Gao ID and Mengjie Tian

School of Minerals Processing and Bioengineering, Central South University, Changsha 410083, China; [email protected] (L.S.); [email protected] (W.S.); [email protected] (Z.G.); [email protected] (M.T.) * Correspondence: [email protected]; Tel.: +86-731-8883-0482

Received: 31 October 2017; Accepted: 4 December 2017; Published: 7 December 2017

Abstract: Tajikistan has abundant copper–tin resources. In this study, mineralogical analysis of copper–tin ores from the Mushiston deposit of Tajikistan indicates that tin mainly occurred in mushistonite, cassiterite, and stannite, while copper mainly occurred in mushistonite, malachite, azurite, and stannite. The total grades of tin (Sn) and copper (Cu) were 0.65% and 0.66%, respectively, and the dissemination size of copper–tin minerals ranged from 4 µm to over 200 µm. Coarse particles of copper–tin minerals were partially recovered by shaking table concentrators with a low recovery rate. Based on the mineralogical analysis, flotation recovery was used for the first time on the fine particles of copper–tin minerals, including mushistonite, from shaking table tailings. Single factor flotation experiments, open circuit flotation tests, and closed circuit flotation tests were performed to determine the optimized flotation conditions. Results indicated that benzohydroxamic acid (C6H5CONHOH) and lead nitrate could effectively recover the mushistonite, cooperating with other depressants. The final concentrate contained 13.28% Sn, with a recovery rate of 61.56%, and 18.51% Cu, with a recovery rate of 86.52%. This method proved effective for the exploitation and use of this type of copper–tin resource in Tajikistan.

Keywords: mushistonite; copper–tin tailings; flotation; benzohydroxamic acid

1. Introduction Tajikistan has abundant mineral resources. More than 400 different mineral deposits have been confirmed through systematic geological investigations. The main deposits of tin ores are found within the Zarafshan–Hisar geological and structural zone, in Central and South Pamir of Tajikistan. The tin resources total about 73 thousand tons [1]. The primary raw material deposit of tin in the Zarafshan–Hisar zone is found in Mushiston. This unique species of tin ore was originally discovered in the Penjikent region of Tajikistan by geologists in the 1960s, but was not named at that time [2,3]. Later, in 1984, the deposit was officially called mushistonite, after the name of the place where it was first discovered in Penjikent, Tajikistan [4–6]. Mushistonite is a malachite-green mineral with a dull luster, and recognized as a copper–tin– hydroxide mineral. Its eneral molecular formula is (Cu,Fe,Zn)Sn(OH)6, with a framework of corner- 0 linked BB (OH)6 octahedra [7]. Mushistonite has also been found in Sichuan, China and South Dakota, U.S., with slightly different chemical element content [5,8,9]. Mushistonite is a rare mineral, distributed in several areas around the world, and is not exploited on a large commercial scale currently. Few research reports have studied the exploitation and use of mushistonite [10,11]. In recent years, some companies and researchers have tentatively exploited the mine and established a small-scale minerals concentration factory. Coarse particle size fractions of cassiterite, chalcopyrite, and malachite were recovered by gravity separation using shaking tables.

Minerals 2017, 7, 242; doi:10.3390/min7120242 www.mdpi.com/journal/minerals Minerals 2017, 7, 242 2 of 11 Minerals 2017, 7, 242 2 of 11

The factory attempted to recover the fine particle size fraction of copper and tin minerals, including mushistoniteThe factory attempted, using the to sulfidation recover the flotation fine particle method, size but fraction failed. of copper and tin minerals, including mushistonite,Research usingstudies the have sulfidation indicated flotation that oxide method, minerals but failed., such as cassiterite and scheelite, could establishResearch hydroxylated studies have surfaces indicated in the that flotation oxide minerals, pulp. Benzohydroxamic such as cassiterite and acid scheelite, (BHA) couldand lead establish ions reacthydroxylated with the surfaces hydroxyl in groups the flotation and co pulp.-absorb Benzohydroxamic on the hydroxylated acid (BHA) surface. and leadAs a ions result, react cassiterite with the andhydroxyl scheelite groups could and be co-absorb selectively on recovered the hydroxylated by flotation surface., using As BHA a result, as a cassiterite collector andand scheelitelead nitrate could as anbe selectivelyactivator [12 recovered–16]. According by flotation, to the using chemical BHA ascomposition a collector andand leadcrystal nitrate structure as an activatorof mushistonite[12–16]. Accordingmentioned toabove, the chemicalhydroxyl compositiongroups are present and crystal within structure the mineral’s of mushistonite own natural mentioned structure. above, The strengthhydroxyl of groups the hydrogen are present bonds within betwee the mineral’sn hydroxyl own naturalgroups structure.is weak, Theand strength easily broken of the hydrogen to form hydroxylatedbonds between surfaces hydroxyl with groups grinding. is weak, We andconsidered easily broken that hydroxyl to form hydroxylatedgroups on the surfaces surface with of mushistonitegrinding. We could considered react with that hydroxylBHA and groupslead ions, on and the surfacebe selectively of mushistonite recovered couldby flotation. react with Based BHA on thisand hypothe lead ions,sis and, process be selectively mineralogy recovered and the by flotation flotation. of Based copper on–tin this tailings hypothesis, were processstudied mineralogy to recover mushistonite.and the flotation Experiment of copper–tinal results tailings confirmed were studied that flotation to recover of copper mushistonite.–tin tailings Experimental, using BHA results as the collectorconfirmed, was that feasible flotation. It of provid copper–tined an tailings,economical using and BHA realistic as the m collector,ethod for was the feasible.use of mushistonite. It provided an economical and realistic method for the use of mushistonite. 2. Materials and Methods 2. Materials and Methods 2.1. Minerals and Reagents 2.1. Minerals and Reagents The raw ore samples were collected from the crushing workshop of the Yukuang Panjakent copperThe–tin raw mine ore concentration samples were plant collected in fromNorthwest the crushing Tajikistan. workshop The samples of the Yukuangused for Panjakentflotation experimentscopper–tin mine were concentration obtained from plant the in Northwestshaking table Tajikistan. tailings The of samplesthis plant used (Figure for flotation 1). Toexperiments ensure the wererepresentativeness obtained from of the the shaking samples, table sampling tailings of work this plantwas (Figureperformed1). To during ensure the the representativenessnormal producing periodof the samples,in the plant. sampling The sampling work was work performed was sustained during for the 30 min normal each producing time, one time period every in thehalf plant. day, forThe three sampling continuous work was days. sustained All samples for 30 min were each fully time, mixed one time after every being half dried day, forin threethe natural continuous air environment.days. All samples were fully mixed after being dried in the natural air environment. The collector BHA waswas 98%98% pure,pure,and and other other reagents reagents used used in in this this study, study including, including lead lead nitrate nitrate as asactivator, activator, sodium sodium carbonate carbonate and and hydrochloric hydrochloric acid asacid pH as modifiers, pH modifiers, carboxymethyl carboxymethyl cellulose cellulose (CMC) (CMC)and sodium and sodium fluorosilicate fluorosilicate (SSF) as (SSF) depressants, as depressants as well, asas terpineolwell as terpineol as a frother, as a were frother all analytical, were all analyticalgrade. All reagentsgrade. All were reagents separately were dissolved separately into deionizeddissolvedwater into todeionized prepare thewater compound to prepare solutions the compoundused in each solutions test. used in each test.

FFigureigure 1. SSamplesamples used used for for flotation flotation test testss were obtained from the shaking table tailings.tailings.

22.2..2. Analysis Methods Mineral identifications identifications of the raw ore were performed using X-rayX-ray diffraction, transmission polarized optical microscopy, and scanning electron microscopy (SEM) with energy dispersive analytical attachment attachment.. Quantitative Quantitative mineralogical mineralogical composition compositionss could could also also be calculated calculated,, but using special methods [17,18 [17,18]].. Chemical Chemical compositions compositions of of raw raw ore ore and flotation flotation ore samples were detected with X-rayX-ray fluorescencefluorescence spectrometry spectrometry and and chemical chemical analysis, analysis which, which was wa sufficients sufficient to evaluate to evaluate the result the resultof flotation. of flotation.

Minerals 2017, 7, 242 3 of 11

2.3. Flotation Experiments Flotation experiments on the gravity process tailings were performed in a series of XFG type flotation machines (Jilin Prospecting Machinery Factory, Changchun, China). Single factor experiments were performed in a 3.0 L flotation cell, while open circuit and closed circuit experiments were conducted in the 3.0 L, 1.0 L, 500 mL, 250 mL, and 100 mL flotation cells. A 1000 g dosage of ore sample was treated for each experiment. Flotation parameters were controlled at a pulp temperature of 293 K and a stirring speed of 1910 r/min. After each flotation experiment, the products were collected, filtered, dried, weighed, and detected by Axios X-ray fluorescence spectrometry (PANalytical B.V., Almelo, The Netherlands), after which the recovery rates of Sn and Cu were calculated.

3. Results and Discussion

3.1. Mineralogical Analysis Mineral composition, chemical formula, specific gravity (a range of values found in the literature), and dissemination size of the raw ore are listed in Table1. Valuable minerals in the raw ore included mushistonite, cassiterite, stannite, malachite, and azurite. Gangue minerals included dolomite, quartz, calcite, talcum, limonite, sericite, and other impurities.

Table 1. Mineral composition, chemical formula, specific gravity, and dissemination size of the raw ore.

Mineral Composition Chemical Formula Specific Gravity Dissemination Size (mm)

Mushistonite (Cu,Fe,Zn)Sn(OH)6 4.0–4.4 0.01–0.5 Cassiterite SnO2 6.8–7.0 0.014–0.2 Stannite Cu2FeSnS4 4.3–4.5 0.004–0.6 Malachite Cu2[CO3](OH)2 3.9–4.0 0.004–0.8 Azurite Cu3[CO3]2(OH)2 3.7–3.9 0.01–0.3 Dolomite CaMg(CO3)2 2.8–2.9 0.06–0.25 Quartz SiO2 2.2–2.6 >0.004 Calcite CaCO3 2.7 >0.004 Talcum Mg3[Si4O10](OH)2 2.6–2.8 0.004–0.2 Limonite Fe2O3·nH2O 3.3–4.0 0.004–0.2 Sericite K{Al2[AlSi3O10](OH)2} 2.6–2.7 0.004–0.1 Others - - -

Tin was distributed among three different minerals: mushistonite, cassiterite, and stannite. The total grade of Sn was 0.65%. As observed in transmission polarized optical microscopy DM4 (Leica Microsystems Inc., Wetzlar, Germany), mushistonite was the result of stannite alteration, and presented with a yellowish-brown color and a greasy luster. It had a homogeneous colloidal distribution and a dense disseminated structure along the crevice of the stannite, mixing with malachite, azurite, and cassiterite, as shown in Figure2a,c,e. Stannite presented with a black color and a metallic luster, and was distributed in mushistonite and malachite, as illustrated in Figure2a,b. Cassiterite presented with a brownish-black to light brown color, and had a greasy luster with an aggregate or scattered disseminated distribution, as exhibited in Figure2c. Copper was mainly distributed among four different minerals: malachite, azurite, mushistonite, and stannite. The total grade of Cu was 0.66%. Malachite presented with an emerald green color and a glassy luster. It had an inhomogeneous, disseminated distribution when confounding with stannite, or an aggregate, thin vein-type distribution along the crevice of quartz, as shown in Figure2b,d. Azurite presented with an azure color and a glassy luster, and had an aggregate distribution when confounding with mushistonite and quartz, as exhibited in Figure2e. The main gangue minerals were dolomite, quartz, and calcite. The gangue mineral with the highest content in the raw ore was dolomite, with a mass content of 61.20%. Dolomite presented with a light gray color and a glassy luster, and its crystal was turbid. It had a hypidiomorphic granular texture, and the granules were closely inlaid with each other, as shown in Figure2f. Quartz presented MineralsMinerals2017, 20177, 242, 7, 242 4 of 114 of 11

a light gray color and a glassy luster, and its crystal was turbid. It had a hypidiomorphic granular withtexture a white, and color the andgranule a glassys were closely luster. inlaid It existed with each as veins, other, includingas shown ingranules Figure 2f. ofQuartz varying presented sizes and inhomogeneouswith a white distributions,color and a glassy as shown luster. inIt existed Figure 2ase,f. veins Calcite, including also presented granules of with varying a white sizes color and and a glassyinhomogeneous luster. It had distribution an allotriomorphics, as shown in granular Figure 2e texture,f. Calcite and also an presented aggregate with vein-type a white color distribution and alonga theglassy crevice luster of. It the had raw an ore,allotriomorphic as exhibited granular in Figure texture2f. and an aggregate vein-type distribution along the crevice of the raw ore, as exhibited in Figure 2f.

Figure 2. Optical microscope images of the raw ore, dissemination characteristic of main minerals. Figure 2. Optical microscope images of the raw ore, dissemination characteristic of main minerals. Mineralogical analysis of the raw ore indicated that the dissemination size of the copper–tin Mineralogicalminerals ranged analysis widely, from of the 4 μ rawm to ore more indicated than 200 thatμm. theCoarse dissemination particles of copper size of–tin the minerals copper–tin mineralscould ranged be recovered widely, using from the 4 µ gravitym to more concentration than 200 µprocessm. Coarse. Cassiterite, particles malachite, of copper–tin and azurite minerals were could partially recovered by shaking table concentrators in the producing plant. The recovery rate of Sn be recovered using the gravity concentration process. Cassiterite, malachite, and azurite were partially through this process was around 10.91%, and that of Cu was 13.89%, which are both low overall recovered by shaking table concentrators in the producing plant. The recovery rate of Sn through this recovery rates. Therefore, a new method to improve the recovery rates of Sn and Cu in the shaking processtable was tailings around was 10.91%,required. and that of Cu was 13.89%, which are both low overall recovery rates. Therefore,Yiel ad, new grade method, and distribution to improve of the Cu recoveryand Sn in rateseach ofparticle Sn and size Cu fraction in the of shaking the shaking table table tailings was required.tailings are listed in Table 2. As the particle size decreased, the grades of both Cu and Sn increased. Yield,Although grade, the yield and of distribution the +75 μm offraction Cu and was Sn over in 40%, each both particle the grade size fractionand the distribution of the shaking of Cu table tailingsand are Sn w listedere low in Tablein this2 fraction,. As the which particle indicated size decreased, that the copper the grades–tin mine ofrals both in this Cu co andarse Sn fraction increased. Althoughmostly the amassed yield ofto the the shaking +75 µm table fraction concentrate was overs. Similarly, 40%, both although the grade the grade ands of the Cu distribution and Sn in the of Cu and Sn were low in this fraction, which indicated that the copper–tin minerals in this coarse fraction mostly amassed to the shaking table concentrates. Similarly, although the grades of Cu and Sn in the −18 µm fraction were higher than in other fractions, the yield of this fraction was only 4.41%, and the distribution of Cu and Sn in this fraction were both less than 8%. Hence, over 88% of tin minerals Minerals 2017, 7, 242 5 of 11 Minerals 2017, 7, 242 5 of 11 −18 μm fraction were higher than in other fractions, the yield of this fraction was only 4.41%, and the distribution of Cu and Sn in this fraction were both less than 8%. Hence, over 88% of tin minerals and and over 86% of copper minerals in the shaking table tailings were distributed in the size fractions of over 86% of copper minerals in the shaking table tailings were distributed in the size fractions of 18 µm to 75 µm, for which the shaking table concentrator was ineffective. 18 μm to 75 μm, for which the shaking table concentrator was ineffective. Table 2. Grade and distribution of Cu and Sn in each particle size fraction of shaking table tailings. Table 2. Grade and distribution of Cu and Sn in each particle size fraction of shaking table tailings.

SnSn Cu Cu SizeSize Fraction Fraction (µ m)(μm) YieldYield (%) (%) GradeGrade (%)(%) Distribution Distribution (%)(%) Grade Grade ( (%)%) Distribution Distribution (% (%)) +75+75 40.2640.26 0.054 3.75 3.75 0.081 0.081 5.76 5.76 −75−75 + + 38 38 24.3724.37 0.89 37.41 37.41 0.86 0.86 37.04 37.04 −38 + 18 30.96 0.96 51.27 0.91 49.79 −38 + 18 30.96 0.96 51.27 0.91 49.79 −18 + 10 3.66 0.99 6.25 0.95 6.14 −−1810 + 10 0.753.66 0.99 1.02 6.25 1.32 0.95 0.96 6.14 1.27 Total−10 1000.75 1.02 0.58 1.32 100 0.96 0.57 1.27 100 Total 100 0.58 100 0.57 100 The content percentage of the elements in the shaking table tailings are listed in Table3. The contentThe content of Sn andpercentage Cu were of 0.58% the elements and 0.57%, in respectively.the shaking Zntable was tailings only distributed are listed in mushistoniteTable 3. The insteadcontent ofof inSn an and independent Cu were 0.58% zinc mineral,and 0.57% and, respectively. could not be Zn separately was only recovered. distributed in mushistonite instead of in an independent zinc mineral, and could not be separately recovered. Table 3. Percentage contents of main elements in the shaking table tailings. Table 3. Percentage contents of main elements in the shaking table tailings.

Component Sn Cu Zn CaO SiO2 MgO Al2O3 FeO K2OS Component Sn Cu Zn CaO SiO2 MgO Al2O3 FeO K2O S ContentContent (%) (%) 0.580.58 0.570.57 0.48 27.76 27.76 22.59 22.59 12.08 12.08 3.84 1.58 0.83 0.83 0.12 0.12

3.2. Flotation Experiments of Copper–TinCopper–Tin MineralsMinerals Based on mineralogical analysis, the valuablevaluable minerals being recovered were distributed in the 18–7518–75 µμmm sizesize fractions.fractions. The liberationliberation degree of tin-tin- and copper-bearingcopper-bearing minerals inin this fraction was high,high, and no further grindinggrinding waswas necessary.necessary. The effectseffects of each reagent dosage on copper–tincopper–tin minerals flotationflotation were investigated following thethe processprocess shownshown inin FigureFigure3 3.. TheThe pulppulp temperaturetemperature waswas controlledcontrolled atat 293293 K.K. CMCCMC andand SSFSSF were were added into the pulp simultaneously andand mixedmixed forfor 55 min.min. LeadLead nitrate and BHA were added into the pulppulp simultaneouslysimultaneously and mixed for 4040 min.min. The terpineol dosage was moderated,moderated, according to the foamfoam volume,volume, andand mixedmixed for for 3 3 min. min. Air Air filling filling and and flotation flotation time time was was 5 min.5 min. The The grade grade and and recovery recovery of Snof Sn and and Cu Cu in thein the concentrates concentrates and and tailings tailings were were detected detected and and calculated. calculated.

Figure 3.3. Rough flotation flotation flowchart flowchart of the shaking table tailings (BHA—Benzohydroxamic(BHA—Benzohydroxamic acid; CMC—carboxymethylCMC—carboxymethyl cellulose;cellulose; SSF—sodiumSSF—sodium fluorosilicate).fluorosilicate).

3.2.1.3.2.1. Effect of Pulp pH Value Sodium carbonatecarbonate and and hydrochloric hydrochloric acid acid were were used used to adjust to adjust the pulp the pH pulp value pH [19 value]. The [19] flotation. The recoveryflotation recovery and grade and of grade Sn and of Sn Cu, and as Cu a function, as a function of pulp of pHpulp value pH value from from 5.0 to 5.0 9.0, to 9.0 are, are exhibited exhibited in in Figure 4. As the pulp pH value was adjusted to approximately 7.2, the recovery and grade of Sn

Minerals 2017, 7, 242 6 of 11

Minerals 2017, 7, 242 6 of 11 Minerals 2017, 7, 242 6 of 11 Figure4. As the pulp pH value was adjusted to approximately 7.2, the recovery and grade of Sn both both reached their peak values, with a recovery of 58.53% and a grade of 4.66%. At the same pH reachedboth reached their peaktheir values,peak values with, awith recovery a recovery of 58.53% of 58.53% and a and grade a ofgrade 4.66%. of 4.66% At the. At same the pHsame value, pH value, the recovery and grade of Cu were 78.92% and 6.18%, respectively. As the pulp pH value thevalue recovery, the recovery and grade and of grade Cu were of Cu 78.92% were and78.92% 6.18%, and respectively. 6.18%, respectively. As the pulp As pHthe valuepulp increasedpH value increased from 7.2 to 7.5, the recovery and grade of Cu both reached their peak values, with a fromincreased 7.2 to from 7.5, the 7.2 recovery to 7.5, the and recovery grade of Cuand both grade reached of Cu their both peak reached values, their with peak a recovery values of, 80.58%with a recovery of 80.58% and a grade of 6.23%. The recovery and grade of Sn slightly decreased, but the andrecovery a grade of 80.58% of 6.23%. and The a grade recovery of 6.23% and grade. The ofrecovery Sn slightly and decreased, grade of Sn but slightly the recovery decreas remaineded, but the at recovery remained at 57.22% and the grade at 4.52%. Experiments on the pulp pH value indicated 57.22%recovery and remained the grade at at57.22% 4.52%. and Experiments the grade at on 4.52%. the pulp Experiments pH value on indicated the pulp that pH the value optimum indicated pH that the optimum pH value range for both elements was 7.2–7.5. valuethat the range optimum for both pH elements value range was for 7.2–7.5. both elements was 7.2–7.5.

Figure 4. Flotation recovery and grade of Sn and Cu as a function of pulp pH value (carboxymethyl Figure 4.4. FlotationFlotation recovery and grade of SnSn andand CuCu asas aa functionfunction ofof pulppulp pHpH valuevalue (carboxymethyl(carboxymethyl cellulose (CMC) 350 g/t; sodium fluorosilicate (SSF) 250 g/t; lead nitrate 200 g/t; Benzohydroxamic cellulosecellulose (CMC)(CMC) 350350 g/t;g/t; sodiumsodium fluorosilicatefluorosilicate (SSF)(SSF) 250250 g/t;g/t; leadlead nitratenitrate 200200 g/t;g/t; BenzohydroxamicBenzohydroxamic acid (BHA) 2000 g/t; Terpineol 25 g/t). acidacid (BHA)(BHA) 20002000 g/t;g/t; Terpineol Terpineol 25 g/t) g/t).. 3.2.2. Effects of Depressant CMC 3.2.2.3.2.2. Effects of DepressantDepressant CMCCMC CMC was reported to be an effective depressant for carbonate minerals, including dolomite and CMC was reported to be an effective depressant for carbonate minerals, including dolomite and calciteCMC [20– was22]. reportedThe flotation to be recovery an effective anddepressant grade of Sn for and carbonate Cu, as a minerals,function of including dosage dolomiteof CMC from and calcite [20–22]. The flotation recovery and grade of Sn and Cu, as a function of dosage of CMC from calcite0 to 450 [20 g–/t22, are]. The exhibited flotation in recoveryFigure 5. and When grade the of CMC Sn and dosage Cu, aswas a function 300 g/t, ofthe dosage recovery of CMC of Cu from reached 0 to 0 to 450 g/t, are exhibited in Figure 5. When the CMC dosage was 300 g/t, the recovery of Cu reached 450its peak g/t, arevalue exhibited of 80.81 in%. Figure As 5the. When dosage the of CMC CMC dosage increased was 300 to g/t,350 theg/t, recoverythe grade of Cuand reached recovery its of peak Sn its peak value of 80.81%. As the dosage of CMC increased to 350 g/t, the grade and recovery of Sn valueand the of 80.81%.grade of As Cu the reached dosage high of CMC value increaseds, at 4.52%, to 350 57.22% g/t, the and grade 6.23% and, respectively. recovery of SnThe and recovery the grade of and the grade of Cu reached high values, at 4.52%, 57.22% and 6.23%, respectively. The recovery of ofCu Cu was reached still 80.58%, high values, with atno 4.52%, significa 57.22%nt variation and 6.23%, as respectively.the dosage Theof CMC recovery increased of Cu wasfrom still 300 80.58%, g/t to Cu was still 80.58%, with no significant variation as the dosage of CMC increased from 300 g/t to with350 g/t. no Experiments significant variation on the aseffects the dosage of the depressant of CMC increased CMC indicated from 300 g/tthat tothe 350 optimum g/t. Experiments dosage of on CMC the 350 g/t. Experiments on the effects of the depressant CMC indicated that the optimum dosage of CMC effectswas 350 of g/t. the depressant CMC indicated that the optimum dosage of CMC was 350 g/t. was 350 g/t.

Figure 5.5. Flotation recovery and grade of Sn and Cu as a functionfunction of dosage of CMC (pH = 7.2–7.5;7.2–7.5; Figure 5. Flotation recovery and grade of Sn and Cu as a function of dosage of CMC (pH = 7.2–7.5; SSF 250 g/t; l leadead nitrate nitrate 200 200 g/t; g/t; BHA BHA 2000 2000 g/t; g/t; terpineol terpineol 25 25 g/t) g/t).. SSF 250 g/t; lead nitrate 200 g/t; BHA 2000 g/t; terpineol 25 g/t).

Minerals 2017, 7, 242 7 of 11 Minerals 2017, 7, 242 7 of 11

Minerals 2017, 7, 242 7 of 11 3.2.3.3.2. Effects3. Effects of of Depressant Depressant SSF SSF 3.2.3.SSF Effects was of applied Depressant as an SSFeffective depressant for silicon-containing minerals (including quartz) in SSF was applied as an effective depressant for silicon-containing minerals (including quartz) in flotation [23–25]. The flotation recovery and grade of Sn and Cu as a function of SFF dosage from 0 flotationSSF [23 wa–25s ].applied The flotation as an effective recovery depress andant grade for ofsilicon Sn and-containing Cu as a minerals function (including of SFF dosage quartz from) in 0 to 350 g/t are shown in Figure 6. When the dosage of SSF was 250 g/t, the recovery and grade of Sn, to 350flotation g/t are [23 shown–25]. The in f Figurelotation6. recovery When the and dosage grade ofof SSFSn and was Cu 250 as g/t, a function the recovery of SFF anddosage grade from of 0 Sn, as well as the recovery of Cu, all reached their peak values of 57.22%, 4.52%, and 80.58%, respectively. to 350 g/t are shown in Figure 6. When the dosage of SSF was 250 g/t, the recovery and grade of Sn, as wellAs the as dosage the recovery of SSF increased of Cu, all from reached 250 g/t their to 300 peak g/t, values the grade of 57.22%, of Cu reached 4.52%, andits peak 80.58%, value respectively. of 6.40%, as well as the recovery of Cu, all reached their peak values of 57.22%, 4.52%, and 80.58%, respectively. Asa then inconspicuous dosage of SSF increase increased, but from the 250 recovery g/t to300 and g/t, grade the gradeof Sn and of Cu the reached recovery its peakof Cu value all sharply of 6.40%, As the dosage of SSF increased from 250 g/t to 300 g/t, the grade of Cu reached its peak value of 6.40%, andecrea inconspicuoussed. Experiments increase, buton the the effects recovery of andthe depressant grade of Sn SSF and indicated the recovery that of the Cu optimum all sharply dosage decreased. of an inconspicuous increase, but the recovery and grade of Sn and the recovery of Cu all sharply ExperimentsSSF was 250 on g/t. the effects of the depressant SSF indicated that the optimum dosage of SSF was 250 g/t. decreased. Experiments on the effects of the depressant SSF indicated that the optimum dosage of SSF was 250 g/t.

Figure 6. Flotation recovery and grade of Sn and Cu as a function of dosage of SSF (pH = 7.2–7.5; CMC Figure 6. Flotation recovery and grade of Sn and Cu as a function of dosage of SSF (pH = 7.2–7.5; 350 g/t; lead nitrate 200 g/t; BHA 2000 g/t; terpineol 25 g/t). CMCFigure 350 6. g/t; Flotation lead nitrate recovery 200 and g/t; grade BHA of 2000 Sn and g/t; Cu terpineol as a function 25 g/t). of dosage of SSF (pH = 7.2–7.5; CMC 350 g/t; lead nitrate 200 g/t; BHA 2000 g/t; terpineol 25 g/t). 3.2.4. Effects of Activator Lead Nitrate 3.2.4. Effects of Activator Lead Nitrate 3.2.4.Lead Effects nitrate of Activator was applied Lead asNitrate the most common and effective activator for tin or tungsten oxide mineralsLead nitrate (including was cassiterite applied as and the scheelite most common) in flotation and [26 effective–28]. The activator flotation for recovery tin or tungsten and grade oxide of Lead nitrate was applied as the most common and effective activator for tin or tungsten oxide mineralsSn and (includingCu, as a function cassiterite of the and dosage scheelite) of lead in nitrate flotation from [ 260 to–28 300]. Theg/t, are flotation shown recoveryin Figure and7. When grade minerals (including cassiterite and scheelite) in flotation [26–28]. The flotation recovery and grade of of Snthe anddosage Cu, of as lead a function nitrate was of the150 dosageg/t, the ofrecovery lead nitrate and grade from of 0 Cu to both 300 g/t,reached are showntheir peak in Figurevalues 7. Sn and Cu, as a function of the dosage of lead nitrate from 0 to 300 g/t, are shown in Figure 7. When Whenof 82.00% the dosage and 6.27% of lead, respectively. nitrate was As 150 the g/t, dosage the of recovery lead nitrate and increased grade of Cufrom both 150 reachedg/t to 200 their g/t, the peak the dosage of lead nitrate was 150 g/t, the recovery and grade of Cu both reached their peak values valuesrecovery of 82.00% and grade and 6.27%,of Sn both respectively. reached their As thepeak dosage values of of lead 57.22% nitrate and increased4.52%, respectively from 150. g/tThe to of 82.00% and 6.27%, respectively. As the dosage of lead nitrate increased from 150 g/t to 200 g/t, the recovery and grade of Cu slightly decreased, but the variation was not significant. Experiments on 200recovery g/t, the and recovery grade and of gradeSn both of reached Sn both reachedtheir peak their value peaks of values 57.22% of 57.22%and 4.52% and, respectively 4.52%, respectively.. The the effects of the activator lead nitrate indicated that the optimum dosage of lead nitrate was 200 g/t. Therecovery recovery and and grade grade of of Cu Cu slightly slightly decreas decreased,ed, but but the the variation variation was was no nott significa significant.nt. Experiments Experiments on on thethe effects effects of of the the activator activator lead lead nitrate nitrate indicated indicated thatthat thethe optimum dosage dosage of of lead lead nitrate nitrate was was 200 200 g/t. g/t.

Figure 7. Flotation recovery and grade of Sn and Cu as a function of the dosage of lead nitrate (pH = 7.2~7.5; CMC 350 g/t; SSF 250 g/t; BHA 2000 g/t; terpineol 25 g/t). FigureFigure 7. 7.Flotation Flotation recovery and and grade grade of ofSn Snand and Cu Cuas a asfunction a function of theof dosage the dosage of lead ofnitrate lead (pH nitrate = (pH7.2~7.5; = 7.2~7.5; CMC CMC 350 g/t; 350 SSF g/t; 250 SSF g/t; 250 BHA g/t; 2000 BHA g/t; 2000 terpineol g/t; terpineol 25 g/t). 25 g/t).

Minerals 2017, 7, 242 8 of 11 Minerals 2017, 7, 242 8 of 11

3.2.5.3.2.5. Effects ofof CollectorCollector BHABHA BHABHA hashas been been used used as anas effectivean effective collector collector for tin for or tungstentin or tungsten oxide minerals, oxide minerals including, cassiteriteincluding andcassiterite scheelite, and andscheelite has begun, and has to bebegun used to in be the used cassiterite in the cassiterite and scheelite and scheelite flotation flotation industries industr [29–31ies]. The[29– flotation31]. The f recoverylotation recovery and grade and of Sngrade and of Cu, Snas and afunction Cu, as a offunction the dosage of the of dosage BHA from of BHA 0 to from 2500 g/t,0 to is2500 shown g/t, inis Figureshown8 .in When Figure the 8. dosage When ofthe BHA dosage was of 1500 BHA g/t, was the grade1500 g/t, of Cuthe reached grade of its Cu peak reached value ofits 6.31%.peak value As the of 6.31%. dosage As of the BHA dosage increased of BHA from increased 1500 g/t from to 20001500 g/t,g/t to the 2000 grade g/t, the of Sngrade reached of Sn itsreached peak valueits peak of 4.52%,value andof 4.52% the grade, and of the Cu grade decreased of Cu slightly, decreased although slightly not significantly., although no Thet significa recoveryntly of. bothThe Snrecovery and Cu of had both an Sn increasing and Cu trendhad an as increasing the dosage trend of BHA as the increased, dosage andof BHA reached increased high values,, and reached 57.22% andhigh 80.58%, values, respectively,57.22% and 80.58% at the dosage, respectively, of 2000 at g/t. the Thedosage dosage of 2000 of BHA g/t. T increasedhe dosage continuouslyfrom of BHA increased 2000continuously g/t to 2500from g/t; 2000 the g/t recovery to 2500 of g/t Sn; andthe recovery Cu exhibited of Sn no and significant Cu exhibited variation, no significa but thent grade variation of Cu, andbut the Sn droppedgrade of obviously.Cu and Sn Experimentsdropped obviously on the. effectsExperiments of BHA on indicated the effects that of theBHA optimum indicated dosage that the of BHAoptimum was 2000dosage g/t. of BHA was 2000 g/t.

FigureFigure 8.8. Flotation recovery andand gradegrade ofof SnSn andand CuCu asas a function ofof the dosage ofof BHA (pH(pH = 7.2 7.2–7.5;–7.5; CMCCMC 350350 g/t;g/t; SSF 250 g/t; lead lead nitrate nitrate 200 200 g/t; g/t; terpineol terpineol 25 25 g/t) g/t)..

3.2.6.3.2.6. Closed Circuit FlotationFlotation TestsTests TheThe bench-scalebench-scale closedclosed circuit flotationflotation tests of thethe shakingshaking tabletable tailingstailings from TajikistanTajikistan werewere performedperformed basedbased onon singlesingle factorfactor experimentsexperiments andand openopen circuitcircuit tests,tests, withwith optimizedoptimized flotationflotation conditions.conditions. AfterAfter furtherfurther adjustmentsadjustments andand optimizingoptimizing thethe flotationflotation conditions,conditions, thethe closedclosed circuitcircuit flotationflotation flowchartflowchart of of the the shaking shaking table table tailings tailings was establishedwas establish followinged follow theing process the process shownin shown Figure 9in. ByFigure adopting 9. By single-stageadopting single roughing,-stage roughing, two-stage two scavenging,-stage scavenging and four-stage, and four cleaning-stage flotation, cleaning the flotation results, werethe results obtained were and obtained calculated and as calculated shown in Table as shown4. The in concentrate Table 4. The was concentrate a copper–tin was mixture a copper product–tin thatmixture contained product 13.28% that contain Sn, withed a 13.28% recovery Sn, rate with of a 61.56%,recovery and rate a Cuof 61.56%, content and of 18.51%, a Cu content with aof recovery 18.51%, ratewith of a 86.52%.recovery Experimental rate of 86.52%. results Experiment confirmedal thatresults selective confirmed flotation that of selectiv copper–tine flotation minerals, of copper including–tin mushistonite,minerals, including from the mushistonite shaking table, from tailings the inshaking Tajikistan, table using tailing BHAs in as Tajikistan the collector, using and BHA lead nitrateas the ascollector the activator, and lead in cooperationnitrate as the with activator other, depressants, in cooperati ison feasible. with other These depress resultsant provideds, is feasible a basis. These for theresults exploitation provided and a basis use offor this the typeexploitation of mineral and in use Tajikistan. of this type of mineral in Tajikistan.

Table 4.4. ResultsResults ofof thethe bench-scalebench-scale closedclosed circuitcircuit flotationflotation experimentsexperiments ofof shakingshaking tabletable tailings.tailings. Sn Cu Product Yield (%) Sn Cu Product Yield (%) Grade (%) Distribution (%) Grade (%) Distribution (%) Grade (%) Distribution (%) Grade (%) Distribution (%) Concentrates 2.71 13.28 61.57 18.51 86.52 ConcentratesFinal Tailings 2.7197.29 0.23 13.28 38.43 61.57 0.08 18.51 13.48 86.52 Final Tailings 97.29 0.23 38.43 0.08 13.48 Feed 100 0.58 100 0.57 100 Feed 100 0.58 100 0.57 100

Minerals 2017, 7, 242 9 of 11 Minerals 2017, 7, 242 9 of 11

Figure 9. The bench-scale closed circuit flotation flowchart of the shaking table tailings. Figure 9. The bench-scale closed circuit flotation flowchart of the shaking table tailings.

4. Conclusions 4. Conclusions Mineralogical analysis of the copper–tin minerals from the Mushiston deposit in Tajikistan Mineralogical analysis of the copper–tin minerals from the Mushiston deposit in Tajikistan indicated that tin mainly occurred in mushistonite, cassiterite, and stannite, while copper mainly indicated that tin mainly occurred in mushistonite, cassiterite, and stannite, while copper mainly occurred in mushistonite, malachite, azurite and stannite. The total grade of Sn was 0.65%, and that occurred in mushistonite, malachite, azurite and stannite. The total grade of Sn was 0.65%, and that of of Cu was 0.66%. The main gangue minerals were dolomite, quartz, and calcite. Dissemination size Cu was 0.66%. The main gangue minerals were dolomite, quartz, and calcite. Dissemination size of of the copper–tin minerals ranged widely, from 4 μm to more than 200 μm. the copper–tin minerals ranged widely, from 4 µm to more than 200 µm. Copper–tin minerals in coarse particles were partially recovered by shaking table concentrators Copper–tin minerals in coarse particles were partially recovered by shaking table concentrators in in a gravity concentration process, but with an overall low recovery rate. The valuable minerals to be a gravity concentration process, but with an overall low recovery rate. The valuable minerals to be recovered were mainly distributed in the 18 μm to 75 μm size fractions of the shaking table tailings. recovered were mainly distributed in the 18 µm to 75 µm size fractions of the shaking table tailings. Flotation was applied in this study, for the first time, to selectively recover the copper–tin minerals Flotation was applied in this study, for the first time, to selectively recover the copper–tin minerals of of shaking table tailings in Tajikistan. Results of the bench-scale closed circuit flotation tests indicated shaking table tailings in Tajikistan. Results of the bench-scale closed circuit flotation tests indicated that that selective flotation of copper–tin minerals, including mushistonite, using BHA as the collector selective flotation of copper–tin minerals, including mushistonite, using BHA as the collector and lead and lead nitrate as the activator, in cooperation with other depressants, is feasible. The resulting nitrate as the activator, in cooperation with other depressants, is feasible. The resulting concentrate concentrate contained 13.28% Sn with a recovery of 61.56%, and the concentrate contained 18.51% of contained 13.28% Sn with a recovery of 61.56%, and the concentrate contained 18.51% of Cu with Cu with a recovery of 86.52%. a recovery of 86.52%. Flotation of copper–tin minerals from the shaking table tailings in Tajikistan is an effective Flotation of copper–tin minerals from the shaking table tailings in Tajikistan is an effective method method for the exploitation and use of this type of resource. Separation of Cu and Sn in the for the exploitation and use of this type of resource. Separation of Cu and Sn in the mushistonite will mushistonite will be the next step of research, based on the work completed by Yang et al. [32]. be the next step of research, based on the work completed by Yang et al. [32].

Minerals 2017, 7, 242 10 of 11

Acknowledgments: This research was financially supported by the National 111 Project (No. B14034), Collaborative Innovation Center for Clean and Efficient utilization of Strategic Metal Mineral Resources, and Innovation Driven Plan of Central South University (No. 2015CX005). Author Contributions: Yuehua Hu and Wei Sun provided the research ideas and guidance; Lei Sun designed and performed the experiments, and wrote the paper as well; Zhiyong Gao revised the manuscripts; Mengjie Tian contributed the experiment performances. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Rahimov, F.K.; Mamadjonov, Y.M. Mineral resource potential of Tajikistan: As an important component of sustainable development of the silk road economic belt. J. Resour. Ecol. 2015, 6, 125–128. [CrossRef] 2. Fleischer, M.; Cabri, L.J.; Chao, G.Y.; Pabst, A. New mineral names. Am. Mineral. 1980, 65, 1065–1070. 3. Fleischer, M.; Cabri, L.J.; Chao, G.Y.; Mandarino, J.A.; Pabst, A. New mineral names. Am. Mineral. 1982, 67, 1074–1082. 4. Dunn, P.J.; Ferraiolo, J.A.; Fleischer, M.; Gobel, V.; Grice, J.D.; Langley, R.H.; Shigley, J.E.; Vanko, D.A.; Zilczer, J.A. New mineral names. Am. Mineral. 1985, 70, 1329–1335. 5. Dunn, P.J.; Roberts, W.L. Cuprocassiterite discredited as mushistonite and an unnamed tin mineral from the Etta mine. Mineral. Rec. 1986, 17, 383. 6. Ivanov, O.P.; Yeremenko, L.Y.; Voronov, A.I.; Zorin, Y.M.; Kuzovenko, A.I.; Khar’Kevich, K.A.; Kanoatov, S.I. Mineralogy and technology of new economic types of tin ores. Int. Geol. Rev. 1993, 35, 603–612. [CrossRef]

7. Welch, M.D.; Wunder, B. A single-crystal X-ray diffraction study of the 3.65 Å-phase MgSi(OH)6, a high-pressure hydroxide perovskite. Phys. Chem. Miner. 2012, 39, 693–697. [CrossRef] 8. Ottens, B. Xuebaoding: Pingwu county, Sichuan province, China. Mineral. Rec. 2005, 36, 45–47. 9. Zhang, D.; Peng, J.; Coulson, I.M.; Hou, L.; Li, S. Cassiterite U–Pb and muscovite40Ar–39Ar age constraints on the timing of mineralization in the Xuebaoding Sn–W–be deposit, western China. Ore Geol. Rev. 2014, 62, 315–322. [CrossRef] 10. Levine, R.M.; Bond, A.R. Tin reserves and production in the Russian federation. Int. Geol. Rev. 1994, 36, 301–310. [CrossRef] 11. Yuan, W.; Yang, Y.; Jin, Z.; Liu, T.; Li, H. Review and prospection on process mineralogy research. Yunnan Metall. 2013, 42, 1–4. 12. Vlcek, L.; Zhang, Z.; Machesky, M.L.; Fenter, P.; Rosenqvist, J.; Wesolowski, D.J.; Anovitz, L.M.; Predota, M.; Cummings, P.T. Electric double layer at metal oxide surfaces: Static properties of the cassiterite-water interface. Langmuir 2007, 23, 4925–4937. [CrossRef][PubMed] 13. Mamontov, E.; Vlcek, L.; Wesolowski, D.J.; Cummings, P.T.; Wang, W.; Anovitz, L.M.; Rosenqvist, J.; Brown, C.M.; Garcia Sakai, V. Dynamics and structure of hydration water on rutile and cassiterite nanopowders studied by quasielastic neutron scattering and molecular dynamics simulations. J. Phys. Chem. C 2007, 111, 4328–4341. [CrossRef] 14. Tian, M.; Hu, Y.; Sun, W.; Liu, R. Study on the mechanism and application of a novel collector-complexes in cassiterite flotation. Colloids Surf. A 2017, 522, 635–641. [CrossRef] 15. Gao, Z.; Li, C.; Sun, W.; Hu, Y. Anisotropic surface properties of calcite: A consideration of surface broken bonds. Colloids Surf. A 2017, 520, 53–61. [CrossRef] 16. Gao, Y.; Gao, Z.; Sun, W.; Hu, Y. Selective flotation of scheelite from calcite: A novel reagent scheme. Int. J. Miner. Process. 2016, 154, 10–15. [CrossRef] 17. Piga, L. Thermogravimetry of a kaolinite-alunite ore. Thermochim. Acta 1995, 265, 177–187. [CrossRef] 18. Ippolito, N.M.; Belardi, G.; Piga, L. Determination of mineralogical composition of spent fluorescent powders by coupling ICP-spectroscopy and electronic microprobe analyses. TrAC Trends Anal. Chem. 2017, 94, 14–20. [CrossRef] 19. Xu, L.; Hu, Y.; Tian, J.; Wu, H.; Yang, Y.; Zeng, X.; Wang, Z.; Wang, J. Selective flotation separation of spodumene from feldspar using new mixed anionic/cationic collectors. Miner. Eng. 2016, 89, 84–92. [CrossRef] 20. Ejtemaei, M.; Gharabaghi, M.; Irannajad, M. A review of zinc oxide mineral beneficiation using flotation method. Adv. Colloid Interface 2014, 206, 68–78. [CrossRef][PubMed] Minerals 2017, 7, 242 11 of 11

21. Sun, W.; Ke, L.; Sun, L. Study of the application and mechanism of benzohydroxamic acid in the flotation of cassiterite. J. China Univ. Min. Technol. 2013, 10, 1367–1382. 22. Kangal, O.; Güney, A. Effects of modifying reagents on huntite flotation. Miner. Process. Extr. Metall. Rev. 2006, 28, 117–126. [CrossRef] 23. Pearse, M.J. An overview of the use of chemical reagents in mineral processing. Miner. Eng. 2005, 18, 139–149. [CrossRef] 24. Song, S.; Lopez-Valdivieso, A.; Martinez-Martinez, C.; Torres-Armenta, R. Improving fluorite flotation from ores by dispersion processing. Miner. Eng. 2006, 19, 912–917. [CrossRef] 25. Fuerstenau, D.W. Design and development of novel flotation reagents for the beneficiation of Mountain Pass rare-earth ore. Miner. Metall. Process. 2013, 30, 1–9. 26. Zhao, G.; Wang, S.; Zhong, H. Study on the activation of scheelite and wolframite by lead nitrate. Minerals 2015, 2, 247–258. [CrossRef] 27. Zhou, Y.; Tong, X.; Song, S.; Wang, X.; Deng, Z.; Xie, X. Beneficiation of cassiterite fines from a tin tailing slime by froth flotation. Sep. Sci. Technol. 2014, 49, 458–463. [CrossRef] 28. Houot, R.; Desbrosses, Y. Is the cassiterite contained in complex sulphide polymetallic ore recoverable? Int. J. Miner. Process. 1991, 32, 45–57. [CrossRef] 29. Wu, X.Q.; Zhu, J.G. Selective flotation of cassiterite with benzohydroxamic acid. Miner. Eng. 2006, 19, 1410–1417. [CrossRef] 30. Sun, L.; Hu, Y.; Sun, W. Effect and mechanism of octanol in cassiterite flotation using benzohydroxamic acid as collector. Trans. Nonferr. Met. Soc. 2016, 26, 3253–3257. [CrossRef] 31. Han, H.; Liu, W.; Hu, Y.; Sun, W.; Li, X. A novel flotation scheme: Selective flotation of tungsten minerals from calcium minerals using Pb–BHA complexes in Shizhuyuan. Rare Met. 2017, 36, 533–540. [CrossRef] 32. Yang, J.; Wu, Y.; Zhang, X. Study on separation of tin from a low-grade tin concentrate through leaching and low-temperature smelting processes. Miner. Process. Extr. Metall. 2014, 123, 228–233. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).