Accepted Manuscript

The utilization of waste by-products for removing silicate from mineral processing
wastewater via chemical precipitation

Jianhua Kang, Wei Sun, Yuehua Hu, Zhiyong Gao, Runqing Liu, Qingpeng Zhang,
Hang Liu, Xiangsong Meng

PII: S0043-1354(17)30709-1

DOI: 10.1016/j.watres.2017.08.047

Reference: WR 13170

To appear in: Water Research

Received Date: 15 May 2017

Revised Date: 11 August 2017

Accepted Date: 20 August 2017

Please cite this article as: Jianhua Kang, Wei Sun, Yuehua Hu, Zhiyong Gao, Runqing Liu,
Qingpeng Zhang, Hang Liu, Xiangsong Meng, The utilization of waste by-products for removing
silicate from mineral processing wastewater via chemical precipitation, Water Research (2017), doi:
10.1016/j.watres.2017.08.047

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Graphical abstract
 ACCEPTED MANUSCRIPT

1 The utilization of waste by-products for removing silicate from

2 mineral processing wastewater via chemical precipitation

3 Jianhua Kang, Wei Sun, Yuehua Hu, Zhiyong Gao, Runqing Liu, Qingpeng Zhang,

4 Hang Liu, Xiangsong Meng

5 School of Minerals Processing and Bioengineering, Central South University,

6 Changsha, 410083, China

7 Abstract

8 This study investigates an environmentally friendly technology that utilizes waste

9 by-products (waste acid and waste alkali liquids) to treat mineral processing

10 wastewater. Chemical precipitation is used to remove silicate from scheelite (CaWO4)

11 cleaning flotation wastewater and the waste by-products are used as a substitute for

12 calcium chloride (CaCl2). A series of laboratory experiments is conducted to explain

13 the removal of silicate and the characterization and formation mechanism of calcium

14 silicate. The results show that silicate removal reaches 90% when the Ca:Si molar ratio

15 exceeds 1.0. The X-ray diffraction (XRD) results confirm the characterization and

16 formation of calcium silicate. The pH is the key factor for silicate removal, and the

17 formation of polysilicic acid with a reduction of pH can effectively improve the silicate

18 removal and reduce the usage of calcium. The economic analysis shows that the

19 treatment costs with waste acid (0.63 $/m3) and waste alkali (1.54 $/m3) are lower than

 Corresponding author. E-mail address: sunmenghu@csu.edu.cn (W. Sun); hyh@csu.edu.cn (Y. Hu).
1
 ACCEPTED MANUSCRIPT

1 that of calcium chloride (2.38 $/m3). The efficient removal of silicate is confirmed by

2 industrial testing at a plant. The results show that silicate removal reaches 85% in the

3 recycled water from tailings dam.

4 Keywords: Waste by-products; Silicate removal; Wastewater treatment; Precipitation.

5 1. Introduction

6 The exploitation and utilization of mineral resources have increased over the past

7 decades as the global demand for metal products has increased. However, the excessive

8 exploitation of mineral resources also triggers environmental pollution and large

9 amounts of wastewater are generated during the mineral processing. Because of the

10 increasing demand for water resource and the restriction on wastewater discharge, most

11 of the mineral processing wastewater is recycled and reused for flotation systems after

12 various treatments (Seneviratne 2007). The residual reagents in recycled wastewater

13 are beneficial for the flotation of minerals in their circuits but have a great impact on

14 the flotation of other minerals in different flotation systems (Chen et al. 2009). The

15 recycling of wastewater will lead to the enrichment of ions in the water, which will

16 change the water quality and affect the flotation of minerals (Liu et al. 2013). The

17 change of water chemistry will influence the froth stability and flotation surface

18 chemistry and, thus, the minerals’ flotation performance (Bıçak et al. 2012). Sodium

19 silicate (NaO·mSiO2) is widely used as a dispersant and depressant in flotation to

20 separate minerals from specific types of gangue minerals. It plays an important role in

21 the flotation of scheelite (Gao et al. 2016, Wang et al. 2016, Han et al. 2017), fluorite

2
 ACCEPTED MANUSCRIPT

1 (CaF2) (Gao et al. 2015), phosphate (Dho and Iwasaki 1990) and xenotime (YPO4)

2 (Zhang and Anderson 2017). Acidified sodium silicate is a kind of modified sodium

3 silicate and has selective depression effects on the flotation separation of scheelite from

4 calcite (CaCO3) (Feng et al. 2015). However, when the amount of sodium silicate

5 exceeds a certain limit, it will affect the flotation of valuable minerals.

6 The flotation plant in Luanchuan (Henan, China) is the largest producer of low

7 grade scheelite in China. It is owned and operated by China Molybdenum Co. Ltd.

8 (CMOC). The scheelite is recovered from the tailings of molybdenite flotation. The

9 roughing of scheelite is conducted at normal temperature using sodium carbonate as the

10 regulator and fatty acids as the collector. The cleaning of scheelite is conducted at a

11 high temperature (85-90°C) and by using sodium silicate (50-70 kg/t) as the dispersant

12 and depressant. The underutilized sodium silicate in the cleaning flotation circuit will

13 reduce the settling velocity of tailings. There is still a large amount of water glass in the

14 recycled water after natural sedimentation, which has a great impact on the scheelite

15 roughing flotation circuit. The roughing recovery is reduced by 5-8% in the winter and

16 1-2% in other seasons compared with another flotation workshop where the silicate in

17 the recycled water is lower than 100 mg/L. Lime is used as a flocculant to accelerate

18 the settlement of tailings, but silicate removal fails because of the limited solubility of

19 calcium hydroxide in strong alkaline conditions. Considering the limitations of using

20 lime in sodium silicate removal, new treatment methods or substitutes need to be

21 developed to achieve a sufficient purification of silicate. The use of waste by-products

3
 ACCEPTED MANUSCRIPT

1 as substitutes for lime has been researched in the literature. Studies on replacing lime

2 with cement kiln dust (CKD) (Mackie and Walsh 2012) and by-products from

3 quicklime manufacturing (Tolonen et al. 2014) have demonstrated the use of these

4 neutralization agents for treating acid mine water.

5 The method of precipitation is widely used for removing inorganic anions. There

6 are few investigations on silicate removal, but many have reported the removal of

7 phosphate, fluoride and sulphate precipitated by the salts of calcium and magnesium.

8 The inorganic anions will form undissolved precipitates with calcium or magnesium.

9 Ammonia, phosphate and fluoride from semiconductor wastewater can be

10 simultaneously removed by chemical precipitation with magnesium salts (Huang et al.

11 2017). Phosphate and fluoride from thin-film transistor liquid crystal display (TFT-

12 LCD) wastewater can be removed by chemical precipitation with calcium salts (Lu and

13 Liu 2010). The sulphate from mine water can be removed by precipitation with

14 ettringite (Tolonen et al. 2016). High fluoride-content wastewater in various industries,

15 such as fertilizer, aluminium and semiconductor, can be removed by the addition of

16 calcium chloride (Liu and Liu 2016). In the same way, silicate may be removed by

17 precipitating with the salts of calcium or magnesium. The waste acid produced by the

18 decomposition of low grade scheelite concentrate with hydrochloric acid contains many

19 underutilized hydrochloric acid and generated calcium chloride. Many studies

20 (Tomaszewska et al. 2001, Xu et al. 2009, Kim et al. 2012, Jeong et al. 2005) have

21 studied the recovery and utilization of waste acid, but it is very difficult for the recovery

4
 ACCEPTED MANUSCRIPT

1 of low concentration waste acid due to the limited technology and high costs. The waste

2 acid is usually neutralized by lime and discharges to rivers. The waste alkali is produced

3 by the reaction between the ammonium chloride that is generated during sodium

4 carbonate production and lime. The ammonia is reused to produce sodium carbonate,

5 but the calcium chloride is discarded as a waste by-product.

6 In the paper, we provide details on the application of waste by-products to remove

7 sodium silicate from mineral processing wastewater. The influence of sodium silicate

8 is eliminated and the recycled water is reused for the flotation system at the Luanchuan

9 plant. The mechanism of sodium silicate removal by precipitation is studied by

10 speciation modelling and XRD. The technical and economic feasibility of the treatment

11 and the reuse of wastewater is also investigated by tests at the plant.

12 2. Materials and methods

13 2.1 Materials

14 2.1.1 Mineral processing wastewater

15 The wastewater sample is taken from a plant located in Luanchuan, China. The

16 collected water is mixed samples taken continuously every 2 h for 48 h from the

17 scheelite cleaning flotation circuit. The concentration of Si is 2600-2800 mg/L. The

18 chemical compositions of the wastewater are shown in Table 1. A typical dosage of

19 sodium hydroxide (200-400 g/t) is used for pH adjustment, and a large amount of

20 sodium silicate (50-70 kg/t) is used as the dispersant and gangue depressant in the

21 scheelite cleaning flotation. The underutilized sodium silicate in the cleaning tailing

5
 ACCEPTED MANUSCRIPT

1 stream makes it difficult to settle for the tailings. Lime is used as a flocculant at a dosage

2 of 30-40 t/d to promote the settlement of tailings. However, the silicate removal fails

3 because of the solubility limitation of calcium hydroxide in strong alkaline conditions.

4 2.1.2 Waste by-products

5 The two by-products sampled are waste acid and waste alkali liquids. The

6 collected by-products are mixed samples taken continuously every 2 h for 48 h. The

7 waste acid is produced by the decomposition of low grade scheelite concentrate (WO3,

8 25-30 wt%) by hydrochloric acid and collected from a tungsten plant located in

9 Luanchuan, China. Its main ingredients are underutilized hydrochloric acid and

10 generated calcium chloride. The waste alkali is the by-product from sodium carbonate

11 production and collected from a sodium carbonate plant located in Luoyang, China. Its

12 main ingredients are calcium chloride and sodium chloride. The present application

13 options for the two by-products are limited by the cost and technology, and the by-

14 products are mainly treated by neutralization. Their chemical compositions are shown

15 in Table 1. The concentration of Ca in waste acid and waste alkali are 23000-25000

16 mg/L.

17 Table 1. Results of chemical analysis of the wastewater and by-products [mg/L].

Samples pH Na Mg Al Si P S K Ca Fe Mo W

Wastewater 11.86 2700 17.3 0.642 2730 1.08 379 ± 5 38.2 63.4 5.42 27.6 10.2

Waste acid 4.68 2270 187 67.5 232 322 253 ± 2 37.5 23200 456 29.7 136

Waste alkali 10.04 3140 7.18 0.316 6.38 1.47 268 ± 1 298 24800 0.486 0.0836 0.264

6
 ACCEPTED MANUSCRIPT

1 2.2 Experimental procedures

2 2.2.1 Precipitation tests

3 The silicate removal experiments were performed with the jar test, in which a

4 wastewater sample volume of 500 mL was used for each jar. Experiments were

5 conducted at room temperature (20 ± 2 °C). The experimental procedures are described

6 as follows: 500 mL of the wastewater was first added to a jar placed on a magnetic

7 stirrer (78HW-3, Hangzhou Instrument Electric Machine Co. Ltd., China), followed by

8 the addition of hydrated lime (Ca(OH)2), calcium chloride, waste acid and waste alkali

9 at a desirable Ca:Si molar ratio (0.1-1.2). Then, the wastewater was stirred for 3 min.

10 After the stirring was completed, the mixture was left for 8 h to sediment, and 5 ml of

11 the supernatant was removed and filtered through 0.22 μm filter membranes for silicate

12 analysis. The produced sludge was washed thrice with deionized water and air dried for

13 XRD analysis. The silicate removal was quantified by using the following equation:
Ci  C f
14 Silicate removal [%] =  100 (1)
Ci

15 where Ci is the initial concentration of silicate and Cf is its final concentration after the

16 treatment.

17 The experimental procedures for examining the effects of pH and reaction time on

18 the silicate removal were similar to those described above. The pH of wastewater was

19 controlled using a pH meter (pHS-3C, China) from 7.5 to 13 with 1 N NaOH or 1 N

20 HCl before the addition of calcium chloride, waste acid and waste alkali at a Ca:Si

21 molar ratio of 0.6 and 1.0. The reaction time was designed from 30 s to 600 s during

7
 ACCEPTED MANUSCRIPT

1 the precipitation and the silicate removals were analysed every 30 s after stirring for 60

2 s.

3 2.2.2 Calcium silicate formation tests

4 The prepared sodium silicate (m=1.03±0.03) aqueous solution ([Si]=3000 mg/L,

5 pH=12.84) was mixed with calcium chloride after adjusting the pH from 9 to 13.5. The

6 molar ratio of calcium to silicate was controlled at 0.6, 0.8 and 1.0. After the addition

7 of calcium chloride, the aqueous solution was stirred for 3 min and left for 8 h to

8 sediment. The supernatant was removed and filtered through 0.22 μm filter membranes

9 for silicon and calcium analysis. The precipitates were washed thrice with deionized

10 water and air dried for silicon and calcium analysis.

11 2.2.3 Industrial tests

12 The industrial tests were performed at a scheelite flotation plant located in

13 Luanchuan, China. A comparison analysis related to the removal efficiency of silicate

14 and the economic estimation of the operation for the usage of calcium chloride, waste

15 acid and waste alkali as calcium sources was performed. The waste acid was used from

16 December 1, 2015 to February 28, 2016 to treat scheelite cleaning flotation wastewater.

17 A brief flow diagram of the industrial test process is shown in Fig. 1. The dose of waste

18 acid was 70-80 t/d, and the flow rate was 50-60 L/min. The dose of lime was 15-20 t/d,

19 and the flow rate was 10-15 kg/min. The reaction time was approximately 5 min. The

20 wastewater was pumped to the tailings dam for natural sedimentation after the addition

21 of waste acid and lime. The water in the tailings dam was recycled to the flotation plant.

8
 ACCEPTED MANUSCRIPT

1 The silicate in the recycled water and the parameters of the scheelite roughing flotation

2 were analysed to confirm the treatment efficiency.

3
4 Fig. 1. Brief flow diagram of industrial test process for the treatment and recycle of

5 wastewater by waste acid.

6 2.3 Equilibrium speciation modelling

7 To obtain the distribution of silicate species under the pH conditions, a silicate

8 speciation model is set up according to the hydrolysis reactions and material balance.

9 The formation constants are given by the literature (Felmy et al. 2001) for dilute

10 solution. Silicate anions were hydrolysed depending on the solution pH in the aqueous

11 solution. The species including H4SiO4, H3SiO4- and H2SiO42- had different

12 distributions as a function of pH conditions. The reactions described by Eqs. (2) - (3)

13 are used to model (Yang et al. 2008) the hydrolysis speciation of the dependence of

14 dilute silicate aqueous solution on pH conditions. In addition, the speciation modelling

9
 ACCEPTED MANUSCRIPT

1 is applied to establish the distributions of calcium cation hydrolysis species versus pH

2 conditions. The hydrolysis reactions are described by Eqs. (4) - (5).

H 2 SiO4 2  H  =H 3SiO4  K1  1013.45

3 (2)

H 3SiO4   H  =H 4 SiO4 K 2  109.82

4 (3)

Ca 2   OH - =CaOH K 3  101.40
5 (4)

Ca 2  2OH  =Ca  OH  2（） K 4  102.77

6 aq
(5)

7 2.4 Analytical methods

8 The analysis of wastewater and waste by-product samples were performed in a

9 reliable laboratory (Modern analysis and testing center of Central South University)

10 with inductively coupled plasma optical emission spectrometry (ICP–OES,

11 SPECTROBLUE, Germany) according to the standard of HJ 776-2015 (China-MEP

12 2015). The precipitated solids were analysed with XRD equipment (Rigaku D/max

13 2500, Japan). The analysis of silicon and calcium in the supernatant was performed

14 with ICP–OES. The contents of silicon and calcium in the precipitate were examined

15 with XRF analysis (AxiosmAX, PANalytical, Holland). The laboratory tests were

16 performed in triplicate and the presented numbers were average values.

17 3. Results and discussions

18 3.1 The results of precipitation tests

19 3.1.1 Removal of silicate by chemical precipitation

20 A series of experiments was conducted to determine the performance of hydrated

21 lime, calcium chloride, waste acid and waste alkali on the silicate removal. Fig. 2(a)

10
 ACCEPTED MANUSCRIPT

1 shows the quantitative results. When the Ca:Si molar ratio is 0.8, the silicate removals

2 reach 85% with precipitation by calcium chloride, waste acid and waste alkali.

3 Moreover, the silicate removals remain at approximately 90% when the Ca:Si molar

4 ratio exceeds 1.0. Silicate removal fails when using hydrated lime because of the

5 solubility limitation of calcium hydroxide in strong alkaline conditions. When Ca2+ is

6 added to wastewater that is rich in silicate, if the ionic product of Ca2+ and H2SiO42- is

7 greater than solubility product (Ksp) of calcium metasilicate, H2SiO42- will be removed

8 as insoluble precipitates, according to Eq. (6). The silicate removal by waste acid gives

9 a higher efficiency than calcium chloride and waste alkali because the presence of

10 hydrochloric acid in the waste acid can lower the pH of wastewater and improve the

11 formation of orthosilicic acid, which is conducive to silicate removal. The settleability

12 of the precipitates formed by Ca2+ was observed during the experiments. The tailings

13 provide a carrier for the precipitates and compress the sedimentation bed, which is

14 conducive to the settlement of precipitates and efficient silicate removal. It forms an

15 obvious solid-liquid separation interface, and the supernatant of the resulting solution

16 system is limpid after free sedimentation for 8 h.

17 The influences of reaction conditions such as pH and sedimentation time on the

18 silicate removal were investigated. The effect of pH on the silicate removal was

19 performed at Ca:Si molar ratios of 0.6 and 1.0. Fig. 2(b) shows the changes in the

20 silicate removal with pH. The pH of wastewater is 11.86 without adjustment, and the

21 silicate removals are 60%-70% with the addition of calcium chloride, waste acid and

11
 ACCEPTED MANUSCRIPT

1 waste alkali. When the pH is adjusted from 8 to 13 at Ca:Si=1.0, the silicate removals

2 remain stable, as observed under the natural pH. When the pH is adjusted to 12.5-13 at

3 Ca:Si=0.6, the silicate removal is the same as when the pH is 11.86. However, the

4 silicate removal increases with the reduction of pH and improves to 90% when the pH

5 is below 11. The results demonstrate that the reduction of solution pH is conducive to

6 silicate removal. The effect of sedimentation time on the silicate removal was examined

7 at a Ca:Si molar ratio of 1.0. The silicate removals remain steady at approximately 90%

8 with sedimentation time. Fig. 2(c) shows the quantitative results which demonstrate that

9 the precipitation reaction finishes in a short time.

10

11 Fig. 2. The results of precipitation tests (a) and effects of pH (b) and sedimentation

12 time (c) on silicate removal precipitated by Ca2+ at different pH and Ca:Si molar

13 ratios.

14 3.1.2 Characterization of the precipitates

15 The suspended solid particles in wastewater and precipitates formed by the

16 treatment with calcium chloride, waste acid and waste alkali at Ca:Si=1.0 were

17 identified from the XRD patterns (Fig. 3). The main components of scheelite cleaning

12
 ACCEPTED MANUSCRIPT

1 tailings are calcite (CaCO3) (47.25 wt%), fluorite (CaF2) (27.84 wt%) and quartz (SiO2)

2 (8.65 wt%). The main phases of suspended solid particles are calcite, fluorite, andradite

3 and quartz according to the X-ray diffraction pattern. The peaks of the precipitates are

4 very similar to that of suspended solid particles, except for the peak (C1) at 31.68.

5 Amorphous calcium silicate (CaO·mSiO2) is generated by treatment with the calcium

6 cation.

8 Fig. 3. X-ray diffraction patterns for the suspended solid particles (1) in wastewater

9 and the precipitates formed by the treatment with calcium chloride (2), waste acid (3)

10 and waste alkali (4) at Ca:Si=1.0.

11 3.2 The formation mechanism of calcium silicate

12 3.2.1 Equilibrium speciation of silicate and calcium

13 The assignments related to the distribution of anionic species in sodium silicate

14 solution, especially very dilute solutions, have been researched in the literature (Halasz

15 et al. 2007, Osswald and Fehr 2006). In this modelling, the precipitation of amorphous

16 silica under thermodynamic conditions is ignored. In this context, it should be

17 mentioned that the kinetics of the formation of silica precipitate is rather slow and

13
 ACCEPTED MANUSCRIPT

1 strongly limited (Conrad et al. 2007), especially at a high pH, low ionic strength, and

2 low silica concentration. Therefore, this implies that the formation of amorphous silica

3 precipitate can be ignored at pH>11. Equilibrium protonation parameters and the

4 formation constants are described in Eq. (1) - (4), along with the chemical formula of

5 each species, and the calculated equilibrium distributions are plotted in Fig. 4. At pH<7,

6 the dominant species of silicate hydrolysis is H4SiO4. With an increasing pH from 7 to

7 9.8, the distribution of H4SiO4 decreases. At pH between 9.8 and 13.4, the dominant

8 species of silicate hydrolysis is H3SiO4-. At pH>13.4, the dominant silicate hydrolysis

9 species is H2SiO42-. For calcium hydrolysis, at pH<10, the dominant species is the Ca2+

10 cations. At pH>12.7, the dominant calcium hydrolysis species is Ca(OH)2. The

11 distribution of Ca(OH)- is the highest at pH=12.7. At pH between 10 and 12.7, the three

12 calcium species can be formed.

13 The pH of wastewater is 11.86 (Table 1), the dominant species of silicate is

14 H3SiO4-, and the dominant species of calcium is Ca2+. With the addition of calcium

15 cations, the silicate will be removed to form precipitates of calcium metasilicate. The

16 reaction of precipitation is predicted as Eq. (6).

Ca 2+  H 3SiO4  =CaO  mSiO2 (s)

17 (6)

14
 ACCEPTED MANUSCRIPT

2 Fig. 4. Results of speciation modelling for silicate (a) and calcium (b) hydrolyses.

3 3.2.2 Effect of pH on the calcium silicate formation

4 The experimental results of silicate removal as affected by the molar ratio of

5 calcium to silicon and pH are shown in Fig. 5(a). The silicate removal is significantly

6 affected by the molar ratio of calcium to silicon and pH. It is found that silicate removal

7 increases with an increase in the Ca:Si molar ratio and a decrease in pH. When the pH

8 is 12.85 (without adjustment), the silicate removal increases from 65% to 97% as Ca:Si

9 molar ratio increases from 0.6 to 1.0. When the pH is adjusted to 13.5, the silicate

10 removal is similar to that of 12.85. The silicate removal remain stable as the changes of

11 pH when the Ca:Si molar ratio is 1.0. The silicate removal increases to 97% from 92%

12 as pH is reduced to 12 when the Ca:Si molar ratio is 0.8. When the pH is adjusted to

13 approximately 12 and Ca:Si molar ratio is 0.6, the silicate removal sharply increases to

14 97% from 65%. Fig. 5(b) shows the calcium utilization at different pH and Ca:Si molar

15 ratios. The calcium utilization decreases gradually with the decrease of pH, which

16 indicates that a small amount of calcium can achieve high removal efficiency of silicate

17 by decreasing the solution pH. Fig. 5(c) shows the SiO2:CaO molar ratio in the

15
 ACCEPTED MANUSCRIPT

1 precipitates at different pH and Ca:Si molar ratios. The SiO2:CaO molar ratios in the

2 precipitates increase with the decrease of solution pH at different Ca:Si molar ratios.

3 The results indicate that solution pH is the key factor for silicate removal and the

4 formation of calcium silicate. The literature (Matijevic and Robert J 2012) has

5 expounded the mechanism of orthosilicic acid formation and polymerization. The

6 silicate does not simply exist in sodium silicate solution; other hydrolysis components

7 form as the pH is reduced. The protonation reactions are described in Eq. (7). When the

8 pH is further reduced, H4SiO4 will combine with H3SiO4- to form dimers. The dimers

9 will combine with H3SiO4- to form polysilicic acid. The polymerization reactions are

10 described in Eq. (8) - (9). When the pH is reduced to 9-10, the SiO2:CaO molar ratios

11 in the precipitates reach 6-15, which indicates that the silicate mainly exists in the form

12 of polysilicic acid. The formation of polysilicic acid can reduce the dosage of calcium

13 and improve the silicate removal.

  
14 H 2 SiO4 2 
H
H 3SiO4  
H
H 4 SiO4 
H
H 5 SiO4  (7)

15 (8)

16 (9)

16
 ACCEPTED MANUSCRIPT

4 Fig. 5. The silicate removal (a), calcium utilization (b) and SiO2:CaO molar ratio in

5 the precipitates (c) precipitated by calcium chloride at different pH and Ca:Si molar

6 ratios.

17
 ACCEPTED MANUSCRIPT

1 3.3 Industrial treatment of scheelite flotation wastewater

2 3.3.1 Comparison analysis of waste by-products

3 In this research, a comparative analysis of the usage of calcium chloride, waste

4 acid and waste alkali as the calcium sources in the removal of silicate was performed.

5 The analysis is mainly related to the removal efficiency of silicate and the economic

6 estimation of the operation. In the analysis of the removal efficiency of silicate, the

7 usages of calcium sources and the corresponding silicate removal are presented as

8 follows: as shown in Fig. 2(a), at the Ca:Si molar ratio of 1.0, the silicate removal

9 reaches 89.48% at a calcium chloride dose of 10.83 g/L, the silicate removal reaches

10 92.37% at a waste acid dose of 168.13 g/L and the silicate removal reaches 89.25% at

11 a waste alkali dose of 157.66 g/L. The results show that waste acid has a better removal

12 efficiency of silicate compared to the waste alkali and calcium chloride. In the economic

13 estimation, only the costs of calcium sources utilized were considered; the energy

14 consumption, manpower, and equipment, among other costs, were ignored. The costs

15 of waste acid and waste alkali include mainly the transportation cost (0.15 $/t·km)

16 because of the free usage of waste by-products. The results of the economic estimation

17 are shown in Table 2. The cost for treating scheelite flotation wastewater with waste

18 acid is revealed to be 0.63 $/m3, which has a better economic advantage compared with

19 calcium chloride (2.39 $/m3) and waste alkali (1.54 $/m3). The comparison of the

20 removal efficiency of silicate and the economic estimation confirms that treating

21 scheelite flotation wastewater while using waste acid as the calcium source is optimal.

18
 ACCEPTED MANUSCRIPT

1 Table 2. Results of the silicate removal efficiency and economic evaluation by using

2 different calcium sources in Ca:Si=1.

Calcium sources Doses [g/L] Doses Ca [g/L] Removal [%] Prices [$/t] Costs [$/m3]

Calcium chloride 10.83 3.9 89.48 220 2.39

Waste acid 168.13 3.9 92.37 3.75 0.63

Waste alkali 157.66 3.9 89.25 9.75 1.54

3 3.3.2 Results of the industrial tests

4 Sodium silicate is used as a dispersant and gangue depressant in the scheelite

5 cleaning flotation circuit in Luanchuan. The wastewater in the tailings dam is reused

6 after natural sedimentation at the scheelite flotation plant. The residual sodium silicate

7 (1000-1200 mg/L) in recycled water has a great impact on the scheelite roughing

8 recovery, especially in winter. The scheelite roughing recovery decreased by 5-8% in

9 the winter and 1-2% in other seasons compared with another flotation workshop where

10 the silicate in the recycled water is lower than 100 mg/L. The recycled water in the

11 other workshop comes from a different tailings dam, which contains only scheelite

12 roughing wastewater. Less sodium silicate is used in the scheelite roughing circuit, so

13 there is less silicate in the recycled water from this tailings dam. To determine the

14 feasibility of silicate removal from scheelite flotation wastewater by chemical

15 precipitation with waste acid, the industrial test was performed in the plant. The results

16 of the industrial test are shown in Fig. 6, and the recycled water before and after

17 treatment by waste acid is analysed, as shown in Table 3. When the treated wastewater

18 pumps to the tailings dam, it takes a long time for the wastewater to recycle back. The

19
 ACCEPTED MANUSCRIPT

1 treated wastewater mixes with untreated wastewater in the tailings dam causes a silicate

2 removal delay at the beginning of waste acid addition. After the natural sedimentation,

3 the mixed wastewater is recycled to the flotation circuit. With the continuous addition

4 of waste acid, all of the untreated wastewater in the tailings dam is recycled back, and

5 the silicate in the recycled water reaches a balance. With the continuous cycle of

6 wastewater in the tailing dam, silicon in the recycled water decreases to 200 mg/L from

7 1200 mg/L after adding waste acid and the silicate removal reaches 85%. The

8 consumption of waste acid is 75.24 t/d and the cost is 282.15 $/d based on the

9 transportation cost of 0.15 $/t·km. With the removal of silicate in recycled water, the

10 scheelite roughing recovery in the winter of 2016 remains stable and consistent with

11 the other workshop. The results of wastewater treatment and improvement of scheelite

12 recovery are significant.

13

14 Fig. 6. The silicate removal in recycled water at the plant with continuous waste acid

15 addition from December 1, 2015 to February 28, 2016.

16 During the industrial tests, the possible negative effects and trend of impurity

20
 ACCEPTED MANUSCRIPT

1 elements such as sulphur, phosphorus, tungsten, iron, aluminium and magnesium in the

2 waste acid were analysed. Sulphur and phosphorus exist in the form of sulfate and

3 phosphate, and other metal elements exist in the form of cations in the water. The pH

4 of recycled water is 10-11 and decreases to 9-10 after adding waste acid. In alkalescent

5 condition, the metal cations will be converted into hydrate precipitates, and sulphur and

6 phosphorus will be converted into calcium salt precipitates by the reaction of lime and

7 stay in the tailings dam. The concentration of impurity elements dose not increase after

8 adding waste acid, as shown in Table 3. Harmful elements such as sulphur and

9 phosphorus in scheelite concentrate are not exceeded, and the flotation indexes of the

10 cleaning circuit remain normal.

11 Table 3. Analysis results of the recycled water before and after treatment by waste

12 acid [mg/L].

Samples pH Na Mg Al Si P S K Ca W Fe

Untreated water 10.62 1340 3.06 0.942 1130 0.282 6.48 ± 0.02 11.9 26.8 3.68 0.942

Treated water 9.43 1320 0.384 0.874 218 0.168 4.56 ± 0.02 10.1 23.5 3.42 0.168

13 4. Conclusions

14 In this study, an efficient, original and environmentally friendly technology that

15 utilizes waste by-products (waste acid and waste alkali liquids) to treat mineral

16 processing wastewater was presented. Chemical precipitation was used to remove

17 silicate from scheelite cleaning flotation wastewater. Laboratory experiments regarding

18 precipitation, equilibrium speciation modelling, XRD were conducted to explain the

19 removal of silicate and provide the characterization and formation mechanism of

21
 ACCEPTED MANUSCRIPT

1 calcium silicate. The precipitation results showed that silicate removals reached 90%

2 when the Ca:Si molar ratio exceeded 1.0. The results of XRD confirmed the

3 characterization and formation of calcium silicate. During the formation of calcium

4 silicate, solution pH was the key factor for silicate removal. Polysilicic acid formed

5 with the reduction of pH, which was beneficial for the silicate removal and reduced the

6 usage of calcium at the same time. The results of the comparison analysis showed that

7 the removal efficiency of silicate using waste by-products or calcium chloride was

8 proportional and that the treatment costs with waste acid (0.63 $/m3) and waste alkali

9 (1.54 $/m3) were lower than that of calcium chloride (2.38 $/m3). This efficient removal

10 of silicate was confirmed by the industrial test work at a plant. The results showed that

11 silicate removal reached 85% in the recycled water when the consumption of waste acid

12 was 75.24 t/d, and the effect of silicate on sheelite roughing recovery was eliminated.

13 This innovative and environmentally friendly approach that utilizes industrial

14 waste by-products to treat flotation wastewater can maximize the utilization of

15 industrial wastes and realize the sustainable development of resources and environment.

16 In the future, it will develop into a new technique and gradually supersede traditional

17 methods, with their high energy consumption and cost. Furthermore, the approach not

18 only has practical significance for mineral processing wastewater treatment but may

19 also be used in other industrial applications.

20 Acknowledgements

21 The research was supported by Natural Science Foundation of China

22
 ACCEPTED MANUSCRIPT

1 (No.51374247 and 51634009), Innovation Driven Plan of Central South University

2 (No. 2015CX005), the National 111 Project (No. B14034), and Collaborative

3 Innovation Center for Clean and Efficient Utilization of Strategic Metal Mineral

4 Resources.

5 References

6 Bıçak, Ö., Ekmekçi, Z., Can, M. and Öztürk, Y. (2012) The effect of water chemistry

7 on froth stability and surface chemistry of the flotation of a Cu–Zn sulfide ore.

8 International Journal of Mineral Processing 102, 32-37.

9 Chen, J., Liu, R., Sun, W. and Qiu, G. (2009) Effect of mineral processing wastewater

10 on flotation of sulfide minerals. Transactions of Nonferrous Metals Society of

11 China 19(2), 454-457.

12 China-MEP (2015) Water Quality-Determination of 32 Elements-Inductively Coupled

13 Plasma Optical Emission Spectrometry, China Environmental Sciences Press,

14 Beijing.

15 Conrad, C.F., Icopini, G.A., Yasuhara, H., Bandstra, J.Z., Brantley, S.L. and Heaney,

16 P.J. (2007) Modeling the kinetics of silica nanocolloid formation and precipitation

17 in geologically relevant aqueous solutions. Geochimica et Cosmochimica Acta

18 71(3), 531-542.

19 Dho, H. and Iwasaki, I. (1990) Role of sodium silicate in phosphate flotation. Minerals

20 & Metallurgical Processing 7(4), 215-221.

21 Felmy, A.R., ChoJames, H., RustadMarvin, R. and Mason, J. (2001) An Aqueous

23
 ACCEPTED MANUSCRIPT

1 Thermodynamic Model for Polymerized Silica Species to High Ionic Strength.

2 Journal of Solution Chemistry 30(6), 509-525.

3 Feng, B., Luo, X., Wang, J. and Wang, P. (2015) The flotation separation of scheelite

4 from calcite using acidified sodium silicate as depressant. Minerals Engineering

5 80, 45-49.

6 Gao, Y., Gao, Z., Sun, W. and Hu, Y. (2016) Selective flotation of scheelite from

7 calcite: A novel reagent scheme. International Journal of Mineral Processing 154,

8 10-15.

9 Gao, Z., Bai, D., Sun, W., Cao, X. and Hu, Y. (2015) Selective flotation of scheelite

10 from calcite and fluorite using a collector mixture. Minerals Engineering 72, 23-

11 26.

12 Halasz, I., Agarwal, M., Li, R. and Miller, N. (2007) Vibrational spectra and

13 dissociation of aqueous Na2SiO3 solutions. Catalysis Letters 117(1), 34-42.

14 Han, H., Hu, Y., Sun, W., Li, X., Cao, C., Liu, R., Yue, T., Meng, X., Guo, Y., Wang,

15 J., Gao, Z., Chen, P., Huang, W. and Liu, J. (2017) Fatty acid flotation versus BHA

16 flotation of tungsten minerals and their performance in flotation practice.

17 International Journal of Mineral Processing 159, 22-29.

18 Huang, H., Liu, J., Zhang, P., Zhang, D. and Gao, F. (2017) Investigation on the

19 simultaneous removal of fluoride, ammonia nitrogen and phosphate from

20 semiconductor wastewater using chemical precipitation. Chemical Engineering

21 Journal 307, 696-706.

24
 ACCEPTED MANUSCRIPT

1 Jeong, J., Kim, M.S., Kim, B.S., Kim, S.K., Kim, W.B. and Lee, J.C. (2005) Recovery

2 of H2SO4 from waste acid solution by a diffusion dialysis method. J Hazard Mater

3 124(1), 230-235.

4 Kim, J.Y., Shin, C.H. and Choi, H. (2012) Recovery of phosphoric acid from mixed

5 waste acids of semiconductor industry by diffusion dialysis and vacuum

6 distillation. Separation and Purification Technology 90, 64-68.

7 Liu, C. and Liu, J.C. (2016) Coupled precipitation-ultrafiltration for treatment of high

8 fluoride-content wastewater. Journal of the Taiwan Institute of Chemical

9 Engineers 58, 259-263.

10 Liu, W., Moran, C.J. and Vink, S. (2013) A review of the effect of water quality on

11 flotation. Minerals Engineering 53, 91-100.

12 Lu, N.C. and Liu, J.C. (2010) Removal of phosphate and fluoride from wastewater by

13 a hybrid precipitation–microfiltration process. Separation and Purification

14 Technology 74(3), 329-335.

15 Mackie, A.L. and Walsh, M.E. (2012) Bench-scale study of active mine water treatment

16 using cement kiln dust (CKD) as a neutralization agent. Water Res 46(2), 327-334.

17 Matijevic, E. and Robert J, G. (2012) Surface and colloid science, Plenum Press, New

18 York.

19 Osswald, J. and Fehr, K.T. (2006) FTIR spectroscopic study on liquid silica solutions

20 and nanoscale particle size determination. Journal of Materials Science 41(5),

21 1335–1339.

25
 ACCEPTED MANUSCRIPT

1 Seneviratne, M. (2007) A Practical Approach to Water Conservation for Commercial

2 and Industrial Facilities, Elsevier, Burlington, USA.

3 Tolonen, E.-T., Sarpola, A., Hu, T., Rämö, J. and Lassi, U. (2014) Acid mine drainage

4 treatment using by-products from quicklime manufacturing as neutralization

5 chemicals. Chemosphere 117, 419-424.

6 Tolonen, E.T., Hu, T., Ramo, J. and Lassi, U. (2016) The removal of sulphate from

7 mine water by precipitation as ettringite and the utilisation of the precipitate as a

8 sorbent for arsenate removal. J Environ Manage 181, 856-862.

9 Tomaszewska, M., Gryta, M. and Morawski, A.W. (2001) Recovery of hydrochloric

10 acid from metal pickling solutions by membrane distillation. Separation and

11 Purification Technology 22, 591–600.

12 Wang, J., Gao, Z., Gao, Y., Hu, Y. and Sun, W. (2016) Flotation separation of scheelite

13 from calcite using mixed cationic/anionic collectors. Minerals Engineering 98,

14 261-263.

15 Xu, J., Lu, S. and Fu, D. (2009) Recovery of hydrochloric acid from the waste acid

16 solution by diffusion dialysis. J Hazard Mater 165(1), 832-837.

17 Yang, X., Roonasi, P. and Holmgren, A. (2008) A study of sodium silicate in aqueous

18 solution and sorbed by synthetic magnetite using in situ ATR-FTIR spectroscopy.

19 J Colloid Interface Sci 328(1), 41-47.

20 Zhang, Y. and Anderson, C. (2017) A comparison of sodium silicate and ammonium

21 lignosulfonate effects on xenotime and selected gangue mineral microflotation.

26
 ACCEPTED MANUSCRIPT

1 Minerals Engineering 100, 1-8.

27
 ACCEPTED MANUSCRIPT

Highlights

Treatment of flotation wastewater using waste acid and waste alkali is investigated.

Ca containing by-products for removing silicate show good performances as CaCl2.

The reduction of pH is beneficial to improve the silicate removal efficiency.

Waste acid is used in the plant and silicate removal reaches 85% in recycled water.