sustainability

Article The Prospective Approach for the Reduction of Mobility in Industrial Waste by Creating Products of Commercial Value

Valdas Rudelis 1, Tadas Dambrauskas 2 , Agne Grineviciene 2 and Kestutis Baltakys 2,* 1 JSC “Lifosa”, Juodkiskio g. 50, LT-57502 Kedainiai, Lithuania; [email protected] 2 Department of Silicate Technology, Kaunas University of Technology, Radvilenu 19, LT-50270 Kaunas, Lithuania; [email protected] (T.D.); [email protected] (A.G.) * Correspondence: [email protected]; Tel.: +370-730-0163; Fax: +370-730-0152

 Received: 20 December 2018; Accepted: 23 January 2019; Published: 25 January 2019 

Abstract: In this work, we present the possibility to reduce the amount of fluoride ions in silica gel waste by using different techniques or to immobilize these ions by creating products of commercial value. The leaching of fluoride ions from silica gel waste to the liquid medium was done under static and dynamic conditions. It was determined that the removal of fluoride ions from this compound depends on various factors, such as dissociation, , the w/s ratio, reaction temperature, leaching conditions, the adsorption properties of silica gel waste, and others. The obtained results showed that, by applying different techniques, the quantity of fluoride ions can be reduced by 60%, while obtained was neutralized by . Additionally, it was determined that silica gel waste is a promising raw material for the hydrothermal synthesis of a stable compound containing fluoride ions–cuspidine.

Keywords: fluoride ions; silica gel waste; leaching; hydrothermal synthesis; cuspidine

1. Introduction The disposal of industrial by-products is becoming an increasing concern for many industries due to the large amounts of wastes generated and increasing costs of operating landfills in combination with the scarcity of landfill sites [1–8]. Accordingly, the recycling and reutilization of industrial waste and by-products are a subject of great importance [9–12]. Large amounts of silica gel waste contaminated with fluoride ions are generated in the industrial production of aluminum fluoride [13,14]. In fact, one of the main fertilizer producers in Lithuania, a joint-stock company “Lifosa”, generates approximately 15 thousand tons per year of the mentioned waste during the manufacture of 17 thousand tons of AlF3 [15]. In the process of obtaining AlF3, the latter compound is formed in the reaction of neutralizing hexafluorosilicic with aluminum hydroxide [13,14]: H2SiF6 + 2Al(OH)3 → 2AlF3 + SiO2·nH2O + H2O + Q (1) The structure of the precipitated silica gel waste depends on many variables: the duration of the reaction, the temperature, the way in which reagents are dosed, the structure of Al(OH)3 used in the process, and the purity of the applied fluorosilicic acid [13]. However, due to the strong bonding of fluoride ions to the of the latter compound, the purification of silica gel waste poses a great challenge. Thus, because of the high cost of silica gel waste processing, it is preferably discharged in landfill sites [13,16–18]. For this reason, long-term storage and maintenance of silica gel waste are currently facing many environmental concerns, which are associated with the dumps and disposal sites of this by-product due to the leaching of fluoride into the surface and underground water [13,16].

Sustainability 2019, 11, 634; doi:10.3390/su11030634 www.mdpi.com/journal/sustainability Sustainability 2019, 11, 634 2 of 18

According to literature [12,19–24], the content of toxic ions can be reduced in two ways: by removing them from waste or by reducing their mobility in the environment–immobilization. The removal or reduction of fluoride ions from silica gel waste is a complex process, thus there is a lack of data in the literature. To our knowledge, in 2014, Iljina et al. tried to purify silica gel waste by performing leaching experiments [17]. However, the results of these experiments demonstrated that the concentration of F− ions was only decreased from 8.64% to 8.03% and the remaining amount of fluoride ions was adsorbed by the main compound–silica dioxide. In the case of success, silica gel with a low content of F− ions and water contaminated with F− ions can be obtained. The silica gel with − a lower content of F ions (<5%) can be used as an SiO2 source in some industries [25–29], and only LTD “Alufluor AB” [30] sells this product worldwide. Meanwhile, wastewater contaminated with F− ions can be neutralized by adding Ca(OH)2 or CaCO3 [31,32]. During neutralization, the obtained CaF2 may be reused as a raw material or an additive in several industrial applications. Mainly, CaF2 is used for the manufacture of hydrofluoric acid and as a flux in or ceramic industries [31,33]. The manufacture of HF requires a high purity of calcium fluoride, while steel or ceramic industries require only 60–97% purity. According to the literature, untreated silica gel waste can be immobilized in stone [14,19]; however, due to effect of F− ions on the environment, the application in industry is limited. Also, F− ions can be bound into stable compounds during the hydrothermal synthesis of lower basicity calcium silicates hydrates [17,18,34]. For these reasons, it may be possible to use silica gel waste for the synthesis of cuspidine (3CaO·2SiO2·CaF2). This compound is generally used as mold flux, which controls the horizontal mold heat transfer and lubricates the solidified steel shell from the oscillating mold [35,36]. According to Jung et al. [37], cuspidine is a key ingredient in the continuous casting process to improve the quality of the cast slabs. As a matter of fact, in the context of reducing environmental issues and enhancing economic benefits, technologies for converting waste materials into products of commercial value are in great demand [14,17,18,29,34,38]. However, the literature concerning the successful application of silica gel waste contaminated with F− ions is scarce. For this reason, the aim of this work is to reduce F− ions in silica gel waste by using different techniques and/or immobilize these ions by creating products of commercial value.

2. Materials and Methods

2.1. Raw Materials

In this work, the following reagents were used: silica gel waste (SGW), i.e., a waste product of AlF3 production in the chemical plant of Lifosa (Kedainiai,˙ Lithuania), which was dried for 48 h at 50 ◦C, 2 with a specific surface area Sa of 281.01 m /kg, by a CILAS LD 1090 granulometer; and calcium ◦ from Ca(OH)2 (“Stanchem”, Poland), additionally burned at 550 C temperature for 1 h and ground for 30 s in a vibrating cup “Pulverisette 9” mill at 600 rpm, with a quantity of free CaO equal to 98.7%.

2.2. Methodology

2.2.1. The Chemical Analysis

A standard method of SiO2 chemical analysis. A total of 1 g of the sample was mixed with sodium and potassium mixture (6 g) and put in a platinum crucible. The crucible was placed in a furnace and melted at a 900–1000 ◦C temperature for 1 h. After fusion, in order to solidify the melt on the crucible walls, the sample was placed on a 200 cm3 porcelain plate with distilled water. Then, the distilled water was removed and a required amount of 10% of HCl was added and mixed until almost all the CO2 gas had been released from the solution. After that, the solution in the porcelain plate was heated on the sand bath until it evaporated. The residue formed was mixed with a few drops of concentrated HCl acid and left to cool down. After 30 min, in order to dissolve chlorides, Sustainability 2019, 11, 634 3 of 18 hot distilled water was added to the porcelain plate. Then, the solution with residue consisting of SiO2 was filtered through an ash-free filter and rinsed with distilled water to eliminate chloride ions (the presence of chloride ions was tested by adding one drop of clean filtrate and a drop of 1% AgNO3 solution on a glass). The obtained filtrate was used for the second chemical analysis of SiO2 and, finally, ◦ both filtrate papers were put in the crucibles and heated at 1000 C for 1 h. The amount of SiO2 was calculated by using the equation: SiO2 = (a1·100)/a, % (2) where a1 is the mass of heated residue after the experiment, g; and a is the initial mass of the sample, g. The determination of fluoride. A total of 1 g of sample was put on a platinum plate and mixed with 10 g of sodium and mixture (5:7 of NaOH and KOH). After that, the platinum plate was put on the sand bath and heated till the mixture was melted. During the heating, the melt was vigorously stirred with a platinum spatula and then, it was left to cool down. Afterwards, 150 cm3 of distilled water was poured onto the platinum plate and heated on the sand bath till the formed salts were melted. Then, in order to bind the fluoride ions, 15 g of chemically pure (NH4)2CO3 powder was added into the solution and evaporated till the dry salt was formed. After that, the residue was mixed with 150 cm3 of distilled water and heated. A hot solution with precipitates was poured into a 250 cm3 flask and cooled down. Then, it was diluted with water to the indicated level and filtered off. Before measurements, TISAB II was added to each sample (volume ratio 1:1). The concentrations of fluoride ions in the solution were measured by using a Metler Toledo T70 potentiometer. The error of the selective electrode for F− ions is ±1 ppm (0.0001%). The concentration of F− was calculated as the arithmetic mean of the three individual results. The measurement deviations were below 3%.

2.2.2. The Leaching of Fluoride Ions The leaching of F− ions from silica gel waste to the liquid medium was done by applying different techniques: (1) Leaching of F− ions under static conditions; in order to reach the water-to-solid (w/s) ratios of 2, 4, 6, 8, and 100, 10 g of silica gel waste was mixed with distilled water and kept for 24 h at 25 and 55 ◦C temperatures. After leaching, the suspensions were filtered off, and the products were dried at a 50 ◦C ± 5 temperature for 24 h. (2) Leaching of F− ions under dynamic conditions by using continuous distilled water (25 ◦C, by applying vacuum 0.6–0.7 bar) flow, which was applied on 10 g of silica gel waste till the water-to-solid (w/s) ratio reached the value of 25, 50, 100, and 200. After the process, the products were dried at a 50 ◦C ± 5 temperature for 24 h. (3) F− ions leaching in cycles under dynamic conditions. A total count of 20 cycles were carried out during this experiment, in which a total amount of 10 g of SGW was treated with 50 mL of distilled water in each step till the w/s ratio reached 100. Different temperature water was used (25, 35, 45, and 55 ◦C) for this experiment. After each cycle, the obtained products were filtered off and dried (50 ◦C ± 5; 24 h) at the end of the process.

2.2.3. The Application of Silica Gel Waste The synthesis of dibasic hydrates by using silica gel waste was performed in the mixture, with a molar ratio of C/S (CaO/SiO2) that was equal to 2. The hydrothermal synthesis was carried out in unstirred suspensions, under saturated steam pressure at 200 ◦C temperature for 16, 24, 48, and 72 h by applying extra argon gas (10 bar) [39].

2.3. Instrumental Analysis The samples were characterised by powder X-ray diffraction (XRD; with a D8 Advance X-ray diffractometer), differential scanning calorimetry (DSC; with Netzsch DSC214 Polyma instrument), X-ray fluorescence spectroscopy (XRF; with a Bruker X-ray S8 Tiger WD spectrometer), a grain-size Sustainability 2018, 10, x FOR PEER REVIEW 4 of 19

The samples were characterised by powder X-ray diffraction (XRD; with a D8 Advance X-ray diffractometer), differential scanning calorimetry (DSC; with Netzsch DSC214 Polyma instrument), Sustainability 2019, 11, 634 4 of 18 X-ray fluorescence spectroscopy (XRF; with a Bruker X-ray S8 Tiger WD spectrometer), a grain-size analyzer (CILAS 1090 LD), and scanning electron microscopy (SEM; with a JEOL JSM-7600F analyzer (CILAS 1090 LD), and scanning electron microscopy (SEM; with a JEOL JSM-7600F instrument) [17,18,34,39]. instrument) [17,18,34,39]. The densityThe of samples of samples was was measuredmeasured using using a Quantachrome a Quantachrome “Ultrapyc 1200e” “Ultrapyc gas pycnometer 1200e” gas pycnometerunder under helium helium atmosphere; atmosphere; the number the of measurements number of was measurements 5 and accuracy—0.1 was kg/m 5 and3. The accuracy value of – 0.1 kg/m3. ThepH value was measured of pH withwas a Hannameasured instrument with (Hia 9321,Hanna microprocessor instrument pH (Hi meter, 9321, Hanna microprocessor Instruments, pH meter, HannaWoonsocket, Instruments, RI, USA). TheWoonsocket, measurements RI, of theUSA) value. The of pH measurements were repeated three of times, the andvalue deviations of pH were were below 2%. repeated three times, and deviations were below 2%. 3. Results and Discussion 3. Results and Discussion 3.1. The Properties of Silica Gel Waste (SGW) 3.1. The PropertiesIn the of first Silica part Gel of thisWaste work, (SGW) the properties of silica gel waste (SGW) were examined in detail. ◦ − The chemical analysis data showed that dried (50 C, 48 h) SGW consists of 79.0% SiO2 and 10.0% F In theions first (Figure part1 a).of this Yet almost work, the the same properties tendency of was silica observed gel waste by applying (SGW) XRF: were it was examined determined in detail. 3+ The chemicalthat SGWanalysis contains data 36.2% showed silica, whichthat dried is equivalent (50°C, to 48 78.9% h) SGW SiO2 (Figureconsists1a). of Moreover, 79.0% SiO 2.5%2 ofand Al 10.0% F- ions (Figureions 1a). and Yet traces almost of other the elements same aretendency also present wa ins observed the mentioned by compoundapplying (FigureXRF: it1a). was As otherdetermined authors’ work showed, the chemical composition of SGW strongly depends on the conditions of AlF3 that SGW contains 36.2% silica, which is equivalent to 78.9% SiO2 (Figure 1a). Moreover, 2.5% of manufacture [17,18]. Al3+ ions and traces of other elements are also present in the mentioned compound (Figure 1a). As The previous results were in a good agreement with the data of XRD analysis. AlF3·3H2O (PDF other authors’00-035-0827, work showed,d-spacing =the 0.545; chemical 0.386; 0.329; composit 0.244ion nm) of and SGW the amorphousstrongly depends dioxide, on the which conditions ◦ of AlF3 manufacturecorresponds to[17,18]. a broad basal reflection in a 18–37 diffraction angle range, were observed in the XRD The previouspattern of SGWresults (Figure were1b). in a good agreement with the data of XRD analysis. AlF3·3H2O (PDF 00-035-0827, dMoreover,-spacing=0.545; the results 0.386; of particle 0.329; size 0.244 distribution nm) analysisand the showed amorphous that the diameter silicon ofdioxide, the latter which compound particles varied in a 0.03–170 µm range, while the particles with a size of 34–72 µm were correspondsdominant to a broad (Figure 1basalc). It was reflection also determined in a 18–37º that the diffraction surface area and angle density range, of silica were gel wasteobserved were in the XRD patternequal of to SGW 281.01 (Figure m2/kg and1b). 2141 kg/m3, respectively.

100 SiOby chemical analysis 80 79.0 78.9 Alby XRF analysis

60

40 Quantity, Quantity, %

20 10.0 2.5 0 - SiO SiO F Al3+ 22 a Figure 1. Cont.

Sustainability 2019, 11, 634 5 of 18 Sustainability 2018, 10, x FOR PEER REVIEW 5 of 19

40 A 35 A 30 25 A A 20 A 15 A AA 10 A AA Intensity, Intensity, cps. A 5 0 3 1323334353 θ 2 , deg. b

100 10 90 9 80 8 70 7 60 6 50 5 40 4 30 3 20 2 Cumulative Cumulative % volume, 10 1 0 0 3 5 Differencial volume, [x10] % volume, Differencial 16 25 43 71 0.6 1.4 8.5 125 200 355 0.03 Particle diameter, µm c

Figure 1. FigureThe chemical 1. The chemical composition composition (a), (aXRD), XRD pattern pattern (b), (b), andand particle particle size size distribution distribution (c) of dried(c) of dried silica gel waste. Indexes: A—AlF3·3H2O. silica gel waste. Indexes: A – AlF3·3H2O. The results of chemical composition were proved by the data of differential scanning calorimetry. ◦ Moreover,In the DSCthe curve,results a of two-step particle decomposition size distributi of AlFon3 ·analysis3H2O was showed observed that at 146 the and diameter 165 C of the latter compoundtemperatures, particles in which varied the in quantity a 0.03–170 of the μ heatm range, flow was while equal the to: particles 2.78 J/g with for AlF a size3·3H 2ofO 34–72→ μm AlF ·0.5H O + 2.5H O and 33.62 J/g for AlF ·0.5H O → AlF + 0.5H O, respectively (Figure2a) [ 40,41]. were dominant3 (Figure2 1c).2 It was also determined3 2 that 3the surface2 area and density of silica gel Meanwhile, the third endothermic effect, which is assigned to the dehydration of silica gel waste were equal to 281.01 m2/kg and 2141 kg/m3, ◦respectively. (SiO2·nH2O → SiO2 + nH2O), was noticed at 188 C. These results were confirmed by the experiment The ofresults water vaporof chemical adsorption: composition when the ratio ofwere relative prov watered vaporby the pressure data (p /ofp0 )differential was equal to 1.0,scanning calorimetry.the dehydrationIn the DSC heat curve, of silica a geltwo-step waste significantly decomposition increased of from AlF 593·3H to 1892O J/g was (Figure observed2b), and thisat 146 and · 165°C temperatures,process overlapped in which with the the endothermic quantity effectof the assigned heat flow to the was decomposition equal to: of 2.78 AlF3 J/g3H2 forO (Figure AlF32·3H). 2O → AlF3·0.5H2O + 2.5H2O and 33.62 J/g for AlF3·0.5H2O → AlF3 + 0.5H2O, respectively (Figure 2a) [40,41]. Meanwhile, the third endothermic effect, which is assigned to the dehydration of silica gel (SiO2·nH2O → SiO2 + nH2O), was noticed at 188 °C. These results were confirmed by the experiment of water vapor adsorption: when the ratio of relative water vapor pressure (p/p0) was equal to 1.0, the dehydration heat of silica gel waste significantly increased from 59 to 189 J/g (Figure 2b), and this process overlapped with the endothermic effect assigned to the decomposition of AlF3·3H2O (Figure 2).

Sustainability 2019, 11, 634 6 of 18 Sustainability 2018, 10, x FOR PEER REVIEW 6 of 19

Heat: 2.78 J/g Peak: 146 ºC Onset: 138 ºC End: 154 ºC Exo

Heat flow, Heat flow, mW/mg Heat: 33.62 J/g Heat: 59.33 J/g ΔQ Peak: 165 ºC Peak: 188 ºC Endo Onset: 159 ºC Onset: 182 ºC End: 170 ºC End: 196 ºC 20 120 220 320 420 520 Temperature, ºC a

Peak: 157 ºC Heat: 189 J/g Heat flow, Heat flow, mW/mg Exo Peak: 175 ºC ΔQ Onset: 160 ºC Endo End: 197 ºC

20 120 220 320 420 520 Temperature, ºC b

Figure 2. DSC curves of silica gel waste samples before (a) and after water vapor adsorption (p/p0 = 1.0) (b). Figure 2. DSC curves of silica gel waste samples before (a) and after water vapor adsorption (p/p0 = 1.0) (b). To fully understand the thermal behavior of SGW, the calcination of this compound was carried out in a 50–1000 ◦C temperature range (Figure3). Although the typical diffraction maximum characteristic ◦ To fullyof AlF 3understand·3H2O slightly the decreased thermal with behavior increasing temperature,of SGW, the it remained calcination stable tillof 165thisC compound and fully was ◦ ◦ carried outdecomposed in a 50–1000 at 188 °CC temperature (Figure3). In a range higher temperature(Figure 3). Although range (188–550 the typicalC), a low-intensity diffraction peak maximum (d-spacing = 0.355 nm) at a ~25◦ diffraction angle was noticed and could be assigned to other anhydrous characteristic of AlF3·3H2O slightly decreased with increasing temperature, it remained stable till aluminum fluoride phases. Besides, when the temperature of calcination was increased to 1000 ◦C, 165 °C and fully decomposed at 188 °C (Figure 3). In a higher temperature range (188–550 °C), a due to the reaction between containing components and SiO2, mullite (Al4.95·Si1.05·O9.52; low-intensityPDF 00-015-0776; peak (d-spacingd-spacing = =0.355 0.341; 0.541;nm) at 0.270 a ~25 nm) was° diffraction formed in calcinationangle was products noticed (Figure and3 ).could be assignedIt to is worthother mentioning anhydrous that Faluminum− ions are strongly fluoride bound phases. in the structure Besides, of silicawhen gel waste,the temperature because of ◦ calcinationthe performedwas increased chemical analysisto 1000 data °C, of calcineddue to SGW the (188 reaction and 550 C)between showed thataluminium the quantity containing of fluoride ions in a solid is equal to ~9.8%. components and SiO2, mullite (Al4.95·Si1.05·O9.52; PDF 00-015-0776; d-spacing= 0.341; 0.541; 0.270 nm) was formed in calcination products (Figure 3). It is worth mentioning that F- ions are strongly bound in the structure of silica gel waste, because the performed chemical analysis data of calcined SGW (188 and 550 °C) showed that the quantity of fluoride ions in a solid is equal to ~9.8%.

SustainabilitySustainability 20182019, 10, 11, x, 634FOR PEER REVIEW 7 of 7 18 of 19

1000 m m m m m

550 0.355

188 0.355 Intensity, Intensity, cps.

A A 165 A A A A A A A AA A A 145 A A A A A A A AA A A 105 A A A A A A A A A

3 1323334353 θ 2 , deg. ◦ FigureFigure 3. 3. XRDXRD patternspatterns of of calcined calcined in ain 50–1000 a 50–1000C temperature °C temperature range forrange 1 h silicafor 1 gel h wastesilica samples.gel waste Indexes: A—AlF ·3H O, m—mullite. samples. Indexes: 3A – 2AlF3·3H2O, m – mullite. 3.2. The Leaching Peculiarities of F− Ions under Static and Dynamic Conditions 3.2. The Leaching Peculiarities of F- Ions under Static and Dynamic Conditions In the second part of this work, the leaching peculiarities of F− ions under static and dynamic - conditionsIn the second were examined part of this by usingwork, dried the leaching SGW, when peculiarities the water-to-solid of F ions (w/s) under ratios static varied and fromdynamic 2 conditionsto 200 and were the temperature examined by of using the reaction dried wasSGW, equal when to 25,the 35, water-to-solid 45, and 55 ◦C. (w/s) It was ratios determined varied from that 2 ° tounder 200 and static the conditions, temperature the of water-to-solid the reaction was (w/s) equal ratio, to as 25, well 35, as 45, the and reaction 55 C. temperature, It was determined strongly that − underaffected static the conditions, stability of the mainwater-to-solid compound (w/s) containing ratio, as F wellions as in SGW,the reaction AlF3·3H temperature,2O (Figure4). strongly At a ◦ – affected25 C temperature, the stability when of the the main w/s compound ratio reached containing 100, the intensity F ions ofin diffraction SGW, AlF peaks3·3H2O characteristic (Figure 4). ofAt a ° 25AlF C3 temperature,·3H2O decreased when by 3.1the timesw/s ratio (Figure reached4). It was 100, determined the intensity that, of evendiffraction when thepeaks w/s characteristic ratio was ofequal AlF3·3H to 200,2O decreased a small amount by 3.1 of times the mentioned (Figure 4). compound It was de wastermined present that, in SGW. even Meanwhile, when the thew/s increase ratio was equalof reaction to 200, temperature a small amount positively of the affected mentione the decompositiond compound was of AlF present3·3H2O: in when SGW. the Meanwhile, temperature the ◦ increasewas equal of toreaction 45–55 C,temperature and AlF3·3H positi2O wasvely fully affected removed the from decomposition the solid (Figure of 4AlF). 3·3H2O: when the temperatureFurthermore, was equal the datato 45–55 of the °C, chemical and AlF analysis3·3H2O ofwas silica fully gel removed waste treated from underthe solid static (Figure conditions 4). at 25 ◦C (w/s = 100) showed that the amount of F− ions released into the liquid medium was equal to ~49% from the total amount of mentioned ions in SGW (Table1). However, when the water-to-solid ratio was increased two times, the amount of F− ions released into the liquid medium increased slightly. It was also observed that, by increasing the w/s ratio from 2 to 100, the value of the pH of the liquid medium rose by two times (Table1). It is worth noting that a higher reaction temperature has a positive effect on the quantity of released F− ions as it increased by 4% and 9% after treatment at a 45 ◦C and 55 ◦C temperature, respectively. In spite of the mentioned fact, the value of the pH of the liquid medium was kept almost the same (Table1). It is worth mentioning that Iljina et al. [ 17], reported that during the leaching of SGW, AlF3·3H2O fully decomposed when the water to solid ratio

Sustainability 2019, 11, 634 8 of 18 was equal to 500; however, the concentration of F− ions decreased slightly (~6% from total amount of F− ions). The differences between the amount of F− ions released from SGW (~8 times lower according to Iljina et al. [17]) can be explained by the short time of leaching (1 h), and as a result, compounds containing fluoride ions were not destructed. In addition, more successful results were obtained by Krysztafkiewicz et al. [42], where the F− ions concentration was reduced by eight times; however, the initialSustainability concentration 2018, 10, x FOR of mentioned PEER REVIEW ions was only equal to 1.6%, and there is no data about the water 8 of to 19 solid ratio.

w/s = 100 55 °C

w/s = 200 25 °C A A A

w/s = 100 25 °C A A Intensity, Intensity, cps. A A A A

A w/s = 8 A 25 °C A A A A A A A A

A w/s = 2 25 °C A AA A A AA A A A

3 1323334353 θ 2 , deg. FigureFigure 4.4. XRD patternspatterns of silicasilica gelgel wastewaste samplessamples afterafter 2424 hh ofof leachingleaching underunder staticstatic conditions.conditions.

Indexes:Indexes: A—AlFA – AlF33··3H3H2O.

Furthermore,Table the 1. dataThe parametersof the chemical of fluoride analysis ions leaching of silica under staticgel waste conditions. treated under static conditions at 25°C (w/s = 100) showed that the amount of F– ions released into the liquid medium − was equal to ~49% from* the totalThe amount Amount of of ment F ioned TheionsAmount in SGW of (Table 1).pH However, of Liquid when the Sample Name − water-to-solid ratio was increasedIons two in times, Solid, the % amountReleased of F¯ FionsIons, released % into theMedium liquid medium ◦ increased SGW-25slightly. C-2It was also observed -that, by increasing the w/s ratio from 2 to 100, 1.53 the value of ◦ the pH of SGW-25the liquidC-4 medium rose by two - times (Table 1). It is worth noting that a 1.76 higher reaction SGW-25 ◦C-6 - 1.95 temperatureSGW-25 has a◦ C-8positive effect on the - quantity of released F¯ ions as it increased 2.06 by 4% and 9% after treatmentSGW-25 at◦C-100 a 45°C and 55°C temperature, 5.08 respectively. 49.3 In spite of the mentioned 2.94 fact, the value of theSGW-25 pH ◦ofC-200 the liquid medium 4.71 was kept almost the same 53.1 (Table 1). It is worth 3.02 mentioning ◦ that IljinaSGW-35 et al. [17],C-100 reported that during 4.92 the leaching of SGW, 50.9 AlF3·3H2O fully decomposed 2.94 when SGW-45 ◦C-100 4.68 53.3 2.92 the water to solid ratio was equal to 500; however, the concentration of F¯ ions decreased slightly SGW-55 ◦C-100 4.21 58.0 2.90 (~6% from total amount of F¯ ions). The differences between the amount of F¯ ions released from *: the sample names were chosen according to the conditions of the experiment: SGW-25 ◦C-2—silica gel waste SGWsample, (~8 times when lower leaching according temperature to was Iljina equal et to al. 25 [17])◦C and can w/s be ratio—2; explained SGW-25 by◦ C-4—silicathe short geltime waste of sample,leaching (1 h), andwhen as leaching a result, temperature compounds was equal co tontaining 25 ◦C and w/sfluoride ratio—4 ions and etc.were While no thet destructed. label SGW-55 ◦InC-100 addition, meant the more silica gel waste sample, when leaching temperature was equal to 55 ◦C and w/s ratio—100. successful results were obtained by Krysztafkiewicz et al. [42], where the F¯ ions concentration was reduced by eight times; however, the initial concentration of mentioned ions was only equal to 1.6%, and there is no data about the water to solid ratio.

Table 1. The parameters of fluoride ions leaching under static conditions.

Sample name* The amount of F- The amount of pH of liquid ions in solid, % released F- ions, % medium SGW-25°C-2 - 1.53 SGW-25°C-4 - 1.76

Sustainability 2018, 10, x FOR PEER REVIEW 9 of 19

SGW-25°C-6 - 1.95 SGW-25°C-8 - 2.06 SGW-25°C-100 5.08 49.3 2.94 SGW-25°C-200 4.71 53.1 3.02 SGW-35°C-100 4.92 50.9 2.94 SGW-45°C-100 4.68 53.3 2.92 SGW-55°C-100 4.21 58.0 2.90 Note: the sample names were chosen according to the conditions of the experiment: SGW-25°C-2 – silica gel waste sample, when leaching temperature was equal to 25 C and w/s ratio – 2; SGW- 25°C-4 – silica gel waste sample, when leaching temperature was equal to 25 C and w/s ratio – 4 Sustainabilityand etc.2019 While, 11, 634 the label SGW-55°C–100 meant the silica gel waste sample, when leaching 9 of 18 temperature was equal to 55 C and w/s ratio – 100.

− TheThe effect effect of of F¯ F ionsions leaching leaching under under static static conditions conditions is clearly is clearly seen seenin SEM in SEMmicrographs: micrographs: the · untreatedthe untreated SGW sample SGW sample showed showedan amorphous an amorphous mass of SiO mass2·nH2 ofO and SiO 2uncertainnH2O and form uncertain agglomerates form ◦ –agglomerates–crystals crystals (Figure 5a). Meanwhile, (Figure5a). in Meanwhile, the sample inafter the the sample treatment after (w/s the treatment = 100, 25°C), ( w/s the = mentioned 100, 25 C), µ · agglomeratesthe mentioned were agglomerates disrupted were and disrupted only globules and only with globules a size with of a~0.5–1 size of μ ~0.5–1m of SiOm2 of·nH SiO2O2 nHwere2O − observedwere observed (Figure (Figure 5b). Besides,5b). Besides, the results the results of EDX of EDXanalysis analysis showed showed that F that– ions F areions distributed are distributed over theover whole the whole structure structure of silica of silica gel waste gel waste (Figure (Figure 5c);5 c);while while in ina case a case of of the the treated treated sample, sample, they they are are removedremoved from from its its surface surface and and are are more more concentrated concentrated in in the the center center of of the the particles particles (Figure (Figure 55d).d).

a b

c d

FigureFigure 5. 5. SEMSEM micrographs micrographs (a, (a, bb)) andand EDX EDX mapping mapping (c (c,,d) d) of SGWof SGW (a, c(a,) and c) and treated treated SGW SGW (under (under static ◦ staticconditions, conditions, w/s =w/s 100, = 100, 25 C)25 (°C)b,d (b,) samples. d) samples.

According to the previous results given in Figure3, the influence of F − ions leaching on the ◦ properties of SGW was examined at 50 and 188 C temperatures. It was determined that both the density and loss on ignition strongly depend on the w/s ratio and temperature of the reaction. In comparison with untreated samples, due to the structural alterations of silica gel waste during the leaching of F− ions, the density of the latter compound significantly increased from 2140 to 2708 kg/m3 (Table2). Meanwhile, the loss on ignition decreased with an increasing w/s ratio and temperature of the reaction (Table2). Sustainability 2019, 11, 634 10 of 18

Table 2. The density and loss on ignition of silica gel waste before and after the leaching and calcination in a 50–188 ◦C temperature range.

Loss on Ignition, % ◦ Sample Name * Density, kg/m3 Temperature, C 50 188 Untreated SGW 2141 50.81 53.39 SGW-25 ◦C-2 2263 44.33 49.70 SGW-25 ◦C-100 2457 35.16 39.23 SGW-55 ◦C-100 2708 32.28 37.55 *: the sample names were chosen according to the conditions of the experiment: SGW-25 ◦C-20—silica gel waste sample, when leaching temperature was equal to 25 ◦C and w/s ratio—20; SGW-25 ◦C-100—silica gel waste sample, when leaching temperature was equal to 25 ◦C and w/s ratio—100; and SGW-55 ◦C-100—the silica gel waste sample, when leaching temperature was equal to 55 ◦C and w/s ratio—100.

Thus, it is clearly seen that after leaching under static conditions, the amount of F− ions in SGW can be reduced from 10% to lower than 5% and such a product can be sold worldwide [30]. However, the leaching process under static conditions requires a high amount of water (w/s = 100) and is time-consuming. For this reason, in order to reduce the water to solid ratio and shorten the interaction time, in the next stage of this research, the leaching of F− ions was performed under dynamic conditions. It was examined that the change in leaching conditions had an effect on the stability of AlF3·3H2O: under static conditions; when the w/s ratio was equal to 100, the intensity of diffraction peaks characteristic of AlF3·3H2O decreased by 3.1 times (Figure4), while under dynamic conditions, by 1.23 times (w/s = 100) and 1.73 times (w/s = 200) (Figure6). The XRD results were verified by the chemical analysis. It was determined that the leaching of F− ions into the liquid medium proceeded heavily, as only ~16.9% (w/s = 100) and ~32.3% (w/s = 200) of F− ions from the total amount in silica gel waste were released into the reaction medium (Table3). This observation can be explained by the short interaction duration between the latter compound and liquid medium (Table3).

Table 3. The parameters of F− ions leaching under dynamic conditions at 25 ◦C temperature by using different w/s ratios.

The Amount of F− The Amount of pH of Liquid The Duration of Sample Name * Ions in Solid, % Released F− Ions, % Medium Interaction, s SGW-25 9.60 4.2 2.45 40 SGW-50 - - 2.67 69 SGW-100 8.33 16.9 2.90 108 SGW-200 6.78 32.3 3.11 215 *: the sample names were chosen according to the conditions of the experiment: SGW-25—silica gel waste sample, in which w/s ratio was equal to 25; SGW-50—silica gel waste sample, in which w/s ratio was equal to 50; SGW-100—silica gel waste sample, in which w/s ratio was equal to 100; and SGW-200—silica gel waste sample, in which w/s ratio was equal to 200.

In order to prolong the duration of the interaction, the leaching under dynamic conditions was performed in cycles. A total count of 20 cycles were carried out during this experiment, in which a total amount of 10 g of SGW was treated with 50 mL of H2O in each step (till the w/s ratio reached 100). The duration of each cycle lasted approximately 25 s, i.e., the total duration of the reaction was equal to ~500 s. It is worth mentioning that after each step, the liquid medium obtained was neutralized by passing it through calcium hydroxide. The neutralized water was returned to the leaching of SGW. Sustainability 2018, 10, x FOR PEER REVIEW 10 of 19

According to the previous results given in Figure 3, the influence of F¯ ions leaching on the properties of SGW was examined at 50 and 188 °C temperatures. It was determined that both the density and loss on ignition strongly depend on the w/s ratio and temperature of the reaction. In comparison with untreated samples, due to the structural alterations of silica gel waste during the leaching of F¯ ions, the density of the latter compound significantly increased from 2140 to 2708 kg/m3 (Table 2). Meanwhile, the loss on ignition decreased with an increasing w/s ratio and temperature of the reaction (Table 2).

Table 2. The density and loss on ignition of silica gel waste before and after the leaching and calcination in a 50-188 °C temperature range.

Loss on ignition, % Sample name* Density, kg/m3 Temperature, °C 50 188 Untreated SGW 2141 50.81 53.39 SGW-25°C-2 2263 44.33 49.70 SGW-25°C-100 2457 35.16 39.23 SGW-55°C-100 2708 32.28 37.55 Note: the sample names were chosen according to the conditions of the experiment: SGW-25°C-20 – silica gel waste sample, when leaching temperature was equal to 25 C and w/s ratio – 20; SGW- 25°C-100 – silica gel waste sample, when leaching temperature was equal to 25 C and w/s ratio – 100; and SGW-55°C-100 – the silica gel waste sample, when leaching temperature was equal to 55 C and w/s ratio – 100.

Thus, it is clearly seen that after leaching under static conditions, the amount of F¯ ions in SGW can be reduced from 10% to lower than 5% and such a product can be sold worldwide [30]. However, the leaching process under static conditions requires a high amount of water (w/s=100) and is time-consuming. For this reason, in order to reduce the water to solid ratio and shorten the interaction time, in the next stage of this research, the leaching of F¯ ions was performed under dynamicSustainability conditions.2019, 11, 634 11 of 18

A A w/s = 200 A A A A A A A

w/s = 100 A A A A A A A AA A A Intensity, Intensity, cps. w/s = 50 A A A A A A A AA A A w/s = 25 A A A A A A A A A

3 1323334353 θ 2 , deg. Figure 6. XRD patterns of silica gel waste samples after leaching under dynamic conditions at 25 ◦C. Indexes: A—AlF ·3H O. 3 2 As expected, due to the ~5 times longer interaction time, at a 25 ◦C temperature, the decomposition of AlF3·3H2O is more intensive compared with the previous results (Figures6 and7). To completely destroy the structure of crystalline compounds, the reaction temperature was increased to 35, 45, ◦ and 55 C. It was determined that AlF3·3H2O became unstable in SGW, when the reaction temperature ◦ ◦ ◦ reached 45 and 55 C, because only a broad peak typical to amorphous SiO2 in a 18 –26 diffraction angle range was visible in the XRD pattern (Figure7). It is worth mentioning that a similar tendency was obtained under static conditions (Figure4); however, in this case, the interaction time is extremely shortened, i.e., from 24 h at static conditions to 0.14 h at dynamic conditions in cycles. The results of the XRD analysis were in good agreement with the data of differential scanning calorimetry. When the reaction temperature was equal to 25 ◦C, the amount of adsorbed heat attributed ◦ to the dehydration of AlF3·3H2O at 180 C decreased by almost two times: from 33.62 to 18.26 J/g, and at 35 ◦C—to 12.69 J/g (Figure2; Figure7). However, at a higher leaching temperature (45 and 55 ◦ C), the dehydration of AlF3·3H2O was no longer observed (Figure7). This fact allowed us to state that the latter compound is fully removed from silica gel waste. Moreover, due to the structural alterations of the mentioned compound, an endothermic effect at a ~182 ◦C temperature, which is assigned to the dehydration of silica gel waste, shifted to a higher temperature range (Figure2; Figure7). Besides, the amount of adsorbed heat during this process decreased from 59.33 to 39.68 J/g as a smaller quantity of moisture was present in this compound after leaching. At the beginning of leaching in cycles, the value of the liquid medium pH rapidly increased until it reached a w/s ratio of 25 (Figure8). However, when the duration of the reaction was extended, the mentioned values only slightly increased and at the end of the process, they were equal to 4.4–4.7. Presumably, at the beginning of the leaching, hexafluorosilicic acid and/or HF was removed from SGW and by continuing the mentioned process, the amount of released F− ions strongly depended on the solubility of compounds, which contained F− ions. SustainabilitySustainability 20182019, 10, 11, x, FOR 634 PEER REVIEW 12 12 of of 18 19

55 °C

45 °C Intensity, Intensity, cps. 35 °C A A

25 °C A A A A A A A

3 1323334353 θ 2 , deg. a

55 °C

Heat: 39.68 J/g Peak: 225 ºC Onset: 219 ºC End: 233 ºC 45 °C

Heat: 40.04 J/g Heat: 12.69 J/g Peak: 220 ºC Onset: 144 ºC Onset: 215 ºC End: 189 ºC End: 227 ºC 35 °C

154 175 Heat: 41.82 J/g Peak: 224 ºC Heat flow, Heat flow, mW/mg Onset: 218 ºC End: 232 ºC 25 °C

Heat: 18.26 J/g Heat: 48.44 J/g Exo Peak: 181 ºC Peak: 205 ºC ΔQ Onset: 170 ºC Onset: 170 ºC Endo End: 186 ºC End: 220 ºC 20 120 220 320 420 520 Temperature, ºC b

FigureFigure 7. 7.XRDXRD patterns patterns (a) (a )and and DSC DSC curves curves (b) (b) of silica gelgel wastewaste samplessamples after after leaching leaching in in cycles cycles (w/s = 100) under dynamic conditions at different temperatures. Indexes: A—AlF3·3H2O. (w/s=100) under dynamic conditions at different temperatures. Indexes: A – AlF3·3H2O.

The results of the XRD analysis were in good agreement with the data of differential scanning calorimetry. When the reaction temperature was equal to 25 °C, the amount of adsorbed heat attributed to the dehydration of AlF3·3H2O at 180 °C decreased by almost two times: from 33.62 to 18.26 J/g, and at 35 °C – to 12.69 J/g (Figure 2, Figure 7). However, at a higher leaching temperature

Sustainability 2018, 10, x FOR PEER REVIEW 13 of 19

(45 and 55 °C), the dehydration of AlF3·3H2O was no longer observed (Figure 7). This fact allowed us to state that the latter compound is fully removed from silica gel waste. Moreover, due to the structural alterations of the mentioned compound, an endothermic effect at a ~182 °C temperature, which is assigned to the dehydration of silica gel waste, shifted to a higher temperature range (Figure 2, Figure 7). Besides, the amount of adsorbed heat during this process decreased from 59.33 to 39.68 J/g as a smaller quantity of moisture was present in this compound after leaching. At the beginning of leaching in cycles, the value of the liquid medium pH rapidly increased until it reached a w/s ratio of 25 (Figure 8). However, when the duration of the reaction was extended, the mentioned values only slightly increased and at the end of the process, they were equal to 4.4-4.7. Presumably, at the beginning of the leaching, hexafluorosilicic acid and/or HF was removed from SGW and by continuing the mentioned process, the amount of released F– ions stronglySustainability depended2019, 11, 634 on the solubility of compounds, which contained F¯ ions. 13 of 18

5 4.5 4 3.5 25° C 3 ° 2.5 35 C Value pHof 2 45 °C 1.5 55 °C 1 135791113151719 Number of cycles FigureFigure 8. 8. TheThe change change in the values ofof pHpH ofof liquidliquid mediummedium during during leaching leaching in in cycles. cycles.

The chemical analysis data of SGW showed that, after leaching in cycles at 25 ◦C, the quantity of The chemical− analysis data of SGW showed that, after leaching in cycles− at 25 °C, the quantity ofreleased released F F¯ions ions was was equal equal to ~45.6% to ~45.6% (w/s (w/s = 100) = 100) from from the total the amounttotal amount of F ions of F¯ in ions silica in gel silica waste, gel which was almost the same as under the static conditions (Table1) and ~2.7 times greater than under waste, which was almost the same as under the static conditions (Table 1) and ~2.7 times greater dynamic conditions (Table3). The same tendency was observed at a higher temperature (Tables3 than under dynamic conditions (Table 3). The same tendency was observed at a higher temperature and4). Moreover, the reaction temperature had a positive effect on the amount of F − ions released (Tables 3 and 4). Moreover, the reaction temperature had a positive effect on the amount of F¯ ions in the liquid medium (Table4), because the best results (55.7%) were obtained after leaching at a released in the liquid medium (Table 4), because the best results (55.7%) were obtained after 55 ◦C temperature. leaching at a 55 °C temperature. Table 4. The amount of F− ions in solid after leaching in cycles at different reaction temperatures, when Tablethe w/s 4. ratioThe amount was equal of toF¯ 100. ions in solid after leaching in cycles at different reaction temperatures, when the w/s ratio was equal to 100. − * The Amount of F Ions The Amount of Released Sample Name The amount of F- The− amount of Sample name* in Solid, % F Ions, % - SGW-25 ◦Cions 5.45 in solid, % released 45.6 F ions, % SGW-35SGW-25°C◦C 4.925.45 50.9 45.6 ◦ SGW-45SGW-35°CC 4.684.92 53.3 50.9 ◦ SGW-55SGW-45°CC 4.444.68 55.7 53.3 ◦ *: the sample namesSGW-55°C were chosen according to the conditions4.44 of the experiment: SGW-25 55.7C—silica gel waste sample, when the reaction temperature was equal to 25 ◦C; SGW-35 ◦C—silica gel waste sample, when the reaction Note:temperature the sample was equal names to 35 were◦C; SGW-45 chosen◦C—silica according gel waste to the sample, conditions when the of reaction the experiment: temperature wasSGW-25°C equal to – ◦ ◦ ◦ silica45 C; gel and waste SGW-55 sample,C—silica when gel wastethe reaction sample, whentemperature the reaction was temperature equal to was25  equalC; SGW-35°C to 55 C. – silica gel waste sample, when the reaction temperature was equal to 35 C; SGW-45°C – silica gel waste sample,As it was when mentioned the reaction before, temperature in order was to neutralize equal to 45 the  liquidC; and medium SGW-55°C obtained – silica after gel leachingwaste experiments,sample, when calcium the reaction hydroxide temperature was used was [31 equal,32]. Itto was55  determinedC. that calcium hydroxide reacted − 3+ with both F and Al ions as calcium fluoride (CaF2; PDF 04-002-2191, d-spacing = 0.315, 0.193, 0.165As nm) it was and mentioned katoite (Ca 3before,Al2.85O in2.55 order(OH) 9.45to neutralizePDF 04-017-1504, the liquidd-spacing medium = 0.509,obtained 0.333, after 0.279 leaching nm) experiments,were identified calcium in the XRD hydroxide pattern afterwas theused experiment [31,32]. (FigureIt was9 ).determined The mentioned that compounds calcium hydroxide formed under all experimental conditions. The obtained results were in good agreement with the literature data [31,32,42]. Meanwhile, the different results obtained by Iljina et al. [17] can be explained by their use of a different leaching technique: SGW with CaO additive (6.5–20%) was treated under static conditions (1 h, w/s-500), thus CaF2 crystallized in solid and cannot be distinguished. By summarizing the leaching results, it can be stated that the removal of fluoride ions from this compound to the liquid medium depends on various factors, such as dissociation, solubility, the w/s ratio, reaction temperature, leaching conditions, the adsorption properties of silica gel waste, and others. Despite the mentioned factors, the quantity of F− can be reduced more than two times and the − obtained products of SGW with a lower content of F (>5%) and CaF2 can be reused in other industries. Sustainability 2018, 10, x FOR PEER REVIEW 14 of 19 reacted with both F- and Al3+ ions as calcium fluoride (CaF2; PDF 04-002-2191, d-spacing = 0.315, 0.193, 0.165 nm) and katoite (Ca3Al2.85O2.55(OH)9.45 PDF 04-017-1504, d-spacing = 0.509, 0.333, 0.279 nm) were identified in the XRD pattern after the experiment (Figure 9). The mentioned compounds formed under all experimental conditions. The obtained results were in good agreement with the literature data [31,32,42]. Meanwhile, the different results obtained by Iljina et al. [17] can be explained by their use of a different leaching technique: SGW with CaO additive (6.5–20%) was treated under static conditions (1 h, w/s-500), thus CaF2 crystallized in solid and cannot be distinguished. Sustainability 2019, 11, 634 14 of 18

40 a 35 30 a a a a 25 a a 20 n a 15 n a 10 a Intensity, Intensity, cps. a n 5 0 10 20 30 40 50 60 θ 2 , deg. FigureFigure 9. 9. XRDXRD patternsof of calcium calcium hydroxide hydroxide after after the neutralization the neutralization experiment. experiment. Indexes: n—calciumIndexes: n – calciumfluoride, fluoride, a—katoite. a – katoite. 3.3. Application of SGW for the Direct Hydrothermal Synthesis of Cuspidine By summarizing the leaching results, it can be stated that the removal of fluoride ions from this compoundIt is knownto the thatliquid toxic medium ions can depends be reduced on invariou two ways:s factors, by removing such as dissociation, them or by reducing solubility, their the − w/smobility–immobilization. ratio, reaction temperature, As the leaching 3.2 part showed, conditio thens, quantity the adsorption of F was properties successfully of reducedsilica gel to waste, less andthan others. 5% in Despite SGW, while the mentioned wastewater factors, was neutralized the quan withtity calciumof F¯ can hydroxide. be reduced Meanwhile, more than in ordertwo times to expand the application areas of SGW, in the third part of this work, the possibility to reuse SGW for and the obtained products of SGW with a lower content of F¯ (> 5%) and CaF2 can be reused in the synthesis of a stable compound containing F− ions was determined. other industries. It was determined that in a CaO–silica gel waste–H2O mixture, when the molar ratio of CaO/SiO2 was equal to 2, after 16 h of isothermal curing at 200 ◦C, the main compound containing F− ions in silica 3.3. Application of SGW for the Direct Hydrothermal Synthesis of Cuspidine gel waste, AlF3·3H2O, was unstable and decomposed (Figure 10, curve 1). For this reason, due to the interactionIt is known between that fluoride,toxic ions calcium, can be andreduced silicon in ions, two cuspidineways: by (PDFremoving 04-015-0711, them ord -spacingby reducing = 0.306, their mobility0.290, 0.287 – immobilization. nm) was formed As (Figure the 3.2 10 part, curve showed, 1). Together the quantity with this of compound, F¯ was successfully 1.1 nm tobermorite reduced to less(PDF than 04-011-0271, 5% in SGW,d-spacing while wastewater = 1.139, 0.308, was 0.298 neut nm),ralized katoite with (PDF calcium 00-024-0217, hydroxide.d-spacing Meanwhile, = 0.513, in order0.230, to 0.204expand nm), the and application traces of grossular areas of SGW, (PDF 00-038-0368,in the third dpart-spacing of this = 0.276,work, 0.226,the possibility 0.200 nm) to were reuse SGWalso for identified the synthesis in XRD of patterns a stable (Figure compound 10, curve containing 1). Furthermore, F¯ ions was basic determined. reflections (d -spacing = 0.493; 0.193;It was 0.179 determined nm) of partially that unreactedin a CaO–silica gel (PDFwaste–H 01-078-0315)2O mixture, were when also noticedthe molar (Figure ratio 10 , of CaO/SiOcurve 1).2 was However, equal whento 2, after the duration 16 h of ofisothermal synthesis curing was extended at 200 to°C, 48 the h, 1.1main nm compound tobermorite containing became metastable and recrystallized to other synthesis products (Figure 10, curve 3). Meanwhile, cuspidine F¯ ions in silica gel waste, AlF3·3H2O, was unstable and decomposed (Figure 10, curve 1). For this reason,remained due stableto the under interaction all hydrothermal between fluoride, synthesis calcium, conditions and (Figure silicon 10 ).ions, The datacuspidine of liquid (PDF medium 04-015- − 0711,analysis d-spacing showed = that0.306, during 0.290, the 0.287 hydrothermal nm) wassynthesis, formed (Figure F ions were10, curve not released 1). Together into the with liquid this medium, but instead, were combined into a stable compound. compound, 1.1 nm tobermorite (PDF 04-011-0271, d-spacing = 1.139, 0.308, 0.298 nm), katoite (PDF Thus, the experimental studies showed that the reduction of mobility and/or removal of F− 00-024-0217, d-spacing = 0.513, 0.230, 0.204 nm), and traces of grossular (PDF 00-038-0368, d-spacing ions from the SGW can be carried out under dynamic or hydrothermal conditions. Based on the = 0.276, 0.226, 0.200 nm) were also identified in XRD patterns (Figure 10, curve 1). Furthermore, data obtained in this study and already existing equipment in a Joint stock Company “Lifosa” [15], basic reflections (d-spacing = 0.493; 0.193; 0.179 nm) of partially unreacted portlandite (PDF 01-078- a principal technological scheme for reducing the quantity of F− ions in silica gel waste or for its 0315) were also noticed (Figure 10, curve 1). However, when the duration of synthesis was immobilization has been designed (Figure 11). For the synthesis of cuspidine, the required amount extended to 48 h, 1.1 nm tobermorite became metastable and recrystallized to other synthesis of SGW and is weighed (CaO/SiO2 = 2.0), and supplied to a mixer (3), in which an appropriate productsquantity (Figure of water 10, (w/s curve = 10) 3). is addedMeanwhile, from the cuspid reservoirine remained (2). The mixture stable ofunder raw materialsall hydrothermal is then synthesissupplied conditions to an autoclave (Figure (4) and 10). the The hydrothermal data of liquid treatment medium is carried analysis out: the showed synthesis that temperature during the ◦ and duration are equal to 200 C and 16 h, respectively. After isothermal curing, the product is filtered off by a vacuum belt filter (6). For the leaching of silica gel waste, SGW is supplied from the weight dispensers to a vacuum belt filter (6), where it is treated by adding a required amount of water (55 ◦C, w/s = 100). The obtained liquid medium with F− ions is neutralized with calcium hydroxide on the second vacuum belt filter (6). Later on, liquid medium is returned to the water reservoir (2). SustainabilitySustainability2019 2018, ,11 10,, 634 x FOR PEER REVIEW 1515 of of 18 19 hydrothermal synthesis, F¯ ions were not released into the liquid medium, but instead, were The obtained products (cuspidine, silica gel with a lower content of fluoride ions, and calcium fluoride) combined into a stable compound. are stored in products silos (7).

c

c c jp c j p k c k k k cc c c c c c

c p p c j c k cp k pc p k k kj cc c c c c c

p p c Intensity, Intensity, cps. j c T p c pc p k k kc Tj c k c c c c p c p p c c p T p j c p k kc T k c cc p c T k cc c

3 132333435363 2θ, deg. FigureFigure 10.10.XRD XRDpatterns patterns ofof synthesissynthesis productsproducts afterafter 1616 hh (curve(curve 1),1), 2424 hh (curve(curve 2),2), 4848 hh(curve (curve 3),3),and and 7272 hh (curve(curve 4)4) ofof isothermalisothermal curing.curing. Indexes:Indexes: c—cuspidine,c – cuspidine, p—portlandite,p – portlandite, k—katoite,k – katoite, j—grossular,j – grossular, T—1.1T – 1.1 nm nm tobermorite. tobermorite. Sustainability 2018, 10, x FOR PEER REVIEW 16 of 19

Thus, the experimental studies showed that the reduction of mobility and/or removal of F¯ ions from the SGW can be carried out under dynamic or hydrothermal conditions. Based on the data obtained in this study and already existing equipment in a Joint stock Company “Lifosa” [15], a principal technological scheme for reducing the quantity of F¯ ions in silica gel waste or for its immobilization has been designed (Figure 11.). For the synthesis of cuspidine, the required amount of SGW and lime is weighed (CaO/SiO2 = 2.0), and supplied to a mixer (3), in which an appropriate quantity of water (w/s = 10) is added from the reservoir (2). The mixture of raw materials is then supplied to an autoclave (4) and the hydrothermal treatment is carried out: the synthesis temperature and duration are equal to 200 °C and 16 h, respectively. After isothermal curing, the product is filtered off by a vacuum belt filter (6). For the leaching of silica gel waste, SGW is supplied from the weight dispensers to a vacuum belt filter (6), where it is treated by adding a required amount of water (55 °C, w/s =100). The obtained liquid medium with F¯ ions is neutralized with calcium hydroxide on the second vacuum belt filter (6). Later on, liquid medium is returned to the water reservoir (2). The obtained products (cuspidine, silica gel with a lower content of fluoride ions, and calcium fluoride) are stored in products silos (7).

Figure 11.11. AA principalprincipal technologicaltechnological schemescheme forfor silicasilicagel gel waste waste neutralization/utilization: neutralization/utilization: 1—raw1 –raw materials silos, silos, 2 2—water– water reservoi reservoir,r, 3 – 3—mixer, mixer, 4 – 4—autoclave, autoclave, 5 – 5—pump,pump, 6 – 6—vacuumvacuum belt beltfilter, filter, 7 – 7—productsproducts silos. silos.

4. Conclusions 1. It was determined that silica gel waste mainly consists of silicon dioxide (79%) and other compounds containing 10% F¯ and 2.5% Al3+ ions. It was observed that three endothermic effects are characteristic of this compound: a two-step decomposition of AlF3·3H2O at 146 and 165 °C temperatures and the dehydration of silica gel at 188 °C. Moreover, the evaluation of the thermal stability of SGW showed that AlF3·3H2O was fully decomposed at 188 °C, while at a higher calcination temperature (1000°C), mullite was formed due to the reaction between aluminium- containing components and SiO2. 2. The findings revealed that, by applying different leaching techniques of F¯ ions (leaching under static and dynamic in cycles conditions), it is possible to reduce the amount of mentioned ions in silica gel waste to less than 5%. The obtained water contaminated with F¯ ions was neutralized with calcium hydroxide and as a result, calcium fluoride and katoite were produced. 3. Silica gel waste is a promising raw material for the hydrothermal synthesis of cuspidine (3CaO·2SiO2·CaF2), because this compound formed after 16 h and remained stable till 72 h of synthesis. The data of the liquid medium analysis showed that during synthesis, F¯ ions were not released into the liquid medium, but instead, were combined into a structure of cuspidine. A principal technological scheme has been designed for the products of commercial value.

Author Contributions: This paper was written using contributions from all of the authors: V.R performed the experiments and results interpretation; T.D. and A.G. performed instrumental analysis; K.B. performed project oversight; and the manuscript was written by all authors. Funding: This research was partially supported by a grant (No. S-MIP-17-92) from the Research Council of Lithuania and by JSC “Lifosa”. Conflicts of interest: None.

References

Sustainability 2019, 11, 634 16 of 18

4. Conclusions 1. It was determined that silica gel waste mainly consists of silicon dioxide (79%) and other compounds containing 10% F− and 2.5% Al3+ ions. It was observed that three endothermic effects ◦ are characteristic of this compound: a two-step decomposition of AlF3·3H2O at 146 and 165 C temperatures and the dehydration of silica gel at 188 ◦C. Moreover, the evaluation of the thermal ◦ stability of SGW showed that AlF3·3H2O was fully decomposed at 188 C, while at a higher calcination temperature (1000 ◦C), mullite was formed due to the reaction between aluminium-containing components and SiO2. 2. The findings revealed that, by applying different leaching techniques of F− ions (leaching under static and dynamic in cycles conditions), it is possible to reduce the amount of mentioned ions in silica gel waste to less than 5%. The obtained water contaminated with F− ions was neutralized with calcium hydroxide and as a result, calcium fluoride and katoite were produced. 3. Silica gel waste is a promising raw material for the hydrothermal synthesis of cuspidine (3CaO·2SiO2·CaF2), because this compound formed after 16 h and remained stable till 72 h of synthesis. The data of the liquid medium analysis showed that during synthesis, F− ions were not released into the liquid medium, but instead, were combined into a structure of cuspidine. A principal technological scheme has been designed for the products of commercial value.

Author Contributions: This paper was written using contributions from all of the authors: V.R. performed the experiments and results interpretation; T.D. and A.G. performed instrumental analysis; K.B. performed project oversight; and the manuscript was written by all authors. Funding: This research was partially supported by a grant (No. S-MIP-17-92) from the Research Council of Lithuania and by JSC “Lifosa”. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Angin, D. Utilization of activated carbon produced from fruit juice industry solid waste for the adsorption of Yellow 18 from aqueous solutions. Bioresour. Technol. 2014, 168, 259–266. [CrossRef][PubMed] 2. Bujak, J.W. Thermal utilization (treatment) of plastic waste. Energy 2015, 90, 1468–1477. [CrossRef] 3. Moh, Y.C.; Abd Manaf, L. Solid waste management transformation and future challenges of source separation and recycling practice in Malaysia. Resour. Conserv. Recy. 2017, 116, 1–14. [CrossRef] 4. Shen, C.; Tran, P.P.; Ly, P.T.M. Chemical waste management in the U.S. Semiconductor industry. Sustainability 2018, 10, 1545. [CrossRef] 5. Stonys, R.; Kuznetsov, D.; Krasnikovs, A.; Škamat, J.; Baltakys, K.; Antonoviˇc,V.; Cernašˇ ejus,˙ O. Reuse of ultrafine mineral wool production waste in the manufacture of refractory concrete. J. Environ. Manag. 2016, 176, 149–156. [CrossRef][PubMed] 6. Yilmaz, O.; Kara, B.Y.; Yetis, U. Hazardous waste management system design under population and environmental impact considerations. J. Environ. Manag. 2017, 203, 720–731. [CrossRef][PubMed] 7. Wo´zniak,J.; Pactwa, K. Overview of polish mining wastes with circular economy model and its comparison with other wastes. Sustainability 2018, 10, 3994. [CrossRef] 8. Siddique, R. Utilization of industrial by-products in concrete. Procedia Eng. 2014, 95, 335–347. [CrossRef] 9. Dyachenko, A.N.; Petlin, I.V.; Malyutin, L.N. The research of sulfuric acidic recycling of aluminum industry fluorine-containing waste products. Procedia Chem. 2014, 11, 10–14. [CrossRef] 10. Li, Y.; Zhang, H.; Zhang, Z.; Shao, L.; He, P. Treatment and resource recovery from inorganic fluoride-containing waste produced by the pesticide industry. J. Environ. Sci. 2015, 31, 21–29. [CrossRef] 11. De Beer, M.; Doucet, F.J.; Maree, J.P.; Liebenberg, L. Synthesis of high-purity precipitated during the process of recovery of elemental sulphur from waste. Waste Manag. 2015, 46, 619–627. [CrossRef] 12. Beak, C.; Seo, J.; Choi, M.; Cho, J.; Ahn, J.; Cho, K. Utilization of CFBC fly ash as a binder to produce in-furnace desulfurization sorbent. Sustainability 2018, 10, 4854. [CrossRef] Sustainability 2019, 11, 634 17 of 18

13. Krysztafkiewicz, A.; Rager, B.; Maik, M. Silica recovery from waste obtained in hydrofluoric acid and aluminum fluoride production from fluosilicic acid. J. Hazard. Mater. 1996, 48, 31–49. [CrossRef] 14. Vaiˇciukyniene,˙ D.; Vaitkeviˇcius,V.; Kantautas, A.; Sasnauskas, V. Utilization of by-product waste silica in concrete-based materials. Mat. Res. 2012, 15, 561–567. [CrossRef] 15. Joint-Stock “Lifosa” Official Website. Available online: https://www.alufluor.com/alusilica/ (accessed on 1 February 2018). 16. Sener, S.; Sener, E.; Karagüzel, R. Solid waste disposal site selection with GIS and AHP methodology: A case study in Senirkent–Uluborlu (Isparta) Basin, Turkey. Environ. Monit. Assess. 2011, 173, 533–554. [CrossRef] [PubMed] 17. Iljina, A.; Baltakys, K.; Baltakys, M.; Siauciunas, R. Neutralization and removal of compounds containing fluoride ions from waste silica gel. Rev. Rom. Mater. 2014, 44, 265–271.

18. Iljina, A.; Baltakys, K.; Bankauskaite, A.; Eisinas, A.; Kitrys, S. The stability of formed CaF2 and its influence on the thermal behavior of C–S–H in CaO–silica gel waste-H2O system. J. Therm. Anal. Calorim. 2017, 127, 221–228. [CrossRef] 19. Kaminskas, R.; Kubiliute, R. Artificial pozzolana from silica gel waste-clay- composite. Adv. Cem. Res. 2014, 26, 155–168. [CrossRef] 20. Li, X.; He, S.; Feng, C.; Zhu, Y.; Pang, Y.; Hou, J.; Luo, K.; Liao, X. Non-competitive and competitive adsorption of Pb2+, Cd2+ and Zn2+ ions onto SDS in process of micellar-enhanced ultrafiltration. Sustainability 2018, 10, 92. [CrossRef] 21. Zhu, Y.; Du, X.; Gao, C.; Yu, Z. Adsorption behavior of inorganic and organic phosphate by manganese plaques on reed roots in wetlands. Sustainability 2018, 10, 4578. [CrossRef] 22. Chen, L.; Chen, Q.; Rao, P.; Yan, L.; Shakib, A.; Shen, G. Formulating and Optimizing a Novel Biochar-Based Fertilizer for Simultaneous Slow-Release of Nitrogen and Immobilization of Cadmium. Sustainability 2018, 10, 2740. [CrossRef] 23. Zhang, X.; Wang, X.; Wang, D. immobilization of heavy in sewage sludge during land application process in China: A review. Sustainability 2017, 9, 2020. [CrossRef] 24. Kim, J.; Tae, S.; Kim, R. Theoretical study on the production of environment-friendly recycled cement using inorganic construction wastes as secondary materials in South Korea. Sustainability 2018, 10, 4449. [CrossRef] 25. Ivanets, A.I.; Rat, A.I.; Azarova, T.A.; Azarov, S.M.; Al-Khowaiter, S.H.; Al-Harbi, O.; Shemchonok, S.V.; Dobysh, V.A.; Tarasevich, V.A.; Agabekov, V.E.; et al. Preparation and properties of microfiltration

membranes based on natural crystalline SiO2. Ceram. Int. 2014, 40, 12343–12351. [CrossRef] 26. Koner, S.; Pal, A.; Adak, A. Utilization of silica gel waste for adsorption of cationic surfactant and adsolubilization of organics from textile wastewater: A case study. Desalination 2011, 276, 142–147. [CrossRef] 27. Rat’ko, A.I.; Ivanets, A.I.; Azarov, S.M. Effect of additives on the pore structure of ceramics based on

crystalline SiO2. Inorg Mater. 2008, 44, 778–784. [CrossRef] 28. Unob, F.; Wongsiri, B.; Phaeon, N.; Puanngam, M.; Shiowatana, J. Reuse of waste silica as adsorbent for removal by iron oxide modification. J. Hazard. Mater. 2007, 142, 455–462. [CrossRef] 29. Vaiˇciukyniene,˙ D.; Kantautas, A.; Vaitkeviˇcius,V.; Jakeviˇcius,L.; Rudžionis, Ž.; Paškeviˇcius,M. Effects of ultrasonic treatment on zeolite NaA synthesized from by-product silica. Ultrason. Sonochem. 2015, 27, 515–521. [CrossRef] 30. LTD “Alufluor AB” Official Website. Available online: http://www.lifosa.com/en/products-and-services (accessed on 10 February 2018). 31. Aldaco, R.; Garea, A.; Irabien, A. Calcium fluoride recovery from fluoride wastewater in a fluidized bed reactor. Water Res. 2007, 41, 810–818. [CrossRef] 32. Gogoi, S.; Nath, S.K.; Bordoloi, S.; Dutta, R.K. Fluoride removal from groundwater by limestone treatment in presence of phosphoric acid. J. Environ. Manag. 2015, 152, 132–139. [CrossRef] 33. Liu, W.T.; Li, K.C. Application of reutilization technology to calcium fluoride sludge from semiconductor manufacturers. J. Air Waste Manag. Assoc. 2011, 61, 85–91. [CrossRef][PubMed] 34. Baltakys, K.; Iljina, A.; Bankauskaite, A. Thermal properties and application of silica gel waste contaminated with F− ions for C-S-H synthesis. J. Therm. Anal. Calorim. 2015, 121, 145–154. [CrossRef] 35. Casasola, R.; Pérez, J.M.; Romero, M. Effect of fluorine content on glass stability and the crystallisation

mechanism for glasses in the SiO2–CaO–K2O–F system. J. Non-Cryst. Solids 2013, 378, 25–33. [CrossRef] Sustainability 2019, 11, 634 18 of 18

36. Seo, M.D.; Shi, C.B.; Wang, H.; Cho, J.W.; Kim, S.H. Non-isothermal melt crystallization of cuspidine in

CaO–SiO2–CaF2 based glasses. J. Non-Cryst. Solids 2015, 412, 58–65. [CrossRef] 37. Jung, I.H.; Lehmann, J.; Jak, E. Treatise on Process Metallurgy. Volume 2: Process Phenomena. Chapter 5.3—Applications; Elsevier: Oxford, UK, 2014; pp. 675–798. 38. Rungchet, A.; Chindaprasirt, P.; Wansom, S.; Pimraksa, K. Hydrothermal synthesis of calcium sulfoaluminate–belite cement from industrial waste materials. J. Clean. Prod. 2016, 115, 273–283. [CrossRef] 39. Dambrauskas, T.; Baltakys, K.; Eisinas, A.; Siauciunas, R. A study on the thermal stability of kilchoanite synthesized under hydrothermal conditions. J. Therm. Anal. Calorim. 2017, 127, 229–238. [CrossRef] 40. Delong, X.; Yongqin, L.; Ying, J.; Longbao, Z.; Wenkui, G. Thermal behavior of aluminum fluoride trihydrate. Thermochim. Acta 2000, 352–353, 47–52. [CrossRef] 41. Yang, G.Y.; Shi, Y.C.; Liu, X.D.; Mujumdar, A.S. TG-DTG analysis of chemically bound moisture removal of

AlF3·3H2O. Dry Technol. 2007, 25, 675–680. [CrossRef] 42. Krysztafkiewicz, A.; Swit,´ Z.; Jesionowski, T. Evaluation of waste silica precipitated in the process of hydrofluoric acid production from fluosilicic acid. Physicochem. Probl. Miner. Process. 2005, 39, 165–176.

© 2019 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/).