Skip to main content

Low-temperature conversion of Fe-rich sludge to KFeS2 whisker: a new flocculant synthesis from laboratory scale to pilot scale


Herein, a KFeS2 whisker was formed in mass production at a low temperature, with waste cold-rolling sludge as Fe source, which exhibited good performance in the removal of Zn/Ni from real electroplating effluent. At laboratory scale, KFeS2 was generated at 80 °C by the hydrothermal method, and KFeS2 whisker grew radially with the extension of the reaction time. This method was applied at pilot scale, where a similar KFeS2 whisker was also produced with waste cold-rolling sludge as Fe source, and a residual brownish supernatant was observed after the reaction and then completely recycled in the next round for KFeS2 synthesis. After recycling five times, the produced KFeS2 whisker did not change. The drying and storage of KFeS2 have also been verified. Freeze drying and vacuum drying were applicable, whereas air drying was not profitable. Moreover, the efficiency of Zn/Ni removal using undried KFeS2 was similar to that of dried KFeS2. The efficiencies of Zn/Ni removal using KFeS2 were apparently higher those of common reagents for wastewater treatment.


KFeS2 is a fibrous Fe/S-bearing mineral [1] crystallised under high potassium activity and sulphur fugacity [2]. Such conditions are extreme; thus, KFeS2 has not been detected in natural rocks. However, its derivate, rasvumite, co-exists with pegmatites in the mafic environment [3]. KFeS2 is usually synthesised artificially, and it has a special structure, in which one Fe atom is covalent with four sulphur atoms to form a stable tetrahedral (FeS2)nn− bond [4]. The free electrons in large spaces located at edge-sharing (FeS2)2 chains are neutralised by K+ [5] to form a stable Fe-S structure. In the synthesis of KFeS2, when cations, such as Ag+, Ca2+, Sr2+ and Ba2+, are introduced [6, 7], they will be embedded into the inner lattice via the space channel of (FeS2)2 [6], and K+ is released and escapes, resulting in the replacement reaction between cations and K+. Thus, KFeS2 is an important raw material [8, 9] to produce new ternary metal thioferrate products, including RbFeS2 [10, 11], AgFeS2 [7, 12] and CuFeS2 [12]. Such products are essential photovoltaic materials [13, 14] and a photothermal platform for medical therapy [15, 16], thereby increasing the demand for KFeS2.

In the early research, the solid burning method was widely used in the synthesis of KFeS2. For the solid-phase reaction, iron, as a raw material, was mixed with sulphur and K source (e.g., K2CO3 and K2S2) [10, 17] and then burned in a reducing atmosphere [18]. During cauterisation, the oxidation of iron and sulphur occurred, followed by solid conversion to form a (FeS2)2 structure; K+ was located at edge-sharing (FeS2)2 chains to form KFeS2 needle-shaped crystals. For instance, Bronger et al. [18] prepared KFeS2 samples by reacting K2CO3 with iron under a H2S atmosphere at 1000 K for 6 h. After extraction with water and alcohol, well-developed needle-shaped violet KFeS2 crystals were generated. The burning method consumes considerable energy to maintain a high temperature, and it has strict requirements on the reaction system such as a reducing atmosphere, which leads to high costs. Thus, KFeS2 cannot be produced on a large scale. Compared with the burning method, the solvothermal method shows a clear advantage in lowering the reaction temperature. Han et al. [9] mixed KNO3, Fe (NO3)3·9H2O and S powder in ethylenediamine solution; after heating at 190 °C for 18 h, a black KFeS2 powder was obtained. However, in the presence of deionised water, Fe2O3 was predominant in the product [9]. The extensive use and consumption of organic solvents limit their production on a large scale. To date, a more economical method of mass production of KFeS2 remains unknown.

For mass production of KFeS2, reducing the costs is necessary. The high cost of KFeS2 synthesis is primarily due to the chemical reagents used in the raw materials and high temperature. Firstly, weakly crystalline iron compounds can dissolve and release Fe (OH)4 [19] in alkaline solutions, which is the intermediate product before HS substitution reaction in the synthesis of KFeS2. Therefore, the cost of raw materials can be reduced by replacing chemically pure iron sources with iron-containing wastes. Herein, cold-rolling sludge, which was precipitated from Fe-bearing picking wastewater by adjusting the wastewater to pH 8 in a weakly crystallised form, was used as a raw material to explore the potential application of Fe-rich sludge as an Fe source to synthesise KFeS2. Secondly, the solvothermal method can reduce the synthesis temperature compared with the sintering method, but it will consume a large amount of organic solvents and increase the synthesis cost of KFeS2. In the previous study, it was found that KFeS2 can be synthesised in KOH solution at 160 °C. However, Han et al. reported that Fe2O3 was generated simultaneously [9]. Therefore, the synthesis of KFeS2 using the hydrothermal method whilst suppressing the formation of Fe2O3 must be developed. According to the research of hydrothermal synthesis of Fe2O3, the reaction temperature is between 100 and 200 °C [20,21,22,23,24,25]. This finding shows that high temperature is conducive to the formation of Fe2O3, and reducing the temperature may avoid the formation of Fe2O3. Therefore, a method of lowering the reaction temperature to reduce the cost and avoid the formation of Fe2O3 is necessary. Thirdly, given the high potassium activity and sulphur fugacity for the formation conduction of KFeS2 [2], the supernatant was rich in K+ and HS, and it has high alkalinity after KFeS2 synthesis reaction. If the supernatant could be recycled for the next synthesis reaction, then the synthesis cost will be greatly reduced. Finally, the drying and storage methods of KFeS2 must also be considered from the perspective of cost reduction.

In neutral or weakly alkaline solution, the skeleton structure of KFeS2 hydrolysis is spontaneous, and the decomposition of (FeS2)nn− generates several Fe/S-bearing flocs [26]. Such Fe/S-bearing flocs are rich in Fe-SH and Fe-OH groups, which show good affinity for heavy metals in solution [27]. The behaviour of KFeS2 decomposition is similar to the common flocculants, namely, poly aluminium chloride and polymeric ferric sulfuric. However, no reports have been found on KFeS2 applied in wastewater treatment. Electroplating wastewater contains significant quantities of heavy metals, organic compounds and surfactants [28,29,30], which is considered a hazardous source. Given the use of plating additives, heavy metals are complexed with organics to form stable organic-heavy metal ligands [31,32,33,34]. Thus, they are refractory to be removed, although the precipitates (e.g., lime and polymeric ferric sulfuric) are added. Thus, electroplating wastewater is used as a pollutant to verify the performance of KFeS2 in wastewater treatment.

Here, the pilot-scale conversion of waste Fe-rich sludge to KFeS2 was successfully implemented at a low temperature and atmospheric pressure. The upcycling of supernatant in the synthesis of KFeS2 was explored. The storage of prepared KFeS2 was also optimised in the range of freeze drying, vacuum drying, air drying and wet storage. The produced KFeS2 showed superior efficiencies in the removal of Zn/Ni from real electroplating effluent.

Materials and methods


Ferric nitrate nonahydrate (Fe (NO3)3·9H2O, AR) was purchased from Tianjin Yongsheng Fine Chemical Co. Potassium sulphide (K2S, AR) was purchased from Aladdin Co. Potassium hydroxide (KOH, AR) was purchased from Tianjin Hengxing Chemical Reagent Manufacturing Co. Ferrihydrite-bearing sludge (known as sludge) was collected from the sludge warehouse of Guixi Cold-rolling Company (Changchun, China). The sludge contained 41.2% Fe, 50.5% water content and less than 10% carbon. Deionised water was used as experimental water in all lab-scale experiments. In pilot scale, tap water was used.

Laboratory-scale experiment for KFeS2 whisker synthesis

In laboratory-scale experiments, ferrihydrite was synthesised by adding 40 g of Fe (NO3)3·9H2O to 500 mL of deionised water and then mixed with 330 mL of 1 M KOH under stirring at 120 rpm. After 1 h, a mixed solution was generated and placed on the laboratory bench for another 2 h. A brownish sediment was produced at the bottom of the solution, which was collected and washed with deionised three times. The sediment (known as Ferr) was air dried at 60 °C for subsequent use.

The laboratory-scale conversion of Ferr to KFeS2 whisker was performed as follows. Firstly, 1 g of Ferr, 3.3078 g of K2S and 30 mL of 6 M KOH solution were mixed in a 50 mL beaker. Secondly, the beaker was sealed with parafilm and then magnetically stirred at 200 rpm at 50 °C. Thirdly, after 10 h, the beaker was cooled to room temperature and placed at a table for another 2 h. Fourthly, a black deposit was formed at the base of the beaker, which was collected and freeze dried at − 80 °C for 24 h. The dried product was named as E50–10, where E represents the experimental-scale synthesis; 50 is the heating temperature, and 10 is the heating time.

The effect of heating temperature was investigated by changing the heating temperature from 50 to 80 °C following the above-mentioned steps. Then, the corresponding product was named as E80–10. A control experiment was also performed following the above-mentioned steps, where the heating time was extended from 10 h to 24 h, and the products were named as E50–24 and E80–24.

Pilot-scale production of KFeS2 whisker

A pilot-scale vessel was made to synthesise KFeS2 whisker at mass production (Fig. 1). In step 1, sludge (0.3 kg), K2S (1.10 kg) and KOH (3.36 kg) with a molar ratio of Fe: K2S: KOH = 1:5:30 and 10 L of tap water were added to a bucket with a volume of 15 L under stirring at 120 rpm for 1 h to generate a black suspension. In step 2, the suspension was transferred into a sealed vessel and then heated at 80 °C for 24 h. In step 3, the solid sediment and suspension in the vessel were transferred into the suction filter. After filtration, the solid fraction and filter liquor were collected separately. In step 4, the solid fraction (P80–24) was stored in a bucket, in which a small portion of the solid was freeze dried at − 80 °C overnight.

Fig. 1
figure 1

Flow chart of pilot-scale production process

The supernatant generated in step 3 was recycled in the next round (Fig. 1). The total volume of supernatant was adjusted to 10 L with supplementary tap water, followed by adding 0.3 kg of sludge, 0.44 kg of K2S and 0.11 kg of KOH under stirring at 120 rpm. After 1 h, a suspension was generated and then treated following the above-mentioned steps. The generated product was named as P80–24-1 (1 represents the recycle number of supernatant), whilst the generated supernatant was collected again for further use. The recycle experiment of the supernatant was performed four times at pilot scale, and the corresponding products were named as P80–24-2, P80–24-3, P80–24-4 and P80–24-5.

The drying method of KFeS2 was also optimised. Two typical drying methods, vacuum drying at 60 °C and air drying at 105 °C, were investigated using P80–24 as the targeted KFeS2 product, in comparison with that of freeze drying.

Application of KFeS2 whisker in real electroplating wastewater treatment

Electroplating wastewater was treated with polymeric aluminium chloride and precipitant (e.g., diethyldithiocarbamate) in the wastewater plant of Jitong Machinery Company (Changchun, China). The effluent discharge from the wastewater plant was collected and used in this study to determine the performance of the synthesised KFeS2 products. The effluent contained 7.8 mg L− 1 Zn and 0.6 mg L− 1 Ni at pH 7.42, which was treated as follows. Approximately 0.2 g of P80–24 was mixed with 1000 mL of effluent in a 2000 mL beaker under stirring at 100 rpm for 2 h. Subsequently, the beaker was placed on the laboratory bench for 2 h to settle particles, whilst 1 mL of supernatant was sampled for characterisation. Control experiments were performed by changing the P80–24 dosage from 0.2 to 0.5, 1, 3, 5 and 10 g. Other products, including the undried P80–24, P80–10, P80–24-1 and P80–24-5, were also used to treat the effluent in accordance with the above-mentioned method and then compared with common reagents, such as Na2S·9H2O, polymeric ferric sulfuric, sodium diethyldithiocarbamate and lime.

The zeta potential and hydrodynamic radius of P80–24 were also investigated. Approximately 0.1 g of P80–24 was dispersed in 100 mL of effluent under constant stirring at 150 rpm to form a mixture solution. At a given interval, 5 mL of solution was sampled and then determined by a zeta potentiometer (Nano-ZS, Malvern, UK). The experiment of P80–24 in deionised water was also performed in accordance with the above-mentioned method.


The morphologies of the samples were observed by scanning electron microscope (SEM, JSM-6400, Jeol, Japan), and the surface of the samples were sputter-coated with gold prior to observation. The crystallography properties of the samples were characterised by X-ray diffractometer (XRD, Rint2200, Rigaku Corporation, Japan) using Cu-Kα radiation. The valance state of surface elements of the samples was investigated by X-ray photoelectron spectroscopy (XPS, ADES-400, VG Scientific, Britain).

Results and discussion

Laboratory-scale synthesis of KFeS2 whisker

The synthesis of KFeS2 whisker at low temperature was optimised at lab scale. At 50 °C, the product was weakly crystallised, which showed a small rod-shaped precursor (Fig. 2 (E50–10)), although the heating time was extended from 10 to 24 h (Fig. 2 (E50–24)). By increasing the temperature from 50 to 80 °C, the product E80–10 appeared as sharp whisker particles with 0.2 μm diameter and 0.5–1 μm length, which indicated the representative peaks of KFeS2 (Fig. 2 (E80–10)). After the reaction for 24 h, the product E80–24 showed that the peaks of KFeS2 were sharp (Fig. 2 (E80–24)), and its whisker grew radially to 1–4 μm. This finding demonstrated that 80 °C was an optimal temperature for sharp KFeS2 synthesis.

Fig. 2
figure 2

(A) SEM photomicrographs and (B) XRD patterns of E50–10, E50–24, E80–10 and E80–24

Mass production of KFeS2 whisker at pilot scale

Pilot-scale synthesis of KFeS2 was performed at 80 °C for 24 h, and the results are shown in Fig. 3. The sludge was an irregular block (Fig. 3A sludge) that showed typical peaks of ferrihydrite and carbon (Fig. 3B sludge). After the reaction, the product P80–24 was a well-formed whisker that showed sharp peaks of KFeS2 (Fig. 3B (P80–24)), which was similar to E80–24 synthesised at lab scale (Fig. 2 (E80–24)). Although impure carbon was mixed with ferrihydrite in the sludge, the XRD peaks of carbon were not recorded after the reaction, revealing that it was covered by KFeS2 whisker and was not observed by an XRD diffractometer. The above-mentioned findings indicated that mass production of KFeS2 whisker was successfully achieved.

Fig. 3
figure 3

(A) SEM photomicrographs and (B) XRD patterns of the sludge and product P80–24 at pilot scale

Upcycling of supernatant during KFeS2 synthesis

At pilot scale, the supernatant was recycled as an alkaline solution for KFeS2 synthesis in the next round, and the results are shown in Fig. 4. In the first round, the product P80–24-1 showed sharp peaks of KFeS2 and well-formed whisker (Fig. 4), which was similar to that without supernatant recycling (Fig. 3 (P80–24)). After recycling for five times, typical KFeS2 whisker was also observed for the product P80–24-5 (Fig. 4 (P80–24-5)), suggesting that the recycling route of supernatant was applicable for KFeS2 synthesis. The supernatant was highly alkaline; recycling not only reduced KOH consumption and used sufficient HS and S2− for KFeS2 synthesis but also avoided the generation of waste alkaline wastewater.

Fig. 4
figure 4

(A) SEM photomicrographs and (B) XRD patterns of products synthesised in the repeated experiments

Optimisation of the drying method

The prepared P80–24 was dried in three ways, namely, freeze drying, air drying and vacuum drying. In freeze drying, the product was in the form of KFeS2 whisker (Fig. 3 (P80–24)). In comparison with freeze drying, the product from vacuum drying also exhibited well-formed sharp whisker and XRD pattern of KFeS2, although a small portion of broccoli-shaped aggregates was recorded (Fig. 5A (vacuum drying)). Such aggregates were generated by the oxidation of structural S in KFeS2 whisker. However, after air drying at 105 °C, KFeS2 peaks were observed (Fig. 5B (air drying)), but abundant broccoli-shaped aggregates were generated, demonstrating that the oxidation of S was accelerated during air drying. These results demonstrated that freeze drying and vacuum drying were effective for KFeS2 whisker dewatering. Wet P80–24 without dewatering was stored in a sealed bucket for a week, dehydrated and freeze dried again to investigate the storage of KFeS2 whisker. The corresponding product was also in the form of a sharp whisker with clear KFeS2 peaks (Fig. S1), demonstrating that the wet storage of KFeS2 was a desirable route. The wet sample of P80–24 was also used in the wastewater treatment as discussed below.

Fig. 5
figure 5

(A) SEM photomicrographs and (B) XRD patterns of products dried by vacuum-drying and air-drying

Application in raw electroplating wastewater treatment

KFeS2-bearing products were used in the treatment of real electroplating effluent (Fig. 6A). The effluent had a pH of 7.42, and it contained 7.8 mg L− 1 Zn and 0.6 mg L− 1 Ni; it was discharged from the electroplating wastewater plant after the addition of precipitant and coagulant. In the effluent, Zn/Ni was at high concentrations, which should be further removed in accordance with the discharge standard of the electroplating industry (electroplating pollutant emission standards [GB21900–2008]). By adding P80–24, Zn/Ni was apparently removed from 0.33 and 0.21 mg L− 1 with 0.2 g to 0.22 and less than 0.1 mg L− 1 with 1 g, which could not be detected with 10 g, thereby meeting the concentration of Zn and Ni (1 and 0.5 mg L− 1, respectively) in electroplating wastewater discharge standards of China. This result indicated that P80–24 was effective in removing Zn/Ni. The optimal dosage of P80–24 was 1 g, where approximately 97% Zn and 84% Ni were removed, whereas the residual Zn/Ni met the discharge standard of electroplating wastewater (electroplating pollutant emission standards [GB21900–2008]).

Fig. 6
figure 6

Treatment of raw electroplating wastewater. (A) the dosage effect of P80–24 and corresponding final pH; (B) P80–24 compared with the wet P80–24 and other synthesised products; (C) P80–24 compared with other reagents (experimental condition: dosage = 1 g L− 1 (except Fig. 6A) and initial pH = 7.42)

P80–24 and the products with recycling supernatant showed similar removal efficiencies of Zn/Ni (Fig. 6B), suggesting that the supernatant was recyclable in the preparation of KFeS2. The removal of Zn/Ni using the products from vacuum drying and air drying was also investigated (Fig. 6B). The residual Zn/Ni levels were 0.15 and 0.076 mg L− 1 when using the product of vacuum drying and steadily increased to 0.16 and 0.16 mg L− 1 when using the product of air drying, demonstrating that air drying was not desirable in P80–24 drying. The removal performance of undried P80–24 was also investigated, where it had 55% water content; thus, its dosage was 2.22 g after calculating the optimal dosage of dried P80–24. By adding wet P80–24, the residual Zn/Ni levels were 0.16 and 0.051 mg L− 1, which were close to that of dried P80–24; these results revealed that wet P80–24 was efficient in Zn/Ni removal, and freeze drying could be completely omitted. Other common reagents, for example, Na2S·9H2O, lime, polymeric ferric sulfuric and sodium diethyldithiocarbamate, were also used in the removal of Zn/Ni (Fig. 6C), but they did not show desirable removal efficiencies in comparison with P80–24. Thus, P80–24 is an applicable reagent in electroplating wastewater treatment.

After using P80–24, the sharp peaks of KFeS2 disappeared, and only weak peaks of Fe-bearing compound appeared (Fig. 7A). Accordingly, a well-formed whisker was not observed, and only irregular blocks were generated (Fig. 7B), indicating the decomposition of KFeS2 in the effluent. P80–24 was also characterised by X-ray photoelectron spectroscopy before and after use (Fig. 8). For the Fe 2p spectra, a typical peak was recorded at the binding energy of 708.4 eV before use, which belonged to structural Fe in (FeS2)nn− [35], but it varied to the binding energy of 710.5 eV after use; this phenomenon was in agreement with the decomposition of KFeS2 and the formation of Fe/S-bearing compound [36]. For S 2p, four peaks at the binding energies of 160.3, 161.2, 163.2 and 167.4 eV were recorded before use, which were affiliated with structural S in the Fe-S bond, S2−, S and sulphate, respectively. However, two peaks disappeared after use because of the decomposition of KFeS2. A new peak at the binding energy of 162.6 eV appeared, along with the peaks of elemental S and sulphate, which demonstrated the formation of the Fe-S-Zn/Ni bond in the decomposed product of KFeS2 after use.

Fig. 7
figure 7

(A) XRD pattern and (B) SEM photomicrographs of used PK80–24 after electroplating effluent treatment

Fig. 8
figure 8

High-resolution (A) Fe 2p and (B) S 2p XPS curves of P80–24 before and after use

Formation and hydrolysis mechanism of KFeS2

KFeS2 whisker had a one-dimensional linear structure, in which an Fe atom was coordinated with four S atoms. It was stable in alkaline solution at pH > 13.6 and commonly formed in strong alkaline solution. Firstly, when Fe3+ was added in the alkaline solution, it was rapidly polymerised to form Fe-bearing precipitates in weakly crystallised form. The sludge acquired from the cold-rolling company showed characteristics similar to the Fe-bearing precipitates; a small portion of carbon was obtained from the dropped emulsion oil [37]. With the addition of KOH and the hydrolysis of K2S, free OH was abundantly generated in the solution, which eroded the Fe surface of sludge to generate and release Fe (OH)4 into the solution. Accordingly, the Fe concentration increased in the supernatant. The concentration of Fe was 0.045 mg L− 1 at pH 7 (Fig. 9), which rapidly increased to 5.60 mg L− 1 at pH 15.6, suggesting the dissolution of Fe-bearing precipitates and sludge in strong alkaline solution. Secondly, free SH was generated at mass production from the hydrolysis of K2S, which spontaneously replaced OH of free Fe (OH)4 to form Fe (OH)3HS. The replacement reaction continued, where Fe/S-bearing products, for example, Fe (OH)3HS and Fe (OH)2(HS)2, were generated. Thirdly, the conjunction reaction between two newly formed Fe/S-bearing products occurred to form (FeS2) Fe (OH)3HS2−. Such products were sparingly soluble in alkaline solution and precipitated from the solution. The concentration of Fe was residual at 0.184 mg L− 1 in the supernatant after the reaction (Fig. 9); this residual level was lower than that in pure KOH solution, demonstrating that Fe/S-bearing products were formed and spontaneously precipitated from the solution. The conjunction reaction continued, which accelerated the polymerisation of Fe/S-bearing products, with the generation of linear (FeS2)nn− as the final product. Fourthly, in the (FeS2)nn− structure, the negative charge was neutralised by free K+, resulting in the formation of one-dimensional KFeS2 whisker. The related formation process is shown in Fig. 10. However, the reaction of HS replaced OH of Fe (OH)4 to form Fe (OH)3HS was commonly endothermic, which was slow at 50 °C, thereby KFeS2 was not formed even for 24 h.

Fig. 9
figure 9

Fe concentration of the supernatant after (A) KFeS2 synthesis and (B) alkaline leaching of Fe-bearing sludge

Fig. 10
figure 10

Schematic representation of the formation and hydrolysis mechanism of KFeS2

Some impurities such as Cr, Mn, Si and Al showed some characteristics in the presence of S2− and alkaline solution, which affected the formation of the (FeS2)nn− group. For instance, when impurities such as Cr and Mn were present in the sludges, the redox reaction between Cr/Mn and S2− occurred, with the generation and release of free OH to solution. Therefore, the released OH were accumulated in the solution, which promoted the formation of Fe (OH)4, polymerisation of the (FeS2)nn− group and crystallisation growth [27, 38]. Conversely, the impurities of Si/Al-bearing minerals were easily dissolved in the alkaline solution, which not only consumed extra OH, but also spontaneously polymerised and crystallised new Si/Al-bearing products, such as sodalite and cancrinite. Other elements, for example, Ca and Co, were reacted with S2− to form corresponding sulphide [39]. Such impure elements did not coordinate into the crystal structure of (FeS2)nn−. Here, impure carbon was not involved in KFeS2 synthesis, and it did not accumulate in the supernatant. After the reaction, the supernatant was alkaline and rich in HS, which could serve as a cyclable resource to prepare KFeS2 with supplementary K2S and KOH. Thus, the dosage of K2S and KOH was considerably reduced. Temperature was an important parameter in KFeS2 synthesis. As the temperature increased from 50 to 80 °C, the reaction between OH and the surface Fe of sludge and the release of Fe (OH)4 to solution accelerated, which used sufficient Fe (OH)4 for the polymerisation and crystallisation of KFeS2 whisker. Accordingly, high temperature was an important route to reduce the reaction time. For instance, linear KFeS2 particles were generated after hydrothermal treatment at 190 °C for 18 h [9]. The drawback of high-temperature treatment in water was the formation of hematite from the rapid polymerisation of the surface of the Fe-OH group of sludge [9, 40, 41].

Before its application in wastewater treatment, the storage of KFeS2 was a key step. Wet KFeS2 particles remained stable for a week and showed a similar effect to freeze-dried KFeS2 in the removal of heavy metals from effluent. Apart from wet storage, vacuum drying could be used as an alternative method to KFeS2 storage, where the dried product showed a similar effect to freeze drying in Zn/Ni removal. During air drying, the redox reaction between oxygen and structural S of KFeS2 occurred. This phenomenon led to the consumption of KFeS2 and accordingly decreased Zn/Ni removal efficiency in comparison with freeze drying.

In the effluent, heavy metals were complexed with organics to form stable organic-heavy metal ligands; therefore, they were refractory to be removed, although the precipitates (e.g., lime and polymeric ferric sulfuric) were added. When KFeS2 was added in the electroplating effluent, it was spontaneously decomposed to generate Fe/S-bearing flocs with numerous Fe-SH and Fe-OH groups [26]. Such flocs were negatively charged (Fig. 11A), which had an average hydrodynamic radius of 600 nm (Fig. 11B). Subsequently, heavy metals, for example, Zn and Ni, were coordinated onto the Fe-S/Fe-O groups, resulting in the removal of Zn/Ni from effluent (Fig. 10). In comparison with the hydroxyl group, the new -SH group had strong affinity for complex heavy metals because S had a larger atomic radium than O, and it was more electronegative to from the -S-Me group than O [42]. After heavy metal coagulation, the zeta potential of flocs apparently increased from − 50 to − 35 mV, where its radium considerably increased to 3500 nm, demonstrating the polymerisation of flocs in the removal of Zn/Ni. Floc polymerisation continued when stirring was slow and/or stopped, resulting in the generation of heavy metal-bearing sludge.

Fig. 11
figure 11

(A) Zeta potential and (B) hydrodynamic radius of PK80–24 during its hydrolysis in deionised water and electroplating wastewater

Environmental application

The conversion of cold-rolling sludge to KFeS2 was performed at pilot scale, and the product KFeS2 whisker showed superior efficiency in the treatment of real electroplating effluent containing Zn/Ni. The total cost of KFeS2 synthesis was calculated (Table 1). In the first round, the conversion of sludge to KFeS2 whisker was performed on the basis of the optimal molar ratio of Fe: K2S: KOH = 1:5:30. This conversion required 3.45 t of K2S, 10.56 t of KOH, 31.5 t of water and 840.5 kWh power, which amounted to USD 7075. After collecting the produced KFeS2 whisker, the remaining supernatant was rich in K2S and KOH, which was recycled completely in the second round. 1.38 t of K2S and 0.34 t of KOH were supplemented to maintain the optimal molar ratio for KFeS2 synthesis, the reagent cost was only USD 1073, nearly 15% of that in the first round, and the total cost was USD 1254. In addition, the wet KFeS2 showed similar performance to the dried one, suggesting that drying can be omitted, and the wet storage of KFeS2 was also acceptable. This behaviour apparently reduced the total cost in the second round for KFeS2 synthesis. However, about USD 373.3 was spent in the disposal of 1000 kg of sludge [43], and this amount can be deducted from the total cost of KFeS2 synthesis. The KFeS2-bearing product was marketable because of its performance in electroplating wastewater treatment. Therefore, the recycling of cold-rolling sludge as a KFeS2-bearing product was profitable.

Table 1 Total cost of the recycling supernatant for KFeS2 synthesis from cold-rolling sludge

Other Fe3+-bearing sludge was also produced as a solid waste in the steel-making, dye chemical and mineral industries; it could function as an Fe3+-bearing resource, which can be recycled as a KFeS2 whisker. Such recycling not only saved the disposal cost of sludge but also produced new a Fe/S-bearing product, thereby exhibiting acceptable application in these industries.


The conversion of Fe3+-bearing sludge to KFeS2 whisker at low temperature was performed successfully. In the lab-scale experiment, ferrihydrite was used as an Fe resource; after treatment at 80 °C for 24 h, a well-formed KFeS2 whisker was obtained. At pilot scale, the cold-rolling sludge was used as Fe3+-bearing resource, and KFeS2 whisker was also produced at mass production. The generated supernatant was completely recycled by the supplement of K2S, KOH and tap water for KFeS2 synthesis in the next round. After supernatant cycling for five times, the product was also in the form of KFeS2 whisker. Freeze drying and vacuum drying were desirable methods to dry KFeS2, except air drying. The KFeS2 whisker was effective for treating real electroplating effluent containing Zn and Ni. Moreover, wet KFeS2 and the product obtained from supernatant recycling showed similar removal efficiencies of Zn/Ni to KFeS2 whisker obtained freeze drying, which exhibited a simple and convenient method of storing KFeS2 products.

Availability of data and materials

All data generated or analyzed during this study are available from the corresponding author on reasonable request.


  1. Tiwary SK, Vasudevan S. Single crystal magnetic susceptibility of the quasi-one-dimensional antiferromagnet KFeS2. Solid State Commun. 1997;101:449–52.

    Article  Google Scholar 

  2. Osadchii VO, Voronin MV, Baranov AV. Phase equilibria in the KFeS2–Fe–S system at 300–600 °C and bartonite stability. Contrib Mineral Petr. 2018;173:44.

    Article  Google Scholar 

  3. Amthauer G, Bente K. Mixed-valent iron in synthetic rasvumite, KFe2S3. Naturwissenschaften. 1983;70:146–7.

    Article  Google Scholar 

  4. Debiasi RS, Taft CA. Magnetic resonance in KFeS2 single crystals. J Mater Sci. 1978;13:2274–75.

    Article  Google Scholar 

  5. Allali N, Favard JF, Rambaud M, Goloub A, Danot M. Low-temperature reactions using potassium iron disulfide as a precursor. Mater Res Bull. 1994;29:135–42.

    Article  Google Scholar 

  6. Boller H. Faserförmige Erdalkali-thioferrate [Fibrous alkaline earth thioferrates]. Monatshefte für Chemie. 1978;109:975–85 [in German].

    Article  Google Scholar 

  7. Galembeck A, Alves OL. Thermal behavior of α-Ba (FeS2)2 and AgFeS2: quasi-unidimensional and 3D network compounds prepared from ion-exchange on the same precursor. J Mater Sci. 1999;34:3275–80.

    Article  Google Scholar 

  8. Guo JG, Chen XL, Wang G, Jin SF, Zhou TT, Lai XF. Effect of doping on electrical, magnetic, and superconducting properties of KxFe2-yS2. Phys Rev B. 2012;85:054507.

    Article  Google Scholar 

  9. Han I, Jiang ZL, dela Cruz C, Zhang H, Sheng HP, Bhutani A, et al. Accessing magnetic chalcogenides with solvothermal synthesis: KFeS2 and KFe2S3. J Solid State Chem. 2018;260:1–6.

    Article  Google Scholar 

  10. Guy JK, Spann RE, Martin BR. Solid state ion exchange chemistry of the solid solution KxRb1−xFeS2. Solid State Ionics. 2008;179:409–14.

    Article  Google Scholar 

  11. Arguello Z, Torriani I, Furtado NC, Arsenio TP, Taft CA. The growth of single crystals of KFeS2 and RbFeS2 by the Bridgman method. J Cryst Growth. 1984;67:483–5.

    Article  Google Scholar 

  12. Boon JW. The crystal structure of chalcopyrite (CuFeS2) and AgFeS2: the permutoidic reactions KFeS2 → CuFeS2 and KFeS2 → AgFeS2. Recl Trav Chim Pay-B. 1944;63:69–80.

    Article  Google Scholar 

  13. Sciacca B, Yalcin AO, Garnett EC. Transformation of Ag nanowires into semiconducting AgFeS2 nanowires. J Am Chem Soc. 2015;137:4340–3.

    Article  Google Scholar 

  14. Nsude HE, Nsude KU, Whyte GM, Obodo RM, Iroegbu C, Maaza M, et al. Green synthesis of CuFeS2 nanoparticles using mimosa leaves extract for photocatalysis and supercapacitor applications. J Nanopart Res. 2020;22:352.

    Article  Google Scholar 

  15. Kiiamov A, Lysogorskiy Y, Seidov Z, von Nidda HAK, Tsurkan V, Tayurskii D, et al. Vibrational properties and lattice specific heat of KFeS2. AIP Conf Proc. 2018;2041:040002.

    Article  Google Scholar 

  16. Ding BB, Yu C, Li CX, Deng XR, Ding JX, Cheng ZY, et al. cis-Platinum pro-drug-attached CuFeS2 nanoplates for in vivo photothermal/photoacoustic imaging and chemotherapy/photothermal therapy of cancer. Nanoscale. 2017;9:16937–49.

    Article  Google Scholar 

  17. Johnston DC, Mraw SC, Jacobson AJ. Observation of the antiferromagnetic transition in the linear chain compound KFeS2 by magnetic susceptibility and heat capacity measurements. Solid State Commun. 1982;44:255–8.

    Article  Google Scholar 

  18. Bronger W, Kyas A, Muller P. The antiferromagnetic structures of KFeS2, RbFeS2, KFeSe2, and RbFeSe2 and the correlation between magnetic moments and crystal field calculations. J Solid State Chem. 1987;70:262–70.

    Article  Google Scholar 

  19. Diakonov II, Schott J, Martin F, Harrichourry JC, Escalier J. Iron (III) solubility and speciation in aqueous solutions. Experimental study and modelling: part 1. Hematite solubility from 60 to 300 °C in NaOH-NaCl solutions and thermodynamic properties of Fe (OH)4−(aq). Geochim Cosmochim Ac. 1999;63:2247–61.

    Article  Google Scholar 

  20. Yu J, Saada H, Sojic N, Loget G. Photoinduced electrochemiluminescence at nanostructured hematite electrodes. Electrochim Acta. 2021;381:138238.

    Article  Google Scholar 

  21. Tadic M, Kopanja L, Panjan M, Lazovic J, Tadic BV, Stanojevic B, et al. Rhombohedron and plate-like hematite (α-Fe2O3) nanoparticles: synthesis, structure, morphology, magnetic properties and potential biomedical applications for MRI. Mater Res Bull. 2021;133:111055.

    Article  Google Scholar 

  22. Gou XL, Wang GX, Park J, Liu H, Yang J. Monodisperse hematite porous nanospheres: synthesis, characterization, and applications for gas sensors. Nanotechnology. 2008;19:125606.

    Article  Google Scholar 

  23. Liang HF, Chen W, Yao YW, Wang ZC, Yang Y. Hydrothermal synthesis, self-assembly and electrochemical performance of α-Fe2O3 microspheres for lithium ion batteries. Ceram Int. 2014;40:10283–90.

    Article  Google Scholar 

  24. Lu Y, Sun Z, Huo MX. Fabrication of a micellar heteropolyacid with Lewis-Bronsted acid sites and application for the production of 5-hydroxymethylfurfural from saccharides in water. RSC Adv. 2015;5:30869–76.

    Article  Google Scholar 

  25. Ruiz-Gomez MA, Rodriguez-Gattorno G, Figueroa-Torres MZ, Obregon S, Tehuacanero-Cuapa S, Aguilar-Franco M. Role of assisting reagents on the synthesis of α-Fe2O3 by microwave-assisted hydrothermal reaction. J Mater Sci-Mater El. 2021;32:9551–66.

    Article  Google Scholar 

  26. Wang ZH, Liu YW, Qu Z, Su T, Zhu SY, Sun T, et al. In situ conversion of goethite to erdite nanorods to improve the performance of doxycycline hydrochloride adsorption. Colloid Surface A. 2021;614:126132.

    Article  Google Scholar 

  27. Liu YW, Khan A, Wang ZH, Chen Y, Zhu SY, Sun T, et al. Upcycling of electroplating sludge to prepare erdite-bearing nanorods for the adsorption of heavy metals from electroplating wastewater effluent. Water-Sui. 2020;12:1027.

    Google Scholar 

  28. Belova L, Vialkova E, Glushchenko E, Burdeev V, Parfenov Y. Treatment of electroplating wastewaters. In: E3S Web of Conferences. Blagoveshchensk: EDP Sciences; 2020.

  29. Scarazzato T, Panossian Z, Tenorio JAS, Perez-Herranz V, Espinosa DCR. A review of cleaner production in electroplating industries using electrodialysis. J Clean Prod. 2017;168:1590–602.

    Article  Google Scholar 

  30. Chen D, Zhang CS, Rong HW, Zhao MH, Gou SY. Treatment of electroplating wastewater using the freezing method. Sep Purif Technol. 2020;234:116043.

    Article  Google Scholar 

  31. Andrus ME. A review of metal precipitation chemicals for metal-finishing applications. Met Finish. 2000;98:20–3.

    Article  Google Scholar 

  32. Wang H, Wang H, Zhao H, Yan Q. Adsorption and Fenton-like removal of chelated nickel from Zn-Ni alloy electroplating wastewater using activated biochar composite derived from Taihu blue algae. Chem Eng J. 2020;379:122372.

    Article  Google Scholar 

  33. Chen C, Chen AQ, Huang XF, Ju R, Li XC, Wang J, et al. Enhanced ozonation of Cu (II)-organic complexes and simultaneous recovery of aqueous Cu (II) by cathodic reduction. J Clean Prod. 2021;298:126837.

    Article  Google Scholar 

  34. Wang Q, Yu JX, Chen XY, Du DT, Wu RR, Qu GZ, et al. Non-thermal plasma oxidation of Cu (II)-EDTA and simultaneous Cu (II) elimination by chemical precipitation. J Environ Manage. 2019;248:109237.

    Article  Google Scholar 

  35. Bronold M, Pettenkofer C, Jaegermann W. Surface analysis investigations on the reaction of FeS2 with alkali metals. Ber Bunsen Phys Chem. 1991;95:1475–9.

    Article  Google Scholar 

  36. Li J, Xu YL, Zhang Y, He C, Li TT. Enhanced redox kinetics of polysulfides by nano-rod FeOOH for ultrastable lithium-sulfur batteries. J Mater Chem A. 2020;8:19544–54.

    Article  Google Scholar 

  37. Liu B, Zhang SG, Pan DA, Chang CC. Synthesis and characterization of micaceous iron oxide pigment from oily cold rolling mill sludge. Procedia Environ Sci. 2016;31:653–61.

    Article  Google Scholar 

  38. Zhu SY, Liu YW, Huo Y, Chen Y, Qu Z, Yu Y, et al. Addition of MnO2 in synthesis of nano-rod erdite promoted tetracycline adsorption. Sci Rep-UK. 2019;9:16906.

    Article  Google Scholar 

  39. Hu TK, Wang HM, Ning RY, Qiao XL, Liu YW, Dong WQ, et al. Upcycling of Fe-bearing sludge: preparation of erdite-bearing particles for treating pharmaceutical manufacture wastewater. Sci Rep-UK. 2020;10:12999.

    Article  Google Scholar 

  40. Zhu SY, Song X, Chen Y, Dong G, Sun T, Yu HB, et al. Upcycling of groundwater treatment sludge to an erdite nanorod as a highly effienct activation agent of peroxymonosulfate for wastewater treatment. Chemosphere. 2020;252:126586.

    Article  Google Scholar 

  41. Zhu SY, Lin X, Dong G, Yu Y, Yu HB, Bian DJ, et al. Valorization of manganese-containing groundwater treatment sludge by preparing magnetic adsorbent for Cu (II) adsorption. J Environ Manage. 2019;236:446–54.

    Article  Google Scholar 

  42. Chen Y, Li H, Wang ZP, Tao T, Hu C. Photoproducts of tetracycline and oxytetracycline involving self-sensitized oxidation in aqueous solutions: effects of Ca2+ and Mg2+. J Environ Sci-China. 2011;23:1634–9.

    Article  Google Scholar 

  43. Zhu SY, Li T, Wu YQ, Chen Y, Su T, Ri KH, et al. Effective purification of cold-rolling sludge as iron concentrate powder via a coupled hydrothermal and calcination route: from laboratory-scale to pilot-scale. J Clean Prod. 2020;276:124274.

    Article  Google Scholar 

Download references


The authors wish to thank the National Natural Science Foundation of China Grant (No. 52070038), National Key Research and Development Program of China (Grant No. 2019YFE0117900) and the Science and Technology Program of Jilin Province (Grant No. 20190303001SF).


This work was supported by the National Natural Science Foundation of China (Grant No. 52070038), the National Key Research and Development Program of China (Grant No. 2019YFE0117900) and the Science and Technology Program of Jilin Province (Grant No. 20190303001SF).

Author information

Authors and Affiliations



Dongxu Liang: Writing - Original Draft. Yu Chen: Writing - Review & Editing. Suiyi Zhu: Conceptualization, Resources, Funding acquisition, Project administration. Yidi Gao: Data Curation, Formal analysis. Tong Sun: Investigation, Validation. Kyonghun Ri: Visualization, Validation. Xinfeng Xie: Conceptualization, Supervision. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Suiyi Zhu.

Ethics declarations

Competing interests

The authors declare they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liang, D., Chen, Y., Zhu, S. et al. Low-temperature conversion of Fe-rich sludge to KFeS2 whisker: a new flocculant synthesis from laboratory scale to pilot scale. Sustain Environ Res 31, 25 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: