1 Introduction

Pollution of the ecosystem due to toxic non-degradable heavy metals presents a serious health hazard (Doabi et al., 2018; Khan et al., 2021; Pakade et al., 2017). One such heavy metal is chromium, which exhibits toxicity by being able to penetrate biological cell membranes resulting in cancer (Deepak et al., 2021; Fu et al., 2020; Zhang et al., 2020). Chromium can be discharged in several ways some of which include industrial activities and natural causes such as volcanic eruptions, which lead to seepage of chromium compounds into the environment (Pakade et al., 2016; Rahman & Singh, 2019). Chromium has a variety of oxidation states. The rare oxidation states include ‘ + 1’, ‘ + 4’ and ‘ + 5’ (Bansal et al., 2019; Farouk et al., 2020). The most common oxidation states include the trivalent Cr(III) and hexavalent Cr(VI) states (Park, 2020; Edwards et al., 2020). Hexavalent chromium is the most stable and persistent form of chromium in the environment (Zhao et al., 2018; Zhu et al., 2020). Hexavalent chromium is also the most toxic form of chromium as it is highly mobile and can easily pass through cell membranes, where it is reduced to Cr(III), making it carcinogenic, resulting in stomach ulcers and mitochondrial functional damage (Zhao et al., 2018; Bakshi & Panigrahi, 2018; Sadeghi et al., 2019; Noah et al., 2020; Zhu et al., 2020; Wang et al., 2021; Huang et al., 2022).

Agencies, such as the World Health Organisation (WHO), have set regulations for maximum tolerable concentration limits of 0.05 mg L−1 and 0.1 mg L−1 for Cr(VI) in drinking water and inland surface water, respectively (WHO, 2003; Maitlo et al., 2021). Furthermore, a local study by Edokpayi et al. (2018) reported that the carcinogenic risk due to Cr exposure, in a number of boreholes studied in Muledane village situated in Limpopo, South Africa, exceeded the 50-µg L−1 risk limit for chromium in drinking water as set by the Department of Water Affairs and Forestry (DWAF) (1996) and WHO (2006) (Loock-Hattingh, 2016). Sources of pollution in various water bodies such as rivers, lakes and boreholes include discharge of wastewater from domestic, agricultural and industrial activities (Jalees et al., 2021; Buba & Maina, 2020). Other causes of pollution in groundwater sources include the geology of the aquifers, climate and anthropogenic activities (Edokpayi et al., 2018; Kurwadkar et al., 2020). Different techniques have been used for the removal of heavy metals which include ion exchange, solvent extraction and chemical precipitation (Alguacil et al., 2008; Bashir et al., 2019; Chen et al., 2018; Dong et al., 2018; Feng et al., 2019; Sorouraddin et al., 2017; Xie et al., 2019; Ye et al., 2019). These methods have been accompanied by setbacks; for instance, ion exchange often requires that concentrations of more species need to be measured to ascertain it (Crist et al., 2002), solvent extraction is generally expensive and is often results in poor extraction efficiencies (Nguyen & Lee, 2018), while chemical precipitation suffers from poor solubility of chromium sulphides and requires large doses of precipitation agents (Pohl, 2020). Adsorption with the use of a suitable adsorbent can be a more effective technique for the removal of heavy metals in aqueous systems (Saleh et al., 2020; Xiaodong et al., 2019). Ferric oxide (Son et al., 2018) and alumina (Chang et al., 2020) have been used as adsorbents for heavy metals due to their high adsorption capacity, but they are difficult to separate from the aqueous system after use and they are expensive. Therefore, there is a need to develop low cost and readily regenerative adsorbents.

Coal fly ash (CFA) is an example of a low-cost adsorbent, generated as a waste material during combustion of coal (Mokgehle et al., 2019a; Vu et al., 2020). The abundance of CFA makes it difficult to dispose. Due to the voluminous nature of CFA, stockpiled CFA during heavy rains leach heavy metal ions into water bodies contaminating and disrupting the ecosystem (Mokgehle et al., 2019b; Hwang et al., 2020). Some of the toxic effects of CFA include increased risk of developing cancer and other diseases (Whiteside & Herndon, 2018). Efforts have been undertaken to recycle CFA. Approximately, 20% of the CFA produced is being used as building material (Sanjuán et al., 2021). Despite the uses of CFA, a large portion of this material remains at landfill sites and still poses an environmental hazard. As a remediation strategy, CFA has therefore been studied as a sorbent for heavy metal ions (Lei et al., 2020; Xie et al., 2014). However, application of CFA in adsorption has often been accompanied by setbacks which include limited adsorption sites. To address these limitations and improve the adsorption performance, this work was directed at the application of hydrothermally treated CFA-derived zeolites for the adsorption of chromium via the batch technique. The hydrothermal synthesis method is suitable as it firstly uses a green solvent, water, for producing zeolites. Secondly, it is an appropriate method for using an abundant waste material (CFA) for synthesis of zeolites.

2 Experimental

2.1 Chemicals and Reagents

CFA was collected from a Modderfontein (Johannesburg, South Africa) steam plant. Al2(SO4)3.18H2O was purchased from Associated Chemical Enterprises (Johannesburg, South Africa), and SiO2 was purchased from Sigma-Aldrich (Johannesburg, South Africa), which were used as standards in the analysis of dissolved Si4+ and Al3+ species. NaOH pellets, HNO3 and K2Cr2O7, were purchased from Rochelle Chemicals (Johannesburg, South Africa). The NaOH pellets were used as alkaline-activating agents for dissolution of the aluminosilicate matrix in CFA. Ultra-pure water was purified using a Direct-Q 5UV distiller (MA, USA) with a conductivity and resistance of 0.055 µS cm−1 and 18.2 mΩ respectively. A magnetic stirrer was used to stir CFA and NaOH. Filtration was done using a filter paper to separate the slurry from the NaOH-activating solution.

2.2 Instruments

Analysis of the dissolved ions from the CFA aluminosilicate matrix were performed on a PinAAcle 900 T flame atomic absorption spectrometer (AAS), purchased from PerkinElmer (MA, USA). A Heidolph temperature probed magnetic stirrer from Heidolph Instruments (Schwarzenberg, Germany) was used to stir CFA and the alkaline activating solution. Filtration was done using a Millipore 0.2-μm membrane filter to separate the slurry from the NaOH-activating solution. A CN-2060 centrifuge acquired from Monitoring and Control Laboratories (Johannesburg, South Africa) was used to separate the slurry from the NaOH-activating solution. Analysis of the dissolved ions from the CFA aluminosilicate matrix was performed on a PinAAcle 900 T graphite furnace atomic absorption spectrometer (AAS) procured from PerkinElmer (MA, USA).

A 278AC Brass Parr Bomb purchased from the Parr Instrument Company (IL, USA) was used to perform crystallization of the CFA slurry. The Parr bomb was placed in a Labotec EcoTherm oven (Johannesburg, South Africa). Molecular interactions were characterized using Fourier transform infrared (FT-IR) spectroscopy from Bruker (MA, USA). A Bruker S1 Titan/Tracer XRF analyser (Cramerview, South Africa) was used to determine the chemical composition of the materials.

2.3 Ageing Studies

During ageing studies, the concentrations of 0.5–1.5 mol L−1 NaOH were prepared. NaOH solutions were prepared and then mixed with 15 g CFA in a 250-mL polypropylene sealable bottles and stirred at 400 rpm for 24 h at 50 °C to determine the extent of dissolution of the CFA aluminosilicate matrix. After the stirring period had elapsed, the slurry was filtered. The filtrate was then analysed for Si4+ and Al3+ using FAAS. The next study evaluated the effect of ageing time where the optimized concentration of NaOH was used to determine the optimum time for complete dissolution of the CFA Al matrix into the activating agent. The optimized NaOH concentration, 50 mL, was mixed with 15 g CFA in a 250-mL polypropylene sealable bottle and stirred at 400 rpm for a variety of time intervals (6–48 h) at 50 °C. After the stirring periods had elapsed, the slurry was filtered. The filtrate was then analysed for Si4+ and Al3+ using FAAS. The concluding ageing study looked at the effect of the solid/liquid ratio. The optimized NaOH concentration and ageing time were used to investigate the dissolution of Si4+ and Al3+ as a function of solid/liquid ratio (0.15–0.5 g mL−1). These mixtures were placed in 250-mL polypropylene sealable bottles and stirred at 400 rpm at 50 °C. After stirring, the slurry was filtered, and the filtrate was analysed for Si4+ and Al3+ content using a FAAS. Following the optimized ageing conditions, crystallization studies were conducted where the effect of crystallization time was examined.

2.4 Crystallization Studies

A 15-mL slurry from ageing studies in Sec 2.3 was poured into a parr bomb and crystallized in an oven at 140 \(^\circ{\rm C}\) for 6–72 h to form zeolites. Thereafter, temperature studies were conducted by pouring 15 mL slurry from the ageing studies, synthesized from the optimum conditions, into a parr bomb. This unit was then placed in an oven for crystallization at temperatures ranging from 35 to 200 °C under a period equivalent to the optimum time. Analyses were then done on the synthesized zeolites via XRF and FTIR. This was followed by a study evaluating the effect of water content on crystallization where the solid (slurry):liquid (water) ratios were 5:10, 7.5:7.5 and 15:0 (v/v). The mixtures were poured into the parr bomb and placed into an oven for crystallization. This was done under the optimum conditions of time and temperature. FT-IR analysis was then done on the resultant zeolites.

2.5 Adsorption of Chromium

Batch adsorption studies were then done evaluating the effect of mass of zeolite, which ranged from 10 to 80 mg, on the extraction of chromium. The conditions for chromium adsorption were 0.5 g mL−1 Cr, at a stirring rate of 100 rpm at 25 °C for 3 h each. After the stirring period had elapsed, the mixture was filtered, and the filtrate was analysed for chromium using FAAS. The next batch study evaluated the effect of contact time (30–420 min) which was performed using the optimum mass of zeolite from the previous study. After the stirring period had elapsed, the mixture was filtered, and the filtrate was analysed for chromium using FAAS. This was followed by the effect of concentration (0.1–0.9 mg L−1) studies. The concluding study evaluated the effect of temperature (25–75 °C) on chromium adsorption. All the filtered solutions were analysed on the FAAS. All the adsorption experiments were conducted in duplicates.

3 Results and Discussion

3.1 Synthesis of Zeolite via Hydrothermal Treatment

3.1.1 Ageing Studies

Effect of Concentration of NaOH on the Dissolution of the Aluminosilicate Matrix of CFA.

The surface of the CFA particle was composed of a highly reactive aluminosilicate layer. The outer layer of CFA composed of the aluminosilicate layer was mainly made of SiO2 while the aluminium content was concentrated in the interior, also referred to as the mullite layer. The trend observed amongst the three concentrations studied indicated that as the concentration of NaOH was increased, the dissolution of the CFA aluminosilicate layer into NaOH increased (Fig. 1) (Mokgehle et al., 2019a). The disproportionately larger concentration of Si4+ dissolved compared to Al3+, indicating that a large portion of CFA was composed of aluminosilicate, which had a high silica content. It was also observed that as the alkali concentration was increased, dissolution of Al3+ increased. This was mainly due to the high pH of the alkaline solution (1.5 M NaOH), which penetrated the inner mullite phase resulting in an increased concentration of Al3+ compared to the 0.5-M and 1.0-M NaOH alkaline solutions (Fig. 1). Mokgehle et al. (2019a) and Inada et al. (2005) also reported on a similar trend. An optimum dissolution of 1.5 M NaOH was then used in subsequent experiments.

Fig. 1
figure 1

Concentration of Al3+ and Si.4+ at different NaOH concentrations. (Experimental conditions; reaction duration — 24 h, temperature — 50 \(^\circ{\rm C}\), mass of CFA — 15 g, volume of NaOH — 50 mL)

Full size image

Effect of Time on the Dissolution of the Aluminosilicate Matrix of CFA

Studies were performed to investigate the effect of time on the dissolution of the Al3+ and Si4+ matrix from CFA. Ageing times ranging from 6 to 48 h were studied using the optimum concentration of 1.5 M NaOH. FTIR analysis of the dried CFA residuals after different times, with CFA as a reference, is shown in Fig. 2. The band occurring at 930 cm−1 was associated with symmetric stretching of Al–O and Si–O (Fernández-Jiménez & Palomo, 2005; Musyoka et al., 2012; Mokgehle et al., 2019a). Additionally, it was also noted that these stretches appeared sharper and slightly shifted towards higher frequencies compared to CFA (Fig. 2), as observed for the 48-h peak. This suggested that the incorporation of Na+ into the CFA aluminosilicate from the dissolution, NaOH, matrix influenced the asymmetric stretching of Al–O and Si–O (Mokgehle et al., 2019a). Hence, the 48-h ageing time was taken as the optimum and used in subsequent experiments. Experiments conducted for longer than 48 h resulted in damage to the base of the polypropylene bottle.

Fig. 2
figure 2

FTIR spectra of residual products as time was varied

Full size image

In Table 1, the XRF data comparing the percentage composition of elements in the residual product as a function of time is shown. Lower percentages of Al2O3 were observed in the CFA residues after 48 h in comparison to CFA (Table 1). This indicated that during ageing, Al initially present on the mullite phase of CFA dissolved into the NaOH-activating agent solution (Mokgehle et al., 2019a). Similar studies reporting on the dissolution of the aluminium containing mullite phase into NaOH were reported by Joseph et al. (2020), Monasterio-Guillot et al. (2020), Ren et al. (2020), and Cao et al. (2021). Therefore, a higher dissolution of the alumino-silicate layer occurred at 48 h and was in agreement with the data presented in Fig. 2.

Table 1 XRF data for the effect of time on the dissolution of the aluminosilicate layer of CFA
Full size table

Effect of Volume of NaOH on the Dissolution of the Aluminosilicate Matrix of CFA

Studies were performed to investigate the effect of volume of NaOH used for the dissolution of Al3+ and Si4+ from CFA. As the volumes of 1.5 M NaOH increased from 30 to 100 mL, a proportional increase of Al3+ was observed, while concentrations of Si4+ generally remained the same (Fig. 3). Therefore, this indicated that more of the aluminosilicate matrix dissolved into solution when a larger volume of NaOH was used (Kaze et al., 2018; Mokgehle et al., 2019a, b, Kuenzel & Ranjbar, 2019). Therefore, the 100-mL volume of 1.5 M NaOH was the optimum.

Fig. 3
figure 3

Concentration of Al and Si for three different volumes of NaOH. (Experimental conditions; reaction duration — 48 h, temperature — 50 \(^\circ{\rm C}\), mass of CFA — 15 g, concentration of NaOH — 1.5 M)

Full size image

FTIR studies on the CFA dried residual products evaluating the effect of volume are shown in Fig. 4. Bands at 930 cm−1 associated with Al–O and Si–O were most intense for the residual where 100 mL of NaOH was used (Mokgehle et al., 2019b). This indicated that the inclusion of Na+ from the NaOH-activating agent, into the aluminosilicate layer of the CFA matrix, influenced the asymmetric stretching of Al–O and Si–O (Mokgehle et al., 2019a, b). Hence, the residue corresponding to when 100 mL of NaOH was used was observed as the optimum and is in consensus with the results from the FAAS studies in Fig. 3.

Fig. 4
figure 4

FTIR spectra of CFA residual products where the effect of volume of NaOH was evaluated

Full size image

3.1.2 Crystallization Studies

Effect of Time, Temperature and Variation in Water Content on Crystallization

An evaluation on the effect of crystallization on time was performed at 6, 24, 48 and 72 h. Bands associated with Al–O and Si–O symmetric stretching at 930 cm−1 appeared to be most intense for 72 h. This indicated that more of the glassy phase present in CFA reacted with the alkaline activator (NaOH) to form zeolite as time was increased as shown in Fig. 5a. Therefore, the optimum time was 72 h.

Fig. 5
figure 5

a FTIR spectra of zeolites residues synthesized at different time intervals. b FTIR spectra for zeolite residues synthesized at different temperatures. c FTIR spectra for zeolites synthesized with different solid/liquid ratios

Full size image

A subsequent study was done investigating the effect of temperature performed at 35 \(^\circ{\rm C}\), 120 \(^\circ{\rm C}\) and 200 \(^\circ{\rm C}\) for 72 h. At 35 °C, the Al–O and Si–O symmetric stretching vibrations at 930 cm−1 appeared to be almost non-existent compared to the residue crystallized at 120 and 200 °C (Fig. 5b). This indicated that the dissolved glassy phase (from CFA) present in the NaOH alkaline solution failed to effectively form the aluminosilicate gel and eventually crystallize into zeolite at lower temperatures (Wałek et al., 2008; Musyoka et al., 2012; Mokgehle et al., 2019a; Kobayashi et al., 2020). Hence, the optimal temperature for zeolite crystallization was 200 °C.

Thereafter, the analysis on the water content of zeolite crystallization was examined at H2O/SiO2 = 0, H2O/SiO2 = 1, H2O/SiO2 = 2 corresponding to additions of 0, 7.5 and 10 mL of ultra-pure water to 15 mL slurry, respectively. The most intense peak for Al–O and Si–O asymmetric vibrations was observed at a H2O/SiO2 = 1. This suggested that water addition improved the conversion of the CFA glassy phase matrix to zeolite, by forming a supersaturated medium which subsequently allowed for more of the NaOH dissolved Si4+ and Al3+ species to crystallize on the ash surface (Tajunnisa et al., 2017; Mokgehle et al., 2019a, b; Lu et al., 2021). However, when larger amounts of water are added, the formation of the zeolite crystals is interrupted (Mokgehle et al., 2019a). Therefore, the optimal result was H2O/SiO2 = 1.

3.2 Adsorption Studies

3.2.1 Effect of Mass of Zeolite

Adsorption studies were performed to evaluate the effect of mass of zeolite. Batch adsorption studies were performed on different amounts of zeolites (10–80 mg) with 5 mL of 0.5 mg L−1 of chromium solution for 180 min each. In Fig. 6, it is shown that as the mass of zeolite was increased, the percentage removal of chromium increased. This suggested that larger doses of zeolite resulted in a larger surface area for adsorption of larger amounts. A similar effect was reported by Saranya et al. (2017), Mokgehle et al. (2019b), Bai et al. (2020), and Qiu et al. (2020). The adsorption capacities for the 10, 20, 40, 60 and 80 mg zeolites were 0.245, 0.100, 0.058, 0.035 and 0.025 mg g−1, respectively. Hence for further studies, 80 mg of zeolite was used as the optimum dose for removing Cr.

Fig. 6
figure 6

Percentage removal of chromium while varying mass of zeolite used in batch adsorption studies (experimental conditions; concentration of chromium — 0.5 g mL.−1 Cr, contact time — 3 h, temperature — 25 °C)

Full size image

3.2.2 Effect of Contact Time

Batch adsorption studies were performed to evaluate the effect of contact time on zeolite. Adsorption studies were performed using 80 mg of zeolite, with 5 mL of 0.5 mg L−1 of chromium solution at different contact times (30–420 min). From Fig. 7, a general trend of an increased percentage removal of Cr was observed as contact time was increased. This revealed that at lower contact times many adsorption sites on the zeolite structure were left unoccupied; however, with an increase in the adsorption period, more of these available sites were filled by Cr. This trend was also reported by Saranya et al. (2017) and Bai et al. (2020). Therefore, 420 min was applied as the optimum time for future studies. The low removal percentages at 300 min could probably be due to agglomeration of the zeolite in solution, preventing its dispersion and subsequent availability of sites for adsorption (Dhiman & Sharma, 2019; Javadian et al., 2020).

Fig. 7
figure 7

Percentage removal of chromium at various contact times (experimental conditions; mass of zeolite — 80 mg, concentration of chromium — 0.5 mg L.−1, temperature — 25 °C)

Full size image

3.2.3 Effect of Concentration

Adsorption studies were performed to evaluate the effect of the concentration of chromium on adsorption. Batch adsorption studies were performed at different concentrations of chromium from 0.1 to 0.9 mg L−1. From Fig. 8, a proportional increase in percentage removal of chromium was observed with increasing concentration. This suggested that mass transfer could have been influential during adsorption under these conditions. A large initial chromium concentration provided a large driving force for chromium sorption onto the zeolite surface via cation exchange, as seen for the 0.9-mg L−1 study (Fig. 8). Furthermore, at initial concentrations of 0.7 and 0.9 mg L−1, the removal of Cr showed a tendency of flattening out due possibly to exhaustion of available adsorption sites. This was in agreement with what was observed by Kulkarni et al. (2018) and Hayati et al. (2018). Therefore, 0.9 g mL−1 was the optimum concentration.

Fig. 8
figure 8

Percentage removal of chromium at different chromium concentrations (experimental conditions; mass of zeolite — 80 mg, contact time — 7 h, temperature — 25 °C)

Full size image

3.2.4 Effect of Temperature

Adsorption studies were performed to evaluate the effect of temperature on the adsorption of Cr(VI). The temperatures studied ranged from 25 to 75 °C with 80 mg of zeolite and 5 mL of 0.9 mg L−1 chromium solution for 420 min each. As temperature was increased, the percentage removal of chromium was observed to also increase (Fig. 9). This indicated that there was an increase in the uptake capacity of chromium by the zeolite as the temperature rose from 25 °C. This suggested that the adsorption process was endothermic in nature. This increased capacity was probably due to the increased number of binding sites that may have been generated by firstly the possible breakage of some internal bonds near the surface of binding sites and secondly due to the increased collision frequency by the adsorbate onto the adsorption sites (Saini & Melo, 2013; Vijayalakshmi et al., 2017). Therefore, 75 °C was identified as the optimum temperature in this study.

Fig. 9
figure 9

Percentage removal of chromium after adsorption under a variety of temperatures (experimental conditions; mass of zeolite — 80 mg, contact time — 7 h, initial chromium concentration — 0.9 mg L.−1)

Full size image

3.3 Application Studies

Application studies were conducted on water collected from the Thulamela Municipality wastewater treatment site, as shown in Table 2. Analysis of chromium in the wastewater samples revealed concentrations that were lower than the limit of detection. This suggested that Cr contamination was unlikely as the Cr effluent concentration was well below the regulatory threshold set by the WHO 2013. Besides Cr, other heavy metals were detected which included Fe, Hg, Mn and Se. The removal percentage in increasing order was as follows: Hg < Se < Fe.

Table 2 Some of the heavy metals detected using ICP-OES in the influent and effluent at the Thulamela water treatment site
Full size table

4 Conclusion

CFA-derived zeolites were shown to be efficient for the removal of heavy metals from wastewater. CFA-derived zeolites were synthesized based on the hydrothermal treatment method. The optimum conditions for ageing was a concentration of 1.5 M, at a time of 48 h and NaOH volume of 100 mL. Thereafter, crystallization studies yielded optimum conditions at 72 h, temperature of 200 °C and H2O/SiO2 = 1. Characterization of the synthesized zeolites indicated symmetric stretching of Al–O and Si–O due to the incorporation of Na+ into the CFA aluminosilicate layer, indicating zeolite formation. XRF data showed that the SiO2 and Al2O3 content was highest in the zeolite synthesized for 48 h. The optimum conditions from the batch adsorption studies were mass = 80 mg, contact time = 420 min, influent Cr concentration = 0.9 M and temperature of 75 °C, where removal percentages improvements of 77% were obtained from the first to the last optimization study. The optimised batch adsorption parameters were then applied for the removal of chromium and other selected heavy metals in tap water from the Thulamela municipality. Though the removal of chromium could not be determined during application studies, based on ICP-OES studies, CFA-derived zeolites were efficient in the removal of Fe, Mn, Hg and Se with percentage removals of 96.46, 72.97, 25.00 and 72.00%, respectively. Furthermore, ICP-OES analysis indicated that the concentration of chromium was below the detection limit, signifying that the tap water from the studied municipality was adequately treated.