Phosphorus removal from ore waste in aqueous solution with different mass of ore waste adsorbent from the Johor mine site

The most significant environmental problem affecting lakes, reservoirs, rivers, and many other aquatic ecosystems is eutrophication, which is a reason for deteriorating water quality and severely limiting water usage (Preisner & Smol 2022). Eutrophication is caused by the excessive addition of nutrients, most notably phosphorus, to water bodies. When eutrophic circumstances occur, poisonous-reduced compounds increase, causing foul odours and tastes and hypolimnetic oxygen depletion (Nguyen et al. 2022). Eutrophication decreases the aquatic life population. An increase in nutrients (phosphorus) promotes the growth of aquatic plants and the formation of organic matter in the body of water, thus resulting in other aquatic life such as fish dying because of the high oxygen demand in the water body (Erhunmwunse et al. 2013). Phosphorus in domestic wastewater has been a big problem in big cities and also small villages (Chrispim et al. 2019). The main source of this pollution could be traced back to the household from soap, shampoo, and many other cleaning supplies that are used without a second thought and discharged away. Moreover, the excessive use of fertilizer in agriculture to provide nutrients for plants will cause the phosphorus pollutant to infiltrate into ground and surface water. It causes the beginning of the eutrophication process in the river due to uncontrolled human activities.

Phosphorus is one of the elements that can be removed through chemical or biological treatment at tertiary wastewater treatment. Due to the high maintenance and operational cost of phosphorus removal, this study approached the possibility of waste material, which is ore waste, to remove phosphorus in a solution (Martí et al. 2021). Ore waste is one of the products used as a train rock ballast in cement aggregate and asphalt aggregate. Other than that, ore waste does not have significant use over time, leading to an abundance of unused ore waste (Pashkevich & Petrova 2019). Despite its uselessness, ore waste contains elements that can contribute to removing phosphorus from the wastewater. One of the components, calcium dialuminium diaquaoctahydroxide 1.84-hydrate, has the ability to remove phosphorus from this wastewater and speed up the adsorption process (Nobaharan et al. 2021). Ore is also very cheap in the market, making it one of the most cost-efficient materials for removing phosphorus if the study is to be a definite success.

The possibility of ore waste in removing phosphorus from water is one of the novel causes that can be used in the adsorption treatment process. Although the adsorption of phosphorus is practically used various materials in previous studies, this study focuses on analysis to verify the theoretical and experimental data. The mathematical models apply the adsorption of solute through surface porous adsorbent, mass of adsorption, and initial concentration to develop the prediction on removing pollutants in water (Chung et al. 2015). The use of kinetic and isotherm models as well as a prediction on removal efficiency through different initial concentrations and required mass of adsorbent still need to be investigated due to ore waste, one of the novel adsorbents in the water treatment process. The objectives of this study are (1) to investigate the removal performance of ore waste in removing phosphorus from synthetic wastewater, (2) to verify the theoretical and batch experimental data using kinetic (pseudo-first-order (PFO) and pseudo-second-order (PSO)) and isotherm (Langmuir and kinetic) modelling studies, and (3) to create the prediction contour of removal efficiency and the mass of adsorbent needed under the different initial concentrations of phosphorus solution.

Adsorbent was prepared by first collecting 1 kg of ore waste at the Johor mine site. The ore waste was then sun-dried for 2 days. In order to obtain complete drying, the dried ore waste was dried again in a drying oven at 30 °C for 2 days. The dry absorbent was then sieved to achieve a size range of 1.18–2.46 mm and weighed using an analytical balance for 2, 4, 6, 8, and 10 g for batch experiments.

In a volumetric flask, 0.1433 g of potassium dihydrogen phosphate was dissolved in 1 L of deionised water to make 100 mg/L solution. The solution was then diluted to 5, 10, 15, 20, and 25 mg/L concentrations.

The amino acid method 8178 programmed was used to assess the initial and final phosphorus concentrations in aqueous solutions using the HACH DR 6000 UV-VIS Spectrophotometer. The researchers employed the COXEM EM – 30 AXE PLUS SEM to assess SEM testing for surface morphology of the membrane and EDXRF for identify the chemical elements group on the ore waste. The Perkin Elmer Spectrum Two FTIR Spectrometer was used to characterise the functional groups of the ore and display the infrared adsorption spectrum before and after the adsorption process. The researchers employed a second-generation BRUKER D2 Phaser Benchtop XRD to analyse the crystallisation or crystal-phase composition of the shell in XRD analysis.

Two batch experiments were carried out for the adsorption study using an aqueous solution. The first batch experiment investigated the initial concentration and adsorbent mass effects by introducing 2, 4, 6, 8, and 10 g of adsorbent into a conical flask of 100 mL with varying concentrations. Each conical flask was mixed with a synthetic solution with 5, 10, 15, 20, and 25 phosphorus concentrations for 5,760 min and shaken at 170 rpm.

A series of second-batch tests explore the influence of absorbent mass and time. The influence of adsorbent mass was investigated in the second group using 2, 4, 6, 8, and 10 g of absorbent in a 100 mL conical flask with varying contact lengths of 30, 120, 300, 1,440, and 5,760 min and shaken in 170 rpm. After filtering each sample with a filtration pump to separate the solution from the suspended particles, the phosphate content was analysed using the DR6000 UV-Spectrophotometer, and the data were verified through kinetic and isotherm models.

The adsorption isotherm is critical in identifying the adsorbent and the adsorbent response. This process will provide adequate evidence that an efficient adsorbent may be absorbed. The form of the isotherm expresses both the durability of the adsorbent–adsorbate interactions and the adsorption affinities of the molecules.

EDXRF was used to evaluate the elemental compositions of mining waste. Table 1 lists the elements that make up ore waste. Before the reaction between the ore waste and phosphorus, the major elements in the ore waste were Si and O, with measurements of 26.22 and 52.38%, respectively. In contrast, the major elements in the ore waste of the reaction with phosphorus afterwards were Si and Ca, with measurements of 48.40 and 24.79%, respectively. It is also shown that O was not in the ore waste anymore after the reaction with the phosphorus. Elements with calcium often have the ability to remove phosphorus from the aqueous solution (Nguyen et al. 2022). This means it is confirmed that ore waste can potentially remove phosphorus from wastewater. Table 1 also shows that after the reaction between the ore and phosphorus, there is an increase in the percentage of the elemental composition in the ore.

The cordierite line with a triangle symbol is the substance with the second-highest dominating peak visible in the XRD pattern in Figure 2(a). A magnesium–iron aluminium cyclosilicate, cordierite is sometimes called dichroite and iolite (Guo et al. 2022). Because phosphate is a cation, the presence of positive ions such as and might enhance the adsorption of phosphate. Since cordierite is currently employed extensively in the electronic and information fields thanks to its superior electrical, thermal, and mechanical qualities, ore waste can now be used more broadly for purposes other than only water treatment (Othman et al. 2021).

The prominent peak in the XRD pattern in Figure 2(b) is the line with the circle symbol. The triangle and square symbols represent the second and third dominating peaks. The triangle and square symbols represent anorthite and calcite, respectively, whereas the circle represents zeolite Y (purely siliceous). Crystalline-hydrated aluminium silicates make up zeolite. The hydrated zeolites display a consistent three-dimensional crystalline aluminosilicate structure based on tetrahedrons, where silicon or aluminium cations are surrounded by four oxygen atoms at the vertices, signifying and (Farro et al. 2023). Any zeolite is highly microporous because these structures create a network of pores and channels. Depending on the width of the pores, any zeolite can adsorb water and cations (Pashkevich & Petrova 2019). As a result, zeolite-containing ore waste can be used to remove phosphate from water.

Calcium silicate has recently emerged as an adsorptive material with tremendous potential for phosphate removal from other adsorbents. Calcium silicate can be used for phosphate recovery in biological wastewater treatment because it acts as a calcium ion donor under low supersaturation and alkaline conditions (pH = 8–9). Anorthite that contains calcium silicate can promote the nucleation process as a crystal seed, which is easier to precipitate and recover than the conventional calcium phosphate compound (Guo et al. 2022).

Calcite has been shown in studies to eliminate phosphate successfully. Extensive studies on phosphate adsorption by calcite have been done at low phosphate concentrations (Li et al. 2021). Langmuir adsorption isotherms can be used to describe phosphate adsorption by calcite, which is most likely due to monolayer adsorption. According to research, over 80% of phosphate anions were adsorbed within 10 s, probably because phosphate ions replaced adsorbed water molecules, bicarbonate ions, or hydroxyl ions from calcite surfaces.

The frequencies of the ore waste sample before the adsorption process were 658.447, 756.514, and 940.118 cm⁻¹. The frequencies significantly change after the reaction with aqueous phosphorus solution to 687.640, 755.900, and 946.297 cm⁻¹. The difference between the frequencies of 29.193 cm⁻¹ (658.447–687.640 cm⁻¹) was attributed to C–Cl stretching (Cao et al. 2022). The adsorption of phosphorus changes the frequency of C–H bending to 0.61 4 cm⁻¹ (756.514–755.900 cm⁻¹). The reaction between the ore waste and the phosphorus solution caused the frequency of C–C bending to change from 940.118 to 946.297 cm⁻¹, with the differences of −6.179 cm⁻¹ in frequency (Possenti et al. 2021; Sambuu 2021). As we can see from Table 2, there are other frequencies that do not have any changes or appeared after the reaction such as the frequency of 1,216.910 cm⁻¹ associated with C–O stretching (Abdullah et al. 2023). Another one is at frequency 1,372.840 cm⁻¹, which is in the functional group of S = O stretching. Next on the table is frequency 1,434.395 cm⁻¹, which is detected in group of O–H bending (Li et al. 2022). After that, the frequency of 1,738.286 cm⁻¹ is associated with the functional group of C = C = C stretching (Giachet et al. 2021; Hien et al. 2022). After the reaction of ore waste and phosphorus, there is appearance of the frequency of 1,954.640 cm⁻¹, which is detected in the functional group of C = O stretching, and the last one is the frequency of 3,019.126 cm⁻¹, which is in the functional group of C–H stretching (Metlenkin et al. 2022).

The rate of phosphorus adsorption on ore waste was tracked over time to estimate the equilibrium-corresponding contact period. This was carried out so that the link between monitoring adsorption uptake over time at a particular pressure or concentration can be observed in adsorption kinetics, which is used to quantify the quantity of diffusion adsorbate that contacts pores (Debnath & Das 2023). Due to the active surface sites present at the start of the adsorption and those that remain after a certain period, the first few minutes may be defined as the speed of adsorption kinetic presence in a substantial quantity, as adsorption is the transition from liquid to solid phase (Martí et al. 2021).

Figure 4 shows the adsorption capacity performance for five different particle masses of mining waste. The adsorption capacity for each particle mass achieved equilibrium at 1,440 min, and the adsorption capacity remained constant for the following 4,320 min. For each particle mass, the adsorption capacity, ⁠, was 0.119 mg/g (2 g), 0.0860 mg/g (4 g), 0.0582 mg/g (6 g), 0.0471 mg/g (8 g), and 0.0434 mg/g (10 g).

Table 3 demonstrates that the particle mass 10 g has an excellent adsorption capacity and removal efficiency, with the highest (0.5242) and a low (1.2215). In contrast, the particle mass 8 g has poor adsorption capacity but a good removal efficiency, with the lowest value (0.3885) and the lowest value (0.9870) for PFO analysis.

Table 3

Kinetic parameter of PFO and PSO model parameters

Kinetic parameter of the PFO model .
Mass of adsorbent (g) .	(mg/g) .	(min−1) .	.	.	(mg/g) .
2	0.2868	0.0066	0.4402	1.1247	0.1190
4	0.2763	0.0071	0.4776	1.3330	0.0860
6	0.2410	0.0086	0.5083	1.3060	0.0582
8	0.1889	0.0079	0.3885	0.9870	0.0471
10	0.2126	0.0097	0.5242	1.2215	0.0434
Kinetic parameter of the PSO model .
Mass of adsorbent (g) .	(mg/g) .	(min-1) .	.	.	(mg/g) .
2	0.1219	0.0657	0.9976	0.0433	0.1190
4	0.0901	0.0458	0.9944	0.0190	0.0860
6	0.0674	0.0195	0.8616	0.0228	0.0582
8	0.0490	0.0979	0.9944	0.0119	0.0471
10	0.0469	0.0530	0.9768	0.0097	0.0434

Kinetic parameter of the PFO model .
Mass of adsorbent (g) .	(mg/g) .	(min−1) .	.	.	(mg/g) .
2	0.2868	0.0066	0.4402	1.1247	0.1190
4	0.2763	0.0071	0.4776	1.3330	0.0860
6	0.2410	0.0086	0.5083	1.3060	0.0582
8	0.1889	0.0079	0.3885	0.9870	0.0471
10	0.2126	0.0097	0.5242	1.2215	0.0434
Kinetic parameter of the PSO model .
Mass of adsorbent (g) .	(mg/g) .	(min-1) .	.	.	(mg/g) .
2	0.1219	0.0657	0.9976	0.0433	0.1190
4	0.0901	0.0458	0.9944	0.0190	0.0860
6	0.0674	0.0195	0.8616	0.0228	0.0582
8	0.0490	0.0979	0.9944	0.0119	0.0471
10	0.0469	0.0530	0.9768	0.0097	0.0434

View Large

The PFO kinetic model, which assumes that chemical sorption or chemisorption is the rate-limiting phase, predicts the behaviour across the whole adsorption range. In this case, adsorption capacity, not adsorbate concentration, dictates the adsorption rate. Figure 6(b) depicts the linear regression analysis for the PSO kinetic model. Table 3 indicates that all particle masses have an excellent coefficient correlation that is close to 1 for PSO analysis. However, particle mass 2 g has the greatest value (0.9976) of all particle masses, indicating that it has the highest adsorption rate. It proves that ore waste reacts better to phosphorus through chemical than physical reactions.

It has been concluded that the PSO model is the better kinetic model to reflect adsorption kinetics since it has the highest value (0.9976) compared to the PFO model (0.5242). Furthermore, the value of of the PSO model (0.0097) is smaller than that of the PFO model (0.9870). This demonstrates that the data fit better with the PSO model than the PFO model. In addition, the best fit demonstrates that the adsorption process depends on the adsorbate and adsorbent quantities (Debnath & Das 2023). Once again, the value of shows that the results support the intra-particle diffusion hypothesis, which sheds light on the adsorption mechanism. The adsorption process is better represented by PSO kinetics if the PSO rate constant is greater than the PFO rate constant. This would imply that rather than only the surface reaction, the adsorption process is predominantly governed by both surface reaction and diffusion (Kalam et al. 2021).

The adsorption mechanism is commonly explicated by applying isotherms, which are mathematical functions that establish a relationship between the quantities of the adsorbate present on the adsorbent material. Various isotherm models, including Langmuir and Freundlich, can characterise the partitioning of metal ions between the liquid and solid phases. The Langmuir isotherm model postulates that adsorption occurs in a single layer on a surface with a limited number of adsorption sites with uniform characteristics and that the adsorbate has no lateral movement on the surface. Once a site has reached its maximum sorption capacity, it can no longer undergo additional sorption processes. This means that the surface attains a state of saturation wherein the highest level of adsorption possible for the surface is attained.

The results indicate that ore residue can absorb from the synthetic solution. This ability is because calcium oxide on the surfaces of raw ore waste and phosphate ions has a strong affinity for one another due to their distinct charges. Table 1 displaying EDXRF analysis shows that ore waste can remove phosphorus from wastewater because materials containing Ca can remove phosphorus from the aqueous solution. The calcination process alters the physical and chemical properties of calcium carbonate and ore residues (Preisner & Smol 2022).

Linear plots were obtained in both cases, indicating the suitability of these isotherms for the current adsorption process. The present study showcases the Langmuir and Freundlich plots of figures and exhibits the adsorption of calcium onto ore waste. Table 4 enumerates the distinct Langmuir and Freundlich constants derived from these plots.

Table 4

Freundlich and Langmuir of the isotherm model parameters

Freundlich model .			Langmuir model .
n .	(mg/g) .	.	(mg/g) .	(mg/g) .	.
0.3970	0.0015	0.8418	−0.0478	−0.12689	0.8277

Freundlich model .			Langmuir model .
n .	(mg/g) .	.	(mg/g) .	(mg/g) .	.
0.3970	0.0015	0.8418	−0.0478	−0.12689	0.8277

View Large

Table 4 presents the Langmuir and Freundlich adsorption constants and the correlation coefficients denoted by ⁠. In order to determine the optimal model for the adsorption of calcium, the data were subjected to fitting using the Langmuir and Freundlich isotherm models. The results indicate satisfactory conformity with the Freundlich model (⁠ = 0.8418) in contrast to the utilisation of the Langmuir isotherm model (⁠ = 0.827). The Freundlich equation yielded a n-value of 0.3970, as presented in Table 4. This suggests that the utilisation of ore waste involves a chemical process.

Even though more precise values could be calculated by explicitly applying the developed equations, plots similar to those in Figure 8 would be useful for quickly determining the removal efficiency trend as a function of the initial adsorption conditions. Such plots enable a clearer understanding of the relationship between removal efficiency and initial adsorption conditions.

The present work evaluated the removal performance of phosphate using ore waste and showed that this natural adsorbent could be considered an efficient material for phosphorus removal from wastewater. Ore waste, mainly composed of calcium, is a potential material for wastewater treatment since precipitation with carbonate is commonly employed for phosphate removal from wastewater. Based on the batch test experiments, all parameters significantly impact phosphate removal efficiency, and mass adsorbent has proved to be the key variable in this study. Experiment results showed that when mass adsorbent increased, phosphate removal percentage also increased due to the increase in the surface area of the adsorbent particles. When the adsorbent dosage increased, the removal of metal ions increased, stimulating a greater phosphate removal process from the aqueous solution. The objectives of this study were achieved. Ore waste can remove phosphate in wastewater because it is rich in calcium. These findings demonstrated the successful removal of ore waste of up to 54.3% from the aqueous solution. The results indicate that the PSO kinetic model is well described with the absorption kinetic of phosphate onto ore waste due to the higher value of ⁠, which is 0.9976. It proves that ore waste reacts better to phosphorus through chemical reactions than through physical reactions. On the other hand, the Freundlich model was considered a more versatile system due to a greater degree of heterogeneity in the surface of absorbent material. Therefore, it can be concluded that more waste could be used because of its effectiveness as a natural adsorbent, which is easily accessible and economical. Ore waste can reduce the cost of treating wastewater, especially in removing phosphate, and promote an environmental-friendly solution of obtaining a cleaner water source.

This research was funded by Tier 1 Grant Vot Q407 and GPPS Grant Q317 provided by the Universiti Tun Hussein Onn Malaysia (UTHM). The authors would like to thank the Neo Environment Technology (NET), Centre for Diploma Studies (CeDS), Research Management Centre, UTHM, for their support.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.