Investigating the efficiency of oak powder as a new natural coagulant for eliminating polystyrene microplastics from aqueous solutions

Polystyrene (PS) is a commonly used plastic material in disposable containers. However, it readily breaks down into microplastic particles when exposed to water environments. In this research, oak powder was used as a natural, inexpensive, and eco-friendly coagulant. The present study aims to determine the effectiveness of oak powder in removing PS from aquatic environments. The Box-Behnken model (BBD) was used to determine the optimal conditions for removal. The removal efficiency was evaluated for various parameters including PS concentration (100–900 mg/L), pH (4–10), contact time (10–40 min), and oak dosage (100–400 mg/L). The maximum removal of PS microplastics (89.1%) was achieved by using an oak dose of 250 mg/L, a PS concentration of 900 mg/L, a contact time of 40 min, and a pH of 7. These results suggest that oak powder can effectively remove PS microplastics through surface adsorption and charge neutralization mechanisms, likely due to the presence of tannin compounds. Based on the results obtained, it has been found that the natural coagulant derived from oak has the potential to effectively compete with harmful chemical coagulants in removing microplastics from aqueous solutions.


Preparation of oak coagulant
The discarded oak powder were collected from western Iran.The discarded powder were crushed, washed with tap water, and then dried in an oven at 60 °C for 24 h.The resulting dry material was ground into a powder with a fineness of less than 100 mesh.

Preparation of PS microplastic
The PS granules underwent a washing process with 0.1 M HCl, followed by drying at 60 °C for 24 h.They were then crushed using a grinder and meshed into sizes of 300 µm, 425 µm, and 850 µm.PS particles were stored in a dark and sealed container.Exposure to light may cause decomposition and alterations in the physical properties and molecular weight of the substance.This causes errors in the test results.

Characteristics and measurements
The characteristics of oak and PS particles were investigated by FTIR, FESEM, and EDX tests.FTIR spectroscopy was utilized to measure the surface functional groups of oak and PS, as well as the functional groups present after the experiment and removal process.The FESEM test was used to determine the surface morphology of oak and PS.Additionally, EDX test was utilized to determine the elemental composition the particles.

Fourier Transform Infrared Spectrometer (FTIR)
FTIR analysis was utilized to compare and identify the chemical compounds, bonds, and functional groups present before and after the synthesis of the adsorbent 29 .This test was conducted using a PerkinElmer spectrometer (FT-IR/NIR Frontier).

Design of experiments
The efficiency of removal was evaluated based on the initial PS concentrations, pH, contact time, and oak dosage (refer to Table 1).All experiments were conducted in a 250 mL flask containing 200 mL of the reaction mixture.
The size of the PS particles was less than 100 µm.All tests were conducted at room temperature (25 °C).A jar test apparatus was utilized to conduct the tests.A 200 cc of the reaction solution was agitated using a Jar test apparatus at 400 rpm for 1 min to ensure direct contact between PS particles and oak particles.The speed was then reduced to 100 rpm for 15 min to facilitate the formation of larger flocs.After the reaction time ended, the suspension of PS and oak was poured into the Imhoff funnel.The flocs that had formed were allowed to settle under static conditions for 30 min.Afterward, the supernatant was collected and filtered through a 45 µm Whatman filter paper.The filtered solution was then dried in an oven at 60 °C for 24 h.The PS removal efficiency was determined using the weighing method and the following formula: In this formula, M1 represents the initial weight of PS before the removal process, while M2 is the weight of PS on the Whatman filter after the removal process.

Modeling the removal of PS
The present study was designed using BBD model to optimize the process of removing PS microplastics.BBD is a popular RSM design that offers several advantages.A significant advantage of this design is its ability to achieve efficient testing by requiring fewer test runs compared to other designs, such as the full factorial design.This decrease in the number of runs makes BBD a practical choice when resources are limited.BBD excels in studying quadratic response surfaces and enables researchers to efficiently analyze and estimate main effects, interactions, and quadratic effects 31 .By capturing these fundamental factors, the BBD model facilitates a comprehensive understanding of complex systems and their non-linear relationships.The quadratic model used by RSM is expressed as the following equation: where Y, β 0 , β i , βii, β ij , and X i or X j represent the predicted response, constant coefficient, regression coefficients for linear impacts, quadratic coefficients, interaction coefficients, and coded values of the parameters, respectively 31 .

Characterization
Figure 1a depicts the FT-IR spectra of PS particles before and after the removal process.The bands observed at 3391.52 and 2926.62 cm −1 are likely related to the H-OH vibration of water molecules, stretching of C-H bonds, and bending vibrations, respectively.The peak at 1657.29 cm −1 is attributed to the C=O stretching band.During the industrial manufacturing process of PS, these particles combine with oxygen and moisture at high temperatures to form carbon-oxygen (C=O) bonds.The C=O functional group connects PS particles to oak powder via hydrogen bonds.It is therefore conceivable to remove PS particles using this method 32 .The peak at approximately 702.38 cm −1 indicates the presence of benzene rings in PS 27 .Figure 1b displays the FT-IR spectra following the coagulation process.Compared to the PS spectrum before removal, the peak corresponding to H-OH shifted to 3408.29 cm −1 , indicating the hydrolysis of the coagulant 33 .Similarly, the intensity of the peak associated with the C=O functional group has changed to 1648.98 cm −1 .Furthermore, the peak at 2924.78 cm −1 , which corresponds to C-H groups, did not exhibit significant changes compared to its state prior to removal.Generally, it can be said that the displacement of the peaks is related to the formation of new bonds between the functional groups present in oak powder, acting as coagulant particles, and PS.FESEM images of PS and the flocs of PS and oak powder are presented in Fig. 2. From Fig. 2a, cracks can be seen on the surface of these particles, which increases their contact surface with the oak particles.According to the findings presented in Fig. 2b, PS particles accumulate on the surface of oak particles.Given that the surface charges of oak and microplastic particles are often opposite, it can be inferred that a bridge forms between them, resulting in a charge neutralization process that leads to the formation of large flocs 34 .As can be seen from the FASEM image, the connection between oak and microplastic particles is well-established.After the removal process, the connection between the oak particles and the rough surfaces of the PS particles has resulted in the formation of larger flocs, thereby increasing the sedimentation rate 32,35 .This finding demonstrates that oak powder are effective as coagulants.
Figure 3 shows the EDX spectrum before and after PS removal.Each peak in the EDX spectrum corresponds to a specific energy level that represents a particular element.The concentration of each element affects the intensity of the peak 36 .Based on the EDX image (Fig. 3a), the levels of carbon, nitrogen, oxygen, sulfur, phosphorus, and calcium elements in PS are displayed.Figure 3b illustrates the changes in element values after binding oak particles to PS.The carbon content depicted in Fig. 3a indicates that PS particles exhibit a strong affinity towards   www.nature.com/scientificreports/binding to the surface of oak particles 37 .According to a study, microplastics have been found to adhere to grain surfaces due to their high carbon content 38 .
According to Fig. 3b, oak and microplastic flocs contain 41.38% oxygen and 48.95% carbon.Other elements, including calcium, phosphorus, sulfur, and nitrogen, make up a small percentage of this composition.The high oxygen and carbon content in the flocs derived from oak and microplastics indicates that oak powder, which act as a natural coagulant, are highly effective in eliminating pollutants from aqueous solutions.High carbon content can cause oak particles to adhere to microplastics and form flocs, thereby increasing the efficiency of PS removal 39 .

Response model
In this study, we investigated the effect of oak powder on the removal of PS from aqueous solutions.Table 2 displays the removal of PS by oak powder.
According to Table 2, the PS removal efficiency ranged from 3.26 to 89.1%.To select the best model, we statistically evaluated the experimental findings for linear, 2FI, quadratic, and cubic models.Table 3 represents the evaluation of the statistical adequacy of the models.
Table 4 displays the coefficients of the quadratic model for PS removal using oak particles.The quadratic model (QM) was fitted to the experimental data as shown in Table 4.
Based on Table 4, the QM calculation of PS removal, according to the coded parameters, is presented in Eq. ( 3):  www.nature.com/scientificreports/According to Eq. ( 3), each model comprises of two fixed and variable components.The removal rate was predicted to be 46.24%based on the various test variables.The coded parameters A, B, C, and D had coefficients of + 22.91, + 4.33, − 0.285, and + 4.95, respectively.The concentration of PS (code A) had the greatest effect on removal, with a coefficient of + 22.24.The interaction effect with the highest coefficient was related to AD, with a value of + 8.35.The highest square effect was correlated with D2, with a coefficient of + 17.75.Table 5 illustrates the analysis of variance (ANOVA) for the QM of PS removal by oak powder.
The values of R 2 , adjusted R 2 , predicted R 2 , and adequacy precision were obtained as 0.95, 0.90, 0.78, and 19.25, respectively.According to Table 5, the P-values for the variables of PS concentration, oak dose, and contact time are all less than 0.05, indicating that these variables are statistically significant and have a significant effect on PS removal.As can be seen from Table 5, the p-value for C and BC, BD, and CD was greater than 0.05.This indicates that the results are not statistically significant, and these factors do not have a significant impact on the removal of PS.Whereas, the p-values for A 2 , B 2 , C 2 , D 2 , and AD were less than 0.01, indicating their significant effect on the removal of PS.In this model, the difference between R 2 and predicted R 2 was less than 0.2, which is considered to be good.In general, the concentration of PS, oak dose and contact time are the most important factors that affect the removal of PS by oak powder.The R-squared value is 0.95, indicating that the model explains 95% of the variability in PS removal 40 .
Figure 4 displays a comparison between the actual removal and the predicted removal.According to Fig. 4, it is evident that the model is adequate in providing a good prediction for PS removal.

Effect of main factors on removal efficiency
The impact of significant variables, including PS concentration, oak dosage, pH, and contact time, on the removal efficiency was drawn by Design Expert 13 software, and the findings are illustrated in Fig. 5.
It is important to note that in order to explain the impact of one parameter on a response, other factors must be held constant at a zero level.For example, when the contact time variable increases from level 1 to + 1, the other variables such as PS concentration, oak dose, and pH remain constant at the zero level.
To calculate the removal efficiency of PS, oak powder was used with doses ranging from 100 to 400 mg/L.According to Fig. 5a, it can be observed that the removal efficiency increases as the oak dose increases.At low doses, oak particles form small and unstable flocs with PS, which reduces the efficiency of removal 41 .By increasing the dose of oak, the flocs come together with sufficient strength and create a larger surface area per unit volume of microplastic particles 42 .This causes PS to be adsorbed through intermolecular or van der Waals attraction 43 .With the bridging formation between the oak and PS particles, the flocs become heavier and increase sedimentation 44 .It should be noted that high doses of oak can cause the flocs to become heavy and break apart, leading to the redispersion of particles.As a result, the removal efficiency may decrease 41 .In the study conducted was increased from 50 to 100 mg/L.Finally, by increasing the dose to 200 mg/L, the adsorption efficiency can be improved to 92.1%.However, with a further increase in Mg(OH) 2 dosage up to 250 mg/L, the removal efficiency decreased by 10% 45 .In another study conducted by Yingying Duan et al., the PET removal efficiency increased from 50.67% to 92.05% as the coagulant dose was increased from 0 to 100 mg/L.When the coagulant dose was increased to 140 mg/L, the removal efficiency reached 84% 46 .One important parameter for increasing removal efficiency during the coagulation process is the initial concentration of PS.According to Fig. 5a, the removal efficiency significantly increases as the initial concentration of PS.According to the results obtained, as the concentration of PS, the likelihood of collision with the coagulant also increases.Conversely, when the concentration of microplastic particles is low, the probability of collision  www.nature.com/scientificreports/decreases.The positive ions on the surface of the oak particles undergo hydrolysis, leading to compression of the bilayer structure of PS and a reduction in the repulsive force between particles of the same charge.As a result, the likelihood of collision between microplastics and coagulants increases.This process causes a significant portion of microplastics to lose stability and settle 47 .In another study, it was found that increasing the coagulant dose from 0 to 100 mg/L resulted in an increase in the removal efficiency of PET from 50.67 to 92.05%.However, at 140 mg/L, the removal efficiency initially increased and then decreased 32 .From Fig. 5b, it is evident that the contact time has a direct impact on the removal efficiency.By increasing the time from 25 to 40 min, the removal efficiency improves by almost 14% (p-value < 0.05).By increasing the reaction time, microplastic particles have a greater opportunity to form flocs when in contact with oak particles, thereby enhancing the efficiency of PS removal.This factor can be seen as a sign of potential interaction between coagulants and PS microplastics.If coagulation is successful, the turbidity is effectively reduced through the settling of flocs 16 .
According to Fig. 5b, the removal efficiency decreased as the pH increased from 4 to 7, but then increased as the pH increased to 10 (p-value > 0.05).These observations can be related to the mechanism of surface charge neutralization.In fact, oak particles function as a natural polyelectrolyte.These particles have an opposite charge to that of microplastics and are typically attached to the surface of plastic particles.This attachment eventually neutralizes the surface charge 27,28 .

Conclusion
In this research, oak powder were used as a natural, inexpensive, and eco-friendly coagulant.The Box-Behnken model was used to determine the optimal conditions for removal.FESEM, FTIR, and EDX analysis were used to characterize the properties of oak and PS particles.The FESEM image illustrates the impact of surface morphology and verifies the presence of PS particles accumulated on the surface of oak seed particles.The FTIR spectrum showed the bonds formed between PS and oak.EDX analysis determined the elements involved in the process of coagulation.The maximum PS microplastics (89.1%) was achieved by using oak powder under specific conditions.These conditions included an oak dose of 250 mg/L, a PS concentration of 900 mg/L, a contact time of 40 min, and a pH of 7. The results suggest that oak powder can effectively remove PS microplastics through surface adsorption and charge neutralization mechanisms, which are attributed to the presence of tannin compounds.Based on the results obtained, it has been found that the natural coagulant derived from oak powder possesses the ability to effectively compete with harmful chemical coagulants in removing microplastics from aqueous solutions.

Figure 1 .
Figure 1.FTIR Spectra of (a) before and (b) after PS removal.

Figure 2 .
Figure 2. FESEM of (a) PS and (b) flocs of PS and oak powder.

Figure 4 .
Figure 4. Distribution of experimental vs. predicted removal for PS by oak powder.

Figure 5 .
Figure 5. Response surface plots illustrating the effects of dose vs. concentration (a) and pH vs. time (b), generated using Design Expert 13 software.

Table 1 .
Parameters required for the removal of PS microplastics.

Table 2 .
BBD matrix for the removal of PS by oak powder.

Table 3 .
Statistical adequacy evaluation of models.

Table 4 .
Estimation coefficients for the quadratic model of PS removal using oak powder.

Table 5 .
ANOVA results for the quadratic model of PS removal by oak powder.