Use of Extracted Proteins from Oak Leaves as Bio-Coagulant for Water and Wastewater Treatment: Optimization by a Fractional Factorial Design

Abstract

The present work sheds light on the potentiality of proteins extracted from oak (Quercus robur) leaves to treat both drinking water and industrial oily wastewater. The work was structured in three steps: firstly, oak leaves in powder form were analyzed by FTIR, XRD and SEM, thus showing the presence of proteins acting as bio-coagulants; secondly, an experimental design was conducted. According to the design of experiences based on fractional design (2⁸⁻⁴), the highest protein concentration (4.895 mg/g) was obtained for the following operating parameters: no filtration, pH of 12, temperature of 20 °C, stirring speed of 300 rpm, stirring time of 60 min, maceration time of 4 h, centrifugation speed of 400 rpm, centrifugation time of 10 min. Finally, a jar test apparatus was used to study the effects of proteins from oak leaves on the characteristics of both drinking water and industrial oily wastewater. In drinking water, the turbidity was reduced from 15.7 to 4.82 NTU when 0.098 mg/L of oak leaves protein was added, thus satisfying the requirements of the national drinking water standards; whereas, in industrial oily wastewater turbidity, total suspended solids, chemical oxygen demand and organic matter were reduced by 96.87, 89.86, 96.39 and 46.28%, respectively, when 0.538 mg/L of oak leaves protein was added. This study opens new perspectives related to the research and development of organic coagulants applicable to industrial wastewater treatment.

1. Introduction

Plants are a source of interest for scientists in relation to the effectiveness of some compounds extracted from them as active agents for different purposes, such as bio-coagulants for water treatments. Bio-coagulants have several advantages such as being renewable, easy to obtain, biodegradable, low or non-toxic and of a relatively low cost [1]. The extraction of compounds from plants is commonly based on their chemical or physical properties [2]. Several methods are used for extracting natural substances from plants. These include traditional extraction techniques (e.g., infusion, decoction or maceration) [2,3], modern extraction techniques (e.g., hydrodistillation, supercritical CO₂, centrifugation) [4,5] and others. The proper extraction method is chosen accordingly to the characteristics of the raw material, as well as those of the desired end products (e.g., proteins, essential oils and total sugars) [6], and their uses in the different fields, such as medical (e.g., preparation of drugs) [7], parachemistry (e.g., manufacturing of cosmetics) [8] and environmental (e.g., water treatment) uses [9,10].

In general, the extraction of bio-coagulants from plants is achieved by three steps: (i) mechanical and physical treatments; (ii) extraction of the coagulant agent by using distilled water [11], either a saline solution [12,13,14,15,16] or a suitable solvent (i.e., acid or base) [13,17]; (iii) purification of the coagulant agent. This agent can belong to carbohydrates, proteins, flavonoids, tannins, total phenolics or cellulose, depending on the plant used to extract the active agent [18,19,20,21,22,23,24]. Several studies have focused on searching for active compounds from plants capable to promote the coagulation process for treating drinking water, as well as industrial wastewater. According to Rasool et al., the coagulant agent extracted from Ocimum basilicum leaves is a polysaccharide [25], whereas Ndabigengesere et al., Taiwo et al. and Bouchareb et al. proved that the active component from Moringa oleifera seeds is a protein [26,27,28]. Conversely, Bello et al. successfully tested spruce tannin as a bio-coagulant for water clarification [29].

The present work is aimed at testing the potential effectiveness of a coagulant protein extracted from the Algerian local oak (Quercus robur) leaves in performing the coagulation/flocculation process used to treat drinking water as well as industrial oily wastewater. This work contributes to the valorization of the natural resources of our country through the development of biobased products for water treatment. Conversely, it allows the possibility of introducing new biodegradable products (oak leaves) during the physicochemical treatment in the coagulation/flocculation process. We hope to use this work to replace some mineral coagulants (such as ferric chloride and aluminum sulfate) widely used in the field of wastewater and drinking water treatment because they have adverse effects on the environment, especially human health [30,31]. Another objective of this study is the industrial oily wastewater treatment, which was generally discharged directly into the environment using a natural product (oak leaves) that is biodegradable and non-toxic for the environment, where this industrial water was responsible for nuisances and harmful impacts on the fauna and flora. This study was carried out in three steps: (i) a chemical analysis of the raw material; (ii) extraction of the protein responsible for promoting the coagulation/flocculation process according to a preliminary study conducted with Minitab software; (iii) use of the bio-coagulant extracted from oak leaves for water treatment. In the first step, oak leaves were analyzed by FTIR, XRD and SEM with the purpose to assess the presence of active agents. In the second step, proteins contained in oak leaves were extracted after the determination of the main factors influencing the extraction and subsequent purification of the coagulant agents according to the results of a preliminary experimental design. An ANOVA study was performed to reduce the number of experiments to be conducted. Finally, the quality of water was evaluated pre and post coagulation/flocculation treatment by measuring several water parameters, such as turbidity, pH, total alkalinity (TA), organic matter (OM) content, salinity, electrical conductivity (Ec), total suspended solids (TSS) and chemical oxygen demand (COD) concentration. The experimental part of this study was carried out at the Laboratory of Process Engineering for Sustainable Development and Health Products (GPDDPS), of the National Polytechnic School of Constantine.

2. Materials and Methods

2.1. Raw Material for Bio-Coagulant Agent Extraction

Most plant-based coagulants are not all affordable due to cost and availability in some areas, making water treatment difficult to implement. Year-round availability is also an important factor to consider. This led us to choose a plant very abundant in Algeria, Quercus robur. Based on our previous studies conducted on the use of oak (Quercus robur) leaves for drinking water treatment, this treatment has also been of great effectiveness for the reduction of turbidity [18].

2.2. Extraction and Purification of Proteins from Oak Leaves

The coagulant agent based on proteins was extracted from oak leaves according to the procedure showed in Figure 1.

The powder coagulant was prepared as follows:

i.

Washing with tap water;

ii.

Drying at a relatively low temperature of 50 °C for 24 h to avoid denaturation of the active compounds responsible for coagulation [32];

iii.

Grinding for particle size reduction and sieving up to obtain a homogeneous fine powder;

iv.

Sieving through a 0.35 mm pore sieve.

A mass of 0.1 g of powdered dried oak leaves was suspended in 100 mL of distilled water to extract the active compounds responsible for coagulation (proteins) using an Erlenmeyer flask (250 mL) and mixed by a magnetic stirrer (RO-TILABO^® MH 15).

This extraction was carried out at a variable pH, temperature and stirring time, and speed depending on the test matrix (see

Table 1. Statistical distribution of experimental data (real values of factors).

Parameters	Unit	Minimum	Maximum
pH (X1)	/	2	12
Temperature(X2)	°C	20	60
Stirring speed(X3)	rpm	300	1500
Stirring time(X4)	min	10	60
Maceration time(X5)	hr	1	4
Filtration(X6)	/	Without	With
Centrifugation speed(X7)	rpm	1000	4000
Centrifugation time(X8)	min	10	60

). After a maceration time variable between 1 and 4 h, the coagulating proteins were purified using two techniques, which are filtration and centrifugation, parts of the solutions are centrifuged with a variable speed and time (see Table 1) after their filtration by standard filter papers (porosity < 8 μm), other parts of the solutions are centrifuged directly after the maceration that is without filtration. Once the extracts are recovered, the proteins are determined and stored in a cold room at ±4 °C.

The optimization of the procedure was obtained through a fractional experimental design. According to the international literature [24,33,34], and taking into account that proteins are sensitive to the surrounding environmental conditions and are easily denatured [35,36], the operational parameters and the relating range of values reported in Table 1 were considered.

According to Table 1, the experimental study would have tested the effect of 8 parameters varying from the 2 extremes of a range of value. This experimental design would have required 2⁸ (i.e., 256) tests to be conducted. With the aim to reduce the number of tests, a fractional design obtained by Minitab software, called 2⁸⁻⁴, and composed of only 16 trials, was considered (

Table 2. Fractional experimental design.

Standard Run	pH	Temperature (°C)	Stirring Speed (rpm)	Stirring Time (min)	Maceration Time (h)	Filtration *	Centrifugation Speed (rpm)	Centrifugation Time (min)
1	2	20	300	10	1	0	1000	10
2	12	20	300	10	1	1	4000	60
3	2	60	300	10	4	0	4000	60
4	12	60	300	10	4	1	1000	10
5	2	20	1500	10	4	1	4000	10
6	12	20	1500	10	4	0	1000	60
7	2	60	1500	10	1	1	1000	60
8	12	60	1500	10	1	0	4000	10
9	2	20	300	60	4	1	1000	60
10	12	20	300	60	4	0	4000	10
11	2	60	300	60	1	1	4000	10
12	12	60	300	60	1	0	1000	60
13	2	20	1500	60	1	0	4000	60
14	12	20	1500	60	1	1	1000	10
15	2	60	1500	60	4	0	1000	10
16	12	60	1500	60	4	1	4000	60
Min	2	20 °C	300 RPM	10 min	1 h	0	1000 RPM	10 min
Max	12	60 °C	1500 RPM	60 min	4 h	1	4000 RPM	60 min

Note: * 0 means there is no filtration and 1 means that filtration was performed.

)

2.3. Water Samples Collection

Proteins extracted from oak leaves were used to test their effectiveness in treating drinking water as well as industrial oily wastewater. Drinking water was collected from the water treatment plant of Oued El Athmania, Mila (36°27′1.01″ N; 6°15′51.98″ E). Industrial oily wastewater was collected from the industrial plant of Sonatrach, Ain Amenas (28°02′18″ N; 9°33′54″ E). The main characteristics of water samples are reported in

Table 3. Quality of water samples.

Parameters	Industrial Oily Wastewater	Algerian Standards [37]	Drinking Water	Algerian Standards [38]
Turbidity (NTU)	187	20	15.7	5
pH	9	6.5–8.5	7.75	6.5–9
OM (mgO2/L)	17.5	/	2.1	5
TSS (mg/L)	969	35	/	/
COD (mg/L)	784.45	120	/	/
TA (mg CaCO3/L)	/	/	134	500
Salinity (g/L)	/	/	0.6	/
Ec (µs/cm)	/	/	782	2800

.

2.4. Jar Test Experiments

The experimental coagulation/flocculation tests were performed using a jar test apparatus (LI-JTA-125, LABARD, Labard instruments, Bengala, India) under the operating conditions summarized in Table 1. The test is aimed at evaluating the optimal coagulant protein concentration. A series of 6 beakers filled with 1000 mL of water (i.e., drinking water and industrial oily wastewater, separately) were used to conduct the jar tests [32,39]. This was conducted in three steps: (i) coagulation at 160 rpm for 3 min; (ii) flocculation at 30 rpm for 20 min; (iii) settling for 30 min. After that, the supernatant was analyzed for pH, salinity, Ec, turbidity, TA and OM for drinking water, whereas turbidity, pH, OM, TSS and COD were measured for industrial oily wastewater. The removal efficiency was calculated according to the following Equation (1):

$$\mathrm{Removal} \mathrm{efficiency}\left( \%\right)=\frac{\left(\mathrm{Initial} \mathrm{value}-\mathrm{Residual} \mathrm{value}\right)\times 100}{\mathrm{Initial} \mathrm{value}}$$

(1)

where initial and residual values are, respectively, the value of the specific parameter investigated (e.g., turbidity, COD) with its relative unit (e.g., NTU, mg/L) pre and post jar test performance.

2.5. Analytical Methods

The Bruker’s X-ray diffraction system (K-Alpha (1.54), 40 kV, 30 mA, D8-Discover, Nancy, France), the Fourier-Transform Infrared Spectrometer (SHIMADZU Code: HI 98713, Chimadzu, Clujnapoca, Romania) and scanning electron microscope (Hitachi TM3400, France) were used to determine the presence of protein in the oak leaves.

The protein content was determined by the Bradford method [40]. pH, salinity, and Ec were evaluated by a multi-parameters device (Jenway model 3540, Camlab, Cambridge, United Kingdom). Turbidity and COD concentration were measured by using a turbidity meter (HANNA Code: HI 98713, Hanna instruments, Cluj-Napoca, Romania) and the Digital Reactor Block 200 (Jenway model 3540, Camlab, Cambridge, United Kingdom), respectively. TA, OM and TSS concentrations were evaluated by standard titrimetric methods [41].

3. Results and Discussions

3.1. Characterization of Oak Leaves Powder

3.1.1. FTIR and XRD Analysis

According to Figure 2a, FTIR spectra of the oak leaves powder shows different peaks corresponding to the active functional groups responsible for the coagulation. Among them, the functional group of protein, according to Padhiyar et al., the C=O function of the ester group at 1721.2 cm⁻¹, proves the presence of proteins [42]. Conversely, Magalhaes et al., in their study, reported that proteins were also observed at 1614.1 cm⁻¹ [43].These functional groups are responsible for the destabilization of colloidal particles, thus promoting their coagulation and flocculation, and the subsequent removal of various organic and inorganic pollutants [9,44].

Figure 2b shows the X-ray diffraction pattern of oak leaves powder. The results reveal a crystallinity rate of 29.46% and the presence of proteins between 15 and 40 °C. These active agents are capable to remove various pollutants present in drinking water, municipal and industrial wastewater such as physical pollutants (turbidity and TSS), organic pollutants (COD, BOD₅, OM) and biological pollutants (total coliforms) through a coagulation/flocculation process [28,45,46].

The studies carried out by Benalia et al. and Padhiyar et al. have, respectively, proved the effectiveness of coagulant proteins when Aloe vera and Moringa oleifera were used as organic coagulants to treat water [17,42].

3.1.2. SEM Analysis

The surface morphology of oak leaves was used to characterize their surface properties. SEM was used as a source of photographic images to evaluate the nature of the biomaterial for coagulation of various pollutants present in the water.

Figure 3 illustrates the surface morphology of oak leaves showing that it is characterized by a heterogeneous external surface (in terms of shape and size). This structure favors the coagulation process, due to the presence of several functional groups such as proteins in the oak leaves [46]. The presence of a fibrous structure favors the coagulation of pollutants present in water [47]. Thus, based on these characteristics, it can be concluded that oak leaves have a suitable morphological profile to facilitate the coagulation/flocculation process by improving the adsorption and bridging capacity.

3.2. Factorial Design

3.2.1. Effect of Operational Parameters on Protein Extraction

The effects of the 8 parameters investigated on the extraction efficiency of proteins from oak leaves are observable in Figure 4.

Stirring speed, centrifugation time and speed clearly result to be irrelevant (their effects are negligible), whereas protein extraction was found to be slightly affected by temperature, as protein concentration decreased when the temperature increased. At high temperatures, hydrogen bonds are broken and the proteins are unable to resume their original structure [48,49]. Therefore, stirring speed, centrifugation time, centrifugation speed and temperature have been considered as constants (Figure 4).

Moreover, Figure 4 shows that protein extraction efficiency is strongly affected by pH. Oak leaves present a very low solubility at acid pH (i.e., pH 2), whereas the solubility is high at basic pH (i.e., pH 12), thus reaching a maximum of protein extraction of approximately 3.8 mg/g at pH 12. Generally, the protein extraction rate and solubility are higher at alkaline pH than acidic. Nevertheless, extremely alkaline pH conditions could cause dissociation and disintegration of the proteins [50]. Proteins own a neutral charge at the isoelectric pH (pH_i). At this pH, strong electrostatic repulsions take place between the ionized groups, thus promoting the denaturation of protein macromolecules [48]. Conversely, the presence of lipids in oak leaves is responsible for a decrease in protein dissolution rate because of the formation of an emulsion during the dissolution stage that avoids the protein extraction [48,51]. Stirring time, maceration time and filtration also affect the extraction process. The amount of protein extracted increased when the maceration time decreased. Thus, the maximum number of proteins was obtained at 10 min of stirring time and 4 h of maceration time. This result is a consequence of the presence of lipids and phenolic compounds in oak leaves that prevents the isolation of proteins and causes a decay in their chemical characteristics [51,52], and therefore, of their effectiveness as bio-coagulants.

According to Figure 4, the results clearly show that the extraction process performs better when the filtration step is not carried out, because a large protein amount remained on the filter paper. Several researchers have studied and confirmed the effect of pH, temperature and maceration time on the extraction of proteins from different plants [49,53].

3.2.2. Optimization of Protein Extraction

The number of proteins extracted from oak leaves corresponding to the 16 performed tests is reported in Figure 5.

According to Minitab software, the highest extracted protein concentration (4.895 mg/g) was obtained for the following operating parameters: no filtration, pH of 12, temperature of 20 °C, stirring speed of 300 rpm, stirring time of 60 min, maceration time of 4 h, centrifugation speed of 400 rpm, centrifugation time of 10 min.

The fractional design study resulted in the following empirical Equation (2), where the extracted protein concentration (Y) is expressed as a function of the only relevant factors:

Y = 3.0537 + 0.7432 X₁ − 0.2092 X₂ − 0.1781 X₃ − 0.4001 X₄ − 0.3754 X₅ − 0.3873 X₆+ 0.0993 X₇ − 0.2674 X_1×2 − 0.1888 X₁ X₃ + 0.1143 X₁ X₅ − 0.4863 X₁ X₆

(2)

In this study, the correlation coefficient R² and R² adjusted were 99.75 and 99.05%, respectively. Of equal interest, the p-value was less than 0.05. The quality of the model prediction will also be assessed by the residual curves (see Figure 6a,b) to visualize the difference between the model values and the actual (experimental) values. Figure 6a shows a distribution of residuals reasonably close to the zero axis. According to Figure 6b, all the residuals were well aligned with the Henry line.

The protein concentration extracted from oak leaves in this study was higher than that extracted from leucaena seed kernels (i.e., 1.59 mg/g) when a 3M NaCl solution was used as the solvent (5 g of seeds in 100 mL of 3M NaCl solution) [54], and that extracted from mustard crude (i.e., 4.1 mg/g) [55].

3.3. Application for Water and Wastewater Treatment

The optimal conditions obtained for maximum protein extraction were applied to the jar test experiments to evaluate the effectiveness of the coagulant protein on the drinking water and industrial wastewater quality.

3.3.1. Effect of the Coagulant Protein Dosage on Turbidity Removal from Water

Results regarding turbidity removal from industrial oily wastewater and drinking water by varying coagulant dosage are reported in Figure 7a and Figure 7b, respectively. A decrease in turbidity efficiency removal corresponding to an increase of coagulant dosage (Figure 7a for a dosage of 0.6 mg/L and Figure 7b for dosages higher than 0.17 mg/L) is a consequence of an increase of turbidity due to a combination of the following factors: increase of particle solids concentration (increase of coagulant dosage), re-stabilization of colloidal particles and potential occurrence of phenomena preventing the inter-particle bridges formation [27]. The optimal protein dosage to treat industrial oily wastewater resulted to be 0.538 mg/L, thus achieving a reduction of turbidity by 96.87%, as the value of this parameter dropped from 187 NTU down to 5.85 NTU (Figure 7a). The bio-coagulant was less efficient in treating drinking water, achieving a reduction of turbidity by 69.29%, as the turbidity dropped from 15.7 NTU down to 4.82 NTU (Figure 7b).

Coagulant proteins promote colloids coagulation and, therefore, contribute to their removal, thanks to adsorption and bridging phenomena [9,24]. Actually, proteins are molecules characterized for a long chain length as well as high molecular weight, and both factors are responsible for the formation of bridges involving colloidal particles. Additionally, the fine powder of the coagulant is surface charged and, therefore, is able to be adsorbed by the surface of colloids. This results in the neutralization of negatively charged colloidal particles and the consequent decrease of the zeta potential and suspension stability [33,56].

Several works from the literature have demonstrated the effectiveness of coagulants extracted from plants in removing turbidity from water. For instance, Bouchareb et al. reached a turbidity reduction by 97.14% from wastewater with an initial turbidity of 118 NTU [28]. Moreover, Landázuri-Rojas et al. tested proteins from Moringa oleifera seeds and obtained a turbidity removal by 40.77% [27]. Finally, Muthuraman et al. investigated the effectiveness of proteins extracted from Strychnos potatorum, Moringa oleifera and Phaseolus vulgaris to treat a wastewater with an initial turbidity of 250 NTU and reached a turbidity removal efficiencies of 90.3%, 93.2% and 84.3%, respectively [33].

3.3.2. Effect of Coagulant Protein Dosage on Other Water Parameters

Table 4. Effect of protein dosage on water characteristics.

Parameters	Industrial Oily Wastewater		Drinking Water
	Pre-Treatment	Post-Treatment	Pre-Treatment	Post-Treatment
Protein dosage (mg/L)	/	0.538	/	0.098
Turbidity (NTU)	187	5.85	15.7	4.82
Turbidity removal (%)	/	96.87	/	69.29
pH	9	9.68	7.75	8.46
OM (mgO2/L)	17.5	9.4	2.1	5.2
TSS (mg/L)	969	100	/	/
TSS removal (%)	/	89.86	/	/
COD (mg/L)	784.45	28.278	/	/
COD removal (%)	/	96.395	/	/
TA (mg CaCO3/L)	/	/	134	142
Salinity (g/L)	/	/	0.6	0.6
Ec (µs/cm)	/	/	782	935

reports the variation of physical and chemical characteristics of industrial oily wastewater and drinking water after treatment with the bio-coagulant dosed according to the best performing dosage (i.e., 0.538 mg/L for industrial oily wastewater and 0.098 mg/L for drinking water). In Table 4, it can be noticed that the coagulant proteins were responsible for increasing the water pH from 7.75 to 8.46 in drinking water and from 9.0 to 9.68 in industrial oily wastewater. This increase is likely due to the bio-coagulant powder that was prepared in a basic medium (pH = 12) by adding NaOH.

Ec of drinking water increased from 782 µS/cm to 935 µS/cm. This variation is due to the natural presence of ions in the bio-coagulant (Na⁺ and OH⁻), and all of them contribute to an increase in the Ec value. Similar results were obtained by Benalia et al. and Shan et al. [17,57]. The values of OM content after treatment were 5.2 mg O₂/L and 9.4 mg O₂/L for drinking water and industrial oily wastewater, respectively. The increase of OM content from 2.1 mg O₂/L to 5.2 mg O₂/L in drinking water is due to the supply of OM from the bio-coagulant, whereas the decrease of OM content from 17.5 mg O₂/L to 9.4 mg O₂/L in industrial oily wastewater is due to the turbidity removal induced by the coagulant that produced an OM content reduction greater than its increase due to the addition of bio-coagulant.

Several researchers have found that the use of coagulant proteins to treat drinking water is responsible for an increase of OM concentration [17,58].

Salinity in drinking water did not change as a consequence of the bio-coagulant addition, thus remaining equal to 0.6 g/L. This result is due to the absence of production or consumption of mineral salts responsible for water salinity, such as sodium chloride, potassium chloride, etc.

Adding 0.098 mg/L of bio-coagulant resulted in a small increase of TA, which moved from 134 mg CaCO₃/L to 142 mg CaCO₃/L. This result is a consequence of the basic nature of the bio-coagulant (pH = 12) that released hydroxide ions (OH⁻) in water.

Concerning TSS and COD concentrations, the maximum removal efficiency obtained for 0.538 mg/L of the coagulant protein resulted in being 89.86% and 96.39%, respectively. This result is due to the destabilization of the colloidal suspension and the following aggregation of solids, promoted by charge neutralization and bridging mechanisms, as mentioned in a previous section, which favored their removal by settling [28,56].

Similar findings were observed by Shan et al., Benalia et al. and Bouchareb et al. [17,28,57].

4. Conclusions

The fractional design allowed to evaluate and reduce the number of factors affecting the extraction and purification of proteins. The quality of drinking water and industrial wastewater were also evaluated before and after treatment with proteins of oak (Quercus robur) leaves. Several water parameters were measured. In this case, the maximum protein extraction yield was applied for the water treatment. The maximum turbidity removal performances were 69.29% and 96.87% for drinking water and industrial oily wastewater, respectively. Moreover, the removal efficiency for COD, TSS and OM concentrations in industrial oily wastewater were 96.39%, 89.86% and 46.28%, respectively.