Characterization and application of biochar-immobilized crude horseradish peroxidase for removal of phenol from water
Mirjana Petronijevic´ a, *, Sanja Panic´ a, Saˇsa Savic´ b, Jasmina Agbaba c, Jelena Molnar Jazic´ c, Marija Milanovic´ a, Nataˇsa Đuriˇsi´c-Mladenovic´ a
aUniversity of Novi Sad, Faculty of Technology Novi Sad, 21000, Novi Sad, Bulevar cara Lazara 1, Serbia
bUniversity of Niˇs, Faculty of Technology, 16000, Leskovac, Bulevar Oslobođ enja 124, Serbia
cUniversity of Novi Sad, Faculty of Science, 21000, Novi Sad, Trg Dositeja Obradovi´ca 3, Serbia

A R T I C L E I N F O

Keywords: Biochar
Horseradish peroxidase Phenol removal
Water treatment Toxicity
A B S T R A C T

Biochar (BC) has attracted much attention as an environmentally friendly material for application in wastewater treatment. In this study, a suitability of wood-derived BC as a support for covalent immobilization of horseradish peroxidase (HRP) across glutaraldehide as crosslinker, known for the capability to remove phenol from water, was investigated. The efficiency of the immobilized HRP in removal of phenol (2 mM) from water at different reaction conditions (varying dosages of polyethylene glycol (PEG300) 0-750 mg/L; H2O2 1.5–3.5 mM, as well as reaction time 5-120 min) and the general toxicity of bio-treated water (Allium cepa test) were measured. All analyzes were performed for free enzyme as well. The immobilized enzyme showed the highest activity at temperature 30 ◦ C and pH 7.0. The greatest efficiency of immobilized enzyme in phenol removing (90 %) was obtained by applying 2.5 mM H2O2 and 1.5 mg/L of PEG300 at pH 7.0 after 2 h of reaction period. After 4 washings, immobilized HRP retained more than 79 % activity with phenol removal of 64 %. Utilizing immo- bilized enzyme significantly reduces the toxicity of the tested water (80 %), which further suggested that it might be considered as an environmentally acceptable process for wastewater treatment. Possible degradation products remained in treated water were analyzed in water samples by liquid and gas chromatography with mass spec- trometry, including also analysis of volatiles by solid phase microextraction technique; different phenol-base compounds were detected.

1.Introduction
Phenols and their derivates can be found as pollutants in wastewater from different industries (such as gas and coke oven industries, petro- leum refineries, phenol–formaldehyde resin manufacture, pharmaceu- ticals, etc.) in wide range of concentrations 1.1–7000 mg/L [1]. If such wastewater is not treated properly, it can pose a potential risk to the environment and human health [2]. The application of conventional wastewater treatments achieves only a degree of water purification at affordable cost, so new alternative technologies for the treatment of phenolic wastewater are developed. Biological methods of water treat- ment (e.g. by enzyme catalytic reaction) have the advantage of almost complete removal of pollutants from the water, at a lower cost [3].
Horseradish peroxidase (HRP) is widely used as an effective catalyst for phenol removal [4,5], and can be utilized in free and immobilized form. In the presence of hydrogen peroxide HRP can catalyze the

formation of organic phenoxy radicals through which subsequent polymerization of phenolic compounds occurs, whereby the toxicity of phenolic compounds may be reduced [6]. Enzyme immobilization is the process of converting an enzyme from a soluble to an insoluble form and one of the ways in which immobilization is performed is binding an enzyme onto a solid material. Some limitations that exist with the use of free enzyme (non-reusability, low stability during time, etc.), can be overcome by immobilization. Covalent immobilization of enzyme onto the solid support gives a stable biocatalyst with higher activity than the free enzyme, and it can be used several times in repeated cycles [7,8]. The possibility of enzyme reuse significantly reduces the consumption of enzymes and the cost of the production process. In order to increase the stability of the enzyme various additives, such as polyethylene glycol, gelatin and certain polyelectrolytes, can be used.
Selecting a suitable support for the immobilization of enzymes is very important, because it affects the filling of the enzyme, operational

* Corresponding author.
E-mail address: [email protected] (M. Petronijevi´c). https://doi.org/10.1016/j.colsurfb.2021.112038
Received 30 March 2021; Received in revised form 22 July 2021; Accepted 12 August 2021 Available online 14 August 2021
0927-7765/© 2021 Elsevier B.V. All rights reserved.

stability and overall process costs [2]. A variety of solid supports (carbon nano-tubes, graphene oxide, activated alumina, chitosan, etc) have been evaluated for removal of phenolic compounds from water [3,9–12]. However, biochar (BC) has been rarely investigated as a support for HRP in phenol removal [13], even though it is known as an efficient, economical and environmentally-friendly material for soil and water remediation [14]. Due to its large specific surface area and porosity, it exhibits good adsorbent capacity and also can provide a large surface area for the catalytic reaction. The selection of the appropriate material which will be used as a support as well as the choice of the enzyme immobilization method largely depends on the characteristics and goals of the application process itself.
This paper explores BC as a support for HRP by investigating the stability and efficiency of the immobilized HRP in phenol removal from water under different reaction conditions. Under the same range of the experimental conditions, immobilized HRP is compared with the free HRP. In order to determine the effect of the HRP treatment on the overall water toxicity, Allium cepa seeds were exposed to both untreated and treated water. Hence, this study complements the existing results on the applicability of BC as an environmentally friendly material for waste- water treatment.

2.Materials and methods
2.1.Materials
Glutaraldehyde, phenol 99 %, polyethylene glycol (PEG 300), 4-ami- noantipyrine (4-AAP) and all other chemicals were obtained from Fisher Scientific or Sigma-Aldrich.
BC was produced by biochar producing company Basna doo, ˇCaˇcak, Serbia, using sawdust of beech and oak wood mixture, which was py- rolyzed at 700 ◦ C under atmospheric pressure. BC was ground in a planetary ball mill (FRITCH) at 200 rpm for 5 min and the obtained particles were sieved (500-800 μm). BC was washed with hot deionized water until the dark leachate ceased to be separated.
Horseradish root was used as the raw material for horseradish peroxidase (HRP) production. The plant samples were collected at a local market (Novi Sad, Serbia). The preparation of the HRP extract was performed according to the procedure given in the paper of Savic´ et al. [4] with minor modifications. Horseradish root (4 g) was extracted by stirring with 20 mL phosphate buffer (50 mM, pH 7.0) for 30 min, centrifuged at 2000 rpm for 15 min and used as concentrated crude HRP.

2.2.BC functionalization and enzyme immobilization
The functionalization of BC with concentrated HNO3 was performed according to the method given by Naghdi et al. [8]. About 20 g of the BC was dispersed in 400 mL of 85 % HNO3 and mixed at 1000 rpm and 70 ◦ C for 4 h. Functionalized BC (FBC) suspension was washed with deionized water to achieve pH 7.0, dried overnight at 80 ◦ C and stored at room temperature.
FBC was prepared for covalent immobilization of HRP using glutaraldehyde as a cross linker [11]. FBC was placed in 50 mL of 50 mM phosphate buffer (pH 7.0) for 24 h to achieve its balancing. After that, the same volume of 1.0 % w/v glutaraldehyde was added to the solution and shacked at 300 rpm for 12 h at room temperature (25 ◦ C). The excess glutaraldehyde was washed with deionized water and biochar (marked as GBC, the full molecular form appears BC-O-CHOH-(CH2)3– CHO) was dried overnight at 80 ◦ C and stored until enzyme immobilization.
Enzyme immobilization was performed by mixing the GBC with the crude enzyme in ratio of solid:liquid = 1:5 at 25 ◦ C with mixing (300 rpm). This mixture was incubated for 3 h. The enzyme-loaded biochar (BC-HRP) was separated, washed twice using the buffer and the enzyme activity was measured. BC-HRP was dried at room
temperature and used in water treatment.

2.3.Characterization of BC samples
To determine characteristic parameters such as moisture, dry matter, volatile matter, ash and fixed carbon of pristine BC, the ASTM Interna- tional standards were applied [15,16]. Fixed carbon content was calculated according to the equation given in Zhao et al. [17].
The pH of BC was measured in triplicates with a pH meter (model Accumet 910, Fisher Scientific) according to the standard method [18].
Zeta potential of the samples was measured in triplicates using a Zetasizer Nano ZS (Malvern Panalytical Ltd., Malvern, UK) with deion- ized water as the carrier fluid (sample concentration was 0.1 mg/mL, refractive index 2.420 and an absorbance 0.9000 were used) [19].
Textural characteristics of the samples were determined by means of low temperature N2 adsorption/desorption (LTNA) method, using He as a carrier gas (Micromeritics ASAP 2010). Specific surface area was calculated by BET equation, while mean pore diameter and pore volume were determined from desorption part of the N2 isotherm and calculated by Barrett-Joyner-Halenda (BJH) method [20]. Pores were classified according to Brunauer-Deming-Deming-Teller method based on hys- teresis loops of adsorption-desorption isotherms [21]. Surface functional groups were detected by Fourier transform infrared spectroscopy (FTIR) (Bruker Vertex 70) at a spectral range of 4000 cm-1 to 400 cm-1 with a resolution of 8 cm-1. The crystallographic structure of the biochar samples was analyzed by X-ray diffraction (XRD) using Rigaku MiniFlex 600 diffractometer (CuKα radiation, λ = 1.5406 Å) in 2θ range 10–80 ◦ with a scan rate of 0.03 ◦ /s. Scanning electron microscopy (SEM) (JEOL JSM-6460LV on 25 kV) was used to investigate the morphology of the samples. The method of microanalysis using X-ray energy dispersion (EDS) is performed in combination with the SEM method and provides insight into the elementary composition of the sample.
2.4.Enzyme assay
The activity of free and immobilized HRP were measured in triplicate spectrophotometrically in the presence of 4-AAP/phenol and hydrogen peroxide as substrates at 25 ◦ C according to Worthington method [22]. The results of activity values for free and immobilized enzyme were presented as relative values (%, where the highest value of enzyme ac- tivity was used as 100 %).
To determine the effect of temperature on enzyme activity, free HRP and BC-HRP were tempered at different temperatures (20–80 ◦ C) for 15 min at pH = 7.0 and then activities were measured.
For evaluation of the impact of pH on enzyme activity, the investi- gation was performed using the buffers with different pH values (4.0–8.0) at 25 ◦ C.
The storage stability of the free and BC-HRP was evaluated by keeping the samples at 25 ◦ C for one month and measuring their activity every three days.

2.5.Phenol removal experiments
The concentration of phenol was determined by a spectrometric method using 4-AAP and ferricyanide [23,24]. The assay mixture con- sisted of 0.25 mL of ferricyanide solution (83.4 mM of K3Fe(CN)6 in 0.25 M NaHCO3), 0.25 mL of 4-AAP solution (20 mM in 0.25 M NaHCO3) and 2 mL of phenolic sample. Red color is released and measured at 510 nm.
The impact of free HRP and BC-HRP on phenol removal from water during time was investigated. The total phenol content in initial reaction mixture was 2 mM (i.e. 188 mg/L in deionized water), which might be regarded as the model wastewater representative for petroleum re- fineries regarding the phenol content [1]. The batch reactions were set up with phenol and enzyme with/without PEG in 50 mM phosphate buffer pH = 7.0 (at 25 ◦ C) and the reaction continued with addition of

H2O2. The applied dose of immobilized enzyme was 0.5 U/mL. The treatment was performed with an equivalent amount of free enzyme. The investigated PEG dosages were in the range 0-750 mg PEG/L [4], while keeping the H2O2 concentration at c(H2O2):c(phenol) = 1:1. The impact of the H2O2 dosages on the phenol removal efficiency was investigated in the range 1.5–3.5 mM [4,9]. Aliquots of the reaction mixture were taken at 5-120 min and assayed for phenol content.
2.6.Reusability
Enzyme activity retained after washing by recovering BC-HRP with phosphate buffer (50 mM; pH 7.0) over 7 cycles was investigated.
The reusability of BC-HRP for phenol removal was measured at 25 ◦ C over 5 cycles. Tests were performed by 0.5 U/mL BC-HRP at optimal dose of H2O2 and PEG. In running the second and subsequent cycles, a new amount of phenol solution is applied and processed as described above. After each cycle, BC-HRP was separated and washed with the phosphate buffer. The results are presented in % of phenol removal in each cycle of the BC-HRP use.

2.7.Toxicity assay
Assessment of general water toxicity after treatment using BC-HRP was performed using the Allium test according to the method given in the study of Rank and Nielsen [25]. The general toxicity of water sample was evaluated based on the inhibition of root growth after its exposure to the sample for a certain period of time. The values of root length (mm) for control (deionized water), untreated phenol-containing water (as a model for phenol-polluted wastewater) and water treated at optimal conditions (with free and BC-HRP) were investigated.

2.8.Analytical methods for phenol intermediates determination in biotreated water samples
Determination of the possible products of phenol enzymatic degra- dation was performed in water samples biotreated with BC-HRP for 30 and 120 min of reaction time; additionaly, 2 mM phenol solution, BC- HRP extract in phosphate buffer, and BC extract in deionized water, were analyzed for the content of organic compounds in order to differ- entiate possible phenol-degradation products from those originating from the added phenol solution, or leached from the BC samples (BC- HRP or BC).
The samples were prepared for gas chromatography/mass spec- trometry (GC/MS) analysis of phenol and catechol, as possible oxidation intermediate, by derivatization with acetic anhydride (acetylation – derivatization of the polar OH group with an acetyl group) and liquid- liquid extraction with hexane [26,27]. An Agilent Technologies 7890B gas chromatograph with a 5977A quadrupole mass spectrometer (Santa Clara, California, USA) with HP-5MS column (30 m x 0,25 mm x 0, 25 μm, J&W Scientific) was used for analysis. The chromatographic conditions for SIM/SCAN method were as follows: inlet temperature: 210 ◦ C (splitless injection mode); initial oven temperature was 50 ◦ C (1 min), then 12 ◦ C/min to 160 ◦ C (2 min); 35 ◦ C/min to 300 ◦ C (5 min). MS Source temperature was 230 ◦ C, and MS Quad temperature was 150 ◦ C. The MS was set at scan mode covering range from 50 to
extraction, the fiber was conditioned in accordance to the recommen- dation of the manufacturer. SPME working conditions were as follows: equilibration time 10 min, equilibration temperature 50 ◦ C, extraction time 40 min, and extraction temperature 50 ◦ C, while the solution was stirred at 400 rpm. Desorption was performed in the GC injector in split mode (split ratio 5:1) at 250 ◦ C for 3 min. Helium was used as a carrier gas at the flow rate of 1 mL/min at a constant flow. The oven was maintained at 40 ◦ C for 1 min, then the temperature increased to 150 ◦ C (8 ◦ C/min), followed with the increase of 15 ◦ C/min to 200 ◦ C, then with 10 ◦ C/min to 290 ◦ C, which was maintained for 2 min. The injector temperature was 250 ◦ C. The ion source, quadrupole, and transfer line were held at 230 ◦ C, 150 ◦ C, and 280 ◦ C, respectively. The data acqui- sition and analysis were done in GC/MS MassHunter, with the NIST 11 reference library used for unknown identification using an 80 % simi- larity match cut-off value.
UHPLC-DAD-ESI-MS/MS analysis was performed by using a Thermo Scientific liquid chromatography system composed of a quaternary pump with a degasser, a thermostated column compartment, an auto- sampler, and a diode array detector connected to LCQ Fleet Ion Trap Mass Spectrometer (Thermo Fisher Scientific, San Jose, California, USA) equipped with electrospray ionization (ESI). Xcalibur (version 2.2 SP1.48) and LCQ Fleet (version 2.7.0.1103 SP1) software were used for instrument control, data acquisition, and data analysis. Separations were performed on a Hypersil gold C18 column (50 × 2.1 mm, 1.9 μm) obtained from Thermo Fisher Scientific. The mobile phase consisted of (A) water (Fisher Chemical, HPLC grade) and 0.1 % formic acid (Carlo Erba, Italy) and (B) acetonitrile (Fisher Chemical, LC–MS grade). A linear gradient program at flow rate of 0.25 mL/min was used 0–2 min from 10 to 20 % (B), 2–4.5 min from 20 to 90 % (B), 4.5–4.8 min 95 % (B), 4.8–4.86 min from 90 to 10 % (B) and 4.86–12 min 10 % (B). The injection volume and temperature of the column were 5 μL and 25 ◦ C, respectively. The analysis was followed at four wavelengths: 210, 254, 310 and 354 nm. The mass spectrometer was operated in positive ion mode as the method of identification. ESI-source parameters were as follows: source voltage 4.5 kV, capillary voltage 39 V, tube lens voltage 115 V, capillary temperature 275 ◦ C, sheath and auxiliary gas flow (N2) 50 and 8 (arbitrary units), respectively. MS spectra were acquired by full range acquisition in the 150–1000 m/z range. For the fragmentation study, a data-dependent scan was performed by deploying the collision- induced dissociation (CID). The normalized collision energy of the collision-induced dissociation (CID) cell was set at 25 eV. Compounds were identified according to the corresponding spectral characteristics: mass spectra, accurate mass and characteristic fragmentation.
3.Results and discussion
3.1.Pristine BC characterization
The basic and textural characteristics of pristine BC are presented in Table 1.
Based on the results obtained by gravimetric analysis, BC has a high

Table 1
Physic-chemical and textural characteristics of BC.

450 m/z. Monitored ions m/z for phenol were 94, 136, and 66, and for catechol 110 and 152.
For qualitative analysis of volatile organic compounds (VOCs) pre- sent in the samples, the same GC/MS configuration was used (Agilent 7890B GC/Agilent 5977 MS with an HP-5MS capillary column (30 m, 0.25 mm i.d., 0.25 film thickness), but this time equipped with solid- phase microextraction (SPME) unit. SPME is a solvent-free sample pre- treatment technique that integrates sampling, isolation and enrich- ment of analytes into one step. VOCs were extracted by 65 μm poly- dimethylsiloxane/divinylbenzene (PDMS/DVB) SPME fiber (23 GA,
Parameter pH
Zeta potential Dry matter Moisture Volatile matter Ash
Fixed carbon BET
Mean pore diameter Pore volume
Units
/
mV
%
%
%
%
% m2/g nm cm3/g
Value ± SD 6.6 0.1
±
-15.5 ± 1.04 84.8 ± 0.27 15.2 ± 0.35 78.3 ± 0.28 4.74 ± 0.52 1.75 ± 0.65 230.2
4.6
0.039

Supelco 57284-U, Bellefonte, Pennsylvania, USA). Prior to the SD – standard deviation for n number of measurements.

content of volatile matter and low content of ash and fixed carbon (Table 1). The literature data show that content of volatile matter (28.0–74.1 %), ash contents (1.50–65.7 %), as well as fixed carbon (0–47.8 %) may significantly vary in different wood-derived biochars [28].
BC has a monomodal pore size distribution profile and high specific surface area (230.2 m2/g, Table 1). Oak and hickory wood-derived BC obtained elsewhere at temperature of~500 ◦ C had much lower surface area (5.41 m2/g) [28], compared to BC used in this work. Porosity and surface area of BCs increase with increasing treatment temperatures [29]. Based on the results in this study, it can be presumed that the pyrolysis process (700 ◦ C) applied on beech and oak sawdust mixture influenced development of rather high specific surface area of BC compared to those from the cited literature source.
The adsorption-desorption isotherm and the pore size distribution of BC are presented on Fig. 1. Adsorption-desorption isotherm of BC is characterized by H4 type of hysteresis loop which indicates the presence of a certain portion of micropores [21]. The hysteresis loop is not closed, i.e. in the area of the lowest relative pressures, some particles were not subjected to desorption. This phenomenon is caused by blocking the pores due to their narrow openings in relation to the dimensions of the interior and indicates the presence of a specific “ink bottle” shape of the pores.
The pH value of untreated BC is almost neutral (Table 1). According to the literature data, the pH value of wood-derived BC could be in the range of 6.24–8.8 [28–30].
The value of the zeta potential is affected by the pH and ionic strength of the solution. Since the value of BC zeta potential (-15.5 mV, Table 1) is between -10 and -30 mV which means that the particles are little unstable [31].
Further characterization of pristine BC is described leather compar- atively to FBC, GBC and BC-HRP.

3.2.Covalent HRP immobilization onto biochar

Immobilization of the enzyme by the covalent attachment provides a permanent bond between the enzyme and the support. However, in order to achieve adequate immobilization of enzymes, it is necessary to make a chemical modification of BC that would lead to the formation of reactive functional groups through which covalent binding would take place. Functional groups, such as hydroxyl and carboxylic groups,
provide adequate points for covalent binding of enzymes using cross- linkers [11,32,33].
The covalent immobilization of HRP onto BC takes place in two steps and the mechanism are present on Fig. 2. The first part of the process is the chemical modification of BC using nitric acid, which results in the formation of reactive hydroxyl groups on the surface of BC. Glutaral- dehyde (cross-linker) is covalently bonded with an aldehyde group at one end of its chain to the hydroxyl group onto BC, while the other aldehyde group of glutaraldehyde remains free. The second part of the process involves the immobilization of the enzyme onto BC across glutaraldehyde, i.e. the binding of the amino group of the enzyme to the free aldehyde group of glutaraldehyde using strong covalent bonds.
The activity of HRP immobilized onto BC was 6.46 ± 0.22 U/g BC at pH 7. The BC itself showed peroxide-like activity (0.47 U/g BC), reacting with phenol/H2O2 during the test, however, significantly lower compared to the activity of BC-HRP.
The FBC (-25 mV) and G-BC (-20 mV) have more negative charge than BC, which was expected after activation process and generation of negative functional groups on the surface of BC. The zeta potential of BC-HRP was less negative (-15 mV), probably because the -NH2 group from HRP was bound.
3.2.1.Comparative characterization of biochar samples
The surface morphology and the elemental composition of the pre- pared BC samples (BC, FBC, GBC and BC-HRP) and free HRP are shown in Fig. 3, respectively. All biochar samples are characterized by het- erogeneous surface and developed porous structure (Fig. 3A). Biochar samples contain a large portion of arranged, mostly macroporous structures in the shape of a honeycomb. According to the SEM results, the BC activation process increases the porosity of the surface of BC. It could be noticed that oxidation treatment with nitric acid, in addition to the generation of functional groups, had an erosive effect on the struc- ture of BC (i.e. FBC) and led to damage of pore wall. Other authors have come to similar conclusions that activation of BC by nitric acid cause erosive effects on the carbonaceous structure [34] and leads to the opening of microporous structures, increasing the surface area of the material [35]. On the surface of BC after immobilization (BC-HRP), small round lumps can be observed, which indicates the bound enzyme to the surface of the support. However, no clear and visible change in the surface texture of BC was observed after HRP immobilization.
Based on the EDS results, all tested biochar samples mainly consist of

Fig. 1. A. N2 adsorption-desorption isotherm and B. BJH pore size distribution from the desorption part of the isotherm obtained for BC used for enzyme immobilization.

Fig. 2. Mechanism of covalent immobilization of HRP onto biochar.

Fig. 3. Characterisation of BC samples. A. SEM images of pristine biochar (BC), functionalized biochar (FBC), functionalized biochar with glutaraldehyde (GBC), immobilized HRP onto biochar (BC-HRP) and free HRP. B. Elemental composition of BC samples obtained by EDS measurements.

Fig. 4. A. XRD and B. FTIR spectra of BC samples.

carbon and oxygen (Fig. 3B). The remaining elements are trace ele- ments: non-metallic (P, N, S and Cl) and metallic (K, Ca, Na, Al). In BC after the functionalization (FBC) and after the binding of glutaraldehyde (GBC), the oxygen content increases (up to 26.88 %), which can be associated with the formation of oxygen functional groups on the surface of biochar. The obtained results are in agreement with previous studies [28–30]. The free enzyme is rich in both non-metallic elements, including nitrogen, phosphorous and sulphur, as well as metallic ele- ments (K ~ Na > Ca > Mg), which can also be found in BC-HRP. However, nitrogen was not detected in BC-HRP, probably as a conse- quence of a very small content of enzyme compared to the content of carbon in biochar.
The XRD spectra of biochars are shown in Fig. 4A and it could be seen that pyrolysis of the beech and oak sawdust mixture at 700 ◦ C led to the formation of a turbostratic carbon structure. Main peaks of turbostratic phase are (002) at 23.6 ◦ and (100) at 43.5 ◦ , respectively [36,37]. This structure has an ordering in between that of amorphous carbon phase and crystalline graphite phase, which explains the broad diffraction peaks. Moreover, the BC sample had additional sharp peaks that indi- cated the presence of inorganic components. The peak at 2θ around 26 ◦ corresponds to the crystalline quartz, while the other sharp peaks belong to the calcite phase [38,39]. These peaks most likely originated from the ash content and tarry materials trapped within the pores of the BC sample [40]. In other samples, these peaks were eliminated or signi- ficantly reduced by the applied procedure, confirming the removal of the majority of ash content. The FBC, GBC and BC-HRP samples were characterized by similar XRD patterns suggesting that the presence of functional groups, glutaraldehyde and HRP cannot be identified by this method.
FTIR adsorption spectra of BC samples (BC, FBC and BC-HRP) are presents on Fig. 4B. All biochar samples were characterized by a wide vibrational band at 3400 cm-1, which can be attributed to the phenol groups or tensile vibrations of the -OH group [33]. In the spectrum of BC a few peaks were observed, including: the bond at 2920 cm-1 present stretching vibrations -CH2 (asymmetric –C–H stretching vibrations – aliphatic), the bond at 2851 cm-1 as stretching vibrations -CH3 (sym- metric –C–H stretching vibrations – aliphatic) [41], the peak at about 1600–1700 cm-1 is associated with the bond of carbonyl groups to the aromatic ring [42], the peak at 1556 cm-1 presents aromatic –C = C– vibrations, the picks at 1077 cm-1 presents symmetric stretching vi- brations of carbonate ion and band at 880 cm-1 (bending vibrations of carbonate ion out of the plane) [41]. The FBC have almost the same bonds but of higher intensity: a band at 2977 and 2880 cm-1; the peak at about 1600–1700 cm-1; the sharp peak shown at 1091 cm-1; peak at 880 cm-1, while the band at 1046 cm-1 can be attributed to aliphatic ether, alcohol C–O or aromatic stretching peak, O–H deformation vi- brations, b-glycosidic bond in cellulose and hemicellulose [42]. BC-HRP sample showed a low peak at 1384 cm-1 and a peak at 1100 cm-1,
corresponding to C–N stretching vibration of amines [43], while the bonds at 3000–3500 cm-1 can be ascribed to OH vibration and CONH [33]. Also, the broad peak appeared after enzyme immobilization at 1586 cm-1 can be attributed to the Amide I in proteins [33,44]. Those suggest that there is an interaction between HRP and BC support.

3.3.Characteristics of free and immobilized HRP
The activities of free and immobilized HRP at different temperatures and pH values and their stability during time are present in Fig. 5.
The activity of free and immobilized HRP firstly increased from 20 ◦ C to 40 ◦ C and from 20 ◦ C to 30 ◦ C, respectively, then decreased to 80 ◦ C in both cases, but different decreasing trends might be observed (Fig. 5A). Unlike expected, the free HRP showed a broader temperature profile (at 30-60 ◦ C activity was higher than 80 %) compared to the immobilized HRP. But at a temperature of 80 ◦ C, the free enzyme showed almost no activity while immobilized HRP still retained 20 % activity. At temperatures above 40 ◦ C, the free enzyme activity begins to decline due to denaturation. Similar results that enzyme activity became more dependent on the temperature after immobilization, were ob- tained by Chagas et al. [45], who performed immobilization of peroxi- dase onto cytosal via glutaraldehyde as a cross-linker. The process of enzyme immobilization onto solid support has a variety of effects on protein conformation and could cause some changes in the properties of the enzyme molecule (e.g. their catalitic activity, thermal stability), compared to the properties of free enzyme [46,47]. The active site of the enzyme can be distorted and/or destroyed when the carrier has a high coefficient of expansion, leading to contraction or expansion during temperature changes [48].
The activities of free and immobilized HRP were sensitive to pH change in similar manner and optimal value was achieved at pH 7.0 for both enzymes (Fig. 5B). Enzyme activity in an acidic medium was very low and increases with increasing pH. These results are consistent with the results obtained by Chen et al. [12], although some authors indicate that pH 6.0 is optimal for HRP [49].
The activity of free enzyme decreased to 50 % after three days, while after 12 days no enzyme activity was observed (Fig. 5C). On the other hand, the BC-HRP retains up to 80 % of its activity in the first 6 days, after which the activity decreases to 60 % and stay unchanged until the end of the examined period of one month. Other researchers come to the same conclusion on higher storage stability of immobilized than free HRP even though they used different carbon supports (graphene oxide and nitrogen-doped carbon dots, respectively) for the enzyme [11,12].

3.4.Phenol removal

To examine the practical application of immobilized HRP onto bio- char, the efficiency of both free and BC-HRP in phenol removal from

Fig. 5. The impact of A. temperature and B. pH values on activities of free and immobilized HRP. C. The stability of free and immobilized HRP at 25 ◦ C during time.

Fig. 6. A. The efficiency of phenol removal from water by free HRP and BC-HRP during time (without PEG). B. The impact of PEG on phenol removal after 2 h. C. The impact of H2O2 on phenol removal after 2 h. Reaction conditions: 0.5 U/mL enzyme, 2 mM phenol, 150 mg/L PEG, pH 7.0.

aqueous solution was investigated. The efficiency of bio-catalytic water treatment during reaction time of 2 h was present on Fig. 6.
The use of immobilized HRP achieves a greater decrease in the content of phenol (up to 69 %) compared with the free HRP (up to 42 %) (Fig. 6A). The low rate of phenol degradation by free HRP is probably because the phenoxy radicals or phenolic polymers occupy the active centers of the enzyme during the reaction of biodegradation, which in- activates the enzyme [50]. Utilizing an appropriate amount of biochar achieves the 50 % of phenol removal after 2 h (Table S1). The hydro- phobic groups on carbon materials increase the affinity of enzymes to the phenolic compounds [12]. On the other hand, the immobilization can adapt HRP to expose more active sites to the substrate [11] and create an apparent synergetic effect between carbon support and HRP [12]. Chen et al. [12] showed that after half an hour of reaction about 55
% of the phenol was removed by free HRP, while 65 % was removed by HRP immobilized on nitrogen-doped carbon dots. The removal effi- ciency at high concentration of phenol (2500 mg/L) was 100 % for the HRP immobilized onto graphene oxide and 55 % for free HRP [11].
The presence of PEG increases the stability of enzymes, and thus their efficiency in removing of pollutants [4,51]. This can also be seen in the results for free HRP and BC-HRP (Fig. 6B) In the case of the free enzyme, the highest degree of phenol removal of 58 % was achieved at a PEG dose of 300 mg/L, while a further increase in the PEG concentration reduces the efficiency of the enzyme. Removal of phenol using an immobilized enzyme is also accelerated by the addition of PEG, at a dose of 150 mg/L a reduction of 85 % was achieved. However, it was not possible to establish a clear trend with a further increase in PEG dose. With a further increase in PEG dose efficacy of phenol removal slightly decreased and does not change significantly in the range of 300-750 mg/L.
The efficiency in phenol removal of both free and immobilized HRP follow the same trend with increasing H2O2 dose (Fig. 6C). The degree of phenol removal increases with increasing dose of H2O2 up to 2.5 mM (up to 90 % using BC-HRP and 48 % using free HRP). By further increasing the dose of H2O2, the efficiency of the process decreases. The reason for this phenomenon could be that an excessive amount of H2O2 results in higher contents of intermediates that inhibit enzyme activity, and/or inactivation of the enzyme in an excess of H2O2 [9]. However, H2O2 is applied in rather narrow range (c (H2O2):c (phenol) = 0.75–1.75), which might be the reason why no large difference in the efficiency of the process was observed with respect to the tested doses of H2O2. Other authors showed that the optimal conditions in phenol removal (2 mM) was 0.8 U/g of support of HRP and H2O2/phenol ratio 1.15 [9]. The use of free HRP in low concentration (0.3 U/mL) led to the 40–60 % phenol removal [51]. For more efficient phenol removal by BC-HRP, it is necessary to increase the degree of enzyme immobilization onto BC (e.g. by optimizing the functionalization process of BC via glutaraldehyde) or
by applying a higher concentration of the immobilized enzyme.

3.5.Reusability
Stabilization of HRP due to binding to BC enabled the easy reuse of the immobilized enzyme in many cycles (Fig. 7).
The BC-HRP retained 67 % of the initial activity after six washing times (cycle VII, Fig. 7A). The reuse of BC-HRP allows the removal of phenols by 84 % under optimal reaction conditions, while when immobilized enzyme was used again in the fifth cycle the efficiency is reduced to 64 % (Fig. 7B).
The reusability of HRP covalently bonded onto graphene oxide was 70 % of initial activity retained after 10 cycles [11]. Zhang and Hay [36]
showed that HRP immobilized onto biosolid-delivered BC retained 51.5
% of their initial activities after 10 cycles. Lai et al. [10] for HRP immobilized on another kind of support, it was found that 50 % of the initial activity is lost after five cycles.

3.6.General toxicity

After both bio-treatments (with free and BC-HRP), the length of the roots increases compared to length recorded in untreated phenol- containing water, which indicates that the treatment achieves a reduc- tion in the general toxicity of water (Fig. S1). Allium cepa roots grown in free HRP and BC-HRP treated waters, showed increase in root length up to 73 % and 80 % compared to untreated water, respectively. Bilal et al. [52] also showed that the application of peroxidase can significantly reduce the toxicity of waste water and that the use of immobilized peroxidase achieves better results compared to free enzyme.

3.7.Analyses of phenol and intermediates in treated water

In the presence of H2O2, peroxidases can effectively catalyze the oxidation of phenolic compounds and generate various phenoxy radi- cals. The generated free radicals can subsequently couple with each other and form insoluble polymers, which could be easily separated from solution by filtration or decantation [43,53]. To determine the possible phenol intermediates formed during treatment with BC-HRP and remained in treated water, the samples were analyzed by (SPME) GC/MS and UHPLC-DAD-ESI-MS/MS as previously described; the results are presented in Tables S2 and S3, and in Fig. S2.
GC/MS analysis confirmed the presence of phenol remained in bio- treated water samples in the form of its acetic derivate (acetic acid phenyl ester) (Fig. S2), which peak gradually decreased in samples with the treatment period. Catechol was not identified in treated water samples (Fig. S2). Using UHPLC-DAD-ESI-MS/MS analysis of the liquid phase, two phenol polymerization products, biphenol and 4-

Fig. 7. Reusability of BC-HRP in consecutive cycles expressed as A. retained enzyme activity and B. maximal efficiency in phenol removal under optimal conditions (0.5 U/mL BC-HRP, 2 mM phenol, 150 mg/L PEG, pH 7.0, after 2 h).

phenoxyphenol, were identified; they were probably formed in accor- dance to the mechanism proposed by Huang et al. [43] and Du et al. [53]. The SPME-GC/MS method revealed the presence of a dozen of volatiles in the gaseous phase that was in equilibrium with the water samples in a form of phenol derivatives such as 2-methoxy-5-methylphe- nol; 2,4-bis(1,1-dimethylethyl)-phenol; and 2-(3-hydrox- y-3-methyl-1-butenyl)-phenol (Table S2).

4.Conclusion

The preparation of covalently immobilized HRP on BC via glutaral- dehyde as the crosslinking agent and its application for the removal of phenol from aqueous solution were investigated. The obtained results show that binding of the enzyme onto BC improves its characteristics, increases its stability over time and allows its reusability in a larger number of cycles.
The application of peroxidase immobilized onto BC has shown high efficiency in the removal of phenols from model wastewater, signifi- cantly reducing the toxicity of water after treatment compared to the quality of the initial water. The study also reveals that the Allium test is a useful and reliable tool for testing the general toxicity and the quality of wastewater that could be used before these effluents are eventually released into the environment.

CRediT authorship contribution statement

Mirjana Petronijevi´c: Conceptualization, Investigation, Writing – original draft. Sanja Pani´c: Supervision. Saˇsa Savi´c: Software, Visual- ization. Jasmina Agbaba: Supervision. Jelena Molnar Jazi´c: Valida- tion, Visualization. Marija Milanovi´c: Software, Visualization. Nataˇsa Đuriˇsi´c-Mladenovi´c: Writing – review & editing.

Declaration of Competing Interest
The authors report no declarations of interest. Acknowledgements
This work was supported by the Ministry of Education, Science and Technological Development of Republic of Serbia (Project OI 172059, grant No. 451-03-68/2020-14/200134, 451-03-9/2021-14/200134, 451-03-9/2021-14/200133 and 451-03-9/2021-14/ 200125).
The authors wish to thanks Dr. Igor Anti´c, University of Novi Sad, Faculty of Technology Novi Sad, for performing the SPME-GC/MS analysis.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.colsurfb.2021.112038.
References
[1]D.M. Naguib, N.M. Badawy, J. Environ. Chem. Eng. 8 (2020), 103592.
[2]A. Hussain, S. Ali, M. Rizwan, M.Z. ur Rehman, M.R. Javed, M. Imran, S.A. S. Chatha, R. Nazir, Environ. Pollut. 242 (2018) 1518.
[3]N. Singh, J. Singh, Prep. Biochem. Biotechnol. 32 (2002) 127.
[4]S. Savi´c, S. Stojmenovi´c, M. Petronijevi´c, ˇZ. Petronijevi´c, Appl. Biochem. Microbiol. 50 (2014) 214.
[5]X. Fan, J. Xu, M. Lavoie, W.J.G.M. Peijnenburg, Y. Zhu, T. Lu, Z. Fu, T. Zhu, H. Qian, Environ. Pollut. 233 (2018) 633.
[6]M. Rizwan, S. Ali, M.Z. ur Rehman, M. Adrees, M. Arshad, M.F. Qayyum, L. Ali, A. Hussain, S.A.S. Chatha, M. Imran, Environ. Pollut. 248 (2019) 358.
[7]T.Md. Souza Bezerra, J.C. Bassan, V.T. de Oliveira Santos, A. Ferraz, R. Monti, Process Biochem. 50 (2015) 417.
[8]M. Naghdi, M. Taheran, S.K. Brar, A. Kermanshahi-Pour, M. Verma, R. Y. Surampalli, Sci. Total Environ. 584–585 (2017) 393.
[9]I. Alemzadeh, S. Nejati, J. Hazard. Mater. 166 (2009) 1082.
[10]Y.C. Lai, S.C. Lin, Process Biochem. 40 (2005) 1167.
[11]M. Besharati Vineh, A.A. Saboury, A.A. Poostchi, A.M. Rashidi, K. Parivar, Int. J. Biol. Macromol. 106 (2018) 1314.
[12]Q. Chen, H. Man, L. Zhu, Z. Guo, X. Wangc, J. Tu, G. Jin, J. Lou, L. Zhang, L. Ci, Catal. Commun. 134 (2020), 105847.
[13]F. Shakerian, J. Zhao, S.-P. Li, Chemosphere 239 (2020), 124716.
[14]Y. Yang, S. Ye, C. Zhang, G. Zeng, X. Tan, B. Song, P. Zhang, H. Yang, M. Li, Q. Chen, J. Hazard. Mater. 404 (2021), 124052.
[15]ASTM D1762-84, Standard Test Method for Chemical Analysis of Wood Charcoal, ASTM International, 2013, https://doi.org/10.1520/D1762.
[16]ASTM D3172-13, Standard Practice for Proximate Analysis of Coal and Coke, ASTM International, 2013, https://doi.org/10.1520/D3172.
[17]L. Zhao, X. Cao, O. Maˇsek, A. Zimmerman, J. Hazard. Mater. 256-257 (2013) 1.
[18]EBC, European Biochar Certificate – Guidelines for a Sustainable Production of Biochar, Arbaz, Switzerland, 2012, https://doi.org/10.13140/RG.2.1.4658.7043. Version 6.1 of 19th June 2015, http://www.europeanbiochar.org/en/download.
[19]J. Marshall, R. Muhlack, B.J. Morton, L. Dunnigan, D. Chittleborough, C.W. Kwong, Soils Soil Syst. 3 (2019) 27.
[20]E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373.
[21]S. Lowell, J.E. Shields, M.A. Thomas, M. Thommes, Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density, Kluwer Academic Publishers, Dordrecht/Boston/London, 2004.
[22]K. Worthington, V. Worthington, Worthington Enzyme Manual, Biochemical Corporation, Worthington, 2011.
[23]C. Nicole, J.K. Bewtra, N. Biswas, K.E. Taylor, Water Res. 33 (1999) 3012.
[24]A.E. Greenberg, L.S. Clesceri, A.D. Eaton, Standard Methods for the Examination of Water and Wastewater, 18th edn, APHA.AWWA.WEF, Washington, 1992.
[25]J. Rank, M.H. Nielsen, Hereditas 118 (1993) 49.
[26]T. Apostolovi´c, J. Triˇckovi´c, M. Kragulj Isakovski, B. Jovi´c, S. Maleti´c, A. Tubi´ J. Agbaba, J. Environ. Sci. 98 (2020) 134.
[27]US EPA, Method PC-97, NCASI, West Coast Regional Center Organic Analytical Program, 1997.
[28]E.N. Yargicoglu, B.Y. Sadasivam, K.R. Reddy, K. Spok, Waste Manage 36 (2015) 256.

[29]R.C. Pereira, J. Kaal, M.C. Arbestain, R.P. Lorenzo, W. Aitkenhead, M. Hedley, F. Macías, J. Hindmarsh, J.A. Maci´a-Agull´o, Org. Geochem. 42 (2011) 1331.
[30]P.J. Mitchell, T.S.L. Dalley, R.J. Helleur, J. Anal. Appl. Pyrolysis 99 (2013) 71.
[31]R. Pashley, M. Karaman, Applied Colloid and Surface Chemistry, John Wiley &
Sons Ltd., 2004.
[32]S.S. Wong, E. Joselevich, A.T. Woolley, C.L. Cheung, C.M. Lieber, Nature 394 (1998) 52.
[33]M. Naghdi, M. Taheran, S.K. Brar, A. Kermanshahi-pour, M. Verma, R. Y. Surampalli, Int. J. Biol. Macromol. 115 (2018) 563.
[34]G. Stavropoulos, P. Samaras, G. Sakellaropoulos, J. Hazard. Mater. 151 (2008) 414.
[35]P.R. Estupi˜nan, L. Giraldo, J.C. Moreno, Rev. Colomb. Quím. 40 (2011) 349.
[36]Ml. Yeboah, X. Li, S. Zhou, Materials 13 (2020) 625.
[37]S. Hashemian, K. Salari, H. Salehifar, Z.A. Yazdi, J. Chem. 2013 (2013) 1.
[38]M. Lawrinenko, D.A. Laird, Green Chem. 17 (2015) 4628.
[39]K.W. Jung, M.J. Hwang, K.H. Ahn, Y.S. Ok, Int. J. Environ. Sci. Technol. 12 (2015) 3363.
[40]A.M. Dehkhoda, N. Ellis, E. Gyenge, J. Appl. Electrochem. 44 (2014) 141.
[41]B. Singh, Y. Fang, C.T. Johnston, Soil Sci. Soc. Am. J. 80 (2016) 613.
PEG300
[42]K. Komnitsas, D. Zaharaki, I. Pyliotis, D. Vamvuka, G. Bartzas, Waste Biomass Valor. 6 (2015) 805.
[43]J. Huang, Q. Chang, Y. Ding, X. Han, H. Tang, Chem. Eng. J. 254 (2014) 434.
[44]N. Misra, V. Kumar, N.K. Goel, L. Varshney, Polymer 55 (23) (2014) 6017.
[45]P.M.B. Chagas, J.A. Torres, M.C. Silva, A.D. Corrˆea, Int. J. Biol. Macromol. 81 (2015) 568.
[46]S.A. Costa, T. Tzanov, A. Paar, M. Gudelj, G.M. Gubitz, A. Cavaco-Paulo, Enzyme Microb. Technol. 28 (2001) 815.
[47]J.M. Guisan, Immobilization of Enzymes and Cells, 2 ed., Humana Press, Totowa, 2006. Chapter 1-98.
[48]J.F. Kennedy, C.A. White, Principles of immobilization of enzymes, in:
Alan Wiseman (Ed.), Handbook of Enzymes Biotechnology, John Whiley &Sons, New York, 1985. Chapter 4.
[49]H. Zhang, A.G. Hay, J. Hazard. Mater. 384 (2020), 121272.
[50]B. Park, K. Ko, D. Yoon, D. Kim, Enzyme Microb. Technol. 51 (2012) 81.
[51]S. Singh, R. Mishra, R.S. Sharma, V. Mishra, J. Hazard. Mater. 334 (2017) 201.
[52]M. Bilal, M. Iqbal, H. Hu, X. Zhang, Biochem. Eng. J. 109 (2016) 153.
[53]P. Du, H. Zhao, H. Li, D. Zhang, C.-H. Huang, M. Deng, C. Liu, H. Cao, Chemosphere 144 (2016) 1674.