Adsorption of Acid Red 35 (AR35) dye onto Mandarin Biochar-TETA (MBT) derived from Mandarin peels

In this study, a new biochar was produced from mandarin peel residues by dehydration with 50% sulfuric acid followed by decoration with oxidation then reaction with TETA. The effect of the obtained new biochar on the ability to remove AR35 dye from the aqueous solution was investigated. The prepared Mandarin Biochar-TETA (MBT) was characterized by FT-IR, BJH, BET, SEM, DSC, TGA, XRD and EDX analyses. The optimum pH value for AR35 dye adsorption was determined as 1.5. The highest removal percentage of AR35 dye was 97.50% using 300 mg L −1 AR35 dye initial concentration and 2.5 g L −1 MBT dose. The MBT had a maximum adsorption capacity (Q m ) of 476.19 mg g −1 . The data obtained were analyzed with Langmuir, Freundlich, Tempkin, Dubinin-Radushkevich and Jovanovic isotherm models. In addition, the data obtained from these isotherm models were tested using different error functions (hybrid error function (HYBRID), average percent errors (APE), the sum of the absolute errors (EABS), Chi-square error (X 2 ), the root mean square errors (RMS) and Marquardt's percent standard deviation (MPSD)) equations. The Dubinin-Radushkevich isotherm model was best �tted to the experimental data of MBT. Kinetic data were evaluated by pseudo-rst-order (PFO), pseudo-second-order (PSO), elovich, intraparticle diffusion and �lm diffusion models. The adsorption rate was primarily controlled by a pseudo-second-order rate model with a good correlation (R 2 > 0.99). The adsorption mechanism process of AR35 dye by MBT mainly involves the adsorption of anions via the electrostatic attraction forces that develop with the increase in the number of positively charged regions at acidic pH values. The results indicate that MBT is promising for the removal of AR35 dye from water and could be repeatedly used without signi�cant loss of adsorption e�ciency.


Introduction
Concern about water scarcity in some regions around the world is constantly increasing with the continuous pollution of existing waters in different regions.Chemical compounds that cause major pollution in the ecosystem can be listed as dyes ( In particular, dyes are easily detected in wastewater due to their color.Synthetic dyes are the leading dyes that are widely used in paint, leather, textile and different industries (El Nemr 2012a; Lin et al. 2017).
Ecological balance and human health are adversely affected by this pollution as most paints are carcinogenic, toxic and non-biodegradable (Rafatullah et al. 2010; Iqbal 2016).The amount of untreated dyestuffs discharged into water bodies, approximately 10-20%, is estimated to average (0.7-2.0) ×10 5   tonnes per year (Dawood and Sen 2012).
Among the synthetic dyes, azo dyes come rst because of their features such as having the most color variety, being the largest and being versatile.Carcinogenic compounds are formed as a result of excessive use of these chemicals (Rauf and Ashraf 2009).
There are many techniques for the treatment of dyeing wastewater, and the main ones can be listed as chemical oxidation (Karthikeyan et Lin et al. 2016Lin et al. , 2017;;Salama et al. 2015).Removal of dyes by adsorption method using activated carbon is one of the most preferred among these techniques due to its high e ciency (El Nemr 2012c).However, the high production and processing cost of commercial activated carbon has led scientists to seek to synthesize cheaper and more effective adsorbent materials (Abdelwahab et al. 2007;Heibati et al. 2015;Song et al. 2015).For this, the trend towards biochar production as a cheaper and environmentally friendly alternative is increasing day by day.Biochars obtained by using biomass and waste materials as starting materials also prevent the waste of scarce resources.Carbonaceous solid materials obtained by gasi cation or pyrolysis of biomass at temperatures above 350 o C in a nitrogen atmosphere are de ned as biochar (Kołodyńska et al. 2017; Song et al. 2015).Güzel et al. (2017), in their study, found that activities for commercial activated carbon production are generally more expensive than activities for biochar production.In addition to their low cost, biochars also have advantages such as reducing secondary environmental pollution, renewability and creating high value-added adsorbents (Liu et al. 2014).In addition to these, the use of biochar as an adsorbent also reduces the amount of carbon dioxide released into the atmosphere (Ahmad et al. 2014; Abdelhafez and Li 2016).Although biochars have more functional groups on the carbonaceous surface, their surface areas and pore volumes are smaller than activated carbons (Liu et al. 2011; El-Nemr et al.

2022a, b).
In order to improve the practical applications of biochars for the removal of dyestuffs from wastewater, it is possible to further increase the number of functional groups by chemical changes on their surfaces.
Modi cations such as impregnation with minerals, oxidation, nanoscale formation and reduction of the biochar surface are generally efforts to increase the adsorption capacity of biochar (Wang et al. 2019).
Impregnation with mineral elements takes place by adding amino groups to the pores of the adsorbent to increase the functionality of the biochar (Yao et al. 2014).In the surface oxidation method, it is aimed to increase the number of acidic functional groups by using various bases (KOH or NaOH), acids (H 3 PO 4 , HNO 3 or H 2 SO 4 /HNO 3 ) and certain oxidizing reagents (NH Mandarin is one of the temperate climate fruits belonging to a kind of citrus family.According to the data published by the United Nations Food and Agriculture Organization (FAO) in 2021, the annual production of citrus is approximately 80 million tons (FAO 2021).Countries such as China, Turkey, Brazil, Egypt, Spain, Japan, Italy and South Korea are the countries with the highest production.The peels of the mandarins used in fruit juice factories, which are thrown into the environment, constitute approximately 8-14% of their total weight.These peels are mostly used in solid fuel, fertilizer, cosmetics and animal feed industries (Koyuncu et al. 2018).It produces a large amount of fruit peel as biomass waste due to mandarin consumption (Boluda-Aguilar et al.2010).Organic carbon components such as hemicellulose, cellulose and pectin in its structure allow the production of environmentally friendly biochars from tangerine peels by pyrolysis.Thus, materials with an excellent adsorption capacity are obtained (Dhillon et al. 2004).A comprehensive study on the effect of physicochemical properties of biochars obtained from mandarin peels on the removal of dyestuffs in wastewater has not been published yet.No studies have been conducted on the adsorption performance of biochars obtained by rst dehydration with H 2 SO 4 , followed by ozonation in water, and nally, amination with Triethylenetetramine (TETA), using mandarin peel waste materials as a suitable precursor for the removal of Acid Red 35 dye from wastewater.In this study, Mandarin-Biochar-O 3 -TETA (MBT) produced from mandarin peels, which is a low-cost agricultural waste material, by dehydration process was investigated for its e ciency in Acid Red 35 dye removal from aqueous environment.Parameters such as initial adsorbate concentration, solution pH, contact time between adsorbent and adsorbate and the effect of adsorbent dose were investigated as the removal conditions of Acid Red 35 dye from aqueous solution.Adsorption kinetics and isotherms for the removal of Acid Red 35 dye on MBT adsorbent were investigated to determine the structure and maximum adsorption capacity of the adsorption.
2 Materials And Methods

Materials and equipment
Mandarin orange (Citrus reticulata) peels obtained from a local market were used for the production of Mandarin Biochar-O 3 -TETA (MBT).Sulfuric acid (H 2 SO 4 , M.W. = 98.07 g, 99%) and Acid Red 35 (AR35) dye (C 20 H 16 N 4 O 9 S 2 Na 2 ) were supplied from Sigma Aldrich (Fig. 1).AR35 dye standard stock solution was prepared by dissolving one gram of dye in one litter distilled water.
The Ozonator instrument was used to produce O 3 in water from the air and Triethylenetetramine (TETA) was obtained from Sigma Aldrich.Analytic Jena (SPEKOL1300 UV/Visible spectrophotometer with glass cells of 1 cm optical path, a shaker (JSOS-500) and a pH meter JENCO (6173) were used.

Preparation of biochars
Mandarin orange (Citrus reticulata) peels were properly washed many times with water to eliminate dust and then dried for 48 hours at 105°C.Crushed and ground-dried Mandarin peels were used in this recipe.For this experiment, 200 g of crushed mandarin peels were boiled for 3 hours in 1.0 L of 50 percent H 2 SO 4 in a re uxed system, after which they were diluted with distilled water, ltered and washed with water until the ltrate became neutral, and then washed with ethanol and dried overnight at 105°C and weighed to produce 85 g.The processes of carbonization and sulphonation occurred as a result of this method of preparation.The mandarin biochar produced (75 g) as a result of this reaction was suspended in 200 ml water and ozonated for 30 min.The ozonated mandarin biochar was ltered, washed with water and dried overnight at 105°C and weighed to give 80 g.The oxidized mandarin biochar (40 g) was then heated in a 120 mL solution of TETA for 3 hours in a re uxed system.The reaction mixture was ltered and rinsed twice with distilled water, ethanol and dried at 70°C overnight to give 47 g.TETA was added to the product's label, which read Mandarin-Biochar-O 3 -TETA (MBT).

Batch adsorption experimental
A stock solution of AR35 dye (1000 mg L −1 ) was obtained by dissolving 1.0 g of AR35 dye in 1 L of pure water, and this solution was diluted to the desired concentrations for the calibration standard curve and adsorption tests.The adsorption capabilities, thermodynamic and kinetic characteristics of MBT were determined using batch adsorption studies.A series of Erlenmeyer asks (300 mL) was shaken at 200 rpm for a speci ed duration with 100 mL of various concentrations of AR35 dye solution and varying volumes of MBT.With 0.1M HCl or 0.1M NaOH, the pH of the sample was changed to the appropriate levels.Concentration measurement of AR35 dye was performed by taking 1 mL sample from the solution in the Erlenmeyer ask and separating from the adsorbent, at various intervals and equilibrium.The concentration of AR35 dye was measured using spectrophotometry λ max = 505 nm.The equilibrium adsorption capacities (q e ) were estimated using equation (1): Where the adsorption capacity (q t ) (mg adsorbate/g adsorbent) is the adsorbent's ability to remove AR35 dye from a solution at a certain time.C 0 (mg/L) is the initial concentration of AR35 dye; C t (mg/L) is the residual concentration of the AR35 dye after the adsorption process had been completed for a given time.
The following equation can be used to calculate the elimination % of AR35 dye from an aqueous solution (2).
The in uence of pH on AR35 dye adsorption was examined by mixing 0.1 g of the adsorbent with 100 mL of AR35 dye solution of a concentration of 100 mg/L for MBT (0.1 g) with initial pH values varying between 1.5 and 12. 0.1M HCl and 0.1M NaOH solutions were used to modify the pH levels.At 25°C, the suspensions were agitated at 200 rpm for 180 min before being sampled for AR35 dye measurement.For the isotherm investigation, 100 mL of AR35 dye solutions were mixed at 200 rpm for 3 hours at 25°C with varying initial concentrations of AR35 dye solutions (100-400 mg/L) and various amounts of MBT (50 to 250 mg).At 25°C, the effect of adsorbent dosage and contact time on AR35 dye removal was investigated by shaking 100 mL of initial AR35 dye concentration for MBT with varied adsorbent dosages of (50, 100, 150, 200 and 250 mg) at different interval times.

MBT characterization
The adsorption-desorption isotherm of N 2 on MBT was calculated at the boiling point of nitrogen gas.
The surface area and pore analyzer (BELSORP -Mini II, BEL Japan, Inc.) was used to assess the BET surface area (S BET ) of the biochar using nitrogen adsorption at 77 K (Gregg and Sing 1982;Rouquerolet al. 1999).The BET plot was used to calculate surface area (S BET ) (m 2 /g), monolayer volume (V m ) (cm 3   (STP) g −1 ), total pore volume (p/p 0 ) (cm 3 /g), mean pore diameter (nm) and energy constant (C) for the isotherm.The following equation was used to compute the average pore radius (3).
To determine the mesopore surface area (S mes ), micropore surface area (S mi ), mesopore volume (V mes ) and micropore volume (V mic ) of MBCT, Barrett-Joyner-Halenda (BJH) method was used by using BELSORP analysis program software.The BJH method (Barrett et al. 1951) is used to compute the pore size distribution from the desorption isotherm.
The surface morphology of the MBCT sample was investigated using a Scanning Electron Microscope (SEM) (QUANTA 250) in conjunction with an Energy Dispersive X-ray Spectrometer (EDX) for elemental analysis.
The functional groups on the MBT surface were investigated using Fourier Transform Infrared (FTIR) spectroscopy (VERTEX70) and ATR unit model V-100.
Thermal analyzes were carried out using the SDT650-Simultaneous Thermal Analyzer device at a temperature range of 25°C to 1000°C, at a temperature increase rate of 10°C/min.
3 Results And Discussions

The characteristics of MBT
Fourier Transform Infrared Spectroscopy (FT-IR) was used to analyze the generated biochar sample in order to detect changes in the functional groups of the sample.The raw Mandarin peels and MBT FT-IR spectra are shown in Fig. 2. Speci cally, the strong band at 3252.8 cm −1 corresponds to the O-H stretching vibration that existed in Mandarin peels, whereas the broad adsorption peak about 3234.5 cm −1 is indicative of the presence of the -OH group of glucose and the -NH of TETA in the MBT (Fig. 2).The presence of this new band suggested that the amino group had been introduced into the biochar surface as a result of the reaction with TETA.According to this theory, the -CH 2 stretching vibration that existed in Mandarin peels and the MBT existed at 2921.6-2854.1 cm −1 and 299.4-2853.0cm −1 , respectively.There is no adsorption peak about 1710.9 cm −1 , demonstrating that the C=O stretching of the carboxyl group that appeared in Mandarin peels has disappeared in the prepared biochar MBT (Fig. 2).The presence of the band at 1646 cm −1 in Mandarin peels and at 1635.95 cm −1 in MBT indicated the presence of amide groups in both materials.In MBT, the N-H stretching vibration in TETA was observed at a frequency of 1560.4 cm −1 , indicating that TETA modi cation may have increased the N-H functional group of MBT.The adsorption peak at 1419.3 cm −1 indicates the presence of the C-O functional group in Mandarin peels, whereas the strong adsorption peaks at 1440.9 and 1363.4 cm −1 were attributable to the stretching vibration of the -N=C=O group in MBT.The appearance of this new peak on the MBT surface indicates that amino groups were successfully introduced following the treatment with TETA.The band at 1034.6 cm −1 represents that the C-O-H functional group existed in MBT while it is strong in Mandarin peels at 1090.2 cm −1 .Furthermore, it seems there was an obvious difference between Mandarin peels and MBT in the peak strength of 1029-1093 cm −1 , indicating TETA modi cation could affect the C-O-H functional group of MBT (Fig. 2).Also, the OH vibration that appeared at 609.1 cm −1 in Mandarin peels was completely disappeared on the surface of the MBT.
The textural properties were calculated by the (BET) and (BJH) methods including the BET speci c surface area, total pore volume, mean pore diameter, mono layer volume, mesopore area, mesopore volume and mesopore distribution peak for the MBT and are presented in Fig. 3.As shown in Fig. 3, the BET-speci c surface area of MBT (5.65 m 2 /g) and the mono layer volume value of MBT was 1.3419 cm 3   (STP)/g.The total pore volume value of MBCT was 0.0175 cm 3 /g and the mean pore diameter of MBT biochars was 11.74 nm (mesopores).The meso surface area of MBT was 6.18 m 2 /g and the meso pore volume value of MBT was 0.018 cm 3 /g.The meso pore distribution peak value of MBT was 1.22 nm.It was reported that the pores in the prepared modi ed biochar may be blocked by the TETA reaction (El- SEM micrographs of the MBT were examined.Fig. 4 shows the surface morphology of MBT and according to the image, most of the pores and caves had been blocked by the TETA decoration (El-Nemr et al. 2020a, 2020b).
The EDX analysis was carried out for the MBT for its chemical composition.The chemical composition of MBT was reported in Fig. 5, which showed the EDX analysis of MBT proved the presence of 16.05% sample weight for nitrogen element.The major elements in the MBT were carbon (57.24%) followed by oxygen atoms (26.34%) and nitrogen (16.05%).A small amount of sulfur atom (0.37%) was recorded as a result of the dehydration step with H 2 SO 4 .
The thermo gravimetric pro le of the raw materials Mandarin peels and MBT as a function of temperature is shown in Fig. 6.The decomposition of the raw material Mandarin peels occurs in four processes, whereas the decomposition of the MBT occurs in two steps, as shown in Fig. 6.The rst process, which takes place at temperatures between 50 and 150°C, involves the loss of surface-bound water and moisture in the sample, with weight losses of 4.5 and 8.80% for raw material Mandarin peels and MBT, respectively.The second weight-loss stage, with Mandarin, peels losing 25.27% at 150-300°C and MBT losing 45.83% at 150-1000°C, respectively.At 300-385°C, the Mandarin peel loses about 25.04% of its weight in the third weight-loss stage followed by losing 14.23% of its weight in the fourth and nal weight-loss stage between 385 and 1000°C (El-Nemr et al. 2020a, 2020b).
Differential thermal analysis (DTA) can be used solely for identi cation purposes, although it is most commonly employed for phase diagram determination, heat change measurements, and decomposition in various atmospheres (Fig. 6).The DTA curve of the Mandarin-peels sample exhibits three peaks at a temperature of ow T f (80.19, 243.82 and 342.32°C).However, the pyrolysis of the Mandarin peels shows three well-resolved degradation peaks.The DTA analysis of the MBT sample showed manly two wellresolved degradation peaks at a temperature of ow T f (78. ) and removal of these dyes; carried out at room temperature (25±2°C), at an initial AR35 dye concentration of 100 mg L −1 and using 0.5 g•L −1 MBT as adsorbent.The adsorption of AR35 dye was studied for 2 hours at pH values between 1.5 and 12 and pH changes are shown in Fig. 9.For the removal of AR35 dye using MBT, it is seen in Fig. 9a that the highest AR35 dyes removal (97.5%) occurred at pH 1.5.In the study conducted for the removal of AR35 dye, decreasing the pH value continuously from 1.5 to 9 caused the adsorption rate to decrease from 97.5-2.6%.This decrease in the percentage of adsorption removal occurred sharply from pH 1.The presence of excess OH − ions in a high solution pH (alkaline environment) reduces the adsorption e ciency as they compete with the anions of AR35 dye, an anionic dye, for adsorption sites.In addition, MBT adsorbent wants to adsorb high concentration and high mobility OH − ions more than dye anions.
What facilitates the adsorption of anions is the electrostatic attraction forces that develop with the increase in the number of positively charged regions at acidic pH values.The very high adsorption e ciency in the strongly acidic region with a pH of 1.5 can be explained as follows: In fact, there are negatively charged surface regions on MBT and these regions do not support the adsorption of anionic AR35 dye molecules due to electrostatic repulsion.This is due to the hydrophobic nature of biochar.When the MBT adsorbent is immersed in water, hydrogens attach to the surface of the carbon and charge it positively.Therefore, adsorption becomes possible by creating attractive forces between the positively charged MBT and the negatively charged AR35 dye.
Determining the pH value at which the adsorbent surface has net electrical neutrality is possible by determining the zero charge point (pH PZC ) (El Qada et al. 2006).To nd out pHpzc, the experiments were conducted at different pH ranges from 2 to 12, by adding 50 mg of adsorbent to 100 mL of solution and agitated on a magnetic stirrer.The pH of the solution was measured after 25 min.As shown in Fig. 9b, the pH PZC of MBT was determined to be 9.64.At the solution pH > pH PZC , the MBT surface is negatively charged and electrostatic repulsive forces are formed between it and negatively charged anionic dyes.However, in our study, when the pH value increased from 9 to 12, there was a slight increase in AR35 dye removal.

Effect of contact time
Contact time is an important parameter for MBT adsorbent and AR35 dye to provide the necessary interaction.For this reason, the effect of contact time was investigated using MBT at pH 1.5, at an AR35 dye initial concentration ranging from 100 to 400 mg•L −1 .It can be seen from Fig. 10 that the adsorption process takes place very rapidly in the rst 5 minutes, and it gradually increases after this minute.Fig. 10 shows that approximately 17-55% of the adsorption of AR35 dye takes place in the rst 15-30 minutes.
The removal of the AR35 dye increased continuously with the increase of the contact time, and after 3 hours, depending on the initial concentration (100, 150, 200, 300 and 400 mg•L −1 ) of the AR35 dye, the removal was 84%, 67%, 42%, 31% and 25%, respectively.
In the removal of AR35 dye with a low initial concentration on the MBT adsorbent, since the dye concentration of the empty active sites is low, most of these ions will be able to hold on to the MBT adsorbent and there will be high removal.On the other hand, in the removal of AR35 dye with a high initial concentration on MBT adsorbent, the removal percentage will remain low, since the empty active sites cannot adsorb other dyes after they are lled with a certain amount of AR35 dyes.El Nemr et al. (2020c) observed a similar trend in their study on the removal of Acid Yellow 11 dye.

Effect of initial AR35 dye concentration
The initial concentration of the adsorbed substance is an important parameter in the adsorption process to examine the effect of the initial AR35 dye concentration on the adsorption capacity at equilibrium (q e ).
In order to determine the effect of MBT dosage on adsorption capacity at equilibrium (q e ), the adsorbent concentration (0.5, 1.0, 1.5, 2.0 and 2.5 g L −1 ) and the initial AR35 dye concentration (100,150,200,300 and 400 mg•L −1 ) was studied at room temperature (25±2°C) at a pH of 1.5.Fig. 11 shows that the adsorbed amount of AR35 dye at the same initial concentration of AR35 dye increases at equilibrium (q e ) with the decrease in MBT doses.The adsorption capacities at equilibrium (q e ) in AR35 dye removal were determined by using MBT adsorbents at different doses (0.5-2.5 g L −1 ) as shown in Fig.  (100,150,200, 300 and 400 mg•L −1 ), respectively.As can be seen from Fig. 11, the adsorption capacity (q e ) of AR35 dye on MBT at equilibrium is higher in solutions with higher initial AR35 dye concentration.It was observed that it decreased as the adsorbent dose increased.Therefore it indicates that the adsorption of AR35 dye from its aqueous solution was dependent on its initial concentration.Khaled et al. (2009) observed a similar trend in their study on the removal of Direct Yellow 12 dye.
In the adsorption process of AR35 dye on MBT adsorbent, AR35 dye molecules rst encounter the boundary layer effect, then diffuse from the boundary layer lm to the surface of the MBT adsorbent and nally are attached by the porous structure of the adsorbent.

Effect of adsorbent dosage on AR35 dye adsorption
Experimental conditions to examine the effect of adsorbent dosage on AR35 dye removal, initial concentration of AR35 dye (100 -400 mg•L −1 ), MBT adsorbent dosages (0.5 -2.5 g•L −1 ), solution temperature (25±2°C), the adsorption time (180 minutes) and solution pH were adjusted to 1.5, and the results are shown in Fig. 12. Experimental results show that with the increase of MBT adsorbent dosage, the percentage of AR35 dye removal (%) increases slightly (in the range of 97.5-99%) (Fig. 12a,b), whereas the amount of AR35 dye adsorbed at equilibrium (q e ) decreases with the increase of MBT adsorbent dosage (Fig. 12c).The reason for the release in the case where the adsorbent dosage is 0.5 g•L −1 and the initial AR35 dye concentration is 300-400 mg•L −1 is the rapid lling of the active sites on the MBT surface in the presence of highly concentrated dye molecules.As a result, dye removal was limited to around 58-71%.
The amount of AR35 dye adsorbed at equilibrium (q e ) decreases from 195.3 to 39.5, 294.0 to 59.5, 390.2 to 78.0, 428.6 to 119.0 and 465.7 to 158.5, mg•g −1 with increasing of the amount of MBT adsorbent from 0.5 to 2.5 g•L −1 for initial AR35 dye concentrations 100, 150, 200, 300 and 400 mg•L −1 , respectively.It was determined that the maximum removal percentage of AR35 dye and the minimum adsorption amount at equilibrium (q e ) were obtained by using 2.5 g•L −1 MBT dose.

Adsorption isotherms
In order to explain the state of the adsorbate molecules dispersed between the solid-liquid phases, the correlation between the mass of the adsorbent (q e in mg•g −1 ) at the equilibrium time and the adsorbate concentration (C in mg•L −1 ) is used, which is called the adsorption isotherm (El Nemr et al. 2010; Fu et al.

2015)
. Isotherm data are used to determine the optimum amount of adsorbent to be used and to determine the molecular fraction of adsorbate distributed in equilibrium (q e ) between solid-liquid phases.
The values obtained as a result of the adsorption of AR35 dye on MBT adsorbent are shown in Table 1, where the constants of the Langmuir isotherm model are the a nity of the adsorption sites (K L ) and the saturated monolayer adsorption capacity (Q m ).In the removal of AR35 dye, MBT adsorbent with 0.5 g•L −1 dose showed a high correlation coe cient (R 2 ≥ 0.998) in the linear form of the Langmuir model, and the maximum monolayer capacity (Q m ) was calculated as 476.19 mg•g −1 .The intersection point and slope of the C e /q e versus C e plot shown in Fig. 13a gave the 1/Q m K L and 1/Q m values of the Langmuir model, respectively.High correlation coe cient (R 2 ≥ 0.998) and equilibrium adsorption constants (K L ) ranging from 0.06 to 0.84 L•mg −1 can be cited as strong evidence of the adsorption of AR35 dye on MBT.Applicability of AR35 dye for adsorbing on MBT adsorbent appears to be possible according to the Langmuir isotherm model.Therefore, it was concluded that AR35 dye was adsorbed on the MBT adsorbent surface as a single layer.
Another model applied for the removal of AR35 dye by MBT adsorbent is the Freundlich isotherm model.The Freundlich isotherm model was applied to determine how effective MBT adsorbent was in the removal of AR35 dye.Linear tting parameters obtained from the Freundlich isotherm model, which constitutes the adsorption process as a heterogeneous phenomenon, are given in Table 1.The intersection point and slope of the log(q e ) versus log(C e ) plot shown in Fig. 13b give the logK F and 1/n F values of the Freundlich isotherm model, respectively.
The value of K F (L•g −1 ), which represents the binding energy, expresses the amount of AR35 dye adsorbed on the adsorbent for unit equilibrium concentration and is one of the Freundlich constants as a distribution or adsorption coe cient.The higher the K F value, the higher the adsorption capacity of the adsorbent.In addition, a value of 1/n less than 1 means that the adsorbate is easily adsorbed by the adsorbent.Therefore, when 1/n is less than 1, removal of AR35 dye by MBT adsorbent is a physical process.If the 1/n values in Table 1 are examined, it can be concluded that the adsorption of AR35 dye on the MBT adsorbent is appropriate since all values are less than one.In addition, the n F value expresses the degree of nonlinearity between the solution concentration and the adsorption process, and when this value is greater than 1, the adsorption of AR35 dye on the MBT adsorbent takes place physically.
The variation in log(q e ) as a function of log(C e ) successfully de nes the Freundlich isotherm correlation coe cient values (Fig. 13b).The better adsorbability of AR35 dye to MBT adsorbent depends on the high Q m value, which is 350.16 mg•g −1 and belongs to the adsorbent with a 0.5 g•L −1 concentration as given in Table 1.Freundlich correlation coe cient (R 2 ≥ 0.965) for MBT adsorbent was lower than Langmuir's correlation coe cient.
The Temkin Model, which explains the effects of indirect adsorbent/adsorbate interactions on the adsorption process, is another isotherm model and has been applied to experimental data.The Temkin isotherm model takes into account the heat exchange that occurs during the adsorbate's adsorption on the adsorbent surface.As a result of the adsorption process, it is assumed that the heat of adsorption of all molecules in the layer decreases linearly with time.Temkin isotherm parameters (A T and B T ) of the adsorption of AR35 dye by MBT adsorbent are calculated from the linear relationship between q e and lnC e as seen in Fig. 13c.The slope of the graph is used to calculate the equilibrium bonding constant A T (g•L −1 ), and the intercept is used to calculate the B T , which expresses the adsorption heat coe cient.
Table 1 gives the calculated Temkin model constants.The Temkin isotherm correlation coe cient (R 2 ≥ 0.990) obtained in the removal of AR35 dye from the adsorbent with 1.5 g•L −1 of MBT dosage is quite high, and it is concluded that the model is suitable for examining the temperature changes in the adsorption process.The very low heat of sorption caused the removal of the AR35 dye by physisorption and the very weak ionic interaction between the adsorbent and the adsorbate.The adsorbent-adsorbate interaction is related to the heat of adsorption (B T ), and the coating of the AR35 dye on the MBT adsorbent is affected by this heat.If Table 1 is examined, this value increased continuously with the increase in the dosage of MBT from 0.5 to 1.5 g•L −1 , while after this dosage it decreased to 2.5 g•L −1 .
The values obtained for the AR35 dye removal of MBT adsorbent at different dosages in the D-R model range between 0.687 and 0.980, and it seems to be more compatible with the experimental data at most adsorbent dosages compared to the Langmuir isotherm model (Fig. 13d; Table 1).
In addition to the isotherm models mentioned so far, the experimental data is also adapted to the Jovanovic isotherm model.In addition to the assumptions of the Langmuir isotherm model, this model assumes that there may be some mechanical contact between adsorbing and desorbing molecules (Salarirad and Behnamfard 2011).The Jovanovic constants obtained from the C e versus ln(q e ) plot shown in Fig. 13e are summarized in Table 1.The fact that the R 2 values are quite low (except for the 2 g L −1 concentration) shows that it is not appropriate to adapt the experimental data to this model.A comparison of four different isotherm models applied with experimental data is given in Fig. 13f.

Adsorption Kinetic studies
For the kinetic models of the adsorption of AR35 dye on MBT, pseudo-rst-order (PFO) equation, pseudosecond-order (PSO) equation, Elovich equation, Intraparticle diffusion and Film diffusion equations were applied (El Nemr et al. 2010).The correlation coe cients (R 2 ) of the kinetic models given in Tables 3-4 take a value between zero (0) and one (1), and the model's acceptance as a suitable model is directly related to the closeness of the R 2 value to one (1).As seen in Fig. 14a, rate constant, k 1 and equilibrium adsorption capacity (q e ) is calculated from the linear graph of lg(q e −q t ) values against time (t).
The signi cant deviation of the calculated q e values from the experimental values is related to the low R 2 values.Therefore, considering the values in Table 3, it means that the pseudo-rst-order (PFO) kinetic equation is not very suitable for the adsorption of AR35 dye on MBT adsorbent.Table 3 shows that there is no regular increase or decrease in R 2 correlation coe cient values with the increase of MBT adsorbent concentration from 0.5 to 2.5 g•L −1 .
The adsorption of AR35 dye on the MBT adsorbent was also analyzed using the pseudo-second-order (PSO) kinetic equation.As shown in Fig. 14b, it is possible to calculate the pseudo-second-order kinetic constant, k 2 (g mg −1 min −1 ) and the amount of AR35 dye adsorbed at equilibrium (q e ) by plotting t/q e versus t.The PSO kinetic plot of MBT adsorbent for adsorption of AR35 dye is shown in Fig. 14b.In addition, the kinetic constant (k 2 ) values, the experimentally and theoretically estimated q e values and the corresponding correlation coe cient (R 2 ) values of the PSO equation are given in Table 3.When Table 3 is examined, it is seen that the pseudo-second-order model is the model with R 2 values closest to 1. Therefore, the most suitable kinetic model is the pseudo-second-order model.Thus, the q e values calculated using the PSO model plot and the experimental q e values exactly overlap for all initial AR35 dye concentrations studied.
Another kinetic model analyzed in the adsorption of AR35 dye on MBT adsorbent is the Elovich model, and Fig. 14c shows the correlation curve between q t and ln(t).The intersection point and slope of Fig. 14c were applied to the calculation of Elovich constants α and β, respectively, and the obtained values are given in Table 4.When the R 2 values are compared, it can be said that the R 2 values of the Elovich model are higher than the PFO kinetic model values and lower than the PSO kinetic model values (Tables 3 & 4).
According to the results from Tables 2 & 3, it is clear that chemical adsorption can determine the adsorption rate of AR35 dye on MBT adsorbent in some cases.
The intraparticle diffusion model is used to explain the transfer of solute in solid-liquid adsorption.The intraparticle diffusion model identi es and explains all the steps involved in the sorption process.In an adsorption process, the adsorbate is transferred onto the adsorbent in three successive steps: (i) The rst step is concerned with the movement of ions or molecules transported from the solution through the liquid lm to the adsorbent surface.(ii) The second step is that the ions or molecules attached to the adsorbent surface are then diffused into the adsorbent.(iii) The last step is the chemical reaction step that takes place in the active groups of the adsorbent.Each of these three steps takes place at a different rate, and the slowest step is also the step that determines the adsorption rate.
According to the theory put forward by Weber and Morris (1963), if the lines drawn in the graph of q t and root time (t) in Fig. 14d pass through the origin, it is suggested that the adsorption is controlled by the intraparticle diffusion step.On the other hand, in the case where the drawn lines do not pass through the origin (where the C value is large), it is assumed that the rate of the adsorption process is determined by lm diffusion.The Webber-Morris adsorption line of the adsorption of AR35 dye on MBT adsorbent at different adsorbent doses and different initial AR35 dye concentrations is shown in Fig. 14d.The K dif and C values shown in Table 4 were calculated from the slope and intercept point of the plot of qt versus t 0.5 , respectively.The straight lines of all adsorbent concentrations in Fig. 14d do not pass through the origin due to their high C intersection.The reason for this situation can be shown that the rate of adsorption of AR35 dye on MBT adsorbent increases gradually over time, and as a result, this rate is controlled by lm diffusion (Fig. 14e).The decrease in the pore volume and surface area of the MBT adsorbent during the adsorption process is the reason for this situation.

Comparison with results reported in the literature
In the literature review summarized in Table 5, since there is no study on the removal of Acid Red 35 dye the effectiveness of the removal of azo dyes using different adsorbents was compared with the MBT adsorbent and showed that MBT adsorbent was effective in removing AR35 dye.

Regeneration of biochars
Desorption experiments of AR35 dye from the loaded MBT were performed using 0.1M NaOH followed by 0.1N HCl as eluted mediums to investigate the economic feasibility and reusability of MBT as an adsorbent.In this condition.the desorption % decreased with rising regeneration cycles Fig. 15.The regenerated MBT was applied in six consecutive cycles of adsorption/desorption.The adsorption amount presented was consistent through the cycles and experienced the adsorption capacity decreased by 12.7% after six generations.which suggests it may be used as a sustainable AR35 dye removal (Fig. 15).

Conclusion
In this study, it has been shown that mandarin peels, which are agricultural waste, can be used in the production of cheap and effective adsorbent material.MBT, which is prepared to be used in the removal of AR35 dye, an azo dye, was prepared by rst treating dry mandarin peels with 50% H 2 SO 4 at boiling point, then oxidizing with ozone in water, and nally amination with Triethylenetetramine (TETA).The adsorption of AR35 dye was found to be dependent on adsorbent dose, initial concentration, contact time between the adsorbent and adsorbate and the pH.It was determined that the optimum pH value of adsorption of AR35 dye by MBT adsorbent was 1.5.It was observed that the maximum removal of the AR35 dye and the minimum adsorption amount (q e ) in the equilibrium were observed when the 2.5 g L −1 concentration of MBT adsorbent dose was used.In the removal of AR35 dye, the Dubinin-Radushkevich model is a better sorption process than other models.The maximum adsorption capacity calculated by using the Langmuir isotherm is 476.The effect of AR35 dye initial concentration (100-400 mg•L -1 ) using MBT doses (0.5-2.5 g•L -1 ) on q e (mg g −1 ) (Temperature = 25± 2 °C).
5 to pH 4, but slightly from pH 4 to pH 9.With the increase of the pH value from 9 to 12, the adsorption rate increased slightly and reached around 16.2%.Considering the studies on the removal of azo dyes,Khaled et al. (2009) found that the adsorption e ciency decreased from 98.1-11.1% by increasing the pH of the solution from 1.5 to 11.1% in their study on the removal of Direct Yellow 12 dye.In addition,Aboua et al. (2015) reported that the adsorption e ciency decreased from 98-56% by increasing the pH value from 2 to 11 in the removal of Methyl Orange dye.Song et al. (2015) reported that the adsorption capacity decreased sharply by increasing the pH value from 2 to 4 in the removal of Sunset Yellow dye.It was determined that the optimum pH value for the removal of AR35 dye for MBT was 1.5.

19 mg g − 1 .
Considering the adsorption energy values obtained from Dubinin-Radushkevich isotherm models, it was determined that chemical adsorption may take place.As a result of this study, it is possible to use MBT as a cheap and effective potential adsorbent for the removal of AR35 dye from wastewater.Declarations Funding This work was partially funded by the Science and Technology Development Fund (STDF) of Egypt (Projects Nos.CB-4874 and CB-22816).

Figures Figure 1 Figure 2 FTIR
Figures

Figure(
Figure

Figure
Figure

Figure 15 Desorption
Figure 15 (Hassaan et al. 2021;7.41°CusingDSC.Two other phase transitions were reported for the DSC of MBCT at 393.04°C as endothermic and at 728.57 as exothermic phase transition (El-Nemr et al. 2020a, 2020b).Figure8shows the XRD of the MBT.The broad peak in the region of 2θ = 10-30 and 40°-50° is indexed as C (002) and (101) planes diffraction peaks indicating an amorphous carbon structure with randomly oriented aromatic sheets(Yeboah et al. 2020).There is small peak around 2θ = 43.669indicatedthat the adsorbent had an amorphous structure including cellulose, hemicellulose and lignin(Hassaan et al. 2021; 73ermal transitions can be used to compare materials using differential scanning calorimetry (DSC).The DSC analyses of Mandarin peels and MBT are shown in Fig.7.Both of the samples had crystallization temperatures T C below 100°C (88.15 and 92.51°C, respectively), which can be attributed to water molecule crystallization.The other two exothermic phase transitions were reported for the DSC analysis of Mandarin peels at 263.73and 347.82°C.The other crystallization temperature T C of raw Mandarin peels was found Textile industry wastewater has very different pH values from each other.The adsorption process is highly affected by the solution pH, which affects the carboxyl, hydroxyl and amino groups on the biochar surface.Determination of the amount of Acid Red 35 (AR35) dye adsorbed at equilibrium (q e 11.These values range from 195.3 to 465.7, 98.2 to 375.7, 65.6 to 261.8, 49.3 to 197.2 and 39.5 to 158.5 mg g −1 for initial AR35 dye concentrations Dubinin-Radushkevich (D-R) isotherm model is the model in which the equilibrium data are applied to determine whether the adsorption of AR35 dye on the MBT adsorbent occurs chemically or physically.In this model, considering the Polanyi potential theory, it is assumed that the adsorption process continues until the pores are lled.The correlation coe cients and D-R isotherm constants obtained in the adsorption of AR35 dye on the MBT surface at different dosages are given in Table1.The apparent (Chowdhury et al. 2011;Mobasherpour et al. 2012adsorption.It is possible to determine the adsorption type according to the value of the apparent energy (E) (E < 8 kJ mol −1 "physical adsorption", 8 kJ mol −1 < E < 16 kJ mol −1 "ion exchange", E > 16 kJ mol −1 "chemical adsorption")(Chowdhury et al. 2011;Mobasherpour et al. 2012).When the estimated binding energy (E) values in Table1are examined, it is seen that all MBT adsorbent dosages are greater than 16 kJ mol −1 , and in this case, it is concluded that the adsorption of AR35 dye on MBT adsorbent is by chemical adsorption.The correlation coe cient (R 2 )

Table 2 A
is to compare several different error function values.Error functions such as average percent errors (APE), Chi-square error (X 2 ), hybrid error function (HYBRID), Marquardt's percent standard deviation (MPSD), sum of absolute errors (EABS) and root mean square errors (RMS) are the main functions used to determine the error distribution between the equilibrium values and the estimated isotherm models (El Nemr et al. 2010).The comparison of the error functions, which express the similarity between the experimental data of the MBT adsorbent and the values calculated using the theoretical isotherms, is given in Table2.When Table2is examined, it is clear that the most suitable isotherm model belongs to the Temkin model, which has the lowest (APE), (X 2 ), (RMS), (HYBRID), (EABS) and (MPSD) error function values.Therefore, while Langmuir and Dubinin-Radushkevich isotherm models are the most suitable isotherm models in terms of correlation coe cients, it is clear that the Temkin isotherm model is compared in terms of error functions.few error function values of the isotherm models best suited to the experimental equilibrium data in the adsorption of AR35 dye on MBT.
Using Langmuir, Freundlich, Tempkin and Dubinin-Radushkevich isotherm models, correlation coe cients (R 2 ) were compared to determine the most suitable model for the adsorption of AR35 dye to MBT adsorbent, to the experimental equilibrium data.Another way to determine the optimal isotherm model for experimental data

Table 5
Comparison of the maximum adsorption capacities of azo dyes of different adsorbents