ADSORPTION OF ATRAZINE BY BIOCHARS PRODUCED FROM BYPRODUCTS OF THE WOOD INDUSTRY

Atrazine is a herbicide widely used in agriculture; however, because of its toxicity, alternatives are needed to remove this compound from the environment. In this study, we investigated the adsorption process and the atrazine-adsorbing ability of three biochars produced at different pyrolysis temperatures from byproducts of the wood industry. The biochars were characterized by Fourier-transform infrared spectroscopy (FTIR) and scanning electron microscopy (SEM), and their physical characteristics were evaluated by Brunauer-Emmett-Teller (BET) analysis. The biochar pyrolyzed at 350 °C (BM350) adsorbed 26.04% of the herbicide, showing the best performance among the three biochars tested, even though it had the lowest specific surface area (1.467 m g). It also had a greater quantity of organic functional groups, which may have influenced the adsorption. The kinetics of the adsorption process were best explained by the pseudofirst-order model and by the Freundlich isotherm model. However, the biochar is not suitable for the removal of the herbicide atrazine, having a lower adsorption capacity than those of others described in the literature.

In humans, atrazine is capable of interacting with the endocrine and reproductive systems (Sass & Colangelo, 2006), and may alter the protein structure of serum albumin (Zhu et al., 2018). In addition, it is classified as potentially carcinogenic by several studies (Donna et al., 1989;Clapp et al., 2007;Komtchou et al., 2019). In nature, research on rats shows problems in the morphology of the male genital system due to contact with atrazine, such as increased estrogen and decreased progesterone levels (Victor-Costa et al., 2010). In addition, several problems have been described in other animals, such as lobsters (Mac Loughlin et al., 2016), fish (Wirbisky & Freeman, 2017), and amphibians (Gonçalves et al., 2017).
Adsorption is one of the most versatile and effective processes for the removal of contaminants, and when using low-cost adsorbents or industrial wastes, it becomes an economically and environmentally viable alternative (Shaheen et al., 2018). With the increase in the use of plant products by industry, the utilization of byproducts in a sustainable manner has gained more attention (Matos et al., 2013).
The use of byproducts from the wood industry is important for the environment. Some examples include the use of shavings in poultry and pig farms (Sousa et al., 2017), confection of agglomerates with sawdust (Iwakiri, 2005), and the production of biochar, which can be used for the removal of heavy metals and agrochemicals from the environment (Matos et al., 2013).
The properties of the biochar, such as its elemental composition, structure, and chemical stability, depend on the raw material and the pyrolysis temperature. The result can be a porous and fine granular carbon product (Shaheen et al., 2018), with a high specific surface area, which makes Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020 it efficient for the adsorption of pesticides (Liu et al., 2015a;Ren et al., 2016).
Considering the benefits of the use of biochar as an adsorbent, and the potential contamination risk of atrazine, the aim of the present study is to evaluate the capacity of three biochars, produced from byproducts of the wood industry at different pyrolysis temperatures, to adsorb the herbicide atrazine in aqueous medium.

MATERIAL AND METHODS
The biochars used were provided by Embrapa Florestas, located in the Colombo city, state of Paraná, Brazil, made from byproducts of sawmills located in Rio Branco city, state of Acre, Brazil, in the Amazon region, with a pyrolysis temperature of 350 °C (BM350), 450º °C (BM450), and 550 °C (BM550).

Characterization of Biochar
The samples BM350, BM450, and BM550 were ground and sieved to 0.075 mm for the adsorption and other characterization tests.
The three biochars were also characterized by Fourier-transform infrared spectroscopy (FTIR) using the Thermo® Nicolet™ IR 200 FT-IR spectrophotometer and following the methodology proposed by Stevenson (1994), by scanning electron microscopy (SEM) using the Tescan® VEGA3™ equipment, and for surface area, pore diameter, total pore volume, and volume of micropores by the Brunauer-Emmett-Teller (BET) method using Quantachrome® NovaWin™ equipment.

Kinetic Tests
Kinetic adsorption tests were performed in triplicate, in a batch system. In each test, 20 mg of biochars was packed in a 10-mL capped conical tube, and 10 mL of 4 mg L -1 atrazine solution was added. The mixture was shaken at 180 rpm, with the temperature controlled to 22 ± 0.5 °C. The contact times were 3, 6,12,24,48,96,192,317,384,480, and 576 h. The samples were then centrifuged at 3500 rpm for 15 min, and the supernatant was immediately filtered through a 0.45-μm membrane.
The concentration of atrazine in the supernatant was determined by high-performance liquid chromatography using the Chromatograph Shimadzu® Prominence GC-2010, according to the methodology proposed by Zheng et al. (2010) and Liu et al. (2015a). The amount of atrazine adsorbed on the biochars in each test was calculated using [eq. (1)]: Where: Qad is the amount of atrazine adsorbed per gram of biochar (mg g -1 ) at time t (h); V is the suspension volume (L); Ceq is the atrazine equilibrium concentration (mg L -1 ); Ci is the atrazine initial concentration (mg L -1 ), M is the biochar mass (g).
From the results of the kinetic tests, the mechanisms governing the adsorption process were evaluated using the pseudo-first-order model described in [eq. (2)] and the pseudo-second-order model described in [eq. (3)]: Where: K1 is the rate constant of pseudo-first-order adsorption; K2 is the rate constant of pseudo-second-order adsorption; qt is the amount adsorbed in time t (mg g -1 ), qe is the amount adsorbed at equilibrium (mg g -1 ).

Adsorption Isotherm Tests
The adsorption tests for the determination of isotherms were performed in triplicate, in a batch system, at concentrations of 2, 4, 6, 8, and 10 mg L -1 . The tests followed the same methodology as that used in the kinetic study. The time of contact, stipulated by the equilibrium point in the kinetic test, was 317 h for BM450 and BM550 and 480 h for BM350.
Using the kinetic test results, the Langmuir (1918) and Freundlich (1928) isotherm models were applied. The Langmuir isotherm model was calculated using [eq. (4)], which can be linearized in the form of [eq. (5)]: Where: Qad is the amount of atrazine adsorbed on the solid phase (mg g -1 ); KL is the Langmuir adsorption constant; qmax is the maximum cover capacity of the monolayer (mg g -1 ), Ceq is the concentration at equilibrium in the liquid phase (mg L -1 ). The Freundlich isotherm model is described by [eq. (6)], which can be linearized in the form of [eq. (7)]: Where: Ceq is the concentration at equilibrium in the liquid phase (mg L -1 ); Kf is the Freundlich constant referring to adsorption capacity, and n is the Freundlich exponent referring to the efficiency of the adsorption process.
Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020 The validity of the models was interpreted by the linearity of the graphs of C eq Q ad vs. C eq and ln Q ad vs. ln C eq , respectively. Table 1 shows the physical and chemical properties of the biochars studied. Increasing the pyrolysis temperature increased the carbon content, and, according to Shaheen et al. (2018), the higher the carbon content, the lower the yield in grams of the biochar. It is also possible to observe the reduction in the O/C and H/C ratios with increasing temperature, revealing the process of carbon structure defunctionalization. Evaluating the physical characteristics, it was observed that BM350 has a smaller specific surface area than BM450 and BM550. However, the specific surface area is associated with the production temperature, because all organic matter does not tend to degrade at low temperatures (Liu et al., 2015a).

Physical and Chemical Properties of the Biochars
The porosity of the biochars has a strong effect on adsorption, and it appears that most of the pores of the three biochars are in the class of mesopores, but with macropores and micropores also present (Webb & Orr, 1997).

Fourier-Transform Infrared (FTIR) Spectroscopy
Owing to the complex composition of the materials, different intensities of FTIR peaks were detected among the biochars. This variety of functional groups directly influences adsorption behavior (Dumanli & Windle, 2012). Figure 1 shows the FTIR results for the three biochars. All the biochars have a broad band ranging from 3347 to 2991 cm -1 , indicating the presence of hydroxyl groups (OH). This is less pronounced in BM450 and BM550, indicating significant water loss from the materials produced at higher temperatures (Liu et al., 2015a;Liu et al., 2015b). In BM350, peaks are observed in the region between 1545 and 1383 cm -1 , corresponding to aromatic C-C and C=O bond stretching vibrations, conjugated with ketones and quinones (Liu et al., 2015b). There is also a band at 1290 cm -1 , which suggests a group of nitric (NO-) bonding compounds (Yadav, 2005), and two bands in the region from 1207 to 1076 cm -1 , the first peak being associated with C-H bonds in the aromatic structure of the benzene ring (Szymanski, 2013), and the second peak with primary or secondary alcohol C-O bonds (Keller, 1986). One region of great differentiation between BM350 and the other biochars is the presence in the former of a band at 1026 cm -1 , attributed to symmetrical C-O stretching, related to cellulose, hemicellulose, and lignin (Liu et al., 2015b), and bands in the region between 910 and 776 cm -1 , which can be identified with functional groups of alcoholic C-O and with polysaccharide, silicate, and phosphate bonding (He & Ohno, 2012;Liu et al., 2015b). This difference is the result of incomplete combustion, because biochars produced at lower temperatures tend to preserve their functional groups and structures, which are not found in biochars produced at high temperatures (Liu et al., 2015a).
Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020  As shown in Figures 2A and 2D, the structures are not completely defined, owing to the presence of organic matter (Rehraha et al., 2016). In the images of BM450 and BM550, a better organization, with more clearly defined structures, was observed, evidencing different geometric formations and opening of the pores, as shown in Figures  2E and 2F.

Kinetic Study: Determination of the Equilibrium Time of the Adsorption Process
The adsorption of atrazine in biochars occurs in three phases: (1) an instantaneous adsorption phase, (2) a subsequent slow adsorption phase, and (3) the equilibrium phase (Liu et al., 2015a). Figure 3 shows the amount of atrazine adsorbed as a function of time. It was observed that Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020 atrazine was adsorbed more slowly in the first 48 h for BM350, 96 h for BM450, and 317 h for BM550, indicating a decrease in the adsorption capacity over time for the biochars, which may be related to gradual blockage of the micropores and mesopores (Patel et al., 2015), and was more evident in BM450 and BM550, which obtained lower responses.

FIGURE 3. Equilibration of atrazine adsorption, at initial concentrations 4 mg L -1
Several studies have shown that the time to reach equilibrium in the adsorption process is affected by the pyrolysis temperature, because of de-functionalization of the carbon structure as well as the presence or absence of organic functional groups (Zheng et al., 2010;Matos et al., 2013;Liu et al., 2015a).

Kinetic Models of Pseudo-First and Pseudo-Second Order
The pseudo-first-and pseudo-second-order models were evaluated for the three biochars, and the results are presented in Table 2. The results revealed that for the biochar produced at 350 °C (BM350), the kinetics of the adsorption process are best represented by the pseudo-first-order model (R 2 = 0.969), applied to systems whose model of propulsion forces is linear. The adsorption rate is related to the difference in saturation concentration and the number of adsorbent active sites, indicating that the overall adsorption rate is proportional to the initial strength of the process (Febrianto et al., 2009;Campos et al., 2018).
Kinetic studies revealed that the pseudo-secondorder model provided the best fit to the experimental data for BM450 (R 2 = 0.963) and BM550 (R 2 = 0.997), suggesting that control of the velocity mechanism is by chemical adsorption, where the overall adsorption rate is proportional to the square of the initial strength of the process (Febrianto et al., 2009), and the molecules bind to the surface of the adsorbent (Gupta & Suhas, 2009). Thus, the low adsorption observed in these materials can be attributed to the small number of organic functional groups and other compounds present. Table 3 presents the Langmuir and Freundlich parameters calculated for the three biochars in the atrazine adsorption process to determine the most appropriate isotherms. Using the determination coefficients, it is possible to verify that the model that best describes the process of atrazine adsorption in the biochars studied is the Freundlich model, more clearly in BM350 (R 2 = 0.998) than in BM450 (R 2 = 0.987) and BM550 (R 2 = 0.892). According to Sposito (1989) and Yang et al. (2018), this is associated with the presence of organic matter.

Adsorption Isotherms
Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020 Figure 4 shows the equilibrium isotherms for the adsorption of atrazine. It was observed that the amount of atrazine adsorbed was higher in BM350 than in the other biochars. The values of Qad obtained at the end of the process were 1.053 mg g -1 for BM350, 0.321 mg g -1 for BM450, and 0.239 mg g -1 for BM550. Many studies involving biochars and atrazine are described in the literature, and some of the results are summarized in Table 4. The efficiency of the process is related to several characteristics (Ren et al., 2016), including the adsorbent raw material, specific surface area, total pore volume, and the presence of organic functional groups (Zheng et al., 2010;Liu et al., 2015a). For adsorbates, the physical characteristics and chemical properties of the molecule should be considered (Song & Guo, 2012;Ren et al., 2016). Some authors have correlated the adsorption capacity of biochars with high surface area and mean pore diameter (Zheng et al., 2010;Liu et al., 2015a;Wang et al., 2019) However, this correlation does not apply to this study, because the biochar with the highest adsorptive capacity has the smallest surface area, pore diameter, total pore volume, and micropore volume.
Engenharia Agrícola, Jaboticabal, v.40, n.6, p.769-776, nov./dec. 2020 From the difference between the biochars, mainly demonstrated by the peak at 1026 cm -1 (Figure 1), it is estimated that the presence of organic functional groups may influence the adsorption process (Mudhoo & Garg, 2011;Kasozi et al., 2012), helping adsorption by the biochar produced at 350 °C, because the influence of organic functional groups on the adsorption is known (Kim et al., 2018). However, the biochars under study were not as effective in adsorbing atrazine as those described in Table 4.

CONCLUSIONS
Among the biochars produced at the three pyrolysis temperatures studied, the one produced at 350 °C adsorbed 26.04% atrazine, and was more efficient than those produced at 450 °C and 550 °C due to the influence of organic functional groups on adsorption, explained by the pseudo-first-order model and the Freundlich isotherm model. The adsorptive capacity of the biochars studied was not considered efficient for adsorbing atrazine when compared to other biochars already described in the literature.

ACKNOWLEDGMENTS
To Embrapa Florestas and Dr. Claudia Maria Branco de Freitas Maia for the biochars, and for elementary and chemical analyses. To CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior e Tecnológico) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for financial assistance, and UFPR (Universidade Federal do Paraná -Campus Palotina) for the use of equipment for the characterization of biochars.