Grain Farmers of Ontario (GFO), Guelph, Ontario, Canada provided field contaminated corn grain as a feedstock for this study. The concentration of deoxynivalenol (DON) in the raw feedstock was measured by SGS Canada Inc., Hensall, Ontario using an Enzyme-Linked Immunoassay (ELISA), Diagnostix, EZ-TOX DON (with a detection limit 0.5 ppm) and was found to be between 5 and 7 ppm. The proximate analysis showed the amount of volatile matter (VM), fixed carbon (FC), and ash content based on ASTM D1762. Ultimate analysis was conducted to determine carbon, hydrogen, nitrogen, and oxygen content, using Thermo Flash EA 1112 elemental analyzer (CHNSO) (Table 2).

Slow pyrolysis experiments were carried out in a batch mechanically mixed reactor equipped with a paddle mixer rotating at 30–40 rpm, as shown in Figure 1. The reactor consists of a 316 stainless steel horizontal cylinder 33 cm long, and 20 cm in internal diameter, with a capacity of 8.5 L. An induction furnace (5–100 kW, Superior Induction Company, CA, United States) with an on-off controller was used to heat the reactor. The pressure inside the reactor was measured using a pressure gauge (max 15 psi) connected to a pressure release valve (McMaster-Carr, Cleveland, OH, United States). A stainless-steel shell and tube condenser kept in a water bath filled with tab water and used to collect condensable vapours (bio-oil). Non-condensable gases exiting the condenser would then pass through a cotton filter to collect the residual mist, and eventually be directed to the exhaust line.

To investigate the adsorption potential of bio-char as an alternative to commercial activated carbons derived from fossil fuels, samples of bio-char were physically activated using an SS316 tubular reactor with 5 cm O.D and 100 cm long. A 240 V electric tube furnace (Lindberg/Blue M, Ashville, United States) was used to heat the activation unit. The reactor was equipped with a built-in temperature control system (Figure 2). The bio-char activation experiments were performed in the tube furnace reactor at 900°C using a sweeping CO₂ flowrate of 0.5 L/min, as recommended based on a previous study (Gopu et al., 2018). In each experiment, 50 g of bio-char was loaded into the furnace. Once the furnace reached 900°C, it was maintained at that temperature for holding times of 0.5, 1, 2, or 3 h.

Adsorption studies were performed using solutions of methylene blue (MB), methyl orange (MO), and ibuprofen as model compounds. The concentration of methyl orange and methylene blue solutions utilized in the experiments was 250 mg/L to which 1 g of bio-char, or activated bio-char, or commercial activated carbon (CAC) was added in 100 ml samples. CAC samples (Norit^® SX2, CAC number 7440-44-0) were purchased from Sigma Aldrich, Canada. They were produced from peat, activated with steam and acid washed. However, the ibuprofen concentration was 120 mg/L and the experiments were performed by adding 100 mg of adsorbents to 30 ml samples (Kannan and Sundaram, 2001; Jan et al., 2012).

All the adsorption experiments were carried out at room temperature (25°C) in a stirrer (INTLLAB) with a constant agitation rate for 3 h. The concertation of samples was measured using a UV-VIS spectrophotometer (Thermo Scientific Evolution 220). The amount of MB, MO, and ibuprofen components adsorbed was calculated using Eq. 1: Q = ( C 0 − C t ) V m (1) Where C₀ (mg/L) is the initial concentration of the solution, C_t is the solution concentration after adsorption. V (ml) is the dye solution volume, and m (g) is the mass of adsorbent (Savci and Uysal, 2017).

Table 3 shows that, by activating the bio-char, its structure becomes more stable as the fixed carbon percentages increased from 61% (bio-char produced at 500°C) to 88% (activated bio-char after an activation time of 3 h). The ash content increased after activation due to the concentration effect.

Yields of activated bio-chars were calculated by loading the same amount of bio-chars (50 g) produced at various temperatures of 450, 500, 600, and 650°C and then subtracting the remained weight of bio-char after the activation divided by the original weight. The holding time for all four samples was 3 h, during which the activation conditions were held constant. Since bio-char samples produced at higher temperatures 600 and 650°C contain less volatiles, 7 and 15.4% respectively, no significant weight loss occurred after activation. Therefore, higher yields were observed for activated bio-char using bio-char pyrolyzed at 650°C (ACB-650-3 h) and ACB-600-3 h compared to ACB-500-3 h and ACB-450-3 h (Figure 3). Although the yield of activated bio-char per gram of raw bio-char increased from 48 to 85%, the yield of activated bio-char per gram of biomass, as shown in Figure 3, remained approximately constant (20%) over the experiments at different temperatures.

Table 4 illustrates the elemental analysis for the four activated bio-chars produced at different holding times of 0.5, 1, 2, and 3 h and plain bio-char. The initial bio-char used for all these activation processes was bio-char produced at a pyrolysis temperature of 500°C and similar in all the experiments. The carbon content of the activated bio-chars was 84%, approximately 10% higher compared to its original bio-char (73.2%). Activated carbons with high carbon contents have potential use in catalytic applications. Also, activated bio-char with high nitrogen percentages is practical as a fertilizer in soil (Mohammad-Khah, Ansari). Adding activated bio-char to the soil has been proven to be beneficial. The high number of pores and increased surface area of the activated bio-char are two parameters that make it valuable as a soil amendment by increasing porosity and ability to store minerals, pH, and water retention capabilities (Batista et al., 2018), (Gopu et al., 2018).

The Brunauer-Emmet-Teller (BET) technique is used to examine activated bio-char’s specific surface area with the purpose of explaining dye adsorption on its surface. It also includes the measurement of pore size distribution and volume (Wani et al., 2020). High surface area along with the pore size are two major parameters which significantly contribute in adsorption process. CO₂ is a well-known activator for developing microporosity, and as a result, increasing the surface area of activated bio-char (Gopu et al., 2018). Table 5 indicates that the surface area of the activated bio-char increased significantly from 63 to 419 m²g⁻¹ by increasing the activation time from 0.5 to 3 h. The maximum BET surface area achieved during these experiments approaches that of CAC used as a comparison (595 m²g⁻¹).

The surface morphology of raw corn, bio-char produced at 500°C, and activated bio-chars are shown in the SEM images of Figure 4, respectively. The SEM images show that the raw corn surface is smooth and homogenous with no pores and that the porosity develops as a result of pyrolysis. At 500°C, some pores are detected, however, the presence of volatiles does not allow for the formation of more pores (Figure 4B). Activation affects the number of pores and also the appearance of the cracks (channel structure and honeycomb), increasing the available surface area of the activated bio-char (El-Shorbagy et al., 2021). Figure 4 shows that, by increasing the activation time from 1 h to 3 h, an obvious evolution of pores and surface heterogeneity was observed corresponding to the increase in the BET surface area from 63 to 419 m²g⁻¹ and in the pore volume from 0.04 to 0.23 cm³g^-1, confirmed by the visual observation of the SEM images.

FTIR is utilized for both structural determinations of organic compounds and for bio-char bond characterization (Chen et al., 2016). The FTIR patterns for raw corn and bio-char products are shown in Figure 5. Table 6 shows the wavenumber corresponding to the main peaks and the functional groups attributed to each region. In the corn starch spectrum, O–H, CH₂, and –C–C– stretching vibrations are attributed to 3200–3400, 2850–2900, and 1600–1700 wavenumber (cm⁻¹), respectively. Peaks with a wavenumber of 1300–1450 cm⁻¹ and 850–900 cm⁻¹ in raw corn represent O–H and –C–C–bending vibrations, respectively (Oromiehie et al., 2013). Raw corn displays the highest number of peaks at the single bond and fingerprint region. The intensity of light components (3252 cm⁻¹ bands) with hydroxyl (–OH) and carboxyl (–COOH) groups diminishes with increasing pyrolysis temperature and disappears at higher temperatures. When compared to other samples, raw corn has two peaks with high absorbance intensity of 3252 cm⁻¹, and 1014 cm⁻¹, indicating the presence of a great number of alcohols, hydroxyls, and aliphatic compounds in its structure. As the temperature increases, only complex compounds within the fingerprint region remain. For instance, two peaks of 1581 cm⁻¹ and 868 cm⁻¹ show the aromatic rings, and 1155 cm⁻¹ shows the amine compounds with CN stretch. Figure 5 illustrates that complex products, such as aromatic and aliphatic compounds, can be seen even at high temperatures (900°C). The results are in accordance with other studies by A. R. Oromiehie et al. (Oromiehie et al., 2013) and Kizil et al. (Kizil et al., 2002).

XRD analysis was performed on two samples of bio-char produced from the pyrolysis of corn grains at 600 and 650°C and all the activated bio-char samples with activation time changing from 0.5 to 3 h (Figure 6). The results show two main diffraction peaks at 2ϴ = 22–28° and 2ϴ = 42°, representative of amorphous structure, presence of carbon, and graphite (Chen et al., 2017; Pusceddu et al., 2017). A high peak at 2ϴ = 24° is a representative of carbon (C₁₂ to C₆₀), because at high temperatures the carbon content of bio-char increases (Chen et al., 2016; Assirey and Altamimi, 2021).

The adsorption performance of original bio-chars and of the activated bio-chars produced from the pyrolysis of corn grains was measured using dyes (MB and MO), as model molecules, and ibuprofen, as a real molecule representing a typical pharmaceutical. The results were also compared with the performance of commercial activated carbon (CAC). The main reasons for choosing methylene blue (MB), methyl orange (MO), and ibuprofen (IBU) for adsorption experiments are their UV activity and their different molecular sizes (MB = 2 nm, MO = 1.2 nm, IBU = 1 nm) (Zurutuza et al., 2020; Elgarahy et al., 2021; Wei et al., 2021). Methylene blue is a cationic molecule and its structure consists of six carbon aromatic rings, nitrogen, and sulphur (Santoso et al., 2020). The molecular size of this dye is ∼2 nm (Ma et al., 2012) and its adsorption on the surface of the bio-char reveals the presence of mesopores in the structure of the adsorbent (Foo and Hameed, 2011).

Adsorption tests of methylene blue molecules were performed using a constant bio-char sample mass (1 g). Figures 7, 8 illustrate the comparison between the performance of bio-char produced at 500°C (BC-500) and of four activated bio-chars with holding times of 0.5 (ACB-500-0.5 h), 1 (ACB-500-1 h), 2 (ACB-500-2 h), and 3 (ACB-500-3 h) hours. CO₂ penetration into the internal structure of bio-char during activation leads to widening of the pores obtained during the pyrolysis step (Colomba, 2015). Figure 7 illustrates that the ACB-500-2 h sample adsorbed about 92% of MB after 1 h of contact time, which was higher compared to the other two samples in that group. Among all activated bio-char samples produced with activation times of 0.5, 1, 2, and 3 h, ACB-500-3 h showed the highest adsorption capacity, with a total adsorption (25 mg) of MB after 5 min (Figures 8, 9). This result is in good agreement with the porosity measurements, which show that the BET surface area of activated bio-char (3 h) was higher (419 m²g⁻¹) compared to other activated carbons bio-chars (63, 151, and 253 m²g⁻¹). The performance of ACB-500-3 h was also compared to that of commercial activated carbon (CAC). Figure 8 shows CAC and ACB-500-3 h exhibit almost the same adsorption behavior with only 3 min time difference, which could be due to the higher surface area of CAC (595 m² g⁻¹) compared to that of ACB-500-3 h (419 m²g⁻¹).

Figure 10 shows the adsorption of methyl orange (MO) using original samples of the same type of bio-char and the activated bio-chars used for the experiments described earlier. The initial concentration of the stock solution was 250 mg/L, which is similar to that of the MB experiments. The results show that 83% of MO was adsorbed using 1 g of ACB-500-2 h after 2 h contact time. Comparing the MB adsorption graphs in Figure 7 with Figure 10, the rate of MB adsorption in the first 30 min of the experiments was higher for the same experiment and activated samples.

Figure 11 shows the comparison between the adsorption performance of the ACB-500-3 h sample and of the CAC sample. Total adsorption of MO (25 mg) occurred after 2 min using the CAC. However, the contact time required for the total adsorption increased to 8 min, while using the ACB-500-3 h sample.

Ibuprofen with the molecular formula of C₁₃H₁₈O₂ is an anti-inflammatory drug with widespread global consumption. However, due to its chemical structure, it also spreads and accumulates widely in aquatic environments, creating a serious concern for both humans and the environment (Mestre et al., 2007; Essandoh et al., 2015). There are various chemical, physical, and biochemical processes for removing ibuprofen. The existing processes have two downsides: they are expensive processes, and secondly, they are not sufficiently effective (Ternes et al., 2002). Bio-chars and activated carbons acquired from agricultural by-products offer good alternatives for drinking water treatments because of their strong interactions with organic contaminants in the aqueous phase (Jung et al., 2013).

Similar experiments to those with MB and MO were performed with ibuprofen (IBU), starting with the ACB-500-3 h sample and exploring the adsorption profile of this pharmaceutical. Figure 12 shows that ACB-500-3 h could adsorb 76% of the ibuprofen after 0.5 h and 95% after 1 h. To reduce the adsorption contact time, a bio-char sample was activated with a holding time of 5 h, to determine if this would improve the ibuprofen adsorption profile. After increasing the activation time to 5 h (Figure 13), the adsorption contact time decreased from 3 h to 12 min. The BET surface area of ACB-500-5 h was 551 m²g⁻¹, very similar to that of CAC (595 m²g⁻¹).

There are many parameters that can affect the adsorption process, such as initial dye concentration, solution pH, amount of adsorbent, and temperature (Elgarahy et al., 2021). In this study, the focus was the effect of pH. The effect of pH is different depending on the type of adsorbent. Kannan et al. (Kannan and Sundaram, 2001) discussed the pH effect on methylene blue (MB) adsorption and the results showed increased adsorption rates after increasing the pH for various adsorbents suspensions in water, such as activated carbons from bamboo dust, coconut shells, rice husks, and straw. According to the pK_a value of MB, MO, and IBU, which are 3.8, 3.5, and 4.9, respectively, their molecular charges are different at pH > pK_a (Zhu et al., 2004). The MB molecule is positively charged at pH > 3.8, MO and IBU molecule are both negatively charged at pH > 3.5, and pH > 4.9, respectively (Zhang et al., 2017; Sumalinog et al., 2018; Ocampo-Perez et al., 2019).

Table 7 indicates how the pH of activated bio-char samples increased after increasing the activation time from 0.5 to 3 h, which increased the adsorption rate. The activated carbon characteristics, such as surface area, porosity, pore-volume, pore distribution, and functional groups, significantly affect the adsorption process (Heidarinejad et al., 2020).

Studies (Vargas et al., 2011; Ma et al., 2012) show that, apart from the parameters discussed above, the presence of functional groups such as aromatic rings –C=O, –C=N, –C–O–C–, –OH, –NH₂, –C=S, and –S=O on the surface of the adsorbent (activated bio-char) play a significant role in MB adsorption. The FTIR results of ACB-500-3 h illustrate the presence of –C=O, –C=N functional groups, which improved the MB adsorption mechanism.

One reason for the different results between MB and MO, while using dye for both experiments could be related to the size of the molecules (MO: 1.2 nm, MB: 2 nm). Also, the adsorption of MB and MO occurs via electrostatic interactions, electron donor-acceptor relationships, or π-π interactions (Zhu et al., 2004). These forces develop between the functional groups on the surface of the carbon and of the dye molecules and any of them could be responsible for the different dye adsorption behavior between two dyes (Santoso et al., 2020).

According to the global activated carbon market forecast, the market will grow from USD 3.12 billion in 2021 to USD 4.5 billion in 2028 (Report ID FBI102175, 2021). Currently, both coal-based activated carbon and biomass-based activated carbon are available in the market. Table 8 shows a comparison between these two types of activated carbons. Both samples have similar characteristics in terms of BET surface area and iodine number, but the one wood based is cheaper. More importantly, the activated carbon that is made from waste biomass is sustainable and environmentally friendly.