Metal composition in the biomass influences the surface properties of activated carbons in the adsorption of different dyes Open Access

Dyes have long been used in most industries, including food, textiles, plastics, and cosmetics. They are classified broadly as natural and synthetic dyes, and more specifically as reactive, disperse, acidic, vat, and direct dyes based on the fibers to which they are applied and on the chemical nature of each dye. Though natural dyes are eco-friendly, synthetic dyes have complex aromatic molecular structures, which make them more stable and difficult to biodegrade (Yasin et al. 2007). Some dyes are toxic to certain microorganisms and may cause direct destruction or inhibition of their enzymatic capabilities when released into the environment. Conventional methods for dye removal include coagulation and flocculation, oxidation or ozonation and membrane separation (Ahmad et al. 2015). However, the application of these methods is expensive. Reactive dyes, which are the most dominant class of dyes used in the textile industry, are difficult to remove by coagulation and activated sludge because of their high solubility in water (Ozdemir et al. 2004). Adsorption technique, which is a surface phenomenon, is by far the most versatile, widely used, and presents a practical alternative for removing pollutants (Kandisa et al. 2016; Mohammed et al. 2022). The adsorption probability of dye molecules depends on both the material (chemistry) and the crystallographic orientation (morphology) of the adsorbent. There is an increasing interest in innovative and affordable adsorbents. Common adsorbent materials with much interest include clay, zeolites, metal oxides (Ozdemir et al. 2004; Von-Kiti et al. 2023) and activated carbon (AC) (Rashed 2013).

AC is a microporous inert carbon matrix with a large internal surface typically ranging from 500 to 3,000 m²/g, which is an effective, non-specific adsorbent of a wide variety of organic pollutants (Rashed 2013; Chen et al. 2016). AC is widely used in the extraction or pre-concentration of trace elements because of its ease of use, large internal surface area, high porosity, large adsorption capacity and eco-friendly nature. Obtaining AC from biomass waste or plant residues is notably inexpensive. This is due to its widespread availability, affordability, ease of conversion into highly porous carbon, and potential for regeneration for repeated usage (Mohammed et al. 2022). Also, the characteristics of the AC are dependent upon the raw material used and the method of activation (Alves et al. 2021). Plant tissues possess unique microstructures that are tailored to their function, such as the uptake of nutrients and water (Van Norman 2015). Likewise, environmental conditions are reported to have an effect on the microstructure of plant tissue (Young & Schadel 1990). These unique microstructures provide a template for the formation of AC with tailored morphology (internal cavity size/shape, surface area), and pore volume (micro/meso/macro) and adsorptive capabilities. Studies have shown the effectiveness of AC even in the adsorption of microorganisms (Busscher et al. 2006; Li et al. 2011).

The choice of an activation method used for deriving ACs is dependent on the intended application and the properties of the precursor material (Ahmad et al. 2024). Physical or steam activation is mostly preferred due to its environmental friendliness and ability to reduce the risk of chemical contamination from the use of catalysts, particularly in industrial applications like potable water treatment and pharmaceutical use (Pallarés et al. 2018). Not many studies, however, have been conducted to relate the surface properties of the ACs obtained from different biomass matrices using physical activation and the adsorption capacity for organic adsorbates with different structural properties. A study, comparing the adsorption of methylene blue (MB) and 8-hydroxyquinoline (8-HQ) on zeolites, has been conducted, indicating that the surface characteristics of the adsorbent tune the adsorption mechanism for different dyes (Von-Kiti et al. 2023).

Is the adsorption mechanism entirely influenced by the functionality of the adsorbent, or is it wholly driven by the surface area? This paper investigates the efficacy of steam AC from different agricultural waste sources (palm kernel (PK), coconut husk (CH), and corncob), with varying morphologies of cavities and channels, in the removal of MB, a cationic dye with topological polar surface area (TPSA) of 43.9 Ų (the National Center for Biotechnology Information 2023a, b) and 8-HQ, a relatively smaller zwitterionic dye (TPSA, 33.1 Ų) (the National Center for Biotechnology Information 2023b) from water at short contact times. The findings from this research contribute significantly to the understanding of the adsorption of dyes by biomass ACs for diverse environmental applications.

Proximate analysis was conducted to evaluate the moisture content, volatile matter, fixed carbon, and ash content of the as-produced ACs, using the ASTM International Standard Test Method for Chemical Analysis of Wood Charcoal (ASTM D1762-84 2021).

The elemental composition of the ACs was investigated using an Oxford Twin-X X-ray fluorescence (XRF) spectrophotometer. The elemental composition from Na to Ca was conducted in a helium environment at 5 kV voltage and a current of 600 μA using a Focus 5 detector. The elemental composition from Sc to U was conducted in an air environment at 5 kV voltage and a current of 90 μA using a Focus 5 detector.

2 g of powdered samples was used for this analysis. A Bruker Tensor 27 Fourier transform infrared (FTIR) spectrometer (Bruker Optics, Billerica, MA, USA) equipped with a Golden Gate attenuated total reflectance sample cell was also used to study the bond structure and interactions present between the dyes and the ACs. The spectrum was obtained at a resolution of 4 cm⁻¹ over the range of 500–4,000 cm⁻¹.

Scanning electron microscopy (ZEISS EVO 50 equipment, Oberkochen, Germany) was used to study the surface morphology of the synthesized adsorbents, with images taken at an accelerating voltage of 2 kV.

XRD analysis was conducted using a Bruker D2 Phaser model (Bruker AXS GmbH, Karlsruhe, Germany). The scans were taken from 2θ = 5 to 76.057° with a continuous scan of step size 0.010° from a radiation source of wavelength (λ = 1.5406 Å), and the diffractograms were compared with references in the American Mineralogist Crystal Structure database.

20 g of AC was weighed into a 500 ml conical flask. 200 ml of MB solution with a concentration of 0.37 mol/L was added. The mixture was continuously agitated using a Thermo Scientific MAXQ 4450 mechanical shaker at a constant speed of 150 rpm and a room temperature of 25 °C for 120 min. At 10-min intervals, 5 ml aliquots were drawn from the flasks and filtered through a Whatman No. 1 filter paper and then analysed. The dye concentration in the filtrate was measured using an Agilent Cary 60 UV–vis spectrophotometer at the maximum absorbance wavelength of 610 nm.

A number of studies have observed rapid adsorption of adsorbates within the first 30 min of adsorption experiments (Sekar et al. 2004; Ademiluyi & Ujile 2013; Khawaja et al. 2015; Aminu et al. 2020; Von-Kiti et al. 2023). Aliquots of AC, about 0.2 g, were weighed into a 250 ml conical flask and recorded to 0.1 mg. Six different concentrations of aqueous solutions of the dyes, ranging from 0.02 to 3.0 mol/L for 8-HQ and 0.02–0.42 mol/L for MB, were prepared. 100 ml of the dye solutions at different concentrations was added to the 0.2 g aliquot of AC, labelled, and maintained at a constant temperature of 25 °C while shaking for 30 mins. The mixture was then filtered through a Whatman No. 1 filter paper, and the concentration of the adsorbate was determined using the Agilent Cary 60 UV vis spectrophotometer at a maximum absorbance wavelength of 610 nm for MB and 320 nm for 8-HQ. Adsorption isotherms (Freundlich, Langmuir, and Temkin) were used to study the relationship between the amount of adsorbate and its concentration in equilibrium solution. Earlier studies showed that pH affects the adsorption equilibrium (Wang et al. 2005; Li et al. 2013). The earlier studies determined that increasing the pH of the solution deprotonates acid groups on the carbon surface, which increases the uptake of the cationic form of dye molecules from the solution (Malarvizhi & Ho 2010). Thus, to study the interaction of the different dyes on the surface of the as-produced carbons, the pH of the adsorbent surface was not modified, maintaining at neutral (pH = 7) for 8-HQ adsorption studies, and a pH range of 3.5–7.0 which reflected the pH for the serially diluted concentration range, 0.42–0.02 mol/L, and 3–0.02 mol/L for MB and 8-HQ, respectively.

Proximate analysis gives the general composition of the biomass in terms of gross components such as MC, VM, ash (Ash), and FC, which are presented in Table 1. During the activation process, molecules that saturate the carbon surface area are removed, creating more unsaturated sites. Thus, this activation process is expected to reduce the VM in the biomass, thereby increasing porosity and surface area (Cao et al. 2006).

Generally, there was a significant change in all the proximate parameters considered, comparing the precursor biomass and the derived AC, apart from the ash content of PK shell and CC, where no significant differences were observed between the AC and the precursor biomass (Table 1). Hence, the activation process led to a significant opening of pores in the cell structure, leaving a higher number of carbon atoms to be surface atoms. Also, VM in carbon (gases, low-boiling-point organic compounds) and MC (water) will most likely occupy the uncoordinated carbon atoms, available pores and channels, which are the active sites for adsorption. These active sites are mostly at the surface; thus, reducing the VM content, MC, and increasing the FC content during activation permits access to the adsorptive (active) sites.

Studies have shown that in electronically driven adsorption processes, mineral composition is of importance in determining the sticking probability of an adsorbate (Zhang et al. 2020; Li et al. 2021). In such cases, ash content may have some desirable properties, as the metal core influences electron movement at the interface of a carbon shell where the adsorption takes place. However, excessive ash content could lead to clogging of pores, thus reducing the surface area of the AC. The results (see Table 1) show PKAC had the least ash content, while there is no significant difference between the ash content in CCAC and CHAC.

The elemental composition of the AC with XRF shows CHAC and CCAC had a high Potassium content, 9.24 and 8.22%, respectively, while Silica was the dominant mineral in the PKAC (4.55%). High mineral content in AC increases the adsorption of gaseous adsorbates, while low mineral content AC is efficient in liquid phase adsorption (Nowicki 2016). Table 2 presents the mineral composition of the as-produced ACs. The high Fe content in CHAC (see Table 2) is expected to increase back donation of electrons to the carbon surface, altering the electronic configuration at the surface of the carbon, which can influence adsorbate orientation and sticking probability (Zhang et al. 2020). The high mineral composition to carbon ratio of CHAC is expected to contribute some electronic effect in the adsorption process compared to PKAC and CCAC. The results from the mineral composition analysis by XRF (Table 2) agree with those of the proximate analysis (Table 1), which shows PKAC with the highest FC content and the lowest ash content.

Based on the kinetic studies presented in Figure 5, the adsorption of MB onto ACs indicates that, after 10 min, equilibrium is reached between molecules of the dye in the solution and that on the surface of the adsorbent at a solid/liquid ratio of 1:10. Thus, the observation suggests that the AC has reached saturation and achieved its maximum adsorption capacity. Therefore, a contact time of 30 min for determining the adsorption isotherm was deemed sufficient.

Table 3 presents the surface area of the ACs used in the study and their adsorption capacity for MB and 8-HQ using different isotherms. The results are compared to the adsorption on zeolites from earlier studies (Von-Kiti et al. 2023).

The surface area of the adsorbent plays an important role in the adsorption process. The method of monolayer adsorption of acetic acid on an adsorbent with an assumed homogeneous surface (the Langmuir adsorption isotherm model) has been used by earlier researchers to determine the surface area of the adsorbent (Adewumi 2009; Hasdemir et al. 2022). The surface area determined from this method differs significantly from that determined using BET. The BET method using N₂ (a cross-sectional area of 0.162 nm²) adsorption measures the specific surface area, which includes the outer and inner surfaces because it is able to access smaller pores and surface irregularities), while the acetic acid (cross-sectional area of 0.21 nm²) adsorption method largely measures the outer surface area and may be excluded from smaller pores. Hence, the acetic method can still be a useful tool for recognizing distinctions in terms of outer surface area adsorption of larger molecules. The surface area of the AC produced from PK shells had the highest surface area of 577.52 m²/g, as expressed in Table 3. This corroborates results from the proximate analysis (Table 1), which showed PKAC, was more amenable to having a higher surface area due to favourable parameters, such as having the highest FC content and lower VM, moisture, and ash content.

Iteration of the adsorption processes on the AC using a smaller molecule, 8-HQ (33.1 Ų), with both an electrophile and a nucleophile in its resonance structure, can give further indication about the effect of pore size. The adsorption of 8-HQ on all as-produced ACs followed a Freundlich Isotherm (see Table 3). As expected, the adsorption capacities of 8-HQ on all the as-produced AC were lower than for MB (an electrophilic molecule with a larger TPSA). There was no significant difference between the adsorption capacity of PKAC and CHAC (p-value of 0.4347), but CCAC was significantly lower than PKAC (p = 0.0034) and CHAC (p = 0.0298), as shown in Figure 6(a). Like the observed trend in the adsorption pattern in MB, CCAC had the least adsorption capacity (26.72 mg/g). It further confirmed that though CHAC may have a lower surface area than PKAC and CHAC, the electronic effect in CHAC due to d-orbital metals in the mineral composition of the AC far more than compensates for the loss in its surface area than those of PKAC and CCAC in the adsorption of cationic MB and the zwitterionic 8-HQ at low concentrations (i.e., MB below 0.42 mol/L and 8-HQ below 3.0 mol/L). The lower adsorption capacity of the zwitterionic 8-HQ could be attributed to electrostatic attraction and repulsion at the surface of carbon, influenced by the orientation of the molecule, as it approaches the surface of the carbon.

The surface area of all the ACs produced was not significantly different (p-value of 0.8107) from that of zeolitic materials used in a previous study (Von-Kiti et al. 2023) (see Table 3). From the previous study (Von-Kiti et al. 2023), the adsorption capacity of the zeolites for both MB (76.44–79.43 mg/g) and 8-HQ (32.93–33.61 mg/g) is also not significantly different from that of the ACs 76.11–80.17 mg/g and 26.72–36.40 mg/g for MB and 8-HQ, respectively.

The adsorption profile was fitted to a Freundlich non-linear isotherm model, as shown in Figure 6(b). The fit shows that the error variance and normality assumptions in the linearized regression may not be overly significant.

The sticking probability of an adsorbate to an adsorbent depends on how much of the surface of the adsorbent is uncovered, which is evident by the absence of specific functional groups. This value reduces as the surface sites are filled with the adsorbate. Hence, steric hindrance arising from the orientation of the dye molecule has a huge influence on the number of molecules that could be admitted onto the adsorbent surface (Khalfaoui et al. 2002; Hadi et al. 2016). The orientation of the dye molecule, however, is influenced by both electronic and geometric factors (Oduro et al. 2011). Larger molecules, by their mode of adsorption, could wrap around active sites for adsorption, making it inaccessible, thus limiting the sticking probability. Electronic tuning of the surface of the adsorbent thus permits electrostatic forces, which allows an orientation of the molecule that accommodates more molecules at the surface.

ACs produced from agricultural waste precursors, using a steam activation method, possess unique morphological properties determined by the cell structures of the parent biomass and electronic properties (driven by metal composition in the biomass), which influence the adsorption characteristics of cationic and zwitterionic dyes in aqueous solutions. The outer surface area of the ACs determined using the acetic acid adsorption method was 577.52, 333.75, and 246.32 m²/g for PK shell, CC,, and CH ACs, respectively. Adsorption capacities of as-produced ACs by cationic dyes, MB (72–85 mg/g) were higher than those of zwitterionic dyes, 8-HQ (33 and 36 mg/g). The adsorption process suggests that the initial adsorption on the surface of the carbon is driven by electronic tuning at the interface through back donation of electrons from the d-orbitals of metals like Fe present in the mineral composition of the biomass, which makes surface carbon nucleophilic. This allows for an electrostatic attraction between the carbon and the electrophilic centres on the dye molecules. Consequent adsorption, at higher adsorbate concentrations, permits multilayer adsorptions with geometric factors (macropores and heterogeneity at the surface) becoming preponderant in the adsorption process. Thus, electronic effects are mainly experienced in the initial interaction at the interface of the adsorption processes. Subsequent molecules, approaching the surface during the adsorption process, are shielded from the electronic influence of the AC surface by the molecules closest to the adsorbent surface.

The established adsorption capacity of the ACs towards the cationic dyes, implies their effectiveness in the potential removal of organic molecules from industrial effluents.

E.V.-K. developed the methodology, investigated the project, rendered support in data curation, formal analysis, wrote the original draft. W.O.O. conceptualized the process, developed the methodology, investigated the project, rendered support in formal analysis, wrote the original draft, wrote and reviewed and edited the article. M.E.S investigated the project, rendered support in formal analysis. L.B.O. investigated the project, rendered support in data curation. E.L. investigated the project, rendered support in data curation. B.K.-A. conceptualized the process, contributed to resources, wrote and reviewed, and edited the article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.