Sustainable Hydrochar from Orange Peel and Bagasse: A Wet Pyrolysis Approach for Efficient Fe²⁺ and Mn²⁺ Removal from Water Using a Factorial Design

Abstract

Water pollution is a global concern, especially due to iron and manganese, which, at high concentrations, affect water quality by altering taste, odor, and color. This work explores the sustainable synthesis of hydrochar from orange peel and bagasse using hydrothermal carbonization (HTC) and a 2³ factorial design to optimize Fe²⁺ and Mn²⁺ removal for water treatment polishing. HTC was performed by varying (1) temperature (100–200 °C), (2) residence time (8–14 h), and (3) activation agent (H₃PO₄ or NaOH), with a central point at 150 °C for 11 h without activation. Characterization was performed using FTIR, TGA, SEM, nitrogen adsorption (BET) for surface area determination, elemental analysis, Brønsted acidity measurements, and zeta potential analysis. The hydrochar synthesized at 100 °C for 14 h with NaOH (HC6) showed the best Fe²⁺ and Mn²⁺ removal performance. The equilibrium time was 400 min, with pseudo-first-order kinetics best fitting the Fe²⁺ adsorption data, while pseudo-second-order kinetics provided the best fit for Mn²⁺ adsorption. The adsorption process was best described by the Freundlich and Langmuir isotherms, with maximum adsorption capacities (q_max) of 21.44 and 33.67 mg g⁻¹ for Fe²⁺ and Mn²⁺, respectively. It can be concluded that HTC-derived hydrochars offer a sustainable and efficient solution for Fe²⁺ and Mn²⁺ removal. This strategy presents a potentially valuable approach for sustainable water treatment, offering advantages for industrial application by operating at lower temperatures and eliminating the need for biomass drying, thereby reducing energy consumption and environmental impact.

1. Introduction

Several treatment processes have been proposed for the removal of metals from aqueous systems, with adsorption standing out as a low-cost and environmentally friendly technology [1]. This method has garnered significant interest due to its high removal efficiency, cost effectiveness, and operational simplicity [2]. Among the materials used as adsorbents for iron and manganese, biochars emerge as a promising alternative.

Biochar is a carbon-rich material produced through the thermal decomposition of organic matter under limited or no oxygen conditions. Its structure features a graphene-like core and various functional groups on its surface, such as carboxylic acids, phenolic groups, carbonyl groups, and quinones, among others. Biochars, characterized by their high surface area and hierarchical porous structure comprising micropores, mesopores, and macropores, offer excellent potential for adsorption applications [3]. They exhibit rapid adsorption kinetics and high efficiency in removing metals from aqueous systems, making them highly promising adsorbents [4].

Among the various biochar production techniques, hydrothermal carbonization (HTC) has gained attention as a simple and promising method. The biochar produced through this process is known as hydrochar. HTC, a process known as wet pyrolysis, is a thermal conversion process conducted in an aqueous reaction medium, typically at temperatures ranging from 180 to 250 °C, under subcritical water conditions and autogenous pressure. Under subcritical conditions, water acts as a solvent due to changes in its properties, facilitating the decomposition of hemicellulose, cellulose, and lignin [5].

Considering this context, efforts are being made to enhance the production engineering of biochars with tailored properties, utilizing residues from agribusiness, one of Brazil’s key economic sectors. This becomes crucial considering the constant increase in agricultural production in Brazil, making the reuse of by-products an indispensable practice. Currently, various types of biomasses and their derivatives are used in the adsorption of toxic metals, such as orange [6], banana [7], melon [8], and rice straw [9].

Among agro-industrial residues, the orange juice sector stands out due to the large volume of waste generated, with pulp and peel accounting for about 50% of the fruit’s weight. This by-product receives significant attention due to its widespread global production and consumption. Brazil is one of the largest producers and exporters of orange juice in the world, reaching an annual production of 16 million tons [10]. Given this, valorizing agro-industrial residues to produce value-added materials presents a promising option, supporting the Sustainable Development Goals (SDGs) and fostering the circular economy [11].

The pollution of water resources due to the indiscriminate disposal of heavy metals has raised global concern in recent decades. Metals that are highly toxic to both humans and the environment include chromium, copper, lead, mercury, manganese, cadmium, nickel, zinc, and iron [12]. The presence of high concentrations of iron and manganese in water causes economic and technological issues. Upon contact with air, these metals precipitate as a dark sludge, promoting the growth of ferruginous and manganese-oxidizing bacteria on pipe walls, which in turn accelerates corrosion [13]. The World Health Organization (WHO) has established guideline values of 0.3 mg L⁻¹ for iron and 0.05 mg L⁻¹ for manganese in drinking water [14].

Based on the above, this work aimed to optimize the synthesis of hydrochar from orange pulp and peel through hydrothermal carbonization for the removal of iron and manganese from aqueous systems. To achieve this, a 2³ factorial design with a central point was applied, evaluating three variables: (1) temperature, (2) residence time, and (3) activating agent. These parameters were selected due to their critical impact on the physicochemical characteristics of the resulting hydrochars. Temperature and residence time influence the degree of carbonization, development of surface area, and the formation of oxygen-containing functional groups, features that are essential for effective metal adsorption. The inclusion of different activating agents (acid, base, or none) aimed to explore the role of chemical treatment in enhancing surface chemistry and porosity. In addition, the synthesized hydrochars were comprehensively characterized in terms of their physicochemical, textural, morphological, and structural properties. Their adsorption performance toward Fe²⁺ and Mn²⁺ ions was then evaluated, highlighting their potential for application in water treatment and environmentally sustainable remediation technologies.

2. Materials and Methods

2.1. Standards and Reagents

Sodium hydroxide (CAS 1310-73-2, 99%) was purchased from NEON, phosphoric acid (CAS 7664-38-2, 85%) from ALPHATEC, and hydrochloric acid (CAS 7647-01-0, 37%) from VETEC. Additionally, ammonium ferrous sulfate (CAS 7783-85-9, 98.5–101.5%), and manganese sulfate (CAS 10034-96-5, 98.0–101.0%), were also used, both purchased from Exôdo Cientifica. The iron and manganese standard solutions were purchased from SpecSol, Sao Paulo, Brazil, with a concentration of 9563 and 9539 mg kg⁻¹, respectively.

All solutions were prepared using type II water and stored at room temperature, protected from light.

2.2. Biomass Acquisition and Preparation

The raw material used in this work consisted of orange peel and pulp (Citrus sinensis L. Osbeck), which were collected from snack bars and restaurants in the city of Viçosa, Minas Gerais, Brazil. The material was thoroughly washed with running water, dried in an oven at 105 °C for 24 h, and then ground and sieved through a 32-mesh screen.

2.3. Hydrochar Synthesis

A total of 6.25 g of pretreated biomass was combined with 50 mL of the activating agent solution inside the 100 mL Teflon-lined stainless-steel autoclave to perform hydrothermal carbonization. The autoclave was heated to a specified temperature and maintained for a predetermined duration, as outlined in the following section. After carbonization, the product was washed with distilled water until a neutral pH was achieved, centrifuged at 4000 rpm for 20 min, and then dried at 105 °C for 24 h. After carbonization, the hydrochar was stored in a desiccator at room temperature, and its yield was subsequently calculated using Equation (1).

$$\mathrm{Y}\left(\%\right)={\displaystyle \frac{\mathrm{Hydrochar} \mathrm{mass}}{\mathrm{Biomass} \mathrm{mass}}}\ast 100$$

(1)

2.4. Experimental Design

The hydrochar synthesis parameters were determined using a factorial design with a central point. Three variables were evaluated: (1) temperature, (2) residence time, and (3) activating agent type, as described in

Table 1. Experimental design for the synthesis of hydrochars from orange peel and pulp using hydrothermal carbonization.

Assay/(Hydrochar Code)	Temperature (°C)	Residence Time (h)	Activating Agent (0.100 mol L−1)
1 (HC1)	200	14	H3PO4
2 (HC2)	200	14	NaOH
3 (HC3)	200	8	H3PO4
4 (HC4)	200	8	NaOH
5 (HC5)	100	14	H3PO4
6 (HC6)	100	14	NaOH
7 (HC7)	100	8	H3PO4
8 (HC8)	100	8	NaOH
9 (HC9)	150	11	H2O
10 (HC10)	150	11	H2O

. The removal efficiency of manganese and iron from water was used as the response variable, and the data were processed using Statistica^® software (version 7).

2.5. Thermal Activation of Hydrochars

A screening was conducted using the hydrochars described in Table 1 to evaluate their efficiency in removing Fe and Mn. Since hydrochars HC5 (100 °C; 14 h; H₃PO₄), HC6 (100 °C; 14 h; NaOH), and HC9 (150 °C; 11 h; without an activating agent) exhibited good performance for both cations, these materials were thermally activated. The selection was also based on the intention to compare the influence of different activation conditions: acid (HC5), base (HC6), and no chemical activation (HC9). Although HC9 did not show the highest removal efficiency, its inclusion provides an important reference to assess the effect of chemical activation on the adsorptive properties of the hydrochars. The biomass (BM) and hydrochars HC5, HC6, and HC9 were subjected to thermal treatment following the procedure outlined by Fernandez et al., (2015) [15]. These new materials were designated as BMA, HCA5, HCA6, and HCA9, respectively. For this, 1.000 g of sample was placed in a covered porcelain crucible and inserted into a muffle furnace at 300 °C for 1 h. After this period, the hydrochars and activated biomass were cooled to room temperature and stored in a desiccator. The stages involved in the synthesis of the biochars were systematically outlined in a process flow diagram (Figure 1).

2.6. Characterization of the Hydrochars

2.6.1. Determination of Brønsted Acid Sites

The Brønsted acid sites were identified using acid–base titration. For this, 100 mg of each hydrochar was added to 10.00 mL of a NaOH solution (0.100 mol L⁻¹) and maintained under constant stirring at 200 rpm for 3 h. After this period, the mixture was centrifuged at 4000 rpm for 10 min, and three 5.00 mL aliquots were collected, each placed in separate Erlenmeyer flasks. To each flask, a few drops of 1% (m/v) phenolphthalein were added. The samples were then titrated with HCl solution (0.100 mol L⁻¹). A control NaOH sample, without hydrochar, was titrated for comparison. The assay was performed in duplicate, and the number of acidic groups (mmol g⁻¹) was calculated according to Equation (2).

$$C\left(H^{+}\right)={\displaystyle \frac{\left(V_C-V_T\right)\ast C_T}{m_B}}$$

(2)

where V_C and V_T are, respectively, the volumes (L) of the titrant used to neutralize the control solution and the sample; C_T is the concentration (mol L⁻¹) of the titrant; and m_B is the mass of hydrochar used (g).

2.6.2. Elemental Analysis (CHNS)

An elemental CHNS analysis was performed using a TruSPec Micro elemental (LECO Corporation, St. Joseph County, MI, USA) analyzer. Sulfanilamide was used as the standard. The combustion tube was maintained at 1150 °C, and the reduction tube at 850 °C.

2.6.3. Fourier Transform Infrared (FTIR) Spectroscopy Analysis

Fourier Transform Infrared (FTIR) Spectroscopy was employed to investigate the functional groups present in the biomass and hydrochars. The analysis was conducted using a Varian 660-IR instrument (Varian, Inc.; Palo Alto, CA, USA) equipped with a PIKE GladiATR accessory for total attenuated reflectance and a diamond crystal. The transmittance was evaluated over a wavenumber range of 400 to 4000 cm⁻¹.

2.6.4. X-Ray Diffraction (XRD) Analysis

To determine the crystalline structure of the biomass and hydrochar, a diffraction system was used, specifically the D8-Discover model (Bruker AXS GmbH, Karlsruhe, Germany), employing Cu-Kα radiation (λ = 0.1541 nm) and an angular range of 2θ from 10° to 50°.

2.6.5. Nitrogen Physisorption Analysis

The specific surface area of the biomass (BM), thermally activated biomass (BMA), and hydrochars HC6 and HCA6 were determined using the nitrogen physisorption technique. The samples were previously dried under vacuum at 120 °C for 4 h. The specific surface area of each sample was obtained from the N₂ adsorption–desorption isotherms using the Brunauer, Emmett, and Teller (BET) method, and the pore size distribution was determined using the Barrett–Joyner–Halenda (BJH) method. The analyses were performed using a Nova 600 Series instrument (Anton Paar, Graz, Austria).

2.6.6. Determination of the Point of Zero Charge (pH_PZC)

The zero point of charge (pH_PZC) was determined by adding 0.200 g of the BC to 20.00 mL of NaCl solution (0.100 mol L⁻¹). The initial pH was adjusted to different values (2, 4, 6, 8, and 10) following the procedure of, using 0.100 mol L⁻¹ solutions of HCl or NaOH [16]. The system was stirred for 24 h at 100 rpm and maintained at room temperature (~25 °C). After this period, the mixture was centrifuged (4000 rpm for 10 min), and the pH of the supernatant was measured using a pH meter (Lab 1000, model mPA 210).

2.6.7. Determination of Zeta Potential

The surface charge of the hydrochars was determined through zeta potential measurements. Hydrochar (10 mg) was added to NaCl solutions (0.100 mol L⁻¹), and the pH was adjusted to different values (2, 4, 6, 8, 10, and 12). The system was stirred for 24 h at 200 rpm and maintained at room temperature (~25 °C), followed by ultrasonic treatment for 1 h. The zeta potential was measured in duplicate for each pH. The analyses were carried out using a Litesizer 500 analyzer (Anton Paar, Graz, Austria).

2.6.8. Scanning Electron Microscopy (SEM) Analysis

The hydrochars were analyzed using Scanning Electron Microscopy (SEM) with a JSM-6010LA electron microscope (JEOL Ltd.; Tokyo, Japan), operating at an accelerating voltage of 15 kV. For SEM analysis, the samples were coated with a thin gold layer using the Quorum Q150RS equipment.

2.6.9. Thermogravimetric Analysis (TGA)

The thermogravimetric analysis (TGA) was performed using a DTG-60H (SHIMADZU, Kyoto, Japan) instrument. The hydrochar and biomass samples were heated from 35 to 900 °C at a rate of 10 °C min⁻¹ under a nitrogen atmosphere with a flow rate of 50 mL min⁻¹. Prior to the analysis, the samples were dried in an oven for 24 h at 105 °C.

2.6.10. Immediate Analysis

Immediate analysis was conducted on the biomass (BM), thermally activated biomass (BMA), HC6, and BCA6 to determine the moisture, ash, volatile matter, and fixed carbon contents, following ASTM D1762-84 (2013).

2.7. Adsorption of Fe²⁺ and Mn²⁺ by the Hydrochars

The adsorption assays were conducted using 25.00 mL of either an iron or manganese solution (10 mg L⁻¹) and 50 mg of each hydrochar, stirred at 180 rpm at room temperature (~25 °C) without any pH adjustment. After 24 h, an aliquot was taken from each system and filtered through a 0.45 μm cellulose acetate membrane.

The adsorption efficiency of iron and manganese was monitored by analyzing the residual concentrations of these metals in the solutions after contact with hydrochar. The concentrations of iron and manganese were obtained using Microwave-Induced Plasma–Optical Emission Spectrometry (MIP-OES) with the 4100 MP-AES model (Agilent Technologies, Mulgrave, VIC, Australia). The spectral lines used for metal determination were 259.940 and 371.993 nm for iron and 260.568 and 403.449 nm for manganese.

2.8. Evaluation of Initial pH Effects

The effect of the initial pH on the adsorption capacity was assessed by adjusting the pH values to 2.0, 3.0, 4.0, and 5.0 using NaOH and/or HCl solutions (0.100 mol L⁻¹). The assays were performed in duplicate, with 20.00 mL of an iron or manganese solution (10 mg L⁻¹) and 50 mg of hydrochar. The system was stirred continuously at 180 rpm at room temperature (~25 °C) for 24 h.

2.9. Adsorption Kinetics

The kinetic assays were conducted in duplicate, using a volume of 20.00 mL of iron and manganese solution, with a concentration of 100 mg L⁻¹. Times of 0, 0.15, 0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 12, 24, 36, and 48 h were evaluated in batch mode. In each system, 50 mg of hydrochar (HC6) was added, maintaining stirring at 200 rpm at a controlled temperature (25 °C). At the specified times, an aliquot from each system was taken and filtered through a cellulose acetate membrane with a pore size of 0.45 μm. Adsorption was assessed using MIP-OES.

The kinetic models described by Equations (3)–(5) were fitted to the experimental data to characterize the kinetic behavior of the adsorption process. The pseudo-first-order (PPO) model (Equation (3)), the pseudo-second-order (PSO) model (Equation (4)), and the Elovich model (Equation (5)) were employed.

$${qt=qe(1-e}^{-k_{1.}t})$$

(3)

$$qt={\displaystyle \frac{k_2{qe}^{2}t}{1+k_{2}qet}}$$

(4)

$$qt={\displaystyle \frac{1}{b}}ln\left(1+abt\right)$$

(5)

where qt is the amount of solute adsorbed per unit mass of adsorbent at a given time (μg mg⁻¹), qe is the amount of solute adsorbed per unit mass of adsorbent at equilibrium (μg mg⁻¹), k₁ = is the pseudo-first-order adsorption rate constant (h⁻¹), k₂ is the pseudo-second-order adsorption rate constant (mg μg⁻¹ h⁻¹), the coefficient “a” (mg g⁻¹ min⁻¹) corresponds to the initial adsorption rate, and the coefficient “b” (mg g⁻¹) is the desorption constant, with “t” representing time (h).

The Akaike Information Criterion (AIC) was used to assess the goodness of fit. The model with the lowest AIC value was considered the most suitable for describing the experimental data [17].

2.10. Adsorption Isotherms

Adsorption isothermal assays for Fe²⁺ and Mn²⁺ removal using hydrochar (HC6) were performed. For this, 50 mg of hydrochar was added to 20.00 mL of metal solutions at varying concentrations (50, 75, 100, 125, 150, 175, and 200 mg L⁻¹) under controlled conditions at 25 °C. The system was stirred at 200 rpm for 7 h to ensure equilibrium. Afterward, the solution was filtered through a 0.45 μm cellulose acetate membrane, and metal removal was monitored using MIP-OES.

The Langmuir and Freundlich isotherm models were fitted to the experimental data according to Equations (6) and (7), respectively.

$$q_e={\displaystyle \frac{q_{max}{K}_L{C}_e}{1+K_L{C}_e}}$$

(6)

$$q_e=K_F{Ce}^{1/n}$$

(7)

where ${\mathit{q}}_{\mathit{m}\mathit{a}\mathit{x}}$ is the maximum adsorption capacity (mg g⁻¹), ${\mathit{K}}_{\mathit{L}}$ (L mg⁻¹) is the Langmuir adsorption constant, ${\mathit{K}}_{\mathit{F}}$ (mg g⁻¹) is the Freundlich adsorption constant, and n is the constant related to the intensity of adsorption.

2.11. Evaluation of Potential Interferents in the Removal Efficiency of Fe²⁺ and Mn²⁺

The adsorption assay followed the previously described procedure, with the difference that Fe²⁺ and Mn²⁺ solutions were prepared using tap water with potential interferents and stirred for 7 h. Water quality parameters, including conductivity and redox potential, were analyzed using an conductivity meter, model 86503 (AZ^®, Taichung, Taiwan) and a platinum ORP electrode, model SRR03 (Sensoglass, Barcelona, Spain) connected to a potentiometer, respectively.

2.12. Regeneration and Reuse of Hydrochar

The regeneration of hydrochar HC6 (NaOH at 100 °C, 14 h) after the first adsorption cycle involved an ion desorption step. Following use, the material was centrifuged at 4000 rpm for 10 min and then treated with 10 mL of either hydrochloric acid (0.100 mol L⁻¹) or citric acid (0.100 mol L⁻¹) solution. This treatment was performed under stirring for 1 min and repeated three times. The supernatant was analyzed by MIP-OES to determine the desorbed metal content.

After identifying the most effective desorbing agent, the regenerated hydrochar was reused in successive adsorption cycles for iron or manganese removal (200 mg L⁻¹), with efficiency evaluated over three consecutive cycles.

3. Results

3.1. Physicochemical Characterization of the Hydrochars

Figure 2 shows the different hydrochars produced by the hydrothermal process.

The yellowish appearance of the biomass changes when the HTC carbonization process is applied, resulting in blackened hydrochars, which strongly indicate successful carbonization. Temperatures ≥ 150 °C led to the formation of darker and finer hydrochars, while the hydrochars produced at 100 °C had a flaky texture, very rigid and difficult to grind.

The different hydrochars were used for the removal of Fe²⁺ and Mn²⁺ (Figure S1). HC6 exhibited the best removal performance for both metals. Statistical analysis of the removal data from the 2³ factorial design was performed to assess the variables influencing metal removal by the produced hydrochars (Tables S1 and S2, Figure 2). As shown in Figure 3, temperature had a significant and negative effect on the adsorption of both metals, at a 95% confidence level. These findings suggest that lower synthesis temperatures favor metal adsorption. In contrast, time showed a significant positive effect, indicating that longer synthesis periods enhance metal removal. The activating agent also showed a significant and negative effect, suggesting that activation with sodium hydroxide promotes adsorption. The interactions between temperature × activating agent and time × activating agent also exhibited significant negative effects. Other variables did not show statistical significance.

HC5, HC6, and HC9 were selected for thermal activation based on their good performance in ion removal and the distinct activation conditions applied: acidic, basic, and no chemical activation, respectively. The inclusion of HC9 enables the assessment of the effect of chemical activation on the adsorptive properties. The images of the thermally activated hydrochars are shown in Figure S2. A darker coloration can be observed in the biochars compared to their precursors. This is likely due to the thermal activation process at 300 °C, which probably enhances carbonization.

Despite the expectation that thermal activation could improve adsorptive performance by increasing surface area and porosity, the thermally treated samples (HCA5, HCA6, and HCA9) did not exhibit significantly better removal efficiency. This may be associated with the partial degradation of surface functional groups, such as hydroxyl, carboxyl, and carbonyl, which play a key role in metal ion adsorption. Therefore, while thermal treatment can modify the textural properties, it may also reduce the density of oxygen-containing groups essential for effective complexation with Fe²⁺ and Mn²⁺ ions.

Although the BMA sample showed similar removal capacities, HC6 offers some practical advantages. For instance, the biomass used in HC6 does not require a drying step before processing, which simplifies the preparation and reduces energy input. Moreover, the BMA process requires significantly higher temperatures, leading to increased energy consumption. From this perspective, HC6 can be considered a more energy-efficient and operationally simple alternative.

For the HTC process, the yields were approximately 40%, except at 100 °C and with the use of phosphoric acid, where the yield decreased to around 20% (Table S3). The addition of acids to the hydrothermal carbonization (HTC) process provides H⁺ ions that catalyze the hydrolysis of lignocellulosic materials [18]. However, an excess of H⁺ can intensify dehydration, leading to a complete degradation of the biomass components, forming low-molecular-weight soluble compounds. This process ultimately results in a reduction in hydrochar yield [19]. A similar result was also found by Zhou, Chen, Feng, et al., 2017 [20]. The authors synthesized hydrochars from banana peels using phosphoric acid concentrations ranging from 0 to 50%. They observed a decrease in yield with increasing phosphoric acid concentration.

The properties and yield of hydrochar can be significantly influenced by various HTC process conditions, including temperature, residence time, and biomass type. For instance, watermelon peel residues were carbonized at 190 and 260 °C for 1, 6, and 12 h, yielding between 46 and 95% [21]. In another work, sunflower stalks exposed to 175 °C for 1 h resulted in a yield of 69.4% [22].

The carbon, hydrogen, and oxygen contents are shown in Table S3, while the H/C and O/C ratios are best observed in the Van Krevelen diagram (Figure 4). This diagram offers a clear representation of the chemical transformations occurring during biomass carbonization. Demethylation leads to methane (CH₄) formation, while dehydration results in water (H₂O) production, both contributing to a lower H/C ratio. Meanwhile, decarboxylation releases carbonyl compounds, such as carboxylic acids, reducing the O/C ratio [23].

The biomass is located in a region with higher H/C and O/C values, highlighting its lignocellulosic composition. All hydrochars showed a reduction in the H/C ratio, indicating a decrease in hydrogen, which is characteristic of demethylation processes. Similar results were reported by Erdogan et al., 2015 [24]. The authors synthesized hydrochars from orange pulp using the hydrothermal carbonization process. The reactions were conducted at temperatures ranging from 175 to 260 °C, with reaction times of 30, 60, 90, and 120 min, resulting in a decrease in the H/C ratio. Except for HC6 and HC8, all hydrochars showed a significant reduction in the O/C ratio, indicating oxygen loss during carbonization and a subsequent decrease in functional groups. This leftward and downward shift in the graph is characteristic of materials subjected to thermal processing [25], while HC6 and HC8, produced at 100 °C, retained their oxygenated groups.

Except for HCA5, the thermally activated hydrochars showed a reduction in both O/C and H/C ratios. HC5, HC6, and HC9 exhibited a higher degree of carbonization than their precursors, as expected due to the elevated processing temperature. Furthermore, hydrochars HC1, HC2, HC3, HC4, HC5, HC9, HC10, and HCA6 (HC6 thermally activated) are positioned within the lignite region, a classification of fossil coal.

Regarding pH_PZC (Table S3e, Figure S3), it can be observed that the hydrochars exhibited values between 4.29 and 5.60, except for HC1 (4.00), HCA6 (6.76), HCA9 (4.00), and BMA (6.17). Thermal activation tends to increase the ash content, which explains the higher pH_PCZ values for HCA6 and BMA. Both temperature and ash content are factors that influence the surface alkalinity of the hydrochar [26]. The decrease in pH_PCZ for hydrochars HC1 (4.00) and HC7 (4.23) may be associated with the use of H₃PO₄, which could result from a higher abundance of oxygenated functional groups on their surface [26]. On the other hand, the synthesis of HCA9 (4.00) did not involve the use of an activating agent. Therefore, the variations in pH_PZC values among the hydrochars can be attributed to differences in thermal activation, ash content, and the presence of oxygenated functional groups, highlighting the impact of synthesis conditions on the surface chemistry of the materials.

The hydrochars exhibited a Brønsted acid concentration ranging from 2.52 mmol g⁻¹ (HC7) to 5.94 mmol g⁻¹ (HCA9), with no clear trend linked to the synthesis conditions. The concentration of Brønsted acids in hydrochars can be influenced by several factors, including the type of raw material used and the synthesis conditions, such as temperature, time, and activating agent. Silva and coworkers [27] report that hydrothermal processes typically yield materials with more acidic characteristics than those produced by pyrolysis. Similarly, Fontoura et al. [16]. observed that higher temperatures generally lead to a reduction in oxygenated functional groups.

The infrared spectra (Figure S4) showed characteristic bands specific to the hydrochars. The band at 3341 cm⁻¹ may be associated with the stretching of the hydroxyl group (νOH) due to the presence of water, carboxylic acids, and alcohols in the materials. The bands at 2920 and 2850 cm⁻¹ are attributed to the stretching vibrations of the C-H bonds present in aliphatic compounds [28,29]. A band is observed at 1636 cm⁻¹ that can be attributed to the stretching of the carbonyl bond (νC = O), characteristic of ketones, aldehydes, lactones, and carboxylic groups. The bands at 1028 cm⁻¹ are attributed to the stretching of the C-O bond [30]. This band is notably diminished when the biomass undergoes thermal treatment at 200 °C, regardless of the activating agent. This result is linked to the decrease in the O/C ratio (Table S3), which is attributed to dehydration and decarboxylation reactions. The small shifts in the bands, attributed to the different processes, suggest chemical modifications occurring within the hydrochars. Regarding the thermally treated hydrochars, a decrease in the 1028 cm⁻¹ bands, along with a reduction in the 3341 cm⁻¹ band, is observed. These changes can be attributed to the dehydration and decarboxylation processes [31].

Since hydrochars HC5 (100 °C; 14 h; H₃PO₄), HC6 (100 °C; 14 h; NaOH), and HC8 (100 °C; 8 h; NaOH) demonstrated the best performance in iron and manganese removal, thermogravimetric analysis (TGA) was performed on these samples, as well as on the orange peel and pulp biomass (Figure S5). Although HC7 (100 °C; 8 h; H₃PO₄) showed similar results to HC5 and was prepared under comparable conditions, it was not included in the TGA analysis to avoid redundancy. The results show that the hydrochars exhibit slightly higher thermal stability than the biomass, which is likely attributed to the carbonization process.

The differential thermal analysis (DTG) is shown in Figure S6. It shows that the biomass undergoes a greater mass loss at lower temperatures, which can be attributed to the presence of volatile components and organic materials. This behavior is characteristic of the biomass structure, which includes hemicellulose and cellulose, both of which decompose at relatively low temperatures (between 200 and 300 °C), as well as lignin, which decomposes within the temperature range of 200 to 500 °C [32]. This mass loss occurs due to dehydration processes and the release of CO₂ and CH₄ gases.

HC5 activated with H₃PO₄ for 14 h demonstrated superior thermal stability compared to the other materials, exhibiting less mass loss in the temperature ranges typically associated with the degradation of hemicellulose and cellulose. This may indicate that phosphoric acid imparted increased thermal resistance, possibly due to the formation of cross-links or modification of the polymer chains, resulting in more stable structures [33,34]. HC6, activated with NaOH for 14 h, exhibited degradation at a higher temperature compared to the biomass, closely resembling the behavior of HC8, which was also treated with NaOH for a shorter duration (8 h). This suggests that NaOH treatment enhances thermal stability and contributes to the removal of volatile components, though it is less effective than H₃PO₄.

Considering the TGA results, which showed that HC6 exhibited greater thermal stability compared to the other materials, X-ray diffraction analysis was performed on HC6, HCA6, the biomass, and the thermally activated biomass (Figure S7). All the materials displayed broad peaks in the 2θ range between 10° and 30°, a characteristic feature of hydrochars, indicating their amorphous nature. However, HC6 exhibits a profile more similar to that of the biomass (BM), likely due to the milder carbonization process. These results are consistent with the findings from the elemental analysis and FTIR. Silva et al. [27] reported results similar to those found in the present work, synthesizing hydrochar from malt residues via HTC at 150 °C. The authors observed that the biomass initially exhibited a broad peak, which diminished in intensity following hydrothermal treatment. This reduction indicates a decrease in crystallinity, attributed to depolymerization and the breaking of intermolecular and intramolecular hydrogen bonds within the structural components.

The materials were also analyzed by nitrogen physisorption, and the results are shown in Figure S8. It can be observed that all the biochars exhibited type IV isotherms, with H3-type hysteresis [35]. This phenomenon is typical of materials composed of flat particles and slit-shaped pores. The specific surface area, pore diameter, and pore radius are summarized in Table S4. It can be observed that HC6 exhibited a slightly higher specific surface area (0.5990 m² g⁻¹) compared to HCA6 (0.1313 m² g⁻¹). Both materials showed low specific surface areas, which aligns with the findings in the literature for hydrochars derived from bagasse and orange peel. For instance, Abdelhafez & Li [36] reported an orange peel hydrochar with a low surface area of 0.21 m² g⁻¹ produced via pyrolysis. Generally, a higher surface area enhances adsorption. However, other factors, such as the presence of oxygenated functional groups on the hydrochar surface, must also be considered in the adsorption process.

The materials BM, BMA, HC6, and HCA6 were analyzed for zeta potential (Figure S9). The results show that the surfaces are negatively charged, with the negative potential becoming more pronounced as the pH increases [37]. Hydrochar HC6 was activated with a NaOH solution (100 °C for 14 h) and further thermally activated (HCA6). It is evident that NaOH treatment enhances the negative charge on HC6 by introducing more oxygenated functional groups to the surface, a phenomenon also confirmed by elemental analysis and FTIR. These groups are most probably carboxylate groups. These negatively charged functional groups promote electrostatic attraction and complexation with metal ions, compensating for the relatively low surface area and contributing significantly to the overall adsorption performance.

The biomass exhibited a smooth surface with irregular sizes, as shown in the Scanning Electron Microscopy (SEM) results (Figure 5). Thermal activation did not induce major alterations in the biomass structure. However, a significant difference was observed when compared to the HTC carbonized materials (HC6) and HTC followed by thermal activation (HCA6). HC6 displayed a morphology consisting of stacked plates, while thermal activation led to greater fragmentation of the structure. These changes are likely attributed to the release of volatile gases generated during devolatilization and the decomposition of chemical bonds throughout the carbonization process. Similar observations were reported by [38], who synthesized a hydrochar from orange peel.

Energy-Dispersive Spectroscopy (EDS) analysis (Figure S10) predominantly reveals the presence of carbon (C) and oxygen (O), along with potassium (K), in the BM and BMA samples, which are associated with the natural composition of orange peel and pulp residue. The contents are presented in Table S5; however, it is important to note that SEM/EDS analyses are semi-quantitative, and the results may vary depending on the specific region analyzed in the sample. Sodium (Na) is detected in the HC6 and HCA6 samples and is directly linked to the use of NaOH in the chemical treatment. Additionally, calcium (Ca) is identified in these samples, possibly originating from the mineral composition of the biomass or interactions induced by the chemical treatment. Lastly, the presence of gold (Au) is attributed to the gold coating applied during the analysis.

The immediate analysis results (Table S6) indicate that the biomass (BM) exhibited a moisture content of 10.2%, while HC6 showed a reduced moisture content of 6.1%. This reduction is a consequence of the hydrothermal carbonization process, which leads to the partial loss of the water originally present in the biomass. Additionally, HC6 displayed a lower fixed carbon content than the biomass, likely due to an increase in volatile matter. According to Ribeiro et al. [39], when biomass has a high mineral content, mineral enrichment can surpass the degree of carbonization, as evidenced by the rise in ash content. The ash content increased due to mineral formation and condensation, along with the volatilization of organic matter. However, BMA and HCA6 exhibited lower volatile content than HC6, a result attributed to their processing conditions. Both underwent pyrolysis at 300 °C, leading to a more complete carbonization, as previously discussed [40]. Consequently, these materials displayed a higher fixed carbon content, with HCA6 exhibiting an even greater content than BMA.

The obtained characterizations, particularly the complementary analyses for HC6, which demonstrated superior performance in Fe²⁺ and Mn²⁺ removal from aqueous systems, indicate that its higher efficiency is associated with the preservation of oxygenated groups and specific surface area. Carboxylic, phenolic, and amine groups can serve as coordination sites for Fe²⁺ and Mn²⁺ ions [41].

3.2. Application of Hydrochars in the Removal of Fe²⁺ and Mn²⁺

The pH of the system influences both the surface charge of the adsorbent and the ionization state of the species targeted for adsorption. The effect of pH on the adsorption of Mn²⁺ and Fe²⁺ was assessed within the pH range of 2–5 for hydrochar HC6 (100 °C/14 h/NaOH), as shown in Figure 6. Values above pH 5.0 were not evaluated since under these conditions, Fe²⁺ and Mn²⁺ tend to precipitate as metal hydroxides, which would compromise the accuracy of adsorption measurements [41]. Maximum adsorption of both metals occurred within the pH range of 4 to 5, highlighting the strong influence of initial pH on removal efficiency. In highly acidic conditions (pH 2), adsorption capacity was significantly reduced, with removal rates of approximately 12% for Mn and 8% for Fe. At pH 4.0, Mn removal efficiency reached nearly 96%, while Fe removal was approximately 83%.

These results can be attributed to the competition between H⁺ ions and Mn²⁺/Fe²⁺ ions for the active sites on the material. Similar findings have been reported in studies on manganese removal using tamarind hydrochar [42] and rice straw hydrochar [11]. A slight decrease in Fe ion adsorption was observed as the pH increased from 4 to 5, possibly due to the formation of hydroxy complexes [Fe(OH)(H₂O)₅]⁺, which can hinder the interaction between ferrous ions and the active sites of the hydrochar [43]. The shift in chemical equilibrium, coupled with modifications to the adsorbent’s surface, are key factors contributing to the decrease in adsorption efficiency.

Notably, the point of zero charge (pH_PZC) of HC6 is 5.2, indicating that at pH values below this threshold, the material’s surface predominantly carries positive charges. This suggests the possibility of electrostatic repulsion between the metal ions and the adsorbent. However, despite the predominance of positive charges below pH 5.2, residual negative charges may still be present on the surface. According to the results obtained, these residual charges are sufficient to facilitate metal removal [2].

The adsorption kinetics of Fe and Mn using HC6 were evaluated, and the results are shown in Figure 7. Equilibrium for both metals was achieved within 400 min (less than 7 h).

Kim et al. (2020) synthesized a banana peel hydrochar for Fe and Mn removal, reporting an equilibrium time of 10 h for both metals [44]. Similarly, Abdić and coworkers developed a tangerine peel hydrochar for the removal of eight metals (Cd, Co, Cr, Cu, Mn, Ni, Pb, and Zn), including manganese, and observed that equilibrium was reached in just 20 min [45].

The pseudo-first-order (PFO), pseudo-second-order (PSO), and Elovich kinetic models were applied to the experimental data (Figure 7A and Table S7). The PFO model provided the best fit for Fe²⁺, as indicated by the high coefficient of determination (R² = 0.99977) and the lowest AIC value (58.5773). This suggests that the adsorption process is predominantly governed by physical mechanisms, such as diffusion into the active sites of the adsorbent. A similar trend was observed in the removal of Pb²⁺ using hydrochar derived from coconut fiber, where the PFO model also demonstrated the best fit [46]. On the other hand, the adsorption of Mn²⁺ was best described by the PSO model, which exhibited a higher coefficient of determination (R² = 0.99703) and the lowest AIC value (13.50689). The results suggest that chemisorption may be the predominant mechanism involved in Mn²⁺ adsorption. Comparable results were reported by Surovka and Pertile (2017), who evaluated orange peel as an adsorbent for copper, iron, and manganese in aqueous solutions and found that the PSO model provided the best description of manganese adsorption kinetics [47].

The adsorption isotherms of iron and manganese by HC6 were evaluated (Figure 8), and the Langmuir and Freundlich models were applied to the experimental data (Table S8). Based on the correlation coefficients (R²) and the lowest AIC values, the Freundlich model provided the best fit for iron adsorption (R² = 0.99689). Analysis of the parameter n indicates that both metals exhibit favorable adsorption onto the hydrochar, with a stronger affinity for iron. The Freundlich isotherm is an empirical model that assumes a heterogeneous adsorbent surface with various types of adsorption sites, describing a multilayer adsorption process [48]. Similar findings were reported by Oyedeji and Osinfade [12], who used coconut shells for Fe and Cu removal. For manganese, the Langmuir model provided the best fit (R² = 0.99943). This isotherm describes adsorption as a monolayer process, assuming that molecules occupy specific, equivalent sites on the adsorbent surface without interactions between them [49].

The q_max parameter represents the theoretical maximum adsorption capacity of the adsorbent. Manganese exhibited a q_max of 33.67 mg g⁻¹, while iron showed a q_max of 21.44 mg g⁻¹. This indicates a higher adsorption capacity for manganese, suggesting a greater affinity or availability of adsorption sites for this metal ion. The results obtained in this work were compared with those from other relevant studies, as shown in

Table 2. Comparison of the q_max values for Mn²⁺ and Fe²⁺ obtained from HC6 with data from the literature.

Material	qmax (Mn) mg g−1	qmax (Fe) mg g−1	Reference
Biochar from coconut	75.65	81.89	[13]
Activated carbon from agro-industrial residues	6.66	10.64	[50]
Orange peel hydrochar	15.95	8.35	[47]
Date palm biochar	3.57	Not evaluated	[42]
Activated biochar derived from Colocasia esculenta	Not evaluated	6.19	[43]
Cattle manure biochar	6.65	Not evaluated	[51]
Poultry manure biochar	2.84	Not evaluated
Pecan shell-based activated charcoal	Not evaluated	41.67	[52]
Activated charcoal from Bombax ceiba fruit shell	Not evaluated	105.26	[53]
Biochars from modified sugarcane bagasse	13.68	Not evaluated	[54]
Hydrochar from orange peel and bagasse	33.67	21.44	This work

. As observed in the table, the best performance was achieved using biochar derived from coconut; however, this material was produced via pyrolysis in a furnace under an inert atmosphere at 700 °C for 12 h. The second-best result was reported for Bombax ceiba peel, although its preparation involved multiple time-consuming steps, including several washing and drying stages, impregnation with ZnCl₂, carbonization, and sequential thermal and chemical activation. Altogether, this process is significantly more complex and energy intensive than the one proposed in the present study. These results underscore that the material developed in this study is indeed a promising alternative, offering competitive adsorption performance through a more sustainable and less complex route.

3.3. Application of Hydrochars in the Removal of Fe²⁺ and Mn²⁺ in the Presence of Interferents

The previous experiments were conducted in controlled systems. Therefore, an experiment was carried out using Fe²⁺ and Mn²⁺ solutions prepared with tap water. The water samples used to prepare the solutions were analyzed for redox potential and conductivity (Table S8). It can be observed that tap water exhibited higher conductivity (189 µS/cm) compared to Type 1 water (0.95 µS/cm) due to the presence of various dissolved ions such as chloride, sodium, and sulfate, among others [55]. Furthermore, despite the presence of other ions, the removal efficiency remained nearly unchanged, highlighting the promising potential of HC6 for Fe and Mn adsorption in aqueous systems.

3.4. Regeneration and Reuse of HC62

The regeneration of HC6 was assessed using two extracting solutions, hydrochloric acid and citric acid, with the results presented in Figure S11a. Hydrochloric acid demonstrated superior desorption efficiency for Fe and Mn ions, as indicated by the higher concentrations recovered. Consequently, hydrochloric acid was selected for the regeneration of HC6 to further investigate its reusability. The reuse results are shown in Figure S11b. A decline in adsorption capacity was observed for both ions in the second cycle, which remained stable in the third. This reduction in removal efficiency is likely due to the blockage of hydrochar adsorption sites by hydrochloric acid, hindering the interaction between the material’s surface and the ions [54]. Although the regeneration protocol was limited to three cycles under fixed pH conditions, it provides a preliminary evaluation of the adsorbent’s reusability. Despite this decrease, it is important to note that the experiments were conducted with high Fe and Mn concentrations, which do not necessarily reflect real-world conditions, where lower concentrations could favor the material’s reuse. Similar findings were reported by Silva et al. (2022) in studies on the regeneration and reuse of hydrochar derived from citrus residues via hydrothermal processing [56]. The authors identified hydrochloric acid solution (0.5 mol L⁻¹) as the most effective desorption solution for Cu(II) ions, although a reduction in removal efficiency was observed, from 45 mg g⁻¹ to 9 mg g⁻¹.

4. Conclusions

In this work, the efficiency of various hydrochars derived from orange peel and bagasse was evaluated for the removal of iron and manganese from aqueous systems using a hydrothermal process. Factorial design played a crucial role in identifying the key variables influencing the material’s adsorption properties. The optimal conditions for Fe and Mn ion removal involved a lower temperature and a longer residence time, with NaOH as the activating agent. These parameters facilitated material carbonization while preserving its surface functional groups. Additionally, these conditions are economically more viable due to the lower energy consumption of the process. Among the evaluated hydrochars, HC6 exhibited the highest adsorption performance for Fe and Mn. This material represents a promising and sustainable alternative for water treatment, utilizing organic waste as a raw material. Finally, this work highlights the importance of developing eco-friendly solutions based on organic waste to address environmental challenges, contributing to water resource preservation and the advancement of cleaner technologies. Furthermore, it aligns with the United Nations (UN) Sustainable Development Goals (SDGs).