Iron- and Nitrogen-Modified Biochar for Nitrate Adsorption from Aqueous Solution

Abstract

Nutrient pollution poses a significant global environmental threat, and addressing this issue remains an ongoing challenge. Biochar has been identified as a potential adsorbent for environmental remediation. However, raw biochar has a low nitrate adsorption capacity; thus, biochar modification is necessary for targeted environmental applications. This work explored and compared the performance of Fe-doped, N-doped, and N-Fe-co-doped biochars from Douglas fir toward nitrate removal from an aqueous solution. A central composite experimental design was used to optimize processing variables, maximizing the surface area and nitrate adsorption capacity. Proximate analysis, elemental composition, gas physisorption, XPS, SEM, TEM, FTIR, and XRD were used to characterize the biochar’s properties. Pyrolysis under NH₃ gas generated more pores in biochar than conventional pyrolysis. Doping biochar with N and Fe improved nitrate adsorption capacity from aqueous solutions. The maximum nitrate adsorption capacity of Fe-N-doped biochar produced at 800 °C was 20.67 mg g⁻¹ in sorption tests at pH 3.0. The formation of N-containing functional groups and Fe oxides on the biochar surface enhanced the nitrate removal efficiency of N-Fe biochar. The results indicate that biochar’s adsorption capacity for NO₃⁻ is largely affected by the solution’s pH and biochar’s surface chemistry. Electrostatic attraction is the primary mechanism for nitrate adsorption.

1. Introduction

Nutrient pollution has become a critical environmental issue of global concern. Although nitrogen (N) is an essential element in aquatic ecosystems, excessive use of nitrogenous fertilizers and disposal of untreated industrial wastes can lead to elevated levels of N in natural water systems, thus impairing water quality [1,2,3,4]. High concentrations of N and P in water lead to eutrophication, threatening the ecosystem and human health [5]. Nitrate (NO₃⁻) is a monovalent anion with lone electrons and strong electronegativity [6] that is highly soluble in water, making it one of the most widespread groundwater contaminants [7]. High NO₃⁻ concentrations in water can cause some types of cancer and a fatal disorder called methemoglobinemia [8,9]. The U.S. Environmental Protection Agency (US EPA) set the maximum contamination level of NO₃⁻ in drinking water to be 10 mg L⁻¹ [10], which is approximately equivalent to the World Health Organization (WHO) guideline (i.e., 50 mg L⁻¹ as NO₃⁻, equivalent to ~11 mg L⁻¹ as nitrate nitrogen NO₃-N) [11]. However, the concentrations of NO₃⁻ in groundwater in some regions can surpass 80 mg L⁻¹ [12] and 200 mg L⁻¹ of NO₃⁻ [13], by far exceeding the WHO’s guidelines. Therefore, there is an urgent necessity to develop effective and low-cost solutions to remove NO₃⁻ from water sources [14]. Groundwater contamination by NO₃⁻ can originate from various sources, such as wastewater treatment plants, poorly managed septic systems, industrial activities, leaky urban sewers, urban runoff, fertilizers, and animal waste [15]. For example, a study conducted in California showed that, for over 50 years, NO₃⁻ from fertilizer and animal waste had contaminated the Tulare Lake Basin and Salinas Valley aquifers [16]. Similarly, Adimalla et al. conducted a study on the chemistry, distribution, and potential health risks of NO₃⁻ in groundwater in a semi-urban area of South India [12]. The study findings indicated that 43.3% of groundwater samples exceeded the safety limit of 45 mg L⁻¹ according to Indian guidelines [12]. The authors suggested that the main sources of NO₃⁻ in groundwater are non-point sources, such as fertilizers, animal waste, and poor sanitation facilities [12].

Different technologies and processes have been employed to remove NO₃⁻ from water, including reverse osmosis, electrodialysis, ion exchange, biological denitrification, chemical denitrification, and adsorption [17]. Biological denitrification is a cost-effective and environmentally friendly method [18,19], but the process could be long, limiting its use [20]. Additionally, this method is sensitive to environmental conditions [21]. Ion exchange is also effective, but it requires careful waste brine disposal and costly ion exchange resins. Reverse osmosis, although effective, is also an expensive method [17]. Chemical denitrification is not feasible at a large scale due to the excessive operational costs involved [22]. Electrodialysis has some drawbacks that still need to be considered. Depending on the level and type of pollutants present in the feed water, a pre-treatment stage may be required. Additionally, pH regulation may be necessary for the water obtained after the electrodialysis process [23]. Adsorption offers the potential to be an efficient and cheaper method for removing NO₃⁻. Its easy operation makes adsorption one of the best technologies for removing pollutants from water [24,25]. In the case of agricultural runoff, adsorption seems to be the only viable path. Operational parameters such as solution pH, contact time, temperature, adsorbent dosage, and NO₃⁻ concentration should be carefully controlled for efficient NO₃⁻ adsorption [6]. Different materials, such as metal oxides/hydroxides, carbon-based materials, organic polymer adsorbents, and agricultural wastes, are used as NO₃⁻ adsorbents [6].

Biochar is one of the main products of biomass pyrolysis under oxygen-limited atmospheres. Biochar’s unique properties, such as a high surface area, porous structure, and tunable surface chemistry, make it a suitable adsorbent for a wide range of contaminants. [4,25]. Positively charged surfaces containing exchangeable anions, multivalent metal ions, and/or hydrogen atoms can be active sites for NO₃⁻ adsorption. However, despite the biochar’s tunable properties, most commercially available biochars have a deficient adsorption capacity. The lack of suitable surface functional groups and the required porosity results in poor NO₃⁻ adsorption. Therefore, biochar modification is essential for NO₃⁻ adsorption to obtain tuned structural and chemical properties. The surface charge of biochar can be largely influenced by the pretreatment of biomass, the pyrolysis conditions, and the post-treatment of biochar. Therefore, these factors need to be considered and controlled when producing biochar for various applications [26]. Biochar produced via conventional pyrolysis under inert gas has high oxygen functional groups [17,27] and typically has a negative surface charge at neutral pH, which is not favorable for the adsorption of anions (e.g., NO₃⁻) [28,29]. Different modification strategies have been used to improve biochar’s anion adsorption capacity. Positively charged primary functional groups such as amides, aromatic amines, and pyridinic groups could be suitable adsorption sites for negatively charged nitrate ions [30,31]. Introducing heteroatoms to biochar carbon polyaromatic ring systems can improve electronic properties, including readily available lone pairs of electrons, durability, thermal stability, charging and polarizability, and conductivity [32]. Heteroatom-containing functional groups can alter the surface properties of biochar since these groups are generally polar and possess sites for hydrogen bonding, ion–dipole, and dipole–dipole interactions [33]. In contrast, hydrophobic carbon-rich surfaces of biochar are suitable sites for non-polar moieties [34].

N doping can be an effective way to introduce these basic functional groups on biochar surfaces [31,35,36]. The role of N in biochar to create/modify functional groups relies on the fact that N has more electronegativity than C, which enables electron transfer from neighboring C, resulting in a high charge density of C atoms [36,37], as shown in previous studies. Yoo et al. produced N-doped activated carbon to remove NO₃⁻ from an aqueous solution and found that N-doping increased the adsorption capacity of activated carbon from 0.38 to 0.75 mmol g⁻¹ [38]. One method for introducing N functional groups involves modifying the environment employed for producing biochar. Pyrolysis of biomass under NH₃ gas was shown to introduce basic N-containing functional groups on the biochar surface due to the reduction in the electron-withdrawing acidic groups and the increase in π electron-dense and oxygen-containing basic groups [39]. This method produced N-doped chars from anaerobically digested fiber with an improved affinity towards anionic pollutants [28]. The increase in the adsorption capacity of the modified biochar was explained by the formation of basic functional groups [31,35,40]. It has also been indicated that introducing metal oxides on biochar surfaces might lead to a positively charged surface that can increase the NO₃⁻ adsorption rate [41]. Several methods have been used to introduce Fe to biochar structures [42,43,44]. Fe can form different Fe oxides on biochar surfaces that improve NO₃⁻ adsorption [42,43,44].

The previously mentioned studies indicate that N doping and Fe doping have been used individually to remove NO₃⁻ from water. However, the combined impact of biochar N and Fe doping has not been addressed in the literature, to the best of the authors’ knowledge. Therefore, the aim of this work is to evaluate the influence of N doping, Fe doping, and Fe-N co-doping on the adsorption capacity of biochar obtained from Douglas fir wood. To the best of the authors’ knowledge, it is the first time that the combined effect of N and Fe on NO₃⁻ adsorption has been investigated. Fe-N-co-doped biochars were produced through one-step NH₃ pyrolysis of FeCl₃-loaded Douglas fir. Response Surface Methodology (RSM) with central composite design was used to examine the effects of biochar processing parameters and Fe content on the adsorption capacity towards NO₃⁻ in an aqueous solution.

2. Materials and Methods

2.1. Biochar Production

Douglas fir wood (DF) residues were used as feedstock for biochar production. DF was air-dried to reduce the moisture content below 8 wt.%, ground, and sieved to obtain particles that passed through a 0.6 mm sieve. The following four types of biochars were produced using N₂ or NH₃ gases: (1) biochar produced through pyrolysis under N₂ gas (herein referred to as DF biochar), (2) biochar produced through pyrolysis under NH₃ gas (i.e., N-DF biochar), (3) biomass impregnated with FeCl₃ solution pyrolyzed under N₂ gas (i.e., Fe-DF biochar); and (4) biomass impregnated with FeCl₃ pyrolyzed under NH₃ gas (i.e., Fe-N-DF biochar). All biochars were produced in duplicate. Figure 1 depicts the processes to produce the biochars. Regarding the fourth type of biochar (Fe-N-DF), biochar was first produced at one temperature (800 °C), metal loading (1 g of FeCl₃ per 15 g of DF), and pyrolysis time (1 h) to compare the NO₃⁻ adsorption performance with other types of biochars. Then, the process (for type 4 biochar) was optimized to maximize the biochar surface area and NO₃⁻ adsorption capacity.

The DF biochar (the first type of biochar) was produced through pyrolysis of the DF particles under N₂ gas at 800 °C in a quartz tube furnace reactor that was 1000 mm long with a 50 mm external diameter (Thermo Scientific Lindberg/Blue Hinged tube furnace HTF55000 Series, Asheville, NC, USA) with three heating zones (controlled independently) that work at similar temperatures. The materials were placed inside the tube using three quartz combustion boats. Approximately 5 g of ground DF was placed in each boat (i.e., ~15 g in total). The heating-up period was performed under N₂. Briefly, the DF particles were in contact with N₂ for 30 min at 25 °C. Afterward, the temperature was increased from 25 °C to 800 °C at a heating rate of 10 °C min⁻¹ and kept at 800 °C for one hour. A flow rate of 500 mL min⁻¹ of N₂ was employed. Then, the biochar produced was cooled down to 100 °C or below under N₂ before removal. No recovery of the pyrolysis liquid by-product was conducted.

The second type of biochar, N-DF biochar, was produced following the same procedure, but instead of N₂, NH₃ gas was used (at a flow rate of 600 mL min⁻¹). For the third type of biochar, Fe-doped biochar (Fe-DF biochar), 15 g of DF, 1 g of FeCl₃ (Sigma Aldrich St. Louis, MO, USA), and 200 mL of water were mixed using a magnetic stirrer at 60 °C for 6 h. The mixture was then filtered using filter paper to separate Fe-containing DF from FeCl₃ solution (after filtration, some FeCl₃ content remains in the solution), and then Fe-containing DF was dried at 80 °C for 48 h. The remaining solid was subjected to pyrolysis under N₂ gas. Finally, the Fe- and N-co-doped biochar (Fe-N-DF biochar) was prepared through pyrolysis of Fe-containing DF under NH₃ gas in a similar way to that employed for the N-DF biochar. In all cases, the final biochar products were washed thoroughly with ionized water until a neutral pH was achieved.

2.2. Char Characterization

Proximate analysis: Proximate analysis was performed in triplicates, using a thermogravimetric analyzer (TGA), SDTA851e (Mettler Toledo, Columbus, OH, USA), to determine char moisture, fixed carbon, volatiles, and ash content. Biochar was heated under N₂ gas in a crucible from 25 to 120 °C and held at 120 °C for 3 min to determine moisture content. The biochar was then heated to 950 °C under N₂ gas (50 mL min⁻¹) and held for 5 min to assess the volatile matter content. Then, the biochar was cooled down to 450 °C. Ash was measured as the remaining mass after biochar burning at 600 °C under an oxygen gas environment for 8 min [28].

Elemental analysis: Elemental analysis was carried out using a TRUSPEC-CHN^® (LECO, St. Joe, MI, USA) elemental analyzer to measure total nitrogen (N), Hydrogen (H), and carbon (C), following the process described previously [28]. Oxygen (O) was computed by difference. These analyses were conducted in triplicates.

Gas physisorption analysis: N₂ and CO₂ adsorption isotherms were measured at 76.85 K and 273 K on a TriStar II PLUS Surface Area and Porosity Analyzer (Micromeritics, Norcross, GA, USA). The micropore volumes were analyzed from the CO₂ adsorption using the Dubinine–Radushkevich (DR) equation. A non-localized density functional theory (NLDFT) was employed using commercial software (MicroActive™ v.1.01, Micromeritics, Norcross, GA, USA) to determine the pore size distribution. Density functional theory also provided an independent assessment of the volume of pores [45,46].

X-ray photoelectron spectroscopy (XPS): XPS was carried out using a Kratos AXIS ULTRADLD XPS (Manchester, UK) system equipped with an Al Ka X-ray source and a 165 mm mean radius electron energy hemispherical analyzer. Vacuum pressure was kept below 3 × 10⁻⁹ torr during the acquisition.

Char morphology: Scanning electron microscope (SEM) imaging analysis was performed using a Tescan Vega3 instrument (Brno, Czech Republic) combined with energy dispersive spectroscopy (EDS). Biochar particles were mounted on a stub and gold-coated before visualization. SEM and EDS were applied to examine the structure and surface characteristics of the char before and after adsorption. A transmission electron microscope (TEM) was used to study biochar’s surface morphology and micro and nanostructure. For the test, each biochar sample was ground into powder. A suspension of distilled (DI) water and fine powder samples were prepared and dispersed onto copper grids. Imaging was conducted at 200 kV under vacuum conditions with an FEI Tecnai G2 20 Twin [47].

X-ray powder diffraction (XRD): The crystallinity of biochars was studied using XRD (Miniflex 600 benchtop X-ray diffractometer Rigaku, Tokyo, Japan) with Cu K α radiation and operated at 40 kV, 15 mA, with 0.01 degree-steps and a scanning rate of 0.5° min⁻¹. The scan range of interest for this analysis was 20–100°.

Fourier Transform Infrared Spectroscopy (FTIR): FTIR analysis was conducted to identify the functional groups on the biochars’ surfaces. FTIR spectra were obtained using a Shimadzu IRPrestige 21 spectrometer (Tokyo, Japan) equipped with a MIRacle single reflection ATR Ge probe with the spectra recording from 900 to 3900 cm⁻¹.

2.3. Nitrate Adsorption Study

Adsorption of NO₃⁻ was carried out to obtain isotherms to evaluate the adsorption efficiency of the four types of biochars. These experiments were conducted in duplicate. In each case, 0.1 g of biochar was mixed with 35 mL of nitrate solution with different concentrations of nitrate–nitrogen (NO₃-N) ranging from 10 to 30 mg L⁻¹ in 50 mL tubes at room temperature (25 °C). Nitrate solutions were prepared by dissolving NaNO₃ (Fisher Scientific,Fair Lawn, NJ, USA) in deionized water. Then, the tubes were shaken at 130 rpm in a mechanical shaker (orbital shaker, Lab-Line Instruments, Inc., Melrose Park, IL, USA) for 24 h. The samples were then filtered using 0.45 μm filters to measure the corresponding nitrate equilibrium concentration. The pH of the solutions was measured using a pH meter (Mettler Toledo, SevenEasy S20) before and after NO₃⁻ adsorption. NO₃⁻ adsorption capacities were determined by measuring the initial and the final concentrations of the NO₃⁻ in the aqueous solution, which were evaluated by the Chromotropic acid method [48]. To explore the effect of pH on NO₃⁻ adsorption, the pH of the solution was set to three different values (i.e., 3, 7, and 8). The pH of the solutions was adjusted using 0.1 M HCl or 0.1 M NaOH.

2.4. Experimental Design

Multiple variables influence the pyrolysis of Fe-impregnated Douglas fir. Therefore, the process needs to be optimized to identify independent variables (pyrolysis temperatures, pyrolysis time, and metal loading) and maximize biochar surface area and NO₃⁻ adsorption capacity. A central composite design (CCD) was applied to optimize the preparation conditions of Fe-N-DF biochar. Statistical software Design Expert, version 13 (Stat-Ease, Inc., Minneapolis, MN, USA), was used in our analysis. The independent factors and their coded levels for the CCD are shown in

Table 1. Independent factors and their coded levels for the CCD.

Factors	Codes	Coded Factors’ Level
		Low (−1)	Center (0)	High (+1)
Pyrolysis temperature (°C)	A	600	700	800
Pyrolysis time (min)	B	30	60	90
Metal loading * (g/per 15 g of DF)	C	2	4	6

* Metal loading is defined as the amount of FeCl₃ in grams per 15 g of biomass in 200 mL of water.

.

3. Results and Discussion

3.1. Biochar Production and Characterization

Fe and N significantly affected the pyrolysis process, biochar yield, and biochar structure. The yields of biochars produced (at 800 °C for 1 h in all cases) were 21.1, 20.5, 28.2, and 24.3 wt.% for DF, N-DF, Fe-DF, and Fe-N-DF biochar, respectively. The presence of FeCl₃ contributed to an increase in biochar yields, which is consistent with previous findings [28,29,49]. The results of the proximate and elemental analysis are presented in

Table 2. Results on the proximate analysis of the four types of biochar.

Sample	Volatile Matter	Fixed Carbon	Ash
DF biochar	7.23 ± 0.18	92.23 ± 0.65	0.53 ± 0.47
N-DF biochar	7.06 ± 0.21	92.7 ± 0.25	0.24 ± 0.04
Fe-DF biochar	8.07 ± 0.02	89.01 ± 1.08	2.92 ± 0.07
Fe-N-DF biochar	13.40 ± 0.47	83.24 ± 0.62	3.37 ± 0.01

and

Table 3. Elemental composition (wt. %) and C/N and C/O ratios of the biochars produced.

Sample	C	H	N	O *	C/O	C/N
DF biochar	93.62 ± 0.2	0.51 ± 0.04	0.60 ± 0.05	4.74	19.77	156.07
N-DF biochar	80.09 ± 0.1	0.93 ± 0.03	6.65 ± 0.05	12.08	6.63	12.05
Fe-DF biochar	91.91 ± 3.3	0.6 ± 0.02	0.61 ± 0.1	3.96	23.18	151.21
Fe-N-DF biochar	80.73 ± 1.2	0.85 ± 0.05	8.03 ± 0.1	7.02	11.5	10.05

* By difference.

, respectively. Impregnation of DF with FeCl₃ solution resulted in biochar with higher ash content and lower fixed carbon than DF biochar. As shown in Table 3, N content in Fe-N-DF biochar is higher (8 wt.%) than in N-DF biochar (6.6 wt.%), which results from the presence of Fe because Fe in biochar catalyzes the formation of the N dopant [28,50]. The N and O contents increased after Fe-N co-doping, indicating that Fe and N were introduced into the biochar matrix and modified the chemical composition of biochar. According to Luo et al., introducing N-containing functional groups can occur through the reaction of NH₃ with oxygen-containing functional groups [51]. Elemental analysis results suggest that introducing Fe to DF increased the relative O content in the resulting biochar, possibly due to the formation of O-containing functional groups, such as Fe oxides [52,53].

Biochar’s surface areas and pore volumes obtained with CO₂ and N₂ adsorption are shown in

Table 4. Surface areas and pore size volumes as determined by N₂ and CO₂ adsorptions.

Sample	SaN2 (m2 g−1)	SaCO2 (m2 g−1)	Micro Volume (cm3 g−1)
DF Biochar	478 ± 9.8	533	0.21
N-DF Biochar	713 ± 14.6	694	0.28
Fe-DF Biochar	431 ± 5	494	0.20
Fe-N-DF Biochar	438 ± 4.1	477	0.19

. The N₂ adsorption method is performed at a temperature of −196 °C. This usually leads to lower surface area values for materials with micropores (<2 nm) because of the kinetic diameter of the gas molecule (0.36 nm) [25,54]. Slow diffusion of N₂ at −196 °C can limit the use of N₂ in the range between 0.7 and 1.47 nm. A combination of CO₂ and N₂ adsorption should be used to characterize biochar [25,55]. N-DF biochar exhibited the highest specific surface area and pore volume. The results suggest that pyrolysis under NH₃ gas generates more pores in biochar structure than conventional pyrolysis under inert gas (e.g., N₂). Previous studies have reported that N doping through NH₃ treatment can increase the specific surface area of biochar [51]. Pyrolysis under ammonia gas results in carbon skeleton etching by gas–solid reactions, producing biochar with a high surface area [50,56]. At around 500 °C, NH₃ decomposes into ˙NH₂, ˙NH, H atoms, and N atoms. The radicals produced in this process react with active substances such as C=O or –OH to create N-containing functional groups. These radicals also take part in gas–solid reactions and etch the surface of the biochar, which increases the pore volume and specific surface area through the pathway shown in equation (1) [57].

(C_biomass) + NH₃ → C* + H* + NH* + NH*₂ → H₂ + C-NH + C-NH₂

(1)

The surface areas of Fe-DF and Fe-N-DF biochars obtained from CO₂ and N₂ adsorption are lower than that of pristine biochar, which can be attributed to the blockage of pores by Fe oxides and the resulting decrease in the surface area [58]. Pores can be occupied by Fe nanoparticles and reduce the surface area of biochar [59]. The pore size distribution of biochar is shown in Figure 2, indicating that the pore structure of biochar (N-DF) has been improved by pyrolysis under NH₃ gas. Moreover, most of the pores in all types of biochar have a width between 0.30 and 0.70 nm. The CO₂ and N₂ adsorption isotherms obtained from gas physisorption analyses are shown in Figure 3 and Figure 4. The isotherm in Figure 3 shows mainly the presence of micropores. IUPAC conventions offer a standardized method for categorizing pore sizes and gas sorption isotherms, providing valuable insights into the correlation between porosity and sorption. The adsorption–desorption isotherm obtained from the N₂ method provides information about porosity, specific surface area, and pore size distribution [60]. Figure 3c for the Fe-DF biochar clearly shows a typical IUPAC type IV isotherm with an H3 hysteresis loop, indicating that the Fe-DF biochar is mainly mesoporous [61,62]. The isotherms presented in Figure 3a,b (e.g., N-DF) can be categorized as type I isotherms, which suggests that biochar is considered a microporous solid [63]. Moreover, the isotherm shown in Figure 3d can be classified as a type II isotherm. CO₂ adsorption isotherms (Figure 4) indicate that the amount of CO₂ adsorbed at p/p₀ = 0.03 also increased when pyrolysis of DF was conducted under an NH₃ environment, showing an increase in micropores in the biochar.

The combined effect of pyrolysis temperature, metal loading, and pyrolysis time on the surface area of Fe-N-co-doped Douglas fir biochar is shown in Figure 5. We tested experimental optimization strategies with a central composite experimental design for biochar in which we varied the metal loading by 2 to 6 g per 15 g of Douglas fir wood, pyrolysis time (between 30 and 90 min), and the final pyrolysis temperature (between 600 and 800 °C) as input variables to identify the conditions resulting in a product with the highest surface area. As seen in Figure 5, the highest surface area can be achieved at the pyrolysis temperature of 698 °C (~700 °C), metal loading of 2 g per 15 g of DF, and the pyrolysis time of 90 min. In Figure 5, it can be observed that the pyrolysis time has a minimal effect on the biochar surface area.

FTIR was used to study the characteristics of functional groups in biochars (Figure 6). Bands between 3723 and 3176 cm⁻¹ were attributed to –OH and N–H functional groups. However, these peaks cannot be observed in Figure 6 since the –OH group disappears in biochars produced at high temperatures (e.g., 800 °C). The rise observed at 1080 cm⁻¹ can be attributed to C–O/C–N stretching vibration. The peak at around 1220 cm⁻¹ could be attributed to C–N stretching. Peaks between 1400 and 1680 cm⁻¹ could be ascribed to C=O, C=C aromatic, and C=N stretching. The N-DF and Fe-N-DF biochars were produced under NH₃ gas; therefore, there is a high possibility of C=N formation. Moreover, peaks between 2800 and 2900 cm⁻¹ can be observed, which might be attributed to aliphatic C–H [64,65].

The SEM images (Figure 7a,b) show that fine particles were dispersed on the surface of Fe-N-DF biochar (produced at 800 °C), which could be due to the formation of metal oxides during the pyrolysis. The EDS result shows the presence of primarily N, Fe, and O, in addition to C (the main constituent) in biochar structures (see Figure 7c). The presence of N and Fe suggests that these elements were successfully doped in biochar structures. TEM images of Fe-N-DF (Figure 7d,e) showed irregular and amorphous morphology on the surface. Small spherical particles can be seen in the biochar structure, which could be due to the formation of nanoparticles of Fe oxides, showing that Fe was embedded in the carbon structure. TEM images also show the presence of nanoparticles between 20 and 50 nm. This result suggests that in addition to the Fe integrated into the biochar structure, some Fe also precipitated on the surface. Iron was detected on the biochars’ surface (Figure 7c,f).

3.2. Nitrate Adsorption

Nitrate adsorption studies were conducted at three different pH levels (3, 7, and 8), using four types of produced biochars. Four different isotherms, including Langmuir, Freundlich, Langmuir–Freundlich, and Redlich–Peterson (Equations (1)–(4)), were applied to explain the adsorption characteristics observed for NO₃⁻ of each material. In the Langmuir equation, K (L mg⁻¹) and Q (mg g⁻¹) denote the Langmuir bonding term, which is associated with interaction energies and Langmuir maximum capacity, respectively, and Ce (mg L⁻¹) is the equilibrium solution concentration of the sorbate. K (mg⁽¹⁻ⁿ⁾ Lⁿ·g⁻¹) and n (dimensionless) in Freundlich represent the Redlich affinity coefficient and the Freundlich linearity constant, respectively. K (Lⁿ mg⁻ⁿ) and n denote the Langmuir–Freundlich affinity parameter and heterogeneity index, respectively. K (L g⁻¹), a (Lⁿ mg⁻ⁿ), and n in the Redlich–Peterson equation represent Redlich–Peterson isotherm constants. The Langmuir model corresponds to monolayer adsorption on a homogeneous adsorbent surface with no interactions between the adsorbed molecules [66,67]. The Freundlich model considers a heterogeneous sorption process with a non-uniform distribution of adsorption heat, which can be applied to multilayer adsorption [68]. The Langmuir–Freundlich model is a combined isotherm model of Langmuir and Freundlich isotherms. Thus, the Langmuir–Freundlich model can be used to model both homogeneous and heterogeneous binding surfaces [69]. Likewise, the Redlich–Peterson model can be applied to either homogeneous or heterogeneous systems since this model is a hybrid isotherm that features Langmuir and Freundlich isotherms [70,71]. Equations (2)–(5) describe the mentioned models.

$$\mathrm{q}=\frac{\mathrm{K}\mathrm{Q}\mathrm{C}\mathrm{e}}{1+\mathrm{K}\mathrm{C}\mathrm{e}}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\mathrm{Langmuir}$$

(2)

$$\mathrm{q}={\mathrm{K}\mathrm{C}\mathrm{e}}^{\mathrm{n}}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\mathrm{Freundlich}$$

(3)

$$\mathrm{q}=\frac{\mathrm{K}\mathrm{Q}{\mathrm{C}\mathrm{e}}^{\mathrm{n}}}{1+\mathrm{K}{\mathrm{C}\mathrm{e}}^{\mathrm{n}}}\hspace{1em}\hspace{1em}\mathrm{Langmuir–Freundlich}$$

(4)

$$\mathrm{q}=\frac{\mathrm{K}\mathrm{C}\mathrm{e}}{1+\mathrm{a}{\mathrm{C}\mathrm{e}}^{\mathrm{n}}}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\mathrm{Redlich–Peterson}$$

(5)

The adsorption isotherms of DF, N-DF, Fe-DF, and Fe-N-DF biochars (produced at 800 °C, metal loading of 1 g of FeCl₃ to 15 g of DF, and pyrolysis time of 1 h), at pH 7, are shown in Figure 8. The maximum NO₃⁻ adsorption capacity for Fe-N-DF biochar at pH 7 was 9.37 mg g⁻¹, which is approximately two-fold higher than that of pristine biochar (4.82 mg g⁻¹). Generally, the surface of pristine biochars is negatively charged, limiting its binding affinity toward NO₃⁻. The results suggest that introducing N and Fe to the biochar structure improves NO₃⁻ adsorption on the biochar surface. It also appears that although N-DF biochar exhibited the highest surface area, the role of N and Fe functional groups in Fe-N-DF biochar is more important than the surface area for NO₃⁻ adsorption. The isotherms in Figure 8 indicate that the Langmuir–Freundlich and Redlich–Peterson models provided the best fit. The results also show that decreasing pH caused improved NO₃⁻ adsorption capacity, which might be attributed to the protonation of surface functional groups. The maximum NO₃⁻ adsorption capacity for the Fe-N-DF biochar at pH 3 was 19.4 mg g⁻¹ (

Table 5. Isotherm parameters for adsorption of NO₃⁻.

Biochar	pH	Langmuir				Freundlich
		K (L mg−1)	Q (mg g−1)		R2	K (mg(1−n) Ln g−1)	n	R2
DF	3	0.28	17.0		0.98	6.06	0.23	0.86
Fe-DF		0.38	16.1		0.99	6.38	0.23	0.88
N-DF		0.40	18.9		0.97	7.26	0.24	0.87
Fe-N-DF		0.83	19.4		0.97	9.01	0.20	0.85
DF	7	0.12	4.8		0.88	1.47	0.25	0.81
Fe-DF		0.11	6.3		0.97	1.69	0.28	0.98
N-DF		0.15	8.7		0.90	2.47	0.28	0.95
Fe-N-DF		1.44	9.37		0.97	5.17	0.15	0.86
DF	8	0.09	3.5		0.88	0.94	0.27	0.87
Fe-DF		0.12	5.0		0.96	1.64	0.23	0.80
N-DF		0.39	6.3		0.83	2.94	0.18	0.90
Fe-N-DF		0.53	7.3		0.92	3.49	0.18	0.99
Biochar	pH	Langmuir–Freundlich				Redlich–Peterson
		K (Ln mg−n)	Q (mg g−1)	n	R2	K (L g−1)	a (Ln mg−n)	n	R2
DF	3	0.23	16.1	1.3	0.99	4.37	0.22	1.04	0.98
Fe-DF		0.38	16.1	1.0	0.99	6.34	0.41	0.99	0.99
N-DF		0.38	18.4	1.2	0.98	7.78	0.43	0.99	0.98
Fe-N-DF		0.89	18.9	1.2	0.98	17.1	0.94	0.98	0.98
DF	7	0.07	4.5	1.3	0.89	0.57	0.11	1.00	0.89
Fe-DF		0.15	8.4	0.6	0.99	1.28	0.41	0.84	0.99
N-DF		0.13	19.4	0.4	0.95	4.13	1.22	0.79	0.96
Fe-N-DF		1.09	9.8	0.7	0.98	14.83	1.73	0.98	0.97
DF	8	0.14	4.1	0.7	0.90	0.44	0.21	0.90	0.89
Fe-DF		0.04	4.6	1.5	0.98	0.40	0.04	1.17	0.99
N-DF		0.31	11.3	0.3	0.91	9.06	2.63	0.86	0.91
Fe-N-DF		0.34	12.8	0.3	0.99	12.41	2.99	0.86	0.99

). The NO₃⁻ adsorption depends on pH, which is consistent with electrostatic attraction. Deprotonation of surface functional groups occurs as the pH increases. Moreover, OH⁻ ions become dominant adsorption sites with increasing pH and compete with NO₃⁻ [72]. NO₃⁻ can also form hydrogen bonds to protonated surface hydroxyls on the Fe oxides [44].

High-resolution XPS analysis was performed to understand biochar’s surface chemistry (See Figure 9). The peak N1s could be observed in the spectrum of N-DF and Fe-N-DF biochars. The N 1s spectra were deconvoluted into the following five peaks: pyridinic-N, pyridone-N, pyrrolic-N, graphitic-N, and oxidized-N. The pyridine functional group at 398.2 eV is the main nitrogen functional group in N-DF and Fe-N-DF biochars (Figure 9). Pyrrolic and pyridone groups on the biochar surface are neutral. However, pyridinic groups protonate at pH below pK_a, resulting in +1 formal charge [34]. The pH of a solution is a crucial parameter influencing biochar surface charge because it can affect the protonation and deprotonation of surface functional groups, leading to surface charge change [73]. Protonation of biochar surface functional groups occurs at low solution pH, resulting in increased positive surface charge with the consequent impact on NO₃⁻ adsorption [74,75]. Pyridinic-N could contribute with one electron to the aromatic π-system and shows Lewis basic characteristics [76]. Previous studies showed that N doping altered the N-configuration on the char surface and caused a positive effect on the adsorption of negatively charged ions [28,35,47].

XRD results are shown in Figure 10. The diffraction peak at 2θ of 26.3° was assigned to the (002) diffraction of amorphous carbon [28]. The presence of α-Fe₂O₃, Fe₃O_4, and Fe₃C can be observed in the XRD results. The formation of Fe₃C resulted from the reduction of Fe oxide in the presence of N [77,78]. Diffraction peaks around 2θ of 40.7°, 42.8°, 44.6°, and 75.8° were attributed to crystalline phases of Fe₃C. Fe₃O₄ can be formed during the pyrolysis of FeCl₃-loaded DF under NH₃ gas [79]. FeCl₃ is hydrolyzed and converted to FeO(OH) during the drying of FeCl₃-loaded biomass. Therefore, FeO/(OH) was converted to Fe₃O₄ by the reducing components H₂ and CO. H₂ is produced from NH₃ at high temperatures, promoting the formation of Fe₃O₄ (Reactions (6)–(9)) [79]. The XRD results show no major changes in peak locations resulting from NO₃⁻ adsorption, suggesting that the adsorption process does not significantly alter the crystalline structure in biochar structure. This result also suggests that electrostatic attraction is the primary mechanism of NO₃⁻ adsorption.

FeCl₃+ 3H₂O → Fe(OH)₃ + 3HCl

(6)

Fe(OH)₃ → FeO(OH) + H₂O

(7)

6FeO(OH) + 4H₂ → 2 Fe₃O₄ + 4H₂O

(8)

6FeO(OH) + 4CO → 2 Fe₃O₄ + 4CO₂

(9)

The effects of pyrolysis temperature, metal loading, and pyrolysis residence time on the NO₃⁻ adsorption capacity of Fe-N-DF biochar were also investigated. For the experimental optimization strategies with a central composite experimental design for Fe-N-DF biochar, we varied the metal loading from 2 to 6 g per 15 g of Douglas fir wood, pyrolysis time between 30 and 90 min, and final pyrolysis temperature between 600 and 800 °C as input variables to identify the conditions resulting in a product with the highest NO₃⁻ adsorption capacity (see Figure 11).

The optimum pyrolysis temperature, metal loading, and pyrolysis time were 800 °C, 6 g, and 30 min, respectively. The NO₃⁻ adsorption capacity of Fe-N-DF biochar at pH 3 and 7 are 20.67 and 9.4 mg g⁻¹, respectively, showing that the adsorption capacity of Fe-N-DF biochar (9.4 mg g⁻¹) at a pH solution of 7 is higher by around 96% compared to pristine biochar (DF biochar: 4.8 mg g⁻¹). The higher NO₃⁻ adsorption at pH 3 is attributed to the protonation of biochar surface functional groups. At low pH, protonation of biochar surface functional groups occurs, which causes an increasing positive surface charge [74,75]. Generally, the surface of raw biochar is negatively charged. Nevertheless, oxides of metals (e.g., Al₂O₃, MnO₂, Ca₂O₃, Fe₂O₃, MgO₂) and N-containing functional groups such as the pyridinic group in the biochar structure can make biochar more positively charged and, therefore, they can contribute to the adsorption of NO₃⁻ [80].

Table 6. Overview of results reported in the literature on NO₃⁻ removal using different types of biochar.

Feedstock	Treatment Method	Surface Area (m2 g−1)	pH of the Solution	Adsorption Capacity (mg.g−1)	References
Oak sawdust	Feedstock impregnated with LaCl2	23.08	-	8.7	[81]
Oak sawdust	-	24.94	-	2.81	[81]
Sugarcane bagasse	Chemical modification using epichlorohydrin, N, N-dimethylformamide, ethylenediamine and trimethylamine	41.67	4.64	28.21	[17]
Corncob	-	-	-	14.46	[82]
Bamboo	-	28	4	5	[83]
Bamboo	Feedstock impregnated with Clay	156	4	9	[83]
Pinewood waste	-	204.2	2	4.2	[84]
Douglas fir	-	478	7	4.8	This study
Douglas fir	Ammonia treatment	713	7	8.7	This study
Douglas fir	Feedstock impregnated with FeCl3	431	7	6.3	This study
Douglas fir	Feedstock was impregnated with FeCl3 and pyrolyzed under ammonia gas	438	7	9.4	This study
Douglas fir	Feedstock was impregnated with FeCl3 and pyrolyzed under ammonia gas	438	3	20.67	This study

compares this work’s experimental results with those reported in the literature for other materials. The results suggest that doping biochar with N and Fe greatly improves NO₃⁻ adsorption capacity from aqueous solutions. Therefore, co-doping biochar with Fe and N appears as a viable and easy-to-conduct strategy for improving the capacity of biochar to remove NO₃⁻ from water sources.

4. Conclusions

Novel Fe-N-co-doped biochar was produced via impregnation of Douglas fir with FeCl₃ solution, followed by pyrolysis under NH₃ gas. XRD, XPS, SEM, and EDS analyses revealed the presence of metal oxides and nitrogen-functional groups on the Fe-N-DF biochar. Although N-doped biochar exhibited a higher surface area (713 m² g⁻¹) than Fe-N-doped biochar (438 m² g⁻¹), the combined role of N and Fe functional groups in Fe-N-co-doped biochar is more important than large surface areas for NO₃⁻ adsorption. Therefore, the introduction of Fe and N in the biochar structure enhanced NO₃⁻ adsorption capacity. The presence of N-functional groups, especially pyridinic N, and Fe oxides on the synthesized Fe-N biochar provide more adsorption sites for NO₃⁻ adsorption, contributing to the high performance of NO₃⁻ removal from aqueous solution. The results show that decreasing the pH of the sorption medium caused improved NO₃⁻ adsorption capacity. The optimum process condition for producing Fe-N-DF biochar was found to be at a pyrolysis temperature of 800 °C, with a metal loading of 6 g per 15 g of biomass and a pyrolysis time of 30 min. These conditions resulted in a maximum NO₃⁻ adsorption capacity of 20.67 milligrams per gram at a pH of 3. Electrostatic attraction has been identified as the primary mechanism of NO₃⁻ adsorption.