Graphene-like Carbon Materials from King Grass Biomass via Catalytic Pyrolysis Using K₃[Fe(CN)₆] as a Dual Catalyst and Activator

Abstract

The potential of king grass biomass as a precursor for carbon-based materials was evaluated through comprehensive physicochemical characterization. The biomass showed high fixed carbon content, reactive oxygenated groups, and favorable atomic ratios, supporting its suitability for conversion into porous carbon structures. This study focused on the synthesis of graphene-like materials via high-temperature pyrolysis (~1000 °C), employing FeCl₃ and potassium ferricyanide (K₃[Fe(CN)₆]) as catalytic agents. Although FeCl₃ is widely studied, it showed limited capacity to promote graphitic ordering. In contrast, K₃[Fe(CN)₆] exhibited a synergistic effect, combining iron-based catalytic species (Fe, Fe₃C) and potassium-derived activating compounds (K₂CO₃), which significantly enhanced graphitization and porosity. Characterization by Raman spectroscopy, XRD, and SEM confirmed that materials synthesized with K₃[Fe(CN)₆] presented improved crystallinity, lower defect densities (I_D/I_G = 0.37–1.11), and distinct 2D bands (I_2D/I_G = 0.32–0.80), indicating the formation of few-layer graphene domains. The most promising structure was obtained from cellulose treated with alkaline peroxide and deoxygenated prior to pyrolysis with K₃[Fe(CN)₆], showing properties comparable to commercial graphene. BET analysis revealed surface areas up to 714.50 m²/g. While non-catalyzed samples yielded higher mass, the catalytic approach with K₃[Fe(CN)₆] demonstrates a sustainable and efficient pathway for producing graphene-like carbon materials from lignocellulosic biomass.

1. Introduction

Graphene and its derivatives, such as graphene oxide and reduced graphene oxide, have attracted significant attention due to their outstanding physicochemical properties, including high surface area, thermal and chemical stability, and tunable surface functionality [1,2,3]. These features make them highly suitable for advanced applications in adsorption, catalysis, and energy storage [4,5,6,7]. However, the performance of these materials is closely tied to their structure, which is in turn influenced by the synthesis route.

Among the various strategies for graphene synthesis, bottom-up approaches using carbon-rich precursors—including polymers, hydrocarbons, and biomass—have emerged as sustainable alternatives to top-down exfoliation or fossil-based methods [1,3]. Biomass offers key advantages such as low cost, global availability, and high carbon content (typically 45–55 wt.%) [4,8]. Recent studies have explored the use of lignocellulosic biomass for producing graphene-like carbon materials, often requiring pretreatments to release cellulose and the addition of metal catalysts and chemical activators during pyrolysis to promote graphitization and porosity development [8,9,10].

Iron-based catalysts (e.g., FeCl₃) have been widely reported to enhance graphitic structure formation under inert conditions, especially through the formation of intermediates such as Fe₃C [11,12,13,14,15,16]. However, iron alone typically produces materials with limited crystallinity or low surface area unless combined with activating agents like KOH or K₂CO₃ [1,2,13,17]. These potassium-based compounds induce pore development and improve the textural properties of carbon materials [1,2,5,13,17]. Most conventional methods rely on multi-step procedures, involving sequential impregnation, thermal activation, and carbonization, which increases chemical usage, energy input, and process complexity [1,2,5,13,17].

To address these limitations, this study proposes a novel and simplified route using potassium ferricyanide (K₃[Fe(CN)₆]) as a dual-function precursor, capable of simultaneously releasing iron (for catalytic graphitization) and potassium (for chemical activation) during biomass pyrolysis. This integrated approach allows for the formation of porous, graphene-like carbon materials in a single thermal step, eliminating the need for multiple reagents and treatments. Compared to previous studies that used FeCl₃ or separate activators [1,2,5,13,17], our strategy offers a technically and economically efficient alternative by reducing synthesis steps, processing time, and reagent consumption.

The selected biomass is king grass (Pennisetum hybridum), a fast-growing, non-edible perennial grass with high productivity (40–60 tons/ha/year in Colombia), easy transport, and low economic value—making it an ideal candidate for thermochemical valorization [18,19,20]. In addition to evaluating the catalytic performance of K₃[Fe(CN)₆], this study also compared it against FeCl₃ to assess the effectiveness of single-component versus dual-function precursors.

2. Materials and Methods

2.1. General Procedure for Carbonaceous Material Synthesis

This study tested king grass biomass to produce graphene-like carbon materials using ultrasound and microwave-assisted methods. This study aimed to investigate the influence of factors such as the type of alkaline peroxide treatment used for cellulose extraction from lignocellulosic herbaceous biomass and the type of catalyst applied for its deoxygenation. These factors were analyzed in terms of their impact on the yield of the final material (calculated as the percentage ratio between the final and initial masses obtained at each synthesis stage) and the physicochemical characteristics of the resulting carbonaceous materials. Additionally, to establish a comparative benchmark, the same physicochemical characterizations were performed on commercial cellulose and graphene samples.

2.2. Biomass Feedstock Collection

Cellulose extracted from purple king grass (Pennisetum purpureum) was employed as the carbon source. The lignocellulosic herbaceous biomass was collected at Guarne, Antioquia, Colombia (Alto de la Virgen; coordinates: 6.3236° N, 75.4573° W). Subsequently, a manual cleaning process was conducted to remove roots, followed by drying in a convection oven at 60 °C for 48 h. The dried biomass was then mechanically milled using a blade grinder and sieved through an ASTM No. 30 mesh (600 µm) to obtain a homogeneous powdered material, ensuring uniformity for subsequent processing stages.

2.3. Cellulose Extraction

Cellulose extraction from king grass biomass (UB-KG) was performed using two distinct methodologies: (i) a conventional alkaline peroxide process (DB-CAP) and (ii) an ultrasound-assisted alkaline peroxide process (DB-CAU).

The conventional alkaline peroxide process involved the following steps: Multiple batches of 20.00 g of dried powdered biomass were placed in cylindrical glass reactors (approximately 15 cm diameter × 7 cm depth). The material was evenly dispersed, followed by dropwise addition of 160 mL of 10 wt% hydrogen peroxide (H₂O₂). The pH was then adjusted to 11.5 by adding 2 M NaOH solution (6–8 mL) and maintained under static conditions at room temperature for 2 h to promote delignification. The extracted cellulose was washed with deionized water until the rinse water reached pH 7 (approximately 1.6 L total volume), dried in a convection oven at 60 °C for 48 h, milled using a blade grinder, and sieved through an ASTM No. 30 mesh (600 µm). Materials derived from this procedure were labeled as DB-CAP.

The ultrasound-assisted alkaline peroxide treatment was carried out in batches of 20.00 g of the dried biomass powder. Each batch was placed in a 1 L Erlenmeyer flask, to which an alkaline peroxide solution (2 M NaOH + 10 wt% H₂O₂) was added dropwise under constant stirring until a pH of 11.5 was reached. The flask was then placed in an ultrasonic bath preheated to 70 °C, and the mixture was sonicated at a frequency of 40 kHz for 2 h. The resulting material underwent the same washing, grinding, and drying procedures as those applied in the conventional alkaline peroxide process. Materials obtained through this method were labeled as DB-CAU.

2.4. Synthesis of Carbonaceous Materials

Catalyst Solution Preparation: Catalytic solutions of iron(III) chloride (FeCl₃) and potassium hexacyanoferrate(III) (K₃[Fe(CN)₆]) were prepared by separately dissolving 0.6 g of each compound in 15 mL of deionized water, yielding a solute-to-solvent mass-to-volume ratio of 1:25. Homogenization was achieved via magnetic stirring at 600 rpm for 30 min under ambient conditions (25 ± 2 °C, 1 atm). This procedure was repeated identically for both catalysts to ensure methodological consistency.

Preparation of the Precursor Solution and Deoxygenation: A total of 3.75 g of cellulose (either extracted or commercial) was mixed with 15 mL of a previously prepared catalyst solution, maintaining a solid-to-liquid ratio of 1:4 (mass-to-volume). The mixture was homogenized using an Ultra-Turrax homogenizer (Heidolph RZR 2021, Schwabach, Germany) at 150 rpm for 30 min at room temperature, resulting in a uniform precursor solution. To remove oxygen-containing functional groups present in the cellulose, the mixture was subjected to a deoxygenation treatment using a microwave digester (Sineo TANK eco, Shanghai, China) operated at 1 kW for 60 min, maintaining an approximate temperature of 170 °C. This procedure was systematically applied to all experimental combinations to obtain the different materials according to the type of cellulose and catalyst used.

Catalytic Pyrolysis: The precursor obtained in the previous step was placed in a stainless-steel batch reactor, in which the air atmosphere was replaced with nitrogen. The reactor was then placed in a muffle furnace (Centricol, Bogota, Colombia) for a pre-carbonization step, heating from room temperature to 300 °C at a rate of 10 °C/min. Once the target temperature was reached, the sample was held at 300 °C for 3 h to produce a carbon-rich material and remove volatile components. Subsequently, the pre-carbonized product was subjected to pyrolysis in the same reactor, heating to the desired final temperature (1000 °C) at a rate of 10 °C/min and holding for 3 h at that temperature to promote the graphitization of the material.

Acid Purification: In order to remove residues primarily of metallic nature and to purify the materials, the samples were washed with 1 M HNO₃, followed by multiple centrifugation cycles at 13,000 rpm using a Biobase TH16 II (BIOBASE Biodustry Co., Ltd., Jinan, Shandong, China) centrifuge for 30 min. Subsequently, the samples were washed with deionized water and dried at 105 °C for 2 h.

Figure 1 illustrates the fundamental stages involved in the different synthesis routes implemented: (1) collection and pretreatment of the raw material, (2) cellulose extraction, (3) cellulose deoxygenation, (4) catalytic pyrolysis through primary and secondary carbonization, and (5) final purification via acid washing. All materials were synthesized in triplicate for each synthesis pathway. Final yields are reported as the means ± standard deviations (SDs), and coefficients of variation (CV) were calculated to evaluate the reproducibility of the processes.

Additionally,

Table 1. Synthesis description and code for each material.

Code	Synthesis Description
CG-ND-ND	Carbonaceous material from non-delignified, non-deoxidized biomass grass and pyrolyzed at 1000 °C.
CG-ND-DW	Carbonaceous material from non-delignified biomass grass, deoxidized without catalyst and pyrolyzed at 1000 °C.
CG-DCAP-FeCl	Carbonaceous material from delignified biomass grass with conventional alkaline peroxide, deoxidized with FeCl3 and pyrolyzed at 1000 °C.
CG-DCAP-KFe	Carbonaceous material from delignified biomass grass with conventional alkaline peroxide, deoxidized with K3[Fe(CN)6] and pyrolyzed at 1000 °C.
CG-DUAP-FeCl	Carbonaceous material from delignified biomass grass with ultrasound alkaline peroxide, deoxidized with FeCl3 and pyrolyzed at 1000 °C.
CG-DUAP-KFe	Carbonaceous material from delignified biomass grass with ultrasound alkaline peroxide, deoxidized with K3[Fe(CN)6] and pyrolyzed at 1000 °C.
CC-FeCl	Carbonaceous material from commercial cellulose, deoxidized with FeCl3 and pyrolyzed at 1000 °C.
CC-KFe	Carbonaceous material from commercial cellulose, deoxidized with K3[Fe(CN)6] and pyrolyzed at 1000 °C.
CC	Commercial cellulose (Sigmacell Cellulose Type 101, Highly purified, Fibers Sigma Aldrich (St. Louis, MO, USA).
CG	Commercial graphene (Graphene nanoplatelets, Sigma Aldrich).
OG	Graphene oxide (Sigma Aldrich).

provides the codes along with corresponding descriptions of the different synthesis conditions employed.

2.5. Experimental Conditions for the Physicochemical Characterization of the Carbonaceous Materials

The physicochemical properties and chemical composition of the carbonaceous materials were determined by various techniques including Raman, XRD, H₂-TPR, AAS, BET surface area, Zeta potential and HRTEM.

The quantification of extractives, structural carbohydrates and lignin composition was carried out using the gravimetric method following the protocol established by the NREL/TP-510-42619 standard [21,22]. Lignin was quantified according to the stipulations of the NREL/TP-510-42618 standard [22]. To complete the determination of chemical composition, holocellulose and cellulose were quantified using the ASTM D-1104 method [23]. The details of the procedure have been described in previous studies [18,24].

Thermal decomposition behavior, including moisture content, volatile matter, fixed carbon, and ash content, was evaluated using a TGA Q500 analyzer, (TA Instruments, New Castle, DE, USA). The temperature protocol included: (1) equilibration at 30 °C under nitrogen (N₂, 50 mL min⁻¹), (2) heating to 120 °C at 40 °C min⁻¹ with a 12 min isothermal hold, (3) heating to 800 °C at 40 °C min⁻¹ with a 10 min isothermal hold under N₂, and (4) oxidation under air (50 mL min⁻¹) with a 15 min isothermal hold at 800 °C.

Bulk elemental composition (C, H, N, S) was determined using a TruSpec CHNS analyzer (LECO Corporation, St. Joseph, MI, USA). K and Fe content in the carbonaceous materials was determined by Atomic Absorption Spectroscopy (AAS) in an atomic absorption spectrophotometer, with an air–acetylene flame, Agilent 240 FS (Agilent Technologies, Santa Clara, CA, USA).

Textural properties were measured by N₂ physisorption at 77 K using a Quantachrome Autosorb Automated Gas Sorption System (Quantachrome Instruments, Boynton Beach, FL, USA). The samples were degassed under vacuum at 380 °C for 12 h prior to measurement.

X-ray diffraction (XRD) analysis was performed via a Rigaku D-Max 2200 Series (Rigaku Corporation, Tokyo, Japan) equipped with Cu–Kα radiation (λ = 1.54 Å) at a scanning rate of 3° per minute. The tube voltage and the current were 40 kV and 40 mA, respectively. The intensity was determined over a 2θ ° angular range of 5–70°. The relative crystallinity index (RCI) was estimated with Equation (1). The crystalline and amorphous peaks were deconvoluted (using a peak fit Gaussian function) for estimation of the area under the peak of the crystalline region (CR) and the amorphous regions (AR).

$$\mathrm{Relative} \mathrm{Cristallinity} \mathrm{Index} \left(\%\right)=\left({\displaystyle \frac{I_{CC}}{I_{CC}+I_{CA}}}\right)\times \left(100\right)$$

(1)

where I_CC is the intensity peak of the crystalline plane of the carbonaceous material, ICA is the intensity peak of the amorphous cellulose, and I_Total is the total area under the XRD peaks (I_Total = I_CC + I_CA).

High-resolution transmission electron microscopy (HRTEM) analyses were carried out by using a Si (Li) model INCA (Oxford Instruments, High Wycombe, UK), with techniques available for a JEOL 2100F electron microscope (JEOL Ltd., Tokyo, Japan), which was operated at 200 kV and a 0.19 nm resolution.

H₂-Temperature programmed reduction (H₂-TPR) of the materials was carried out at 10 °C/min from 25 °C up to 900 °C in the presence of 10% H₂/Ar (Infra, Mexico City, Mexico). The gas flow rate was 25 mL/min and it was kept constant using a mass flow controller. The apparatus was calibrated by reduction of 30 mg of CuO powder (99%, Merck, Rahway, NJ, USA) under the same experimental conditions.

Raman analyses were conducted at room temperature using a Horiba Jobin Yvon confocal Raman spectrometer (Horiba, Longjumeau, France) high resolution Labram HR model, with a 632.81 nm laser, in the range from 600 to 4000 cm⁻¹. Zeta potential measurements were performed with a Malvern ZEN 2600 (Malvern Panalytical, Malvern, UK).

3. Results and Discussion

3.1. Characterization of King Grass Biomass and Extracted Cellulose

3.1.1. TGA/DTA

As shown in Figure 2, TGA/DTA analyses revealed three well-defined thermal transitions in the king grass biomass, corresponding to the degradation of hemicellulose (221–325 °C), cellulose (325–400 °C), and lignin (400–600 °C). These thermal events reflect the sequential decomposition of the main structural components of the biomass, with lignin being the last to degrade due to its complex molecular structure. Accurate knowledge of these decomposition ranges is essential for the design and control of thermal processes aimed at the synthesis of carbonaceous materials using biomass as a carbon precursor.

3.1.2. Proximate and Elemental Analysis

Proximate analysis revealed that 60.66% of the biomass consisted of volatile matter, indicating a high susceptibility to thermal decomposition—an effect also reflected in the mass loss observed between 70 and 100 °C. In addition, a relatively high fixed carbon content was recorded (21.55%), which is advantageous for its use as a precursor material in the synthesis of carbonaceous compounds. This suggests a higher presence of stable aromatic structures that may contribute to the formation of a robust carbon matrix during pyrolysis. The moisture content was moderately low (9.48%), which favors improved thermal process yields, while the ash content (5.63 percent) was also low and falls within the expected range for lignocellulosic biomass, as reported in the literature [17,25].

Elemental (ultimate) analysis was consistent with the previous findings, revealing a high oxygen content (53.31%), followed by carbon (38.20%), hydrogen (4.87%), nitrogen (3.12%), and sulfur in minimal concentrations (0.50%). Based on these results, the atomic ratios O/C, H/C, and (O+N)/C were calculated as 1.05, 1.52, and 1.12, respectively. The elevated O/C ratio indicates a high abundance of oxygen-containing functional groups, which are typically removed to achieve higher levels of graphitization during carbonization processes, particularly those aimed at producing graphene-like carbon materials.

Conversely, the H/C ratio greater than 1.5 suggests low aromaticity and a high degree of unsaturation in the precursor, a common characteristic of uncarbonized lignocellulosic materials. This ratio decreases significantly during thermal treatment and is thus a useful parameter for tracking the structural evolution toward more graphitic or aromatic materials. Finally, the (O+N)/C ratio of 1.12 confirms the high polarity of the precursor, which is advantageous for applications requiring surface interactions, such as adsorption, catalysis, or electrochemical energy storage [26]. According to the literature, these atomic ratios fall within the characteristic ranges of lignocellulosic biomass considered suitable as precursors for the synthesis of activated or functionalized carbon materials, particularly those designed for environmental and energy-related applications [8,27,28,29,30,31].

3.1.3. Extractives, Structural Carbohydrates and Lignin Composition

The evaluation of carbohydrate composition in king grass biomass and the delignified materials reveals significant contrasts between the two alkaline peroxide procedures, both in terms of efficiency and their impact on cellulose content (

Table 2. Extractives, structural carbohydrates and lignin composition of untreated and delignified biomass.

Type of Biomass	Moisture (%)	Extractives (%)		Carbohydrates (%)
		Water	Ethanol	Lignin	Holocellulose	Hemicellulose	Cellulose
UB-KG	9.90	16.13	9.87	3.62	80.40	42.03	38.37
DCAP	7.65	0	0.00	0	84.81	39.33	45.48
DUAP	6.69	0	8.70	0	94.00	28.24	65.80

). The raw biomass (UB-KG) exhibits a holocellulose content of 80.40%, comprising 42.03% hemicellulose and 38.37% cellulose, along with a relatively low lignin content (3.62%). These results indicate that the material is initially well-suited for fractionation treatments aimed at producing cellulose-rich carbonaceous precursors [5].

The conventional alkaline peroxide treatment (DCAP) resulted in a material with a slight decrease in hemicellulose content (39.33%) compared to the raw biomass, but a notable increase in cellulose content to 45.48%, indicating partial removal of hemicellulose and a more effective release of cellulose chains. This trend is consistent with previous studies [24,32,33, which reported that conventional alkaline treatments effectively cleave ester bonds between lignin and hemicellulose, facilitating cellulose liberation. The complete absence of lignin and extractives suggests that the process was suitable for removing non-cellulosic components while preserving the target matrix.

In turn, the ultrasound-assisted alkaline peroxide treatment (DUAP) demonstrated an even greater capacity for cellulose extraction, reaching 65.80%, which represents a 71.5% increase relative to the original biomass. Unlike the conventional method, this treatment achieved a more significant removal of hemicellulose (28.24%), which can be attributed to the synergistic effect of ultrasound-induced cavitation. This phenomenon facilitates the disruption of the lignocellulosic network and promotes the selective cleavage of glycosidic and ether bonds within the hemicellulosic fractions. These findings are consistent with other studies conducted on different types of biomass, where sonication was similarly found to enhance the efficiency of chemical treatments by generating microcurrents and localized high-energy zones, thereby improving reagent penetration and reaction rates [18,24,32,33].

Both methods are effective in obtaining a delignified biomass, free of undesired components such as lignin and extractives. However, the ultrasound-assisted treatment presents additional advantages in terms of increased cellulose yield and greater hemicellulose removal, which is favorable when the objective is to produce high-purity carbonaceous materials with more homogeneous structures. Several studies support these observations, highlighting that a higher proportion of cellulose in sugarcane bagasse leads to greater carbonization efficiency and better development of pores and specific surface areas in the resulting carbon materials [1]. Both procedures enable high cellulose content to be obtained from king grass, with the ultrasound-assisted method offering higher efficiency in terms of structural polysaccharide concentration. This feature makes it a more suitable alternative for carbon material synthesis processes that require a precursor with high purity, improved thermal stability, and optimized H/C and O/C ratios for applications in adsorption, catalysis, and energy storage [26,29,34].

3.1.4. Yields of the Production of Carbonaceous Materials

Table 3. Average yield, standard deviation and coefficients of correlation for each stage of the synthesis.

Code	Obtaining Biomass (%)			Cellulose Extraction (%)			Cellulose Deoxidation (%)			Catalytic Pyrolysis (%)			Acid Purification (%)			Overall Yield (wt.%)
	Yield	SD	CV	Yield	SD	CV	Yield	SD	CV	Yield	SD	CV	Yield	SD	CV
CG-ND-ND	13.39	0.03	0.21	-	-	-	-	-	-	27.39	0.08	0.31	-	-	-	3.67
CG-ND-DW	13.21	0.08	0.61	-	-	-	84.07	7.48	8.90	32.58	0.40	1.22	84.57	0.21	0.25	3.06
CG-DCAP-FeCl	13.38	0.23	1.72	44.01	0.11	0.24	75.81	1.01	1.34	27.94	0.39	1.41	94.26	0.27	0.29	1.18
CG-DCAP-KFe	13.27	0.08	0.58	45.13	0.43	0.95	94.70	1.15	1.22	34.07	0.11	0.33	41.74	0.47	1.12	0.81
CG-DUAP-FeCl	13.26	0.07	0.54	35.15	0.45	1.27	75.08	0.76	1.01	27.26	0.46	1.69	74.67	0.50	0.67	0.71
CG-DUAP-KFe	13.17	0.21	1.57	34.86	0.33	0.94	87.02	5.19	5.96	31.54	0.11	0.34	44.82	3.04	6.79	0.56
CC-FeCl	-	-	-	-	-	-	79.60	3.09	3.88	21.96	0.46	2.10	73.98	0.43	0.58	1.71
CC-KFe	-	-	-	-	-	-	90.54	4.81	5.31	20.30	0.52	2.59	63.11	2.11	3.35	1.55
CG	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
OG	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-

shows the yield of each stage and the overall yield, along with their respective standard deviations and coefficients of variation, for the different synthesized carbonaceous materials. The low values of standard deviation and coefficient of variation indicate the high reproducibility and reliability of the synthesis process. This suggests effective control over operational variables, minimizing fluctuations in the quality of the materials obtained. The consistency of these parameters is crucial to ensuring uniform physicochemical properties of carbonaceous materials.

The average overall yield for obtaining dry powdered biomass from the collected king grass was 13.28%, with standard deviations ≤ 0.23% and coefficients of variation ≤ 1.72%. The yield in biomass production shows that this stage is homogeneous and does not have a significant impact on subsequent yield variations among the different treatments. This consistency indicates that the differences observed in later stages are not due to variability in the initial biomass quantity, allowing focus on the delignification, deoxygenation, and pyrolysis processes to better understand the discrepancies in overall yield.

On the other hand, the average yield for delignification by the conventional alkaline peroxide method and ultrasound-assisted method was 44.57% and 35.01%, respectively, both with significantly low standard deviations and coefficients of variation (≤0.45% and 1.27%, respectively). Although there are no reports in the literature on the yields of biomass recovery and delignification for this type of lignocellulosic biomass, the results obtained are comparable to those reported for other types of biomass [8,25,27,28].

The delignification stage is fundamental to optimizing efficiency in the production of carbonaceous materials. Conventional alkaline peroxide delignification methods (DCAP) show notably higher yields compared to ultrasound-assisted methods (DUAP), with values ranging from 44.01% to 45.13% for DCAP versus 35.15% to 34.86% for DUAP. These results indicate that, although sonication represents a promising strategy by reducing processing time and reagent consumption, its effectiveness in terms of final solid yield may be limited in certain cases. This could be explained by the fact that, while ultrasound facilitates the disruption of the cell wall, which is mainly composed of lignin, and promotes cellulose release, prolonged exposure to extreme mechanical energy and cavitation conditions can induce partial degradation of the material, resulting in loss of the final solid mass. Additionally, the higher energy generated during sonication may favor the dissolution of certain components into the extraction medium, which also contributes to the decrease in the mass of the recovered solid biomass [24,35,36].

The microwave-assisted deoxygenation stage also shows notable differences. Deoxygenation yields are higher when K₃[Fe(CN)₆] is used as a catalyst, reaching values up to 94.70% in CG-DCAP-KFe and 91.54% in CC-KFe, compared to materials treated with FeCl₃, which present yields ranging from 75.08% to 79.60%. These results suggest that K₃[Fe(CN)₆] is a more efficient catalyst for oxygen removal under microwave conditions, possibly due to its ability to form more stable complexes or greater interaction with oxygenated functional groups present in the biomass. This has important implications for reducing the amount of residual oxygen in carbonaceous materials, which could likely improve their physical and chemical properties, such as conductivity and thermal stability [37]. Therefore, catalyst selection is a determining factor in enhancing deoxygenation yield, and K₃[Fe(CN)₆] appears to be the most efficient option for this study.

Microwave-assisted synthesis offers several advantages over traditional thermal treatments for the preparation of carbon-based catalysts. While single-layer graphene can be typically produced at temperatures as high as 3000 °C, microwave radiation enables a higher degree of graphitization at significantly lower temperatures, often below 1000 °C. This enhanced effect may be attributed to the interaction of microwave energy with metal oxides present in the material, which induces local distortion through electron migration. Such mechanisms improve the conversion of electromagnetic energy into thermal energy, rapidly increasing local temperatures by several tens to hundreds of degrees.

Unlike conventional heating, microwave radiation generates an alternating electromagnetic field that interacts directly with polar molecules and conductive particles within the material. This field causes molecular realignment and friction-like effects that efficiently convert electromagnetic energy into heat. As a result, the material is heated more uniformly and rapidly from the inside, enhancing reaction kinetics.

Microwave-assisted synthesis offers several advantages over traditional thermal treatments for the preparation of carbon-based catalysts. While single-layer graphene can typically be produced at temperatures as high as 3000 °C, microwave radiation enables a higher degree of graphitization at significantly lower temperatures, frequently close to 1000 °C. This enhanced effect may be attributed to the interaction of microwave energy with metal oxides present in the material, which induces local distortion through electron migration. Such mechanisms improve the conversion of electromagnetic energy into thermal energy, rapidly increasing local temperatures by several tens to hundreds of degrees.

Microwave radiation also leads to significant improvements in the development of porous structures. The rapid and localized heating promotes the formation of abundant micropores with a suitable size distribution for ion transport, which is particularly beneficial for catalytic and electrochemical applications. By extending the irradiation time, further pore expansion can be achieved, increasing micropore diameters and in some cases forming small mesopores due to the collapse of unstable pore walls under high localized temperatures. Therefore, microwave-assisted processes not only reduce energy requirements and synthesis time but also improve structural properties such as surface area, graphitization degree, and pore architecture, making them a promising alternative to traditional carbonization methods [37,38,39,40,41,42].

Catalytic pyrolysis, both primary and secondary, shows a clear dependence on the previous deoxygenation treatment of the biomass. The materials obtained from extracted cellulose exhibit higher yields compared to those derived from commercial cellulose, as observed in CC-FeCl and CC-KFe, with yields of 21.96% and 20.30%, respectively. In contrast, the delignified materials, such as CG-DCAP-KFe and CG-DCAP-FeCl, show yields of 27.94% and 34.07%, respectively. This indicates that lignin removal not only facilitates the thermal decomposition of biomass but can also enhance the carbonization of the material, allowing for greater conversion of precursors into carbon. This behavior is likely due to the removal of structural barriers imposed by lignin, which enables a more controlled and efficient degradation of cellulose and hemicellulose during pyrolysis. The reduced yield in non-delignified materials also suggests a higher retention of volatile compounds and non-carbonizable residues.

In the purification stage (washing and drying), the yields are notably higher for the materials treated with FeCl₃, such as CG-DCAP-FeCl (94.26%) and CC-FeCl (73.98%), compared to those treated with K₃[Fe(CN)₆], such as CG-DCAP-KFe (41.74%) and CG-DUAP-KFe (44.82%). This difference may be due to FeCl₃ promoting the formation of more stable products that are less susceptible to being removed during washing. On the other hand, the use of K₃[Fe(CN)₆] may induce the formation of secondary products that, although they enhance the deoxygenation process, result in greater material loss during washing. This is particularly associated with the reduction in particle size, which could be related to the more reactive nature of the catalyst or to a greater formation of ashes. This difference in the stability of the final products has implications for the purity of the carbonaceous material obtained, where FeCl₃ appears to offer an advantage in terms of material retention.

The analysis of the overall yield reveals that the materials with the highest efficiencies are CG-ND-DW and CG-ND-DW, which were not subjected to delignification or catalytic pyrolysis processes. This suggests that, in certain cases, process simplicity may compensate for the losses associated with additional chemical treatment steps. However, the final characteristics of the material and its intended application will provide complementary information to support the selection of the most suitable synthesis route, establishing a balance between cost, process simplicity, and application.

On the other hand, it is interesting to note that among the materials subjected to deoxygenation and pyrolysis process, those obtained from commercial cellulose (CC-FeCl and CC-KFe) exhibited the highest overall yields. In other words, optimizing the production process of carbonaceous materials from biomass largely depends on the choice of delignification method and the catalyst used for deoxygenation [14,17]. Although ultrasound-assisted delignification may offer advantages in terms of sustainability, its lower yields compared to the conventional process limit its efficiency.

3.2. Physicochemical Characterization of the Carbonaceous Materials

3.2.1. Textural Properties

One of the most striking aspects of the results is the notable variability in the textural properties of the materials (

Table 4. Physicochemical properties of materials.

Code	BET Surface Area (m2/g)	Average Pore Volume BJH (cm3/g)	Average Pore Diameter (nm)	Fe (wt.%) AAS	K (wt.%) AAS	pH Dispersion	Zeta Potential (mV)
CG-ND-ND	42.73	0.028	8.97	-	-	10.12	−34.0
CG-ND-DW	275.89	0.082	6.75	-	-	9.93	−28.8
CG-DCAP-FeCl	400.22	0.123	6.32	3.76	-	6.63	−37.3
CG-DCAP-KFe	449.06	0.124	4.92	5.90	2.73	9.99	−38.0
CG-DUAP-FeCl	388.41	0.123	5.10	4.22	-	5.51	−24.8
CG-DUAP-KFe	414.29	0.110	4.40	6.02	2.75	10.20	−37.4
CC-FeCl	314.22	0.098	3.97	5.90	-	6.59	−31.2
CC-KFe	714.50	0.140	3.18	8.06	1.88	9.93	−39.4
CG	49.03	0.096	11.97	-	-	8.15	−47.9

). The values range from 42.73 m²/g for CG-ND-ND to 714.50 m²/g for CC-KFe. Materials treated with K₃[Fe(CN)₆] consistently exhibit the highest specific surface areas. This is evident in CC-KFe (714.50 m²/g) and CG-DCAP-KFe (449.06 m²/g), suggesting that this catalyst promotes greater microporosity or mesoporosity during the deoxygenation and pyrolysis processes. In contrast, the material that was not subjected to delignification or deoxygenation, such as CG-ND-ND, shows the lowest specific surface area at only 42.73 m²/g, indicating that the residual lignin and lack of effective oxidation limit pore formation.

According to the nitrogen adsorption-desorption isotherms (Figure 3), materials treated with K₃[Fe(CN)₆], such as CG-DCAP-KFe, CG-DUAP-KFe, and CC-KFe, exhibit type IV isotherms (IUPAC) with H4-type hysteresis loops, characteristic of materials combining micropores with narrow mesopores. This matches the average pore diameters between 3.18 and 4.92 nm and the high surface areas (>400 m²/g), indicating a hierarchical structure dominated by narrow mesopores and significant microporosity. In contrast, FeCl₃-treated materials like CG-DCAP-FeCl and CG-DUAP-FeCl also display type IV isotherms but with H3-type hysteresis loops, associated with wider mesopores and ink-bottle-shaped pores, reflected in larger average pore diameters (~5–6 nm). Finally, materials without delignification or without catalyst, such as CG-ND-ND and CG-ND-DW, show isotherms intermediate between type II and IV, with minimal or no visible hysteresis and larger pore diameters (>6.75–8.97 nm), suggesting predominantly macroporous or wide mesoporous structures with low microporosity and low surface area.

The pore size distribution curves (Figure 4) reveal significant differences depending on the catalyst type and pretreatment applied. Materials synthesized with K₃[Fe(CN)₆], such as CG-DCAP-KFe, CG-DUAP-KFe, and CC-KFe, show broader, often bimodal or multimodal distributions with main peaks in the narrow mesopore region (~2–5 nm) and extended tails toward micropores (<2 nm) and wider mesopores (~5–8 nm). This hierarchical porosity contributes to high surface areas (up to 714.50 m²/g) and enhanced accessibility for catalytic and adsorption processes.

In contrast, materials activated with FeCl₃ (CG-DCAP-FeCl, CG-DUAP-FeCl, CC-FeCl) exhibit narrower distributions centered around ~4–6 nm, indicating a predominance of wider mesopores and lower microporosity, consistent with their lower BET areas and slightly larger average pore diameters (≈4–6 nm).

Finally, materials without delignification or catalyst (CG-ND-ND, CG-ND-DW, CG) show distributions shifted toward larger diameters (~6–9 nm and even >10 nm) with narrower, asymmetric profiles, suggesting mainly macroporosity or uncontrolled wide mesoporosity, matching their low BET areas and pore volumes [12,26,29].

This hierarchical pore structure in K₃[Fe(CN)₆]-treated materials is directly related to the catalyst’s dual effect: Fe promotes graphitization, while potassium chemically activates the carbon, creating cavitation and expanding the porous network during pyrolysis. Thus, the coexistence of micropores and narrow mesopores results in more reactive and versatile materials, whereas materials without activation remain denser, with larger, less reactive pores.

The literature reports that the specific surface area of carbonaceous materials synthesized from lignocellulosic biomass or through thermal treatments such as pyrolysis can vary significantly [1,2,13,17]. For example, studies utilizing methods like chemical activation with KOH or ZnCl₂, or controlled pyrolysis, often report surface areas exceeding 500 m²/g [1,2,13,17]. The highest values obtained in this work, such as for CC-KFe (714.50 m²/g), align with these studies, suggesting that the use of K₃[Fe(CN)₆] as a deoxidizing agent favors the development of a more advanced microporous structure, thereby increasing the surface area. In contrast, non-delignified materials such as CG-ND-ND exhibit significantly lower surface areas (42.73 m²/g), indicating that delignification and chemical activation play a critical role in enhancing the surface accessibility of the material.

A comparison with the literature indicates that pore volumes for biomass-derived carbonaceous materials may range between 0.10 and 0.30 cm³/g, depending on synthesis conditions and applied treatments [1,2,13]. The relatively high value observed for CC-KFe suggests that activation and catalysis with K₃[Fe(CN)₆] also contributes to the generation of a more porous structure, consistent with studies reporting the ability of potassium-containing activating agents such as KOH, K₂CO₃, or KNO₃ to enhance microporosity and pore volume in carbonaceous materials derived from diverse biomass sources [17,31,38,43]. For example, KOH has been used to remove amorphous carbon from rice husk, coconut shell, corncob, straw, and black sesame, among others, to induce porosity in carbonaceous materials at varying temperatures [1,17,31,38,43].

Regarding pore diameter, values below 10 nm observed in nearly all materials indicate predominantly mesoporous structures. However, the CG material exhibited a pore diameter of 11.97 nm, suggesting a macroporous architecture, potentially linked to the structural properties of the cellulose feedstock employed in the synthesis process. Prior studies utilizing lignocellulosic biomass for graphene-like material synthesis report analogous surface area and pore size characteristics, particularly when chemical or thermal activation methods are implemented [11,30,44,45].

The obtained results demonstrate that the utilization of K₃[Fe(CN)₆] as a catalyst produces materials with enhanced porosity compared to FeCl₃ or the absence of a catalyst. This divergence arises from distinct chemical and structural mechanisms promoted during pyrolysis. K₃[Fe(CN)₆] acts simultaneously as a graphitization catalyst and an activating agent. During pyrolysis at 1000 °C, it thermally decomposes into reactive iron (Fe) and potassium (K) species. Fe forms metallic nanoparticles that serve as templates for carbon reorganization into graphitic sheets, thereby reducing the activation energy required for graphitization. Conversely, K generates gases (e.g., CO₂, H₂O) during its oxidation or reaction with carbon, inducing microporosity and increasing the surface area phenomenon well-documented in alkali activation processes. This synergistic effect explains the formation of ordered and porous structures in samples such as CG-DCAP-KFe and CG-DUAP-KFe. In contrast, FeCl₃ decomposes into Fe and Cl₂. Although Fe partially catalyzes graphitization, the absence of K limits activation, resulting in amorphous or semicrystalline materials with reduced porosity (CG-DCAP-FeCl, CG-DUAP-FeCl). Furthermore, Cl₂ may introduce defects within the carbon network, inhibiting the development of extended graphitic domains. In the absence of a catalyst (CG-ND-DW), the lack of agents promoting structural reorganization leads to entirely amorphous carbons [1,17,31,38,43].

3.2.2. Fe and K Content

The Fe and K content, measured by AAS, provides important information about the presence of these elements in the resulting carbonaceous materials, which is crucial for various applications, including catalytic and carbocatalytic processes [26,28]. Materials treated with K₃[Fe(CN)₆] show higher Fe contents, as seen in CC-KFe with 8.06 wt.% Fe and CG-DUAP-KFe with 6.02 wt.%. This suggests greater retention of this element within the material structure during pyrolysis, which could enhance its catalytic properties. In contrast, materials treated with FeCl₃, such as CG-DUAP-FeCl (4.22 wt.% Fe) and CG-DCAP-FeCl (3.76 wt.% Fe), display lower Fe contents, indicating a lower efficiency in retaining this element within the carbon structure. The difference in Fe retention between K₃[Fe(CN)₆] and FeCl₃ may be attributed to the greater stability of the complexes formed from K₃[Fe(CN)₆], which favors the integration of the metal into the carbon matrix. This result aligns with studies showing that the inclusion of Fe in graphene can enhance its catalytic activity, particularly in applications such as oxygen reduction [46,47].

3.2.3. Dispersion pH and Zeta Potential

The pH of the material dispersions is a key parameter for determining their acidic or basic nature in solution, which directly influences their applications in adsorption, catalysis, and carbocatalysis [26,28]. The dispersion pH of the materials ranges from 5.51 for CG-DUAP-FeCl to 10.20 for CG-DUAP-KFe. In the literature, carbonaceous materials typically exhibit an alkaline pH when functionalized with surface oxygen-containing groups or when they contain metals such as potassium [48,49]. The pH values of materials treated with K₃[Fe(CN)₆] tend to be higher (close to 10), indicating that these materials may have a greater content of basic functional groups, such as potassium oxides, which increase the dispersion pH.

Materials treated with K₃[Fe(CN)₆] exhibit relatively higher pH values, such as CG-DUAP-KFe (10.20) and CG-ND-ND (10.12), indicating a more basic nature, likely related to the retention of basic oxygenated functional groups or the presence of potassium species. In contrast, materials treated with FeCl₃ tend to show lower pH values, such as CG-DUAP-FeCl (5.51) and CG-DCAP-FeCl (6.63), suggesting a greater retention of acidic species, possibly associated with the incorporation of Fe into the carbon matrix. These results are relevant for applications in which the material’s reactivity is influenced by the pH environment, such as in contaminant adsorption processes or acid–base catalysis [29,39,40].

Zeta potential is a measure of the electrostatic behavior of materials in dispersion and provides information about colloidal stability and particle–particle interactions [50]. Materials with the most negative Zeta potentials, such as CC-KFe (−39.4 mV) and CG-DCAP-KFe (−38.0 mV), tend to be more stable in dispersion compared to others. However, when dispersed in water, they do not form stable suspensions over time. In comparison, materials treated with FeCl₃ tend to exhibit less negative Zeta potentials, such as CG-DUAP-FeCl (−24.8 mV), indicating lower colloidal stability.

3.2.4. FT-IR Analysis

The FTIR spectra of the synthesized carbonaceous materials (Figure 5) reveal three main spectral regions exhibiting significant differences in the presence and intensity of absorption bands. In the region between 3600–375 cm⁻¹, the CC-KFe material displays two well-defined bands that can be attributed to the stretching vibrations of free or terminal hydroxyl groups (–OH) associated with surface defects in the carbon framework. These bands appear with lower intensity in the CC-FeCl material, while in the CG-DCAP-KFe and CG-DCAP-FeCl samples, their intensity decreases markedly and becomes completely absent in the CG-ND-ND, CG-DUAP-FeCl, and CG-DUAP-KFe materials. This progressive attenuation suggests a reduction in the surface density of oxygen-containing functional groups, likely due to the combined influence of the precursor type (gel-based or carbonized), the presence or absence of catalytic agents, and the thermal treatment applied prior to pyrolysis.

In contrast, a broad and intense band centered around 3400 cm⁻¹ appears exclusively in the CC-FeCl material. This feature may be related to collective vibrational modes arising from structural defects or residual hydrogen bonding interactions among oxygenated functional groups at highly defective sites within the graphitic network.

In the region near 2300 cm⁻¹, a weak band is observed in the CG-ND-ND, CG-DCAP-KFe, CG-DCAP-FeCl, CC-FeCl, and CC-KFe materials, and is almost imperceptible in CG-DUAP-FeCl and CG-DUAP-KFe. This band can be associated with stretching vibrations of C≡C or C≡N bonds, indicating the possible formation of multiple bonds or nitrogen-containing defect sites during synthesis, particularly in the presence of nitrogenous precursors or residual metal salts [13,17,39,51].

In the 1650 and 1500 cm⁻¹ regions, medium- and high-intensity bands, respectively, are identified. The band at 1650 cm⁻¹, present in both CC-KFe and CC-FeCl (although with lower intensity in the former), is attributed to the stretching vibrations of conjugated carbonyl (C=O) groups or surface defects. The 1500 cm⁻¹ band likely corresponds to C=C vibrational modes within extended aromatic domains, characteristic of partially ordered graphene-like structures. In the CG-DUAP-KFe, CG-DUAP-FeCl, CG-DCAP-KFe, CG-DCAP-FeCl, and CG-ND-ND materials, the 1500 cm⁻¹ band is present but with lower relative intensity, suggesting a reduced density of graphitic structures or a lower degree of aromatic conjugation. This difference is likely a consequence of the precursor type, the activation pathway, and interactions with the catalyst [13,17,39,51].

Finally, in the 1100 cm⁻¹ region, an intense band is detected in the CG-DCAP-FeCl sample, with lower intensities observed in CG-DCAP-KFe, CG-DUAP-FeCl, and CG-DUAP-KFe. This band can be assigned to C–O or C–C vibrational modes within deformed rings or ether/epoxide-type defects in the carbon network. Its presence appears to strongly depend on catalyst interactions and the chemical environment of the precursor during the pre-pyrolysis stages. The absence of this band in CC-KFe, CC-FeCl, and CG-ND-ND supports the hypothesis that these materials possess a more homogeneous and less functionalized carbon network, possibly due to more efficient heteroatom removal during carbonization or reduced formation of surface defects [13,17,39,51].

3.2.5. H₂-TPR Analysis

Figure 6 and Figure 7 present the general and deconvoluted H₂-TPR profiles of the materials, respectively, providing insight into their reducibility. Furthermore,

Table 5. H₂ consumption associated with each reduction event for the different materials.

Code	Reduction Temperature (°C)					Hydrogen Consumption (mmol H2/g Material)					Total
	T1	T2	T3	T4	T5	T1	T2	T3	T4	T5
CG-ND-ND	-	-	-	-	-	-	-	-	-	-	-
CG-ND-DW	-	-	-	-	-	-	-	-	-	-	-
CG-DCAP-FeCl	526.0	697.2	610.3	-	-	0.005	0.005	0.005	-	-	0.015
CG-DCAP-KFe	337.1	459.6	576.6	810.6	-	0.005	0.005	0.005	0.005	-	0.020
CG-DUAP-FeCl	536.8	715.5	-	-	-	0.061	0.007	-	-	-	0.068
CG-DUAP-KFe	544.1	601.5	708.2	-	-	0.032	0.008	0.009	-	-	0.049
CC-FeCl	462.3	554.3	687.2	823.5	-	0.015	0.023	0.011	0.010	-	0.059
CC-KFe	470.0	540.7	576.1	-	-	0.005	0.005	0.005	-	-	0.015
CG	506.4	586.7	-	-	-	0.009	0.010	-	-	-	0.019
GO	474.5	542.2	608.7	705.5	-	0.042	0.027	0.043	0.023	-	0.134

summarizes the H₂ consumption associated with each reduction event, enabling a quantitative comparison of the materials reduction behavior. The comparison between the synthesized materials with CG and GO revealed fundamental differences in their redox properties and, consequently, in their surface composition, resulting in advantages and/or limitations depending on the potential applications for which they may be intended.

CG exhibits two reduction peaks at 506.4 °C and 586.7 °C, with a total hydrogen consumption of 0.019 mmol H₂/gCat. This low consumption reflects its high purity and highly graphitic structure, lacking significant oxygenated functional groups or metallic species. The observed peaks could be attributed to residual structural defects or trace impurities adsorbed during its industrial synthesis, as reported for non-functionalized graphenes [26]. The thermal stability and absence of redox-active sites make CG ideal for applications requiring high electrical conductivity, although these features may limit its usefulness in catalytic processes that depend on surface interactions. In contrast, OG shows four reduction peaks (between 474.5 and 705.5 °C) and an exceptionally high H₂ consumption (0.134 mmol/g), the highest among all materials. This behavior is typical of oxidized graphene, where epoxy, hydroxyl, and carbonyl groups are reduced in a stepwise manner [34]. For instance, the peak at 474.5 °C corresponds to epoxy group reduction, while the higher-temperature peaks (542–705 °C) reflect the decomposition of carbonyls and more stable oxidized structures. OG represents a highly functionalized material, ideal for applications requiring chemical modification or nanoparticle anchoring, although its conductivity is lower than that of pure graphene [26,52].

A relevant finding regarding the synthesized materials is the consistent correlation between the TPR profiles and the Fe and K contents quantified by AAS, confirming the direct influence of metal composition on the surface reactivity of the materials. Thus, materials treated with K₃[Fe(CN)₆], such as CG-DCAP-KFe and CG-DUAP-KFe, showed reduction peaks at lower temperatures (337.1–708.2 °C) compared to those treated with FeCl₃, such as CG-DCAP-FeCl and CG-DUAP-FeCl, whose peaks were located between 526.0 and 715.5 °C. Similarly, the materials CG-DCAP-KFe, CG-DUAP-KFe, and CC-KFe exhibited higher Fe contents (5.90–8.06 wt.%) and K contents (1.88–2.75 wt.%) compared to CG-DCAP-FeCl and CG-DUAP-FeCl (3.76–6.02 wt.% Fe; K not detected). The simultaneous presence of Fe and K in CG-DCAP-KFe and CG-DUAP-KFe is associated with TPR profiles at lower reduction temperatures ([53] between 337–708 °C) and moderate total H₂ consumption (0.020–0.049 mmol/g), suggesting a synergistic effect between the two metals. In other words, the use of K₃[Fe(CN)₆] facilitates the formation of reducible species at lower temperatures. According to the literature, the peaks at lower temperatures (around 337 °C) could correspond to the reduction Sof Fe₂O₃ to Fe₃O₄ [54], while those between 460–577 °C may reflect the reduction of Fe₃O₄ to metallic Fe processes that are promoted by the presence of K [55,56]. Previous studies have shown that K acts as a promoter by lowering the activation energy for Fe oxide reduction, facilitating electron transfer, and modifying the distribution of metallic species [53,57,58].

In contrast, the materials synthesized with FeCl₃ (CG-DCAP-FeCl, CG-DUAP-FeCl, and CC-FeCl) show reduction peaks at higher temperatures (526–823 °C) and variable H₂ consumption values (0.015–0.068 mmol/g). In particular, CG-DUAP-FeCl exhibits the highest H₂ consumption (0.068 mmol/g), which could be related to the reduction of highly dispersed or more crystalline iron oxides, possibly due to ultrasound-assisted delignification that increases the accessibility of active sites. The presence of a peak at 715 °C in this material may be associated with the reduction of iron oxides encapsulated within the carbon matrix, a phenomenon reported in biomass-derived carbons treated with intensive physical methods [53,57,58].

The materials derived from commercial cellulose (CC-FeCl and CC-KFe) exhibit divergent behaviors. CC-FeCl shows four reduction peaks (462–823 °C) and a relatively high H₂ consumption (0.059 mmol/g), indicating a greater diversity of reducible functional groups. In contrast, CC-KFe, despite its high Fe content (8.06 wt. %), registers a low H₂ consumption (0.015 mmol/g), possibly due to the formation of less accessible Fe phases or to particle sintering during pyrolysis—a phenomenon accentuated by the presence of K. This discrepancy highlights the importance of the interaction between the carbon precursor and the catalyst in the formation of carbonaceous materials with distinct structures and compositions.

The total hydrogen consumption was higher in materials treated with ultrasound and FeCl₃, such as CG-DUAP-FeCl (0.068 mmol H₂/g material), indicating a greater amount of reducible oxidized species. This result correlates with the Fe content determined by AAS, where CG-DUAP-FeCl showed 4.22 wt.% Fe. On the other hand, CG-DUAP-KFe, with a Fe content of 6.02 wt.% and K content of 2.75 wt.%, showed a total hydrogen consumption of 0.049 mmol H₂/g, suggesting that the presence of K may influence the reducibility of the formed species. Materials derived from commercial cellulose also reflected this trend. CC-FeCl, with an Fe content of 5.90 wt.%, exhibited a total hydrogen consumption of 0.059 mmol H₂/g, whereas CC-KFe, with an Fe content of 8.06 wt.% and K content of 1.88 wt.%, showed a significantly lower consumption of 0.015 mmol H₂/g. This could indicate that despite the higher Fe content, the presence of K in CC-KFe may stabilize certain species, making them less susceptible to reduction. The presence of a peak at high temperature (823.5 °C) indicates the existence of more stable and harder-to-reduce functional groups. In contrast, CC-KFe showed peaks at lower temperatures (470.0, 540.7, and 576.1 °C), suggesting a lower amount of oxygenated groups or a more ordered structure.

The results indicate that combining ultrasound-assisted alkaline peroxide delignification with K₃[Fe(CN)₆] as a catalyst (CG-DUAP-KFe) yields materials with enhanced reducibility and a favorable distribution of metallic species or oxygenated functional groups. However, the presence of residual metal oxides suggests that, depending on the intended application, an additional purification step may be required to eliminate inorganic species and optimize the carbonaceous structure. Recent studies propose post-pyrolysis acid treatments for this purpose, which have been shown to produce materials with high conductivity and elevated surface areas [8,13].

The TPR profiles and metal contents indicate that the choice of catalyst and carbon source significantly influences the redox properties of the resulting carbonaceous materials. Furthermore, the combination of K₃[Fe(CN)₆] treatment and ultrasound appear to promote the formation of more readily reducible species, whereas FeCl₃ tends to yield structures that are more stable and resistant to reduction. Previous studies have demonstrated that iron-based catalysts, such as FeCl₃, facilitate biomass graphitization at relatively low temperatures. For instance, the in-situ generation of Fe₃C nanoparticles has been reported to catalyze the formation of interconnected graphitic nanotubes from lignocellulosic biomass [41,43,59]. Additionally, the combination of chemical treatments and hydrothermal carbonization in the presence of iron has been shown to enhance the properties of biomass-derived carbonaceous materials [41,43,59].

3.2.6. XRD Analysis

The XRD analysis of the synthesized carbonaceous materials (Figure 8) reveals notable structural differences, including in the crystalline phase, degree of crystallinity, and crystallite size, which depend on both the catalyst type and the pretreatments applied (

Table 6. Structural parameters from XRD and Raman for commercial graphene and synthesized materials.

Code	XRD			Raman
	Angle 2θ	Crystallite Size (nm)	Crystallinity (%)	Allotropic Form	Intensity Band Ratio		Layer Number
					I2D/IG	ID/IG
CG-ND-ND	21.96	0.43	14.0	Amorphous carbon	0	1.12	-
CG-ND-DW	21.96	0.38	6.30	Amorphous carbon	0	1.15	-
CG-DCAP-FeCl	21.78	0.33	15.0	Amorphous carbon	0	1.13	-
CG-DCAP-KFe	26.31	0.11	24.3	Graphene-like material	0.80	0.37	2
CG-DUAP-FeCl	21.86	0.25	12.8	Amorphous carbon	0	1.16	-
CG-DUAP-KFe	26.43	0.11	21.3	Graphene-like material	0.32	1.11	3
CC-FeCl	25.94	0.11	18.8	Amorphous carbon	0	1.20	-
CC-KFe	26.41	0.23	24.0	Graphene-like material	0.46	0.88	3
CG	26.59	0.26	54.0	Graphene	0.39	0.09	3

). The reference material, CG, exhibits a highly ordered crystalline structure with peaks at 2θ = 26.59°, 42.32°, and 44.46°, and an additional peak at 54.50°, characteristic of well-organized graphitic materials [12].

Solids lacking delignification and deoxidation treatments, such as CG-ND-ND, exhibit a characteristic structure of predominantly amorphous carbonaceous materials, with poorly defined peaks at 2θ = 21.96°, 28.41°, 30.33°, and 40.64°. The absence of the (002) plane peak near 2θ ≈ 26° aligns with the lack of crystalline organization, suggesting that this treatment does not induce the formation of significant graphene- or graphite-like crystalline carbon phases. This observation may arise from the absence of catalytic species required to promote molecular-level structural ordering [13,60]. Similarly, in the CG-ND-DW sample—subjected to deoxidation without catalyst addition—the (002) peak near 2θ ≈ 26° is also absent. The low crystallinity of this sample underscores the critical role of the catalyst in carbon structural organization, as CG-ND-DW displayed fewer peaks (2θ = 21.82° and 36.00°) and a reduced structural order. This indicates that deoxidation without a catalyst similarly fails to generate significant crystalline carbon phases, retaining an amorphous and poorly defined structure [13,51].

On the other hand, the absence of the characteristic peak at 2θ ≈ 26° (002 plane) in all materials synthesized with FeCl₃ during the deoxidation stage—except for CC-FeCl—suggests that this catalyst did not promote the formation of crystalline carbon structures. These results contrast with those reported by Zhang et al. [60], who obtained high-quality graphene using glucose as the carbon precursor and FeCl₃ under carbonization at 700 °C. Nonetheless, the appearance of this peak in the case of CC-FeCl indicates that the carbon source also influences the structural development of the material. Even so, this material retained features typical of an amorphous phase. Additionally, a trend toward the formation of specific crystalline Fe phases was observed, particularly in the case of CC-FeCl. This material exhibited several characteristic peaks at 2θ = 30.21°, 35.58°, 43.05°, 44.16°, 44.76°, and 46.03°, mainly associated with iron oxides such as Fe₃O₄ or Fe₂O₃, which are common in systems where FeCl₃ was used. The peaks around 35° and 44° support this finding, indicating a crystalline structure influenced by the Fe content, which is considerably high in this sample (5.90 wt.% Fe). The samples CG-DCAP-FeCl and CG-DUAP-FeCl, although showing fewer diffraction peaks, exhibited similar patterns that suggest the presence of Fe compounds, but in lower quantity and order, which could be related to differences in the pretreatment methods.

In contrast, the materials synthesized with K₃[Fe(CN)₆] exhibited a structural organization more similar to that of CG, as evidenced particularly by the presence of peaks at 2θ around 26° and 44°. These peaks are characteristic of graphene-like structures, indicating that the use of K₃[Fe(CN)₆] as a deoxidation catalyst promotes a more ordered arrangement of carbon layers. This suggests that these materials may have a structure comparable to commercial graphene, at least in terms of carbon layer organization—an important achievement in the synthesis of graphene-type materials from biomass. The CG-DCAP-KFe and CG-DUAP-KFe samples showed peaks at 2θ = 30.13°, 35.64°, 43.68°, and 44.60°, while CC-KFe displayed even more peaks at similar angles, with additional reflections at 37.70°, 42.81°, and 49.09°. The Fe and K content in these samples (e.g., 6.02 wt.% Fe and 2.75 wt.% K in CG-DUAP-KFe) may be promoting the formation of graphitic-like phases, in addition to minor crystalline compounds. These results suggest that K₃[Fe(CN)₆] not only contributes to the organization of carbon atoms into layered structures but may also promote the formation of minor potassium-based crystalline phases, distinguishing these materials from those treated with FeCl₃.

The results obtained are consistent with the structures and morphologies revealed in the HRTEM micrographs analysis described in Section 3.2.8 (HRTEM Analysis), and also support the patterns observed in the Raman spectra (Figure 6), which will be discussed in later sections. Taken together, these findings confirm that the type of catalyst used plays a decisive role in the structural organization of the synthesized carbonaceous materials. In this context, the use of K₃[Fe(CN)₆] proves to be particularly effective in promoting the formation of graphene-like structures in materials derived from Pasto king grass biomass. In contrast, the FeCl₃ catalyst tends to favor the formation of amorphous materials, as well as the presence of crystalline phases associated with iron oxides. These results highlight the high potential of K₃[Fe(CN)₆] as a structuring agent in the synthesis of graphene-type materials under the synthesis conditions applied in this study and underscore the importance of proper catalyst selection according to the desired crystalline architecture in the materials.

Regarding the crystallite size (determined from the peak at 2θ ≈ 26°, plane 002) and peak shape, the carbonaceous materials classified as graphene-type (CG-DCAP-KFe, CG-DUAP-KFe, CC-KFe, and CG) exhibited diffraction peaks at ~26.3–26.6° (002 plane), which are characteristic of graphitic structures. For instance, CG-DCAP-KFe showed a diffraction angle of 26.31°, with a crystallinity of 24.3% and a crystallite size of 0.11 nm, values close to commercial graphene (26.59°, 54% crystallinity, 0.26 nm). These results suggest an intermediate structural order between few-layer graphene (3 layers in CC-KFe) and graphite, consistent with studies that associate pyrolysis temperatures ≥ 1000 °C with the formation of graphitic domains in biomass-derived materials [13].

In contrast, the amorphous materials (CG-ND-ND, CG-DCAP-FeCl, CC-FeCl) exhibited broader peaks at ~21.8–21.96°, indicative of disordered structures with low crystallinity (6.3–15%). The absence of well-defined peaks in these cases is associated with the retention of heteroatoms (oxygen, hydrogen) and the lack of preferential layer orientation, a phenomenon commonly reported in pyrolyzed untreated biomass [61].

The use of K₃[Fe(CN)₆] as a catalyst promoted the formation of graphene in biomass-derived samples (CG-DCAP-KFe, CG-DUAP-KFe) and commercial cellulose (CC-KFe), with significant Fe contents (5.90–8.06%). This metal acts as a graphitization catalyst by lowering the energy barriers for carbon rearrangement, as previously reported in iron-assisted graphene synthesis from lignin [16]. In contrast, FeCl₃ showed lower efficacy, predominantly yielding amorphous carbon (CG-DCAP-FeCl, CC-FeCl), likely due to its tendency to form oxides that inhibit structural ordering.

Ultrasound-assisted delignification (DUAP) did not significantly enhance crystallinity compared to the conventional method (DCAP). For instance, CG-DUAP-KFe achieved only 21.3% crystallinity, whereas CG-DCAP-KFe reached 24.3%, suggesting that ultrasonic energy may excessively fragment cellulose fibers, hindering the formation of continuous carbon layers. This outcome contrasts with studies in which ultrasound improves delignification in biomass [18,24], underscoring the need to optimize parameters such as exposure time.

According to our literature review, previous studies have also investigated the impact of catalysts such as FeCl₃ and Fe(NO₃)₃ on the synthesis of carbon-based materials and their crystallization into graphene-like structures [16,61]. These works agree on the significant influence of these compounds on the formation of both carbonaceous and iron oxide crystalline phases when used in the thermal processing of biomass or other carbon precursors. For instance, studies such as Zhang et al. [61], have demonstrated that FeCl₃ promotes the formation of iron oxide crystalline phases during the thermal treatment of carbon precursors, which is consistent with the results observed for the CC-FeCl, CG-DCAP-FeCl, and CG-DUAP-FeCl samples, where diffraction peaks were detected at characteristic positions for Fe₃O₄ and Fe₂O₃ (approximately at 2θ = 30°, 35°, and 44°). This behavior can be attributed to the partial decomposition and oxidation of FeCl₃ at high temperatures, which facilitates the integration of iron into the carbon matrix and enhances the material’s conductivity, an advantageous property for electrochemical and catalytic applications [26,29].

On the other hand, the use of K₃[Fe(CN)₆] as an activating agent and iron precursor has been scarcely explored. However, in the study conducted by Fan Jiang et al. [62], it was demonstrated that this compound not only induces the organization of carbon layers but also facilitates the formation of graphene-like structures with some crystalline features comparable to those observed in the present work. In our study, the samples synthesized with K₃[Fe(CN)₆], such as CG-DCAP-KFe, CG-DUAP-KFe, and CC-KFe, exhibited prominent diffraction peaks at 2θ ≈ 26° and 44°, which are associated with graphitic carbon phases and indicate higher crystallinity compared to the materials synthesized without a catalyst or with FeCl₃. This behavior is consistent with the findings of Fan Jiang et al. [62], who highlighted the ability of K₃[Fe(CN)₆] to promote the formation of well-organized structures, possibly due to the presence of both potassium and iron, which act as structuring agents during the pyrolysis process.

Similarly, several studies [38,43], have reported that the treatment of biomass with iron-based catalysts and potassium-containing activating agents can lead to the formation of crystalline phases of graphene-like compounds with a high degree of structural organization, as evidenced by diffraction peaks at positions similar to those of commercial graphene. This supports the peaks observed in the CG sample and in the K₃[Fe(CN)₆]-catalyzed samples in our work, suggesting that this compound is particularly effective in producing materials with an organized microstructure, a desirable outcome for energy storage applications such as supercapacitor electrodes and lithium ion batteries [29,43].

3.2.7. Raman Analysis

In Figure 9, the characteristic bands found in the Raman spectra of the obtained carbonaceous materials are shown. The D, G, and 2D bands are fundamental for characterizing the structure of this type of material. The D band, observed around 1360 cm⁻¹, is associated with structural defects (edges, vacancies, or disorder) and the presence of small domains in the sp² carbon network; it usually appears in materials with disordered or nanocrystalline structures. For example, this band is not observed in perfect monolayer graphene. The G band, located near 1580 cm⁻¹, is common in graphene- or graphite-based materials and represents the vibrations of the C–C bonds with sp² hybridization; its intensity reflects the amount of ordered sp² carbon domains. On the other hand, the 2D band corresponds to the second harmonic of the D band, generated by double-resonant phonon scattering; its shape and intensity depend on the number of layers and stacking of the graphitic material. This band is key to differentiating between graphene and graphite. The absence of this band in carbonaceous materials corresponds to an amorphous structure [63,64,65,66].

A lower value of the I_D/I_G ratio indicates fewer structural defects, fractures, or topological imperfections in the carbon network, which may be related to a higher degree of graphitization [64,65,67]. On the other hand, the I_2D/I_G ratio is a critical parameter to determine the type of carbonaceous material and the number of layers in graphite- or graphene-like materials, where values close to or greater than 1.6 indicate monolayer graphene, values around 0.8 suggest few-layer structures (~2 layers), values near 0.3 correspond to multilayer graphene (~3 layers), and a further reduction to approximately 0.07 reflects behavior similar to graphite [63,68,69,70]. Additionally, an I_2D/I_G ~ 0 indicates a disordered structure lacking regular layer stacking, a characteristic parameter for amorphous carbon [6].

The reference material CG exhibited an I_D/I_G ratio of 0.09 and an I_2D/I_G ratio of 0.39, indicating a high-quality graphitic structure with a low defect density. These values also suggest the presence of few-layer graphene (approximately three layers). The CG reference material showed Raman intensity ratios of I_D/I_G = 0.09 y I_2D/I_G = 0.39, values characteristic of multilayer graphene (approximately 3 layers) with high structural integrity and minimal defect density. These results confirm its highly ordered graphitic nature, consistent with standardized commercial materials, where a low I_D/I_G (<0.1) reflects a nearly defect-free crystalline network, while I_2D/I_G ratios between 0.3–0.5 are indicative of 3–5 layer stacking [63,64,65,66].

The Raman parameters in Table 6 show variations in the I_D/I_G and I_2D/I_G ratios depending on three factors: biomass pretreatment (conventional delignification, ultrasound, or untreated), carbon source (king grass biomass or commercial cellulose), and catalyst (FeCl₃, K₃[Fe(CN)₆], or none). As illustrated in Figure 6, the materials CG-ND-ND and CG-ND-DW, synthesized without delignification or catalyst, completely lack the 2D band (I_2D/I_G = 0) while exhibiting high I_D/I_G ratios (1.12 and 1.15, respectively). These results, consistent with XRD analysis (2θ ≈ 21.96°, crystallinity ≤ 15%), confirm the amorphous nature of these materials, characterized by a high density of structural defects and negligible graphitic organization.

In delignified, deoxidized, and pyrolyzed materials treated with FeCl₃ (CG-DCAP-FeCl, CG-DUAP-FeCl, and CC-FeCl), the 2D Raman band was not detected (I_2D/I_G = 0), and the I_D/I_G ratios ranged from 1.13 to 1.20, indicating a high density of structural defects. These results demonstrate that neither alkaline peroxide treatment (conventional or ultrasound-assisted) nor the use of FeCl₃ as a catalyst significantly enhanced graphitic ordering, preserving a predominantly amorphous structure. This behavior suggests that FeCl₃, likely due to its tendency to form metal oxides during pyrolysis, acts as a graphitization inhibitor by limiting carbon reorganization into ordered sheets [63,64,65,66].

Using K₃[Fe(CN)₆] instead of FeCl₃ led to 2D bands (I_2D/I_G = 0.32–0.80) and fewer defects (I_D/I_G = 0.37–1.11) in all samples. These features, particularly evident in CG-DCAP-KFe (I_2D/I_G = 0.37, 2 layers), confirm that K₃[Fe(CN)₆] acts as an effective promoter for the formation of graphitic domains. The presence of K⁺ ions, known to intercalate between carbon layers and facilitate their orderly stacking [6,11,12], explains the improvement in structural organization and the formation of 2–3 layer graphene. This finding aligns with previous studies in which the use of KOH as an activating agent and FeCl₃ as a catalyst enhanced crystallinity in carbons derived from paper cups [12].

When comparing carbon sources, materials derived from commercial cellulose (CC-FeCl and CC-KFe) replicate the catalyst dependency observed in king grass-derived analogs but exhibit inferior structural organization. For instance, CC-FeCl displayed a high defect density (I_D/I_G) 1.20) and absence of the 2D band (I_2D/I_G) 0), characteristic of amorphous carbons. Although the use of K₃[Fe(CN)₆] in CC-KFe reduced defects (I_D/I_G = 0.88) and induced a nascent 2D signal (I_2D/I_G = 0.46), its crystallinity (24.0%) and graphitic order remain lower than those of king grass-derived counterparts (e.g., CG-DCAP-KFe: I_D/I_G = 0.37, I_2D/I_G = 0.80). This disparity underscores the suitability of king grass as a precursor, likely attributable to its heterogeneous lignocellulosic matrix, which facilitates the formation of graphitic domains during catalytic pyrolysis. In contrast, purified cellulose’s ordered fiber structure may restrict three-dimensional carbon reorganization, limiting structural evolution [6,16].

CG-DCAP-KFe had the best performance: I_D/I_G = 0.37 and I_2D/I_G = 0.80, close to commercial graphene. These values not only surpass those of other materials in this study but also align with those reported for biomass-derived graphene synthesized using iron-based catalysts [16]. The efficacy of K₃[Fe(CN)₆] is attributed to its ability to promote biomass graphitization and ordered structural evolution during pyrolysis. This catalyst facilitates carbon layer intercalation, mitigates van der Waals forces between layers, and enables the production of ultrathin graphene-like sheets [16]. These findings demonstrate that the synergistic combination of conventional alkaline peroxide delignification (DCAP) and the K₃[Fe(CN)₆] catalyst optimizes graphene formation from king grass, yielding structural properties comparable to commercial counterparts. This approach underscores the potential of unconventional biomass feedstocks for synthesizing high-quality graphene-like carbon materials.

Comparison of the present results with previously published studies shows several points of convergence and a few distinctive features in the production of graphene from grass-derived biomass, particularly when advanced techniques such as ultrasound and microwave irradiation are combined with catalysts [38,39,42]. King grass, in particular, has proven to be an effective carbon source owing to its high cellulose content, which facilitates the synthesis of high-quality graphene. Relative to analogous investigations on graphene-type materials [63,64,65,66], king grass displays similar I_D/I_G ratios, indicating a defect density comparable to that of graphene obtained from other grasses. However, the I_2D/I_G values reported here point to a higher degree of structural ordering, especially when K₃[Fe(CN)₆] is used as the catalyst. This finding suggests that king grass may constitute a competitive and efficient alternative for graphene synthesis.

Delignification, a crucial step in enhancing the purity and structural organization of graphene, has been widely explored in the literature [18,24,32,33]. The alkaline peroxide delignification treatments employed in this study yielded results consistent with investigations demonstrating that alkaline delignification increases crystallinity and reduces the number of defects in the graphitic structure. In the present work, the combination of ultrasound with alkaline peroxide proved particularly beneficial, as it facilitated the removal of lignin and hemicellulose, thereby promoting a more ordered structure in the resulting carbon material [18,24,32,33]. This finding aligns with other studies showing that ultrasound enhances reagent penetration into biomass, improving delignification efficiency and contributing to greater graphitic organization.

The catalytic agents employed during deoxidation also markedly influence the quality of the synthesized graphene. In this study, the use of K₃[Fe(CN)₆] resulted in a reduced I₂/I_G ratio and an elevated I_2D/I_G, ratio, indicative of lower defect density and enhanced graphitic ordering compared to FeCl₃. This finding aligns with prior research by Fan et al. [62], who similarly reported that K₃[Fe(CN)₆] facilitates the formation of well-structured C–C bonds in graphitic materials. Furthermore, the ordered structure of the materials synthesized here, particularly CG-DCAP-KFe and CG-DUAP-Kfe, demonstrates structural parity with commercial graphene benchmarks, underscoring the viability of king grass as a sustainable precursor for graphene-based applications.

The synergistic integration of ultrasound and microwave irradiation applied in this study represents a significant advancement toward energy efficiency and process sustainability in graphene synthesis. Recent literature highlights that ultrasound and microwave techniques enable faster, more efficient biomass processing, reducing processing times and energy consumption compared to conventional thermal methods [41,42]. The results obtained here suggest that combining ultrasound and microwave assistance with an optimal catalyst not only enhances the quality of the synthesized graphene but also reinforces the sustainability of the approach. This strategy promotes a scalable and resource-efficient alternative for graphene production, aligning with green chemistry principles while maintaining cost-effectiveness.

In other words, research on the synthesis of graphene from king grass using ultrasound and microwave techniques in combination with K₃[Fe(CN)₆] as a catalyst supports and expands upon findings in the literature. The results obtained show that lignocellulosic herbaceous biomass such as king grass is a promising resource for the production of high-quality graphene, and that the applied treatments improve the structural organization of the carbonaceous material, making it comparable to that derived from other biomass precursors [40]. Furthermore, the methodology employed highlights the potential for implementing low-cost and environmentally friendly processes, aligning with the current interest in the production of advanced materials from renewable resources within the framework of sustainability.

The results summarized in Table 6 demonstrate that the incorporation of the dual precursor K₃[Fe(CN)₆] significantly alters the structure of the carbonaceous materials synthesized from king grass. Raman spectroscopy analysis reveals that the I_D/I_G ratios for the treated samples range approximately from 0.37 to 1.11, which are clearly lower than those observed for untreated samples. This reduction indicates a lower defect density and a higher degree of partial graphitization. Simultaneously, the I_2D/I_G values, ranging from 0.32 to 0.80, are relatively higher in the materials synthesized with the dual precursor, suggesting the presence of more extensive or fewer-layered graphene-like domains.

This improvement in structural ordering can be attributed to the catalytic role of iron, which during pyrolysis forms Fe₃C. This phase facilitates the reorganization of carbon layers and promotes the development of graphitic domains. Meanwhile, the potassium content in K₃[Fe(CN)₆] transforms into K₂CO₃, acting as a chemical activating agent that enhances microporosity and increases the specific surface area without severely compromising crystallinity.

The XRD results complement this analysis. Treated samples display sharper and more intense diffraction peaks around 2θ ≈ 26°, corresponding to the (002) plane of graphite, which supports the formation of more ordered structures. Additionally, characteristic peaks corresponding to Fe₃C and K₂CO₃ are detected in the treated samples but are absent or significantly less prominent in the control materials. These findings further support the interpretation that the combined in situ generation of a catalyst and activating agent enables more efficient partial graphitization alongside controlled porosity development.

Taken together, the detailed analysis of I_D/I_G and I_2D/I_G ratios, along with XRD patterns, confirms that the use of K₃[Fe(CN)₆] synergistically balances the increase in surface area—via chemical activation—and the enhancement of structural order via catalytic graphitization. This results in graphene-like materials with a reduced defect density, more ordered domains, and increased porosity, characteristics that make them highly promising for applications in adsorption, catalysis, and energy storage.

Compared to previous studies using biomass-derived precursors for graphene-like material synthesis, the materials produced in this work exhibit BET surface areas up to ~714.5 m²/g and lower defect indices (I_D/I_G = 0.37–1.11), indicating superior structural order. These values are comparable to, or even exceed, those reported for other biomass sources such as rice husk or sugarcane bagasse, which typically yield BET areas in the range of 400–600 m²/g through conventional multi-step chemical activation processes [1,17,31,36,38,43].

3.2.8. HRTEM Analysis

Figure 10 presents the HRTEM micrographs of the synthesized materials, along with the aqueous dispersions prepared to evaluate their colloidal stability. All dispersions, including the commercial graphene reference, exhibited rapid sedimentation. While dispersions of the carbonaceous materials required slightly longer settling times, all samples completely sedimented within 10 min.

The HRTEM micrographs of CG reveal thin sheets with well-defined, homogeneous edges characteristic of multilayer graphene. These sheets consist of one to three layers, consistent with XRD (54% crystallinity) and Raman spectroscopy (I_2D/I_G = 0.39; I_D/I_G = 0.09), which confirm a highly ordered structure with minimal defects. No impurities or secondary phases are observed, further corroborating the materials’ high purity and quality.

In contrast, the HRTEM micrographs of CG-DCAP-FeCl and CG-DUAP-FeCl exhibit poorly defined structures with irregular aggregates and amorphous morphologies devoid of crystalline order. Dark spots, likely corresponding to iron oxide nanoparticles, are also observed. These findings align with XRD data, which indicate low crystallinity (<15%) despite larger crystallite sizes, and Raman spectra, where the absence of the 2D band (I_2D/I_G = 0) and elevated I_D/I_G (>1.1) reflect a material with substantial structural defects. The detection of iron via AAS (3.76–4.22 wt.%) is visually corroborated by dispersed metallic phases.

Conversely, CG-DCAP-KFe and CG-DUAP-KFe show thin, folded graphene-like sheets with 2–3 layers, confirmed by Raman analysis (I_2D/I_G = 0.80 y 0.32, respectively) and XRD data (crystallite size = 0.11 nm; crystallinity = 24.3% and 21.3%). The I_D/I_G (0.37 and 1.11) indicate lower defect densities compared to FeCl₃-treated counterparts. Notably, the micrographs reveal clean surfaces without visible metallic impurities, suggesting that K₃[Fe(CN)₆] promotes the formation of more ordered graphitic architectures.

Additionally, the images corresponding to the CC-KFe material exhibit overlapping sheets with wavy edges and regions of variable transparency, suggesting a multilayer structure (around three layers), in agreement with the Raman results (I_2D/I_G = 0.46) and XRD data (crystallinity of 24%). The sheets show greater uniformity and cleanliness compared to those obtained from king grass, which could be attributed to the higher purity of the commercial cellulose used as the precursor. However, a slight presence of impurities was detected (K = 1.88%; Fe = 8.06%), possibly related to catalyst residues.

Finally, the CC-FeCl material exhibits morphologies similar to those observed in the CG-FeCl group, with amorphous structures and the presence of dark metallic aggregates. The high I_D/I_G ratio (1.20) and low crystallinity (18.8%), along with the absence of the 2D band in the Raman spectra, suggest a highly disordered structure, indicating that the FeCl₃ treatment did not promote the formation of well-organized graphenic structures.

3.3. Mechanistic Discussion of the Dual Precursor Effect on Structure and Porosity

The enhanced porosity and structural ordering observed in the carbon materials synthesized using K₃[Fe(CN)₆] is attributed to a dual-function mechanism involving catalytic graphitization and chemical activation. Upon pyrolysis at 1000 °C under an inert atmosphere, K₃[Fe(CN)₆] undergoes thermal decomposition, producing metallic iron (Fe), potassium cyanide (KCN), elemental carbon, and nitrogen gas, as described in Equation (2).

$$K_3\left[Fe{\left(CN\right)}_6\right]\stackrel{\Delta }{\to }Fe+3KCN+3C+{\displaystyle \raisebox{1ex}{$2$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}{N}_2\uparrow$$

(2)

The Fe atoms released during decomposition aggregate into metallic nanoparticles that catalyze the transformation of amorphous carbon into graphitic domains [12,17,71]. This occurs through two complementary pathways: direct catalytic reorganization of the carbon matrix on the surface of Fe nanoparticles (Equation (3)), and the transient formation of iron carbides, such as Fe₃C, which later decompose, leaving behind graphitized carbon layers (Equation (4)) [12,17,71].

$$F{e}_{\left(S\right)}+C_{amorphous}\to F{e}_{3}C$$

(3)

$$F{e}_{3}C\stackrel{\Delta }{\to }3Fe+C_{graphitic}$$

(4)

In parallel, the K-containing species—mainly KCN—undergo oxidation and secondary transformations, producing potassium carbonate (K₂CO₃), potassium oxide (K₂O), and potassium hydroxide (KOH). These intermediates play a key role in the activation process by generating gases and reacting with carbon, thereby inducing porosity [8,30,31,72]. At elevated temperatures, K₂CO₃ begins to decompose according to Equation (5).

$$K_{2}C{O}_3\to K_{2}O+C{O}_2\uparrow$$

(5)

The released CO₂ can gasify the carbon matrix (Equation (6)).

$$C{O}_2+C\to 2CO\uparrow$$

(6)

Additionally, both K₂CO₃ and K₂O can be reduced by carbon, generating metallic potassium vapor, which intercalates into the carbon lattice and facilitates the formation of micropores (Equations (7) and (8)) [8,30,31,72].

$$K_{2}C{O}_3+2C\to 2K+3CO\uparrow$$

(7)

$$C+K_{2}O\to 2K+CO\uparrow$$

(8)

The vapor-phase potassium penetrates between graphitic layers and expands the carbon structure, creating micropores and enhancing the surface area. In some cases, the formation of KOH (via reaction of K₂O with moisture or surface oxygen) can enrich the surface chemistry of the carbon materials with oxygen-containing functional groups such as –OH, C–O–C, C=O, and C–H. However, as the temperature increases (>800 °C), these oxygen-containing groups become thermally unstable and decompose, releasing additional gases (CO and CO₂) and contributing further to porosity development [72].

The porosity development and graphitization processes are not strictly sequential but partially overlapping. Porosity begins to form during the early stages of pyrolysis (below 800 °C), as K-derived species react with the carbonaceous matrix and generate gaseous products that etch the surface. Graphitic ordering, on the other hand, initiates at higher temperatures (above 850–900 °C), catalyzed by the presence of Fe nanoparticles and the formation–decomposition cycle of iron carbides. The porosity promotes gas diffusion and atomic mobility, which favors graphitization, while the emerging graphitic domains help stabilize the evolving porous network. This synergistic interplay leads to the formation of hierarchical porous carbon materials with a moderate to high degree of structural ordering.

This dual mechanism explains the significant differences in morphology and performance observed in samples prepared with K₃[Fe(CN)₆] compared to those synthesized using FeCl₃ or in the absence of a catalyst.

3.4. Limitations and Future Perspectives

While the proposed method has proven effective for synthesizing graphene-like carbon materials with improved graphitic domain ordering and high porosity from king grass, it is important to acknowledge several limitations that open pathways for future research aimed at optimizing and broadening its applicability. First, the present study was conducted at the laboratory scale, and the scalability of the process to industrial levels remains to be evaluated. Key factors for scale-up include the precise control of the pyrolysis temperature, the homogeneous release and interaction of Fe and K species during thermal decomposition, and the potential need for continuous feed and product handling systems.

Another critical aspect involves the long-term stability of the structural and textural properties of the synthesized materials. Their performance under real operating conditions—such as those encountered in adsorption, catalysis, or energy storage—must be assessed to determine their functional durability over extended periods.

From an environmental and safety perspective, attention should be given to the residual content of metallic species, particularly Fe₃C and potential iron oxides, which may affect material purity and pose environmental risks depending on the application. Additionally, the thermal decomposition of K₃[Fe(CN)₆] releases gaseous byproducts such as HCN, CO, and N₂, necessitating strict process control and the use of inert atmospheres to ensure both operational and environmental safety. At elevated temperatures, excess potassium vapor can lead to structural degradation of the carbon matrix, while residual potassium-containing compounds (such as K₂CO₃ and K₂O) may remain embedded in the structure, potentially altering electrochemical behavior unless properly removed through post-treatment processes like acid washing.

Furthermore, although king grass was selected as a model lignocellulosic biomass due to its high cellulose content and availability, it is essential to evaluate the reproducibility and robustness of the method using a wider range of biomass feedstocks. Such efforts would help establish the generalizability and adaptability of the synthesis approach for various agricultural or forestry residues with diverse compositions and thermal behaviors.

Future investigations should also focus on process optimization strategies aimed at reducing the pyrolysis temperature, improving control over the micro- and mesopore distribution, and directly evaluating the functional performance of the resulting materials in practical applications. These may include pollutant adsorption, catalytic activity in advanced oxidation processes, and energy storage in supercapacitors or rechargeable batteries.

Addressing these limitations will contribute to a deeper mechanistic understanding of the process and will support the demonstration of its industrial competitiveness and environmental viability.

4. Conclusions

The combined results of the characterization of the forest residue indicate that king grass biomass possesses physicochemical characteristics favorable for its valorization as a precursor in the production of carbon-based materials. Its high fixed carbon content and the presence of reactive oxygenated structures and specific atomic ratios provide a solid foundation for its transformation into porous materials with tunable properties. These characteristics are essential for the development of catalytic supports or high-value-added adsorbent materials, particularly in contexts where sustainability and the integral use of lignocellulosic waste are priorities.

This investigation into the synthesis of graphene-like carbon materials from king grass biomass revealed significant differences in the structural and morphological properties of the products, depending on the catalyst employed during deoxidation. Notably, the use of K₃[Fe(CN)₆] proved more effective in forming graphitic structures with higher crystallinity compared to FeCl₃ or catalyst-free conditions. At ~1000 °C, K₃[Fe(CN)₆] decomposes, releasing gases and forming Fe, Fe₃C, and K₂CO₃. Fe and Fe₃C catalyze graphitization; K₂CO₃ promotes pore formation and increases the surface area.

In contrast, FeCl₃-derived materials exhibited amorphous structures with reduced crystallinity. While FeCl₃ may catalyze dehydration and aromatization reactions, its efficacy in promoting graphene-like ordering is limited. Catalyst-free pyrolysis resulted in entirely amorphous carbons due to the absence of active species to drive structural reorganization.

Materials synthesized with K₃[Fe(CN)₆] displayed structural and morphological characteristics closer to commercial graphene, including higher crystallinity and pronounced porosity. These properties arise from the synergistic interplay of iron and potassium during pyrolysis, which concurrently enables graphitization and carbon activation. In comparison, FeCl₃-treated or catalyst-free materials showed less ordered structures and lower surface areas, limiting their resemblance to commercial graphene.

The use of K₃[Fe(CN)₆] as a catalyst for synthesizing carbon materials from king grass biomass represents an effective strategy for producing graphene-like architectures with high crystallinity and porosity. The synergistic action of iron and potassium during pyrolysis facilitates both graphitization and activation, yielding materials with superior structural and morphological properties compared to FeCl₃ or catalyst-free approaches. These findings align with emerging trends in biomass valorization and underscore the potential of lignocellulosic resources in advancing material nanoscience.