Tailoring Porosity and CO₂ Capture Performance of Covalent Organic Frameworks Through Hybridization with Two-Dimensional Nanomaterials

Abstract

This study reported covalent organic frameworks (COFs) and their hybrid composites with two-dimensional materials, graphene oxide (GO), graphitic carbon nitride (g-C₃N₄), and boron nitride (BN), to examine their structural, textural, and gas adsorption properties. Material characterization confirmed the crystallinity of COF-1 and the preservation of framework integrity after integrating the 2D nanomaterials. FT-IR spectra exhibited pronounced vibrational fingerprints of imine linkages and validated the functional groups from the COF and the integrated nanomaterials. TEM images revealed the integration of the two components, porous, layered structures with indications of interfacial interactions between COF and 2D nanosheets. Nitrogen adsorption–desorption isotherms revealed the microporous characteristics of the COFs, with hysteresis loops evident, indicating the development of supplementary mesopores at the interface between COF-1 and the 2D materials. The BET surface area of pristine COF-1 was maximal at 437 m²/g, accompanied by significant micropore and Langmuir surface areas of 348 and 1290 m²/g, respectively, offering enhanced average pore widths and hierarchical porous strcuture. CO₂ adsorption tests were investigated showing maximum adsorption capacitiy of 1.47 mmol/g, for COF-1, closely followed by COF@BN at 1.40 mmol/g, underscoring the preserved sorption capabilities of these materials. These findings demonstrate the promise of designed COF-based hybrids for gas capture, separation, and environmental remediation applications.

1. Introduction

The rapid rise of carbon dioxide (CO₂) in the atmosphere, caused by human activities such as the combustion of fossil fuels and industrial operations, poses a significant threat to global environmental stability, leading to irreversible climate change [1]. With the expansion of global populations and ongoing industrial development, CO₂ emissions are expected to increase, exacerbating climate change [2,3]. Confronting this challenge necessitates immediate and efficient solutions, with carbon capture and sequestration (CCS) technologies emerging as crucial tactics for reducing CO₂ emissions, primarily until renewable energy sources are adequately prevalent to satisfy global demand [4,5]. The success of CCS depends on creating new materials that can capture CO₂ efficiently, selectively, and cost-effectively [6]. Liquid adsorbents for CO₂ capture typically include aqueous amine solutions, ionic liquids, and amino acid salts, each offering high chemical affinities for CO₂. These liquids exhibit advantages such as high adsorption capacity and ease of regeneration under controlled conditions via chemical adsorption. However, challenges such as solvent degradation, high energy requirements for regeneration, and corrosiveness, especially in amine-based systems, limit their long-term sustainability and economic viability [7,8]. These limitations have intensified the search for alternative materials that exhibit enhanced performance and durability. Key targets for CCS technology include stationary point sources such as power plants and biogas upgrading systems, where CO₂ must be extracted from the gas mixture [9,10]. An optimal CO₂ adsorbent must demonstrate substantial adsorption capability, superior selectivity against other gases, durability in humid and high temperatures, and facile regeneration with little energy requirements [11,12,13].

Covalent organic frameworks (COFs) constitute an innovative category of porous materials, entirely composed of light elements such as carbon, hydrogen, boron, nitrogen, and oxygen, interconnected by robust covalent bonds [14,15,16,17]. The frameworks are formed by strategically linking organic moieties through strong covalent bonds, e.g., single bonds (C–C, C–N, C–O, or C–Si) or double bonds (e.g., C=N and N=N) [18,19,20]. They have experienced swift and extensive expansion, leading to the creation of diverse chemical architectures ranging from one-dimensional (1D) chains to intricately two-dimensional (2D) and three-dimensional (3D) networks [21,22,23,24]. The extensive array of reaction routes facilitates the creation of COFs with varied dimensions, topologies, and functional characteristics. COFs encompass low density, high specific surface area, high pore architectures, and stacking configurations that enhance electrical conductivity [24,25]. Moreover, the modular and adjustable characteristics of COF synthesis facilitate the precise design of pore dimensions, functions, and framework structures. The distinctive characteristics of COFs have established them as exceptionally adaptable materials, garnering considerable interest from the scientific community. They have exhibited high potential for various applications, such as environmental control (removal of pollutants from air, water, soil) [17], gas storage and separation [26], photocatalysis [27,28], sensing [29], drug delivery [30], and energy storage [31,32,33]. COFs were intensively investigated as adsorbents for CO₂ capture [34,35]. They can capture CO₂ from the open air [36]. They can be applied as catalysts for the chemical conversion of CO₂ into valuable compounds via several methods, including fixation [37,38], electrocatalytic reduction [39], and photocatalysis [40,41]. COF-based cobalt porphyrins enabled catalytic CO₂ reduction in water [42]. However, they have a high aggregation tendency [43]. Combining COF materials with other nanomaterials enables the formation of membranes [44]. Thus, they were conjugated with different materials for MOFs [45], polymers [46], and MXene [47]. COF-based nanocomposites may solve the materials’ challenges and enhance their performance or processing into commercial products, enabling further steps toward commercialization.

Two-dimensional nanomaterials, including carbon-based, e.g., graphene (G), graphene oxide (GO), graphitic carbon nitride (g-C₃N₄), and boron nitride (BN), received significant attention due to their extraordinary physicochemical properties and wide range of potential applications [48,49]. These materials possess exceptionally high surface areas, facilitating increased interaction with target molecules, making them promising for applications in sensors and biosensors [50], catalysis [51], energy storage [52], and environmental remediation or monitoring [53,54,55]. Their unique anisotropic structures contribute to directional physical and chemical behaviors that can be tailored for specific uses. Furthermore, their ease of integration into functional devices and prototype systems enhances their commercial viability. The structural versatility, high reactivity, and diverse applications make 2D nanomaterials a valuable platform in both academic research and industrial development. COF nanocomposites of 2D nanomaterials were investigated as adsorbents for removing anionic and cationic dyes, offering a 100% adsorption efficiency with superior selectivity relative to the pristine COF [56]. Conjugation of 2D with COF-1 may prevent the aggregation tendency of COFs.

This study involved the synthesis and modification of 2D nanomaterials, such as GO, g-C₃N₄, and BN, for use in COF-1. The materials were synthesized using a one-pot procedure via the solvothermal method. They were evaluated using X-ray diffraction (XRD), Fourier transform infrared (FT-IR), transmission electron microscopy (TEM), and nitrogen adsorption–desorption isotherms. The materials were investigated as adsorbents for CO₂ adsorption.

2. Results and Discussion

2.1. Material Characterization

Figure 1 and Figure 2 show the synthesis procedure for carbon nanomaterials (GO and g-C₃N₄), BN exfoliation, COF-1, and their composites. Hummer’s method produced graphene oxide (GO) (Figure 1a). Graphite powder and sodium nitrate were gradually incorporated into concentrated sulfuric acid while maintaining constant agitation in an ice bath for 4 h. Subsequently, potassium permanganate was added incrementally while maintaining continuous stirring. The solution was diluted with water, and the temperature was sustained at 37 °C for 2 h. To neutralize unreacted permanganate and finalize oxidation, hydrogen peroxide was introduced while stirring for 1 h, yielding a yellow solution. The resulting suspension was centrifuged, and the GO precipitate was rinsed with double-distilled water to eliminate residual salts and unreacted reagents. GO was dehydrated in a vacuum oven at 60 °C for 2 h.

g-C₃N₄ was obtained using direct thermal polymerization of melamine in a semi-closed environment to inhibit sublimation (Figure 1b). In a standard synthesis, melamine was added in a sealed crucible and heated at 550 °C at 10 °C/min in a muffle furnace for 2 h. The resultant yellow solid was rinsed and dried overnight.

BN exfoliation was conducted using a straightforward aqueous reflux technique (Figure 1c). Bulk BN was refluxed in deionized water with stirring at approximately 120 °C for 72 h. The exfoliated BN flakes were obtained by depositing the suspension onto a glass plate, drying, and carefully scraping (Figure 1c).

COF-1 was prepared using a solvothermal method (Figure 2). Melamine and terephthaldehyde were solubilized in DMSO under a nitrogen environment and subjected to rapid stirring. The mixture was conveyed to a closed reaction chamber and heated at 180 °C for 72 h. The product was obtained via filtration, rinsed with THF, acetone, and DMF, and subjected to vacuum drying at 120 °C overnight (Figure 2).

To prepare COF-based nanocomposites (COF-1@GO, COF-1@BN, and COF-1@g-C₃N₄), each 2D nanomaterial (GO, BN, or g-C₃N₄) was initially dispersed in DMSO using ultrasonication for 30 min (Figure 2). Subsequently, melamine and terephthaldehyde were included in the dispersion. The mixture was agitated and heated under solvothermal conditions at 180 °C for 72 h. The solid products were filtrated and rinsed with organic solvents to remove unreacted species. The synthesized materials were evaluated by XRD (Figure 3), TEM images (Figure 4 and Figure 5), FT-IR (Figure 6a), DRS (Figure 6b), and nitrogen sorption isotherms (Figure 7;

Table 1. Surface area and pore volume using different techniques.

Materials	Surface Area (m2/g)				Pore Size (Å)
	Micropore Surface Area	BET Surface Area	Langmuir Surface Area	t-Plot External Surface Area	Adsorption Average Pore Width (4 V/A by BET)	Desorption Average Pore Width (4 V/A by BET)	BJH Adsorption Average Pore Diameter (4 V/A)	BJH Desorption Average Pore Diameter (4 V/A)	D-H Adsorption Average Pore Diameter (4 V/A)	D-H Desorption Average Pore Diameter (4 V/A)
COF-1	437	348	1290	231	6.6	7.6	13.0	36.9	13.0	37.5
COF@GO	141	129	823	126	113.6	132.0	173.3	162.5	165.4	165.3
COF@g-C3N4	309	262	1196	211	85.5	102.5	145.0	163.5	155.5	166.6
COF@BN	362	305	1341	361	82.6	93.7	140.9	210.4	141.2	213.4

).

The structural characteristics and crystallinity of the produced materials, GO and COF-1, and their nanocomposites (COF-1@GO and COF-1@BN), were investigated via XRD, as illustrated in Figure 3a,b. The XRD pattern of GO has a distinct diffraction peak at a Bragg angle of 2θ = 10.9°, associated with the (001) plane. Employing Bragg’s law (nλ = 2dsinθ) with Cu Kα radiation (λ = 0.15418 nm), this peak corresponds to an interplanar spacing (d-spacing) of 0.81 nm, confirming the successful oxidation of graphite and the formation of GO characterized by interlayer expansion resulting from oxygenated functional groups and water intercalation (Figure 3a).

For COF-1, a unique diffraction pattern is evident, featuring significant peaks at 2θ = 9.0°, 16.8°, 18.3°, and 27.2° (Figure 3a). The associated d-spacings are 0.98 nm, 0.53 nm, 0.48 nm, and 0.33 nm, respectively. The peak at 9.0° is significant due to its occurrence at a low angle, indicating a considerable interlayer distance, characteristic of supramolecules such as COF. This reflection corresponds to the (100) plane in a lattice structure and endorses the establishment of an organized, polymeric COF network. The systematic arrangement of building blocks in two dimensions and the maintenance of long-range order further validate the efficacy of the condensation process between melamine and terephthaldehyde, resulting in a crystalline polymeric structure rather than an amorphous polymer (Figure 3a).

The XRD pattern of the COF-1@GO composite has a prominent diffraction peak at a Bragg angle (2θ) of 26.5° with a d-spacing of 0.34 nm, attributable to the Miller index of (002) reflection of graphitic carbon (Figure 3a). The inclusion of GO shows an insignificant change in the COF structure but notably alters its diffraction profile, indicating the integration of the two components.

The XRD pattern of the COF-1@BN composite displays a distinct peak at 2θ = 26.8° (d = 0.33 nm), aligning with the (002) plane of COF-1 (Figure 3b). It verifies the preservation of COF’s crystalline structure and its stable integration with BN. The high intensity may indicate the highly organized structure of the BN phase inside the hybrid material. Alongside strong diffraction peaks, the large hump within the 20–50° range indicates the existence of amorphous or semi-crystalline domains inside the hybrid composites. Such characteristics generally result from structural disorder, partial exfoliation, or strain effects from the interaction between the COF and the 2D nanomaterials (GO and BN). These disruptions may decrease long-range order while maintaining local crystallinity, which is frequently beneficial for applications such as adsorption or photocatalysis. The XRD data indicate the effective synthesis of crystalline COF-1 with long-range polymeric order, as demonstrated by the low-angle peak at 9.0° and its associated large d-spacing. Incorporating GO and BN into the COF matrix introduces supplementary crystallographic characteristics and alters the diffraction pattern by hybrid material development. These observations confirm the structural integrity and successful production of COF-based composite materials.

TEM images for pristine materials (Figure 4) and their composite (Figure 5) are investigated. Figure 4a displays the TEM image of GO, with a characteristic wrinkled and crumpled sheet-like shape. The distinctive properties of TEM images reflect the thin nature of GO nanosheets, which are prone to developing folds and wrinkles due to structural imperfections and oxygenated functional groups introduced during oxidation. The sheets display pores, presumably due to these imperfections. GO displays wavy particles with gray contrast, indicative of the standard features seen in chemically exfoliated GO (Figure 4a). Figure 4b presents a TEM image of exfoliated BN. The image illustrates a multi-layered structure (Figure 4b). The sheet edges are sharply separated, indicating effective exfoliation of BN following extended reflux in water. The exfoliated edges indicate larger layer spacing, which amplifies BN’s surface area and reactivity, facilitating its accessibility for interaction or composite production. Figure 4c depicts the TEM image of COF-1, illustrating irregular spherical particles with a relatively consistent size of around 20 nm. The small particle size and spherical shape indicate the creation of nanoscale COF particles.

Figure 5 offers more insights into the composite materials. TEM images represent the morphology of the COF-1@GO (Figure 5a) and COF-1@g-C₃N₄ (Figure 5b). It illustrates that the spherical COF-1 nanoparticles are uniformly dispersed into the crumpled GO sheets (Figure 5a). Figure 5b illustrates the TEM image of the COF-1@g-C₃N₄ nanocomposite. COF-1 nanoparticles are dispersed throughout the layered structure of g- C₃N₄, which exhibits particles with flat morphology. The contrast difference verifies the effective deposition of COF-1 onto the g-C₃N₄ surface. TEM images presented in Figure 4 and Figure 5 validate the synthesis and structural incorporation of COF-1 with diverse two-dimensional nanomaterials. The distinct visualization of morphology, particle size, and hybrid interfaces offers compelling evidence for these composite formations and structural integrity.

The functional groups and chemical interactions in COF-1 and its composites, COF-1@GO, COF-1@g-C₃N₄, and COF-1@BN, were examined via FT-IR spectroscopy (Figure 6a). The FT-IR spectrum of COF-1 shows a band at 2920 cm⁻¹ and 3400 cm⁻¹ corresponding to N–H/C–H symmetric stretching and adsorbed O–H groups, presumably from moisture, respectively. The presence of bands at 1540 and 1480 cm⁻¹, indicative of C=N stretching vibrations of the triazine ring, confirms the polymerization and production of the COF (Figure 6a). The characteristic peaks for the aldehyde are absent, indicating the aldehyde’s condensation with the amine. No peaks related to the aldehydes were observed, specifically at 2870 cm⁻¹ (C-H stretching) and 1690 cm⁻¹ (C=O stretching). Notably, when COF-1 is combined with two-dimensional nanomaterials such as GO, g-C₃N₄, and BN, the FT-IR spectra of COF-1@GO, COF-1@g-C₃N₄, and COF-1@BN exhibit an insignificant change in the FT-IR spectrum of pure COF-1. These observations suggest that integrating these 2D nanomaterials during COF’s synthesis shows no effect on the polymerization process. The preservation of characteristic COF-1 peaks in all composite spectra verifies the integrity of the COF, suggesting that the interactions between COF-1 and the 2D nanomaterials are predominantly physical interactions with no observation of new bonds (no chemical interactions). The noncovalent bonds are mainly hydrogen bonds and π-π interactions based on the benzene rings present in both COF-1 and 2D nanomaterials.

The DRS spectra illustrated in Figure 6b elucidate the optical absorption properties of the synthesized materials: COF-1, GO, COF-1@GO, COF-1@g-C₃N₄, and COF-1@BN. COF-1 displays a broad absorption spectrum from 200 to 490 nm, with a peak absorption at 245 nm, indicative of π–π* transitions in its conjugated structure (Figure 6b). GO exhibits an absorption peak at 250 nm, corresponding to π–π* transitions of sp² hybridized carbon domains, characteristic of GO (Figure 6b). The spectra of the COF-1@GO, COF-1@g-C3N4, and COF-1@BN composites primarily exhibit the absorption profile of COF-1, suggesting that the addition of GO, g-C₃N₄, or BN does not substantially modify the optical properties of the COF-1 framework in the ultraviolet region. COF-1@g-C₃N₄ has an extra absorption band at around 450 nm, absent in the other composites (Figure 6b). The absorption at 450 nm is due to the inherent absorption of g-C₃N₄, a yellow-colored substance recognized for its activity in visible light. The existence of this peak validates the effective incorporation of g-C₃N₄ into the COF-1 matrix and suggests an expanded light absorption spectrum. The DRS results confirm the structural integrity of COF-1 in all composites and illustrate the functional effect of g-C₃N₄ in shifting optical absorption toward the visible spectrum.

Figure 7a illustrates the nitrogen adsorption–desorption isotherms for COF-1 and its composites with 2D materials (GO, g-C₃N₄, and BN). All materials display Type I isotherms characterized by a high uptake at low relative pressures (P/P₀ < 0.1), suggesting their microporous characteristics. A hysteresis loop at high relative pressures (P/P₀ = 0.8–0.9) is evident in the composites, highlighting the formation of supplementary mesopores or interparticle pores between COF-1 particles and 2D materials (i.e., GO, g-C₃N₄, and BN). The BET (Brunauer, Emmett and Teller) plots in Figure 7b validate linearity within the suitable relative pressure range, hence endorsing precise surface area calculations, as shown in Table 1.

Table 1 summarizes the specific surface areas (e.g., external surface area, micropore surface area, BET, and Langmuir) and pore sizes (e.g., BET, Barrett-Joyner-Halenda (BJH), and Dollimore-Heal (DH) method). COF-1 has a substantial BET surface area of 348 m²/g and a notable microporosity of 437 m²/g, characterized by small pore diameters ranging from 6.6 to 7.6 Å. Hybridization with GO, g-C₃N₄, or BN significantly decreases BET surface area (129–305 m²/g), presumably due to partial pore obstruction or structural distortion. These composites exhibit markedly increased average pore sizes (up to approximately 213 Å), as demonstrated by BJH and D-H techniques, highlighting the development of micro and meso-pore structure i.e. hierarchical porous structure. Among the composites, COF/BN has the highest surface area and external surface area (305 and 361 m²/g, respectively), but COF/GO, despite possessing the lowest BET area, features the most prominent pores, indicating a structure characterized by interfacial pores or voids. Data analysis suggests the formation of a hierarchically porous structure suitable for multifunctional applications, including adsorption.

2.2. Carbon Dioxide Adsorption

COF-1 and their composites (e.g., COF-1@GO, COF@g-C₃N₄, and COF@BN) were investigated for CO₂ adsorption (Figure 8a,b). Figure 8a illustrates the CO₂ adsorption–desorption isotherms for COF-1 and its composites with GO, g-C₃N₄, and BN. All materials have characteristic Type I isotherms, characterized by rapid adsorption at low relative pressures (P/P₀ < 0.03), indicating strong interactions with CO₂ molecules and the existence of microporous regions. The well-aligned adsorption and desorption branches at low pressures indicate reversible physisorption. As pressure increases, the adsorption capacity progressively increases, signifying the high adsorption capacities at high pressure.

These isotherms’ quantitative CO₂ adsorption capabilities further substantiate the material trends (Figure 8b). Pristine COF-1 exhibits the maximum CO₂ adsorption at 1.47 mmol/g, owing to its high microporosity and well-defined pore architecture. COF@BN exhibits a near value of 1.40 mmol/g, owing to its hierarchical pore structure that enhances CO₂ transport and retention. Conversely, COF@GO and COF@g-C₃N₄ demonstrate marginally reduced adsorption capabilities of 0.997 and 0.945 mmol/g, respectively (Figure 8b). This decrease may come from partial pore obstruction or a modified framework arrangement resulting from the incorporation of 2D nanosheets. Despite their low capacity, these composites exhibit commendable performance and provide supplementary functions (e.g., structural stability or improved light absorption) relevant to hybrid material applications. The material can be recycled for CO₂ adsorption in three successive cycles without a drop in its adsorption capacity (Figure 8c).

Carbon dioxide (CO₂) can be removed chemically via amine-forming carbamate, which undergoes decomposition over time, decreasing its capturing capacity over time [57]. Furthermore, adsorption is mainly a chemically governed mechanism. Thus, physical adsorption using porous materials like zeolites [58], porous carbons [59], metal–organic frameworks (MOFs) [60], and COFs have attracted considerable interest for their potential in CO₂ capture applications (

Table 2. Material adsorbents for CO₂ adsorption.

Materials	Synthesis Method	Conditions	CO2 Adsorption Capacities (mmol/g)	Adsorption Conditions	Ref.
HKUST-1@BNNS-PEI	Ball-milling	6 h, 400 rpm	4.47	289 K, 1 bar	[61]
MIL-100@BNNS-PEI			1.51
ZIF-8@BNNS-PEI			0.96
TDCOF-5	Solvothermal method	A mixture of mesitylene and dioxane at 120 °C for 5 days	2.09 mmol/g (92 mg/g)	273 K, 1 bar	[62]
Activated carbon	Hydrothermal carbonization of waste biomass	Carbonization at 800 °C for 2 h	1.45	298 K, 10 kPa	[63]
COF-1	Hydrothermal	Reflux, 180 °C, 72 days	1.47	298 K, 1 bar	Herein
COF-1@BN			1.40
COF-1@GO			0.997
COF-1@g-C3N4			0.945

). Zeolites offer chemical and thermal stability; however, they exhibit performance degradation in humid conditions and require substantial energy for regeneration. Porous carbons facilitate high-pressure CO₂ extraction and moisture resistance, although they exhibit selectivity and structural regulation challenges. MOFs characterized by high porosity, structural adaptability, and low energy demands for regeneration exhibit exceptional capabilities for CO₂ capture. Nonetheless, their efficacy may decrease in humid environments because water preferentially binds to active metal sites. COFs, as a category of porous crystalline materials, offer a compelling alternative by integrating the advantageous characteristics of MOFs with improved stability and recyclability, even under humid conditions. Two-dimensional materials, such as boron nitride, enable the self-assembly and formation of porous foams with adsorbents like MOFs, resulting in effective CO₂ adsorption [61].

A comparison among different materials was tabulated, including COF [62], MOF [61], and carbon materials [63] (Table 2). Assessing materials for CO₂ adsorption is complex, as efficacy relies on multiple qualities rather than a singular characteristic. Evaluating materials only based on surface area or porosity may oversimplify performance results and overlook critical factors. There is no absolute correlation or relation between BET surface area and the adsorption capacities of CO₂ absorption. Despite the extensive surface areas of activated carbons, their capacity to adsorb CO₂ molecules may be constrained if their pore dimensions are inadequately sized or inaccessible. Zeolites and MOFs featuring ultramicropores, possessing moderate surface areas yet tailored micropore structures, can excel in low-pressure applications owing to enhanced van der Waals or electrostatic interactions. CO₂ adsorption is significantly influenced by pore size, particularly in ultramicropores (<0.7 nm), where potential overlap enhances binding. Materials with minimal pore volume or surface area can exhibit considerable adsorption owing to the presence of narrowly distributed and accessible micropores. Chemical functionality is essential for CO₂ capture, especially at low partial pressures. In diluted CO₂ streams, amine-functionalized materials such as COF materials can enhance adsorption via acid–base interactions. Nonetheless, these interactions may influence desorption efficiency or stability. Materials that are purely physisorptive exhibit greater reversibility but lower efficacy under ambient conditions due to their weaker interactions. Composite or hybrid materials, including polymer–inorganic frameworks, can improve CO₂ adsorption through synergistic mechanisms. These systems cannot be evaluated against homogeneous materials such as pure zeolites or carbon adsorbents using a singular property metric, as functional domain distribution and interactions influence performance. El-Kaderi et al. reported the synthesis of a two-dimensional mesoporous COF (TDCOF-5) originating from triptycene [62]. The substance was synthesized by condensing 1,4-benzenediboronic acid and hexahydroxytriptycene, resulting in boronate links via a solvothermal process. The synthesis occurred in a mixed solvent system of mesitylene and 1,4-dioxane at 120 °C for five days. The resultant TDCOF-5 exhibited CO₂ adsorption capability of 92 mg/g at 273 K and 1 bar pressure [62]. A “self-supporting foam” was reported to create a thermally conductive composite foam consisting of MOFs and boron nitride nanosheets (BNNSs), interconnected by polyethyleneimine (PEI) cross-linking, referred to as MOFs@BNNS-PEI [61]. The adaptable approach facilitates the effective integration of diverse representative MOFs: HKUST-1, MIL-100(Fe), and ZIF-8. Due to these synergistic effects, the MOFs@BNNS-PEI composite demonstrates a markedly enhanced CO₂ adsorption capacity, 1.35 to 1.42 times greater than pure MOFs, and a 5.0- to 5.7-fold increase in CO₂ desorption rate (Table 2) [61]. HKUST-1@BNNS-PEI, MIL-100@BNNS-PEI, and ZIF-8@BNNS-PEI exhibited CO₂ capacities of 4.47, 1.51, and 0.96 mmol/g compared to pristine MOFs (3.15, 1.10, and 0.71 mmol/g), respectively [61]. Activated carbons were synthesized via hydrothermal carbonization of waste biomass [63]. The material was used for CO₂ adsorption, offering capacities of 1.45 mmol/g at a low CO₂ partial pressure of 10 kPa and a temperature of 0 °C [63]. Our materials exhibit comparable CO₂ adsorption compared to reported materials, e.g., COF [62], MOF [61], and carbon materials [63] (Table 2). Our materials exhibit high porosity, enabling high adsorption capacities with high potential for use in the adsorption of other molecules.

3. Experimental

3.1. Materials and Methods

Melamine (Merck, Rahway, NJ, USA), dimethyl sulfoxide (SDFCL, Mumbai, India), and sulfuric acid (SDFCL, Mumbai, India) were used without further purification. Graphite flakes (+100 mesh, ≥75%) and terephthaldehyde were purchased from Fluka (Germany). Boron nitride (BN, powder, size of 1 μm, 98%) was purchased from Sigma-Aldrich (Steinheim, Germany).

3.2. Synthesis of GO

GO was synthesized according to Hummer’s method (Figure 1a) [64,65,66]. Graphite powder (2 g) and sodium nitrate (2 g) were introduced to concentrated sulfuric acid (100 mL). The addition was performed in an ice bath and stirred continuously for 4 h. Thereafter, potassium permanganate (12 g) was gradually introduced while maintaining constant agitation. Afterward, 200 mL of double-distilled water was carefully added, and the temperature was maintained at 37 °C for 2 h. The reaction mixture was further diluted with additional double-distilled water, and the temperature was then elevated to 98 °C and maintained for 2 h. To reduce the excess potassium permanganate, 40 mL of hydrogen peroxide was added while stirring for 1 h, yielding a yellow solution. The suspension was subsequently centrifuged (6000 rpm, 10 min) to separate the GO, which was washed thrice with double-distilled water to eliminate unreacted substances and residual salts.

3.3. Synthesis of g-C₃N₄

g-C₃N₄ was synthesized via polymerization of melamine (Figure 1b) [67,68]. A semi-closed system was employed to prevent the sublimation of melamine during heating. Specifically, 5 g of melamine powder was added into a covered alumina crucible and heated in a muffle furnace. The initial heating was conducted at 550 °C for 2 h with a heating rate of 10 °C/min. The material was then washed with water and ethanol, followed by overnight drying.

3.4. Exfoliation of Boron Nitride

The exfoliation of boron nitride was carried out via a water-based method (Figure 1c). Briefly, a bulk boron nitride (2 g) was refluxed (heating at 120 °C) in water for 3 days. The exfoliated material in water was dropped onto a glass plate and then allowed to dry. The flakes were collected and used for COF/BN synthesis.

3.5. Synthesis of COF-1

COF-1 was synthesized via a solvothermal approach (Figure 2) [69]. In a typical procedure, melamine (4 mmol, 0.5 g) and terephthaldehyde (6 mmol, 0.8 g) were dissolved in 25 mL of degassed dimethyl sulfoxide under a nitrogen atmosphere. The reaction mixture was stirred vigorously and heated at 180 °C for 72 h. The resulting precipitate was filtered and thoroughly washed with tetrahydrofuran and N,N-dimethylformamide before being dried under vacuum at 120 °C overnight.

3.6. Synthesis of COF-1-Based Composites

To prepare the nanocomposites, 300 mg of GO, BN, or g-C₃N₄ was dispersed in 25 mL of DMSO using ultrasonication [56]. Subsequently, melamine (4 mmol) and terephthaldehyde (6 mmol, 0.8 g) were dispersed to the prepared solutions. The mixture was then subjected to solvothermal treatment at 180 °C for 3 days. After cooling, the precipitates were isolated and washed, as mentioned above.

3.7. Characterization

XRD patterns were collected using X’Pert PRO PANAlytical (Bruker, Billerica, MA, USA). TEM (Philips, Amstelplein, the Netherlands) was used to determine the morphology and particle size. Functional groups on the surface were studied using FT-IR (IRTracer-100, SHIMADZU, Kyoto, Japan). Diffuse reflectance spectroscopy (DRS) spectra were recorded using UV-2600i (SHIMADZU, Kyoto, Japan).

3.8. Gases (Nitrogen and CO₂) Adsorption

The specific surface areas (Brunauer–Emmett–Teller (BET), Langmuir, micropore surface area, t-plot external surface area, m²/g) were determined using N₂ isotherms (at 77 K) using an ASAP 2000 analyzer (Micrometric, Lincoln, UK). The sample was degassed at 110 °C under vacuum for 3 h. Pore size (Å) was evaluated using adsorption average pore width (4 V/A by BET), desorption average pore width (4 V/A by BET), BJH (Barrett, Joyner, and Halenda) adsorption average pore diameter (4 V/A), BJH desorption average pore diameter (4 V/A), D-H (Dollimore–Heal) adsorption average pore diameter (4 V/A), and D-H desorption average pore diameter (4 V/A).

4. Conclusions

This study emphasizes the synthesis and characterization of COF-1 and its composite variants with GO, g-C₃N₄, and BN. XRD and FT-IR structural analysis validated the integrity of the COF and identified functional groups from both components in the composites. TEM images provided morphological insights, revealing well-organized polycrystalline particles and the efficient incorporation of two-dimensional additives, which facilitated the development of heterogeneous interfaces. Nitrogen sorption analyses showed that the pure COF possessed a high surface area and microporosity, but the integration of 2D materials resulted in the formation of additional mesopores, as demonstrated by hysteresis in the isotherms and enlarged average pore sizes. The results demonstrate the effectiveness of integrating 2D materials in modifying the textural and adsorption properties of COFs, positioning these hybrids as viable options for gas storage and separation technologies.