Preparation and CO₂ Adsorption Performance of Nitrogen-Doped Carbon Derived from Phenolic Resin

Abstract

Carbon dioxide emissions, particularly from large point sources such as fossil-fuel power plants, represent a primary driver of global warming. Although various carbon-based adsorbents have been developed for carbon capture applications, most existing materials exhibit limited CO₂ adsorption capacity at flue gas-relevant partial pressures and are susceptible to interference from impurity components. In this study, a series of nitrogen-doped carbons was prepared from commercial phenolic resin and melamine via a two-step carbonization–activation process. The effects of precursor-to-dopant ratio and thermal conditions on CO₂ adsorption were systematically investigated. The results indicated that CO₂ uptake was influenced by specific surface area, nitrogen content, micropore volume, and total pore volume, with a maximum adsorption capacity of 2.455 mmol·g⁻¹ and selectivity over 28 at 25 °C and 1 bar. The series also exhibited excellent cycling stability (<1% loss after 5 cycles) and fast kinetics (>90% uptake within 3 min), suggesting its potential applicability in flue gas CO₂ capture.

1. Introduction

Since the Industrial Revolution, the demand for fossil fuels has expanded rapidly, leading to massive energy consumption and greenhouse gas emissions. Among these emissions, carbon dioxide (CO₂), the primary greenhouse gas, accumulates from fossil fuel combustion and directly causes global warming and ocean acidification [1,2,3]. To meet the Paris Agreement’s 2 °C goal, global CO₂ emissions must be cut 50% by 2050 [4]. Therefore, it is imperative to advance low-carbon technologies such as carbon capture, utilization, and storage (CCUS) [5]. Carbon capture stands as one of the most critical singular mitigation strategies available today [6].

Common carbon capture methods include amine scrubbing, cryogenic distillation, membrane separation, and adsorption [7,8,9]. Despite its high efficiency and wide use, amine scrubbing suffers from solvent volatility, degradation, corrosivity, and high regeneration energy costs [10,11,12,13,14,15,16,17]. Cryogenic distillation produces high-purity liquid CO₂ suitable for storage and transport but is energy-intensive and less cost-effective [18,19]. Membrane systems offer modularity, operational simplicity, and lower energy use, yet are limited by plasticization, fouling, aging, and high replacement costs [18,20,21,22,23,24,25]. In contrast, adsorption-based processes utilizing solid sorbents to separate CO₂ from flue gas offer a promising low-energy alternative for carbon capture. This approach is recognized for its process simplicity, low corrosivity, cost-effectiveness, and high efficiency, making it a viable technology option [3,26,27,28].

In adsorption-based CO₂ capture processes, the performance of the adsorbent is a critical determinant of overall system efficiency [29]. Conventional adsorbents are often constrained by limitations such as insufficient selectivity, low adsorption capacity, and inadequate thermal stability [30]. Therefore, significant attention has been devoted to the adsorption of CO₂ on porous materials, such as metal–organic frameworks (MOFs), amine-grafted silica, zeolites, carbon-based materials, and mesoporous alumina [31,32,33]. Among these, carbon materials have emerged as highly promising candidates for CO₂ capture because of their wide availability, low cost, excellent hydrophobicity, tunable pore structures, high stability, and low energy requirements for regeneration [34,35,36]. Activated carbons are typically classified by precursors into coal-derived, biomass-derived, resin-based, and nitrogen-doped types. Key textural properties, including specific surface area, micropore volume, and pore size distribution, can be tuned through precursor selection and synthesis conditions, directly governing their CO₂ adsorption performance [37,38,39]. To improve CO₂ separation on carbon adsorbents, amine functionalization and nitrogen doping are commonly employed. Despite its high capacity, amine grafting suffers from limited regenerative stability. Nitrogen doping, which integrates nitrogen atoms into the carbon framework to enhance surface polarity and basicity, offers a more robust and widely adopted alternative [5,40,41].

YUE et al. synthesized a series of modified carbon materials using urea-treated coconut shell activated carbon, achieving a CO₂ adsorption capacity of 3.71 mmol·g⁻¹ at 1 bar and 25 °C [42]. LIU et al. prepared N-doped porous carbon from industrial phenolic resin using urea as a nitrogen source and KOH as an activating agent, reporting a CO₂ uptake of 5.01 mmol·g⁻¹ under the same conditions [43]. Souza et al. utilized formaldehyde and phloroglucinol as carbon precursors to synthesize phenolic resin-based carbons via a soft-templating method, followed by KOH activation to enhance microporosity. The activated sample demonstrated a CO₂ adsorption capacity of 4.4 mmol·g⁻¹ at ambient pressure and 25 °C [44]. Lucas Spessato et al. developed nitrogen-doped adsorbents using Brazil nut shells (BNSs) as the carbon source, with melamine, hexamethylenetetramine, and tetramethylammonium hydroxide as nitrogen agents, achieving CO₂ adsorption capacities ranging from 4.16 to 5.30 mmol·g⁻¹ [45]. Wang et al. synthesized N-doped porous carbon using polypyrrole prepared from FeCl₃, KOH, and pyrrole as a precursor, achieving a CO₂ adsorption capacity of 3.61 mmol·g⁻¹ at 30 °C and 100 kPa [46]. Liu et al. carbonized low-cost urea-formaldehyde resin followed by KOH activation, yielding a sample with a CO₂ uptake of 1.44 mmol·g⁻¹ at 1 bar and 25 °C [47]. Rao et al. produced nitrogen-rich activated carbon via a single-step sodium amide activation of carbonized water chestnut shell, which exhibited a CO₂ adsorption capacity of 4.5 mmol·g⁻¹ at 25 °C and 1 bar [48]. Dang et al. prepared N-doped microporous activated carbon using starch as the carbon source, phenolic resin as a nitrogen modifier, and KOH as the activating agent, reporting a high CO₂ uptake of 6.54 mmol·g⁻¹ at 0 °C and 100 kPa [49].

Although these nitrogen-doped carbon materials exhibit promising CO₂ adsorption performance, their synthesis often involves complex procedures and cumbersome precursor processing, and poses scaling challenges. Moreover, most of these materials are produced via chemical activation, which requires extensive washing processes, leading to resource consumption and potential chemical pollution [50]. In this study, we employ low-cost phenolic resin with well-controllable porosity as the precursor [51] and melamine as the nitrogen dopant [52]. A nitrogen-doped carbon-based adsorbent is synthesized using CO₂ as the physical activating agent. The effects of the dopant-to-precursor mass ratio, carbonization temperature, and activation temperature on the adsorption performance are systematically investigated. The adsorbent is further evaluated in terms of its CO₂/N₂ selectivity, adsorption kinetics and cyclic stability.

2. Materials and Methods

2.1. Preparation

The commercial phenolic resin used in this study was obtained from Zhejiang Juhua Co., Ltd., Quzhou, China melamine was sourced from Tianjin Kermel Chemical Reagent Co., Ltd., Tianjin, China. and both CO₂ and N₂ gases were supplied by Tianjin Best Gas Corporation., Tianjin, China.

2.1.1. Carbonization Under N₂ Atmosphere

Carbonization is a pyrolysis process conducted under an inert and oxygen-free atmosphere, where the precursor undergoes thermal decomposition to eliminate volatile components, yielding carbon with a rudimentary pore structure. This intermediate carbon is subsequently transformed into a highly developed porous architecture during the activation stage [53,54].

In this study, melamine and phenolic resin were thoroughly mixed in an agate mortar at various mass ratios and then transferred into a porcelain boat. (The mass ratio of melamine to phenolic resin is denoted as R, where R = 0 corresponds to pure phenolic resin. The carbonization and activation temperatures are designated as X and Y, respectively, in degrees Celsius (°C)). The mixture was subjected to carbonization in a horizontal tube furnace under a N₂ atmosphere with a flow rate of 0.6 L/min. The thermal treatment was carried out at a heating rate of 5 °C/min up to a carbonization temperature X (e.g., 500 °C), held for 2 h, followed by natural cooling to room temperature. The N₂ atmosphere was maintained throughout this cooling period. The cooled char was subsequently activated.

2.1.2. CO₂ Activation

Physical activation involves the reaction of carbonized char with gaseous activating agents at elevated temperatures, typically employing steam or carbon dioxide as the agents [53,55]. After cooling to room temperature following the carbonization stage, the atmosphere was switched to CO₂ at a flow rate of 0.8 L/min and maintained for 30 min to allow for complete gas purging. The activation process was then initiated at a heating rate of 5 °C/min up to the activation temperature Y (e.g., 600 °C) and held for 1 h. The CO₂ atmosphere was retained during the subsequent cooling stage. After cooling to room temperature, the resulting material was repeatedly washed with deionized water to remove ash, and then dried at 120 °C for 2 h in a constant-temperature drying oven. The final black N-doped carbon powder was designated as NC_R-X-Y.

In this study, the melamine-to-phenolic resin mass ratio R was selected as 0, 0.5, 1, 2, and 3; carbonization temperatures X were set at 450 °C, 500 °C, 550 °C, and 600 °C; and activation temperatures Y were tested at 550 °C, 600 °C, 650 °C, and 700 °C.

2.2. Adsorption Performance Evaluation

2.2.1. Static Volumetric Method

The CO₂ adsorption isotherms of the nitrogen-doped activated carbons were measured using custom-built static volumetric apparatus, as schematically illustrated in Figure 1. The system consists of gas cylinders, a reference tank, an adsorption tank, pressure transducers, a vacuum pump, a thermostatic water bath, and associated valves and tubing.

The tubing volume between the inlet valve V₁ and valve V₂ was attributed to the reference tank volume, while the volume between V₂ and the evacuation valve V₃ was assigned to the free volume of the adsorption tank. The volume of the reference tank (V_r) was pre-calibrated. A sample of mass m (g) was loaded into the adsorption tank. After opening the equalization valve V₂, the vacuum pump and valve V₃ were activated to evacuate the system for 3 h.

The experiment consisted of the following two steps:

• Measurement of the Free Volume in the Adsorption Tank

After evacuation, the pressures in both the adsorption tank and the reference tank were equal, recorded as the initial system pressure P₀. Valve V₂ was then closed, and a certain amount of helium was introduced into the reference tank via valve V₁ before closing V₁. After the pressure reading stabilized, the elevated pressure was recorded as P₁. Subsequently, valve V₂ was opened, allowing gas to flow from the high-pressure reference tank to the low-pressure adsorption tank until pressure equilibrium was reached. The resulting equilibrium pressure was denoted as P₂. Since helium is not adsorbed by the adsorbent, the total amount of substance remained unchanged before and after pressure equalization. The free volume V₀ was then calculated using Equation (1):

$$V_0={\displaystyle \frac{P_1-P_2}{P_2-P_0}}{V}_r$$

(1)

where V₀ represents the free volume (mL) remaining in the adsorption tank after loading the adsorbent, V_r is the volume (mL) of the reference tank, P₀ denotes the initial system pressure (Pa), P₁ indicates the pressure (Pa) after pressurization of the reference tank, and P₂ is the equilibrium pressure (Pa) after opening the equalization valve.

2.

Measurement of Adsorption Isotherms

The procedure for determining the adsorption isotherm followed a method similar to that used for free volume measurement. After determining the free volume, valves V₂ and V₃ and the vacuum pump were opened to evacuate helium from the system. The resulting initial system pressure was recorded as P₃. Valve V₂ was then closed, and the gas inlet valve V₁ was opened to pressurize the system to P₄. Subsequently, valve V₂ was opened to allow gas to enter the adsorption tank, initiating the adsorption process. Once the pressure stabilized and no further changes were observed, the equilibrium pressure was recorded as P₅. This procedure was repeated to collect 9–12 pressure points. The amount adsorbed was calculated based on the change in the number of moles before and after adsorption. Prior to pressure equalization, the change in the number of moles introduced into the system satisfies Equation (2):

$$\left(P_4-P_3\right)V_r/1000=n_{1}RT$$

(2)

Following the opening of the equalization valve and the attainment of adsorption equilibrium, the amount of gas remaining in the system is given by the following Equation (3):

$$\left(P_5-P_3\right)\left(V_r+V_0\right)/1000=n_{2}RT$$

(3)

The static adsorption capacity at equilibrium pressure P₅ is defined as the difference in the number of moles before and after pressure equalization, given by:

$$q_i={\displaystyle \frac{n_1-n_2}{m}}$$

(4)

In Equation (4), q_i denotes the equilibrium adsorption capacity (mmol·g⁻¹) at equilibrium pressure P₅; n₁ represents the amount of substance (mmol) of gas introduced into the system; n₂ is the amount of substance (mmol) of gas remaining in the system after adsorption equilibrium; and m refers to the mass (g) of the adsorbent loaded in the adsorption tank.

2.2.2. Selectivity

Selectivity is an important parameter of adsorbents, and its definition is provided in Equation (5):

$$s={\displaystyle \frac{x_i/x_j}{y_i/y_j}}$$

(5)

where: s refers to the separation performance of the adsorbent for binary-component adsorbates; a larger value of s indicates a greater ease of separating the mixed system. x_i and x_j denote the mole fractions of component i and component j in the adsorbed phase at adsorption equilibrium, respectively. y_i and y_j denote the mole fractions of component i and component j in the gas phase at adsorption equilibrium, respectively.

The adsorption isotherms of CO₂ and N₂ were fitted using the Langmuir equation:

$$q={\displaystyle \frac{q_m\ast b\ast p_i}{1+b\ast p_i}}$$

(6)

where q refers to the equilibrium adsorption capacity, with the unit of mmol·g⁻¹; q_m denotes the saturation adsorption capacity, with the unit of mmol·g⁻¹; b represents the adsorption constant, with the unit of Pa⁻¹; p_i is the partial pressure of component i, with the unit of Pa.

Based on the parameters of the adsorption isotherms, the adsorption capacity of each component at a specified pressure was calculated first, and the resulting values were then substituted into Equation (5) to determine the selectivity of the adsorbent.

3. Results and Discussion

3.1. Porosity Characterization

Figure 2 shows the N₂ adsorption–desorption isotherms of the synthesized NC_R-X-Y series samples. A sharp increase in adsorption capacity was observed at low relative pressure (P/P₀ < 0.01), followed by a nearly horizontal plateau, which is characteristic of a Type I isotherm and indicates the predominantly microporous nature of the materials [56].

As illustrated in

Table 1. Porous structure of series samples.

Samples	SBET (m2/g)	VO (cm3/g)	Vt (cm3/g)	daverage
NC0-500-600	542.846	0.256	0.194	1.884
NC0.5-500-600	556.227	0.256	0.198	1.840
NC1-500-600	576.757	0.283	0.204	1.965
NC2-500-600	232.245	0.142	0.085	2.454
NC3-500-600	143.828	0.125	0.058	3.481
NC2-450-600	195.730	0.145	0.068	2.955
NC2-500-600	232.245	0.142	0.085	2.454
NC2-550-600	334.090	0.196	0.116	2.344
NC2-600-600	272.812	0.141	0.100	2.073
NC2-500-600	232.245	0.142	0.085	2.454
NC2-500-650	507.477	0.259	0.187	2.045
NC2-500-700	709.741	0.363	0.265	2.045

, with increasing melamine doping ratio, the specific surface area of the samples initially increased slightly and then decreased significantly. This trend can be attributed to the fact that a moderate amount of melamine produces nitrogen-containing gases and other volatile compounds during pyrolysis. These gases react with carbon atoms on the carbon surface, and their subsequent release etches the carbon framework, thereby synergistically promoting the formation and expansion of micro- and mesopores. However, excessive melamine leads to the accumulation of solid carbonaceous residues that block the pores, resulting in a noticeable reduction in specific surface area.

The specific surface area of the samples increased with rising carbonization and activation temperatures, with the activation temperature exhibiting a more pronounced influence. This trend can be attributed to the distinct roles of each process: carbonization primarily involves the thermal decomposition of the precursor, eliminating volatile components and facilitating structural reorganization. Since N₂ serves solely as an inert protective gas and does not participate in the reaction, the carbonization temperature mainly affects the preliminary formation of the pore structure, contributing only modestly to the final specific surface area. In contrast, during activation, CO₂ reacts with carbon atoms on the material surface, resulting in etching and substantial development of the porous architecture. Higher activation temperatures accelerate the reaction kinetics and intensify the etching effect, thereby significantly enhancing the porosity and specific surface area of the activated carbon.

The pore size distribution (PSD) curves of the samples are shown in Figure 3. As the melamine mass increased, the PSD initially broadened and then narrowed, indicating that melamine can, to some extent, promote pore development at moderate amounts, whereas excessive amounts lead to pore blockage. Both higher carbonization and activation temperatures resulted in broader PSD profiles. A comparison between samples NC_2-450-600 and NC_2-500-550 revealed that even at the lowest carbonization temperature, some microporosity was developed, whereas at the lowest activation temperature, virtually no micropores were detected. This further confirms that the activation temperature plays a more critical role in determining the final porous structure of the activated carbon.

3.2. Elemental Composition and Surface Chemistry

To investigate the influence of synthesis conditions on the composition and properties of the adsorbents, elemental analysis (EA) was performed on the series of samples, with selected representative samples further analyzed by X-ray photoelectron spectroscopy (XPS). As summarized in

Table 2. Elemental composition and CO₂ uptake of series samples.

Samples	C	N	H	O	CO2 Uptake (mmol/g)
					101.325 kPa	15 kPa
NC0-500-600	84.020	1.090	1.267	13.153	2.162	0.831
NC0.5-500-600	73.090	4.750	1.383	20.603	2.306	0.831
NC1-500-600	71.430	6.870	1.648	19.871	2.208	0.827
NC2-500-600	61.490	21.050	1.395	15.95	1.973	1.037
NC3-500-600	57.530	25.860	1.402	15.088	1.683	0.915
NC2-450-600	60.250	23.520	1.285	14.805	1.676	0.766
NC2-500-600	61.490	21.050	1.395	15.95	1.973	1.037
NC2-550-600	62.030	22.490	1.419	13.946	1.927	0.946
NC2-600-600	63.840	20.720	1.455	13.873	1.916	0.812
NC2-500-550	58.620	25.9	1.388	14.95	1.491	0.835
NC2-500-600	61.490	21.05	1.395	15.95	1.973	1.037
NC2-500-650	59.980	19.1	1.64	16.95	2.298	1.051
NC2-500-700	64.390	10.37	1.911	17.95	2.455	0.575

, the nitrogen content of the N-doped sample NC_0-500-600 was 1.09 wt%, while the melamine-modified samples exhibited significantly higher N content, ranging from 4.75 wt% to 25.8 wt%. This indicates that the pristine precursor contains negligible nitrogen, and melamine effectively serves as a source for nitrogen incorporation into the carbon framework.

With the increase in the melamine-to-precursor mass ratio (R) from 0 to 3, the nitrogen content in the samples increased from 1.09 wt% to 25.86 wt%, while the carbon content decreased from 84.02 wt% to 57.53 wt%. This inverse trend can be attributed to the partial substitution of carbon atoms by nitrogen within the carbon framework, along with the chemical adsorption of nitrogen-containing gaseous species on the carbon surface during pyrolysis, collectively leading to enhanced nitrogen incorporation at the expense of carbon content. Furthermore, the oxygen content in the nitrogen-doped samples was generally higher than that in the undoped sample, though it exhibited a declining trend with increasing R. The undoped sample had an oxygen content of 13.153 wt%. At R = 0.5 and 1, the oxygen content increased to 20.603 wt% and 19.871 wt%, suggesting that nitrogen doping promotes the formation of oxygen-containing functional groups. However, as R increased further to 3, the oxygen content decreased to 15.088 wt%, implying that excessively high nitrogen loading suppresses oxygen introduction and functional group formation.

With the mass ratio R fixed at 2 and the activation temperature held at 600 °C, variations in carbonization temperature were investigated. As the carbonization temperature increased, the carbon content of the samples rose from 60.25 wt% to 63.84 wt%, while the nitrogen content decreased from 23.52 wt% to 20.72 wt%. These relatively minor alterations in elemental composition suggest that the carbonization temperature has limited influence on the chemical structure of the samples, which is consistent with the BET surface area results. Similarly, with R fixed at 2 and the carbonization temperature maintained at 500 °C, an increase in activation temperature led to a noticeable decrease in nitrogen content from 25.9 wt% to 10.37 wt%, accompanied by an increase in both carbon and oxygen contents. This behavior can be attributed to accelerated thermal decomposition of melamine at higher temperatures, promoting the release of nitrogen in gaseous forms and reducing its incorporation into the carbon framework. Additionally, nitrogen species chemically adsorbed on the carbon surface are prone to desorption as temperature rises. Concurrently, the heterogeneous reaction between CO₂ and the carbon surface may enhance the chemical adsorption of oxygen, contributing to the observed increase in oxygen content [57].

Based on the CO₂ adsorption capacity and selectivity under partial pressure, three representative samples were selected from the series for further investigation into the effects of synthesis conditions on adsorbent performance and structure: NC_2-500-600, NC_2-550-600, and NC_2-500-700.

To elucidate the nitrogen speciation on the carbon surface, XPS analysis (Figure 4) was conducted on the representative samples. All three samples exhibited distinct peaks at binding energies of 398.3–398.6 eV and 400.1–400.2 eV, which are attributed to pyridinic-N (N-6) and pyrrolic-N (N-5), respectively. These results indicate that nitrogen is primarily present in the form of N-5 and N-6 configurations, with negligible contributions from oxidized nitrogen or quaternary nitrogen species.

As shown in Figure 4a,b, under the same carbonization temperature, an increase in the activation temperature from 600 °C to 700 °C led to a noticeable increase in the content of pyrrolic-N (N-5). Comparison between Figure 4a,c further reveals that, at a fixed activation temperature, raising the carbonization temperature promoted a partial transition from pyridinic-N (N-6) to pyrrolic-N (N-5), although this change was not pronounced. This transformation suggests that during thermal treatment, nitrogen species undergo a transition from the less stable N-6 state to the more energetically favorable N-5 configuration, accompanied by partial nitrogen loss through decomposition. These findings are consistent with the elemental analysis results and the applied synthesis conditions.

3.3. Surface Morphology and Phase Analysis

The SEM and TEM characterization results of the representative samples NC_2-500-600, NC_2-500-700, and NC_2-550-600 are presented in Figure 5. The images indicate that the melamine-doped adsorbents possess well-developed porous structures with an abundance of fine micropores.

Figure 6 shows the X-ray diffraction (XRD) pattern of sample NC_2-500-600, which exhibits a broad peak centered at 2θ = 22.5° and a weaker diffuse peak around 2θ = 43°. These features correspond to the (002) and (100) diffractions of amorphous carbon. These features indicate the presence of short-range ordered structures with low crystallinity, which is characteristic of amorphous carbon materials. This amorphous nature was further confirmed by TEM analysis (Figure 5d), where numerous randomly distributed worm-like micropores were observed, unequivocally verifying the disordered structure of the prepared N-doped carbon. Such a disordered structure is conducive to the development of a high level of porosity, thereby facilitating CO₂ adsorption [58,59].

3.4. Adsorption Performance

3.4.1. CO₂ and N₂ Adsorption Isotherms

The CO₂ and N₂ adsorption isotherms of the adsorbents were measured at 25 °C using a static volumetric adsorption apparatus, as shown in Figure 7a,b. As the nitrogen doping ratio increased, the nitrogen content of the samples rose, while the specific surface area initially increased slightly and then decreased significantly. Accordingly, the adsorption capacities for both CO₂ and N₂ under identical pressures first increased and then decreased, making it difficult to discern the individual contributions of nitrogen content and specific surface area to the CO₂ adsorption capacity.

Figure 7c,d demonstrate that CO₂ adsorption capacity first increases and then decreases with rising carbonization temperature, with an optimum of 1.973 mmol·g⁻¹ observed for NC_2-500-600. Notably, as the carbonization temperature increased, the specific surface area increased while the nitrogen content decreased, indicating that the CO₂ adsorption capacity is a multi-factorial process.

From Figure 7e,f, it is evident that higher activation temperatures generally lead to greater CO₂ and N₂ adsorption capacities. However, the CO₂ adsorption isotherm in Figure 7e shows that although the sample activated at 700 °C achieved the highest adsorption capacity at high pressure, it exhibited the lowest capacity in the low-pressure region. This behavior can be attributed to excessive activation temperature, which promotes pore widening and reduces the proportion of narrow micropores, thereby diminishing low-pressure adsorption performance.

As summarized in Table 2, the CO₂ adsorption capacities of the adsorbents at 1 bar ranged from 1.491 to 2.455 mmol·g⁻¹, while the capacities at 15 kPa varied from 0.575 to 1.051 mmol·g⁻¹. For most samples, the adsorption capacity at 15 kPa reached nearly half of that at 100 kPa. The high adsorption performance in the low-pressure region can be attributed to the filling of CO₂ in narrow micropores [60]. It is noteworthy that sample NC_2-500-700, which showed the highest adsorption capacity at 100 kPa, exhibited the lowest at 15 kPa. This indicates that higher activation temperatures promote more thorough reaction between CO₂ (as an activating agent) and the carbon product, resulting in a reduction in narrow micropores suitable for low-pressure adsorption, which is disadvantageous for separation under flue gas conditions. Given that the CO₂ partial pressure in typical flue gas is approximately 15 kPa, the selection of the optimal sample was based on a combination of low-pressure adsorption capacity and separation selectivity rather than performance at 1 bar. Thus, instead of choosing NC_2-500-700, which shows the highest capacity at 1 bar, NC_2-500-600 was selected for further investigation of adsorption kinetics and cyclic stability. This decision was based on its superior performance under flue gas conditions, achieving an adsorption capacity of 1.037 mmol·g⁻¹ and a separation selectivity of up to 40.

Based on the analysis of CO₂ adsorption isotherms and structural characterization data of the NC_R-X-Y series samples, it can be concluded that the CO₂ adsorption capacity is influenced by multiple factors, including the pore size distribution, specific surface area, micropore volume, and nitrogen doping level. The correlation between CO₂ uptake and individual structural parameters is illustrated in Figure 8. The results indicate positive yet non-linear relationships between the CO₂ adsorption capacity and the specific surface area, micropore volume, and total pore volume. Under identical carbonization and activation temperatures, the CO₂ adsorption capacity initially increases and then decreases with rising nitrogen content. The sample with the highest CO₂ uptake, NC_2-500-700, has a nitrogen content of 10.37 wt%. Notably, the sample with the lowest nitrogen content, NC_0-500-600, does not exhibit the lowest adsorption capacity. Further analysis incorporating specific surface area and pore size distribution data of adsorbents with different nitrogen doping ratios suggests that while nitrogen doping enhances CO₂ adsorption, excessive amounts of melamine can lead to pore blockage. Beyond an optimal point, this negative effect outweighs the beneficial contribution of nitrogen functional groups to CO₂ capture.

3.4.2. CO₂/N₂ Selectivity

The separation selectivity toward CO₂ over N₂ is a critical performance metric for adsorbents in gas separation applications. Higher selectivity values correspond to more efficient CO₂ capture from mixed gas streams. The CO₂ and N₂ adsorption isotherms were well-fitted by the Langmuir model, and the corresponding parameters are summarized in

Table 3. Langmuir equation fitting parameters for series samples.

Samples	CO2			N2
	qm	b	R2	qm	b	R2
NC0-500-600	3.397	1.723 × 10−5	0.9981	1.824	2.267 × 10−6	0.9999
NC0.5-500-600	3.334	2.215 × 10−5	0.9968	1.839	2.37 × 10−6	0.9999
NC1-500-600	3.111	2.414 × 10−5	0.9966	1.89	2.039 × 10−6	1
NC2-500-600	2.339	5.312 × 10−5	0.9964	1.228	2.382 × 10−6	1
NC3-500-600	1.97	5.783 × 10−5	0.9928	1.096	2.637 × 10−6	0.9998
NC2-450-600	2.143	3.559 × 10−5	0.9925	1.198	3.212 × 10−6	0.9998
NC2-550-600	2.351	4.487 × 10−5	0.9936	1.262	2.171 × 10−6	1
NC2-600-600	2.51	3.816 × 10−5	0.9952	1.335	2.673 × 10−6	1
NC2-500-550	1.727	6.234 × 10−5	0.9859	0.8561	2.875 × 10−6	0.9994
NC2-500-650	2.895	3.8 × 10−5	0.9867	1.661	2.163 × 10−6	1
NC2-500-700	5.675	7.523 × 10−6	0.9962	2.025	1.87 × 10−6	1

. The high fitting accuracy suggests that the adsorption behaviors of both CO₂ and N₂ on the NC_R-X-Y materials comply with the assumptions of ideal monolayer adsorption under the Langmuir framework.

The CO₂/N₂ selectivity of the NC_R-X-Y series samples for dry flue gas containing 15% CO₂ was calculated based on the Ideal Adsorbed Solution Theory (IAST), with the results presented in Figure 9.

As illustrated in Figure 9a, which depicts the relationship between selectivity and the nitrogen doping ratio, all nitrogen-doped adsorbents exhibited higher selectivity compared to the undoped sample, indicating that nitrogen incorporation enhances the separation performance. The selectivity increased consistently as the doping ratio rose from 0 to 2. However, at a doping ratio of 3, the selectivity decreased, which can be attributed to excessive melamine leading to pore blockage and consequently impairing CO₂ adsorption. Figure 9b shows that the selectivity also exhibited an initial increase followed by a decrease with rising carbonization temperature, a trend consistent with the variation in adsorption capacity. The optimal performing sample, NC_2-500-600, achieved a selectivity exceeding 40. Figure 9c reveals that higher activation temperatures resulted in reduced selectivity. Although elevated activation temperatures promote the development of porosity and increase overall adsorption capacity, they also cause the enlargement of narrow micropores, thereby reducing molecular sieving ability and ultimately diminishing separation performance.

3.4.3. Cyclic Stability

Cyclic stability is a critical performance indicator for adsorbents in industrial applications. To evaluate this property, five consecutive adsorption–desorption cycles were conducted on the optimal sample, NC_2-500-600, at 25 °C. The resulting adsorption isotherms, shown in Figure 10, nearly overlap across all cycles, demonstrating exceptional regeneration. The CO₂ uptake measured at 100 kPa was 1.973 mmol·g⁻¹ in the first cycle and 1.965 mmol·g⁻¹ in the fifth cycle, representing a capacity loss of less than 1%. These results confirm that NC_2-500-600 possesses outstanding cyclic stability and excellent reversibility.

3.4.4. CO₂ Adsorption Kinetics

Adsorption kinetics play a crucial role in determining the productivity of adsorbents and the separation efficiency of gas mixtures. An ideal CO₂ adsorbent must exhibit rapid adsorption kinetics.

The kinetic performance of the optimal sample, NC_2-500-600, was evaluated at 25 °C, as shown in Figure 11. The adsorbent demonstrated exceptionally fast adsorption rates, reaching 90% of its equilibrium adsorption capacity within 3 min and 95% within 5 min. Such rapid kinetics significantly shorten the adsorption cycle duration and enhance process efficiency, underscoring the high potential of NC_2-500-600 for industrial CO₂ capture applications.

3.5. Comparison with Other Resin-Derived Carbons

The properties of carbon materials derived from various resins in different studies are summarized in

Table 4. The adsorption performance of other resin-derived carbons.

Materials	Activation	Adsorption Conditions	CO2 Adsorption Capacity	Selectivity	Ref
		Temperature/Pressure
MFRC-650-1	KOH	25 °C/1 bar	3.52 mmol/g	24	[62]
WIRAC	KOH	20 °C/1 bar	0.729 mmol/g	-	[63]
MR (0.25)/GO-500	KOH	25 °C/500 kpa	5.21 mmol/g	19	[64]
RNK-550-2	KOH	25 °C/1 bar	5.01 mmol/g	19	[43]
RUK-600-3	KOH	25 °C/1 bar	4.61 mmol/g	12	[65]
MFC-700-0.2	KOH	25 °C/1 bar	3.30 mmol/g	17	[66]
UFCA-2-600	KOH	25 °C/1 bar	3.21 mmol/g	-	[47]
NC2-500-600	CO2	25 °C/1 bar	1.973 mmol/g	28	This paper
NC2-500-650	CO2	25 °C/1 bar	2.298 mmol/g	23	This paper
NC2-500-700	CO2	25 °C/1 bar	2.455 mmol/g	11	This paper

. While some of the reported carbons demonstrate considerable CO₂ uptake, they generally share common limitations with most existing works, such as reliance on chemical activation routes or complex synthesis procedures. In contrast, the N-doped carbon developed in this work through physical activation not only achieves competitive CO₂ adsorption capacity and high selectivity, but also benefits from a straightforward and cost-effective process [61]. These advantages position our material as a promising candidate for practical industrial implementation.

4. Conclusions

Nitrogen-doped activated carbons were synthesized using phenolic resin as the precursor and melamine as the nitrogen source. By varying the melamine doping ratio, carbonization temperature, and activation temperature, a series of samples was prepared and systematically characterized in terms of structure, composition, and adsorption performance:

(1)

An appropriate amount of nitrogen doping enhances CO₂ adsorption performance, whereas excessive doping leads to pore structure deterioration and reduced adsorption capacity.

(2)

Carbonization temperature had a limited effect on pore development, and higher activation temperatures promoted more extensive porosity but also broadened the pore size distribution, reducing the proportion of narrow micropores and thus diminishing low-pressure adsorption capacity and separation selectivity.

(3)

The obtained materials were amorphous activated carbons with abundant micropores and high specific surface area. Nitrogen was primarily incorporated as pyridinic-N (N-6), and its content increased with the doping ratio but decreased at elevated carbonization and activation temperatures.

The results indicate the CO₂ adsorption performance was collectively influenced by microporous volume, total pore volume, specific surface area, micropore size distribution, nitrogen content, and nitrogen configuration. Under the conditions of 25 °C and 1 bar, the optimal sample achieved a CO₂ adsorption capacity of 2.455 mmol·g⁻¹ and a separation factor exceeding 28. In addition, the adsorbent exhibited excellent cyclic stability, and rapid adsorption kinetics. Furthermore, the proposed synthesis route offers notable advantages for large-scale application. The precursors, phenolic resin and melamine, are inexpensive and readily available. The process itself is straightforward, involving only carbonization followed by physical activation using CO₂, thereby avoiding the equipment corrosion, chemical pollution, and wastewater issues. These attributes underscore the industrial potential of the developed adsorbents, particularly for CO₂ capture under realistic flue gas conditions.