Selective Adsorption of VOCs/Water Vapor on Activated Carbon: The Role of Adsorbent and VOC Molecular Polarity

Abstract

The presence of abundant water vapor in industrial organic waste gases greatly reduces the selective adsorption of volatile organic pollutants (VOCs). The polarity of the adsorbent and VOC molecules plays an important role in the adsorption process, especially in the presence of water vapor. In this paper, commercial coconut shell activated carbon (CSC) was modified by a thermal reduction treatment to obtain heat-treated coconut shell activated carbon (HCSC). CSC and HCSC exhibited similar pore structure characteristics but differed significantly in surface oxygen content (10.97% and 7.55%, respectively). Dynamic adsorption breakthrough experiments were conducted to determine the dynamic adsorption capacities of toluene on both adsorbents under varying relative humidity levels. HCSC demonstrated superior toluene/water vapor adsorption selectivity. Further analyses of toluene adsorption kinetics, activation energy, and water vapor adsorption isotherms revealed that the lower surface oxygen functional group content of HCSC resulted in a weaker surface polarity, facilitating the adsorption of weakly polar toluene. This was attributed to stronger toluene–HCSC interactions and weaker water–HCSC interactions. The dynamic adsorption capacities of three VOCs with varying polarities were also tested on HCSC. The observed VOC/water vapor adsorption selectivity had the following order: toluene > n-heptane > 1,2-dichloroethane. Grand Canonical Monte Carlo (GCMC) simulations were employed to quantify the relationship between the adsorption selectivity of eight VOCs with varying polarities and their molecular polarity. The results indicated a decrease in adsorption selectivity with increasing VOC polarity. A mechanistic analysis suggests that more polar VOCs prefer to adsorb polar oxygen-containing functional groups, competing with water molecules for adsorption sites. Under high humidity, hydrogen bonding leads to the formation of water clusters, exacerbating this competition. This research holds significant implications for the efficient selective adsorption of VOCs with varying polarities in humid industrial conditions.

1. Introduction

Volatile organic compounds (VOCs) are important precursors, playing a crucial role in the formation of complex atmospheric pollution [1,2]. Many organic pollutants are carcinogenic, mutagenic, and teratogenic. Long-term exposure to VOCs may pose significant risks to health even at low concentrations [3]. Sources of VOC emissions primarily include industrial exhaust, solvent use, and the combustion of coal and biomass fuels [4,5]. Increasing emissions pose a significant threat to air quality and human health [6]. Controlling VOC emissions is a major research focus, particularly in industrial processes [7].

Industrial VOCs typically have low concentrations and high gas flow. Adsorption technology is widely used for VOC abatement due to its high effectivity in adsorption in low-concentration, high-volume gas streams [8]. Activated carbon, with its high surface area, well-developed pore structure, and chemical stability, is a widely used and highly effective adsorbent in gas and water purification [9,10]. However, complex industrial waste gases usually contain diverse VOCs and water vapor. It is necessary to enhance the adsorptive selectivity of activated carbon. The adsorption selectivity of activated carbon towards VOCs is governed by both the VOCs’ molecular structure and physicochemical properties, as well as the surface functional groups of activated carbon. Prior research indicates that the surface chemistry of activated carbon, specifically the quantity and type of polar functional groups (e.g., carboxyl and hydroxyl), significantly influences its adsorption capacity for various polar VOCs [11]. Yang et al. reported that O/N doping enhances the surface polarity of activated carbon, improving the capture of acetone [12]. Water vapor further complicates the adsorption process. The competition for micropores and polar adsorption sites between water vapor and VOCs reduces the VOC adsorption capacity. Furthermore, the polarity of water molecules alters the surface physicochemical properties of activated carbon, influencing its VOC adsorption selectivity [13,14,15]. Yan et al. [16] investigated the adsorption performance of activated carbon fibers under humid conditions, observing a significant negative impact of humidity. This impact varied among different VOCs, which is more remarkable for polar VOCs.

The hydrophobic modification of the adsorbent is a common approach to enhancing the selectivity of activated carbon for VOCs/water vapor. Hydrophobic coating is a widely used modification technique. Li et al. [17] modified walnut shell activated carbon using polydimethylsiloxane (PDMS), finding an increase in surface Si and O with the formation of Si-O-Si. At 90% relative humidity (RH), the adsorption capacity showed a 55.9% decrease in benzene on unmodified activated carbon, compared to only a 19.3% decrease on the modified material. Park et al. [18] demonstrated that PDMS coating could reduce the specific surface area of activated carbon by 7–58%, depending on the treatment temperature. While PDMS coatings significantly enhance hydrophobicity, the substantial changes in pore structure also make it difficult to assess the influence of adsorbent and VOC polarity on competitive adsorption. Thermal reduction is another effective hydrophobic modification method, due to its ability to reduce surface oxygen functional groups while preserving pore structure. Unlike chemical modification methods, thermal treatment avoids introducing foreign elements and maintains the intrinsic hydrophobicity of carbon. Studies have shown that high-temperature reduction treatment can improve its toluene adsorption capacity; at 50% relative humidity, its capacity is 83% of the dry state value (for 4100 mg/m³ toluene). Considering the damage of pores due to high temperatures, the treatment temperature is generally limited to below 800 °C [19]. Existing research primarily focuses on the adsorption of porous carbon for high-concentration VOCs [20], with limited studies on low-concentration VOCs. Furthermore, a systematic understanding of the relationship between the selectivity of VOC/water vapor on activated carbon and the polarities of both the adsorbent and VOC molecules under humid conditions is worth exploring.

This study utilizes a thermal reduction treatment to prepared two adsorbents with different polarities but similar pore structures to investigate the adsorption selectivity of various VOCs/water vapor on activated carbon, focusing on the polar influence of adsorbent and VOC molecules. Furthermore, Grand Canonical Monte Carlo (GCMC) simulations are employed to quantify the correlation between VOC/water vapor adsorption selectivity and VOC molecular polarity. This study could enhance the adsorption efficiency and selectivity of activated carbon for target VOCs.

2. Materials and Methods

2.1. Samples and Reagents

Commercial coconut shell activated carbon was obtained from Jiangsu Zhuxi Activated Carbon Co., Ltd. (Changzhou, China). Toluene was purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China) Commercial coal-based activated carbon was boiled in deionized water for 60 min and then dried at 110 °C for 12 h to obtain sample CSC. Sample CSC was subjected to calcination at 700 °C for 1 h under a nitrogen atmosphere, with a heating rate of 10 °C/min, obtaining sample HCSC.

2.2. Characterization of Adsorbents

Nitrogen adsorption–desorption isotherms were measured at 77 K using a fully automatic surface area analyzer (ASAP 2020C, Micromeritics, GA, USA). Specific surface area was determined using the BET equation, and total pore volume was calculated from the adsorption isotherm data at P/P₀ = 0.99. Micropore size distribution and micropore volume were determined using the Horvath–Kawazoe (HK) method. Water vapor adsorption isotherms were measured at 40 °C using the same instrument. X-ray photoelectron spectroscopy (XPS, ESCLAB 250xi, Thermo, Waltham, MA, USA) was used to determine the surface elemental composition (C, N, and O) and relative content of oxygen-containing functional groups using Al Kα radiation (1486.6 eV) as the excitation source.

2.3. Dynamic Adsorption Capacity Evaluation

Dynamic breakthrough experiments were used to investigate the adsorption behavior of various VOCs on the adsorbents on a fixed-bed adsorption system. As illustrated in Figure 1, the system comprised four main components: a VOC generator, a water vapor generator, an adsorption reactor, and a detection system. VOCs were generated through a temperature-controlled heater, vaporized, and carried by nitrogen gas steam. This stream was then mixed with a balanced nitrogen flow in a mixing chamber to achieve the desired VOC concentration. Humidity was controlled using a water vapor generator (micro-injection pump + heater) to achieve target relative humidity (RH) values. The gas mixture was passed through a temperature/humidity sensor (Figure 1) to monitor and stabilize RH during experiments. The adsorption reactor had an inner diameter of 1 cm, with a gas flow rate of 1 L/min and a toluene concentration of 120 mg/m³. Outlet VOCs were analyzed using a non-methane total hydrocarbon analyzer (PF-300, Pollution, Budrio, Italy) equipped with a flame ionization detector (FID). All of experiments were repeated at least three times. Adsorption capacity was calculated using the following equation:

$${\mathrm{Q}}_{\mathrm{t}}={\displaystyle \frac{1}{\mathrm{m}}}{\int }_0^{\mathrm{t}}\mathrm{F}({\mathrm{C}}_0-\mathrm{C})\mathrm{d}\mathrm{t}$$

(1)

where Q_t is the adsorption capacity at time t (mg/g); t is the adsorption time (min); F is the gas flow rate (L/min); C₀ is the inlet VOC concentration (mg/m³); and C is the outlet VOC concentration at time t (mg/m³).

2.4. Molecular Simulation Method

Grand Canonical Monte Carlo (GCMC) simulations are widely used to study VOC adsorption on porous materials at specified temperatures and pressures [21,22]. Using Materials Studio software 2020 and GCMC method, competitive adsorption of VOCs and water vapor was investigated using the Sorption module with a fixed-pressure adsorption isotherm approach. A slit-pore model was used to represent the activated carbon, with the simulation temperature at 313 K. In each Metropolis Monte Carlo step, creation, rotation, and translation were attempted with equal probability. Initially, 5 × 10⁶ Monte Carlo steps were used for energy minimization, followed by 5 × 10⁶ steps for production. Adsorbate conformations were restricted to 5° rotations and 1 Å translations within the Metropolis Monte Carlo framework. The Compass force field was used for energy calculations, employing the Ewald method for electrostatics and an atom-based method for van der Waals interactions. A 12.5 Å cutoff was used for non-bonded interaction calculations. The minimum-energy configuration of the adsorbate molecules within the adsorbent model was obtained.

3. Results and Discussion

3.1. Adsorbent Characteristics

Figure 2a presents the N₂ adsorption–desorption isotherms and pore size distributions of the samples. Both CSC and HCSC exhibit type I isotherms, a characteristic of microporous materials. The presence of hysteresis loops in both isotherms suggests the presence of mesopores. The nitrogen adsorption capacity at saturation for HCSC is slightly lower compared to that for CSC, indicating a minor reduction in surface area and total pore volume of CSC after the heat treatment. This may be due to the collapse of some pores at high temperatures. Figure 2b shows the specific surface area and pore structure parameters. Both samples are predominantly microporous (approximately 75%) and have a high proportion of narrow micropores. The similarity in pore structure between the CSC and HCSC allows us to analyze the impacts of the adsorbent and molecular polarity better.

The element relative content of the adsorbents surface was determined by X-ray photoelectron spectroscopy (XPS), with the C1 high-resolution spectra shown in Figure 2c,d. The relative content of C, N, O, and various oxygen-containing functional groups was also obtained. The C1 peak was deconvoluted into four components with binding energies of 284.7 eV (C=C, graphitic carbon), 285.5 eV (C-O, alcohol and ether carbons), 287.1 eV (C=O, carbonyl carbons), and 289.4 eV (O=C-O, carboxyl or ester carbons) [23]. The heat treatment resulted in a significant decrease in the surface oxygen concentration from 10.12% to 6.72% and a corresponding increase in graphitic carbon. The decrease in the relative content of O=C-O and C-O suggests the thermal decomposition of oxygen-containing functional groups. This aligns with previous reports that the surface functional groups could decompose at high temperatures [24,25]. The decomposition temperatures range from 150 to 300 °C for carboxyl groups, 300 to 500 °C for anhydrides and lactones, and 500 to 800 °C for ethers, phenolic hydroxyls, and carbonyls [25]. The high-temperature treatment also enhanced the graphitization degree of the activated carbon, increasing its hydrophobicity.

3.2. Effect of Adsorbent Polarity on the Adsorption Performance of Toluene/Water Vapor

Toluene was used as the adsorbate to investigate the effect of the adsorbent polarity. Figure 3 shows the toluene breakthrough curves on CSC and HCSC under varying humidity conditions with 0.5 g adsorbent loading. With increasing relative humidity, toluene adsorption is increasingly affected. At 60 RH%, the adsorption capacity of toluene significantly decreased. Compared to 20 RH%, the adsorption capacity decreased 17% on CSC and 8% on HCSC. At 80 RH%, the decreases were 68% and 54%, respectively. The heat-treated HCSC with low polarity outperformed the pristine activated carbon (CSC) in toluene adsorption under different humidities, exhibiting greater toluene/water vapor selectivity. Compared to the other reported adsorbents, HCSC also showed excellent water resistance in the process of toluene adsorption, as listed in

Table 1. The ratio of wet/dry adsorption capacities of HCSC and some other materials for toluene.

Samples	RH	Q/QRH = 0%	Reference
HCSC	60%	86%	This work
	80%	40%	This work
CSC	60%	74%	This work
	80%	23%	This work
UiO-66	70%	63%	[26]
CTAB-U-0.5	70%	17%	[27]
ZSM-5	50%	36%	[28]
TS−1@LDH	50%	44%	[28]

.

To further elucidate the effect of adsorbent polarity, the adsorptions of toluene and water vapor were investigated separately. Figure 4 shows the breakthrough curves for toluene adsorption (120 mg/m³) at 40 °C on both adsorbents under dry conditions. The toluene adsorption on HCSC exhibited a longer breakthrough time and higher capacity. HCSC and CSC exhibited similar surface areas, total pore volumes, and micropore volumes, suggesting that these parameters do not primarily govern low-concentration toluene adsorption onto the heat-treated activated carbon. This enhancement of toluene adsorption on HCSC is likely attributed to the significant reduction in oxygen-containing functional groups after heat treatment, leading to decreased surface polarity and the improved adsorption of nonpolar toluene. To further investigate low-concentration toluene adsorption kinetics, nonlinear fitting was performed using pseudo-first-order (Equation (2)), pseudo-second-order (Equation (3)), and Elovich (Equation (4)) kinetic models [29,30].

$${\mathrm{Q}}_{\mathrm{t}}={\mathrm{Q}}_{\mathrm{e}}\times (1-{\mathrm{e}}^{-\mathrm{k}\mathrm{t}})$$

(2)

$${\displaystyle \frac{\mathrm{t}}{{\mathrm{Q}}_{\mathrm{t}}}}={\displaystyle \frac{1}{\mathrm{k}{\mathrm{Q}}_{\mathrm{e}}^2}}+{\displaystyle \frac{1}{\mathrm{Q}\mathrm{e}}}\mathrm{t}$$

(3)

$${\mathrm{Q}}_{\mathrm{t}}={\displaystyle \frac{1}{\mathsf{\beta }}}\mathrm{l}\mathrm{n}\mathsf{\alpha }\mathsf{\beta }+{\displaystyle \frac{1}{\mathsf{\beta }}}\mathrm{l}\mathrm{n}\mathrm{t}$$

(4)

where Q_t is the amount adsorbed at time t (mg/g); Q_e is the equilibrium adsorption capacity (mg/g); k is the diffusion rate constant; α is the initial adsorption rate constant (mg·g⁻¹·min⁻¹); and β is the desorption rate constant (g·mg⁻¹).

Figure 4 shows the fitted curves, and the fitting parameters are summarized in

Table 2. Fitting parameters for three kinetic models describing toluene adsorption on activated carbon.

Models	Parameter	CSC	HCSC
Pseudo-first order	Qe (mg·g−1)	83.9476	124.0922
	K (min−1)	0.0065	0.0050
	r2	0.9917	0.9868
Pseudo-second order	Qe (mg·g−1)	122.1190	187.4180
	K (g·mg−1·min−1)	0.00003998	0.00001885
	r2	0.9859	0.9815
Elovich	α (mg·g−1·min−1)	1.5835	1.8511
	Β (g·mg −1)	0.0437	0.0307
	r2	0.9400	0.9215

. The pseudo-first-order kinetic model could better fit the breakthrough curves, suggesting that diffusion is the rate-limiting step [31]. The diffusion rate constant of toluene on HCSC is lower than that on CSC, which is related to the content of surface oxygen-containing functional groups. At low toluene concentrations (120 mg/m³), the oxygen-containing functional groups have a negative effect on toluene adsorption. The higher concentration of oxygen-containing functional groups on CSC hinders toluene adsorption, reducing diffusion resistance and resulting in a higher diffusion rate constant. Overall, at low toluene concentrations, the adsorption rate constant correlates with the concentration of oxygen-containing functional groups, indicating a positive correlation between the adsorption rate constant and adsorbent polarity.

The adsorption performance of low-concentration toluene at different temperatures is shown in Figure 4b,c. The toluene adsorption capacity decreased with increasing temperature. Pseudo-first-order kinetic fitting (

Table 3. The diffusion parameters and activation energies of toluene on samples.

Adsorbent	40 °C		60 °C		80 °C		Ea (kJ·mol−1)
	K (min−1)	r2	K (min−1)	r2	K (min−1)	r2
CSC	0.0065	0.9917	0.0088	0.9915	0.0119	0.9951	13.87
HCSC	0.0050	0.9868	0.0069	0.9873	0.0094	0.9878	14.49

) revealed that the adsorption rate constant increased with temperature. This is because surface adsorption involves molecular thermal motion, and higher temperatures result in faster molecular diffusion. The Arrhenius equation [32] (Equation (5)) was used to calculate the activation energies for toluene adsorption on CSC (13.87 kJ/mol) and HCSC (14.49 kJ/mol) (Figure 4d). The higher activation energy for toluene adsorption on HCSC indicates stronger toluene-adsorbent interactions. Meanwhile, it suggests that toluene has a larger diffusion barrier and a lower diffusion rate constant on the weak polar HCSC.

$$\ln \mathrm{k}=\mathrm{l}\mathrm{n}\mathrm{A}-{\displaystyle \frac{{\mathrm{E}}_{\mathrm{a}}}{\mathrm{R}\mathrm{T}}}$$

(5)

where k is the diffusion rate constant (min⁻¹), R is the gas constant, A is the Arrhenius pre-exponential factor, and E_a is the activation energy (kJ/mol).

To further clarify the adsorption performance of water molecules on the samples, water adsorption isotherm tests were performed, and the results are shown in Figure 5. A significant increase in water adsorption was observed above a relative pressure of 60%, consistent with the results of toluene adsorption at 60 RH%. HCSC exhibited lower water adsorption than CSC across the entire relative pressure range, demonstrating enhanced hydrophobicity. The enhanced hydrophobicity at low pressures is attributed to the reduction in oxygen-containing functional groups, decreasing water binding sites and increasing surface basicity. According to the hard–soft acid–base (HSAB) theory [33], the enhancement in basicity hinders the binding with hard base water.

The water adsorption isotherms were fitted using the Do and Do (DD) model (6) [34] and the cluster formation induced micropore filling (CIMF) model (7) [35], as shown in Figure 5. The relevant parameters are listed in

Table 4. Fitting parameters of adsorption isotherms of water vapor.

Model	Adsorbent	S0 (mmol/g)	Cμs (mmol/g)	Kf	Kμ	n	m	r2
CIMF	CSC	1.92	10.09	9.92	23.30	9	6.49	0.9997
	HCSC	0.77	12.33	7.60	15.34	9	7.25	0.9996
DD	CSC	1.50	14.07	18.75	8.22	8.48	5	0.9994
	HCSC	0.33	18.02	95.00	4.27	9.43	5	0.9978

. The DD model is not suitable for strongly hydrophobic activated carbons, while the CIMF model fits better [23]. The DD model is not an ideal fit for the curves of HCSC. The fitting correlation coefficients of the CIMF model for HCSC are above 0.999, confirming the strong hydrophobicity of HCSC. The parameter S₀ reflects the degree of surface oxidation. The lower S₀ value for HCSC indicates its lower surface oxidation. This is because the heat treatment removes some surface acidic oxygen-containing functional groups, reducing the surface oxidation degree.

$${\mathrm{C}}_{\mathsf{\mu }}={\mathrm{S}}_0{\displaystyle \frac{{\mathrm{K}}_{\mathrm{f}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{i}·\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{i}}}{1+{\mathrm{K}}_{\mathrm{f}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{i}}}}+{\mathrm{C}}_{\mathsf{\mu }\mathrm{s}}{\displaystyle \frac{{\mathrm{K}}_{\mathsf{\mu }}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^6}{{\mathrm{K}}_{\mathsf{\mu }}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^6+\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}}$$

(6)

$${\mathrm{C}}_{\mathsf{\mu }}={\mathrm{S}}_0{\displaystyle \frac{{\mathrm{K}}_{\mathrm{f}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{i}·\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{i}}}{1+{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{i}}}}+{\mathrm{C}}_{\mathsf{\mu }\mathrm{s}}{\displaystyle \frac{{\mathrm{K}}_{\mathsf{\mu }}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{m}+1}}{{\mathrm{K}}_{\mathsf{\mu }}{\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}^{\mathrm{m}+1}+\left(\frac{\mathrm{P}}{{\mathrm{P}}_0}\right)}}$$

(7)

where C_μ is the water adsorption capacity (mmol/g) at relative pressure P/P₀; S₀ is the concentration of primary adsorption sites (mmol/g); C_μs is the maximum water adsorption capacity (mmol/g); K_f and K_μ are the equilibrium constants for water adsorption onto functional groups and micropore filling, respectively; n is the average number of water molecules fully adsorbed per surface functional group; and m is the size of the water clusters adsorbed into the micropores.

3.3. The Influence of VOCs’ Molecular Polarity on Adsorption Capacity

The dynamic adsorption performance of three VOCs with different polarities was tested on HCSC under dry conditions and at 60 RH% to investigate the effect of molecular polarity on the adsorption selectivity of the VOCs/water vapor (Figure 6). The three VOCs were 1,2-dichloroethane, toluene, and n-heptane with dipole moments of 6.1, 1.0, and 0 Debye, respectively. The similar molecular mass and boiling points of these VOCs allowed us to better investigate the influence of polarity. Under both 0 RH% and 60 RH%, toluene and n-heptane exhibited significantly higher adsorption capacities than 1,2-dichloroethane, highlighting the significant influence of adsorbate polarity. This preferential adsorption is attributed to the nonpolarity of HCSC, favoring the adsorption of weak polar molecules like toluene and n-heptane. The toluene adsorption slightly exceeded that of n-heptane due to π-π interactions between the toluene aromatic ring and the abundant graphitic carbon on the HCSC. At 60 RH%, the adsorption capacity decreased compared to that of 0 RH%, and the amount of decrease varied among different VOCs. The adsorption capacity decreased by 23%, 17%, and 12% for 1,2-dichloroethane, toluene, and n-heptane, respectively, demonstrating the greater negative impact of water vapor on polar VOCs. Therefore, the adsorption selectivity order of VOCs/water vapor is toluene > n-heptane > 1,2-dichloroethane.

To further elucidate the relationship between molecular polarity and adsorption selectivity, single-component constant-pressure adsorption experiments were conducted using eight VOCs with varying polarities on an activated carbon model with a surface oxidation degree similar to HCSC. The VOC adsorption capacity and average isosteric heat of adsorption were determined. Combined with the obtained water adsorption capacity and isosteric heat of adsorption, the VOC/water vapor adsorption selectivity at 70 RH% was calculated using the DIH equation [36] (

Table 5. Physicochemical properties of adsorbates and VOC/water vapor adsorption selectivity.

Adsorbates	Relative Molecular Mass (g/mol)	Boiling Point (°C)	Dipole Moment (a Debye)	Fugacity (kPa)	Capacity (mmol/g)	Adsorption Heat (kJ/mol)	Adsorption Selectivity
n-heptane	100.2	98.5	0	0.00304601	1.7493	94.8070	442.58
P-xylene	106.17	138.4	0	0.00293503	1.6276	95.1376	446.77
Toluene	92.14	110.6	1.0	0.00337821	1.9531	95.1251	488.57
Trichloroethylene	88.11	87.0	2.7	0.00235842	2.9553	92.8275	438.12
Chlorobenzene	112.56	132.0	5.4	0.00276981	2.0542	89.9859	247.08
1,2-Dichloroethane	98.96	83.50	6.1	0.00315711	3.5298	89.0024	282.89
Ethyl acetate	88.11	77.50	6.1	0.00350493	2.5452	87.4874	195.03
Methyl isopropyl ketone	86.13	78.30	9.2	0.00360366	1.8398	79.1091	52.37
Cyclohexanone	98.14	155.00	10.3	0.00318117	1.6388	82.6454	80.39
Water	18.02	100.00	5.9	5.15575000	3.9653	47.5458	-

^a Dipole moment in coulombmeter (10⁻³⁰ C·m) measured in benzene, tetrachloromethane, and 1,4-dioxane or n-hexane at 20–30 °C. 1Debye = 3.336 × 10⁻³⁰ C·m [38].

). The relationship between VOC molecular properties (molecular mass, boiling point, and dipole moment) and adsorption selectivity was further analyzed and is presented in Figure 7. No significant correlation was observed between VOC molecular mass, boiling point, and adsorption selectivity. Previous studies have indicated that adsorbates with a larger molecular mass and higher boiling point exhibit a greater adsorption capacity due to increased interaction within pores of activated carbon [37]. However, this observation primarily applies to dry conditions and is not applicable to high-humidity VOC adsorption. A strong negative correlation was observed between dipole moment and adsorption selectivity (R² = 0.9276). Adsorption selectivity decreased with increasing dipole moment, which is because at a low VOC concentration, surface chemistry plays a more important role in adsorption. The relatively low carboxyl group content and thus nonpolar nature of the AC leads to unfavorable adsorption for strongly polar VOCs. Furthermore, as illustrated in Figure 8, strongly polar VOCs preferentially adsorb near polar oxygen-containing functional groups, leading to competition between adsorption sites and water molecules. The water clusters formed via hydrogen bonding under high-humidity environments, resulting in a larger negative effect. Therefore, the adsorption selectivity decreases with increasing VOC polarity.

4. Conclusions

This study prepared two adsorbents with different polarities but similar porous structures via thermal reduction. Toluene was used as the adsorbate to investigate the effect of adsorbent polarity. Dynamic adsorption capacity measurements revealed that the heat-treated, weaker-polarity activated carbon outperformed the pristine activated carbon in toluene adsorption under various humidity levels, indicating superior toluene/water vapor selectivity. The selective adsorption of VOCs/water vapor on activated carbon is governed by the interplay between adsorbent polarity and VOC molecular polarity, with weaker-polarity adsorbents favoring nonpolar VOCs through reduced competitive interactions with water. Molecular dynamics and a diffusion activation energy analysis indicated a significant decrease in the relative content of oxygen-containing functional groups on HCSC, resulting in reduced surface polarity and the enhanced adsorption of nonpolar toluene. The weaker-polarity adsorbent also exhibited a higher diffusion activation energy, suggesting stronger adsorbate–adsorbent interactions and a lower diffusion rate constant. The water vapor adsorption isotherm analysis confirmed that the reduced content of oxygen-containing functional groups led to decreased surface oxidation and a weaker affinity for water vapor. At 60 RH%, the dynamic adsorption capacity measurements of three VOCs with varying polarities on HCSC revealed the following selectivity order: toluene > n-heptane > 1,2-dichloroethane, implying the preferential adsorption of weaker-polarity toluene and n-heptane on the weaker-polarity HCSC. GCMC simulations of eight VOCs with varying polarities revealed a strong negative correlation between the adsorption selectivity and dipole moment. A mechanistic analysis indicated that, under high humidity levels, water molecules form clusters via hydrogen bonding, thereby significantly influencing the adsorption of stronger-polarity VOCs. Thermal reduction offers a scalable approach to improve VOC/water selectivity in industrial adsorbents. Future works should explore multi-component adsorption systems and long-term stability under cyclic operations.