The Influence of Carboxyl-Functionalized Carbon Dots on Ethanol Selectivity in Gas Sensing

Abstract

For semiconductor tin dioxide (SnO₂) materials, the oxygen adsorption theory often struggles to explain their selectivity towards specific gases. Therefore, it is worth considering altering the surface functional groups of SnO₂ to modify its surface state and enhance its selectivity towards specific gases. Due to the rich functional groups on the surfaces of carbon dots, this study employed a hydrothermal method to prepare three types of carbon dots with varying carboxyl functional group contents by adjusting the hydrothermal time. These carbon dots were then used as dopants and combined with SnO₂ to create composite gas-sensitive devices. The gas-sensing test results indicate that the introduction of carboxyl functional groups can enhance the selectivity of SnO₂ towards ethanol. Furthermore, at any operating temperature within the range of 150–300 °C, the higher the carboxyl functional group content on the surface of carbon dot-doped SnO₂, the higher the sensitivity towards ethanol. By employing density functional theory (DFT), the interaction energies between the surfaces of carbon dots and surface carboxyl groups with the target gas were calculated. These calculations validated the gas-sensing test results, confirming that the presence of carboxyl functional groups enhances the selectivity towards ethanol. The results of this study can provide new insights into the research on the selective mechanism of gas-sensitive materials.

1. Introduction

Gas sensors based on metal oxide semiconductors (MOSs) possess advantages such as high sensitivity, fast response time, good anti-interference performance, and low manufacturing cost. As a result, they are promising gas sensors with broad application prospects [1,2,3]. In recent years, numerous gas sensors based on MOSs have been reported, including ZnO, SnO₂, TiO₂, WO₃, and In₂O₃. Among them, SnO₂ is a typical n-type semiconductor material with a bandgap width of 3.6 eV at 300 K. It exhibits excellent chemical stability and outstanding electrical performance. In particular, when SnO₂ is used as the sensing material, gas sensors can exhibit advantages such as low resistance, high sensitivity, and fast response-recovery speeds. They are commonly employed for monitoring certain combustible, toxic, and pollutant gases [4,5].

Currently, there are various strategies to improve the performance of gas sensors based on SnO₂ as the substrate. The oxygen adsorption theory model explains the gas detection principle of SnO₂, which involves the reaction between the target gas being adsorbed and the surface-adsorbed oxygen ions [6,7]. It is evident that a larger specific surface area of SnO₂ allows for a greater interaction area with both oxygen and target gases, thereby enhancing its gas-sensing performance. Therefore, constructing nanostructured SnO₂ can significantly increase the material’s specific surface area, providing more active surface sites for reactions, as well as facilitating smooth gas diffusion channels. This can greatly improve the sensor’s sensitivity characteristics [4,8]. Li et al. achieved remarkable results with SnO₂ nanosheets synthesized via a simple hydrothermal method. The nanosheets exhibited a response value as high as 494 towards 20 ppm NO at 200 °C. Additionally, they demonstrated excellent selectivity and stability towards NO gas [9]. Zhang et al. fabricated a porous layered SnO₂ nanostructure using ultrasound-assisted shaking and calcination methods. The resulting material exhibited a response value of 50 towards 10 ppm formaldehyde gas at 200 °C. Furthermore, it demonstrated excellent selectivity and stability [10]. The excellent gas-sensing performances of both sensors mentioned above can be primarily attributed to the SnO₂ gas-sensing materials prepared with a small grain size, large specific surface area, and abundant pore structures. These characteristics contribute to enhanced gas adsorption and interaction, resulting in improved sensitivity and selectivity.

In addition to adjusting the microstructure and morphology of SnO₂, ion doping can induce changes in its lattice parameters, as well as alter the number of dangling bonds on the surface and the abundance of surface defects. This, in turn, promotes the adsorption of gas molecules and enhances the gas-sensing performance of the material. N. Tammanoon et al. achieved a high response of 3626 towards 400 ppm acetone by doping La₂O₃ into nanostructured SnO₂. The doped material also exhibited excellent selectivity towards acetone [11]. Thanks to the unique chemical and electronic sensitizing properties of noble metals, surface modification with noble metals can also enhance the gas-sensing performance. Lin et al. synthesized Pd nanoparticle-loaded SnO₂ nanofibers through electrospinning and wet modification techniques. The resulting nanofibers exhibited a significant improvement in their response to formaldehyde gas [12].

Apart from surface modification with noble metals, other approaches to enhance the performance of SnO₂ gas-sensing materials revolve around altering the surface state of SnO₂. Current research widely accepts the oxygen adsorption theory as the sensing mechanism for semiconductor gas-sensitive materials. According to this theory, at a certain temperature, when the target gas reaches the surface of the sensitive material, it undergoes a chemical reaction with the oxygen adsorbed on the surface, resulting in a change in the material’s resistance. While the oxygen adsorption theory can explain how increasing the contact area, enhancing surface oxygen vacancies, and increasing active adsorption sites can improve the sensitivity characteristics of SnO₂, it has limitations in explaining the selective mechanism for specific gases. Additionally, it cannot facilitate the development of gas sensors that exclusively enhance the selectivity for specific gases. In addition to altering surface oxygen vacancies and adsorption sites, changing the surface functional groups of SnO₂ is also an effective method to modify its surface state.

Carbon dots (CDs) refer to nanomaterials with a carbon-based framework and surface-active functional groups, such as carboxyl (−COOH), amino (−NH₂), and hydroxyl (−OH) groups, among others, with a size smaller than 10 nm [13]. The surfaces of carbon dots possess abundant functional groups, and different surface functional groups can be obtained by changing the carbon source, synthesis method, and preparation conditions. Carbon dots can be easily doped into other semiconductor gas-sensing materials to modify their surface functional groups, thereby altering the selectivity of the gas-sensing materials towards different target gases. Cheng et al. fabricated hierarchical lychee-shaped In₂O₃ nanospheres decorated with carbon dots, achieving high sensitivity and selectivity for NO₂ detection [14]. Yu et al. prepared flower-shaped ZnO structures through a hydrothermal method and subsequently introduced carbon dots. The resulting composite sensing material exhibited excellent response and selectivity to NO [15]. Currently, there have been advancements in designing and synthesizing composite materials of metal oxides and carbon dots to enhance the gas-sensing performance. However, the impact of surface functional groups introduced by carbon dots on the gas-sensing performance, especially selectivity, has not been explored.

In this work, a simple one-step hydrothermal method was employed to prepare three types of carbon dots with different surface carboxyl functional group contents. These carbon dots were then doped into SnO₂ to fabricate composite gas-sensing materials. Considering that carboxyl functional groups act as Lewis acid sites and can react with ethanol, which acts as a Lewis base, the influence of carbon dot-doped SnO₂ gas-sensing materials on the selectivity towards ethanol was investigated. The results indicate that the introduction of carboxyl functional groups through carbon dot doping on the surface of SnO₂ enhances its sensitivity and selectivity towards ethanol. Furthermore, the sensitivity of the gas-sensing material towards ethanol increases with an increase in the carboxyl functional group content. The gas-sensing test results were verified through density functional theory (DFT) calculations, which provided further insights into the mechanism of enhanced selectivity of SnO₂ towards ethanol from the perspective of surface functional groups.

2. Materials and Methods

2.1. Chemical Reagents

Citric acid monohydrate (C₆H₈O₇·H₂O, ≥99.5%) was purchased from China National Pharmaceutical Group Chemical Reagent Co., Ltd, Shanghai, China. Nanoscale tin dioxide (SnO₂, AR, 99.9%) was obtained from Aladdin Chemical Reagent Co., Ltd, Shanghai, China. Deionized water and anhydrous ethanol were used during the experimental process.

2.2. Preparation of Carbon Dots with Different Carboxyl Functional Group Contents on the Surfaces

Preparation of carbon dots using a hydrothermal method was conducted as follows: an amount of 2.1 g of citric acid monohydrate was dissolved in 20 mL of deionized water with vigorous stirring for 30 min until fully dissolved. The resulting solution was then transferred to a 50 mL stainless-steel reaction vessel lined with polytetrafluoroethylene. The reaction vessel was maintained at 180 °C for 60 min. Upon completion of the reaction, the solution was naturally cooled to room temperature. Subsequently, centrifugation was carried out at 5000 rpm for 20 min to remove larger impurities. Finally, the supernatant was transferred into a dialysis bag and dialyzed for 24 h to obtain a carboxyl-functionalized carbon dot solution (CDs-60). Under identical conditions, CDs-90 was obtained by maintaining the reaction at 180 °C for 90 min, and CDs-120 was obtained by maintaining the reaction for 120 min.

2.3. Characterization

The morphology and microstructure of the prepared carbon dots (CDs) and SnO₂/CDs were analyzed using transmission electron microscopy (TEM) (JEOL JEM-3010, JEOL, Ltd., Tokyo, Japan) and high-resolution transmission electron microscopy (HRTEM) (JEOL JEM-3010, JEOL, Ltd., Tokyo, Japan). The chemical bonds and functional groups of the carbon dots were analyzed using a Fourier transform infrared spectrometer (Nicolet iS10, Thermo Electron Scientific Instruments, Madison, WI, USA). The chemical composition and electronic state information on the surfaces of the carbon dots were analyzed using X-ray photoelectron spectroscopy (XPS) (EscaLab 250, Thermo Electron Scientific Instruments, Madison, WI, USA).

2.4. Sensor Fabrication and Gas-Sensing Testing

After mechanically grinding 50 mg of nano-size SnO₂ powder for 30 min, 300 μL of three different carbon dot solutions (CDs) were added dropwise. The grinding process was continued until a homogeneous slurry was obtained. The SnO₂/CDs mixture was uniformly coated onto a ceramic tube with gold electrodes and cured at 150 °C for 1 h. Then, a Ni−Cr resistance wire was inserted into the ceramic tube as a heater to control the temperature. The ceramic tube and Ni−Cr resistance wire were soldered onto a hexagonal base. After aging, the gas-sensing performance of the material was studied using a static testing system consisting of a gas-sensing element, gas cylinder, Fluke 8846A digital multimeter, GPD3303S DC power supply, and a computer workstation. Sensitivity is defined as R_a/R_g (reducing gas), where R_a and R_g represent the resistance of the device in air and the target gas, respectively. Response time and recovery time are defined as the time required for the device to reach 90% of the final stable resistance change value, both during gas exposure and after removing the gas, respectively [16].

3. Results and Discussion

3.1. Characterization of Carbon Dots

The SEM image of SnO₂/CDs-60 is shown in Figure 1a. It can be observed that the sample is mainly composed of small-sized SnO₂ nanoparticles. These tiny nanoparticles have a diameter of approximately 10–20 nm and exhibit uniform size distribution. Due to their small size, the nanoparticles possess a large surface area, which is advantageous for enhancing the performance of the sample in gas adsorption and other aspects. Because the prepared CDs have an extremely small particle size, their morphology, as well as the presence of CDs in the SnO₂/CDs-60 composite material, cannot be observed using SEM. Therefore, the characterization of CDs-60 was conducted using transmission electron microscopy (TEM).

The TEM image of CDs-60 is shown in Figure 1b. It can be observed that CDs-60 exhibits good dispersion and uniform size distribution, with a particle size of approximately 5 nm for the carbon nanoparticles. The HRTEM image of CDs-60, as shown in Figure 1c, reveals a lattice spacing of 0.21 nm, which falls within the range of lattice spacing for carbon dots (0.20–0.35 nm) [17]. This lattice spacing corresponds to the distance between the (100) crystal planes of sp² carbon materials [18]. In addition to that, the internal structure of the SnO₂/CDs-60 composite material was also observed. In the HRTEM image of the composite material shown in Figure 1d, lattice fringes with spacings of 0.176 nm and 0.264 nm can be attributed to the (211) and (101) crystal planes of SnO₂, respectively. Furthermore, striped patterns with a lattice spacing of 0.21 nm, corresponding to the (100) crystal planes of carbon materials, can also be observed. However, there is no mutual fusion or heterojunction structure formed between the carbon dots (CDs) and SnO₂, which indicates that the CDs were only doped into the SnO₂ through mechanical mixing and a successful blending of CDs and SnO₂ was achieved.

Powder X-ray diffraction (XRD) was used for the crystalline phase analysis of the prepared carbon dots (CDs). The XRD pattern of CDs-60 is shown in Figure S1, exhibiting a broad diffraction peak at 2θ = 20°~30°, which is indicative of the characteristic diffraction pattern of amorphous carbon materials [19,20]. Furthermore, the broadening of the diffraction peaks in the sample indicates the low content and small size of the CDs, suggesting their poor crystallinity and good dispersion.

The surface functional group information of the three prepared carbon dots was analyzed using Fourier transform infrared (FT-IR) spectroscopy. As shown in Figure 2, the absorption peaks of the three carbon dots around 3420 cm⁻¹ correspond to the stretching vibration of O−H bonds [21], the broad absorption peak in the range of 3000–2800 cm⁻¹ corresponds to the stretching vibration of saturated C−H bonds [22], and the absorption peak observed at 1703 cm⁻¹ corresponds to the stretching vibration of the C=O bond in the −COOH group on the surfaces of the carbon dots [23]. The absorption peak at 1643 cm⁻¹ corresponds to the stretching vibration of C=C bonds. Additionally, the absorption peak observed around 1400 cm⁻¹ corresponds to the bending vibration of C−OH bonds in the −COOH group [24]. It is evident that the stretching vibration peak of the C=O bond in the −COOH group and the bending vibration peak of C−OH in CDs-60 are significantly stronger than those in CDs-90 and CDs-120. Preliminary conclusions can be drawn from the analysis: The prepared carbon dots exhibit carboxyl functional groups on their surfaces, and the content of carboxyl functional groups decreases with an increasing hydrothermal time. Among the three prepared carbon dots, CDs-60 has the highest content of carboxyl functional groups on its surface. The FT-IR analysis of citric acid monohydrate was also conducted, as shown in Figure S2. The absorption peak at 3369 cm⁻¹ corresponds to the stretching vibration of O−H bonds. The absorption peak around 3000 cm⁻¹ corresponds to the stretching vibration of saturated C−H bonds. The absorption peaks at 1756 cm⁻¹ and 1728 cm⁻¹ correspond to the stretching vibration of C=O bonds in citric acid. The absorption peak at 1213 cm⁻¹ corresponds to the stretching vibration of carboxylic acid C−OH bonds. In comparison to citric acid monohydrate, the three types of prepared carbon dots exhibit a new absorption peak at around 1640 cm⁻¹, which indicates the presence of C=C bonds. This suggests that the carboxyl-functionalized carbon dots are obtained through the carbonization of citric acid monohydrate.

To further determine the content of carboxyl functional groups on the surfaces of the three carbon dots, X-ray photoelectron spectroscopy (XPS) analysis was performed on the samples. Figure 3a–c show the XPS survey spectra of CDs-60, CDs-90, and CDs-120. It can be observed that the three carbon dots are composed of carbon (C) and oxygen (O) elements, with no other impurities present. Figure 3d–f present the high-resolution analysis of the C 1s peak. The C 1s peak was fitted with five peaks centered at approximately 284.4, 285, 286.3, 287.9, and 289.2 eV, corresponding to (C=C), (C−C), (C−O), (C=O), and (O−C=O) bonds, respectively [25,26]. By calculating the peak areas of each component in the C 1s spectrum, the relative percentages of carboxyl functional groups (O−C=O) on the surfaces of the carbon dots were determined to be 27.04%, 19.05%, and 13.44% for CDs-60, CDs-90, and CDs-120, respectively. The relative percentages of the other components are listed in

Table 1. The percentages of different components of the three carbon dots in the C1s region.

Samples	C=C	C−C	C−O	C=O	O−C=O
CDs−60	24.84%	27.01%	15.26%	5.85%	27.04%
CDs−90	17.97%	36.73%	19.49%	6.76%	19.05%
CDs−120	19.27%	36.05%	16.55%	14.69%	13.44%

. XPS analysis further confirmed the results obtained from FT-IR spectroscopy that showed that all three types of prepared carbon dots have carboxyl functional groups on their surfaces. Moreover, the XPS results show that the content of carboxyl functional groups gradually decreases with an increasing hydrothermal time. Among the three samples, CDs-60, prepared with a hydrothermal time of 60 min, exhibits the highest percentage of carboxyl functional groups, which is 27.04%.

3.2. Study on the Influence of Surface Carboxyl Functional Groups on Selectivity

To investigate whether the introduction of carboxyl groups enhances the selectivity of SnO₂ towards ethanol, sensitivity tests were conducted at the optimal operating temperature of 225 °C. The tests compared the sensitivity of pristine SnO₂ and SnO₂/CDs-60 composite sensing materials to different common VOCs, including acetone, formic acid, toluene, and ethanol, at a concentration of 100 ppm, as shown in Figure 4. It can be observed that the sensitivity of SnO₂ towards ethanol significantly increased after doping with CDs-60, while the sensitivity towards acetone, formic acid, and toluene remained almost unchanged. This result is consistent with the expected outcome, indicating that the introduction of carboxyl groups onto the surface of SnO₂ can enhance the selectivity towards ethanol without affecting the sensitivity to other VOCs. Additionally, the sensitivity of SnO₂/CDs-60 to 100 ppm of acetone, formic acid, and toluene was tested within the temperature range of 150–300 °C, as shown in Figure S3. It can be observed that within this temperature range, the sensitivity of SnO₂/CDs-60 to these three VOC gases is significantly lower compared to its sensitivity to 100 ppm of ethanol, further demonstrating the excellent selectivity of SnO₂/CDs-60 towards ethanol.

The interference resistance of the SnO₂/CDs-60 sensing material was tested at 225 °C, as shown in Figure S4, except for selectivity. It can be observed that the sensitivity of SnO₂/CDs-60 remains unchanged when 100 ppm of acetone, formic acid, toluene, or a combination of the three VOCs is added to 100 ppm of ethanol, indicating its excellent interference resistance.

To determine the influence of the carboxyl group content on the sensitivity of SnO₂ towards ethanol detection, the sensitivity to 100 ppm ethanol of the three sensors was tested at different operating temperatures, as shown in Figure 5. Figure 5a illustrates that, at arbitrary operating temperatures ranging from 150 to 300 °C, SnO₂/CDs-60 exhibits higher sensitivity towards ethanol compared to SnO₂/CDs-90 and SnO₂/CDs-120. SnO₂/CDs-90 also demonstrates higher sensitivity towards ethanol than SnO₂/CDs-120. At the optimal operating temperature of 225 °C, the sensitivities of the three sensors towards 100 ppm ethanol are 52.1, 38.62, and 26.3, respectively. This indicates that, as the surface carboxyl group content on SnO₂ increases, its sensitivity towards ethanol also increases. The sensor prepared by incorporating CDs-60 with the highest surface carboxyl group content into SnO₂ exhibits the best gas-sensing response to ethanol. Figure 5b shows the linear relationship between the sensitivities of the three sensors to different concentrations of ethanol at 225 °C. It can be observed that SnO₂/CDs-60 has the highest slope, followed by those of SnO₂/CDs-90 and SnO₂/CDs-120. This indicates that the higher the surface carboxyl group content on SnO₂, the greater the sensitivity change in detecting ethanol, thus indicating a higher sensitivity to ethanol detection.

In order to further demonstrate the feasibility of incorporating carbon dots (CDs) into SnO₂ to introduce carboxyl functional groups onto the surface of SnO₂ and enhance ethanol selectivity, the response and recovery of SnO₂/CDs-60 to different ethanol concentrations were tested, as shown in Figure 6. Figure 6a presents the dynamic response-recovery curves of the SnO₂/CDs-60 sensor to different ethanol concentrations at 225 °C. It can be observed that the sensor exhibits excellent response-recovery characteristics at various ethanol concentrations. Additionally, the response increases with the increasing ethanol concentration. Figure 6b evaluates the response-recovery time of the SnO₂/CDs-60 sensor to 100 ppm ethanol based on the dynamic response-recovery curves. The calculated response time for 100 ppm ethanol is 7 s, while the recovery time is 82 s, indicating that the sensor exhibits a relatively fast response-recovery speed.

Based on the above test results, we can conclude that the introduction of carbon dots into SnO₂ to create surface carboxyl functional groups enhances the sensitivity and selectivity of SnO₂ towards ethanol gas. The higher the surface carboxyl functional group content on SnO₂, the greater the sensitivity to ethanol.

3.3. Gas-Sensing Mechanism and Density Functional Theory (DFT) Calculations

Currently, it is widely believed that the sensing mechanism of MOS gas sensors is mainly based on gas adsorption and chemical reactions that occur on the surface of the sensing material [27,28]. For SnO₂-based gas sensors, when exposed to air, oxygen molecules, due to their higher electron affinity than the work function of SnO₂, can extract electrons from the conduction band of SnO₂, resulting in the formation of adsorbed oxygen ions (O⁻ (150–400 °C)) [29]. This leads to the formation of an electron-depleted layer, causing the SnO₂-based gas sensor to exhibit a high-resistance state. The reaction process can be represented by Equation (1):

$$O_2\left(ads\right)+2e^{-}\leftrightarrow 2O^{-}\left(ads\right)$$

(1)

For ethanol sensors, the currently widely accepted mechanism is as follows [30,31,32,33]: When the SnO₂-based gas sensor is exposed to ethanol gas, the ethanol gas undergoes oxidation-reduction reactions with the adsorbed oxygen ions (O⁻) on the surface of the SnO₂ material, and the released electrons from the reaction re-enter the conduction band of the SnO₂ material, resulting in a narrowing of the electron depletion-layer width. At this point, the resistance of the SnO₂-based gas sensor decreases significantly. This process can be represented by Equation (2):

$$C_2{H}_{5}OH\left(g\right)+6O^{-}\left(ads\right)\to 2C{O}_2\left(g\right)+3H_{2}O\left(g\right)+6e^{-}$$

(2)

Although this reaction mechanism has gained general acceptance, it has not been experimentally verified due to the inability to measure the precise changes in the gas-phase composition in the atmosphere before and after the reaction.

The schematic diagram is shown in Figure 7. According to this mechanism, improving the gas-sensing performance of SnO₂ can be achieved by increasing its contact area, increasing the surface oxygen vacancies, and enhancing the active adsorption sites. However, these modifications do not necessarily enhance the selectivity towards specific gases, and the selective mechanism remains unclear.

This paper explains the selectivity mechanism from the perspective of surface functional groups on SnO₂. By doping carbon dots into an MOS and introducing highly active surface functional groups, the surface state of the MOS can be altered, thereby enhancing its gas-sensing performance [34,35]. The introduction of carbon dots with surface carboxyl functional groups can enhance the selectivity towards alcohol gases [36]. The carboxyl functional groups on the surface of SnO₂ will undergo a protonation reaction, losing a hydrogen ion from the hydroxyl group and releasing a proton. This transforms the oxygen atom on the carboxyl group into an acidic site and increases the surface acidity of SnO₂. Studies have shown that as the surface acidity of an MOS increases, the adsorption of ethanol molecules by the MOS becomes stronger [37]. This may be attributed to the reaction between the carboxyl functional groups, acting as Lewis acid sites, and ethanol, acting as a Lewis base. Therefore, the introduction of carboxyl groups can enhance the adsorption of ethanol by SnO₂. Carboxyl functional groups can form hydrogen bonds (O−H...O) with ethanol molecules, which further enhances the adsorption of ethanol molecules on the carboxyl functional groups on the SnO₂ surface [38].

To verify the experimental results, density functional theory (DFT) calculations were performed. DFT calculations were performed using the Dmol3 module in Material Studio 2020. Graphene fragments with hydroxyl and carboxyl functional groups at the edges were used to represent the crystal facets of the carbon dots. VOC molecules, including ethanol, formic acid, acetone, and toluene, were placed on this model. The optimized geometries of the structures are shown in Figure 8. The simulation method is as follows: In this work, the density functional theory (DFT) calculation was performed using the Dmol3 code [39]. The exchange-correlation interaction was treated by the generalized gradient approximation (GGA) with a PBE functional [40]. A double-numerical quality basis set with a d-type polarization function (DNP [3]) was utilized for all the geometric optimizations and total energy calculations. The core electrons were modeled using the effective core pseudopotentials (ECPs) of Dolg [41] and Bergner [42]. The calculations employed a spin-unrestricted methodology. The atomic positions underwent complete relaxation until the specified convergence criteria were satisfied: forces below 0.002 Ha/Å, total energy below 10⁻⁵ Ha, and displacements below 0.005 Å. A real-space cutoff radius of 4.1 Å was employed. The self-consistent field computations were terminated upon reaching a criterion of 10⁻⁶ Ha. The adsorption energy (E_ad) is defined as follows:

$$E_{ad}=E_{Surf+adsorbates}-E_{adsorbates}-E_{Surf}$$

(3)

where E_{Surf+adsorbates} represents the total energy of ethanol, formic acid, acetone, and toluene adsorbed on the surface crystal plane of the CDs, E_adsorbates represents the energy of individual gas molecules, and E_Surf represents the energy of the CDs. After structure optimization, the calculated adsorption energies (E_ad) for each model were found to be negative, indicating an exothermic process of molecule adsorption and the formation of a more stable system [43].

The calculated adsorption energies of the four VOCs on the surfaces of carbon dots (CDs) are listed in

Table 2. Adsorption energies (E_ad) of VOCs adsorbed on CDs.

Complex	Ead (eV)
Ethanol–CDs	−0.1088
Acetone–CDs	−0.1633
Formic acid–CDs	−0.1361
Toluene–CDs	−0.2177

.

The geometrically optimized structures of the VOC molecules adsorbed on the carboxyl groups introduced in carbon dots (CDs) are shown in Figure 9.

The calculated adsorption energies of four VOCs on the carboxyl functional groups on the carbon dot surfaces are listed in

Table 3. Adsorption energies (Ead) of VOCs adsorbed on carboxyl-functionalized CDs.

Complex	Ead (eV)
Ethanol–COOH	−0.5878
Acetone–COOH	−0.4844
Formic acid–COOH	−0.4109
Toluene–COOH	−0.2041

.

Based on the calculation results, it can be concluded that, after introducing carboxyl functional groups onto the surfaces of carbon dots, the negative adsorption energy of ethanol molecules on the carboxyl functional groups is greater compared to the negative adsorption energy on the surfaces of the carbon dots. Furthermore, it exhibits the highest negative adsorption energy (a more negative E_ad value indicates stronger interactions between the adsorbed molecule and the surface) when compared to acetone, formic acid, and toluene. The density functional theory calculations successfully validated the gas-sensing test results, indicating that the carboxyl functional groups on the surfaces of carbon dots enhance the adsorption of ethanol. By doping carbon dots into SnO₂ and introducing carboxyl functional groups onto the surface of SnO₂, a composite gas sensor can be prepared, which enhances the selectivity towards ethanol.

4. Conclusions

In this study, carbon dots (CDs) with different surface carboxyl functional group contents were synthesized using a hydrothermal method, and their characterization was performed using the TEM, FI-IR, and XPS techniques. The results showed that all three types of carbon dots (namely, CD-60, CDs-90, and CDs-120) possessed carboxyl functional groups on their surfaces. The percentages of carboxyl functional groups were found to be 27.04%, 19.05%, and 13.44% for CDs-60, CDs-90, and CDs-120, respectively. Among them, CDs-60 exhibited the highest content of carboxyl functional groups, good dispersibility, and a uniform size. Three types of carbon dots were separately doped into SnO₂ to fabricate gas-sensing devices. The gas-sensing test results demonstrated that the introduction of carboxyl groups onto the surface of SnO₂ enhanced its selectivity towards ethanol. Moreover, the sensitivity of the sensor towards ethanol increased with a higher carboxyl group content on the surface of SnO₂. These results were further confirmed by density functional theory calculations, which indicated that the negative adsorption energy of ethanol on the carboxyl functional groups on the carbon dots’ surfaces was greater than that on the carbon dots’ surfaces alone. Additionally, it was found to be higher than that of acetone, formic acid, and toluene.

In summary, this method of utilizing different active functional groups on the surfaces of carbon dots to modify the surface state of SnO₂ gas-sensing materials has been shown to effectively enhance the selectivity of the sensitive material towards specific gases. It provides a new approach for the targeted design of semiconductor gas sensors.