SnO₂/a-C. The 34 mmol of D-glucose was dissolved in 170 mL of deionized water and stirred for 30 min. The mixed solution was placed in Teflon inner container and performed hydrothermal synthesis at 200°C for 12 h. After performing centrifuge from hydrothermally synthesized solution were dried in the vacuum oven overnight. The dried powder was conducted in a tube furnace filled with Ar atmosphere at 500°C to obtain the a-C. SnO₂ decoration was again performed via hydrothermal synthesis with SnCl₄⋅5H₂O and a-C. The solution of SnCl₄⋅5H₂O and a-C in deionized water was stirred for 30 min and hydrothermal synthesis was performed at 200°C for 12 h and sequentially dried the centrifuged powder in the vacuum oven overnight. The obtained powder was calcined under an Ar atmosphere at 500°C for 1 h.

SnO₂/C. SnO₂ decoration on carbon was performed via hydrothermal synthesis using SnCl₄⋅5H₂O with nano-sized carbon (Sigma-Aldrich, nano powder carbon <100 nm) under 200°C for 12 h. The solution of SnCl₄⋅5H₂O and carbon nanoparticles in deionized water was stirred for 30 min, and hydrothermal synthesis was performed under 200°C for 12 h. The resulting powder was calcined under an Ar atmosphere at 500°C for 1 h.

PtO_x/SnO₂/a-C and PtO_x/a-C. PtO_x decoration on the SnO₂/a-C or a-C particles was carried out using microwave synthesis for 30 s with H₂PtCl₆∙H₂O (Sigma-Aldrich, 99.9% trace metals basis) dissolved in deionized water. After drying the synthesized powder overnight in the oven, we obtained the PtO_x/SnO₂/a-C and PtO_x/a-C powder.

Morphological measurements were conducted by scanning electron microscopy (SEM, JEOL-7800F, JEOL Ltd.) and transmission electron microscopy (TEM, JEM-F200, JEOL Ltd.). The crystal structure characterization was performed using X-ray diffraction (XRD, Smart Lab, Rigaku) with Cu Kα radiation. Chemical bonding states were analyzed via X-ray photoelectron spectroscopy (XPS, K-alpha, Thermo Fisher Scientific Co.) Raman spectroscopy (LabRam Aramis, Horiba Jovin Yvon) was utilized to confirm the carbon vibration mode. The water contact angles for a-C and carbon were measured by Attension Theta Lite (Biolin Scientific).

A 2-probe electrode configuration was employed for the gas sensing analysis. SnO₂/a-C, SnO₂/a-C, PtO_x/SnO₂/a-C, and PtO_x/a-C properties were uniformly dispersed in ethanol at identical concentrations and subsequently drop-cast onto gold electrodes positioned on an alumina substrate. The gas-sensing performance of the fabricated sensors was evaluated within a custom-built chamber equipped with mass flow controllers, maintaining a fixed flow rate of 500 standard cubic centimeters per minute using air as the carrier gas. The sensors were exposed to target gas concentrations ranging up to 20 ppm for 500 s, followed by a recovery period in air for 1,000 s at a controlled temperature of 30°C. The resistance values in air (R_a) and upon exposure to the target gases (R_g) were recorded, and the sensor response (R_g/R_a) or R_a/R_g) was determined by calculating the ratio of resistance values in ambient air to those under gas exposure conditions Gas sensing measurements were performed for various gases, including Com HCHO, acetone, p-Xylene, Benzene, NO₂, and SO₂.

The emergence of gas adsorption sites on SnO₂ nanoparticles depends on the bonding characteristics of the carbon supports. As shown in the SEM results in Figure 1A, we used various nano-sized carbon and a-C particles for support. The intensity ratio between sp³ carbon bonding at the D site and sp² carbon bonding at the G site, noted as the I_D/I_G ratio in the Raman spectra (Figure 1B), is 0.77 for a-C, indicating a higher G band (sp² peak of carbon) bonding in a-C. This implies a higher C vacancy site than in the carbon particles (I_D/I_G = 1.08) and suggests a higher -OH group concentration than in the carbon particles. Consequently, the favorable sp² bonding character in a-C results in lower water contact angles compared to the carbon particle surface, as shown in Figure 1C.

SnO₂ nano-particles decorated on carbon supports (SnO₂/C) and a-C (SnO₂/a-C) are shown in the HR-TEM and EDS results (Figures 2A, B). Approximately 10 nm-sized SnO₂ particles are evenly distributed on all Carbon supports. The X-ray diffraction (XRD) results in Figure 2C show the SnO₂ single-phase peaks in all samples. However, the unobservable carbon peaks in the XRD results suggest lower crystallinity or an amorphous phase of the carbon supports. The XPS results for SnO₂/C and SnO₂/a-C (Figures 2D–F) demonstrate different chemical bonding characteristics. The carbon XPS results exhibit a larger asymmetric peak in SnO₂/a-C binding energy. The remaining XPS intensity at higher binding energy than 284.8 eV indicates the existence of C-O single bonding or C=O double bonding, suggesting stronger bonding between a-C and Oxygen in SnO₂ particles than in the Carbon supports (Kalish et al., 1999; Chokradjaroen et al., 2021). The Sn 3d and O 1s results display 496.4 eV for Sn 3d _3/2, 487.9 eV for Sn 3d _5/2, and 531.8 eV for O 1s in SnO₂/a-C, whereas SnO₂/C exhibits 496.3 eV for Sn 3d _3/2, 487.8 eV for Sn 3d _5/2, and 531.7 eV for O 1s. The higher binding energies of the peak positions indicate that the Sn on decorated SnO₂ nanoparticles maintains the preferred +4 charge state on both the C and a-C supports, consistent with the single-phase SnO₂ XRD pattern. Despite similar maximum peak positions for all samples, the higher Oxygen vacancy (V_O) region compared to the symmetric shape of the O 1s peak is observed in the SnO₂/C sample.

The higher VO density in SnO₂/a-C provides more activated gas adsorption sites compared to SnO₂/C samples. The chemiresistive gas sensing functionality, shown in Figure 3, measured the electrical resistance differences before and after exposure to 10 ppm analyte gases (CO, HCHO, acetone, p-Xylene, NO₂, and SO₂) for SnO₂/C and SnO₂/a-C at room temperature. Unlike SnO₂/C, which shows no significant electrical resistance difference under any of the analyte gases, SnO₂/a-C exhibits a response unit (R_a/R_g, where R_a is the resistance in air and R_g is the resistance under analyte gas) of less than five for all gases. However, SnO₂/a-C shows ambiguous gas selectivity, with a response of 4.5 for CO gas, which is higher compared to less than three for other gases.

The ambiguous gas selectivity of SnO₂/a-C particles is modulated by decorating them with PtO_x nanoparticles using hydrothermal synthesis methods. The PtO_x nanoparticles are evenly distributed on the SnO₂/a-C powder, as observed in the HR-TEM/EDS measurements in Figure 4A. The XRD results in Figure 4B show that the few wt% of PtO_x nanoparticles are decorated while maintaining the crystal structure of SnO₂/a-C powder. The Raman spectra of PtO_x/SnO₂/a-C in Figure 4C show a higher sp² vibration peak (G) compared to the sp³ vibration mode peak (D), indicating a higher carbon vacancy on the a-C support. The XPS results for PtO_x/SnO₂/a-C particles in Figures 4D–G show measurements for Pt 4f, C 1s, Sn 3d, and O 1s, respectively. Compared to SnO₂/a-C, the PtO_x/SnO₂/a-C samples exhibit a higher binding energy region on a-C, indicating C-O and C=O bonding sites between SnO₂ and a-C. The Sn and O measurements show peaks at 495.5 eV for Sn 3d _3/2, 487.1 eV for Sn 3d _5/2, and 531.1 eV for O 1s, respectively. The similar peak positions of the Sn 3d peaks indicate a consistent charge state of Sn in SnO₂/a-C samples, remaining in the 4+ valence state. The lower V_O region in the O 1s XPS results after PtO_x decoration on SnO₂/a-C suggests complex O bonding with Sn and Pt. The peaks at 73.2 eV for Pt 4f _7/2 and 76.5 eV for Pt 4f _5/2 indicate the fully ionized state of Pt cations as PtO_x on the SnO₂ and a-C support. Furthermore, the oxidation state of Pt in PtO_x/SnO₂/a-C is similar to that in PtO_x/a-C, independent of the matrix materials. However, the 533.3 eV O 1s peak, attributed to covalent bonding on the carbon tape, indicates a lower concentration of PtO_x nanoparticles, suggesting that they are physically decorated on the surface without strong chemical bonding with SnO₂ or carbon. The gas selectivity is primarily generated by the PtO_x particles when they are decorated on SnO₂ nanoparticles.

To further investigate the role of PtO_x, we conducted gas-sensing measurements under the same conditions (room temperature, 10 ppm concentration of CO, HCHO, acetone, p-Xylene, NO₂, and SO₂). We also synthesized PtO_x/a-C particles and evaluated their gas-sensing performance. The results, shown in Figure 5A, reveal a gas-sensing response of nearly seven under CO gas. Additionally, the CO gas sensing response of PtO_x/SnO₂/a-C remained consistent 1 month after synthesis, demonstrating higher selectivity for CO compared to other gases (HCHO, acetone, p-Xylene, NO₂, and SO₂). In contrast, PtO_x/a-C powder showed no gas-sensing functionality for all of the tested gases. Comparing the gas selectivity for SnO₂/C, SnO₂/a-C, and PtO_x/SnO₂/a-C, as plotted in Figure 5B, demonstrates that the decorated PtO_x particles impart CO gas selectivity to SnO₂/a-C, which has an activated gas adsorption site. The response and recovery times (t_Res. and t_Rec.) of SnO₂/a-C and PtOx/SnO₂/a-C under 10 ppm of CO gas are shown in the inset. The PtO_x/SnO₂/a-C sample displays a response time of 231 s and a recovery time of 101 s, while the SnO₂/a-C sample exhibits a response time of 125 s and a recovery time of 149 s, respectively. A detailed analysis reveals that PtO_x/SnO₂/a-C demonstrates a faster t_Res. than SnO₂/a-C up to 70% of the maximum response, with the t_Res. of PtO_x/SnO₂/a-C surpassing that of SnO₂/a-C as it approaches the maximum gas response.

The reversible gas-sensing functionality of PtO_x/SnO₂/a-C during long-term gas-sensing measurements, shown in Figure 5C, demonstrates that PtO_x/SnO₂/a-C is an n-type gas-sensing device and exhibits a stable resistance range of 10¹¹∼10¹² Ω. The higher CO gas selectivity of PtO_x/SnO₂/a-C, compared to the lack of gas response on PtO_x/a-C, demonstrates that PtO_x on SnO₂ is the primary gas adsorption site, and PtO_x-.SnO₂-a-C provides a sequential external carrier conduction route, as described in Figure 5D.