1. APS Materials

The experimental conditions, synthesis methods, and structural characterization of APS materials are based on previous reports ^[6]. This paper focuses on the efficient electrical evaluation of APS films, referencing earlier studies for experimental procedures and structural data.

2. Synthesis of Pd-decorated ZnO Nanoparticles

ZnO NPs, 50 nm in size, were purchased from Sigma-Aldrich. A total of 0.25 grams of these nanoparticles (NPs) were dispersed in 20 ml of ethanol and stirred for 3 hours. To enhance sensing performance, 3 nm thick Pd films (mass thickness) were deposited onto the nanoparticles using a shadow mask and electron beam evaporation at a base pressure of $5.0\times 10^{-8}$ Torr ^[6].

Synthesis of Anti-poisoning rGO Layer with CeO${}_{2}$ Nanoparticles Composite

The anti-poisoning rGO layer with CeO${}_{2}$ NPs composite layered material was prepared using an impregnation method. Ce(NO${}_{3}$)${}_{3}\cdot$6H${}_{2}$O, sourced from Sigma-Aldrich, served as the precursor for CeO${}_{2}$, while rGO (Standard Graphene, V20-100) was used without modification. 1 wt% solution of Ce(NO${}_{3}$)${}_{3}\cdot$6H${}_{2}$O and 0.5 g of rGO were dispersed in 50 mL and 70 mL of distilled water, respectively, and stirred vigorously for 1 hour. After mixing the solutions, the pH was adjusted to 10 by adding NH${}_{4}$OH ^[5,6]. The mixture underwent ultrasonic dispersion for 1 hour, and then was stirred at 60$^\circ$C for 6 hours. It was then filtered, washed several times to remove unreacted materials and impurities, and dried using a vacuum rotary evaporator. The resulting 1 wt% CeO${}_{2}$-rGO composite was dispersed in a 1:1 ratio of NMP and IPA solvents, spin-coated onto the Pd-decorated ZnO NPs ^[6].

4. Material Characterization

Metal oxide film materials were characterized by field-emission scanning electron microscopy (FE-SEM, FEI, and Nova Nano SEM). To measure the electrical characteristics of the various metal oxide materials, a bias sweep from $-1$ to $1$ V was applied across the electrodes at 250$^\circ$C and the resistance of each NPs was measured using a Keithley 2636A source measurement unit ^[6].

5. Fabrication of Anti-poisoning Material

For the design of APS sensors, electrodes were fabricated by depositing Ti/Au (20/200 nm) onto a SiO${}_{2}$ platform. Afterwards, ZnO NPs were deposited. Subsequently, a 3 nm thick Pd film was applied onto the nanoparticles within the active region using a shadow mask. The CeO${}_{2}$/rGO composite mixture was then spin-coated onto the decorated nanoparticles. Finally, the fully prepared samples underwent rapid thermal annealing at 400$^\circ$C under a nitrogen atmosphere for 1 minute to enhance the Ohmic contact between the nanoparticles and metal electrodes ^[6].

Fig. 1. Structure of the anti-poisoning sensor for HMDSO removal and H${}_{2}$ detection.

Fig. 1 illustrates the layered structure of the APS element, which underwent drying under standard atmospheric conditions and aging at the operational temperature. This evaluation particularly focuses on comparing notable resistance changes between the APS sensor and the non-APS sensor.

The APS sensors were placed in a reaction chamber and exposed to pulses of a single gas, such as H${}_{2}$ at 200 ppm, mixed with air in 2000 sccm flow for 40 minutes. They were measured in a probe chamber (approx. 125 cm${}^{3}$) with an electrical system, vacuum, temperature control, and gas flow setup. Pure air was introduced between pulses for a few minutes. HMDSO at a concentration of 10 ppm, mixed with air, was introduced into the chamber for several minutes to evaluate any potential poisoning or deterioration of the sensing material by the silicone vapors. Resistance was measured using a Keithley 2636A source measurement unit with a 1 V bias applied to the electrodes. The sensor operating temperature was controlled and varied from room temperature up to 400$^\circ$C, with the optimal sensing performance observed at 250$^\circ$C. Sensor resistance changes are calculated as (R${}_{a}$ $-$ R${}_{g}$) / R${}_{a} \times 100$, where R${}_{a}$ and R${}_{g}$ represent the steady-state resistance values with and without the analyte gas, respectively. This comparison allows for quantifying resistance changes in the presence of the analyte, crucial for evaluating sensor effectiveness in gas detection. All the gases were controlled using mass flow controllers, as shown in Fig. 2.

Fig. 2. Schematic representation of gas flow line to the sensing chamber.

1. APS Film Materials Development

The experimental section of the previous report detailed the fabrication process of anti-poisoning materials. The cross-sectional scanning electron microscopy (SEM) image of the Pd-decorated ZnO nanoparticles and the CeO${}_{2}$-rGO composite layers are shown in Fig. 3 ^[5,6]. The size of the Pd (3 nm) decorated ZnO nanoparticles layer ranges over several tens of nanometers due to the role of the fixing site on the 600-800 nm CeO${}_{2}$-reduced graphene surface. Annealing is performed to prevent particle aggregation and to achieve smooth film formation.

Fig. 3. SEM images of CeO${}_{2}$/rGO-coated Pd/ZnO nanoparticles with scale bars of 1 $\mu$m.

2. Anti-poisoning ability of APS layer

In this work, we applied previous recommendations to significantly optimize the performance of APS sensors for resistance to poisoing of HMDSO ^[6]. APS layers, featuring CeO${}_{2}$-rGO coatings on Pd-decorated ZnO nanoparticles, aid in eliminating HMDSO poisoning and detecting H${}_{2}$. The resistance variations of ZnO, Pd-ZnO, and the APS layer in response to HMDSO exposure were noted, with all sensors reacting to HMDSO. Measurements were conducted on the response of ZnO, Pd-decorated ZnO, and CeO${}_{2}$-rGO/Pd/ZnO under both air/HMDSO (10 ppm) conditions at 250$^\circ$C. Notably, surface modification with APS materials resulted in the most significant increase in resistance. The measured resistance changes value, defined as (R${}_{a}$ $-$ R${}_{g}$)/R${}_{a} \times 100$, exhibited a notable changes as shown in Fig. 4 ^[5,6,20]. Fig. 4(a) shows a 1.25% change for the CeO${}_{2}$-rGO APS layer when exposed to both air/HMDSO environments. In comparison, the Pd/ZnO sensor shows an 8.5% change in Fig. 4(b), and the ZnO sensor shows a 15.3% change in Fig. 4(c) ^[6].

Fig. 4. Illustrates the response of (a) CeO${}_{2}$-rGO decorated Pd/ZnO APS sensor and (b) Pd/ZnO sensor, along with (c) ZnO sensor, for H${}_{2}$ (100 ppm) detection. Measurements were taken at 250$^\circ$C under exposure to both air and HMDSO (10 ppm). A study on resistance changes was carried out to investigate the impact of surface modification on HMDSO.

The greatest resistance is achieved with surface modification using the APS layer, showing almost no change in resistance. In the APS layer, CeO${}_{2}$ supplies an abundance of oxygen molecules to the surface of ZnO nanoparticles, reducing sensitivity to HMDSO. The rGO layer functions as a filter, reducing the SiO${}_{2\ }$thickness on the nanoparticles, thus enhancing the resistance of APS against HMDSO ^[8,21]. HMDSO cannot be fully decomposed within the inner layer of Pd/ZnO, while the resistance of the APS layer remains nearly unchanged. The APS layer fully decomposes HMDSO without affecting H${}_{2}$ detection, ensuring gas sensing performance remains intact. The anti-poisoning mechanism is briefly elaborated upon. APS sensing materials CeO${}_{2}$-rGO/Pd/ZnO effectively prevent poisoning gases. In this APS technology, we will apply it in ECS applications, examining its lifespan, sensitivity to food poisoning gases, and developments in refrigerator sensor technology. The detection of spoilage gases/particles such as HMDSO, ammonia, hydrogen sulfide, ethanol, methanol, and carbon dioxide ensures reliable monitoring of food freshness and safety with sensors. We will conduct experimental studies and research on detecting food poisoning gases and developing stable refrigerator sensors for long-term monitoring and detection.

3. Anti-poisoning mechanism

HMDSO, commonly found as a toxic agent in gas sensors, presents the greatest risk to the effectiveness of sensors used for detecting flammable gases or oxidizing volatile organic compounds (VOCs) ^[9,22]. It generates organic silicon compounds, further degrading sensing performance through a surface-chemical process on the metal/metal-oxide surface, terminating in the formation of a SiO${}_{2}$ layer. Previous studies on HMDSO-treated sensors have shown that the thin and porous SiO${}_{2}$ layer blocks the diffusion of analyte gas molecules ^[21]. This study involved the surface modification of Metal-oxide sensors with an anti-poisoning composite layer. This involved applying an APS layer to enhance the sensor's resistance against HMDSO deactivation. The anti-poisoning mechanism is explained in detail below.

The sensitivity of both metal/metal-oxide sensors decreases due to silicon contamination from HMDSO, leading to the formation of a SiO${}_{2}$ layer on the sensor surface and thereby degrading their ability to detect analyte gases. In the presence of moist atmosphere, these siloxane bonds in HMDSO hydrolyze, creating Si$\mathrm{-}$OH groups ^[23]. These groups react with hydroxyl groups on the ZnO surface, reducing their number. At elevated temperatures, a porous and thin SiO${}_{2}$ layer forms. This SiO${}_{2}$ formation also degrades the catalytic activity of Pd nanoparticles by reducing active sites. The SiO${}_{2\ }$layer effectively damages the ZnO and Pd/ZnO sensors, as illustrated in Fig. 5(a). H${}_{2}$ molecules are catalytically dissociated on the Pd surface, producing H atoms that rapidly diffuse into the ZnO bulk ^[6,24]. These H atoms react with surface oxygen to form hydroxyl groups (OH${}^{-}$), which can further react with more H atoms to produce water, leaving behind oxygen vacancies ^[23,24]. These vacancies function as donor states, increasing the electron concentration in ZnO. Furthermore, the H atoms form hydrogen-based species within ZnO, creating additional donor states and further increasing electron concentration. Understanding these anti-poisoning mechanisms is crucial for improving sensor design.

Fig. 5. This illustrates the response of three sensors: (a) CeO${}_{2}$-rGO decorated Pd/ZnO APS sensor, which showed minimal resistance changes compared to (b) Pd/ZnO sensor and (c) ZnO sensor, during H${}_{2}$ detection. Measurements were conducted at 250$^\circ$C under exposure to air and HMDSO (10 ppm). The study focused on investigating how surface modification affects HMDSO-induced resistance changes.

Anti-poisoning sensor designs utilize CeO$_2$-rGO anti-poisoning layer coatings on Pd-decorated ZnO NPs for HMDSO poison elimination and H$_2$ (100 ppm) detection at 250$^\circ$C. It effectively decomposes HMDSO without interfering with H$_2$ detection, thereby preserving gas sensing performance. In CeO${}_{2}$/rGO/Pd/ZnO APS, CeO${}_{2}$ provides sufficient oxygen to the surface of ZnO nanoparticles, reducing sensitivity to HMDSO in surface chemical processes, while the rGO layer functions as an HMDSO filter. Surface modification with the anti-poisoning CeO$_2$-rGO layer was used to improve the tolerance of the Pd/ZnO sensor to HMDSO deactivation, as shown in Fig. 5(b). In this anti-poisoning mechanism, we employed previous references to significantly optimize the performance of APS sensors against poisoning gases. HMDSO preferentially contacts the APS layer and is decomposed on the CeO${}_{2}$-rGO surface, thereby protecting the inner sensor layer from HMDSO deactivation.

We introduced the APS CeO$_2$/rGO layer coated onto the Pd/ZnO layer and annealed it at 400$^\circ$C. We measured the change in resistance in a 10 ppm HMDSO/air environment at 250$^\circ$C. It was clearly observed that there was no change in the response or the resistance, as shown in Fig. 5(c). This means that HMDSO did not reach the Pd/ZnO through the layer, and therefore, the SiO$_2$ layer is negligible ^[5,6,24]. The anti-poisoning sensor layer can completely decompose HMDSO without depleting H$_2$, and thus it will not affect gas sensing performance. In this paper, we demonstrated the use of CeO$_2$-rGO coatings on Pd NP-decorated ZnO NPs for the simultaneous removal of HMDSO poison and detection of H$_2$. We observed the conditions outlined in a previous report, making minor adjustments to achieve optimal performance. Developments in APS technology are promising for the successful commercialization of MOS sensors.

We will study refrigerator sensor mechanisms and the activity of APS materials. Fig. 6 illustrates APS tools for electrochemical sensors. This work leveraged APS technology to develop long-lasting MOS gas sensors for HMDSO resistance and H$_2$ detection. The objective of these APS material systems is to improve the lifespan and reliability of electrochemical sensors (ECS) for future applications in detecting food spoilage gases such as ammonia, methane, carbon monoxide, and H$_2$ in refrigerator environments ^[25].

Fig. 6. Schematic of APS technology applied to electro-chemical refrigerator sensors for detecting food spoilage gases and improving lifespan: future work.

The unique APS characteristics of this work, in comparison to our previous report, are highlighted through the studies presented in Table 1.

Table 1. Comparison of material composition, HMDSO resistance, and application focus in gas detection.