Effect of Pt Loading on the Colorimetric Hydrogen-Sensing Properties of MoO₃ Nanorods

2.1 Synthesis of MoO₃ NRs

MoO₃ NRs were synthesized using a hydrothermal method [12]. Molybdenum powder (2.0 g, -250 mesh, 99.9%, Thermo Fisher Scientific Inc.) was dissolved in 20 mL of H₂O₂ solution (35 wt%, Thermo Fisher Scientific Inc.) with stirring at 500 rpm for 20 min. The solution was heated at 230°C for 12 h in a 90 mL Teflon-lined autoclave. The product was washed with deionized water and ethanol by centrifugation, dried at 60°C overnight, and calcined in air at 300°C for 2 h.

2.2 Preparation of Pt-MoO₃ NRs

Pt-decorated MoO₃ NRs were prepared by a pyrolysis-assisted deposition method [13]. Chloroplatinic acid hexahydrate (0.1 g, Sigma-Aldrich) was dissolved in 2 mL of deionized water by sonication for 40 min. MoO₃ NRs (0.1 g) were dispersed in 10 mL of methanol (≥99.8%, Thermo Fisher Scientific Inc.), and the Pt solution was added to obtain Pt loadings of 1, 3, and 5 wt%. The mixtures were stirred at 800 rpm for 24 h, freeze-dried for 12 h, and calcined in air at 300°C for 2 h. The samples were denoted as 1Pt-MoO₃, 3Pt-MoO₃, and 5Pt-MoO₃ based on Pt loading.

2.3 Characterization

The morphology and microstructure were examined by field-emission scanning electron microscopy (FE-SEM, Gemini 500, Zeiss) and high-resolution transmission electron microscopy (HR-TEM, Titan G2 Cube 60-300, FEI) equipped with energy-dispersive X-ray spectroscopy (EDS). The Pt contents were quantitatively determined by inductively coupled plasma mass spectrometry (ICP-MS, NexION-350X, PerkinElmer). The crystal structures were analyzed by X-ray diffraction (XRD, LabX XRD-6100, Shimadzu) using Cu Kα radiation (λ = 0.1541 nm). The specific surface area (SSA) was determined by Brunauer-Emmett-Teller analysis of nitrogen adsorption isotherms (BELSORP MINI X, MicrotracBEL). Surface chemical states were analyzed by X-ray photoelectron spectroscopy (XPS, Sigma Probe, Thermo VG Scientific) using an Al Kα source. Optical absorption was measured using an ultraviolet–visible–near-infrared (UV-Vis-NIR) spectrophotometer (V-770, JASCO).

2.4 Sensor Fabrication

The synthesized powder (0.1 g) was mixed with polydimethylsiloxane (PDMS, Sylgard 184 Silicone Elastomer Kit, Dow Corning) at a weight ratio of 1:15 using a paste mixer (AR-100, Thinky) at 1000 rpm for 10 min. A curing agent was added at a PDMS-to-curing-agent weight ratio of 10:1, followed by mixing for 3 min. The mixture was spin-coated onto a 50 × 50 mm² glass substrate at 500 rpm for 60 s and cured at 70°C for 30 min. The colorimetric sensor was cut into 10 × 10 mm² pieces for colorimetric sensing measurements. For the H₂ leakage test, the film was cut into 10 × 50 mm² strips and used as tape-type colorimetric sensors.

2.5 Gas-Sensing Measurements

Colorimetric sensing measurements were conducted in an 8 mL quartz chamber. All analyte gases were supplied as N₂-balanced gas mixtures and further diluted with N₂ (99.999%, Hankook Special Gases) to the required concentrations. H₂ sensing was evaluated over a concentration range of 0.5–5 vol%, and selectivity was examined against CH₄ (5 vol%), NH₃ (100 ppm), acetone (5 ppm), ethanol (5 ppm), and CO (5 ppm). The total gas flow rate was fixed at 500 mL min^-1, and all measurements were performed at room temperature (~25°C). The quartz chamber was placed inside a dark enclosure to minimize variations in external lighting conditions. The color change of the sensor was monitored in real time using a digital microscope (HiMax Tech.) positioned approximately 3 cm above the sensor surface. Illumination inside the enclosure, camera exposure, and white-balance settings were kept constant throughout the measurements to ensure consistent image acquisition. The acquired images were processed using a custom-developed image-analysis program. The color change was quantified using the color difference (ΔE) in the International Commission on Illumination (CIE) L*a*b* color space according to the following equation:

where L, a, and b represent lightness, the green–red coordinate, and the blue–yellow coordinate before gas exposure, respectively, and L′, a′, and b′ represent the corresponding values after gas exposure. A ΔE value of 3 has been reported to represent a clearly visible color difference [14]. Accordingly, the response time was defined as the time required for ΔE to reach 3 after gas exposure.

3.1 Characterization of Pure MoO₃ and Pt-MoO₃ NRs

The morphologies of pure MoO₃ and Pt-MoO₃ NRs were examined by SEM (Fig. 1(a)). All samples exhibited similar one-dimensional NR morphologies, with no noticeable changes as Pt loading increased. TEM analysis of 3Pt-MoO₃ revealed nanoparticles attached to the MoO₃ NR surfaces (Fig. 1(b)). HR-TEM analysis revealed lattice fringes with interplanar spacings of 0.35 and 0.20 nm, corresponding to the (040) plane of orthorhombic MoO₃ [15] and the (200) plane of metallic Pt [16], respectively (Fig. 1(c)). EDS elemental mapping further showed the distribution of Pt over the MoO₃ NR surfaces (Fig. 1(d)). ICP-MS analysis determined Pt contents of 0.873, 3.24, and 4.58 wt% for 1Pt-MoO₃, 3Pt-MoO₃, and 5Pt-MoO₃, respectively. These results confirm the successful decoration of metallic Pt nanoparticles on the MoO₃ NRs.

Fig. 1.

Structural and morphological characterization of Pt-MoO3 NRs. (a) FE-SEM images of MoO3, 1Pt-MoO3, 3Pt-MoO3, and 5Pt-MoO3. (b) TEM, (c) HR-TEM, and (d) EDS elemental mapping images of 3Pt-MoO3 NRs. (e) XRD patterns of all samples. (f) Pore size distribution and SSA (inset) of 3Pt-MoO3 NRs.

XRD patterns showed that all samples exhibited the orthorhombic MoO₃ phase (JCPDS #05-0508) (Fig. 1(e)), consistent with the HR-TEM results. No noticeable shifts in the major diffraction peaks were observed with increasing Pt loading, suggesting that Pt was not incorporated into the MoO₃ lattice. No diffraction peaks corresponding to Pt were observed, likely due to the low Pt content. These results a indicate that Pt decoration did not alter the crystal structure of the MoO₃ NRs.

Pore-size analysis of 3Pt-MoO₃ revealed pores spanning a broad size range from several to tens of nanometers (Fig. 1(f)), which can be attributed to void spaces formed between the assembled NRs. The SSA of 3Pt-MoO₃ was 4.39 m² g^-1 (inset of Fig. 1(f)). These inter-rod void spaces provide pathways for gas transport through the sensing material, which is favorable for gas sensing.

3.2 Colorimetric Hydrogen Sensing Properties of Pt-MoO₃ Sensors

The colorimetric H₂-sensing properties of pure MoO₃ and Pt-MoO₃ sensors were evaluated using 5 vol% H₂ at room temperature. Pure MoO₃ initially appeared gray and showed little color change even after H₂ exposure for more than 10 min (Fig. 2(a)). In contrast, all Pt-MoO₃ sensors exhibited a clear color change to dark navy upon H₂ exposure and did not return to their original color after H₂ removal, demonstrating that Pt decoration is essential for the colorimetric H₂ response. The initial color of the Pt-MoO₃ sensors became progressively darker with increasing Pt loading.

Fig. 2.

Colorimetric responses of Pt-MoO3 NRs to 5 vol% H2 at room temperature. (a) Time-dependent optical images. (b) Time-resolved changes in ΔE. (c) ΔE values at selected exposure times. (d) Response time (τres) as a function of Pt loading.

The color changes were quantitatively evaluated using the time-dependent ΔE (Fig. 2(b)). The results revealed contrasting effects of Pt loading on the magnitude and rate of the color change. The ΔE decreased in the order of 1Pt-MoO₃, 3Pt-MoO₃, and 5Pt-MoO₃ over the measured exposure times (Fig. 2(c)), indicating that higher Pt loading resulted in a smaller color difference.

In contrast, the response time decreased with increasing Pt loading (Fig. 2(d)). Based on the time required to reach ΔE = 3, the response time decreased from approximately 17 s for 1Pt-MoO₃ to 15 and 9 s for 3Pt-MoO₃ and 5Pt-MoO₃, respectively. Thus, increasing Pt loading resulted in a smaller color difference but a faster response, demonstrating its contrasting effects on the colorimetric sensing characteristics of Pt-MoO₃.

3.3 Colorimetric Sensing Mechanism of Pt-MoO₃ Sensors

To examine the H₂-induced coloration mechanism of Pt-MoO₃, the surface chemical states and optical properties of 3Pt-MoO₃ were analyzed before and after H₂ exposure. The Mo 3d XPS spectrum [17,18] of the pristine sample consisted of Mo⁶⁺ and Mo⁵⁺ components with relative fractions of 66.6 and 33.4%, respectively (Fig. 3(a)). The O 1s spectrum [19] showed lattice oxygen (O_L, 81.8%), oxygen vacancies (O_V, 16.4%), and surface hydroxyl species (O_H, 1.8%) (Fig. 3(b)). Upon H₂ exposure, the Mo⁵⁺ fraction increased to 71.4%, while the Mo⁶⁺ fraction decreased to 28.6% (Fig. 3(c)). Concurrently, the O_V and O_H fractions increased to 30.7 and 4.3%, respectively, while the O_L fraction decreased to 65.0% (Fig. 3(d)). These changes indicate that H₂ exposure reduces MoO₃, accompanied by increases in oxygen vacancies and surface hydroxyl species.

Fig. 3.

Surface chemical and optical analyses of 3Pt-MoO3 NRs. (a, b) Mo 3d and O 1s XPS spectra before H₂ exposure. (c, d) Mo 3d and O 1s XPS spectra after H₂ exposure. UV-Vis-NIR absorbance spectra (e) before and (f) after exposure to 5 vol% H₂.

The H₂-induced chemical changes were accompanied by a pronounced change in optical absorption. Compared with the pristine sample (Fig. 3(e)), the H₂-exposed sample showed substantially increased absorption across the visible and NIR regions (Fig. 3(f)). This increase can be attributed to reduced Mo species and to defect-related electronic states formed upon H₂ exposure. In particular, intervalence charge transfer between adjacent Mo⁵⁺ and Mo⁶⁺ species can contribute to absorption in the visible region [20], accounting for the characteristic navy coloration. Taken together, the XPS and optical results indicate that H₂ exposure reduces MoO₃, increasing Mo⁵⁺ and oxygen vacancies and consequently altering its electronic structure and optical absorption. These changes produce the colorimetric H₂ response of Pt-MoO₃.

Pt nanoparticles facilitate this process by promoting H₂ activation at room temperature [21]. Because direct H₂ activation on MoO₃ is limited, Pt promotes H₂ dissociation and the transfer of activated hydrogen species to the MoO₃ surface through hydrogen spillover [8]. This accounts for the negligible color change of pure MoO₃ and the pronounced coloration of Pt-MoO₃ upon H₂ exposure. The faster response with increasing Pt loading can therefore be reasonably attributed to enhanced H₂ activation and subsequent hydrogen transfer to MoO₃.

In contrast, increasing Pt loading decreased ΔE. As shown in Fig. 2(a), the initial color became progressively grayer with increasing Pt loading. This change is likely associated with the optical contribution of Pt nanoparticles [22], as it differs from the characteristic navy coloration induced by H₂ exposure. The darker initial color reduces the optical contrast between the initial and H₂-exposed states, resulting in a smaller ΔE at higher Pt loadings. Therefore, increasing Pt loading accelerates the colorimetric response but decreases the resulting color difference. Considering these contrasting effects, 3Pt-MoO₃ was selected for further evaluation because it retained a relatively large ΔE while exhibiting a shorter response time than 1Pt-MoO₃ (Table 1).

Table 1.

Colorimetric H2-sensing characteristics of MoO3 NRs with different Pt loadings.

3.4 Gas Sensing Performance of the 3Pt-MoO₃ Sensor

The H₂-sensing performance of the selected 3Pt-MoO₃ sensor was further evaluated under various conditions. Fig. 4(a) shows the time-dependent ΔE as the H₂ concentration was decreased from 5 to 0.5 vol%. As the H₂ concentration decreased, the color change became smaller and slower. Nevertheless, at 0.5 vol% H₂, ΔE gradually exceeded the visible color-difference criterion (ΔE = 3) upon prolonged exposure. This demonstrates that 0.5 vol% H₂ can be visually detected after prolonged exposure.

Fig. 4.

Colorimetric H2 sensing performance of 3Pt-MoO3 NRs. (a) Time-resolved ΔE changes at 0.5–5 vol% H2. (b) Selectivity against interfering gases. (c) Time-dependent optical images under 5 vol% H2 at RH 0–90%. (d) Sensor-to-sensor reproducibility under 5 vol% H2.

The selectivity of the 3Pt-MoO₃ sensor was evaluated against CH₄, NH₃, acetone, ethanol, and CO (Fig. 4(b)). H₂ produced a pronounced color change, whereas all interfering gases showed only minor changes, even after prolonged exposure. The ΔE values of the interfering gases remained below 3, demonstrating that the visible color change was highly selective to H₂ under the tested conditions.

The effect of relative humidity (RH) on the H₂ response was examined over an RH range of 0–90% (Fig. 4(c)). The sensor exhibited comparable color changes upon H₂ exposure across the tested RH range, showing that its colorimetric response was not significantly affected by humidity. This humidity tolerance is favorable for H₂ detection under varying environmental conditions.

Finally, sensor-to-sensor reproducibility was evaluated using ten 3Pt-MoO₃ sensor pieces (Fig. 4(d)). All ten sensors exhibited similar ΔE values and visually consistent color changes upon H₂ exposure. The small variation among the sensors confirms the reproducibility of the fabricated colorimetric sensors. Collectively, these results demonstrate that 3Pt-MoO₃ maintains reliable colorimetric H₂-sensing performance across different H₂ concentrations, interfering gases, humidity conditions, and individual sensor units.

3.5 H₂ Leakage Detection Using the 3Pt-MoO₃ Colorimetric Sensor

The practical applicability of the 3Pt-MoO₃ sensor was examined using a simple H₂ leakage test (Fig. 5). The sensor film was attached near a gas-line fitting, which was slightly loosened to create a controlled leakage point, and 5 vol% H₂ was subsequently supplied through the gas line. A commercial H₂ sensor was used to monitor the surrounding

Fig. 5.

Demonstration of visual H2 leakage detection using a 3Pt-MoO3 sensor before and 2 min after 5 vol% H2 leakage from a gas-line fitting.

H₂ concentration and ensure safety during the experiment. Before H₂ exposure, the sensor appeared gray. Upon leakage from the fitting, the sensor region near the leakage point changed to dark navy within 2 min, allowing the leakage to be readily identified by visual inspection without additional signal readout. This demonstration shows that the 3Pt-MoO₃ sensor can be directly attached near potential leakage points in H₂ gas lines to provide a simple visual indication of leakage.