Development of a Colorimetric H₂ Sensor Based on PtWO₃ Nanoparticles

2.1 Material and synthesis of PtWO₃ nanoparticles

All chemicals were of analytical grade, purchased from Sigma-Aldrich (USA), and used as received without further purification. The precursors included tungsten hexachloride (WCl₆, 99.9%) as the tungsten source and hexachloroplatinic acid as the platinum source. Ethanol (99.9%) and deionized (DI) water were used as solvents throughout the synthesis.

The Pt-doped tungsten trioxide (PtWO₃) nanoparticles were synthesized via controlled thermal decomposition under a nitrogen atmosphere.

First, 17.4 mmol of WCl₆ and 11.9 mmol of hexachloroplatinic acid were dissolved in 36 mL of ethanol under vigorous stirring until a homogeneous suspension was obtained. The mixed solution was then dried at 120°C for 12 h under a nitrogen atmosphere to remove the residual solvent and initiate precursor decomposition.

Subsequently, the dried powder was calcined at 300°C for 1 h to reach the crystalline PtWO₃ phase, optimizing its color-change behavior toward hydrogen gas. A slow heating rate (approximately 2°C min⁻¹) was maintained to prevent agglomeration and promote uniform Pt incorporation into the WO₃ lattice.

After cooling naturally to room temperature, the resulting grayish-blue PtWO₃ powder was finely ground in an agate mortar to yield well-dispersed nanoparticles (20–50 nm). The powders were stored in airtight glass vials to prevent moisture absorption and were directly used to fabricate the PDMS-based colorimetric membranes described in Section 2.2.

2.2 Fabrication of colorimetric membrane on alumina substrates

The PtWO₃/PDMS composite solution was prepared by dispersing the as-synthesized PtWO₃ nanoparticles in a pre-polymerized PDMS matrix (Sylgard 184, base:curing agent = 10:1 wt ratio). The mixture was magnetically stirred for 1 h to achieve a uniform dispersion of the nanoparticles, followed by vacuum degassing for 30 min to remove entrapped air bubbles. The absence of air voids is critical for ensuring film uniformity and optical transparency, which directly affects the reproducibility of the colorimetric response.

After degassing, approximately 100 µL of the composite solution was drop-cast onto a pre-cleaned alumina (Al₂O₃) substrate (10 × 10 mm² ). The substrates were ultrasonically cleaned in ethanol and DI water for 10 min each and dried under nitrogen flow before coating. The film was then spin-coated at four different rotation speeds (600, 700, 800, and 900 rpm) for 60 s to investigate the influence of coating thickness and surface uniformity on optical performance.

Following spin coating, the samples were thermally cured at 80°C for 2 h in a convection oven to complete the PDMS cross-linking and immobilize the PtWO₃ nanoparticles within the elastomeric network. The resulting membranes exhibited smooth and flexible surfaces with good adhesion to the alumina substrate. The film thickness decreased from approximately 120 µm at 600 rpm to 60 µm at 900 rpm, as estimated using a surface profilometer. The coating prepared at 800 rpm showed the best balance between uniformity and mechanical robustness.

The fully cured membranes were subsequently stored in a desiccator prior to the gas-sensing experiments to prevent ambient moisture adsorption, which can alter the baseline color of PDMS.

2.3 Gas exposure and colorimetric measurement

All test gases, including H₂, CO, NO, NO₂, and CO₂, were supplied by Korea Standard Gas (Republic of Korea) as N₂-balanced certified calibration gases. A 10% H₂ (N₂-balanced) cylinder was used as the parent source and was diluted with high-purity N₂ and dry air using mass flow controllers (MFCs) to achieve final hydrogen concentrations of 1%, 2%, 3%, 4%, and 5%. The total flow rate was maintained at 1000 sccm, which consisted of 500 sccm of dry air and 500 sccm of a mixed stream of N₂ and the target gas. Before each run, all gas lines were purged with pure N₂ for at least 10 min to remove any residual species.

Each prepared gas mixture was collected in a Tedlar bag to ensure compositional accuracy and flow stability. Gas exposure tests were conducted in a custom-built acrylic chamber (6 × 4 × 8 cm³ ) that could be evacuated to a near-vacuum state before gas introduction. After evacuation, the pre-filled Tedlar bag gas was injected through a controlled inlet until the chamber pressure reached the ambient level, ensuring a homogeneous gas distribution around the sensor.

All measurements were performed at a controlled temperature of 23 ± 1°C and relative humidity of 30 ± 2%, with continuous monitoring using a temperature and humidity sensor (BLUETEC, BO-807). The PtWO₃/PDMS-coated alumina sensor was placed at the center of the chamber and exposed to the target gas for 15 min (reaction phase).

Immediately after the exposure phase, the recovery behavior was evaluated by opening the chamber directly in ambient air, allowing natural desorption and color restoration under atmospheric conditions. The color change process during both exposure and recovery was recorded using a Galaxy Note 8 smartphone camera, which provided real-time visual documentation under ambient lighting.

The captured images were analyzed using ImageJ software (version 1.54, NIH, USA) to extract the corresponding RGB intensity values, which were then converted to CIELAB (L*, a*, b*) coordinates. The overall color difference (ΔE) was calculated as follows:

where the subscripts 1 and 2 represent the RGB values before and after gas exposure, respectively. A ΔE > 2.3 was defined as the threshold for a color change perceptible by the human eye [17]. Each experiment was repeated thrice to verify the reproducibility and stability of the colorimetric response. To ensure consistent image capture, all photographs were captured under controlled lighting conditions with fixed white balance and exposure settings.

3.1 Spin-coating speed optimization

The effect of the spin coating rate on the optical response of the PtWO₃/PDMS colorimetric membrane was systematically investigated, as illustrated in Fig. 1.

Fig. 1.

Optimization of spin-coated PtWO3/PDMS membranes for H2 sensing. (a) Colorimetric images of PtWO3/PDMS membranes before/after 3% H2 exposure and after recovery at different spin rates (600–900 rpm), (b) ΔE variations showing the highest contrast at 800 rpm.

Fig. 1(a) shows the photographic images of the sensor surfaces before/after exposure to 3% H₂, and after the recovery phase at four different coating speeds (600, 700, 800, and 900 rpm).

At lower speeds (600 and 700 rpm), the films appeared relatively thick and slightly opaque, resulting in weaker and slower color transitions during H₂ exposure.

In contrast, the film coated at 800 rpm exhibited the most distinct and rapid color change from light gray to deep blue, indicating optimal film thickness and uniform nanoparticle dispersion.

When the spin speed was further increased to 900 rpm, the coating became too thin and partially nonuniform, reducing the active volume for gas interaction and thus diminishing the visible color contrast.

The quantitative colorimetric responses corresponding to these visual results are shown in Fig. 1(b).

The ΔE value increased with the spin speed up to 800 rpm and then decreased beyond that point, forming a bell-shaped curve, which is a characteristic of diffusion-limited thin-film systems.

The sensor fabricated at 800 rpm achieved the highest ΔE of approximately 11.3 upon exposure to 3% H₂, which is well above the perceptibility threshold (ΔE > 2.3), confirming a clearly visible color change to the human eye.

This optimal behavior can be attributed to a balanced combination of light transmittance, gas permeability through the PDMS matrix, and effective optical path length within the PtWO₃-dispersed layer.

After the exposure phase, all samples showed reversible color recovery when the chamber was opened to ambient air.

However, the 800 rpm specimen produced the strongest color contrast and exhibited the fastest and most complete recovery within approximately 5 min, demonstrating the high sensitivity and stable structural integrity of the membrane.

These results establish 800 rpm as the optimal spin-coating condition for the fabrication of reproducible and responsive PtWO₃/PDMS colorimetry.

3.2 Colorimetric behavior of PtWO₃/PDMS membranes at different H₂ concentrations

3.2.1 Response characteristics

The overall colorimetric response of the PtWO₃/PDMS membrane to various hydrogen concentrations is shown in Fig. 2(a).

Fig. 2.

Colorimetric response of the PtWO3/PDMS membrane to different H2 concentrations. (a) Membrane photographs before/after exposure and after recovery at various H2 concentrations (1%–5%), (b) ΔE–time profiles during reaction and (c) recovery, showing rapid, reversible color changes.

When exposed to H₂ gas (1%–5%), the initially light-gray surface gradually turned blue, and the coloration intensity increased proportionally with the gas concentration.

A noticeable color transition occurred even at 1% H₂, with the film becoming distinctly deep blue at concentrations above 3% H₂, thus confirming high optical sensitivity.

The temporal evolution of ΔE during exposure is presented in Fig. 2(b).

A rapid increase in ΔE was observed within the first few minutes, followed by a gradual saturation behavior as equilibrium was reached between H₂ adsorption and surface reduction of WO₃.

The membrane reached 90% of its final ΔE value within approximately 4 min, demonstrating fast response kinetics.

The maximum ΔE values were approximately 3.2, 7.8, 11.3, 14.2, and 19.7 for 1%, 2%, 3%, 4%, and 5% H₂, respectively, indicating a clear concentration-dependent response.

This coloration originates from the catalytic dissociation of H₂ on Pt nanoparticles and subsequent proton spillover to the WO₃ lattice, leading to the formation of tungsten bronze (H_xWO₃). The resulting intervalence charge transfer between W⁵⁺ and W⁶⁺ ions generates the characteristic blue color [20]. This mechanism provides both high sensitivity and rapid optical transition under ambient conditions without requiring external power.

3.3 Selectivity against interfering gases

The selectivity of the PtWO₃/PDMS membrane toward hydrogen gas was investigated by comparing its colorimetric response to various potentially interfering gases under identical experimental conditions (Fig. 3). The gas tested included CO₂ (5%), CO (1000 ppm), NO₂ (1000 ppm), and NO (1000 ppm), with all mixtures balanced in nitrogen and supplied by Korea Standard Gas (Republic of Korea).

Fig. 3.

ΔE responses of the PtWO3/PDMS membrane to different gases, confirming high selectivity toward H2.

As shown in Fig. 3, the membrane exhibited a strong optical response only to hydrogen exposure, with negligible color changes from the other gases. The corresponding ΔE values were approximately 11.3 for 3% H₂, whereas the other gases—CO₂, CO, NO₂, and NO—produced ΔE values of only 0.21, 0.52, 0.49, and 0.17, respectively. All of these were far below the perceptible color-difference threshold (ΔE = 2.3), demonstrating excellent selectivity for hydrogen.

This selective response originated from the unique redox chemistry of the PtWO₃ composite. Under hydrogen exposure, the Pt nanoparticles catalyze the dissociation of H₂ molecules, followed by proton spillover to the WO₃ lattice. The resulting reduction process leads to the formation of tungsten-bronze (H_xWO₃) species, causing the pronounced blue coloration via intervalence charge transfer (W⁶⁺ ↔ W⁵⁺). In contrast, other test gases (CO₂, CO, NO, and NO₂) neither dissociated effectively from Pt at room temperature nor induced sufficient reduction of WO₃, resulting in minimal color variation.

Moreover, the hydrophobic PDMS matrix served as a gas-permeable yet chemically inert host that limited the adsorption of polar or oxidizing species, further enhancing the hydrogen-specific response. The combination of Pt-catalyzed dissociative activation and WO₃ optical transition provided a composite film with high selectivity, chemical stability, and repeatable colorimetric performance for hydrogen leak visualization in ambient environments.