*3.3. Humidity Sensing Using Transmittance Measurements*

It is well known that in general case measuring the transmittance is easier, more accurate, and less expensive than measuring the reflectance. Therefore, it will be more advantageous to use transmittance measurements as optical read-out for detecting humidity. To perform transmittance measurements, the sensitive medium should be deposited on transparent substrate. Usually these are glass or plastic with approximate refractive index in the range 1.4–1.5 that is very close to refractive index of the polymers used for detection. Thus, the small optical contrast will lead to low sensitivity of detection, because it will be difficult to distinguish the thin film from the substrate because of the match of their refractive indices. We have already shown that quarter-wavelength multilayers stacks (Bragg stacks) and glass covered with thin semitransparent metal overlayer are suitable transparent substrates for optical detecting of humidity in transmittance mode [25,26]. When planar Bragg stacks are used for substrates, the sensitivity of detection increases with thickness of the sensitive medium deposited on top and it is the highest for films thicker than 250 nm [25]. However, as already mentioned an ambiguity exists when films with thicknesses higher than 100–150 nm are used as sensitive media. Therefore, in this study we use thin film with thickness of 80 nm deposited on Au–Pd covered glass substrate. The thickness of the metal overlayer is selected to be 30 nm thus guaranteeing transmittance to be around 50%.

The transmittance of Ac24 thin film with thickness of 80 nm, deposited on glass covered with Au–Pd overlayer with thickness of 30 nm as a function of relative humidity is shown in Figure 4. The observed percentage of hysteresis of 3.6% is very close to the value obtained when reflectance as a function of humidity is used (3.8%) (Figure 2b). Furthermore, similarly to the case of reflectance measurements (Figure 2b), two well-distinguished linear parts of the *T*-vs-RH curve are observed. The calculated sensitivities are comparable to the case of silicon substrate: 0.03 in the range 5%–70% *RH* and 0.14 for *RH* > 70%. However, because of the higher accuracy in transmittance measurements (errT is 0.1% as compared to errR = 0.3% when R is measured) the accuracy of sensing ΔRH is 3 times higher (Equation (2)). Thus, using the configuration polymer film/metal layer/glass, less than 1% RH could be distinguished in the range of high humidity (*RH* > 70%) and 3% for RH < 70% (Table 2).

**Figure 4.** Transmittance versus relative humidity curve for PVA film (80 nm thick) of 24% acetal content, deposited on Au–Pd (30nm) covered glass substrate, pre-annealed at 60 ◦C measured for increasing (solid black symbols) and decreasing (open blue symbols) humidity.
