(562d) Acoustoelectric Effect in Hydrogen Surface Acoustic Wave Sensors (Saw) with Phthalocyanine-Palladium Sensing Bi-Layers

Hydrogen has a role as an important chemical commodity that will continue to increase with the developments of the hydrogen economy. With a lower explosive limit of 4.73% by volume fast and accurate sensors are needed. Our recent work has shown that bi-layers of phthalocyanine-palladium structures can be optimized to construct effective SAW sensors for hydrogen. In a bilayer sensing film structure, we can use the much stronger acoustoelectric effect in the SAW sensor response as the main detection mechanism more effectively than with a single sensing layer. This effect can be many times greater than the mass effect which can be dominant in nonconductive polymer films and simple metal and dielectric films typically employed in SAW gas sensors. The ?work point? of such a structure must be shifted to the high sensitivity region, where small variations in conductivity (under the influence of gas molecules) cause remarkable changes in the wave velocity (see Figure 1). Thus, to take full advantage of the high sensitivity offered by the SAW sensor, the conductivity of sensing film must be tailored to a particular range.

In a recent set of papers, we have explored the possibility of utilizing a bilayer structure comprised of a thin palladium (Pd) film and a copper-, nickel- and or metal-free-phthalocyanine layer on a LiNbO₃ Y-Z substrate for the detection of hydrogen in a medium concentration range. The best results were achieved using metal-free phthalocyanine and Pd. In these films, the top layer consisted of a very thin Pd film ((~20 nm) and the bottom layer consisted of the semi-conducting phthalocyanine material. We have experimentally established that a ~160 nm thick metal-free-phthalocyanine film and the above 20 nm Pd film bilayer gives optimal response. Large frequency shifts of about 15 kHz were observed at hydrogen concentrations above 2 vol% in synthetic dry air at ~50 ^oC. The acoustoelectric nature of the response was established by simultaneous measurements of electrical conductivity on samples prepared by the same process.

Our repeatable measurements (see Figure 2 for results at 31 ^oC) corresponding to changes in film resistance were observed at hydrogen concentrations above 2%, indicative of an acoustoelectric interaction. For hydrogen concentrations between 0.5% and 1.5% in air the variation in frequency Df is rather small, with negligible changes in observed film resistance. These frequency shifts are likely caused by the mass-loading mechanism. At a temperature of ~500 ^oC, very large shifts on the order of 15 kHz were observed for hydrogen concentrations above 2 vol% in synthetic air. Quantitative comparison with perturbation theory and equivalent circuit based models that predict the mass and acoustoelectric contributions to the SAW sensor response are possible and will provide a rapid way of optimizing the Pd and phthalocyanine film thicknesses. We hypothesize that the thin Pd film serves to collect and provide a constant and high concentration boundary condition for hydrogen diffusion into the phthalocyanine film, which can be verified by response-time measurements with and without this film, on resistance-optimized films. Some of these results will be presented in this contribution.