Nanoplasmonic In Situ Spectroscopy for Catalysis Applications

Review article, 2012

Indirect nanoplasmonic sensing, INPS, as the key step forward, facilitates the use of nanoplasmonic sensor technology in highly demanding environments in terms of temperature (up to 850 °C, so far), chemical harshness (strongly oxidizing and reducing atmospheres), and pressure for in situ and real time probing of catalyst and other functional nanomaterials. Furthermore, INPS allows for almost infinite material combinations. We also note that the pressure range within INPS can be used is not limited by the sensor or readout principle itself, but rather, by the design of the measurement cell; hence, experiments above atmospheric pressure should be straightforward. The INPS sensor chip features a dielectric spacer layer physically separating the nanoplasmonic sensors from the probed nanomaterial and serving several additional key functions, including (i) protection of the Au nanosensors from the environment and from structural reshaping at high temperature, (ii) providing tailored surface chemistry (support mateial) for the nanomaterial/catalyst to be studied, (iii) being chemically inert or (iv) participating actively in the process under study, e.g., in spillover effects during a catalytic reaction. In principle, any other dielectric material (oxides, nitrides, carbides) that can be deposited as a thin flat or porous film-but also polymers-can be used as the spacer layer/support material for an INPS experiment, depending on the needs of the specific probed system. To date, we have successfully applied the INPS sensing platform to investigate structural and chemical changes of nanomaterials, such as in catalyst sintering processes,19 the oxidation/reduction of Pd nanoparticles, or the storage of NOx species in BaO. We have also applied INPS to scrutinize size effects in the hydride formation process in nanoparticles in the sub-10 nm size range1,16 or to measure in situ changes in adsorbate surface coverage on heterogeneous catalysts at atmospheric pressure.2 Optical nanocalorimetry has been used to measure local temperature changes at the nanolevel and relate the latter, for example, to the activity of a catalyst.1 Furthermore, we have recently applied INPS to study dye molecule adsorption/impregnation of 10-μm-thick mesoporous TiO2 photoanodes in dye-sensitized solar cells by placing the INPS sensor at the hidden, internal interface between the support and the mesoporous TiO2.18 This approach provides a unique opportunity to selectively follow dye adsorption locally in the hidden interface region inside the material and inspires a generic and new type of nanoplasmonic hidden interface spectroscopy that makes highly time-resolved measurements inside a material possible. This first application of hidden interface INPS has thus also prepared grounds for studies of even more realistic catalyst structures comprising a micrometers-thick mesoporous washcoat-like support structure on the INSP chip, loaded with "real" catalyst nanoparticles. Finally, we have also demonstrated first experiments toward single particle INPS spectroscopy in the example of hydride formation in individual Pd and Mg nanoparticles. In summary, owing to its sensitivity, versatility, robustness, compatibility with harsh environments and high temporal resolution in the millisecond range, INPS constitutes a very promising novel experimental platform for the in situ spectroscopy of functional nanomaterials such as catalysts under close-to or real application conditions. The lack of specificity of the readout signal, that is, shifts in the spectral position of the localized plasmon peak of the INPS sensor, requires careful design of experiments and, in some cases, combinations with complementary techniques, such as AFM/ SEM/TEM, or other spectroscopic techniques, such as XPS. Hence, as one important future direction for further development, we identify the direct integration of the INPS function on a sample compatible with simultaneous additional readouts (such as the aforementioned ones but also others, such as quartz crystal microbalance,42 or nonlinear optical spectroscopies, such as SFG) as a high priority. Furthermore, we believe that more efforts directed toward the probing of individual catalyst nanoparticles during a catalytic reaction are well motivated, because of both the promising first proof-ofprinciple experiments already presented and the potential to efficiently circumvent inhomogeneous sample material artifacts. As the main challenges, here, we identify on one hand the optimization of the utilized microspectroscopy for compatibility with high temperatures and, on the other hand, the further optimization of sensitivity and geometrical arrangement of sensor and probed nanoparticle to ultimately be able to probe individual nanoparticles in the sub-10 nm size range under realistic application conditions. © 2012 American Chemical Society.