Fabrication of homogeneous shell-isolated sers substrates for catalytic applications

Chemically synthesized (CS) metal-nanoparticles (MNPs) for surface enhanced Raman spectroscopy (SERS) provide high orders of electric field enhancement that are useful for applications in real-time monitoring of chemical reactions.[1] However, a limitation is the inhomogeneous SERS signals over large areas due to the random distribution of MNPs. For applications in catalysis, where the MNPs could be active in the catalytic reaction, the MNPs also need be coated with an insulating shell. This insulating shell leads to a reduction in the enhancement, but provides a higher thermal stability to the MNPs and limits the Raman signals from undesired side-products. This approach, which is known as shell-isolated nanoparticle-enhanced Raman spectroscopy (SHINERS), has been successfully applied for catalysis, but controlling the shell thickness and homogeneity and ensuring that it is pin-hole free is challenging. [2]

For in-situ monitoring of catalytic reactions, it is therefore critical to have homogeneous SHINERS substrates that result in strong SERS signals over large areas. Additionally, it is also important to have homogeneous, stable and pin-hole free shells to achieve proper isolation. In this work, we report and investigate two improved methods to fabricate lithographic SHINERS substrates with an application in real-time monitoring of CO2 hydrogenation. As shown in our previous work [3], lithographically fabricated SERS substrates not only provide high orders of enhancement factors (EFs) (~ 108) but also contribute to the homogeneity of the SERS signal with only ~ 4% variance in the average EF. To fabricate lithographic SHINERS, we investigate two methods for shell-isolation that can be directly applied to the lithographically fabricated SERS substrates.

For method 1, we synthesize a shell on a lithographic MNP nanocone substrate using chemical precursors while for method 2, we use an atomic layer deposition (ALD) process to form a shell on lithographically fabricated nanodots. Figure 1 shows the Rhodamine 6G spectra for a CS shell on MNP-nanocone substrate and an ALD shell on lithographically fabricated nanodots. Here, a decrease in the Raman intensity for the shell-isolated substrates compared to their non-isolated counterparts, can be expected due to the presence of an insulating layer. The presence of this insulating shell increases the distance of the sensing molecule from the enhancing surface, therefore reducing the local electric field intensity where the molecule is detected. This can also be evidenced from the finite-difference-time-domain (FDTD) simulations, as shown in Figure 2. When compared to a CS shell, we find that the ALD shell is conformal, controlled and reproducible and shells as thin as 2.5 nm can be formed. The SEM images of Figure 3 show the differences between a pin-hole rich and pin-hole free ALD film, after being subjected to gold etchant. As a proof of concept, we show the ability to fabricate lithographic SHINERS using two different methods and prove that ALD combined with MNP-Nanocone is a better choice for applications in the field of catalysis. The combination of a low-variance SERS substrate with a conformal ALD coating will ensure homogeneous SHINERS sensing capabilities for catalytic reactions.