UC Santa Barbara

Although surface-enhanced Raman spectroscopy (SERS) is a spectroscopic technique with unusually high sensitivity and molecular specificity, few practical analytical applications have been implemented that take advantage of its power. Based on what is understood about SERS from the experimental and theoretical research of the past forty years, we developed a few well-defined design principles on the basis of which a reliable and reproducibly manufacturable SERS-active substrate could be fabricated that is highly enhancing, highly uniform, stable, and based on a broad range of metals so that various chemical processes could be probed. Finally, we restricted ourselves to using only readily scalable fabrication techniques. The resulting SERS-active device was a metal over silica, two-dimensional nano-grating that was shown to produce enhancements of ~10⁷ when compared to a smooth surface of the same metal. This SERS substrate also shows unprecedented signal uniformity over square centimeters, and is fabricated using commonly-available foundry-based approaches exclusively.

Initially, we explored the properties of a gold-coated substrates in which a first-order grating resonance due to long-range symmetry is augmented by a local resonance due to the individual core-shell grating elements. The SERS properties of such grating systems were systematically studied as a function of various structural parameters such as the grating pitch, the inter-element gap and the thickness of the metal layer. The most enhancing substrates were found to have a grating parameter with a radiative, rather than evanescent, first-order resonance; a sufficiently small gap between nearest neighbor grating elements to produce near-field interactions; and a gold layer whose thickness was larger than the electronic mean-free-path of the conduction electrons, so as to ensure a high conductivity for the metal layer to sustain strong surface plasmons.

We applied these same architectural principles to metals other than gold, and concluded that every workable metal (and virtually any material with sufficiently high electrical conductivity) when appropriately nanostructured, has the potential to be an efficient SERS substrate. The use of materials for SERS beyond silver and gold, has significant advantages, most importantly, allowing SERS to be used to study the surface chemistry and catalysis taking place on metals with more interesting chemistries than those of Au and Ag. Additionally, SERS substrates can be fabricated from high natural abundance, low cost materials. This was illustrated by producing SERS substrates using copper, aluminum, and nickel in addition to silver and gold which were used as benchmarks. All five metals were found to yield high SERS intensities. The variation of the SERS enhancement among them is ascribed mainly to local field effects, with the (larger) grating-based enhancement making an approximately equivalent contribution to the SERS enhancement of the five metals. This conclusion is supported by local electric field simulations.

The utility of these new grating-based SERS substrates was demonstrated by implementing them in chemical analysis in both aqueous and gas phases for which, for example, we were able to readily detect opioids such as fentanyl and morphine at concentrations as low as less than one part per billion. Additionally, we have made good progress toward integration of this substrate architecture into a microfluidic channel for a higher degree of sample workup and control.