A shorter path to circumventing the low reproducibility of SERS spectra through variable screening and optimization...

Currently, several SERS methodologies are focused on quantitative goals, and, as they move toward becoming a suitable analytical alternative, their main drawback remains the low reproducibility of SERS signals. As a result, strategies to circumvent this are highly desirable, but given the number of experimental variables that affect reproducibility, the task can sometimes be endless. Here, we selected a series of C5-substituted uracil derivatives (5-fluoro-, 5-chloro-, and 5-bromouracil) adsorbed on silver nanoparticles. These compounds are of analytical and pharmacological interest due to their role as biomarkers, pharmaceuticals, or environmental pollutants. They are expected to be found at very low levels in bioanalytical matrices; thus, to reach low limits of detection, they were paired with adenine, which has a well-known high affinity for silver nanoparticles. Considering the complexity of these pairs and the experimental variables that affect SERS reproducibility, a two-stage strategy was adopted to select the best experimental conditions and minimize the standard deviation (STD) of the signals from the three base pairs. This included a variable screening procedure followed by an optimization stage. After optimization, the STD values decreased by 100-, 300-, and 1000-fold for 5-fluorouracil, 5-bromo, and 5-chlorouracil, respectively. The validation experiments yielded STD values at the same order of magnitude as those achieved under the optimized conditions, indicating that our approach can be applied to improve signal reproducibility. In perspective, this strategy can be easily adapted to other SERS-based systems, especially for analytical purposes.