Digital surface enhanced Raman spectroscopy for quantifiable single molecule detection in flow

Abstract

Surface enhanced Raman scattering (SERS) provides a label free method of analyzing molecules from diverse and complex signals, potentially with single molecule sensitivity. The chemical specificity inherent in the SERS spectrum can identify molecules; however signal variability arising from the diversity of plasmonic environments can limit quantification, particularly at low concentrations. Here we show that digitizing, or counting SERS events, can decrease the limit of detection in flowing solutions enabling quantification of single molecules. By using multivariate curve resolution and establishing a score threshold, each individual spectrum can be classified as containing an event or not. This binary "yes/no" can then be quantified, and a linear region can be established. This method was shown to lower the limit of detection to the lowest physical limit, and lowered the limit of detection by an order of magnitude from the traditional, intensity based LOD calculations.

Fig. 1. Background subtracted spectra of Nile Blue A at concentrations ranging from 10 pM to 100 nM. Inset plot shows calibration curve created using the area of the peak at 593 cm⁻¹. A limit of detection of 8 ± 3 nM was calculated using three times the standard error in y over the slope of the line of best fit. Error bars are the standard deviation from triplicate measurements.

Fig. 2. Distribution of scores for each sample concentration. The dashed line shows the event threshold (24.8), which was calculated using the three times the standard deviation of the scores of the blank. Error bars are the standard deviation from triplicate measurements.

Fig. 3. Curve showing the percent of spectra from each concentration that are considered to be events. The curve levels off to 100% around 10 nM. The inset plot shows the linear region of this curve. A digital limit of detection of 500 ± 200 pM was calculated using 3 times the standard error in y over the slope of the line of best fit. Error bars are the standard deviation from triplicate measurements.