(H-308) Investigation on methods to improve the limit of detection of an optical cavity-based biosensor

Introduction:: One of the clearest ways to improve the survivability of diseases is to diagnose them earlier so that treatment can begin sooner. Many highly sensitive diagnostic methods for a variety of biomarkers are being developed, including the gold standard enzyme-linked immunosorbent assay (ELISA). However, like ELISA, many of these diagnostic methods are labor-intensive and necessitate the use of specially trained personnel at facilities external to the point of care. These drawbacks increase cost and decrease accessibility, making it more difficult for people to effectively monitor their health status, thereby decreasing the chances of early detection. The development of low cost, label-free, and highly sensitive point-of-care (POC) biosensors is a promising alternative. The label-free feature is particularly important as it simplifies the detection process significantly, allowing for the development of detection methods viable for use at home and in low-resource environments. To this end, we have proposed an optical cavity-based biosensor (OCB) as a POC biosensor for the early diagnosis of diseases. Our OCB is based on an optical cavity structure with two partially reflective silver layers and consists of low-cost components and its sensitivity is enhanced by using a differential detection method. We successfully demonstrated the detection of a biomarker, C-Reactive Protein (CRP), with a limit of detection (LOD) of 377 pm. In this presentation, we will share results from our investigation of the reflectance of the partially reflective layers of the OCB and the equation used for the differential detection method for the purpose of improving this LOD further.

Materials and Methods:: The inside of the optical cavity structure is functionalized with bioreceptors such that analytes attach to the surface, causing a local refractive index (RI) change that shifts the resonance response. The cavity size is optimized such that, during this shift, the intensity of two different wavelengths changes linearly in the opposite direction. The intensities of the two wavelengths are measured using low-cost laser diodes and a CCD camera. A differential detection method is employed to enhance the sensitivity. To improve the LOD of the OCB further, we investigated two methods: (1) increasing the reflectance of the partially reflective layer using thicker silver layers and (2) improving the differential equation used for the differential detection method. With a thicker silver layer, the reflectance increases which in turn increases the quality (Q) factor for the optical resonator. A higher Q-factor makes the OCB more responsive to the shift thereby yielding a higher differential value change. Regarding the differential equation, the old equation calculates the difference of the intensity changes (or efficiency in simulation) divided by the baseline intensity. To improve the LOD, we proposed an improved differential equation to use both intensity changes more effectively. For this investigation, we used FIMMWAVE/FIMMPROP software and employed a fixed index model by which the analyte binding is approximated as a sensing layer with a fixed refractive index and certain thickness.

Results, Conclusions, and Discussions:: Previously, using the old differential equation, we demonstrated a LOD of 377 pM for CRP using a silver thickness of 20 nm and a cavity size of 8 μm. At that cavity size and silver thickness, the simulated differential value produced by a 20 nm sensing layer was 0.525. We further optimized the cavity size for this silver thickness and improved responsivity (the differential value change from a 20 nm sensing layer) with a cavity size of 10.8495 μm for wavelengths of 880 nm and 904 nm. Then, we simulated the resonance characteristics of 5 other silver thicknesses up to 45 nm with an increment of 5 nm. Figure 1 shows the responsivity of six silver thicknesses with both the old and improved differential equations. With the old equation, at a silver thickness of 20 nm, the responsivity of the OCB is 0.529. As the silver thickness increases to 45 nm, the optimized cavity size becomes 11.2019 μm and the responsivity increases to 3.239, a 6.12 times increase. This means the LOD could be decreased from 377 pM to 62 pM by using a silver thickness of 45 nm assuming an equivalent noise level. As shown in the figure, the responsivities calculated with the improved differential equation are higher than those with the old differential equation. This is because, in the improved equation, the term for the wavelength that reduces in intensity when the resonance response shifts is more responsive since the denominator uses the reduced intensity rather than the initial value, which would be higher. When the improved equation is used with 45 nm of silver, the responsivity is increased 1.56 times to 5.067, indicating that the LOD could be lowered even further from 62 pM to 40 pM. Currently, we are working to experimentally demonstrate this LOD improvement with a thicker silver layer and the improved equation. In conclusion, using a thicker silver layer and the improved equation, we demonstrated that the LOD of our POC OCB biosensor could be improved 9.4 times, equivalent to 40 pM for CRP detection.

Acknowledgements (Optional): : Research reported in this presentation was supported by NIGMS of the National Institutes of Health under award number R15GM146233.


References (Optional): :