Improving the performance of label-free optical biosensors

Summary form only given. Rapid and real-time detection of antigens, bacteria, viruses, etc., for medical diagnostics and environmental monitoring requires the development of highly sensitive biosensors. Traditionally, labeled sensors, such as ELISA assays and fluorescent immunoassays, have been used for as highly sensitive detection platforms in complex environments, such as serum or whole blood. These sensors are particularly successful in such environments due to their high specificity to the target molecule. However, labeled sensors detect the presence of the fluorescent probe rather than the molecule of interest, requiring a priori knowledge of the target. In direct contrast, label-free sensors, such as cantilever sensors, optical waveguide sensors, and surface plasmon resonators, detect the molecule itself, but many times with less sensitivity or with difficulty in minimizing false positives in complex environments. To overcome these limitations, improvements in both the sensitivity and specificity of label-free sensors must be simultaneously addressed. Fortunately, a new class of high-performance optical sensors, developed from whispering gallery mode microcavities, has shown improved capabilities for sensing. Although these devices were initially designed for telecommunications applications, their very low optical loss makes them prime candidates for biosensing platforms. This is due to the correspondence between the low material loss of the microcavities and the long circulating lifetimes of photons within the cavity, which enables "optical amplification" of otherwise undetectable signals, and therefore improves the signal to noise ratio. The primary metric used to quantitatively describe the optical loss is the quality factor (Q) of the optical resonator. The two most commonly used optical resonators, the microsphere and the microtoroid, have Q-factors in excess of 100 million, corresponding to photon storage times greater than 100 ns. The performance of these highly sensitive optical platforms may be further improved by the addition of a component that adds specificity to the device. This would improve the ability of these platforms to compete with labeled assays in complex environments. Typically, this is done via physical adsorption to the surface, but this technique may not be stable to environmental changes, such as temperature or pH fluctuations. An improved method is the covalent attachment of probe molecules that specifically detect the target molecule of interest. However, to date, a general strategy for the bioconjugation of label-free, high sensitivity whispering gallery mode resonators via covalent attachment has not been developed, primarily due to the difficulties of surface conjugation without adversely impacting the sensitivity of the device. Therefore, it is crucial to develop a uniform, covalent surface functionalization process that is capable of maintaining the optical device's performance metrics, such as the quality factor. In this approach, we use the silica ultra-high-Q microtoroid microcavity as the test platform, as it is the only microcavity fabricated on a silicon substrate which has achieved Q factors in excess of 100 million. We selectively functionalize the surface of these silica microtoroids using a three step process: 1) hydroxylation, 2) amination, and 3) biotinylation. Optical and scanning electron microscopy are used to qualitatively characterize the as-fabricated and surface-modified devices. Microcavitiy analysis techniques are used to quantitatively probe the effects of the surface modifications on the quality factor of the devices. Together, these techniques enabled the identification of the conditions best suited to ensuring the devices' high performance. Additionally, the surface chemistry and properties of these devices are explored via X-ray photoelectron spectroscopy, contact angle measurements, and fluorescent imaging at each reaction step, and show uniform surface coverage of only the devices themselves, and not the surrounding substrate. Based on these analyses, the optimal parameters for fabrication and surface modification protocols have been found, and result in the creation of devices whose surface functionalities have uniform packing, low optical absorption, specificity only to the target molecule, and high stability in air and water, as well as high device sensitivity. This work demonstrates one of the first examples of covalent bioconjugation of microtoroidal optical resonators that may be used for the label-free detection of biomolecules. The primary advantage of a device that has been functionalized using the methods described above is that a wide array of biomolecules can be detected and studied, since the functionalization process does not limit the study to a specific ligand-substrate pair, or even a specific type of molecule. These methods can used in conjunction with a variety of silica-based optical sensors, such as microdisk and microsphere resonators, and waveguide sensors, resulting in an array of highly specific and sensitive sensor devices. Using these techniques, the overall performance of label-free optical biosensors can be improved for use in complex environments. This represents a significant step forward in our ability to improve the performance of label-free sensors.