Optofluidic Flow-Through Biosensor Sensitivity - Model and Experiment

Abstract

We present a model and simulation for predicting the detected signal of a fluorescence-based optical biosensor built from optofluidic waveguides. Typical applications include flow experiments to determine pathogen concentrations in a biological sample after tagging relevant DNA or RNA sequences. An overview of the biosensor geometry and fabrication processes is presented. The basis for the predictive model is also outlined. The model is then compared to experimental results for three different biosensor designs. The model is shown to have similar signal statistics as physical tests, illustrating utility as a pre-fabrication design tool and as a predictor of detection sensitivity.

Keywords: Anti-resonant reflecting optical waveguides (ARROW); integrated waveguides; model design; optofluidics; predictive simulation.

Fig. 1.

Overview of the biosensor fabrication process. (a) The SU8 sacrificial core is patterned and developed. (b) Pedestal is patterned and etched by DRIE. (c) A layer of silicon dioxide is grown over the pedestal and SU8 core. (d) The core of the ridge waveguide is etched into the silica layer by RIE. (e) A second silica layer of lower refractive index is grown as a cladding, (f) The sacrificial core is etched out to make the hollow channel.

Fig. 2.

Overview diagram of biosensor chip. The fluorescent particles, bonded to pathogen targets will pass through the blue-shaded ARROW. An excitation mode is guided in a ridge waveguide that orthogonally intersects the ARROW. Fluorescence photons generated by the particles passing through the waveguide mode are collected into another ridge waveguide that couples with the ARROW, to be detected by an off-chip single-photon detector.

Fig. 3.

A close-up view of the excitation region in the biosensor. Fluorescent bioparticles flow through the active hollow core and pass through an excitation waveguide mode. Fluorescent light then guides down the length of the hollow core for detection.

Fig. 4.

Overview of model consturction process. The final excitation region profile is created by the matrix multiplication of power and time factors to create a profile that represents the energy that would be incident on a particle passing through the excitation region. The incident energy is multiplied by a coupling efficiency matrix to create the excitation region profile. This matrix is then sampled to simulate a physical device’s behavior.

Fig. 5.

Example Excitation Waveguide Mode Profile. Generated in Lumerical® MODE Solutions. (a) Mode profile with 0.2×0.2 μm² mesh size. (b) Example of a profile with a very fine mesh size (1×1 nm²).

Fig. 6.

Example of a summed-power vector. The mode power profile is condensed to a single-column vector that contains the summed power values of every row.

Fig. 7.

(a) Velocity profile of ARROW channel with an average velocity that can range from 0.8 to 3 cm/s. (b) Time spent in excitation region by x-y position in channel, defined as the depth of the excitation region divided by the flow velocity.

Fig. 8.

Incident excitation energy profile of ARROW channel.

Fig. 9.

The FDTD-calculated coupling efficiency profile of the channel based on a fluorescent particle’s x-y position.

Fig. 10.

(a) The model of the final excitation region profile is the product matrix of energy density and coupling efficiency matrices. This profile will change based on simulated optical mode, flow velocity profile, and ARROW cross-section dimensions. The outer rows and columns of values are removed as particles typically do not flow there in experimental findings. (b) Example simulated signal graph over chip run time. The signal magnitude is measured in “Counts/0.1ms”, the number of photons detected in a 0.1ms time bin. (c) Example distribution of signal magnitudes. The number of events (No. Events) is the number of particles detected within a signal bin.

Fig. 11.

(a) Standard biosensor excitation ridge profile. (b) Excitation region profile of ARROW channel for a standard design with a double-lobe excitation mode. (c) Excitation region profile for a standard design with a single-lobe excitation mode.

Fig. 12.

(a) Geometry of “Three-Micron” excitation ridge waveguide. (b) Excitation region profile of ARROW channel for “Three-Micron” design.

Fig. 13.

(top) Geometry of “sandwich” excitation waveguide. (bottom) Excitation region profile of an ARROW channel for a “sandwich” design.

Fig. 14.

Signal distribution comparisons of simulated and experimental signals. The signal magnitude is measured in “Counts/0.1ms”, the number of photons detected in a 0.1ms time bin; the number of events (No. Events) is the number of particles detected within a signal bin. (a) Standard design. (b) Three-Micron design. (c) “Sandwich” design.