Nanogap traps for passive bacteria concentration and single-point confocal Raman spectroscopy

Abstract

A microfluidic device enabling the isolation and concentration of bacteria for analysis by confocal Raman spectroscopy is presented. The glass-on-silicon device employs a tapered chamber surrounded by a 500 nm gap that serves to concentrate cells at the chamber apex during sample perfusion. The sub-micrometer gap retains bacteria by size exclusion while allowing smaller contaminants to pass unimpeded. Concentrating bacteria within the fixed volume enables the use of single-point confocal Raman detection for the rapid acquisition of spectral signatures for bacteria identification. The technology is evaluated for the analysis of E. cloacae, K. pneumoniae, and C. diphtheriae, with automated peak extraction yielding distinct spectral fingerprints for each pathogen at a concentration of 10³ CFU/ml that compare favorably with spectra obtained from significantly higher concentration reference samples evaluated by conventional confocal Raman analysis. The nanogap technology offers a simple, robust, and passive approach to concentrating bacteria from dilute samples into well-defined optical detection volumes, enabling rapid and sensitive confocal Raman detection for label-free identification of focused cells.

FIG. 1.

Nanogap-enabled bacteria trapping, concentration, and detection. (a) Sample solution containing bacteria is perfused into an open chamber separated into two volumes by a 500 nm tall V-shaped gap. Rigid bacteria cells cannot enter the nanogap and are forced to accumulate along the gap entrance. (b) Side view of the sample perfusion step. Cells roll along the trap entrance until reaching the tip where they are trapped. (c) After perfusing the desired sample volume, fresh buffer is introduced through the inlet to complete the capture process and rinse the focused bacteria. (d) Air is flushed through the chip to dry the focused sample. (e) Single-point confocal Raman detection is performed to extract spectral signatures from the concentrated bacteria at the chamber tip.

FIG. 2.

(a) Nanogap chip fabrication process comprising nanogap patterning, microchannel and focusing chamber patterning, port etching, and glass/silicon anodic bonding. (b) Fabricated glass/silicon chip with an array of devices with different trap chamber volumes. (c) SEM image of a single trap. (d) Magnified view of the trap apex where Raman detection is performed.

FIG. 3.

Simulation of bacteria trapping performed via COMSOL particle tracing. The inflow was set as fully developed flow and an average linear velocity of 20 μm/s was applied to ensure numerical stability. The characteristic transport length scale given by the product of mean flow velocity (u) and focusing time (t) is provided for each frame.

FIG. 4.

(a) Optical micrograph showing trapped E. cloacae bacteria at the nanogap tip. Raman detection was performed at a point 3–4 μm behind the tip to reduce background signal from the silicon surface. (b) Typical background spectrum collected from the silicon trap surface and used during background subtraction in subsequent bacteria analyses.

FIG. 5.

Raman spectra acquired following capture of (a) C. diphtheriae, (b) E. cloacae, and (c) K. pneumoniae bacteria in a nanogap chip. Matching spectra for samples deposited on a bare cover slip are shown for each bacterium. Peaks identified in either the nanogap or reference spectrum but absent from the corresponding spectrum are denoted with an asterisk (*).