Optofluidic microscope shrinks to fit on a chip

According to Yang, the simple design means that tens or even hundreds of these microscopes can be built on a single chip and operate in parallel to speed up the imaging process at a cost per chip of around $10 (€7). This portable and cheap device is particularly appealing for third-world applications where it could be used in the field to analyse blood samples for malaria or check water supplies for pathogens.

"We can build very cheap, iPod-sized, rugged microscopes that fit easily into a health worker's back pocket," said Yang. "We can potentially furnish each scientist or clinician with hundreds or thousands of microscopes. Implantable blood analysis devices could even be built to continuously monitor blood in vivo to screen for circulating tumour cells or white blood cell population ramps, which is indicative of infection."

Optical microscopy pervades almost all aspects of modern bioscience research and clinical procedures. Current microscope designs are limited by relatively low throughput, high cost and high space requirements in which lenses are required to focus and magnify images. Since these optical elements are difficult and expensive to miniaturize, the Caltech team turned to optofluidics (an emerging field that combines microfluidics and optics) to create the new device.

Microfluidic flow

Turning to optofluidics
The approach is based on optofluidic microscopy (OFM) and uses a microfluidic flow to deliver specimens across an array of micrometre-sized apertures etched onto a metal-coated CMOS sensor. This direct imaging approach has several advantages. The lack of optical elements in the arrangement implies that there are no aberrations to worry about. Also, this is an intrinsically space-conserving method.

The device is fabricated by coating a linear sensor array with a layer of metal to block out light. A line of holes is then punched into the metal layer. Finally, a microfluidic channel is added on top.

"By covering the sensor grid with a thin metal layer and etching small apertures onto the layer at the centre of each pixel, the sensor pixel will be sensitive only to light transmitted through the aperture," explained Yang. "Although the method is non-magnifying, the resolution is determined only by the aperture size and not by the pixel size of the CMOS sensor."

When operating, the device is uniformly illuminated from above with white light from a halogen lamp (20 mW/cm², approximately the intensity of sunlight). The target object flows across the array of holes via the microfluidic channel and the time-varying light transmission through each hole forms a transmission image line trace across the object. By stacking the line traces from all of the holes together, a transmission image of the object can be constructed.

Going with the flow
Although the group first conceived the OFM method four years ago, until now it had no means of flowing the cells across the imaging system in a controlled manner.

Yang realized that by applying an electric field across the microfluidic channel, the sample could be drawn across the imaging system. "We found that a constant voltage applied to a pair of platinum electrodes at the channel inlet and outlet provides a simple and direct way to control the motion of biological cells on-chip," he said.

In the set-up, a voltage of 25 V is applied across the inlet and outlet of a microfluidic channel that is 2.4 mm long, 40 µm wide and 13 µm high. The electric field draws the specimen across the aperture array in a steady stream. The array consists of 120 holes with a diameter of 0.5 µm and separation of 10.4 µm, fabricated on a 2D CMOS imaging sensor. The sensor comprises a grid lattice of 1280 x 1024 square pixels with a pixel size of 5.2 µm.

The grid is tilted at a small angle to create a diagonal line with respect to the flow direction causing the images to overlap slightly. All of the images are then pieced together to create a precise two-dimensional picture of the object with a resolution of 800 nm.

Yang is now working on translating the technology into a commercially available product and is optimistic that the resolution of the device could be improved. "In principle, we can push the resolution down past the diffraction limit to around 50 nm as the resolution is limited only by the size of the holes that we can punch," he said. "The issue with this refinement is that at that resolution, the device will only image the surface of cells, not the interior. We can work on ways to pattern the holes so that we get better resolution as well as the ability to image the interior of cells."

The group is also working on adding fluorescence and phase imaging capability to the OFM. "There are a lot of potential applications that we would like to evaluate. Our main obstacle lies in getting the word out there and soliciting interested bioscientists and clinicians to help us to evaluate this technology," concluded Yang.

• For more details on the device go to the Biophotonics Laboratory, California Institute of Technology webpage at www.biophot.caltech.edu/people/yang.html or e-mail Changhuei Yang direct at chyang@caltech.edu.

• This article originally appeared in the October 2008 issue of Optics & Laser Europe magazine.

View pdf of article