‘Lab on a chip’ concept takes another step forward

Gary Boas

A variety of applications exist for single-molecule spectroscopy and for fluorescence correlation spectroscopy in particular, including detection of protein conformational changes, DNA hybridization, membrane binding or protein association interactions.

Researchers have described full implementation of fluorescence correlation spectroscopy on a chip, paving the way for a number of advances, from improvements in existing instrumentation to development of new capabilities. In the chip, the excitation beam enters the liquid-core waveguide along the X-direction, and light is confined in liquid and solid core via dielectric multilayers shown in different shades of gray.

Still, although a number of single-molecule spectroscopy methods have been reported, none has involved full optical integration; such integration could contribute to a variety of advances, from improvements in existing instrumentation to the development of new capabilities. Some studies have reported partial integration with micro- and nanofabrication techniques. With these, though, at least one microscope objective is used to define the optical excitation and/or detection volume.

Full integration requires guiding light through microfluidic channels small enough to enable single-molecule detection. Investigators have come up with several ways in which to achieve this, including confining modes in hollow antiresonant reflecting optical waveguides (ARROW).

In the June 11 issue of Optics Express, a group with the University of California, Santa Cruz, and with Brigham Young University in Provo, Utah, reported a fully integrated fluorescence correlation spectroscopy device made possible by the development of liquid-core ARROWs with micron-scale dimensions.

The design of the chip includes a liquid-core waveguide surrounded by dielectric confinement layers. Molecules either can diffuse freely through the waveguide or be controlled by electric fields. When excited molecules fluoresce, the emission is captured and guided along the liquid channel toward an avalanche photodiode detector. ARROW = antiresonant reflecting optical waveguide.

The study represents a collaboration between Holger Schmidt’s group at the University of California, which designed the chip and performed the measurements, and Aaron R. Hawkins’ group at Brigham Young, which developed the fabrication technology. “The main challenge [in designing and fabricating the chip] was to make the excitation volume as small as possible while keeping the waveguide loss low enough to collect enough light,” Schmidt said. “These two requirements are not compatible and represent a trade-off. We optimized our waveguide design and fabrication to end up with a suitable chip.” He added that the ARROWs allowed them to tune the properties of the excitation volume by changing the confinement layer properties.

The design offered a variety of advantages for fluorescence correlation spectroscopy. First, the planar integrated optics enabled the researchers to route all of the signals in the chip plane. “This eliminates the third dimension and should [allow] for much smaller devices which can also be more robust and less expensive than confocal microscopy methods,” Schmidt said.

The design provided for use of a single-pump waveguide to excite multiple channels in parallel on the same chip. Furthermore, because the excitation geometry is defined precisely during fabrication by the waveguide dimensions and the wavelengths used, it is not susceptible to alignment problems or to mechanical disturbances, for example.

The chip is expandable; that is, more integrated optical elements can be added to the same platform. Finally, it is compatible with standard optical fiber technology to deliver the light to, and collect it from, the waveguides.

There are some disadvantages. “It is very difficult for us to go to excitation volumes on the order of 1 fl or lower, which one can achieve with confocal microscopy or metal nanoholes,” Schmidt said. “This leads to an upper concentration limit for single-molecule detection.”

The researchers demonstrated the chip by performing fluorescence correlation spectroscopy measurements. They filled the hollow waveguides with a solution of water and Alexa 647 dye molecules at varying concentrations. Then, using a single-mode fiber, they coupled 633-nm pulses from an optical parametric oscillator made by Coherent Inc. of Santa Clara, Calif., with a pulse width of 200 fs, into a solid-core waveguide. After removing residual excitation light with a filter from Omega Optical Inc. of Brattleboro, Vt., they detected the fluorescence emitted into the detection arm using a single-photon avalanche photodiode made by PerkinElmer Inc. of Waltham, Mass.

Subsequent analysis showed excellent agreement between the experimental data and an ARROW-based fluorescence correlation spectroscopy model, confirming the utility of the chip for FCS measurements.

The researchers continue developing the chip. Plans include detection of single viruses, further integration of the chip (adding filters and sources, for example), expanded functionality (adding single-molecule control), and more. In a Lab on a Chip paper published online on June 27, they reported use of the chip to detect and analyze liposomes at the single-particle level. They showed that they could control the liposomes electrically and then use fluorescence correlation spectroscopy to determine concentration and dynamic properties, including velocity and the diffusion coefficient.

Schmidt noted that the chip could be used with single-molecule detection methods other than fluorescence-based methods, including Raman scattering and capillary electrophoresis.