The Nano Sensors Group has recently demonstrated a fundamentally new approach for resonant optical biosensors in which an active optical resonator – in the form of a Distributed Feedback (DFB) laser - is used to generate its own high intensity narrow bandwidth output without any requirement for precise incoupling of excitation illumination or precise outcoupling to collect emitted light, and is shown conceptually in Figure 1. Lasers are well-known to produce a high-intensity narrow bandwidth output through the process of stimulated emission within a resonant optical cavity incorporating a gain medium. The resonant optical cavity of a laser has a Q-factor, as defined by Δνcav , but the spectral output range of the laser oscillator, Δνosc, will be more narrow than Δνcav, as shown in Equation (1) below:
(1)
where Δνosc is the FWHM of the laser output, Pout is the output power, g1 and g2 are cavity parameters (g1,2 = 1-(d/R1,2)) and N1(s) and N2(s) are the density of lower and upper energy states for electrons. Equation (1), which applies to any laser, shows that bandwidth of the light output decreases with the square of the Q-factor of the laser cavity. We must, therefore, clearly differentiate the Q-factor of a cavity, Qcav, from the Q-factor of a laser oscillator, Qosc. Because extra linewidth narrowing is provided by the laser, it was our hypothesis that, for a biosensor incorporating an active resonator structure in which shifts in laser emission wavelength are monitored as the sensed output, high resolution may be achieved while maintaining a high level of detection sensitivity.
In most lasers, simple mirrors are used as feedback elements – however, it is possible to distribute the feedback if light is forced to propagate through a medium comprised of periodic partial reflections, where the fraction of light reflected at each small boundary is small. The total field propagating in the negative direction does not remain small when the phase delay between individual small wave packets is chosen correctly. Therefore, the feedback is frequency selective, since only frequencies satisfying the condition
, (where L is the period and m corresponds to the order) generate components that add in phase, and thus result in feedback of significant amplitude. Such a feedback system can be realized by producing a periodic structure along the emission axis of a semiconductor laser diode, and in molded polymer structures with large surface areas. While the m=1 order results in laser emission in the plane of the grating, the m=2 order results in outcoupling in the direction normal to the grating surface. DFB laser structures have been studied extensively, and their design/operation is summarized in many textbooks, including.
Of the many different types of resonant optical cavities that could be selected for use as a laser-based biosensor, we have selected the DFB structure for the following reasons: 1). The DFB structure is comprised of a subwavelength periodic grating that may be inexpensively fabricated from polymer materials over large surface areas by replica molding. 2). The DFB laser may be fabricated as a surface structure for which part of the oscillating mode resides within the dielectric medium in contact with the surface structure, and therefore provides the opportunity for laser wavelength tuning by attachment of biomolecules to the surface. 3). DFB lasers may be designed to emit only a single mode (i.e. a single wavelength) that will be directed normal to the surface by first order diffraction[69]. 4). Qosc of DFB lasers of 10,000 are typically demonstrated. 5). DFB laser emission may be achieved using low power optical pumping.
The implications of these features are of enormous practical importance for the development of a sensor technology to be used in life science research laboratories. It will be possible to fabricate large surface areas of DFB laser biosensor material on flexible, inexpensive plastic substrates and to incorporate them into disposable labware. With an entire surface comprised of usable sensor material, there is no necessity to precisely aim immobilized ligand to bind only upon certain regions. As this technology develops further, it will be possible to simultaneously measure many regions on the sensor surface in parallel, and to easily incorporate references and controls into every experiment. Using optical pumping to excite the laser emission, only millimeter-level precision is needed to aim the pump source at the desired detection area. The laser emission itself is of sufficient intensity for easy detection with inexpensive sensors, and is captured with a simple optical fiber. Performing a single measurement is also extremely fast. The optical pump duration is ~10 nsec, resulting in an output pulse of approximately equal duration. Further, if the pump laser is focused down to a ~1 mm diameter spot on the DFB laser biosensor surface, the resulting emission wavelength will be determined by only a very small region around the excitation spot, so that by rastering the excitation spot sequentially across a DFB surface, a spatial map of adsorbed biochemical or cellular binding density may be obtained.
Fig.1.
Schematic diagram of the sensor structure and experimental setup.

DFB laser biosensor incorporated with bottomless 96-well microplate (under UV lamp).
Fig.3.

Spontaneous emission and laser spectra for the DFB laser recorded for pump fluences below and above threshold. The inset displays the dependence of the relative laser output power on the pump fluence. Observed laser spectrum when the sensor surface is immersed in DI water.
Fig.4.

The kinetic detection of wavelength shift of the streptavidin immobilization and the biotin binding with the streptavidin process.
Fig.5.

Detection of alternating layers of positive and negative charged polymer. Inset depicts shifts of the Intensity Spectrum collected by the Spectrometer

Binding kinetics of different Human IgG concentration to Protein measured on the biosensor. (a) Laser Wavelength shift respective to time with different concentration. (b) Stabilized wavelength shift respective to Human IgG concentration on log scale.
Conferences: