The design for the oxygen sensor is based upon the quenching effect of oxygen on a fluorescent material. A fluorescent material absorbs light in a certain wavelength range and emits light (fluoresces) over a different range. Figure 1 shows the excitation and emission spectrums for Pt(TfPP). After absorbing light, fluorescent molecules can relax back to the ground state via two main types of energy loss: emission of a photon and by a non-radiative energy loss due to strong collisions. The rate at which excited atoms relax back to the ground state via collisions is oxygen dependent.  An increase in oxygen concentration will lead to more collisions thus increasing the relaxation rate for the collision energy loss mechanism. This, in turn, will decrease the number of photons being emitted. Thus by knowing the rate at which photons are emitted by the fluorescent material, we can determine the oxygen concentration in that material.

The scheme used to detect the rate of fluorescence is phase modulation fluorometry. The fluorescent sample is illuminated with excitation light whose intensity is sinusoidally modulated at a frequency w.  The emission will also be sinusoidal, but will be shifted by a phase angle, as shown in Figure 2. If we assume the use of only pure fluorophores and a homogenous environment, the following relationship between the fluorescent lifetime and the phase shift can be derived from the rate equations:

Figure 3 shows a traditional set-up for this detection scheme.  A LED emits light in the excitation spectrum, illuminating the sensor material.  The sensor material, absorbing the excitation light, fluoresces in all directions. A filter and a lens are incorporated into the system to focus some of the emission light onto the photodiode. The photodiode’s output would go to a lock-in to determine the phase shift from the input signal.

We have explored the use of different optical systems to increase the signal to noise ratio of the device, while attempting to reduce the size and cost. As shown in Figure 4, one such system incorporates a waveguide into the detection scheme. If a thin layer of fluorescent material is adhered onto a microscope slide it creates a waveguide.  The excited fluorescent material emits light in all directions.  The index of refraction of the fluorescent material and the glass microscope slide is higher then that of the surrounding air.  Thus at certain critical angles, the light in the microscope/fluorescent material complex will be internally reflected until it exits out one of the ends of the waveguide.  Thus, by placing a detector at the end of the waveguide, we can detect the collected emission light.

It is important to note the excitation light from the LED has a small divergence angle and is approximately normal to the waveguide.  Thus the light that has not been absorbed will transmit through the waveguide. Since the excitation light is not internally reflected to the detector, we have eliminated the need for a red-pass filter. In the future, we hope to incorporate other optical schemes to futher increase the single to noise ratio while maintaining the objective of a low-cost system.