Google Custom Search



Get our eNewswire Email:

Fiber Optic Oxygen Sensor Frequently Asked Questions

Fiber Optic Oxygen Sensors use a fluorescence method to measure the partial pressure of dissolved or gaseous oxygen. Optical fiber carries excitation light produced by the Blue LED to the thin-film coating at the probe tip. Fluorescence generated at the tip is collected by the probe and carried by the optical fiber to the high-sensitivity spectrometer. When oxygen in the gas or liquid sample diffuses into the thin-film coating, it quenches the fluorescence. The degree of quenching correlates to the level of oxygen pressure.

What is the wavelength of light measured in detecting fluorescence?

The sensor formulation absorbs excitation energy from a blue LED (470 nm peak output). The fluorescence emission peak is near 600 nm.

How exactly does the oxygen quench the fluorescence?

Oxygen as a triplet molecule is able to quench efficiently the fluorescence and phosphorescence of certain luminophores. This effect (first described by Kautsky in 1939) is called "dynamic fluorescence quenching." Collision of an oxygen molecule with a fluorophore in its excited state leads to a non-radiative transfer of energy. The degree of quenching is related to the frequency of collisions, and therefore, to the concentration, pressure and temperature of the oxygen-containing media.

How fast does the probe respond?

The probe response is limited by the speed of diffusion of oxygen into the sensor. Our standard films are very fast (<1 second) in gases and slightly slower in liquids. In viscous samples, the diffusion through the sample will determine the response rate. The probe will respond quickly in samples such as water, and more slowly in viscous samples such as oils, emulsions and creams. Unlike an electrode, the optical sensor will not consume oxygen. This means that stirring the sample will increase the response rate, but will not affect the final equilibrium reading. Also, optional probe overcoatings, applied to exclude ambient light and improve chemical resistance, will slow probe response to 20-30 seconds in gases and 30-50 seconds in liquids.

What chemicals are known to affect the probe coating or measurement accuracy?

The glass is susceptible to long-term attack by high-alkalinity environments and HF (hydrogen fluoride), as well as to severe physical abrasion. The glass is immune to common organic solvents. In the event that the coating is damaged or destroyed, the probe can be re-coated at a nominal charge. We also provide witness samples (coated glass cover slips) for testing susceptibility to unusual or hazardous solvents, or to conditions we can't reproduce in our lab. Measurement accuracy is affected by compounds that competitively quench the fluorescence. For details on each sensor coating's chemical compatibility, click here.

Why does the probe last only 1 year?

The probe is warranted for 1 year as a cautious approach to a new product. It is expected that probes in benign environments may last for a very long time. However, it is too soon in the probe's product development to discern with great certainty.

Does anything slow or accelerate probe degradation?

Because the sensor formulation may be subject to photodegradation, the total exposure to the excitation energy should be minimized. The operating software will turn the LED on and off so that it is illuminating the probe only during a measurement cycle. Also, severe biofouling, physical abrasion, and chemical etching of the glass may erode the sensing surface. Cleaning and protection from these environments will extend the life of the sensor.

Is the bifurcated fiber assembly used only to connect the probe to the instrument and excitation source, or is the probe itself a bifurcated optical fiber?

There are several probes available as off-the-shelf items, and others available as custom designs. The standard probe -- the "-R" Probe -- is a 1 mm diameter, silica-core, silica-clad fiber in a 1/16" OD stainless steel ferrule. The tip is polished at a 30-degree angle and is coated with the sensor formulation. The other end is terminated with a standard SMA connector. This probe couples to a bifurcated 600 um diameter fiber with 1 leg used for excitation and the other for bringing the fluorescent signal to the spectrometer. For a list of all probe options, click here.

Can I use a sensor probe to measure the headspace in packaged products?

Yes, Ocean Optics' needle probes are built into a special non-coring needle for insertion through septa on serum vials. If the probes are supplied with an overcoat, they can be used to test both headspace and dissolved oxygen in the sample. The headspace in pharmaceutical packaging and in food and beverage packaging are among the samples our customers have monitored.

Can a smaller diameter fiber be used in the probe?

The diameter size of the fiber determines the signal strength and photometric signal-to-noise. Any size fiber can be coated.

Is the probe influenced by ambient light?

The probe will be affected by ambient light. All probe types should be shielded from ambient light by installation in a closed vessel, or by shielding the probe with a dark housing. Our -T1000 probe has a screw-on cap to block the optical field of view of the fiber, while allowing easy access to the probe tip. Also, an overcoat of silicone can be applied to the probe (at no charge) to eliminate ambient light effects.

What is the probe measurement range?

The probe will respond to gaseous samples from 0-100% (mole percent) at 1 atmosphere pressure (0-760 mm Hg partial pressure). In water, the probe will respond from 0-40.7 ppm (0-760 mm Hg partial pressure). In other liquids, the probe responds from 0 to the saturation level for oxygen in that particular liquid.

Our specifications use the terms "low" and "high" oxygen concentrations. In determining resolution, what oxygen concentration separates "low" from "high"?
The output of the probe (in terms of voltage or fluorescent intensity) is a non-linear function described by the Stern-Volmer equation. The probe is more sensitive at low concentrations. The sensitivity is shown below for units in gas and water, assuming a photometric signal-to-noise ratio of 1800:1. This represents a "typical" setup, averaging across 200 pixels in the fluorescence peak that is about 1800 counts for zero oxygen and 1000 counts for the high standard. A better signal-to-noise ratio can be realized by averaging through time, but at the expense of the sampling frequency.

What is the probe accuracy?

The accuracy of the system is limited by resolution (random noise), deviations from the Stern-Volmer relationship, and the accuracy of the calibration experiments. The Stern-Volmer relationship is valid for low-pressure oxygen levels, such as found in water. Deviations from the Stern-Volmer relationship occur primarily at higher-pressure oxygen levels. These deviations can be quantified, and the calibration adjusted, by obtaining several calibration points instead of the normal high and low.

The accuracy of the calibration is itself important. The figure below shows the magnitude of the error that results from a 1% error in recording either I₀ (intensity of fluorescence at zero-pressure oxygen) or the intensity at the high calibration point.

wpe6.jpg (12157 bytes)

How sensitive is the probe's temperature dependence?

Temperature affects both fluorescence intensity and the collisional frequency of the oxygen molecules with the fluorophore. Temperature change also affects the solubility of oxygen in analytes. The net effect is seen as a change in the calibration slope. For accurate results, the sample must be held at a constant temperature (+/-1°C).

If this is impractical, then temperature should be measured concurrently with a thermistor or thermocouple. A -T-MOD1 or -T-MOD-K controller couples to the RS-232 port of the computer and acquires the temperature data. The USB-LS-450 LED module, which attaches to the USB4000-FL Spectrometer, includes the electronics for reading an RTD type of sensor; OOISensors software uses the temperature data to correct the oxygen readings automatically.

The temperature response of our sensors can be determined by the user, or can be supplied by a factory calibration. Alternatively, data from your own temperature sensor can be read (sensed) by the spectrometer's A/D function, if you provide a 0-10 volt signal.

wpe5.jpg (23421 bytes)

Also, testing reveals that the sensor fluorescence appears to be quenched permanently after 120^o C, indicating an upper temperature limit of 110^o C. Though the low temperature limit has not been tested, we can reasonably deduce that the sensor operates successfully well below the freezing point of water.

What technical references are available in support of the Fiber Optic Oxygen Sensors?

The following are the most significant references to date:

Krihak, M.; Shahriari, M.R. A Highly Sensitive, All Solid State Fiber Optic Oxygen Sensor Based on the Sol-gel Coating Technique. Electronics Letters, 1996, Vol. 32, No. 3.

Wang, W.; Reimers, C.E.; Wainright, S.C.; Shahriari, M.R.; Morris, M.J. Applying Fiber-Optic Sensors for Monitoring Dissolved Oxygen. Sea Technology, March 1999, Vol. 40, No. 3, pp. 69-74.

Krihak, M.; Murtaugh, M.T.; Shahriari. M.R. Fiber Optic Sensors Based on the Sol-gel Coating Technique. Chemical, Biochemical and Environmental Fiber Sensors VIII, 1996, Vol. 2836, pp. 87-98.

Shahriari, M.R.; Murtaugh, M.T.; Kwon, H.C. Ormosil Thin Films for Chemical Sensing Platforms. Chemical, Biochemical and Environmental Fiber Sensors IX, 1997, Vol. 3105, pp. 40-51.

Contact an Ocean Optics Applications Scientist