Optical and electrochemical (EC) oxygen sensors are both proven technologies, so choice of a sensor really has more to do with how they each function and what you want from your analyzer. In this post we’ll focus on optical technology, and I’ll follow up with an EC post next time.
Let’s start by talking about one of the characteristics common to both types of probes. With the exception of fuel cell probes measuring dry gases in super, super low concentrations of gas-phase oxygen, all oxygen probes (as far as I’m aware) measure the partial pressure of the oxygen.
So what exactly is partial pressure? It’s a more precise way of saying “percent O2.” For example, whether you are measuring on top of Mt. Everest or in the depths of Death Valley, the sensor will tell you there is about 21% oxygen. But if the percentage of oxygen is the same in both places, why is it that most people do fine in Death Valley, but can’t survive the summit of Mt. Everest without extreme conditioning or supplemental oxygen? It’s because that 21% is relevant to the other gases in the air. On Mt. Everest there are fewer molecules total, and that means there isn’t enough O2 to survive.
So both optical and electrochemical oxygen sensors measure partial pressure, and when measuring oxygen dissolved in a liquid – as you are with beer – the partial pressure is used in conjunction with Henry’s Law to calculate the solubility of gases that can dissolve in the liquid. The units of measurement are expressed as a weight per volume of liquid, usually mg/L or µg/L. Most brewers, however, differ from this a bit. They use units of ppm instead of mg/L, and ppb in place of µg/L.
An optical sensor works by exposing it to the sample you want to measure. (For example: beer, wort, or the CO2 blanket in a bright beer tank.) The sensor contains a polymer coated with florescent properties that is easily permeated by oxygen in the environment touching the sensor. To work, the sensor flashes blue light on the florescent coating; it then fluoresces and emits red light back to a detector that is proportional to the amount of oxygen in the florescent coating.
The wonderful thing about optical sensors is that the florescent matrix doesn’t consume the oxygen, so the sample flow can be very low. Since there are no membranes to change, the sensors don’t need much maintenance and can be calibrated infrequently. Pressure spikes and CIP don’t cause sensor drift, so they are exceptionally robust.
The only downside to optical sensors is that they don’t have the same dynamic range as traditional electrochemical sensors. Optical sensors that are tuned for measurements in the ppb range lose their linearity above 2 ppm and become inaccurate. Sensors that are tuned to measure above 2 ppm lose their sensitivity below about 0.2 ppm and have a typical uncertainty of 0.15 ppm.
My final thought is that your choice of sensor will most likely depend upon your particular application, although an optical sensor will always have the advantage of being easier to maintain.