The Membrane-Covered Clark Dissolved Oxygen Sensor
Before 1956, dissolved oxygen sensors consisted of bare, un-insulated noble metal electrodes. These sensors had a major drawback in that they were quickly coated with the measured medium, which degraded the measuring result.
American scientist Leland C. Clark (1918–2005) invented a polarographic oxygen sensor that overcame this issue. The Clark sensor consists of two electrodes, a silver anode and a platinum cathode. Both electrodes are immersed into a half-saturated potassium chloride (KCl) electrolyte chamber. The electrolyte chamber is separated from the sample solution by a Teflon membrane that is held in place by a rubber ring. Oxygen molecules from the sample solution diffuse through the membrane into the electrolyte. The platinum cathode is completely insulated by a glass cylinder, and only the tip, with a diameter of approximately 20 μm, is exposed to the electrolyte. The electron flow between the two electrodes, when polarized with a negative potential of -600 mV, determines the oxygen concentration in the measured medium.
The Clark oxygen sensor is a two-electrode sensor, i.e., the anode is combined with the reference electrode. To fulfill the role of a reference electrode, the anode must be coated with AgCl or silver bromide (AgBr). The electrolyte surrounding the anode must also contain the coating material.
The most important development of the Clark oxygen sensor is the protection of the electrodes from the measured medium by a membrane. This membrane must have the ability to allow oxygen to diffuse through it from the measured medium to the cathode. The membrane material is normally polytetrafluoroethylene (PTFE) or fluorinated ethylene propylene (FEP), with a thickness of between 10 and 50 μm. These plastics possess a high oxygen permeability, are highly resistant to chemical corrosion, and can be exposed to high temperatures. The membrane must be fitted to the sensor in such a manner that it will lie flat against the cathode tip so that all the diffused oxygen is reduced.
The membrane-covered Clark sensor requires a constant flow of new liquid in order to supply oxygen molecules to the sensor cathode. If the reduction at the cathode is low, the flow velocity can also be low, but if the reduction is fast, the flow velocity of the measured medium must be increased accordingly.
The oxygen consumption at the cathode depends on the thickness of the membrane and the size of the cathode tip area. A thicker membrane decreases the oxygen consumption and increases the response time. A smaller cathode tip area also decreases the oxygen consumption.
By optimizing these parameters, it is possible to slow the flow velocity of the measured medium down to less than 5 mm/s. However, a low flow velocity increases the sensor response time beyond acceptable levels. Achieving a fast response time requires a thin membrane (approximately 10 μm). The oxygen consumption at the cathode again depends on the thickness of the membrane and the size of the cathode tip area; therefore, a sufficiently thin membrane to achieve fast response time requires a high flow velocity of up to 300 mm/s.
Prior Article - Principles of Polarographic Measurement
Next Article - Sensor Output Signal
Download Our O2 Measurement Guide
Get a better understanding of O2 measurement in Hamilton’s comprehensive O2 Measurement Guide.