Polarographic oxygen measurement

For the measurement of oxygen in continuous process analysis, several technologies are available. Because they differ widely in terms of application coverage, field performance and ease of use, the right technology has to be carefully chosen.

In many processes in the chemical, petrochemical and pharmaceutical industries, the level of oxygen at different stages of the manufacturing process must be controlled in order to prevent explosive conditions developing. In other applications, product degradation or further process reaction resulting from exposure to oxygen, must be avoided during final or intermediate storage. In these applications, the oxygen level must be kept below a certain limit in order to ensure safe and reliable process conditions. This is achieved using an inert gas, usually nitrogen, to blanket the vessel or tank. An intermittent or sometimes continuous inflow of nitrogen is then applied, maintaining a constant positive pressure in the tank during all stages. In this way, ingress of detrimental or even dangerous levels of air can be avoided.

Inerting approaches

The inflow of inert gas is necessary in order to ensure that the tank is inerted under all transient phases such as filling and emptying of the tank. Additionally, in- and out-breathing of the tank, due to changes in atmospheric pressure and temperature, must be compensated for. These changes in level or tank pressure require the inflow of nitrogen to be correspondingly adapted, leading to a varying oxygen concentration in the tank. If not measured, the effective oxygen concentration in the vessel remains unknown. In order to err on the side of caution, more nitrogen is pumped into the tank, leading to possibly unnecessary inerting costs. The preferred situation is to measure and control oxygen within an optimal range, conserving inert gas usage while at the same time maintaining a margin of safety (Figure 1).

Figure 1: Oxygen concentration control during inerting.

Historically, extractive measurement systems (paramagnetic or fuel-cell) have been used for such inerting tasks. While accuracy and long-term performance of such technologies have been up to most users’ expectations, their reliability and cost of ownership are being questioned. Due to the fact that in these technologies the measurement cell must be protected from the process conditions by a gas sampling and conditioning system, their overall performance and reliability are heavily dependent on the performance and reliability of the sampling system. Any failure in the sampling equipment leads to unexpected downtimes and loss of process uptime.

Therefore, there has been an increase in demand for cost-efficient, oxygen measurement solutions for tank blanketing and inerting. Recently, polarographic (amperometric) solutions have gained interest because of their ability to withstand a wide range of gas compositions, to measure in-situ and because of their low cost of ownership.

Polarographic technology and sensor design

Polarographic sensors have been in use for decades for the measurement of dissolved oxygen in blood analysis, biotechnology fermentation, brewing processes, wastewater treatment and many other applications. These sensors are essentially electrochemical cells that measure the partial pressure of oxygen in the liquid or gas phase. This partial pressure is directly proportional to the current produced by the following reduction-oxidation (redox) reactions:

While at the reference the following reaction occurs: A cross-sectional view of a polarographic sensor is shown in Figure 2. The current that flows between the anode and the cathode is measured and is proportional to the amount of oxygen present in the process. During construction, the anode/cathode assembly is placed into an electrolytic solution in order to complete the current loop. The polarisation voltage that is necessary for a stable and linear relationship between the oxygen partial pressure and the resulting current is applied between a silver/silver chloride reference electrode and the platinum cathode.

Figure 2: Functional overview of a polarographic oxygen sensor.

Anode, cathode, reference electrode and electrolyte are hermetically sealed from the process by a membrane. This membrane is selectively permeable to gases and volatile substances, but not to ions, and prevents detrimental oxidising or reducing agents influencing the measurement of oxygen. For improved pressure resistance, the composite membrane is reinforced by an embedded steel mesh. The outer layer of the membrane is made of dirt-repellent PTFE and is designed to withstand a wide range of process conditions, including moisture and dust.

The current produced during operation depends on several parameters:

The produced current is in the range from 0 to 100 nA for measurements between an oxygen-free environment and air. The practical temperature measurement range for polarographic sensors is from 0 to 80°C. Above this temperature, the conversion rate of the electrochemical reaction becomes unstable due to changes in reference potential and residual current, as well as variations in membrane permeability. If the polarisation voltage was to be swept over the zero to 1.5 V range, the resulting sensor current response would be an S-shaped line called a polarogram. The polarogram always shows a flat section in the middle where small changes in the polarisation voltage do not affect the resulting current. This is why the polarisation voltage is usually chosen to be in the middle of the flat section. This plateau results from the current created by O2-diffusion through the sensor membrane being limited at the polarisation voltage. This effect is crucial for the linearity of the measurement signal.

Figure 3: Polarogram showing variation in current against O2 partial pressure for nitrogen, air and oxygen.

Polarisation voltages for industrial-grade polarographic sensors are in the range of 500 to 700 mV. Having now selected the polarisation voltage, what is the variation of the current in response to changes in the oxygen partial pressure in the process? In Figure 3, the polarogram taken in 100% nitrogen shows that the flat section of the curve is at 0 nA; the polarogram for air has a plateau at approximately 70 nA; and for 100% oxygen this value climbs to 350 nA. Figure 4 illustrates that the plot of sensor output current as a response to oxygen concentration shows a nearly 100% linear relationship over the whole measurement range. This demonstrates two important features of polarographic sensors:

Figure 4: Plot of sensor output current against oxygen concentration.

• Firstly, the high dynamic range of polarographic sensors opens many application possibilities. Not only can polarographic sensors measure accurately at very low or medium concentrations that are typically used in blanketing and inerting applications, they can also measure in air. Therefore, they can be used as gas detectors for oxygen measurement in potentially hazardous areas for personnel to work in. This feature is especially valuable in applications such as glove boxes, where the area that is usually inerted must be frequently accessed by technicians.
• Secondly, the calibration of polarographic sensors can be performed without the use of specially produced calibration gases of certified composition. Due to its high linearity, the sensors can be calibrated in air at about 20-20.95% O₂ (depending on the humidity of the calibration air) and achieve highly accurate measurements at much lower oxygen concentrations. In fact, as we will see below, air is the only calibration gas that is needed for polarographic sensors as their zero point remains mainly unaffected.

Maintenance aspects

Polarographic sensors for industrial applications have been the gold standard in applications such as biotech fermentation control or filler line monitoring in breweries. Such environments require that each item of measuring equipment comply with the highest standards of dependability, ruggedness and ease of use. Next to measurement accuracy requirements as low as 2.5 ppb dissolved oxygen (50 ppm gaseous oxygen), emphasis over recent years has included the serviceability and predictability of service operations such as component exchange and calibration.

Typically, polarographic sensors need membrane and electrolyte exchange at regular intervals for sustained performance and reliability. Although the electrolyte itself is not in contact with the process media, it is recommended to exchange it, together with the membrane, due to the dissolution of silver chloride into the electrolyte and the fact that when operated at high temperatures, the electrolyte may evaporate outside of the membrane cartridge and impurities will be in higher concentration. Membrane exchange can be performed at the measurement point by a non-specialist and requires less than five minutes. For simple maintenance, the membrane itself is pre-installed into a membrane cartridge that also serves as a container for the electrolyte. The frequency of the membrane replacement depends on the process conditions, but in typical inerting applications membranes should be replaced at 3-9 month intervals.

Intelligent diagnostics

The latest developments in process analytical equipment aim to improve the level of diagnostic information delivered from the sensor in order to better estimate, in real time, the sensor’s residual operational time and the time to the next calibration. To do this, polarographic sensors have been equipped with on-board electronics with analog-to-digital conversion, diagnostics software and data storage capabilities to continuously monitor the sensor’s state and warn the user pre-emptively when early signs of sensor failure are detected. For example, if the electrolyte level falls below a pre-set tolerance, this is sensed by the on-board intelligent system and an alarm is triggered so that the user can schedule a maintenance call before sensor performance is impacted.

Polarographic sensors can be mounted in-situ and therefore are able to measure reliably within the inerted zone itself. As a consequence, they must be used in conjunction with a safe process adaption that allows extraction without process interruption, providing insulation between the inerted zone and the environment. Retractable housings with an integrated flushing chamber for automated sensor cleaning and calibration are available.

In-situ measurement

Thanks to their moisture- and dust-resistant membranes combined with a rugged sensor design for process conditions (some suppliers offer a wide selection of alloys and O-ring materials for improved corrosion resistance), polarographic sensors do not need a gas sampling or conditioning system, as paramagnetic systems do. In day-to-day use, sampling and conditioning systems frequently cause problems, such as line clogging, condensation and air leaks, and require regular off-line maintenance. Polarographic sensors are considered to be more dependable and easier to maintain than extractive systems. When an ROI calculation is performed, payback times are usually very short due to the reduced system complexity and the fast and easy maintenance.

Regulatory compliance

Inerting processes require a high level of safety compliance due to the risk of explosion. When built into the process, sensors must provide the same level of safety as all other components such as flow meters, valves and level detectors. Today’s polarographic-based measurement systems routinely offer ATEX zone 0/1, FM and CSA Class 1 Div I and IECEx approval for the sensor and the housing as well as the analyser. For pharmaceutical applications, attention must also be paid to FDA-compatible materials and sealants.

Conclusion

Polarographic sensors are highly suitable for use in applications such as inerting and blanketing in tanks and vessels. Safety in the workplace can be increased as oxygen is part of the fire triangle and therefore its presence should be measured in-situ. Polarographic sensors offer several key advantages in terms of maintenance effort, predictability of sensor behaviour and ability to measure in-situ.