Dissolved oxygen refers to the amount of oxygen that has been dissolved into a water source, which could be anything from the ocean to a bioreactor.
The necessity of dissolved oxygen measurement depends on what application you’re using it for. If your company is involved in aquaculture in some regard, low dissolved oxygen levels will result in fish suffocating. As for sewage treatment, bacteria is necessary for decomposing the solid waste. Lower DO levels result in bacteria dying off, which means that the decomposition of waste will stop. No matter the kind of industrial or aquaculture application you’re involved in, dissolved oxygen measurement is highly important. Being able to react quickly to changes in DO levels may allow you to save ample amounts of time and money.
The various types of sensors that can be used to help with the measurement of dissolved oxygen include galvanic DO sensors and optical DO sensors, both of which provide precise results and allow for quick calibration.
In this article we discuss the importance of aeration, the oxygen mass transfer coefficient (kLa), working of DO sensors within a bioreactor.
Delivering oxygen to cells
One of the most important functions of bioreactors is providing dissolved oxygen to cells continuously through a process called aeration.
Aeration in the bioreactor typically occurs when:
Oxygen diffuses through overlay to the cell culture medium interface.
Oxygen from the spargers dissolves in the cell culture through convection with the help of agitation.
Agitation disperses the oxygen bubbles and promotes mass transfer of the gas bubbles through the gas‑liquid (cell culture medium) interface. The rate of oxygen transfer (OTR) from gas to liquid interface is a function of physicochemical properties of the cell culture medium, the geometrical parameters of the bioreactor, and presence of cells.
Diagram of a gas bubble in liquid, showing how the bubble is released, solubilized, and transferred to a cell.
Due to its low solubility in liquid phase and increasing metabolic consumption by the cells with time, oxygen is supplied continuously to the cell culture. Oxygen supply is carefully controlled for optimal cell growth by manipulating bioreactor parameters.
During batch cell culture, OUR (or OTR) is initially low during the lag phase, where cells are self‑synthesizing and there is little gain of cell density. As cell density increases during the exponential phase, OUR increases until OTR becomes a limiting rate, as determined by the mass transfer of oxygen into the bulk liquid.
Phases of cell growth
The OTR and OUR rates are correlated by the oxygen mass transfer coefficient, kLa. Therefore, the OTR, through its correlation to kLa, defines a theoretical maximum cell density that could be achieved in cell culture.
Equation for kLa
Imagine a gas bubble in liquid. For this discussion the gas bubble contains oxygen, and the liquid is the liquid in a bioreactor. kLa can be represented by the following equation:
kLa = kL × a
Where kLa is the mass transfer coefficient from the gas to liquid phase, given in sec‑1
kL = liquid side mass transfer coefficient (resistance in gas side film can be neglected)
a = bubble surface (available for diffusion)
Key variables that impact kLa values
Any change to process and engineering parameters or to physical characteristics will have an impact on kLa and should be considered when evaluating bioreactor platforms and performing scaling calculations.
Here are four key variables that can affect kLa values:
1. Gas bubble size
When gas bubble size decreases, surface area and gas residency time increases, causing bubbles to stay in the culture longer. Thus, there is a greater opportunity for oxygen to release mass transfer into the cell culture medium. An increase in this oxygen residence time improves kLa.
2. Mixing
In a bioreactor, mixing is used to eliminate gradients of concentration (cell, gas, medium, and nutrient), temperature, and other properties. Mixing time is widely used to characterize mixing efficiency in a bioreactor. Mixing efficiency is one of the most significant factors affecting both performance and scale‑up in a bioreactor.
Gas bubble size and residency time are highly dependent upon three mixing conditions: impeller type, speed, and location(s). kLa values generally increase as tip speed increases. However, tip speed is proportional to shear forces that can lead to cell death. Bioreactors, therefore, are designed with different impeller types, combinations, and locations to achieve target kLa values without creating these shear forces.
Generally, kLa values are closely associated with impeller design, with Rushton typically higher than paddle, which is typically higher than marine and pitched impeller.
3. Air flow rate
Higher oxygen availability drives kLa increases. Increasing oxygen supply to a bioreactor drives this availability and can be controlled by modifying concentration (air vs O2 enrichment) and volumetric flow. Although high kLa values are desirable, it is important to consider the actual operating conditions and implications to cell viability and associated process costs.
For example, high air flow rates can cause cell damage due to shear forces. Excessive foam might also be generated, requiring a high concentration of antifoam that could hinder downstream processing. Additionally, higher air flow rates require a larger exhaust filter area, driving consumable cost increases.
4. Properties of the liquid or medium
During cell culture, small bubbles collide and coalesce to form larger bubbles, decreasing surface area (a) and subsequently kLa. Be aware of reported kLa values in which high salt concentrations are used, because this can prevent bubble coalescing. Antifoaming agents are used to influence surface tension, resulting in reduced bubble coalescence and foaming.
However, this principle does not always lead to increases in OTR wherein antifoam also reduces bubble mobility, which subsequently reduces the kLa
Other factors that affect cell culture kLa
5. Measurement method
Several different methods are used. Most commonly the nitrogen stripping (i.e., gassing‑out) method is employed.
When scaling a process within the same platform, it is important to use an identical method for measuring kLa. kLa, when combined with process engineering parameters (i.e., tip speed, power input), can be used to experimentally determine the cell density in a larger bioreactor compared with a smaller bioreactor.
6. Temperature
Increasing temperatures inversely affects both the volumetric mass transfer coefficient and oxygen solubility in culture medium. Oxygen solubility in pure water falls with increasing temperature (i.e., ‑0.5 × 10‑3 kg/m‑3 between 35°C and 30°C.
Therefore, it is important to note the temperature conditions from vendor‑supplied characterization data.
7. Sparger characteristics
kLa values will vary widely with sparger characteristics, including number, pore size, and surface area, because these factors affect bubble size, gas velocity, and flow rates.
Measuring dissolved oxygen concentration
Commercially available dissolved oxygen sensors typically fall into 3 categories:
Galvanic dissolved oxygen sensors
Polarographic dissolved oxygen sensors
Optical dissolved oxygen sensors
Each type of dissolved oxygen sensor has a slightly different working principle. Therefore, each dissolved oxygen sensor type has advantages and disadvantages depending on the water measurement application where it will be used.
Electrochemical dissolved oxygen sensor working principle:
Both galvanic DO sensors and polarographic DO sensors are types of electrochemical dissolved oxygen sensors. In an electrochemical DO sensor, dissolved oxygen diffuses from the sample across an oxygen permeable membrane and into the sensor. Once inside the sensor, the oxygen undergoes a chemical reduction reaction, which produces an electrical signal. This signal can be read by a dissolved oxygen instrument.
Polarographic vs. galvanic DO sensors:
The difference between a galvanic DO sensor and a polarographic DO sensor is that a polarographic DO sensor requires a constant voltage to be applied to it. It must be polarized. By contrast, a galvanic DO sensor is self-polarizing due to the material properties of the anode (zinc or lead) and cathode (silver). This means that while galvanic DO sensors can be used immediately after calibration, polarographic sensors require a 5-15 minute warm up time.
Electrochemical DO Sensor
Optical dissolved oxygen sensor working principle:
An optical dissolved oxygen sensor does not have an anode or cathode, and oxygen is not reduced during measurement. Instead, the sensor cap contains a luminescent dye, which glows red when exposed to blue light. Oxygen interferes with the luminescent properties of the dye, an effect called “quenching.” A photodiode compares the “quenched” luminescence to a reference reading, allowing the calculation of dissolved oxygen concentration in water.
Optical DO sensor
Optical vs. galvanic DO sensors:
Both optical dissolved oxygen measurement and galvanic dissolved oxygen measurement have advantages and advantages. The good news is that both technologies offer a similar level of accuracy when measuring dissolved oxygen concentration. This holds true across a wide range of measurement values: field tests have shown similar results for optical and galvanic DO sensors from ~1mg/L up to 14mg/L.
One point of differentiation between optical and galvanic DO sensors is that galvanic DO sensors exhibit flow dependence. This means that a minimum inflow velocity (2 inches/sec for Sensorex models) is required to maintain measurement accuracy. Optical DO sensors require no minimum inflow velocity.
Some sample constituents may interfere with measurement accuracy. Hydrogen sulfide, for example, a compound found in wastewater, lake bottoms, and wetlands, can permeate the galvanic sensor membrane. An optical dissolved oxygen sensor would make a better choice in these environments, as these sensors are not susceptible to interference by H2S.
One advantage of galvanic DO sensors over optical DO sensors is that galvanic DO sensors have a faster response time. Galvanic DO sensors respond 2-5x faster than optical DO sensors depending on the membrane material. This limitation of optical DO sensors is more cumbersome in applications where a high number of sample measurements will be taken. Response time is typically not a limiting factor when choosing a DO sensor for continuous monitoring applications.
Comparison of polarographic, galvanic, and optical DO sensors:
The table below summarizes advantages and disadvantages of the 3 primary methods for measuring dissolved oxygen concentration in water:
0 Comments
Comments here