CO2 Solubility in Beer: Does Quantifying it Precisely Matter?

My recent posts have prompted a couple of interesting questions about the solubility of dissolved CO2 in beer, tea and soda, so that’s what we’ll focus on today, keeping in mind that the points we cover can be applied to the solubility of O2 or N2 as well.

First, let’s define exactly what we mean by solubility. To understand solubility, we need to understand another term: partial pressure. The partial pressure of a gas is the percentage of gas present in any particular environment, relative to any other gasses in that environment.

In general, dissolved gasses are described in units of weight per volume. But in North America, when measuring in liquids like beer or soft drinks or even water, we use units of volume per volume. Volume per volume as it relates to liquids is not a very scientific unit, but it’s still the standard. (In Europe you’ll see it expressed as weight per weight – i.e., grams per kilogram.) So when we talk about solubility in beer, we are discussing the mass of a gas that will dissolve in a known volume of that beer, even though the units specify “volume per volume.”

The things that affect this are the liquid temperature, the liquid type, and the type of gas. For example, CO2 is exceptionally soluble in water, O2 less so, and N2 even less so. And the gravity (density) of a liquid will also have a big impact, as we can see by examining the graph below showing the solubility curves of dissolved CO2 in water versus a low-gravity beer. This graph also shows our third parameter, temperature. The higher the gravity, the lower the solubility. Likewise, a higher temperature correlates with lower solubility.


Now let’s look at our questions. One customer wanted to validate multiple CO2 analyzers with a single dissolved CO2 liquid source, but was concerned that if he used water as the test liquid it wouldn’t relate back to beer. He also wondered if his analyzers should be calibrated individually (customized calibration) for the different gravities of different beers. And then I had a soft drink customer who was wondering about using data regarding the solubility of CO2 in beer, and whether or not it could translate to measuring CO2 in their soft drink products.

These questions may seem very different, but my answer to all of them was the same. The main point we needed to know was whether or not the absolute concentration of CO2 – as opposed to partial pressure — in whatever liquid we were measuring was important. Both customers were measuring pressure and temperature to calculate their CO2 concentration. Both customers were happy with their products, but were concerned that not knowing the absolute concentration might matter to the quality of the product.

I don’t think that it is necessary to know the absolute concentration of the CO2. In fact, I fear that if you knew the absolute concentration for every beer type and tried to monitor based on that, it could leave the potential for too many mistakes and could ultimately be more harmful than helpful. That’s because consistency in measurement technique is more important than absolutely quantifying the amount of CO2 in the liquid.

Regardless of the type of sample, be it carbonated water, low-gravity lager, or high-gravity ale, most CO2 monitors will give you a concentration based on one solubility at equal partial pressures of the product. While it may not be the quantitatively correct concentration, it does represent a consistent percentage of the CO2 in the sample. If your high-gravity beer appears to need more CO2 to get the same sensory effect as your low-gravity beer, it may be because you need a higher pressure of CO2 to get the same carbonation level.

My final thought is to keep it simple when measuring CO2 concentration. What you really need to trust is your sensory analysis and understand that if the measured CO2 value needsto be different on some products it may be because their gravity and other parameters are different.



Tips for Getting Accurate Portable Dissolved Gas Measurements


My last post was about getting accurate in-line gas measurements. There is a link to that post here. Today I’m going to follow-up with some tips on how to achieve accurate portable results.

Portable measurements are a tad more forgiving and easier to accomplish than in-line measurements. You don’t have as many sensor placement restrictions and you can measure in vessels, as opposed to using an in-line probe that needs flow. Here are the main things to remember:

  1. Just as with in-line probes, all of the gasses in your beer — the CO2 and O2 that will be in contact with the sensors — need to be clear and in solution. If there is degassing in the flow chamber then you will probably get lower than expected results.
  2. Even though you are not measuring in-line, you still need to deal with flow in the flow chamber of your instrument. Portable electrochemical oxygen sensors have specified flow rates. If you deliver the sample too slowly or too quickly, the readings will be underestimated. Slow flow will not feed sufficient product to the sensor to satisfy the requirements of the sensor’s electrochemistry. High flow may result in product degassing.
  3. Optical sensors are much more forgiving and require less flow than electrochemical sensors. The instruments used with most optical probes are designed to free you from worry about flow as long as there’s sufficient backpressure on the systems to keep them from degassing.

My final thought is to remember that different analyzers have different functions, and you may need more than one to instrument to get full control over your dissolved gasses. You can’t go wrong with a good electrochemical analyzer, but optical sensors give you a different type of flexibility, and might be the next step in a well-rounded instrument collection.

Tips for Getting Accurate In-Line Dissolved Gas Measurements


The most crucial thing to remember when measuring the content of any gas in a process pipe is to make sure that all of the gas you want to measure is in solution. Here are the main points:

  1. There can be no CO2 bubbles. The liquid must be clear. If there are CO2 bubbles then your dO2 or CO2 results will be lower than your actual value.
  2. Measurement points need to be as far as possible from any source of cavitation, turbulence, or gas injection. When measuring after a carbonator or a wort oxygenation source, place your probes as far as you can from the injection point.
  3. Measurements in tanks and in process lines are not the same. What you measure in a process line should reflect what is in the pipe where the sensor is located. Once your beer is in the tank, several other factors may come into play and cause your instruments to show either higher or lower values than what you see in your pipe.

Lets talk for a minute about this last point. For example, how could a value go lower? In the case of wort or beer, you could lose gas in the tank (O2 for wort, CO2 for beer) if the head pressure on the tank is less than the pressure in the process pipe. It’s not always possible to keep some gas from escaping, so it’s good to be aware of this and not immediately assume there’s a problem with your instruments.

Here are some things we recommend with regard to in-line measurements.

  • Never place a sensor in a descending pipe. Liquids in descending pipes can degas if the line does not have adequate pressure. If this happens it pulls a vacuum on the probe, causing it to read too low.
  • Place sensors in horizontal or ascending pipes.
  • All sensors should be as far from pipe bends as possible, but never closer than five pipe diameters from any elbow.
  • Sensors should never be on the suction side of a pipe.
  • Place probes as far from pumps as possible. Placing them right after pumps can cause false low readings.
  • When placing a sensor before a rotary filler, keep it out of the drip line of the filler.

My final thought about in-line measuring is that you’ll always do best with an installation if you think in terms of maximizing the ability of all of the gases to stay in the liquid.

%d bloggers like this: