Dissolved Oxygen in Beer: How It Compares to Total Package Oxygen

When it comes to questions about oxygen in beer, I think the one I’m asked most often is, “What is the difference between dissolved oxygen and total package oxygen (TPO)?”  The main source of this confusion is that when measuring O2 in packages, the O2 in the headspace is often overlooked. If you don’t take headspace oxygen into account, then you are measuring a partial concentration, period. So let’s talk about the differences and what each one tells you.

A significant number of craft brewers have a dissolved oxygen (dO2) analyzer they use to measure the dO2 content of their beer in process. The most common point of measurement is the finished beer tank. The beer in a finishing tank will have O2 pickup from the empty vessel and from the filtration process, plus it will pickup more O2 as it goes through packaging.

Once the beer is packaged, however (assuming good packaging,) rapid O2 pickup from outside sources all but stops. So what can we tell about how much oxygen actually made it into the package?  It is not a simple matter of measuring the O2 in the beer.  The package must be shaken to equilibrate the oxygen in the beer and the headspace before the 02 in the beer is measured, and that number must then be used to calculate your TPO. Let’s think about what it is possible to measure and what each thing tells you.

Package dO2 –

The easiest measurement to take on packaged beer is the dO2 of a package just off the filler without shaking the beer. It is important to measure as quickly as possible, so the product does not “consume” the oxygen in the beer. (Residual or live yeast may be hungry, plus oxidation by trace metals, etc.) In some packages there is a measurable difference within five minutes and in other packages the rate of oxygen consumption takes significantly longer, sometimes hours. It is always best to measure as quickly as possible.

This unshaken package measurement represents the combination of the dO2 of the beer at the base of the filler and the oxygen pickup of the filler. Oxygen picked up at the filler can be quite variable. Most fillers run at about 25 to 50 percent deviation, but in some cases it can be up to 100 percent deviation. The best way to measure the percent deviation is to determine the dO2 at the base of the filler and then measure six to ten packages and determine the variation of each package as compared to the average of all the containers. But remember: this measurement only tells you what is in the liquid. When measuring unshaken packages, any gas in the headspace is left uncounted.

Shaken Package dO2 –

When you shake a package of beer so that the partial pressure of the oxygen in the liquid is equal to the partial pressure in the headspace, it changes the characteristics of the oxygen partitioning in the package. If most of the oxygen in the package is locked in the liquid, then shaking the container will move the O­2 from the liquid to the headspace until equilibrium is reached.

So, you have measured the dO2 and then shaken the package. Now what do you do with the data? If you really want to quantify the TPO of the package you have to take into account the headspace oxygen. To do this accurately you need to know the headspace volume and the package temperature.

Total Package Oxygen –

When using the dissolved oxygen measurement, the TPO can only be calculated from a shaken package. To do this calculation you also need to know the headspace volume, liquid volume and the package temperature. The temperature and the headspace volume are critical values and small inaccuracies can alter the results significantly, but the liquid volume may be estimated by using the average fill volume. Once you have your figures, then you can use a TPO calculator to determine the concentration from your initial DO2 measurements.

My final thought is to not skimp on how much you shake the packages. Cold containers should be shaken for five minutes and room temperature cans or bottles need about three minutes. If you’d like a copy of a TPO calculator built into an Excel spreadsheet, then please click here to request one.



Beer Sensory – Creating a Basic Plan for Beer Storage


I was talking to a brewer about TPO and he asked if minimizing oxygen pickup at the filler might help extend his shelf life and limit sensory changes as his beer ages. He was mainly concerned with controlling beer in the distribution channel, but when I quizzed him about how he calculates the pull dates for his product, he admitted it was mostly a guess. He said there were significant differences in his beer after it had been sitting on a shelf in a store, but he hadn’t tried to mimic that aging and study the changes in a controlled setting.

Oxygen pickup is definitely the most critical component in the development of off-flavors in beer, but many other things can be a factor too, so careful studies of the way your beer ages can help you make informed decisions. Here are some ideas that might help.

First, if you haven’t already done it, then consider getting formal training in sensory. You no-doubt know your product very well, but learning the standard terms and techniques of sensory can be a lot of fun and can give you a good foundation to stand on as you start keeping records. I was with a company that hired a sensory specialist to come in and work with us over the course of a year and it was some of the most valuable training I’ve ever received.

Next, set up a system for comparing your stored beer in different environments over time. I once worked with a closure manufacturer to develop oxygen-scavenging polymers for crown closures. We established a taste panel to evaluate everything from how the polymer affected beer taste to how well the crowns worked to minimize oxidative flavor formation. Depending upon what we wanted to learn, we had different methods for beer storage and comparison.

Our most basic protocol called for storing a set quantity of the same beer (all from one run, one filler) in different ways, typically one case stored cold, one at room temperature, and another upside-down at room temperature.

The cold beer was the control, since sensory changes are accelerated by heat. The room temperature samples were to see what changes developed in non-harsh conditions. And the upside-down bottles at room temperature were to minimize effects of oxygen ingress through the closure, so we could see how well they worked. This can be done with canned beer too, but you shouldn’t need to store the beer upside down unless you are concerned with poor seaming.

Next, you can see how your beer responds to harsh conditions and other variables. Punish it by exposing it to heat, like might happen in a hot delivery truck. Or maybe you’re concerned about closure polymers affecting taste. To analyze this, gently heat lined caps on a hot plate and remove the liner material with tweezers as soon as it gets soft, then put them in the beer, fob it, and re-seal. When you taste it days or weeks out, you should be able to detect things like flavor scalping or off-flavor pickup.

Once you’ve established a system for storing and evaluating your beer under different conditions, make sure you check it on a regular basis. It’s probably not necessary to do it every day, but in addition to tracking sensory changes over time and in different conditions, it can be helpful if there is something in your process that changed, like a new source of ingredients or filling issues.

My final thought is to get to know your beer. Compare it with both beer packaged at the same time and with beer that is older or younger. By understanding sensory changing factors, you may be able to comfortably lengthen the time you beer is in the distribution channel by weeks or even months.

Creating a TPO Validation Standard


The most difficult thing about using Total Package Oxygen instrumentation is creating a proper validation standard. In other words, you need a solution with a known concentration of oxygen in order to validate your TPO instrumentation, but how do you get that?

Since the concept of TPO was first published widely, the most referenced paper on the subject is the 1985 article in Brauwelt by Carlos Vilachá and Klaus Uhlig, “The Measurement of Low Levels of Oxygen in Bottled Beer.” It covers the best calculation to get TPO, but doesn’t talk about how to validate the results.

Your TPO validation standard must be repeatable (for statistical purposes you’ll be using multiple packages of your standard,) but it’s not easy to repeatedly put a known dO2 concentration in a package and not have the oxygen content change over time. You can package water, but microbes or can corrosion can decrease oxygen after awhile, plus it’s hard to find a can filler capable of getting the oxygen levels in the headspace low enough to reflect the values found in most freshly filled beer packages. (That’s because when beer foams, it displaces the oxygen out of the headspace, and water doesn’t foam.)

So the best solution I know is to use old pasteurized beer and inject the cans with a known concentration of air. The method is fairly simple, but there are a few tricks that make getting accurate recovery levels possible. Here’s how it works:

  • Use room temperature beer that is at least 30 days old and preferably pasteurized, but not can-conditioned.
  • Place a sticky-back septum (most are pretty small, less than an inch in diameter) on the can and hold the septum on the can with a large hose clamp. I recommend putting the septum on the upper edge of the can, where the can begins to neck into the seam, because the can is stronger in that location and will flex less. Tighten the hose clamp so you can still rotate it if you use some force. Make sure to center one of the hose clamp holes over your septum.
  • Using a 500 micro liter gas-tight syringe, pull 100 micro liters of deaerated water into the syringe and then add an additional 250 to 350 micro liters of air to the contents. Make sure the water stays touching the plunger.
  • With the syringe, pierce through the septum and into the can and inject the air and water into the container. Then rotate the hose clamp so it blocks the hole where you injected the air and shake the can.
  • After shaking the can for three minutes, measure the dO2 and calculate the TPO. If you have a TPO analyzer, the TPO result from the instrument is the proper value to use for comparisons.
  • Next, calculate the mass of the oxygen in the air you used (based on whatever volume you used.) This will give you a value to compare to your TPO.

Hach sells a kit that contains all the pieces you need, including detailed instructions for the procedure, as Part Number DG33373. You can also get septums, syringes, and needles from distributers like VWR and the hose clamp from your local hardware store. Next week I’ll post a step-by-step pictorial and a way for you to get the calculations.

My final thought is two-fold. First, make sure you test your ability to do this procedure on multiple packages and get statistics on your percent recovery. If you’re getting great repeatability but not good accuracy, it may be the calibration of your instrument. Second, some TPO instrumentation will not get valid or repeatable values due to foam in the headspace. While these instruments work very well on water samples, foam messes them up and can affect their overall accuracy.



%d bloggers like this: