The IBU Assay

The spectrophotometer is among a small set of tools and equipment that are essential for a quality control lab to be adequately productive and accurate in a brewery setting. The reason for this is the BU assay which, apart from the HPLC, offers the best and most direct way to analytically measure the bitterness of beer that is available to brewers today. Relative to HPLC, the BU assay lacks precision, accuracy and sensitivity. But what makes it appealing are a number of things, foremost being the price: a couple thousand dollars can net you a new UV/Vis spectrophotometer and some supplies, whereas an HPLC can be an order of magnitude more expensive. Convenience, simplicity, and to some extent reliability are also among the benefits of this method, since HPLC, indeed chromatography in general, can be fickle and prone to error if rigorous procedures are not followed. Here, we’ll explore various aspects of the BU assay, including its origins and the fundamentals behind it.

The Test:

The Back Story:

The origins of today’s IBU assay started as a way to measure the bitterness of beer using an analytical technique. Most of the early methods attempted to measure the iso-alpha acids by themselves. In 1952, Rigby and Bethune attempted to use a particular method of liquid-liquid extraction (called Countercurrent Distribution, or CCD) to measure the bitterness of beer. This method was a precursor to modern HPLC technologies and involved a series of 100 glass tubes which extracted and back-extracted the sample to the point where various components were separated from each other. When applying this method to beer, they were able to partially separate one of the types of iso-alpha acids from the other two, and they named it iso-cohumulone. When applied to hops, they managed to separate all three alpha acids from each other, and named them: n-humulone, cohumulone, and adhumulone. But this method had one main drawback: it took nearly an entire day to run one sample. Other various methods were developed by other researchers which had their own positives and negatives. Spectrophotometric methods using chloroform were found to extract too much confounding materials. Similar methods using acidified methanol and trimethylpentane gave artificially inflated values relative to the CCD method. In 1955, Moltke and Meilgaard decided to take a different approach by attempting to measure the perceived bitterness rather than the amount of iso-alpha acids. They speculated that some of the bitterness of beer came from non-iso-alpha acid sources and derived a new method that they believed would be a good approximation of the bitterness. They adjusted the wavelength used to 275nm instead of the 255nm that was used in previous methods and measured the trimethylpentane directly instead of using acidified methanol. They knew that this method measured things beyond just the iso’s, but apart from this inflated value they found that it correlated well with perceived bitterness and actual iso-alpha acid concentration. They also admitted that there were other factors at play here, such as pH and protein interactions, but felt that their new method was the best to date. However, the methods which were actually adopted at first differed depending on what part of the world you were from. In the late 1950’s, Europeans were using a modified version of the Moltke Meilgaard method, while North American breweries used a later version of Rigby and Bethune’s system. Disagreements, discussions, and various tests kept researchers busy through the late 1960’s as they attempted to reconcile these differences and agree on a standardized method. Finally, in 1968, after collaborative efforts between Rigby, Bethune, Moltke and Meilgaard, the industry adopted the version we know today as the standard method for finding the International Bitterness Units of beer.

The How-To Version:

The methodology for the test for International Bitterness Units (IBU, or BU) is now essentially the same between two main scientific organizations involved in brewing research: the American Society of Brewing Chemists (ASBC) and the European Brewing Convention (EBC). Both these entities have references tomes of standardized methodology for a great many analytical procedures, from raw materials analysis to the testing of various production parameters and sensory protocols and methods, to name a few. The ASBC Methods of Analysis method Beer (10-C) and the EBC Analytica method 9.8 are both in agreement as to the parameters of the test: a volume of wort or beer is acidified with HCl then shaken for 15 minutes with twice the volume of 2,2,4-trimethylpentane (iso-octane). This shaking should be done on a mechanical wrist-action shaker to ensure repeatability, but it is possible to shake it by hand. It is not recommended. This trimethylpentane organic phase should then be read at 275nm in an ultraviolet spectrophotometer. A conversion of the measured absorbance (x50) gives you the Bitterness Units.

The Colloquial Version:

Essentially, what is happening in the test is a liquid-liquid extraction of the various components of beer that are bitter. The semi-hydrophobic (water-hating) iso-alpha acids and their companions are relatively comfortable being in solution in the wort/beer, that is until you lower the pH of the matrix with some hydrochloric acid. This forces the iso’s to become more hydrophobic, to the point where pretty much all of them would prefer to be in an organic (non-polar) liquid. This is conveniently provided by the trimethylpentane, and the shaking speeds their transfer right along. After the specified shaking time the organic phase is read in the spec at a specific wavelength. 275 nanometers is chosen due to it being the best approximation of the maximum absorbance of the various bitter molecules, each of which actually has its own specific maximum. Overall, it works well enough.

The Deeper Version:

We can look at this test a little deeper by trying to get a better idea of what this test is actually measuring. It’s not just iso-alpha acids that it’s measuring, although they are often the headlining attraction when discussing beer bitterness. In fact, an addition of 1ppm iso-alpha acids only increases the BUs by 0.7. In addition to the iso’s, this test measures the various oxidation products of beta-acids, or “lupulone” more generally. These lupulones are found along side the alpha acids in the lupulin glands of the hop cones and are generally of secondary importance to brewers – but they do have an impact on beer flavor, particularly when they degrade. As hops age their potential affect on the BU assay varies since the BU methodology was originally calibrated on hops that had their own level of oxidation, which was rather high by today’s standards, as well as their own specific alpha/beta acid ratio. Beta-acid oxidation products increase as hops age, thus increasing the measured BU’s. This is not a wholly undesired change since these oxidation products actually have some bitterness. Acting in the opposite direction of this change is the degradation of the iso-alpha acids, which leads to a loss of measured BU’s. Again, this does not really affect the accuracy of the assay since the degradation products of iso-alpha acids and alpha acids do not have much bitterness, so the perceived bitterness will drop along with the BU result as the iso’s are lost. Ultimately, it’s alpha-beta acid ratio in the hops and how it relates to the ratio in the hops originally used to calibrate the test that dictates how the BU’s change as the hops or beer degrade. Hops with more beta acids will contribute more oxidation products during storage than beers with less beta acids. This rise in beta acid oxidation products will inflate the BU results at the same time that the loss of alpha’s and iso’s will deflate the value. Depending on the alpha-beta ratio, the level of BU’s could theoretically be stable. In fact, some research which used hops from a single lot aged at different temperatures showed BUs to be rather stable (13 BUs and 11 BUs from hops stored at -20F and 70F respectively) while the final iso-alpha acid content plummeted (19.8 and 2.9ppm respectively). Most beers brewed today use hops that have been stored far better than the hops used in the 1950’s and 60’s. As such, iso-alpha acids tend to be at higher levels than the measured BU’s, for fresh beers. As beers age however, iso-alpha acid levels drop much faster than BU’s do, so they eventually converge to a similar level.

Now believe it or not, this still doesn’t tell the whole story of Bitterness Units, but 1500 words is about all I can do at the moment. I may explain more next time about what I’ve found in my research and production experiences. And I just realized that there really isn’t any sensory science in this article…

Source: Peacock, V., The International Bitterness Unit, its Creation and What it Measures., Hop Flavor and Aroma, Proceedings of the 1st International Brewers Symposium, ASBC/MBAA, p 157-166.


13 responses to “The IBU Assay

  1. Great read! Fine with me that you don’t always talk sensory science. I love learning about everything that happens in the brewery lab. Thanks!

  2. Miguel Angel Martínez Trejo

    Hi there! great article! I work in a brewery, and it’s amazing your interest in this topics, I’m searching for IBUS information… specifically WHY TRIMETHYLPENTANE and not other solvent ¿? do you have some information about that ¿?

    See you! Kind regards.

  3. Pingback: Purpose of Corn in Flanders recipes? - Page 3 - Home Brew Forums

  4. Pingback: Best Online Colleges: 6 Beer Terms Every Beer Expert Should Know

  5. I used to work at a brewery (with the hopes of becoming the microbiologist) and they never did any sort of testing besides taste. Now I work as R&D for a company called StellarNet that makes miniature spectrometers. It is funny how related everything in the world is. I really enjoyed this article, especially because I am in the beginning stages of IBU analysis for my homebrews (luckily, I work for a company that makes the would-be limiting factor of my project). I had a friend who was brewing his own beer and boasting a specific IBU content. When I asked him how he knew the IBU’s of his beer, he told me that he knows how many hops he put into the beer; how unscientific!? Now I am going to try to convince our lead software engineer to include an IBU macro to our software. Technically the software already has a chemometrics engine but I would much prefer a plug and play system.

  6. Pingback: Boley Brew Blog » Blog Archive » Quality Analysis – AHS Oaked Imperial Whiskey Stout

  7. I’d like to give this a try. Can you be more specific about the acidification process? Presumably a small volume of concentrated HCl is preferable so that the original sample is not significantly diluted. Thanks

    • Well, actually any dilution of the sample shouldn’t have much of an effect, so long as the pH doesn’t shift too much. The main things to keep precise are the volume of the wort/beer in the assay, and the volume of the organic solvent. As long as those are correct, you can fudge the rest a fair bit I would guess. The process is effectively a mass-transfer of bitter substances from the beer into the organic phase, so as long as your volumes of beer and solvent are good, you should be fine. The reason for the acid is to drop the pH so that the bitter substances prefer to be in the organic solvent rather than the aqueous beer. If you are off a little bit with the acid addition, it should make little difference.

      10ml wort/beer into centrifuge tube (w/cap), add 1ml 3N HCl, then 20ml trimethylpentane, shake for 15 minutes (mechanical, ideally). You can add 1 drop of octanol to keep foam down if you have a problem with that (not required).

      After shaking, read the absorbance at 275nm, multiply result by 50 to get value in IBU’s. Be sure to blank (zero) the spectrophotometer with pure trimethylpentane before taking readings, and to only use a QUARTZ cuvette of 1cm in size (important!).

      If volume is an issue, you can cut all the above volumes in half with no problems.

  8. Pingback: Lo amargo (IBU): percepción y costumbre | Radical OH

  9. I have a B&L SPEC 20 photometer that goes between 340 -660. Is there any way I can modify this to get to 275?

    • I seriously doubt you’d be able to modify the device to read that low. You’d need a new bulb, for one (don’t know if the voltage supplied to the bulb would be suitable), and you’d probably need to alter some of the interface or whatever to allow the meter to be able to be set that low. Ultimately, I give it a very low chance of success, but I don’t know for sure.

  10. Sorry if I’ve this piece of info, but what chemical standard do people use to calibrate their instrument and method?

    • The only “calibration” that is needed in the typical IBU assay is to first blank the instrument with a cuvette of acidified trimethylpentane/isooctane. After that, read samples as normal.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s