IBUs and the SMPH Model

Introduction
When I first started brewing, the software I used had three options for predicting IBUs: Tinseth, Rager, and Garetz. The word “Tinseth” had a nice sound to it, so I chose that one. I was quite happy with that option until I became interested in flameout additions and hop stands, where the Tinseth formula predicts zero IBUs. Then I found out that I could get IBUs professionally measured at a very reasonable cost. So I did one small experiment to measure IBUs in finished beer with and without a hop stand. And then another experiment, and then another. Just when I thought I could predict IBUs reasonably well, I’d get results that challenged my assumptions. I wrote detailed blog posts about almost all of my experiments, so that anyone can (hopefully) replicate my findings. More than seven years and well over 300 measured IBU values after my first experiment, I put the finishing touches on a new model for predicting IBUs. The purpose of this post is not to go into the gory details, but to give an overview of the model’s higher-level concepts and to address some common misconceptions about the IBU. I also give test results that compare this model with four other IBU models on 18 different beers ranging from 20 to 70 IBUs.

This new model, called SMPH, is available at https://jphosom.github.io/alchemyoverlord/. Even if you don’t use it for recipe planning, I encourage you to play around with it to see how different brewing conditions can yield very different (or sometimes not so different) IBU predictions.

It’s important to note that I’ve been pretty obsessive about measuring or estimating volumes, alpha-acid ratings, weights of hops, hop steep times, wort cooling times, pH, and any other factor that seems relevant. While I’m not saying that you need to be this obsessive in your brewing (it’s supposed to be fun, right?), realize that small measurement or estimation errors might have a large impact on predicted IBUs. If your post-boil volume measurement is off by 10%, then your IBU prediction will also be off by 10%. There is also 10% to 15% variation in alpha-acid content within the same bale of hops [Verzele and De Keukeleire, p. 331], and so the AA rating on your package may not be an accurate indicator of the amount of alpha acids that are in your hops. If your predicted IBU value is off by 20%, that might reduce your prediction from 40 to 32 IBUs. If you’re unlucky, all of these measurement errors can add up and make the prediction meaningless; if you’re lucky, they can cancel each other out. Your mileage may vary.

If predicting IBUs is such an imprecise and difficult art form, why bother? Obviously, you don’t need to care about predicting IBUs if you’re happy with the bitterness levels of the beers that you make. Or, if you find that a beer turns out more (or less) bitter than you’d like and you don’t mind brewing it again with a different amount of hops (or adding iso-alpha acid extract, or blending several beers), then you don’t need to worry about it. But, if you find that your first attempt at a beer can sometimes yield a bitterness that isn’t quite right, you may want to get the best prediction of bitterness that you can before brewing. That prediction might still be a bit off, but an in-the-ballpark estimate is still better than no estimate at all. By way of analogy, even though pH test strips aren’t as accurate as a digital pH meter, if you don’t have a pH meter it’s still better to use test strips than to pretend that pH doesn’t matter and ignore it.

IBUs: IAAs and ABCs
The IBU is a measurement of the amount of infrared light absorbed by a sample of processed beer [Thermoscientific; Anon.]. It is often (and incorrectly) reported that one IBU equals one part per million (ppm) of isomerized alpha acids (IAAs). However, as Val Peacock explains, the IBU was developed in the 1950s and 1960s to measure the combination of both IAAs and “auxiliary bittering compounds” (ABCs) [Peacock, pp. 158-161]. The researchers at that time knew that there are bitter substances in beer other than IAAs, and they deliberately included them in the IBU measurement. The IBU combines the concentration of IAAs and ABCs in beer into a single measure of approximate bitterness. The confusion about one IBU equaling one ppm of IAAs has come up because they scaled the IBU measurement so that the two numbers would often be close to each other. However, this rough correspondence only holds under specific brewing circumstances that were common in the 1960s and are less common today. When the IBU was developed, IAAs contributed to about 70% of the IBU value, and ABCs contributed the remaining 30%. The proportion of IAAs contributing to the IBU can change greatly depending on brewing techniques and how well the hops have been stored. In the 18 beers used in testing the SMPH model (described in more detail below), I estimate that IAAs contribute to between 50% and 75% of the IBU. (A West-Coast IPA with lots of late-hop and dry-hop additions has an IAA contribution of 50%, and a more traditional beer with one early and one late addition has an IAA contribution of 75%.) In the data used for finding SMPH parameter values, the estimated IAA contribution ranges from 0% to over 80% of the IBU.

I have found that the largest fraction of ABCs are oxidized alpha acids (oAAs) that are produced when hops are added to hot wort [Algazzali, p. 17]. I estimate that about 10% of the available alpha acids oxidize quickly in boiling wort, producing oAAs. In most beers, the second-largest contributors to ABCs are malt and hop polyphenols, followed by oxidized beta acids. In her Master’s thesis, Christina Hahn (advisor: Tom Shellhammer) notes that “individually, iso-alpha acids and [oxidized alpha acid] concentrations are relatively poor predictors of sensory bitterness, while the sum of iso-alpha acids and [oxidized alpha acids] is almost as good a predictor of sensory bitterness as [the IBU]” [Hahn, p. 48]. She found a strong correlation (R² = 0.86) between sensory bitterness and the IBU, and a strong correlation (R² = 0.80) between sensory bitterness and the combination of IAAs, oAAs, and alcohol (ABV) [Hahn, p. 50]. In short, the concentrations of IAAs and oAAs are, together, very good predictors for both sensory bitterness and the IBU. These findings support the claim that oAAs are the largest component of the auxiliary bittering compounds.

SMPH Model: The Big Picture
The SMPH model was developed to have one key advantage over other IBU models: it separates out the contribution of isomerized alpha acids (IAAs) from auxiliary bittering compounds (ABCs). The conversion of alpha acids to IAAs takes place relatively slowly (e.g. over the course of an hour-long boil), but ABCs are quickly produced or dissolved in the wort. These different time scales mean that IAAs and ABCs should be modeled separately. (The mIBU calculator includes some approximations in this regard, but it is inherently limited in its ability to accurately separate the two.) While IAAs contribute the most to the IBU in “typical” beers (if there is such a thing as a typical beer anymore), ABCs can contribute a significant amount, especially when using hops late in the boil, when using a hop stand, and/or when dry hopping (techniques commonly used in brewing IPAs).

The starting point for development of the SMPH model was an understanding of Val Peacock’s explanation that IBUs are a specific proportion of the concentrations of IAAs and ABCs in beer [Peacock, p. 161], and realizing that Mark Malowicki’s model of the production and degradation of IAAs in boiling wort [Malowicki, p. 27] could be combined with rough estimates of the concentration of each ABC and different loss factors to predict IBUs. At that point there was the skeleton of a model but a lot of missing factors and unknown parameter values. These values were determined (or found to be irrelevant) by controlled experiments in which only the factor in question was varied. Data from those experiments were gradually added to the set of model training data. The process of estimating a loss factor and then minimizing the mean-squared error on the remaining parameters was iterated until the error on a cross-validation set was reduced to an acceptable level.

The SMPH model first makes a prediction of the concentration of IAAs in wort using Malowicki’s model of alpha-acid isomerization. It then estimates of the concentration of each auxiliary bittering compound in wort. The concentrations of IAAs and ABCs are then modified by various factors (described below) occurring during the boil, fermentation, and conditioning. Finally, it uses an equation proposed by Val Peacock [Peacock, p. 161] to convert from these estimated concentrations in beer to a final IBU value.

Like the Garetz model, the SMPH model can account for a large number of factors that influence IBUs. In the SMPH model, that means accounting for the boiling point of water, wort gravity, wort pH, wort clarity (e.g. a careful vorlauf vs. brew-in-a-bag wort collection), form of the hops (whole cones or pellets), hopping rate, hop freshness, krausen loss, flocculation, finings, filtering, and age of the beer. Basically every step in the brewing process seems to have some influence on IBUs.

The SMPH model uses approximations of all of the known factors that might influence IBUs. (Unknown factors are probably still waiting to be discovered.) The goal has not been to precisely quantify each of the myriad factors (I only have one life to live), but to put all of the approximations together into one imperfect but reasonable model. Where even approximations have been difficult to come by, I used over 300 measured IBU values to find the parameter values that give the best fit to the data.

Figure 1 illustrates the different components of the SMPH model. Measured IBU values from finished beer are shown at 10-minute intervals during a 90-minute boil. The green area shows the contribution from isomerized alpha acids (using the Malowicki model), the blue area shows the contribution from oxidized alpha acids, and the red area shows the contribution from malt and hop polyphenols. The SMPH model output is the sum of these contributions. (In this example, the hops were well preserved and so the contribution from oxidized beta acids is negligible.)

Figure 1. Measured IBUs and the components of the SMPH model.

A much more detailed explanation of the concepts and factors used in the SMPH model is described in a separate blog post, A Summary of Factors Affecting IBUs.

Factors Influencing IBUs
The SMPH model accounts for a number of factors that influence IBUs. These factors can be put into one of three groups for the purposes of discussion: “large-impact”, “medium-impact”, and “small-impact” factors.

Large-Impact Factors
The factors that can have a large impact on IBUs are (a) hops added to hot wort (kettle hops) vs. ambient-temperature or “cold-side” wort (dry hops), (b) form of the hops (whole cones or pellets), (c) hopping rate, (d) wort pH, and (e) wort clarity.

Kettle vs. Dry Hops: Hops added to hot wort in the kettle undergo alpha-acid isomerization, which produces the majority of bitterness in most beers. Dry hopping will produce no IAAs, but in large amounts it can produce significant bitterness from ABCs [Parkin, pp. 33-34; Maye and Smith, p. 135], especially from oxidized alpha acids created during hop storage. Oddly enough, at higher IBUs the use of dry hopping also reduces the concentration of IAAs from kettle additions [Parkin, p. 34; Maye and Smith, p. 135]. The IBUs from a dry-hop addition are difficult to estimate, but the difference between adding hops to the boil kettle or to the fermentation or conditioning vessel will have the largest impact on the IBU value.

Form of Hops: Hop pellets produce more IBUs than whole cones. With pellets, the production of oxidized alpha acids when hops are added to the boiling wort is about double that of whole cones. This factor seems to be variety specific, with some varieties producing very little increase from pellets, and other varieties producing a large increase. The rate of alpha-acid isomerization appears to be the same when using pellets or whole cones.

Hopping Rate: It is well known that doubling or tripling the amount of hops generally won’t produce a doubling or tripling of the IBU. As the concentration of hops increases, the resulting IBU value increases more slowly. An alpha-acid solubility limit is a reasonable explanation for this effect, with all alpha acids dissolving up to about 200 ppm and a reduction in the percent that dissolves as the alpha-acid concentration increases. Mark Garetz incorporated a hopping-rate factor into his model, but I suspect that he underestimated the effect.

Wort pH: I’ve found that lowering the pH from 5.75 (the approximate pH of a mash made from untreated low-alkalinity water and two-row malt) to 5.25 (within the recommended range of 5.2 to 5.4) can reduce IBUs by 15% to 35%. Most of the decrease in IBUs appears to come from a loss of ABCs, with only a small loss of IAAs.

Wort Clarity: Much to my surprise, I’ve found that the clarity of the wort can have a significant impact on IBUs. In this case, “clarity” refers to how visually clear or cloudy the wort is when it is transferred to the fermentation vessel (FV), ignoring the effect of hop matter. Cloudy wort yields relatively fewer IBUs. In other words, wort produced using the brew-in-a-bag technique with no filtering of the grain bed can yield a much lower IBU value than clear wort produced with a careful vorlauf and good grain-bed filter. (This is not to say that one method is better than the other, just that they may yield different IBUs.) Likewise, stirring the wort just before transferring into the FV can produce a lower IBU value than letting the wort settle and racking only the clear wort into the FV. I’ve observed very clear wort producing 30% more IAAs than typical wort, and very cloudy wort producing 30% fewer IAAs than typical wort. The reason for IBUs being affected by wort clarity is unknown, but wort protein levels do not seem to be a factor.

Medium-Impact Factors
Factors that often have only a medium impact on IBUs are: (a) how well the hops have been stored (hop freshness), (b) wort specific gravity, (c) the use of a hop stand, (d) losses to krausen deposits, and (e) the age of the beer.

Hop Storage Conditions: The storage conditions of hops can have a large impact on the amount of alpha acids remaining in those hops. As the amount of alpha acids decreases due to poor storage conditions and/or longer storage duration, the amount of oxidized alpha and beta acids increases, somewhat mitigating the reduction in IBU values [Peacock, p. 162]. (Nitrogen-flushed packaging and cold storage are the best ways to preserve hops.) While differences in storage conditions may not have a large effect on the IBU, I think storage conditions do have a large impact on overall beer quality.

Wort Gravity: Wort gravity is one of the factors common to all IBU prediction models. On average, the difference in IBUs between a 1.030 wort and a 1.080 wort is about 15%. The difference between a 1.040 wort and a 1.070 wort is about 10%.

Hop Stands: During a hop stand, alpha acids continue to isomerize in the hot wort, increasing the IBU. The amount of impact from a hop stand depends a lot on the duration of the stand and when hops are added to the wort, so I’ve classified this as a medium-impact factor.

Krausen: Most brewers let krausen deposits accumulate on the sides of the fermentation vessel. If you skim off the krausen as it is produced (which is sometimes recommended to produce a “smoother” beer [e.g. Troester; Hough et al., pp. 652-653]), the resulting IBU value can be about 5% to 10% lower. If you use a blow-off tube and remove a lot of the krausen, the IBU value may be 25% lower. If you mix the krausen back into the beer (or use an anti-foaming agent) during fermentation then the IBU may be about 10% higher. I’ve classified krausen as a medium-impact factor because the loss of lots of krausen through a blow-off tube is quite possible but perhaps not so common.

Age of the Beer: After primary fermentation, IBUs will decrease as the beer conditions. I have noticed a 20% decrease in IBUs as a beer ages from 1 week to 13 weeks at about 60°F (16°C). While a lot of the decrease seems to happen in the first several weeks, most beers aren’t conditioned for months at cellar or room temperature, and if a beer is conditioned or stored at cold temperatures, IBUs are probably much better preserved. Therefore, I’ve put this factor in the “medium-impact” category, but it’s probably a small impact for cold-conditioned lagers.

Small-Impact Factors
The factors that usually have a minor impact on IBUs are (a) the boiling point of water, (b) the rate at which wort is force-cooled after flameout or a hop stand, and (c) flocculation, finings, and filtering.

Boiling Point of Water:
The difference in IBUs when brewing at sea level compared with Boulder, Colorado or Johannesburg, South Africa is about 20% for typical beers. This would be a large-impact factor, but most cities are at 1000 feet (300 meters) or less, in which case the impact is 4% or less. (In a typical beer, the majority of IBUs come from alpha-acid isomerization, and we can use the Malowicki model of temperature-dependent isomerization to estimate the impact of altitude.)

Rate of Wort Cooling: After flameout or a hop stand, alpha acids continue to isomerize in the hot wort while it is force-cooled, down to about 140°F (60°C). These post-boil IAAs increase the IBU. While there may be a large difference in IBUs when going (for example) from an ice bath to a Hydra^™ wort chiller, smaller differences in cooling technique may have only a small impact on IBUs.

Flocculation, Filtering, and Finings: These factors are each estimated to influence the IBU by about 5% or less [Garetz, pp. 140-140; Fix and Fix, p. 129].

No-Impact Factors
There is one more group of factors that aren’t in the SMPH model because I don’t believe that they have any meaningful impact on IBUs. Such factors include the kettle size and kettle geometry, containing hops in a mesh bag, and the use of malt extract instead of wort from all-grain brewing. Kettle size or geometry is sometimes claimed to have an impact on IBUs, but one explanation for the correlation between kettle size and IBUs is the time it takes to cool a large volume of wort and the isomerization that happens while the wort is being cooled. My experiments have used a wide range of volumes, and I’ve seen no effect of volume or kettle size on IBUs. However, it is possible that hydrostatic pressure is a factor that may increase IBUs; an experiment by Brülosophy found a significant perceptual difference resulting from a change in hydrostatic pressure.  Further tests of IBUs and hydrostatic pressure may yield interesting results.  Putting hops in a mesh bag is sometimes claimed to reduce IBUs, but experiments conducted by both Brülosophy and me have shown no meaningful difference in measured IBUs. I’ve also heard that brewing with malt extract can yield different IBUs than with all-grain brewing, but my direct comparison of beers brewed with Briess^™ Pilsen Dried Malt Extract and Great Western^™ Premium two-row malt showed no meaningful difference in measured IBU values. (Also, I can think of no plausible mechanism through which the concentration of wort into dried malt extract could affect alpha-acid isomerization or the concentration of ABCs.) Future experiments may show some relationship for some of these factors under different conditions, for example with hops in a fine-mesh bag or specific brands of malt extract, but for now there is no known difference worth modeling.

SMPH Parameter Estimation
For most parameters in the SMPH model, estimated values could be obtained from the literature, direct experimentation, or reasonable assumptions. For a few parameters, though, there was no good estimate: (a) the loss of IAAs to trub during the boil, (b) what percent of the available alpha acids are quickly oxidized when added to hot wort, and (c) what percent of the alpha acids that oxidize during storage are dissolved when added to wort. In addition, I wanted to use all available data to get better estimates of the two parameters used in a hopping-rate correction model. A set of 347 measured IBU and IAA values were used to estimate values for these five parameters. (Four IBU and four IAA values were taken from Val Peacock’s reported numbers [Peacock, p. 162]. The other values came from my experiments.)

While this may seem like a lot of data for estimating five parameters, the estimation was complicated by the fact that I often didn’t have precise estimates of the alpha-acid content on brew day and/or how the hops had degraded during storage. Each measured value was therefore associated with a small parameter search for these experiment-specific values as well as the five common values.

Optimizing the parameter values to fit the data resulted in a root-mean-square (RMS) error of 1.6 IBUs and a maximum difference of 7.1 IBUs (for a condition that had 81 measured IBUs). The estimated loss factor for IAAs during the boil is 0.51. The percent of available alpha acids that quickly oxidize when added to hot wort is estimated at 11%. The percent of storage-generated oxidized alpha acids that dissolve in the wort is estimated at 33%. The solubility of alpha acids (for hopping-rate correction) is estimated to have a minimum limit of 200 ppm (below which all alpha acids are dissolved) and a maximum of 580 ppm.

Test Results
To evaluate and compare different IBU models, I collected an additional set of 18 IBU values that were not used in parameter estimation or for cross-validation of the SMPH model. These values ranged from 20.2 to 70.0 IBUs, including a variety of ale styles (two stouts, one ESB, one Kölsch, an English IPA, a West-Coast IPA, and twelve single-malt-and-single-hops (SMASH) beers with different timings of the hop additions). All IBU values were measured from finished beer.

The table below shows, for five IBU models, the RMS error and maximum difference between a measured and modeled IBU value on this set of 18 data points.

Model	RMS Error (IBUs)	Max. Error (IBUs)
SMPH	2.4	5.2
Tinseth	20.4	70.5
Rager	39.6	137.9
Garetz	12.34	28.14
mIBU	11.4	33.2

Table 1. A numerical comparison of measured and predicted values for five IBU models.

Figure 2 compares measured IBUs and predicted IBUs for the five models, with measured IBUs on the horizontal axis and predicted IBUs on the vertical axis. The straight dashed line from lower left to middle right indicates where predicted and measured IBUs are equal. It can be seen that on this set of data, the Tinseth, Rager, and mIBU models all have very large predicted IBUs for the higher-IBU beers. The Garetz model has a good fit with the higher-IBU samples, but predicts values about 50% too low in the range of 20 to 25 IBUs.  The SMPH model has an RMS error 4.5 times lower than the next-best model (Garetz) and 12 times lower than the Tinseth model.

Figure 2. A visual comparison of measured and predicted values for five IBU models.

When I conducted the analysis for Table 1 and Figure 2, I believed that all of these models predict IBUs in beer. I have since found that when developing his model, Prof. Tinseth measured IBUs from samples of wort, not beer, and so the Tinseth model may be better suited for predicting IBUs in wort.  (He also collected “data … that small breweries provided” [Hieronymous, p. 185], and it’s unknown if this data was for wort or beer, or to what degree his model was developed from his samples of wort or the data from other breweries. Therefore, it’s not entirely clear if it’s appropriate to apply his model to the prediction of IBUs in beer, even though that is how his model is used.) I’ve been unable to find out if the Rager model was based on data from wort or beer, but the general similarity of the utilization curves implies that it was based on the same type of data as the Tinseth model. The Garetz model was developed for finished beer, not wort [Garetz, p. 124].

There is a significant difference between IBUs measured from wort and IBUs measured from beer. Not only does fermentation reduce the IBU level through a reduction of isomerized alpha acids (by about 15%), but alpha acids that are present in the wort (having not yet undergone isomerization) are not present in the finished beer [Lewis and Young, p. 259]. These alpha acids are not bitter [Shellhammer, p. 169], but they do add to the measured IBU value by absorbing light at 275 nm. The concentration of these alpha acids at the end of the boil will vary greatly, depending on the hop schedule and how much isomerization has occurred up until flameout. The IBUs of wort can be 30% to 50% higher than the IBUs of beer [Justus, p. 72], with this wide variation possibly explained by different concentrations of alpha acids in the wort. The IBU was designed to be correlated with the bitterness of finished beer [Peacock, pp. 157-161], and the presence of alpha acids in unfermented wort will make IBUs measured from wort less reliable as a predictor of bitterness than IBUs measured from beer.

If you use the (unmodified) Tinseth, Rager, or mIBU models to predict IBUs in your beer (which is the standard use of these models), then Table 1 and Figure 2 provide an accurate evaluation of the prediction results. If you consider these models to be better suited for wort and not beer, and you modify these models to account for losses of alpha acids and the reduction in isomerized alpha acids that occurs during fermentation, then Figure 3 may provide a better comparison of the Tinseth, Rager, and mIBU models with measured IBU values. In this example, I’ve reduced the Tinseth, Rager, and mIBU utilization by 40%, providing a better fit with measured IBU values. If you use a different scaling factor, your results may differ.

After modifying the Tinseth formula to reduce predicted IBUs by 40%, this model has an RMS error of 7.9 IBUs and a maximum error of 14.3 IBUs, and the Rager model has an RMS error of 15.4 IBUs and a maximum error of 54.8 IBUs. The mIBU model, which is based on the Tinseth model, has an RMS error of 7.5 IBUs and a maximum error of 12.7 IBUs. Figure 3 shows a much better overall fit for these models with measured IBUs, implying that these models should be modified in order to predict IBUs in beer. However, at IBUs less than 30 the Tinseth model now consistently underpredicts IBUs by an average of 6.6 IBUs, Rager underpredicts by 5.2 IBUs, and mIBU underpredicts by 7.3 IBUs. At IBUs greater than 30, the Tinseth model now overpredicts by an average of 7.5 IBUs, Rager overpredicts by 17.5 IBUs, and mIBU underpredicts by 6.1 IBU. The RMS error of the SMPH model is still three times lower than that of the Tinseth model (2.5 vs 7.9 IBUs, respectively), and the SMPH model underpredicts at all IBUs by an average of only 1.7 IBUs, demonstrating that the SMPH model has significant advantages over other models even after modifying other models to predict IBUs in beer and not wort.

Figure 3. A visual comparison of measured and predicted values for three IBU models after reducing the predicted IBUs for these models by 40%.

Other Considerations
Some people are more sensitive to bitterness than others [Reed et al., p. 215]. From what I’ve observed, people who are very sensitive to bitterness find it unpleasant, and therefore they don’t tend to drink high-IBU beers. Also, the perception of bitterness changes with each sip. Therefore, I wouldn’t worry much about minor IBU differences; getting somewhere in the ballpark is probably just fine.

The IBU scales linearly with the concentrations of IAAs and ABCs. Bitterness, like most perceptual phenomena, does not increase linearly with the strength of the stimulus (as noted by Fechner’s law). Therefore, there is a divergence from the linear relationship between IBU values and the perception of bitterness, starting at about 60 IBUs [Hahn, p. 50]. However, as noted earlier, there is a strong correlation between IBUs and perceived bitterness, even at high IBUs. Hahn has developed a quadratic equation to map between IBUs and perceived bitterness, accounting for this non-linearity [Hahn, p. 50]. The SMPH calculator includes Hahn’s perceived bitterness value (or “bitterness intensity”) as an additional output.

Oxidized alpha acids are perceived as being about 34% less bitter than isomerized alpha acids [Algazzali, p. 45]. They absorb about 8.5% less infrared light than IAAs when measuring the IBU [Maye et al., p. 25, Figure 7], and so their perceptual bitterness is about 28% less than their measured contribution to the IBU (0.66/0.915 = 0.72). This is enough of a difference that if a beer containing only oxidized alpha acids (no IAAs) has 40 measured IBUs, it might be perceived as having the bitterness of a beer with only 29 IBUs.  This difference of 11 IBUs is above the perceptual threshold of 5 IBUs [Daniels, p. 76].

If the concentration of residual sugars in a beer is low and the IBU is large, the resulting beer may be perceived as overly bitter. Likewise, if there are a lot of residual sugars and a low IBU, the beer may be considered too sweet. Hahn’s perceptual study did not control for residual sugars, and yet panelists were able to fairly consistently judge a beer’s bitterness. The perception of bitterness and sweetness are different, but we prefer some relationship between them in our beers. The ratio of IBU to original-gravity points can be a useful (if imprecise) way to estimate this bitter/sweet balance and design a pleasing beer. Personally, I find that an IBU/OG ratio of about 0.5 creates a “balanced” beer a bit on the sweeter side, and an IBU/OG ratio of about 1.0 creates a pleasantly bitter (e.g. West-Coast) IPA.

One of the advantages of the Tinseth, Rager, and Garetz models is that no computer is needed to estimate IBUs. You just need to look up some values in tables and do basic math. These models are also easy to program, which has contributed to their popularity in brewing software. Unfortunately, the SMPH calculator is quite complex, using thousands of lines of code to compute concentrations and loss factors. This calculator is, however, available online to anyone who wants to use it.

Summary
An IBU value is determined by measurement of the amount of infrared light absorbed by (acidified) beer. The IBU deliberately includes the effects of both isomerized alpha acids and auxiliary bittering compounds. Even at higher IBUs, there is a strong correlation between IBUs and the perception of bitterness. IBU prediction usually doesn’t need to be very precise, because many people aren’t really all that good at detecting minor (or sometimes even moderate) differences in bitterness.

The SMPH model is a new method for estimating IBUs, which may be useful when trying to predict a beer’s bitterness before brewing. A key difference between the SMPH model and other IBU models is that it accounts separately for the contribution of IAAs and ABCs. Predicting IBUs is a bit of a “black art”, because there are so many variables and there is so much variability. The only way to really know the IBU level of a beer is to have it professionally tested, which is something I highly recommend.

Acknowledgments
I’d like to give a big shout-out to Dana Garves at Oregon BrewLab for the IBU measurements (as well as protein, polyphenol, and other measurements) used in developing the SMPH model.  I can always rely on the accuracy of the measured values and Dana’s cheerfulness. Scott Bruslind at Analysis Laboratory was also hugely supportive, helpful, and encouraging with my initial experiments. Zach Lilla at AAR Lab has been a friendly and reliable source for measuring alpha and beta acids (and the hop storage index) in my hops. I’d also like to thank Glenn Tinseth and Randy Mosher for prompt and encouraging answers to my out-of-the-blue questions. I greatly appreciate the spirit of cooperation and support that is a critical part of the homebrewing culture.

The SMPH model would not have been possible without the excellent research and publications by Tom Shellhammer (and his graduate students) at Oregon State University, Mark Malowicki (in particular), and Val Peacock at Hop Solutions, Inc.  While the model would not have been possible without their previous work, they had no input on its development, and so the name “SMPH” is simply a sequence of four letters, not an acronym.

References

• V. A. Algazzali, The Bitterness Intensity of Oxidized Hop Acids: Humulinones and Hulupones, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• Anon., “The IBU Assay” in Beer Sensory Science. https://beersensoryscience.wordpress.com/2011/04/05/the-ibu-assay/. Accessed Oct. 2, 2021.
• R. Daniels, Designing Great Beers: The Ultimate Guide to Brewing Classic Beer Styles. Brewers Publications, 2000.
• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques. Brewers Publications, 1997.
• M. Garetz, Using Hops: The Complete Guide to Hops for the Craft Brewer. HopTech, 1st edition, 1994.
• C. D. Hahn, A Comprehensive Evaluation of the Nonvolatile Chemistry Affecting the Sensory BItterness Intensity of Highly Hopped Beers.  Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2017.
• S. Hieronymus, For the Love of Hops: The Practical Guide to Aroma, Bitterness, and the Culture of Hops. Brewers Publications, 2012.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science. Volume 2: Hopped Wort and Beer. Springer-Science+Business Media, B. V., 2nd edition, 1982.
• A. Justus, “Tracking IBU Through the Brewing Process: The Quest for Consistency”, in MBAA Technical Quarterly, vol. 55, no. 3, pp. 67-74, 2018.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• J. P. Maye, R. Smith, and J. Leker, “Humulinone Formation in Hops and Hop Pellets and Its Implications for Dry Hopped Beers”, in MBAA Technical Quarterly, vol. 51, no. 1, pp. 23-27, 2016.
• J. P. Maye and R. Smith, “Dry Hopping and Its Effects on the International Bitterness Unit Test and Beer Bitterness”, in MBAA Technical Quarterly vol. 53, no. 3, pp. 134-136, 2016.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures”, in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• D. R. Reed, T. Tanaka, and A. H. McDaniel. “Diverse Tastes: Genetics of Sweet and Bitter Perception”, in Physiology & Behavior, vol. 88, no. 3, pp. 215-226, 2006.
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• Thermoscientific, “Beer Analysis — Bitterness”. https://assets.thermofisher.com/TFS-Assets/MSD/Application-Notes/SN53084-analyze-beer-bitterness.pdf. Accessed Oct. 2, 2021.
• K. Troester, “Should the Kraeusen fall back into the beer?”. http://braukaiser.com/blog/blog/2010/02/14/should-the-kraeusen-fall-back-into-the-beer/. Accessed Oct. 2, 2021.

Four Pilot Studies for Maximizing Hop Flavor with Late-Hop Additions

Abstract
The purpose of the experiments described here was to estimate at what point in the boil, and at what temperature, hops should be added in order to maximize hop flavor.  The first two perceptual tests were conducted using beers with the same amount of hops added at different times before flameout (from 1 to 20 minutes).  The third test was conducted with the same amount of hops added at 10 minutes before flameout and the kettle covered or not covered.  The fourth test was conducted with hops added at 10 minutes before flameout to boiling wort or to wort held at 170°F (77°C) .  The bitterness of the beers within each perceptual test was kept constant by adjusting the amount of a 40- or 45-minute hop addition.  These experiments were pilot studies due to the small number of test comparisons, the use of a single test subject, and the use of a single variety of hops.  The results indicate that hop flavor may be most pronounced with a 1-minute steep time, that evaporation has a gradual effect on hop flavor (with 10 minutes probably corresponding to a just-noticeable difference), and that the difference between a 1-minute and 20-minute steep time with an uncovered kettle was the most easily perceived of the conditions tested.  The 10-minute hop stand at 170°F (77°C) showed no perceptual difference from a 10-minute boil.  The results suggest that a “best practice” for maximizing hop flavor may be to add the hops very close to flameout, but that other late-hopping techniques may produce results that are perceptually very similar.

1. Introduction
The purpose of the experiments described here was to estimate at what point in the boil, and at what temperature, hops should be added for maximum hop flavor.  The term “hop flavor” can mean different things to different people.  For example, George Fix says that it has been traditionally (and not quite correctly) believed that the hop resins (which are responsible for bitterness) contribute to hop flavor, while the hop oils (including flavor compounds) contribute to hop aroma [Fix and Fix, p. 33 (emphasis mine)].  In this case, because the resins are responsible for bitterness, the term “hop flavor” is associated with the taste of bitterness.  Somewhat more recently, it has been recognized that hop oils contribute to hop “flavor and aroma” [Oliver, p. 539] and that “late-hopping [is] a well-accepted technique for adding hop flavor and aroma” [Oliver, p. 539], and so “hop flavor” can refer not to a bitter taste, but to a distinct non-bitter flavor.  Mark Garetz uses the term “character” to define this non-bitter flavor [Garetz, p. 14].  In this post, I use the term “flavor” for the non-bitter hop flavor that comes from the hop oils, with typical descriptions such as “floral,” “citrus,” “spicy,” “grapefruit,” or “earthy.”  These oils are also responsible for hop aroma [Oliver, p. 539], and so the terms “flavor” and “aroma” are often used together to describe their sensory impact.  I will use the term “flavor” with the understanding that flavor and aroma are intertwined.

It is usually said that hops should be added earlier in the boil for bitterness and later in the boil for flavor and/or aroma [e.g. Fix and Fix, p. 33; Garetz, pp. 10-11; Noonan, p. 160; Oliver, p. 539]. Therefore, the experiments in this blog post focus on late-hop additions ranging from 1 to 20 minutes before flameout and forced cooling.  (The distinction between “early” and “late” hopping is at around 30 minutes before flameout [Oliver, p. 539].)

While the belief in late hopping for flavor is nearly universal, it is difficult to find in the literature a “best” time for maximizing flavor or a quantified relationship between hop steep time and flavor.   Greg Noonan says that “flavoring hops are commonly added ten or fifteen minutes before the end of the boil for lager beer” [Noonan, p. 159].  Charlie Papazian is the only source I know of who provides a graph of the relationship between steep time and hop flavor, with a peak at 10 minutes before flameout (and a separate peak at 0 minutes for aroma) [Papazian, p. 68], but it’s unclear what set of data was used to produce this graph.  It is possible that chemical reactions between boiling wort and hop oils require some amount of time to produce the most hop flavor in finished beer.  Because flavor and aroma are intertwined, and the oils responsible for hop aroma are lost with evaporating steam [e.g. Lewis and Young, p. 271], it’s also possible that  peak hop flavor comes from flameout additions.  The use of hop stands, with hops steeped at below-boiling temperatures, are common in hop-forward ales and might also contribute to increased hop flavor.

Attempting to answer the question of when to add hops for maximum flavor presents two logistical challenges.  The first challenge is that the bitterness of beer increases with hop steep time and temperature, and so simply adding the same amount of hops at different times or temperatures will change the bitterness level in addition to any flavor changes.  This topic is discussed more in Section 2.  The second challenge is how to measure hop flavor in order to know when it has been maximized.  The perceptual-testing approach used here is discussed in more detail in Section 3.

I’ve created a separate web page as an interactive tutorial for the mathematics behind perceptual difference testing, including significance testing, the power of a test, likelihood ratios, estimating the effect size (d’), and confidence intervals.  These different analysis methods can be used to obtain a detailed interpretation of the results, which can be especially useful when the number of samples per trial is small and/or the statistical power of the test is low.

The perceptual experiments described below used only a single test subject and a single hop variety (Amarillo).  In addition, the number of test samples used in these experiments was too small to reliably detect minor perceptual differences. These experiments are therefore pilot studies; results are tentative and these results may or may not be supported by future studies.  Having tentative results is at least a first step toward having more conclusive results.

2. Controlling for Bitterness
In order to control the bitterness level of the beers in these experiments, I used up to two hop additions in each condition.  One addition was the same weight of hops added at different times or temperatures before flameout.  Another addition (if used) was always made at 40 or 45 minutes before flameout (40 minutes for the first two experiments; 45 minutes for the second two), and the weight of this other addition was varied in order to target the same IBU value across all conditions within a test.  Because additions at 40 or 45 minutes are considered to be primarily for bittering and not for flavor, the goal was to change the flavor with the timing of the late-hop addition but to keep total bitterness of each condition the same with the smaller but earlier addition.

To predict IBU values for each condition, I used the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements.  This technique is used, with Mark Malowicki’s model of alpha-acid isomerization [Malowicki], to estimate two parameters for modeling IBUs: scaling_IAA and scaling_nonIAAhops.  The scaling_IAA parameter indicates how much of the isomerized alpha acids (IAA) are lost during the boil and fermentation, and the scaling_nonIAAhops parameter indicates (a) what percent of the weight of the hops becomes auxiliary bittering compounds during the boil and (b) to what degree these compounds are lost during the boil and fermentation.  I obtained initial estimates of these two parameters from a preliminary study.   I used these values, along with wort volume, weight of the hops, AA rating, pH, and original gravity to predict IBUs.  The preliminary study and all experiments described here used hops from the same one-pound (0.45 kg) bag, to keep the alpha-acid (AA) rating and alpha-acid decay factor [e.g. Garetz, pp. 103-118] as equal as possible across conditions.

For the late-hop addition, I targeted an initial alpha-acid concentration close to the estimated alpha-acid solubility limit of about 200 ppm.  The IBU prediction technique estimates a certain IBU value from this amount of hops, wort, temperature, and steep time (ranging from 1 up to 20 minutes).  I then adjusted the weight of another hops addition, always added at 40 or 45 minutes before flameout, so that the model predicted the same total IBU value across all conditions within an experiment.  The goal was to have all of the conditions in a perceptual comparison within 5 measured IBUs of each other, as 5 IBUs has been reported to be the perceptual threshold [Daniels, p. 76].  Up to about 50 or 60 IBUs there is a strong linear relationship between IBUs and perceived bitterness [Hahn, p. 50], and so for beers in this range the IBU is a good (and linear) metric for perceived bitterness.

3. Flavor Testing Methodology
3.1 Overview
To measure hop flavor, I used the triangle test (also used at Brülosophy) in order to judge whether two conditions can be distinguished from each other [e.g. Angevaare; Society of Sensory Professionals].  In the triangle test, a test subject tastes three samples of beer where two of the samples are from the same condition and one is from a different condition.  The subject is asked which one of the three beers is different.  This test is repeated a number of times.  If the number of correct answers is above a threshold, then the two conditions can be considered perceptually different.  It is important to note that if the number of correct answers is below the threshold, nothing can be concluded from a standard significance test; standard significance testing can not accept the hypothesis that there is no perceptual difference between two conditions.  However, likelihood ratios can be used to estimate the relative strength of the evidence for whether two beers are perceptually the same or different.  We can also estimate the effect size (d’), which indicates the amount of difference between the two conditions.  A d’ of 0 indicates identical conditions, a d’ of 1.0 corresponds to a just-noticeable difference, and larger values of d’ indicate greater perceptual differences.

In this test, the beer judged as different was also rated by the subject as having either “more hop flavor” or “less hop flavor” than the others.  By comparing beers at a range of steep times, one can first determine which steep times can be distinguished from each other.  Then, for those samples that are correctly identified as different, one can look at how often one steep time is judged more flavorful than the other.

3.2 Testing Details
These experiments used a single subject or taster (this author).  This single-subject design has advantages and disadvantages.  One significant disadvantage of using a single subject is that the results from these experiments may or may not generalize to the larger population.  One significant advantage of using a single subject is that there is probably a lower threshold for detecting perceptual differences, compared with a larger group of subjects.  (Even if the one subject has a high threshold compared with the average population, the variance in the responses will be less for one subject than for many subjects due to individual threshold differences.  This variance in responses negatively affects the effect size (lowering the value of d’), making it more difficult to distinguish between conditions in a study with many subjects.)

In the first two perceptual studies, both experiments had four conditions for different hop steep times, labeled A, B, C, and D in Experiment #1 and E, F, G, and H in Experiment #2. This resulted in six comparisons between conditions (in Experiment #1, Condition A vs. B, A vs. C, A vs. D, B vs. C, B vs. D, and C vs. D).  Each comparison was tested eight times, for a total of 48 tests per experiment.  The third experiment had two conditions: (J) kettle covered or (K) uncovered during the 10-minute late-hop addition.  The fourth experiment also had two conditions: hops added to (L) boiling or (M) 170°F (77°C) wort for 10 minutes.  Each of the comparisons in the third and fourth experiments was tested 24 times.  The final two perceptual studies were conducted simultaneously, for a total of 48 tests.

A computer program was written to arrange the tests in random order with random ordering of conditions within a test.  Tests were conducted up to four times per day with at least an hour between tests (to reduce order effects), and so each experiment took about two weeks to test.  A second person poured samples for two to four tests every morning according to an instruction sheet with the randomized order of conditions.  Each test sample was 1.5 oz (44 ml), and so more than 74 oz (2.2 liters) of each condition were required for testing.  While the beers were stored close to freezing to preserve flavor, each sample of beer came up to room temperature before tasting.

The subject marked their responses (i.e. indicated the beer that was judged different, and if they thought this beer was more or less flavorful than the others) on a separate sheet.  Testing was conducted in a quiet room with as much time as needed for making a decision.  The subject did not know the correct answers until the end of the experiment.

3.3 Evaluating Results
With eight tests of a comparison and a significance level of 0.05, six tests need to be correctly identified in order to reach statistical significance and reject the null hypothesis of “no perceptual difference.”  At the same significance level, seven of the eight comparisons need to be correctly identified in order to reach statistical significance rejecting the null hypothesis of a just-noticeable difference (JND).  Unfortunately, with only eight results per trial, the power of a significance test comparing no perceptual difference against the JND is an abysmal 6%, meaning that 94% of the time that there really is a just-noticeable perceptual difference, a statistically-significant result will not be obtained.  (This is one reason why a test result that does not show significance should not be used to conclude that the conditions are perceptually equal.  These experiments were conducted with the expectation that there would be more than a just-noticeable difference in at least one comparison.)

With 24 tests of a comparison and the same significance level, 13 tests need to be correctly identified to reach statistical significance and reject the null hypothesis of “no perceptual difference”, and 16 tests need to be correctly identified in order to reject the null hypothesis of a just-noticeable difference.  The power of a test comparing no perceptual difference against the JND is still a miserable 15%, meaning that 85% of the time that there really is a just-noticeable perceptual difference, a statistically-significant result will not be obtained.

In order to obtain more information from the test results, the likelihood ratios and maximum-likelihood estimates of the effect size (d’) with a 95% confidence interval were computed, in addition to significance testing.  For those less familiar with these concepts, there is an interactive tutorial on the terminology and mathematics of perceptual testing.

4. Experiment #1: Varying Steep Times with an Uncovered Kettle
4.1 Experiment #1: Experimental Overview
In this experiment, a late-hop addition was made at 1, 5, 10, or 20 minutes.  The kettle was uncovered during the final 20 minutes of the boil, allowing volatile hop oils to evaporate.

4.2 Experiment #1: Experimental Methods
All conditions used 2.55 lbs (1.16 kg) of Briess Pilsen Dried Malt Extract with 3.37 G (12.75 liters) of 120°F (49°C) water to create 3.50 G (13.25 liters) of room-temperature wort with specific gravity 1.031.  The wort sat for about 90 minutes to let the pH stabilize, at which point the pH was adjusted with phosphoric acid to 5.30.  The wort was boiled (uncovered) for 5 minutes to reduce the foam associated with the start of the boil.  A 12-oz (0.35 liter) sample was taken for measuring specific gravity and a 40-minute timer was started.  The first addition of Amarillo hops (AA rating 8.8%) was made with the weight listed in Table 1 (using a weighted coarse-mesh bag).  The kettle was covered for the first 20 minutes of the boil to reduce evaporation, after which time the cover was removed to allow evaporation.  At each target time, the second addition of 0.850 oz (24.1 g) of the same Amarillo hops (with the steep time listed in Table 1) was added in a weighted coarse-mesh bag.  At flameout the wort was quickly cooled with an immersion chiller to 75°F (24°C) and the hops were removed.  Sterilized, room-temperature water was added to bring the volume up to about 3.0 G (11.36 liters).  The wort was stirred and then sat for about 15 minutes, covered, to settle the heavier trub.  Then, 0.813 G (3.08 liters) of wort was transferred to a sanitized fermentation vessel.  This wort was aerated for 1 minute by vigorous shaking, and 0.08 oz (2.20 g) of Safale US-05 yeast was added.  A final sample was taken from the kettle for measuring specific gravity.

The wort fermented for one week, after which time 92 oz (2.72 liters) were decanted, leaving the trub behind.  From that, a 4-oz (0.12 liter) sample was taken for IBU measurement by Oregon BrewLab.  The remainder was stored at close to freezing with minimal exposure to oxygen until the results from Oregon BrewLab confirmed that the samples were all within 5 IBUs of each other.  Except when bringing samples up to room temperature for tasting, the beers were kept at near freezing and with minimal exposure to oxygen.

The perceptual experiment was conducted as described in Section 3.2.  Conducting up to four tests per day took 17 days.  Due to the difficulty in detecting clear differences between samples, tasting of each sample was spaced out by about 30 seconds and small sips of water or a tiny amount of dry bread was taken between tastings to reset the palate.

Condition:	A	B	C	D
weight of 1st addition:	0.379 oz / 10.75 g	0.289 oz / 8.20 g	0.185 oz / 5.25 g	0 oz / 0 g
steep time of 2nd addition:	1 min.	5 min.	10 min.	20 min.
pre-boil specific gravity (SG):	1.031	1.031	1.031	1.031
pre-boil volume: (measured, room temp.)	3.51 G / 13.30 liters	3.49 G / 13.22 liters	3.50 G / 13.25 liters	3.50 G / 13.26 liters
SG at 1st addition:	1.033	1.033	1.034	1.033
volume at 1st addition: (estimated from SG)	3.30 G / 12.50 liters	3.27 G / 12.38 liters	3.26 G / 12.34 liters	3.29 G / 12.46 liters
post-boil SG: (after volume correction)	1.036	1.036	1.036	1.036
post-boil volume: (estimated from SG)	3.03 G / 11.46 liters	2.99 G / 11.32 liters	3.02 G / 11.43 liters	3.02 G / 11.44 liters
measured IBUs	23.6	24.5	23.0	22.8

Table 1.  Measured and estimated (where indicated) values for the four conditions with an uncovered kettle.

4.3 Experiment #1: Results and Analysis
The IBU levels from the four conditions were well within the perceptual threshold of 5 IBUs.  The average was 23.5 IBUs, with a standard deviation 0.66 IBUs.  The maximum difference between two conditions was 1.7 IBUs.  These results indicate that the beers were not perceptually different in terms of bitterness.

The results of the perceptual test are shown in Table 2.  The top-right corner of the table provides the number of correct responses, the p value associated with this response rate (with the value in bold font if significance was reached), the likelihood ratio for a just-noticeable difference relative to no perceptual difference, and the low, maximum-likelihood, and high estimates of d’ (using a 95% confidence interval; a d’ of 0 corresponds with no perceptual difference, and a d’ of 1 corresponds with a just-noticeable difference).  The bottom-left corner of the table shows the identity of the preferred sample for each correct response.

The expected amount of variability in the results is quite large, given only 8 samples per trial (standard deviation 1.4 samples).  Two trends in the correct-response rate are visible, however: (1) Condition A is more likely to be distinguished from the other conditions, and (2) other comparisons indicate that no perceptual difference is approximately just as likely as a just-noticeable difference.

One unusual result is that the comparison of A vs. B demonstrates a significant difference, and A vs. D also demonstrates a significant difference, but A vs. C does not demonstrate significance.  Jumping ahead a little bit in the story in order to explain these results, the experiment described in Section 6 (to test the impact of evaporation on a 10-minute steep time) has results which indicate that the true underlying trend is probably that A and B actually have the least perceptual difference, A and C probably have a not significant and just-noticeable difference, and A and D have the largest perceptual difference.  In other words, evaporation and steep time probably affects perception, but the effect is more likely to be a gradual change over a period of about 10 or 20 minutes.

For the preferences, all of the correct responses involving Condition A were associated with a preference for Condition A.  For comparison B vs. C, the preference was equally split.  For B vs. D, the single correct response favored D.  For C vs. D, four out of the five favored Condition C.  Shorter steep times appear to be somewhat preferred over longer steep times, but the only universal preference was for the shortest steep time of 1 minute.

These results (and taking into account the results from Section 6) suggest that the shortest hop steep time has the most perceived hop flavor, and that evaporation probably affects hop flavor gradually over a 10- to 20-minute period.  Based on these results, one should keep hops in the wort for the shortest time possible in order to maximize flavor.

Comparison:	A:	B:	C:	D:
A:		6 / 8 correct p = 0.020 LR: d’=1/d’=0 = 2.98 d’ (low, ML, high) = 0.68, 2.79, 4.68	2 / 8 correct p = 0.805 LR: d’=1/d’=0 = 0.69 d’ (low, ML, high) = 0, 0, 2.10	7 / 8 correct p = 0.003 LR: d’=1/d’=0 = 4.28 d’ (low, ML, high) = 2.10, 3.75, 4.68
B:	more flavor: AAAAAA		4 / 8 correct p = 0.259 LR: d’=1/d’=0 = 1.44 d’ (low, ML, high) = 0, 1.46, 3.75	1 / 8 correct p = 0.961 LR: d’=1/d’=0 = 0.48 d’ (low, ML, high) = 0, 0, 0.68
C:	more flavor: AA	more flavor: BB CC		5 / 8 correct p = 0.088 LR: d’=1/d’=0 = 2.07 d’ (low, ML, high) = 0, 2.10, 3.75
D:	more flavor: AAAAAAA	more flavor: D	more flavor: CCCC D

Table 1.  Results from perceptual testing with an uncovered kettle.  The top-right corner shows analysis of the number of correct responses.  The bottom-left corner shows, for those samples correctly identified as different, which sample was considered to have more hop flavor.

5. Experiment #2: Varying Steep Times with a Covered Kettle
5.1 Experiment #2: Experimental Overview
The experiment with an uncovered kettle showed that hop flavor is probably maximized with the shortest possible steep time.  There are two likely explanations for this: (1) the hop oils degrade when they’re in boiling wort, and/or (2) the hop oils are removed from the wort through evaporation.  If the first explanation is true, then one may be able to vary the temperature of the wort in order to minimize degradation and maximize flavor.  If the second explanation is true, then one only needs to cover the kettle in order to prevent the loss of hop oils.  The experiment described here tested the second explanation by covering the kettle during the boil.  If there is no perceptual difference between any of the conditions, that would suggest that the oils are lost primarily through evaporation.  If results are similar to the experiment with the uncovered kettle, that would suggest that oils are mostly degraded in boiling wort.

5.2 Experiment #2: Experimental Methods
This experiment was conducted using the same general methods as the first experiment.  The first addition of Amarillo hops was made with the weight listed in Table 3 (using a weighted coarse-mesh bag).  The kettle was covered during the entire 40-minute steep time, except for brief stirring and to add the second hop addition.  At each target time, the second addition of 0.765 oz (21.7 g) of Amarillo hops (with the steep time listed in Table 3) was added in a weighted coarse-mesh bag.

The perceptual experiment was conducted as described in Section 3.2.  Unfortunately, a bug in the randomization yielded between 7 and 12 samples per trial, instead of always 8 samples per trial.  Conducting up to four tests per day took 16 days.  Due to the difficulty in detecting clear differences between samples, tasting of each sample was spaced out by about 30 seconds and small sips of water or a tiny amount of dry bread was taken between tastings to reset the palate.

Condition:	E	F	G	H
weight of 1st addition:	0.363 oz / 10.30 g	0.274 oz / 7.77 g	0.181 oz / 5.14 g	0.096 oz / 2.71 g
steep time of 2nd addition:	1 min.	5 min.	10 min.	15 min.
pre-boil specific gravity (SG):	1.031	1.032	1.032	1.031
pre-boil volume: (measured, room temp.)	3.48 G / 13.18 liters	3.48 G / 13.18 liters	3.48 G / 13.19 liters	3.50 G / 13.25 liters
SG at 1st addition:	1.033	1.033	1.033	1.034
volume at 1st addition: (estimated from SG)	3.31 G / 12.54 liters	3.33 G / 12.62 liters	3.18 G / 12.04 liters	3.28 G / 12.42 liters
post-boil SG:	1.034	1.0345	1.036	1.035
post-boil volume: (estimated from SG)	3.18 G / 12.03 liters	3.18 G / 12.04 liters	3.01 G / 11.41 liters	3.14 G / 11.88 liters
measured IBUs	20.2	21.4	21.2	18.7

Table 3.  Measured and estimated (where indicated) values for the four conditions with a covered kettle.

5.3 Experiment #2: Results and Analysis
The IBU levels from the four conditions were well within the perceptual threshold of 5 IBUs.  The average was 20.4 IBUs with standard deviation 1.07 IBUs.  The maximum difference between two conditions was 2.7 IBUs.  These results indicate that the beers were not perceptually different in terms of bitterness.

The results of the perceptual test are shown in Table 4.  The top-right corner of the table provides the number of correct responses, the p value associated with this response rate (none of the results reached significance), the likelihood ratio for a just-noticeable difference relative to no perceptual difference, and the low, maximum-likelihood, and high estimates of d’.  The bottom-left corner of the table shows the identity of the preferred sample for each correct response.

In this experiment, condition E (the shortest steep time) does not demonstrate any significant differences against the other conditions.  Overall, the likelihood ratios show no clear trend; for example, conditions with a greater difference in steep time are not more likely to have a just-noticeable difference than conditions with a small difference in steep time.  Unlike the first experiment, all of the 95% confidence intervals include a d’ of 0, or no perceptual difference.

For the preferences, there is also no clear preference for any one steep time.  The number of correct responses is quite small in most comparisons, and the only comparison with more than four correct responses was evenly split in preference between the two conditions.

While it’s not possible to demonstrate that two conditions are perceptually the same using standard significance testing, the set of results here suggests that all conditions in this experiment have at most a just-noticeable difference and quite likely no perceptual difference. In the previous experiment, Condition A had greater perceptual differences from other conditions and was universally preferred over other conditions; those patterns were not observed in this experiment.  These results suggest that hop oils lost through evaporation are an important component of hop flavor.

Comparison:	E:	F:	G:	H:
E:		3 / 8 correct p = 0.532 LR: d’=1/d’=0 = 1.00 d’ (low, ML, high) = 0.0, 0.68, 2.79	2 / 7 correct p = 0.737 LR: d’=1/d’=0 = 0.79 d’ (low, ML, high) = 0, 0, 2.58	3 / 8 correct p = 0.532 LR: d’=1/d’=0 = 1.00 d’ (low, ML, high) = 0.0, 0.68, 2.79
F:	more flavor: EE F		1 / 9 correct p = 0.974 LR: d’=1/d’=0 = 0.42 d’ (low, ML, high) = 0, 0, 0	4 / 8 correct p = 0.259 LR: d’=1/d’=0 = 1.44 d’ (low, ML, high) = 0, 1.46, 3.75
G:	more flavor: GG	more flavor: G		7 / 12 correct p = 0.066 LR: d’=1/d’=0 = 2.48 d’ (low, ML, high) = 0, 1.89, 3.38
H:	more flavor: E HH	more flavor: BB DD	more flavor: GGG HHHH

Table 4.  Results from perceptual testing with a covered kettle.  The top-right corner shows analysis of the number of correct responses.  The bottom-left corner shows, for those samples correctly identified as different, which sample was considered to have more hop flavor.

6. Experiment #3: Covered vs. Uncovered Kettle with 10-Minute Addition
6.1 Experiment #3: Experimental Overview
The first experiment demonstrated an unexpected result: a significant difference between 1 and 5 minutes (A vs. B comparison with 6 correct responses out of 8 tests), no significant difference between 1 and 10 minutes (A vs. C with 2 out of 8 correct), and a significant difference between 1 and 20 minutes (A vs. D with 7 out of 8 correct).  It is mathematically more likely that the lack of perceptual difference in the A vs. C comparison is an incorrect conclusion, which implies that hop oils quickly evaporate with steam.  However, the number of data points in this experiment was small and therefore the uncertainty is large.  A third experiment was conducted to test this hypothesis with more data.  This experiment had two conditions, J and K, both with a 10-minute late-hop addition.  The primary difference between the two conditions was that in Condition J the kettle was covered during the final 10 minutes and in Condition K the kettle was uncovered (allowing steam to escape).  If the tentative conclusion from the first experiment is correct and hop oils are quickly lost with evaporating steam, then there should be a perceptual and significant difference between Conditions J and K.  (With an estimated d’ of 2.79 in the A vs. B comparison and 3.75 in the A vs. D comparison, an estimate of d’ for a 10-minute steep time is about 3.  With 24 tests and a d’ of 3.0, the power of the test is close to 1.0.)

6.2 Experiment #3: Experimental Methods
This experiment was conducted using the same general methods as the first and second experiments.  Wort for each condition was created using 2.47 lbs (1.12 kg) of DME and 3.27 G (12.38 liters) of water, yielding 3.43 G (13.0 liters) of wort with specific gravity 1.031.  The first addition of 0.176 oz (5.0 g) of Amarillo hops (AA rating 8.8%) was made at 45 minutes before flameout (in a weighted coarse-mesh bag).  Both conditions had 0.811 oz (23.0 g) of Amarillo hops added in a weighted coarse-mesh bag at 10 minutes before flameout.  Safale S-04 yeast was used for fermentation.

For Condition J, the kettle was uncovered for the first 10 minutes after the initial hop addition, and then covered for the remaining 35 minutes of the boil (with the brief exception of adding the 10-minute hop addition).  For Condition K, the kettle was uncovered during the first 10 minutes, covered during the next 25 minutes, and uncovered during the final 10 minutes (after the second hop addition was made).

The perceptual experiment was conducted as described in Section 3.2.  Conducting 24 tests with up to four tests per day, along with the 24 tests in the fourth experiment, took 17 days.  With the expectation of less difficulty in detecting a clear difference between samples and a desire to balance memory effects with adaptation effects, tasting of each sample was spaced out by about 10 seconds and only small sips of water were taken between tastings to reset the palate.

6.3 Experiment #3: Results and Analysis
The measured IBUs were 24.7 for Condition J and 28.9 for Condition K.  The difference between these IBU levels, 4.2, is within the perceptual threshold of 5 IBUs.  These results indicate that the beers were not perceptually different in terms of bitterness.

The results of the perceptual test were that 11 out of the 24 tests were correctly identified, and of those correct responses, 3 times Condition J was preferred and 8 times Condition K was preferred.  The p value associated with this response rate is 0.14 (not significant at a threshold of 0.05), and the likelihood ratio for a just-noticeable difference relative to no perceptual difference is 2.07. The low, maximum-likelihood, and high estimates of d’ are 0.0, 1.24, and 2.32, respectively.

These results were very much unexpected, in the low estimate of d’, the lack of significance, and the general preference for the uncovered late-hop addition over the covered late-hop addition.  These results imply that in the first experiment the A vs. B comparison (1 min. vs. 5 min.) yielded an incorrect result that supported a perceptual difference, and that the A vs. C comparison (1 min. vs. 10 min.) was actually correct in not demonstrating significance.  Given the strength of the A vs. D comparison (1 min. vs. 20 min., with 7 out of 8 correct and consistent responses), it seems prudent to continue to assume that the result of that comparison was correct.

The preference for Condition K over Condition J might be due to (a) difficulty in distinguishing these two conditions (with a fairly low d’) (b) small differences in the perceptual testing methodology that may have had an unexpectedly large effect , (c) the use of a different strain of yeast, and/or (d) flavor changes over time due to the transformation of hop oils in the hot wort in addition to the loss of oils through evaporation.  The simple explanation that hop oils are simply lost through evaporation may or may not be the complete explanation.

Considering the set of results of the first three experiments, it appears that hop flavor does decrease with longer steep times, but only relatively slowly.  We can estimate the perceptual change over time (with an uncovered kettle) as a d’ of roughly 1.0 after 10 minutes (a just-noticeable difference) and a d’ of roughly 3.0 (with a maximum-likelihood estimate of 3.75) at 20 minutes.  With the preference for the shortest steep time in the first experiment not consistent with the preference for the uncovered kettle in the third experiment, it is unclear if flavor changes occur only through evaporation, through additional mechanisms, or if testing differences or statistical variation in the third experiment caused a different result.  The universal preference for the shortest steep time in the first experiment leads to the tentative conclusion that flavor is maximized with the shortest steep time.

7. Experiment #4: Boiling vs. Sub-Boiling Hop Addition
7.1 Experiment #4: Experimental Overview
A comparison of the results from the first and second experiments indicates that covering or not covering the kettle can be responsible for a noticeable change (or lack of change) in hop flavor.  The results of the third experiment suggest that the effect of covering the kettle is only a just-noticeable difference at a 10-minute steep time.  Other than volatile hop oils evaporating with steam, another likely explanation for a change in hop flavor is a transformation of hop oils in contact with boiling wort.  The fourth experiment tested the effect of wort temperature on hop flavor, comparing a 10-minute steep time at boiling (Condition L) with a 10-minute steep time at 170°F (77°C)  (Condition M).

7.2 Experiment #4: Experimental Methods
This experiment was conducted using the same general methods as the previous three experiments.  Dried malt extract was used to create 3.43 G (13.0 liters) of wort with pre-boil specific gravity 1.031.  The first addition of Amarillo hops (AA rating 8.8%) was made at 45 minutes before flameout (in a weighted coarse-mesh bag).  Condition L used 0.176 oz (5.0 g) of hops in the first addition and was identical with Condition J in Experiment #3.  Condition M used 0.388 oz (11.0 g) of  hops in the first addition.   Both conditions had 0.811 oz (23.0 g) of Amarillo hops added in a weighted coarse-mesh bag at 10 minutes before flameout, and the kettle was covered during the final 10 minutes.  In Condition L the wort was kept at boiling; in Condition M, the wort was cooled from boiling to 170°F (77°C) during the 11th minute before flameout using an immersion chiller, and the target temperature was maintained (to within a few degrees) during the final 10 minutes before flameout.  Safale S-04 yeast was used for fermentation.

The perceptual experiment was conducted as described in Section 3.2.  Conducting 24 tests with up to four tests per day, along with the 24 tests in the third experiment, took 17 days.  As in the third experiment, tasting of each sample was spaced out by about 10 seconds and only small sips of water were taken between tastings to reset the palate.

7.3 Experiment #4: Results and Analysis
The measured IBUs were 25.1 for Condition L and 29.6 for Condition M.  The difference between these IBU levels, 4.5, is within the perceptual threshold of 5 IBUs.  These results indicate that the beers were not perceptually different in terms of bitterness.

The results of the perceptual test were that 5 out of the 24 tests were correctly identified, and of those correct responses, 3 times Condition L was preferred and 2 times Condition M was preferred.  The p value associated with this response rate is 0.94 (not significant at a threshold of 0.05), and the likelihood ratio for no perceptual difference relative to a just-noticeable difference is 4.29. The low, maximum-likelihood, and high estimates of d’ are 0.0, 0.0, and 0.8, respectively.

While it is not possible to conclude that two conditions are perceptually identical using significance testing with a null hypothesis of no difference, it would be difficult to get results that more clearly indicate no perceptual difference between the two conditions.  Even random guessing would result in, on average, 8 of the 24 tests being correctly identified.  The result of 5 correct responses is not so low that one should be concerned about experimental error, but low enough that the likelihood of there being no perceptual difference is more than four times greater than there being a just-noticeable difference.  The maximum-likelihood estimate of d’ is 0, indicating no perceptual difference.  In short, there is no evidence that there is a perceptual difference between hops boiled for 10 minutes and hops kept at 170°F (77°C)  for 10 minutes.  I will abuse the mathematics a bit and conclude that a sub-boiling hop stand produces no noticeable difference in hop flavor, at least for a 10-minute steep time and these experimental conditions.

8. Conclusions
8.1 Summary of Results
The results from these experiments indicate that hop flavor is lost primarily through evaporating steam while the hops are steeped in hot wort.  After about 10 minutes of steeping there may be a just-noticeable difference in hop flavor; after about 20 minutes the difference may be more easily perceived.  Flavor appears to be lost through the evaporation of hop oils, but it is also possible that other factors also affect the flavor compounds over time.

The best-practice recommendation resulting from these experiments is to keep hops in boiling wort for as short a time as possible in order to preserve hop flavor, but a difference of 10 minutes or a decrease in wort temperature may not have a perceptible impact, especially with a covered kettle.  This recommendation might be paraphrased as: minimize the time that the hops are in hot wort, but (in the words of Charlie Papazian) relax, don’t worry, and maybe have a homebrew.

One potential concern with a covered kettle is the production of dimethyl sulfide (DMS) which can then not be removed by evaporation.  Most ales, however, “have DMS levels well below threshold” [Fix and Fix, p. 50].  Because the precursor S-methylmethionine (SMM) and DMS are reduced more at ale fermentation temperatures than at lager fermentation temperatures, “any hint of DMS in ales is likely from technical brewing errors, most notably contamination” [Fix, p. 75].  In lagers, the increase in DMS caused by a covered kettle can be counteracted with a longer (uncovered) boil time and/or faster wort cooling [Fix and Fix, pp. 50-51].  (The other option is to not worry about DMS and brew lager in the style of Rolling Rock [Bamforth, p. 18].)

8.2 Comments on Perceptual Testing
In general it was very difficult to tell the beers in these conditions apart, despite the nearly ideal testing conditions.  This difficulty was compounded (or caused) by the first taste of a beer being the most perceptually distinctive and subsequent tastes of other samples having less sensory impact.   There was therefore a balance between waiting long enough to reset the palate but not waiting so long that the specifics of the flavor were forgotten.  Taking small sips of water or eating a tiny amount of dry bread to reset the palate in between tastings seemed to help, but in most cases the differences between conditions were very subtle (or nonexistent).

My general preference for the flavor obtained from a 1-minute steep time with Amarillo hops may or may not be shared by others.  As a counterexample, my wife thinks that every IPA she has ever encountered tastes and smells disgusting.  Another hop variety might yield different results.  In short, your perceptions and preferences may be different from the results of these experiments.

9. Acknowledgment
I would like to sincerely thank Dana Garves at Oregon BrewLab for the IBU measurements in these experiments.  Oregon BrewLab has been a pleasure to work with, and I can always rely on the accuracy of the measured values.

References

• J. Angevaare,  A New Triangle Test Calculator.  https://onbrewing.com/a-new-triangle-test-calculator/.  Accessed Apr. 21, 2021
• C. Bamforth.  Beer is Proof God Loves Us.  FT Press, 1st edition, 2011.
• R. Daniels, Designing Great Beers: The Ultimate Guide to Brewing Classic Beer Styles. Brewers Publications, 2000.
• G. Fix, Principles of Brewing Science. Brewers Publications, 2nd edition, 1999.
• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques. Brewers Publications, 1997.
• M. Garetz, Using Hops: The Complete Guide to Hops for the Craft Brewer. HopTech, 1st edition, 1994.
• C. D. Hahn, A Comprehensive Evaluation of the Nonvolatile Chemistry Affecting the Sensory BItterness Intensity of Highly Hopped Beers.  Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2017.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• G. J. Noonan, New Brewing Lager Beer. Brewers Publications, 1996.
• G. Oliver, The Oxford Companion to Beer, Oxford University Press, 2011.
• C. Papazian, The Home Brewer’s Companion.  William Morrow / HarperCollins,  1st edition, 1994/2002.
• Society of Sensory Professionals, Triangle Test.  https://www.sensorysociety.org/knowledge/sspwiki/Pages/Triangle%20Test.aspx. Accessed Apr. 21, 2021.
• Wikipedia.  Hops.  https://en.wikipedia.org/wiki/Hops.  Accessed Apr. 21, 2021.

Why Do Hop Pellets Produce More IBUs Than Hop Cones?

Abstract
Hop pellets are usually described as having higher utilization than hop cones.  A separate blog post looks at the amount of increase in IBUs caused by using pellets instead of cones. It finds that the amount of increase is constant over a range of hop steep times, instead of increasing with steep time.  This means that the increase in IBUs is not caused by an increase in the rate of alpha-acid isomerization or availability of alpha acids, which would result in longer steep times having a greater increase in IBUs. The first experiment in this post looks at whether this constant increase is more likely to be caused by a greater concentration of isomerized alpha acids (IAA) produced soon after a hop addition, or by other bittering compounds (nonIAA, also called “auxiliary bittering compounds”).  This experiment analyzes the rate at which IAA and nonIAA are removed from beer over time, and a comparison is made with the rate at which the increase in IBUs from pellets decreases over time.  The results indicate that pellets yield increased IBUs from an increase in auxiliary bittering compounds, not from increased IAA.  In other words, the concentration of isomerized alpha acids in finished beer is the same for beer made with cones or pellets, but the concentration of nonIAA is greater in beer made from pellets.  Data from a second experiment indicate that while the concentration of polyphenols is greater with the use of pellets, this greater polyphenol concentration cannot explain the observed increase in IBUs.  In this experiment, the increase in IBUs from pellets does not increase linearly with the amount of hops added, which is consistent with the IBU increase being caused by oxidized alpha acids.  (The same alpha-acid solubility limit that explains relatively lower IAA at higher alpha-acid concentrations can explain the relatively lower production of oxidized alpha acids at higher concentrations.)  The most likely explanation for the increase in IBUs when using pellets is that the pelletization process gives the alpha acids greater surface area, and that these exposed alpha acids oxidize quickly when brought into contact with hot wort, creating an increase in the concentration of oxidized alpha acids during the boil.

1. Background: Utilization, Reported Differences, and IBU Models
1.1 Utilization
Hop utilization, U, is the amount of isomerized alpha acids (IAA) in finished beer divided by the amount of alpha acids added to the kettle, and then multiplied by 100 to convert to percent [e.g. Lewis and Young, p. 266]:

U = 100 × (isomerized alpha acids in beer) / (alpha acids added to kettle)

[1]

Utilization refers only to the relative amount of isomerized alpha acids, not to IBUs.  While IAA and IBUs can be considered roughly equivalent as a quick rule of thumb, IBUs measure a number of bitter components in addition to IAA.  These other bitter components are called nonIAA or “auxiliary bittering compounds”.  With short boil times, high hopping rates, low steeping temperatures, improperly-stored hops, and other factors, one can see significant differences between measured IBUs and the concentration of IAA.

1.2 Reported Differences Between Cones and Pellets
Hop pellets are almost always described as having greater utilization than hop cones [e.g. Daniels p. 78].  According to Michael Lewis and Tom Young, “the alpha acids dissolve most easily from extracts, less easily from pellets …, and least with whole hops” [Lewis and Young, p. 266].  It is said that the higher rate at which alpha acids from pellets “dissolve,” compared with whole cones, is because “the pelletization process ruptures the lupulin glands and spreads the resins over the hop particles, giving a larger surface area for isomerization” [Hall, p. 58].  Greg Noonan says that “with pelletized hops, ruptured and better-exposed lupulin glands give greater utilization” [Noonan, p. 154].

1.3 Modeling IBUs from Pellets with Scaling Factors
A previous blog post describes a model of IBUs based on equations from Val Peacock [Peacock, p. 157] and Mark Malowicki [Malowicik, p. 27].   This model can be used to estimate the scaling factors for isomerized alpha acids (IAA) and auxiliary bittering compounds (nonIAA) in beer made from either cones or pellets.  Another blog post used those scaling factors to show that the increase in IBUs is modeled well by an increase in the concentration of nonIAA, or by some process that adds IBUs at the beginning of the boil but not during the rest of the boil.

2. Introduction
Although the other blog post on pellet-based IBUs found that the increase in IBUs resulting from the use of pellets was modeled well by an increase in nonIAA concentrations, it is still possible that this increase is actually caused by the rapid production of isomerized alpha acids close to the start of a hop addition, instead of the usual time-dependent alpha-acid isomerization.  The model referred to in Section 1.3 groups all compounds that are produced near the beginning of a hop addition as nonIAA compounds, under the assumptions that isomerization is a fairly slow process and that nonIAA compounds are produced quickly.  If IAA are also somehow produced quickly after adding hops, this model would not be able to distinguish these IAA from nonIAA.

Perhaps the process of manufacturing pellets (which includes heat [Srečec, pp. 141-143], a primary factor in isomerization [Verzele and De Keukeleire, pp. 102-109]) transforms alpha acids into an intermediate compound which then quickly results in IAA when the pellets are added to boiling wort.  Such a process would mean that pellets show increased IBUs because of greater utilization, even if this increase in utilization happens much more quickly than the typical isomerization process.  (The existence of such an intermediate compound is postulated simply to explain how the increase in IBUs seen with the use of pellets might be caused by isomerized alpha acids, since the rate of isomerization or availability of alpha acids is not affected when using pellets.)

The rest of this blog post addresses the question of whether the increase in IBUs observed with the use of pellets is more likely to be the result of (a) IAA that are produced soon after a hop addition (i.e. greater utilization), (b) oxidized alpha acids produced when the hops are added to the boiling wort [Algazzali, p. 17; Dierckens and Verzele, p. 454], or (c) hop polyphenols.  (It is highly unlikely that this increase is related to oxidized beta acids because of the negligible impact that oxidized beta acids have on the IBU when using well-preserved hops.)

3. Experiment #1: Experimental Overview and Methods
3.1 Overview of Experiment #1
The IBU level and the concentration of IAA in beer decrease over time, especially at room temperature [Peacock, p. 164].  This decrease may be caused by IAA and possibly nonIAA transforming over time into different products or binding with other compounds and precipitating out of solution.  In either way, IAA and possibly nonIAA compounds are removed from beer over time.  The current analysis assumes that the rate at which IAA and nonIAA compounds are removed from beer is different.  By transforming multiple IBU measurements taken from a single beer at multiple points throughout the boil (both fresh beer and aged beer) into estimates of IAA and nonIAA factors in a model of IBUs, we can evaluate how these factors (and therefore IAA and nonIAA concentrations) change with the age of the beer.  If the increase in IBUs produced by the use of pellets decreases over time at the same rate as IAA loss, we can conclude that this increase in IBUs is probably produced by IAA.  Conversely, if the decrease matches the rate of nonIAA loss, we can conclude that nonIAA compounds are most likely responsible for the increase in IBUs with pellets.  (If the different-rate-of-decay assumption is wrong, then the decrease in IAA will be the same as the decrease in nonIAA, and no conclusions will be possible.)

A picture may help to illustrate the overall concept.  Figure 1 shows hypothetical cases of (a) IBUs produced using hop cones (solid dark blue line), (b) IBUs produced using hop pellets (solid dark green line), (c) the same cone-produced beer after 10 weeks of aging (dotted light-blue line), and (d) the same pellet-produced beer after 10 weeks (dotted light-green line).  This set of hypothetical data is based on two assumptions: (1) the change in IBUs over 10 weeks is due entirely to the loss of IAA; nonIAA compounds do not decrease in beer over time, and (2) the increase in IBUs caused by the use of pellets comes entirely from nonIAA compounds.  These assumptions produce a particular pattern in the IBU levels in Figure 1: (a) the solid green line and solid blue line are different by a constant factor (due to nonIAA compounds), (b) the dotted blue line starts at the same value as the solid blue line at 0 minutes, and then gradually decreases relative to the solid blue line (because only IAA levels decrease with age), and (c) the dotted green line and dotted blue line are different by the same constant factor (because the increase in IBUs with pellets comes only from nonIAA, which does not decrease over time).  Neither of these assumptions may be true, but we can analyze real IBU data using the model mentioned in Section 1.3 to estimate scaling factors.  The scaling factors, which could be used to produce graphs like Figure 1, will tell us how much loss occurs in both IAA and nonIAA over 10 weeks. We can then compare the change in pellet-related IBUs over the 10 weeks to the IAA and nonIAA scaling factors.  Comparing the rates at which losses occur will help us determine if the increase in IBUs from the use of pellets is more likely caused by IAA or nonIAA.

Figure 1. Hypothetical IBU levels from fresh and aged beer made with cones and pellets.  The data in this figure are made up in order to illustrate the patterns one might see as IBUs change over time in both types of beer.

I previously brewed two batches of beer that were nearly identical in all respects except for the use of cones in one case and pellets in the other, as part of a previous blog post (Hop Cones vs. Pellets: IBU Differences, Experiment #5).  For each batch, I took samples of wort at 10-minute intervals during a 60-minute boil.  Each sample was fermented into beer and 4 oz of each was sent to Oregon BrewLab for IBU analysis about 10 days after the start of fermentation.  I kept whatever wasn’t sent to Oregon BrewLab at room temperature for aging.  Those additional 12 samples were sent to Oregon BrewLab for IBU analysis at 10 weeks after the start of fermentation.

3.2 Methods for Experiment #1
All data for this experiment consisted of two batches of beer brewed on the same day, one batch using hop cones and the other using hop pellets.  I used 7.0 lbs (3.18 kg) of Briess Pilsen DME in 8.0 G (30.28 l) of water, yielding about 8.5 G (32 l) of wort with a specific gravity of about 1.037.  I did not adjust the water profile or pH, which resulted in a pre-boil wort pH of 5.80.

In this experiment, I used Comet cones from Hops Direct (stored in my freezer soon after harvest for about 4 months) and Comet pellets from YCH Hops (lot P92-ZLUCOM5216, about 2½ years old at the time of the experiment).   The previous blog post concluded that the age of the hop pellets did not have any impact on the pellet-based increase in IBUs.

I added hops (i.e. started the steep time at 0) after the wort had been boiling for 5 minutes, to avoid the foam associated with the start of the boil.  The hops were boiled for a total of 60 minutes with the cover on the kettle (except for taking samples) to minimize evaporation and the resulting changes in specific gravity.  I used 1.939 oz (54.96 g) of hop cones (alpha-acid rating 9.70%) and 2.147 oz (60.86 g) of hop pellets (alpha-acid rating 8.76%) to target an initial alpha-acid concentration of 170 ppm in both batches.

Samples were taken every 10 minutes from the start of steeping.  Each sample was taken from the boil in a measuring cup and then transferred to an aluminum cup using a wire mesh sieve to remove larger hop particles.  For the cones condition, 32-oz (0.95-liter) samples were taken; for the pellets condition, 16-oz (0.44-liter) samples were taken.  The aluminum cup was placed in an ice bath and the contents were stirred to cool quickly.  Once cooled to 75°F (24°C), the sample was transferred to a sanitized, sealed, and labeled quart (liter) container.  I aerated each sample by vigorous shaking for 60 seconds, then added .008 or 0.017 oz (0.24 or 0.48 g) of Safale US-05 yeast (depending on the volume of the sample) to target 750,000 viable cells per ml and degree Plato [Fix and Fix, p. 68].  After all samples were taken, the containers were cracked open to vent, and they fermented for eight days.  I swirled the samples every day to remove most of the krausen deposits on the sides of the containers.  After fermentation, I sent 4 oz (0.12 l) of each sample to Oregon BrewLab for IBU measurement.  The remainder of each sample then proceeded to age for 10 weeks at room temperature.  After 10 weeks, another 4 oz (0.12 l) was sent to Oregon BrewLab for IBU measurement.

4. Experiment #1: Results
The estimated room-temperature volume at the start of steeping was 8.34 G (31.57 liters).  The specific gravity after 10 minutes of steeping was about 1.0384.  The specific gravity after a 60-minute steep time was 1.0396.  The small change in specific gravity during the boil (due to keeping the lid on the kettle) means that there is little difference between using the measured IBU values for analysis or normalizing these IBUs by the volume when the sample was taken.  For simplicity and clarity, the measured IBU values are used below.

Figure 2 shows the measured IBU values from this experiment.  The average  difference in IBUs between cones and pellets is shown for both the fresh and aged beer.

Figure 2. Measured IBU data from beer made with cones or pellets, at 1 and 10 weeks after the start of fermentation.  The average IBU difference between cones and pellets at week 1 is 11 IBUs, and the average difference at week 10 is 8.5 IBUs.

5. Experiment #1: Analysis
5.1 Average Differences and Visual Analysis
The increase in IBUs caused by the use of pellets decreases from an average of 11.02 IBUs at week 1 to 8.50 IBUs at week 10.  This implies that whatever is causing this increase in IBUs, it does decay as the beer ages.  This pellet-based increase in IBUs decayed by a factor of 0.77 over the 10-week period (0.77 = 8.50/11.02).

It appears that the slope of the line changes between weeks 1 and 10 for both cones and pellets, with less of a difference at 10 minutes and more of a difference at 60 minutes, but the effect is subtle.  This change in slope is caused by the loss of IAA; a 10% loss of IAA will have less of an absolute effect on 20 IBUs than it will on 40 IBUs.  Because the data do not extend back to a steep time of 0, it is difficult to see if the vertical-axis offset of the lines changes with the age of the beer, which would correspond with a decrease in nonIAA concentrations.

In short, whatever is causing the increase in IBUs does decrease with age, and both IAA and nonIAA might decrease with age.  To get a more conclusive answer, we need to distill the data in this graph into a smaller set of numbers for easier comparison.

5.2 Model and Scaling Factors
We can use the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to split the IBU value into estimates of (a) the concentration of IAA and (b) the concentration of other bitter substances measured with the IBU that are called nonIAA.  Since nonIAA are predominately oxidized alpha acids (oAA), we can use existing models of the other factors (polyphenols and oxidized beta acids) and focus on estimating the concentration of oAA.  (The assumption of oAA as the primary source of nonIAA differences between cones and pellets is examined in Experiment #2.  Even if this assumption is incorrect, the model uses a direct translation between the concentration of oAA and total nonIAA, and so the results of this experiment will still be valid for nonIAA although off by a constant scaling factor.)

We can use multiple IBU values from the same batch of beer, along with an equation that describes the isomerization of alpha acids as a function of time and temperature [Malowicki, p. 27], an equation that describes the IBU as a combination of IAA and nonIAA in the finished beer [Peacock, p. 161], and models of polyphenols and oxidized beta acids, to estimate two scaling factors: scaling_IAA and scaling_oAA.  The scaling_IAA parameter is the scaling factor that accounts for losses of IAA during the boil, fermentation, and aging; scaling_oAA is the scaling factor from the initial concentration of alpha acids in the wort to the concentration of oxidized alpha acids in the beer.  With scaling_IAA and scaling_oAA, as well as the volume of wort, weight of the hops, initial alpha-acid concentration, steep time, original gravity, and models of polyphenols and oxidized beta acids, we can map from IBU value to IAA and oAA concentrations, and vice versa.  The IBU values resulting from this analysis are listed in Table 1.

	10	20	30	40	50	60
cones, week 1 (meas., estimate)	16.4, 15.7	21.2, 21.8	26.6, 27.0	31.3, 31.4	35.2, 35.3	39.0, 38.5
cones, week 10 (meas., estimate)	13.5, 12.8	17.5, 18.1	22.1, 22.6	26.6, 26.6	30.0, 30.0	33.2, 32.8
pellets, week 1 (meas., estimate)	26.0, 26.4	33.6, 32.6	37.8, 37.9	41.6, 42.5	46.9, 46.5	49.9, 49.8
pellets, week 10 (meas., estimate)	20.7, 21.4	27.9, 26.6	31.2, 31.2	33.6, 35.1	40.0, 38.4	40.5, 41.3

Table 1. Measured and estimated IBUs for each sample in each condition. Samples are identified by the duration of hop boiling, in minutes (column headings). The type of hops (cones or pellets) and the age of the beer are identified by row headings. Each cell in the table shows measured IBUs followed by estimated IBUs. Estimates are from the model described in Section 5.2.

5.3 Analysis of Cones Data
The analysis of IBU data of the beer made with hop cones and aged one week resulted in scaling_IAA = 0.472 and scaling_oAA = 0.067.  These results indicate that somewhat less than half of the isomerized alpha acids from this batch made it into the finished beer, and about 7% of the alpha acids added to the wort ended up as oxidized alpha acids in the beer.  The analysis of beer made with hop cones and aged 10 weeks resulted in scaling_IAA = 0.414 and scaling_oAA = 0.049.

From these results, we can estimate that IAA levels decayed by a factor of 0.877 over the 10 weeks (0.877 = 0.414/0.472), and nonIAA levels decayed by a factor of 0.731 (0.049/0.067).  The decrease over time attributed to pellet-specific factors (0.77 from Section 5.1) is closer to 0.73 than it is to 0.88, and so this suggests that the pellet-based increase in IBUs is more likely to be caused by oxidized alpha acids.

5.4 Analysis of Pellet Data
We can perform a similar analysis on the set of pellet data.  However, we don’t want to include the effect of the increase in IBUs caused by pellets in our analysis results, so when we estimate values for scaling_IAA and scaling_oAA, we add 11.02 IBUs to the model of week-1 data and 8.50 IBUs to the model of week-10 data.  (Or, equivalently, we can subtract 11.02 from the measured values from week 1 and 8.50 from the measured values at week 10.)  When this is done, the beer made with hop pellets and aged one week results in scaling_IAA = 0.484 and scaling_oAA = 0.059.  The beer made with hop pellets and aged 10 weeks results in scaling_IAA = 0.411 and scaling_oAA = 0.047.

From these results of pellet-based IBUs, we can estimate that IAA levels decayed by a factor of 0.849 over the 10 weeks (0.849 = 0.411/0.484) and nonIAA levels decayed by a factor of 0.797 (0.047/0.059).  While this difference between IAA and nonIAA degradation is smaller than that estimated for cones, the decrease over time attributed to pellets (0.77) is even slightly less than the estimated nonIAA decay factor for pellets (.797).  This indicates again that the pellet-based increase in IBUs is more likely to be caused by nonIAA compounds than by IAA.

The IAA scaling factor (scaling_IAA), oxidized alpha acid scaling factor (scaling_oAA), and root-mean-square (RMS) error resulting from this analysis are listed in Table 2.

	IAA scaling factor	oAA scaling factor	RMS error
cones, week 1	0.472	0.067	0.429
cones, week 10	0.414	0.049	0.461
pellets, week 1	0.484	0.059	0.613
pellets, week 10	0.411	0.047	1.106

Table 2. Estimated IAA and oAA scaling factors, and the associated RMS error, for each condition.

5.5 Averaged Analysis
The results in this study rely on parameter estimation that is subject to errors in the model, in the “known” values used in this model (i.e. the concentration of alpha acids at the start of steeping), and in the measured IBU values.  The pellet-based decay factor (0.77) is somewhat higher than the estimated nonIAA factor for cones (0.73), and the pellet-based decay factor is somewhat lower than the estimated nonIAA factor for pellets (0.80).

Assuming that these differences in results for cones and pellets are due to errors in the “known” or measured values, we can average the IAA and nonIAA decay factors (or the scaling factors) to arrive at a more robust combined estimate.  This averaging yields an IAA decay factor of 0.86 and a nonIAA decay factor of 0.76.  From these averaged values, we can conclude that the increase in IBUs caused by pellets (with a decay factor of 0.77) is most likely due entirely to nonIAA.

6. Experiment #2: Experimental Overview and Methods
6.1 Overview of Experiment #2
Having concluded in Experiment #1 that the increase in IBUs is more likely to come from nonIAA than from IAA, Experiment #2 looked at which of the components that are collectively referred to as nonIAA might be responsible for the increase. While Experiment #1 modeled the nonIAA increase assuming oxidized alpha acids are the unknown scaling factor, it is possible that this assumption is not correct, and that (for example) oxidized alpha acids are constant while the concentration of polyphenols is actually responsible for the increase in IBUs.

Malt polyphenols can obviously not be responsible for a change in IBUs caused by the type of hops used, and oxidized beta acids have a negligible impact on IBUs when using well-preserved hops.  This leaves oxidized alpha acids and hop polyphenols as the possible contributors.  It is possible that, even though hop polyphenols normally contribute only a small amount to the IBU [e.g. Shellhammer, p. 177; Almaguer, p. 300], the pelletization process produces such an increase in soluble hop polyphenols that this increase can explain the IBU differences between cones and pellets.

In order to test this theory, Oregon BrewLab measured the polyphenol concentrations in beer made with varying concentrations of hop cones and varying concentrations of hop pellets.  While isomerized alpha acids do not increase linearly with an increase in alpha acids, polyphenols should not have a solubility limit at even fairly high hopping rates.  We can then plot the change in polyphenol levels as a function of concentration to determine (a) the concentration of malt polyphenols, (b) the rate of increase of hop polyphenols when using cones, (c) the rate of increase of hop polyphenols when using pellets, and (d) whether any differences in the polyphenol concentrations between cones and pellets might explain the observed increase in IBUs from pellets.

6.2 Methods for Experiment #2
The data for this experiment consisted of five batches of beer brewed on the same day, two batches using hop cones and the other three batches using hop pellets.  Batch A used 0.76 oz (21.68 g) of cone hops with AA rating 8.32%.  Batch B used 2.04 oz (57.81 g) of the same cone hops.  Batch C used 0.67 oz (18.89 g) of pellet hops with AA rating 9.55%.  Batch D used 1.78 oz (50.37 g) of the same pellet hops.  Finally, Batch E used 2.66 oz (75.55 g) of the same pellet hops.  These weights, when used with the expected volume of wort when hops were added and with the estimated alpha-acid ratings, were designed to result in initial alpha-acid concentrations of 150 ppm, 400 ppm, 150 ppm, 400 ppm, and 600 ppm for Batches A through E, respectively.  Therefore Batches A and C can be directly compared, and Batches B and D can be directly compared.

For each batch, I used 2.88 lbs (1.31 kg) of Briess Pilsen DME in 3.32 G (12.57 l) of water, yielding about 3.47 G (13.14 l) of wort with a specific gravity of about 1.036.  I did not adjust the water profile or pH, which resulted in a pre-boil wort pH of 5.77.

The hops used in this experiment were from the same source as in Experiment #1.  This experiment was conducted nine months after the first; during that time, the hops were stored at about −9°F (−23°C) in vacuum-sealed bags.  I added hops (i.e. started the steep time at 0) after the wort had been boiling for 5 minutes, to avoid the foam associated with the start of the boil.  Samples were taken every 10 minutes from the start of the hop addition, for a total steep time of 40 minutes (4 samples).  Each 15-oz (0.44-liter) sample was taken from the boil in a measuring cup and then transferred to an aluminum cup using a wire mesh sieve to remove larger hop particles.  The aluminum cup was placed in an ice bath and the contents were stirred to cool quickly.  Once cooled to 75°F (24°C), the sample was transferred to a sanitized, sealed, and labeled quart (liter) container.  I aerated each sample by vigorous shaking for 60 seconds, then added .009 oz (0.25 g) of Safale US-05 yeast.  After all samples were taken, the containers were cracked open to vent, and they fermented for nine days. After fermentation, I sent 4 oz (0.12 l) of each sample to Oregon BrewLab for IBU measurement.  The sample taken at 10 minutes of steep time was also analyzed by Oregon BrewLab for polyphenol concentration.

7. Experiment #2: Results
The measured polyphenol levels were, for Batches A through E respectively: 112 mg/L, 125 mg/L, 112 mg/L, 130 mg/L, and 141 mg/L.  Figure 3 shows these polyphenol concentrations plotted as a function of the estimated concentration of total hop matter in the wort at the time the sample was taken (10 minutes of steeping).  The cone polyphenol concentrations are shown with green points and connecting dashed lines, and the pellet concentrations are shown with red points and connecting dashed lines.  The cones data and pellets data were each fit to a linear function (referred to as “model” in Figure 3), which are plotted in lighter green and red with solid lines.

Figure 4 shows the measured IBU values from this experiment, with cones in green and pellets in red.  The average difference in IBUs between Batches A and C is 8.3 IBUs, and the average difference between Batches B and D is 10.1 IBUs.

Figure 3. Concentration of polyphenols as a function of the concentration of total hop matter. Data for cones are plotted in green; data for pellets are plotted in red. The raw data are shown with triangles and dashed lines. The best linear fit to the data is shown using solid lines.

Figure 4. Measured IBU values for the five batches of beer in Experiment #2. Values for hop cones are shown in green, and pellets are shown in red. The average difference between Batches A and C is 8.3 IBUs, and the average difference between B and D is 10.1 IBUs.

8. Experiment #2: Analysis
In Experiment #2, the results from cones indicate a malt polyphenol concentration of 104.20 mg/L (when the hop polyphenol concentration is zero), and the results from pellets indicate a malt polyphenol concentration of 102.84 mg/L.  On average, the results indicate that in this experiment the malt contributed 103.5 mg/L of polyphenols.  The model of polyphenols developed in The Contribution of Malt Polyphenols to the IBU predicts 97.66 mg/L from the specific gravity and boil time, which is within 6% of the measured values.  The model of IBUs developed in that blog post predicts 0.81 IBUs from the malt polyphenols, based on the specific gravity and wort pH.

We can use the slope of the lines in Figure 3 to estimate what percent of the weight of the hops comes from polyphenols.  First, we assume that 20% of polyphenols dissolve in wort [Forster, p. 124] and that there is a fermentation loss factor of 0.70 (estimated in The Contribution of Malt Polyphenols to the IBU and assuming the same loss factor for malt and hop polyphenols).  From those assumptions and the slope of the lines in Figure 3, the hop cone polyphenols are 3.1% of the weight of the hops, and the pellet polyphenols are 4.5% of the weight of the hops.  Both of these values are within published estimates that hop polyphenol levels are in the range from 2% to 6% of the weight of the hops [Shellhammer, p. 169; Hough et al., p. 422; Algazzali, p. 5; Verzele and De Keukeleire, p. 9].  In general, the hop pellets here demonstrate a 43% increase in polyphenol concentrations, compared with hop cones (0.00628 / 0.00439 = 1.43 or 43% increase).

We can then use the slope of the lines in Figure 3 to estimate the IBUs contributed by the hop polyphenols.  Ellen Parkin reports that “an increase of 100 mg/L of polyphenols was predicted to increase the [IBU] value by 2.2” [Parkin, p. 28], and so the increase in hop polyphenols in Figure 2 can be mapped to an increase in IBU levels using a conversion factor of 0.022 from concentration (in mg/L) to IBUs.  Using this conversion results in estimates of 0.17, 0.21, 0.46, 0.57, and 0.85 IBUs for Batches A through E, respectively.

Figure 4 shows the measured IBU values from this experiment.  The average difference between Batches A and C is 8.2 IBUs, and the average difference between Batches B and D is 10.1 IBUs.  The first point of interest is that the observed increase in IBUs from using pellets is at least an order of magnitude greater than the expected increase in IBUs caused by hop polyphenols.  This effectively rules out hop polyphenols as being the primary cause of the increase in IBUs observed with pellets.  The second point of interest is that even though the concentration of hops increased by a factor of 2.67 between Batches A and B and between Batches C and D, the IBUs associated with the use of pellets increased only from 8.2 to 10.1 (on average) with the increase in hop concentration, or a factor of 1.22.  This implies that the increase in IBUs associated with pellets is subject to a solubility limit somewhere between 150 ppm and 400 ppm.  Such a solubility limit is already expected with alpha acids, but is not expected with other auxiliary bittering compounds.  This implied solubility limit is consistent with the hypothesis that the increase in IBUs with pellets is caused by the production of oxidized alpha acids when hops are added to the kettle; this oAA production would be restricted by the same solubility limit that limits the isomerization of alpha acids.

9. Summary and Conclusion
The results of the first experiment indicate that the increase in IBUs associated with the use of pellets is caused by an increased concentration of auxiliary bittering compounds, not by increased availability of alpha acids that quickly become isomerized alpha acids.

Of the possible auxiliary bittering compounds (oxidized alpha acids, oxidized beta acids, hop polyphenols, and malt polyphenols), oxidized alpha acids (especially those produced during the boil [Algazzali, p. 17]) are the only likely candidate, with the increase in IBUs from pellets as a function of hopping rate consistent with an alpha-acid solubility limit.  Oxidized beta acids produced during the boil are highly unlikely because of their very low presence in finished beer when using well-preserved hops.  Hop polyphenols are estimated to contribute about an order of magnitude less to the IBU than observed differences, and the contribution of malt polyphenols is obviously unrelated.

Based on the results of these experiments, oxidized alpha acids appear to be the source of the increase in IBUs when using pellets.  Why would the use of pellets increase the concentration of oxidized alpha acids?  Maye et al. found that oxidized alpha acids make up less than 0.5% by weight of hop pellets before being added to wort [Maye, p. 24], which is not enough to explain the observed increase in IBUs.  However, the “creation of [oxidized alpha acids] occurs when hops are added to boiling wort” [Algazzali, p. 17].  The pelletization process ruptures the luplin glands [Hall, p. 58], and therefore the alpha acids of pellet hops have a much greater surface area (compared with cones).  It seems plausible that the oxidation of alpha acids that happens during the boil is limited by both the initial available surface area of the alpha acids and their solubility; in other words, only those alpha acids that are initially exposed to (and dissolve in) the boiling wort are quickly oxidized.  Therefore, the greater surface area of alpha acids in hop pellets allows more production of oxidized alpha acids during the boil, thereby increasing the IBU value.

10. Acknowledgements
I would, as usual, like to thank Dana Garves at Oregon BrewLab for the IBU and polyphenol analyses for these experiments.  The conclusions reached by these experiments would not be possible without the level of accuracy that Oregon BrewLab provides.

References

• V. A. Algazzali, The Bitterness Intensity of Oxidized Hop Acids: Humulinones and Hulupones, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• C. Almaguer, C. Schönberger, M. Gastl, E. K. Arendt, and T. Becker, “Humulus lupulus – a story that begs to be told: A review,” in Journal of the Institute of Brewing, vol. 120, pp. 289-314, 2014.
• R. Daniels, Designing Great Beers: The Ultimate Guide to Brewing Classic Beer Styles.  Brewers Publications, 2000.
• J. Dierckens and M. Verzele, “Oxidation Products of Humulone and Their Stereoisomerism,” in Journal of the Institute of Brewing, vol. 75, pp. 453-456, 1969.
• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques.  Brewers Publications, 1997.
• A. Forster, “Influence of Hop Polyphenols on Beer Flavor,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• M. L. Hall, “What’s Your IBU,” in Zymurgy.  Special Edition, 1997.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science. Volume 2: Hopped Wort and Beer. Springer-Science+Business Media, B. V., 2nd edition, 1982.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• J. P. Maye, R. Smith, and J. Leker, “Humulinone Formation in Hops and Hop Pellets and Its Implications for Dry Hopped Beers”, in MBAA Technical Quarterly, vol. 51, no. 1, pp. 23-27, 2016.
• G. J. Noonan, New Brewing Lager Beer. Brewers Publications, 1996.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• S. Srečec, T. Rezić, B. Šantek, and V. Marić, “Hop Pellets Type 90: Influence of Manufacture and Storage on Losses of α-Acids,” in Acta Alimentaria. Vol. 38, no. 1, pp. 141–147, 2009
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids.  Developments in Food Science 27.  Elsevier, 1991.

 

Hop Cones vs. Pellets: IBU Differences

Abstract
Hop pellets are described as having greater utilization than hop cones.  The predicted amount of increase, however, varies quite a bit between different reports.  This blog post compares the IBUs from cones and pellets in a series of five experiments.  While the IBUs from pellets were found to be consistently higher than IBUs from cones, it seems that this increase in IBUs is not caused by an increase in the rate of isomerization (as is typically claimed), but by a greater concentration of bitter substances produced soon after a hop addition.  A separate blog post finds that these bitter substances are most likely not isomerized alpha acids, but probably oxidized alpha acids that are produced when hops are added to the boiling wort.  Furthermore, the amount of increase in IBUs seems to be dependent on the hop variety.   When controlling for the initial concentration of alpha acids, some hop varieties show very little increase (average 1.5 IBUs from 170 ppm of alpha acids), while others have a very large increase (average 10.8 IBUs from 170 ppm of alpha acids).  Because of this variety-dependent increase, predicting IBUs from hop pellets is even more challenging than predicting IBUs from hop cones. For hop cones, it is estimated that about 6% of the alpha acids added to the wort are quickly oxidized and survive into the finished beer.  (One ppm of oxidized alpha acids contributes about 0.7 IBUs.) Considering only the three hop varieties studied here, the increase in oxidized alpha acids from the use of pellet hops varies from a factor of 1.2 to a factor of 3.2. A rough (variety-independent) approximation for predicting IBUs from pellets is that the concentration of oxidized alpha acids produced during the boil doubles in beer made with pellets, from 6% to 12%.

1. Introduction: Reported Differences and IBU Models
1.1 Utilization
Hop utilization, U, is the ratio of the amount of isomerized alpha acids (IAA) in finished beer divided by the amount of alpha acids added to the kettle, and then multiplied by 100 to convert to percent [e.g. Lewis and Young, p. 266]:

U = 100 × (isomerized alpha acids in beer) / (alpha acids added to kettle)

[1]

Utilization refers only to the relative amount of isomerized alpha acids, not to IBUs.  While IBUs can be considered roughly equivalent to the concentration of IAA as a quick rule of thumb, IBUs measure a number of bitter compounds in addition to IAA.  With short boil times, high hopping rates, low steeping temperatures, improperly-stored hops, and other factors, one can see significant differences between IBUs and the concentration of IAA.

1.2 Reported Differences Between Cones and Pellets
Hop pellets are almost always described as having greater utilization than hop cones [e.g. Daniels p. 78].  According to Michael Lewis and Tom Young, “the alpha acids dissolve most easily from extracts, less easily from pellets …, and least with whole hops” [Lewis and Young, p. 266].  The higher rate at which alpha acids from pellets “dissolve,” compared with whole cones, is because “the pelletization process ruptures the lupulin glands and spreads the resins over the hop particles, giving a larger surface area for isomerization” [Hall, p. 58].  Greg Noonan says that “with pelletized hops, ruptured and better-exposed lupulin glands give greater utilization” [Noonan, p. 154].

Expressing pellets as more efficient than whole hops, Noonan provides a pellet correction factor (in table form) that varies from 1.0 to 1.5, based on boil time and gravity [Noonan, p. 215].  Mark Garetz recommends a pellet correction factor of 1.10 for boil times up to 30 minutes, otherwise a correction factor of 1.0 [Garetz, p. 131, 141].  Hieronymus says that hop pellets are 10% to 15% more efficient than cones [Hieronymus, p. 188].  According to Michael Hall, Randy Mosher specifies a correction factor of 1.33 [Hall, p. 62].  This is a wide range of relative increase, from 0% to 50% according to Noonan, and from 0% to 33% according to other sources.

The purpose of this blog post is to get a better understanding of how large an IBU increase there is when using pellets and how this increase can be modeled.  A separate blog post looks at whether this increase is more likely to be the result of a greater concentration of isomerized alpha acids or an increase in other bittering compounds; it finds that the IBU increase is most likely caused by an increased concentration of oxidized alpha acids.

1.3 A Model of the Isomerization of Alpha Acids
Mark Malowicki [p. 27] provides a formula for the concentration of isomerized alpha acids (IAA) as a function of steep time (t, in minutes), temperature, and initial alpha-acid concentration ([AA]₀, in ppm):

[IAA]wort = [AA]0 × (k1/(k2–k1)) (e–k1t-e–k2t)

[3]

where [IAA]_wort is the concentration of isomerized alpha acids in the wort at time t, and e is the constant 2.71828.  The parameters k₁ and k₂ are two temperature-dependent rate constants.  At boiling, k₁ = 0.0125 and k₂ = 0.0031.

1.4 Modeling IBUs from Pellets with a Scaling Factor
Figure 1 shows theoretical IBU values based on several scenarios described in this section.  These IBU values are based on Val Peacock’s model of IBUs [Peacock, p. 157], in which

IBU = 5/7 × ([IAA] + [nonIAA])

[2]

where [IAA] is the concentration of isomerized alpha acids in the finished beer and [nonIAA] is the concentration of other bittering substances in the beer.  (This model is described in more detail in the blog post Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements.)  In Figure 1, the black line shows theoretical IBU values from hop cones using Peacock’s model.  The concentration of isomerized alpha acids (IAA) increases from 8.5 ppm at 10 minutes into the boil to 35.0 ppm at 60 minutes (using the IAA model from Section 1.3 and a loss factor of 0.5), and the dotted gold line (constant at 5 IBUs) shows the contribution of nonIAA in this model.

The effect of pellets is usually expressed in the literature as a scaling factor [Hall, p. 62], for example a factor of 1.20 that is applied to the IBU value predicted for hop cones.  In this case, if an IBU model developed for hop cones predicts 30 IBUs, a pellet correction factor of 1.20 would yield 36 IBUs (36 = 30 × 1.20).  In Figure 1, the blue line shows theoretical IBU values predicted using a scaling factor of 1.20.  Because this scaling factor depends on the IBU value, smaller “cone” IBU values result in a smaller increase, and larger “cone” IBUs result in a larger increase.  For example, in Figure 1 the increase in IBUs is 2.2 IBUs at 10 minutes and 6.0 IBUs at 60 minutes.

Another way to model an increase in IBUs is with a scaling factor that depends on the concentration of isomerized alpha acids.  Because IBUs are correlated with [IAA], the net effect is similar.  In Figure 1, the dashed green line shows theoretical IBU values for pellets using a scaling factor of 1.25 applied to the concentration of IAA.

A third way to model an increase in IBUs is with a scaling factor that doesn’t depend on IBUs or [IAA], but on the concentration of nonIAA (which is also proportional to the total concentration of hops in the boil).  In Figure 1, the red line shows theoretical IBU values predicted by scaling the nonIAA concentration by a factor of 2.0.  In this case, every IBU value is simply increased by 5, because the concentration of nonIAA doesn’t vary with boil time.

Figure 1.  Hypothetical IBU values based on (a) Peacock’s model (black line), (b) IBU scaling with a factor of 1.20 (blue line), (c) [IAA] scaling with a factor of 1.25 (dashed green line), (d) [nonIAA] scaling with a factor of 2.0 (red line).  The IBUs predicted when [IAA] is zero are shown with a dotted gold line.

2. Experimental Overview and Methods
2.1 Overview
Five experiments were conducted to look at the relative difference in IBUs between hop cones and pellets.  Within each experiment, two batches of beer were designed to be identical in all respects, except for the use of hop cones in one case (referred to as cones) and hop pellets in the other (referred to as pellets).  The five experiments looked at (a) three varieties of hops, (b) the impact of krausen, and (c) the age of the pellets.

In all experiments, the alpha-acid rating of the cones and pellets was comparable, and adjusted when necessary to yield the same concentration of approximately 170 ppm of alpha acids at the start of the hop addition.  For each batch, I took samples of wort at 10-minute intervals and quickly cooled them in an ice bath.  Each sample was fermented into beer and sent to Oregon BrewLab for IBU analysis.

This set of experiments yielded 37 pairs of IBU values, with the values within a pair being directly comparable in terms of hop variety, boil gravity, initial alpha-acid concentration, boil time, and fermentation conditions.

The first experiment used Citra hops, the second and third used Willamette hops, and the fourth and fifth experiments used Comet hops.  The Comet pellets were very fresh in the fourth experiment and about 2½ years old in the fifth experiment.

2.2 Procedures Common to All Experiments
Each experiment consisted of two batches brewed on the same day.  I used as large a batch size as I dared in my 10 G (38 l) kettle, in order to minimize the effect of measurement errors and evaporation rate.  I used 7.0 lbs (3.18 kg) of Briess Pilsen DME in 8.0 G (30.28 l) of water, yielding about 8.5 G (32 l) of wort with a specific gravity of about 1.037.  I did not adjust the water profile or pH; the local water here in Portland, Oregon has relatively low alkalinity and hardness.  This resulted in a pre-boil wort pH of about 5.70 to 5.80.

I added hops (i.e. started the steep time at 0) after the wort had been boiling for 5 minutes, to avoid the foam associated with the start of the boil.  The hops were boiled for a steep time of 60 to 90 minutes with the cover on the kettle (except for taking samples) to minimize evaporation and the resulting changes in specific gravity.  I did not use a mesh bag with the cones, because I think that it is more standard practice to have the hops freely floating in the wort.  I targeted an initial alpha-acid concentration of 170 ppm in order to not exceed the solubility limit of approximately 200 ppm at boiling, using an estimated volume of about 8.28 G (31.36 l) when adding the hops and the experiment-specific alpha-acid (AA) ratings.  For Experiment #1, an AA rating of about 14.1% for both cones and pellets translated into a hop addition of 1.333 oz (37.81 g).  For Experiments #2 and #3, an AA rating of about 5.05% for both cones and pellets translated into an addition of 3.724 oz (105.57 g).  For Experiment #4, an AA rating of about 10.0% for both cones and pellets translated into an addition of 1.880 oz (53.31 g).  For Experiment #5, an AA rating of 9.70% for cones and 8.76% for pellets translated into additions of 1.939 oz (54.96 g) and 2.147 oz (60.86 g), respectively.

Samples were taken every 10 minutes from the start of steeping.  Each sample (about 15 oz (0.43 l)) was taken from the boil in a measuring cup and then transferred to an aluminum cup using a wire mesh sieve to remove larger hop particles.  The aluminum cup was placed in an ice bath and the contents were stirred to cool quickly.  Samples were cooled below 140°F (60°C) within about 45 seconds.  Once cooled to 75°F (24°C), the sample was transferred to a sanitized, sealed, and labelled quart (liter) container.  I aerated each sample by vigorous shaking for 60 seconds, then added about .01 oz (0.28 g) of Safale US-05 yeast to target 750,000 viable cells per ml and degree Plato [Fix and Fix, p. 68].  (The process of taking a sample, cooling it, transferring it to a sanitized container, aerating, and pitching yeast took between 5 and 10 minutes.)  (For the “cones” condition in Experiment #5, 32-oz (0.95-liter) samples were taken and transferred into 1.6 quart (1.5 liter) sanitized containers for fermentation with 0.017 oz (0.48 g) of Safale US-05 yeast.)  After all samples were taken, the containers were cracked open to vent, and they fermented for nine to ten days.  For every experiment except Experiment #2, I swirled the samples every day to remove most of the krausen deposits on the sides of the containers (mixing the krausen back into the beer).  For Experiment #2, I let krausen deposits accumulate on the sides of the containers.  After fermentation, I sent 4 oz (0.12 l) of each sample to Oregon BrewLab for IBU measurement.

2.3 Hops in Experiment #1
The hop cones in Experiment #1 were Citra from Hops Direct.  The pellets were Citra from Yakima Valley Hops.  Both were from the same harvest year (2017), and were about 3 months old at the time of the experiment.  I purchased both the cones and the pellets soon after they became available and stored them in my freezer until the experiment.  I sent samples to both Alpha Analytics and Brew Laboratory for analysis within 3 weeks of the experiment.  Alpha Analytics used the spectrophotometric method ASBC 6A; Brew Laboratory used high-performance liquid chromatography (HPLC).  The package ratings and analysis results are listed in Table 1.  It can be seen that the analysis results are very consistent with the package ratings, except for the pellets result from Alpha Analytics.  Verzele and De Keukeleire note that “there are easily differences up to 15-20% in alpha acids content between and within bales of a single hop delivery” [Verzele and De Keukeleire, p. 331], and so even this “outlier” (13.3%) is well within the expected variation.  Because of the small number of samples, it is more appropriate to take the median than the mean for a representative value of the alpha acids.  Therefore, the alpha-acid rating on brew day was about 14.2% for cones and 14.0% for pellets.

	Cones: Package Rating	Cones: Alpha Analytics	Cones: Brew Laboratory	Pellets: Package Rating	Pellets: Alpha Analytics	Pellets: Brew Laboratory
alpha acids	14.3%	14.2%	14.1%	14.0%	13.3%	14.0%
beta acids	N/A	3.6%	3.4%	N/A	3.9%	3.8%
HSI	N/A	0.265	N/A	N/A	0.293	N/A

Table 1. Results of hops analysis for Experiment #1, including alpha acids, beta acids, and (where available) the Hop Storage Index (HSI).

2.4 Hops in Experiments #2 and #3
In Experiments #2 and #3, I used Willamette hops from Yakima Chief Hops.  The cones  were from lot PR2-ZKUWIL5041 and the pellets were from lot P92-ZKUWIL5170, both about 2½ years old at the time of the experiment.  Analysis was performed by Brew Laboratory within two weeks of the experiment.  The package ratings and analysis results are listed in Table 2.  The alpha-acid rating on brew day was about 5.0% for both cones and pellets.

The reason for conducting Experiment #3 was that the results from Experiment #2 were so surprising to me (see Section 3) that I wanted to replicate the results.  In addition, in Experiment #2 I did not remove krausen deposits by swirling, and in Experiment #3 I made sure that fermentation conditions were the same as in Experiments #1, #4, and #5.

	Cones: Package Rating	Cones: Brew Laboratory	Pellets: Package Rating	Pellets: Brew Laboratory
alpha acids	5.0%	5.0%	4.8%	5.1%
beta acids	3.8%	3.1%	4.0%	3.2%
HSI	0.252	N/A	0.298	N/A

Table 2. Results of hops analysis for Experiment #2, including alpha acids, beta acids, and (where available) the Hop Storage Index (HSI).

2.5 Hops in Experiments #4 and #5
In Experiment #4, I used Comet hops from Hops Direct.  (The customer service representative at Hops Direct was very helpful, and they were able to fulfill my request for both hop cones and pellets at close to 10% AA from the most recent (2018) harvest.) These hops were stored in vacuum-sealed packaging in my freezer.  Analysis was performed by Advanced Analytical Research (AAR Lab) within one week of the experiment.  I used an AA rating of 10.0% for both cones and pellets as the best estimates at the time of the experiment.

	Cones: Package Rating	Cones: AAR Lab	Pellets: Package Rating	Pellets: AAR Lab
alpha acids	9.9%	10.8%	10.0%	9.84%
beta acids	N/A	3.92%	N/A	3.69%
HSI	N/A	0.25	N/A	0.33

Table 3. Results of hops analysis for Experiment #4, including alpha acids, beta acids, and (where available) the Hop Storage Index (HSI).

In Experiment #5, conducted four months later, I used the same hop cones, but Comet pellets from YCH Hops (lot P92-ZLUCOM5216) that were 2½ years old at the time of the experiment.  Analysis was performed by AAR Lab with three weeks of the experiment.  Because of the age of the hops, I used the analysis results from AAR Lab (9.70% for cones and 8.76% for pellets) as the best estimates for the AA ratings at the time of the experiment.

	Cones: Package Rating	Cones: AAR Lab	Pellets: Package Rating	Pellets: AAR Lab
alpha acids	9.9%	9.70%	9.5%	8.76%
beta acids	N/A	3.17%	4.3%	3.22%
HSI	N/A	0.35	0.326	0.42

Table 4. Results of hops analysis for Experiment #5, including alpha acids, beta acids, and (where available) the Hop Storage Index (HSI).

3. Results
The estimated room-temperature volume at the start of steeping was 8.28 G (31.36 liters) for all conditions and all experiments.  The average specific gravity after 10 minutes of steeping was 1.0384 (minimum 1.0378, maximum 1.0392).  The specific gravity after a 90-minute steep time was about 1.0404.  The small change in specific gravity during the boil (due to keeping the lid on the kettle) means that there is little difference between using the measured IBU values for analysis or normalizing these IBUs by the volume when the sample was taken.  For simplicity and clarity, the measured IBU values are used below.

Figures 2, 3, and 4 show the measured IBU values from Experiments 1 through 5.  The average difference in IBUs between cones and pellets is provided in each figure.

Figure 2. Measured IBU values for Citra cones and pellets. The average difference is 5.2 IBUs.

Figure 3.  Measured IBU values for Willamette cones and pellets, in two separate experiments.   The average difference in Experiment #2 is 1.7 IBUs, and the average difference in Experiment #3 is 1.3 IBUs.

Figure 4.  Measured IBU values for Comet cones and pellets, in two separate experiments.  In Experiment #4, both cones and pellets were recently harvested.  In Experiment #5, the pellets were 2.5 years old at the time of the experiment.  The average difference in Experiment #4 is 10.6 IBUs, and the average difference in Experiment #5 is 11.0 IBUs.

In Experiment #2, krausen was allowed to build up on the sides of the fermentation vessels, which  explains the overall lower IBU values when compared with Experiment #3.  (Another blog post looks at the impact of krausen on IBUs; it finds that krausen that adheres to the sides of the fermentation vessel can cause a significant decrease in IBUs.)

The increase in IBUs in Experiment #5 (compared with Experiment #4) may have been caused by the greater weight of hops used in this experiment.  A greater weight of hops in the same volume was used to target the same initial alpha-acid concentration of 170 ppm.  This may have resulted in greater IBU values because (a) the estimated decrease in alpha-acid content over time was greater than the actual decrease, and so the greater weight of hops over-compensated for the decrease in AA levels, (b) variation in AA levels in the hops, (c) the greater weight of hops increased the concentration of nonIAA compounds, thereby increasing IBU levels, or (d) some combination of all of these reasons.

4. Analysis
4.1 Visual Analysis of the Figures
It is easily seen in Figures 2, 3, and 4 that the increase in IBUs from the use of pellets is closer to the pattern associated with nonIAA scaling in Figure 1 than to the pattern of IAA or IBU scaling.  This constant offset is difficult to explain as a relative increase in IBUs or IAA (as illustrated in Figure 1), but very easy to explain as a relative increase in nonIAA concentration.  This may explain why Noonan used different utilization factors for cones and pellets at different steep times [Noonan, p. 215], resulting in a roughly constant increase for pellets regardless of steep time.

It is also clear that the increase in IBUs changes with the use of different hop varieties.  There is an average increase of 5.2 IBUs, 1.5 IBUs, and 10.8 IBUs for Citra, Willamette, and Comet pellets, respectively.  Within a variety, the increase in IBUs from cones to pellets is quite similar.  This topic is discussed more in Section 4.3.  Across varieties, the cone IBU values are much more similar than the pellet IBU values.  For example, at a 10-minute steep time, the cone IBU values are 14.0, 13.9, 14.8, 14.2, and 16.4 (standard deviation 0.9 IBUs) for Experiments #1 through #5, respectively, while the pellet IBU values are 18.7, 16.6, 16.3, 24.0, and 26.0 (standard deviation 4.0 IBUs).

4.2 Modeling Analysis
We can use the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to split IBU values into estimates of (a) the concentration of IAA and (b) the concentration of other bitter substances measured with the IBU that are called nonIAA.  In brief, we can use multiple IBU values from the same batch of beer, along with (a) the equation in Section 1.3 that describes the isomerization of alpha acids as a function of time and temperature [Malowicki, p. 27] and (b) the equation in Section 1.4 that describes the IBU as a combination of IAA and nonIAA in the finished beer [Peacock, p. 161], in order to estimate two scaling factors: scaling_IAA and scaling_nonIAAhops.  The scaling_IAA parameter is the scaling factor that accounts for losses of IAA during the boil, fermentation, and aging; scaling_nonIAAhops is the scaling factor from concentration of total hop particles in the wort to the concentration of hop-related nonIAA in the beer (excluding malt-related nonIAA).  With scaling_IAA and scaling_nonIAAhops, as well as the weight of the hops, initial alpha-acid concentration, steep time, and original gravity, we can map from IBU value to IAA and nonIAA concentrations, and vice versa.

A separate blog post investigates the reason for the increase in IBUs associated with hop pellets, and concludes that this increase in IBUs is most likely caused by an increase in the concentration of oxidized alpha acids produced when the hops are added to the kettle.  Using models that predict IBUs due to malt polyphenols, hop polyphenols, and oxidized beta acids, we can change the scaling_nonIAAhops parameter from a single parameter estimating the combined effect of all hop-related auxiliary bittering compounds to a parameter estimating the effect of only oxidized alpha acids, scaling_oAA.

By searching over a large number of values of scaling_IAA and scaling_oAA to minimize the error on the cones batch of IBU values in Experiment #1, we get scaling_IAA = 0.417 and scaling_oAA = 0.057.  These results indicate that somewhat less than half of the isomerized alpha acids from this batch made it into the finished beer, and about 6% of the alpha acids were oxidized and survived into the finished beer.  These scaling factors yield a root-mean-square (RMS) error of 0.77 IBUs on the nine IBU values, with a maximum difference of -1.39 IBUs at 90 minutes.  We can do the same search for scaling_IAA and scaling_oAA using the set of nine values of pellets IBU data from Experiment #1.  In this case, we get scaling_IAA = 0.406 and scaling_oAA = 0.109, with an RMS error of 0.74 IBUs and a maximum difference of -1.42 IBUs at 50 minutes.  These results indicate that nearly the same percentage of IAA were produced and made it into the finished beer in both the cones and the pellets batches (i.e. hop utilization was the same in both cases), but that the concentration of oxidized alpha acids nearly doubled in the pellets batch.

Table 5 lists the IAA scaling factor (scaling_IAA) for both cones and pellets in the five experiments.  It can be seen that the IAA scaling factor is very similar between cones and pellets for all five experiments, slightly higher in some cases and slightly lower in other cases.  (The average cone-to-pellet IAA ratio is 1.05.)  These small differences are probably due to measurement error, and it seems most likely that the IAA scaling factor is basically the same for both cones and pellets.  (The IAA scaling factor in Experiment #2 is expected to be lower than all of the others because the krausen was not mixed back into the beer in this experiment, resulting in greater loss of both IAA and oAA.)

	Exp. #1	Exp. #2	Exp. #3	Exp. #4	Exp. #5
IAA scaling factor: cones	0.417	0.339	0.416	0.461	0.472
IAA scaling factor: pellets	0.406	0.307	0.359	0.465	0.476

Table 5. IAA scaling factors for cones and pellets in each experiment.  These values were estimated using the model described in Section 4.2.

We can then set the IAA scaling factor within each experiment to be the average of the IAA scaling factors for cones and pellets, and re-estimate the oAA scaling factors.  Table 6 shows the new estimates for IAA scaling factors (scaling_IAA) and oAA scaling factors (scaling_oAA).  Table 7 shows the measured IBU values and estimated IBU values using the model and scaling factors from Table 6.  The RMS errors are as follows: Exp #1 cones: 0.78, pellets 0.75; Exp #2 cones: 0.44, pellets 0.39; Exp #3 cones: 1.16, pellets: 1.17; Exp #4 cones: 0.91, pellets: 0.58; Exp #5 cones: 0.43, pellets: 0.62. The RMS error over all experiments is 0.80 IBUs.  Note that the average oAA scaling factor for cones estimated here (6.3%) is close to the value estimated in Section 8.2 of The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU (5.9%).

	Exp. #1	Exp. #2	Exp. #3	Exp. #4	Exp. #5
IAA scaling factor	0.4115	0.323	0.3875	0.463	0.474
oAA scaling factor: cones	0.059	0.072	0.071	0.046	0.066
oAA scaling factor: pellets	0.107	0.088	0.084	0.148	0.150

Table 6. Averaged IAA scaling factor and oAA scaling factors for cones and pellets in each experiment.

	10 min	20 min	30 min	40 min	50 min	60 min	70 min	80 min	90 min
Exp 1: cones (meas., est.)	14.0, 13.8	18.8, 19.1	24.2, 23.7	26.6, 27.6	31.6, 31.0	32.8, 33.8	36.3, 36.3	37.9, 38.3	41.6, 40.0
Exp 1: pellets (meas., est.)	18.7, 19.1	24.2, 24.3	29.7, 28.8	32.1, 32.8	37.5, 36.2	39.5, 39.0	41.5, 41.5	42.4, 43.5	44.7, 45.3
Exp 2: cones (meas., est.)	13.9, 14.1	17.5, 18.2	22.3, 21.8	25.0, 24.9	27.3, 27.6	30.3, 29.8
Exp 2: pellets (meas., est.)	16.6, 15.8	19.9, 19.9	23.2, 23.5	26.4, 26.5	29.4, 29.2	31.0, 31.4
Exp 3: cones (meas., est.)	14.8, 14.9	19.1, 19.9	23.7, 24.1	26.7, 27.8	30.4, 30.9	35.1, 33.6	34.8, 35.9	40.1, 37.8
Exp 3: pellets (meas., est.)	16.3, 16.3	21.5, 21.2	27.6, 25.4	28.3, 29.0	33.2, 32.2	34.7, 34.8	35.1, 37.1	38.5, 39.0
Exp 4: cones (meas., est.)	14.2, 13.5	19.7, 19.6	25.1, 24.8	28.6, 29.3	31.7, 33.2	35.9, 36.5	39.6, 39.3	43.3, 41.7
Exp 4: pellets (meas., est.)	24.0, 24.5	31.4, 30.4	34.6, 35.6	40.2, 40.0	44.2, 43.8	46.7, 47.0	49.4, 49.8	52.4, 52.1
Exp 5: cones (meas., est.)	16.4, 15.7	21.2, 21.7	26.6, 26.9	31.3, 31.4	35.2, 35.3	39.0, 38.6
Exp 5: pellets (meas. est.)	26.0, 26.6	33.6, 32.7	37.8, 38.0	41.6, 42.5	46.9, 46.5	49.9, 49.8

Table 7. Measured and estimated IBUs for each sample in each experiment. Samples are identified by the duration of hop steeping, in minutes (column headings). Experiments and condition (cones or pellets) are identified by row headings. Each cell in the table shows measured IBUs followed by estimated IBUs. Estimates are from the model described in Section 4.2.

The change in oAA factor between cones and pellets for each experiment is listed in Table 8, expressed as a ratio of pellets/cones.  It can be seen that these factors vary from an 18% to 222% increase for pellets compared with cones, and that this increase is approximately the same within a hop variety but different between varieties. For example, for Willamette the ratios 1.22 and 1.18 are very similar, and for Comet the ratios 3.22 and 2.27 are more similar to each other than they are to the ratios of other varieties.  The average ratios for each variety are 1.81, 1.20, and 2.74 for Citra, Willamette, and Comet, respectively.  Over all three varieties, the pellet-to-cone ratio is 1.9, representing an approximate doubling in the concentration of oxidized alpha acids in the finished beer.

	Exp. #1	Exp. #2	Exp. #3	Exp. #4	Exp. #5
oAA pellet-to-cone ratio	1.81	1.22	1.18	3.22	2.27

Table 8. oAA pellet-to-cone ratios estimated for the five experiments. This ratio expresses the relative increase in oxidized alpha acids that contribute to the observed increase in IBUs with the use of hop pellets.

5. Predicting an Increase in IBUs
5.1 Variety-Specific Factors
The results from Sections 3 and 4 indicate that the increase in IBUs and oAA concentration that results from using hop pellets is dependent on the hop variety. We can check if any quantitative descriptions of these varieties might allow us to predict the amount of increase in oxidized-alpha acid concentration.

Table 9 lists a variety of quantitative descriptions of the three varieties used here. The alpha-acid and beta-acid levels are taken from the averages of cones and pellets in Section 2, and the other descriptions are taken from The Hops List [Healey]. Each cell shows the typical composition (in percent or ml/100g) and the approximate concentration used in these experiments. If a descriptor is associated with the oAA pellet-to-cone ratio, we would expect a correlation between the concentration of this descriptor and the oAA pellet-to-cone ratio for this variety. In other words, we are looking for concentration values that increase in order from Willamette to Citra to Comet. None of these descriptors show such a trend, meaning that we can not currently predict the oAA pellet-to-cone ratio from knowledge of the hop variety or characteristics.

	alpha acid (%)	beta acid (%)	cohumulone	total oil	storability
Citra	14.1% 170 ppm	3.67% 44 ppm	27.5% 47 ppm	2.25 ml/100g 3% v/v	75%
Comet	10.0% 170 ppm	3.5% 60 ppm	41% 70 ppm	1.98 ml/100g 3% v/v	49%
Willamette	5.0% 170 ppm	3.5% 120 ppm	32.5% 55 ppm	1.25 ml/100g 4% v/v	62%

Table 9. Quantitative descriptions of the three hop varieties used in these experiments. The descriptions are provided as both typical composition (in percent or ml/100g) and approximate concentration in these experiments (in ppm or %v/v). The exception is “storability,” which is the percent of alpha acids remaining after storage for six months at room temperature.

Another possibility is that there is a transformation (other than oxidation) which happens while pellets age in their nitrogen-flushed packaging, and this hypothetical transformation causes less of a pellet-based IBU increase with older hops. The purpose of Experiment #5 was, in fact, to test this hypothesis. In Experiment #1, the Citra pellets were about 3 months old at the time of the experiment, and the pellet-based IBU increase was moderate. In Experiments #2 and #3, the Willamette pellets were several years old at the time of the experiment, and the pellet-based IBU increase was minor. In Experiment #4, the Comet pellets were extremely fresh and the increase was quite large. Therefore, Experiment #5 used Comet pellets that were deliberately several years old at the time of the experiment. If the results of Experiment #5 showed an increase in IBUs similar to that of the Willamette hops, then this hypothesis of older pellet hops having less increase would have been supported. However, the results showed just as large an increase in IBUs as in Experiment #4, indicating that the age of the (properly stored and nitrogen flushed) pellets has no impact on the increase in IBUs.

This leaves us with measuring the variety-specific ratio for each variety of hops. With hundreds of available hop varieties (e.g. [Healey]), this is a nearly impossible task. The more practical but less accurate approach is to treat all hop varieties as having the same increase as the average of the three varieties studied here, i.e. an oAA pellet-to-cone ratio of about 1.9.  (It is also possible that there is no variety-specific increase, but that the differences in the ratios are due to differences in the pellet-production process at each manufacturer.  Checking this hypothesis would require further study of both variety-specific and manufacturer-specific pellets.)

5.2 Modeling an Increase in IBUs
As seen in Table 6, when using hop cones, about 6% of the alpha acids added to the wort are oxidized and survive into the finished beer.  For pellet hops, about 12% of the alpha acids added to the wort are oxidized and survive into the beer.  There is a factor of 0.9155 for scaling the light absorption at 275 nm from oxidized alpha acids to isomerized alpha acids, as seen in Figure 7 of Maye et al. [Maye, p. 25], and a scaling factor of 51.2/69.68 to convert the light absorption of isomerized alpha acids to IBUs [Peacock, p. 161].  Therefore, 1 ppm of oxidized alpha acids will produce 0.67 IBUs.  Let’s consider an example to see how to model the use of pellet hops.  If we have a beer made with 1.50 oz (42.52 g) of 10% AA hops boiled for 60 minutes in 5.50 gallons (20.82 liters) of wort (and ignoring evaporation), when we add the hops we have 204 ppm of alpha acids added to the wort (204 ppm = 0.10 × 42.52 g × 1000 / 20.82 l).  From the 0.150 oz (4.252 g) of alpha acids added to the wort, with hop cones we get 0.009 oz (0.2551 g) of oxidized alpha acids, or 12.25 ppm, in the finished beer (12.25 ppm = 4.252 g × 0.06 × 1000 / 20.82 l), increasing the IBU by 8.24 (0.673 × 12.25 ppm).  With hop pellets, we get 0.018 oz (0.5102 g) of oxidized alpha acids, or 24.51 ppm, increasing the IBU by 16.48.  These oxidized-alpha-acid IBUs are in addition to the IBUs from isomerized alpha acids (e.g. 30 IBUs) and the IBUs from malt and hop polyphenols (e.g. 2 IBUs), resulting in 40 IBUs for cones and 48 IBUs for pellets.  In this example, then, pellets demonstrate a 20% increase in IBUs compared with cones.

6. Summary and Conclusion
The IBU data from these five experiments showed an unexpected but consistent pattern: the increase in IBUs from pellets is constant over a range of steep times, instead of increasing with steep time.  It therefore seems that the increase in IBUs when using pellets is not caused by an increase in the rate of isomerization or availability of alpha acids, and should not be modeled with a multiplication factor applied to [IAA] or IBUs. Instead, this increase in IBUs can be modeled by an increase in the concentration of oxidized alpha acids produced during the boil, as discussed in a separate blog post.  The amount of increase appears to be dependent on the hop variety and is not easily predicted from characteristics within each variety.  Therefore, the most practical way to model this increase in IBUs is to treat the isomerization of alpha acids in the same way as hop cones, but to double the concentration of oxidized alpha acids ending up in the finished beer.

7. Acknowledgement
I greatly appreciate the high-quality IBU analysis provided by Dana Garves at Oregon BrewLab. This accuracy can be seen in the smooth and consistent shape of the IBU plots in Figures 2, 3, and 4.  Without such consistent accuracy, it would not be possible to draw meaningful conclusions from the data.

References

• R. Daniels, Designing Great Beers: The Ultimate Guide to Brewing Classic Beer Styles.  Brewers Publications, 2000.
• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques. Brewers Publications, 1997.
• M. Garetz, Using Hops: The Complete Guide to Hops for the Craft Brewer. HopTech, 1st edition, 1994.
• M. L. Hall, “What’s Your IBU,” in Zymurgy.  Special Edition, 1997.
• J. Healey, The Hops List: 265 Beer Hop Varieties From Around the World. Healey, 1st edition, 2016.
• S. Hieronymus, For the Love of Hops: The Practical Guide to Aroma, Bitterness, and the Culture of Hops.  Brewers Publications, 2012.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• J. P. Maye, R. Smith, and J. Leker, “Humulinone Formation in Hops and Hop Pellets and Its Implications for Dry Hopped Beers,” in Master Brewers Association of the Americas Technical Quarterly, vol. 53, no. 1, pp. 23-27, 2016.
• G. J. Noonan, New Brewing Lager Beer. Brewers Publications, 1996.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids.  Developments in Food Science 27.  Elsevier, 1991.

How Lautering and Wort Clarity Affect IBUs

Abstract
Lautering is the process of separating wort from spent grains.  Two of the three experiments described here look at wort produced using different lautering techniques and how the resulting wort characteristics impact IBUs.  The third experiment looks at the impact of wort clarity after lautering.  The results indicate that very turbid (cloudy) wort can yield about 30% fewer IBUs than average, and very clear wort (free from smaller particles) can yield about 30% more IBUs than average.  A 10-minute rest period to allow hot and cold break to settle before transfer into the fermentation vessel can have an 8% impact on IBUs, suggesting that it’s not the lautering technique itself but the final clarity of the wort going into the fermentation vessel that affects IBUs.  The reason for the difference in IBUs may be that isomerized alpha acids (IAA) bind to small particles in the wort during fermentation and precipitate out of solution.  A proposed model of the impact of wort clarity on IAA uses a linear scaling factor with a value of 1.30 for very clear wort, 1.0 for average wort, and 0.70 for very cloudy wort.  While the results indicate that clear wort yields more IBUs than cloudy wort, the adage that “clear wort produces clear beer” was not confirmed.

1. Introduction
A number of factors are typically listed as having an impact on IBUs and/or isomerized alpha acids, including alpha-acid concentration, boil time, wort temperature, hopping rate, wort pH, form of the hops (e.g. cones or pellets), and wort gravity.  Lewis and Young state that “iso-alpha-acids react with proteins of wort whence they are partially removed as trub or hot break” [Lewis & Young, p. 266], implying that a wort higher in protein might result in reduced isomerized alpha acid concentration and therefore lower IBUs.  In the first experiment described in this blog post, I was looking for the effect that wort protein levels might have on IBUs.  Because the teig layer on top of the grain bed that results from recirculation is high in proteins [Lewis & Young, p. 216, p. 247], I thought that it would be interesting to compare protein levels from lautering with and without recirculation, and to compare the IBUs from these different worts when using the same amount of hops.  When this experiment (described in Section 3) did not show the expected results, but other preliminary data indicated the possible impact of wort clarity instead of proteins, I performed the second experiment with lautering techniques that produced different levels of wort clarity, and the third experiment with the same post-boil wort but different levels of clarity going into the fermentation vessel. When looked at as a single set of data, the first two experiments provide six examples of wort lautered in five different ways, producing wort of varying clarity.  These experiments present the first data I’m aware of that look at the impact of lautering technique and wort clarity on IBUs.

2. Terminology
2.1 The IBU
The IBU is a measure of the concentration of a number of different bitter compounds in beer.  (To be more precise, the IBU is a measure of the absorbance of light at 275 nm through acidified beer.  A number of bitter compounds in beer absorb light at this frequency.  The greater the concentration of these compounds, the more light is absorbed, and the higher the IBU.)  In typical beers, the IBU value represents mostly the concentration of isomerized alpha acids (IAAs) [Peacock, pp. 164-165], which are produced during the boil from alpha acids (AA).  The other bitter compounds, known as “auxiliary bittering compounds” (ABCs), or nonIAA, are polyphenols, oxidized alpha acids, and oxidized beta acids.  These compounds can be considered to be present in the wort soon after the hops addition [e.g. Dierckens and Verzele, p. 454; Askew, p. 18].  The majority of ABC appear to be oxidized alpha acids, which are produced when hops are added to the boiling wort [Parkin, p. 11, Algazzali, p. 17; Dierckens and Verzele, p. 454; Oliver p. 471] as well as during storage [Lewis and Young, p. 265].

2.2 Lautering
Lautering is the process of separating wort from spent grains.  There are several possible steps in lautering: mash-out, recirculation, sparging, and drawing (or run-off) of the liquid wort.  All of these steps are optional except for the drawing of wort.  If performing a mash-out, the temperature of the wort is raised to 170°F (77°C) to stop (or greatly reduce) enzymatic activity.  If performing recirculation (also called vorlauf), some of the wort is drawn out and recirculated back onto the grain bed.  The grain bed acts as a filter to remove larger particles from the wort [Oliver, p. 819].  There are two common types of sparging: fly sparging and batch sparging; it is also possible to omit this step entirely for “no sparge” wort collection. While the drawing of wort is not typically discussed much, in my experience the speed at which the wort is drawn, in combination with how well the grain bed filter has set, can have a large effect on wort clarity.  There is an adage that “clear wort makes clear beer“, and so some brewers aim for the clearest wort possible.

2.3 Brew-in-a-Bag (BIAB)
The Brew-in-a-Bag (BIAB) technique is one method of lautering.  In its most basic form, the mash is performed while the grains are inside a mesh bag; when the mash is complete, the bag is removed (with the grains), leaving the wort behind.  There are several variations on this technique, including with and without mash-out, and with and without sparging.  The standard BIAB approach is essentially lautering with an optional mashout, no recirculation, no sparge, and fast drawing of the wort.

2.4 Teig and Wort Proteins
Recirculation and/or continuous sparging can form a layer of teig (“top dough”) on the top of the grain bed [Palmer, p. 306].  This teig layer is high in proteins [Lewis & Young, p. 216, p. 247].  Because the teig layer is separate from the liquid wort, wort that is produced leaving a layer of teig on the grain bed should have a lower concentration of proteins than wort in which no teig layer is formed.

2.5 Wort Clarity
While there are instruments for measuring turbidity, I judged wort clarity/turbidity in these experiments on an entirely subjective basis.  (Tubidity meters do exist but are not cheap.)  If I were to re-do these experiments, I would more strongly consider investing in such a meter.

3. Experiment #1
3.1 Experiment #1: Experimental Overview
The intended difference between conditions in the first experiment was the level of protein in the wort, which was expected to differ according to the amount of teig on the grain bed, which in turn was controlled by lautering with and without recirculation.

This first experiment had three conditions.  Condition A used wort produced with mash-out, without recirculation, and no sparge (in a manner similar to BIAB). Condition B used wort produced with mash-out, with recirculation, and no sparge.  Condition C used wort produced from Briess Pilsen Light Dried Malt Extract (DME).  The intended difference between Conditions A and B was a teig layer in Condition B that would decrease the concentration of proteins in the wort relative to Condition A.  Condition C was used as a reference condition, as most of the experiments conducted for this blog have used wort produced using DME.

For lautering Conditions A and B, I used a 13 G (49 l) picnic-cooler lautering vessel with a 12″ (30 cm) mash screen connected to a ball valve and tubing.  To prevent very large particles from ending up in the wort of Condition A, I covered the mash screen with a fine mesh bag.   In Condition B, I performed the recirculation for 10 minutes, gently recirculating the wort to the other end of the lautering vessel.  In both Condition A and Condition B, I kept the ball valve approximately one-third open, allowing the wort to flow at a moderate rate.

In other respects, all three conditions were intended to be identical, having the same target specific gravity, volume, and hops.  I did not expect the wort pH to differ significantly between conditions, and so I did not control for pH.

3.2 Experiment #1: Methods
For Conditions A and B, 7.06 lbs (3.20 kg) of two-row malt was added to 4.87 G (18.42 l) of low-alkalinity water in a 10 G (38 l) kettle.  The crushed grains were heated to 153°F (67°C) and this temperature was held for 60 minutes.  The kettle was then heated to 170°F (77°C) for 10 minutes.  The mash was transferred to the lauter tun, and for Condition B a 10-minute recirculation step was performed.  The wort was drawn into the boil kettle with a moderate flow to collect 3.58 G (13.55 l) of wort at ~160°F (~70°C), corresponding to a room-temperature volume of 3.50 G (13.25 l).

For Condition C, 3.58 lbs (1.62 kg) of Briess Pilsen Light DME was added to 3.26 G (12.36 l) of 120°F low-alkalinity water to target 3.54 G (13.40 l) of wort at 120°F (49°C), corresponding to a room-temperature wort volume of 3.50 G (12.25 l).  The wort then sat for 2 hours to let the pH stabilize.

For all conditions, the wort was brought to a boil, a 33-oz (0.98-liter) sample of wort was taken (the “pre-boil” sample), the wort was boiled for 5 minutes, and 0.72 oz (20.4 g) of Comet hops were added.  These hops were analyzed shortly after harvest and found to have an AA rating of 10.8%.  The hops were 7.5 months old at the time of the experiment, and so the estimated alpha-acid rating at the time of the experiment was 9.73% using the Garetz formula [Garetz].  The volume of wort and amount of hops were designed to yield an alpha-acid concentration of 170 ppm when the hops were added to the kettle.  The kettle was covered during the boil (except for the initial 5 minutes and for taking samples) in order to minimize evaporation and the resulting changes in volume and specific gravity.  A 15-oz (0.44-liter) sample was taken every 10 minutes after the hop addition, for a total of 40 minutes (4 samples per condition).  Each sample was quickly cooled in an aluminum cup and ice bath to 75°F (24°C) and then transferred to a sanitized quart (liter) container.  The wort in each container was aerated for 1 minute by vigorous shaking, and 0.0085 oz (0.24 grams) of Safale US-05 yeast (age 9 months) was pitched to target 750,000 cells per ml and degree Plato.  At the end of the 40-minute boil, another sample was taken (the “post-boil” sample) and cooled to room temperature.

The 10, 20, 30, and 40-minute samples fermented for about one week (with a small opening to vent CO₂).  The pre- and post-boil samples were stored in sanitized and refrigerated sample containers during this time.  The krausen of the fermenting samples was left to deposit on the sides of the vessel during fermentation.  I removed the krausen deposits one day before taking samples for analysis by Oregon BrewLab.  Oregon BrewLab analyzed the unfermented pre- and post-boil samples for protein levels, the 40-minute sample from Condition C for protein, and all fermented samples for IBUs.

3.3 Experiment #1: Results
The subjective clarity of wort increased between Conditions A, B, and C.   The wort from Condition A was very cloudy, as expected; perhaps the word “murky” would be a good descriptor.  Despite my attempt at using recirculation to produce clear wort, Condition B was judged probably cloudier than Condition C.

The specific gravity at the start and end of the boil was nearly the same for all three conditions, with a starting gravity of 1.045 and a post-boil gravity of 1.046 (Conditions A and C) or 1.047 (Condition B).  The pre-boil pH, however, was unexpectedly low for the all-grain conditions, at 5.55 and 5.54 for Conditions A and B, respectively; the DME condition had a more expected pre-boil pH (given the specific gravity) of 5.74.

The protein levels at the start of the boil were 2.4, 2.3, and 2.7 g/12oz for Conditions A, B, and C, respectively.  The post-boil protein levels were 2.2, 2.4, and 2.9 g/12oz for Conditions A, B, and C.  The protein level of the finished beer of Condition C was 2.5 g/12oz.  The protein levels between Conditions A and B are nearly identical, and certainly not the difference that was expected.

The IBU values are plotted in Figure 1.  It can be seen that Condition A has significantly lower IBU values than Condition B, and Condition B has lower values than Condition C.  For the 40-minute samples, Condition C (34.8 IBUs) has 55% more IBUs than Condition A (22.4 IBUs) .

Figure 1. Measured IBU values from the first experiment.  Condition A used lautering with no recirculation and a medium rate of drawing wort, Condition B used lautering with recirculation and a medium rate of drawing wort, and Condition C used dried malt extract.

4. Experiment #2
4.1 Experiment #2: Experimental Overview
While Experiment #1 did not show the expected change in protein levels, it did show a remarkable difference in IBU levels.  Some additional preliminary experiments (not described here) suggested that rather than protein, the wort clarity might be correlated with IBUs.  The purpose of Experiment #2 was to obtain additional data for evaluating the relationship between wort clarity and IBUs.  By the time I conducted Experiment #2, I was able to successfully produce very clear wort by resting the mash for 30 minutes, using a 10-minute recirculation, and drawing the wort very slowly from the lauter tun.  I again used a fine mesh bag over the mash screen to filter out large particles from the wort.

This experiment had three conditions.  Condition D used (like Condition A in Experiment #1) a mash-out, no-recirculation, no-sparge lauter, but this time I opened the ball value fully to quickly collect the wort.  Condition E used a 30-minute rest of the mash after mash-out, 10-minute recirculation, and very slow drawing of the wort (over one hour to collect less than 2 G (7.6 l) of wort), to produce a very clear wort.  Condition F used (like Condition C in Experiment #1) DME, and again serves as a point of reference with other experiments.

In other respects, all three conditions were intended to be identical, having the same target specific gravity, volume, and hops.  This time, I adjusted the pre-boil wort pH of Condition F (using phosphoric acid) to be the same as that in Conditions D and E.

4.2 Experiment #2: Methods
In this experiment, I created one high-gravity mash for Conditions D and E.  I added 16.46 lbs (7.47 kg) of two-row malt to 6.33 G (23.96 liters) of low-alkalinity water at 105°F (40.5°C).  I heated this mash in a kettle to 153°F (67.2°C) and held this temperature for 60 minutes.  Then I raised the mash temperature to 170°F (76.7°C) for 15 minutes.  I then transferred all of the mash to the lauter vessel and immediately drew 1.79 G (6.78 l) of hot wort with the ball valve completely open for fast transfer.  I added 1.75 G (6.62 l) of room-temperature water to this wort to create (room-temperature-normalized) 3.50 G (13.25 l) of wort for Condition D.  I stirred the mash remaining in the lauter vessel and let it sit for 30 minutes, followed by 10 minutes of recirculation with the ball valve open just enough to let through a steady trickle of wort.  After the recirculation, I drew the wort slowly into another boil kettle, taking over an hour to collect 2 G (7.6 l) of wort.  I decanted 1.76 G (6.66 l) of this warm wort and added 1.75 G (6.62 l) of room-temperature water to create 3.50 G (13.25 l) of wort (normalized to room temperature) for Condition E.

The wort for Condition F was created using 3.15 lbs (1.43 kg) of DME with 3.30 G (12.49 l) of 120°F (49°C) water, corresponding to a room-temperature wort volume of 3.50 G (13.25 l).  This wort sat for over two hours to let the pH stabilize.  I then adjusted the wort of Condition F to be the same as Conditions D and E, using phosphoric acid.

For all conditions, the wort was brought to a boil, a 12 oz (0.35 l) sample of wort was taken for measuring specific gravity and pH, the wort was boiled for 5 minutes, and 1.33 oz (37.57 g) of Cascade hops (AA rating 6.4%, analyzed soon after harvest) were added.  The hops were 18.5 months old at the time of the experiment, and so the estimated alpha-acid rating at the time of the experiment was 5.53% using the Garetz formula [Garetz].  The volume of wort and amount of hops were designed to yield an alpha-acid concentration of 170 ppm when the hops were added to the kettle, the same as in Experiment #1.  The kettle was covered during the boil (except for the initial 5 minutes and for taking samples) in order to minimize evaporation and the resulting changes in volume and specific gravity.  A 15-oz (0.44-liter) sample was taken every 10 minutes after the hop addition, for a total of 40 minutes (4 samples per condition).  Each sample was quickly cooled in an aluminum cup and ice bath to 75°F (24°C) and then transferred to a sanitized quart (liter) container.  The wort in each container was aerated for 1 minute by vigorous shaking, and 0.010 oz (0.29 grams) of Safale US-05 yeast (age 14 months) was pitched to target 750,000 cells per ml and degree Plato.  At the end of the 40-minute boil, another sample was taken for measuring specific gravity and pH.

The 12 samples fermented for about one week (with a small opening to vent CO₂).  The krausen was left to deposit on the sides of the vessel during fermentation.  I removed the krausen deposits one day before taking samples for analysis by Oregon BrewLab.  Oregon BrewLab analyzed all samples for IBUs and the 40-minute samples for both protein and polyphenol concentrations.

4.3 Experiment #2: Results
While the intent with Condition D was to create a cloudy wort, I found that by completely opening the ball valve, the initial wort was cloudy but the grain bed quickly compacted.  By the time I was collecting the second gallon of wort, the wort had become noticeably clearer.  Therefore, Condition D was cloudy (and more cloudy than Conditions E or F), but subjectively less cloudy than Condition A even though neither method used recirculation.  The wort for Condition E was, as hoped, subjectively much clearer than the wort for Conditions D and F.

The specific gravity at the start and end of the boil was nearly the same for all three conditions, with a gravity at the start of the boil of 1.041 to 1.042 and a post-boil gravity of 1.041 to 1.044.  The measured pre-boil pH was 5.69 for Condition D and 5.67 for Condition E.   The initial pH of Condition F was 5.86, and this was lowered to 5.68 using phosphoric acid.  While the pH of Conditions D and E was still lower than expected, it was much greater than the pH observed in Experiment #1 for the all-grain conditions.

The protein levels of the finished beer were 2 g/12oz for all conditions, indicating that the protein levels did not vary significantly between the conditions or between the experiments.

The polyphenol concentrations were 156, 165, and 138 mg/L for Conditions D, E, and F, respectively.  The recirculated wort therefore had a somewhat higher polyphenol concentration than the non-recirculated wort.  This difference can be attributed entirely to the higher specific gravity of Condition E (1.0436) relative to Condition D (1.0411) using a previously-developed model of malt polyphenols. (A higher-gravity wort contains not only more sugars, but also a higher concentration of malt polyphenols.)

The IBU values are plotted in Figure 2.  It can be seen that Condition D (cloudy wort) has on average somewhat lower IBU values than Condition F (DME), and that both of these conditions have much lower IBU values than Condition E (clear wort).  For the 40-minute samples, Condition E (37.5 IBUs) has 21% more IBUs than Condition D (30.9 IBUs).

In this experiment, I took note of the subjective clarity of the finished beer, expecting the clear wort to produce clear beer.  In contrast, the beer from Condition D (cloudy wort) was noticeably clearer than the other two, and the beer from Condition E (clear wort) was not noticeably clearer than the beer from Condition F (DME).

Figure 2. Measured IBU values from the second experiment.  Condition D used lautering with no recirculation and fast drawing, Condition E used lautering with recirculation and slow drawing, and Condition F used dried malt extract.

5. Experiment #3
5.1 Experiment #3: Experimental Overview
Both Experiments #1 and #2 demonstrated an impact of lautering technique on IBUs.  Because Mark Malowicki showed that the production of IAA is not greatly affected by factors such as brewing-range pH or maltose levels [Malowicki, p. 31], it is likely that the impact of lautering technique on IBUs is caused by a reduction in the levels of IAA (and/or auxiliary bittering compounds) after they have been produced.  This leaves us with the question of whether the reduction in IBUs occurs primarily during the boil or during fermentation.

The third experiment looked at the impact of wort settling on IBUs, using the same post-boil wort in all conditions.  Four samples were created with either “settled” wort or “stirred” wort, and IBUs were measured from the beers produced from these samples.  If the IBUs of the four conditions are quite similar, then we don’t need to worry about settling or wort clarity going into the fermentation vessel; the IAA (and possibly nonIAA) are already reduced during the boil.  If the IBUs are different, then the final wort clarity is more important than the specific lautering technique, and the reduction happens primarily during fermentation.

Condition G (settled wort) used a 32-oz (0.946-liter) sample of wort taken from the top of the wort at the end of the boil.  This sample sat undisturbed in an ice bath to cool and settle, after which 16 oz (0.473 liters) of clear wort were decanted for fermentation.  Condition H (stirred wort) used a 16-oz (0.473-liter) sample of wort taken after Condition G and after the hot wort had been thoroughly stirred.  Condition J (stirred wort) used a 16-oz (0.473-liter) sample of wort taken immediately after forced cooling of the full volume of wort to room temperature, and Condition K (settled wort) used a 16-oz (0.473-liter) sample taken from the top of the full volume of wort after forced cooling and a 10-minute rest.

5.2 Experiment #3: Methods
In this experiment, I created one set of wort for all samples.  I added 2.55 lbs (1.16 kg) of DME to 3.370 G (12.755 liters) of low-alkalinity water at 120°F (49°C) to yield 3.50 G (13.25 l) of room-temperature wort at specific gravity ~1.030. This wort sat for 90 minutes to let the pH stabilize, after which I adjusted the pH to 5.30 using phosphoric acid.  I heated the wort to boiling, boiled it for 5 minutes to reduce the foam associated with the start of the boil, and took a 12-oz (0.355-liter) sample for measuring specific gravity and pH.  I then added 1.24 oz (35.13 g) of Cascade hops with an average AA rating at harvest of 7.7% and estimated decay factor of 0.80, targeting about 180 ppm of alpha acids at the start of the steep.  The hops were contained in a coarse mesh bag and boiled in the wort for 40 minutes.

At the end of the boil, 32 oz (0.946 liters) were taken from the top of the kettle and set aside in an ice bath to both cool and settle for Condition G.  The wort was stirred, and 16 oz (0.473 liters) were taken and set in an ice bath for Condition H.  The remaining wort was then quickly cooled to 75°F (24°C) using an immersion chiller.  When the full volume of wort reached 75°F (24°C), it was stirred and a 16-oz (0.473-liter) sample was immediately transferred to a sanitized quart (liter) container for Condition J.  The full volume of wort then sat undisturbed for 10 minutes (with the lid on the kettle), after which a 16-oz (0.473-liter) sample was taken from the top of the wort and transferred to a sanitized quart (liter) container for Condition K.  When Condition G had reached 75°F (24°C), 16 oz (0.473 liters) of clear wort were decanted to a sanitized quart (liter) container.  When Condition H reached the same temperature, all of it was transferred to a sanitized quart (liter) container.

Each sample was then aerated for 1 minute by vigorous shaking, and 0.0085 oz (0.24 g) of Safale US-05 yeast (age 11 months) was pitched to target 750,000 cells per ml and degree Plato.  Each sample fermented for about one week (with a small opening to vent CO₂).   Each sample was swirled once a day to reduce krausen deposits.  Oregon BrewLab analyzed all samples for IBUs.

5.3 Experiment #3: Results
Condition G (settled wort) had 29.2 IBUs.  Condition H (stirred wort) had 26.9 IBUs.  Condition J (stirred wort) had 26.5 IBUs.  Condition K (settled wort) had 28.5 IBUs.  Comparing the two conditions taken before immersion chilling, Condition H had 7.8% lower IBUs than Condition G.  Comparing the two conditions taken after immersion chilling, Condition J had 7.0% lower IBUs than Condition K.  While the reduction in IBUs is not as pronounced as in the first two experiments, (a) the use of DME in all conditions (which had average or better-than-average wort clarity in the previous experiments) might have prevented a larger difference in wort clarity from being achieved through settling, and (b) the differences might have become larger with a settling time longer than 10 minutes.  At any rate, it appears that clarifying the wort by settling does yield relatively greater IBUs than stirred wort.  This result indicates that it is probably the clarity of the wort going into the fermentation vessel that is important.

6. Analysis
6.1 IBUs, IAA, and oAA
While there are differences between the first two experiments in terms of wort pH, specific gravity, and variety and weight of hops used, the IBU results from the two DME conditions are quite similar (the green lines in Figure 3), indicating that a comparison between two conditions from the different experiments is not unwarranted.  (Note that one would expect Condition F to have somewhat lower IBU values than Condition C, as observed, because the pH of the wort in Condition F was lowered to 5.68, and lower pH is associated with lower IBU values.)

The following analysis combines the results of the first two experiments, yielding 6 data points produced using five techniques: Condition A, with very cloudy wort produced without recirculation and a moderate drawing rate; Condition B, with moderately cloudy wort produced with recirculation and a moderate drawing rate; Condition C, with moderately clear wort produced from DME; Condition D, with moderately cloudy wort produced without recirculation and a fast drawing rate; Condition E, with very clear wort produced with recirculation and a slow drawing rate; and Condition F, with moderately clear wort produced from DME.  Figure 3 shows the IBU results from all six conditions.

Figure 3. Measured IBU values from the first and second experiments.    Condition A used lautering with no recirculation and a medium rate of drawing wort, Condition B used lautering with recirculation and a medium rate of drawing wort, and Condition C used dried malt extract.  Condition D used lautering with no recirculation and fast drawing, Condition E used lautering with recirculation and slow drawing, and Condition F used dried malt extract.

Figure 3 shows a general correspondence between subjective wort clarity and IBUs; the more turbid the wort, the lower the IBU levels.  The increase in IBUs from the average IBU at a time point (over the six conditions) to the IBU of the clearest wort (Condition E) ranges from 20% (at 40 minutes) to 29% (at 30 minutes).  The decrease in IBUs from the average to the most turbid wort (Condition A) ranges from 21% (at 10 minutes) to 29% (at 40 minutes).  The IBU values from the two DME conditions are within 6% of each other, which (a) is well within the possible range of alpha-acid variation between and within bales, estimated at 15% to 20% [Verzele and De Keukeleire, p. 331] and (b) follows the expected trend of lower-pH worts having lower IBUs.  Therefore, the IBU differences between the two DME conditions are well within expected variation.

We can use the IBU values from each condition to estimate an IAA scaling factor and an oxidized-alpha acid (oAA) scaling factor, using the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements and assuming that the other auxiliary bittering compounds can be modeled reasonably well with existing formulas (as described in Section 5 of that blog post).  This estimation yields the scaling factors specified in Table 1.  The most turbid wort (Condition A) has the lowest IAA scaling factor, and the clearest wort (Condition E) has the highest IAA scaling factor.  There is, however, no clear pattern to the oAA scaling factor, other than the two DME conditions having the two lowest values. The oAA scaling factor has a mean of 0.085 with standard deviation 0.012.

Because the oAA scaling factor shows no clear pattern correlating with the IBU measurements, and because the standard deviation is fairly small relative to the mean, the variation in oAA scaling factor may be due to measurement error, modeling error, and/or a limited number of data points.  We can then set this value to the average oAA scaling factor, 0.085, and re-estimate the IAA scaling factors.  These results are listed in Table 2.  With these results, the most turbid wort (Condition A) (still) has the lowest IAA scaling factor and the clearest wort (Condition E) (still) has the highest IAA scaling factor.  These estimated IAA scaling factors have a strong positive correlation of 0.975 with the measured IBU values at 40 minutes.

Condition:	A	B	C	D	E	F
IAA scaling factor:	0.289	0.417	0.507	0.432	0.523	0.503
oAA scaling factor:	0.086	0.096	0.077	0.081	0.102	0.069

Table 1.  Estimated IAA and oAA scaling factors for each of the six conditions in the combined set of experiments.  Each scaling factor was estimated from four measured IBU values.

Condition:	A	B	C	D	E	F
IAA scaling factor:	0.291	0.441	0.483	0.419	0.575	0.453

Table 2.  Estimated IAA scaling factors for each of the six conditions in the combined set of experiments, holding the oAA scaling factor constant at 0.085.  Each IAA scaling factor was estimated from four measured IBU values.

The average of the six IAA scaling factors, 0.433, is close to the average of the four “mid-range” scaling factors, 0.444.  If we take 0.444 as the scaling factor for “average” wort turbidity, we can create a linear scaling factor for the impact of wort turbidity on IAA concentration.  This scaling factor has a value of 1.0 for average wort, 1.30 (1.30=0.575/0.444) for very clear wort (Condition E), and 0.66 (0.66=0.291/0.444) for very cloudy wort (Condition A).  While precisely quantifying wort turbidity from subjective descriptions is an additional challenge, the impact on IBUs may be roughly estimated using general descriptors and associated scaling factors.  For example, wort descriptions and scaling factors might be “very clear” (1.30), “clear” (1.20), “somewhat clear” (1.10), “average” (1.0), “somewhat cloudy” (0.90), “cloudy” (0.80), and “very cloudy”  (0.70).  Mapping from scaling factor to description, Condition A is even more than “very cloudy” (0.66 < 0.70), Condition B is “average” (0.993 ≈ 1.00), Condition C is “somewhat clear” (1.088 ≈ 1.10), Condition D is in between “somewhat cloudy” and “average” (0.90 < 0.944 < 1.00), Condition E is “very clear” (1.295 ≈ 1.30), and Condition F is close to “average” (1.020 ≈ 1.00).  While this turbidity scale ranges from 30% less than average to 30% more than average, the four non-extreme cases here all fall within 10% of average turbidity.

Why might wort turbidity affect the IAA concentration?  Lewis and Young say that “iso-alpha-acids, being surfactants, react with inert surfaces of all sorts” [Lewis & Young, p. 267].  It is therefore possible, especially given the results of the third experiment, that IAA react or bind with the fine particles in turbid wort during fermentation and then precipitate out of solution, lowering the IBU.  The wort clarity going into the fermentation vessel (instead of the pre-boil clarity resulting from the lautering technique) appears to be the most relevant factor.

6.2 Protein Levels
For Experiments #1 and #2, I had the protein levels in the wort and beer measured, in order to check the hypothesis that protein levels and IBUs are negatively correlated.   The results listed in Sections 3.3 and 4.3 demonstrate that the protein levels were very similar across all three pre-boil worts in Experiment #1.  The protein levels in Experiment #1 didn’t change significantly during the boil or during fermentation.  The protein levels in Experiment #2 showed no difference from each other or the values in Experiment #1.  Therefore,  the expected difference in protein levels did not materialize, indicating that recirculation and teig formation did not lower the protein concentration in the wort.  While there might be a relationship between protein levels and IBUs, these experiments do not evaluate such a relationship.

6.3 Polyphenols
I also had polyphenol concentrations measured in Experiment #2, as a way of testing the model of malt polyphenols proposed in the blog post The Contribution of Malt Polyphenols to the IBU and the model of hop polyphenols described in the blog post The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU. Table 3 lists (a) the measured total polyphenol concentration in each condition, (b) the predicted polyphenol concentration derived from malt according to the model, (c) the estimated IBUs obtained from this model of malt polyphenols, (d) the predicted polyphenol concentration derived from hops, (e) the estimated IBUs obtained from these hop polyphenols, (f) the total polyphenol concentration from both malt and hops (the sum of the individual model concentrations), and (g) the total IBUs predicted to come from polyphenols according to the model.   The measured polyphenol concentrations are greater than the model concentrations by 21 mg/L in both Conditions D and E, and less than the model by 2 mg/L in Condition F.  The last row of Table 3 estimates the impact of the observed and model differences on IBUs, by changing the model’s malt polyphenol concentration to be equal to the measured concentration minus the estimated hop concentration, estimating the IBUs from the sum of the changed malt polyphenols and the hop polyphenols, and taking the difference between this new IBU estimate and the original IBU estimate.  Although the measured polyphenol concentration is up to 14% greater than the model concentration, the estimated impact of this difference is less than 0.2 IBUs.

Condition:	D	E	F
Measured total polyphenol concentration:	156 mg/L	165 mg/L	138 mg/L
Model malt polyphenol concentration:	118.0 mg/L	126.5 mg/L	122.4 mg/L
Model malt IBU contribution:	0.84	1.00	1.29
Model hops polyphenol concentration:	16.9 mg/L	17.6 mg/L	17.3 mg/L
Model hops IBU contribution:	0.37	0.39	0.38
Model total polyphenol concentration:	134.9 mg/L	144.1 mg/L	139.7 mg/L
Model total IBU contribution:	1.21	1.39	1.67
Estimated model error, in IBUs:	0.13	0.14	0.0

Table 3.  Measured polyphenol concentrations, predicted polyphenol concentrations and IBUs (from malt, hops, and the combination), and estimated error in IBUs.

6.4 Wort and Beer Clarity
It would be interesting to repeat these experiments using a turbidity meter to obtain precise measurements of turbidity.  The subjective classifications I’ve used might be sufficient for describing a general trend, but they limit the objectiveness, and therefore the usefulness, of these results.

In the second experiment I found that Condition D, which had moderately cloudy wort, ended as an exceptionally clear beer.  Condition E, which had a very clear wort, ended as a somewhat cloudy beer after one week of fermentation.  Condition F, which like Condition D was neither exceptionally turbid nor exceptionally clear, also ended as a somewhat cloudy beer.  These results therefore show no relationship between wort clarity and beer clarity, in contrast with expectations.

6.5 Wort pH
From the blog post Some Observations of Mash and Wort pH, a wort made from either two-row malt or DME and with low-alkalinity water to a specific gravity of 1.040 to 1.045 should have a pH of about 5.75 to 5.85.  The pre-boil pH of the all-grain conditions in the first experiment (Conditions A and B) was much lower than expected given the specific gravity, at about 5.55.  The pH of the wort produced from DME in this experiment (Condition C) was close to the expected range, at 5.74.  The pH of Conditions D and E in the second experiment was higher than in the first experiment, at about 5.68, but still lower than Condition F (DME) at 5.86.  The reasons for the low pH values and differences in pH are unclear.  The all-grain conditions in both experiments used malt from the same 55-lb (25 kg) bag of Great Western two-row malt.  John Palmer and Colin Kaminski note that different lots of malt from the same maltster can produce variation in mash pH, and that pH can vary with growing conditions and microflora [Palmer and Kaminski, p. 76], which might explain the lower-than-expected pH from the all-grain conditions in both experiments.  (Another beer produced from this same bag of base malt also had a lower-than-expected unadjusted pH.)  The grist ratio (volume of water per weight of grain) was different between the two experiments, which might explain some of the difference between the pH observed in the two experiments [Palmer and Kaminski, p. 70].

7. Conclusion
Two of the experiments described here show that wort turbidity can affect IBUs by as much as 60%, with a ~30% increase in IBUs for very clear wort and a ~30% decrease for very cloudy wort.  The estimated IAA loss factor varies from 0.66 to 1.30 in the data from this experiment.  The third experiment indicates that this change in IBUs probably occurs during fermentation instead of during the boil.  In other words, the IAA levels appear to be affected by overall wort clarity during fermentation, not just the pre-boil wort clarity (from lautering technique).  The wort protein levels did not change with wort turbidity, and so the impact of protein on IBUs is still unclear.

Wort clarity was not well correlated with beer clarity, which does not support the adage that “clear wort makes clear beer”.

The combination of a previously-described model of malt polyphenols and model of hop polyphenols predicted the measured total polyphenol concentration reasonably well, with relative differences of -14%, -13%, and 1% for the three samples.

8. Acknowledgment
I would like to sincerely thank Dana Garves at Oregon BrewLab for the IBU, protein, and polyphenol measurements for these experiments.  Oregon BrewLab has always been a pleasure to work with.

References

• V. A. Algazzali, The Bitterness Intensity of Oxidized Hop Acids: Humulinones and Hulupones, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• H. O. Askew, “Changes in Concentration of α and β Acids and of Iso-Compounds on Heating Extracts of Hops in Aqueous Solutions and Unhopped Wort,” in Journal of the Institute of Brewing, vol. 71, pp. 10-20, 1965.
• J. Dierckens and M. Verzele, “Oxidation Products of Humulone and Their Stereoisomerism,” in Journal of the Institute of Brewing, vol. 75, pp. 453-456, 1969.
• M. Garetz, “Hop Storage: How to Get – and Keep – Your Hops’ Optimum Value” in Brewing Techniques, January/February 1994, hosted on morebeer.com.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• G. Oliver, The Oxford Companion to Beer, Oxford University Press, 2011.
• J. J. Palmer, How to Brew: Everything You Need to Know to Brew Beer Right the First Time. 3rd edition, Brewers Publications, 2006.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids.  Developments in Food Science 27.  Elsevier, 1991.

 

The Production of Oxidized Alpha Acids at Hop-Stand Temperatures

Abstract
The IBU combines the concentration of isomerized alpha acids (IAAs) and the concentration of “auxiliary bittering compounds” (ABCs) in beer into a single measure of approximate bitterness.  While IAAs contribute the most to the IBU in typical beers, ABCs play a significant role and may have contributions greater than IAAs in very late-hopped beers.  The auxiliary bittering compounds are composed of polyphenols, oxidized alpha acids, and oxidized beta acids.  Oxidized alpha acids are produced as the hops age, but they are also produced in fairly large quantities during the boil.  (There is evidence that the oxidized alpha acids produced during the boil are the second-greatest contributor to the IBU, after IAA.)  It is known that temperature has a large effect on how quickly alpha acids isomerize, but it is not clear what impact wort temperature has on the production of oxidized alpha acids.  This blog post estimates the concentration of oxidized alpha acids in finished beer from hops steeped at three temperatures: boiling (100°C or 212°F), 90°C (194°F), and 80°C (176°F).  The results, while not definitive, indicate that these different temperatures do not yield significant differences in the production of oxidized alpha acids.  Polyphenol levels in the beer samples were also measured in order to check previously-developed polyphenol models and provide supporting evidence that most of the ABCs are, as expected, probably coming from oxidized alpha acids and not from polyphenols.

1. Introduction
The IBU is a measure of the concentration of a number of different bitter compounds.  (To be more precise, the IBU is a measure of the absorbance of light at 275 nm through acidified beer.  A number of bitter compounds in beer absorb light at this frequency.  The greater the concentration of these compounds, the more light is absorbed, and the higher the IBU.)  In typical beers, the IBU value reflects mostly the concentration of isomerized alpha acids (IAAs) [Peacock, pp. 164-165], which are produced during the boil from alpha acids (AA).  The other bitter compounds, known as “auxiliary bittering compounds” (ABCs), or nonIAA, are polyphenols, oxidized alpha acids, and oxidized beta acids.  These compounds can be considered to be present in the wort soon after the hops addition [e.g. Dierckens and Verzele, p. 454; Askew, p. 18].

Alpha acids (without isomerization) “do not survive to any significant extent into beer” [e.g. Lewis and Young, p. 259] and are not bitter [Shellhammer, p. 169], but as they age and become oxidized, the resulting oxidized alpha acids (oAAs) are both soluble in wort and bitter [Algazzali, pp. 14-15, p. 19, p.45; Maye et al, p. 23; Hough et al., pp. 435-436; Hough et al., p. 439; Lewis and Young, p. 265].  Oxidized alpha acids are also produced during the boil [Parkin, p. 11, Algazzali, p. 17; Dierckens and Verzele, p. 454; Oliver p. 471].  A previous blog post has estimated that oxidized alpha acids (oAA) are the second-largest contributor to the IBU, after isomerized alpha acids, and that a typical beer may have equal contributions of IAA and oAA after about 10 minutes of boiling hops in wort.  In beers brewed with large additions of hops at flameout, the IBU may be a measurement of mostly oxidized alpha acids.

While the impact of wort temperature on the rate of isomerization is well known [Malowicki, p. 27], the impact of temperature on the production of oxidized alpha acids is not known.  (At room temperature, dry hopping will contribute oxidized alpha acids to the finished beer [Parkin, p. 30; Maye, p. 25], but it seems unlikely that there is much production of oAA.  Instead, it is more likely that during dry hopping (most of) the oAA already present in the hops (coming from oxidation during storage) dissolve into the beer [Maye, p. 25].)  If boiling transforms x% of the available alpha acids to oxidized alpha acids in the finished beer, then does steeping hops at 80°C (176°F) transform only 0.80 × x% of the alpha acids?  Or, more generally, how do temperatures typically encountered in hop stands affect the oxidization of alpha acids, relative to oxidation at boiling? The purpose of this blog post is to answer this question.

When collecting data to answer this question, I also measured polyphenol concentrations as a way of testing the model of malt polyphenols proposed in the blog post The Contribution of Malt Polyphenols to the IBU and the model of hop polyphenols described in the blog post The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU.  To the extent that the polyphenol models accurately predict polyphenol concentration, we can have confidence in the models’ estimate of the contribution of polyphenols to the IBU, and support (or contradict) the claim that oxidized alpha acids contribute much more to the IBU than other auxiliary bittering compounds.

2. The Concentration of Isomerized Alpha Acids in Beer
Mark Malowicki developed formulas to estimate the concentration of IAAs in the wort from the initial concentration of alpha acids [Malowicki, p. 27]:

k1(T) = 7.9×1011 e-11858/T	[1]
k2(T) = 4.1×1012 e-12994/T	[2]
[IAA]wort = [AA]0 × (k1(T)/(k2(T) − k1(T))) × (e–k1(T)t − e–k2(T)t)	[3]

where k₁(T) and k₂(T) are empirically-derived rate constants, T is the temperature in Kelvin (i.e. 373.15 K for boiling), t is the steep time (in minutes), e is the constant 2.71828, and [AA]₀ is the initial concentration of alpha acids in the wort (in ppm).  This concentration of IAA in the wort, [IAA]_wort, decreases as IAAs are lost to trub and krausen during the boil and fermentation.  The concentration of IAAs in beer ([IAA]_beer) can then be expressed as the IAA concentration in wort multiplied by a loss scaling factor, scaling_IAA:

[IAA]beer = [IAA]wort × scalingIAA

[4]

3. The IBU Expressed as Concentrations of Bitter Compounds
Val Peacock [Peacock, p. 161] provides an equation to express the IBU as a combination of the concentration of IAAs and ABCs (also called nonIAA):

IBU = 5/7 × ([IAA]beer + [ABC]beer)

[5]

where IBU is the IBU value of the beer, [IAA]_beer is the concentration of isomerized alpha acids in the finished beer (in ppm, from Section 2), and [ABC]_beer is the concentration of all other bittering compounds (also in ppm).

The concentration of ABCs in beer ([ABC]_beer) can be expressed as the sum of the concentrations of the individual ABC components multiplied by appropriate scaling factors that relate each concentration to absorption at 275 nm:

[ABC]beer = [PPmalt]beer × scalePPmalt + [PPhops]beer × scalePPhops + [oAA]beer × scaleoAA + [oBA]beer × scaleoBA

[6]

where [PPmalt]_beer, [PPhops]_beer, [oAA]_beer, and [oBA]_beer are the concentrations in the beer of malt polyphenols, hop polyphenols, oxidized alpha acids, and oxidized beta acids, respectively, and scale_PPmalt, scale_PPhops, scale_oAA, and scale_oBA are the scaling factors that relate concentration to absorption at 275 nm for these compounds.

Alternatively, we can express the concentration of ABCs in beer as the concentration of total hop particles added to the wort, multiplied by a single scaling factor that accounts for (a) the proportion of each ABC compound to total hop matter, (b) different absorption factors of these compounds, and (c) losses of each ABC to trub and during fermentation:

[ABC]beer = [hops]wort × scalingABC

[7]

where [hops]_wort is the concentration of hops added to the wort and scaling_ABC is the above-mentioned scaling factor.

We can estimate the scaling factors scaling_IAA and scaling_ABC from equations [3], [4], [5], and [6] and measured IBUs of beer samples fermented from wort taken at different time points during the boil.  This technique is described in the blog post Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements.

4. Experimental Overview
To evaluate the production of oxidized alpha acids at hop-stand temperatures, I brewed three batches of beer (A, B, and C) with identical wort, hops, and yeast, and varied only the temperature at which the hops steeped in the wort.  The total steep time was 40 minutes, and I took samples every 10 minutes.  I fermented these 12 samples into beer, and had the IBU values of the finished beer measured by Oregon BrewLab.

I then used the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to model the concentrations and scaling factors of IAA and ABC that contribute to the measured IBU values.  In particular, the temperature T in Equations [1], [2], and [3] (above) was set to the steep temperature in order to account for the reduced rate of production of isomerized alpha acids at below-boiling temperatures.  Using estimates of malt and hop polyphenol concentrations and their contribution to the IBU (described in The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU) and ignoring the contribution of oxidized beta acids (due to their negligible concentration in finished beer when using well-preserved hops), I estimated the contribution of oxidized alpha acids to the IBU in each batch.  These estimates were then examined for a relationship with wort temperature.

In addition, Oregon BrewLab measured the polyphenol concentrations at the 10-minute sample of all three batches and at the 40-minute sample of Batch A.  I computed the expected malt polyphenol concentration using the model developed in The Contribution of Malt Polyphenols to the IBU and the expected hop polyphenol concentration using the model in The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU, and compared the predicted with measured polyphenol values.

5. Experimental Methods
Each batch of wort was prepared from 2.53 lbs (1.147 kg) of Briess Pilsen Light Dried Malt Extract and 3.35 G (12.68 liters) of 120°F (49°C) low-alkalinity water, yielding 3.49 G (13.21 liters) of room-temperature wort.  This wort sat for 90 minutes to let the pH stabilize.  The pH was then adjusted to about 5.30 (at room temperature) using phosphoric acid.  The measured pre-boil specific gravity was 1.032 for all three batches.  The wort was boiled for 5 minutes before adding hops, in order to reduce the foam associated with the start of the boil.  A 12-oz (0.35 l) sample of wort was taken after this 5-minute period to measure specific gravity and pH at around the time of the hop addition.

Just before adding hops, the temperature was reduced with the use of a wort chiller to the target steep temperature.  For Batch A, the target temperature was boiling (100°C, 212°F) and no temperature reduction was made.  For Batch B, the target temperature was 90°C (194°F).  For Batch C, the target temperature was 80°C (176°F).  This target was held as closely as possible throughout the hop steeping time.

I used 0.868 oz (24.6 grams) of Amarillo hops in this experiment with a package AA rating of 8.8%.  These hops were harvested in Fall 2019 for the experiment in January 2020.  The hops were analyzed by AAR Lab shortly after I received them, and they showed an alpha-acid (AA) rating of 9.56% and a beta-acid (BA) rating of 5.84%, with a hop storage index (HSI) of 0.272.  To estimate the alpha-acid content at the time of brewing, I used the Garetz formula for estimating alpha-acid decay [Garetz] to obtain a decay factor of 0.96 and an alpha-acid rating at the time of the experiment of 9.21%.

After adding the hops, 16-oz (0.473 l) samples were taken every 10 minutes and quickly cooled in an aluminum cup and ice bath.  The kettle was covered during the boil (or hop stand) to minimize evaporation and the resulting changes in specific gravity.  Each sample was transferred to a sanitized quart (liter) container after it was cooled to 75°F (24°C).  The wort in each container was aerated for 1 minute by vigorous shaking, and 0.008 oz (0.24 grams) of Safale US-05 yeast (age 11 months) was pitched to target 750,000 cells per ml and degree Plato.  At the end of the 40-minute boil (or hop stand), another sample was taken for measuring specific gravity and pH.

Each sample fermented for 10 days (with a small opening to vent CO₂).  The krausen was left to deposit on the sides of the vessel during fermentation.  I removed the krausen deposits one day before taking samples for IBU and polyphenol analysis by Oregon BrewLab.

6. Results
Unfortunately, I kept the hops in Batch B at boiling for the first 3 minutes of the boil, and only then decreased the heat to the target temperature.  To correct for this, the models below use instantaneous temperatures and integrate IAA levels over time for Batch B, with a temperature of boiling for the first 3 minutes and the target of 90°C (194°F) after that, and so the effect of this mistake should be accounted for in the models.

The measured IBU values are plotted in Figure 1 with solid lines.  The polyphenol concentrations are listed in Table 1.

Figure 1. Measured IBU values at steep times of 10, 20, 30, and 40 minutes (horizontal axis) and at steep temperatures 100°C (212°F) (red line), 90°C (192°F) (blue line), and 80°C (176°F) green line.

6. Analysis
6.1 Model #1: Batch-Specific IBU Analysis
The first analysis (Model #1) used the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to estimate scaling factors for IAA and ABC and use these scaling factors to model IBU values.  We can identify the portion of the IBU value that comes from oxidized alpha acids by subtracting estimates of the malt and hop polyphenol IBU contributions and the estimate of the IAA contribution from the model IBU value.  The results of this analysis are IAA scaling factors of 0.40, 0.49, and 0.59 for Batches A, B, and C, respectively, and a ABC scaling factors of 0.0048, 0.0047, and 0.0044 for Batches A, B, and C, respectively.   The estimated oAA levels (expressed as the oAA contribution to the IBU) are 6.8, 6.7, and 6.2 IBUs for Batches A, B, and C, respectively.  The RMS error over all values is 0.40 IBUs.  The estimated IBU levels from Model #1 are plotted in Figure 2 with solid lines (red for Batch A, blue for Batch B, and green for Batch C).  The estimated oAA levels decrease slightly with decreasing temperature.  However, the IAA scaling values increase as the temperature decreases, and there is no clear reason for IAA scaling values to increase in this way.  This suggests that the model is overfitting to the data, and that this trend is an artifact of the data and analysis technique.

Figure 2. IBU values from Model #1 at temperatures 100°C (212°F) (red line), 90°C (192°F) (blue line), and 80°C (176°F) green line.  The measured IBU values (from Figure 1) are plotted with gray markers and dashed lines for reference.

6.2 Model #2: IBU Analysis with Constant IAA Scaling
The results from analyzing each batch independently (Model #1) showed an unexpected increase in the IAA scaling factors as temperature decreases.  All of these batches should, in theory, have the same scaling value for IAA; the kettle temperature should not influence the loss of isomerized alpha acids.  I therefore repeated the estimation of IAA and ABC scaling factors, but constrained the value of the IAA scaling factor to be the same for all three batches.  This analysis is called Model #2.  The results of this analysis are an IAA scaling factor of 0.43 and ABC scaling factors of 0.0041, 0.0055, and 0.0051 for Batches A, B, and C, respectively.  The estimated oAA levels (expressed as the oAA contribution to the IBU) are 5.8, 7.8, and 7.2 IBUs for Batches A, B, and C, respectively.  The RMS error over all values is 0.60 IBUs.  The trend that we are looking for, where oAA levels are constant or decrease as the temperature decreases, is not apparent.  The estimated IBU levels from Model #2 are plotted in Figure 3.

Figure 3. IBU values from Model #2 at temperatures 100°C (212°F) (red line), 90°C (192°F) (blue line), and 80°C (176°F) green line.  The measured IBU values (from Figure 1) are plotted with gray markers and dashed lines for reference.

6.3 Model #3: IBU Analysis with Constant IAA and ABC Scaling
The results from enforcing a constant IAA scaling value across all three batches show no clear trend in oAA values with temperature.  If the oxidized alpha-acid levels do not change significantly with temperature and the observed differences are all caused by experimental error, then we can constrain both IAA scaling and ABC scaling to be the same for all three batches, and evaluate if the resulting model error might be explained by experimental errors.  This analysis, called Model #3, resulted in an IAA scaling factor of 0.40 and a ABC scaling factor of 0.0053, with an RMS error of 0.88 IBUs (about double that of Model #1 with no constraints).  The estimated oAA levels (expressed as the oAA contribution to the IBU) are 7.5 IBUs for all batches. This constraint causes the model values for Batch A to be somewhat higher than observed values, and the model values for Batch B to be a bit lower than observed values at higher steep times. The estimated IBU levels from Model #3 are plotted in Figure 4.

Figure 4. IBU values from Model #3 at temperatures 100°C (212°F) (red line), 90°C (192°F) (blue line), and 80°C (176°F) green line.  The measured IBU values (from Figure 1) are plotted with gray markers and dashed lines for reference.

Can we explain this measured error (0.88 IBUs, compared with 0.40 IBUs for the unconstrained model) with plausible experimental errors?  If we hypothesize that the differences are because the steep temperatures did not reach their targets, then it would take a steep temperature of 93°C (199.4°F) in Batch B (instead of the target 90°C/194°F) and a steep temperature of 83°C (181.4°F) in Batch C (instead of the target 80°C/176°F) to reduce the error to 0.44 IBUs.  It seems possible that the temperature in the kettle was not uniform, but possibly higher at the bottom of the kettle (near the heat source) and lower at the top of the kettle (where I measured temperature).  (It is also very likely that I did not reach exactly the target temperature over the entire duration of steeping, even after accounting for the mistake in Batch B.)  Such effects might conceivably lead to a 3°C (5.4°F) difference between the average actual temperature and the measured temperature, although my gut feeling is that this is stretching the bounds of plausibility.  If we hypothesize that the alpha-acid ratings varied between batches, then Batch A would have to be 8% lower, Batch B 7% higher, and Batch C 8% higher than expected in order to reduce the error to 0.50 IBUs.  Alpha-acid ratings can vary 15% to 20% within the same bale of hops [Verzele and DeKeukeleire, p. 331], and it is worth noting that the measured AA rating of 9.56% is 9 percent higher than the package rating of 8.8%.  So, it is also conceivable that the actual AA ratings in each batch varied by up to 8% from the expected value, but once again my gut feeling is that this is pushing the bounds of plausibility.  If we have a combination of smaller differences in both temperature and AA rating (namely that Batch A has 4% lower AA, Batch B has a temperature of 91.5°C (196.7°F) and 3.5% higher AA, and Batch C has a temperature of 81.5°C (178.7°F) and 4% higher AA) then the error is 0.46 IBUs.  Such errors seem plausible, but they are of course a hypothetical “post-mortem” explanation of the data.

6.4 Summary of IBU Analysis
A pattern of decreasing oAA levels with decreasing temperature is slight in the first model and does not exist in the other two models.  There is also no easy explanation for the increase in IAA scaling factors with decreasing temperature in the unconstrained model (Model #1).  If the model is adjusted to account for several small potential experimental errors, both IAA scaling and oAA levels can be held constant and result in a low overall error.  Therefore, the data suggest, but do not prove, that the production of oxidized alpha acids is not greatly influenced by steep temperature.  It is estimated that in this experiment, oxidized alpha acids contributed about 7 of the measured IBUs, regardless of the model, steep time, or steep temperature.

6.5 Analysis of Polyphenol Concentrations
In order to test the model of polyphenols developed in a previous blog post, Oregon BrewLab measured the polyphenol concentrations of four conditions: the 10-minute sample for Batches A, B, and C, and the 40-minute sample for Batch A.  Results are listed in Table 1; these values are the sum of both malt and hop polyphenols.

Sample	Measured Polyphenol Concentration (ppm)	Modeled Polyphenol Concentration (ppm)
Batch A, 10 minutes	91	95.2
Batch B, 10 minutes	92	95.2
Batch C, 10 minutes	92	95.2
Batch A, 40 minutes	107	106.7

Table 1.  Measured and modeled polyphenol levels of four samples in this experiment.

We can use the model of malt polyphenols with the model of hop polyphenols described in Section 4 of The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU to estimate the total polyphenol concentration.  For using the model of malt polyphenols at steep time 10, we have a boil time of 10 minutes, a specific gravity of 1.0342 and a pH of 5.31.  For the model of hop polyphenols, we have 24.60 grams (0.868 oz) of hops added to 12.306 liters (3.25 gallons) of wort.  These models predict 84.0 ppm of malt polyphenols  and 11.2 ppm of hop polyphenols at 10 minutes, for a total of 95.2 ppm.  Since this model does not depend on wort temperature, it predicts the same value for each batch at the 10-minute sample time.  Changing the boil time to 40 minutes yields 95.5 ppm of malt polyphenols and the same 11.2 ppm of hop polyphenols, for a total of 106.7 ppm.

The model values of 95.2 ppm for all batches at 10 minutes and 106.7 ppm at 40 minutes are in good agreement with the measured values in Table 1.  These results provide some confidence in how well these models predict malt and hop polyphenol levels.

9. Conclusion
While not conclusive, the data from this experiment indicate that hop steeping temperatures from boiling to as low as 80°C (176°F) do not result in significant differences in the production of oxidized alpha acids.  Also, the measured polyphenol concentrations correspond well with the models of malt and hop polyphenols proposed in previous blog posts.

It is estimated that in this experiment, the oxidized alpha acids produced during the boil contribute more to the IBU than isomerized alpha acids up to about a 15-minute steep time.  At the 10-minute steep time and boiling, it is estimated that the 14.9 measured IBUs reflect 5.6 IBUs from isomerized alpha acids, 7.8 IBUs from oxidized alpha acids, 1.3 IBUs from malt polyphenols, and 0.2 IBUs from hop polyphenols.  At the 40-minute steep time and boiling, it is estimated that the 29.1 measured IBUs reflect 19.8 IBUs from IAA, 7.8 IBUs from oAA, and 1.5 IBUs from polyphenols.

10. Acknowledgment
I would like to sincerely thank Dana Garves at Oregon BrewLab for her attention to quality and detail that is reflected in the IBU and polyphenol measurements presented here and in previous posts.

References

• V. A. Algazzali, The Bitterness Intensity of Oxidized Hop Acids: Humulinones and Hulupones, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• J. Dierckens and M. Verzele, “Oxidation Products of Humulone and Their Stereoisomerism,” in Journal of the Institute of Brewing, vol. 75, pp. 453-456, 1969.
• M. Garetz, “Hop Storage: How to Get – and Keep – Your Hops’ Optimum Value” in Brewing Techniques, January/February 1994, hosted on morebeer.com.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science.  Volume 2: Hopped Wort and Beer.  Springer-Science+Business Media, B. V., 2nd edition, 1982.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• J. P. Maye, R. Smith, and J. Leker, “Humulinone Formation in Hops and Hop Pellets and Its Implications for Dry Hopped Beers”, in MBAA Technical Quarterly, vol. 51, no. 1, pp. 23-27, 2016.
• G. Oliver, The Oxford Companion to Beer, Oxford University Press, 2011.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids.  Developments in Food Science 27.  Elsevier, 1991.

The Contribution of Malt Polyphenols to the IBU

Abstract
This blog post analyzes the contribution of malt polyphenols to the IBU, in order to enable better estimates of the relative contributions of all auxiliary bittering compounds.  These estimates may be used to develop better models and explanations of the factors that influence IBUs.  The data from the experiment described here indicate that malt-derived IBUs (but not polyphenol concentrations) increase as the wort pH decreases, that the process of fermentation decreases polyphenol levels and IBUs by about 30%, and that the polyphenol concentrations (but not IBUs) increase during the boil.

1. Introduction
The IBU is mostly a measurement of the concentration of isomerized alpha acids (IAA) in beer, but it also includes concentrations of auxiliary bittering compounds (ABC, also referred to as nonIAA), which include oxidized alpha acids, oxidized beta acids, hop polyphenols, and malt polyphenols.  Malt polyphenols contribute a small amount to the total IBU value of a beer.  This contribution has been estimated at 1 to 3 IBUs [Shellhammer, p. 177].  I have no illusions of modeling IBUs with an error less than 3 IBUs, and so the purpose of this study is not to develop a hyper-accurate IBU formula.  Instead, the purpose here is to better understand and model the relative contribution of all auxiliary bittering compounds to the IBU.  This better modeling, including the contributions of oxidized alpha acids and hop polyphenols, may result in better IBU predictions, even if the average error is still greater than 3 IBUs.  Having a better understanding of the relative contributions of different components may also enable better explanations of why IBUs change under certain circumstances, such as with the use of pellet hops as opposed to whole-cone hops.

There are not many references in the literature to malt-derived IBUs.  Tom Shellhammer has said that the IBU “measurement yields a finite value in the range of 1 − 3 [IBU] for unhopped beer” [Shellhammer, p. 177].  Emily Parkin’s thesis looked at the impact of dry hopping on an unhopped beer, and so she analyzed the control condition of an unhopped beer for both IBUs and polyphenol concentrations [Parkin, p. 23].  She found that an increase of 100 mg/L of hop polyphenols in the beer resulted in a 2.2 increase in IBUs [Parkin, p. 28].  I have not been able to find a similar published relationship between malt polyphenols and IBUs.

Yeast has only a minor effect on polyphenol concentrations during fermentation.  As Steve Alexander states, “the impact of yeast on the amounts of phenolics in beer is quite small” [Alexander].   Leiper and Miedl also note that “little is lost during fermentation”, but that “polyphenols are lost in cold break, and later in cold conditioning” [Leiper and Miedl, p. 136].  It is then unclear how much, if any, of the malt polyphenols are lost to trub and krausen deposits during fermentation.

In a previous blog post, I estimated that malt-derived IBUs could be approximated by the original gravity of the beer (in gravity points) multiplied by 0.025. Using this method, a beer with OG 1.050 is estimated to have 1.25 IBUs from malt polyphenols.

2. Experimental Overview
One purpose of the experiment described here was to answer some questions about the relationship between malt polyphenols and IBUs.  For example, do malt-derived IBUs and/or malt polyphenol levels change with pH, boil time, or fermentation?  A second purpose was to use the answers to these questions to develop a quantitative model of malt-derived IBUs.  This new quantitative model can be compared with the model from the previous blog post.  The degree of similarity between the two models (developed using different sets of data) provides a corresponding degree of confidence in the accuracy of the models.

This experiment produced of only six samples of wort or beer; the small number of data points means that any conclusions from this set of data are only preliminary.  Three of the samples were designed to evaluate the relationship between specific gravity, IBUs, and polyphenols, with target gravities of 1.030 (Condition A), 1.050 (Condition D), and 1.070 (Condition E).  Two samples were designed to evaluate the impact of pH on IBUs and polyphenols, with target pH levels of 5.80 (Condition A) and 5.30 (Condition B).  Two samples were designed to evaluate the impact of fermentation, one of pre-fermentation wort (Condition C) and one of post-fermentation beer (Condition D).  Finally, two samples were designed to evaluate the impact of boil time on IBUs and polyphenols, with boil times of 10 minutes (Condition E) and 60 minutes (Condition F).

In all conditions except for Condition B, the pH of the wort was not adjusted, and therefore the pH decreased as the specific gravity increased.  In all conditions except for Condition F (with a 60-minute boil), the wort was boiled for 10 minutes (primarily to sterilize it).

3. Experimental Methods
Three batches of wort were created using the amounts of Briess Pilsen Light Dried Malt Extract (DME) and 120°F (49°C) low-alkalinity water listed in Table 1.  The first batch was used to create Conditions A and B, the second batch to create Conditions C and D, and the third batch to create Conditions E and F.  The resulting estimated volumes of room-temperature wort are listed in Table 1.  All batches of wort sat for three hours to let the pH stabilize.  For Conditions A and B, the wort was separated into two equal parts.  One part (Condition A) was not modified, and the pH of the other part (Condition B) was adjusted with phosphoric acid to a room-temperature pH of 5.30. The measured (and temperature-corrected) pH levels and specific gravities are listed in Table 1.

Conditions C and D were boiled together as one batch of wort.  After cooling, 4 oz (0.118 liters) were transferred to a sanitized sample container for analysis, becoming Condition C.  This sample was refrigerated while the other samples fermented.  The remainder of this batch became Condition D.  Conditions E and F were boiled together for 10 minutes as one batch of wort, after which 15.5 oz (0.458 l) were transferred for cooling (becoming Condition E) and the remainder was boiled for another 50 minutes with tape sealing the lid to bring further evaporation close to zero (becoming Condition F).

The kettle was covered (but not completely sealed shut) during the 10-minute boil to reduce evaporation and the resulting changes in specific gravity.  After boiling for 10 minutes (or 60 minutes for Condition F), the wort was quickly cooled in an ice bath. Once samples reached 75°F (24°C), they were transferred to sanitized quart (liter) containers.  The five samples being fermented were aerated for 1 minute by vigorous shaking, and the amount of Safale US-05 yeast (age 12 months) listed in Table 1 was pitched in order to target 750,000 cells per ml and degrees Plato.  These five samples fermented for 14 days with a small opening to vent CO₂. (Condition D had a boil-over, and about 2 oz of wort were lost.  The amount of yeast that was pitched was adjusted to account for the lost volume.  Other samples had about 15 oz (0.44 liters) of wort available for fermentation.)  The krausen was left to deposit on the sides of the vessel during fermentation.  These deposits were removed one day before decanting analysis samples.  All samples were analyzed for polyphenols and IBUs by Oregon BrewLab.

	Conditions A, B	Conditions C, D	Conditions E, F
DME	2.70 oz / 76.54 g	2.88 oz / 81.65 g	6.31 oz / 178.88 g
added water (at 120°F / 49°C)	30.72 oz / 0.908 l	19.00 oz / 0.562 l	28.62 oz / 0.846 l
resulting wort volume	32.0 oz / 0.946 l	20.0 oz / 0.591 l	32.0 oz / 0.946 l
specific gravity	1.0314	1.0508	1.0715
unadjusted pre-boil pH	5.84	5.77	5.73
final pre-boil pH	5.84 (Condition A), 5.30 (Condition B)	5.77	5.73

Table 1. Measured (and temperature-corrected) values of the three batches of wort.

	Condition A	Condition B	Condition C	Condition D	Condition E	Condition F
description	SG 1.03, no pH adjust, 10-min boil	SG 1.03, low pH, 10-min boil	SG 1.05, no pH adjust, 10-min boil, pre-ferment	SG 1.05, no pH adjust, 10-min boil	SG 1.07, no pH adjust, 10-min boil	SG 1.07, no pH adjust, 60-min boil
target yeast pitched	0.0078 oz / 0.22 g	0.0078 oz / 0.22 g	N/A	0.0102 oz / 0.29 g	0.0173 oz / 0.49 g	0.0173 oz / 0.49 g
estimated volume for fermentation	15.3 oz / 0.452 l	15.1 oz / 0.447 l	N/A	13.1 oz / 0.387 l (loss due to boil-over)	15.0 oz / 0.444 l	15.0 oz / 0.444 l
estimated post-boil gravity	1.0328	1.0333	1.0538	1.0538	1.0763	1.0763
estimated post-boil pH	5.802	5.283	5.733	5.733	5.694	5.500
measured polyphenols (mg/L)	70	67	179	142	222	275
measured IBUs	0.7	1.6	1.3	1.0	1.6	2.3

Table 2. Estimated and measured values for each condition.

4. Experimental Results
4.1 Measured Results
Figure 1 shows the measured IBU values from this experiment, plotted as a function of the estimated post-boil specific gravity (expressed as gravity points).  It can be seen that IBUs increase with specific gravity, as expected.  IBUs increase by 128% (from 0.7 to 1.6) as the pH drops from 5.80 to 5.28 (as seen at specific gravity 1.033).  There is a 23% decrease in IBUs as a result of fermentation, dropping from 1.3 to 1.0 (as seen at specific gravity ~1.052).  There is a 44% increase in IBUs between a 10-minute and 60-minute boil time, from 1.6 to 2.3 (as seen at specific gravity 1.076).

Figure 1. Measured IBUs plotted as a function of specific gravity, with different values of specific gravity, pH, fermentation, and boil time.

Figure 2 shows the measured polyphenol values, also plotted as a function of the estimated post-boil specific gravity (expressed as gravity points).  The polyphenol concentration shows an increase with specific gravity that is similar to the increase in IBUs.  However, polyphenol levels don’t change with pH, with a negligible 4% decrease from pH 5.80 to 5.28 (as seen at specific gravity 1.033).  The change in polyphenol levels with fermentation is very similar to that of IBUs, with a 21% decrease from 179 mg/L to 142 mg/L (as seen at specific gravity ~1.052).  There is a 24% increase in polyphenol levels as the boil time changes from 10 minutes to 60 minutes, from 222 mg/L to 275 mg/L (as seen at specific gravity 1.076).

Figure 2. Measured polyphenol concentration plotted as a function of specific gravity, with different values of specific gravity, pH, fermentation, and boil time.

4.2 Estimated Results
The post-boil specific gravity for each condition (listed in Table 2) was estimated from the measured pre-boil specific gravity, pre-boil volume, and post-boil volume.

The post-boil pH for each condition (also in Table 2) was estimated using a formula derived from the set of pH and boil time values plotted in Figure 3 of the blog post “Some Observations of Mash and Wort pH“.  This formula is:

pHfinal = (pHinit × ((-0.00393 × t) + 1.0) + (0.01862 × t)

[1]

where pH_final is the post-boil pH, pH_init is the pre-boil pH, and t is the boil time, in minutes.

5. Analysis and Modeling
5.1 Preliminary Analysis
As expected, there is a strong correlation between malt polyphenols, specific gravity, and IBUs.  Wort with higher gravity has a greater concentration of malt polyphenols, and these polyphenols are the primary source of IBUs in unhopped beer [Shellhammer, p. 177].  It should be noted that this relationship will therefore be weaker for worts made with adjuncts that increase the original gravity but do not contribute any polyphenols.

By comparing the results from Conditions A and B, it is clear that IBUs increase as pH decreases, but that polyphenol levels don’t change with pH.  We also know that the wort pH naturally decreases as the specific gravity increases.  This effect is seen in the current set of data, with pre-boil gravity increasing from 1.031 to 1.071 and pre-boil pH decreasing from 5.84 to 5.73.  It is then not obvious how much of the change in IBUs between the conditions is due to the different levels of specific gravity (caused by different concentrations of malt polyphenols) or the different pH levels (correlated with different levels of specific gravity).  This difficulty is addressed in the next sections.

The results indicate that both IBUs and polyphenols are decreased by fermentation.  While Figures 1 and 2 show an increase in both IBUs and polyphenols with boil time, the increase in IBUs may be due, at least in part, to the decrease in pH that occurs during the boil.

5.2 IBUs as a Function of pH
To account for the effect of pH changes on malt-derived IBUs, we can model the effect of pH that is observed between Conditions A and B as a scaling factor.  First, we compute the change in IBU per pH unit observed between Conditions A and B:

slope = −1 × (IBUA − IBUB) / (pHA − pHB) = (0.7 − 1.6) / (5.802 − 5.283) = 1.734

[2]

where slope is the change in IBU per pH unit, IBU_A and IBU_B are the measured IBUs from Conditions A and B, respectively, and pH_A and pH_B are the wort pH values estimated at the time each sample was taken, from Conditions A and B, respectively.  It is convenient to think of IBUs increasing with a decrease in pH, and so the use of −1 in Equation [2] changes the decrease in IBUs per (positive) pH unit to an increase in IBUs per pH unit, with the understanding in subsequent equations that we are modeling a decrease in pH.

Next, we use this slope to compute the change in IBUs that occurs with any change in pH, given that we start at a specific gravity and pH that produces a value of 0.7 IBUs:

deltaIBU0.7 = slope × (pH1 − pH2)

[3]

where deltaIBU_0.7 is the change in IBUs caused by a change in pH, given the starting point of 0.7 IBUs.  The value pH₁ is the pH of the wort before any pH adjustment or boiling.  We usually don’t know pH₁, because pH adjustments are often made during the mash, but if necessary we can approximate it as a single value, 5.75, which is a reasonable compromise over a range of specific gravity levels for two-row malt and low-alkalinity water.  The value pH₂ is the pH of the wort after pH adjustment and boiling.  For example, if the pH drops from 5.75 to 5.30, we predict an increase of 0.78 IBUs from our starting point of 0.7 IBUs, or a total of 1.48 IBUs.

Then we can normalize by 0.7 IBUs to convert the absolute change (at 0.7 IBUs) to a relative change (at any IBU):

IBUrel = slope × (pH1 − pH2) / 0.7

[4]

where IBU_rel is the relative change in IBUs.  (A value of 0.0 represents no change, and 1.0 represents an increase of 100%).  The relative change caused by a decrease in pH from 5.75 to 5.30 is then 0.78 / 0.70, or 1.11, which is a 111% increase in IBUs from 0.70 to 1.48.  We can convert from a relative change to a scaling factor by adding 1:

IBUfactor = (slope × (pH1 − pH2) / 0.7) + 1.0

[5]

where IBU_factor is a multiplication factor that accounts for pH changes.  If we have a pH change from 5.75 to 5.30 when the malt-derived IBU level is 1.50, then this reduction in pH increases the IBUs by a factor of 2.11, from 1.50 to 3.16.

5.3 Modeling IBUs as a Function of pH and Specific Gravity
We can now create a model of how IBUs change as a function of specific gravity.  The problem from Section 5.1, that we don’t know how much of the change in measured IBUs is due to the effect of specific gravity or the effect of pH, can now be addressed by removing the estimated influence of pH from the measured IBU values.

For this model, we can use the five conditions of fermented beer, namely Conditions A, B, D, E, and F.  For each condition, we can (a) hypothesize a scaling factor that models IBUs as a function of specific gravity, (b) use the estimated post-boil specific gravity with this scaling factor to predict an IBU level before pH changes, (d) modify this IBU level by the change in pH caused by boiling (and, in Condition C, lowering of the pH with phosphoric acid), and (e) compute the difference (error) between this predicted IBU value and the measured IBU value for this condition.  We can then search over a large number of scaling factors and find the factor that minimizes the mean squared error (or RMS error) over all five conditions.  Using gravity points to represent specific gravity, this results in (a) a scaling factor of 0.0190 with a root-mean-square (RMS) error of 0.13 IBUs when using a constant value of 5.75 for pH₁ and (b) a scaling factor of 0.0193 with a RMS error of 0.07 when using pH₁ estimated from the data.  Therefore, we can predict IBUs with an unadjusted wort pH as follows:

IBUbase = (OG − 1.0) × 1000) × 0.019

[6]

where IBU_base is the IBU level predicted with an unadjusted wort pH, and OG is the post-boil specific gravity, also known as the original gravity.  The subtraction of 1.0 and multiplication by 1000 converts from specific gravity to gravity points.

We can then modify this base IBU value by the change in pH caused by pH adjustments and/or boiling, by combining Equations [5] and [6]:

IBUmalt = ((OG − 1.0) × 1000) × 0.019) × ((1.734 × (pH1 − pH2) / 0.7) + 1.0)

[7]

where IBU_malt is the final predicted malt-derived IBU value, pH₁ (if unknown) can be approximated as 5.75, and pH₂ is the post-boil pH.  (Equation [7] could obviously be greatly simplified, but I’ll leave it in this form to preserve a record of the derivation.)

Figure 3 shows the IBU values from this experiment as a function of original gravity, after removing the effect of pH using Equation [5] (i.e. dividing the measured IBU value by the result of Equation [5]).  The dashed gray line in Figure 3 shows the IBUs predicted by Equation [6].

Figure 3. IBU values normalized by pH, plotted as a function of specific gravity.  The dashed gray line shows a linear fit to the data with scaling factor 0.019.

5.4 IBUs as a Function of Boil Time
The green line in Figure 3 (at specific gravity 1.076) shows the IBUs from Conditions E and F, with boil times of 10 and 60 minutes, respectively, after accounting for the effect of pH.  From these two points it is easy to conclude that all of the increase in IBUs seen with a boiling time of 60 minutes can be accounted for by the reduction in pH during the boil.  Therefore, the boiling of wort has no impact on malt-derived IBUs other than the associated change in pH.

5.5 Polyphenol Concentrations as a Function of Boil Time
In the same way that malt-derived IBUs appear to be affected by wort pH, malt-derived polyphenol concentrations appear to be affected by boil time.   After a 10-minute boil time the polyphenol concentration is 222 mg/L (Condition E) and after 60 minutes the concentration increases to 275 mg/L (Condition F).

We can take the same process that was used to normalize and model the impact of pH on IBUs (Sections 5.2 and 5.3), and apply this process to normalize the impact of boil time on polyphenols.  This process yields a factor for the effect of boil time and a prediction of polyphenol concentration without the effect boil time as follows:

PPfactor = (1.060 × t / 222.0) + 1.0	[8]
PPbase = ((OG − 1.0) × 1000)2 × .01) + ((OG − 1.0) × 1000) × 2.0)	[9]

where PP_factor is the effect of boil time on polyphenol concentrations, t is the boil time (in minutes), and PP_base is the estimated polyphenol level before boiling.  The RMS error between measured values and values predicted from PP_factor and PP_base is 8.02 mg/L.

Figure 4 shows the polyphenol concentrations from this experiment as a function of original gravity, after removing the effect of boil time using Equation [8] (i.e. by dividing the measured polyphenol concentration by the result of Equation [8]).  The dashed gray line in Figure 4 shows the polyphenol concentration predicted by Equation [9].

The normalized polyphenol concentration, when plotted as a function of specific gravity, does not extend in a straight line back to the origin.  (A straight line intersects the horizontal axis at around 13 gravity points, implying that a specific gravity of 1.013 would have a polyphenol concentration of 0.)  It is logical that a specific gravity of 1.000 has a polyphenol concentration of 0, and so it is not clear if the relationship really is linear and there is an issue with the data, or if there is some (unknown) factor that causes polyphenols to not increase linearly with specific gravity.  (Switching from gravity points to degrees Plato does not significantly improve the linearity back to the origin.)   Therefore, I have modeled the polyphenol concentration using the quadratic function in Equation [9] and shown with a dashed gray line in Figure 4.  This function provides a compromise between the expected linear function and the best fit to the data.

Figure 4. Polyphenol concentration, normalized to boil time of 0 minutes, plotted as a function of specific gravity.  The dashed gray line shows a quadratic fit to the data.

5.6 Polyphenol Concentrations as a Function of pH
It is clear from the red lines in Figures 2 and 4 (Conditions A and B, at specific gravity 1.033) that wort pH has no effect on malt polyphenol concentrations.

5.7 The Impact of Fermentation on IBUs and Polyphenols
The orange line in Figure 3 at around specific gravity 1.052 shows the IBUs from Conditions C and D, pre- and post-fermentation, respectively.  After accounting for the effects of pH on IBUs (using the measured value of 5.773 for pH₁), fermentation appears to cause a 29% reduction in IBUs (from 1.30 to 0.925 IBUs).

The orange line in Figure 4 at around specific gravity 1.052 shows the polyphenol concentrations for the same Conditions C and D.  After accounting for the effects of boil time on polyphenol concentration, fermentation appears to cause a 33% reduction in polyphenols (from 179 mg/L to 120 mg/L)

Fermentation therefore seems to reduce both the malt-derived polyphenol concentration and the IBU level by about 30%.

5.8 The Relationship Between IBUs and Malt Polyphenols
Figure 5 shows IBUs as a function of malt polyphenol concentration, using the models in Equations [7], [8], and [9] and a pH₁ value of 5.75.  Because IBUs are dependent on pH, and polyphenol concentrations are dependent on boil time, the relationship can not be described with a single line.  Figure 5 shows several examples, with post-boil pH levels 5.10 and 5.80 and no boiling, and with post-boil pH ranging from 5.10 to 5.80 with a 60-minute boil.

Figure 5. Predicted IBUs plotted as a function of predicted polyphenol concentration.  Several cases are shown with different wort pH levels and different boil times.

For a typical beer with a 60-minute boil, original gravity of 1.050, and post-boil pH of 5.15, the estimated polyphenol concentration of 161 mg/L is predicted to yield 2.36 IBUs.

6. Comparisons
6.1 A Comparison of IBUs Derived from Malt and Hop Polyphenols
Ellen Parkin’s thesis looked at the increase in polyphenol levels at different hopping rates when dry hopping [Parkin].  She found that “an increase of 100 mg/L of [hop] polyphenols was predicted to increase the [IBU] value by 2.2” [Parkin, p. 28], so that 1 mg/L of hop polyphenols increases the IBU by 0.022.  Using Equations [7], [8], and [9], we can estimate the rate of IBU increase in her beer per 1 mg/L of malt polyphenols, and compare the two rates of increase (from hops and from malt) under the same conditions.

Parkin’s beer had a pre-boil pH of 5.10 and pre-boil specific gravity of 1.0525 (13° Plato) [Parkin, p. 19].  The preparation of wort and the fermentation are described without any mention of wort boiling.  She measured 111 mg/L of polyphenols before the addition of any hops (with standard deviation 5.1), 4.5 IBUs (standard deviation 0.7), 1.7 mg/L of isomerized alpha acids (IAA) (standard deviation 0.03), 0 mg/L of alpha acids, and 0 mg/L of humulinones [Parkin, p. 23].  The presence of IAA in the unhopped beer is not easily explained; presumably there was a small amount of contamination from a previous batch.  Using Peacock’s conversion factor of 0.734 (51.2/69.7) to translate from IAA concentration to IBU value [Peacock, pp. 161, 163], there were 3.3 malt-derived IBUs in Parkin’s unhopped beer.

From her description of the beer preparation and assuming no wort boiling, Equations [7], [8], and [9] predict 125 mg/L of malt polyphenols and 2.60 IBUs, and so the addition of 1 mg/L of malt polyphenols increases the IBU by 0.021.  While the predicted polyphenol and IBU values are somewhat different from her observed values (125 mg/L vs. 111 mg/L of malt polyphenols; 2.6 vs. 3.3 IBUs), the predicted scaling factor of 0.021 is remarkably close to her scaling factor of 0.022.  (The predicted IBU value is (just) within one standard deviation of the mean from her mean observed value, indicating a reasonable chance that these two values are not significantly different.)  Whether this closeness indicates that malt and hop polyphenols have a very similar impact on IBUs and both are affected by pH and boil time, or whether this closeness is coincidence, remains to be seen.

6.2 A Comparison of Current and Previous Results
I previously developed a formula to predict IBUs from the concentration of malt polyphenols, using original gravity as a proxy for polyphenol concentration:

IBUwort = (OG − 1.0) × 1000) × 0.025

[10]

where IBU_wort is the expected IBU level from wort polyphenols and OG is the original gravity of the beer.   While this previous scaling factor of 0.025 is 30% higher than the base scaling factor of 0.190 estimated here, the current base scaling factor is for a fairly high wort pH of 5.75.  We can use the current model to account for the change in pH in the previous experiment, and compare the (pH-adjusted) scaling factors.  Unfortunately, I did not measure the wort pH in the previous experiment.  We can, however, approximate it by fitting the pre-boil specific gravity and pH observed in this experiment to a function, and use this function to predict pH from specific gravity:

pH1 = 4.288 × exp(−32.815 × (SG − 1.0)) + 5.692

[11]

where pH₁ is the pre-boil wort pH and SG is the pre-boil specific gravity.  From Equation [11] we estimate that at the observed specific-gravity levels 1.0467, 1.0706, and 1.0878 in the previous experiment, the pre-boil pH values were 5.785, 5.734, and 5.716, respectively.  With Equation [1] we can estimate the post-boil pH values as 5.703, 5.656, and 5.640, respectively.  We can then use Equation [7] to predict IBU values of 1.07, 1.60, and 1.99, respectively.  These IBU values divided by the specific gravity (in points) leads to a scaling factor of 0.023.  This scaling factor is within 10% of the previous scaling factor of 0.025, indicating a good degree of correspondence between the results of the two experiments.

7. Conclusions
7.1 Modeling Malt-Derived IBUs and Polyphenol Concentrations
If we know the original gravity and post-boil pH, we can estimate the malt-derived IBU level using Equation [7]; if the value of pH₁ in Equation [7] is unknown, a value of 5.75 can be used. If we know the original gravity and boil time, we can estimate the concentration of malt-derived polyphenols using Equations [8] and [9].  For a beer with a 60-minute boil time, OG 1.050, and post-boil pH 5.15, we predict 2.36 malt-derived IBUs and 161 mg/L of malt polyphenols.

7.2 The Effect of pH
From the data obtained for this experiment, the malt-derived IBU level increases with a decrease in pH.  There seems to be no effect of pH on the concentration of malt polyphenols.  It may be that a decrease in pH changes the chemical structure of some malt polyphenols, giving them more light absorption at 275nm and potentially more bitterness.

The effect of pH on malt-derived IBUs is the opposite of the effect found for isomerized alpha acids and for the general class of auxiliary bittering compounds (ABCs).  This implies that the ABCs must contain a relatively small proportion of malt-derived IBUs.  This implication is supported by an analysis of the relative contribution of different ABCs, which concluded that in a typical beer using well-preserved hops, about 85% of the IBUs from ABCs are from oxidized alpha acids, 10% are from malt polyphenols, 4% from hop polyphenols, and 1% from oxidized beta acids.

7.3 The Effect of Boil Time
While the boil time has no effect on malt-derived IBUs other than the associated decrease in pH, the concentration of malt polyphenols increases during the boil.  It may be that some malt polyphenols that do not absorb light at 275nm (and are therefore potentially not bitter) become more soluble with extended heating.

7.4 The Effect of Fermentation
Both malt-derived IBUs and polyphenol concentrations decrease with fermentation, by about 30%.  While the literature indicates that this loss is not due to the effect of yeast on polyphenols [Alexander; Leiper and Miedl, p. 136], it seems reasonable that polyphenols may be lost in the trub and/or krausen deposits produced during fermentation.

7.5 Future Work
The small number of data points available for analysis means that all of the conclusions reached here are preliminary and further research is needed based on a much larger set of data.

8. Acknowledgements
I am greatly appreciative of the high-quality IBU and polyphenol analysis provided by Dana Garves at Oregon BrewLab.  Without such good and consistent measurements, it would not be possible to draw meaningful conclusions.

References

• S. Alexander, “Fear of Phenols,” in Brewing Techniques, 5 (6), 1997.  This article is kindly hosted by MoreBeer! at https://www.morebeer.com/articles/Beer_Phenols (accessed on Apr. 28, 2020).
• K. A. Leiper and M. Miedl, “Colloidal Stability of Beer” in Beer: A Quality Perspective (Handbook of Alcoholic Beverages), ed. C. Bamforth, I. Russell, and G. Stewart.  Academic Press, 1st edition, 2009.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer. Master Brewers Association of the Americas, 2009.
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer. Master Brewers Association of the Americas, 2009.

Specific Gravity and IBUs

Abstract
The specific gravity of wort is thought to have a significant impact on IBUs, with an increase in gravity associated with a reduction in IBUs.  The purpose of the two experiments in this blog post was to evaluate existing formulas of specific-gravity effects on experimental data. The results of the first experiment showed no clear trend in IBUs as a function of specific gravity, and the second experiment showed a pattern only after a 40-minute hop steep time.  From this set of data, it seems that specific gravity has no effect on IBUs at shorter steep times; it may be that previous work looked only at longer steep times.  The data from these experiments were used to construct a scaling factor for IBUs based on specific gravity: F_G = 1 − 2×exp(−1 / (slope × (SG − 1))), where F_G is the relative change in IBUs, SG is the specific gravity, slope = 1 for steep times less than 30 minutes, slope = 4.9 for steep times greater than or equal to 40 minutes, and slope = 0.39 × (t − 30) + 1 for time t between 30 and 40 minutes.  This scaling factor at 40 minutes and above is close to the average of other formulas for this effect.

1. Introduction
1.1 Terminology
A number of terms are used to refer to the specific gravity (SG) of wort as it relates to IBUs.  One term is “boil gravity,” the specific gravity during the boil (as opposed to the specific gravity before the boil) [Hall, p. 62].  Another is “original gravity,” the specific gravity at the end of the boil (as opposed to the specific gravity at the beginning of, or during the middle of, the boil). Glenn Tinseth uses the term “average gravity” to denote the average specific gravity of the wort during the boil [Tinseth, web page].  While the gravity does change during the boil, the use of different terms by different authors serves more to confuse the issue than to distinguish subtle nuances between the methods.  Therefore, all of these terms will be grouped together in this blog post under the common term “specific gravity,” or SG.

Utilization, or the kettle utilization rate, is the concentration of isomerized alpha acids (IAAs) that end up in the finished beer, divided by the concentration of alpha acids added to the kettle [e.g. Fix and Fix, p. 47].  For the same amount of alpha acids that are added, a change in utilization often corresponds to a similar change in IBUs, even though IBUs are not the same as isomerized alpha acids.  For the sake of simplicity, the two terms are used interchangeably in this blog post.

1.2 References in the Literature
References in the literature for a relationship between gravity and IBUs go back at least as far as 1965.  John Hudson noted at that time that “for the same hop rate, increase in the original gravity … results in less hop substance in beer” [Hudson, p. 482].  James S. Hough et al. say that “in wort boiling higher utilization is obtained from weak worts than from strong worts” [Hough, p. 489].  Michael Lewis and Tom Young say that “losses [of isomerized alpha acids] depend on many factors including … wort composition especially its gravity” [Lewis and Young, p. 266].  George and Laurie Fix note that the amount of utilization depends on wort gravity, and that higher gravity wort is associated with lower utilization [Fix and Fix, pp. 47-48].  These descriptions note the same trend (decreasing utilization with increasing gravity), but none of them are specific enough to enable even a rough mathematical or quantitative description.

Ian McMurrough et al. provide specific data about the effect of gravity on utilization [McMurrough, p. 106].  They found that wort gravities of 1.024, 1.040, 1.061, and 1.083 had utilizations of 57%, 47%, 45%, and 43%, respectively, with an average pH of 5.5.  While these results conform to the trend reported elsewhere in the literature, the relative difference in utilization between an SG 1.040 wort and an SG 1.083 wort is less than 10%; it is only the very low-gravity wort (SG 1.024) that shows relatively high utilization.  (The original data were presented in degrees Plato; I have converted those values to specific gravity for consistency with the rest of this post.  The data fit well to an equation with exponential decay, U = 0.888 × exp(-78.4 × (SG−1)) + 0.434, where U is the utilization and SG is the specific gravity.)  While the authors noted an average pH of 5.5, they did not indicate the range of pH values or if they adjusted the pH of the wort.  They also did not state the steep time of the hops.

Mark Malowicki looked at the rate of production of isomerized alpha acids in a buffer system (pH 5.2) with specific gravities 1.0 and 1.040 [Malowicki, pp. 38-41].  He found no difference in the production of IAAs at these two gravity levels, and speculated that the “experience-based knowledge that hop utilization decreases with increasing wort strength” could be due to losses of IAAs to trub rather than a decrease in the rate of production of IAAs.  He considered that these losses might be explained by high-gravity worts having a greater concentration of proteins that remove bitter acids from the wort [Malowicki, p. 41].

Sebastian Kappler et al. boiled 100 ppm of isomerized alpha acids for 60 minutes in worts of different strengths and looked at how much of the IAAs could be recovered from the wort after the boil [Kappler].  They found, consistent with Malowicki’s theory that proteins in the wort bind with IAAs and precipitate out of solution, that higher specific gravities resulted in less recovery of IAAs.  At SG 1.040 about 90% of the IAAs could be recovered, while at SG 1.074 only about 50% could be recovered.  The recovery rate decreased linearly with increasing gravity.

1.3 Quantitative Models
Michael Hall has summarized previous work on modeling the impact of specific gravity on IBUs [Hall].  Hall reviewed several methods, including those by Jackie Rager [Rager, pp. 53-54 (as referenced by Hall)], Randy Mosher [Mosher, pp. 108-109], Glenn Tinseth [Tinseth, web page], and Greg Noonan [Noonan, p. 215]. (Other authors of IBU formulas, such as Mark Garetz and Ray Daniels, use the Rager gravity correction factor.)  In the Rager and Noonan methods, SG has no impact on IBUs at gravities less than 1.050.  Mosher and Tinseth show increasing IBUs at gravities lower than 1.050.  At higher gravities, all four authors show a similar trend of decreasing IBUs with increasing gravity.  For example, at SG 1.075, the scaling factor of the different formulas ranges from 0.8 to 0.9 times the factor at SG 1.050.

The tabular data provided by Noonan suggest that at shorter steep times (15 minutes or less) there is less impact (and sometimes no impact) of gravity on IBUs [Noonan, p. 215], but the data are quite noisy at these shorter steep times.

1.4 Experimental Control
While these formulas (and Noonan’s table) are more useful than qualitative descriptions, it is not clear what parameters were controlled for when the models were developed.  For example, the pH of wort naturally changes with the specific gravity.  A wort made from low-alkalinity water to specific gravity 1.080 may have a pH of 5.70.  If this wort is diluted with the same low-alkalinity water to SG 1.030 (or if fresh wort is prepared with the same water to SG 1.030), the pH may increase to 5.95.  This 0.25 increase in pH can cause an 8% increase in IBUs, depending on the alpha-acid rating of the hops and the boil time.  Tinseth did not control the wort pH when he made his measurements [Tinseth, email], and so his measured changes in IBUs at different gravities may have been affected, at least to some degree, by changes in pH.  (As Prof. Tinseth has noted, there is not much variation in the production [Malowicki, p. 41] or losses [Kappler, p. 334] of isomerized alpha acids in the pH range of interest [Tinseth, email].  The pH-dependent changes I’ve observed in IBUs seem to be caused more by auxiliary bittering compounds.)  It is unknown if the other models, or McMurrough’s or Kappler’s experiments, controlled for wort pH.

1.5 Summary of Previous Work
The literature describes an increase in specific gravity as resulting in lower utilization.  The reports that quantify this effect are, however, quite varied in their results.  In the set of data provided by McMurrough et al., the effect was most pronounced at lower gravities (e.g. SG 1.024) and there was a fairly small effect (less than 10%) between SG 1.040 and SG 1.083.  Malowicki did not observe any change in the rate of production of IAAs at different gravities.  Kapper et al. found a fairly steep decrease in the recovery of IAAs as specific gravity increased, with a relative 44% decrease from SG 1.040 to SG 1.074.  Of the four available quantitative models, two describe a lower limit of 1.050 for the effect of gravity on IBUs, which contradicts the data provided by McMurrough.  At higher specific gravities, all of these models predict much less of an effect than Kappler’s results and more of an effect than McMurrough’s results.  Three of the four models are not dependent on the hop steep time, and one of them is. The lack of published procedures and experimental details makes it difficult to determine under what conditions, and to what degree, there is a relationship between gravity and IBUs.

2. Experimental Overview
The purpose of the two experiments described here was to evaluate the available models on experimental data.  The conditions within each experiment were designed to be as similar as possible with the exception of the variable being tested (specific gravity).  Each experiment consisted of four batches of beer at different specific gravity levels. The conditions of the second experiment were identical to the first, but the treatment of krausen was different.

3. Experimental Methods
For each batch of beer, wort was prepared to a target specific gravity using the amounts of Briess Pilsen Light Dried Malt Extract and 120°F low-alkalinity (49°C) water listed in Tables 1 and 2, yielding about 3.47 G (13.13 liters) of room-temperature wort in each condition.  (Within each experiment, the DME was from the same lot number.)  This wort sat for at least 90 minutes to let the pH stabilize before the pH was adjusted with phosphoric acid to a room-temperature pH of about 5.60. The measured (and temperature-corrected) pH levels and specific gravities are listed in Tables 1 and 2.

Hops were added to target about 170 ppm of alpha acids at the time of the hop addition.  This meant 1.181 oz (33.485 g) of Cascade with an alpha-acid rating at harvest of 7.05% (measured by AAR Lab) and an estimated degradation factor of 0.908 from being stored in the freezer in vacuum-sealed packaging for 11 months (using the Garetz formula [Garetz, pp. 111-114]).

The wort was boiled for 5 minutes before adding the hops, to avoid the foam associated with the start of the boil.  Immediately before the hop addition, a 12-oz (0.35 l) sample was taken for later measurement of specific gravity.  The loose hop cones were then added to the wort, defining time t = 0.  Every 10 minutes, 15-oz (0.44 l) samples were taken from the boiling wort, quickly filtered through a sieve, and cooled in an aluminum cup and ice bath.  Once they reached 75°F (24°C), the cooled samples were transferred to sanitized quart (liter) containers.  Each container was aerated for 1 minute by vigorous shaking, and the amount of Safale US-05 yeast (age 9 or 10 months) listed in Tables 1 and 2 was pitched in order to target 750,000 cells per ml and degree Plato.

The kettle was covered during the boil to minimize evaporation and the resulting changes in specific gravity.  The total hop steeping time was 40 minutes.  Soon after obtaining the last 15-oz (0.44 l) sample, another 12-oz (0.35 l) sample was taken for subsequent measurement of specific gravity.

Each of the 15-oz (0.44 l)  samples fermented for 8 days (with a small opening to vent CO₂) and was then analyzed for IBUs by Oregon BrewLab.

In Experiment #1, the krausen that formed was mixed back into the fermenting wort every day using a sanitized thin spatula.  In Experiment #2, the fermentation vessels were left undisturbed and krausen was allowed to build up on the sides of the container.

	Condition A	Condition B	Condition C	Condition D
DME	90.87 oz / 2.576 kg	72.67 oz / 2.060 kg	54.47 oz / 1.544 kg	36.27 oz / 1.028 kg
added water	3.11 G / 11.76 l	3.19 G / 12.08 l	3.28 G / 12.41 l	3.36 G / 12.73 l
measured pre-boil pH	5.60	5.60	5.61	5.60
measured pre-boil SG	1.0718	1.0589	1.0438	1.0288
measured SG at t = 0	1.0775	1.0630	1.0460	1.0300
measured SG at t = 40	1.0785	1.0635	1.0465	1.0303
target yeast pitched	0.016 oz / 0.455 g	0.013 oz / 0.369 g	0.010 oz / 0.280 g	0.007 oz / 0.189 g

Table 1. Measured and estimated values for each condition in Experiment #1.

	Condition A	Condition B	Condition C	Condition D
DME	90.87 oz / 2.576 kg	72.67 oz / 2.060 kg	54.47 oz / 1.544 kg	36.27 oz / 1.028 kg
added water	3.11 G / 11.76 l	3.19 G / 12.08 l	3.28 G / 12.41 l	3.36 G / 12.73 l
measured pre-boil pH	5.59	5.62	5.61	5.60
measured pre-boil SG	1.0725	1.0580	1.0425	1.0265
measured SG at t = 0	1.0750	1.0607	1.0460	1.0303
measured SG at t = 40	1.0768	1.0624	1.0473	1.0310
target yeast pitched	0.016 oz / 0.455 g	0.013 oz / 0.369 g	0.010 oz / 0.280 g	0.007 oz / 0.189 g

Table 2. Measured and estimated values for each condition in Experiment #2.

4. Experimental Results
4.1 Experiment #1
The left-hand side of Figure 1 shows the measured IBU values from Experiment #1.  This set of data shows the opposite trend from what is expected based on the literature: an increase in IBUs with higher specific gravity.  The effect is small but consistent across sample times, and is therefore unlikely to be due to random variation.

IBUs can be measured in unhopped beer [Shellhammer, p. 177].  These IBUs come from malt polyphenols that are one of the auxiliary bittering compounds (also called nonIAA).  Worts with higher gravity will therefore have their IBU values increased slightly by the greater concentration of malt polyphenols.  I previously developed a simple formula to predict IBUs from the concentration of malt polyphenols, based on original gravity:

IBU_wort = (OG − 1.0) × 25.0

where IBU_wort is the expected IBU level from wort polyphenols and OG is the original gravity of the beer.  (A gravity of 1.050 is predicted to yield 1.25 IBUs, which is generally consistent with levels reported by Tom Shellhammer [Shellhammer, p. 177].)

The right-hand side of Figure 1 shows adjusted IBU values that have had the estimated contribution of wort polyphenols removed.  The IBU values between conditions on the right-hand side of Figure 1 are closer to each other, with only Condition A (highest wort gravity) showing noticeably larger IBU values.  Whether this is due to an effect of specific gravity, an underestimation of the impact of malt polyphenols, or some other factor (such as Condition A having, by chance, a slightly higher concentration of alpha acids) can not be determined from this set of data.

Figure 1. IBU values from Experiment #1 at different steep times (horizontal axis) and original gravity levels. Figure 1(a) shows the measured IBU values and Figure 1(b) shows the values after removing the estimated effect of malt polyphenols on IBUs.

4.2 Experiment #2
Thinking that mixing the krausen back into the fermenting wort might have had a significant impact on the results from Experiment #1, I conducted Experiment #2 with the same conditions but letting krausen deposits build up on the sides of the fermentation vessels.  The hypothesis in this case was that utilization and IBUs really do decrease with increasing gravity, but that higher-gravity worts have (for some unknown reason) relatively more IBUs in the krausen, and so by mixing the krausen back into the wort the effect of gravity on IBUs was not seen in Experiment #1.

The measured IBU values from Experiment #2 (shown in Figure 2 on the left-hand side) show no clear difference in IBUs with a change in specific gravity, with the possible exception of an 8% relative difference between SG 1.030 (Condition D) and SG 1.075 (Condition A) at 40 minutes. After removing the expected IBUs coming from malt polyphenols (as shown in Figure 2 on the right-hand side), the difference between the conditions at 10, 20, and 30 minutes becomes even less.  The difference at 40 minutes becomes more pronounced, with a 13% relative difference between Conditions A and D.

Figure 2. IBU values from Experiment #2 at different steep times (horizontal axis) and original gravity levels.  Figure 2(a) shows the measured IBU values and Figure 2(b) shows the values after removing the estimated effect of malt polyphenols on IBUs.

5. Analysis
The data from the 40-minute steep time in Experiment #2 can be fit to an equation: F_G = 1 − 2×exp(−1 / (slope × (SG − 1))) where F_G is the relative level of IBUs (relative to the IBUs at SG 1.030), SG is the specific gravity, and slope = 4.9.  These data points and a graphical representation of the equation are shown in Figure 3.  While there is no explicit limit at SG 1.050 (as in some other methods), the impact of gravity is very close to 1.0 up to SG 1.030 and is fairly close to 1 (0.966) at SG 1.050.  The general shape of the equation is similar to that of the Mosher data (see Figure 4).

Figure 3.  Data points from Experiment #2 at a 40-minute steep time, modeling a decrease in IBUs with increasing gravity.  The data are the IBU value at the gravity indicated on the horizontal axis divided by the IBU value at gravity 1.030.  The model is a best fit to these data points.

If one accepts that there is little to no impact of gravity on IBUs at steep times less than 40 minutes, it is possible to modify this formula to reflect this time-dependent nature.  A slope value of 1.0 effectively removes any impact of gravity, and a linear interpolation between 30 and 40 minutes can smooth the transition from “no effect” to “full effect”.  This interpolation can be modeled with slope = 0.39 × (t − 30) + 1 for time t between 30 and 40 minutes.

Figure 4 shows the relative impact of gravity using the four available quantitative methods (with two steep times, 30 minutes and 60 minutes, for the Noonan method).  This figure also shows the formula derived from the data in Experiment #2 at the 40-minute steep time, labeled ‘Hosom’.  It can be seen that all of these formulas have results similar to each other, with the exception of the Tinseth formula at lower gravities.  (I labeled the new formula ‘Hosom’ to be consistent with the existing naming convention.)

Figure 4.  A visual comparison of existing methods (and the formula developed here, labeled ‘Hosom’) for accounting for the effect of specific gravity on IBUs.  The Noonan data were derived from his Table 18.  The Noonan and Mosher data are plotted relative to the predicted IBUs or utilization at the lowest gravity.

6. Conclusions
6.1 Effect of Steep Time
The data from the two experiments show no effect of specific gravity at steep times of 30 minutes or less.  It seems possible that the experiments that were conducted for developing the previous formulas (with the exception of Noonan) did not evaluate short steep times, but used a traditional boil time closer to 60 minutes.  The lack of an effect at less than 40 minutes implies that it can take a long time for the proteins in the wort to bind with the isomerized alpha acids (IAAs) (and/or auxiliary bittering compounds (ABC)) and precipitate out of solution.

It is also possible that, instead of steep time, the effect of gravity only applies at higher concentrations of IAAs and/or ABCs, and therefore only at higher IBU values.  In this case, according to Figure 2, the effect of gravity might only be observed in beers with more than 18 IBUs.  Or, there might be an interaction between time and concentration, and the combination of these two may be needed to predict when gravity will have an effect on IBUs.

6.2 Controlling for pH
Both the data from McMurrough and the formula from Tinseth show relative utilization decreasing with gravity using an exponential decay factor.  This exponential decay factor means that very low SG values have much greater relative utilization than higher SG values.  Although McMurrough et al. reported the average pH over all conditions, they did not report the pH levels at each gravity.  Tinseth did not control for pH when developing his formula because IAA levels in beer are not greatly influenced by pH.  Unless the wort pH is adjusted (e.g. by the addition of acid), the pH of wort will increase exponentially as the specific gravity decreases, and increased pH is associated with larger IBU values.  I therefore suspect that the relatively high utilization at low gravity values noted by McMurrough and Tinseth is due to a confounding of the effects of pH and gravity on IBUs.  While there may be an effect of gravity at lower SG values (e.g. less than 1.050), the effect appears to be minor.

6.3 Effects of Krausen
The IBU values from Experiment #1 are more spread out at each time point than the results of Experiment #2.  In Experiment #1, the condition with the highest SG has the highest IBU levels, and the condition with the lowest SG has the lowest IBU levels, even after accounting for the IBUs contributed by malt polyphenols.  Experiment #2 shows, except for the 40-minute steep time, very little difference between conditions after accounting for malt polyphenols.  The only (intentional) difference between Experiments #1 and #2 was in the treatment of krausen.  In Experiment #1, krausen was mixed back into the fermenting beer once a day; in Experiment #2, krausen deposits were allowed to form on the sides of the fermentation vessel.  The overall differences between Experiments #1 and #2 show an  expected increase in IBUs from mixing krausen back into the fermenting beer, but the SG-dependent pattern in Experiment #1, while slight, is unexpected.

If this effect is real, it seems that (a) there is less loss of IBUs with higher-gravity worts during fermentation, or (b) there is greater production of (auxiliary) bittering compounds in the krausen during fermentation with high-gravity worts.  It is difficult to envision why a lower-gravity wort would lose more IBUs than a higher-gravity wort during fermentation.  It is possible that alpha (or beta) acids still present in the krausen are oxidized and transformed into bitter substances (or that IAAs may transform back into oxidized alpha acids [Verzele and De Keukeleire, p.116]), and that the greater amount of foamy krausen in higher-gravity beers facilitates this transformation, but this is pure conjecture.  It is also quite possible that these subtle differences are due to unintended variation between the experimental conditions (such as slightly more alpha acids ending up in Condition A than in Condition D).  Additional experiments would be required to replicate the observed pattern and identify the reason for this apparent  trend.

6.4 Future Work
While this blog post was originally intended to be a fairly straightforward evaluation of existing formulas on experimental data, the results bring up more questions than answers.  The experiments described here are therefore just a first step to a better understanding of how specific gravity affects IBUs.  In the future, it would be interesting to evaluate data with: (a) boil times ranging from 10 to at least 60 minutes, (b) a greater range of specific gravity levels, (c) different initial concentrations of alpha acids, (d) different wort pH levels, and (e) different treatment of krausen.

7. Acknowledgements
I greatly appreciate the helpfulness and quick responses of both Glenn Tinseth and Randy Mosher in response to my out-of-the-blue questions.  That both of these luminaries were happy to answer my questions is a testament to the spirit of cooperation and support that makes homebrewing a wonderful hobby.

I am also always greatly appreciative of the high-quality IBU analysis provided by Dana Garves at Oregon BrewLab.  Without such consistent accuracy, it would not be possible to draw meaningful conclusions from the data.

References

• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques.  Brewers Publications, 1997.
• M. Garetz, Using Hops: The Complete Guide to Hops for the Craft Brewer. HopTech, 1st edition, 1994.  Also see “Hop Storage: How to Get – and Keep – Your Hops’ Optimum Value” in Brewing Techniques, January/February 1994, hosted on morebeer.com.
• M. L. Hall, “What’s Your IBU,” in Zymurgy.  Special Edition, 1997.
• J. R. Hudson, “The Rationalization of Hop Utilization — A Review,” in Journal of the Institute of Brewing, vol. 71, pp. 482-489, 1965.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science.  Volume 2: Hopped Wort and Beer.  Springer-Science+Business Media, B. V., 2nd edition, 1982.
• S. Kappler, M. Krahl, C. Geissinger, T. Becker, M. Krottenthaler, “Degradation of Iso-alpha-Acids During Wort Boiling,” in Journal of the Institute of Brewing, vol. 116, no. 4, pp. 332-338, 2010.
• I. McMurrough, K. Cleary, F. Murray, “Applications of High-Performance Liquid Chromatography in the Control of Beer Bitterness,” in Journal of the American Society of Brewing Chemists, vol. 44, no. 2, pp. 101-108, 1986.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• R. Mosher, The Brewer’s Companion.  Alephenalia Publications, Seattle, WA 1995.
• G. J. Noonan, New Brewing Lager Beer. Brewers Publications, 1996.
• J. Rager, “Calculating Hop Bitterness in Beer,” Zymurgy Special Issue 1990 (vol. 13, no. 4), pp. 53-54. (as referenced by M. L. Hall)
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• G. Tinseth, web page, “Glenn’s Hop Utilization Numbers”.  Accessed most recently on Dec. 6, 2019.  http://realbeer.com/hops/research.html
• G. Tinseth, email: personal e-mail communication with Glenn Tinseth on September 3, 2018.
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids, vol. 27, 1st edition, Elsevier,  ISBN 0-444-88165-4, eBook ISBN 9781483290867, 1991.

 

The Relative Contribution of Oxidized Alpha- and Beta-Acids to the IBU

Abstract
The IBU combines the concentration of isomerized alpha acids (IAAs) and the concentration of “auxiliary bittering compounds” (ABCs) in beer into a single measure of approximate bitterness.  While IAAs contribute the most to the IBU in typical beers, ABCs play a significant role and may have concentrations greater than IAAs in very late-hopped beers.  The auxiliary bittering compounds are composed of polyphenols, oxidized alpha acids, and oxidized beta acids.  This blog post estimates the relative contribution of oxidized alpha acids and oxidized beta acids to the IBU, using data from beer brewed with four varieties of hops.  The data indicate that when using well-preserved hops, the concentration of oxidized alpha acids in beer is much greater than the concentration of oxidized beta acids. It is estimated that the auxiliary bittering compounds in most beers made with well-preserved hops are composed primarily of oxidized alpha acids, with much lower contributions from malt polyphenols, hop polyphenols, and finally oxidized beta acids. When hops have been stored for long periods with exposure to oxygen, however, the oxidized beta acids may contribute significantly to the IBU.

1. Introduction
The IBU is a measure of the concentration of a number of different bitter compounds.  (To be more precise, the IBU is a measure of the absorbance of light at 275 nm through acidified beer.  A number of bitter compounds in beer absorb light at this frequency.  The greater the concentration of these compounds, the more light is absorbed, and the higher the IBU.)  In typical beers, the IBU value represents mostly the concentration of isomerized alpha acids (IAAs) [Peacock, pp. 164-165], which are produced during the boil from alpha acids (AA).  The other bitter compounds, known as “auxiliary bittering compounds” (ABCs), or nonIAA, are polyphenols, oxidized alpha acids, and oxidized beta acids.  These compounds can be considered to be present in the wort soon after the hops addition [e.g. Dierckens and Verzele, p. 454; Askew, p. 18].  (Alpha acids, which are present in wort [Hough et al., p. 491] and also absorb light at 275 nm [Hough et al., p. 434, p. 491], are not bitter [Shellhammer, p. 169] but also not typically present in the fermented beer [e.g. Lewis and Young, p. 259, Hough et al., p. 491].  Therefore, they do not contribute to the measured IBU value of beer except when dry hopping.)

While “typical” beers (if there is such a thing anymore) have a much greater concentration of IAAs compared to ABCs, beers produced with a large amount of hops added very late in (or after) the boil can have ABC concentrations greater than IAAs.  In modeling IBUs and getting a better understanding of the bitterness qualities of late-hopped beers, it is beneficial to have an estimate of the relative concentrations of the compounds that are collectively referred to as ABCs.  While the contribution of malt and hop polyphenols to the IBU is generally known (as discussed in Section 4), a quantitative analysis of the relative contribution of oxidized alpha- and beta-acids is not easily found in the literature.

This blog post estimates the relative concentrations of oxidized alpha- and beta-acids in beer by analysis of IBU values.  These IBU values were obtained by ASBC Beer-23A analysis of samples taken at different points in the boil from four beers brewed with different varieties of hops (and different alpha-acid and beta-acid ratings).

2. The Concentration of Isomerized Alpha Acids in Beer
Mark Malowicki developed formulas to estimate the concentration of IAAs in the wort from the initial concentration of alpha acids [Malowicki, p. 27]:

k1(T) = 7.9×1011 e-11858/T	[1]
k2(T) = 4.1×1012 e-12994/T	[2]
[IAA]wort = [AA]0 × (k1(T)/(k2(T) − k1(T))) × (e–k1(T)t − e–k2(T)t)	[3]

where k₁(T) and k₂(T) are empirically-derived rate constants, T is the temperature in Kelvin (i.e. 373.15 K for boiling), t is the steep time (in minutes), e is the constant 2.71828, and [AA]₀ is the initial concentration of alpha acids in the wort (in ppm).  This concentration of IAA in the wort, [IAA]_wort, decreases as IAAs are lost to trub and krausen during the boil and fermentation.  The concentration of IAAs in beer ([IAA]_beer) can then be expressed as the IAA concentration in wort multiplied by a loss scaling factor, scaling_IAA:

[IAA]beer = [IAA]wort × scalingIAA

[4]

3. The IBU Expressed as Concentrations of Bitter Compounds
Val Peacock [Peacock, p. 161] provides an equation to express the IBU as a combination of the concentration of IAAs and ABCs (also called nonIAA):

IBU = 5/7 × ([IAA]beer + [ABC]beer)

[5]

where IBU is the IBU value of the beer, [IAA]_beer is the concentration of isomerized alpha acids in the finished beer (in ppm, from Section 2), and [ABC]_beer is the concentration of all other bittering compounds (also in ppm).

The concentration of ABCs in beer ([ABC]_beer) can be expressed as the sum of the concentrations of the individual ABC compounds multiplied by appropriate scaling factors that relate each concentration to absorption at 275 nm:

[ABC]beer = [PPmalt]beer × scalePPmalt + [PPhops]beer × scalePPhops + [oAA]beer × scaleoAA + [oBA]beer × scaleoBA

[6]

where [PPmalt]_beer, [PPhops]_beer, [oAA]_beer, and [oBA]_beer are the concentrations in the beer of malt polyphenols, hop polyphenols, oxidized alpha acids, and oxidized beta acids, respectively, and scale_PPmalt, scale_PPhops, scale_oAA, and scale_oBA are the scaling factors that relate concentration to absorption at 275 nm for these compounds.

Alternatively, we can express the concentration of ABCs in beer as the concentration of total hop particles added to the wort, multiplied by a single scaling factor that accounts for (a) the proportion of each ABC compound to total hop matter, (b) different absorption factors of these compounds, and (c) losses of each ABC to trub and during fermentation:

[ABC]beer = [hops]wort × scalingABC

[7]

where [hops]_wort is the concentration of hops added to the wort and scaling_ABC is the above-mentioned scaling factor.

We can estimate the scaling factors scaling_IAA and scaling_ABC from these equations and measured IBU values of beer samples fermented from wort taken at different time points during the boil.  This technique is described in the blog post Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements.

4. The Concentration of Polyphenols in Beer
Polyphenols in beer come from both malt and hops. This section describes how to estimate their concentration in beer and their impact on IBUs.

4.1 Malt Polyphenols
According to Tom Shellhammer, IBUs are in the range of 1 to 3 for unhopped beer [Shellhammer, p. 177].  In one experiment, I brewed several unhopped beers and developed a formula for predicting IBUs that come from malt polyphenols:

IBUPPmalt = ((OG − 1.0) × 1000) × 0.019) × ((1.734 × (5.75 − pH) / 0.7) + 1.0)

[8]

where IBU_PPmalt is the IBU value obtained from malt polyphenols, OG is the original gravity of the beer, and pH is the post-boil pH.  This formula predicts 2.4 IBUs from an OG of 1.050 and post-boil pH of 5.15, which is generally in line with Shellhammer’s statement.  We can convert this formula from a prediction of IBUs to a prediction of scaled malt polyphenol concentrations if we multiply by 7/5 (from Equations [5] and [6]):

[PPmalt]beer × scalePPmalt = 7/5 × IBUPPmalt

[9]

where [PP_malt]_beer is the concentration of malt polyphenols in the finished beer (in ppm) and scale_PPmalt is the scaling factor for light absorption at 275 nm.  This scaling factor is dependent on both pH and boil time, but we often don’t need to determine the separate values of [PP_malt]_beer and scale_PPmalt; knowing their product, as specified by 7/5 × IBU_PPmalt, is usually sufficient.  (The implication of a non-constant scaling factor is that not all of the malt polyphenols contribute to the IBU; those that do contribute are affected by pH and time.)

4.2 Hop Polyphenols
Hop polyphenol levels are often reported in the range from 2% to 6% of the weight of the hops [Shellhammer, p. 169; Hough et al., p. 422; Algazzali, p. 5; Verzele and De Keukeleire, p. 9].  After having been added to the wort, polyphenols are removed “extensively by precipitation with proteins during wort boiling”; 80% of hop flavanols are removed in the trub when boiling hopped wort [McLaughlin, p. 7]. Yeast has only a minor effect on polyphenol concentrations during fermentation [Leiper and Miedl, p. 136; Alexander], but losses to trub and krausen may be similar to those of malt polyphenols, estimated at 30%.  Finally, we need a scaling factor, scale_PPhops, to use with the concentration of hop polyphenols in Equation [6].  According to Ellen Parkin, “an increase of 100 mg/L of polyphenols was predicted to increase the [IBU] value by 2.2” [Parkin, p. 28], so that 1 ppm of hop polyphenols should increase the IBU by 0.022.  We can multiply this by 7/5 to convert from an IBU value to a concentration scaling factor (using Equations [5] and [6]).

From this, we can construct a model of the concentration of hop polyphenols in beer and the associated scaling factor, with an initial level of polyphenols at 4% of the weight of the hops, a loss factor (or combined solubility and loss factor) for polyphenols during the boil estimated at 0.20 (corresponding to 80% loss), and a loss factor of 0.70 (corresponding to 30% loss) during fermentation:

[PPhops]beer = 0.04 × 0.20 × 0.70 × W × 1000 / V	[10]
scalePPhops = 7/5 × 0.022 = 0.0308	[11]

where [PP_hops]_beer is the concentration of hop polyphenols in the beer (in ppm), W is the weight of hops added (in grams), V is the final wort volume (in liters), and scale_PPhops is the scaling factor for light absorption at 275 nm.

4.3 Relative Contributions of Malt and Hop Polyphenols
The majority of polyphenols in beer come from malt.  According to Steve Alexander, “roughly 75% of total beer phenolics come from malt, and the remaining 25% come from hops” [Alexander].  Denis De Keukeleire states that “hops may contribute up to about one third of the total polyphenols in beer” [De Keukeleire, p. 109].  According to Cynthia Almaguer et al., “about 20–30% of the polyphenols found in the wort come from the hop material” [Almaguer, p. 300].

5. Oxidized Alpha- and Beta-Acids in Beer
Alpha acids (before isomerization) “do not survive to any significant extent into beer” [e.g. Lewis and Young, p. 259] and are not bitter [Shellhammer, p. 169], but as they age and become oxidized, the resulting oxidized alpha acids (oAAs) are both soluble in wort and bitter [Algazzali, pp. 14-15, p. 19, p.45; Maye et al, p. 23; Hough et al., pp. 435-436; Hough et al., p. 439; Lewis and Young, p. 265].  Oxidized alpha acids are also produced during the boil [Parkin, p. 11, Algazzali, p. 17; Dierckens and Verzele, p. 454; Oliver p. 471], but the amount of oxidized alpha acids produced in this way is unclear.

Oxidized beta acids (oBAs) are also soluble [Algazzali, p. 16] and may be produced and contribute to bitterness in the same way as oxidized alpha acids [Malowicki, p. 2; Peacock, p. 157; Fix, p. 36; Lewis and Young, p. 265; Hall, p. 55; Oliver, p. 132; Oliver, p. 470; Parkin, p. 11; Algazzali, p. 17; Hough et al., p. 489]. Val Peacock says that “the beta oxidation products contribute to analytical IBUs” [Peacock, p. 161] and that “much of the nonIAA material originates from the oxidation of the beta acids as hops age” [Peacock, p. 164].  Oxidized beta acids may also be produced during the boil [Spetsig, p. 350].  Stevens and Wright provide an estimate of the amount of oxidized beta acids produced during the boil, noting that “as much as 10% of the beta acid had been converted into cohulupone” [Stevens and Wright, p. 500].  If 10% of the beta acids are oxidized during the boil, and if oxidized beta acids contribute to analytical IBUs, then oxidized beta acids can contribute to a significant portion of ABCs even when using fresh hops. If beta acids are not oxidized during the boil, then the contribution of oxidized beta acids may depend on how well the hops have been stored.  In this case, poorly-preserved hops may have a greater contribution of oxidized beta acids to the IBU than well-preserved hops.  Almaguer notes that “further oxidation of the [oxidized beta acids] will result in the non-bitter hulupinic acid” [Almaguer, p. 295].  It is not clear how quickly the oxidized beta acids are converted into hulupinic acid.  This process might occur very quickly, in which case even oxidized beta acids from poorly-stored hops will not contribute to the IBU.  Or, this transformation might occur over weeks or months, in which case oxidized beta acids may contribute significantly to the IBU.

In addition to the concentrations of these products in beer, [oAA]_beer and [oBA]_beer, we’re also concerned with how their concentrations relate to IBU measurement via the scaling factors scale_oAA and scale_oBA.  According to the data in Maye et al. (Figure 7), the scaling factor for scale_oAA is 0.0130 / 0.0142 = 0.9155 [Maye, p. 25].   According to Hough et al., “[oxidized beta acids] exhibit 80-90% of the absorption of the iso-alpha-acids at [275 nm in acid solution]” [Hough et al., p. 491].  Therefore, the oxidized beta-acid scaling factor is about 0.85.

scaleoAA = 0.0130 / 0.0142 = 0.9155	[12]
scaleoBA = 0.85	[13]

Although oxidized alpha- and beta-acids are both bitter and soluble, potentially contributing to the IBU measurement, the literature is not clear on the relative proportion of oxidized alpha acids to oxidized beta acids produced during the boil and ending up in the finished beer.

6. Experimental Overview
To evaluate the relative proportion of oxidized alpha- and beta-acids in beer, I designed two new experiments and took IBU data from three earlier experiments. The new experiments used Teamaker hops, known for having almost no alpha acids and a very high concentration of beta acids.

The three other experiments are described in the blog posts The Impact of Krausen Loss on IBUs, The Effect of pH on Utilization and IBUs, and Hop Cones vs. Pellets: IBU Differences.  I took data from Condition B of the Krausen Loss experiment, Condition A of the Effect of pH experiment, and Condition A of the Cones vs. Pellets experiment because those conditions were most similar to the new experiment with the exception of hop variety.  The primary difference between the different sets of data was in the variety of hops used, with different alpha- and beta-acid ratings, and in the storage conditions of the Teamaker hops.

7. Experimental Details and Results
7.1 New Experiment #1 (Experiment #1)
The first new experiment used well-preserved Teamaker hops.  Wort was prepared from 4.13 lbs (1.87 kg) of Briess Pilsen Light Dried Malt Extract and 3.22 G (12.20 liters) of 120°F (49°C) low-alkalinity water, yielding 3.48 G (13.17 liters) of room-temperature wort.  This wort sat for 90 minutes to let the pH stabilize.  The measured pre-boil specific gravity was 1.054.  The wort was boiled for 5 minutes before adding hops in order to reduce the foam associated with the start of the boil.

I used 2.01 oz (57.0 grams) of Teamaker hops in this experiment.  The hops were analyzed by AAR Lab and showed an alpha-acid (AA) rating of 0.41% and a beta-acid (BA) rating of 11.93%, in line with expectations.  (I had harvested and dried these hops four weeks earlier and stored them in a vacuum-sealed bag in my freezer.  The resins were a deep orange color, in contrast with the normal bright yellow.  The AAR results also showed 13% moisture and a hop storage index (HSI) of 0.188.)

When the hops were added to the kettle (defined as time t = 0), the (temperature-corrected) specific gravity was 1.0565 and the pH was 5.74.  The estimated volume at the time when hops were added was 3.36 G (12.71 liters), estimated from the initial volume and gravity and the measured gravity at time 0.

After adding the hops, 15-oz (0.44 l) samples were periodically taken from the boiling wort and quickly cooled in an aluminum cup and ice bath.  Samples were taken after the hops had been steeping for 5, 10, 20, 40, and 60 minutes.  The kettle was covered during the boil to minimize evaporation and the resulting changes in specific gravity.  Once they reached 75°F (24°C),  the cooled samples were transferred to sanitized quart (liter) containers.  The wort in each container was aerated for 1 minute by vigorous shaking, and 0.011 oz (0.31 grams) of Safale US-05 yeast (age 7 months) was pitched to target 750,000 cells per ml and degree Plato.  At the end of the 60-minute boil, the specific gravity was 1.0577 and the pH was 5.69.

Each sample fermented for 8 days (with a small opening to vent CO₂).  The krausen was left to deposit on the sides of the vessel during fermentation.  After a week the krausen was more “billowy” than normal, still covering the beer.  I removed this krausen one day before taking samples for IBU analysis by Oregon BrewLab.

The measured IBU values at 5, 10, 20, 40, and 60 minutes were 3.3, 3.6, 3.8, 3.8, and 4.0, respectively.  These values are plotted with a solid red line in Figure 1.

7.2 Results from Experiment on Krausen Loss (Experiment #2)
The full details of the experiment on krausen loss are given in that blog post.  In summary, for Condition B, the gravity when the hops were added (t = 0) was 1.037 and the estimated volume was 8.38 G (31.72 liters).  I added 1.88 oz (53.31 grams) of Comet hops to the wort with an alpha-acid rating of 9.7% and a beta-acid rating of 3.7%, for an initial alpha-acid concentration of about 160 ppm.  The gravity at the end of the boil was 1.039 and the estimated volume was 8.06 G (30.49 liters).  The measured IBU values of the resulting beer are plotted with a solid green line in Figure 1.

7.3 Results from Experiment on Impact of pH on Utilization and IBUs (Experiment #3)
The full details of the experiment on the effects of pH on IBUs are given in that blog post.  In summary, for Condition A, the gravity when the hops were added (t = 0) was 1.038 and the estimated volume was 4.25 G (16.09 liters).  I added 0.65 oz (18.28 grams) of Citra hops to the wort with an alpha-acid rating of 14.2% and a beta-acid rating of 3.35%, for an initial alpha-acid concentration of about 160 ppm.  The gravity at the end of the boil was 1.039 and the estimated volume was 4.08 G (15.43 liters).  The measured IBU values of the resulting beer are plotted with a solid blue line in Figure 1.

7.4 Results from Experiment on Cones and Pellets (Experiment #4)
The full details of the experiment on the differences between hop cones and pellets are given in the blog post Hop Cones vs. Pellets: IBU Differences.  In summary, for Condition A, the gravity when the hops were added (t = 0) was 1.039 and the estimated volume was 8.28 G (31.36 liters).  I added 3.72 oz (105.57 grams) of Willamette hops to the wort with an alpha-acid rating of 5.0% and a beta-acid rating of about 3.4%, for an initial alpha-acid concentration of about 170 ppm.  The gravity at the end of the boil was 1.039 and the estimated volume was 8.16 G (30.88 liters).  The measured IBU values of the resulting beer are plotted with a solid orange line in Figure 1.

Figure 1. Measured IBUs (solid lines) and model IBUs (dotted lines) from the four experiments in this blog post. Experiment #1 (red line) used 57.0 g of Teamaker hops with an alpha-acid rating of 0.4%. Experiment #2 (green line) used 53.3 g of Comet hops with an alpha-acid rating of 9.7%. Experiment #3 (blue line) used 18.3 g of Citra hops with an alpha-acid rating of 14.2%.  Experiment #4 (orange line) used 105.6 g of Willamette hops with an alpha-acid rating of 5.0%.

7.5 New Experiment #2 (Experiment #5)
The second new experiment had two conditions: (A) one condition using Teamaker hops stored in oxygen-barrier packaging in a freezer for six months (well-preserved hops), and (B) one condition using the same harvest of Teamaker hops stored with exposure to oxygen and at room temperature for six months (poorly-preserved hops).

Wort for each condition was prepared from 3.02 lbs (1.37 kg) of Briess Pilsen Light Dried Malt Extract and 3.31 G (12.53 liters) of 120°F (49°C) low-alkalinity water, yielding 3.50 G (13.25 liters) of room-temperature wort. This wort sat for 90 minutes to let the pH stabilize. The measured pre-boil specific gravity was 1.037. The wort was boiled for 5 minutes before adding hops in order to reduce the foam associated with the start of the boil.

I used 2.0 oz (56.7 grams) of Teamaker hops in each condition. The hops were analyzed by AAR Lab within one week of the experiment and showed an alpha-acid (AA) rating of 0.64% and a beta-acid (BA) rating of 10.92% for the well-preserved hops, and an AA rating of 0.57% and a BA rating of 3.61% for the poorly-preserved hops. (The AAR results also showed 9.8% moisture and a hop storage index (HSI) of 0.291 for the well-preserved hops, and 7.5% moisture and an HSI of 1.050 for the poorly-preserved hops.)

When the hops were added to the kettle (defined as time t = 0), the specific gravity was 1.041 and the pH was 5.87. The estimated volume at the time when hops were added was 3.21 G (12.15 liters).

After adding the hops, 15-oz (0.44 l) samples were periodically taken from the boiling wort and quickly cooled in an aluminum cup and ice bath. Samples were taken after the hops had been steeping for 10, 20, and 40 minutes. The kettle was covered during the boil to minimize evaporation and the resulting changes in specific gravity. The cooled samples were transferred to sanitized quart (liter) containers once they reached 75°F (24°C). The wort in each container was aerated for 1 minute by vigorous shaking, and 0.009 oz (0.25 grams) of Safale US-05 yeast (age 10 months) was pitched to target 750,000 cells per ml and degree Plato. At the end of the 60-minute boil, the specific gravity was 1.042 and the pH was about 5.78.

Each sample fermented for 10 days (with a small opening to vent CO₂). Very little krausen accumulated on the sides of the vessel during fermentation. IBU values were measured by Oregon BrewLab.

The measured IBU values for the well-preserved hops at 10, 20, and 40 minutes were 4.2, 5.1, and 5.7, respectively. The measured IBU values for the poorly-preserved hops at 10, 20, and 40 minutes were 18.0, 22.1, and 25.9, respectively. These values are plotted with blue (well-preserved hops) and red (poorly-preserved hops) lines in Figure 2.

Figure 2. Measured IBU values from well-preserved Teamaker hops (blue line) and poorly-preserved Teamaker hops (red line).

8. Analysis
We now have all of the information we need to estimate the concentrations of oxidized alpha acids and oxidized beta acids in finished beer.  First, we will consider oxidized beta acids produced during the boil using well-preserved hops.  Then we will consider what happens when using poorly-preserved hops

8.1 Analysis Step 1
The first step is to estimate the concentrations of IAAs, malt polyphenols, and hop polyphenols in the beer and (using the associated scaling factors) subtract their effects from the measured IBUs, yielding the “remaining” IBUs that come from oxidized alpha- and beta-acids.  This step is repeated for all four experiments.  The rest of this section explains this step in more detail and provides the analysis results from each experiment.

To estimate the IAA concentration, we can use the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to determine the values of scaling_IAA and scaling_ABC for each experiment.  The results of fitting the data to this model are shown with the dashed red, green, blue, and orange lines in Figure 1 for Experiments #1, #2, #3, and #4, respectively.

We can combine Equations 4, 5, and 6 to be explicit about all of the contributions to the IBU, and highlight variables with “known” values in green and variables with “unknown” values in red:

IBU = 5/7 × ([IAA]wort × scalingIAA + ([PPmalt]beer × scalePPmalt) + [PPhops]beer × scalePPhops + [oAA]beer × scaleoAA + [oBA]beer × scaleoBA)

[14]

where “known” variables can have their values measured, computed, or estimated from the experimental data and equations described above. We can define IBU_remaining to represent the difference between the measured IBU and the other “known” portion of Equation [14] as follows:

IBUremaining = IBU − 5/7 × ([IAA]wort × scalingIAA + [PPmalt]beer × scalePPmalt + [PPhops]beer × scalePPhops)

[15]

and note that, by rearranging terms from Equations [14] and [15], IBU_remaining is the sum of the (scaled) contributions from oxidized alpha- and beta-acids:

IBUremaining = 5/7 × ([oAA]beer × scaleoAA + [oBA]beer × scaleoBA)

[16]

We can define the IBUs that are derived from oxidized alpha- and beta-acids as IBU_oAA and IBU_oBA, and note their relationship with IBU_remaining:

IBUoAA = 5/7 × ([oAA]beer × scaleoAA)	[17]
IBUoBA = 5/7 × ([oBA]beer × scaleoBA)	[18]
IBUremaining = IBUoAA + IBUoBA	[19]

While we don’t (yet) know [oAA]_beer or [oBA]_beer, we can compute IBU_remaining from Equation [15].

For each of the four experiments, Table 1 lists (a) the root-mean-square (RMS) error between the measured IBU values and the model that solves for scaling_IAA and scaling_ABC, (b) the IBUs at time 0 (when the hops are added to the boiling wort) according to this model, at which point the ABCs have been added to the wort but there is not yet any production of isomerized alpha acids, (c) the estimated IBUs from malt polyphenols (determined from the original gravity and Equation [8]), (d) the estimated IBUs from hop polyphenols (determined from Equations [10] and [11]), and (e) the remaining IBUs according to Equation [15].

In Experiment #1, there was an initial alpha-acid concentration of 18 ppm and an initial beta-acid concentration of 535 ppm.  In Experiment #2, the initial alpha-acid concentration was 163 ppm and the initial beta-acid concentration was 53 ppm.  In Experiment #3, the initial alpha-acid concentration was 161 ppm and the initial beta-acid concentration was 38 ppm.  In Experiment #4, the initial alpha-acid concentration was 168 ppm and the initial beta-acid concentration was 116 ppm.

	IBU model RMS error	model IBUs at time 0	IBUs from malt polyphenols	IBUs from hop polyphenols	remaining IBUs
Experiment #1	0.136	3.21	1.23	0.55	1.42
Experiment #2	0.314	6.42	0.86	0.21	5.35
Experiment #3	0.295	8.11	0.85	0.14	7.12
Experiment #4	0.399	8.78	0.86	0.42	7.50

Table 1. Results from the IBU model for each of the four experiments.  The “RMS error” is the root-mean-square error between the measured IBUs and the model.  The “model IBUs at time 0” show the model with the effect of ABC but no isomerized alpha acids.  The “IBUs from malt polyphenols” shows the estimated IBUs contributed by the malt.  The “IBUs from hop polyphenols” shows the estimated IBUs contributed by the hop polyphenols.  The “remaining IBUs” shows the difference between the model IBU at time 0 and the effect of polyphenols, which is the effect of oxidized alpha- and beta-acids.

8.2 Analysis Step 2
Assuming that the hops have been preserved well and have undergone very little oxidation during storage, we can express the concentration of oxidized alpha acids in the beer as the initial concentration of alpha acids added to the wort multiplied by a scaling factor, lossFactor_oAA, that accounts for (a) the percentage of alpha acids that oxidize during the boil and (b) the percentage of oxidized alpha acids that remain after boiling and fermentation.  The same formulation can be applied to the beta acids:

[oAA]beer = [AA]0 × lossFactoroAA	[20]
[oBA]beer = [BA]0 × lossFactoroBA	[21]

where [AA]₀ is the initial concentration of alpha acids (in ppm), [BA]₀ is the initial concentration of beta acids (in ppm), lossFactor_oAA is the scaling factor for alpha acids, and lossFactor_oBA is the scaling factor for beta acids.

We can then note the following relationships by combining Equations [17, 18, 20, 21]:

IBUoAA = 5/7 × ([AA]0 × lossFactoroAA × scaleoAA)	[22]
IBUoBA = 5/7 × ([BA]0 × lossFactoroBA × scaleoBA)	[23]

All of this preparation has laid the groundwork for the formula that will be useful to us, obtained by combining Equations [19, 22, 23]:

IBUremaining = 5/7 × [AA]0 × lossFactoroAA × scaleoAA + 5/7 × [BA]0 × lossFactoroBA × scaleoBA

[24]

This formula expresses the remaining IBU value as a combination of the initial concentrations of oxidized alpha- and beta-acids multiplied by scaling factors and loss factors.  We know IBU_remaining, [AA]₀, scale_oAA, [BA]₀, and scale_oBA, and we want to estimate lossFactor_oAA and lossFactor_oBA.

We can solve for these two loss factors by using the data from the four experiments (obtained in Analysis Step 1) to construct four equations (based on Equation [24]) with the two unknowns:

1.42 = 5/7 × 18.38 × lossFactoroAA × 0.9155 + 5/7 × 534.87 × lossFactoroBA × 0.85	[25]
5.35 = 5/7 × 163.03 × lossFactoroAA × 0.9155 + 5/7 × 53.28 × lossFactoroBA × 0.85	[26]
7.12 = 5/7 × 161.35 × lossFactoroAA × 0.9155 + 5/7 × 38.06 × lossFactoroBA × 0.85	[27]
7.50 = 5/7 × 168.32 × lossFactoroAA × 0.9155 + 5/7 × 116.14 × lossFactoroBA × 0.85	[28]

We can perform a least-squares estimation, minimizing the root-mean-square (RMS) error between predicted and observed remaining IBU values, to solve for lossFactor_oAA and lossFactor_oBA.  The results of this analysis are that lossFactor_oAA = 0.059 and lossFactor_oBA = 0.0023, with RMS error 0.76 IBUs.

Independent of any particular batch of beer, lossFactor_oAA and lossFactor_oBA describe our best estimate of the contribution of oxidized alpha- and beta-acids to the IBU.  For any batch of beer where we know [AA]₀ and [BA]₀, we can use lossFactor_oAA and lossFactor_oBA to estimate [oAA]_beer and [oBA]_beer, and/or their contributions to the IBU.   (These estimates may be further influenced by wort pH, age of the beer, the treatment of krausen, among other factors.)

8.3 Analysis of Aged Hops
The experiment comparing well-preserved and poorly-preserved hops clearly indicates that oxidation of the beta acids produced during storage can yield significant IBUs. (I find it remarkable that hops with only 0.57% alpha acids and 3.61% beta acids can yield a beer with more than 20 IBUs.  This beer not only tasted more bitter or astringent, it had a more “herbal” quality, similar to Ricola “natural herb” throat drops.)  In Figure 2 it can be seen that the IBU values increase with boil time, which was unexpected. Is there some isomerization of oxidized beta acids that occurs along the same time-frame as the isomerization of alpha acids? Because there is no well-motivated explanation for why these IBU values increase, I will simplify the analysis by focusing on a single “representative” boil time, i.e. 20 minutes.

The increase in oxidized beta acids produced during aging at room temperature can be computed from the decrease in beta acids between the two conditions. The beta acids decreased from 10.92% to 3.61%, and so the increase in oxidized beta acids was 7.31% of the weight of the hops. With a volume of 3.16 G (11.96 liters) at the end of the boil and a weight of 2.0 oz (56.70 grams) of hops, the 4741 ppm of hops translates into 346.55 ppm of oxidized beta acids in the post-boil wort. If we assume various relative losses during fermentation and aging (fermentation loss of 0.85, krausen relative increase of 1.38 compared with normal krausen deposits, and an age-related loss factor of 0.96), there are 391 ppm of oxidized beta acids in the finished beer. The increase in IBUs between the two conditions (at a 20-minute boil time) is 17.0 IBUs. We can then map between IBUs and an “oxidized beta acid boil factor”, boilFactor_oBA, which expresses the reduction in storage-produced oxidized beta acids during the boil:

IBUoBA = 5/7 × ([oBA]0 × boilFactoroBA × fermentFactoroBA × scaleoAA)	[29]
17.0 = 5/7 × (346.55 × boilFactoroBA × 1.126 × 0.85)	[30]

and so boilFactor_oBA equals 0.07. This implies that about 7% of the beta acids that oxidize during storage survive the boil and end up in the wort. This factor is about the same as the factor for oxidized alpha acids produced during the boil (0.06), indicating that both can contribute significantly to the IBU.

9. Conclusion
The estimated scaling factor lossFactor_oAA (0.059) being 25 times larger than the scaling factor lossFactor_oBA (0.0023) for well-preserved hops means that the oxidized alpha acids contribute much more to the IBU than oxidized beta acids, as long as the hops are well preserved.  The estimated contribution of oxidized beta acids is so low in this case, it seems quite likely that oxidized beta acids are not produced during the boil at all.

For poorly-preserved hops, however, the contribution from oxidized beta acids produced during storage appears to be roughly equal to the contribution from oxidized alpha acids produced during the boil. (The better the storage conditions, the less impact that beta acids will have on the IBU.)  With oxidized beta acids produced during aging and present in finished beer, the reduction in IBUs from using older hops with a lower alpha-acid content is offset by the oxidation products. As Hough et al. say, “The level of alpha-acid in hops falls during storage but the bittering potential of the hops does not fall to the same extent. This is because many of the oxidation products of both alpha- and beta-acids … are capable of bittering beer” [Hough et al., p. 489].

If we consider a “typical” beer produced with well-preserved hops, such as an American Pale Ale as described by Ray Daniels [Daniels, pp. 167-172], we might have an original gravity of 1.050, a post-boil volume of 5.25 G (19.87 liters), and a post-boil pH of 5.25.  This beer might have five hop additions, all with 9% AA and 5.0% beta acids: one of 0.75 oz (21.26 g) at the start of the 60-minute boil, a second of 0.75 oz (21.26 g) at 45 minutes before flameout, a third of 0.50 oz (14.18 g) at 20 minutes before flameout, a fourth of 0.75 oz (21.26 g) at 10 minutes before flameout, and a dry-hop addition of 1.0 oz (28.35 g).  In this case, we might get 50.0 IBUs in total, 33 of those from isomerized alpha acids and 17 from ABCs.  Of these 17 ABC IBUs, using the model described above, oxidized alpha acids contribute 14.0 IBUs, oxidized beta acids contribute 0.3 IBUs, malt polyphenols contribute 2.1 IBUs, and hop polyphenols contribute 0.7 IBUs.  The oxidized alpha acids are therefore by far the greatest component of the auxiliary bittering compounds (at 82% of all ABC), followed by malt polyphenols (12%), then hop polyphenols (4%), and finally oxidized beta acids (2%). A single hop addition at 10 minutes before flameout will have about as many IBUs coming from oxidized alpha acids as from isomerized alpha acids.  This can be seen in Figure 1, with about 13.5 IBUs at a 10-minute steep time averaged over Experiments #2, #3, and #4, and about 7.5 IBUs estimated at time 0 when isomerization begins.

Even if you are uncomfortable with some (or all) of the assumptions made in this model, it is still clear from the measured IBUs in Figure 1 that oxidized beta acids produced during the boil can not contribute significantly to the IBU.  First, there is the fact that 535 ppm of beta acids in Experiment #1 yielded less than 4 IBUs, and at least some of those IBUs come from polyphenols.  Additionally, if we compare the results of Experiments #3 and #4, the IBU values are very similar.  Although the alpha-acid ratings of the hops in the two experiments were very different (14.2% and 5.0%), the amount of hops added to the kettle was set to target a similar alpha-acid concentration (160 to 170 ppm).  This resulted in Experiment #3 having an initial beta-acid concentration of 38 ppm and  Experiment #4 having an initial beta-acid concentration of 116 ppm.  The similarity of the IBU values between the two experiments, with three times the beta-acid concentration in Experiment #4, can best be explained by the alpha acids (both isomerized and oxidized) contributing to the vast majority of the IBU and the beta acids contributing very little to the IBU value.

10. Discussion
From the results of these experiments, it appears that (a) there is little or no production of oxidized beta acids during the boil, (b) about 7% of the oxidized beta acids produced during storage end up in the wort and finished beer, and (c) any transformation of oxidized beta acids to hulupinic acid ([Almaguer, p. 295]) occurs slowly (e.g. over the course of weeks or months).

It is interesting to note similarities between the Rager [Rager] and Tinseth [Tinseth] IBU formulas and the model described here.  The Rager formula predicts 5% utilization even for a steep time of 0 minutes, which correlates extremely well with the estimated lossFactor_oAA of 0.059 in this blog post.  In both cases, before there is any significant isomerization, about 5% of the available alpha acids contribute to the IBU (in the form of oxidized alpha acids).  Tinseth, on the other hand, knew that isomerization can be modeled with a first-order reaction [Tinseth], and so the shape of the Tinseth utilization curve is similar to the rate of alpha-acid isomerization determined by Malowicki (Equation [3]).

The slopes of the lines formed by the IBU values in Experiments #2 and #3 imply that Experiment #2 had a higher concentration of alpha acids than Experiment #3 or less loss of isomerized alpha acids.  The AA rating of the hops would have to be increased from 9.7% to 12.6% for the loss of IAAs to be the same.  While AA levels can be highly variable even within the same bale of hops [Verzele and DeKeukeleire, p. 331], a 30% variation is larger than one would normally expect.  It is therefore unclear why there is such a difference between the results of these two experiments.  My best guess is that small differences in my degassing procedure when preparing samples for IBU analysis resulted in less loss of IAAs in Experiment #2, and that this effect was combined with normal variation in AA levels.  (I have seen a 12% difference in IBUs from the same beer that was degassed in slightly different ways.)  On the other hand, the similarity of the results between Experiments #3 and #4 is remarkable given the difference in AA ratings, amount of hops used, and other various differences.

11. Acknowledgment
I would like to thank Dana Garves at Oregon BrewLab for her attention to quality and detail that is reflected in the IBU measurements presented here.  The change in measured IBU values over time very closely follows the expected trend, even to the point of a fractional increase between 5 minutes and 60 minutes for the well-preserved Teamaker hops with an AA rating less than 1%.  An analysis can only be as good as the data it is based on, and so I greatly appreciate the data of such high caliber.

References

• S. Alexander, “Fear of Phenols”, in MoreBeer! articles, Nov. 30, 1999. Accessed most recently on Nov. 8, 2019.  https://www.morebeer.com/articles/Beer_Phenols
• V. A. Algazzali, The Bitterness Intensity of Oxidized Hop Acids: Humulinones and Hulupones, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• C. Almaguer, C. Schönberger, M. Gastl, E. K. Arendt, and T. Becker, “Humulus lupulus – a story that begs to be told: A review,” in Journal of the Institute of Brewing, vol. 120, pp. 289-314, 2014.
• H. O. Askew, “Changes in Concentration of α and β Acids and of Iso-Compounds on Heating Extracts of Hops in Aqueous Solutions and Unhopped Wort,” in Journal of the Institute of Brewing, vol. 71, pp. 10-20, 1965.
• R. Daniels, Designing Great Beers.  Brewers Publications, 1st edition, 1996.
• D. De Keukeleire, “Fundamentals of Beer and Hop Chemistry”, in Química Nova, vol. 23, no. 1, pp. 108 – 112, 2000.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science. Volume 2: Hopped Wort and Beer. Springer-Science+Business Media, B. V., 2nd edition, 1982.
• K. A. Leiper and M. Miedl, “Colloidal Stability of Beer” in Beer: A Quality Perspective (Handbook of Alcoholic Beverages), ed. C. Bamforth, I. Russell, and G. Stewart.  Academic Press, 1st edition, 2009.
• J. Dierckens and M. Verzele, “Oxidation Products of Humulone and Their Stereoisomerism,” in Journal of the Institute of Brewing, vol. 75, pp. 453-456, 1969.
• G. Fix, Principles of Brewing Science. Brewers Publications, 2nd edition, 1999.
• M. L. Hall, “What’s Your IBU,” in Zymurgy.  Special Edition, 1997.
• J. S. Hough, D. E. Briggs, R. Stevens, and T. W. Young, Malting and Brewing Science.  Volume 2: Hopped Wort and Beer.  Springer-Science+Business Media, B. V., 2nd edition, 1982.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• J. P. Maye, R. Smith, and J. Leker, “Humulinone Formation in Hops and Hop Pellets and Its Implications for Dry Hopped Beers”, in MBAA Technical Quarterly, vol. 51, no. 1, pp. 23-27, 2016.
• I. R. McLaughlin, Bitterness Modifying Properties of Hop Polyphenols, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• G. Oliver, The Oxford Companion to Beer, Oxford University Press, 2011.
• E. J. Parkin, The Influence of Polyphenols and Humulinones on Bitterness in Dry-Hopped Beer, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2014.
• V. Peacock, “The International Bitterness Unit, its Creation and What it Measures,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• J. Rager, “Calculating Hop Bitterness in Beer”, in Zymurgy Special Issue, vol. 13, no. 4, pp. 53-54, 1990.
• T. H. Shellhammer, “Hop Components and Their Impact on the Bitterness Quality of Beer,” in Hop Flavor and Aroma: Proceedings of the 1st International Brewers Symposium, ed. Thomas H. Shellhammer, Master Brewers Association of the Americas, 2009.
• L. O. Spetsig, “The Bitter Substances of Spent Hops, Trub, and Yeast Cover: A Chromatographic Study,” in Journal of the Institute of Brewing, vol. 74, pp. 346-351, 1967.
• R. Stevens and D. Wright, “Evaluation of Hops [Part] X. Hulupones and the Significance of β Acids in Brewing,” in Journal of the Institute of Brewing, vol. 67, 1961.
• G. Tinseth, “Glenn’s Hop Utilization Numbers”.  Accessed most recently on Nov. 8, 2019.  http://realbeer.com/hops/research.html
• M. Verzele and D. De Keukeleire, Chemistry and Analysis of Hop and Beer Bitter Acids.  Developments in Food Science 27.  Elsevier, 1991.

The Impact of Krausen Loss on IBUs

Abstract
It is often recommended to remove the krausen during fermentation for a “smooth bitterness.”  Some brewers accomplish this through the use of a blow-off tube and a small headspace in the fermentation vessel.  Many brewers do nothing about krausen, allowing most of it to fall back into the beer.  This post looks at how the removal of krausen affects IBUs by measuring IBUs resulting from different amounts of krausen loss and different hop steep times.  The data show that losses of krausen to deposits on the walls of the fermentation vessel can have a small (5% to 10%) impact on IBUs, and that the loss of krausen through a blow-off tube can result in more than a 25% reduction in IBUs.  This effect on IBUs is quantified with separate adjustments for isomerized alpha acids (IAA) and the auxiliary bittering compounds (nonIAA) that both contribute to measured IBUs.  The results indicate that both IAA and nonIAA are lost with the removal of krausen, but that the loss of nonIAA is about three times greater than the loss of IAA.

1. Introduction
Krausen (or kraeusen, kräusen, or barm) is the foam that forms on top of fermenting beer, in varying shades of white, off-white, brown, and even green (from the hops).  This foam consists of yeast, hop resins, and wort proteins [Palmer, p. 89] and some of it tends to adhere to the sides of the fermentation vessel (FV).

It is often, but not always, recommended that the krausen be removed.  Kai Troester, citing Narziss and Kunze, says that “if a smooth bitterness is desired, [krausen] should be removed via blow-off tube [or] skimming and not allowed to fall back into the beer”  [Troester (pH)].  However, Troester also states that “common brewing advice in American home brewing is to let the Kraeusen fall back into the beer after primary fermentation finishes” [Troester (Krausen)].  Letting krausen fall back into the beer during fermentation appears to also currently be the norm in the UK.

According to John Palmer, “these compounds [in krausen] are very bitter and if stirred back into the wort, would result in harsh aftertastes. Fortunately these compounds are relatively insoluble and are typically removed by adhering to the sides of the fermentor as the krausen subsides” [Palmer, p. 89].

According to Mark Garetz, “if [krausen] is stirred back into the wort at the proper time, hop utilization is increased by some 18%.  But what usually happens is that these iso-alpha acids are lost.  In commercial practice, this head may be skimmed.  On the homebrew level, it may be blown out if the brewer uses the ‘blow-off’ method.  Otherwise it is pushed to the sides of the fermenter where it sticks.   … that which doesn’t stick to the fermenter walls will fall back through the beer, but not be redissolved.  So regardless of the fermentation method, these alpha acids are lost” [Garetz, p. 126-127].

Lewis and Young say (in 2001) that the “world’s biggest brewing operation” in the U.S. removes krausen through the use of a sloping roof on their fermenter and a small headspace that forces the krausen into a “foam chamber,” and that this is not unlike the Burton Union System [Lewis and Young, p. 304].

Hough et al. say that “the head gradually collapses, leaving a dark-colored, bitter-tasting scum which should be separated from the beer by skimming or suction.  Some breweries arrange for this scum to stick to the roof of the fermenter and then be removed by special chutes at the side of the vessel.” [Hough et al., pp. 652-653].

Noonan says that during the late krausen stage, all of the krausen can be “floated, siphoned, or skimmed off, even as more is forming, so that it does not fall back through the beer” [Noonan, p. 184].

Troester tested the impact of krausen removal on the taste of beer.  He found that some people were able to detect a difference, but that others were not.  (Some people are better at detecting differences in bitterness levels than others, and this difference is thought to be genetic.)  Those who could detect a difference preferred the beer with krausen removed, describing it as having a “cleaner aftertaste” that Troester states is desired in “any German style beer” [Troester (Kraeusen)].

Jake Huolihan at Brülosophy looked at the impact of skimming krausen during fermentation [Huolihan] and found that, similar to the results from Troester, nearly half of the participants in his experiment were able to detect a difference between beers with and without krausen removal.  He did not look at the perception of bitterness quality by those who were able to detect a difference.  Huolihan was able to reliably detect the beer with krausen removal, perceived that  beer to be less bitter, and had a slight preference for that beer.

Given that the krausen is bitter, as indicated by Hough et al. [p. 652] and Palmer [p. 89], it seems plausible that some of the isomerized alpha acids (IAA) and possibly the auxiliary bittering compounds (nonIAA) adhere to the proteins and/or yeast in the krausen, and so the removal of krausen reduces the level of IBUs and bitterness in the remaining beer.  The experiment described here tests this hypothesis of a reduction in IBU levels caused by the removal of krausen.

2. Overview of the Experiment
This blog post looks at the impact of krausen removal on IBUs.  Using a single batch of wort, several conditions were created:

1. A: All of the krausen was gently mixed back into the fermenting beer,
2. B: The krausen was skimmed off once a day,
3. C: The beer was fermented in a FV with a large headspace and gently swirled to reduce krausen deposits,
4. D: The beer was fermented, undisturbed, in a FV with a large headspace,
5. E: The beer was fermented, undisturbed, with a small headspace that encouraged more krausen deposits to be created on the sides and top of the FV, and
6. F: The krausen was removed through the use of a FV with a blow-off tube and very small headspace.

For conditions A and B, small samples of wort were taken at 10-minute intervals during the 60-minute boil, and each sample was fermented into beer.   For other conditions, a larger single sample was fermented.  IBUs were measured for all samples in each condition.  The measured IBU values indicate the impact of each method of krausen removal on IBUs.  Having multiple samples from the same condition (specifically, conditions A and B) allows us to use multiple IBU measurements to estimate the loss factors for IAA and nonIAA separately, which allows us to compare the impact of krausen loss on IAA and nonIAA.

3. Experimental Procedures
In order to minimize any possible effects caused by removing samples of wort during the boil, I used as large a batch size as I dared in my 10 G (40 l) kettle.  I used 7.0 lbs (3.18 kg) of Briess Pilsen DME in 8.0 G (30.28 l) of water, yielding about 8.45 G (31.99 l) of room-temperature wort with a specific gravity of about 1.037.

I used hops from a 1 lb (0.45 kg) bag of Comet hops from Hops Direct that were purchased soon after harvest and subsequently stored in a vacuum-sealed bag in my freezer.  This bag had an alpha-acid rating on the package of 9.9%.  I had these hops analyzed by Advanced Analytical Research one month before brew day (which was about six months after harvest), and the result showed an alpha-acid content of 9.70%.

I added hops (i.e. started the steep time at 0) after the wort had been boiling for 5 minutes, to avoid the foam associated with the start of the hot break.  The hops were boiled for a steep time of 60 minutes with the cover on the kettle (except for the first few minutes and for taking samples) to minimize evaporation and volume changes.  I did not use a mesh bag with the hop cones.  Just prior to adding hops, I took samples for pH measurement.  The pH at 65°F (18°C) was 5.82, which can be normalized to a room-temperature pH of 5.80.

I targeted an initial alpha-acid concentration of less than 170 ppm in order to not exceed the solubility limit of around 200 ppm at boiling.  With an expected evaporation loss of 0.17 G (0.63 l) from time spent heating the wort and waiting 5 minutes before adding hops, I expected 8.38 G (31.72 liters) of wort when the hops were added.  With an AA rating of about 9.70%, 1.88 oz (53.31 g) were added to achieve an initial concentration of about 163 ppm.

Samples for Conditions A and B were taken every 10 minutes.  Each sample (about 32 oz (0.95 l)) was taken from the boil in a measuring cup and then quickly transferred to a large aluminum cup using a wire mesh sieve to remove larger hop particles.  The aluminum cup was placed in an ice bath, and the wort was stirred to cool quickly.  Once cooled to 77°F (25°C), the sample was transferred to a sanitized, sealed, and labeled 1.8-quart (1.7-liter) container.  I aerated each sample by vigorous shaking for 60 seconds, then added about .016 oz (0.47 g) of Safale US-05 yeast (age 7 months) to target 750,000 viable cells per ml and degree Plato [Fix and Fix, p. 68].  (The process of taking a sample, cooling it, transferring it to a sanitized container, aerating, and pitching yeast took almost 10 minutes.)  At the end of the boil, an additional sample was taken for measuring original gravity and pH.  The OG and room-temperature pH of the wort at the end of the boil were 1.039 and 5.66, respectively.

The wort was then quickly chilled (using a hydra wort chiller) to 75°F (24°C) and 3.5 G (13.25 l) of settled wort were transferred to a sanitized carboy.  This wort was aerated for 90 seconds using a mix-stir, and 0.233 oz (6.60 g) of the same Safale US-05 yeast was pitched.  Within 12 hours the wort was separated into four different 1-gallon (128-oz or 3.78-liter) fermentation vessels for Conditions C, D, E, and F.

For Conditions A and B, after all samples were taken during the boil, equal amounts (about 16 oz or 0.48 l) of each sample were transferred to two sanitized quart (liter) containers for fermentation.  The container lids were cracked open to allow carbon dioxide to vent.  For Condition A, once a day I used a sanitized thin rubber spatula to gently remove deposits from the sides of the six containers.  For Condition B, once a day I used sanitized paper towels (heated in an oven at 300°F (150°C) for 10 minutes and then stored in a ziplock bag) to remove the krausen by folding and then gently skimming the towels over each of the six samples.  (By heating the paper towels in this way, the outermost towel (in a layer of six) turned brown but did not burn.  Your experience in using this technique may be different, and I strongly recommend caution if using this approach.)

Condition C consisted of 64 oz (1.89 l) of wort with an airlock, and once a day I gently swirled the fermenting wort to try to reduce the deposits on the side of the FV.  Condition D consisted of 64 oz (1.89 l) of wort with a blow-off tube and was left undisturbed during fermentation.  Condition E consisted of 111 oz (3.28 l) of wort and a blow-off tube, so that the krausen would come into contact with, and hopefully stick to, the top of the FV.  Condition F consisted of 128 oz (3.78 l) of wort with a tiny headspace and blow-off tube, so that a large amount of the krausen would be forced through the blow-off tube. Conditions E and F were undisturbed during fermentation.  The blow-off tubes for Conditions D, E, and F all went into a container of Saniclean (diluted to the recommended level) to prevent air from flowing back into the FV.

After 11 days of fermentation, 4 oz (0.12 l) of each sample was measured for IBUs by Oregon BrewLab.  The final gravity of all samples was about 1.0020 (minimum 1.0015; maximum 1.0030).  The final pH of all samples was about 4.10 (minimum 4.08, maximum 4.15).

4. Experimental Results
Mixing the krausen deposits back into the wort worked as expected for Condition A.  Skimming the krausen worked as expected for Condition B, but new krausen formed fairly soon after removal.  Skimming off krausen several times a day might yield different results.

I wasn’t able to fully remove the krausen deposits from Condition C using gentle swirling, so Conditions C and D seemed to have very similar amounts of krausen deposits.   The krausen from Condition E did not stick as much as I had expected to the top of the FV, but seemed to have somewhat more krausen deposits than Conditions C or D.  Condition F pushed a significant amount of krausen through the blow-off tube for the first few days, and then the volume of fermenting beer was lowered enough that additional krausen remained in the FV.

IBU results for Conditions A and B are shown in Table 1 and Figure 1.   For Conditions C, D, E, and F, IBU values were 38.2, 37.4, 36.7, and 30.3, respectively.

	10 min	20 min	30 min	40 min	50 min	60 min
Condition A (krausen mixed in)	16.8	21.6	27.7	32.7	35.5	41.0
Condition B (krausen removed)	13.4	18.7	23.4	27.8	31.2	35.1

Table 1. Results of IBU analysis for Conditions A and B.  In Condition A, the krausen deposits were gently mixed back into the fermenting beer.  In Condition B, the krausen was removed by skimming once a day.

Figure 1.  Plot of the measured IBU values for Condition A (dark-blue line and points) and Condition B (dark-green line and points). In Condition A (dark blue), the krausen deposits were gently mixed back into the fermenting beer.  In Condition B (dark green), the krausen was removed by skimming once a day.  This figure also shows the estimated IBU values from the best fit to the model described in Section 5, with the light-blue line for Condition A and the light-green line for Condition B.

5. Analysis
In this section, an IBU loss of X% is the same as a loss factor F, where F = (100% − X%)/100%.  For example, if Condition X has 50 IBUs and Condition Y has 45 IBUs, the loss factor F from X to Y is 45 / 50 = 0.90 (in other words, Y = F × X), and this is the same as a 10% loss (0.90 = (100% − 10%)/100%).

The IBU values of Conditions A and B at the 60-minute sample time can be generally compared with the IBU values from the other conditions, because the wort for the other conditions was cooled very quickly after the 60-minute steep time.

If we consider Condition A at 60 minutes (41 IBUs) to have no krausen loss, we can compare other conditions with Condition A as a baseline.  In this case, a ring of krausen deposits yielded an IBU loss of about 8% (with 7% in Condition C and 9% in Condition D), and at least some krausen stuck to the top of the FV (Condition E) yielded a loss of 10%.  Presumably, more krausen stuck to the top of the FV would yield greater loss. Daily removal of krausen by skimming (Condition B) yielded a loss of 14%, and removal of krausen through a blow-off tube (Condition F) yielded an IBU loss of 26%.

We can use the technique described in Estimating Isomerized Alpha Acids and nonIAA from Multiple IBU Measurements to obtain separate estimates of the isomerized alpha acid (IAA) and nonIAA contributions to the IBU for Conditions A and B.  This technique uses a model of IAA production (and degradation) developed by Mark Malowicki [Malowicki, p. 27] and a model of nonIAA components that doesn’t vary with boil time.  The technique finds the best fit of scaling factors for losses of IAA and nonIAA given the model’s estimate of production of IAA and the measured IBU values.  We can then estimate how much of the decrease in IBUs between Conditions A and B comes from IAA losses and how much from nonIAA losses.  Because IBUs measure concentration but the volume of the wort decreases during the boil, we can normalize the measured IBUs to a single volume in order to directly compare IBU values from all samples. Using this estimation technique and this IBU normalization, the best estimate for Condition A is an IAA scaling factor of 0.52 and a nonIAA scaling factor of 0.0063.  (In other words, just over half of the IAA that was produced during the boil was lost during the boil and fermentation, and of the total concentration of hop matter in the wort, 1/159 ended up contributing to the IBU as polyphenols and oxidized alpha- and beta acids.)  These scaling factors yield a root-mean-square (RMS) error of 0.73 IBUs between the model and the measured IBU values, indicated by the light-blue line in Figure 1.  The best estimate for Condition B is an IAA scaling factor of 0.47 and a nonIAA scaling factor of 0.0045, with an RMS error of 0.30 IBUs, indicated by a light-green line in Figure 1.  These four scaling factors imply that the removal of krausen in Condition B caused a 9.6% loss of IAA and a 28.6% loss of nonIAA.  While the amount of krausen and IBU loss varies with each condition, if we assume that the relative loss of IAA and nonIAA is constant, this suggests that with krausen loss, the loss of nonIAA is 2.97 times greater than the loss of IAA.

Some might say that because the measured IBU value of Condition A at 60 minutes (41.0 IBUs) is greater than the model value (40.0 IBUs), and because the model represents the combination of six data points instead of a single data point, then a better representation of Condition A in comparison with other conditions is 40 IBUs.  While this may be true, it also took slightly longer to cool the 6.5 G (24.6 l) of wort (even using the hydra wort chiller) than the 32 oz of wort in Conditions A and B, which would slightly increase the IBUs for Conditions C through F.  Also, the relationships drawn from the data in this blog post must necessarily be preliminary, as they are based on a very tiny amount of data.  So, I wouldn’t put a lot of emphasis on 8% loss compared with 10% loss, or 31% loss compared with 29% loss.

6. Conclusion
In general, it seems that the more krausen is lost to deposits, skimming, or blow-off, the greater the reduction in IBUs.  Mixing the krausen back into the beer yields the highest IBU levels.  Leaving krausen deposits on the side of the FV will reduce the IBU level somewhat relative to mixing in the krausen; in this study, there was an IBU loss of 7% to 10%.  Removing krausen by daily skimming results in more loss, with a loss of 14% seen here.  Removing krausen through a blow-off tube results in the most loss, at 26% in this study.  The loss of nonIAA to krausen is about three times greater than the loss of IAA.

In contrast with some of the literature,  it appears that simply letting the krausen stick to the walls of the FV does not remove krausen (or lower the bitterness level) as effectively as skimming or blow-off.  In normal homebrewing practice, with minor krausen deposits on the FV walls, the impact on IBUs is similar to mixing the krausen back into the fermenting beer.  Removing krausen through the use of a blow-off tube was the most effective at reducing IBUs in these experiments, although skimming might have had more of an impact if performed more often than once a day.

It seems plausible that both IAA and nonIAA components bind to proteins and/or yeast in the krausen, and when the krausen is removed the IBUs therefore decrease.  The data suggest that losses of nonIAA components are about three times those of IAA.

The reason given to remove krausen is to promote a “smoother” bitterness [Troester (Kraeusen)].  (And, presumably the “world’s largest brewery” wouldn’t remove krausen [Lewis and Young, p. 304] without some kind of purpose.)  This change in bitterness quality may not be noticeable to all drinkers and might be more pronounced in lagers than in ales [Troester (Kraeusen)].  While I am fairly sure that I am not very sensitive to bitterness, I tasted samples from Conditions C (38.2 IBUs) and F (30.3 IBUs) to see if I could detect a difference.  I performed four blind tastings at different times in the same day, using 2 samples from Condition C and 1 sample from Condition F in two tastings, and 1 sample from C and 2 from F in the other two tastings.  The goal was to detect which beer of the three was different, and if I succeeded, which beer I preferred.  The samples were served uncarbonated and at room temperature.  I found myself guessing which one was Condition C and which was Condition F, knowing that one had more IBUs than the other.  I had a consistent preference for the beer that I thought was less bitter.  I correctly identified 10 of the 12 samples, but the two errors in two different tastings meant that I was able to identify the odd-beer-out only half the time.  As to the question of whether F had a “smoother bitterness” or if F was simply less bitter, I couldn’t say.  (It would be interesting to compare two beers with very different amounts of krausen removal but the same measured IBU levels.  Such an experiment would be well suited to the folks over at Brülosophy.)  My wife, who prepared the samples for tasting but otherwise knew nothing of the different conditions, interestingly described Condition C as having an “overripe peach” aroma that Condition F didn’t have.  This suggests that in addition to reducing bitterness, krausen removal may also decrease the level of aromatic hop compounds.  These observations are, of course, very preliminary and in no way conclusive.

It should be noted that it is difficult to quantify how much krausen is deposited or lost.  It is also difficult to predict how much krausen will be produced during fermentation.  In particular, the krausen deposits on the FV for Condition E were not as large as hoped; more impact may have been seen with a more vigorous production of krausen.  In general, though, one may be able to put krausen deposits into one of several categories, such as “light deposits,” “heavy deposits,” “moderate blow-off loss,” and “large blow-off loss.”  Then, the impact on IBUs can be approximated from the specified category.

Acknowledgement
Many thanks (again) to Dana Garves at Oregon BrewLab for the IBU analysis.  Without such consistent analysis results within and across conditions, this blog post would not have been possible.

References

• G. J. Fix and L. A. Fix, An Analysis of Brewing Techniques. Brewers Publications, 1997.
• M. Garetz, Using Hops: The Complete Guide to Hops for the Craft Brewer. HopTech, 1st edition, 1994.
• J. S. Hough, D. E. Briggs, R. Stevens, T. W. Young, Malting and Brewing Science: Volume II Hopped Wort and Beer.  Springer-Science+Business Media, 2nd edition, 1982.
• J. Huolihan, The Impact of Skimming Kräusen During Fermentation, Brülosophy.  Accessed July 8, 2019.
• M. J. Lewis and T. W. Young, Brewing. Springer Science+Business Media, 2nd edition, 2001.
• M. G. Malowicki, Hop Bitter Acid Isomerization and Degradation Kinetics in a Model Wort-Boiling System, Master of Science thesis (advisor: T. H. Shellhammer), Oregon State University, 2005.
• G. J. Noonan, New Brewing Lager Beer. Brewers Publications, 1996.
• J. Palmer, How to Brew.  Brewers Publications, 2006, 3rd edition.
• K. Troester (Kraeusen), “Should the Kraeusen fall back into the beer?”.  http://braukaiser.com/blog/blog/2010/02/14/should-the-kraeusen-fall-back-into-the-beer/.  Accessed June 9, 2019.
• K. Troester (pH), “How pH affects brewing”.  http://braukaiser.com/wiki/index.php/How_pH_affects_brewing.  Accessed June 9, 2019.