Understanding the Science Behind Beer Foam

When logic and proportion have fallen sloppy dead/And the White Knight is talking backwards/and the Red Queen’s “off with her head”/ Remember what the dormouse said/ “Feed your head/Feed your head.”
— “White Rabbit” by Jefferson Airplane

There are two aspects of good beer foam: formation and stability. Beer foam is formed by the interaction of various malt proteins, isomerized alpha acids, metal ions, and carbon dioxide (and beer). The three main proteins are protein Z, hordein species and lipid transfer protein 1 (LPT1). If you think about the structure of foam being like building a skyscraper, protein Z and hordeins are the girders, LPT1 are the struts, isomerized alpha acids are the gusset plates and metal ions are the rivets. The iso-alpha acids and metal ions act to connect the proteins and bind the structure together.

When a beer is first poured, the foam is considered to be “wet” and contains a relatively high proportion of beer. As the beer drains from the foam, the foam dries out, and the bubbles begin to collapse or coalesce. The coalescence of the bubbles is driven in large part by the surface viscosity of the beer, which is partly driven by alcohol content and the type of gas in the bubble. Highly stable foams will exhibit more lacing on the sides of the glass. The stability of the foam depends upon the relative amounts and types of the various components listed above, the surface viscosity and on any foam inhibiters present. The primary foam destabilizers are lipids. Lipids are a broad class of fat-soluble natural compounds, including fats, oils, waxes, sterols, glycerides and fatty acids. If you are wearing lipstick or eating a bag of chips, the head on your beer will dissipate quickly.

Foam proteins: protein Z

The malt proteins are the building blocks of foam. The first malt protein to be specifically identified with beer foam was protein Z, an albumin (water soluble) with a molecular weight of 40,000 that is often glycosylated (combined with a sugar molecule) as a result of Maillard (browning) reactions during malting and boiling. Protein Z has been found to have the highest surface viscosity and elasticity properties of the malt proteins, and these properties are further enhanced by glycosylation. Modern malting barley varieties with high diastatic power typically have high amounts of protein Z, and it appears, along with LPT1, to be the primary building block in beer foam. The amount of protein Z in the wort is primarily dependent on barley variety and is not significantly affected by proteolytic enzymes in the mash, but it is does seem to depend on a minimum degree of malt modification. Experiments have shown that malts with a Kolbach Index (ie. soluble-to-total protein ratio) less than about 40% generally have low amounts of protein Z, while malts with a ratio greater than 40% (i.e., well-modified) have generally high levels. In less-modified malts, the contribution of protein Z to foam is substantially aided or supplanted by malt hordein proteins.

Foam proteins: hordein

The hordein family of proteins are the major storage proteins of barley, and are generally insoluble until hydrolyzed by enzymatic action, either during malting or mashing. The amount of hordein in the malt, like protein Z, is fairly consistent for each malting barley variety. Various hordeins have been isolated from beer foam, and these range in size from a molecular weight of less than 5,000 to more than 50,000, but one in particular that has been found to be concentrated in beer foam versus the beer has a molecular weight of 23,000.

While large hordeins can be broken down into smaller and potentially more foam-active hordeins by the mash, it needs to be pointed out that haze-active hordeins span the same size range as foam-active hordeins, so using a protein rest on highly modified malts to promote foam will also tend to promote beer haze.

Interestingly, research has shown that some hordein species foam more easily than protein Z, and appear to preferentially displace protein Z in the foam structure, but the hordein-based foams are less stable. Going back to our analogy, protein Z is a better girder, but hordeins are faster and cheaper.

Foam proteins: LPT1

Lipid transfer protein 1 is expressed in the aleurone layer of the malt during germination and has a molecular weight of about 9,700. LPT1 has two roles — in its native form, it is an effective foam former, but not a foam stabilizer. In the boil, however, it slowly denatures, and denatured LPT1 is an effective stabilizer when combined with protein Z or hordein. To use our skyscraper analogy, it works like a strut to strengthen the overall structure. The native form of LPT1 also acts as a lipid binding protein, preventing the lipids from destabilizing the foam structure. You could think of this as the struts being covered with guard wires to keep the pigeons off. Overly long boils can denature most of the LPT1 in the wort and impair foam formation and lipid scavenging ability.

Other lipid binding proteins are puroindolines (PIN) from wheat and hordoindolines (HIN) from barley. These proteins are very hydrophobic and should be strong foam promoters, but experiments have shown that these proteins do not survive the brewing process and are not detected in beer or its foam. Their value as foam promoters may be realized during the wort boil, removing lipids that would later reduce the foam capability. However, certain types of lipids in the wort are essential for yeast nutrition. The key is to strike the appropriate balance between removing lipids that impair foam, and not impairing fermentation.

Hops and metal ions

Isomerized alpha acids have been known to be foam promoters for many years. Several experiments have shown that foam stability is enhanced by an increase in IBUs. The iso-alpha acids are thought to facilitate cross-linking of the malt proteins via hydrogen bonds. Their function can be likened to the gusset plates that reinforce a skyscraper’s girders and strut joints.

The molecular structure of the iso-alpha acids is an important factor. Isohumulone is a much more effective foam stabilizer than isocohumulone, and so it stands to reason that today’s high alpha hops, which were bred for low cohumulone percentage of alpha acids, have better foam stability than older hop varieties. Ironically, the light-stable hydrogenated pre-isomerized alpha acid extracts have the greatest foam stability by far. The use of these extracts in beers will yield whipped egg white type foam that will last until it is time to wash the dishes. These are often referred to as tetra and hexa hop extracts. Note that there are both hydrogenated and non-hydrogenated pre-isomerized alpha acid extracts available.

The rivets for the gusset plates and girders are divalent metal ions, such as manganese, aluminum, nickel, tin, magnesium, zinc, calcium, and barium. Metal ion hydrides form hydrogen bonds between hop acids and proteins and stabilize the molecular structure. Zinc additions of as little as 2 ppm have been shown to have significant increases in foam stability. Unfortunately, most of the metal ions in the wort are lost to the trub during wort boiling, and the only ion that doesn’t impart off-flavors in the list above is calcium. Other ions that will benefit foam with a low risk of off-flavors are zinc, aluminum and magnesium. Aluminum is the strongest foam promoter, but the hardest to get into solution. Divalent metal ions like copper and iron are known to catalyze beer staling reactions.

What are lipids?

Lipids are naturally occurring substances that include fats, waxes, oils, gums, fatty acids, glycerides, sterols and sterol esters. Most of the lipids in wort come from the malt though some come from the hops. Yeast will synthesize short fatty acids, but these have been determined to have little effect on foam stability. One of the most damaging types of lipids are the hydroperoxides formed by the malt enzyme lipoxygenase acting on linolenic acid (a common malt lipid). Lipids and their hydroperoxides are highly hydrophobic and will cause the bubble film to rupture, leading to rapid coalescence of the bubbles and collapse of the foam. These same hydroperoxides are also a primary cause of beer staling.

Most of the lipids in wort are removed during the brewing process, either in the spent grains due to vorlauf, in the hot break or utilized by the yeast. Unfortunately, any hydroperoxide forms that make it into the fermenter will not be utilized by the yeast and will need to be dealt with by LPT1 and other yet-to-be-identified lipid binding proteins.

Brewing process effects

Malts and adjuncts

Barley, wheat and rye malts and adjuncts contain lots of the necessary proteins for good foam. The other adjuncts like maize, rice and refined sugars, do not. Initial levels of protein Z, hordeins and LPT1 depend on the barley variety, but modern varieties can all be considered adequate. Barley varieties grown in wetter regions are known to have higher levels of LPT1, which biologically is a plant defense protein, but these same conditions are also known to stimulate the level of lipoxygenase, so barley selection on this basis may be self-defeating. New varieties are being selected by barley breeders with reduced lipoxygenase levels, and these may be a good option. The kilning of malt denatures protein Z and LPT1, and so high-color malts (most everything beyond Munich) don’t contribute significant levels of foam active protein. However, the melanoidin these malts contain has been demonstrated to produce stable foams.


Once in the mash, hordein can be broken down into smaller foam-promoting proteins, but protein Z and LPT1 are mostly unaffected. A multi-step mashing schedule that includes a protein rest can be used with less-modified malts to bring hordein levels to the same levels as well modified malts. However, overlong protein rests can breakdown the foam-promoting proteins and in turn produce excess FAN and basic amino acids, which are known foam destablizers.

In general, the foam of beer brewed from highly modified malts benefits from high mash temperatures. When the malt is mashed in at temperatures above 150°F (65°C), a greater proportion of foam-promoting hordeins survive into the beer.

The detrimental enzyme lipoxygenase is most active from 95–140°F (35–60°C), so that region should be avoided if you are brewing with highly modified malt — it is unnecessary and most likely detrimental. If you are brewing with a high proportion of unmalted adjuncts, or with less-modified malt, then you will need to use rests in that region, but take care not to aerate the mash while stirring because this will promote lipid hydroperoxide formation and staling.

In addition, industrial high gravity brewing practices are known to impact foam quality.


There are several foam positive and negative reactions that occur during the wort boil, and this is the step that really levels the playing field. A lot of protein and lipids are removed from the wort by thermal denaturing and coagulation in the hot break. The impact of wort boiling is such that even though proportions and trends of protein levels are carried through the boiling process, the concentration of all species is reduced by an order of magnitude, and so large differences become smaller differences. In other words, a malt that is 50% higher in LPT1 will make a wort that is only 5% higher in LPT1 compared to a normal malt.

The severity and gravity of the boil are the primary factors. Although not a concern for homebrewing systems, the high hydrostatic pressure in deep boil kettles in professional breweries can elevate the boiling temperature past 217 °F (103 °C). This greatly increases the thermal stress on the wort, and can further reduce the final LPT1 levels from 20 ppm to 2 ppm, having a large impact on foam stability. In short, only boil as long as necessary to achieve the hot break, reduce dimethyl sulfide precursor, and isomerize the hops. Do not oversparge and overboil to concentrate the wort.

As noted above, a large difference in wort protein before the boil becomes a smaller difference in wort protein after the boil. The data would seem to indicate that there is a relatively small range of post-boil protein concentrations that are apparently independent of sugar concentration. This issue is especially pertinent to extract brewers who are doing concentrated boils on the stove and diluting in the fermenter. The foam-active proteins will also be diluted, and result in poor head retention.


If we assume that healthy yeast produce a healthy fermentation and thereby produce a typically high quality foam, all else being equal, then factors that stress the yeast will also affect the foam. Yeast produce a variety of byproducts and waste products, including lipids, during fermentation, and produce even more when they are stressed. Common examples of stressed yeast byproducts are elevated levels of fusel alcohols, esters, acetaldehyde and vicinal diketones (VDKs). Yeast also excrete an enzyme called proteinase A, which particularly effects proteins around 10,000 Daltons in size, including LPT1. It has been shown that stressed yeast produce higher levels of proteinase A and lower levels of foam stability as a result. Factors that are known to contribute to increased proteinase A secretion are low FAN levels, and high levels of alcohol, dissolved carbon dioxide and hydrostatic pressure. Proteinase A will continue to be secreted by yeast after fermentation is complete, and will remain active in the beer, even if the yeast is filtered, unless the beer is pasteurized. Experiments have shown that foam active proteins will continue to be degraded in the bottle and that the enzyme is more active at room temperature than at refrigerated temperatures. So store your beer cold if you want to preserve the foam.

Finally, if the yeast are stressed to the point of autolysis, the rupture of the yeast cell will release a variety of foam degrading lipids and enzymes into the beer, greatly impacting foam stability and flavor.

Foam finale

The road to good foam seems to be full of pitfalls and barriers; it’s a wonder we can produce any foam at all! Hopefully this discussion has presented the structure of beer foam and the levers you can use for promoting good foam. Remember: Don’t use protein rests on well-modified malts. Avoid overlong boils and avoid excessive thermal loading. Malt extract brewers should avoid concentrated boils and diluting in the fermenter. (Use the “extract late” method instead.) And finally, avoid stressing the yeast during fermentation.

Issue: May-June 2008