Skip to main content

What Is the Proton Sea?

Leonard Kelley holds a bachelor's in physics with a minor in mathematics. He loves the academic world and strives to constantly explore it.

Sometimes it is dangerous to refer to something as fundamental, especially in the realm of science. Upon the discovery of atoms and their parts – electrons, protons, and neutrons – the fundamental building blocks of the Universe were supposedly uncovered.

But then we uncovered antimatter and other types of boson, fermions, and mesons, and in the process found that these too have components to them in the form of quarks and gluons. Depending on the mix of quarks you have, a different particle will be present. Delving into the contents of a proton has yielded a major surprise.

But, I am speeding through this, so let’s pump the brakes and go back to the origins of the quark/gluon picture as discovered by man.

What’s That?

In 1964, Murray Gell-Mann and Georg Zweig released their quark model of subatomic particles. It showed how the supposedly fundamental particles were really sets of two or three quarks, and these could be from one of 6 total types and all of which have a fractional charge value to them.

The model first received evidence in 1970 when the SLAC accelerator sent electrons crashing into protons, causing pieces to fly away and confirm the composite nature of these particles. But as further studies of the contents of a proton were done, further surprises were revealed (Wolchover).

Those three quarks, 2 of which are up (+2/3 charge) and 1 of which is down (-1/3 charge) that make up a proton do not make up the mass of a proton. In fact, it’s not even close to being a significant portion of it. But what else was there? Higher energy electron collisions (with greater speed and therefore smaller wavelengths, increasing collision rates) with protons showed particle scattering indicating more than the three quarks were present.

As it turns out, the quark model was just an approximate solution of quantum chromodynamics, or QCD, which was fully fleshed out in 1973. It introduced the idea of gluons which…well…glue the quarks together (Ibid).

These gluons actually feel the strong nuclear force as well as carry it (which is not like a photon, which merely carries the EM force). This self-interaction creates a storm inside the proton, which gives “gluons free reign to arise, proliferate and split into short-lived quark-antiquark pairs.”

These pairs usually go unnoticed because they exist for such a short span of time, with three unbalanced quarks giving the total charge count we see when measuring the proton (Ibid).

This constantly changing entity is known as the proton sea, and somehow it implies that despite this inner turmoil, an average of 2 up and 1 down quarks happen! It is because of this complex interplay between the merging and disappearing pairs that the QCD equations are currently unsolved, which means meaningful predictions are difficult to make (Ibid).


This is why further tests to gain insights were important, but they themselves revealed even more surprises. In 1991, the New Muon Collaboration in Geneva sent muons, a heavier electron, into proton and deuterium targets, the latter being a proton-neutron nucleus. Surprisingly, it showed that there were more down antiquarks than up antiquarks in the proton sea. Why was this inherent imbalance happening? Shouldn’t there be a fair split of antiquarks, not some preference? (Ibid)

Scientists developed models in an attempt to account for this. The pion model makes use of the pion particle, which is readily exchanged between protons and neutrons as they exist in close quarters i.e. a nucleus. Normally, a pion and a neutron merge to become a proton but if a proton could send to itself then we could measure the fluctuation in the expected particle count. In fact, a proton could instead go the other way, decaying into a pion and a neutron (Ibid).

This would establish a quark count of 2 down and 1 up for the neutron and an up quark and down antiquark for a pion. This would still total to 2 up and 1 down as we expect for a proton but still establish the discrepancy we see in the proton sea. Another possible solution is the statistical model, wherein the particles inside the proton act much like gas in a room, with their velocity dictated by the spin value they possess. When the experiment is fine-tuned via the results, the model indicates a down antiquark excess (Ibid).

Now, these models clearly do not convey the same ideas and so have different predictions, making testing them much easier. One of these tests is hunting for the antiquark momentum fraction. Recall that the mass of a proton is correlated to the energy of those particles popping in and out. The models indicate differences for the ratio of down to up antiquarks should naturally occur, depending on the energy of the antiquarks present. Each model points to different relations of the energy to the ratio, allowing for definite testing (Ibid).


And the Results Are…

One of the first tests done to gauge this ratio was the NuSea experiment at Fermilab. First brought online in 1999, it looked at the down-to-up ratio based on the antiquark momentum. NuSea found that down antiquarks were more numerous than up antiquarks in the proton sea, but if the up antiquarks had 30% or more momentum than the down antiquarks, then the up were more numerous.

This was strange because, as Johanna Miller puts it, “gluons don’t couple to quark flavor, so they should produce up-antiup and down-antidown pairs with equal probability.” Many QCD theories also predict a flavor-symmetric sea as well, but no attempts at explaining the matter invariance have been successful at explaining the weird behavior at the high fractional momentum ratios (Wolchover, Miller).

Therefore, SeaQuest was developed as a follow-up to see if the noisy results from NuSea were in fact real. In this, protons were fired at hydrogen or deuterium. Occasionally, via the Drell-Yan process, one of the valence quarks (one of the up or down quarks not in the sea) annihilates with one of the antiquarks in the hydrogen or deuterium’s protons.

This creates a virtual photon which then becomes s a muon and an antimuon along with a bunch of other byproducts that make this interaction easy to spot. SeaQuest would study this potential interaction by using iron to isolate the muon, measure its speed, and work backwards to determine the type of interaction that created it along with the momentum fraction the antiquarks had (Wolchover, Miller).

With this in hand, one compares the hydrogen and deuterium results, because the former is just a proton/electron pair while the latter includes a neutron. That difference in atomic makeup allows for a comparison of the proton and neutron, which mirror each other in terms of quark composition. So by comparing the ratio of up-to-down antiquarks between them we can get a feel for the true momentum fraction (Ibid).

After 20 years of running this, the results pointed to an average 5:7 ratio of up-to-down antiquarks, or that for every up antiquark there are about 1.4 down antiquarks. This matches well with the pion cloud theory…but it also doesn’t contradict potential results from the statistical model, either. This could be because of the mathematical nature of the analysis and description of the results, so follow-ups will still be needed, sadly (Wolchover, Miller)

Works Cited

Miller, Johanna L. “The puzzling asymmetry of the proton sea.” DOI:10.1063/PT.6.1.20210412a.

Wolchover, Natalie. “Decades-Long Quest Reveals Details of the Proton’s Inner Antimatter.” Quanta, 24 Feb. 2021. Web. 20 Jul. 2021.

This content is accurate and true to the best of the author’s knowledge and is not meant to substitute for formal and individualized advice from a qualified professional.

© 2022 Leonard Kelley