Ok, testing you to see if you’re on your toes: when was the first Higgs particle produced by humans? We can’t know the exact date, but make your best estimate. 🙂
[Hint: how many Higgs particles will be produced at the Large Hadron Collider in 2012?]
Quick Higgs Quiz: The First One
POSTED BY Matt Strassler
ON 07/26/2012
Share via:
Twitter
Facebook
LinkedIn
Reddit
Buy The Book
Reading My Book?
Got a question? Ask it here.
Media Inquiries
For media inquiries, click here.
Related
Particle physicists describe how elementary particles behave using a set of equations called their “Standard Model.” How did they become so confident that a set
POSTED BY Matt Strassler
ON 11/20/2024
If you’re curious to know what my book is about and why it’s called “Waves in an Impossible Sea”, then watching this video is currently
POSTED BY Matt Strassler
ON 11/04/2024
82 Responses
Matt, thanks, lets do this again–with the first likely man made everything– like the first top and the first W, first man made neutrino (I will guess under the stadium in Chicago, 1942, or could have it been in Rome 1934 with Fermi’s early work??) The top may have first been made in the same bunch crossing that made the first higgs in Run 0??
So if in my ignorance I was to make a stab in the dark, say weak boson fusion at LEP, how does one go about estimating the cross-section to make a Higgs at a particular collision energy.
Hard to estimate things like that without expertise: best to use Google or INSPIRE to search for papers on “weak boson fusion LEP Higgs”.
Because Prof. Strassler said there’s a gap in the history listed in the comments, can I add the semi-random and amateurish guess of a Higg-like being a singular chance event happening in a test fusion reactor (Tokamak)?
Looking up the energies, this seems improbable if not impossible, but as we’re going from 300,000 BCE to 2012, I thought I might as well throw that in and see if I win the T-shirt!
🙂 nope. Fusion reactions, like nuclear explosions, occur at energies per particle that are a tens of thousands of times too small. The sun does not make Higgs bosons.
” The sun does not make Higgs bosons.”
This immediately makes the Curious George in me wonder: which if any stellar events would produce Higgses? Supernovae being the obvious answer that first comes to mind, but apart from that?
How about the CDF run 0 at Fermilab, 1988-89 (before the first Tevatron upgrade): http://prd.aps.org/abstract/PRD/v43/i7/p2070_1
The run had an integrated luminosity of 4.4/pb at 1.8 TeV. Cross-section for ppbar Higgs production something like 0.85pb, so perhaps 4 Higgs particles produced in that period.
So the first one is most likely to have been produced in 1988?
Ohhh, I like that answer.
Ding Ding Ding Ding Ding Ding Ding — you just barely got it in time — in fact I wrote my post at 6 this morning before your reply came in. But your reply came in before 8:45 when the post went out! Congrats.
Thanks 🙂 That was a very informative process. It’s a great question for exploring the history of these experiments – I’d never heard of CDF run 0 before today.
My random guess December 1995 (LEP).
Well it’s obviously not the LHC. Tevatron did see little hints of the Higgs, remember, so there must have been zillions made there underneath the background. It has to be Tevatron or earlier. That was already suggested when the prof said nobody had suggested the one that he thought was it. SPS and HERA are the only other colliders with sufficient energy listed on Wikipedia. The prof shot down SPS himself, and remained curiously silent about HERA other than offering a leading question. HERA is the only suggestion offered between his comment about one thing not being suggested yet and his comment about all the ideas being out there. So Maybe I’m reading into things, but from what I can tell, all arrows point to HERA.
ok all the ideas are out there, what is it?? was it in a linear machine or in a ring?
I’m getting the idea that it has to be at either the LHC or Tevatron, as evidently nothing else has enough per-particle energy. Are there any machines of any energy that haven’t been mentioned? The only other thing I can think of is spacecraft which encountered cosmic rays outside of the atmosphere, which might allow for enough energy, but probably not enough collisions to have a chance of producing a Higgs. Gosh I feel dumb now 🙂
You’re right, all the ideas are out there. There’s just a gap in the history.
What about when quarks were discovered? Wasn’t that similar to Bohr’s discernment of the atomic nucleus, where lumpy particles collided and diverse recoil angles were observed? The first time protons (or mesons or pions?) were smashed, was there enough GeV? Small enough barns?
I would guess the answer you’re thinking of is early April 2010. Since the graph you helpfully link to in your hint shows that the cross section is order 1 (for some processes at least). I don’t know the beam flux and I am a bit confused why the conversion factor would be 1000 up to June 2011 but 5000 for Dec 2011 (I don’t see how that scaling works)… but seem like Higgses would have been produced within the first days of operating at 7TeV.
P.S. on the RHIC, if gluon fusion is the dominant production channel for Higgs, doesn’t heavy ion collision produce lots of gluon fusion? That would give a date of 2000 or a little later.
The issue is not “how often does gluon fusion occur?” but “how often does gluon fusion with a gluon-gluon center-of-mass energy of 125 GeV occur?”. The answer to the latter question, at RHIC, is “not often.” RHIC didn’t observe W bosons (much easier to make) until very recently.
More precisely, the flux of cosmic rays of 10000 GeV, needed to produce a c.o.m. energy of 125 GeV, is one per square meter per week. Given that a human is about a square meter, and there are many millions of them, and millions of weeks, there will have been a huge number of Higgs bosons produced, and they would never have existed without humans.
Marc, I don’t think most of those comic rays reach the ground; they hit something high in the air and make a cosmic-ray shower. The losses due to the atmosphere from the cosmic ray flux are exponentially large. If this weren’t true, most of them would pass right through a human body without hitting anything.
Moreover, 10000 GeV isn’t enough; that’s enough for a 150 GeV proton-proton collision, yes, but you know the probability of making a Higgs boson that way is exponentially small. You need something closer to Tevatron energies, and that means you need 1,000,000 GeV or more, and you pay a big price for that extra energy in terms of the numbers.
So I think the probability of a human being intersecting with such a cosmic ray is very small indeed.
And moreover, if the humans weren’t there, the Higgs bosons would have been made in the ground, so it isn’t true they would not have existed without humans; the numbers would have been the same.
So you’ve had your fun 😉 but I don’t think your argument stands. I don’t think a single Higgs boson has been made by a cosmic ray hitting a human.
Just for the sake of mischief, how about a cosmic ray hitting one of the Apollo astronauts in the late 1960s? 😉
A cosmic ray of energy 3×10^20 eV has been observed (according to Wikipedia, which speculates that 750 TeV of that would be available for conversion if it collided with another particle)
300,000 BC. Cosmic rays of high energy strike humans all the time. Some of the collisions will be energetic enough that a Higgs will be produced. The cosmic-ray human interaction cross section is probably high enough, given the time scale, cross sectional area of the human population, and the known cosmic ray flux.
What about RHIC? It’s newer than the Tevatron but there’s sure a lot of energy in those collisions.
There’s a lot of energy, but not as much energy per proton — only an average of 200 GeV per proton-proton collision inside the nucleus-nucleus collision. That’s not nearly enough to find gluons energetic enough to make a Higgs regularly.
Well, with that teaser, I’ll go for the HERA electron->proton collider at DESY in Hamburg, which started operation in 1992. It operated at up to 318GeV for 15 years, so I guessing it had plenty of opportunity to produce a Higgs – oops sorry, a Higgs-like particle. I’ll go for 1993.
The dominant way of making a Higgs at a hadron collider is in gluon + gluon –> Higgs. The dominant way of making a Higgs at a lepton collider is in electron + positron –> Z + Higgs. The dominant way of making a Higgs at an electron proton machine is… ?
Well, it’s got to be vector boson fusion. The cross-section for that at the LHC is a few pb, and at the Tevatron a few fb. HERA only took a few hundred pb^-1 of data in the 1990s, so there’s probably not much chance of it producing one.
I don’t know because I’m not a physicist, but this link seems to say it’s WW or ZZ fusion.
http://www.desy.de/~heraws96/proceedings/electroweak/Kniehl2.ps.gz
That’s the right paper to look at, yes… good work.
Ok, up to now we’ve had lots of very good suggestions, and at least one of them might be right, but there’s at least one suggestion that hasn’t been made that I think is probably the right answer.
My naive guess:
The Tevatron was probably producing Higgs particles occasionally by the mid 90’s, but since it was not tuned to detect them, does that really count?
Now I’ll read the preceding comments to see what less naive readers think…
Yes, it counts. You can work backwards, from how many you’ve detected, and from your correct predictions which tell you your equations are more or less right, to figure out how many you must have made that you didn’t detect. That’s the power of science, it allows you to work this kind of thing out and be confident in the answer.
I agree with others. So far only a Higgs like particle has been produced. The kind of Higgs or if it is a Higgs at all has yet to be determined at 5 sigma. This will require measuring further properties like spin, not just decays and decay fractions.
No one is disagreeing with you. However, I’ve pointed out that there are already two or three pieces of information in favor of the Higgs-like particle being a Higgs of some type, and while of course that is not a proof, it doesn’t prevent taking it as the most reasonable working assumption for the moment. There is no alternative reasonable proposal, as far as I can tell as a theorist; every other alternative requires at least two accidents to happen, whereas no accidents are required if it is a Higgs of some type.
Which accidents?
Suppose X is not in any way like a Higgs particle (say it has spin 2, or it is pure CP odd). Then it is an accident that
(1) the cross section for proton+proton –> X –> photon + photon comes out within a factor of 2 of what is expected for a Standard Model Higgs
(2) the cross section for proton+proton –> X –> four leptons from real and virtual Z particles comes out within a factor of 2 of what is expected for a Standard Model Higgs
(3) the new particle X does not screw up precision tests of the Standard Model using measurements from the LEP and Tevatron colliders
Moreover, if you believe the relevant Tevatron and LHC measurements, you can add
(4) it is an accident that the process proton+antiproton –> W + X, Z + X where X –> bottom quark + bottom anti-quark is within a factor of 2 of the SM Higgs
(5) it is an accident that the process proton + proton –> X –> lepton+antilepton+undetected particles (from real and virtual W’s) comes out to within a factor of 2 of the SM Higgs
(6) it is an accident that there appears to be something resembling VBF production with roughly the rate expected for an SM Higgs.
A few of these things are related, so if one accident occurs another is automatic; so the number of accidents isn’t quite as large as listed. But the number of accidents is definitely larger than 1.
Now if X is a Higgs particle of some type, none of these things are accidents. It is very, very common in theories that go beyond the Standard Model that one of the Higgs particles will be roughly similar to the SM Higgs.
Before I finish reading the other answers, and after trying to peek at the textbook, here’s my answer, but I expect only partial credit if I’m right.
The top quark is 172GeV, and it was created in the Tevatron, was it not? If they were making particles up to 172GeV, they probably made the lighter Higgs particles, just not enough to hope to see them above the background.
I should go look up the quarks and where they were discovered, to find out where the colliders were that went above 127GeV, so I can give a precise answer. Still, this is a pre-class quiz, right? So we can get the answer and get on with class? (Just had to clown around, that’s part of quizzes, too.)
I’ll go with 25000 BC +/- 27012, simply because we haven’t experimentally disproven that a person could create non-standard-model Higgs simply by bashing two rocks together. Anything else would be speculation.
If everything we know about nature had to be determined by literally do precisely the experimental test, we’d never have learned anything. We didn’t have to do the experiment of running the LHC to know how to design it in advance; that’s because we understand physics. Similarly, I don’t have to experimentally prove that I can’t produce a 125 GeV particle by bashing rocks together. We know enough about energy conservation, quantum mechanics and atomic physics (and rocks) to know that if all the humans that had ever lived had spent their entire lives, every hour of every day, dropping rocks off of cliffs, no 125 GeV particles would ever have been produced.
I never said anything about 125 GeV, but I accept that there is probably a good reason that nobody is looking for it in whatever range mashing rocks together would produce. What would you estimate that range to be, anyway?
Very rough estimate: if the rocks have a mass of 1 kg each, and we smash them together at 10 m/s each, then there is 100 joules of energy in the collision. A nucleon weighs roughly 1 gram per mole, so 2000 grams of rock is roughly 10^27 nucleons. So that’s 10^-25 joules per nucleon, or less than a millionth of an eV, or a millionth-billionth of a GeV. 🙁
At these energies, the atoms just bounce off one another; the nuclei never get anywhere near each other.
The “produced by humans” phrase is giving me an unscratchable itch!
I would say that we humans have produced none. We -well, they- have created an environment where the particle could most likely be observed having into account the known elements (you said previously that if it was not observed then it only meant there are other unknown forces/elements yet to be discovered). Then the first time we humans confirmed the observation of a Higg’s particle was at the beginning of this month 🙂
I like the way Pam thinks. I find the moment of first production dissipates as we try to observe it. But the decay products are fascinating!
No, this is all too wishy washy. Physics is a very powerful tool; you’re underestimating it. We do know what’s going on pretty well (after all, we knew how to look for this thing, and about how often it would be produced and how it would decay if it were more or less like a Standard Model Higgs, long before we found it.) Granted there are things we know and things we surmise, and I am assuming this Higgs-like particle is indeed a type of Higgs particle. But I’ve argued that’s far more likely than that it’s something that’s there to fool us into thinking it’s a Higgs of some type.
There are very confusing statements in your comment.
First, yes, I said that if the Higgs particle was not observed (over many years, because you’d have to convince yourself first that it was unobserved because it didn’t exist, rather than because you didn’t look for it in the right way) then it meant that there are other forces or particles. But it HAS apparently been observed, so this is irrelevant now.
Second, do you not agree that humans are responsible for building the particle accelerators? And that without the particle accelerators, no Higgs particles (or W particles, Z particles or top quarks) would have been created? If you’re making a semantic comment (well, nature made the Higgs particles, not humans, who just arranged for nature to make them) then — that’s semantics, and not worth further discussion.
Finally, once you have confirmed something exists in nature, you can work backwards in time; that’s part of how science works. Once you discover that AIDS is caused by a virus, you can assume that all the people who previously were afflicted by AIDS had that virus, even though you didn’t explicitly test their blood to check. Once you discover that supernova 1987a emitted neutrinos, you can infer that all previous supernovas of similar type emitted neutrinos also, and you can go looking for signs of those neutrinos. And now that we know there is a Higgs-like particle at 125 GeV, we can use our knowledge of how it is produced to figure out where it was produced in the past.
My appreciation of Pam’s comment was more about how a question like “first Higgs particle” can be hard to pin down precisely. We start to discover more about the question as we try to answer it.
Also, we can’t produce a Higgs particle on demand, the way we could produce a proton on demand. We can say: “To get a proton, do this, and your proton will appear here.” We can’t say “To get a Higgs boson, do this, and pick up your Higgs particle.” We have to deduce its existence indirectly, so it’s kind of hard to say that we “made it.” I think I understand Pam’s nuance, that we made inverse barns that had Higgs particles in them. Like we can’t make a quark, but we can make mesons that have quarks in them.
So have I thoroughly mixed apples and basketballs now?
Do you bake a cake, or do you create the environment for it to exist?
December 1992. X-section at Run I would have been about 0.5 pb; December 1992 is when about 2 pb-1 integrated at Tevatron (per D0 run I luminosity plot)
Lep II was too late
SPS cross section is closer to fb, they never got close to an inverse femtobarn on tape.
2011 CERN
Definitely not. Even CERN had enough collisions in 2010 to make hundreds of them. And they weren’t the first.
Tevatron Run I would have required a few /fb for each Higgs and only reached 150/pb so assume it did not make one. LEP would have needed to reach about 215 GeV to produce a Z plus a Higgs but it stopped at 209 GeV. Tevatron run II produced one Higgs for every 1/pb. It got there around beginning of July 2001, so that is the answer 🙂
You say Run 1 would have required a few /fb of collisions to make a Higgs, while Run II would have required 1/pb to make a Higgs.
Are you sure?
See Figure 1, http://lss.fnal.gov/archive/2004/conf/fermilab-conf-04-027-e.pdf
If Tevatron run I production cross-section was more like 0.1 pb as these plots suggest then it would have started producing them around beginning of 1993 with about 10/pb.
http://www-d0.fnal.gov/runcoor/RUN/images/d0_run1_lum.png
Other cross-section plots I looked at did not exptrapolate well.
Well, a plot isn’t really needed. Run II was at 1.96 TeV; Run I was at 1.8. With only a 10% change in energy, you’re not going to see more than a factor of 2 change (typically more like 1.5) in the production rate for something whose mass isn’t at the edge of the accelerator’s reach, including a particle produced resonantly at 125 GeV. That’s clear from the parton distribution function for the gluons, which has a roughly power-law behavior for energies of order 1/10 or less than the proton’s energy. You only get exponential fall-off at significantly higher energy fractions. [Same is true for the other pdfs, it’s just a question of where the fall-off sets in.]
And indeed, if you look closely, the cross-section for Run I at 125 GeV is about 0.6 per pb, not 0.1. And that’s actually important in the history.
I understand a GeV, but the inverse barns still elude my visual imagination.
I see where the discussion is going: we need not only enough GeV, but also enough inverse barns to reach 50% probability of a single higgs.
Would Professor Strassler, or anyone, like to help me visualize an inverse barn, and why it takes part of one to make a Higgs particle? (Not necessarily now, but sometime when it fits into the discussion.)
Barns are kind of a useless unit unless you are using them both for the luminosity of your beam-beam collisions (i.e., how much data are you collecting) and the cross-section for a process (i.e., how hard is it for this process to occur.) The product of the two is what gives you things that are easy to think about, such as the number of times a certain process occurs per second, or the number of times it has occurred throughout the LHC run.
Physically: the cross-sectional area of a proton (as viewed before a 10 TeV collision) is about .1 barn.
A beam of protons is characterized by how many protons are in the beam, how fast the beam is moving (speed of light in this case) and how tightly the protons are packed together (and whether they come in bunches, too). These things are complicated to keep track of, especially when we have two beams, so instead physicists use a different way of thinking, one that doesn’t require knowing the details of the beams.
The idea is this. We might say: our two beams correspond to 100 per barn per second — we say this is their “luminosity” — then since the effective area (or cross-section) of a proton is 0.1 barns, that means we will be making 10 proton-proton collisions happening per second. Or if we run our beams for a minute, we’ll have had a total of 60 seconds * 100 per barn per second = 6000 barns — we say this is the “integrated luminosity” — and that means (with proton-proton collisions having a cross-section of 0.1 barns) 600 proton-proton collisions total.
So both ATLAS and CMS now each have about 10 inverse femtobarns of data — that means 10 per femtobarn, or 10 * 10^15 per barn — 10^16 per barn of integrated luminosity. That corresponds to about 10^15 proton proton collisions since a proton-proton collision has a cross-section of 0.1 barns. And it corresponds to nearly 200,000 Higgs particles, since the cross-section of a Higgs particle of 125 GeV is on the order of 20 picobarns = 20,000 femtobarns, and the total amount of data at CMS and ATLAS is 10 per femtobarn.
It’s easy once you get the hang of it. Just remember that the purpose of barns is to allow you to talk about what the collider is doing without knowing the details of the beams, and that you use barns mainly to combine a quantity of data with the likelihood of a process to get a total number of events of a particular type.
I’m not sure if nuclear weapons have enough energy _density_. They should have enough overall energy, but that might not help here…
I would guess 1995, the first time LEP went over 125 GeV c.o.m. energy in electron-positron collisions.
Indeed, nuclear weapons, despite the huge energy release, do not come close to putting that energy into a small enough space to make a heavy particle like a W, Z, top or Higgs. Binding energies in nuclei are small fractions of a GeV, so that’s the most you can get out per nucleon; and even if you somehow converted an entire nucleus’s mass to energy (which you can’t) you’d be getting densities which are GeV per nuclear volume, which is many (8 or so) orders of magnitude too small.
You need a particle accelerator, and you need the center-of-mass energy of the collisions created using that accelerator to be greater than the mass-energy of the particle you want to produce, in this case 125 GeV.
So far we have mid-40s, mid-60s, mid-70s, early 80s, and 90s. Any other votes?
In 1983 at Tevatron or in 1981 at SPS 🙂
do you want to defend this argument? After all, SPS had enough energy in principle, yes, but you also have to have enough collisions.
Probably Higgs particle were produced at SPS. But we know nothing about them.
Then Higgs particle were produced at Tevatron for sure because of excess in 120 GeV region and now we know of existance of Higgs particle confirmed by ATLAS and CMS. Then CDF starts their program of Higgs search at 02/07 and D0 at 04/07.
1964, by Peter Higgs, adding it to his second paper after the first submission of the same was rejected. I believe Nambu gave the recommendation to do so.. 🙂
I’m asking about the particle, not the idea of the particle. If you’re trying to address the history questions, let’s do that elsewhere…
There may have been the odd one or two produced in the Fermilab’s 500GeV experiments of May 1976; or in SPS later the same year.
Hmm. Are you talking about “fixed-target” experiments here [where one slams a beam of protons into a stationary object]? Or is there a program here I don’t know about. In any case, you’d need an enormous number of proton-proton collisions at center-of-mass energy of 500 GeV to make a single Higgs particle. They didn’t make that many back then.
You’re right, I take that back – they were stationary target experiments.
SPS did proton-antiproton collisions at 270+270 GeV from 1981. They accumulated 28/nb of collisions in 1982 (then roughly 150, 500 and 750/nb in the next three years). You couldn’t call those “enormous numbers”, so presumably the probability of there being a Higgs among them is pretty tiny. Still…
Almost correct; the energy was 320+320, I believe. So let’s call the integrated luminosity 1 per picobarn. What is the cross-section to create a 125 GeV Higgs particle in proton-proton collisions at a center of mass energy of 640 GeV?
I think there was an increase from 270 to 320 GeV in 1984. The cross-section for Higgs at 640 GeV would be of the order 10fb? That would mean a 1% chance of Higgs production at SPS by 1985. (Am I in the right order of magnitude there?)
I think so. Even with multiple experiments, the chance that SPS made a Higgs particle is certainly very small.
I have to wonder, what probability are we going to accept as “created the first Higgs particle?”
I think the idea is to look at the distribution of probabilities and come up with a best guess as to when the first one was produced. (Working on the assumption that a Higgs is either created or it isn’t, rather than worrying about superpositions of states that haven’t been definitively identified or anything like that. That could be argued too, I guess.)
Think median, if you like. Then you’re looking for the time by which it’s 50% likely to have been produced in the colliders operating up to that point. From what’s been said below, it looks likely to have been 1992.
Certainly higher than 50%; but as it turns out, there are some accelerators for which the probability is a few percent, some for which it’s a tiny tiny fraction of a percent, some for which it’s zero, and some for which it’s basically 100%. And there’s only a couple of cases which are transitional.
No confirmed Higgs particle has been produced to date it remains hypothetical pending further data. The July 4th announcement from CERN noted a Higgs-like particle.
Fine — replace “Higgs” with “Higgs-like”. We still know when this particle was first produced. [I gave you my strong reasons for thinking it is a Higgs of some type: http://profmattstrassler.com/articles-and-posts/the-higgs-particle/the-discovery-of-the-higgs/higgs-discovery-is-it-a-higgs/ . After 3 weeks of thinking I stand by this argument.]
Touche, Andrew!
I assume you don’t mean virtual Higgses. What about the off-shell process e+e- -> Z* H at LEP? The final LEP energy was 209 GeV, leaving 125 GeV for the Higgs and 84 GeV for the virtual Z. So probably some on-shell Higgses were produced in the final days of LEP, i.e. by 2000.
But surely there were some produced at Fermilab before that, buried in the background.
So I’ll say “mid 90’s”
The first nuclear device ever detonated was an implosion-type bomb at the Trinity test, conducted at New Mexico’s Alamogordo Bombing and Gunnery Range on 16 July 1945.
A nuclear device isn’t NEARLY enough to make a 125 GeV particle. The energy per particle (which is what counts) is a thousand times too small.
Ok, how about this one?
On May 10, 1752 Thomas-François Dalibard of France conducted Franklin’s experiment using a 40-foot (12 m)-tall iron rod instead of a kite, and he extracted electrical sparks from a cloud.
NASA has already proven that electrical storms are one of the best sources for antiparticles. Is there enough energy to release the Higgs boson as well?
On a different subject, will you have a moment to read my reply on the “Mass and Energy” post?
Thank you Professor. Sometimes I jest only for the good and i certain congratulate you for giving of you valuable time to educate the curious minds.