For many years now discussion in the HEP community of what might be the appropriate next machine to try and finance and build after the LHC has centered around the idea of a linear electron-positron collider. The logic has been that an electron-positron machine would provide a much better environment that the LHC for detailed studies of physics at the TeV scale. At these energies, synchrotron radiation losses when accelerating electrons are so high in a circular geometry that such a machine would have to be a linear collider to keep the power needed something plausible. The two main proposals under study have been the ILC (250 GeV + 250 GeV, later upgradeable to 500 GeV + 500 GeV) and, a less mature technology, CLIC (1.5 TeV + 1.5 TeV). These would be very expensive machines to build and operate ($10 billion and up?), requiring completely new technology, tunnels and detectors.

The operating assumption has been that initial results from the LHC would show the existence of new supersymmetric or other particle states in the region of multiple hundreds of GeV to small numbers of TeV, and the linear collider designs could then be chosen and optimized to study this new physics. The other main goal of such a collider would be detailed study of the Higgs, and knowing the mass of the Higgs is also highly relevant to what kind of linear collider to build.

The initial LHC results are now in (125 GeV Higgs, nothing new at the TeV scale) and they are rather discouraging for the linear collider idea, providing no strong motivation for studying electron-positron collisions around 1 TeV. A “Higgs Factory” capable of producing and studying large numbers of Higgs particles is an attractive idea, but the production cross-section for a 125 GeV Higgs is dominated by the process e+ + e- -> Z + H, which starts to get large around 220 GeV and reaches a maximum value around 255 GeV. So, for most Higgs studies, the right energy for an electron-positron machine is around 250 GeV, not 1 TeV.

This realization is driving a new proposal that is getting a lot of attention: the idea of going back to circular electron-positron colliders, building a new machine in the LHC tunnel, optimized as a Higgs factory, and designed to operate at 120 GeV + 120 GeV. This is being called “LEP3″, since it would be in many ways similar to LEP2, the predecessor machine to the LHC, which operated in the same tunnel, reaching an energy of 209 GeV. There would be huge cost advantages to building such a machine over the ILC or CLIC, since it can use the LHC tunnel, infrastructure, and, crucially, the CMS and ATLAS detectors (the detectors are a large part of the cost of a new accelerator).

Space was left in the LHC tunnel to allow another ring, so there are various possibilities for having a LEP3 and the LHC cohabitate. Until now, the assumption has been that the LHC would be upgraded to the”HL-LHC”, operating at higher luminosity throughout the 2020s, then perhaps an “HE-LHC”, operating at higher energies during the 2030s. This plan is being challenged by the LEP3 proposal, with the argument that it might turn out that Higgs physics is where the only action will be, and a long period of LHC operation at higher luminosities and modestly higher energies might be less worthwhile than building a LEP3 Higgs factory.

There’s a very good article about this at PhysicsWorld. For more detail about LEP3, see here, here and here. John Ellis is one of the co-authors of the latest document proposing study of the LEP3 possibility, and the Physics World article has him arguing that, after waiting to see if LHC14 turns up anything:

LEP3 could be a more secure option than the ILC if only a Higgs is discovered…If the LHC does not discover anything beyond the Higgs, then would you keep running it for years?

Lyn Evans, who led construction of the LHC and is now director of the linear-collider project argues against the LEP3 concept:

The first job is to fully exploit the LHC and all its upgrades, This is at least a 20 year programme of work, so I think that it is very unlikely that the LHC will be ripped out and replaced by a very modest machine with little scope apart from studying the Higgs.

The problem is, what if, as seems increasingly likely, “studying the Higgs” is the only new physics accessible in these energy ranges? Dreams of superpartners and extra dimensions may die hard.

This entry was posted in Experimental HEP News. Bookmark the permalink.

30 Responses to LEP3

  1. Bernhard says:

    This is interesting, but I think Ellis is dreaming. No way the LHC will be shut down until is completely explored, the pressure to keep the machine running and achieve as high integrated luminosity (including upgrades) as possible is enormous. Building a Higgs playground would be nice but at the cost of the LHC, no way.

    Furthermore I hear the Japanese are strongly pushing for the ILC to be built (in Japan) and they are willing to pay for a considerable cost of it. In any case, before the LHC running a considerable time at design energy an luminosity these are speculations. And I believe very few people would be excited about working in a “Higgs machine” and ultimately this is a key aspect.

    The ILC running at 1 TeV can in any case still easily be defended as a discovery machine as e.g. virtual corrections from all sorts of models make the discovery reach for large range of particles (models) to be much higher than the machine center-of-mass and or even the LHC for that matter.

  2. HE-LHC vs HL-LHC says:

    If all they find is the Higgs boson up to 14 TEV by 2020, no SUSY or DM or other particles, or extra dimensions what would be the point of spending money to build a HL-LHC upgrade around 2020? What would higher luminosity offer? Couldn’t the LHC continue to run at a lower luminosity but take more time to gather more data?

  3. Peter Woit says:

    HE v. HL,

    The idea of the HL-LHC is to increase the luminosity by something like an order of magnitude, and run for a decade. If you stay at current design luminosity, you’d have to run for a century to match that.

    One motivation for the high luminosity is to look for new physics via rare events. But, even just for the Higgs, higher luminosity means much better measurements. The crucial question is just how much better a LEP3 could do than this.

  4. DEQ says:

    Why the idea of a muon collider not gaining traction? Is that technology just too far out right now, i.e. something 20 years away?

  5. Peter Woit says:


    Yes, muon collider technology is still a long ways off (although you’d need someone more expert than me to give a time estimate), requiring some very new technologies. LEP3 however would use what appear to be relatively straight-forward extensions of proven technologies.

  6. Tmark48 says:

    My opinion is that the future of HEP is going to be in astronomy/astrophysics.
    Space is the only environment where we can witness (with suitable detectors/telescopes) physics beyond the SM. Investing is astronomy is going to be much more productive than building a YAA (yet another accelerator) for untold tens of billions of €/$/£/Yens when we don’t even know at what energy scale beyond SM physics kicks in.

    As for the LHC, for pete’s sake it has been in operation for what less than 4 years and we want to build another one ? LHC is going to be useful for a good decade if not 20 full years. It is going to be useful even if it throws in the dust bin all the speculative ideas that are found to have no experimental evidence.

  7. Peter Woit says:


    The LHC was designed about 20 years ago, with approval of funding and work on starting to build magnets beginning around 1993. So, if you believe it has a lifetime of 20 years, now is about the time for starting work on whatever will be next. These projects have a very long lead time.

    So far, instruments in space (which aren’t cheap either…) have yet to tell us what’s beyond the SM. I wouldn’t be so sure this will change in the forseeable future.

    The interesting thing about the LEP3 idea for a new machine is that it’s not necessarily that expensive. Instead of order of magnitude $ten billion in new money, it might be doable for a lot less, fitting within the CERN budget with no significant increase. It definitely seems worthwhile to investigate the idea and produce some realistic cost estimates, as well as a serious investigation of the physics case for the machine.

  8. Mitchell Porter says:

    “Dreams of superpartners and extra dimensions may die hard.”

    In my opinion, now that there’s a Higgs, the next big issue is the hierarchy problem. The economical way to test known forms of unification, would be to test their solutions to the hierarchy problem, e.g. the existence of new “partner” particles for the top. That way you test at the same time, both standard versions of supersymmetry (with light stops) and nonstandard models with top-partners. And it seems that this is what the experimental collaborations are doing…

  9. Tmark48 says:

    The LHC was designed about 20 years ago, with approval of funding and work on starting to build magnets beginning around 1993. So, if you believe it has a lifetime of 20 years, now is about the time for starting work on whatever will be next. These projects have a very long lead time.

    You are correct, but big projects take decades to come to completion. Even the Hubble was like this. But the thing is, the LHC was designed for one purpose, finding the HIGGS.
    Even now that the LHC has found the HIGGS, we still don’t know what kind of HIGGS particle it is. Further experiments are warranted.
    You say we need to start considering a new accelerator. Ok, to search what ? What is the goal ? It cannot be a moving goalpost as supersymmetry is. Isn’t it better to upgrade the LHC and continue using it instead a starting a wild goose chase ?

    So far, instruments in space (which aren’t cheap either…) have yet to tell us what’s beyond the SM. I wouldn’t be so sure this will change in the forseeable future.

    Astronomical instruments are way, way, way less costly than building particle accelerators and they bring more science to the table. As technology improves, detectors improve and our ability to see and interpret high energy astrophysical phenomena gets better and better. No such thing is possibile for particle accelerators.
    We simply cannot increase energy levels to tens of thousands of TeV. What if beyond SM physics kicks in at that scale ? High Energy Physicists are screwed.
    We need to start considering that maybe beyond SM physics is simply unattainable here on earth.
    Furthermore all interesting fundamental physics is coming from astronomy.
    Dark matter ? check.
    Dark energy ? Check. Particle accelerators are “in my opinion” becoming obsolete mainly because technology doesn’t give us the tools to increase energy levels arbitrarily at which to probe the microscopic scale.

    The interesting thing about the LEP3 idea for a new machine is that it’s not necessarily that expensive. Instead of order of magnitude $ten billion in new money, it might be doable for a lot less, fitting within the CERN budget with no significant increase. It definitely seems worthwhile to investigate the idea and produce some realistic cost estimates, as well as a serious investigation of the physics case for the machine.

    At the pace at which technology improves, astronomical instruments can become competitive with particle accelerators in the short term. Nature has no problem orchestrating high energy events in the cosmos. But we here on earth we are limited in our ability to create high energy events. This is a problem that the HEP community has to start thinking about.

  10. Bernhard says:


    you are perhaps right, but these experiments would have to seriously deal with the problem of luminosity. The thing with accelerators is that it gives you a way to make a controlled experiment. If the next energy frontier is at the GUT scale, we are screwed, yes, but we don’t know that, so we keep looking. How are you going to build detectors to look for astrophysical sources and have statistical evidence of BSM physics? And I’m not talking about dark matter, I’m talking about filtering an interesting process from huge SM background and studying a new resonance. Or are you perhaps saying that the sort of thing CAST does is the way to go? Then you need to look for a specific thing that requires a “moving goalpost”, like axions.

  11. Roger says:

    I’d like to know what the prospects for the “desktop experiments” when it comes to falsifying the Standard Model. For example, what precision is needed on g-2 of the muon or the EDM neutron limits in order to explore the multi-TeV region ? Is it a better use of resources to invest in these types of experiments ?

  12. Peter Woit says:


    Precision tests of the SM like the ones you mention can only give you very limited information about the multi-TeV region, and you’re not going to learn anything at all about the Higgs from them. Definitely worth doing, but they’re on a different cost scale and looking at different physics, not really alternatives to accelerator experiments.


    The interesting thing about the LEP3 proposal is that it does have a well-defined goal (detailed study of the Higgs), and the cost looks doable (proponents are talking about .1 x cost of a linear collider). I know of no conceivable way astrophysical observations can study this kind of physics.

    As for whether astrophysics can tell you about the multi-TeV scale, Bernhard points out the problems, but in any case, that’s a different issue.

  13. Roger says:

    Peter – I was thinking more about their ability to detect “new physics” rather than their ability to characterise it. I’m curious as to how much investment (experimentaly and theoretically) would be needed such that it would be possible with one of these experiments to get sensitivity to non-SM processes at a scale beyond the LHC.

  14. Peter Woit says:


    There has been lots of work, theoretical and experimental, on these ideas. Having a convincing violation of the SM prediction would be quite interesting, some sort of clue, and people are working hard on this. But the problem is that typical BSM models are complicated, with vast numbers of undetermined parameters (more than 100 for the MSSM). So, if you see nothing, you put a constraint on a complicated function of 100 or so parameters, which is where we are now, these experiments do give non-trivial constraints (although arguably the muon g-2 value is marginally non-SM, so maybe there is something there). If you see something, and you try and interpret it as for instance SUSY, you are getting very little about the underlying theory, since you just get one number, and it’s a very complicated function of the numbers you are interested in.

  15. David Nataf says:


    There’s a tremendous future in astronomy/astrophysics in terms of fundamental physics, and all competent physics departments are recognizing this: Dark matter, dark energy / modified gravity, number of neutrino species, inflation, anthropic principle, etc. However, we shouldn’t make a bandwagon and pretend that this is where the entire future is going to be.

    Keep in mind that if the particle physics community doesn’t build another particle accelerator, the know how to build another particle accelerator will be lost. You need to have supervisors training the next generation. It is possible that the SM goes up to 1 PeV or whatever but you don’t know. I’ll also point out that null results are scientific results. If somebody has a proposal for a particle accelerator that is simply to see if the SM extends up to 10x the energies probed by the LHC, with no prior that they should not, I would say that this sounds like a legitimate scientific experiment.

    Your proposal that funding be cut from future particle accelerators and shifted to satellites is also politically impossible. First, let’s recognize that the most likely outcome is that money is going to be cut from all pure science sectors. LISA and SIM have been eliminated for example due to lack of funds. Further, they are different budgets. The astrophysics experiment budget is largely an industrial policy for the high-tech sector in the USA. The mirror in the Hubble Space Telescope is the same size as that in many spy satellites, for example. There are different corporate blocks supporting different science experiments.

  16. uiop says:

    The next generation astronomical observatories such as ALMA, SKA, and the thirty-metre class telescopes cost more than a billion dollars, and flagship-class satellite missions are even more expensive (even small ones cost a couple of hundred million). Are accelerators really way, way, way more costly?

  17. David Nataf says:


    The cost of the James Webb Space Telescope is now 6 billion dollars and counting.

    Anyway it’s a false dichotomy. The combined costs are under 1% of the US federal budget. I think it’s safe to assume that if money is cut from one column it won’t be moved to the other column.

  18. Roger says:

    I buy the argument that such experiments are weak in many respects but they may have a crucial advantage in being able to falsify the SM at a higher energy scale than that of the LHC- I should add that I’m an experimentalist at one of the LHC experiments btw.

    I’ve yet to see the limits in precision on calculations and measurement which we think can realistically be achieved on, eg, g-2_muon and other related experiments etc. If it turns out that there is a realistic gain to be had by using “desktop experiments” to access higher energies then investing in them that would be a punt worth taking in my opinion (we’ve spent a *lot* of manpower on looking for black holes and other nonsense). The key argument for me would be whether or not the SM calculations and measurement errors would be believable such that a Nobel prize could be given. Very high precision and accuracy are hard to get right.

    Regarding the limits one could set on, eg, SUSY that’s of small interest (to me). The key thing is, for the first time, we could unambiguously falsify the SM (ok, excluding the neutrino stuff..).

  19. Anon says:

    Hi Roger –The naive standard model prediction for neutron EDM is very small of order 10^{-32} ecm.  This assumes just std model with say PQ symmetry to solve the strong CP problem.  Experiments will probe the region 10^{-26} ecm to 10^{-29} ecm in the next few years.  If it is found experimentally it would be a deviation from just the standard model + PQ symmetry that is the usual way of thinking of neutron edm. For a prediction of lower bounds on neutron EDM based on P and CP symmetries that have been used to solve the strong CP problem (in place of PQ symmetry) please see the paper http://lanl.arxiv.org/abs/1203.2772 — even if new physics is beyond collider reach — say beyond 100′s or 1000′s of TeV this solution still predicts a deviation from the standard model in a large region of its parameter space.

  20. Patrick says:

    To Peter:

    The first argument that comes to mind in favour of LEP3 is certainly the cost, as well stated in your post, but the physics arguments are also numerous. Not only LEP3 can do the same measurements as a linear collider of the same energy, but it does them better. Indeed, most of the Higgs coupling and Higgs mechanism related measurements are statistically limited, and the statistics foreseen for LEP3 is about one order of magnitude larger than that foreseen at a linear collider at 240 GeV, two orders of magnitude larger at 160 (the WW threshold) and three orders larger of magnitude at the Z pole.

    The much larger repetition rate and the number of detectors (up to 4), unique to circular machines, are the reasons for this extraordinary performance. If – God forbid ! – the LHC found nothing beyond the Higgs after three years running at 13 TeV, LEP3 would open the door to unprecedented and unchallengeable precision measurements, including measurements of virtual effects at the Z pole, which in the past determined the top quark mass before its discovery at the Tevatron, which a much more modest luminosity. (LEP3 could repeat the LEP1 programme in … 10 minutes!) Now that the top quark and the Higgs mass are known, very small effects due to heavy particles can be unveiled with these precision measurements, and show the direction to follow for the next-to-next machine.

  21. Bernhard says:


    but at a linear collider one could go in principle to 1 TeV of CME and this would beat LEP3 as a discovery machine. I agree with “LEP3 can do the same measurements as a linear collider of the same energy,” but I guess a good argument for a linear collider is its high CME and luminosity which it´s sensitive to all sorts of very heavy virtual corrections.

    I agree though that to build a “cheap” machine (and much faster, which to me is much more important the the cost) to see what makes sense to build next and at which CME makes sense too.

  22. Roger says:

    If, after 3 years running at 13/14 TeV, the LHC finds nothing other than the Higgs then the last thing CERN should do is close down the LHC and replace it with LEP3. The LHC at that point will still only have delivered a fraction of the luminosity intended for that device. Limits on the direct production of heavy new particles could still be extended a lot (this is especially the case for electroweakly interacting particles).

    Another point worth making is that a lot of innovation takes place with the development of new accelerator technologies. This is not a reason in itself to go for a costlier option but it is something which ought not to be forgotten.

  23. Roger says:

    Another point springing to mind is that LEP3 would be unique among the major facilities in recent years in being essentially a single topic machine. LEP1,2, HERA, the Tevatron, the LHC all had/have a very broad physics program in addition to their core priority areas. Given the success of LEP1,2 in measuring (very precisely) everything under the sun and the comparatively small increase in centre-of-mass energy in going to LEP3 its unclear whether or not a broad physics program would be in anyway attractive or useful. It would also perforce mean a lot of particle physics research groups moving away from collider physics. That may or may not be a bad thing but I’m pretty sure that it would happen.

  24. anonomous says:

    With our without the LEP3, it sounds like we will have 20 years to stew over the results of the LHC before new HEP results come. I think that astrophysics will be the hot area for uncovering new phenomena but as we have no way to control these events it will be hard to confirm models. It looks like we will have a lot of time to theorize on what BSM physics will look like.

  25. Low Math, Meekly Interacting says:

    Thanks in advance for humoring an utter non-expert…

    How realistic is the cohabitation scenario? If it’s doable, shouldn’t LEP3 be a no-brainer at such a low cost compared to, say, the ILC, which is perhaps decades away from being built in even the most optimistic estimates. It’s hard to see how the LEP3 would cannibalize the ILC’s budget if it could be built relatively soon. I can’t imagine there wouldn’t be a wonderful professional symbiosis between the scientists working on two such impressive machines operating in such close proximity. Would the LEP3 somehow preclude the construction of the next generation LHC? Again, if no harm comes to the LHC, why not, BSM physics or no? Precision Higgs physics too pedestrian?

  26. Peter Woit says:


    Politically, I’d guess there’s a problem that deciding to build LEP3 would put the idea of an LC on hold for many years.

    Practically, even if LEP3 and the LHC can cohabit, building LEP3 would probably involved shutting down the LHC for multiple years during construction, as well as during times LEP3 is running. The cost would presumably come out of the CERN budget, forcing a delay or cancellation of plans for higher luminosity (HL-LHC) or (HE-LHC).

    The machines would use the same detectors, and building and running the machines would be the same people as at the LHC now. So, it would be the same people working on both experiments. Right now, I think skepticism about LEP3 is partly skepticism that the thing could be built cheaply, without interfering with the LHC, as advertised, together with worry that it would put a stop to HL-LHC, HE-LHC or LC possibilities.

    But some of the commenters here know quite a bit more about this than I do…

  27. Yatima says:

    Well, I didn’t that Fermilab has already a website on a Muon Collider Project. I hope to see this running before I kick the bucket!


  28. Low Math, Meekly Interacting says:

    Thanks for your perspective!

  29. Pingback: After the Large Hadron Collider, what? | Uncommon Descent

  30. yyy says:

    Building of full-scale Muon Collider will require the following to be built before:
    1. Neutrino Factory [non-colliding beam of circa 30 GeV muons]
    2. Higgs Factory [two colliding beams of muons 63 GeV each];

    It will take few decades.