One of Fermilab’s recent colloquia was by James Rozensweig of UCLA on the topic of Reinventing the Accelerator for the High-Energy Frontier. Video and Powerpoint slides of the talk are now available.
Current accelerator technologies are up against very fundamental physical limits. In the case of circular proton colliders like the LHC, the fundamental limiting factor is the strength of the magnets and the size of the ring. The LHC uses a 27km ring and 8.36 Tesla superconducting magnets, and the energy scales linearly with the magnet strength and ring size. So one could get beams an order of magnitude more energetic than those in the LHC by using a 270km ring, but the cost of such a thing is likely to be prohibitive. One could also try and design higher field magnets, but the current record for this kind of magnet is only about 16 Tesla.
For circular electron colliders, the limiting factor is the energy loss to synchrotron radiation and these energy losses scale as the fourth power of the energy. LEP was probably the highest energy collider of this kind anyone is ever likely to build, since it already was using an amount of power a sizable fraction of that of the city of Geneva. One could try and use muons, which are much heavier so synchrotron radiation is not a problem, but they decay quickly so there are lots of problems with storing them in a collider.
These considerations mean that there is only one viable route to much higher energies, a linear collider, and this is the path that the ILC project is pursuing. What limits the energy in a linear collider like the ILC is the combination of the energy gradient one can achieve and the length of the machine. The superconducting RF cavities being studied for use in the ILC are inherently limited to gradients of less than 40-50 MV/m, with something like 35 MV/m a likely realistic number. With this gradient, to get up to a TeV in energy would require a length of about 33 km, about at the outer limits of what is possibly affordable. Realistically, to get to higher energies than this, one needs to find some way to get much higher energy gradients.
Rozensweig’s talk covers this material, but goes on to discuss various exotic new technologies that in principle can provide these much higher gradients. He describes progress on a long list of these, the most advanced of which is the CLIC project at CERN which uses the wake-field from a drive beam (a second accelerator). Much more exotic are various proposals involving lasers and plasma waves, some of which have been used to achieve gradients of 40 GV/m over short distances in the laboratory.
So, now all one has to do is to achieve a stable, high luminosity beam and make this work over kilometers not centimeters…. Not going to happen any time soon, but the distant future of high energy physics may depend on this kind of technology.
Update: I should also have mentioned here an article on this topic in the current (February) Scientific American entitled Plasma Accelerators.
Peter says “… One could try and use muons, which are much heavier so synchrotron radiation is not a problem, but they decay quickly so there are lots of problems with storing them in a collider.These considerations mean that there is only one viable route to much higher energies, a linear collider, and this is the path that the ILC project is pursuing. …”.
As an advocate of muon colliders ( see my comments on Peter’s blog entry at http://www.math.columbia.edu/~woit/wordpress/?p=328 ) I have some biases here, but it seems to me that Fermilab is putting all its eggs in the ILC basket at least in part because of possible radiation hazard to Chicago of a muon collider at the Fermilab site.
Unfortunately for Fermilab, there are a lot of uncertainties about building the ILC in the first place, and about its site if it is ever built.
For example, a Charles Seife article in Science 308 (1 April 2005) 38-40 (and no, it does not seem to be an April Fool joke, although some physicists probably wish it were):
“… Fermilab’s Tevatron, due to shut down around 2010, could be the last large particle accelerator in the United States. …
… Fermilab’s Butler says, “a large number of U.S. physicists at the Tevatron are already planning to work at the LHC; they have exit strategies.”
But Butler isn’t happy about the new venue. “This field is being outsourced,” he says.
The one big hope for U.S. accelerator physics is the ILC. “We’re going to go for
the linear collider,” says Orbach. …
For Butler and other physicists, the projected completion date for the ILC in the middle of the next decade is another huge obstacle. … a timetable that puts the ILC at the end of the next decade or beyond would leave an entire generation of physics students without access to an accelerator in the United States. …
But the leader of the majority party in the U.S. House of Representatives isn’t ready to make a firm commitment. “If [the ILC] fits within certain parameters, we’d like to keep it in the U.S.,” Hastert says. The biggest of those parameters is the cost, estimated by DOE at $12 billion, of which the host country would presumably pick up half. …”.
Even if the ILC is built, if it is part of an international collaboration involving countries like Japan and China (which hold ots of dollars due to the inability of the USA to compete in trade of manufactured goods), then it may be that the money people in Japan and China will insist on an Asian site. As a Physics Today article at http://www.physicstoday.org/pt/vol-54/iss-9/p22.html about Snowmass 2001 said:
“… But one cannot see the Pacific Ocean from the Illinois prairie. This geographic truth was pointed out rather bluntly at Snowmass by KEK director Hirotaka Sugawara. He reminded his audience that the joint work on the copper linac design in the US and Japan was undertaken with the understanding that the machine would be sited somewhere on the Pacific Rim … Sotoru Yamashita of Tokyo University was more specific: Europe already has the LHC … it was now Japan’s turn to build an energy-frontier machine. …”.
In short, the future of high energy experimental physics in the USA looks to me to be highly uncertain at best, which is doubly sad because the USA theoretical high-energy physics community seems to be insistent on following a program that is Not Even Wrong.
Tony Smith
http://www.valdostamuseum.org/hamsmith/
The main problem of a TeV-scale muon collider is it would emit neutirnos, giving unacceptable radiation hazards, see e.g. hep-ex/0005006.
Tony,
I think it’s true that high energy physicists are putting all their eggs in one basket, the ILC. The problem is that, for reasons outlined in the posting, there isn’t really any alternative. Things like a muon collider or exotic acceleration technologies are a long ways away from the point where one could come up with a viable design and try and get someone to fund it. The ILC is really the only possibility for at least the next ten years or so, and probably longer.
“Much more exotic are various proposals involving lasers and plasma waves, some of which have been used to achieve gradients of 40 GV/m over short distances in the laboratory.”
It seems to me that accelerator physics is frozen in time; it hasn’t changed much for the last 25 years. Aren’t these “exotic” proposals at least that old? Am I being fair in accusing HEP of a shameful, inexcusable failure to innovate in accelerator physics? Isn’t this failure a mayor cause for the failure of HEP theorist to produce predictive theories?
I was going to write a reply to Quantoken about electric v magnetic fields, but the post got deleted. Oh well … probably for the best I think. I was also going to say essentially the same thing as Woit in previous post …. the scale of machines for TeV energies is such that there can really be only one in the world. The ILC is about the only scheme which is buildable with today’s technology (muon colliders are far future). Even then it will take many years. Exotic acceleration technologies are probably the only hope in the long term, but the key word is “long”.
Peter said “… The ILC is really the only possibility for at least the next ten years or so, and probably longer. …”.
If so, then:
1 – What odds do USA high-energy physicists give that it might get built ?
2 – If it were to be built, what odds do they give that it would get built in the USA ?
3 – If it were to be built outside the USA, what do they think would be the consequences for USA high-energy experimental physics ?
If they aren’t willing to give reasonable odds for the sake of discussion, does that mean that the pattern of wishful thinking that has corrupted the superstring community has entered the experimental community ?
Tony Smith
http://www.valdostamuseum.org/hamsmith/
I read anon’s post just after submitting the previous.
If I may reply to anon ….. yes, acceleration techniques haven’t changed much in 25 years (though superconducting rf was an innovation). The exotic ideas (use of plasmas) have indeed been around for that long, if not more. Shameful failure to innovate? The plasma wakefields etc are difficult ideas, not easy to make practical, even after all these years. But there has not been a lack of trying.
Tony:
Good question. But I guess no one is willing to discuss the odds that things they wish for will be built. It’s painfully clear, if one is honest to himself in admitting, the odds are pretty low at this moment.
Frankly I think it is now a good time to mossball things up for a few decades, and let the public money be spend on something more useful in short terms, like alternative energy research. Later, when you guys figure out a better way of doing experiments, then we can fund you again.
Quantoken
There has been tons of discussion about the challenges of building the ILC. See for example Turner’s talk below.
http://www.linearcollider.org/cms/?pid=1000069.
The odds of the ILC depend very much on what happens in the next few years: it is likely that if the LHC does not find anything but a Higgs, then the ILC will not be built. If there are lots of surprises, then the chances are a lot better.
Site selection is usually put off as much as possible, so nobody knows the odds that will be built in the US.
If it isn’t built in the US, then US experimentalists will rack up a lot of frequent flyer miles. But they’ll cope.
Tony and Quantoken,
I don’t think anyone has a good estimate what the odds are that the ILC will be funded and built in the US. It certainly is true that if it is not built in the US, that will be very bad news for US high energy physics, and specifically for Fermilab if it is not built there.
My guess is that the decision whether to build the ILC won’t be made until 2009-2010, at which point initial results should be coming in from the LHC. A lot depends on what the LHC finds. If it finds something interesting that would require the ILC to fully understand, that will help a lot. If it finds nothing at all and it then looks as if there is nothing new for the ILC to study up to its highest energy of 1 Tev, that will be a big problem. No one knows which of these is more likely, that’s the biggest question in HEP right now.
On top of this, everything depends on the US budget situation several years down the road. I don’t think anyone can predict that either, except that the US will still be struggling with what to do about budget deficits. Bush’s announcement last night that he wants to double the budget for fundamental physics research may or may not lead to anything and may or may not go to HEP. We should know soon when his administration submits their FY2007 request. But the bottom line is that the budget situation is unclear for later this year, much less for 2009-2010.
And, no, even if the idea made any sense, you can’t “mossball” an entire scientific field. Much of the budget pays scientists and technicians. If you throw all of them out of their jobs and have them move into other careers, they’re not going to be coming back.
anon:
You’ve got to be kidding me by implicating accelerator physics as the major cause for HEP theorists not producing predictive theories. “Oh so sorry Mr./Miss String Theorist, we can’t provide the gazillions of TeV you need to verify your theory, so it’s our fault your theories don’t predict anything.”
And as for innovation, do you have any idea of how incredibly hard it is, for example, to build an experiment to test plasma wakefield acceleration, using an electron beam with the right properties? Sure these ideas existed on paper for a long time, but the technology needed is incredible.
The real problem is money. We could build a linear TeV collider now if we want, but we would need a LOT of real estate and a LOT of power and a LOT of money. This is obviously why accelerator physicists are investigating high-gradient alternatives, but these alternatives are technologically challenging. Even building a “warm” linear collider at X-band (11.424 GHz) to achieve higher-gradients as compared to the SLC at S-band (2.856 GHz) is quite challenging. The high fields alone lead to RF breakdown and pulsed heating that prevent high gradients from being achieved. These phenomena opened up a lot of study involving material science and surface physics. There are just so many impediments to technologically achieve high gradients that it requires years of experiments, analyses and simulations. But more importantly, it requires a lot of money which is hard to obtain. The accelerator field is making a lot of progress, but this progress is not really seen by an outside observer because they usually only see the big machines that finally get built.
Agreeing with sunderpeeche, there has been a lot of progress in acclerator research over the last two decades. Part of the apparent slowdown compared to the blinding progress of the preceding four or five decades is that all the low-hanging fruit is gone. The new developments are harder to realize, just like in any field of technology.
Regarding the ‘eggs in one basket’ criticism, I don’t think there is much choice politically. If they don’t get the ILC, they can push for something more exotic perhaps, but they need to get everyone on board now, with no apparent alternative and present a unified front to have any hope of landing the ILC.
As for the odds, you make the best proposal you can as a team and move on with it. The numbers I’ve seen don’t seem unreasonable (e.g. not terribly out of scale with currently funded large science projects), the biggest problem is to make all of the politicians agree. THere’s not much point in doing statistics and hand-wringing when you have no real alternative in any case, is there?
Peter said “… Bush’s announcement last night that he wants to double the budget for fundamental physics research may or may not lead to anything and may or may not go to HEP. We should know soon when his administration submits their FY2007 request. …”.
Deb Riechmann wrote an AP article dated 1 Feb 2006 and carried on a Forbes web page at http://www.forbes.com/entrepreneurs/feeds/ap/2006/02/01/ap2492673.html that said in part:
“… In addition to the money for the tax credit, Bush wants to double over 10 years the investment in agencies that support basic research in the physical sciences and engineering, a $50 billion commitment.
The research money – $910 million in fiscal 2007 – would fund research programs at the National Science Foundation, pay for 500 new researchers at the Commerce Department’s National Institute of Standards and Technology and 2,600 new researchers at facilities operated or assisted by the Energy Department. …”.
If the ILC costs $12 billion ( a DOE estimate according to a Charles Seife article in Science 308 (1 April 2005) 38-40 ), then
it might fit comfortably within a $50 billion increase over the next ten years
and
if the USA is the major funding government, then the USA should be able to dictate the site location, and it is possible that Fermilab might get the ILC and give high-energy experimental physicists a home in the USA.
Of course, that presupposes that the Bush budget idea would be supported by the next administration. Since Bush cannot run again, Fermilab will have to hope that the new president will not do to the ILC what Clinton did to the SSC (i.e., campaign promises to support it, followed by pulling the rug from under it).
Tony Smith
http://www.valdostamuseum.org/hamsmith/
The implication that the SSC was viable in 1993 is not too credible. Many members of both parties in congress had the long knives out for that project for years before they finally succeeding in shutting it down – it was the prior Bush administration that kept it alive by lobbying so hard.
As for whether the SOTU numbers for DOE, NIST and NSF will ever materialize, I would suggest looking at past promises and comparing them to realities to develop one’s own prediction.
Peter:
It’s wishful thinking that if LHC finds something interesting then ILC may be founded. Just look at Hubble Telescope, it’s finding something interesting every day but yet it’s let to retire early. The bottom line is you need to look at priorities. And high energy particle physics research is clearly NOT the priority right now.
Don’t you guys listen to the Bush address carefully? The predorminant theme is “energy”, “alternative energy research”, “Get rid of America’s oil addiction”. That certainly has something to do with department of energy. But high energy physics is now just a step son of department of energy. Bush emphasised nanotechnology (material science), super computing, and alternative energy research.
Predictably, money will be shifted away from un-important fields, to areas that are now considered priority. The priority is alternative energy research, not dark energy research. The important things is replacing black gold, not researching black holes. For high energy physics research, the party will soon be over.
Quantoken
J. F. Moore said “… As for whether the SOTU numbers for DOE, NIST and NSF will ever materialize, I would suggest looking at past promises and comparing them to realities …”.
For two examples:
during the 2000 campaign, Bush promised tax cuts for big business, and he delivered; and
during the 2003 SOTU, Bush promised to remove a threat of Iraq under Saddam Hussein, and he did indeed remove Saddam Hussein as ruler of Iraq.
Therefore, it seems to me that if Bush feels strongly about a big idea, he will move to support it. That leaves the question as to whether or not Bush feels strongly about high-energy physics as a big idea.
Quantoken says “… high energy physics is now just a step son of department of energy. Bush emphasised nanotechnology (material science), super computing, and alternative energy research. …”.
However,
as J. F. Moore said, “… it was the prior Bush administration that kept … the SSC … alive by lobbying so hard …”, and it seems to me that with respect to high-energy physics it might be like-father-like-son.
Even though I agree with some things said by J. F. Moore, his statement about my comment:
“The implication that the SSC was viable in 1993 is not too credible”
is inaccurate and easily refuted as follows:
I did NOT state, and did NOT imply, that “the SSC was viable in 1993”.
In fact, I said “… what Clinton did to the SSC ..[was]… campaign promises to support it, followed by pulling the rug from under it …”.
Clinton was elected in 1992, and when he took office in 1993 he pulled the rug from under the SSC and it was terminated. According to an article by David Ritson in Nature 366 (16 December 1993) 607-610:
“… even with Bush’s personal support, the SSC had a narrow escape in 1992.
… in the summer [of 1993] … Clinton publicly supported the SSC. But how far this support extended in private is questionable .
… The [USA] house [of representatives] instructed its representatives in the conference committee to terminate the SSC. This action ended the SSC early in October [1993] …”.
There are lessons that the ILC supporters should learn from the SSC.
First, unless regional rivalries about sites within the USA (such as Illinois v. Texas) are fully reconciled, Clintoneque backstabbing can be fatal.
Second, the ILC management must avoid undisciplined cost controls and arrogant behaviour, both of which were severe problems with the SSC.
All things considered, it seems to me that maybe Bush might feel strongly enough about high-energy physics to support the ILC, but since Bush cannot run in 2008, it will be necessary for his successor also to support it.
Tony Smith
http://www.valdostamuseum.org/hamsmith/
PS – Just so that nobody will have grounds to accuse me of being a pro-Bush propagandist, I will state for the record that in 1992, 1996, 2000, and 2002 I voted for Perot, Perot, Libertarian, and Libertarian for USA President.
My point was simply that congress was ready to kill SSC for years. Whatever motives you want to ascribe to Clinton notwithstanding, the vultures were circling well before he was elected, which is supported by your pull quote from Nature. If a line item desperately needs vigorous presidential support to save it, it doesn’t require “backstabbing” to end it, it requires folding your arms and doing nothing.
More importantly, I think you are fooling yourself if you think the Bush administration has any commitment to physics, or even science at any level. Most of the things in these speeches are pure fluff. How about evaluating the hundreds of promises all presidents have made in SOTU addresses, including the promise of _every_ president since Nixon, some of them multiple times, to reduce our dependence on foreign oil, when over the same period the fraction has gone from under 30% to about 60%?
By the way, a quick calculation shows that this bold doubling initiative, even if it happens, will result in a 7% annual increase. Between payroll, insurance, and facilities charges, most places have a hard time keeping status quo with only a 5% increase. I am being kind and not counting overhead, which always rises faster.
I had a feeling that particle accellerators were in for a redesign. I knew that circular accellerators did not scale very well, but to the fourth power! Gad-zooks!
I wonder has anyone tried some genetic algorithms to “evolve” a better energy gradient process, rather like an evolved antenna.
It is a consequence of basic (classical) electrodynamics that the radiated power goes like P ~ E^2B^2 where E= energy and B = magnetic field. Solve the Lorentz eqn for ultrarelativistic motion in a circle of radius R and you get B ~ E/R, so P ~ E^4/R^2. Nobody has found a way around this. So to reduce the power output one needs big rings (large R). The circumference of LEP was 27 km. FWIW LEP dissipated 18 MW at its top energy of sqrt(s) = 209 GeV. Also FWIW the SSC (20 TeV protons) would also have dissipated synchrotron radiation (I think several kW), the heat load of which would go into the superconducting magnets … another problem for the machine/magnet design.
An AP article said “… Bush wants to double over 10 years the investment in agencies that support basic research in the physical sciences and engineering, a $50 billion commitment. …”.
J. F. Moore said “… a quick calculation shows that this bold doubling initiative, even if it happens, will result in a 7% annual increase.
Between payroll, insurance, and facilities charges, most places have a hard time keeping status quo with only a 5% increase.
I am being kind and not counting overhead, which always rises faster. …”.
Although I am not sure exactly how the AP article’s “bold doubling” deals with inflation etc,
if I use for the sake of argument J. F. Moore’s method, it seems to me that, based on a present $50 billion budget “in agencies that support basic research in the physical sciences and engineering”:
$50 b x (1.07)^10 = $98 b, giving roughly a “new” $48 b over the initial $50 b, or doubling as described by J. F. Moore;
and
$50 b x (1.05)^10 = $81 b for J. F. Moore’s “status quo”,
so
that the “bold doubling” from $50 b to $50 b plus $ 50 b = $100 b would give $19 b over and above a 5% “status quo” amount,
and
$19 b is still more than the $12 b cost of the ILC,
so
even under J. F. Moore’s own reasoning, “basic research in physical sciences and engineering” could:
maintain its “status quo”;
build the ILC with USA money at a USA site; and
spend $7 b on new and different stuff.
I should also mention that if ILC were sited at Fermilab, funds that now go to the Tevatron (within the present $50 b) could be assigned to the ILC, thus reducing the ILC’s demand for new money.
Of course, J. F. Moore could move the goal posts by pleading about ” overhead, which always rises faster”.
However,
maybe faster-rising-overhead is a symptom of poor cost control management, which is one of the factors that killed the SSC,
and
maybe refusal to restrain overhead is a symptom of management arrogance, which is another of the factors that killed the SSC.
Tony Smith
http://www.valdostamuseum.org/hamsmith/
PS – If one is asking a President for support for ILC, and a President makes public statements about a budget that could support ILC, it seems to me to be impolitic to say things like “… you are fooling yourself if you think the Bush administration has any commitment to physics, or even science at any level …”, particularly in view of the strong support for the SSC by the first Bush administration. Why bite the hand from which you are asking to be fed?
Tony and J.F. Moore,
I’d suggest postponing further detailed speculation about exactly what the Bush administration is going to do for a few days, until they release their 2007 budget figures. In any case, discussing what this means for 10 years from now is pointless: Bush will be long gone, things will be very different, and the history of promises like this is that they are forgotten after a few years. But soon we’ll know something solid that means something, whether and how Bush’s rhetoric will be reflected in next year’s budget request.
Very good points, Peter. I totally agree about trying to predict the politics of this a decade out.
I will simply clarify at this stage that in general, budget increases are not pre-normalized to inflation, let alone expected rising costs. I agree that even so there is room for ILC. We will have to see what happens.
As I recall it wasn’t only politicians who were anti-SSC. Phil Anderson had a long letter in NY Times attacking it. And it’s very hard to justify spending billions on research that is so removed from people’s everyday lives. There’s the problem.
Phil Anderson persistently attacked the SSC. (What is hypocritical is that after the SSC was cancelled solid-state physicists expected some of the $8B to go to their field. It didn’t. It’s not as if there was an $8B pot and “how shall we spend this on physics”?) It is also true that the initial cost estimate was approx 1.5B, this increased to about 4.5B when the Central Design Group (which leased space at LBL) developed their design. When SSC took over from CDG (and relocated to Texas) the revised cost estimate was 8B, with a possibility of 11B. This was simply too much for anyone and the thing was cancelled. Also the SSC was advertised (at least initially) to “regain the lead in HEP for the USA”, and for the USA to have the “world’s highest energy accelerator”. It was hard to get other countries to contribute so that the USA could wave the flag.
It’s called due process. The paper was endorsed by a several people independently. I have a PhD in theoretical physics from the University of California and unlike some of you who are being outsourced I will not have to be driving cabs or working at Burger King. 🙂
Well, J. Sarfatt did some publications on physics during the years 1963-1973, and I guess he had a theoretical phycsics PhD 🙂