Barry Barish and the GDE: mission achievable

22 February 2008

Barry Barish likes a challenge. He admits to a complete tendency to go for the difficult in his research – in his view, life is an adventure. Some might say that his most recent challenge would fit well with a certain famous TV series: “Your mission, should you choose to accept it… is to produce a design for the International Linear Collider that includes a detailed design concept, performance assessments, reliable international costing, an industrialization plan, and siting analysis, as well as detector concepts and scope.”


Barish did indeed accept the challenge in March 2005, when he became director of the Global Design Effort (GDE) for a proposed International Linear Collider (ILC). He started in a directorate of one – himself – at the head of a “virtual” laboratory of hundreds of physicists and engineers around the globe. To run the “lab” he has set up a small executive committee, which includes three regional directors (for the Americas, Asia and Europe), three project managers and two leading accelerator experts. There are also boards for R&D, change control and design cost.

Barish operates from his base at Caltech, where he has been since 1962 and ultimately became Linde Professor of Physics (now emeritus). His taste for research challenges became evident in the 1970s, when he was co-spokesperson with Frank Sciulli (also at Caltech) of the “narrow band” neutrino experiment at Fermilab that studied weak neutral currents and the quark substructure of the nucleon. He later became US spokesperson of the collaboration behind the Monopole, Astrophysics and Cosmic Ray Observatory, which operated from 1989 to 2000 in the Gran Sasso National Laboratory (LNGS). The experiment did not find monopoles, but it set the most stringent upper limits so far on their existence.

In 1991 he also began to lead the design of the GEM detector for the Superconducting Super Collider project, together with Bill Willis of Columbia University. In October 1993, however, the US congress infamously shut down the project and Barish found himself in search of a new challenge. He did not have to look far, as Caltech was already involved in the Laser Interferometer Gravitational-wave Observatory (LIGO), conceived to search for effects even more difficult to detect than neutrinos. The project was already approved and just beginning to receive funding. Barish became principal investigator in 1994 and director of the LIGO Laboratory in 1997.

Here was an incredibly challenging project, Barish explains, that was “making the audacious attempt to measure an effect of 1 in 1021“. It has indeed achieved this precision, but has not yet detected gravitational waves. “Now it’s down to nature,” says Barish, who found the work on LIGO very satisfying. “There is no way I would have left it except for an exciting new challenge – and the ILC is certainly challenging.” He says that it was hard to move on, “but I felt I could make a difference”. Moreover, he adds: “The likelihood is that the ILC will be important for particle physics.”

At 72 years old, Barish does not expect to participate in the ILC – the earliest it could start up would be in the 2020s. “The plan is short term. The question was whether I could pull together a worldwide team to conceive of a design that will do the job,” he says. With no background in accelerator physics, Barish may not seem the obvious choice for the task. However, he points out that “coming in from the outside, not being buried in the forest, can be very useful”. In addition he believes that he is a good student, and that a good student can be a good leader: “If you do your homework, if the people you work with respect you, then it’s possible.”

An important factor in building the team behind the GDE is that there is not as much history of collaboration in accelerator physics as there is in experimental particle physics. Barish points out that many of the members of the accelerator community have met only at conferences. There has never been real collaboration on accelerator design, so the GDE is a learning process in more than one sense. There are also interesting sociological issues, as the GDE has no physical central location, and meetings usually take place via video and tele-conferencing. Barish likens his job as director to “conducting the disparate instruments in an orchestra”.

In February 2007, the GDE reached a major milestone with the release of the Reference Design Report (RDR) for a 31 km long electron–positron linear collider, with a peak luminosity of about 2 × 1034 cm–2s–1, at a top centre-of-mass energy of 500 GeV, and the possibility of upgrading to 1 TeV. The report contains no detailed engineering; it might state, for example, that a magnet is needed for a certain task, but it does not describe how to build it. The report also contains a preliminary cost estimate, of some $6700 m plus 13,000 person-years of effort.

The final goal will be to produce a strong engineering design, and to optimize costing to form a serious proposal. An appealing deadline is the 2010 ICHEP meeting in Paris. By then there should be results from the LHC that could justify the project. “The main job,” says Barish “is to design a good machine and move once it’s justified.”

In the meantime there is important R&D to be done. Two key areas concern the high-voltage gradient proposed in the machine – an average of 31.5 MV/m – and the effects of electron clouds. Electrons from the walls of the beam pipe cause the positron beam to blow up, thereby reducing the luminosity, and ultimately the number of events. The clouds decay naturally and cease to be a problem if there is sufficient time between bunches, but this reduces the collision rate. The conservative option to keep the rate high would be to have two positron rings to inject alternate pulses into the linac. However, this has huge cost implications so, as Barish says: “There is huge motivation to solve the problem.” One attractive possibility that needs further investigation involves grooving and coating the beam pipe, which could reduce the electron cloud a hundredfold.

However, just before the end of 2007, bad news on funding in both the US and the UK struck a major blow to the plan foreseen at the time that the RDR was released. The UK dealt the first strike, stating that it would “cease investment” in the project, while the US reduced funding for the ILC from $60 m to $15 m as part of a hastily agreed compromise budget for FY2008. Barish recalls the complete surprise of the congressional decision on a budget that President Bush had put forward in February 2007. “We went to bed as normal on Friday (14 December), and woke up on Monday to find the project axed out.”

The cuts in the two countries are both quantitatively and qualitatively different. In one sense the UK’s decision is more serious, as it appears to be a policy decision taken with no input from the community (see Particle physics in the UK is facing a severe funding crisis). Barish says that the main loss here to the GDE is in intellectual leadership. He hopes that continued funding in the UK for general accelerator R&D will mean that the project does not lose people that he says are irreplaceable. In contrast, he expects to see the R&D for the ILC revived in the US budget for FY2009 (starting October 2008), albeit at a level lower than the $60 m originally promised for FY2008. Here the problem is how to cope with the loss of people over the coming months, as there is no funding left to support them in the current budget. Where it hurts most, says Barish, is that the US will not be able to develop the same level of home-grown expertise in the technology required for the ILC, compared with Japan or Europe.

A revival of the ILC in the US budget was a key assumption when Barish and the GDE executive committee met for a relatively rare face-to-face meeting at DESY on 12 January to formulate a new plan. At least the collaboration that Barish has forged is “strong enough to give us the ability to adjust and move on, even with reduced goals”. The aim of the new plan that has emerged is to reduce the scope of the R&D work, but maintain the original schedule of completion by 2010 for items with the highest technical risk, while stretching other parts of the programme to 2012.

The work on high-gradients, underway globally, and tests at Cornell University on reducing the electron cloud will remain high priorities for part one of the newly defined Technical Design Phase, to be ready for 2010. Part two, which will focus on the detailed engineering and industrialization, should be ready by 2012.

Looking further ahead, Barish acknowledges that an ILC-like machine could be the end of the line for very high-energy accelerators, but he points out that accelerators for other applications have a promising future. The GDE itself is already providing an important role in teaching accelerator physics to a new generation. “There is no better way to train them than on something that is pushing the state of the art,” he says. In fact, he sees training as a limiting factor in breeding new experts – whether young people or “converts” from other areas of physics, as many accelerator physicists now are. One problem that he is aware of is that “accelerator people are not revered – but they should be!”.

Despite the recent setbacks with the GDE, Barish remains determined to achieve his mission. “In these ambitious, long-range projects you are going to hit huge bumps in the road, but you have to persevere,” he says. What is vital in his view, is that the agenda should remain driven by science, and that this alone should determine if and when the ILC is built on the firm foundations laid by the GDE. Let us hope that those who fund particle physics have the vision to ensure that one day he can say: “Mission accomplished.”

bright-rec iop pub iop-science physcis connect