Accelerator physicists have a vision: an energy-efficient X-ray source that can make high-resolution movies of molecules in chemical reactions. And the US National Science Foundation (NSF) has backed the dream — since 2005, it has invested more than US$50 million to develop such a source, most likely beneath the campus of Cornell University in Ithaca, New York.

Ten people who mattered in science this year Why swine flu, and a vaccine against it, may have triggered cases of narcolepsy Best evidence yet that our Universe is a hologram

But there is one big problem: despite the inflow of cash, no US government agency has any plans to build the machine.

The source, called an energy recovery linear accelerator (ERL), would be a hybrid of a synchrotron, in which electrons emit X-rays while whirling around a ring, and a free-electron laser, in which straight beams of electrons are induced to produce bright pulses of X-ray light.

The Cornell project is currently receiving $27 million in a single award from the NSF’s materials division — by far the division’s largest grant for instrument development. But in July, the ERL concept was ranked the lowest of three potential next-generation X-ray sources by an advisory panel to the US Department of Energy. And in December, officials at the NSF told Nature that the agency has no plans to move forward with construction.

Despite all this, Thomas Rieker, the NSF programme manager for the ERL materials grant, says that the research effort has been a success, providing component designs that would allow an accelerator to be built quickly. “We wanted to keep our options open,” he says. “That was the impetus for funding it.”

Credit: Source: Cornell Univ.

An NSF advisory panel had strongly recommended in 2008 that the NSF invest in an ERL. So why the turnaround? Agency officials now say that the NSF’s priorities and the budgetary climate have changed, and that a machine costing upwards of $1 billion would not be a good use of taxpayers’ money.

Some physicists are expressing frustration over seeing so much research money apparently going nowhere. “The NSF should really decide if there’s a real need for this in the country,” says Sunil Sinha, a condensed-matter physicist at the University of California, San Diego, who advised on the energy-department panel.

The idea for an ERL was developed in 1965 by Cornell physicist Maury Tigner. It involves injecting electrons into a linear accelerator (linac) and then wiggling the particles to prompt the emission of X-ray pulses. The energy-recovery aspect comes from a loop that ushers the electrons gently around to enter the linac a second time. Their arrival is timed so that their energy is transferred to a new bunch of electrons that will then be accelerated.

The approach has several advantages. For starters, it would be vastly more energy-efficient than a free-electron laser, which recovers no energy. That makes it practical to keep electrons streaming continuously, rather than in widely separated bunches. An ERL can also focus its electron beam, and hence the resulting X-rays, to a tighter spot than the beams in current synchrotron rings, which spread out as they lose energy going around in circles. This would allow for more-advanced studies of the atomic energy levels in materials.

Japan and the United Kingdom have both expressed interest in building an ERL, and there is a small demonstration version of an infrared ERL at the Thomas Jefferson National Accelerator Facility in Newport News, Virginia. But Cornell’s plan is the most advanced X-ray ERL effort in the United States.

The NSF should really decide if there’s a real need for this.

Grant documents stress that the ERL research is not site-specific, meaning that it could feed into projects elsewhere. But most experts think that, if an ERL gets built, it would be at Cornell, where it could reuse the tunnels of an existing NSF-funded X-ray light source, the Cornell High Energy Synchrotron Source (see ‘Electrons in recovery’). “We feel the construction of an ERL can go right ahead,” says Georg Hoffstaetter, an accelerator physicist who is leading the Cornell effort.

The capabilities of an ERL would overlap with those of other planned light sources. In California, the Department of Energy has plans to build a free-electron laser, perhaps by upgrading the Linac Coherent Light Source at the SLAC National Accelerator Laboratory in Menlo Park, California. This machine would provide images of materials with unprecedented resolution in space and time, using fast pulses of high-energy X-ray beams.

Pulses of X-ray light from an ERL would not be as fast, but they would be gentler, and nearly continuous — more appropriate for probing sensitive samples such as biological specimens. However, next-generation ring-shaped light sources, such as an upgrade planned at the Advanced Photon Source near Chicago, Illinois, will also stream continuous light. Although less bright and lower in energy than an ERL, such sources would still prove useful for biological imaging.

The energy department’s decision to go with the other machines will make it harder for the ERL to justify itself scientifically, says Paul Evans, a materials scientist at the University of Wisconsin–Madison. “Defining the niche they’re headed for is the critical challenge.”

Even if the ERL is not built, Cornell scientists say that the research has been useful. Their work is aiding design of superconducting cavities for the Deparment of Energy’s future free-electron laser, which could also one day have energy-recovery loops tacked on. Cornell has also developed a high-current electron gun that could be used in other accelerators to generate X-rays or to study particle collisions.

But although he takes satisfaction in the spin-off possibilities, Hoffstaetter is not ready to give up on the ERL’s construction. “The ERL is a wise investment,” he says.