Medium beta cryomodule

JLab staff prepare to load the medium β cryomodule onto a flatbed semi for its road test.

Jefferson Lab technology, capabilities take center stage in construction of portion of DOE's Spallation Neutron Source accelerator

Jefferson Lab is once again taking center stage, as Lab scientists, engineers and technicians mobilize to provide 81 niobium cavities for 23 cryomodules for the Spallation Neutron Source under construction in Oak Ridge, Tennessee

Thermos bottles usually don't weigh nearly five tons or measure almost 26 feet end-to-end. But these aren't run-of-the-mill containers for soup or coffee. Rather, they're the complex, state-of-the-art supercooled components in which particle beams are accelerated for scientific research.

The Department of Energy's Jefferson Lab technology is once again taking center stage, as Lab scientists, engineers and technicians mobilize to provide what eventually will be 81 niobium cavities for 23 cryomodules for a new federal laboratory, the Spallation Neutron Source, or SNS, under construction in Oak Ridge, Tennessee. JLab is part of a team of federal laboratories — including Argonne, Brookhaven, Lawrence Berkeley, Los Alamos and Oak Ridge — assisting in the design, engineering and construction of the $1 billion-plus SNS, which will provide the most intense pulsed-neutron beams in the world for scientific research and industrial development. JLab is located in Newport News, Virginia.

"We're definitely world leaders in this kind of technology. We've been trailblazers," says Isidoro Campisi, senior scientist with JLab's Institute for Superconducting Radio Frequency Science & Technology. "We're helping to make another generation of machines become practical."

The Lab's engineers and technicians are creating two types of cryomodules. One is known as the "medium β (beta)," version with three cavities per module, and is thus shorter and lighter than its four-cavity "high β" sibling. As at JLab, superconducting radiofrequency techniques and advanced cryomodule design are being incorporated within the SNS accelerator complex to enable low-cost, high-efficiency operation.

Because the speed (represented by the β symbol) of the SNS's negative hydrogen-ion beam will be slightly less than the electron beam in JLab's accelerator, the internal structure of the cavities was slightly adjusted, or "graded," to match the reduced velocities. Therefore, the shape of SNS niobium cavities are flatter ellipses, more like oversized pancakes than their fatter CEBAF predecessors. The operating frequency of RF cavities is measured in cycles per second, given the name hertz to honor the 19th century German physicist Heinrich Hertz, who carried out numerous experiments to clarify the nature of electromagnetic radiation. Most RF cavities operate at very high frequencies, where the appropriate unit is millions of hertz, or megahertz (MHz). SNS cavities operate at 805 MHz, compared to 1497 MHz for CEBAF cavities.

Electromagnetic power is fed into the superconducting cavities via RF couplers. The SNS couplers have been designed to handle a much higher power level than the original CEBAF couplers and so provide a pulsed power — SNS operates in pulses and not in a continuous wave as CEBAF does — with a peak value of over 100 times that of the CEBAF couplers. Campisi was responsible for development of the high-power RF coupler required for the SNS modules.

"For the first time, we're making superconducting elliptical cavities not matched to the speed of light," he points out. "Everybody is pretty excited. We've been working hard to get to this point. We're the youngest of the federal labs involved in the SNS project, so we're happy that we can deliver on what we promised."

Medium beta cryomodule on the flatbed

On the flatbed (top to bottom): Jeff Saunders, Manny Nevarez, Jeff Campbell and Ken Worland load the medium β cryomodule, while on the floor Brian Hannah watches and Frank Humphry manages the overhead crane. Saunders and Hannah are SNS employees from Oak Ridge. Nevarez, Campbell, Worland and Humphry are all Accelerator Division staff.

Prototype takes "test" road trip

In October, a prototype SNS cryomodule was taken on a road trip from Newport News to Virginia's mountains near Charlottesville. Campisi says the sojourn was essential to test the modules' many sensitive parts — among them the "window" that allows for the introduction of radio waves, the vacuum seals between the inner cavity and the outer cryomodule, and the welds that hold everything together — to withstand the inevitable insults of highway travel.

"All cryomodules made here will go by truck to Oak Ridge," he explains. "So we had to evaluate all the factors in that trip. What would happen when you go over bumps in the road, or if you stopped quickly or had to accelerate suddenly? We have to make sure all parts are working correctly before we dare put power in."

In most respects, the SNS cryomodules are virtually identical to their JLab cousins. The innermost components of the cryomodules' three-part system includes the superconducting cavities, a cooling tank to hold the liquid helium, and a Thermos-bottle-like structure known as a cryostat that provides insulation — allowing the cavities to remain cooled to two Kelvin, or nearly absolute zero. At such a temperature the surface currents associated with the introduced radio waves lose all electrical resistance, and provide acceleration with a power dissipation of less than a millionth of that used in energizing accelerators made of normally-conducting materials, such as copper.

At the SNS, four different linear accelerators, or linacs, will accelerate a beam of hydrogen ions to 1 billion electron volts, or 1 GeV. The first three accelerators, the Radio Frequency Quadrupole (RFQ), the drift-tube linac and the coupled-cavity linac, will be made from copper, operate at room temperature and accelerate the beam to 187 million electron volts. The fourth accelerator will make use of JLab cryomodules and accelerate the hydrogen-ion beam an additional 813 MeV. Sixty times per second, the accelerator will produce a one millisecond-long burst of hydrogen ions, a pulse that would stretch for about 270 kilometers (or 168 miles) if the beam were not intercepted. Instead, the beam is wrapped around a ring of magnets, called an accumulator ring, until the whole pulse has been captured in roughly 1,000 turns, like wire on a spool. Then, the opening of an electromagnetic gate allows all the accumulated ions to be delivered to the mercury target in a single microsecond-long pulse. The resultant short, sharp bursts of neutrons are what researchers will be using for their neutron-scattering investigations.

When the SNS facility is complete, researchers will be able to obtain detailed snapshots of material structure, and stop-action images of molecules in motion. Like a strobe light providing high-speed illumination of an object, the SNS will produce pulses of neutrons every 17 milliseconds, with more than 10 times more neutrons than are produced at the most powerful pulsed-neutron sources currently available. The neutrons will scatter from materials under study in such a way as to reveal that material's subatomic structure and properties.

JLab's first SNS cryomodule, the medium-beta prototype, passed its travel test and has been shipped to Oak Ridge. Other production models will follow, with the first tested perhaps as early as January 2003 and shipped the following month. SNS construction is slated to be complete by 2006.