These units are described in detail below. An over-riding consideration in choosing these units was their ease-of-use. Especially in the early years of HELIOS, all users will be inexperienced with its characteristics. Further risk to the users� experiments caused by complexities of cryostat control, for example, are important to avoid. Displex units also provide for more convenient computer interfacing, and promise a greater degree of automatic control of data acquisition.

We request funds for item 1 only. We will obtain funds for items 2-4 from other agencies, possibly the National Science Foundation, if the HELIOS spectrometer is itself funded.

3.3.1 Sample rotation stage

......

......

A second degree of freedom for specimen orientation could be provided by a tilting stage that mounts within the sample space. Such a tilt axis will likely be important for data acquisition from single crystal samples. The first such experiments will be performed at room temperature within the thimble described in section 3.3.2. With increased experience in single crystal experiments, we will design tilt axes for other sample environments.

3.3.2 Thimble for specialized sample environments (Collin)

You might mention that with the oscillating radial collimators, there will be little measurable scattering from the thimble itself.

3.3.3 Oscillating radial collimator (Collin)

3.3.5 Low temperature closed-cycle refrigerator (3.6 - 350 K)

A closed-cycle refrigerator unit is already in place at LANSCE for use on PHAROS, and we expect this unit to be adaptable to HELIOS. Since low temperatures are required for much of the scientific program, it is important to acquire another displex unit. We may choose to duplicate the PHAROS unit, but our preference is to obtain a unit with extended low temperature capability. One approach is taken by APD Cryogenics in their commercial Heliplex design. It uses a closed-cycle refrigerator in combination with a Joule-Thompson heat exchanger to achieve base temperatures below 3.6 K.

3.3.4 High temperature closed-cycle refrigerator(<30 - 650 K)

Ross W. Erwin of NIST has modified a closed-cycle refrigerator to enable it to reach temperatures of 650 K. The trick is to use a sapphire spacer between the cooling head and heating stage. The thermal conductivity of sapphire is temperature-dependent so that it provides better heat conduction at low temperatures than at higher temperatures. The NIST group is agreeable to providing us with drawings and technical advice. We plan to build our own unit, however, which is based on a closed-cycle refrigerator unit from APD or Leybold, a heating stage from Air Products, and a sapphire link between them. Although the closed-cycle refrigerator has a base temperature of 6.5 K, we expect that with the sapphire spacer the base temperature should be between 25 and 30 K.

3.3.5 Temperature Controller / Sample Vacuum??

We will purchase a computer-interfaced temperature controller for use the displex systems, such as a Lake Shore 340 unit.

?? High vacuum pumping station, turbopumped, interlocked, and computer interfaceable... do we need to propose this, Collin? I expect $20,000.

3.3.6 Furnace (600 - 2073 K)

We have evaluated the performance and reliability of several high temperature furnaces that have been used previously for neutron scattering work. The clear choice is the �ILL� furnace manufactured by AS Scientific Products, Abingdon, England, and shown below in Fig. X. This unit uses a cylindrical Nb heating element around the sample, which can be as large as 4.5 cm diameter and 10 cm high. Cylindrical Nb radiation shields surround the heating element, and the water-cooled housing has thin Al windows for 360� neutron access around the sample. Neutron access is �20� out of the plane of the spectrometer. This unit has relatively low background, but the background will be essentially negligible with incident beam collimation and the use of radial collimators in the HELIOS vacuum vessel. The unit is reasonably priced, about 65 k$ with vacuum system, power supply, interlocks, and miscellaneous spare parts.

An important advantage of this ILL furnace is that it is a mature product, with an integrated vacuum system, temperature controller, power supply, and safety interlocks. It will likely be useful for other experimental facilities at LANSCE � certainly it will be useful on PHAROS. There is a possibility that an ILL high temperature furnaces will be procured for the PHAROS spectrometer. Depending on timetable and funding for HELIOS, it may be possible to find partial external support for this furnace as it is being purchased for work on PHAROS.

Fig. X. The ILL high temperature furnace. The thin Nb heating element and Nb radiation shields are in the neutron beam, but we have had little trouble with spurious scattering in reactor experiments with collimated beams at ORNL.

3.3.7 Issues with pressure cells

Pressure experiments with a samples having masses of tens of grams and greater are often performed by mounting the sample in a canister that can be compressed in a hydraulic press, and then locked in a compressed position. Upon removal from the hydraulic press, the pressure is maintained in this "pressure cell", which is a compact unit that can be moved into the specimen region of the HELIOS spectrometer, for example. An obvious feature of the pressure cell is that it is made of high strength materials such as steel, and typically more such material when higher pressures are required. For pressures of 10 kbar or higher, spurious scattering from the pressure cell can overwhelm the scattering from the sample itself. This problem can be alleviated considerably by collimation. The incident beam is collimated to have a width comparable to the sample, and the scattered beam is collimated with the oscillating radial collimators. The small size of the vacuum vessel of HELIOS facilitates such scattered beam collimation with oscillating radial collimators. For this reason we expect it to be easier to perform high pressure experiments with HELIOS than with the PHAROS spectrometer.