Sample preparation of intermetallic compounds for PAC measurements: We arc-melt pure metals under argon together with carrier-free 111In activity. Shown is Bonner Walsh melting a sample. Melting process takes about one second and results in a spherical sample about 3-mm in diameter and having a 100-mg mass.
More melting. Welder's goggles are a must.
Bulletin board outside the measurment laboratory
Data acquisition computer with monitor showing four double-sided PAC coincidence spectra. Racks of modules at the far ends of the table hold electronics for the two PAC spectrometers.
Display showing raw data PAC spectrum for 111Cd probes in ferromagnetic nickel at room temperature. Coincidence counts are recorded versus nuclear lifetimes. "Wiggles" show Larmor precessions of magnetic moments of probe nuclei in the internal magnetic field in Ni, superimposed on the exponential lifetime decay of the 120-ns PAC level. A typical measurement takes about a day.
John Bevington at the console.
Sample holder in one of our PAC vacuum furnaces. The sample sits in a central cup of Mo. Three tungsten filaments spot-welded between the two wire rings are used to heat the sample incandescently. The two rings and filaments can independently be brought to a potential of about -1000 volts relative to the grounded central cup, leading to thermionic emission of electrons from the filaments and electron-beam heating of the cup. Before bolting the vacuum housing on top, the entire source assembly is covered with two concentric, cylindrical heat shields of thin molybdinum foil to reduce radiant heat losses. Temperatures are measured using a thermocouple junction at the bottom of the cup that is in contact with the sample. About 15-20 watts of joule heating is sufficient to raise the temperature to 1000 C.
Closeup of a new sample holder for the vacuum oven. Tungsten filaments will be welded between the two wire rings for use as heating elements. Thermocouple leads are in holes in the central ceramic support.
Each PAC spectrometer has four BaF2 scintillation crystals mounted on fast Hamamatsu photomultipler tubes. The PM tubes are mounted on Amperex base assemblies. The sample is at the center of the aluminum vacuum housing. Water flows through the copper tubing to cool the vacuum housing.
Rack of NIM modules for one spectrometer, including the high-voltage power supply, two discriminators, four amplifiers, eight single channel analysers, and a time-to-amplitude converter. Additional electronics are the EG&G digital delay generator shown below the rack and a homebuilt logic module (not shown) that sorts out detector pairs from which timing pulses signaling formation and decay of the PAC level were derived.
Online data reductions calculated in real time show progress of experiments. The above image shows the approximate perturbation function for 111Cd impurities in an Al4Sr compound. Shown is a double-sided spectrum with time t=0 at the center of the spectrum and equivalent data at positive and negative times out to limits of plus and minus 500 ns.
Fitted perturbation function for 111Cd in CoGa3 measured at 400 C, in a double-sided spectrum extending from -200 to +200 ns . The fitting shows that the Cd-probe occupies just one of two inequivalent Ga-sites. There are three frequency components in the time series, all which arise from a single site (see following image of a fourier transform). The fit shows that the probes have a fundamental quadrupole precession frequency of 334 Mrad/s, and an EFG asymmetry parameter of 0.56. As can be seen, the signal is essentially undamped, as indicated by the fitted relaxation frequency of 0.91 MHz. Such a low relaxation frequency might be due in part to diffusional jumps of probe atoms in the crystal structure at the 0.9 MHz frequency, but it might be due in part to instrumentation limitations that create an effective lower limit.
Fourier amplitude transform of the time-domain PAC perturbation function measured at 400 C, shown in the previous image. The spectrum shows the signal from a single site having three frequency components at 333, 485 and 810 Mrad/s, for the spin 5/2 PAC level of 111Cd. When the three frequency harmonics are in proportion to 1 : 2 : 3, the electric field gradient is said to be axially symmetric, meaning that there is an axis of three (or higher) symmetry through the probe atom site. That is not the case for the indium site occupied in CoGa3. Looking for additional signals, note the teeny-tiny peak at 640 Mrad/s that appears to be above the level of "digital white noise". That small peak might be part of another signal frequency-triplet that has not yet been identified.
Fitted perturbation function for 111Cd in CoGa3 measured at 500 C, in a double-sided spectrum extending from -300 to +300 ns . The fitting shows that the Cd-probe occupies just one of two inequivalent Ga-sites. There are three frequency components in the time series, all which belong to this one site. The fit shows a fundamental quadrupole precession frequency of 329 Mrad/s, and an EFG asymmetry parameter of 0.56. As can be seen, the signal is damped and was fitted with a relaxation frequency of 6.10 MHz, much more damping than in the spectrum shown before measured at 400 C. This damping is almost certainly caused by diffusional jumps of probe atoms in the crystal structure at the 6 MHz frequency, although further theoretical work will be needed to interpret the detailed diffusional jump mechanism.
We can make PAC measurements down about 20 K using the closed-cycle He refrigerator shown above. Recent studies have not needed this capability.
A Mössbauer spectrometer setup from Ranger Scientific is shown for measurements in transmission at room temperature. The electromechanical drive, which incorporates a laser interferometer for absolute velocity measurements, is at left. The shield for the intense radioactive source is at center and a proportional counter with krypton gas, especially suitable for experiments using the 14.4 keV level of 57Fe, is shown at right.
Closed cycle He refrigerator available for making Mössbauer measurements on sample down to about 20 K.
Dale Brewer setting up for an XAFS measurements at the Advanced Photon Source at Argonne National Laboratory in 2002. A few exploratory experiments were carried out to determine lattice locations of solute atoms in intermetallic compounds.
Circular tunnel at APS synchrotron at Argonne, 2002.
