Perturbed angular correlation of gamma rays (PAC)
PAC can discriminate among local environments of probe atoms in
solids. Internal fields in solids exert torques
on nuclear moments:-- a magnetic field exerts
a torque on the magnetic dipole moment and an electric-field
gradient (EFG) exerts a torque on the electric quadrupole moment. These
nuclear hyperfine
interactions lead to frequencies of precession of probe nuclei that
are
proportional to the internal fields and are characteristic of the
probe's lattice location. .
Many PAC probes have an excited nuclear state reached through a
gamma-gamma
cascade. Start and stop gamma rays signal creation
and
decay of an intermediate state and (importantly) there is an
anisotropic angular
correlation
between directions of emission of the two gamma rays. Such is the
case, for example, following decay of 60Co, for which the
lifetime of the intermediate nuclear state is very short.
A classic experiment in advanced teaching laboratories involves
measuring the anisotropy of the angular correlation of radiations from 60Co
by measuring the coincidence counting rate of the two gammas as a
function of the angle subtended by the two detectors. For
longer-lived states, the nuclear spins can precess through
appreciable
angles and interaction frequencies can be measured with good
precision. A small number of probes have long lifetimes and other
favorable nuclear properties for "table top" PAC experiments, including
111In,
decaying
into 111Cd, and 181Hf, decaying into 181Ta.
Internal fields in solids are produced mostly by charges and spins
within the first few atomic shells, with more distant charges
and spins only contributing to inhomogeneous signal broadening.
As a result, interaction frequencies can be used to
characterize
the local atomic environments in which probe atoms find
themselves.
Once a frequency has been identified with an underlying environment, it
can be monitored in measurements made following diffferent methods of
sample
preparation, or for changing conditions and temperature.
PAC Methodology
In the figure below, a radioactive probe atom in a
crystal is shown at top. Start
and stop gamma rays signal
formation and decay of the
intermediate
PAC nuclear state and are detected in scintillation detectors. A
clock and histogramming memory is used to record time intervals between
start
and stop gamma rays. In general, this would give the distribution
of individual nuclear
lifetimes, but here the measured lifetime decay curve is modulated by
spin
precessions
of the nuclei. The modulating, or perturbation function G2(t)
shown at the bottom of the figure contains all information about
internal fields in the
sample.
It is a kind of "spin rotation" pattern for probe
nuclei.

The perturbation function G2(t) shown in the graphic
above comes from
measurements on 111In/Cd
probe atoms in the ordered intermetallic compound NiAl. NiAl has
the CsCl
crystal structure, with interpenetrating simple cubic lattices
of Ni and Al atoms. The sample was almost perfectly
stoichiometric, with the In probe sitting on the aluminum sublattice,
surrounded
in the perfect crystal by eight Ni atoms in the first shell, six Al
atoms
in the second shell, and so forth. The spectrum displays a
superposition of signals corresponding to two different environments of
the probes:
- Defect-free sites having cubic point symmetry and, as a
consequence,
zero precession frequency.
The vertical offset of 0.5 units indicates that about 50% of the
probe
atoms are at this site.
- Sites having one Ni-vacancy in the first-neighbor shell exhibit
the periodic, 50-nanosecond period. The amplitude indicates
that about
40%
of probes are in that environment. A schematic 2D diagram showing
a
probe next to a vacancy is here.
The reasons for the large fraction of probe atoms having a vacancy
neighbor
is that some Ni-vacancies are quenched into the structure and there is
significant binding between oversized, indium impurities
and
Ni-vacancies.
The binding enthalpy has been determined
in other experiments to be about 0.20
eV.
General areas of application of PAC
PAC is widely applied in
condensed-matter and materials
physics studies. Applications include the following:
- phase transformations; phase analysis
- structural and magnetic phase transitions
- static and dynamic critical behavior and
exponents
- magnetism: spin dynamics, stability of atomic magnetic moments in
different
hosts, exchange interactions
- diffusion and other atomic movement in solids, dynamical
interactions detected via nuclear relaxation
- surfaces, interfaces and grain boundaries, thin films,
nanocrystals
- point defects: identification, thermodynamic properties,
interactions,
- point defect production by plastic deformation, radiation and
implantation
damage, and quenching
- lattice location of impurities in compounds
For additional applications, consult PAC
worldwide
links. PAC has favorable characteristics as a
spectroscopy.
Measurements are unrestricted in temperature, unlike in NMR and
Mössbauer effect for which signal amplitudes are reduced at high
temperature by Boltzmann factors. Concentrations of probe
atoms are typically as low as ~10 parts-per-billion in 111In/Cd
PAC experiments.
On the
downside, PAC has a limited number of isotopes
with favorable nuclear properties for "table-top"
experiments, the most common ones being 111In/Cd
and 181Hf/Ta, followed by 99Mo/Tc,
99Rh/Ru, and 100Pd/Rh. Additional probes
having parent isotopes with short lives are available at major
facilities that produce radioactive ion beams, such
as ISOLDE at CERN, ISAC
at TRIUMF. and, perhaps in the future, the rare-ion beam facility at NSCL
in Michigan. We are part of a research group that was
awarded beam time at ISOLDE for diffusion studies in February
2009; the spokesperson for the group is Manfred Deicher of the
University of the Saarlandes. Experiments are planned using the
isotope 117Cd, which decays into 117In.
Applications at WSU
1. Point Defects in Solids
Point defects nearby probe nuclei modify internal fields. All of
the following types of defects have been detected in this laboratory
over
the past two decades:
- lattice vacancies, forming complexes with probe atoms
- equilibrium defect combinations in compounds such as the triple
defect in NiAl, comprised of one NiAl antisite
atom and two VNi vacancies.
- substitutional impurities and interstitial impurities such as
hydrogen next to probe atoms
- antisite atoms in compounds
To illustrate a study of point defects, consider the study of vacancy-probe complexes in FeAl.
in which bound states of indium impurities with from one to four
near-neighbor vacancies were observed to form and dissolve reversibly
by "condensation" and "evaporation" as the temperature was lowered and
increased. The very first PAC study
of lattice defects was made in the early 1960's by
George Hinman, who, by coincidence, is currently Emeritus Professor of
Environmental
Science at WSU. Click on the following link for a
well-written
brief
history
of PAC by Frits Pleiter of the Groningen hyperfine group.
2.
Site Preferences of Impurities in Compounds
When a compound contains sites havingl different symmetries, one
can
identify the site occupied by an impurity PAC probe atom by measuring
the quadrupole
interaction. A particularly simple and common structure having
this feature is the Cu3Au,
or L12, crystal structure.
Atoms at corners of the unit cell have cubic point symmetry, which
makes the quadrupole interaction frequency zero. Atoms
at face-centers of the cubic unit cell have tetragonal symmetry,
leading to a
non-zero quadrupole interaction. Therefore, simple
observation of a zero or non-zero quadrupole interaction serves to
identify the site occupied by the impurity.
We did a large study of site
preferences of indium solutes in GdAl2, which has the Cu2Mg , or
C15 structure and was studied in exactly the same way. Green
atoms in the pictured structure
are at cubic sites while brown atoms are not. We found that
indium solutes can appreciably occupy both sites in thermal
equilibrium and that they move from Gd sites (green) to Al sites
(brown)
as temperature increases. The enthalpy to transfer an indium
solute from the Gd-site to Al-site was measured to be 0.347 eV.
3.
Jump Frequencies of PAC Tracer Atoms in Solids
If tracer atoms jump on a sublattice whose sites have non-cubic
symmetry and if the axis of symmetry reorients in each jump, then the
quadrupole interaction will lose coherence over time. Roughly
speaking,
quadrupole moments precess around the principal axis of the local EFG
tensor., so that reorientation of the EFG leads to signal
decoherence. In a slow fluctuation regime, in which the jump
frequency is less than the quadrupole interaction frequency, the
decoherence is exhibited in good approximation as exponential
damping of the quadrupole perturbation pattern. In many
situations, the damping time is equal to the mean residence time of the
probe on a site, and can be fitted very accurately to obtain the jump
frequency. Amazingly, this simple approach to the study of
diffusion in solids went unnoticed in the PAC community for fifty years
before
we happened upon it. Our first paper, in Physical
Review Letters, involved PAC probe atoms jumping on the
In-sublattice in the compound In3La,
which has the Cu3Au,
or L12, crystal structure. A great deal of additional research has been carried out and is in progress.
Recently, we showed, also in Physical
Review Letters, that one can gain insight into diffusion
mechanisms in highly ordered binary compounds by comparing jump
frequencies measured at opposing boundary compositions. In
most solids, atom movement is made possible by the presence of
vacancies into which neighboring atoms can jump. Consider
for simplicity two samples of a compound A3B, one prepared
to be A-rich and the other B-rich. It turns out that the
concentration of A-vacancies must increase as the composition becomes
more B-rich and, similarly, the B-vacancy concentration increases
as the composition becomes more A-rich. Since jump
frequencies are proportional to the number of vacancies available, it
is a simple matter to determine which type of vacancy is involved in
the predominant difffusion mechanism. For example, if the
principal diffusion mechanism involves B-vacancies, then a greater jump
frequency will be observed in the A-rich sample. As a
case in point, we presented measurements on an entire series of In3R phases having the L12 structure,
in which R is a rare-earth element, that show that diffusion on the
A-sublattice is dominated by A-vacancies in the "heavy" lanthanide
indides (e.g., Lu, Tm, Er,...) but by B-vacancies in the "light"
lanthanide indides (e.g., La, Ce, Pr). This change in the
dominant mechanism along the series is very surprising given the great
chemical similarity of rare-earch elements.
References for PAC spectroscopy
Other web descriptions and graphics picturing PAC spectroscopy
PAC theses and monographs on the web
(please notify me of others)
- Atomic
Slide Puzzle, Raoul van Gastel, thesis,
Leiden, 2001.
- Defect
Chalcopyrites, M. Dietrich, dissertation, Freiberg, 2001.
- Phase Transition in Zircon, Matthew P. Rambo, dissertation, Miami U., Ohio, 2005
- Synthetic Zircon and Hafnon, Sean McBride, thesis, Miami U., Ohio, 2005
PACMAN in the universe
May 2009, GSC. Back to methods.
Back to our Collins group home page..
Please send suggestions and corrections to collins-at-wsu.edu.