Search

The penetration depth λ in the filled-skutterudite heavy-fermion superconductor PrOs4Sb12 has been measured using transverse-field muon spin rotation (TF- μ SR ). It is found to be temperature-independent at low temperatures, consistent with a nonzero gap for quasiparticle excitations. In contrast, radiofrequency (RF) measurements yield a stronger temperature dependence of λ ( T ) , indicative of point nodes in the gap. A ∼ 10 % discrepancy is found at low temperatures. This may be due to mechanisms that modify the vortex-state field distribution, or to the surface scattering which breaks pairs in an odd-parity superconductor. Alternatively, it may be a matter of field orientation of nodal gap structure in the μ SR measurements.

We present the first μ SR measurements ever performed on elemental Pu, and set the most stringent upper limits to date on the magnitude of the ordered moment μ ord in α -Pu and δ -stabilized Pu (alloyed with 4.3 at% Ga). Assuming a nominal hyperfine coupling field of 1 kOe/ μ B we find μ ord ⩽ 10 - 3 μ B at T ≅ 4 K .

We present transverse-field (TF) measurements of the in-plane magnetic-field penetration depth λ ( T ) in single-crystalline PuCoGa 5 ( T c = 18.5 K ) for 0.06 T applied field ( ≈ 2 – 5 × H c 1 ) . We find that the temperature dependence of the superfluid density, and therefore Δ λ ( T ) = λ ( T ) - λ ( 0 ) , is ∝ T for T / T c ⩽ 0.5 . We estimate that the measured λ ( 0 ) = 241 ( 3 ) nm . ZF measurements find no evidence for time-reversal symmetry violation. The ZF and TF measurements are consistent with an even-parity (pseudo-spin singlet), d-wave pairing state.

Recently it has been demonstrated that low energy spin polarized radioactive nuclei can be used to probe the electronic and magnetic properties of ultra-thin samples using β detected NMR. However, use of the technique requires an accurate understanding of ion penetration depth and range straggling as a function of beam implantation energy and target density. Range straggling measurements of 8 Li were conducted with thin films of metallic Al and Au on sapphire and SrTiO3 substrates and the results were compared to SRIM code predictions.

We report μ + -Knight shift measurements as a function of temperature in U 1− x Th x Be 13 , x =0 and 0.035, single crystals. An anomalous change in the hyperfine coupling at low temperatures is observed which is similar to that reported in nuclear magnetic resonance studies of U- and Ce-based heavy fermion compounds. In the superconducting phase the static spin susceptibility, χ spin , determined from the μ + -Knight shift decreases continuously with decreasing temperature in pure UBe 13 , but saturates below the second transition temperature in U 0.965 Th 0.035 Be 13 . We briefly discuss these behaviors in terms of a two-component spin-triplet order parameter and propose μSR experiments that can test our hypothesis.

We have commissioned a polarized low-energy 8 Li ion beam line, which together with a high-field β-NMR spectrometer, can act as sensitive new probe of thin films and interfaces. The implantation energy can be continuously adjusted from 1 to 90 keV and the maximum polarization achieved thus far is 80%. This instrument opens up new applications for β-NMR which parallel and complement efforts with low-energy muons. For example, it is possible to probe the magnetic field distribution near the surface of a material by stopping a polarized 8 Li beam in a thin overlayer of Ag. Since the 8 Li adopts a site with cubic symmetry in Ag there is no quadrupolar splitting of the resonance, and the 8 Li acts as a purely magnetic sensor.

The effective superconducting penetration depth measured in the vortex state of PrOs 4 Sb 12 using transverse-field muon spin rotation ( TF - μ SR ) exhibits an activated temperature dependence at low temperatures, consistent with a nonzero gap for quasiparticle excitations. In contrast, Meissner-state radiofrequency (rf) inductive measurements of the penetration depth yield a T 2 temperature dependence, suggestive of point nodes in the gap. A scenario based on the recent discovery of extreme two-band superconductivity in PrOs 4 Sb 12 is proposed to resolve this difference. In this picture a large difference between large- and small-gap coherence lengths renders the field distribution in the vortex state controlled mainly by supercurrents from a fully gapped large-gap band. In zero field all bands contribute, yielding a stronger temperature dependence to the rf inductive measurements.

In some zero-field μSR experiments the outcome depends on reducing stray external magnetic fields to very low levels. These fields may originate from the Earth, nearby accelerators, and residual fields from spectrometer magnets. We describe a method of determining appropriate trim coil currents to zero the field, which avoids the observation of very low unwanted muon precession frequencies. This method depends on measuring the muonium precession frequencies in a set of x-, y- and z-axis weak transverse field scans, and then calculating the necessary currents to achieve practically zero net field.

We report a study of muon Knight shifts to investigate the formation of the heavy-fermion state in single crystals of ${({\text{Ce}}_{1\ensuremath{-}x}{\text{La}}_{x})}_{2}{\text{IrIn}}_{8}$. Two different kinds of Knight-shift anomalies (deviations from a linear relation between the Knight shift and the susceptibility) are found: (1) a high-temperature effect arising from depopulation of crystalline electric field levels with temperature, and (2) a lower-temperature anomaly arising from the onset of the heavy-fermion state below a characteristic temperature ${T}^{\ensuremath{\ast}}$, in agreement with the ``two-fluid'' model of heavy-fermion formation. In ${\text{Ce}}_{2}{\text{IrIn}}_{8}$, we find ${T}_{c}^{\ensuremath{\ast}}=20.0(6)\text{ }\text{K}$ and ${T}_{a}^{\ensuremath{\ast}}=15.2(1.2)\text{ }\text{K}$ for applied field $H\ensuremath{\parallel}c$ axis and $H\ensuremath{\parallel}a$ axis, respectively. For the Ce diluted systems ${({\text{Ce}}_{1\ensuremath{-}x}{\text{La}}_{x})}_{2}{\text{IrIn}}_{8}$, $x=0.1$, 0.25, 0.50, 0.70, and 0.90, ${T}^{\ensuremath{\ast}}$ decreases linearly for $x\ensuremath{\le}0.5$, reaching zero near $x=0.7$, indicating the reduction in intersite $f$-spin correlations with Ce dilution. A comparison with nuclear magnetic-resonance measurements of Knight-shift anomalies in several other heavy-fermion compounds suggests that the observed small anisotropy in ${T}^{\ensuremath{\ast}}$ may be induced by the applied field, and that ${T}^{\ensuremath{\ast}}$ may be inherently isotropic, even in highly anisotropic materials.

We have measured the temperature dependence and magnitude of the superfluid density ${\ensuremath{\rho}}_{s}(T)$ via the magnetic field penetration depth $\ensuremath{\lambda}(T)$ in $\mathrm{Pu}\mathrm{Co}{\mathrm{Ga}}_{5}$ (nominal critical temperature ${T}_{c0}=18.5\phantom{\rule{0.3em}{0ex}}\mathrm{K}$) using the muon spin rotation technique in order to investigate the symmetry of the order parameter, and to study the effects of aging on the superconducting properties of a radioactive material. The same single crystals were measured after $25\phantom{\rule{0.3em}{0ex}}\text{days}$ $({T}_{c}=18.25\phantom{\rule{0.3em}{0ex}}\mathrm{K})$ and $400\phantom{\rule{0.3em}{0ex}}\text{days}$ $({T}_{c}=15.0\phantom{\rule{0.3em}{0ex}}\mathrm{K})$ of aging at room temperature. Penetration depths $\ensuremath{\lambda}(0)=265(5)$ and $\ensuremath{\geqslant}498(10)\phantom{\rule{0.3em}{0ex}}\mathrm{nm}$ are derived for the fresh and aged samples, respectively. The temperature dependence of the superfluid density ${\ensuremath{\rho}}_{s}(T)∕{\ensuremath{\rho}}_{s}(0)=\ensuremath{\lambda}{(0)}^{2}∕\ensuremath{\lambda}{(T)}^{2}$ is well described in both materials by a model using $d$-wave gap symmetry. Within the context of a strong-coupling dirty $d$-wave model a zero-temperature gap value ${\ensuremath{\Delta}}_{0}=3.0{k}_{B}{T}_{c0}$ is obtained in the fresh sample for a scattering rate $\ensuremath{\Gamma}=0.005\ensuremath{\pi}{k}_{B}{T}_{c0}$, which is consistent with Abrikosov-Gor'kov (AG) pair-breaking theory. This ${\ensuremath{\Delta}}_{0}$ should be compared to the weak-coupling value for a clean $d$-wave gap ${\ensuremath{\Delta}}_{0}=2.14{k}_{B}{T}_{c0}$. In the aged sample the same model yields ${\ensuremath{\Delta}}_{0}=2.4{k}_{B}{T}_{c0}$ for $\ensuremath{\Gamma}=0.010\ensuremath{\pi}{k}_{B}{T}_{c0}$. This value of $\ensuremath{\Gamma}$ is much less than required by the AG pair-breaking formalism. Furthermore, the aged ${\ensuremath{\rho}}_{s}(0)$ is reduced by at least 70% compared to the fresh sample, which is also incompatible with $\ensuremath{\Delta}{T}_{c}∕{T}_{c0}\ensuremath{\sim}20%$, according to AG theory. We conclude that the data in aged $\mathrm{Pu}\mathrm{Co}{\mathrm{Ga}}_{5}$ support the postulate that the scattering from radiation-induced defects is not in the limit of the AG theory of an order parameter which is spatially averaged over impurity sites, but rather in the limit of short-coherence-length superconductivity. We show that a model calculation consistent with this assumption fits ${\ensuremath{\rho}}_{s}(T)$ in the aged sample rather well.

We present $\ensuremath{\mu}\mathrm{SR}$ measurements on Pu metal which set stringent upper limits on the magnitude of the ordered moments ${\ensuremath{\mu}}_{\mathrm{ord}}$ in $\ensuremath{\alpha}\text{\ensuremath{-}}\mathrm{Pu}$ and $\ensuremath{\delta}$-stabilized Pu (alloyed with $4.3\phantom{\rule{0.3em}{0ex}}\mathrm{at}.%$ Ga). The magnitude of the low-temperature $(4\phantom{\rule{0.3em}{0ex}}\mathrm{K}lTl100--150\phantom{\rule{0.3em}{0ex}}\mathrm{K})$ $\ensuremath{\mu}\mathrm{SR}$ rate in zero applied field is independent of temperature and consistent with nuclear dipolar broadening alone. The Knight shift in $\ensuremath{\alpha}\text{\ensuremath{-}}\mathrm{Pu}$ in $2\phantom{\rule{0.3em}{0ex}}\mathrm{T}$ applied field is also independent of temperature below about $100\phantom{\rule{0.3em}{0ex}}\mathrm{K}$, but increases abruptly at higher temperatures, an effect likely caused by muon diffusion to paramagnetic impurities. A rough estimate for the muon hopping rate $1∕{\ensuremath{\tau}}_{h}$ at $150\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ in $\ensuremath{\alpha}\text{\ensuremath{-}}\mathrm{Pu}$ yields $1∕{\ensuremath{\tau}}_{h}\ensuremath{\approx}3\ifmmode\times\else\texttimes\fi{}{10}^{7}\phantom{\rule{0.3em}{0ex}}{\mathrm{s}}^{\ensuremath{-}1}$; no estimate is yet obtained for $\ensuremath{\delta}\text{\ensuremath{-}}\mathrm{Pu}$. The principal results reported here are limits for ${\ensuremath{\mu}}_{\mathrm{ord}}$ in both $\ensuremath{\alpha}$- and $\ensuremath{\delta}\text{\ensuremath{-}}\mathrm{Pu}\ensuremath{\leqslant}{10}^{\ensuremath{-}3}{\ensuremath{\mu}}_{\mathrm{B}}$ at $\mathrm{T}\ensuremath{\cong}4\phantom{\rule{0.3em}{0ex}}\mathrm{K}$. These limits are discussed in terms of recent models of the electronic properties of $\ensuremath{\delta}\text{\ensuremath{-}}\mathrm{Pu}$, where magnetism has been previously predicted. Our results are consistent with, but cannot unambiguously distinguish between, cases where the moments in $\ensuremath{\delta}\text{\ensuremath{-}}\mathrm{Pu}$ are essentially zero, possess very weak interatomic exchange, are reduced by hybridization, or are washed out by dynamical fluctuations.

Longitudinal-field muon spin relaxation experiments have been carried out in the paramagnetic state of single-crystal Gd_2Ti_2O_7 just above the phase transition at T_m = 1.0 K. At high applied fields the exponential relaxation time T_1 is proportional to field, whereas T_1 saturates below a crossover field B_c that is ~2.5 T at 1.5 K and decreases as T_m is approached. At low fields the relaxation rate increases markedly as the freezing temperature is approached, as expected for critical slowing down of the spin fluctuations, but the increase is suppressed by applied field. This behavior is consistent with the very long autocorrelation function cutoff time implied by the low value of B_c., 4 pages, 2 figures, submitted to 10th International Conference on Muon Spin Rotation, Relaxation, and Resonance, Oxford, UK, August 2005

We report muon spin rotation Knight shift and susceptibility studies for 1 T applied field along the crystalline c- and a-axes of the heavy fermion compound Ce2IrIn8. Below a characteristic temperature T* one observes a `Knight-shift anomaly' in which the Knight shift constant K no longer scales linearly with susceptibility chi. This anomaly is consistent with a scaling law in which chi is composed of a high-temperature component corresponding to non-interacting local moments and a low-temperature component chi_cf proportional to (1-T/T*)\ln(T*/T) which characterizes the heavy-electron state below T*. We find that T* is anisotropic, with T_a* = 59(3)K and T_c* = 24(1)K, and derive the magnitudes of chi_cf for H along the a- and c-axes., 4 pages, 5 figures, Submitted to 10th International Conference on Muon Spin Rotation, Relaxation and Resonance, Oxford, UK, August 2005

A new facililty for conducting beta-detected nuclear magnetic resonance (beta-NMR) investigations of condensed matter systems has recently been constucted at TriUniversity Meson Facility in Vancouver, Canada. The unique features of this new facility are described, and some preliminary results on Li-8(+) in GaAs are presented. (C) 2003 Elsevier B.V. All rights reserved.

Zero-field (ZF) μSR measurements were performed on single crystals of the high temperature superconductor YBa 2 Cu 3 O 6+ x , for x ⩾0.67. Contrary to early ZF-μSR studies, we find that the muon-spin relaxation function evolves significantly as a function of temperature and oxygen content. The relaxation rate of the ZF-μSR time spectra exhibits a local minimum near 55 K and a subsequent increase with decreasing temperature. These features weaken considerably with decreasing x . A comparison to the results of nuclear quadrupole resonance experiments on highly doped samples suggests that these observations are associated with the build up of long-range charge correlations and not spontaneous orbital moments as suggested by Sonier et al. (Science 292 (2001) 1692).

We have performed muon spin rotation ( μ SR ) measurements in the radioactive Pu-based superconductor PuCoGa 5 to study the effects of radiation damage on the superconducting properties. The same single crystals were studied 25 days (fresh) and 400 days (aged) after initial preparation. We find that the μ SR rate σ v , which is proportional to superfluid density ρ s , is steeply reduced in the aged sample compared with its reduction of T c . This result is inconsistent with conventional Abrikosov–Gor’kov pair-breaking theory, but can be explained by the local suppression of the order parameter in a strong-scattering, short-coherence-length superconductor.

Zero-field muon spin relaxation (μSR)(μSR) in non-Fermi liquid Ce0.2La0.8RhIn5 reveals disordered spin freezing with a sharp transition at 0.16 K. The frozen moment at T=0T=0 is estimated as ∼0.13μB/Ce, an order of magnitude smaller than found from thermodynamic data at higher fields and temperatures. Longitudinal-field μSRμSR exhibits no dynamic relaxation at any temperature, showing that the spin freezing is not accompanied by the slow spin fluctuations found in spin glasses above the glass temperature and in other disordered non-Fermi liquid systems.

The local magnetic susceptibility wloc around a positive muon in quasi-one-dimensional spin 1/2 antiferromagnetic chain compounds dichlorobis KCuF3 has been investigated using muon spin rotation/relaxation muSR: We compare chi_loc with the bulk magnetic susceptibility chi measured in a SQUID magnetometer. In all cases there is a dramatic difference between chi_loc and chi; as predicted by theory.