No abstract

Angular-dependent magnetic Barkhausen noise (MBN) measurements were performed on a pipeline steel sample for various values of applied uniaxial stress at three angles with respect to the sample’s zero stress magnetic easy axis direction. It was observed that the response of the MBN signal to stress was dependent upon the direction of the stress with respect to the zero stress easy axis. The stress response of the MBN signal was greatest for (i) tensile stresses oriented perpendicular to the zero stress easy axis direction and (ii) compressive stresses applied parallel to the easy axis direction. The modification of the MBN signal under an applied stress was attributed primarily to a change in the 180° domain wall population in the material investigated. Results were described by a model that considered regions of locally correlated domain behavior, termed ‘‘interaction regions,’’ that were typically the size of grains within the steel material. A basic result of the model was the stress required to modify the number of 180° domain walls within an interaction region. Theoretical calculations of these threshold stresses for a typical grain size were found to be in agreement with the range of applied stresses that was observed to modify the angular-dependent MBN signal obtained from the sample.

The metal-binding affinity of the anionic poly-y-D-glutamyl capsule of Bacillus licheniformis was investigated by using Na+, Mg2+, Al3+, Ca2+, Cr3+, Mn2+, Fe3+, Ni2+, and Cu2+. Purified capsule was suspended in various concentrations of the chloride salts of the various metals, and after dialysis the bound metals were analyzed either by graphite furnace atomic absorption spectroscopy or by inductively coupled plasma-mass spectrometry. Exposure of purified capsule to excess concentrations of Na+ revealed it to contain 8.2 ,umol of anionic sites per mg on the basis of Na binding. This was confirmed by titration of the capsule with HCI and NaOH. Other metal ions were then added in ionic concentrations equivalent to 25, 50, 75, 100, 200, and 400% of the available anionic sites. The binding characteristics varied with the metal being investigated. Addition of Cu2+, Al3+, Cr3+, or Fe3+ induced flocculation. These metal ions showed the greatest affinity for B. licheniformis capsule in competitive-binding experiments. Flocculation was not seen with the addition of other metal ions. With the exception of Ni2+ and Fe3+ all capsule-metal-binding sites readily saturated. Ni2+ had low affinity for the polymer, and its binding was increased at high metal concentrations. Fe3+ binding resulted in the development of rust-colored ferrihydrite which itself could bind additional metal. Metal-binding characteristics of B. licheniformis capsule appear to be influenced by the chemical and physical properties of both the capsule and the metal ions.

The angular dependence of magnetic Barkhausen noise (MBN) on eight surfaces through the thickness of a 2% Mn steel pipeline sample was investigated. The MBN signal was analyzed by integrating the square of the MBN voltage signal with respect to the time axis. The resulting value, referred to as the MBN energy signal, was modeled by considering the irreversible motion of 180° domain walls, under the influence of an oriented magnetic field. An expression for the angular dependence of the MBN energy signal was derived and was given by energy=α cos2 θ+β, where α and β are the fitting parameters and θ is the angle between the maximum MBN signal and the applied sweep field. The α parameter was associated with the irreversible motion of 180° walls that contributed to the net macroscopic easy axis near the surface of the sample, while the β parameter was associated with the isotropic background MBN signal. The energy equation could be used to fit the data for all sweep field amplitudes in which the MBN spectrum was observed. A dependence of the α and β parameters on the applied sweep field amplitude was observed.

The consistency of magnetic Barkhausen noise (MBN) measurements under applied sinusoidal magnetic field control and sinusoidal magnetic circuit flux control was investigated under variable circuit permeability conditions. A U-core electromagnet was used to provide the alternating magnetic excitation. The magnetic circuit permeability was changed by varying excitation magnet lift-off and by using samples with known magnetic anisotropy. By controlling the circuit magnetic flux, measured as the flux in one of the U-core poles near the sample, MBN measurements were found to be consistent and independent of the excitation magnet lift-off in both a Si-Fe steel sample and an interstitial free (IF) steel sample at peak sample flux densities greater than 1.16 T and 0.29 T respectively. Consistency within a 95% confidence level was demonstrated for lift-off values of 0.6 mm or less, with decreasing sensitivity to lift-off observed at higher fluxes. MBN anisotropy measurements were also performed using both field control and flux control. Under field control conditions a component of the anisotropy signal was found to be dependent on the magnetic circuit permeability. This permeability dependence was absent when using flux control. The results demonstrated that flux control should be used when performing MBN measurements on samples where lift-off may be an issue, as values obtained will have less dependence on the excitation magnet characteristics than when field control is used.

No abstract

An analytical model was developed to determine the length of a surface-breaking defect along the direction of the applied field when using the magnetic flux leakage (MFL) non-destructive technique. The theoretical model fits the experimental MFL results from simulated defects. The extreme positions of the normal magnetic leakage field component were subsequently used for a quantitative evaluation of the defect length. Permeability variations were neglected by employing a flux density close to sample saturation. Four different defect geometries were experimentally investigated and the validity of the analytical model was verified. Good agreement between theoretical and experimental results suggests that this method can be used as an inverse MFL data interpretation technique.

No abstract