M. P. Schulze

Pressure dependencies of magnetostrictive strain and d coefficient in Terfenol‐D after thermal or magnetic annealing

The pressure dependence of the magnetostrictive strain coefficient, d33, and maximum strains in applied fields of 120 kA m−1 have been measured in samples of Terfenol‐D before and after thermal or magnetic annealing. Application of an annealing field, Ha, parallel to the 〈111〉 axes which are normal to grains oriented along the [112] axis, leads to an increase in d33 of as much as 81%, with applied uniaxial prestresses as low as 3 MPa. The variation of optimum d33 values as a function of stress applied along the [112] axis shows anomalous oscillations after magnetic annealing. The thermal and magnetic effects of the annealing procedures are discussed and a mechanism to explain magnetic annealing is proposed.

The effects of magnetic annealing and compressive stress on the magnetic properties of the rare earth-iron compound Terfenol-D

Abstract The rare earth-iron compound Terfenol-D (Tb x Dy 1− x Fe y with x ≈ 0.3 and y ≈ 1.92) is known to be highly magnetostrictive (1500–2000 ppm) and to possess a magnetomechanical coupling coefficient of around 0.7, properties which are affected by the application of compressive stress. Measurements have been made of the variation in the magnetostriction (λ), induction ( B ), magnetostrain coefficient ( d 33 ), permeability (μ σ ) and magnetomechanical coupling ( k 33 ) as a function of applied field and uniaxial compressive stress in the range 0 to 20 MPa and remeasured after being subjected to a magnetic heat treatment. After magnetic annealing, increases in the magnetostriction in the absence of an applied pre-stress are evident. However, under the conditions used in these annealing experiments, this improvement in the magnetostriction is at the expense of a reduced μ σ and d 33 .

Enhanced differential magnetostrictive response in annealed Terfenol-D

The field and pressure dependencies of the magnetostriction of Tb0.316Dy0.684Fe1.982 have been measured in a grain‐oriented rod after thermally annealing for 1 day at 850 °C and for 4 days at 950 °C in an argon atmosphere. The results of the heat treatment are a fivefold increase in the strain coefficient d33(=dλ/dH) and a 100% increase in the maximum strain (λ). There was also an increase in the λ‐vs‐H hysteresis. Under compressive uniaxial stress there was virtually no bulk change in magnetostrictive strain until the field exceeded a critical value which depended on the applied stress, for instance ∼12 kA/m under a stress of 6 MPa.

The stress dependence of k/sub 33/, d/sub 33/, lambda and mu in Tb/sub 0.3/Dy/sub 0.7/Fe/sub 1.95/

DC field measurements of lambda , d/sub 33/, k/sub 33/, and mu under uniaxial prestresses up to approximately 20 MPa have made on a 152-mm-long 6-mm-diameter grain-oriented rod of Tb/sub 0.3/Dy/sub 07/Fe/sub 1.95/ prepared by a free float zone technique. The lambda -vs.-H curve shows a negative dip at low fields which persist to approximately 20 MPa. The maximum negative strains approximately -120 ppm occur under approximately 2 MPa, which is also the pressure at which optimum permeability and magnetomechanical coupling occur. The origin of the negative strains is discussed in the light of earlier predictions based on an anisotropic rotational model for magnetization and magnetostriction, and the practical consequences are examined. >

Characterisation of magnetostrictive materials with computer modelling

A plane wave model (PWM) which enables magnetostriction parameters to be determined is fitted to simulated and experimental impedance curves, including an analysis of the inaccuracies due to random and systematic error. The PWM method is applied to a free standing rod of Terfenol-D. The magnetomechanical coupling coefficient, k/sub 33/, is found to have a maximum value of 0.509+or-0.002 at 25 kAM/sup -1/. In comparison a resonance technique which only employs values for the resonance (f/sub R/) and antiresonance (f/sub A/) frequencies, with a correction for the skin effect, yields values for k/sub 33/ which are 15% larger. The difference is attributable to frequency dependent losses which influence not only f/sub R/ and f/sub A/ but the shape of the whole resonance curve. >

The characterisation of transducer materials

Abstract Efficient transducers rely on the incorporation of high quality magnetostrictive or piezoelectric materials. However their behaviours, if they are to be quantified, have to be represented by specific measurable parameters. Preferably these should be identified with particular physical properties, for instance defect density, such that device performance and lifetimes can be gauged. To this end, different approaches to modelling material or device response are briefly reviewed and their benefits and limitations discussed. As part of a characterisation process the method of plane wave modelling (PWM) is used to predict the resonant response of a transducer element and through a comparison of the predicted and observed behaviour, values for six characterisation parameters are obtained. These data are then used to determine a reliable value for the magnetomechanical coupling coefficient that is representative of material quality regardless of the sample size and geometry or frequency of operation.

Nondestructive evaluation of Terfenol-D for transducer applications

A method and techniques have been developed for the nondestructive evaluation of grain oriented highly magnetostrictive rare-earth compounds, such as Terfenol-D, for use in actuator construction. Detection of an asymmetric flux distribution and localized fluctuations in ac permeability have been correlated with gradients in chemical composition and departures from the ideal grain structure produced by free float or contained zone preparation techniques. The data provide valuable information to gauge material quality, the response of individual samples to applied stress, and the reproducibility of transducer properties from one sample to another.

Non-destructive testing of Terfenol-D

Abstract Chemical composition and grain structure have a direct effect on the magnetostrictive behaviour of the compound Terfenol-D. The rare-earth (RE):iron and Dy:Tb ratios affect not only the magnetostriction but also the magnetocrystalline anisotropy and its temperature dependence. Grain structure has a critical effect on the strain coefficient, d 33 (dλ/d H ) and its dependence on applied uniaxial prestress as a function of applied field. For transducer applications these factors are significant because gradients in composition and grain misorientation, leading to a reduction in strain coefficient, will cause inefficiencies. In both cases, the full potential of Terfenol-D can only be realised by applying larger fields. However, this in turn leads to higher induction losses in the Terfenol. The present paper describes a method and non-destructive techniques for a qualitative investigation into compositional and grain structure variations along the length of a magnetostrictive rod. The results of scans along different samples show how gradients in chemical composition and grain misorientations can be identified without recourse to sectioning samples. The application of uniaxial stress and DC bias fields along the length of a rod enable the scanning response to optimised. Features of the scan can then be correlated with the main defects observed by sectioning the sample.

Variations in strain amplitude and phase ina a cylindrical specimen of Tb/sub 0.3/Dy/sub 0.7/Fe/sub 1.95/ in an AC magnetic field

In the light of differences between magnetostrictive strains measured on the surface of Terfenol with strain gauges and from a dilatometer, an optical interferometer has been used to probe the spatial variations of the magnetostrictive strains across the plane surface of a polished disc (6 mm diameter, 1 mm thick) of grain-orientated Tb/sub 0.27/Dy/sub 0.73/Fe/sub 1.95/ DC and AC fields ( approximately 15 kA/m and approximately 440 A/m RMS, respectively) were applied normal to the disk surface parallel to the