The design of a system of adjustable superconducting plates for balancing a gradiometer J.A. Overweg and M.J. Walter-PetersIn a SQUID-magnetometer system external magnetic fields are coupled into the SQUID by a fluxtransformerJ If the magnetic fields to be measured are much weaker than other low gradient ambient fields, eg in our case of magnetocardiography in an unshielded environment, this fluxtransformer is usually a first or second order gradiometer 2 as sketched in Fig. 1. When the gradiometer is ideal only inhomogeneous fields such as that of the human heart induce a current in the fluxtransformer and therefore give a flux in the SQUID. The total magnetic flux through all loops due to a homogeneous field should be zero, so that these fields are not measured.However any real gradiometer will always be more or less sensitive to uniform fields too. This imbalance can be caused by imperfect alignment and non-identical areas of the pickup loops. Magnetic fields can be picked up by the wires that connect the loops. Even a perfectly balanced gradiometer will respond to a homogeneous field when this is distorted by a superconducting body or by eddy-currents in normal metals, for example of the insulation of the cryostat.There are various ways to reduce the sensitivity of a gradiometer to uniform magnetic fields. One of these methods is to change the effective area of the gradiometer loops by small adjustable superconducting plates, a This paper deals with the design of such a balancing system. Improving gradiometer balance with superconducting platesBecause magnetic fields are expelled from superconducting bodies (the Meissner-effect) a superconducting plate will distort a magnetic field. There are two ways to use this effect for adjusting the effective area of a gradiometer loop:The authors are at

A new. sQc1D-based, measuring system has been developed for the investigation of the dynamic properties of magnetic materials. Its main advantages. compared to conventional mutual inductance systems, are its high sensitivity and its nearly frequency independent response. down to extremely low frequencies. With the SQUID system it is now possible to measure directly and rapidly the frequency dependence of the dynamic susceptibility of weakly magnetic materials in the range from 0.002 Hz to 5 kHz at a fixed value of a constant background field of up to 5 T.The frequency-swept operation allows experiments which are beyond the capabilities of conventional systems. for example the investigation of thermally isolated samples or the study of the dynamics at magnetic phase transitions in cases where the magnetic properties change very rapidly in a small field interval.

A digital phase sensitive detector (PSD) has been designed and built which may be used in conjunction with a SQUID-based AC susceptibility measurement system. The system operates in a useful frequency range from 2 mHz-2 kHz and measures both the in-phase and the quadrature component. The digital PSD, which is based on a small minicomputer. corrects for DC drift and flux quantum jumps of the SQUID and operates in 'real time'. An analysis of its design and several properties and measuring results are presented.

Abstract.-The effective area of a circular wire loop to a uniform field is reduced considerably if it is placed close to a large superconducting disc. The effective area to the field of a nearby magnetic dipole is reduced much less. Such a selective pickup loop is used to construct a small first order gradiometer, with a baseline of 3 cm. Its sensitivity to fields of nearby sources is comparable to that of a much larger symmetric second order gradiometer wheras its sensitivity to inhomogeneous magnetic noise is much less. Another advantage of the unsymmetric configuration is, that it has a smaller selfinductance than a conventional gradiometer, so that a better flux transfer to the SQUID is possible.

Dynamic susceptibility curves observed as a function of frequency often deviate considerably from curves predicted by the simple Casimir-Du Pr6 model. In the case of cerium magnesium nitrate these deviations are mainly caused by the poor thermal conductivity of the liquid or gaseous helium that surrounds the sample and becomes extremely large when the external magnetic field becomes larger than approximately 1 T. We found that, even in this extreme situation, the influence of the environment can be accurately described with the thermal conduction model developed by Plokstra et al.We also discovered that the lattice temperature. is stabilized much more effectively if the sample is placed in helium gas at its saturation pressure. The thin helium film that is formed on the surface of the sample provides, through condensation and reevaporation, a far better thermal stabilization of the lattice than bulk liquid helium does. The susceptibility curves that are observed under these conditions differ much less from the Casimir-Du P& curves.