Magnetic Pebbles – Materials with Controllable Magnetism for Compact, Low-Power Shim Units
David Otto Brunner1, Simon Gross1, Jonas Reber1, and Klaas Paul Pruessmann1

1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland

Synopsis

B0 shimming with very high channel count encounters many implementation problems due to the size and current handling requirements of the shim units. Here we present an approach using distributions of ferromagnetic materials with controllable magnetic moments to generate shim fields. These units are small, require only low currents and can hence be implemented in large numbers into RF receive arrays.

Introduction

Susceptibility induced off-resonances challenge many cutting-edge applications using single shot read-outs, balanced acquisitions or high-resolution spectroscopy, in particular at ultra-high fields. B0 shimming with increased number of channels yields critical improvements [1], but significantly reduces free bore diameter. Coil conductors in close proximity to the subject [2] or even on the RF coils [3] reduce space and power requirements. However, unwanted interactions with RF and gradient operation was reported as a main issue and the handling is aggravated by the amount of conductors in the unit and the large number of high-current wires routed through the bore. Furthermore highly stabilized current supplies need to be fitted in the technical room. As an alternative we present the integration of ferromagnetic materials whose magnetization can be accurately controlled in-situ as opposed to traditional passive shims [4]. Thereby the secondary fields produced by this material is used to shim highly localized. The proposed geometric arrangement allows producing fields with both polarities.

Methods

The magnetic moment density (m) of a ferromagnetic particle in a strong external magnetic field can be controlled by its temperature ([5], Fig. 1a). At Curie temperature (TC) the ferromagnetism vanishes and renders the material paramagnetic. By heating its magnetic moment is gradually reduced by more than an order of magnitude, correspondingly the secondary fields nearly vanishes. Hence by controlling the material temperature a shim field can be tailored. To operate in a reasonable temperature range, Nickel and Copper were alloyed to shots in a Ni75Cu25 stoichiometry [6] resulting in a TC of 350°K (Fig. 1b). Each shot of ~(3mm)3 is restively heated, its temperature is measured by a Pt500 thermistor and it is thermally isolated and shielded (Fig 3ab). To produce fields of both polarities using materials with only positive susceptibilities, the magnetic material is arranged in a matrix such that its net secondary field is uniform when the magnetization of the heated particles is roughly halved (eg. Fig 2). In a first example, 35 particles were arranged on the axis of the main magnetic field by pressfitting in a wooden bar with 16mm distance (Fig 3c). 3 controllable magnetic particles (having 2m at low temperatures) were mounted in the center-slots, obtaining from each a net field outside the cylinder of approximately a dipole with a magnetic moment of –m to +m. The secondary field of the shim unit was measured by B0 mapping (Philips 3T Achieva, Best Netherlands) with 3 ms echo-spacing in a phantom bottle placed directly on top of the unit. 3 magnetic field probes (Skope MRT, Zurich, Switzerland) were each placed about 1 cm from each unit and 2 on top of the bottle.

Results

Fig. 4a&b show the net B0 fields induced by the 3 shim units temporally and spatially. Voltages from 0-10 V were applied in steps resulting in a temperature range from 293-420°K. The surface of the unit did warm up hardly noticeably. The slew rate was about 1 µT/s. Figure 4b shows field profiles in about 4 cm distance from the units. The ripples in the shim field next to the unit in the sagittal images result from the discretization of the magnetic moment distribution from a continuous cylinder into individual shots. They are expected to be drastically reduced once the NiCu is cast into a cylindrical geometry.

Discussion

Particles with controllable magnetism produce shim-field patterns with high spatial degrees of freedom. Since the source of the field is not an electric current but the magnetism of the material, smaller form factors and lower current consumptions are achieved and the particles are well decoupled from gradient and shim as well as from RF coils. Opposed to passive shims, rearranging the material is not required to fit subject specific susceptibility distributions. The heat required to control the units can be administered by DC and AC currents as well as optically tunable materials can be employed [7]. Furthermore the power delivery for the heating can be efficiently modulated by switched mode schemes such as by PWM current sources similarly as used for LED lightings where tens of channels can be housed in a single IC package. This allows placing the required electronics in the bore which dramatically reduces the involved cabling efforts. Very large numbers of independent shim channels can be integrated in RF coils with low additional weight and space requirements. Thereby high degrees of freedom can be obtained. Therefore the approach is expected to be well suited for shimming of susceptibility induced off-resonances i.e. in the prefrontal cortex, ear channels or the spine.

Acknowledgements

No acknowledgement found.

References

1) Pan JW et al. Role of very high order and degree B0 shimming for spectroscopic imaging of the human brain at 7 tesla., MRM 68(4) 2012, 2) Juchem C et al. Dynamic multi-coil technique (DYNAMITE) shimming for echo-planar imaging of the human brain at 7 Tesla, Neuroimage 105 2015 3) Stockmann JP, A 32-channel combined RF and B0 shim array for 3T brain imaging. MRM 2015 4) Hoult, D.I., Lee, D., 1985. Shimming a superconducting nuclear-magnetic-resonance imaging magnet with steel. Rev. Sci. Instrum. 56, 131–136. 5) C. Kittel, Introduction to Solid State Physics. 6) Ahern SA et al. The spontaneous magnetization of nickel+copper alloys, Proc. Royal Soc. of London A, 248(1253) 7) Munekata H, Photoinduced Magnetism in Semiconductor-Based Structures, J Sup.Cond 14(2) 2001.

Figures

Figure 1: a) Typical temperature dependence of the magnetic moment density (m) of a ferromagnetic particle in a strong external field. Towards the Curie temperature (TC) the strong ferromagnetic moment diminishes. Above TC the paramagnetism scales still down with temperature. b) TC of nickel alloys can be adjusted by the admixture of cupper. Thereby the temperature range in which a magnetic particle has to be controlled can be adjusted.

Figure 2: Spatial distribution of magnetic material. The distribution of the static material (m) and the controllable units (mc(T)) settled at half of their maximum moment provides zero magnetic field outside the cylindrical structure given it is much longer than the volume of interest. Lowering one unit’s temperature provides then a net field of a paramagnet and increasing the temperature that of a net strong diamagnet. The shim units can thereby adjust the fields with both signs (bipolarly).

Figure 3: Setup. The controllable units (a) were made of a (3 mm)3 Ni75Cu25 shot thermally coupled to SMD resistors. A Pt500 measures its temperature. 50 µm diameter wires are used for connecting. A wrapper of foam is applied for isolation and aluminum foil for RF shielding. b) Photograph of a unit. c) setup for the measurements with 3 controllable units. The static material distribution has been discretized into shots of half the volume than the units because no wire material was available.

Figure 4: a) Field as recorded by the field probes heating m1-m3 sequentially. b) B0 field maps in s1-s4 acquired as marked in a). Isolines at 0 ±15 ±30 ±50 ±100 ±200 Hz. The fields show a distinct dipolar pattern in the range of ±300Hz. Close range field ripples result from the discretization of the magnetization distribution into shots. Casting the material into cylinders should remove this problem. c) off-resonance line profiles 4cm distant from the shim units.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
0492