Ziming Huang1,2, Alex T. L. Leong1,2, and Ed X. Wu1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, China, 2Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China
Synopsis
Halbach permanent magnet array offers lightweight and
mobility but its field is sensitive to temperature, thus hindering its robust
point-of-care MRI applications. This work presents a temperature-insensitive
Halbach magnet array design for ultra-low-field MRI. The proposed array is of
adequate bore diameter and reasonable weight (~50kg), and provides a 30 mT B0 within
a 20 cm imaging field of view (FOV). Numerical simulation is conducted to test
the proposed design, demonstrating a drastic reduction in temperature-induced
B0 drifting and inhomogeneity.
Introduction
There is an increasing interest in ultra-low-field
(ULF) MRI for low-cost, low-power and point-of-care clinical applications [1-4].
The Halbach magnet array features low-cost, lightweight and easy-to-maintain advantages, making it an ideal choice for portable ULF magnets [1-3]. However, all rare earth magnet materials are sensitive to temperature. Temperature-induced B0 drifting and inhomogeneity
can easily arise during scanning due to environments or gradient coil heating.
One may use more temperature-insensitive materials for
Halbach array construction [5-7]. However, such rare earth materials are
difficult to access and expensive. Here we present a temperature-insensitive
Halbach array design, which can provide a stable B0 field in presence of global
or local temperature changes.Theory and Method
The field strength of
magnets generally behaves as a linear function of temperature, as determined by
a negative reversible temperature coefficient α (unit: ppm/ºC). The temperature compensation
approach [4] is adopted here. This
design interleaves two Halbach arrays that are constructed by two rare earth
magnet materials of very different temperature coefficients and with
opposite magnetization directions. This is to cancel out the
temperature-induced changes in B0 at a minor cost to magnet efficiency. In this
study, SmCo magnets (primary array) and NdFeB magnets
(compensation array) are chosen for their large difference in temperature coefficient
( αSmCo is -300 ppm/ºC and αNdFeB is -1400 ppm/ºC). Consequently, the two arrays’ ratio of B0 is determined based on their
material’s temperature coefficient, satisfying that B0SmCo/B0NdFeB=αNdFeB/αSmCo.
The proposed cylindrical
array consists of 23 rings. In each ring, 18 SmCo magnets and 18 NdFeB magnets
are alternately arranged (FIGURE 1). To improve homogeneity, genetic
algorithm (GA) is applied here to find optimal choices of magnet grades, sizes
of individual magnet elements, and their ring radii. SmCo and NdFeB arrays are
optimized by GA independently. Halbach array length is chosen to be ~50 cm. The
individual element volume is empirically chosen to be 10.4-11.5 cm3 and 2.19-2.92 cm3 and for SmCo and NdFeB elements, respectively. See
TABLE1 for other key design parameters.
At 26.85°C (300 K), the SmCo array generates a
B0 at ~38mT in the -y direction (see FIGURE 2a), and the NdFeB array
generates a B0 at ~8mT in the +y direction (see FIGURE 2b). B0 strength
of the final assembly is ~30 mT (see FIGURE 2c). The overall inhomogeneity
is ~1500 ppm. For comparison, a pure SmCo Halbach is designed by modifying the
SmCo array of the temperature-sensitive design. The only difference is that the size of each magnet element is scaled by a factor of 0.8, so this array
also generates a B0 field along the -y direction at about 30 mT. Its inhomogeneity
is ~1200 ppm. The weight of the proposed design is 49.4 kg (SmCo array 40.8 kg,
NdFeB array 8.6 kg). In comparison, the weight of pure SmCo Halbach array is 32 kg.
Finite element simulation using CST is conducted to study
the design performance with regard to spatially uniform and non-uniform
temperature changes. Three types of non-uniform temperature profiles (linear
distribution along x, y and z directions; spatially isotropic Gaussian
distribution; and exponential distribution along x direction) are simulated. Their
temperature distributions or maps are shown in FIGURE 3 (left) and
simulation details are given in TABLE 2. B0 changes are acquired by
comparing Halbach arrays in presence of these temperature distributions with
Halbach arrays at 26.85°C (uniform). Results
Numerical simulation
indicates that the overall B0 drifting due to global uniform temperature
changes is generally removed in the proposed design,
while small B0 drifting still can be observed when non-uniform temperature
changes occur. In cases of three non-uniform temperature change profiles, the
drifting of average B0 is suppressed to <50 ppm. The temperature-induced inhomogeneity
in the proposed design is ~10% of that in the pure SmCo Halbach array. The B0
changes caused by non-uniform temperature changes are shown in FIGURE 3
(right). They are quantified and summarized in TABLE 2. Discussion and Conclusions
This temperature-insensitive
Halbach array design features significantly improved stability to both spatially
uniform and non-uniform temperature changes, yielding a very stable B0 field under
various temperature change profiles. With such high immunity to temperature
changes, the proposed temperature-insensitive design is made suitable for air-conditioning-free
or/and outdoor scenarios. The configuration of the SmCo and NdFeB arrays
can be optimized. For example, a more uniform distribution of SmCo and NdFeB
magnets may further reduce the intrinsic B0 inhomogeneity (in absence of any
temperature changes) and temperature-induced B0 inhomogeneity. In this design,
SmCo and NdFeB magnets are chosen for Halbach array construction. Without loss
of generality, one can choose other materials to build temperature-insensitive
arrays in the same scheme. A further improvement of magnet efficiency is
possible when materials with higher temperature coefficient differences are
available. Acknowledgements
This
work was supported in part by Hong Kong Research Grant Council (R7003-19F,
HKU17112120 and HKU17127121 to E.X.W., and HKU17103819, HKU17104020 and
HKU17127021 to A.T.L.L.), Lam Woo Foundation, and Guangdong Key Technologies
for Treatment of Brain Disorders (2018B030332001) to E.X.W..References
[1] O’Reilly, Thomas, et al. "In vivo 3D brain and extremity
MRI at 50 mT using a permanent magnet Halbach array." Magnetic
resonance in medicine 85.1 (2021): 495-505.
[2] McDaniel, Patrick C., et al. "The MR
cap: A single‐sided MRI system designed for potential point‐of‐care limited
field‐of‐view brain imaging." Magnetic resonance in medicine 82.5
(2019): 1946-1960.
[3] Cooley, Clarissa Z., et al. "A
portable scanner for magnetic resonance imaging of the brain." Nature
biomedical engineering 5.3 (2021): 229-239.
[4] Liu Y, Leong, A.T.L., Zhao, Y., Xiao, L.,
Mak, H.K.F., Tsang, A., Lau G.K.K., Leung G.K.K. & Wu, E.X. A Low-cost and
Shielding-free ultra-low-field brain MRI scanner. Nature Communications in
press, MS#28454 (2021).
[5] Danieli, E., et al. "Highly stable and
finely tuned magnetic fields generated by permanent magnet
assemblies." Physical review letters 110.18 (2013):
180801.
[6] Li, Dong, et al. "The 2-17 type Sm 2-x
HRE x Co 10 Cu 1.5 Fe 3.2 Zr 0.2 (HRE= Gd, Tb, Dy, Ho, Er) magnets with low
temperature coefficient." IEEE Transactions on Magnetics 16.5
(1980): 988-990.
[7] Jones, F., and M. Tokunaga. "Low
temperature coefficient cobalt-rare earth magnets." IEEE
Transactions on Magnetics 12.6 (1976): 968-970.
[8] Narasimhan, K. S. V. L., M. G. H. Wells,
and D. V. Ratnam. "Rare earth–cobalt magnets with low reversible temperature
coefficient." Journal of Applied Physics 49.3 (1978):
2072-2074.