Development of a Magnetic-Susceptibility-Measurement Apparatus Used under High Pressure in Pulsed High Magnetic Fields
PI of Joint-use project: K. Nihongi
Host lab: Hagiwara, Kindo, and Uwatoko Groups
Host lab: Hagiwara, Kindo, and Uwatoko Groups
The combination of extreme conditions such as low temperature, high magnetic field, and high pressure provide insights into electrical and magnetic physical properties in condensed-matter materials. To date, magnetization measurements have been reported with an induction method using a non-destructive pulse magnet and a metallic piston-cylinder cell (PCC) made by Cu-Be or Ni-Cr-Al alloy under high pressure of up to 0.95 GPa in pulsed high magnetic fields of up to 50 T [1-3]. In this method, magnetization signal was detected by winding pick-up coils with approximately 100 turns around the exterior of the PCC (left panel of Fig. 1). This measurement apparatus is an effective tool for observing large abrupt transition phenomena, but the following factors may interfere with accurate measurements: (a) low sample-filling rate inside the pick-up coil, (b) extrinsic magnetization signals from the pressure cell, and (c) Joule heating caused by eddy currents in the metallic parts of the pressure cell in pulsed high magnetic fields. To deal with these problems, we have designed a new PCC in use for pulsed high magnetic fields and developed a magnetic-susceptibility-measurement apparatus using a proximity detector oscillator (PDO) under high pressures in pulsed high magnetic fields [4].
The PDO is an inductance (L)-capacitance (C) self-resonating LC tank circuit based on a widely available proximity detector chip used in modern metal detectors [5,6]. When a magnetic insulator is put into the sensor coil, this circuit detects the change in the resonance frequency (Δf) corresponding to the change in the dynamic magnetic susceptibility ((ΔM/ΔH). Hereafter, we refer to this technique as an LC method. The absolute value of Δf increases as the sample filling rate increases against the sensor coil. The sensor coil is typically wound only 5–30 turns with a diameter as small as 300 μm. Therefore, the sensor coil including the sample can be inserted in the pressure cell as shown in the right panel of Fig. 1. This setting prevents the magnetization signal of the pressure cell from being superimposed to the measurement signal.
The PCC used in our magnetic-susceptibility measurements in pulsed high magnetic fields was made of Ni-Cr-Al alloy with a low conductivity and high tensile strength compared with a PCC made of Cu-Be alloy. The sensor coil is put into the PCC in this LC method, and therefore the outer diameter of the PCC can be expanded up to 8.6 mmφ, resulting in increasing the applied pressure of up to 2.10 ± 0.02 GPa.
To evaluate the effect on the Joule heating at the sample position in pulsed high magnetic fields, we investigated the temperature change in the sample space using a magnetic-field and temperature-calibrated thermometer. Figure 2 shows the temperature changes at the sample position inside the PCC, starting from an initial temperature of 1.4 K, in pulsed high magnetic fields as a function of time. For the maximum field of 51 T, the temperature at the sample position remained almost 1.4 K until nearly 6.5 ms (approximately 40 T during the field-ascending process). After approximately 6.5 ms, the temperature slowly increased, reaching approximately 8 K at 40 ms (around 0 T). Since the sample is covered with a Teflon tube and immersed in Daphne 7373 as a pressure medium, the Joule heating from the metallic parts of the PCC is transmitted to the sample position with some delay.
To verify the performance of our developed magnetic-susceptibility-measurement apparatus in pulsed high magnetic fields under high pressures, we investigated the pressure dependence of the magnetic susceptibility of Ba3CoSb2O9, which is one of the typical S = 1/2 triangular lattice antiferromagnets and exhibits successive phase transitions below the Néel temperature TN = 3.8 K [7]. Figure 3 shows the Δfsub-H curve of Ba3CoSb2O9 for H || ab plane in pulsed magnetic fields of up to 51 T under pressures of up to 1.97 GPa. Δfsub is the frequency difference obtained by subtracting the frequency at T = 10 K as background from Δf at 1.4 K. At ambient pressure in the PCC during the field-ascending process, Δfsub-H curve was in good agreement with the field derivative magnetization dM/dH in a previous report [7]. Under several pressures, we also observed the anomalies at phase transition fields in the field-ascending process. In the field-descending process at ambient pressure in PCC and at 1.10 GPa, Δfsub-H curves did not show any anomalies at phase transition fields observed in the field-ascending process owing to the temperature increase of the sample above TN. As aforementioned,Δfsub up to the saturation field (Hsat) of approximately 32 T in the field-ascending process (Fig. 3(a)) is not affected by the increase in the sample temperature due to Joule heating. The present study demonstrated that our developed magnetic-susceptibility-measurement apparatus is a powerful tool for investigating magnetic properties of a frustrated magnet with a small spin value at low temperatures down to 1.4 K in pulsed high magnetic fields of up to 40 T under high pressures of up to 2.1 GPa. We hope that this apparatus will help us discover unconventional physical phenomena in quantum and frustrated spin systems.

Fig. 3. (a) Δfsub-H curves for H||ab plane of Ba3CoSb2O9 at 1.4 K at various pressures. Gray lines show Δfsub-H curves in the field-descending process at ambient pressure in PCC and at 1.10 GPa. Other lines are Δfsub-H curves in the field-ascending process. The dotted lines are guidelines indicating the pressure dependence of transition fields (Hc1, Hc2, Hc3 and Hsat). (b) Enlarged view of the Δfsub-H curves around 22 T. The curves in Figs. 3(a) and 3(b) are arbitrarily shifted from the ambient-pressure curve with increasing pressure for clarity.
References
- [1] T. Hamamoto, K. Kindo, T. Kobayashi, Y. Uwatoko, S. Araki, R. Settai, and Y. Onuki, Physica B 281-282, 64 (2000).
- [2] T. Tahara, T. Kida, Y. Narumi, T. Takeuchi, H. Nakamura, K. Miyake, K. Kindo, and M. Hagiwara, J. Phys. Soc. Jpn. 89, 064711 (2020).
- [3] T. Matsunaga, T. Kida, S. Kimura, M. Hagiwara, S. Yoshida, I. Terasaki, and K. Kindo, J. Low Temp. Phys. 159, 7 (2010).
- [4] K. Nihongi, T. Kida, Y. Narumi, N. Kurita, H. Tanaka, Y. Uwatoko, K. Kindo, and M. Hagiwara, Rev. Sci. Instrum. 94, 113903 (2023).
- [5] M. M. Altarawneh, Ph.D. thesis, The Florida State University, 2009.
- [6] S. Ghannadzadeh, M. Coak, I. Franke, P. A. Goddard, J. Singleton, and J. L. Manson, Rev. Sci. Instrum. 82, 113902 (2011).
- [7] T. Susuki, N. Kurita, T. Tanaka, H. Nojiri, A. Matsuo, K. Kindo, and H. Tanaka, Phys. Rev. Lett. 110, 267201 (2013).