Relationship of magnetic ordering and crystal structure in the lanthanide ferromagnets Gd, Tb, Dy, and Ho at high pressures

The pressure dependence of the magnetic ordering temperatures for the lanthanide ferromagnets Gd, Tb, Dy, and Ho has been investigated in the pressure region up to 18 GPa by two types of magnetic measurements using a superconducting quantum interference device (SQUID). The present magnetic measurements enabled us to investigate the pressure dependence of the magnetization intensity at low magnetic fields as well as the magnetic ordering temperatures. Their results are interpreted in the light of such previous experiments as magnetic susceptibility, magnetization, electrical resistance, neutron diffraction, and Mössbauer spectroscopy measurements. All of the magnetic orderings in the above four elements were suppressed down to less than the detection level, being related to the structural transition. The ferromagnetic ordering in Gd, Tb, Dy, and Ho is stabilized in the hcp structure. The magnetic anomalies due to the helimagnetic ordering of Tb and Dy disappear at the Sm-to-dhcp transition and the hcp-to-Sm transition, respectively, while that of Ho disappears in the Sm-type phase near the Sm-to-dhcp transition.

Figure 1

Structural transformation in $4f$ lanthanide ferromagnets Gd–Tm under pressure from hcp $\to$ Sm-type $\to$ double-hcp (dhcp) $\to$ fcc structure. The change in the stacking of hexagonal planes is expressed by naming their planes the A, B, and C planes.

Figure 2

$T$ dependence of $M_{\mathrm{AC}}^{\prime }$ for Gd at $H_{\mathrm{DC}}=0$ (a) and 1 kOe (b), measured by method (3). The number in parentheses gives the order of the measurement number for each run.

Figure 3

$P$ dependence of $T_{\mathrm{C}}$ for Gd. The data in the literature on magnetic measurements such as by Mito et al. [8, 10] and Jackson et al. [9] and on electrical measurements by Samudrala et al. [11] and Lim et al. [14, 16] are also presented.

Figure 4

$T$ dependence of $M_{\mathrm{AC}}^{\prime }$ for Tb at $H_{\mathrm{DC}}=0$, measured by method (3). The $T_{\mathrm{N}}$ (high-$T$ side) and $T_{\mathrm{C}}$ (low-$T$ side) are marked with arrows, and their changes are traced with lines.

Figure 5

$P$ dependence of $T_{\mathrm{N}}$ and $T_{\mathrm{C}}$ for Tb. The data in the literature on magnetic measurements by Mito et al. [10] and Jackson et al. [9], electrical resistance ($R$) measurements by Thomas et al. [12] and Lim et al. [15, 16], and neutron diffraction [12] are also presented.

Figure 6

$T$ dependence of $M_{\mathrm{AC}}^{\prime }$ in Dy at $H_{\mathrm{DC}}=0$. $T_{\mathrm{N}}$ (high-$T$ side) and $T_{\mathrm{C}}$ (low-$T$ side) are marked with arrows, and their changes are traced with lines. (a) $M_{\mathrm{AC}}^{\prime }$ in the first run, (b) $M_{\mathrm{AC}}^{\prime }$ in the second run, and (c) normalized $M_{\mathrm{AC}}^{\prime }$ after subtracting the normalized background contribution. In (a) and (b), the number in parentheses gives the order of the measurement number for each run.

Figure 7

$P$ dependence of $T_{\mathrm{N}}$ and $T_{\mathrm{C}}$ for Dy. The data in the literature on magnetic measurements by Mito et al. [10] and Jackson et al. [9] and on electrical measurements by Samudrala et al. [13] and Lim et al. [15, 16] are also presented. Referring to Fig. 6, the anomalies determined by Lim et al. appear at around the positions of the minimum $M_{\mathrm{AC}}^{\prime }$ between $T_{\mathrm{N}}$ and $T_{\mathrm{C}}$. The transition temperature determined from $R$ by Samudrala et al. [13] follows $T_{\mathrm{N}}$ determined in the present data and, for $P>7$ GPa, it approaches the broken extrapolated line of $T_{\mathrm{C}}$ evaluated from the $M_{\mathrm{DC}}$ data at $H_{\mathrm{DC}}=0.3$ T by Mito et al. [10]. For reference, in the neutron diffraction experiment by Perreault et al. [18], the temperature at which the nuclear peak with a magnetic origin appears is also marked. It is also on the extrapolated line of $T_{\mathrm{C}}$ evaluated from the $M_{\mathrm{DC}}$ data by Mito et al. [10].

Figure 8

$T$ dependence of $M_{\mathrm{AC}}^{\prime }$ in the first run for Ho at $H_{\mathrm{DC}}=0$, measured by method (3). $T_{\mathrm{N}}$ (at high-$T$ side) is marked with an arrow, and $T_{\mathrm{C}}$ (at low-$T$ side) at $P=0$ is marked with a dotted line. The number in parentheses gives the order of the measurement number. The data for sequences of (1)–(5) and (6)–(11) are shown in (a) and (b), respectively.

Figure 9

$T$ dependence of $M_{\mathrm{AC}}^{\prime }$ in the second (a) and third (b) runs for Ho at $H_{\mathrm{DC}}=0$, measured by method (3). In (a), $T_{\mathrm{N}}$ (at high-$T$ side) is marked with an arrow, and $T_{\mathrm{C}}$ (at low-$T$ side) at $P=0$ is marked with a dotted line. The number in parentheses gives the order of the measurement number. In (b), $T_{\mathrm{N}}$ at (high-$T$ side) and $T_{\mathrm{C}}$ (at low-$T$ side) at $P=0$ are marked with the arrows.

Figure 10

$T$ dependence of $M_{\mathrm{VCM}}$ in Ho, measured by method (4). $M_{\mathrm{VCM}}$ at the $H_{\mathrm{DC}}$ of 2.0 Oe was measured during the warming process after zero-field cooling over two runs [(a) the first and (b) the second run].

Figure 11

$P$ dependence of $T_{\mathrm{N}}$ and $T_{\mathrm{C}}$ in Ho. The data on $T_{\mathrm{N}}$ and $T_{\mathrm{C}}$ in the literature on magnetic measurements by Mito et al. [10] and Jackson et al. [9] are also presented. For reference, in the neutron diffraction experiment by Perreault et al. [18], the temperatures at which magnetic diffraction appears are also marked.

Figure 12

The relationship between the critical pressure $P_{\mathrm{c}}$ for the disappearance of both FM and HM magnetizations and the change in the crystal structure. The pressure regions in which the FM and HM magnetizations survive are displayed with light blue and light green bars, respectively. LT and HT in parentheses stand for the low-temperature and high-temperature phases, respectively.