Pressure-Induced Structural Phase Transition in EuNi2P2.

EuNi2P2 was studied with a diamond anvil cell (DAC) and X-ray diffraction (XRD). The structural evolution of powder crystal EuNi2P2 under high pressure up to 137 GPa and its single-crystal structure up to 9 GPa were reported. The unique structural phase transition of this 122-type crystal occurred above 70 GPa in powder crystal EuNi2P2. The diffraction data from single-crystal EuNi2P2 revealed the coordinate change of the P atom, and the stability of the crystal at 9 GPa was confirmed. The crystal EuNi2P2 remained stable with a tetragonal phase without obvious symmetry changes during compression to 137 GPa.

Introduction

There is a rich variety of compounds similar to the ThCr2Si2-type crystal (I4/mmm). Its unique tetragonal structure can accommodate a variety of atoms with large radius differences, which leads to its special structural and electromagnetic transitions that occur under high pressure.[1−5] According to previous works on high-pressure study of EuNi2P2, the crystal had no obvious structural change in the pressure range of 32–45 GPa.[5,6] Interestingly, this type of crystal often exhibits a transition to a collapsed tetragonal phase at high pressure, such as EuCo2As2, EuFe2As2, or EuCo2P2. In the phase transition, the abnormal sharp decrease in the c-axis parameter has attracted much attention.[3,7,8] The compressibility along the c-axis is higher than that along the a- and b-axes, and high-pressure-induced amorphization may even occur with significant compression along the c-axis.[9] In this work, a high-pressure powder crystal X-ray diffraction (XRD) experiment on EuNi2P2 was carried out to investigate whether there was a similar phase transition to a collapsed tetragonal phase. We investigated the compressibility of EuNi2P2 under high pressures up to 137 GPa at room temperature. The equation of state (EOS) of EuNi2P2 was also evaluated by fitting pressure–volume data that had been collected and Rietveld refined. Above 66 GPa, the c parameter and crystal volume of EuNi2P2 suddenly changed and a possible phase transformation occurred.

The valence state of the rare earth element Eu has always been a hot topic in physics and is closely related to the transfer of some 4f-electrons to the conduction band and near the Fermi surface.[10] In recent years, EuNi2P2 has been regarded as a rare earth compound with a mixed valence state, and the valence state of Eu is closely related to both pressure and temperature.[11] The evolution of the volume of single-crystal EuNi2P2, which is somewhat connected to its valence state, is an interesting study.[12] High-pressure study of EuNi2P2 has shown that there is a transition from Eu2+ to Eu3+ and that a new P–P bond parallel to the c-axis can be formed.[5] Additionally, the electrical resistivity of EuNi2P2 has been reported to abnormally change above 8 GPa.[6] The changes in the mixed valence state and resistivity under high pressure may be accompanied by subtle changes in the structure; however, as reported, there is no obvious change in the structure based on the powder crystal data. Therefore, by taking advantage of high-pressure single-crystal XRD,[13] this work conducted a detailed study on the high-pressure structure of crystal EuNi2P2.

Experimental Details

The EuNi2P2 sample was synthesized using the tin-flux method.[5,14] The samples used in the two independent experiments were processed from the same batch of single-crystal samples.

Powder X-ray Diffraction

The in situ high-pressure XRD measurement for powder EuNi2P2 was performed up to 137 GPa, using synchrotron X-rays with a wavelength of 0.3344 Å at 13-ID-D in APS (Argonne National Laboratory). The powder sample was loaded into a diamond anvil cell (DAC) with a diamond culet size of 100 μm. A Re chip was used as a gasket, and Au was used as a pressure marker.[15] Without a pressure-transmitting medium, the effects of deviatoric stress present at high pressure will exist.[9,16]

Single-Crystal X-ray Diffraction

For the sample preparation, EuNi2P2 crystals were processed into 30 μm blocks for later use. The DAC was used as a pressurizing device with a diamond culet of 500 μm. A 300 μm diameter hole was drilled in T301, used as the gasket, to form the sample chamber, and ruby was used as the pressure marker.[17] In this experiment, we loaded neon as a pressure-transmitting medium to offer an approximate hydrostatic environment. The light source used for the experiments was a Bruker D8 VENTURE, a single-crystal X-ray diffractometer with a Ag Kα IμS microfocus source (Ag Kα, λ= 0.56087 Å). The collected data were refined with SHELX through APEX3 and Olex2.[18]

Results and Discussion

Powder Crystal

Figure shows the diffraction patterns of the powder EuNi2P2 sample at high pressure, together with those of the Au and Re phases. No obvious new peaks are observed at pressures up to 137 GPa, which proves the stability of this series of crystals.[5−7] No phase transitions are observed up to 137 GPa in the XRD experiments. As shown in Figure , the crystal is refined according to the (I4/mmm, No. 139) symmetry to obtain the lattice parameters of EuNi2P2 under pressure. Figure shows the evolution of the lattice parameters during compression and the data from the GSAS and PowderCell program packages.[19] Since there was no pressure-transmitting medium, the initial pressure was 40 GPa. When fitting the volume of the crystal under high pressure using the third-order Birch–Murnaghan EOS,[20] in which the bulk modulus is B0 and its pressure derivative is B0′, the Birch–Murnaghan formulation is valid only when the latter value is sufficiently close to 4,[21] and two red and blue fitting lines can be obtained.

Figure 1

Selected powder synchrotron XRD patterns of EuNi2P2, with Re and Au pressure markers, up to 137 GPa. X-ray diffraction patterns of EuNi2P2 at high pressures, which are indexed to the tetragonal ThCr2Si2-type structure (I4/mmm).

Figure 2

Rietveld refinements of powder EuNi2P2 (Re, Au) at 0 and 40 GPa showing that no phase transitions are observed up to 137 GPa in the XRD experiments. The crystal is refined according to the (I4/mmm, No. 139) symmetry.

Figure 3

(a) Variations of lattice constants under high pressure. (b) Pressure dependence of the unit cell volume compression (V) for the tetragonal phase of EuNi2P2 up to 137 GPa. The red and blue solid curves are the Birch–Murnaghan EOS fitted to the two phases. The bulk modulus of EuNi2P2 is B0 = 72(3) GPa and 205(13) GPa. The error for pressure determination gradually increases with increasing pressure.

Birch–Murnaghan equation:

In this work, to better compare the bulk modulus of EuNi2P2, we fitted the obtained volume data to the second-order Birch–Murnaghan EOS (fixing B0′ = 4).[22,23] The correlation coefficient of B0 and its pressure derivative B0′ in the first fitting is greater than 99%, and that in the second fitting is greater than 98%.[24] However, the bulk modulus B0 increases to 205(13), which is not only due to the increase in atomic packing in the crystal but also related to the increase in deviatoric stress. The existence of deviatoric stress will reduce the compressibility of the crystal.[9,16] According to Figure b, the crystal shows a rapid volume decrease of approximately 19% with the pressure increasing from 66 to 80 GPa; at the same time, the drastic decrease in the lattice parameter c in the crystal is more than 10% according to Figure a, which indicates a tetragonal phase-to-collapsed tetragonal phase transition in EuNi2P2 above 66 GPa. The same type of crystal structure has Eu–Eu exchange interactions in the ab-plane and a helical antiferromagnetic structure with the helix axis along the c-axis direction,[25] where c/a is greater than 2. The material is more compressible along the c-axis under high pressure. Similar to the cases of EuCo2P2, EuCo2As2, and EuFe2As2, in which there is an isostructural collapse in EuCo2P2 at 3.1 GPa,[3,8] EuCo2As2 at 4.7 GPa, and EuFe2As2 at 8.5 GPa,[7] the above structural collapse occurs within 15 GPa. The isostructural collapse of this series is accompanied by negative compressibility of lattice parameter a and reduction of the polar velocity of c, while the high-pressure collapse of EuNi2P2 only shows rapid compression of c, without negative compressibility of a. According to previous reports, the bond interaction of the P atom and the change in the electronic structure might be the main reasons for the collapse under high pressure.[3] Therefore, more detailed research is needed.

Single Crystal

The high-pressure single-crystal diffraction data are refined to obtain the crystal CIF files within 10 GPa. The evolution of the high-pressure single-crystal parameters is consistent with previous work within 45 GPa.[5,6] As shown in Figure , the changes in the a of EuNi2P2 are relatively stable, while c fluctuates to some extent, which may lead to instability of the c-axis changes. The single-crystal structure is refined with Olex2 to produce the corresponding CIF files,[18] and the lattice parameters, lattice volumes, and lengths of Ni–P and P–P bonds at different pressures are determined with pubICIF.[26] With the Birch–Murnaghan equation,[20] the P–V curve is fitted, and the bulk modulus B0 is obtained as B0 = 103(2) GPa. The red solid line in Figure is the Birch–Murnaghan EOS fitted to the phase. Compared to previous reports, the bulk modulus of EuNi2P2 up to 9.02 GPa is smaller in this work; the previously obtained bulk modulus is B0 = 154.9(6) GPa in the range of 0–32 GPa[5] and B0 = 147.9(3) GPa in the range of 0–45 GPa.[6] Interestingly, the single-crystal EuCo2As2 under nonhydrostatic pressure exhibits an isostructural phase transition from 3.2 to 4.7 GPa with a collapsed tetragonal structure, resulting in a smaller tetragonal structure with bulk modulus B0 = 48(4) up to 4.7 GPa and B0 = 111(2) GPa after the collapse.[3] Single-crystal EuFe2As2, which is layered, also exhibits a collapsed structural transformation, with a smaller bulk modulus of 39(1) GPa before the collapse compared to the value of 134(1) GPa after the collapse.[7] In comparison with the bulk modulus of the same type of crystal, B0 of single-crystal EuNi2P2 in the range below 9 GPa of 103(2) GPa has a certain credibility.[23]

Figure 4

Lattice parameters a and c for the tetragonal phase as a function of pressure; (inset) pressure dependence of the unit cell volume compression (V) for the tetragonal phase of single-crystal EuNi2P2 up to a pressure of 9.02 GPa. The red solid curves are the Birch–Murnaghan EOS fitted to the phases.

The refined high-pressure crystal structure was obtained by Vesta,[27] from which the bond lengths and atomic positions were derived, as shown in Figure . The z fractional coordinate of the P atom is the unique changed parameter under high pressure, and the evolution process of the crystal structure under pressure can be better analyzed by studying the evolution of P, which is also an important basis for influencing the fluctuation of the mixed valence state of Eu. The bond length decreases with increasing P atomic coordinate, as shown in Figure . There are possible mutations in the crystal during compression due to changes in the valence state that may cause collapsed phase transformation.[28] Possible structural phase transitions can be found by meticulously studying the high-pressure structures.

Figure 5

(Blue squares) z fractional coordinate of the P atom; (black squares) evolution trend of the P–P bond length during compression.

Medvedev et al. found that the resistivity of EuNi2P2 obviously changed above 8 GPa.[6] For ThCr2Si2-type structure BaTi2Sb2O, a new Sb–Sb bond appears according to the Sb p to Ti 3d bands,[29] and the number of d-electrons is decisive for the order of new bonds.[1] Interestingly, in EuNi2As2, the existence of Ni vacancies does not affect the crystal structure and related physical properties,[30] but the change in the components will change the lattice volume to some extent and affect the charge transfer efficiency.[31] According to previous work on EuCo2P2, the crystal structure underwent a collapsed phase transition under 4.5 GPa, and the high-pressure-induced redistribution of the electron density to form a new weak P–P bond. Continuous pressure increased the electron density, and chemical bonds such as Eu–P and P–P were significantly enhanced. For EuCo2P2 under high pressure, the P–P bond was shortened and the electron redistribution was changed, which eventually led to an obvious change in the crystal microstructure and properties.[4,8] However, in the single-crystal experiment, no fine structural abnormalities of EuNi2P2 were found during compression to 9 GPa.

As shown in Figure , during compression, there appears to be an obvious alteration of the electron density as the pressure increases. This electron density alteration coincides with the appearance of charge transfer in previous reports.[3,5−7] Regarding the important role of the d-electrons, in homologous structure EuFe2P2, both Fe–P and Fe–Fe exhibit covalent and ionic bond characteristics and Fe 3d-electrons also play an important role in the electronic structure.[32] The electronic structure of the Ni (Fe, Co) atom has an important influence on the bond interaction of the anion.

Figure 6

Electron density distribution of atoms (Eu, Ni, P) projected perpendicular to [100].

Conclusions

In this work, the isostructural phase transition of EuNi2P2 above 66 GPa is found by high-pressure diffraction measurements, and the change in the P–P bond length is found to correspond to the coordinate of the P atom according to the single-crystal XRD experiment. Due to the large deviatoric stress in the nonhydrostatic environment, the compressibility of EuNi2P2 is reduced, so we calculated a large bulk modulus B0 of 205(13) GPa. Combined with single-crystal diffraction data, the symmetry of EuNi2P2 is found to have good stability but c exhibits a sudden change trend during compression.