Synthesis, Structure, and Complex Magnetism of MIr2In

Mar 7, 2016 - ABSTRACT: We report the synthesis, crystal structure, and physical properties of two new polar intermetallic compounds,. EuIr2In8 and Sr...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

Synthesis, Structure, and Complex Magnetism of MIr2In8 (M = Eu, Sr) Nicholas P. Calta,† Sergey L. Bud’ko,‡,§ Alexandra P. Rodriguez,† Fei Han,∥ Duck Young Chung,∥ and Mercouri G. Kanatzidis*,†,∥ †

Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States Department of Physics & Astronomy, Iowa State University, 12 Physics Hall, Ames, Iowa 50011, United States § Division of Materials Science & Engineering, Ames Laboratory, Ames, Iowa 50011, United States ∥ Materials Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States ‡

S Supporting Information *

ABSTRACT: We report the synthesis, crystal structure, and physical properties of two new polar intermetallic compounds, EuIr2In8 and SrIr2In8. Both were synthesized in good yield using In metal as a reactive flux medium, enabling the growth of large crystals for physical property measurements. They crystallize in the orthorhombic space group Pbam with the CeFe2Al8 structure type, which is sometimes also referred to as the CaCo2Al8 structure type. The two analogues have unit cell parameters of a = 13.847(3) Å, b = 16.118(3) Å, and c = 4.3885(9) Å for M = Eu and a = 13.847(3) Å, b = 16.113(3) Å, and c = 4.3962(9) Å for M = Sr at room temperature. SrIr2In8 is a diamagnetic metal with no local magnetic moments on either the Sr or Ir sites, and the diamagnetic contribution from core electrons overwhelms the expected Pauli paramagnetism normally seen in intermetallic compounds. Magnetism in EuIr2In8 is dominated by the local Eu moments, which order antiferromagnetically at 5 K in low applied fields. Increasing the field strength depresses the magnetic ordering temperature and also induces a spin reorientation at lower temperature, indicating complex competing magnetic interactions. Lowtemperature heat capacity measurements show a significant enhancement of the Sommerfeld coefficient in EuIr2In8 relative to that in SrIr2In8, with estimated values of γ = 118(3) and 18.0(2) mJ mol−1 K−2, respectively.

■

INTRODUCTION Intermetallic compounds that contain rare-earth elements with two stable oxidation states (specifically Ce, Eu, Sm, and Yb) exhibit a wide variety of interesting and unusual physical properties including superconductivity,1,2 heavy fermion behavior,3 the Kondo effect,4−6 and non-Fermi liquid behavior.7 Of particular interest for our investigation of the Eu−Ir−In system are the so-called heavy fermion materials, intermetallics that exhibit carrier effective masses (m*) far larger than can be explained by the weakly interacting electron models that describe typical metals. Heavy fermion behavior arises at low temperature as a consequence of strong coupling between localized f-electron moments and conduction electrons. This coupling is known as the Kondo effect,6 and while it can have different consequences in different situations, it is generally destroyed by the onset of long-range magnetic order. The family of compounds most relevant to the Eu Ir−In system in this context, CeTIn51 and structurally related Ce2TIn8 (T = Co, Rh, Ir),8,9 exhibit unconventional superconductivity that arises from a mechanism believed to be similar to that of high-Tc superconductors. RTIn5 (R = rare earth) analogues to the Ce compounds have been reported for all of the rare-earth elements apart from Pm and Eu. Therefore, our initial goal was to synthesize and characterize the as-yet-unreported EuIrIn5. © XXXX American Chemical Society

The chemistry of Eu−Ir−In systems is relatively unknown, with EuIrIn410 and Eu3Ir2In1511 the only reported ternary compounds. Both were synthesized using a reactive In flux technique very similar to the one we use in this report. EuIrIn4 crystallizes in the YNiAl4 structure type with a threedimensional Ir−In framework that has voids filled by Eu atoms. It is an antiferromagnet that exhibits multiple magnetic ordering transitions, suggesting the presence of low-energy magnetic states very near to the ground state in energy. Eu3Ir2In15 adopts a vacancy-filled variant of the Sc5Co4Si10 structure type that also features a three-dimensional Ir−In framework and channels filled by relatively isolated Eu atoms. Similar to EuIrIn4, Eu3Ir2In15 also exhibits rather unusual magnetic behavior, with multiple ordering transitions at low temperature all attributed to changes in the magnetic configuration of the local Eu magnetic moments.11 Instead of the targeted EuIrIn5 phase or previously reported phases, our synthesis yielded EuIr2In8, a new compound that crystallizes in the CeFe2Al8 structure type,12,13 which is also sometimes referred to as the CaCo2Al8 structure type.14 Over 35 intermetallic compounds are known to adopt this structure Received: January 11, 2016

A

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Powder X-ray Diffraction. The identity and purity of synthetic mixtures was checked by powder X-ray diffraction. Samples were ground in a mortar and pestle, and the resulting powder was affixed to a borosilicate glass slide with double-sided tape. The diffraction pattern was measured using an Inel CPS120 diffractometer using Cu Kα radiation (λ = 1.5406 Å). Single-Crystal X-ray Diffraction. All single-crystal X-ray diffraction data were collected at room temperature with a Stoe IPDS2 or IPDS2T single-crystal diffractometer using graphitemonochromated Mo Kα radiation (λ = 0.71073 Å). Multiple single crystals were selected from the reaction mixture, cut to an appropriate size for single-crystal X-ray diffraction, and mounted on glass fibers with glue. These crystals were screened to determine the unit cell and crystal quality by measuring a small number of frames. Full-sphere data were collected on best crystals. Data collection, integration, absorption correction, and reduction were performed using the X-AREA software suite provided by Stoe.31 The structure was solved with direct methods and refined based on Fo2 using the SHELX suite.32 Refined structures were standardized using the Structure Tidy function in PLATON.33 Each refinement was checked for partial and mixed occupancy, but all atomic positions were fully occupied. A summary of the least-squares refinement result is shown in Table 1. Scanning Electron Microscopy−Energy-Dispersive X-ray Spectroscopy (SEM−EDS). SEM images were obtained and elemental compositions acquired using a Hitachi S3400N-II instru-

including the two archetypal members as well as many other aluminides including RFe 2 Al 8 (R = La, Pr, Eu), 15,16 RFe2Al8−xMgx (R = La, Ce, Pr, Nd, Sm),17 RCo2Al8 (R = La, Ce, Pr, Sm, Yb),18−21 CaNi2Al8,22 and YbNi2−xFexAl8,23 the gallides EuM2Ga8 (M = Co, Rh, Ir),24 RFe2Ga8 (R = La, Ce, Pr, Nd, Sm),25 RCo2Ga8 (R = Ca, Ce, Pr, Yb),25−27 RRu2Ga8 (R = La, Ce, Pr, Nd), and EuM2Ga8 (M = Rh, Ir),27 and the indides SrRh2In828 and EuRh2In8.27 The structure appears to be relatively stable and can be synthetically accessed by a variety of routes including arc melting, direct combination by radiofrequency induction heating, and flux synthesis. Chemical substitution is possible on the transition-metal site (M) and can tune the valence of the rare-earth atom (R) for elements such as Ce that exhibit mixed valence.29 While not all reports of compounds in this ternary intermetallic family include physical property measurements, many CeFe2Al8-type compounds exhibit unstable magnetic ground states and physical properties that arise as a consequence of competing magnetic interactions. For example, both CeFe2Al8 and CeCo2Al8 exhibit clear signs of strong electron correlations in both magnetic and transport measurements,30 and a solid solution of the two with the composition CeCoFeAl8 exhibits evidence of Ce valence fluctuations.29 The series RFe2Al8−xMgx (R = La, Ce, Pr, Nd, Sm)17 was synthesized in Mg/Al flux, resulting in Mg substitution on the Al sites. All analogues exhibit rather unusual physical properties ranging from evidence for valence fluctuation in the Sm analogue to low-temperature upturns in resistivity for the La and Pr analogues, suggesting the that the Kondo effect plays a role in the behavior of these systems when Mg substitution alters the Fermi level. The closest analogues to the title compounds, EuRh2Ga8 and EuRh2In8, both exhibit complex magnetic behavior at low temperature in which competing magnetic couplings stabilize slightly different magnetic structures.27 In this Article, we report the synthesis, crystal structure, and physical properties of two new CeFe2Al8-type compounds, EuIr2In8 and SrIr2In8. Both were synthesized using In flux as large single crystals suitable for physical property measurements. The Sr analogue behaves as a normal metal without localized magnetic moments, and the Eu analogue exhibits an enhanced Sommerfeld coefficient and antiferromagnetic (AFM) order at low temperatures.

■

Table 1. Single-Crystal X-ray Diffraction Refinement Details for MIr2In8 (M = Sr, Eu) empirical formula, Z fw temperature (K) wavelength (Å) cryst syst space group unit cell dimens a (Å) b (Å) c (Å) volume (Å3) calcd density (g cm−3) abs coeff (mm−1) F(000) cryst size (μm) θ range for data collection (deg) index ranges

EXPERIMENTAL DETAILS

Synthesis. All manipulations of air-sensitive Eu and Sr were performed in a nitrogen glovebox. For each reaction, 1 mmol of Eu metal powder (hand filed from lump, 0.152 g, 99.9%, Chinese Rare Earth Information Center) or Sr metal pieces (0.088 g, 99%, Aldrich), 0.5 mmol of Ir powder (0.096 g, 99.95%, American Elements), and 15 mmol of In pieces (1.722 g, 99.999%, Plasmaterials) were placed in an alumina crucible. A stainless steel filter (McMaster-Carr, 100 mesh woven wire) and a short alumina tube were placed on top of the crucible, and the entire assembly was sealed under vacuum (∼5 × 10−4 mTorr) in a fused-silica tube. The sealed tube was placed in a tube furnace, then heated to 1000 °C over 12 h, held at that temperature for 48 h, slowly cooled to 550 °C over 48 h, held at 550 °C for a few hours, then removed from the furnace and quickly centrifuged to remove excess In flux. Remaining In metal was then removed from the product crystals by etching for 7−10 h in ∼5% HCl. The MIr2In8 crystals are stable in this concentration of HCl for at least 4−5 days, and occasionally longer etchings were required to completely remove In from the surface of the crystals. Typical yields for reactions using these conditions are 60−70%, calculated by mass based on the limiting reagent Ir.

reflns collected indep reflns completeness (%) refinement method data/restraints/param GOF final R indices [I > 2σ(I)] R indices (all data) extinction coeff largest diff peak, hole (e Å−3) weighting scheme

EuIr2In8, 4 1454.92

SrIr2In8, 4 1390.58 293(2) 0.71073 orthorhombic Pbam

13.8468(6) 16.1177(7) 4.3885(2) 979.42(7) 9.867 51.720 2436 90 × 30 × 10 2.92−34.87

13.847(3) 16.113(3) 4.3962(9) 980.9(3) 9.417 50.747 2336 120 × 20 × 15 2.93−34.83

−22 ≤ h ≤ 22 −25 ≤ k ≤ 25 −6 ≤ l ≤ 6 14394 2335 (Rint = 0.0485) 99.1 full-matrix least 2335/0/70 1.213 R1obs = 0.0343, wR2obs = 0.0471 R1all = 0.0443, wR2all = 0.0488 0.00134(4) 2.373, −3.213

−22 ≤ h ≤ 22 −24 ≤ k ≤ 25 −7 ≤ l ≤ 6 14078 2337 (Rint = 0.0678) 99.4 squares on F2 2337/0/70 0.995 R1obs = 0.0329, wR2obs = 0.0474 R1all = 0.0560, wR2all = 0.0511 0.00232(4) 2.510, −2.884

w1 = 0.0163, w2 = 3.9867

w1 = 0.0184, w2 = 0

R1 = ∑||Fo| − |Fc||/∑|Fo|, wR2 = (∑[w(|Fo|2 − |Fc|2)2]/∑[w(| Fo|4)])1/2 and calcd w = 1/[σ2(Fo2) + (w1P)2 + w2], where P = (Fo2 + 2Fc2)/3 and w1 and w2 are refined weights. B

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry ment equipped with an Oxford Instruments INCAx-act EDS detector. Samples consisted of unpolished single crystals, and EDS spectra were collected from clean, flat surfaces of crystals using an accelerating voltage of 20 kV and a collection time of 60 s. Collected EDS spectra were consistent with an elemental composition of MIr2In8 within the expected experimental error for all measured crystals. Resistivity Measurements. Resistivity as a function of temperature was measured using a Quantum Design Physical Property Measurement System (PPMS) on oriented single crystals along the a axis using the four-probe method. Copper wire contacts were attached to the samples using conductive Dupont 4929N silver paste, and SEM images were used to determine sample dimensions. Magnetization Measurements. Field- and temperature-dependent magnetization data were collected using a Quantum Design MPMS SQUID magnetometer. Each sample consisted of ground single crystals secured with super glue in a gel capsule wrapped with Teflon tape and contained in a straw. Heat Capacity Measurements. The temperature-dependent specific heat capacity was measured using the hybrid adiabatic relaxation technique of the heat capacity option on a Quantum Design PPMS. The data were taken on several crystals assembled on a heat capacity platform, with thermal contact to the platform provided using Apiezon N grease.

Structure. MIr2In8 (M = Sr, Eu) crystallizes in the orthorhombic space group Pbam with the CeFe2Al8 structure type with unit cell edges of a = 13.847(3) Å for both b = 16.113(3) and 16.118(3) Å and c = 4.3962(9) and 4.3885(9) Å, respectively. The overall refined structure is shown in Figure 2,

■

Figure 2. (a) Structure of MIr2In8 viewed along the c axis and (b) multiple cells viewed from the same perspective, with three IrIn9 units highlighted. (c) Single unit cell viewed along the a axis. (d) View of the IrIn9 columns viewed perpendicular to the c axis.

RESULTS AND DISCUSSION Synthesis. Our synthetic approach utilized the metal flux technique in order to overcome the very high melting point of Ir metal (2446 °C)34 coupled with the relatively low boiling point of Eu metal (1529 °C).34 In our reaction, a large excess of In acts as a solvent and dissolves both Eu and Ir, facilitating fast reactant diffusion and permitting the growth of large, rodshaped single crystals,35−37 which are shown in Figure 1. Our

Figure 1. SEM micrographs of typical flux-grown crystals of (a) EuIr2In8 and (b) SrIr2In8.

Figure 3. Local coordination environments around the (a) Ir atom and (b) Eu or Sr atom.

optimized synthetic conditions use an excess of Eu metal in order to avoid the formation of IrIn3, an unwanted side product, and produce yields of 60−70%. Some reactions with excess Eu still contain a small amount of IrIn3, on the order of 5%. SrIr2In8 syntheses appear phase-pure by powder X-ray diffraction, but one resistivity measurement showed a small drop in the resistivity at the superconducting critical temperature of SrIr2, indicating that this phase is present in some reaction mixtures in amounts below the detection threshold of our powder diffractometer, roughly 5 mol %. Reactions where M = Ca or Ba, however, yield primarily CaIrIn438 and BaIr2In939 rather than the MIr2In8 structure, suggesting that the stability of the MIr2In8 structure is very sensitive to the ionic radius of the M cation. Because Eu and Sr are nearly identical in size, it is expected that they would adopt the same structure. In addition to the relative atomic size, flux chemistry plays an important role in stabilizing the MIr2In8 structure. Slightly different heating profiles and reactant concentrations dissolved in the In flux medium yield either EuIrIn410 or Eu3Ir2In15,11 highlighting the rich chemistry of the Eu−Ir−In system when explored with In flux.

with local coordination shown in Figure 3. Atomic positions for both the Sr and Eu analogues are presented in Table 2. As is the case with all known M−Ir−In (M = Sr, Eu) compounds, the structure is composed of an extended Ir−In framework broken up by M2+ ions that occupy channels along the c axis. The Ir−In framework contains no Ir−Ir bonds and is completely described as a series of two crystallographically independent [IrIn9]capped triangular prisms very similar to those observed in BaIr2In9.39 The nine-coordinate environment of the Ir atom is higher than the six- and eight-coordinate Ir environments observed in SrIrIn240 and MIrIn 4 (M = Eu, Sr),10,41 respectively. The more complete coordination of Ir and the higher dimensionality of the Ir−In network facilitate the incorporation of more In into the structure, resulting in the more In-rich composition relative to the 1−1−2 and 1−1−4 compounds with similar composition. As in BaIr2In9, the IrIn9 prisms stack along the c axis, parallel to the channels containing M2+ ions. The IrIn9 prisms connect in the ab plane by either face- or corner-sharing (Figure 2d). The Ir−In bond lengths in these prisms range from 2.5985(8) and 2.596(1) Å to 2.9387(6) and 2.9314(7) Å in EuIr2In8 and SrIr2 In8 , respectively. The short Ir−In bond lengths correspond to the C

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 2. Fractional Atomic Coordinates and Isotropic Displacement Parameters for MIr2In8 (M = Eu, Sr)a EuIr2In8

a

SrIr2In8

label

x

y

z

Uisob

x

y

z

Uisob

Ir(1) Ir(2) Eu/Sr In(1) In(2) In(3) In(4) In(5) In(6) In(7) In(8) In(9)

0.0345(1) 0.1536(1) 0.3426(1) 0.0193(1) 0.1671(1) 0.2400(1) 0.3306(1) 0.4487(1) 0.1002(1) 0.3400(1) 0 0

0.4045(1) 0.0964(1) 0.3192(1) 0.1339(1) 0.3779(1) 0.1778(1) 0.4881(1) 0.1833(1) 0.2535(1) 0.0443(1) 1 /2 0

0 0 0 1 /2 1 /2 1 /2 1 /2 1 /2 0 0 1 /2 0

7(1) 5(1) 13(1) 8(1) 9(1) 8(1) 8(1) 8(1) 11(1) 10(1) 9(1) 10(1)

0.345(1) 0.1536(1) 0.3421(1) 0.0203(1) 0.1665(1) 0.2399(1) 0.3301(1) 0.4488(1) 0.0999(1) 0.3399(1) 0 0

0.4044(1) 0.0965(1) 0.3191(1) 0.1336(1) 0.3783(1) 0.1769(1) 0.4888(1) 0.1829(1) 0.2534(1) 0.0443(1) 1 /2 0

0 0 0 1 /2 1 /2 1 /2 1 /2 1 /2 0 0 1 /2 0

6(1) 4(1) 12(1) 7(1) 8(1) 8(1) 8(1) 8(1) 9(1) 9(1) 7(1) 9(1)

All atomic positions are fully occupied. bUiso is defined as one-third of the trace of the orthogonalized Uij tensor and uses units Å2 × 10−3

⎛ T ⎞4 ρ = ρ0 + AT ⎜ ⎟ ⎝ ΘD ⎠

two-coordinate bridging bonds and are shorter than that observed in binary IrIn2 and IrIn3, indicating strong M−M bonding between the IrIn9 clusters in the ab plane. The long bond lengths are observed in the two axial Ir−In bonds that link stacks of polyhedra along the c axis. While these are longer than most observed Ir−In bonds reported in intermetallic solids, they are short to be considered genuine bonding interactions, albeit comparatively weak ones. The M2+ ions occupy channels that are pseudo-5-fold-symmetric, and each M2+ ion is coordinated by 10 In ligands with bond lengths between 3.3722(7) and 3.5004(8) Å for Eu and 3.391(1) and 3.512(1) Å for Sr. These bond lengths are similar in magnitude to the sum of covalent radii for Sr/Eu and In, which suggests that M−In bonding interactions are nonnegligible in this compound, similar to that observed theoretically in EuT2Ga8 (T = Co, Rh, Ir).24 Resistivity. Typical zero-field resistivity as a function of the temperature for both the Eu and Sr analogues is shown in Figure 4. We observe significant variability in the residual

∫0

T / ΘD

x 4 dx + KT 3 (e x − 1)(1 − e−x) (1)

where ρ0 is the residual resistivity, ΘD is the Debye temperature, A and K are scale factors, and x represents all possible phonon energies in the system. The abrupt drop in resistivity at ∼5 K for EuIr2In8 is caused by AFM order and is present in all measured crystals. When the Eu2+ magnetic moments order, they reduce the amount of electron scattering off of magnetic moments by reducing the disorder of the system. Above this ordering temperature, EuIr2In8 resistivity also exhibits BGM behavior, indicating that electron−phonon scattering dominates electron transport. This further suggests that the localized f-electron moments on Eu atoms do not interact strongly with conduction electrons and that electron−electron interactions do not make a significant contribution to the resistivity. Further details of BGM fits can be found in the Supporting Information. Magnetic Behavior. Magnetization as a function of temperature for EuIr2In8 and SrIr2In8 can be found in Figure 5. SrIr2In8 is diamagnetic with a magnetization value of −0.00872(7) emu mol−1 Oe−1 at 300 K. This behavior is a bit unexpected because metals normally exhibit temperatureindependent Pauli paramagnetism arising from delocalized conduction electrons. In this case, the combination of a diamagnetic contribution from the large number of core Ir

Figure 4. (a) Resistivity as a function of temperature for typical EuIr2In8 and SrIr2In8 single crystals measured along the a axis. The fits shown are to the BGM equation discussed in the main text. (b) Lowtemperature detail of the resistivity of EuIr2In8 at a variety of fields. The abrupt drop at 5 K is a consequence of AFM order.

resistivity ratio and residual resistivity from crystal to crystal, which is caused by differences in crystal quality. The resistivity increases as the temperature increases for both compounds, as expected for metals. SrIr2In8 follows Bloch−Grüneisen−Mott (BGM) behavior (eq 1) over the entire temperature range, as expected for a normal metal without any strong electron correlations. The resistivity as a function of the temperature in BGM metals follows the equation

Figure 5. (a) Magnetization as a function of temperature for EuIr2In8 and SrIr2In8, with a modified Curie−Weiss fit to the data. (b) Temperature-dependent magnetization at a variety of fields at low temperature. The second magnetic transition is most obvious in the 0.5 and 1 T curves. D

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

in heat capacity measurements is very small (see below), suggesting that this is a spin reorientation from one AFM state to another that causes only a small change in the overall energy of the system. This indicates that EuIr2In8 can adopt two magnetic states that are nearly identical in energy such that a small change in the applied field stabilizes one over the other. Further experiments are necessary to more thoroughly understand these two magnetic ground states and their relationship with one another, although this may be complicated by the large neutron absorption cross section of Eu atoms. These two transitions are likely very similar in character to those observed in EuRh2X8 (X = Ga, In), which were attributed to very small crystal-field splitting and unusual exchange interactions between the Eu magnetic moments.27 It should be noted that, because a collection of small crystals with random orientation with respect to the applied field were used for these measurements, the obtained field dependence of the AFM and suggested spin reorientation transitions in Figure 6 should be considered as an indication of average behavior. Additional measurements on oriented single-crystal samples would be needed to obtain anisotropic H−T phase diagrams. Heat Capacity. Zero-field heat capacity data between 1.8 and 50 K for both title compounds are shown in Figure 7. Above ∼8 K, both materials are well described by the Debye model:

electrons and Landau diamagnetism likely overwhelms the Pauli paramagnetism, yielding overall diamagnetic behavior. The bulk diamagnetism in SrIr2In8 also confirms that the Ir atoms carry no magnetic moment. EuIr2In8 exhibits modified Curie−Weiss behavior above 50 K. This fit yields a Curie constant of C = 7.14(2) emu K mol−1, corresponding to a magnetic moment of 7.55(2) μB per Eu atom, a bit smaller than the 7.93 moment expected for spin-only free Eu2+ ions. This suggests that the Ir atoms carry no local magnetic moment, as is the case in the Sr analogue, and also raises the possibility that the Eu atoms are mixed valent, with a small fraction of them adopting the Eu3+ oxidation state, for which no magnetic moment is expected. In addition to the Eu moments, the χ0 term is −0.01096(7) emu mol−1 Oe−1, indicating that temperature-independent diamagnetic contributions overwhelm the Pauli paramagnetism, presumably for the same reasons as those in the the case of the Sr analogue discussed above and similar to that observed in BaIr2In9.39 The Weiss constant is −2.1(3) K, indicating that weak AFM correlations dominate at high temperature. The first AFM ordering transition, at 5 K, is consistent with this result. At some applied fields, a second transition occurs at lower temperature, suggesting a complicated magnetic phase diagram at low temperature. In order to further investigate the magnetic ordering observed in this compound at low temperature, we measured magnetization as a function of both the field and temperature. Magnetization as a function of temperature at a variety of fields is shown in Figure 5b. As expected for an AFM material, increasing the applied field suppresses the temperature of the first AFM transition, and by 7 T, the transition is below 1.8 K, the lowest temperature of our measurement. At intermediate fields, however, a second transition occurs at lower temperatures, first appearing at 0.2 T, occurring at the highest temperature between 1 and 2 T and finally moving below the base temperature of our measurement before 4 T. A temperature versus field phase diagram is shown in Figure 6,

⎛ T ⎞3 Cp = γT + n × 9R ⎜ ⎟ ⎝ ΘD ⎠

∫0

T / ΘD

x 4 dx (e x − 1)2

(2)

where γ is the Sommerfeld coefficient, ΘD is the Debye temperature, x represents the available phonon energies in the system, and n is related to the number of atoms in the unit cell. The linear term describes the electronic contribution, while the cubic term describes the phonon contribution. SrIr2In8 behaves as expected for a normal metal. The Sommerfeld coefficient of 18.0(2) mJ mol−1 K−2 is well within the range of values typically observed for normal intermetallic compounds. For comparison, γ = 6.85 mJ mol−1 K−2 for Yb3Ga7Ge342 and ranges between 12.8 and 30.2 mJ mol−1 K−2 for RFe2 (R = Gd, Tb, Dy, Ho, Er, Lu).43 The Debye temperature is 188(2) K, a typical value for heavy-atom intermetallic compounds with relatively soft lattice modes. Measurements of EuIr2In8, on the other hand, exhibit significantly different behavior. The phase transition at ∼5 K corresponds to the AFM ordering transition observed in both resistivity and magnetization measurements. Precise determinations of the Debye temperature present a significant challenge because of the large discontinuity at low temperature. It is likely slightly lower than the Debye temperature of SrIr2In8 because Eu is heavier than Sr, but precise determination proved impossible. Consequently, we used the approximation in eq 3 to fit the data:

Figure 6. Sketched magnetic phase diagram of EuIr2In8. Filled symbols are based on heat capacity measurements, while open symbols are based on magnetization measurements. Lines are guides to the eye.

Cp = γT + βT 3

(3)

At temperatures sufficiently below the Debye temperature, the simple coefficient β can approximate the integral in eq 2. The most significant difference between the two analogues beyond the obvious magnetic ordering transition is that for EuIr2In8 the Sommerfeld coefficient above the ordering transition is 118(3) mJ mol−1 K−2, an order of magnitude larger than that observed in the Sr analogue. Reliable determination of the Sommerfeld coefficient below the transition is impossible without measurement to significantly lower temperatures to rule out the

and it clearly indicates a dome, labeled AFM2, at lower temperatures than the first AFM transition between roughly 0.2 and 3 T. While the exact location of this dome varies slightly from sample to sample, all measured samples show this behavior. Magnetization measured as a function of the field at 2 K (Figure S1 in the Supporting Information) shows only a very small kink in the curve slightly above 1 T, which we attribute to the second ordering transition. The signature of this transition E

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 7. (a) Temperature dependence of the heat capacity of EuIr2In8 and SrIr2In8 at zero field. (b) C/T versus T2 plot for both compounds. The fits shown are linear fits to estimate the Sommerfeld coefficient. (c) Temperature dependence of the EuIr2In8 heat capacity at a variety of fields. The signatures of the second magnetic phase transition are most easily seen in the 1 and 2.5 T curves.

temperatures is an incorrect approximation of the true heat capacity, or both. It is possible that the enhanced Sommerfeld coefficient arises from Eu2+/Eu3+ mixed valence. At low temperatures, contraction of the lattice can induce the reduction of Eu2+ to the smaller Eu3+, releasing a conduction electron. This can adjust the Fermi level to a location with very high density of states or high degeneracy, leading to an enhanced Sommerfeld coefficient. The enhancement could alternatively be due to magnetic fluctuations above the magnetic ordering temperature that influence the carrier effective mass. Both of these scenarios are plausible, and therefore the exact origin of the enhancement in γ in the Eu analogue relative to the Sr analogue remains unclear. To investigate the change in the ground state as a function of the field, we measured Cp as a function of temperature at a variety of fields, as shown in Figure 7c. Measurements at 1 and 2.5 T indicate very small anomalies near the temperature of spin reorientation observed by magnetization measurements. The very small magnitude of this anomaly indicates that the spin reorientation has only a very minor impact on the overall energy of the system, suggesting that these two ground states are very close to one another in energy and are likely the result of very finely balanced competing magnetic exchange interactions. A detailed further investigation is required to more definitively understand the difference between these two states.

possibility of additional transitions. To further understand the origin of the enhanced Sommerfeld coefficient in the Eu analogue, we estimated the magnetic contribution to the EuIr2In8 heat capacity, Cmag. To isolate the magnetic contribution, we subtracted the “nonmagnetic” part of Cp, obtained by taking the heat capacity of SrIr2In8 and adding 13% to each point in order to adjust for the different cell volumes and mass differences between Sr and Eu. From Cmag, we calculated the change in the magnetic entropy, Smag, using the following equation: Smag =

∫0

T

Cmag(T ′) dT ′ T′

(4)

Cmag and Smag are plotted as a function of temperature in Figure 8. The Smag calculation assumed a linear extrapolation to zero

■

CONCLUSIONS The new orthorhombic, CeFe2Al8-type intermetallic compounds, MIr2In8 (M = Eu, Sr), grow as large single crystals in good yield using In flux. SrIr2In8 is a diamagnetic metal, and low-temperature heat capacity measurements indicate a Sommerfeld coefficient of 18.0(2) mJ mol−1 K−2, typical for polar intermetallic compounds. EuIr2In8 exhibits AFM ordering of the Eu2+ magnetic moments at 5 K, a transition that is suppressed at higher magnetic fields. Between 0.2 and 3 T, a spin reorientation occurs below this AFM ordering, suggesting that two nearly degenerate ground states dominate the lowtemperature behavior of this material. Heat capacity measurements also suggest that EuIr2In8 exhibits significant enhancement of the Sommerfeld coefficient relative to the nonmagnetic Sr analogue, with γ = 118(3) mJ mol−1 K−2 at temperatures just above the AFM ordering. The interesting interplay of magnetic ground states and evidence of strong electron correlation with enhanced γ make EuIr2In8 an intriguing compound for further

Figure 8. Estimated magnetic component of the heat capacity Cmag and total magnetic entropy change Smag for EuIr2In8. R ln 8 is also plotted. Smag exceeds R ln 8 at around 8 K, indicating that our estimate of Cmag is not accurate, most likely because of the use of a linear extrapolation to zero to estimate the behavior at temperatures below 1.8 K, the lowest temperature of our measurement.

from our lowest temperature point in order to estimate the entropy change below our measurement. The expected magnetic entropy change from magnetic order in the general case is Smag = R ln(2S + 1), where R is the ideal gas constant and S is the spin quantum number. S = 7/2 for Eu2+, so R ln 8 is the maximum entropy change associated with Eu2+ magnetic moments. Our estimate of the magnetic entropy exceeds R ln 8 at around 7 K. This result should not be possible and is likely because either our estimate of the nonmagnetic contribution is a poor one or the linear interpolation to zero at low F

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Nakamura, N.; Sugiyama, K.; Takeuchi, T.; Kindo, K.; Namiki, T.; Aoki, Y.; Sato, H.; O̅ nuki, Y. J. Phys. Soc. Jpn. 2004, 73, 649−655. (10) Sarkar, S.; Gutmann, M. J.; Peter, S. C. Dalton Trans. 2014, 43, 15879−15886. (11) Sarkar, S.; Banerjee, S.; Jana, R.; Siva, R.; Pati, S. K.; Balasubramanian, M.; Peter, S. C. Inorg. Chem. 2015, 54, 10855− 10864. (12) Yarmolyuk, Y. P.; Rychal, R. M.; Zarechnyuk, O. S. Second AllUnion Conference on the Crystal Chemistry of Intermetallic Compounds, Book of Abstracts; American Chemical Society: Washington, DC, 1974. (13) Kolenda, M.; Koterlin, M. D.; Hofmann, M.; Penc, B.; Szytuła, A.; Zygmunt, A.; Ż ukrowski, J. J. Alloys Compd. 2001, 327, 21−26. (14) Czech, E.; Cordier, G.; Schäfer, H. J. Less-Common Met. 1983, 95, 205−211. (15) Manjako, N. B.; Stec, I. N.; Kavich, J. V.; Zarechnjuk, O. S.; Janson, T. I. Dopov. Akad. Nauk Ukr. RSR Ser. B: Geol. Khim. Biol. Nauki 1983, 6, 41−45. (16) Klesnar, H.; Rogl, P. J. Mater. Res. 1991, 6, 53−56. (17) Ma, X.; Chai, P.; Chen, B.; Lochner, E.; Latturner, S. E. J. Solid State Chem. 2015, 229, 181−187. (18) Zarechnjuk, O. S.; Rykhal, R. M.; Korin, V. V. Dopov. Akad. Nauk Ukr. RSR Ser. A: Fiz.-Mater. Techn. Nauki 1980, No. 1, 86−89. (19) Rykhal, R. M.; Zarechnjuk, O. S.; Protasov, V. C. Dopov. Akad. Nauk Ukr. RSR Ser. A: Fiz-Mater. Techn. Nauki 1985, 12, 73−75. (20) Manyako, N. B.; Yanson, T. I.; Bodak, O. I.; Cerny, R.; Yvon, K. Z. Kristallogr. - Cryst. Mater. 1996, 211, 216. (21) Watkins-Curry, P.; Burnett, J. V.; Samanta, T.; Young, D. P.; Stadler, S.; Chan, J. Y. Cryst. Growth Des. 2015, 15, 3293−3298. (22) Manjako, N. B.; Janson, T. I.; Zarechnjuk, O. S. Izv. Akad. Nauk SSSR Metally 1988, 3, 185−189. (23) Wu, X.; Francisco, M.; Rak, Z.; Bakas, T.; Mahanti, S. D.; Kanatzidis, M. G. J. Solid State Chem. 2008, 181, 3269−3277. (24) Sichevych, O.; Kohout, M.; Schnelle, W.; Borrmann, H.; Cardoso-Gil, R.; Schmidt, M.; Burkhardt, U.; Grin, Y. Inorg. Chem. 2009, 48, 6261−6270. (25) Sichevych, O. M.; Lapunova, R. V.; Grin, Y.; Yarmolujk, Y. P. Izv. Akad. Nauk SSSR Metally 1985, 6, 117−118. (26) Kodera, Y.; Chan, H.-P.; Doi, K. Phys. Med. Biol. 1983, 28, 841− 852. (27) Fritsch, V.; Bobev, S.; Moreno, N. O.; Fisk, Z.; Thompson, J. D.; Sarrao, J. L. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 70, 052410. (28) Muts, I. R.; Zaremba, V. I.; Pöttgen, R. Z. Anorg. Allg. Chem. 2007, 633, 2234−2237. (29) Treadwell, L. J.; Watkins-Curry, P.; McAlpin, J. D.; Rebar, D. J.; Hebert, J. K.; DiTusa, J. F.; Chan, J. Y. Inorg. Chem. 2015, 54, 963− 968. (30) Ghosh, S.; Strydom, A. M. Acta Phys. Pol., A 2012, 121, 1082− 1084. (31) X-AREA, version 1.39; Stoe & Cie GmbH: Darmstadt, Germany, 2006. (32) Sheldrick, G. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (33) Gelato, L. M.; Parthe, E. J. Appl. Crystallogr. 1987, 20, 139−143. (34) CRC Handbook of Chemistry and Physics, 86th ed.; Taylor and Francis: Boca Raton, FL, 2006. (35) Canfield, P. C.; Fisk, Z. Philos. Mag. B 1992, 65, 1117−1123. (36) Kanatzidis, M. G.; Pöttgen, R.; Jeitschko, W. Angew. Chem., Int. Ed. 2005, 44, 6996−7023. (37) Phelan, W. A.; Menard, M. C.; Kangas, M. J.; McCandless, G. T.; Drake, B. L.; Chan, J. Y. Chem. Mater. 2012, 24, 409−420. (38) Hoffmann, R.-D.; Pöttgen, R. Chem. - Eur. J. 2000, 6, 600−607. (39) Calta, N. P.; Han, F.; Kanatzidis, M. G. Inorg. Chem. 2015, 54, 8794−8799. (40) Hoffmann, R.-D.; Rodewald, U. C.; Pöttgen, R. Z. Naturforsch., B: J. Chem. Sci. 1999, 54, 38−44. (41) Muts, I. R.; Zaremba, V. I.; Pöttgen, R. Z. Anorg. Allg. Chem. 2007, 633, 2234−2237.

study to understand the complex physics that arise in this structure type.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b00059. Further information about magnetization and resistivity measurements (PDF) Single-crystal X-ray diffraction data in CIF format (CIF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Work at the Ames Laboratory (S.L.B.) was supported by the U.S. Department of Energy (DOE), Basic Energy Sciences, Division of Materials Sciences and Engineering, under Contract DE-AC02-07CH11358. A.P.R. acknowledges support through an Undergraduate Research Grant administered by the Northwestern University Office of Undergraduate Research. This work made use of the EPIC facility (NUANCE Center, Northwestern University), which has received support from the MRSEC program (NSF Grant DMR-1121262) at the Materials Research Center, the International Institute for Nanotechnology (IIN), and the State of Illinois, through the IIN. The work at Argonne National Laboratory was supported by Materials Sciences and Engineerinng Division, Basic Energy Sciences, Office of Science, U.S. DOE. We acknowledge Prof. Danna Freedman and Samantha Clarke for assistance with magnetic measurements, which were supported by Northwestern University’s IIN and the State of Illinois Department of Commerce and Economic Opportunity under Award 10203031.

■

REFERENCES

(1) Petrovic, C.; Pagliuso, P. G.; Hundley, M. F.; Movshovich, R.; Sarrao, J. L.; Thompson, J. D.; Fisk, Z.; Monthoux, P. J. Phys.: Condens. Matter 2001, 13, L337−L342. (2) Bauer, E.; Hilscher, G.; Michor, H.; Paul, C.; Scheidt, E. W.; Gribanov, A.; Seropegin, Y.; Noël, H.; Sigrist, M.; Rogl, P. Phys. Rev. Lett. 2004, 92, 027003. (3) Hiranaka, Y.; Nakamura, A.; Hedo, M.; Takeuchi, T.; Mori, A.; Hirose, Y.; Mitamura, K.; Sugiyama, K.; Hagiwara, M.; Nakama, T.; O̅ nuki, Y. J. Phys. Soc. Jpn. 2013, 82, 083708. (4) Francisco, M. C.; Malliakas, C. D.; Macaluso, R. T.; Prestigiacomo, J.; Haldolaarachchige, N.; Adams, P. W.; Young, D. P.; Jia, Y.; Claus, H.; Gray, K. E.; Kanatzidis, M. G. J. Am. Chem. Soc. 2012, 134, 12998−13009. (5) Han, F.; Wan, X.; Phelan, D.; Stoumpos, C. C.; Sturza, M.; Malliakas, C. D.; Li, Q. a.; Han, T.-H.; Zhao, Q.; Chung, D. Y.; Kanatzidis, M. G. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 92, 045112. (6) Kondo, J. Prog. Theor. Phys. 1964, 32, 37−49. (7) Stewart, G. R. Rev. Mod. Phys. 2001, 73, 797−855. (8) Chen, G.; Ohara, S.; Hedo, M.; Uwatoko, Y.; Saito, K.; Sorai, M.; Sakamoto, I. J. Phys. Soc. Jpn. 2002, 71, 2836−2838. (9) Ueda, T.; Shishido, H.; Hashimoto, S.; Okubo, T.; Yamada, M.; Inada, Y.; Settai, R.; Harima, H.; Galatanu, A.; Yamamoto, E.; G

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (42) Peter, S. C.; Malliakas, C. D.; Nakotte, H.; Kothapilli, K.; Rayaprol, S.; Schultz, A. J.; Kanatzidis, M. G. J. Solid State Chem. 2012, 187, 200−207. (43) Butera, R. A.; Clinton, T. J.; Moldovan, A. G.; Sankar, S. G.; Gschneidner, K. A. J. Appl. Phys. 1979, 50, 7492−7494.

H

DOI: 10.1021/acs.inorgchem.6b00059 Inorg. Chem. XXXX, XXX, XXX−XXX