Anisotropic Near-Zero Thermal Expansion in REAgxGa4–x (RE = La

Oct 3, 2018 - E; Energy & Fuels · Environmental Science & Technology · Environmental .... The anisotropic zero thermal expansion (ZTE) behavior was ob...
0 downloads 0 Views 8MB Size
Article pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

Anisotropic Near-Zero Thermal Expansion in REAgxGa4−x (RE = La− Nd, Sm, Eu, and Yb) Induced by Structural Reorganization Vidyanshu Mishra, Udumula Subbarao, Soumyabrata Roy, Saurav Ch. Sarma, Dundappa Mumbaraddi, Shreya Sarkar, and Sebastian C. Peter*

Inorg. Chem. Downloaded from pubs.acs.org by UNIV OF LOUISIANA AT LAFAYETTE on 10/03/18. For personal use only.

New Chemistry Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Bengaluru 560064, India S Supporting Information *

ABSTRACT: In this work, we have discovered the anisotropic near-zero thermal expansion (NZTE) behavior in a family of compounds REAgxGa4−x (RE = La−Nd, Sm, Eu, and Yb). The compounds adopt the CeAl2Ga2 structure type and were obtained as single crystals in high yield by metal flux growth technique using gallium as active flux. Temperaturedependent single crystal X-ray diffraction suggests that all the compounds exhibit near zero thermal expansion along c direction in the temperature range of 100−450 K. Temperature-dependent X-ray absorption near-edge spectroscopic study confirmed ZTE behavior is due to the geometrical features associated within the crystal structure. The anisotropic NZTE behavior was further established by anisotropic magnetic measurements on selected single crystals. The atomic displacement parameters, apparent bond lengths, bond angles, and structural distortion with respect to the temperature reveal that geometric features associated with the structural distortion cause the anisotropic NZTE along c-direction. The preliminary magnetic studies suggest all the compounds are paramagnetic at room temperature except LaAgGa3. Electrical resistivity study reveals that compounds from this series are metallic in nature.

1. INTRODUCTION Thermal expansion (TE) property and differences in thermal expansion coefficients of materials are important in many technological applications ranging from electronic devices to space research. It is a common observation that most of the materials expand upon an increase in temperature, which is known as positive thermal expansion (PTE). The materials with temperature-dependent zero thermal expansion (ZTE) or negative thermal expansion (NTE) are scarce and mostly occur over a relatively narrow temperature window. PTE can be explained well on the basis of population of the higher energy vibrational levels which cause the bond expansion due to asymmetric nature of a typical interatomic potential.1 However, many devices that are expected to operate under extreme conditions, for example, low or very high temperature, need to be engineered without any volume expansion and degradation. The materials with NTE or ZTE are highly desirable for both exactitude engineering of various components for nanodevices and of complex bulk systems. The materials demonstrating regular PTE can be tuned with NTE materials to fabricate composites in order to attain an overall ZTE. The ZTE feature hinders materials undergoing thermal shock on hasty cooling or heating. Usually, the NTE feature in materials is crystallographic direction specific. Most of these materials contain elemental silicon and germanium,2 titania− silica glasses,3 carbon fibers, Kevlar and Invar alloys,4 ZrW2O8 oxides,5 and certain molecular networks.6 The materials Ag3[Co(CN)6],7 Sm3C60,8 BiNiO3,9 MnF2,10 CuO,10 Mn3Cu1−xGexN11 are also well-known compounds exhibiting NTE. © XXXX American Chemical Society

The materials with anisotropic mechanical, magnetic, optical, and elastic properties are well-exploited, but the materials with intrinsic anisotropic thermal properties are very rare. Rare earth based intermetallics are noteworthy in this direction; YbGaGe,2 Yb4TGe8 (T = Cr, Ni, and Ag),12 and Yb8Ge3Sb513 are a few examples reported with anisotropic TE and noncubic systems with intrinsic anisotropic thermal properties (PTE, ZTE, or NTE) need to be discovered. The measure of expansion in a material is given by its coefficient of thermal expansion (CTE). Generally, the compounds having volume CTE in the range of αv ≈ 10−5 K−1 and linear CTE αl < |2| × 10−6 K−1 are considered to show very low thermal expansion.14,15 There are several aspects which give rise zero and or negative thermal expansion such as structural flexibility in the compounds like LiAlSiO416,17 and α- and β-ZrW2O8,5 magnetovolume effect in the compounds like Fe−Ni and Ni− Cu alloys,18 Tm2Fe16Cr,19 and Dy2Fe17−xMnx,20 and phase transition in compounds like ZrV2O7.21 Herein, we report the αv ≈ 4.6−6.7 × 10−5 K−1 and αl ≈ −1.3−6.6 × 10−6 K−1 for REAgxGa4−x (RE = Ce, Nd, Yb, and Eu) systems in the range of 100−500 K which originate due to structural flexibility. We have chosen the compounds with the general formula RETX3 (RE = rare earth and T = transition metals, and X = Al, Si, Ga, Ge, and Sb) to study TE behavior. These compounds show interesting physical phenomena due to their adaptation associated with different structure types.22−34 These structural Received: June 22, 2018

A

DOI: 10.1021/acs.inorgchem.8b01650 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. SEM images of typical REAgxGa4−x (RE = La−Nd, Sm, Eu, and Yb) single crystals grown by metal flux technique. product were detected. Residual gallium was removed by immersing the products in 2−4 M solution of iodine in dimethylformamide (DMF) over 3 h followed by sonication at room temperature followed by drying with acetone. The gray plate-like crystals of REAgxGa4−x were 2−4 mm in size obtained in high yield (>90%). Several crystals, which were grown as shining as metallic silver plates, were carefully selected for the elemental analysis, single crystal XRD analysis, and anisotropic magnetic measurements. The structure refinement using single crystal XRD resulted in nonstoichiometric composition, which has been taken for the synthesis of the polycrystalline sample in large quantity using the following methods: 2.2.2. Arc Melting Technique. The compounds REAgxGa4−x (RE = La−Nd, and Sm) were synthesized using Edmund Bühler GmbH MAM-1 arc melter. The precursors were weighed in the respective stoichiometric ratio (obtained after single crystal refinement) in the Ar atmosphere and repeatedly arc melted (3 times) for maintaining homogeneity. The yield of final products were found to be almost 99.99%. The samples were obtained in the form of polycrystalline ingots. After crushing the ingots into a fine powder, samples were used for further characterization. 2.2.3. High-Frequency Induction Heating. The compounds EuAgGa3 and YbAgGa3 were synthesized by high-frequency induction heating. For the synthesis of EuAgGa3, the precursors europium, silver, and gallium were taken in respective atomic ratios (obtained after single crystal refinement) and kept inside a niobium ampule under an argon atmosphere and subsequently sealed in Edmund Bühler GmbH MAM-1 arc melter. The sealed ampule was then kept inside a water-cooled sample chamber of Easy Heat induction heating system, Model 7590). The reaction was carried out at 183 A (ca. 1200−1350 K) for 50 min. Finally, the reaction was rapidly cooled to room temperature by reducing the current to 0 A. The light gray polycrystalline product EuAgGa3 was obtained by cutting the niobium tube. No side products or reactions with the ampule could be detected. Several tests were run for confirming the stability of EuAgGa3 at 2 weeks interval for three months and was found to be stable at standard conditions. No more than 1% weight loss was observed in the product. For the synthesis of the YbAgGa 3 compound, the metals ytterbium, silver, and gallium were mixed in a tantalum ampule. The reaction was carried out with the procedure similar to that for EuAgGa3, but applying 170 A current. No side products were detected. The compound YbAgGa3 was found to be stable for several months. The obtained samples were used for crystal structure determination and physical property studies.

variations in these types are due to the various parameters including valence fluctuations or other electronic transition and or unique geometrical features associated with the structure. These observations motivated us to investigate the temperature-dependent thermal properties of these compounds, especially on REAgxGa4−x compounds. The crystal structure of these compounds was proposed as tetragonal (SG: I4/mmm) earlier by Grin et al.35 using powder XRD data. However, our single crystal XRD data suggest that they showed unusual thermal parameters in different crystallographic directions. The temperature-dependent single crystal XRD on these compounds showed ZTE along c direction and PTE along ab plane due to the unique geometrical arrangement of atoms in crystal structure in the certain temperature range. A detailed temperature-dependent single crystal XRD, anisotropic magnetic studies on single crystals, and X-ray absorption near-edge spectroscopy (XANES) studies confirm that the geometrical adjustments in the crystal structure along ab plane and c-direction cause the ZTE.

2. EXPERIMENTAL SECTION 2.1. Reagents. The RE metals (La−Nd, Sm, Eu, and Yb pieces, 99.99%, Alfa Aesar), Ag (shots, 99.99%, Sigma-Aldrich) and Ga (pieces, 99.99%, Alfa Aesar) were used as purchased without any further purification. The metals were stored and cut in small pieces inside the glovebox. 2.2. Synthesis. 2.2.1. Single Crystal Growth. Single crystals were grown (Figure S1) through the metal flux growth technique using gallium as active flux. Plate-shaped single crystals of REAgxGa4−x were obtained by combining RE metal (0.3 g), Ga pieces (2 g), and Ag shots (0.4 g) in a niobium ampule. The ampule was inserted in a 13 mm fused silica tube and was flame-sealed under a vacuum of 10−5 Torr to avoid any oxidation during heating. The reactants were heated to 1273 K over 10 h to form the melt and then soaked for 2 h. Next, the melt was cooled down to 1073 K in 2 h. To get the desired phase formed, the reaction was maintained at this temperature for 48 h. Finally, to grow the crystals, the reaction was allowed to cool gradually to 300 K over 48 h. Since Ga was used as a flux, the product was heated at 673 K and subsequently centrifuged at this temperature through a coarse frit to isolate excess Ga from the product. No reactions between the niobium ampule and starting materials or B

DOI: 10.1021/acs.inorgchem.8b01650 Inorg. Chem. XXXX, XXX, XXX−XXX

C

a

402 0.05 × 0.08 × 0.07 3.761−24.918°

404 0.06 × 0.05 × 0.09 3.757−24.990°

397 0.07 × 0.05 × 0.08 3.75−31.28°

a = 4.3633(4) c = 10.8576(10) 206.71(4) 7.357 39.179

EuAg0.72(2)Ga3.28(2) 458.31

406 0.10 × 0.05 × 0.08 3.711−24.957°

a = 4.2422(4) c = 10.9824(17) 197.64(5) 7.962 49.020

YbAg0.57(2)Ga3.43(2) 473.86

102/0/10 1.393 Robs = 0.0345, wRobs = 0.0813 Rall = 0.0345, wRall = 0.0813 0.076(11) 1.231 and −3.186

70/0/10

1.125 Robs = 0.0483, wRobs = 0.1059 Rall = 0.0497, wRall = 0.1082 0.015(5) 2.825 and −1.457

80.3%

1.188 Robs = 0.0525, wRobs = 0.1409 Rall= 0.0527, wRall = 0.1409 0.014(4) 6.411 and −3.277

129/0/10

100%

1.088 Robs = 0.0206, wRobs = 0.0517 Rall= 0.0212, wRall = 0.0530 0.016(2) 1.613 and −1.612

1.488 Robs = 0.0666, wRobs = 0.1793 Rall= 0.0666, wRall = 0.1793 0.048(2) 4.246 and −3.147

100% 100% full-matrix least-squares on F2 71/0/10 67/0/10

1.187 Robs = 0.0214, wRobs = 0.0441 Rall = 0.0214, wRall = 0.0441 0.105(7) 2.120 and −1.693

104/0/10

0.919 Robs = 0.0299, wRobs= 0.0779 Rall = 0.0299, wRall = 0.0779 0.020(3) 2.311 and −1.556

71/0/10

100%

83.1%

390 0.07 × 0.05 × 0.08 3.777−33.233°

a = 4.2133(5) c = 10.847(2) 192.56(6) 8.061 40.466

SmAg0.97(1)Ga3.03(1) 467.38

92.1%

391 0.10 × 0.07 × 0.09 3.77−30.78°

388 0.10 × 0.07 × 0.04 5.041−24.864°

a = 4.2835(5) c = 10.7889(15) 197.96(5) 7.519 37.310

NdAg1.05(1) Ga2.95(1) 463.18 0.71073 Å tetragonal I4/mmm a = 4.2660(6) c = 10.835(2) 197.19(7) 7.801 37.698

−6 ≤ h ≤ 6, −5 ≤ k ≤ 6, −5 ≤ h ≤ 4, −5 ≤ k ≤ 5, −15 ≤ l ≤ 12 −12 ≤ l ≤ 12 1108 664 104 [Rint = 0.0721] 71 [Rint = 0.0788]

a = 4.3106(3) c = 10.803(1) 200.74(3) 7.456 35.841

a = 4.3637(17) c = 10.768(6) 205.0(2) 7.265 34.451

PrAg0.75(2) Ga3.25(2) 448.21

−5 ≤ h ≤ 4, −5 ≤ k ≤ 5, −6 ≤ h ≤ 5, −6 ≤ k ≤ 6, −6 ≤ h ≤ 5, −6 ≤ k ≤ 5, −4 ≤ h ≤ 5, −5 ≤ k ≤ 5, −4 ≤ h ≤ 4, −4 ≤ k ≤ 4, −12 ≤ l ≤ 6 −14 ≤ l ≤ 15 −16 ≤ l ≤ 11 −12 ≤ l ≤ 10 −12 ≤ l ≤ 12 652 1035 915 773 645 70 [Rint = 0.1443] 102 [Rint = 0.0644] 129 [Rint = 0.0944] 71 [Rint = 0.0718] 67 [Rint = 0.0995]

CeAg0.83(2)Ga3.17(2) 450.66

LaAg0.80(2)Ga3.20(2) 448.50

R = Σ||F0| − |Fc||/Σ|F0| for F02 > 2σ(F02), wR = {Σ[w(|F0|2 − |Fc|2)2]/Σ[w(|F0|4)]}1/2 and calcd w = 1/[σ2(F0 2) + (0.0359p)2 + 6.1794p], where p = (F02+2 Fc2)/3.

ext. coefficient largest diff. peak and hole (e·Å−3)

reflections collected independent reflections completeness ref. method data/restraints/ parameters goodness-of-fit final R indicesa [I > 2σ(I)] R indices [all data]

volume (Å3) density (g/cm3) absorption coefficient (mm−1) F(000) crystal size (mm3) θ range for data collection index ranges

empirical formula formula weight wavelength crystal system space group unit cell dimensions (Å)

Table 1. Crystal Data and Structure Refinement for REAgxGa4−x (RE = La−Nd, Sm, Eu, and Yb) at 293 K

Inorganic Chemistry Article

DOI: 10.1021/acs.inorgchem.8b01650 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry 2.2.4. Elemental Analysis. The visibly clean surfaces of single crystals obtained from the flux techniques were analyzed by semiquantitative microprobe analyses SEM equipped with Bruker 129 eV EDAX instrument with an accelerating voltage of 20 kV in 90 s accumulation time. The typical image of a plate-like crystal of REAgxGa4−x (RE = La−Nd, Sm−Gd, and Yb) grown from the flux synthesis is shown in Figure 1. The atomic composition of each crystal was found to be close to 1:1:3 by the analysis, which is in good agreement with the single crystal XRD data (Figure S2). 2.2.5. Powder XRD. The phase identity and purity of the REAgxGa4−x (RE = La−Nd, Sm−Gd, and Yb) powder samples were determined using the Bruker D8 Discover diffractometer, set at Cu Kα radiation (λ = 1.5406 Å) over the angular range of 20° ≤ 2θ ≤ 80°, with a step size of 0.02° at room temperature (297 K) calibrated against corundum standard. The comparison between the experimentally observed and the simulated pattern generated from the single crystal XRD refinement (Figure S3.1) confirmed the phase purity. The lattice parameters were confirmed by Pawley profile fitting for powder samples of CeAgGa3 and YbAgGa3 (Figure S3.2) using TOPAS software.36 2.2.6. Single Crystal XRD. Single crystal XRD data of REAgxGa4−x (RE = La−Nd, Sm−Gd, Yb) were first collected at room temperature on selected plate-shaped single crystals using a Bruker Smart−CCD diffractometer equipped with a usual focus, 2.4 kW sealed tube X-ray source with graphite monochromatized Mo Kα radiation (λ = 0.7107 Å) operating at 50 kV and 30 mA, with ω scan mode. Temperaturedependent data were collected in the range of 100−500 K using a Bruker D8 Venture diffractometer (Photon CMOS detector) equipped with a micro focus, 2.4 kW sealed tube X-ray source with graphite monochromatic Mo Kα radiation (λ = 0.71073 Å) operating at 50 kV and 30 mA, with ω scan mode. Commercially available super glue was used to mount the very tiny plate shaped single crystals on a thin glass (∼0.1 mm) fiber. A full sphere having 60 frames was acquired up to 73.28° in 2θ. The individual frames were collected with steps of 0.50° and an exposure time of 10 s per frame. SAINT37 and SADABS38 programs were used to integrate the diffraction profiles and absorption correction, respectively. The analyses of systematic absences led us to decide that systems were crystallizing in centrosymmetric tetragonal space group I4/mmm. Platon program in the WinGx system, version 1.80.0539 was employed to check the additional symmetry. 2.2.7. Magnetic Measurements. A Quantum Design MPMSSQUID magnetometer was used to perform the magnetic measurements. For the measurements, polycrystals obtained from arc melting and induction heating were used after grinding followed by verification of phase identity and purity by powder XRD. Under the applied magnetic field (H) of 1 kOe, the temperature-dependent data were collected for Field Cooled mode between 2 and 300 K. Magnetization data were also collected for REAgxGa4−x at 2 and 300 K with field sweeping from −60 000 to 60 000 Oe (Figure S4). The magnetic measurements were also performed anisotropically on single crystals of REAgxGa4−x (RE = La−Nd, Sm−Gd, Yb) grown using metal flux technique. The measurement parameters were the same as were for the polycrystalline samples. 2.2.8. Electrical Resistivity. The conventional ac four probe setup was used to perform the resistivity measurements in the 1T field on REAgxGa4−x pellets. A strongly conducting silver epoxy paste was adhered to give contacts on the pressed pellets with four very thin copper wires. Commercial QD-PPMS was used to collect the data in the range of 3−300 K. Reproducible results were obtained for several batches. 2.2.9. X-ray Absorption near Edge Spectroscopy. XANES experiments at various temperatures on REAgxGa4−x (RE = La, Ce, Nd, Eu, and Yb) were performed at PETRA III, P65 beamline of DESY, Germany. Measurements at the Eu LIII edge and Yb LIII edge were performed in transmission mode, while measurements at the Ce LIII, Nd LIII, and La LIII edge were performed in fluorescence mode at ambient pressure using gas ionization chambers to monitor the incident and transmitted X-ray intensities. A Si [111] double crystal monochromator was used to obtain the monochromatic X-rays,

calibrated by defining the inflection point (first derivative maxima) of Cu foil as 8980.5 eV; ionization chambers filled with suitable gases were used to record the incident (I0) and transmitted (It) photon intensities simultaneously at different edges. A PIPS detector was used for the fluorescent signals. Pellets for the measurements were made by homogeneously mixing the sample with an inert cellulose matrix to have an X-ray absorption edge jump close to 1.

3. RESULTS AND DISCUSSIONS 3.1. Structure Refinement. SHELXS 97 was used to solve the crystal structures, and SHELXL was used to refine them with anisotropic atomic displacement parameters (ADPs) for all atoms employing full matrix least-squares method.40 The crystallographic parameters in the centrosymmetric space group (I4/mmm) of the previously reported compound CeAgGa341 were taken in the initial step of the refinement to solve and refine the structures of all other compounds REAgxGa4−x (RE = La−Nd, Sm, Eu, and Yb). To confirm the precise composition, a discrete series of least-squares cycles was used to refine the occupancy parameters. The refinement value (