Metal Flux Crystal Growth Technique in the Determination ofOrdered Superstructure in EuInGePublished as part of the Crystal Growth & Design virtual special issue In Honor of Prof. G. R. Desiraju

Udumula Subbarao,† Ashly Sebastian,† Sudhindra Rayaprol,‡ C. S. Yadav,§,⊥ Axel Svane,#

G. Vaitheeswaran,∥ and Sebastian C. Peter*,†

†New Chemistry Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur, Bangalore, 560064-India‡UGC-DAE Consortium for Scientific Research, Mumbai Centre, BARC, R-5 Shed, Trombay, Mumbai-400085, India§DCMP & MS, Tata Institute of Fundamental Research, Mumbai-400005, India#Department of Physics and Astronomy, Aarhus University, DK-8000 Aarhus C, Denmark∥Advanced Centre of Research in High Energy Materials (ACRHEM), University of Hyderabad, Prof. C. R. Rao Road, Gachibowli,Hyderabad-500046, India

*S Supporting Information

ABSTRACT: High quality single crystals of EuInGe were grown from thereaction run with excess indium. X-ray diffraction investigations showed thatEuInGe crystallizes with a pronounced subcell structure, superstructure of theThSi2 type: Pnma space group, a = 4.9066(10) Å, b = 3.9834(8) Å and c =15.964(3) Å. However, the powder X-ray pattern reveals weak superstructurereflections, and the inclusion of additional reflections in the analysis points to anew type of structural arrangement, in a monoclinic system, P21/c space group, a= 7.9663(16) Å, b = 4.9119(10) Å, c = 16.465(5) Å, and β = 104.03°.Magnetization measurements carried out as a function of temperature showmultiple magnetic transitions at 13, 25, 44, and 70 K. In the temperature regionabove 100 K, the Curie−Weiss law is followed indicating a paramagnetic state ofthe sample. Magnetic moments deduced from this region suggest europium tobe in a divalent state, which was further confirmed by 151Eu Mossbauerspectroscopic measurements. Experiments were accompanied by first-principles density functional calculations using the full-potential linear muffin-tin orbital method within the local density approximation (LSDA) and including the onsite Coulombinteraction (LSDA+U) for the Eu-f states. The density of states shows a pronounced pseudo gap feature around the Fermi level.The inclusion of a Hubbard U has only a minor effect on the band structure. From the calculated total energies the P21/cstructure is favorable by 25 meV per formula unit when compared to the Pnma subcell structure.

1. INTRODUCTION

Intermetallic compounds are usually synthesized by directreaction of the elements heated in a vacuum or in an inertatmosphere. The required reaction temperatures are generallyhigh, often above 1000 °C necessitating the use of conventionaltechniques such as arc melter or high frequency inductivefurnace. Though these methods have been widely used for thesynthesis of intermetallics, the fabrication of high quality singlecrystals using these techniques is very limited. On a fewoccasions single crystals can be obtained by annealing orquenching the product, but in most situations only powdersamples are obtained. This situation often makes crystalstructure determination difficult and limits proper character-ization. In a recent review, it has been demonstrated thatmolten metal fluxes represent an excellent alternative to theconventional synthetic methods for the exploratory synthesis ofnew rare earth intermetallic compounds, as well as single crystalgrowth of already reported compounds.1 In general, aluminum,

gallium, and indium metal fluxes have been exploited widely asa synthetic medium for the synthesis of new intermetalliccompounds.1−29

Recently, new rare earth based intermetallic compounds havebeen synthesized using indium as active and nonactive metalflux. The rare-earth metals chosen were particularly those whichcan show mixed valent behavior such as Ce, Eu, and Yb.30−34

Every new compound discovered displayed a novel situation intheir physical properties. Some notable examples are zerothermal expansion behavior in Yb4TGe8 (T = Cr, Ni, andAg),35 metamagnetism in Eu2AuGe3,

36 structural phasetransition in RE2AuGe3 (RE = Eu and Yb),36,37 structuraldisorder induced magnetic ordering in YbCu6−xIn6+x (x = 0, 1,2),38 valence fluctuations in Yb4TGe8,

35 low temperature

Received: October 18, 2012Revised: November 23, 2012Published: November 30, 2012

Article

pubs.acs.org/crystal

© 2012 American Chemical Society 352 dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359

Fermi-liquid behavior in Yb3Ga7Ge3,39 non-Fermi-liquid

behavior in Yb5Ni4Ge10,40 heavy fermion behavior in

Yb2AuGe3,37 etc. These properties are associated with the

presence of an unstable electronic 4f-shell, as several rare-earthspossess two electronic configurations that are closely spaced inenergy: for Ce, the magnetic Ce3+ (4f1) and the nonmagneticCe4+ (4f0), for Eu, the magnetic Eu2+(4f7) and the nonmagneticEu3+(4f6) and for Yb, the magnetic Yb3+(4f13) and thenonmagnetic Yb2+(4f14). Here we synthesize EuInGe singlecrystals using the flux technique and investigate the crystalstructure and physical properties of this compound.The crystal structure of EuInGe was previously reported

within the orthorhombic system.41 EuInGe was synthesized in ahigh temperature reaction of the pure elements in equiatomicratio, and the reactions were done within Ta tubes in an inertatmosphere and by the high frequency induction method. Anew type orthorhombic structure with space group Pnma waselucidated from the single crystal data, with lattice constants a =4.921(1) Å, b = 3.9865(9) Å, and c = 16.004(3). The crystalstructure of EuInGe was explained on the basis of the Zintl−Klemm concept42 with an unusual anionic network. However,during the structure refinement a large disorder was facedwhich showed up in the largely unsatisfactory residuals (R1 >10%), large electron density residuals, and high thermalellipsoid at the In atomic position. In order to solve thoseproblems, a split atom model was used in the indium position,however, creating unacceptable In−In bond distances of 0.8 Å.All these factors and our recent experiences on this kind ofproblem in intermetallic systems35 pointed toward a probablesuperstructure of the EuInGe compound. Hence, to answer theunresolved questions about the structure, we set out to producehigh quality single crystals of EuInGe using the metal fluxtechnique, with indium as the active flux, and to determine theactual structure of this compound. The crystal structure wasstudied using single crystal X-ray diffraction. The stability of thecrystal structure and the nature of the chemical bonding inEuInGe were investigated by first-principles density functionalcalculations. The valence state of europium was determinedusing 151Eu Mossbauer spectroscopy and magnetic suscepti-bility measurements. Physical properties such as magnetism andtransport measurements are also reported.

2. EXPERIMENTAL SECTION2.1. Synthesis. 2.1.1. Method 1. The EuInGe compound was

obtained by combining 4 mmol of europium metal, 4 mmol ofgermanium, and 20 mmol of indium in an alumina crucible under aninert nitrogen atmosphere inside a glovebox. The crucible was placedin a 13 mm fused silica tube, which was flame-sealed under a vacuumof 10−4 Torr, to prevent oxidation during heating. The reactants werethen heated to 1000 °C over 10 h, maintained at that temperature for5 h to allow proper homogenization, followed by cooling to 850 °C for2 h, and held there for 48 h. Finally, the system was allowed to slowlycool to 50 °C in 48 h. The reaction product was isolated from theexcess indium flux by heating at 350 °C and subsequent centrifugationthrough a coarse frit. Any remaining flux was removed by immersionand sonication in glacial acetic acid for 48 h. The final crystallineproduct was rinsed with water and dried with acetone. Several crystals,which grow as metallic silver rods, were carefully selected for elementalanalysis, structure characterization, and the measurements of physicalproperties.2.1.2. Method 2. Elemental europium, indium, and germanium

were mixed in the ideal 1:1:1 atomic ratio and sealed in tantalumampules under argon atmosphere in an arc-melting apparatus. Thetantalum ampules were subsequently placed in a water-cooled samplechamber of an induction furnace (Easy Heat induction heating system,

model 7590), first rapidly heated by applying 180 Amperes current(reaching ca. 1000−1100 °C in temperature), and kept at thattemperature for 30 min. Finally, the reaction was rapidly cooled toroom temperature by switching off the power supply. The productcould easily be removed from the tantalum tubes. No reactions withthe crucible material could be detected. EuInGe is in polycrystallineform and light gray in color and was found to be stable in moist air forseveral weeks. The weight losses of the final material were found to beless than 1%. The samples obtained from the high frequency inductionheating method were used for the resistivity studies.

2.2. Scanning Electron Microscope (SEM)/Energy DispersiveSpectrum (EDS). Semiquantitative microanalyses were performed onthe single crystals obtained from the flux techniques using a scanningLeica 220i electron microscope (SEM) equipped with Bruker 129 eVenergy dispersive X-ray analyzer (EDS). Data were acquired with anaccelerating voltage of 20 kV and in 90 s accumulation time. The EDSanalysis performed on visibly clean surfaces of the single crystals andpolycrystals showed the atomic composition was close to 1:1:1, whichwas in good agreement with the results derived from the refinement ofsingle crystal X-ray diffraction data. The typical EDS spectra for singlecrystal and polycrystalline sample are available in the SupportingInformation (S1 and S2).

2.3. Powder X-ray Diffraction (PXRD). Phase identity and purityof the EuInGe sample was determined by powder XRD experimentsthat were carried out with a Bruker D8 Discover diffractometer usingCu-Kα radiation (λ = 1.54187 Å) over the angular range 10° ≤ 2θ ≤90°, with a step size of 0.014903° at room temperature calibratedagainst corundum standards. Le Bail profile analysis in the Fullprofsuite was used to refine the X-ray diffraction data. The background wasestimated by the six-coefficient polynomial function consisting of ninecoefficients, and the peak shapes were described by a pseudo-Voightfunction varying nine profile coefficients. A scale factor, a zero errorfactor, and shape were refined. The experimental powder pattern ofEuInGe and the XRD pattern simulated from the single-crystal X-raystructure refinement were found to be in good agreement (Figure S3in Supporting Information).

2.4. Single Crystal X-ray Diffraction (SCXRD). The X-rayintensity data were collected at room temperature on rod shapedsingle crystals of EuInGe using a STOE IPDS 2T (with additionalcapability of 2θ swing of the detector) diffractometer with graphite-monochromatized Mo Kα (λ = 0.71073 Å) radiation. A crystal ofsuitable size (0.11 × 0.07 × 0.06 mm) was cut from a larger crystal andmounted on a glass fiber. A full sphere of 180 frames was acquired upto 63° in 2θ. The individual frames were measured with w steps of0.30° and an exposure time of 60 s per frame. The X-AREA (includingX-RED and X-SHAPE) package suite was used for the data extractionand integration and to apply empirical and analytical absorptioncorrections. The WinGX package43 was used for refinement andproduction of data tables, and ORTEP44 was used for structurevisualization. Packing diagrams were generated with Diamond.45

Indexing of the first few frames obtained from the diffractometer led toa primitive monoclinic cell. The intensity of the superstructurereflections increased when the exposure time increased, and we used 6min for better data. This clearly confirms the weak additional peaks inthe powder XRD. Initially, the systematic absences led to the non-centrosymmetric space group P21. A combined general and recemictwinning with three BASF parameters were applied for the successfulrefinement for this non-centrosymmetric space group. However, thePlaton program within WinGx system, ver. 1.70.01.43 was used tocheck the additional symmetry, and this led to space group P21/c. Thestructure was solved by direct method and refined using Shelxl-97(full-matrix least-squares on F2)46 with anisotropic atomic displace-ment parameters for all atoms. As a check for the correct composition,the occupancy parameters were refined in a separate series of least-squares cycles. All sites were fully occupied within two standarduncertainties and in the final cycles the ideal occupancies wereassumed again. All bond lengths are within the acceptable rangecompared to the theoretical values. In addition we omitted thesuperstructure reflections and refined the substructure in theorthorhombic system within the Pnma space group. A better residual

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359353

obtained using a split model did not solve the unacceptable In−Inbond distance (0.8164(10) Å). The details of the data collection andcomplete refinement are shown in Table 1. The positional parameters

and interatomic distances are listed in Tables 2 and 3. Furtherinformation on the structure refinements is available fromFachinformationszentrum Karlsruhe, D-76344 Eggenstein-Leopold-shafen (Germany), by quoting the Registry No. CSDs 425244 and425245 for substructure and superstructure, respectively.

2.5. Mossbauer Spectroscopy. The 151Eu Mossbauer measure-ments were carried out in the transmission geometry using standardPC based spectrometer at room temperature using Eu2O3 as thereference sample.

2.6. Physical Properties. Magnetic measurements of EuInGe as afunction of temperature (magnetic susceptibility) and field (magnet-ization) were carried out on a Quantum Design made MPMS-SQUIDmagnetometer. The resistivity measurements were performed on theEuInGe compound with a conventional AC four probe setup. Fourvery thin copper wires were glued to the pellet using a stronglyconducting silver epoxy paste. The data were collected between 3 and300 K using Quantum Design Physical Property Measurement System(QD-PPMS).

2.7. Theoretical Calculations. The electronic structure ofEuInGe was calculated with the linear muffin-tin orbital (LMTO)method47 in the full potential (FP) implementation of ref 48. Theexchange and correlation energy was treated either in the local spindensity approximation (LSDA)49 or with the inclusion of a Hubbard Ucorrelation parameter on the Eu atoms (LSDA+U) in the fullylocalized limit as implemented in ref 50. In the FP-LMTO method thecrystal volume is divided into two regions, with nonoverlappingmuffin-tin spheres around each atom and the remaining interstitialspace between the atoms.48 A double κ spdf LMTO basis is used todescribe the valence electrons, i.e., Hankel-functions characterized bydecay constants κ are smoothly augmented inside the atomic sphereswith numerical radial functions. The basis set was further augmentedwith local orbitals to describe the semicore states of Eu 5p and In 4dand with floating orbitals centered on interstitial positions.48 The k-point sampling used a regular mesh of 6 × 10 × 4 in the P21/cstructure or 8 × 8 × 4 in the Pnma structure (corresponding to 84 and170 k-points in the irreducible wedge of the Brillouin zone,respectively). The atoms are treated in the scalar-relativisticapproximation with the spin−orbit interaction added within theatomic spheres.

3. RESULTS AND DISCUSSION3.1. Reaction Chemistry. Single crystals of EuInGe were

grown in indium flux generally as metallic silver rods. An SEMimage of a typical single crystal of EuInGe is shown in Figure 1.Reaction byproducts were small amounts of recrystallizedgermanium as well as very small amounts of EuGe2,

51 whichdue to very different crystal morphology (polygonal shape)could be easily distinguished and removed when necessary. TheEuInGe compound also synthesized by direct combination ofthe reactants in primarily gray polycrystalline form and piecesmade up from packed crystals. EuInGe is stable in air and nodecomposition was observed for weeks. The elemental analysisof this compound with SEM/EDS gave the atomic composition33(1):34(1):33(1) for Eu:In:Ge, in excellent agreement with

Table 1. Crystal Data and Structure Refinement for EuInGe

substructure superstructure

formula weight 339.37temperature 293 Kwavelength 0.71073 Åcrystal system orthorhombic monoclinicspace group Pnma P21/cunit cell dimensions a = 4.9066(10) Å a = 7.9663(16) Å

b = 3.9834(8) Å b = 4.9119(10) Åc = 15.964(3) Å c = 16.465(5) Å

β = 104.03 degvolume 312.02(11) Å3 625.1(3) Å3

Z, calculated density 4, 7.224 g/cm3 8, 7.213 g/cm3

absorption coefficient 36.42 mm−1 36.36 mm−1

F(000) 576 1152crystal size 110 × 70 × 60 μmTheta range for datacollection

2.55 to 31.71 deg 2.55 to 31.69 deg

limiting indices −7 ≤ h ≤ 6 −11 ≤ h ≤ 11−5 ≤ k ≤ 5 −7 ≤ k ≤ 6−23 ≤ l ≤ 23 −24 ≤ l ≤ 24

reflections collected/unique

2601/585 [R(int) =0.0438]

5641/2122 [R(int) =0.0548]

completeness to theta =28.99

98.0% 95%

refinement method full-matrix least-squares on F2

data/restraints/parameters

585/0/23 2122/0/57

goodness-of-fit on F2 1.114 1.062final R indices [I >2σ(I)]a

R1 = 0.0251 R1 = 0.0467

wR2 = 0.0496 wR2 = 0.1135largest diff peak and hole 1.853 and −1.366 e

Å−39.733 and −4.244 e Å−3

aR = Σ∥Fo| − |Fc∥/Σ|Fo|, wR = {Σ[w(|Fo|2 − |Fc|2)2]/Σ[w(|Fo|4)]}1/2

and w = 1/(σ2(I) + 0.0016I2).

Table 2. Atomic Coordinates (× 104) and Anisotropic Displacement Parameters (Å2 × 103) for EuInGea

Wyckoff Site x y z U11 U22 U33 U23 U13 U12 Ueq

SubstructureEu1 4c 191(1) 2500 3925(1) 9(1) 8(1) 8(1) 0 0(1) 0 8(1)In1 8d 5026(1) 1475(2) 7105(1) 7(1) 13(1) 6(1) −1(1) 0(1) 0(1) 9(1)Ge1 4c 4238(2) 2500 5422(1) 10(1) 7(1) 6(1) 0 1(1) 0 7(1)SuperstructureEu1 4e 8215(2) 177(4) 8927(1) 6(1) 5(1) 9(1) 1(1) 0(1) 0(1) 7(1)Eu2 4e 6793(2) 9791(4) 1077(1) 7(1) 12(1) 7(1) 0(1) −1(1) −1(1) 9(1)In1 4e 5716(2) 4934(4) 7895(1) 14(1) 8(1) 10(1) 0(1) 1(1) 0(1) 11(1)In2 4e 314(2) 4984(5) 2105(1) 15(1) 9(1) 10(1) −1(1) 3(1) 0(1) 11(1)Ge1 4e 6001(4) 5765(3) 9581(1) 6(1) 8(1) 6(1) −6(2) 1(1) 1(1) 6(1)Ge2 4e 9012(4) 4235(3) 424(1) 4(1) 9(1) 6(1) −3(2) 2(1) 0(1) 6(1)

aThe anisotropic displacement factor exponent takes the form: −2π2[h2a*2U11 + ... + 2hka*b*U12. Equivalent isotropic displacement parameters Ueqare defined as one third of the trace of the orthogonalized Uij tensor.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359354

the results obtained from the single crystal refinement. Singlecrystals were selected from the reaction mixture, and theircrystal structure was crystallographically refined. The poly-crystalline sample obtained from the high frequency inductionfurnace method was used for the physical property measure-ments.3.2. Crystal Chemistry. The crystal structure of EuInGe is

shown in Figure 2, panels a and b as viewed along the b-axisand a-axis, respectively. Mao et al.41 proposed that a structuralrelationship exists between BaGe2 (tetragonal) and EuInGe(orthorhombic Pnma) through the intermediate EuInGe(GdSi2, orthorhombic Imma).41 Our refinement suggests thatEuInGe crystallizes in an ordered monoclinic structure, whichcan be explained in a similar way as the previously reportedorthorhombic structure except that the doubling of the b axis isobserved in the crystal obtained from the flux reactions,together with a change in the symmetry from tetragonal (I41/amd, ThSi2 type structure) to monoclinic (P21/c space groupand its own type structure). Group−subgroup relationships and

group analysis can predict a lowering symmetry fromorthorhombic Pnma to monoclinic P21/c (see Figure 3). Thebonds to the Eu atoms were omitted to emphasize the three-dimensional [InGe] framework and its channels. The structurecan also be described as an intergrowth of two independentslabs of parallel zigzag chains of In and Ge atoms that runparallel to the a or b axis (Figure 2c,d) with two Eu atomsresiding within the channels formed by the connection of thelayers.The refined crystal structure is fully ordered with acceptable

bond distances. The nearest Ge−Ge distances are 2.465(6) Åand 2.575(5) Å and comparable to the Ge−Ge distances inEuGe2.

51 The shortest distance between the Eu and Ge atomsis 3.094(3) Å, which is common in Eu−Ge basedintermetallics.52,53 Crystallographically, there are six differentsites in EuInGe, two each of europium, indium, and germaniumatoms. Both In atoms are 4-fold coordinated and have almosttetrahedral geometries, while germanium atoms are trifold

Table 3. Interatomic Distances (Å) in the Structure of EuInGe, Calculated with the Powder Lattice Parameters

Substructure

Eu: 1 Ge 3.1077(9) In: 1 In 0.8164(14) Ge: 2 Ge 2.5188(9)2 Ge 3.1265(7) 1 Ge 2.7444(10) 2 In 2.7443(10)2 In 3.2741(8) 2 In 2.7584(7) 2 Eu 3.1265(7)2 In 3.3106(8) 2 In 2.8767(8) 2 Eu 3.5389(8)2 In 3.4297(8) 1 In 3.1670(15)2 Ge 3.5389(8) 1 Eu 3.2741(8)1 In 3.7379(8) 1 Eu 3.3106(8)

1 Eu 3.4297(8)1 Eu 3.7379(8)1 Eu 3.7699(9)

Superstructure

Eu1: 1 Ge2 3.094(3) Eu2: 1 Ge1 3.103(2) In1: 1 Ge1 2.760(2) In2: 1 Ge2 2.736(2)1 Ge2 3.113(2) 1 Ge1 3.118(3) 2 In1 2.882(1) 2 In2 2.879(1)1 Ge1 3.145(3) 1 Ge2 3.158(3) 1 In2 3.163(2) 1 In1 3.163(2)1 In1 3.267(2) 1 In1 3.310(2) 1 Eu2 3.310(2) 1 Eu1 3.298(3)1 In2 3.298(3) 1 In2 3.315(2) 1 Eu1 3.442(3) 1 Eu2 3.306(2)1 In2 3.414(2) 1 Ge1 3.522(3) 1 Eu2 3.726(2) 1 Eu1 3.414(2)1 In1 3.441(3) 1 Ge2 3.557(3) 1 Eu1 3.777(2) 1 Eu1 3.743(2)1 Ge2 3.523(3) 1 In1 3.726(2) 1 Eu2 3.900(2) 1 Eu2 3.865(3)1 Ge1 3.567(3) 1 In2 3.743(3) Ge1: 1 Ge1 2.465(6) 1 Ge2 2.461(6)1 In2 3.773(2) 1 In2 3.865(3) 1 Ge2 2.575(5) 1 Eu1 3.094(3)1 In1 3.777(2) 1 In1 3.900(2) 1 Eu2 3.102(3) 1 Eu2 3.158(3)1 Eu1 3.965(2) 1 Eu1 3.974(3) 1 Eu1 3.146(3) 1 Eu1 3.523(3)

1 Eu2 3.522(3)

Figure 1. A typical single crystal of EuInGe grown using the metal fluxtechnique.

Figure 2. Crystal structure view of monoclinic EuInGe along [010]and [100] shown in (a) and (b), respectively. (c) Two dimensionalzigzag layer of In atom. (d) One dimensional Ge layer.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359355

coordinated (Figure 4) making this compound of Zintl typeEu2+(4b-In)−(3b-Ge)− as explained by Mao et al.41 Our

magnetic susceptibility and Mossbauer spectroscopic studiesalso revealed the divalent nature of europium atoms in EuInGe.The coordination environments of all atoms are shown inFigure 5. Eu1 and Eu2 are located in 15-membered prismaticcages. Although the overall nature of both environments is thesame, strong interactions of Eu1−In (3.41−3.44 Å) and weakinteractions of Eu2−In (3.86−3.90 Å) led to cages withdifferent distortions. The coordination environment of Inatoms are formed as tetracapped triangular prismatic,composed of 10 atoms of which six are Eu atoms, three areIn atoms, and one is a Ge atom. On the other hand, thecoordination environment of Ge atoms is formed as tricappedtriangular prismatic, composed of nine atoms of which six areEu atoms, one is an In atom, and two are Ge atoms.3.3. Mossbauer Spectroscopy. The 151Eu Mossbauer

spectrum of EuInGe at room temperature is presented inFigure 6 together with transmission integral fits. The spectrashow one line with an isomer shift of around −10 mm/s with

respect to the Eu2O3 source, which indicates divalence ofeuropium in EuInGe consistent with the magnetic susceptibilitymeasurements and the chemical bonding analysis. Theexperimentally observed line widths are close to the usualline width of 2.3 mm/s of europium.

3.4. Physical Properties. Magnetic susceptibility (χ = M/H) was measured as a function of temperature under twodifferent applied fields of H = 100 Oe and 5 kOe. χ(T) was alsomeasured in the zero field cooled (ZFC) and field cooled (FC)conditions of the sample. In Figure 7, a plot of ZFC and FCχ(T) measured in a field of 100 Oe is shown. Overall, thefeatures of χ(T) observed here are in good agreement withthose reported by Mao et al.41 However, on close examinationof the profile, signatures of multiple magnetic transitions areobserved. As temperature (T) is lowered, χ increasesmonotonically but undergoes a sudden change below 70 K asif undergoing a ferromagnetic transition. On further lowering ofthe temperature, more structures are observed in χ(T) at 44,25, and 13 K, respectively. To highlight the anomalies in χ(T),the first derivative of the magnetization (dM/dT) is plotted asan inset in Figure 7. The temperatures at which anomalies areobserved are marked as T1, T2, T3, and T4.In Figure 8 the molar magnetic susceptibility measured in a

field of 5 kOe in the ZFC condition of the sample is plotted asa function of temperature. The inverse susceptibility plot,χ−1(T), is linear above 100 K corresponding to the Curie−

Figure 3. Group−subgroup scheme of subcell and the superstructureof EuInGe. The indices for the translationengleiche (t) transition andthe unit cell transformations are given. The evolution of the atomicparameters is shown at the right.

Figure 4. Tetra-coordinate indium and tricoordinate germanium in thebasic unit of the monoclinic structure of EuInGe.

Figure 5. The local coordination environments of (left) Eu atoms,(right) In atoms, and (middle) Ge atoms in the monoclinic structureof EuInGe.

Figure 6. Experimental and simulated 151Eu Mossbauer spectra ofEuInGe at room temperature. The spectrum is measured relative to aEu2O3 source.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359356

Weiss Law. The deviation from Curie−Weiss behavior is seenbelow 100 K. The values of the paramagnetic Curietemperature (θp) and the effective moment (μeff) obtainedfrom the linear fit of the χ−1(T) data in the 150−30 K range areθp = 12 K and μeff = 7.61 μB, respectively. The positive value ofθp indicates predominant ferromagnetic interactions. The valueof μeff is close to the expected free ion moment for Eu2+ of 7.94μB. Several intermetallic compounds exhibit similar values fordivalent europium.54 These values are slightly different from theones reported by Mao et al.41 and may be attributed to samplepreparation conditions. In order to verify the field inducedchanges to magnetic susceptibility, we have shown in Figure 9χ(T) measured in two different magnetic fields (H = 100 Oeand 5 kOe). Though the essential features are similar in bothcases, the low field data are sharper and the anomalies areclearly visible. The ordering temperatures observed in low fielddata are also seen in 5 kOe. It is interesting to note that above70 K, the two curves overlap as if the behavior is field-independent.In Figure 10, the isothermal magnetization curves (M vs H)

are shown for EuInGe measured at T = 1.8, 50, and 100 K. Thenotable observations made here at all three temperatures are

that there is no hysteresis observed and there is no saturation ofmagnetization. For initial application of field, M varies linearlywith H. However, for T = 1.8 and 50 K, there is slight deviationfrom this linearity above a field of 10 kOe. This means that themagnetic correlations persist up to 50 K, but less than 100 K. AtT = 100 K, M varies linearly with H, indicating theparamagnetic state of the sample. The normal state temperaturedependent resistivity of EuInGe is shown in Figure 11. Theresistivity decreases almost linearly with decreasing temper-ature, which is typically seen in metallic systems55,56 withoutany long-range magnetic ordering.

Figure 7. Zero field cooled (ZFC) and field cooled (FC) magneticsusceptibility (χ = M/H) for EuInGe measured in a field of 100 Oe.The first derivative of the observed magnetization (M) is plotted as afunction of temperature in the inset of the figure, to highlight themultiple magnetic transitions.

Figure 8. Magnetic susceptibility and inverse susceptibility of EuInGemeasured in a field of 5 kOe. The red line passing through the datapoints is the linear fit of the χ−1(T) data exhibiting the Curie−Weissbehavior. The intercept of this line on T-axis marks the paramagneticCurie temperature (θp).

Figure 9. Magnetic susceptibility of EuInGe measured in H = 100 Oeand 5 kOe is plotted on the same scale to highlight the influence offield on the moment formation.

Figure 10. Isothermal magnetization curves (M vs H) for EuInGemeasured at T = 1.8 K, 50 and 100 K. The straight thick lines passingthrough the data points are a guide for eyes only, shown to highlightthe deviation of the data from linearity.

Figure 11. Electrical resistivity (ρ) of EuInGe measured as a functionof temperature.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359357

3.5. Band Structure Calculations. The calculatedelectronic density of states of EuInGe in the experimentalP21/c crystal structure of the present work is shown in Figure12. A Hubbard parameter of U = 0.5 Ry was used. The density

of states reveals that a significant hybridization takes place inEuInGe, with a distinct pseudogap occurring at the Fermi level.The lowest valence bands, from −11 eV to about −7.5 eVrelatively to the Fermi level, are dominated by the Ge states; inthe region from −7.5 eV to −4 eV the In states dominate, whileall constituent atoms contribute to the states between −4 eVand the Fermi level. Most notably the Eu f states fall in thisregion. The exact position of the Eu f level cannot be predictedaccurately within the LSDA+U as it depends on the valueadopted for the U parameter. With U = 0, i.e., in the LSDA, thef-states fall just below the Fermi level, while for U = 0.5 Ry, thef-peak has moved to about −2.5 eV below the Fermi level. Inboth cases the f-occupation is close to seven corresponding todivalent Eu ions and spin-moments of 7 μB per Eu atom, whichis consistent without magnetic susceptibility and Mossbauerdata. The pseudogap around the Fermi level is also present inLSDA (i.e., with U = 0) but less pronounced (not shown). Incomparison of the total energy of the crystal with ferromagneticand antiferromagnetic ordering of the Eu moments theferromagnetic arrangement was found to be most favorableboth in LSDA and LSDA+U, by about 0.17 eV per formula unitin LSDA. The energy was also calculated for a EuInGecompound assuming the Pnma parent crystal structure with halfoccupation of the 8d positions of In. The energy is in this case0.40 eV higher per formula unit. By doubling the Pnma unit cellalong the crystallographic b-axis and allowing for different Inoccupations in the two cells, the energy difference to the P21/cstructure comes down to 25 meV, still in favor of the P21/cstructure. These results show that the subcell picture largely isvalid; i.e., the observed P21/c structure energetically is close tothat of a Pnma structure with a double unit cell andrearrangement of the In atoms.

4. CONCLUDING REMARKS

EuInGe single crystals were obtained from reactions in moltenIn over a broad range of synthetic conditions. The flux methodproves to be an excellent tool for growing intermetalliccompounds to understand the complex crystal structures. Theroom temperature structure was refined from single crystaldiffractometer data, revealing a new monoclinic structure type.

Multiple magnetic transitions as a function of temperature havebeen observed in EuInGe from magnetic susceptibilitymeasurements. Europium is found to be in the divalentoxidation state. Further, we believe that low temperaturestructural, Mossbauer, and other spectroscopic studies such asXMCD, in the temperature range of 2−50 K, would be quiterewarding in understanding the microscopic origins of themultiple magnetic transitions observed in the magneticsusceptibility measurements.

■ ASSOCIATED CONTENT*S Supporting InformationCrystallographic information files (CIF), Rietveld refinement ofX-ray powder diffraction pattern, and typical EDS spectra ofsingle crystal and polycrystals. This information is available freeof charge via the Internet at http://pubs.acs.org/.

■ AUTHOR INFORMATIONCorresponding Author*Phone: 080-22082298. Fax: 080-22082627. E-mail:sebastiancp@jncasr.ac.in.Present Address⊥Currently at Department of Condensed Matter Physics, Ecolede Physique, 24, quai Ernest-Ansermet, 1211, Geneva 4,Switzerland.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank Prof. C. N. R. Rao for his support and guidance.S.C.P. is grateful to Prof. Mercouri G. Kanatzidis for thediscussion on the scope of this work and critical comments.Financial support from the DST (Grant SR/S2/RJN-24/2010),Sheik Saqr Laboratory and JNCASR is gratefully acknowledged.U.S. thanks CSIR for the research fellowship and S.C.P. thanksDST for the Ramanujan fellowship. Authors thank BharathRajeswaran (JNCASR) for resistivity measurements and Prof.A. Gupta and Dr. V. R. Reddy (UGC-DAE Indore) forMossbauer spectroscopic studies. Authors also thank DST forproviding the financial support for carrying out experiments atIndian Beamline, Photon Factory, Japan.

■ REFERENCES(1) Kanatzidis, M. G.; Pottgen, R.; Jeitschko, W. Angew. Chem., Int.Ed. 2005, 44, 6996−7023.(2) Chen, X. Z.; Sportouch, S.; Sieve, B.; Brazis, P.; Kannewurf, C. R.;Cowen, J. A.; Patschke, R.; Kanatzidis, M. G. Chem. Mater. 1998, 10,3202−3211.(3) Sieve, B.; Chen, X. Z.; Henning, R.; Brazis, P.; Kannewurf, C. R.;Cowen, J. A.; Schultz, A. J; Kanatzidis, M. G. J. Am. Chem. Soc. 2001,123, 7040−7047.(4) Chen, X. Z.; Larson, P.; Sportouch, S.; Brazis, P.; Mahanti, S. D.;Kannewurf, C. R.; Kanatzidis, M. G. Chem. Mater. 1999, 11, 75−83.(5) Zhuravleva, M. A.; Kanatzidis, M. G. Z. Naturforsch., B: J. Chem.Sci. 2003, 58, 649−657.(6) Zhuravleva, M. A.; Pcionek, R. J.; Wang, X. P.; Schultz, A. J.;Kanatzidis, M. G. Inorg. Chem. 2003, 42, 6412−6424.(7) Zhuravleva, M. A.; Evain, M.; Petricek, V.; Kanatzidis, M. G. J.Am. Chem. Soc. 2007, 129, 3082−3083.(8) Chen, X. Z.; Small, P.; Sportouch, S.; Zhuravleva, M.; Brazis, P.;Kannewurf, C. R.; Kanatzidis, M. G. Chem. Mater. 2000, 12, 2520−2522.(9) Latturner, S. E.; Bilc, D.; Mahanti, S. D.; Kanatzidis, M. G. Inorg.Chem. 2003, 42, 7959−7966.

Figure 12. Density of states of EuInGe in the experimental P21/ccrystal structure as calculated with the LSDA+U method with U = 0.50Ry. The Fermi level is marked with a dashed line, and the Eu, In, andGe partial contributions are illustrated.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359358

(10) Wu, X. U.; Latturner, S. E.; Kanatzidis, M. G. Inorg. Chem. 2006,45, 5358−5366.(11) Latturner, S. E.; Kanatzidis, M. G. Inorg. Chem. 2008, 47, 2089−2097.(12) Latturner, S. E.; Bilc, D.; Mahanti, S. D.; Kanatzidis, M. G.Chem. Mater. 2002, 14, 1695−1705.(13) Salvador, J. R.; Gour, J. R.; Bilc, D.; Mahanti, S. D.; Kanatzidis,M. G. Inorg. Chem. 2004, 43, 1403−1410.(14) Salvador, J. R.; Bilc, D.; Gour, J. R.; Mahanti, S. D.; Kanatzidis,M. G. Inorg. Chem. 2005, 44, 8670−8679.(15) Salvador, J. R.; Kanatzidis, M. G. Inorg. Chem. 2006, 45, 7091−7099.(16) Chondroudi, M.; Balasubramanian, M.; Welp, U.; Kwok, W.-K.;Kanatzidis, M. G. Chem. Mater. 2007, 19, 4769−4775.(17) Salvador, J. R.; Hoang, K.; Mahanti, S. D.; Kanatzidis, M. G.Inorg. Chem. 2007, 46, 6933−6941.(18) Canfield, P. C.; Fisk, Z. Z. Philos. Mag. B 1992, 65, 1117−1123.(19) Bud’ko, S. l.; Islam, Z.; Wiener, T. A.; Fisher, I. R.; Lacerda, A.H.; Canfield, P. C. J. Magn. Magn. Mater. 1999, 205, 53−78.(20) Fisher, I. R.; Islam, Z.; Canfield, P. C. J. Magn. Magn. Mater.1999, 202, 1−10.(21) Nicklas, M.; Sidorov, V. A.; Borges, H. A.; Pagliuso, P. G.;Petrovic, C.; Fisk, Z.; Sarrao, J. L.; Thompson, J. D. Phys. Rev. B 2004,67, 020506−020510.(22) Hundley, M. F.; Sarrao, J. L.; Thompson, J. D.; Movshovich, R.;Jaime, M.; Petrovic, C.; Fisk, Z. Phys. Rev. B 2001, 65, 024401−024407.(23) Macaluso, R. T.; Sarrao, J. L.; Moreno, N. O.; Pagliuso, P. G.;Thompson, J. D.; Fronczek, F. R.; Hundley, M. F.; Malinowski, A.;Chan, J. Y. Chem. Mater. 2003, 15, 1394−1398.(24) Bailey, M. S.; McCuire, M. A.; DiSalvo, F. J. J. Solid State Chem.2005, 178, 3494−3499.(25) Benbow, E. M.; Latturner, S. E. Inorg. Chem. 2006, 179, 3989−3996.(26) Klunter, W.; Jung, W. J. Solid State Chem. 2006, 179, 2880−2888.(27) Lukachuk, M.; Galadzhun, Y. V.; Zaremba, R. I.; Dzevenko, M.V.; Kalychak, Y. M.; Zaremba, V. I.; Rodewald, U. C.; Pottgen, R. J.Solid State Chem. 2005, 178, 2724−2733.(28) Macaluso, R. T.; Sarrao, J. L.; Pagliuso, P. G.; Moreno, N. O.;Goodrich, R. G.; Browne, D. A.; Fronczek, F. R.; Chan, J. Y. J. SolidSate Chem. 2002, 166, 245−250.(29) Zaremba, V. I.; Dubenskiy, V. P.; Rodewald, U. C.; Heying, B.;Pottgen, R. J. Solid State Chem. 2006, 179, 891−897.(30) Peter, S. C.; Kanatzidis, M. G. J. Solid State Chem. 2010, 183,2077−2081.(31) Chondroudi, M.; Peter, S. C.; Malliakas, C. D.;Balasubramanian, M.; Li, Q.; Kanatzidis, M. G. Inorg. Chem. 2011,50, 1184−1193.(32) Peter, S. C.; Kanatzidis, M. G. Z. Anorg. Allg. Chem. 2012, 638,287−293.(33) Peter, S. C.; Salvador, J.; Martin, J. B.; Kanatzidis, M. G. Inorg.Chem. 2010, 49, 10468−10474.(34) Iyer, A. K.; Peter, S. C. Eur. J. Inorg. Chem. 2012, 11, 1790−1794.(35) Peter, S. C.; Chondroudi, M.; Malliakas, C. D.;Balasubramanian, M.; Kanatzidis, M. G. J. Am. Chem. Soc. 2011, 133,13840−13843.(36) Peter, S. C.; Malliakas, C. D.; Chondroudi, M.; Schellenberg, I.;Rayaprol, S.; Pottgen, R.; Kanatzidis, M. G. Inorg. Chem. 2010, 49,9574−9580.(37) Peter, S. C.; Sarkar, S.; Kanatzidis, M. G. Inorg. Chem. 2012, 51,10793−10799.(38) Subbarao, U.; Peter, S. C. Inorg. Chem. 2012, 51, 6326−6332.(39) 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.(40) Peter, S. C.; Rayaprol, S.; Francisco, M. C.; Kanatzidis, M. G.Eur. J. Inorg. Chem. 2011, 2011, 3963−3968.

(41) Mao, J. G.; Goodey, J.; Guloy, A. M. Inorg. Chem. 2002, 41,931−937.(42) Corbett, J. D. In Chemistry, Structure and Bonding of Zintl Phasesand Ions; Kauzlarich, S. M., Ed.; VCH Publishers: New York, 1996; p139.(43) WinGX: A Windows Program for Crystal Structure Analysis:Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.(44) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.(45) CRYSTAL IMPACT, Version 3.g; Crystal Impact GbR: Bonn,Germany, 2011.(46) SHELXTL; 5.10 ed.; Bruker Analytical X-ray Systems, Inc.:Madison, WI, 1998.(47) Andersen, O. K. Phys. Rev. B 1975, 12, 3060−3083.(48) Methfessel, M.; van Schilfgaarde, M.; Casali, R. A. Lecture Notesin Physics; Springer: Berlin, 2000; Vol. 353, p 114.(49) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200−1211.(50) Larson, P.; Lambrecht, W. R. L.; Chantis, A.; van Schilfgaarde,M. Phys. Rev. B 2007, 75, 045114−045128.(51) Bobev, S.; Bauer, E. D.; Thompson, J. D.; Sarrao, J. L.; Miller, G.J.; Eck, B.; Dronskowski, R. J. Solid State Chem. 2004, 177, 3545−3552.(52) Felner, I.; Nowik, I. J. Phys. Chem. Solids 1978, 39, 763−765.(53) Wortmann, G.; Nowik, I.; Perscheid, B.; Kaindl, G.; Felner, I.Phys. Rev. B 1991, 43, 5261−5268.(54) Buschow, K. H. J. Handbook of Magnetic Materials; Elsevier: UK,2009; Vol. 18.(55) Dremov, R. V.; Koblyuk, N.; Mudryk, Y.; Romaka, L.;Sechovsky, V. J. Alloys Compd. 2001, 317, 293−296.(56) He, J.; Ling, G.; Jiao, Z. Physica B 2001, 301, 196−202.

Crystal Growth & Design Article

dx.doi.org/10.1021/cg301532b | Cryst. Growth Des. 2013, 13, 352−359359