Synchrotron X-ray diffraction study of ErMn2D2

Synchrotron X-ray diffraction study of ErMn2D2

Journal of Alloys and Compounds 437 (2007) 140–145 Synchrotron X-ray diffraction study of ErMn2D2 d ˙ J.P. Maehlen a , V.A. Yartys a,∗ , A.B. Riabov ...

724KB Sizes 0 Downloads 13 Views

Journal of Alloys and Compounds 437 (2007) 140–145

Synchrotron X-ray diffraction study of ErMn2D2 d ˙ J.P. Maehlen a , V.A. Yartys a,∗ , A.B. Riabov b , A. Budziak c , H. Figiel d , J. Zukrowski a

Institute for Energy Technology, P.O. Box 40, Kjeller No-2027, Norway Physico-Mechanical Institute of the National Academy of Science of Ukraine, 5 Naukova Street, 79601 Lviv, Ukraine c The H. Niewodniczanski Institute of Nuclear Physics, Radzikowskiego 152, 31-342 Krakow, Poland d Department of Solid State Physics, Faculty of Physics and Applied Computer Science AGH, University of Science and Technology, 30 Mickiewicza av., 30-059 Krakow, Poland b

Received 29 June 2006; received in revised form 19 July 2006; accepted 20 July 2006 Available online 7 September 2006

Abstract ErMn2 D2 deuteride has been studied by high-resolution synchrotron X-ray powder diffraction at temperatures between 150 and 298 K. Below ˚ c = 9.05368(7) A; ˚ T = 298 K) 210 K a transformation from the hexagonal C14 Laves phase type structure (space group P63 /mmc; a = 5.55357(3) A; ˚ b = 5.57558(4) A; ˚ c = 9.07102(8) A; ˚ β = 90.5451(5)◦ ; T = 150 K) into the monoclinic one crystallising in the space group C2/m (a = 9.61247(9) A; takes place. A rather large two-phase region of coexistence of both structural modifications was observed between 185 and 205 K. © 2006 Elsevier B.V. All rights reserved. Keywords: Magnetically ordered materials; Hydrogen absorbing materials; Crystal structure; Phase transitions; Synchrotron radiation

1. Introduction The ErMn2 H2 (ErMn2 D2 ) hydride has earlier been studied using standard X-ray diffraction technique [1,2]. At room temperature the hexagonal C14 Laves phase-related structure, space group (s.g.) P63 /mmc, was found, while the pattern obtained below ∼210 K was attributed to a triclinic distortion to s.g. P1 [1]. Because of rather low quality of the data, attempts to refine them with structures of higher symmetry did not give satisfactory results. In present work we have performed synchrotron X-ray powder diffraction (SR-XRD) studies to solve the structure of the low temperature ErMn2 D2 deuteride unambiguously. 2. Experimental details 2.1. Sample preparation The sample was prepared from high purity metals. Induction melting was followed by annealing of the sample placed into an evacuated quartz ampoule at 1070 K for 5 days and then quenching into a mixture of ice and water, yielding a nearly pure intermetallic alloy ErMn2 . The deuteration was performed using standard technique described, e.g. in [3]. Dideuteride ErMn2 D2 was synthesised



Corresponding author. Tel.: +47 63 80 64 53; fax: +47 63 81 29 05. E-mail address: [email protected] (V.A. Yartys).

0925-8388/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2006.07.088

by absorbing the amount of deuterium gas necessary to reach stoichiometry D/ErMn2 = 2.0. Synthesis was accompanied by annealing at 453 K for a few hours to improve the homogeneity of the material. The X-ray diffraction data collected using conventional technique showed a formation of close to single-phase deuteride. The sample contained small amounts of impurities, Er2 O3 (approxi¯ a = 10.5480(7) A) ˚ and ErD2±x (approximately 2 wt.%, mately 2 wt.%, s.g. Ia3, ¯ a = 5.1158(8) A), ˚ which were both introduced into the refinements s.g. Fm3m, of the X-ray powder diffraction data. At temperatures applied in present work (150–298 K), the studied deuteride is stable; no changes of the deuterium content in the sample take place.

2.2. The X-ray measurements The preliminary X-ray diffraction measurements were performed on a ˚ with a SIEMENS D5000 diffractometer using Cu K␣ radiation (λ = 1.54056 A) continuous-flow cryostat supplied by Oxford Instruments. SR-XRD data were collected at the Swiss-Norwegian Beam Line (BM01B) ˚ at ESRF, Grenoble (Si(1 1 1) channel-cut monochromator, λ = 0.37504(2) A, scintillation detectors) at temperatures 150, 185, 190, 195, 200, 205, 210, 215 and 298 K. The sample was enclosed inside a quartz capillary of 0.3 mm in diameter. The experiment was performed in following steps: 1. Measurements were first done at 298 K for 2 h. 2. The sample was cooled down to 150 K and measured for 2 h. 3. A series of measurements (20 min each) with a stepwise increase in temperatures (185–215 K with a 5 K step) was performed. Prior to every measurement step, the sample was equilibrated at constant temperature for approximately 30 min.

J.P. Maehlen et al. / Journal of Alloys and Compounds 437 (2007) 140–145

Fig. 1. Diffraction pattern of the hexagonal ErMn2 D2 (s.g. P63 /mmc) collected at 293 K showing observed (crosses), calculated (upper line) and difference (bottom line) plots. The positions of the Bragg peaks are shown as ticks (upper: ErD2±x , middle: Er2 O3 , bottom: ErMn2 D2 ). Inset: selected region of the diffraction pattern, 7.5–9.0◦ , with Bragg indexes of three peaks given.

3. Results All powder diffraction data were analysed by the Rietveld whole-profile refinement method [4] using the General Structure Analysis System (GSAS) software [5]. Peak shapes were described by a multi-term Simpson’s rule integration of the pseudo-Voigt function [6,7], which includes the asymmetry correction according to Finger et al. [8]. 3.1. Crystal structure of ErMn2 D2 at room temperature The refinements of the room temperature data confirm the formation of the hexagonal C14 type Laves type hydride: s.g. ˚ c = 9.05368(7) A ˚ and are shown P63 /mmc; a = 5.55357(3) A; in Fig. 1. The metal sublattice of the ErMn2 D2 crystal structure is presented in Fig. 2 while Table 1 summarises the results obtained from the refinements of the diffraction data. The interatomic metal–metal distances change ˚ (Er–Er), ∼3.10 A ˚ (Er–Mn), and ∼2.65 A ˚ from ∼3.24 A ˚ (Mn–Mn) in the intermetallic compound ErMn2 [a = 5.294 A, ˚ (Er–Er), ˚ s.g. P63 /mmc] to 3.376(1)–3.4068(4) A c = 8.664 A, ˚ (Er–Mn), and 2.677(3)–2.876(3) A ˚ 3.2500(3)–3.2890(8) A (Mn–Mn) in the deuteride. ErMn2 belongs to the group of Laves phases RMn2 (R = rare earth) compounds containing magnetic sublattices formed by 3d magnetic moments of the Mn atoms. The 3d shells lose their intrinsic magnetic moments when firstneighbour Mn–Mn distance d becomes smaller than some crit˚ [9]. The shortest Mn–Mn distances ical distance of dc ∼ 2.7 A ˚ explain the absence of present in the metal sublattice (2.678 A) a long-range magnetic order of ErMnD2 at room temperature. 3.2. Crystal structure of ErMn2 D2 at low temperatures Structural solution of the low temperature diffraction pattern (150 K) was achieved in the following manner. First, an individual peak fitting was performed for the selected reflections (40 peaks using the WinPLOTR software [10]) with the

141

Fig. 2. The metal sublattice in the hexagonal crystal structure of ErMn2 D2 at room temperature. The Mn4 tetrahedra are slightly elongated along the c-axis ˚ distances out of (Mn–Mn interatomic distances in the (a, b) plane are 2.677 A; ˚ the plane, along [0 0 1] are 2.741 A).

aim at obtaining as-accurate-as-possible peak positions. Then the unit cell was identified utilising indexing software packages (CRYSFIRE indexing suite [11]) yielding a monoclinic unit cell as the best choice. Systematic extinctions observed (for the hkl reflections a condition h + k = 2n was the case) indicated that the unit cell was a C-centred cell with the possible groups among C2, Cm, and C2/m. The whole-profile Rietveld fitting was performed using starting atomic positions obtained through a general transformation from description of the crystal structure in s.g. P63 /mmc to s.g. C2/m (no. 12). Lowering the symmetry (both space groups C2 (no.5) and Cm (no.8) were considered) did not give any improvements of the fit. The fitted pattern obtained at 150 K is presented in Fig. 3. Table 1 summarises the fitted parameters, while interatomic distances are presented in Table 2. The labelling of the sites presented in Tables 1 and 2 is chosen according to the following scheme (hexagonal → monoclinic): Er → Er1 and Er2; Mn1 → Mn1 and Mn2; Mn2 → Mn3 and Mn4. The scheme of the transformation from the room temperature to the low temperature structures is presented in Fig. 4. During this transformation, the irreducible number of atoms doubles (increasing from three in the hexagonal to six in the monoclinic cell). During the transformation of the hexagonal modification into the monoclinic one it undergoes a deformation creating “enlarged” Mn–Mn interatomic distances. Even the shortest Mn–Mn distance in the monoclinic ErMn2 D2 phase becomes ˚ As a result, all dMn–Mn disquite long, being equal to ∼2.71 A. ˚ tances become longer than the critical distance dc of ∼2.7 A observed for the magnetically ordered RMn2 (R = rare earth) compounds [9]. 3.3. Region of coexistence of the hexagonal and monoclinic modifications Within a certain temperature window, two structural modifications of ErMnD2 coexist forming a two-phase region (see

142 Table 1 ˚ 2 ), unit cell dimensions, phase fraction (by weight), and goodness-of-fit parameters) derived from the Rietveld refinements of Crystal structure data (atomic coordinates, isotropic temperature factors (Uiso × 100 A the SR-XRD data for ErMn2 D2 at selected temperatures T (K) 150

185

190

195

200

205

0.6641(5) 0.9333(5) 0.1(1) 0.3327(5) 0.4393(6) 0.1(1) 0.1(2) 0.1(2) 0.1634(18) 0.7501(16) 0.1(2) 0.0804(12) 0.7531(16) 0.2551(15) 0.1(2) 9.60364(32) 5.57279(14) 9.07047(25) 90.4231(18) 485.430(24) 0.65(5)

0.6650(5) 0.9335(6) 0.26 0.3319(5) 0.4383(6) 0.26 0.09 0.09 0.1610(17) 0.7501(16) 0.09 0.0835(11) 0.7580(17) 0.2514(17) 0.09 9.6034(3) 5.57240(15) 9.06965(27) 90.4206(19) 485.341(26) 0.63(5)

0.6642(5) 0.9324(7) 0.4 0.3325(6) 0.4393(7) 0.4 0.16 0.16 0.1641(21) 0.7476(20) 0.16 0.0823(13) 0.7575(19) 0.2528(18) 0.16 9.6017(4) 5.57189(17) 9.0685(3) 90.4059(23) 485.148(29) 0.57(5)

0.6656(6) 0.9319(7) 0.45 0.3321(7) 0.4395(8) 0.45 0.2 0.2 0.1627(24) 0.7483(22) 0.2 0.0819(15) 0.7580(21) 0.2520(20) 0.2 9.6001(4) 5.57122(18) 9.0674(3) 90.3922(28) 484.95(3) 0.50(5)

0.664(1) 0.933(2) 0.5 0.334(1) 0.439(2) 0.5 0.6 0.6 0.169(4) 0.749(4) 0.6 0.078(3) 0.753(4) 0.252(3) 0.6 9.5965(6) 5.56972(25) 9.0639(5) 90.391(5) 484.45(5) 0.26(5)

0.4357(4) 0.5(3) 0.5(3) 0.840371(13) 0.680744(26) 0.5(3) 5.54831(15) 9.0492(3) 241.247(13) 0.33(5)

0.4360(4) 0.5(3) 0.5(3) 0.8401(11) 0.6801(21) 0.5(3) 5.54823(16) 9.0492(3) 241.241(13) 0.39(5)

0.4358(4) 0.5(3) 0.5(3) 0.8403(9) 0.6806(19) 0.5(3) 5.54829(16) 9.0495(3) 241.254(13) 0.47(5)

0.43616(19) 0.5(3) 0.5(3) 0.8403(4) 0.6806(8) 0.5(3) 5.54848(12) 9.04943(24) 241.268(10) 0.71(5)

210

215

298

– – – – – – – – – – – – – – – – – – – – 0

– – – – – – – – – – – – – – – – – – – – 0

– – – – – – – – – – – – – – – – – – – – 0

cella

Ambient temperature phase, hexagonal cellb Er (z) – 0.4354(4) Er (Uiso ) – 0.5(3) Mn1 (Uiso ) – 0.5(3) Mn2 (x) – 0.8408(12) Mn2 (z) – 0.6816(24) Mn2 (Uiso ) – 0.5(3) ˚ a (A) – 5.54891(15) ˚ c (A) – 9.0492(3) ˚ 3) V (A – 241.299(13) Weight fraction 0 0.30(5) Goodness-of-fit parameters Rwp (%) 8.26 Rp (%) 6.04

8.43 6.36

8.83 6.68

9.33 7.00

9.64 7.42

9.74 7.37

0.43594(15) 0.29(8) 0.2(1) 0.8403(3) 0.6806(6) 0.2(1) 5.54833(22) 9.0489(4) 241.241(16) 0.97(5) 10.40 7.45

0.43595(16) 0.31(8) 0.03(12) 0.8395(3) 0.6791(7) 0.03(12) 5.54820(10) 9.04840(18) 241.216(8) 0.96(5) 11.12 8.13

0.43642(6) 0.713(18) 0.53(3) 0.83931(15) 0.67861(29) 0.53(3) 5.55357(3) 9.05368(7) 241.8241(27) 0.96(2) 9.74 6.74

Calculated/estimated standard deviations in parentheses. During the refinements the isotropic temperature factors for the atoms the same type, Er or Mn, were constrained to be equal. No other constraints were applied in the refinements. a Atomic positions in the monoclinic cell, space group: C2/m (no. 12): Er1, Er2, and Mn3 in 4i (x, 0, z), Mn1 in 2a (0, 0, 0), Mn2 in 2c (0, 0, 1/2), and Mn4 in 8j (x, y, z). b Atomic positions in the hexagonal cell, space group: P6 /mmc (no. 194): Er in 4f (1/3, 2/3, z), Mn1 in 2a (0, 0, 0), and Mn2 in 6h (x, 2x, 1/4). 3

J.P. Maehlen et al. / Journal of Alloys and Compounds 437 (2007) 140–145

Low temperature phase, monoclinic Er1 (x) 0.66341(13) Er1 (z) 0.93520(12) Er1 (Uiso ) 0.32(1) Er2 (x) 0.33296(13) Er2 (z) 0.43652(12) Er2 (Uiso ) 0.32(1) Mn1 (Uiso ) 0.24(3) Mn2 (Uiso ) 0.24(3) Mn3 (x) 0.1639(4) Mn3 (z) 0.7479(4) Mn3 (Uiso ) 0.24(3) Mn4 (x) 0.0818(3) Mn4 (y) 0.7569(4) Mn4 (z) 0.2520(3) Mn4 (Uiso ) 0.24(3) ˚ a (A) 9.61247(9) ˚ b (A) 5.57558(4) ˚ c (A) 9.07102(8) β 90.5451(5) ˚ 3) V (A 486.140(7) Weight fraction 0.96(1)

J.P. Maehlen et al. / Journal of Alloys and Compounds 437 (2007) 140–145

143

Table 2 Interatomic distances in the monoclinic phase (T = 150 K) and hexagonal phase (T = 298 K) derived from the Rietveld refinements of SR-XRD data for ErMn2 D2 T = 150 K Vector

T = 298 K Length

Vector

Length

˚ distance below dc = 2.7 A ˚ is (a) Mn–Mn interatomic distances (in A, shown in bold) Mn1–Mn3 2.789(4) Mn1–Mn2 2.7409(8) Mn1–Mn4 2.766(3) Mn2–Mn3 2.733(3) Mn2–Mn4 2.748(3) Mn3–Mn4 2.723(3) Mn2–Mn2 2.677(3) Mn3–Mn4 2.833(3) Mn2–Mn2 2.876(3) Mn4–Mn4 2.711(5) Mn4–Mn4 2.865(5) ˚ (b) Er–Er interatomic distances (in A) Er1–Er1 3.364(3) Er1–Er1 3.4479(12) Er1–Er2 3.3728(10) Er2–Er2 3.401(2) Er2–Er2 3.417(2) ˚ (c) Er–Mn interatomic distances (in A) Er1–Mn1 3.2559(7) Er1–Mn1 3.2832(11) Er2–Mn2 3.2575(13) Er2–Mn2 3.2655(6) Er1–Mn3 3.265(2) Er1–Mn3 3.306(3) Er2–Mn3 3.2513(2) Er2–Mn3 3.273(4) Er1–Mn4 3.228(3) Er1–Mn4 3.286(4) Er1–Mn4 3.311(3) Er2–Mn4 3.223(3) Er2–Mn4 3.264(4) Er2–Mn4 3.265(3)

Er–Er

3.3756(11)

Er–Er

3.4068(4)

Er–Mn1

3.25761(10)

Er–Mn2 Er–Mn2 Er–Mn2

Calculated standard deviations are given in parentheses.

3.2500(3) 3.2890(8) 3.2500(3)

Fig. 3. Diffraction pattern of the monoclinic ErMn2 D2 (s.g. C2/m) collected at 150 K showing observed (crosses), calculated (upper line) and difference (bottom line) plots. The positions of the Bragg peaks are shown as ticks (upper: ErD2±x , middle: Er2 O3 , bottom: ErMn2 D2 ). Inset: selected region of the diffraction pattern, 7.5–9.0◦ ; splitting of three originally single peaks (1 1 0)hex , (1 0 3)hex and (2 0 0)hex [compare with Fig. 1] is clearly seen. The Bragg indexes for the monoclinic cell are given.

Fig. 5). An example of a fitted pattern (collected at 200 K) is presented in Fig. 6, while the results from the refinements are summarised in Table 1. The two-phase region was experimentally observed at temperatures between 185 and 205 K. The appearance of the monoclinic phase at 205 K is clearly visible in the diffraction pattern (Fig. 7). The upper temperature limit of the existence of the monoclinic ErMn2 D2 agrees rather well with the temperature of the magnetic ordering transition reported in [1] (∼214 K). It cannot be ruled out that the two-phase region in the low temperature area extends to temperatures below 185 K before a complete transformation is achieved at 150 K. The weight fraction ratio of the two phases present in the mixture and the changes

Fig. 4. Scheme of the transformation of the unit cell of the hexagonal phase of ErMn2 D2 observed at room temperature (left) into the monoclinic low temperature one (right; the original hexagonal cell is shown with the dashed lines). During this transformation, the Er position splits into two positions (Er → Er1 and Er2), and the two Mn positions both split into two (Mn1 → Mn1 and Mn2; Mn2 → Mn3 and Mn4).

144

J.P. Maehlen et al. / Journal of Alloys and Compounds 437 (2007) 140–145

Fig. 5. Evolution of the in situ SR-XRD patterns of the ErMn2 D2 as a function ˚ This plot of temperature (T = 185–215 K in steps of T = 5 K; λ = 0.3750 A). contains a smoothened subset of the graphs to give an impression of the time development of the process.

Fig. 8. Relative abundances of the hexagonal and monoclinic phases (a), and changes of unit cell volumes (b) as a function of temperature. For the hexagonal phase the doubled values of V are presented to have them in the same scale with the values for the monoclinic cell.

Fig. 6. Example of the refinements of the diffraction pattern of ErMn2 D2 collected in the two-phase region. This plot shows the data obtained at T = 200 K showing observed (crosses), calculated (upper line) and difference (bottom line) plots. The positions of the Bragg peaks are shown as ticks (from top to bottom: hexagonal ErMn2 D2 , ErD2±x , Er2 O3 , monoclinic ErMn2 D2 ).

Fig. 7. Diffraction pattern of the ErMn2 D2 sample collected at 205 K. The presence of the monoclinic phase is clearly visible. Observed intensities are represented as crosses, the calculated intensities (upper line) and difference (bottom line) as lines.

of the unit cell volumes as a function of temperature are shown in Fig. 8. We note opposite trends observed in changes of the unit cell volumes on cooling. For the hexagonal phase, a contraction takes place (Vhexagonal is −0.24% upon cooling from 298 down to 200 K). In contrast, for the monoclinic phase the cooling is associated with a rather pronounced expansion (Vmonoclinic equals to 0.35% upon cooling from 205 down to 150 K).

Fig. 9. Changes of the unit cell parameters and angle β of the monoclinic unit cell of ErMn2 D2 as a function of temperature between 150 and 205 K.

J.P. Maehlen et al. / Journal of Alloys and Compounds 437 (2007) 140–145

Fig. 9 shows in detail the changes of the unit cell parameters of the monoclinic and hexagonal ErMn2 D2 between 150 and 205 K. The driving force for this transformation is probably strongly related to the behaviour of hydrogen (deuterium) atoms in the structure. At high temperature, the hydrogen atoms freely diffuse between the most preferable for hydrogen accommodation interstitial sites in the hexagonal structure. Lowering of the temperature leads to hydrogen ordering in the lattice via “freezing” of hydrogen atoms in the specific interstitial sites, which, in turn, creates a distortion of the structure. It is not possible to find the ordering of hydrogen in the material using SR-XRD. Neutron diffraction measurements on this sample aiming to resolve the type of the ordering of deuterium in this compound are required. Acknowledgements This paper is in part supported by the Polish Research Council, Norwegian Research Council and Nordic Energy Research (project NORSTORE). We wish to thank H. Emerich and other members of the scientific staff at the Swiss-Norwegian Beam Line for their skilful assistance during the SR-XRD experiments. We are grateful to Dr. M. Sato (Tokai University, Japan and IFE) and T. Foerde (IFE) for their help in the experimental measurements.

145

References ˙ [1] H. Figiel, A. Budziak, P. Zachariasz, J. Zukrowski, G. Fischer, E. Dormann, J. Alloys Compd. 368 (2004) 260. ˙ [2] H. Figiel, A. Budziak, J. Zukrowski, G. Wiesinger, B. Ouladdiaf, J. Magn. Magn. Mater. 272–276 (2004) 585. [3] H. Figiel, J. Przewo´znik, V. Paul-Boncour, A. Lindbaum, E. Gratz, M. Latroche, M. Escorne, A. Percheron-Guegan, P. Mietniowski, J. Alloys Compd. 274 (1998) 29. [4] H.M. Rietveld, J. Appl. Crystallogr. 2 (1969) 65. [5] A.C. Larson, R.B. von Dreel, General Structure Analysis System, LANL, Los Alamos, 1994. [6] C.J. Howard, J. Appl. Crystallogr. 15 (1982) 615. [7] S.P. Thompson, E.D. Cox, J.B. Hastings, J. Appl. Crystallogr. 20 (1987) 79. [8] L.W. Finger, D.E. Cox, A.P. Jephcoat, J. Appl. Crystallogr. 27 (1994) 892. [9] O.L. Makarova, I.N. Goncharenko, A.V. Irodova, I. Mirebeau, E. Suard, Phys. Rev. B 66 (2002) 104423. [10] T. Roisnel, J. Rodriguez-Carvajal, WinPLOTR: a Windows tool for powder diffraction patterns analysis. Material Science Forum, Proceedings of the Seventh European Powder Diffraction Conference (EPDIC7), 2000, p. 118–123. Ed. R. Delhez and E. J. Mittenmeijer. [11] R. Shirley, The CRYSFIRE System for Automatic Powder Indexing: User’s Manual, The Lattice Press, Guildford, Surrey, England, 2000.