Chromium Environment within Cr-Doped BaAl2O4: Correlation of X

Nov 20, 2015 - ABSTRACT: Powder BaAl2O4 samples doped with 0 and 1.76 atom % Cr in relation to Al were hydrothermally prepared. Both samples were char...
1 downloads 12 Views 2MB Size
Article pubs.acs.org/IC

Chromium Environment within Cr-Doped BaAl2O4: Correlation of X‑ray Diffraction and X‑ray Absorption Spectroscopy Investigations Martina Vrankić,† Biserka Gržeta,*,† Dirk Lützenkirchen-Hecht,‡ Sanja Bosnar,§ and Ankica Šarić§ †

Division of Materials Physics and §Division of Materials Chemistry, Ruđer Bošković Institute, Bijenička 54, P.O. Box 180, HR-10002 Zagreb, Croatia ‡ Fachbereich C − Physik, Bergische Universität Wuppertal, Gauss Strasse 20, D-42097 Wuppertal, Germany ABSTRACT: Powder BaAl2O4 samples doped with 0 and 1.76 atom % Cr in relation to Al were hydrothermally prepared. Both samples were characterized by X-ray diffraction and synchrotron based X-ray absorption spectroscopy at the Cr K- and the Ba L3-edge. Diffraction patterns indicated that samples were nanocrystalline with a hexagonal crystal structure, space group P63. Chromium doping of barium aluminate caused an increase of the unit-cell volume and diffraction line broadening. The doped sample contained a small amount of an impurity phase, namely, BaCrO4. Analyzed Cr K-edge X-ray absorption near edge structure for the doped sample showed the presence of chromium in 6+ and 3+ oxidation states: Cr6+ was characteristic for chromium in the impurity phase BaCrO4, while Cr3+ participated in the formation of the doped phase BaAl2O4:Cr. Extended Xray absorption fine structure suggested an unusual tetrahedral coordination of Cr3+ ions within the BaAl2O4 host phase. The structure of samples was refined by the Rietveld method, simultaneously with the analysis of diffraction line broadening. Rietveld structure refinement showed that in doping the Cr3+ ions likely substituted for Al3+ ions on Al1 tetrahedral sites of barium aluminate crystal lattice. Crystallite sizes in the samples decreased with chromium doping, from 32 nm for pure BaAl2O4 to 24 nm for Cr-doped BaAl2O4. The dopant Cr3+ cations acted as defects in the barium aluminate structure that increased lattice strain from 0.02% for pure BaAl2O4 to 0.14% for doped BaAl2O4 and disturbed the crystallites to grow. derived from β-tridymite by substituting Si4+ cations in SiO4 tetrahedra by trivalent cations, followed by the introduction of large divalent cations into the channels running perpendicular to the layers of six-sided rings of the tetrahedra. In the case of the BaAl2O4 structure, Al3+ cations replace Si4+ cations in the tetrahedra of β-tridymite, and large Ba2+ cations occupy sites in the hexagonal channels formed by corner-sharing AlO4 tetrahedra.15,6 Huang et al. reported the existence of two hexagonal phases of BaAl2O4 with a reversible phase transition at 123 °C.6 At room temperature, the ferroelectric phase (hexagonal crystal system, space group P63, a = 10.449(1) Å, c = 8.793(1) Å) is observed, while at high temperature the paraelectric phase (hexagonal crystal system, space group P6322 a = 10.447(2) Å, c = 8.799(1) Å) is present.6 In the structure of the RT barium aluminate phase two different barium sites, Ba1 and Ba2, are positioned on Wyckoff positions 6c and 2a respectively, and coordinated by nine oxygen ions with average Ba−O distances of 2.86 Å for Ba1 and 2.87 Å for the Ba2 site. There are four different aluminum sites, Al1 to Al4, with average Al−O distances of 1.77, 1.74, 1.72, and 1.83 Å, respectively. Considering the ferroelectric room temperature phase, there are two different kinds of AlO4 tetrahedra: the first

1. INTRODUCTION Barium aluminate, BaAl2O4, is a material that has many applications in the fields of electronics and optical communications. It is used in the production of modern fluorescent lamps, cathode ray tubes, field emission displays (FEDs), plasma display panels (PDPs) and fiber amplifiers.1,2 BaAl2O4 doped with transition metal and/or rare earth ions displays a long afterglow luminescence.3,4 This makes BaAl2O4 interesting as a potential phosphor host material for the production of a new generation of phosphorescent materials.5 Furthermore, BaAl2O4 is a high melting point (1815 °C) material with good ferroelectric,6,7 catalytic,8 and advantageous hydraulic hardening properties,9,10 which makes it useful in various technologies. Barium aluminate belongs to the family of compounds with a stuffed β-tridymite structure.11 Tridymite is known as a polymorph of silicon dioxide, SiO2. It can occur in seven different crystalline forms.12−14 The two most common forms at standard pressure are a low-temperature orthorhombic αtridymite phase and a high-temperature hexagonal β-tridymite phase. The structure of β-tridymite is based on layers of sixsided rings made of SiO4 tetrahedra with the individual tetrahedra alternately pointing up and down. The layers are stacked so that the upward pointing tetrahedra from one layer share oxygen atoms with the downward pointing tetrahedra from the layer above. The stuffed β-tridymite structure is © XXXX American Chemical Society

Received: June 19, 2015

A

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

composition of the doped sample was also checked by synchrotron based X-ray fluorescence (XRF) spectroscopy using the setup described below. Prepared powder samples were characterized by X-ray powder diffraction (XRD) at room temperature using a Philips MPD 1880 counter diffractometer with graphite-monochromatized Cu Kα1α2 radiation (λCu‑Kα1 = 1.54056 Å, λCu‑Kα2 = 1.54439 Å). Two data sets were recorded for each prepared powder sample: (i) XRD pattern of the sample mixed with a molybdenum powder (99.999%, Koch-Light Lab. Ltd., UK) as an internal standard reference material, scanned in steps of 0.02° (2θ) in the 2θ range from 10° to 100° with a fix counting time of 7 s per step, for the purpose of the precise determination of unit-cell parameters; and (ii) XRD pattern of the pure sample, scanned in steps of 0.02° (2θ) in the 2θ range from 10° to 120° with a fix counting time of 7 s per step, for the purpose of the Rietveld structure refinement39 and size-strain analysis. Unit-cell parameters a and c of the prepared powder samples were determined using the UNITCELL program40 and refined by the whole-powderpattern fitting method using the WPPF program.41 Crystal structure refinement and size-strain analysis were performed by the Rietveld method with the X’Pert HighScore Plus program42 using a pseudoVoigt profile function and a polynomial background model. Isotropic vibration modes were assumed for all atoms. For the purpose of sizestrain analysis, the silicon powder (99.999%, Koch-Light Lab. Ltd., UK; spherical particles with diameter of 1 μm) was used as a standard for instrumental diffraction line broadening. The crystallite size and the lattice strain in the same sample were determined simultaneously with the Rietveld structure refinement. X-ray absorption spectroscopy (XAS) at the Cr K-edge for sample S1 was used to determine the oxidation state of absorbing chromium and to investigate a local short-range order around it in the sample. Therefore, X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) regions obtained from the room temperature XAS measurements were analyzed. Also, the Ba L3-edge XANES spectrum for both samples was measured for the purpose of identifying possible impurity phases containing barium in the samples, and to determine their contents. XANES and EXAFS measurements at the Cr K-edge and the Ba L3-edge were performed on the hard X-ray beamline BL10 at the superconducting asymmetric wiggler at the 1.5 GeV Dortmund Electron Accelerator DELTA, equipped with a channel-cut Si(111) monochromator,43 and on the Materials Science Endstation of the Rossendorf Beamline (BM20)44 at the European Synchrotron Radiation Facility (ESRF) in Grenoble under dedicated ring operating conditions (6 GeV, 130−200 mA). Transmission mode experiments were not productive for the Crdoped BaAl2O4 sample due to the overlap of the Ba L1-edge and Cr Kedge with an identical energy of 5989 eV. Thus, fluorescence mode experiments on the Cr K-edge (5989 eV) of Cr-doped BaAl2O4 sample were conducted using a Peltier-cooled silicon drift diode with a multichannel analyzer as detector (Amptek, Bedford, USA). Repeated scans each of typically 1 h acquisition time were performed and averaged in order to improve the data statistics. XANES data were analyzed by means of a linear combination fitting procedure using the WinXAS software.45 The Cr K-edge spectra of BaCrO4 and Cr2O3 were used as references for the linear fits. Structural information for the sample S1 were extracted by the Fourier-filtering of experimental EXAFS oscillations data into distance-space, giving a radial distribution function from which the interatomic distances and coordination numbers were determined.46 EXAFS data reduction and analysis were performed using the IFEFFIT package,47,48 and phases and amplitudes calculated by FEFF 8.049 were used for the quantitative fits.

one containing Al1 and Al2 atoms forming the Al1−O−Al2 angle close to 156°, and the second one containing the Al3 and Al4 atoms along the 3-fold axis and constituting the Al3−O− Al4 angle of 180°.6,16 Chromium doped systems are used in a production of tunable solid-state lasers,17 high temperature sensors,18,19 and high-pressure calibrants.20 Optical properties of Cr-doped barium aluminates are relatively well investigated, but the structure of these doped materials has not yet been experimentally studied. Red fluorescence from BaAl2O4:Cr3+ as well as the influence of the surface area on fluorescence was reported by Singh et al.4 It is known that Cr3+ dopant ions in inorganic compounds exhibit a strong tendency to possess an octahedral coordination. On the contrary, Cr3+ ions in tetrahedral coordination are very rare.21−26 The four-coordinated Cr3+ ions have been shown to exist in systems such as Cr-bearing blue-colored diopsides, Cr-heteropolytungstate, Na4CrO4, several Cr-containing inverted spinels, Cr-doped Bi12SiO20, and several organometallic compounds.27−34 One of the fundamental problems in the structure determination of Cr-doped materials is the potential coexistence of several oxidation states of chromium in the materials under study (Cr2+, Cr3+, Cr4+, Cr5+, or Cr6+), making the spectroscopic fingerprinting of tetrahedral Cr3+ difficult.35−37 To the best of our knowledge, no experimental study has been performed on the Cr-doped BaAl2O4 structure up to now. In the present work we report the synthesis of the Cr-doped BaAl2O4 powder sample via a hydrothermal route and the structural characterization of the prepared sample by means of X-ray powder diffraction and X-ray absorption spectroscopy (XANES and EXAFS). The main emphasis is put on determination of the oxidation state of the Cr dopant and on its coordination environment in the structure of the doped sample.

2. EXPERIMENTAL SECTION 2.1. Materials and Synthesis. Powder samples of pure BaAl2O4 (sample S0) and one doped with chromium (sample S1) in the amount of 1.76 atom % in relation to aluminum were prepared using a hydrothermal route followed by a thermal treatment. Analytical reagent grade barium nitrate, Ba(NO3)2 (Fisher Chemical, USA), aluminum nitrate nonahydrate, Al(NO3)3·9H2O (Fisher Chemical, USA), chromium nitrate nonahidrate, Cr(NO3)3·9H2O (SigmaAldrich, UK), citric acid monohydrate, C6H8O7·H2O (Kemika, Croatia), and ammonium hydroxide, NH3·aq (25%) (Kemika, Croatia), were used for the preparation of samples. Stoichiometric amounts of barium nitrate, aluminum nitrate nonahydrate, and chromium nitrate nonahydrate were dissolved in Milli-Q water (prepared in the own laboratory) at room temperature. The solutions were used to prepare two mixtures: one for producing sample S0, and the other for producing sample S1. Citric acid was added to each mixture, and the mixtures were homogenized. Then the pH value of the mixtures was adjusted to 10 by addition of ammonium hydroxide, and they were subjected to precipitation in an autoclave at 170 °C during 24 h. Precipitates were separated from supernates by centrifugation, washed several times with Milli-Q water, and dried at 60 °C subsequently. The obtained powders were heated up to 1100 °C in a furnace with static air, with a heating rate of 10 °C/min, and calcined at that temperature for 4 h. 2.2. Measurements and Characterization. The chromium concentration in Cr-doped BaAl2O4 sample was determined by means of particle induced X-ray emission (PIXE) spectroscopy, using a nuclear microprobe facility with a 3 MeV proton beam and a semiconductor Si(Li) X-ray detector.38 The K-series of emitted radiation from thin powder samples were used for the analysis. The

3. RESULTS AND DISCUSSION 3.1. XRD Characterization of Samples. XRD patterns of the prepared samples S0 and S1 (Figure 1) indicated that the samples possess a hexagonal structure characteristic for the room temperature, ferroelectric, BaAl2O4 phase (space group P63).6 No impurities were detected in sample S0, while sample S1 contained a small amount of an impurity phase, namely, B

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. XRD patterns of samples S0 and S1 and graphical representation of ICDD powder diffraction data50 for BaAl2O4 (card no. 82-1349). Triangles denote visible diffraction lines of BaCrO4 (ICDD card no. 78-1401).

BaCrO4 (orthorhombic, space group Pnma).51 Crystalline phases were identified according to the data existing in the ICDD Powder Diffraction File:50 card no. 82-1349 for BaAl2O4 and card no. 78-1401 for BaCrO4. Diffraction lines were broadened indicating that prepared samples were nanocrystalline. Refined unit-cell parameters a and c of pure BaAl2O4 (denoted as phase BA, sample S0) and of Cr-doped BaAl2O4 (denoted as phase BAC in sample S1) are listed in Table 1 along with the literature data for BaAl2O4 at RT.6 On Crdoping, the unit-cell parameter a slightly increased, the parameter c slightly decreased, while unit-cell volume slightly increased. If we consider the ionic radii for the four-coordinated Al3+ (0.39 Å),53 four-coordinated Cr3+ (0.47−0.53 Å)22 and four-coordinated Cr6+ (0.26 Å),53 such behavior of the cell parameters may indicate a possibility that Cr3+ or both Cr3+ and Cr6+ were substituted for Al3+ in the BAC phase of sample S1. It should be noted here that a substitution of some of chromium cations for the Ba2+ cation in the structure of BaAl2O4 would be rather impossible according to the first Goldschmidt’s rule.54 Namely, a difference in the ionic radius of nine-coordinated Ba2+ (1.47 Å), and the radius of any conceivable chromium cation (Cr2+, Cr3+, Cr4+, Cr5+, and Cr6+) possessing any possible coordination reported by Shannon53 is larger than 38%. However, a dramatic change in the unit-cell volume of the barium aluminate phase on Crdoping was not observed, and thus the replacement of Ba2+ ions by chromium ions in the BaAl2O4 lattice is very unlikely and can be ruled out here. 3.2. X-ray Absorption Spectroscopy (XAS). XAS at the Ba L3-edge. As found by XRD, the sample S1 contained the

Figure 2. (a) Comparison of the Ba L3-edge XANES spectra of pure BaAl2O4, pure BaCrO4 and sample S1. (b) Linear combination fit for the experimental XANES spectrum of sample S1 using the Ba L3-edge spectra of pure BaAl2O4 and pure BaCrO4 as references for the fit.

BaAl2O4-type phase and a small amount of the impurity phase BaCrO4. The Ba L3-edge XANES spectra of pure BaAl2O4, pure BaCrO4 and the sample S1 are compared in Figure 2a. Differences are obvious in the intensity of the white line at about 5248 eV photon energy, where the most intense signal is detected for the BaAl2O4 sample, while BaCrO4 reveals a substantially lower white line intensity. Further distinct differences between these two spectra are also visible in the energy range between 5250 and 5270 eV; i.e., BaCrO4 has an absorption minimum at about 5255 eV, where BaAl2O4 features a visible maximum. Also, a minimum is obvious for BaAl2O4 at 5264 eV, whereas the absorption shows an opposite behavior for BaCrO4 at this energy. As can be seen in this comparison, the Ba L3-edge XANES spectrum for sample S1 mainly follows the spectrum of the pure BaAl2O4 sample, however with a

Table 1. Sample and Doped BaAl2O4 Phase Notation, Chromium Doping Level and Refined Values of Unit-Cell Parameters for Doped Phases, along with the Literature Data for Pure BaAl2O4a sample BaAl2O4 S0 S1 a

b

Cr content in sample (atom %)

doped BaAl2O4 phase

Rp

Rwp

a (Å)

c (Å)

V (Å3)

0 0 1.76(5)

BA BAC

0.078 0.061 0.058

0.170 0.081 0.079

10.449(1) 10.4488(4) 10.4539(5)

8.793(1) 8.7959(5) 8.7954(4)

831.44(1) 831.66(1) 832.42(1)

Rp and Rwp are the discrepancy factors that characterize a quality of the fit.52 bData cited from Huang et al., 1994.6 C

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 3. X-ray fluorescence spectra of sample S1 (containing 1.76 atom % Cr in relation to Al) obtained with different excitation energies in the vicinity of the Cr K-edge/Ba L1-edge as indicated. Specific emission lines from Ba and Cr are assigned. Emission intensity in the range between ∼5300 eV and ∼5600 eV is mainly caused by the Crfluorescence, denoted as a region of interest (ROI).

Figure 5. (a) Fitting result for EXAFS data of pure BaCrO4 at Cr Kedge using a single Cr−O shell to model the experimental data. (b) Fitting result for EXAFS data of sample S1 at Cr K-edge. A model consisting of two different oxygen shells was used to fit the experimental data: the first one to model the Cr−O shell in the BaCrO4 phase (with a Cr6+−O2− distance of ∼1.64 Å) and the second one to model the Cr−O shell in Cr-doped BaAl2O4 phase BAC (with a possible Cr3+−O2− distance of ∼1.73 Å). The increase of the Cr−O bond length for the second shell can be followed by the vertical dotted lines.

the experimental spectrum of sample S1 can be well fitted by a superposition of the spectra of the pure barium reference compounds, giving the result that sample S1 contains 2.9(1) wt % of the impurity phase BaCrO4. XAS at the Cr K-Edge. As already mentioned in section 2.2, the analysis of the Cr K-edge X-ray absorption data is difficult due to the overlap of the Cr K-edge and the Ba L1-edge. Figure 3 presents the fluorescence spectra of sample S1 measured for different excitation energies as indicated. Besides the contributions from the elastic scattering, intense Ba signals are detectable at 4460 eV (Ba Lα1 and Ba Lα2), 4830 eV (Ba Lβ1), and 5150 eV (Ba Lβ2), and a weaker feature related to the Ba Lγ1 was found at 5530 eV. For the excitation energies above the Cr K-edge, a new feature related to the chromium centered at about 5450 eV develops. It should be stressed that this spectral feature is absent in the X-ray fluorescence spectra of pure BaAl2O4. Thus, we have used a region of interest (ROI) from 5300 to 5600 eV, to minimize the influence of Ba on the Cr K-edge data.

Figure 4. (a) Cr K-edge XANES spectra of sample S1, pure BaCrO4, and pure Cr2O3. The data are normalized and shifted with respect to each other along the ordinate. (b) Difference XANES spectrum between the XANES spectra of sample S1 and of BaCrO4, generated to enable estimation of Cr3+ content in sample S1. XANES spectrum of Cr2O3 with a reduced IF/I1 value is presented for comparison.

small, but significant, change in the white line intensity, where a slightly reduced absorption is detectable, as well at 5255 eV. We thus performed a linear-combination fit of the experimental spectrum of sample S1 with those of pure BaAl2O4 and pure BaCrO4, and the result is shown in Figure 2b. As can be seen, D

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 2. Results of XANES and EXAFS Analysis at the Cr K-Edge for Pure BaCrO4 and Sample S1 Cr K-edge XANES sample BaCrO4 S1

Cr species 6+

Cr Cr6+ Cr3+

Cr K-edge EXAFS

E0 (eV)

share of Cr species in total Cr content (%)

Ri (Å)

Ni

σi2 (Å2)

share of Cr species in total Cr content (%)

6006.6 6006.5 5998.6

100 87(1) 13(1)

1.64(2) 1.64(2) 1.73(9)

3.75(2) 3.49(2) 0.73(2)

0.0020 0.0020 0.0022

100 83(6) 17(3)

indications from X-ray diffraction that Cr2O3 was present in sample S1. Nevertheless, this first guess hints to the presence of Cr3+ within the sample. We have thus determined a difference spectrum between the XANES spectrum of sample S1 and that of the pure BaCrO4 reference in order to provide a direct proof for the existence of Cr3+ species within sample S1. Assuming that the pre-edge peak intensity for the Cr3+ species around 5990 eV is generally small compared to that of the tetrahedral Cr6+ (see, e.g., Beale et al., 2006,21 Tromp et al., 2007,55 Wei et al., 200956), the concentration of BaCrO4 was adjusted so that the pre-edge intensity matches well the experimental data of sample S1. It should be noted here that the difference spectrum derived for the substituted Cr (shown in Figure 4b) is similar to that of the Cr3+-species in Cr2O3 because of its edge-position of 5998.6 eV, which is approximately identical to that of Cr2O3 (5999.1 eV) and substantially smaller than that of Cr6+ (6006.6 eV), and also due to the absence of intense pre-edge features. However, the overall shape above the edge is different from the Cr2O3reference spectrum. For example, a broad local absorption minimum is detected at about 6035 eV for Cr2O3, while the derived difference spectrum for Cr in the doped BaAl2O4 phase reveals a maximum in this energy range. It can thus be assumed that the chromium incorporated in the BaAl2O4 structure is likely present as Cr3+ ion, but with a different coordination than in Cr2O3. From the determined contribution of BaCrO4 in sample S1 and the difference spectrum determined in Figure 4b, the concentrations of both Cr species in the sample S1 could be determined. It follows that ∼87% of the total chromium content in sample S1 is present as Cr6+, forming 2.9 wt % BaCrO4, while the remaining ∼13% of the total chromium content is related likely to Cr3+ incorporated in the BaAl2O4 host lattice. One may argue that, in addition to the Cr3+ species incorporated into the BaAl2O4 host phase and the Cr6+ species in BaCrO4 impurity phase, an additional Cr6+ species may be present in sample S1. However, such a species would reduce the contribution of BaCrO4 in the Cr K-edge spectra, in contradiction to the results obtained at the Ba L3edge and the results of the X-ray diffraction. On the basis of the results of the XANES data evaluation, we may therefore assume that two chromium species are present in the Cr-doped sample S1, namely, the Cr6+-species in the BaCrO4 phase and the Cr3+-species incorporated in the BaAl2O4 host. To obtain more detailed information on the Cr-dopant characteristics within the sample S1, EXAFS data of sample S1 and of pure BaCrO4 were analyzed. Extracted EXAFS functions χ(k)*k2 of sample S1 and of the impurity phase BaCrO4 were Fourier-transformed into R-space, and the first shell was isolated, transformed back into k-space and fitted with a Cr−O coordination. The interpretation of the EXAFS data for BaCrO4 was straightforward. The appropriate fitting result is shown in Figure 5a in the form of the Fouriertransform of the k2-weighted EXAFS function. Comparing the experimental data and Fourier-transformed data for sample S1

Figure 6. Graphical results of Rietveld refinement and line broadening analysis for samples S0 and S1. Triangles denote visible diffraction lines of BaCrO4. Presented FWHM values belong to pure BaAl2O4 (sample S1) and to Cr-doped BaAl2O4 phase (BAC phase in sample S1).

XANES spectra of reasonable quality could be obtained after averaging over 3−4 scans as can be seen in Figure 4a, where the XANES spectrum of sample S1 is shown in comparison to the spectrum of pure BaCrO4 (in which Cr6+ ion is tetrahedrally coordinated with O2− ions) and to the spectrum of pure Cr2O3 (in which Cr3+ ion is octahedrally coordinated with O2− ions, respectively). The presence of BaCrO4 in sample S1 was already deduced from the spectra obtained at the Ba L3-edge (see Figure 2). Fitting the XANES data of sample 1 with those of pure BaCrO4 showed that the white line feature of BaCrO4 at 5990 eV is much stronger than that of sample S1, indicating that the structures of these two samples are different. Accordingly, a second Cr-species seems to be present in sample S1. Using Cr2O3 as a reference for Cr3+ ion, possibly incorporated into the doped sample S1, the fit improved significantly, however still with distinct deviations between the experimental XANES data for sample S1 and the linear combination fit with BaCrO4 and Cr2O3 XANES data. Such a discrepancy could however be expected because there were no E

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Table 3. Results of Rietveld Structure Refinement and Size-Strain Analysis for BA Phase (Pure BaAl2O4, Sample S0) and for BAC Phase (Cr-Doped BaAl2O4 Phase in Sample S1)a doped BaAl2O4 phase

a

Cr content (atom %)

Rp

Rwp

BA

0

0.098

0.123

BAC

0.3

0.098

0.129

atom site

Wyck. position

occupancy

x

y

z

Ba1 Ba2 Al1 Al2 Al3 Al4 O1 O2 O3 O4 O5 O6 Ba1 Ba2 Al1 Cr Al2 Al3 Al4 O1 O2 O3 O4 O5 O6

2a 6c 6c 6c 2b 2b 6c 6c 6c 6c 6c 2b 2a 6c 6c 6c 6c 2b 2b 6c 6c 6c 6c 6c 2b

1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.992 0.008 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

0 0.504(1) 0.159(7) 0.155(5) 1/3 1/3 0.183(6) 0.674(5) 0.4918(1) 0.180(10) 0.119(3) 1/3 0 0.504(4) 0.159(8) 0.159(8) 0.155(6) 1/3 1/3 0.182(5) 0.676(2) 0.4920(6) 0.183(2) 0.116(2) 1/3

0 0.002(2) 0.335(5) 0.330(9) 2/3 2/3 0.007(11) 0.0018(2) 0.169(6) 0.503(12) 0.304(9) 2/3 0 0.002(1) 0.335(5) 0.335(5) 0.331(7) 2/3 2/3 0.006(2) 0.002(4) 0.169(6) 0.506(4) 0.304(9) 2/3

1/4 0.259(1) 0.061(7) 0.4442(9) 0.942(3) 0.551(3) 0.987(6) 0.030(5) 0.997(1) 0.000(9) 0.249(12) 0.747(10) 1/4 0.259(3) 0.061(9) 0.061(9) 0.444(2) 0.942(4) 0.552(4) 0.988(4) 0.030(5) 0.997(1) 0.000(9) 0.252(5) 0.748(9)

Biso

(Ǻ 2)

0.50(2) 0.50(2) 0.79(8) 0.79(8) 0.79(8) 0.79(8) 0.82(12) 0.82(12) 0.82(12) 0.82(12) 0.82(12) 0.82(12) 0.62(5) 0.62(5) 0.88(7) 0.88(7) 0.88(7) 0.88(7) 0.88(7) 0.97(10) 0.97(10) 0.97(10) 0.97(10) 0.97(10) 0.97(10)

crystallite size (nm)

lattice strain (%)

32.1(1)

0.02(1)

23.9(1)

0.14(1)

Rp and Rwp are the discrepancy factors that characterize a quality of the fit.52

Table 4. Metal−Oxygen Distances (Å) along with Al1−O5−Al2 and Al3−O6−Al4 Angles (deg) of BA Phase (Pure BaAl2O4, Sample S0) and of BAC Phase (Cr-Doped BaAl2O4 Phase in Sample S1)

(Figure 5b) with those of pure BaCrO4 (Figure 5a), it is evident that they differ from each other. Accordingly, we performed a modeling for the interpretation of the obtained EXAFS data for sample S1. It should be noted

here that all the peaks in the FT are generally shifted toward smaller radial distances R compared to their crystallographic distances because of the k-dependence of the phase shift in the scattering processes of the photoelectron.57 In the case of F

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

BaCrO4 impurity (as already seen in the first coordination shell) and the complex structure of the BaAl2O4 host. Assuming that both Cr6+ and Cr3+ ions are 4-fold coordinated with oxygen, the concentration of Cr3+ ions (substituting for Al3+ in BaAl2O4) and of Cr6+ ions (present in the impurity phase BaCrO4) in sample S1 can therefore be determined from the obtained coordination numbers N1 and N2. Namely, N2/(N1 + N2) and N1/(N1 + N2) give the concentrations of Cr3+ and Cr6+ (i.e., the parts of total Cr content) in sample S1, respectively. It follows that ∼17.3% of the total chromium content in sample S1 is present as Cr3+ incorporated in the host BaAl2O4 structure, while ∼82.7% of the total chromium content is present as Cr6+ in the impurity phase BaCrO4. The obtained Cr3+ concentration determined from the EXAFS data evaluation is slightly larger compared to that determined from the XANES data, as seen in Table 2. Taking the uncertainties of both evaluation procedures into account, it appears reasonable that about 15% of the total Cr content in sample S1 is 4-fold coordinated Cr3+ present in Crdoped BaAl2O4 (BAC phase). This means that the BAC phase was doped with ∼0.3 atom % Cr in relation to Al. Therefore, the chemical formula for the BAC phase may be proposed as IX [Ba2+]IV[Al3+0.997Cr3+0.003]2O4. 3.3. Rietveld Structure Refinement and Analysis of Diffraction Line Broadening. The Rietveld refinement of the prepared samples S0 and S1 was performed for several purposes: (1) to confirm that both phases BA (pure BaAl2O4, sample S0) and BAC (Cr-doped BaAl2O4 phase in sample S1) possess a hexagonal structure, space group P63, (2) to confirm the quantitative phase composition of sample S1 determined by X-ray absorption spectroscopy, (3) to elucidate the mode of Cr3+ incorporation into barium aluminate BaAl2O4, and (4) to determine the crystallite size and lattice strain in the phases BA and BAC. The refinement was started using a structure model for barium aluminate BaAl2O4 according to Huang et al. 19946 (ICSD card no. 75426)61 and formula for the Cr-doped BaAl2O4 phase as resulted from the XANES and EXAFS investigations of sample S1. Accordingly, the structure model for BAC phase in sample S1 included a partial substitution of Al3+ by Cr3+ in the structure of the host BaAl2O4. Also, in the Rietveld structure refinement of sample S1, the impurity phase BaCrO4 (orthorhombic, space group Pnma, ICSD card no. 62560)51 was included as a second phase. The refinement procedure for sample S0 involved refinement of background parameters, diffraction-line profile parameters, lattice parameters a and c, atomic position parameters, and temperature factors for all present atoms. Isotropic temperature factors (Biso) were used in the refinement; for the same kind of atoms they were constrained to change unanimously during the refinement. The refinement procedure for sample S1 included refinement of the above-mentioned parameters for the Crdoped BaAl2O4 phase, as well as the refinement of its scale factor. Furthermore, the isotropic temperature factors for Al3+ and Cr3+ cations sharing the same site were constrained to change identically during the refinement. For the BaCrO4 impurity phase only the scale factor, line profile parameters, and lattice parameters were refined. In the refinement for sample S1, the lowest values of discrepancy factors Rp and Rwp were achieved with the structure model for BAC phase in which Cr3+ substituted for Al3+ on the Al1 site solely. The results of the Rietveld structure refinement confirmed that both phases BA (pure BaAl2O4, sample S0) and BAC (Crdoped BaAl2O4 phase in sample S1) have the hexagonal

sample S1 (Figure 5b), the leading peak at a radial distance of ∼1.2 Å belongs to the tetrahedrally coordinated Cr6+ in BaCrO4 phase with a bond length R1 of 1.64 Å,58 but with a different coordination number than for BaCrO4. So we assumed an additional Cr−O shell for sample S1 to represent the contribution of the second chromium species in the sample, and the fits of the EXAFS data were performed using phases and amplitude functions calculated by FEFF.49 The k2-weighted EXAFS function χ(k)*k2 in the k-range from 2.03 Å−1 to 10.25 Å−1 was used for the fit, and the data in the radial distribution from 0.9 to 1.8 Å were fitted by a variation of the following parameters: bond lengths (R1, R2), nearest neighbor coordination numbers (N1, N2), disorder (σ1, σ2), and inner potential shift ΔE0. We have used the same value of ΔE0 for both shells here. Furthermore, the values for R1 = 1.64(2) Å and σ12 = 0.0020 Å2 were taken from the fit of the pure BaCrO4 sample; i.e., only R2, N1, N2, σ2, and ΔE0 were varied in the fitting procedure. The fitting result for sample S1 is shown in Figure 5b. As can be seen, the second Cr−O shell is required to model the data adequately; otherwise an unreasonably broadened main peak with an unphysical disorder σ1 would result from a sole single-shell model. Consequently, the broadened peak representing the oxygen shell again proves the presence of two different Cr species (with different Cr−O distances) in sample S1. Compared to the nearest neighbor coordination number determined for pure BaCrO4 (N1 = 3.75), a reduced N1 = 3.49 was found for BaCrO4 in sample S1, indicating on the coexistence of Cr in BaCrO4 and Cr in the host BaAl2O4 lattice. The second Cr−O shell with a substantially smaller coordination number N2 = 0.73 is located at a larger radius of R2 = 1.73(9) Å, which is close to the average Al−O bond distance in BaAl2O4 (RAl−O = 1.76(5) Å)6 suggesting that the second chromium species in sample S1 may be Cr3+ located on Al-sites in the barium aluminate host material substituting for Al3+. Such a substitution of tetrahedral Al3+ by Cr3+ has already been reported for crystalline aluminophosphates, where a 4-fold coordination of Cr3+ with an average Cr−O bond distance of 1.89 Å was determined.21 It is important to emphasize here that for the octahedral Cr− O bonds such as in Cr2O3, the Cr−O bond length is typically on the order of ∼2 Å.59,60 From our previous EXAFS experiments for pure Cr2O3, we derived values of 1.96(1) and 2.03(1) Å for the two slightly different sets of Cr−O bonds in the distorted coordination octahedron around Cr3+ ion in the Cr2O3 structure, which is in a good agreement with reported literature data.59,60 On the other hand, the presently found substantially reduced Cr K-edge energy (5998.6 eV) for Cr3+ in the BAC phase of sample S1, and the short Cr−O bond of 1.73(9) Å in that phase, indicated the presence of tetrahedrally coordinated Cr3+ in the Cr-doped BaAl2O4 phase substituted for Al3+, as well as the absence of Cr species of higher coordination substituted for Ba2+ (in agreement with the results and discussion in section 3.1.). One might argue that the position of Cr3+ within the BaAl2O4 lattice may be further refined using coordination shells at larger bond distances, that are obviously present in the Fourier transforms shown in Figure 5. Comparing the FT of the BaCrO4 sample and that of sample S1, distinct differences are visible for radii larger than ∼2 Å, indicating that the atomic short-range order around Cr is different in these two samples. However, a clear assignment of the EXAFS peaks to the individual coordination shells in sample S1 is difficult, due to the dominating signals of the G

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry structure, space group P63. Refinement for sample S1 confirmed the chemical formula Ba(Al0.997Cr0.003)2O4 for the BAC phase. The quantitative phase analysis based on Hill and Howard formalism62 showed that sample S1 also contained a small content of the impurity phase, namely, 2.8(1) wt % BaCrO4 which is in agreement with the result obtained by XANES investigations of sample S1 at the Ba L3-edge. The observed and calculated XRD patterns for samples S0 and S1 are presented in Figure 6 along with the values of full-widths at half-maximum (FWHM) for the BA phase (sample S0) and the BAC phase (in sample S1) in the wide range of the Bragg angle. Table 3 lists the refined structural parameters for the phases BA and BAC, and the results of diffraction-line broadening analysis performed simultaneously with the structure refinement. The refined metal−oxygen distances along with the values of Al1−O5−Al2 and Al3−O6−Al4 angles in the structure of the phases BA and BAC are listed in Table 4. As seen in Table 4, the average distances in (Al1)O4 tetrahedra increased with Cr3+ incorporation on the Al1 site, the average distances in (Al3)O4 tetrahedra decreased, while the average distances in (Al2)O4 and (Al4)O4 tetrahedra as well as average distances in (Ba1)O9 and (Ba2)O9 polyhedra were constant within the standard deviation. Also, the Al1−O5−Al2 angle decreased on Crdoping from the value of 156.0(8)° to 154.4(6)°. Such behavior of the interatomic distances and of Al1−O5−Al2 angle caused an increase of the unit-cell parameter a, shortening of the parameter c, but the overall increase of the unit-cell volume for the Cr-doped BaAl2O4 in comparison to the pure BaAl2O4. It should be noted here that the average (Al1/Cr)−O distance as obtained by the Rietveld refinement and the Cr3+−O distance resulted previously from the EXAFS analysis of sample S1 agreed within the standard deviation. The line broadening analysis showed that both samples S0 and S1 were nanocrystalline. The crystallite size in barium aluminate BaAl2O4 decreased from ∼32 nm to ∼24 nm on doping with only 0.3 atom % Cr in relation to Al. Simultaneously, the lattice strain increased from 0.02% to 0.14% upon doping. Obviously, the Cr3+ dopant ions acted as strong defects in the barium aluminate structure that increased lattice strain and disturbed crystallites to grow.

of transition metal ions to cope with unfavorable coordination environments in a host lattice, i.e., the barium aluminate BaAl2O4 in the present case. An important goal for the future investigations is the reduction of the BaCrO4 impurity phase, hopefully leading to more efficient chromium doping of BaAl2O4.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; tel.: +385 1 4561120; fax: +385 1 4680114. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Financial support from the Ministry of Science, Education and Sports of Republic of Croatia within the framework of the Project No. 098-0982886-2893, and the financial support from the Federal Ministry of Education and Research (BMBF) of Germany under Project No. 05K10PX1 are gratefully acknowledged. We would like to thank Ralph Wagner and Stefan Balk for help with the experiments performed at the DELTA storage ring, and Pascal Becker, Oliver von Polheim, and Carsten Baehtz for their help at the ESRF BM20. We also appreciate DELTA and the ESRF for the provision of beamtime.



REFERENCES

(1) Hyland, R. W. Jr.; Quintenz, J. P.; Dunville, B. T.; Subrahmanyam, G. U. S. Patent 6969475 B2, 2005. (2) Kim, C.; Kwon, I.; Park, C.; Hwang, Y.; Bae, H.; Yu, B.; Pyun, C.; Hong, G. J. Alloys Compd. 2000, 311, 33−39. (3) Lin, Y.; Zhang, Z.; Tang, Z.; Zhang, J.; Zheng, Z.; Lu, X. Mater. Chem. Phys. 2001, 70, 156−159. (4) Singh, V.; Chakradhar, R. P. S.; Rao, J. L.; Zhu, J.-J. Mater. Chem. Phys. 2008, 111, 143−148. (5) Lou, Z.; Hao, J.; Cocivera, M. J. Phys. D: Appl. Phys. 2002, 35, 2841−2845. (6) Huang, S. Y.; Von Der Mühll, R.; Ravez, J.; Chaminade, J. P.; Hagenmuller, P.; Couzi, M. J. Solid State Chem. 1994, 109, 97−105. (7) Bush, A. A.; Laptev, A. G. Sov. Phys.-Solid State (Engl. Transl.) 1989, 31, 535−536. (8) Lin, H.; Li, Y.; Shangguan, W.; Huang, Z. Combust. Flame 2009, 156, 2063−2070. (9) Ion, T.; Ciocea, N. Cemento 1980, 77, 3−10. (10) Ali, M. M.; Agarwal, S. K.; Agarwal, S.; Handoo, S. K. Cem. Concr. Res. 1995, 25, 86−90. (11) Glasser, F. P.; Glasser, L. S. D. J. Am. Ceram. Soc. 1963, 46, 377−380. (12) Kihara, K.; Matsumoto, T.; Imamura, M. Z. Kristallogr. 1986, 177, 27−38. (13) Graetsch, H. Am. Mineral. 1998, 83, 872−880. (14) Deer, W. A.; Howie, R. A.; Wise, W. S. Rock-Forming Minerals: Framework Silicates: Silica Minerals, Feldspathoids and the Zeolites; The Geological Society Publishing House: London, 2004; pp 22−44. (15) Hörkner, V. W.; Müller-Buschbaum, H. Z. Anorg. Allg. Chem. 1979, 451, 40−44. (16) Huang, S.-Y.; Von Der Mühll, R.; Ravez, J.; Couzi, M. Ferroelectrics 1994, 159, 127−132. (17) Struve, B.; Huber, G. J. Appl. Phys. 1985, 57, 45−48. (18) Zhang, Z. Y.; Grattan, K. T. V.; Palmer, A. W.; Fernicola, V.; Crovini, L. Phys. Rev. B: Condens. Matter Mater. Phys. 1995, 51, 2656− 2660. (19) Grattan, K. T. V.; Selli, R. K.; Palmer, A. W. Rev. Sci. Instrum. 1988, 59, 1328−1335. (20) Wamsley, P. R.; Bray, K. L. J. Lumin. 1994, 59, 11−17.

4. CONCLUSIONS Preparation, XRD and XAS examinations of the BaAl2O4 powder sample and the one doped with 1.76 atom % Cr (in relation to Al) have been reported. The prepared samples were nanocrystalline. They possessed hexagonal structure in the space group P63. Unit-cell parameter a increased, parameter c decreased, while the overall unit-cell volume increased on chromium doping. XANES and EXAFS investigation of the doped sample revealed that chromium incorporated in the BaAl2O4 host structure as tetrahedral Cr3+ ion, in the content of only ∼15% of the total chromium present in the doped sample. The remaining content of chromium formed an impurity phase in that sample, namely, 2.9 wt % BaCrO4. Rietveld structure refinement showed that in doping Cr3+ ion substituted for tetrahedral Al3+ ion on the Al1 site of the BaAl2O4 structure. The line broadening analysis exhibited that the lattice strain of BaAl2O4 increased with chromium doping while the crystallite size decreased, indicating that the Cr3+ dopant ions acted as strong defects in the BaAl2O4 structure, which disturbed crystallites to grow. The existence of tetrahedrally coordinated Cr3+ was directly proven by the structural investigations, and thus the results of the performed study illustrate the versatility H

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(59) Newnham, R. E.; de Haan, Y. M. Z. Z. Kristallogr. 1962, 117, 235−237. (60) Sawada, H. Mater. Res. Bull. 1994, 29, 239−245. (61) Inorganic Crystal Structure Database (ICSD); Fachinformationszentrum: Karlsruhe, Germany, 2007. (62) Hill, R. J.; Howard, C. J. J. Appl. Crystallogr. 1987, 20, 467−474.

(21) Beale, A. M.; Grandjean, D.; Kornatowski, J.; Glatzel, P.; de Groot, F. M. F.; Weckhuysen, B. M. J. Phys. Chem. B 2006, 110, 716− 722. (22) Ikeda, K.; Yagi, K. Contrib. Mineral. Petrol. 1977, 61, 91−106. (23) Ikeda, K.; Yagi, K. Contrib. Mineral. Petrol. 1982, 81, 113−118. (24) Schreiber, H. D. Am. Mineral. 1977, 62, 522−527. (25) Burns, R. G. Geochim. Cosmochim. Acta 1975, 39, 857−864. (26) Reinen, D. Struct. Bonding (Berlin) 1969, 6, 30−51. (27) Burns, R. G. Mineralogical Applications of Crystal Field Theory, 2nd ed.; Cambridge University Press: Cambridge, UK, 1993; pp 353− 395. (28) Brown, D. H. J. Chem. Soc. 1962, 25, 3322−3324. (29) Doumerc, J. P.; Vlasse, M.; Pouchard, M.; Hagenmuller, P. J. Solid State Chem. 1975, 14, 144−151. (30) Sinha, K. P.; Sinha, A. P. B. J. Phys. Chem. 1957, 61, 758−761. (31) Sviridov, D. T.; Sviridova, R. K. J. Appl. Spectrosc. 1981, 34, 431−433. (32) Ahmad, I.; Marinova, V.; Vrielinck, H.; Callens, F.; Goovaerts, E. J. Appl. Phys. 2011, 109, 083506. (33) Bottomley, F.; Chen, J.; MacIntosh, S. M.; Thompson, R. C. Organometallics 1991, 10, 906−912. (34) Allen, D. P.; Bottomley, F.; Day, R. W.; Decken, A.; Sanchez, V.; Summers, D. A.; Thompson, R. C. Organometallics 2001, 20, 1840− 1848. (35) Sunandana, C. S.; Phaninath, D. Solid State Commun. 1986, 58, 115−119. (36) Hartmann, M.; Kevan, L. Chem. Rev. 1999, 99, 635−664. (37) Weckhuysen, B. M.; Rao, R. R.; Martens, J. A.; Schoonheydt, R. A. Eur. J. Inorg. Chem. 1999, 1999, 565−577. (38) Jakšić, M.; Bogdanović, I.; Dujmić, D.; Fazinić, S.; Tadić, T. Strojarstvo 1996, 38, 249−254. (39) Rietveld, H. M. J. Appl. Crystallogr. 1969, 2, 65−71. (40) Toraya, H. J. Appl. Crystallogr. 1993, 26, 583−590. (41) Toraya, H. J. Appl. Crystallogr. 1986, 19, 440−447. (42) X′Pert HighScore Plus Program, Version 2.1; PANalytical Almelo: Netherlands, 2004. (43) Lützenkirchen-Hecht, D.; Wagner, R.; Szillat, S.; Hüsecken, A. K.; Istomin, K.; Pietsch, U.; Frahm, R. J. Synchrotron Radiat. 2014, 21, 819−826. (44) Matz, W.; Schell, N.; Bernhard, G.; Prokert, F.; Reich, T.; Claußner, J.; Oehme, W.; Schlenk, R.; Dienel, S.; Funke, H.; Eichhorn, F.; Betzl, M.; Pröhl, D.; Strauch, U.; Hüttig, G.; Krug, H.; Neumann, W.; Brendler, V.; Reichel, P.; Denecke, M. A.; Nitsche, H. J. Synchrotron Radiat. 1999, 6, 1076−1085. (45) Ressler, T. J. Synchrotron Radiat. 1998, 5, 118−122. (46) Sayers, D. E.; Stern, E. A.; Lytle, F. W. Phys. Rev. Lett. 1971, 27, 1204−1207. (47) Newville, M. J. Synchrotron Radiat. 2001, 8, 322−324. (48) Ravel, B.; Newville, M. J. Synchrotron Radiat. 2005, 12, 537− 541. (49) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys. Rev. B: Condens. Matter Mater. Phys. 1998, 58, 7565−7576. (50) Powder Diffraction File, PDF-2 Database; International Centre for Diffraction Data: Newtown Square, Pennsylvania, USA, 2003. (51) Lentz, A.; Büchele, W.; Schöllhorn, H. Cryst. Res. Technol. 1986, 21, 827−833. (52) Young, R. A.; Prince, E.; Sparks, R. A. J. Appl. Crystallogr. 1982, 15, 357−359. (53) Shannon, R. D. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976, 32, 751−767. (54) Goldschmidt, V. M. J. Chem. Soc. 1937, 655−673. (55) Tromp, M.; Moulin, J.; Reid, G.; Evans, J. AIP Conference Proc. 2006, 882, 699−701. (56) Wei, F.; Chen, Z. W.; Gibson, W. M. X-Ray Spectrom. 2009, 38, 382−385. (57) Koningsberger, D.; Prins, R. X-Ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS and XANES; Wiley: New York, 1988; pp 211−256. (58) Duesler, E. N.; Foord, E. E. Am. Mineral. 1986, 71, 1217−1220. I

DOI: 10.1021/acs.inorgchem.5b01379 Inorg. Chem. XXXX, XXX, XXX−XXX