Gadolinium ytterbium trifluoride, Gd_0.81Yb_0.19F₃

A new gadolinium ytterbium trifluoride has been grown for the first time by the Czochralski technique. Although GdF₃ and YbF₃ both present a high-temperature phase transition, the mixed compound Gd_0.81Yb_0.19F₃ maintains its crystallographic structure upon cooling to room temperature. Taking into account that both Gd³⁺ and Yb³⁺ ions are distributed randomly on a single site (Wyckoff position 4c), this is attributed to a mean cationic radius coincident with that of the Tb³⁺ ion, so that the stability of the crystal structure resembles that of TbF₃. The grown crystal melts noncongruently at ∼1413 K, it is transparent and colourless, and it has a high density.

S1. Comment

Rare earth trifluorides (REF₃) have long been studied as hosts for scintillation applications (Kobayashi et al., 2003). These crystals present a series of important characteristics, such as easy doping by isovalent Ce³⁺ substitution, high density, and high visible transparency. Among the near-UV–visible transparent REF₃ (RE = La, Ce, Gd, Yb1 and Lu), only the well known LaF₃ and CeF₃ do not present a phase transition upon cooling.

Initial studies of the synthesis of REF₃ identified a structural dimorphism along this series of compounds, which can crystallize in the tysonite hexagonal form and/or in the orthorhombic β-YF₃ form (Zalkin & Templeton, 1953; Thoma & Brunton, 1966). Later authors have concluded that REF₃ possess three different crystal structures. Depending on the occurrence or absence of a phase transition, REF₃ are classified into four morphotropic series as a function of decreasing lanthanide ionic radius (Fedorov & Sobolev, 1995; Spedding et al., 1974; Sobolev et al., 1976; Greis & Cader, 1985; Petzel & Rathjen, 1994):

(1) LaF₃ to NdF₃ (trigonal LaF₃-type, P3c1),

(2) SmF₃ to GdF₃ (hexagonal Schlyter type, P6₃/mmc; trigonal LaF₃-type, P3c1; and orthorhombic β-YF₃-type, Pnma),

(3) TbF₃ to HoF₃ (orthorhombic β-YF₃-type, Pnma) and

(4) ErF₃ to LuF₃ (trigonal α-UO₃-type, P3m1; and orthorhombic β-YF₃-type, Pnma).

[There is a controversy about the existence of the high-temperature hexagonal Schlyter-type phase. On the one hand, its appearance is claimed (Greis & Cader, 1985; Schlyter, 1953), while on the other it has been argued that it corresponds to twinned crystals in space group P3c1 with balanced volume ratios (Maximov & Schulz, 1985).]

The transitions between these series are a consequence of their relative structural instabilities. According to previous studies regarding the stabilization of each structural type [see Sobolev et al. (1977), and references therein], the mean cationic radius is the geometric factor that plays a decisive role. In that particular paper, Sobolev dealt with the stability of the orthorhombic β-YF₃-type in the phase diagrams of the GdF₃–LnF₃ systems (with Ln = Tb, Ho, Er or Yb). Taking into account the ionic radii (Shannon, 1976), one can calculate the stability range of this phase, which extends from peritectic to eutectic composition. In all four diagrams this range relates to a mean cationic radius which lies between the ionic radii of Tb and Ho, i.e. it matches the range (3) above where no phase transition is observed.

In the present work, we intended to synthesize a binary REF₃ which was transparent and did not present any phase transitions. For this purpose, we chose Gd³⁺ and Yb³⁺ as cations and considered the GdF₃–YbF₃ phase diagram (Sobolev et al., 1977). Due to the noncongruent nature of mixed Gd_1-xYb_xF₃ compounds, the initial composition of the melt was chosen to be x = 0.30, so that, according to the phase diagram, a crystal with the approximate composition Gd_0.91Yb_0.09F₃ should crystallize at a temperature of about 1443 K with an orthorhombic structure.

A transparent crack-free single crystal was grown by the Czochralski technique. Chemical analysis carried out by inductively coupled plasma indicated that the crystal composition was Gd_0.81Yb_0.19F₃. According to the phase diagram, this concentration of Yb1 should correspond to a starting melt value of x = 0.43, instead of the nominally used x = 0.3. X-ray diffraction measurements confirmed that the grown crystal has the expected β-YF₃ structure, characterized by two anionic sites and a single cationic site. Consequently, both Gd³⁺ and Yb³⁺ occupy the same site randomly (Wyckoff position 4c), as shown in Fig. 1. Furthermore, as the two cations are very similar to each other, it is not possible to distinguish them by X-ray diffraction measurements. The lattice parameters obtained correspond precisely with those known for TbF₃. Considering Vegard's law (Reference?) and the ionic radii for coordination number 9 for the cationic site, we found that the average radius coincides with that of Tb₉³⁺, namely 1.235 Å. Therefore, in agreement with previous observations about average ionic radius for the stabilization of the orthorhombic structure, our experiment reinforces the hypothesis that the Tb³⁺ ionic radius represents the maximum average limit.

The thermal behaviour of the as-grown crystal was investigated by differential scanning calorimetry and is shown in Fig. 2. Similar to the GdF₃–YbF₃ phase diagram, three transitions at high temperature can be clearly distinguished. The first one, at 1413 K, corresponds to the nucleation temperature, the second one, at 1527 K, with the peritectic temperature and the highest one, at 1540 K, with the completely liquid phase. These values are shifted from those indicated in the phase diagram for the measured crystal composition (x = 0.19), namely 1398, 1453 and 1466 K for the nucleation temperature, eutectic phase and total melting, respectively.

In summary, we have presented the first report of the growth of a single crystal of Gd_0.81Yb_0.19F₃ by the Czochralski technique. Due to its noncongruent nature, the growth of large crystals will require non-standard techniques to counteract the continuous shift of the melt composition during growth. On the other hand, this growth behaviour could be an advantage for the synthesis of core–shell nanoparticles based on REF₃ luminescent ions, which have specific applications in the fields of biology and optoelectronics (see e.g. Wang et al., 2006). Crystalline Gd_0.81Yb_0.19F₃ is transparent and colourless, it crystallizes in the orthorhombic β-YF₃ structure and it does not present any phase transition upon cooling. Furthermore, the density is as high as 7.25 Mg m^-3. Therefore, Gd_0.81Yb_0.19F₃ represents a new REF₃ with attractive properties for scintillation and core–shell nanoparticle applications.

Related literature

For related literature, see: Fedorov & Sobolev (1995); Greis & Cader (1985); Kobayashi et al. (2003); Maximov & Schulz (1985); Petzel & Rathjen (1994); Schlyter (1953); Shannon (1976); Sobolev et al. (1976, 1977); Spedding et al. (1974); Thoma & Brunton (1966); Wang et al. (2006); Zalkin & Templeton (1953).

S2. Experimental

A single crystal of the title compound was grown by the Czochralski (Cz) technique under a high-purity CF₄ (99.999%) atmosphere. Commercial GdF₃ (99.99%) and YbF₃ (99.99%) powders were mixed with a nominal composition of Gd_0.7Yb_0.3F₃, and loaded into a Pt crucible. This was surrounded by refractory carbons and heated inductively using a 30 kW generator. Using a Pt wire as a seed, a single crystal was pulled up at a speed of 2 mm h^-1 and a rotation of 10 r min^-1.

The chemical composition of the grown crystal was determined by inductively coupled plasma (ICP) atomic emission spectroscopy using an IRIS Advantage from Nippon Jarrell–AshCo. Thermal analysis of the crystal was performed by differential scanning calorimetry (DSC) using a DSC Rigaku Thermoplus 8270. The heating rate was set to 10 K min^-1 and the measurements were carried out under a 200 ml min^-1 flow of CF₄ in order to prevent sample oxidation.

Computing details

Data collection: SMART (Bruker, 2002); cell refinement: SAINT-Plus (Bruker, 2003); data reduction: SAINT-Plus (Bruker, 2003); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: DIAMOND (Brandenburg, 2000); software used to prepare material for publication: SHELXL97 (Sheldrick, 2008).

Figures

	Fig. 1. A view of the unit cell of Gd0.81Yb0.19F3, in a polyhedral representation.
	Fig. 2. Differential scanning calorimetry curve of Gd0.81Yb0.19F3 under a CF4 atmosphere.

Gadolinium ytterbium trifluoride

Crystal data

Gd0.81Yb0.19F3	Dx = 7.258 Mg m^−3
Mr = 217.24	Mo Kα radiation, λ = 0.71073 Å
Orthorhombic, Pnma	Cell parameters from 866 reflections
a = 6.518 (1) Å	θ = 4.6–40.3°
b = 6.950 (1) Å	µ = 35.83 mm^−1
c = 4.389 (1) Å	T = 293 K
V = 198.82 (6) Å^3	Plate, colourless
Z = 4	0.16 × 0.16 × 0.1 mm
F(000) = 368.560

Data collection

Bruker SMART APEX CCD area-detector diffractometer	656 independent reflections
Radiation source: fine-focus sealed tube	646 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.035
ω scans	θmax = 40.3°, θmin = 5.5°
Absorption correction: empirical (using intensity measurements) (SADABS; Sheldrick, 1996)	h = −11→11
Tmin = 0.023, Tmax = 0.086	k = −11→12
4098 measured reflections	l = −7→7

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.020	w = 1/[σ^2(Fo^2) + (0.0235P)^2 + 0.5681P] where P = (Fo^2 + 2Fc^2)/3
wR(F^2) = 0.052	(Δ/σ)max < 0.001
S = 1.23	Δρmax = 2.34 e Å^−3
656 reflections	Δρmin = −1.69 e Å^−3
23 parameters	Extinction correction: SHELXL97 (Sheldrick, 2008), Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
0 restraints	Extinction coefficient: 0.058 (2)

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
F1	0.1649 (3)	0.0642 (3)	0.1159 (5)	0.0131 (3)
F2	0.0208 (4)	0.2500	0.5841 (6)	0.0142 (4)
Gd1	0.36752 (2)	0.2500	0.43768 (4)	0.00909 (8)	0.81
Yb1	0.36752 (2)	0.2500	0.43768 (4)	0.00909 (8)	0.19

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
F1	0.0141 (6)	0.0108 (7)	0.0145 (7)	0.0008 (6)	−0.0031 (6)	−0.0020 (6)
F2	0.0135 (10)	0.0151 (12)	0.0140 (11)	0.000	0.0016 (7)	0.000
Gd1	0.00965 (11)	0.00723 (11)	0.01039 (11)	0.000	−0.00085 (4)	0.000
Yb1	0.00965 (11)	0.00723 (11)	0.01039 (11)	0.000	−0.00085 (4)	0.000

Geometric parameters (Å, º)

F1—Gd1	2.325 (2)	Gd1—F2^iv	2.325 (3)
F1—Gd1^i	2.329 (2)	Gd1—F1^v	2.325 (2)
F1—Gd1^ii	2.341 (2)	Gd1—F1^vi	2.329 (2)
F2—Gd1^iii	2.325 (3)	Gd1—F1^vii	2.341 (2)
F2—Gd1	2.349 (3)	Gd1—F2^viii	2.498 (3)
F2—Gd1^ii	2.498 (3)
Gd1—F1—Yb1^i	140.93 (9)	F1^ix—Gd1—F2	79.68 (5)
Gd1—F1—Gd1^i	140.93 (9)	F1^viii—Gd1—F2	145.48 (5)
Gd1—F1—Yb1^ii	103.00 (8)	F1^vii—Gd1—F2	145.48 (5)
Yb1^i—F1—Yb1^ii	114.09 (8)	F2^iv—Gd1—F2^viii	130.97 (12)
Gd1^i—F1—Yb1^ii	114.09 (8)	F1^v—Gd1—F2^viii	70.76 (7)
Gd1—F1—Gd1^ii	103.00 (8)	F1—Gd1—F2^viii	70.76 (7)
Yb1^i—F1—Gd1^ii	114.09 (8)	F1^vi—Gd1—F2^viii	110.13 (5)
Gd1^i—F1—Gd1^ii	114.09 (8)	F1^ix—Gd1—F2^viii	110.13 (5)
Yb1^iii—F2—Gd1	131.33 (12)	F1^viii—Gd1—F2^viii	64.96 (7)
Gd1^iii—F2—Gd1	131.33 (12)	F1^vii—Gd1—F2^viii	64.96 (7)
Yb1^iii—F2—Gd1^ii	130.97 (12)	F2—Gd1—F2^viii	129.45 (9)
Gd1^iii—F2—Gd1^ii	130.97 (12)	F2^iv—Gd1—Yb1^ii	142.27 (7)
Gd1—F2—Gd1^ii	97.71 (9)	F1^v—Gd1—Yb1^ii	38.66 (5)
Yb1^iii—F2—Yb1^ii	130.97 (12)	F1—Gd1—Yb1^ii	38.66 (5)
Gd1^iii—F2—Yb1^ii	130.97 (12)	F1^vi—Gd1—Yb1^ii	94.04 (5)
Gd1—F2—Yb1^ii	97.71 (9)	F1^ix—Gd1—Yb1^ii	94.04 (5)
F2^iv—Gd1—F1^v	142.42 (6)	F1^viii—Gd1—Yb1^ii	133.92 (5)
F2^iv—Gd1—F1	142.42 (6)	F1^vii—Gd1—Yb1^ii	133.93 (5)
F1^v—Gd1—F1	67.48 (10)	F2—Gd1—Yb1^ii	42.68 (6)
F2^iv—Gd1—F1^vi	74.68 (6)	F2^viii—Gd1—Yb1^ii	86.76 (6)
F1^v—Gd1—F1^vi	132.31 (6)	F2^iv—Gd1—Gd1^ii	142.27 (7)
F1—Gd1—F1^vi	68.38 (5)	F1^v—Gd1—Gd1^ii	38.66 (5)
F2^iv—Gd1—F1^ix	74.68 (6)	F1—Gd1—Gd1^ii	38.66 (5)
F1^v—Gd1—F1^ix	68.38 (5)	F1^vi—Gd1—Gd1^ii	94.04 (5)
F1—Gd1—F1^ix	132.31 (6)	F1^ix—Gd1—Gd1^ii	94.04 (5)
F1^vi—Gd1—F1^ix	139.28 (10)	F1^viii—Gd1—Gd1^ii	133.92 (5)
F2^iv—Gd1—F1^viii	74.62 (8)	F1^vii—Gd1—Gd1^ii	133.93 (5)
F1^v—Gd1—F1^viii	135.70 (4)	F2—Gd1—Gd1^ii	42.68 (6)
F1—Gd1—F1^viii	95.91 (7)	F2^viii—Gd1—Gd1^ii	86.76 (6)
F1^vi—Gd1—F1^viii	65.91 (8)	F2^iv—Gd1—Yb1^viii	91.36 (7)
F1^ix—Gd1—F1^viii	128.76 (5)	F1^v—Gd1—Yb1^viii	103.47 (5)
F2^iv—Gd1—F1^vii	74.62 (8)	F1—Gd1—Yb1^viii	103.47 (5)
F1^v—Gd1—F1^vii	95.91 (7)	F1^vi—Gd1—Yb1^viii	103.45 (5)
F1—Gd1—F1^vii	135.70 (4)	F1^ix—Gd1—Yb1^viii	103.45 (5)
F1^vi—Gd1—F1^vii	128.76 (5)	F1^viii—Gd1—Yb1^viii	38.34 (5)
F1^ix—Gd1—F1^vii	65.91 (8)	F1^vii—Gd1—Yb1^viii	38.34 (5)
F1^viii—Gd1—F1^vii	66.95 (10)	F2—Gd1—Yb1^viii	169.05 (7)
F2^iv—Gd1—F2	99.59 (5)	F2^viii—Gd1—Yb1^viii	39.61 (6)
F1^v—Gd1—F2	67.66 (7)	Yb1^ii—Gd1—Yb1^viii	126.372 (10)
F1—Gd1—F2	67.65 (7)	Gd1^ii—Gd1—Yb1^viii	126.372 (10)
F1^vi—Gd1—F2	79.68 (5)

Symmetry codes: (i) −x+1/2, −y, z−1/2; (ii) x−1/2, y, −z+1/2; (iii) x−1/2, y, −z+3/2; (iv) x+1/2, y, −z+3/2; (v) x, −y+1/2, z; (vi) −x+1/2, −y, z+1/2; (vii) x+1/2, −y+1/2, −z+1/2; (viii) x+1/2, y, −z+1/2; (ix) −x+1/2, y+1/2, z+1/2.