Rietveld refinement of the crystal structures of Rb₂XSi₅O₁₂ (X = Ni, Mn)

The synthetic leucite silicate framework mineral analogues Rb₂XSi₅O₁₂ {X = Ni [dirubidium nickel(II) penta­silicate] and Mn [dirubidium manganese(II) penta­silicate]} have been prepared by high-temperature solid-state synthesis. The results of Rietveld refinements, using X-ray powder diffraction data collected using Cu Kα X-rays, show that the title compounds crystallize in the space group Pbca and adopt the cation-ordered structure of Cs₂CdSi₅O₁₂ and other leucites. The structures consist of tetra­hedral SiO₄ and XO₄ units sharing corners to form a partially substituted silicate framework. Extraframework Rb⁺ cations sit in channels in the framework. All atoms occupy the 8c general position for this space group. In these refined structures, silicon and X atoms are ordered onto separate tetra­hedrally coordinated sites (T-sites). However, the Ni displacement parameter and the Ni—O bond lengths suggest that for the X = Ni sample, there may actually be some T-site cation disorder.

1. Chemical context

Synthetic analogues of the silicate framework minerals leucite KAlSi₂O₆ (Mazzi et al., 1976) and pollucite CsAlSi₂O₆ (Dimitrijevic et al., 1991) can be prepared with the general formulae ABSi₂O₆ and A₂CSi₅O₁₂. A is an alkali metal cation (K, Rb, Cs), B is a trivalent cation (Al, B, Fe³⁺) and C is a divalent cation (Be, Mg, Mn, Fe²⁺, Co, Ni, Cu, Zn, Cd). The title compounds are leucite analogues with A = Rb and C = Ni and Mn, these structures are in the space group Pbca and are isostructural with Cs₂CdSi₅O₁₂ (Bell et al., 1994b).

These leucite structures all have the same topology, a silicate framework structure with B or C cations partially substituting on the tetra­hedrally coordinated silicon sites (T-sites). A cations sit in the extraframework channels, these extraframework cations can be removed by ion exchange which makes them of technological inter­est as a possible storage medium for radioactive Cs from nuclear waste (Gatta et al., 2008).

2. Structural commentary

For the X = Ni refinement, the Ni site isotropic temperature factor was larger than expected [B_iso = 7.5 (9) Ų]. Also the mean Ni—O bond length for the NiO₄ tetra­hedron is 1.90 (2) Å, shorter than that seen in tetra­hedrally coordinated NiO₄ units. NiCr₂O₄ has the cubic spinel structure with Ni in tetra­hedral coordination. A single-crystal structure refinement (Crottaz et al., 1997) gives the Ni—O distance as 1.967 (3) Å. An EXAFS/XANES study (Farges et al., 2001) gives the Ni—O distance as 1.96 (1) Å. The mean Si—O bond length for the SiO₄ tetra­hedra in the title structure is 1.643 (18) Å, which is slightly larger than the range of Si—O distances for silicates [1.59–1.63 Å; Inter­national Tables for X-ray Crystallography (1975, Vol. III, Table 4.1.1)]. These differences in the Ni—O and Si—O distances suggest some possible T-site cation disorder with some Si on the Ni site. A future higher quality neutron/synchrotron X-ray powder diffraction study may show if there really is Ni/Si T-site cation disorder.

Fig. 1 shows the Rietveld difference plot for Rb₂NiSi₅O₁₂. The crystal structure of Rb₂NiSi₅O₁₂ is displayed in Fig. 2 and consists of a framework of corner-sharing tetra­hedral SiO₄ and NiO₄ units, and Rb⁺ cations sitting in the extraframework channels.

Figure 1 Rietveld difference plot for the single-phase refinement of Rb2NiSi5O12. The red, blue and green lines show, respectively, the observed, calculated and difference plots. Calculated Bragg reflection positions are indicated by crosses.

Figure 2 The crystal structure of Rb2NiSi5O12. Green spheres show Rb cations, blue polyhedra show SiO4 units, pink polyhedra show NiO4 units and red spheres represent O atoms.

For the X = Mn refinement, unlike in the X = Ni sample, the mean Mn—O distance is 2.02 (1) Å. This is in agreement with the mean Mn—O distance for Cs₂MnSi₅O₁₂ (1.98 (3) Å; Bell & Henderson, 1996). This would suggest that this structure has complete T-site cation ordering. However, the isotropic temperature factors for the Mn site and the O sites could not be refined to chemically sensible positive values, so these were both fixed at B_iso = 0.1 Ų. This refinement was done assuming that Mn was present as Mn²⁺ and that all sites were fully occupied. If some or all of the Mn atoms were present with a higher oxidation state, then this could account for the problem with refining these temperature factors. However, a higher quality neutron/synchrotron X-ray powder diffraction study may also be needed for a more precise determination of the state of Mn in this structure.

Fig. 3 shows the Rietveld difference plot for Rb₂MnSi₅O₁₂. The crystal structure of Rb₂MnSi₅O₁₂ is displayed in Fig. 4 and consists of a framework of corner-sharing tetra­hedral SiO₄ and MnO₄ units; Rb cations sit in the extraframework channels. Note how inclusion of the larger Mn cation in the silicate framework compared to Ni causes the central channel of the crystal structure to be slightly more distorted for Rb₂MnSi₅O₁₂ (Fig. 4) compared to Rb₂NiSi₅O₁₂ (Fig. 2).

Figure 3 Rietveld difference plot for the single-phase refinement of Rb2MnSi5O12. The red, blue and green lines show, respectively, the observed, calculated and difference plots. Calculated Bragg reflection positions are indicated by crosses.

Figure 4 The crystal structure of Rb2MnSi5O12. Green spheres show Rb cations, blue polyhedra show SiO4 units, purple polyhedra show MnO4 units and red spheres represent O atoms.

3. Database survey

Many different leucite analogue crystal structures are known at ambient temperature. Table 1 gives compositions, space groups, lattice parameters, and references for some known ambient temperature leucite crystal structures. In addition, a high-temperature structure for Cs₂ZnSi₅O₁₂ in the space group Pa has been reported above 566 K (Bell & Henderson, 2012).

Table 1
Crystal structure parameters (Å°) for room-temperature leucite analogues

Stoichiometry	SG	a	b	c	β	V
K2MgSi5O12^a	Iad	13.4190 (1)	13.4190 (1)	13.4190 (1)	90	2416.33 (5)
K2MgSi5O12^a	P21/c	13.168 (5)	13.652 (1)	13.072 (5)	91.69 (5)	2348 (2)
Cs2CdSi5O12^b	Pbca	13.6714 (1)	13.8240 (1)	13.8939 (1)	90	2625.83 (6)
Cs2CuSi5O12^c	Pbca	13.58943 (6)	13.57355 (5)	13.62296 (4)	90	2512.847 (13)
Cs2CuSi5O12^c	Iad	13.6322 (4)	13.6322 (4)	13.6322 (4)	90	2533.4 (2)
Cs2MgSi5O12^d	Pbca	13.6371 (5)	13.6689 (5)	13.7280 (5)	90	2559.0 (2)
Rb2MgSi5O12^d	Pbca	13.422 (1)	13.406 (1)	13.730 (1)	90	2470.6 (4)
Cs2ZnSi5O12^d	Pbca	13.6415 (9)	13.6233 (8)	13.6653 (9)	90	2539.6 (3)
Rb2CdSi5O12^e	Pbca	13.4121 (1)	13.6816 (1)	13.8558 (1)	90	2542.51 (5)
Cs2MnSi5O12^e	Pbca	13.6878 (3)	13.7931 (3)	13.7575 (3)	90	2597.4 (2)
Cs2CoSi5O12^e	Pbca	13.6487 (4)	13.7120 (4)	13.6828 (4)	90	2560.7 (2)
Cs2NiSi5O12^e	Pbca	13.6147 (3)	13.6568 (5)	13.6583 (5)	90	2539.5 (1)
Rb2ZnSi5O12^f	Iad	13.4972 (1)	13.4972 (1)	13.4972 (1)	90	2458.86 (3)
KFeSi2O6^g	I41/a	13.2207 (3)	13.2207 (3)	13.9464 (3)	90	2437.6 (2)
RbFeSi2O6^g	I41/a	13.4586 (1)	13.4586 (1)	13.9380 (1)	90	2524.63 (5)
CsFeSi2O6^g	Iad	13.8542 (1)	13.8542 (1)	13.8542 (1)	90	2653.98 (3)
CsBSi2O6^h	I41/a	13.019 (2)	13.019 (2)	12.899 (3)	90	2186 (1)
CsAlSi2O6^i	Iad	13.647 (3)	13.647 (3)	13.647 (3)	90	2541.6 (6)
KAlSi2O6^j	I41/a	13.09 (1)	13.09 (1)	13.75 (1)	90	2356 (5)
Cs0.814B1.092Si1.977O6^k	Ia-3d	13.009 (8)	13.009 (8)	13.009 (8)	90	2202 (1)
Rb2MgSi5O12^l	Iad	13.530 (1)	13.530 (1)	13.530 (1)	90	2476.8 (2)
Cs2BeSi5O12^m	Iad	13.406 (1)	13.406 (1)	13.406 (1)	90	2409.3 (2)
Notes: SG = space group; all α and γ angles = 90°; (a) Bell et al. (1994a); (b) Bell et al. (1994b); (c) Bell et al. (2010); (d) Bell & Henderson (2009); (e) Bell & Henderson (1996); (f) Bell & Henderson (1994a); (g) Bell & Henderson (1994b); (h) Agakhanov et al. (2012); (i) Dimitrijevic et al. (1991); (j) Mazzi et al. (1976); (k) Bubnova et al. (2004); (l) Torres-Martinez & West (1986); (m) Torres-Martinez et al. (1984).

4. Synthesis and crystallization

The samples were made from stoichiometric mixtures of Rb₂CO₃, SiO₂ and NiO (X = Ni) or MnO (X = Mn). These mixtures were ground together and then heated overnight at 873 K to decompose the carbonates, then melted in platinum crucibles at 1573 K for 1.5 h (X = Ni) or 1673 K for 2 h (X = Mn) before quenching to form glasses. The glasses were dry-crystallized at ambient pressure and 1193 K for 12 d.

5. Refinement

Crystal data, data collection and structure refinement details are summarized in Table 2. For each sample, a small amount of powder was ground and mounted on a low-background silicon wafer with a drop of acetone. These were mounted in flat-plate mode on a PANalytical X'Pert Pro MPD diffractometer. X-ray powder diffraction data were collected at 293 K using CuKα X-rays over the range 10–80°/2θ using a PANalytical X'Celerator area detector. The powder diffraction data collection time for each sample was 8 h 20 min.

Table 2
Experimental details

	Rb2NiSi5O12	Rb2MnSi5O12
Crystal data
Chemical formula	Rb2NiSi5O12	Rb2MnSi5O12
Mr	562.06	557.92
Crystal system, space group	Orthorhombic, Pbca	Orthorhombic, Pbca
Temperature (K)	293	293
a, b, c (Å)	13.469 (3), 13.480 (3), 13.442 (2)	13.4085 (10), 13.6979 (11), 13.5761 (10)
V (Å^3)	2440.7 (8)	2493.5 (3)
Z	8	8
Radiation type	Cu Kα, λ = 1.540560 Å	Cu Kα, λ = 1.540560 Å
Specimen shape, size (mm)	Irregular, 10 × 10	Irregular, 10 × 10
Data collection
Diffractometer	PANalytical X'Pert Pro MPD	PANalytical X'Pert Pro MPD
Specimen mounting	Flat plate	Flat plate
Data collection mode	Reflection	Reflection
Scan method	Step	Step
2θ values (°)	2θmin = 9.897 2θmax = 79.883 2θstep = 0.017	2θmin = 10.139 2θmax = 80.125 2θstep = 0.017
Refinement
R factors and goodness of fit	Rp = 9.048, Rwp = 12.007, Rexp = 4.263, RBragg = 10.421, χ^2 = 7.935	Rp = 6.527, Rwp = 8.847, Rexp = 3.909, RBragg = 5.989, χ^2 = 5.121
No. of parameters	73	71
No. of restraints	24	24
Computer programs: X'Pert Data Collector (PANalytical, 2006), FULLPROF (Rodríguez-Carvajal, 1993), VESTA (Momma & Izumi, 2008) and publCIF (Westrip, 2010).

All Bragg reflections in both of the powder diffraction patterns could be indexed in the space group Pbca with similar lattice parameters to that for the Cs₂CdSi₅O₁₂ leucite (Bell et al., 1994a). The crystal structures (Bell & Henderson, 1996) of Cs₂NiSi₅O₁₂ (X = Ni) and Cs₂MnSi₅O₁₂ (X = Mn) were respectively used as starting models for Rietveld (1969) refinements. In both cases, Rb⁺ replaced Cs⁺ as the extraframework cation. Isotropic atomic displacement parameters were used for all atoms in these phases. In both refinements, the isotropic atomic displacement parameters were constrained to be the same for all sites occupied by the same element, each Rb site had the same displacement parameter as did each Si site and each O site. Soft constraints were used for both refinements, in both cases the Si—O distances were constrained to be 1.61±0.02 Å. For X = Ni, the Ni—O distances were constrained to be 1.88±0.02 Å, the mean Ni—O distance for Cs₂NiSi₅O₁₂. For X = Mn, the Mn—O distances were constrained to be 1.98±0.02 Å, the mean Mn—O distance for Cs₂MnSi₅O₁₂. In both cases, Si and X atoms were ordered onto separate tetra­hedrally coordinated sites, both refined structures are similar.

Supporting information

S1. Chemical context

Synthetic analogues of the silicate framework minerals leucite KAlSi₂O₆ (Mazzi et al., 1976) and pollucite CsAlSi₂O₆ (Dimitrijevic et al., 1991) can be prepared with the general formulae AB₂O₆ and A₂CSi₅O₁₂. A is an alkali metal cation (K, Rb, Cs), B is a trivalent cation (Al, B, Fe³⁺) and C is a divalent cation (Be, Mg, Mn, Fe²⁺, Co, Ni, Cu, Zn, Cd). The title compounds are leucite analogues with A = Rb and C = Ni and Mn, these structures are in the space group Pbca and are isostructural with Cs₂CdSi₅O₁₂ (Bell et al., 1994b).

These leucite structures all have the same topology, a silicate framework structure with B or C cations partially substituting on the tetra­hedrally coordinated silicon sites (T-sites). A cations sit in the extraframework channels, these extraframework cations can be removed by ion exchange which makes them of technological inter­est as a possible storage medium for radioactive Cs from nuclear waste (Gatta et al., 2008.)

S2. Structural commentary

For the X = Ni refinement, the Ni site isotropic temperature factor was larger than expected [B_iso = 7.5 (9) Ų]. Also the mean Ni–O bond length for the NiO₄ tetra­hedron is 1.90 (2) Å, shorter than that seen in tetra­hedrally coordinated NiO₄ units. NiCr₂O₄ has the cubic spinel structure with Ni in tetra­hedral coordination. A single-crystal structure refinement (Crottaz et al., 1997) gives the Ni—O distance as 1.967 (3) Å. An EXAFS/XANES study (Farges et al., 2001) gives the Ni—O distance as 1.96 (1) Å. The mean Si—O bond length for the SiO₄ tetra­hedra in this structure is 1.643 (7)Å, which is slightly larger than the range of Si—O distances for silicates [1.59–1.63 Å; Inter­national Tables for X-ray Crystallography (1975, Vol. III, Table 4.1.1)]. These differences in the Ni—O and Si—O distances suggest some possible T-site cation disorder with some Si on the Ni site. A future higher quality neutron/synchrotron X-ray powder diffraction study may show if there really is Ni/Si T-site cation disorder.

Fig. 1 shows the Rietveld difference plot for Rb₂NiSi₅O₁₂. The crystal structure of Rb₂NiSi₅O₁₂ is displayed in Fig. 2 and consists of a framework of corner-sharing tetra­hedral SiO₄ and NiO₄ units, and Rb cations sitting in the extraframework channels.

For the X = Mn refinement, unlike in the X = Ni sample, the mean Mn—O distance is 2.02 (1) Å. This is in agreement with the mean Mn—O distance for Cs₂MnSi₅O₁₂ (1.98 (3) Å; Bell & Henderson, 1996). This would suggest that this structure has complete T-site cation ordering. However, the isotropic temperature factors for the Mn site and the O sites could not be refined to chemically sensible positive values, so these were both fixed at B_iso = 0.1 Ų. This refinement was done assuming that Mn was present as Mn²⁺ and that all sites were fully occupied. If some or all of the Mn atoms were present with a higher oxidation state, then this could account for the problem with refining these temperature factors. However, a higher quality neutron/synchrotron X-ray powder diffraction study may also be needed for a more precise determination of the state of Mn in this structure.

Fig. 3 shows the Rietveld difference plot for Rb₂MnSi₅O₁₂ . The crystal structure of Rb₂MnSi₅O₁₂ is displayed in Fig. 4 and consists of a framework of corner-sharing tetra­hedral SiO₄ and MnO₄ units; Rb cations sit in the extraframework channels. Note how inclusion of the larger Mn cation in the silicate framework compared to Ni causes the central channel of the crystal structure to be slightly more distorted for Rb₂MnSi₅O₁₂ (Fig. 4) compared to Rb₂NiSi₅O₁₂ (Fig. 2).

S3. Database survey

Many different leucite analogue crystal structures are known at ambient temperature. Table 1 gives stoichiometries, space groups, lattice parameters, and references for some known ambient temperature leucite crystal structures. In addition, a high-temperature structure for Cs₂ZnSi₅O₁₂ in the space group Pa3 has been reported above 566 K (Bell & Henderson, 2012).

S4. Synthesis and crystallization

The samples were made from stoichiometric mixtures of Rb₂CO₃, SiO₂ and NiO (X = Ni) or MnO (X = Mn). These mixtures were ground together and then heated overnight at 873 K to decompose the carbonates, then melted in platinum crucibles at 1573 K for 1.5 h (X = Ni) or 1673 K for 2 h (X = Mn) before quenching to form glasses. The glasses were dry-crystallized at ambient pressure and 1193 K for 12 d.

S5. Refinement

Crystal data, data collection and structure refinement details are summarized in Table 2. For each sample, a small amount of powder was ground and mounted a low-background silicon wafer with a drop of acetone. These were mounted in flat-plate mode on a PANalytical X'Pert Pro MPD diffractometer. X-ray powder diffraction data were collected at 293 K using CuKα X-rays over the range 10–80°2θ using a PANalytical X'Celerator area detector. The powder diffraction data collection times for each sample were 8 h 20 min.

All Bragg reflections in both of the powder diffraction patterns could be indexed in the space group Pbca with similar lattice parameters to that for the Cs₂CdSi₅O₁₂ leucite (Bell et al., 1994a). The crystal structures (Bell & Henderson, 1996) of Cs₂NiSi₅O₁₂ (X = Ni) and Cs₂MnSi₅O₁₂ (X = Mn) were respectively used as starting models for Rietveld (1969) refinements. In both cases, Rb replaced Cs as the extraframework cation. Isotropic atomic displacement parameters were used for all atoms in these phases. In both refinements, the isotropic atomic displacement parameters were constrained to be the same for all sites occupied by the same element, each Rb site had the same displacement parameter as did each Si site and each O site. Soft constraints were used for both refinements, in both cases the Si—O distances were constrained to be 1.61±0.02 Å. For X = Ni, the Ni–O distances were constrained to be 1.88±0.02 Å, the mean Ni—O distance for Cs₂NiSi₅O₁₂. For X = Mn, the Mn—O distances were constrained to be 1.98±0.02 Å, the mean Mn—O distance for Cs₂MnSi₅O₁₂. In both cases, Si and X atoms were ordered onto separate tetra­hedrally coordinated sites, both refined structures are similar.

Related literature

For the crystal chemistry of leucite analogues, see: Agakhanov et al. (2012); Dimitrijevic et al. (1991); Bell & Henderson (1994a, 1994b, 1996, 2009, 2012); Bell et al. (1994a, 1994b, 2010); Bubnova et al. (2004); Mazzi et al. (1976); Torres-Martinez & West (1986); Torres-Martinez et al. (1984);

Atomic coordinates as starting parameters for the Rietveld refinement (Rietveld, 1969) of the present phases were taken from Bell & Henderson (2009).

Structure description

Synthetic analogues of the silicate framework minerals leucite KAlSi₂O₆ (Mazzi et al., 1976) and pollucite CsAlSi₂O₆ (Dimitrijevic et al., 1991) can be prepared with the general formulae AB₂O₆ and A₂CSi₅O₁₂. A is an alkali metal cation (K, Rb, Cs), B is a trivalent cation (Al, B, Fe³⁺) and C is a divalent cation (Be, Mg, Mn, Fe²⁺, Co, Ni, Cu, Zn, Cd). The title compounds are leucite analogues with A = Rb and C = Ni and Mn, these structures are in the space group Pbca and are isostructural with Cs₂CdSi₅O₁₂ (Bell et al., 1994b).

These leucite structures all have the same topology, a silicate framework structure with B or C cations partially substituting on the tetra­hedrally coordinated silicon sites (T-sites). A cations sit in the extraframework channels, these extraframework cations can be removed by ion exchange which makes them of technological inter­est as a possible storage medium for radioactive Cs from nuclear waste (Gatta et al., 2008.)

For the X = Ni refinement, the Ni site isotropic temperature factor was larger than expected [B_iso = 7.5 (9) Ų]. Also the mean Ni–O bond length for the NiO₄ tetra­hedron is 1.90 (2) Å, shorter than that seen in tetra­hedrally coordinated NiO₄ units. NiCr₂O₄ has the cubic spinel structure with Ni in tetra­hedral coordination. A single-crystal structure refinement (Crottaz et al., 1997) gives the Ni—O distance as 1.967 (3) Å. An EXAFS/XANES study (Farges et al., 2001) gives the Ni—O distance as 1.96 (1) Å. The mean Si—O bond length for the SiO₄ tetra­hedra in this structure is 1.643 (7)Å, which is slightly larger than the range of Si—O distances for silicates [1.59–1.63 Å; Inter­national Tables for X-ray Crystallography (1975, Vol. III, Table 4.1.1)]. These differences in the Ni—O and Si—O distances suggest some possible T-site cation disorder with some Si on the Ni site. A future higher quality neutron/synchrotron X-ray powder diffraction study may show if there really is Ni/Si T-site cation disorder.

Fig. 1 shows the Rietveld difference plot for Rb₂NiSi₅O₁₂. The crystal structure of Rb₂NiSi₅O₁₂ is displayed in Fig. 2 and consists of a framework of corner-sharing tetra­hedral SiO₄ and NiO₄ units, and Rb cations sitting in the extraframework channels.

For the X = Mn refinement, unlike in the X = Ni sample, the mean Mn—O distance is 2.02 (1) Å. This is in agreement with the mean Mn—O distance for Cs₂MnSi₅O₁₂ (1.98 (3) Å; Bell & Henderson, 1996). This would suggest that this structure has complete T-site cation ordering. However, the isotropic temperature factors for the Mn site and the O sites could not be refined to chemically sensible positive values, so these were both fixed at B_iso = 0.1 Ų. This refinement was done assuming that Mn was present as Mn²⁺ and that all sites were fully occupied. If some or all of the Mn atoms were present with a higher oxidation state, then this could account for the problem with refining these temperature factors. However, a higher quality neutron/synchrotron X-ray powder diffraction study may also be needed for a more precise determination of the state of Mn in this structure.

Fig. 3 shows the Rietveld difference plot for Rb₂MnSi₅O₁₂ . The crystal structure of Rb₂MnSi₅O₁₂ is displayed in Fig. 4 and consists of a framework of corner-sharing tetra­hedral SiO₄ and MnO₄ units; Rb cations sit in the extraframework channels. Note how inclusion of the larger Mn cation in the silicate framework compared to Ni causes the central channel of the crystal structure to be slightly more distorted for Rb₂MnSi₅O₁₂ (Fig. 4) compared to Rb₂NiSi₅O₁₂ (Fig. 2).

Many different leucite analogue crystal structures are known at ambient temperature. Table 1 gives stoichiometries, space groups, lattice parameters, and references for some known ambient temperature leucite crystal structures. In addition, a high-temperature structure for Cs₂ZnSi₅O₁₂ in the space group Pa3 has been reported above 566 K (Bell & Henderson, 2012).

For the crystal chemistry of leucite analogues, see: Agakhanov et al. (2012); Dimitrijevic et al. (1991); Bell & Henderson (1994a, 1994b, 1996, 2009, 2012); Bell et al. (1994a, 1994b, 2010); Bubnova et al. (2004); Mazzi et al. (1976); Torres-Martinez & West (1986); Torres-Martinez et al. (1984);

Atomic coordinates as starting parameters for the Rietveld refinement (Rietveld, 1969) of the present phases were taken from Bell & Henderson (2009).

Synthesis and crystallization

The samples were made from stoichiometric mixtures of Rb₂CO₃, SiO₂ and NiO (X = Ni) or MnO (X = Mn). These mixtures were ground together and then heated overnight at 873 K to decompose the carbonates, then melted in platinum crucibles at 1573 K for 1.5 h (X = Ni) or 1673 K for 2 h (X = Mn) before quenching to form glasses. The glasses were dry-crystallized at ambient pressure and 1193 K for 12 d.

Refinement details

Crystal data, data collection and structure refinement details are summarized in Table 2. For each sample, a small amount of powder was ground and mounted a low-background silicon wafer with a drop of acetone. These were mounted in flat-plate mode on a PANalytical X'Pert Pro MPD diffractometer. X-ray powder diffraction data were collected at 293 K using CuKα X-rays over the range 10–80°2θ using a PANalytical X'Celerator area detector. The powder diffraction data collection times for each sample were 8 h 20 min.

All Bragg reflections in both of the powder diffraction patterns could be indexed in the space group Pbca with similar lattice parameters to that for the Cs₂CdSi₅O₁₂ leucite (Bell et al., 1994a). The crystal structures (Bell & Henderson, 1996) of Cs₂NiSi₅O₁₂ (X = Ni) and Cs₂MnSi₅O₁₂ (X = Mn) were respectively used as starting models for Rietveld (1969) refinements. In both cases, Rb replaced Cs as the extraframework cation. Isotropic atomic displacement parameters were used for all atoms in these phases. In both refinements, the isotropic atomic displacement parameters were constrained to be the same for all sites occupied by the same element, each Rb site had the same displacement parameter as did each Si site and each O site. Soft constraints were used for both refinements, in both cases the Si—O distances were constrained to be 1.61±0.02 Å. For X = Ni, the Ni–O distances were constrained to be 1.88±0.02 Å, the mean Ni—O distance for Cs₂NiSi₅O₁₂. For X = Mn, the Mn—O distances were constrained to be 1.98±0.02 Å, the mean Mn—O distance for Cs₂MnSi₅O₁₂. In both cases, Si and X atoms were ordered onto separate tetra­hedrally coordinated sites, both refined structures are similar.

Computing details

For both compounds, data collection: X'Pert Data Collector (PANalytical, 2006). Program(s) used to refine structure: FULLPROF (Rodríguez-Carvajal, 1993) for Rb2NiSi5O12; FULLPROF (Rodríguez-Carvajal, 2001) for Rb2MnSi5O12. For both compounds, molecular graphics: VESTA (Momma & Izumi, 2008); software used to prepare material for publication: publCIF (Westrip, 2010).

Figures

	Fig. 1. Rietveld difference plot for the single-phase refinement of Rb2NiSi5O12. The red, blue and green lines show, respectively, the observed, calculated and difference plots. Calculated Bragg reflection positions are indicated by crosses.
	Fig. 2. The crystal structure of Rb2NiSi5O12. Green spheres show Rb cations, blue polyhedra show SiO4 units, pink polyhedra show show NiO4 units and red spheres represent O atoms.
	Fig. 3. Rietveld difference plot for the single-phase refinement of Rb2MnSi5O12. The red, blue and green lines show, respectively, the observed, calculated and difference plots. Calculated Bragg reflection positions are indicated by crosses.
	Fig. 4. The crystal structure of Rb2MnSi5O12. Green spheres show Rb cations, blue polyhedra show SiO4 units, purple polyhedra show show MnO4 units and red spheres represent O atoms.

(Rb2NiSi5O12) Dirubidium nickel(II) pentasilicate

Crystal data

Rb2NiSi5O12	Z = 8
Mr = 562.06	Dx = 3.059 (1) Mg m^−3
Orthorhombic, Pbca	Cu Kα radiation, λ = 1.540560 Å
Hall symbol: -P 2ac 2ab	T = 293 K
a = 13.469 (3) Å	purple-blue
b = 13.480 (3) Å	irregular, 10 × 10 mm
c = 13.442 (2) Å	Specimen preparation: Prepared at 1193 K and 100 kPa
V = 2440.7 (8) Å^3

Data collection

PANalytical X'Pert Pro MPD diffractometer	Data collection mode: reflection
Radiation source: X-ray tube	Scan method: step
Specimen mounting: flat plate	2θmin = 9.897°, 2θmax = 79.883°, 2θstep = 0.017°

Refinement

Rp = 9.048	4189 data points
Rwp = 12.007	Profile function: T-C-H Pseudo-Voigt
Rexp = 4.263	73 parameters
RBragg = 10.421	24 restraints
χ^2 = 7.935	Background function: cubic splines interpolation

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

	x	y	z	Uiso*/Ueq
Rb1	0.1367 (17)	0.1209 (19)	0.134 (2)	0.128 (4)*
Rb2	0.4100 (17)	0.3755 (18)	0.4037 (19)	0.128 (4)*
Ni1	0.3641 (11)	0.8370 (10)	0.9075 (10)	0.095 (12)*
Si2	0.1383 (12)	0.6631 (11)	0.6170 (13)	0.082 (6)*
Si3	0.5913 (12)	0.1289 (16)	0.6577 (12)	0.082 (6)*
Si4	0.6419 (14)	0.5703 (13)	0.1057 (11)	0.082 (6)*
Si5	0.9207 (11)	0.3702 (12)	0.8320 (14)	0.082 (6)*
Si6	0.8342 (11)	0.9051 (13)	0.3647 (13)	0.082 (6)*
O1	0.4680 (13)	0.383 (4)	0.155 (5)	0.017 (3)*
O2	0.143 (5)	0.4529 (14)	0.415 (3)	0.017 (3)*
O3	0.378 (5)	0.130 (5)	0.499 (2)	0.017 (3)*
O4	0.7369 (14)	0.436 (3)	0.618 (6)	0.017 (3)*
O5	0.658 (2)	0.7205 (19)	0.353 (6)	0.017 (3)*
O6	0.416 (3)	0.628 (6)	0.7182 (13)	0.017 (3)*
O7	0.9625 (14)	0.861 (6)	0.691 (2)	0.017 (3)*
O8	0.633 (5)	0.977 (2)	0.860 (6)	0.017 (3)*
O9	0.903 (2)	0.633 (6)	0.961 (2)	0.017 (3)*
O10	0.2157 (13)	0.928 (2)	0.146 (5)	0.017 (3)*
O11	0.169 (3)	0.2028 (17)	0.898 (5)	0.017 (3)*
O12	0.881 (5)	0.096 (5)	0.223 (2)	0.017 (3)*

Geometric parameters (Å, º)

Rb1—O1^i	3.51 (6)	Rb2—O11^ii	3.41 (5)
Rb1—O2^ii	3.11 (5)	Rb2—O12^iv	4.15 (7)
Rb1—O3^iii	3.84 (7)	Ni1—O4^xvii	1.93 (4)
Rb1—O3^iv	3.92 (7)	Ni1—O7^xvi	1.90 (3)
Rb1—O4^v	3.68 (8)	Ni1—O9^xviii	1.89 (3)
Rb1—O5^vi	3.08 (4)	Ni1—O11^xix	1.87 (3)
Rb1—O6^vii	3.64 (8)	Si2—O1^xx	1.64 (4)
Rb1—O7^viii	2.72 (4)	Si2—O3^xix	1.66 (4)
Rb1—O8^viii	3.37 (7)	Si2—O5^xii	1.64 (4)
Rb1—O9^viii	3.59 (8)	Si2—O10^xxi	1.66 (3)
Rb1—O10^ix	2.81 (4)	Si3—O1^xxii	1.67 (2)
Rb1—O11^x	3.39 (7)	Si3—O2^xxiii	1.63 (4)
Rb1—O12^xi	3.65 (7)	Si3—O6^xiii	1.67 (2)
Rb1—O12^iv	3.84 (6)	Si3—O11^xxiv	1.63 (5)
Rb2—O1	3.43 (7)	Si4—O2^xxv	1.61 (3)
Rb2—O2	3.75 (7)	Si4—O3^xxvi	1.64 (4)
Rb2—O3	3.58 (7)	Si4—O4^xv	1.64 (3)
Rb2—O4^viii	3.23 (4)	Si4—O12^xxvii	1.65 (4)
Rb2—O5^vi	4.14 (7)	Si5—O5^xxviii	1.64 (3)
Rb2—O5^viii	3.64 (8)	Si5—O7^xxix	1.61 (3)
Rb2—O6^viii	2.85 (4)	Si5—O8^xxx	1.65 (4)
Rb2—O7^xii	3.84 (8)	Si5—O12^xxii	1.62 (4)
Rb2—O8^xiii	3.51 (8)	Si6—O6^xxxi	1.64 (4)
Rb2—O8^xiv	3.65 (6)	Si6—O8^xxxii	1.65 (4)
Rb2—O9^xv	2.63 (4)	Si6—O9^xiv	1.67 (4)
Rb2—O9^xvi	3.92 (8)	Si6—O10^xxv	1.63 (2)
Rb2—O10^i	3.92 (7)
O4^xvii—Ni1—O7^xvi	104 (3)	O5^xxviii—Si5—O7^xxix	127 (3)
O4^xvii—Ni1—O9^xviii	102 (4)	O5^xxviii—Si5—O8^xxx	109 (3)
O4^xvii—Ni1—O11^xix	119 (3)	O5^xxviii—Si5—O12^xxii	99 (5)
O7^xvi—Ni1—O9^xviii	115 (2)	O7^xxix—Si5—O8^xxx	123 (5)
O7^xvi—Ni1—O11^xix	106 (4)	O7^xxix—Si5—O12^xxii	99 (4)
O9^xviii—Ni1—O11^xix	110 (4)	O8^xxx—Si5—O12^xxii	80 (4)
O1^xx—Si2—O3^xix	95 (4)	O6^xxxi—Si6—O8^xxxii	93 (5)
O1^xx—Si2—O5^xii	115 (4)	O6^xxxi—Si6—O9^xiv	94 (3)
O1^xx—Si2—O10^xxi	101 (2)	O6^xxxi—Si6—O10^xxv	131 (4)
O3^xix—Si2—O5^xii	121 (5)	O8^xxxii—Si6—O9^xiv	100 (5)
O3^xix—Si2—O10^xxi	96 (4)	O8^xxxii—Si6—O10^xxv	94 (4)
O5^xii—Si2—O10^xxi	123 (3)	O9^xiv—Si6—O10^xxv	132 (4)
O1^xxii—Si3—O2^xxiii	110 (4)	Si2^vii—O1—Si3^ii	147 (2)
O1^xxii—Si3—O6^xiii	87 (3)	Si3^v—O2—Si4^iv	124 (2)
O1^xxii—Si3—O11^xxiv	134 (4)	Si2^i—O3—Si4^vi	158 (3)
O2^xxiii—Si3—O6^xiii	128 (4)	Ni1^xiii—O4—Si4^xxviii	130 (2)
O2^xxiii—Si3—O11^xxiv	82 (3)	Si2^xxxi—O5—Si5^xv	131 (2)
O6^xiii—Si3—O11^xxiv	120 (5)	Si3^xvii—O6—Si6^xii	129 (2)
O2^xxv—Si4—O3^xxvi	110 (4)	Ni1^xxiv—O7—Si5^xxxiii	124.2 (18)
O2^xxv—Si4—O4^xv	88 (3)	Si5^xxvii—O8—Si6^xxxiv	165 (3)
O2^xxv—Si4—O12^xxvii	112 (4)	Ni1^xxxv—O9—Si6^xxi	129 (2)
O3^xxvi—Si4—O4^xv	106 (5)	Si2^xiv—O10—Si6^iv	117.0 (19)
O3^xxvi—Si4—O12^xxvii	133 (3)	Ni1^i—O11—Si3^xvi	118 (2)
O4^xv—Si4—O12^xxvii	96 (4)	Si4^xxx—O12—Si5^ii	171 (3)

Symmetry codes: (i) −x+1/2, y−1/2, z; (ii) x, −y+1/2, z−1/2; (iii) −x+1/2, −y, z−1/2; (iv) x−1/2, y, −z+1/2; (v) x−1/2, −y+1/2, −z+1; (vi) −x+1, y−1/2, −z+1/2; (vii) −x+1/2, −y+1, z−1/2; (viii) −x+1, −y+1, −z+1; (ix) x, y−1, z; (x) x, y, z−1; (xi) x−1, y, z; (xii) x−1/2, −y+3/2, −z+1; (xiii) −x+1, y−1/2, −z+3/2; (xiv) x, −y+3/2, z−1/2; (xv) −x+3/2, −y+1, z−1/2; (xvi) x−1/2, y, −z+3/2; (xvii) −x+1, y+1/2, −z+3/2; (xviii) x−1/2, −y+3/2, −z+2; (xix) −x+1/2, y+1/2, z; (xx) −x+1/2, −y+1, z+1/2; (xxi) x, −y+3/2, z+1/2; (xxii) x, −y+1/2, z+1/2; (xxiii) x+1/2, −y+1/2, −z+1; (xxiv) x+1/2, y, −z+3/2; (xxv) x+1/2, y, −z+1/2; (xxvi) −x+1, y+1/2, −z+1/2; (xxvii) −x+3/2, y+1/2, z; (xxviii) −x+3/2, −y+1, z+1/2; (xxix) −x+2, y−1/2, −z+3/2; (xxx) −x+3/2, y−1/2, z; (xxxi) x+1/2, −y+3/2, −z+1; (xxxii) −x+3/2, −y+2, z−1/2; (xxxiii) −x+2, y+1/2, −z+3/2; (xxxiv) −x+3/2, −y+2, z+1/2; (xxxv) x+1/2, −y+3/2, −z+2.

(Rb2MnSi5O12) Dirubidium managnese(II) pentasilicate

Crystal data

Rb2MnSi5O12	Z = 8
Mr = 557.92	Dx = 2.975 (1) Mg m^−3
Orthorhombic, Pbca	Cu Kα radiation, λ = 1.540560 Å
Hall symbol: -P 2ac 2ab	T = 293 K
a = 13.4085 (10) Å	pale brown-pink
b = 13.6979 (11) Å	irregular, 10 × 10 mm
c = 13.5761 (10) Å	Specimen preparation: Prepared at 1193 K and 100 kPa
V = 2493.5 (3) Å^3

Data collection

PANalytical X'Pert Pro MPD diffractometer	Data collection mode: reflection
Radiation source: X-ray tube	Scan method: step
Specimen mounting: flat plate	2θmin = 10.139°, 2θmax = 80.125°, 2θstep = 0.017°

Refinement

Rp = 6.527	4189 data points
Rwp = 8.847	Profile function: T-C-H Pseudo-Voigt
Rexp = 3.909	71 parameters
RBragg = 5.989	24 restraints
χ^2 = 5.121	Background function: cubic splines interpolation

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

	x	y	z	Uiso*/Ueq
Rb1	0.1311 (8)	0.1348 (8)	0.1535 (6)	0.0544 (19)*
Rb2	0.3811 (10)	0.3968 (7)	0.3953 (7)	0.0544 (19)*
Mn1	0.3707 (9)	0.8342 (7)	0.9432 (6)	0.00127*
Si2	0.1328 (11)	0.6601 (10)	0.6011 (10)	0.024 (2)*
Si3	0.5921 (11)	0.1146 (10)	0.6255 (10)	0.024 (2)*
Si4	0.6449 (10)	0.5993 (11)	0.1054 (10)	0.024 (2)*
Si5	0.9006 (10)	0.3583 (11)	0.8208 (10)	0.024 (2)*
Si6	0.8312 (11)	0.9317 (10)	0.3329 (11)	0.024 (2)*
O1	0.4758 (12)	0.355 (3)	0.1607 (19)	0.00127*
O2	0.083 (2)	0.4973 (12)	0.399 (3)	0.00127*
O3	0.397 (2)	0.150 (3)	0.4890 (13)	0.00127*
O4	0.7328 (12)	0.4230 (19)	0.620 (2)	0.00127*
O5	0.659 (2)	0.7326 (14)	0.360 (2)	0.00127*
O6	0.363 (4)	0.632 (2)	0.7605 (17)	0.00127*
O7	0.9822 (15)	0.877 (3)	0.645 (2)	0.00127*
O8	0.659 (3)	0.9530 (13)	0.860 (3)	0.00127*
O9	0.893 (2)	0.635 (2)	0.9119 (13)	0.00127*
O10	0.2099 (14)	0.9281 (18)	0.135 (3)	0.00127*
O11	0.132 (3)	0.1885 (12)	0.9545 (17)	0.00127*
O12	0.892 (3)	0.139 (3)	0.2068 (12)	0.00127*

Geometric parameters (Å, º)

Rb1—O1^i	4.09 (4)	Rb2—O10^i	3.76 (4)
Rb1—O2^ii	3.51 (3)	Rb2—O11^iii	3.63 (4)
Rb1—O2^iii	3.95 (4)	Rb2—O12^iv	3.80 (4)
Rb1—O3^iv	3.69 (3)	Mn1—O4^xvii	2.04 (2)
Rb1—O4^v	3.46 (3)	Mn1—O7^xvi	2.00 (3)
Rb1—O5^vi	3.12 (3)	Mn1—O9^xviii	2.03 (2)
Rb1—O6^vii	3.51 (3)	Mn1—O11^xix	2.002 (19)
Rb1—O7^viii	3.13 (3)	Si2—O1^xx	1.68 (2)
Rb1—O8^viii	3.07 (4)	Si2—O3^xix	1.58 (2)
Rb1—O9^viii	3.29 (3)	Si2—O5^xii	1.60 (2)
Rb1—O10^ix	3.03 (3)	Si2—O10^xxi	1.66 (3)
Rb1—O11^x	2.80 (2)	Si3—O1^xxii	1.68 (2)
Rb1—O12^xi	3.29 (4)	Si3—O2^xxiii	1.57 (2)
Rb1—O12^iv	3.98 (4)	Si3—O6^xiv	1.68 (3)
Rb2—O1	3.48 (3)	Si3—O11^xxiv	1.58 (3)
Rb2—O2	4.23 (3)	Si4—O2^xxv	1.63 (2)
Rb2—O3	3.62 (4)	Si4—O3^xxvi	1.56 (3)
Rb2—O4^viii	2.91 (3)	Si4—O4^xv	1.68 (2)
Rb2—O5^vi	4.17 (3)	Si4—O12^xxvii	1.56 (3)
Rb2—O5^viii	3.80 (3)	Si5—O5^xxviii	1.57 (3)
Rb2—O6^vii	3.77 (5)	Si5—O7^xxix	1.66 (3)
Rb2—O6^viii	4.05 (5)	Si5—O8^xiii	1.61 (3)
Rb2—O7^xii	3.43 (4)	Si5—O12^xxii	1.55 (2)
Rb2—O7^xiii	3.86 (3)	Si6—O6^xxx	1.60 (3)
Rb2—O8^xiv	3.45 (4)	Si6—O8^xxxi	1.63 (2)
Rb2—O9^xv	3.07 (3)	Si6—O9^xxxii	1.63 (3)
Rb2—O9^xvi	4.19 (3)	Si6—O10^xxv	1.68 (3)
O4^xvii—Mn1—O7^xvi	94.7 (15)	O5^xxviii—Si5—O7^xxix	121 (3)
O4^xvii—Mn1—O9^xviii	112.6 (18)	O5^xxviii—Si5—O8^xiii	106 (2)
O4^xvii—Mn1—O11^xix	128.0 (19)	O5^xxviii—Si5—O12^xxii	109 (3)
O7^xvi—Mn1—O9^xviii	114.1 (18)	O7^xxix—Si5—O8^xiii	105 (3)
O7^xvi—Mn1—O11^xix	111 (2)	O7^xxix—Si5—O12^xxii	110 (3)
O9^xviii—Mn1—O11^xix	97.8 (16)	O8^xiii—Si5—O12^xxii	106 (3)
O1^xx—Si2—O3^xix	104 (2)	O6^xxx—Si6—O8^xxxi	134 (3)
O1^xx—Si2—O5^xii	98 (2)	O6^xxx—Si6—O9^xxxii	94 (2)
O1^xx—Si2—O10^xxi	109 (2)	O6^xxx—Si6—O10^xxv	117 (3)
O3^xix—Si2—O5^xii	117 (3)	O8^xxxi—Si6—O9^xxxii	111 (2)
O3^xix—Si2—O10^xxi	111 (3)	O8^xxxi—Si6—O10^xxv	93 (2)
O5^xii—Si2—O10^xxi	116 (2)	O9^xxxii—Si6—O10^xxv	108 (2)
O1^xxii—Si3—O2^xxiii	103 (3)	Si2^vii—O1—Si3^iii	134.2 (18)
O1^xxii—Si3—O6^xiv	92 (3)	Si3^v—O2—Si4^iv	142.2 (18)
O1^xxii—Si3—O11^xxiv	111 (3)	Si2^i—O3—Si4^vi	137.6 (19)
O2^xxiii—Si3—O6^xiv	111 (3)	Mn1^xiv—O4—Si4^xxviii	120.5 (14)
O2^xxiii—Si3—O11^xxiv	120 (2)	Si2^xxx—O5—Si5^xv	136.4 (19)
O6^xiv—Si3—O11^xxiv	115 (2)	Si3^xvii—O6—Si6^xii	138 (2)
O2^xxv—Si4—O3^xxvi	100 (3)	Mn1^xxiv—O7—Si5^xxxiii	146.0 (16)
O2^xxv—Si4—O4^xv	110 (2)	Si5^xxvii—O8—Si6^xxxiv	138.2 (19)
O2^xxv—Si4—O12^xxvii	100 (3)	Mn1^xxxv—O9—Si6^xxi	132.5 (15)
O3^xxvi—Si4—O4^xv	122 (2)	Si2^xxxii—O10—Si6^iv	134.1 (18)
O3^xxvi—Si4—O12^xxvii	117 (2)	Mn1^i—O11—Si3^xvi	125.5 (15)
O4^xv—Si4—O12^xxvii	106 (3)	Si4^xiii—O12—Si5^iii	155.7 (19)

Symmetry codes: (i) −x+1/2, y−1/2, z; (ii) −x, y−1/2, −z+1/2; (iii) x, −y+1/2, z−1/2; (iv) x−1/2, y, −z+1/2; (v) x−1/2, −y+1/2, −z+1; (vi) −x+1, y−1/2, −z+1/2; (vii) −x+1/2, −y+1, z−1/2; (viii) −x+1, −y+1, −z+1; (ix) x, y−1, z; (x) x, y, z−1; (xi) x−1, y, z; (xii) x−1/2, −y+3/2, −z+1; (xiii) −x+3/2, y−1/2, z; (xiv) −x+1, y−1/2, −z+3/2; (xv) −x+3/2, −y+1, z−1/2; (xvi) x−1/2, y, −z+3/2; (xvii) −x+1, y+1/2, −z+3/2; (xviii) x−1/2, −y+3/2, −z+2; (xix) −x+1/2, y+1/2, z; (xx) −x+1/2, −y+1, z+1/2; (xxi) x, −y+3/2, z+1/2; (xxii) x, −y+1/2, z+1/2; (xxiii) x+1/2, −y+1/2, −z+1; (xxiv) x+1/2, y, −z+3/2; (xxv) x+1/2, y, −z+1/2; (xxvi) −x+1, y+1/2, −z+1/2; (xxvii) −x+3/2, y+1/2, z; (xxviii) −x+3/2, −y+1, z+1/2; (xxix) −x+2, y−1/2, −z+3/2; (xxx) x+1/2, −y+3/2, −z+1; (xxxi) −x+3/2, −y+2, z−1/2; (xxxii) x, −y+3/2, z−1/2; (xxxiii) −x+2, y+1/2, −z+3/2; (xxxiv) −x+3/2, −y+2, z+1/2; (xxxv) x+1/2, −y+3/2, −z+2.