Kalyuzhnyite-(Ce), NaKCaSrCeTi(Si₈O₂₁)OF(H₂O)_3, a new mineral from the Darai-Pioz alkaline massif, Tien-Shan mountains, Tajikistan: mineral description, crystal structure and a new double (Si₈O₂₁) sheet Open Access

This paper reports the description and the crystal structure of kalyuzhnyite-(Ce) [Russian Cyrillic: калюжныит-(Ce)], ideally NaKCaSrCeTi(Si₈O₂₁)OF(H₂O)₃, a new mineral from the well-known Darai-Pioz alkaline massif, Tien-Shan Mountains, Central Tajikistan. Kalyuzhnyite-(Ce) is a sheet-silicate mineral with large channels and can potentially be used as a model for synthesis of microporous materials of industrial interest. Single-crystal X-ray diffraction revealed a double Si₈O₂₁ sheet of ten-membered rings of SiO₄ tetrahedra that has never been described in minerals (Hawthorne et al., 2019). The name is in honour of Vasily Avksentievich Kalyuzhny (Russian Cyrillic: Василий Авксентьевич Калюжный) (1899–1993), a prominent Russian geologist, an authority on the geology of the Komi Republic and its ore deposits; his pioneering work resulted in the discovery of the Yaregskoe Ti-deposit in oil-bearing sandstones of the Komi Republic. Dr. Kalyuzhny was also a member of the Pamir-Tajik geological expedition (1934–1937), and he studied granite pegmatites of the Turkestan ridge around rivers of the Karavshin system and Sn-bearing pegmatites of the Kalbinsky ridge in Kazakhstan. The new mineral and its name have been approved by the Commission on New Minerals, Nomenclature and Classification, International Mineralogical Association (IMA2022-133, Agakhanov et al., 2023). The holotype material is deposited in the systematic collection of the Fersman Mineralogical Museum, Moscow, Russia, catalogue number 98144. The IMA mineral symbol is Kalu-Ce.

Kalyuzhnyite-(Ce) occurs in the moraine of the Darai-Pioz glacier in the upper reaches of the Darai-Pioz River, Tien-Shan mountains, near the junction of the Turkestansky, Zeravshansky and Alaisky ridges (39°30′N 70°40′E). This area is in the Rasht (formerly Garm) district, Central Tajikistan. The Darai-Pioz alkaline massif belongs to the Upper Paleozoic Alaysky (Matchaisky) intrusive complex. Information on the geology of the Darai-Pioz massif can be found in Pautov et al. (2019) with reference to relevant earlier publications. Kalyuzhnyite-(Ce) was found in boulders (up to 2 m across) of quartz rock, i.e. silexite boulders composed of 90–95% medium- to coarse-grained quartz of ice-like appearance (quartz grains vary from 2 mm to 2 cm) characteristic for moraine deposits of the Darai-Pioz glacier. The following minor and accessory minerals are present: large (up to 10 cm across) golden-brown tabular and lamellar crystals of polylithionite, pink plates of sogdianite–sugilite, pale-yellow to orange aggregates and tabular crystals of reedmergnerite, black crystals of aegirine, orange–brown semi-transparent lenticular crystals of stillwellite-(Се), grass-green or yellowish-green semi-transparent and transparent crystals of leucosphenite, dark-green crystals of turkestanite and aggregates of large white grains of microcline. Also present are fine-grained brown or greyish-brown aggregates of Mn-bearing pectolite, quartz, Sr-bearing fluorite and a variety of rare minerals. Kalyuzhnyite-(Ce) occurs in these pectolite-rich aggregates (Fig. 1a,b), associated with quartz, fluorite, pectolite, baratovite, aegirine, leucosphenite, neptunite, reedmergnerite, orlovite, sokolovaite, mendeleevite-(Ce), odigitriaite, pekovite, zeravshanite, kirchhoffite and garmite. The origin of the silexite boulders with pectolite aggregates, in which kalyuzhnyite-(Ce) was found, is debatable as these rocks have not been investigated in situ. The problem of their genesis has been discussed by Pautov et al. (2022).

Kalyuzhnyite-(Ce) occurs as equant grains with poorly developed faces in a quartz–pectolite aggregate (Fig. 1a,b). The two grains of kalyuzhnyite-(Ce) up to 70 μm were found by scanning electron microscopy in a small hand-sample. One part of grain 1 (Fig. 1a) was used for the crystal-structure work (crystal 1) and the second part of grain 1 was used for the electron-microprobe analysis (crystal 2). The microprobe mount of crystal 2 was deposited as a holotype Kalyuzhnyite-(Ce), catalogue number 98144. The second smaller grain 2 up to 50 μm (Fig. 1b) was used for Raman spectroscopy (in a thin section).

The grains are colourless and have a vitreous lustre. Kalyuzhnyite-(Ce) has an uneven fracture and does not fluoresce under cathode waves or ultraviolet light. The cleavage is {010} perfect and no parting was observed. The hardness of kalyuzhnyite-(Ce) was not measured due to the very small size of grain 2. The mineral is brittle and D_calc. = 3.120 g/cm³.

Individual grains show no visible twinning. Kalyuzhnyite-(Ce) is nonpleochroic. In reflected light, kalyuzhnyite-(Ce) is grey. Reflectance values were measured with a UMSP-50 Opton microspectrophotometer using the Opton SiC standard 474251 (with a spectral slot width of 10 nm) and are given in Table 1. Kalyuzhnyite-(Ce) has very weak bireflectance.

The Raman spectrum of kalyuzhnyite-(Ce) (Fig. 2) was obtained on a randomly oriented crystal (polished section, Fig. 1b) using Thermo DXR2xi Raman imaging confocal microscope with a green laser (532 nm) at room temperature. The output power of the laser beam was 10 mW (at 100% power), holographic diffraction grating was used with 1800 lines cm^–1, spectral resolution was 2 cm^–1, the range was from 50 to 1800 cm^–1. The diameter of the focal spot on the sample was 2 μm. The back-scattered Raman signal was collected with a 100× objective; signal acquisition time for a single scan of the spectral range was 2.0 s and the signal was averaged over 20 scans. The spectrum was processed using Omnic software.

Lines in the Raman spectrum of kalyuzhnyite-(Ce) can be grouped into three frequency ranges by analogy with data obtained for some multi-ring silicates of Na, Sr, Ti and REE (Sitarz et al., 1998; Frost and Xi, 2012; Tobbens et al., 2008). The 800–1100 cm^–1 range bands are due to Si–O bond-stretching modes, the 500 to 800 cm^–1 vibrations are due to O–Si–O bending modes. The spectral region lower than 500 cm^–1 corresponds to Me–O stretching vibrations and lattice modes, while some bonds from 300 to 500 cm^–1 can correspond to SiO₄-unit vibrations.

The chemical composition of kalyuzhnyite-(Ce) was determined using a JEOL 773 electron microprobe (energy dispersive spectroscopy mode, an accelerating voltage of 20 kV, a specimen current of 2 nA and a beam diameter of 5 μm; Fersman Mineralogical Museum). The following standards were used: microcline USNM 143966 (Si,K), albite 107 (Na), SrTiO₃ (Sr), BaSO₄ (Ba), PbTiO₃ (Pb), CsTbP₄O₁₂ (Cs), ilmenite USNM 96189 (Ti), anorthite USNM 137041 (Ca), LiNbO₃ (Nb), LaPO₄ (La), CePO₄ (Ce), PrPO₄ (Pr), NdPO₄ (Nd), SmPO₄ (Sm), GdPO₄ (Gd), HoPO₄ (Ho), ErPO₄ (Er), and MgF₂ (F). The data (10 analyses) were reduced and corrected by the PAP method of Pouchou and Pichoir (1985). The chemical composition of kalyuzhnyite-(Ce) is the mean of 10 determinations and is given in Table 2. The empirical formula calculated on 26.11 (O + F) apfu (atoms per formula unit) is Na_1.07K_0.37Cs_0.30Sr_1.21Ca_0.37Pb_0.24Ba_0.06(Ce_0.43Nd_0.41Pr_0.08La_0.06Sm_0.03Gd_0.01Er_0.01Ho_0.01)_Σ1.04(Ti_0.97Nb_0.04)_Σ1.01Si_8.06O_25.21F_0.90H_6.42 with Z = 4. The structural formula based on assigned site-populations is (Na_0.80□_0.20)_Σ1(K_0.37Cs_0.30)_Σ0.67[(Ca_0.25Pb_0.22Sr_0.14)Na_0.25Ln³⁺_0.08]_Σ0.94(Sr_0.98Pb_0.02)_Σ1(Ln³⁺_0.96Ca_0.04)_Σ1(Ti_0.97Nb_0.03)_Σ1.00(Si₈O₂₁)OF_0.50[F_0.40(H₂O)_0.10](H₂O)_3.11, where Ln³⁺_1.04 = (Ce_0.43Nd_0.41Pr_0.08La_0.06Sm_0.03Gd_0.01Er_0.01Ho_0.01)_Σ1.04, □ = vacancy and Ba_0.06 does not belong to this mineral species (see below). The simplified formula is (Na,□)(K,Сs)(Ca,Pb,Sr,Na)SrLn³⁺Ti(Si₈O₂₁)OF(H₂O)₃, where Ce is the dominant lanthanoid. The ideal formula of kalyuzhnyite-(Ce), NaKCaSrCeTi(Si₈O₂₁)OF(H₂O)₃ (see below), requires (wt.%) Na₂O 3.06, K₂O 4.58, CaO 5.46, SrO 10.08, Ce₂O₃ 15.97, TiO₂ 7.78, SiO₂ 46.78, F 1.85, H₂O 5.26, O ≡ F –0.78, Total 100.00.

Powder X-ray diffraction data were obtained by collapsing single-crystal experimental data into two dimensions. Data (in Å for MoKα) are listed in Table 3. Unit-cell parameters are therefore the same as from the single-crystal data (Table 4).

X-ray single-crystal data for kalyuzhnyite-(Ce) were collected from a twinned crystal with a Bruker APEX II ULTRA three-circle diffractometer with a rotating-anode generator (MoKα), multilayer optics and an APEX II 4K CCD detector. The intensities of reflections with –26 ≤ h ≤ 26, –15 ≤ k ≤ 15 and –20 ≤ l ≤ 20 were collected with a frame width of 0.3° and a frame time of 6 s up to 2θ ≤ 60.21°, and an empirical absorption correction (SADABS, Sheldrick, 2015) applied. There were few observed reflections at high 2θ, and refinement of the structure was done for 2θ ≤ 55°, −24 ≤ h ≤ 24, –14 ≤ k ≤ 15 and –19 ≤ l ≤ 19. The crystal-structure solution by direct methods and refinement were done with the Bruker SHELXTL Version 2014/3 software (Sheldrick, 2015) in space group P2/c. We refined the structure as two components related by the TWIN matrix (⁠|$\bar{1}$| 0 0, 0 |$\bar{1}$| 0, 1 0 1). The crystal structure of kalyuzhnyite-(Ce) was refined to R₁ = 2.74%, the twin ratio being 0.5037(9) : 0.4963(9) (Table 4). Details of data collection and structure refinement are given in Table 4. The occupancies of ten cation sites were refined with the following scattering curves: M1 site: Ti; M2 site: Nd; M3 site: Sr; M4 site: Na; M(5,6) sites: Pb; M(7,8) sites: Ca; A(1,2) and B(1,2) sites: K and Cs, respectively. The occupancies of the F and X sites were refined with the scattering curve of F, and W(1–9) sites, with the scattering curve of O. Refinement of the F, X and W1 site-occupancies converged to integer values (within 3 e.s.d.) and were subsequently fixed at full occupancy. Scattering curves for neutral atoms were taken from the International Tables for Crystallography (Wilson, 1992). Final atom coordinates and anisotropic displacement parameters are given in Table 5, selected interatomic distances and angles in Table 6, refined site-scattering values and assigned site-populations in Table 7, and bond-valence values in Tables 8 and 9. A list of observed and calculated structure factors and a Crystallographic Information File (CIF) have been deposited with the Principal Editor of Mineralogical Magazine and are available as Supplementary Material (see below). The structure diagrams were drawn using ATOMS 64 software (Dowty, 2016).

Si is assigned to the eight tetrahedrally coordinated Si(1–8) sites, with <Si–O> = 1.614 Å (Tables 5, 6).

The four M sites in the heteropolyhedral sheet (Table 5) are considered next. At the [6]-coordinated M1 site, the refined scattering is 21.97 electrons per formula unit (epfu) and the mean bond-length <M1–φ> = 1.983 Å (φ = O, F). In accord with the grand mean bond-length <^[6]Ti⁴⁺–O> = 1.971 Å> observed in inorganic structures (Gagné and Hawthorne, 2020), we assign all available Ti (Table 2) and minor Nb to the M1 site: Ti_0.97Nb_0.03 apfu, with calculated site-scattering of 22.57 epfu (Table 7). At the [8]-coordinated M2 site, the refined scattering is 57.1 epfu and the mean bond-length <M2–φ> = 2.494 Å (φ = O, F). In accord with the grand mean bond-length <^[8]Ce³⁺–O> = 2.495 Å> observed in inorganic structures (Gagné, 2018), we assign REE = Ln³⁺_0.96, where Ce is the dominant lanthanoid (Ln = lanthanoids), and minor Ca: 0.04 apfu, to the M2 site: Ln_0.96Ca_0.04 pfu [^[8]Ce³⁺: r = 1.143 Å, ^[8]Ca²⁺: r = 1.12 Å, Shannon (1976)] (Tables 5–7). There is a good agreement between refined and calculated site-scattering for the M2 site: 57.1 and 57.59 epfu, respectively (Table 7). On the basis of the refined site-scattering values and observed bond-distances, we assign Sr_0.98Pb_0.02 and Ca_0.80□_0.20 pfu to the M3 and M4 sites, respectively (Table 7).

Consider the interstitial M(5–8), A(1,2) and B(1,2) sites. The refined site-scattering value at the M5 site is slightly higher than at the M6 site: 15.3 versus 14.7 epfu; the mean bond-length values: 2.515 and 2.531 Å, are very similar. Therefore, we allocate more Pb (the heaviest scatter) to the M5 site: Pb_0.16Ca_0.11□_0.23 pfu, and less Pb to the M6 site: Sr_0.14Ln_0.08Pb_0.06Ca_0.11□_0.22 pfu (Table 7). Based on the refined site-scattering values and observed bond-distances, we assign Na_0.25□_0.75 and Ca_0.14□_0.86 pfu to the M7 and M8 sites, respectively (Table 7).

The refined site-scattering values at the [11]- and [|${9}$|]- coordinated A(1,2) and B(1,2) sites sum to 21.01 epfu (Table 7). The cations to be assigned to these sites are Cs_0.30, K_0.37 and Ba_0.06, with individual calculated scattering of 16.5, 7.03 and 3.36 epfu, respectively, and a total calculated scattering 26.89 epfu (Table 2). The A(1,2) sites, with total refined site-scattering of 15.7 epfu and derived cations’ radii of 1.85 and 1.88 Å, respectively [e.g. calculated as <A1–O> – r(^[4]O^2–): 3.320–1.38 = 1.85 Å] must be occupied by Cs [^[11]Cs: r = 1.85 Å, Shannon (1976)]. Hence, we assign Cs_0.20□_0.30 and Cs_0.10□_0.40 pfu to the A1 and A2 sites, with calculated site-scattering values of 11.0 and 5.50 epfu, respectively, with a sum of 16.5 epfu (Table 7). The B(1,2) sites, with total refined site-scattering of 5.4 epfu and derived cation radii of 1.74 and 1.60 Å, respectively, must be occupied by K [^[11]K: r = 1.62 Å, ^[9]K: r = 1.55 Å, Shannon (1976)]. Hence, we assign K_0.20□_0.30 and K_0.17□_0.33 pfu to the B1 and B2 sites, with calculated site-scattering values of 3.80 and 3.23 epfu, respectively, and a sum of 7.03 epfu (Table 7). There is a good agreement between refined and calculated site-scattering values for (1) the Cs-bearing A(1,2) sites: 15.7 versus 16.50 epfu; (2) the K-bearing B(1,2) sites: 5.4 versus 7.03 epfu and (3) A + B sites: 21.1 versus 23.53 epfu (Table 7). We were not able to assign Ba_0.06 apfu (calculated scattering is 3.36 epfu; ^[11]Ba: r = 1.57 Å, ^[9]Ba: r = 1.47 Å) to the A and B sites and suggest that Ba_0.06 does not occur in the crystal structure of kalyuzhnyite-(Ce).

The crystal structure of kalyuzhnyite-(Ce) contains three groups of cation sites: M sites of the heteropolyhedral sheet, Si sites of the Si–O sheet and interstitial M(5–8), A(1,2) and B(1,2) sites.

There are four cation sites in the heteropolyhedral sheet: the Ti-dominant M1 site, the Ce-dominant M2 site, the Sr-dominant M3 site, and the Na-dominant M4 site (Fig. 3a). The M1 site is occupied by Ti_0.97Nb_0.03 apfu, ideally Ti apfu. The M1 atom is octahedrally coordinated by six anions: five O atoms and one F atom, <M1–φ> = 2.983 Å (φ = O and F) (Tables 5–7). The M2 site is occupied by Ln_0.96Ca_0.04 apfu, ideally Ce apfu (Ce is the dominant lanthanoid). The M2 atom is coordinated by eight anions: seven O atoms and one X anion, <M2–φ> = 2.494 Å (φ = O, F and H₂O) (Tables 5–7). The M3 site is occupied by Sr_0.98Pb_0.02 pfu, ideally Sr apfu. The M3 atom is coordinated by eight anions: seven O atoms and one F atom, <M3–φ> = 2.631 Å (φ = O and F) (Tables 6, 7). The M4 site is occupied by Na at 80%, its ideal composition is Na apfu. The M4 atom is coordinated by seven anions: six O atoms and one X anion, <M4–φ> = 2.591 Å (φ = O, F and H₂O) (Tables 6, 7). Ideally, cations of the heteropolyhedral sheet, ^M4Na + ^M3Sr + ^M2Ce + ^M1Ti, sum to (NaSrCeTi)¹⁰⁺ apfu (Table 7).

In the Si–O sheet (Fig. 3b), there are eight Si(1–8) tetrahedrally coordinated sites occupied by Si with a <Si–O> distance of 1.614 Å (Tables 5, 6). We consider the Si–O sheet a complex anion (see below).

Interstitial cations occur at the eight partly occupied M(5–8), A(1,2) and B(1,2) sites within the double Si–O sheet (Fig. 4a,b; Table 5). There are short distances between cations which occur at the four interstitial sites (Table 6): M6–B1 = 0.65(3) Å and B2–A2 = 1.03(2) Å, so these sites can be only alternately occupied. There is a short distance between two points of the M7 site: M7–M7′ = 0.88(4) Å (Table 6), hence the M7 site must be occupied at ≤ 50%. The M5 and M6 sites are 54% and 56% occupied; their compositions are Pb_0.16Ca_0.11□_0.23 and Sr_0.14Ln_0.08Pb_0.06□_0.22 pfu, respectively. The M5 and M6 sites are each coordinated by four basal O atoms of Si tetrahedra, an X ligand and H₂O groups at the partly occupied W(3,7) sites [M5] and W(4,8) sites [M6] (Fig. 5a; Table 6). Because of short O–O distances [W3–W7 = 1.67 Å and W4–W8 = 1.79 and 1.68 Å (Table 6)] the W3–W7 and W4–W8 sites can be only alternately occupied. The partial occupancies of the W3 site (28%) + W7 site (26%) (Table 5) sum to 54%, equivalent to the 54%-occupancy of the M5 site. The partial occupancies of the W4 site (32%) + W8 site (24%) (Table 5) sum to 56%, equivalent to the 56%-occupancy of the M6 site. Therefore, we consider [W3 + W7] and [W4 + W8] as two aggregate ligands (H₂O groups) for the M5 and M6 atoms, respectively. Hence M5 and M6 atoms are each coordinated by four O atoms of Si tetrahedra, an X ligand (see Anion sites below) and an H₂O group, with <M5–φ> = 2.515 Å and <M6–φ> = 2.531 Å (Tables 6, 7). The M7 and M8 sites are occupied by Na at 25% and Ca at 14%, respectively (Fig. 5b; Table 7). The M7 and M8 atoms are coordinated by four O atoms and two H₂O groups at the W(2,4) sites [M7] and W(1,6) sites [M8], with <M7–φ> = 2.53 Å and <M8–φ> = 2.380 Å (Tables 6, 7). The composition of the four M(5–8) sites is [(Ca_0.25Pb_0.22Sr_0.14)Na_0.25Ln_0.08)]_Σ0.94, ideally Ca apfu (Table 7). The two [11]-coordinated A1 and A2 sites are occupied by Cs at 40% and 20%, respectively (Fig. 5c; Tables 5–7). The Cs atoms are coordinated by eight O atoms, two H₂O groups (W1 site) and a F atom, with <A1–φ> = 3.230 Å and <A2–φ> = 3.255 Å (Tables 6, 7). The [9]-coordinated B1 and [11]-coordinated B2 sites are occupied by K at 40% and 34%, respectively (Fig. 5d,e; Tables 5–7). The B1 atom is coordinated by six O atoms, two H₂O groups (W2 site) and one X ligand, with <B1–φ> = 3.12 Å and the B2 atom is coordinated by eight O atoms, two H₂O groups (W1 site) and a F atom, with <B2–φ> = 2.983 Å (Tables 6, 7). The A(1,2) + B(1,2) sites are occupied by (K_0.37Cs_0.30)_Σ0.67, ideally K apfu (Table 7). Ideally, interstitial cations, ^M(5–8)Ca + ^A(1,2)B(1,2)K, sum to (KCa)³⁺ pfu.

Cations of the heteropolyhedral sheet and interstitial complex ideally sum to (NaSrCeTi)¹⁰⁺ + (KCa)³⁺ = (NaKCaSrCeTi)¹³⁺.

There are 21 anion sites, O(1–21), occupied by O atoms which form the tetrahedral coordination of the Si(1–8) atoms (Tables 5, 6). The Si(1–8) and O(1–21) atoms form a distinct complex Si–O oxyanion, (Si₈O₂₁)^10– per formula unit.

The M(1–4) cations of the heteropolyhedral sheet are coordinated by O atoms shared with the Si–O sheet, and three other ligands. There is one O atom (O22), which is a common vertex for the M1 [Ti], M2 [Ce] and M3 [Sr] polyhedra, it receives bond-valence of 1.90 vu (valence units) (Table 8). The F atom receives bond-valence contributions from two M1 [Ti] atoms, two M3 [Sr] atoms and interstitial cations A(1,2) [Cs], and B2 [K], with a total sum of 0.93 vu (Table 8). The F site gives F_0.5 apfu (Table 7). The X anion (Tables 5, 9) is involved in short-range order (SRO). SRO-1 (80%) occurs where the X anion receives bond-valence contributions from two M4 [Na] atoms [note that the M4 site is occupied by Na at 80%, Tables 5, 7, 9], two M2 [Ce] atoms and interstitial cations: M5 [Pb], M6 [Sr] and B1 [K], with a total sum of 0.92 vu (Table 9), hence X is a F atom. SRO-2 (20%) occurs where the X anion does not receive bond-valence contributions from two M4 [Na] atoms [note that the M4 site is vacant at 20%, Tables 5, 7] and M(6) [Sr], but receives bond-valence contributions from two M2 [Ce] atoms and interstitial cations: M5 [Pb] and B1 [K], with a total sum of 0.61 vu (Table 9), hence X is an H₂O group. We write the composition of the X ligand as F_0.40(H₂O)_0.10 pfu, ideally F_0.5 apfu (Table 7). Hence ideal composition of the anions/H₂O groups of the heteropolyhedral sheet not shared with the Si–O sheet is ^O22O^2– + ^FF_0.5^0.5– +^XF_0.5^0.5– = (OF)^3– pfu.

The O atoms of H₂O groups at the W(1–9) sites occur within the Si–O sheet in the large channels at y ≈ 0 (Fig. 4a,b; Table 5). The O atom at the W1 site receives 0.20 vu from interstitial cations (Table 8) and it is an O atom of an H₂O group. The W1 site is 100%-occupied by an H₂O group, giving (H₂O)_1.0 pfu (Tables 5, 7; note positions of the two H atoms in Table 5). The eight W(2–9) sites are partly occupied by H₂O groups at 14–42% (Table 5), in total giving (H₂O)_2.11 pfu (Table 7). The W(1–9) sites give (H₂O)_3.11 pfu, ideally (H₂O)₃ pfu.

The complex Si–O anion, and the simple anions of the heteropolyhedral sheet and interstitial complex (charge is given in brackets), (Si₈O₂₁) (10^–) + (OF) [O22,X,F] (3^–) + (H₂O)₃ [W(1–9)] (0), sum to (Si₈O₂₁)OF(H₂O)₃, with a total charge of 13^–.

There are two main structural units in the crystal structure of kalyuzhnyite-(Ce): a heteropolyhedral sheet and a double Si–O sheet (Fig. 3a,b). The heteropolyhedral sheet parallel to (010) is composed of Ti-dominant M1 octahedra, Ce-dominant M2 polyhedra, and Sr-dominant M3 polyhedra and Na-dominant M4 polyhedra which share edges and vertices (Fig. 3a, Tables 5–7). We sum the cation and anion parts of the heteropolyhedral sheet to derive its ideal composition: (NaSrCeTi)¹⁰⁺ + (OF)^3– = [NaSrCeTi(OF]⁷⁺. The double Si–O sheet parallel to (010) is composed of ten-membered rings of SiO₄ tetrahedra (Figs 3b, 4b). The ten-membered rings occur at y ≈ + 0.265 and y ≈ –0.265 and connect via two SiO₄ tetrahedra which form an Si4–O15–Si5 bridge oriented along b (Fig. 4a; Table 6). The composition of this double Si–O sheet is (Si₈O₂₁)^10–. Such a double Si–O sheet is new – a sheet with this topology and composition has never been described in minerals before (Hawthorne et al., 2019); we call this the kalyuzhnyite sheet. The heteropolyhedral and Si–O sheets alternate along b; there is one heteropolyhedral sheet and one Si–O sheet per b unit-cell parameter (Fig. 4a). The sheets connect via common vertices of M(1–4) polyhedra and SiO₄ tetrahedra to form a framework. Summation of the ideal compositions of the heteropolyhedral sheet and double Si–O sheet, gives ideal composition of the framework: [NaSrCeTiOF]⁷⁺ + (Si₈O₂₁)^10– = NaSrCeTi(Si₈O₂₁)OF]^3–. The interstitial complex is composed of M(5–8), A(1,2) and B(1,2) cations and H₂O groups at the W(1–9) sites. The interstitial complex is located within the double Si–O sheet (Figs 4, 5). The larger cations, Pb and Sr at the M(5,6) sites, Cs at the A(1,2) sites and K at the B(1,2) sites occupy centres of ten membered rings of SiO₄ tetrahedra (Fig. 4b). Smaller cations, Na and Ca at the M7 and M8 sites and H₂O groups at the W(1–9) sites occur in the large channels along c (Figs 4a, 5b). The ideal composition is derived by a sum of cation and anion/H₂O parts of the interstitial complex to give: (KCa)³⁺ + (H₂O)₃ = [(KCa) (H₂O)₃]³⁺.

The structural formula on the basis of assigned site-populations (Table 7) is (Na_0.80□_0.20)_Σ1(K_0.37Cs_0.30)_Σ0.67[(Ca_0.25Pb_0.22Sr_0.14)Na_0.25Ln³⁺_0.08]_Σ0.94(Sr_0.98Pb_0.02)_Σ1(Ln³⁺_0.96Ca_0.04)_Σ1(Ti_0.97Nb_0.03)_Σ1.00(Si₈O₂₁)OF_0.50[F_0.40(H₂O)_0.10](H₂O)_3.11, where Ln³⁺_1.04 = (Ce_0.43Nd_0.41Pr_0.08La_0.06Sm_0.03Gd_0.01Er_0.01Ho_0.01)_Σ1.04 and Ce is the dominant lanthanoid. The simplified formula is Na(K,Cs,□)(Ca_,Na,Pb,Sr)SrLn³⁺Ti(Si₈O₂₁)OF(H₂O)₃. On the basis of the structure refinement results and bond-valence calculations, the ideal formula of kalyuzhnyite-(Ce) can be written as the sum of the cation and anion components: (NaKCaSrCeTi)¹³⁺ + [(Si₈O₂₁)OF(H₂O)₃]^13– = NaKCaSrCeTi(Si₈O₂₁)OF(H₂O)₃, Z = 4. The validity of the ideal formula is supported by the good agreement between the total charges for cations in the ideal and empirical formulae: 1⁺ (Na) + 1⁺ (K) + 2⁺ (Ca) + 2⁺ (Sr) + 3⁺ (Ce) + 4⁺ (Ti) = 13⁺versus 1.07⁺ (Na_1.07) + 0.67⁺ (K_0.37Cs_0.30) + 1.64⁺ (Ca_0.37Pb_0.24Sr_0.21) + 2⁺ (Sr_1.00) + 3.12⁺ (Ln_1.04) + 4.08⁺ (Ti_0.97Nb_0.04) = 12.58⁺.

The ideal formulae of kalyuzhnyite-(Ce), above, was based on the sum of the cation and anion components. The ideal formula of kalyuzhnyite-(Ce) (Z = 4) based on ideal compositions of cation and anion sites ^M4Na, ^A(1,2)B(1,2)K, ^M(5–8)Ca, ^M3Sr, ^M2Ce, ^Si(1–8)Si₈, ^O(1–21)O₂₁, ^O22O, ^F+XF and ^W(1–9)(H₂O)₃, can be written as NaKCaSrCeTi(Si₈O₂₁)OF(H₂O)₃.

Kalyuzhnyite-(Ce) has no analogues. The double (Si₈O₂₁)^10– sheet consists of ten-membered rings of SiO₄ tetrahedra. It has unique topology and composition, and its Si–O double sheet has not previously been described in minerals (Hawthorne et al., 2019). We can relate kalyuzhnyite-(Ce) to penkvilksite, Na₂TiSi₄O₁₁⋅2H₂O (Bussen et al., 1974; Merlino et al., 1994; Cadoni and Ferraris, 2008) and tumchaite, Na₂(Zr,Sn)Si₄O₁₁⋅2H₂O (Subbotin et al., 2000), two silicate minerals with a single corrugated sheet of ten-membered rings of SiO₄ tetrahedra; the composition of this single sheet is (Si₄O₁₁)^6–.

The supplementary material for this article can be found at https://doi.org/10.1180/mgm.2023.74.

We thank two anonymous reviewers and Principal Editor Stuart Mills for their help. FCH was supported by Discovery grants from the Natural Sciences and Engineering Research Council of Canada.

The authors declare none.