Occurrence

Almagreraite was found by one of the authors (GF) on the dump of the República Romana mine, Sierra Almagrera, Cuevas del Almanzora, Almería, Andalusia, Spain (37°16’59’’N, 1°45’51’’W). The República Romana mine was an iron mine operated from the 1870s to the beginning of the 20^th Century. The dumps are rich in Mn oxides with some Zn in chalcophanite and hetaerolite. The new mineral is very rare, having been found so far in only one small fragment, the best portion of which is deposited as the holotype. Almagreraite is a secondary mineral intimately associated with well-crystallized blades of malachite on matrix consisting of well-crystallized goethite and massive friable hematite.

Although details about the República Romana mine specifically are very limited, the mines of Sierra Almagrera in general are well known. The mines exploit hydrothermal vein-type deposits containing base-metal sulfides and Pb-Sb-Cu-Ag sulfosalts. They date back to the period of Phoenician colonization and were worked extensively for lead and silver during Roman times. The modern mining period extending from the mid-19^th to the mid-20^th century. Mindat (accessed on 31 October 2025) lists 183 mineral species as occurring in the Sierra Almagrera deposits, including the type minerals ferberite (Niña mine), jarosite (Jaroso Ravine) and zinkosite (Jaroso Ravine). Details regarding the geology and ore mineralization of Sierra Almagrera deposits can be found in Martinez-Frías (Reference Martínez-Frías1991) and Martinez-Frías et al. (Reference Martínez-Frías, Lunar, Rodríguez-Losada, Delgado and Rull2004). Details regarding the mineralogy can be found in Valladares et al. (Reference Valladares, Barral, Le Gaillard and Favreau2021).

Morphology, physical properties and optical properties

Almagreraite crystals are tapering prisms or blades up to ∼0.4 mm in length, ubiquitously multiply twinned (by 60° and 120° rotations on [001]) and commonly forming subparallel to divergent aggregates (Figs 1, 2 and 3). Crystals are elongate on [001]. The curved faces and complex twinning made the determination of crystal forms impossible. The morphology appears to be produced by at least two prism forms and one or two steep pyramids. The Bravais–Friedel–Donnay–Harker principle (Donnay and Harker, Reference Donnay and Harker1937) predicts that the most likely prism forms are {010} and {110}. Combining these with the pyramid form {501}, approximates the observed morphology reasonably well.

Figure 1.

Almagreraite crystals on malachite and hematite; field of view 0.48 mm across; J.M. Johannet photo.

Figure 2.

Almagreraite crystals with malachite blades on hematite; field of view 0.36 mm across; holotype specimen #76458.

Figure 2 Long description

A color photomicrograph shows a cluster of elongated, dark mineral crystals resting within a reddish-brown granular matrix. The crystals are slender, needle-like to blade-shaped, and occur in a radiating arrangement near the center of the image. Several lighter, translucent crystals intersect the darker crystals at different angles. The surrounding matrix has a coarse texture with small mineral grains distributed throughout. The crystals vary in length and orientation, producing a star-like aggregate against the earthy background. The image documents the natural appearance of the mineral assemblage under reflected light at a submillimeter scale.

Figure 3.

Electron micrograph displaying elongated twinned crystal aggregates with sharp edges and layered surfaces.

Figure 3 Long description

A grayscale scanning electron micrograph presents sharply defined elongated crystals on a rough substrate. The crystals exhibit flat faces, straight edges, and prominent longitudinal striations. Several crystals intersect to form a twinned aggregate, producing angular V-shaped and overlapping arrangements. Surface details are clearly visible, including layered growth features and fracture planes along the crystal faces. The surrounding substrate appears granular and uneven, contrasting with the smooth geometric crystal surfaces. A scale bar near the lower edge indicates a magnified view of the crystal morphology and structural relationships within the aggregate.

Crystals are black and opaque with submetallic lustre. The streak is dark grey with a brownish tint. The tenacity is brittle and the fracture is splintery. There are probably two cleavages based on splintery fracture (possibly {100} and {010}). The Mohs hardness is ∼5 based on scratch tests. The density could not be measured because crystals exceed the density of available density fluids and there is insufficient material for physical measurement. The calculated density is 5.174 g·cm^–3 based upon the empirical formula and 5.259 g cm^–3 for the ideal formula. The mineral is insoluble in room-temperature concentrated HCl.

The average index of refraction of almagreraite calculated using the Gladstone–Dale relation Mandarino (Reference Mandarino1976) is 2.61. Despite the very high average index of refraction and the very dark colour of the mineral, some transmitted-light optical properties were obtained using extremely thin slivers, which are brownish orange in colour. In plane-polarized light, crystals are nonpleochroic and, under crossed polars, crystals are seen to be length fast, providing the partial optical orientation X = c. Based on the crystal system, the mineral is biaxial.

In reflected light, the mineral is bluish grey in colour. Bireflectance is moderate and pleochroism is slightly bluish grey to a lighter bluish grey. Weak anisotropy was observed with steely grey rotation tints. Crystals exhibit orange internal reflections. Reflectance data for an acicular grain are reported in Table 1. A SiC standard was used and measurements were made on an 8×8 µm area. The Commission on Ore Mineralogy (COM) standard wavelengths in Table 1 are in bold.

Table 1.

Reflectance data for almagreraite

Table 1 Long description

The table lists minimum and maximum reflectance values for almagreraite at wavelengths from 400 to 700 nanometers. Both the minimum and maximum reflectance generally decrease as wavelength increases. At 400 nanometers, reflectance ranges from 27.6 to 34.0, while at 700 nanometers it ranges from 22.9 to 29.0. Mid-range examples include 500 nanometers at 24.9 to 31.1 and 600 nanometers at 23.7 to 29.8. The spread between minimum and maximum stays fairly consistent, narrowing slightly from about 6.4 at 400 nanometers to about 6.1 at 700 nanometers. Values near 546 and 589 nanometers follow the same downward pattern and can be used as reference points if those wavelengths are treated as standards.

Note: COM standard wavelengths are in bold.

Raman spectroscopy

Raman spectroscopy was conducted on a Horiba XploRA PLUS. Efforts to record spectra using a 532 nm laser were unsuccessful because of the sensitivity of almagreraite to that wavelength even under minimal power. The spectrum was successfully recorded using a 785 nm diode laser, 2 mW power, 200 μm slit and 1800 gr/mm diffraction grating and a 100× (0.9 NA) objective. The spectrum was featureless between 2000 and 800, so only the spectrum from 800 to 60 cm^–1 is shown in Fig. 4.

Figure 4.

Raman spectrum of almagreraite recorded using a 785 nm laser.

Figure 4 Long description

Raman spectrum of almagreraite recorded using a 785 nm laser. A line graph shows a single red spectrum trace. The x-axis is labeled Wavenumber (cm superscript minus 1) with labeled ticks at 800, 700, 600, 500, 400, 300, 200, 100 and 0. The y-axis is labeled Intensity, with no units shown. Peak labels are printed next to local maxima on the trace.

Previous Raman studies of manganese oxides with structures based on linkages of edge-sharing Mn⁴⁺O₆ octahedra (Post et al., Reference Post, McKeown and Heaney2020; Post et al., Reference Post, McKeown and Heaney2021) suggest that the prominent bands in the almagreraite spectrum at 681 and 649 cm^–1 are probably assignable to Mn⁴⁺–O stretching modes, thus confirming the tetravalent state of Mn in almagreraite. The band at 547 cm^–1 may correspond to vibrations of the short Cu–O bond of 1.949 Å (Siidra et al., Reference Siidra, Vladimirova, Tsirlin, Chukanov and Ugolkov2020). The assignment of modes to the other Raman bands in the spectrum is not straightforward.

Chemical composition

Electron probe microanalyses (EPMA; 6 points) were performed at Caltech on a JXA-iHP200F electron probe in WDS mode. Analytical conditions were 15 kV accelerating voltage, 20 nA beam current and 1 μm beam diameter. No other elements were detected by EDS or WDS. Manganese is assigned 4+ valence based on bond-valence sums for the Mn sites and by analogy with the isostructural compound Co²⁺₂Mn⁴⁺₃O₈ (see the X-ray crystallography section below). Analytical data are given in Table 2.

Table 2.

Analytical data (in wt.%) for almagreraite

Table 2 Long description

The table reports average composition in weight percent for almagreraite, listing each constituent with its observed range, standard deviation, and the reference material used for calibration. Manganese oxide is the major component with a mean of 63.02 wt percent and a range from 62.70 to 63.56, showing low variability with a standard deviation of 0.29. Copper oxide averages 17.60 wt percent with a range from 17.43 to 17.72 and a standard deviation of 0.11. Zinc oxide is similar at 17.54 wt percent, ranging from 17.36 to 17.66 with a standard deviation of 0.12. Alumina is minor at 0.59 wt percent, ranging from 0.56 to 0.63 with a standard deviation of 0.03. The reported total is 98.75 wt percent, indicating the analysis does not sum to a full 100 percent, which may reflect unmeasured components or analytical uncertainty.

S.D. – standard deviation

The empirical formula based on O = 8 atoms per formula unit is (Mn⁴⁺_3.05Cu²⁺_0.93Zn_0.91Al_0.05)_Σ4.94O₈. The empirical formula derived using OccQP (see below) is (Cu²⁺_0.93Mn⁴⁺_0.05)_Σ0.98(Zn_0.91Mn⁴⁺_0.04)_Σ0.95(Mn⁴⁺_1.99Al_0.01)_Σ2.00(Mn⁴⁺_0.97Al_0.03)_Σ1.00O₈. For the simplified and ideal formulas, the two similar Mn sites are combined. The simplified formula is (Cu²⁺,Mn⁴⁺)(Zn,Mn⁴⁺)(Mn⁴⁺,Al)₃O₈ and the ideal formula is Cu²⁺ZnMn⁴⁺₃O₈, which requires CuO 18.86, ZnO 19.30, MnO₂ 61.84, total 100 wt.%.

X-ray crystallography

Powder X-ray diffraction (PXRD) data were recorded using a Rigaku R-Axis Rapid II curved imaging plate microdiffractometer with monochromatized MoKα radiation. A Gandolfi-like motion on the φ and ω axes was used to randomize the sample. Observed d values and intensities were derived by profile fitting using JADE Pro software (Materials Data, Inc.). The powder data are presented in Supplementary Table S1 (see below). The unit-cell parameters refined from the powder data using JADE Pro with whole pattern fitting are (space group Pmn2₁) a = 5.727(2), b = 4.976(2), c = 9.339(4) Å and V = 266.14(18) ų.

The single-crystal structure data were collected at room temperature using the same diffractometer and radiation noted above. The selection of a crystal suitable for structure study was very challenging because crystals are ubiquitously multiply twinned and generally occur in subparallel intergrowths. The twinning is by 60° and 120° rotations on [001]; however, the crystal chosen for data collection consisted of only two twin components. The small size of suitable crystals and the difficulty in the integration of reflections significantly limited the number of independent reflections, providing a poor data to parameter ratio (Alert level A in the checkcif).

The TwinSolve program within the Rigaku CrystalClear software package was used for processing of HKLF 5 structure data, including the application of an empirical multi-scan absorption correction using ABSCOR (Higashi, Reference Higashi2001). The |E ²–1| statistics indicated the structure to be noncentrosymmetric and the systematic absences were consistent with the space group Pmn2₁. Using non-overlapping reflections from one twin individual, the structure was solved by direct methods using SIR2011 (Burla et al., Reference Burla, Caliandro, Camalli, Carrozzini, Cascarano, Giacovazzo, Mallamo, Mazzone, Polidori and Spagna2012) and was determined to be identical to that of synthetic Co₂Mn₃O₈ (Riou and Lecerf, Reference Riou and Lecerf1975). The atom coordinates were transformed to correspond to those reported for that structure. SHELXL-2016 (Sheldrick, Reference Sheldrick2015) was used for the refinement of the structure. On crystal-chemical grounds, Mn was assigned to the octahedrally coordinated Mn1 and Mn2 sites, Cu was assigned to the elongated octahedrally coordinated Co1 site (labelled Cu) and Zn was assigned to the tetrahedrally coordinated Co2 site (labelled Zn). These sites all refined to somewhat less than full occupancy. In the final refinement cycles, all sites were successfully refined with anisotropic displacement parameters. Data collection and refinement details are given in Table 3, atom coordinates and displacement parameters in Table 4, selected bond distances in Table 5 and a bond-valence analysis in Table 6. The crystallographic information file (cif) has been deposited and is available in the Supplementary material (see below).

Table 3.

Data collection and structure refinement details for almagreraite

Table 3 Long description

The table summarizes single crystal X ray diffraction data collection conditions and structure refinement results for almagreraite. Data were collected on a Rigaku R Axis Rapid II using molybdenum K alpha radiation at 293 K. The refined composition is Mn2.86 Cu0.94 Zn0.90 O8 in space group Pmn21, with unit cell a 5.7298 angstrom, b 4.9715 angstrom, c 9.3479 angstrom, volume 266.28 cubic angstrom, and Z 2. The crystal measured 150 by 40 by 30 micrometers, with absorption coefficient 14.282 per millimeter and density 5.031 grams per cubic centimeter. The theta range was 4.10 to 25.01 degrees with index limits h 0 to 6, k 0 to 5, and l 0 to 11, yielding 522 unique reflections and 423 above two sigma intensity. Completeness to the maximum theta was 97.4 percent. Refinement used full matrix least squares on F squared with 75 parameters and no restraints, giving goodness of fit 1.032 and final R1 0.0297 and weighted R2 0.1065 for observed data, rising to R1 0.0404 and weighted R2 0.1103 for all data. Additional indicators include absolute structure parameter 0.22 with uncertainty 0.09, extinction coefficient 0.15 with uncertainty 0.07, and residual electron density peaks of plus 1.07 and minus 0.80 electrons per cubic angstrom.

* Note that the processing of twinned data using TwinSolve provides an averaged dataset, so it is not possible to give R _int.

GoF = S = {Σ[w(F _o²–F _c²)²]/(n–p)}^1/2. R ₁ = Σ||F _o|–|F _c||/Σ|F _o|. wR ₂ = {Σ[w(F _o²–F _c²)²]/Σ[w(F _o²)²]}^1/2; w = 1/[σ²(F _o²)+(aP)²+bP] where a is 0.0769, b is 0 and P is [2F _c²+Max(F _o²,0)]/3.

Table 4.

Atom coordinates and displacement parameters (Ų) for almagreraite

Table 4 Long description

The table lists fractional atomic coordinates along the a, b, and c directions and atomic displacement parameters for Mn1, Mn2, Cu, Zn, and oxygen sites O1 through O6 in almagreraite. The metal sites have closely clustered equivalent displacement values, about 0.007 to 0.008, indicating similarly low atomic motion or disorder among Mn, Cu, and Zn. Oxygen sites generally show higher equivalent displacement values, ranging from about 0.010 to 0.018, with O6 the highest and O4 among the lowest. Several atoms lie on special coordinate values such as zero or one half for one coordinate, including Mn2 and Zn at one half in the a direction and Cu at zero in the a direction. The largest anisotropic component among the listed values occurs for O3 in the c-direction term, consistent with oxygen showing more directional displacement than the metals. Some cross terms are reported as zero for symmetry-restricted sites, so comparisons of those terms should consider site constraints. Values in parentheses indicate measurement uncertainty, and refined site occupancies for the metal positions are slightly below full occupancy, which may influence the displacement parameters.

* Refined occupancies: Mn1 = 0.95(4); Mn2 = 0.96(4); Cu = 0.93(4); Zn = 0.90(3)

Table 5.

Selected bond distances (Å) for almagreraite

Table 5 Long description

The table lists selected metal–oxygen bond lengths in angstroms for Mn1, Mn2, Cu, and Zn sites in almagreraite. Mn1–O distances range from 1.842(17) to 2.007(19), with an average Mn1–O distance of 1.914. Mn2–O distances range from 1.87(3) to 1.947(17), with an average Mn2–O distance of 1.916. Zn–O distances cluster tightly from 1.951(15) to 1.96(3), giving an average Zn–O distance of 1.956. Cu–O distances show a clear split between shorter bonds, including Cu–O1 twice at 1.949(18) and Cu–O2 twice at 2.049(15), and longer bonds at 2.19(3) and 2.43(2); the listed mean short Cu–O is 1.999 and mean long Cu–O is 2.31. Values marked “times two” indicate two equivalent bonds, and numbers in parentheses indicate measurement uncertainty in the last digits.

Table 6.

Bond valence analysis for almagreraite. Values are expressed in valence units*

Table 6 Long description

The table reports bond-valence contributions from Mn1, Mn2, Cu, and Zn to six oxygen sites, with a per-oxygen total at the end of each row and per-cation totals in the last row. Oxygen-site totals are close to two valence units across all sites, ranging from 1.80 at O3 to 2.11 at O4, with O1 at 1.86, O2 at 2.03, O5 at 1.99, and O6 at 2.05. Mn1 contributes to every oxygen site and is strongest at O3 and O6, each listed as 0.78 with a twofold multiplicity, giving Mn1 a total of 3.92. Mn2 contributes mainly to O1, O2, O4, and O5, with its largest single listed value at O5 of 0.73, totaling 3.87. Cu contributes to O1, O2, O3, and O5 and totals 2.04, while Zn contributes to O2, O4, and O6 and totals 1.98. Some entries indicate repeated bonds by multiplicity, so individual cell values may represent more than one equivalent interaction, and the parameters assume ideal full cation occupancies.

* Multiplicity is indicated by ×→↓. Bond-valence parameters are from Gagné and Hawthorne (Reference Gagné and Hawthorne2015) and are based on full cation occupancies by the ideal constituents.

Description of the structure

In the structure of almagreraite, Mn⁴⁺O₆ octahedra share edges forming an octahedral sheet parallel to {001}. One in every four octahedra in the sheet is vacant (Fig. 5), providing the sheet formula [Mn⁴⁺₃O₈]^4-. Successive sheets are connected via bonds to Cu²⁺ and Zn²⁺ located in a layer between the sheets. The Cu²⁺ bonds to three O atoms in each adjacent sheet forming a Jahn-Teller distorted Cu²⁺O₆ octahedron. The Zn bonds to three O atoms in one sheet and one atom in the other forming a ZnO₄ tetrahedron. Each of the vacant octahedra in the sheet has a Cu²⁺O₆ octahedron on one side and a ZnO₄ tetrahedron on the other. Thus, the structure can be considered as alternating sequence of [Mn⁴⁺₃O₈] octahedral sheets and layers containing Cu²⁺O₆ octahedra and ZnO₄ tetrahedra (Figs 5 and 6).

Figure 5.

Octahedral sheets (top) and octahedral-tetrahedral layers (bottom) in the structures of almagreraite and kamiokite viewed along c. Unit-cell outlines are shown with dashed lines. The structure drawings were created using ATOMS, version 6.5 (Shape Software, Kingsport, Tennessee, USA).

Figure 5 Long description

The image consists of two distinct parts, each illustrating the structures of almagreraite and nolanite. The top section shows octahedral sheets, with the left side labeled as almagreraite and the right side as nolanite. In the almagreraite structure, the octahedra are colored purple and two are labeled Mn1 and Mn2, with dashed lines indicating unit-cell outlines. In the nolanite structure, the octahedra are similarly colored, with labels V1 and V2 and dashed lines marking the unit-cell outlines. The bottom section displays octahedral-tetrahedral layers. For almagreraite, green octahedra labeled Zn and Cu are interspersed with orange tetrahedra. The nolanite structure features green octahedra labeled V2 and V3, surrounded by orange tetrahedra. Dashed lines again indicate unit-cell outlines. The arrangement highlights the alternating sequence of octahedral sheets and layers containing octahedra and tetrahedra in both structures.

Figure 6.

Clinographic projections of the structures of almagreraite and kamiokite. The structure drawings were created using ATOMS, version 6.5 (Shape Software, Kingsport, Tennessee, USA).

Figure 6 Long description

The image consists of two clinographic projections side by side, illustrating the structures of almagreraite on the left and kamiokite on the right. In the almagreraite structure, purple octahedra are arranged in layers, with green Mn4 plus O6 octahedra forming sheets parallel to the base. Some octahedra are vacant and these sheets are interspersed with layers containing orange Cu2 plus O6 octahedra and ZnO4 tetrahedra. The Cu2 plus octahedra are bonded to three oxygen atoms in adjacent sheets, while the Zn tetrahedra bond to three oxygen atoms in one sheet and one in the other. The kamiokite structure on the right similarly features purple octahedra, but with green Fe2 plus and orange MoO4 tetrahedra interspersed between the layers. The Fe1 and Fe2 labels indicate different positions within the structure. Both structures are shown with axes labeled a, b and c, indicating orientation. The labels for each structure are placed at the bottom of their respective projections.

The structure has distinct similarities to the structure of minerals of the nolanite supergroup (Chukanov et al., Reference Chukanov, Gridchina, Rastsvetaeva, Zubkova and Pekov2025) and most notably the kamiokite group. The structure of kamiokite, Fe²⁺₂Mo⁴⁺₃O₈ (Kanazawa and Sasaki, Reference Kanazawa and Sasaki1986), also has edge-sharing octahedral sheets in which one in every four octahedral sites is vacant, providing the sheet formula [Mo⁴⁺₃O₈]^4–. The sheets are linked through Fe²⁺O₆ octahedra and Fe²⁺O₄ tetrahedra that are similarly placed above and below the vacant octahedra in the sheets. Both structures are based on the same close-packed anion framework with a ABCBA sequence; however, the arrangement of vacant octahedrally coordinated sites differs, and with it the arrangement (and linkage) of octahedra and tetrahedra in the intervening layer (Fig. 5).

As noted above, the cation sites are exhibited lower than ideal occupancies for their ideal constituents. The EMPA also provided less than 5 cations per formula unit: (Mn⁴⁺_3.05Cu²⁺_0.93Zn_0.91Al_0.05)_Σ4.94O₈. To some extent, the deficiencies in the occupancies can be explained by the lower cation content indicated by the EPMA and by substitution of lower atomic number cations, e.g. Mn for Cu. Even so, the low occupancies cannot be completely accounted for in this way and we suspect that, to some extent, they are artefacts of the refinement (note poor data to parameter ratio). Based on bond-length and bond-valence considerations, it is most likely that Al is accommodated at the Mn sites, while the excess Mn in the EPMA is presumably accommodated at the Cu and Zn sites. Assuming that all Cu is located at the Cu site, and all Zn is located at the Zn site and Al is accommodated only at the Mn sites, an analysis using OccQP (Wright et al., Reference Wright, Foley and Hughes2000) found the best fit of the EPMA to the cation sites (Mn1, Mn2, Cu, Zn) to be as follows: Mn_0.993Al_0.007, Mn_0.965Al_0.035, Cu_0.930Mn_0.051 and Zn_0.906Mn_0.044. The site scattering values compare as follows (assigned/refined): Mn1(99.66/95.00), Mn2(49.17/48.00), Cu(56.50/53.94), Zn(56.55/54.00).