Geological Setting

Berlin project is located in the Colombian Andean orogenic system (Figure 1), specifically in the Cajamarca-Valdivia terrane in the Central Continental Sub-plate realm (^{Cediel et al., 2003}), or in the Tahami Terrain in the sense of ^{Restrepo and Toussaint (2020)}. The Berlin project is located in the area of influence of the Palestine Fault system, which is associated with a crustal weakness zone ranging from 0.5 to 30 km wide (^{Page, 1986}). Palestine fault is considered to have had a dextral displacement during the Neogene to Paleogene (^{Feininger, 1970}).

Figure 1 Left, location of the Berlin project in the Caldas Department. Right, a generalized geological map of the study area. 

The basement consists of metamorphic rocks of the Cajamarca Group, which has also been named Cajamarca Series (^{Nelson, 1962}) or Complex (^{Maya and González, 1995}), that is considered of Early Paleozoic age (^{Bürgl, 1966}).

A sedimentary sequence that corresponds to a transitional continental-marine-lagoonal environment is found overlying discordantly the Cajamarca Group (Figure 2).

Figure 2 Stratigraphic column Berlin project and drill core description. 

Four successions were recognized in the study area. Succession A: a terrigenous lithology (K1-Sj) of basal conglomerates interbedded with sandstones recorded the beginning of a back-arc extensional event at 150 Ma (^{Zapata et al., 2019}). The clasts present within the conglomerate layers consist mainly of mafic and intermediate igneous fragments, metamorphic fragments, and milky white quartz. The upper part of this succession corresponds to a massive silicified, medium to coarse grain size, with poorly sorted and subangular grains of sandstone. The average thickness of the entire succession A is 36 m (^{Naranjo, 1983}). Succession B: presents a gradation contact with succession A and corresponds to a fossiliferous calcareous mudstone, with the presence of well-preserved bivalves and gastropods (Figure 3A), deposited during basin deepening and the change to a marine environment that took place until Albian (100 Ma). The upper part of this succession is defined by a major erosional surface where the different grain size is easily observable between succession B and C (Figure 3B). This disconformity marked not just the change to a compressional tectonic event but the lower limit of the mineralization. Succession C: a muddy wackestone with the presence of reworked fossil fragments, displaying wavy lamination and a high content of authigenic apatite and bitumen (Figure 3C). This is consistent with either the increasing water energy or probably the relative shallowing of the basin (^{Edelman-Furstenberg, 2009}; ^{Zapata et al., 2019}). This succession constitutes the mineralized level, which varies in thickness from 1 to 3.5 m, with a maximum thickness of 9.2 m (in a drillhole) and 6.7 m (in a trench). Succession D: the upper succession (b6-Be) corresponds to a deformed black carbonaceous shale with bioclast (Figure 3D), which may locally reach 600 m in thickness. This was deposited in a lagoonal environment and is related to the enclosing of the marine basin caused by the accretion of the Romeral terrane and the uplifting of the Cajamarca-Valdivia terrain during Aptian to Albian (^{Cediel et al., 2003}) or even Albian to Campanian (Zapata et al., 2019). All data were confirmed on fossil analysis, according to personal communication by Etayo-Serna, in ^{Cáceres (2012)}.

Figure 3 Cross-polarized transmitted light (A, B, and C). Plane-polarized transmitted light (D). A. Photomicrograph showing fossiliferous mudstone with bivalves and gastropods (succession B). B. Photomicrograph illustrating erosive surface (disconformity). Contact between successions B and C. C. Photomicrograph shows the muddy wackestone with reworked fossil fragments (succession C). D. Photomicrograph illustrating bioclast within succession D. 

A set of dikes and sills of andesitic to dacitic composition are found injecting the metamorphic and sedimentary rocks described above. Leucocratic (alaskites) and mesocratic (ranging from tonalite to granodiorite) plutons are recognized as the Samaná Igneous Complex, which is considered of Aptian-Albian age (119±10 Ma - K/Ar in hornblende) according to ^{Barrero and Vesga (1976)}.

Sample Preparation and Analytical Methods

Six hundred fifty-six samples were taken from succession B, C, and D, during the drilling program of the U3O8 Corp during the resource estimation for the Berlin project (^{De Klerk et al., 2013}). Nine polished thin sections of the succession C (mineralized layer) from boreholes DDB11, DDB16, DDB17, and DDB18 were used for petrographic analysis using reflected and transmitted light and scanning electron microscopy to establish the mineral assemblage. Sampling localities were selected considering a locality where the sequence is not inverted or intensively deformed (Figure 1, Table 1). Polished thin sections were also used for electron probe microanalysis (EPMA) with a JEOL JXA-8230 with five wavelength-dispersive spectrometers (WDS). EPMA using a JEOL JXA-8230 equipped with five wavelength-dispersive spectrometers (WDS). Acquisitions were obtained with a 15-kV accelerating voltage, 10-nA beam current, and 5 to 10 urn diameter defocused beam. Typical counting times were 10 s for major elements, 20 s for minor elements, and 30 to 40 s for trace elements.

Uraninites were dated by laser ablation using the polished thin section DDB13 by in situ U-Pb isotope ratios using a ThermoFinnigan Neptune high-resolution multicollector ICP-MS (HR-MC-ICP-MS), following the method of ^{Chipley et al. (2007)}.

Isotopic compositions of carbon and oxygen were measured in the carbonates of the borehole DDB16 using a Finnigan MAT 252 isotope ratio mass spectrometer (IRMS). δ ¹³C results are relative to the PDB standard, while δ ¹⁸O results are relative to the Vienna Standard Mean Ocean Water (V-SMOW). All analyzes were conducted at Queen's Facility for Isotope Research Lab.

Paragenetic sequence

The paragenetic sequence recognized in the lithology that constitutes the mineralized level (succession C) is depicted in Table 2, emphasizing the diagenetic process.

Table 2 Generalized paragenesis for the Berlin project. 

Quartz and muscovite conform to the detrital minerals that remained in the samples. Calcite Sage 1 occurs as fossil replacement and as cement formed during burial.

Petrographic analysis reveals textural and structural features and the mineral composition within the rocks that contain the mineralization. The occurrence of detrital quartz provides the primary porosity that allows diagenetic fluids to percolate and precipitate calcite (stage 2), authigenic apatite, and bitumen (Figure 4A). Also, many samples revealed intense fracturing of the clastic quartz, which can be associated with tectonic efforts caused by the inversion of the extensional regime during the collision of the oceanic lithosphere which started in the late Mesozoic (^{Cediel et al., 2003}). Detrital muscovite is observed with alteration to the edge (Figure 4B).

Figure 4 Photomicrographs of the textual microstructural features Berlin project. A. Fracture quartz grains in an interstitial matrix with calcite (stage 2), apatite, and bitumen. B. Detrital muscovite with alteration to the edge. Quartz (Qtz), Calcite Stage 2 (Cal 2), Apatite (Ap), Bitumen (BT), Muscovite (Ms) and Chernykhite (Chk) (adapted and modified after ^{Pimiento, 2011}). 

Electron microprobe and SEM disclose the diagenesis process in detail. Authigenic apatite (Figure 5A) and bitumen (Figures 5B, 5E, and 5F), were incorporated into the mineralized layer, along with uraninite and chernykhite (Ba-V-rich mica). Electron probe microanalysis allowed identifying the replacement of muscovite by chernykhite (Figures 5C and 5D). Bitumen is observed filling pore spaces between the apatite (Figure 5E) and calcite stage 2 (Figure 5F). Authigenic apatite is zoned (Figure 5G), showing an increase in Y and U from core to rim. Pyrite, Sphalerite (Figure 5H), and other sulfides, as well as bitumen (Figure 5H), fill the spaces formed between the apatite, calcite stage 2, and detrital quartz. Late diagenesis (3) is characterized by the presence of Ni-As sulfides, which replace sphalerite (Figure 5I), and a second generation of calcite, which occurs as grooves and veinlets that cross-cut apatite and bitumen. The formation of this calcite can be attributed to the dissolution and re-precipitation processes of carbonate.

Figure 5 A. Back-scattered electron image of textural relationships between uraninite and apatite. B. Back-scattered electron image of uraninite within bitumen being cross-cut by calcite. C. Back-scattered electron image of uraninite and chernykhite. D. Back-scattered electron image of muscovite replaced by chernykhite. E. Plane-polarized reflected light image of bitumen filling spaces between the apatite. F. Plane-polarized reflected light image of bitumen and organic matter filling spaces between calcites. G. Back-scattered electron image of apatite uranium-yttrium zonation. H. Plane-polarized reflected light image of sphalerite following the pattern of bitumen. I. Sphalerite was replaced by Ni-As sulfides. U: Uraninite; Ap: Apatite; BT: Bitumen; M.O: Organic matter; Cal: Calcite; Chk: Chernykhite; Ms: Muscovite; ZnS: Sphalerite. 

Mineral Chemistry

Uraninite grains were analyzed by electron microprobe analysis and scanning electron microscopy (Figure 6). Uraninite contains up to 90 wt% U and other elements, such as rare earth elements (REE), Ca, and radiogenic Pb (^{Alexandre and Kyser, 2005}). Electron back-scatter images show that the average size for uraninite is less than 10 um, so some results obtained using electron microprobe analysis reveal "contamination" from surrounding material due to the small size of uraninite, which will complicate isotopic dating. The chemical composition of uraninite is quite variable, with UO₂ from 81.47 to 94.44 wt% (Table 3).

Figure 6 Back-scattered electron images of uraninite grains. 

Chemical compositions of the uraninite indicate negative correlations between UO₂ and SiO₂, V₂O₃, Cr₂O₃, MnO, P₂O₅, and FeO, and positive correlations between UO₂ and ThO₂ and PbO (Figure 7).

Figure 7 Correlations between UO2 and other oxides (adapted and modified after ^{Cáceres, 2012}). 

These correlations that the oxides that positively vary with UO₂ correspond to the original mineralogy of the uraninite present, whereas those that are negatively correlated correspond to contamination, which would occur if uraninite particles were small enough (<10 µm) so that the microprobe beam diameter (5 µm) encompassed surrounding elements of other material as well as the uraninite. Uraninite may incorporate Th, Ca, REE, and Y during its initial formation, and Pb is added later through decay. It is assumed that the uraninite does not initially contain Si, Fe, P, Mn, Cr, and V, but rather that these elements were introduced by contamination of the microprobe beam. Results of microprobe analysis reveal that the concentration of contaminants is negatively correlated with UO₂ so pure uraninite is when contaminant concentration is zero. By extrapolating the concentration of contaminants to zero, we calculated the best estimation for uraninite composition (Figure 8).

Figure 8 Negative correlations between oxides introduced by contamination (SiO₂, V₂O₃, Cr₂O₃, MnO, P₂O₅ and FeO) and some oxides initially in uraninite: A. UO₂, B. PbO and C. Y₂O₃ (adapted and modified after ^{Cáceres, 2012}). 

The estimated age of the Berlin project is 67±15 Ma. This age calculation is based on the extrapolated chemical composition of uraninite, and the assumption that all Pb contained in the analysis of the uraninite is radiogenic (^{Bowles, 1990}). Pb coming from Th is negligible, as Th contents are low. The amount of ²⁰⁶Pb (atom %wt) was calculated from the PbO content, and the amount of ²³⁸U (atom %wt) was calculated from the total UO₂ content, as determined from the microprobe analysis best-estimated values of uraninite (UO₂ = 94.3 wt%; U = 94.3*1.134¥ = 83.16, and PbO = 0.8 wt%; Pb = 0.8* 0.928§ = 0.76). Age calculations were carried out using the empirical formula of ^{Ranchin (1968)}, which is the best approach for ages less than 200 Ma (^{Alexandre and Kyser, 2005}).

Age estimates based on correlations

The age of formation of the uraninite in the Berlin project can also be deciphered by extrapolating the chemical U-Pb ages to the time when the content of the contaminating elements is zero; the inferred age of 75±15 Ma, estimated from the intercept of the total contaminant elements (Figure 9), is similar to the age of 67 Ma obtained by the best-estimated U-Pb values from the uraninite.

Figure 9 Combined concentration of contaminating uraninite elements (SiO₂, V₂O₃, Cr₂O₃, MnO, P₂O₅ and FeO contents) as a function of the estimated chemical U-Pb age (adapted and modified after ^{Cáceres, 2012}). 

U/Pb geochronology

U-Pb dating on uraninites was performed on one polished thin section by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS). According to ^{Cáceres (2012)}, uranium mineralization occurred at 53±11 Ma based on ²⁰⁴Pb/²³⁸U data obtained by laser ablation of two uraninites (Figures 10 and 11), which coincides with the calculated chemical U/Pb ages (67±15 Ma).

Figure 10 Plane-polarized reflected light image showing uraninite chosen for ablation. 

Figure 11 U-Pb inverse Concordia diagram from isotopic analysis by LA-ICPM-MS of uraninites from Berlin project. 

Stable Isotopes

Calcite from fossils and the matrix were sampled from a mineralized section of drill core DDB16. Around 3 mg were taken using a scalpel under a binocular microscope. The matrix calcite crystals were very small (<0.5 mm); therefore, it is possible that samples contained some contamination from the surrounding diagenetic apatite and organic matter. Calcite samples were dissolved in 100% phosphoric acid at 25°C, and the CO₂ gas released was analyzed in an isotope ratio mass spectrometer for carbon and oxygen isotopic compositions. The values obtained are reported in units per mil (%o). Calcite within the fossils showed δ¹³C and δ ¹⁸O values of -0.4 and +22.7, respectively, whereas calcite in the matrix, showed the following values: δ ¹³C = -1.8 and δ ¹⁸O = +18.1 (Figure 12). These stable isotopic data for calcite within the fossils correspond to calcite formed between 10 and 25°C in a marine environment, whereas those values of calcite in the matrix correspond to calcite formed at 40°C in a fresh-water environment (^{Keith and Weber, 1964}; ^{Kim and O'Neil, 1997}) during diagenesis. The high δ ¹³C value of calcite in the matrix also indicates a minimal contribution from organic matter.

Figure 12 Carbon and oxygen isotopic compositions of calcite from fossils and calcite from the matrix in the mineralized sample from drill hole DDB16. Samples are represented by yellow squares. The field for carbonates from magmatic CO₂ is from ^{Des Marais and Moore (1984)} and the field for marine carbonates at different temperatures is from ^{Kim and O'Neil (1997)}.