The data of this study were acquired from Burkina Faso national geological survey. It consists of twelve drill cores from “Projet Minier Gaoua” (PMG) studies between 1982 and 1985. The drill cores were logged and one hundred and ten representative samples were taken in regards with the previous PMG geochemical copper anomaly contour, lithology, hydrothermal variation and mineralization type. The samples were described, labelled and put in some bags for further several laboratories studies.

Polished thin sections were done in the laboratory of Geosciences and Environment, in the department of Earth Sciences of University Joseph KI-ZERBO and Burkina Faso national geological survey (BUMIGEB). We used a polarizing and metallographic microscope for sample description and analyses.

Twenty-nine rock samples were analyzed for whole rock geochemistry. Sixteen samples were crushed in Bigs Global laboratory in Burkina Faso, before being sent eleven to Bureau Veritas laboratory in Canada and five in “Laboratoire Magmas et Volcans” (France) for whole rock analysis (Table 1 and Table 2).

Table 1. Major and traces elements of geochemical results of whole rock analyses.

Table 2. Suite of Table1.

In the two laboratories, major elements were analyzed by inductively coupled plasma-atomic emission spectrometry (ICP-AES), whereas traces elements data were obtained by inductively coupled plasma mass spectrometry (ICP-MS).

These data were supplemented by thirteen whole rock analyses results from [27] and [42] studies in the study area. All of these samples were analyzed in ALS Chemex laboratories in Spain and Canada by ICP-AES for major elements and ICP-MS for trace elements.

Four rock samples containing pyrite and chalcopyrite were sent to Actlabs in Vancouver (Canada) for extraction and analyses. Dried samples are weighted into tin cups for 34S analysis. Approximately 0.3 mg of sample is used for 34S analysis, with 3 mg of niobium pentoxide added to each sample to ensure complete sample combustion. Samples are loaded into a Fisons Instruments elemental to be flash combusted at 980° C. Released gases are carried by ultrapure helium through the analyzer, then separated by gas chromatography. Clean SO2 gas is carried into the Mat 253, Thermo Scientific, IRMS for analysis. Data is corrected and normalized using three international standards, IAEA SO6, IAEA SO5, NBS 127, and two calibrated internal standards that bracket the samples. Standards are at the beginning and end of every run. The δ34S values are reported relative to the Vienna Cañon Diablo Troilite (VCDT) standard.

Microprobe analyses by CAMECA SX100 were carried out at the “Magmas et Volcan” Laboratory of the Clermont-Ferrand University (France). They involved a total of four core samples from the study area. These analyses aim to determine the different mineral phases of these rocks in order to determine their petrogenetic characteristics and their paragenetic sequence. Epidotes, chlorites, feldspars, carbonates and amphiboles are the hydrothermal alteration minerals which have been analyzed.

Preliminary factor analyses were conducted to define factors related to hydrothermal alteration and mineralization using SPSS 16.0 software to analyze eleven rock samples. The trace elements used for this analysis are copper (Cu), gold (Au), silver (Ag), molybdenum (Mo), selenium (Se), antimony (Sb), nickel (Ni), lead (Pb), zinc (Zn), mercury (Hg), barium (Ba), arsenic (as), strontium (Sr), rubidium (Rb), cesium (Cs), bismuth (Bi). The major elements are represented by Fe₂O₃, TiO₂, loss on ignition (LOI), total carbon (TOT/C) and total sulfur (TOT/S). Alteration indexes were also used. These are those of [45] (AI, SI), [46] (CCPI) and [47] (IP or potential index). The AI generally indicates chlorite-sericite alteration, the CCPI indicates chlorite-carbonate-pyrite alteration and SI indicates the silicification index.

Hydrothermal alteration is an important factor in the genesis of hydrothermal deposits in magmatic and metamorphic contexts. In the Gongondy and Diénéméra deposits, several types of alteration have been identified on the basis of petrographic observations.

4.1.1. Sericitic Alteration

Sericitic alteration affected spherulitic dacite (Gongondy) and, to a lesser extent, Quartz porphyry microdiorite (Diénéméra) (Figure 3(A)). This alteration is marked by abundant sericite associated with secondary quartz, carbonates and chlorite in minute quantities, accompanied by deposition of abundant metallic opaque crystals, either in fine disseminations or fractured coarse-grains, cemented by quartz and carbonates. Sericite occurs in fine lamellae (40 wt% - 50 wt%), resulting from partial or total destruction of the feldspars. In facies affected by these types of alteration, sericite is also found in the form of veinlets deposited around other veinlets, particularly early quartz-carbonate veinlets. Greenish chlorite is present in small quantities (5 wt% - 15wt%). It is practically associated with disseminated sulfides, which represent around 15-20wt% of the rock minerals (Figure 4(A)).

In dacite and Quartz porphyry microdiorite, a second type of sericite alteration occurs late or coexists with the first. This latter type of alteration is dominated by white micas (50 wt% - 55 wt%) associated with accessory carbonates (5 wt% - 10 wt%). Rocks affected by these types of alteration take on a light green or yellowish color (Figure 3(B)). This type of hydrothermal alteration can also be found in fractures or veinlets (Figure 3(C)). Micas occur in more or less oriented lamellae. Sulfides (5wt%) are rare, fine (80 µm) and automorphs, disseminated in the rock, moulded by phyllites consisting mainly of white mica (Figure 4(B)).

4.1.2. Propylitic Alteration

Propylitic alteration affects quartz porphyry microdiorite, spherulitic microdiorite, intersertal microdiorite and quartz diorite. At a macroscopic scale, altered rocks take on a light green color (Figure 3(D)). At a microscopic scale, it is characterized by an epidote-chlorite-sericite assemblage, with varying content of carbonates, quartz and albite (Figure 4(C)). Epidote (15 wt% - 45 wt%) is the result of the destabilization of plagioclases and amphiboles. Chlorite (10 wt% - 35 wt%) is largely derived from ferromagnesian minerals (biotite, amphibole) and plagioclases, which are also altered into sericite. Carbonates come from the decomposition of feldspars and amphibole. Slight albitization (5 wt% - 10 wt%) related to saussuritisation of the plagioclases is associated with propylitic alteration in these intermediate rocks (Figure 4(H)).

4.1.3. Phyllitic-Carbonate Alteration

The Phyllitic-carbonate alteration is marked by the predominance of phyllites (chlorite and sericite) (50 wt% - 60 wt%) over carbonates (15 wt% - 20 wt%) and, conversely, by the predominance of carbonates (55 wt% - 60 wt%) over phyllites (30 wt% - 35wt%) for the carbonate-phyllitic type. At Gongondy, where these types of alteration are most pronounced, a certain zonation is observed according to the dominance of one of the three phases. Chlorite dominates at the surface, sericite in the intermediate part and carbonates at deeper zones. Depending on the dominance of a particular mineral, the rock can take on a white, grey or light green color (Figure 3(E)). Sericite (15 wt% - 35 wt%) comes mainly from the partial destabilization of feldspars and is present in fine flakes. Chlorite (10 wt% - 25 wt%) results mainly from the alteration of ferromagnesian minerals (amphiboles, biotite) or, to a lesser extent, feldspars. Carbonates (10 wt% - 60 wt%) come from the destabilization of feldspar or ferromagnesian minerals (amphibole, biotite) (Figure 4(D)).

4.1.4. Chloritic Alteration

Chloritic alteration affects microdiorite porphyry and microgabbro (Figure 3(F)). Microscopically, massive chlorite alteration (40 wt% - 60 wt%) almost completely invades the rocks in question and is associated with sericite (10 wt% - 15 wt%), carbonates (10 wt% - 15 wt%) and quartz (10 wt%). Accessory elements (10 wt%) are represented by opaques (pyrite, rutile, etc.). This chlorite can be distinguished from that observed in the mineralized zones by its bluish hue, suggesting a different composition. Chlorite is the product of the destabilization of feldspar or ferromagnesian minerals (amphibole, biotite) (Figure 4(E)).

4.1.5. Potassic Alteration

Traces of potassic alteration were observed in borehole S16G at Gongondy. It affects a pinkish microgabbro-diorite dyke injected into an ophitic microdiorite. Microscopic evidence of this alteration shows the transformation of amphibole to biotite and the perthitisation of plagioclase (Figure 4(F)). This type of alteration is also suspected in a late intrusion of microdiorite porphyry, which shows a similar reddish color.

4.1.6. Silicification

Silicification of rocks is common in the study area. The alteration is frequently visible in the Diénéméra deposit. It is characterized by quartz veins and veinlets, or silica partially or completely invading quartz microdiorite porphyries, spherulitic microgabbros, microgranodiorites and microgranites. Microscopically, silicification leads to the formation of ribbons of quartz, often polygonal, associated with coarse carbonate and opaque crystals (Figure 4(G)). It also appears in the form of coarse quartz veinlets in quartz microdiorites. It forms the cement for the coarse cracked pyrite crystals.

Figure 3. Macroscopic view of altered rocks from Diénéméra-Gongondy copper deposits. Sericitic alteration 1 affecting microdiorite porphyry at Gongondy (A). Sericite alteration 2 affecting dacitic rocks: Pervasive alteration (B) and alteration saw in veinlet (C). Propylitic alteration affecting Quartz porphyry microdiorite in Diénéméra (D). Phyllito-carbonate alteration affecting microdiorite in Gongondy deposit (E). Chloritic alteration affecting porphyry microdiorite in Gongondy (F).

Figure 4. Microscopic view of altered rocks from Diénéméra-Gongondy copper deposits. Sericitic alteration 1 (A), Sericitic alteration 2 (B), Propylitic alteration (C), Phyllitic-carbonate alteration (D), Chloritic alteration (E), Residual potassic alteration (F), Silicification (G) and Albitization (H). Minerals abbreviations: Amp: Amphibole, Bt: biotite, Cb: carbonate, Chl: chlorite, Ep: epidote, Ms: muscovite, Op: Opaques; Pl: plagioclase; Py: pyrite, Qtz: quartz, Ser: sericite, Sulf: sulfides.

Finally, silicification is associated with shearing in the form of centimetric to decimetric quartz veins.

Macroscopic and microscopic studies of hydrothermal alteration in the study area show that all samples are subject of hydrothermal alteration. Based on the alteration mineral assemblages, several patterns of hydrothermal alterations have been defined. The geochemical analysis of major and trace elements can be used to define the mobility of some elements such as K₂O, Na₂O. The binary diagram of [48] modified by [49] discriminated less altered rocks from the very altered ones (Figure 5).

Figure 5.Binary diagram of [48] modified by [49] discriminating the fresh rocks to altered rocks.

Figure 6. CCPI and AI for different types of alteration observed in the study zone. Phyllitic-carbonate alteration (A) (B), Propylitic alteration (C) (D), chloritic alteration (E) (F), Sericite alteration 1 (I), Sericite alteration 2 (G) (H).

Most of the samples analyzed are altered, but the degree of alteration varies. The ophitic rocks are characterized by an exclusive sodic alteration. The alteration indexes of [45] and [46] have been important to distinguish the different types of hydrothermal alteration. Propylitized rocks according to petrographic observations show a low CCPI (60 to 75) and medium AI (25 to 35) (Figure 6(C) and Figure 6(D)) whereas phyllitic-carbonate altered rocks show a relatively high CCPI (75 to 85) and AI (50 to 60) (Figure 6(A) and Figure 6(B)). These two types of alteration are clearly distinguishable from sericitic and chloritic alteration, which have higher AI and CCPI values respectively. Chloritic alteration in relation with mafic rocks has a low AI like propylitic alteration, with high CCPI between 85 and 100 (Figure 6(E) and Figure 6(F)). Sericite alteration 1, on the other hand, shows an AI (50 to 70) that is similar to that of phyllitic-carbonate alteration, for a low CCPI between 20 to 30 (Figure 6(G)). Sericitic alteration 2 differs from the sericitic alteration 1 by a lower value of CCPI which is less than 30 (Figure 6(H) and Figure 6(I)).

In the ternary diagram of [50], an evidence of sericitization is observed in all rocks with a tendency for chloritization, notably for mafic rocks (Figure 7(A)). However, the diagram of [51] shows a clear segregation of rocks in different alteration fields. Dacitic rocks fall in the sericitic and propylitic fields. Gabbroic rocks are generally observed in the propylitic field only. Intermediate rocks are observed in the propylitic fields dominantly quartz microdiorite porphyry, quartz diorite porphyry, and ophitic microdiorites. Porphyric microdiorites, porphyric diorites and andesites are observed in the K-silicate fields in relation with K-Feldspar alteration (Figure 7(B)).

Figure 7. Hydrothermal alteration diagram using major elements. Ternary diagrams of [50] (A) and [51] (B) which define some fields of hydrothermal alterations.

The chemical analyses of some minerals which are frequent in the alteration zone of Diénéméra-Gongondy porphyry copper (plagioclase, chlorite, epidote, amphibole, carbonate) enabled us to define the chemical composition of these minerals and the consequently determine the type of fluid responsible for the alteration and the deposition of the mineralization. The details of chemical analyses of minerals are found in Table3 and Table4.

4.3.1. Plagioclase

Four samples were used to study the chemical composition of the plagioclase found in the study area. All analyzed points show a high proportion of SiO₂ (66.43 wt% - 71.69 wt%), Al₂O₃ (18.08 wt% - 20.27 wt%), Na₂O (8.81 wt% - 10.91 wt%) and a low proportion of K₂O (0.006 wt% - 0.31 wt%) and CaO (0.07 wt% - 0.60 wt%). The percentage of K and Ca are less than 1 wt%, whereas the proportion of Na is higher, varying between 3.08 and 3.66 wt%. According to the ternary Ab-An-Or diagram of [52] for the classification of plagioclase, all the samples analyzed fall in the albite field (Figure 8(A)).

4.3.2. Epidote

Five points of epidotes were analyzed. These epidotes have a homogeneous chemical composition, with higher content of Al₂O₃ than FeOt and Cr₂O₃. Moreover, the content of CaO is high, ranging from 22.87 to 23.77 wt%. The Al₂O₃ content shows very little variation between 21.77 and 22.90 wt%. Total iron content range from 12.82 and 13.82 wt%, whereas the content of chromium is below than 0.1%. The Al-Cr-Fe ternary diagram of [53] shows that the epidotes have a composition of clinozoisite (Figure 8(B)).

4.3.3. Carbonate

Carbonate are fairly ubiquitous in the Diénéméra and Gongondy deposits. Seven carbonates points from three rock samples were analyzed chemically. A wide variation in CaO (28.99 wt% - 58.29 wt%), MgO (0.02 wt% - 9.45 wt%) and FeOt (0.10 wt% - 13.98 wt%) was observed depending on the type of carbonate present in the rocks. The carbonate of sample E11G is distinguished from the other carbonates by high MgO and FeOt contents (9.45 wt.% and 13.98 wt.%, respectively and a correspondingly lower CaO content (28.99 wt%). The Mg-Ca-Fe ternary diagram of [54] shows that all carbonates have a composition close to calcite, except the carbonates of sample E11G, which lies between dolomite and ankerite composition (Figure 8(C)). In the Mg-Fe-Mn diagram of [54], the carbonate of sample E11G falls in the field of Fe-dolomite (Figure 8(D)).

4.3.4. Chlorite

Chlorite is an abundant mineral in the rocks in the study area. The chemical composition of the mineral shows a high proportion of SiO₂ (23.83 wt% - 29.78 wt%), Al₂O₃ (13.18 wt% - 23.30 wt%) and low proportions of CaO (0 wt% - 0.38 wt%), K₂O (0 wt% - 0.82 wt%) and Na₂O (< 0.20 wt%). The proportions of MgO (9.12 wt% - 20.01 wt%) and FeOt (18.03 wt% - 33.03 wt%) vary depending to the samples. The lowest proportions of MgO are found in the chlorite of samples E22G where this proportion of of FeOt (31.04 wt% - 33.03 wt%) than other chlorites. Other samples have a relatively same proportion of FeOt and MgO. The content of TiO₂ of most samples is less than 0.1% except for samples E11G where the content is more than 0.4%. In binary diagram of [55], the chlorites from sample E22G fall into the ripidolite field. whereas the chlorites from the other samples are of types of ripidolite and pycnochlorite fields (Figure 8(E)).

4.3.5. Amphibole

The chemical composition of 3 amphiboles points was analyzed. The contents of SiO₂ (52.93 - 54.89 wt%), FeOt (9.07 wt% - 11.32 wt%), MgO (16.63 wt% - 18.01 wt%) and CaO (11.58 wt% - 12.28 wt%) are high. Low contents are observed for TiO₂ (0.25 wt% - 0.27 wt%), Na₂O (0.46 wt% -0.52 wt%) and K₂O (0.03 wt% - 0.11 wt%). The diagram of [56] also shows two different sources of amphiboles: primary and secondary (Figure 8(F)). The diagram of [57] shows that amphiboles are from two types: actinolite and magnesio-hornblende (Figure 8(G)).

Figure 8.Diagrams of alteration minerals chemistry. Ab-An-Or ternary diagram for classifying plagioclase [52] (A). Al-Cr-Fe ternary diagram for epidote classification [53] (B). The Mg-Ca-Fe ternary diagram of carbonates classification [54] (C). the Mg-Fe-Mn diagram of [54] is used to discriminate Fe-carbonates (D). Binary diagram of chlorite classification [55] showing two groups of chlorites (E). Classification binary diagram of amphibole according to the origin [56] (F) and the composition [57] (G).

Table 3. Chemical composition of minerals: major elements.

Table 4. Chemical composition of minerals: trace elements.

Using some common metals in mineralization such as copper, gold, silver, molybdenum, lead, zinc in association with some indices such as CCPI, AI and IP, it is possible to correlate the mineralization and hydrothermal alteration types.

Factor 1 shows an association of copper, silver, gold, molybdenum, selenium and antimony mineralization with CCPI and AI (Figure 9(A)). In factor 5 the same mineralization shows a positive correlation with CCPI, TOT/S, FeO₃, and TiO₂ and a negative correlation with AI (Figure 9(E)). In factor 2, the initial mineralization shows a negative correlation with the phenomena. On the other hand, zinc, nickel, barium mineralization is identified and shows a positive correlation with CCPI (Figure 9(B)). Factor 3 translates a positive correlation of copper and gold with the phenomena to which a strong positive correlation with alkaline minerals (Rb, Cs, IP) can be associated. The positive correlation with AI and negative with CCPI is also remarkable (Figure 9(C)). Factor 4 shows an association of gold and silver positively correlated with the phenomenon, as well as AI where the correlation is positive, but negative with CCPI (Figure 9(D)). Factor 6 is quite similar to factor 4 but differs by a negative correlation of the mineralization with gold, silver, selenium with the phenomena (Figure 9(F)). In the four factors where the mineralization with Cu, Au, Ag shows a positive correlation with the different phenomena, Sr on the other hand shows a negative correlation.

Figure 9. Diagram showing the different factors associated with metals, hydrothermal minerals and alteration index. Evident association of Cu-Au mineralization with alteration indexes is observed in Factors 1 (A), 3 (C) and 5 (E). Factors 2 and 6 show lack of correlation of Cu-Au mineralization with the phenomena (B) (F), whereas, factor 4 shows an association of AI with Se, Au and Ag (D).

The sulfide isotopes for four samples of pyrite are listed in Table 5. The samples consist of one semi-massive pyrite in fractures of quartz veins with few chalcopyrite inclusions which exhibits negative δ³⁴S sulfide (−4,1) for the pyrite. The three other samples consist of disseminated pyrite associated with hydrothermal minerals (carbonates ± chlorite ± epidote). In these sulfides, the δ³⁴S values range from 2.5 to 4.7.

Table 5. Result of δ³⁴S analyses of four samples of Diénéméra-Gongondy deposit.

The use of factor analysis of hydrothermal alteration and mineralization highlighted the association of mineralization and hydrothermal alteration. Indeed, six phenomena are highlighted. The first phenomenon, related to factor 1 correspond to propylitic alteration associated with the Cu-Au-Ag-Mo-Sb mineralization in dioritic intrusions. The same mineralization association is observed in the factor 5 which could be related with phyllitic-carbonate alteration and basic mineralized intrusions represented by microgabbro. The sericitic alteration consists of two types. The first one is represented by factor 3 which is associated with Cu-Au-Mo mineralization and is remarkable by a positive correlation with IP, Rb, Cs and AI. The second is associated with Au-Ag mineralization and Hg, As and SI. This phenomenon can be associated with late orogenic mineralization highlighted by previous studies in the zone [29]-[31]. The last two factors are not associated with the Cu-Au mineralization. The first can be associated with the chloritic alteration regarding the positive correlation with CCPI and negative correlation with AI. The factor 6 corresponds to late non altered diorite which is barren in Cu-Au mineralization. This analysis is in accordance with the ore microscopy studies of the copper mineralization and the study of the hydrothermal alteration at microscopic scale.

The ore microscopy indicated that copper mineralization is associated with hydrothermal alteration. The chalcopyrite is observed in dissemination and associated with alteration minerals, dominated by chlorite and carbonates at Gongondy. However, at Diénéméra, chlorite, carbonates, epidote and muscovite are the most frequent alteration minerals which are associated with dioritic rocks (Figure 10(A) and Figure 10(B)).

In felsic grinding zones, it can be found in subautomorphic crystals and cemented by felsic recrystallization products including carbonates and silica. Moreover, chalcopyrite is also frequently associated with pyrite in crushing zones (Figure 10(C)) or in the form of a trail associated with chlorite and carbonates (Figure 10(D)). The fine crystals are most often crumbled or chamoised while the coarser ones are cracked and filled with chlorite which often acts as cement (Figure 10(E)). In intergranular interstices in dioritic rocks at Diénéméra, chalcopyrite and pyrite are frequently associated with chlorite, carbonates, epidote and muscovite (Figure 10(F)).

Figure 10. Hand specimens and photomicrographs showing the association of sulfides with hydrothermal alteration. The ore microscopy showed a relation between chalcopyrite and pyrite with some quartz in inclusions in pyrite fractures (A). Sulfides in relation with chlorite and cristal of K-feldspar (B). Defromed and fragmented chalcopyrite associated with chloritic alteration (C). Fragmented pyrite in deformed zone associated with quartz in shadow pressure and carbonate (D). Massive chalcopyrite in quartz fractures (E). Chalcopyrite and pyrite observed in intergranular joint observed in quartz microdiorite (F). Silicified microdiorite associated with semi-massive sulfides (G). Fractured quartz with sulfides in the fractures (H).

In silicified zones, chalcopyrite and pyrite are massive to semi-massive associated with silica and chlorite.

The study of Gaoua Porphyry copper deposit showed that potassic alteration is absent. The greenschist facies metamorphism or late hydrothermal alteration can explain this lack, regarding the intense albitization of potassic feldspar and possible chloritization of hydrothermal biotite. The albitization is highlighted by microprobes analyses which indicated that the feldspars in the study area have a composition of albite. Even if [21] thought that this albitization can be related to the chemical factor of Precambrian rock composition, we think that this albitization can be associated with late hydrothermal alteration, notably the propylitic alteration generally associated with albitic phase or with retrograde greenschist facies metamorphism. This latter hypothesis can be justified by the posteriority of potassic alteration in relation with the deposition of the porphyry copper deposit around 2161 ± 23 Ma, whereas the greenschist facies metamorphism occurred during the D2Ga deformational phase which took place after −2150 Ma [28][30][32][58].

The most important alteration minerals frequently associated with the copper mineralization of Gaoua porphyry copper deposit consists of chlorite, carbonates and locally silica, sericite and epidote. The composition of some of these hydrothermal minerals such as chlorite and epidote is an important tool for porphyry copper mineralization exploration [59]-[63]. The analyses of chlorite in the study area enabled us to define two types of chlorite which are low-magnesium chlorite and the relatively high iron-magnesium chlorite, respectively with Ripidiolite and pycnochlorite composition. The composition of chlorite can permit to discriminate the two chlorites and to see the link of these chlorites with the mineralization. Based on its association and other minerals such as epidote, carbonates and sericite, we classify chlorite depending on the propylitic, phyllitic-carbonates and chlorite alteration zones. In this study, Fe and Mg rich chlorite is found in propylitic alteration zone, possibly in phyllito-carbonates alteration zone, whereas, the low Mg chlorite is observed in the chloritic alteration zone where no relation was observed with the mineralization. In some parts of Diénéméra area, the relatively high Mg content of secondary amphibole (actinolite), associated with chlorite and epidote is a proof of origin of Fe-Mg-chlorite in link with the propylitic alteration. This is in accordance with the chlorite study in Naruo porphyry copper deposit [64] where the binary MgO versus FeO diagram determinated a porphyry copper origin of Mg-chlorite. However, Fe-rich chlorite can be associated with a late fluid, notably in relation with greenschist facies metamorphism or orogenic gold mineralization event which overprinted on the porphyry copper mineralization [28][29]. The presence of Fe-rich dolomite associated with chlorite can permit to confirm this hypothesis regarding the absence of carbonates in greenschist metamorphic facies [65][66].

Alkalis such as Sr, Cs, Rb are the most important trace elements to evaluate the impact of hydrothermal fluids on rocks. The mobility of alkalis due to hydrothermal alteration, the primitive mantle normalized formula of Leach and Campbell [47] (Cs + Rb)/Th can be used as alteration index to determine the potential mineralization of a lithology. According to this formula, most of the samples analyzed, show a value of (Cs + Rb)/Th of more than 5 units confirming the potentiality of Diénéméra-Gongondy mineralization (Figure 11(A)).

In the binary diagram of [67] based on the excess of Aluminium content of plagioclase used to determine the fertility of the intrusions, all of the samples fell in the field of fertile intrusion porphyry (Figure 11(B)). The clinozoisite composition of epidote is also a proof of the Al-rich origin of plagioclase in the area zone.

Figure 11. Graph of potential mineralisation index using the primitive mantle normalized formula of [47]. The red line corresponds to limit of potential mineralization which must be more than 5 units (A). Binary diagram of [66] for the determination of the fertility of intrusions. Most of the samples fell in the fertile intrusion field (B).

The preliminary δ³⁴S sulfide value of Diénéméra-Gongondy Cu-Au deposit is close to the δ³⁴S mantle value and confirms the magmatic origin of the copper and gold mineralization in comparison with other porphyry copper in the world (Figure 12). Indeed, this conclusion is in accordance with studies of fluid inclusions in quartz associated with the mineralization in the study area [28][30]. The early mineralization is disseminated associated with dioritic intrusions. During the emplacement of the preserved-structure Fe-rich (magnetite) microgabbro and the possible contact with meteoric groundwater induced a probably sericitic alteration and precipitation of copper mineralization, which corroborates with the study of [68]. The emplacement of several generations of late diorites in the form of dykes and especially of a late gabbro adjacent to the mineralization at Diénéméra and Gongondy, could be the origin of the mobilization of hydrothermal fluids rich in iron and magnesium which caused the chloritic alteration observed in certain zones. It is not excluded that this chloritic alteration often associated with carbonates in relation with the emplacement of the orogenic gold mineralization.

Figure 12. Comparison of the δ34S compositions in the Diénéméra-Gongondy Cu-Au deposit with the other well-known porphyry Cu deposits. Data from [70].

The late fluid which caused the chloritic alteration and released acidic fluids, origin of the dissolution of the rock and carbonates and sulfates filled cavities or breccia. The acidic fluids are the result of interaction of late fluids with the host rocks, responsible of the alteration of albite into chlorite with liberation of acidic fluids according to [69].

The mineralization of Gaoua is however fortunate to have been preserved despite the different phases of late intrusions and the level of erosion that could have destroyed a part of the mineralization. Although the association of copper mineralization with propylitic alteration in Diénéméra is obvious, it does not contradict the conclusions of [1] who defined the propylitic zone as barren. Indeed, the propylitic zone is characterized by disseminated mineralization, and does not constitute an important zone of economically valuable mineralization because the grades are generally low, less than 0.2% of copper. However, the presence of significant mineralization associated with breccia, cavities, intergranular interstices and silicification (Figure 10), reflects a last-stage phenomenon that would be the cause of the remobilization of the copper mineralization or the reconcentration of this mineralization as [25] had stated. The first-order structures associated with the D3Ga deformation phase would have effectively contributed to this remobilization by channeling the fluids rich in gold and copper released by mafic intrusions (Microgabbro) into more permeable and lower pressure zones or zones likely to contain the mineralization such as intergranular interstices, breccia zones or cavities (Figure 13) This conclusion is in accordance with studies of Cu-Au-Mo in Oyu Tolgoï [71][72] and Cu ± Mo ± Au ± Ag in Imourkhssen [73].

Figure 13. Lithological log of drill cores from Diénéméra and Gongondy deposits showing the relationships between lithologies, copper and gold mineralization, hydrothermal alteration and shearing. Important mineralization (more than 0.2% of copper) can be observed in breccia, cavities, veins and veinlets, intergranular interstices or as massive patches. Abbreviations: Br.: Breccia, Cav.: Cavities, Diss.: Dissemination, Frac.: Fracture, Inters. Intergranular joints, Mas.: Massive, Stock.: Stockwerk.