The origin and evolution of the ore-forming fluids at the Manondo-Choma gold prospect, Kirk range, southern Malawi

Joshua Chisambi; Bjorn von der Heyden

doi:10.1515/geo-2022-0494

Artikel Open Access

The origin and evolution of the ore-forming fluids at the Manondo-Choma gold prospect, Kirk range, southern Malawi

Joshua Chisambi und Bjorn von der Heyden

Veröffentlicht/Copyright: 28. September 2023

Veröffentlicht von

Veröffentlichen auch Sie bei De Gruyter Brill

Manuskript einreichen Informationen für Autor*innen

Aus der Zeitschrift Open Geosciences Band 15 Heft 1

Abstract

This study provides an in-depth analysis of fluid inclusions (FIs) and stable isotopes from the Manondo-Choma gold prospect in southern Malawi to understand the ore-forming mechanisms, genesis, and evolution of the hydrothermal fluids responsible for gold mineralization at the Manondo-Choma gold prospect. FIs and microthermometry studies were carried out on mineralized quartz veins from the area. The Manondo-Choma gold prospect is located in the southern Malawi within the Kirk range. Gold is mostly found in quartz veins within metamorphic rocks, such as gneiss and schists, and is structurally controlled by the NE–SW ductile shear zones. Gold mineralization is linked to quartz sulfide veins. Mineralization occurs in the following three stages: early, middle, and late, of which the middle stage is more plorific. The following three FI types were recognized in the quartz veins: pure carbonic, aqueous carbonic (CO₂–H₂O) and aqueous (H₂O–NaCl) inclusions. Gold and associated mineralization were likely precipitated due to the lowering of pressure and fluid immiscibility. Oxygen isotope data indicate that the source of ore-forming fluids at the Manondo-Choma gold prospect was largely metamorphic in origin with minor magmatic input.

Keywords: fluid inclusion; gold mineralization; Kirk range; hydrothermal fluids

1 Introduction

The Manondo-Choma gold prospect located in the Kirk range area, southern Malawi, has the potential to be an important gold-producing region. Placer gold is currently being mined by small-scale artisanal miners along the rivers. Gold is found in hydrothermal quartz veins that break through biotite schist rocks where it occurs as flakes ranging in size between 0.24 and 4 mm. Little geological research has been conducted in the area since the discovery of gold occurrences in the 1900s. The area was subjected to a regional geological survey (1934–1935) reported by Bloomfield and Garson [1] and subsequent geochemical stream sediment sampling campaigns [2,3]. More recently, Chisambi and von der Heyden [4,5] have shown that gold is hosted in schistose metamorphic rocks in which mineralization is associated with SE dipping rocks that have experienced multiple generations of folding and shearing. Gold mineralization is confined to ductile shear structures trending NE–SW direction [4]. The Manondo-Choma thrust is proposed to be the controlling structure for gold mineralization [6]. These previous studies mostly concentrated on geology and the controls of the primary gold mineralization. The mechanism of gold precipitation has remained largely unknown until now, owing to a lack of comprehensive geochemical data for the deposit. A thorough investigation of fluid inclusions (FIs) (microthermometry) and stable isotopes (oxygen of quartz) from the Manondo-Choma gold prospect in southern Malawi is presented in this article to understand the genesis and evolution of the hydrothermal fluids responsible for gold mineralization and to understand the mechanism that led to the formation of the Manondo-Choma gold prospect. This is the first study of its kind in the area that focuses on the character of mineralizing ore fluids.

2 Regional geological outline

Malawi has a complex geological history and is dominated by lithologies and structures related to the formation of three major orogenic belts, viz. the Ubendian belt (2,200–1,800 Ma), the Irumide belt (1,050–950 Ma), and the Mozambique belt (800–500 Ma) [7,8,9]. The Ubendian orogeny represents the amalgamation of eastern Africa and Archean cratonic blocks/nuclei [10,11] and has resulted in NW–SE trending structures confined predominantly to the northern regions in Malawi [9]. A subsequent tectono-metamorphic event (the Irumide orogeny) gave rise to the Irumide belt, which stretches from Zambia through central and southern Malawi to the north of Mozambique [12,13,14]. The main structural trend of the Irumide belt is NE–SW and is related to extensive crustal shortening during the main stage of the Irumide orogeny [15,16,17]. It is characterized by amphibolite-grade metamorphism and accompanied by widespread granite magmatism and anataxis [15]. This orogenic belt is subdivided by crustal-scale shear zones into the Irumide sensu stricto, the Southern Irumide, the Unango, and the Nampula sub-provinces (Figure 1) [16,18]. The Pan-African orogeny (∼800–500 Ma) formed the Mozambique belt [17,19], which underwent metamorphic conditions of 750–800°C at roughly 12–13 kbar and subsequent amphibolite facies retrogression at 550–700°C and 5–8 kbar [20]. Pan-African tectonism and high-grade metamorphism caused significant overprinting and reactivation of older orogenic features from the Ubendian, Irumide, and Kibaran orogenies [9] and resulted in the development of predominantly N–S trending structures and fabrics.

Figure 1

Regional geology for southern Malawi and surrounding countries. Black dots indicate the location of identified gold deposits. Our study area is indicated by the blue rectangle.

This regional scale understanding of the Malawian basement geology is derived largely from detailed work conducted in the east of Zambia [21] and the north of Mozambique [16,18] with only little focus on Malawian rocks [1,22,23]. Importantly, Kronner and Collins [23] have shown that southern Malawi’s high-grade gneissic assemblage (covering the Kirk range study area) reached peak metamorphic conditions at 900 ± 70°C and 9.5 ± 1.5 kbar during the Neoproterozoic Pan-African orogeny. The region is bounded by Mbwembeshi shear to north, Sanangoè shear belt to the south, and Lurio shear belt to the east and thus coincides with a continuation of the Irumide belt in the south (Figure 1), which represents the tectonized edge of Niassa Craton [24]. Pan-African magmatism is manifested through the Carbonatitic intrusions named the Chilwa Alkaline Province such as the Kangankunde Hill in the Kirk range [25].

2.1 Local geology

The Manondo-Choma gold prospect is situated in Malawi’s southern region in the Balaka district Phalula area about 50 km south of Blantyre. Biotite schists and gneisses are host rocks for gold mineralization. These rocks were affected by the southern Irumide orogeny and were later reworked during the Pan-African orogeny. Zircon U-Pb dating indicates that these rocks were metamorphosed at 1,040 and 650 ± 12 Ma (unpublished data). Structurally, the Manondo-Choma gold prospect area is characterized by NE–SW trending – faults and shear zones, and gold mineralization are associated with these structures.

Mineralization is associated with assemblages of sulfide minerals dominated mainly by pyrite, pyrrhotite, chalcopyrite, sphalerite, and gold (Figure 2a–e). Gold is present both as free Au in wall rock and in association with pyrite (Figure 2c–e). Quartz, feldspar, sericite, chlorite, and clay minerals make up the bulk of gangue minerals.

Figure 2

(a and b) Ore minerals identified using a reflected light microscope at the Manondo-Choma gold prospect composed of pyrrhotite, pyrite, chalcopyrite, and sphalerite. (c–e) Gold (Au) mineralization identified in thin section occurring in association with pyrite (Py) and as free gold. (f) Late-stage milky veins with patches of sulfide mineralization. (g) Carbonate early-stage barren quartz vein crosscut by the main-stage mineralized smoky vein. (h and i) Main-stage mineralized quartz veins with sulfide mineralization in yellow.

Three stages of mineralization (early, medium, and late stages) were recognized in the area based on field evidence, petrographic observation, and crosscutting relationships.

Early-stage mineralization is represented by the development of milky quartz veins with no gold mineralization (Figure 2g). The main or middle stage is characterized by quartz sulfide veins that crosscut the early milky veins and denote the main stage of gold mineralization. In this stage, quartz is fine grained and smoky and contains sulfides (Figure 2g–i). The quartz + carbonate assemblage, with trace pyrite and no gold, characterizes the late stage, mostly occurring as veinlets along and crosscutting the foliation of altered wall rocks (Figure 2f).

3 Materials and methods

3.1 FI

The mineralized quartz vein samples were cut into double-polished sections (about 250–300 μm thick) for FI studies and examined with a petrographic microscope. FIs were carefully observed to identify their genetic and composition types, vapor–liquid ratios, and spatial clustering. The criteria and recommendations of authors in [26,27,28] were used to select the FI assemblages. Microthermometry was carried out at the University of Stellenbosch’s FI Laboratory using the Linkam THMSG 600 heating-freezing stage with a temperature range of −196 to 600°C. Microthermometric measurements were carried out by initially cooling down the stage to −150°C before subsequent heating to observe phase changes. During the initial stages of each heating run, the heating rate was 10–30°C/min and then dropped to 0.5–1°C/min toward the phase change points.

Final ice-melting temperature (T _m-ice), melting temperature of clathrate (T _m-cla), melting temperature of solid CO₂ (T _m-CO₂), homogenization temperature of CO₂ liquid and vapor (T _h-CO₂), and total homogenization temperature (T _h) were the phase transformation parameters that were observed during microthermometry, depending on the types of FIs. The equations of authors in [29] were used to calculate salinities (wt% NaCI equivalents) for aqueous FIs and for aqueous-carbonic inclusions. Salinities were calculated from the last ice-melting temperatures and clathrate (T _m,clath) using the program Aqso5e, included in the package FLUIDS [30]. Densities and pressures were estimated according to microthermometric data for CO₂–H₂O–NaCl system. The stage was calibrated using synthetic FIs supplied by Fluid Inc. against the triple point of pure CO₂ (−56.6°C), the freezing point of water (0.0°C), and the critical point of water (374.1°C) to ensure the accuracy of the measurements.

3.2 Oxygen isotope analysis

Twelve samples of mineralized quartz veins were carefully chosen for oxygen isotope analyses. The samples were crushed to 40–60 mesh. Quartz grains were handpicked under a binocular microscope after crushing. Analysis was done at the stable isotope laboratory at the University of Cape Town, South Africa, using a Finnigan MAT 253 mass spectrometer. Oxygen was extracted from quartz by the BrF₅ method [31]. The results were normalized and reported relative to Vienna standard mean ocean water and the analytical precisions were ±0.2‰. The oxygen isotope values of water were calculated using the fractionation formula 1,000 ln α _quartz−H2O = 3.38 × 106T ⁻² − 3.40 reported by authors in [31], where T is the mean value of the homogenization temperature of FIs from the same quartz samples.

4 Results

4.1 FI petrography

These FIs in quartz are being studied because this quartz is related to the Manondo-Choma gold ore formation. The quartz is associated with pyrite and the pyrite is associated with gold mineralization (Figure 2a–i). When conducting FI petrography, the shape of the FIs, their genetic composition, vapor-to-liquid ratio and spatial distribution of the inclusions were carefully examined. Based on FI components and phase behavior at room temperature (25°C), three types of inclusions were recognized, as liquid-rich aqueous inclusions (H₂O–NaCl), aqueous carbonic (CO₂–H₂O) inclusions, and pure carbonic (CO₂) inclusions.

4.1.1 Liquid-rich aqueous inclusion

Liquid-rich aqueous inclusions are either one phase (LH₂O) or two phase (VH₂O + LH₂O) (Figure 3a–d). Inclusions with two phases are more prevalent, with the volume of vapor taking up 10–40% of the overall cavity. These inclusions range in size from 10 to 20 μm in diameter and come in a variety of forms, from irregular to elliptical and negative crystal. These inclusions are seen as arrays or trails along healed fractures and create many planes, most of which crosscut one another. They are thought to be of secondary origin.

Figure 3

(a–k) Types of FIs recognized at the Manondo-Choma gold prospect from mineralized quartz veins, liquid-rich aqueous inclusions (H₂O–NaCl), aqueous carbonic (CO₂–H₂O) inclusions, and pure carbonic (CO₂) inclusions.

4.1.1.1 Aqueous carbonic inclusions

These FIs are made up of H₂O and CO₂ phases whereby the carbonic phase is about 5–90 vol% (Figure 3e and f). At room temperature, the inclusions can be further subdivided into two phases (VCO₂ + LH₂O) and into three phases (VCO₂ + LCO₂ + LH₂O) with a volumetric proportion of CO₂ ranging from 5 to 90%. These inclusions appear as trails along healed fractures but also occur in clusters or in isolation. These features suggest that they are primary or pseudo-secondary. These inclusions vary in size ranging from 10 to 30 μm.

4.1.1.2 Pure carbonic inclusion

These FIs have sub-rounded to negative crystal shapes and are generally dark in color. At room temperature, they are composed of almost pure carbonic phase without any visible H₂O. They are usually dark with oval crystal morphology (Figure 3g and h). They occur either in isolation or in linear array. They range from 2 to 15 µm in diameter. A few of these inclusions have a small proportion of aqueous solution that fills up to 10% of the inclusion volume. The fact that these inclusions are generally isolated and dispersed implies that they are primary inclusions.

4.2 Microthermometry study of the FIs

The FI numbers within the quartz veins differ depending on the stage of mineralization. The early mineralization stage contains liquid-rich aqueous and aqueous carbonic FI. The main mineralization stage contains all three types of FIs. The late mineralization stage is dominated by liquid-rich aqueous inclusions along fractures.

Detailed microthermometric analysis was conducted on the FIs and the microthemometric data are summarized in Table 1 and Figures 4–6.

Table 1

Summary of the microthermometry data from FI from the three fluid flow stages of the Manondo-Choma gold prospect

Stage	FI type	T _m,CO₂ (°C)	T _m,cla (°C)	T _h,CO₂ (°C)	T _m,ice (°C)	T _h (°C)	Salinity	Density (g/cm³)
Early	Carbonic	−60.8 to −56.8	5.3–6.6	24–31.2		300–349	6.06–8.68	0.23–0.63
	Aqueous				−9.1 to −1.3	303–390	2.23–12.95	0.62–0.78
Main	Carbonic	−60.1 to −56.7	3.7–9.8	17.2–32		285–350	0.038–7.67	0.15–0.80
	Pure carbonic	−59.8 to −56.8		25.6–31.3		288–365		0.24–0.35
	Aqueous				−9.1 to −0.1	222–301	0.18–13. 1	0.70–0.90
Late	Aqueous				−9.8 to −0.8	148–220	1.39–13.72	0.87–0.99

FI type: fluid inclusion type, T _m,CO₂: melting temperature of solid CO₂, T _m,cla: temperature of CO₂ clathrate dissociation; T _hCO₂: homogenization temperature of CO₂, T _m,ice: ice-melting temperature, T _h: total homogenization temperature.

Figure 4

Histograms of ore-bearing FIs in the early stage of mineralization. (a) Total homogenization temperatures for aqueous carbonic and liquid-rich aqueous fluids. (b) Salinity for aqueous carbonic and liquid-rich aqueous fluids. (c) Homogenization temperature of CO₂. (d) Melting temperature of solid CO₂.

Figure 5

Histograms of ore-bearing FIs in the main stage of mineralization. (a) Melting temperature of solid CO₂. (b) Homogenization temperature of CO₂ liquid and vapor. (c) Total homogenization temperatures for aqueous carbonic, pure carbonic, and liquid-rich aqueous fluids. (d) Melting temperature of clathrate. (e) Salinity for aqueous carbonic and liquid-rich aqueous fluids. (f) Final melting temperature of ice.

Figure 6

Histograms of ore-bearing FIs in the late stage of mineralization. (a) Final melting temperature of ice. (b) Salinity for liquid-rich aqueous fluids. (c) Total homogenization temperature. (d) Melting temperature of clathrate.

Aqueous carbonic-type and liquid-rich aqueous-type FIs are present in early mineralization stage quartz veins. The aqueous carbonic-type FIs have solid CO₂-melting temperatures (T _m,CO₂) of −60.8 to −56.8°C during microthermometry, indicating that the carbonic phase contains CH₄ or N₂, and melting temperatures of CO₂ clathrate (T _clath,CO₂) of 5.3 to 6.6°C (Figure 6d), yielding salinities of 6.06–8.68 wt% NaCl equivalent. At temperatures (T _h,CO₂) ranging from 24.0 to 31.2°C, the carbonic phases homogenize to a liquid phase, producing densities of 0.23–0.63 g/cm³. The bulk of these FIs homogenizes into the liquid phase at temperatures 300–349°C (Th,TOT) (Table 1, Figure 4).

The final ice-melting temperatures (T _m,ice) of the aqueous liquid-rich-type FIs range from −9.1 to −1.3°C, producing salinities 2.23–12.95 wt% NaCl equivalent. The majority of these FIs homogenize into the liquid phase at temperatures (T _h, TOT) 303–390°C, corresponding to densities 0.62–0.78 g/cm³ (Table 1, Figure 4). The main mineralization stage is associated with aqueous carbonic, liquid-rich aqueous – and pure carbonic – FI types. FI of the aqueous carbonic and pure carbonic have T _m,CO₂ values ranging from −60.1 to −56.7°C, close to the CO₂ triple point (−56.6°C), indicating that the carbonic phase contains minor quantities of other volatiles, and have Tclath,CO₂ values from 3.7 to 9.8°C, yielding salinities of 0.038–7.67 wt% NaCl equivalent. The carbonic phases homogenize to liquid at temperatures between 17.2 and 32°C, producing 0.15–0.80 g/cm³ as carbonic phase densities. At temperatures 285–350°C, numerous pure carbonic-type FIs homogenized into the vapor phase. The liquid-rich aqueous, FIs have ice-melting temperatures (T _m,ice) values −9.1 to −0.3°C, giving salinities 0.18–13.1 wt% NaCl equivalent. These fluids homogenize at 222–301°C (Th,TOT), producing densities 0.70–0.90 g/cm³ (Table 1, Figure 5).

Only liquid-rich aqueous-type FIs were observed in the late stage of mineralization with T _m,ice ranging from −9.8 to −0.8°C, which corresponds to salinities 1.39–13.72 wt% NaCl equivalent and they homogenize at temperatures (T _h,TOT) 148–220°C into the liquid phase, yielding densities 0.87–0.99 g/cm³ (Table 1, Figure 6).

4.3 Oxygen stable isotopes

Table 2 shows the stable isotopic data for oxygen. The oxygen isotopic compositions of hydrothermal fluids were determined using the fractionation formula 1,000 ln α _Q−H2O = 3.38 × (10⁶/T ⁻²) − 3.40 [32] and homogenization temperatures T of the FIs of the same samples were used. The analyzed δ¹⁸O quartz values range from 7.0 to 12.6‰, yielding calculated δ¹⁸O_H₂O values between 4.3 and 6.9‰ and δD values are in the range of −65 to −44‰.

Table 2

Oxygen isotope composition from the Manondo-Choma gold prospect. δ¹⁸O_m measured δ¹⁸O value of quartz; δ¹⁸O_w value of ore-forming fluids in equilibrium with quartz calculated according to the equation 1,000 ln α _Q−H₂O = 3.38 × (10⁶/T ⁻²) − 3.40 reported by Clayton and Mayeda [31]. The temperatures used in calculation are the homogenization temperature values of the FI quartz samples

Sample	δ¹⁸O_m	δ¹⁸ O_W	δD	T _h
JJ01-1	12.6	4.3	−44	390
JJ01-2	10.1	4.3	−44	388
JJ01-3	11.4	4.4	−45	385
JJ01-4	11.1	4.9	−49	364
JJ01-5	9.2	5	−50	360
JJ01-6	12.7	5.3	−52	351
JJ02-1	10.6	5.3	−52	350
JJ02-2	8.7	5.7	−56	337
JJ02-3	7.1	6.8	−64	303
JJ02-4	7	6.9	−65	301
JJ02-5	8.7	6.9	−65	300

5 Discussion

5.1 Nature and immiscibility of ore-forming fluids

At the Manondo-Choma gold prospect, FI microthermometry results reveal a gradual decrease in temperature from the early to late mineralization stages. FI temperatures of homogenization vary around 301–390°C in the early mineralization stage, 222–301°C in the main mineralization stage, and 148–220°C in the late stage, which is consistent with fluid immiscibility [33]. Furthermore, pure carbonic-type, liquid-rich aqueous-type, and aqueous carbonic-type FIs coexist in the main stage of mineralization, implying that when these FIs were trapped, the ore fluids were in a heterogeneous thermal condition and the initial homogeneous fluids underwent fluid immiscibility. This suggests that gold at the Manondo-Choma prospect was precipitated from immiscible fluids. In many hydrothermal deposits, fluid immiscibility is one of the most important mechanisms for ore formation [34,35,36,37,38] and many gold deposits have been reported to have H₂O–NaCl–CO₂ fluid immiscibility associated with gold mineralization [39,40].

It is inferred that fluid immiscibility at the Manondo-Choma gold prospect might have caused phase separation. The initial NaCl–H₂O–CO₂ fluids split into H₂O-rich and CO₂-rich fluids, leading to the capture of low-salinity CO₂-bearing inclusions and aqueous inclusions. The escape of CO₂ during phase separation caused significant changes in fluid composition, as well as physical and chemical conditions, resulting in a drop in Au solubility in hydrothermal fluids and rapid gold precipitation.

5.2 Pressure–temperature (PT) conditions of ore deposition and metallogenic depth

PT conditions during mineralization were estimated from the FI assemblages. The estimated pressures of the FIs at the Manondo-Choma gold prospect are shown in Figure 7 based on the CO₂–H₂O–NaCl system. Representative density isochores that were used to constrain the trapping pressures were calculated using Hokieflincs H₂O–NaCl [41] and an algorithm for calculating H₂O–NaCl₂–CO₂ fluids developed by MacInnis et al. [42]. To constrain the minimum trapping pressure range, intersections of isochores of FIs with the minimum and maximum densities and homogenization temperature ranges were used [40,43,44]. Plotting the homogenization temperatures as a function of salinity of the aqueous carbonic inclusions [45], representative density isochores were estimated that were used to constrain the trapping pressures. In the early mineralization stage, the trapping pressures of the Manondo-Choma gold prospect range from 240 to 340 MPa, in the main mineralization stage, the trapping pressures range from 200 to 280 MPa, and in the late stage of mineralization, the trapping pressure ranges from 180 to 240 MPa, showing a gradual decrease during the evolution of the ore-forming fluids. We do posit that decreasing pressure and fluid immiscibility led to the precipitation of gold at the Manondo-Choma gold prospect. The calculated trapping depth for the gold at Manondo-Choma corresponds to a depth of 7.6–10.7 km in the main mineralization stage, applying a ground pressure gradient of 26 Mpa/km (mean density of upper crust). This shows that the ore deposition occurred at mesothermal depth ranges at temperatures 222–301°C, pressures 200–280 MPa, and depths 7.6–10.7 km. Ore-forming fluids at the Manondo-Choma gold prospect are therefore low- to medium-temperature, low-salinity fluids.

Figure 7

Representative isochores for minimum and maximum CO₂ densities for H₂O–NaCl–CO₂ inclusions and solvus for H₂O–NaCl–CO₂ fluids containing 6 wt% equivalent NaCl and 15 mol% CO₂. (a) The early-stage mineralization estimated trapping PT diagram, (b) the main-stage mineralization estimated trapping PT diagram, and (c) the late-stage mineralization estimated trapping PT diagram.

5.3 Source of the ore-forming fluids

Table 2 provides the analytical data on oxygen isotopes in quartz veins as well as the estimated δ¹⁸O water values of ore fluids. Oxygen isotope data fall in the metamorphic water field, with some data in the primary magmatic water field (Figure 8a). In some metamorphic environments, magmatic activity is prevalent, and some orogenic gold deposits are thought to have formed as a result of a mix of metamorphic processes and magmatic inputs [46]. However, ore-forming fluids in the Manondo-Choma gold prospect most likely came from a metamorphic source due to their low salinity, which differs from the typically high-salinity magmatic-hydrothermal fluids. The fluid composition (low salinity and H₂O–CO₂–N₂–CH₄ in composition) indicates a metamorphic origin and is characteristic of most orogenic gold deposits across the world. This is further substantiated by the salinity versus homogenization diagram (Figure 8b), FI data from the Manondo-Choma gold prospect plot in the orogenic goldfield.

Figure 8

(a) δD vs δ¹⁸O plot of the ore-forming fluids at the Manondo-Choma gold prospect, they plot in the metamorphic and primary magmatic water fields. (b) Summary homogenization temperature–salinity diagram indicating FI composition plot in the orogenic gold field.

5.4 Ore genesis of the Manondo-Choma gold prospect

The foregoing explanation implies that the fluid system at the Manondo-Choma gold prospect evolved from the early stage of CO₂-rich fluids through the very late stage of low CO₂ fluids, which was accompanied by fluid boiling and escape of CO₂ gas. Thus, the Manondo-Choma gold prospect was created by metamorphic fluids and may be genetically categorized as orogenic, owing to the fact that the original ore-forming fluid is CO₂ rich and low salinity, which are diagnostic characteristics of orogenic mineral systems [45–49]. Further evidence to substantiate the categorization of the Manondo-Choma gold prospect as orogenic type is that the area has similar metamorphic and structural features that are characteristic of orogenic-type deposits [50]. This is further substantiated by the fact that the gold prospect area occurs along shear zones and we interpret it to have developed in the orogenic cycle of the Irumide and Pan-African belts [4]. Aqueous fluids that were very hot and low salinity circulated through these shear zones (Figure 9). These fluids were derived under amphibolite facies conditions from deeply buried rocks as they were being metamorphosed (Figure 9). The metal-rich hydrothermal fluids migrated upwards during the mineralization, along fractures/shear zones and lost pressure. The migration of the fluids along the fractures brought about major changes in its physico-chemical condition and this led to fluid immiscibility and gold precipitation [51]. A reduction in pressure ensued from the opening and expansion of the shear zones, as well as the upward migration of the fluids, resulting in the separation of carbonic and aqueous phases.

Figure 9

Mineralization model for the Manondo-Choma area in Kirk range. Mineralization occurred as a result of loss in pressure of the metamorphic fluids derived from devolatilization reactions.

6 Genetic model for gold-deposition mechanism at Manondo-Choma area

Gold is primarily transported in hydrothermal solutions as gold chloride (AuCl²⁻) and gold bisulfide complexes (Au(HS)₂ ⁻). The AuCl²⁻ complexes are commonly found in high salinity, near-neutral to slightly alkaline fluids whereas the Au(HS)₂ ⁻ complexes are more prevalent in transporting gold at near-neutral to slightly acidic pH in sulfur-bearing fluids and at relatively low temperatures (400°C). Considering that the ore-forming fluids at the Manondo-Choma gold prospect were rich in CO₂ and relatively low temperatures (148–390°C), we infer that the gold was probably transported as Au(HS)₂ ⁻ complexes in the ore-forming fluids, which is also consistent with the fact that gold is usually accompanied with sulfides (especially pyrite, Figure 2). Thus, gold deposition at the Manondo-Choma prospect probably is related to the breakdown of gold bisulfide complexes (Au(HS)₂ ⁻).

Fluid immiscibility might trigger H₂S loss from ore-forming fluids and Au(HS)₂ ⁻ destabilization [52,53]. The temperature of the fluid dropped from the early mineralization stage through the main and late mineralization stages accompanied by a significant pressure drop. This process might have led to the destabilization of Au(HS)₂ ⁻ and the deposition of gold.

Field evidence indicates that hydrothermal alteration, mainly assemblage of sericite–chlorite, is prevalent in the Manondo-Choma gold prospect. The aqueous-carbonic fluids interacted with the country rocks across a wide temperature range (390–148°C) as they moved to shallower depths through the shear zones. When ore-bearing fluids interacted with iron-bearing minerals in wall rocks to create pyrite, the solubility of Au(HS)₂ ⁻ decreases, resulting in gold deposition. All this suggests that the fluid–wall-rock interaction also played a significant role in gold precipitation.

7 Conclusions

(1) The mineralization of the Manondo-Choma gold prospect is structurally controlled and divided into the following three stages: early, main, and late.

(2) Aqueous-carbonic, pure carbonic, and liquid-rich aqueous FIs were recognized as the three kinds of FIs. The early mineralization stage contains aqueous carbonic and liquid-rich aqueous FIs, the main mineralization stage has all three types of FIs, and the late mineralization stage only contains liquid-rich aqueous inclusions.

(3) The initial ore fluids were metamorphic and derived from metamorphic devitalization with a magmatic component. These fluids were channeled/transmitted through deep faults from the deep crust.

(4) Drop in pressure, fluid–rock interaction, and temperature decrease caused large-scale fluid immiscibility, which led to gold deposition.

(5) The Manondo-Choma gold prospect and its fluid chemistry have many of the same characteristics as orogenic gold deposits across the world; thus, we conclude that this gold prospect area is orogenic.

jjchisambi@gmail.com

Acknowledgments

The authors would like to thank the Malawi University of Business and Applied Sciences (MUBAS) for funding this work and the Central Analytical Facility staff at the Stellenbosch University in South Africa for the analysis of the data.

Funding information: This research was funded by the Malawi University of Business and Applied Sciences. The institution provided essential resources and support that facilitated the successful completion of this work. The authors express their sincere gratitude for the resources invested in this work.
Conflict of interest: Authors state no conflict of interest.

References

[1] Bloomfield K, Garson MS. The geology of the kirk range-lisungwe valley area. Ministry of natural resources. Geological Survey Department. Bulletin No. 17. The government printer, Zomba Malawi; 1965.Suche in Google Scholar

[2] Malunga G. Geochemical exploration of gold in the likudzi block. Geological Survey of Malawi. Unpublished report; 1992.Suche in Google Scholar

[3] Hori T, Tomita K. Regional geochemical reconnaissance of kirk range-lisungwe, Malawi: Kagoshima University; 1997.Suche in Google Scholar

[4] Chisambi J, von der Heyden B. Primary gold mineralization in the lisungwe valley area, kirk range, southern Malawi. South Afr J Geol. 2019;122:505–18. 10.25131/sajg.122.0039.Suche in Google Scholar

[5] Chisambi J, Von Der Heyden B, Tshibalanganda M, Le Roux S. Gold exploration in two and three dimensions: improved and correlative insights from microscopy and x-ray computed tomography. Minerals. 2020;1–20.10.3390/min10050476Suche in Google Scholar

[6] British Geological Survey. Mineral Potential of Malawi; 2009.Suche in Google Scholar

[7] De Waele B, Mapani B. Geology and correlation of the central Irumide belt. J Afr Earth Sci. 2002;35:385–97.10.1016/S0899-5362(02)00149-5Suche in Google Scholar

[8] Lenoir JL, Liégeois JP, Theunissen K, Klerkx J. The palaeoproterozoic ubendian shear belt in Tanzania: Geochronology and structure. J Afr Earth Sci. 1995;19:169–84.10.1016/0899-5362(94)90059-0Suche in Google Scholar

[9] Ring U, Kröner A, Toulkeridis T. Palaeoproterozoic granulite-facies meta-morphism and granitoid intrusions in the Ubendian-Usagaran Orogen ofnorthern Malawi, east-central Africa. Precambrian Res. 1997;85:27–51.10.1016/S0301-9268(97)00028-4Suche in Google Scholar

[10] Boniface N, Schenk V, Appel P. Mesoproterozoic high-grade metamorphism in pelitic rocks of the northwestern Ubendian Belt: Implication for the extension of the Kibaran intra-continental basins to Tanzania. Precambrian Res. 2014;249:215–28. 10.1016/j.precamres.2014.05.010.Suche in Google Scholar

[11] McConnell RB. Outline of the geology of Ufipa and Ubende. Bull Geol Surv Tanganyika. 1950.Suche in Google Scholar

[12] De Waele B, Liegeois J, Nemchin AA, Tembo F. Isotopic and geochemical evidence of proterozoic episodic crustal reworking within the irumide belt of south-central Africa, the southern metacratonic boundary of an Archaean Bangweulu Craton. Precambrian Res. 2006;148:225–56. 10.1016/j.precamres.2006.05.006.Suche in Google Scholar

[13] Ackermann EH, Forster A. Grundzuge der stratigraphie und struktur des irumide orogen. 21st International Geological Congress; 1960. p. 182–92.Suche in Google Scholar

[14] Ackermann EH. Ein neuer faltengürtel in Nordrhodesien und seine tektonische stellung im Afrikanischen grundgebirge. Geologische Rundsch. 1950;38:24–39.10.1007/BF01766570Suche in Google Scholar

[15] Macey PH, Miller JA, Rowe CD, Grantham GH, Siegfried P, Armstrong RA, et al. Geology of the Monapo Klippe, NE Mozambique and its significance for assembly of central Gondwana. Precambrian Res. 2010;233:259–81. 10.1016/j.precamres.2013.03.012.Suche in Google Scholar

[16] Boyd R, Nordgulen Ø, Thomas RJ, Bingen B, Bjerkgård T, Grenne T, et al. The geology and geochemistry of the east African Orogen in Northeastern Mozambique. Geol Soc South Afr. 2010;113:87–129. 10.2113/gssajg.113.1.87.Suche in Google Scholar

[17] De Waele B, Fitzsimons ICW, Tembo F, Mapani B. The geochronological framework of the Irumide Belt: A prolonged crustal history along the margin of the Bangweulu Craton. Am J Sci. 2009. 10.2475/02.2009.03.Suche in Google Scholar

[18] Bingen B, Jacobs J, Viola G, Henderson IHC, Skår Ø, Boyd R, et al. Geochronology of the Precambrian crust in the Mozambique belt in NE Mozambique, and implications for Gondwana assembly. Precambrian Res. 2009;170:231–55. 10.1016/j.precamres.2009.01.005.Suche in Google Scholar

[19] De Waele B, Kampunzu AB, Tembo F. The mesoproterozoic irumide belt of Zambia. J Afr Earth Sci. 2006;46:36–70. 10.1016/j.jafrearsci.2006.01.018.Suche in Google Scholar

[20] Sommer H, Kröner A. Lithos Ultra-high temperature granulite-facies metamorphic rocks from the Mozambique belt of SW Tanzania AR. Lithos. 2013;170–171:117–43. 10.1016/j.lithos.2013.02.014.Suche in Google Scholar

[21] Karmakar S, Schenk V. Mesoproterozoic UHT metamorphism in the Southern Irumide Belt, Chipata, Zambia: Petrology and in-situ monazite dating Mesoproterozoic UHT metamorphism in the Southern Irumide Belt, Chipata, Zambia: Petrology and in situ monazite dating. Precambrian Res. 2016;275:332–56. 10.1016/j.precamres.2016.01.018.Suche in Google Scholar

[22] Achille M, Andreoli G. Petrochemistry, tectonic evolution and metasomatic mineralisations of Mozambique belt granulites from S Malawi and Tete (Mozambique). Precambrian Res. 2001;25:161–86.Suche in Google Scholar

[23] Kroner A, Collins AS. The East African Orogen: New Zircon and Nd ages and implications for Rodinia and Gondwana supercontinent formation and dispersal. Gondwana Res. 2001;4:179–81.10.1016/S1342-937X(05)70685-7Suche in Google Scholar

[24] Andreaoli M. Petrochemistry, tectonic evolution and metasomatic mineralisation of Mozambique belt granulites from South Malawi and Tete (Mozambique). Precambrian Res. 1984;25:161–86.10.1016/0301-9268(84)90031-7Suche in Google Scholar

[25] Bloomfield K, Garson MS. The geology of the kirk range-lisungwe valley area. Zomba: Government print, Bulletin No. 26; 1965.Suche in Google Scholar

[26] Roedder E. Fluid inclusions. Reviews in mineralogy Vol. 12. Washington DC: Mineralogical Society of America; 1984.10.1515/9781501508271Suche in Google Scholar

[27] Randive KR, Hari KR, Dora ML, Malpe DB, Bhondwe AA. Study of fluid Inclusions: Methods, techniques and applications. Gondwana Geol Mag. 2014;29:19–28.Suche in Google Scholar

[28] Goldstein RH. Fluid inclusions in sedimentary and diagenetic systems. Lithos. 2001;55:159–93.10.1016/S0024-4937(00)00044-XSuche in Google Scholar

[29] Bodnar RJ. Revised equation and table for determining the freezing point depression of H2O–NaCl solutions. Geochim Cosmochim Acta. 1993;57:683–4.10.1016/0016-7037(93)90378-ASuche in Google Scholar

[30] Bakker RJ. Package FLUIDS 1. Computer programs for analysis of fluid inclusion data and for modelling bulk fluid properties. Chem Geol. 2003;194:3–23.10.1016/S0009-2541(02)00268-1Suche in Google Scholar

[31] Clayton R, Mayeda TK. The use of bromine pentafluoride in the extraction of oxygen from oxides and silicates for isotopic analysis. Geochim Cosmochim Acta. 1963;27:43–52.10.1016/0016-7037(63)90071-1Suche in Google Scholar

[32] Clayton RN, O'Neil JR, Mayeda TK Oxygen isotope exchange between quartz and water. J Geophys Res. 1972;77:3057–67.10.1029/JB077i017p03057Suche in Google Scholar

[33] Coulibaly Y, Boiron MC, Cathelineau M, Kouamelan AN. Fluid immiscibility and gold deposition in the Birimian quartz veins of the Angovia deposit (Yaoure, Ivory Coast). J Afr Earth Sci. 2008;50:234–54. 10.1016/j.jafrearsci.2007.09.014.Suche in Google Scholar

[34] Roedder E, Bodnar RJ. Fluid inclusionstudies of hydrothermal ore deposits. In: Barnes HL, editor. Geochemistry of hydrothermal ore deposits. 3rd edn. New York: Wiley & Sons, Inc; 1997. p. 657–98.Suche in Google Scholar

[35] Chen Y, Pirajno F, Li N, Guo D, Lai Y. Isotope systematics and fl uid inclusion studies of the Qiyugou breccia pipe-hosted gold deposit, Qinling Orogen, Henan province, China: Implications for ore genesis. Ore Geol Rev. 2009;35:245–61. 10.1016/j.oregeorev.2008.11.003.Suche in Google Scholar

[36] Li L, Sun J, Men L, Chai P. Origin and evolution of the ore-forming fluids of the Erdaogou and Xiaobeigou gold deposits, Jiapigou gold province, NE China. J Asian Earth Sci. 2016;129:170–90. 10.1016/j.jseaes.2016.08.009.Suche in Google Scholar

[37] Yang L, Guo L, Wang Z, Zhao R, Song M, Zheng X. Timing and mechanism of gold mineralization at the Wang’ ershan gold deposit, Jiaodong Peninsula, eastern China. Ore Geol Rev. 2016;88:491–510. 10.1016/j.oregeorev.2016.06.027.Suche in Google Scholar

[38] Yang L, Deng J, Guo L, Wang Z, Li X, Li J. Origin and evolution of ore fluid, and gold-deposition processes at the giant Taishang gold deposit, Jiaodong Peninsula, eastern China. Ore Geol Rev. 2015;72:585–602. 10.1016/j.oregeorev.2015.08.021.Suche in Google Scholar

[39] Chai P, Sun J, Xing S, Li B, Lu C. Ore geology, fluid inclusion and 40 Ar/39 Ar geochronology constraints on the genesis of the Yingchengzi gold deposit, southern Heilongjiang Province, NE China. Ore Geol Rev. 2016;72:1022–36. 10.1016/j.oregeorev.2015.09.026.Suche in Google Scholar

[40] Chai P, Sun J, Hou Z, Xing S, Wang Z. Geological, fluid inclusion, H–O–S–Pb isotope, and Ar–Ar geochronology constraints on the genesis of the Nancha gold deposit, southern Jilin Province, northeast China. Ore Geol Rev. 2016;72:1053–71. 10.1016/j.oregeorev.2015.09.027.Suche in Google Scholar

[41] Steele-Macinnis M, Lecumberri-sanchez P, Bodnar RJ. HOKIEFLINCS_H2O-NACL: A Microsoft Excel spreadsheet for interpreting microthermometric data from fluid inclusions based on the PVTX properties of H 2 O–NaCl. Comput Geosci. 2012;49:334–7. 10.1016/j.cageo.2012.01.022.Suche in Google Scholar

[42] MacInnis MS, Sanchez PL, Bodnar RJ. H2O–NaCl, fluid inclusions, microthermometry, PVTX, isochores. Comput Geosci. 2018;49:334–7.Suche in Google Scholar

[43] Zhang L, Chen H, Liu C, Zheng Y. Ore genesis of the Saridala gold deposit, Western Tianshan, NW China: Constraints from fluid inclusion, S-Pb isotopes and 40 Ar/39 Ar dating. Ore Geol Rev. 2018;100:63–76. 10.1016/j.oregeorev.2017.06.011.Suche in Google Scholar

[44] Zhou Z, Chen Y, Jiang S, Hu C, Qin Y, Zhao H. Isotope and fl uid inclusion geochemistry and genesis of the Qiangma gold deposit, Xiaoqinling gold fi eld, Qinling Orogen, China. Ore Geol Rev. 2015;66:47–64. 10.1016/j.oregeorev.2014.10.020.Suche in Google Scholar

[45] He W, Yang L, Brugger J, Mccuaig TC, Lu Y, Bao X, et al. Hydrothermal evolution and ore genesis of the Beiya giant Au polymetallic deposit, western Yunnan, China: Evidence from fl uid inclusions and H–O–S–Pb isotopes. Ore Geol Rev. 2016;90:847–62. 10.1016/j.oregeorev.2016.10.035.Suche in Google Scholar

[46] Gleeson SA, Yardley BWD, Munz IA, Boyce AJ. Infiltration of basinal fluids into high-grade basement, South Norway: sources and behaviour of waters and brines. Geofluids. 2007;3:33–48.10.1046/j.1468-8123.2003.00047.xSuche in Google Scholar

[47] Goldfarb RJ, Groves DI. Orogenic gold: Common or evolving fl uid and metal sources through time. Lithos. 2015;233:2–26. 10.1016/j.lithos.2015.07.011.Suche in Google Scholar

[48] Goldfarb RJ, Groves DI, Gardoll S. Orogenic gold and geologic time: a global synthesis. Ore Geol Rev. 2001;18:1–75.10.1016/S0169-1368(01)00016-6Suche in Google Scholar

[49] Aliyari F, Rastad E, Goldfarb RJ, Abdollahi J. Geochemistry of hydrothermal alteration at the Qolqoleh gold deposit, northern Sanandaj – Sirjan metamorphic belt, northwestern Iran: Vectors to high-grade ore bodies. J Geochem Explor. 2014;140:111–25. 10.1016/j.gexplo.2014.01.007.Suche in Google Scholar

[50] Groves DI. Orogenic gold deposits: A proposed classification in the context of their crustal distribution and relationship to other gold deposit types. Ore Geol Rev. 1998;13:7–27.10.1016/S0169-1368(97)00012-7Suche in Google Scholar

[51] Yang L, Deng J, Li N, Zhang C, Ji X, Yu J. Isotopic characteristics of gold deposits in the Yangshan Gold Belt, West Qinling, central China: Implications for fl uid and metal sources and ore genesis. J Chem Explor. 2016;168:103–18. 10.1016/j.gexplo.2016.06.006.Suche in Google Scholar

[52] Chaumba JB, Mundalamo HR, Ogola JS, Cox JA, Fleisher CJ. Petrography, sulfide mineral chemistry, and sulfur isotope evidence for a hydrothermal imprint on Musina copper deposits, Limpopo Province, South Africa: Evidence for a breccia pipe origin? J Afr Earth Sci. 2016;120:141–59. 10.1016/j.jafrearsci.2016.05.003.Suche in Google Scholar

[53] Cox SF, Knacksted MA, Braun J. Principles of structural control on permeability and fluid flow in hydrothermal systems. Soc Econ Geol. 2001;11:1–24.10.5382/Rev.14.01Suche in Google Scholar

Received: 2021-09-21

Revised: 2023-05-10

Accepted: 2023-05-10

Published Online: 2023-09-28

This work is licensed under the Creative Commons Attribution 4.0 International License.

Artikel in diesem Heft

https://doi.org/10.1515/geo-2022-0494

Schlagwörter für diesen Artikel

fluid inclusion; gold mineralization; Kirk range; hydrothermal fluids

Creative Commons

BY 4.0