In Situ Analyses of Sulphides from the Tomingley Gold Project, Central-West NSW, Australia: Pathfinder Textures and Trace Elements

Abstract

This study investigated sulphide textures and trace element chemistry from the Tomingley Gold Project (TGP) region of Central-West NSW, eastern Australia, using in situ techniques. In particular, the study focused on pyrite and arsenopyrite to gain insights into ore-forming processes and determine which trace elements within these minerals can be used as potential pathfinder elements for mineral exploration in the TGP. A total of 41 drill core samples from a variety of lithologies (volcaniclastic, monzodiorite, graphitic siltstone, dacite, andesite) were described and analysed using reflected light microscopy, high-resolution microscopy (via Scanning Electron Microscope or SEM), elemental mapping (via Electron Probe Micro Analysis or EPMA) and targeted trace element analysis of sulphide grains (via Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry or LA-ICP-MS). Findings show that pyrite and arsenopyrite are the major sulphides that host fracture-fill/inclusions of native gold and ‘invisible gold’. Pyrite rich in groundmass inclusions should be evaluated due to their characteristic high concentrations of both As and Au. Pyrite trace element chemistry (Sn, Bi, W, Sb, Au and Se) was able to delineate mineralised from unmineralised samples in volcaniclastics, graphitic siltstones and andesites but was much more challenging for lithologies like dacites and monzodiorites. The study also found that Au may have been introduced into the system earlier and existed as ‘invisible gold’ in earlier generations of pyrite. This study highlighted the utility of in situ techniques to discriminate mineralised signatures from unmineralised samples, and this has proven to be far more effective compared to whole-rock techniques, emphasising the benefits of such datasets in mineral exploration.

1. Introduction

The Tomingley Gold Project (TGP) comprises a cluster of orogenic gold deposits and prospects within an Ordovician volcanic belt (Mingelo Volcanics) near the village of Tomingley, approximately 50 km southwest of Dubbo in the Central West region of NSW, eastern Australia [1,2,3]. The Peak Hill high-sulphidation epithermal deposit occurs in the same belt but is likely older and not genetically related to the orogenic deposits [4]. Modern mining of the orogenic Au deposits began in 2014 by Alkane Resources Ltd. and initially exploited the Wyoming One, Wyoming Three, Caloma One and Caloma Two deposits [5], followed by the more recently discovered San Antonio and Roswell (SAR) deposits [6].

Previous work on the TGP focused on regolith studies [7], the distribution of pathfinder elements around the deposits [8], the biogeochemical potential of native vegetation to detect the Au deposits [9,10,11], lithogeochemistry and alteration [12], paragenesis of mineralization [2], and thermodynamic modelling of ore formation [3]. Previous paragenetic studies [2,3] have only superficially examined the textures and mineral chemistry of the sulphides associated with Au mineralisation in the TGP. No previous studies have attempted to collect trace element compositional data in the context of sulphide chemistry and paragenesis. As a result, the relationship between Au and other trace elements (as potential pathfinders) in the various host lithologies of Au mineralisation in the TGP is still poorly understood. These problems present knowledge gaps within the TGP and form the basis of this study.

A wide variety of trace elements can be incorporated into pyrite through substitution within the lattice structure or as inclusions of other minerals. Elements such as Ni, Co, As, Se and Te can be incorporated into pyrite through direct substitution [13,14,15,16,17]. Nickel, Co and Mo substitute for Fe²⁺ whereas As, Se, and Te substitute for S²⁻ [16,17]. Antimony and Cu occur in the form of coupled substitution, where Sb and Cu and another element such as Au and Tl are substituted for 2Fe²⁺ [14,15]. Some elements like Pb have a larger ionic size than Fe or S, making it difficult to be incorporated into the lattice structure of pyrite, instead leading to the formation of mineral inclusions like galena [18,19]. Nanoparticles of a number of trace elements (As, Pb, Sb, Bi, Cu, Co, Ni, Zn, Au, Ag, Se and Te) may also be present in pyrite, in particular, hydrothermal pyrite [20]. Recent machine learning applications on extensive pyrite datasets have yielded important information on trace element systematics in orogenic gold deposits. For example, trace element occurrences in pyrite in orogenic gold deposits can generally be divided into four types: (1) Au-As; (2) Sb-Cu; (3) Ag-Pb-Bi; and (4) Co-Ni-Te-Se-Mo [21]. A study comparing pyrite trace elements in five different classes of gold deposits (porphyry, low/high-sulphidation epithermal and Carlin) by [22] revealed that orogenic Au deposits generally have the lowest average concentrations but with high variance in almost all of the trace elements tested (Au, As, Bi, Co, Cu, Ni, Pb, Sb, Se, Te, Tl and Zn). Studies of sulphides in the Macraes orogenic Au-W deposit in Otago, New Zealand, found that invisible gold within the lattice structure of pyrite had a strong positive correlation with As but weak correlations with Cu, Ag, Bi and Te and no obvious correlations with Pb, Zn or Ni [23].

This study aims to identify the sulphides that are present and characterise their textures and chemistry in Au-hosting lithologies in the TGP. Any potential relationship between Au mineralisation and sulphide texture and/or chemistry would then be explored for its vectoring potential. It will examine whether there are specific trace elements in the sulphides that could help discriminate between mineralised and barren zones and whether they correlate with whole-rock chemistry. This research will help to further understand the mineralising sequence through time and ore-forming processes related to the partitioning of elements in sulphides [24,25,26], specifically within the Tomingley gold deposits. This study will emphasise the application of sulphide chemistry and texture as a tool to explore for additional Au resources within the TGP. The findings of this study may have the potential to be applied more widely for the exploration of orogenic gold in other districts [27]. The main aims of this project were to:

• Identify sulphides and characterise major sulphide textures within the Au-hosting lithologies within the TGP.
• Investigate the relationship between Au mineralisation and sulphide textures and chemistry.
• Assess the capabilities of sulphide trace element chemistry as mineral exploration vectors.
• Compare and evaluate the effectiveness of data obtained from in situ and whole-rock techniques for mineral exploration vectors.

2. Geological Background

2.1. Regional Geology

The geology of Eastern Australia comprises a set of geologic terranes (the Delamerian, Lachlan, Thomson, and New England Orogens, known collectively as the Tasmanides) that were accreted to the Pacific margin of Gondwana during the Palaeozoic [28,29,30]. The Lachlan Orogen (LO) represents a remnant of a subduction-accretionary system that formed from the Cambrian to the Carboniferous (520 Ma–350 Ma). The main lithological units are metasedimentary and metavolcanic rocks such as quartz-rich turbidites, cherts, mafic-intermediate volcanics and younger continentally derived sequences, along with voluminous granitic intrusions [31]. These sequences are interpreted to have similarities with modern Pacific oceanic island arc and back-arc basin environments [31]. The LO can be subdivided into western, central and eastern subprovinces due to differences in rock types, metamorphic grade, structural features and geological setting [31].

The eastern subprovince of the Lachlan Orogen (Figure 1) includes four discontinuous volcanic belts. These four belts (Junee-Narromine, Molong, Rockley-Gulgong and Kiandra Volcanic Belts) collectively comprise the Macquarie Arc and are commonly interpreted to be the segmented remnants of an intra-oceanic island arc that developed off the coast of the early Australian continental margin [32]. While more recently recognised to have initiated during the Cambrian [33], the bulk of the igneous and volcaniclastic rocks was emplaced during the Middle-Late Ordovician to Early Silurian [34]. Subsequent deformation broke up and juxtaposed the segments of the island arc between multiple other terranes (e.g., Ordovician turbidite fans and Silurian metamorphic belts and basins) [35,36].

The Macquarie Arc in the Eastern Lachlan Orogen is best known for hosting porphyry Au-Cu (e.g., Cadia, Northparkes; [37]) and epithermal Au (±Cu) (e.g., Cowal, Peak Hill; [4]) mineralisation. Extending over 200 km from north to south, the Junee-Narromine Volcanic Belt (JNVB) is the longest and westernmost belt in the Macquarie Arc. The JNVB is dominated by Ordovician calc-alkaline to shoshonitic volcanic rocks, primarily andesites, dacites and basalts, intercalated with volcaniclastic and sedimentary units [38]. The JNVB hosts significant porphyry-epithermal deposits as well as younger orogenic Au mineralisation, the largest of which is at Tomingley [1,39].

2.2. Geology of the Tomingley Gold Project

2.2.1. District Scale Geology

The TGP sits near the eastern margin of the JNVB, centred on the Mingelo Volcanics, a ~40 km long and up to 1.5 km wide Ordovician belt of andesite and basaltic andesite, volcaniclastic sandstone and siltstone, and lesser intrusions and black mudstones (Figure 2) [1,40]. The Mingelo Volcanics trends N-S and has faulted contacts with the surrounding Late Ordovician-Early Silurian Cotton Formation (deep marine siltstones and conglomerates) [41]. The Mingelo Volcanics and Cotton Formation occur, along with several other narrow volcanic belts, in a 10–15 km wide zone of high strain called the Kiandra-Narromine Structure [38].

The rocks of the Kiandra-Narromine Structure have been metamorphosed to greenschist facies [42] and deformation manifests as a steep penetrative foliation with tight to isoclinal folding in places [1,38]. The volcanic belts and Cotton Formation deformed by the Kiandra-Narromine Structure are separated from the rest of the JNVB to the west by the Parkes Thrust, a major intra-belt fault, and an unnamed fault bounding the Narromine Igneous Complex (Figure 2). The volcanic belts and Cotton Formation are bounded to the east by Ordovician turbidites (Kirribilli Formation, Mugincoble Chert) and then rift-related Devonian volcanics (Dulladerry Volcanics) and Late Devonian Hervey Group quartz sandstones [42]. Devonian dolerite dykes that crosscut the Mingelo Volcanics are believed to be related to the rifting to the east that produced the Dulladerry Volcanics [43].

Figure 2. Geological setting of the Tomingley-Peak Hill-Parkes area. The orogenic gold deposits and prospects of the Tomingley Gold Project are marked as mine symbols and stars within the Mingelo Volcanics [44]. Note that the Peak Hill Gold Mine is a high-sulphidation epithermal gold deposit and is unrelated to the orogenic gold system in the same belt [45,46].

Figure 2. Geological setting of the Tomingley-Peak Hill-Parkes area. The orogenic gold deposits and prospects of the Tomingley Gold Project are marked as mine symbols and stars within the Mingelo Volcanics [44]. Note that the Peak Hill Gold Mine is a high-sulphidation epithermal gold deposit and is unrelated to the orogenic gold system in the same belt [45,46].

2.2.2. Deposit Geology

Multiple orogenic Au deposits and prospects occur within the Mingelo Volcanics, mainly towards the northern (more extensively explored) part of the belt. Most orogenic Au mineralisation appears to be associated with the rheological contrast between narrow (sub-100 m thick) lavas and shallow intrusive sills, and the surrounding volcaniclastic sedimentary rocks and graphitic mudstones [1]. Most of the deposits appear to be directly hosted by the igneous units or in the adjacent sedimentary rocks [5]. The TGP currently consists of the Wyoming One, Wyoming Three, Caloma, Caloma Two, and San Antonio and Roswell (SAR) deposits. Mineralisation in the Wyoming and Caloma deposits is hosted in a mix of shallow intrusions and volcaniclastic sedimentary rocks. Mineralisation in the SAR deposits is mainly hosted in andesite lavas, with only minor mineralisation in the surrounding volcaniclastics [41]. A summary table on the geological characteristics of the main deposits in the TGP is provided in

Table 1. Summary table of sites considered in this study and their characteristic mineralisation styles. MZD: monzodiorite, VCN: volcaniclastics, ANT: andesite, BAND: basaltic andesite, GSILT: graphitic siltstone, DAC: dacite, Py: pyrite, Apy: arsenopyrite, Qz: quartz, carb: carbonate, diss.: disseminated, selv: selvages.

Site (Bold = Au Bearing)	Deposit or Prospect	Lithologies Present	Mineralisation Style	Mineralisation Assemblage		Type of Samples Collected
				Sulphides	Other Minerals (Vein or Alteration)
Wyoming One	Deposit	VCN and GSILT	Stockwork-like vein system, poddy quartz veining, brecciated cap of main host intrusion	Py (diss., vein); Apy (vein)	Qz (vein)	Mineralised vein (Py+Apy) in MZD; Unmineralised (Py) in MZD
Wyoming One South	Prospect, no significant mineralisation	MZD	Narrow isolated sheeted veins	Cpy (vein), Py (diss., vein)	Qz (vein)	Mineralised (Py) in MZD
San Antonio-Roswell	Deposit	MZD, ANT, DAC, and VCN	Stockworks, breccias	Py (vein); Apy (vein)	Qz-carb (vein, altered)	Unmineralised (Py) in MZD; Mineralised vein and non-vein (Py+Apy) in ANT; Unmineralised (Py+Apy) in ANT; Mineralised vein and non-vein (Py+Apy) in DAC; Unmineralised (Py+Apy) in DAC; Unmineralised (py) in VCN
Plains	Prospect; no significant mineralisation	BAND	Not applicable	Py (diss.); Apy (diss.)	Qz-carb (vein, altered)	Mineralised (Py) in BAND
Caloma 2	Deposit	MZD and VCN	Sheeted quartz veins, en echelon veins, potential saddle reef structure	Py (vein, selv); Apy (vein)	Qz-carb (vein, altered)	Mineralised (Py) in GSILT
Tomingley Two	Prospect; no significant mineralisation	VCN and GSILT	Not applicable	Py (diss., vein, selv); Apy (vein)	Qz (vein, altered)	Mineralised (Py) in VCN and GSILT
Smiths/Cotton	Prospect, Unmineralised	VCN	Not applicable	Py (diss., selv)	Qz (vein)	Unmineralised (Py) in VCN

. Further details on known mineralisation characteristics are provided in Supplementary S4.

The volcaniclastic sedimentary rocks (well-bedded sandstones and siltstones with minor breccias, lithic conglomerates and black mudstones) mainly comprise plagioclase ± augite grains and/or lithic fragments; detrital quartz is rare [1]. The lavas (mainly around Roswell and San Antonio) usually have andesite to basaltic-andesite compositions [47] and can have similar chemistries to the intrusive sills (at Wyoming and Caloma) [12]. The andesites predominantly consist of minute tabular crystals of feldspars in the groundmass, forming a trachytic texture [47]. The intrusive porphyrytic (phyric) bodies usually have coarse feldspar ± augite crystals and trachy-andesitic to mafic trachy-andesite affinities with rare peperitic textures, suggesting they were shallowly emplaced [1]. The regional metamorphic alteration assemblage is typified by chlorite and epidote alteration. Hydrothermal alteration overprints the regional assemblage with sericite, carbonate, albite and silica [41].

3. Materials and Methods

3.1. Sample Selection and Rationale

A total of 41 drill core samples were obtained from 17 drill holes. The drill core samples were obtained from multiple deposits and prospects, including: (a) Wyoming 1; (b) Wyoming 1 South; (c) San Antonio and Roswell; (d) Plains; (e) Caloma 2; (f) Tomingley 2; and from the barren Cotton Formation at the (g) Smiths prospect. We targeted drill core samples with different rock types (e.g., monzodiorite, graphitic siltstone, dacite, andesite, and volcaniclastic;

Table 2. Sample count for each rock type.

Rock Type	Number of Samples Obtained
	Mineralised	Unmineralised
Monzodiorite	3	9
Graphitic siltstone	3	3
Volcaniclastics	4	4
Andesite	9	3
Dacite	1	2

). Samples were also characterised based on the degree of mineralisation (mineralised and unmineralised). Previous whole-rock analyses and a previous honours project [3] on the characterisation of the host lithologies informed sampling. The degree of mineralisation of samples was ascertained using whole-rock Au assay data. Whole-rock Au assay values less than and greater than 0.1 g/t were classified as unmineralised and mineralised, respectively. The uneven number of samples was due to limitations on available materials in the sample collection for the project.

3.2. Microscopic and Analytical Techniques

Hand specimens were initially examined for physical descriptions (i.e., drill core lithology, alteration assemblages and their characteristic sulphide morphologies). Areas with significant sulphides were marked for polished mount preparation (30 mm diameter polished epoxy solid cylinders). These polished resin blocks were investigated under reflected light microscopy (RL), scanning electron microscopy (SEM), electron probe microanalysis (EPMA) and laser-ablation inductively coupled plasma spectrometer (LA-ICP-MS).

Reflected light microscopy used the Leica DM2500 P (Leica Microsystems, Wetzlar, Germany), equipped with a 5-megapixel CCD sensor, live image speed of 18 frames per second and a SXGA resolution of 1280 × 960 pixels. Photomicrographs were captured using a Leica MC170 digital microscope camera (Leica Microsystems, Wetzlar, Germany) attached to the microscope at UNSW Sydney. Parameters used included captured format set to highest resolution (2592 × 1944), and varying magnification scales (2.5 μm, 10 μm, and 40 μm). Textures that could not be resolved using standard microscopy techniques (smaller than 40 μm) required a higher resolution microscopy. Such textures were further explored using SEM and EPMA. Nine polished resin blocks were chosen for SEM, while three polished blocks were targeted for elemental mapping using the EPMA to target specific questions. Both techniques required the polished blocks to be carbon-coated with a thickness of ~20–25 nm to enhance sample conductivity. These investigations were performed at the Electron Microscope Unit (EMU) at the Mark Wainwright Analytical Centre in UNSW Sydney.

Scanning Electron Microscopy (SEM) was done using a Hitachi TM4000Plus (Hitachi High-Tech Corporation, Tokyo, Japan) equipped with a Bruker X-Flash 630Hc energy-dispersive X-ray spectroscopy (EDS) detector (Bruker Corporation, Billerica, MA, USA). It was operated with an accelerating voltage of 15–20 kV in high vacuum mode. Data acquisition and preparation were done using Bruker Esprit Compact software (ver. 2.7). Electron probe microanalysis (EPMA) qualitative elemental distribution mapping was done using a JEOL JXA- 8500F Hyperprobe (JEOL Ltd., Tokyo, Japan) equipped with four wavelength-dispersive spectrometers (WDS) and a JEOL EDS detector. Mapping was conducted at an accelerating voltage of 20 kV and current of 40 nA, with a pixel/step size of 1–2 μm. WDS and EDS X-ray peak analysis positions were optimised using the pyrite samples themselves for S and Fe and using in-house high-concentration standards for Co (cobaltite), Ni (heazlewoodite), As (arsenopyrite) and Mn (Rhodonite). X-ray lines mapped include Kα of S, Fe, Co, Ni, Lα of As via WDS, and Kα of Mn via EDS.

Laser Ablation-Inductively Coupled Plasma Mass Spectrometer (LA-ICP-MS) spot analyses and imaging were carried out at CODES Analytical Laboratories, University of Tasmania. Analyses were conducted on a Resolution LR laser ablation system with a Coherent COMPexPro 110 Excimer ArF laser (Coherent Corp. Saxonburg, PA, USA) operating at 193 nm wavelength and pulse width of ~20 ns coupled with an Agilent 8900 quadrupole mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). The sample ablation was performed in a He atmosphere flowing at 0.35 L/min and immediately combined with Ar flowing at 1.05 L/min. A circular laser beam of 19 µm was used for spot analyses of arsenopyrite, while a 29 µm spot size was used for pyrite. A square laser beam of 7 µm was used for imaging of pyrite. Imaging of the whole area of the grain was covered by a raster of parallel lines with a rastering speed of 7 µm/s. The following masses were measured: 23Na, 24Mg, 27Al, 29Si, 34S, 39K, 43Ca, 49Ti, 51V, 53Cr, 55Mn, 57Fe, 59Co, 60Ni, 65Cu, 66Zn, 75As, 77Se, 88Sr, 90Zr, 107Ag, 111Cd, 121Sb, 125Te, 157Gd, 178Hf, 181Ta, 195Pt, 197Au, 202Hg, 205Tl, 208Pb and 209Bi. Total sweep time was equal to 0.65 s for spot analyses and 0.26 s for imaging. STDGL3 [48] and GSD-1G [49] glasses analysed at 51 µm were used as a primary standard for quantification and instrument drift correction. Quantification was performed using 57Fe as the internal standard element with analyses normalised to 100 wt.% total using calculated sulphur. Ablation was carried out using a laser frequency of 5 Hz for spot analyses and 10 Hz for imaging. A laser beam energy density of approximately 2.7 J/cm² was used for STDGL3 and sulfides and 3.5 J/cm² for GSD-1G glass. “Squid” smoothing device was used as an interface between the laser ablation system and ICP-MS for spot analyses, while straight nylon tubing 3 m long was used for imaging. LADR software version 1.1.07 (Norris Scientific) was used for quantification of spot analyses and to determine cps/ppm conversion factors for imaging. Final processing for the images was done with an in-house developed Python script. Visualisation of TE data was done using the ioGAS software (ver. 8.3.) —an industry-standard software for geochemical data.

4. Results

4.1. Drill Core Observations

A total of 41 drill core samples were examined and described for observed rock type, alteration assemblages and sulphide morphology. The following subsection describes representative features of mineralised and unmineralised samples from each rock type.

Monzodiorite (MZD)

An example of mineralised monzodiorite comprises intense sericite-carbonate-albite alteration with multi-generational quartz veins (Figure 3a), which have collectively destroyed the original rock fabric. Pyrite distribution within the sample was observed to be strongly controlled by a quartz vein, with pyrite mainly occurring in the selvage around the vein. Figure 3b shows a light to dark green unmineralised monzodiorite unit with tabular plagioclase (average = 0.4 cm) phenocrysts in a fine-grained groundmass. Pyrite and arsenopyrite occur in a quartz-chlorite-sericite veinlet that cuts through the sample. Fine disseminated pyrite was also observed. Less intense sericite-carbonate alteration pervades the sample.

Graphitic siltstone (GSILT)

The sample shown in Figure 4a is of an intensely altered mineralised graphitic siltstone. This unit was characterised by its bleached appearance and differently oriented quartz veins. Pyrite was the primary sulphide and usually occurs as coarse grains in quartz veins. Minor arsenopyrite was present, though intergrown with pyrite. A wedge of unmineralised graphitic siltstone (Figure 4b) was collected from an interval within a sheared monzodiorite unit with multiple parallel quartz-carbonate veins. The wedge of dark-coloured siltstone may have been sheared into the monzodiorite or could represent some of the surrounding country rock. Within the marked zone, pyrite banding was observed overprinting a carbonate vein and a pyrite-rich vein was also present.

Andesite (ANT)

The sampled mineralised andesite (Figure 5a) has been brecciated and infilled with quartz-carbonate and intense silica alteration. Coarse-grained arsenopyrite (1 cm) is intergrown with minor pyrite and sits within the quartz veins with brecciated textures. The selected unmineralised andesite (Figure 5b) sample has preserved the original igneous texture and has only a relatively weak chlorite-sericite alteration affecting the groundmass and feldspar phenocrysts. Pyrite-rich veinlets cut through the andesite.

Dacite (DAC)

The mineralised dacite sample (Figure 6a) has a patchy, sericite-dominated alteration assemblage but with the porphyritic igneous texture still preserved. Two differently oriented veinlets cut through the unit; an earlier milky quartz vein is cut by a dark green veinlet. Pyrite was the primary sulphide, and the grains vary from coarse- to medium- grained. Much of the pyrite appears to be controlled by veinlets. Lesser medium-grained arsenopyrite was also present. The unmineralised dacite (Figure 6b) was less intensely altered (mainly chlorite-sericite), depicted by the muted greenish grey colour. An irregular quartz-carbonate veinlet cut through the dacite. The unit has mostly pyrite with minor arsenopyrite. Pyrite occurs in the vein sparingly, but pyrite and arsenopyrite disseminations occur throughout the sample.

Volcaniclastics (VCN)

The selected mineralised volcaniclastic sample (Figure 7a) is strongly carbonate-sericite altered (indicated by the brownish grey colour). A quartz vein cuts the sample. The main sulphide was pyrite, followed by minor arsenopyrite, both of which were disseminated throughout the rock mass. The selected unmineralised volcaniclastic sample (Figure 7b) is a light to dark greenish grey and comprises interbedded sandstone-siltstone with pervasive epidote-chlorite alterations. Pyrite in this unit mostly occurs within the bands of pale-green epidote.

4.2. Sulphide Textural Observations

All 41 samples were examined for textural characterisation using reflected light microscopy and the documented textures have been compiled in Supplementary Information (S8). This includes a spreadsheet with a list of the samples, including drill hole number, depth, and the texture types observed in each sample. The dominant sulphides in the orogenic gold deposits of the TGP are pyrite (FeS₂) and arsenopyrite (FeAsS). Pyrite occurs in all samples in variable amounts and forms (e.g., anhedral-subhedral, euhedral, etc.). Arsenopyrite (e.g., bladed, acicular, etc.) was less common but typically occurs in association with pyrite. Other sulphide minerals observed in this study include chalcopyrite (CuFeS₂), sphalerite (ZnS), galena (PbS), and, rarely, pyrrhotite (Fe₇S₈). In the following subsections, the dominant textures of major sulphides are described.

Pyrite

Framboidal pyrite: Framboidal pyrite was the least common type of pyrite in this study and was only found in two graphitic siltstone samples. The framboids usually occur as relicts, overgrown by a later pyrite (Figure 8a,b). In some instances, single clusters of framboids were also found to be disseminated within the siltstone matrix (Figure 8c,d).

Anhedral-subhedral pyrite: Anhedral and subhedral pyrite textures were the most common occurrence among all pyrite textures. This texture usually occurred as disseminations in the groundmass, in quartz veins (Figure 9a) or concentrated just outside quartz veins (in the selvage or alteration halo). Grain sizes ranged from ~10 µm to ~1 mm. This type of pyrite also formed aggregates (Figure 9b) and commonly had base metal sulphide (i.e., chalcopyrite, sphalerite, galena) inclusions or fracture-fill.

Euhedral-subhedral pyrite: The texture of euhedral pyrite in the TGP varied from ‘glossy’ (inclusion-poor) to inclusion-rich (typically non-reflective silicate phases). Euhedral-subhedral pyrite mainly occurred as cubic or pyritohedron shapes, sometimes aggregated into clusters (Figure 10a), disseminated isolated grains (Figure 10b) and fine- to coarse-grains (~1 µm to ~4 mm), in quartz veins (Figure 10c). Fractures and caries (voids within a sulphide grain with irregular dissolution/corroded shapes) were common. Visible Au was rarely associated with this type of pyrite.

Core-rim pyrite: Pyrite crystals with distinct core-rim zonation, visible under reflected light, were found in the mineralised dacite and volcaniclastic samples. These multi-generational pyrite crystals typically had wider cores (~150 µm diameter) and narrower rims (<50 µm thick) with the boundary often marked by trails of inclusions (Figure 11a–c). This type of pyrite was investigated further for trace element differences between its core and rim using LA-ICP-MS spot analyses and compositional mapping under the EPMA. Results will be discussed later (Section 4.4).

Minor pyrite textures: Minor pyrite textures included differently shaded pyrite grains intergrown with each other, distinguished due to the lighter and darker contrast under reflected light. This pyrite was investigated further using BSE imaging and EPMA to observe any compositional relationship that possibly caused the colour contrast (see Supplementary S5). Other textures, including filled fractures/inclusions of Au and core-rim and framboidal pyrites, were investigated further via BSE imaging to confirm observations made under reflected light microscopy, details of which are included in (see Supplementary S5).

Arsenopyrite

Arsenopyrite grains usually occurred as acicular, bladed or prismatic crystals and had sizes ranging from 5 µm to 1 mm. In this study, arsenopyrite crystals were observed both in and outside of quartz veins. Arsenopyrite in quartz veins usually occurs individually as bladed or acicular crystals but is spatially associated with pyrite. Arsenopyrite grains were observed overgrown by later pyrite and even occasionally partially replaced by pyrite (Figure 12).

Other sulphides

Other sulphides observed were sphalerite, chalcopyrite, pyrrhotite and galena, and some (galena, chalcopyrite and sphalerite) were texturally associated with Au (Figure 13a–d). Sphalerite, galena and chalcopyrite occurred as free grains in the host rock and as inclusions or more commonly as fracture-fill phases (Figure 13a–c). Galena also occurred as inclusions in pyrite/arsenopyrite but less commonly compared to sphalerite and chalcopyrite. In some instances, galena was observed to be associated with gold. Pyrrhotite was the rarest among all the sulphides and was observed to be in association with pyrite and chalcopyrite (Figure 13a).

Au-sulphide association

In the mineralised volcaniclastics (n = 4), visible Au was observed in two samples. Visible Au and chalcopyrite co-occurred together by infilling fractures in euhedral pyrite (Figure 14a). Some Au was visible as inclusions within aggregated anhedral pyrite but without any associated chalcopyrite (Figure 14b).

Among mineralised monzodiorite samples (n = 3), two contained visible Au, which was associated with pyrite as fracture-fill (Figure 15a,b). Gold was rarely observed in the mineralised graphitic siltstone (n = 1/3), occurring in a fracture with sphalerite along a fracture (a) and the edge of an anhedral pyrite grain (Figure 16b).

Four mineralised andesite samples were observed to contain visible Au. The Au particles typically occurred as inclusions or fracture-fill in anhedral aggregated pyrite (Figure 17a,b). Gold was also present as free grains at the edge of bladed arsenopyrite grains in sample RWD016 @ 265.7 m (Figure 17c,d). Visible Au in mineralised andesite was not observed in association with base metal sulphides.

Gold was also observed associated with galena, as shown in Figure 18a–c, in a few samples, implying that pre-existing sulphides played a role in precipitating Au. In this instance, it is likely that an electron transfer process from galena to Au atoms aided in the stabilisation of the Au (0) [50].

4.3. Trace Element Compositional Mapping (EPMA and LA-ICP-MS)

Two mineralised samples (a dacite and volcaniclastic) with core and rim textures (visible under reflected light) were further investigated to determine their composition (Figure 19, Figure 20 and Figure 21) and potential zoning owing to varying compositions. Elemental mapping using EPMA was also done on shaded (light and dark) pyrite (see Supplementary S5). A mineralised dacite sample (RWD016-273.9 m) with a core and rim pyrite texture showed elevated As in the core relative to the rim. Cobalt and Ni appeared to be slightly elevated in the core but both (including Mn) generally have moderate concentrations throughout (Figure 19).

Pyrite from a mineralised volcaniclastic (WY831D0-166.8 m) appeared to be homogenous in Fe and S, but subtle zonation in Co and Ni was evident in Figure 20 and Figure 21, mainly within the complex-As-zoned cores. The innermost pyrite in Figure 20 and Figure 21 appeared to be slightly elevated in As, followed by periods of depleted As. The euhedral-shaped As-poor core was bounded by As-rich rims.

Using a LA-ICP-MS imaging technique, an unmineralised graphitic siltstone sample with relict framboidal pyrites was scanned to observe differences in trace element composition (Figure 22). Trace element distribution enabled determining different pyrite generations (py 1, py 2, py 3, py 4, py 5). The relict framboids (py 1) appeared enriched in Au, As, Mn, Mo, Co, Cu, Pb, Tl, Sb, Mo, Co, Ag and Mn. Py 2 was enriched in As, Mn and Mo. Py 3 was enriched in Tl, Sb and Mo along with high As and Co towards the later stage of py 3. Py 4 was high in Ni and As rimmed by a later Co enrichment. This was followed by py 5 with high Ni rimmed with another late-stage Co enrichment.

4.4. Trace Element LA-ICP-MS Spot Data for Pyrite and Arsenopyrite

A total of 402 LA-ICP-MS spot analyses were conducted on pyrite (n = 372) and arsenopyrite (n = 100). In the following sections, comparative boxplots were used to show the distribution of pyrite TEs (Mn, Co, Ni, Cu, Zn, As, Ag, Se, Mo, Cd, Sn, Sb, Te, W, Au, Tl, Pb, Bi) in mineralised and unmineralised samples within each rock type. A complete comparison between mineralised and unmineralised samples for arsenopyrite TEs was only possible for the dacites and monzodiorites. Therefore, arsenopyrite TEs for other rock types will not be discussed.

Boxplots represent the interquartile range (IQR) from the 25th to the 75th percentile, with lines extending below and over the box towards the maximum and minimum. The IQR of both the mineralised and unmineralised samples were compared relative to each other to determine their enrichment and depletion in trace elements. We refer to enrichment when the mean/median values of mineralized samples are higher than their unmineralized counterparts and vice versa.

Volcaniclastics: Only As, W and Au were elevated in the mineralised volcaniclastics (Figure 23). Most other elements (Mn, Co, Ni, Cu, Zn, Se, Mo, Ag, Cd, Sn, Sb, Pb and Bi) appear to be depleted in the mineralised samples (Figure 23). Significantly different elements (W, Au, Mn, Cu, Zn, Mo, Sb, Pb, and Bi) could potentially be used as pathfinder elements specifically for volcaniclastic samples.

Monzodiorite: The IQR of Mn, Co, Ni, Cu, Zn, Se, Mo, Cd, Sn and Bi were generally lower in the ‘mineralised’ pyrite relative to the ‘unmineralised’ pyrite (Figure 24), whereas the boxes for As, Ag, Sb, W, Au, and Pb in ‘mineralised’ pyrite were higher than ‘unmineralised’ pyrite (Figure 24).

In the arsenopyrite from the monzodiorite samples, Mn, Zn, Mo, W, Tl, Pb, and Bi were enriched in ‘mineralised’ (Figure 25) compared to ‘unmineralised’, but were depleted in Co, Se, Sn, Sb, and Au (Figure 25).

Graphitic siltstone: Pyrite from mineralised graphitic siltstone samples has higher concentrations of Se, Te, W, Au, Pb, and Bi (Figure 26) but lower concentrations of Mn, Co, Zn, Mo, Cd, Sn, Sb, and Tl (Figure 26) relative to unmineralised samples.

Dacite: Cobalt and Se were the only two elements in pyrite that were elevated in the mineralised dacites (Figure 27), and these were depleted in Mn, Ni, Sn, Sb, Pb, Au, and Bi relative to unmineralised dacites (Figure 27). However, arsenopyrite was enriched in Se, Ag, Cd, Te, W, Au and Tl in the mineralised samples (Figure 28) but was depleted in Mo, Sb, and Bi (Figure 28) when compared to the unmineralised samples.

Andesite: Mineralised andesite appears to be elevated in As, W, Au and Bi (Figure 29) while depleted in Mn, Co, Cu, Zn, Mo, Cd, and Pb (Figure 29) when compared to unmineralized andesite.

5. Discussion

5.1. Textures and Their Trace Element Composition

In this subsection, we will explore sulphide textures in the mineralised and unmineralised equivalents of each of the rock types investigated in this study (

Table 3. Summary table of textural trends present in mineralised and unmineralised samples across all rock types.

Rock Type	Texture Occurrences
	Mineralised	Unmineralised
Volcaniclastics	Euhedral to anhedral pyrite Abundant arsenopyrite Abundance of chalcopyrite and sphalerite inclusions	Euhedral to anhedral pyrite Less arsenopyrite Less chalcopyrite and sphalerite inclusions
Monzodiorite	Abundant arsenopyrite Abundant sphalerite, and rare chalcopyrite and galena	Less arsenopyrite Less sphalerite, and rare chalcopyrite and galena
Graphitic siltstone	Framboidal pyrite, fractured pyrite Abundant fracture-fill sphalerite and chalcopyrite	Framboidal pyrite, porous pyrite Less sphalerite and chalcopyrite inclusions
Dacite	Anhedral to subhedral pyrite (diss., caries) Arsenopyrite present Sphalerite inclusions/fracture-fill	Anhedral to subhedral pyrite (diss., caries) Arsenopyrite present Chalcopyrite free grains and galena inclusions
Andesite	Euhedral to anhedral pyrite Abundant sphalerite, chalcopyrite and galena inclusions in pyrite	Euhedral to anhedral pyrite Less sphalerite and rare chalcopyrite and galena inclusions in pyrite

). This was mainly done to observe if certain sulphide textures were related to Au mineralisation.

Core and rim pyrite: Some pyrite textures exhibited well-developed core and rim textures that were easily distinguishable under reflected light. In the mineralised volcaniclastic sample (WY831D—166.8 m), laser ablation spot analyses on the porous core and ‘clean’ rims of coarse euhedral pyrite (Figure 30a) showed significant differences in Au values between the two (Figure 30b). Gold was enriched in the porous core by at least two orders of magnitude compared to the ‘clean’ rim. This relationship was also reflected in the elements Cu, As, Se, and Te. This type of textural relationship between a porous pyrite core surrounded by a homogeneous pyrite rim has been frequently documented in orogenic gold systems globally and is interpreted to record coupled dissolution-reprecipitation reactions in sulphide minerals [51,52].

Anhedral and euhedral pyrite: Anhedral and euhedral pyrite textures in the same sample were compared to see if they exhibited similar or different trace element signatures. Anhedral pyrite was enriched in Mn, Co, Ni, Cu, Zn, Se, Mo, Cd, Sb, Pb and Bi relative to euhedral pyrite but no differences were observed for Au and As concentrations (Figure 31).

While some textures were characteristic of mineralised samples across the five rock types, we also observed many similarities between mineralised and unmineralised samples. Based on our investigations, we discuss specific textures that should be targeted for future research as they may be useful for exploration. For instance, the presence of anhedral-subhedral-euhedral pyrites was generally consistent between mineralised and unmineralised samples across all rock types studied, suggesting that these pyrite textures were widespread within the TGP. Slight differences, such as the intensity of fractures, are more likely to be attributed to local events/host rock/sites. However, as explained above, while there were similarities in the textures observed, the trace element compositions of most textures can be used to differentiate between mineralised and unmineralised samples. We recommend targeting all texture types (euhedral/subhedral/anhedral) to account for the trace element complexities and a minimum of 10 spot analyses for each sample. Secondly, base metal sulphide inclusions were a common observation in the mineralised samples but were rare in the unmineralised samples. Ample base metal sulphide inclusions in mineralised samples suggest variation in hydrothermal fluid composition compared to the unmineralised samples and may be related to Au mineralisation.

Finally, an inclusion-rich (porous) texture was observed to be enriched in Au and As concentration in some of the mineralised volcaniclastic and graphitic siltstone samples. Such observations have been previously reported to contain significant Au [24] and As concentration [53], especially grains with pores, microfractures and non-sulphide inclusions compared to less porous pyrite/arsenopyrite. The high Au concentration in the porous pyrite cores texture observed in this study also highly suggests that some of the Au may have been introduced into the system early and then later remobilised. It is important to note that higher Au concentrations are only found in the cores or inclusion-rich zones of the sedimentary units (volcaniclastics and graphitic siltstones) and not the igneous units. This suggests an earlier sedimentary origin for the Au, i.e., Au-rich hydrothermal fluid activity during the deposition of these sedimentary units, which were later remobilized. The LA-ICP-MS trace element imaging also strongly suggests multiple generations of pyrite with varying compositions in response to the hydrothermal fluid pulses. Trace element imaging in the igneous host rocks should be further explored in detail for TGP-type deposits to note if different generations of pyrite are present.

5.2. Pyrite Association with Mineralisation

Paragenesis refers to the sequence of formation of mineral assemblages. These mineral assemblages can be attributed to minerals of interest (related to ore deposits) or events like hydrothermal alteration or veining. A paragenetic sequence within the TGP was first made by [2] and built upon by [3] as seen in Figure 32.

Cherry (2013) [2] interpreted major sulphides like pyrite and arsenopyrite to occur before the main Au mineralisation event. This was mainly because most Au was seen infilling late fractures in pyrite and arsenopyrite and was typically associated with later chalcopyrite and sphalerite. Cherry (2013) [2] further investigated Au remobilisation at the TGP and proposed deformation as the main driver. Deformation causes fracturing, which creates pathways for fluids, leaching trace elements from the pyrite and carrying mobilised trace elements. These fractures become nucleation sites for Au precipitation due to rapid pressure fluctuations. Such mechanisms have also been observed in other types of Au deposits, such as Carlin [54] and sediment-hosted deposits [55]. Cherry (2013) [2] also concluded that Au did not occur as invisible gold in the pyrite and arsenopyrite grains, but instead that Au was brought in by later hydrothermal fluids. This interpretation was not consistent with the two-stage Au mineralisation mechanisms in orogenic deposits [56,57], where the first stage of gold mineralisation comprised invisible gold incorporated during initial growth zones of pyrite or arsenopyrite, and the second stage involved a later event of visible gold precipitation as fracture-filling and rims around pyrite-arsenopyrite grains.

However, results from this study suggest that not all of the Au-rich pyrites were post- Au mineralization. As per this study, Au was observed in the TGP as native Au inclusions and fracture-fill within and associated with pyrite and arsenopyrite. Plotting the pyrite laser spot data on the Au solubility curve of [58] suggests that any invisible Au is likely hosted within the pyrite lattice and not as micro- or nano-inclusions (Figure 33). The gold solubility line (black) is from [58]. Data points below the gold solubility line occur as solid solution (Au+) in the pyrite lattice structure, while those above likely exist as gold nanoparticles (Au0). This is consistent to some extent with orogenic gold systems in metasedimentary rock sequences, which typically contain Au as “invisible gold” via coupled Au-As redox reactions within As-rich sulphides [59]. In metavolcanic rocks, however, Au predominantly manifests as polymetallic (Au-Ag-Te-Bi) inclusions within As-poor sulphides, indicating distinct fluid mechanisms for gold precipitation [60].

Other evidence of early Au mineralization includes the following but should be corroborated using future studies:

• Spot analyses (LA-ICP-MS) comparison between porous pyrite core and inclusion-poor rims showed significantly higher Au concentrations in the cores (9.93–84.01 ppm) compared to the rims (0.028–0.029 ppm) in the metasedimentary units of this study.
• Gold has a strong association with As in the mineralised volcaniclastic samples (r = 0.87). This brings significance to the slightly elevated As subhedral-shaped core observed (Figure 20 and Figure 21), suggesting early incorporation of Au.
• The LA-ICP-MS trace element image of a pyrite grain with framboids (Figure 22) also showed high Au concentration in framboidal pyrite, suggesting Au incorporation during the formation of sedimentary pyrite. Sedimentary pyrite from black shales has been suggested to be a likely source of Au [61,62]. Given the various host lithologies in the TGP, graphitic siltstone could be one of many lithologies to contribute as the source of gold.
• Based on trace element patterns, distinct discrimination between mineralised and unmineralised samples suggested that mineralisation was associated with specific elements, not limited to Au. These pathfinder elements may vary depending on their rock types.

5.3. Pathfinder Elements for Each Rock Type

‘Pathfinder’ elements refer to trace elements in pyrite that could indicate the presence of an economically significant mineral or deposit. In this context, trace elements that showed a considerable difference in the average values of trace elements between mineralised and unmineralised samples could be potential pathfinder elements to indicate mineralisation. The purpose of this subsection is to explore the effectiveness of a few potential pathfinder elements in differentiating mineralised from unmineralised samples using binary plots and a multivariate statistical analysis (the Mahalanobis ellipses) built into the ioGAS software.

The Mahalanobis ellipses represent the mean and variance of a set of data (e.g., mineralised samples, unmineralised samples). These ellipses (p = 0.5, 0.95, 0.99) can have different orientations and sizes, which could inform on the correlation between variables (e.g., different trace elements) and the variance of data along different dimensions, respectively. No overlap between mineralised and unmineralised ellipses suggests that the geochemical signatures are distinct from each other within reason. Within each group, narrower and more elongated ellipses indicate stronger correlation and lower variability, whereas broader and more circular ellipses represent weaker correlation and greater variance. In this pilot study, we have used the Classical Mahalanobis ellipses with a confidence region of p = 0.5 (50% confidence level) that has led to the successful characterisation of mineralized vs. unmineralized samples. However, a higher sample size and no. of analyses will allow for further statistical treatment and higher confidence intervals (p = 0.90 or 0.95). The following subsections only discuss volcaniclastics, graphitic siltstone and andesite, as the pyrite data looked promising. Data on dacite and monzodiorite can be found in the Supplementary S6.

Volcaniclastics: Potential pathfinder elements for volcaniclastics include W, Au, Mn, Cu, Zn, Mo, Sb, Pb, and Bi as there were reasonable differences (for p = 0.5) in IQR between mineralised and unmineralised samples. As an example, two trace elements (W and Au) were chosen. Gold generally increased with increasing Bi for both groups but suffered from wide ellipses (Figure 34). Despite this, both ellipses were geochemically distinct from each other. There was a stronger relationship for mineralised pyrite in W-Sb and a negative correlation for unmineralised samples, but both ellipses seem to overlap slightly. For ratio biplots between Au/Bi and W/Sb, ‘mineralised’ samples were able to pull out from ‘unmineralised’ samples but suffered from more circular ellipses (Figure 34). Nonetheless, the elongated ellipse in unmineralised pyrite suggested less variance and could still be used to differentiate from mineralised pyrite.

Graphitic siltstone: Bismuth and Se were used as an example for graphitic siltstone. Selenium biplots against Sn revealed two distinct As populations in the unmineralised samples (e.g., Figure 35), which indicated that there is additional complexity in the pyrite of this lithology and it may not be suitable to discriminate between mineralised and unmineralised pyrites. The mineralised pyrites, however, were less varied. Both Bi and Sb generally increased with increasing Au concentrations, with the mineralised sample having better confidence (Figure 35). Nonetheless, the ratios of Bi/Sb and Se/Sn were able to further discriminate between mineralised and unmineralised clusters (Figure 35).

Andesite: Manganese and As were good pathfinder elements for delineating mineralised and unmineralised samples. There was a strong positive relationship between Bi and Pb for both mineralised and unmineralised samples (Figure 36). An overlap caused difficulty in differentiating mineralised from unmineralised samples, but a degree of clustering can still be observed from both mineralised and unmineralised samples. This was similar between Cu and Bi (Figure 36), but the ellipses were much larger, indicating a weak positive relationship.

A binary plot using As, Mn and Mo only revealed a moderate positive relationship for the unmineralised sample, but the tight ellipses of ‘unmineralised’ pyrite suggest a strong confidence in the data. Although plotting ratios of these elements (e.g., Figure 36) enabled us to pull out between mineralised and unmineralised samples, the ellipses were quite big, indicating that precaution must be exercised when using these ratios.

5.4. Implications for Exploration

In the previous section, specific elements were shown to be able to differentiate between mineralised and unmineralised samples for some of the rock types. The dacite and monzodiorite samples were found to be more challenging to discriminate than andesite, graphitic siltstone and volcaniclastic lithologies. We speculate that this disparity is owing to the potential of the lithologies to be a Au source. The inclusion-rich cores (Au) of pyrites in the sedimentary host rocks imply that while there may have been other sources of gold, an earlier sedimentary origin is highly likely. In contrast, based on analyses from this study, the pyrites in the dacite and monzodiorite do not suggest that they may have been a likely source of Au. The main control appears to be the brittle rheology of the igneous host rock rather than its composition (or pyrite chemistry). The following subsection mainly focuses on validating the reliability of LA-ICP-MS data by comparing the technique with the whole-rock assay technique. Two examples from volcaniclastics are shown below.

Binary plots between W and Bi (Figure 37) revealed that mineralised and unmineralised samples from both LA-ICP-MS and whole-rock techniques showed a clear separation. The main difference lies in element pair correlation, which was weak but clear between mineralised and unmineralised using the LA-ICP-MS technique. There was no trend observed for both element pairs in the whole-rock data. For W-Sb, the mineralised and unmineralised samples had a slight overlap using the LA-ICP-MS technique, but it was statistically significant in the whole-rock technique (Figure 38). Similarly, relationships between element pairs were much clearer for data using LA-ICP-MS compared to whole-rock. The reason for this may be that whole-rock assays are typically of a whole metre of drill core, meaning that Bi and W concentration could be averaged out by many different minerals other than pyrite. The separation between mineralised and unmineralised samples in the pyrite analyses could be attributed to the targeted nature of the spot analyses.

6. Conclusions

Based on microscopic observations and LA-ICP-MS analyses of this study, we conclude the following. However, these conclusions remain to be tested and refined by future studies.

• Pyrite was identified to be the dominant sulphide in the TGP, followed by arsenopyrite.
• Gold was present as inclusions or filling fractures and within the lattice structure of pyrite/arsenopyrite as invisible gold.
• Common observations of pyrite textures across all lithologies suggested that the paragenetic sequence can be applied to most deposits/prospects within the TGP.
• The best texture to investigate for Au is porous pyrite.
• Gold may have been incorporated early as sedimentary pyrite in graphitic siltstone, but the various host lithologies in the TGP suggest that graphitic siltstone may have been one of many Au sources. However, S isotopes are needed to confirm the early source of Au.
• Volcaniclastics, graphitic siltstone and andesite were the best Au-hosted lithologies that showed high vectoring potential using pyrite trace elements (Au, Bi, W, Sb, Se, Sn, As, Mn, Mo).
• Arsenopyrite trace elements were more suited for vectoring in dacite and monzodiorite (Se, Ag, Cd, Te, W, Mo, Sb, Bi, Tl, Pb, Co, Sn).
• Whole-rock data can distinguish between mineralised and unmineralised samples, but relationships between trace elements were unclear despite the high number of data points. On the other hand, fewer data obtained using the in situ LA-ICP-MS technique helped to further refine relationships between trace element pairs and differentiation between mineralised and unmineralised samples.

This study emphasised the benefits of utilising sulphide texture and chemistry not just as a mineral exploration toolkit but also as a valuable framework to better understand and gain more insights into Au-fluid-related processes.