Ninety-eight underground diamond holes (~102 km) drilled by Far Southeast Gold Resources Inc. at the Far Southeast porphyry Cu-Au deposit, Philippines, from 2011 to mid-2013, provide a three-dimensional exposure of the deposit between 700- and –750-m elevation, with surface at ~1,400-m elevation. Far Southeast contains an inferred resource of 891.7 million tonnes (Mt) averaging 0.7 g/t Au and 0.5 wt % Cu, equivalent to 19.8 Moz Au and 4.5 Mt Cu. This contribution reports the spatial and temporal distribution of alteration and mineralization at Far Southeast, notably a white-mica–chlorite-albite assemblage that formed after early secondary biotite and before late quartz–white-mica–pyrite alteration and that is associated with the highest copper and gold grades.

Alteration assemblages were determined by drill core logging, short-wavelength infrared (SWIR) spectral analysis, petrographic examination, and a quantitative evaluation of materials by scanning electron microscopy (QEMSCAN) study. Alteration is limited around sparse veins or pervasive where vein density is high and the alteration halos coalesce. The alteration and mineralization zones with increasing depth are as follows: (1) the lithocap of quartz-alunite–dominated advanced argillic-silicic alteration that hosts part of the Lepanto high-sulfidation Cu-Au epithermal deposit (mostly above ~700-m elevation), (2) an aluminosilicate-dominated zone with coexisting pyrophyllite-diaspore ± kandite ± alunite and white mica (~700- to ~100-m elevation), (3) porphyry-style assemblages characterized by stockwork veins (below ~500-m elevation), (4) the 1 wt % Cu equivalent ore shell (~400- to –300-m elevation), and (5) an underlying subeconomic zone (about –300- to –750-m elevation, the base of drilling). The ore shells have a typical bell shape centered on a dioritic intrusive complex.

The paragenetic sequence of the porphyry deposit includes stage 1 granular gray to white quartz-rich (± anhydrite ± magnetite ± biotite) veins with biotite-magnetite alteration. These were cut by stage 2 lavender-colored euhedral quartz-rich (± anhydrite ± sulfides) veins, with halos of greenish white-mica–chlorite-albite alteration. The white mica is largely illite, with an average 2,203-nm Al-OH wavelength position. The albite may reflect the mafic nature of the diorite magmatism. The quartz veins of this stage are associated with the bulk of copper deposited as chalcopyrite and bornite, as well as gold. Thin Cu sulfide (chalcopyrite, minor bornite) veins with minor quartz and/or anhydrite (paint veins), with or without a white-mica halo, also occur. These veins were followed by stage 3 anhydrite-rich pyrite-quartz veins with white-mica (avg 2,197 nm, illite)–pyrite alteration halos.

Combined with previous studies, we conclude that this porphyry system, including the Far Southeast porphyry and Lepanto high-sulfidation Cu-Au deposits, evolved over a period of 0.1–0.2 m.y. Three diorite porphyry stocks were emplaced, and by ~1.4 Ma biotite-magnetite–style alteration formed with quartz-anhydrite veins and deposition of ≤0.5% Cu and ≤0.5 g/t Au (stage 1); coupled with this alteration style, a barren lithocap of residual quartz with quartz-alunite halo plus kandite ± pyrophyllite and/or diaspore formed at shallower depth (>700-m elevation). Subsequently, lavender quartz and anhydrite veins with bornite and chalcopyrite (high-grade stage, avg ~1 wt % Cu and ~1 g/t Au) and white-mica–chlorite-albite halos formed below ~400-m elevation (stage 2). They were accompanied by local pyrite replacement, the formation of hydrothermal breccias and Cu sulfide (paint) veins. Stage 2 was followed at ~1.3 Ma by the formation of igneous breccias largely along the margins of the high-grade zones and stage 3 pyrite-quartz-anhydrite ± chalcopyrite veins with white-mica (mostly illitic) halos. At shallower depths in the transition to the base of the lithocap, cooling led to the formation of aluminosilicate minerals (mainly pyrophyllite ± diaspore ± dickite) with anhydrite plus high-sulfidation-state sulfides and pyrite veinlets. Consistent with previous studies, it is likely that the lithocap-hosted enargite-Au mineralization formed during this later period.

Porphyry copper systems, which include porphyry deposits, account for about three-quarters of global copper, half of molybdenum, and about one-fifth of gold production (Sillitoe, 2010). However, the majority of porphyry deposits that crop out have been discovered, requiring deeper exploration (Arndt et al., 2017), meaning that knowing and understanding alteration signatures and zonation, particularly toward deeper levels, is crucial to improve exploration efficiency and success. This study examines the alteration and ore paragenesis and zoning of the Far Southeast porphyry Cu-Au deposit, northern Luzon, Philippines, which lies ~1 km below surface (Fig. 1) and is in an advanced exploration stage (Gold Fields, 2019).

The Far Southeast porphyry deposit was discovered in 1980 (Concepcion and Cinco, 1989), subjacent to a lithocap (Sillitoe, 1995) of residual quartz with a quartz-alunite alteration halo that hosts the Lepanto enargite-gold deposit (Fig. 1B; Gonzalez, 1959; Claveria, 2001). Mining of the Lepanto deposit between 1938 and 1996 produced 36.3 million tonnes (Mt) of ore at 2.9 wt % Cu, 3.4 g/t Au, and 14 g/t Ag, with recovery of 3.2 Moz Au, 13.8 Moz Ag, and 0.74 Mt Cu (Chang et al., 2011). The discovery histories of both deposits, plus that of the adjacent Victoria epithermal veins (initial resource of 11 Mt at 7.3 g/t Au), were reviewed by Chang et al. (2011). From 2010 to mid-2013, the Far Southeast Gold Resources, Inc. (FSGRI) company, a joint venture between Gold Fields Limited and Lepanto Consolidated Mining Company, drilled over 102 km of diamond core in 98 new holes collared underground from numerous cuddies at 700-m elevation, about 700 m below the surface (Gaibor et al., 2013). This work established an inferred resource of 891.7 Mt of ore at 0.7 g/t Au and 0.5% Cu (Gold Fields, 2019), with the top of the resource at ~400-m elevation (1 km below the present surface) and the base at about –200- to –500-m elevation. The availability of the deep drill core provided the opportunity to document a complete world-class porphyry copper deposit over a vertical interval of ~1.5 km.

Previous studies of the Lepanto epithermal deposit and its host lithocap demonstrated a temporal and genetic linkage with the deeper Far Southeast porphyry deposit (Arribas et al., 1995; Hedenquist et al., 1998). Lepanto is located northwest of the porphyry deposit and its associated quartz diorite intrusions (Fig. 1B; Chang et al., 2011; Hedenquist et al., 2017). The historical alteration sequence described at Far Southeast includes potassic alteration followed by sericite-clay-chlorite and final sericite-dominant assemblages. Pyrophyllite-dominant assemblages occur at shallower elevations in the transition to the Lepanto epithermal deposit (Hedenquist et al., 1998, and references therein). Evidence that much of the copper and gold in the Far Southeast deposit was precipitated during formation of the postpotassic white-mica alteration was presented by Hedenquist et al. (1998), who also noted that the hydrothermal fluids responsible for these alteration stages were magmatic in origin.

In this study, we focused on the evolution of crosscutting veins, their mineralogy, and associated alteration halos. Drill core observations were made from the base of the lithocap at ~700-m depth, through a transition zone hundreds of meters thick to the top and through the porphyry ore zone, to the low-grade roots at –750-m elevation, more than 2 km below the present surface at 1,400-m elevation. Mineralogical observations consisted of core logging, a petrographic study, and microprobe, short-wavelength infrared (SWIR) and quantitative evaluation of minerals by scanning electron microscopy (QEMSCAN) analyses provided a 3-D characterization of the deposit, which contributed to our genetic interpretations.

Knowledge of porphyry copper deposits has accumulated for nearly 100 years, with the fundamental geologic elements and ore environments of porphyry deposits known by the 1930s (Lindgren, 1933). Lowell and Guilbert (1970) recognized common zonation patterns of porphyry copper deposits, and Gustafson and Hunt (1975) published a detailed study of alteration at the El Salvador, Chile, porphyry copper deposit; they introduced the Anaconda mining company’s observations of the different styles of veins and the time sequence of various zoning patterns, from early and hot to later and cooler and generated in conditions of either lithostatic or hydrostatic pressure, respectively. Recent reviews of porphyry copper deposits have focused on the overall architecture of the alteration/mineralization system (Sillitoe, 2010) and on the nature of the magmas and hydrothermal fluids involved (Audétat and Simon, 2012; Kouzmanov and Pokrovski, 2012; Audétat, 2019).

Various alteration types (mineral assemblages related to vein types originally defined by Gustafson and Hunt, 1975) are typically present in porphyry copper systems associated with calc-alkaline arc magmas (summarized by Seedorff et al., 2005, 2008; Sillitoe, 2010, and references therein). Sodiccalcic alteration is located on the deep margin, below the ore zone, and is composed of albite-actinolite-magnetite as halos of magnetite ± actinolite (M) veinlets. Sodic alteration is located deep, proximal to the ore zone, up to the level of white-mica alteration and consists of albite-chlorite-epidote ± specularite, commonly as halos of quartz-pyrite-tourmaline veins. Propylitic alteration occurs on deposit margins with variable amounts of Fe-rich amphibole-epidote-chlorite-pyrite ± albite ± carbonate. Sodic-calcic, sodic, and propylitic alteration types are typically barren. Potassic alteration is located deep, in the core of the ore deposit with magnetite-biotite ± K-feldspar, as halos of quartz-magnetite-biotite ± anhydrite ± Cu sulfide (A) veins and coincident with many ore zones. Paragenetically later chlorite–white-mica alteration is located in the upper portions of the deposit core with quartz-sulfide veins (pyrite, chalcopyrite) and remnant magnetite, martite, or specularite ± anhydrite and is commonly associated with ore minerals. White-mica alteration is common as quartz–white-mica ± pyrite halos to quartz-pyrite (D) veins and locally with upward increase in chalcopyrite, tennantite, and enargite abundance (Gustafson and Hunt, 1975). Sericitic and phyllic are synonyms of white mica; however, Meyer and Hemley (1967) advocated for the mineralogy to be determined and use of the term sericite to be discontinued.

Transitional to the epithermal environment, above the porphyry ore zone, aluminosilicate alteration (pyrophyllite-diaspore ± kandite ± alunite) typically cuts white mica, locally with shallow pyrite and traces of high-sulfidation-state sulfides. A lithocap occurs at shallow levels, up to the paleosurface, and can be a flat-lying body consisting of a core of residual quartz, with advanced argillic alteration halos of quartz-alunite outward to kandite (kaolinite and/or dickite), locally with pyrophyllite and/or diaspore in feeder structures. Lithocaps form over porphyry deposits (Sillitoe, 1995, 1999) at the time of deep potassic alteration (Arribas et al., 1995), which can be explained by the linkage between vapor and hypersaline liquid, respectively (Henley and McNabb, 1978). The lithocap is barren on formation (Chang et al., 2011; Hedenquist and Taran, 2013) but is a potential host for the introduction of subsequent intermediate- to high-sulfidation-state sulfides and gold (Hedenquist et al., 1998, 2000; Sillitoe and Hedenquist, 2003).

In some porphyry Cu deposits, such as Bajo de Alumbrera, Argentina, the highest metal grades appear to be associated with the early period of potassic alteration (Proffett, 2003). In contrast, there is evidence for copper and gold deposition to occur subsequent to biotite alteration, including at Chuquicamata and Escondida, Chile (Ossandón et al., 2001; Padilla-Garza et al., 2004), Butte, United States (Reed et al., 2013, and references therein), Batu Hijau, Indonesia (Clode et al., 1999; Garwin, 2002; Arif and Baker, 2004), and Pebble, Alaska (Lang and Gregory, 2012; Gregory, 2017). Sillitoe (2010) noted that chlorite–white-mica (sericite) alteration (Sillitoe and Gappe, 1984) is prevalent in the shallower parts of some porphyry Cu deposits, particularly those that are gold rich, and overprints preexisting potassic assemblages, with magnetite converted to hematite, and pyrite plus chalcopyrite deposited. Although copper and/or gold tenors of the former potassic zones may undergo depletion during the chlorite–white-mica overprint (e.g., Esperanza, northern Chile; Perelló et al., 2004), metal introduction is also recognized, such as at Escondida and Rosario, Chile (Padilla-Garza et al., 2001; Masterman et al., 2005). Indeed, this type of alteration at a few deposits is considered to account for much of the contained copper (e.g., Cerro Colorado, northern Chile; Bouzari and Clark, 2006).

Consistent with this recognition, cathodoluminescence (CL) imaging also indicates that early quartz veins that formed at high temperature are typically reopened and filled by quartz and copper sulfides at lower temperature, e.g., at Butte (Rusk and Reed, 2002), as well as at Bingham Canyon (Redmond et al., 2004; Redmond and Einaudi, 2010) and Santa Rita (Tsuruoka et al., 2021), United States, plus at El Salvador (Watanabe et al., 2018). In some cases, copper precipitation is not accompanied by quartz deposition (i.e., sulfide-only veinlets of chalcopyrite or bornite, referred to as paint veins by some; e.g., Stefanova et al., 2014), due to the fluid being within the field of retrograde solubility of quartz at ~380°–550°C (Fournier, 1985; Shock et al., 1989), overlapping with the temperature of copper deposition at Bingham Canyon (350°–420°C; Landtwing et al., 2005).

The Far Southeast porphyry deposit is located in the Mankayan district of the Cordillera Central of northern Luzon (Fig. 1A). The N-trending Philippine archipelago is flanked by two convergent zones, the Manila trench to the west and the Philippine trench to the east (Queano et al., 2007; Hollings et al., 2011). When mineralization occurred at 2–1 Ma in the Mankayan district, eastward subduction of the aseismic Scarborough sea ridge beneath the region was ongoing (Cooke et al., 2005).

The geology of the district has been described by Gonzalez (1959), Sillitoe and Angeles (1985), Concepción and Cinco (1989), Garcia (1991), and Chang et al. (2011). More recent geologic mapping by FSGRI (P. Dunkley, unpub. report, 2015) has refined the stratigraphy, which comprises a basement of pre-middle Miocene volcanic and intrusive rocks, overlain by a cover sequence of Pleistocene dacitic tuffs and breccias intruded by domes that are broadly andesitic in composition. The Pleistocene volcanism was accompanied by the shallow intrusion of diorite complexes (within only a few kilometers of the paleosurface), with which the mineralization in the district is related (Fig. 1B).

Data for recent radiometric dating results (Table 1; P. Dunkley, unpub. report, 2015) are presented in the digital Appendix Table A1, including methods. The oldest exposed stratigraphic unit, the Lepanto volcanic unit, comprises a thick sequence of basaltic pillow lavas, hyaloclastites, and tuffaceous rocks, with a minor rhyolitic component toward the top, and was intruded by basaltic and rhyolitic dikes. Uranium-Pb zircon dates of 35.3 ± 1.0 and 33.7 ± 0.7 Ma have been obtained for the rhyolites (Table 1; P. Dunkley, unpub. report, 2015). Uncertainties listed for dates are all at 2σ or 95% confidence level. A large gabbro-diorite-tonalite pluton, the Bagon complex, dated by U-Pb on zircons at 34.8 ± 0.7 and 34.5 ± 0.5 Ma (Table 1; P. Dunkley, unpub. report, 2015), intruded the Lepanto volcanic unit and forms part of the more extensive Central Cordillera batholith that runs along the spine of the Luzon Cordillera Central. The Balili volcaniclastic unit unconformably overlies the Lepanto volcanic unit and consists of poorly sorted epiclastic volcanic breccias, conglomerates, and sandstones. The breccias and conglomerates are mainly composed of polymictic clasts of andesite, minor clasts of diorite and granite, and rare clasts of limestone. Based on fossil evidence, Maleterre (1989) assigned a late Oligocene-early Miocene age to the Balili unit.

The Pleistocene volcanic sequence comprises the Imbanguila pyroclastic unit overlain by the Bato pyroclastic unit. Both units consist of massive to poorly stratified dacitic tuff breccias and fine tuffs containing beds with accretionary lapilli. The tuff breccias and tuffs contain lithic blocks and lapilli of diverse composition, including widespread clasts of diorite with porphyry-style veinlets and mineralization (Sillitoe, 1983; P. Dunkley, unpub. report, 2015). The Imbanguila unit is a host to the Lepanto enargite orebody and is widely affected by strong alteration. It is intruded by the Imbanguila diorite porphyry (also known as Imbanguila dacite porphyry), which has yielded K-Ar hornblende dates from 2.19 ± 0.62 to 1.82 ± 0.36 Ma (Arribas et al., 1995) and an Ar-Ar date on groundmass plagioclase of 1.94 ± 0.26 Ma (Table 1; P. Dunkley, unpub. report, 2015). The younger Bato units are syn- to postmineralization in age and less affected by hydrothermal alteration. Arribas et al. (1995) obtained a K-Ar date of 1.43 ± 0.21 Ma on hornblende from the matrix of a Bato tuff breccia. Numerous domes and laccoliths of broadly dacitic composition that intruded the Bato pyroclastic sequence at shallow depth yielded biotite and hornblende K-Ar dates of 1.18 ± 0.08 and 0.96 ± 0.29 Ma, respectively (Arribas et al., 1995) and Ar-Ar dates on groundmass plagioclase ranging from 1.68 ± 0.33 to 1.16 ± 0.09 Ma (Table 1; P. Dunkley, unpub. report, 2015).

The structure of the district is dominated by a complex zone of N- to NW-trending faults of the Abra River fault system (Fig. 1A), which is the main branch of the Philippine fault running along the length of the Luzon Cordillera Central (Barrier et al., 1991). Geologic mapping by FSGRI shows the NW-trending Mankayan fault is a major strand of the system and the largest fault in the district. Northwest- and E-trending secondary faults (Lepanto, Hatakazawa, and Imbanguila faults; Fig. 1A) splay from the Mankayan fault and were primary controls on the emplacement of the enargite lodes of the Lepanto mine as well as the Far Southeast porphyry deposit.

The E-trending Imbanguila fault coincides with a strong E-trending gravity gradient, interpreted as a deep-rooted basement structure thought to have controlled the emplacement of the Far Southeast intrusive complex, the main host of the Far Southeast porphyry Cu-Au deposit (Gaibor et al., 2013; Fig. 2). The Far Southeast complex consists of several diorite bodies that intruded the Lepanto volcanic unit and the Imbanguila diorite porphyry, which are cut across by large pipe-like breccia bodies (Fig. 2). Hydrothermal biotite from the Far Southeast deposit reported an average K-Ar date of 1.41 ± 0.05 Ma (n = 6), essentially the same as that of alunite within the overlying lithocap of 1.42 ± 0.08 Ma (n = 5; Arribas et al., 1995). Molybdenite from the Far Southeast deposit provided Re-Os dates of 1.50 ± 0.01 and 1.48 ± 0.01 Ma (Table 1; P. Dunkley, unpub. report, 2015). The subsequent overprint of white mica was dated with the K-Ar method; illite returned dates of 1.37 ± 0.05 to 1.22 ± 0.06 Ma (n = 10; Arribas et al., 1995).

The oldest and deepest host rocks intersected at Far Southeast are basaltic rocks of the Lepanto volcanic unit (Figs. 1B, 2). These were intruded by the Imbanguila diorite porphyry (Figs. 3A, 4), which was in turn intruded by the younger Far Southeast diorite intrusive complex. The complex is divided into three diorite units by FSGRI workers (P. Dunkley, unpub. report, 2015), based on different granularity, color, groundmass/phenocryst ratios, and phenocryst mineralogy (quartz, plagioclase, hornblende). An early quartz diorite porphyry stock (Figs. 3B, 4), defined as undifferentiated diorite porphyry, is characterized by ~10% quartz, ~20% hornblende, and abundant plagioclase phenocrysts. This is cut by the dark diorite porphyry with less abundant quartz phenocrysts (<5%) and similar proportions of plagioclase and hornblende (Figs. 3C, 4). The youngest intrusion, the light diorite porphyry (locally equigranular), is characterized by <5% quartz, ~75% plagioclase, and ~20% hornblende phenocrysts (Figs. 3D, 4). The individual intrusive phases can be distinguished at their boundaries, locally with chilled margins (Fig. 3E), truncated veins (Fig. 3F), or abrupt changes in quartz vein density. Most appear to be intermineralization in age, since xenoliths of the Lepanto volcanic unit with truncated quartz veins (Fig. 3G) are present. Away from the contacts, the intrusions are difficult to differentiate, particularly where altered, which posed a challenge to core logging and cross-section construction (Fig. 4). Some late-stage dikes are weakly altered by illite-chlorite but without veins.

The intrusive complex is cut by several breccia pipes (Figs. 2, 4). Within the drill hole intervals logged, two categories of breccia are recognized. One is a polymictic igneous breccia with a fine-grained igneous matrix, with clasts having blocky to wispy shapes and mostly porphyritic with andesite and dacite composition (Fig. 5A). Some clasts of diorite porphyry contain porphyry copper-style veins (Fig. 5B), some being fragments of quartz-magnetite veins (Fig. 5C), whereas other clasts are xenoliths of residual quartz with vuggy texture, derived from shallow lithocap levels or from an older lithocap at deeper position when this brecciation event occurred (Fig. 5D). These breccia bodies have a pipe shape with a slight angle from vertical and random azimuthal directions and widen upward, as shown in Figure 2. This type of breccia is interpreted to be of phreatomagmatic origin based on the wispy, likely juvenile clasts (Wright et al., 1980); the flaring-upward pipe shape is compatible with this hypothesis. The second breccia style is polymictic and clast supported (without apparent juvenile material); clasts have various alteration styles, some with clastconfined veins (Fig. 5E). The cement consists of hydrothermal quartz, anhydrite, chalcopyrite, and pyrite, leading to a hydrothermal breccia interpretation. The phreatomagmatic breccias cut through the diorite intrusions and/or the wall rock of Imbanguila and Lepanto volcanic unit and extend for an unknown distance above the intrusion complex (Figs. 2, 4). Some of the phreatomagmatic breccias are located just outside of the 1% Cu and 1 ppm Au ore shells. The hydrothermal breccia is hosted by the intrusion, adjacent to a magmatic breccia, and inside the 1% Cu and 1 ppm Au ore shell (Fig. 2).

Mineral assemblages (Table 2) were characterized by hand lens during core logging (e.g., Fig. 6) and SWIR analyses in the core shed, followed by laboratory study, including optical microscopy (Fig. 7), backscattered electron (BSE) imaging using a scanning electron microscope (SEM), and QEMS-CAN examination of polished thin sections. Methodology is discussed further in the digital Appendix.

In the main porphyry deposit, a complex assemblage of veins and their alteration halos occurs. In places, vein alteration halos may be only a few millimeters wide. Halo widths increase with density of the related vein type, with halos around vein stockworks ultimately coalescing to form continuous alteration for tens to >100 m. Three stages of vein and alteration assemblages are recognized at Far Southeast: (1) veins of granular quartz-magnetite ± anhydrite ± biotite, with shreddy biotite + magnetite alteration halos, (2) veins of lavender to white quartz plus anhydrite and sulfides, with alteration halos of white-mica-chlorite-albite, and (3) sulfide-rich veins ± anhydrite ± quartz with bleached halos of white mica (Fig. 6). The spatial distribution of these assemblages is shown in Figure 8. Copper sulfides are observed in these three stages, within veins and alteration halos and disseminated in pervasive alteration. Pervasive propylitic alteration occurs outside the area of this study (Garcia, 1991).

There is also a veinlet-associated pervasive alteration of largely aluminosilicate minerals, locally with alunite, which occurs mostly above the top of the main porphyry deposit (above ~100- to 200-m elevation) up to ~700-m elevation (Fig. 8). At shallow depths (above ~700-m elevation, i.e., above the level of this study), a residual quartz-bearing part of the lithocap occurs at the unconformity between the Imbanguila and Balili or Lepanto volcanic unit. The residual quartz, locally with a vuggy texture, and a halo of quartz-alunite ± dickite and kaolinite, hosts the Lepanto high-sulfidation epithermal Au-Cu deposit (Garcia, 1991; Fig. 1B).

The locations of the 10 representative samples studied in detail with QEMSCAN are shown on the northwest-southeast section (Fig. 8). Sample information and the summary results are presented in Table 3. The results for five of these samples are presented here, with results from the other five samples included in Appendix Figures A1-A5.

Stage 1: Biotite-magnetite alteration halo to granular quartz veins

Stage 1 biotite alteration marks the earliest recognized hydrothermal event, as evidenced by crosscutting relationships of veins (Fig. 6A). Veins are typically wavy in appearance (Fig. 6A, B) and characterized by granular quartz with common magnetite, local anhydrite, and coarse-grained biotite (Fig. 6B, C). Most are dominated by quartz, but hairline magnetiterich veins also occur. Biotite crystals (<100 µm in size) with a shreddy texture and magnetite occur scattered throughout the groundmass of the intrusions (Fig. 7A) and surrounding Lepanto volcanic unit and within granular quartz veins (Fig. 7B). Copper sulfides are typically not abundant, but bornite has been attributed to this stage. This alteration style is not closely associated with the high-grade Cu-Au zone.

Stage 1 is principally preserved below –250-m elevation, just below the 1 ppm Au grade shell and ~100 m below the 0.5 wt % Cu grade shell in drill holes on the northwest-southeast section (Fig. 8). This style of alteration mostly occurs within intrusions but has been observed up to 250 m into the Lepanto volcanic unit. It is locally preserved at shallower levels (up to 500-m elevation), mostly outside of the 0.5 g/t Au and 0.5 wt % Cu grade envelopes.

Stage 2: White-mica–chlorite-albite alteration associated with lavender quartz

Stage 2 veins are composed of euhedral lavender quartz, anhydrite, hematite plus minor magnetite, pyrite, and Cu sulfides (Fig. 6F). Hematite commonly replaces magnetite with the reverse locally observed (e.g., magnetite has locally replaced hematite as in Fig. 7C). In the high-grade zone, chalcopyrite is commonly replaced by bornite and later by covellite (Fig. 7D). The pyrite in the veins appears to be replaced by chalcopyrite and bornite, with a Swiss cheese texture, locally with incipient chalcocite and digenite replacement (Fig. 7E). Chlorite from stage 2 commonly replaces stage 1 hydrothermal biotite (Fig. 7B). The lavender color of these quartz veins may be caused in part by hematite daughter minerals in hypersaline fluid inclusions trapped in quartz crystals (Fig. 7F) coexisting with vapor-rich inclusions. As well as occurring in the rock matrix, this alteration replaced primary igneous minerals, so that hornblende has been replaced by chlorite, magnetite, and chalcopyrite (Fig. 7G). Copper sulfides also occur as thin (~1 mm) straight veinlets, referred to as paint veins, containing few or no other minerals (Fig. 6E).

Stage 2 veins have halos of white-mica–chlorite-albite wall-rock alteration. This is the dominant alteration type between 500- and –300-m elevation and is spatially associated with the high-grade copper and gold zone (Fig. 8). This alteration style has been observed in all lithologic units and occurs as narrow vein halos (Fig. 6F) but can become pervasive throughout the wall rock where stage 2 veins are abundant and halos coalesce (Fig. 6H). Late-mineralization breccias contain quartz diorite porphyry clasts affected by stage 2 alteration. Paint veins are typically hosted by stage 2 altered rock but can also have white-mica–rich alteration halos (Fig. 6F). Examination of stage 2 alteration by QEMSCAN (Fig. 9A) indicates that pervasive white mica and chlorite patches in distinct zones (0.5–2 mm in size), probably reflecting the original igneous texture and composition of the rock unit; white mica and chlorite likely replaced felsic and mafic minerals, respectively.

Gaibor et al. (2013) first recognized albite associated with the highest density of stockwork veins that coincide with the highest copper and gold grades. Hydrothermal albite can be difficult to distinguish by optical microscopy alone due to its typical fine-grained and pseudomorphous character (as it maintains the original shape and optical features when igneous feldspars or groundmass are replaced). Typical hydrothermal albite (as blocky to tabular crystals, with characteristic parallel twinning) is coeval with anhydrite, magnetite, hematite, and chalcopyrite (Fig. 7H); the composition has been measured by microprobe (Fig. 10). The high albite content of stage 2 alteration (consisting of ~50% of sample MB45) is shown by QEMSCAN analyses to occur in two areas of study, 1 and 2 (Fig. 9B), more so in the former (73 and 43%, respectively, with plagioclase remaining in the latter).

The Al-OH absorption peak position of stage 2 white mica ranges from 2,188 to 2,225 nm, but is mostly >2,200 nm, with a mean of 2,203 nm (Fig. 11). The higher wavelength position indicates a higher Fe and Mg content (phengitic), accounting for the light-green color of the white mica (Uribe-Mogollon and Maher, 2018). The green color of stage 2 alteration (Fig. 6F, H, I) is likely caused by the dark-green chlorite plus light-green phengitic white mica. About 90% of illite crystallinity (IC) values of the white mica of this stage vary between 0.4 and 2.6 (App. Fig. A6), indicating that most of the white mica is illite to interlayered illite/smectite.

Stage 3: White-mica–quartz alteration associated with anhydrite-pyrite-quartz veins

Sulfide-rich (pyrite ± chalcopyrite)-quartz-anhydrite veins plus a white-mica alteration halo cut stage 1 and 2 assemblages (Fig. 6D, F, I, J). This assemblage occurs over a wide vertical range, from the deepest drill core (–750 m) up to ~500-m elevation, particularly associated with fractures within both the Lepanto volcanic unit and the Far Southeast diorite complex. White-mica alteration is also pervasive within and in part surrounding the vertical (pipe-like) breccia body with igneous matrix (Fig. 8), which formed relatively late, with lower metal grades. Stage 3 pyrite-anhydrite-quartz or anhydrite-chalcopyrite veins with a white-mica halo may cut and overprint the two previous stages, including white-mica–chlorite-albite (Fig. 9C) and biotite-magnetite–altered rock (Fig. 9D). There is also evidence for white-mica precipitation prior to or intergrown with chalcopyrite (Fig. 7I, J).

The Al-OH absorption peak position of stage 3 white mica ranges from 2,188 to 2,207 nm, mostly <2,198 nm, with a mean of 2,197 nm (Fig. 11B). Such lower wavelength positions typically indicate elevated Na content (paragonitic) and a low Fe-Mg content, which is consistent with the typical white color of the stage 3 white mica (Fig. 11A). Although the value ranges overlap, compared to stage 2 white mica, the Al-OH absorption feature of stage 3 white mica has a lower wavelength position (Fig. 11B).

A plot of white-mica composition (2,200-nm feature) of samples from all stages on the northwest-southeast and southwest-northeast sections (Fig. 12A) shows values of less than ~2,197 nm above ~300-m elevation on the southwest-northeast section. Both sections record values >2,203 nm within the high-grade zones (>1% Cu and >1 g/t Au), although there are high as well as low values outside the high-grade centers of the deposit. The lower white-mica wavelength features on both sections tend to correlate with the location of aluminosilicate minerals, particularly on the southwest-northeast section.

Aluminosilicate alteration

At levels above the main porphyry orebody (mostly above 250-m elevation, within Imbanguila diorite porphyry), aluminosilicate alteration is pervasive, with variable assemblages that include pyrophyllite, diaspore, kandite, quartz, anhydrite, and/or illite, and minor to rare alunite, woodhouseite (an aluminum phosphate-sulfate mineral of the alunite group), andalusite, and dumortierite. Optical microscopy reveals diaspore associated with aluminosilicate minerals in the rock matrix or replacing original feldspars (Fig. 7K). QEMSCAN analyses shows bleached and mottled pervasive alteration of quartz and white mica with pyrophyllite plus diaspore, cut by a quartz vein reopened by an anhydrite-rich vein with white mica, pyrite, bornite (replaced by chalcocite), barite, and woodhouseite (Fig. 13). The mineralogy of the aluminosilicate minerals was confirmed by SWIR spectroscopy (Fig. 12B, C); locally aluminosilicates were also identified at depth, along narrow structures. Aluminosilicate alteration is associated with veins of anhydrite-quartz ± hematite that variably contain pyrite, chalcopyrite, bornite, chalcocite, covellite, digenite, enargite, sphalerite, or molybdenite (Fig. 7L).

Alteration types: Overprinting and evolution

The key characteristics of the various alteration and mineralization stages recognized at Far Southeast are summarized in Table 2. The main veins and their associated alteration minerals (Fig. 14) follow the general temporal sequence of early biotite (stage 1) to white-mica–chlorite-albite (stage 2) and then white mica (stage 3); at shallower depth, there is a transition from white mica to aluminosilicate alteration, up to the base of the earlier formed lithocap (during the biotite stage). Key minerals associated with these alteration types follow the long-recognized paragenetic sequence of biotite → chlorite → white mica → pyrophyllite, consistent with a sequence predicted by a decrease in temperature as well as an evolution in fluid composition (Hemley and Jones, 1964; Hemley et al., 1980; Seedorff et al., 2005; Sillitoe, 2010).

At depth (below –250-m elevation), early biotite represents the highest temperature alteration (500°–550°C; Hedenquist et al., 1998); K-feldspar (Fig. 9B), and magnetite, commonly replaced by hematite, is also present. This high temperature is consistent with the sinuous nature of the associated quartz veins (Fig. 6A, B), indicating a plastic rheology of the rock (Gustafson and Hunt, 1975). Subsequently, biotite is widely replaced by chlorite (Fig. 7B), interpreted to be due to system cooling (Seedorff et al., 2005). Early sulfides associated with the biotite stage are dominated by bornite, as well as digenite and covellite, whereas chalcopyrite dominates stage 2.

A large amount of albite is associated with stage 2 alteration, up to ~40–70% of the rock, as quantified by QEMSCAN (Fig. 9B), indicating Na mobility during this period of alteration. In stage 2 alteration, albite follows anhydrite, as seen in thin fractures (Fig. 9A) and within the rock matrix (Fig. 9B; App. Figs. A1, A2, A5). Compositional analyses of hydrothermal plagioclase associated with stage 2 alteration confirms the presence of albite (Or1Ab96–98An1–3; Ab = albite, An = anorthite, Or = orthoclase; Fig. 10). This alteration style identified at Far Southeast (Gaibor et al., 2013) is partly analogous to the sericite-clay (illite)-chlorite alteration first noted in Philippine porphyry copper deposits (previously referred to as SCC; Sillitoe and Gappe, 1984). Thus, it is possible that other deposits with previously identified chlorite-sericite alteration (Sillitoe, 2010) may also contain significant but unrecognized albite, due to its fine grain size and tendency to replace and retain the shape of igneous minerals.

Stage 2 alteration style is overprinted by white-mica alteration, which typically occurs as halos to straight anhydrite and pyrite-rich quartz veins (Fig. 9C). This hydrothermal event represents the waning period of the mineralizing system. The SWIR Al-OH wavelength of the white mica decreases from white-mica–chlorite-albite to white-mica-only alteration (Fig. 11), caused by a trend from more Fe-Mg–rich toward more K-Na–rich illitic minerals.

The IC determined by SWIR correlates with IC determined from X-ray diffraction (XRD) (Doublier et al., 2010), which is known to be related to formation temperatures (e.g., Frey, 1987). Based on our experience, SWIR-IC values >3.5 are typically obtained from coarse-grained muscovite associated with porphyry, greisen, and pegmatite deposits, which may form at >350°C. By contrast, IC values <0.5 are typically smectite that forms at temperatures of ~150°C (Hedenquist and Arribas, 2022). Most IC values at Far Southeast are between 0.8 and 2.5 (App. Fig. A6). Samples from a shallow depth, with IC values of 0.8–1.5, are from an area on the porphyry margin. Nearby, a fluid inclusion assemblage in quartz with white mica—suspected to be illite—returned a value of 280°C from the edge of the high-grade zone (Hedenquist et al., 1998).

Aluminosilicate alteration occurs mainly at ~700 m to as deep as ~100- to 200-m elevation (Fig. 8), including pyrophyllite, diaspore, and kandite minerals along with minor alunite. This alteration is typically present with quartz-anhydrite veinlets that contain pyrophyllite, diaspore, and woodhouseite, which cut white-mica alteration (Fig. 13), consistent with cooling from white-mica conditions (Hemley et al., 1980; Watanabe and Hedenquist, 2001). The sulfide assemblage (pyrite-bornite-chalcocite; Figs. 7L, 13) indicates a high-sulfidation state, consistent with the mineralogy of the overlying lithocap-hosted enargite-luzonite-gold (telluride) deposit (Gonzalez, 1959; Claveria, 2001). Cooling of fluid originally in equilibrium with white mica continued through the aluminosilicate zone and into the residual quartz lithocap that hosts the Lepanto high-sulfidation deposit, with cooling—in an environment of limited rock buffer capacity— again being the main cause of the shift from intermediate- to high-sulfidation-state sulfides (Einaudi et al., 2003). This ascent of deep liquid contrasts with the earlier vapor condensates that initially formed the barren residual quartz lithocap and quartz-alunite alteration halo (Hedenquist et al., 1998). Distinguishing the features of both styles (i.e., formation environments) of advanced argillic alteration (shallow lithocap due to vapor condensation vs. deeper aluminosilicate cooled from white mica) is essential to better understand their position relative to a potential porphyry deposit (Hedenquist and Arribas, 2022).

Criteria for the timing of this aluminosilicate ± minor alunite alteration type are not straightforward. There is a partial overlap of the mineralogy of this style with that of the lithocap, including kandite and alunite plus pyrophyllite. Macroscopically, this assemblage is observed to cut and overprint white-mica–chlorite-albite alteration, consistent with it being contemporaneous with (but shallower than) most white-mica alteration, which distinguishes it from the early formed and shallow lithocap alteration. This distinction is also supported by isotopic analyses of the pyrophyllite (Hedenquist et al., 1998), with pyrophyllite and white mica sharing a similar formational fluid of largely magmatic origin, indicating that the aluminosilicate-dominated alteration is the shallow equivalent of stage 3 white mica.

Alteration mineralogy and grade distribution

Information from drill core logging, along with SWIR analyses to help clarify mineralogy, and copper and gold grades are plotted on lithology cross sections (Fig. 8), encompassing a volume of ~2.5 km3. The higher copper and gold grades tend to overlap, as is typical of porphyry Cu-Au deposits world-wide (Esperanza, northern Chile, Perelló et al., 2004; Pebble, Alaska, Gregory et al., 2013; Grasberg, Indonesia, Leys et al., 2019), and are located at the top of the Far Southeast diorite complex (between an elevation of about –300 and 500 m), confirming the relationship between the three observed intrusions and metal distribution. High grades of copper and gold (>1% Cu and 1 g/t Au) are most closely associated with intense white-mica–chlorite-albite alteration, whereas biotite alteration is associated with grades that are typically ≤0.5 g/t Au and ≤0.5 wt % Cu (Fig. 8).

As determined from detailed logging by company geologists, an approximate estimate of copper deposition at Far Southeast is about 30% associated with the biotite stage, 50% in the high-grade white-mica–chlorite-albite stage, and perhaps 20% in the white-mica stage. Most Cu sulfides are associated with postbiotite alteration, but they have, nonetheless, been observed in multiple alteration stages. This, along with observed contradicting paragenetic relationships, is testament to fluid pulses and fluctuations of fluid conditions, similar to that modeled by Weis et al. (2012). Major and high-grade metal (copper and gold) deposition (Gaibor et al., 2013) was associated with lavender quartz-anhydrite veins and related white-mica–chlorite-albite alteration, as well as chalcopyrite or bornite paint veins (Fig. 6E, F; with or without a white-mica alteration halo).

Stage 2 alteration was overprinted by white-mica alteration and the aluminosilicate style (pyrophyllite, diaspore, dickite) at shallower levels as the system cooled to <300°C, the temperature limit deduced for this assemblage (Hemley et al., 1980; Watanabe and Hedenquist, 2001). Both white-mica-only and aluminosilicate styles of alteration tend to be associated with lower grades (perhaps due to remobilization of earlier deposited metals), although halos containing these alteration assemblages can occur adjacent to quartz-chalcopyrite veins and high-grade intervals (white-mica halos, App. Figs. A3, A4, and aluminosilicate veins, Fig. 13). This period was also contemporaneous with the emplacement of the breccias, possibly triggered by the intrusion of even deeper diorite bodies (below drilled depths), as evidenced from the juvenile clasts in some of the breccias with felsic igneous matrix (Fig. 5A). At least two of the phreatomagmatic breccias appear to have intruded preferentially along the margins of the preexisting diorite intrusions (Fig. 2), suggesting this was a zone of weakness with increased permeability for magmatic-hydrothermal activity. The emplacement of the breccia with a felsic matrix diluted the metal grades (note effect on copper and gold contours in Fig. 4), whereas the breccia with a hydrothermal matrix is within the high-grade (>1 ppm Au) zone.

Several questions remain, including the following: How much metal, if any, that was precipitated during stage 2 was remobilized from deeper potassic-altered levels, rather than directly from the magmatic-hydrothermal source? How much metal in the Lepanto epithermal deposit was derived directly from a deep magmatic source versus remobilized from the porphyry Cu-Au deposit during the relatively late white-mica alteration overprint?

Alteration paragenesis of and metal introduction to porphyry Cu-Au deposits

Debates on the timing of copper introduction in porphyry deposits have long existed (Schwartz, 1947). In the case of Far Southeast, we have documented that copper and gold deposition was mostly associated with postbiotite stage 2 white-mica–chlorite-albite alteration, particularly that of the higher grades (>0.5 wt% Cu and >0.5 g/t Au). This is indicated by the spatial association of the high-grade ore shell and intense stage 2 alteration, by the large quantity of Cu sulfides in related lavender quartz veins, and by the abundance of paint veins with this stage and the subsequent white-mica alteration.

The dominance of postbiotite mineralization was also found to be the case at the gold-rich porphyry copper deposit of Batu Hijau, where Schirra et al. (2019, 2022) noted that the sulfides hosted by the A and B veins were deposited in crosscutting veinlets that reopened A and B veins, similar to C (paint) veins with narrow selvages of chlorite-sericite-albite (Clode et al., 1999). The albite-related alteration is not only post-A and B veins, it appears to also be associated with the event that deposited high copper and gold grades, like at Far Southeast. Garwin (2002) suggested that this sodic metasomatism at Batu Hijau was due to the presence of low-K tonalite intrusions, which promoted the formation of oligoclase ± albite alteration in the system. The dioritic intrusions at Far Southeast have a similar composition (App. Table A3) to the Batu Hijau tonalite, supporting the hypothesis that a mafic magmatic contribution was associated with the albite that formed during the main, high-grade stage of copper and gold mineralization.

At the young Far Southeast porphyry deposit (Arribas et al., 1995), shallow intrusion of diorite porphyry stocks into a ~2 Ma pyroclastic and porphyritic volcanic unit (Imbanguila) preceded the development of a magmatic-hydrothermal system, which by ~1.4 Ma formed biotite-magnetite alteration with sinuous veins of quartz and anhydrite plus deposition of ≤0.5% Cu and ≤0.5 g/t Au. Coupled with this early stage, acidic vapor condensates formed a lithocap (barren residual quartz core with quartz-alunite-kandite halo) at shallower depth, <1 km (Fig. 15A). As the system cooled, brittle veins of lavender quartz and anhydrite veins formed with alteration halos of white mica-chlorite-albite. This white mica is characterized by an Al-OH 2,200-nm peak position with higher SWIR wavelength than subsequent white-mica-only alteration, with means of 2,203 versus 2,197 nm, respectively (Fig. 11). Chalcopyrite and bornite paint veins also formed during these later stages, giving rise to the high-grade (~1 wt % Cu and ~1 g/t Au) mineralization of the porphyry deposit (Fig. 15B). At shallower depths, in the transition to the base of the lithocap, subsequent cooling from white-mica stability led to the formation of aluminosilicate minerals (pyrophyllite ± diaspore ± dickite) with anhydrite plus high-sulfidation-state sulfides and pyrite veinlets (Fig. 15C). This later stage was likely the timing of metal transport to the lithocap, which formed the Lepanto enargite Au deposit (Hedenquist et al., 1998; Chang et al., 2011). There are at least three subvertical breccia bodies centered around the subvolcanic intrusive complex (Figs. 2, 15C); alteration of the breccia pipes is white mica only, with pyrite ± chalcopyrite.

The recognition of albite in the widespread white-mica– chlorite assemblage at Far Southeast (Gaibor et al., 2013; this study) is consistent with observations of postbiotite albite in other young, little-eroded porphyry deposits such as Batu Hijau (Clode et al., 1999; Garwin, 2002). This event (of an intermediate hydrothermal timing, postbiotite and pre-white mica-dominant) is associated with the highest concentrations of copper and gold and was a fundamental stage in the development of the Far Southeast deposit. We predict that on close examination albite, formed during high-grade metal introduction, will be found in other porphyry deposits, particularly those that are gold rich and associated with more mafic intrusions.

We acknowledge Bryan Yap, president of the Lepanto Consolidated Mining Company, and Frederick Louw, formerly president of Far Southeast Gold Resources Inc. (FSGRI), for permission to conduct this study, and Mr. Yap plus Gary Snow, vice president of exploration for Gold Fields Australia, for their permission to publish the results. We thank FSGRI for logistical support to MFC, ZC, AA, KK, and JWH during field visits to Lepanto, and Glacialle Tiu, formerly of FSGRI, for her generous contributions and assistance. This study was funded by James Cook University, Australia, the University of Geneva, Switzerland, Akita University, Japan, the SEG Foundation, and the Swiss Federal Social Insurance Office. We thank Dave Selby, Sebastian Meffre, and Fred Jourdan for the radiometric analyses, funded by FSGRI, Bill Chavez, John Muntean, and Richard Sillitoe, who provided constructive comments on early drafts, and Lisard Torró, Yi Sun, and David Cooke, whose careful manuscript reviews helped to clarify our presentation.

Michael F. Calder has conducted research on magmatic-hydrothermal deposits in Armenia and the Philippines, affiliated with the University of Geneva, Switzerland, and James Cook University, Australia. After working in an energy consulting firm, he has focused on ore deposit research and is based in Geneva. He acknowledges the importance of metals to help decarbonize the world economy. Recently, he has been involved in Ore Deposits Hub, an online economic geology communication platform. He enjoys traveling, arts and culture, the outdoors, and hiking in the mountains.

Gold Open Access: This paper is published under the terms of the CC-BY-NC 3.0 license.

Supplementary data