Samples of tourmaline supergroup minerals from seven mineralized porphyry systems (Cu, ±Au, ±Mo), including Casino (Yukon Territory, Canada), Coxheath (Nova Scotia, Canada), Donoso breccia-Los Bronces (Chile), Highland Valley Copper (British Columbia, Canada), New Afton (British Columbia, Canada), Schaft Creek (British Columbia, Canada), and Woodjam (British Columbia, Canada), were examined at a variety of scales to evaluate their relationships with mineralization. Data from paragenetic observations show that tourmaline supergroup minerals are generally early hydrothermal minerals that predate both mineralization and alteration (e.g., overgrown and crosscut by). In general, tourmaline supergroup minerals occur as sub- to euhedral crystals that are black in hand sample and can be found in a variety of mineralized settings (including breccias, veins, and disseminations) and alteration assemblages (including potassic, sodic-calcic, phyllic, propylitic, and argillic). As tourmaline supergroup minerals are physically and chemically resilient and occur throughout a given porphyry system, they are comprehensive recorders of the type and extent of various geochemical processes that exist during the complex genesis of these systems. Data from BSE imaging shows two primary zonation types: concentric and sector. These are interpreted to reflect conditions of rapid crystallization and disequilibrium. Results from SEM-EDS analyses show that most tourmaline supergroup minerals are dravite (∼80% of grains), with the remainder being primarily classified as schorl. Porphyry tourmaline supergroup minerals exhibit remarkably consistent ∼2.0 apfu Mg values (range: 0.69–2.89), with the majority of tourmaline supergroup minerals plotting along the oxy-dravite–povondraite trend, reflecting the predominance of the Al3+ ↔ Fe3+ substitution at constant Mg values. This pattern starts from the povondraite side (reflecting the oxidizing nature of early porphyry mineralizing fluids) and trends toward oxy-dravite as a porphyry system evolves, a feature that can, in turn, be interpreted to reflect relative emplacement depths. In mineralized porphyry systems, tourmaline supergroup minerals exhibit remarkably similar physical and chemical characteristics among the systems examined, suggesting that the source and geological processes must be extraordinarily similar. Unfortunately, these characteristics are not unique to porphyry systems and such observations should be integrated with additional data, such as trace element mineral chemistry, to effectively discriminate tourmaline supergroup minerals that have formed in porphyry systems.

Tourmaline supergroup minerals (TSM) are common accessories in many geological environments and deposit types (Henry & Guidotti 1985, Slack 1996, Slack & Trumbull 2011). The major, minor, and trace elements and isotopic features of TSM have been widely used as indicators of the primary environments of TSM formation, primarily owing to the wide variation in elements that may be accommodated (Henry & Guidotti 1985, Bosi & Lucchesi 2004, 2007, Marschall & Jiang 2011). In general, TSM occur in many hydrothermal deposits worldwide, including volcanogenic massive sulfide (VMS) (Slack & Coad 1989, Slack et al. 1993, Griffin et al. 1996, Slack 1996), sediment-hosted base metal (Jiang et al. 1996), orogenic gold (Kalliomäki et al. 2017, Manéglia et al. 2018, Sciuba et al. 2021), Sn-W veins and greisens (Launay et al. 2018, Codeço et al. 2019, Codeço et al. 2020), porphyry Cu-Mo±Au (Clarke et al. 1989, Yavuz et al. 1999, Frikken et al. 2005, Baksheev et al. 2010, 2011, 2012, 2017, Byrne et al. 2020), and emerald mineralization (Galbraith et al. 2009). Although several mineral groups have been evaluated as potential indicators of mineralized porphyry systems (e.g., apatite, epidote, and chlorite group minerals; Ahmed et al. 2020, Baker et al. 2020, Cooke et al. 2020a, c, Pacey et al. 2020, Wilkinson et al. 2020), few papers have discussed TSM as an exploration guide for porphyry deposits (Clarke et al. 1989, Baksheev et al. 2012, Li et al. 2022). The study described herein is focused on mineralized porphyry-type settings and builds upon the approaches, data sets, and interpretations made through allied studies, while also comparing and contrasting the features of porphyry TSM to those forming in other deposits, as well as in unmineralized environments.

Studies of TSM from mineralized porphyry systems to date have been deposit-specific, sporadic, and generally lacking a comprehensive overview of how these minerals integrate into the genesis and evolution of these systems. A detailed review of TSM in porphyry systems is absent in the literature and, as such, is summarized here.

Occurrences and textures

Of roughly 550 porphyry systems worldwide that report grade (using the USGS porphyry Cu database; Singer et al. 2008 and data compiled in this study Appendix 11), nearly 30% are reported to contain TSM (Fig. 1). Notable deposits that contain significant concentrations of TSM include the Río Blanco-Los Bronces copper deposit and the El Teniente deposit. Importantly, TSM breccias at the Río Blanco-Los Bronces copper deposit have been empirically considered to be the most favorable host for mineralization due to the porosity and open space created by radiating TSM crystals (Warnaars et al. 1985). Paragenetically, TSM were identified as the first hydrothermal minerals to crystallize in the breccia matrix (Skewes et al. 2003), and a transition from a biotite cement upward into a TSM cement and finally into a quartz-sericite-TSM cement has been noted at the Sur-Sur breccia Río Blanco-Los Bronces copper deposit, Chile (Frikken et al. 2005). At El Teniente, TSM-cemented breccias are observed, which formed by the exsolution of magmatic fluids; TSM are also found in later veins (Skewes et al. 2002). At this locality, TSM occur as black fracture coatings and veinlets with sharp borders, while in the breccias they occur as replacements of wall-rock fragments (also black), or, in some cases, replacing plagioclase feldspars (Skewes et al. 2002). At El Teniente, TSM are found in variable amounts throughout the numerous breccias but are most abundant in the “tourmaline-breccia” (Vry et al. 2010). Veins of TSM commonly cut early A-type veins (defined as early sinuous quartz ± bornite or chalcopyrite following the nomenclature of Sillitoe 2010, and references therein), but definitive relationships with later vein types are unclear (Gustafson & Hunt 1975). Some B-type veins are also reported to commonly contain TSM (Gustafson & Hunt 1975). Molybdenite rarely occurs in proximity to TSM, and most TSM veins do not exhibit sulfide or alteration halos, nor is TSM present in late pyritic veins (Gustafson & Hunt 1975). Little is known about the vertical zonation of TSM in a given porphyry. At the El Salvador deposit (Chile) vertical zoning of TSM is strong with rare TSM veins on the 2400 ft level and an increase in abundance of both TSM veins and disseminations toward the surface where TSM are abundant and widespread as well as reportedly contemporaneous with sulfides (Gustafson & Hunt 1975).

TSM species and major-element variations

Major- and minor-element variations of TSM are dependent on a myriad of factors including (but not limited to) fluid/melt composition (Trumbull et al. 2011), nature of co-crystallizing minerals (Taylor & Slack 1984, Taylor et al. 1999), P-T conditions (van Hinsberg et al. 2011), oxygen fugacity (van Hinsberg et al. 2011), and crystal structure constraints (Bosi 2018). Therefore, the only way to completely understand the variation in the overall elemental composition of TSM is to holistically understand the interplay among the inter- and intra-variables that are in operation during (in the simplest case) crystallization. However, it must also be appreciated that recognizing the most important factors influencing the chemical composition of a TSM may be recognized or recorded in TSM physical or chemical characteristics. For example, processes including oxidation (i.e., through examination of the redox-sensitive elements, e.g., Fe2+/Fe3+), fractionation (variations in Mg#), alteration (changes in alkali content), etc., can all be investigated through examination of the variations in the chemical compositions of TSM. Further, the potential use of such chemical variations to discriminate between TSM arising in differing environments of formation becomes an important, feasible outcome, and it is this latter aspect, specifically with respect to those characteristics of TSM from mineralized porphyry systems, that serves as the locus of the present study.

The most comprehensive study of major-element chemistry of TSM is from seven mineralized porphyry Cu systems in Russia (Baksheev et al. 2012). Results from their study demonstrated that TSM from porphyry systems generally follow an Fe-Mg trend, initially being Fe-rich. They also noted the common occurrence of Fe3+ ↔ Al3+ coupled substitution, leading to the development of an oxy-dravite [Na(MgAl2)MgAl5(Si6O18)(BO3)3(OH)3O] to povondraite [NaFe3+3(Mg2Fe3+4)(Si6O18)(BO3)3(OH)3O] trend (here in referred to as an “O–P trend”), which suggests the presence of Fe3+ (also identified by Fe3+(calc) and confirmed by Mössbauer spectroscopy). This chemical trend is directly relevant to understanding the evolving chemistry of TSM, in that it points toward the role of oxidation. The calculated Mg contents were found to be variable among the differing types of mineralized porphyry systems: in those strictly considered to be Cu-dominant, the TSM have ∼2 Mg apfu, while those that are considered to be Au-porphyries have a lesser amount (1–2 Mg apfu), and those that are Sn-porphyries have even lesser amounts (0–1 Mg apfu) (Baksheev et al. 2012). Tourmaline from the Coxheath porphyry deposit (Nova Scotia, Canada) is restricted to regions of sodic alteration that overprint potassic alteration and conforms to the O–P trend (Lynch & Ortega 1997). It has also been suggested that the TSM demarcate the boiling front in that system (Lynch & Ortega 1997). Variations in TSM major-element chemistry as a function of depth were investigated at the San Jorge porphyry Cu deposit, Argentina (Dill et al. 2012). It was determined that the overall composition of TSM was homogeneous over a 100 m interval but with TSM from so-called fertile breccias showing stronger Fe-Mg correlations (R2 > 0.8) than those from barren breccias (Dill et al. 2012). Results from an examination of TSM from the Hadamio porphyry Au deposit (China) showed the existence of two broad groups of TSM, which were proposed to be related to magmatic and hydrothermal processes (Qiao et al. 2019). Those TSM associated with the magmatic stage of formation occur in the matrix of granite porphyry dikes as disseminated dark brown anhedral grains, whereas those associated with the hydrothermal stage TSM form veins of euhedral radiating aggregates (Qiao et al. 2019). In terms of major-element chemistry, both TSM types classify as schorl–dravite with the hydrothermal grains having lower Ca and higher vacancy (Vac) components (0.089 and 0.206 avg. apfu, respectively) than the magmatic TSM (0.094 and 0.245 avg. apfu, respectively) (Qiao et al. 2019).

This study examined TSM from seven mineralized porphyry systems: Casino (Yukon Territory, Canada), Coxheath (Nova Scotia, Canada), Donoso breccia (Río Blanco-Los Bronces copper deposit, Chile), Highland Valley Copper (British Columbia, Canada), New Afton (British Columbia, Canada), Schaft Creek (British Columbia, Canada), and Woodjam (British Columbia, Canada). These deposits were selected to cover a diverse range in mineralization types (Cu, Cu-Au, Cu-Au-Mo) as well as varying ages, host lithologies, porphyry classifications (calc-alkalic and alkalic), associated alterations, and tectonic settings (Table 1). Host rocks of calc-alkaline affinity vary from diorite (Coxheath) to granodiorite (Highland Valley Copper), with one deposit of alkaline affinity hosted in a monzonite (New Afton). Ages for the deposits examined range from Miocene to Triassic, with one deposit of Precambrian age also being included (Coxheath). As TSM are found to occur in a variety of alteration environments, samples were selected to include potassic (e.g., biotite, K-feldspar), phyllic (e.g., quartz, sericite), sodic-calcic (e.g., albite, actinolite, magnetite), propylitic (e.g., chlorite, epidote, albite, carbonate), and argillic (e.g., quartz, alunite, clays) (Fig. 2). Unfortunately, TSM are not observed in all alteration styles at a single porphyry deposit. Samples of TSM are investigated from ore zones as well as zones of alteration, barren of mineralization.

Casino (Yukon Territory, Canada)

The Casino deposit, located ∼300 km northwest of Whitehorse, Yukon Territory, Canada, is a calc-alkaline porphyry hosted in Late Cretaceous quartz-monzonite and associated breccias (Casselman & Brown 2017). It is unique among Canadian deposits as it has a relatively well-preserved leached cap and supergene oxide zone, which overlie hypogene mineralization (Casselman & Brown 2017). Tourmaline at the Casino deposit is relatively widespread and was first noted by Godwin (1975), who observed it in potassic and phyllic alteration zones. Godwin (1975) also produced a contour diagram based on the observed modal mineralogy that empirically demonstrated a correlation between TSM and Fe-oxides (magnetite/hematite), with the highest concentration of TSM in the core of the deposit and concentrations decreasing away from the core (their Fig. 3.10; Godwin 1975). Breccias with TSM cement are also observed at Casino. These are found at the contact between the older Dawson Range Batholith and the Casino Suite rocks that are the primary ore host and are interpreted to be products of late-stage hydrothermal activity (Bower et al. 1995, Casselman & Brown 2017). The Casino Suite rocks consist of the Patton Porphyry (porphyritic plagioclase intrusive of rhyodacite composition) and associated breccias (Bower et al. 1995, Casselman & Brown 2017). The most common historic report of TSM is in D veins (late quartz-pyrite ± chalcopyrite veins with feldspar-destructive alteration selvages), followed by disseminated TSM in zones of phyllic alteration (Bower et al. 1995). A review of drill logs, conducted in this study, shows that TSM are noted throughout the current deposit outline (including in the leached cap, oxide supergene, and hypogene zones), but are also present in areas distal to the deposit, where mineralization is absent. In hand sample, TSM occur as small millimeter-scale black radiating prismatic grains as disseminations, in veins, and as a breccia cement. Under transmitted light, grains show a range in colors from blue-green to brown and, in some cases, colorless. Grains commonly occur in clusters or aggregates, or more rarely as individual grains.

Coxheath (Nova Scotia, Canada)

The Coxheath deposit is located in Nova Scotia, Canada (Kontak et al. 2001). This deposit of calc-alkaline affinity is hosted in a hornblende-diorite of Precambrian age (621 Ma) and is characterized by Cu-Mo-Au mineralization (Lynch & Ortega 1997). Results from previous studies show that TSM are mainly restricted to the diorite host and occur in veins that create a stockwork associated with albite, which defines a zone of sodic alteration that overlaps the central potassic alteration (Lynch & Ortega 1997, Kontak et al. 2001). In these zones of sodic alteration, quartz, TSM, and albite occur with accessory muscovite, chlorite, and epidote. Rarely, TSM are also found as replacement minerals in samples of wall rock (Kontak et al. 2001). The TSM are spatially associated with sulfide minerals and are thought to have formed synchronously with them (Lynch & Ortega 1997, Kontak et al. 2001). The TSM are massive in hand sample and black in color and can be found as a breccia cement or vein infillings within sodic alteration zones that overlap with potassic alteration. Some late propylitic alteration crosscuts TSM, which supports the early paragenesis of TSM. Under transmitted light TSM are sub- to euhedral and form clots of interlocking crystals that are strongly pleochroic from light brown to green, with some grains also being blueish under plain polarized light.

Donoso breccia – Río Blanco-Los Bronces copper deposit (Chile)

The Donoso breccia is found within the Río Blanco-Los Bronces copper deposit, located in the Andes of central Chile, and represents one of the world's most prolific porphyry districts (Warnaars et al. 1985). Both barren and mineralized breccias are found in the district (Skewes et al. 2003), but those containing TSM or biotite are empirically observed to be associated with most of the copper mineralization. Tourmaline is observed to be paragenetically the earliest mineral to crystallize in the breccias (Skewes et al. 2003). Other significant matrix minerals include quartz, chalcopyrite, bornite, hematite var. specularite, and minor anhydrite (Skewes et al. 2003). The Donoso breccia pipe (late Miocene to early Pliocene age, 5.2–4.9 Ma) is the youngest of the more than 15 mineralized breccias in the region (Warnaars et al. 1985, Skewes et al. 2003). The TSM are dark brown acicular crystals with Fe/Mg ratios between 1.2 and 2.4 (Skewes et al. 2003). In some cases, there is also evidence for TSM replacing primary mafic minerals (e.g., biotite and hornblende; Skewes et al. 2003). The TSM from the Donoso breccia, examined in this study, form as a black (not brown as previous authors have stated) fine-grained breccia cement along with quartz and sulfides. The breccias occur within phyllic alteration zones of the deposit with TSM grains forming as euhedral radiating clots overgrown by sulfides.

Highland Valley Copper (British Columbia, Canada)

The Highland Valley Copper (HVC) district is located 65 km southwest of Kamloops, British Columbia (Byrne et al. 2020). It consists of five major deposits, hosted in the Late Triassic calc-alkaline Guichon Creek batholith: Bethlehem, Highmont, J.A., Lornex, and Valley, all of which have been reported to contain TSM (McMillan & Panteleyev 1988). In the HVC district, TSM generally occur in mineralized (chalcopyrite) and unmineralized veins (with K-feldspar, white mica, chlorite, and sometimes epidote, surrounded by albite alteration halos), as well as TSM-cemented breccias with albite-altered clasts (Byrne 2019). The TSM are found primarily associated with sodic-calcic alteration and more rarely in phyllic alteration zones. Associated minerals include quartz, hematite var. specularite, epidote, calcite, and copper sulfides (McMillan & Panteleyev 1988). The TSM in these deposits are considered to be genetically syn- to post-ore formation, based on the occurrence of TSM breccias containing mineralized clasts (McMillan & Panteleyev 1988). The highest concentrations of TSM have been noted at the Bethlehem and Highmont deposits, where they occur in and around breccia bodies, replacing clasts and as breccia cements (McMillan & Panteleyev 1988). At Bethlehem, TSM are widespread but erratic in distribution (Briskey & Bellamy 1976). They are present within and adjacent to all the ore zones, including quartz-rich breccias (Briskey & Bellamy 1976). Two samples have been investigated, one from the Highmont pit and the other from the Bethlehem pit. In the hydrothermal breccia at the Highmont deposit, TSM forms as a black breccia cement with quartz and hematite. This breccia overprints and is crosscut by sulfide mineralization (Reed & Jambor 1976). Sodic alteration, in the form of albite, is prevalent. Grains of TSM are small (∼50 × 500 μm), sub- to euhedral, and form radiating masses of pleochroic blue-green to brown grains. The material from the Iona pit (a subset of the Bethlehem pit) contains TSM that are small (∼50 × 200 μm), sub- to euhedral, and form radiating masses of pleochroic green to brown grains in plane polarized light.

New Afton (British Columbia, Canada)

The New Afton Cu-Au deposit is located 13 km west of Kamloops, British Columbia (Kwong 1981). It is a silica-saturated alkalic deposit hosted by the Late Triassic Iron Mask batholith. The TSM have been reported in the historic Afton pit (613–640 m level) in association with siderite, prehnite, and magnetite (Kwong 1981). Kwong (1981) stated that TSM only occur in areas adjacent to intense carbonate alteration. While TSM are generally considered accessory minerals, little detail has been paid to the presence of TSM here, and they may be more common than has been reported (D. Wade, pers. commun. 2019). In the current study, samples of TSM were obtained from both the center of the phyllic zone and the inner phyllic zone located near the boundary with the potassic zone. At New Afton, TSM are massive black grains in hand sample that occur as vein fill and breccia cement. In plane polarized light, subhedral grains are green to light brown and weakly pleochroic. In some samples, TSM grains are broken, while others appear dissolved and replaced by carbonate (dolomite). Some grains are overgrown by pyrite, suggesting the TSM are paragenetically earlier.

Schaft Creek (British Colombia, Canada)

The Schaft Creek deposit is located in northwestern British Columbia (Scott et al. 2008). It is a calc-alkaline porphyry Cu-Mo-(Au) deposit hosted by Late Triassic volcanics (Stuhini Group-Stikine Terrane) and associated with the Hickman batholith and associated porphyritic granodiorite dikes. The abundance of TSM is minor in the Main and Paramount zones, but they are slightly more abundant in the West Breccia zone (Scott et al. 2008). In the Main zone, TSM are observed as discrete disseminated grains, while the West zone mineralization comprises sericite-chlorite altered brecciated volcanics cemented by TSM-quartz-carbonate ± sulfate ± chalcopyrite ± pyrite. The TSM are associated with several alteration styles, including potassic, sericitic/chloritic, and propylitic. Samples of TSM are investigated from the central Paramount zone where they occur as massive disseminated black grains in hand sample, vein fill, and breccia cement. In plane polarized light, grains are prismatic, sub- to euhedral, and pleochroic, showing a range of colors including brown, green, and blue.

Woodjam (British Columbia, Canada)

The Woodjam deposits are a cluster of porphyries located ∼50 km east-northeast of Williams Lake (del Real et al. 2017). They include five Mesozoic-aged, high-K monzonite porphyries with calc-alkaline affinities: the Southeast Zone, Deerhorn (DH), Megabuck, Takom (TK), and Three Firs. Deerhorn (Au-Cu) and Takom (Cu-Au) have characteristics of high K calc-alkalic intrusions (del Real et al. 2017). The only deposits that contain TSM are DH and TK, and samples from both were examined. At Deerhorn, TSM are observed in syn- to post-ore Py-Qz-Chl±Hem±TSM veins with distinctive white halos dominated by illite (del Real et al. 2017). The TSM are found intergrown and overprinting potassic alteration associated with minor chalcopyrite mineralization. In the Nicola Group volcanic rocks, which overlie the Takom deposit, a texturally destructive TSM-albite-epidote alteration is present. Tourmaline-cemented breccias occur in the mineralized upper portion of the Takom porphyry as well as TSM is found in quartz-TSM-pyrite veins with albite alteration halos. In hand sample, TSM are black, whereas in plane polarized light they are pleochroic green-brown and, more rarely, blue. Grains are sub- to euhedral radiating prismatic crystals, sometimes replacing (occasionally pseudomorphously) primary magmatic feldspars. Paragenetically, TSM are early, as evidenced by crosscutting relationships with alteration assemblages and their presence as inclusions within sulfides (i.e., pyrite and chalcopyrite).

Chemical analyses (major and minor elements) of 2873 points from ∼750 individual TSM grains in 29 different samples were conducted by scanning electron microscopy energy-dispersive spectroscopy (SEM-EDS) on a JEOL 6400 in the MicroAnalytical Centre at Laurentian University. Operating conditions include an accelerating voltage of 20 kV, a beam current of 1.000 nA, and count times of 30 s. The beam current was consistently monitored and adjusted to be within ± 0.005 nA. The primary standards used included well-characterized materials: diopside (MgKα, CaKα, SiKα), albite (AlKα, NaKα), chalcopyrite (FeKα), and CaTiO3syn (TiKα). Elements commonly occurring in low concentrations in TSM, including K, Mn, V, Zn, F (unreliable due to FeLα-FKα overlap), and Cl, were rarely observed above the limit of detection (LOD) (<0.1 wt.%) and are not reported here. The average internal errors for the elements analyzed include (wt.%) Na2O ± 0.07, MgO ± 0.07, Al2O3 ± 0.13, Si ± 0.17, Ca ± 0.07, Ti ± 0.07, and Fe ± 0.19. The standards were analyzed at the beginning and end of each day as well as periodically throughout the run to ensure precision and accuracy of the analysis. Following strict SEM-EDS protocols, the concentrations fall within ±5% relative of the expected absolute value, which is comparable to the accuracy of electron microprobe analysis (Newbury & Ritchie 2013). Data were processed using Aztec software. Chemical data were reduced to apfu based on 15 cations (Y+Z+T) following the recommendation of Henry et al. (2011) using the program WinTcac (Yavuz et al. 2014). Values of Fe2+/Fe3+ were calculated using the same software using a charge balance approach. Analyses were conducted on TSM grains oriented parallel to the c-axis, given the propensity of the mineral to develop complex chemical zonation about this axis. Based on trace element results (presented in Beckett-Brown et al. 2021), Li is known to be <100 ppm and has therefore not been considered during the formula calculation. The V and W site compositions have been determined following the procedure of Henry et al. (2011).

Tourmaline physical and optical characteristics

The TSM from the porphyry systems in this study exhibit a wide range of colors, sizes, morphology, textures, zonation, mineral associations, and inclusions. A summary of these features is presented in Table 2. The significance of TSM's physical and chemical characteristics is critical in constraining the timing of TSM in relation to sulfides and alteration assemblages in mineralized porphyry systems. Sulfides are commonly found overgrowing prismatic to acicular, sub- to euhedral TSM crystals, suggesting that sulfides post-date TSM formation (Fig. 3). In all porphyry deposits examined in this study, TSM in general forms paragenetically earlier than the associated mineralization and alteration. Development of later generations of TSM have been observed in some deposits; it is rare and is generally the result of recrystallization/dissolution-reprecipitation of pre-existing TSM.

Color (hand sample and optical microscopy).

Macroscopically, TSM ubiquitously occur as black masses in hand sample from mineralized porphyry systems. In rare cases, TSM can be colorless, but only when viewed perpendicular to the c-axis using a binocular microscope. While TSM color can show a wide range in hand sample (blue, green, red, pink, yellow, etc.), those examined in this study do not show any of these other colors, a situation quite different from many TSM developing in felsic pegmatite environments (Hawthorne & Dirlam 2011). The strong coloration (black in hand sample) of TSM from porphyry systems reflects the predominance of Fe and Mg. Under plane-polarized light, the TSM are pleochroic and exhibit a range of colors, most commonly green to brown and blue, although, more rarely, they can be colorless and non-pleochroic. Grains also exhibit core–rim zonation that is most dominantly defined (looking at cross-sections perpendicular to the c-axis) by shades of green and more rarely brown, blue, and colorless. A range of zonation types exist, with general abundance decreasing in the order concentric > sector > patchy/irregular > overgrowth/replacement, with these features best observed under backscattered electron imaging (BSEI). The concentric zonation exhibits sharp boundaries between distinct color zones and very rarely shows gradational boundaries between zones. No identifiable patterns among samples and deposits or core–rim zonation from transmitted light color could be discerned. Grains are observed forming with light green to colorless cores which sharply transition to green zones and back to lighter green rims, while other grains (in some cases in the same sample) exhibit dark green to brownish cores and sharply transition to green rims.

Individual grains: Size, morphology, occurrence.

In hand sample, TSM typically occur as radiating aggregates of fine-grained, sub- to euhedral crystals (μm scale), and more rarely as coarse grains (millimeter scale). The TSM develop in three distinct textural styles within a porphyry system, including breccias, veins, and disseminations (Fig. 4). In breccias, TSM develop in the matrix, cementing wall-rock breccia fragments (Fig. 4a–e). It is also quite common to observe both quartz and sulfides (i.e., chalcopyrite and pyrite) in the matrix, but these post-date the TSM. The vein-style TSM (Fig. 4f–j) are the smallest, typically ∼20 μm in diameter and rarely >100 μm. The veins are mineralogically simple, containing TSM with quartz and sulfides (i.e., pyrite and chalcopyrite), although other minerals (e.g., epidote, apatite, and calcite) may also be present in minor proportions. The TSM commonly nucleate on the wall rock and project into the vein, more rarely nucleating within the vein matrix but not exhibiting any preferential growth orientation. The veins (millimetric to centimetric in vein thickness) also contain paragenetically later quartz and sulfide. In hand sample, the veins can exhibit bleached white selvages, consisting of primarily alkali feldspar and quartz, although this is not universal (Fig. 4g, i). Some TSM veins exhibit sharp wall-rock boundaries, whereas others exhibit irregular contacts. The presence of TSM in veins is not useful in determining porphyry vein style classification, as they can form in multiple types, including B and D veins (Gustafson & Hunt 1975, Sillitoe 2010). Disseminated-style TSM are recognized by their occurrence as isolated clots/patches or, in rare cases, as individual grains (Fig. 4k–o). The crystals of TSM are the coarsest observed, with grains up to several centimeters in length. Texturally, disseminated TSM closely resemble those of breccias (i.e., acicular to prismatic radiating masses, the interstices of which are commonly infilled by quartz), but with TSM developing in discrete individual subhedral clots/patches, rather than large radiating interlocking masses as in TSM breccias. Here, TSM forms clots/patches/masses of radiating sub- to euhedral prismatic interlocking (or in rare cases acicular) grains. Rarely, TSM are observed as isolated individual grains. Clusters of TSM have been observed pseudomorphically replacing (or infilling) primary porphyritic feldspars and mafic minerals (mainly biotite, and more rarely amphibole). Owing to their chemical resilience, TSM are unaffected by subsequent alteration; only in cases where earlier-formed TSM have been brecciated is there evidence of recrystallization. This can be explained by the modification of primary grains that are broken/dissolved (or possibly melted, akin to the textures observed by van Hinsberg 2011) and recrystallized.

Mineral associations and relationship of TSM to alteration style.

The types of minerals forming in association with TSM in porphyry deposits in this study are highly variable and partially dependent on the host rock composition, as well as the style of alteration in which the TSM are present. Important associated minerals include (but are not limited to): quartz, feldspars (alkali feldspars and plagioclase feldspars), rutile, epidote, apatite, ilmenite, hematite, biotite, chlorite, carbonates, and sulfides (mainly chalcopyrite and pyrite) (Table 2).

Tourmaline can be observed throughout the typical alteration profile of a porphyry deposit but is most common in the potassic (e.g., Woodjam) and phyllic (e.g., Casino) zones, and more rarely occurs within the sodic-calcic (Highland Valley Copper), propylitic (Woodjam), and argillic (Casino) alteration zones (Fig. 2). In zones of potassic alteration, TSM occur as disseminations or mineral replacements and as veins with potassic alteration halos. Veins of TSM with potassic (bleached white alkali feldspar) alteration halos are most common (e.g., Byrne et al. 2020, Escolme et al. 2020). Additionally, TSM can be found less commonly in regions of significant biotite alteration as disseminated grains or clots of multiple grains. In zones of phyllic alteration, TSM form as veins and pods associated with quartz, white mica, and pyrite (e.g., Baksheev et al. 2012, Bozkaya et al. 2020). In the calcic-sodic zone, TSM form in veins with quartz, albite, and actinolite (e.g., Carten 1986, Byrne et al. 2020). In zones of propylitic alteration, TSM form as veins and selective alteration in association with epidote, chlorite, quartz, and carbonate (e.g., Baksheev et al. 2012, Li et al. 2022). Finally, TSM are also found in zones of argillic alteration, in association with clay minerals (Li et al. 2022).

Mineral inclusions.

Mineral inclusions in TSM from porphyry deposits are relatively limited (<1% of TSM grains investigated) in terms of type and abundance based on the samples examined in this study. Common mineral inclusions in the samples in this study are zircon, rutile, Fe-oxides, apatite, albite, quartz, titanite, epidote, baryte, chalcopyrite, and molybdenite (Table 2, Fig. 5). Due to the paucity and irregularity of the inclusions and inclusion populations, definitive statements cannot be made. Zircon and rutile were noted in TSM in multiple samples from multiple deposits, while other mineral inclusions are considerably rarer. It is noteworthy that no sulfide inclusions were observed, further supporting the early paragenesis of TSM. Only sulfides developed on later-formed fractures are noted.

Internal textures in TSM: Backscattered electron imaging.

The TSM generally display a high degree of chemical heterogeneity in terms of both their major- and minor-element characters. Examination of TSM using BSEI reveals four distinct textural features, with two considered to be primary in origin: (1) oscillatory/concentric growth zoning, recognized by the development of sharp chemical boundaries developed parallel to [001] (Fig. 6a), and (2) sector zoning, recognized by the partitioning of elements into distinct sectors (Fig. 6b). Oscillatory zonation is interpreted to record rapidly changing fluid conditions at the mineral surface scale, potentially due to surface-limited diffusion from fast crystallization rates (Dutrow & Henry 2018). Similarly, sector zonation forms under conditions of rapid growth such that equilibrium between different sectors cannot be maintained (Henry et al. 1999). These sectors initially form when elements are preferentially incorporated at specific growth surfaces and survive throughout crystal growth (van Hinsberg 2006). Further, uncommon textures considered to be secondary in origin were also observed: (1) patchy/irregular growth common in recrystallized/brecciated TSM, with diffuse grain boundaries, developed perpendicular to [001] (Fig. 6c), and (2) overgrowth or replacement features (Fig. 6d). The development of regions of high porosity, giving rise to textures similar to those observed in plagioclase (Putnis 2009), are considered products of coupled dissolution-reprecipitation. Such textures could be included in the patchy/irregular group. These dissolution-reprecipitation textures correspond to major-element changes, primarily involving Fe and Mg (i.e., Mg# variations), typically reflected in the replacement of early Fe-rich (schorl) by Mg-rich (dravite) overgrowths. This observation is somewhat counterintuitive in that Mg is the smaller cation and would preferentially be incorporated before larger ones (in this case Fe). Variations in chemical zonation could reflect the relative abundance of these two elements in the fluid, competition for Fe with other co-crystallizing phases (e.g., sulfides or oxides?), or a fast rate of crystallization.

Mineral chemistry

Results from SEM-EDS analyses and the recalculated empirical formula (summary of data in Table 3) show that the majority of the TSM analyzed (>80%) are dravite, with the remainder being schorl. All belong to the Na-dominant (Fig. 7a), alkali TSM group, with the majority following the oxy-dravite–povondraite (O–P) trend (Fig. 7b) and, to a lesser extent, the schorl–dravite trend (primarily the Casino samples; Fig. 7c). A significant portion of the TSM analyzed show Fe3+(calc), suggesting a moderate relationship with prevailing oxidizing conditions. Analyses with the highest Fe3+(calc) concentrations are associated with the most Fe-rich and Al-depleted regions of individual grains. The TSM from all the porphyry systems analyzed contain some Fe3+(calc), with those from Casino having the fewest number of analyses with calculated Fe3+(calc) (∼2% of analyses, totaling 23 grains). A limited number of analyses (n = 203) from the Donoso breccia contain the highest calculated Fe3+(calc) values (1.38 apfu avg.), with a rare number (n = 3) that contain sufficient Fe3+ to classify as povondraite. The predominance of Fe3+ could reflect the oxidized nature of early porphyry fluids or could arise via boiling mechanisms (Yang & Jang 2002).

The variation in major- and minor-element concentrations in the TSM from the various deposits examined in this study is remarkably small, despite there being relatively large differences in age, host lithology, porphyry classification (calc-alkalic and alkalic), associated alteration, tectonic setting, and metal associations. The average major-element composition for all porphyry-related TSM (n = 2873 analyses from ∼750 grains) analyzed (range in wt.%) is as follows: Na 1.71 (0.64–2.36), Ca 0.69 (<LOD–2.64), Ti 0.34 (<LOD–2.15), Mg 4.76 (1.60–7.29), Fetot 6.47 (0.20–24.85), Al 15.95 (6.45–20.09), and Si 16.91 (14.42–18.68). The averages (range) in apfu are as follows: Na 0.74 (0.28–1.04), Ca 0.15 (<LOD–0.70), Vac 0.12 (nil–0.54), Ti 0.05 (<LOD–0.45), Mg 1.94 (0.69–2.89), Fetot 1.17 (0.03–4.97), Al 5.85 (2.69–7.14), and Si 5.98 (5.53–6.15). Results show that, based on occupancy of the X-site, the TSM analyzed predominantly plot in the alkali-dominant field, with a limited number (<2%) plotting in the X-vacant and calcic fields (Fig. 7a). The greatest compositional variation in terms of major and minor-element chemistry is observed in the Y- and Z- crystallographic sites; these sites house Al as well as other metals, predominantly Mg and Fe (Fig. 7b, 7c). Calculated Fe2+ and Fe3+ contents show an average of Fe2+ 0.94 apfu (range: 0.03–2.41) and Fe3+ 0.24 apfu (range: <LOD–3.20). No regular or systematic core–rim chemical changes could be discerned at the deposit scale or among the entire data set, based on the identification of grains with core–rim zonation, and comparing the Al-Mg-Fe values did not show any repeatable patterns. Additionally, no difference in TSM chemistry exists when the data were sorted based on the TSM occurrence (i.e., breccia, vein, and disseminated), suggesting that macroscopic textures have no control on major- or minor-element chemistry.

The TSM exhibit variable chemical zonation within a single bedrock sample, between different samples from the same deposit, and between deposit types. Among the TSM analyzed from the different deposits examined, some minor differences can be discerned (Fig. 7 and Fig. 8): (1) the TSM from Casino show the highest average Al (6.31 apfu) among the deposits examined and the greatest proportion of X-site vacancies (0.17 apfu), along with the lowest average concentrations of Ti (0.03 apfu), Fe (0.62 apfu), and Ca (0.06 apfu); (2) those from Coxheath are not remarkable in their chemistry and are intermediate schorl–dravite in composition; (3) those from the Donoso breccia show the highest average Na (0.81 apfu), Fetot (2.95 apfu), and Fe3+calc (1.38 apfu) and the lowest Al (4.47 apfu); (4) those from HVC are intermediate in composition; (5) those from New Afton show a high average Ti (0.09 apfu); (6) those from Schaft Creek contain the highest average Ca (0.30 apfu) and Mg (2.15 apfu) and lowest X-site vacancies (0.02 apfu), as well as high average Ti (0.09 apfu); and (7) those from Woodjam contain the lowest average Na (0.65 apfu).

Samples of TSM are observed in a wide range of alteration styles (Fig. 2). Although hand sample and petrographic observations indicate TSM to be, in general, paragenetically earlier than all types of alteration (in the samples in which they form), an examination of major and minor elements was made to investigate whether any correlations exist among alteration styles (Fig. 9). Some minor differences in major and minor elements can be observed among TSM and the alteration assemblages in which they are found. Analyses of TSM from zones of potassic alteration do not show any unique differences in major or minor-element chemistry in comparison to TSM from other alteration zones. However, those from Na-Ca alteration zones exhibit unique mineral chemistry, containing the highest average Fe (1.85 apfu) as well as high Ca (0.26 apfu) coupled with the lowest average Al (5.26 apfu). Grains forming in phyllic alteration zones show the highest average Na (0.76 apfu), coupled with low average Ca (0.10 apfu). Grains of TSM forming in propylitic alteration show the highest Mg (2.14 apfu) and high Ca (0.31 apfu) values. Finally, argillic alteration shows the highest Al (6.47 apfu) and Vac (0.20 apfu), coupled with the lowest average Fe (0.61 apfu) and Ca (0.08 apfu) values. The contrast in major-element chemistry between phyllic and propylitic TSM is comparable to that presented by Baksheev et al. (2012); here, TSM associated with phyllic alteration were noted to be enriched in Na, Vac, and Al, while propylitic TSM has a higher average Ca and Fe.

Tourmaline textural variations

The physical characteristics, including color, grain size, morphology, color in plane-polarized light, mineral associations, and inclusions found in TSM, are highly similar among the seven mineralized porphyry systems examined in this study. In hand sample, all TSM appear black, but can range from colorless to dark brown to black under the binocular microscope. Although TSM can occur in a wide range of colors in other geologic environments (Hawthorne & Dirlam 2011), color alone cannot be used to identify TSM derived from a mineralized porphyry system, but it can be an effective first pass in separating prospective grains (i.e., dark brown to black in hand sample). However, as color is, in part, related to chemistry, and considering the uniformity in color observed in the TSM analyzed from the seven porphyry systems examined in this study, the implication suggests a uniformity in TSM genesis in mineralized porphyry systems. TSM occur in relatively uniform grain sizes ranging from microns to millimeters in diameter, form most commonly as sub- to euhedral radiating prismatic grains in groups or clusters, and are very rarely found as isolated individual crystals. The uniformity in grain size could reflect similar cooling rates or degrees of undercooling (Kirkpatrick 1975). Under polarized light, TSM are moderate to weakly pleochroic and range in color from light green to blue, light brown to brown, and are even colorless and non-pleochroic in some rare cases. Optical zonation is ubiquitous as fine-scale (μm) concentric zonation perpendicular to the c-axis. As mentioned, TSM exhibit a range of internal textures (zonation), which in some cases can be recognized using plane-polarized light, noting changes in color, but is most evident with BSEI, which highlights both major and more subtle changes in mineral chemistry. The two primary zonation types (concentric and sector) are observed. Concentric zonation (most common) is evident as distinct fine (micrometer scale) chemical zones (primarily Al-Mg-Fe exchange) that are separated by sharp and not gradational chemical variations. Concentric zonation in minerals can result from (1) changes in the chemical composition of a melt or fluid or (2) crystal face kinetics and diffusion limitations of a chemical species (García-Ruiz & Otálora 2015). Sector zonation results from the preferential incorporation of elements on specific growth surfaces (van Hinsberg & Schumacher 2006). Additionally, sector zonation indicates rapid crystallization such that equilibrium has not been maintained (Henry et al. 1999). Considering the variation in chemistry between zones (Fig. 6a, b), it is unlikely to reflect changes in bulk fluid composition (based on the inconsistency of TSM zoning chemistry of paragenetically simultaneous grains), but more likely records local changes in chemistry at the mineral surface during rapid crystal growth. The bulk of TSM formed during a single, major fluid episode with negligible/minor remobilization and precipitation during later fluid events, which is reflected in the abundance of primary zonation features in TSM and the low abundance of secondary zonation features.

Samples of TSM recovered from till and stream sediments surrounding known porphyry deposits have been shown to contain abundant inclusions, which provide critical information about their potential source (Beckett-Brown et al. 2021). Mineral inclusions are rare in occurrence (<1% of grains) for TSM in mineralized porphyry systems examined in this study, but, when present, zircon and rutile are the most common. As such, the integration of such physical (such as inclusion types presented here) and chemical data (major and minors discussed below and traces in Beckett-Brown et al. 2023) of these minerals in mineralized porphyry systems could provide valuable insight into both the formation of these systems and the exploration for them (e.g., their use as indicator minerals in sediment samples). Inclusions in TSM are only significant if they are hydrothermal in origin and are coeval with TSM formation. Mineral inclusions in TSM are also present in other environments. Quartz inclusions in TSM from granites, schists, and gneiss have been observed (van Hinsberg & Schumacher 2011, Marger et al. 2019, Cheng et al. 2021). Samples of TSM from metamorphic rocks such as tourmalinites and serpentinites have been reported to contain inclusions of zircon and titanite (Dietrich 1985). Pegmatitic TSM have also been reported to contain inclusions such as apatite and rutile (Dunn 1977, Dietrich 1985). Grains of TSM from orogenic Au deposits occasionally contain inclusions of quartz, calcite, chlorite, and amphibole (e.g., Sciuba et al. 2021). Inclusions of apatite, rutile, monazite, and xenotime have also been reported in TSM from granite-related Sn deposits, e.g., the world-class San Rafael deposit in Peru (Harlaux et al. 2020). In VMS deposits, TSM are reported to contain inclusions of quartz, zircon, and sphalerite (Slack & Coad 1989). Thus, the inclusion assemblage could be an important first pass filter to examine large quantities of TSM, which may indicate porphyry mineralization, therefore warranting further investigation, whereas those that contain abundant quartz inclusions should be omitted from further investigation.

Tourmaline chemical variations

Given the complexity of the crystal structure of TSM and their ability to incorporate a wide range of elements (alkalis, HFSE, semi-metals, etc.), it is necessary to consider the various sites possible and their occupants as revealed in this study. Beginning with the X-site, TSM chemistry in porphyry deposits may, in principle, be linked to the specific alteration assemblage (potassic, calcic-sodic, phyllic, propylitic, or argillic) with which they are associated even though, in many cases, the TSM form prior to the alteration minerals that characterize these assemblages. The contents of the X-site (0.74 Na apfu avg.) in TSM forming in porphyry systems are similar, with >95% of analyses plotting in the alkali-dominant field (Fig. 7a). However, correlations exist between X-site populations and alteration assemblage. For example, TSM associated with sodic-calcic and propylitic alteration contain higher average proportions of Ca than other alteration types with contents of 0.26 and 0.31 apfu, respectively, potentially reflecting the Ca-rich nature of the fluids. Phyllic alteration TSM can be identified by high average concentrations of Na (0.76 apfu) coupled with low Ca (0.10 apfu) that could reflect the low Ca and higher Na contents of the fluid. The TSM associated with argillic alteration zones contain the highest average Vac (0.20 apfu), reflecting the lower alkali contents of these environments. There are no distinct major-element differences for TSM developing in potassic alteration. This lack of difference could reflect the timing of TSM formation (i.e., before potassic alteration), but could also reflect the extreme difficulty of incorporating K into TSM, which is a function of P, with high P needed to incorporate any meaningful concentrations of K (Berryman et al. 2015). Thus, in mineralized porphyry systems, concentrations of K in TSM cannot be used as a guide for identifying zones of potassic alteration. The X-site composition of TSM forming in other environments can be similar, with grains (on average) from orogenic Au deposits (0.64 apfu Na: Sciuba et al. 2021), VMS (0.64 apfu Na: Slack & Coad 1989, Slack et al. 1993), orbicular tourmaline granites (0.74 apfu Na: Andreozzi et al. 2020), W-Sn-Cu polymetallic deposits (0.65 apfu Na: Codeço et al. 2017; a similar Na proportion was observed by Harlaux et al. 2020), and even many pegmatitic TSM (0.61 apfu Na: Gadas et al. 2012, López de Azarevich et al. 2021) dominated by Na in the X-site. X-site vacant dominant TSM, including Mg-foitite, foitite, rossmanite, and oxy-foitite, form in unique environments (e.g., alkali-deficient portions of pegmatites), but their occurrence is extremely rare (e.g., Bačík et al. 2015) and they are not found in mineralized porphyry systems. Calcium dominant X-site TSM (e.g., uvite, feruvite, fluor-uvite, fluor-liddicoatite, lucchesiite, and adachiite) form in Ca-rich environments generally associated with mafic to ultramafic rocks (e.g., Jiang et al. 1996). The X-site of TSM can contain Na, Ca, and Vac components in major concentrations, with those forming in mineralized porphyry systems being exclusively sodic dominant.

Guides for exploration

In many systems, TSM have been regarded as passive accessory phases that reflect changes in the local environment and do not control it (Henry & Dutrow 2018). The majority of porphyry TSM major-element variability involves Al, Mg, and Fe at the Y- and Z-sites and has served as the basis for the development of the Al–Fe(tot)–Mg classification ternary diagram (Henry & Guidotti 1985), which has in turn been used as an environment of formation discriminator for TSM. As these fields only denote TSM forming in non-mineralized settings, what becomes key are the trends in the data for porphyry TSM (Fig. 7b). Results from SEM-EDS show that TSM from porphyry systems plot on the O–P trend, reflecting the substitution of Al3+ for Fe3+ with reasonably constant Mg contents (∼2 apfu, 1.94 ± 0.29 apfu). This finding is consistent with those from previous studies (e.g., Lynch & Ortega 1997, Rabbia et al. 2003, Baksheev et al. 2012), indicating that TSM from all porphyry systems must arise from similar geochemical and geological processes. The O–P trend can be observed in other non-porphyry environments (Henry et al. 1999, Kalliomäki et al. 2017), but it is distinctive relative to the more common schorl–dravite substitution trend observed for TSM in other environments, e.g., orogenic (Sciuba et al. 2021), W-Sn-Cu polymetallic deposits (Codeço et al. 2017, 2019, 2020), pegmatitic (Novak et al. 2011, Gadas et al. 2012), and VMS (Slack & Coad 1989). The O–P trend that characterizes TSM that forms in porphyry systems may distinguish it from other deposit types listed above, but not from other environments that also display the O–P trend (e.g., Orogenic TSM). It is important to note that the type locality for povondraite is from a schist near Villa Tunari, Bolivia, with analyses plotting along the O–P trend (Grice et al. 1993). The fact that TSM from porphyry deposits chemically follow the O–P trend highlights the essential nature of the Al3+ ↔ Fe3+ substitution and at least the partial role in their formation. If the porphyry systems are grouped based on their emplacement depth, it reveals that deposits that form at depths >2 km contain higher Fe3+(calc) (i.e., plot closer to povondraite, Fig. 7b) than samples that form closer to the surface (<2 km) which plot closer to oxy-dravite (i.e., contain less Fe3+ and more Al3+). This O–P trend displays a transition from early oxidized conditions toward more reducing conditions, potentially reflecting the change in oxidation of arc magmas during their evolution and ascent (Sun et al. 2015). Conversely, this trend could reflect conditions or initiation of boiling, similar to the interpretations of other authors (Lynch & Ortega 1997, Baksheev et al. 2011). Given that Al is relatively fluid-immobile, the observed enrichment in TSM could suggest a local source rather than derived from an external fluid. The hydrothermal alteration of primary magmatic minerals (e.g., feldspars, biotite, and amphibole) could contribute to the Al required to form TSM.

Challenges with tourmaline in porphyry Cu systems

There exists debate within the literature regarding the advent of subduction and specifically its occurrence in the Precambrian (Wyman et al. 2002, Bédard et al. 2013). Porphyry deposits in the Precambrian have been used to infer the presence of subduction (e.g., Malanjkhand, India; Stein et al. 2004). A comparison of the TSM from a Precambrian porphyry-like system may shed light on the link between Precambrian and more recent “typical” porphyry systems. Due to the robust nature and high closure temperature, TSM are ideal phases to assess similarities and differences among Phanerozoic and Precambrian systems. The Precambrian Coxheath Cu-Mo-Au system age is problematic when it comes to the initiation of modern-style subduction. Importantly, Coxheath shares many features similar to typical porphyry systems (e.g., large concentric zoned alteration, continental margin, stockwork mineralization, mineralized breccias, etc.). The TSM from the Coxheath deposit share many similarities with TSM characteristics from typical porphyry systems. This similarity includes color in hand sample and under transmitted light, grain size and morphology, inclusion types, absolute major- and minor-element chemistry, and porphyry major-element chemical trends (i.e., the O–P trend), as well as trace element analyses (Beckett-Brown et al. 2023). The overwhelming similarity in TSM physical and chemical characteristics indicates that they have similar geological processes of formation. The implication is that porphyry or porphyry-like systems may be more common in the Precambrian than was previously thought and that the TSM found in these systems have developed under broadly similar conditions to Phanerozoic porphyry systems. Subsequently, TSM could help identify possible porphyry systems in the Precambrian where there still exists debate as to their origin.

Although historically they have not been regarded as a significant accessory phase, considering the number of deposits including major deposits that report them (Fig. 1), TSM should be considered important accessory phases. The recognition of boron-bearing minerals, notably TSM, is likely under-reported considering the abundance of boron (∼500 ppm avg) observed in porphyry fluid inclusions (Wagner et al. 2016). There are no apparent correlations of Cu concentration or age versus ore tonnage when the presence or absence of TSM is considered (Fig. 10). It is important to note the binary relationship of these diagrams and that modal abundance, chemistry, texture, etc. are not assessed, thus the interpretations being made are limited. Considering the direct overlap of data for deposits that contain TSM and those that do not, it begs the question of why some deposits do not contain TSM. Two reasons could explain this. The first is observational, such as TSM are not reported in available literature, potentially misidentified for other minerals (epidote, amphibole, biotite), or simply overlooked. The second reason, process related and equally as likely, is that the magmas are not sufficiently boron-rich to form TSM, which is a direct reflection of the source of melting (Leeman et al. 1994, Jones et al. 2014). A detailed study, especially of notable systems which do not contain TSM (e.g., examining drill logs or re-logging key intervals), could significantly increase the number of porphyry systems that have been reported to contain TSM as an accessory phase. If TSM prove to be a significant phase in mineralized porphyry systems, their presence in turn reflects a contribution of boron in the system. Considering that boron (in the form of borax) is used as a flux to reduce the melting point of metals for metallurgical processes (Davies 1992), perhaps it plays a similar role in the complexing of metals in porphyry mineralizing fluids. Further work is needed to confirm or refute the potential of boron as an important ligand for transporting metals in porphyry systems.

Tourmaline should be considered a significant accessory phase in mineralized porphyry systems due to its repeated occurrence in many porphyry systems worldwide (Fig. 1) and its presence highlights the importance of hydrothermal boron in crustal ore-forming processes.

The physical characteristics of TSM (e.g., color, grain size, morphology, color in PPL) can provide some diagnostic, but not decisive, information that can be used in distinguishing porphyry-related TSM from those forming in other environments. Although TSM major-element chemistry is commonly reported, it has not been demonstrated to be a definitive tool in discriminating TSM forming in mineralized versus unmineralized settings. The following are new insights from this study about TSM forming in mineralized porphyry systems that may be useful in discriminating between environments:

  1. All porphyry-associated TSM studied are exclusively black to dark brown at the macro-scale, occurring as radiating euhedral to subhedral masses of interlocking grains, and are almost never observed as individual isolated grains. In transmitted light, grains are weakly pleochroic and range from green, brown, and blue, to colorless. While inclusions are rarely observed, when present, zircon and rutile are most common.

  2. In most porphyry systems, TSM form paragenetically early and exhibit a wide range of zonation textures including concentric, sector, patchy, and overgrowths, with concentric being by far the most common.

  3. Major-element chemical trends of TSM forming in porphyry systems are not unique compared to TSM forming in other systems (e.g., some orogenic Au TSM), but among the porphyry systems presented in this study, and those presented in the literature, they are remarkably consistent. The grains are dominated by Na in the X-site but can contain minor concentrations of Ca and Vacancy components. They are Mg-Fe rich, ranging from predominantly dravite to schorl in composition with remarkably consistent Mg apfu values of ∼2.0 (1.94 ± 0.29) and follow the O–P trend.

  4. Physical and major-element data alone cannot definitively distinguish TSM forming in mineralized porphyry systems. However, the integration of physical and chemical data (including trace element data, e.g., Beckett-Brown et al. 2023) provides an effective method for distinguishing TSM that has formed in mineralized porphyry systems from that originating in any other geologic environment.

The following companies provided sample material: Western Copper and Gold, New Gold, and Gold Fields Ltd. (now owned by Consolidated Woodjam Copper Corporation). Scott Casselman, Yukon Geological Survey, is thanked for providing geological information, advice, and access to samples from deposits in Yukon. John Chapman, Martin McCurdy, and Dave Sinclair (Geological Survey of Canada) are thanked for their contributions. Western Copper and Gold are thanked for providing logistical support to access the Casino deposit and for sharing geological information. This study was funded by the Geological Survey of Canada Targeted Geoscience Initiative – 5. The lead author was additionally funded by the NSERC-CGS-Doctoral (2018–2021) and additional support from the Society of Economic Geology Research Grant (2018–2019) and Student Fellowship (2017). This paper benefited from detailed reviews by D. Petts, M. Harlaux, and N. Zajzon as well as the editorial handling of Dr. Camprubi. Natural Resources Canada contribution number 20220492.

Supplementary Data are available from the Depository of Unpublished Data on the MAC website (, document “Porphyry Tourmaline, CM61, 22-00011”.
This manuscript was handled by Associate Editor Antoni Camprubí and Editors Lee Groat and Stephen Prevec.