Geology, Geochemistry, and Potential Origins of the Basin Volcano-Sedimentary Lithium Deposit, Kaiser Spring Volcanic Field, Northwestern Arizona, U.S.A. Open Access

Globally, lithium (Li) has been obtained from four deposit types. The first major, and more traditional, source of high-grade Li and other rare metals is the lithium-cesium-tantalum (LCT) family of granitic pegmatites with Li ore minerals including spodumene, petalite, and lepidolite. Examples of well-known deposits include Tanco, Canada, and Kings Mountain, North Carolina, USA (London, 2008). A second source is Li-F–rich granite (Tkachev et al., 2015, 2018). A third major Li source are low-grade, large-tonnage continental brines, with the Li dissolved in hypersaline solutions hosted in closed, evaporitic basins, such as Salar de Atacama, Salar de Uyuni, Salton Sea, and Silver Peak (Munk et al., 2016). In recent years, a fourth Li source, volcano-sedimentary deposits, is intermediate in both grade and tonnage between pegmatite and brine deposits (Fig. 1; Table 1). Examples include the Li-rich clays at McDermitt caldera/Thacker Pass, Nevada; Big Sandy Valley, Arizona; Sonora, Mexico; Clayton Valley, Nevada; and the subject of this paper, Basin, Arizona.

In fact, the largest Li resource identified in the United States is the hectorite- and illite-type clay mineralized in the Nevada’s McDermitt caldera (Vikre et al., 2016; Benson et al., 2017, 2023; Castor and Henry, 2020; Ingraffia et al., 2020). In Jadar, Serbia, one of the largest single ore-grade Li deposits in the world is hosted in jadarite-bearing lacustrine evaporite-type sediments (Putzolu et al., in press). Other examples include sequences within the Mud Hills and Calico Mountains of the Barstow Formation in California (Gagnon et al., in press), the Cave Spring Formation in Cave Spring basin near Nevada’s Rhyolite Ridge (Chafetz, 2021; Ogilvie, 2022; Darin et al., in press), the Big Sandy Formation in Big Sandy Valley, Arizona (Sheppard and Gude, 1972; Ferguson et al., 2023; Gootee and Johnson, 2023; Gootee et al., 2024), and polylithionite clay mineralized in tuffaceous sediments in Sonora, Mexico (Pittuck et al., 2018). Much attention has been devoted to investigating the geology, structure, sedimentology, and sedimentary architecture of these, and other volcano-sedimentary depositional environments, to better understand the variable geologic settings in which Li-mineralized sediments can form—including alkaline volcanogenic lakes and playa lakes in extensional basins proximal to or containing felsic volcanic rocks.

In northwestern Arizona, there are three volcano-sedimentary Li deposits located within 70 km of one another: Big Sandy, Lyles clay (now referred to as Thompson Valley), and Basin, Kaiser Spring volcanic field (Fig. 2; App. Table A1). The Big Sandy deposit is hosted within the late Miocene to earliest Pliocene mudstone-dominated lithofacies of the Big Sandy Formation, which accumulated in a modified extensional half graben (Sheppard and Gude, 1972; MacFadden et al., 1979; Hawkstone Mining, 2019a, b; Ferguson et al., 2023; Gootee and Johnson, 2023; Garcia et al., 2024). The Thompson Valley deposit is hosted with late Miocene to Quaternary (?) lacustrine sedimentary and tuffaceous rocks (Norton, 1965; Eyde, 2012; Köster and Gilg, 2018), but nothing is known about the basin geometry in which these rocks were deposited. The same was true for the generically named Basin volcano-sedimentary Li deposits until new geologic mapping (Thompson et al., 2023) and recent drilling by Bradda Head Lithium, Ltd. revealed sedimentological and stratigraphic evidence for a buried Miocene-aged maar crater filled with Li-rich clay, pyroclastic rocks, and basalt flows. Herein, we present a framework geologic study of the Basin stratigraphic architecture, the sedimentology and geochemistry of the Li-rich clay deposits within, and evidence for maar crater basin geometry. Existing and new surface and drill core geochemical data and borehole sedimentological logs have been integrated with geologic mapping to characterize the Basin stratigraphic package, which importantly may contain economically extractable Li-, K-, and Mo-rich claystone.

Basin is a small (~70 km²) depocenter located at the southeastern margin of Kaiser Spring volcanic field (~400 km²) at the southeastern end of the Aquarius Mountains (Fig. 2).

It lies within both the Transition zone physiographic province and the boundary zone between the Mojave and Yavapai Paleoproterozoic tectonic provinces (Spencer et al., 2001; Duebendorfer, 2015). The Mojave-Yavapai boundary zone has Pb isotope signatures intermediate between the two provinces and is defined by locally exposed shear zones and/or multiple isotopic, geochemical, metamorphic, and geophysical discontinuities (Duebendorfer, 2015). The bedrock through which the Basin has been excavated is composed of Paleoproterozoic-granitic rocks (~1.73–1.66 Ga) that are locally separated by screens of older Paleoproterozoic metamorphic rocks (Bryant et al., 2001). A belt of amphibolite-grade metamorphic rocks extends from the Bagdad porphyry copper mine south-southwest to the Poachie Range, with a defined stratigraphy from oldest to youngest of the mafic volcanic rocks of the Bridle Formation, felsic volcanic rocks of the Butte Falls tuff, and the Hillside Mica schist (Bryant et al., 2001). A rhyolitic unit called the Dick rhyolite was interpreted to intrude this sequence near Bagdad by Anderson et al. (1955), but it was reinterpreted as extrusive and overlying the Bridle Formation by Conway et al. (1986). The metamorphic rocks to the southwest of Bagdad host seven known Proterozoic volcanogenic massive sulfide (VMS) deposits (Conway et al., 1986). Similar metamorphic rocks are present in less well-defined belts to the south and southwest of the study area. Mesoproterozoic granitic plutons intrude the older rocks and form several well-defined plutons, with the 1410 ± 4 Ma Signal granite to the southwest and 1418 ± 2 Ma granite of Olea Ranch located to the southeast. At least two distinct Proterozoic deformational events are recorded across the region: (1) an early (~1.74–1.715 Ga) deformation event producing map-scale and mesoscopic, W- to SW-vergent recumbent folds with axial planar foliation dipping gently to moderately east-northeast, and (2) a younger event producing a subvertical NE-striking foliation representing the Yavapai/Ivanpah orogeny (Duebendofer, 2015). Duebendorfer et al. (2001) suggest the early event marks the collision of the Yavapai arc with the eastern part of the Mojave province.

Approximately 5 km southwest of Bagdad, the ~77 to 76 Ma Grayback tuff unconformably overlies Proterozoic rocks, demonstrating the removal of Mesozoic or Paleozoic strata prior to the Late Cretaceous (Anderson et al., 1955; Greig, 2021). The Bagdad porphyry Cu-Mo deposit is associated with one of several slightly younger Late Cretaceous granodioritic- to-quartz monzonitic stocks (Anderson et al., 1955). Barra et al. (2003) report Re-Os molybdenite ages of 75.9 ± 0.2 to 71.8 ± 0.2 Ma, suggesting multiple pulses of hydrothermal fluid flow and mineralization. The Bagdad porphyry Cu-Mo deposit has an estimated metal endowment (production + reserves + resources) of 14.5 million metric tonnes (Mt) Cu, 0.7 Mt Mo, 0.6 Moz Au, and 253.4 Moz Ag (Leveille and Stegen, 2012).

To the southwest of the Basin, exhumation of mylonitized midcrustal rocks in the Buckskin-Rawhide metamorphic core complex initiated ~22 to 21 Ma and continued until ~12 to 11 Ma (Singleton et al., 2014). The high- and low-silica rhyolites and basalts of the bimodal Kaiser Spring volcanic field erupted between ~15 and 8 Ma (Moyer and Esperança, 1989), and these eruptions overlapped with deposition of the Tule Wash Formation in Big Sandy Valley to the northwest. Deposition of the Tule Wash Formation may be synextensional with a poorly exposed low-angle, E-dipping normal fault bounding the Hualapai Mountains (Gootee et al., 2024). The Late Miocene to Early Pliocene Big Sandy Formation, which also hosts an Li-mineralized clay lithofacies, is deposited on gentle angular unconformity with the underlying Tule Wash Formation (Sheppard and Gude, 1972; MacFadden et al., 1979; Gootee and Johnson, 2023).

The earliest known exploration of the clay deposits in the Basin area was conducted by GSA Resources first exploring for magnesian (saponite-type) clays in 1983. They identified a small, cosmetic-grade saponite deposit that is presently operated by BYK Additives, Inc. (T.H. Eyde and D. Eyde, unpub. report, 1983; Pittuck, 2023). In 2016, Zenith Minerals Limited initiated its exploration program of the area and subsequently drilled 14 reverse circulation holes in 2018. In late 2019, an initial inferred mineral resource for Basin (then called Burro Creek East) was announced, with 42.6 Mt of mineralized rock with grades of 818 ppm Li and 3.3% K. Bradda Head Lithium, Ltd. acquired the property from Zenith in 2021 and drilled 10 diamond drill holes, with two updated indicated and inferred mineral resources released in 2022. An additional 28 sonic holes were drilled between 2022 and 2023 (Pittuck, 2023). In late 2023, using a cutoff grade of 550 ppm Li, an indicated mineral resource of 17 Mt of mineralized rock with grades of 940 ppm Li and 3.4% K and an inferred mineral resource of 210 Mt of mineralized rock with grades of 900 ppm Li and 2.8% K was announced (Pittuck, 2023). In July 2024, a new mineral resource estimate was released using a cutoff grade of 550 ppm Li, with an inferred mineral resource of 506 Mt of mineralized rock with grades of 808 ppm Li, an indicated mineral resource of 122 Mt of mineralized rock with grades of 860 ppm Li, and a measured mineral resource of 20 Mt of mineralized rock with 929 ppm Li (Bradda Head Lithium, 2024b). The combined indicated, inferred, and measured mineral resource at the Basin is currently 641 Mt of mineralized rock with grades of 823 ppm lithium (Li), containing 2,809 kt of Li carbonate equivalent (LCE) or 527.5 kt of lithium metal. Potassium values are not referenced in the most recent mineral resource.

New 1:24,000-scale geologic mapping of the Kaiser Spring volcanic field was completed from the fall of 2021 through the spring of 2023 (Thompson et al., 2023). Additional 1:5,000-scale mapping of hydrothermal deposits in Basin East, West, and North, augmented by high-resolution infrared data collected by Jet Propulsion Laboratory Hyperspectral Thermal Emission Spectrometer (HyTES), was completed in the spring of 2024 (A. Huff, unpub. data, 2024). The two scales of maps were reconciled into a compilation map.

Published geochemical data (Schreiner, 1985; Moyer, 1986, 1990) was augmented by new sample analysis. Whole-rock geochemical samples were prepared and analyzed at AGAT Labs and SGS Labs under contract with the U.S. Geological Survey’s Mineral Resource Program for a suite of 61 elements via inductively coupled plasma-optical emission spectroscopy (ICP-OES), inductively coupled plasma-mass spectrometry (ICP-MS), and X-ray fluorescence (XRF). Bradda Head Lithium, Inc.’s surface and core samples and cuttings were prepared and analyzed for 48 element assays at ALS labs via an inductively coupled plasma-mass spectrometer with four-acid digestion (ICP-AES), where lower and upper Li detection values are 0.2 and 10,000 ppm. Drill core, both sonic and diamond drill, from 2023 forward, was analyzed by SGS Labs, also using ICP-AES.

Published XRD data were augmented with new analyses. Samples were analyzed at the Illinois State Geological Survey’s XRD/XRF Materials Characterization Laboratory, where 10 to 15 g of the sample (<250 μm) was immersed in deionized water and disaggregated. The sample was stirred in an electronic mixer to isolate clays in suspension. The particles settled, and the salts that remained in solution were poured off. The sample was stirred again and allowed to settle until only the <2-μm particles were in suspension. Using an eye dropper, the <2-μm clays were placed onto a glass slide which was then air-dried, glycolated, and scanned by XRD. Step-scanned data was collected from 2° to 34° 2θ with a fixed rate of 1°/min with a step size of 0.02° 2θ.

Sieve and microscope analyses were conducted on 14 samples from one borehole, BES-23-07, that intersected clay-bearing strata in Basin North. Core samples were described and weighed prior to disaggregating in water. Odor, texture, and degree of clay swelling were noted, then sieved using 10, 35, 60, and 100 mesh screens, separate from pan silt and clay. Dried sieve samples were weighed to calculate weight percent clay, followed by a description of primary minerals, grains and secondary minerals, and alteration of nonclay portions using a binocular microscope.

The Miocene Kaiser Spring volcanic field is the youngest bimodal volcanic field in Arizona (~14–8 Ma; Moyer, 1990). It comprises two geochemically distinct suites of felsic volcanic domes, lava flows, and pyroclastic deposits: (1) older high-silica rhyolites of the Eastern volcanic belt and (2) younger low-silica rhyolites of the Western volcanic belt (Moyer, 1986, 1990) (Fig. 3). Both belts of volcanic rocks are interbedded with three suites of basaltic lava flows (Moyer, 1986; Thompson et al., 2023) and the entire volcanic field has a silica gap from 57 to 69 wt % SiO₂ (Moyer and Esperança, 1989). New major and trace element geochemical data for relevant rhyolite lava and rhyolitic tuff samples in the study area are reported in Appendix Table A2.

The timing of Kaiser Spring volcanic eruptions is not well constrained. Dates reported herein are from K-Ar radiometric ages in previous studies (summarized in table 2 of Thompson et al., 2023). Tholeiitic volcanism initiated ca. 15 Ma (Moyer, 1990) in the southern portion of the Kaiser Spring volcanic field, Tholeiitic basalts unconformably rest on Proterozoic igneous and metamorphic rocks and are interbedded with basinal pyroclastic and sedimentary rocks. From 15 to 8 Ma, tholeiitic volcanism transitioned to mildly alkaline (Moyer, 1990). The youngest basalt lava flow in the suite has not been dated, but stratigraphic relationships suggest it may be younger than or contemporaneous with Late Miocene to Pliocene basin fill and correlative with the Big Sandy Formation (Gootee and Johnson, 2023; Thompson et al., 2023).

Felsic Eastern volcanic belt dome eruptions form a N-S–oriented belt bordering Basin and were active by at least 12 Ma (Moyer, 1990). All Eastern volcanic belt lava domes are the culminating phase of singular felsic eruptions in which lavas filled a tuff ring crater. The 12 Eastern volcanic belt domes are partially hydrated and devitrified and characterized by perlitic merakanite-bearing (Apache Tear) bases with abundant megaspherulites grading into flow-banded high-silica rhyolite (mean 76.9 wt % SiO₂; Fig. 4A, B) with abundant lithophysae (Moyer, 1990). Notably, two of these domes, Red Knob and N. Ed, are garnet- and topaz-bearing rhyolites (Burt et al., 1981). Associated pyroclastic rocks include lithic lapilli tuffs, ash tuffs, and lapillistones (Fig. 4C, D). Western volcanic belt domes were active by ~10 Ma (Moyer, 1990). Like Eastern volcanic belt domes, the 12 Western volcanic belt domes overlie related pyroclastic rocks. Western volcanic belt domes are distinguished by their lower silica content (mean 71.6 wt % SiO₂; Moyer, 1990), phenocryst assemblages, and larger width to height ratios, with Western volcanic belt flows being thinner and more aerially extensive. Mafic blob inclusions are noted in at least one Western volcanic belt dome, which suggests mixing of felsic and mafic magmas (Moyer, 1990).

Basin stratigraphic architecture was first fully described by Thompson et al. (2023), and this study follows those names and map units herein (Fig. 5A-C; Table 2). Neogene basin fill stratigraphy includes multiple nonvolcanic and volcaniclastic conglomerates and sandstones (Nc1, Nc2, Nvc); felsic lithic lapilli tuffs, ash tuffs, breccias, and lapillistones (Ntu); basalt lava flows, scoria, and basaltic tephra (Nb2); claystones; dolomite, magnesite, and agate (Nsin); and minor sandstones (Ns, Nvs), all of which unconformably overlie Proterozoic bedrock (YXu). All units are included within one or more complex landslides (Ql) or rock slab avalanches (Nx).

Basin mafic volcanism is correlative to the volcanic Wilder formation based on regional stratigraphic relationships. The informally named Wilder formation is an undated group of basalt lava flows, dikes, and scoria cones deposited on Proterozoic rock or interbedded with local conglomerates and ash tuffs that crop out in Wilder Canyon north of the Bagdad porphyry copper mine ~13 km to the east of the Basin (Fig. 2; Anderson et al., 1955). The Wilder formation is interbedded with coarse alluvial conglomerates shed from Proterozoic terrain and capped by a regionally extensive basalt lava flow that also caps Basin sediments. Stratigraphic relationships with capping basalt lava flows and young conglomerates indicate the Basin Miocene stratigraphic package is also correlative with the Miocene Tule Wash Formation in neighboring Big Sandy Valley (Ferguson et al., 2023; Gootee and Johnson, 2023).

The earliest record of basin sedimentation is a W-SW–imbricated nonvolcanic coarse cobble conglomerate exposed within small canyons cut into Proterozoic crystalline bedrock on the northern basin margin (Nc1). This conglomerate (≥5 m thick) was deposited prior to the initiation of the Kaiser Spring volcanism and records pre-Middle Miocene westerly flow directions of an older, local drainage system. A second, coarse cobble conglomerate (Nc2), which unconformably overlies Nc1, records deposition during the onset of Kaiser Spring volcanism. The second conglomerate (≥6 m thick) is composed of angular to subrounded Proterozoic igneous and metamorphic cobbles, boulders, and grus with angular to subrounded scoria, basalt, and sparse high-silica rhyolite clasts. South-southeast cobble imbrication from the northern basin margin and northerly cobble imbrication from the southern basin margin indicate the basin was closed at the time of Nc2 deposition. Small pockets of moderately sorted to well-sorted sandstone (Ns) occupy space between fan remnants. In general, the Nc2 conglomerate underlies all other basin sedimentary rocks. However, on the border of Basin East and Basin North, just north of Burro Creek, 8 to 11 m of N-NE–imbricated, clast- and matrix-supported conglomerate composed of >99% Proterozoic clasts is interbedded with clay and lithic lapilli tuff (Fig. 5). The same conglomerate also caps Proterozoic bedrock that crops out ~95 m to the south-southwest, indicating aggradation of a coarse alluvial fan into the basin depocenter.

As volcanism intensified through the Middle Miocene, numerous tuff rings developed around the basin margins through the felsic phreatomagmatic and volcanic excavation of shallow craters. These craters were subsequently filled with rhyolite lava domes of the Eastern volcanic belt (Fig. 6A-D). The early coarse- to medium-grained basin sedimentation was punctuated by an unknown number of multisourced, ash and lapilli tuffs, lithic lapilli tuffs, lapillistones, and tuff breccias from explosive rhyolitic eruptions and possible dome collapse (Ntu), and at least three 2- to 8 m-thick basalt lava flows (Nb2). Individual tuff beds range from 1 to 20 m thick, and the thickest beds form tuff rings that underly rhyolitic lava domes. Older tuffs are lithic rich and are typically massive to low-angle crossbedded. Younger tuffs lack Proterozoic lithic fragments, are massive, and are pumice and ash rich. Many pumices have been partially or wholly chloritized. Some basalt lava flows were intermixed with wet sediment (Nvc) as they were being deposited, indicated by peperitic margins and soft-sediment deformation located beneath basaltic scoria breccia (Fig. 7A). Basalt flows are fresh, saprolitic, or partially altered to clay (Fig. 7B, C), and at least one flow caps scoria and basaltic tephra indicating eruptions were explosive within the basin (Fig. 7D, E). Some hydrovolcanic explosions are evidenced by tuffs bearing accretionary lapilli (Moyer, 1986) and numerous basalt block sags in alluvial conglomerates (Fig. 7F-H).

The phreatomagmatic and pyroclastic eruptions that generated the numerous lithic lapilli tuffs were both penecontem-poraneous and concurrent with deposition of the multicolored magnesian smectite-group clays (Nca) hosting the Basin volcano-sedimentary Li deposit. The clay is the slip surface for large, complex landslides in the basin (Ql) and most of the area over which clay is exposed is within these landslides, including the Calico landslide (Basin West) and the Little Calico landslide (Basin East). On the surface, exposed clay beds are a maximum of ~18 m thick and buttress up against Proterozoic bedrock. Notably, clay does not laterally or vertically grade into marginal sedimentary facies, such as well-sorted sand deposits that are expected in a lacustrine depositional setting.

All clay exposures are associated with volcanic sinter mounds and aprons (Nsin) (Fig. 8A, B). The sinter is composed of interbedded laminated to massive or mounded strontium-rich dolomite, strontium-rich cryptocrystalline magnesite, cryptocrystalline silica, and/or chaotic cryptocrystalline silica breccia. Bedding surfaces display elephant-skin weathering (Fig. 8C) and dolomite beds and mounds overlie spring conduits (Fig. 8D). The sinter is also host to the abundant, irregularly shaped, banded white-clear, orange-brown, blue, purple, gray, and red chalcedonic agate, which is often collected by prospectors for lapidary use (Fig. 8E, F). All sinter deposits are spatially related to the Li-rich claystone, saprolitic or clayaltered basalt lava flows, and clay-altered tuffs, and sinter deposits generally align with high-angle normal faults.

Kaiser Spring volcanism waned in the Late Miocene (Moyer and Esperança, 1989), and the interbedded sedimentary and volcanic sequence was covered by a thick (≥60 m), SW-imbricated, coarse cobble boulder volcaniclastic conglomerate (Nvc). The conglomerate, dominantly composed of Eastern volcanic belt-sourced high-silica rhyolites, basalt, and Proterozoic igneous and metamorphic rocks, represents locally derived alluvial fan aggradation from the Aquarius Mountains and Eastern volcanic belt rhyolite domes.

There are no basin-bounding faults controlling the formation of accommodation space in the Basin (Fig. 5; see also Thompson et al., 2023). Alluvial fanglomerates (Nc1, Nc2) onlap Proterozoic terrain on the north and south side of the basin. The younger conglomerate (Nc2) forms the base of the basin sedimentary and volcanic package, which is generally conformable except for volcanic sinter mounds and aprons fed by spring conduits that cut claystone (Nca) and tuff (Ntu). Bedding attitudes indicate the entire Miocene sedimentary and volcanic package has centroclinal dips, focused on the locus of a basin-wide gravitational low (Fig. 5), creating a radial synclinal basin with average (nonfaulted) dips of 7° to 9°. The basin fill package (units Nc2 through youngest Nvc and Nb2; Fig. 5C; Table 2) is cut by numerous small-displacement (≥40 m) high-angle normal faults. Mapped relationships in the greater Kaiser Spring volcanic field show that NE-SW–striking normal faults are generally cut by younger NW-SE–striking high-angle normal faults (Thompson et al., 2023).

The Basin Li-rich volcano-sedimentary deposit is a heterogeneous, massive- to planar-bedded, multipastel-colored magnesian smectite-group claystone, tuffaceous claystone, and sandy claystone with a distinct popcorn weathering texture (Nca; Fig. 9A-F). Clay colors change rapidly over decimeters and include white beige brown, medium to dark brown, yellow gray, olive gray green, pale olive, light gray, and grayish black. At the surface, clay is exposed over ~19 km². Claystone outcrops are not vegetated, but they are often partially obscured by broken outcrops in irregular patches underneath rotated, mass-wasted blocks of lithic lapilli tuff and basalt, or under a chalcedonic agate and dolomite clast lag. The heterogeneous outcrop nature of the claystone is a function of multiple complex mass wasting events. Nearly everywhere clay is exposed in the Basin, it is the slip surface for landslides or rock slab avalanches, which have been reactivated many times. Stratigraphy within the slide is slumped, rotated, and overturned, and many young slumps and debris flows emanate from the margins of the greater disturbed area. Thus, the landslides are named for their tan-brown-black (or tuff-clay-basalt) patchy calico appearance in satellite imagery: Calico landslide in Basin West and Little Calico landslide in Basin East.

Basin claystone is divided into two distinct stratigraphic horizons, herein referred to as the Upper and Lower Clay, separated by a laterally continuous non-Li-bearing zeolitic lithic lapilli tuff 4 to 25 m thick (Fig. 10). The thickness of the Upper Clay in Basin North ranges from 82 to 89 m. The Lower Clay is a minimum of 36 m thick; thickness range is unknown because most drill holes do not penetrate the entire unit. Both clay units are composed of greater than 84% clay and silt size material (<100 mesh or 0.149 mm in size), and most sieved horizons contain up to 94% clay and silt size fraction (App. Table A3). Nine out of 10 sieved horizons in the Upper Clay and one sieved horizon in the Lower Clay contained residual euhedral to subhedral biotite, hornblende, and quartz phenocrysts and glass with trace light orange-brown pumices, trace magnetic grains, and anhedral dolomite/magnesite grains or nodules (Fig. 10; horizons a-h, j-k). The Upper Clay contains rare subrounded detrital quartz sand (horizon c) and rare euhedral cubic pyrite in veinlets and disseminated in the crystal ash matrix (horizon i). Secondary gypsum, carbonate nodules, and carbonate cementation is common throughout the Upper and Lower Clay deposits.

The Upper and Lower Clay units are not stratigraphically distinct and are only distinguishable when considered in stratigraphic context with the lithic lapilli tuff marker bed that separates them. In general, the Lower Clay is more massive, while the Upper Clay is massive to wavy and parallel laminated. Both the Upper and Lower Clay display a range of pastel coloring and abundant carbonate, dolomite, or magnesite nodules or masses. Most drill holes terminate at the base of the Upper Clay because it hosts significantly higher-grade Li than the Lower Clay. To date, borehole logs show the base of the Upper Clay extends to a maximum depth of 208 m (hole BES-23-14; Fig. 11) in Basin North and the unit thickens 11.8 m from south to north over 1 km.

Though both the Upper Clay and Lower Clay units host Li, only the Upper Clay data will be discussed herein because of the paucity of drill holes in the Lower Clay. Clay mineralogy is limited to a few surface samples of the Upper Clay. No clay mineralogy or XRD data are available for any borehole samples in the Upper or Lower Clay.

A study by Schreiner (1985) analyzed nine clay samples using XRF for whole-rock geochemistry and XRD for mineralogy. Samples were collected on the surface in what is now Basin West and within the Upper Clay (Fig. 5). That study concluded the clay species is dominantly trioctahedral magnesian smectite, or saponite-type clay, with varying amounts of additional minerals including dolomite, phillipsite, quartz, kaolinite, gypsum, muscovite, and potassium feldspar. The only sample analyzed for clay mineralogy by Thompson et al. (2023) (sample 2) showed the Upper Clay, in that location, is composed of 85% smectite, 13% illite, with subordinate kaolinite and chlorite. Sample 2 is from the same general location as Schreiner (1985) samples (Fig. 5), presumably from the same horizon of Upper Clay. More recent, XRD mineralogical analysis of one sample, collected from the Upper Clay at an unknown location in Basin East, showed the dominant clay species was 2 to 10 wt % swinefordite (LiCa_0.5 Na_0.1 Al_1.5 Mg_0.5 Si₃ O₁₀ (OH)_1.5 F_0.5 4(H₂O) and 2 to 10 wt % illite, with possible associated petalite (Li(AlSi₄O₁₀)) (Pittuck, 2023). Notably, all these samples are coincident with abundant chalcedonic agate and located proximal to volcanic sinter deposits of dolomite, magnesite, and cryptocrystalline silica (unit Nsin).

Borehole assay data show the Upper Clay has a consistently higher Li grade when compared to the Lower Clay. Maximum Li values in the high-grade zone reach 2,791 ppm, and average Li in the entire Upper Clay ranges from 778 ppm to 983 ppm. Average Li in the Lower Clay is 690 ppm. In the Upper Clay, Li is concentrated within a 5- to 39-m-thick high-grade zone (Li ≥ 1,200 ppm) in the approximate middle of the unit (Fig. 10; Table 3). Overall, Li mineralization corresponds to increased Mg, and Mg is inversely correlated with Al. The distribution of Mg in clay mineral species is unknown, because samples were analyzed for their geochemistry using a four-acid digestion with an ICP-AES finish and claystone often contains dolomite and magnesite masses or nodules. In July 2024, Bradda Head Lithium Ltd. announced a combined inferred, indicated, and measured mineral resource of 641 Mt of mineralized rock with mean grades of 823 ppm Li. Basin contains 2,809 kt of Li carbonate equivalent (Bradda Head Lithium, 2024b). The former late 2023 mineral resource (combined indicated and inferred) of 227 Mt of mineralized rock had grades of 2.8% K containing 6,370 kt K. Maximum Li values for all Upper Clay surface samples and boreholes are shown in Figure 11. Notably, surface samples are depleted in Li when compared to clay sampled from drill core.

The high-grade Li interval in the Upper Clay also contains highly anomalous molybdenum. Molybdenum is nearly absent throughout the entire Upper Clay package, typically below detection (<1 ppm), and then abruptly increases proximal to the top/bottom of the high-grade Li sequence in the Upper Clay (Fig. 10; Table 3). Molybdenum concentration is highly variable in the high-grade interval, ranging from 641 ppm to 4 ppm over less than 1 m. The high-grade Li interval varies in thickness from 5 to 39 m thick and has weighted averages of Mo between 69 ppm and 206 ppm. Although sulfur is present in the system with values close to 1%, sulfur does not correspond to any of the high molybdenum intervals; thus, molybdenite (MoS₂) is not the molybdenum host species. Neither optical microscopy nor electron probe microanalysis has successfully identified with which mineral the molybdenum is located.

We propose that phreatomagmatic explosions of mafic magma excavated an ~3.5-km-wide, deep crater (>300 m) cut into Proterozoic igneous and metamorphic rocks, forming accommodation space for subsequent deposition of the basin stratigraphic sequence (Fig. 12). Phreatomagmatic mafic explosions occur when rising mafic magma encounters groundwater or shallow surface water and steam-driven explosions ensue, which fragments and ejects rock and/or sediment overlying the magma. The resulting crater, a maar volcano, is the surface expression of a much deeper cone-shaped diatreme filled with primary volcanic material and brecciated country rock (Fig. 13). The crater is surrounded by a thin (commonly <50 m thick), wide apron primarily composed of ejected country rock with lesser primary magmatic material. After eruption, the crater is typically filled with shallow water and lacustrine sediments, punctuated by crater wall mass wasting or alluvial fans. Over time, lacustrine deposition transitions to subaerial as the lake is filled with sediment or other volcanic material (Ollier, 1967; Lorenz, 1986; White, 1990; Vespermann and Schmincke, 2000; Pirrung et al., 2003; Zelawski, 2011).

We favor this interpretation for many reasons. First, volcanism in the Kaiser Spring volcanic field initiated ca. 15 Ma with the eruption of tholeiitic basalt lava flows that unconformably overlie Proterozoic igneous and metamorphic rocks (Nb; Fig. 3). Vent locations for these lava flows, such as scoria cones, have not been located (Moyer, 1990), indicating the flows likely erupted from now-covered fissures/dikes. Basin sedimentation initiated before rhyolite domes were emplaced (Nc2 conglomerate only contains rhyolite clasts only at the top of the section), and the Basin sedimentary package is overall younger than the earliest basalt lava flows (Nc2 contains basalt clasts at the base of its section). In addition, Wilder formation basalt flows, dikes, and scoria cones, which are stratigraphically correlative with the Basin basalt flows, crop out ~13 km to the east. Anderson (1955) interprets horizontally bedded basaltic tephra of the Wilder formation as “water-deposited.” A basaltic dike related to these flows or cones could have intercepted shallow groundwater or surface water at or near a paleoriver channel, initiating phreatic explosions, ultimately excavating a deep crater into Proterozoic bedrock.

Second, two units totaling at least 117 m (drill hole BES-23-07; Basin North) of laterally continuous clay (Nca1, Nca2), punctuated by a lithic lapilli tuff (Ntu1), accumulated in a small yet deep depocenter. There are no transitional sedimentary facies (i.e., equally thick sand deposits) present that would be expected in a typical lacustrine depositional environment (Fig. 12). In addition, claystone is interbedded at basin margins with coarse alluvial conglomerates (Nc2) and rock-avalanche slabs (Nx) derived from Proterozoic rock terrain. Due to their phreatic origin, maar craters often fill with shallow lakes posteruption and many are depocenters for thick accumulations of claystone, pelite, or peat (Pirrung et al., 2003). Maar crater walls dip 75° to 85°, allowing sediments to accumulate in layer cake fashion across the crater lake floor (Suhr et al., 2006). The Miocene-Pliocene Hopi Buttes volcanic field in northern Arizona contains the greatest concentration of maar craters in any volcanic field in the world (n = 300; White, 1990) and many of those craters are filled with clay interbedded with mass wasted deposits and coarse alluvial gravels shed from steep crater walls (White, 1990; Zelawski, 2011). Gutmann (2002) documented in the Pliocene-Holocene Pinacate volcanic field on the Arizona-Sonora border that maar craters were excavated after scoria cones were built, and those craters are located within possible ancient river channels with permeable river gravels. An alluvial system was through-flowing prior to Kaiser Spring volcanic field eruptions (units Nc1), and cobble imbrication shows those streams were diverted or completely dammed by eruptions (unit Nc2); therefore, the water table could have been at or near the surface.

Third, detailed geologic mapping of the Kaiser Spring volcanic field (Thompson et al., 2023) did not identify any basin-bounding faults that could be responsible for the formation of hundreds of meters of accommodation space.

Mapped high-angle normal faults are, overall, younger than basin sediments and show no more than ~40 m of displacement (most faults have only a few meters of displacement). Stratigraphic units do not display fanning dips or significant thickening toward the basin center, and thus, dips do not fit a half-graben depositional model. Yet, bedding attitudes and core logs indicate a main depositional center that is at least 3.5 km wide (along the NW-SE axis) and >300 m deep (depth to diameter ratio of 0.09) (Fig. 12). The largest maar crater on Earth is ~5 km wide (mean crater diameter, which is the average of the large and small axes), but most maar craters have a mean crater diameter of 600 to 800 m. Maar depths range from 5 to 400 m with an average depth to diameter ratio of 0.10 (Graettinger, 2018). The depth to diameter ratio of Basin is an order magnitude smaller than the global average, but this could be the result of poorly defined maar crater boundaries that could include a preexisting alluvial plain or multiple, coalesced craters. To date, only Basin East and a portion of Basin North have been drilled; therefore, the crater extent is not well defined.

Fourth, the basin sedimentary package displays centroclinal postdepositional subsidence, tilting all units 7° to 9° into the locus of a semicircular gravitational low (Figs. 5, 12). Post-depositional subsidence over a maar diatreme is well documented and centroclinal dips range from 5° to 20° (Suhr et al., 2006). Maar crater diatremes—or the root zones of maar craters—are filled with unconsolidated heterogeneous breccia composed of primary tephra and country rock. Posteruptive mechanical and chemical compaction, and diagenetic or hydrothermal alteration of diatreme fill, initially occurs rapidly and decreases over time (Suhr et al., 2006). As a result, the maar crater sediment overburden—in this case, claystones (Nca), tuffs (Ntu), basalt lava flows (Nb), and alluvial sediments (Nvc, Ns)—which itself would also mechanically and chemically compact over time, subsides into the diatreme, forming a radial synclinal fold that outlines the diatreme diameter. Maar diatremes create strong Bouguer anomalies that increase with diatreme depth ranging from 1.2 to 6 mGal (Lorenz, 1986). The Basin gravity survey (Bradda Head Lithium, Ltd., 2024a) indicates 2 mGal of relief over <1.7 km in a semicircular shape (Fig. 5). In addition, the Basin sedimentary package does not dramatically thicken toward the basin center (i.e., there are no fanning dips) and thus, dips can only be explained by overburden subsidence within a semicircular, low-void space. Diatreme compaction-induced subsidence over the Early Miocene Kleinsaubernitz maar, Germany, has been shown to continue for millions of years, as the subsidence has affected even the youngest Holocene alluvium (Suhr et al., 2006). The amount of subsidence generally increases with diatreme width, and documented cases have ranged from 50 to 1,250 m (Hearn Jr., 1968; Suhr et al., 2006). Assuming claystone was deposited horizontally, we estimate 100 to 225 m of postdepositional subsidence from the projection of the present 7° to 9° dips of the clay deposits to the center of the gravitational low (Fig. 5).

Finally, sedimentological evidence shows water/magma interaction during and after mafic eruptions elsewhere in the Basin. One basalt lava flow interbedded with tuff and volcaniclastic conglomerate displays a peperitic upper contact and a baked lower contact (Fig. 7A). Peperite is a rock formed when lava flows into saturated sediment and flashes pore waters to steam, causing small phreatic explosions to mix basalt with sediment at the flow margins (Batiza and White, 2000). In addition, Basin West volcaniclastic conglomerate contains numerous basalt block sags (Fig. 7F-H). Block sags form when phreatic explosions excavate a crater, and country rock is ballistically ejected onto the surrounding landscape. When blocks land on water-saturated sediment, the beds undergo soft sediment deformation. The blocks in Basin West range from a few tens of centimeters to ~3 m. The source(s) of the explosion is unknown, but at least one basalt lava flow does crop out locally.

For these reasons, we hypothesize that a maar crater, ~3.5 km wide and >300 m deep, may be the origin of accommodation space for deposition of the Basin sedimentary and volcanic package. Maar craters are relatively deep for their size, creating large amounts of accommodation space and hydrologically closed basins where water, sediments, and other volcanic material accumulate. Their phreatic explosions eject country rock mixed with primary magmatic material onto the surrounding landscape in discrete pulses, resulting in a rim of thin discrete tephra beds. These tephra beds have not been identified in the stratigraphic record, but they could have easily been resedimented and/or eroded by a through-flowing alluvial drainage system. A maar crater is favored over a tuff ring, many of which were built in other parts of the Kaiser Spring volcanic field, because maars are defined by a deep excavational crater cut into the pre-eruption surface and tuff rings are defined by small craters at or above ground level (Vespermenn and Schmincke, 2000). Within our current state of knowledge, the presence of one or more coalesced maar craters cannot be ruled out.

The Kaiser Spring volcanic field domes were active contemporaneously with sedimentation in the Basin. Multiple lithic lapilli tuffs are interbedded with claystone (Figs. 5, 10). These nonmineralized tuffs, which are deposited across the area, are very lithic rich (varying from lithic lapilli tuffs to lapillistone; Fig. 4D), and lithic fragments are either angular Proterozoic igneous and metamorphic rocks or rhyolite lavas. The lapilli tuffs, therefore, must have been created by the excavation of the shallow tuff ring craters in Proterozoic bedrock, forming pyroclastic flows and surges, and/or the collapse of oversteepened domes, forming block and ash flows (Moyer, 1986; Thompson et al., 2023). Notably, lithic-poor, ash- and pumice-rich deposits are absent in drill core, even though ash clouds must have accompanied the pyroclastic density currents and ash- and pumice-rich beds are present throughout the rest of the landscape (Fig. 4C), mantled on Proterozoic basement rocks and interbedded with lithic-rich tuffs on basin margins. This begs the question: What happened to all the volcanic ash that must have accompanied the numerous lithic lapilli tuff deposits in the basin depocenter?

We propose that lake water was initially enriched with Li and Mo (and other trace elements in minor quantities) due to (1) hydration and dissolution of suspended or sedimented Li-rich volcanic ash, (2) hydration of rhyolitic vitrophyre at the base of nearby Eastern volcanic belt rhyolitic domes, and/or (3) hydrothermal fluid mixing of groundwater and surface water. It is likely Li/Mo enrichment occurred because of all three of these processes. Eastern volcanic belt rhyolite and lithic lapilli tuff is elevated in Li (54–222 ppm; App. Table A2) several times that of the mean value for rhyolitic obsidian (45 ppm; MacDonald et al., 1992) and crustal levels (30–40 ppm; Levinson, 1974; Rose et al., 1979; Price et al., 2018). Felsic ash tuff beds are conspicuously absent in the basin depocenter stratigraphic package, even though they are deposited on slopes interbedded with lithic lapilli tuff throughout the drainage basin. Moreover, most clay-sieved fractions are crystal-rich (App. Table A3), lack detrital material, and have felsic ash tuff chips within them. Therefore, volcanic ash from many felsic dome eruptions, inevitably distributed over the entire basin, must have settled out of suspension from lake water. In alkaline conditions, volcanic glass will easily dissolve, liberating Li (and other trace elements including Mo), and simultaneously increasing water alkalinity (Fisher and Schmincke, 1984; Price et al., 2018; Ellis et al., 2022).

Hydration of volcanic glass suspended in or settled out of lake water would have been contemporaneous with hydration of rhyolite dome vitrophyre throughout the drainage basin. Ellis et al. (2022) showed that an average of 22% of Li can be mobilized during posteruption glass hydration and that the degree and rate of Li loss is constrained to a finite period dependent on the deposit’s cooling history. Eastern volcanic belt unaltered rhyolite lava samples from Red Knob and N. Ed domes have 143 to 247 ppm Li (App. Table A2; Fig 5). Hydrated (perlitic) Eastern volcanic belt lavas have a significantly lower Li content (Northern Burro dome = 16 ppm; unnamed rhyolite flow near Highway 93 = 11 ppm): evidence that Li loss due to glass hydration throughout the volcanic field was likely significant. Additionally, Keith and Shanks (1988) showed that Mo was easily leached from hydrated volcanic glass in ash deposits at the Pine Grove porphyry Mo deposit. Molybdenum concentrations for high-silica rhyolites in the Eastern volcanic belt are <3 ppm (map no. 10 (devitrified), 24 (hydrated), and 28; App. Table A2). Although these concentrations are low, the magmatic concentration of Mo at Pine Grove was suggested to be 2 ppm (Keith and Shanks, 1988).

It is also likely that Li concentration was further enriched via hydrothermal fluids interacting with alkaline lake water or circulating as groundwater, as has been shown at McDermitt caldera (Benson et al., 2023). Laminated cryptocrystalline silica beds, interbedded with strontium-rich dolomite (Sr > 2,500 ppm; App. Table A2) mounds, aprons, and spring conduits (unit Nsin; Figs. 5, 8), and cap and crosscut claystone, are colocated with saprolitic or clay-altered basalt lava flows (Fig. 7B, C). The sinter deposits are interpreted as hydrothermal spring deposits because of their morphologies, textures, and anonymously high strontium.

Hydrothermal fluid circulation is further supported by a positive correlation between Li and the trace elements fluorine (F) and molybdenum (Mo) (Fig. 14; App. Table A4) in the Upper Clay, which are anonymously high when compared to average crustal values (Parker, 1967). There is only a small data set with F—two holes drilled in 2021 in Basin East (holes BCE21-01 and -03)—but F is positively correlated with Li (F = 9,320 ppm is coincident with Li = 1,910 ppm) (hole BCE21-03). Other trace elements commonly associated with hydrothermal fluid alteration, antimony (Sb), arsenic (As), potassium (K), rubidium (Rb), sulfur (S), and thallium (Tl), have weak to moderate correlations to high Li. This could be due to variations in the source of hydrothermal fluids or lateral permeability constraints within the clay sequence, although Castor and Henry (2020) suggested that Sb, As, F, K, and Mo could simply be mobilized through diagenesis/hydration of felsic volcanic glass. In the Upper Clay, antimony (Sb) absolute values are relatively low (max = 11 ppm) with the highest values between 0.5 and 2.9 ppm (hole BES-23-04). The highest detectible level of As in the Upper Clay at Basin is 1,300 ppm (BES-23-04); this horizon is correlated with increased S (1.97%) but relatively low Li (288 ppm) and Tl (2.4 ppm). High K values frequently bracket the high-grade Li horizon in the Upper Clay (K ranges from 5 to 9%). Both K and Rb levels are highest at the top and bottom of the Upper Clay (Rb ~250 ppm) and are negatively correlated with Li (ranging only from 50–100 ppm) (hole BES-23-05). Lastly, all surface Li concentrations are significantly lower than cored clay at depth (Fig. 11). Therefore, it is possible that Li concentrations at depth continue to be enriched through diagenesis as surficial rocks weather and Li is transported via meteoric waters above the groundwater table.

The origin of Li-rich magnesian smectite-dominated claystones at Basin is unresolved, largely because there are only a few mineralogical studies to ascertain the mineral species in which Li is contained or identify zeolite mineral associations, and no systematic mineralogical analyses has been performed on drill core. The claystones are thick and stratigraphically continuous and contain >84% clay and silt size fractions, with most sieved horizons up to 94%. There are no well-sorted volcanic ash beds in the Basin stratigraphic package, even though volcanic ash beds are prevalent throughout the drainage basin and claystone is interbedded with lithic lapilli tuff. Sieved clay fractions are dominantly fragmented, euhedral to subhedral biotite, hornblende, and quartz crystals associated with Eastern volcanic belt eruptions, with little to no detrital input, suggesting volcanic ash was the parent material for the smectite (App. Table A3).

Maar crater basin morphology and basin-wide sedimentary imbrication is suggestive of a hydrologically closed basin, where highly alkaline, Li-rich, and silica-saturated water ponded, creating adequate conditions for precipitation, or neoformation, of Li-rich smectite clays, as has been documented at McDermitt caldera/Thacker Pass (Benson et al., 2023) and in other alkaline lacustrine environments (Furquim et al., 2008). Because of the dearth of mineralogical studies, only the zeolite phillipsite, which forms principally in saline, lacustrine environments with pH values of 9 or greater, has been described at Basin (Schreiner, 1985; Miller et al., 1987). Primary wavy and parallel laminations throughout both clay beds suggest at least partial settling via suspension, whereas very thin gradational contacts with interbedded nonmineralized lithic-lapilli tuffs suggest that postdepositional diagenesis alone could not have caused near complete alteration to Li-rich magnesian smectite in two beds (Upper and Lower Clay), while leaving the interbedded lithic lapilli tuff nonmineralized.

Postdeposition hydrothermal fluid circulation likely partially illitized smectite (Morrissete, 2012; Benson et al., 2023) in the Upper Clay. Limited XRD data show one sample retrieved proximal to hydrothermal spring mounds, aprons, and spring conduits that crosscut claystone (sample no. 2, Fig. 5) was a mixed smectite/illite claystone (illite = 13%; Thompson et al., 2023) and a separate sample from an unknown location in Basin East was 2 to 10 wt % illite (Pittuck, 2023). Previous workers have also suggested the clay was a product of hydrothermal alteration of tuffs based on field relationships and proximity to masses of magnesite and dolomite (T.H. Eyde and D. Eyde, unpub. report, 1983; Schreiner, 1985, Miller et al., 1987). These deposits, lumped as volcanic sinter (dolomite and laminated cryptocrystalline silica beds, mounds, and spring conduits) are always spatially associated with claystone and abundant chalcedonic agate (Fig. 8E, F) indicating silicasaturated solutions. Additionally, high K₂O/Na₂O ratios for one sample of the upper lithic lapilli tuff (>11; map no. 6, App. Table A2, Fig. 5) indicate K contents could have been enriched through hydrothermal K metasomatism (Roddy et al., 1988; Ellis et al., 2022). However, potassium enrichment via Na loss in tuffs and rhyolite lavas is not pervasive throughout the Basin (Thompson et al., 2023), indicating K metasomatism, if it occurred, may be a local phenomenon proximal to hydrothermal springs.

To date, there are no documented volcano-sedimentary Li deposits within a bimodal volcanic field or within maar craters. The Basin is singular; therefore, it represents a new type of closed, volcanogenic hydrologic basin in which Li-rich claystone accumulated. While basin morphology may be unique, the Basin Li deposit is analogous to other, regional volcano-sedimentary Li-rich clay deposits and it sits squarely in the middle of the field as a viable economic resource for Li (Fig. 1; Table 1). All known occurrences of volcano-sedimentary Li occur in formerly saline, highly alkaline closed basins, and all occurrences have proximal rhyolitic tuffs and/or lavas which are the Li source rocks (Ames et al., 1958; Price et al., 2000; Benson et al., 2017; Pittuck et al., 2018; Ingraffia et al., 2020; Ellis et al., 2022; Benson et al., 2023; Darin et al., in press; Gangon et al., in press; Putzolu et al., in press). Additionally, the Li resource is dominantly found in magnesian smectite claystones, where Li enrichment appears to be emergently, positively correlative to Mo and F (Castor and Henry, 2020; Ingraffia et al., 2020) based on available assay data (App. Table A4). Conversely, K, which is associated with hydrothermal fluid alteration (Ingraffia et al., 2020) or K metasomatism (Roddy et al., 1988), is negatively correlated with zones of Li enrichment, a phenomenon which needs further investigation.

In addition to Basin, there are two other volcano-sedimentary Li deposits in northwestern Arizona: the Big Sandy deposit and the Lyles clay deposit (now referred to as Thompson Valley) (Figs. 1, 2). Additionally, the Anderson uranium deposit in the Date Creek basin has elevated Li values within clays (Mueller and Halbach, 1983) and recent exploration drilling in the Willcox playa, southeastern Arizona, encountered hectorite and saponite-dominated claystones at depths of ~21 to 366 m, with grades of 507 to 570 ppm Li over 45- to 56-m intervals and grades ranging from 620 to 850 ppm Li over 1.5- to 12.7-m intervals (Max Power, 2024a, b). No systematic effort has been undertaken at any of the deposits to ascertain the lithium-bearing clay ore mineralogy, and therefore the clay origins at all these locations remain undetermined. Broadly, smectite-group clays (saponites, swindefordite, and hectorite) appear to be the main hosts for lithium mineralization at Basin, Thompson Valley, and Anderson. At Big Sandy, one high-grade (2,450 ppm Li) sample was dominated by 88% smectite with 10% illite; however, it is unknown from which clay mineral the lithium is derived. Arizona Li deposit commonalities are summarized in Appendix Table A1.

Clay and zeolite mineral species and their compositions are virtually unknown for both the Upper and Lower Clays. The only samples analyzed by XRD to date are a few surface samples of Upper Clay from Basin West and Basin East (Schreiner, 1985; Pittuck, 2023; Thompson et al., 2023; this study). Systematic sampling of all units, particularly in drill core where stratigraphic relationships are preserved for XRD analysis for clay smectite/illite mineralogy and for zeolite alteration products, will elucidate the level to which postdepositional hydrothermal alteration affected clay mineralogy (Benson et al., 2023; Emproto et al., in press). Detailed mineral chemistry and compositions of the clays could potentially be determined with combined electron microprobe analyzer (EMPA) and laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS), as demonstrated by recent work on Li-bearing micas in the Erzegbirge region of the Czech Republic and Germany (Brieter et al., 2019). Understanding the distribution of Li within the ores will be critical to processing and metallurgy as the deposit approaches production.

Additional whole-rock geochemistry should be obtained for all Kaiser Spring volcanic domes to improve understanding of rhyolite Li sources. Eastern volcanic belt dome vitrophyre is partially to wholly hydrated and many lavas are devitrified. Geochemical results will improve understanding of Li and Mo content in fresh and altered dome facies, identifying whether there was significant Li or Mo loss during cooling. Whole-rock geochemistry will improve understanding of petrological variations in Li-rich rhyolites. Systematic trace element geochemistry will greatly improve clay provenance interpretations, including possible hydrothermal origins. Whole-rock geochemistry can also be used to improve stratigraphic correlations within and across the complex landslides in Basin West and Basin East. Lithium isotope geochemistry could be used to attempt to discriminate potential sources and constrain the different contributions to the Li budget in sedimentary Li mineralization (Meixner et al., 2020).

No new geochronological data were obtained during this study. The only age control on Kaiser Spring volcanic field rocks is from K-Ar studies (i.e., Reynolds et al., 1986; Moyer and Esperança, 1989) and most of those samples were taken outside of the Basin. New ⁴⁰Ar/³⁹Ar geochronology of rhyolite lavas, tuffs, and basalts on the surface and in drill core will constrain the timing of individual dome eruptions, will aid in calculating depositional rates for interbedded clay units, and can be used to correlate subsurface stratigraphy to help constrain basin geometry for future drilling.

Most Bradda Head Lithium, Ltd. drill holes were terminated at the base of the Upper Clay, and therefore, the stratigraphic architecture of the entire basin sedimentary and volcanic package is unknown. In addition, only Basin East and a portion of Basin North have currently been drilled. At the time of writing of this manuscript, Bradda Head Lithium, Ltd. continues to drill in Basin North, though the products of that drilling campaign are currently not publicly available. At least one deep drill hole (>500–600 m) near or at the center of the gravitational low would test the presence of diatreme breccias beneath the basin floor and shed light on the maar crater hypothesis. Additional drill holes to the base of the Lower Clay throughout the basin would constrain stratigraphy, determine Li and Mo geochemical trends, and identify the margins of the Li-mineralized clays. Drill holes at basin margins could further identify mass wasting or alluvial fan deposits related to possible maar crater wall instability and would constrain maar crater geometry.

Two large, complex landslides (Calico and Little Calico landslides) occupy much of Bradda Head Lithium Ltd.’s Basin East and Basin West claims (Fig. 5). To complicate matters more, high-angle normal faults cut the stratigraphic units within the slides prior to mass wasting events. For these reasons, stratigraphic relationships within the slides are very poorly understood. Lithologic units change over just a few meters, many beds are vertical or overturned, and the entire area is covered in an angular basalt lag, originating from the capping basalt that has been highly extended due to mass wasting. Mapping at 1:5,000 scale within these slides would do the following: (1) greatly improve stratigraphic correlations across the basin, (2) identify fault block domains (which are now obscured by mass wasting) that could be isolated for future drilling, and (3) help constrain basin geometry on in the areas of the slides.

Presently very few units can be correlated from Basin North across Basin West into the Eastern volcanic belt.

Our integration of the first 1:24,000-scale geologic map of the Miocene Kaiser Spring volcanic field, northwestern Arizona (Thompson et al., 2023), with detailed sedimentology and stratigraphy, geochemistry, and borehole sedimentological logs and assay data, provides critical insights into the formation of the Basin in the Eastern Kaiser Spring volcanic field and the origins of Li-rich volcano-sedimentary deposit within. The Basin is a small, deep depocenter that contains a stratigraphically continuous Neogene package of multiple nonvolcanic and volcaniclastic conglomerates and sandstones; felsic lithic lapilli tuffs, ash tuffs, breccias, and lapillistones; basalt lava flows, scoria, and basaltic tephra; claystones; dolomite, magnesite, and agate; and minor sandstones—all of which unconformably overlie Proterozoic bedrock. All these units are included within one or more complex landslides or rock slab avalanches, which often obscure stratigraphic relationships at the surface. Cobble-pebble imbrication in alluvial conglomerates indicate the Basin was a closed hydrologic system prior to deposition of the volcano-sedimentary Li deposit.

Bradda Head Lithium, Ltd., drill core reveals the Basin volcano-sedimentary Li deposit consists of two thick, dominantly magnesian smectite-clay units (Upper and Lower Clay) interbedded with nonmineralized lithic lapilli tuff. Multicolored clay units are laterally continuous, display consistent thicknesses across the Basin, and do not grade into equally thick sand deposits or other basin margin facies. Instead, clay overlies and is locally capped by coarse volcanic and nonvolcanic fanglomerates which represent aggradation from steep basin margins into the basin depocenter.

The Basin sedimentary and volcanic package displays centroclinal dips focused downward into the locus of a semicircular gravitational low with 2 mGal of relief over 1.7 km, indicating postdepositional subsidence over a similarly shaped subsurface void space—possibly a maar crater and diatreme cut into Proterozoic bedrock. This interpretation is supported by several other lines of geologic evidence including the following: (1) local Middle Miocene basaltic lava flows and cinder cones that unconformably overlie Proterozoic bedrock; (2) a lack of basin-bounding faults that could have formed hundreds of meters of accommodation space; and (3) numerous basalt block sags sunk into local alluvial deposits indicating local phreatic explosive activity.

Alkaline lake water in the closed hydrologic basin was likely enriched with Li, Mo, and other trace elements via the hydration of Eastern volcanic belt rhyolite dome vitrophyre, the hydration/dissolution of volcanic ash, and hydrothermal fluid circulation/mixing with meteoric water. All 12 Eastern volcanic belt rhyolite domes have partially to completely hydrated (perlitic) bases with low Li concentrations (11–16 ppm), whereas nonhydrated rhyolite lavas and tuffs are enriched in Li (54–247 ppm) beyond average crustal levels. Volcanic sinter, composed of laminated cryptocrystalline silica and Sr-rich dolomite mounds and spring conduits crosscut and cap claystone, and are always associated with abundant chalcedonic agate. The sinter is interpreted as hydrothermal spring deposits, an explanation that is further supported by positive correlations with Mo and F in the Upper Clay. Other major trace element concentrations commonly found in hydrothermal fluids including Sb, As, K, Rb, S, and Tl are elevated above and below the Li-rich interval (Li > 1,200 ppm) in the Upper Clay.

Structural and sedimentological evidence suggest that the Basin was a closed hydrologic system likely flooded with alkaline, silica-saturated water. Given the abundance of Eastern volcanic belt volcanic ash deposits throughout the drainage basin, and the presence of lithic lapilli tuffs interbedded with clay, it is likely volcanic ash fell over and was suspended in this alkaline lake environment. Indeed, microscopic analyses of sieved Upper and Lower Clay reveal the nonclay fraction is dominantly composed of broken crystals of quartz, biotite, and hornblende likely derived from explosive Eastern volcanic belt dome eruptions; each of these domes had one or more pyroclastic and phreatomagmatic eruptions contemporaneous with basin sedimentation. Li-rich magnesian smectite clays could have formed in several ways: via neoformation—or direct precipitation—from silica-saturated, alkaline, saline lake water enriched in Li and other trace elements through glass hydration; via diagenesis of smectite or felsic volcanic ash dominantly composed of glass shards; or via hydrothermal alteration of volcanic ash. It is likely that Li/Mo enrichment occurred through a combination of all these possibilities. Specific clay origins as well as dominant Li-host mineral species, remains unresolved, however, due to the dearth of systematic mineralogical analyses on drill core.

Although the volcano-sedimentary deposit within the Basin is the first to be described in a bimodal volcanic field and in a maar crater, the Li-, K-, and Mo-mineralized claystone within is analogous to known Li-rich deposits at McDermitt caldera/Thacker Pass, Nevada, Clayton Valley, Nevada, Rhyolite Ridge, Nevada, and Sonora, Mexico, in that it occurs in a closed hydrologic basin, proximal to Li-rich felsic lavas and pyroclastic rocks, and it is composed dominantly of magnesian smectite clays. Other volcano-sedimentary deposits in northwestern Arizona are proximal to the Kaiser Spring volcanic field, including the Big Sandy Valley deposit, currently being explored by Arizona Lithium, and the Lyles clay (Thompson Valley) deposit, currently unexplored.

This research was funded in part by the U.S. Geological Survey Earth MRI (Mapping Resources Initiative) and National Cooperative Geological Mapping Program under cooperative agreement G20AC00166. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. government. This manuscript is submitted for publication with the understanding that the U.S. government is authorized to reproduce and distribute reprints for governmental use.

We thank Bradda Head Lithium Ltd. for permission to use their assay and drill core data. Hugo Zuniga (Bradda Head) and John Keller (formerly Bradda Head) provided valuable insight and collegial data sharing, which greatly helped improve this study’s outcomes. Alexandra Huff (Arizona State University; ASU) processed and interpreted the Hyperspectral Thermal Emission Spectrometer (HyTES) data, which refined geologic mapping of hydrothermal deposits. Thank you to Calvin Mako (Arizona Geological Survey; AZGS), Tawnya Wilson (AZGS), Tom Benson, and other anonymous reviewers for their constructive and detailed critiques, which greatly improved the structure and poignancy of this manuscript. Lastly, we thank Phil Pearthree (AZGS) and Randi Bellassai (AZGS) for their professional support of this project.

Lisa Thompson is a research scientist at the Arizona Geological Survey (AZGS), University of Arizona. Prior to her appointment, she held a senior lecturer position at Northern Arizona University, where she taught geographic information systems (GIS), field methods, geologic disasters, and volcanology. She received her M.S. degree at Northern Arizona University (2005) and her B.S. degree at Temple University (2002). Lisa’s interests include physical volcanology, sedimentology and stratigraphy, field geology, and GIS, with applications to carbon sequestration and sedimentary lithium mineralization. Her current research includes investigating the carbon sequestration potential of mineralizing CO₂ in glassy, mafic rocks ex situ, CO₂ storage in stacked saline reservoirs, H₂ storage in salt bodies, and a variety of volcano-sedimentary depositional environments in Arizona.