Tephras are significant markers in the stratigraphic record and play a key role in establishing paleoenvironmental and paleoclimate histories worldwide. Despite burgeoning research focused on tephra characterization and correlation techniques, there are still few techniques that allow for the direct dating of tephra, particularly below the lower age limit of K/Ar and Ar/Ar dating methods.
Radiation Measurements 153 (2022) 106731 Contents lists available at ScienceDirect Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas Exploring the application of blue and red thermoluminescence for dating volcanic glasses K Rodrigues a, *, S Huot b, A Keen-Zebert a a b Division of Earth and Ecosystem Sciences, Desert Research Institute, 2215 Raggio Parkway, Reno, NV, 89512, USA Illinois State Geological Survey, Prairie Research Institute, University of Illinois at Urbana-Champaign, Champaign, IL, 61820, USA A R T I C L E I N F O A B S T R A C T Keywords: Luminescence Tephrochronology Thermoluminescence Volcanic glass TL Tephras are significant markers in the stratigraphic record and play a key role in establishing paleoenvir onmental and paleoclimate histories worldwide Despite burgeoning research focused on tephra characterization and correlation techniques, there are still few techniques that allow for the direct dating of tephra, particularly below the lower age limit of K/Ar and Ar/Ar dating methods In this study, we test different thermoluminescence (TL) dating approaches on the 4–11 μm volcanic glass constituents of three different independently different tephras By comparing against independent age control, we demonstrate the utility of both blue (320–450 nm) and red (587–651 nm) TL emissions for dating volcanic glasses using single aliquot regenerative (SAR) dose techniques We find that both blue and red TL emissions from the volcanic glass shards are dim but reproducible and show no evidence for significant sensitivity changes occurring between the natural TL and the first test dose during the SAR protocol Fading tests on the blue TL signal show that g-values range from 1.6 ± 1.0 to 2.9 ± 1.1%/decade and are statistically indistinguishable with zero at 2σ for the red TL Bleaching experiments show that both blue and red TL signals are sensitive to light exposure, with sensitivity corrected signals declining by ~40% over a 2-h period For all three tephras, both the fading-corrected blue and red SAR-TL ages are consistent with age expectation These successful results demonstrate the effectiveness of TL techniques for determining the eruption ages of tephra deposits in primary position between ~1 and at least 30 ka Introduction from other eruption events (Lowe, 2011) A major obstacle in the application of tephrochronology is accurate age determination The age of tephras is most commonly established either directly by radiometric methods (commonly K/Ar or Ar/Ar, e.g., Van den Bogaard, 1995), or fission track dating of primary mineral constituents for older deposits (≳100 ka, e.g., Seward, 1974) or indirectly by radiocarbon dating of associated organic material for younger tephras (75% SiO2) on the basis of major and minor elemental analysis of volcanic glass shards by electron microprobe (Benson et al., 1997; Kuehn and Negrini, 2010; Bursik et al., 2014; Pouget et al., 2014) Adams, 2003) The only local age constraint of the Turupah Flat tephra at Salt Wells is based on a single radiocarbon age (650–920 cal years BP) determined from a charcoal sample collected immediately below the tephra layer within ~20 m of the site sampled in this study At the site selected for sample collection, the Turupah Flat tephra is ~5 cm in thickness and situated 0.4 m below the ground surface The tephra layer has an abrupt lower contact and grades upward into parallel laminae of silt and tephra throughout the upper cm (Fig 2A) Adams (2003) documents the depositional environment, geomorphology, stratigraphic succession, and relevant geochronology at this site 2.1 Turupah Flat tephra The Turupah Flat tephra bed is comprised of a series of geochemi cally similar tephras dated between 0.6 and 2.0 ka years ago (Table 1, Wood, 1977; Davis, 1978; Miller, 1985; Sieh and Bursik, 1986; Adams, 2003) The Turupah Flat tephra sampled in this study is exposed at Salt Wells on the landward (south) side of a beach barrier that borders a small playa in northern Nevada The beach barrier—originally mapped as the Fallon lake shoreline by Morrison (1964) and since referred to as the Salt Wells beach barrier (Adams, 2003)—has been interpreted to represent a lake stand that covered most of the Carson Sink (a remnant of Lake Lahontan) during the late Holocene The Turupah Flat tephra is thought to have been deposited through overwash processes on the backside of the beach barrier at the time of this highstand (Davis, 1978; 2.2 Trego Hot Springs tephra The Trego Hot Springs tephra is widespread throughout the Lahon tan basin and has been independently dated to 23.4 ka (Table 1, Davis, 1983; Berger, 1991, Benson et al., 1997) For this study, the Trego Hot Springs tephra was sampled at a locality on Squaw Creek (the southern amphitheater of Davis, 1983), a site that has been studied thoroughly over the last 40 years (Davis, 1983; Benson et al., 1997; Adams, 2010) Here, Squaw Creek has exposed parts of a delta belonging to the Sehoo Formation of Morrison (1964) An ~12–15 cm thick exposure of the Trego Hot Springs tephra is situated near the base of the delta and grades upward into an ~30 cm thick siltier unit with prominent reworked K Rodrigues et al Radiation Measurements 153 (2022) 106731 tephra (Fig 2B) Table The SAR-TL protocol applied to the samples in this study 2.3 Wono tephra Step Approximately km southeast of the Turupah Flat tephra sampled at Salt Wells is a prominent exposure of the Wono tephra, located at the northern end of the Bunejug Mountains The Wono tephra has a wellestablished age of 27.3 ka (Table 1, Davis, 1983; Benson et al., 1997; Zic et al., 2002) At the site selected for sample collection, the Wono tephra is interbedded with coarse beach gravels and has been inter preted to represent deposition in backset beds on the margin of a former lake (Adams, 2010) At this location, the Wono tephra forms an ~15 cm thick bed with abrupt upper and lower contacts (Fig 2C) a Treatment Observation a Give dose, Di Preheat (200 ◦ C for 10 min) TL measurement to 450 ◦ C at ◦ C/s Background TL measurement to 450 ◦ C Administer test dose (50 Gy) Preheat (200 ◦ C for 10 min) TL measurement to 450 ◦ C at ◦ C/s Background TL measurement to 450 ◦ C Return to step Lx Tx For measurement of the natural signal, i = a heating rate of ◦ C/s The background TL was recorded in a second measurement on the same aliquot immediately after signal readout and subtracted channel-wise to obtain net signals TL measurements were made on 9.8 mm diameter stainless steel discs mounted with ~1 mg of sample by settling in acetone Methods 3.1 Sample collection and preparation At each of the three sites, luminescence samples were taken directly from the tephra bed by hammering steel tubes (for Wono and Trego Hot Springs tephras: ~20 cm L x ~5 cm D, for Turupah Flat tephra: ~10 cm L x cm D) into freshly cleaned vertical sections The ends of tubes were then wrapped to avoid light exposure during transport To account for the heterogeneous gamma dose rate environments at each of the lumi nescence sample sites, sediment samples were collected from the upper and lower bounding layers for dose rate assessment A third ‘average’ dose rate sample was collected from sediment within a 30 cm radius surrounding the luminescence sample tube This third dose rate sample was used exclusively to check for equilibrium conditions in the 238U decay chain Luminescence sample preparation was conducted at the DRI Lumi nescence Laboratory (DRILL) Preparation of the tephra for TL mea surement adapted the methodology from Berger (1991) Following this protocol, the tephra samples were chemically treated to remove car bonates and organic material (10% HCl and 30% H2O2, respectively), and the fine grained (4–11 μm) fraction was separated from the bulk sample by extraction from suspension at appropriate settling velocity times according to Stokes’ Law Volcanic glass was isolated from bulk tephra using a solution of lithium heteropolytungstate in methanol prepared with a specific gravity of 2.45 g/cm3 and centrifuged at 3000 rpm for 10 The heavy liquid separation protocol was carried out a minimum of two times with the float The effectiveness of the separation technique was evaluated by visual inspection under both petrographic and scanning electron microscope (SEM) at the University of Nevada, Reno Microbeam Laboratory Panchromatic SEM-cathodoluminescence (SEM-CL) was also applied to individual glass shards in effort to char acterize them further 3.3 Initial testing and measurement protocols To define the thermally stable part of the TL glow curve and deter mine an appropriate preheat temperature for experimentation, a plateau test was carried out on three aliquots of each sample by using the ratio of natural TL to the TL after laboratory bleaching (herein defined as heating to 450 ◦ C to remove the signal) and subsequently administering a β-dose (Aitken, 1985) Standard SAR procedures were carried out following the methods of Murray and Wintle (2000) and incorporated a preheat of 200 ◦ C for 10 to eliminate the thermally unstable part of the TL glow curve prior to TL readout, 4–5 regeneration doses bracketing the natural dose including a recycled dose and a zero dose, in addition to sensitivity correction with a 50 Gy test dose (Table 2) SAR-TL dose recovery tests were carried out for both blue and red TL to further test the reliability of the SAR protocol and determine the spread of the recovered doses Blue TL dose recovery tests (Murray and Wintle, 2003) were carried out on 10 aliquots of each sample that were first bleached and then β-irradiated with a known dose approximating the natural equivalent dose (natural De) Red TL dose recovery tests followed the same protocol but for only aliquots of Trego Hot Springs All previously published TL dating work on volcanic glass (Berger and Huntley, 1983; Berger, 1985, 1987, 1991; Berger and Davis, 1992) has employed a multiple aliquot additive dose (MAAD) approach using the methodology described in Aitken (1985) However, MAAD ap proaches require significantly more prepared material and yield values of De that are based on extrapolation and yield lower age precision relative to interpolative approaches like the single aliquot regenerative protocol (SAR, Murray and Wintle, 2000) In order to test whether a SAR approach would be appropriate for both blue and red TL measurements on volcanic glass, a single-aliquot regeneration and added dose (SARA, Mejdahl and Bøtter-Jensen, 1997) approach was applied to assess for potential sensitivity changes occurring between natural and regenera tive TL readout Owing to sample availability, the SAR-SARA experi ments were only carried out for Trego Hot Springs The SAR-SARA experiments included four to five groups of aliquots which were given a different additive prior to the determination of De by our SAR-TL pro tocol (Table 2) These SAR-TL measured doses were then plotted against the known added doses, and the value of De was obtained by a linear extrapolation of the data to the dose axis at the intercept A line with a slope of was considered to reflect insignificant sensitivity change be tween measurement of the natural and first regenerative dose The ‘Luminescence’ package (Kreutzer, 2021) for R was used to calculate De values using the SAR-TL protocol The appropriate TL in tegral used for De calculation was determined by identifying a stable region showing a plateau in De values (Aitken, 1985) All results of TL 3.2 Luminescence measurements TL measurements on each sample were performed using two different methods, each utilizing a different range of emission spectra for measurement: blue TL and red TL Blue TL measurements on separated volcanic glass were conducted using a Risø TL/OSL-DA-20 reader housed at the DRILL with an integrated 90Sr/90Y β-source delivering a dose rate of 0.10 Gy/s Blue TL signals were detected with an EMI 9235QA photomultiplier tube (PMT) after passing through a blue filter pack (4 mm of Corning 7–59 in combination with mm Schott BG 39) Red TL measurements were conducted on a Lexsyg Smart TL/OSL reader housed at the Illinois State Geologic Survey OSL Dating Lab with a90Sr/90Y β-source delivering a dose rate of 0.074 Gy/s Red TL emission was detected through a thermoelectrically cooled H7421-40 Hama matsu PMT after passing through a combination of Chroma ET 620/60 and Schott KG filters The beta sources for both systems were cali brated with 4–11 μm quartz prepared by Risø All TL measurements were conducted in an N2 atmosphere and using K Rodrigues et al Radiation Measurements 153 (2022) 106731 Fig Images of Wono bulk tephra pre- (A) and post-volcanic glass separation (B) (C) and (D) are images of the same glass shard displaying a series of unidentifiable elongated or acicular microlites under SEM (C) and as observed under cathodoluminescence (CL) Note the dim visible range luminescence emitted from these microlites under CL measurements were required to pass the following criteria for further analysis: >40 ◦ C plateau range,