Different optically stimulated luminescence (OSL) single aliquot regenerative (SAR) saturation characteristics have been previously reported for quartz of diverse grain sizes. This preliminary electron spin resonance (ESR) study documents on paramagnetic signals in natural sedimentary quartz extracts of different grain sizes (4–11 μm, 63–90 μm, 90–125 μm, 125–180 μm, 180–250 μm) extracted from loess and sand samples that are amendable to OSL and ESR dating.
Radiation Measurements 120 (2018) 59–65 Contents lists available at ScienceDirect Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas Electron spin resonance characterisation of sedimentary quartz of different grain sizes T A Timar-Gabora,b,∗ a b Interdisciplinary Research Institute on Bio-Nano-Science, Babeş-Bolyai University, Cluj-Napoca, Romania Faculty of Environmental Sciences and Engineering, Babeş-Bolyai University, Cluj-Napoca, Romania A R T I C LE I N FO A B S T R A C T Keywords: Quartz Electron spin resonance (ESR) Al-hole centre Ti-centres E1′ centre Peroxy centre Grain size Different optically stimulated luminescence (OSL) single aliquot regenerative (SAR) saturation characteristics have been previously reported for quartz of diverse grain sizes This preliminary electron spin resonance (ESR) study documents on paramagnetic signals in natural sedimentary quartz extracts of different grain sizes (4–11 μm, 63–90 μm, 90–125 μm, 125–180 μm, 180–250 μm) extracted from loess and sand samples that are amendable to OSL and ESR dating Spectra recorded at room temperature and low power (< mW) are dominated by the E1’ centre, whose intensity in natural samples was found to decreases with increasing grain size The signature of titanium centre could not be detected in fine (4–11 μm) quartz For coarse fractions (> 63 μm) titanium-lithium [TiO4/Li+]0 signals increase with increasing grain size Aluminium-hole ([AlO4]0) signals are observed in all natural and laboratory irradiated investigated samples when spectra are recorded at low temperature (90 K) The intensity of these signals appears to decrease with increasing grain size, however, room temperature measurements show that these signals are highly interfered by a variety of signals tentatively attributed to peroxy (g ≈ 2.007 and g ≈ 2.004) with significantly higher intensities in fine grains (4–11 μm) A decrease of their intensity is reported when grain size increases and partial evidence that these defects are preferentially located in damaged regions of the grains is presented A dose dependent paramagnetic signal at g ≈ 2.011 was detected only in 4–11 μm quartz The stronger signature of the [TiO4/Li+]0 signals in larger grains coupled with the weaker signals of peroxy signals interfering with [AlO4]0 signal measurement is suggesting that coarser fractions should be preferred for conventional ESR dating using aluminium-hole and titanium signals The understanding of the implication of these defects in OSL dating alongside with their concentration dependency on grain size requires further investigations Introduction By analysing different grain sizes of quartz from different sedimentary contexts around the world it was concluded that the saturation characteristics of single aliquot regenerative (SAR) optically stimulated luminescence (OSL) signals are anticorrelated with grain diameter (Timar-Gabor et al., 2017) While significantly higher saturation characteristics are reported for the fine (4–11 μm) quartz fraction compared to coarser fractions (> 63 μm), there is increasing evidence that laboratory saturation at higher doses does not provide a corresponding extended age range in luminescence dating (Timar-Gabor and Wintle, 2013; Timar-Gabor et al., 2017) Saturation of luminescence signals in quartz can be dictated by various physical processes related to both charge trapping and recombination However, defects giving rise to luminescence (thermoluminescence (TL) and optically stimulated luminescence (OSL)) in quartz have not yet been unambiguously identified Electron spin resonance is one of the few methods sensitive enough to monitor defects in quartz It well known that a large number of ∗ paramagnetic defects are present in quartz such as E′- centres, peroxy centres and oxygen hole centres, [H3O4]0, [AlO4]0, [AlO4/H+]+, [GeO4/ Li+]0, [TiO4/M+]0 (where M+ denotes an alkali metal or hydrogen ion), etc (Preusser et al., 2009) These ubiquitous defects can be involved through different mechanisms in optically stimulated luminescence as electron traps, recombination centres or competitors for trapping or recombination Al is always present in quartz, and is giving rise to [AlO4/M+]0 (where + M denotes an alkali metal or hydrogen ion), and [AlO4]0, respectively (Malik et al., 1981) Studies on radiation-induced ionic conductivity of quartz (Weil, 1984; Martini et al., 1986) showed that irradiation at room temperature leads to the dissociation of [AlO4/M+]0 resulting in the formation of [AlO4/H+]0 and [AlO4 ]0 centres In addition, Mondragon et al (1988) suggested that the relative proportion of [AlO4/H+]0 and [AlO4]0 might be dose-rate dependent While studies performed in the late 1980s (e.g McKeever, 1991) suggested that electron-hole recombination at [AlO4]0 sites yields the TL emissions near 470 nm, whereas recombination Interdisciplinary Research Institute on Bio-Nano-Sciences, Babeş-Bolyai University, Treboniu Laurean 42, 400271, Cluj-Napoca, Romania E-mail address: alida.timar@ubbcluj.ro https://doi.org/10.1016/j.radmeas.2018.06.023 Received December 2017; Received in revised form 15 May 2018; Accepted 26 June 2018 Available online 28 June 2018 1350-4487/ © 2018 The Author Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/) Radiation Measurements 120 (2018) 59–65 A Timar-Gabor at the [H3O4]0 site yields the 380 nm TL emissions, more recent studies correlate [AlO4]0 to the luminescence centres involved in the production of OSL (emissions around 380 nm) (Martini et al., 2009, 2012 a,b, 2014) [AlO4]0 signal is also frequently used in ESR dating of sediments (Tissoux et al., 2010, Bahain et al., 2012, Moreno et al., 2012, Voinchet et al., 2013, Tsukamoto et al., 2015 etc.) E1’ centre is a paramagnetic oxygen vacancy formed from a neutral oxygen vacancy which traps an electronic hole It was proposed that these holes can be released upon heating by Al-hole defects (Rudra and Fowler, 1987) Experimental results (Woda et al., 2002, Schilles et al., 2001) support the idea that this centre may act as a non-radiative recombination centre acting in competition with the emission bands in the near UV and visible region Ti4+ may substitute Si4+ in quartz with no charge compensation, creating [TiO4]0, the precursor for the Ti centre (Weil, 1984) Upon room temperature irradiation, Ti4+ may trap an electron together with an alkali ion M+ for charge compensation, forming [TiO4/M+]0, where M+ can be either Li+, H+ or Na+ (Toyoda, 2015) While signals form Ti-centres received considerable attention lately in ESR dating of sediments (Beerten et al., 2006; Duval and Guilarte, 2015; Duval et al., 2017), to our knowledge, their possible involvement in OSL production has been less documented ESR signals of the Ge centres are observed usually in irradiated quartz samples (Toyoda, 2015) From the precursor [GeO4]0, the paramagnetic defect [GeO4]- is formed through electron trapping This centre is not stable at room temperature, and recent studies present strong evidence that this defect can be correlated to the 110 °C TL peak electron trap (Vaccaro et al., 2017) [GeO4/M+]0 centres, and most notably [GeO4/Li+]0 have a higher degree of thermal stability (Vaccaro et al., 2017) Oxygen –associated trapped hole centres have been identified in silica and denoted “dry OHC” and “wet OHC” according to the OH− content of the material (Stapelbroek et al., 1979) Dry OHC (Friebele et al., 1979) is an oxygen interstitial in the form of ≡Si-O-O∙ (where “≡” represents the three Si-O bonds and “∙” represents an unpaired electron), and is referred to as “peroxy centre” by Ikeya (1993), a nomenclature further used here Wet OHC is an oxygen hole in form of ≡Si-O∙, denoted in the same nomenclature as non-bonding oxygen hole centre (NBOHC), while OHC term is used to denote a dose dependent signal, sometimes ascribed to the ozonide radical (O3−) (Ikeya, 1993) Although their existence in quartz was proposed decades ago, their characteristics in crystalline quartz and their implication for dating has been less documented than the other signals described above However, a combination of five peroxy signals was identified in natural quartz that has not been subjected to additional artificial irradiation by Botis et al (2005) and their structure has been further investigated by Nilges et al (2009) and Pan and Hu (2009) In this study we complement previous luminescence investigations with electron spin resonance (ESR) investigations on different grain sizes (4–11 μm, 63–90 μm, 90–125 μm, 125–180 μm, 180–250 μm) of sedimentary quartz A dependence of the most important paramagnetic defects listed above as function of grain size is presented allowing for some preliminary conclusions regarding their implication in dating to be drawn Materials and methods 2.1 Samples Sedimentary quartz samples of different grain sizes extracted mostly from loess have been investigated Quartz has been extracted according to standard OSL preparation techniques under subdued red light (see further information in the supplementary file) During ESR measurements the samples have not been exposed to direct sunlight and their exposure to weak natural light was restricted to minimum The selection of samples was based on: (i) the availability of sufficient material for ESR investigation, (ii) the high purity of the quartz extracts as confirmed by routine tests in OSL dating as well as by scanning electron microscopy imaging coupled with chemical analysis of local area by energy dispersive X-ray spectroscopy (EDX) using a FEI Quanta 3D FEG dual beam microscope (see Fig S1), (iii) (caption on next page) 60 Radiation Measurements 120 (2018) 59–65 A Timar-Gabor Fig Examples of Al-hole (a), titanium (b) spectra recorded at 90 K and E1′ signals (c) measured at room temperature for 125–180 μm quartz (sample STY 1.10) For this particular sample a dose of 100 Gy was given on top of the naturally accrued dose Measurement parameters are given in Section the excellent behavior of the samples in the single aliquot regeneration protocol, with no aliquot rejected due to poor recycling, IR depletion or recuperation (Murray and Wintle, 2003) Sample Rox 1.14 originates from Roxolany loess palaeosol section, Southern Ukraine and was collected below the Brunhes/Matuyama polarity transition An electron spin resonance age of over Ma was obtained for this sample using a multiple centre (Al and Ti) approach (Duval et al., 2017) This result, along with optically stimulated investigations using both the standard single aliquot regenerative (SAR) multi grain OSL procedure as well as single grain investigations are presented in detail in Anechitei-Deacu et al., this issue Sample Rox 1.9 is quartz extracted from a loess sample collected just above the Eemian soil (MIS 5e) from the same section Quartz samples Sty 1.10 and Sty 2.4, were extracted from samples collected from Stayky loess palaeosol section, Northern Ukraine For Sty 2.4 equivalent doses of 69.7 ± 0.8 Gy and 59.7 ± 1.5 Gy were obtained using standard SAR-OSL dating using 4–11 μm and 63–90 μm quartz respectively For sample Sty 1.10 an equivalent doses of 492 ± 30 Gy was obtained on 4–11 μm quartz, however in the case of coarse grains the natural signals were found close (about 85%) to SAR laboratory saturation The OSL chronology of this section as well as extended SAR-OSL dose response curves on these samples are presented in detail in Veres at al., submitted For further experiments, calibration quartz (Cal Q) provided by Risø National Laboratory (Hansen et al., 2015) was used, due to the availability of high amounts of sample The 180–250 μm quartz fraction was separated from aeolian sand from Rømø, Jutland, Denmark, through conventional sample preparation techniques (Hansen et al., 2015) It is important to note that this quartz sample has only undergone physico-chemical preparation, without the sensitisation (annealing, dosing, annealing) 2.2 Instrumentation and measurement protocols Electron spin resonance analyses have been carried out using an X band Bruker EMX plus spectrometer A high sensitivity cavity (ER 4119HS Bruker resonator) with an unloaded quality factor of 15,000 and loaded factor of about 7800 for the investigated samples was used Low temperature measurements were performed using a variable temperature unit Care was taken that all samples are centered inside the cavity and have the same height The mass of one sample was approximately 200 mg in the case of coarse grains (> 63 μm) and about 120 mg in the case of fine grains, with variations of 10% Mass normalisation was applied for the measured signals, where relevant Samples have been rotated in the cavity for collecting several spectra using a programmable goniometer If not otherwise mentioned measurement parameters employed for recording Al-hole ([AlO4]0) signals were: temperature of 90 K, modulation frequency 100 kHz, modulation amplitude of G, 3350 G centerfield, 300 G sweep width, 120 s sweep time, 40 ms conversion time, 40.96 ms time constant Microwave power was mW and the sample has been rotated times (every 1200) in the cavity A single scan per angle was carried out and the average of the three spectra was computed For titanium centres ([TiO4/M+]) the following measurement parameters have been used: measurement temperature 90 K, modulation frequency 100 kHz, modulation amplitude 1G, 3490 G centerfield with a 220 G sweep width, 22s sweep time, 10 s conversion time, 20.48 ms time constant Microwave power was 10 mW and 30 scans have been performed E1' spectra have been recorded at room temperature at an optimal power of 0.02 mW selected based on power saturation experiments 12 rotations (every 300) have been carried out The modulation amplitude was 0.1 G, centerfield was set to 3360 G, the sweep width was 20 G with a sweep time of 40 s, conversion time of 20 ms and time constant of 20.48 ms Measurements have generally been repeated 2–5 times at a few weeks interval in order to check the reproducibility of the (caption on next page) 61 Radiation Measurements 120 (2018) 59–65 A Timar-Gabor Fig Dependece of Al-hole (a), titanium (b) E1’ (c) natural signals magnitude as function of the grain size of quartz for the available grain sizes from samples Rox 1.14 (squares), STY 1.10 (up triangles) Rox 1.9 (circles) and Sty 2.4 (down triangles) Measurement parameters are given in Section Mass normalisation to a mass of 100 mg was performed for peak to peak intensities of the signals measured as shown in Fig we have investigated so far despite applying repeated scans or irradiating the samples The lack of these signals is further depicted in Fig S4, for a fine grained sample irradiated with 10,000 Gy Titanium signals intensity linearly increases with increasing grain size in the case of coarse grains (> 63 μm) Results reporting an increase in sensitivity with increasing grain size have also been reported for fluvial sediments by Liu et al (2015) As far as Al-hole and E1′- centres are concerned, both their concentration seems to decrease as the grain size increases Larger grain sizes are characterized by lower E1′- centre concentration as well as apparent lower Al-hole concentrations The differences observed between grain sizes could be attributed to the presumably different origins of the fine and coarse quartz material, as the fine material (4–11 μm) could originate from more distal sources However, it can be observed that the trend mentioned above is valid for samples of different origins Stayky is located on the upper Dnieper River that is only draining the East European platform comprising very old mainly metamorphic and magmatic rocks of Proterozoic age, while Roxolany sits at the mouth of Dniester river, that is has its origin in the Outer Carpathian Mezozoic flysh deposits, as such the sources of the sedimentary quartz from Sty and Rox are expected to be of different origin 3.2 Contribution of peroxy signals to Al-hole signal measurement It is important to note that in the case of Al-hole signal the signature recorded for finer grains displays a different shape (Fig 3) This was observed in the case of all samples investigated in this study and all fine grain quartz samples investigated so far in our laboratory For the sake of completeness Fig S5 additionally presents a comparison of the natural signals for sample Sty 2.4 It can be noted that the hyperfine signature is less pronounced for the signals of fine quartz and the signal is reaching its maximum value at g ≈ 2.008, while in the case of coarse grains for natural signals peaks at g ≈ 2.018 as expected for typical aluminium-hole signals This difference can be more easily observed when comparing natural signals to signals measured after laboratory doses are given on top of the naturally accrued dose After adding laboratory doses the signal of fine grains gradually increases and reaches its maximum at g ≈ 2.018 (Fig S6) As this observed apparent difference in shape could be attributed to interfering signals, measurements using the same parameters as in the case of aluminium-hole measurements but conducted at room temperature were carried out, as is well known that when measuring at room temperature Al-hole signals can no longer be observed due to the very short spin-lattice relaxation time of the defect As such, if the interfering signal could be detected at room temperature, its signature should be discernible in these room temperature spectra In Fig a such room temperature spectra recorded for fine (4–11 μm) quartz are presented in comparison to spectra collected in the same manner for coarse (90–125 μm) quartz It can be noted that signals at g ≈ 2.0084–2.0063 (indicated by “A” in Fig a)) and g ≈ 2.0047–2.0022 (indicated by “B” in Fig a)) are present in both fine and coarse quartz grains For better visualization a comparison between room temperature spectra and liquid nitrogen spectra for fine (4–11 μm) and (90–125 μm) quartz respectively is presented in Fig S7 These signals are not increasing upon gamma irradiation as shown in Fig b) for fine quartz and Fig c) for coarse quartz respectively We tentatively attribute these signals to peroxy radicals (Botis et al., 2005; Nilges et al., 2009; Pan and Hu, 2009) The signature of these signals ascribed to peroxy centres are visible in room temperatures spectra as well as spectra recorded at liquid nitrogen after the samples were heated to 500 °C, a temperature for which Al-hole signals are annealed (Toyoda and Ikeya, 1991) (see Figs S8–S9) Further examination of spectra recorded at room temperature reveal the weak presences of Ge centres (Ikeya, 1993), namely [GeO4/Li+]0 at g ≈ 1.996 both in the case of fine and coarse grains when spectra are recorded at room temperature (Fig 4b) and c)) Notably, only in the case of fine (4–11 μm) quartz a dose dependent paramagnetic signal at g ≈ 2.011 was detected Ikeya (1993) reported such a signal (see their Fig Dependece of Al-hole natural signals shape as function of the grain size of quartz for sample Rox 1.14 Natural signals are represented Measurement parameters are given in Section system but also for quantifying average values and associated errors Examples of such reproducibility tests are presented in supplementary material (Figs S2 and S3) Irradiations have been performed at Centre for Nuclear Technologies, Technical University of Denmark (DTU NUTECH) using a calibrated 60Co60 gamma cell, with a dose rate of Gy/s (dose rate to water) at the time of irradiation Dose rate to quartz was estimated to be 96% of dose rate to water based on Monte Carlo simulation considering the irradiation geometry used Results and discussion 3.1 ESR signals and their dependence on grain size Typical examples of spectra showing Al-hole, titanium and E1’ signals are presented in Fig Baseline correction was not performed unless otherwise mentioned Signals have been quantified using peak to peak height as indicated in the figure, namely form g = 2.018 to g = 1.993 in the case of Al-hole signals as recommended by Toyoda and Falguères (2003) and widely used in ESR dating studies, and form g = 1.978 to g = 1.913 in the case of Ti signals respectively, denoted as option A in Duval and Guilarte (2015) Fig presents the dependence of the magnitude of natural signals as function of the grain size for available grain sizes of samples Rox 1.14, Rox 1.9, Sty 1.10 and Sty 2.4 There is no titanium signature in the case of fine (4–11 μm) grains These signals were not found present in neither of the samples presented in this study or in any other fine grained quartz sample 62 Radiation Measurements 120 (2018) 59–65 A Timar-Gabor Fig (a) Comparison between natural signals of sample Rox 1.9 fine (4–11 μm) and coarse (90–125 μm) quartz Please note that signals have not been mass normalised and the mass of the coarse material is two times higher than the fine counterpart (b) Dose dependece of the signals presented in panel a) for fine grains; (c) Dose dependecy of the siganls presented in panel a) for coarse grains; For all measurements the same measurement parameters as for Al-hole spectra (see section 2) were used, except the measurement being carried out at room temperature Fig 9.5) for silty quartz (20–60 μm) from a Scotland windblown sediment and tentatively ascribed it to OHC The dependence of the magnitude of these signals attributed to peroxy centres as function of grain size was analysed by recording spectra of natural signals at room temperature using the measurement parameters employed for Al-hole signal measurement Signal intensity Fig Dependece of peroxy natural signals (denoted as A and B in Fig 4a) intensity as function of the grain size of quartz Measurements have been carried out at room temperature using the same parameters as for measuring Al-hole signals (section 2) Peak to peak intensity values measured as depicted in Fig 4a) were mass normalised to a mass of 100 mg 63 Radiation Measurements 120 (2018) 59–65 A Timar-Gabor was quantified as peak to peak height as depicted in Fig 4a (g ≈ 2.0084–2.0063 and g ≈ 2.0047–2.0022, respectively) The dependence of signal intensity as function of grain size is presented in Fig (panel a) for signal denoted as “A” and panel b) for signal denoted as “B” in Fig 4a)) A significant decrease with increasing grain size can be observed In the case of Al-hole signals measured at 90 K, the ratio between the magnitude of the signals in the case of fine (4–11 μm) grains and coarse grains (> 63 μm) ranges from 1.1 to 1.8 depending on the sample and grain size considered (Fig panel a)) Higher natural signals in the case of finer grains could be expected due to the fact that this fine grain fraction has received an alpha dose contribution and larger grain sizes are expected to receive a smaller beta dose due to attenuation of beta radiation in the grain However, it is highly unlikely that solely this cause could be responsible for the trends observed for all signals Moreover, for the interfering signals measured at room temperature the ratio between the magnitudes in the case of fine grains to coarse grains is significantly higher, ranging from to even 15 depending on the sample and grain size considered (Fig 5a and b) More quantitative measurements involving signal deconvolution are needed, however, this indicates that care must be taken when Al-hole signals are being quantified in the presence of these overlapping signals As it can be observed that their intensity is higher in the case of finer grains, this observation, along with the fact that titanium signals could not be detected in fine grains (see section 3.1) is suggesting that coarser grains should be preferred for the application of standard ESR dating methods, especially if one attempts to derive equivalent doses in a multiple centre approach using both Al and Ti signals as suggested by Toyoda et al (2000), Rink et al (2007) or Duval et al (2017) Fig Intensity of peroxy signals (denoted as “A” and “B” in this study), E1′, Alhole and titanium-lithium signals respectively, as function of etching (HF 40%) time Values were normalised to the intensity measured following a 40 acid attack Datapoints represent the average of 2–4 measurements The continuous line represents unity and serves as an eye guide Dotted lines represent a 10% deviation from unity Scanning electron microscopy imaging of the grains after the h acid attack (Fig S10) and 24 h HF attack (Fig S11) show that etching does not progressively remove a uniform outer layer of the grain as generally thought but shows that for extended etching times the acid is rather digging inside the grain, preferentially removing the more damaged regions of the grains, while crystalline phases seem to be less affected Results of the etching experiment presented above along with the higher concentration of these defects reported for fine grains (which have received full alpha irradiation contrary to coarser fractions) brings further evidence that these signals attributed to peroxy centres come from defects localised in regions characterized by crystal damage, as suggested by Rink and Odom (1991) or Botis et al (2005) Furthermore, the reduction of the E′1 intensity with etching can also be explained assuming the argument that this defect is concentrated in regions associated with deformation of the lattice (Ikeya et al., 1992), irrespectively of the mechanism assumed for oxygen vacancy formation in nature, namely by high linear energy transfer radiations (Rink and Odom, 1991) or by low linear energy transfer radiations (Toyoda, 2005; Toyoda et al., 2005) 3.3 Effect of prolonged etching Peroxy radicals have been initially detected in amorphous silica (Friebele et al., 1979) and are generally accepted as being structural damage defects Griscom and Friebele (1981) presented radiation damage as a possible mode for formation of Frenkel defects, resulting in peroxy and E1′ formation, a model later adopted by Rink and Odom (1991) when discussing corresponding ESR signals in crystalline quartz If this model is correct one would expect these defects to have a significantly higher concentration in fine grains (4–11 μm) that have received alpha irradiation throughout their existence For coarser grains these defects should be located primarily on the surface of the grains For etched coarse grains it was suggested these centres are concentrated in radiation damaged areas and can be the product of alpha radiolysis of water incorporated in damaged areas by diffusion (Botis et al., 2005) We have undertaken extensive etching experiments on (180–250 μm) Cal Q grains A batch of calibration quartz was separated in six parts, each batch being subjected to hydrofluoric acid attack (40% concentration) for a different durations: 40 as generally performed in luminescence and ESR dating, h, h, h, h and 24 h, respectively The main ESR signals were measured as described above and their mass normalised signals (normalisation to 100 mg) are plotted as function of etching time in Fig No significant effect could be observed when measuring Al-hole signals as function of etching time For titanium signals the values obtained following different etching times are consistent within error, although for long etching times (24 h) the baseline becomes distorted and a quantification of the signal intensity as described above could not be performed As titanium and Al-hole signals are extrinsic impurity related defects it is to be expected that these defects have a relative homogenous distribution in the volume of a sedimentary quartz grain On the other hand, our results indicate a pronounced decrease for peroxy signals as well as a reduction in the case of E′1 signals as function of etching time Summary Electron spin resonance investigations were presented for 4–11 μm, 63–90 μm, 90–125 μm, 125–180 μm and 180–250 μm quartz respectively, extracted from loess Titanium ([TiO4/Li+]0) signals are reported in coarse grains and their intensity increases for larger grains This signal could not be detected however in fine (4–11 μm) quartz in any of the samples investigated A decrease in E1′ signal intensity with increasing grain size was observed Aluminium-hole [AlO4]0 signals are present in all investigated samples, however, these signals are interfered during their measurement by a variety of other signals Signals at g ≈ 2.007 and g ≈ 2.004 (tentatively attributed to peroxy) can be detected in all grain sizes, but with significantly higher intensities in the fine grains Extended etching experiments resulted in obtaining partial evidence that these defects are concentrated in damaged areas of the grains A radiation sensitive paramagnetic signal at g ≈ 2.011 was detected only in 4–11 μm quartz As far as implications for electron spin resonance dating are concerned the stronger signature of the [TiO4/Li+]0 signals in larger grains 64 Radiation Measurements 120 (2018) 59–65 A Timar-Gabor coupled with the weaker signals of peroxy signals interfering with [AlO4]0 signal measurement is suggesting that coarser fractions should be preferred for conventional ESR dating using aluminium-hole and titanium signals Further work for understanding the implication of these paramagnetic defects in luminescence dating as well as for unravelling the processes that govern the observed dependence of their intensity as function of grain size is in progress and electrodiffused quartz J Appl Phys 60, 1705–1708 Martini, M., Fasoli, M., Galli, A., 2009 Quartz OSL emission spectra and the role of [AlO4]° recombination centres Radiat Meas 44, 458–461 Martini, M., Fasoli, M., Galli, A., Villa, I., Guibert, P., 2012a Radioluminescence of synthetic quartz related to alkali ions J Lumin 132, 1030–1036 Martini, M., Fasoli, M., Villa, I., Guibert, P., 2012b Radioluminescence of synthetic and natural quartz Radiat Meas 47, 846–850 Martini, M., Fasoli, M., Villa, I., 2014 Defect studies in quartz: composite nature of the blue and UV emissions Nucl Instrum Methods Phys Res Sect B Beam Interact Mater Atoms 327, 15–21 McKeever, S.W.S., 1991 Mechanisms of thermoluminescence production: some problems and a few answers? International journal of radiation applications and instrumentation Part D Nucl Tracks Radiat Meas 18, 5–12 Mondragon, M.A., Chen, C.Y., Halliburton, L.E., 1988 Observation of a dose-rate dependence in the production of point defects in quartz J Appl Phys 63, 4937–4941 Moreno, D., Falguères, C., Pérez-González, A., Duval, M., Voinchet, P., Benito-Calvo, A., Ortega, A., Bahain, J.-J., Sala, R., Carbonell, E., 2012 ESR chronology of alluvial deposits in the Arlanzón Valley (Atapuerca, Spain): contemporaneity with Gran Dolina archaeological site Quat Geochronol 10, 418–423 Murray, A.S., Wintle, A.G., 2003 The single aliquot regenerative dose protocol: potential for improvements in reliability Radiat Meas 37, 377–381 Nilges, M.J., Pan, Y., Mashkovtsev, R.I., 2009 Radiation-induced defects in quartz III Single-crystal EPR, ENDOR and ESEEM study of a peroxy radical Phys Chem Miner 36, 61–73 Pan, Y., Hu, B., 2009 Radiation –induced defects in quartz IV, Thermal properties and implications Phys Chem Miner 36, 421–430 Preusser, F., Chithambo, M.L., Götte, T., Martini, M., Ramseyer, K., Sendezera, E.J., Susino, G.J., Wintle, A.G., 2009 Quartz as a natural luminescence dosimeter Earth Sci Rev 97, 184–214 Rink, J., Odom, 1991 Natural alpha recoil particle radiation and ionizing radiation sensitivities in quartz detected with EPR: implications for geochronometry Nucl Tracks Radiat Meas 18, 163–173 Rink, W.J., Bartoll, J., Schwarcz, H.P., Shane, P., Bar-Yosef, O., 2007 Testing the reliability of ESR dating of optically exposed buried quartz sediments Radiat Meas 42, 1618–1626 Rudra, J.K., Fowler, W.B., 1987 Oxygen vacancy and the E1' center in crystalline SiO2 Phys Rev B 35, 8223–8230 Schilles, T., Poolton, N.R.J., Bulur, E., Bøtter-Jensen, L., Murray, A.S., Smith, G.M., Riedi, P.,C., Wagner, G.A., 2001 A multi-spectroscopic study of luminescence sensitivity changes in natural quartz induced by high temperature anneling J Phys D Appl Phys 34, 722–731 Stapelbroek, M., Griscom, D.L., Friebele, E.J., Sigel, G.H., 1979 Oxygen associated trapped holecenters in high purity fused silicas J Non-Cryst Solids 32, 313–326 Timar-Gabor, A., Wintle, A.G., 2013 On natural and laboratory generated dose response curves for quartz of different grain sizes from Romanian loess Quat Geochronol 18, 34–40 Timar-Gabor, A., Buylaert, J.-P., Guralnik, B., Trandafir-Antohi, O., Constantin, D., Anechitei-Deacu, V., Jain, M., Murray, A.S., Porat, N., Hao, Q., Wintle, A.G., 2017 On the importance of grain size in luminescence dating using quartz Radiat Meas 106, 464–471 Tissoux, H., Valladas, H., Voinchet, P., Reyss, J.L., Mercier, N., Falguères, C., Bahain, J.J., Zöller, L., Antoine, P., 2010 OSL and ESR studies of Aeolian quartz from the Upper Pleistocene loess sequence of Nussloch (Germany) Quat Geochronol 5, 131–136 Toyoda, S., Ikeya, M., 1991 Thermal stabilities of paramagnetic defect and impurity centers in quartz: basis for ESR dating of thermal history Geochem J 25, 437–445 Toyoda, S., Voinchet, P., Falguères, C., Dolo, J.M., Laurent, M., 2000 Bleaching of ESR signals by the sunlight: a laboratory experiment for establishing the ESR dating of sediments Appl Radiat Isot 52, 1357–1362 Toyoda, S., Falguères, C., 2003 The method to represent the ESR signal intensity of the aluminium hole center in quartz for the purpose of dating Adv ESR Appl 20, 7–10 Toyoda, S., 2005 Formation and decay of the E1′ center and its precursor in natural quartz: basics and applications Appl Radiat Isot 62, 325–330 Toyoda, S., Takeuchi, D., Asai, T., Komuro, K., Horikawa, Y., 2005 Spin-spin relaxation times of the E1` center in quartz with and without irradiation: implications for the formation process of the oxygen vacancies in nature Radiat Meas 39, 503–508 Toyoda, S., 2015 Paramagnetic lattice defects in quartz for applications to ESR dating Quat Geochronol 30, 498–505 Tsukamoto, S., Toyoda, S., Tani, A., Oppermann, F., 2015 Single aliquot regenerative dose method for ESR dating using X-ray irradiation and preheat Radiat Meas 81, 9–15 Vaccaro, G., Panzeri, L., Paleari, S., Martini, M., Fasoli, M., 2017 EPR investigation of the role of germanium centers in the production of the 110° C thermoluminescence peak in quartz Quat Geochronol 39, 99–104 Veres, D., Tecsa, V., Gerasimenko, N., Zeeden, C., Constantin, D., Hambach, U., TimarGabor, A, OSL dating constraints for a key Middle-to-Late Pleistocene loess-paleosol sequence in Eastern Europe (Stayky, Ukraine) Quat Sci Rev., submitted for publication Voinchet, P., Yin, G., Falguères, C., Liu, C., Han, F., Sun, X., Bahain, J.J., 2013 ESR dose response of the Al center measured in quartz samples from the Yellow River (China): implications for the dating of Upper Pleistocene sediment Geochronique 40 (4), 341–347 Weil, J.A., 1984 A review of electron spin spectroscopy and its application to the study of paramagnetic defects in crystalline quartz Phys Chem Miner 10, 149–165 Woda, C., Schilles, T., Rise, U., Mangiani, A., Wagner, G.A., 2002 Point defects and the blue emission in fired quartz at high doses: a comparative luminescence and EPR study Radiat Protect Dosim 100, 261–264 Acknowledgement This project received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme ERC-2015-STG (grant agreement No [678106]) An anonymous reviewer is acknowledged for pertinent comments that improved the clarity and quality of the manuscript Mark Bailey (Risø DTU) is highly acknowledged for performing Co-60 irradiations of the samples Vicky Hansen and Jan Pieter-Buylaert helped us by providing the Cal Q sample The author thanks Viorica Tecsa, Valentina Anechitei and Oana Antohi-Trandafir for their help with chemical preparation of samples Robert Begy is acknowledged for technical assistance Adriana Vulpoi-Lazar is thanked for further checking the purity of quartz extracts using SEM Dan Veres, Natalia Gerasimenko and Ulrich Hambach are acknowledged for collecting Sty and Rox samples The author would like to express her gratitude to Professor Simion Simon for very useful scientific discussions as well as his significant administrative support in implementing electron spin resonance measurements in our institute Appendix A Supplementary data Supplementary data related to this article can be found at http://dx doi.org/10.1016/j.radmeas.2018.06.023 References Anechitei-Deacu, V., Timar-Gabor, A., Thomsen, K.J., Buylaert, J-P., Jain, M., Bailey, M., Murray, A Single and multi-grain OSL investigations in the high dose range using coarse quartz Radiat Meas., in this issue, https://doi.org/10.1016/j.radmeas.2018 06.008 Bahain, J.-J., Falguères, C., Laurent, M., Shao, Q., Dolo, J.-M., Garcia, T., Douville, E., Frank, N., Monnier, J.-L., Hallégouët, B., Laforge, M., Huet, B., Auguste, P., Liouville, M., Serre, F., Gagnepain, J., 2012 ESR and ESR/U-series dating study of several middle paleolithic sites of pléneuf-val-andré (brittany, France): piégu, les vallées and nantois Quat Geochronol 10, 424–429 Beerten, K., Lomax, J., Clémer, K., Stesmans, A., Radtke, U., 2006 On the use of Ti centres for estimating burial ages of Pleistocene sedimentary quartz: multiple-grain data from Australia Quat Geochronol 1, 151–158 Botis, S., Nokhrin, S., Pan, Y., Xu, Y., Bonli, T., 2005 Natural radiation induced damage in quartz I Correlations between cathodoluminescencecolors and paramagnetc defects Can Mineral 43, 1565–1580 Duval, M., Guilarte, V., 2015 ESR dosimetry of optically bleached quartz grains extracted from Plio-Quaternary sediment: evaluating some key aspects of the ESR signals associated to the Ti-centers Radiat Meas 78, 28–41 Duval, M., Arnold, L.J., Guilarte, V., Demuro, M., Santonja, M., Perez-Gonzalez, A., 2017 Electron spin resonance dating of optically bleached quartz grains from the Middle Palaeolithic site of Cuesta de la Bajada (Spain) using the multiple centres approach Quat Geochronol 37, 82–96 Friebele, E.J., Griscom, D.,L., Stapelbroek, M., Weeles, R.A., 1979 Fundamental defect centers in glass, the peroxy radical in irradiated, high purity silica Phys Rev Lett 42, 1346–1349 Griscom, D.,L., Friebele, E.J., 1981 Fundamental defect centers in glass 29Si hyperfine structure of the non-bridging oxygen hole center and the peroxy radical in SiO2 Phys Rev B 24, 4896–4898 Hansen, V., Murray, A., Buylaert, J.-P., Yeo, E.-Y., Thomsen, K., 2015 A new irradiated quartz for beta source calibration Radiat Meas 81, 123–127 Ikeya, M., Kohno, H., Toyoda, S., Mizuta, Y., 1992 Spin-Spin relaxation time of E` centers in neutron irradiated quartz (SiO2) and in fault gouge J Appl Phys 31, L1539–L1541 Ikeya, M., 1993 New Applications of Electron Spin Resonance: Dating, Dosimetry and Microscopy World Scientific Publishing, Singapore, pp 520 Liu, C.-R., Yin, G.-M., Han, F., 2015 Effects of grain size on quartz ESR dating of Ti-Li center in fluvial and lacustrine sediments Quat Geochronol 30, 513–518 Malik, D.M., Kohnke, E.E., Sibley, W.A., 1981 Low-temperature thermally stimulated luminescence of high quality quartz J Appl Phys 52, 3600–3605 Martini, M., Spinolo, G., Vedda, A., 1986 Radiation-induced conductivity of as-grown 65 ... investigations with electron spin resonance (ESR) investigations on different grain sizes (4–11 μm, 63–90 μm, 90–125 μm, 125–180 μm, 180–250 μm) of sedimentary quartz A dependence of the most important... dependence of the magnitude of natural signals as function of the grain size for available grain sizes of samples Rox 1.14, Rox 1.9, Sty 1.10 and Sty 2.4 There is no titanium signature in the case of. .. Y., 1992 Spin- Spin relaxation time of E` centers in neutron irradiated quartz (SiO2) and in fault gouge J Appl Phys 31, L1539–L1541 Ikeya, M., 1993 New Applications of Electron Spin Resonance: