the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Jan 2025
Further investigations into the accuracy of infrared-radiofluorescence (IR-RF) and its inter-comparison with infrared photoluminescence (IRPL) dating
Abstract. Infrared radiofluorescence (IR-RF) is an alternative dating technique for potassium feldspar grains, offering a higher signal stability and based on a simpler underlying mechanism than more common luminescence dating approaches. However, its accuracy when tested on known-age samples has so far shown inconsistent results. In this study, we present a refined accuracy assessment using samples that have previously produced unreliable IR-RF ages. Our approach incorporates two major methodological advancements developed over the past decade: elevated temperature measurements using the IR-RF70 protocol and sensitivity change correction by vertical sliding. To expand the dose range comparison, we included two additional samples: one expected to be in saturation and another of modern age. Additionally, we evaluated the effect of using a narrower bandpass filter to exclude any signal contributions from potentially contaminating shorter wavelength emissions. Our results following the IR-RF70 protocol with sensitivity corrections show an improvement over the original room temperature results. For four out of the seven tested known-age samples spanning 20–130 ka, we obtained results in keeping with the expected doses. Two additional modern samples, however, yielded slight age underestimations. Introduction of a multiple-aliquot regenerative dose (MAR) protocol improved the accuracy of two out of three samples with large sensitivity changes. Finally, we also compared the new IR-RF equivalent doses (De) to those obtained with the newer trap-specific dating method infrared-photoluminescence (IRPL) for the same samples, including previously published values and new measurements. We observe that with the new improvements the success rate of IR-RF is comparable to that of IRPL.
- Preprint
(1597 KB) - Metadata XML
-
Supplement
(2228 KB) - BibTeX
- EndNote
Status: open (until 26 Feb 2025)
-
RC1: 'Comment on gchron-2024-36', Anonymous Referee #1, 20 Jan 2025
reply
Summary
Over the past two decades, two groundbreaking luminescence new dating techniques have emerged: infrared radiofluorescence (IR-RF) on feldspar specimens and infrared photoluminescence (IR-PL). Although IR-RF is the older method, it has only recently gained more adoption. Since the introduction of IR-PL, there has been
ongoing debate regarding its superiority over IR-RF. The primary advantage of both techniques lies in their ability to measure the charge density of a principal trap. This measurement provides a more direct estimate of the accumulated dose.The manuscript presents in essence a long-overdue systematic comparison of IR-RF and IR-PL that should have been done long ago on a set of samples. The present contribution goes even beyond as it also re-analysed older results from a study that has questioned the reliability of IR-RF.
The approach is systematic, and the manuscript is carefully prepared. It takes a somewhat neutral standpoint, although most of the tests are clearly related to IR-RF. Nevertheless, what is refreshing is that it does not claim superiority of one method over the other but tries to point out differences and challenges where applicable.
The manuscript clearly aligns with the scope of Geochronology and should be published.
I have only a few general remarks and more minor technical comments, but I am confident
that they can be addressed by the study authors easily.General remarks
- The manuscript is generally well-structured and well-written, with most sections being easy to read. However, I had the impression that the authors added more experiments along the way and somewhat forgot the original purpose of the manuscript. While this is common, I suggest that the authors revise the introduction to make it easier for readers by clearly marking hypotheses that can be quantified and tested. For instance, the title suggests a straightforward comparison of IR-RF and IRPL, but the study then presents a diverse range of tests (including pIRIR measurements). In other words, the study lacks some rigour
and could benefit from a little streamlining. - The MAR test is intriguing, but it comes unexpectedly and lacks further substantiation. This presents a missed opportunity. The authors should either extend this section or remove it, as it will not significantly alter the outcome but blur the story. By doing so, they would also have the opportunity to design more robust experiments and incorporate modelling.
- The same goes for the discussion about the detection window. This is given quite some attention, but at the end, it seems to have no effect. If so, this can be shortened to two or three lines and with the rest of the data in the supplement.
- The authors provide access to raw data, which is highly appreciated; however, they should also add full information on the source rate calibration (see below). Otherwise, the results are of limited use to others.
- The definition of uncertainties appears somewhat ambiguous, often I am not sure whether they are truly comparable. For instances, sometimes it seems to be the standard deviation, sometimes the standard error of the mean, for the independent age control something else(?).
Minor comments
- L37: Because you explain basics, you should also addd a suitable reference for IRSL
- L38: The reference to Krbetschek et al. (2000) seems incorrect. The authors wrote "Fading tests (storage over periods of several months at room temperature) have shown signal stability." Krbetschek et al. (2000, p. 497). They further stated: "Further investigation is necessary to ascertain what this tells us about the mean life of the trap population" (2000, p. 497). They did not write anything about "lower anomalous fading rates". They wrote about signal stability and the mean life(time) of the trap population. This implies that they meant the thermal not the athermal lifetime.
- L46: Technically, the sliding approach goes back to Prescott et al. (1993) (or even earlier) under the name "Australian slide". Buylaert et al. (2012) describe horizontal sliding in their Fig. 4; the method/tool is described in Lapp et al. (2012) where they describe a time-shift. Kreutzer et al. (2017) (see also in Murari et al. 2018) first used the approach; Murari et al. (2018) formally introduced it. However, Buylaert et al. (2012) indeed mention horizontal and vertical adjustments, but it is unclear what their conclusion was and why they did not test it. But I agree, credit should be given to them because they mention the idea.
- L86-L95: This paragraph is very muddled. You start re-analysing 16 samples but present 10 new IR-RF and then again "eight samples originally used". Please rephrase to improve readability or make a list for your experiments or a workflow graph.
- L86-L95: The introduction should explicitly state a research hypothesis that will be tested in the contribution, rather than presenting a list of experiments that may or may not yield a specific outcome; some of them even unrelated to the study title.
- L102-L103: The HF etching of feldspars is challenging, and still it is unlikely to remove any other alpha-irradiated layer uniformly and in the desired manner (Duller, 1992; Porat et al., 2015; Duval et al., 2018). Given that sample preparation cannot be altered retrospectively, I propose discussing this issue and its potential impact on the final results later in the manuscript.
- L116-L125: Please provide detailed information on the calibration of the other machines, including the aliquot size, sample carrier, dose rate, and calibration date. This information is essential for cross-checking your results by others without the need for additional inquiries. If necessary, please indicate whether you had to correct calibrations based on previous measurements (as described in Autzen et al., 2022). I attempted to recalculate a few results using the data provided on Zenodo. However, without the dose rate (available for some samples, though), I am unable to effectively compare the results.
- L136: Please state the number of ignored channels, or the dose (you do that later). I tried to recalculate, for instance, A8. But my result is consistently 0, however, I can get any result (also the one you report) by ignoring a certain number of channels. Means, this information matters.
- L139: A more detailed description with of the initial rise can be found in Frouin et al. (2017) (their supplement). From this analysis, it becomes evident that the response appears to be dose-dependent and exhibits a range of responses within a given dose range.
- L163: What is the justification for the double-exponential fit? Wouldn't the GOK model (Guralnik et al., 20215) be a better candidate for feldspar?
- L176: Can you show such a distribution?
- L215: I think that the number of channels matters more than the dose; please also check the supplement by Frouin et al. (2017) where this investigated (although it seems only with horizontal sliding).
- L430: I concur with this conjecture, albeit with a slight distinction. I think that the relationship lies not solely with the dose but also with the number of channels. Your objective should be to identify a plateau of equivalent dose values rather than distinct segments. To achieve this, you can segment your natural dose and incrementally add channels to the RF natural dose until a plateau is detected. While this approach addresses the issue of channel-related variations, it still presents a challenge: if the regenerated and natural curves indeed differ, the results may be inaccurate when compared to an independent age control.
Nevertheless, this approach eliminates the possibility of arbitrary channel selection. - L440-L443: They yield 0 Gy because the algorithm has no other choice to match the curves given the shapes and the starting points and then sets it to 0. This is not coincidental; this is by design. See your own arguments a few lines below.
- L437: This somewhat contradicts your chain of arguments trying to emphasise good arguments and put more weight on one or the other. The 35 Gy is an arbitrary choice and sample dependent; it seems a good fit for your samples, but I suggest refraining from generalising this observation. The best approach seems to reject the very first channel and keep the rest (with a certain number of minimum channels)
- L461-L464: In Murari et al. (2021) all measurements (Risoe and Freiberg readers) seem to have been used 70ºC as recording temperature; please rephrase or remove.
- L465-L470: Agreed, but you should also point to the different protocols with no less than 17 to 18 steps. I am wondering how sensitive the equivalent dose is to certain parameters. If you cannot test this, you should at least discuss it.
- L483-L484: I do not believe that the comparison to the quartz model is valid. While the observation may share some similarities, the underlying mechanism is unknown and likely distinct in quartz. Unless you can provide a model and demonstrate that the mechanism is indeed similar, I recommend removing this speculative comparison. The subsequent comparison is more appropriate and logical, although it is purely descriptive.
Figures
- Figure : The figure I am missing is a distribution plot for equivalent doses. Perhaps this can be added for suitable samples.
- Figure 1: Please add information on the aliquot number and which measurement window was used. Also, here contrary to what was written in the M&M section no initial channel was discarded.
- Figure 2: Please colour-code the samples and use shapes to denote the methods. This will prevent readers from having to guess which sample is shown. If you run out of distinguishable colours, please use labels. For the 600 Gy exposure, the quantity of channels is the primary parameter of concern, rather than the dose (the information remains beneficial regardless).
- Figure 3: What is the central new information conveyed by these figures? The sliding method, particularly requires offsetting for short segments and less curvature. Please condense to a single key figure with a succinct message.
- Figure 5: To compare IR-RF and IRPL, it is necessary to include a third figure that compares both techniques with the method you believe performs most effectively. Additionally, you should compare the distribution of the relative residuals from the 1:1 regression line to assess whether there is a significant difference between the two methods or if they are merely random.
- Figure 6: Do the grey bands make sense? The optimal range would be with the highest number of channels. However, if you wish to retain the current settings, you should also experiment with different integration values for the other methods. But I guess then it becomes very confusing.
- Figure 9: It requires an illustration of the separated dose signal components. Currently, it appears a little bit arbitrary and descriptive.
- Figure S4: The offsets of the curves are a little bit difficult to see, perhaps you can show the residuals?
- Figure S9: What does this figure add to the manuscript? Your concern is the comparison of two methods, here you compare all kind of protocols and procedures on top of two types of IR-RF and IRPL. I can somewhat understand your Fig. 6 in the main text, but this seems too much.
Tables
- Table 1: Instead of 'se' that refers to the standard error (of what?), please use confidence intervals.
- Table 3: What do the uncertainties represent? The standard deviation? For such a low 'n' you should rather calculate confidence intervals using
the t-distribution unless you can show that the normal distribution approximation is valid. - Table 4: Please explain the meaning of the uncertainties and align them. I suggest calculating consistently 95% confidence intervals.
References
Autzen, M., Andersen, C.E., Bailey, M., Murray, A.S., 2022. Calibration quartz: An update on dose calculations for luminescence dating. Radiation Measurements 106828. [https://doi.org/10.1016/j.radmeas.2022.106828](https://doi.org/10.1016/j.radmeas.2022.106828)Buylaert, J.P., Jain, M., Murray, A.S., Thomsen, K.J., Lapp, T., 2012. IR-RF dating of sand-sized K-feldspar extracts: A test of accuracy. Radiation Measurements 47, 759–765. [https://doi.org/10.1016/j.radmeas.2012.06.021](https://doi.org/10.1016/j.radmeas.2012.06.021)
Duller, G.A.T., 1992. Luminescence chronology of raised marine terraces, south-west north island, New Zealand (PhD thesis). University of Wales, Aberystwyth.
Duval, M., Guilarte, V., a, I.C. n, Arnold, L.J., Miguens, L., Iglesias, J., lez-Sierra, S.G. a, 2018. Quantifying hydrofluoric acid etching of quartz and feldspar coarse grains based on weight loss estimates: implication for ESR and luminescence dating studies. Ancient TL 36, 1–14. [https://doi.org/10.26034/la.atl.2018.522](https://doi.org/10.26034/la.atl.2018.522)
Erfurt, G., Krbetschek, M.R., 2003. IRSAR - A single-aliquot regenerative-dose dating protocol applied to the infrared radiofluorescence (IR-RF) of coarse-grain K-feldspar. Ancient TL 21, 35–42. [https://doi.org/10.26034/la.atl.2003.358](https://doi.org/10.26034/la.atl.2003.358)
Frouin, M., Huot, S., Kreutzer, S., Lahaye, C., Lamothe, M., Philippe, A., Mercier, N., 2017. An improved radiofluorescence single-aliquot regenerative dose protocol for K-feldspars. Quaternary Geochronology 38, 13–24. [https://doi.org/10.1016/j.quageo.2016.11.004](https://doi.org/10.1016/j.quageo.2016.11.004)
Guralnik, B., Jain, M., Herman, F., Ankjærgaard, C., Murray, A.S., Valla, P.G., Preusser, F., King, G.E., Chen, R., Lowick, S.E., Kook, M., Rhodes, E.J., 2015. OSL-thermochronometry of feldspar from the KTB borehole, Germany. Earth and Planetary Science Letters 423, 232–243. [https://doi.org/10.1016/j.epsl.2015.04.032](https://doi.org/10.1016/j.epsl.2015.04.032)
Krbetschek, M.R., Trautmann, T., Dietrich, A., Stolz, W., 2000. Radioluminescence dating of sediments: methodological aspects. Radiation Measurements 32, 493–498. [https://doi.org/10.1016/s1350-4487(00)00122-0](https://doi.org/10.1016/s1350-4487(00)00122-0)
Kreutzer, S., Murari, M.K., Frouin, M., Fuchs, M., Mercier, N., 2017. Always remain suspicious: a case study on tracking down a technical artefact while measuring IR-RF. Ancient TL 35, 20–30. https://doi.org/10.26034/la.atl.2017.510
Lapp, T., Jain, M., Thomsen, K.J., Murray, A.S., Buylaert, J.P., 2012. New luminescence measurement facilities in retrospective dosimetry. Radiation Measurements 47, 803–808. [https://doi.org/10.1016/j.radmeas.2012.02.006](https://doi.org/10.1016/j.radmeas.2012.02.006)
Murari, M.K., Kreutzer, S., Fuchs, M., 2018. Further investigations on IR-RF: Dose recovery and correction. Radiation Measurements 120, 110–119. [https://doi.org/10.1016/j.radmeas.2018.04.017](https://doi.org/10.1016/j.radmeas.2018.04.017)
Murari, M.K., Kreutzer, S., Frouin, M., Friedrich, J., Lauer, T., Klasen, N., Schmidt, C., Tsukamoto, S., Richter, D., Mercier, N., Fuchs, M., 2021. Infrared Radiofluorescence (IR-RF) of K-Feldspar: An Interlaboratory Comparison. Geochronometria 48, 95–110. [https://doi.org/10.2478/geochr-2021-0007](https://doi.org/10.2478/geochr-2021-0007)
Peng, J., Wang, X., Adamiec, G., 2022. The build-up of the laboratory-generated dose-response curve and underestimation of equivalent dose for quartz OSL in the high dose region: A critical modelling study. Quaternary Geochronology 67, 101231. [https://doi.org/10.1016/j.quageo.2021.101231](https://doi.org/10.1016/j.quageo.2021.101231)
Prescott, J.R., Huntley, D.J., Hutton, J.T., 1993. Estimation of equivalent dose in thermoluminescence dating - the Australian slide method. Ancient TL 11, 1–5. [https://doi.org/10.26034/la.atl.1993.204](https://doi.org/10.26034/la.atl.1993.204)
Porat, N., Faerstein, G., Medialdea, A., Murray, A.S., 2015. Re-examination of common extraction and purification methods of quartz and feldspar for luminescence dating. Ancient TL 33, 22–30. [https://doi.org/10.26034/la.atl.2015.487](https://doi.org/10.26034/la.atl.2015.487)
Citation: https://doi.org/10.5194/gchron-2024-36-RC1 - The manuscript is generally well-structured and well-written, with most sections being easy to read. However, I had the impression that the authors added more experiments along the way and somewhat forgot the original purpose of the manuscript. While this is common, I suggest that the authors revise the introduction to make it easier for readers by clearly marking hypotheses that can be quantified and tested. For instance, the title suggests a straightforward comparison of IR-RF and IRPL, but the study then presents a diverse range of tests (including pIRIR measurements). In other words, the study lacks some rigour
Data sets
Further investigations into the accuracy of infrared-radiofluorescence (IR-RF) and its inter-comparison with infrared photoluminescence (IRPL) dating (v1.0.0) [Data set] M. Sontag-González et al. https://doi.org/10.5281/zenodo.14507179
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 105 | 16 | 4 | 125 | 13 | 1 | 2 |
- HTML: 105
- PDF: 16
- XML: 4
- Total: 125
- Supplement: 13
- BibTeX: 1
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1