#### NOTE:

The following is a guest post by Graham Hagen-Peter (even though I -Bence- am listed as the author, it is Graham’s work; I am just posting it here). This post is intended to accompany the Sr Isotopes (Universal) DRS, which is available here. Please note that there is additional information about the updated version (v2) at the bottom of this post.

## Introduction and Background

The “Sr Isotopes Universal” DRS provides a number of user-selectable options for reducing Sr isotope data measured by LA-MC-ICP-MS. Since the first Sr isotope measurements by LA-MC-ICP-MS (to our knowledge, Christensen et al., 1995), there have been a number of articles that have outlined and evaluated various potential sources of bias (predominately the multitude of interferences on all Sr isotope masses) in the measurements and proposed various corrections for these sources of bias (e.g., Woodhead et al., 2004; Vroon et al., 2008; Kimura et al., 2014; Müller & Anczkiewicz, 2016). All of these employ “peak-stripping” of interferences (monitoring a nominally interference-free isotope of the interfering element and using canonical isotopic abundances to subtract its interferences on the Sr isotopes of interest) at some point in the correction scheme. Many different atomic, molecular, and doubly charged interferences have been considered (most of which are covered in Vroon et al., 2008, Kimura et al., 2014, and Müller & Anczkiewicz, 2016), and they are matrix-dependent. For example, doubly charged REE (“REE2+”) may yield significant interferences in apatite but be negligible in plagioclase and carbonates. The Sr Isotopes Universal DRS allows users to choose different combinations of interferences to subtract from their data, depending on the expected and/or observed interferences for their samples. There are also several other corrections that can be performed. Users can choose to subtract different combinations of REE2+, CaAr-CaCa, NaNi-CaAlO inteferences; calculate an adjusted Rb mass-bias factor to improved the accuracy of the 87Rb interference correction; correct for Rb-Sr elemental fractionation, which may affect age-correction; and/or correct for residual biases in 87Sr/86Sr following other corrections, which could be attributed to uncorrected 87(CaPO). This document describes the options in the DRS and the results calculated based on the selected options.

Krypton and Rb interferences are always applied in the DRS. The former is corrected thorugh on-peak baseline (which can be fit by one of Iolite’s many spline types) subtraction (see Kimura et al., 2014 for an alternative approach) and the latter from monitoring 85Rb and peak-stripping. Users can additionally choose to subtract REE2+, CaAr-CaCa, NaNi-CaAlO inteferences. The corrections can be selectively applied by selecting or deselecting check-boxes in the DRS settings. If all these (or some combination of these) interferences are assumed to be present and chosen to be subtracted, they are necessarily subtracted that order. For example, doubly charged Er (± Dy) should be subtracted from masses 83 and 82 (in addition to the Sr masses) before carrying out a CaAr-CaCa correction using 83 or 82 as a monitor mass. Nominal mass-bias factors based on 88Sr/86Sr are applied to these interferences. For each step in the correction (e.g., after 172Yb2+, and 176Yb2+ and 176Lu2+ have been subtracted from 86 and 88, respectively), an updated nominal 88Sr/86Sr mass-bias factor is calculated and applied to subsequent corrections. The CaAr-CaCa interferences can be monitored on either mass 82 or 83 (there are isotopologues of these molecules with both these masses). Notice that CaAr and CaCa are considered together. This is because both molecules have isotopologues with mass 82 and 83. Therefore, the user must assume and enter a proportion of the signal on mass 82-83 attributable to CaAr and CaCa in order to subtract these from the Sr masses based on canonical isopologue proportions. In theory, the 83/82 signal ratio could be used to calculate the proportion, but this ratio is likely very imprecise due to low signal intensities. The same combined subtraction applies to NaNi and CaAlO, although these have significant isotopologues only of mass 83 and not 82. Several other settings relevant to these optional corrections are described in the setting-by-setting descriptions below.

Interferences of 87Rb on 87Sr are significant, even in matrices with low to moderate Rb/Sr (e.g., carbonates and plagioclase). Because Rb has only one nominally interference-free isotope (85Rb), a mass-bias factor cannot be directly determined for Rb. It is common practice to apply the Sr mass-bias factor based on 88Sr/86Sr (“ϐSr”) to Rb (and other interference corrections). However, this has been demonstrated to fail at moderate to high Rb/Sr, with a pseudo-linear relationship between 87Sr/86Sr bias and Rb/Sr after 87Rb subtraction (e.g., Jackson & Hart, 2006), which can be attributed to differences in the magnitude of Sr and Rb mass bias. Previous studies (e.g., Jackson & Hart; Zhang et al., 2018) have used glass standards with moderately high Rb/Sr or Rb-doped Sr isotope solutions to calibrate ϐRb or an adjusted 85Rb/87Rb (which are equivalent in effect on the correction). This DRS allows the user to calculate ϐRb using measurements of a reference material with moderately high Rb/Sr and known 87Sr/86Sr (BCR-2G used by default) and solving for ϐRb in the equation below. Obviously, measurements of the reference glass must be included in the sequence of analyses that are being reduced.

$$^{87}Sr/^{86}Sr_{ref} = [ ^{87}(Sr,Rb) – ^{85}Rb * (^{87}Rb/^{85}Rb_{ref}) * (mass_{87Rb}/mass_{85Rb})^{ϐRb}] * \left( \frac {mass_{87Sr}} {mass_{86Sr}}\right)^{ϐSr} * (1/^{86}Sr_{cor})$$

where:

\(^{87}Sr/^{86}Sr_{ref}\) is the Sr isotope ratio of the bracketing standard

\(^{87}(Sr,Rb)\) is the signal on mass 87 with all other interferences subtracted

\(^{85}Rb\) is the raw signal on mass 85

\(^{87}Rb/^{85}Rb\) is a canonical natural Rb isotope ratio

\(^{86}Sr_{cor}\) is the interference-corrected \(^{86}Sr\) signal

Significant elemental fractionation between Sr and Rb has been observed to occur in LA-MC-ICP-MS measurements (Zhang et al., 2018). In samples that are old and/or have high Rb/Sr, accurately determining the 87Rb/86Sr of the sample is critical to accurate age-correction (i.e., calculating the 87Sr/86Sr at the time of isotopic equilibrium —e.g., crystallization— of the sample). Users can use measurements of a reference material with known Rb/Sr (BCR-2G by default) to calculate an elemental-fractionation-corrected 87Rb/86Sr to use in online or offline age-correction. Note that various mineral matrices will not necessarily yield the same Rb-Sr fractionation as the calibration reference material, but the correction utilizes this assumption. Also note that *currently* no downhole-correction for the Rb/Sr is performed by the DRS.

Finally, the DRS allows for the correction of residual bias in 87Sr/86Sr ratios following all other interference corrections and direct normalization to a primary reference material. This may appear to be “fudging” the data, but the approach may be practical in the case of interferences that cannot be independently monitored and subtracted from the Sr masses. A relevant example is the interference of 87(CaPO) — the major isotopologue of a molecule which may obviously be present in apatite analyses and which does not have isotopologues that are practically measurable on other detectors in the array. Therefore, there is no practical way to correct for this interference aside from an empirical secondary correction following other peak-stripping corrections. The correction is done by fitting a linear or exponential model to the data in 87Sr/86Sr bias vs Sr signal space, assuming that the magnitude of an interference from a stoichiometric component —like CaPO in apatite— is inversely proportional to Sr signal (i.e., 1/Sr signal if the CaPO signal is constant).

As a final and obvious note, the user of the DRS must use some intuition in deciding which corrections to apply to their data. Some of the interferences have only briefly been considered in the literature (in-particular the NaNi-CaAlO interferences discussed in Kimura et al., 2014) and may not be relevant to any common matrices. Applying unnecessary corrections may increase measurement uncertainties (e.g., applying the REE2+ corrections if there is no baseline signal on the half-masses) or introduce bias in the corrected 87Sr/86Sr. The corrections that can be applied also depend on the masses measured on the particular detector array. At minimum, the DRS requires 88, 87, 86, 85, and 84 for the basic mass-bias and Rb-interference corrections (perhaps we will eventually remove 84 from this list). Other corrections rely on monitoring interferences on particular masses. If the relevant masses are not measured, the user will get an error message (e.g., if CaAr-CaCa correction is selected but there are no data for the monitor mass 82 or 83).

As a final, final note, this DRS (as is the case with most Iolite DRS) is a work in progress. It is likely that it will be changed (hopefully improved) in the future, perhaps with your input. We will try to accompany all updates with notes.

## DRS Settings

**Index channel**: iolite allows for the possibility that some channels have not been recorded in all files. If this is the case, a channel that is omitted in some mass spec files will have a different number of points to all the other channels, and will need to be interpolated to have the same number of points as the baseline and fractionation splines etc (to perform baseline subtraction etc). The Index channel is just a channel that is measured in all files in the session so that its time array can be used for interpolation. If the channels measured are the same in all files, you can ignore this parameter as it will have little effect. In this DRS, it is set to m88 (the most abundant Sr isotope) by default.

**Reference material**: Primary standard for final normalization of 87Sr/86Sr after all interference and mass-bias corrections have been done

*Mask and related settings*: Enables a signal-threshold criterion to be used for masking data in the Time Series window. The mask is not applied to Input channels but is applied to all Intermediate and Output channels. If the masked values overlap with RM or sample selections, the masked values will not be used in the calculation of the stats for these selections.Masking is most-commonly used to avoid very large scale changes in ratio channels, where the ratio values of the baselines are potentially ± inf, thus making it very hard to view the sample data.

**REE subtraction ?**: If checked, REE2+ will be subtracted from masses 84, 85, 86, 87, and 88 (and 82 and 83 if those channels are present). The corrections rely on signals for the half-masses: 83.5, 86.5, 87.5, so these channels need to be present to carry out the calculations.

**Dy/Er ratio (default is approximately chondritic)**: Doubly charged Er interferes on mass 83, and doubly charged Er and Dy interfere on mass 82, the latter of which can be selectively used for subsequent correction of CaAr-CaCa. Erbium2+ is monitored on “half-mass” 83.5 and can be subtracted from masses 82 and 83, but Dy2+ is not independently monitored on many faraday cup configurations. Therefore, a defined Dy/Er ratio is used to subtract Dy2+ from mass 82. The default (1.5) was very loosely chosen from the ordinary chondrite composition of Nakamura (1974). This should be constrained for individual samples if the REE2+ interferences are significant.

**Scale REE Beta (1 = BetaSr)**: The nominal 88Sr/86Sr mass-bias factor (I say “nominal” because it is calculated before all the interferences are subtracted) can be scaled if the user things the mass-bias factor for the REE is different. In theory, the REE mass-bias could be directly determined by monitoring 173Yb2+ and 171Yb2+, but this would likely be very imprecise for small signals on the half mass. If signals are small, the difference in mass-bias is likely to have a negligible effect anyway. The default (1) is that the REE mass-bias equals nominal Sr mass-bias.

**CaAr-CaCa_subtraction?**: If “yes”, CaAr and/or CaCa interferences will be subtracted from all relevant channels. The corrections rely on signals for either 82 or 83 as a monitor mass, so one of these channels need to be present to carry out the calculations.

**Use 83(CaAr,CaCa) as monitor? (uses 82 as default otherwise)**: If checked, the signal on mass 83 will be used as the monitor mass for the CaAr-CaCa peak-stripping. Otherwise (default), 82 will be used. If 83 is used, this assumes that the signal on mass 83 (the only potential monitor mass for NaNi-CaAlO) is due entirely to CaAr-CaCa, so a subsequent NaNi-CaAlO correction cannot be done. Also note that if mass 82 is used as the monitor following an REE2+ correction, the assumed Dy/Er of the sample (explained above) will affect the residual 82 signal used for the CaAr-CaCa correction, whereas the residual 83 signal depends only on the Er2+ monitored on 83.5.

**Proportion CaAr (0 to 1)**: Because CaAr and CaCa are monitored on the same monitor mass (82 or 83), the signal on this mass may be due to a mixed contribution from each molecule. Therefore, a proportion of the monitor signal from each molecule needs to be assumed to subtract the interferences on the Sr masses using canonical abundances of the isotopologues of the different molecules. A value of 1 (default) assumes that 100% of the signal is due to CaAr. A value of 0 assumes that 100% of the signal is due to CaCa. If a value outside the range of 0 to 1 is entered, you will get an error message. Differences in assumed proportion of these molecules will have the largest effect on the subtractions from masses 85, 86, and 87 because CaCa has significant isotopologues on 85 and 87, whereas CaAr does not, and CaCa has a greater abundance of isotopologues 86 than CaAr. Whether differences in assumed proportions have a significant effect on most datasets is yet to be explored.

**Scale CaAr-CaCa Beta (1 = BetaSr)**: The same as for the REE mass-bias scalar.

**NaNi-CaAlO subtraction?**: If “yes”, NaNi and/or CaAlO interferences will be subtracted from all other signals.

**Proportion NaNi (0 to 1)**: The same principle as for “Proportion CaAr”. A value of 1 (default) assumes that 100% of the signal is due to NaNi. A value of 0 assumes that 100% of the signal is due to CaAlO. If a value outside the range of 0 to 1 is entered, you will get an error message. There are significantly different abundances of the isotopologues of these molecules on the Sr masses, but whether differences in assumed proportions have a significant effect on most datasets is yet to be explored.

**Scale NaNi-CaAlO Beta (1 = BetaSr)**: The same as for the REE mass-bias scalar.

**Rb Beta adjust?**: Gives you the option to carry out the calculation of the Rb mass-bias factor based on a standard (“Reference material Rb beta”). Do not check this if you haven’t measured a standard with moderately high Rb/Sr (e.g., BCR-2G) for this calculation.

**Reference material Rb beta**: Reference material used for calculating an adjusted Rb mass-bias factor using the equation outlined above. The calculation depends on the 85Rb signal, among other things, and so it’s important to have a reference material with relatively high Rb content. BCR-2G is the default and seems to work well.

**Scale Rb beta (if Rb Beta adjust unchecked; 1 = BetaSr)**: If no moderately high Rb/Sr reference material is measured, a scaling factor can be applied to ϐSr to yield an adjusted ϐRb. However, instead of directly calculating ϐRb directly, you need to iteratively try different ϐSr scalars to minimize the bias in 87Sr/86Sr for moderately high Rb/Sr reference materials. If “Rb Beta adjust?” is checked, this scalar will not be used.

**Rb-Sr fractionation?**: Gives you the option to normalize the measured (interference- and mass-bias-corrected) 87Rb/86Sr to an external reference material. This may be important for age-correcting samples that are old and have relatively high Rb/Sr. Keep in mind that minerals and rock glasses may not yield the same Rb-Sr fractionation (i.e., because they are not matrix-matched), yet this correction (if using a rock glass as a reference material) relies on the assumption that they are. We assume that the correction will nonetheless improve the accuracy of the determined Rb/Sr, assuming that the significant interelement fractionation between Rb and Sr in the measurement (albeit tuning-condition-dependent; Zhang et al., 2018) is greater than that induced by matrix effects. This has not been tested though, due to a paucity of mineral standards with uniform, known Rb/Sr.

**Reference material Rb/Sr fractionation**: Reference material for correcting for Rb/Sr elemental fractionation in the measurement.

**Scale Rb/Sr (if no Rb/Sr standard)**: If you haven’t measured a standard with moderately high and well-characterized Rb/Sr, the measured Rb/Sr can be scaled by entering a value other than 1 here. However, if no reference materials with well-characterized Rb/Sr are measured, it will be impossible to evaluate the accuracy of the correction.

**Age (Ma)**: The age of your sample in Ma for calculating age-corrected 87Sr/86Sr (87Sr/86Sri). The DRS uses the “final” 87Rb/86Sr (elemental-fractionation-corrected if selected) and the 87Rb decay constant of Villa et al. (2015). Currently, the DRS uses the present-day 87Sr/86Sr after interference (including Rb with or without an adjusted ϐRb) but before the optional CaPO correction. We will soon fix this. Regardless, it is often more practical to do an age-correction offline on the Iolite output (therefore applying the age correction to the average final 87Sr/86Sr of the selections instead of point-by-point in Iolite), especially if you are measuring multiple samples with different ages.

**CaPO Fit Eqn**: The form of the function (linear or exponential decay) used to fit the data in 87Sr/86Sr bias vs Sr signal space for correction of residual 87Sr/86Sr bias after all other interference and mass-bias corrections. As the name implies, this may be relevant to apatite because 87CaPO cannot not be independently peak-stripped. Because it is a stoichiometric component in apatite, one might expect a relationship between 87Sr/86Sr bias and Sr signal (or 1/Sr signal; proportional to CaPO/Sr signal).

**RMs for CaPO correction**: A dropdown that will populate with all reference materials from your session that have reference 87Sr/86Sr. The reference materials used to establish the relationship (the model fit) between 87Sr/86Sr bias and Sr signal are chosen here. Keep in mind that if all reference materials from a session are used for the correction, they are all in a sense self-normalized, and there are not reference materials left to treat as quality-control or “secondary” standards. If applying this additional correction, it is probably good to measure as many reference materials (apatite in this case) as possible to define the fit as best as possible and to assure that it applies to a wide range of apatite.

**CaPO Correction fit**: Graph that shows the 87Sr/86Sr bias vs Sr signal for individual selections of the reference materials chosen above. Each dot of a particular color represents one selection of a particular reference material selection group.

**Propagate Errors?**: This is the same as other Iolite DRS. It calculates excess variance in selections of a reference material and propagates additional uncertainty into the measurement uncertainties. See the Iolite manual for more information.

## DRS Outputs (Intermediate and Output Channels)

**mXX.X_cps**: Baseline-subtracted signals for each input channel.

**Sr84,…Sr88, Rb85**: Signals for the Sr and Rb isotopes following the all interference corrections.

**Sr87_86_Corr**: Interference- and mass-bias-corrected 87Sr/86Sr calculated by applying ϐSr to Rb but before final normalization of 87Sr/86Sr to a primary reference material. This result is included so the 87Sr/86Sr calculated with and without the adjusted ϐRb can be compared.

**BetaSr**: The final Sr mass-bias factor, calculated from “Sr88” and “Sr86” following all interference corrections. This is the mass-bias factor applied to the final 87Sr/86Sr.

**BetaRb**: The Rb mass-bias factor, either directly calculated from a reference material if “Rb Beta adjust” is selected, scaled from ϐSr if a value other than 1 is entered for “Scale Rb Beta”, or equal to ϐSr if “Rb Beta adjust” is not selected and the scalar is left at 1 (default).

**Sr87_86_CorrRb**: Interference- and mass-bias-corrected 87Sr/86Sr calculated by applying ϐRb to Rb but before final normalization of 87Sr/86Sr to a primary reference material. This will be the same as “87Sr_86_Corr” if “Rb Beta adjust” is not selected and the scalar is left at 1 (i.e., because ϐRb = ϐSr in this case).

**Ho_Sr_ppm**: (Only if “REE subtraction” is selected). Calculates the Ho signal (monitored on half-mass 82.5) relative to the total Sr signal in parts per million (totalHo/totalSr \(\times\) 10^{6}). Holmium is monoisotopic and is not used in any corrections, but can be used as a proxy for REE/Sr content of samples. This could also potentially be used (along with Er_Sr_ppm and Yb_Sr_ppm to calculate REE slopes to constrain Dy/Er).

**Er_Sr_ppm**: (Only if “REE subtraction” is selected). Calculates the total Er signal (monitored on half-mass 83.5 and calculated from canonical Er isotope abundances) relative to the total Sr signal in parts per million (totalEr/totalSr \(\times\) 10^{6}). This can be used as a proxy for REE/Sr content of samples. This could also potentially be used (along with Ho_Sr_ppm and Yb_Sr_ppm to calculate REE slopes to constrain Dy/Er).

**Yb_Sr_ppm**: (Only if “REE subtraction” is selected). Calculates the total Er signal (monitored on half-mass 86.5 and calculated from canonical Yb isotope abundances) relative to the total Sr signal in parts per million (totalYb/totalSr \(\times\) 10^{6}). This can be used as a proxy for REE/Sr content of samples. This could also potentially be used (along with Ho_Sr_ppm and Er_Sr_ppm to calculate REE slopes to constrain Dy/Er).

**StdCorr_Sr87_86**: Interference- and mass-bias-corrected 87Sr/86Sr calculated by applying ϐSr to Rb and normalized to 87Sr/86Sr of a primary reference material. This result is included so the 87Sr/86Sr calculated with and without the adjusted ϐRb can be compared.

**Sr84_86_Corr**: Interference- and mass-bias-corrected 84Sr/86Sr. This stable Sr isotope ratio is sometimes used as an additional check to evaluate the effectiveness of the interference and mass-bias corrections. It is not normalized to a primary reference material, as this partly defeats the purpose of using it as an independent check.

**Sr84_88_Corr**: Interference- and mass-bias-corrected 84Sr/88Sr. This stable Sr isotope ratio is sometimes used as an additional check to evaluate the effectiveness of the interference and mass-bias corrections. It is not normalized to a primary reference material, as this partly defeats the purpose of using it as an independent check.

**Rb87_Sr86_Final**: Final 87Rb/86Sr after interference, mass-bias, and elemental fractionation (if selected) corrections.

**StdCorrRb_Sr87_86**: Interference- and mass-bias-corrected 87Sr/86Sr calculated by applying ϐRb to Rb and normalized to 87Sr/86Sr of a primary reference material. This will be the same as “StdCorr_Sr87_86” if “Rb Beta adjust” is not selected and the scalar is left at 1 (i.e., because ϐRb = ϐSr in this case).

**Sr87_86_AgeCorr**: Age-corrected 87Sr/86Sr. The result is always calculated with “Rb87_Sr86_Final”, “StdCorrRb_Sr87_86”, the 87Rb decay constant of Villa et al. (2015), and the age of the sample in Ma entered by the user in the DRS settings. As of now, the calculation is applied to “StdCorrRb_Sr87_86” (i.e., before the CaPO correction, if selected), but this may be changed in the future. If a user prefers to use a different corrected 87Sr/86Sr result (e.g., StdCorr_Sr87_86) in the calculation, this can be done offline (i.e., with the data exported from iolite). As mentioned above, doing the age correction offline may be more practical in many cases.

**CaPO_corrAmt**: Model bias in 87Sr/86Sr according to the linear or exponential fit to the 87Sr/86Sr bias vs Sr signal (expressed as biased 87Sr/86Sr over “true” 87Sr/86Sr). This is a function of total Sr signal intensity and is the correction factor that is applied the 87Sr/86Sr of the unknown (sample or secondary reference material) measurement.

**CaPOCorr_Sr8786**: Final 87Sr/86Sr after the CaPO correction. The CaPO correction is applied to StdCorrRb_Sr87_86” before age-correction.

## Sr Isotopes Universal DRS version 2.0 updates

Several modifications have been made to the DRS to add slightly more versatility for different cup configurations on MC-ICP-MS:

In version 1.0, it was necessary to monitor 175Lu^{2+} on mass/charge 87.5 to subtract 176Lu^{2+} from mass/charge 88 before the first βSr estimate. 176Lu^{2+} likely contributes a very small amount to the 88 beam, even in apatite, and may not be monitored in all cup configurations. Now, the REE^{2+} correction can be done even if 87.5 is not measured. If 87.5 is measured, the DRS will still do the 176Lu^{2+} correction directly using a canonical 176Lu/175Lu. If 87.5 is not measured, the DRS will calculate 176Lu^{2+} based on the Yb^{2+} monitored at 86.5 and a user-defined Lu/Yb ratio. Like the user-defined Dy/Er in version 1.0, the default is roughly chondritic, but can be changed if you have an idea of the Lu/Yb in your sample. You can set this equal to zero if you think a Lu correction is not needed at all. 173Yb^{2+} monitored at 86.5 and 167Er^{2+} monitored at 83.5 are still both required for the REE^{2+} correction. This is only relevant if and REE^{2+} correction is selected.

In version 1.0, a correction for 164Dy^{2+} on 82 (which can subsequently be used for a CaAr/CaCa correction) was based on the Er^{2+} monitored at mass/charge 83.5 and a user-defined Dy/Er (this was described in the explanation of version 1.0). This is because 81.5 (163Dy^{2+}) was not measured on the cup configurations used when originally creating this DRS. Now, the 164Dy^{2+} can be done directly if mass/charge 81.5 is measured. If 81.5 is measured, the DRS will directly calculate 164Dy^{2+} using a canonical 164Dy/163Dy by default. If it is not measured, 164Dy^{2+} will be calculated from the Er monitor and the user-defined Dy/Er as previously described. This is only relevant if and REE^{2+} correction is selected.

In version 1.0, there was a calculation for the Ho^{2+}/Sr as a proxy for the REE^{2+}/Sr. Holmium is monoisotopic (mass 165) and so does not interfere on any Sr masses as a doubly charged species, so it was only used as a monitor. However, this required that mass 82.5 (165Ho^{2+}) be measured if the REE^{2+} was selected. To increase versatility in cup configuration, this calculation was deleted so the REE^{2+} correction can be done even if 82.5 is not measured. This is only relevant if and REE^{2+} correction is selected.

In version 1.0, there was an error in the mass channel that was defined as 82.5 (used to monitor 165Ho^{2+}). The error has been fixed, but the Ho^{2+}/Sr has been removed, so it doesn’t affect any calculations anyway.

In version 1.0, the variable names Sr84, Sr86, etc. were used for interference-corrected intermediate channels. Raw data files for some instruments (particularly Neptunes, I think) may have the same names, leading to an error when attempting to define the intermediate channels that are already defined as input channels (Sr84, Sr86, etc.). In version 2.0, the interference-corrected intermediate channels are re-named Sr84_corr, Sr86_corr, etc.

## References

Christensen, J. N., Halliday, A. N., Lee, D. C., & Hall, C. M. (1995). In situ Sr isotopic analysis by laser ablation. Earth and Planetary Science Letters, 136(1-2), 79-85.

Müller, W., & Anczkiewicz, R. (2016). Accuracy of laser-ablation (LA)-MC-ICPMS Sr isotope analysis of (bio) apatite–a problem reassessed. Journal of Analytical Atomic Spectrometry, 31(1), 259-269.

Nakamura, N. (1974). Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary chondrites. Geochimica et cosmochimica acta, 38(5), 757-775.

Villa, I. M., De Bièvre, P., Holden, N. E., & Renne, P. R. (2015). IUPAC-IUGS recommendation on the half life of 87Rb. Geochimica et Cosmochimica Acta, 164, 382-385.

Vroon, P. Z., Van Der Wagt, B., Koornneef, J. M., & Davies, G. R. (2008). Problems in obtaining precise and accurate Sr isotope analysis from geological materials using laser ablation MC-ICPMS. Analytical and Bioanalytical Chemistry, 390(2), 465-476.

Woodhead, J., Swearer, S., Hergt, J., & Maas, R. (2005). In situ Sr-isotope analysis of carbonates by LA-MC-ICP-MS: interference corrections, high spatial resolution and an example from otolith studies. Journal of Analytical Atomic Spectrometry, 20(1), 22-27.

Zhang, W., Hu, Z., Liu, Y., Wu, T., Deng, X., Guo, J., & Zhao, H. (2018). Improved in situ Sr isotopic analysis by a 257 nm femtosecond laser in combination with the addition of nitrogen for geological minerals. Chemical Geology, 479, 10-21.