Assessing Matrix and Nonmatrix, Single, and Multipoint Calibration of Trace Elements Using LA‐ICP‐MS on a Tropical Speleothem

ABSTRACT Rationale Suites of trace elements are routinely used in speleothems as proxies to understand periods of past climate change. Laser ablation techniques are regularly implemented to acquire high resolution (50‐μm) trace element concentrations in carbonate archives for paleoclimatology. There exists limited research investigating Laser Ablation‐Inductively Coupled Plasma‐Mass Spectrometry (LA‐ICP‐MS) protocols using speleothem samples. This study investigates the difference between using matrix (carbonate) and nonmatrix (silicate) matched reference materials and the utility of 1‐point versus multiple point calibration curves. Methods Following an extensive review of published literature on speleothem LA‐ICP‐MS analyses, we conducted two laser ablation experimental runs 8 months apart on a 2.7‐cm section of a natural speleothem using matrix and nonmatrix matched reference materials. We used a 193‐nm wavelength Analyte G2 laser attached to a X‐Series‐2 ICP‐MS, a silicate reference material, and three carbonate reference materials. Next, we calculated concentrations using a 1‐point calibration curve, a 2‐point calibration curve, and a 3‐point calibration curve. Results The analysis of matrix and nonmatrix matched reference materials demonstrates that the trends of trace elements/Ca are minimally impacted by the matrix material of the standard. We also show that 2‐ and 3‐point calibration curves bracket the range of sample concentrations compared to a 1‐point (silicate) calibration curve. The calculated cave‐air temperatures using Mg/Ca concentrations fall within error of each other regardless of the calibration curve approach applied. Conclusions Our experiments provide a proof of concept on the conventional setup of standards during LA‐ICP‐MS speleothem analysis. We suggest the use of at minimum a 2‐point (silicate plus carbonate or carbonates) calibration curve that crucially bracket the range of sample concentrations rather than relying on a 1‐point silicate standard that does not bracket the sample concentration. Finally, our results have implications for both speleothem studies that use LA‐ICP‐MS analytical techniques and additional carbonate archives.