Effect of laser parameters and tooth type on the ablation of trace metals from mammalian teeth

Publisher Website
Google Scholar
Abstract
This study has shown that the analysis of trace metals in tooth material by laser ablation (LA) ICP-MS is wavelength-, beam energy- and matrix-dependant. Ultraviolet (UV) light offers more promise for laser work with teeth, because it results in Ca-normalized signals for a number of metals which are greater than, or equal to, those obtainable with green light, and with less mass ablated and so less intra-instrument Ca deposition. Two distinct types of ablation processes appear to occur with tooth material. Bulk ablation of metals, indicated by Ca-normalized signals that are generally independant of beam energy, occurs at green light energies of > 2–3 mJ and UV energies of > 3–3.5 mJ. Below these energy levels, selective thermal desoprtion (fractional ablation) of metals is indicated by increasing Ca-normalized signals associated with declining rates of matrix removal. If fractional ablation could be achieved reproducibly, it may have potential for reducing Ca interference effects during the ionization phase of LA–ICP-MS analysis of calcified tissues. Walrus dentine and beluga cement appear to ablate in fundamentally different ways; these matrix effects are probably species-related, because beluga dentine and cement ablate similarly. In beluga teeth, increasing amounts of matrix are removed with increasing UV and green beam energies up to the maximum levels of our laser, while the mass of walrus dentine removed is constant above UV energies of 3.8 mJ and green energies of 7.0 mJ. This suggests the existence of a bulk ablation power density threshold in walrus dentine which, once attained, produces an ablating plasma that is not affected by increasing beam energy. Pulse rate is an important variable, with evidence of greatly reduced fractional ablation and more consistent Ca-normalized signals of metals at a pulse frequency of 5 Hz compared with 10 Hz.

This publication has 0 references indexed in Scilit: