2007 GSA Denver Annual Meeting (28–31 October 2007)

Paper No. 9
Presentation Time: 1:30 PM-5:30 PM

LASER-INDUCED BREAKDOWN SPECTROSCOPY OF MINERALS: CARBONATES AND SILICATES


MCMILLAN, Nancy J., Geological Sciences, New Mexico State University, Box 30001, MSC 3AB, Las Cruces, NM 88003, HARMON, Russell S., U.S. Army Research Office, PO Box 12211, Research Triangle Park, Research Triangle Park, NC 27709, DE LUCIA, F., Weapons and Materials Research Directorate, U.S. Army Rsch Lab, Aberdeen Proving Ground, Aberdeen, MD 21005 and MIZIOLEK, Andrzej, Army Research Laboratory, Aberdeen Proving Ground, Aberdeen, MD 21005, russell.harmon@us.army.mil

Laser Induced Breakdown Spectroscopy (LIBS) is an atomic emission spectroscopic technique with the inherent capability to detect, identify, and quantify the chemical composition of any material. Broadband LIBS, in which the full plasma emission spectrum from 200 to 980 nm over which all elements emit is captured with a single laser shot, is particularly attractive for a wide spectrum of geological and environmental applications because of its potential for spatially-resolved, real-time analysis under both bench-top laboratory and field-portable conditions. LIBS provides an alternative chemical analytical technique that obviates the issues of sample preparation and sample destruction common to most laboratory-based analytical methods. The broadband LIBS spectrum contains information about essentially the entire periodic table, although LIBS is more sensitive for the light elements.

This study explores the use of LIBS to rapidly identify carbonate and silicate minerals. Fifty-two mineral samples (18 carbonates, 9 pyroxenes and pyroxenoids, 6 amphiboles, 8 phyllosilicates, and 11 feldspars) were analyzed. Two composite broadband spectra (averages of 10 single laser shots) were calculated for each sample to produce two databases, each containing the composite LIBS spectra for the same 52 mineral samples. By using correlation coefficients resulting from the regression of the intensities of pairs of LIBS spectra, all 52 minerals in the database were correctly identified. If the LIBS spectra of each sample were compared to a database containing the other 51 minerals, 5% were identified as a mineral of similar composition from the same mineral family. The remaining minerals were misidentified for two reasons: (i) the mineral had high concentrations of an element not present in the database; and (ii) the mineral was identified as a mineral with similar elemental composition from a different family. For instance, the Ca-Mg carbonate dolomite was misidentified as the Ca-Mg silicate diopside. This pilot study suggests that LIBS has potential in mineral identification and field-based chemical analysis that could enhance the role of geochemistry in field studies.