Quantagenetics® analysis of laser-induced breakdown spectroscopic data: Rapid and accurate authentication of materials

Highlights: • Quantagenetics® rapidly and accurately authenticates materials using data only from the material itself. • Quantagenetics® uses spectra from Laser-Induced Breakdown Spectroscopy (LIBS) analysis. • Bayesian statistics are applied to the Euclidian distance between LIBS spectra. • 99.4% of spectra from 510 Colombian emeralds from 18 mines were correctly classified. • Quantagenetics® distinguishes composition and structure of steel bars with various heat treatments. Many industrial and commercial issues involve authentication of such matters as the manufacturer or geographic source of a material, and quality control of materials, determining whether specific treatments have been properly applied, or if a material is authentic or fraudulent. Often, multiple analytical techniques and tests are used, resulting in expensive and time-consuming testing procedures. Laser-Induced Breakdown Spectroscopy (LIBS) is a rapid laser ablation spectroscopic analytical method. Each LIBS spectrum contains information about the concentration of every element, some isotopic ratios, and the molecular structure of the material, making it a unique and comprehensive signature of the material. Quantagenetics® is a multivariate statistical method based on Bayesian statistics that uses the Euclidian distance between LIBS spectra of materials to classify materials (US Patents 9,063,085 and 8,699,022). The fundamental idea behind Quantagenetics® is that LIBS spectra contain sufficient information to determine the origin and history of materials. This study presents two case studies that illustrate the method. LIBS spectra from 510 Colombian emeralds from 18 mines were classified by mine. Overall, 99.4% of the spectra were correctly classified; the success rate for individual mines ranges from 98.2% to 100%. Some of the mines are separated by distances as little as 200 m, indicating that the method uses the slight but consistent differences in composition to identify the mine of origin accurately. The second study used bars of 17–4 stainless steel from three manufacturers. Each of the three bars was cut into 90 coupons; 30 of each bar received no further treatment, another 30 from each bar received one tempering and hardening treatment, and the final 30 coupons from each bar received a different heat treatment. Using LIBS spectra taken from the coupons, the Quantagenetics® method classified the 270 coupons both by manufacturer (composition) and heat treatment (structure) with an overall success rate of 95.3%. Individual success rates range from 92.4% to 97.6%. These case studies were successful despite having no preconceived knowledge of the materials; artificial intelligence allows the materials to classify themselves without human intervention or bias. Multivariate analysis of LIBS spectra using the Quantagenetics® method has promise to improve quality control and authentication of a wide variety of materials in industrial enterprises.