Poster Contributed Presentation
ATOM
Bruna Bossard
PhD Student
University of Sao Paulo
São Paulo, Sao Paulo, Brazil
João M. de Lima Júnior
PhD student
University of São Paulo
São Paulo, Sao Paulo, Brazil
Juliana Naozuka, Prof. Dr.
Associate Professor
Federal University of Sao Paulo
Sao Paulo, Sao Paulo, Brazil
Maciel Santos Luz
Researcher
IPT
SAO PAULO, Sao Paulo, Brazil
Cassiana S. Nomura, Prof. Dr.
Associate Professor
University of Sao Paulo
Sao Paulo, Sao Paulo, Brazil
Bruna Bossard
PhD Student
University of Sao Paulo
São Paulo, Sao Paulo, Brazil
Columbite ores are the main source of Nb and Ta widely used in micro-electronics, aerospace, chemical processing and medical devices. The application, however, depends on the amount of those elements and others that can change material properties. Quality control has been done by traditional spectrometric methods such as induced coupled plasma optical emission spectrometry (ICP OES) in which tedious and time-consuming sample preparation processes are required. Laser-Induced Breakdown Spectroscopy (LIBS) emerges as an interesting tool that allows rapid, direct and multi-elemental determination. However, high matrix effect makes quantitative analysis challenging. This research proposes the feasibility of different approaches of calibration for direct quantitative measurement of Nb, Ta and Pb in columbite ores by LIBS: matrix matching calibration (MMC), partial matrix matching multi energetic calibration (PMM-MEC) and slope rate calibration (SRC). The use of carbon as internal standard (IS) to improve accuracy was also checked. Samples and standards were ground and pressed with 30 % m/m of cellulose. For measurements, the following LIBS instrumental parameters were used: spot size = 65 µm, delay time = 0.25 µs, laser energy = 20 mJ, Ar flow rate = 0.5 L min⁻¹ and number of pulses = 500. For MMC approach, 6 references were used as standards and the accuracy obtained were about 91-82 % for Nb and 94-83 % for Ta. For Pb, the use of C as IS was needed. PMM-MEC and SRC approaches are interesting for demanding only 1 standard. For PMM-MEC the accuracy was 84-98 % for Nb, 82-91 % for Ta and 92-97 % for Pb. When IS was used, accuracy increased about 88-98 % for Nb and for Pb. For SRC the accuracy was 85-97% for Nb, 87-100 % for Ta and 87-99 % for Pb. All evaluated calibration approaches were successful, but SRC is simpler since for demanding only one standard and monitoring only one emission line while MEC demands the monitoring of at least 3 emission lines and MMC needs at least 3 standards for calibration.