Plasma characteristics and element analysis of steels from a nuclear power plant based on fiber-optic laser-induced breakdown spectroscopy. (5th November 2018)

• Record Type: Journal Article
• Authors: Wu, Jian; Yu, Han; Qiu, Yan; Zhang, Zhi; Liu, Tao; Xue, Fei; Yu, Weiwei; Li, Xingwen; Qiu, Aici
• Is Part Of: Journal of physics. Volume 52:Number 1(2019)
• Publication Date: 2018-11-05
• Subjects: laser-produced plasma -- laser-induced breakdown spectroscopy -- fiber-optic LIBS system -- steel analysis -- support vector regression -- random forest regression; Physics -- Periodicals; 530
• DOI: 10.1088/1361-6463/aae7b4
• Languages: English
• ISSNs: 0022-3727

Abstract

Abstract: The laser-produced plasma and laser-induced breakdown spectroscopy (LIBS) of two steels from a nuclear power plant, including the Z3CN20-09M steel used for the main reactor coolant pipes, and the 16MND5 steel used for nuclear reactor pressure vessels. These were investigated based on a fiber-optic LIBS system. Under the same input laser energy, the laser-produced plasma from the Z3CN20-09M steel had a weaker emission intensity compared to the 16MND5 steel, probably due to a greater content of Cr. Thus, plasma from the Z3CN20-09M steel had a larger kinetic energy and induced a faster expansion shockwave in air. Calibrations of the fiber-optic LIBS system by the means of internal standardization, support vector regression and random forest regression were set up. For the Cr element, as a major element in Z3CN20-09M (19.33wt.%), it was determined with the relative error controlled under 3.5%; as a trace element (0.11wt.%) in 16MND5 it was measured with the best root-mean-square error of prediction as 0.032wt.%, and the Mn content (1.19–1.51wt.%) in both steels was measured with the relative error around 10%.