Discrimination of Copper Molten Marks through a Fire Reproduction Experiment Using Microstructure Features

Area of Science:

• Materials Science
• Forensic Science
• Metallurgy

Background:

• Copper molten marks are crucial forensic evidence for fire investigation.
• Existing quantitative methods for analyzing these marks are limited to laboratory settings.
• The applicability of these methods to actual fire sites requires validation.

Purpose of the Study:

• To validate the use of four quantitative discriminant factors for copper molten marks in real fire scenarios.
• To develop reliable methods for distinguishing primary and secondary arc beads.
• To enhance the accuracy of fire cause analysis and fire recurrence prevention.

Main Methods:

• A fire reproduction experimental system was established to simulate actual fire conditions.
• Electron backscatter diffraction (EBSD) was employed to analyze the microstructure of molten copper.
• Four discriminant factors were measured: (001) component fraction, grain aspect ratio, Σ3 boundary fraction, and maximum grain size fraction.

Main Results:

• The four discriminant factors exhibited similar characteristics in reproduced fire marks as in laboratory settings, confirming their applicability.
• Discriminant equations and processes were successfully derived to differentiate primary and secondary arc beads.
• A probabilistic discrimination method and a machine learning classification model were developed.

Conclusions:

• The four quantitative factors are applicable to copper molten marks found at actual fire sites.
• The developed discriminant equations and machine learning models improve the reliability of fire investigation.
• This research contributes to preventing fire recurrence by enhancing the accuracy of fire cause determination.