Investigation of ancient wall paintings in Mogao Grottoes at Dunhuang using laser-induced breakdown spectroscopy. (December 2019)

• Record Type: Journal Article
• Authors: Yin, Yaopeng; Sun, Duixiong; Su, Maogen; Yu, Zongren; Su, Bomin; Shui, Biwen; Wu, Chao; Han, Weiwei; Shan, Zhongwei; Dong, Chenzhong
• Is Part Of: Optics & laser technology. Volume 120(2019)
• Publication Date: 2019-12
• Subjects: Mogao Grottoes -- Wall paintings -- Laser-induced breakdown spectroscopy -- Principal component analysis -- Green pigments; Optics -- Periodicals; Lasers -- Periodicals; Electronic journals; 621.366
• DOI: 10.1016/j.optlastec.2019.105689
• Languages: English
• ISSNs: 0030-3992

Abstract

Highlights: The optimized laser pulse energy was fixed as 10 mJ. The classifiable model for green pigments was built and applied successfully. The relationship of ablation depth and number of laser pulses was obtained. Abstract: Laser-induced breakdown spectroscopy (LIBS) is utilized to analyze the ancient wall paintings in the Dunhuang Mogao Grottoes. First, based on a combination of the signal-to-noise ratio (SNR) of the LIBS spectra and the sizes of the ablation craters on simulated samples, a laser ablation threshold is suggested to preventing excessive damage on the ancient wall paintings. Second, coupling with the principal component analysis (PCA) method, LIBS is used to differentiate atacamite and malachite green pigments on simulated sample, the classifiable model is successfully used to differentiate and classify the three unknown green pigments with different tones on a real mural piece, which dates back to the Yuan Dynasty. Third, in combination with a cross-sectional analysis performed via optical microscopy, LIBS is employed to depth profile analysis of green painted layers, a fitting relationship of the ablation depth and the number of laser ablation pulses is obtained. Finally, the thickness of the green painted layers on the mural piece is calculated by means of the relationship, which is close agreement with the result of the cross-sectional analysis.