• Spectroscopy and Spectral Analysis
  • Vol. 39, Issue 6, 1724 (2019)
WANG Jin-hua1、*, CAO Lan-jie1, XU Guo-qiang2, FENG Xiao-xin3, WU Bing1, and ZHANG Bo1
Author Affiliations
  • 1[in Chinese]
  • 2[in Chinese]
  • 3[in Chinese]
  • show less
    DOI: 10.3964/j.issn.1000-0593(2019)06-1724-07 Cite this Article
    WANG Jin-hua, CAO Lan-jie, XU Guo-qiang, FENG Xiao-xin, WU Bing, ZHANG Bo. Research on Hyper-Spectral Test of Concrete Corrosion Product under Sodium Sulfate Attack[J]. Spectroscopy and Spectral Analysis, 2019, 39(6): 1724 Copy Citation Text show less

    Abstract

    The current testing of concrete chemical etching mainly includes scanning electron microscope, energy spectrum, X-ray diffraction,which has many disadvantages for the concrete in service such as easy to damage, longertest time, poorer continuity. It was proposed that a novel non-destructive testing of concrete corrosion products based on hyper-spectral technology could be used to overcome this shortage. Hyper-spectral test of concrete has many merits such as non-contact, convenient, and continuous lossless, but there are problems in practical application for the unmixed process and analysis of concrete measured spectrum. It was assumed that the spectrum of the concrete under sodium sulfate was a linear mixture of the two component spectra: one component was the spectral reflectivity of the cement hydration product, and another component was the spectral reflectivity of gypsum and ettringite, according toits features of hydration product in concrete, which was proposed the spectral unmixing model of corrosion products of concrete based on the derivative model of the ratio spectroscopy. The hyper-spectral test of concrete was carried out, and the test objects were two concrete sample groups, which were soaked in water and 10%Na2SO4 solution and measured the reflection spectrume very 15 days for 30 to 120 day age of concrete. Themeasured data was smoothed and continuum removed, and then was processedby spectral unmixing model of the corrosion product of the concrete in order to know the features of unmixingspectrum, which was fitted with the standard spectral curve of the mixture of gypsum and ettringite by SFF matching, and observed ettringite and gypsumcorrosion products by SEMand X-ray diffraction in theconcretesamples. the secondary unmixing of the measured spectrum was used the ratio spectrum curve of concrete divided by the gypsumto study ettringiteformation during the 30 to 120 dayscorrosion processing.The result shows that: (1) With the increase of corrosion time, there is obvious difference in the reflectance spectrum curve of two concrete sample groups at different ages in clean water and 10% Na2SO4 solution. Both of them have strong absorption near 1 445 and 1 945 nm, but the absorption depth and absorption area of the spectral curve of 10% Na2SO4 solution concrete are significantly larger than that of clear water. (2) After the once-processed about concrete spectral unmixing model, there are clear absorption valley at around 1 450 and 1 945 nm in the ratio spectrum curve, which is consistent with the standard spectrum of the mixture of gypsum and ettringite, and matching fitting value of the SFF reached 0.96; (3) The spectral characteristics of the corrosion product ettringite are well highlighted by the curve after the ratio spectrum divided by the gypsum, which indicate that the abundance of ettringite produced show down first and then upfor 30 to 120 day age of concrete.(4) The results ofhyper-spectral, SEM, and X-ray diffraction ofconcrete corrosion product undersodium sulfate attackare identical, which indicates that the type and abundance of corrosion products of the concrete can be calculated correctly with the spectral unmixing modelin the non-destructive test, and provides a theoretical basis for the research of hyper-spectral nondestructive testing of concrete.
    WANG Jin-hua, CAO Lan-jie, XU Guo-qiang, FENG Xiao-xin, WU Bing, ZHANG Bo. Research on Hyper-Spectral Test of Concrete Corrosion Product under Sodium Sulfate Attack[J]. Spectroscopy and Spectral Analysis, 2019, 39(6): 1724
    Download Citation