Speaker
Description
X-ray microtomography has been established as a fundamental technique for studying porous media across scales ranging from nanometers to centimeters. It is widely used in the routine characterization of materials, particularly rocks, and dynamic processes. However, the complexity of X-ray beam interactions with imaged materials, especially when using polychromatic beams in laboratory applications, introduces significant uncertainties in correlating microtomography data with the sample's compositional features. Such data are often presented in grayscale values without direct physical correspondence to any material property that enables quantitative evaluation.
In this work, we employ three standards—nested cups made of Teflon, aluminum, and quartz—imaged simultaneously with the samples of interest during X-ray microtomography acquisition.
First, the obtained images were used to evaluate and correct beam-hardening effects. Without the use of reference standards, the analysis and correction of such effects are typically performed subjectively, undermining the reproducibility of measurements. In this study, the distribution of attenuation coefficients within the standards is systematically analyzed to identify the optimal filters for mitigating beam-hardening effects in the images. These filters directly impact the calculated attenuation coefficients. The effects of these filters were quantified, and their influence on porosity and effective atomic number determination via the dual-energy technique (see [1]) demonstrated that the objective selection of filters based on reference standards is essential for quantitative applications.
Second, the images from the reference standards were utilized to assess image noise. In laboratory settings, acquisition conditions are often subjectively determined by the microtomography operator. In high-throughput environments, this frequently results in poor-quality images due to the lack of an objective metric to identify the issue. This work demonstrates that reference standard images can be quickly analyzed to quantify image noise, facilitating decision-making for optimal acquisition conditions tailored to specific microtomography applications.
Finally, after systematically acquiring over 500 microtomography images of plugs, we used these images as input for advanced algorithms that directly compute porosity and permeability from the images, as described in [2]. The results show that systematic application of these data significantly improves the accuracy of such algorithms, highlighting the value of incorporating reference standards into quantitative X-ray microtomography workflows.
References
[1] Alves, H., Lima, I., de Assis, J.T., Neves, A.A. e Lopes, R.T.. “Mineralogy evaluation and segmentation using dual-energy microtomography”. In: European X-Ray Spectrometry Conference, Bologna, Italy, 15–20 June 2014.
[2] dos Anjos, C.E.M.N, de Matos, T.F., Avila, M.R.V., Fernandes, J. C. V., Surmas, R.; Evsukoff, A.G. “Permeability estimation on raw micro-CT of carbonate rock samples using deep learning”. Geoenergy Science and Engineering, v. 222, p. 211335, 2023.
| Country | Brazil |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
