Fig 15 - uploaded by Sanja Vranjes-Wessely
Content may be subject to copyright.
(a) SPM surface scan of a vitrinite particle in sample SK10 (map 1). The nanoindents are color-coded according to the k-means clustering results shown in (b). (c) Load displacement curves for the indents highlighted by black circles in (a).
Source publication
Nanoindentation is a valuable tool, which enables insights into the material properties of natural, highly inhomogeneous composite materials such as shales and organic matter-rich rocks. However, the inherent complexity of these rocks and its constituents complicates the extraction of representative material parameters such as the reduced elastic m...
Citations
... In microscale testing, particle size may sometimes be overlooked. However, even when examining sufficiently large particles, it can be challenging to fully eliminate the influence of underlying substances on the target particle (Vranjes-Wessely et al., 2021;Yang et al., 2023). Also, due to the characteristics of the nanoscale methods, the calculated modulus may not be affected by all the components of the sample . ...
Geomechanical properties of rocks are essential for understanding their elastic behavior. These parameters find their applications in various fields such as petroleum engineering, geological storage sites, of nuclear waste and carbon dioxide and other geotechnical operations. There are several methods to obtain these parameters, including experimental methods, mathematical upscaling, and numerical simulation while experimental approaches are more trusted. The aim of this study is to estimate Young's modulus of a shale sample from several major upscaling mathematical methods and compare the results to polyaxial compressive strength test and nanoindentation measurements on the same piece of sample. To achieve this, five major theoretical upscaling models, including DEM, MT, SCA, KTF, and DM, have been utilized to calculate the Young's modulus based on constituent components of the sample that was obtained from routine XRD and geochemical analysis. The calculations for each of the five methods were performed in two scenarios: without incorporating the organic matter in the model and with it, in a range of porosity values while input parameters for each constituent component were found from the literature. Young's modulus with organic materials following the upscaling schemes was in the range of 24.0661 [GPa] to 27.0001 [GPa]. Likewise, excluding the organic substances Young's modulus was calculated 25.1784 [GPa] and 29.1394 [GPa], as the smallest and largest values, respectively. These values were compared to the Young's modulus obtained from the nanoindentation (34.15 GPa) and polyaxial test (24.36 GPa). The Young's modulus decreases of approximately 7% with considering organic matter around 2.45% and decrease of less than 5% with variation of 1% porosity. Based on the results, it was concluded that upscaling methods overall provide results in an acceptable range while they are more cost-effective and time-efficient when compared to expensive and time-consuming experimental approaches and can replace them. Furthermore, the use of upscaling methods reduces the need for destructive testing and also eliminates the requirement for expensive equipment with limited access to the proper size of the samples and required preparations that should be followed. However, the lithology of the sample and the presence of certain types of pore spaces and fractures should not be taken lightly in this process.
... As a result, micro-and nano-mechanical characterization techniques, such as nanoindentation and atomic force microscopy (AFM), are widely used in materials science because they require small sample volumes only. They have been gradually introduced to measure mechanical data of organic matter in shale (Ahmadov et al., 2009;Eliyahu et al., 2015;Fender et al., 2020;Kumar et al., 2012;Li et al., 2018bLi et al., , 2018cShukla et al., 2013;Tan et al., 2020;Vranjes-Wessely et al., 2021;Zeszotarski et al., 2004;Zhang et al., 2022). ...
... For example, it has been reported that the elastic modulus of kerogen in the Bakken shale decreases with maturity from immature to mature stages (Zargari et al., 2016), while latest studies show the opposite results for the organic matter from the Bakken shale (Khatibi et al., 2018;Li et al., 2018b). Some works show even more complex dependence where the elastic modulus of organic matter generally decreases with increasing maturity (from 1.33 to 1.96% vitrinite reflectance, R o ), and then increases slightly between 2.00 and 2.23% R o from the Lower Cretaceous Shahezi Formation rock in the Songliao Basin, China (Vranjes-Wessely et al., 2021). For the different kinds of organic matter in the same shale, it is reported that the elastic modulus of kerogen is larger than that of bitumen (Vranjes-Wessely et al., 2021;Zargari et al., 2013;Zhao et al., 2020). ...
... Some works show even more complex dependence where the elastic modulus of organic matter generally decreases with increasing maturity (from 1.33 to 1.96% vitrinite reflectance, R o ), and then increases slightly between 2.00 and 2.23% R o from the Lower Cretaceous Shahezi Formation rock in the Songliao Basin, China (Vranjes-Wessely et al., 2021). For the different kinds of organic matter in the same shale, it is reported that the elastic modulus of kerogen is larger than that of bitumen (Vranjes-Wessely et al., 2021;Zargari et al., 2013;Zhao et al., 2020). At the same maturity level (R o = 1.10%), the stiffness of organic matter of lacustrine shales in Shahejie Formation, Bohai Bay Basin, China follows: fusinite > inertinite (macrinite) > micrinite > vitrinite (Gao et al., 2023). ...
The quantification of mechanical properties of organic matter in shale is of significance for the fine prediction and characterization of shale reservoir’s mechanical properties. Due to the micron-sized and dispersed distribution of organic matter particles in shale, the accurate evaluation of the actual mechanical response remains challenging. This work focuses on shale from Wufeng-Longmaxi Formation, which is the main shale gas exploration and development formation in China. A method based on atomic force microscopy (AFM) with an optical microscope (i.e., in-situ AFM technique) is presented to locate the organic matter in-situ and then visualize and quantify its mechanical properties using AFM Young’s modulus mapping. The merits and limitations for determining the mechanical properties of organic matter in shale between the AFM and the more conventional nanoindentation technique are discussed. Results show that combining in-situ nanoindentation and in-situ AFM mapping provides more accurate description of the mechanical properties of organic matter in shale than traditional grid indentation methods with low spatial resolution. The Young’s moduli of organic matter calculated from nanoindentation are around twice smaller than those obtained from AFM measurements mainly because the elasto-plastic deformation zone of organic matter in nanoindentation tests is larger and can be additionally affected by the presence of inorganic particles and/or larger micro-pores in organic matter. The Young’s modulus and hardness of graptolite in the shale obtained by nanoindentation are slightly larger than those of solid bitumen at the same thermal maturity. Both in-situ AFM and in-situ nanoindentation results show that the mechanical strength of organic matter increases with increasing maturity. Overall, the presented approach shows a great potential for accurate and in-situ measurement of the mechanical properties of organic
matter in shale at the nanoscale, which may be beneficial to the development of rock mechanical models for the accurate evaluation of the actual mechanical properties of shale.
... As further proof of the postulate that considers, in this review, highspeed nanoindentation as the only suitable probing technique for the intrinsic heterogeneities in minerals, and natural materials in general, with the solid intention for designing energy carriers materials and nuclear waste disposal technologies, Varnjes-Wessely et al. [52] employed HSNM to study the properties of natural composite materials such as shales and organic matter-rich rocks. This work aimed to find alternatives to standard macroscopic measurements, which provided low precision and inconsistent results without any capability to map the individual component. ...
... Advanced blind and «Unsupervised» high-speed mapping data analysis: a) 2-D Gaussian interpolation for Al-Cu intermetallics (reprinted and adapted from Xiao et al.[47], Copyright(2020), with permission from Elsevier) and b) k-means clustering applied to thermal spray coatings by Vignesh et al. (reprinted from Vignesh, B. et al.[49]); c) unsupervised k-means clustering for organic matter-rich rocks with identification 1:1 correlation of load-depth behavior (reprinted from Vranjes-Wessely, S. et al.[52]) and d) comparison between phase-identification efficiency on Al-Cu eutectic alloy of classical blind statistical analysis and advanced deconvolution techniques as a function of indent spacing (reprinted and adapted from Besharatloo, H. et al.[23]). ...
High-Speed Nanoindentation Mapping (HSNM) has been recently developed and established as a novel enabling technology for fast and reliable assessment of small-scale mechanical properties of heterogeneous materials over large areas. HSNM allows for one complete indentation cycle per second, including approach, contact detection, load, unload, and movement to the nth indent location, thus enabling high-resolution, spatially resolved hardness (H) and elastic modulus (E) mapping.
This article reviews the recent advancements in HSNM and its application to support the design, synthesis, and characterization of advanced materials, potentially impacting the ongoing digital and green transitions. A comprehensive review is given of (a) the main experimental features and critical issues of the protocols in comparison with traditional quasi-static nanoindentation, (b) the advanced data analysis tools employed, and (c) the combination with other microscopy and spectroscopy methods for multi-technique correlative applications. Finally, the relevance of HSNM for selected classes of materials is discussed, including (i) additively manufactured metals, (ii) advanced alloys, (iii) composite materials and cement, highlighting the potential for matrix-reinforcement mechanical characterization and optimization routes, (iv) coatings for industrial components and energy/transportation, discussing damage progression identification at the micro-structural level, and (v) natural materials. Ultimately, future perspectives are presented and discussed.
... On a similar category of research topics, a practical correlative approach is represented by the work of Vranjes-Wessely et al. [73], on the extraction of representative mechanical properties by nanoindentation in highly inhomogeneous composite materials, such as shales and organic matter-rich rocks, by combining high-speed nanoindentation mapping, comprehensive optical petrography, and high resolution-imaging methods, including scanning electron microscopy and helium ion microscopy, as well as ML algorithms. ...
... All articles reviewed in this section [71][72][73][74][75][76][77][78] clearly show the two main advantages of the use of ML tools, in comparison with conventional statistics, since (a) ML are flexible and adaptive, helping to identify hindered effects in the data (like hidden mechanical phases, non-linear relationships between variables, phases with non-normal statistical distribution), (b) give the information on the spatial distribution of the mechanical phases (information that is not given by conventional statistical deconvolution), (c) offer the potential for automation and scalability: once a machine learning model is trained on a set of nanoindentation data, it can be applied to new datasets with minimal human intervention, making the analysis process more efficient and less time-consuming. Moreover, as more data becomes available, machine learning models can be easily retrained and updated to incorporate the latest information, ensuring continuous improvement in analysis accuracy. ...
The solution of instrumented indentation inverse problems by physically-based models still represents a complex challenge yet to be solved in metallurgy and materials science. In recent years, Machine Learning (ML) tools have emerged as a feasible and more efficient alternative to extract complex microstructure-property correlations from instrumented indentation data in advanced materials. On this basis, the main objective of this review article is to summarize the extent to which different ML tools have been recently employed in the analysis of both numerical and experimental data obtained by instrumented indentation testing, either using spherical or sharp indenters, particularly by nanoindentation. Also, the impact of using ML could have in better understanding the microstructure-mechanical properties-performance relationships of a wide range of materials tested at this length scale has been addressed.
The analysis of the recent literature indicates that a combination of advanced nanomechanical/microstructural characterization with finite element simulation and different ML algorithms constitutes a powerful tool to bring ground-breaking innovation in materials science. These research means can be employed not only for extracting mechanical properties of both homogeneous and heterogeneous materials at multiple length scales, but also could assist in understanding how these properties change with the compositional and microstructural in-service modifications. Furthermore, they can be used for design and synthesis of novel multi-phase materials.
... The sample (~1629 m; quartz 31 wt.%, clay mineral 39 wt.%) was polished using a Hitachi ArBlade 5000 broad ion beam (BIB) system and imaged using a Tescan Clara field emission scanning electron microscope (SEM) before and after the nanoindentation. Nanoindentation property mapping was performed using a Hysitron TS 77 Select Nanoindenter in load-controlled mode (see also Vranjes-Wessely et al., 2021). A total of eight array maps (7 × 7 indents, 6 μm spacing) were indented, covering both grain and matrix areas (Fig. 1). ...
Mudstones and shales are fine-grained sedimentary rocks that can serve as top seals of geological reservoirs in various geoenergy applications. Apart from traditional oil and gas exploration, the urgent need for underground storage of energy carriers (e.g., H2) and climate-relevant gases (e.g., CO2) facilitated extensive research on pore structural and mechanical parameters and their influence on the seal capacity of these rocks. In this contribution, high-speed nanoindentation mapping was combined with machine learning data analysis as a feasible high throughput tool for the mechanical characterization of mudstones and similar fine-grained sedimentary rocks. The presented approach is planned to be applied to an extensive set of mudstone samples from the Vienna Basin with the purpose to link mechanical property changes to burial diagenesis.
... Shale oil and gas are unconventional hydrocarbon resources and constitute an important focus for modern petroleum exploration (Curtis, 2002). With their rapid development all over the world, the characterization of organic matter (OM) mechanics in shale has gradually attracted attentions from scholars in shale geophysics (Zargari et al., 2013(Zargari et al., , 2016, mechanics (Kumar et al., 2012a;Hull et al., 2015;Alstadt et al., 2016;Mashhadian et al., 2018;Liu et al., 2019;Patel et al., 2022), and petroleum science (Eliyahu et al., 2015;Emmanuel et al., 2016;Alstadt et al., 2016;Yang et al., 2017a;Khatibi et al., 2018a;Liu et al., 2018a;Zhao et al., 2020;Vranjes-Wessely et al., 2021;Wang, 2021;Wang et al., 2021;Wang et al., 2022a) in recent years. This is because OM, as an important component of shale, is not only the parent material for generating and storing shale oil and gas (e.g., Loucks et al., 2012;Zhang et al., 2012;Hackley and Cardott, 2016), but also affects the shale mechanics (Sayers, 2013;Qin et al., 2014;Kumar et al., 2015;Abedi et al., 2016a;Li et al., 2018b;Slim et al., 2019) and fracture performance during reservoir simulations (Brochard et al., 2013;Daigle et al., 2017;Khatibi et al., 2018b) in different ways and to varying degrees. ...
... Zeszotarski et al. (2004) creatively introduced the nanoindentation technology widely used in materials science to measure the mechanical data of the micron-sized kerogens in Woodford shale without destroying the rock, when he tried to interpret the influence of formation and fluid pressures on kerogen maturation and hydrocarbon primary migration through detecting the possible mechanical anisotropy of kerogen in source rocks. Since then, scholars in shale oil and gas field have gradually used the nanoindentation to in-situ measure the mechanical data of the micron-sized organic particles in shales (Ahmadov et al., 2009;Kumar et al., 2012a;Bennett et al., 2015;Alstadt et al., 2016;Mashhadian et al., 2018;Khatibi et al., 2018a;Zhao et al., 2020;Vranjes--Wessely et al., 2021) and coals (Kossovich et al., 2016a;Yu et al., 2018;Zhang et al., 2018;Hou et al., 2020). This application of nanoindentation into mechanical characterization of OM in source rocks represents a major progress in rock mechanics and organic petrology, and it is foreseeable that relevant research will continue to increase in the future with the global increase of shale hydrocarbons and coalbed methane development. ...
... Kerogen is the most abundant form of naturally occurring OM (Durand, 1980a). It can act not only as a generator of shale oil/gas (Vranjes-Wessely et al., 2021) but can also affect the macro-mechanical characteristics of shale, especially for shales of high carbon content (Li et al., 2018a(Li et al., , 2018bPrasad et al., 2009;Zhao et al., 2020). When the total organic carbon (TOC) content (weight %) of Bakken Formation shale increases from 7 to 20%, that is, its volume percentages (v.%) can be up to 14-40%, the Young's modulus decreases from 45.67 to 20.08 GPa in the bedding plane parallel direction (Li et al., 2018a). ...
... In shales, OM is usually scattered, with particle sizes ranging from a few microns to hundreds of microns, and it is impossible to characterize OM mechanical properties using conventional uniaxial/triaxial compression tests. Therefore, micromechanical techniques such as nanoindentation and atomic-force microscopy (AFM) have been introduced as alternative techniques for the characterization of OM mechanical properties (Eliyahu et al., 2015;Emmanuel et al., 2016;Kumar et al., 2012;Li et al., 2018b;Vranjes-Wessely et al., 2021;Zargari et al., 2016;Zhao et al., 2019Zhao et al., , 2020. The Young's modulus of organic matter generally varies between 0 and 25 GPa, with a much lower range than those of inorganic minerals in shale (Khatibi et al., 2018;Zhao et al., 2020). ...
... Solid bitumen is produced during kerogen cracking, and its interaction may affect mechanical properties of kerogen. Previous studies have suggested that solid bitumen is retained by kerogen, forming an over-mature entity (Craddock et al., 2015;Vranjes-Wessely et al., 2021). It is very difficult to distinguish between solid bitumen and kerogen and, although they were not distinguished in this study, indentation positions of similar reflection color and surface features were sited on kerogen or kerogen-bitumen mixtures. ...
Kerogen is the most abundant form of naturally occurring organic matter in organic-rich shale, and it can affect the macro-mechanical characteristics of shale reservoirs during the whole shale thermal evolution process. However, the magnitude and mechanism of this effect are not clear. Here, a relatively low-maturity kerogen isolated from an organic-rich shale was used to prepare a series of kerogen samples of different maturities in an artificial pyrolysis experiment. Raman spectroscopy and solid-state ¹³C nuclear magnetic resonance (NMR) spectroscopy were used to determine changes in the chemical structures of the kerogen samples, and in situ nanoindentation analysis was employed to characterize the evolution of their mechanical properties. Raman spectral analyses indicate that Raman band separation (RBS) is very sensitive to thermal maturity, increasing with maturity and displaying a strong linear correlation with mechanical properties of kerogen such as hardness and Young's modulus. Based on the solid-state ¹³C NMR and nanoindentation analysis, the evolution of kerogen mechanical properties with maturity can be divided into two stages: oil-generation wet-gas stage (corresponding to 0.7%–2.5% EasyRo), and dry-gas stage (2.5%–4.5% EasyRo). In the former, kerogen exhibits a soft nature with relatively low hardness and Young's modulus, possibly attributable to its relatively large proportion of aliphatic carbon and long-chain alkanes. When maturity reaches the gas stage, kerogen becomes increasingly stiff due to the markedly increased proportion of aromatic and bridgehead carbon. Considering the strong positive correlation between mechanical parameters and chemical-structure parameters (RBS, and aromatic and bridgehead carbon ratios) of kerogen during the overall thermal evolution process, we suggest that the stiffening of kerogen may result from the decreasing content of aliphatic chains and increasing aromatic carbon. This study will be beneficial to the development of rock mechanical models that are critical for accurately evaluating borehole stability and optimizing hydraulic fracture design.
... According to Ref. [19]; the main pore type detectable at the scale of BIB-SEM was intergranular. Only minor organic matter-hosted porosity could be determined here, which agrees with the results of this study, as well as data reported by Ref. [114] for a sub-set of samples from the Shahezi Formation which is also covered here. ...
Free hydrogen detected in the Songke-2 well (Songliao Basin, China) has a strong crustal contribution. Here we evaluate whether the source could be the organic matter in Lower Cretaceous coals and shales, and extend our findings regionally. We could establish the rapid growth of aromatic ring systems, forming hydrogen, methane and pyrobitumen, using high resolution mass spectrometry. Molecular hydrogen is generated after late hydrocarbon gas generation is complete, concluding at Rr = 5.0%. The kinetic parameters of molecular hydrogen formation were constructed by subtracting the hydrogen associated with hydrocarbon formation from the total hydrogen, as measured by extensive open system pyrolysis experiments. This new insight was achieved using a CH4–H2 stoichiometric balance. Generalised calculations indicate that the yield per unit rock volume closely resembles that of economic shale gas in the Barnett Shale, though storage in organic matrices is unlikely in this depositional setting. While the prolific generation of hydrogen from organic sources appears to be a reality in the Songliao basin, the free H2 in the Songke-2 mudstream coming from this source must most likely have migrated into the basement rocks mainly from lateral equivalents of the Shahezi rather than from the drilled section itself.
... It also decreases thermal drift effects [1]. The large datasets obtained using this method can utilize advanced statistical methods and artificial intelligence algorithms [2][3][4]. High-speed indentation is suitable for all materials [5,6], although specifically, for multi-phase materials [7]. ...
High–speed nanoindentation rapidly generates large datasets, opening the door for advanced data analysis methods such as the resources available in artificial intelligence. The present study addresses the problem of differentiating load–displacement curves presenting pop-in, slope changes, or instabilities from curves exhibiting a typical loading path in large nanoindentation datasets. Classification of the curves was achieved with a deep learning model, specifically, a convolutional neural network (CNN) model implemented in Python using TensorFlow and Keras libraries. Load–displacement curves (with pop-in and without pop-in) from various materials were input to train and validate the model. The curves were converted into square matrices (50 × 50) and then used as inputs for the CNN model. The model successfully differentiated between pop-in and non-pop-in curves with approximately 93% accuracy in the training and validation datasets, indicating that the risk of overfitting the model was negligible. These results confirmed that artificial intelligence and computer vision models represent a powerful tool for analyzing nanoindentation data.
... The idea was further enhanced by using simultaneous analysis of optic and SEM gathered images [52]. The idea of the usage of optical microscopy obtained images for automated macerals identification was also considered by other researchers [53][54][55][56][57][58]. Młynarczuk and Skiba proposed the usage of machine learning (ML) and artificial intelligence methods in maceral identification [59]. ...
The study of the petrographic structure of medium- and high-rank coals is important from both a cognitive and a utilitarian point of view. The petrographic constituents and their individual characteristics and features are responsible for the properties of coal and the way it behaves in various technological processes. This paper considers the application of convolutional neural networks for coal petrographic images segmentation. The U-Net-based model for segmentation was proposed. The network was trained to segment inertinite, liptinite, and vitrinite. The segmentations prepared manually by a domain expert were used as the ground truth. The results show that inertinite and vitrinite can be successfully segmented with minimal difference from the ground truth. The liptinite turned out to be much more difficult to segment. After usage of transfer learning, moderate results were obtained. Nevertheless, the application of the U-Net-based network for petrographic image segmentation was successful. The results are good enough to consider the method as a supporting tool for domain experts in everyday work.
