Spalled Surface Research Articles

Experimental study on the influence of non-penetrating crack spatial distribution on brittle fracture process

To gain insights into the spatial distribution of non-penetrating cracks during the rock fracture process, a comprehensive uniaxial compression test is conducted on cubic gypsum specimens (100 mm × 100 mm × 100 mm) containing two non-penetrating cracks. The two pre-formed cracks are rectangular, with dimensions of 25 mm length, 2 mm width, and depths of 80 mm and 35 mm on adjacent sides of the specimen. The depth of the 80 mm crack can be adjusted from 0° to 150° in increments of 30°, while the other is fixed at a 45° angle. The results show that the spatial distribution of non-penetrating cracks can significantly influence the strength of the specimen. Initially, the strength of the specimen exhibits an upward trend and subsequently declines as the pre-crack inclination angle of the main rupture plane increases, ultimately reaching its pinnacle at 90°. The total percentage of tensile cracks in specimens with different inclinations are found to be 57%, 57%, 63%, 77%, 68%, and 61%, respectively. This change aligns seamlessly with the fluctuation in specimen strength as influenced by the angle of inclination. Non-penetrating cracks can also induce spalling on the specimen surface and give rise to anti-wing cracks, thereby exacerbating the spalling on the specimen surface. The inclinations of non-penetrating cracks can inevitably exert a certain influence on the propagation of neighboring non-penetrating cracks. Additionally, the macro-scale shear fracture of the specimen often occurs on the side of the non-penetrating crack that is deeper. The curved tensile fracture surface formed by the extension of the non-penetrating crack bears resemblance to the non-penetrating region in its ability to somewhat restrain the propagation of new cracks. Even under uniaxial compression, the spalling surface of the specimen containing spatial non-penetrating cracks frequently exhibits fracture characteristics belonging to I-III mode fracture, while its interior may display characteristics belonging to I-II-III mode fracture. These findings hold significant implications for comprehending and elucidating the genuine fracture process and three-dimensional fracture theory of rocks.

Dynamic mechanical response and constitutive model of (Ti37.31Zr22.75Be26.39Al4.55Cu9)94Co6 high-entropy bulk metallic glass

In this work, the mechanical response and fracture characteristics of (Ti37.31Zr22.75Be26.39Al4.55Cu9)94Co6 high-entropy bulk metallic glass (HE-BMG) were investigated in detail over a wide range of strain rates (10−4–105 s−1). The HE-BMG exhibited a negative strain rate sensitivity under uniaxial compression, with the strength showing more significant rate dependence under dynamic conditions. The shear band behavior translated from the dominance of multiple shear bands propagations under quasi-static compression to the rapid propagation of a single shear band to form cracks under dynamic compression. Under dynamic loading, the shearing velocity increased along an arc-shaped displacement path, and the local stress state on the shear fracture surface shifted from compressive-shear to tensile-shear, accompanied by changes in fracture morphologies. The spall strength of HE-BMG decreased as flyer impact speed increased, while the long-term dependence of spall strength on strain rate may be positive. With the increase of impact speed, the main microstructural features of the spalling surface translated from flat regions accompanied by dimples to cup-cone structures. Furthermore, the complete parameters of the Johnson–Holmquist II (JH-2) model for HE-BMG were obtained based on experimental data. Numerical simulations of planar impact, penetration, and hypervelocity impact are in good agreement with experimental results, demonstrating the validity of the JH-2 model parameters. The current work has important guiding value for the application of HE-BMG in space debris protection.

DEVELOPMENT OF A MACHINE VISION SYSTEM FOR DAMAGE AND OBJECT DETECTION IN TUNNELS USING CONVOLUTIONAL NEURAL NETWORKS

Abstract. Tunnel inspection, i.e. detection of damages and defects on concrete surfaces, is essential for monitoring structural reliability and health conditions of transport facilities, thus providing safe and sustainable urban transportation infrastructures. In this study, an innovative visual-based system is developed for damage and object detection tasks in roadway tunnels based on deep learning techniques. The main components of the developed Machine Vision System such as industrial cameras, flash-based light sources, controller, the synchronization unit and corresponding software programs are designed to collect high-resolution images with sufficient quality from dimly lit tunnel environments in normal traffic flows with an operating speed of 30–50 km/h. Unlike recent studies, the training data includes multiple types of damage such as cracks, spalling, rust, delamination and other surface changes. Furthermore, 10 classes of common tunnel objects including traffic signs, traffic cameras, traffic lights, ventilation ducts, various sensors and cables are labeled for object detection. As state-of-the-art Convolutional Neural Networks, DeepLab and U-Net are trained and evaluated using accuracy metrics for image segmentation. The results highlight the most important parameters of the discussed Machine Vision System as well as the performance of DeepLab and U-Net for object and damage detection.

New Damage Accumulation Model for Spall Propagation Mechanism in Bearing Raceways

The aim of this study was to investigate the spall propagation mechanism in ball bearing raceways using physics-based models. Spalling is one of the most common types of bearing failures that can lead to catastrophic failure. This research takes a step forward toward developing a prognostic tool for ball bearings. It is first necessary to understand the spall progression process in order to formulate a constitutive law of spall deterioration and to estimate the amount of remaining useful life. Fragment formation in the vicinity of the spall edge was found to consist of surface and sub-surface cracks that eventually coalesce, and a fragment is released from the raceway, based on naturally-developed spalls. Here, we describe a physics-based model, integrating a dynamic model with a finite element one to simulate this process. A continuum damage mechanics (CDM) approach and fracture mechanics tools were embedded into the finite element model to simulate the damage propagation. The formation of cracks in the vicinity of the spall (surface and sub-surface cracks) were studied using this effective stress CDM model, and the propagation of the cracks was examined using two approaches: a fracture mechanics approach and an accumulated inelastic hysteresis energy CDM approach. The latter also predicts the overall process of a single fragment release. The simulation results of the spall propagation models are supported by experimental results of spalls from both laboratory experimental bearings and an in-service Sikorsky CH-53 helicopter swashplate bearing. The results obtained show that the impact of the ball on the spall edge affects the crack propagation and the appearance of the surface and sub-surface cracks. Both release the residual stresses and cause crack propagation until a fragment is released.

Shock wave characteristics and spalling behavior of non-coherent Cu/Nb multilayers

We investigate shock wave characteristics and spalling failure of Cu/Nb nanolaminates under shock loading via molecular dynamics simulations. The features of stress wave propagation in the Cu/Nb multilayers, including the reflection-loading and reflection-unloading of wave-front stress, and the strong interfacial discontinuities of transverse normal stress components, maximum shear stress and average kinetic energy are revealed by atomistic simulations. Continuum models are proposed to explain these shock wave evolution characteristics. In addition, the simulations well reproduce experimentally observed abnormal ductile damage distribution, i.e., voids nucleate in the Cu phase rather than in the Nb phase or interfaces. Such abnormality is intrinsic dynamic characteristics of nanolaminates, independent of layer thickness and shock intensity. Differences in theoretical spall strengths of Cu, Nb and the interfaces are calculated using the Cauchy-Born kinematic constraint. Moreover, we bring novel insight into the cause of the abnormality from a dynamic perspective that the maximum tensile stress in the stronger Nb phase is restricted by the spall strength of the weaker Cu phase, sound velocity, strain rate and layer thickness. The simulations also indicate that dynamic ductility can be enhanced by reducing the layer thickness, due to that more spall fracture surfaces are generated in nanolaminates with finer layer thickness.

High‐Efficiency Solar Cells Grown on Spalled Germanium for Substrate Reuse without Polishing

AbstractRadical reduction of III–V device costs requires a multifaceted approach attacking both growth and substrate costs. Implementing device removal and substrate reuse provides an opportunity for substrate cost reduction. Controlled spalling allows removal of thin devices from the expensive substrate; however, the fracture‐based process currently generates surfaces with significant morphological changes compared to polished wafers. 49 single junction devices are fabricated across the spalled surface of full 50 mm germanium wafers without chemo‐mechanical polishing before epitaxial growth. Device defects are identified and related to morphological spalling defects—arrest lines, gull wings, and river lines—and their impact on cell performance using physical and functional characterization techniques. River line defects have the most consistent and detrimental effect on cell performance. Devices achieve a single junction efficiency above 23% and open‐circuit voltage of 1.01 V, demonstrating that spalled germanium does not need to be returned to a pristine, polished state to achieve high‐quality device performance.

Morphological features and fractography analysis for in situ spalling in the China Jinping underground laboratory with a 2400 m burial depth

Brittle spalling of hard rock is a typical failure in deep underground engineering that induces challenges for construction safety. Seventy marble spalling samples were collected from the China Jinping Underground Laboratory Phase-II (CJPL-II) to analyse their mesoscopic morphology and failure mechanism. The 3D scanning and corresponding statistical analysis indicated that the spalling surfaces were not smooth but with low roughness and evident anisotropy, and that the morphological transition (i.e. from the mirror zone to the mist zone then to the hackle zone) on the spalling surface was observed. The fractography of the CJPL-II spalling samples showed typical features of brittle fracture dominated by the granular morphology of intergranular fracture based on the Scanning Electron Microscope technology. What’s more, the simulated spalling by the laboratorial true triaxial unloading tests showed similar morphology with that of CJPL-II spalling. Moreover, by comparing the marble spalling samples with other granite spalling and basalt sapling samples, the effect of the mineral particle size of the surrounding rock on the roughness of the spalling slice was discussed. This study deepens the understanding of the in-situ spalling mechanism of hard rock under high geostress conditions.

Damage characteristics of YAG transparent ceramics under different loading conditions

YAG (Y3Al5O12) transparent ceramics have attractive application prospects for transparent armor protection modules because of their excellent light transmittance and anti-ballistic capability. Understanding the fracture behavior and damage mechanism of YAG is necessary for armor design. To explore the damage characteristics of YAG under compression and tension, shock compression and shockless spalling experiments with soft recovery technique are conducted. The spall strength of YAG is obtained and the recovered samples are observed by CT and SEM. It is shown that the macroscopic damage characteristic of YAG under compression is vertical split cracks with oblique fine cracks distributed in the entire sample, while that under tension is horizontal transgranular cracks concentrated near the main spall surface. The cracks generated by macroscopic compression, tension and shear stress extend in similar tensile form at the microscale. The proportion of transgranular fractures on spall surfaces is higher than that of cracks induced by macroscopic compression. Meanwhile, higher loading rate and longer loading duration increase the transgranular fracture percentage.

Impacts of Mode Mixity on Controlled Spalling of (100)-Oriented Germanium

Controlled spalling is a technology to prepare single-crystal thin films of semiconductors by fracture with a subsurface crack propagating nearly parallel to the substrate surface. Practical applications require uniform thickness and a smooth surface across the whole film. Both wafer-scale and patterned-stressor-defined small-area spalling of germanium substrates are conducted experimentally and numerically. River line features are observed on spalled surfaces close to lateral edges of the spall, regardless of the spall direction and the size of the spalled area. Three-dimensional finite element method modeling shows the river lines are caused by mixed mode I + III loading near the lateral edges of spall and predicts a spall depth variation near the lateral edges of spall due to mixed mode I + II loading. The absolute range of river lines increases with lateral size of spall, while the relative range of river lines decreases, consistent with variations in mode mixity.

Numerical simulation on dynamic damage evolution of high pure copper with different grain sizes

High-purity (HP) copper targets with grain sizes of 50, 130 and 200 μm are constructed by using the Voronoi method. Damage nucleation points are randomly prefabricated at the grain boundaries. A two-dimensional axisymmetric finite element model is established to simulate the spallation experiment of HP copper target. The effects of grain size and loading stress on the macro- mechanical response and meso-damage evolution of HP copper spallation are studied and compared with the relevant experimental results. Based on the analysis of free surface velocity profiles, the effects of grain size on the location of pull back velocity rebound point, velocity rebound slope and velocity rebound amplitude are revealed. It is demonstrated that the spalling strength corresponds to the peak value of tensile stress in the damage zone, which essentially represents the critical stress of micro damage nucleation or early growth. Based on the characteristic analysis of damage evolution nephogram, the evolution process of localized plastic strain field around the micro-voids in the growth and coalescence process is reproduced, and the strong dependence of micro-void coalescence behavior on grain size is clarified. The loading stress amplitude has little effect on the location of pull back velocity rebound point, but has a significant effect on the growth and coalescence behavior of micro-voids. The slope and amplitude of pull back velocity rebound increase with loading stress increasing, which is consistent with the relevant experimental result. With the increase of the loading stress, the micro-voids grow from independent growth to coalescence, thus forming spalling surface. The physical process of damage evolution determines the wave oscillation characteristics after the pull-back rebound point. The numerical simulation results reproduce the physical process of damage evolution and its influence on the macroscopic mechanical response, which is of great significance for further understanding spall damage evolution mechanism and theoretical model construction.

Long-term effects of wildfire on rock weathering and soil stoniness in the Mediterranean landscapes

The severe wildfire at Mt. Carmel, Israel, in 2010, caused massive destruction of carbonate rocks. The thermal shock caused extreme exfoliation, producing large and flat clasts, affecting rocks to a depth of up to 20 cm. A decade after the fire, most flakes and spalls disappeared from the rock outcrops and adjacent soils. From these observations, this study pursued two objectives: (a) to monitor and analyze the spatio-temporal distribution of the disintegrated flakes 10 years after the fire and (b) to test the hypothesis that fires contribute to increased soil stoniness via physical and chemical flake erosion.The studied area included five lithostratigraphic units composed of chalk, limestone, and dolomite. The Schmidt Hammer test showed that after a decade, most of the spalled surface on the burned outcrops was lost, exposing new rock surfaces to atmospheric and weathering processes. The spalls and flakes were broken down and pulverized.The most prominent effects were changes in surface stoniness on the rendzina soils over the chalks, while there was less impact on the dolomite and limestone samples. The stoniness of the non-burned chalk was 23–39% and increased significantly to 69–86% in the burned area. Chalk erosion produced large (>16 mm, median 8–16 mm) and abundant gravel, suggesting fragmentation of large spalls, and particles that lost their bladed shapes becoming oblate and equant.While earlier works suggested that increasing rock fragment cover is often associated with the removal of fine particles, our results showed a substantial increase in rock fragments due to fire-induced exfoliation of rock surfaces, leading to long-term changes in soil properties. We therefore propose that the size, shape, and spatial distribution of rock fragments should be considered when examining the effects of rock fragments on hydrological and geomorphological processes or on post-fire soil rehabilitation.

Analysis of the Ways to Identify Rail Running Surface Defects by Means of Vibration Signals

The article discusses a preliminary concept of a method enabling the identification of chosen rail running surface defects, such as squats, spalling, and running surface defects, by analysing the parameters of vibration signals. It features a description of the methodology of the conducted tests, the scope thereof, and the selection of the measurement points with specific defect types. The article covers selected results of vibration tests, the results of analyses of recorded signals for defective track sections and those for control track sections. The presented measurement results have been obtained for the technical–operating conditions occurring on railway line no. 213 Reda – Hel and line no. 131 Chorzów Batory – Tczew. The preliminary test results and conclusions included in the article show that it is reasonable to pursue further research into the phenomena involving the utilisation of vibroacoustics in rail performance diagnostics Keywords: vibroacoustics, squat, spalling, running surface defect

Fracture mechanism of steel plate loaded by explosive-induced shock waves

The fracture mechanism of in-plane notched steel plate were experimentally investigated under the loading of the triangle shock waves. A series of 20 mm thick steel plates were loaded by explosive charges with different weights and the steel plates were notched in the load surface and free surface. The explosives cut the central part of the plate along the notches and the recovered specimens were processed and analyzed by the scanning electron microscope (SEM). It was found that the explosive charges and the notch properties had a significant effect on the notch fracture and the spall observed in the experiments. Thick explosive charge plate generated deeper, smaller and more dimples on the spall fracture surface which means the ductile of the fracture was increased and the crack extension resistibility was strengthened. This led to a step-like morphology on the spall surface. Moreover, the fractures of two sides of the notch were different. The shear fracture emerged on the loading side and the tensile fracture appeared to the free surface side. However, when there was no notch in the loading surface, the shear fracture appeared to the free surface side and extended to the loading surface. At the same time, the spall and notch fractures interrelated near the notch. Based on the experimental results, a new damage level function was proposed considering the statistic of crack length instead of the void volume. The theoretical analyses of spalls under all six loadings were performed and compared with the experimental results. The comparison showed a good agreement and the reliability of the proposed model was verified. The fracture behavior and mechanism obtained from the experiments provided a good foundation for the further theoretical investigations and the improved damage model was applicable in engineering analysis.

Impacts of temperature and surface finish upon steam oxidation of austenitic steel TP347HFG

The effects of the steam oxidation process on an austenitic steel (TP347HFG) exposed under isothermal conditions between 600 and 800 °C for up to 2500 h have been investigated. Samples with both as-received and ground surfaces have been exposed and the impact of surface finish on the oxidation process analysed using scanning electron microscopy with energy dispersive X-ray analysis. Exfoliated oxide flakes have also been examined to characterise their microstructures on fractured sections as well as external and spalled surfaces. Microscopic analyses demonstrated that ground surfaces possess better steam oxidation resistance than as-received surfaces due to their ability to form a more protective chromium-rich layer. The formation of regions of thicker multi-layered oxides was noted on both types of surface finish, covering large areas on as-received surfaces and only nodules on ground surfaces (spreading with increasing exposure temperature and time).

Laser-driven flyer application in thin film dissimilar materials welding and spalling

This paper applied a low cost method to pack and drive laser-driven flyer in the applications of welding and spalling. The laser system has the maximum energy of 3.1J, which is much lower than that used in the previous study. The chemical release energy from the ablative layer was estimated as 3.7J. The flying characteristic of laser-driven flyer was studied by measuring the flyer velocity at different locations with photonic Doppler velocimetry (PDV). The application of laser-driven flyer in welding Al and Cu was investigated at different laser spot size. Weld strength was measured with the peel test. Weld interface was characterized with optical microscopy (OM) and scanning electron microscopy (SEM). The study of application of laser-driven flyer in spalling was carried out for both brittle and ductile materials. The impact pressure was calculated based on the Hugoniot data. The amount of spalling was not only related to the impact pressure but also related to the duration of impact pressure. The fractography of spalled fracture surface was studied and revealed that the fracture mode was related to the strain rate. The spall strength of Cu 110, Al 1100 and Ni 201was measured and was consistent with the literature data.

Plastic behavior of steel and iron in high strain rate regime

The shock response of anti-hydrogen steel (HR-2) and iron was studied in a series of laser-driven shock wave experiments. A line-imaging optical recording velocity interferometer system for any reflector was used to record the free surface velocity histories of shock loaded samples, 100–300 \(\upmu \hbox {m}\) thick and with an initial temperature ranging from 296 to 1073 K. Based on the recorded free surface velocity profiles, the elastic precursors, dynamic yield and tensile (spall) strengths of HR-2 and iron were calculated. The dependence of the measured HEL stresses on the propagation distance for HR-2 and polycrystalline iron is approximated by a power law relationship.But, that for the single crystal iron with orientation of (110) seems to be constant. Spall strengths \((\upsigma _{\mathrm{sp}})\) of HR-2 estimated from the magnitude of the pull-back signal show that the spall strength dependence on the strain rate \((\dot{\upvarepsilon })\) is approximated by a power law relationship \(\upsigma _{\mathrm{sp}} =0.24\left( \dot{\upvarepsilon } \right) ^{0.24}\,\left( {\hbox {GPa}} \right) \). The spall strength of HR-2 and single crystal iron at the initial temperatures of 296–1073 K decreases slightly with increasing temperature and that of poly crystal iron abnormally increases at a temperature of 873 K. The X-ray diffraction results on the recovered poly crystal samples indicate significant changes in the relative peak intensity and the change in the crystal orientation may be the reason for the abnormal increasing at 873 K. The spall fracture surfaces of HR-2 were observed using a 3D laser scanning confocal microscope. The spall surface contains many dimples, suggesting that the fracture mode is that of ductile fracture. At ambient temperatures, the dimples and crowns were evenly distributed at the fracture surface. At high temperatures, many large crowns appeared and were unevenly distributed at the fracture surface.

Experimental investigation on the effect of operating speeds on wear and rolling contact fatigue damage of wheel materials

The objective of this study is to explore the wear and rolling contact fatigue (RCF) characteristics of wheel materials under different operating speeds conditions using a small scale wheel/rail facility. The results indicate that the operating speed plays a vital role in the wear and RCF of wheel materials. As the operating speed increases, both the surface hardness and the hardness of the plastic flow layer on the wheel roller decrease while the worn surfaces become rougher. Shear strain hardening of wheel roller decreases with increasing operating speed. Furthermore, the primary damage mechanism of the wheel roller transforms from spalling and slight surface fatigue to severe fatigue cracks as the operating speed increases. Fatigue cracks initiate from the wear surface and grow along with the soft ferrite lines within the plastically deformation layer. RCF damage become more severe and the wear rate of the wheel roller decreases as operating speed increases. Low wear rate can not adequately remove cracks and leads to a visible increase in the angle of fatigue crack propagation and the crack depth. Furthermore, the wear debris is composed of metallic flakes whose primary compositions are iron, Fe2O3, and Fe3O4 and the size of wear debris decreases, but the thickness increases as operating speed increases.

Concrete Crack Repair with Polymer Modified Mortars: The Status Quo in the South African Construction Industry and the Way Forward

In recent years, concrete repair has become an integral part of the construction industry. With the vast quantity of concrete used in the South African construction industry over the past 100 years, one can expect an increase in repair and rehabilitation requirements during the extended lifecycle of exposed concrete structures. Crack repair, re-profiling of spalled areas and surface sealing with polymer related materials forms the bulk of such repair and rehabilitation operations. Due to the complexity of these projects and the variety of professionals and other stakeholders involved from the diagnostics to the implementation phase (specialists consultants, contractors, suppliers and owners of the structures), considerable problems seem to have surfaced to ensure cost-effective but sustainable and durable outcomes. It has been found that in many concrete repair projects, the responsibility for the repair work, adequate quality control and the assessment of successful patch repairs are not fully embraced by the various stakeholders.This concern has led to the research as reported in this paper. The research entailed a series of questionnaires drawn up specifically for the four different stakeholder sectors of the concrete repair industry. The results indicate that, although there is agreement that concrete repair is a highly specialized field, there is not enough training in the correct use of the repair materials, nor enough knowledge regarding the diagnostics or material specification and selection processes. Knowledge on polymer modified mortar are also minimal. These problems are compounded by inadequate quality control and lack of ongoing monitoring of patch repair failure. The paper concludes with suggestions on the way forward.

A closed-form criterion for dislocation emission in nano-porous materials under arbitrary thermomechanical loading

Our traditional view of void nucleation is associated with interface debonding at second-phase particles. However, under extreme dynamic loading conditions second-phase particles may not necessarily be the dominant source of void nucleation sites. A few key experimental observations of laser spall surfaces support this assertion. Here, we describe an alternative mechanism to the traditional view, namely shock-induced vacancy generation and clustering followed by nanovoid growth mediated by dislocation emission. This mechanism only becomes active at very large stresses. It is therefore desirable to establish a closed-form criterion for the macroscopic stress required to activate dislocation emission in porous materials. Following an approach similar to Lubarda and co-workers, we derive the desired criterion by making use of stability arguments applied to the analytic solutions for the elastic interactions of dislocations and voids. Our analysis significantly extends that of Lubarda and co-workers by accounting for a more general stress state, finite porosity, surface tension, as well as temperature and pressure dependence. The resulting simple stress-based criterion is validated against a number of molecular dynamics simulations with favorable agreement.

Effects of substrate material and TBC structure on the cyclic oxidation resistance of TBC systems

The effects of different substrates (CMSX-4 and IN738LC) on cyclic oxidation behavior of two thermal barrier coatings (TBCs) were investigated in this study. Cyclic oxidation test was conducted at a peak temperature of 1080°C with 30min hold time. The TBC systems tested consisted of two nickel-base superalloy substrates (CMSX-4 and IN738LC), a platinum aluminide bond coat and two types of 8YSZ top coat (vertical cracks (VC) and columnar structure). Samples with CMSX-4 substrate showed greater cyclic oxidation lifetimes than that with IN738LC, and VC YSZ on CMSX-4 offered longer cyclic life than those with columnar structure. After the oxidation test, the samples were analyzed using a scanning electron microscope (SEM) and an X-Ray diffractometer. The failure location of all TBC systems examined was found to be in the vicinity of TGO, consistent with past research of similar coating systems. Alumina and spinels (CS) were detected on all spalled surfaces although the amount of each phase differs from sample to sample. Ta-rich oxide was found on spalled coating surfaces with CMSX-4 substrate while islands of YSZ on the exposed TGO surface of samples with IN738LC substrate. It is believed that alloy composition, TGO rumpling and stresses induced by TGO growth and coefficient of thermal expansion (CTE) mismatch between different layers all contributed to the reduced cycles to failure of TBC systems with IN738LC substrate.