Laurence J. Jacobs

This research considers the development of non-collinear wave mixing techniques to image internal microscale damage throughout the interior volume of relatively large specimens. This measurement technique can uniquely provide localized values of material nonlinearity, which is scanned from multiple points in a material of interest. Under certain resonance conditions, cross interaction between two elastic waves can generate a scattered wave whose amplitude is dependent upon the material nonlinearity at the point of mixing. By exploiting the underlying mechanics of nonlinear wave mixing, it is possible to mix two incident waves with frequencies low enough to propagate without being scattered by an inherently heterogenous microstructure such as concrete, while still being sensitive to damage features with length scales well below these incident wavelengths. For this study, scanning and imaging is accomplished with both single element transducers and arrays. Knowledge of the wave speeds in the specimen plus synchronization identifies the location of the mixing zone – the specific volume of being imaged. The viability of the proposed technique is demonstrated from the various experimental iterations before arriving at a successful phased array based nonlinear wave mixing measurement method.

In the search for techniques to distinguish between merely good and excellent bond conditions, linear reflection based methods are simple to apply, but their sensitivity decreases rapidly with increasing bond quality. As an alternative, this research reports on the experimental validation of analytical and computational models of interface properties using non-collinear shear wave mixing. The interface between two solids in dry contact under varying contact pressures serves as a model for diffusion bonds with different degrees of imperfection. Both nonlinear shear–shear wave mixing and linear reflection and transmission measurements are considered. The results show that under increasing load, the nonlinear interface interaction coefficients decrease more slowly in comparison to analogous linear coefficients, demonstrating the higher sensitivity of nonlinear coefficients to subtle interface imperfections in high quality bonds. This comparison is achieved by integrating approximate phenomenological models with the experimentally measured data. Overall, these findings demonstrate the utility of non-collinear shear wave mixing for localized nonlinear interface measurements, and highlight advantages over linear methods for interface characterization.

It is well established in the scientific literature that a material's nonlinear response is much more sensitive to increasing dislocation density, microcrack nucleation, and other types of early material degradation than its linear response. The nonlinear elastic behavior of materials can be studied using various nonlinear ultrasonic techniques (NLU). However, they are all significantly more complex than their linear counterparts; therefore, they are often limited to a laboratory environment, and their field of application in industry is very narrow. In recent years, numerous publications have proposed new techniques based on the so-called Sideband Peak Count (SPC) method that utilizes relatively simple ultrasonic measurements to evaluate the degree of nonlinearity in materials. In contrast to conventional NLU NDE techniques, such as harmonic generation or wave mixing, SPC currently lacks a rigorous theoretical basis. To fill this gap, this paper presents computational results obtained under the assumption of classical quadratic nonlinearity using the COMSOL Multiphysics finite element software package. Parametric studies of four relevant variables ‒ excitation level, material nonlinearity, localized nonlinearity caused by a defect, and linear scattering caused by a geometrical feature ‒ were conducted. All the results of this numerical parametric study indicate that the SPC method and the SPC damage parameter named SPC-Index (SPC-I) offer limited sensitivity to changes in the level of classical acoustic nonlinearity at typical excitation levels used in ultrasonic NDE.

Previous research established a nonlinear stiffness model for imperfect interfaces and analyzed non-collinear wave mixing at such interfaces for two incident shear waves. This study extends those results to encompass five possible non-collinear wave mixing modalities involving longitudinal and shear vertical waves, considering both material and interfacial nonlinearities. By simultaneously applying the conditions for both bulk resonance and interface phase-matching, the two required incident wave angles are determined, allowing the mixing efficiency to be directly characterized by the frequency ratio. The analytical predictions for both types of nonlinearities are numerically validated using COMSOL finite element simulations. Quantitative comparisons across different wave mixing scenarios reveal the relative contributions of material and interface nonlinearity. These findings offer valuable insights and guidelines for designing and interpreting future experimental studies involving non-collinear ultrasonic wave mixing.

Abstract Early detection of alkali-silica reaction (ASR) with nondestructive evaluation (NDE) is important for the monitoring of damage and assessment of repair performance, while effective NDE in large-scale reinforced concrete elements remains challenging. Nonlinear acoustic methods have shown promise for this application due to their sensitivity to the microscale damage characteristic of early-stage ASR. The objective of this investigation is to examine if a recently introduced NDE method employing ultrasonic non-collinear wave mixing may be used to detect and quantify internal ASR damage in large-scale reinforced concrete elements. Measured nonlinearity parameters from an undamaged concrete column are compared with those from reinforced and unreinforced concrete columns both experiencing ASR damage. This study incorporates corrections for attenuation and diffraction in the measured amplitudes of the mixed wave signals to allow for direct comparisons among the three columns, and at two different depths. The nonlinearity parameters measured for the undamaged and ASR-affected columns show clear differences, which are independently validated through scanning electron microscopy (SEM) and ultrasonic second harmonic wave generation (SHG) testing on an extracted core. While severities of damage at two depths were not clearly discernible by SEM, the nonlinearity metrics from both wave mixing and SHG measurements exhibit similar trends of higher nonlinearity at greater depths. These experimental results demonstrate that the non-collinear wave mixing technique is a promising NDE method for detecting and quantifying internal microscale damage in large-scale concrete members.

The potential of additive manufacturing is often limited by qualification issues, particularly due to process defects such as lack-of-fusion porosity and highly anisotropic elastic properties. This research demonstrates the ability of ultrasonic measurement techniques to assess these elastic properties, process defects, and microstructural characteristics. Ultrasonic velocity measurements are used to evaluate the impact of various process parameters and heat treatments (HTs) on the elastic anisotropy of laser powder bed fusion 316 L stainless steel. These variations are linked to material characteristics through microstructural analysis and porosity measurements. By characterizing the orthotropic elastic behavior, this study quantifies the errors that can arise in the design and analysis of additively manufactured parts by assuming isotropic or transversely isotropic elastic properties. Furthermore, HTs are used to isolate and quantify the individual contributions of process defects such as lack-of-fusion defects and microstructural factors-including crystallographic texture and grain morphology-to elastic anisotropy. The findings of this research highlight the potential of ultrasonic techniques for monitoring and qualifying additively manufactured materials.

The objective of this research is to demonstrate the effectiveness of a phased array-based nonlinear wave mixing technique to characterize internal, localized microscale damage in an additively manufactured (AM) component. By using phased arrays for the generation of the incident waves, it is possible to produce a nonlinear wave mixing scanning technique without the need for immersion or changing coupling conditions. The phased arrays can be configured to generate incident waves in multiple directions that meet the resonance conditions required for nonlinear wave mixing at a variety of internal locations. This allows for the scanning of a specimen without the removal and re-coupling of the source transducers, leading to greater scanning speed and repeatability. To demonstrate the accuracy of this phased array wave mixing approach, measurements of acoustic nonlinearity in an AM component are first made with a bulk wave second harmonic generation through thickness measurement. Next, nonlinear wave mixing measurements are made with single element transducers to confirm the sensitivity of the proposed nonlinear wave mixing approach to lack-of-fusion porosity in AM metals. Finally, phased arrays are used to highlight the effectiveness of the proposed nonlinear wave mixing technique in these same AM components.

This paper proposes a framework of using deep learning-assisted methods for the prediction of interfacial conditions in coated plates using guided wave data. The coating-substrate interface is modeled as a linear spring layer of zero thickness, and the mechanical behavior of this spring layer is characterized by the spring compliance. Both tangential and normal spring compliances are introduced to characterize the bond quality. Numerical simulations are conducted for a wide range of spring compliances to generate the corresponding dispersion curves. A Long Short-Term Memory (LSTM) network is utilized to predict the interfacial conditions. In addition, we consider the delamination cases where the coating layer is completely separated from the substrate over the delaminated region. Finite element simulations are carried out to model guided wave generation, propagation, interaction with delamination, and reception. The time-space images are formed by measuring the time-domain signals by receivers at several locations downstream from the source transducer, which are then fed into the developed Convolutional Neural Network (CNN). Once trained, this Deep-Learning (DL) model enables the accurate prediction of delamination location and size. Results of this paper demonstrate that the proposed methodologies have tremendous potential for characterizing interfacial conditions in practical Nondestructive Evaluation (NDE) and Structural Health Monitoring (SHM) applications.

There are no nondestructive evaluation (NDE) tools capable of characterizing microscale damage throughout the thickness of concrete components, due to the multiphase, heterogeneous and multiscale nature of concrete. Ultrasound is only scattered by features at the same length scale, or smaller, than the wavelength of a wave’s dominant frequency. Successful imaging of microscale damage using ultrasound requires that the ultrasonic wavelength be on the order of a few millimeters (or smaller), yet the heterogeneous microstructure of concrete, with its fine and coarse aggregates, is on this same (and higher) micrometer/millimeter length scale, causing excessive ultrasonic wave scattering even in “good” concrete. The proposed solution applies non-collinear wave mixing to spatially image microscale damage, while still maintaining penetration through thick concrete components. This microscale imaging is possible by combining nonlinear wave mixing, with advanced phased array hardware and software to develop a breakthrough tool that will bring revolutionary changes in NDE of concrete infrastructure in terms of image resolution and depth of penetration. This work uses non-collinear wave mixing which exploits the physics that material nonlinearities such as microscale damage, cause interactions between two intersecting ultrasonic waves due to cross-mixing, which can lead to the generation of a third wave with a frequency and wave number of the sum or difference of the incident waves. The concrete material volume at this mixing point is characterized/imaged. This project delivered a single-sided nonlinear ultrasonic phased array imaging device, that can image microscale damage (microcracks of 100 micrometers) through a 0.5 m thick concrete component and assessed the commercial feasibility of such arrays for improved crack detection.