Engineering microstructural defects, like grain boundaries, offers superior control over transport properties in energy materials. However, technological advancement requires establishing microstructure‐property relations at the micron or finer scales, where most of these defects operate. Here, the first experimental evidence of thermal resistance for individual silicon grain boundaries, estimated with a Gibbs excess approach, is provided. Coincident site lattice boundaries exhibit uniform excess thermal resistance along the same boundary, but notable variations from one boundary to another. Boundaries associated with low interface energy generally exhibit lower resistances, aligning with theoretical expectations and previous simulations, but several exceptions are observed. Transmission electron microscopy reveals that factors like interface roughness and presence of nanotwinning can significantly alter the observed resistance, which ranges from ∼0 to up to ∼2.3 m²K/GW. In stark contrast, significantly larger and less uniform values ‐ from 5 to 30 m²K/GW ‐ are found for high‐angle boundaries in spark‐plasma‐sintered polycrystalline silicon. Further, finite element analysis suggests that boundary planes that strongly deviate from the sample vertical (beyond ∼45°) can show up to 3‐times larger excess resistance. Direct correlations of properties with individual defects enable the design of materials with superior thermal performance for applications in energy harvesting and heat management.

Non-propagating evanescent fields play an important role in the development of nano-photonic devices. While detecting the evanescent fields in far-field can be accomplished by coupling it to the propagating waves, in practice they are measured in the presence of unwanted propagating background components. It leads to a poor signal-to-noise ratio and thus to errors in quantitative analysis of the local evanescent fields. Here we report on a plasmonic near-field scanning optical microscopy (p-NSOM) technique that incorporates a nanofocusing probe for adiabatic focusing of propagating surface plasmon polaritons at the probe apex, and for enhanced coupling of evanescent waves to the far-field. In addition, a harmonic demodulation technique is employed to suppress the contribution of the background. Our experimental results show strong evidence of background free near-field imaging using the new p-NSOM technique. Furthermore, we present measurements of surface plasmon cavity modes, and quantify their contributing sources using an analytical model.

Elastic guided waves were generated in mixed-culture bacterial biofilms for characterizing its viscoelastic properties.

Deep generative models have demonstrated effectiveness in learning compact and expressive design representations that significantly improve geometric design optimization. However, these models do not consider the uncertainty introduced by manufacturing or fabrication. The past work that quantifies such uncertainty often makes simplifying assumptions on geometric variations, while the “real-world,” “free-form” uncertainty and its impact on design performance are difficult to quantify due to the high dimensionality. To address this issue, we propose a generative adversarial network-based design under uncertainty framework (GAN-DUF), which contains a deep generative model that simultaneously learns a compact representation of nominal (ideal) designs and the conditional distribution of fabricated designs given any nominal design. This opens up new possibilities of (1) building a universal uncertainty quantification model compatible with both shape and topological designs, (2) modeling free-form geometric uncertainties without the need to make any assumptions on the distribution of geometric variability, and (3) allowing fast prediction of uncertainties for new nominal designs. We can combine the proposed deep generative model with robust design optimization or reliability-based design optimization for design under uncertainty. We demonstrated the framework on two real-world engineering design examples and showed its capability of finding the solution that possesses better performance after fabrication.

A high-sensitivity frequency domain laser based ultrasonic system is presented which uses a low power, amplitude modulated continuous wave (cw) laser source for acoustic wave generation. The acoustic signals are detected using a path stabilized Michelson interferometer coupled to a rf lock-in amplifier. The modulation frequency of the generation laser is scanned over the bandwidth of interest, and transient acoustic signals are reconstructed from the frequency domain data. The effects of measurement frequency resolution, bandwidth, and time domain aliasing on the reconstructed transient response are discussed. Experimental results on thin plates, where diffuse acoustic wave fields lasting several hundred microseconds are seen as a result of multiple reflections off of sample boundaries, demonstrate that the time domain signal can be unambiguously reconstructed through appropriate selection of frequency resolution. Time domain reconstructions of acoustic signals over a bandwidth of $200\mathrm{MHz}$ demonstrate the utility of the approach for acoustic microscopy; individual acoustic transients can be detected with extremely high signal to noise ratio and time gated for analysis.