Understanding the mechanisms by which vertical arrays of carbon nanotube (CNT) forests terminate their growth may lead to the production of aligned materials of infinite length. We confirm through calculation of the Thiele modulus that several prominent systems reported in the literature to date are not stunted by diffusion limitations. Evidence also suggests that, for many systems, the growth-termination mechanism is spatially correlated among nanotubes, making spontaneous, random catalytic poisoning unlikely as a dominant mechanism. We propose that a mechanical coupling of the top surface of the film creates an energetic barrier to the relative displacement between neighboring nanotubes. A Monte Carlo simulation based on this premise is able to qualitatively reproduce characteristic deflections of the top surface of single- and doubled-walled CNT (SWNT and DWNT) films near the edges and corners. The analysis asserts that the coupling is limited by the enthalpy of the carbon-forming reaction. We show that for patterned domains, the resulting top surface of the pillars is approximately conic with hyperbolic cross sections that allow for empirical calculation of a threshold force (F(max) = 34-51 nN for SWNTs, 25-27 nN for DWNTs) and elastic constant (k, 384-547 N/m for SWNTs and 157-167 N/m for DWNTs) from the images of experimentally synthesized films. Despite differences in nanotube type and precursor chemistry, the values appear consistent supporting the validity of the model. The possible origin of the mechanical coupling is discussed.

Preparation of nanomaterial dispersion or nanofluids requires good characterization techniques, including particle size and morphological measurements. A reliable and straight-forward process to characterize and quantify the degree of dispersion and agglomeration is needed. A wet-cell transmission electron microscope (TEM) technique has been developed to make comparisons between sonicated and hand-shaken solutions of both aluminum oxide nanoparticles and multi-walled carbon nanotubes. In each case, the wet-cell TEM technique reveals images of nanoparticles well dispersed in aqueous solutions due in part to the use of ultrasonic power instead of simply manual shaking and stirring. The technique is currently qualitative and shows great potential for a host of nanotechnology applications.

The objective of this article is to introduce a fairness interpretability framework for measuring and explaining the bias in classification and regression models at the level of a distribution. In our work, we measure the model bias across sub-population distributions in the model output using the Wasserstein metric. To properly quantify the contributions of predictors, we take into account favorability of both the model and predictors with respect to the non-protected class. The quantification is accomplished by the use of transport theory, which gives rise to the decomposition of the model bias and bias explanations to positive and negative contributions. To gain more insight into the role of favorability and allow for additivity of bias explanations, we adapt techniques from cooperative game theory.

Designing a high efficiency thermoelectric material for thermal to electric energy conversion means simultaneously optimizing multiple properties of the material. Although it might seem straightforward to maximize the electrical power and minimize thermal losses, the convoluted relationship between these properties makes optimization complex, requiring a more sophisticated algorithm. The Accelerated Insertion of Materials (AIM) methodology developed to engineer the mechanical properties of complex multiphase steel alloys provides a framework for optimization that can be applied to engineer the thermal and electrical transport properties of a multiphase thermoelectric material. The AIM methodology can be utilized in creating a high figure of merit (zT) material by considering the effects of each structural parameter, such as grain size and grain boundary properties, precipitate volume fraction, and doping and defect concentration of the matrix phase on the zT of the material using a variety of analytical models. The combination of these models provides a way to accelerate the design of high zT materials.