Search for: All records

Understanding and exploiting material flexibility through phenomena such as the bending and twisting of molecular crystals has been a subject of increased interest owing to the number of applications that benefit from these properties, such as optoelectronics, mechanophotonics, soft robotics, and smart sensors. Here, we report the growth of spontaneously bent and twisted ammonium urate crystals induced by the keto–enol tautomerism of the urate molecule. The major tautomer is native to biogenic crystals, whereas the minor tautomer functions as an effective crystal growth modifier to induce naturally bent and twisted ammonium urate crystals. We show that the degree of curvature can be tailored based on the judicious selection of growth conditions. A combination of state-of-the-art microscopy and spectroscopy techniques are used to characterize the origin of bending. Spatially resolved nano-electron diffraction and high-resolution electron microscopy of naturally bent crystals show nearly single crystallinity with local lattice deformations generated by a combination of screw and edge dislocations. These observations are consistent with photoinduced force microscopy and contact resonance atomic force microscopy, which confirmed spatially resolved changes in the intermolecular interactions and the mechanical properties throughout the cross-sectional and axial regions of bent crystals. A mechanism of bending involving the generation of regionally specific dislocations is proposed as an alternative to more commonly reported models. These findings highlight a unique characteristic of tautomeric crystals that may have broader implications for other biogenic materials. 

Abstract The SOL2013-10-28T02:02:58L133C110 flare occurred on the western limb, acquiring the GOES class of X1.0, and we focus on an oscillatory phenomenon detected at 34 GHz by the Nobeyama Radioheliograph (NoRH) during this flare. The oscillation is less obvious at 17 GHz and is unseen for a hard X-ray source detected by the RHESSI. In the 94 Å images from the Atmospheric Imaging Assembly on board the Solar Dynamics Observatory, we traced the evolution of the extreme ultraviolet (EUV) images capturing an eruption around 01:58 UT over the location of the RHESSI 50–100 keV source. We located the microwave emitting loop inferred from the 17/34 GHz maps within the complex EUV loop systems, and performed a model calculation of the dynamic evolution of the microwave brightness, including the radiative transfer in a magnetically asymmetric loop and evolving nonthermal electrons. The results demonstrate that a quasiperiodic injection of energetic electrons at a fixed spatial point is sufficient to reproduce such an oscillatory motion, without an actual shift of the nonthermal electron injection point, and that the magnetic environment required for the microwave loop model is consistent with the observed EUV activities related to the overall reconnection geometry. 

Machine unlearning---efficiently removing the effect of a small "forget set" of training data on a pre-trained machine learning model---has recently attracted significant research interest. Despite this interest, however, recent work shows that existing machine unlearning techniques do not hold up to thorough evaluation in non-convex settings. In this work, we introduce a new machine unlearning technique that exhibits strong empirical performance even in such challenging settings. Our starting point is the perspective that the goal of unlearning is to produce a model whose outputs are statistically indistinguishable from those of a model re-trained on all but the forget set. This perspective naturally suggests a reduction from the unlearning problem to that of *data attribution, where the goal is to predict the effect of changing the training set on a model's outputs. Thus motivated, we propose the following meta-algorithm, which we call Datamodel Matching (DMM): given a trained model, we (a) use data attribution to predict the output of the model if it were re-trained on all but the forget set points; then (b) fine-tune the pre-trained model to match these predicted outputs. In a simple convex setting, we show how this approach provably outperforms a variety of iterative unlearning algorithms. Empirically, we use a combination of existing evaluations and a new metric based on the KL-divergence to show that even in non-convex settings, DMM achieves strong unlearning performance relative to existing algorithms. An added benefit of DMM is that it is a meta-algorithm, in the sense that future advances in data attribution translate directly into better unlearning algorithms, pointing to a clear direction for future progress in unlearning. 

An optimization study is presented to enhance optical dielectrophoretic (ODEP) performance for effective manipulation of an oil-immersed droplet in the floating electrode optoelectronic tweezers (FEOET) device. This study focuses on understanding how the droplet’s position and size, relative to light illumination, affect the maximum ODEP force. Numerical simulations identified the characteristic length (Lc) of the electric field as a pivotal factor, representing the location of peak field strength. Utilizing 3D finite element simulations, the ODEP force is calculated through the Maxwell stress tensor by integrating the electric field strength over the droplet’s surface and then analyzed as a function of the droplet’s position and size normalized to Lc. Our findings reveal that the optimal position is xopt= Lc+ r, (with r being the droplet radius), while the optimal droplet size is ropt = 5Lc, maximizing light-induced field perturbation around the droplet. Experimental validations involving the tracking of droplet dynamics corroborated these findings. Especially, a droplet sized at r = 5Lc demonstrated the greatest optical actuation by performing the longest travel distance of 13.5 mm with its highest moving speed of 6.15 mm/s, when it was initially positioned at x0= Lc+ r = 6Lc from the light’s center. These results align well with our simulations, confirming the criticality of both the position (xopt) and size (ropt) for maximizing ODEP force. This study not only provides a deeper understanding of the position- and size-dependent parameters for effective droplet manipulation in FEOET systems, but also advances the development of low-cost, disposable, lab-on-a-chip (LOC) devices for multiplexed biological and biochemical analyses. 

Historically Black Colleges and Universities (HBCUs) operate and are centered within the nexus of concerted nationwide efforts to advance the participation and success of Black students within the sciences, technology, engineering, and mathematics disciplines (STEM). Through an institution-levelasset-based approach, this study aimed to further elucidate how undergraduate STEM preparation and success at HBCUs is linked to the transition into (and experiences within) graduate education. One hundred and fifty-one HBCU alumni from 37 unique HBCUs completed our HBCU Alumni Success survey. Factor analysis revealed 13 emerging components along three main touchpoints along alumni's graduate pathway: their HBCU undergraduate experiences, graduate application, and decision-making, as well as graduate school experiences. Cluster analysis further identified five unique clusters of alumni, revealing variation regarding the individual, institutional, and cultural factors that contributed to HBCU alumni's experiences within their graduate pathway. Specific attributes that characterized each unique cluster included (Cluster 1) experiencing challenges throughout theirgraduate pathway, (Cluster 2) variation in the sources of motivation that influenced graduate school choice, (Cluster 3) deliberation around attending graduate school, (Cluster 4) high commitment, success, and support in pursuit of a graduate degree, and (Cluster 5) high personal agency as well as faculty and research support within engineering. Implications for practice include capitalizing on the areas of success such as the impact of faculty mentorship and research opportunities. 

We report a power-efficient analog front-end integrated circuit (IC) for multi-channel, dual-band subcortical recordings. In order to achieve high-resolution multi-channel recordings with low power consumption, we implemented an incremental ΔΣ ADC (IADC) with a dynamic zoom-and-track scheme. This scheme continuously tracks local field potential (LFP) and adaptively adjusts the input dynamic range (DR) into a zoomed sub-LFP range to resolve tiny action potentials. Thanks to the reduced DR, the oversampling rate of the IADC can be reduced by 64.3% compared to the conventional approach, leading to significant power reduction. In addition, dual-band recording can be easily attained because the scheme continuously tracks LFPs without additional on-chip hardware. A prototype four-channel front-end IC has been fabricated in 180 nm standard CMOS processes. The IADC achieved 11.3-bit ENOB at 6.8 μW, resulting in the best Walden and SNDR FoMs, 107.9 fJ/c-s and 162.1 dB, respectively, among two different comparison groups: the IADCs reported up to date in the state-of-the-art neural recording front-ends; and the recent brain recording ADCs using similar zooming or tracking techniques to this work. The intrinsic dual-band recording feature reduces the post-processing FPGA resources for subcortical signal band separation by >45.8%. The front-end IC with the zoom-and-track IADC showed an NEF of 5.9 with input-referred noise of 8.2 μVrms, sufficient for subcortical recording. The performance of the whole front-end IC was successfully validated through in vivo animal experiments.