Search

Conventional static radiative coolers have an unadjustable cooling capacity, which often results in overcooling in low temperature environment. Therefore, there is a great need for an adaptive dynamic radiative cooler. However, such adaptive coolers usually require complex preparation processes. This paper proposes an adaptive radiative cooler based on a Fabry–Perot resonant cavity. By optimizing the structural parameters of the radiative cooler, this adaptive radiative cooler achieves a modulation rate of 0.909 in the atmospheric window band. The net radiative cooling performance difference between low and high temperatures is nearly eight times. Meanwhile, the device is easily prepared, has a high tolerance, and can effectively prevent W–VO2oxidation. This study provides new insights into adaptive radiative cooling with potential for large-scale applications.

Accurately perceiving location and scene is crucial for autonomous driving and mobile robots. Recent advances in deep learning have made it possible to learn egomotion and depth from monocular images in a self-supervised manner, without requiring highly precise labels to train the networks. However, monocular vision methods suffer from a limitation known as scale-ambiguity, which restricts their application when absolute-scale is necessary. To address this, we propose SelfOdom, a self-supervised dual-network framework that can robustly and consistently learn and generate pose and depth estimates in global scale from monocular images. In particular, we introduce a novel coarse-to-fine training strategy that enables the metric scale to be recovered in a two-stage process. Furthermore, SelfOdom is flexible and can incorporate inertial data with images, which improves its robustness in challenging scenarios, using an attention-based fusion module. Our model excels in both normal and challenging lighting conditions, including difficult night scenes. Extensive experiments on public datasets have demonstrated that SelfOdom outperforms representative traditional and learning-based VO and VIO models.

Accurately and reliably positioning pedestrians in satellite-denied conditions remains a significant challenge. Pedestrian dead reckoning (PDR) is commonly employed to estimate pedestrian location using a low-cost inertial sensor. However, PDR is susceptible to drift due to sensor noise, incorrect step detection, and inaccurate stride length estimation. This work proposes ReLoc-PDR, a fusion framework combining PDR and visual relocalization using graph optimization. ReLoc-PDR leverages time-correlated visual observations and learned descriptors to achieve robust positioning in visually degraded environments. A graph optimization-based fusion mechanism with the Tukey kernel effectively corrects cumulative errors and mitigates the impact of abnormal visual observations. Real-world experiments demonstrate that our ReLoc-PDR surpasses representative methods in accuracy and robustness, achieving accurate and robust pedestrian positioning results using only a smartphone in challenging environments such as less-textured corridors and dark nighttime scenarios.

Overcoming ice accretion on external aircraft wing surfaces plays a crucial role in aviation, and developing environmentally friendly passive anti-icing surfaces is considered to be a promising strategy. Superhydrophobic surfaces (SHSs) have attracted increasing attention due to their potential advantages of keeping the airframe dry without causing additional aerodynamic losses. However, the passive anti-icing performances of SHSs reported to date varied a lot under different icing test conditions. Therefore, a systematic investigation is necessary to elucidate the icing conditions where SHSs can remain effective and pave the way for SHSs toward practical anti-icing applications. Herein, we designed and fabricated a typical type of SHS featuring dual-scale hierarchical structures with arrayed micromountains (with both spacings and heights of tens of micrometers) covered by single-scale sandy-corrugation-like periodic structures (with both spacings and heights of only several micrometers) (termed SS1). Its anti-icing performances under three representative icing conditions, including static water freezing, dynamic supercooled-droplet impinging, and icing wind tunnel conditions, were comparatively investigated. The SS1 SHS maintained a lower static ice-adhesion strength (<60 kPa even after 50 deicing cycles at temperatures as low as −25 °C), which was attributed to a cumulative cracking effect facilitating the ice detachment. Within the laboratory dynamic icing tests, the SS1 SHSs with micromountain heights of 20–30 μm performed optimally in the antiadhesion of supercooled droplets (at an impinging velocity of 3.4 m/s and temperatures of −5 to −25 °C). In spite of the significant anti-icing performances of the SS1 SHSs in both static and dynamic laboratory tests, they could hardly sustain reliable passive anti-icing performances in harsher icing wind tunnel tests with supercooled droplets impinging their surfaces at velocities of up to 50 m/s at a temperature of −5 °C for 10 min. This study can inspire the development of improved SHSs for achieving satisfactory anti-icing performances in real-aviation conditions.

Capillary-fed thin-film evaporation viamicro/nanoscale structures has attracted increasing attention for its high evaporation flux and pumpless liquid replenishment. However, maximizing thin-film evaporation has been hindered by the intrinsic trade-off between the heat flux and liquid transport. Here, we designed and fabricated nanostructured micro-steam volcanoes on copper surfaces featuring triple-level super-wicking routes to overcome this trade-off and boost water evaporation. The triple-level super-wicking routes enable the continuous formation of a 3D thin film for highly efficient evaporation by continuous self-driven liquid replenishment and extending the thin-film region. The micro-steam volcanoes increased the surface area by 225%, improving the evaporation rate by 141%, with a rapid self-pumping water transport speed up to 80 mm s−1. A remarkable solar-driven water evaporation rate of 3.33 kg m−2h−1under one sun vertical incidence was achieved, which is among the highest reported values for metal-based evaporators. When attached to electric-heating plates, the evaporator realized an electrothermal evaporation rate of 12.13 kg m−2h−1. Moreover, it can also be used for evaporative cooling with enhanced convective heat transfer, reaching a 36.2 °C temperature reduction on a heat source with a heat flux of 6 W cm−2. This study promises a general strategy for designing thin-film evaporators with high efficiencies, low costs, and multi-functional compatibilities.

Based on geometrical characteristics, all surface microstructures are categorized into two types: closed-cell and open-cell structures. Closed-cell structures are well-known to have more stable and durable superhydrophobicity at room temperatures. However, in low-temperature environments where massive environmentally induced physical changes emerge, whether closed-cell surfaces can maintain good anti-icing performances has not yet been confirmed, and thus how to design optimal superhydrophobic anti-icing microstructures is rarely reported. Here, we apply an ultrafast laser to fabricate superhydrophobic surfaces with tunable patterned micro-nanostructures from a complete closed-cell to different ratios and to a complete open-cell. We discover that droplets on closed-cell structures completely degrade to the high-adhesion Wenzel state after icing and melting cycles while those on the open-cell structures well recover to the original Cassie-Baxter state. We propose an improved ideal gas model to clarify the mechanisms that the decreased air pocket pressure and the air dissolution on closed-cell structures induce easy impalement during icing and the difficult recovery during melting, paving the way for optimizing the anti-icing structure design. The optimized open-cell surfaces exhibit over 33 times lower ice adhesion strengths (1.4 kPa) and long-term icephobic durability (<20 kPa after 33 deicing cycles) owing to the increased air pocket pressure at low temperatures. Significant dewetting processes during condensation endow the open-cell structures with more remarkable high-humidity resistance and anti-frosting properties. Our study reveals the general design principle of superhydrophobic anti-icing structures, which might guide the design of superhydrophobic anti-icing surfaces in practical harsh environments.

The wicking phenomenon, including wicking and hemiwicking, has attracted increasing attention for its critical importance to a wide range of engineering applications, such as thermal management, water harvesting, fuel cells, microfluidics, and biosciences. There exists a more urgent demand for anisotropic wicking behaviors since an increasing number of advanced applications are significantly complex. For example, special-shaped vapor chambers and heating atomizers in some electronic cigarettes need liquid replenishing with various velocities in different directions. Here, we report two-dimensional anisotropic hemiwicking behaviors with elliptical shapes on laser structured prismatic microgrooves. The prismatic microgrooves were fabricated via one-step femtosecond laser direct writing, and the anisotropic hemiwicking behaviors were observed when utilizing glycerol, glycol, and water as the test liquid. Specifically, the ratios of horizontal wicking distance in directions along short and long axes were tan 0°, tan 15°, tan 30°, and tan 45° for samples with cross-angles of 0°, 30°, 60°, and 90°, respectively. The vertical water wicking front displayed corresponding angles under the guidance of laser structured prismatic microgrooves. Theoretical analysis shows that the wicking distance is mainly dependent on the cross-angle θ and surface roughness, in which the wicking distance is proportional to cos(θ/2). Driven by the capillary pressure forming in the narrow microgrooves, the liquid initially filled the valleys of microgrooves and then surrounded and covered the prismatic ridges with laser-induced nanoparticles. The abundant nanoparticles increased the surface roughness, leading to the enhancement of wicking performance, which was further evidenced by the larger wicking speed of the sample with more nanoparticles. The mechanism of anisotropic hemiwicking behaviors revealed in this work paves the way for wicking control, and the proposed prismatic microgrooved surfaces with two-dimensional anisotropic hemiwicking performance and superhydrophilicity could serve in a broad range of applications, especially for the advanced thermal management with specific heat load configurations.

Perineural invasion (PNI) is a biological characteristic commonly observed in pancreatic cancer. Although PNI plays a key role in pancreatic cancer metastasis, recurrence, and poor postoperative survival, its mechanism is largely unclarified. Clinical sample analysis and endoscopic ultrasonographic elasticity scoring indicated that cancer-associated fibroblasts (CAFs) were closely related to the occurrence of PNI. Furthermore, CAF-derived extracellular vesicles (EVs) were involved in PNI in dorsal root ganglion coculture and mouse sciatic nerve models. Next, we demonstrated that CAFs promoted PNI through extracellular vesicle transmission of PNI-associated transcript (PIAT). Mechanistically, PIAT specifically bound to YBX1 and blocked the YBX1-Nedd4l interaction to inhibit YBX1 ubiquitination and degradation. Furthermore, PIAT enhanced the binding of YBX1 and PNI-associated mRNAs in a 5-methylcytosine (m5C)–dependent manner. Mutation of m5C recognition motifs in YBX1 or m5C sites in downstream target genes reversed PIAT-mediated PNI. Consistent with these findings, analyses using a KPC mouse model demonstrated that the PIAT/YBX1 axis enhanced PNI through m5C modification. Clinical data suggested that the PIAT expression in the serum EVs of patients with pancreatic cancer was associated with the degree of neural invasion and prognosis. Our study revealed the important role of the PIAT/YBX1 signaling axis in the tumor microenvironment (TME) in promoting tumor cell PNI and provided a new target for precise interference with CAFs and RNA methylation in the TME to suppress PNI in pancreatic cancer.

Dynamical models estimate and predict the temporal evolution of physical systems. State-space models (SSMs) in particular represent the system dynamics with many desirable properties, such as being able to model uncertainty in both the model and measurements, and optimal (in the Bayesian sense) recursive formulations, e.g., the Kalman filter. However, they require significant domain knowledge to derive the parametric form and considerable hand tuning to correctly set all the parameters. Data-driven techniques, e.g., recurrent neural networks, have emerged as compelling alternatives to SSMs with wide success across a number of challenging tasks, in part due to their impressive capability to extract relevant features from rich inputs. They, however, lack interpretability and robustness to unseen conditions. Thus, data-driven models are hard to be applied in safety-critical applications, such as self-driving vehicles. In this work, we present DynaNet, a hybrid deep learning and time-varying SSM, which can be trained end-to-end. Our neural Kalman dynamical model allows us to exploit the relative merits of both SSM and deep neural networks. We demonstrate its effectiveness in the estimation and prediction on a number of physically challenging tasks, including visual odometry, sensor fusion for visual-inertial navigation, and motion prediction. In addition, we show how DynaNet can indicate failures through investigation of properties, such as the rate of innovation (Kalman gain). [ABSTRACT FROM AUTHOR]

Due to the sparse rewards and high degree of environmental variation, reinforcement learning approaches, such as deep deterministic policy gradient (DDPG), are plagued by issues of high variance when applied in complex real-world environments. We present a new framework for overcoming these issues by incorporating a stochastic switch, allowing an agent to choose between high- and low-variance policies. The stochastic switch can be jointly trained with the original DDPG in the same framework. In this article, we demonstrate the power of the framework in a navigation task, where the robot can dynamically choose to learn through exploration or to use the output of a heuristic controller as guidance. Instead of starting from completely random actions, the navigation capability of a robot can be quickly bootstrapped by several simple independent controllers. The experimental results show that with the aid of stochastic guidance, we are able to effectively and efficiently train DDPG navigation policies and achieve significantly better performance than state-of-the-art baseline models. [ABSTRACT FROM AUTHOR]

Developing cost-effective and highly efficient oxygen evolution reaction (OER) electrocatalysts based on earth-abundant elements is vital to hydrogen production from electrocatalytic water splitting. Herein, a three-dimensional and in situ-activated electrocatalyst derived from stainless steel is successfully fabricated via a two-step laser direct writing strategy. The electrocatalyst appears in the form of nanoparticle-stacked porous clusters on the multiscale stainless steel with irregular microcone arrays and microspheres, which exposes more active sites and facilitates the mass transport. Especially, the clusters undergoe a self-optimizing morphological and compositional reconfiguration induced by the leaching of Cr species under OER conditions for favorable charge transfer and enhanced intrinsic catalytic activity. As a result, the in situ-activated, Ni/Cr-doped Fe3O4electrocatalyst exhibits an outstanding OER performance with a small overpotential of 262 mV to reach 10 mA cm–2, a low Tafel slope of 35.0 mV dec–1, and excellent long-term stability of 120 h, among the best spinel Fe-rich OER electrocatalysts. Finally, we also verify the feasibility of the affordable and efficient electrocatalyst coupled with the commercial Ni cathode in the practical water electrolysis. This work may open up a new avenue to design nanostructured metal oxides for various energy applications and beyond.

Evaporation concentration of target analytes dissolved in a water droplet based on superhydrophobic surfaces could be able to break the limits for sensitive trace substance detection techniques (e.g. SERS) and it is promising in the fields such as food safety, eco-pollution, and bioscience. In the present study, polytetrafluoroethylene (PTFE) surfaces were processed by femtosecond laser and the corresponding processing parameter combinations were optimised to obtain surfaces with excellent superhydrophobicity. The optimal parameter combination is: laser power: 6.4 W; scanning spacing: 40 μm; scanning number: 1; and scanning path: 90 degree. For trapping and localising droplets, a tiny square area in the middle of the surface remained unprocessed for each sample. The evaporation and concentration processes of droplets on the optimised surfaces were performed and analyzed, respectively. It is shown that the droplets with targeted solute can successfully collect all solute into the designed trapping areas during evaporation process on our laser fabricated superhydrophobic surface, resulting in detection domains with high solute concentration for SERS characterisation. It is shown that the detected peak intensity of rhodamine 6G with a concentration of 10−6min SERS characterisation can be obviously enhanced by one or two orders of magnitude on the laser fabricated surfaces compared with that of the unprocessed blank samples.

Lymphangiogenesis is considered the rate-limiting step for tumor lymph node (LN) metastasis by sprouting the new lymphatic network from preexisting lymphatic vessels to favor the invasion and entry of cancer cells into the lymphatic system. Over the past decade, numerous studies have investigated the cellular roles and molecular mechanisms of lymphangiogenesis in cancers, and corresponding molecular control strategies targeting lymphangiogenesis to treat LN metastatic cancers have been developed. Nevertheless, due to the crucial role of lymphatic vessels in the infiltration of immune cells and immune activation in cancers, there have been several diverse findings regarding the biological roles and strategies of targeting tumor lymphangiogenesis as cancer immunotherapy has emerged and rapidly developed. Herein, we summarize the progress in characterizing the molecular mechanisms and therapeutic strategies involved in tumor lymphangiogenesis and highlight the potential of targeting lymphangiogenesis in combination with immunotherapy for the treatment of cancer, providing opportunities to improve the efficacy of targeting lymphangiogenesis to prevent cancer metastasis.

Castration-resistant prostate cancer (CRPC) that occurs after the failure of androgen deprivation therapy is the leading cause of deaths in prostate cancer patients. Thus, there is an obvious and urgent need to fully understand the mechanism of CRPC and discover novel therapeutic targets. Long noncoding RNAs (lncRNAs) are crucial regulators in many human cancers, yet their potential roles and molecular mechanisms in CRPC are poorly understood. In this study, we discovered that an lncRNA HOXD-AS1 is highly expressed in CRPC cells and correlated closely with Gleason score, T stage, lymph nodes metastasis, and progression-free survival. Knockdown of HOXD-AS1 inhibited the proliferation and chemo-resistance of CRPC cells in vitro and in vivo. Furthermore, we identified several cell cycle, chemo-resistance, and castration-resistance-related genes, including PLK1, AURKA, CDC25C, FOXM1, and UBE2C, that were activated transcriptionally by HOXD-AS1. Further investigation revealed that HOXD-AS1 recruited WDR5 to directly regulate the expression of target genes by mediating histone H3 lysine 4 tri-methylation (H3K4me3). In conclusion, our findings indicate that HOXD-AS1 promotes proliferation, castration resistance, and chemo-resistance in prostate cancer by recruiting WDR5. This sheds a new insight into the regulation of CRPC by lncRNA and provides a potential approach for the treatment of CRPC.

Reducing unfavorable ice accretion on surfaces exposed in cold environment requires effective passive anti-icing/deicing techniques. Icephobic surfaces are widely applied on various infrastructures due to their low ice adhesion strength and flexibility, whereas their poor mechanical durability, common liquid infusion, weak resistance to contamination, and low bonding strength to substrates are the major remaining challenges. According to the fracture mechanics of ice layer, initiating cracks at the ice-solid interfaces via the proper design of internal structures of icephobic materials is a promising way to icephobicity. Herein, a crack initiating icephobic surface with porous PDMS sponges sandwiched between a protective, dense PDMS layer and a textured metal microstructure was proposed and fabricated. The combination of high- and low- stiffness PDMS layers anchored by the structured metal surface give the sandwich-like structure excellent icephobicity with both high durability and low ice adhesion (5.3 kPa in the icing–deicing cycles). The porosity and the elastic modulus of the PDMS sponges and the periodicity of the metal surface structures can both be tailored to realize enhanced icephobicity. The sandwich-like icephobic surface remained insignificantly changed under solid particle impacting and the durability characterized via linear abrasion tests was elevated compared with PDMS coating on flat metal surfaces. Additionally, the trilayer icephobic surface possesses durability, low ice adhesion strength, and improved resistance to contamination and is applicable on various surfaces.

Precise patterning with microscale lateral resolution and widely tunable heights is critical for integrating colloidal nanocrystals into advanced optoelectronic and photonic platforms. However, patterning nanocrystal layers with thickness above 100 nm remains challenging for both conventional and emerging direct photopatterning methods, due to limited light penetration depths, complex mechanical and chemical incompatibilities, and others. Here, we introduce a direct patterning method based on a thermal mechanism, namely, the thermally activated ligand chemistry (or TALC) of nanocrystals. The ligand cross-linking or decomposition reactions readily occur under local thermal stimuli triggered by near-infrared lasers, affording high-resolution and nondestructive patterning of various nanocrystals under mild conditions. Patterned quantum dots fully preserve their structural and photoluminescent quantum yields. The thermal nature allows for TALC to pattern over 10 μm thick nanocrystal layers in a single step, far beyond those achievable in other direct patterning techniques, and also supports the concept of 2.5D patterning. The thermal chemistry-mediated TALC creates more possibilities in integrating nanocrystal layers in uniform arrays or complex hierarchical formats for advanced capabilities in light emission, conversion, and modulation.

Anti-icing superhydrophobic surfaces have attracted tremendous interests due to their repellency to water and extremely low ice affinity, whereas the weak durability has been the bottleneck for further applications. Surface durability is especially important in long-term exposure to low-temperature and high-humidity environments. In this study, a robust micro–nano-nanowire triple structure-held PDMS superhydrophobic surface was fabricated via a hybrid process: ultrafast-laser-prepared periodic copper microstructures were chemically oxidized, followed by modification of PDMS. The hedgehog-like surface structure was composed of microcones, densely grown nanowires, and tightly combined PDMS. The capillary force difference in micro–nanostructures drove PDMS solutions to distribute evenly, bonding fragile nanowires to form stronger composite cones. PDMS replaced the commonly used fragile fluorosilanes and protected nanowires from breaking, which endowed the surfaces with higher robustness. The ductile PDMS–nanowire composites possessed higher resiliency than brittle nanowires under a load of 1 mN. The surface kept superhydrophobic and ice-resistant after 15 linear abrasion cycles under 1.2 kPa or 60 icing–deicing cycles under −20 °C or 500 tape peeling cycles. Under a higher pressure of 6.2 kPa, the contact angle (CA) was maintained above 150° until the abrasion distance exceeded 8 m. In addition, the surface exhibited a rare spontaneously optimized performance in the icing–deicing cycles. The ice adhesion strength of the surface reached its lowest value of 12.2 kPa in the 16th cycle. Evolution of surface roughness and morphology were combined to explain its unique U-shaped performance curves, which distinguished its unique degradation process from common surfaces. Thus, this triple-scale superhydrophobic surface showed a long-term anti-icing performance with high deicing robustness and low ice adhesion strength. The proposed nanostructure-facilitated uniform distribution strategy of PDMS is promising in future design of durable superhydrophobic anti-icing surfaces.

WD repeat domain 5 (WDR5) plays an important role in various biological functions through the epigenetic regulation of gene transcription (Wysocka et al., 2005 [1]; Sandstrom et al., 2014[2]; Ang et al., 2011[3]). Recently, our study found that WDR5 was upregulated in bladder cancer tissues, promoted bladder cancer cell proliferation, self-renewal and chemoresistance to cisplatin in bladder cancer cells in vitro, and tumor growth in vivo(Chen et al., 2015). To gain a molecular understanding of the role of WDR5 in promoting bladder cancer, we performed a genome-wide analysis on WDR5 knockdown by microarray gene expression profiling. Here we provide detailed experimental methods and analysis for the microarray data, which have been deposited into Gene Expression Omnibus (GEO): GSE59132.

Lymphatic metastasis constitutes a leading cause of recurrence and mortality in bladder cancer. Accumulating evidence indicates that lymphangiogenesis is indispensable to trigger lymphatic metastasis. However, the specific mechanism is poorly understood. In the present study, we revealed a pathway involved in lymphatic metastasis of bladder cancer, in which a circular RNA (circRNA) facilitated lymphangiogenesis in a vascular endothelial growth factor C (VEGF-C) independent manner. Novel circRNA circEHBP1 was markedly upregulated in bladder cancer and correlated positively with lymphatic metastasis and poor prognosis of patients with bladder cancer. CircEHBP1 upregulated transforming growth factor beta receptor 1 (TGFBR1) expression through physically binding to miR-130a-3p and antagonizing the suppression effect of miR-130a-3p on the 3’ UTR region of TGFBR1. Subsequently, circEHBP1-mediated TGFβR1 overexpression activated the TGF-β/SMAD3 signaling pathway, thereby promoting the secretion of VEGF-D and driving lymphangiogenesis and lymphatic metastasis in bladder cancer. Importantly, administration of VEGF-D neutralizing antibodies remarkably blocked circEHBP1-induced lymphangiogenesis and lymphatic metastasis in vivo. Our findings highlighted that the circEHBP1/miR-130a-3p/TGFβR1/VEGF-D axis contributes to lymphangiogenesis and lymphatic metastasis of bladder cancer independent of VEGF-C, which might lead to the development of circEHBP1 as a potential biomarker and promising therapeutic target for lymphatic metastasis in bladder cancer.

Supplemental Digital Content is available in the text