Nanoscience News (<front>)

Title tbc

Tbc

• Speaker: Joni Teräväinen (University of Cambridge)
• Wednesday 28 May 2025, 13:30-15:00
• Venue: MR4, CMS.
• Series: Discrete Analysis Seminar; organiser: Julia Wolf.

Add to your calendar or Include in your list

Deep Learning many-body electronic structure

Deep learning with neural network wave functions is a powerful new approach for calculating the many-body electronic structure of molecules, materials, and physical models, addressing shared problems of strongly correlated electronic structure in fields of condensed matter physics and chemistry. In this talk, I will provide an overview of the many-body electronic structure problem and the development of related deep learning methods, and discuss our recent progresses, including a series of algorithmic developments, accurate ab initio solutions of molecular and solids, and exploring the fractional quantum Hall effect and correlated topological states of matter.

• Speaker: Prof. Ji CHEN (Peking University)
• Friday 06 June 2025, 14:00-15:30
• Venue: Seminar Room 3, RDC.
• Series: Theory of Condensed Matter; organiser: Bo Peng.

Add to your calendar or Include in your list

Title to be confirmed

Abstract not available

• Speaker: Prof. Ji CHEN (Peking University)
• Friday 06 June 2025, 14:00-15:30
• Venue: Seminar Room 3, RDC.
• Series: Theory of Condensed Matter; organiser: Bo Peng.

Add to your calendar or Include in your list

Understanding the drivers of antimicrobial resistance evolution

The evolution of antimicrobial resistance is a growing threat to human lives that is becoming progressively harder to manage. This is because there are many different mechanisms that allow bacteria to develop resistance to antimicrobial treatments. In this talk, I will explore the complex dynamics of genetic, phenotypic and mobile resistance mechanisms in microbial populations. I will show under which conditions we would expect genetic or phenotypic antimicrobial resistance to determine pathogen success and how this leads to different mutational pathways during chronic and acute infections. Further, I will discuss the genetic and environmental conditions that determine the spread of mobile resistance elements in heterogeneous bacterial populations. Overall, I aim to demonstrate the complexity of microbial resistance dynamics but that there is hope to understand them by studying their mechanistic interactions.

The seminar will be held in JDB Seminar Room, Department of Engineering, and online (zoom): https://newnham.zoom.us/j/92544958528?pwd=YS9PcGRnbXBOcStBdStNb3E0SHN1UT09

• Speaker: Claudia Igler, University of Manchester
• Thursday 29 May 2025, 14:00-15:00
• Venue: JDB Seminar Room, Department of Engineering and online (Zoom).
• Series: CUED Control Group Seminars; organiser: Fulvio Forni.

Add to your calendar or Include in your list

Prebiotic Chemistry, Exoplanets and Stellar Flaring

Abstract not available

• Speaker: Lukas Rossmanith
• Tuesday 03 June 2025, 11:15-12:00
• Venue: Martin Ryle Seminar Room, Kavli Institute.
• Series: Hills Coffee Talks; organiser: David Buscher.

Add to your calendar or Include in your list

Energy Environ. Sci., 2025, Advance Article
DOI: 10.1039/D5EE00227C, PaperDigen Ruan, Yanru Wang, Jiasen Guo, Zhuangzhuang Cui, Qingshun Nian, Zhihao Ma, Dazhuang Wang, Jiajia Fan, Jun Ma, Bingqing Xiong, Qi Dong, Ruiguo Cao, Shuhong Jiao, Xiaodi Ren
An oscillating Li+-acceptor fluorine-donor electrolyte incorporating a new co-solvent with an asymmetric super-lithiophilic fluorine group resolves the dilemma between fast electrolyte Li+ transport and stable interphases.
To cite this article before page numbers are assigned, use the DOI form of citation above.
The content of this RSS Feed (c) The Royal Society of Chemistry

Positional encodings in LLMs

Positional encodings are essential for transformer-based language models to understand sequence order, yet their influence extends far beyond simple position tracking. This talk explores the landscape of positional encoding methods in LLMs and reveals surprising insights about how these architectural choices shape model behavior.

We begin with the fundamental challenge: why attention mechanisms require explicit positional information. We then survey the evolution of encoding strategies, from sinusoidal approaches to modern techniques like ALiBi and RoPE, examining their architectural implications and trade-offs.

The talk delves into how these different encoding strategies fundamentally shape model architectures and representations. We analyze the specific limitations and trade-offs of each approach, examining how positional information propagates through transformer layers and influences the learned representations.

• Speaker: Valeria Ruscio
• Thursday 04 June 2026, 17:00-17:45
• Venue: Lecture Theatre 2, Computer Laboratory, William Gates Building.
• Series: Foundation AI; organiser: Pietro Lio.

Add to your calendar or Include in your list

Energy Environ. Sci., 2025, Accepted Manuscript
DOI: 10.1039/D5EE01686J, PaperLiang Zeng, Rong Hu, Ming Zhang, Seunglok Lee, Qingyuan Wang, Shixin Meng, Qi Chen, Jiangang Liu, Lingwei Xue, Liwei Mi, Changduk Yang, Zhi-Guo Zhang
Tethered small-molecule acceptors (SMAs), featuring multiple SMA subunits connected to an aromatic core via flexible chains, effectively suppressing thermodynamic relaxation and enhancing stretchability in polymer solar cells (PSCs). However, these...
The content of this RSS Feed (c) The Royal Society of Chemistry

AI Meets Economics: The Case of Auction-Assisted AI Systems in Cloud-Edge Continuum

Join via MS Teams

Many cloud and edge AI services today perform machine learning inference in real time on end user requests. Over time, however, models could degrade in accuracy due to data and concept drifts, and full retraining can be infeasible because of limited training data, long training delay, and prohibitive computational overhead. A promising solution is for the AI service to incorporate externally supplied pre‑trained models to maintain resilience and accuracy in the face of evolving inputs. To incentivize third‑party model providers, who alone possess the requisite resources and data, to produce and contribute models, an economic mechanism is required to monetize their contributions. Auction formats naturally suggest themselves, yet they introduce fundamental challenges in this circumstance: the interdependence of sequential auctions, the trade‑off between system overhead and inference performance, and the need to balance economic properties with sustained participation. In this talk, firstly, I will formulate the repeated model‑procurement auctions as a non‑linear mixed‑integer social cost minimization problem, design a suite of polynomial‑time approximation algorithms that jointly solve this problem in an online manner, and describe the multiple performance guarantees of our approach, including per‑auction truthfulness and individual rationality, an upper bound on inference loss, and a parameterized‑constant competitive ratio for social cost, all supported by empirical evaluations. Afterwards, I will briefly survey our other efforts on auction‑assisted AI systems, including edge AI inference over auctioned resources and foundation model fine‑tuning with auction‑based pricing. Finally, I will conclude with a vision for future research.

Biography: Lei Jiao received his Ph.D. in computer science from the University of Göttingen, Germany, in 2014. He is currently a faculty member at the University of Oregon, USA , and was previously a member of the technical staff at Nokia Bell Labs, Ireland. He researches networking and distributed computing, spanning AI infrastructures, cloud/edge networks, energy systems, cybersecurity, and multimedia. His work integrates mathematical methods from optimization, control theory, machine learning, and economics. He has authored over 80 peer-reviewed publications in journals such as IEEE Transactions on Networking, IEEE Transactions on Mobile Computing, IEEE Transactions on Parallel and Distributed Systems, and IEEE Journal on Selected Areas in Communications, and in conferences such as INFOCOM , MOBIHOC, ICDCS , SECON, ICNP , ICPP, and IPDPS , garnering over 6,000 citations according to Google Scholar. He is a recipient of the U.S. National Science Foundation CAREER Award, the Ripple Faculty Fellowship, the Alcatel-Lucent Bell Labs UK and Ireland Recognition Award, and several Best Paper Awards including those from IEEE CNS 2019 and IEEE LANMAN 2013 . He has served in various program committee roles, including as a track chair for ICDCS , as a member for INFOCOM , MOBIHOC, ICDCS , and WWW , and as a chair for multiple workshops with INFOCOM and ICDCS .

• Speaker: Prof. Lei Jiao
• Friday 23 May 2025, 15:00-16:00
• Venue: Computer Lab, FW11 and Online (MS Teams link below).
• Series: Computer Laboratory Systems Research Group Seminar; organiser: Richard Mortier.

Add to your calendar or Include in your list

A multi-generational compositional design creates a printable low-loss dielectric composite that achieves over 16 W m−1 K−1. This breakthrough is enabled by thermal post-processing, which promotes templated crystallization in a polymer matrix from surface-modified particles, creating a “hetero-percolated network.” The resulting material is three dimensionally printed into heatsinks that perform as effectively as metallic heatsinks while being electrically insulative and  transparent to radio frequency signals.

Abstract

As electronic devices become simultaneously more powerful and compact, thermal management is increasingly critical. Optimizing components like heatsinks is increasingly required, which has recently leveraged additive manufacturing. There is growing demand to move away from incumbent metallic materials for dielectric materials that are electrically insulative and  transparent to radio frequency signals. Thermally conductive polymer composites containing phonon-conducting ceramics offer a low-density dielectric solution compatible with fused filament fabrication. However, these materials have struggled to exceed thermal conductivities of 4 W m−1 K−1 due to challenging rheological flow effects at high filler volume fractions that prevent stable material extrusion. In this work, multi-generational compositional design is conducted to develop a printable low-loss dielectric composite that achieves over 16 W m−1 K−1, comparable to stainless steel. This breakthrough is enabled by thermal post-processing, which promote templated crystallization in a poly-lactic acid  matrix from surface-modified boron nitride platelets, creating a “hetero-percolated network”. The resulting material is three dimensionally printed into heatsinks that perform as effectively as metallic heatsinks while being electrically insulative and RF transparent.

This work presents a computationally guided approach to discovering stable amorphous ceramics, using yttrium tungsten nitride (Y-W-N) as a prototype. By combining first-principles random structure sampling with thin-film synthesis, the authors identify and experimentally realize previously unreported amorphous nitrides, demonstrating their exceptional thermal stability, tunable properties, and superior diffusion barrier performance.

Abstract

Amorphous materials offer unique functional characteristics, which are not observed in their crystalline counterparts making them invaluable for many applications in science and technology, such as electronic and optical devices, solid-state batteries, and protective coatings. However, finding compositions that are stable against crystallization and/or phase separation, and at the same time offer the needed functionality in the amorphous phase is still largely done by serendipity or trial-and-error. In this work, using yttrium tungsten nitride as a prototype, it is shown how computational random structure sampling provides a robust method to identify compositions that exhibit a highly corrugated potential energy surface with many narrow local minima, which are consequently hard to crystallize and remain stable in the amorphous phase. Synthesis experiments prove that the predicted nitride is readily synthesized in an amorphous phase with no detectable precipitates. High-throughput and conventional characterization of structural, physical, and functional properties of the discovered amorphous nitride compound reveal its attractive properties and possible application potential. The proposed workflow combining theory and experiment is broadly applicable to the discovery of a wide range of amorphous ceramic materials, paving the way for advanced amorphous materials for diverse emerging technologies.

Modulation of the N-alkyl chain length in polysulfobetaine-based zwitterionic micelles enables precise tuning of nanocarrier interactions with proteins and cell membranes. The optimized micelles comprising poly(sulfobetaine methacrylate)-block-poly(ɛ-caprolactone) with N-butyl substituent (PSB4-PCL), which are protein-resistant yet cell membrane-philic, achieve a unique balance between prolonged circulation and efficient tumor transcytosis, leading to enhanced intratumoral accumulation, deep penetration, and potent antitumor efficacy.

Abstract

Long blood circulation and fast cellular uptake are essential yet paradoxical requirements for efficient tumor-targeted drug delivery carriers. For instance, polyzwitterions, generally nonfouling to proteins and cells, have been extensively explored as long-circulating drug delivery carriers but suffer ultraslow cell internalization, making them inefficient in delivering drugs to cells. Protein-resistant yet cell membrane-binding polymers will simultaneously achieve long blood circulation and fast cellular internalization, but their designs are generally complicated, such as introducing cell-membrane binding groups. Here, it is shown that the N-alkyl chain length of zwitterionic poly(sulfobetaine) can be used to tune its affinity toward proteins and cell membranes. A poly(sulfobetaine) with a moderately long N-alkyl chain became cell membrane-philic while retaining protein resistance, leading to long blood circulation and fast cellular uptake, which further triggered efficient tumor cell transcytosis and intratumor penetration. Thus, its paclitaxel (PTX)-loaded micelles demonstrated potent antitumor efficacy in triple-negative breast cancer models. This study showcases a paradigm of designing polyzwitterions harmonizing long blood circulation and fast cellular uptake properties as tumor-active drug delivery carriers.

Field-effect passivation is achieved by selectively extruded tetraphenylphosphonium cations for suppressing the interfacial non-radiative recombination, yielding an improved power conversion efficiency of 21.0% for hole transport layer-free carbon-based printable mesoscopic perovskite solar cells.

Abstract

Mesoporous electron transport layer (ETL) in printable mesoscopic perovskite solar cells (p-MPSCs) enables rapid and selective extraction of photogenerated electrons and facilitates device fabrication without a hole transport layer (HTL). However, the inherent mesoporous architecture introduces abundant interfacial defects that promote undesired non-radiative recombination, limiting the power conversion efficiency (PCE). To address this challenge, an interface field-effect passivation strategy is implemented, leveraging spatially selective cation extrusion. By incorporating tetraphenylphosphonium cations, sterically bulky organic ions that migrate to the perovskite/ETL interface during crystallization, a robust interfacial electrostatic field is introduced. This field simultaneously suppresses the non-radiative recombination by inducing field-effect passivation and enhances the charge extraction through optimizing energy alignment. The synergistic effects yield a PCE enhancement from 19.4% to 21.0%. This work underscores the potential of cation-engineered interfacial fields to improve the performance of HTL-free carbon-electrode perovskite photovoltaics.

Schematic illustration of a low energy synthesized FePc–MOF–MXene nanozyme-based microneedle patch coupled with an iontophoretic sampling interface for real-time in vivo L-Cysteine detection. The nanozyme enhances electron transfer and catalytic efficiency, while the microneedles and iontophoretic patch enable minimally invasive access to interstitial fluid. The platform demonstrates continuous, on-body monitoring performance in a live murine model.

Abstract

Advancing clinical diagnostics requires platforms that combine catalytic efficiency, biocompatibility, and real-time, in vivo accessibility. Herein, this study reports a structurally integrated FePc–ZIF-8–MX nanozyme that combines the redox activity of FePc, the porous confinement of ZIF-8, and the electrical conductivity of MX. Synthesized via a low-energy, ambient-condition process, this hybrid enables efficient electron transfer, enhanced analyte enrichment, and sustained catalytic activity in physiological environments. To translate this functionality into a wearable diagnostic format, the hybrid is seamlessly incorporated into a microneedle array, offering minimally invasive access to interstitial fluid for continuous L-cysteine (L-Cys) monitoring. The resulting platform exhibits high selectivity and sensitivity across complex biological matrices, including serum, urine, cultured cells, and a murine model of myocardial infarction. This study presents a multifunctional electrochemical platform that enables on-body metabolite monitoring through a microneedle-integrated nanozyme interface. To the best of our knowledge, it constitutes the first realization of real-time, in vivo L-Cys sensing in this format, setting a new benchmark for precision biosensing in translational healthcare.

Elemental iodine (I2) is used to create an iodine-rich environment for inhibiting the formation of iodine vacancy defects. This strategy also modifies the perovskite's surface energy, regulating its crystallization kinetics and resulting in a more well-crystallized perovskite with a preferred (001)-orientation. Consequently, PeLEDs with efficiencies of 32.5% for deep-red (678 nm) and 29.5% for pure-red (649 nm) have been achieved.

Abstract

Perovskite light-emitting diodes (PeLEDs) face a tough challenge that halogen vacancy defects limit device performance, while the introduction of additional agents to passivate defects may potentially compromise the stability of the structure and increase the complexity of the system. Here, elemental iodine (I2) is employed as an additive, utilizing its ability to create an iodine-rich condition and to transform into I− ions for passivating iodine vacancy defects, while its volatile nature ensures no residue and avoids the introduction of extraneous elements. This approach also alters the perovskite's surface energy and subsequently regulates its crystallization kinetics, which results in a more well-crystallized perovskite with vertically-aligned organic spacer layers, in turn promoting the transport of charge carriers. On the basis of this strategy, PeLEDs with efficiencies of 32.5% and 29.5% for deep-red (678 nm) and pure-red (649 nm) have been achieved, respectively.

A multifunctional dual-anion synergistic strategy that combines acetate anions (Ac−) from formamidinium acetate and thiocyanate anions (SCN-) from guanidinium thiocyanate to produce high-quality MA-free Cs0.1FA0.9Pb0.5Sn0.5I3 perovskite films via a simplified antisolvent-free spin-coating process. Ac− anions retard crystallization and mitigate Sn2+ oxidation through the formation of intermediate phases and anion exchange process, while SCN- anions further promote crystal growth and stabilize Sn2+ via strong coordination interactions, improving energy level alignment and grain boundary management.

Abstract

Mixed tin-lead (Sn-Pb) perovskites are integral to all-perovskite tandem solar cells (TSCs), offering significant potential to surpass the theoretical efficiency limits of single-junction solar cells. However, the rapid crystallization of Sn-Pb perovskite thin films and the propensity of Sn2+ to oxidize into Sn4+ remain critical challenges, hindering device performance and stability. Herein, it is demonstrated that a multifunctional dual-anion synergistic regulation strategy to fabricate high-quality MA-free Cs0.1FA0.9Pb0.5Sn0.5I3 perovskite thin films with superior morphology and crystallinity via a simplified antisolvent-free spin-coating process. Acetate anions (Ac−) derived from formamidinium acetate (FAAc) effectively regulate crystallization kinetics and mitigate Sn2+ oxidation via intermediate phase formation and anion exchange process. Simultaneously, the combination of Ac− and thiocyanate anions (SCN−) from guanidinium thiocyanate (GuaSCN) promotes larger crystal grain growth and stabilizes Sn2+ via strong coordination interactions. The dual-anion strategy effectively minimizes grain boundaries, suppresses non-radiative recombination, and optimizes the energy level alignment at interfaces. As a result, the champion single-junction Sn-Pb perovskite solar cell (PSC) achieves an impressive power conversion efficiency (PCE) of 23.26%, setting a new benchmark for Sn-Pb PSCs fabricated without antisolvent. While all-perovskite TSCs reach 28.07% efficiency with remarkable operational stability, retaining 81% of initial performance after 600 h under maximum power point tracking.

Researchers developed ultra-stable near-infrared perovskite LEDs with peak external quantum efficiencies (EQEs) of ≈24.7% and EQEs of >20% from 70 to 1200 mA cm−2, resulting in an ultra-high peak radiance of 2270 W sr−1 m−2. Furthermore, these PeLEDs exhibit outstanding operational stability at high current densities. The high stability is enabled by a multifunctional stabilizer containing formamidine groups, which prevent phase transition and decomposition of the α-FAPbI3 perovskite under elevated temperatures.

Abstract

Perovskite light-emitting diodes (PeLEDs) have emerged as a promising candidate for next-generation display technologies, owing to their high efficiency and color purity. However, the operational instability of PeLEDs at high current densities (>100 mA cm−2) remains a significant challenge. Here, near-infrared (≈797 nm) PeLEDs are reported with peak external quantum efficiencies (EQEs) of ≈24.7% and EQEs of >20% across a wide range of current densities (70–1200 mA cm−2), resulting in an ultra-high peak radiance of 2270 W sr−1 m−2. These PeLEDs exhibit outstanding operational stability with operational lifetimes (T 50) of 6.6, 12.7, 19.2, 49.5, 238.6, and 350 h at current densities of 500, 400, 300, 200, 100, and 50 mA cm−2, respectively. The high stability is enabled by a multifunctional stabilizer containing formamidine groups, which prevent phase transition and decomposition of the α-FAPbI3 perovskite under elevated temperatures. This work demonstrates the feasibility of achieving efficient and stable PeLEDs at high current densities, providing strategies for the development of high-power optoelectronic devices based on halide perovskites.

This work presents a solid-solution strategy to modulate the mechanoluminescence of the composite elastomer with the physical principles proposed. With no need for power supply and circuit design, the developed CaBa4(PO4)3Cl:Eu/PDMS elastomer yields bright and repeatable light for over 20 000 cycles under various rapid and continuous stretching conditions, significantly breaking the performance limitations of the existing materials.

Abstract

Mechanoluminescence (ML) elastomer shows tremendous potential for the next generation of flexible displaying, imaging, and sensing devices. However, inadequate repeatability and cyclic stability are the current bottlenecks. In this work, a solid-solution strategy is reported to regulate the interfacial interactions of ML elastomer, in which the (Ca,Ba)5(PO4)3Cl:Eu solid solutions and polydimethylsiloxane (PDMS) are employed, respectively. The results suggest that the solid-solution strategy can effectively modulate the energy position of the valence band and the charge density distributions of the lowest unoccupied band, as well as the interfacial triboelectrification. Consequently, the ML performance of the composite elastomer in terms of intensity, spectral characteristic, and repeatability has been significantly improved. Particularly, the optimum CaBa4(PO4)3Cl:Eu/PDMS elastomer exhibits stable and repeatable ML for over 20 000 cycles under various rapid and continuous stretching conditions. This work not only provides a highly repeatable and cyclically stable ML elastomer but also clarifies the intrinsic physical principles, showing high guiding significance for future ML design and applications.

Energy Environ. Sci., 2025, Accepted Manuscript
DOI: 10.1039/D5EE00618J, Paper Open Access &nbsp This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.Saurabh Parab, Jonathan Lee, Matthew Miyagishima, Qiushi Miao, Bhargav Bhamwala, Alex Liu, Louis Ah, Bhagath Sreenarayanan, Kun Ryu, Mingqian Li, Neal Arakawa, Robert D. Schmidt, Mei Cai, Fang Dai, Ping Liu, Shen Wang, Ying Shirley Meng
Lithium-sulfur (Li-S) batteries hold significant promise for electric vehicles and aviation due to their high energy density and cost-effectiveness. However, understanding the root causes of performance degradation remains a formidable...
The content of this RSS Feed (c) The Royal Society of Chemistry

Nature Materials, Published online: 23 May 2025; doi:10.1038/s41563-025-02215-9

Treatments for traumatic brain injury are lacking owing to the limited regenerative capacity of neurons. Now, ultrasound-activated piezoelectric nanostickers that attach to cell membranes are shown to promote the neuronal differentiation of transplanted stem cells, leading to substantial brain tissue repair in rats with traumatic brain injury.