To date, neuroimaging has provided a wealth of information on how the human brain works in health and disease. With functional magnetic resonance imaging (fMRI), we can obtain spatially precise information about long-lasting brain activations whereas electro- and magnetoencephalography (EEG/MEG) can track transient cortical responses at millisecond resolution. However, none of these methods excel in time-resolved detection of sustained cortical activations, which are typically reflected as bursts of gamma-range (30–150 Hz) oscillations, frequently present in invasive recordings in patients. Although we have recently demonstrated that in exceptional situations MEG can detect even single gamma responses, their signal-to-noise ratio is usually prohibitively low, largely due to the substantial distance (4–5 cm) between cortex and sensors. Here, I propose to exploit recent advances in a novel magnetic sensor technology—atomic magnetometry—to construct a new kind of MEG system that allows capturing cerebral magnetic fields within millimetres from the scalp. Our simulations show that this proximity leads up to a 5-fold increase in the signal amplitude and an order-of-magnitude improvement of spatial resolution compared to conventional MEG. Therefore, a high-resolution MEG (HRMEG) system based on atomic magnetometers should enable non-invasive recordings of cortical activity at unprecedented sensitivity and detail level, which I propose to capitalize on by characterizing cortical responses, particularly gamma oscillations, during complex cognitive tasks. Additionally, since atomic magnetometers can recover within milliseconds from fields of several tesla, I also propose to combine transcranial magnetic stimulation (TMS) with MEG, leveraging the reciprocity of TMS and MEG and thus allowing better-than-ever characterization of TMS-evoked responses. This proposal comprises the research towards a HRMEG system and its application to study the working human brain in a new way.

Phase change materials (PCMs) has attracted much attention of researchers during the past two decades. However, current PCMs still have seriously potential problems on the fire safety mainly due to the flammability of the PCM substances. In this context, fire safe PCMs offer some advantages over conventional PCMs, leading to safe PCMs become very promising candidate for advanced energy saving field; but they still have a lot of challenges. In this project, aiming at solving some key scientific problems of PCMs and providing theoretical basis to the development of new generation fire safe PCM, based on the sophisticatedly structure design and fabrication the fellow will focus on the study of advanced (Sustainable, form-stable, and fire-safe) PCMs. The main objectives include: (1) To design and realize support materials for phase change materials with low cost, biomass source, good compatibility with organic PCM substances, uniform open pores, and convenience to realize large-scale preparation; (2) To design and prepare biobased phase change materials with fire safety, high thermal conductivity, and form stability: realize adjustable phase change temperature and ultra-high phase change enthalpy; (3) Develop the practical application of this new type of high-performance PCM in solar energy harvesting and storage, energy conservation and temperature regulation abilities; and (4) To expand the application of PLA, ionic liquid, other bio-based molecules (e.g., phytic acid, tannic acid, Chitosan, and Sodium alginate) in the new generation of PCMs. This proposal presents an interdisciplinary and innovative approach with great significance. These original ideas will bring us bran-new multi-functional and sustainable PCMs. It also provides a practical technical approach for the comprehensive design of PCMs.

The continuous growth of global data traffic pushes the data center technology boundaries and their energy consumption. Data centers already use more than 2% of the global electricity and are expected to consume 20% by 2025. The new generation hyperscale data centers offer a more energy-efficient alternative to conventional data centers but are limited by the current technology restrictions. Silicon photonics has been deemed to solve current bottlenecks for building green hyperscale data centers. However, this technology is challenged by the complicated and expensive fabrication methods of the laser sources, which represent 40% cost of the entire silicon transceiver market. Using a simple fabrication method invented in our ERC Advanced project, we have designed a waveguide light source that can be fabricated onto the silicon-integrated platform with a single-step, straightforward, low-cost, low-temperature and scalable process that lowers the fabrication cost ~10 times compared to the state-of-the-art light sources. Such technologic breakthrough will push the implementation of silicon photonics in hyperscale data centers, allowing them to maintain a low energy footprint while meeting the ever-increasing bandwidth requirements of the growing big data societal challenge. Our project aims to further develop the novel technology and demonstrate its effectiveness for data center applications, to create the first-of-its-kind low-cost and scalable fabrication technology for the silicon photonics market, compatible with the current systems used in the industry. Furthermore, we will prepare and validate the business concept with value chain players, strengthen our IPR strategy, and prepare the commercialization to establish a start-up aiming at serving this significantly growing market (40% Compound Annual Growth Rate).

Kelvin Probe Force Microscopy (KPFM) is one of the newest scanning probe microscopy techniques, that enables us to obtain information about electrostatics and charge transfer on a surface, measured via very sharp tip moving above a sample. However, the theory behind the KPFM measurements and all physical interactions between the tip and sample are not fully understood, especially for very close scans. We plan to use density functional theory calculations to reveal the unknown physics of close KPFM scans. We will prepare multiscale simulation package for the KPFM, which will work on quantum theory level as well as simplified fast mechanistic model level and which will cover a wide range of experimental conditions. This work will enable us to get additional information about the physics going on the scanned sample from the KPFM measurements and to employ KPFM as an additional source of information for structural identification. Finally, it can lead to general theory for chemical resolution in scanning probe microscopy.

Real-world optimization problems pose major challenges to algorithmic research. For instance, (i) many important problems are believed to be intractable (i.e. NP-hard) and (ii) with the growth of data size, modern applications often require a decision making under {\em incomplete and dynamically changing input data}. After several decades of research, central problems in these domains have remained poorly understood (e.g. Is there an asymptotically most efficient binary search trees?) Existing algorithmic techniques either reach their limitation or are inherently tailored to special cases. This project attempts to untangle this gap in the state of the art and seeks new interplay across multiple areas of algorithms, such as approximation algorithms, online algorithms, fixed-parameter tractable (FPT) algorithms, exponential time algorithms, and data structures. We propose new directions from the {\em structural perspectives} that connect the aforementioned algorithmic problems to basic questions in combinatorics. Our approaches fall into one of the three broad schemes: (i) new structural theory, (ii) intermediate problems, and (iii) transfer of techniques. These directions partially build on the PI's successes in resolving more than ten classical problems in this context. Resolving the proposed problems will likely revolutionize our understanding about algorithms and data structures and potentially unify techniques in multiple algorithmic regimes. Any progress is, in fact, already a significant contribution to the algorithms community. We suggest concrete intermediate goals that are of independent interest and have lower risks, so they are suitable for Ph.D students.