Fastepo

---

Title: Discovery of materials for enhanced PV performance (Ref NSGPVPS2023) (University of Liverpool jobs)

Funding: Funded studentship

Application deadline: 17.03.2024

Eligibility: Open to students worldwide

Subject area: Chemical Sciences

Description:

An exciting opportunity is available for a 3.5-year PhD position supported by NSG Group, focused on the computational discovery of novel materials to enhance the performance of photovoltaic (PV) devices. This project is part of a broader collaboration with NSG, aimed at discovering innovative materials for the glass industry.

In this PhD project, we will explore the utilization of existing computer science methodologies and algorithms, as well as the development of new ones, to automate the analysis of material features and their combinations. This automation will enable the prediction of various material properties. The project may involve creating models to identify new chemical compositions or regions within the periodic table where these desired properties may manifest. It may also involve finding ways to enhance the properties of existing materials.

The ultimate goal of this project is to discover and develop new materials suitable for use as coatings on glass, with the aim of improving the performance of solar cell devices. For instance, this could involve enhancing heat dissipation from the top surface of the glass. Machine learning and data analytics will play a key role in proposing new materials with the desired optical, electrical, and thermal properties. These materials can then be synthesized by other researchers as part of our collaborative efforts.

The successful student will collaborate closely with experts in computer science, inorganic chemistry, physics, and materials science to develop tools for predicting new materials with desirable properties. While prior experience in developing machine learning models or chemistry is not a requirement, candidates should possess strong mathematical and programming skills.

In addition to gaining knowledge and experience in machine learning and data analytics, the student will have the opportunity to develop teamwork and scientific communication skills, as they will work closely with both computational and experimental researchers within the team.

Applications are encouraged from students with a preferably 1st or 2.1 class BSc (or equivalent) in Mathematics, Computer Science, or Physics. Previous experience in machine learning models (or chemistry) is not mandatory, but candidates with robust mathematical and programming abilities are desired.

More Details and Apply.

---

Title: Discovery of new materials for applications on glass using Mathematical Optimisation and Machine Learning methodologies (Ref NSGPVCS2023) (University of Liverpool jobs)

Funding: Funded studentship

Application deadline: 17.03.2024

Eligibility: Open to students worldwide

Subject area: Maths and Computing

Description:

The functionalization of flat glass is a thriving industry worth billions of dollars, encompassing various applications like display technology, energy-efficient windows, and photovoltaic energy generation. To keep advancing in these fields and others, there’s a need for novel materials that outperform existing ones.

Developing such materials, which can be applied as thin films on glass and exhibit stability and unique functional properties, is a complex endeavor. It involves intricate considerations such as structure, bonding, and defect chemistry. Mathematical modeling and optimization, which are fundamental in machine learning and artificial intelligence, are also crucial in addressing the intricate challenges in chemistry and material science. These challenges include predicting new functional materials with desired properties, statistically characterizing data, designing experiments, and managing production.

This PhD project will investigate existing optimization techniques and propose innovative methods and mechanisms. The primary focus will be on black-box optimization methods capable of tackling complex problems that are challenging to model directly. Examples of such methods include grid, coordinate, and pattern searches, as well as metaheuristics, model-based approaches, surrogate models, and evolutionary optimization techniques like genetic algorithms. Additionally, methods utilizing natural gradient and information geometry, as well as Bayesian optimization, will be explored. Many of these methods heavily rely on various machine learning and statistical mechanisms to adapt to the optimization landscape and automatically gather relevant data.

The student will collaborate closely with computer scientists, inorganic chemists, physicists, and material scientists to develop tools for predicting new materials with desirable properties. The optimization algorithms developed or enhanced during this research will surpass the classical approaches employed by computational scientists in the existing literature.

Qualifications: We welcome applications from students holding a preferably first-class or upper second-class BSc degree (or equivalent) in Mathematics, Computer Science, or Physics. Prior experience in optimization, machine learning, or chemistry is not mandatory, but successful candidates must possess strong mathematical and programming skills.

More Details and Apply.

---