## Title: Development of 3d-printed microwave metamaterial absorbers

This project relates to the development of broadband metamaterial absorbers for mm-wavelength radiation. Such absorber technologies are being considered for next-generation telescopes measuring the polarization of the cosmic microwave background (CMB). The project involves the study of potential plastic resin candidates doped with conductive particles to increase mm-wavelength absorption properties as well development of code that generates geometries suitable for 3D printing. As part of the project, the student will develop a familiarity with a resin 3D printer and test fabrication of absorbing geometries. The project will conclude with measurement of material properties and comparison with expectations.

Supervisor: Jon Gudmundsson

## Title: Machine learning algorithms and optics design

The project aims to explore and develop machine learning algorithms to design optical instrumentation for astronomical observatories automatically. High-fidelity astronomy requires designing ever-increasing complex optical instruments that meet the researcher’s demands. Traditional optical systems have consisted only of limited pre-machined convex or concave shapes. In recent years, it has become feasible to produce so-called freeform lenses of arbitrary surface shapes, allowing for a wider variety of optical tasks that can be addressed with them.

This opens the opportunity to design more complex optical instruments that satisfy specified criteria. However, the ability to machining such new lenses comes at the expense of an increased design task. This thesis will explore various machine learning approaches to search for optimal optical designs that satisfy predefined criteria and requirements of scientific instrumentation.

Supervisors: Jens Jasche and Jon Gudmundsson

## Title: Method of moments approach to optical modeling for mm-wavelength telescopes

In this project, we will develop an algorithm that makes use of method of moments approach for electromagnetic scattering—a concept often employed in stealth technology—to study realistic optical systems and their impact on the overall noise budget of our bolometers. Although the method of moments approach to electromagnetic scattering problems has been around for a few decades, it has not seen much use in modeling of large telescope systems operating at radio and mm wavelengths. Some of the basic concepts are introduced in textbooks by Harrington.

We will review the algorithm and implement it on a few simple electromagnetic scattering problems before developing code that applies the algorithm to realistic telescopes used by CMB experiments.

Contact Jón Gudmundsson for more information

## Title: Avoided level crossings in cosmology

There is strong observational evidence that most of the matter in the universe is invisible, or ’dark’. It is not known what particles comprise this dark matter, or how they came into being. An ‟avoided level crossing” is a quantum phenomenon, well-known in atomic physics and neutrino physics, where two eigenvalues of a time-dependent Hamiltonian become almost degenerate, and transitions between the eigenstates can become ‟resonant” and unsuppressed.

This project will explore if avoided level crossings can explain the production of dark matter in the early universe, and if so, what additional predictions this entails. This project requires an active interest in theoretical quantum physics, astroparticle physics, and cosmology.

Contact: David Marsh

## Title: Axion-photon conversion and correlation functions

The QCD axion and more general axion-like particles (ALPs) are theoretically very well-motivated extensions of the Standard Model of particle physics. A prediction of this theory is that ALPs mix with photons in the presence of magnetic fields. The mixing equation is classical, but can be written as a Schrödinger equation for a 3-level system. This makes it possible to translate results from quantum mechanics to learn about the predictions of ALP theories.

This project will develop the formalism of axion photon mixing as applied to one of the most promising targets for axion searches: the X-ray emitting intracluster medium of galaxy clusters. This project will include theoretical and numerical work that may contribute to the foundation for future ALP searches by the next generation of satellite missions.

Contact: David Marsh

## Title: Geometric destabilisation of cosmological fields

Inflation provides the leading paradigm for the origin of cosmic structure. In this framework, quantum scalar field fluctuations froze into the fabric of space during a hypothetical period of accelerated expansion in the early universe. Inflation can be realised with rather simple ingredients, such as one or more scalar fields slowly ‟rolling” down a potential. Theories with more than one field are theoretically very well-motivated, and includes new features such as non-trivial curvature of the field space itself.

This project will investigate a potential instability of multifield inflationary theories in which the field space curvature induces an instability for certain fluctuations around the inflationary background. After carefully characterising this effect geometrically, this project may investigate whether it can be realised in the type of complex Kähler geometries that are relevant in theories of supergravity. This project involves theoretical work in cosmology and field theory, including aspects of geometry.

Contact: David Marsh

## Title: Modified dispersion relations in cosmology

Can we use cosmological observations to observe, e.g. quantum effects of gravity?

If we can measure the gravitational lensing effect and/or the group velocity through the light travel time for single sources at different energies, we are able to investigate modified dispersion relations.

First, review systems that have observations at different wavelengths. Second, use available data to constrain the energy scale of modifications to the dispersion relation. Last, interpret results in terms of modifications to general relativity.

Is it possible that modified dispersion relations can help easing the tension between the inferred lensing of the cosmic microwave background and lensing of galaxies observed in optical light? How do these limits compare to time delay constraints in, e.g., https://arxiv.org/abs/2109.07850?

Contact Edvard Mörtsell for more information

## Title: Lensed gravitational waves

Cook up a lens that explains the model of lensed gravitational waves in https://arxiv.org/abs/2007.12709.

Also, is it possible that many other gravitational wave events are lensed and in fact at much higher redshifts than commonly believed, see https://arxiv.org/abs/2006.13219 and https://arxiv.org/abs/2106.06545?

Contact Edvard Mörtsell for more information

## Title: Gravitational waves in modified gravity

In extended gravity theories, e.g., massive graviton theories, the gravitational wave velocity will be energy dependent. This means that the higher frequency signal from the later stages of a coalescing event may catch up with the early low frequency part causing a gravitational wave sound bang at the detector, or even an inverted signal. In principle this could be searched for in archival data. The project aims at deriving possible highly distorted gravitational wave signals very different from the ones nominally searched for and therefore possibly missed by common detection pipelines.

Contact Edvard Mörtsell for more information

## Title: Micro lensing bias in gravitationally lensed supernovae and quasars

In cases of multiply imaged supernovae and quasars, we expect the magnification of individual images to be affected by micro lensing from stars and possibly compact dark matter in the lens galaxy. In principle, this can be corrected for on a statistical basis. However, the fact that it is easier to detect high magnification events may cause a bias towards such events in the observed data, possibly invalidating such simple corrections. The project aims at quantifying this bias through the use of simulated events.

Contact Edvard Mörtsell for more information

## Title: Modified gravity and rotation curves

Can we fit galactic rotation curves if the acceleration scale of Modified Newtonian Dynamics (MOND, or some similar theory) is mass dependent? An example is bimetric theory where the so called Vainshtein radius within which general relativity is restored, . References: e.g. https://arxiv.org/abs/1401.5619 and https://arxiv.org/abs/1705.02366.

Contact Edvard Mörtsell for more information

## Title: The prior dependence in Bayseian model selection

When doing model selection using Bayesian statistics, the assumed prior on the parameters of the model is very important. Especially, the range of the prior can have a large impact on the validity of the model in question. As an example, the case for a non-zero cosmological constant, , is expected to beyond doubt given how much better the fit is to observed data compared to the case with . However, given that the theoretical prior on the possible value of is huge, it is not obvious that it will be preferred given a strict Bayesian analysis. This, and simliar cases, should be investigated in the project. See also https://arxiv.org/abs/2102.10671 and https://arxiv.org/abs/2111.04231.

Contact Edvard Mörtsell for more information