Microtechnology and Nanoscience

Title: <span style="font-size: 12.0pt; font-family: 'Aptos',sans-serif; mso-fareast-font-family: Aptos; mso-fareast-theme-font: minor-latin; mso-bidi-font-family: Aptos; mso-ligatures: standardcontextual; mso-ansi-language: EN-US; mso-fareast-language: SV; mso-bidi-language: AR-SA;" lang="EN-US">Design and characterisation of terahertz planar&nbsp;hybrids
Date and time: 2 June 2025, 13.15 Location: A604, MC2
Supervisor: Jan Stake&nbsp; Examiner: Jan Stake Opponent: David Klerebladh
Abstract: Power splitting and combining are essential functions in electronic and optical instruments. These functions are typically implemented using a directional coupler, which is a passive component that comes in various forms across the electromagnetic spectrum. These couplers are used to combine power sources, distribute signals for balanced and sideband-separating mixers, or redirect portions of signals for monitoring. At terahertz frequencies, the multi-section branch guide coupler, implemented in E-plane split block waveguide technology, is the only established version. However, the fabrication of these couplers is constrained by tight tolerances and high aspect ratio features, making them scale poorly at higher frequencies. There is a need to revisit this simple component and explore alternative implementations to support semiconductor devices that are rapidly being developed at terahertz frequencies.
In this work, three different planar hybrids operating in the 750 GHz to 1100 GHz frequency range are presented. Specifically, the branch line, broadside coupled line, and Lange are evaluated through electromagnetic simulations. The 3-dB branch line coupler is fabricated on a 3-µm-thin silicon substrate using microfabrication and SOI technology. Gold beam leads are used to mechanically support the planar circuits and an E-plane waveguide split block.&nbsp;
The branch line coupler shows a simulated bandwidth of 18%, with a return loss of better than 15 dB in a WM-250 waveguide system. The broadside coupled line and Lange designs provide a flatter amplitude response across the band and improved phase balance. These results demonstrate the promising potential of planar couplers and hybrids at terahertz frequencies and will create opportunities for more advanced receiver and transmitter circuits.
&nbsp;

Title: D-Band Transition from eWLB to Polymer Microwave Fiber: Design, Fabrication, and Characterization
Date and time: 2 June 2025, 15.00 Location: C511, MC2
Supervisor: Frida Strömbäck<span style="font-size: 12.0pt; font-family: 'Aptos',sans-serif; mso-fareast-font-family: Aptos; mso-fareast-theme-font: minor-latin; mso-bidi-font-family: Aptos; mso-ansi-language: SV; mso-fareast-language: SV; mso-bidi-language: AR-SA;">&nbsp; Examiner: Herbert Zirath Opponent: Aditya Goturu
Abstract: This thesis investigates the design, fabrication, and characterization of coupling solutions between embedded wafer-level ball grid array (eWLB) packages and polymer microwave fibers (PMFs) for sub-terahertz communication, specifically targeting the 130–150 GHz portion of the D-band frequency range (110–170 GHz). PMFs offer a promising low-loss alternative to traditional waveguides and free-space transmission for short-range, high-data-rate applications. However, the current coupling solution—using an in-package Vivaldi antenna—suffers from high transition losses.
To address this, a dielectric coupling structure is proposed, designed to enhance electromagnetic field confinement and better match the modal profile of the PMF. The coupling system, including a back-to-back simulation setup, is modeled and optimized using the full-wave electromagnetic simulation tool High Frequency Structure Simulator (HFSS). The physical structure is fabricated using 3D printing in acrylonitrile butadiene styrene (ABS), and its performance is verified through vector network analyzer (VNA) measurements.
The average measured coupling improvement over the 130–150 GHz range is 0.87 dB, which is in good agreement with the simulation results. A maximum measured improvement of 2.2 dB is observed at 134 GHz.

Title: Quantum Computation with Giant Atoms Beyond Single-Photon Processes
Date and time: 3 June, 2025, 13.00 Location: Luftbryggan, MC2
Supervisors: Guangze Chen and Ariadna Soro Álvarez Examiner: Anton Frisk Kockum
Abstract: In quantum technology, one of the main goals is the construction of a universal quantum computer which requires one to be able to do fast and lossless quantum gates. Currently this can be achieved through superconducting qubits which are highly tunable and can perform high fidelity quantum gates. This, however, often requires additional resources such as tunable couplers and, in the case quantum simulation, also additional qubits that can act as an environment.
In recent years, artificial atoms that break the dipole approximation have emerged as a potential solution to this resource problem. These are called giant atoms and they can exhibit interference effects between different decay channels which can result in decoherence-free dynamics. In particular, this allows for decoherence-free interaction (DFI) between different giant atoms when they are connected in certain configurations. This allows for the mostly lossless transfer of population between different giant atoms without the need for tunable couplers or other additional components.
DFI has previously been shown for both 1D and 2D structured environments, which are lattices of coupled cavities. These waveguides allow us to study giant atoms beyond the Markovian regime, which has use in the simulation of several quantum systems and in chiral quantum optics. Previous studies have shown the DFI required for the iSWAP gate to be constructed, however to make an efficient quantum computer the CZ gate which is impossible to construct using two levels of a giant atom is also required. One way of constructing the CZ gate with giant atoms is through the use of the third level of the atoms, this in turn requires us to go beyond common assumptions in the field such as the single-excitation subspace. In this thesis we therefore use the exact diagonalization of the Hamiltonian to perform simulations of two-photon processes in giant atoms to show DFI between several different levels of the atoms. We show DFI that allow for both the iSWAP gate as well as the CZ gate, while also increasing the number of frequencies it can be constructed for by adjusting the individual coupling points.

Title: Optimizing and Scheduling Quantum State Tomography Experiments with Graph Colouring Heuristics
Date and time: 3 June, 2025, 14.00 Location: Luftbryggan, MC2
Supervisors: Akshay Gaikwad and Anton Frisk Kockum Examiner: Anton Frisk Kockum
Abstract: Quantum State Tomography (QST) is the standard tool for reconstructing an unknown quantum state-typically represented by a density matrix. It is carried out through the measurement of expectation values of a set of observables (data acquisition step), followed by suitable post-processing. One of the key challenges in QST is to determine an optimal set of quantum circuits (unitary operations) to measure a tomographically complete set of observables. It is a task that becomes computationally hard due to the exponential scaling of the Hilbert space with the number of qubits. In this work, we recast the task of identifying an optimal set of quantum circuits in QST as a graph colouring problem. Within this framework, we explore a variety of algorithms and heuristic approaches to optimize QST experiments, including Degree of Saturation (DSATUR), Recursive Largest First (RLF), Integer Linear Programming (ILP), Graph Neural Networks (GNNs), Spectral Clustering (SC), and a range of hybrid methods that aim to balance solution optimality with computational efficiency. Our results demonstrate that the graph colouring heuristic not only determines and substantially reduces the number of required quantum circuits but also guides appropriate experiment scheduling for efficient full or partial QST. Furthermore, the optimization process completes within minutes for systems of up to five qubits on a standard computer, offering a substantial speedup over brute-force methods, which can take several hours. This highlights the protocol’s potential as a practical and scalable solution for characterizing noisy intermediate-scale quantum (NISQ) devices.&nbsp;

Title: Optimal Control of a SNAIL-resonator
Date and time: 3 June, 2025, 15.00 Location: Luftbryggan, MC2
Supervisors: Tahereh Abad Examiner: Göran Johansson Opponent: Joel Sandås
Abstract: Quantum State Tomography (QST) is the standard tool for reconstructing an unknown quantum state-typically represented by a density matrix. It is carried out through the measurement of expectation values of a set of observables (data acquisition step), followed by suitable post-processing. One of the key challenges in QST is to determine an optimal set of quantum circuits (unitary operations) to measure a tomographically complete set of observables. It is a task that becomes computationally hard due to the exponential scaling of the Hilbert space with the number of qubits. In this work, we recast the task of identifying an optimal set of quantum circuits in QST as a graph colouring problem. Within this framework, we explore a variety of algorithms and heuristic approaches to optimize QST experiments, including Degree of Saturation (DSATUR), Recursive Largest First (RLF), Integer Linear Programming (ILP), Graph Neural Networks (GNNs), Spectral Clustering (SC), and a range of hybrid methods that aim to balance solution optimality with computational efficiency. Our results demonstrate that the graph colouring heuristic not only determines and substantially reduces the number of required quantum circuits but also guides appropriate experiment scheduling for efficient full or partial QST. Furthermore, the optimization process completes within minutes for systems of up to five qubits on a standard computer, offering a substantial speedup over brute-force methods, which can take several hours. This highlights the protocol’s potential as a practical and scalable solution for characterizing noisy intermediate-scale quantum (NISQ) devices.&nbsp;

Title: Exploring the Mapping Between Continuous and Discrete Quantum Computing.&nbsp;Comparing CV Gates to DV Gates through GKP encoding and the Stabilizer Subsystem Decomposition
Date and time: 3 June, 2025, 16.00 Location: Luftbryggan, MC2
Supervisor: Cameron Calcluth Examiner: Giulia Ferrini Opponent: Hampus Brunander
Abstract: This thesis project investigates how operations in continuous variable (CV) quantum computing relate to their discrete variable (DV) counterparts through the framework of GKP encoding and the stabilizer subsystem decomposition. Focusing on how CV quantum gates map to their DV counterparts, and what the error is when simulating one with the other. The analysis combines analytical work and numerical simulations, and the results offer insight into the structure and simulatability of continuous variable quantum systems.
&nbsp;

Title: <span style="font-size: 12.0pt; font-family: 'Aptos',sans-serif; mso-fareast-font-family: 'Times New Roman'; mso-bidi-font-family: Aptos; mso-ansi-language: SV; mso-fareast-language: SV; mso-bidi-language: AR-SA;">AI-assisted circuit design with active learning
Date and time: 4 June 2025, 09.00 Location: Fasrummet A820, MC2
Supervisor: Martin Sjödin, Ericsson Examiner: Christian Fager, Chalmers Opponent: Andreas Führ
Abstract: This thesis explores the use of active learning (AL) to reduce the amount of training data required for machine learning (ML) models used for circuit design by selectively sampling the most informative data points. A convolutional neural network (CNN) is trained to predict scattering parameters (S-parameters) from pixelated representations of 2-port circuits, enabling the ML model to act as a fast surrogate to electromagnetic (EM) solvers. The goal is to speed up the training process for the ML models using AL. The performance of the AL-based model is compared to a baseline model trained using random sampling. Evaluation is conducted on a fixed test set, as well as across different frequency ranges and S-parameters. Results show that AL consistently outperforms the baseline in terms of root-mean-square error (RMSE), particularly at higher frequencies where EM behavior becomes more complex. Ensemble models were also investigated to assess their potential in improving the sampling strategy; however, they did not yield better results, and each ensemble run required over two weeks of computation, limiting further experimentation. Finally, the models were tested in a design task where a genetic algorithm (GA) generated circuits from targeted S-parameters. The AL model achieved a 32.9 % lower mean RMSE than the baseline when comparing predicted and simulated S-parameters. These findings highlight active learning as a promising approach for improving data efficiency in ML-based circuit design.
&nbsp;

Title: Wideband MMIC Power Limiters for Integrated Receivers: Design and Simulation
Date and time: 4 June 2025, 15.00 Location: Direktorn D209, MC2
Supervisors: Gregor Lasser Examiner: Gregor Lasser
Abstract: This thesis investigates the possibility of designing wideband power limiter circuits that can be manufactured using MMIC technologies, with a main focus on GaN. Three complete limiter circuits have been designed and simulated. One limiter needing external activation is also presented. Design and simulation were performed using Keysight ADS. Measurements of the circuits have not been performed. Three designs are made for a 150 μm GaN process, and the last design is made for a GaAs process incorporating PIN diodes. The GaN-based limiters use switch-transistors to achieve the limiting behavior with different means of activation. These limiters are designed to operate in the 2−6 GHz frequency band. The minimum frequency is set by the limiter leakage, and the upper frequency by small-signal insertion loss. The best complete GaN limiter keeps the leakage below 25 dBm for input powers up to 50 dBm The GaAs limiter uses PIN diodes in an anti-parallel configuration to achieve limiting. This limiter works in the 2−18 GHz frequency band, and is thus a good option when this type of process is available. The leakage is below 17.5 dBm for input powers up to 48 dBm. It is, however, unclear whether the limiter itself will survive powers that large. Although the results presented in this study are based solely on simulations, they appear reasonable when compared to similar designs found in other works. The designs presented here perform well in terms of power and frequency band, but the insertion losses are slightly higher than what is generally accepted.

Title: <span style="font-size: 12.0pt; font-family: 'Aptos',sans-serif; mso-fareast-font-family: 'Times New Roman'; mso-bidi-font-family: Aptos; mso-ansi-language: SV; mso-fareast-language: SV; mso-bidi-language: AR-SA;">A vacuum integrated fiber Fabry-Pérot cavity setup for optical and optomechanical characterisation
Date and time: 5 June 2025, 15.00 Location: Kollektorn, MC2
Supervisor: Hannes Pfeifer&nbsp; Examiner: Witlef Wieczorek Opponent: Pontus Lindgren
Abstract: Fibre Fabry–Pérot cavities are miniaturised Fabry–Pérot cavities integrated on optical fibres. Their small size, open mode volume and intrinsic integration with optical fibres make them well suited for optomechanical experiments in which one studies the interaction between light and mechanical motion. The short cavity length enables enhancement of the vacuum optomechanical coupling strength and the open mode volume allows for the easy insertion of mechanical resonators or other quantum systems directly into the cavity. These features make fibre Fabry–Pérot cavities suited for many applications such as levitation of nano particles and quantum-enhanced inertial and force sensing. In this masters thesis, a test platform was developed with the goal to enable optical and optomechanical characterisation of fibre Fabry–Pérot cavities consisting of an optical fibre mirror and a reflective mechanical resonator on a chip. To this end, an existing experimental setup for optical characterisation was improved and a vacuum-integrated setup for optomechanical experiments was built and tested. The resulting test platform enables measurements of the resonance linewidth and free spectral range of fibre Fabry–Pérot cavities with a high accuracy. Within the vacuum-integrated setup the cavity length can be locked to the drive laser. This setup will in future work be used for optomechanical experiments measuring the optomechanical coupling strength and controlling mechanical resonator modes utilising the optical spring effect and optomechanical damping.

Title: Quasiparticle Tunneling Detection in Superconducting Qubits
Date and time: 10 June, 2025, 13.15 Location: Kollektorn, MC2
Supervisors: Linus Andersson Examiner: Simone Gasparinetti Opponent: Elias Rasmussen
Abstract: Superconducting quantum circuits are sensitive to decoherence caused by quasiparticle tunneling events, where broken Cooper pairs tunnel across Josephson junctions. While these events typically affect individual qubits, high-energy cosmic ray impacts can generate correlated relaxations and decoherence events across multiple qubits on the same chip, a detrimental problem for quantum error correction protocols that assume independent errors. In this thesis, we studied correlated quasiparticle tunneling using charge-sensitive sensors. The quasiparticle detectors consisted of single-island qubits directly coupled to a waveguide.&nbsp;
By recording simultaneous time traces from both QPD sensors, we identified burst events with elevated tunneling rates consistent with cosmic ray impacts. Our correlation analysis found evidence that these high-energy events affect both sensors within the same timeframe, supporting the hypothesis of correlated quasiparticle generation across the chip. The main limitation was relatively low signal-to-noise ratios which affected our ability to distinguish tunneling events from background noise.
Additionally, we explored the feasibility of using a conventional transmon qubit as a quasiparticle sensor by exploiting the enhanced charge sensitivity of higher-order energy transitions. We successfully demonstrated control of the 3-4 energy transition, but found that the charge dispersion was below our measurement resolution limit for the device tested, highlighting the need for qubits with lower EJ/EC ratios for this approach.

Title: A critical comparison of current quantum gate optimization techniques — Deep reinforcement learning with an ansatz, filtering of piece-wise constant controls, and analysis of quantum gate dynamics
Date and time: 10 June, 2025, 14.00 Location: Luftbryggan, MC2
Supervisors: Tahereh Abad and Anton Frisk Kockum Examiner: Anton Frisk Kockum
Abstract: Better quantum gates are likely key to enabling fault-tolerant, useful quantum computers. This study has focused on comparing different approaches to single-qubit and two-qubit gate optimization. The primary focus is on deep reinforcement learning. It is shown that using an ansatz can enhance the performance of deep reinforcement learning, both for single-qubit and two-qubit gates, but only significantly for single-qubit gates. Two different approaches to including an ansatz are compared, and adding the ansatz to the output from the neural network might be preferred over pre-training. The study is unable to show a significant improvement of quantum gates using deep reinforcement learning relative to using a state-of-the-art black-box optimizer, despite the black-box optimizer being easier to implement experimentally. For any quantum gate defined by piece-wise constant controls, it seems likely that a low-pass filter can enhance the performance, at least if considered by the optimizer. This is significant as the bandwidth of experimental devices is limited, meaning that low-pass-like distortions might be expected in experiments. Finally, the study highlights the importance of ZZ-coupling to understanding two-qubit gates, a phenomenon not typically considered when analyzing two-qubit gates analytically.

Title: Quantum Geometric Effects in Magnetic topological Insulator MnBi2Te4
Date and time: 11 June 2025, 14.00 Location: Luftbryggan, MC2
Supervisors: Saroj P. Dash and Lars Sjöström Examiner: Saroj P. Dash Opponent: Ludvig Rodung
Abstract: MnBi2Te4 is a recently discovered van der Waals magnetic topological insulator with layer-dependent ferromagnetic and antiferromagnetic properties. It has attracted significant attention for its potential to explore quantum geometrical effects. In particular, its nonlinear transport properties, enabled by broken time-reversal symmetry, have emerged as a powerful tool for probing quantum geometry. Recent theoretical work suggests that higher-order harmonic responses can reveal underlying contributions related to Berry curvature and quantum metric multipoles.&nbsp;
In this thesis, the nonlinear transport in MnBi2Te4 was investigated up to the seventh harmonic under varying applied magnetic field, temperature and crystal angle. Devices were fabricated via exfoliation and electron-beam lithography and characterized using lock-in techniques. Higher harmonic voltage responses up to seventh order were successfully measured, revealing nonlinear transport behaviour. Suppression of even-order harmonics was observed in the presence of inversion symmetry, while odd-order signals remained measurable. Additionally, scaling analysis of the higher harmonics were performed, indicating the presence of multiple contributing mechanisms, beyond the nonlinear Drude and quantum metric terms.

Title: Transfer of wafer scale single crystalline graphene
Date and time: 13 June, 2025, 10.00 Location: Luftbryggan, MC2
Supervisor: Johanna Huhtasaari Examiner: Samuel Lara-Avila Opponents: Vera Hamlet Louice Andersson and Pontus Ericstam
Abstract: Discoveries and studies on the properties of two-dimensional (2D) materials have attracted significant attention. Among them, graphene has emerged as one of the most promising materials, gaining traction in the fields of electronics, sensing and energy storage.
Epitaxial graphene, grown by high-temperature thermal decomposition of silicon carbide (SiC), is a promising 2D material for high-end electronics, offering highly reproducible and homogeneous wafer-scale single-crystal graphene. Graphene was successfully delaminated from such a 4-inch SiC substrate and transferred on to a 4-inch SiO2/Si wafer. This was achieved by optimizing transfer of graphene from 7 mm x 7 mm SiC chips. Studies on the surface and transport properties of these transferred graphene samples indicate that crystalline quality was preserved during the transfer process.

Architecture and Civil Engineering

Bachelor’s thesis Chemistry and Chemical Engineering

Chemistry and Chemical Engineering

Communication and Learning in Science

Computer Science and Engineering

Electrical Engineering

Industrial and Materials Science

Life Sciences

Mathematical Sciences

Mechanics and Maritime Sciences

Physics

Space, Earth and Environment

Technology Management and Economics

Presentation of master thesis

Choose subject

Course syllabus for

Subject

Each academic year contains different courses

See programme page at universityadmissions.se

Educational area

Permissible substitutions of compulsory courses

To see your programme syllabus – select the academic year in which you started the programme.

[en: page.studyprogrammepage.syllabusesview.coursesdescription]

Presentation of master thesis

Master and Bachelor thesis presentations at Microtechnology and Nanoscience

2 June 2025

13:15 Aditya Goturu, MPWPS

15:00 David Klerebladh, MPWPS

3 June 2025

13:00 Walter Rieck, MPCAS

14:00 Sumukh Moudghalya, MPNAT

15:00 Nils Nerpin

16:00 Ludvig Rodung, MPPHS

4 June 2025

09:00 Omar Bark, MPCAS

15:00 Josef Jörgensen, MPWPS

5 June 2025

15:00 Johanna Örgård, MPPHS

10 June 2025

13:15 Hampus Brunander, MPNAT

14:00 Pontus Lindgren, MPCAS

11 June 2025

14:00 Elias Rasmussen, MPNAT

13 June 2025

10:00 Joyal Jain Palakulam, MPNAT