Events: Fysikhttp://www.chalmers.se/sv/om-chalmers/kalendariumUpcoming events at Chalmers University of TechnologyFri, 05 Jun 2020 10:32:56 +0200http://www.chalmers.se/sv/om-chalmers/kalendariumhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-Lina-Aberg-200608.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-Lina-Aberg-200608.aspxLina Åberg, Applied Physics and Biomedical Engineering<p>Online via Zoom</p><p>​Title of thesis: Machine learning for classifying the early stage of Osteoarthritis, based on biological data. Follow the presentation online</p><h2 class="chalmersElement-H2">​Abstract: </h2> <div><span style="background-color:initial">Osteoarthritis, or OA, is a chronic joint disease and the most common form of arthritis. It is a very common disease in human athletes, but also the most common reason for lameness and poor performance in animal athletes, such as racehorses. The traditional standard for diagnosing OA is by radiographic measurements. Unfortunately, clinically recognizable changes do not appear until the chronic destruction of the articular cartilage has progressed too far and the disease is irreversible.</span></div> <div> </div> <div><br /></div> <div> </div> <div>In order to diagnose the disease earlier, the focus has been shifted from imaging biomarkers to biological biomarkers. Several promising biological biomarkers have been found, each representing a different stage of the destruction process. One specific biomarker has shown to increase in both blood and synovial fluid in horses with acute lameness, corresponding to an early stage of OA. If this early OA could be identified, it would be possible to intervene in time and the chronic and painful destruction of the joint tissues could be prevented, which could greatly improve the equine welfare.</div> <div> </div> <div><br /></div> <div> </div> <div>The aim of this thesis was to investigate different machine learning approaches in order to classify OA and find a promising method to be used in a future decision support system for practitioners. The future system should be able to help diagnose OA, and specifically identify the different progression stages of structural changes in the joint, based on biological data from a simple blood sample. </div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Elin-Lorin_Niklas-Westman-200608.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Elin-Lorin_Niklas-Westman-200608.aspxElin Lorin, Product Development Niklas Westman, Production Engineering<p>Online via Zoom</p><p>​Titel på masterarbetet: Development and testing of a concept for analysing kinematics in show jumping. Follow the presentation online</p><h2 class="chalmersElement-H2">​Abstract:</h2> <div><div>Technology is today used within a vast range of sports and sport equipment. This trend is increasing in both the number of solutions available and the different sports it can be applied on. The area of equestrian sports starts to see different products utilizing technologies such as motion sensors for data collection. However, there is an opportunity for further development of products within different areas of equestrian sports. Show jumping has been an area focused on with yearly projects such as the Chalmers Technical Fence at Gothenburg Horse Show, and stakeholders have voiced an interest in further development of a product that can be used outside of this competition.</div> <div><br /></div> <div>Therefore, the aim of this thesis is to map the voice of the customer in order to create an initial specification. Further, the aim is to choose and develop a concept based on said specification. The concept is then tested and evaluated where test results also aims at increasing the knowledge surrounding the kinematics in show jumping.</div></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Otto-Magnusson-200608.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Otto-Magnusson-200608.aspxOtto Magnusson, Biotechnology<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: &quot;Investigating flow related effects of Chronic Kidney Disease on renal drug toxicity in a human-derived proximal tubule microphysiological system&quot; Follow the presentation online​ Password: 912347</p><br />Abstract: To be announced.<br /><br />https://www.chalmers.se/en/departments/physics/calendar/Pages/Licentiateseminar-Sean-Miller-200609.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Licentiateseminar-Sean-Miller-200609.aspxSean Miller, Physics<p>Online via Zoom</p><p>​ Title of thesis: &quot;Nucleon-nucleon scattering in a wave-packet formalism&quot;​ Follow the presentation online​</p><h2 class="chalmersElement-H2">Abstract:</h2> <div><span style="font-size:14px"><span></span>​“In this thesis I analyse the prospect of leveraging statistical analyses of the strong nuclear interaction by using the wave-packet continuum discretisation (WPCD) method to efficiently compute nucleon-nucleon (NN) scattering observables on a graphics processing unit (GPU). The WPCD method gives approximate solutions to the S-matrix at multiple scattering energies at the cost of a single eigendecomposition of the NN channel Hamiltonian. In particular, I demonstrate and analyse the accuracy and inherent parallelism of the WPCD method by computing the most common NN scattering observables using a chiral Hamiltonian at next-to-next-to-leading order. I present an in-depth numerical study of the WPCD method and the GPU acceleration thereof. Additionally, I discuss which windows of opportunity are open for studying the strong nuclear interaction using data from few-nucleon scattering experiments.”</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Jacob-Olander-200610.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Jacob-Olander-200610.aspxJacob Olander, Physics and Astronomy<p>Online via zoom</p><p>​ Title of Master&#39;s thesis: &quot;Constrained space MCMC methods for nested sampling Bayesian computations&quot; Follow the presentation online Password: 105329</p><h2 class="chalmersElement-H2">Abstract:</h2> <div><span style="background-color:initial">Natural phenomena can in general be described using several different scientific models, which creates a need for systematic model selection. Bayesian model comparison as- signs relative probabilities to a set of possible models using the model evidence (marginal likelihood), obtained by an integral that in general needs to be evaluated numerically. Nested sampling is a conceptual framework that efficiently estimates the model evidence and, additionally, provides samples from the model parameter posterior distribution used in Bayesian parameter estimation. A vital step of nested sampling is the likelihood- constrained sampling of the model parameter prior distribution, a task that has proven particularly difficult and that is subject to ongoing research. In this thesis we implement, evaluate and compare three methods for constrained sampling in conjunction with a nested sampling framework. The methods are variants of Markov chain Monte Carlo algorithms: Metropolis, Galilean Monte Carlo and the affine-invariant stretch move, respectively. The latter is applied in the context of nested sampling for the first time in this work. The performances of the methods are assessed by their application to a reference problem that has a known analytical solution. The problem is inspired by effective field theories in subatomic physics where the model parameters take the form of coefficients that are of natural size. We conclude that the efficiency and computational accuracy of nested sampling is strongly dependent on the choice of sampling method and the settings of its associated hyperparameters. In certain cases, especially for high-dimensional parameter spaces, the implementations of this work are seen to achieve better computational accu- racy than MultiNest, a state-of-the-art nested sampling implementation extensively used in astronomy and cosmology. Generally for nested sampling, we observe that it is possible to obtain an inaccurate result without receiving any clear warning signs indicating that this is the case. However, we demonstrate that the validity of the computational results can be assessed by monitoring the sampling process. </span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentationer-fysik-200610.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentationer-fysik-200610.aspxMaster's thesis presentations<p>Online via Zoom</p><p>​Follow the presentations online Password: 176802</p><h3 class="chalmersElement-H3">​<span>Damien Pierce, </span>Applied Physics</h3> <div><span style="background-color:initial"><strong>Title of Master's thesis:</strong> An Investigation Into the Effects of Ozone on a Nanoplasmonic Gas Sensor</span></div> <div> </div> <div></div> <div> </div> <div><br /><strong>Abstract: </strong><br />The importance of gas sensing is of vital importance to the continued health and well being of the populations in urban areas. The sensing of Nitrogen Dioxide and Ozone are of particular importance due to their negative effects on the respiratory system, using nanoplasmonic sensing can provide advantageous sensitivity and response time over other sensors. The nanoplasmonic sensor in this study has already been successful in detecting Nitrogen Dioxide. However its behaviour with Ozone has yet to be investigated. Here we show that using an experimental approach as well as a Langmuir isotherm model can be useful in determining the physical characteristics of the system. It was found that the optimum temperature for Ozone was 300 degrees and 200 degrees for Nitrogen Dioxide. The detection of both Ozone &amp; Nitrogen Dioxide was possible up to 40 ppb of Ozone background. Pure Ozone showed rapid kinetics and saturation, with a five fold increase in sensitivity over Nitrogen Dioxide and an average saturation time of 6 minutes. This study presents a starting point for refining the model presented to further understand the behaviour of nanoplasmonic gas sensors.<br /><br /><div><strong>Supervisor</strong>: Dr. Olof Andersson, Insplorion AB</div> <div><strong>Examiner</strong>: Christoph Langhammer</div> <div><strong>Opponent</strong>: Sarah Zulfa Khairunnisa​<br /><br /><h3 class="chalmersElement-H3">Björn Lönn, <br /><span>Applied Physics</span></h3> <div><strong>Title of Master's thesis:</strong> <span></span><span style="background-color:initial">Need for Speed - Integrating the Worlds Fastest Hydrogen Sensor onto a Miniaturized Platform<br /><br /></span><div><strong>Abstract: </strong><br />The hydrogen revolution may finally be around the corner, with fuel cell driven vehicles and hydrogen gas turbines to mention just a few examples of applications. With such a transformation comes a need of fast and sensitive hydrogen detection systems. Nanoplasmonic hydrogen sensors have been a studied topic at Chalmers University for over a decade, and it offers the fastest hydrogen sensing equipment to date. In order to become commercially attractive, the technology need to be altered and integrated onto a platform consisting of cheap electronic components. This thesis demonstrates a first such attempt for a miniaturized nanoplasmonic hydrogen sensor, by combining the sensing technology developed at Chalmers, with Insplorion’s low-cost NPS-platform. The optimized device developed here meets the requirements of detection range and accuracy, stated by the U.S Department of Energy, but exhibits a response time considerably slower than what is required.</div></div> <div><br /></div> <div><strong>Supervisor</strong>: Dr. Irem Tanyeli Insplorion AB &amp; Christoph Langhammer</div> <div><strong>Examiner</strong>: Christoph Langhammer</div> <div><strong>Opponent</strong>: Linnéa Strandberg</div> <div><br /><br /></div> <div><span></span><h3 class="chalmersElement-H3">Sarah Zulfa Khairunnisa, Nanotechnology</h3> <div><span style="font-weight:700;background-color:initial">Title of Master's thesis:</span> Systematic Study of PdAuCu Alloy Nanoplasmonics for Hydrogen Gas Detection</div> <div><br /></div> <div><strong>Abstract: </strong><br />Hydrogen has great potential as an important energy carrier in the future, due to its ability to generate clean and sustainable electricity in a fuel cell. However, hydrogen is flammable even at low concentration (4 vol.%) in ambient air. Furthermore, it is odorless and transparent. Therefore, hydrogen sensors are needed as a warning system for leak detection along the entire value chain.  </div> <div>In this study, I have investigated the composition-dependent response of palladium-alloy-based plasmonic hydrogen sensors. Palladium is of interest due to its inherent selectivity towards hydrogen gas, but it also has inherent disadvantages, such as hysteretic response, and high susceptibility towards CO poisoning. To this end, it has been known that hysteresis and CO-poisoning issues can be addressed by alloying Pd with Au and Cu, and a recent study showed that a ternary alloy (PdAuCu) is able to combine the best features from both PdAu and PdCu binary alloys. However, there is a trade-off between the ternary alloy sensor’s sensitivity and CO-poisoning resistance. Therefore, in this thesis, I have systematically screened a broader Cu and Au concentration range in PdAuCu ternary alloys to find the optimum alloyant concentration in this respect.  </div> <div>As the main results, I found that; (i) CO-poisoning resistance is achieved with a Cu content of minimum 10 at.%; (ii) hysteresis-free and linear response is obtained when the Au content is minimum 25 at.%; (iii) the identified champion system Pd65Au25Cu10 shows reasonably fast response/recovery time of 6.8 s and 10.0 s, respectively at room temperature and in synthetic air background.<br /><br /></div></div> <div>​<span style="background-color:initial;font-weight:700">Supervisor</span><span style="background-color:initial">: Iwan Darmadi &amp; Christoph Langhammer</span><br /></div> <span></span><div><span style="font-weight:700">Examiner</span>: Christoph Langhammer</div> <div><span style="font-weight:700">Opponent</span>: Damien Pierce​</div></div></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Hillevi-Wachtmeister-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Hillevi-Wachtmeister-200611.aspxHillevi Wachtmeister, Physics<p>Online via Zoom</p><p>​ Title of Master&#39;s thesis: &quot;Tracking marine micro organisms using deep learning</p><h2 class="chalmersElement-H2">​Abstract:</h2> <div><span style="background-color:initial">The goal</span><span style="background-color:initial"> of this project is to develop a software that can be used to study swimming patterns of marine micro organisms. The software is based on a neural network, which is trained to recognize different types of plankton. The predictions from the network are then used to find the positions of the plankton, and then track their movements.</span></div> <div> </div> <div><br /></div> <div> </div> <div>The project is divided into two parts. First, videos containing only one type of plankton, Lingulodinium polyedra and Alexandrium tamarense respectively, are analyzed. A type of neural network, called U-net, is trained to segment the input images into background and plankton sections. From the segmented images, positions can be obtained and then connected to form a trajectory for each plankton. The drift of the plankton movements is calculated and subtracted from the trajectories, and finally the speed and net displacement is calculated. The results from the single plankton experiments are compared to a previous analysis that was made using the algorithmic method TrackMate.</div> <div> </div> <div><br /></div> <div> </div> <div>Secondly, videos containing two types of plankton are analyzed. Two experiments are conducted using Strombidium arenicola and Rhodomonas baltica in the first experiment, and Alexandrium tamarense and Rhodomonas baltica in the second. The segmented images, obtained from the U-net, consists of an additional plankton section for the second type of plankton present in the experiment.</div> <div> </div> <div><br /></div> <div> </div> <div>The analysis of the single plankton experiments yields longer and fewer trajectories using the U-net method, compared to the previous TrackMate results. This shows that the TrackMate method is losing plankton at more positions, compared to the U-net method. The U-net method is therefore able to track each plankton for a longer time. The multi-plankton experiments proves the network's ability to distinguish and track multiple plankton at the same time.</div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Robin-Pfeiffer-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Robin-Pfeiffer-200611.aspxRobin Pfeiffer, Materials Engineering<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: Pt / Pt3Y Fuel Cell Nanoparticle Catalyst Fabrication via Sputter Deposition onto Liquid Substrates Follow the presentation online​ Password: 198470</p><h2 class="chalmersElement-H2">Abstract:</h2> <div><span style="background-color:initial">Thi</span><span style="background-color:initial">s thesis pres</span><span style="background-color:initial">ents an overview on the fabrication of Pt and Pt3Y nanoparticles via sputtering onto liquid substrates for potential use in the catalysis of the oxygen reduction reaction in modern fuel cells. It could be shown that spherical nanoparticles with diameters in the range of 1 - 4 nm can be formed by sputtering onto different liquids. Ionic liquids as well as polyethylene glycol could be proved to be suitable for Pt nanoparticles. However, Pt3Y seemed to react with ionic liquids to form undesirable structures and compounds. Sputtering of Pt3Y onto polyethylene glycol resulted in the formation of nanoparticles without any unwanted reactions. Compared to an ionic liquid substrate, polyethylene glycol produced nanoparticles with a slightly broader particle size distribution due to less effective stabilization mechanisms. Summarized, the liquid polymer turned out to be the best liquid substrate candidate for the formation of Pt3Y nanoparticles.</span></div> <div> </div> <div><br /></div> <div> </div> <div>The size of the nanoparticles varied slightly with the applied liquid substrate without exhibiting significant trends. Furthermore, the size appeared to increase by roughly 20 - 30 % by means of a post heat-treatment at 165 °C. An elevated sputtering power resulted in larger particle sizes but an unchanged particle concentration, which points out a particle formation and growth mechanism at the liquid surface. This suggestion could be validated by an investigation of the visual appearance of the sputtered wafers. The nanoparticles fabricated in polyethylene glycol were shown to be catalytically active for the oxygen reduction reaction in an acidic environment by rotating disc electrode measurements. Alloying the nanoparticles with Y led to an electrocatalytic activity enhanced by a factor 1.67 compared to pure Pt nanoparticles. This proves the particles’ potential application in fuel cells and provides motivation for further research on this topic.</div> <div> </div> <div></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Linnea-Strandberg-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Linnea-Strandberg-200611.aspxLinnéa Strandberg, Applied Physics<p>Online via Zoom</p><p>​Titel på masterarbetet: &quot;Fabrication and Characterisation of Fuel Cell Catalyst Nanoparticles via Laser Ablation&quot; Follow the presentation online​ Password: 891639</p><h2 class="chalmersElement-H2">​Abstract: </h2> <div><span style="background-color:initial">This report investigates the feasibility of using laser ablation in liquid (LAL) to produce nanoparticles to be used as catalysts in fuel cells, with a focus on producing Pt3Y-particles for use in proton exchange membrane fuel cells (PEMFC). Energy-dispersive X-ray spectroscopy (EDX) and X-ray photoelectron spectroscopy (XPS) measurements indicate that the yttrium was oxidised during the fabrication process, and, thus, not forming the alloy with platinum as aimed for. Similarly, when ionic liquids were used as the ablation liquid, much of the yttrium reacted with fluorine present in the liquid. Taken together, the lack of metallic yttrium indicates that the particles were formed by evaporation and condensation of atoms from the target surface in the liquid phase rather than from ejection of droplets.</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-Sebastian-Lundquist-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-Sebastian-Lundquist-200611.aspxSebastian Lundquist, Physics and Astronomy<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: &quot;Bayesian Model Averaging of Nuclear Mass Models&quot; Follow the presentation online​ Password: 812042</p><h2 class="chalmersElement-H2">Abstract:</h2> <div>In this thesis we investigate the performance of Bayesian inference and Bayesian model averaging (BMA) applied to two nuclear mass models, the Duflo-Zuker 10 parameter model (DZ10) and the semi-empirical mass formula (SEMF). The DZ10 and SEMF models both have theoretically and experimentally motivated terms but the relative importance of them is less clear. Using Bayesian inference and BMA we have attempted to quantify model uncertainties and improve inference about nuclear masses. To explore the robustness of our BMA analysis we compare the results using different choices of parameter priors, and vary the assumed model discrepancy. The main focus is on the DZ10 model. We employ the Atomic Mass Evaluation from 1983 (AME83) for parameter estimation, then we evaluate the predictive power of the model using the Atomic Mass Evaluation from 2016 (AME16). In an attempt to determine the limits of stability of visible matter we also make a prediction for the neutron drip lines, in the tin isotopic chain (Z=50) using the DZ10 model trained on AME16. The 1-neutron drip line is predicted to neutron number N=123 [95, 125], and the 2-neutron drip line at N=115 [103, 125]. Where the error bar corresponds to a 67% degree of belief.​<br /></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Christoffer-Henriksson-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Christoffer-Henriksson-200611.aspxChristoffer Henriksson, Applied Data Science<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: &quot;Predicting Cycle Life of NMC Cells by Discharge Capacity Voltage Curves&quot; Follow the presentation online​</p><h2 class="chalmersElement-H2">​Abstract: </h2> <div><span style="background-color:initial">The biggest issue with rechargeable batteries is arguably their limited lifetime. They suffer from capacity degradation and power fade, and their performance decreases as they age. Estimating the remaining useful life is therefore an important task. However, the complex internal aging mechanisms are difficult to model. Recently, machine learning has become a promising approach for predicting remaining useful life. This thesis evaluates whether a new elastic net machine learning model trained on data from LFP cells can be used to predict cycle life of NMC cells. The model uses capacity and voltage data during discharge phases to derive a feature highly correlated to cycle life. Four commercial NMC cells were cycled in Chalmers Electric Power Battery Lab to collect cycling data. The model was able to make useful cycle life predictions for these cells, which suggests that the approach is applicable to other lithium-ion cells.</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Jakub-Fojt-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Jakub-Fojt-200611.aspxJakub Fojt, Applied Physics<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: Hot-carrier generation and transfer across nanoparticle-molecule interfaces Follow the presentation online​</p><h2 class="chalmersElement-H2">​Abstract: </h2> <div><span style="background-color:initial">Metallic nanoparticles are important materials for emerging sensing and catalysis technologies. Their special properties stem from the presence of a localized surface plasmon resonance (LSPR) mode that can couple to visible light. The LSPR causes the nanoparticle to scatter and absorb more light at frequencies that match the plasmon energy. The plasmon excitation has a lifetime of a few femtoseconds before it dephases into a distribution of electrons and holes with</span></div> <div> </div> <div>a strongly athermal energy distribution. In this thesis, time-dependent density functional theory has been employed to study these phenomena in Ag nanoparticles.</div> <div> </div> <div><br /></div> <div> </div> <div>In the first part of this thesis, the photoabsorption spectra were systematically calculated for a series of Ag nanoparticles between N=13 and 586 atoms in size. The main findings are that the LSPR peak frequency depends linearly on N-1/3.</div> <div> </div> <div><br /></div> <div> </div> <div>When a plasmon forms in a nanoparticle in the vicinity of a molecule it may dephase into a transition of an electron from the nanoparticle to the LUMO state of the molecule, or from the HOMO state of the molecule to an unoccupied state in the nanoparticle. These processes are termed direct hot-electron transfer and direct hot-hole transfer, respectively. In the second part of the thesis, a systematic study was carried out in which a CO molecule was placed at different distances from the nanoparticle, and the system was excited with a laser pulse. The results indicate that for this system direct hot-electron transfer happens with a probability of around 1% and is only weakly dependent on the molecule-nanoparticle separation until it decays to zero at large distances. Meanwhile, hot-hole transfer is between 0.2 and 0.3% at a distance of 1.8Å and decays monotonically. Contributing factors to the differences are that the molecular LUMO state is much more delocalized at large separations than the HOMO state. The most important criterion for transfer to occur is an alignment in energy between the nanoparticle and molecule densities of state.</div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Victor-Rosendal-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Victor-Rosendal-200611.aspxVictor Rosendal, Applied Physics<p>Online via Zoom</p><p>​ Title of Master&#39;s thesis: Optical response of nanoalloy hydrogen sensors from first-principles Follow the presentation online​</p><h2 class="chalmersElement-H2">​​Abstract: </h2> <div>Hydrogen shows promise as a replacement for conventional fossil fuels. However, its high flammability and gas permeability pose high demands on sensors, which must respond quickly and accurately. Nanoscaling improves the kinetics and allows for optical hydrogen sensing. A nanoscaled metallic sensor typically shows a well-defined extinction peak in the optical regime and one proposed sensing technique is to detect the shift in said peak due to hydrogenation.</div> <div> </div> <div><br /></div> <div> </div> <div>The aim of this thesis is to, from first-principles, study the optical response of PdAu nanodisks as a function of hydrogenation. PdAu:H was mainly treated as a random alloy but thermodynamic structures were also investigated. Cluster expansions were used in combination with Monte Carlo simulations to generate thermodynamically representative PdAu:H structures. The dielectric functions for the random and the thermodynamic structures were calculated by applying static and time-dependent density functional theory. Optical extinction spectra of PdAu:H nanodisks were obtained via electromagnetic finite-difference time-domain simulations using the previously calculated dielectric functions.</div> <div> </div> <div><br /></div> <div> </div> <div>The extinction peak of nanodisks with a diameter of 100 nm and height 20 nm showed a redshift due to hydrogenation over the entire range of gold concentrations of 0 to 42% considered here, and the redshift is approximately linear with respect to hydrogen. Even though there is non-trivial ordering in the thermodynamic PdAu:H structures, no clear difference between the random and thermodynamic case was observed in the optical response.</div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Thesis-defense-Cecilia-Fager-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Thesis-defense-Cecilia-Fager-200611.aspxCecilia Fager, Materials Science<p>Online via Zoom</p><p>​Title of doctoral thesis: &quot;Quantitative 3D reconstruction of porous polymers using FIB-SEM tomography: Correlating materials structures to properties of coatings for controlled drug release&quot;  Follow the thesis defence live on Chalmers Physics&#39; YouTube channel​​</p><h2 class="chalmersElement-H2">Abstract:</h2> <div><span style="background-color:initial">Understanding the correlation between materials structures and properties enables the optimisation of materials and tailoring them for specific applications. This work concerns porous networks in polymer coatings and in particular films for controlled drug release. The porous networks act as transport paths for drugs. To tailor the drug release, the correlation between the porous network and the transport properties is crucial. The network needs to be characterised in three dimensions (3D) and high spatial resolution 3D data can be acquired using a focused ion beam combined with scanning electron microscope (FIB-SEM) tomography. The FIB-SEM utilises an ion beam to perform serial sectioning and the electron beam to image the cross-section surface.</span></div> <div> </div> <div><span style="font-size:14px">The aim of this work was to develop a generic protocol for optimised FIB-SEM tomography for soft, porous and poorly conducting materials and to use the quantitative experimental data to simulate transport properties. The protocol was used for model porous polymer films and polymer film coated pellets representative of structures used for controlled drug release in pharmaceuticals. The 3D reconstruction and quantitative evaluation of the porous network provided information about important structural characteristics such as pore connectivity, tortuosity and geodesic paths. The structural information was used to simulate transport properties and explained the experimentally measured diffusion properties of different porous polymer films.</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Andreas-Hellstrom-200611.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Andreas-Hellstrom-200611.aspxAndreas Hellström, MPSES<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: ”Hydrogen production and storage at Renova – A preliminary analysis”</p>​<b>Abstract</b>: To be annonunced.<br /><br /><a href="https://chalmers.zoom.us/j/65837412045" target="_blank"><img class="ms-asset-icon ms-rtePosition-4" src="/_layouts/images/icgen.gif" alt="" />Follow the presentation online​</a><br />Password: 658 3741 2045​ ​​<br />https://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-presentation-Lukas-Nystrom-200612.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-presentation-Lukas-Nystrom-200612.aspxLukas Nyström, MPCAS<p>Online via Zoom, (Veterans General Hospital, Taipei, Taiwan)</p><p>​Title of Master&#39;s thesis: Inter-hospital brain tumour diagnostics using Private Federated Learning Follow the presentation online Meeting ID:94422537863 password: 20200612</p><h2 class="chalmersElement-H2">​Abstract: </h2> <div><span style="background-color:initial">I </span><span style="background-color:initial">have implemented and analysed the possibility of using Private Federated Learning as a tool to perform decentralised, collaborative training of Deep CNN models for brain tumour segmentation. The general goal is to develop an AI system that can assist medical professionals in their diagnoses of patients, thereby reducing the time and resources required. The main obstacle in medical AI is the lack of high quality data which implies that it would be beneficial to train models across several instituitons, since each individual hospital has insufficent amounts of cases. However, due to the sensitive nature of patient journals extensive privacy regulations prohibit the sharing of such samples outside of the hospital. Federated Learning is an attempt to circumvent this issue by sharing and aggregating model updates instead, thus artifically increasing the apparent size of the data set without actually breaching the integrity of any individuals. Besides empirically evaluting different models and configurations in order to find one that provides human level performance, I have also conducted an extensive theoretical review of additional privacy mechanism that can further strengthen the integrity of the system against malicous attackers. This includes the use of Differential Privacy, Homomorphic Encryption and other kinds of Secure Multiparty Computation techniques. The study concludes that it is indeed feasible to create such a federated model, at a minor 11% performance cost relative to if the data had been centralised. However the federated model is 30% better than any of the models created by each institution on their own which shows that Federated Learning is the best available option for a real world scenario.</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Thesis-defense-Mikael-Valter-200612.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Thesis-defense-Mikael-Valter-200612.aspxMikael Valter, Physics<p>Online</p><p>​Title of doctoral thesis: &quot;Modelling electrooxidation of glycerol and methanol on close-packed transition metal surfaces&quot;. Follow the thesis defence live on Chalmers Physics&#39; YouTube channel​​​</p><h2 class="chalmersElement-H2">Abstract:</h2> <div>Burning fossil fuels leads to excess CO2 in the atmosphere, causing global warming, threatening civilisation and ecosystems worldwide. As a step in making the society fossil-independent, we need to replace oil, coal, and gas in the transportation sector with fuels originating from sustainable energy sources. Biodiesel is one such option, from which glycerol is a byproduct. With the help of electrooxidation, glycerol can be used as a feedstock to extract hydrogen gas, which may be used for upgrading biofuels or in proton exchange membrane (PEM) fuel cells. Methanol is a possible fuel in direct methanol fuel cells (DMFCs) and can, moreover, be used as a simple model for glycerol in some respects.</div> <div> </div> <div>The primary focus of this thesis is to study the reaction thermodynamics of glycerol electrooxidation on Au(111) and other close-packed late transition metal surfaces. This provides routes and products that are thermodynamically favourable, information on steps that are difficult to overcome, and at what theoretical limiting potential the reaction becomes spontaneous. Using scaling relations for adsorption energies, these results can be generalised to alloys and other possible electrode materials. We use density functional theory to model the system, and to some extent experimental verification by cyclic voltammetry. Long range dispersion (van der Waals), which have been neglected in computations until recently, is investigated by assessing density van der Waals functionals. This is of particular importance for an inert metal such as gold. Another aspect that has commonly been ignored is solvent effects, which we study for the model system of methanol electrooxidation on Au(111). This includes an implicit model - a continuous dielectric -and an explicit model of water molecules. ​                    </div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Oskar-Lindroos-200615.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Oskar-Lindroos-200615.aspxOskar Lindroos, Physics and Astronomy<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: &quot;Prospects for the Discovery of General Dark Matter-Induced Atomic Responses&quot; Follow the presentation online​ Password: 572245</p><h2 class="chalmersElement-H2">​Abstract:</h2> <div>One of the major mysteries of modern physics is dark matter. Proposed by Fritz Zwicky in 1933, dark matter stands as the answer to several astronomical observations, but its constituents remain unkown. Since current observations only show dark matter interacting gravitationally, one can assume that the dark matter particle is a massive, weakly interacting particle also referred to as a WIMP. In an attempt to find the dark matter particle, large underground detectors have been constructed. These detectors utilize the principle of direct detection in order to detect weakly interacting particles. As the Earth passes through the halo of dark matter within the Milky Way one expects the flux of dark matter particles that arises to produce a signal within these experiments. However, so far a signal that cannot be anything other than dark matter remains to be found. This thesis will present the sensitivities for future direct detection experiments under the assumption of general dark matter-electron interactions. We provide the underlying theory of a model describing general dark matter-electron interactions in the non-relativistic frame which we use to simulate event rates in active experiments. Based on the null results in current experiments, we perform a statistical analysis in order to find the lowest detectable coupling constant for different sets of interactions corresponding to a specific dark matter particle mass. We express the sensitivity of future direct detection experiments in terms of statistical significance for signal discovery.​</div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-Thesis-presentation-Ida-Svenningsson-200615.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-Thesis-presentation-Ida-Svenningsson-200615.aspxIda Svenningsson, Applied Physics<p>Online via Zoom</p><p>​ Title of Master&#39;s thesis: &quot;Hot-tail runaway electron generation in cooling fusion plasmas&quot; Follow the presentation online​</p><h2 class="chalmersElement-H2">​​Abstract:</h2> <div><span style="font-size:14px"></span><span></span><div><span style="font-size:14px">Runaway electrons pose a threat to safe operation of magnetic confinement fusion reactors due to the damage they can cause on the reactor wall. During a fast cooling of a fusion plasma, the electric field strength increases and high-energy electrons are accelerated to relativistic speeds, a process called hot-tail runaway generation. To mitigate their effect, reliable and efficient theoretical models to predict generation of hot-tail electrons are of importance. Current numerical methods are computationally expensive and the accuracy of available analytical models has not been found satisfactory. In this work, analytical and simplified numerical models for hot-tail generation including a self-consistent description of the electric field are proposed. The models are benchmarked against numerical simulations and their regions of validity are explored. </span></div></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-presentation-Christoffer-Olofsson-200615.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masters-thesis-presentation-Christoffer-Olofsson-200615.aspxChristoffer Olofsson, Physics and Astronomy<p>Online via Zoom</p><p>​ Title of Master&#39;s Thesis: &quot;Models of sub-cycle electromagnetic pulse generation in laser-plasma interaction&quot; Follow the presentation online​</p><h2 class="chalmersElement-H2"><font style="vertical-align:inherit"><font style="vertical-align:inherit">​Abstract: </font></font></h2> <div><span style="background-color:initial"><font style="vertical-align:inherit"><font style="vertical-align:inherit">Sub-cycle pulses are ultra-short laser pulses containing less than a single oscillation and are essential tools in the study of matter at the shortest timescales. It has been recently proposed that such pulses can be attained by letting laser pulses interact with a plasma to generate amplified and compressed pulses. In the scheme of laser wakefield driven amplification (LWDA), an initial seed pulse is modulated by traveling electron plasma waves, forming amplified sub-cycle pulses. In this thesis we investigate the underlying mechanism of sub-cycle pulse generation in the scheme of LWDA. An analytical approach using the method of Green's functions is used in conjunction with particle-in-cell simulations. Moreover, a custom code solving Maxwell's equations with a source term given by a non-linear plasma wave model is implemented and its results compared with particle-in-cell simulations.</font></font></span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Mastersthesis-Tobias-Sandstrom_Lars-Jansson-200615.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Mastersthesis-Tobias-Sandstrom_Lars-Jansson-200615.aspxTobias Sandström and Lars Jansson, MPCAS<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: Graph Convolutional Neural Networks for Brain Connectivity Analysis​​ Follow the presentation online​</p><h2 class="chalmersElement-H2">​Abstract:</h2> <div><span style="background-color:initial">We explore the strengths and limitations of Graph Convolutional Neural Networks (GCNs) for classification of graph structured data. GCNs differs from regular Artificial Neural Networks (ANNs) in that they operate directly on graph structures by defining convolutional operators in a non-euclidean space. We show that GCNs perform well on graph structured data, where regular ANNs typically fail due to the arbitrary ordering of nodes. Different GCN architectures are examined and compared to simplistic ANNs. Tests are initially performed on simulated data sets with implicit class-dissimilarities in regards to graph structures. We demonstrate that GCNs is vital in accurately classifying the simulated data. Network performance is later evaluated on structured MRI-data, displaying cortical thicknesses for 68 regions in the brain of patients with Alzheimer's disease and a healthy control group. On the structured MRI-data, both GCNs and regular ANNs are shown to be able classifiers. However, it is crucial for the performance of ANNs that an order of nodes can be imposed on the MRI-data from labeled brain regions.</span></div>https://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Benjamin-Midtvedt-200615.aspxhttps://www.chalmers.se/en/departments/physics/calendar/Pages/Masterpresentation-Benjamin-Midtvedt-200615.aspxBenjamin Midtvedt, Engineering Mathematics and Computer Science<p>Online via Zoom</p><p>​Title of Master&#39;s thesis: &quot;DeepTrack: A comprehensive deep learning framework for digital microscopy&quot; Follow the presentation online​ Password: 952882​</p><h2 class="chalmersElement-H2">​Abstract:</h2> <div><span style="background-color:initial">Despite the rapid advancement of deep-learning methods for image analysis, they remain underutilized for the analysis of microscopy images. State of the art methods require expertise in deep-learning to implement, disconnecting the development of new methods from end-users. The packages that are available are typically highly specialized, challenging to reappropriate, and almost impossible to interface with other methods. Finally, training deep-learning models often requires large datasets of manually annotated images, making it prohibitively difficult to procure training data that accurately represents the problem. DeepTrack is a deep-learning framework targeting optical microscopy, designed to account for each of these issues. Firstly, it is packaged with an easy-to-use graphical user interface, solving standard microscopy problems with no required programming experience. Secondly, it bypasses the need for manually annotated experimental data by providing a comprehensive programming API for creating representative synthetic data, designed to exactly suit the problem. DeepTrack creates physical simulations of samples described by refractive index or fluorophore distributions, using fully customizable optical systems. To accurately represent the data to be analyzed, DeepTrack supports arbitrary optical aberration and experimental noise. Thirdly, many standard deep-learning methods are packaged with DeepTrack, including architectures such as U-NET, and regularization techniques such as augmentations, decreasing the barrier to entry. Finally, the framework is fully modular and easily extendable to implement new methods, providing both longevity and a centralized foundation to deploy new deep-learning solutions. We demonstrate the versatility of DeepTrack by training networks to solve a broad range of common microscopy problems, including particle tracking, cell-counting in dense biological samples, multi-particle 3-dimensional tracking, and cell segmentation and classification. </span></div>https://www.chalmers.se/en/departments/mc2/calendar/Pages/ws-prospects-of-ultrastrong.aspxhttps://www.chalmers.se/en/departments/mc2/calendar/Pages/ws-prospects-of-ultrastrong.aspxWorkshop on Prospects of Ultrastrong light-matter interactions<p>Hjortviken conference centre</p><p>​Ultrastrong coupling between light and matter has been a topic of immense studies over the last decade. Already, ultrastrong and deep strong coupling regimes have been experimentally achieved in a variety of platforms (intersubband polaritons, THz-resonators, superconducting circuits, optomechanics, plasmonic cavities, etc.). Nowadays, the focus of the community is shifting to observing the rich and exotic phenomena predicted before, which are promising for future quantum technologies. Such as; breakdown of the Purcell effect, optical higher order processes, polariton and QED chemistry, among others.​</p>​<span style="background-color:initial">This workshop aims at gathering researchers from the USC community, working in interdisciplinary </span><span style="background-color:initial">related fields, to shed light upon where the field is now and where it is heading. The objectives </span><span style="background-color:initial">are to stimulate discussions on the latest developments in different platforms for USC, as </span><span style="background-color:initial">well as to encourage exploration of the prospects of the field. This workshop will bring together </span><span style="background-color:initial">theoretical and experimental researchers from various disciplines (Quantum Optics, Plasmonics, </span><span style="background-color:initial">Nanophotonics, Condensed Matt</span><span style="background-color:initial">er physics, ...) to enrich discussions, broaden perspectives and </span><span style="background-color:initial">encourage potential collaborations that will drive the field forward.</span><div><br /><span style="background-color:initial"></span><div><span style="background-color:initial"></span><span style="background-color:initial"><strong>Topics:<img src="/SiteCollectionImages/Institutioner/MC2/AQPL/hjortviken.jpg" class="chalmersPosition-FloatRight" alt="" style="margin:5px;width:250px;height:186px" /><br /></strong></span><div><ul><li>Recent development of fundamental models in USC</li> <li>Different architectures and platforms to harness the potential of USC</li> <li>Prospects and applications of USC</li></ul></div> <div><strong>Location:</strong> <a href="http://www.hjortviken.se/en">Hjortviken conference center</a>, <span style="background-color:initial">Gothenburg, Sweden</span></div> <div><strong style="background-color:initial">Dates:</strong><span style="background-color:initial"> 7-9 September, 2020</span><br /></div> <div><br /></div> <div><strong style="background-color:initial">Format</strong><span style="background-color:initial">: </span><span style="background-color:initial;font-size:14px">The workshop consists of 3 days of presentations and discussions. There will be 4 sessions with presentations and a round-table discussion at the end to encourage discussions. </span></div> <div><span style="font-size:14px">To adapt to the current coronavirus situation, the format of the workshop will be a mix between a remote and physical event. </span><span style="background-color:initial">People who can travel will be welcome to join us in Hjortviken. Given the small dimensions of the gathering, it will be possible to keep a safe distance. Otherwise, we will have the possibility of attending via Zoom (to listen, give the talks and join discussions).</span></div> <div><span style="font-size:14px">Due to the difficulty to predict the travel recommendations status in September, we will open registrations for physical attendance in July. In case it is necessary, we will have the event fully online, via zoom. We will announce any updates here if that is the case.</span></div> <div>​<br /></div> <div><span style="background-color:initial"></span></div> <div><div><strong>Accommodation</strong>: Accommodation and all meals at Hjortviken are included for all attendees.</div> <div>The location can be found at <a href="https://goo.gl/maps/WjGh9RZ2ie9YafdW6" target="_blank">https://goo.gl/maps/WjGh9RZ2ie9YafdW6.</a></div> <div>Hjortviken is 10 minutes away from Landvetter International Airport and free shuttle service is <span style="background-color:initial">available between 08:00-20:00.</span></div> <div>Invited speakers will stay in single rooms and students will share double rooms with a fellow <span style="background-color:initial">student.</span></div></div> </div></div>https://www.chalmers.se/en/centres/gpc/calendar/Pages/Lise-Meitner-award-2020.aspxhttps://www.chalmers.se/en/centres/gpc/calendar/Pages/Lise-Meitner-award-2020.aspxPOSTPONED - Ceremony and lecture – Gothenburg Lise Meitner Award 2020<p></p><p>​Due to the Coronavirus situation, the ceremony will be postponed. The organizers will be back with a new date during autumn 2020. Professor Anne L’Huillier, Lund University, Sweden is the winner of the Gothenburg Lise Meitner Award 2020. She will visit the Gothenburg Physics Centre to receive the award and give the traditional award lecture in honour of the Austrian-Swedish physicist Lise Meitner. ​ Professor L’Huillier receives the award “For pioneering contributions to attosecond laser science and technology”. ​</p><img src="/SiteCollectionImages/Centrum/Fysikcentrum/Gothenburg%20Lise%20Meitner%20Award/Lise%20Meitner%202020/150_Anne_H_.jpg" class="chalmersPosition-FloatRight" alt="" style="margin:5px;width:110px;height:146px" />Professor Anne L'Huillier has been at the forefront of ultrafast laser science since its inception, with her pioneering contributions to high-order harmonic light generation, which is a base technology for attosecond science. Her research has helped foster the field of attosecond science, allowing scientists to visualize the movements of electrons in light-induced processes, which can be used to understand chemical reactions on the atomic level.​<br /><br /><a href="https://portal.research.lu.se/portal/en/persons/anne-lhuillier%28266ecd6e-b257-4a8e-988f-8d232b31abb3%29.html" target="_blank"><img class="ms-asset-icon ms-rtePosition-4" src="/_layouts/images/ichtm.gif" alt="" />Read more about<strong> </strong><span style="font-weight:300"><strong>Professor L'Huillier's</strong> </span>research​</a><br /><a href="/en/centres/gpc/activities/lisemeitner/Pages/default.aspx" target="_blank"><img class="ms-asset-icon ms-rtePosition-4" src="/_layouts/images/ichtm.gif" alt="" />Read more about The Gothenburg Lise Meitner Award and previous laureates​​</a>