Skip directly to content

Coming dissertations at Uppsala university

  • Time-resolved Photoelectron Spectroscopy of Lead Halide Perovskites Author: Birgit Kammlander Link: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-540756 Publication date: 2024-11-15 08:29

    Lead halide perovskites are promising for applications such as solar cells or LEDs. Their success partly stems from the tunablity of their properties by varying their composition. For efficient devices, interfaces between the lead halide perovskites and transport materials play a crucial role. However, stability remains a challenge for commercializing these thin film devices. Further development requires understanding the fundamental properties of both the material itself and of interfaces with transport materials, ideally at an atomic level. This includes understanding charge dynamics and chemical changes under external impacts.

    In this thesis, the method photoelectron spectroscopy (PES) was used and developed for investigating both model systems (via single crystals) and thin film devices. This includes studying dynamics at timescales from pico- to nanoseconds and seconds to minutes. Different single crystal compositions were studied and compared.

    Degradation under X-rays and heat was studied. X-rays induced AX radiolysis, ion migration and metallic lead formation while heat degradation was dominated by AX radiolysis. The interface formation and energy alignment between different single crystal compositions and a potential transport material was characterized.

    Electron and ion dynamics were investigated via laser-pump-X-ray-probe PES and simpler laser ON/OFF experiments. Timescales for electron and ion dynamics were determined and discussed, including surface band bending effects.

    This work further includes a more applied use of this method on thin film devices. Charge dynamics at different interfaces with transport layers in a quantum dot solar cell were studied and the gold contact was found to be crucial for large and long-lived illumination-induced voltage. Degradation of blue perovskite LEDs under applied bias was followed via hard X-ray PES. Chloride migration into the top transport layer and metallic lead formation were found. Overall, this thesis gives insight into chemical changes and electron dynamics of surfaces and interfaces.

  • On how Firms can use Artificial Intelligence for Economic Value Creation : A Business Model Approach Author: Ricardo Costa-Climent Link: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-540703 Publication date: 2024-11-14 13:26

    This research provides a novel answer to the following question: How can firms use artificial intelligence (AI), specifically predictive AI, to create economic value? A novel theory is developed by integrating, in a specific way, two complementary theoretical bodies: the theory of data network effects and business model theory. The former accounts for AI’s unique learning ability to generate users’ perceived value but offers no insight into economic value creation at the firm level. To remedy this limitation, business model theory accounts for economic value creation from digital technology use, while not accounting for AI’s unique learning ability. The process of integrating these two theories is carried out using a multi-method research approach, presented here in nine papers. Regarded together, the papers manifest an evolutionary research process with conceptual and empirical work and exploratory and hypothesis-based studies. Spanning several industrial contexts, these studies jointly develop and test the new theory proposed in this thesis. The findings show that when AI is used to create perceived value for its users, and thereby realise data network effects, and when such AI use is positioned adequately within a firm’s business model architecture, one or more of the four business model themes (novelty, efficiency, complementarity and lock-in) can be activated to achieve economic value creation. The key contribution of this thesis lies in explaining how a firm can use AI to create economic value. The findings have practical implications for managers’ investment decisions regarding the use of AI.

  • The Dynamic Structure of the Escherichia coli Chromosome Author: Konrad Gras Link: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-540764 Publication date: 2024-11-14 13:19

    The Escherichia coli chromosome is a dynamic molecule, exhibiting choreographed reorganizations across spatial scales. The chromosome is involved in a number of essential processes, such as DNA replication and chromosome segregation. These processes ensure that there are at least two copies of the genetic material at cell division, one for each daughter cell to inherit. However, maintaining a cycle-dependent chromosome organization is no small feat. To understand how the chromosome organization is regulated over the cell cycle and to understand the functional importance of the chromosome structure, we investigated the dynamics and intracellular positioning of various chromosomal loci in live E. coli using fluorescence microscopy. 

    Our efforts to understand when and where in the cell chromosomal loci are replicated were based on fluorescently labeling a chromosomal locus and a subunit of the replisome in the same cell. With this labeling strategy, we followed the intracellular positioning of the replisome and various loci relative to each other, as well as their short-time-scale movements. We found that as loci were replicated their short-time-scale movements slowed down momentarily. Mapping the short-time-scale movements over different intracellular positions showed a clear repositioning of several loci towards the replisome to be replicated, which led us to conclude that the chromosome moves to the replisome during DNA replication. 

    To investigate the three-dimensional positioning of chromosomal loci, we performed time-lapse imaging of E. coli strains with fluorescently labeled loci using a microscope with an astigmatic fluorescence emission path. To determine the 3D coordinates of the emitters, we developed a neural network-based algorithm trained on simulated images of E. coli cells with fluorescent foci. Applying this neural network to different loci showed distinct 3D localization patterns over the cell cycle. 

    To study the 3D chromosome organization, we imaged a collection of 83 E. coli strains, each with a different locus label. Using in situ genotyping of barcode sequences to determine each strain’s identity, we mapped 3D localization phenotypes to multiple chromosomal loci in a single experiment. We captured known longitudinal chromosome reorganization, as well as radial localization patterns that had not been observed previously. Finally, we used the experimental location distributions to inform a polymer model of the chromosome, which showed how Mbp-sized domains form dynamically over the cell cycle. Using this approach to study the 3D chromosome organization in live E. coli, we hope to gain further insights into the regulation and functional importance of the chromosome structure.

Pages