Mathematisch-Naturwissenschaftliche Fakultät

Exploring Phase-Change Materials for Heat-Storage from First Principles

Fri, 26 Jun 2026 00:00:00 GMT

Exploring Phase-Change Materials for Heat-Storage from First Principles Jütten, Stefan The transition towards a carbon-neutral energy landscape necessitates the development of efficient thermal energy storage systems to bridge the temporal gap between renewable energy supply and thermal demand. The polymorphic ceramic trititanium pentoxide (Ti₃O₅) has emerged as a promising candidate for latent heat storage, capable of storing thermal energy in a metastable high-temperature phase indefinitely. Here, a comprehensive first-principles investigation of the Ti₃O₅ heat-storage system is presented, ranging from the electronic structure of the bulk material to the complex thermodynamic and kinetic behavior of doped systems, interfaces, surfaces and nanoparticles.
The initial part of this work establishes a robust theoretical framework for describing the open-shell transition metal oxide. It is demonstrated that the meta-GGA functional r²SCAN, augmented with the D3 dispersion correction, provides an accurate description of the structure of Ti₃O₅ polymorphs, superior to standard hybrid functionals. In the bulk, the thermodynamic ground state of the β-phase is identified as an antiferromagnetic semiconductor, while the metastable λ-phase is shown to be a ferromagnetic semiconductor. The transition state is characterized by a rotation of a central Ti-dimer, and predicted r²SCAN-D3 phase transition enthalpy and phase transition temperature are in good agreement with experiment.
Building on this foundation, the modulation of heat-storage properties via aliovalent cation doping (Sc, Al, Mg) is explored. The results reveal that doping lowers the phase transition temperature and enthalpy, primarily through local lattice distortions rather than direct electronic effects. Substitution turns the semiconducting bulk materials into metals, with significant electron density accumulation at the defect site. At low dopant concentration the β-Ti₃O₅ to λ-Ti₃O₅ barrier remains unchanged while the substitution stabilizes the λ-phases relative to β-Ti₃O₅. This provides a theoretical basis for tuning the operational temperature window of the material for specific waste-heat recovery applications.
A central challenge in the theoretical description of Ti₃O₅ has been the discrepancy between calculated and experimental pressures required to induce the phase transition. This thesis resolves this issue by moving beyond bulk models. First, the close degree of lattice matching between β- and λ-phases results in stable optimized phase interfaces along the (100), (010) and (001) grain boundaries between the phases. A more refined picture of the phase transition pathway is obtained by varying the β/λ ratio in large supercell models, revealing a distinct anisotropy in the phase transition pathway involving the (001) interface, which proceeds with a significantly lower barrier as compared to the other considered interfaces, confirming experimental trends. Hydrostatic pressure simulation on these mixed-phase models reveals pressure to destabilize high λ-phase fraction systems, however, these pressures of several GPa still overestimate experimental values by orders of magnitude. Second, the influence of particle size and morphology is quantified. By calculating surface free energies and applying the Wulff construction, it is shown that surface effects stabilize the λ-phase in particles smaller than 43 nm in diameter, providing a thermodynamic explanation for the experimentally observed thermal hysteresis and the persistence of the metastable phase at room temperature. The phase transition temperature is also shown to be influenced by particle size, with nanoparticles exhibiting diameters in the experimentally synthesized regime displaying phase transition temperatures in excellent agreement with experiment.
Finally, the atomistic mechanism of the pressure-induced phase transition is elucidated using machine-learned potentials driven on-the-fly probability enhanced sampling simulations. Simulated annealing simulations reveal a favorable surface reconstruction of the (001) λ-phase surface, which is then subject to repulsive harmonic potentials to model pressure effects. A comprehensive, universal and transferable framework for the translation of the slab compression to a pressure value is introduced. By modeling the uniaxial compression of nanoparticle surfaces rather than hydrostatic bulk compression, the predicted transition threshold of ≈700 bar is brought into better agreement with experimental values (≈600 bar). The mechanism is shown to involve a sequential, system size independent nucleation and a resulting layer-by-layer transformation, which is resolved in detail from direct molecular dynamics simulations under pressure at ambient temperatures.
Collectively, this work bridges the gap between quantum-chemical predictions and experimental observations in Ti₃O₅. It establishes a validated computational workflow for the discovery and optimization of phase-change materials, highlighting the critical importance of the correct choice of computational method, uncovering the fundamental effects of cation substitution in the modulation of material properties and the use of realistic models accounting for finite-size effects for accurate predictions of thermodynamic material properties.

Model inference and uncertainty quantification in complex resistivity imaging

Fri, 19 Jun 2026 00:00:00 GMT

Model inference and uncertainty quantification in complex resistivity imaging Hase, Joost This thesis contributes to the solution of the complex resistivity imaging (CRI) problem by introducing an inversion workflow that emphasizes accurate model inference and uncertainty quantification in a probabilistic framework. In applied geophysics, CRI is a frequency-domain (FD) technique for the analysis of tomographic induced polarization (IP) measurements, inferring the complex electrical conductivity distribution in the subsurface from measurements of the complex electrical impedance. Applications can be found across earth sciences. As earth scientists gain a better understanding of how the macroscopic geoelectric properties of earth materials relate to the structures and mechanisms at the pore scale, CRI must meet the increasing demand for accuracy and uncertainty quantification, to ensure that advances made in laboratory studies translate into the field setting.
Despite an increase in the practical utilization of FD measurements during recent years, time-domain (TD) measurements remain the predominant methodology used in field surveys. The first contribution of this thesis is an approach for the conversion of IP measurements from the TD into the FD, which is based on the Debye decomposition of the transients into relaxation-time distributions and the subsequent calculation of the equivalent spectra. The conversion scheme is tested in a synthetic study, confirming its accuracy and the validity of propagated measurement uncertainties. The field application is demonstrated on a data set from Kamchatka (Russia), followed by the inversion of the obtained complex electrical impedances into subsurface images using the established CRI technique.
The second contribution of this thesis introduces a probabilistic framework for the solution of non-linear geophysical inverse problems in complex variables, specifically focusing on the application to CRI. By formulating the likelihood and prior terms as complex probability distributions and combining them into a posterior distribution using Bayes' theorem, the approach can simultaneously account for individual data errors of the real and imaginary data parts, independently regularize the real and imaginary parts of the complex model, and still take into account cross-sensitivities that originate from a complex forward calculation. The complex conductivity image with the highest probability is determined using a Gauss-Newton scheme. The variances and covariances of the inversion result are approximated locally under the simplifying assumption of a linearized forward calculation and normally distributed model parameters. In a synthetic study, the advantages of the probabilistic framework over the established inversion approach are demonstrated.
In the third contribution, the probabilistic framework is used as the basis for a probabilistic inversion of CR measurements using the Hamiltonian Monte Carlo (HMC) method, aiming at a comprehensive characterization of the posterior distribution and an accurate global uncertainty quantification, both of which have not been fully achieved within the second contribution. Convergence criteria are monitored to assess the quality of the probabilistic inversion result, and the final sample is analyzed. Based on the HMC inversion result, the validity of the locally approximated variances and covariances obtained on the basis of the deterministic inversion result is assessed.
The collective contribution of this thesis is an enhanced workflow for CRI based on TD and FD IP measurements.

Titanocene-Catalyzed Hydrosilylation of Cyclic Ethers to Stereochemically Enriched Anti-Markovnikov Alcohols and Fatty Alcohols

Fri, 19 Jun 2026 00:00:00 GMT

Titanocene-Catalyzed Hydrosilylation of Cyclic Ethers to Stereochemically Enriched Anti-Markovnikov Alcohols and Fatty Alcohols Goli, Harie In this thesis, titanocene-catalyzed hydrosilylations are investigated as tools towards pharmaceutically relevant motifs. First, this work establishes a straightforward approach to stereochemically enriched anti-Markovnikov alcohols from a diastereomeric mixture of trisubstituted olefins via enantioselective epoxidation followed by the diastereoconverging Ti-catalyzed hydrosilylation. Central to this strategy is the use of a suitable enantioselective epoxidation method. In this context, the Shi-epoxidation fulfilled all requirements. It controls the absolute configuration of the less-substituted stereocenter regardless of the olefin's geometry. Moreover, it transforms both olefin isomers into their respective epoxides with high enantioselectivity. The subsequent titanocene-catalyzed hydrosilylation erases the stereochemical information at the higher-substituted carbon and establishes a new relative configuration with respect to the less-substituted stereocenter, resulting in a single stereoisomer of the desired anti-Markovnikov alcohol.
Second, this work focuses on optimizing the regioselectivity for the titanocene-catalyzed hydrosilylation of monosubstituted alkyl epoxides. The relatively slow radical opening of these substrates by Cp₂Ti(III)H allows the hydricity of the active species to promote a competing nucleophilic pathway, yielding unwanted Markovnikov alcohols. To suppress this nucleophilic mechanism, the hydricity of the active species is reduced by modifying the cyclopentadienyl ligands of titanocene dichloride with strongly Lewis-acidic silyl groups, which intramolecularly coordinate to the titanium-bound hydride. By additionally using 1,4-dioxane as a cosolvent to precipitate Lewis-acidic Mg salts, the regioselectivity of epoxide opening is improved from a regioisomeric ratio of 63:37 to 93:7.
Finally, in this work, the titanocene-catalyzed hydrosilylation was applied to oxetanes. In contrast to the previous approach using Cp₂Ti(III)Cl as the active species and an external HAT reagent, the hydrosilylation system efficiently reduces oxetanes towards the anti-Markovnikov alcohols without the formation of an elimination side product. The β-hydride elimination is avoided due to the fast intramolecular HAT step. Moreover, the higher stability of oxetanes suppresses the competing nucleophilic side reaction observed for the opening of monosubstituted epoxides. Therefore, no elaborate catalyst design is necessary to achieve high regioselectivity in the opening of monosubstituted alkyl oxetanes. The regioselectivity (r.r. 99:1) is even higher compared to the corresponding optimized epoxide opening reaction (r.r. 93:7) due to the stronger contact of the catalyst's ligands with the substrate in the transition state of oxetane opening.

From Encounters to Guidance: Privacy-Preserving Risk Modeling on Temporal Contact Networks for Proactive Digital Contact Tracing

Fri, 19 Jun 2026 00:00:00 GMT

From Encounters to Guidance: Privacy-Preserving Risk Modeling on Temporal Contact Networks for Proactive Digital Contact Tracing Diallo, M. Diaoulé COVID-19 highlighted the promise and limits of digital contact tracing. Decentralized deployments protected privacy, but alerts were reactive, binary, and often triggered broad quarantines. Network-based targeting can reduce and delay transmission by focusing on high-risk individuals, yet typical models require full visibility of the population contact network, which conflicts with privacy. This thesis investigates which benefits a network-based digital contact tracing policy can retain under a strictly privacy-preserving setup. It designs and evaluates a proactive scheme that combines realistic temporal contact networks with local risk signals and addresses three questions: how to synthesize realistic temporal networks at scale, what can be inferred about individual spreading potential under local visibility, and how effective a proactive local policy can be.
First, it introduces a pipeline that synthesizes temporal contact networks by calibrating human mobility models to empirical contact datasets with Bayesian optimization, and then embeds the calibrated models in an agent-based framework to generate multi-venue temporal networks at scale. The resulting networks reproduce structural and epidemic signatures and provide high-resolution testbeds for evaluation.
Second, the thesis quantifies what can be inferred about an individual's spreading potential from locally visible contacts. Across diverse networks, multiple vital node identification methods are benchmarked under k-hop visibility. Simple degree, which uses only minimal local information, already yields solid estimates of spreading potential. Modestly extending visibility beyond the immediate neighborhood lets more expressive methods approach the accuracy of full-network information. With modest information sharing, lightweight machine learning models that aggregate local neighbor features add gains without sharing raw topology.
Finally, these insights are operationalized as Network-based Proactive Contact Tracing. In simulations on the generated and empirical networks, individuals map recent encounter intensity to a spreading potential-based risk score and compare it to a single epidemic-aware threshold. Exceeding it triggers a graded intervention that reduces future contacts. Evaluation shows infection peak reductions up to 40% while suppressing about 20% of contacts. It outperforms non-adaptive baselines and is robust to parameter choices.
The contributions are: 1) a calibrated generator for realistic multi-venue temporal contact networks; 2) a quantified privacy-accuracy trade-off for ego-bounded risk estimation; and 3) a privacy-preserving, proactive scheme with measured cost-benefit, robustness, and distributional properties. Together, these results provide methods, metrics, and empirical evidence to guide the design and evaluation of alternative digital contact tracing policies based on realistic temporal contact structures and local, privacy-preserving risk estimation.