Tesis Doctorals - Facultat - Física

URI permanent per a aquesta col·leccióhttps://hdl.handle.net/2445/41381

Estadístiques

Examinar

Mostrant 1 - 20 de 243

Design and Synthesis of Out-of-Equilibrium Phospholipid Vesicles
(Universitat de Barcelona, 2026-02-16) Venugopal, Akhil; Battaglia, Giuseppe; Kumar, Mohit; Universitat de Barcelona. Facultat de Física
[eng] This thesis focuses on the design, synthesis and application of out-of-equilibrium (OOE) phospholipids constructed from simple, water-soluble building blocks. Phospholipids are fundamental amphiphiles that spontaneously organize into supramolecular membranes, providing the structural and functional framework of all living cells. Life itself depends on the chemical and physical transformation of diverse biomolecules such as proteins, nucleic acids, carbohydrates, and lipids. Yet these transformations are orchestrated within highly organized compartments defined by phospholipid bilayers of only a few nanometers thick. These membranes are far more than passive barriers; they enable selective transport, spatial organization, and continuous remodeling, thereby supporting essential processes such as signaling, metabolism, and replication across a range of length scales, from nanometers to entire cellular architectures. In this thesis, I draw inspiration from these natural architectures to investigate how OOE processes of living cells can be incorporated into the design of synthetic phospholipid vesicles to achieve life-like features. By integrating metabolic-like fueling strategies, the work demonstrates how artificial membranes can be generated, remodeled, and functionally sustained, offering insights into the design of adaptive systems for applications in nanomedicine. In essence, these dynamic vesicles follow the philosophy of being “of biology, by biology, for biology,” as they are inspired by cellular processes, operate through life-like principles, and aim to engage more effectively with biological systems. Hence, this thesis aims to push systems chemistry towards biology, focusing on how fuel-driven reaction networks can generate and remodel phospholipid compartments that are central to life itself. Chapter 1 provides a comprehensive literature survey on vesicular self-assembly under OOE conditions. It begins with the basic principles of chemically fueled OOE cellular membranes, highlighting how biological processes rely on continuous energy dissipation to maintain dynamic structures and functions. The chapter then briefly reviews synthetic vesicles under equilibrium conditions, which have long served as models for membrane formation but remain inherently static in nature, i.e., under thermodynamic control. Here, we will mainly highlight vesicular assemblies that emerge from in situ lipid-forming systems in aqueous media, generated through chemical reactions between simple water-soluble precursors. The discussion then shifts to recent advances in fuel-driven strategies for the formation of vesicles under OOE, covering both 1) non-phospholipid systems that showcase the breadth of dissipative chemistry and 2) phospholipid-based systems that most closely resemble natural membranes. Together, these studies set the stage for the design of synthetic lipid systems with adaptive, life-like features and thereby laying the conceptual groundwork for the experimental work described in the following chapters. Chapter 2 describes the chemically fueled, OOE self-assembly of synthetic phospholipids into vesicles inspired by the metabolic processes in natural membranes. In living systems, phospholipid turnover is continuously regulated by metabolic cycles that consume fuel to regulate lipid synthesis and degradation, thereby controlling membrane properties. Drawing inspiration from this principle, we developed a minimal chemical reaction network in which simple, water-soluble precursors undergo fuel-driven transformation into double-tailed phospholipids that spontaneously self-assemble into vesicles. Crucially, these assemblies are transient; their lifetime can be tuned from minutes to hours by modulating the kinetic competition between anabolic-like formation and enzymatic catabolism. Beyond their fundamental significance, these nanoscale vesicles offer translational potential through the encapsulation and regulated release of cargo molecules under OOE conditions, opening avenues for adaptive nanomedicine and controlled drug delivery. This chapter represents a cornerstone of the thesis, establishing that phospholipid-based vesicles can indeed be sustained and controlled by artificial metabolic cycles. Chapter 3 extends this concept to the membrane formation and OOE membrane remodeling of phospholipid vesicles. Here, we systematically varied the hydrophobic tails of lipid precursors and uncovered how tail length directly influences membrane formation. Longer hydrophobic chains accelerated vesicle formation, while shorter chains slowed the process. We also allowed the system to compete and self-select by mixing different fuels, thereby revealing the inherent preferences of lipid assembly. Under equilibrium conditions, the system always converged to a mixed population of asymmetric lipids, giving rise to uniquely shaped compartments. In contrast, under dissipative conditions, kinetic asymmetries reshaped the outcome where fast forming lipid species were rapidly depleted, biasing the system towards distinct lipid populations. This chapter highlights, for the first time, how simple chemical design coupled with non-equilibrium conditions can emulate the adaptive remodelling of biological membranes. Finally, Chapter 4 presents an overall summary of the work. It also, provides an outlook on how these findings can be applied toward the generation of more integrated, artificial cellular systems. By combining lipid metabolism with other synthetic reaction networks, it may become possible to design compartments that grow, divide, and remodel in response to external stimuli. Such advances would not only shed light on the chemical origins of life but also pave the way for engineering soft, adaptive materials for nanomedicine. Also, this chapter discusses the challenges of this field, more specifically related to the characterization of dynamic supramolecular events in solution, and highlights the importance of the real-time, in situ visualization of these dynamic supramolecular systems.
A computational study of the Low-Density Lipoprotein Receptor-Related Protein 1 and its ligands
(Universitat de Barcelona, 2026-01-23) Tuveri, Gian Marco; Battaglia, Giuseppe; Franzese, Giancarlo; Universitat de Barcelona. Facultat de Física
[eng] This thesis investigates the structure, dynamics, and molecular recognition properties of the low-density lipoprotein receptor–related protein 1 (LRP1), a key receptor at the blood–brain barrier (BBB). Through a combination of molecular modeling, molecular dynamics simulations, docking analyses, and coarse-grained nanoparticle studies, we provide an integrated description of how modular motifs, hydration, glycans, and nanocarrier electrostatics define receptor function and therapeutic potential. We first generated structural models of LRP1 by integrating domain-by-domain neural network predictions with homology modeling from LRP2. These models revealed the modular repetition of calcium-binding, β -propeller, and EGF-like motifs, and highlighted the role of glycosylation in shaping receptor architecture. Molecular dynamics showed that the flexible R domains undergo large-scale coiling, emphasizing that LRP1 is not a static scaffold but a dynamic receptor whose organization depends on solvent and steric effects. Hydration analysis revealed that conserved amino acids display heterogeneous water interactions depending on their structural context. By introducing a corrected radial distribution function method and solving the Ornstein–Zernike closure iteratively, we constructed free-energy maps of residue–water interactions. This approach was extended to estimate desolvation enthalpies for ligand binding: for example, the binding of angiopep2 to CB19 and CB25 was found to require the removal of hydration shells corresponding to ∼ 20 kBT , resulting in a predicted binding timescale of ∼ 0.6 ms. These results demonstrate that hydration is not passive but an active regulator of binding specificity and kinetics at the BBB. Docking analyses further showed that LRP1 presents extended binding regions rather than a single site, consistent with its ability to recognize multiple ligands. Glycans were found to stabilize the dimeric canopy through protein–glycan and glycan–glycan hydrogen bonds, while their steric repulsion modulates ligand access. Finally, coarse-grained simulations of PEG–PLA micelles reproduced their self-assembly into PLA cores and PEG coronas, while zeta-potential calculations demonstrated how spontaneous hydrolysis at the PEG–PLA junction can generate charged fragments that tune micelle electrostatics over time. Together, these findings emphasize that LRP1 function emerges from the interplay of structure, dynamics, solvation, and glycosylation, and that nanoparticle design must consider not only affinity but also hydration and charge dynamics. This integrated perspective provides both a molecular basis for LRP1-mediated transport and guidelines for the rational engineering of nanocarriers for brain delivery.
Topological defects in out-of-equilibrium systems
(Universitat de Barcelona, 2026-04-09) Rouzaire, Ylann; Levis Sotomayor, Demian Francisco; Pagonabarraga Mora, Ignacio; Universitat de Barcelona. Facultat de Física
[eng] We study topological defects in two-dimensional non-equilibrium systems, focusing on active extensions of the XY model, including activity, mobility and non-reciprocity. In a noisy Kuramoto lattice with short-range coupling, intrinsic frequency heterogeneity destroys quasi-long-range order and fragments the system into finite domains. Defects unbind at all temperatures and exhibit superdiffusive random walks, advected by evolving domain boundaries. By contrast, when oscillators are allowed to move in space, the system undergoes a Berezinskii–Kosterlitz–Thouless transition and regains quasi-long-range order, revealing the fundamental role of motility in sustaining coherence. We also analyse a non-reciprocal O(2) model with vision-cone couplings and derive a continuum theory that captures the same large-scale physics. Non-reciprocity selects defect shapes, enriches the annihilation process, and reshapes patterns through advection. Together, these results elucidate the fundamental role of activity and non-reciprocity in shaping topological defects and ordering in non-equilibrium systems.
Development of a splenic chip model for RBC mechanical Properties Evaluation
(Universitat de Barcelona, 2026-04-17) Mellalakari, Aishwarya; Samitier i Martí, Josep; Rodríguez Trujillo, Romén; Universitat de Barcelona. Facultat de Física
[eng] The spleen plays a critical role in immune defense and red blood cell (RBC) filtration, yet ethical and technical barriers limit in vivo studies, particularly for disorders like sickle cell disease (SCD). This thesis addresses these challenges by developing a biomimetic spleen-on-a-chip platform to replicate splenic microarchitecture, RBC deformability, and macrophage interactions under controlled oxygen conditions. The hypothesis shows that a microfluidic device with narrow interendothelial slits (1.5 µm) and rounded channels (50-139 µm height) can mimic splenic filtration, revealing hypoxia-induced RBC rigidity in SCD and enabling phagocytosis assays. Objectives include designing and fabricating the device using novel hybrid lithography, preparing RBCs and macrophages, and validating RBC biomechanics under normoxia (150 mmHg PO₂) and hypoxia (14-36 mmHg PO₂). Methods involved iterative microfabrication (standard, two-layer, backside, and hybrid lithography), RBC isolation from healthy and SCD donors, monocyte differentiation into M1/M2 macrophages, and analysis via ImageJ, TrackMate, and flow cytometry. Results demonstrated successful fabrication of Hess-Murray-compliant channels with high-fidelity slits, showing healthy RBCs maintained deformability (transit time: 0.0042-0.0060 s, shear modulus: 14.6-20.8 µN/m) while SCD RBCs exhibited sickling, increased transit times (0.0082-0.0154 s) under hypoxia. Macrophages phagocytosed SCD RBCs, validating immune clearance. This platform advances organ-on-a-chip technology, offering insights into SCD pathophysiology and potential for drug screening, bridging gaps in traditional models.
Enhancing Quantum Simulation with Quantum Resources and High Performance Computing
(Universitat de Barcelona, 2026-03-17) Masot Llima, Sergi; García Sàez, Artur; Universitat de Barcelona. Facultat de Física
[eng] In the advent of quantum computing, the technology has constantly had to distinguish its strengths from what we can do with classical means. To a great extent, this is because we have really powerful simulation tools, and the notion of “quantumness" has proved quite hard to capture. Quantum resources, which describe quantum systems from the point of view of which transformations are costly and which are easy, are the key to describe with accuracy the behaviour of quantum systems and characterize this ellusive property. Hence, exploring simulation methods from this point of view stands to gain useful insight, but also produce novel techniques that can be used in high-performance computing to reach new heights in the simulation realm. On the other hand, this also contributes to our understanding of quantum algorithms, and allows us to understand what kind of problems can benefit the most from being solved with a quantum computer. Lastly, bringing these two sides of the problem together, that is, powerful simulation with classical means and useful quantum algorithms, offers the opportunity to develop hybrid classic-quantum algorithms, a direction that some high-performance computing centres are starting to pursue. In this thesis we approach the simulation problem from multiple fronts, from entanglement applied to fermionic systems to resource mixin, with contributions in the problems above. We begin with an introduction on the most important and well-studied quantum resources, and also the basics of high-performance computing. We also present Tensor Networks, an entanglement-based technique that is at the core of our research. Using this tool, we prepare large-scale aimed simulation frameworks and analyze their behaviour with hetereogeneous structures. We put the previous pieces together with the introduction of a new approach that incorporates magic (one such quantum resource) to tensor networks, enhancing its simulation capabilities. This extension, our most novel contribution, successfully explores the intricate interaction between entanglement and magic to simulate quantum circuits and other quantum systems in new ways. Additionally, we use it as a basis for a generalized method that can similarly incorporate other resources to tensor networks. Overall, our analysis on this family of tools and the different methods that work with them reveals an approach that could become crucial to an ultimate description of the frontier between classical and quantum algorithms. Entanglement is also under the spotlight for the analysis of the nuclear shell model, a fermionic many-body system, with surprising results on the nuclear structure. This study naturally leads to the proposal of a hybrid algorithm that improves the performance of current variational algorithms and further contributes to defining this quantum-classical frontier. The dicotomy of simulation methods and quantum algorithms is placed in a common frame with our last contribution of this work, using a classification of quantum circuits based on their output functions in the context of quantum learning models. Such a perspective presents us with the opportunity to discuss how all the moving pieces of this thesis find its place in the current, fast-paced and exciting mainstream of quantum computing.
Challenging Lepton Flavour Universality with rare b-baryon decays at LHCb
(Universitat de Barcelona, 2026-04-24) López Huertas, Albert; Graugés Pous, Eugeni; Marin Benito, Carla; Universitat de Barcelona. Facultat de Física
[eng] This thesis aims to test one of the fundamental properties of the Standard Model of particle physics: the universality of leptonic taste, according to which the interactions of leptons with mediating particles are independent of their taste. Despite the success of the Standard Model in describing numerous phenomena, it has important limitations, such as the lack of a description of gravity, the absence of a dark matter candidate or the impossibility of explaining the asymmetry between matter and antimatter observed in the universe. For this reason, the study of processes sensitive to possible contributions of new physics is essential to advance in the fundamental understanding of nature. The work focuses on the study of rare decays of hadrons containing a b quark, in particular b-to-s type transitions with two leptons at the end of the state. These decays are strongly suppressed in the Standard Model and only occur by second-order processes, which makes them especially sensitive to the effect of new particles not described by the model. In recent years, several measurements related to these processes have shown discrepancies with respect to theoretical predictions, although these are affected by hadronic uncertainties that are difficult to control. To reduce the impact of these theoretical uncertainties, constructed observables are studied as ratios between identical decays except for the flavor of the final lepton. Specifically, this thesis presents a new measure of the universality ratio of leptonic flavor in decays Lb at pK with two leptons, comparing the channels with muons and with electrons. In the Standard Model, this ratio should be very close to unity, with theoretical uncertainties of the order of 1%. A significant deviation would be a clear signal of new physics. The analysis has been carried out with data collected by the LHCb experiment during the Run 1 and Run 2 periods of the Large Hadron Collider, corresponding to the years 2011–2012 and 2016–2018, with a total integrated luminosity of 8.4 fb⁻¹. The main objective is to update a previously published measurement by adding the data from 2017 and 2018, as well as to completely reoptimize the analysis strategy to improve statistical accuracy. In addition, the measurement of the observable is carried out for the first time in two differentiated regions of the moment transferred to the dileptonic system. The measurement is based on a double ratio that uses a resonant channel with J/psi meson decay as normalization. This strategy allows to cancel out a large part of the systematic effects associated with the differences between electron and muon reconstruction. The number of events is determined by adjustments to the invariant mass distributions, while the efficiencies are calculated from simulations corrected with real data. To avoid experimental biases, a blinding method is applied to the efficiencies of rare channels during the development of the analysis. Electron reconstruction presents specific difficulties with respect to muons, due to the emission of bremsstrahlung photons and a lower energy resolution than the momentum measurement system. These differences affect the mass resolution of the electronic channels and require more careful selection and calibration. Strict particle identification criteria, contaminant channel vetoes, and multivariate classifiers are developed to separate the signal from the background. As a validation of the method, the ratio of branch fractions of the resonant channel in muon and electron modes is measured, checking its compatibility with the unit, as predicted by the universality of the leptonic flavor in the decay of J/psi. The results obtained are compatible with this hypothesis within the statistical uncertainties. The ratio of universality of the leptonic flavor is extracted by a simultaneous adjustment to all the samples considered. Results are presented for the combined region of transferred moment and for two subregions, showing a clear improvement in statistical accuracy compared to the previously published measurement. This improvement is higher than the factor expected only by the increase in luminosity, which validates the reoptimization of the analysis strategy. In addition, a competitive measurement is achieved for the first time in the regions of low and medium moment transferred. Finally, the thesis includes a preliminary study of end-state electron decays using data from Run 3, after the upgrade of the LHCb detector. This work constitutes the first analysis of an electronic channel with the new detector and allows to validate the behavior of electron reconstruction and identification in the new experimental environment, characterized by higher occupancy and a completely software-based trigger system. Overall, this thesis provides a more accurate observable measurement of leptonic flavor universality in rare decays, extends the study to new kinematic regions and lays the foundations for future measurements with data from Run 3. These results contribute significantly to the coherence tests of the Standard Model and to the indirect search for new physics.
Chirality as a Control Parameter in Active Microswimmer Systems
(Universitat de Barcelona, 2026-04-29) Lara Chavero, Zaida Zuleica; Pagonabarraga Mora, Ignacio; Universitat de Barcelona. Facultat de Física
[eng] Understanding how microscopic swimming strategies shape the macroscopic behaviour of active suspensions is one of the central challenges of active matter physics. Biological microswimmers — from flagellated bacteria to ciliated protozoa and algal colonies — navigate viscous environments at low Reynolds number through surface deformations that generate self-propulsion without net external forces. The squirmer model, introduced by Lighthill and Blake, captures this physics through a prescribed surface slip velocity and has become a standard theoretical and computational tool for studying microswimmer hydrodynamics. However, the classical squirmer is axisymmetric: it translates along a fixed axis without rotation, an idealisation that excludes the helical and circular trajectories observed in many biological swimmers. This thesis lifts that constraint and asks what new physics emerges when a microswimmer is allowed to simultaneously translate and rotate in three spatial dimensions. The answer, developed across four research chapters using fully resolved lattice-Boltzmann simulations, turns out to be qualitatively richer than any axisymmetric model can predict. The second chapter examines collective behaviour in suspensions of axisymmetric pushers and pullers — swimmers whose far-field flow is extensile or contractile, respectively. We demonstrate that suspension composition is a universal control parameter for polar order: even a minority pusher fraction (χA ≈ 0.2–0.4) acts as an alignment catalyst, organising the majority puller population through its long-range extensile stresses and driving polar order beyond what either pure component achieves at the same activity and packing fraction. This non-monotonic, composition-controlled enhancement is specific to the hydrodynamic semidilute regime and disappears once steric interactions dominate at φ3D ;:: 0.18. The result holds across three-dimensional and quasi-two-dimensional geometries, establishing composition as a design parameter for collective order in binary active suspensions analogous to activity or Péclet number in single-component systems. The third chapter introduces the chiral squirmer model, a minimal extension of the Lighthill-Blake framework that incorporates azimuthal surface velocity modes following the generalised squirmer formulation of Pak and Lauga. The resulting swimmer is characterised by eight fundamental squirming modes that collectively enable simultaneous translation and rotation, producing straight-line, helical, or circular trajectories depending on the chirality angle χ and the propulsion-to-rotation ratio Rcircle. Systematic lattice-Boltzmann simulations of chiral squirmer pairs identify eight distinct hydrodynamic interaction states — extending previous five-state far-field taxonomies by distinguishing three bounded-state variants and introducing the transitory interaction class. The most striking result is a fundamental dichotomy: bounded states, representing stable or quasi-stable swimming pairs that persist indefinitely, form exclusively in asymmetric pusher-puller combinations, where extensile and contractile flow fields combine complementarily when modulated by chiral rotation. Symmetric pusher-pusher and puller-puller pairs produce no bounded states under any parameter combination studied. The rotlet dipole mode C02 plays an essential near-field role in stabilising bounded states and mediating reorientation during close encounters, a contribution that is entirely inaccessible to stresslet-based far-field approximations. The fourth chapter investigates the interaction of a single chiral squirmer with a solid no-slip boundary. The complete phase diagram in the chirality-stresslet (χ, β) plane reveals that the wall acts as a passive chirality sorter governed by two orthogonal control axes. The chirality angle χ selects the qualitative interaction state: above a sharp geometric threshold at χ = 5π/12, the swimmer is deflected by its own self-rotation before the stresslet field can trap it, producing robust scattering that is independent of hydrodynamic character. Below this threshold, all swimmers are permanently trapped at the wall, with β determining whether the trapped state is a stationary bound configuration or a sustained circular orbit, and controlling the orbit radius within the orbiting regime. This orthogonality suggests that a microfluidic channel tuned to the critical chirality threshold would passively separate swimmers by trajectory geometry without any active control. Beyond the (χ, β) phase diagram, activating the lateral squirming mode B11 alongside rotational modes C01 and C02 generates four qualitatively distinct surface locomotion modes — sliding, wall rolling, donut rolling, and reorientation trapping — that have no counterpart in axisymmetric models and arise from specific combinations of the non-axisymmetric mode set. The fifth chapter examines the gravitational sedimentation of chiral run-and-tumble squirmers, combining the geometric effects of helical trajectories with hydrodynamic and stochastic reorientation. The central result is that three control parameters act as largely orthogonal axes in the space of sedimentation behaviour. The activity ratio χ1 sets the overall sedimentation length scale without altering the profile structure. The stresslet β selects the collective structural phase: neutral swimmers form a lamellar phase (circular) or smooth exponential profiles (helical and straight-line); pushers homogenise the suspension and suppress the lamellar phase; and pullers drive the formation of a sustained supernatant above the main sedimentation layer, accompanied by pronounced upper-wall trapping. The rotlet dipole C02 acts as a structural switch that destroys the lamellar phase of neutral circular swimmers by introducing slow orbital axis precession, without affecting the sedimentation behaviour of helical or straight-line swimmers. The spontaneous lamellar phase of neutral circular swimmers — a genuinely collective structural state with layer periodicity ≈ 2a, entirely confined to z/a ,:S 15, and insensitive to swimming activity — is the most unexpected finding of this chapter, with no passive analog and no counterpart in straight-line or helical swimmers. Taken together, the studies consistently support a unified organisational principle: the behaviour of chiral microswimmer systems across all settings examined is governed by the interplay of three largely independent parameters — trajectory geometry, hydrodynamic character, and chiral hydrodynamic stress — that act as orthogonal design axes. This orthogonality reflects the distinct spatial decay rates and symmetry properties of the underlying flow singularities and provides a principled framework for engineering active matter systems by tuning each axis independently. The quantitative phase diagrams and interaction taxonomies produced in this work offer concrete targets for the design of programmable active materials, chirality-selective microfluidic sorters, and synthetic microrobots with controlled surface locomotion modes.
Neutron star matter at finite temperature and implications for neutron star mergers
(Universitat de Barcelona, 2026-01-29) Kochankovski, Hristijan; Ramos Gómez, Àngels; Tolós Rigueiro, Laura; Universitat de Barcelona. Facultat de Física
[eng] This PhD thesis investigates the properties of neutron star (NS) matter, with a primary focus on the appearance and effects of hyperons at finite temperature. The work aims to contribute to resolving the "hyperon puzzle" in a two-fold way. On the one hand, in the thesis a theoretical framework necessary to describe hot and dense baryon matter is developed. The relativistic mean field approach is used, as it naturally incorporates relativistic effects, explicitly includes meson degrees of freedom, and can be generalized to finite temperature. The FSU2H∗ equation of state (EoS) is developed, which integrates the experimental data on hyperon potentials in symmetric nuclear matter and ΛΛ bond energy to ensure consistency with nuclear and hypernuclear physics constraints. Furthermore, two extreme parametrizations of the models are constructed so as to cover the uncertainties in the hyperon-meson couplings arising from the scarce experimental data. The two additional parametrizations, FSU2H∗L and FSU2H∗U, represent the EoSs that satisfy the lower and upper limits of these uncertainties, respectively. It is demonstrated that the models are encompassing the uncertainty range in the total hyperonic content and in the EoS itself in the conditions relevant for supernovae, proto-neutron stars (PNSs), and binary neutron star (BNS) mergers. It is shown that the pressure is especially sensitive to the variation of the hyperon-meson couplings, which in the densest part of the stars shows a 10% difference between the different parametrizations. Other thermodynamical quantities like the energy per particle and entropy per particle only show small variations. With the constructed parametrizations, several stellar observables are computed both in cold and hot stars. For cold NS properties, hyperonic uncertainties significantly impact the maximum NS mass (a difference of 0.14 M between FSU2H∗U and FSU2H∗L) and the radius of massive stars (up to 10% difference for a fixed mass). The tidal deformability of very massive cold neutron stars is particularly affected, with predictions differing by a factor of approximately 2.8 at high NS masses due to these uncertainties. For hot NS properties, calculations for stars resembling phases of the PNS evolution demonstrate that higher temperatures at lower densities lead to lower compactness. The influence of hyperonic uncertainties on hot star observables mirrors that for cold stars, with a greater impact observed in later PNS evolution phases due to lower lepton fractions promoting hyperon appearance. On the other hand, the thesis also searches for model-independent signatures of exotic baryonic matter. Inspecting 14 different hyperonic EoSs and 23 nucleonic EoSs, it is demonstrated that the appearance of hyperons induces a characteristic drop in the thermal pressure and, consequently, the thermal index. This behavior is attributed to the redistribution of degeneracy pressure as new, less degenerate heavy baryons are thermally populated. Furthermore, hypernuclear matter generally exhibits a higher specific heat at constant pressure compared to purely nucleonic matter, indicating an enhanced capacity to store thermal energy due to the opening of new degrees of freedom. With an extensive simulation campaign using all EoSs at our disposal, the imprints of hyperons on BNS merger observables and remnant evolution quantities are studied. The main finding is a systematic positive shift (2-4%) in the dominant post-merger GW frequency, which is identified when hyperonic models are compared to reference simulations that assume an ideal nucleonic thermal behavior. This shift is anti-correlated with the average thermal index in matter and is robust across variations in total binary mass and mass asymmetry. The shift, estimated to be between 50 and 150 Hz, is potentially detectable by future gravitational-wave observatories such as the Einstein Telescope or Cosmic Explorer. In addition, the secondary GW peak f2 0, exhibits similar positive shifts, while fspiral remains largely unaffected by the thermal behavior. The numerical robustness of the identified GW shifts is confirmed through comparisons with different numerical schemes and resolution studies. Some non-spectral observables are studied too. Hyperonic models predict larger dynamical mass ejecta (up to a factor of two) compared to purely nucleonic models with similar cold NS properties. The presence of hyperons also leads to a reduction in the threshold mass for prompt black hole collapse by approximately 0.05 M , attributed to the thermal softening of the EoS.
Exploiting hPSC-derived kidney organoids to study nephrogenesis and Congenital Anomalies of the Kidney and the Urinary Tract
(Universitat de Barcelona, 2026-04-17) Amato, Gaia; Montserrat Pulido, Núria; Garreta Bahima, Elena; Universitat de Barcelona. Facultat de Física
[eng] Human pluripotent stem cells (hPSCs) possess the capacity for unlimited self-renewal and the potential to differentiate into all human cell types. These properties have been extensively exploited to recapitulate key morphogenetic processes in vitro, such as human kidney development. Although animal models have been indispensable for elucidating the cellular and molecular mechanisms governing nephrogenesis, species-specific differences in developmental timing, transcriptional regulation, and tissue architecture often result in divergent outcomes when comparing orthologous gene manipulations between animals and humans. These limitations underscore the importance of hPSC-based models, which enable the study of human kidney morphogenesis under defined in vitro conditions while allowing precise genetic manipulation. The advent of CRISPR/Cas9-mediated genome editing has greatly expanded the potential of hPSCs, allowing the generation of isogenic lines for detailed analysis of developmental processes and disease initiation. When combined with differentiation into kidney organoids, this approach provides a versatile system to interrogate kidney development and model the onset of Congenital Anomalies of the Kidney and Urinary Tract (CAKUT). Within this framework, reporter knock-in lines represent powerful tools for dynamically tracking and isolating differentiated populations for molecular characterization, whereas gene knockout lines provide a complementary approach to dissect gene function during nephrogenesis modeled in kidney organoids. In this thesis, four distinct hPSC-derived reporter lines (GATA3-mOrange, SIX2-GFP, MEIS1-mPlum, and LRP2-GFP) were established, each marking a distinct compartment of the developing nephron. Together, they constitute a modular resource to study nephrogenesis across spatial and temporal scales. In parallel, loss-of-function mutations of PAX2 and HNF1B were introduced into the GATA3-mOrange background, enabling simultaneous lineage tracing and phenotypic assessment of selected CAKUT conditions. These knockout models revealed distinct, gene-specific requirements for nephron formation, elucidating the role of PAX2 and HNF1B in human nephrogenesis and the developmental defects arising from their loss. Overall, this work establishes a comprehensive set of hPSC-based reporter and knockout lines that illuminate fundamental aspects of human nephrogenesis and provides mechanistic insight into CAKUT-associated gene function. Beyond advancing basic developmental biology, these models offer a robust platform to refine differentiation protocols, model disease phenotypes in a human context, and support translational applications in drug discovery, regenerative medicine, and therapeutic innovation.
Soft Polymer Gels and Structure / Dynamics Decoupling through 3D-Dynamic Light Scattering
(Universitat de Barcelona, 2026-01-30) Arenas Gullo, Adrián Antonio; Fernández-Nieves, Alberto; Universitat de Barcelona. Facultat de Física
[eng] Soft matter—such as colloidal suspensions, polymer gels, and even our own bodies—is as ubiquitous in our daily lives as it is challenging to describe within a unified theoretical framework. The diverse phenomena exhibited by these systems emerge from the interplay between entropy and relevant interactions, such as van der Waals forces. To be largely self-contained, this thesis first introduces the key soft matter concepts relevant to the presented experimental work and simulations, including the physics and phase behavior of colloids and the prop-erties of polymer gels. This is followed by an overview of light scattering theory within the Rayleigh-Gans-Debye approximation, and an in-depth description of the 3D-dynamic light scattering (3D-DLS) cross-correlation scheme, which is used to “suppress” multiple-scattering contributions to the scattered intensity, which become significant in turbid colloidal sus-pensions. The scheme uses two beams and two detectors, producing four scattering vectors, q—two of which are identical. When computing the correlation of scattered intensities associated with the different pairs of vectors, only the term for the identical pair is correlated and, as demon-strated by Schätzel, the corresponding multiple-scattering contribution is negligible. Attention is given to the experimental implementation, including optical alignment and limitations related to signal detection and correlation measurement. This is especially relevant for real experimental systems, which may differ substantially from idealized models. After introducing the theoretical and experimental aspects of 3D-DLS, we will present the main results of this thesis. Colloidal suspensions con-sisting of thermoresponsive microgels are probed in static and dynamic light scattering (SLS and DLS) experiments, showing that the connection between the q-dependent relaxation time and the static structure factor—known as de Gennes narrowing—breaks down under certain conditions, in which the relaxation times remain essentially invariant with respect to the structural length scales. In particular, this takes place at low temperatures, where the gel morphology suggests increased softness. The effect of single-particle elasticity on the coupling between structure and dynam-ics is then confirmed in simulations, where a multi-Hertzian potential is used to emulate the compressible and deformable nature of these micro-gels. As the potential is tuned—with respect to the extent of each layer and their elastic moduli—to follow the actual morphology of the micro-gels, a qualitatively similar behavior is observed. Finally, the approach to the glass transition is analyzed for the experiments and simulations, resulting in a change from an Arrhenius-like behavior to a VFT law, for the conditions where the gels are softer and stiffer, respectively. This apparent change in fragility—from strong to fragile—however, seems to be caused by the use of a generalized volume fraction based on the dilute particle size, instead of the actual volume fraction. The second investigation in this work focuses on macroscopic thermore-sponsive polymer gels, which undergo a volume phase transition when the temperature rises above their lower critical solution temperature (LCST). In particular, when subjected to a rapid temperature quench above the LCST, an impermeable skin forms, constraining the volume and induc-ing phase separation between solvent-rich and solvent-poor regions. This separation, originating from the thermodynamic instability at play, can lead to large deformations of the gel, hence the term extreme thermo-dynamics. For cylindrical gels—or more generally, gels with a circular cross-section—a polarization, p, is introduced to describe the deviation of the solvent distribution from axisymmetry. This polarization reduces the free energy and, when the gel’s centerline is curved, couples to the cur-vature, generating a feedback loop in the gel’s evolution. Beyond these deformations, additional instabilities arise from temperature quenches, producing patterns that depend on geometry and temperature. Finally, some preliminary results on complementary techniques are presented. On the one hand, multi-speckle light scattering (MSLS) is intro-duced along with representative results and experimental considerations. This technique enables the probing of colloidal suspensions both tempo-rally and spatially, which is particularly useful for investigating hetero-geneous or time-evolving systems, such as colloidal glasses. On the other hand, preliminary small-angle X-ray scattering (SAXS) measurements are performed and analyzed. These calibration measurements are intended to support future studies on the internal structure of macroscopic alginates.
Emergent Complex Socio-economic Phenomena in Urban Environments. Hierarchies, Segregation and Cooperation
(Universitat de Barcelona, 2026-05-18) Sadurni Parera, Marc; Perelló, Josep, 1974-; Montero Torralbo, Miquel; Universitat de Barcelona. Facultat de Física
[eng] Urban environments are complex socio-economic systems in which inequalities, segregation, and cooperation emerge from the interplay between individual decisions, group dynamics, and institutional structures. This thesis explores how computational and data-driven tools from complexity science can be used to ex-plain, simulate, and quantify such emergent large-scale social patterns. Grounded in a perspective specific to statistical physics, it integrates agent-based modelling, stochastic processes, network science, and game theory to investigate the micro-motives behind urban inequality, segregation, and collective action. The first part of the thesis models the spontaneous emergence of social hierarchies through agent-based simulations. It shows that even symmetric groups can evolve into highly unequal status distributions through self-reinforcing local interactions. The system exhibits identifiable phase transitions and scaling laws, independent of system size, highlighting how dominance structures can emerge endogenously through inter-group dynamics in competitive societies. The second part reconstructs co-residence network using high-resolution ad-ministrative registration data from the city of Vienna, covering local and migrant populations. The analysis reveals socio-spatial segregation driven by neighbourhood wealth, neighbourhood diversity, and nationality-based homophily. This network-based clustering uncovers latent group structures and offers a generalisable and novel method for mapping urban socio-economic segregation patterns in multicultural urban contexts beyond traditional spatial metrics. The last part draws on a citizen science experiment conducted in Barcelona, where participants engaged in a collective-risk dilemma for air pollution mitigation action under unequal initial resources to reproduce real-world socioeconomic heterogeneity. The study analyses decision-making behaviour and finds that less-endowed individuals contribute proportionally more to the public good, high-lighting asymmetries in perceived responsibility and capacity. By combining real behavioural data with agent-based simulations led by conditional probabilities, the experiment challenges assumptions in classical models of cooperation and underscores the role of inequality in shaping strategic behaviour. Together, these case studies show how tools from complexity science can advance both theoretical and applied understanding of urban social dynamics. By bridging abstract modelling, empirical data, and behavioural insights, the thesis contributes to the study of statistical physics applied to modern cities, offering a deeper understanding of the structural mechanisms driving inequality, segregation, and cooperation. It proposes a multi-method framework for diagnosing feedback loops in urban complex systems and guiding policy interventions to-ward more inclusive and equitable urban environments, combining quantitative rigour with societal relevance. Agent-based modelling, network science, and game theory share key foundational principles with statistical physics, particularly the emphasis on emergence, collective dynamics, and bottom-up interactions. In this thesis, these frameworks are not treated as subsets of physics but as complementary perspectives aligned by a shared epistemological philosophy: that complex social phenomena arise from the interactions of many simple units. These approaches highlight the potential of physics-inspired models to support the co-production of knowledge among researchers, institutions, and citizens.
Motion of Enzyme-Powered Nanomotors in Complex Media
(Universitat de Barcelona, 2025-12-19) Ruiz González, Noelia; Sánchez Ordóñez, Samuel; Universitat de Barcelona. Facultat de Física
[eng] In recent years, the field of NMs has experienced rapid growth in biomedical applications, largely due to their ability to navigate complex physiological environments. However, unlocking their full clinical potential requires a deeper understanding of NMs modulate with and interact with biological interfaces. This doctoral thesis aims to address this need by focusing on the design, synthesis, and evaluation of enzyme-powered NMs capable of propelling in viscous biological environments. To this end, this PhD thesis investigates a variety of NPs platforms, ranging from inorganic and polymer-based nanocarriers, functionalized with enzymes to produce enzyme-powered NMs. These NMs are evaluated across clinically relevant scenarios, including navigation in biological barriers, tissue regeneration or drug delivery. The first part of the thesis focuses on the development of inorganic-based NMs, such as those based on MSNPs, demonstrating a dual-enzyme NMs strategy to enhance diffusion through highly viscous SF. For that, two distinct NM actuating in “troops” were developed, capable of reducing SF viscosity and self-propel more effectively. The synergistic interaction between these NM systems significantly improved the transport of macromolecules across SF-mimicking environments, offering a potential strategy for intra-articular drug delivery in joint diseases. Building on these findings, the second part of this thesis continues within the same biomedical application framework but introduces key advancements in NM design. In this part, a new generation of nanogel-based NMs (NGs-NMs) composed of synthetic polymers are introduced to substitute the inorganic core of NMs. These NGs are soft and flexible, and enable motion in viscoelastic fluids while preserving the bulk rheological properties of SF. Their structural flexibility and tunable responsiveness to environmental changes (e.g., temperature, pH, or redox conditions) are achieved through controlled polymer crosslinking, enabling dynamic modulation of size and density allowing for improved interaction with ECMs. After surface functionalization with urease, NGs-NMs achieved efficient propulsion in viscous media at low urea concentrations. Importantly, they exhibited rapid cellular uptake and were further evaluated as active transport carriers for growth factors, as IGF-1, preserving the bioactivity of the conjugated IGF-1 and demonstrated pro-regenerative effects in chondrogenic cell model. The third part of this thesis further explores the application of NG-NMs for antimicrobial therapy, extending their use to other biomedical applications that require navigation across mucosal barriers for effective drug delivery. In this case, hyaluronic acid (HA), a naturally occurring mucoadhesive biopolymer, was employed as the core material to construct enzyme-powered nanomotors (HA-NMs) functionalized with urease. These HA-NMs were designed to actively cross mucin environments and deliver therapeutic agents, such as antibiotics. Their performance was assessed in mucosal models, highlighting their potential to overcome barriers associated, for instance, with antimicrobial resistance and enable targeted treatment at mucosal interfaces. Their performance was evaluated in vitro using mucosal models, including transwell assays, which confirmed their ability to penetrate mucin barriers and significantly reduced bacterial proliferation. When loaded with antibiotics, HA-NMs exhibited enhanced antibacterial activity against Escherichia coli compared to both free antibiotic and reference MSNPs-NMs from the first part of the thesis. The results presented in this thesis highlight the transformative potential of enzyme-powered NMs as highly versatile platforms for targeted therapeutic delivery. Their ability to actively navigate complex and viscous biological environments and enhance therapeutic efficacy represents an important breakthrough in the field of nanomedicine. These findings open the door for a new generation of self-propelled nanotherapeutics, bringing us closer to their clinical translation and opening new avenues for precision treatment in challenging pathological contexts.
Influence of the Quasi-Biennial Oscillation on the tropical troposphere at interannual timescales
(Universitat de Barcelona, 2026-01-09) Rodrigo Sánchez, Mario; García-Serrano, Javier, 1980-; Bladé, Ileana; Universitat de Barcelona. Facultat de Física
[eng] The Quasi-Biennial Oscillation (QBO), an atmospheric phenomenon characterized by alternating westerly and easterly winds that descend through the equatorial stratosphere, is the dominant mode of tropical stratospheric variability. Several studies have explored its connection with the extratropical stratosphere and the troposphere, including its connection to El Niño-Southern Oscillation (ENSO), the main driver of interannual variability in the tropical troposphere. While it is accepted that ENSO exerts an upward impact on the QBO, manifested through changes in the QBO amplitude, period and downward propagation rate, the potential downward influence of the QBO on the tropical troposphere, and on ENSO in particular, has received less attention. Clarifying this downward influence is a difficult scientific problem because the QBO signal in the troposphere is modest and often masked by the much stronger variability associated with ENSO and other variability modes. Moreover, isolating the QBO influence requires long records and/or specifically designed model experiments to separate its effects from those of concurrent variability. This thesis examines the downward impact of the QBO on tropical convection and large-scale circulation on interannual timescales, using reanalyses and a hierarchy of climate model experiments, principally the European Consortium Earth system model (EC-EARTH). As a preliminary step, the performance of EC-EARTH in simulating the stratospheric circulation and the main features of the QBO was assessed, to provide the necessary validation for the subsequent analyses. Overall, the QBO in EC-EARTH is realistic, although its amplitude is underestimated in the lower stratosphere. Reanalysis data and coupled simulations with EC-EARTH were analyzed to examine the QBO-induced changes in the tropical circulation under climatological conditions and during El Niño events. Results show that the QBO affects upper-tropospheric divergence over the Maritime Continent, resulting in reduced outflow during the westerly phase as compared to the easterly phase. Furthermore, during El Niño events, which are characterized by a weakened zonal overturning circulation, i.e. the Walker cell, the QBO signal extends downward into the troposphere and influences El Niño evolution. The westerly QBO phase acts to further suppress summer tropical convection over the Maritime Continent and the western Pacific, thereby accentuating the weakening of the Walker circulation during El Niño. These results highlight the importance of considering the QBO for improving El Niño prediction and projection, especially for extreme events known as super El Niños. While results from that first study reveal a QBO influence, they also show that disentangling the QBO teleconnections in the tropical troposphere from the dominant influence of ENSO is challenging. To better isolate the QBO signal, an atmosphere-only experiment with climatological boundary conditions was then analyzed and treated as an ENSO-neutral state. This part of the thesis focused on identifying QBO-induced changes in temperature and zonal wind in the upper troposphere-lower stratosphere (UTLS) and their subsequent impact on static stability, wind shear and relative vorticity. The results confirm that the QBO effectively modifies vertical velocity and precipitation over the Maritime Continent region, but further show that the QBO affects both the Walker circulation and, more notably, the Hadley circulation. These impacts are highly seasonally-dependent, being strongest in summer. The vertical structure of the QBO signal in the UTLS is found to be zonally asymmetric, with anomalies descending into the upper troposphere only over the Indo-Pacific region. The timing of the QBO influence on tropical convection and precipitation is primarily due to the QBO-induced changes in static stability, which descend into the UTLS earlier than those in wind shear and vorticity. To assess how ENSO modulates the QBO teleconnection to the troposphere, two additional atmosphere-only experiments with perpetual El Niño and La Niña conditions were analysed, complementing the ENSO-neutral experiment. Results confirm that the QBO affects summer tropical convection even under a strong oceanic forcing such as ENSO, but its influence depends on the position of convection. During La Niña, when convection remains close to its climatological position but intensifies, the QBO signal closely resembles that in ENSO-neutral conditions, with the westerly QBO phase reducing summer convection north of the equator over the Maritime Continent. Instead, during El Niño, when convection shifts equatorward over the Indian-Pacific region, the QBO-related reduction in convection likewise shifts equatorward. ENSO also affects the QBO period and downward propagation rate, with a longer period and slower descent occurring under La Niña conditions, and a shorter period and faster descent under El Niño conditions. The latter causes the QBO phase to transition more quickly during El Niño, leading to a reversal of its impact on tropical convection from early to late summer: in early summer, anomalous descent dominates over the Maritime Continent, whereas anomalous ascent prevails in late summer over the western tropical Pacific. The results based on EC-EARTH thus suggest that ENSO modulates the QBO teleconnection in the tropical troposphere through changes in the position of convection and the QBO period/downward-phase propagation. In the last part of the thesis, the robustness of these results was evaluated using the multi-model ensemble from the QBO initiative (QBOi) of Atmospheric Processes And their Role in Climate (APARC). Most QBOi models reproduce similar patterns in the three experiments—ENSO-neutral, El Niño, and La Niña—though the lower-tropospheric signal is generally underestimated. The QBO impact on the tropical circulation is spatially robust across models, but its timing varies, ranging from May to November.
Development of a fast-timing detector for Time-of-Flight Applications: enabling new opportunities in Positron Emission Tomography and Velocity Map Imaging Mass Spectrometry
(Universitat de Barcelona, 2025-11-14) Mariscal Castilla, Antonio; Guberman, Daniel Alberto; Gómez Fernández, Sergio; Universitat de Barcelona. Facultat de Física
[eng] Silicon photomultipliers (SiPMs) have driven recent advances in time-of-flight radiation detectors owing to their single-photon sensitivity, excellent time resolution, and compact form factor. To support the development of fast and scalable detection systems, new front-end electronics are required that can process the fast signals from SiPMs while maintaining low power consumption and minimal size. Application-specific integrated circuits (ASICs) represent the most promising solution, offering a high degree of customization that enables the design of power-efficient, compact, and cost-effective electronic systems. This thesis presents the design and testing of a novel SiPM-based radiation detector that employs the FastIC ASIC. FastIC is an analog front-end that extracts the arrival time and peak amplitude from eight SiPM channels, with a power consumption of approximately 12 mW per channel. This allows for the development of compact and scalable detector systems. The performance of the proposed detector is evaluated in two applications: time-of-flight positron emission tomography (TOF-PET) and time-of-flight mass spectrometry, with a particular focus on velocity map imaging mass spectrometry (VMImMS). In time-of-flight positron emission tomography (TOF-PET), the com-bination of SiPMs and fast scintillators has enabled the achievement of coincidence time resolutions (CTR) below 200 ps FWHM in commercial scanners and below 100 ps FWHM in laboratory settings. However, sub-100 ps performance typically requires bulky, high-power electronics, limiting the scalability of these results for clinical systems. In this context, FastIC emerges as a promising candidate for achieving high timing performance with low power consumption. To evaluate FastIC capabilities, I measured the CTR of single-channel detectors using modern SiPMs and scintillators. Using a 3 mm-thick LSO crystal coupled to an FBK SiPM, CTR values of approximately 76 ps FWHM were obtained. When using LYSO crystals with dimensions comparable to those in commercial scanners, CTR values around 127 ps FWHM were achieved. Measurements with pure Cherenkov radiators demonstrated that FastIC, in combination with SiPMs, can process prompt light with CTR performance comparable to that of high-power electronics reported in the literature. In time-of-flight mass spectrometry (TOF-MS), SiPM-based detec-tors offer a promising alternative to conventional microchannel plate (MCP)-based detectors by addressing key limitations such as restricted ion rates, aging effects, the need for high vacuum conditions and low sensitivity for high mass-to-charge ratio (m/z) ions. Furthermore, due to limitations in MCP readout systems, current detectors are unable to achieve simultaneous sub-nanosecond time and sub-millimeter spa-tial resolution. This constraint prevents the measurement of full three-dimensional velocity distributions of molecular fragments in velocity-map imaging mass spectrometry (VMImMS). The proposed ion detector consists of a fast scintillator, an array of SiPMs, and FastIC-based readout electronics. Two configurations are considered: direct ion detection using the scintillator, or enhanced signal amplification by employing first a single-stage MCP to convert the ions to thousand of electrons. A small prototype was developed, comprising a 16-channel SiPM array, a fast organic scintillator, and FastIC readout electronics. This prototype was tested in a custom-built velocity-map imaging TOF-MS instrument and used to acquire TOF-MS spectra of C3H6 and CF3I molecules. The detector successfully measured ions with m/z of 196 and 18, achieving time resolutions of approximately 3.3 ns and 2.5 ns FWHM, respectively, thus demonstrating its ability to reconstruct TOF-MS spectra with high precision. A key feature of the detector is its potential to operate at ion fluxes up to 109 cm−2 s−1, exceeding the performance of conventional MCP-based detectors. The single-stage MCP configuration enhances sensitivity to ions with high m/z ratios while maintaining a high maximum ion processing rate. Two critical effects were investigated to assess the feasibility of using SiPMs for MCP readout: the impact of SiPM dark count rates (DCRs) on detector sensitivity, and the influence of time walk introduced by fluctuations in the MCP electron yield on time resolution. The first effect was mitigated by adjusting the FastIC discriminator threshold to achieve DCRs comparable to those of MCPs. An optimal threshold was identified at the 8 photoelectron (phe) level for an SiPM bias voltage of 55 V using a Hamamatsu device. To address the second effect, time walk was corrected by measuring signals of varying intensities from a picosecond laser, yielding a corrected time resolution of approximately 120 ps FWHM. To achieve the sub-millimeter spatial resolution required for full 3D velocity imaging, a novel camera design was proposed. This design employs an MCP for signal amplification and an optical window to distribute scintillation light across multiple SiPMs. Monte Carlo simulations indicated that spatial resolutions below 200 µm can be attained, demonstrating the system capability to achieve sub-millimeter precision. Additional VMImMS simulations confirmed that, with the achieved time and spatial resolution, the detector can successfully measure the three-dimensional velocity distributions of molecular fragments. The detector technologies developed in this thesis present significant opportunities across a broad range of applications. In medical imaging, improved time-of-flight resolution in PET detectors enhances image quality and sensitivity, enabling faster scans and supporting the development of more accessible and cost-effective systems. In mass spectrometry, a SiPM-based ion-to-photon detector facilitates the de-sign of compact and portable instruments suitable for high-throughput screening in fields such as drug discovery and toxicology. When combined with a MCP, the detector can achieve enhanced sensitivity to ions with high m/z ratios, expanding the potential of TOF-MS for pesonalized medicine. In velocity-map imaging, the capability to record three-dimensional molecular movies broadens its applicability to the study of more complex molecular systems.
Hyperbolic Cartography of Complex Networks. Designing Maps for Unipartite, Bipartite, and Feature-Enriched Graphs
(2025-09-30) Jankowski, Robert; Serrano Moral, Ma. Ángeles (María Ángeles); Boguñá, Marián; Universitat de Barcelona. Facultat de Física
[eng] Cartography is the science and practice of creating and utilizing maps. Originating in ancient times, it flourished notably during the Age of Discovery and remains essential today through digital mapping applications. Historically, maps have served both to depict the world generally and to facilitate navigation and wayfinding. Nowadays, we regularly rely on tools such as Google Maps to commute within cities through transportation networks, including metro, bus, and tram systems. These physical networks are naturally embedded in Euclidean space. However, many social, biological, or technological networks lack such apparent spatial structures. In this Thesis, we develop methods to chart multidimensional hyperbolic maps for a diverse range of complex systems, encompassing unipartite, bipartite, and feature-enriched networks. These maps provide new means to interpret, analyze, and visualize these systems. Studying interactions between discrete entities has emerged as the study of complex networks in recent years. Many networks share similar topological properties, such as the small-world effect, heavy-tailed degree distributions, high clustering coefficients, and sparsity. The network geometry framework was proposed to explain these properties. In this approach, nodes are located in the hidden latent space, where each node’s coordinates reflect its similarity to other nodes and its intrinsic popularity, thereby intertwining the geometry with the network topology. Models from this family are able to explain many properties of real networks. These latent-space models accurately replicate empirical observations by interpreting connection probability as a decreasing function of hyperbolic distance. The quest to reverse-engineer these geometric models gave rise to model-based hyperbolic embeddings. By projecting complex networks into low-dimensional hyperbolic spaces, graph operations, such as community detection, node classification, and link prediction, become efficient vector computations. To date, however, these methods have been limited to one dimension, even though there is no reason for this to be the case. In this Thesis, we develop three multidimensional hyperbolic embed-dings: D-Mercator for unipartite networks, FiD-Mercator for feature-enriched networks, and B-Mercator for bipartite networks, demonstrating that these embeddings significantly enhance our understanding of network structures. With D-Mercator, we estimated the intrinsic dimensionality of real networks in terms of navigability and community structure. FiD-Mercator emphasized the importance of quantifying the correlations among node labels, graph topology, and node features for downstream tasks. Whereas, B-Mercator revealed patterns of linguistic and geographic diversity, and also improved the node classification task. Moreover, we bridged the gap between network geometry and graph machine learning by introducing HypBench, a benchmarking framework for graph neural networks. HypBench provided concrete guidance for selecting the optimal model for specific graph datasets. Together, these contributions establish a flexible, multidimensional framework for mapping and analyzing complex networks, offering novel insights and practical tools for advancing research across multiple disciplines.
The Multisymplectic Geometry of Classical Field Theories on Finite-Dimensional Covariant
(Universitat de Barcelona, 2025-07-24) Guerra IV, Arnoldo; Román-Roy, Narciso; Universitat de Barcelona. Facultat de Física
[eng] This thesis presents new developments in the De Donder–Weyl formulation of classical field theories, which treats space and time on an equal footing. The formal Lagrangian geometric construction of field theories takes place on manifolds of jets, giving rise to the standard Euler–Lagrange equations, while the equivalent Hamiltonian formulation of De Donder–Weyl takes place on affine dual jets, giving rise to the Hamilton–De Donder–Weyl equations. Manifolds of jets (and their duals) act as finite-dimensional phase spaces, and their multisymplectic geometry acts as a generalization of symplectic geometry in classical mechanics, underpinning variational calculus and providing a powerful set of tools for the investigation of Noether symmetries and bonds. In addition, the multisymplectic construction of relativistic field theories preserves covariance throughout the analysis of these theories and, by working on sections of jet manifolds (and their duals) over space-time, it is possible to derive the covariant phase space formalism from Zuckerman [152], Crnković and Witten [34], and Lee and Wald [108], best known in physical literature and, in general, of infinite dimension. The original contributions of this thesis include new properties of De Donder–Weyl bonds and natural symmetries, which we present as recently proven mathematical propositions and can be found in the following publications: [63, 75, 76]. This work also provides a multisymplectic construction of the Poisson parentheses introduced by Marsden et al. [116] to obtain the Hamilton–De Donder–Weyl equations from the action functional, in direct analogue with the analysis of classical mechanics. In addition, we show how our geometric interpretation of these Poisson parentheses also leads to field equations and discuss their relationship to the infinite-dimensional covariant formalism mentioned above. Finally, we offer a first step towards understanding the multisymplectic geometry associated with the BRST–BV analysis of gauge theories in the Lagrangian framework. After constructing the BRST symmetry, we find the corresponding multimoment application and show that it provides the standard BRST charge except for a total derivative. The novelties presented in this thesis offer a new selection of geometric techniques that allow us to move from the infinite-dimensional language of local functionals to the finite-dimensional language of vector fields and differential forms, treating space and time on an equal footing.
B radiative decays at LHCb: measurement of B→ K∗γ isospin asymmetry and preparation of Run 3 analyses
(2025-09-08) Lobo Salvia, Aniol; Graugés Pous, Eugeni; Marin Benito, Carla; Universitat de Barcelona. Facultat de Física
[eng] The Standard Model (ME) of particle physics is the theory that describes fundamental particles and their interactions, except gravity. Despite its success and accuracy in predicting phenomena, it is known that it is not a complete theory. It does not describe dark matter or energy, the asymmetry between matter and antimatter in the Universe, or the origin of neutrino mass. Consequently, it is vitally important to investigate possible extensions of this model and measure processes that may reveal new physics. An experiment that contributes to this task is the LHCb, one of the four main detectors located in the LHC particle accelerator operated by CERN. His study focuses on high-precision measurements of particles involving heavy quarks b and c. Rare radiative decays of B mesons are an ideal platform to test EM and study its possible extensions. These decays cannot occur through tree diagrams within the description of the ME, but take place through loops in which unknown particles could participate, thus altering the observable quantities. This thesis presents three interconnected topics within the study of rare radiative decays of B mesons: the experiment's calorimeter monitoring tools, the radiative inclusive selections in the trigger system, and the measurement of the isospium asymmetry of the B→K*γ decay. The measurement of the isospium asymmetry of the B→K*γ decay has been done using data from Run 2, which covers the years 2015 to 2018. The other two tasks have been carried out within the framework of Run 3, starting in 2021. Between these two Runs, an ambitious improvement program, called Upgrade I, has made it possible to increase the capacity of the detector. Some of the subdetectors have been completely renewed, the data reading system has been improved to facilitate 40 MHz reading and the hardware phase of the trigger has been dismantled, becoming the first large detector with a completely software-based trigger. This set of improvements has made it possible to increase the instantaneous luminosity at which the detector operates by a factor of five, thus gaining sensitivity and its precision measurements. The monitoring of the calorimeter makes it possible to detect anomalies in the data collection and provides information on the origin of the problem in order to be able to solve it quickly. It has been a critical function during the commissioning period of Run 3, until the system has reached its stable operation. Trigger selections inclusive of radiative processes make it possible to capture events of this type and rule out those that constitute background noise, thus keeping the volume of data stored within the limits of the system. The selections for the new instantaneous high luminosity environment have been optimized and the BDT multivariable method has been used in order to obtain the best possible signal efficiency. Finally, the measurement of the isospy asymmetry of the B→K*γ decay is obtained for the first time with data from the LHCb and with an accuracy comparable to the measurements of other experiments, validating the ME and applying constraints on its possible extensions, in particular extensions of supersymmetry and extra Higgs sectors.
Artificial 3D Skeletal Muscle Tissues to Model and Investigate Disease Onset and Progression Mechanisms
(Universitat de Barcelona, 2025-07-22) Mughal, Sheeza; Ramón Azcón, Javier; Fernández-Costa, Juan Manuel; Universitat de Barcelona. Facultat de Física
[eng] This thesis investigates the innovation and application of three-dimensional (3D) in vitro skeletal muscle tissue models to study and model muscle pathophysiology, particularly in idiopathic and corticosteroid-induced myopathies. Based on the historical evolution of cell culture techniques, this work highlights the limitations of conventional two-dimensional systems in reproducing native muscle structure, bioenergetics, and functional features. In the first section of this work, 3D human skeletal muscle constructs were developed using human muscle progenitor cells and exposed these tissues to sera from patients with Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and long COVID (LC-19). Structural, functional and transcriptomic analyses post-treatment revealed disease-specific, time-dependent disturbances in mitochondrial function, protein homeostasis, and inflammatory signaling. The disease pathology in these models demonstrated mitochondrial stress, IL-6–mediated catabolism, and metabolic reprogramming toward glycolysis, with ME/CFS samples showing particularly severe contractile deficits. In the second part, a steroid myopathy model was produced by exposing 3D muscle tissues to dexamethasone. Taurine supplementation effectively restored contractile function by promoting protein synthesis and suppressing proteolysis by regulating the AKT/mTOR pathway modulation, even in the continued presence of dexamethasone. Together, these models provide powerful platforms for investigating disease mechanisms and testing therapeutic interventions, offering new mechanistic insights into muscle dysfunction. These insights are particularly important in conditions with unknown etiologies. Future prospects include developing an in vitro construct together with immune and neural components, increasing patient cohorts, and improving these models for clinical translation, particularly for testing patient stratification.
Some considerations about the climate variability and change in the Mediterranean region
(Universitat de Barcelona, 2025-09-04) Cos i Espuña, Pep; Doblas-Reyes, Francisco Javier; Marcos Matamoros, Raúl; Universitat de Barcelona. Facultat de Física
[eng] Climate change has many effects, and while global warming caused by human activity is widely recognised, regional-scale changes are particularly important yet uncertain. To effectively communicate, plan protective measures, and adapt to these changes, it is crucial to improve the regional climate information and expand the knowledge of regional climate variability. This thesis focuses on the Mediterranean region, not only because it is where we are located, but also due to its high population density, distinct climate, and the expected intensification of climate change beyond the global average. The thesis begins with an analysis of temperature and precipitation projections from the latest phase of the Coupled Model Intercomparison Project (CMIP6). It considers multiple sources of uncertainty, including different climate models and emission scenarios, and compares the results with those from the previous CMIP5 multi-model set. Findings confirm that the Mediterranean is a climate-change hotspot, showing significant warming and drying trends, especially in summer. Despite differences in global warming projections, CMIP6 projects a warmer future than CMIP5, the pat-tern of amplified regional warming remains consistent. Uncertainty increases toward the end of the 21st Century, driven by the differences in emission scenarios, under-scoring the urgent need for mitigation efforts. Without emission reductions, summer mean temperatures (JJA) could rise by up to 8◦CZ by 2100. Precipitation projections vary widely, but drying trends become clearer under higher emission scenarios, highlighting the region’s future vulnerability to drought and water shortages. The study employs weighting methodologies on the multi-model ensembles to constrain their sampling and performance issues. The thesis then examines the near-term climate in the region, focusing on methods to estimate Mediterranean summer temperatures over the next 20 years from CMIP6 projections. At this timescale, internal climate variability plays a dominant role in uncertainty. The study evaluates different methods that incorporate internal variability by selecting simulations that match the recent climate state. This comparative assessment of constraining methods for short-term climate projections is novel and provides a framework for testing them against observational data. Results show that effectiveness varies depending on the method used and the area considered within the Mediterranean region. Selection approaches based on sea surface temperatures appear promising for improving the estimates but require further refinement for reliability. The research highlights the importance of evaluating the methods to estimate future climate against past observations. Given the complexity of the constraining methodologies that attempt to capture internal variability in uninitialized simulations, the study proposes a framework to improve the understanding of the results and guide future constraining techniques. The thesis concludes addressing another process that affects the Mediterranean climate and has received little attention: the Saharan warm air intrusions, which are not necessarily linked to dust storms and transport. These intrusions can impact extreme temperatures across large parts of the Euro-Mediterranean region throughout the year. The study identifies a rising trend in these events during summer, winter, and autumn over the historical period. It also identifies large-scale atmospheric circulation patterns associated with their occurrence. Understanding these mechanisms could improve climate estimates and serve as an additional metric for evaluating climate models. Initial results indicate that CMIP6 models perform in a variety of ways when reproducing the historical frequency, seasonality, and trends of the intrusion events. This research explores open questions about the Mediterranean climate, enhances understanding of its variability, and proposes ways to improve the quality of available climate information using climate simulations and observational products.
Toroidal nematics: Point defects, disclination lines and walls
(Universitat de Barcelona, 2025-05-29) Rojo-González, Javier; Fernández-Nieves, Alberto; Universitat de Barcelona. Facultat de Física
[eng] Confining partially order fluids can lead to frustration if the locally preferred state can not be achieved everywhere simultaneously. When this happens, the configuration of the material can not be given by the local minimum, and a plethora of compatible frustrated configurations become possible. In this case, the ground state configuration results from the minimization of the total energy. This is specially exemplified by nematic liquid crystals where the anisotropic building blocks reduce their free energy by aligning parallel to each other. Perfect parallel alignment can become impossible everywhere when the nematic is confined within different shapes while fixing the orientation at the boundaries. How this frustration is resolved and which configuration the liquid crystal acquires depends on the geometry and topology of the boundary and might require the presence of regions with no defined alignment, called defects. In this thesis we mainly focus on experimentally studying the configurations acquired by a nematic liquid crystal confined to a stable toroidal droplet with different boundary conditions. We first study the case with tangential alignment at the surface. In this case, we find stable defect populated configurations when the torus is heated to the isotropic phase and cooled back to the nematic phase. They consist mainly on defect pairs whose number and location is random and that turn out to be metastable with an energetic barrier that is high enough for them to not spontaneously annihilate. We also report the appearance of a configuration not fully elucidated that we call ”X” pattern and that can have a solitonic character. Note that all these features are observed even when there is no topological requirement for the torus to have defects. We then study the effect of applying magnetic fields on planar nematic tori with and without defect pairs and find that stable inversion walls form. In the defect free case, two splay-bend walls form, while when a defect pair is present, either one of the walls does not appear or an extra wall is formed depending on the initial location of the defect pair. Interestingly, we find that the mobility of the defects allows for the defect pair wall to reorganize into different types of walls, becoming a wall of the twist type to reduce energy. We also show how, by heating and cooling our nematic droplets in an external magnetic field, the number of defect pairs generated can be controlled. We finish our work on nematic tori by studying the case with perpendicular anchoring at the surface. We recover previous experimental results and find that the anchoring strength plays a significant role in the appearance of a configuration characterized by four dark brushes forming a spiral. Moreover, for certain conditions, stable configurations filled with disclination lines are found and experimentally studied using epifluorescence microscopy. Aside from our experimental work, from a theoretical perspective, we use the decomposition of a director field distortion into four normal modes to derive the compatibility conditions for a 2D and 3D nematic. For that, we use a simple approach based on imposing that the director field variation has to be the same when going through two different paths up to second order. This approach gives an intuitive interpretation of the different equations and provides a simple way to relate them to their counterpart in curved space through the concept of holonomy. Finally, we also discuss some side projects collaterally related to the main thesis work. We first discuss our studies on water evaporation of chitosan hydrogels. We describe the constructed setup to generate the gels and our experiments that track the mass evolution of the gels in time. Our results suggest that the evaporation rate for our hydrogels is slower than pure water due to an increase in the energy barrier for the water molecules to escape the gel, thus bringing about a kinetic character to the process. We also show the design, construction and first tests on an upscaled prototype for the production of these chitosan hydrogels. We end by briefly describing our work with thermorresponsive hydrogels that present a constant volume phase transition leading to phase coexistence between a swollen and shrunken phases when the temperature is increased rapidly. We mainly show how cylindrical gels suddenly bend due to a spontaneous polarization of the region that shrinks, which is in accordance with the current theory for this process.

Examinar

Enviaments recents