Doctoral School

Reading date: 16/03/2026

• ALONSO, MATÍAS: Hydro-mechanical modelling of a sealing concept for a deep geological radioactive waste repository
  
  Author: ALONSO, MATÍAS
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN GEOTECHNICAL ENGINEERING
  Department: Department of Civil and Environmental Engineering (DECA)
  Mode: Normal
  Deposit date: 15/12/2025
  Reading date: 16/03/2026
  Reading time: 11:00
  Reading place: ETSECCPB.UPC, Campus NordBuilding C1. Classroom: 002C/Jordi Girona, 1-308034 Barcelona
  Thesis director: VAUNAT, JEAN | OLIVELLA PASTALLE, SEBASTIAN
  Thesis abstract: Deep Geological Disposal (DGD) has emerged as the most viable solution for the final disposal of radioactive waste, offering the potential for the permanent containment and isolation of waste from the biosphere over extended timescales. Several countries have made significant progress in developing Deep Geological Repository (DGR) concepts for the permanent disposal of such waste. The long-term safety of these facilities relies primarily on the host rock—the natural barrier that plays the central role—supplemented by engineered components collectively referred to as the engineered barrier system (EBS). The EBS includes containers, backfills, buffers, and other structures designed to ensure favourable conditions for the long-term isolation of radioactive waste. The design, performance, and safety assessment of a DGR—and particularly of its EBS components—are therefore essential for the sustainable development of nuclear energy, making their study a key research area within geotechnical engineering.In this context, the main objective of this research is to contribute to the understanding and assessment of the long-term performance of a large-diameter sealing concept developed within the framework of the Cigéo project, led by the French National Radioactive Waste Management Agency (Andra). To achieve this objective, a multi-scale and multi-step numerical modelling strategy has been adopted. The approach combines detailed material characterisation with advanced constitutive modelling of the expansive core, backfill materials, and host rock, accounting for features such as inherent anisotropy and double structure. The modelling framework incorporates coupled hydro-mechanical processes, enabling the analysis of key phenomena such as the natural hydration of the sealing core, the development of swelling pressure, the resaturation and recompression of the excavation-damaged zone (EDZ), the global equilibrium of the sealing system, and the potential deconfinement of the sealing core and its associated loss of swelling capacity. The simulations address the complexity of the problem by integrating large-scale three-dimensional geometries, advanced constitutive formulations, and critical geometric details at the decimetre scale. These challenging simulations provide valuable insights into the performance and long-term integrity of the sealing structures, establishing a robust framework and a powerful tool to enhance the understanding of the behaviour of these EBS, contributing to the optimisation of repository design and safety.
  

• GHAFOURI JESHVAGHANI, NAVIDEH: Reinforcement Learning-Based Energy Efficient Network Orchestration, Slicing, and Resource Management for 6G Intelligent Networks
  
  Author: GHAFOURI JESHVAGHANI, NAVIDEH
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
  Department: Department of Signal Theory and Communications (TSC)
  Mode: Normal
  Deposit date: 16/02/2026
  Reading date: 16/03/2026
  Reading time: 09:30
  Reading place: ETSETB B3 Teleensenyament
  Thesis director: VARDAKAS, JOHN | VERIKOUKIS, CHRISTOS
  Thesis abstract: This dissertation responds to the emerging challenges of AI-driven 6G network management, aiming to devise smart, adaptive strategies that optimize network performance while reducing complexity and energy consumption. Through the lens of Reinforcement Learning, the study lays the foundation for self-organizing, cost-efficient, and scalable wireless communication systems that can meet the diverse demands of next-generation applications.
  

• RAMIREZ RODRIGUEZ, MARIA SARAY: Fisheries Sustainability: Environmental, Economic, and Social Vulnerabilities, Addressing the Problem Through Food Waste
  
  Author: RAMIREZ RODRIGUEZ, MARIA SARAY
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN SUSTAINABILITY
  Department: University Research Institute for Sustainability Science and Technology (IS.UPC)
  Mode: Normal
  Deposit date: 11/02/2026
  Reading date: 16/03/2026
  Reading time: 10:30
  Reading place: Sala de Graus del EEABB (Campus del Baix Llobregat)
  Thesis director: REIG PUIG, MARIA LOURDES | FERNÁNDEZ GUERRERO, DAVID
  Thesis abstract: Food security and a resilient food system are aims to ensure the future of humanity in a scenario of several threats, such as climate change, the decreasing trend in available resources, the emergence of virulent diseases, global conflicts affecting supply chains, etc. Providing solutions to these issues is so vital that within the Sustainable Development Goals (SDG), sustainable production and consumption is a stated goal (nº 12), and the reduction of Food Losses and Waste (FLW) has become one of the tools for achieving this goal (nº12.3). . This study aims to contribute in FLW reduction for the fish sector, decreasing its negative impacts and increasing the sustainability of the fish sector activity. Moreover, the goal of the thesis is also to reduce fish sector vulnerability, by identifying weak points and possible areas of improvements with a focus on economic and social sustainability, but without forgetting an environmental approach.The methodology combines qualitative and quantitative methods. Public databases were used but also own data was generated by including fish sector stakeholders in the study’s development by surveys, interviews and participatory dynamics. Spain was chosen for the geographical scope due to several factors, such as its high volume of catches, consumption and international trade in fishery and aquaculture products, both in the European and global contextThe thesis is structured in the following way. Chapter 1 was developed as a diagnosis of FLW in the fisheries sector. FLW were quantified and causes, and prevention measures were detected through the use of a standardized methodology. Furthermore, two species were selected to represent Spanish fish sector based on consumption and catches. Anchovy (Engraulis encrasicolus) was chosen to represent oily fish and Hake (Merluccius merluccius) for representing lean fish. As a result, we provided FLW estimations using standardized methodologies, which allows for comparison with future research. Furthermore, our contribution enables the formulation of mitigation and prevention actions. After the diagnosis stage, interventions were designed and tested, trying to approach the problems identified in the diagnosis stage.Chapter 2 tries to improve the resilience of the fisheries sector in Catalonia by developing a market intervention for the expansion of target species. Priority was given to local species, favored by climate change, and which are currently being wasted both nutritionally and economically, due to their low commercial interest. The selected species was Sardinella aurita, which was also validated by local fishermen as a good candidate for the market intervention. The market intervention contributes to increase the economic value of catches by offering alternative products made from low-value species. Moreover, it approached consumers toward local species by adding value to proximity and offering more convenient consumption methods for unskilled consumers. On the other hand, in Chapter 3 another mitigation action was proposed in order to reduce FLW and the economic devaluation of products due to consumer concerns about food safety and aesthetic reasons. As was detected during the diagnostic stage, Anisakis negatively affects the fish sector. It is necessary to inform consumers correctly by developing communication tools that focus on risk avoidance and the social and health benefits of consuming fish. This chapter proposed and tested communications to decrease the perception of food risk towards Anisakis and increase fish consumption.Finally, recommendations have been formulated for each stage of the fish FSC to reduce FLW and increase economic profit, as well as to propose strategies to promote sector’s resilience. These recommendations include strong collaboration and better communication between stages and withpolicymakers, as well as promoting fish culture to increase consumption and reduce food safety concerns.
  

Reading date: 17/03/2026

• BLANCO CASARES, ANTONIO: A Numerical Framework for Solving Complex Flow Regimes with Continuous Galerkin
  
  Author: BLANCO CASARES, ANTONIO
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN COMPUTATIONAL AND APPLIED PHYSICS
  Department: Department of Physics (FIS)
  Mode: Normal
  Deposit date: 18/02/2026
  Reading date: pending
  Reading time: pending
  Reading place: pending
  Thesis director: LEHMKUHL BARBA, ORIOL | MIRA MARTÍNEZ, DANIEL
  Thesis abstract: This thesis investigates a stable and high-order numerical formulation for solving a variety of flow problems using the continuous Galerkin method, including reacting, compressible, and incompressible flows. Reliable and accurate numerical tools for such problems are crucial in many real-world applications, ranging from aerodynamics and industrial processes to medical studies, where high-fidelity and computational efficiency are simultaneously required. Achieving stable solutions in advection-dominated problems, particularly on nonuniform meshes, remains a fundamental challenge.To address this, the thesis develops a stabilization strategy well-suited for high-order elements, such as the spectral ones, designed to effectively eliminate numerical instabilities while introducing minimal numerical dissipation to achieve a high accuracy solution. Based on recent literature, we have focused on stabilization methods based on gradient projection, which allows to detect the numerical fluctuations to effectively stabilize the system. The numerical stencil features a combination of a high- and low-order stabilization ruled by a smoothness sensor. High-order stabilization is applied in smooth regions to maintain physical profiles, while low-order stabilization is activated only near strong gradients to suppress non-physical oscillations. This methodology is formulated for a generic conservation law and tested across linear advection, scalar transport in reacting flows, compressible, and incompressible Navier–Stokes problems.The numerical results show that, for turbulent premixed combustion under the low-Mach assumption, the approach handles tabulated chemistry with the flamelet model, delivering a proper representation of the flame-front. For compressible flows, a Mach-number-based smoothness sensor has been developed, effectively handling complex flow features such as shock waves. For incompressible flows, the algorithm employs a fractional step method with well defined boundary conditions, maintaining the same stabilization principles inherent to this method. In all cases, the stabilization introduces diminishing numerical dissipation with increasing polynomial order, achieving optimal convergence under both mesh refinement and polynomial order increase.These developments contribute to practical, large-scale numerical simulations that balance computational efficiency with high-fidelity results, enabling accurate prediction of real-world physical phenomena. The proposed framework provides a robust and scalable tool for computational fluid dynamics applications, with potential impact across scientific research, and industrial modeling.
  

• CAGETTI, MARTA: Integrated Charge Sensing and Electromechanics in Suspended Carbon Nanotube Quantum Devices
  
  Author: CAGETTI, MARTA
  Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
  Programme: DOCTORAL DEGREE IN PHOTONICS
  Department: Institute of Photonic Sciences (ICFO)
  Mode: Normal
  Deposit date: 11/02/2026
  Reading date: 17/03/2026
  Reading time: 15:00
  Reading place: Elements Room & Teams: Meeting ID: 382 823 036 636 54 Passcode: rd7ja2kU
  Thesis director: BACHTOLD, ADRIAN | FORSTNER, STEFAN
  Thesis abstract: In this thesis we present an ultrasensitive, fast and widely tunable charge detection architecture, suitable for the readout of both electronic states and mechanical motion. The platform is implemented in suspended carbon nanotube (CNT) devices. Our approach employs a radiofrequency (RF) readout scheme operating without impedance matching, thereby avoiding one of the main practical limitations of conventional reflectometry. The readout achieves charge sensitivities exceeding state of the art RF detection techniques, while relying on a comparatively simple measurement setup. The device is based on an integrated single nanotube platform, in which a system of gate defined quantum dots and a proximal quantum dot based charge sensor are hosted in the same suspended CNT and separated by a short metallic drain electrode. The drain is connected to an RLC resonator with a resonance frequency fRLC approximately 1.25 MHz and a bandwidth of 50 kHz, enabling RF readout of the charge sensor current at the circuit resonance. This geometry provides strong capacitive coupling while maintaining independent electrostatic control of the sensor operating point and of the target quantum dots. Using this platform, we achieve self charge sensitivities of order 10^-7 e/sqrt(Hz) and an exceptionally low single shot infidelity, 1 - F approximately 10^-15, for an integration time tau approximately 3.5 microseconds. Beyond the readout of electronic charge transitions in the target quantum dots, the same charge sensor provides highly sensitive access to the mechanical degrees of freedom of the suspended nanotube in the system region. Mechanical displacement is transduced into variations of the charge sensor quantum dot conductance, enabling measurements ranging from driven nonlinear dynamics to thermomechanical motion in the few phonon regime. Crucially, our platform allows operation in a regime where electromechanical backaction, which is typical of suspended carbon nanotubes hosting quantum dots, is strongly suppressed. This addresses one of the central challenges of CNT based nanomechanics: in single dot electromechanical architectures, achieving strong or ultrastrong coupling generally requires operation near charge degeneracy, where coupling to electronic reservoirs and stochastic tunneling lead to excessive dissipation, frequency noise and a pronounced reduction of the mechanical quality factor. Indeed, in previous experiments in the ultrastrong coupling regime, measurement backaction broadened the mechanical response to the point of obscuring access to the intrinsic mechanical properties. In contrast, in our devices we maintain high readout sensitivity without any observable degradation of the mechanical quality factor Q, enabling quantitative spectroscopy of the resonator while preserving its intrinsic mechanical properties. This capability to perform quantitative spectroscopy of a nanomechanical resonator coupled to a two level system in the few phonon regime constitutes a key requirement for advancing towards experiments in the quantum regime, where preserving intrinsic mechanical coherence is essential. The high degree of tunability of our platform enables precise control of charge occupation, tunnel couplings and electrostatic potentials, allowing systematic studies of electromechanical coupling from the single electron regime in a simple quantum dot to the double quantum dot configuration. We demonstrate ultrastrong electromechanical coupling, opening the door to future work on nonlinear nanomechanics, mechanical qubits, quantum delocalization and carbon nanotube based quantum simulation.