SearcharxivSearch

arXiv subjects

Denis Rideau

Publications and source records attributed to Denis Rideau.

5 recordsLinked to original sources

Diffusion Schr\"odinger Bridges with enhanced posterior sampling for metasurface inverse design

Metasurface inverse design is challenged by the intricate relationship between structural parameters and electromagnetic responses, as well as the high dimensionality of the optimization space. Local models, while commonly employed, quickly become infeasible for complex and locally coupled structures. Conventional iterative optimization techniques, on the other hand, are computationally intensive, time-consuming, and susceptible to convergence in local minima. This study explores a versatile generative methodology based on enhanced posterior sampling within the Schr\"odinger Bridge framework. By decomposing posterior sampling into amplitude and directional contributions, we effectively integrated different kind of posterior sampling. This approach is further supported by refined training strategies to enhance performance and reduce the complexity of hyperparameter optimization. The proposed framework demonstrates exceptional accuracy and robustness, representing a significant advancement in metasurface design. Notably, it enables high-precision inverse design for large-scale configurations of up to $350 \times 350$ pillar arrays, despite being trained on significantly smaller arrays of $23 \times 23$ pillars.

physics.optics

Enhanced posterior sampling via diffusion models for efficient metasurfaces inverse design

The inverse design of metasurfaces faces inherent challenges due to the nonlinear and highly complex relationship between geometric configurations and their electromagnetic behavior. Traditional optimization approaches often suffer from excessive computational demands and a tendency to converge to suboptimal solutions. This study presents a diffusion-based generative framework that incorporates a dedicated consistency constraint and advanced posterior sampling methods to ensure adherence to desired electromagnetic specifications. Through rigorous validation on small-scale metasurface configurations, the proposed approach demonstrates marked enhancements in both accuracy and reliability of the generated designs. Furthermore, we introduce a scalable methodology that extends inverse design capabilities to large-scale metasurfaces, validated for configurations of up to $98 \times 98$ nanopillars. Notably, this approach enables rapid design generation completed in minute by leveraging models trained on substantially smaller arrays ($23 \times 23$). These innovations establish a robust and efficient framework for high-precision metasurface inverse design.

physics.optics

Quenching statistics in Si and Ge SPADs using particle Monte Carlo simulation

Si- and Ge-based single-photon-avalanche-diodes (SPAD) are investigated by using self-consistent 3D Monte Carlo simulation, in a mixed-mode approach including the presence of a passive quenching circuit. This approach of transport allows us to capture all stochastic features of carrier transport and SPAD operation, not only for the avalanche triggering but also for the quenching process. Beyond the comparison of Si and Ge devices, we show in particular the strong inverse correlation between avalanche and quenching probabilities when tuning the bias voltage, which highlights the importance to find a tradeoff between these two probabilities for the optimization of SPAD operation.

cond-mat.other

Contact resistances in trigate and FinFET devices in a Non-Equilibrium Green's Functions approach

We compute the contact resistances $R_{\rm c}$ in trigate and FinFET devices with widths and heights in the 4 to 24 nm range using a Non-Equilibrium Green's Functions approach. Electron-phonon, surface roughness and Coulomb scattering are taken into account. We show that $R_{\rm c}$ represents a significant part of the total resistance of devices with sub-30 nm gate lengths. The analysis of the quasi-Fermi level profile reveals that the spacers between the heavily doped source/drain and the gate are major contributors to the contact resistance. The conductance is indeed limited by the poor electrostatic control over the carrier density under the spacers. We then disentangle the ballistic and diffusive components of $R_{\rm c}$, and analyze the impact of different design parameters (cross section and doping profile in the contacts) on the electrical performances of the devices. The contact resistance and variability rapidly increase when the cross sectional area of the channel goes below $\simeq 50$ nm$^2$. We also highlight the role of the charges trapped at the interface between silicon and the spacer material.

cond-mat.mes-hall

Quantum calculations of the carrier mobility in thin films: Methodology, Matthiessen's rule and comparison with semi-classical approaches

We discuss the calculation of the carrier mobility in silicon films within the quantum Non-Equilibrium Green's Functions (NEGF) framework. We introduce a new method for the extraction of the carrier mobility that is free from contact resistance contamination, and provides accurate mobilities at a reasonable cost, with minimal needs for ensemble averages. We then introduce a new paradigm for the definition of the partial mobility $μ_{M}$ associated with a given elastic scattering mechanism "M", taking phonons (PH) as a reference ($μ_{M}^{-1}=μ_{PH+M}^{-1}-μ_{PH}^{-1}$). We argue that this definition makes better sense in a quantum transport framework as it is free from long range interference effects that can appear in purely ballistic calculations. As a matter of fact, these mobilities satisfy Matthiessen's rule for three mechanisms [surface roughness (SR), remote Coulomb scattering (RCS) and phonons] much better than the usual, single mechanism calculations. We also discuss the problems raised by the long range spatial correlations in the RCS disorder. Finally, we compare semi-classical Kubo-Greenwood (KG) and quantum NEGF calculations. We show that KG and NEGF are in reasonable agreement for phonon and RCS, yet not for SR. We point to possible deficiencies in the treatment of SR scattering in KG, opening the way for further improvements.

cond-mat.mes-hall