Studies of stochastic thermodynamics with optical tweezers

• Ano de defesa: 2022
• Autor(a) principal: Kamizaki, Lucas Prado
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Conversão informação-energia; Finite-time thermodynamics; Information-to-energy conversion; Optimal protocols; Protocolos ótimos; Termodinâmica em tempo finito
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/76/76134/tde-13072022-121617/

In the last thirty years, experimental and theoretical advancements allowed the investigation of the thermodynamics of small systems far from equilibrium. In 1998, Ken Sekimoto showed that work and heat can be associated with individual trajectories of a Brownian particle. In this context, work becomes a stochastic quantity with a probability distribution associated, respecting important relations as the Jarzynski equality. A paradigmatic study case in stochastic thermodynamics is the fluid-immersed particle trapped in a harmonic potential, a routinely achieved situation using optical tweezers. Using the light-matter interaction, optical tweezers can trap and control colloidal particles. Thus, optical tweezers are powerful and versatile tools when analyzing the thermodynamics of small systems. In this dissertation, we have simulated the dynamics of a colloidal particle trapped in an optical tweezer in different stochastic thermodynamic contexts. By simulating the experimental system, we can verify the feasibility and the adequate parameters to study stochastic thermodynamics in practice. The two main topics studied are optimization of protocols and information-to-energy conversion. Because the work probability density function depends on the protocol, different protocols have different average work values. Among all protocols with a certain intensity and duration, the one that stands out is the optimal protocol, i.e., the process that has the minimum average work. Generally, it is hard to find the optimal protocol analytically, and often other methods are necessary. The first main result is the numerical determination of the performance of the protocols found by approximate methods (for slowly varying processes and weak processes) for different protocol times and intensities. In addition to controlling the system through the protocol, information about its state allows Maxwells demon-like experiments. The second main result is that we propose a new feedback experiment, simplifying the ideas presented in previous works. By doing so, we were able to calculate the dependency of the information-to-energy conversion and the delay time analytically. The simplifications made allow the study of feedback experiments using our actual experimental capabilities.