Fig. 6: Energy-harvesting performances for non-thermal (NT) and quasi-thermal (QT) states.
From: Efficient heat-energy conversion from a non-thermal Tomonaga-Luttinger liquid
a Electromotive force Vemf. b Maximum idealized efficiency \({\bar{\eta }}_{{{\rm{Z}}}}\) in the zero power limit. c Maximum normalized energy-recovery rate R =  PM/JTΓ, where electric power PM is generated from total waste heat JT by a quantum-dot (QD) heat engine with overall tunnel rate Γ. d Idealized efficiency \({\bar{\eta }}_{{{\rm{M}}}}\) at the condition of R. Data were taken at transmission coefficient g = 0.03–0.05 (red circles) for NT states and g = 0.3–0.5 (blue squares) for QT states obtained with setup I (open symbols) and II (filled symbols). Error bars represent typical uncertainty in the determination of drain current ID, effective bias Veff, and QD level ε for each data (see Methods). Simulations with the binary Fermi distribution function fbin for the NT state (red lines) and the thermalized Fermi distribution function fth for the QT state (blue lines) are obtained for single- (solid lines) and multiple-level (dashed lines) QD models. e Schematic energy diagram of the QD heat engine in the presence of excited states with level spacings Δ1 and \({\Delta }_{1}^{{\prime} }\). Transport through the level \(\varepsilon +{\Delta }_{1}^{{\prime} }\) increases the thermoelectric current, but that through ε − Δ1 decreases the current.
