Thermoelectric contributions to electronic transport and THz responsivity of TeraFETs –simulations and experiments

天线耦合场效应晶体管(FET)成功地成为太赫兹(THz)辐射的灵敏探测器。探测机制是基于场效应管沟道中等离子体激元增强的分布电阻混合(Dyakonov-Shur机制)和热电热载流子效应，这些热载流子效应是由于能量和电荷传输的扩散贡献而产生的。对考虑热效应的流体力学Dyakonov-Shur模型进行了扩展，在本项目的第一阶段，我们从Boltzmann方程出发，用矩方法推导了电荷和能量输运的流体动力学模型。基于场效应管的电流/电压特性，我们还建立了石墨烯场效应管的电荷-电压模型(对于其他材料系统，我们已经在前面推导了该模型)。在电路模拟工具Keysight ADS中实现了描述电荷和能量输运的完整的物理模型，其中包括描述电荷和能量输运的四个微分方程。作为模拟的输入参数，只需要从晶体管的直流测量和天线模拟中提取的数据。该模型以接近定量的方式再现了在不同材料系统中测量到的太赫兹响应率和噪声当量功率，范围从0.1太赫兹到至少4太赫兹。可以确定在III/V材料系统和石墨烯中制造的晶体管中热电效应的贡献；对于III/V，在栅极电压的工作点上热电效应是有害的。在项目的第二阶段，我们现在正在申请资金，我们的目标是各种模型的改进。其中之一涉及在直流电和太赫兹频率下在通道边界处的电势分布的细节。交流电势分布有望强烈地影响扩散输运，从而影响热电效应。研究的另一个主题是高辐射强度下的饱和效应。进一步的改进主要是针对石墨烯设备，并涉及与芬兰阿尔托大学的一个团队的密切合作。通过特别设计的器件结构，我们的目标是提取更稳健的电荷载流子能量弛豫速率数据。使用先进的模型，我们将着手设计优化的基于石墨烯的探测器结构，由阿尔托团队实施，并在阿尔托和我们进行测量。该项目的很大一部分还将用于准备出版物(I)关于模拟工具、其基本物理模型假设及其在ADS中的算法实施、(Ii)关于模拟工具在设备性能分析中的应用、以及(Iii)关于将其用于开发改进的探测器的预测性使用。

Antenna-coupled field-effect transistors (FETs) establish themselves successfully as sensitive detectors of terahertz (THz) radiation. The detection mechanism is based on plasmon-enhanced distributed resistive mixing in the FET's channel (Dyakonov-Shur mechanism) in combination with thermoelectric hot-carrier effects arising from diffusive contributions to energy and charge transport. Extending the hydrodynamic Dyakonov-Shur model, which does not take thermal effects into account, we derived in the first phase of this project a hydrodynamic model for charge and energy transport from the Boltzmann equation using the method of moments. Based on the current/voltage characteristics of the FETs, we also developed a charge-voltage model for graphene FETs (for other material systems, we had derived them earlier). The complete physical model with its four differential equations describing charge and energy transport was implemented in the circuit simulation tool Keysight ADS for the material systems Si, GaN, GaAs and Graphene. As input parameters for the simulations, only data extracted from DC measurements of the transistors and antenna simulations are needed. The model was found to reproduce the measured THz responsivity and noise-equivalent power of detectors in various material systems in a near-quantitative manner from 0.1 THz up to at least 4 THz. The contribution of the thermoelectric effect in transistors made in the III/V material systems and in Graphene could be determined; for III/Vs, it is detrimental at the operation points of the gate voltage.In the second phase of the project, for which we apply for funding now, we aim at various model refinements. One of them relates to the details of the potential distribution at the channel’s boundaries both at DC and THz frequencies. The AC potential distribution is expected to strongly influence the diffusive transport and thus the thermoelectric effect. Another subject of the studies will be saturation effects at high radiation intensities. Further refinements are mainly for graphene devices and involve a close cooperation with a team at Aalto University, Finland. With specifically designed device structures, we aim for the extraction of more robust data for the energy relaxation rates of the charge carriers. Using the advanced model, we will then embark on the design of optimized graphene-based detector structures, to be implemented by the Aalto team and to be measured at Aalto and by us. A significant part of the project will also be devoted to the preparation of publications (i) on the simulation tool, its underlying physical model assumptions and its algorithmic implementation in ADS, (ii) on the application of the simulation tool for the analysis of device performance, and (iii) on the predictive use of it for the development of improved detectors.