Numerical and semi-analytical models are presented for photon-enhanced-thermionic-emission (PETE) devices. The models take diffusion of electrons, inhomogeneous photogeneration, and bulk and surface recombination into account. The efficiencies of PETE devices with silicon cathodes are calculated. Our model predicts significantly different electron affinity and temperature dependence for the device than the earlier model based on a rate-equation description of the cathode. We show that surface recombination can reduce the efficiency below 10% at the cathode temperature of 800 K and the concentration of 1000 suns, but operating the device at high injection levels can increase the efficiency to 15%.

Infrared (IR) radiation detectors are used in numerous applications from thermal imaging to spectroscopic gas sensing. Obtaining high speed and sensitivity, low-power operation, and cost-effectiveness with a single technology remains to be a challenge in the field of IR sensors. By combining nano-thermoelectric transduction and nanomembrane photonic absorbers, we demonstrate uncooled IR bolometer technology that is material-compatible with large-scale CMOS fabrication and provides fast and high sensitivity response to long-wavelength IR (LWIR) around 10 µm. The fast operation speed stems from the low heat capacity metal layer grid absorber connecting the sub-100 nm-thick n- and p-type Si nano-thermoelectric support beams, which convert the radiation induced temperature rise into voltage. The nano-thermoelectric transducer-support approach benefits from enhanced phonon surface scattering in the beams, leading to reduction in thermal conductivity, which enhances the sensitivity. We demonstrate different size nano-thermoelectric bolometric photodetector pixels with LWIR responsitivities, specific detectivities, and time constants in the ranges 179 V/W–2930 V/W, 1.5 × 10⁷ cm Hz^1/2/W–3.1 × 10⁸ cm Hz^1/2/W, and 66 µs–3600 µs, respectively. We benchmark the technology against different LWIR detector solutions and show how nano-thermoelectric detector technology can reach the fundamental sensitivity limits posed by phonon and photon thermal fluctuation noise.

Vibration energy harvesters scavenge energy from mechanical vibrations to energise low power electronic devices. In this work, we report on vibration energy harvesting scheme based on the charging phenomenon occurring naturally between two bodies with different work functions. Such work function energy harvester (WFEH) is similar to electrostatic energy harvester with the fundamental distinction that neither external power supplies nor electrets are needed. A theoretical model and description of different operation modes of WFEHs are presented. The WFEH concept is tested with macroscopic experiments, which agree well with the model. The feasibility of miniaturizing WFEHs is shown by simulating a realistic MEMS device. The WFEH can be operated as a charge pump that pushes charge and energy into an energy storage element. We show that such an operation mode is highly desirable for applications and that it can be realised with either a charge shuttle or with switches. The WFEH is shown to give equal or better output power in comparison to traditional electrostatic harvesters. Our findings indicate that WFEH has great potential in energy harvesting applications.

The state-of-the-art infrared (IR) detection uses quantum photodetectors and bolometers. Quantum IR photodetectors are expensive and require cooling, and exotic and toxic materials. Whereas, bolometers are cost-efficient and uncooled, but they are much slower and less sensitive. Recently we have demonstrated that ultra-thin, highly-doped silicon membranes can be used to build fast and highly-sensitive thermoelectric bolometers. We present the fabrication of these devices, electro-thermal characterization results, and estimate the full potential of this technology.

We investigate the predictable quantum efficient detector (PQED) in the visible and near-infrared wavelength range. The PQED consists of two n-type induced junction photodiodes with Al₂O₃ entrance window. Measurements are performed at the wavelengths of 488 nm and 785 nm with incident power levels ranging from 100 µW to 1000 µW. A new way of presenting the normalized photocurrents on a logarithmic scale as a function of bias voltage reveals two distinct negative slope regions and allows direct comparison of charge carrier losses at different wavelengths. The comparison indicates mechanisms that can be understood on the basis of different penetration depths at different wavelengths (0.77 μm at 488 nm and 10.2 μm at 785 nm). The difference in the penetration depths leads also to larger difference in the charge-carrier losses at low bias voltages than at high voltages due to the voltage dependence of the depletion region.

The state-of-the-art infrared (IR) photodetectors are either thermal detectors (bolometers) or quantum detectors (photovoltaic and photoconductive detectors). Compared to quantum IR photodetectors, IR bolometers are slower and less sensitive but in turn, they offer lower cost without need for cooling and exotic materials (e.g. HgCdTe). Phonon/photon engineered materials offer interesting routes for enhancing room-temperature IR bolometers. We have recently demonstrated experimentally a nano-thermoelectric bolometer for long-wave IR detection. The technology utilizes efficient thermoelectric transducers based on silicon nanomembranes, which have an enhanced thermoelectric figure of merit arising from the low thermal conductivity stemming from the nano-scale thickness. For the absorption of the IR radiation the nano-thermoelectric bolometer utilizes a nanomembrane based quarter-wave resistive absorber, which is also known as the Salisbury screen. The use of nanomembranes in both the thermoelectric transducer and the absorber results in a very small thermal mass, and thereby high speed for the detector. In this article, we present an analytical model for quarter-wave resistive absorbers (i.e. Salisbury screens). It can be applied both in radio frequency (RF) and optical applications. The results of the analytical model are compared with the ones obtained with the transfer-matrix method using the optical material data available in the literature. We present also a device model of the nano-thermoelectric IR detector and estimate the full performance of this technology.