New strategies for colloidal-quantum-dot-based intermediate-band solar cells.

The intermediate-band solar cell (IBSC) concept promises to increase the efficiency limit in a single-junction solar cell through the absorption of below-bandgap-energy photons. Despite their operating principle having been proposed over 20 years ago, IBSCs have not delivered on this promise yet, and the devices fabricated so far, mainly based on embedded epitaxial quantum dots, have instead operated with lower efficiency than conventional solar cells. A new paradigm, based on the exploitation as the intermediate band of the intragap states naturally occurring in the density functional theory description of colloidal (i.e., chemically synthesized) quantum dots, was suggested recently. Here, we revisit this intriguing concept unveiling its shortcomings and propose two alternative schemes: in the first, the localized electron surface trap states, ubiquitously found in commonly synthesized colloidal quantum dots, are used as intermediate bands in strongly coupled films made of small InAs nanocrystals and, in the second scheme, the intermediate band is provided by the conduction-band-minimum-derived miniband in films of larger InAs nanocrystals. Both schemes yield estimated limiting IBSC efficiencies exceeding Shockley-Queisser's limit for a single absorber.

Fully Parallel Stochastic Computing Hardware Implementation of Convolutional Neural Networks for Edge Computing Applications.

Edge artificial intelligence (AI) is receiving a tremendous amount of interest from the machine learning community due to the ever-increasing popularization of the Internet of Things (IoT). Unfortunately, the incorporation of AI characteristics to edge computing devices presents the drawbacks of being power and area hungry for typical deep learning techniques such as convolutional neural networks (CNNs). In this work, we propose a power-and-area efficient architecture based on the exploitation of the correlation phenomenon in stochastic computing (SC) systems. The proposed architecture solves the challenges that a CNN implementation with SC (SC-CNN) may present, such as the high resources used in binary-to-stochastic conversion, the inaccuracy produced by undesired correlation between signals, and the complexity of the stochastic maximum function implementation. To prove that our architecture meets the requirements of edge intelligence realization, we embed a fully parallel CNN in a single field-programmable gate array (FPGA) chip. The results obtained showed a better performance than traditional binary logic and other SC implementations. In addition, we performed a full VLSI synthesis of the proposed design, showing that it presents better overall characteristics than other recently published VLSI architectures.

Optical Absorption in N-Dimensional Colloidal Quantum Dot Arrays: Influence of Stoichiometry and Applications in Intermediate Band Solar Cells.

We present a theoretical atomistic study of the optical properties of non-toxic InX (X = P, As, Sb) colloidal quantum dot arrays for application in photovoltaics. We focus on the electronic structure and optical absorption and on their dependence on array dimensionality and surface stoichiometry motivated by the rapid development of experimental techniques to achieve high periodicity and colloidal quantum dot characteristics. The homogeneous response of colloidal quantum dot arrays to different light polarizations is also investigated. Our results shed light on the optical behaviour of these novel multi-dimensional nanomaterials and identify some of them as ideal building blocks for intermediate band solar cells.

Optical properties of nanocrystal films: blue shifted transitions as signature of strong coupling.

We present a theoretical study at the atomistic level of the optical properties of semiconductor nanocrystal films. We investigate the dependence of the absorption coefficient on size, inter-dot separation, surface stoichiometry and morphology, temperature, position of the Fermi level and light polarization. Our results show that, counter-intuitively, huge blue shifts are expected in some intra-band transitions for strongly coupled arrays, in contrast with the predicted and observed red shift of the band gap absorption in such systems. Furthermore, we find that the energies of such transitions can be tuned within a range of several hundreds of meV, just by engineering the inter-dot separation in the film through the choice of appropriately sized capping ligands. Finally we discuss the application of this effect to nanocrystal-based intermediate-band solar cells.

Efficient, non-stochastic, Monte-Carlo-like-accurate method for the calculation of the temperature-dependent mobility in nanocrystal films.

We present a new non-stochastic framework for the calculation of the temperature dependence of the mobility in nanocrystal films, that enables speed-ups of several orders of magnitude compared to conventional Monte Carlo approaches, while maintaining a similar accuracy. Our model identifies a new contribution to the reduction of the mobility with increasing temperature in these systems (conventionally attributed to interactions with phonons), that alone is sufficient to explain the observed experimental trend up to room temperature. Comparison of our results with the theoretical predictions of the hopping model and the observed temperature dependence of recent field-effect mobility measurements in nanocrystal films, provides the means to discriminate between band-like and hopping transport and a definitive answer to whether the former has been achieved in quantum dot films.