High Performance Quantum Dot Excitonic Solar Cells

Limitations in the stock of fossil fuels, ever increasing energy demand, and climatic changes have resulted in the search of sustainable clean energy sources world wide. Photovoltaics (PV), the science and technology of solar cells that provide renewable and clean electricity, have developed significantly over the past several decades. Solid-state p-n junction devices profiting from the experience of the semiconductor industry have dominated the field of solar cells so far. Presently commercial solar cell modules (area typical 1 x 2 m2) of efficiency ~ 17% and single cells (area ~1 cm2) of efficiency up to 40 % (under high optical concentration) have been realized from crystalline silicon and multijunction devices, respectively. Solar cells that mimicking the natural solar energy conversion process, known as dye-sensitized solar cells (DSC) or Graetzel cells, are getting increased attention recently due to the possibilities of cost-effectiveness Considerable efforts were devoted in this area that brought the efficiency ~11% in single cells of area ~ 1cm2. Efforts are already initiated at the industry levels to commercialize the DSC.

The schematic diagram of a DSC and the various processes in it is shown in Fig.1. A DSC consists of three functional parts (Fig.1A); viz. a solar light harvester, usually a dye, which converts an absorbed photon into an exciton; an electron acceptor (electrode) that splits the exciton into electrons and holes by the energy difference between the LUMO of the light harvester and the conduction band of the electrode; and a red/ox mixture that injects the electron back to the dye. Figure 1B shows a typical interface problem and schematic exciton generation, diffusion splitting, and electrical transport processes.

There are at least nine fundamental processes that can control the energy conversion efficiency in an excitonic solar cell. These are (i) photon absorption which is determined by the wavelength window where the harvester absorbs, intensity of solar radiation at that window, and absorption cross-section of the dye (); (2) radiative recombination determined by the carrier life time and the probability for radiative recombination in the excited state (); (3) exciton diffusion and its diffusion length () which controlled by the exciton diffusion coefficient () and exciton life time; (4) interfacial electron transfer and its rate (); (5) interfacial charge recombination determined by the rate at the interface (); and (6) the exciton relaxation () through which the exciton lose its energy due to relaxation; (7) electron transport through the electrode with drift (), (8) the phonon relaxation () through which an electron lose its energy via thermalization, (9) the red/ox potential of the electrolyte and rate of electron transfer to the dye. Current status of DSC is such that ~11% efficiency is achieved in laboratories and efforts are initiated at the industrial levels to commercialize the product. Progress in the conversion efficiencies of the excitonic solar cells are still possible because of the large difference between its thermodynamic limit and current status as well as potential of new generation materials, known as quantum dots, to generate many excitons from a single photon of sufficient intensity. This is in principle feasible for “high energy” photons, the application of such a concept to photons with energies less than twice the band- gap energy will result in the reduction of the device voltage and thus not in an increase of the power output of the photovoltaic cells. Therefore, a compromise is to be established between the best power output of the photovoltaic cells and multi-exciton generation.

The thermodynamic limit of 31% could be shifted ~42% if dyes are replaced by inorganic quantum dots due to a possibility of generating multi-exciton from a single photon of sufficient energy. A literature survey reveals that the best reported cell employing quantum dots gave an energy conversion efficiency ~6%,7 which is much lower than the thermodynamic limit. We propose to improve the efficiency up to 15% over a period of five years by paying attention to two issues: (I) to achieve multiple exciton generation with proper quantum dots and spectrum and (II) enhancing the charge collection through new electrodes. The charge collection could be increased by (i) functionalization of the surface of quantum dots using appropriate ligands for high (= 80%) incident-photon-to-electrical conversion (IPCE) efficiencies and (ii) using appropriate choice of new materials and architectures as electrodes. We believe that the five year period for achieving 15% efficiency using quantum dots is realistic because new approaches are proposed rather than simply replacing a dye by a quantum dot. We propose to study systematically the charge separation through appropriate ligands. Besides, multi-exciton generation (MEG) and subsequent carrier multiplication possibilities in quantum dots were not adequately addressed. Efforts to attain high efficiency excitonic solar cells using quantum dots with MEG possibility went in vein because requirement of proper quantum dot is not identified, though considerable progress has been made in understanding the fundamental properties of quantum dots. Our proposal is based on realization of these drawbacks; therefore, we strongly believe that the target could be realized.

I. Efficient light harvesting through multi-excitons:

Inorganic semiconducting quantum dots are proposed for light harvesters excitonic solar cells. We identify that the quantum dots have three significant advantages over conventional organic dyes: (i) being a crystal rather than a molecule, high temporal stability could be expected; (ii) the absorption spectrum of quantum dots is continuous from the onset of the first excitonic peak to higher energy side which improves the absorption wavelength range; (iii) carrier multiplication from a single absorbed photon of sufficient energy, known as multi-exciton generation (MEG), is possible. Quantum dots of CdSe or PbSe have been the main focus in excitonic solar cells so far because of the easiness in their synthesis. The CdSe and PbSe quantum dots have bandgaps 1.8–3.5 eV and 0.3 – 1.2 eV, respectively. The CdSe quantum dots suffer from lower possibility of MEG (maximally ~2 excitons from a single photon of higher energy) whereas for PbSe, the absorption cross-section of these quantum dots of bandgap energy 1.0 eV is very weak because of its smaller size (~1.8 nm).10 If quantum dots (size ~5-15 nm) of first excitonic transition energy (equivalent to its bandgap energy) ~1-2.5 eV with enhanced absorption cross-section is realized, a large spectrum of solar radiation could be captured and converted into electrical energy with high conversion efficiencies. A typical schematic of this concept is shown Fig. 2. Therefore, the present proposal is to optimize quantum dots of II-VI, III-V and IV semiconductors for photovoltaic applications. Colloidal synthesis schemes using organo-metallic reagents and proper coordination solvents will be used for synthesis of quantum dots. A comprehensive investigation on nucleation and growth of these quantum dots, their structural, surface and chemical properties, dynamic distribution of concentration and fluorescence within an ensemble, carrier life time and decay dynamics, absorption cross-section, and multi-exciton generation and its mechanism will be undertaken. Experimental tools for colloidal synthesis and characterization facilities are accessible through the Nanoscience and Nanotechnology Initiative of National University of Singapore (NUSNNI). In addition to the experimental methods, first-principle quantum mechanical calculations will also be employed to deeply understand the structure and properties of quantum dots. Advanced softwares and computers could also be accessed through NUSNNI.

II. Enhancing the charge collection:

The excitons produced in the quantum dots diffuse to the interface between the quantum dots and the electrodes where they are separated into electron and hole. Two detrimental aspects are involved in the processes of charge collection, viz. (i) choice of proper ligand that links the quantum dots to the electrodes and (ii) new electrodes and architectures.

(i) Tailored bonding of quantum dots with electrodes for efficient charge transfer: The quantum dots and the electrodes are inorganic solids. Conventional solid state reaction to bond them together lead to a new compound which is different from the parent reactants; therefore, chemical methods are used for linking functional components in advanced materials technology programs. Appropriate ligands should be used to link the quantum dots with the electrodes (Fig.3). These linkers (i) should be of small length for efficient charge separation; (ii) are not supposed to shift the band edges of the quantum dots or photo-electrodes when bonded with either of them; (iii) should not quench the fluorescence by itself. Our preliminary investigation on the use of conventional electrodes for CdSe quantum dots show that TiO2 and ZnO interact strongly with quantum dots and enhance thermalization possibilities. Proper choice of ligands will be identified systematically by starting from small molecules featuring –COOH and/or –SH groups, for example mercapto-acetic acid, and using them to link the quantum dots and the electrodes. The quantum dot conjugated electrodes will be characterized in detail for structural and/or functional stability, energy and/or charge transfer, and IPCE using experimental methods. Preliminary studies made by us to understand whether the decay dynamics of CdSe quantum dots are affected by the chemical environment using conventional photoluminescence spectroscopy indicated that surface ligands affect the fluorescence. However, nature of this dependence of fluorescence on surface ligands, i.e, whether it is thermalization or charge transfer, was not investigated in detail. The quantum dot conjugated electrode system will also be modeled using first-principle quantum mechanical methods to deeply understand their structure and properties.



If one such simple system is found to be a poor choice from the experimental and theoretical investigations, then longer molecules of the same homologues series or alternate small molecules with different functional groups such as succinic acid with –COOH groups on both sides will be examined to meet the target =80% IPCE.

(ii) Guided Electron Transport through New Electrodes: There are numerous issues in the selection of a suitable electrode for photovoltaics. Some of these issues are (i) relative positions of conduction band energy of the electrode and LUMO energy of the photon absorber, (ii) photo-corrosion due to similarities in the bandgaps of the electrode and the photon absorber, (iii) relative positions of the conduction band energy of the electrode and redox potential of the electrode, etc. The conventional TiO2 or ZnO electrodes in the current excitonic solar cells are meant for LUMOs of dyes that falls above their conduction band edges (~ -4.2 eV with respect to vacuum). For the proposed quantum dots of first exciton energy ~1 eV, these electrodes might not be ideal and should be revisited depending on the LUMO energies of the target quantum dots. Electrodes of appropriate conduction band and bandgap energies will be developed to fit into the exact requirement for efficient charge separation and transfer. The charge transport in conventional semiconducting electrodes is through hopping mechanism and is very inefficient. The proposed idea here is to dope the semiconductors electrodes using heavy metals such that quasi metallic electrical conduction could be expected, thereby improving the efficiency. This proposal also aims to develop one-dimensional nanostructures as electrodes for channeled electron transport such that grain boundary scattering could be minimized, thereby improving the efficiency.

Electrospinning of a polymeric solution and subsequent drying, calcination, and sintering offers large potential to produce one-dimensional nanostructures as well as to fabricate nanostructures on an industrial scale for energy applications. We showed recently that by simply spraying electrospun nanorod based electrodes relatively larger area (=20 cm2) compared to the conventional laboratory made (area ~0.2 cm2) DSC could be realized with a conversion efficiency of ~6%.12 Choice of electrolytes for a given set of quantum dots and electrodes will be investigated to achieve the targets, i.e., ~15% overall cell efficiency and = 80% IPCE.