The concept of intermediate band solar cells (IBSC) promise a drastic increase in single junction efficiency (theoretically up to over 60% under concentration), exceeding theoretical predictions made by the Shockley-Queisser limit. In this concept, an intermediate band (IB) is formed within the forbidden gap of the host material, acting as an additional energy level where photons with energy less than the host bandgap can be absorbed, significantly reducing the transmission loss of a single junction solar cell. Unfortunately, thermal escape from the quantum dots limit the two-step photon absorption (TSPA) photocurrent collection to low temperatures. In this talk, we report on growth, fabrication, and characterization of type-II InP QD/InGaP IBSC via MOCVD and show high temperature TSPA photocurrent collection compared to conventional InAs and GaSb QD based IBSCs. Time resolved photoluminescence (TRPL), atomic force microscopy (AFM), and excitation-power dependent PL measurements reveal homogeneous formation of type-II InP QDs in an InGaP host matrix with long carrier lifetimes of up to 250 ns. TSPA response using two light sources (monochromatic light and an IR laser) show photocurrent collection at temperatures up to 250 K for the InP QD cells compared to less than 120 K for InAs and GaSb QD cells. These results clearly establish InP QD/InGaP as a viable material system for IBSC applications.
State of the art InGaP2/GaAs/In0.28Ga0.72As inverted metamorphic (IMM) solar cells have achieved impressive results, however, the thick metamorphic buffer needed between the lattice matched GaAs and lattice mismatched InGaAs requires significant effort and time to grow and retains a fairly high defect density. One approach to this problem is to replace the bottom InGaAs junction with an Sb-based material such as 0.73 eV GaSb or ~1.0 eV Al0.2Ga0.8Sb. By using interfacial misfit (IMF) arrays, the high degree of strain (7.8%) between GaAs and GaSb can be relaxed solely by laterally propagating 90° misfit dislocations that are confined to the GaAs-GaSb interface layer.
We have used molecular beam epitaxy to grow GaSb single junction solar cells homoepitaxially on GaSb and heteroepitaxially on GaAs using IMF. Under 15-sun AM1.5 illumination, the control cell achieved 5% efficiency with a WOC of 366 mV, while the IMF cell was able to reach 2.1% with WOC of 546 mV. Shunting and high non-radiative dark current were main cause of FF and efficiency loss in the IMF devices. Threading dislocations or point defects were the expected source behind the losses, leading to minority carrier lifetimes less than 1ns. Deep level transient spectroscopy (DLTS) was used to search for defects electrically and two traps were found in IMF material that were not detected in the homoepitaxial GaSb device. One of these traps had a trap density of 7 × 1015 cm-3, about one order of magnitude higher than the control cell defect at 4 × 1016 cm-3.
The operation of solar cells incorporating multiple repeat units of InAs quantum dot structures, as well as those with
and without δ-doping of 4 and 8 electrons per quantum dot, were studied. Room temperature measurements of these
samples revealed high quality devices, but insignificant differences between δ-doped samples and undoped samples.
An IR-pumped quantum efficiency measurement was performed at 6 K to probe the extraction of quantum confined
carriers through a two-photon process while shutting off phonon-assisted extraction. No two-photon signal rose
above the noise, but additional sub-bandgap illuminated IV curves revealed current generation in the quantum dot
devices, suggesting the dominant carrier removal mechanism is through tunneling. Finally, dark-diode data was
taken and fit to determine ideality factor as a function of temperature. Control devices had an overall larger ideality,
while QD devices exhibited variations as a function of temperature, which were attributed to kinetic barriers in the
first QD layers, as well as possible Auger recombination at very low temperature.