Search

The practical realization of epitaxial quantum dot (QD) nanocrystals led before long to impressive experimental advances in optoelectronic devices, as well as to the emergence of new technological fields. However, the necessary capping process is well-known to hinder a precise control of the QD morphology and therefore of the possible electronic structure required for certain applications. A straightforward approach is shown to tune the structural and optical properties of InAs/GaAs QDs without the need for any capping material different from GaAs or annealing process. The mere adjust of the capping rate allows controlling kinetically the QD dissolution process induced by the surface In-Ga intermixing taking place during overgrowth, determining the final metastable structure. While low capping rates make QDs evolve into more thermodynamically favorable quantum ring structures, increasing capping rates help preserve the QD height and shape, simultaneously improving the luminescence properties. Indeed, a linear relationship between capping rate and QD height is found, resulting in a complete preservation of the original QD geometry for rates above ∼2.0 ML s−1. In addition, the inhibition of In diffusion from the QDs top to the areas in between them yields thinner WLs, what could improve the performance of several QD-based optoelectronic devices.

GaAsSbN dilute nitrides are potential candidates for integration in high-performance multi-junction solar cells due to the bandgap tunability in the 1.0–1.15 eV range and the possibility to match the lattice constant to the GaAs substrates. Recently, the use of GaAsSb/GaAsN superlattices (SLs) has been shown as an effective way to enhance photovoltaic efficiency as compared to the quaternary counterparts. Here we apply a combination of HR-XRD and cross-sectional (S)TEM techniques together with theoretical calculations to analyse the compositional distribution in GaAsSb/GaAsN SLs with different periodicity. The measurements of compositional profiles indicate that Sb is strongly segregated into the GaAsN layers while N remains confined where it is deposited. We demonstrate that the Sb profiles run as a shark-fin waveform that can be precisely described with a one-dimensional model where segregation of the supplied Sb is promoted by a three-layer fluid exchange mechanism. Moreover, the role played by the periodicity in the effectiveness of the Sb incorporation adds a new level of complexity. The modelling of Sb segregation in the GaAsSb/GaAsN SLs system could be used to carry out more precise pseudopotential calculations on the band structure in order to understand and predict their electrical and optical behaviour.

The implementation of GaAs0.8Sb0.2 as CL to obtain type-II strain-coupled InAs MQD structures has been examined and compared to similar structures without Sb or without strain coupling. First, it has been demonstrated that capping with GaAsSb prevents the formation of In-rich agglomerations that hampered the QD formation as it has been observed in the sample without Sb. Instead, it promotes the vertical alignment (VA) of almost all QDs with a high density of QD columns. Second, there is a preferential Sb accumulation over the dots together with an undulation of the growth front, contrary to the observed in the uncoupled structure. In case of a deficient covering of GaAsSb, as occurs for giant QDs, In-rich agglomerations may develop. Each VAQD column consists of a sequence of alternating quantum blocks of pyramid-shaped In(Ga)As separated by GaAsSb blocks that rest over them. These Sb-rich blocks are not homogeneous accumulating around the pyramidal apex like a collar. Between the columns, there is an impoverishment of In and Sb compared to the uncoupled sample. These columns can behave as self-aligned nanowires with type II band alignment between self-assembled InAs and GaAsSb quantum blocks that opens new opportunities for novel devices.

The unavoidable presence of the wetting layer (WL) in Stranski-Krastanov quantum dots (QD) has typically a negative impact on the performance of QD solar cells. In this work, a simple method to engineer the WL of InAs/GaAs QD solar cells is investigated. In particular, we show that covering the QDs at high GaAs capping rates reduces In-Ga intermixing and, therefore, In redistribution from the QDs to the WL. This results not only in larger QDs, but also in thinner WLs, with larger quantum confinement energies and reduced potential barriers for electrons and holes. Carrier trapping by the WLs and subsequent recombination is therefore reduced, resulting in larger photocurrent values of the QD solar cells under short circuit conditions.

We demonstrate type-II GaAsSb/GaAsN superlattices as a suitable candidate to form a lattice-matched 1.01.15 eV subcell that could significantly improve the performance of multi-junction solar cells. The spatial separation of N and Sb allows better lattice-matching control, composition homogeneity, crystal quality and interface abruptness. Moreover, the type-II band alignment provides effective bandgap and radiative lifetime tunability through the period thickness. For the impact of period thickness on transport, quantum-kinetic simulations support the experimental finding of efficient carrier extraction at thicknesses up to 6 nm. All this leads to single-junction solar cells with improved efficiency under monochromatic illumination over the equivalent bulk devices.

The impact of using thin GaAs(Sb)(N) capping layers (CLs) on InAs/GaAs quantum dots (QDs) is investigated for their application in solar cell devices. We demonstrate the ability to combine strain-balancing techniques with band engineering approaches through the application of such CLs. Extended photoresponse is attainable by means of an independent tunability of the electron and hole confinements in the QD. Moreover, the CL acts itself as a quantum well (QW), providing an additional photoresponse, so that the devices work as hybrid QD–QW solar cells. The use of a GaAsSb CL is particularly beneficial, providing devices with efficiencies under AM1.5 conditions 20% higher than standard GaAs-capped QDs. This is mainly due to a significant increase in photocurrent beyond the GaAs bandgap, leading to an enhanced short-circuit current density ( J sc ). The addition of N to the CLs, however, produces a strong reduction in J sc . This is found to be related to carrier collection problems, namely, hindered electron extraction and retrapping in the CLs. Nevertheless, the application of reverse biases induces a release of the trapped carriers assisted by a sequential tunneling mechanism. In the case of GaAsN CLs, this leads to a complete carrier collection and reveals an even higher QD–QW-related photocurrent than when using a GaAsSb CL. The hindered carrier collection is stronger in the case of the quaternary CLs, likely due to the faster recombination rates in the type-I GaAsSbN/GaAs QW structure as compared to the type-II ternary counterparts. Nevertheless, alternative approaches, such as the use of a thinner CL or a short-period superlattice CL, lead to significant improvements, demonstrating a great potential for the quaternary CLs under a proper device design.

Trabajo presentado en el SPIE Octo: Physics, Simulation, and Photonic Engineering of Photovoltaic Devices VII, celebrado en San Francisco (California, Estado Unidos), del 27 de enero al 1 de febrero de 2018, Dilute nitride GaAsSbN is an ideal candidate to form the 1-1.15 eV lattice-matched sub-cell that would significantly enhance the performance of 3- and 4-junction solar cells. However, growth problems inherent to this quaternary alloy lead typically to a poor crystal quality that limits its applicability. Better compositional control and crystal quality have been recently reported by growing the material as a GaAsSb/GaAsN superlattice, because of the spatial separation of Sb and N that avoid miscibility problems. Moreover, these structures provide bandgap tunability trough period thickness. Here we study the performance of lattice-matched 1.15 eV GaAsSb/GaAsN type-II superlattice p-i-n junction solar cells with different period thickness and compare them with the bulk and GaAsSbN/GaAs type-I superlattice counterparts. We demonstrate carrier lifetime tunability through the period thickness in the type-II structures. However, the long carrier lifetimes achievable with periods thicker than 12 nm are incompatible with a high carrier extraction efficiency under short-circuit conditions. Only superlattices with thinner periods and short carrier lifetimes show good solar cell performance. Quantum kinetic calculations based on the non-equilibrium Green’s function (NEGF) formalism predict a change in transport regime from direct tunneling extraction to sequential tunneling with sizable thermionic emission components when passing from 6 nm to 12 nm period length, which for low carrier lifetime results in a decrease of extraction efficiency by more than 30%., J. M. Ulloa acknowledges funding from the Spanish MINECO through project MAT2016-77491-C2-1-R and D. Fuertes Marrón from Comunidad de Madrid (S2013/MAE-2780). U. Aeberhard, D. Fuertes Marrón and J.M. Ulloa acknowledge EU COST Action MP1406 “Multiscale in modelling and validation for solarphotovoltaics (MultiscaleSolar)”.

In this work we investigate the effect of rapid thermal annealing (RTA) on the performance of solar cells consisting of different GaAsSbN-based structures and correlate the device results with modifications of the optical and structural properties of the alloy. In particular, bulk layers grown at different growth rates and type-II GaAsSb/GaAsN superlattices with different period thickness are analyzed. We find evidences of material quality improvement after the annealing process such as a reduction of N-related radiative defects and Sb clusters. These RTA-induced changes lead to a notable enhancement of the open circuit voltage (VOC), which results in values of the bandgap-voltage offset (WOC = EG/q-VOC) comparable to that of a non-optimized reference GaAs solar cell with the same device structure (WOC ∼0.63 eV). The decrease in WOC after annealing shows a correlation with the reduced radiative recombination at low energy N-related sub-bandgap states. These results suggest that radiative recombination in a broad band of deep defect states is a source of VOC degradation in GaAsSbN solar cells.

The effect of the capping process on the morphology of InAs/GaAs quantum dots (QDs) by using different GaAs-based capping layers (CLs), ranging from strain reduction layers to strain compensating layers, has been studied by transmission microscopic techniques. For this, we have measured simultaneously the height and diameter in buried and uncapped QDs covering populations of hundreds of QDs that are statistically reliable. First, the uncapped QD population evolves in all cases from a pyramidal shape into a more homogenous distribution of buried QDs with a spherical-dome shape, despite the different mechanisms implicated in the QD capping. Second, the shape of the buried QDs depends only on the final QD size, where the radius of curvature is function of the base diameter independently of the CL composition and growth conditions. An asymmetric evolution of the QDs' morphology takes place, in which the QD height and base diameter are modified in the amount required to adopt a similar stable shape characterized by a averaged aspect ratio of 0.21. Our results contradict the traditional model of QD material redistribution from the apex to the base and point to a different universal behavior of the overgrowth processes in self-organized InAs QDs.

This work reports on the benefits from using thin GaAsSb capping layers (CLs) on InAs/GaAs quantum dot (QD) solar cells. The application of such CLs allows the tunability of the QD ground state, switching the QD-CL band alignment from type I to type II for high Sb contents and extending the photoresponse beyond 1.5 µm. Two different structures with ~10% and ~20% Sb contents in the CL (type-I and type-II band alignments, respectively) are explored, leading to efficiency improvements over a reference InAs/GaAs QD solar cell of 20% and 10%, respectively. In general, a significant increase in short-circuit current density (Jsc) is observed, partially due to the extended photocurrent spectrum and the additional contribution of the CL itself. Particularly, for a moderate Sb content, an improved carrier collection efficiency is also found to be a main reason for the Jsc increase. Calculations from an 8×8 k·p method suggest the attribution of such an improvement to longer carrier lifetimes in the wetting layer-CL structure due to the transition to a type-II band alignment. Open-circuit voltages (Voc) exceeding that of a reference QD solar cell are demonstrated under light concentration using GaAsSb CLs, which proves that the Voc is not limited by the low bandgap CLs. Moreover, the highest value is obtained for the high Sb content type-II structure, despite the higher accumulation of strain and the lower effective bandgap. Indeed, the faster Voc increase with light power found in the latter case leads to an Voc even larger than the effective bandgap., We acknowledge the Spanish MICINN-MINECO for funding through project MAT2013-47102-C2-2-R. JMLL and IA acknowledge the financial support of the Comunidad Autónoma de Madrid (Grant S2013/MAE-2780), TEC2015-64189-C3-2-R (MINECO/FEDER) and MINECO (Grant AIC-B_2011-0806).

Super-resolution (SR) software-based techniques aim at generating a final image by combining several noisy frames with lower resolution from the same scene. A comparative study on high-resolution high-angle annular dark field images of InAs/GaAs QDs has been carried out in order to evaluate the performance of the SR technique. The obtained SR images present enhanced resolution and higher signal-to-noise (SNR) ratio and sharpness regarding the experimental images. In addition, SR is also applied in the field of strain analysis using digital image processing applications such as geometrical phase analysis and peak pairs analysis. The precision of the strain mappings can be improved when SR methodologies are applied to experimental images.

In this manuscript we carry out a comparative analysis of p-i-n junction solar cells based on 10 stacks of InAs/GaAs quantum dots (QDs) capped with GaAs(Sb)(N) capping layers (CLs). The application of such CLs allows to significantly extend the photoresponse beyond 1.3 μm. Moreover, a strong photocurrent from the CLs is observed so that the devices work as QD-quantum well solar cells. The GaAsSb CL leads to the best results, providing a strong sub-band-gap contribution, which is higher than that in a sample containing standard GaAs-capped QDs, despite giving rise to the highest accumulated strain. The use of a GaAsN CL reduces the photocurrent originating from GaAs, pointing to electron retrapping and hindered extraction and/or the introduction of point defects as possible reasons for this. Nevertheless, the addition of N helps to balance the accumulated strain, necessary to stack a higher number of QD layers. In addition, the possibility to independently tune the hole and electron confinements by the simultaneous presence of Sb and N in the CL is also confirmed for 10 stacked QD layers. This not only allows to further extend the QD ground state and, therefore, the photoresponse, but also offers the possibility to design an optimized structure facilitating carrier extraction from the QDs. Nevertheless, carrier losses seem to be stronger under the simultaneous presence of N and Sb in the CL.

An extensive study on molecular beam epitaxy growth conditions of quaternary GaAsSbN as a capping layer (CL) for InAs/GaAs quantum dots (QD) was carried out. In particular, CL thickness, growth temperature, and growth rate were optimized. Problems related to the simultaneous presence of Sb and N, responsible for a significant degradation of photoluminescence (PL), are thereby solved allowing the achievement of room-temperature (RT) emission. A particularly strong improvement on the PL is obtained when the growth rate of the CL is increased. This is likely due to an improvement in the structural quality of the quaternary alloy that resulted from reduced strain and composition inhomogeneities. Nevertheless, a significant reduction of Sb and N incorporation was found when the growth rate was increased. Indeed, the incorporation of N is intrinsically limited to a maximum value of approximately 1.6% when the growth rate is at 2.0 ML s−1. Therefore, achieving RT emission and extending it somewhat beyond 1.3 μm were possible by means of a compromise among the growth conditions. This opens the possibility of exploiting the versatility on band structure engineering offered by this QD-CL structure in devices working at RT. PACS 81.15.Hi (molecular beam epitaxy); 78.55.Cr (III-V semiconductors); 73.21.La (quantum dots)

The application of a GaAsSb/GaAsN short-period superlattice capping layer (CL) on InAs/GaAs quantum dots (QDs) is shown to be an option for providing improved luminescence properties to this system. Separating both GaAsSb and GaAsN ternaries during the growth in 2 monolayer-thick phases solves the GaAsSbN immiscibility-related problems. Strong fluctuations in the CL composition and strain field as well as in the QD size distribution are significantly reduced, and a more regular CL interface is also obtained. Room-temperature (RT) photoluminescence (PL) is obtained for overall N contents as high as 3%, yielding PL peak wavelengths beyond 1.4 μm in samples with a type-II band alignment. High external quantum efficiency electroluminescence and photocurrent from the QD ground state are also demonstrated at RT in a single QD-layer p-i-n device. Thus, it becomes possible to combine and transfer the complementary benefits of Sb- and N-containing GaAs alloys to InAs QD-based optoelectronics.

Type I and type II band alignment InAs/GaAs quantum dot laser diodes (LD) are demonstrated using a 5-nm-thick GaAsSb capping layer with moderate or high Sb contents. The threshold current density, external differential quantum efficiency, and characteristic temperature substantially improve when Sb is used in the capping layer. Nevertheless, in the type II LD, lasing arises from type I-like excited states with much shorter lasing wavelengths than expected. This is likely related to the observed inhibition of the ground state transition in the spontaneous emission, which would also reduce the radiative current and, therefore, the threshold current.