A general design guideline for strain-balanced quantum-wells toward high-efficiency photovoltaics
Introduction
Thanks to highly-radiative and tunable optical property, multiple-quantum-well (MQW) structures have been extensively applied to various optoelectronic devices, such as lasers (Halliday et al., 1992, Nagarajan et al., 1992, Tessler and Eistenstein, 1993), LEDs (Bedair et al., 1984), optical amplifiers (Tiemeijer et al., 1993), modulators (Wood, 1988), photodetectors (Andrews and Miller, 1991), and solar cells (Adams et al., 2011, Ekins-Daukes et al., 1999, Fujii et al., 2014a, Fujii et al., 2012a, Razykov et al., 2011, Yamaguchi et al., 2008). In the tandem solar cells technology in particular, by dividing incoming solar spectrum equally between subcells with optimum absorption thresholds, the conversion efficiency is anticipated to be able to go beyond 50% (Geisz et al., 2018a, Geisz et al., 2018b). For instance, it has been suggested that a quad-junction solar cell with ideal bandgap combinations of 1.99/1.51/1.15/0.66(Ge) eV can achieve an efficiency of 50.1% by concentrator systems under 1000-suns (Thomas et al., 2014). To attain bandgap as low as ~1.0 eV for the middle cell, dilute-nitride alloy GaInNAs has been considered as a good approach for monolithic growth on germanium substrate (Courel et al., 2012a, Courel et al., 2012b, Friedman et al., 1998, Tukiainen et al., 2016, Yamaguchi et al., 2005). Additionally, a strain-balanced (SB) InGaAs/GaAsP MQW, a promising alternative approach, has been developed with an engineered effective bandgap down to the targeted value 1.15 eV successfully (Toprasertpong et al., 2016). However, when incorporating MQW structures into solar cells or the other aforementioned devices, carrier transport in the perpendicular direction suffers due to the introduction of the potential barriers at the heterointerfaces. Without being carefully designed, the collection probability of the photogenerated carriers tends to drop drastically, leading to limited energy yields.
Optimizing an MQW structure is challenging. There are typically four parameters that can be controlled when designing MQW structures based on compound III-V semiconductors: well thickness tw, barrier thickness tb, well material, and barrier material, as illustrated in Fig. 1a. Conventional design techniques involve strain-balance conditions (Ekins-Daukes et al., 2002), reducing the number of free parameters from four to three. This represents a large parameter space, which is challenging to explore systematically. In previous work, the optimization was done only partially, e.g. in reference (Kim et al., 2018) only the well material and the barrier thickness are modified while the well thickness is fixed; in reference (Fujii et al., 2014a) only the barrier material and barrier thickness are optimized, whereas the well thickness is fixed. Those past optimization schemes are essentially not satisfying for three reasons:
- (1)
Not all parameters are explored comprehensively. The optimum design can be found unlikely.
- (2)
It was trial-and-error. Because there were no comprehensive models to aim at a favorable design beforehand, the QW design needed to be further adjusted at the fabrication stage. Such a method is time-consuming and unprecise.
- (3)
MQWs with different absorption thresholds were compared. However, this is the most significant material requirement when designing a subcell with an optimum bandgap for tandem photovoltaic cells.
On the other hand, understanding the carrier dynamics behind is crucial to further improve optoelectronic devices based on QW. So far, conventional diagnosis of such degraded carrier transport in MQWs on the microscopic level has been relying mainly on the escape probability function derived from tunneling and thermionic-emission lifetime models (Fox et al., 1991, Nelson et al., 1993, Schneider and Klitzing, 1988). It was revealed as a good technique interpreting the transport dynamics involved in carrier collection as a solar cell from the macroscopic experiment results qualitatively (Fujii et al., 2012a, Sayed and Bedair, 2019). Nevertheless, the relative contribution of thermionic emission and carrier tunneling remains unclear. Furthermore, without estimating carrier transport in terms of their equivalent mobilities, it is difficult to quantitatively analyze how poor the perpendicular carrier transport could be. This prohibits one to fully exploit the tunability of transport properties of MQWs from the microscopic point of view.
In this research, to tackle these design and analysis conundrums, we propose a novel design flow including three essential steps:
Step 1: Reduce the parameter space from 4 to 2, by setting two important design requirements for well thickness tw and barrier thickness tb: a strain-balance condition and an effective bandgap equal to a target value Eg,target. Then, with their critical layer thicknesses, assess the difficulty of performing heteroepitaxy given the design structures.
Step 2: Use a recently-developed model allowing to quantitatively resolve effective mobilities for MQW structures as a function of the last 2 free parameters: well material and barrier material. This enables mapping of the effective mobilities with different material combinations.
Step 3: By considering probable minority-carrier-lifetime ranges, deduce the cell efficiency. Based on several additional design constraints according to applications, we are able to pin down the optimum range for the composed materials and thicknesses and therefore derive the optimum MQW design.
This flow will define an optimum region to be explored experimentally, in terms of the 4 free parameters. It solves the three drawbacks highlighted above. By comparing only MQWs with the target absorption threshold, we are able to determine the optimum subcell design for high-efficiency tandem photovoltaic cells.
In the following, we first illustrate our design strategy in the specific case of InGaP/InGaP MQW (depicted in Fig. 1b) as a candidate absorber for the top cell in dual-MQW triple-junction solar cells. Such an architecture unveils enhanced tolerance against the fluctuation of the incoming spectrum owing to photon-coupling between MQW cells (Lee et al., 2012). However, replacing the conventional top In0.49Ga0.51P absorber by MQWs implies worse carrier transport. A detailed quantitative investigation of degradation mechanisms has never been reported and is in urgent demand. With this example, we present our full design procedures and resolve the carrier transport mechanism in MQW compared to the In0.49Ga0.51P bulk with identical absorption thresholds for the first time. Secondly, we apply the design guideline to optimize a 1.23 eV-InGaAs/GaAsP MQW absorber (depicted in Fig. 1c), which is significant for not only high-efficiency single-junction photovoltaics but also current-matched tandem cells (Barnham and Duggan, 1990, Katsuyama et al., 1987). Yet, the MQW absorber was studied within a small parameter space in previous research (Fujii et al., 2014a). With this example, we discuss the possibility of further improving the efficiency of the conventional design in terms of its constituent materials. Additional design constraints, such as background doping that hinders carrier collection and requirement for current matching to an ideal In0.49Ga0.51P top cell, are investigated as well. Our design technique is considered to be applicable not only to MQW photovoltaics but also to any optoelectronic device based on MQWs to explore the optimum combination of constituent materials toward better performance.
Section snippets
Design method and transport model
On the basis of two essential requirements, we design the thicknesses of well and barrier layers. The first design requirement is the zero-stress (or strain-balance) condition, which is described by (Ekins-Daukes et al., 2002):where t is denoted for the layer thicknesses. ε|| is for the in-plane strain in the layers (). a is for lattice constant. A is for the stiffness parameter. The subscripts w, b, and sub
Mobility mapping and transport analysis
Our proposal is firstly used to design the MQW equivalent of a bulk In0.49Ga0.51P for tandem solar cells. Owing to its bandgap, InGaP serving as a top cell grown on GaAs or Ge substrates in tandem solar cells is rather important for concentrator and satellite systems in space (Friedman, 2010). In triple-junction technologies, for example, a 40.7% efficiency with the conventional metamorphic InGaP/InGaAs/Ge cell under 240-sun concentration (AM1.5D) has been demonstrated (King et al., 2007).
Reexamination of MQW design for GaAs middle cell
MQW absorbers with an effective bandgap of 1.20–1.25 eV are not only essential for current-matched tandem solar cells but also significant for high-efficiency single-junction photovoltaics (Barnham and Duggan, 1990, Katsuyama et al., 1987). In our previous work, an optimized In0.30Ga0.70As/GaAs0.60P0.40 MQW with bandgap-engineered absorption threshold at 1.23 eV has been successfully developed through several attempts of tuning the barrier material, barrier thickness, and so on individually,
Conclusions
In summary, targeting at a particular optical absorption threshold, we proposed a novel design framework for SB-MQW by which both the structural parameters and constituent materials can be uniquely designed. With the first design step, we reduced the design space to 2 parameters by considering design constraints. Next, we have quantitatively analyzed the degradation of carrier transport for all possible SB-MQW designs with identical absorption threshold. Comparing these MQW designs, we have
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
The authors would like to acknowledge financial support from New Energy and Industrial Technology Development Organization (NEDO), Japan (P15003).
References (74)
Dislocations in strained-layer epitaxy: theory, experiment, and applications
Mater. Sci. Reports
(1991)Progress and challenges for next-generation high-efficiency multijunction solar cells
Curr. Opin. Solid State Mater. Sci.
(2010)- et al.
Suppressed lattice relaxation during InGaAs/GaAsP MQW growth with InGaAs and GaAs ultra-thin interlayers
J. Cryst. Growth
(2012) - et al.
New approaches for high efficiency cascade solar cells
Sol. Cells
(1987) - et al.
Self-consistent modeling of escape and capture of carriers in quantum wells
Phys. E Low-dimensional Syst. Nanostructures
(2006) - et al.
GaAsP/Si tandem solar cells: Realistic prediction of efficiency gain by applying strain-balanced multiple quantum wells
Sol. Energy Mater. Sol. Cells
(2018) - et al.
Controlled intrinsic carbon doping in MOVPE-grown GaAs layers by using TMGa and TBAs
J. Cryst. Growth
(2003) - et al.
Defects in epitaxial multilayers
J. Cryst. Growth
(1974) - et al.
In situ investigation of GaAs (0 0 1) intrinsic carbon p-doping in metal-organic vapour phase epitaxy
J. Cryst. Growth
(2000) - et al.
Solar photovoltaic electricity: Current status and future prospects
Sol. Energy
(2011)
Investigation and modeling of photocurrent collection process in multiple quantum well solar cells
Sol. Energy Mater. Sol. Cells
In situ reflectance monitoring for the MOVPE of strain-balanced InGaAs/GaAsP quantum-wells
J. Cryst. Growth
Novel materials for high-efficiency III–V multi-junction solar cells
Sol. Energy
Multi-junction III–V solar cells: current status and future potential
Sol. Energy
Optical properties of In1-xGaxAs1-yPy alloys
Phys. Rev. B
Recent results for single-junction and tandem quantum well solar cells
Prog. Photovolt. Res. Appl.
Experimental and theoretical studies of the performance of quantum-well infrared photodetectors
J. Appl. Phys.
A new approach to high-efficiency multi-band-gap solar cells
J. Appl. Phys.
A new GaAsP—InGaAs strained-layer super-lattice light-emitting diode
IEEE Electron Dev. Lett.
Characterization and analysis of multi-quantum well solar cells
Stress control of tensile-strained In1−xGaxP nanomechanical string resonators
Appl. Phys. Lett.
Physics of photonic devices
GaAs/GaInNAs quantum well and superlattice solar cell
Appl. Phys. Lett.
An approach to high efficiencies using GaAs/GaInNAs multiple quantum well and superlattice solar cell
J. Appl. Phys.
Monte Carlo simulations of carrier transport in AlGaInP laser diodes
IEEE J. Quantum Electron.
Strain-balanced GaAsP/InGaAs quantum well solar cells
Appl. Phys. Lett.
Strain-balanced criteria for multiple quantum well structures and its signature in X-ray rocking curves†
Cryst. Growth Des.
Quantum well carrier sweep out: relation to electroabsorption and exciton saturation
IEEE J. Quantum Electron.
100-period, 1.23-eV bandgap InGaAs/GaAsP quantum wells for high-efficiency GaAs solar cells: toward current-matched Ge-based tandem cells
Prog. Photovolt. Res. Appl.
Evaluation of carrier collection efficiency in multiple quantum well solar cells
IEEE J. Photovolt.
Building a six-junction inverted metamorphic concentrator solar cell
IEEE J. Photovolt.
Solar cell efficiency tables (Version 53)
Prog. Photovolt. Res. Appl.
Cited by (3)
Gradient bandgap modification for highly efficient carrier transport in antimony sulfide-selenide tandem solar cells
2022, Solar Energy Materials and Solar CellsCitation Excerpt :The schematic view of energy level diagrams of devices with non-gradient and front-gradient mid-cells are presented in Fig. 1g and h, respectively. The wx-AMPS is used to design and simulate SbSSe based solar cells [67–69]. By solving elementary equations of semiconductor devices, including Poisson equation and continuity equation of charge carriers, one can automatically calculate and obtain outputs, including J-V curve, carrier generation rate distribution, carrier recombination rate distribution, and carrier concentration distribution.
Hybrid nano and microbial consortium technologies to harvest biofuel (biomethane) from organic and agri waste
2022, Microbial Resource Technologies for Sustainable DevelopmentInGaP Solar Cell with InGaP Multiple Quantum Wells Grown under Optimized V/III Ratio
2023, Physica Status Solidi (A) Applications and Materials Science