Читать книгу Perovskite Materials for Energy and Environmental Applications - Группа авторов - Страница 28

The ASA program is intended to simulate devices based on amorphous and crystalline semiconductors. The ASA program resolves one dimensional elementary semiconductor equations (Poisson equation and electron continuity equation and hole continuity equation) and utilizes variables such as concentration of free electrons, n, and that of holes, p, and electrostatic potential. Additionally, in order to describe explicit device operation and optoelectronic properties of the material it employs several advanced physical.

The ASA one dimensional (1-D) device simulator is very appropriate for simulating thin film structures of silicon solar cells. The program fulfils the usual requirements for simulating thin-film solar cell devices [3].

For modeling silicon thin-film devices the electronic structure of hydrogenated amorphous silicon (a-Si: H) and hydrogenated microcrystalline silicon (μc-Si: H) need to be considered. The atomic structure of a-Si:H shows spatial disorder leading to the energy band gap with edges of conduction band (CB) and valance band (VB) not well-defined with a continuous density of states (DOS). While studying the transport properties of carriers in a-Si:H, in the DOS distribution, it is necessary to differentiate between the localized states and the extended states. Within the mobility gap, the localized states greatly affect the processes of charge trapping followed by recombination. Thus, as is the case with crystalline semiconductor modeling, the charge trapped inside the localized states cannot be neglected. For the calculation of recombination-generation (R-G) statistics different methods are required for the localized states in the mobility gap of a-Si:H as they can be of different nature. The Shockley-Read-Hall (SRH) recombination is not used for amorphous films in device structures in a-Si:H to model recombination as the states introduced by dopants and/or impurities are negligible compared to the recombination via the tail states or dangling bond states. Though, SRH recombination based on the carrier lifetimes can be used for crystalline materials in the ASA program.

From the optical perspective, in order to achieve greater light conversion efficiencies, both the effective use of solar spectrum and light distribution in the solar cells are significant. Light regulation is accomplished in thin film solar cells through the application of techniques for light trapping. The techniques for trapping of light are focused on substrates with textured surface being introduced and using special (back-) reflector layers. The substrates with textured surface provide rough interfaces to the solar cells. The light incident at rough interfaces is scattered and the simulation of solar cells must consider the rough interface scattering mechanisms so as to precisely assess the generation profile for charge carriers within the solar cell. It needs the design of optical models which take into account propagation of both coherent nonscattered (specular) light and incoherent distributed (diffused) light through a system.

To use the solar spectrum efficiently, a multi-junction approach is required in thin-film silicon solar cells. At an interface between two neighbouring junctions, the tunnelling assisted recombination is accountable for the movement of the charge carriers through a solar cell with several junctions. Such an interface is defined as the tunnel-recombination junction (TRJ). Two methods can be used for modeling TRJ in the ASA program. The Delft approach utilizes the improved transport of carriers in the high-field region of TRJ and the trap-assisted tunnelling model. The Pennsylvania approach utilizes the implementation of a strongly defective layer with a sharply compact bandgap at the interface of n/p and the gradation of the n-layer and p-layer mobility gap in the regions next to the defective layer.

The ASA program’s key characteristics are outlined [4]:

 Amorphous and/or crystalline multilayer semiconductor device modeling

 Models including both the extended and the localized (tail and defect) state that describe a complete DOS as an energy function.

 Use of the defect-pool model for the distribution of defectstates in a-Si:H

 Basic statistics for generation-recombination of donor and acceptor states and of ambipolar states

 Optical models for measuring the absorption profile in flat and/or rough interfaces

 Constant adjustment (gradation) of virtually all input parameters as a function of position in the device or in the gap energy level

 Tunnel-recombination junction model

 a-Si:H solar cell degradation modeling

 Written in language ANSI C

The ASA program can be installed on computers working under Windows or Linux operating systems. The ASA program is protected against unauthorised use and therefore its operation is allowed only with the simultaneous use of the ASA protection key [5–7].

Here, one calculation using the ASA program is followed on the basis of the flowchart in Figure 1.2 [3, 8–12]:

1 The ASA program has a built-in parser to run through the input file. After one statement is read, the appropriate part of the program is invoked.

2 The first part of an input file always consists of statements thatDefine the simulated device, i.e. the layers, including their width and the grid.Set electrical parameters of the material, e.g., the valance tail slope parameters Nv0, Evo. These parameters are set for every electrical layer.Set optical parameters, e.g., the refractive index and extinction coefficient of the layers.Set calculation and model settings, e.g., if the Newton method is to be used.

3 After this initialization, the calculations are started with “solve.”The program generates the grid according to the definition earlier in the input file. The arrays of the simulated device are allocated and filled with the (initial) values defined by the given (electrical) parameters. When the defect pool model is selected, the defect density NDB(x) is calculated.If it is not read from external file, the program calculates the generation rate in the device using the GENPRO1, GENPRO2, or GENPRO3 module.One of the five simulation modes can be selected by a “solve” statement:(a) equil: in this mode, the electron, hole densities and the matching potential are calculated in thermodynamic equilibrium. These calculations are automatically carried out before mode 2.(b) jv: the current-voltage characteristics of the simulated device are computed, by using the Poisson and continuity equations. A selection can be made between the Newton and Gummel method. The calculations can be carried out with or without carriers generation by illumination.(c) cv: the capacitance-voltage characteristics of the simulated device are calculated. It is calculated by using mode 2 to find the currents at an applied voltage of and , which gives rise to a difference in charge density Δρ(x). By calculating the difference in surface charge at the front and back contacts, Δσ is also found. These define the capacitance at voltage V.(d) sr: the spectral response (i.e. the external quantum efficiency EQE) is calculated by using mode 2 and the generation rate profile calculated by one of the optical models. The EQE is defined by ΔI/(qΔnphoton), where Δnphoton is the differential number of photons between the (optional) bias spectrum S and the bias spectrum plus a monochromatic probe light S + Δnphoton.(e) rt: the reflection and transmission are calculated. This is carried out by module GENPRO1 (no scattering included).

4 The desired program output can finally be saved to particular files using the “print” statement.

Figure 1.2 Flowchart showing steps for calculation using ASA adapted from [3].