Wafer ray tracer
This calculator determines the photogenerated current density in a solar cell or test structure under a chosen illumination spectrum. It can be used to assess or improve the optical properties of a cell or test structure.
The wafer optics calculator complements OPAL 2. Whilst OPAL 2 accurately calculates reflection, transmission and thin film absorption at a single surface, the Wafer Optics Calculator incorporates the effect of the wafer bulk and both of its surfaces. The Wafer Optics Calculator therefore permits the assessment of all optical losses and light trapping.
To simulate the optical behaviour of a wafer, the Wafer Optics Calculator must take a more computationally intensive approach than OPAL 2. Therefore, it is typically slower, and provides a less precise estimate of its outputs.
Basic Algorithm
The wafer optics calculator combines Monte Carlo ray tracing with thin film optics. It calculates the photogenerated current in a wafer via the following algorithm:
-
A number of light rays are created above the front surface of the defined structure.
-
Each light ray proceeds along a straight line until it intersects with a facet of the wafer surface.
-
At each interaction:
-
reflectance, transmittance and absorptance are calculated; the value of each depends on the wavelength λ and the electric field (polarisation), as well as the refractive indices of any thin films or materials on either side of the interface.
-
the magnitudes of reflectance and transmittance are translated into probabilities. Next, a random decision, weighted by these probabilities, is taken to either reflect or transmit the ray at this interface.
-
As the ray passes through absorbing media, its intensity is reduced. In a semiconductor like silicon, this absorption process equates to the photogeneration of electron-hole pairs.
-
Steps 2 to 4 are repeated for each ray until:
-
the ray is lost from the structure (it may have been reflected from the front surface, escaped, or transmitted through the wafer);
-
the ray's intensity falls below a threshold (its energy has been absorbed by the wafer or thin films); or
-
the ray has intersected with the maximum allowable number of interfaces.
-
The gains (photogeneration) and losses (reflection, transmission, parasitic absorption) are recorded for each ray. The global gains and losses are determined by averaging a large number of rays. With a sufficiently large number of rays, the Monte Carlo simulation converges to the physical model.
Defining a structure
To define the optical structure of a wafer, you may choose:
-
front and rear texture morphology; and
-
materials, with refractive indices given in the Refractive Index Library, for the following:
-
a surrounding material (eg. air or EVA);
-
front thin film coatings and thicknesses (eg. SiNx);
-
a substrate material and thickness (eg. a silicon wafer); and
-
rear thin film coatings and thicknesses or a rear reflector, with defined internal reflectance Rr,int, transmittance Tr,int and absorptance Ar,int, as well as defined lambertian fraction (Flamb = 1 is a perfect diffuse reflector, while Flamb = 0 is a specular mirror)."
Calculator outputs
The Wafer Optics Calculator determines the optical losses and photogeneration in the wavelength range requested. It weights the magnitudes by the photon flux in the user-defined spectrum, then integrates over the wavelength, in order to calculate photon current density. The photogenerated current JG in a wafer equates to the short circuit current that could be extracted from a perfect solar cell made from the wafer.
Improving precision
The Monte-Carlo algorithm employed by the Wafer Optics Calculator necessarily results in output uncertainties. These uncertainties are calculated by dividing the user-requested number of rays into several sub-simulations, then applying statistical analysis to the set of sub-simulations to arrive at a mean value and 95% confidence interval (about two standard deviations).
Uncertainties are largest when the number of rays simulated is small. It is possible to reduce the displayed uncertainty by choosing fewer wavelengths (increasing the wavelength interval or narrowing the range). Note, however, that the displayed uncertainty does not account for uncertainty caused by smoothing of the spectrum. The best way to reduce uncertainty is to increase the number of rays simulated. Unfortunately, this will inevitably increase computation time.
Note that in the current version, the uncertainties provided do not account for the impact of weakly absorbed well trapped rays that are not traced to their proper conclusion. To minimise adverse impacts on results, choose a low intensity limit and high maximum number of bounces."
Disclaimer
Neither PV Lighthouse nor any person related to the compilation of this calculator make any warranty, expressed or implied, or assume any legal liability or responsibility for the accuracy, completeness or usefulness of any information disclosed or rendered by this calculator.