The two key variables of an Si solar cell, i.e., emitter (n-type window layer) and base (p-type substrate) doping levels or concentrations, are studied using Medici, a 2-dimensional semiconductor device simulation tool. The substrate is ptype and 150 μm thick, the pn junction is 2 μm from the front surface, and the cell is lit on the front surface. The doping concentration ranges from 1 × 1010 cm−3 to 1 × 1020 cm−3 for both emitter and base, resulting in a matrix of 11 by 11 or a total of 121 data points. With respect to increasing donor concentration (Nd) in the emitter, the open-circuit voltage (Voc) is little affected throughout, and the short-circuit current (Isc) is affected only at a very high levels of Nd, exceeding 1 × 1019 cm−3, dropping abruptly by about 12%, i.e., from Isc = 6.05 × 10−9 A·μm−1, at Nd = 1 × 1019 cm−3 to Isc = 5.35 × 10−9 A·μm−1 at Nd = 1 × 1020 cm−3, likely due to minority-carrier, or hole, recombination at the very high doping level. With respect to increasing acceptor concentration (Na) in the base, Isc is little affected throughout, but Voc increases steadily, i.e, from Voc = 0.29 V at Na = 1 × 1012 cm−3 to 0.69 V at Na = 1 × 1018 cm−3. On average, with an order increase in Na, Voc increases by about 0.07 V, likely due to narrowing of the depletion layer and lowering of the carrier recombination at the pn junction. At the maximum output power (Pmax), a peak value of 3.25 × 10−2 W· cm−2 or 32.5 mW· cm−2 is observed at the doping combination of Nd = 1 × 1019 cm−3, a level at which Si is degenerate (being metal-like), and Na = 1 × 1017 cm−3, and minimum values of near zero are observed at very low levels of Nd ≤ 1 × 1013 cm−3. This wide variation in Pmax, even within a given kind of solar cell, indicates that selecting an optimal combination of donor and acceptor doping concentrations is likely most important in solar cell engineering.

Continuous efforts are being made to improve the efficiency of Si solar cells, which is the prevailing technology at this time. As opposed to the standard front-lit solar cell design, the back-lit design suffers no shading loss because all the metal electrodes are placed on one side close to the pn junction, which is referred to as the front side, and the incoming light enters the denuded back side. In this study, a systematic comparison between the two designs was conducted by means of computer simulation. Medici, a two-dimensional semiconductor device simulation tool, was utilized for this purpose. The 0.6 μm wavelength, the peak value for the AM-1.5 illumination, was chosen for the incident photons, and the minority-carrier recombination lifetime (τ), a key indicator of the Si substrate quality, was the main variable in the simulation on a p-type 150 μm thick Si substrate. Qualitatively, minority-carrier recombination affected the short circuit current (Isc) but not the opencircuit voltage (Voc). The latter was most affected by series resistance associated with the electrode locations. Quantitatively, when τ ≥ 500 μs, the simulation yielded the solar cell power outputs of 20.7 mW·cm−2 and 18.6 mW·cm−2, respectively, for the front-lit and back-lit cells, a reasonable 10 % difference. However, when τ < 500 μs, the difference was 20 % or more, making the back-lit design less than competitive. We concluded that the back-lit design, despite its inherent benefits, is not suitable for a broad range of Si solar cells but may only be applicable in the high-end cells where float-zone (FZ) or magnetic Czochralski (MCZ) Si crystals of the highest quality are used as the substrate.

The BCBJ (Back Contact and Back Junction) or back-lit solar cell design eliminates shading loss by placing the pn junction and metal electrode contacts all on one side that faces away from the sun. However, as the electron-hole generation sites now are located very far from the pn junction, loss by minority-carrier recombination can be a significant issue. Utilizing Medici, a 2-dimensional semiconductor device simulation tool, the interdependency between the substrate thickness and the minority-carrier recombination lifetime was studied in terms of how these factors affect the solar cell power output. Qualitatively speaking, the results indicate that a very high quality substrate with a long recombination lifetime is needed to maintain the maximum power generation. The quantitative value of the recombination lifetime of minority-carriers, i.e., electrons in p-type substrates, required in the BCBJ cell is about one order of magnitude longer than that in the front-lit cell, i.e., 5 × 10−4 sec vs. 5 × 10−5 sec. Regardless of substrate thickness up to 150 μm, the power output in the BCBJ cell stays at nearly the maximum value of about 1.8 × 10−2 W·cm−2, or 18 mW·cm−2, as long as the recombination lifetime is 5 × 10−4 s or longer. The output power, however, declines steeply to as low as 10 mW·cm−2 when the recombination lifetime becomes significantly shorter than 5 × 10−4 sec. Substrate thinning is found to be not as effective as in the front-lit case in stemming the decline in the output power. In view of these results, for BCBJ applications, the substrate needs to be only mono-crystalline Si of very high quality. This bars the use of poly-crystalline Si, which is gaining wider acceptance in standard front-lit solar cells.

In photovoltaic power generation where minority carrier generation via light absorption is competing against minority carrier recombination, the substrate thickness and material quality are interdependent, and appropriate combination of the two variables is important in obtaining the maximum output power generation. Medici, a two-dimensional semiconductor device simulation tool, is used to investigate the interdependency in relation to the maximum power output in front-lit Si solar cells. Qualitatively, the results indicate that a high quality substrate must be thick and that a low quality substrate must be thin in order to achieve the maximum power generation in the respective materials. The dividing point is 70 μm/5 × 10−6 sec. That is, for materials with a minority carrier recombination lifetime longer than 5 × 10−6 sec, the substrate must be thicker than 70 μm, while for materials with a lifetime shorter than 5 × 10−6 sec, the substrate must be thinner than 70 μm. In substrate fabrication, the thinner the wafer, the lower the cost of material, but the higher the cost of wafer fabrication. Thus, the optimum thickness/lifetime combinations are defined in this study along with the substrate cost considerations as part of the factors to be considered in material selection.