Design strategies for energy-harvesting photovoltaics in diverse environments

Abstract

Indoor photovoltaics (IPVs) are small and not optimized for versatile environments, making them environmentally sensitive. To expand the application of energy-harvesting photovoltaics, overcoming the current problems and mismatch loss is important. In this study, we found that IPVs are sensitive to changes in current density under low illuminance, and we introduced a protocol to reveal the modules resulting in the smallest standard deviation using current maps. For an IPV of 100 cm², dividing the area into nine cells was most beneficial irrespective of the illuminance considering loss due to contact. Therefore, we introduced the small-area, high-voltage concept to IPVs and revealed the module design that minimizes the mismatch loss. As a result, we demonstrate a rough taxonomic classification into three groups that can be used as fundamental information for selecting an IPV. We expect that scaling from two dimensions to three dimensions is worthy of further research.

Keywords: Applied sciences; Energy application; Energy engineering.

Graphical abstract

Figure 1

Power output analysis under indoor light (A) A photograph and schematic image of the location where measurements were performed. (B) The illuminance ratio from the center measured at A. See also Figures S1 and S2. (C) The maximum power (P_max) ratio from the center measured on the pipe at A. (D) The P_max according to the light intensity. The “high-J” parameter is based on the value obtained under standard test conditions for the cell parameters, and the “low-J” parameter is based on the value obtained under indoor light. (E) The P_max according to the current density, as calculated using LTspice.

Figure 2

EQE map (A) EQE maps of the cells were encapsulated with polydimethylsiloxane (PDMS). From the left, the heterojunction technology solar cell (HJT), passivated emitter and rear cell (PERC), and interdigitated back contact (IBC) solar cell were used. (B) EQE maps of the bare cells. From the left, the HJT, PERC, and IBC solar cell were used. See also Figures S3 and S4.

Figure 3

Power loss according to uniformity (A) Power loss tendency due to resistance when there is no mismatch. (B) Power loss tendency due to resistance when a mismatch exists. See also Figure S5.

Figure 4

The effect of the divided cells in the module (A) The current density (left) and power (right) when the current density of each cell was 0.2 mA cm⁻². (B) The current density (left) and power (right) when the current density of each cell was 0.4 mA cm⁻². (C) The current density (left) and power (right) when the current density of each cell was 0.8 mA cm⁻². See also Figure S6.

Figure 5

Power of the module in four regions in current density (A) Power output assuming the current density of 4 mA/cm². It was assumed to be more than 10,000 lx illuminance. (B) Power output assuming the current density of 1.6 mA/cm². The assumed illuminance ranges from 4,000 lx to 10,000 lx. (C) Power output assuming the current density of 0.8 mA/cm². The assumed illuminance ranges from 2,000 lx to 3,000 lx. (D) Power output assuming the current density of 0.2 mA/cm². It was assumed to be less than 500 lx illuminance. See also Figure S6.

Figure 6

Symmetry design for power imbalance (A) The current-density map (J-map) corresponding to Figure 1B is shown as the original (left top). The J-map distorted from the original with a linear function (right top), a parabolic function (left bottom), and a cubic function (right bottom) with 5% variation in the y-direction. See also Figures S1, S2, S7, and S8. (B) The symmetry module designs of linear (left) and rotational (right). (C) The relative current density according to the number of cells in the module. For the calculation, the linear symmetry module design was adopted. The distortion in the y-direction varied from 1% to 20%. (D) The standard deviation in the y-direction. This graph uses five different variations in the x-direction. See also Figure S8.

Figure 7

Power optimization via SD analysis (A) From Figure 5A, the J-maps distorted with a linear function (left), a parabolic function (middle), and a cubic function (right) with 20% variation in the y-direction. See also Figure S8. (B) The example module designs for high-symmetry conditions. (C) The relative current density according to the number of cells in module with a linear function (left), a parabolic function (middle), and a cubic function (right). σ values indicated standard deviation (SD) for the x-, y-, diagonal, and rotational directions.

Figure 8

Imbalanced current-density maps using polynomial functions The J-maps using polynomial functions in Figure S8. The combined functions are presented at the top of each graph and ordered x-y. Except the L-P1 small x-pattern and small y-pattern, both directions used 50% variation. The L-P1 small x-pattern used 25% variation in the x-direction, and L-P1 small y-pattern used 25% variation in the y-direction from L-P1. The arrow direction in the graphs is the arrangement direction for minimum loss. See also Figure S8.