Use of a solar cell in concentrator PV technology requires reduction in its series resistance in order to minimize the resistive power losses. The present paper discusses a methodology of reducing the series resistance of a commercial c-Si solar cell for concentrator applications, in the range of 2 to 10 suns. Step by step optimization of commercial cell in terms of grid geometry, junction depth, and electroplating of the front metal contacts is proposed. A model of resistance network of solar cell is developed and used for the optimization. Efficiency of unoptimized commercial cell at 10 suns drops by 30% of its 1 sun value corresponding to resistive power loss of about 42%. The optimized cell with grid optimization, junction optimization, electroplating, and junction optimized with electroplated contacts cell gives resistive power loss of 20%, 16%, 11%, and 8%, respectively. An efficiency gain of 3% at 10 suns for fully optimized cell is estimated.

Solar PV technology is gaining importance as one of the major alternative source of energy. This is evident from the increase in solar cell production and demand during the past years [

In case of c-Si modules about 50% is contributed by the base material, Si [

The low-cost potential of concentrator solar cells is due to reduction in cell area for a given power output. Cell area decreases as inverse of the concentration ratio. Due to this inverse relationship, reduction in cell area is about 90% of the 1-sun cell area for concentration ratio of 10 suns. Further increasing the concentration ratio does not result in significant cell area reduction, and hence it does not result in further significant cost reduction when the solar cell efficiencies are assumed to be low, in range of 14 to 16%, as in case of commercially available c-Si cells. The advantage of working with low concentration ratio c-Si cell technology (2 to 10 suns) is that the processes which are used currently in industry can be used to fabricate solar cells suited for concentration. Another advantage of low-concentrator PV systems is that it offers higher tolerance for sun-tracking, both in terms of tracking accuracy and tracking infrastructure [

One of the major issues for the solar cell to operate at concentration levels is its series resistance,

The present paper discusses an analytical approach for designing a c-Si concentrator solar cell for 10 suns application. It is shown that the current industrial solar cell fabrication setup which is used for 1 sun cell fabrication can also be used for 10 suns cell fabrication. Design of solar cell is presented in steps of grid geometry optimization, junction depth optimization and electroplated metal contact, and finally combination of these. Reduction in series resistance at various stages of cell design fabrication is analysed, and effect of individual design steps on solar cell performance is presented.

The current in a solar cell flows vertically in the base, horizontally in the emitter layer, then through the fingers and bus bars. During the flow of the current the solar cell has to overcome various resistances, in the bulk, emitter, and the metal grid region. Most of the resistive losses occur in the top region of the solar cell especially in the thin emitter region and in the metal fingers. In order to estimate the series resistance of a cell under concentration, a simplified resistive network is derived based on the resistance model given by Handy [

Since current flow is one-dimensional, the current generated in the emitter region is assumed to be collected at the fingers and then transported to the bus-bar; hence no direct current flow is considered from the emitter to bus-bar. This eliminates the contact resistance between the bus-bar and the emitter and the emitter resistance between the semiconductor and the bus-bar; thus making the resistive model simplified. Power loss equations are considered for calculating the resistance in the emitter layer and metal grid fingers. Contact resistance between the metal and semiconductor is calculated from the contact resistivity information available from literature.

A solar cell of length,

Resistance network of the solar cells (

Grid optimization is an important exercise for keeping the resistive losses and shading losses to a minimum level. It is more important for concentrator cell application as in this cases that the generated current levels are higher than 1 sun levels which results in higher resistive losses, that is, higher

The size and shape of the front contact grid is tradeoff between the shading and the resistive power loss. More number of lines or wide fingers would reduce the resistance but at the same time will cause shading that will reduce the short-circuit current. Less numbers of fingers would result in increased short-circuit current and higher resistive loss in the emitter. Thus the number of fingers and width of fingers and bus-bars need to be optimized for a given area and power loss.

A model of resistance network explained in Section

From (

The grid optimization is carried out while varying only two parameters: the number of fingers,

Geometrical parameters used in the simulation.

4 | cm | |

4 | cm | |

45 | ||

3 | m | |

150 | ||

2 | mm | |

11 | Nil | |

3.7 | cm | |

3.5 | mm | |

0.1 | m^{2} |

Flow chart for grid geometry optimization.

The flowchart begins with input parameters of unoptimized concentrator solar cell (Table

Optimization of

Maximized values for

Normalized plot of Efficiency against the concentration ratio of the optimized and unoptimized grid solar cell.

From Figure

The experimentally obtained efficiency of commercial 1 sun cell follows similar pattern with respect to concentration level as that of unoptimized cell but shows higher performance over range of concentration ratio. Such higher performance can be attributed to the fact that the commercial cell used for comparison was ^{2} in size obtained by dicing from ^{2} large area cells. The grids of the large area 1 sun commercial cells are designed to carry higher current than what is produced in ^{2} area. Since current generated in ^{2} cells is smaller, the resistive power losses in cells are smaller and hence performance is higher under light concentration. Normally the performance of small cells obtained from large area commercial cells peaks at about 2 to 3 suns concentration [

From the above discussion it is explained that the grid-optimized cell is designed to have lower value of series resistance as compared to the commercial one sun cell and the unoptimized cell, but its series resistance is still higher for its operation at higher concentration levels of more than 2 suns. To overcome these limitations, junction depth optimization and electroplating of the grid contacts, an approach in designing of solar cells is proposed in the next sections.

The cell performance under low concentration levels can further be improved by emitter optimization, as stated earlier. The typical value of the emitter sheet resistance of a commercial Si solar cell is in the range of 40 to 60

Based on the study done in Section ^{−3}:
_{,}

Flow chart for emitter junction depth optimization along with grid optimization.

Also a correlation for

Normalized plot of the Junction depth-optimized solar cells (plot normalized with the highest efficiency).

A normalized efficiency plot of junction-optimized cell, grid-optimized cell, un-optmized cell, and commercial 1 suns cell against the concentration ratio is shown in Figure

Normalized efficiency plot of cell designs for the junction-optimized-electroplated cell, electroplated cell, grid-optimized cell, and unoptimized and commercial 1 sun cell.

Figure

The resistive power loss at 10 suns is reduced from 20% for grid-optimized cell to about 16% for junction-optimized cells. Thus by optimizing the junction depth the resistive power loss in the sheet resistance is reduced; a better performing solar cell at low concentration ratio can be designed.

Front metal grid (fingers and bus-bars) and back Al metal contacts on solar cells are usually fabricated using screen printing of the metal paste. The front contact paste is a mixture of Ag with various organic bonders and additives while the back contact is Al paste. The screen printing paste has organic bonders and additives to make a better contact with Si. These bonders and additives evaporate during the cofiring of the contacts thus leaving behind the vacant spaces which result in reduced metal density of the fingers. This increases the sheet resistance and results in higher resistive power loss. Typically the sheet resistance of the screen printed contacts is in the range of 5 to 2 m

Once the contact printing and firing has taken place, the front contacts could then be electroplated in an Ag bath using light-induced plating techniques as explained by Mette et al. in [

The grid optimization study, described in Section

A comparison of normalized efficiency against concentration ratio for electroplated contacts cell, junction-optimized cell, grid-optimized cell, unoptimized, and commercial 1 sun cell is shown in Figure

As studied in Section

The procedure for optimizing junction depth remains similar to what shown in flow chart of Figure

Normalized efficiency variation with concentration ratio for different junction depth on grid-optimized and electroplated solar cells.

Section of the top region of the solar cell indicating the fingers, bus-bars, and the emitter region.

Indicating the resistive network in the top region of the solar cell.

A normalized plot summarizing the results of all the cells designs (studied from Section

Table

Summary of results of cell design

Cell design | Approximate change in Efficiency at 10 suns | Resistive Power loss with respect to power generated at 10 suns |
---|---|---|

Junction optimized, electroplated cell (case 4) | +2 −3% | 8.7% |

Electroplated cell (case 3) | −1% | 11.25% |

Junction optimized cell (case 2) | −5% | 16.49% |

Grid-optimized cell (case 1) | −10% | 20.15% |

Commercial 1 sun (experimental data) | −30% | Not available |

Unoptimized 1 sun (case 0) | −35% | 42.2% |

It is shown that using the optimization techniques described in this paper for cell design, it is possible to design and fabricate low-concentrator c-Si solar cells (2- to 10 suns) using industrially viable cell processes. Among the four optimization processes described for concentrator solar cell design, the junction depth formation and front contact grid formation are the routine procedure followed during the solar cells fabrication. Which makes them suitable for implementation in case of low-concentrator cells without any additional process step. In case of electroplated contacts of case 3 an additional process of electroplating is required on the front metal contacts to reduce the series resistance. Since the electroplating is an industrial process which is carried out on a mass scale, it is possible to use this process on a commercial solar cell manufacturing. The use of light induce plating (LIP) has made the electroplating a much faster processing technique and hence a higher through put process.

A methodology for step by step reduction in series resistance for design of low level concentrator solar cell (2 to 10 suns) is presented. A resistive model is developed for analysis of concentrator solar cells. A commercial 1 sun solar cell is converted to work under low level concentration by optimizing the front grid and junction depth and varying the front metal lines resistance. These optimization processes can be implemented in commercial fabrication setup of solar cells. At 10 suns, the estimated resistive power loss as compared to generated power is 42%, 20%, 16%, 11% and 8% for unoptimized cell, grid-optimized cell, junction-optimized cell, electroplated cell, and junction-optimized, electroplated cell. The reducing trend of resistive power loss indicates that the proposed optimized commercial cell can be used for increasing concentration ratio up to 10 suns. A commercial cell optimized for grid junction and low metal line resistance has shown an efficiency improvement of 3% at 10 suns concentration.

The calculation of series resistance from the individual resistance is explained in the appendix. Power loss equations in emitter grid are used to calculate the resistance in that region. Continuity equations are used in deriving the power loss equations. Figure

Integrating (

The power loss from (

The overall series resistance is given by the following equation: