Corescan application examples:

Examples from a firing temperature optimisation

Cells were fired in an IR belt furnace at different furnace temperature settings while using the same belt speed for all cells. The belt direction was always perpendicular to the busbars. The furnace set temperature in the hottest zone was varied from T-45 °C to T+45 °C, where T is defined to be the temperature that results in the highest fill factor (FF). The Corescans on these cells are shown below; the short-circuit current density generated within the light beam area was 30 mA/cm², as usual. On the colour scale at the right, the potential is indicated in mV.

The contact resistance is highest in the white regions of the scans; for the underfired cells there is a badly contacted region at the centre of the cell, while the edges are badly contacted in case of the overfired cells. This can be explained by a varying temperature across the belt width, being lowest at the centre of the cell and highest at the edges: at the cool centre it is too cold when underfired and OK when overfired; at the warm edges the temperature is OK for underfiring and too hot in the case of overfiring. At the optimum firing temperature the result is reasonable, although there remain some small bad spots. These bad spots are probably due to the fact that some crystal orientations give rise to higher contact resistance, although the crystal orientations on these specific cells have not been checked (more details about this subject are given in the next section). Due to the large temperature variations over the cells the firing window is much smaller than it could have been with a constant temperature. These scans therefore show that the firing process can be made much less sensitive to the firing temperature when the furnace design is improved.

This 10 x 10 cm multi-crystalline cell seems to have a random pattern of high contact resistance regions. In this case it was found that the increased contact resistance is related to the orientation of the crystals. Cells made from wafers that have undergone alkalic saw damage removal are sensitive to this effect. The contact resistance is high when the normal on the crystallographic planes is near the <100> direction, while it is low when the normal is orientated closer to the <111> or <110> directions. The fundamental cause for this difference is is currently being investigated. Interestingly, when acid saw damage removal was used this effect did not not show up.
The colour scaling is the same as for the firing optimisation examples (the numbers indicate the potential in mV).

The effect of supports below a cell

Four regularly oriented spots of poor contact are seen in the center of this cell. Apparently, the temperature was too low at these 4 locations where the cell was touched by a quartz support.
The colour scaling is the same as for the firing optimisation examples (the numbers indicate the potential in mV).

The Corescan shows a narrow band of high contact resistance along the edges of the cell. The plasma etching of the edges that was used for cell isolation has also removed some of the emitter near the edges, resulting in an increased contact resistance.
The colour scaling is the same as for the firing optimisation examples (the numbers indicate the potential in mV).

This Corescan shows contact resistance non-uniformity caused by non-uniform emitter doping. In this example, a cell on both sides doped by spinning has higher contact resistance at the rim of a circle with the same diameter as the spinning head, where the emitter has been less doped.
The colour scaling is the same as for the firing optimisation examples (the numbers indicate the potential in mV).

The overall contact resistance problem on a cell can be evaluated from the measured Corescan data on the so-called scan evaluation panel. The evaluation is done by determining the percentage of measurement data that falls within each of the 4 potential ranges that are used (and that have adjustable limits). If the percentage of points in a range exceeds its (adjustable) range limit for at least one of the ranges, a red sign is given to indicate that the cell is not acceptable.
In the figures below, a typical scan for a badly contacted cell is shown on the left and its evaluation is shown on the right:

It must be noted that the potential range limits were not optimised for this example , each individual user should determine what is acceptable and what is not. The scan evaluation tool can of course also be used for data from the other methods, but it is most useful for the Corescan method.