DOE OSTI2017
There has long been a desire to replace the front-side silver contacts in silicon solar cells. There are two driving forces to do this. First, silver is an expensive precious metal. Secondly, the process to use silver requires that it be formulated into screen print pastes that need a lead-containing glass frit, and the use of lead is forbidden in many parts of the world. Because of the difficulty in replacing these pastes and the attendant processes, lead exemptions have granted to solar cells. Copper has been the replacement metal of choice because it is significantly cheaper than silver and is very close to silver in electrical conductivity. Using processes which do not use lead, obviates it as an environmental contaminant. However, copper cannot be in contact with the silicon of the cell since it migrates through the silicon and causes defects which severely damage the efficiency of the cell. Hence, a conductive barrier must be placed between the copper and silicon and nickel, and especially nickel silicide, have been shown to be materials of choice. However, nickel must be sputtered and annealed to create the nickel silicide barrier, and copper has either been sputtered or plated. All of these processes require expensive, specialized equipment and plating uses environmentally unfriendly chemicals. Therefore, Intrinsiq proposed using printed nano nickel silicide ink (which we had previously invented) and printed nano copper ink to create these electrodes and barriers. We found that nano copper ink could be readily printed and sintered under a reducing atmosphere to give highly conductive grids. We further showed that nano nickel silicide ink could be readily jetted into grids on top of the silicon cell. It could then be annealed to create a barrier. However, it was found that the combination of printed NiSi and printed Cu did not give contact resistivity good enough to produce efficient cells. Only plated copper on top of the printed NiSi gave useful contact resistivity, and that proved to five to ten times less conductive than the commercial silver grids. Even so, the NiSi layer was a very good barrier to copper migration, even under harsh environmental conditions. Additionally, both plated copper and printed copper could be soldered to. While it may be possible to produce an all printed copper/nickel silicide top electrode for silicon cells, it was not easily demonstrated within the time and monetary constraints of the present project. Additionally, potential customers have told us that having to laser ablate the anti-reflection coating of cells to create a connection for NiSi, and the addition of two printing and annealing (sintering for copper) steps, adds too much expense to compensate for any potential cost savings from using copper. The cost benefits of copper have been further eroded by the facts that over the lifetime of this project, the cost of silver electrodes decreased due to manufacturers finding ways to use less and less silver, and inventing pastes which use less costly silver materials to begin with. All of these factors were considered and led to the decision to stop the program before actual manufacturing scale was attempted.