Abstract of the Disclosure

A method is described for making terminal pads, and more particularly solderable bump terminals on a plurality of semiconductor devices formed in a silicon wafe r. Two spaced aluminum layers have electrical connection to the P-type region and the N-type region, respectively, of a PN junction of each device. A permanent protective glass layer is deposited over the active face of the wafer, having holes opened over the terminal pads. The wafer is covered with a temporary insulative masking layer except over terminal pad portions of the two aluminum layers, respectively. The wafer is submersed in a zinc plating solution and subsequently in a nickel plating solution to form a nickel bump on the terminal pad portions. A dark environment is provided about the PN junctions to avoid deplating aluminum from some of he aluminum pads due to photo-electric currents in the immersion solution. The masking layer prevents formation of unwanted spurious nickel bumps through cracks in the glass layer during the plating step and further reduces stray current paths to other regions of the wafer.

• Figure 1 shows in cross-sectional view the region of a terminal pad of a semiconductor device after nickel plating with a photo resist layer over the quartz.
• Figure 2 shows in cross-sectional view the region of Figure 1 after removal of the overlying photo resist layer.
• Figure 3 shows in cross-sectional view the region of a terminal pad of a semiconductor device after nickel plating with a photo resist layer over the quartz, which photo resist layer has a larger diameter hole than the mutually concentric hole in the underlying quartz.
• Figure 4 shows in cross-sectional view, after removal of the overlying photo resist layer of Figure 3, a region of the terminal pad of a semiconductor device.

 Citations [54]:
  
3,096,262  07/1963  Shockley
3,495,324  02/1970  Guthrie
3,669,734  06/1972  Jacob et al
3,760,238  09/1973  Hamer et al