Alloy Junction

Making a PNP Alloy Junction Transistor

For the full history of Texas Instruments on this site please return to The Early History of Texas Instruments Semiconductors.

Ed Millis kindly provided the following additional details:

“These transistors were assembled manually. Firstly the un-mounted transistor was assembled in multiple wells in graphite boats designed to hold the following components in place:

Indium alloy collector dot

base tab

base tab solder die

Wafer of N type germanium (the transistor base) about 125mils square and 10-15 mils thick

Indium alloy emitter dot held exactly opposite the collector dot

The boat with its load of preforms was then loaded into an oven in order to alloy the indium dots to the base wafer and solder the base tab to the base. This process formed the PNP transistor as the alloying process converted the surfaces of the N-type base to P-type.

The objective was to leave a very thin N-Type base layer: too fat and gain and frequency performance suffered, too thin could lead to shorted transistors due to the P-Type layers meeting. Transistor geometry was controlled by oven time and temperature and dimensions of the wafer and dots.

Then the transistor was mounted on its header. The first step was to spot-weld the transistor's tab to the base lead of the header, which was bent to be centered. Then using a binocular microscope a short piece of gold wire (2 mils in diameter) was welded the header collector connection (and similarly the emitter). These wires, properly installed, would be pointed upward toward the transistor. Then, using tweezers and a needle-pointed soldering iron of low wattage, the loose end of the gold wire was brought to bear against each fused indium dot and soldered into it. This was done on each side to make the collector and emitter connections.

Then the assembly was etched to clean up the transistor and boiled in alcohol as a final clean up.

Canning was done by hand, with a soldering iron, soft solder, and a dab of flux." [Millis 2011]