Understanding the structure and fabrication of tunnel diodes gives additional insight into the operation of different tunnel diode devices.
There are many similarities between the tunnel diode structure and that of the standard PN junction, but also there are some key differences that anble the tunnel diode to operate as it does.
To manufacture tunnel diode devices, the standard fabrication processes can be sued, enabling he devices to be made in an economic fashion.
Tunnel diode structure basics
The tunnel diode is similar to a standard p-n junction in many respects except that the doping levels are very high. Densities of the order of 5x1019 cm-3 are common.
A further difference is that the depletion region, the area between the p-type and n-type areas, where there are no carriers is very narrow. Typically it is in the region of between five to ten nano-metres, which equates to a width of only a few atoms.
With a depletion region that is this small, capacitance levels are very small and this lends itself to high frequency operation, and typically performance extends well into the microwave region.
Tunnels diodes can be fabricated from a variety of different semiconductors, but germanium is the one that has been most widely used. It has the advantage that it has a small energy gap that allows for more efficient tunnelling.
Tunnel diode structures & techniques
Tunnel diodes can be fabricated using a variety of different structures. These structures generally fall into one of three basic types:
- Ball alloy tunnel diode structure: This type of tunnel diode format is fabricated as a mesa structure. To achieve this form of structure, the fabrication technique involves bringing an alloy containing the required dopants into contact with a heavily doped substrate. The temperature used is around 500°C at which point the dopants quickly melt and diffuse into the substrate. The overall structure geometry is then defined by etching the diode to the required proportions.
- Pulsed bond tunnel diode structure: This is a relatively straightforward tunnel diode structure to create, although careful process control is required during the fabrication process. The tunnel diode is created by using a wire coated with an alloy containing the required dopants. This is pressed hard onto the heavily doped substrate, and then a voltage pulse is applied. The effect of this is that the junction forms by a process of local alloying.Despite this, there are drawbacks to this process because it can only produce a small junction, and the exact properties, including the area of the junction cannot be controlled tightly.
- Planar tunnel diode structure: Planar technology can be used to create tunnel diodes. Using this approach for the fabrication process, the heavily doped n+ substrate is masked off by an insulating layer to leave a small area exposed. This exposed area is then open to become the active area of the diode.The doping for the area can be introduced by diffusion, alloying or epitaxial growth. Alternatively it is possible to grow an epitaxial layer over the whole surface and then etch away those areas that are not required to leave a mesa structure.
All three structures enable high performance diodes to be obtained.
Tunnel diodes using all of these structures have been manufactured and are still manufactured, although in lower quantities than they used to be. All three of these tunnel diode structures provide good levels of performance and with new manufacturing techniques and technologies, performance can be improved.