What is diffusion in a semiconductor

Semiconductor technology from A to Z

With ion implantation, charged dopants (ions) are accelerated in an electric field and directed onto the wafer. The depth of penetration can be determined very precisely by reducing or increasing the voltage required to accelerate the ions. Since the process takes place at room temperature, previously introduced doping cannot diffuse out. As with diffusion, areas that are not to be doped are covered with a mask, with a masking made of photoresist being sufficient for the implantation.

An implanter consists of the following components:

  • Ion source: The doping gas (e.g. boron trifluoride BF3) is ionized (electrons are emitted by a hot cathode and collide with the gas particles. Impact ionization always produces positive ions and free electrons)
  • Pre-accelerator: The ions are drawn from the ion source with approx. 30 kiloelectron volts
  • Mass separator: The charged particles are deflected by 90 ° by a magnetic field. Too light / heavy particles are deflected more / less than the desired ions and are intercepted with screens behind the separator
  • Acceleration distance: The particles are accelerated to their final energy with several 100 keV (200 keV accelerate boron ions to approx. 2,000,000 m / s)
  • Lenses: Lenses that focus the ion beam are distributed over the entire system
  • Deflection Devices: Capacitors deflect the ions in order to irradiate the desired location
  • Wafer station: The wafers are brought into the ion beam and irradiated either individually or on large rotating wheels

Representation of an implantation system

Penetration depth of ions in the wafer

In contrast to diffusion, the particles do not penetrate due to their own motion, but are shot into the crystal lattice at high speed. They are slowed down by colliding with the silicon atoms. The silicon atoms are knocked off their lattice sites as a result of the impact, and the doping ions themselves usually accumulate on interstitial lattice sites. There they are not electrically active because there are no bonds with other atoms that could cause free charge carriers. The displaced silicon atoms have to be reintegrated into the crystal lattice and the electrically inactive dopants have to be activated.

Healing of the crystal lattice and activation of the dopants

The dopants are moved to lattice sites by a temperature step at approx. 1000 ° C (previously only approx. 5% of the doping atoms are on lattice sites). The grid damage caused by the collisions is healed at around 500 ° C. Since the doping atoms move in the substrate during the high temperatures, these steps are only carried out for a very short time.

Channeling

The substrate used is a single crystal, i.e. the silicon atoms are arranged regularly and form channels. The injected dopant atoms then run parallel to these channels, are only slightly slowed down and penetrate very deeply into the substrate. There are two ways to prevent this:

  • Wafer alignment: The wafers are deflected by approx. 7 ° to the beam direction. As a result, the ions are not injected parallel to the grid channels and are slowed down early by collisions.
  • Scattering oxide: A thin oxide is applied to the wafer surface, which deflects the ions and prevents them from reaching the substrate perpendicularly

Characteristic

  • The reproducibility of the ion implantation is very high
  • The process at room temperature prevents other dopants from diffusing out
  • Photoresist is used as a mask, oxide as in diffusion is not necessary
  • Ion implanters are expensive and the cost per wafer processed is high
  • The dopants do not spread laterally under the mask (only minimally through collisions)
  • Almost every element can be implanted in the highest purity
  • Similar to the deposits of dopants during diffusion in the quartz tube, ions can be deposited on walls or panels, which are detached during subsequent implantations and get onto the wafer
  • Three-dimensional structures (e.g. trenches) cannot be doped with ion implantation
  • The implantation process takes place under a high vacuum, which has to be generated with several turbo-molecular or cryopumps

There are different types of implanters, mostly medium and high current implants being used. Medium current implanters are suitable for small to medium doses of ions (1 x 1011–1·1015 Ions / cm2), High current implanter for doses of 1 x 1015–1·1017 Ions / cm2.
Ion implantation has for the most part established itself due to its advantages over diffusion.

Alloy doping

For the sake of completeness, it should also be mentioned that in addition to these two processes, there is also doping by means of an alloy. However, since this process has disadvantages such as cracking in the substrate, it is hardly used in today's semiconductor technology.