Dissemination of IT for the Promotion of Materials Science (DoITPoMS)

DoITPoMS Teaching & Learning Packages Physical Vapour Deposition of Thin Films Sputtering
Physical Vapour Deposition of Thin Films AimsBefore you startIntroductionEvaporation theoryEvaporation TechniquesSputteringComparisons and ComplicationsGrowth ModesSummaryQuestionsGoing further
TLP creditsTLP contentsShow all content
Viewing and downloading resourcesAbout the TLPsTerms of useFeedbackCredits Print this page
PreviousNext

Sputtering

Sputtering is an etching process. The source (known as the target) is bombarded with a high energy species, leading to the ejection of a vapour flux. Sputter deposition therefore uses this flux as the vapour source for film growth. It principally consists of atoms, with a range of energies, travelling away from the target at random angles. Sputtering is a purely physical process and is most simply modelled by assuming elastic binary collisions. The Sputter Yield (S) is defined as the average number of sputtered atoms per collision and it is affected by many variables. Forming an exact relationship between these parameters and the sputter yield is very difficult. Some of the contributing factors are:

  • The momentum and energy transfer coefficients
  • The temperature of the target
  • The bond strength of the target atoms
  • The incident particle energy
  • The incident angle at the target surface

Typically, the target is bombarded with noble gas ions such as Argon (with energies in the range of 100-500eV). In this case, it is found that S~1 for most metals.

DC Glow Discharge Sputtering

The most common method of sputter deposition uses a self-sustaining discharge in a low pressure inert gas. The sputtering target is the cathode and ejects secondary electrons. These collide with the inert gas atoms, which become positively charged and accelerate towards the target. This causes sputtering on impact.

DC glow discharge sputtering

In order to maintain the discharge, the gas pressure needs to be high enough that the secondary electrons collide with and ionise gas atoms before they are lost to the surroundings. The sputtered flux then has to travel through this gas in order to reach the substrate. This scatters it, meaning that the setup is quite inefficient.

The following animation shows the different stages of the sputtering process:

 

If the required target is not electrically conductive, then a Radio Frequency voltage can be used to develop a negative potential on an insulating target surface.

Magnetron Sputtering

The principle used here is to add a magnetic field at the target surface. This means that the secondary electrons which are ejected from the target are in a region of crossed electric and magnetic fields, leading to cycloidal motion. This traps the electrons near the target surface, prolonging their residence time and enhancing the probability of collisions such that a denser discharge can be maintained down to lower pressures. This greatly increases the deposition rates.

Magnetron sputtering

Due to the nature of the magnetic field, the electrons are trapped within a specific region on the target, and so this is where sputtering occurs most heavily. This leads to a distinctive ‘racetrack’ region where the target is worn down much faster.

There are many possible magnetron geometries, but the rectangular planar one shown above is very common. Cylindrical cathodes allow uniform erosion of the target surface, improving material utilisation.

PreviousNext