There exists a wide range of applications that makes use of thin coatings of materials to alter the appearance or functionality of a surface. Some common examples are:

1. Coating tool bits to make them last longer.

2. Decorative coatings (“gold” color on faucets or door handles is often zirconium nitride).

3. CD’s and DVD’s. Some DVD’s are actually made from two discs glued together. One disc is partially transmitting so light can pass through it to read the other – a very challenging application.

4. Antireflection coatings for eye glasses.

5. Low-E coatings for windows.

To form these thin films, a form of physical vapor deposition (PVD) known as sputtering is often used. As shown in Figure 1, the goal is to remove material from a target, for example silicon, and have it deposit on a substrate (for example a piece of glass). At the same time a reactive gas can be added, such as nitrogen, which reacts with the deposited silicon forming silicon nitride. In sputtering, high energy particles (ionized Ar) strike the surface of the target material. When they strike the target, these high energy particles can be (1) reflected from the surface of the target, (2) penetrate into the target surface and disrupt the target material atoms, (3) free electrons from the target (called secondary electrons), or (4) eject material from the target. The ejected material becomes the sputtered flux that deposits on the glass.

In order to sputter a material there are several requirements:

1) A vacuum enclosure – low pressures are necessary to generate a plasma, allow effective transport of the sputtered material, and ensure purity of the film created.

2) A means of delivering power to the target – we need to pass electricity through the target to sustain the plasma and create the high-energy particles for sputtering.

3) A way of cooling the target – most of the power going through the target will be converted to heat. The target must therefore be cooled or it will be damaged.

4) A way of supplying pure gases – in order to sputter we need a non-reactive sputter gas (for example argon) and for some processes a reactive gas (nitrogen or oxygen). We must be able to supply this gas in a very controlled precise way.

The simplest sputter configuration is a planar cathode, which consists of a flat piece of target material in contact with a water cooled body. The water cooled body is supported on insulating blocks that electrically isolate the cathode from the coating system. Power is delivered through a direct connection to the cathode body (often the water line). An obvious concern is the water being in direct contact with the electrical power. If the water is conductive, power will be transmitted from the cathode to the water manifold behind the system. This is unacceptable for both safety and process control. By using treated, low electrical conductivity water this is prevented – the low conductivity prevents the passage of power. To increase the performance of the cathode, a magnet array is also present in the cathode. This confines the electrons within the magnetic field, increasing the density of the plasma thereby increasing the rate at which the target can be sputtered. The electrons are confined to a region of the magnet field that defines a racetrack over the surface of the target.

To “start” a cathode we first flow gas into the chamber and apply power to the cathode. Initially no plasma is present, but when the power is applied the cathode will have a very large negative voltage on the target. This negative voltage repels and accelerates any electrons, which would naturally be present. These accelerating electrons collide with atoms in the gas ionizing them and freeing other electrons. This new electron will also be repelled by the voltage and can collide with and ionize another gas atom. The result is an avalanche effect, electrically breaking down the gas and generating a plasma. The plasma provides the charged particles (ions and electrons) necessary to allow electric current to flow. At this point the voltage on the cathode drops and current begins to flow. The voltage on the target determines the amount of energy the ions have when the reach the target and the current is a measure of how many ions are striking the surface. Figure 2 shows plasma over a magnetron cathode surface. The magnetic fields define the region of electron and therefore plasma confinement. This region is where erosion occurs.

There are several issues or limitations with planar magnetron cathodes, both related to the magnetic profile. Since the electrons are confined to a racetrack region over the target surface only this region is effectively sputtered. The target erosion is very non-uniform and develops a “V” shape. This limits the possible utilization of the target. Well designed planar magnetrons can achieve a utilization of 45% but 20-25% is more common.