For 2-4″ Torus guns the maximum thickness is 1/4″, therefore a 1/8″ thick target bonded to a 1/8″ thick backing plate is the most common configuration.
Use 150Â°C for indium bonding and 250Â°C for elastomer bonding.
Magnetron sputtering depends on the presence of a magnetic field above the target’s surface. Ferromagnetic targets (Fe, Ni, Co, Permalloy, etc) of normal thickness have high magnetic permeabilities which trap the flux from the source’s normal magnet set. However, reducing the target’s thickness causes it to saturate allowing the magnetic field to penetrate. A nickel target’s permeability is such that a target thickness of 0.100 (or less) saturates. An iron target, with its much higher permeability, must be approximately 0.002 to saturate in the normal magnet set’s field. Sputtering very thin targets brings its own problems: first the power must be very low to prevent target melting and burn-through; second, thin targets tend to bend away from the cooling well’s surface exacerbating the chances of melting and burn-through; and third there may be no obvious signs when burn-through occurs and the cooling well’s surface starts to sputter. For these reasons, SAM strongly recommends the customer use a high-strength magnet set when sputtering ferromagnetic materials.
Any hot pressed, sintered, or ‘hipped’ ceramics or metals should be bonded.
Most oxide, nitride, and silicide targets crack during normal sputtering. Applying (or removing) power to these types of targets, must be done by slowly ramping up / down with occasional ‘soaks’ to allow the target to recover from thermal stresses. Bonding such targets to a suitable backing plate are highly recommended to the target, once cracked, stays together. A cracked target will still be sputtered providing plasma has no line-of-sight access through the cracks to the bonding agent or the backing plate. Bonding the target to a backing plate may or may not allow it to accept a higher power. Remember, the characteristics of the bonding agent must now be added to the heat dissipation considerations.
The Materials department does not re-melt customer material to make new products because control over the purity level cannot be maintained or certified. Instead, we offer a reclaim service in which the customer sends in a spent precious metal target and receives a credit toward the purchase of a new one.
There are many factors that influence deposition rate. The two common ways to increase rate within an existing sputtering system are to either increase the applied power to the target or reduce the throw distance. There is a linear relationship between sputtering rates and the applied target power. Doubling the applied power will double the sputtering rate. However, increasing the power to the target must be carefully planned since it is easy to damage a fragile target or melt a metallic target if the power is too high. Decreasing the source-to-substrate distance (r) is the most powerful way to influence the deposition rates because distance and rate have a power-law relationship. For a point source, the sputtering rate is a function of 1/r^2. However, decreasing the distance (r) may decrease in film thickness uniformity on the substrate. If neither ‘power’ nor ‘distance’ modification is possible in your vacuum system, contact SAM for further options. Slow deposition rates are expected when RF sputtering oxide or nitride targets. In some instances, rates can be dramatically increased by sputtering the metal target using pulsed DC power and adding O2 or N2 to the argon to produce the intended film composition. Contact SAM with any questions regarding this ‘reactive’ sputtering method.
Electrically insulating targets must be sputtered RF. Electrically insulating films, however, can be made through DC/Pulsed DC reactive sputtering. The electrical conductivity of a sputtering target must be greater than 10-7 (ohm-cm)-1 in order to sputter via a DC power supply . This is an experimentally derived quantity and not subject to much debate. 1. J. Szczyrbowski, G. Brauer, W. Dicken, M. Scherer, W. Maab, G. Teschner, and A. Zmelty, Surf. And Coating Technol., 93, 14 (1997)
Deposition materials that react with air or water vapor are protected during storage and shipment by immersion in a low viscosity (hydrocarbon) mineral oil. Before the material can be installed in a vacuum system, this oil must be completely removed using solvents that do not act as additional contamination sources for the vacuum system or the subsequent thin-film processes.
The preferred deposition methods are: RF sputtering, e-beam evaporation, or reactive pulse DC sputtering of the metal. Consult the SAM deposition table for information specific to a given material.
Roughly three-quarters of the power applied to a sputter gun ends up heating the coolant water. Understanding the heat transfer processes and heat dissipation is critical to successful sputtering. For a given sputter gun, four factors determine the maximum power density the target will accept: (1) the type of target material; (2) its thermal conductivity; (2) heat transfer from target to gun’s cooling water; and (4) the water’s flow rate. The type of material determines its brittleness, thermal expansion coefficient, fragility, etc which determines the target’s response to thermal stress. Applying a high power density to a material with a poor thermal conductivity (compared to, say Cu or Au) creates a large temperature difference between the target’s top and bottom surfaces causing thermal stress. For fragile materials (such as some ceramics and semiconductors) stress may crack the target. For stronger, ductile materials (many metals) the target may melt. The efficiency of target cooling depends on the ‘thermal resistance’ between the target’s back surface (the surface not being sputtered) to the gun’s cooling water. Directly-cooled targets, where the water flow is in contact with the target’s back surface, have low thermal resistance. By comparison, indirectly-cooled targets (which are clamped to a copper plate in the sun and cooled by water flowing across the plate’s back surface, not the target’s) have high thermal resistance. At the atomic level, the interface between target’s back surface and copper plate’s front surface has a small contact area for thermal conduction. Over the remaining area, heat transfer is by radiation. (Interestingly, the thermal resistance of such an interface under vacuum may be 4-10 higher than its thermal resistance in air.) Water flow through the sputtering gun should be sufficient to remove the total power. As a guideline 1 gal/min (~4 L/min) dissipating 4 kW power causes a rise in the water temperature of ~15C. For coolant liquids other than water, remember to factor in the liquid’s specific heat. Suggested guidelines for calculating maximum power are: for a high thermal conductivity, directly-cooled, metal target such as Al, the maximum power density is ~250 W/in^2 for the target’s (front-side) surface area. For other types of (directly-cooled) targets, de-rate the maximum power density based on the target materials thermal conductivity compared to aluminum. As an example, for a target with a thermal conductivity 1/10th that of Al, the maximum power density should be 25 W/in^2. For a high thermal conductivity, indirectly-cooled, metal target such as Al, the power density limit is ~100 W/in^2 for the target’s (front-side) surface area. Again, de-rate this power density for materials with lower thermal conductivities. Two important final points: (1) To prevent target cracking, fragile targets may need to have maximum power density de-rated even further than indicated by the relative thermal conductivities; and (2) some target materials crack no matter what power density is used.
The difference in color of the TiO2 is due to the degree of oxidation. Fully oxidized TiO2 is white. With even a slight reduction, the material becomes grey to black in color. When evaporating TiO2 it is necessary to melt the material before depositing films. Titanium has several stable oxides (TiO, TiO2, Ti2O3, Ti3O5) and when the material is heated it evolves oxygen and turns black. Regardless of which material you use to start (white or black) the material will turn black and evolve oxygen when it is heated up. Once outgassing subsides the films can then be deposited. It is necessary to add 1 x 10^-4 Torr to 2 x 10^-4 Torr of O2 gas during the deposition to maintain stoichiometry of the film.