Electroplating
Alexander E. Braun -- Semiconductor International, 7/1/2003
Electroplating, probably the earliest electrochemical process carried out by mankind, has not changed much since its beginnings. According to established orthodoxy, Alessandro Volta invented the first electric battery, or "voltaic pile," in 1800, opening the way to electrochemical processes. However, that may not be the case.
In 1938, German archaeologist Wilhelm Konig described a clay jar found in the region of Khujut Rabu, near Baghdad, Iraq. Dated as belonging to Mesopotamia's Parthian period, roughly 2000 years ago, it lay unnoticed for years in the Baghdad National Museum. Konig hypothesized that it could be a battery. After World War II, Willard F.M. Gray of the General Electric High Voltage Laboratory (Pittsfield, Mass.) carried out experiments, using duplicates of the Baghdad jar. He determined that, when filled with an electrolyte like grape juice, the device produced about 2 V. This battery, manufactured during a time when Roman Legions guarded Caesar Augustus' stupendous empire, may have been in daily use to electroplate silver with gold utilizing essentially the same process employed today to plate chrome onto car bumpers, or for inexpensive jewelry.
Which brings us to today. For many years, aluminum has been the conductor of choice for device manufacturing because it has the characteristic of adhering well to silicon and SiO2. As shrinks have raced after Moore's Law, however, they have produced progressively more crowded architectures, with currents coursing more slowly though increasingly thinner interconnects. Electrodeposited copper became the logical alternative to aluminum because it is a superior conductor. The electrical resistance of copper interconnects is less than two-thirds that of its tungsten-aluminum counterparts — the series resistance of copper via can run as low as 20% that of tungsten plugs.
When current is passed through the surface to the electrolyte, the metal ions react with the electrons and form metal (Figure ). In the case of a silicon wafer (in this case the cathode), the wafer's surface is electrically connected to an external power source and current is transferred to the surface into the solution. This results in a combination of the metal ions (Mz+) with electrons (e-) to form metal (M):
Mz+ + ze- → M(0)
For copper deposition, the wafer is coated with a conductive seed layer of copper and immersed in a solution containing copper ions. The electrons are then driven through the wafer, producing the reaction at the wafer's surface:
Cu2+ + 2e- → Cu (0)
The water accepts copper ions from the solution. To complete the circuit, another conductive object must accept electrons from the ionic solution, and release positively charged copper ions. In this case, a solid copper anode provides the solution with the necessary ions, completing the circuit. The oxidation reaction occurring at the anode balances the current flow at the cathode, maintaining the solution's electrical neutrality. All the copper ions removed from solution at the wafer cathode are replaced by dissolution from the copper anode.
Because, as described by Faraday's Law of Electrolysis, the current delivered to a conductive surface during electroplating is directly proportional to the quantity of metal deposited, the layer deposited is easily and precisely controlled through variations of plating current and time. Electroplating can be done using a constant current, a constant voltage, or variable waveforms of current or voltage. A constant current is the easiest way to obtain accurate control of the amount of deposited metal. Plating at a constant voltage and using variable waveforms requires more complex equipment and control, but can be useful in tailoring specific thickness distributions and film properties.
Editor's note: This information was compiled with the help of the paper, "Fundamentals of Electroplating," written by Jonathan Reid of Novellus Systems.
