|   |   |   |

To receive the next issue by e-mail, visit My Account and confirm you have opted-in to receive e-mail from AE.

Evaluating Power Supply Quality

Determining the right power supply for your process involves two important choices: process power method and product supplier. You may have chosen the ideal method, and even set it up optimally, but a unit that is prone to failure under real-world conditions can cancel out any potential benefits of these efforts. Therefore, it is critical to choose a reputable, high-quality product.

What Happens When the Gas Turns Off?

If your power supply fails, your process comes to a standstill. A high-quality power supply protects your process throughput and capital investment. It stands up to harsh process demands and even protects your equipment from the potentially devastating effects of maintenance slips, such as failure to replace the gas bottle.

Six Essential Power Supply Performance Tests

If the gas runs out, will your power supply self-destruct? To determine the quality of your existing power supply, or to help choose a new one, consider the following tests. They pose extreme, yet realistic conditions that eliminate inferior units from consideration.

Please note that this procedure may cause power supplies of inadequate quality to self-destruct. If you are unsure of your power supply’s adequacy under these conditions, or if you are considering a new power supply for purchase, you may want to consider the questions below theoretically only. However, all AE power supplies are designed and tested to stand up to the harsh conditions described here.

Test 1

Run your power supply at its full-rated output. Abruptly shut off the gas supply. This replicates real-world conditions in which someone has failed to refill or replace the gas bottle. How does the power supply respond? Does it continue to function properly? Does it simply “go to” high-voltage, open-circuit condition? Or, does it break (“breaking” may involve smoke, flame, or other obvious display of malfunction)?

If your power supply is still fully functional, continue to the next test.

Test 2

Turn the gas back on. Even though it withstood test 1, does your power supply now break, or does it resume normal operations?

If your power supply is still fully functional, continue to the next test.

Test 3

After subjecting your power supply to tests 1 and 2, run it in full metal mode (high voltage, low current).

If your power supply is still fully functional, continue to the next test.

Test 4

Slowly decrease the pressure into the mid-4 Torr range. Your power supply should withstand these gas-starvation conditions.

If your power supply is still fully functional, continue to the next test.

Test 5

To test your power supply’s ability to handle fast, extreme impedance changes (see hysteresis curve below), begin by running it with full argon, in full metal mode (high voltage, low current). Quickly switch to full oxygen in full reactive mode, at full-rated power. Next, switch back to full argon at full-rated power.

Hysteresis curve

If your power supply is still fully functional, continue on to the next test.

Test 6

If your power supply passed tests 1 to 5, observe it under extreme arcing conditions. Does it quickly identify and extinguish the arcs before they damage film quality?

Conclusion

If your power supply doesn’t pass all of these tests, it is not of adequate quality to protect your process throughput and capital investment. To find a power supply that is designed and tested to withstand these harsh conditions, contact AE. All AE power supplies are constructed to maintain performance under real-world conditions, including those described here. And, advanced protection circuitry shuts down the unit, if needed, to help prevent power-supply equipment damage.

Don’t Get Called at 2 A.M.: Troubleshooting the Most Common Process Problems

Sudden plasma loss, the unexpected appearance of pinholes, and a rapid decline in general film quality are the three main reasons a process engineer might be called at 2 A.M. The following section offers common causes of these phenomena and proposes solutions that will hopefully help get your process back on track before a phone call becomes necessary.

Sudden Plasma Loss

Your choice of process power method is very often at the root of sudden plasma loss. When you use a method that does not periodically reverse voltage, such as straight DC in reactive mode, you will likely experience dielectric film buildup on the anode. When this buildup occurs, electrons can’t pass from the cathode to the anode. Impedance then increases, which leads to an increase in voltage. These conditions intensify arcing, which, without intervention, will grow until you lose your plasma and the process breaks down completely.

How do you prevent this arcing “spiral of death” in your process? Consider replacing your power method with AC or pulsed DC. See our process power selection matrix from our Q1 2007 Sputter Spotlight newsletter for more information on the decision-making process.

Other possible causes of plasma loss include weak magnets, poor target materials, and, simply, loss of power. Contact Doug for more information about other possibilities and remedies regarding sudden plasma loss.

Pinholes

The basic cause of pinholes is excessive arc energy. If you are experiencing pinholes, either your power supply can’t maintain low stored energy during arc events, or your arc handling setup is incorrect.

The ability to maintain low stored energy is inherent to the power supply itself, so choose one at the outset that stores very little energy. AE power supplies are leaders in the process power industry in terms of their ability to maintain extremely low levels of stored energy. For example, the Crystal AC power supply stores as little as 1 mJ per kW.

Certain power supplies on the market make arc handling setup so complicated that it may be preferable to simply switch this function off, instead of wasting time trying to figure it out or risking setting it up incorrectly. Choose a power supply that allows you to easily understand and set its arc-handling parameters. AE power supplies enable you to set detect time, off time, and trip level with no hassle.

Your choice of process power method also affects the occurrence of pinholes. For applications in which pinhole prevention is critical, AC and pulsed DC may be excellent solutions. See our process power selection matrix from our Q1 2007 Sputter Spotlight newsletter for more information on the decision-making process.

Overall Poor Film Quality

Sudden-onset problems with overall film quality are most often caused by either old targets, faulty gas and/or pressure gauges, or vacuum chamber leaks.

When planar targets near the ends of their useful lives, they tend to start to cross-sputter across their racetrack grooves. This causes molecular collisions in the plasma, which can increase arcing. These collisions can also lead to re-deposition on the opposite side of the racetrack, which further increases arcing. Arcing continues to escalate, eventually completely overwhelming and shutting down the process. The solution here is to replace your targets.

Racetrack erosion pattern

Gas gauge malfunctions cause errors in the gas mix, and if your pressure gauge fails, pressure may increase above desirable levels. As pressure increases, the deposited film becomes more porous because there are more molecular collisions in the plasma. In either case, to get your film quality back on track, replace your gas and/or pressure gauge.

Vacuum chamber leaks are also a common culprit when it comes to film quality problems, so check the integrity of your chamber.

Ask Doug!

Doug Pelleymounter, AE's senior field application engineer, has more than 32 years, or 224 dog-years, of hands-on experience working with all kinds of challenging sputtering applications. In this column, Doug helps you answer some of your difficult application questions. Submit your question or comment to sputtering@aei.com. 

1. What is the difference in utilization between planar and rotatable targets?
2. How do the different erosion patterns of planar and rotatable targets affect the process over the course of the target's lifetime?
3. You’ve mentioned before that generally, pulsed-DC and AC power produce better films than straight DC. What is the actual difference in film quality?
4. How can I optimize my sputter rate?
5. I’ve heard that an upcoming technology called HPPMS produces extremely flat, uniform films, but is not yet widely available. Are there any alternatives that produce similar results using readily available equipment? 

1. What is the difference in utilization between planar and rotatable targets?
   Answer: You typically get about 35% utilization out of a planar target and about 85% out of a rotatable. These numbers are independent of the process power method or target material you are using. Please note, however, that rotatable cathodes are not compatible with RF power. Generally, they are best used in AC, DC, or pulsed-DC powered processes.
   
   
   
2. How do the different erosion patterns of planar and rotatable targets affect the process over the course of the target’s lifetime?
   Answer: Although planar and rotatable targets do erode differently, there is actually little difference in how you should handle your process power over the course of the target’s lifetime. For rotatables, target thickness decreases in a uniform manner, causing the magnets to move closer to the target surface. This results in an increase in current and a decrease in voltage. Although planar targets erode in a non-uniform manner, you will see a decrease in current and an increase in voltage.
   
   
   
3. You’ve mentioned before that generally, pulsed-DC and AC power produce better films than straight DC. What is the actual difference in film quality?
   Answer: The photos below show that the difference in film quality is quite significant.
   
   
   
   
   

   Film quality produced by straight-DC power (above) vs. pulsed-DC power (below)
   Source: Centre for Advanced Materials and Surface Engineering, University of Salford, U.K.

   
   
   
   
   
   
   
   
   
   

   Film quality produced by straight-DC power (above) vs. AC power (below)

   
   
   
   
4. How can I optimize my sputter rate?
   Answer: This is a great follow up to the previous Ask Doug, where I answered the question, “What sputtering rates can I achieve?”
   
   The general rule of thumb is that the lower the pressure, the better your rate and film quality because there are fewer molecular collisions in the plasma and a longer plasma throw distance (the ability of the sputtered particles to reach the substrate from the target). So, sputter at the lowest pressure you can, but of course, avoid going into gas starvation conditions, which can cause problems for your power supply. (See Evaluating Power Supply Quality above.)
   
   The second thing you can do is use a gauss meter to check the balance of your magnetrons. Unbalanced magnetrons widen plasma throw distance and create excess electrons, which affects substrate heat and film quality. Balanced magnetrons focus throw distance, which helps your sputter rate especially when the distance is great between the cathode and substrate.
   
   
   
   
   
   Unbalanced magnetrons (above) vs. balanced magnetrons (below)
   
   
   
   
   
   
   Third, check the strength of your magnets. As you increase magnet strength, your plasma throw distance increases. One thing to note is that although this results in greater sputter rate and film packing density, stronger magnets deepen the groove in your target, which decreases utilization.
   
   All of that said, sputter rate is a complex and multi-faceted issue. Please feel free to contact me at sputtering@aei.com for advice on your specific situation.
   
   
5. I’ve heard that an upcoming technology called HPPMS produces extremely flat, uniform films, but is not yet widely available. Are there any alternatives that produce similar results using readily available equipment?
   Answer: You’re in luck! There is an existing and easily accessible technique that will give you a similar degree of flatness as HPPMS (high-power, pulsed magnetron sputtering) technology. This technique combines two process power methods: RF and pulsed DC. An added bonus is that, although it is a relatively new method itself, RF-superimposed pulsed DC has been around long enough that a certain amount of information has been developed and made available about it. Please see AE’s power supply selection matrix, as well as our Arc Handling in RF-Superimposed DC Processes application note for details on this method.