On-site supply of fluorine is the most logical choice of cleaning gas for the electronics industry in future, an alternative intensified by the ongoing financial crisis.

Furthermore, it’s not inconceivable to expect to see a growing number of on-site facilities emerging as a result.

Those are the sentiments of Dean O’Connor, Head of Solar Business for Linde Gases Division, as he exclusively spoke to gasworld about the functional role of fluorine (F2) in the TFT-LCD industry.

Back in July, The Linde Group lauded the development of its on-site generation of F2 technology, as an answer to concerns over the environmental damage caused by the use of nitrogen trifluoride (NF3) in electronics production.

The news came after reports had highlighted the negative effects of rapidly increasing NF3 consumption, a stable cleaning gas with a global warming potential (GWP) of 17,000.

Linde now has its on-site fluorine generators installed at more than 20 semiconductor, LCD, and solar cell sites, with O’Connor describing the technology as a ‘fundamentally better’ solution.

So why is this method so significant?
Consumption of NF3 has increased dramatically over the past decade, due in part to the emergence of the TFT-LCD industry.

In the LCD process, the deposition of a silicon film onto a glass substrate also results in a hefty deposit of the film on to the walls of the chamber - requiring a cleaning agent such as NF3.

“In the case of the LCD tool and solar tools,” O’Connor told us, “You’re ending up with quite a lot of silicon deposit on the walls. You’re using quite a lot of cleaning gas to remove it. That explains why consumption is growing so rapidly.”

“The argument for why fluorine is effective from an environmental perspective is very simple: NF3 has a global warming potential of seventeen thousand, fluorine has a global warming potential of zero.”

O’Connor adds, “For fluorine really, it’s a true engineering design solution because it designs the problem out at its most fundamental level. You can mitigate something with a high global warming potential but if your alternative is a zero global warming potential, it is fundamentally better.”

Cost effective
As well as the obvious environmental merits, the use of F2 compared with NF3 offers a number of other performance and cost benefits, which take on added significance as the credit crunch bites.

NF3 itself both begins as F2 gas and ends as such during the TFT-LCD production process. Cutting out the middle man so to speak, therefore makes perfect sense logically and financially. Theoretically, costs should lower.

The energy wasted breaking NF3 down into F2 is a major production factor, with the surplus energy saved using solely F2 enabling a much more efficient process. Activating all the F2 (up to four or five times as much as NF3) in the chamber through the pioneering Linde technology, allows for a much quicker cleaning rate.

O’Connor exclusively told gasworld, “For no extra cost, you can roughly half the cleaning time. It makes sense in many cases to pay that bit extra for even more fluorine and reduce the cleaning time even further. The value of the output is far greater than the cost of the gas.”

“In the case of an LCD fab, especially the very largest scale LCD fabs, the data that we’re getting from some customershearing in the field suggests that somewhere between 5-7% extra throughput may be possible at a fab as a consequence of being able to clean the CVD chambers in half the time.”

Cost and efficiency implications have perhaps never been more relevant than at present, as the financial climate continues to leverage pressure on business operations.

So gasworld asks, is the credit crunch likely to accelerate the adoption of on-site F2 generation in electronics production?

“It’s an interesting point,” O’Connor reflects. “I would say that actually the argument existed even before the current environment. It’s probably going to be exacerbated by it because there’s no doubt that higher interest rates are going to put pressure on the cost of solar cells, in order to maintain the economics of people who are installing them. Higher commercial interest rates would appear to be an inevitable a very likely consequence of the current climate.”

“The reality is that even before this happened there was a tremendous downward pressure, the goal of the entire PV industry is about driving down the cost per watt. There’s no doubt that the financial climate will accelerate the whole process.”

End of the road for NF3?
With the arguments stacking up in such favour of F2, it seems almost inevitable that this method of chamber cleaning will become the standard in large volume LCD and thin film PV processes. Indeed, gasworld understands that the majority of equipment manufacturers have already qualified it for LCD and may not be too far away from making the switch to F2 for solar.

A widespread breakthrough seems relatively imminent.

Which surely spells the end for NF3? Not quite. Not even close in fact.

According to O’Connor, NF3 will maintain a healthy role within the semiconductor and wider electronics industry for the foreseeable future.

“NF3 is a gas that’s reasonably stable and relatively easy to handle, it has some hazardous properties but they are generally manageable hazards. When used in small volumes it’s very difficult to beat the economics. So if you need a source of fluorine in small volumes, the economics for NF3 are very effective.”

“For those well established and smaller scale applications NF3 is still a viable alternative and is still probably the best option.”

“The only reason you would contemplate a change is if the environmental factor became so overwhelming, that it didn’t make sense to use a global warming gas anymore.”