Point of View DUV Optics Contamination Redux

Point of View DUV Optics Contamination Redux


I was intrigued by the "Reader Response: DUV Lithography Contamination Issues," which appeared in A2C2's April 2003 issue. Your reader's comments highlight an important issue in deep UV lithography: molecular contamination measurement and control is becoming increasingly critical, especially as they relate to the risks of optics contamination in shorter-wavelength lithography. I applaud your reader's insight, but would like to provide a few clarifications regarding his comments.

While it is true that lithographers have long been aware of organic contamination issues, it is only now that we begin to see the effects of organic contamination on deep UV (DUV) exposure tool optics. This is an issue precisely because of wavelength changes, not "regardless of wavelength," as your reader states.

Put very simply, the further you go down the electromagnetic spectrum, the greater the risk of optics contamination. Optics contamination happens when high-energy photons interact with organic vapors-a process known as radical polymerization. In severe cases, a damaging polymer layer may be formed on the optical surface.

Now, heavyweight organics like DOP or oils will contaminate optics regardless of the wavelength, but these organics are easily filtered out. The challenge begins with filtering out C6-C12 organics for long periods. These lighter-weight organics do not begin to contaminate optics unless photochemistry is involved.

For example, i-line (365nm wavelength) light is energetic enough to break down only a few iodated components, which are not commonly found in cleanroom air. But 248nm wavelength light reacts with most halogenated organics, and so we begin seeing optics contamination concerns in this generation. And, when we adopt 193nm lithography, the ArF light reacts very efficiently with a wide range of airborne molecular organic contaminants, which means the need to safeguard optics becomes critical.

This is why suppliers of DUV exposure tools begin paying attention to optics contamination concerns with 248nm systems, and why optics contamination becomes a dominating concern for 193nm systems.

Further, optics contamination was not the concern that led IBM to adopt gas-phase filtration in the early 1990s. Rather, as IBM's own paper(1) indicates, they were concerned about the effects of basic organic contaminants (such as amines and NMP) on newly developed chemically amplified resists.

For over a decade, non-basic organics filtration was not required or specified for 248nm tool enclosure filtration. Photo-induced organic contamination was rare. Most external optics contamination was caused by inorganic ionic compounds (acid-base reactions causing haze formation(2).

I'd also like to provide another clarification: it is not true that any filter can be placed in a serial flow design. Here's why:

Lithography tool manufactures have a strict specification for filtration system pressure drop. High-pressure drop leads to increased energy consumption, excessive vibration, noise, etc. Traditional carbontrays have too much of a pressure drop (~ 1 " WC) to be used in an in-series configuration. Modern filtration elements, made of pleated or corrugated composite materials, have much lower pressure drop (~0.1 WC) and can be arranged in-series easily (at the same or higher total load of adsorptive material).

Considering this, an in-series (and in-spec) filter design, can give fabs the ability to reduce the frequency of filter changes while providing the ability to monitor filter degradation, thereby eliminating the risk of breakthrough due to unexpected contamination events. This is especially important as we ramp 193nm exposure tools, where organics contamination can pose serious threats to optics performance and lifetimes.

Your reader did make one salient point which bears repeating: "contamination concerns at 193nm are far from solved." In fact, the industry is working to understand the full scope of contamination concerns for this wavelength. The development of contamination control roadmaps and close customer partnerships are helping us learn more about the specific contamination challenges in today's fabs. Moreover, the ramp of 193nm lithography presents our greatest opportunity to identify and properly characterize many of the pollutants the threaten 193nm optical elements and processes. The rate at which these contaminants and new sources are identified, measured and effectively controlled will accelerate with the accumulation of good data.

1 Proc.Soc.Photo-Opt. Instr. Eng.,1466, pp. 2-12, 1991, S. MacDonald et al.

2 "Targetting Gaseous Contaminants in Wafer Fabs: Fugitive Amines" Pennwell Publishing, Cleanrooms Magazine 1993, D, Kinkead and J. Higley.