Senior Applications Engineer Toby discusses the manufacturing technique of laser interference lithography and how Skylark Lasers can provide optimal results when using this application.
At Skylark Lasers, we specialise in the development of high-power UV single frequency, diode-pumped, solid-state lasers.
Our patented laser architecture allows us to provide unrivalled laser performance to enable the development of the next generation of laser interference lithography set-ups.
Interference lithography is most commonly used for semiconductor manufacturing. However, it can also be used to produce other nanostructures such as optical gratings. Interference lithography is much easier to set up than e-beam lithography.
Lloyd's Interferometry
Lloyd’s interferometry is a common maskless technique and is much a easier set-up than other forms of laser interference lithography. As you can see from the diagram below, the laser beam is expanded before it reaches the mirror and photoresist-coated sample.
The pinhole is used for spatial filtering and an interference pattern is created in the sample by the direct illumination beam and the reflected beam. This produces structured lines of well-defined spacings. In any form of interferometry, coherence length is vitally important.
Narrow linewidth
Our UV lasers have a narrow linewidth of less than 500 kHz, therefore the coherence length will be more than long enough for any set-up. Typically, with Lloyd’s interferometry, the exposure times can be anything from 20 seconds to 10 minutes, and this emphasises the need for wavelength and power stability over this time.
Pointing stability is also paramount as any angular misalignment will lose the interference pattern or will create blurring of the structured line pattern that is created. To ensure that there is a uniform exposure of the photoresist, the gaussian beam is typically expanded to 5x the size of the photoresist.
The small feature size that can be produced is called the critical dimension and is related to the period P. For Lloyd’s interferometric set-up, P is determined by:
Error is a combination of properties of the photoresist, the laser linewidth, the wavelength stability and divergence or changes of the laser beam.
These structures can be obtained from lack of wavelength and pointing stability. In addition, if the laser has a poor coherence length, this can affect the interference pattern that is produced.
In addition to these laser parameters, it is vitally important to have a high level of thermomechanical stability for the whole set-up to allow for reproducible nanostructures. Also, people can experience poor modulation depth after etching because of poor reflectivity of the mirror.
Feature sizes between 160 - 800 nm with Skylark lasers
Feature sizes of anywhere between 160 and 800 nm can theoretically be achieved using our 320 nm NX laser. For a large range of photoresists available around 365 nm, this is because they were all developed using a mercury lamp transition, therefore the 349 nm NX laser is well-suited for these. In addition, the high power of up to 250 mW of our 349 and 320 NX systems allow for large surfaces to be exposed by beam expansion.
Large surfaces can be exposed using other techniques such as Mach-Zehnder interferometry.
As you can see from the graphs below, from a 320 NX system, we’ve achieved a power stability of around 1% over a 25-hour period. The power stability is imperative to ensure accurate illumination times.
This wavelength stability will produce exceptional well-defined gratings on your substrate.
In addition, the spatial mode properties of our lasers will allow for effective reimaging of the beam to ensure a uniform illumination.
Comments