This month part 3 of the series: "MEMS Stepper & Scanner Metrology", the final part of this series probably the most challenging for the lithographer: Front to Back Alignment.

MEMS Metrology Part 3: Front To Back Alignment - the final frontier

Front to Back Alignment (FTBA) lithography has evolved over the past 10 years from contact printers utilizing infrared (IR) alignment and dual sided optical microscopes, to the current state of the art, ASML's “3D align” system.

Historically, FTBA overlay specifications were typically 2 to 3 micron, which defined the “outer limits” for device designers. This has all changed with the introduction of ASML’s “3D align” option, a stepper/scanner based FTBA system that pushes beyond these limits by an order of magnitude. This new capability opens up opportunities to designers, who can now develop high volume, high density, MEMS products that were previously only possible in small R&D quantities. A short animation describing how the ASML “3D align” works is available at this link.

So, what are the challenges? The main challenge, besides keeping the front side of the wafer protected whilst printing patterns on the back, is the verification of FTBA overlay. In the past, optical verniers visible in IR, were used and although error prone were deemed adequate for 2-3 micron design rules. Unfortunately, this is not the case when FTBA overlay performances pushes below 500nm, especially when one considers measuring this over a 600 micron wafer thickness. Researchers have experimented with electrical test methods, but these have proved unsuitable for a production environment. Etching alignment targets through the wafer has also proved not to be error free due to asymmetry effects. So how does one measure the error?

The metrology building blocks discussed in the previous articles suggest that self metrology should be possible. This is the subject of a paper by the Delft University. The paper explores the use of glass wafers, chrome omnimarkers, and rotating the wafers 180 degrees to solve the alignment beam angle and refractive index errors that confound the metrology.

For more discussion about the “final frontier” and other possible solutions, please This e-mail address is being protected from spambots. You need JavaScript enabled to view it.