Super accurate x/y table

[the_anomaly]

I want to do some simple shape e-beam lithography.  I want to build my own beam writer and I'm thinking that accelerating and focusing an electron beam will be a lot easier if I don't deflect it.  Instead of deflection, I want to move the sample table beneath the beam to draw my image.  The goal is better than 50nm resolution.


Any thoughts on building such an x-y table? I'm thinking of using friction wheels to move it to avoid backlash of gears.


I am looking at getting an SEM but setting them up for lithography is not simple and I would like a large writing area of >1cm2.

[klugesmith]

One way, not cheap, is to use picomotor leadscrews for X and Y motion.
https://www.newport.com/f/picomotor-piezo-linear-actuators

[Hydron]

The folks here play around with some high resolution motion stages (in the z direction at least) for extreme focus stacking, but 50nm is pretty small even for that: https://www.photomacrography.net/forum/index.php

[klugesmith]

How do you intend to sense positions, for closed loop motion control?

[johnf]

these are not cheap but certainly are overkill in accuracy


https://www.physikinstrumente.com/en/technology/piezoelectric-drives/piezowalk-piezo-motors/

[the_anomaly]

I knew you guys would have some good suggestions.  Looks like linear motion may not be my greatest challenge.  The Piezo linear actuators appear to meet my desired performance and for my budget they are affordable.

If I do not go with a piezo product, thinking about lead screws, for an 80 TPI lead screw each turn traverses 0.0125" or 317,500 nm.  If I count in degrees of a revolution that gives about 882 nm per degree.  If I further count in minutes of a degree, that would give 14.6 nm per minute..  I would accomplish this using gear reduction (belts/pulleys) counting motor rotation as position feedback.  Sounds good in my head at least.   

The next issue is slide travel accuracy.  If I jog the slide to the end of its travel (1cm or 1"), has it moved off axis more than 50nm.

The 50 nm resolution is negotiable to.  I've read that SEM lithography generally has a 10 nm limit of accuracy.  I thought, let me back off that level of performance but try stay within the nm range.

I really like the Physikinstruments products, particularly their slide tables.  The prices however, the U-723 XY stage is 7k USD..  Their highest precision N-565 linear stage has unidirectional tolerance of +/-50nm.  So I know it is possible but for that stage you need to submit a quote so I assume it is out of my budget.  They do mention crossed roller bearings for the slide - something to look into.

I've ordered a couple of textbooks on SEM and particle beam physics to learn more, maybe deflection is worth trying.