Using OFM for making a probe station

General
1

Hello,

I am a physics PhD student from India working in the field of quantum transport, so we generally fabricate nanoscale devices on small wafers and study quantum phenomena essentially with voltages and currents. Most of the time we measure our devices in cryogenic environments at 4 K and all the way down to a few millikelvins.

I got into 3D printing only a few months ago and I recently came across this incredible project when I was looking to build a simple 3D printed translation stage (of which I didn’t expect much in terms of precision or accuracy). There are so many things in my lab for which we would find it really useful to have a precise stage, and more than building a microscope, I am more interested in incorporating the mechanism of the stage to build something similar to a wafer probe station that can be used to check our devices at room temperature before we load them in our cryostats for long measurements. As I understand it from a couple of posts on this forum, the block stage can be used for making a micromanipulator. For a basic probe station, we would need at least 4 probes which can move independently in X, Y and Z. So is it feasible to have 4 such block stages, each holding a probe that can be adjusted independently? What are the possible challenges I would encounter if I were to attempt to do this? Such a probe station would also require a long working distance microscope with relatively lower magnification, but I guess that can be arranged.

Apart from a probe station, I also would like to build a precise XYZ stage for our samples with at least 20mm translation in XY, but in this case I would also like to know exactly how much (in microns) I am moving my sample in any given direction since this information is needed for an experiment I am working on. Right now, I am using a manual XYZ translation stage with micrometers that has a least count of 10 microns. If I could potentially get even better accuracy with an OFM based stage on top of being motorized, and without breaking the bank on commercially available stages, I would be so happy. While our lab has expensive cryogenic equipment and fab facilities to do our experiments, we just don’t have the funds to invest on these other things unfortunately. As it is, I am thinking of building an OFM microscope just to get a feel for how it works and in what other ways I can incorporate it into my research. We do need more microscopes in our lab as the one and only Olympus microscope does get crowded sometimes. But our devices are opaque (mostly silicon based) and so only reflection based microscopy will work and it also needs to be upright. Any suggestions on what would be the best OFM configuration to build for this?

Apologies for the long post and if I have asked too many questions. I’m just really curious to try this out :slight_smile:

3 Likes
2

Hi. These are all good questions and you have a very interesting application.

The block stage should work well as a component of a probe station. The overall size is quite big, but the moving stage is in one corner, so four could surround a central probe area quite well. More than four would need more thought, but it would be easy to build a couple of stages and play with different arrangements. The range of motion is much less than the microscope stages, only about 2x2x2mm. If you just use the knobs to turn the actuators manually at first that gives a good feel for the motion that you can get. Motors add complexity, and they do move quite slowly, so doing it by hand is often good at first.

For a low magnification, there is a prototype Field Dissection Microscope that you could try, which is also discussed on the forum in First prototype of the Field Dissection Microscope is ready to build.

For x-y-z sample motion the Delta Stage is probably the best existing option. The range of travel is not as large as you would like - it is about 12x12mm. There is some discussion of modifying the stage in the thread “Substage for 2-photon microscopes”. That is about making the stage wider, which should be simply changing the parameter for the stage size, but caused some problems. For a larger range of motion you would want to increase the stage height, which may or may not work by simply changing the height parameter. This gives longer levers and more motion for the same maximum angle of the flexures (in x and y, but I think will not change the z range). It reduces the incremental motion and makes the stage less stable, so there is a limit to how far it would be sensible to go.

All of the microscope versions - the standard (inverted), the upright, or the delta stage can use reflection modules. There are not currently instructions for assembly of the reflection optics module in the standard microscope instructions. The correct STLs are available in the customisations and alternatives tab, but for assembly you would need to look at the Delta Stage instructions. Note that the optics modules for the Microscope and the Delta Stage are very similar, but the camera is rotated 45-degrees relative to each other, so you do need the modules from the right part of the project.

Overall you have a good plan to assemble a standard microscope first, and I would suggest you do that with transmission illumination. This has seen the most development in both development of the hardware and the instructions, so should be the simplest way in. You can build it upright or inverted, and in either case you can use high resolution (with an objective lens) or low cost optics modules.

The build techniques and much of the non-printed hardware are the same for the Block Stage and the Delta Stage, but those two projects are in a much earlier stage of development so it can help to have some experience of the way Openflexure stages go together.

2 Likes
3

Thanks a lot for your input! I think starting with a couple of block stages is a good idea, although I’m definitely a bit concerned about the total range of movement with the block stage. Our wafers are quite small in general (less than 5x5 mm^2) and the 2mm range should be fine technically, but it would mean that the block stages have to be aligned with respect to the wafer so that the probes don’t go out of range. The field dissection microscope looks cool, I didn’t know something like that was also being developed! I’ll definitely take that into consideration.

Regarding the delta stage for keeping track of sample position in XYZ, I don’t really want a bigger stage as much as I want just a little bit more range in the translation, especially in the XY plane. I’ll definitely play around with the stage height and see where that takes me. I’m not familiar with OpenSCAD, so need to get acquainted with that first. But something which I’m still not clear about is whether the software allows me to keep track of the sample coordinates from the “origin”. Could you throw some light on that for me?

Let me get started with building the standard microscope then. I have ordered most of the parts listed in the BOM so I’m looking forward to putting it together!

4

I was also wondering whether any of these stages could be used for setting up a UV mask aligner for photolithography. Granted, the limited translation range could once again be a problem but apart from that all we need is something to hold the photomask above the sample and a collimated UV source for the exposure.

5

The Sangaboard keeps a track of the steps travelled since the zero position was set. That is visible in the microscope software, or available with a serial request. There is some backlash in the stage, some creep/drift and there may be missed steps. I don’t think that there has been a recent test of how close that absolute position is on the Block Stage. With the microscopy applicatoins, fine position tracking comes from the camera images.

1 Like
6

The internals of the Block stage are different from the other stages and I am less familiar with them. In the design intent there is a set standard platform height that it needs to match. The height is low and restricts what can fit in underneath. You do not have that standard height requirement, and I think that as with other stages taller would mean more motion.

I was not very clear. I meant it just as an example of the kind of change that should work automatically, but can throw up errors. The changes for more travel might hit similar issues, or they might work straight away.

1 Like
7

I’m fairly sure I have previously modified the block stage to get 4mm translation in X and Y, and 2mm in Z. That might work well for your application - indeed a probe station was something I always thought would be cool but never got around to making. I think if you got the design of the probe holder right, you might be able to build some coarse alignment in there, so you can get all the probes within the field of view of the microscope at least - then possibly 4x4x2mm would be just about ok.

A 20mm travel stage is something I’d love, but the flexures start getting a bit wobbly once you get much past 12mm in my (limited) experience. It’s definitely possible, I’ve just not figured out how yet!

Lastly, for keeping track of position, you do get an estimate from the number of steps on the motor. However, if you really want an accurate estimate, using the camera (i.e. tracking how far the image has moved) will be significantly more accurate, particularly with regard to mechanical backlash. I’ve used this trick in the past and it works quite well - though that code’s a decade old and for different hardware. I think the most accurate way to keep track would be to first acquire a big tiled image of your sample, then use that as a map - so you can locate the current field of view with respect to your whole sample. That doesn’t give you quite the same quantity as how far the stage has moved, but it’s often a more useful number (i.e. if there’s mechanical drift, even a perfect stage encoder will disagree with the image - but usually what you care about is “where am I in the sample” unless you’re using the stage step count for metrology - which I wouldn’t particularly recommend.

1 Like
8

I’m fairly sure I have previously modified the block stage to get 4mm translation in X and Y, and 2mm in Z.

Oh, that’s cool! I’ll give it a try, 4mm translation should definitely be sufficient if we build some coarse alignment on top of it like you said.

A 20mm travel stage is something I’d love, but the flexures start getting a bit wobbly once you get much past 12mm in my (limited) experience.

I see. Then I shall first build the standard microscope stage and then modify it to get maybe around 15mm which I guess is plausible without making it wobbly. Or perhaps is there another material other than PLA which can be used to get more flexure without hitting the elastic limit? What about something like polypropylene, which is quite flexible but also strong? I’ve read that it can be tricky to print without warping though.

However, if you really want an accurate estimate, using the camera (i.e. tracking how far the image has moved) will be significantly more accurate, particularly with regard to mechanical backlash.

I’ll consider it but tracking movement like this would be a bit cumbersome for what I need it for. I would basically be measuring the sample’s RF response at each step along the X and Y axes, so I would need the X and Y coordinates of each step of the sample’s movement.

Thanks again, @r.w.bowman and @WilliamW. Honestly, I’m grateful for your quick and helpful advice and also for this project!