## Tuesday, October 06, 2015

### Table-top electron microscopes

A quick question in the hopes that some people in this blog's readership have direct experience:  Anyone work with a table-top scanning electron microscope (SEM) that they really like?  Any rough idea of the cost and operations challenges (e.g., having to replace tungsten filaments all the time)?  I was chatting with someone about educational opportunities that something like these would present, and I was curious about the numbers without wanting to email vendors or anything that formal.  Thanks for any information.

(Note: It would really be fun to try to develop a really low-budget SEM - the electron microscopy version of this.  On the one hand, you could imagine microfabricated field emitters and the amazing cheapness of CCDs could help.  However, the need for good vacuum and some means of beam focusing and steering makes this much more difficult.  Clearly a good undergrad design project....)

Josh Einsle said...

Hiya Doug-

We just got a Phenom here at Cambridge. It looks ok and the techs seem to enjoy playing with it. I have not had a good excuse to use it, but I would like to use it to check the quality of my coatings before other higher-res experiments. Might do that in the next week or so. I would suggest contacting Paul Midgley, as I am pretty sure he was involved in purchase of ours in some way and he can chat numbers with you.

As to your other idea, keep it ticking over in your brain. I would be interested in some part of this in a few years once I get through this round of post-docing.....3D printing would offer a lot...

Ted Sanders said...

I've talked to microscopists at Stanford interested in building an ultra cheap SEM. One of their ideas was to make the electron beam path so short that you wouldn't even need a vacuum.

Ted Sanders said...
This comment has been removed by the author.
Ted Sanders said...

Anonymous said...

I have the Hitachi TM-3000 with the Bruker EDS system. A few years ago it was around 120 all in. The EDS detector is very good and makes the instrument a true workhorse. It generates excellent EDS maps very quickly and is a lifesaver for rapid chemical analysis. Hope this helps.

Anonymous said...

Here at Cornell, Dave Muller's group has invented an air-SEM. A graphene window is transparent to electrons, so you can image samples without need for vacuum. Shoot him an email.

Unknown said...

I have been working with table top JEOL FESEM 6500F and 7000F attached with the EDS/WDS by Oxford inst since last >6yrs. I worked as an operator for it. Although, one of the model 6500F is old, it still performing well and able to get high-resolution nanometer-scale images. As far as W-fillament is concerned, it depends on your use. We are using the 6500F filament for almost 8 yrs without any problem and FYI we use it every day. For instance, powder samples with more metallic contain can easily damage filament if its not stick properly to specimen stage. Once these particles enter in FESEM gun chamber, the lifetime of the filament start decaying faster. Yearly, maintenance may involve power-backup, roughing pump, and aperture in an extreme case. I'm sorry, I don't know the cost of the instruments.

Douglas Natelson said...

Thanks, everyone. Josh, I'll take a look. Ted, I had also seen some papers out there looking at MEMS-style fab to do the electron emitter and optics about 20 years ago, but it doesn't seem to have gone anywhere. Seems like some microfab field emitter + electrostatic lenses + piezo-driven rastering + integrated detection could actually be done relatively cheaply if mass-produced.

Anon@8:38, the graphene window is cool, though that still uses a conventional electron optics column as far as I know. Maybe mating this idea with a microfabbed emission/electron optics setup could be done, though it would be fragile.

Unknown, thanks, but neither the 6500F or 7000F are "table-top" - they have full, stand-alone vacuum systems, full-height (1 m or so) high voltage columns, UHV gun chambers, etc. I really mean some kind of smaller system like the Phenom mentioned by Josh up above.

Anonymous said...

Dear Prof. Natelson, this is unrelated but on your blog, you mentioned that the NNIN was not being renewed a year ago. It seems that the NSF finally got their act together and ended up giving out grants.

http://www.nsf.gov/news/news_summ.jsp?cntn_id=136211

I was wondering what your thoughts were.

Amy Blum said...

Hi Doug!

We just got a Delong LVEM5. It does SEM, but we mostly use it as a TEM. It is really great for pre-screening samples before using better EMs, quality control, and routine checks, especially for mid-process testing in near real time. We get publication quality images for some but not all samples. I pay a small fortune in EM user fees (as do many nano people here), so this is a toy that will pay for itself in about 2 years in saved user fees for the groups that went in on it.

Before buying, we checked with a group at NRC that has one, and they say the filament lasts 2-3 years easy with regular use. It is pretty idiot proof compared to a standard EM since the vacuum system is so much smaller and less complicated. We have a technician who trains people on it (she also maintains 3 AFMs), and so far the maintenance has been minimal. It is priced in Euro, so your dollars would go a lot further than ours did! Mail me if you want more specifics.

Douglas Natelson said...

Anon@4:26, yes, I was on the panel that drafted the call for the NNCI, which replaced the NNIN. The new approach involved individual sites proposing themselves, and then a team being assembled after the awards are chosen, as opposed to the old approach of having teams compete. I hope the new approach works well. The whole program is ABSURDLY underfunded given the stated goals of the NSF. There are some people who want to be able to claim that the NNCI will revolutionize research infrastructure in a transformative way, etc., but the annual budget for the whole thing is about $16M, which is less than the annual budget of a single one of the DOE nanocenters user facilities. Bear in mind that a single high end TEM is about$8M. I remain concerned that we are heading further into the regime where research universities will bifurcate into those that are willing to spend on top infrastructure, and everyone else.

Amy, long time! Thanks for the info - I'll definitely take a look!