Slashdot Mirror
Build Your Own Scanning Tunneling Microscope
I don't want to spen writes "For all you fans of nanotech out there, a friend just posted me a link to instructions for building a scanning tunnelling microscope, from the University of Muenster. Interestingly, their licensing terms sound open source-ish to me: '(... We grant everybody the right to construct the microscope using the here-published design for private or educational purposes. On these web pages all necessary diagrams, drawings, material descriptions and software-source-codes are published for free access. While granting the right to build the microscope we make it mandatory that new developments, improvements or other applications of our design are also made openly available for private or educational purposes...)'"
Atomic force microsopes on the other hand can do some very neat work with small organic particals, but seperating something like an HIV from solution is still difficult, and usually involved crystalization.
How we know is more important than what we know.
I dont think so
;-)
With BSD or GPL style licenses you are actually allowed to use the copyrighted work in an commercial setting, just not to sell it. For instance a commercial company might run their web server using GPL licensed software.
With this not only do they retain the exclusive commercial rights, but the license might in fact be read as an attempt to force you to turn over any improvements on their design.
So if you make an improvement, does this mean that you have implicitely granted the University of Muenster the rights for commercial exploitation of your own improvment by accepting their license in the first place?
This does not sound "open source-ish" to me, it sounds like out right theft.
PS: Please ignore any bad spellings/grammar in my english or at least be polite when telling me
This type of microscope requires a target which will conduct electricity. That's why images are typically of gold, semiconductors or items which have been electro-plated.
You're full of shit. Ya can't use an STM with organic samples, let alone live organic samples, let along live organic samples in real time. I know most Slashdot residents are morons but try not to insult the intelligence of those of us who have one.
How we know is more important than what we know.
Your comment makes little sense. My lab does a lot of microscopy. B & L strictly make optical microscopes, which by the wavelengths of light, and properties of glass are restricted to resolve objects down to about 100 nanometers or so (at best - and we're talking with a really really good confocal or deconvolution microscope that runs about $500K). Mediocre electron microscopes visualize objects down to about 1000 angstroms. That's two orders of magnitude better, perhaps more if you have a good EM setup.
I think a similar thing was done with robots that carried out part of the human genenome project. Sorry I forget the group that put the plans out on web. Any how, there is certianly a demand for scientific instruments that can be assembled for relatively low cost. A good thing for universities in developing countries, that can't afford the cost of brand new equipment. I saw a pH meter made of the carbon rods from D size cells made in India. A lot of labs in developing countries either buy second hand equipment, but just like every thing else there is planned obsolesence, and parts get hard to find or make.
I am not aware of any of any instance of a large molecule whose structure was deduced from a scanning tunneling microscope. Things like proteins, enzymes, and viral particles are most generally probed by growing them into crystals and a analyzing their x-ray diffraction patterns. The big problem with this technique is that big molecules are hard o grow into crystals (thus all the grow protein crystals in the micro gravity of orbit effort) Nonetheless, a Scanning tunneling microscope is VERY COOL, and within the engineering capabilities of a dedicated hobbyist. Heck, you can now buy a complete Scanning tunneling microscopes for 20K; pretty cheap for a high-end piece of laboratory equipment. The real trick is to get the stage into a hard vacuum at cryogenic temperatures. Who will be the first person to spell out LINUX in Xenon atoms on a nickel substrate? I will donate money to that cause By the way, in college I used to produce atomically sharp needles for field emission ion sources just by burning tungsten wire in a propane torch. That should be an easy way to make probe tips
I mean this is cool, bet really the cost on older electron microscopes is pretty low (under 2k) Heck theres a phillips 500 for sale on Ebay for $1000 right now. I've often thought about buying one, but dont have 3 phase in the house and the garage is a little damp.
Yes, his post is completely wrong. However, there have been advances in STM so that you can do STM with fixed organic samples now. Here's an example of a recent published study. Also atomic force microscopy (AFM) has become fairly common with fixed organic samples. I'm not calling you wrong - I'm just updating Slashdot readers to the state of the art in biological microscopy.
The design as given requires a ISA slot because of the type A/D converter card they selected. If you do not have an ISA slot available, I am sure a PCI based, or even a USB based analog to digital converter can be found. It would probably be a good idea to change the A/D, as the one used has a 100khz refresh rate. I am sure that there are cards out there that refresh at a much quicker rate, thereby allowing improvements in other areas of the design. Just be aware that the software would have to be modified because of the different card, but that should not be a difficult matter for anyone attempting this project.
SELECT * FROM User WHERE Clue > 0
0 rows returned
Depends on the size of the protein, but unfortunately you're right for most interesting proteins. There are hybrid computational modelling + mass spectrography techniques that can reduce the required computational time by orders of magnatude. When peptides fold up into proteins they make covalent bonds between the aminoacids. When you're doing a computer simulation you can say something about which bonds are more likely than others (and this can reduce your run time to less than blind search) but you can't say with any certainty which aminoacids are bound to which. What these hybrid techniques do is cut up the folded protein into small molecules (5-6 aminoacids) and then messure the mass. From this data you can tell if there are certain crosschain combinations. For example, you can see that there is a Cysteine-Asparagine bond and if there is a Serine-Proline bond or whatever. Armed with this information you can remove a lot of possibilities from your search space.
How we know is more important than what we know.
No, but you can visualize prion proteins within intact cell with a conventional electron microscopy. I kid you not. Here's an article with some really pretty pictures of prion proteins trafficking in cells.
Despite what they call it in English (University of Muenster), the proper name is the Westfaelische Wilhelms Universitaet. Even though I am an American, I studied there for a semester.
/., it isn't a surprise to see good things from Muenster. It is one of those wonderful little secrets - a top notch place few know about.
;)
Muenster is a wonderful college town, as well as a place of historical significance (30 years war ended there). The hospital associated with the university, and thus the medical program, are well respected across Europe. (Comparible to Mayo / Johns Hopkins / Mass. General here in the US).
Anyway, while it is surprising to see this on the front of
Posting AC because I believe in privacy on the net
The GPL only kicks in when you *distribute* copies of the software. In this case if you choose to ignore the GPL, copyright law defaults to 'you are not allowed to distribute the software'. Hence the *only* legal way to distribute (note: I didn't say 'use') GPL'd software is to agree to the GPL.
There is a similar reasonably well-documented homebrew STM that was built by a guy named Jurgen Muller. His site is pretty interesting, and well worth the read.
Obviously there are a lot of articles on STMs in various academic journals. If you're at a university, you might start by searching in Reviews of Scientific Instruments and perhaps the Phys Rev journals.
I was involved with a STM project for a while, and our conclusion was that the 3D piezo setup is quite fragile, and extremely difficult to isolate from vibration, etc. It seemed that a better design was a so-called slip-stick walker, which uses a stage that slides on smooth rails. A tube of piezoelectric ceramic is attached and driven in such a way that it creates a series of small, sharp forces on the stage that momentarily break the static friction between the stage and base, causing it to move in small steps.
This stage is used to approach the sample to the STM tip, which is mounted on another piezo tube, and can be deflected laterally and vertically in order to do a raster scan of a small area of the surface.
The limitation to this method is that you can't scan a very large surface area. You can add a second "walker" unit underneath the first one so that you can move the sample from side to side in addition to moving it towards/away from the tip, so this would allow you to scan a stripe across the surface.
To get full 3D control, there are several designs called "beetles" (IIRC) that are described in the literature, which use a somewhat similar technique that allows more control.
Maybe you should read it a little closer. This is a summer project for rich and technically competent high schoolers, or grad students. This is not cheap. And when it comes to making tips I think it's great to have kids playing with tubs of KOH.
And it's not SEM, it's STM. Sem is great for making pictures of insects and what not, STM is great for tracing out the p-orbitals of graphite. BIG difference (not your error, but as long as I'm clarifing, why not hit that too).
Propose a sputtering chamer or a PVD chamber, they'd probably be much cheaper to build and can be used to make other stuff. Which then one could look at with either an SEM or STM if one chose.
Goto industrial and university auctions too. I've hear tell of people giving TEM's away to whomever was willing to transport them (not that an isolation pad on which to set it is within the means of Young Scientists). But still.
Then about another day to learn how to cut the Pt-Ir needles to be atomic (this isn't that hard, a wire cutter worked). If you think about it there is probably a single atom at the bottom no mater what (there has to be one lowest atom) and since the tunneling falls off rapidly r^4 (?) others can be close but not interfear significantly.
On the third day I had it calibrated with the glassy carbon provided by the manufacturer.
But the biggest deal was the vibrations. I had to clog all the air vents in the room because they all made noise (vibration) and then use the room at the time of lowest building activity. I came to the building at 1 AM and took my worst images only to walk out the door and find a guy buffing the floor with a huge hopping floor buffer. Then I tried durring dinner and everything worked out great (I just ate pizza--typical student food).
If we use the FSF's mode of describing things as a point of reference, this is probably an open source project, but not a free one. If it were a software project, it would violate the very first of the freedoms the FSF considers essential (look here) in that it restricts who may use it for what purposes. Whether or not they can meaningfully impose that restriction is a separate issue.
The technical answer has to do with lots of issues, like most sound cards being for AC signals and the input not so great for DC signals. The simpler answer just asks would you do surgery with a chain saw?
I'm an American. I love this country and the freedoms that we used to have.
I am probably quite qualified to answer your
question seeing as I do STM research for a living.
Your second question is easier to answer, so I'll
do that first:
there are two ways, either you put down enough
of them to assure there will be a molecule in the
range of your scanner whereever you approach
or you use some other technique like lithography
to make small structures then another technique
to deposit your molecules near those structures
then (if you got the microscope that allows it)
position your tip optically near the structure
and spend days on looking around with STM until
you find it.
Now you first question. STM can be used to find
some structural info from large molecules. My lab
has done some research on nanotubes and you can
get atomic resolution on those and then determine
their helicity. People have also imaged bio stuff
and for some smaller molecules have seen the
structure. Even DNA has been imaged. That said,
STM is not a great structure probe, it is a great
probe of electronic states.
Last word of warning: people rarely realize that
STM in air is not going to tell you anything
that you can rely on physics-wise. The reason is
that all surfaces exposed to air are covered in a
thin layer of water which makes the interpretation
of data hard. What they show on that page is a toy
though well-thought-out and maybe even useful to
some. Seeing atomic steps on gold and "atomic"
resolution on HOPG is not hard, just don't hold
your breath for something like atomic resolution
on gold, or silicon, or anything else really.
For that you at least need a UHV system.
Cheers.
Like it says on the tin, building a scanning tunneling microscope was a 1st year undergraduate physics project at Imperial College of Science and Technology back in 1987 when I was there(http://www.imperial.ac.uk/)
In my universe, 1000 Angstrom is the same as 100 nanometers so that makes your statement zero orders of magnitude. My experience with a SEM that's cheaper than the $500k B&L is that you can get well below 100 nanometer.
There are several reasons why you would not want to use a sound card's line inputs. One the most important is that a sound card's refresh rate is not continuously adjustable over the range which it can operate. I have not messed much with sound cards lately as ripping CDs and working with sound is not what I am into, but I seem to remember that sound cards sample as certain predetermined rates like 8k/sec, 16k/sec, 22k/sec and 44k/sec. I am sure sound cards can sample at higher rates, but if you need to sample at an intermediate rate say something like 33k/sec you are just plain out of luck. An A/D card would permit sampling at whatever rate you need up to its maximum refresh rate, which can be in excess of 1 MHz.
The next reason not to use a sound card, is that when it samples the incoming signal, it only samples in the audio ranges. Any analog data above say 25,000 khz is not captured. A sound card has problems at the low frequency end also. What if the incoming data is a DC voltage? As far as I know sound cards can not "read" a 0 hz signal, since 0 hz is not an audion signal. This can be a big reason to rule out the sound card approach.
Another advantage A/D cards have over sound cards is that most, if not all, A/D cards also come with a couple programmable D/A outputs. These D/A outputs permit instant on operation and adjustable output voltages and frequencies. This will allow a computer to control a device, in this case the microscope. While a sound card has D/A outputs, those outputs are tuned toward audio frequencies. Is a sound card capable of outputing a DC signal. I dont know. But if you are trying to use relays to control the timing of events, using an AC signal will just make those relays chatter. Definately not good.
SELECT * FROM User WHERE Clue > 0
0 rows returned