Building a Scanning Tunneling Microscope for less than $100
By Christopher MooreI am a scanning tunneling microscope (STM) guy. I completed my M.S. thesis using a commercial instrument and I have always wanted to build one. Until recently, I haven’t had the time or the money. About 1 month ago, a student here at Longwood University approached me about doing a project. The first thing that came to my mind was “Let’s build an STM!” The student agreed, and off we went. Now building an STM is part of my job!
Commercial research STM systems cost up to $250k or more. Less powerful “student” systems run around $15,000 but offer little to the researcher. The STM that my student Donald and I are building will end up costing ~$100 in parts. More important, the parts can be found at the local hardware store or ordered online.
Over the next few months Donald and I will chronicle the construction of our STM, showing you what parts we have used, how we are constructing it, and how it works. We plan to have a fully-functional analog STM up and running by the end of this semester (early December 2007). Next semester begining in January, we will chronicle the build of our second STM which will be digitally controlled via a computer. The budget for our computer controlled model is less than $200, and the microprocessor and control board are being designed by an undergraduate at Virginia Commonwealth University.
We have already begun. To learn more about STM, this specific project, or current progress, and what’s next, continue reading.
First, what is an STM? Here is a short blurb from that thesis I mentioned above:
STM produces real-space images of a surface with atomic-resolution. Its operation relies on the quantum mechanical tunneling of electrons through a finite potential barrier. A sharp metal tip is brought to within a few tenths of a nanometer of a conducting sample surface. A potential is applied between the tip and surface, causing electrons to tunnel from the tip to the surface. The probability of electron transmission, and therefore the tunneling current, is exponentially dependend on the magnitude of the tip-surface gap. By manipulating the gap size with piezo-electric devices to maintain a constant current and rastering the tip across the surface, information about the topography and electric structure of the surface can be obtained.
Essentially, the STM is a sort of “brail microscope” that allows us to “see” atoms on surfaces. Using principles of quantum mechanics, we can create a device that uses a sharp tip to “feel” what the surface of a conducting or semiconducting surface looks like with a resolution allowing for the determination of individual atom placements. We’re talking VERY small stuff.
We need to build a device that can move a sharp metal tip in three dimensions on a scale of less than 1 micron (less than the thickness of a human hair). The design we have chosen to work with is based off of the model shown on John Anderson’s website. On this site, John outlines the construction of an STM for under $100 using a patented disk-piezo scanner. Visit his site for more details.
For the month of September, our goal was to have the mechanical pieces fabricated. By the middle of the month of October our goal was completion of the scanner. So far, we are right on schedule. First, I will detail the mechanical construction. Then, I will discuss the scanner.
Schematics for mechanical portions of the scanner and sample stage can be found here. Below are some pictures. Click on the picture for a larger version:
Here is a picture of Donald working at the mill. Note that a mill is not necessary; however, we have one so we might as well use it. 
Here is the completed sample stage:
Notice that it is no larger than the plam of Donald’s hand:
We need a means to move or STM tip in three dimensions. Typically, this is accomplished with high-voltage piezo-electric tubes that expand and contract on the nanometer scale with applied voltages. High-voltage is more complicated to work with and typical piezo tubes cost up to $200 a piece. For 3D, that equals at least $600, or six times our budget! John came up with a brilliant solution that uses a small piezo disk from a $1.50 piezo buzzer. Since he invented this type of scanner, I’ll let him explain it. Read here.
The buzzers that John uses are actually $4 a piece now. Also, I do not believe that Panasonic is still manufacturing this specific part. I was able to order several buzzers from Digi-Key, but they have a very limited supply. The part can be ordered here. I am currently researching alternatives and will update if I discover something.
The piezo disk must be separated into 4 quadrants. We used an exacto-knife with a straight edge to do this. I encourage everyone to find there own student for this part, since this is actually rather tedious, difficult, and took more time and finger strength than we thought it would. Donald has sore fingers, I do not!
Here is a picture of our piezo disk:
In the center is the tip holder, which we made out of an IC pin and a small plastic standoff that we found in our electronics lab. In this picture, it is not mounted. We will mount it later after we wire the piezo and build the support. As described on this page, John wired his disk-scanner using silver paste and secured the wires with super glue. He was concerned about soldering the connections since the heat could damage the piezo material. I have soldered piezo material before with great success, so Donald and I were not as timid. However, we did use super glue to secure the wires. Here is a picture of our wired piezo-disk:
To secure the piezo-disk, we created a housing using a 9/16″ washer like the one used by John (details here). The washers I was able to buy at the hardware store ($0.16 a piece) had a larger outer diameter than those John used, so it would not fit between the screws on our scanner. We used our mill to notch out space for the screws.
The wires connected to the piezo-disk need a path out of the housing. John used a file to create trenches in his washers, which I can imagine would be a pain in stainless steel. We have a mill, so we used that. Below are pictures of our scanner housing holding are wired piezo-disk:
Our scanner is not quite complete, since we need a means to control its motion. Essentially, we need to apply equal yet opposite polarity voltages on opposing piezo quandrants to move in a given direction. For this, we use a signal generator that produces a triangle function. Basically, we linearly increase the voltage to a peak and then linearly decrease the voltage to a minimum. This will cause one quadrant to expand and contract. However, we need the opposing side to contract and expand, respectively. To accomplish this we used a series of operational amplifiers set-up as inverters. What an inverter does is it takes an inputed voltage and spits out its inverse (opposite polarity). In the next installment, we will discuss the electronics in more detail. Here is a picture of our control electronics for the x and y motions for our scanner:
The integrated circuit in the middle is composed of 4 op-amps that are used for the -x, +x, -y, and +y quandrants of the piezo-disk. This IC can be bought from Digi-Key here. The schematics can be found here. We have constructed our circuit on a solderless breadboard for testing. In about a month, we will be getting a CNC router for printing circuit boards. When that arrives, we will design and print a board.
Below is an image showing the circuit hooked up to a triangle-wave generator with an oscilliscope in the background showing the signals the -x and +x quandrants would “see”. Notice how the two signals are the same, only inverted:
Next week, we plan to complete our scanner by mounting the tip holder and mounting the entire assembly on the scanner stage. We will also begin contruction of the STM pre-amplifier.
5 Comments »
RSS feed for comments on this post. TrackBack URI
Leave a comment
Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>
I am developing a site. There are hundreds of free math video tutorials already. I am aiming to submit over 1000 video tutorias by the end of 2007. Do you need help with algebra. Take a look at my site. If you have any suggestions, please let me know.
Comment by TuLyn — Wednesday -- October 17th, 2007 @ 12:43 am
Any updates to the STM project? It looks like a really cool project and was wondering if you got it up and running yet?
Comment by Scott — Friday -- January 11th, 2008 @ 11:04 pm
We are still working hard! We ran into a few problems with the electronics and the piezo disk that we are trouble shooting.
My student is getting a grade for this project this semester, so he will be required to keep this blog up-to-date on our progress. You should start seeing more updates in the next week or two.
Also, I have another student building a sputtering system for thin-film deposition. He will be writing about that project here, as well.
Comment by Chris Moore — Thursday -- January 17th, 2008 @ 5:41 pm
man..thats wonderful if u can build an STM for so cheap a price. Are you able to control the distance at the tunnelling regime. how do u manage with the temperature and vacuum?
May I will send my first sample to you. all the best with ur work. If this works, developing countries will have access to STMs too. Good luck!
Comment by Saranyan — Saturday -- February 23rd, 2008 @ 7:11 am
The distance is controlled via a low-voltage unimorph piezo disk and an analog feedback circuit. We can control the distance via an analog set-point voltage on the difference amplifier.
There is no vacuum. This is an air device and is therefore only able to resolve a few systems on the atomic scale (like graphite).
Our STM does not work (yet!). However, we are not the first to construct this design. It has been built and it works elsewhere.
Comment by Christopher Moore — Wednesday -- February 27th, 2008 @ 12:46 pm