Centro de Microelectrónica Universidad de los Andes -

Teoria 1

Teoria 1 cuerpo

Tip preparation for the STM-Uniandes

Probes are a fundamental aspect in the application of SPM techniques. They are the first and most important element composing the transducer that allows us detect very small scale interactions. Since a STM uses tunneling electrons as the interaction, both sample and tip have to be of a metallic material so that they have a constant electronic structure throughout their surface.

The microscope's lateral resolution is a very important parameter since the smaller resolution the better the instrument. the lateral resolution is directly related to sharpeness achieved when fabricating probes, and hence their importance.

A scanning tip or probe can be characterized by the largest circle of radius r that can be inscribed into the tip geometry. The lateral ersolution is then estimated as:

The most popular tip fabrication metheds include[21]:

• Mechanical polishing or grinding
• Electrochemical polishing
• Manual cutting

At Uniandes we have developed latter two methods as explained below:

Electrochemical polishing The references section has two papers describing this procedure in more detail.

This method is used to remove material by electrochemically etching the interface between a conductor and an electrolyte by applying a potential difference accross them. This methos usually requires a post cleaning procedure to ensure a good tunneling interaction. It is mainly made using a tungsten (W) wire (9.95% purity 0.50mm diameter) and a less electronegative metal such as cupper or platinum.

The figure below shows the elements building a electrochemical polishing set-up. It includes an electrolyte, cathod, anode and a potenial difference.

This technique is also referred as lamellae drop-off since the etching geometry is in fact a lamellae of elecrtolyte formed around the conductor. The electrolyte is generally KOH or NaOH.

The tungsten wired used has to be carefully cut and placed inside a tip holder, which for our case is done using the metalic tube in a hypodermic needle 20G × 1/2 in, as shown below:

Additionally, this method needs a electropolishing circuit that can detect a very small change in the etching current and swithc off the voltage as soon as the tip has been formed. This circuit is described in the Tip Fabrication tab of the Publications sections. Using this set-up, we have fabricated scanning tips as the ones shown in the images below. From these images we can determine an estimate of thelateral resolution of the instrument. Since their radious os about 300nm, the average lateral resolution expected is of 18nm.

Manual cutting

This method is very similar to mecanically grinding down a wire until reaching a very sharp tip, however in this case instead of using a grinder we are using a fine wire cutters and we are taking advantage of the ductility of the material. THe most common material used in this case is Platinum Iridium (Pt-Ir). Upon applying the right force to a acute cut, very small tips are produce at the very tip of the wire and eventually one of them will have a small enough radius to produce a stable tunneling current when interacting with the sample.

At Uniandes, we used PtIr 80/20 (99.95% pure 0.25mm diameter) and a preocedure similar to the one described in [23]. The same needle 20G × 1/2 in was used as tip holder and a very precise stainless steel wire cutters. THe procedure is schematically shown below.

Once again we have characterized the fabricated tips and we found that the raidus of the tip at the very end was somewhat between the expected range. The images below show SEM images of the fabricated tips using this method and from there we can quantify an approximate radius of 70nm, that could potentially render a lateral resolution of 7,9nm.

SPM-Uniandes Software

Under construction

STM-Uniandes Mechanical set-up

A precise control of the position of the tip/probe with respect to the sample is necessary if we want to achieve detectable tunneling currents. The positioning process is done in two stages. Firstly, a coarse positioning is achieved by manually placing the tip over the sample at a tunneling distance (~nA). When this is achieved, the tip is scanned around the sample using a pair of piezoelectric elements in a scanner configuration, and a third piezo controls the tunneling distance using the controller described in the electronics section.

Coarse positioning

At Uniandes, we implemented a similar system to the one used in the SXM Project, however the high standard pieces needed to reproduce such system at a low cost are generally difficult to obtain in the local market. This mechanical system is based in the principle of a seesaw in which linear movement in one end can be scaled down to the movement of the other end as explained below.

Seesaw scheme

This principle can be easily implemented using a pair of metallic plates in which one of the metallic plates sits on top of the second one and can be displaced using 3 screws in a tripod fashion. Two screws in the front can perform as the pivot of the seesaw system and the third screw would set the distance d1 above shown. Such a system would schematically look as the one below.

Largo positioning mechanism

Even though this systems would theoretically deliver any amount of distance scaling, in reality L1 is a finite distance hence the scaling values would not reduce more than 30 to 50 times. A standard screw would displace about 1mm per turn then if we manually rotated the screw with a precision of about 2 degrees, the finest approaching precision would be in the order of 120 nanometers.

Obtaining screws that would displace less than the example above would mean elevating the cost of fabrication and maintenance of this system, hence at Uniandes we decided to use a standard 0.5mm per turn screws and a 1/150 reducing gear easily obtained in the local market. This set-up would deliver a fine precision of about 1nm, as it is needed.

Fine positioning

Sample holder

Vibration isolation

A scanning probe microscopes set-up has to be such that the probe can be steadily placed at the order of tenths of nanometers from the sample. When trying to achieve this, mechanical vibrations are a main issue since modern buildings transmit structural vibrations nearly everywhere. To avoid this, a SPM has to include an appropriate mechanical isolation system that can reduce vibrations to about one pico meter (since we are trying to control nanometer positions) [15]. This would mean an attenuation of −120 dB when the input vibrations are in the order of a couple micrometers.

The mechanical isolation set-up not only includes the elements that reduce vibrations transfer from the outside world toward the tip-sample, but also all the elements that can hold wires and connections in place and enhance the robustness of the instrument.

It has been reported that mechanical vibrations range from 10 to 100 Hz [16]. There are many elements, such as motors and fans, that contribute considerably to this mechanical inputs. Their typical frequency is around or a factor of the electrical network frequency (60Hz in Colombia). Other vibration sources include people walking and the movements of buildings themselves. Those range from 1 Hz to 15 and 25 Hz.