Scanning Probe Energy Loss Spectroscopy Spels Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Nanotechnology is one of the youngest but at the same time most promising technologies of today. It has only been few decades since us humans started to work and think in nano scale, however even the idea has shown us lots of possibilities. Ideas such as computers and robots in nano scale (nano-computers and nano-robots) or bottom to top manufacture (constructing items by positioning atoms next together like Lego pieces) which has no waste in making material and has atomic precision, are few examples of what this great technology can bring to us. Progress of nanotechnology meant that we needed to be able to operate in atomic scale and this required new tools and equipment.

<Image from: >Inventing the Scanning tunnelling microscopy (STM) in early 1980's was one of the biggest steps in nanotechnology. STM was invented by Heinrich Rohrer (Figure 1 left) and Gerd Binnig (Figure 1 right) of IBM's Zurich Research Laboratory which later on they won the Nobel Prize in 1986 for their breakthrough. This new device provided researchers with the tool to explore material in atomic scale and even enabled them to play around with atoms. To show an example of what can be done with an STM in 1989 Don Eigler (IBM Almaden research centre California) used an STM to spell "IBM" with 35 xenon atoms.

How does STM work?

Figure -A schematic representation of a scanning tunneling microscopy (STM)

STM is built around the concept of quantum tunnelling. The very sharp tip of the STM (usually about few nano meters in diameter) is brought very close to the surface of the sample that is being examined. Then by applying a voltage difference between the tip and the sample, Electrons will travel through the vacuum area between the tip and surface of the sample creating a tunnelling current (Figure 2). The tunnelling current is what we get most of the information from. The actual tip position is controlled by piezoelectric tubes that by applying voltages move the tip in X, Y and Z directions (Figure 2).eyes.jpg

In general STM works in 2 modes:

Constant current

Constant height

In the constant current mode the STM is programmed so that the current between the tip and the surface is constant. So when the tip moving close to the sample reaches a hill, the tunnelling current changes. So the system makes the tip go higher to keep the current constant. The same way when the tip reaches a pit, tunnelling current changes and the STM will decrease the height of the tip to keep the current constant.

In the constant height mode the tip scans in the same height and records the change in current. In this mode if we choose the height of the tip to low, there is a chance of crashing to a hill and ruining the tip. In both modes by recording information we can get an idea of the shape of the sample that we are examining in atomic scale.


STM's have come a long way since they were first invented. Now we have STM's that work in gas and liquid instead of ultra high vacuum (UHV) and in a range of temperatures as low as zero Kelvin to high as few hundred Celsius.SPELS fig.png

Figure -SPELS geometryEven with all these upgrades STM is still limited in some areas. One of these limitations is that STM is not able to recognise what material is in the sample; therefore we need to know what we are using as the sample material before using the STM. This means we are not able to use multiple materials in our sample since STM cannot distinguish the difference between them. To solve this problem groups around the world have come up with STM based ideas to fix this limitation.

One of these ideas is Scanning probe energy loss spectroscopy that researchers at university of Birmingham nanotechnology department have come up with. Scanning probe energy loss spectroscopy (SPELS) uses an STM working in ultra high vacuum coupled with an electron energy analyser (an Omicron STM 1 with VG 100 AX hemispherical electron energy analyser). The STM's tip in SPELS is used in field emission mode, so the tip produces an incident beam of electrons that is scattered when it hits the surface of the sample. The electron energy analyser will detect the scattered electrons to produce local electron energy loss spectra.


Figure - Inside the STM chamber

Figure -The STM chamber

Figure -A picture of SPELS setup located at University of Birmingham

Electron energy loss spectra

energy loss diagram.png

Figure -An exaggerated schematic representation of the electron energy loss spectrum diagram

Figure 7 is an exaggerated version of the electron energy loss spectrum so that the peaks are easily noticeable.

As you can see in the diagram there is four set of bumps in the spectrum:

Elastic peak

Plasmons or plasmon peaks

Auger electrons

Secondary electrons

Elastic peaks

Elastic peaks occur when the electrons have an elastic collusion with the sample and scatter with the same energy that they collided with, so the electrons have zero energy loss.


These electrons have inelastic collusions with the sample atoms; therefore they will lose some energy before scattering.

With SPELS we usually look for plasmons. The amount of energy loss will help us identify the substance in the sample.

Auger electrons

Figure -The Auger processSometimes when the incident electron hits an atom in the sample, it removes a core electron (electrons from the inner layer [Figure 8-top]). When that electron is removed it leaves a vacancy in the inner layer so an electron from a higher level will release energy to fill the empty space. The released energy is usually in the form of an emitted photon but sometimes the released energy is transferred to an outer shell electron that gets removed from the atom (Figure 8-bottom). This ejected electron is called an auger electron.Auger.png

Secondary electrons

At the end of the spectrum with the highest energy loss are the secondary electrons. These are the scattered electrons from the sample. When the electrons from the incident electron beam collide with the sample atoms, they remove some electrons from the sample atoms. These are called the secondary electrons that have the lowest energy.

<Image from:>Uses of SPELS

One of the most important benefits of invention of SPELS is pricing. Although resolution of SPELS is usually higher than some other equipment, the price is mostly lower. For a rough comparison we can say SPELS will cost about half a million pounds which is half the price, compare to the price of an SEM (Scanning electron microscopy) that cost around a million pounds or it is a lot less compare to few million pounds for a TEM (Transmission electron microscopy). So we can collect the same or better results for less cost.

Figure -A schematic diagram showing the experimental configuration of SPELSSPELS tip is usually about 100 Nano meters from the sample surface, which is much higher than the tip sample spacing of an STM. By moving the tip from section to section and collecting and analysing the scattered electrons we gather local energy loss spectra from the sample. From the energy loss spectra we get information about the sample surface and sample substances. Since SPELS uses lower energies compare to most other equipment, there is less damage made to the surface of the sample and the sample itself. This is really helpful in surface science or any study that the surface is the primary aspect of the research such as study of catalysts. SPELS schematic.png

SPELS has two main advantages compare to a normal STM.

SPELS has the ability to identify different substances from the energy loss of the scattered electrons, therefore we can use multiple materials whiten the sample however with an STM this in not possible.

SPELS is less sensitive to the distance between the sample and the tip compare to a basic STM. So if we increase the distance we still get useful results with SPELS method, but will not get much useful information from a basic STM.

We should keep in mind that we can do all the basic STM procedures with SPELS as well, since the technique is based on an STM.

Although using the STM in field emission mode and collecting the scattered electrons with an electron energy analyser is a great idea with lots of benefits, it still has its down sides.

Figure -The electric field bending the electron pathOne of the important setbacks is, when we are using the STM tip in field emission mode, there is a electric field between the STM tip and the sample. This electric field is sends the incident electrons to the sample, but then again it is an opposing force for the scattered electrons. So when an electron collides with the surface and its scattering back the electric field from the tip is pushing the electron back towards the sample. This means the electrons path will be bent and in some cases the electron will return back to the sample before it is detected by the electron energy analyser (EEA). Bending.png

Figure -The shielded and non shielded tip geometry comparisonThe result of this is less electrons being detected and the electrons being detected at the wrong angle. To reduce the effect of the electric field we can shield the tip by using a microfabricated coaxial tip for SPELS (Figure 11). A microfabricated coaxial tip has a screening layer and an emitting layer with an isolator in between (in case of Figure 11 the isolator is hafnium dioxide which is a very good isolator and the screening and emitting layers are gold). The emitting layer is the layer that produces the electric field and the screening layer is there to make sure the electric field is as narrow as possible so it has the least effect on the scattered electrons (Ideally the field should be excluded to within a few microns). Since the emitting layer has a voltage and the screening layer is grounded we need the isolating layer in between, so the two layers don't make contact. So by using a microfabricated coaxial tip we can slightly eliminate the electron path bending problem.micro tip.png

Even though using the coaxial tip helps to collect better data, however since we are in early stages and we are only testing the equipment, we only use normal tungsten tips since it is easy and cheap to make and it is good for the results we are looking for.


There are 5 main types of imaging that we use SPELS to take






The results shown below are from a sample of silver (Ag) on top of graphite (carbon - C). These results were taken in February of 2011 using SPELS at university of Birmingham by Dr Shane Murphy.


Figure - STM scanning resultThis is just a normal STM scan as explained in the STM section of the report (Pages 2 & 3). We adjust the tip on constant current (or constant height) and scan the sample.

In the result shown across the bright section is the silver island and the dark sections are the graphite sheet.

We can see from the results that the islands are easily distinguishable from the graphite. This is because the silver is sitting on top of the graphite like an island. So it is higher than the graphite, therefore when the tip reaches the silver island the tunnelling current changes and depending on the STM mode we can separate the two substances. In this case we know that in the sample we have silver islands on top of graphite. If we did not have this information we would be unable to find out from this scan.STM result.jpg


Topography result.jpg

This is a quick scan that enables us to get a general idea of the shape of the sample and what does it look like. So later we can compare it to the other scans which take longer and see if the results are realistic.

Figure -Topography scanning resultComparing this to the STM scan you can see that the pixels on the topography image are visible but not in the STM scan. This is due to the fact that when the STM scan is being done the tip just moves in the direction and when needed it will adjust the height, but in the topography imaging since the tip is in field emission mode it stays longer than the STM in each section to collect the scattered electrons.

Counts result.jpg


Counts are a scan that only counts the number of scattered electrons. It does not matter what energy the electrons have, it just counts the number of all the scattered electrons in each section. We can see in the results that the number of scattered electrons is more in the graphite than silver.

Figure -Counts scanning result

Elastic result.jpg


Figure -Elastic scanning resultWe can adjust our program so it only counts electrons with specific energies (like a filtration system). Here the EEA only detects electrons that had elastic collusion with the sample and have not lost any energy. Again here when scanning, the tip stops in each section for few seconds then moves to the next section.

Plasmonplasmon result.jpg

Figure -Plasmon scanning resultThis scan is a scan of the plasmons. So the EEA only detects electrons from the plasmon region (they have lost some of their initial energy in the inelastic collusion with sample). We can see that in all of the scans the two separate materials were easily recognisable and they all matched and were realistic.


Recently we found out by normalising the graph we can get a better graph for our results. The result across is a division of the plasmon graph by the elastic graph. This is useful since instead of the actual peak number now we have the ratio of the two peaks compared to each other.

Figure -Normalised scanning result

last result.png

Figure -Electron energy loss spectrum for graphite and silver taken by SPELS

To analyse the images we use LabVIEW software. The software enables us to select a section from the image, and it will show us the electron energy loss spectrum from that area. Figure 18 shows an energy loss spectra obtained in SPELS from (a) graphite and (b, c) different regions of an evaporated silver film on graphite. The voltage and current in each case has been mentioned in the figure. In figure 18-c you can easily see the elastic peak and a plasmon peak (at 2.2 eV). It is by the size of the plasmon peak that we can identify what substance we are looking at since each material has a different energy loss and different plasmon peaks.

In general we put a sample in the equipment and then spend time scanning it in different modes and then analyse the data using our software.

Summary: What is happening with SPELS now?

SPELS is still in early days and it has not been in use much. The system is now completed. Dr Murphy's duty is to present how SPELS works and how accurate it is, so it can be marketed and sold to different sectors.

The goal is to get reasonable and useful data so when it is marketed university is able to show some hard data of what SPELS is capable of and how accurate it delivers it.20110222_008.jpg

Figure -SPELS in University of BirminghamThe main difficulties in this task are that since the equipment has not been used before, there is not much examples to compare to. Because there is not much information known, everything should be tried until the desired conditions and results are found. So at the moment Dr Murphy is trying different samples in different conditions to gather enough data so the best method of analysing the data and the best conditions of using the system is found.

The future of SPELS seems very bright and promising. Very accurate results with a comparatively low cost, all promises good things to come.