As a part of the course “From Idea to Result”, we had to conduct a research project within a research field completely new to us. Sand is an important source of industrially valuable components such as magnetite (Fe\(_3\)O\(_4\)) or ilmenite (FeTiO\(_3\)). The project I became a part of closely investigated the magnetic properties of a natural sample of heavy sand from Villingebæk beach (North Zealand). In order to do so, we first separated a magnetic fraction from the sample for further analysis (simply by using a strong handheld magnetic and collect magnetic particles).
We then employed a variety of experimental methods to analyse the particles’ chemical components, but our analysis concluded that there were only low concentrations of valuable heavy minerals in the sample – meaning that we didn’t discover a source of income to supply our SU.
The methods we used are described in the menu below (click on the names to read more about each method):
A (handheld) XRF spectrometer uses short wavelength x-rays to ionise the elements of a sample, and determine the amount of elemental compounds of a material with a small detector. The spectrometer works within some precision, as it must be calibrated to the specific type of sample, e.g. dirt, sand, wood, etc.
Mössbauer spectroscopy utilises the crystal structure to determine the Iron compound of the sand. The sample is scanned for nuclear resonance frequencies using an oscillating gamma ray source. The source is radioactive Cobalt-57 that decays into an excited state of Iron-57 which then decays to the ground state of Iron-57 by emission of a gamma ray. If the sample you are scanning contains Iron, these gamma rays will be absorbed and detected as a drop in the intensity of the gamma ray source’s transmittance.
XRPD will also utilise the crystal structure of the sample to determine the mineral compounds of it. Using an x-ray source, the sample is scanned. Now, it is important that the sample is isotropic meaning that the analysis is independent of scanning the sample from a specific angle. Therefore, samples for XRPD analyses are usually ground into a fine powder, making all crystal planes quite random in direction.
The sample is furthermore rotated to make sure we analyse the material from all sides to ensure a more true randomness in the sample orientation.
We then look for a signal that arises because of Bragg’s law – the atoms in the sample will act as a diffractometer with diffraction grating being the atomic distance between the atoms. We should see constructive interference signals when this atomic spacing and the beam angle matches integer values of the beam wavelength. Whenever there is interference (creating a peak in the signal), we can compare the intensity of the peak together with the angle of the beam with experimentally determined table values of pure materials.
SEM works by placing the sample in a vacuum chamber and using an electron beam to scan the surface of the sample. Roughly said, the back-scattered electrons (flying off the sample surface) are used to create a detailed image of the surface and texture of the sample.
The SEM we used was equipped with an EDS detector, described in the next item.
An EDS detector is a detector placed in the SEM which utilises the fact that elements in the sample will be ionised by the electron beam. Ionisation of an element is very temporary – the compounds in the sample will emit radiation that is characteristic to the specific compound. It’s like a very specific XRF spectrometer, allowing researchers to only analyse a tiny, microscopic area of the sample.
The conclusions across the methods are in consistency with the main conclusion, and some of the analysis is tedious if you’re not particularly interested in sand.. Instead of rigorously going through it all, I will just show you the most interesting results.
Below, I included a gallery of all the electron microscope pictures we caught on the magnetic fraction. Lighter areas in the particles correspond to denser constituents, i.e. heavier elemental component. Most of the particles are, as you see, heavy, so they could be filled with valuable heavy compounds (hypothetically speaking, as I’ve already confirmed the opposite).
On one of the SEM pictures, we saw a dark area on the tip of a particle. Using the EDS detector, we analysed the two areas to get a better understanding of their constituents.
In the spectres below, you can see (for the marked spots “Spectrum 11” and “Spectrum 12” on the photo above) the registered constituents of the sand particles. We can see from Spectrum 11, that the dark spot consists of organic materials – in other words dried sea water! This makes sense, as the sand was collected directly from the beach. There had been no cleaning of it whatsoever, before we analysed it.
Now, this is a bit technical to explain thoroughly, but I still want to show you the cool spectrum we obtained for the magnetic fraction when we ran it through the Mössbauer spectrometer.
On the figure below, you can see a lot of green lines kind of packed within a blue line. Should you add each green line with each other, the results would in fact create the blue line. Each green line indicates a potential constituent in the sample which we tried to fit our data points (the red crosses) with. Choosing these is a bit of a guess, mostly based on experience. Luckily, we worked together with a very experienced chemist who helped ‘suggest’ some suitable fit-parameters. Some of the small ‘bumps’ and skewness in the peaks suggest that our material isn’t entirely pure.
Personal notes on the project
This was a really exciting project to partake in! I have never considered myself an experimental physicist, but I could really see the exciting applications of the methods. However, in the end, I have decided to let my experimentalist take a long break while I explore data analysis and computer science 😉