Since 2017 we have a LEEM microscope in our lab. This technique, about three decades old, is really neat for observing dynamics on crystalline surface. We have been using LEEMs starting in 2003. The lack of an instrument (until 2017) gave us the oportunity to develop fruitful collaborations with some LEEM owners, specially with our US colleagues, Kevin F. McCarty(Sandia Nat Labs, USA) who retired in 2015 and whose instrument is now handled by Farid El Gabaly, a former PhD student of our group, and Andreas K. Schmid (Berkeley Lab, USA), who is in charge of the Spin-polarized LEEM at the Nacional Center for Electron Microscopy, at Berkeley lab. If LEEMS in the world are less than a hundred, spin-polarized ones can be counted with one hand with spare fingers.
A LEEM is very similar in concept and design to a Transmission Electron Microscope but with reflected electrons: a beam of electrons at a few tens of keV is focused by several electromagnetic lenses, and slowed to energies of a few electron volts before reaching the sample. The reflected electrons are accelerated again to high energy (aberrations are already bad enough at higher energies), before their distribution is amplified by several lenses and imaged by a set of channel-plates and then a phosphor screen. The best thing is that this technique is very fast, as it is not a scanning technique. Videos of the evolution of the surface with nanometer resolution (ok, 10 nm) can be acquired in real time. You have not seen thin film growth until you have seen monolayer islands grow in real time!. A movie showing the growth of Co on Ru can be reached here. Surprises can arise with this kind of dynamic view of surfaces, such as our reported serpentine growth of Pd on Ru.
Furthermore, this microscope allows you to perform low energy electron diffraction (LEED) in selected (micrometer sized or even smaller) areas of the surface. If you have ever taken a complex LEED pattern of a surface, you know that is it a pain to have all the patterns from every corner of the sample together. With a LEEM microscope, you can get only the patterns from a single terrace (check our publication on the LEED patterns of a single terrace on an hcp substrate). Or you can use diffraction contrast to image only the surface areas with a given diffraction spot (the so-called dark field imaging mode of TEM). We exploited this mode of work when understanding the evolution of two copper layers on ruthenium as presented a our Science paper.
But it gets better: you can use a source of spin-polarized electrons, and do Spin-Polarized Low Energy Electron Microscopy (SPLEEM). Then you can see the magnetization of a monolayer film in real tiem while you change the temperature. Or determine the three components of the magnetization vector in a ferromangetic film, one component at a time. For a good example of the type of research we have done with this, you can look at this Physical Review Focus.
If you have a synchrotron around, you can do Photoemission imaging with the same instrument. For that, we believe a good example is our work on "nanometer-thick magnetite", or our work on cobalt-iron oxides published in Advanced Materials.
We hope to have convinced you that LEEM is much, much better than having a traditional surface science system. So why doesn't everyone devoted to surface science (LEED, XPS,...) have one of this apparatus in their lab (starting with us)? Well, on one hand, it is "resolution challenged" when compared with the scanning tunneling microscope (STM). You get around 10nm resolution in commercial systems (two companies sell them, Specs and Elmitec). For higher resolution you need to do aberration correction (where you can get about 2 nm). And to be frank, it is expensive. Well, not that expensive, the price tag of entry systems is similar to a fully equipped STM one, but while for the latter we can cheat and build most of it ourselves (including the STM), building a LEEM is a bit outside our area of expertise.
Since 2017 we have the only LEEM in Spain devoted uniquely to imaging with electrons as illumination. Actually, there are only two LEEMs in Spain, so competition is not that great. The other one has been running since 2010 at the spanish synchrotron ALBA at the CIRCE beamline under Lucia Aballe and Michael Foerster, with whom we have an ongoing great collaboration. The microscope has been acquired through an infrastructure call of the Spanish Ministry of Economy (MINECO, CSIC15-EE-3056) and cofinanced by the ERDF (European Regional Development Fund) and the support of Adrian Quesada (ICV), Enrique G. Michel from the UAM and Arantzazu Mascaraque and Lucas Pérez from the UCM. You can see the instrument here:
There are a few introductions by the several groups that use these instruments. These are a few of the most relevant ones (without pretending to have a complete list, but with a bit of shameless self-promotion):
- The LEEM Wikipedia web page.
- A "low-energy electron microscopy" chapter (you can find a draft version here) written between J. de la Figuera and K.F. McCarty and publised in the volume "Surface Science Techniques, compiled by G. Bracco and B. Holst, as part of the Springer Series in Surface Sciences, 51 (2013) 531.
- The Bible is the book "Surface Microscopy with Low Energy Electrons" from Ernst Bauer, published by Springer.
Mössbauer spectroscopy is based in the Mössbauer effect, that is, in the emission and resonant absorption of gamma rays by nucleii withouth energy loss due to nuclear recoil. The nuclear resonant absorption has been observed in more than one hundred nuclear transitions of different isotopes of various elements. Mössbauer spectroscopy is only applied to solids or frozen solutions independently of their crystalline or amorphous character. From those all transitions where the Mössbauer effect has been observed the most popular, by far, is the 14.4 keV transition of 57Fe (to which most work is devoted). Other popular isotopes (but less used) are 119Sn, 151Eu and 121Sb.
The characteristics of the 14.4 keV transition of 57Fe and its relative isotopic abundance (2%) make possible to work at reasonable experimental conditions: room presure and temperature using reasonable amounts of sample (usually a few miligrams). The possibility of performing Mössbauer spectroscopy with 57Fe is very fortunate since iron is an element of the most scientific and technological importance. It is involved in magnetism, catalysis, corrosion, biology, mineralogy, metallurgy and many other interesting and important fields.
By using Mössbauer spectroscopy one can quantify the magnitude of the hyperfine interactions. From the quantification of these interactions, which depend on the environment in which the Mössbauer atom is located, chemical, structural and magnetic information can be obtained. For example, the oxidation state, the cordination type or the magnitude of the hyperfine magnetic field, if there is any kind of magnetic ordering, can be easily determined. By recording spectra at different temperatures, information about magnetic ordering temperatures can be inferred and by the application of external magnetic fields the type of magnetic ordering can be deduced.
Each iron species is characterized by three different hyperfine parameters. In complex samples as multiphasic samples or compounds with various iron sites, Mössbauer spectroscopy can be used to identify each phase ("fingerprint" method) or the different iron sites. Our work is mainly dedicated to 57Fe Mössbauer spectroscopy although we also have sources for 119Sn Mössbauer spectroscopy. The spectrometers are located in a separate room in our laboratory area.
Of the three types of Mössbauer spectroscopy (transmission, ICEMS, ILEEMS), in Spain there are about 10 groups that have transmission equipment, less than three with ICEMS (i.e., for thin films) acquisition (including us) and one for ILEEMS.
Transmission spectrometer
Experimentally, the most common mode of operation in Mössbauer spectroscopy is the transmission mode where the gamma quanta rays emmitted from the sample and passing through a thin absorber arrive to an approppriate detector. This mode provides mainly bulk information from the sample, which must be thin enough for the gamma rays to go through the sample. We have an spectrometer devoted to transmission spectroscopy. In this one we often use powder samples, or micrometer-thick foils. The samples can be cooled down with a He-closed cycle cryorefrigerator for the sample which allows us to record spectra at different temperatures between 15 K and 298 K. This is often crucial to detect different iron compounds, as we do for our work on complex oxides.
Integral Conversion Electron Mössbauer Spectrometer (ICEMS)
For some nucleus, such as it occurs for 57Fe, the deexcitation after the nuclear resonant absorption is more likely to occur via an internal conversion process, where electrons are emmitted from the sample. Since the mean free path of the electrons in a solid is usually short (depending on their energy) if these electrons are detected, Mössbauer spectroscopy can be turned surface sensitivity. This mode of operation is called Integral Conversion Electron Spectroscopy (ICEMS). Because of the energy of the electrons involved in the process, information from the uppermost 300 nm of a sample can be obtained, although this information is mainly weighted towards the most external 50 nm. By careful preparation of the samples, that usually involves their enrichment with 57Fe, the method can be sensitive enough as to detect a fraction of a monolayer.
Our second spectrometer is devoted to ICEMS by means of a Parallel Plate Avalanche Counter (although we can also measure in transmission mode in required). This allows us to routinely measure films in the 10-300 nm thickness range, perfect for Pulse Laser Deposition films of single crystal oxides.
Integral Low Energy Electron Mössbauer Spectrometer (ILEEMS)
A final variant of Mössbauer spectroscopy using electrons is based on the detection of electrons of very low energy (Integral Low Energy Electron Spectroscopy, ILEEMS). By the approppriate application of a positive bias voltage at the cone entrance of a channeltron these low energy electrons can be accelerated and counted more efficiently making the technique very surface sensitive. Our third spectrometer belongs to this class, with both the sample and channeltron in ultra-high vacuum. This allows us to measure samples that are not conducting (unlike CEMS). We have another spectrometer being tested for its connection to our multipurpose growth and characterization chamber.
Fe-57 transmission, CEMS and LEEMS spectra recorded from iron-doped niobium titanium phosphorous oxide (Fe0.33NbTiP3O12). The Fe3+ doublet (dashed in the figure) is enhanced in the spectra recorded in the electron detection mode, particularly in the LEEMS case, indicating the surface sensitivity of the technique.
It consists of an horizontal UHV chamber with fast-entry loadlock with a UHV VT-STM and a XPS system, together with a LEED diffractometer and an 2-axis LN2-cooled manipulator. Both the STM and the chamber itself have been home-designed and built (most of the pieces, including the chamber, where made at SEGAINVEX, the UAM machine shop, with the most recent pieces being made at our Institute machine shop).
We also have a Physical Vapour Deposition system with three magnetrons (one of them specially designed for the deposition of ferromagnetic materials) for the growth of coatings, thin films and multilayers. The movement of the sample holder is microprocessor-controlled so that it can rotate many times, in a programmed way, forward and backwards getting to the previous position very reproducibly. This allows to grow multilayers containing many individual layers with different compositions of the individual layers
We currently have a dedicated XPS system for surface characterization comprised of:
- UHV vacuum system,
- XPS VG CLAM 3-channeltron spectrometer,
- Dual X-ray Leybold gun
- manipulator with fast-load-lock entry system
- Ar ion gun for sputter cleaning and depth profiling