Synchrotron Based Scanning Transmission X-ray Microscopy and Microspectroscopy (C-, N-, O-XANES)


BACKGROUND and PERSONAL HISTORY (from an early user's perspective): Advances in X-ray micro-focusing techniques coupled with brilliant, synchrotron derived X-ray sources have lead to the development of Scanning Transmission (Soft) X-ray Microscopy (STXM) (e.g. Jacobsen and Kirz, 1998); a technology that allows the quantitative analysis of bio-organic structure for functional group distributions at spatial resolutions approaching 25 nm. This is an organic molecular structure "ultra-microscope": how cool is that?

The particular STXM I was introduced to in 1994 was designed and built by Janos Kirz and his student Chris Jacobsen and the X-ray Optics group of the Department of Physics, SUNY Stony Brook, this was the world's first STXM (or close to it- BESSY in Germany was nearly coincident in time- but I don't think C-XANES spectroscopy was an option). Beamline X1A utilized the first undulator installed at the National Synchrotron Light Source (NSLS I), a 2.8 GeV electron storage ring. NSLS I is sadly no more, but it served a mighty purpose and has been replaced by the new State-of-the-Art NSLS II beamline. Sadly there is not currently a STXM at NSLS II (yet), a number of proposals to install such have not been successful. Note: I believe that I was the first Geoscientist to ever use STXM- (I could be wrong- so here is my first reference-Cody et al 1995: also note my actual first reference was Botto, Cody et al 1994, but my advisor at the time (Botto) claimed first authorship from me a post Doc, even as I collected these data and wrote the paper- so obviously if some other geochemist used STXM before me- their paper will precede mine -and then I am wrong!- Yes I know this is not a race- but I have heard claims of firsts in this area) so if such is the case- there is my little stake!

X1A was a triumph of vision from the Stony Brook X-ray Optics group lead by Janos Kirz, the X1A worked as a spectrometer primarily because of a small miss-parallelism of the undulator that gave a "pink-ish" beam (spanning some 30 eV) at a given gap setting- X1A was almost a spectrometer by accident. Harald Ade a student of Janos Kirz- clearly saw what a STXM could do and envisioned real improvements- a STXM by design! Note: Chris Jacobsen, also a student of Janos Kirz was also a major developer. And for the early user like me Sue Wirick was our goddess! guiding us to data unthought of and such a great collaborator.

Harald Ade realized that the soft X-ray flux off of bending magnets on a relatively small radius high energy synchrotron (the Advanced Light Source- LLBL, tight radius and third Generation 4.0 GeV!) could work if you could figure out how to capture such energy. Such synchrotron radiation off of a bending magnet is very different than what comes off of an undulator or wiggler. Harald Ade (NCSU, USA) and Adam Hichcock (McMaster Univ. Canada) both recognized the power of STXM and envisioned a dedicated "polymer" STXM and identified some open real estate on the ALS floor that was not recognized to be useful (at that time!).

This group enlisted the exquisite expertise of two amazing designer/builders: A. David Kilcoyne and Tolek Tyliszczak who both designed and built the next generation STXM and all controlling soft ware. These two built the first two premier STXM's beamline 5.3.2 and 11.0.2 at ALS. Adam Hitchcock with his son (priniciple coder developer!) developed the core analysis Soft Ware [Axis2000] - both the hardware and software are now the core of STXM today at every synchrotron world wide that is supported at beam lines in USA, Canada, Europe, and Asia. I believe that the company Xradia bought the designs for the mechanics of the ALS STXM's and these are now the standard. These beautiful instruments have so many elegant control features that I can't describe them fully here- just imagine total laser interferometry spatial location and elegant feed back control on beam drift- BEAUTIFUL! Thank David K and Tolek T. for this!

These new beamlines really opened up STXM to full capability- now you can dance across K and L edges from ~ 220 eV up to ~ 800 eV which provides a lot of real estate spectroscopy wise. A. David Kilcoyne went further expanding STXM capability with a new instrument ( that works from 600 eV to above 2000 eV- this means that with and really you can do anything (meaning just about any edge- imagine Si, Al, Na, Mg, ...etc).

International growth of STXM is also great news- Switzerland, Italy, Germany, and recently Photon Factory - Japan and Diamond in UK. These are work-horse instruments- whether on a bending magnet or a dedicated undulator, these are real data generators!!!!

Figure (Right): This is a STXM image of a very intriguing 45 million year old sample of wood from the Canadian Arctic “Fossil Forest”. The image was acquired at 285 eV. Dark =strong absorption, Light = weak absorption. Absorption intensity at this energy corresponds to aromatic carbon, e.g. in the biopolymer lignin- the bluer the more rich in aromatic carbon. The contrast reflects differences in lignin-polysaccharide abundances.

The intriguing structure shown here records an as yet unknown degradation mechanism that may not have involved fungal and/or bacteria. For scale, the width of narrowest region of the middle lamellae is ~ 50 nm. What you are looking at are collapsed cell lumens, the open cell space is gone- the sample is now solid carbon, but STXM shows that remnant cell structure remains- we use STXM to follow biochemistry far back in time. We are now focusing on structural biopolymers back to the Cambrian.

One reason I love this image is that it both reminds me that we really do not understand organic matter decomposition but also because it reminds of me etchings by the Artist M. C. Escher- who inspired me as a child. This is 45 million year old wood! Makes the head spin.

Soft X-rays are generated via perturbation of the synchrotron electron beam by an undulator, i.e. imagine a series of opposed dipolar magnets placed on either side of the electron storage ring. By varying the undulator gap X-rays in the soft X-ray range ~ 150 – 800 eV are generated. The effective energy bandwidth of the X-ray beam at X1A is on the order of 40 eV, allowing complete access to a given absorption edge region at a given undulator gap setting. Energy selection is obtained via a series of order sorting mirrors, exit and entrance slits, and a spherical grating monochromator with an energy resolution on the order of 0.03 eV (see discussions and references in Jacobsen and Kirz, 1998).

Figure (Right): This awesome image is of compression wood from red spruce. The image contrast is totally based on the concentration of aromatic carbon= Lignin. (black = epoxy) Note that the darker (purple) regions correspond to increased lignin content. This image shows that in compression wood lignin accumulates in the secondary cell wall next to the primary cell wall at concentrations as high as observed in the compound middle lamellae. The finest scale features in this image are on the order of 50 nm.

This is clearly just the tip of the iceberg of applications....

The power of soft x-ray microscopy is that image contrast is derived from chemistry by exploiting various absorption bands that exist in the pre-edge region of carbon 1s absorption edge, e.g. Carbon X-ray Absorption Near Edge Spectroscopy (C-XANES). X-rays with sufficient energy are capable of promoting core level electrons completely far from any columbic interaction with the core hole, i.e., ionization. For carbon the ionization energy threshold occurs at ~ 292-295 eV. X-rays with slightly lower energy are capable of promoting 1s electrons up to various “bound” states, i.e., unoccupied molecular and/or atomic levels. Electrons in these bound states are strongly connected to the core hole through columbic forces. The unoccupied π orbitals (π* states) of unsaturated carbon containing functional groups provide particularly intense absorption bands. These bands are shifted in energy by the electronic perturbation of neighboring atoms, e.g. the electron withdrawing nature of oxygen will impart significant energy shifts of 1s-π* transitions. Aliphatic or sp3 carbon also exhibit absorption bands corresponding (approximately) to 1s-s* transitions; these absorption bands are also shifted by electron withdrawing substituents.

Figure (Right): Note that the spatial distribution of lignin and polysaccharides can be revealed by imaging at different energies.

Figure (Right): This image reveals a thin band of cutinite (C) in a ~ 300 My old organic sediment. Cutinite is derived from the waxy outer coating of leaves. At 285 eV the cutinite is weakly absorbing due to the low concentration of aromatic carbon. At 288.1 eV (corresponding to the relatively strong 1s-3p/s*transition of aliphatic carbon) the cutin absorbs strongly relative to the surrounding vitrinite.

This was in fact the the data set that acheived by Post Doctoral Research goals- that is "to do solid state imaging of coal macerals". It was simply not possible for me (at least) do solid state NMR imaging and achieve anything like this.

Figure (Right): In collaboration with others we have been using STXM to explore the microbial degradation of lingo-cellulosic materials. This series of micro C-XANES spectra clearly reveal the selective degradation of polysaccharides relative to lignin after months of inoculation with the brown rot fungus P. placenta.

STXM is extremely powerful analytical instrumentation and the world recognizes this as STXM instruments are being implemented at all significant synchrotrons.

STXM was the reason why me and Hikaru Yabuta (now at Hiroshima U.) were brought into the NASA Stardust Investigation team and it was STXM that provided the first evidence that the mission was a success in bringing ET organics back to Earth.

As a component of my research I continue to apply STXM to a variety of scientific questions in fields such as paleobotany, organic geochemistry (kerogen evolution), cosmochemistry, and biogeochemistry. See research projects for details or call/email me.

I have worked to help develop the STXM user bases since I was first introduced to STXM in 1993.