Sof X-Ray Spectromicroscopy

47
PUB-786 UC-404 Microscale on a Materials Characterization Soft X-Ray Spectromicroscopy

description

 

Transcript of Sof X-Ray Spectromicroscopy

Page 1: Sof X-Ray Spectromicroscopy

PUB-786UC-404

Microscaleon a

MaterialsCharacterization

Soft X-RaySpectromicroscopy

Page 2: Sof X-Ray Spectromicroscopy

Cover Photo:The new Advanced Light Source building retains thedome that covered the first large accelerator at theLawrence Berkeley National Laboratory, the 184-InchCyclotron built by Laboratory founder Ernest OrlandoLawrence during World War II.

Page 3: Sof X-Ray Spectromicroscopy

A New Opportunity3

7

15

29

41

Synchrotron Radiation

Spectroscopy

Spectromicroscopy

Getting Started

Page 4: Sof X-Ray Spectromicroscopy

A N

ew O

ppor

tuni

ty

Page 5: Sof X-Ray Spectromicroscopy

MicroscopyX-Ray Microanalysis Of High-Tech Materials

X-ray microscopy at photonenergies near characteristicx-ray absorption peaks forcarbon bonds makes it pos-sible to differentiate submi-cron particles of styreneacrylonitrile (SAN) from par-ticles of urethane-basedpolyisocyanate-polyaddition(PIPA) in a polyurethanematrix. [Images made at theALS. Courtesy of A. Hitch-cock, McMaster University;E. Rightor and W. Lidy, DowChemical; and T. Warwick,ALS. Spectra taken at theNational Synchrotron LightSource. Courtesy of H. Ade,North Carolina StateUniversity.]

4

Spectro

Page 6: Sof X-Ray Spectromicroscopy

In today’s world of high-tech materials, researchersneed to examine samples on a local scale oftenmeasured in fractions of a micrometer, rather than

observing averages over large areas. The patterns ofintegrated circuits provide the most obvious example.Domains in magnetic recording media constitute another,as does the interleaved structure of metal, ceramic, orpolymer composite materials. Moreover, many of theadditives that scientists often mix in to improve theperformance of permanent magnets, superconductors,and structural materials end up distributed as second-phase particles with a different structure and compositionthan the matrix in which they are imbedded.

While x-ray absorption and photoelectron spectros-copy of atomic core levels are well-developed tools forobtaining elemental composition, chemical state, andstructural information from complex materials and theirsurfaces, they have not traditionally provided significantspatial resolution. But now, owing to the dramaticallyenhanced brightness of the newest synchrotron radiationsources, such as the Advanced Light Source (ALS) at the

Lawrence Berkeley National Laboratory, researchers areable to carry out these x-ray spectroscopies on a micro-scopic scale (spectromicroscopy), thereby providing thespatial resolution required to address a wide range ofmaterials microcharacterization problems, includinganalysis of the intricate sub-micrometer features on amicrocircuit chip, in-situ chemistry and metallurgy ofcomposites, and measurements of spatially inhomoge-neous chemical reactions.

At the ALS, we are looking forward to a wealth of newapplications resulting from the union of established x-rayspectroscopy techniques with imaging. In the followingpages, we will specifically illustrate the potential of soft(long-wavelength) x-ray spectromicroscopy for providingspatially resolved information about materials andsurfaces. Along the way, we will review the properties ofsynchrotron radiation that make it suitable for this kind ofmaterials microcharacterization, summarize spectroscopytechniques, and describe how spatial resolution andimaging are achieved. We will finish by describing how toobtain information about the ALS and how to initiate a

research program.For a summary of the

spectromicroscopy facili-ties currently available atthe ALS, please turn to theloose-leaf pages inserted inthe pocket of the insideback cover of this bro-chure.

The Advanced Light Source(ALS) at the Lawrence Berke-ley National Laboratory islocated in the hills above Ber-keley, California, adjacent tothe campus of the Universityof California, in the midst ofthe varied high-tech industrialinfrastructure of the SanFrancisco Bay Area.

5

Page 7: Sof X-Ray Spectromicroscopy

Sync

hrot

ron

Rad

iati

on

Page 8: Sof X-Ray Spectromicroscopy

BrightnessThe Competitive Frontier ofSynchrotron Radiation

The locations of existing andplanned ultrabright third-generation synchrotronsources in countries aroundthe world illustrate the im-portance of state-of-the-artmaterialsmicrocharacterization tools inan economically competitiveglobal economy.

8

Page 9: Sof X-Ray Spectromicroscopy

In the 1960s, scientists interested in the propertiesof materials and surfaces began to exploit the ultra-violet and x rays emitted by electrons in synchrotrons

built and operated for high-energy physics research. Bythe early 1980s, new facilities based on electron storagerings were springing up in the major industrializednations. Not only were these dedicated to synchrotronradiation, but many were also designed to achieve muchhigher brightness than previously obtainable.

Users of synchrotron radiation now recognize thatbrightness—defined as the flux of photons per unitsource area and per unit solid angle of emission—is acritical performance specification (see box on page 10).In short, with a high-brightness source, x-ray opticalsystems can use the photons efficiently to get the highestpossible flux onto the sample. For these reasons, within adecade, work began on the third generation of facilities,with great expectations for benefits from still higherbrightness.

These performance curvesdemonstrate the orders-of-magnitude jump in the bright-ness available from the ALSin the soft x-ray spectral re-gion (and those of its sisterfacility, the APS, in the hardor short-wavelength x-rayspectral region) over thatavailable from second-gen-eration synchrotron sourcesand from conventional con-tinuum and line sources inthe laboratory.

The ALS is one of the first third-generation facilities,with construction completed in March 1993. Designedexplicitly for the production of soft x-ray and ultravioletsynchrotron radiation with the highest possible bright-ness, the ALS is now being operated by the LawrenceBerkeley National Laboratory of the University ofCalifornia for the U. S. Department of Energy as anational user facility that is available around the year forresearch by scientists from industrial, academic, andgovernment laboratories.

The ALS organization is committed to the rapiddevelopment and application of spectromicroscopytechniques that exploit the brightness of the source incollaboration with academic and industrial users. Theindustrial connection to synchrotron radiation is particu-larly important, and it is one reason countries around theworld are building high-brightness synchrotron sources.In addition to two in the United States (the ALS and theAdvanced Photon Source at Argonne National Laboratorynear Chicago), Europe and Asia together have five inoperation, and more are planned or under construction.

9

Page 10: Sof X-Ray Spectromicroscopy

10

The statistical-mechanical concept of phase space applies to thesource of the synchrotron radiation (the electron beam) and to theradiation itself. For electrons, the phase-space area (the emittance)is the product of the beam size and divergence. Emittance is some-what like a temperature, so the ALS storage ring is said to have anultralow-emittance or ultracold beam. The emittance sets a lowerlimit for the phase-space area of the light beam, defined as the prod-uct of the effective source size and radiation cone angle. Brightnessis the density of photons in this phase space. Although optical sys-tems can manipulate the source size and angle (e.g., by focusing),they cannot reduce their product without losing some of the light, aswith an aperture, and therefore cannot increase the brightness. Op-tical systems that transport the radiation in general collect light inonly a portion of the phase-space area of the incident radiation, Theuseful phase-space area (the acceptance) is given by the product ofthe maximum source size and radiation angle that the system cancollect. Similarly, there is a sample acceptance defined by the spotsize and beam divergence required at the sample. Since the highestflux is delivered to the sample when the phase-space area of thephoton beam matches the acceptances of the optical system andthe sample, brightness is almost always more important than fluxalone.

Phase Space & Brightness

The key parameter of the photon source for spectromicroscopy is its bright-ness. For a photon beam with a Gaussian density distribution, the bright-ness B(ε) on the optical axis of an undulator is approximately

where F(ε) is the on-axis flux for photon energy ε, Σh and Σv are the RMShorizontal and vertical source sizes, and Σh’ and Σv’ are the RMS horizontaland vertical radiation-cone half-angles.

The electron beam contributes to both the effective source size and diver-gence. For a single electron in an undulator radiating into a cone with anRMS half-angle σr’, there is a corresponding wavelength-dependent, diffrac-tion-limited source size σr = λ/4πσr’. The effective source size Σh or Σv isthe quadrature sum of σr and the size σh or σv of the Gaussian electronbeam [e.g., Σh = (σh

2 + σr2)1/2]. Similarly, the effective divergences Σh’ and

Σv’ are obtained from the divergence σr’ of the photon beam from a singleelectron and the divergences σh’ and σv’ of the electron beam.

Third-generation synchrotron light sources optimize the brightness bygenerating a beam with a small emittance (product of the electron-beamsize and divergence). If the emittance is small enough relative to the wave-length λ, the radiation is diffraction limited (i.e., the phase space occupiedby the electron beam is less than the phase space occupied by a diffraction-limited photon beam)

εh = σh × σh’ < σr × σr’ = λ/4π εv = σv × σv’ < σr × σr’ = λ/4π.

Under these conditions, the radiation has the spatial (transverse) coherenceproperties of a laser and it is possible to collect all of the light and focus itto the smallest possible size without loss. At the ALS, the vertical emittanceεy is about 10-10 m·rad (1 Å·rad) while the horizontal emittance is about4 × 10-9 m·rad. This means that the ALS has laser-like properties in thevertical direction at wavelengths as short as 10 Å.

A Small Beam Emittance Maximizes The Brightness

Page 11: Sof X-Ray Spectromicroscopy

Strobe-light simulation of theelectron beam in the ALSimmediately suggests thelaserlike qualities of theultrabright light produced bythe beam.

11

Page 12: Sof X-Ray Spectromicroscopy

SourcesThe Emphasis Is On Undulators

Third-Generation

The undulator sources of theALS are carefully engineeredarrays of permanent magnetsthat produce precisely peri-odic magnetic fields to gener-ate the brightest possiblebeams of soft x rays.

12

Page 13: Sof X-Ray Spectromicroscopy

In the first synchrotron sources, the radiation camefrom the bend magnets in the curved sectors of anelectron accelerator. Relativistic electrons emit

synchrotron radiation when a magnetic field bends theminto a curved trajectory. At any point on the trajectory,the synchrotron radiation emerges in a narrow conetangential to the path of the electron with a flux andspectral range that depends on the electron energy andthe magnetic field. As the electron sweeps around thecurve, it generates a horizontal fan of light. In practice,an aperture determines how much of the horizontal fanis collected (typically, up to several milliradians). For theALS, the highest beam energy is 1.9 GeV, where thebend-magnet field is about 1.35 Tesla, so that usefulfluxes are available at photon energies above 10 keV.

In third-generation light sources, the storage ring isspecifically designed to include special magneticstructures known as insertion devices (undulators andwigglers). Although designs differ, the most commoninsertion device comprises a linear arrayof dipole magnets withalternatingpolarity

(i.e., N-S-N-S and so on). The array generates a sinusoi-dal vertical field that drives an electron into an oscillatingtrajectory in the horizontal plane with the same period asthe field. Each dipole is a bend-magnet source radiatingalong the axis of the undulator.

The defining feature that separates an undulator froma wiggler is that the maximum angular excursion of anelectron relative to the axis of the device is less than thenatural opening angle of the synchrotron-radiation cone,so that the radiation emitted from each pole interferesconstructively. As a result, the broad radiation spectrumfrom individual bends squeezes into a series of sharpspectral peaks comprising a fundamental and a set ofharmonics. Mechanically opening and closing the verticalgap between the undulator poles adjusts the undulatorfield and thereby the photon energies (wavelengths) atwhich the peaks occur. For the values of the periods (5 to10 cm) and maximum fields (less than 1 Tesla) of theALS undulators, the spectral range is currently fromabout 5 eV to 1500 eV.

At the ALS, the number of poles in an undulatorranges from about 80 to about 180, resulting in a much

larger flux, as compared to a bend magnet. This,combined with the low emittance of the electron

beam, is the source of the high brightness ofthe ALS undulators.

The ALS comprises an accel-erator complex (electron gun,linear accelerator or linac,booster synchrotron, andstorage ring), an experimen-tal area (beamlines and ex-perimental stations), and abuilding to house them. Thebrightest sources of synchro-tron radiation—and thereforebest forspectromicroscopy—are themagnetic structures calledundulators that reside in thestraight sections of thestorage ring.

13

Page 14: Sof X-Ray Spectromicroscopy

Spec

tros

copy

Page 15: Sof X-Ray Spectromicroscopy

ALSNew Power & Versatility For X-RaySpectroscopy

The

The magnetic force betweenthe upper and lower rows ofmagnets in this ALS undula-tor ranges up to 40 tons, yetthe placement of the magnetpoles must be accurate tobetter than 20 micrometersover the 4.5-meter length ofthe device in order to achievethe highest brightness.

16

Page 16: Sof X-Ray Spectromicroscopy

In addition to being at least 100 times brighter thansources available before, the ALS offers the tradi-tional virtues associated with synchrotron radiation as

compared to conventional laboratory sources of x rays:high flux and variable (tunable) photon energy, as well astightly collimated beams and controllable polarization.These properties make the ALS an ideal source for softx-ray spectroscopy based on the atomic core levels incomplex materials.

Since the presence or absence of long-range order doesnot strongly affect core levels, they are particularly suitedfor probing short-range order and local properties (e.g.,atomic coordination and oxidation states). Because oftheir localized nature, they inherently provide elementalidentification in spectroscopy experiments. Complexmaterials can be dissected sequentially by tuning to theabsorption edges of the constituent elements.

Among core-level spectroscopies, laboratories through-out the world use x-ray photoelectron spectroscopy (XPS)as an analytical technique for materials characterization.

Typical XPS instruments available commercially includea Mg- or Al-Kα x-ray source, filtered to send a monochro-matic beam into a spot possibly as small as 30 microme-ters in diameter. In addition to elemental analysis, XPScan also determine the valence state and bonding envi-ronment of atoms near a sample surface; it can identifyorganic functional groups of polymers; and it can charac-terize very thin, layered structures. Already a powerfultechnique in the laboratory, XPS has greatly expandedcapabilities when performed at a modern synchrotronradiation source, such as the ALS, where the high flux,high spectral resolution, and tunable photon energies canbe fully exploited.

Another spectroscopic technique based on excitationof electrons from core levels, near-edge x-ray absorptionfine structure spectroscopy (NEXAFS), provides spectrawith signatures characteristic of the chemical bonding ofthe element excited by the x rays. It is easy to performwhen the x-ray source is tunable. Consequently, the highflux and tunable photon energy available at the ALS not

only make this technique practical butextend its capabilities to include real-time studies of the dynamics of localizedsurface chemical changes. And because itmakes efficient use of every photon andthereby minimizes radiation damage,NEXAFS is also ideal for studies of largeorganic molecules.

Undulator beamline 7.0 atthe ALS serves multiple ex-perimental stations (fore-ground and immediatelybehind) capable of NEXAFSand XPS spectroscopies.

17

Page 17: Sof X-Ray Spectromicroscopy

SpectroscopyHigh Flux, High Resolution, & TunablePhoton Energy

Photoelectron

Photoelectron diffractiongives the atomic geometry ofa monolayer of iron depositedon a tungsten (110) surface.The XPS spectrum for tung-sten 4f photoelectrons hasthree primary components:one peak due to atoms at theinterface next to iron, a sec-ond peak due to next-neigh-bor atoms in the second layerbelow the iron, and a weakfeature due to deeper atoms.Comparison of experimentaldiffraction patterns, such asthose shown for interfacialand next-neighbor atoms,with calculated diffractionpatterns for candidate struc-tures shows that the iron sitsin a bridge site 2.16 Å abovethe interfacial tungsten at-oms. [Data taken at the ALS.Courtesy of E. Tober, IBMAlmaden Research Center,and C. Fadley, University ofCalifornia, Davis.]

18

Page 18: Sof X-Ray Spectromicroscopy

The intense flux of tunable synchrotron radiation thatthe high-brightness undulator beamlines at the ALSsupply to a sample brings enormous benefits to x-

ray photoelectron spectroscopy (XPS). For example, withthe substantial count rates resulting from high flux,researchers can exploit the fine energy resolution avail-able from ALS beamlines (typically 1 part in 5000). Highresolution facilitates fitting complicated multiple spectralpeaks for quantitative analysis.

The photon-energy tunability of synchrotron radiationprovides additional benefits because the cross section forx-ray absorption is much higher when the photon energyis near an absorption edge, resulting in a significantlyenhanced photoemission signal that can make marginalexperiments practical. For example, the cross section forabsorption by carbon 1s electrons is about 100 timeslarger for excitation just above the carbon K edge with320-eV photons than with aluminum-Kα radiation.

Still another aspect of tunability is the ability to adjustthe surface sensitivity of the experiment. X rays canpenetrate deeply into a solid before being absorbed, sothat a photoelectron has some distance to travel in order

to escape from the surface. The distance the photoelec-tron can travel with no energy losses due to scatteringdepends on its kinetic energy. The minimum escapedepth is a few angstroms for kinetic energies near 50 eVand is deeper at both higher and lower energies. Experi-menters using synchrotron radiation can therefore tunethe x-ray photon energy to produce photoelectrons fromvery near the surface or farther down in the bulk, afeature that turns core-level photoelectron spectroscopyinto a technique with variable depth sensitivity.

These features combine to enhance studies of surfacestructure by means of the angular distribution of theintensity of a peak in the photoelectron spectrum. Thisdistribution constitutes a diffraction pattern of the atomssurrounding the emitter (photoelectron diffraction).Selecting the emission energy isolates a particular atomicspecies, and binding-energy shifts due to chemicalbonding or lattice location can further specify the emit-ting atom. The angular-distribution data then contains animmense amount of information about the crystalstructure close to the surface.

Some photoelectrons losekinetic energy on their way tothe surface because of in-elastic collisions. The meanfree path between inelasticscattering events for a photo-electron depends on its ki-netic energy in much thesame way for elements andinorganic compounds. Thisfunctional dependence com-bined with tunable photonenergies provides a way tocontrol the depth below thesurface from which electronscontributing to the XPS spec-trum come. [Adapted fromC.R. Brundle, C.A. Evans, Jr.,and S. Wilson, eds., Encyclo-pedia of MaterialsCharacterization,Butterworth-Heinemann, Stoneham, MA,1992, p. 293.]

19

Page 19: Sof X-Ray Spectromicroscopy

This schematic diagramshows that peaks in the pho-toelectron spectrum of a me-tallic solid correspond to thequantum states from whichthe electrons are emitted.The diagram shows corestates in atomic potential-energy wells, the portion ofthe valence band occupied byelectrons (i.e., below theFermi level), and a state in apotential-energy well associ-ated with the surface. Thereis also a large secondary-electron tail due toinelastically scattered elec-trons. [Adapted from N.V.Smith and F.J. Himpsel, “Pho-toelectron Spectroscopy,” inE.-E. Koch, ed., Handbook ofSynchrotron Radiation, Vol.1B, North-Holland PublishingCompany, Amsterdam, 1983,p. 908.]

20

Page 20: Sof X-Ray Spectromicroscopy

21

In x-ray photoelectron spectroscopy (XPS) of solids, an electron ex-cited from a core level by x-ray absorption makes a transition to anunoccupied state at energies above the ionization threshold. Fromthere, the electron can migrate to the surface and escape into thevacuum. If the electron suffers no collisions on the way, its kineticenergy EK on escaping is related to the binding energy EB (minimumenergy to escape) of the core level and the photon energy ε

EK = ε - EB.

This relation shows that, for a fixed photon energy, the spectrumof photoelectron kinetic energies measured with an electron-energyanalyzer reflects the distribution of occupied core states. Severalother effects also contribute to the spectrum—Auger-electron emis-sion occurs at a fixed energy independent of the photon energy whenan electron from a higher lying state fills the core hole; satellite pho-toelectron peaks appear at lower kinetic energy than the main peakwhen excitation of a second electron to an unoccupied state belowthe ionization threshold drains away some of the incident photonenergy; and the large background at low kinetic energies is due tomultiply scattered (secondary) electrons.

To a first approximation, the binding energy of the core level isindependent of the environment, so it identifies the atomic species.However, there are small “chemical shifts” in the binding energy thatdepend on the local chemical bonding. For example, at the interfacebetween silicon and silicon dioxide, the binding energy of the impor-tant silicon 2p3/2 core level varies approximately linearly with thenumber of oxygen atoms bound to silicon, shifting about 1 eV peroxygen atom from the 99.2-eV binding energy for pure silicon. It isthese energy shifts that provide chemical-state information.

How XPS Provides Chemical Specificity

Page 21: Sof X-Ray Spectromicroscopy

AbsorptionA Chemical Fingerprint From Near-Edge Spectra

This schematic diagram of adiatomic molecule shows thatpeaks observed in the near-edge x-ray absorption spec-trum correspond tounoccupied valence states towhich core electrons are ex-cited. These valence statesmay be modified by crystal-field effects in the solid stateand by their participation inchemical bonding. TheNEXAFS spectrum is thus astriking signature of thestructural environment andchemical state of the absorb-ing atom. [Adapted from J.Stöhr, NEXAFS Spectroscopy,Springer-Verlag, Berlin,1992, p. 85.]

X-Ray

22

Page 22: Sof X-Ray Spectromicroscopy

Chemical information can be derived from thedistinctive spectral features of the absorptioncross section in the region of an x-ray absorption

edge. This technique is known as near-edge x-rayabsorption fine-structure spectroscopy (NEXAFS, butsometimes called XANES for x-ray absorption near-edgespectroscopy).

If the excitation energy is near an absorption edge, anelectron excited from a core level by x-ray absorptioncan make a transition to an unoccupied valence state atenergies near the ionization threshold. If the lifetime ofthe core hole is long, then the intrinsic energy width of acore level is narrow, and the absorption spectrum (peakposition, shape, and intensity) maps the energy distribu-tion of unoccupied states to which the electron isexcited. As it happens, most of the sharpest core levelsof interest for NEXAFS spectroscopy have bindingenergies well within the VUV and soft x-ray range of theALS undulators.

The density of states is a signature of the chemicalbonding and surrounding crystal structure of theabsorbing atom. For example, in organic compounds,there may be a distinctive spectrum due to strong carbonabsorption into orbitals determined by the chemical

bonds in which the carbon atoms are participating. Intransition metals, local symmetry (e.g., octahedral ortetrahedral) and the strength of the crystal field stronglyaffect the absorption. The near-edge absorption thereforeserves as a chemical and structural fingerprint.

Measurement of the transmitted intensity is the mostobvious way to do absorption spectroscopy, if the sampleis thin enough. The penetration depth for soft x-rayphotons is usually in the range 0.1 to 1 µm, so transmis-sion measurements give information about the interior ofmaterials. For thicker samples, one can monitor theemission of either electrons or fluorescence photons froma solid surface, both of which are a measure of absorptionprobability.

A simple photocurrent measurement of the total yieldof electrons (sum of photoelectrons, Auger electrons, andscattered secondary electrons of a few electron volts)monitors electron emission, whereas an energy-resolvingx-ray detector directly measures the fluorescence.Electron emission and fluorescence have differentsampling depths, typically 5 nm for total electron yieldand up to 1 µm for fluorescence. One can enhance thesurface sensitivity by biasing the sample to eliminate thelow-energy electrons (partial electron yield) or by moni-toring the Auger electrons directly with an electron-energy analyzer.

NEXAFS spectra of organicmolecules at the carbon Kedge show peaks characteris-tic of the molecular orbitalsbonding the carbon atoms.Absorption at different siteswith correspondingly differentbonding yields spectral signa-tures of the various mol-ecules. The spectrum ofpolyethylene-terephthalate(PET) is shown here. As adatabase of characteristicspectra builds up, theNEXAFS technique is becom-ing a very important tool forpolymer analysis. [Data takenat the National SynchrotronLight Source. Courtesy of H.Ade, North Carolina StateUniversity.]

23

Page 23: Sof X-Ray Spectromicroscopy

PolarizationEnhancing The Capability of X-RaySpectroscopy

In an elemental ferromagnet,the differential absorption ofleft and right circularly polar-ized x rays propagating paral-lel to an applied magneticfield results from an imbal-ance in the spin occupancy ofthe partially occupied valenceband and from quantum-me-chanical selection rules thatapply to the absorption tran-sitions. From the dichroism,it is possible to extract theseparate spin and orbitalcontributions to the totallocal magnetic moment. Thespin density mD in the figureis an orientation-dependentterm that vanishes in isotro-pic materials and certainexperimental geometries.

24

Page 24: Sof X-Ray Spectromicroscopy

Synchrotron radiation is naturally linearly polarizedin the horizontal plane of the electron orbit. Sincequantum-mechanical transition probabilities

depend on the relative orientation of the electric-fieldvector of the exciting radiation and the anisotropic wavefunctions of the quantum states involved, polarizationprovides symmetry selectivity that is lacking in processesexcited by unpolarized light. For example, in the case ofcore-level excitation of molecular species, the absorptionis proportional to the cosine squared of the angle betweenthe electric-field vector and the orbital to which theelectron is excited. Scientists have used this relation withNEXAFS spectroscopy to determine the orientation ofmolecules chemisorbed on a solid surface and theorientation of polymer fibers in complex structures.

The use of circularly polarized synchrotron radiation isgrowing in materials research to study spin-dependentmagnetic phenomena. Experimental techniques at theALS that make use of circularly polarized x rays are oftenbased on NEXAFS spectroscopy. In x-ray magneticcircular dichroism spectroscopy (XMCD), one measuresthe absorption spectra in a magnetized sample using left-and right-circularly polarized x rays. For example, in the

elemental ferromagnets iron, nickel, and cobalt, thevalence band splits with the majority electrons (thosewhose spin magnetic moment is parallel to the appliedmagnetic field) lying lower in energy relative to theminority electrons with antiparallel spin, so that theunoccupied states in the valence band are primarily ofminority spin. Together with the angular-momentumselection rules appropriate for circularly polarized x rays,this means that the absorption cross section for excitationof electrons from core levels into the valence banddepends on the direction of the field and the helicity ofthe photons.

Analysis of the difference between the absorptionspectra for reversed helicities or field orientations (thedichroism) yields element-specific magnetic properties,such as the size and orientation of magnetic moments. If agiven element exists in more than one oxidation state orsite symmetry, the spin orientations of these individualspecies may be distinguishable. There is also the possibil-ity of extracting the separate spin and orbital contribu-tions to the total local magnetic moment. With the use ofpolarizing optics and linearly polarized synchrotronradiation, similar information about magnetic materials

comes from measuring the rotation ofthe plane of polarization of soft x raystransmitted through (Faraday rotation) orreflected from (Kerr rotation) samples inmagnetic fields.

Linearly polarized synchro-tron radiation is a useful toolfor determining the orienta-tion of adsorbed molecules ona surface. Here, the verticalorientation of a carbon mon-oxide molecule on a molybde-num surface is verified by thestrong absorption due to theπ* orbital when the electricvector is parallel to the sur-face and by the absorptiondue to the σ* orbital whenthe electric vector is nearlynormal to the surface.[Adapted from J. Stöhr,NEXAFS Spectroscopy,Springer-Verlag, Berlin,1992, p. 171.]

25

Page 25: Sof X-Ray Spectromicroscopy

Schematic diagram of an el-liptically polarizing perma-nent-magnet undulator showshow mechanically adjustingthe relative positions of themagnet rows controls thepolarization. The arrows indi-cate the direction of magneti-zation in each block, so thatfour blocks constitute oneperiod λu. The displacementD between the upper frontand lower rear rows relativeto the upper rear and lowerfront rows gives a rowphase = D/λ u. The degree ofpolarization depends on thephase shift.

26

Page 26: Sof X-Ray Spectromicroscopy

27

Bend-magnet radiation out of the horizontal plane of the electronorbit is elliptically polarized; that is, the electric-field vector consistsof horizontal and vertical components that are 90˚ out of phase. Therelative amplitudes of the horizontal and vertical components dependon the observation angle. In the horizontal plane, the vertical compo-nent is zero, so that bend-magnet radiation is linearly polarized withthe electric-field vector in the plane. As the observation angle aboveor below the plane increases, the amplitude of the vertical field com-ponent approaches that of the horizontal component, but the inten-sity of the radiation decreases. In practice, one chooses anobservation angle that maximizes the product of the square root ofthe flux and the degree of circular polarization. At this maximum, thedegree of circular polarization is typically about 70 percent. Thesense of the polarization can be reversed by changing the viewingangle from above to below the plane and vice-versa by means ofmovable apertures.

To increase the flux of circularly polarized photons over that avail-able from a bend magnet, the ALS is developing an elliptically polariz-ing undulator (EPU). In the usual linear undulator, the circularcomponents of the radiation from successive poles reverse direc-tion, so the net effect is linearly polarized radiation. In the EPU, fouridentical rows of permanent-magnet blocks are arranged in twopairs above and below the plane of the electron orbit. If the upperfront and lower rear rows (or upper rear and lower front rows) aremoved longitudinally in the same direction by the correct distancewith respect to the fixed rows, a helical magnetic field—and henceelectron trajectory—is created. Moving the rows in the oppositedirection creates a field of the opposite helicity. This motion istermed a row phase shift. Depending on the phase shift, the undula-tor radiation is polarized linearly in the vertical direction, right or leftcircularly polarized, or linearly polarized in the horizontal plane as ina conventional undulator.

How To Obtain Polarized Synchrotron Radiation

Page 27: Sof X-Ray Spectromicroscopy

Spec

trom

icro

scop

y

Page 28: Sof X-Ray Spectromicroscopy

CombinationPutting X-Ray Microscopy &Spectroscopy Together

A New

XPS imaging of a very smallamount of plutonium oxide(4 µg) demonstrates thatspectromicroscopy can ana-lyze otherwise hazardous ma-terials without specialenvironmental chambers.Here, imaging is accom-plished using an x-ray spot50 µm in diameter by movingthe substrate in the fixedx-ray focal spot. XPS spectraof the plutonium 4f electronsused in the imaging showoxide chemical shifts.NEXAFS spectra also showthe plutonium 5d edge. [Datataken at the ALS. Courtesy ofD.K. Shuh, Lawrence Berke-ley National Laboratory, B.P.Tonner, University of Wiscon-sin–Milwaukee, S.D. Kevan,University of Oregon, andT. Warwick, ALS.]

30

Page 29: Sof X-Ray Spectromicroscopy

Owing to the dramatically enhanced brightness ofthe newest synchrotron radiation sources, such asthe ALS, x-ray spectroscopy and microscopy can

now be combined. Spectromicroscopy provides all thepower of x-ray spectroscopic techniques (e.g., chemicalsensitivity, variable excitation energy, minimum radiationdamage, and variable polarization) with spatial resolutionranging from around one micrometer to as fine as a fewnanometers. With soft x-ray spectromicroscopy at thesesmall length scales, scientists can address a very widerange of materials microcharacterization problemsinvolving complex microstructured systems and surfaces.

For example, the semiconductor industry now fabri-cates devices with features a fraction of a micrometer insize, and the ability of x-ray spectromicroscopy to performchemical analysis at this spatial scale is attracting atten-tion. Similarly, materials analysts can use these tech-

This “phase space” diagramshowing the spectral andspatial resolutions achievablewith electron-beam and x-rayanalytical techniques high-lights the ability of soft x-rayspectromicroscopy to obtainspatially resolved chemicalinformation.

niques to examine the chemical, structural, and magneticproperties of small-scale structures in magnetic recordingmedia and devices. And they can investigate compositematerials with phases of the order of 1 µm in size, evenwhen radiation-sensitive, organic components areinvolved.

At the ALS, near-edge x-ray absorption fine-structurespectroscopy (NEXAFS) and x-ray photoelectronspectroscopy (XPS) are being developed as spectromi-croscopies in the soft x-ray region of the spectrum formaterials microcharacterization. Although soft x-rayspectromicroscopy is in its infancy, microscopes at theALS have already demonstrated 100-nm resolution in theearly stages of a continuing program aimed at the devel-opment of several instruments and techniques residingon multiple beamlines with a spectrum of capabilities.

31

Page 30: Sof X-Ray Spectromicroscopy

Versatility & High Spatial Resolution

MicroscopyScanning

In this example of NEXAFSspectromicroscopy, a sectionof polyacrylonitrile fiber about10 µm in diameter is ana-lyzed by x-ray absorption nearthe carbon K edge. Imagestaken after a heat-treatmentprocess to enhance the fire-retarding capabilities of thismaterial show contrast dueto the distribution of carbon-nitrogen and carbon-oxygenbonds (peaks A, B, and C).Annealing in air causes triplebonds (C≡N), which exist uni-formly throughout the homo-geneous PAN fiber beforetreatment, to become de-pleted in the interior, givingrise to a rim structure. [FromB.P. Tonner, et al., J. Elect.Spectros. Rel. Phenom., 75,309–332 (1995).]

32

Page 31: Sof X-Ray Spectromicroscopy

In a scanning x-ray microscope, optics focus the x-raybeam, and a detector senses photons or electrons asthe x-ray spot moves across the sample surface (or as

the sample rasters through a fixed x-ray spot). In eithercase, the image builds up pixel by pixel during a scan thattypically lasts about a minute at the ALS. The mechani-cal design of the sample stage fixes the spatial scanningrange. For example, the scan range of piezo drives limitsthe maximum scanned areas to about 100 µm × 100 µm.However, the image field can be positioned anywherewithin a much larger area. The primary advantages ofscanning microscopy are its guaranteed spatial resolutionand the versatility resulting from the variety of usefulsignals.

The size of the focused spot determines the spatialresolution of the instrument. Spot sizes of the order of100 nm are routine. Achieving such a small spot dependson special diffraction-limited x-ray optical systems. Inparticular, scanning microscopes have been implementedusing both Fresnel zone-plate lens systems, whosefocusing action is like that of a circular diffraction grat-ings, and reflective focusing from spherical mirrors in theSchwarzschild configuration. Because of the low emit-tance of the ALS storage ring, undulator radiation comes

from a nearly diffraction-limited point source, resulting inacceptance of a large fraction of the x rays by the diffrac-tion-limited optics. The low-emittance also makes itpractical to obtain significant coherent flux from bendmagnets by means of apertures.

Images can be obtained from any signal as a functionof position on the sample surface. This signal could bethe count rate of transmitted photons, the total or partialyield of photoelectrons, the yield from a particularphotoelectron peak in an XPS spectrum , or the yield offluorescence photons at peaks in a NEXAFS spectrum.Each type of signal will give different elemental, chemi-cal, magnetic, or surface/bulk information. Scanningspectromicroscopy has been applied to many materials,including biological systems, organic polymer blends,magnetic media, and semiconductors.

Scanning microscopy is a serial acquisition methodwith inherent speed restrictions. A typical scenario mightconsist of a conventional spectroscopic examination tolocate characteristic spectral features, followed by amapping of the distribution of those features at coarseresolution, a finer mapping of areas of interest, and adetailed spectroscopic examination of the most importantfeatures.

Transmission NEXAFS imagesof a sectioned Kevlar fiberwith radially oriented hydro-gen-bonded sheets betweenadjacent polymer chainsshows the sensitivity to therelative orientation betweenthe x-ray polarization planeand the π* molecular orbitalsof nitrogen in the molecule.Nitrogen 1s electrons can beexcited into these orbitalsonly when the orbitals areoriented parallel to a compo-nent of the electric field ofthe linearly polarized light.The magnitude of the parallelcomponent rises and fallsaround the fiber, giving riseto the distinctive image.[Data taken at the ALS. Cour-tesy of H. Ade, North Caro-lina State University, and B.P.Tonner, University of Wiscon-sin–Milwaukee.]

33

Page 32: Sof X-Ray Spectromicroscopy

Fresnel zone plates arediffrac-tive devices that canfocus the x-ray source to aspot size approximately equalto the width of the outermostzone. As a result of theirproven record of achievinghigh spatial resolution andthe flexibility of multiple de-tection methods, they arewidely used in x-ray micro-scopes at several synchro-tron radiation facilities.

Comprising a central opaque zone surrounded by alternating trans-parent and opaque zones of increasing radius and decreasing width,zone plates are circular diffraction gratings with a diffraction angleincreasing linearly with radius, so that they focus incident radiationto a small spot. For a lens of diameter D with a total of N opaqueand transparent zones and a numerical aperture NA, the focal lengthf is given by

f = D/2(NA) = 4N(∆r)2/λ,

where ∆r is the width of the outermost opaque zone. Lenses with ∆ras small as 30 nm and N equal to several hundred have been made.The depth of focus ∆f (the longitudinal distance occupied by the fo-cused spot) is also wavelength dependent

∆f = 1.22λ/(NA)2 = 4.88(∆r)2/λ,

The finest spatial resolution δ of a zone-plate with monochromaticand spatially coherent illumination and perfect zone placement is

δ = 0.61λ/NA = 1.22∆r.

It is anticipated that, with improvements to zone-plate fabricationtechnology based on electron-beam lithography, resolutions of 25nm or less will ultimately be possible. Although zone plates can focusany wavelength, the amount of light they collect varies as λ2, so thatsignal intensity is highest at long wavelengths and low photon ener-gies. Phase-contrast zone plates are also available with transmissivezones designed to give the appropriate phase shift at a specificwavelength for higher efficiency.

Fresnel Zone-Plate Lenses Offer High Resolution & Tunability

34

Page 33: Sof X-Ray Spectromicroscopy

Mirrors in the Schwarzschildgeometry can focus the x-raysource to a submicron spot.Multilayer interference coat-ings enhance the normal-incidence reflectivity at softx-ray energies. The longworking distance of this ob-jective lens is conven-ient forXPS scanning spectromicros-copy.

35

The Schwarzschild objective comprises two concentric sphericalmirrors, the first presenting a convex face to the incident radiationand the second a concave face, so that the net effect is to focus thelight to a small spot. Since the normal-incidence reflectivity of mirrormaterials at soft x-ray wavelengths is small, recourse is made tomultilayer coatings. These comprise alternate layers of x-ray trans-mitting and absorbing materials, resulting in a coherent addition ofthe reflectivities at each interface. The increased reflectivity, whichmay be as high as 60% for each mirror, comes at the expense of alimited bandwidth of a few percent around a central wavelength givenby a Bragg equation for reflection from the layers

λ = 2d sinθ,

where d is the layer spacing and θ is close to 90°. The limited band-width for a coating with N layers (∆λ/λ ≈ 1/N) means that multipleobjectives are needed for work over a range or wavelengths. Multi-layer coatings can be designed with a variety of materials and spac-ings for a wide range of wavelengths, but the reflectivity decreasesat shorter wavelengths, so that a Schwarzschild instrument is bestsuited for valence-band spectromicroscopy up to a photon energy ofabout 200 eV. Typical systems operate at 95 or 130 eV.

The Schwarzschild objective has a large numerical aperture, whichallows a large amount of light to be collected and results in a longworking distance. The diffraction-limited image size δ is given byδ = 0.61λ/NA, where for the current-generation x-ray microscopes,NA is about 0.2. With this aperture and a wavelength of 13 nm, aresolution of about 90 nm has been experimentally measured.

Schwarzschild Objectives—Efficient At Lower Photon Energies

Page 34: Sof X-Ray Spectromicroscopy

ImagingSpectromicroscopy In Real Time

Thin, hard coatings based onsp3 bonding in carbon havemany uses. Here, micron-sized particles of chemicalvapor-deposited diamond ona silicon substrate are easilyseen in the X-PEEM imagetaken at 292 eV near thecarbon K edge. Moreover,NEXAFS spectra from theparticles reproduce thestructure seen in spectra fordiamond. In particular, theydo not have the characteris-tic resonance below the Kedge associated with sp2

bonding in graphite and othercarbon forms but absent inthe purely sp3 bonding of dia-mond. [Data taken at theStanford Synchrotron Radia-tion Laboratory. Courtesy ofA. Cossy, S. Anders, and A.Garcia, ALS. [Referencespectra from J.F. Morar etal., Phys. Rev. Lett. 54,1960 (1985) and L.J.Terminello et al., Chem. Phys.Lett. 182, 491 (1991).]

Full-Field

36

Page 35: Sof X-Ray Spectromicroscopy

For materials microcharacterization, imaging spectro-microscopy is based on electron optical techniques.In this approach, photon optics condense the beam

to the desired field of view, and an electron microscopeimages the sample surface using photoemitted electrons.The first photoelectron microscopes (UV-PEEM) used anultraviolet source and imaged the electrons with a seriesof electrostatic lenses, obtaining both topographical andwork-function contrast.

Using higher energy photons from a synchrotron lightsource to excite core levels, it is possible to obtainchemical-state information with an X-PEEM. The yieldof electrons over the entire illuminated area is imaged athigh magnification onto an image-intensified camera. Byrecording frames sequentially with an incrementally

increasing photon energy, spatially resolved electronyield, and hence absorption, can be measured in the near-edge region, thereby producing a NEXAFS spectrum foreach image point. In principle, the addition of an elec-tron-energy analyzer allows XPS spectromicroscopy, aswell.

At present the chromatic aberration resulting from thewide energy distribution of the imaged electrons limitsthe spatial resolution of the X-PEEM. For a microscopewith a 10-kV accelerating voltage, a resolution around 300nm has been measured in a field of view 50 µm indiameter. The field size itself is dependent on themagnification of the microscope, since the size of thedetector is fixed. Instruments now under design, in whichan electron-mirror removes the chromatic aberration,ultimately should achieve 5-nm resolution.

For full-field imaging microscopy, the goal is to get thehighest possible flux in the field of view, hence the beamneed only be demagnified enough to illuminate the areato be imaged. Since there is no focusing to a small spot,the illuminating radiation need not be diffraction limited,so both undulators and bend magnets at the ALS canserve as high-quality sources.

Moreover, the full-field photoelectron microscope isinherently fast, and the evolution of surface processes,such as the propagation of surface-reaction fronts at lowcoverage, can be measured at video rates. As compared toa scanning instrument, however, the microscope is lessversatile, owing to the exclusive use of electron detection,and suffers from the added complexity of the electronimaging system and the need for a high-vacuum environ-ment.

Schematic of a simple two-stage X-PEEM showing elec-trostatic objective andprojective lenses that createa magnified image of thephotoemitted electrons fromthe sample surface. A pinholein the back focal plane of theobjective lens limits the angu-lar acceptance to reducespherical aberration.

37

Page 36: Sof X-Ray Spectromicroscopy

38

New concepts in electron optics are making possible an advancedx-ray photoemission electron microscope (X-PEEM) for chemical-state imaging that has very high spatial resolution and greatly im-proved transmission, as compared to previous instruments. Onebreakthrough is the invention of electrostatic hyperbolic-field mirrorsfor the correction of electron-optical aberrations, and chromaticaberrations in particular.* The use of aberration-correcting mirrorshas the potential for pushing down the spatial-resolution limit of theX-PEEM to 5 nm and possibly even to 2 nm.

In the first of an evolving series of microscopes leading to incorpo-ration of this concept at the ALS, photoemitted electrons will beimaged by means of an optical system consisting of a three-elementhigh-voltage objective lens, a transfer lens, intermediate lens, anddeflector lens. With this system, electrons will be imaged at highmagnification onto an image intensifier and CCD camera. Next, cou-pling lenses, a chromatic-aberration correcting electron mirror, anda beam separator will be added. The expected improved spatialresolution would take synchrotron-based spectromicroscopy into theregime previously accessible only by using electron-beam probes intransmission and scanning electron microscopes. Eventually, lensdesigners plan to refine the electronic optics and incorporate energyfiltering (i.e., use an electron-energy analyzer) so that multiple elec-tron spectromicroscopies (i.e., NEXAFS and XPS) will be possiblewith the same instrument

*A number of groups are working on this problem; we are followingthe electron-mirror correction method suggested and demonstratedby Rempfer at Portland State University, Oregon.

Advanced Electron Lenses Reduce Chromatic Aberration

Page 37: Sof X-Ray Spectromicroscopy

CAD drawing of a new 30-kVx-ray photoemission electronmicroscope (X-PEEM) nowunder construction at theALS for full-field imagingshows the multiple lens sys-tem whose high voltage,small aperture, and beam-guiding deflectors should,according to extensive com-puter modeling, give a spatialresolution around 30 nm. Afuture version of the X-PEEMwill incorporate an electronmirror for correction of chro-matic aberrations and stillbetter resolution.

39

Page 38: Sof X-Ray Spectromicroscopy

Get

ting

Sta

rted

Page 39: Sof X-Ray Spectromicroscopy

A Growing Complement Of Tools &Techniques

ALSThe

New research tools oftenhave unforeseen benefits. Forexample, soft x-ray fluores-cence spectroscopy at undu-lator beamlines 7.0 and 8.0(shown here) has blossomedinto a powerful probe of elec-tronic structure at surfacesand in the bulk.

42

Page 40: Sof X-Ray Spectromicroscopy

The characteristics and capabilities of soft x-rayspectromicroscopy reviewed in the previous pagesshow that applications to the microcharacterization

of materials and surfaces are potentially widespread andlimited only by the imagination and creativity of thosewith problems to solve.

But the question may arise: why another form ofmicroscopy? There are already a plethora of microscopiesbased on electron and ion probes that have excellentspatial resolution and even elemental sensitivity. Inaddition, the new generation of scanning probe tech-niques (STM, AFM, etc.) offer a wealth of information.The answer is that no single technique can provide all theanswers that materials scientists seek, so spectromicros-copy should be viewed as bringing a major new capabilityto materials microcharacterization, rather than as replac-

ing established techniques. Moreover, the field ofspectromicroscopy has only recently been born, and weare just setting out to develop applications and thetechniques to carry them out.

Because the ALS is a relatively new facility, ourcomplement of experimental instrumentation continuesto grow as additional research opportunities becomedefined and funding becomes available. Accordingly, weare providing information about specific spectromicros-copy facilities and successful demonstrations of theirapplications in the form of loose-leaf inserts in the pocketon the inside back cover of this brochure. These insertsare modified or added as new information becomesavailable, so that they represent the most up-to-dateinformation about spectromicroscopy opportunities atthe ALS.

43

Page 41: Sof X-Ray Spectromicroscopy

ParticipationHow To Conduct Research At The ALS

44

Please see the ALS Web site for information about publications and proposal submission. A link is provided above.
Page 42: Sof X-Ray Spectromicroscopy

The ALS provides synchrotron radiation year around,except for scheduled shutdowns (usually one majorshutdown each year lasting several weeks) for

installation of new experimental facilities and for accel-erator maintenance. The current operating scheduleprovides 15 shifts per week (8 hours/shift) for users. As aU.S. Department of Energy national user facility, theALS is available without charge to personnel fromuniversity, industrial, and government laboratories fornonproprietary research intended for publication in theopen literature. Proprietary research is also welcome butis subject to a nominal cost-recovery charge for provisionof beam time. Proprietary users have the option to taketitle to any inventions made during the proprietaryresearch program and treat as confidential all technicaldata generated during the program.

Whether nonproprietary or proprietary, there are threemodes of conducting research at the ALS: as a member ofa participating research team (PRT), as an independentinvestigator, or as a collaborator with a PRT. Collaborativegroups comprising research personnel from one or moreinstitutions with common research interests, PRTscontribute to the construction, operation, and mainte-

nance of experimental facilities at the ALS. In return fortheir contributions, PRT members are granted priority fora percentage of the operating time on their facilities. Theremaining operating time on each beamline is available toindependent investigators. The proportion of timeallotted to independent investigators varies from beam-line to beamline. Independent investigators may bringtheir own experimental stations to ALS beamlines or withPRT concurrence use PRT stations.

Proposals for the establishment of new PRTs arereviewed by the Program Advisory Committee. Thoseinterested in collaboration with a PRT, which may be aproductive mode of entry to the ALS for new users,should contact the spokesperson for the appropriate PRT.Proposals from independent investigators are peer-reviewed by the Proposal Study Panel twice a year with1 June and 1 December deadlines for receipt of proposals.For details, consult the ALS Users’ Handbook, which isavailable from the User Administrator (see box). Addi-tional publications available include a safety handbook, abeamline-design guide, and an annual activity reportdescribing the previous year’s accomplishments.

45

Page 43: Sof X-Ray Spectromicroscopy

ReadingAdditional

Publications in the literature relevant to spectromicroscopyfor materials microcharacterization include:

H. Ade (editor), Special issue on spectromicroscopy with x-ray and VUV photons, Journal of Electron Spectroscopy andRelated Phenomena 258 (1) (to be published in spring 1997).

H. Ade et al., “Chemical Contrast in X-Ray Microscopyand Spatially Resolved XANES Spectroscopy of OrganicSpecimens,” Science 258, 972 (1992).

H. Ade and B. Hsiao, “X-Ray Linear Dichroism Micros-copy,” Science 262, 1427 (1993).

G.F. Rempfer, D.M. Desloge, W.P. Skoczylals, and O.H.Griffith, “Simultaneous Correction of Spherical and Chro-matic Aberrations with an Electron Mirror: An ElectronOptical Achromat,” Microscopy and Microanalysis 3, 14(1997).

G. R. Harp, Z. L. Han, and B. P. Tonner, “X-Ray Absorp-tion Near Edge Structures of Intermediate OxidationStates of Silicon in Silicon Oxides During Thermal De-sorption,” Journal of Vacuum Science and Technology A 8, 2561(1990).

J. Kirz, C. Jacobsen, and M. Howells, “Soft X-Ray Micro-scopes and Their Biological Applications,” Quarterly Re-view of Biophysics 28, 33 (1995). Despite the emphasis onbiology, a good summary of microscopy principles.

W. Ng. et al., “High Resolution Spectromicroscopy withMAXIMUM: Photoemission Microscopy Reaches the1000 Å Scale,” Nuclear Instruments and Methods in PhysicsResearch A 347, 422 (1994).

J. Stöhr et al., “Element-Specific Magnetic Microscopywith Circularly Polarized X-Rays,” Science 259, 658 (1993).

B. P. Tonner, G. R. Harp, S. F. Koranda, and J. Zhang, “AnElectrostatic Microscope for Synchrotron Radiation withX-Ray Absorption Microspectroscopy,” Review of ScientificInstruments 63 (1), 564 (1992).

B. P. Tonner, “The Role of High Spectral Resolution inSoft X-Ray Microscopy,” in X-Ray Microscopy and Spectro-microscopy, J. Thieme, G. Schmahl, E. Umbach, and D.Rudolph, eds., Springer-Verlag, Heidelberg, to be pub-lished in 1997. This is the proceedings of XRM ’96 Inter-national Conference on X-Ray Microscopy andSpectromicroscopy, Würzburg, Germany, 19–23 Au-gust 1996.

T. Warwick et al., “Soft X-Ray Spectromicroscopy Devel-opment for Materials Science at the Advanced LightSource,” in the special issue on spectromicroscopy with x-ray and VUV photons, Journal of Electron Spectroscopy andRelated Phenomena 258 (1) (to be published in spring 1997).Also published as Lawrence Berkeley Laboratory ReportLBNL-38906, August 1996.

Synchrotron Radiation News reports on developments insynchrotron radiation at facilities around the world. Articleson x-ray spectromicroscopy for materials analysis include:

H. Ade et al., “Industrial Applications of X-ray Micros-copy,” Synchrotron Radiation News 9, 31 (No. 5, September/October 1996).

H. Ade, “NEXAFS Microscopy of Polymeric Samples,”Synchrotron Radiation News 7, 11 (No. 2, March/April 1994).

W. Ng et al., Study of Surfaces and Interfaces by ScanningPhotoemission Microscopy,” Synchrotron Radiation News 7,25 (No. 2, March/April 1994).

B. P. Tonner, “Photoemission Spectromicroscopy of Sur-faces in Materials Science,” Synchrotron Radiation News 4,27 (No. 2, March/April 1991).

46

Page 44: Sof X-Ray Spectromicroscopy

Recent proceedings of major x-ray microscopy conferencesinclude:

X-Ray Microscopy and Spectromicroscopy, J. Thieme, G.Schmahl, E. Umbach, and D. Rudolph, eds., Springer-Verlag, Heidelberg, to be published in 1997. This is theproceedings of XRM ’96 International Conference on X-Ray Microscopy and Spectromicroscopy, Würzburg, Ger-many, 19–23 August 1996.

X-Ray Microscopy IV, V. V. Aristov and A. I. Erko, eds.,Begorodski Pechatnik Publishing Company, Moscow,1995.

X-Ray Microscopy III, D. A. Michette, G. Morrison, and C.Buckley, eds., Springer Series in Optical Sciences Vol. 67,Springer-Verlag, Berlin/New York, 1992.

National and international conferences on synchrotronradiation are held every two and three years, respectively.Recent proceedings include:

G. K. Shenoy and J. L. Dehmer, Eds., “Proceedings of theNinth National Conference on Synchrotron Radiation In-strumentation,” Review of Scientific Instruments 67 (9),(1996). Available only on CD.

J. B. Hastings, S. L. Hurlbert, and G. P. Williams, Eds.,“Proceedings of the Fifth International Conference onSynchrotron Radiation Instrumentation,” Review of Scien-tific Instruments 66 (2), (1995).

G. G. Long, D. R. Mueller, and S. H. Southworth, Eds.,“Proceedings of the Eighth National Conference on Syn-chrotron Radiation Instrumentation,” Nuclear Instrumentsand Methods in Physics Research A 347 (1994).

R. L. Stockbauer, E. D. Poliakoff, and V. Saile, Eds., “Pro-ceedings of the Seventh National Conference on Synchro-tron Radiation Instrumentation,” Nuclear Instruments andMethods in Physics Research A 319 (1992).

I. H. Munro and D. J. Thompson, Eds., “Proceedings ofthe Fourth International Conference on Synchrotron Ra-diation Instrumentation,” Review of Scientific Instruments 63(1), (1992).

Background material about x-ray spectroscopy and its use inmaterials analysis can be found in:

C. R. Brundle, C. A. Evans, Jr., and S. Wilson, eds., Encyclo-pedia of Materials Characterization, Butterworth-Heinemann, Stoneham, MA, 1992.

N. V. Smith and F. J. Himpsel, “Photoelectron Spectros-copy,” in E.-E. Koch, ed., Handbook of Synchrotron Radia-tion, Vol. 1B, North-Holland Publishing Company,Amsterdam, 1983.

Joachim Stöhr, NEXAFS Spectroscopy, Springer-Verlag, Ber-lin, 1992.

47

Page 45: Sof X-Ray Spectromicroscopy

48

Page 46: Sof X-Ray Spectromicroscopy

This document was prepared as an account of work sponsored by the United States Government. While this document is believed to contain correct information, neither the United States Government nor any agency thereof, nor The Regents of the Uni-versity of California, nor any of their employees, makes any warranty, express or im-plied, or assumes any legal responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commer-cial product, process, or service by its trade name, trademark, manufacturer, or other-wise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or The Regents of the University of California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof, or The Regents of the University of California.

DISCLAIMER

Available to DOE and DOE Contractors from the Office of Scientific and Technical Information, P. O. Box 62, Oak Ridge, TN 37831. Prices available from (615) 576-8401.

Available to the public from the National Technical Information Service U.S. Department of Commerce, 5285 Port Royal Road, Springfield, VA 22161.

Ernest Orlando Lawrence Berkeley National Labo-ratory is an equal opportunity employer.

This work was supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, of the U. S. Department of Energy under Contract No. DE-AC03-76SF00098.

P U B - 7 8 6 • M a y 1 9 9 7

Page 47: Sof X-Ray Spectromicroscopy

May 1997

Ernest Orlando Lawrence Berkeley National LaboratoryAdvanced Light SourceUniversity of CaliforniaBerkeley, California 94720