Angle-Dependent XPS Increasingly Popular with Researchers.
The surface sensitivity and quantitative chemical analysis capabilities of x-ray photoelectron spectroscopy (XPS) have made it one of today's most broadly used surface-analysis techniques. With an ability to nondestructively detect all elements except hydrogen and helium at levels down to about 0.05 atom percent, it's being used extensively to examine the composition of the surface layers of solid materials down to a depth of about 5 nm.
Angle-dependent XPS is a refinement of the technique that lets users zero in on the depth of their choice and even develop detailed compositional depth profiles. Although the amount of angle-dependent XPS experimentation being done remains relatively small, there are several emerging applications for it that promise to increase its usefulness.
"One growing application for angle-dependent XPS involves studying gate insulating films for advanced semiconductor devices," says John Moulder, a staff scientist at Physical Electronics Inc. (PHI), Eden Prairie, Minn. "These films are becoming too thin to be measured by traditional methods."
Polymers are another important application area. If you want to print on a polymer material or make it impermeable as a container liner or prevent static buildup on it as a surgical garment, you need to treat the polymer surface with film layers that are often extremely thin. Angle-dependent XPS is being used to look at these layers.
Other application areas concern films used as adhesion promoters, corrosion inhibitors, heterogeneous catalysts, and passivating layers on metals and semiconductors, says John Watts, a researcher at the Univ. of Surrey, Guildford, U.K. Yet another important use involves mapping the thickness variations of lubricants on hard disks.
XPS is accomplished under ultrahigh vacuum conditions by irradiating a sample with monochromatic x-rays. Absorption of these x-rays by the sample results in emission of photoelectrons. The kinetic energies of the emitted electrons are measured with an electron spectrometer, yielding a spectrum that consists of a series of photoelectron peaks.
The kinetic energy of each emitted electron is related to the binding energy of the electron, by means of Einstein's photoelectric equation. The binding energy, in turn, is approximately equal to the electron's former orbital energy level inside the atom it vacated. Because each atom has a unique set of orbital energy levels, the measured energy spectrum is able to tell researchers about the elemental composition of surface layers.
The electron energy levels of an atom can be thought of as comprising two types: core energy levels belonging to electrons that are tightly bound to the nucleus of an atom and valence levels belonging to those electrons that are weakly bound. Although the valence electrons participate most in forming chemical bonds in molecules and compounds, the energy levels in core electrons are shifted slightly by chemical bonding. XPS, which targets the core energy levels, is able to provide chemical state information by providing a measure of these shifts. Because of this capability, XPS is often referred to as ESCA, electron spectroscopy for chemical analysis.
In practice, peak areas are used to determine the composition of a material's surface. The shapes of each peak along with the binding energies are used to provide chemical bonding information.
Incoming x-rays penetrate the surface of a sample to micrometer depths. However, the emitted photoelectrons, because of their low kinetic energies ([is less than] 1500 eV), can only travel a short distance before being scattered inelastically. This short distance is referred to as the electron's escape depth. Photoelectrons that are close enough to the surface to escape without losing energy participate in the XPS peaks. Those photoelectrons that lose energy before leaving the sample surface contribute to the background of the measured spectra. The escape depth limitation is what restricts XPS to analyzing only topmost layers.
For flat samples, the analysis depth is proportional to the sine of the angle between the axis of the electron analyzer lens and the sample's surface. By varying this angle--usually by tilting the sample--it's possible to get a picture of the compositions of varying surface thicknesses. So at a grazing angle, measurements become more surface-sensitive. As the angle becomes more perpendicular to the analyzer, XPS looks deeper into the sample.
It's important to keep in mind that you aren't viewing different "slices" as you change angle, because you continue to view through the topmost layers. So an angle-dependent XPS experiment conducted at various angles is only suggestive of a compositional depth profile; the technique does not directly provide it. However, it's often very useful to know qualitatively where a species is enhanced or depleted at the surface.
Most XPS vendors provide curve-fitting software that will provide elemental and chemical-state information about the surface volume measured. Software that will tease apart contributions from various levels to provide compositional depth profiles is not, however, commercially available at this time. Actually there is no shortage of this type of software; it's just that it takes the form of algorithms that have been developed on an application-specific basis. The algorithms were developed by research groups, primarily at universities, to address their own research needs.
"At the present time, there is no universally accepted way of interrogating the angle-resolved XPS data to produce a compositional depth profile," says Watts. "We have a very sophisticated algorithm that we set up very carefully for use on polymers. Although we occasionally use it with metal oxides, we are conscious that we need to do some additional work on it to get it to work just right for that application."
The lack of a single piece of software that can address several important application areas has inhibited the use of angle-dependent XPS for depth profiling, but that situation is about to change, thanks to work at the National Physical Laboratory (NPL), Teddington, Middlesex, U.K.
"The NPL work will hopefully end up with a package of different algorithms that can be used for different situations," says Watts. NPL is now validating the software and should release its first version very soon." The ESCA Users Group (www.york.ac.uk/org/esca/home.html) is also actively involved.
In industry, most applications of angle-dependent XPS are performed at one angle, says PHI's Moulder. "They may do some profiling initially to understand what's going on, but then they take data at just one angle, especially if they are performing a process-control function. In that context, angle-dependent XPS doesn't differ from straight XPS, except that a researcher is consciously putting the sample at a specific angle so that a particular analysis depth will be chosen."
With some newer instruments, ordinary XPS and angle-dependent XPS are identical in every respect. That's because the manufacturers designed the transfer lens so that users can make the angle-dependent measurement they wish without tilting the sample. The PHI Quantum 2000 Scanning ESCA Microprobe, from Physical Electronics Inc. (612-828-6100), for example, puts the lens at a 45 [degrees] angle, which is perfect for many applications.
The new Sigma Probe XPS instrument from VG Scientific, East Grinstead, West Sussex, U.K., (+44-1342-327-211) features the Radian Lens, which is not only tilted at 37 [degrees] with respect to the surface of a flat sample, but collects emitted electrons within a cone that subtends an angle of 60 [degrees]. This means that users may collect data at the convenient 37 [degrees] angle or select "slices" from the cone, collecting only those electrons emitted at a near-normal angle of 67 [degrees] or a grazing angle of 7 [degrees]. This allows the collection of a great deal of angular information without tilting the sample at all, which is important with large samples, such as semiconductor wafers, that cannot be tilted.
Actually, tilting a sample is not the problem it used to be with new-generation XPS instruments. The Quantum 2000, for example, has a computer-controlled stage whose software keeps selected points on a sample at the focal point of the analyzer as the sample is tilted.
"The Quantum's software models what's going on when you tilt the stage. So if you want to be at 10 [degrees], it makes some calculations and puts the sample in the right place after the tilt operation," says Moulder.
The AXIS Ultra from Kratos Analytical Inc., Chestnut Ridge, N.Y., (914-426-6700), also offers an automated tilt capability, as does VG's Sigma Probe.
Another reason researchers formerly shunned taking measurements at several takeoff angles has to do with the fact that insulating samples acquire a net charge as photoelectrons leave them. To maximize sensitivity and resolution, this charge must be reduced to a minimum uniformly, since a variable surface charge will result in differences in peak energy, which will degrade the energy resolution, making it more difficult to resolve chemical states.
The conventional method for compensating for the charge buildup involves the use of a focused electron flood gun to supply electrons to the surface. This arrangement can lead to problems when data are being collected at more than one takeoff angle.
"Tilting the specimen means the angle of incidence of the electrons varies from one analysis to another. That can compromise the charge compensation, because it can mean that you're moving from a situation where the charge compensation is good to a situation--often at grazing emission--where the charge compensation is not so good. That may lead to changes in the apparent resolution and sensitivity," says Watts.
This is less of a concern with new-generation instruments. The Kratos AXIS Ultra, for example, uses its magnetic immersion lens as part of the charge neutralization system. Low-energy electrons are injected into the magnetic field from a filament located at the base of the electron input lens. These electrons follow the field lines to the surface of the sample, where they neutralize any surface charge buildup. As these are the same magnetic field lines used to extract and focus the photoelectrons, the neutralizing electrons arrive at the same point from which the photoelectrons are emitted.
The design of the neutralizer enables excess electrons to become trapped in the magnetic field until required for the neutralization process, creating a sufficiently high flux of electrons to ensure that the whole analyzed area has a uniform surface charge.
One disadvantage of taking measurements at several angles that the Sigma Probe is particularly good at overcoming involves x-ray flux density changes as the sample is tilted from one angle to another, causing the count rate to change. This change in count rate exacts a time penalty because you need to gather data longer to acquire the same signal-to-noise statistics.
Did you find this article useful? If so ... Circle 202
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||x-ray photoelectron spectroscopy|
|Publication:||R & D|
|Date:||Jan 1, 1999|
|Previous Article:||Strong U.S. Economy Drives Continued R&D Growth.|
|Next Article:||Analysis Software Shapes Data Into Useable Results.|