Monatomic Depth Profiling
Experimental arrangement for a depth-profile experiment.
Monatomic depth profiling uses an ion beam to etch layers of the surface or surface contamination, revealing subsurface information. Combining a sequence of ion gun etch cycles with XPS analyses provides quantified information as well as layer thicknesses. Before removing material from the sample, a spectrum, or set of spectra, is recorded from the surface of the sample. The surface is etched by rastering an ion beam over a square or rectangular area of the sample. After the etch cycle, the ion beam is blanked and another set of spectra is recorded. This sequence of etching and spectrum acquisition is repeated until profiling has proceeded to the required depth.
If the sample is an insulator, then an equilibration period should be allowed between the ion etch part of the cycle and the data acquisition. This allows the sample’s surface potential to return to its steady state before data collection.
Summary of the experimental procedure for producing XPS depth profiles.
During the profile acquisition, the acceptance area of the transfer lens or the source-defined monochromator beam should be directed at the center of the area rastered by the ion beam, which ensures that the analyzed areas are situated on the flat bottom of the crater.
The experimental data is presented in the following ways:
- Displaying each individual spectrum
- Displaying all spectra from a given spectral region on a montage plot - this reveals peak intensity and position data trends
- Graphing the time or depth on the horizontal axis, and signal or atomic concentration on the vertical axis - this enables comparison of the trends in different elements in the same sample
The sputter yield determines the rate at which material is removed from the sample during a depth profile.
Sputter yield = Number of atoms removed/Number of incident ions
The sputter yield depends on the following factors: the material, ion energy, incidence angle and the mass and nature of the primary ion.
Material: The sputter yield depends on the sample’s elements and their chemical state. Although it is difficult to predict the sputter yield for a material, there are a number of computer simulations available. In general, these simulations predict the sputter yields of elements reasonably well, but the values they provide for compounds are less reliable. When possible, it is better to measure the sputter yield under normal experimental conditions.
Sputter yields of silicon as a function of ion energy for noble gas ions at normal incidence.
lon Energy: When plotting the variation of the sputter yield for silicon as a function of ion energy for each of the noble gases, the sputter yield is shown to be a sensitive function of the beam energy.
The variation of the sputter yield with angle for the three metals. Below approximately 60 degrees, the sputter rate increases with angle before passing through a maximum.
Incidence Angle: The way in which a sputter yield changes with angle depends on the material being sputtered. When considering the change of etch rate with changing angle, it is important to account for any change in the rastered area, which may be caused by tilting the sample.
Nature of the Primary Ion: If the ion beam species is not a noble gas ion, then there could be a chemical interaction between the ion beam and the sample surface. Commonly in SIMS, oxygen or caesium is used as a primary ion source, which chemically changes the surface and the sputter rate. For example, if an oxygen beam strikes a silicon surface at normal incidence, then the surface becomes silicon dioxide and the sputter yield becomes that of silicon dioxide, instead of silicon.
This is the second important consideration when acquiring a depth profile of a sample. Following is a list of some of the important factors that can influence the depth resolution. The relative importance of each one in determining the achieved resolution will depend on the sample being analyzed and the experimental conditions being used. These factors may be classified as physical factors, which are a consequence of the sputtering process, instrumental factors, which are a consequence of the way the experiment has been designed and sample characteristics.
Ion Energy: Depth resolution deteriorates as the ion energy increases, which is caused by the ion beam mixing the atoms within the sample. At increasing energies, the depth range of the ions within the sample increases, allowing mixing at greater depths.
Incidence Angle: As the incidence angle of the ion beam increases from normal incidence, the depth resolution improves. This result is largely because as the angle increases, the depth range of the ions within the sample decreases, which causes mixing to occur at reducing depths.
Primary Ion Species: At a given energy, the larger the ion is striking a surface, the shorter its depth range within the sample, which will result in improved depth resolution.
Crater Quality: The crater bottom must be as flat as possible over the area being analyzed. Otherwise, information is being collected from a range of depths and resolution suffers. As a general rule, the crater dimension should be between 5 and 10 ion beam diameters in each direction to get acceptable flatness over a reasonable distance within the crater.
Beam Impurities: The chemical impurities in the beam can be minimized by choosing a high-purity gas feed. There are other impurities in an ion beam that also must be considered (e.g., neutral species and ions with multiple charges). Neutrals in the beam are caused by charge exchange between the high energy ions and neutral species with thermal energies. The problem with neutrals in the beam is that they are unaffected by the electrostatic lenses and scan plates. These neutrals can sputter the sample in an undefined manner and disrupt the quality of the crater. If an ion has a double charge associated with it, then it will strike the sample’s surface with twice the energy of an ion with a single charge and have a correspondingly longer range within the sample.
Information Depth: In the analysis process, the depth from which information is collected will affect the depth resolution. Generally for electron spectroscopies, the lower the kinetic energy of the electrons, the smaller is their average escape depth. Therefore, if there is a choice of peaks that can be monitored in a profile, then the peak with the lowest kinetic energy should be chosen.
Redeposition: The removal of material from the crater wall and its deposition within the crater in the area of analysis is referred to as redepositon. The smaller the crater, the more important this effect becomes.
Original Surface Roughness: A sample with a rough surface may affect the overall depth resolution, because the roughness will be preserved throughout the profile.
Induced Roughness: The sputtering process can cause topography or roughness to appear during a profile and degrade the depth resolution. This effect can be largely overcome by rotating the sample (azimuthal rotation) during the sputtering part of a profiling cycle. Preferential Sputtering: In a multi-component sample, the sputter yields from different elements can be different. Such conditions will result in roughening that might not be controllable by azimuthal rotation.
Charging: Ion or electron beams can be deflected by a charge building up on the surface of an insulating sample. This effect can either distort the crater or change the analysis position within the crater. Charging across a thin oxide layer can cause migration of species through the layer.
The researcher must decide on a set of conditions that favor either analysis speed or depth resolution. Factors to consider when setting up a depth profile include the following:
- Beam energy: high for speed low for resolution
- Incidence angle (with respect to sample normal): High for smooth samples or low for rough samples
- lon mass: large mass helps depth resolution - Xe is expensive
- Azimuthal rotation: improves depth resolution and reduces etch rate
- Sputtered area: small for speed, large for resolution
- A pure gas feed to the ion gun will improve results
- Analysis area must be small compared with crater size and well centered within the crater
- A good vacuum will minimize the adsorption of reactive species on the surface, which is important when profiling for oxygen, carbon etc.
- The low kinetic energy signal will often give the best depth resolution
- The ideal sample will have a smooth, well-polished surface