RBS Tutorial: Instrumentation


By 1909, Ernest Rutherford had established that alpha particles consisted of helium with +2 charge. The backscattering experiment that bears Rutherford’s name was suggested by Hans Geiger (of Geiger counter fame). However, it remained for twenty-year old undergraduate Ernest Marsden to actually do the first measurements. Marsden observed that the vast majority of alpha particles (He++) passed cleanly through a thin gold foil, but that some were scattered at all angles from the incoming He++ beam. Rutherford proceeded from this observation to propose the existence of the atomic nucleus. The key feature of Rutherford’s nucleus proposal is that a very small volume contains most of the atomic mass. The alpha particles scatter from nuclei as a horde of billiard balls would scatter if propelled at a bowling ball.

Rutherford backscattering spectrometry (RBS) is the measurement of energies of these backscattered particles. These energies depend on the identity of the atom from which the alpha particle scatters, the angle of scatter, and the depth into the sample to which the particle travels before scattering. Thus, RBS can be used for elemental analysis, especially of surfaces.

One early use of RBS (called the alpha-scattering experiment at the time) was elemental analysis of lunar soils as part of the Surveyor V scientific payload in 1967. Most early RBS experiments used radioactive sources of alpha particles. Today, the intense pencil like beam of alpha particles required to produce a modest backscattered signal is most commonly provided by a charged particle accelerator.

RBS Theory - Instrumentation history

The three main components of an RBS instrument are a source of helium ions, an accelerator to convert them to high energy alpha particles, and a detector to measure the energies of the backscattered ions. The type of accelerator determines the configuration of the other components. Single-ended accelerators have the ion source floating at high voltage. Electrical isolation of the megavolt potentials is achieved by housing the terminal in a tank filled with an insulating gas, usually SF6. One disadvantage of locating the ion source within the tank is that it is difficult to change or replenish the source.

RBS Theory - Instruments

The tandem accelerator is a clever innovation. The tandem accelerator uses a positive terminal located in the center of the device. Negatively charged particles are injected into the accelerator and attracted to the terminal where a stripper element removes two or more electrons from each particle. The positive terminal repels the resulting positive ion back toward ground. Thus the particle acquires energy both before and after the terminal.

RBS Theory - Instruments

The tandem configuration has two important advantages over a single-stage setup. First, lower terminal voltages are required, and second, both the source and the ion exit operate near ground potential. The main disadvantage is that inefficiencies of He- production and charge stripping lower He++ beam current to about 100 nA for a tandem versus 1 mA for a single-ended accelerator. Fortunately, most RBS experiments can only use about 100 nA because of detector limitations. A typical RBS installation uses a tandem accelerator, producing a 2.25 MeV He++ beam by removing three electrons from He- at the + 750 KV terminal.

Tandem Accelerators

There are four main components of a tandem accelerator:

  1. A charged particle beam line in a tank containing high voltage components and an insulating gas.
  2. An electrode called the terminal supplied by a high voltage source.
  3. An electron charge stripper located at the terminal.
  4. A vacuum system for the charge stripper and beam line.


Although large accelerators can often fill entire halls, smaller accelerators are suitable for RBS. Such accelerators require investments in the $300,000 range, and suitable end stations for analysis and control cost an additional $50,000 to $200,000. While these are certainly expensive devices, they are not out of line with other analytical instruments.

Voltage Sources

The oldest form of accelerator is the Van de Graaff, named for Robert Van de Graaff, who put a suggestion by Lord Kelvin into practice. Its distinguishing feature is charge transfer on a moving belt with one pulley at ground and the other at the terminal. Charge is placed on the belt by a comb of corona points. A second set of points removes the charge at the terminal. Although the traditional Van de Graaff accelerator is single-ended, the voltage source can be used in a tandem machine.

A well-known variation is the Pelletron which is identical to a Van de Graaff, except that a chain with alternate links (pellets) of metal and insulator replaces the belt. This chain provides more uniform charge transport than the belt, resulting in more stable voltage.

A Tandetron accelerator uses a voltage doubler power supply fed by a radio frequency signal. There are no moving parts, reducing the need for expensive maintenance. The voltage from the Tandetron device is also very stable. Terminal voltage stability influences spectroscopic resolution.

RBS Theory - Instrumentation Voltage Sources
Beam Line and Tank

Electrical isolation of the accelerator terminal obviously requires good insulators. Because air ionizes and becomes conducting in high electric fields, the accelerator components must be arranged to minimize electric fields. Thus, one guiding principle of accelerator design is elimination of any sharp points or edges that produce high electric fields. The high voltage components are housed in a pressure vessel filled with sulfur hexafluoride gas (SF6) because this gas resists electrical breakdown better than air.

The terminal and the beam line require mechanical support, a more demanding task because of the forces exerted by the pulley used as one end of a Van de Graaff or Pelletron voltage source. The beam line consists of alternating insulator and conductor sections connected to the terminal voltage along a resistor chain to linearize the electrical potentials and reduce electric fields between components. The beam line and the terminal are surrounded by a series of smooth conducting rings also connected along a resistor chain to the terminal voltage. These precautions serve to reduce voltage gradients between accelerator components and thus minimize electric fields. The precautions are necessary because the 750 kV terminal has the potential to produce very large sparks.

RBS Theory - Instrumentation Beam Line Tank
Stripper Elements

The He- ions have substantial kinetic energy (typically 750 keV) when they arrive at the positive terminal. High energy ions lose electrons in grazing collisions as they travel through any material. The ions also give up a small amount of kinetic energy for each lost electron. However, the lost energy is small compared to the kinetic energy of the negative ions as they pass the terminal.

Convenient electron stripper materials (located at the terminal) include thin foils and gases. A foil is easier to use since it simply hangs in the beam path. However, the thickness of the foil can increase by carbon deposition (from residual carbon containing gases in the vacuum system) or decrease by sputtering. Foil thickness affects stripper efficiency and beam energy. The foils must occasionally be replaced, which involves opening the tank and the beam line.

Because gas cells operate more predictably and avoid maintenance problems, they have become more common. Gas cells consist of concentric tubes, perhaps 0.3 m long. The ends are closed except for small holes through which the ion beam passes. The outer tube is connected to an extra vacuum pump and gas is leaked from an external source into the inner tube to produce a pressure of 1 to 10 millitorr. This differential pumping in the outer tube reduces the amount of gas leaking into the beam line.

Because of the pressures required to maintain a gas cell, pressures in the flight tube (measured at the end of the accelerator) rise into the 1 to 10 microtorr range. Lower pressures are possible with a stripper foil. However, the modest collisional cross section of high energy ions with gas molecules diminishes the requirement for high vacuum in RBS experiments.

RBS Instrumentation Stripper Elements
Source of Negative Helium Ions

The He+ or He++ ions required for a single ended accelerator come from plasma ion sources. The duoplasmatron starts with a low voltage arc burning between cathode and anode. The helium plasma is geometrically and magnetically confined, and ions are extracted by a strong electric field. The radio frequency plasma source also generates He+ ions. Radio frequency sources are more common in single-ended accelerators, because they produce He++ more efficiently than duoplasmatrons. A typical source produces 1 mA of He+.

RBS Instrumentation Negative helium ion sources

Tandem accelerators require negative helium ions. Helium is the most inert of the inert gases; it tends not to gain or lose electrons. The ion source must overcome this natural tendency of helium not to form negative ions. The source operates in two stages. First, positive ions are produced as described above. An alkali metal channel converts He+ into He- in a second stage. This charge exchange happens as He+ passes through hot alkali metal vapor. Rubidium is used in the illustration, but all of the alkali metals have sufficient reducing power to form He-. The charge exchange process is inefficient; 1 milliamp of He+ leads to 1 microamp of He-. The negative ions are extracted from the alkali metal channel and injected into the tandem accelerator at 20 to 30 keV.

RBS Instrumentation Negative helium ion sources
Focusing Elements

Two ion optical components are usually placed between the accelerator and the sample chamber. A magnetic field separates any He-, He, or He+ from the He++ beam. A quadrupole lens shapes the beam and focuses it into the sample chamber. Relatively strong ion bending and focusing components are required for these high energy (rigid) beams. The high energy beam provides an important analytical advantage.

RBS Instrumentation Focusing elements

Samples in RBS can be insulating, in contrast to most charged particle analytical methods. Even if such samples charge to a few kilovolts, the rigid beam is barely deflected. Sample charging causes only slight perturbations in RBS spectra because the kilovolt charging effect is a small fraction of the megavolt particle energies.

Sample Chamber

The main sample chamber components are a stage, one or more detectors,a beam entrance, and the vacuum system. The chamber can be as simple asa flange with a sample and a single energy dispersive spectrometer attached.

RBS Instrumentation Sample chamber

More typically, the samples mount on a five-axis goniometer, which isconvenient for loading many samples into the vacuum system and analyzingthem sequentially. The goniometer can also tilt and rotate the samples.Comparing spectra obtained at different incident and exit beam angles providesbetter characterization of sample composition as a function of depth. Randomlytilting and rotating a sample relative to the incident beam avoids variationsin the spectrum that arise because the incident beam travels down a channelin the sample’s crystal structure. Two surface barrier detectors (one normal angle, one grazing exit), are included in a typical RBS sample chamber.


Alpha particles come from the accelerator and strike the sample surface. Energies must be measured for those few particles that recoil back into the detector. Surface barrier silicon detectors are used in RBS. Since these devices are essentially diodes, they are often called semiconductor diode detectors. The high energy charged particles produce electron-hole pairs in the semiconducting material. The detector is operated with an electrical potential (typically 4 kV) between the front and back surfaces. In the resulting electric field, the electron-hole pairs produce a current proportional to the energy of the charged particle.

The average energy expended by He++ to produce one electron-hole pair is approximately 3.7 eV. This is sometimes called the ionization energy of the detector. Each 1 MeV particle produces about 2700 electron-hole pairs. The fluctuation or variance in the number of charge carriers affects the spectroscopic resolution. The theoretical minimum variance (which follows Poisson statistics) is equal to the number of charge carriers. The standard deviation equals the square root of the variance. The Fano factor is the ratio of the observed to this theoretical minimum variance. The Fano factor implicates other sources of peak broadening, typically incomplete charge collection and variations in dead layer loss.

Incomplete charge collection is minimized by high purity semiconductors which provide relatively few sites for electron-hole pair recombination. The energy lost before the charged particle reaches the active volume of the detector (dead layer loss) is minimal because this layer is thin (about 100 nanometer) in surface barrier detectors. Since this thickness corresponds to only about 0.4% energy loss for 1 MeV He++, small variations in the energy loss are insignificant for typical RBS experiments. High quality silicon surface barrier detectors are thus nearly ideal for alpha particle spectroscopy.

Particle arrival times at the detector are randomly spaced in time, leading to the possibility of interference between measurements when particles arrive at nearly the same time. This phenomenon, called pulse pile-up, becomes a serious problem at high particle arrival rates. There are two types of pile-up. Tail pile-up involves the superposition of pulses on the long duration tail or undershoot from a preceding pulse, leading to reduced spectral resolution. High quality electronic circuits minimize tail pile-up. Two pulses sufficiently close together to be treated as a single pulse undergo peak pile-up, the second type. Detector dead time is the minimum time between successive ion arrivals if they are to be measured separately. Peak pile-up ultimately limits the rate at which RBS data collection can occur.

RBS Instrumentation - spectroscopy
Related Techniques

By adding accessories to the sample chamber, or by changing the operating procedures, several other experiments can piggy back onto the RBS analysis. For example, measurement of the X-rays produced by ion bombardment is called particle induced X-ray emission (PIXE). Common RBS accessories include detectors for these X-rays, which are always produced, but not always measured. Hydrogen forward scattering (HFS) only requires the addition of a thin foil to separate forward scattered hydrogen from forward scattered primary He++ ions. Heavy ion backscattering spectrometry (HIBS) is the same as RBS, except that heavy ions are used instead of He++. Primary He++ ions with higher energies than the usual 2.2 MeV scatter, but the nuclear interactions become more complicated. In spite of these complications, called resonance effects, higher beam energies often prove useful. In some cases, the incident ions are captured by a target isotope and a different particle is promptly emitted. Measurement of the energies of these reemitted particles is called nuclear reaction analysis (NRA). Finally, charged particle activation analysis (CPAA) uses the accelerator to produce new radioactive nuclides which are measured with the same instruments used in neutron activation analysis.

Particle Induced X-Ray Emission

PIXE stands for Particle Induced X-Ray Emission. Several kinds of excitation beams produce X-rays with energies characteristic of the target elements. Thus, photon excitation (by X-rays) gives rise to X-ray fluorescence spectroscopy. Electron excitation in a scanning electron microscope or an electron microprobe provides energy dispersive or wavelength dispersive X-ray spectroscopy (depending on the X-ray detector). Charged particle beams of He++, or more commonly H+, afford PIXE spectroscopy. In all three cases, the excitation beam removes a core electron and X-rays are emitted with specific energies when outer shell electrons change state to inner shell. The X-ray energies are independent of the excitation process. The PIXE accessory is useful for heavy element identification on RBS instruments. These heavy elements have small differences in RBS energies, but distinct differences in PIXE spectra. Dedicated PIXE instruments usually use H+ rather than He++ because H+ provides higher sensitivity. If helium and hydrogen are mixed in the source, He++-based RBS and H+-based PIXE can be performed with the same accelerator.

PIXE has several advantages as an analytic technique. It offers signal levels similar to its electron beam counterparts, but better signal-to-background ratios. The background in electron spectroscopy arises from bremsstrahlung which is largely absent in PIXE because He++ or H+ ions, even at PIXE energies, have much lower velocities than electrons. Another advantage of PIXE over electron induced spectroscopy is that, like RBS, PIXE works with insulating samples. Finally, a proton beam can pass through a thin window and penetrate several centimeters through air. Because of this, unusual samples, such as valuable art work, need not be subjected to the rigors of a vacuum. PIXE finds applications in geology, archaeology, and art conservation.

Hydrogen Forward Scattering

HFS stands for Hydrogen Forward Scattering. An HFS experiment uses essentially the same apparatus as standard RBS. The analytical information obtained by HFS consists of hydrogen concentration versus depth. The sample is tilted so that the He++ beam strikes at a grazing angle. Helium is heavier than hydrogen, so there is no backscattering of He++ from hydrogen. Instead, hydrogen (H+ and Ho) is knocked forward with significant energy after being struck by He++.

The He++ also scatters toward the detector because of heavy elements in the sample. The number of He++ ions scattering at this low angle is large relative to forward scattered hydrogen. The He++ signal would swamp the H+ signal except that a thin foil (about 8 microns) positioned in front of the detector separates the interfering He++ from the H+. Carbon, mylar, and aluminum foils are commonly used. The foils cause significant energy loss and straggling in the forward scattered hydrogen. Although these effects limit the HFS depth profile resolution to about 50 nanometer, hydrogen (or deuterium) quantitation at surfaces is possible with 5 % accuracy down to 0.01 % detection limits.

Heavy Ion Backscattering Spectrometry

HIBS stands for Heavy Ion Backscattering Spectrometry. This technique uses ions heavier than He++. Collision cross sections are higher for heavy primary ions, and there are no resonance effects at available energies. These heavy ion beams provide advantages in trace heavy element determinations of light element samples. The matrix elements are all scattered forward and cannot contribute interference signals. The forward scattered analyte elements can also be useful for analysis. Measurement of analyte ions scattered forward by heavier primary ions is a general phenomenon called Elastic Recoil Detection (ERD). (HFS) is a special case of ERD.

The accelerators and detectors are the same for HIBS and RBS, but different ion sources are required. A typical heavy ion source uses a sputtering Cs+ beam impinging on a heavy element such as silicon from which negative ions (Si- in this example) are extracted. The charge stripper removes a variable number of electrons from Si-, leaving ions with mixed charge states and energies. The primary ion beam must be separated into components, usually in a magnetic field, before the beam strikes the sample.

Nuclear Reaction Analysis

NRA stands for Nuclear Reaction Analysis. Helium ions with less than 2.2 MeV energy undergo elastic recoil with all elements. Elastic recoil, like colliding billiard balls, can be described by classical mechanics. For ease of interpretation, most RBS analyses use less than 2.2 MeV He++ beams. Higher energy He++ ions collide non-elastically with analyte nuclei. This means that the collision cross sections can be much higher for certain (resonance) energies. At resonance energies, the analyte nucleus seems to absorb and reemit the He++ ion rather than simply scatter it. Quantum mechanics are required to explain these interactions.

An example is the well known resonance effect at 3.045 MeV for alpha particles scattered from 16O. The process is indistinguishable from elastic recoil except that the cross section is many times greater. Although variable cross sections resulting from higher beam energies complicate the analysis, the higher energies are useful in specific cases such as oxygen present as a single impurity in a thin layer.

At some resonance energies, the primary ion is absorbed by the analyte nucleus and a different particle (usually a proton, neutron, alpha particle, or gamma ray) is promptly emitted. Among the light elements, there are several potentially useful reactions such as 19F(p,alpha)16O. In this case a 19F nucleus absorbs a 1.25 MeV proton and promptly emits an 8.114 MeV alpha particle. Measurement of the alpha energy indicates the depth from which it arose. This is the nuclear reaction analysis technique.

To enable certain features and improve your experience with us, this site stores cookies on your computer. Please click Continue to provide your authorization and permanently remove this message.

To find out more, please see our privacy policy.