ion beams and their applications

Ion Beams and their Applications

Ion beams come in many shapes and sizes; different source technologies produce a wide range of different beam types, some have flexible sources that can accommodate a range of elements, while some are broad beams, and some analytical. It can be a mine field if you don’t know what you’re looking for. In this application note, we shed light on Ionoptika’s range of ion beams.

1. Sputter vs. Analytical Ion Beams

Ionoptika’s range of ion beams can be split into two groups based on their applications or purpose; sputter beams, and analytical beams.

Sputter Beams

While all ion beams will sputter a surface, we make this distinction based on the area and speed with which this occurs. Sputter beams have three characteristic features: high current, large spot size, and wide field of view. They are designed to deliver a large dose of ions over a wide area as quickly as possible, in order to optimize etch rates.

Sputter beams are often used to remove material prior to analysis using a separate analytical technique, e.g. SIMS, XPS, SEM, TEM, Auger etc. This can be for cleaning purposes, or used as a means of depth profiling through the sample.

Analytical Beams

Rather than being used to facilitate analysis using a separate technique, analytical beams are designed to perform the analysis themselves. They also have three characteristic features; wide energy range, small spot size, and variable current control. This gives the user incredibly fine control over the beam characteristics, which enables them to optimize their experiment.

Analytical beams are primarily used for mass spectrometry, for example SIMS (secondary ion mass spectrometry), and have also been installed on MALDI (matrix assisted laser desorption ionization) instruments.

2. C60 Beams

Carbon-60, or just C60, is a type of fullerene molecule, consisting of sixty carbon atoms formed into a hollow sphere, not unlike the shape of a soccer ball. The first C60 ion beam was produced by Ionoptika, in collaboration with the University of Manchester in 2002, and since then more than 140 units have been sold worldwide.

Compared to monatomic ion beams, C60 beams result in a much “gentler” sputtering action, greatly reducing the damage caused to sub-surface layers. Preferential sputtering – normally an issue for monatomic beams – is largely non-existent for C60, while the etch rate of is also relatively consistent across different material types. This makes C60 a very powerful, very consistent sputter source.

As an analytical beam, the gentle sputtering action of C60 also reduces the fragmentation of larger molecules, resulting in an enhanced molecular signal intensity. For techniques such as SIMS, this can be incredibly important, as it significantly increases the usable mass range of the technique.

3. Gas Cluster Ion Beams

Gas cluster ion beams (GCIB) are high-energy beams of cluster ions, ideal for the sputtering and analysis of organic matter. They are an incredibly versatile ion source, as both the beam-type and the properties of the beam can be varied as needed. This allows the user to tune the beam to the needs of their experiment.

The source operates through the adiabatic expansion of gas in a vacuum, causing rapid cooling and resulting in cluster formation. The clusters are then ionized through electron bombardment and accelerated through the column. The size of cluster is a vital parameter, and may be tuned over a wide range by adjusting the source conditions.

GCIB for Organic Analysis

GCIBs are the ideal choice for sputtering organic matter. Etch rates of organic materials are several orders of magnitude higher than metallic or semiconductor materials. This makes cluster beams such as the GCIB 10S an excellent tool for cleaning surfaces prior to analysis. The large cluster species also produce very little fragmentation or sub-surface damage – performing better than C60 on both fronts.

In order to maximize the benefits of GCIBs for SIMS, the beam must be operated as high energy. This is because the secondary ion yield shows a non-linear increase as a function of beam energy. We therefore currently offer two variants of analytical beam: the 40kV GCIB 40, and the 70kV GCIB SM.

The J105 SIMS utilizes the benefits of gas cluster beams for organic analysis. The combination of the gentle sputter action of large cluster ions with increased secondary ion yield has extended the usable mass-range to > m/z 2500.

Choice of gas

The versatility of GCIBs comes from having a choice of input gas. Argon has historically been a favorite, as it is an inert gas that forms clusters easily. Ar/CO2 mixtures are also common – the CO2 acting as a cluster formation agent. However pure CO2 gas is now the favored option for many of our customers using GCIBs for SIMS.

The stronger van der Waals forces between CO2 molecules results in much larger clusters than would be available for Ar – up to 20,000 in some cases. This gives users much greater control over the all important E/n value (energy per nucleon). Researchers have shown that optimizing E/n results in an enhancement of the secondary ion signal. It is also thought that the presence of O ions at the surface improves the probability of ionization – further enhancing ion yield.

We have recently developed a water vapor (H2O) source for GCIBs. This is currently available as an optional add-on for our GCIB-40 and GCIB-SM products.

4. Liquid Metal Ion Beams

Liquid metal ion beams, also known as LMIS, or LMIG, are a well established source technology. The source operates by a liquid metal reservoir feeding a blunt tungsten tip, from which ions are extracted by a strong electric field. The source design is elegant and reliable, and hence has been used in focused ion beam (FIB) systems for decades. Ionoptika offer a 25 kV LMIG system in two variants; the IOG 25AU gold-cluster system, and IOG 25GA gallium system.

Liquid metal beams produce monatomic or small-cluster ion beams, such as Au+, Ga+, and Au3+. As such, they are characterized by very small spot sizes (< 100 nm). This makes them ideal for high-resolution analysis applications.

The high-energy smaller ions can penetrate, or channel, far beneath the surface before dissipating their energy. This causes significant sub-surface damage, and as such depth profiling can be unreliable. It also results in significant fragmentation, making LMIG sources much more suited to the analysis of hard materials.

5. Duoplasmatron Ion Beams

Duoplasmatrons are an incredibly high-brightness ion source. The ions are generated as gas is slowly fed into an arc chamber, where it ignites as plasma. The plasma is then extracted and focused through the column. The nature of the source allows for many different source gases to be used. This means that duoplasmatron sources are a highly flexible source of ions.

Duoplasmatron ion beams are monatomic, and do not form clusters. This however results in much lower energy distributions, enabling smaller spot sizes. Combined with a high-brightness source, this leads to a very high current density beam, ideal for hard materials or where fragmentation is not a concern.

Ionoptika produce two variants of the duoplasmatron ion beam: the FLIG® 5 – a low energy sputter source, and the IOG 30D – a 30 kV high-current analytical beam.

FLIG – Floating Low Energy Ion Beam

The FLIG® 5 is a unique ion beam system as it is based on floating column design. The design enables very low energy operation (down to 200 eV), while still delivering high current. Operating at such low impact energies greatly reduces the penetration depth of the beam, which in turn improves the depth resolution. Both O2 and Cs variants of the FLIG are available.

6. Conclusion

The table below compares Ionoptika’s various ion beam products under a number of categories discussed in this article.

Analytical Ion Beams:

Ionoptika ion beam product comparison table

Sputter Ion Beams:

Ionoptika Product Comparison Table Sputter beams

SIMS showing Cr signal of M1.6 screw thread

ToF SIMS of Rough or Uneven Samples

A common problem faced by TOF SIMS analysis is loss of peak resolution and mass accuracy on samples with rough or uneven topography. The J105 SIMS does not suffer from these issues, experiencing no loss in mass accuracy across even the roughest of samples

The ability to obtain consistent, accurate results is key to all analytical techniques, and none more so than ToF SIMS. Providing detailed chemical information about a surface, ToF SIMS is an extremely versatile surface analysis technique. The results, however, can be very sensitive to sample environment; insulating samples, or changes in sample height can affect the measurement and result in inconsistent, or inaccurate results.

In this application note we explain the reasons why sample roughness can be a problem, and show how the unique design of the J105 SIMS has solved this issue.

Why Roughness is a Problem for ToF SIMS

Time of flight mass spectrometers determine the mass-to-charge ratio of ions by measuring the time taken to travel a known distance under an electric field. The accuracy of this measurement – known as the mass-resolution – is determined by how precisely this flight time can be measured.

In conventional ToF SIMS, good measurement accuracy is achieved by generating very short pulses (in nanoseconds) of the primary ion beam. If the sample is completely uniform, the variation in flight time of ions with the same mass will be minimal. Thus the experiment will be quite accurate.

Say that our sample is not uniform however, but has a height variation caused by a slight tilt when mounting. In this case, ions of equal mass coming from different parts of the sample will have to travel slightly different paths, and thus will have slightly different flight times. Our experiment will not be as accurate this time, and will show broader peaks as ions of the same mass arrive at slightly different times.

What about a sample with a lot of surface roughness or height variation, such as the surface of a meteorite, or a coronary stent? In these cases it should be clear that accurate results will be difficult to obtain.

There is also a secondary issue caused by samples with rich topography, particularly those that are naturally insulating. Topographic features may result in the build up of localized surface charge. This localized charge distorts the local electric field, and can thus affect the flight path of ions at these locations.

Solving the Problem: Decoupling The Mass Spectrometer

Animated GIF of J105 SIMS Operation

Schematic of the J105 SIMS operation, illustrating the path of primary (green) and secondary (red and blue) ions through the instrument.

In order to solve this problem, the sample and its environment must be decoupled from the mass spectrometer. However, in order to do this, a new method of producing the first time-focus is required.

The J105 SIMS has been designed such that the primary beam is decoupled from the time-of-flight measurement. Instead, a shaped-field buncher and timed ion gate sit after the electrostatic analyzer, generating the first time-focus for the time-of-flight measurement. By cooling the ions prior to passing them into the buncher, the energy spread is reduced to within 1 eV, removing any effects of the sample.

As a result of this design, the distance traveled by the ions prior to reaching the spectrometer has no effect on the measured flight time. The mass spectrometer is therefore independent of the local environment on the sample.  This thus enables the J105 to analyze otherwise challenging samples, producing consistent, accurate results with no loss of resolution.

The following examples demonstrate the capabilities of the J105 to analyse rough, uneven, topographic samples without compromising mass resolution or mass accuracy.

Example #1: M1.6 Screw Thread

The figure below shows the total ion image as obtained from an M1.6 screw thread. This sample was chosen for its difficult topography, having a thread depth of approximately 300 μm. Despite the challenging nature of the sample, highly consistent results are obtained no matter what point on the surface is chosen.

The spectrum surrounding the Cr+ peak (nominally 51.94051 amu) is shown to the right of the image for a range of selected areas across the sample. The areas chosen span the full 300 μm depth of the screw thread.

Despite this large variation in height across the sample, the variation in peak center across the chosen points is only 0.0002 amu, which corresponds to a mass error of less than 5 ppm across the complete depth range.

Total Ion Image of M1.6 screw thread

Left Total ion image of an M1.6 screw thread, obtained on the J105 SIMS. Right Comparison of the Cr+ peak (nominal mass 51.94051 amu) from spectra obtained at various locations across the screw thread sample (height variation approximately 300 μm). Variation in mass accuracy across the 300 μm depth range of the sample shows less than 5 ppm deviation from the nominal value.

Example #2: Frog Embryo

The figure below shows the total ion image of a frog embryo during multiple stages of development over a 5.5 hr period. The embryo is a sphere approximately 1.1 mm in diameter – a very challenging samples for ToF SIMS.

Despite the challenging nature of the sample, uniform signal is obtained from almost all areas of the hemisphere without affecting the mass calibration.

Total ion image of developing Xenopus laevis embryo

Total ion image of a Xenopus laevis embryo at various stages of development, from 2 cells to many. Embryo is approximately 1.1 mm in diameter, in a field of view of 1200 x 1200 μm. Strong signal is obtained from all parts of the hemisphere, and the mass calibration did not change over the image. This research was originally published in the Journal of Lipid Research. Tian, H. J. Lipid Res. 2014. 55: 1970-1980. Published with permission from original author.

Conclusions

Rough, uneven, or highly topographic samples can cause problems in ToF SIMS analysis, resulting in loss of mass accuracy and mass resolution. Unfortunately, most real world samples tend not to be perfectly uniform.

The J105 SIMS solves this problem by decoupling the mass spectrometer from the primary beam, thereby removing any effects of the sample from the time of flight measurement. The result is a ToF SIMS capable of analyzing rough, topographic samples without compromising mass resolution or mass accuracy.

The J105 SIMS combines flexibility with high performance, delivering unprecedented performance on biological and inorganic samples alike. For more information, visit the J105 product page, read our Application Notes, or Contact Us to discuss your interest further.

3D SIMS overlay of a single Tetrahymena cell in water ice on a silicon substrate.

Depth Profiling in ToF-SIMS with the J105 SIMS

Depth profiling is a powerful technique in surface analysis for examining interfaces and exploring the 3-dimensional structures of a material, to which SIMS is uniquely suited.

Many modern instruments are equipped with a sputter ion beam in addition to their primary analysis beam for depth profiling. This enables users to perform “etch-cycles” in-between analysis cycles, thereby building up a stack of 2D images and generating a 3-dimensional view of the sample.

There are limitations to the use of SIMS for depth-profiling however. Unlike non-destructive techniques such as XPS, SIMS alters the surface during analysis. In fact, fragmentation and migration of species can occur up to 20 nm below the surface. Etch-cycles therefore perform several duties. 1. They enable 3D image capture in a realistic time-frame, and 2. they remove the damaged layers, leaving behind a relatively clean surface for the subsequent analysis cycle. This process has some obvious drawbacks, most notably that a significant amount of material is lost during each cycle. This can severely limit the depth resolution.

As the J105 SIMS uses a dual-stage ToF analyzer, any beam may be employed as the primary analysis beam, including C60 and gas cluster beams. In this case, additional etch-cycles are not required – analysis and low-damage etching are continuous and concurrent. This means that no data is lost and all the material is sampled, making the J105 SIMS an extremely accurate tool for depth profile analysis.

No etch-only cycles on the J105 SIMS

As the J105 uses a DC beam, there is no need to interlace an etching beam with the analysis beam as analysis and low-damage etching are continuous and concurrent.

Organic Samples

In 2015, NPL led a VAMAS study comparing compositional analysis of an organic layered structure using both SIMS and XPS. The stack consisted of layers of Irganox 1010 and either Irganox 1098 (denoted MMK) or Fmoc-pentafluoro-Lphenylalanine (denoted MMF) in well defined ratios.

Chemical structure of Irganox 1010 and Irganox 1098

Chemical structure of (a) Irganox 1010, and (b) Irganox 1098.

Compared with a number of instruments using a Bin+ analysis beam, the J105 SIMS, using a 40 kV Ar4000+ beam, achieved the highest depth resolution of any instrument, demonstrating 12.8 nm. The figure below shows an example depth profile of such an organic stack structure.

VAMAS depth profile using J105 SIMS

Depth profile of a stacked Irganox 1010/Irganox 1098 sample. The J105 SIMS demonstrated a depth resolution of 12.8 nm – the highest of any instrument in the study.

HeLa cells

Depth profiling may also be used to produce 3D chemical maps of complex structures. The example below shows a 3D reconstruction of HeLa cells – an immortal human cervical cell line. The lipid phosphocholine head group (m/z 184.1), shown in green, represents the cell membrane, while the nucleic acid adenine (m/z 135.1), shown in red, represents the nucleus.

3D chemical map of HeLa cells

3D chemical mapping of HeLa cells using a 40 kV C60 beam. Phosphocholine (m/z 184.1 – green) and adenine (m/z 135.1 – red) represent membrane and nucleus respectively. Data courtesy of John Fletcher.

Such 3D reconstructions, detailed by Fletcher, Rabbani, and Henderson et al. here, require extensive post-processing of the data to convert the 2D layers – which have no height information – into a 3D image that is representative of the actual structure. The result, however, is a stunning reconstruction of the cell structure, that is made possible by the ability of the J105 SIMS to capture a complete spectrum from each voxel without loss of data.

The animation below reveals how the cell membrane is gradually removed to reveal the internal contents – in this case, depicted by the void created in cells.

Animated GIF of depth profile through HeLa cells with J105 SIMS

Depth profiling through HeLa cells using a 50 kV (CO2)5000+ beam at 1 μm per pixel. The lipid layer (phosphocholine) @ m/z 184.1 – representing the cell membrane, is gradually etched away, revealing the contents of the nucleus (adenine) @ m/z 135.1. Data courtesy of Hua Tian.

Inorganic Samples

The properties of poly-atomic ion beams such as gas clusters and C60+ are particularly useful for the analysis of organic samples, and have allowed molecular depth profiling to become a reality. However, that does not mean that such ion beams do not also provide benefits on inorganic samples. In fact, C60 ion beams are uniquely useful for use on both organic and inorganic samples.

Sputtering under C60+ has been shown to be less sensitive to variation in material, incidence angle, and crystallinity of the sample and has been show to offer improvements in quantitation and reproducibility on glass/mineral standards, and therefore provide an excellent solution for depth profiling on these types of samples.

NIST Ni:Cr Standard

The performance of 40 keV C60+ on a metal multilayer sample is demonstrated on the Ni:Cr standard reference material from NIST. The sample comprises alternating layers of Ni and Cr on a silicon wafer substrate. The plot on the left shows the signal from the m/z 52, and 58 mass channels corresponding to the Cr and Ni respectively. The Ni layers are 66 nm thick and the Cr layers 53 nm. A depth resolution of ca. 5 nm is calculated (16-84% rising edge of 1st Ni layer). This is close to the expected crater depth of a single C60+ impact.

Depth profile through NIST Ni:Cr standard with J105 SIMS

40 keV C60+ depth profile through the NIST Ni:Cr standard showing < 5 nm depth resolution.

Mixed Metal Oxide Sample

The example below is a depth profile through a mixed metal oxide material of relevance in hydrogen fuel cell fabrication. The experiment was performed using a C60+ 40 keV primary ion beam rastered over a 150 × 150 µm2 field of view. The J105 provides excellent signal to noise as all the material is collected during the depth profile similar to a dynamic-SIMS instrument but with the parallel mass detection and mass range of a time-of-flight analyzer. Decoupling the mass spectrometry from the ion generation process also means that the mass resolution is not compromised, and mass accuracy and calibration does not drift as the sample is eroded. Depth profiles and 3D images are generated retrospectively from the “image stack” with the outermost pixels discarded to remove any crater-wall artifacts.

Depth profile through fuel cell using J105 SIMS

Variation in secondary ion signal as a function of primary ion beam fluence for Cr+, Fe+, CeO+, and TiO+ displayed in the profile and 3D reconstruction in green, grey, red, and blue respectively. 150 x 150 μm2 field of view, z depth approximately 600 nm.

The composition of this sample is in fact much more complex than shown in the above plot and 3D rendering, and contains Al, Mn, Cr, Ce, Fe, Ti and oxides of these metals all of which are observed clearly in the mass spectra through the profile, in addition there is a thin layer of Na and K on the very surface of the sample (not shown).

Conclusions

The J105 SIMS provides an excellent solution for depth profiling of samples. The ability to use a variety of cluster beams for continuous and concurrent analysis and low-damage etching, without the need for “etch-only” cycles, enables high depth resolution on any sample type. If you would like more information, or would like to speak to a member of our team, please get in touch via our Contact page.

SIMS overlay image of rock surface

ToF-SIMS Analysis on Insulating Samples

Performing ToF-SIMS analysis on insulating samples can be particularly challenging. Surface charge can impact the accuracy of the results, or in the worst cases prevent any results being obtained. Thanks to its unique ToF design, the J105 SIMS is capable of imaging highly insulating samples without loss of signal or performance.

The Problem

Insulating samples provide a particular challenge for the ToF-SIMS user. A charged beam of particles impinging a surface necessarily produce a surface charge, both by emission of secondary electrons and by transfer of charge to that surface. For a metallic sample this charge is easily dissipated into the surrounding bulk, however this is not the case for insulating materials where there are no free electrons to neutralize the charge. The result is a build up of surface charge that can have a number of deleterious effects on a ToF-SIMS experiment.

For instruments where the time reference is determined by the primary beam pulse, regions of surface charge can affect the flight time of ions ejected from the surface, thereby negatively impacting on the mass-accuracy of the experiment. This effect is particularly pronounced around regions of topography where field strength is enhanced – in the worst case this leads to highly distorted fields preventing ions from ever reaching the detector. The most prominent effect of surface charge, however, is the suppression of secondary ion signals due to field suppression and/or neutralization prior to extraction.

The Solution

Thanks to its unique ToF design, the J105 SIMS is capable of imaging highly insulating samples without loss of signal or performance.

ToF Design

As the J105 does not rely on a pulsed primary beam as a time reference, temporal aberration in the extraction gap, caused by sample charging or other effects, has no impact on the mass accuracy. This enables a large extraction gap and a low extraction field, which in turn mean the extraction optics can accept ions leaving the sample up to very large angles from the normal, increasing transmission. The overall result is consistent performance across the entire sample, even those areas with lots of topography.

Insulating Sample Fig

Schematic illustration of the various means by which the J105 SIMS can compensate or accommodate charging samples.

Charge Compensation

Charge compensation/neutralization may be used to alter the effects of surface charge and improve signal on charging samples; this can be achieved by sample biasing and/or low-energy electron bombardment. In sample biasing, the sample stage is biased up to ± 100 V to compensate for the field created by the surface charge. This requires that the bias voltage is able to reach the sample surface, and so may not be effective on thicker samples or in situations where conductive tracks are prohibited.

Alternatively, pulses of low-energy electrons may be interlaced between the primary beam, in a pseudo-DC mode, to directly neutralize the charge on the surface. This has advantages in that the electrons are naturally drawn to localized pockets of charge, providing a “self-limiting neutralization” mechanism. One must be careful however, as even low-energy electron beams may result in damage to fragile organic structures. Generally speaking, in most situations, it is best to use a combination of both techniques, experimenting to find the correct balance for each individual sample.

Insulating Samples

Sodium Citrate Crystal

The figure below shows SIMS images of a single sodium citrate crystal on a substrate of double-sided tape – both highly insulating materials. Using a combination of sample bias and electron flood gun to neutralize surface charge, good signal levels are obtained from all areas without loss of mass accuracy or mass resolution, despite the highly topographic surface, enabling the identification of small micro crystals of NaCl within the sample.

Sodium Citrate SIMS image

SIMS images of a single sodium citrate crystal on double-sided tape, obtained with a 40 kV C60+ beam, showing (left) total secondary ion image, (middle) Na2OH (m/z 62.985), and (right) Na2Cl (m/z 80.95 + 82.95).

Polished Rock Surface

The second example shows a SIMS overlay image of a polished rock surface. The image is a 1 x 1 mm area, taken at 1024 x 1024 pixels using a 40 kV C60+ beam. A combination of electron flood gun, and sample bias was used in this instance, with +60 V applied to the stage in order to compensate for the effects of charging. A number of elements and minerals are identified in the image, including K, Ca, SiOH, Mn, Fe, and TiO.

SIMS image of rock surface

SIMS overlay image of a polished rock surface, imaged with a 40 kV C60+ beam. A combination of a low-energy electron flood gun and sample biasing was used to optimise imaging conditions for this insulating sample. Elements and minerals visible include: K (green), Ca (yellow), SiOH (red), Mn (cyan), Fe (white), and TiO (blue). Data courtesy of Dan Graham & Tina Angerer.

Mass Accuracy

Despite the charging nature of this sample, consistent mass accuracy is maintained across the data set. This is illustrated in the figure below, where three peaks, TiO, Na2O, and C5H2 respectively, are identified with sub 5 ppm accuracy. As evidenced by the overlay image accompanying the spectrum, all three peaks are spatially distributed throughout the sample. Spatial resolution, mass resolution, and mass accuracy all combine to make peak identification a routine procedure, even on challenging samples such as this.

Mass accuracy on insulating samples

SIMS overlay image & spectrum of polished rock surface. Even on highly insulating samples such as this, high mass accuracy (<5 ppm) is maintained at all points, enabling identification of numerous mineral fragments and organic species. Data courtesy of Dan Graham & Tina Angerer.

Conclusions

Real-world samples – where sample charging and topography are unavoidable – have always posed a challenge to reliable ToF-SIMS analysis. Now, through innovations in instrument design and engineering on the J105 SIMS, analysis of such samples is becoming routine, giving you more time to focus on solving the important problems. If you would like more information, or would like to speak to a member of our team, please get in touch via our Contact page.