SIMS2015 Session CT1-MoM: Complementary Techniques and Multi-Technique Approaches

While SIMS is amongst the most sensitive surface analysis techniques, like all techniques, it has its limitations. Lateral resolution is limited in commercial instruments to several tens of nanometers with a fundamental limit of around ten nanometers. Quantification is achieved via the use of standards which may not be convenient for some samples. To achieve high spatial resolution, depth resolution must be sacrificed or vice versa. Many of the limitations of SIMS are intrinsic, and simply designing better instruments will no longer be enough to meet the challenges presented by complex samples. Fortunately, many of the limitations of SIMS can be overcome by combining it with other techniques. By combining SIMS data with high resolution microscopy techniques such as Transmission Electron Microscopy (TEM) or Helium Ion Microscopy (HIM) we can combine the advantages of high spatial resolution and high sensitivity. By using other analysis techniques e.g. Energy Dispersive Spectroscopy (EDS) we can facilitate quantification [1]. By combining SIMS with Atomic Force Microscopy (AFM) information, artifact free 3D reconstruction may be achieved even for complex samples with widely varying sputter yields (Fig. 1).

To avoid issues of contamination or sample reconstruction, to gain a maximum of information and to speed up analysis, multiple techniques need to be performed in-situ and ideally simultaneously. We have investigated three different combinations of SIMS based correlative microscope and developed three corresponding prototypes:

SIMS+AFM: Cameca NanoSIMS with integrated AFM module.

SIMS+HIM: Zeiss ORION NanoFab with dedicated SIMS extraction optics and in-house SIMS spectrometer.

SIMS+TEM: customized FEI Tecnai TF20 with a Magnum gallium FIB and in-house SIMS spectrometer (Fig 2).

In order to reach good detection limits when probing very small voxels in imaging applications, the ionization probability of the sputtered particles needs to be maximized. When using primary ion species such as Ga, He or Ne, the intrinsic yields are low compared to the ones found in conventional SIMS. However, the yields may be drastically increased by using reactive gas flooding. Our results show that both negative and positive ion yields obtained with He⁺, Ne⁺ and Ga⁺ bombardment can be increased by up to 4 orders of magnitude when using such reactive gas flooding (Fig 3).

Each of these three combinations offers unique advantages in terms of correlative microscopy. We will present the latest results and perspective for each one.

References

[1] Y. Kudravtsev et al. Nucl. Instrum. Meth. B 343 (2015) 153-157

Identification and localization of specific biomolecules such as peptides and lipids within cellular membranes is currently a major challenge in metabolomics and biological studies. With recent technological and methodological improvements imaging mass spectrometry methods including secondary ion mass spectrometry (SIMS) and matrix-assisted laser desorption ionization (MALDI) are now promising techniques in the field of molecular imaging of biological samples such as tissue sections or cells.

In biological investigations SIMS and MALDI often provide complimentary information. SIMS provides detection of small molecules at high spatial resolution, whereas MALDI is capable of ionizing larger molecules sample, albeit at reduced spatial resolution. Although these MS techniques have associated abilities there are some limitations with each method.

The chemistry of the sample is normally significantly altered in MALDI due to the addition of the matrix. This can be eliminated by using Au nanoparticles as a matrix (Nanoparticle Assisted-LDI) on tissue allowing lipid molecular ion species and relevant mass fragments to be imaged with enhanced spatial resolution.

To extend the useful mass range of SIMS imaging we have applied gas cluster ion beams (GCIBs) that have been developed to analyze and image organic compounds. The high energy Ar GCIB shows potential for ionizing molecular ion lipids and intact neuropeptides as well as high mass fragments of large proteins produced by on surface tryptic digestion. To complement the dual imaging methods, we have developed a sample preparation technique that should provide detection in single cell studies by both MALDI and SIMS techniques.

The major mass spectrometry techniques for nuclear materials analysis (TIMS, SIMS and ICP-MS) have traditionally been positioned against each other, with valid claims that a particular instrument provides the best overall data for a given application. This work takes a combinatorial approach that defocuses on the “who’s the best” rationale and enables complimentary analytical data streams, providing a more comprehensive signature. The intent of the short time frame and small amount of seed funding, an internal investment through Pacific Northwest National Laboratory's Signature Sciences Division Innovation Program, was to provide a proof of concept in direct solid sampling using a laser ablation system and subsequent analysis of that sub-sampled material by ICP-MS, SIMS and TIMS. The three step approach was fairly simple. (1) Simultaneous particle collection and LA-ICP-MS analysis - utilizing an in-line filter, a pulsed femtosecond laser generated particles, a fraction of the particles were collected on the filter membrane and the remaining particles were transported through the membrane and into the ICP for vaporization, atomization, ionization and detection, providing elemental information. (2) The collected particles from step #1 were extracted from the filter membrane by ultrasonication and then dispersed onto a laser scribed carbon planchet with unique position markers for co-location purposes - using dynamic SIMS ion microscope mode, an automated ion imaging (particle search) process was used to identify particles and determine respective isotope ratios. (3) Utilizing the isotopic screening SIMS approach in step #2 and use of the mentioned co-location specific markers, several particles were placed onto a TIMS filament for benchmark isotopic measurements. The proof of concept, experimental approach employed both traditional cellulose filter membranes and conductive carbon-based swipe materials for direct SIMS particle analysis (minimizing the amount of sample preparation necessary); NIST series glass (SRM 612), a luminescent glass and a fulgurite sample (to simulate nuclear debris) were analyzed under this experimental scheme. The work presented will showcase the successes and failures of the research, providing data that indicates that this combinatorial approach could be viable for nuclear debris analysis.

The unique properties of graphene have received substantial attention in recent years as a candidate for next-generation electronic materials. Despite intense interest, widespread use has remained elusive due to the absence of a band gap in intrinsic graphene [1]. Hydrogenation of graphene is considered among the most efficient ways to overcome this obstacle as it allows for tuning of electronic properties in conjugated carbon systems [2]. Recent work on plasma hydrogenation has, however, focused almost exclusively on the formation of sp² lattice defects as determined by the ratio of intensities of D and G peaks (I_D/I_G) in Raman spectroscopy [3], rather than changes in electronic properties. Here, structural and electronic characterization by Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS) and x-ray photoelectron spectroscopy, including x-ray induced Auger emissions, respectively, is coupled to Raman analysis to better understand the effect of defect formation on the electronic properties of plasma-treated graphene.

The fine structure of the carbon KLL emission is strongly affected by sample electronic properties and can be considered a fingerprint of carbon atom arrangement in evaluating extent of hydrogenation [4]. The D-parameter, calculated as the distance, D, between local maxima and minima in the first derivative of the C KLL spectrum, can thus be utilized to determine changes in the electronic structure due to plasma hydrogenation relative to pristine graphene [5]. The results of this study indicate substantial differences in the electronic properties of plasma-treated graphene are observed in the presence of similar sp² defect structures normally attributed to hydrogenation. Multivariate processing of ToF-SIMS images additionally provides information related to the uniformity of hydrogenation.

[1] Elias, D. C., Science, 323, 2009, 610-613.

[2] Jones, J. D. et al., Applied Physics Letters, 97(23), 2010, 233104.

[3] Zheng, L. et al. Chem Commun. 47(4), 2011, 1213-1215.

[4] Jackson, S. T., Nuzzo, R. G. Applied Surface Science, 90, 1995, 195-203.

[5] Mezzi, A., Kaciullis, S., Surface and Interface Analysis, 42(6-7) 2010, 1082-1084.

From the pace of the ice ages to how the carbon cycle has changed through time, much of what we know about earth history and climate dynamics is based on proxies. Recorded as trace element anomalies or as isotopic shifts, proxies are geochemical relationships that connect environmental parameters to the bulk composition of a natural sample. Recent developments in high-resolution imaging like NanoSIMS and atom probe tomography have shown that many environmental and biological materials are characterized by small-scale compositional heterogeneity, complicating the traditional interpretation of proxy-based records. Our lab has been harnessing the rich and newly accessible chemical data encoded in small-scale heterogeneity, developing a set of tools that can probe the mineral growth process and that can use these data to reconstruct past environmental conditions. We focus specifically on biomineralization, a complex chemical process occurring at the biological-mineral interface during skeletal self-assembly. The biomineral target of our study, the CaCO3 skeletons of a single celled organism called foraminifera, was specifically chosen because the preserved skeletons from this organism are widely used to develop climate records. I will present new research where we have been probing how protein-mineral interactions control small-scale heterogeneity. Specifically we have been using a combination of NanoSIMS and atom probe tomography (APT) to map elemental concentrations at the organic-mineral interface. By improving the interpretation of chemical signatures in biominerals, our research aims to improve the precision and accuracy of climate records, towards a better understanding of earth system dynamics. Collectively, our research on the mechanisms controlling small-scale heterogeneity in environmental materials has applications in climate science, in geochemistry, and in the design of complex biomimetic materials.

In a previous work, we showed that a depth resolution of about 4 nm was achievable on amino acids, namely tyrosine (Tyr) and phenylalanine (Phe) multilayers, using low energy (500 eV) Cs⁺ ions as sputtering beam. This is, to our knowledge, one of the best depth resolutions obtained on organic samples so far. However we think it should be possible to improve the depth resolution even further, since the damages produced by low energy Cs⁺ impacts are very shallow.

The depth resolution actually decreases when the order of the amino acids is inverted, the Tyr to Phe interface being always sharper than the Phe to Tyr one. As the depth resolution is largely determined by the beam induced topography, we followed an original multi-technique approach combining AFM, SEM, ToF-SIMS and optical microscopy to probe the topography in depth (roughness) but also laterally. The fields of view of these four complementary imaging techniques are respectively 3 µm, 20 µm, 25 µm and 800 µm, allowing a global outlook of the crater, as well as a local inspection of the phenomena occurring at the interfaces. Images were taken before sputtering, then at all interfaces. The initial rms roughness was lower than 1 nm, indicating very flat evaporated films. On the Tyr/Phe interface (the “good one”), the AFM roughness is about 4 nm, consistently with the measured ToF-SIMS depth resolution. On both the AFM and SEM images, 2 phases with equal proportions and lateral dimensions around 100 nm are observed, which could correspond to Phe and Tyr areas. On the Phe/Tyr interface (the “bad one”), the SEM images revealed 1-2 µm circular features apparently protruding out of a smooth background. These features were even visible with an optical microscope. Surprisingly, AFM images taken on the smooth areas showed no obvious topography and a very low roughness (<2 nm), proving that the low energy Cs⁺ bombardment did not develop significant roughness. High-resolution ToF-SIMS imaging of the protruding structures revealed they were residual islets of Phe in Tyr. A possible explanation for this could be the existence of crystallites in the Phe evaporated layers, with a lower sputtering yield than the surrounding background (presumably amorphous), therefore developing a micrometric topography. One can understand why such objects affect the depth profile, since the analysis raster represents a 100 µm square.