Back to the basics of time-of-flight secondary ion mass spectrometry of bio-related samples. I. Instrumentation and data collection

A. Mounting samples—Conductor or insulator?

Once samples have been prepared, they need to be mounted for analysis. At this point, the first decision to make is whether to mount the samples as insulators or conductors. Mounting as a conductor means that the samples are placed in electrical contact with the sample holder where there is a direct path to ground. This is often done using a screw or clip attached to the sample holder. Typical conducting samples include metals and thin films on silicon wafers. Mounting as an insulator means that the samples are placed on the sample holder using a barrier that prevents direct contact with the sample holder. This is often done using the double sided tape. A large piece of tape is not necessary. Small pieces of tape can be used as long as no part of the sample comes in contact with the sample holder. The decision to treat a sample as an insulator or conductor should be made based on the material properties. Any material that does not conduct or may not conduct electrons should be mounted as an insulator. Mounting a nonconducting sample or semiconducting sample as a conductor can cause odd charging effects and artifacts in the data. Multilayer samples that contain both conducting and insulating materials should be mounted as an insulator during depth profiling. It is also noted that all samples can be treated as insulators and often this is the best choice. However, keep in mind that mounting a sample as an insulator requires the use of charge neutralization, which can cause sample damage.68–7168. S. P. Harvey, J. Messinger, K. Zhu, J. M. Luther, and J. J. Berry, Adv. Energy Mater. 10, 1903674 (2020). https://doi.org/10.1002/aenm.20190367469. I. S. Gilmore and M. P. Seah, Appl. Surf. Sci. 203–204, 600 (2003). https://doi.org/10.1016/S0169-4332(02)00774-270. C. Zhou, A. Trionfi, J. W. P. Hsu, and A. V. Walker, J. Phys. Chem. C 114, 9362 (2010). https://doi.org/10.1021/jp911402u71. R. Havelund, M. P. Seah, A. G. Shard, and I. S. Gilmore, J. Am. Soc. Mass Spectrom. 25, 1565 (2014). https://doi.org/10.1007/s13361-014-0929-5

B. Putting samples into the instrument

ToF-SIMS is a UHV analysis method. Due to this, all samples put into the system must be able to be pumped down to UHV vacuum levels before the analysis can be carried out. Although many instruments will work at pressures in the high 10−8 mbar range, it is best to operate with vacuum levels in the 10−9 or low 10−8 mbar range. This will minimize potential contamination in the chamber and increase the lifetime of vacuum sensitive electronics. For most samples, pump down times from atmospheric pressure to a pressure sufficiently low to place them into the main analytical chamber is 30 min–1 h. This will be true for most solid samples (metals, some polymers, and thin films) and nonporous samples. Samples that are porous or hydroscopic can take extended time to pump down to a sufficiently low vacuum level. In extreme cases, it may be necessary to reduce the sample size in order to achieve an acceptable vacuum level in a reasonable amount of time. For example, some hydrogel samples are capable of storing so much water that even a 2 × 2 × 1 mm3 sample can take 24 h to pump down to the UHV range. In these cases, it is important to schedule pump down time when it will cause minimal disruption to the overall analysis schedule. Samples such as these are good to load on a Friday afternoon and then allow them to pump over the weekend. The other option for such samples is to analyze them frozen. More information on frozen hydrated experiments can be found in the literature including Refs. 55–5855. A. G. Shard, S. J. Spencer, S. A. Smith, R. Havelund, and I. S. Gilmore, Int. J. Mass Spectrom. 377, 599 (2015). https://doi.org/10.1016/j.ijms.2014.06.02756. R. G. Cooks and K. L. Busch, Int. J. Mass Spectrom. 53, 111 (1983). https://doi.org/10.1016/0020-7381(83)85106-757. K. Takahashi, S. Aoyagi, and T. Kawashima, Surf. Interface Anal. 49, 721 (2017). https://doi.org/10.1002/sia.621458. J. E. Baio, D. J. Graham, and D. G. Castner, Chem. Soc. Rev. 49, 3278 (2020). https://doi.org/10.1039/D0CS00181C, 1919. A. M. Piwowar, S. Keskin, M. O. Delgado, K. Shen, J. J. Hue, I. Lanekoff, A. G. Ewing, and N. Winograd, Surf. Interface Anal. 45, 302 (2013). https://doi.org/10.1002/sia.4882, and 72–7472. T. L. Colliver, C. L. Brummel, M. L. Pacholski, F. D. Swanek, A. G. Ewing, and N. Winograd, Anal. Chem. 69, 2225 (1997). https://doi.org/10.1021/ac970174873. M. Taylor, D. Scurr, M. Lutolf, L. Buttery, M. Zelzer, and M. Alexander, Biointerphases 11, 02A301 (2016). https://doi.org/10.1116/1.492820974. R. Metzner, H. U. Schneider, U. Breuer, and W. H. Schroeder, Plant Physiol. 147, 1774 (2008). https://doi.org/10.1104/p.107.109215

C. Positive or negative ion data?

It is not uncommon for a new potential ToF-SIMS user to make an analysis request saying they “only need the positive ion” or “only need the negative ion” data. This often results from the user having read a paper that specified that only one polarity or the other contained “useful” information. Although it may be that for a given sample type, one ion polarity or the other may contain more “characteristic” peaks for the material of interest, it is always important to collect data in both polarities. For example, cations form predominantly positive ions (Na+, Ca+, K+, etc.), while halides form predominantly negative ions (Cl−, Br−, etc.). The main fragments from amino acids show up in the positive ion data, and the main molecular ions and molecular ion clusters from self-assembled monolayers typically show up in the negative ion data. Not taking data in both polarities will omit valuable information about the samples including, in some cases, the presence or absence of contaminants.

D. Data collection parameters

When working with biological samples, timing is important. Biological samples can degrade even while in a vacuum system. For analysis that takes significant time (such as tissue imaging), it is best to load one sample at a time to avoid degradation within the vacuum chamber. It is also recommended to schedule sample analysis times on a consistent basis, so the time between sample preparation and analysis is always the same.1111. L. J. Gamble, D. J. Graham, B. Bluestein, N. P. Whitehead, D. Hockenbery, F. Morrish, and P. Porter, Biointerphases 10, 019008 (2015). https://doi.org/10.1116/1.4907860 This will help minimize variations seen due to sample degradation or component migration. When working with tissues, it is best to keep the tissue frozen until the day of analysis and then cut the necessary sections and take them immediately for analysis. If this is not possible, then store the cut sections in the freezer and use the same time between removing them from the freezer and starting the analysis. When doing this, it is important to take into account the pumping time of the instrument.

It is also important to minimize damage from the analysis beams. This can become challenging when dealing with heterogeneous or patterned surfaces where it is desirable to collect the positive and negative ion data from the same spot. To do this, and maintain the total ion dose below the static limit, it is recommended to collect the “most important” polarity first at a dose of ∼5 × 1011 ions/cm2 and then other polarity at ∼5 × 1011 ions/cm2. To minimize the cycling of instrument electronics and reduce data collection time, it is best to collect all data from one ion polarity and then all data from the other polarity, although this is not required.

Typical spot size for ToF-SIMS spectral analysis is 100 × 100 μm2, although the actual spot size used can depend on the goals of the analysis and the specific samples. For example, when analyzing features that are 100 μm in diameter, it would be recommended to increase the spot size to at least 150 × 150 μm2 or 200 × 200 μm2. When deciding on a spot size, it is important to use other considerations including minimizing topographical artifacts and optimizing charge neutralization. Both topography and inadequate charge neutralization will reduce the obtainable mass resolution. In some ToF-SIMS systems, even a moderate difference in topography can result in double peaks at every mass because ions from the same species will originate from different heights on the sample.

In general, decreasing the spot size will increase local charging, unless the dose of primary ions hitting the surface is also reduced. Corollary to this, improving charge neutralization can often be achieved by reducing the dose (lower current, decrease pixel dimensions) or increasing the spot size. Reducing the spot size can help minimize contributions from areas with different heights when working with samples with high surface roughness or topography. One way of doing this is to start with the largest spot size possible and do an initial scan to locate areas with a uniform signal. Then, focus in on a uniform area using a spot size that fits within that area. It is important to consider the total ion dose when doing this since reducing the spot size will increase the ion dose unless the current, number of shots/pixel, or the pixel dimensions is reduced.

When collecting negative ion data, it is often useful to leak a small amount of argon gas into the analysis chamber along with using the electron flood gun to aid in charge neutralization. We have found that when starting with a base pressure of around 5 × 10−9 mbar, raising the pressure to around 5 × 10−7 mbar with argon gas helps provide adequate charge neutralization with negative ion data. Without the use of argon, it is often difficult to find settings that avoid the dark bands that often show up on the edges of negative ion images due to charging on insulating samples.

Some detectors have a limit to how much signal they can accommodate per pulse of ions. In addition, most modern primary ion guns are very efficient in generating secondary ions. The result of this is that unless settings are adjusted by the user, it is common to saturate peaks when using ToF-SIMS. Peak saturation causes the detector to shut off momentarily to reset. This causes a loss in signal for any peaks that are within the dead time of the detector. Saturation also makes it, so the measured number of counts from the saturated peak is not accurate. It is critical that the user understands that any saturated peak(s) cannot be used for any analysis beyond noting the presence of the peak(s). Saturated peaks should not be used for calibration, in making peak intensity ratios, comparing intensities or in any type of MVA. In addition, when using MVA methods, if any peaks are saturated, they should not only be omitted from the analysis, but users should not normalize to the total intensity since it will be influenced by the saturated peaks. In those cases, it is best to normalize the data to the sum of selected peaks or the total intensity minus the intensity of the saturated peaks.

With some systems, it is possible to allow a small degree of saturation if the software is capable of correcting the intensity based on the known distribution of the data. However, in practice, it is best to not allow peak saturation. One potential exception to this is when the peaks that saturate are not of interest to the main study. For example, the sodium and potassium signals in ToF-SIMS spectra from tissues are often saturated. In order to avoid saturation in these peaks, the beam current sometimes has to be lowered significantly (to around 0.05 pA or lower), which greatly increases the analysis time. In such cases, it is often acceptable to allow the sodium and potassium signals to saturate as long as these signals are not used in any semi-quantitative calculations and the fact that they were saturated is noted in any reported data.

A full discussion of choosing an appropriate primary ion is beyond the scope of this Tutorial; however, there are numerous studies comparing various ion sources.6,75–796. S. Muramoto, D. J. Graham, M. S. Wagner, T. G. Lee, D. W. Moon, and D. G. Castner, J. Phys. Chem. C 115, 24247 (2011). https://doi.org/10.1021/jp208035x75. S. Rabbani, A. M. Barber, J. S. Fletcher, N. P. Lockyer, and J. C. Vickerman, Anal. Chem. 83, 3793 (2011). https://doi.org/10.1021/ac200288v76. S. S. Rabbani, I. B. Razo, T. Kohn, N. P. Lockyer, and J. C. Vickerman, Anal. Chem. 87, 2367 (2015). https://doi.org/10.1021/ac504191m77. J. G. Son, S. Yoon, H. K. Shon, J. H. Moon, S. Joh, and T. G. Lee, Biointerphases 15, 021011 (2020). https://doi.org/10.1116/6.000010578. J. Cheng, J. Kozole, R. Hengstebeck, and N. Winograd, J. Am. Soc. Mass Spectrom. 18, 406 (2007). https://doi.org/10.1016/j.jasms.2006.10.01779. N. Wehbe, T. Tabarrant, J. Brison, T. Mouhib, A. Delcorte, P. Bertrand, R. Moellers, E. Niehuis, and L. Houssiau, Surf. Interface Anal. 45, 178 (2013). https://doi.org/10.1002/sia.5121 In general, cluster ions (Au3+, Bi3+) will generate a higher yield of secondary ions for most samples than monoatomic ions (Au1+, Bi1+). One exception to this is that when analyzing a thin film on a hard substrate (e.g., silanes on silicon or self-assembled monolayers on gold), a monoatomic primary ion will often provide a higher yield of secondary ions and, in particular, molecular ions. This is due to the fact that with a hard substrate, the cluster ions deposit more energy across the surface causing increased fragmentation, while the monoatomic ions penetrate deeper within the surface and allow for molecular emission due to energy reflected back toward the surface from the impact zone. There is some evidence that large clusters from GCIB sources are capable of emitting larger fragments and intact species from biomolecules77,80,8177. J. G. Son, S. Yoon, H. K. Shon, J. H. Moon, S. Joh, and T. G. Lee, Biointerphases 15, 021011 (2020). https://doi.org/10.1116/6.000010580. Y. Yokoyama et al., Anal. Chem. 88, 3592 (2016). https://doi.org/10.1021/acs.analchem.5b0413381. S. Sheraz, H. Tian, J. C. Vickerman, P. Blenkinsopp, N. Winograd, and P. Cumpson, Anal. Chem. 91, 9058 (2019). https://doi.org/10.1021/acs.analchem.9b01390 and, therefore, may be useful for biological samples. The main sources used for collecting data in modern ToF-SIMS instruments are summarized in Table III. Not all types of analysis are shown in the table; however, it should provide a general overview of the current state of the art.Table icon

TABLE III. Typical analysis beams for standard experiments.

IonType of sourceMost common useBi1+, Au1+LMIGaAnalysis of thin films on hard substratesBi3+, Bi3++LMIGaGeneral purpose source for a wide range of samples and high resolution imagingC60+, C60++—Depth profiling of many organic materialsC60+, C60++—Analysis of many organic materialsArgon clustersGCIBbDepth profiling of organic materialsArgon, CO2, or water clustersGCIBbAnalysis of biological materials. Can provide enhanced high mass signal compared to LMIG and C60 sourcesCs+LMIGInorganic analysis. Enhanced signal from negative ion species

留言 (0)

沒有登入
gif