Photoresist(PR) that used in semiconductor etch process consists of aromatic hydrocarbon polymers. Photoresist should be eliminated during the development process after exposure, and photoresist residues that remain after development cause serious problems along the following process. Thus, it is important to analyze photoresist residues accurately in each semiconductor process.

A conventional ToF-SIMS depth profiling using the Cs+ sputter gun is not appropriate for the preservation of molecular information due to its collision cascase effect. Therefore, by Cs+ sputtered depth analysis, information on the source of hydrocarbon in photoresist cannot be obtained accurately because the source of hydrocarbon can be generated not only from photoresist but also from the contamination from environmental factors. Depth profiling using GCIB (Gas cluster ion beam) does not destruct the original molecular structure of residues.[1] Therefore, the method is appropreate for depth profiling of photoresist residue analysis.

KrF is one of the photoresist used in semiconductor manufacturing process which transforms to hydrophilic polymer after developent process. Depth profile analysis using GCIB as sputter gun gives information as its original molecular structure (119m/z). By using GCIB depth profiling method, instead of obtaining destructed molecular information, the original polymer molecular information cas be obtained.

The development of cluster and polyatomic projectiles such as SF₅⁺and C₆₀⁺ has made it possible to depth profile numerous organic materials using SIMS [1,2]. The use of polyatomic projectiles has produced better results when compared to monatomic projectiles for depth profiling of organic and biological materials because the energy of monatomic projectiles penetrates below the sample surface and therefore the sub-surface chemistry of the sample is destroyed. However, the use of C₆₀⁺ ion beam, whilst successful for certain compound classes, is problematic for the analysis of organic electronic materials such as aluminium hydroxyquinolate (Alq₃) and organic light emitting diodes (OLEDs). In recent years it has been reported that the depth profiling of organic semiconductors and OLEDs has been successful with large argon gas cluster ion beams [3].

In this paper we present results from the semiconductor poly(4,4-diphenyl)-(2,4-dimethylphenyl)-amine (PTAA), depth profiled using a variety of primary ion beams at 20 keV including C₆₀⁺, Ar_2000-4000⁺ and (H₂O)_2000-4000⁺. The use of the C₆₀⁺ projectiles produced a rapid decay in the molecular ion of the repeat unit and large characteristic fragment ion signals. Consequently, the depth profile of this sample is unable to be obtained because of carbon deposited on sample and cross linking. More informative depth profiles were produced with the gas cluster beams.

[1] C.M. Mahoney Mass Spec. Rev. (2010) 29, 247– 293

[2] A.G. Shard, R. Havelund, M.P. Seah, S.J. Spencer, I.S. Gilmore, N. Winograd, D. Mao, T. Miyayama, E. Niehuis, D. Rading, and R. Moellers Anal. Chem. ( 2012 )84, 7865-7873

[3] S. Ninomiya, K. Ichiki, H. Yamada, Y. Nakata, T. Seki, T. Aoki and J. Matsuo, Surf. Interface Anal. (2011) 43, 95-98

The development of the large cluster ion beams technique has led to revolutionary progress in surface analysis techniques, such as SIMS and X-ray photoelectron spectroscopy (XPS). Molecular depth profiling and three-dimensional (3D) imaging are widely utilized for polymers, organic semiconductors and biological materials, because of the significant reduction in surface damage caused with using large cluster ion beams. The dual beam analysis technique is widely used these days, because of the difficulty of focusing large cluster ion beams. Furthermore, it is possible to decouple the analysis beam condition and the sputtering beam condition. However, most of the sample material is wasted during sputtering. This is unworthy for molecular depth profiling, but turns out to be a significant issue for mass imaging.

On the other hand, single beam analysis, in which a large cluster ion beam is utilized for both sputtering and analysis, is an alternative choice for organic SIMS, in particular for mass imaging. A fine-focused Ar cluster ion beam down to ~1 mm, has been reported recently [1]. This ion beam was combined with the orthogonal acceleration technique, in which there is no need to pulse a primary ion beam to obtain time-of-flight (TOF) mass spectra. Moreover, the mass resolution is decoupled of the pulse width of the primary ion beam, and high mass resolution is achieved without bunching. It is very important to maintain the ion current density of the primary beam, because the primary ion dose used in single beam analysis exceeds the static limit dose (~10¹² ions/cm²).

Biological materials, such as amino acid and lipids were measured with both the dual and single beam analysis techniques. The ratio between the analysis beam and sputtering beam was optimized to maximize secondary molecular ions. In the case of dual beam analysis, the accumulated molecular ion counts were inversely proportional to the ratio around 1,000. However, the accumulated molecular ion counts were saturated below the ratio of 100, because surface damage could not be completely removed by sputtering. These results indicated that most of sample material was not used for analysis. In case of single beam analysis, the accumulated molecular ion count was simply proportional to the ion dose.

The comparison between the single and dual beam analysis techniques will be reported and its strangeness and weaknesses will be discussed.

Acknowledgements

This work is partially supported by SENTAN of JST.

[1] J. Matsuo, S. Torii, K. Yamauchi, K. Wakamoto, M. Kusakari, S. Nakagawa, M. Fujii, T. Aoki and T. Seki, Applied Physics Express, 7, 056602 (2014)

This contribution reports the application of ToF-SIMS to the study of the chemical and structural features of polystyrene coatings deposited by plasma near atmospheric pressure. This plasma-polymer material is mainly used in the preparation of electronic devices, as protective films and in medical applications. However, a complete control of the influence of the plasma parameters for the tuning of the desired properties of these coatings is not reached yet. The ToF-SIMS technique is the perfect candidate for monitoring the chemistry of these plasma-deposited films, both on the surface and along the depth, the latter being made possible by the advent of large noble gas clusters for “damageless” sputtering. Historically, several structural indicators, given by ratios or normalized sums of secondary ion intensities, have been developed to extract information about the unsaturation, branching and cross-linking of the polymer films from the SIMS surface mass spectra. [1,2] However, their evolution as a function of the plasma parameters depends strongly on the chemical/structural nature of the precursor material and their blind use can lead to contradictory results, like in the case of this study. The present work proposes the application of principal component analysis (PCA) as a reliable method for the data treatment of “bulk” spectra of plasma-polymerized polystyrene (pp-PS) coatings. The pp-PS synthesis was conducted in a dielectric barrier discharge (DBD) from styrene carried by Ar gas into the reactor and varying uniquely the injected power in the range 10-80 W. The depth-profiles were performed by sputtering with 10 keV Ar3000+ projectiles. The related reconstructed “bulk” mass spectra, obtained using 30 keV Bi5+ ions, are expected to be more representative of the film chemistry than those acquired on the surface. The PCA suggests that the increase of the power induces a loss of aromaticity accompanied by an increase of cross-linking and oxidation. Finally, a comparison between the trends of the PCA and the classical structural indicators is performed, in order to deduce more adequate criteria for the study of the investigated plasma-polymer system.

[1] U. Oran et al., Surface and Coatings Technology 2005, 200(1), 463.

[2] D. Merche et al., Plasma Science, IEEE Transactions on 2009, 37(6), 951.