AVS2010 Session SS2-WeM: Electron, Photon and Ion Beam Induced Surface Modification

Electron beam induced deposition (EBID) is attracting increased interest as a single-step, direct-write process capable of depositing free standing, nanometer-sized structures with high spatial resolution (≈ 1 nm). However, the largest single limitation of EBID is that the deposited metallic nanostructures typically contain unacceptable levels of organic contamination which adversely affects the material’s properties, thereby limiting potential applications. A more detailed understanding of the electron stimulated, decomposition of EBID precursors is needed to rationalize the relationships between organometallic ligand architecture and the composition of the deposits. Such mechanistic detail will allow greater control over the composition and function of nanostructures deposited by EBID while providing the necessary rational design criteria to fashion new organometallic precursors. Other scientific issues associated with EBID include a lack of quantitative information on the fundamental surface dynamics and deposition processes as well as overcoming the challenges of implementing successful purification strategies. The application of surface analytical techniques enables changes in the metal oxidation state, surface and gas phase composition as well as structure and chemical composition that accompany electron interactions with organometallic precursors to be studied in situ and in real time. In addition, I will describe how these various electron stimulated processes can be studied to extract data on reaction cross-sections and kinetics, information that would facilitate the development and testing of predictive models that can accurately describe EBID. Mechanistic insights into purification strategies that can be obtained through surface analytical techniques are also detailed. Throughout, I will discuss unresolved challenges and opportunities associated with EBID.

A system of many electrons can display emergent phenomena that cannot be extrapolated from the behavior of independent electrons. Non-adiabatic resonant charge transfer is used to reveal this multi-electron behavior by employing singly charged alkaline earth ions in a low energy ion scattering experiment. The spin of the single valence electron of such an ion behaves as a magnetic impurity that interacts with the continuum of many-body excitations in the metal, resulting in Kondo and mixed valence resonances near the Fermi energy.^1,2 The occupation of these resonances is acutely sensitive to changes of surface temperature, which leads to an anomalous temperature dependence of the ion neutralization properbility. We report a maximum in the neutralization probability near 600 K for 2-4 keV Sr⁺ ions scattered from polycrystalline gold. Correlated electron effects, which have traditionally been manifest at low temperature, are observable well above room temperature because the projectile’s ionization level shifts and crosses the target’s Fermi energy as it approaches the surface where the interaction between the localized and extended electrons is very strong. This interaction becomes more pronounced as the work function of the surface is lowered by Sr implantation into the near-surface region. Further study will involve well-characterized single crystal surfaces that will enable detailed investigation of the relationships between the formation of correlated electron states and the atomic structure of the solid.

¹H. X. Shao, P. Nordlander, and D. C. Langreth, Phys. Rev. Lett. 77, 948 (1996).

²J. Merino and J. B. Marston, Phys. Rev. B 58, 6982 (1998).

The damaging effect of radiation in applications such as immobilization of toxic radionuclides, fission and fusion reactors, and radiation therapy has attracted considerable scientific interest. The search for radiation tolerant materials has revealed that fluorite structural derivatives, such as zirconia and cerium oxide, have the ability to accommodate high energy radiation induced defects. Particularly for biomedical applications, cerium oxide is of more interest because not only it can prevent the damage from radiation but also it has the ability to quench free radicals and reactive oxygen species produced as the result of high energy radiation by regenerative switching between the +3 and +4 valence states of cerium. However the underlying mechanisms of radiation interaction and the resulting physicochemical and structural changes of cerium oxide are not well understood. In order to gain a fundamental understanding of radiation tolerance of cerium oxide nanostructures, we explored the behavior of single and polycrystalline ceria under radiation. High quality single and poly crystalline nanoceria thin films were grown on YSZ and sapphire respectively by using oxygen plasma assisted molecular beam epitaxy and exposed to 2 MeV He⁺ radiation at fluences extending over three order of magnitude (10¹⁴ to 10¹⁷ ions/cm^-2) using the ion beam accelerator at EMSL, in Pacific Northwest National Laboratory. The chemical changes occurring in the thin films due to radiation exposure were characterized in situ by x-ray photoelectron spectroscopy (XPS). The experimentally observed changes in valence state were correlated by simulating the nanoceria thin film by molecular dynamics and studying the displacement cascades produced by cerium and oxygen as primary knock on atom. The structural evolution of nanoceria thin films due to radiation exposure will be discussed in detail with implications for the use of cerium oxide as a radiation tolerant material.

Particle accelerators are a key tool for scientific research ranging from fundamental studies of matter to analytical studies at light sources. Cost-for-performance is critical, both in terms of initial capital outlay and ongoing operating expense, especially for electricity. It depends on the niobium superconducting radiofrequency (SRF) accelerator cavities at the heart of most of these machines.

Presently Nb SRF cavities operate near 1.9 K, well (and expensively) below the 4.2 K atmospheric boiling point of liquid He. Transforming the 40 nm thick active interior surface layer from Nb to delta NbN (Tc = 17 K instead of 9.2 K) appears to be a promising approach. Traditional furnace nitriding appears to have not been successful for this. Further, exposing a complete SRF cavity to the time-temperature history required for nitriding risks mechanical distortion.

Gas laser nitriding instead has been applied successfully to other metals [P.Schaaf, Prog. Mat. Sci. 47 (2002) 1]. The beam dimensions and thermal diffusion length permit modeling in one dimension to predict the time course of the surface temperature for a range of per-pulse energy densities. As with the earlier work, we chose conditions just sufficient for boiling as a reference point.

We used a Spectra Physics HIPPO nanosecond laser (l = 1064 nm, E_max= 0.392 mJ, beam spot@ 34 microns, PRF =15 – 30 kHz) to obtain an incident fluence of 1.73 - 2.15 J/cm² for each laser pulse at the target. The target was a 50 mm diameter SRF-grade Nb disk maintained in a nitrogen atmosphere at a pressure of 550 – 625 torr and rotated at a constant speed of 9 rpm.

The materials were examined by scanning electron microscopy (SEM), electron probe microanalysis (EPMA) and x-ray diffraction (XRD). The SEM images show a sharp transition with fluence from a smooth, undulating topography to significant roughening, interpreted here as the onset of ablation. EPMA measurements of N/Nb atom ratio as a function of depth found a constant value to depths greater than the SRF active layer thickness. Certain irradiation conditions resulted in values consistent with formation of delta NbN. Under certain irradiation conditions, XRD data were consistent only with delta NbN on top of Nb metal.

Funding: authored by Jefferson Science Associates LLC under US DOE Contract De-AC05-06OR23177. We are indebted to Prof. P. Schaaf (Goettingen) for the simulation code and helpful discussions.

A combination of ion channeling, 4.265 ± 0.035 MeV 12C(alpha, alpha;)12C , 3.05 ± 0.005 16O(alpha, alpha)16O MeV Nuclear Resonance and 2.8 MeV Hydrogen Recoil Detection are used for high resolution compositional depth profiling of the first 10 nm of the surface. Tapping Mode Atomic Force Microscopy (TMAFM) provides statistical analysis of the topography of these Si-based surfaces at a length scale ranging from a few nm to several µm. Extended atomic terraces with low edges and defects density on OH(1x1)Si(100) and crystalline silicates such as beta-cristobalite and alpha-quartz can render an insulating surface hydrophobic, while small scale roughness and surface defects makes it more hydrophilic. The water affinity and surface energy is measured using the Sessile Drop method and the Young-Dupré analysis as a function of topographical, compositional and micro-structural surface analysis using TMAFM, IBA combined with ion channeling and X-ray diffraction respectively. This correlation explains the behavior of water condensation at the liquid/air interface of intraocular implants during vitro-retinal surgery after cataract extraction and intraocular lens (IOL) implantation.

Polymer adsorption on surfaces alters their hydroaffinity, can control condensation on silicone IOL’s [1] and enable for hermetic bonding in silica-based sensors in medical electronic implants [2]

[1] US Patent pending: "Molecular films for controlling hydrophobic, hydrophilic, optical, condensation and geometric properties of silicone implants surfaces, including intraoccular lenses used in cataract surgeries." Inventor(s): N. Herbots, J. D. Bradley, M. Hart, D. A. Sell, S. Whaley, Q. Bradley (November 09, 2009)

[2] US Patent Pending: “Methods for Wafer Bonding, and for Nucleating Bonding Nanophases”. Inventor(s): N. Herbots, J. D. Bradley, M. Hart, D. A. Sell, S. Whaley, R. J. Culbertson (April 30, 2010)

The work function of a number of metal oxide thin films was measured using low intensity x-ray and ultraviolet photoemission spectroscopy (LIXPS, UPS).

Our experiments revealed that UPS based work function measurements of a variety of metal oxide surfaces previously exposed to the ambient or in contact with water vapor caused an immediate work function reduction of the order of 0.5 eV. This artifact was demonstrated using LIXPS work function measurements, which expose the sample surface to only a small fraction of the radiation flux encountered during UPS measurements. This enabled the measurement of the work function prior to the UPS characterization, revealing the UPS-related work function reduction. LIXPS measurement performed after the UPS measurements showed that sample charging effects can be ruled out as causes of the work function reduction, and that the work function change is permanent. These results suggest that a photochemical hydroxylation of the surface through photodissociation of water molecules occurs during the UPS measurement, which results in a surface dipole reducing the work function. Additional experiments investigated the influence of standard XPS measurements as well as LIXPS measurements on the work function. These experiments demonstrated that XPS measurements also cause a work function reduction similar to UPS measurements, albeit during a longer exposure time frame. In contrast, LIXPS measurements do not result in a significant work function reduction, i.e. can be used as a reliable work function characterization tool for oxide surfaces.