AVS2011 Session AS-TuP: Applied Surface Science Poster Session

Disposable elastic gloves are ubiquitous in scientific laboratories and are also widely used in many industries during handling of critical surfaces. Disposable gloves are typically made from nitrile, latex, neoprene, or other polymers and offer their users protection from various aqueous acids and bases, biological fluids, organic solvents, and other potentially harmful chemicals. A second major application of disposable gloves is to protect manufactured products and analytical samples from contamination caused by the transfer of skin cells, oils, salts, or other residues resulting from contact with bare hands. However, disposable gloves can also be a potential source of contamination. In addition to the primary polymer structure, many types of common laboratory gloves also contain a variety of inorganic materials in the glove formulation. Mold-release agents that allow the gloves to be easily stripped from the glove formers during fabrication may be present on glove surfaces. Furthermore, many types of disposable gloves have polymeric surface coatings that provide improved donning properties. Contamination resulting from surface residues on gloves can adversely affect materials used in industries where surface cleanliness is essential for optimum product performance and can also interfere with the analysis of samples depending on the specificity and sensitivity of the analytical technique. X-ray photoelectron spectroscopy (XPS) is a qualitative and quantitative surface sensitive technique that can be used to evaluate the surface composition of disposable gloves and to determine if contamination transfer occurs from gloves in a specific process. In this study, XPS was used to characterize the surface compositions of a variety of common laboratory gloves. The transfer of surface components from gloves to other material surfaces and changes in the surface composition of gloves following exposure to several common laboratory solvents were also investigated.

Focused Ion Beam (FIB) techniques are widely used in order to generate cross sections of samples which can then be analyzed using a variety of analytical instrumentation in order to obtain microstructural information. In order to rapidly generate FIB cross sections, small areas (typically 5 to 10 microns) are milled. Time of Flight Secondary Ion Mass Spectrometry (ToF-SIMS) is a powerful surface analytical technique that is able to measure low concentrations of sub surface species. ToF-SIMS depth profile measurements are collected by eroding a large area (50 to 1,000 microns in size) of a sample using an ion beam for a certain amount of time, pausing the erosion, analyzing a smaller (central) region of the eroded area, and repeating the cycle until the desire depth has been reached. Ion images are often recorded at each depth and a full mass spectrum is saved at every volume element allowing for 3 D analysis. Often times, both ToF-SIMS depth profile analysis and high resolution microstructural SEM characterization are required on the same sample. Ideally, these complimentary analyses would be performed in the same region of a sample. In this presentation, we will demonstrate that FIB techniques can be used to clean up the large area craters produced by ToF-SIMS depth profile analysis thereby enabling microstructural characterization over larger areas of samples.

Advantages of this approach are: larger FIB cross sections can be prepared since a significant amount of material was removed during the erosion cycles in the ToF-SIMS analysis, combining the chemical information provided from the ToF-SIMS analysis with the large area microstructural characterization provided by FIB-SEM analysis, and having both the ToF-SIMS and the large area cross sectional analysis performed in the same region of the sample. The ToF-SIMS depth profile craters have a slope of about 45 deg (the ion column is positioned at an angle of 45 deg from the sample normal), and hence an advantage of performing the FIB clean up at this angle is the enhancement in the layer thickness that will be available for microstructural analysis.

Stainless steel coupons approximately 0.5” in diameter and 0.125” thick were passivated with five different surface treatments and an untreated coupon was left as a control. These surface treatments are being explored for use in tritium storage containers. These coupons were made to allow surface analysis of the surface treatments using well-known surface analysis techniques. Depth profiles using Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) were performed on these coupons to characterize the surface and near surface regions. Scanning electron microscope (SEM) images were collected as well. All of the surface treatments studied here appear to change the surface morphology dramatically, as evidenced by lack of tool marks on the treated samples. In terms of the passivation treatment, Vendors A-D appeared to have oxide layers that were very similar in thickness to each other (0.7-0.9 nm thick) as well as the untreated samples (the untreated sample oxide layers appeared to be somewhat larger). Vendor E’s silicon coating appears to be on the order of 200 nm thick.

Discharged firearm cartridges produce unique microscopic particles referred to as gunshot residue (GSR). GSR is primarily composed of the products of combustion of the cartridge primer materials and typically contains varying amounts of Pb, Sb, and Ba, plus other elements. GSR particles can range in size from tens of nanometers to hundreds of micrometers. Because of the high temperatures (~1,500-3,600 °C) and pressures (~14,000-65,000 psi) that result within 1 millisecond of discharging a firearm cartridge, highly complicated chemical interactions are likely to occur that will affect the chemical composition of the GSR particles. Computer-controlled scanning electron microscopy (CCSEM) is the method preferred by the forensic community for the automated analysis of GSR. With CCSEM, large populations of potential GSR particles in the size range of ~1-10 µm are rapidly screened for the characteristic presence of combinations of Pb, Sb, and Ba using energy dispersive X-ray spectroscopy (EDS). Determining the presence of these three elements fused together in a single particle having the correct morphology is all that is normally required for the positive identification of GSR. CCSEM, however, does generally not provide information regarding the population of particles much less than 1 µm. In addition, little is presently known regarding the surface chemistry of GSR. Because of its nanometer-scale sampling depth and the ability to provide detailed chemical state information, X-ray photoelectron spectroscopy (XPS) can provide important information regarding the surface chemistry of GSR. This work examines the fine fraction of GSR particles with high resolution electron microscopy methods and complements the microscopy data with surface chemistry information obtained from XPS.

Research sponsored by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of FreedomCAR and Vehicle Technologies, as part of the High Temperature Materials Laboratory User Program, Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy under contract number DE-AC05-00OR22725.

Ceria-zirconia mixed oxide (CZO) has been incorporated into automotive catalysts as a support material for precious metals (Pt, Rh, Pd) due to its highly desirable redox properties. However, at high operating temperatures, certain interactions between CZO and precious metal particles may adversely affect the performance of the catalytic converter. Thus, it is important to investigate the CZO/precious-metal interactions under redox conditions. Here, the preparation of Pd/CZO thin films as model planar catalyst systems will be described together with a systematic study of an important adverse interaction, the encapsulation of Pd particles by CZO, using a combination of x-ray diffraction (XRD) and high resolution transmission electron microscopy (HRTEM).

Ceria-zirconia (Ce_0.7Zr_0.3O₂) thin films with thickness ranging from 10-200 nm were grown on (111) yttria-stabilized zirconia (YSZ) by oxygen plasma assisted molecular beam epitaxy (OPA-MBE). The epitaxial layer growth of CZO on YSZ(111) was confirmed by in-situ reflection high energy electron diffraction (RHEED). Atomic force microscopy (AFM) images revealed the smooth surface of the films with low roughness values (3-8 Å). Rutherford backscattering spectrometry (RBS) data along channeling and random geometries showed the minimum yield (χ_min) of 13% for Ce with no inter-diffusion of metal atoms at the film/substrate interface. The single crystal nature of the film with CZO(111) orientation was confirmed by XRD data. Following the growth and characterization of thin films, ~1 monolayer of Pd was deposited on CZO(111) by thermal evaporation in a UHV chamber. After air calcination at 600 °C, the Pd/CZO films were reduced in 1% H₂/N₂ at 200°C, followed by annealing in N₂ at 700 °C. Due to the additional reduction-induced compressive stress, the 10 nm and 40 nm films broke up and formed a rough surface with 10-20 nm CZO mesas, as observed by HRTEM. However, films with a thickness above ~50 nm were stable, and Pd particles that formed on the surface became partially encapsulated by migration of CZO. In the films with a thickness above ~50 nm, the native compressive stress induced by the lattice mismatch between the film and substrate was relaxed by misfit dislocations at the interface, as revealed by high resolution scanning TEM images, to varying degree, depending on film thickness, but the extent of encapsulation appeared to be about the same for all film thicknesses. These results suggest that the driving force for the partial encapsulation resides in the nature of CZO, and is not simply a consequence of the lattice-mismatch-induced compressive stress in the films.

Metal oxide semiconductor field effect transistors (MOSFETs) are the dominant logic device in modern electronics. Due to the challenges of scaling Si-MOSFETs alternative materials are being explored to improve device performance. III-V semiconductors are of interest for use in MOSFETs due to their high mobilities, but in order to make these devices competitive the semiconductor-oxide interface needs to have a low density of interfacial traps (D_it­) in order to minimize subthreshold swings and also be atomically flat to allow high mobility at high field strengths. Current Si-MOSFET fabrication uses a gate last process which is an attractive method because it minimizes the gate oxides exposure to harsh processing conditions. In order for gate last processing to provide a nearly defect free semiconductor-oxide interface for surface channel devices, the channel surface must be in pristine condition before the oxide layer is deposited. In addition the oxide deposition process cannot introduce any defects. Recent advances show that a thin indium phosphide layer deposited on top of the InGaAs channel may provide superior electronic performance because the InGaAs/InP interface is defect-free and flat while the defects at the oxide/InP interface are less detrimental to device performance than the defects at the oxide/InGaAs interface.

In order to facilitate gate last processing of the InP/InGaAs channel stack, a four step process was investigated to clean and nucleate atomic layer deposition (ALD) of an InP/InGaAs surface. Samples were grown with an undoped 2 nm InP layer on ~1 µm InGaAs layer doped with 4 X 10¹⁸ cm^-3 of Si on an InP wafer. Using in-situ atomic imaging (scanning tunneling microscopy), electronic measurements (scanning tunneling spectroscopy), and XPS a four step in-situ process was developed for converting an air-exposed InGaAs(100)-4x2 surface into a flat, electronically passivated, mono-layer nucleation template for ALD of gate oxides. The optimized process includes a small dose of atomic hydrogen, annealing to reduce surface roughness, TMA dosing to functionalize the surface, and a final annealing to induce a highly ordered ALD nucleation layer. Following atomic hydrogen cleaning at 380˚C and annealing at 470˚C, the surface was exposed to ~1 x 10^-2 Torr of TMA at room temperature and annealed to 270˚C. The TMA induces a surface reconstruction consistent with a bulk like bonding configuration between the Al atoms and the surface P atoms. The TMA passivation layer has horizontal rows of DMA. The cleaning and nucleation process does not disrupt the InP surface thereby providing a path for further scaling of the InP layer and the gate oxide.

Highly selective and low damage etching of the GaAs cap layer on AlGaAs is essential in fabricating devices such as heterojuction superlattices, field effect transistors, injection lasers, and solar cells, etc. The GaAs on AlGaAs was etched using a low energy Cl₂/O₂ neutral beam and the schottky device characteristics fabricated on the exposed AlGaAs were compared with those fabricated after the etching using wet etching and a Cl₂/O₂ ion beam. The wet etching was conducted by a mixture of H₂O₂ and H₃PO₄. For the Cl₂/O₂ neutral beam etching, a neutral beam system composed of three-grid inductively coupled plasma (ICP)-type ion gun and a reflector installed just in front of the ion gun was used. Using a low energy Cl₂/O₂ ion beam or a Cl₂/O₂ neutral beam, highly selective etching of the GaAs cap layer to AlGaAs similar to wet etching could be achieved through the formation of Al₂O₃ on the exposed AlGaAs during the etching. When the electrical characteristics of the schottky devices were compared, the devices fabricated after the etching using the neutral beam showed the best electrical characteristics such as electrical stability, low leakage current, higher barrier height, etc. by showing low damage to the exposed AlGaAs surface.

[1] “Self consistent ARXPS analysis for multilayer conformal films with abrupt interfaces.” A. Herrera-Gomez. Internal Report 2007 (http://www.qro.cinvestav.mx/~aanalyzer/arxpsAnalysisSharpIntefaces.pdf).

[2] Cumpson PJ. Surf. Interface Anal. 2000; 29: 403.

As the demand for high speed electronics remains ever increasing, the dimensions of MOSFET technology are continuously scaled down to the smallest possible levels. Traditional gate dielectrics such as SiO₂ are unable to effectively stop electron tunneling from degrading device performance at thicknesses below 1-1.2 nm. We propose using high dielectric materials instead, which can effectively limit leakage current and have a band gap close to SiO₂. This study will focus on the dielectrics Al₂O₃, HfO₂, and Al_xHf_yO_z deposited using ALD. Several ratios of aluminum and hafnium in the Al_xHf_yO_z compound will be investigated to find a level that maximizes both electrical and physical properties of the film. The electrical properties of each dielectric will be characterized by XPS, AFM, spectroscopic ellipsometry, and by taking IV and CV measurements of fabricated transistors and capacitors. We will also investigate the effects of different annealing and deposition temperatures on the interface by TEM.