Here, we present an in-depth study of metal–insulator interfaces within granular metal (GM) films and correlate their interfacial interactions with structural and electrical transport properties. Nominally 100 nm thick GM films of Co and Mo dispersed within yttria-stabilized zirconia (YSZ), with volumetric metal fractions (φ) from 0.2–0.8, were grown by radio frequency co-sputtering from individual metal and YSZ targets. Scanning transmission electron microscopy and DC transport measurements find that the resulting metal islands are well-defined with 1.7–2.6 nm average diameters and percolation thresholds between φ = 0.4–0.5. The room temperature conductivities for the φ = 0.2 samples are several orders of magnitude larger than previously-reported for GMs. X-ray photoemission spectroscopy indicates both oxygen vacancy formation within the YSZ and band-bending at metal–insulator interfaces. The higher-than-predicted conductivity is largely attributed to these interface interactions. In agreement with recent theory, interactions that reduce the change in conductivity across the metal–insulator interface are seen to prevent sharp conductivity drops when the metal concentration decreases below the percolation threshold. These interface interactions help interpret the broad range of conductivities reported throughout the literature and can be used to tune the conductivities of future GMs.
AlGaN-channel high electron mobility transistors (HEMTs) were operated as visible- and solar-blind photodetectors by using GaN nanodots as an optically active floating gate. The effect of the floating gate was large enough to switch an HEMT from the off-state in the dark to an on-state under illumination. This opto-electronic response achieved responsivity > 108 A/W at room temperature while allowing HEMTs to be electrically biased in the offstate for low dark current and low DC power dissipation. The influence of GaN nanodot distance from the HEMT channel on the dynamic range of the photodetector was investigated, along with the responsivity and temporal response of the floating gate HEMT as a function of optical intensity. The absorption threshold was shown to be controlled by the AlN mole fraction of the HEMT channel layer, thus enabling the same device design to be tuned for either visible- or solar-blind detection.
The room temperature electronic transport properties of 1 μm thick Bi0.4Sb1.6Te3 (BST) films correlate with overall microstructural quality. Films with homogeneous composition are deposited onto fused silica substrates, capped with SiN to prevent both oxidation and Te loss, and postannealed to temperatures ranging from 200 to 450 °C. BST grain sizes and (00l) orientations improve dramatically with annealing to 375 °C, with smaller increases to 450 °C. Tiny few-nanometer-sized voids in the as-deposited film grain boundaries coalesce into larger void sizes up to 300 nm with annealing to 350 °C; the smallest voids continue coalescing with annealing to 450 °C. These voids are decorated with few-nanometer-sized Sb clusters that increase in number with increasing annealing temperatures, reducing the Sb content of the remaining BST film matrix. Resistivity decreases linearly with increasing temperature over the entire range studied, consistent with improving crystalline quality. The Seebeck coefficient also improves with crystalline quality to 350 °C, above which void coalescence and reduced Sb content from the BST matrix correlate with a decrease in the Seebeck coefficient. Nevertheless, a plateau exists for an optimal power factor between 350 and 450 °C, implying thermal stability to higher temperatures than previously reported.
Neutron sensing is critical in civilian and military applications. Conventional neutron sensors are limited by size, weight, cost, portability and helium supply. Here the microfabrication of gadolinium (Gd) conversion material-based heterojunction diodes for detecting thermal neutrons using electrical signals produced by internal conversion electrons (ICEs) is described. Films with negligible stress were produced at the tensile-compressive crossover point, enabling Gd coatings of any desired thickness by controlling the radiofrequency sputtering power and using the zero-point near p(Ar) of 50 mTorr at 100 W. Post-deposition Gd oxidation-induced spallation was eliminated by growing a residual stress-free 50 nm neodymium-doped aluminum cap layer atop Gd. The resultant coatings were stable for at least 6 years, demonstrating excellent stability and product shelf-life. Depositing Gd directly on the diode surface eliminated the air gap, leading to a 200-fold increase in electron capture efficiency and facilitating monolithic microfabrication. The conversion electron spectrum was dominated by ICEs with energies of 72, 132 and 174 keV. Results are reported for neutron reflection and moderation by polyethylene for enhanced sensitivity, and γ- and X-ray elimination for improved specificity. The optimal Gd thickness was 10.4 μm for a 300 μm-thick partially depleted diode of 300 mm 2 active surface area. Fast detection (within 10 min) at a neutron source-to-diode distance of 11.7 cm was achieved with this configuration. All ICE energies along with γ-ray and K α,β X-rays were modeled to emphasize correlations between experiment and theory. Semi-conductor thermal neutron detectors offer advantages for field-sensing of radioactive neutron sources.
The room temperature electronic transport properties of 100 nm-thick thermoelectric Bi0.8Sb0.2 films, sputter-deposited onto quartz substrates and post-annealed in an ex-situ furnace, systematically correlate with the overall microstructural quality, improving with increasing annealing temperature until close to the melting point for the alloy composition. Furthermore, the optimized films have high crystalline quality with ~99% of the grains oriented with the trigonal axis perpendicular to the substrate surface. Film resistivities and Seebeck coefficients are accurately measured by preventing deleterious surface oxide formation via a SiN capping layer and using Nd-doped Al for contacts. Our resulting values are similar to single crystals and significantly better than previous reports from films and polycrystalline bulk alloys.
Neutron sensing is critical in civilian, military, industrial, biological, medical, basic research, and environmental applications. Conventional neutron sensors are limited by size, weight, cost, portability, and helium supply. Here the microfabrication of Gd conversion material-based heterojunction diodes is described for detecting thermal neutrons using electrical signals produced by internal conversion electrons (ICE). Films with negligible stress were produced at the tensile-compressive crossover point, enabling Gd coatings of any desired thickness by controlling the radiofrequency sputtering power and using the zero-point near p(Ar) of 50 mTorr at 100 W. Post-deposition Gd oxidation-induced spallation was eliminated by growing a residual stress-free 50 nm neodymium-doped aluminum cap layer atop Gd. Resultant coatings were stable for at least six years demonstrating excellent product shelf life. Depositing Gd on the diode surface eliminated air gap, leading to improved efficiency and facilitating monolithic microfabrication. The conversion electron spectrum was dominated by ICE with energies of 72, 132, and 174 keV. Results are reported on neutron reflection and moderation by polyethylene for enhanced sensitivity and g- and X-ray elimination for improved specificity. Optimal Gd thickness was 10.4 um with 300 um thick partially depleted diode of 300 mm2 active surface area. Fast detection within 10 minutes at a neutron source-to-diode distance of 11.7 cm was achieved using this configuration. All ICE energies along with g-ray and Ka X-ray were modeled to emphasize correlations between experiment and theory and to calculate efficiencies. Semiconductor thermal neutron detectors offer advantages for field-sensing of radioactive neutron sources. ACKNOWLEDGEMENTS Sandia National Laboratories is a multi-program laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Company, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. We thank Edward Cole, David Wheeler, Robert Koudelka, and Lyle Brunke for productive interactions and materials support.
We study nanoporous-carbon (NPC) grown via pulsed laser deposition (PLD) as an electrically conductive anode host material for Mg2+ intercalation. NPC has high surface area, and an open, accessible pore structure tunable via mass density that can improve diffusion. We fabricate 2032 coin cells using NPC coated stainless-steel disk anodes, metallic Mg cathodes, and a Grignardbased electrolyte. NPC mass density is controlled during growth, ranging from 0.06-1.3 g/cm3. The specific surface area of NPC increases linearly from 1,000 to 1,700 m2/g as mass density decreases from 1.3 to 0.26 g/cm3, however, the surface area falls off dramatically at lowermass densities, implying a lack of mechanical integrity in such nanostructures. These structural characterizations correlate directly with coin cell electrochemical measurements. In particular, cyclic voltammetry (CV) scans for NPC with density ∼0.5 g/cm3 and BET surface area ∼1500 m2/g infer the possibility of reversible Mg-ion intercalation. Higher density NPC yields capacitive behavior, most likely resulting from the smaller interplanar spacings between graphene sheet fragments and tighter domain boundaries; lower density NPC results in asymmetrical CV scans, consistent with the likely structural degradation resulting from mass transport through soft, low-density carbon materials.
Compositional-homogeneity and crystalline-orientation are necessary attributes to achieve high thermoelectric performance in Bi1-xSbx thin films. Following deposition in vacuum, and upon air exposure, we find that 50%-95% of the Sb in 100-nm thick films segregates to form a nanocrystalline Sb2O3 surface layer, leaving the film bulk as Bi-metal. However, we demonstrate that a thin SiN capping layer deposited prior to air exposure prevents Sb-segregation, preserving a uniform film composition. Furthermore, the capping layer enables annealing in forming gas to improve crystalline orientations along the preferred trigonal axis, beneficially reducing electrical resistivity.
Vertically aligned, untangled planarized arrays of multiwall carbon nanotubes (MWNTs) with Ohmic back contacts were grown in nanopore templates on arbitrary substrates. The templates were prepared by sputter depositing Nd-doped Al films onto W-coated substrates, followed by anodization to form an aluminum oxide nanopore array. The W underlayer helps eliminate the aluminum oxide barrier that typically occurs at the nanopore bottoms by instead forming a thin WO3 layer. The WO3 can be selectively etched to enable electrodeposition of Co catalysts with control over the Co site density. This led to control of the site density of MWNTs grown by thermal chemical vapor deposition, with W also serving as a back electrical contact. Ohmic contact to MWNTs was confirmed, even following ultrasonic cutting of the entire array to a uniform height.
Using galvanostatic pulse deposition, we studied the factors influencing the quality of electroformed Bi1-xSbxnanowires with respect to composition, crystallinity, and preferred orientation for high thermoelectric performance. Two nonaqueous baths with different Sb salts were investigated. The Sb salts used played a major role in both crystalline quality and preferred orientations. Nanowire arrays electroformed using an SbI3-based chemistry were polycrystalline with no preferred orientation, whereas arrays electroformed from an SbCl3-based chemistry were strongly crystallographically textured with the desired trigonal orientation for optimal thermoelectric performance. From the SbCl3 bath, the electroformed nanowire arrays were optimized to have nanocompositional uniformity, with a nearly constant composition along the nanowire length. Nanowires harvested from the center of the array had an average composition of Bi0.75Sb0.25. However, the nanowire compositions were slightly enriched in Sb in a small region near the edges of the array, with the composition approaching Bi0.700.30.
We present the results of a three year LDRD project that focused on understanding the impact of defects on the electrical, optical and thermal properties of GaN-based nanowires (NWs). We describe the development and application of a host of experimental techniques to quantify and understand the physics of defects and thermal transport in GaN NWs. We also present the development of analytical models and computational studies of thermal conductivity in GaN NWs. Finally, we present an atomistic model for GaN NW electrical breakdown supported with experimental evidence. GaN-based nanowires are attractive for applications requiring compact, high-current density devices such as ultraviolet laser arrays. Understanding GaN nanowire failure at high-current density is crucial to developing nanowire (NW) devices. Nanowire device failure is likely more complex than thin film due to the prominence of surface effects and enhanced interaction among point defects. Understanding the impact of surfaces and point defects on nanowire thermal and electrical transport is the first step toward rational control and mitigation of device failure mechanisms. However, investigating defects in GaN NWs is extremely challenging because conventional defect spectroscopy techniques are unsuitable for wide-bandgap nanostructures. To understand NW breakdown, the influence of pre-existing and emergent defects during high current stress on NW properties will be investigated. Acute sensitivity of NW thermal conductivity to point-defect density is expected due to the lack of threading dislocation (TD) gettering sites, and enhanced phonon-surface scattering further inhibits thermal transport. Excess defect creation during Joule heating could further degrade thermal conductivity, producing a viscous cycle culminating in catastrophic breakdown. To investigate these issues, a unique combination of electron microscopy, scanning luminescence and photoconductivity implemented at the nanoscale will be used in concert with sophisticated molecular-dynamics calculations of surface and defect-mediated NW thermal transport. This proposal seeks to elucidate long standing material science questions for GaN while addressing issues critical to realizing reliable GaN NW devices.
Nanoporous carbon (NPC) is a purely graphitic material with highly controlled densities ranging from less than 0.1 to 2.0 g/cm3, grown via pulsed-laser deposition. Decreasing the density of NPC increases the interplanar spacing between graphene-sheet fragments. This ability to tune the interplanar spacing makes NPC an ideal model system to study the behavior of carbon electrodes in electrochemical capacitors and batteries. We examine the capacitance of NPC films in alkaline and acidic electrolytes, and measure specific capacitances as high as 242 F/g.
Nanoporous carbon (NPC) is a purely graphitic material with highly controlled densities ranging from less than 0.1 to 2.0 g/cm3, grown via pulsed-laser deposition. Decreasing the density of NPC increases the interplanar spacing between graphene-sheet fragments. This ability to tune the interplanar spacing makes NPC an ideal model system to study the behavior of carbon electrodes in electrochemical capacitors and batteries. We examine the capacitance of NPC films in alkaline and acidic electrolytes, and measure specific capacitances as high as 242 F/g.
Active cooling of electronic systems for space-based and terrestrial National Security missions has demanded use of Stirling, reverse-Brayton, closed Joule-Thompson, pulse tube and more elaborate refrigeration cycles. Such cryocoolers are large systems that are expensive, demand large powers, often contain moving parts and are difficult to integrate with electronic systems. On-chip, solid-state, active cooling would greatly enhance the capabilities of future systems by reducing the size, cost and inefficiencies compared to existing solutions. We proposed to develop the technology for a thermoelectric cooler capable of reaching 77K by replacing bulk thermoelectric materials with arrays of Bi{sub 1-x}Sb{sub x} nanowires. Furthermore, the Sandia-developed technique we will use to produce the oriented nanowires occurs at room temperature and can be applied directly to a silicon substrate. Key obstacles include (1) optimizing the Bi{sub 1-x}Sb{sub x} alloy composition for thermoelectric properties; (2) increasing wire aspect ratios to 3000:1; and (3) increasing the array density to {ge} 10{sup 9} wires/cm{sup 2}. The primary objective of this LDRD was to fabricate and test the thermoelectric properties of arrays of Bi{sub 1-x}Sb{sub x} nanowires. With this proof-of-concept data under our belts we are positioned to engage National Security systems customers to invest in the integration of on-chip thermoelectric coolers for future missions.
Carbon nanotubes have shown promise for applications in many diverse areas of technology. In this report we describe our efforts to develop high-current cathodes from a variety of nanotubes deposited under a variety of conditions. Our goal was to develop a one-inch-diameter cathode capable of emitting 10 amperes of electron current for one second with an applied potential of 50 kV. This combination of current and pulse duration significantly exceeds previously reported nanotube-cathode performance. This project was planned for two years duration. In the first year, we tested the electron-emission characteristics of nanotube arrays fabricated under a variety of conditions. In the second year, we planned to select the best processing conditions, to fabricate larger cathode samples, and to test them on a high-power relativistic electron beam generator. In the first year, much effort was made to control nanotube arrays in terms of nanotube diameter and average spacing apart. When the project began, we believed that nanotubes approximately 10 nm in diameter would yield sufficient electron emission properties, based on the work of others in the field. Therefore, much of our focus was placed on measured field emission from such nanotubes grown on a variety of metallized surfaces and with varying average spacing between individual nanotubes. We easily reproduced the field emission properties typically measured by others from multi-wall carbon nanotube arrays. Interestingly, we did this without having the helpful vertical alignment to enhance emission; our nanotubes were randomly oriented. The good emission was most likely possible due to the improved crystallinity, and therefore, electrical conductivity, of our nanotubes compared to those in the literature. However, toward the end of the project, we learned that while these 10-nm-diameter CNTs had superior crystalline structure to the work of others studying field emission from multi-wall CNT arrays, these nanotubes still had a thin coating of glassy carbon surrounding them in a sheath-like manner. This glassy carbon, or nano-crystalline graphite, is likely to be a poor conductor due to phonon scattering, and should actually be deleterious for extracting electrons with electric fields. While we did not achieve the field emission reported for single-wall carbon nanotubes that spurred the idea for this project, at the year's very end, we had a breakthrough in materials growth and learned to control the growth of very-small diameter nanotubes ranging from 1.4 to 7 nm. The 1.4-nm nanotubes are single-walled and grow at only 530 C. This is the lowest temperature known to result in single-wall carbon nanotubes, and may be very important for many applications that where certain substrates could not be used due to the high temperatures commonly used for CNT growth. Critically important for field emission, these small diameter nanotubes, consisting of only a few concentric graphene cylindrical walls, do not show the presence of a poorly-conductive sheath material. Therefore, these nanotubes will almost definitely have superior field emission properties to those we already measured, and it is possible that they could provide the necessary field emission to make this project successful. Controlled spacing and lengths of these single-wall nanotubes have yet to be explored, along with correlating their structures to their improved field emission. Unfortunately, we did not discover the methods to grow these highly-crystalline and small diameter CNTs until late in the year. Since we did not achieve the necessary emission properties by mid-year, the project was ''prematurely'' terminated prior to the start of the second year. However, it should be noted that with the late developments, this work has not hit the proverbial ''brick wall''. Clearly the potential still exists to reproduce and even exceed the high emission results reported for randomly-oriented and curly single-wall carbon nanotubes, both in terms of total field emitting currents and perhaps more importantly, in reproducibility.
We have built and tested a miniaturized, thermoelectric power source that can provide in excess of 450 {micro}W of power in a system size of 4.3cc, for a power density of 107 {micro}W/cc, which is denser than any system of this size previously reported. The system operates on 150mW of thermal input, which for this system was simulated with a resistive heater, but in application would be provided by a 0.4g source of {sup 238}Pu located at the center of the device. Output power from this device, while optimized for efficiency, was not optimized for form of the power output, and so the maximum power was delivered at only 41mV. An upconverter to 2.7V was developed concurrently with the power source to bring the voltage up to a usable level for microelectronics.
Multiwall carbon nanotubes are grown via thermal chemical vapor deposition between temperatures of 630 and 830 C using acetylene in nitrogen as the carbon source. This process is modeled using classical thermodynamics to explain the total carbon deposition as a function of time and temperature. An activation energy of 1.60 eV is inferred for nanotube growth after considering the carbon solubility term. Scanning electron microscopy shows growth with diameters increasing linearly with time. Transmission electron microscopy and Raman spectroscopy show multiwall nanotubes surrounded by a glassy-carbon sheath, which grows with increasing wall thickness as growth temperatures and times rise.
A variety of solution deposition routes have been reported for processing complex perovskite-based materials such as ferroelectric oxides and conductive electrode oxides, due to ease of incorporating multiple elements, control of chemical stoichiometry, and feasibility for large area deposition. Here, we report an extension of these methods toward long length, epitaxial film solution deposition routes to enable biaxially oriented YBa{sub 2}Cu{sub 3}O{sub 7-{delta}} (YBCO)-coated conductors for superconducting transmission wires. Recent results are presented detailing an all-solution deposition approach to YBCO-coated conductors with critical current densities J{sub c} (77 K) > 1 MA/cm{sup 2} on rolling-assisted, biaxially textured, (200)-oriented Ni-W alloy tapes. Solution-deposition methods such as this approach and those of other research groups appear to have promise to compete with vapor phase methods for superconductor electrical properties, with potential advantages for large area deposition and low cost/kA {center_dot} m of wire.
Imagine free-standing flexible membranes with highly-aligned arrays of carbon nanotubes (CNTs) running through their thickness. Perhaps with both ends of the CNTs open for highly controlled nanofiltration? Or CNTs at heights uniformly above a polymer membrane for a flexible array of nanoelectrodes or field-emitters? How about CNT films with incredible amounts of accessible surface area for analyte adsorption? These self-assembled crystalline nanotubes consist of multiple layers of graphene sheets rolled into concentric cylinders. Tube diameters (3-300 nm), inner-bore diameters (2-15 nm), and lengths (nanometers - microns) are controlled to tailor physical, mechanical, and chemical properties. We proposed to explore growth and characterize nanotube arrays to help determine their exciting functionality for Sandia applications. Thermal chemical vapor deposition growth in a furnace nucleates from a metal catalyst. Ordered arrays grow using templates from self-assembled hexagonal arrays of nanopores in anodized-aluminum oxide. Polymeric-binders can mechanically hold the CNTs in place for polishing, lift-off, and membrane formation. The stiffness, electrical and thermal conductivities of CNTs make them ideally suited for a wide-variety of possible applications. Large-area, highly-accessible gas-adsorbing carbon surfaces, superb cold-cathode field-emission, and unique nanoscale geometries can lead to advanced microsensors using analyte adsorption, arrays of functionalized nanoelectrodes for enhanced electrochemical detection of biological/explosive compounds, or mass-ionizers for gas-phase detection. Materials studies involving membrane formation may lead to exciting breakthroughs in nanofiltration/nanochromatography for the separation of chemical and biological agents. With controlled nanofilter sizes, ultrafiltration will be viable to separate and preconcentrate viruses and many strains of bacteria for 'down-stream' analysis.
Chemical microsensors rely on partitioning of airborne chemicals into films to collect and measure trace quantities of hazardous vapors. Polymer sensor coatings used today are typically slow to respond and difficult to apply reproducibly. The objective of this project was to produce a durable sensor coating material based on graphitic nanoporous-carbon (NPC), a new material first studied at Sandia, for collection and detection of volatile organic compounds (VOC), toxic industrial chemicals (TIC), chemical warfare agents (CWA) and nuclear processing precursors (NPP). Preliminary studies using NPC films on exploratory surface-acoustic-wave (SAW) devices and as a {micro}ChemLab membrane preconcentrator suggested that NPC may outperform existing, irreproducible coatings for SAW sensor and {micro}ChemLab preconcentrator applications. Success of this project will provide a strategic advantage to the development of a robust, manufacturable, highly-sensitive chemical microsensor for public health, industrial, and national security needs. We use pulsed-laser deposition to grow NPC films at room-temperature with negligible residual stress, and hence, can be deposited onto nearly any substrate material to any thickness. Controlled deposition yields reproducible NPC density, morphology, and porosity, without any discernable variation in surface chemistry. NPC coatings > 20 {micro}m thick with density < 5% that of graphite have been demonstrated. NPC can be 'doped' with nearly any metal during growth to provide further enhancements in analyte detection and selectivity. Optimized NPC-coated SAW devices were compared directly to commonly-used polymer coated SAWs for sensitivity to a variety of VOC, TIC, CWA and NPP. In every analyte, NPC outperforms each polymer coating by multiple orders-of-magnitude in detection sensitivity, with improvements ranging from 103 to 108 times greater detection sensitivity! NPC-coated SAW sensors appear capable of detecting most analytes tested to concentrations below parts-per-billion. In addition, the graphitic nature of NPC enables thermal stability > 600 C, several hundred degrees higher than the polymers. This superior thermal stability will enable higher-Temperature preconcentrator operation, as well as greatly prolonged device reliability, since polymers tend to degrade with time and repeated thermal cycling.
In this investigation, YBa{sub 2}Cu{sub 3}O{sub 7{minus}{delta}} (YBCO) films were fabricated via a metal acetate, trifluoroacetic acid based sol-gel route, and spin-coat deposited on (100) LaAlO{sub 3} with a focus on maximizing J{sub c}, while minimizing processing time. We demonstrate that the use of a low pO{sub 2} atmosphere during the pyrolysis stage can lead to at least a tetiold reduction in pyrolysis time, compared to a 1 atm. O{sub 2} ambient. High-quality YBCO films on LaAlO{sub 3}, with J{sub c} values up to 3 MA/cm{sup 2} at 77 K, can be routinely crystallized from these rapidly pyrolyzed films.
HgBa{sub 2}CaCu{sub 2}O{sub 6+{delta}} (Hg-1212) thin films were fabricated by exchanging the TI cations in TlBa{sub 2}CaCu{sub 2}O{sub 7-{delta}} (Tl-1212) thin films with Hg cations, causing a 30-K increase in Tc. To determine how this exchange effects such a Tc increase, the irreversibility lines, temperature dependence of critical current density, and temperature dependence of Hall angle of Hg-1212 and T1-1212 thin films were measured and then compared. The results strongly suggest that the Tc shift is caused by a doubling of charge carrier density.
Short pulse laser damage and ablation of amorphous, diamond-like carbon films is investigated. Material removal is due to fracture of the film and ejection of large fragments, which exhibit a broadband emission of microsecond duration.
Nanostructural characterization of amorphous diamondlike carbon (a-C) films grown on silicon using pulsed-laser deposition (PLD) is correlated to both growth energetic and film thickness. Raman spectroscopy and x-ray reflectivity probe both the topological nature of 3- and 4-fold coordinated carbon atom bonding and the topographical clustering of their distributions within a given film. In general, increasing the energetic of PLD growth results in films becoming more ``diamondlike'', i.e. increasing mass density and decreasing optical absorbance. However, these same properties decrease appreciably with thickness. The topology of carbon atom bonding is different for material near the substrate interface compared to material within the bulk portion of an a-C film. A simple model balancing the energy of residual stress and the free energies of resulting carbon topologies is proposed to provide an explanation of the evolution of topographical bonding clusters in a growing a-C film.