Search

Cold spray has been used to manufacture more than 800 Cu-In-Ga (CIG) sputter targets for deposition of high-efficiency photovoltaic thin films. It is a preferred technique since it enables high deposit purity and transfer of non-equilibrium alloy states to the target material. In this work, an integrated approach to reliability assessment of such targets with deposit weight in excess of 50 lb. is undertaken, involving thermal-mechanical characterization of the material in as-deposited condition, characterization of the interface adhesion on cylindrical substrate in as-deposited condition, and developing means to assess target integrity under thermal-mechanical loads during the physical vapor deposition (PVD) sputtering process. Mechanical characterization of cold spray deposited CIG alloy is accomplished through the use of indentation testing and adaptation of Brazilian disk test. A custom lever test was developed to characterize adhesion along the cylindrical interface between the CIG deposit and cylindrical substrate, overcoming limitations of current standards. A cohesive zone model for crack initiation and propagation at the deposit interface is developed and validated using the lever test and later used to simulate the potential catastrophic target failure in the PVD process. It is shown that this approach enables reliability assessment of sputter targets and improves robustness.

In this work the Brazilian disc test technique is applied to mechanical characterization of cold-sprayed Cu-In-Ga deposits. The main advantage of the test is that the material can be tested in its end product form while naturally attending to the distribution of micro-structural phases and flaws by choosing an appropriate specimen size. The stress state of the test specimen can be obtained in analytical closed form and testing can be readily extended to obtain the stress-strain curve. While limited to low ductility materials the Brazilian disc test appears ideal for testing as-sprayed cold spray deposits due to their typically brittle nature.

A computer simulation was created to model the transport of sputtered atoms through an ionized physical vapor deposition (IPVD) system. The simulation combines Monte Carlo and fluid methods to track the metal atoms that are emitted from the target, interact with the IPVD plasma, and are eventually deposited somewhere in the system. Ground-state neutral, excited, and ionized metal atoms are tracked. The simulation requires plasma conditions to be specified by the user. Langmuir probe measurements were used to determine these parameters in an experimental system in order to compare simulation results with experiment. The primary product of the simulation is a prediction of the ionization fraction of the sputtered atom flux at the substrate under various conditions. This quantity was experimentally measured and the results compared to the simulation. Experiment and simulation differ significantly. It is hypothesized that heating of the background gas due to the intense sputtered atom flux at the target is primarily responsible for this difference. Heating of the background gas is not accounted for in the simulation. Difficulties in accurately measuring plasma parameters, especially electron temperature, are also significant.

Conventional magnetron sputter deposition with a rf inductively coupled plasma (ICP) has demonstrated that ionized metal fluxes can be effectively utilized to fill trenches and vias with high aspect ratios. The ICP is created with a seven turn (1/2 wavelength), water cooled coil located between the magnetron cathode and the substrate. A large fraction of the metal atoms sputtered from the magnetron cathode are ionized by the ICP. These ions are accelerated across the sheath toward the substrate and deposited at normal incidence, by placing a negative bias on the substrate. A gridded energy analyzer configured with a quartz crystal microbalance is located in the center of the substrate plane to determine the ion and neutral deposition rates. While keeping the magnetron power, rf coil, target to substrate distance, pressure and diagnostic location constant, the ionization fraction was measured for two metal targets: Cu and Ti using three different working gases: Kr, Ar and Ne. Variations in target materials...

A helicon antenna that sits remotely outside the vacuum system is attached to a magnetron sputtering system. This increases the electron temperature, which increases the ionization of the sputter flux for achieving ionized physical vapor deposition (IPVD). There are no shadowing and contamination problems, unlike other IPVD devices with immersed coils, since the helicon antenna is outside the vacuum system. Furthermore, the target to substrate distance can be kept small. At 2 kW magnetron power, 4 kW helicon power, 45 mTorr argon gas, and with a copper target, ionization fractions to the substrate of 51±10% and a deposition rate of 847±42 A/min are measured using a quartz crystal oscillator (QCO) and a multi-grid filter. Without the antenna, the ionization fraction to the QCO is 30±6% and the deposition rate is 815±41 A/min. Multiple remote sources are envisioned to be positioned radially around a sputtering chamber, controlling uniformity while increasing the ionization further. Since 21% additional ionization is achieved using only one source, with no threat of contamination inside the vacuum chamber, the helicon source has good potential for a secondary plasma source in IPVD applications.

A rotating magnet dc planar magnetron with a 33-cm diameter aluminum target is coupled with a secondary plasma source to ionize the sputtered metal neutral flux to control the angular distribution of the flux arriving at the surface of the substrate. For this purpose, a radio-frequency (rf) plasma is created between the sputtering target and substrate by a three-turn coil located in the vacuum chamber. The rf plasma increases the electron temperature and density, which results in significant ionization of the neutral metal flux from the sputtering target. By applying a small negative bias to the substrate, metal ions are drawn to the substrate at normal incidence. A gridded energy analyzer and a quartz crystal microbalance (QCM) were used to determine the ion and neutral deposition rates. From this, the ionization fraction of the flux incident onto the QCM is determined.

Sufficient neutral atom and molecular throughput is essential for the steady-state operation of the proposed Tokamak Physics Experiment tokamak. To predict the throughput, the B2 edge-plasma fluid code and the DEGAS Monte Carlo neutral transport code were coupled globally. For the day 1 low-power (17.5-MW) operation condition, the recycling coefficient for both codes matched at 0.985, implying that for every 1000 ions striking the divertor plate, 15 are ultimately removed down the pump duct. The neutral molecular density was 2.52{+-}0.15x10{sup 19}m{sup 3}, giving a throughput of 92.6{+-}5.6 Torr.l/s. Varying the scrape-off length for the plasma extending into the gap between the baffle and the plate from 0 to 2 cm decreased the throughput by a factor of >2. Moving the strike point away from the gap at first increases the throughput by lessening the pumping efficiency of the plasma in the gap. As the plasma is moved even farther away, the throughput drops due to a lack of source term for neutrals entering the pumped region. Illustrating the importance of moving the source term, moving the strike point away from the gap but retaining the original plasma in the gap lowers the throughput by a factor of 10. Altering themore » curvature of the baffle has little effect on the neutral solution. 7 refs., 9 figs.« less

The Tokamak Physics Experiment is designed to develop the scientific basis for a compact and continuously operating tokamak fusion reactor. It is based on an emerging class of tokamak operating modes, characterized by beta limits well in excess of the Troyon limit, confinement scaling well in excess of H-mode, and bootstrap current fractions approaching unity. Such modes are attainable through the use of advanced, steady state plasma controls including strong shaping, current profile control, and active particle recycling control. Key design features of the TPX are superconducting toroidal and poloidal field coils; actively-cooled plasma-facing components; a flexible heating and current drive system; and a spacious divertor for flexibility. Substantial deuterium plasma operation is made possible with an in-vessel remote maintenance system, a lowactivation titanium vacuum vessel, and shielding of ex-vessel components. The facility will be constructed as a national project with substantial participation by U.S. industry. Operation will begin with first plasma in the year 2000.

The DEGAS neutral atom transport code was used to simulate helium pumping and D/T throughput in ITER. The sensitivity of the simulation to two different reflection models, four transmission probabilities from the exit of the simulation to the pump (0.0625, 0.125, 0.1875 and 0.250), and a 2-D model versus a 3-D model were analyzed. The variation in reflection model changes the densities in the duct and the recycling of D/T by a factor of 1.6. The variation in the transmission probabilities affects these same quantities by a factor of 2.5. The dimensionality of the simulation affects the density profile in the duct. A transmission probability from the exit of the DEGAS simulation to the pump of 0.110 to 0.125 was calculated from the ITER reference drawings. Using this quantity and the DEGAS results, an exhaust rate of 112 to 127 moles/h is predicted, implying that the reference pumping systems may be larger than necessary by a factor of 2.

Summary form only given. An inductively coupled plasma (ICP) RF antenna is added to a commercial magnetron sputtering machine. The power absorbed by this multi-turn water-cooled coil heats the electrons and raises their density, thereby increasing the ionization cross section for the metal sputtered neutrals. This creates up to 78% ionization of the depositing flux with less than 800 W of RF power at 2.5 kW of magnetron power. By applying a bias to the substrate the majority of the metal can be deposited normally, and at energies controllable by the set bias level. Aluminum and copper targets are used with argon and neon working gases. Operating pressures of 15 to 35 mTorr are investigated. Alternate ionization schemes not requiring an in-situ antenna are also investigated. The applications of this design to higher aspect-ratio trenches and sub-.5 /spl mu/m features is discussed.

Summary form only given, as follows. An analytic model is developed for a cylindrically symmetric inductively coupled plasma system in order to find the electron temperature and density distribution. Boltzmann's equations solved by a computer code using a 2-term spherical harmonic expansion. Analytic results are compared to experimental measurements made with a probe. The apparatus is a commercial system donated by Materials Research Corporation with an RF coil inserted between the target and substrate. The RF coil deposits additional power into the system, increasing the electron temperature and density. This increases the amount of metal ionization in the plasma. Far from the target, the resulting plasma is dominated by this ionization source, so the plasma at the magnetron target is not accounted for in the analytic model. In the model, electric and magnetic fields from the RF coil are found as a function of position and the power deposition profile is calculated. Insights gained from this model are used to guide research efforts in ionizing the sputter flux in the magnetron.