Search

The energy distributions of secondary ions for the ion beam sputtering of a Ga 2 O 3 target using O 2 + and Ar + ions are measured in dependence on various process parameters using energy-selective mass spectrometry. The process parameters include sputtering geometry (ion incidence angle α , polar emission angle β , scattering angle γ), the energy of incident ions E ion , and the background pressure of O 2. The main secondary ion species are identified to be Ga + , O + , O 2 + , and, when argon is used as a process gas, Ar +. The changes in the sputtering geometry and the primary ion energy have the most impact on the energy distributions of secondary Ga + and O + ions, giving control over the high-energy tail, which is attributed to anisotropy effects in sputtering. The formation of O 2 + ions is attributed to collisions with background gas molecules, as their energy distributions are not influenced by the sputtering geometry or the primary ion energy. The increase of the O 2 pressure leads to a minor decrease of the energy of Ga + ions due to collisions with the background gas particles. The use of primary Ar + ions with O 2 background pressure does not show any specific effect on energy distributions of Ga + , O + , and O 2 + ions except for the case without additional O 2 background. In the latter case, much fewer O + and O 2 + ions are produced indicative of oxygen depletion of the surface due to preferential sputtering of oxygen. At all considered O 2 pressures, the energy distributions of Ar + ions have a high-energy peak, attributed to direct scattering events. The trends in experimental data show qualitative agreement to simulations using the Monte Carlo code SDTrimSP. [ABSTRACT FROM AUTHOR]

Magnetron sputtering of fluorine-doped tin oxide (FTO) is a scalable deposition method for largearea transparent conducting films used in fenestration, photovoltaics, and other applications. The electrical conductivity of sputtered FTO is, however, lower than that of spray-pyrolized FTO because of the ion damage induced by high energy ions leading to a reduction of the crystal quality in sputtered FTO films. In this study, various ion species present during the reactive sputtering of a metallic tin target in a mixed Ar/O2/CF4 atmosphere are systematically characterized by energy and mass spectrometry, and possible ways of controlling the ion fluxes are explored. Ion energy distribution functions (IEDFs) of the negative ions F- and O- exhibit large peaks at an energy corresponding to the full target voltage. Although the applied partial pressure of CF4 is about 1/30 than that of O2, the obtained IEDFs of F- and O- have comparable peak height, which can be attributed to a higher electronegativity of F. The IEDFs of positively charged O+, O2+, Ar+, and Sn+ species have their peaks around 2-8 eV. To control ion fluxes a solenoid or permanent magnets were placed between the target and the mass spectrometer. The flux of positive ions could be varied by several orders of magnitude as a function of the applied current through the solenoid, whereas the high-energy (>100 eV) negative F- and O- ions were not notably deflected. By using permanent magnets with the B-field orthogonal to the ion trajectory, the flux of O- ions could be decreased by two orders and the exposure to the high-energy F- ions was completely suppressed. [ABSTRACT FROM AUTHOR]

We have measured the band edge energies of CdxZn1–xO thin films as a function of composition by three independent techniques: we determine the Fermi level stabilization energy by pinning the Fermi level with ion irradiation, measure the binding energy of valence band states and core levels by X-ray photoelectron spectroscopy, and probe shifts in the conduction band and valence band density of states using soft X-ray absorption and emission spectroscopy, respectively. The three techniques find consensus in explaining the origin of compositional trends in the optical-bandgap narrowing upon Cd incorporation in wurtzite ZnO and widening upon Zn incorporation in rocksalt CdO. The conduction band minimum is found to be stationary for both wurtzite and rocksalt alloys, and a significant upward rise of the valence band maximum accounts for the majority of these observed bandgap changes. Given these band alignments, alloy disorder scattering is found to play a negligible role in decreasing the electron mobility for all alloys. These band alignment details, combined with the unique optical and electrical properties of the two phase regimes, make CdZnO alloys attractive candidates for photoelectrochemical water splitting applications. [ABSTRACT FROM AUTHOR]

Pulsed emissive probe techniques have been used to determine the plasma potential distribution of high power impulse magnetron sputtering (HiPIMS) discharges. An unbalanced magnetron with a niobium target in argon was investigated for a pulse length of 100 μs at a pulse repetition rate of 100 Hz, giving a peak current of 170 A. The probe data were recorded with a time resolution of 20 ns and a spatial resolution of 1 mm. It is shown that the local plasma potential varies greatly in space and time. The lowest potential was found over the target's racetrack, gradually reaching anode potential (ground) several centimeters away from the target. The magnetic presheath exhibits a funnel-shaped plasma potential resulting in an electric field which accelerates ions toward the racetrack. In certain regions and times, the potential exhibits weak local maxima which allow for ion acceleration to the substrate. Knowledge of the local E and static B fields lets us derive the electrons' E×B drift velocity, which is about 105 m/s and shows structures in space and time. [ABSTRACT FROM AUTHOR]

The plasma over a magnetron's erosion 'racetrack' is not azimuthally uniform but concentrated in distinct dense ionization zones which move in the E×B direction with about 10% of the electron E×B/B2 drift velocity. The ionization zones are investigated with a gated camera working in concert with a streak camera for Al, Nb, Cu, and W targets in Ar or Kr background gas. It is found that each ionization zone has a high plasma density edge, which is the origin of a plasma-generating electron jet leaving the target zone. Each region of strong azimuthal plasma density gradient generates an azimuthal electric field, which promotes the escape of magnetized electrons and the formation of electron jets and plasma flares. The phenomena are proposed to be caused by an ionization instability where each dense plasma zone exhibits a high stopping power for drifting high energy electrons, thereby enhancing itself. [ABSTRACT FROM AUTHOR]

Gas compression and strong rarefaction have been observed for high power impulse magnetron sputtering (HIPIMS) discharges using a copper target in argon. Time-resolved ion saturation currents of 35 probes were simultaneously recorded for HIPIMS discharges operating far above the self-sputtering runaway threshold. The argon background pressure was a parameter for the evaluation of the spatial and temporal development of the plasma density distribution. The data can be interpreted by a massive onset of the sputtering flux (sputter wind) that causes a transient densification of the gas, followed by rarefaction and the replacement of gas plasma by the metal plasma of sustained self-sputtering. The plasma density pulse follows closely the power pulse at low pressure. At high pressure, the relatively remote probes recorded a density peak only after the discharge pulse, indicative for slow, diffusive ion transport. [ABSTRACT FROM AUTHOR]

Dense metal plasmas obtained by self-sputtering far above the runway threshold are well suited to generate intense quiescent ion beams. The dilemma of high current density and charge state purity can be solved when using target materials of low surface binding energy by utilizing nonresonant exchange reactions before ion extraction. Space-charge-limited quiescent beams of Cu+, Zn+, and Bi+ with ∼10 mA/cm2 have been obtained through multiaperture gridded ion extraction up to 45 kV from self-sputtering plasmas. [ABSTRACT FROM AUTHOR]

The plasma of a high power impulse magnetron sputtering system has been investigated using a time-of-flight spectrometer. The target materials included high sputter yield materials (Cu, Ag), transition metals (Nb, Cr, Ti), and carbon (graphite); the sputtering gases were argon, krypton, and nitrogen, and two different target thicknesses were selected to consider the role of the magnetic field strength. Measurements for selected combinations of those parameters give quantitative information on the transition from gas-dominated to metal-dominated (self-sputtering) plasma, on the fractions of ion charge states, and in the case of molecular gases, on the fraction of atomic and molecular ions. [ABSTRACT FROM AUTHOR]

A plasma assistance system was investigated with the goal to operate high power impulse magnetron sputtering (HiPIMS) at lower pressure than usual, thereby to enhance the utilization of the ballistic atoms and ions with high kinetic energy in the film growth process. Gas plasma flow from a constricted plasma source was aimed at the magnetron target. Contrary to initial expectations, such plasma assistance turned out to be contraproductive because it led to the extinction of the magnetron discharge. The effect can be explained by gas rarefaction. A better method of reducing the necessary gas pressure is operation at relatively high pulse repetition rates where the afterglow plasma of one pulse assists in the development of the next pulse. Here we show that this method, known from medium-frequency (MF) pulsed sputtering, is also very important at the much lower pulse repetition rates of HiPIMS. A minimum in the possible operational pressure is found in the frequency region between HiPIMS and MF pulsed sputtering. [ABSTRACT FROM AUTHOR]

Short-pulse high-current discharges in vacuum were investigated with the goal to maximize the ion charge state number. In a direct extension of previous work [G. Y. Yushkov and A. Anders, Appl. Phys. Lett. 92, 041502 (2008)], the role of pulse length, rate of current rise, and current amplitude was studied. For all experimental conditions, the usable (extractable) mean ion charge state could not be pushed beyond 7+. Instead, a maximum of the mean ion charge state (about 6+ to 7+ for most cathode materials) was found for a power of 2–3 MW dissipated in the discharge gap. The maximum is the result of two opposing processes that occur when the power is increased: (i) the formation of higher ion charge states and (ii) a greater production of neutrals (both metal and nonmetal), which reduces the charge state via charge exchange collisions. [ABSTRACT FROM AUTHOR]

Spark plasma sintering (SPS) is a remarkable method for synthesizing and consolidating a large variety of both novel and traditional materials. There are a number of mechanisms proposed to account for the enhanced sintering abilities of the SPS process. Of these mechanisms, the one most commonly put forth and the one that draws the most controversy involves the presence of momentary plasma generated between particles. This study employs three separate experimental methods in an attempt to determine the presence or absence of plasma during SPS. The methods employed include in situ atomic emission spectroscopy, direct visual observation, and ultrafast in situ voltage measurements. Using these experimental techniques, no evidence of plasma was found during the SPS process. This result was confirmed using several different powders across a wide spectrum of SPS conditions. [ABSTRACT FROM AUTHOR]

The commonly used current-voltage characteristics are found inadequate for describing the pulsed nature of the high power impulse magnetron sputtering (HIPIMS) discharge; rather, the description needs to be expanded to current-voltage-time characteristics for each initial gas pressure. Using different target materials (Cu, Ti, Nb, C, W, Al, and Cr) and a pulsed constant-voltage supply, it is shown that the HIPIMS discharges typically exhibit an initial pressure dependent current peak followed by a second phase that is power and material dependent. This suggests that the initial phase of a HIPIMS discharge pulse is dominated by gas ions, whereas the later phase has a strong contribution from self-sputtering. For some materials, the discharge switches into a mode of sustained self-sputtering. The very large differences between materials cannot be ascribed to the different sputter yields but they indicate that generation and trapping of secondary electrons play a major role for current-voltage-time characteristics. In particular, it is argued that the sustained self-sputtering phase is associated with the generation of multiply charged ions because only they can cause potential emission of secondary electrons, whereas the yield caused by singly charged metal ions is negligibly small. [ABSTRACT FROM AUTHOR]

Cathodic arc plasmas are considered fully ionized and they contain multiply charged ions, yet gaseous and metal neutrals can be present. It is shown that they can cause a significant reduction of the ion charge states as measured far from the cathode spots. Several cathode materials were used to study the evolution of the mean ion charge state as a function of time after arc ignition. The type of cathode material, arc current amplitude, intentionally increased background gas, additional surfaces placed near the plasma flow, and other factors influence the degree of charge state reduction because all of these factors influence the density of neutrals. In all cases, it was found that the mean ion charge state follows an exponential decay of first order, [formula], where A is a parameter describing the importance of the decay, τ is the characteristic decay time, and Qss is a steady-state value approached for continuous arc operation. The extrapolated values Q(t→0) indicate surprisingly high mean charge states as produced at cathode spots and not “skewed” by charge exchange collisions with neutrals. [ABSTRACT FROM AUTHOR]

Charge-state-resolved ion energy distribution functions were measured for pulsed cathodic arcs taking the sheath into account that formed between the plasma and the entrance of a combined energy and mass spectrometer. An electron emitting probe was employed to independently determine the plasma potential. All results were obtained by averaging over several individual measurements because the instantaneous energy distributions and the plasma potential show large amplitude fluctuations due to the explosive nature of the arc plasma generation. It was found that the ion energy distribution functions in the plasma were independent of the ion charge state. This is in contrast to findings with continuously operating, direct-current arcs that employ a magnetic field at the cathode to steer the cathode spot motion. The different findings indicate the important role of the magnetic steering field for the plasma properties of direct-current arcs. The results are further supported by experiments with “biased plasmas” obtained by shifting the potential of the anode. Finally, it was shown that the ion energy distributions were broader and shifted to higher energy at the beginning of each arc pulse. The characteristic time for relaxation to steady state distributions is about 100 μs. [ABSTRACT FROM AUTHOR]

The charge-state-resolved ion energy distributions (IEDs) in filtered aluminum vacuum arc plasmas were measured and analyzed at different oxygen and argon pressures in the range of 0.5–8.0 mTorr. A significant reduction of the ion energy was detected as the pressure was increased, most pronounced in an argon environment and for the higher charge states. The corresponding average charge state decreased from 1.87 to 1.0 with increasing pressure. The IEDs of all metal ions in oxygen were fitted with shifted Maxwellian distributions. The results show that it is possible to obtain a plasma composition with a narrow charge-state distribution as well as a narrow IED. These data may enable tailoring thin film properties through selecting growth conditions that are characterized by predefined charge state and energy distributions. [ABSTRACT FROM AUTHOR]

Current fluctuations of cathodic arcs were recorded with high analog bandwidth (up to 1 GHz) and fast digital sampling (up to 5 Gsamples/s). The power spectral density of the arc current was determined by fast Fourier transform clearly showing material dependent, nonlinear features in the frequency domain. These features can be associated with the nonlinear impedance of the conducting channel between cathode and anode, driven by the explosive nature of electron emission and plasma formation. The characteristic times of less than 100 ns can be associated with individual explosive processes, “ectons,” and therefore represent the short-time physical cutoff for the fractal model of cathodic arcs. [ABSTRACT FROM AUTHOR]

The charge-state-resolved ion energy distributions (IEDs) of aluminum vacuum arc plasma species were measured and analyzed for different geometric and magnetic field configurations. The IEDs were fitted by shifted Maxwellian distributions. Plasma expansion in the absence of a magnetic field showed higher ion energies for higher charge states. The introduction of a magnetic field (independent of geometric configuration) resulted in a broader distribution and increased average ion energies. The energy gain was approximately proportional to the charge state, which may be due to the presence of electric fields in the magnetized plasma. The evolution of ion energy distributions is relevant to thin-film growth, and it is shown that the IEDs can be modified by suitable magnetic field configurations. [ABSTRACT FROM AUTHOR]

The species and ion charge state evolution of a pulsed cathodic arc plasma was investigated at different pressures. A zirconium cathode was operated in a nitrogen environment, and the plasma composition was analyzed by time-of-flight charge-to-mass spectrometry. Large plasma chemistry changes were detected with respect to time and pressure. The 250 μs plasma pulse can be divided in two characteristic phases: a transient phase before 150 μs and a steady state phase for all later times. The measured changes in plasma chemistry in the transient phase at p∼10-5 Torr are explained by charge transfer collisions, while the increasing N+ fraction in the p>10-5 Torr range most likely originates from erosion of the nitrided cathode surface. In the steady-state phase, a pressure-induced change from higher to lower charge states was observed, which was mainly due to scattering of self-sputtered metal followed by metal ion-atom charge exchange collisions. These results are of importance for understanding the evolution of thin film composition and microstructure during reactive plasma deposition. [ABSTRACT FROM AUTHOR]

Ions in cathodic arc plasmas have high velocities attained by acceleration at cathode spots. Among other techniques, two time-of-flight setups had previously been used to determine ion velocities. These measurements showed significant discrepancies especially for some light cathode materials like carbon. To reconcile the differences, a thorough investigation was conducted using carbon as the cathode material. It is shown that systematic errors occur when the ion source time-of-flight system is not operated near perveance-matching conditions. The extracted ion beam is not parallel but divergent, and the Faraday cup detector measures only a fraction of the beam. In contrast, plasma source experiments without ion extraction are free of such distortions. In the absence of an external magnetic field, the average carbon ion energy has been determined to be in the range 18.5-20.5 eV for arc currents in the range 100-600 A, in agreement with previous plasma source measurements and other literature data. [ABSTRACT FROM AUTHOR]

The cathodic arc plasmas produced by cathode spots usually include macroparticles, which is undesirable for many applications. A common way of removing macroparticles is to use curved solenoid filters that guide the plasma from the source to the substrate. In this work, an arc source with a relatively small cathode is used, limiting the possible locations of plasma production. The relative position of cathodic arc source and macroparticle filter was systematically varied and the filtered plasma current was recorded. It was found that axis-symmetric plasma injection lead to maximum throughput only if an anode aperture was used, which limited the plasma to near-axis flow by scraping off plasma at larger angles to the axis. When the anode aperture was removed, more plasma could enter the filter. In this case, maximum filtered ion current was achieved when the plasma was injected off-axis; namely, offset in the direction where the filter is curved. Such behavior was anticipated because the plasma column in the filter is known to be shifted by E×B and centrifugal drift as well as by non-axis-symmetric components of the magnetic field in the filter entrance and exit plane. The data have implications for plasma transport variations caused by different spot locations on cathodes that are not small compared to the filter cross section. © 2004 American Institute of Physics. [ABSTRACT FROM AUTHOR]

We describe the temporal development of the plasma composition in a pulsed plasma stream generated by cathodic arc. Cathodes of Zr and Cr were operated at various nitrogen pressures. The time-resolved plasma composition for the cathode materials was analyzed with time-of-flight charge-to-mass spectrometry, and was found to be a strong function of the nitrogen pressure. Large plasma composition gradients were detected within the first 60 µs of the pulse, the nitrogen ion concentration increasing with increasing pressure. The results are explained by the formation and erosion of a compound layer formed at the cathode surface in the presence of a reactive gas. The average charge state was also found to be affected by the reactive gas pressure as well as by the time after ignition. The charge states were highest in the beginning of the pulse at low nitrogen pressure, decreasing to a steady-state value at higher pressure. These results are of importance for reactive plasma processing and for controlling the evolution of thin film composition and microstructure. [ABSTRACT FROM AUTHOR]

Curved magnetic filters are often used for the removal of macroparticles from cathodic arc plasmas. This study addresses the need to further reduce losses and improving plasma throughput. The central figure of merit is the system coefficient κ defined as a filtered ion current normalized by the plasma-producing arc current. The coefficient κ is investigated as a function of continuous and pulsed magnetic field operation, magnetic field strength, external electric bias, and arc amplitude. It increases with positive filter bias but saturates at about 15 V for relatively low magnetic field (∼10 mT), whereas stronger magnetic fields lead to higher κ with saturation at about 25 V. Further increase of positive bias reduces κ. These findings are true for both pulsed and continuous filters. Bias of pulsed filters has been realized using the voltage drop across a self-bias resistor, eliminating the need for a separate bias circuit. Almost 100 A of filtered copper ions have been obtained in pulsed mode, corresponding to κapprox. 0.04. The results are interpreted by a simplified potential trough model. © 2003 American Institute of Physics. [ABSTRACT FROM AUTHOR]

The velocity distribution function of vacuum arc ions can be measured by a time-of-flight technique. The measuring principle makes use of the well-justified assumption that the ion drift velocity from the cathode spot region to a collector is approximately constant. It is shown that the negative time derivative of the collector current is directly proportional to the ion distribution function provided that the time-averaged emission of ions from cathode spots is constant until the arc is rapidly switched off. In the experiment, arc termination took about 700 ns, which is much faster than the decay of the ion current measured at the collector placed more than 2 m from the cathode. The experimental distribution functions for most cathode materials show one large peak with a tail and one or more small peaks at higher ion velocities. The typical peak position is at about 10[sup 4] m/s, with the precise values being material specific. The distribution functions for some materials exhibit not one but several peaks. No conclusive answer can be given about the nature of these peaks. Arguments are presented that the peaks are not caused by different charge states or plasma contamination but rather are due to insufficiently averaged source fluctuations and/or acceleration by plasma instabilities. [ABSTRACT FROM AUTHOR]

Burning voltages of vacuum arcs were measured for 54 cathode materials and compared with literature data. As anticipated, a correlation between the arc burning voltage and the plasma temperature was found. However, more importantly, a correlation between the cohesive energy of the cathode material and the arc burning voltage could be demonstrated. This link between a cathode material property, the cohesive energy, and a discharge property, the arc burning voltage, is essential for the operation of the vacuum arc discharge because is determines the plasma temperature. Energy balance considerations show that this "cohesive energy rule" is responsible for several other secondary relationships, such as the correlation between the mean ion charge state and the boiling temperature of the cathode. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]