B Gelin

Atom assisted sputtering yield amplification

At first sight one might assume that it is unlikely to influence the sputtering yield of a specific ion/substrate combination by any external means. However, we have found that such an influence may well be introduced. The sputtering yield is predominantly determined by the ion/substrate momentum transfer efficiency and the energy of the incoming ion. Sputter erosion of, e.g., carbon atoms by argon ions from a carbon substrate exhibits a very low sputtering yield. Due to the difference in masses between carbon and argon much of the momentum is transferred into the bulk of the carbon substrate. This situation could be changed by simultaneous codeposition of Pt atoms onto the carbon substrate surface during the argon sputtering. Keeping the argon flux at a level well above what is needed to sputter remove all the deposited Pt atoms the following effect occurs. Some of the deposited Pt atoms will be forward implanted by the energetic argon ions into the near surface region of the carbon substrate. Collision ...

Ion assisted selective thin film deposition

Ion bombardment of a growing film is one known way of changing film properties. Also in reactive etching, ions may assist surface reactions. We will present a study where different substrates are bombarded by energetic argon ions during deposition of titanium. If titanium is deposited onto three different substrates without influencing the substrate by any ion irradiation, a thin titanium film, equally thick, will be formed on all three substrates. If, however, the substrates are bombarded by energetic argon ions (500–800 eV) during deposition of titanium, we have found that quite different film thicknesses may be obtained on different substrates. In fact, a balance may be found where actually one substrate is sputter etched while another substrate under identical deposition conditions (same run) is deposited by titanium. Results from rf‐bias sputtering in argon shows selectivity, e.g., between Si and Pt at 10 mTorr pressure, 1600 V target voltage and 800 V substrate bais. Ti is deposited onto silicon but...

Influence of substrate material on the initial thin film growth during ion deposition from a glow discharge

Abstract If the argon is replaced by e.g. CH4 during rf-sputtering, a carbon film will be grown on the rf-excited electrode. Two major processes take place simultaneously at this electrode: physical sputtering of the electrode surface by ionized energetic methane fragments. Ionized energetic methane fragments stick to the electrode surface, thereby forming a carbon film. Depending on the sticking coefficients and sputtering yield values of the methane fragments onto the exposed substrate surface there will be a certain time interval before the substrate surface is completely covered by a carbon layer. The average carbon film thickness d(t) does not increase linearly with time at the initial stage of growth (i.e. before the carbon film completely covers the substrate—electrode surface). The reason for this effect is that during the initial film growth the substrate surface gradually changes from a 100% substrate material surface to a 100% carbon film surface. Since the sticking coefficients of the methane fragments onto these two materials may differ, a non-linear growth rate will occur. Experimental results are presented that verify that the initial carbon film growth rate depends on the substrate material. By considering the combination of sputter etching and deposition a model that describes the general behaviour of the initial growth has been developed. The growth rate reaches a steady state value at an average thickness d0 of only 20–50 A. Since this represents only 10–20 atomic layers, considerable fluctations in film thicknesses at different points may occur due to the statistical nature of the process. To get information about the expected fluctuations a simple Monte Carlo simulation of the process has been carried out. From this simulation we obtain a first order approximation about film roughness, atomic mixing and film formation at the substrate / carbon film interface. These results also confirm the growth mechanism responsible for different growth rates on different substrate surfaces.

Self‐limiting etch depths using simultaneous sputter etching/deposition technique

During sputter etching in argon, time is most commonly used as the parameter determining the total etch depth. If the argon is replaced by CH4 during substrate etching the effect will be that etching stops, CH4 decomposes in the plasma, and a carbon film starts to deposit on the substrate. This effect will also be obtained if a mixture of Ar/CH4 is used during processing. Depending on the concentration of CH4 in the gas mixture it will take different times before a monolayer of carbon is formed at the substrate surface. During this time interval the substrate material is exposed to energetic ion sputter etching. The total amount of original substrate material etched away before the substrate is completely covered by carbon is dependent of Ar/CH4 ratio and on substrate material. The carbon layer can easily be removed by an O2 plasma when the process is completed. A process has thus been obtained that can create a self‐limited etch depth (SLED) into a surface material independent of processing time as long ...

Patterning with the use of ion-assisted selective deposition

Abstract High bias sputtering in inert argon may be used to selectively deposit thin films on patterned substrates. Simultaneous sputter etching and film deposition will, at the initial stage of film formation, cause a newly discovered interface phenomena. The sputtering yield of a monoatomic thin film will strongly depend on the underlying bulk substrate. During ion bombardment of a growing ultra-thin film on top of a patterned substrate the film may be preferentially sputter eroded on areas where the sputtering yield of the thin film has the largest value. If a critical balance between deposition rate and sputter erosion rate is selected, actual selective large area deposition may be obtained. No degradation of the pattern was observed after selective deposition. This rules out the possibility of any contribution from local variations in the glow discharge to this selective deposition effect. We will also show that it is possible to predict the selectivity between different substrate materials by simulation of the sputtering yield values of the thin film by Monte Carlo calculations of the collision cascade process.

Selective deposition of Ti: An interface study

If different substrates are bombarded by energetic (500–1000 eV) Ar ions during deposition of Ti, quite different film thicknesses may result on the substrates. In fact, on some of the substrates there will be no net deposition of Ti, while on others there will be formed a thin film of Ti. This is the bias sputtering selective deposition technique. The advantage with this technique is that the process is carried out in inert argon gas. Furthermore, almost any material may be selectively deposited. Since the substrate is bombarded by energetic argon ions during deposition its surface will be sputter etched. During the buildup of the Ti film a mixed interface may result. We have investigated the sharpness of these interfaces by Rutherford backscattering and by Auger depth profiling. We have found a critical processing interval where soft interfaces are formed by this technique. The experiments indicate that this ‘‘soft interface’’ interval appears when the deposition rate of sputtered Ti dominates only by a...

Ion assisted selective deposition of thin films

Abstract The initial growth mechanism at the substrate/film interface may be affected by inert argon ion bombardment during thin film deposition. If the ion bombardment is very intense, sputter removal may dominate over film growth. However, in the processing region where the deposition rate is very close to the sputter removal rate, we have observed that the net deposition rate is substrate dependent. It may even be that some deposition rates are found to be negative (etching). This effect results in some materials being selectively deposited onto certain substrates. This paper intends to explain the physics of this selective deposition mechanism.

Studies of a parallel-plate electron multiplier

A parallel-plate electron multiplier (PPEM) under the influence of a continuous flow of electrons at the input is analysed theoretically and by measurements. In contrast to the tubular channel electron multiplier the PPEM is equipped with four electrodes and a separate current collector. The four electrodes make it possible to operate the PPEM with multiplication on one of the side walls or both. They also provide the unique possibility of always being able to operate the PPEM at high gain. For some applications it is favourable from an assembly point of view to short-circuit the two back electrodes. The result is less sensitive to variations of gain around the optimum gain setting points.

Selective deposition of thin films by substrate argon ion bombardment

Abstract As an alternative to normal patterning in sub-micron integrated circuit technology, selective processes have gained in interest. One fairly well known process is the thermally activated decomposition of WF6 to deposit selectively a thin tungsten film onto silicon areas of patterned wafers. Some problems with the thermal decomposition technique are the need for a specific metal carrying gas and an elevated process temperature. In a series of articles we have presented a completely new process for selective deposition, using high bias sputtering in an argon atmosphere. If the substrate bias is increased to the point where the net deposition rate becomes almost zero, substrate-dependent effects such as different sputtering yields (because of the difference in binding energy to different substrate atoms) and sticking coefficients for different surfaces come into action. On a patterned silicon wafer, it is thus possible to deposit material on silicon areas but not on the oxide (or vice versa), without the use of any photoresist masking. To make the mathematical model describing this selective process more comprehensible, we have developed a graphical method using a nomogram, that simplifies the prediction of selectivity between various materials. To further point out the possibilities of this selective method, results from ion beam selective deposition of platinum onto patterned silicon wafers are also presented.

Parallel Plate Multichannel Electron Detector With Uniform Response

A new signal collecting technique has been developed useful for multichannel arrays of Parallel Plate Electron Multipliers (PPEMs). Previously the yield in multichannel detector production has been low [1] depending on nonuniformity in detector performance. Also, the arrays suffered from lack of resolution in the individual pulse height distributions. This new technique, based on single active dynode operation, increases the yield in useful multiplier production to almost 100%. Furthermore, good resolution (10%FWHM) in the recorded pulse height spectra can be achieved. Capacitive coupling between adjacent channels is almost eliminated. This technique resulted from measurements of DC dynode currents in PPEMs. Whenever a PPEM is exposed to a continuous flow of particles at the input, continuous field deformations occur in the active dynodes[2-6]. If the resulting changes in DC plate currents are observed, information of the actual electron trajectories and also about saturation in the channels are obtained. Analysis of these DC-current changes resulted in a better understanding of the device.