Specific Process Knowledge/Thin film deposition/Deposition of Tungsten/Sputtering of W in Sputter Coater 3: Difference between revisions

[[Specific_Process_Knowledge/Thin_film_deposition/Sputter_coater#Sputter_coater_03_(Cressington)|Sputter coater 03]] is mainly used for sample preparation before [[Specific_Process_Knowledge/Characterization/SEM: Scanning Electron Microscopy|SEM inspection]]. Usually, [[Specific_Process_Knowledge/Thin_film_deposition/Deposition_of_Gold|Au]] is sputtered on dielectric surfaces to compensate for charge build-up. Alternatively, many other materials can be sputtered if the setup allows (you need to have a turbo pump and Ar gas). Gold can be deposited even in pure oxygen plasma and only with roughing pumps. Fortunately, Sputter Coater 3 located in the basement of building 346, has both a turbo pump and an Ar working gas source. The pressure during the process is measured to be below 0.1 mBar. This page describes deposition of tungsten (W) with this instrument.

In summary, the tungsten films produced by Sputter-Coater 3 are of poor quality. The user can install a big (15 cm high glass chamber). In this case, the deposition rate is very low (0.033 nm/s) and the film will contain a high level of oxygen (up to 50%). This can be partially improved by installing a smaller chamber (6.5 cm high). However, this leads to a very bad uniformity disregardless of which current settings are used.  Depositing on small samples (around 1x1 cm) placed in the middle of the chamber will be the most acceptable procedure. Nevertheless, the elevated level of oxygen and rough and porous structure will limit the applicability.

In the beginning, a big glass chamber was installed in the instrument. It was quickly revealed that the deposition rate is very low, so only the highest current setting (80 mA) was used in the experiment.

The maximum time the user can select is 300s, so if longer depositions are needed the user has to redo it several times to achieve an expected thickness.

The samples with low aspect ratio Si trenches (500 nm deep, 200 nm wide with a pitch of 400 nm) were used to illustrate the deposition morphology (Instrument: [[Specific_Process_Knowledge/Characterization/SEM_Supra_1|SEM Supra 1]]). To make the result more visible it was deposited <b>twice</b> at the maximum allowed deposition time (2x300s). The SEM results are shown below.

The samples used for the XPS analysis were prepared in a big glass chamber with 80mA current and a deposition time of 300s.The pressure is measured to be below 0.1 mBar.

The analysis reveals that the top surface is a native oxide, while the thin layer itself is partially oxidized tungsten. The stoichiometry shows 50%W and 50% O. The stoichiometric tungsten oxide has a form of WO₃, so the layer itself contains metallic W around 33% mixed with up to 67% oxide. This is not a big surprise since this tool does not have shutter mechanisms or a long well controlled pumping system. The deposition rate is also very low due to the limited power. This also contributes to the elevated oxygen level in the film.

In the next part of the experiment, a small (6.5 cm high) chamber was installed in the instrument. The deposition rate is much higher in this case, so the different current settings (20-80 mA) were employed. The consequence of the higher deposition rate was a very bad uniformity, such as the center of the wafer getting a full thickness, while the edges of the wafer were not covered at all. This deposition mode is better suited for small samples/chips. The produced films revealed high roughness and pronounced porosity revealed by XRR.

<gallery caption="Small glass chamber setup." widths="500px" heights="500px" perrow="2">

<gallery caption="Small glass chamber setup. Flat films deposited on a Si wafers at different current settings." widths="300px" heights="300px" perrow="4">

<gallery caption="Small glass chamber setup. Flat films deposted on a Si wafers at different current settings." widths="300px" heights="300px" perrow="4">

<gallery caption="Small glass chamber setup. Deposition rate as a function of current." widths="500px" heights="400px" perrow="2">

Deposition of W in Sputter Coater 3 using a small glass (6.5cm high) glass chamber configuration is fast, but the resulting coatings are still expected to have a significant fraction of oxygen. Although XPS measurements were not performed on these films, the sheet resistance values were very high if compared with pure tungsten layers.  

Another issue is the pronounced roughness. The attempt to perform XRR on the film (which is expected to be in the range of 120-130nm) reveals a very clear rough layer profile, where all oscillations were canceled, and a smooth curve of reflected intensity was observed. These findings support the conclusion that produced films are porous and partially oxidized.