The 2nd Area Selective Deposition workshop (ASD 2017, link) is approaching quickly now. The workshop will take place in Eindhoven on April 21 and my colleague Adrie Mackus is the main organizer. With the upcoming workshop in mind, I thought it would also be good to briefly reflect on the 1st ASD workshop at imec, Leuven a year ago (April 15, 2016).
The 1st ASD workshop last year was a big success. With about 140 people participating, it was well-attended with people coming from all over the world, even despite the fact that it only lasted a single day. The large attendance by industry (about 40% of the attendees) was especially remarkable and it illustrates how important the topic of area-selective deposition is. This is also confirmed by the registrations for this year’s workshop. At the time of writing, the number of registrations is 151 with even more than 50% of the people registered coming from industry. See the figure below for the geographical distribution of the participants as of now (the registration closes on April 12!). It is again a one-day workshop but now with a welcome reception and lab tour on the evening before. The program features inviteds by academia and industry and also some contributed presentations. The invited presentations from industry are somewhat shorter than the other inviteds (20 min. instead of 40 min.). This was done to lower the threshold for accepting the invitation as speakers from industry typically cannot reveal too many details about they are working on. Moreover, in this way simple more presentations could be fit in the program as well.
Registrations for ASD 2017 per country separating out those from academia and industry (status April 10, 2017)
Last year almost all presentations came from academia, the only exceptions being contributions from FEI Company (now part of Thermo Fisher Scientific) and Intel. There was also a panel discussion that involved Tokyo Electron, LAM Research and Applied Materials as industrial members. To me, this discussion made clear that the semiconductor equipment manufacturers really feel the need to work on area-selective deposition but that it is not yet very clear what the first applications will be. Perhaps this has changed this year, we might find out in about 10 days from now. While waiting for ASD 2017, it might be interesting to pay attention to my presentation at ASD 2016 and give some updates.
The title of my presentation of last year was “ALD-enabled nanopatterning: area-selective ALD by area activation” (PDF). On the basis of this title, there are already a few comments to make. First of all, it dealt with ALD as most of the presentations during the workshop. Note however that CVD is also a very interesting method for area-selective deposition. Reports on selective CVD go already back more than 3 decades and selective deposition of tungsten has been adopted in industry. With so much focus on ALD, we should not forget about CVD and in this respect it is good that John Abelson (University of Illinois) and Son Van Nguyen (IBM) will present about area-selective CVD at ASD 2017. Secondly, the presentation was about nanopatterning which basically goes a step beyond area-selective ALD as I will explain below. Thirdly, it dealt with one particular method of generating nanopatterns with area-selective ALD, a method that we have been mostly focusing on in our work at the Eindhoven University of Technology. I will also explain this below.
In the beginning of my talk, I first addressed what area-selective deposition actually is, see the slide below. Suppose that you have a surface with two materials, material A and material B. Deposition is area-selective when material is grown only on one of these materials, e.g., only on material A and not on material B. So in principle for a given surface, a deposition process is selective or it is not selective, at least in this ideal case. Personally, I don’t understand the term “inherent selectivity” that well. This term is sometimes used (also during ASD 2016) to express that no additional step (such as blocking of part of the surface) is required to deposit selectively. To me, inherent selectivity just means that the process is selective.
Slide from ASD 2016 – Definition of area-selective deposition.
After giving some examples of area-selective ALD, I addressed the question how we can do nanopatterning through area-selective ALD. In this case, the starting point is a surface on which no patterns have been defined yet, so a blank surface. Consider, for example, a surface with only material B, see the figure below.
Slide from ASD 2016 – The central question: How to do nanopatterning with area-selective ALD?
One method to come to a nanopattern is to block – or deactivate – part the surface, e.g., by self-assembled monolayers (SAMS) or by a photoresist film. Subsequently, one can do area-selective ALD to grow a film only on the non-blocked part of the surface. We can label this as area-selective ALD by area-deactivation. This is the terminology which we introduced in our review paper about nanopatterning that appeared in 2014 (“The use of atomic layer deposition in advanced nanopatterning”). One needs to realize however that this method requires a separate patterning step to get to the point where the surface is partially de-activated. In many cases, this is a lithography step. Furthermore, it is important to note that the ALD process should be really blocked by the species de-activating part of the surface. This means that no ALD film growth should take place on, e.g., the SAMS or on the photoresist. If growth would take place on these materials, the ALD process would not be really selective. When working with a photoresist, the process would basically just be a lift-off, at least when the ALD film covered photoresist is removed from the surface by a wet chemical step or the like. These two processes, lift-off and area-selective ALD by area-activation, are schematically illustrated in the slide below. This slide shows a key figure from the aforementioned review paper that came back several times during the workshop.
Slide from ASD 2016 – Key figure from the review paper “The use of atomic layer deposition in advanced nanopatterning”
At this stage, I would like to take the opportunity to mention that our paper about area-selective ALD of Pt using photosensitive polyimide appeared at the end of last year (“Area-selective atomic layer deposition of platinum using photosensitive polyimide” ). This paper was also highlighted in my ASD 2016 presentation. As the title of the paper indicates, no ALD of Pt took place on the polyimide that is used for area-deactivation.
Next, I addressed the method of area-selective ALD by area-activation, also indicated in the figure from the review paper shown above. As mentioned earlier, this is the method that has been the focus of our work at the Eindhoven University of Technology in recent years. It is also where the title of my ASD 2016 talk referred to, so it formed the main part of the presentation. In this method, the starting surface is first locally activated and subsequently an area-selective ALD process is applied which only takes place on the activated area. The local activation defines the nanopattern and as indicated by the two examples shown in the presentation (see below), this local activation does not have to involve lithography. Therefore this method does two things, it defines the nanopattern on a blank surface and it deposits the material area-selectively. It is therefore a true nanopatterning method. As indicated before, it is important to distinguish area-selective ALD and nanopatterning. Nanopatterning goes beyond area-selective ALD!
Finally, I want to briefly comment on the two approaches for area-selective ALD that I highlighted in the talk. The first one was the approach in which we combined electron-beam induced deposition (EBID) of a seed layer with area-selective ALD. We demonstrated this approach extensively for Pt, see the figure below. We first published on it in 2010 (“Local deposition of high-purity Pt nanostructures by combining electron beam induced deposition and atomic layer deposition” ) and recently two papers appeared in which the method was demonstrated at the (quasi-)device level: for the contacting of carbon nanotube field-effect transistors (“Resist-free fabricated carbon nanotube field-effect transistors with high-quality atomic-layer-deposited platinum contacts”) and for the contacting of graphene (“Graphene devices with bottom-up contacts by area-selective atomic layer deposition”). But let’s come back to this approach in a separate blog post soon as there are quite a few more things to say about it.
Slide from ASD 2016 – Schematic of the area-selective ALD process of Pt on oxide materials. First (1) nanoscale patterns are defined by depositing a sub-nanometer-thick seed layer of Pt by electron beam-induced deposition. Second (2) the Pt is deposited selectively on the seed layer in a building step. The ALD process consists of two alternating half reactions: MeCpPtMe3 precursor dosing in pulse A and O2 dosing in pulse B.
Slide from ASD 2016 – Carbon nanotube and graphene field-effect transistors fabricated using a bottom-up and resist-free method employing Pt seed layer preparation by electron beam-induced deposition followed by a thickening step by area-selective ALD of Pt.
The second approach that I highlighted was a completely new development and a paper describing it only appeared recently (“Area-Selective Atomic Layer Deposition of In2O3:H using a µ-PlasmaPrinter for local activation” ). This method involves the local activation of a hydrogen-terminated silicon-based surface (e.g., amorphous silicon or silicon nitride) by a micro-plasma printer and the subsequent area-selective ALD of In2O3 on this activated area. The patterns generated by the micro-plasma printer have dimensions which are at least a few hundred micrometers wide so this method would not immediately qualify as nanopatterning (rather micropatterning). However there is no reason why this method could not be scaled down to the nanoscale when using a different activation method.
Slide from ASD 2016 – Schematic of the area-selective ALD process of In2O3:H on H-terminated silicon materials. First (1) microscale patterns are defined by activating the surface with a μ-plasma operated in air or O2. Second (2) the In2O3:H is deposited selectively on the activated areas in a building step. The ALD process consists of two alternating half reactions: InCp precursor dosing in pulse A and a mixture of O2 and H2O dosing in pulse B.
To my opinion, area-selective ALD by area-activation can be considered as the ultimate dream of pure bottom-up processing because the surface on which no film needs to be deposited remains untouched. Moreover, as emphasized before, the two approaches illustrated also work on a blank starting surface and therefore they are also patterning methods. To express this, we have also adopted the term direct-write ALD for this kind of methods.
You can download the presentation (with updated references) here.
“The use of atomic layer deposition in advanced nanopatterning” by A.J.M. Mackus, A.A. Bol and W.M.M. Kessels, Nanoscale 6, 10941 (2014).
“Area-selective atomic layer deposition of platinum using photosensitive polyimide” by R.H. J. Vervuurt, A. Sharma, Y. Jiao, W.M.M. Kessels, and A.A. Bol, Nanotechnology 27, 405302 (2016).
“Local deposition of high-purity Pt nanostructures by combining electron beam induced deposition and atomic layer deposition” by A.J.M. Mackus, J.J.L. Mulders, M.C.M. van de Sanden, and W.M.M. Kessels, J. Appl. Phys. 107, 116102 (2010).
“Resist-free fabricated carbon nanotube field-effect transistors with high-quality atomic-layer-deposited platinum contacts” by A.J.M. Mackus, N.F.W. Thissen, J.J.L. Mulders, P.H.F. Trompenaars, Z. Chen, W.M.M. Kessels, A.A. Bol, Appl. Phys. Lett. 110, 013101 (2017).
“Graphene devices with bottom-up contacts by area-selective atomic layer deposition” by N.F.W. Thissen, R.H.J. Vervuurt, A.J.M. Mackus, J.L.L. Mulders, J-W. Weber, W.M.M Kessels and A.A. Bol, 2D Mater. 4, 025046 (2017).
“Area-Selective Atomic Layer Deposition of In2O3:H using a µ-PlasmaPrinter for local activation” by A. Mameli, Y. Kuang, M. Aghaee, C. K. Ande, B. Karasulu, M. Creatore, A. J. M. Mackus, W.M.M. Kessels and F. Roozeboom, Chem. Mater. 29, 921 (2017).