To close out 2018, we exhibited at the Fall Materials Research Society conference. As we usually do during the show, we took time to attend technical talks. Knowing the course that research is taking helps us to best meet the needs of researchers. Besides, the discoveries themselves are fascinating to take in! We’re reviewing what we learned and building on that knowledge as we kick off 2019.
They say it’s all in the delivery—and this is becoming truer of combinatorial deposition. The relative flux of the sputtered components is not the only determinant of the properties and performance of a combinatorial candidate material. Two of the talks we attended brought this point home:
- Cooling rate of high entropy alloys (HEA). Experiments in HEA often seek to determine which stoichiometries will lead to amorphous structures as opposed to the common cubic crystalline forms. The speaker cautioned, however, that the structure obtained depends also on the cooling rate of the thin film after the deposition is complete. In general, it is known that post-deposition thermal treatments will affect the crystallinity. But, in a gradient deposited film with whole-substrate heating there is no opportunity to vary deposition temperatures with control, post-deposition annealing or cooling rates.
- Nitrogen concentration in combinatorial deposition of high entropy nitrides. In this case, high hardness and corrosion resistance were the application goals. The speaker detailed that she found three different structural regimes, depending on the nitrogen concentration during the reactive growth. However, gas concentration cannot be a parameter in a single run during a gradient combinatorial deposition. A greater number of runs must be executed at the cost of more time, which creates more opportunities for uncontrolled variations to be introduced.
Creating a spread of compositions through gradient deposition is useful in many cases, which is why that ability is offered in our tools. However, if other processing parameters beyond stoichiometry of the sputtered materials are relevant to the exploration, then discrete test pad combinatorial deposition offers much greater flexibility. With a system designed specifically for maximum parameter control, sample-by-sample variations in one run could include:
- Growth temperature
- Annealing profile
- Growth rate
- Reactive chemistry conditions
- Sample bias (DC/RF)
- Substrate pre-cleaning
- Film thickness
- Individual thickness of multi-layer components
Because the "combi" field is becoming more about the delivery, our engineers at PVD create high-quality instrumentation that gives researchers full control over the parameters that are relevant to the space to be explored.
For example, we provide an option for laser-based, local substrate heating that acts only at the deposition mask position. A separate heating profile can be applied to each combinatorial test pad. We observed in one experiment how the surface morphology of HfO2 films depended on the film growth temperature – an experiment that can be facilitated with great precision and time savings by local heating. Our recent article in the SVC Bulletin shares more details on this.
With complete process control, materials science researchers can quickly navigate through a multi-dimensional process space to discover novel materials in a cost- and time-efficient manner. Complete deposition process control also supports automated materials searching, a recognized important next step in the industry, to determine what materials are necessary for application needs and how to make the materials.
For more on current combinatorial deposition methods, download our white paper that discusses the advantages and disadvantages of each method and introduces five developments that improve upon these methods: