Caleb Hustedt and Drs. Robert Davis and Richard Vanfleet, Physics and Astronomy
Graphene is an exciting material that stands to have a large impact on the scientific community. Unfortunately, graphene cannot be fully utilized due to its small scale and time consuming production. Graphene grown by chemical vapor deposition solves these issues but comes with a cost of decreased mechanical and electrical properties due to defects. However, the exact properties of CVD graphene are not well quantified. In order to measure the qualities of CVD graphene it must be suspended. Carbon infiltrated carbon nanotube forests were fabricated with holes 2-20um in size. Imaging showed CVD graphene was suspended over up to 50 percent of the holes.
Graphene is a monolayer sheet of carbon one atom thick. This two dimensional material is tightly packed together in a honeycomb lattice. Graphene can be wrapped up into soccer ball like buckyballs, rolled up into nanotubes, or stacked to become graphite. Some familiar forms of graphite, multi-layer graphene, are charcoal or the material in pencils. Graphene was first theorized in 1962 by Hanns-Peter Boehm. At the time graphene was considered to be a purely academic material. Even though know as a part of three-dimensional materials graphene was not presumed to exist on its own. Graphene was assumed to be unstable as a single sheet and would revert back to other forms such as nanotubes or buckyballs. It was first isolated by Geim and Novoselov in 2004, for which they won the noble prize.
Graphene has been characterized and found to have many interesting and exploitable properties. Among these is the existence of a zero band gap making graphene a semi-metal, “the only known material in which the electronic band structure changes significantly via electric field effect”. Graphene also exhibits the highest tensile strength of any known material; 200 times stronger than steel with a tensile strength of 130GPa(19,000,000psi). Silicon is widely used in electronics because of its high mobility with
The results of the carbon nanotube forests depended on several variables. If the lithography was not done well enough the holes would not be the right size or bleed into each other creating a large open area over which the graphene would not be able to be suspended. The etching process had to be done for a short enough time in order to ensure that the lithography pattern transferred to the iron and didn’t etch all the iron or resist away. Having the holes bleed together created opening too large to suspend a graphene sheet and rendered the sample useless.
Flowing argon while infiltrating allowed for the forests to release on their own and not crack into pieces on release. The holes in the forests theorized to be 1-10um ended up to be approximately twice as big resulting in 2-20um holes.
The graphene had to be handled very delicately to ensure successful transfers to the carbon nanotube forest supports. Being as graphene is only a single atom thick it is very easy broken or torn. Using a clean glass slide with containers mostly full of DI water was integral in the transfer from copper etchant to nanotube forest support. Unfortunately the majority of the holes did not have graphene suspended over them. Drying the membranes with a critical point dryer might have resulted in a higher percentage of suspended graphene membranes. The size of hole appeared to have no correlation to the chance of successful transfer.
However, I found that if the forests were plasma cleaned that the number of holes that were covered with graphene significantly increased. The transfer process proved to not cover the entire sample in graphene but the locations that were covered showed almost 100 percent success of having suspended graphene over the holes.
As can be seen in the above image, the copper grain structure can be seen in the suspended graphene. This shows that all the resist was indeed removed and that there is only the single graphene layer suspended over the hole. The grain structure can be seen due to different crystal orientations in the copper substrate during growth resulting in different orientations in the graphene. These different orientations of graphene have varied electrical properties and therefore interact differently with the electron beam and show contrast in the image.
In this study I fabricated a new technique for transferring thin graphene membranes. Carbon nanotube forests were used as a substrate for transferring and suspending graphene. Electron beam lithography was used to pattern the catalyst for nanotube growth. The carbon nanotubes were grown and infiltrated with carbon to make them strong and durable. The graphene was grown by chemical vapor deposition and transferred using a sacrificial polymer layer. The result was a new way to suspend graphene that was strong and able to have opening through an entire sample. Characterization was performed using optical and scanning electron microscopes.
Graphene was successfully transferred to this new substrate and suspended over holes ranging from 2-20um. The amount of holes that were covered was less than anticipated with coverage of around 14 percent. Samples that were plasma treated had a higher coverage percentage, approximately 50 percent.
These results show that this technique is a viable way to suspend graphene membranes. This technique allows for new studies of CVD graphene that have been heretofore unable to be performed. Pressure testing of CVD graphene is one such option that is available using this technique that could add important information about the qualities of these large-scale graphene membranes.