Wednesday 27 September 2017

Behind the scenes Part 1: Giant fullerene formation through thermal treatment of fullerene soot


We recently had a paper accepted where we showed that heating up small cages of carbon (fullerenes) led to them fusing into giant fullerenes. The smaller fullerenes C60 and C70 usually get all the attention because they can be easily extracted using solvents (e.g. toluene). These small cages have found applications in molecule-based solar cells (organic solar cells), superconductors (if you put some potassium atoms between the cages) and are good antioxidants. The header for this blog features C60 buckminsterfullerene the most famous carbon cage (and also the most symmetric molecule known) which is one of the smallest stable cages. For an overview of the main results of the paper I have embedded a 5 minute long audioslides presentation below. However, most of the results require a bit more of an explanation to really understand the significance of the results for the non-expert. So I thought I would give you a behind the scenes look at putting this paper together through a blog series. In the first instalment, I discuss what other people have done on fullerene formation and mechanisms so you can see the underlying motivations for the paper.

Finding out giant fullerenes are more stable than C60

It all started after I was reading about the formation of carbon fullerenes and trying to understand the high abundance of C60 in soot containing fullerenes. Why the high abundance of C60 is puzzling is the way in which fullerenes are usually made, an electrical arc is produced between two carbon rods which produces a carbon plasma at > 3000 degrees C. At these temperatures, you wonder how such a symmetric and ordered molecule could form. Previous mechanisms assumed a bottom-up approach where small carbon molecules grew and closed to form C60 cage molecules. This was challenged by reactive molecular dynamic computer simulations in 2006 that suggested that the carbon structures close into giant fullerene (cages with a greater number of carbon atoms than sixty) and then in order to cool down they eject carbon and shrink to form the magic number fullerenes C60 and C70. This was shown to be at least feasible in 2007 when a giant fullerene was observed, in an electron microscope (more on that below), to shrink when it was heated.


Curl et al. in 2008 reiterated that small fullerene cages are less stable (more negative means more stable) than larger cages (see the figure below). This makes sense as flat graphite is the most stable form of carbon and any attempt to curve the structure will make the structure less stable (increase it's energy/more positive). This might seem straightforward however the common explanation of why C60 is formed (in the bottom up mechanism) was that C60 is a very stable molecule. 
Modified from Curl et al. 2008
Curl et al. also suggested that carbon cages could exchange carbon. This would allow C60 and C70 to be produced in greater abundance as they are significantly lower in energy than their neighbouring carbon cages (due to their high symmetry) and with fullerenes continually exchanging carbon they could get stuck at C60 or C70 as it would be harder for carbon to be exchanged from these highly stable symmetric cages i.e. there are no weak points. Think of a fullerene as rolling into stable notches which are hard to get out of. Further reactive forcefield simulations in 2011 suggested carbon dimers C2 could be ingested into fullerenes and this was experimentally shown in 2012 using mass spectrometry where fullerenes with metals in them could ingest C2 without losing their cargo.

I ran some reactive forcefield simulations starting in 2011 wanting to observe the carbon capture and ejection considering that two interacting cages would exchange C2 from the smaller to the larger cage, but ended up finding that these structures really wanted to fuse together and coalescence into bigger giant fullerenes. Below is a video of a computer simulation I ran showing two fullerenes fusing.


This coalescence has also been observed before with C60 coalescing together inside carbon nanotubes and many different simulations. I, therefore, started to think that perhaps fullerenes actually would rather become larger if they are not in a higher energy chaotic plasma. But I wondered if I could experimentally observe coalescence of fullerenes to form isolated cages.

Tune in next week for part two where I show you an amazing instrument that can weigh individual molecules and show the higher fullerenes in fullerene soot.

Link to all the blog posts on the paper
Part 1 - Energetically fullerenes are unstable and want to coalescence (this page)
Part 2 - Weighing fullerenes as they grow
Part 3 - Viewing fullerenes coalescing
Part 4 - Simulating how large we can grow giant fullerenes