The molecule on the right is triangulene, and it’s an odd one. You’d think at first that you could fill everything in with alternating double bonds, like a small piece of graphite or a bigger relative of anthracene, but when you try, you find that the geometry won’t let you. You’re either going to end up with a “Texas carbon” (five bonds), or you’re going to have electrons left that aren’t involved in a bond at all (as shown). The molecule has been kicking around for decades on fume sashes and chalkboards/whiteboards, but no one has ever been able to synthesize it. It’s theoretically stable (for some values of “stable”) but should be extremely reactive.
So non-chemical synthesis has stepped in. In a new paper, a team at IBM’s Zürich facility has used their expertise with the scanning tunneling microscope to make single molecules of triangulene by the most direct route imaginable. They start from the dihydro form (prepared at the University of Warwick), with CH2 groups on the side, and then step in on an individual molecule with an STM/AFM tip to rip off two hydrogens like picking apples off a tree branch. The molecule that’s left sitting on the surface is indeed triangulene – it’s perfectly flat, and its EPR behavior is what you would expect from the triplet diradical. As long as you keep it at low temperature, it seems stable – one of the molecules they made hung around for days on a copper surface.
Press coverage of this work has mentioned that triangulene might be useful in quantum computing, and while that’s not wrong, I wouldn’t, uh, spin it that way. This is a molecule that we’ve known for a long time could exist, but no chemist had ever seen it or been able to make it. Now we can reach in and tug on individual atoms, though, and that does the trick – just the thought of direct mechanical synthesis being the way to make an elusive species like this is enough for me.