For the first time, researchers have synthesised a strange and unstable triangle-shaped molecule called triangulene, which physicists have been chasing for nearly 70 years.

Triangulene is similar to the 'wonder material' graphene in that it's only one-atom-thick. But instead of sheet of carbon atoms, triangulene is made up of six hexagonal carbon molecules joined along their edges to form a triangle - an unusual arrangement that leaves two unpaired electrons unable form a stable bond. No one has ever been able to synthesise the molecule until now.

The elusive molecule was created by a team of researchers from IBM, using a needle-like microscope tip to manipulate individual atoms into the desired format.

"Triangulene is the first molecule that we've made that chemists have tried hard, and failed, to make already," lead researcher Leo Gross from the IBM labs in Switzerland, told Philip Ball from Nature.

It's not the first time researchers have been able to synthesise unstable molecules that have been impossible to make conventionally, but triangulene is extra special - not only because of its unique structure, but also because it's predicted to have useful properties in electronics and quantum computers. 

Triangulene was first predicted back in 1950 by Czech scientist Erich Clar. He calculated that - theoretically at least - a triangle-shaped hydrocarbon could be made from six fused circular benzene molecules that would have an even number of atoms and electrons, but would leave two unpaired electrons due to its arrangement.

Clar tried and failed to make triangulene in the lab. It's an incredibly difficult task, because those two unpaired electrons don't like to remain unpaired, so immediately react with anything around them. 

Conventional synthesis techniques involve scientists reacting molecules together to build up larger structures. But for almost 70 years, researchers have struggled to create triangulene this way.

"As soon as you synthesise it, it will oxidise," said one of the researchers, Niko Pavliček.

The IBM team overcame this by using a different technique - instead of building up a structure molecule by molecule, they first created a larger precursor structure and then whittled it down.

The precursor had a couple of extra hydrogen atoms on it to make the molecule was stable. These hydrogen atoms were blasted off using an electron beam, leaving the unstable triangulene molecule.

They were even able to image the structure using a scanning probe microscope.

"To my knowledge, this is the first synthesis of unsubstituted triangulene," Takeji Takui from Osaka City University in Japan, who hass worked on synthesising triangulene, but wasn't involved with the IBM research, told Nature.

The new material is already showing some unique and unexpected properties.

As predicted, the team has shown that the two unpaired electrons in the triangulene molecule have aligned spins - which makes it magnetic on the molecular level, and means the structure could be useful for quantum computer and even creating spintronic devices.

But they also found that it also remained stable on a copper surface - for up to four days in one experiment. It was predicted that triangulene would have reacted with the metal, and the team is now trying to figure out why that didn't happen.

"We were surprised that no bond formed for triangulene on copper," Gross said in a press release. "We think that is because triangulene is a pi-radical, which means it's unpaired electrons are delocalised."

There's still a lot to learn about triangulene, and as with the demonstration of any new material, other teams will now have the chance to poke holes in this discovery as they attempt to independently verify whether what the IBM researchers have created is really the elusive triangle-shaped molecule.

And while more research into this new structure is needed, the new chemical synthesise technique developed by the team should come in handy for making other elusive structures.

It's not something that will be suitable for all types of chemical synthesis - this is a slow and expensive process that probably won't be useful for many materials.

But when we're talking about something as lucrative as the future of quantum computing, then IBM definitely has a vested interested in getting new materials off the ground.

And that's a good thing for the rest of the scientific community, which will also benefit from their finds.

The research has been published in Nature Nanotechnology.