Quantum computing could revolutionise the way we perform computations that are beyond the reach of even the most powerful computing machines we have today. 

But we still don't have large scale error free quantum computers because the quantum bits used to encode the quantum information are still too susceptible to noise from the environment.

But we are working to solve that problem. Imagine I am the nucleus of an atom of phosphorus that has been implanted inside a silicon chip. 

So as a nucleus spin, I am really good at isolating myself from the noise in the environment that would otherwise disturb my quantum operations. This is what makes me an ideal building block for a quantum computer.

But of course, a single nucleus is not enough to build a quantum computer. We need to put many nuclei together and they need to be able to talk to each other.

So here's the problem. A phosphorus nucleus does such a good job at blocking out its surrounding that it can only communicate with other nuclei if they're really close together. 

I am saying the only experiments so far where nuclei have been able to talk to each other, whether once, whether we are really, really close together.

What we have achieved for the first time is to make two nuclear spins talk to each other even when they're very far away. To give you an idea of how far, imagine that the size of nucleus number one is my actual size. So nucleus number two would be a lot further away. 

The actual distance is more like. Sydney to Boston.

So how did we find a way to make these nuclei communicate over such large distances? One could say that we gave each nucleus a kind of electronic telephone.

Each nucleus was attached to its own electron. Now, although we generally think of electrons as extremely small subatomic particles in quantum physics, their influence can actually stretch very far via clouds of probability. 

The two phosphorous atoms were implanted in a silicon chip. With an ion beam, we can rotate the spin of the nuclei and the electrons by applying bursts of oscillating signals at some specific frequencies. In the past, we relied on having the nuclei very close together and coupled to one common electron.

This works very well for proof of principle demonstrations, but of course it's very difficult to scale to a large number of nuclei. 

Now, we showed for the first time that we can entangle two nuclei in a silicon chip that are not coupled to the same electron. This allows us to space the atoms much further apart than has ever been done before.

Each nucleus has its own electron and the probability clouds of the two electrons touch each other ever so slightly, but enough to ensure that the frequency at which one electron responds depends on the quantum state of both nuclei. 

So we prepared the nuclei in a quantum superposition state. We rotated one electron conditional on a specific orientation of the two nuclei, and as a result we ended up in what is called an entangled state, where each nucleus individually does not point anywhere, but it's correlated to the orientation of the other. 

Now that we have this method in place, we can envisage using intermediate couplers that extend the range of the interactions between the electrons even further.

This work is the key step that enables us to create an architecture that can scale up to the millions of cubits necessary to perform useful quantum computations. It really is quite amazing. We can finally start doing some quantum computation at a distance.