Quantum Simulation
One of the problems, todays computers are still struggling with is simulation of quantum systems. Simulating molecules, chemical reactions or processes is so complex that even supercomputers can only simulate them using approximations.
Why are we actually interested in simulation? Simulations of quantum systems can help examine properties, structure, and behaviour of something like a molecule. This way they can help us, for example, develop new and more efficient materials for batteries leading to more efficient mobility. Understand the chemical processes involved in the process of fertilizer production could support finding a process that uses way less energy. And for drug research, scientists would be able to investigate these drugs without actually creating them in a lab synthetically.
And why is simulating something like a molecule so hard to do? Molecules are made up of atoms like Hydrogen (H), Oxygen (O), Nitrogen (N) and so on. Those atoms are themselves made up of protons, neutrons and electrons. In order to simulate a molecule, we need to think about every single electron, their position, their interaction among each other and any forces acting on the molecule. Storing all of this requires a huge amount of computer memory. How much, you can find out by exploring the different molecules below! And just as memory usage increases, so does the number of operations required to simulate the system over time.
Water (H2O) is vital for all forms of life and a rather simple molecule. Something you could simulate with \(16384\) bits or 2 Kilobyte.
Ethanol (C2H6O) is an alcohol and it's what gives our drinks their effect. \(4,398,046,511,104 \) bits or 512 Gigabyte will bring you there.
Benzene (C2H6O) is a natural constituent of crude oil and an important intermediate for the petrochemical industry. \(10^{21}\) bits or 536870912 Terabyte (that's a lot of harddrives) are something even supercomputers cannot deliver anymore.
Glutamine (C5H10N2O3) is an amino acid and important to regulate water retention in cells. During great physical stress it causes an increase in cell volume. \(10^{36}\) bits is what's needed to simulate this amino acid.
Caffeine (C8H10N4O2) is the reason we get active every morning. It is what makes our coffee actual coffee. As a molecule Caffeine contains 24 different atoms, which require \(10^{48}\) bits to simulate classically.
Penicillin (C16H18N2O4S) is used as an antibiotics for infections. It does not only look complex, it would also require \(10^{86}\) bits to simulate classicaly.
Explore the different molecules to continue ...
Simulating quantum systems using a quantum computer
Nature isn't classical, dammit, and if you want to make a simulation of nature, you'd better make it quantum mechanical.
In this famous quote, Richard Feynman (credited with pioneering quantum computing) sketches the reason why it is so hard to simulate nature classically: The effects of nature are determined by quantum mechanics and not classical mechanics. As a reason, the number of bits needed and also the number of operations increases exponentially. The only way to avoid this classicaly would be to use good approximations, which are not always available or face severe limitations.
Thus, quantum computing seems like a natural fit for simulating molecules and their reactions. Quantum computing not only allows us to do simulations much faster but also move the boundary of what's possible. While it is no problem to simulate a molecule like water, the effort and memory needed to simulate a molecule like benzene with its 12 atoms already skyrockets well beyond the limits of what is possible classically. With a quantum computer simulating benzene could be done using 72 qubits. And even simulating penicillin, which would require more bits than there are atoms in the observable universe, would be possible with 286 qubits.
Click here to show the number of qubits required for simulating the different molecules in the above plot.
-
A
Storing the information about electrons, their positions and interactions requires huge amounts of classical memory.
-
B
As each atom needs its own processor to be simulated, we quickly run out of processors.
-
C
Good approximations to describe quantum systems are often not available or face severe limitations.
Exactly! Storing this kind of information requires huge amounts of classical memory and good approximations to describe quantum systems are often not available of face limitations.
Not quite! Actually, storing this kind of information requires huge amounts of classical memory and good approximations to describe quantum systems are often not available of face limitations.