The Power of Quantum Simulations
Since the age of classical computing, simulations have been an increasingly important tool when doing science. Simulations help to discover new chemicals, create new materials and deepen our understanding about the underlying laws of reality. They give humans the ability to test things without needing physical access to the materials being tested.
But some simulations are messy. This is an unfortunate truth, especially in the fields of chemistry and material science.
For example, methods such as Monte Carlo simulations and Density Functional Theory (DFT) are used in chemistry in order to simulate the behavior of molecules. The ways in which these simulations make predictions is a whole topic of its own, but one important thing to note is that both involve significant approximations.
In other words, these simulations do not describe what would actually happen in real life with these chemicals but rather, an idealized version where certain properties of the molecules are ignored. Experiments that utilize these simulation methods usually involve results in which the difference between the simulation and real life, as a result of these ignored properties, are negligible. Because of this, the methods are useful in some situations but they still leave a lot of work left to be done by the chemists in the lab.
What are these properties and why must they be ignored when using these simulations?
At the scale of individual molecules, there are a lot of variables to look at when trying to perfectly simulate them. There are the positions and velocities of all the individual atoms in the molecule which create more variations as the molecule gets bigger in size and complexity. In addition, there are also the bonds between the atoms and the interaction between their electrons.
It’s variables like this which makes accurately simulating a simple 24-atom caffeine molecule on today’s current computers a process that would take longer than the age of the universe. But this is only a problem with current computers.
The Power Of Quantum
Over the past three decades, a new type of computer has been gaining steam in hopes of being able to solve problems such as the extremely long simulation time of molecules. This is where the field of Quantum Computing comes from.
Quantum computers are able to answer questions that would be impossible for classical computers in a human life time through the quantum phenomenon that appear at the smallest of scales, such as Superposition, Entanglement and Interference.
Superposition
Classical computers run on bits, the smallest unit of information, which equate to either a 1 or a 0. Each bit equals one unit of information. For example, with two bits there are two units of information, the state of the first bit, and the state of the second bit.
Many bits can work together to store memory. For example, a byte consists of eight bits and is big enough to store the information required to produce a character such as an “s” or an “e.”
Instead of bits, quantum computers use qubits which utilize the quantum phenomenon of superposition. Superposition is the ability for a particle to exists in two states at the same time. So while a bit can be either a 1 or a 0, a qubit can be a 1, 0 or both at the same time. In fact, a qubit can be represented as the probabilities of each combination occurring (00, 01, 10 and 11) or the qubits’ probability function.
By using these probabilities as information, 2 qubits contain four pieces of information while 2 bits only carry two. This relationship scales by 2^n, meaning for every n qubits, there are 2^n pieces of information possible. Its easy to see how, at large quantities, qubits can hold significantly more information then bits. This is part of the reason why a quantum computer would be much more efficient at simulating a caffeine molecule.
The ability to hold two states at the same time also allows multiple processes to happen at the same time. With each added qubit, the processing power increases exponentially allowing tasks to be processed much faster than in a classical computer.
Entanglement
Entanglement is another weird property that appears at the scale of elementary particles. At this scale, two particles can become entangled with each other meaning they become correlated. For example by knowing the spin of one electron, the spin of another entangled electron is automatically known. If the measured electron is spin up, then that means its entangled partner is spin down.
The weird part about this is that this works no matter the distance. These particles can be on opposite ends of the universe and still must follow the rule of entanglement.
Entanglement allows the qubits in quantum computers to all correlate together and work as one system. This means something that happens in one qubit effects the whole system of qubits which allows for faster communication when compared to classical computers.
Interference
Interference is the ability for the probability functions of the qubits to combine in the same way that waves do. Constructive interference, where two peaks combine, create a probability function with a peak equaling the height of the two peaks added together. Destructive interference, where a peak and a trough combine, creates a canceling out effect resulting in a straight line on the probability function.
Using this phenomena, quantum computers can combine qubits’ probability functions in a way that makes the highest peak on the resulting probability function the answer they are looking for. In other words, imagine all the qubits as waves. Then add all the waves together using interference. The highest peak on the resulting wave would be the answer that the computer spits out.
By manipulating the process of interference, researchers can increase the probability that the highest peak has the answer they are looking for, rather than some arbitrary result.
Types of Qubits
Not only are qubits beneficial for their quantum properties, but they also rely on relatively small and simple components. All that's needed for something to work as a qubit is the ability to be in one state or another, and the ability to be in both states at the same time. While the requirements are simple, the number of objects that can achieve the second one consistently is small. Types of qubits include:
- Electrons — The spin of an electron could be used as a 0 (Spin up), or 1 (Spin down).
- Photons — There are three ways in which a photon can be used as a qubit. You could measure its path as 0 (Goes up), or 1 (Goes down), measure the time it arrives at a certain point as 0 (Arrives early), or 1 (Arrives late) or measure its polarization as 0 (Horizontal), or 1 (Vertical).
- Trapped Ions — Atoms contain electrons that can be in their normal orbital around the nucleus or, through an increase in energy, be in a higher orbital putting it at a higher energy state. These states can represent a 0 (Lower energy state), or 1 (Higher energy state).
- Superconducting Circuits — Circuits can flow in either direction, with each direction being able to represent a 0 (Clockwise), or 1 (Counterclockwise).
What Has Been Simulated?
If we have this awesome way to create more accurate simulations, then what has been done so far?
Well the first big step came when Google simulated a Hydrogen atom and predicted its ground state energy, using superconducting circuits as qubits back in 2016. Things got more complex in 2017 with IBM’s simulation of Lithium hydride and even more so in 2019 when, startup company, ionQ simulated a water molecule, using trapped ion qubits.
Then in 2020 things got more exciting when Google once again hit a new milestone and simulated the first ever chemical reaction on a quantum computer — the reconfiguration of the hydrogens on a diazene molecule, which consists of two hydrogen, and two nitrogen atoms.
Today, these companies are working to create better quantum computers with more qubits in order to simulate more complex molecules and reactions.
While these results might seem cool, they are quite simple. In fact the way these simulations were verified as accurate was by testing them on classical computers. What makes these simulations so impressive and why is it so difficult to simulate more complex molecules on our current quantum computers?
The Obstacles of Quantum Computing
While the advantages of using qubits instead of bits for molecular simulations are huge, the actual ability to do so is very limited right now. This is due to the extremely fragile nature of qubits.
Another quantum property is decoherence, which is one of the biggest obstacle today in creating quantum computers with more qubits. Decoherence occurs when a particles’ probability function collapses and the particle takes on one state. In other words, the particle is no longer in superposition. This prevents the key advantage of qubits from being realized, which is its ability to be multiple things at once.
What causes decoherence? Pretty much everything. Interaction with anything from the outside environment will cause a qubit to decohere. Because of entanglement, this problem becomes worse cause now if any one qubit decoheres, the whole system decoheres since all the qubits are entangled together.
Decoherence creates errors in the calculations that the qubits execute, making any of the information received through them useless. The possibility for errors increase as the amount of qubits increase because there are more chances for part of the system to decohere.
These issues can be diminished by cooling down the environment in which the qubits operate. Quantum computers operate at extremely cold temperatures, reaching just a fraction of a degree above absolute zero. That’s colder then space! Since heat causes particles to move, its crucial that it is limited as much as possible to prevent decoherence from unwanted movement around the qubits.
Another method, is Quantum Error Correction, which essentially uses more qubits to prevent the influence any errors have on the system from significantly effecting the final answer. While this method has been instrumental in allowing quantum computing to become more practice than theory, it has also made progress harder as more qubits are now needed to do the same simulations that fewer, error-free qubits would need.
This is why todays simulations are so simple, because current quantum computers only have around 50 to 100 qubits, with the best quantum computer having 256. Its analogous to being slightly past the vacuum tube stage of classical computing. While the numbers are impressive by todays standards, they aren't yet enough to start making useful simulations.
What is Needed for Useful Simulations?
If simulating molecules at a level useful enough for practical purposes is the goal, how much qubits do we need in order to do so? Well the question depends a lot on the application.
In order for a chemical simulation to be accurate, all the electrons in the molecules have to be accounted for. This makes the amount of qubits reliant on the amount of electrons in a molecule.
For example a Caffeine molecule contains 132 electrons total, so its possible to need somewhere between 100 and 200 qubits in order to simulate the molecule perfectly. Its clear how this number can grow the more molecules there are.
However the amount of qubits also depends on what specifically is being simulated and whether or not all the electrons are important for the accuracy needed. For example, if you are trying to test a specific bond between Caffeine and another molecule, you might only need to accurately simulate the electrons in the bonding orbitals making the necessary amount of qubits lower.
What will really makes a quantum simulation useful is when it can answer questions in a way that classical computers can’t. This is why the current molecular simulations aren't useful in a practical sense, cause they can all be done on classical computers. Despite this fact, quantum computers can still be quite useful, even if they can’t do the entire simulation on their own.
Classical and Quantum Combinations?
It can be easy, when talking about quantum computers, to spend a lot of time comparing their abilities to classical computers but this view ignores the progress that can be made through utilizing both quantum and classical simulation techniques.
For example, if quantum simulations are used for medicine and drug discovery, they are going to involve proteins and other biological molecules which consists of thousands of atoms. Instead of waiting to reap the benefits of quantum computers until they can simulate molecules of that complexity, combining the strengths of smaller quantum computers and classical computers can make them useful earlier.
Since the actual drugs used in these types of experiments consists of around 50 to 100 atoms, you could have quantum computers simulate the drug, along with its’ binding sight on the protein, and then let classical simulations simulate the other thousands of atoms in the protein.
Other applications, such as trying to discover a new material with a specific property, could use a classical computer to narrow down a large amount of combinations, and then use quantum computers to simulate the remaining ones. It may not be able to narrow the pool down to the exact combination that will be best for the experiment but it could help decrease the amount in the pool allowing the chemist to have less physical work to do in the lab.
The Benefits of More Accurate Molecular Simulations
Being able to better simulate molecules and chemical reactions would greatly improve progress in material science, medicine, electronics, sustainability, etc. Even theoretical physics could potentially be aided by quantum simulations.
Medicine
Medicine is obviously a big field that would benefit from quantum simulations, especially in the form of drug discovery. Testing drugs requires a lot of trial and error and tests on humans which takes up a lot of time, money and resources, along with creating risks for the people being tested. Quantum simulations could allow us to do more of the testing virtually which would decrease all of these factors allowing more life saving medicines to be created quicker and safer.
Not only can these simulations help with drug discovery but it can also help increase society’s understanding of diseases at a molecular level. Diseases such as Dementia involve the folding of proteins. Simulating these proteins accurately would add a lot more insight when trying to discover new ways of treating the disease.
Electronics
One of the exciting applications of Quantum Simulations is that of superconductors. Superconductors are conductors that have no resistance allowing zero energy loss from the flow of electrons through the circuit. The problem is that they only work at extremely cold temperatures, near absolute zero.
Studying superconductors is tough as their properties rely heavily on the arrangement and behavior of their electrons. This makes them hard to simulate using classical computers but makes them a perfect candidate for a quantum one. By being able to simulate superconductors and their individual electrons along with different types of materials, it might be possible to discover a way to make ones that function at room temperatures.
Another place where quantum simulations can be used in electronics is with batteries. Batteries rely on electrochemical reactions inside them to keep the energy coming. With quantum simulations, it would be possible to test a large amount of reactions with varying chemicals to a high accuracy which could reveal much more efficient ways to make batteries last longer.
Catalysts
Catalysts are molecules in a chemical reaction which allow the reaction to proceed much quicker. Catalysts help run all the important chemical processes in our bodies. Being able to simulate these processes more accurately would allow us to have a better understanding of them which can further aid in creating drugs and also just better finding out how the body works.
Simulating these processes doesn't just apply to humans though. A common application brought up when talking about quantum simulations is its potential to help create new fertilizer.
The Haber-Bosch process, which uses extreme pressure to force nitrogen to bond with hydrogen, is the current method of creating nitrogen rich fertilizer in the US. Bacteria on the other hand, is able to do this naturally with catalysts that, if understood, could help create a much more efficient way to make fertilizer.
In 2017 Microsoft actually did calculations to test how many qubits would be needed to create such a simulation and found with 200 perfect qubits, the problem of creating better fertilizer could be solve in as low as a few weeks.
Environmental Sustainability
There are many ways in which quantum simulation of chemicals could help the environment. For starters, a significant factor in how effective solar panels and solar cells are is the material they are made out of. The material determines how much light is absorbed and how much energy is lost in the form of heat. Quantum simulations could help create new composite materials that optimizes the light absorbing properties while decreasing its tendency to release heat.
Furthermore, sustainable plastics that can be better degraded by the earth could be developed through the simulation of novel combinations of polymer chains. More energy-fficient fuels could be created through testing different combinations of chemicals. Creating new materials that have the same properties as currently used materials such as concrete or steel but with a smaller carbon footprint from manufacturing could drastically lower carbon emissions. The options are endless in how discovering new materials through quantum simulation could help the environment.
Material Science
Material science is the basis of so many fields. Transportation, architecture, robotics, aerospace, are all heavily influence by the materials available and the properties that come with them.
Material science helps emphasize the long-term impact of quantum simulations. When looking at the periodic table, there are over a 100 different elements, which all can be combined to make a seemingly infinite amount of combinations. Add on to that all the different combinations between different molecules. The properties of materials such as their strength, color, flexibility, conductivity, is heavily dependent on its atomic and molecular make up.
Chemistry started out as a science of trial and error where chemicals were randomly mixed together to figure out what properties the resultant chemical would have. With the invention of computers, this process became a bit more virtual but still came with a lot of approximations. With the ideals of quantum simulation, society will be able to simulate any combination of elements and molecules they want at a much faster rate and with much less resources wasted.
There are likely useful properties of materials that humans don't even know exists yet and quantum simulations will allow the ability to more efficiently discover them. Scientists will truly be able to work from the bottom up and create materials with useful properties by exploring them at a molecular level.
For example, if a material that’s lightweight but as strong as steel is needed, the process of discovering one would be exponentially faster as researchers would have a quantum toolkit of all the possible combinations at their fingertips. They would be equipped to easily try new recipes using the ingredients of the molecular world.
The ability to do molecular simulations using quantum computers is an exciting thing to look forward to. From the possibility of early applications by combing quantum and classical computers to the far future where quantum simulations allow for the discovery of new materials with the click of the button, its important for the world to stay up to date with the technology in order to utilize its benefits and help society progress.
We’re providing opportunities for the next generation of student thinkers, inventors, and learners, to publish their thoughts, ideas, and innovation through writing.
Our writers span from all areas of topics — from Growth to Tech, all the way to Future and World.
So if you feel like you’re about to jump into a rabbit hole of reading these incredible articles, don’t worry, we feel the same way. ;)
That’s why studentsxstudents is the place for getting your voice heard!
Sounds interesting? Why not join us on this epic journey?
