Journey to The Center Of a Quantum Computer

The beating heart of IBM’s quantum computer is no bigger than a quarter. These extravagant machines promise to solve the difficult problems that stumble the best classic computers of today. The chip itself is only part of a bigger puzzle. Unlike the laptops that people use in everyday life, the computing infrastructure that supports the work of a quantum chip is layered like a Russian doll, with convoluted interconnections in a Rube-Goldberg-style machine.

However, even with its complicated construction and stunning design, a quantum computer is still a machine that performs operations using both hardware and software. Some of these actions are similar to those performed by conventional computers.

Curious to understand how they work? Popular Science took a look at the IBM Quantum Center in Yorktown Heights, New York City campus. Take a closer look at what’s going on inside, starting with what’s called the qubit (more in a moment) and zoom in, little by little.

To show quantum qualities, objects must be very small or very cold. For IBM, this chandelier-shaped structure, which looks like an upside-down gold steampunk wedding cake, is called a dilution refrigerator.

It keeps their qubits cool and stable, and is the infrastructure that the company has created for this 50 qubit chip. It contains several plates that become progressively colder the closer they are to the ground. Each plate is a different temperature, with the top layer sitting at room temperature.

The quantum processor is mounted on the lowest and coldest plate of the dilution refrigerator that reaches a temperature of about 10 to 15 milliKelvin, which is about -460 degrees F. The first stage of cooling involves large pieces of copper seen draped in the top layer that are connected to cold heads as part of a closed-cycle helium cryocooler.

More tubes feeding the lower levels introduce another closed cycle of cryogenic material, consisting of a mixture of helium isotopes. At the back of the housing structure is the hidden support infrastructure for the chandelier.

This includes the gas handling system that supports the cryogenic infrastructure, as well as the pumps and temperature monitors. And then there’s the classic custom control electronics. When users run a program using the IBM Quantum Cloud Service, they actually orchestrate a set of doors and their circuits.

These are transformed into microwave pulses that are correctly sequenced, aligned and distributed in the system to control the qubits. And the read pulses retrieve the states of the qubits, which are translated into binary values and returned to users.

Conventional computers represent information using one or zero binary bits. In the case of quantum, the information is represented by qubits, which can come in a combination of zero and one. This is a phenomenon called superposition. There is superposition all the time in the real world.

Music is an overlay of frequencies, for example,” says Zaira Nazario, technical manager for theory, algorithms and applications at IBM Quantum. Because it is a waveform, it provides an amplitude of zero and one.

This means that it comes with a phase, and like all waves, they can interfere with each other. Superconducting qubits are based on the chip and have been packaged in something like a circuit board. The wires and coaxial cables for the input and output signals protrude from the circuit board.

In the new higher Qubit chip models, IBM has been working towards more compact solutions involving cabling and embedded components to be more efficient with space. Less footprint means components would be easier to cool. Currently, it takes about 48 hours to completely cool a quantum computer to the desired temperatures.

For the quantum computer to function properly, each plate must be thermally protected and insulated to prevent black body radiation from affecting it. Vacuum engineers seal the entire device to prevent unwanted photons as well as other electromagnetic radiation and magnetic fields.

The container that contains the dilution refrigerator inside. Qubits are controlled with microwave signals that range from 4 to 7 gigahertz. Conventional electronics generate microwave pulses that move towards the cables to bring the input signals to the chip and carry the output signals back.

As the signal descends into the chandelier, it passes through components such as filters, attenuators and amplifiers. IBM works largely with superconducting qubits. These are small pieces of metal that rest on the wafer, which is used to make the chip.

The metal is composed of superconducting materials such as niobium, aluminum and tantalum. A Josephson junction, made by superimposing a very thin insulation between two superconducting materials, provides the essential non-linear element needed to transform a superconducting circuit into qubit.

“What we’re building are quantum examples of oscillators,” says Jerry Chow, IBM’s director of quantum infrastructure. “ Oscillators convert a direct current from an energy source (in this case, microwave photons) into an alternating current, or a wave.

Unlike typical harmonic oscillators, a non-linear oscillator gives you uneven spacing of energy levels, says Chow. “When you have that, you can isolate the lower two to act like your quantum zero and your quantum one.”

The resonators, seen here, connect the qubits to each other and to the control electronics. IBM, Think of a hydrogen atom. Physically, it has a set of energy levels. The right wavelengths of the light hitting this atom could promote it to different states. When the microwaves hit the qubit, it does something similar.

“You do have this artificial atom,” says Chow. “We have a quantum of energy, which we move by placing the right amount of microwave photons at a certain pulse for a certain time to excite or de-excite a quantum of energy in this non-linear microwave oscillator.”

In a conventional computer, there is a state (one) and an out state (zero). For a quantum computer, the off-state is the ground state of the artificial atom. Adding a pulse of a particular photon of microwave energy would excite it, promoting it to one. If the qubit is hit again with this impulse, it would be returned to ground state.

Let’s say it takes 5 gigahertz for 20 nanoseconds to promote a completely excited qubit-if you were to halve the amount of energy or halve the amount of time, you would actually be driving an overlay state, Chow said. This means that if you were to measure the state of the qubit with a resonator, you would have a 50% chance of it being in zero, and a 50% chance of it being in one.

Users can play with circuit elements, pulse frequencies, duration and energy between different qubits to couple them, swap them, or perform conditional operations such as building entangled states and combining single qubit operations to perform a universal calculation across the device. When the waves cross, it can either amplify or deconstruct the message.

The practical uses of quantum computers have evolved in recent years. “If I look at what people were doing with the system in 2016, 2017, 2018, it was using quantum for quantum research… condensed matter physics, particle physics, that sort of thing,” says Katie Pizzolato, director of strategy and applications research at IBM Quantum.

The key will be to take conventional resources and make them quantum-sensitive. We need to make sure that experts in their field understand where to apply quantum, but are not quantum experts.”

IBM’s interest in quantum problems with their machines can be grouped into three categories: chemistry and materials, machine learning and optimization (finding the best solution to a problem from a set of possible options). The key is not to use a quantum computer in every part of the problem, but on the most difficult parts.

The IBM team is constantly looking for real problems that are difficult to solve for conventional computers because of their structure or the mathematics involved. And there are many interesting places to look for them.

Conventional computers solve basic mathematical problems using binary logic and circuit components such as adders. However, quantum computers are really good at making linear algebraic multiplication matrices, and representing vectors in space. This is due to the unique features of their design.

It allows them to perform functions such as factoring relatively easily, a problem that is extremely difficult for a conventional computer because of the exponential increase in the number of variables and parameters and the interactions between them.

There are structures in this factoring problem that allow you to take advantage of the entanglements, of everything you get with these devices. That’s why it’s different,” Pizzolato says. And with chemistry and material problems, qubits are just better at simulating properties like bonds and connected electrons.

“We think about the kinds of things that we can map to quantum circuits that are not simulable in a conventional way, and then what do we do with them,” Pizzolato says. Much of the discussion about the algorithm is about how to exploit the underlying mechanisms of this device. How do you map the higher-dimensional spaces and how do you use that coordination and those matrix multiplications to give you the answer you want?”

Remember, qubits can have a value of zero, one, or a combination of both. Since qubits are waveforms, engineers can rotate the zero or one to give it a negative amplitude. Qubits can also become entangled, a unique property of quantum mechanics that has no classical analog.

Tangled qubits can contain information not only in zeros and those themselves, but also in interactions between all. In addition, there are gates in quantum circuits that can rotate the qubit to change its phase, and oscillators can tangle these qubits.

“The art of making a quantum algorithm is to manipulate all these entangled states and intervene in such a way that the wrong amplitudes cancel out, and the right amplitudes manifest, and you get your answer,” Nazario says. “You have a lot more flexibility in a quantum algorithm because of all these entangled states and this interference with an algorithm that only allows you to switch between zeros and one.”

Qiskit, IBM’s open-source development kit for quantum computers, contains information on various types of algorithms and quantum programs at different levels of detail.

Still find it difficult to visualize what the qubit does? Let’s turn to a few examples of how IBM partners use quantum computers. For example, biopharmaceutical company Amgen is looking to use quantum computers and machine learning to predict which patients would be best suited for a drug trial based on health records and other factors.

And Boeing applies quantum computing to analyze corrosion coefficients on aircraft. Aircraft wings require a certain density of materials. Engineers make them with different layers of various materials, but need help to understand how they should organize the layers in a way that makes the wings stronger, cheaper and lighter. It comes down to a combinatorial optimization problem.

Goldman Sachs uses them to price options. These are very complex operations that cost a lot of money in calculation. And they have complex distributions,” Nazario says. This involves calculating the derivatives of the variations of these options (a linear algebra operation), which will inform them about the risks. Finally, in the field of natural sciences, research groups were interested in the use of quantum computers to study photosynthesis.

While IBM has steadily increased the size of its quantum computer processors, and built a community of industry partners, national government centres and academic institutions, the company is still finding the best ways to move forward with hardware and software.

The company has already said it will have a machine capable of quantum advantage (in which it can solve a problem reliably and accurately better than a conventional computer) ready by 2025. This means that in addition to developing new components, it needs to address issues and make what is already working more effective.

This is a big part of the goal of the software. We recognized that a lot of tools, error mitigation tools, smart orchestration, the idea of knitting circuits, how to break down problems to extend what we can do on the quantum computer, these technologies are becoming increasingly prolific,” says Pizzolato.