In a paper recently published on Nature Physics, the National Institute of Standards and Technology (NIST) documented the implementation and verification of a two-qubit quantum computer that, according to researchers, is a truly general-purpose machine and could soon be used as a building block for much larger quantum computers.
Building a programmable quantum computer
A number of two-qubit quantum computers - even all-electronic in nature - have already been developed, but the circuits used for these devices are all special-purpose, meaning only the inputs, and not the operations to perform, can be changed. In contrast, the processor developed by the NIST team allows any two-qubit program to run and is therefore a milestone in the history of quantum computing.
The NIST processor represents qubits with a beryllium ion that can be turned into a "0" or a "1" by ultraviolet lasers. The lasers can also place the ions in a superposition of both "1" and "0" at the same time, or "entangle" the two qubits in a process that links their respective values even when the two ions are physically separated from one another.
To test the processor, the researchers elaborated 160 different two-qubit programs. Rather than performing mathematical operations, though, each program simply manipulated the states of the ions through a different combination of value assignments, superpositions and entanglements, and the researchers then compared the results with the theoretically correct ones. According to the team's reports, the processor returned the correct answer 79 percent of the time across 900 runs, with each run lasting about 37 milliseconds.
Quantum computing vs. digital security
The RSA encryption algorithm is widely used in e-commerce and to encrypt sensitive information across a number of domains, and relies on the fact that prime number factorization on very large numbers is very computationally expensive. However, prime factorization is much easier with quantum algorithms, meaning RSA encryption could be easily broken compromising the safety of our data (and finances). Given that the modularity of the NIST processor could make powerful quantum computers possible, should we be worried just yet?
The answer, at least for now, is no. Even though a more complex quantum computer could be built using several of the two-qubit modules demonstrated by the team, the outputs coming from each module still have a very high error rate of 21 percent, meaning partial outputs must be verified and corrected every single time. This requires a substantial computational overhead that currently outweighs the advantages of quantum computation. Future research will aim to improve the accuracy of the processor, but this, researchers say, is no easy task and will likely take several years to achieve.
Meanwhile, the module could find more immediate applications in simulating quantum systems: large quantum simulators could for instance help explain the phenomenon of high-temperature superconductivity, which scientists and engineers plan to exploit for a more efficient storage and distribution of electric power.