Research shows that it takes 300 million qubits to crack bitcoin

In recent years, many people are worried about the risk of quantum computing cracking the current encryption system. At present, the bitcoin network using 256 bit elliptic curve encryption (ECC) is also exposed to the risk. How many qubits can crack bitcoin? The AVS quantum science paper [1] published by British and Dutch researchers on January 25 shows that cracking bitcoin encryption in one hour requires a machine with 317 million qubits. Even if the encryption is cracked in one day, the number will only drop to 13 million qubits.

In this paper, the researchers first discussed the problem of simulating FeMo CO molecule, which is the core component of nitrogenase. Nitrogenase is responsible for converting nitrogen in the atmosphere into ammonia, which is the main raw material of chemical fertilizer. But the traditional Haber Bosch process consumes a lot of energy, so scientists are looking for a better process. The key to solve this problem is to understand the structure of FeMo Co, but FeMo CO is a complex molecule, which is located in the core of enzyme, and it is difficult to model by classical computer.

Researchers used two parallelization methods to simulate FeMo Co catalysts: surface codes and autoccz factories. It is found that autoccz factories perform better in resource estimation. A code cycle time is 1 μ S superconducting device needs 7.5 million qubits to simulate FeMo CO in 10 days, while 235 μ The capture ion device with s code cycle time takes 2450 days.

Then, the researchers applied this method to the logical resource estimation required to crack 256 bit Elliptic Curve Cryptography (ECC) (protect the public key in bitcoin network). They use the logic resource requirements developed by the latest algorithm, which is about 2 orders of magnitude higher than the previous technical level in the following ways.
There is a small time window, about 10 – 60 minutes, in which the public key is available and vulnerable after the transaction starts. They quantified the number of physical qubits needed to crack the encryption in one hour as a function of code cycle time and basic physical error rate.

Using surface code, code cycle time = 1 μ s. Reaction time = 10 μ s. Physical gate error rate = 10-3, which requires about 317 million physical qubits to crack the encryption within 1 hour. In addition, 13 million physical qubits are needed to crack the encryption in one day. If the basic physical error rate is more optimistic 10-4, 33 million physical qubits are required to crack the encryption in one hour. This huge demand for physical qubits means that the bitcoin network will be protected from quantum computing attacks for many years (possibly more than 10 years).

The number of physical qubits required to crack bitcoin 256 elliptic curve encryption, and the maximum running time is fixed. The number of qubits is a function of code cycle time and basic physical error rate.