Understanding nitrogen-vacancy diamonds
The previous blog posts by my colleagues established the importance of and the challenge in having good magnetometers. If you have not done so yet, I encourage you to read them here. At SBQuantum we pride ourselves in developing a new quantum technology that holds the promise of exceeding what is currently achieved by commercial magnetometers: we trust it will correct for drifts, dead zones and heading errors. To achieve this feat, we develop a device centered around special artificial molecules that can be grown, in a laboratory setting, inside a diamond crystal: the so-called Nitrogen-Vacancy colored centers.
Let me explain a thing or two about NV-centers. Recall that the diamond is a carbon crystal with a tetrahedral structure. The NV-center is obtained by replacing a carbon atom with a nitrogen atom and by removing one its neighbors, thus creating the vacancy. This sets up an artificial molecule that electrons can bind to. It is the properties of this bound electron that strongly depend on the magnetic field. As a bonus, the carbon structure surrounding this molecule acts as an effective vacuum that is extremely clean, partly thanks to the well-known very rigid nature of the diamond crystal. In short, we end up with a degree of freedom susceptible to the magnetic field surrounded by an exceptionally clean environment and we thus have our magnetic probe. Moreover, this probe is vectorial because the tetrahedral crystal provides four orientations that enable magnetic field measurements along four axes. The diamonds are grown by element 6 as part of their newly launched DNV-B14 and DNV-B1 lines.
You might interject that quantum magnetometers already exist, like superconducting quantum interference devices (SQUIDs) for example, a cryogenically cooled microscopic circuit that reacts to any change in magnetic environment. One major difference is that our device is hand-held and operates at room temperature, drastically improving operability conditions and costs.
Important questions remain. How are we going to interact with this probe? As any measurement device, we need to initialize it, manipulate it, and measure it. It turns out that the physics of the NV-center enables all of these. For measurement, we shine a green laser on the diamond. It then re-emits red light in a way that depends on the NV-centers’ states. As a bonus, this measurement steers the state to a known one, such that measurement is also used to initialize. Lastly, to manipulate the probe, we require an additional control knob: microwave signals. By bathing the diamond crystal in microwaves in our custom-designed cavity, we can send pulses that modify the state of the NV-center in a controllable and predictable manner. Combining these three modalities, we can effectively measure the magnetic field at the diamond’s position.
A ton of details are diligently being worked out. First, in a single diamond of a few millimeters we are actually probing millions of millions of these artificial molecules. With this ensemble of NV-centers we cannot ignore the statistical nature of the signal. Second, given the structure of the crystal, by combining the reading of NV-centers in many directions, we can measure the magnetic vector and not just its amplitude, enriching the collected data by moving beyond one-dimensional to three-dimensional data. Third, to simplify its operating details, it is sometimes preferable to purposefully add magnets to the device to create an offset in the measurement. You can imagine the crucial importance of characterizing this offset very precisely when aiming to measure magnetic fields to 1 nT precision. This is the critical limiting factor influencing the technology and we have a novel way to get around the problems these magnets can cause.
As a team, we are faced with the challenges brought by both the physics and the integration. Very satisfying results were shown in the lab using an optical bench bigger than your average dinner table, a laser source as big as your typical microwave oven and a voltage source as big as your average desktop computer, each costing tens of thousands of dollars. Now, the question is whether we can reproduce similar performances using integrated chips and diodes such that the whole device is producible at reasonable cost and fits in your hand. The challenge is real, and we are up for it.
By combining the expertise of mathematicians, physicists and electrical, robotics and software engineers, we tackle the problem from all sides. We are optimistic and enthusiastic. So much that we embarked on NGA challenge sponsored by NASA to improve the accuracy of the World Magnetic Model so our sensor needs to be NASA certified by Feb 2023. We dream of the day when our device circles to Earth to improve our scientific knowledge of its magnetic field.