Magnetometry with Ensembles of NV Centers
Ensembles of nitrogen vacancy (NV) centers in diamond make useful sensors of DC and AC magnetic fields, thanks to their ability to perform sensitive, wide field-of-view magnetic imaging at ambient conditions. By using different techniques, experimentalists can use NV centers to sense frequencies from DC up to several GHz with bandwidths of up to roughly 100 kHz. The key figure of merit for this application is the sensitivity, which specifies the weakest field strength that can be detected with a signal-to-noise ratio (SNR) of unity in a given bandwidth. Compared to single NV centers, ensembles comprising N different NV centers in the sensing volume benefit from a √N improvement to the sensitivity, allowing them to reach sensitivities at the level of pT/√Hz.
Furthermore, ensembles of NV centers can be used for wide-field magnetic imaging at room temperature. As NV centers have four possible orientations within the diamond crystal, ensembles can even be used for vector magnetometry. Sensing with NV centers typically relies on optical detection: the strong signals generated by ensembles of NV centers make it possible to use photodiodes instead of the more expensive and cumbersome avalanche photodiodes. The ambient operating conditions and high sensitivities of ensembles of NV centers, alongside their wide frequency range and strong signals, thus make them a powerful tool for magnetometry and other types of sensing.
Techniques for sensing with ensembles of NV centers broadly fall into two categories: pulsed and continuous-wave (cw) approaches. Although pulsed approaches often reach better sensitivities, cw approaches are simpler to implement. The most common cw-based strategy uses a bias magnetic field Bz to split the |ms = ±1> levels as shown in Figure 1, so that the states can be measured independently. Depending on the orientation of the external magnetic field, the |ms = ±1> levels of different NV orientations in the ensemble split by different amounts, thus enabling these NV orientations to be distinguished from each other. In the simplest experiment, a green laser off-resonantly excites the ensemble of NV centers to generate a red fluorescence signal measured by a photodiode, and a signal generator sends microwave (MW) signals to an antenna close to the ensemble to drive spin transitions (see Figure 2).
The MW frequency is then swept while recording the fluorescence signal from the ensemble. As the MW frequency becomes resonant with a transition from the |ms = 0> state to one of the |ms = ±1> states, a dip in the fluorescence signal is observed because the |ms = ±1> states are dim states that have a chance to decay to a long-lived singlet state when optically excited (see Figure 1). The change in fluorescence when the MW frequency is resonant make it possible to record a spectrum. Although simple to implement, this approach to cw magnetometry is extremely slow given that the full spectrum is recorded for each measurement.
A more effective approach to cw magnetometry involves lock-in amplifiers. The MW frequency is first tuned to be nearly resonant with one of the transitions, so that the MW frequency rests on the slope of the transition. Modulating the bias field Bz with a time-varying voltage from a lock-in amplifier leads to a modulated fluorescence signal on the photodiode: this can then be demodulated with the lock-in amplifier to improve the SNR (see Figure 2). Additionally, a PID loop can provide a feedback to the fluorescence signal by supplying a current to shift the bias magnetic field, so that the transition does not drift outside of the measurement region due to temperature fluctuations over time. With its wide frequency range, high output power and dynamic range, and optional PID controllers, the MFLI Lock-in Amplifier can produce the signals needed to modulate the bias magnetic field and to lock-in to the modulated photocurrent.
Depending on the type of sensing experiment performed, many factors play a role in determining the scheme's sensitivity: these include the electron gyromagnetic ratio γNV, the number of NV centers in the sensing volume, and the strength of the fluorescence signal from the NV center ensemble. With cw approaches, the shot-noise-limited sensitivity can be reached by using weak optical and MW driving fields to avoid power broadening, and by choosing lasers and electronics with low noise.
User-friendly and reliable data loggers and plotters are also essential for monitoring how the lock-in measurement signals vary over time; the demodulated data can even be used to trigger measurements, e.g. when measuring suddenly changing magnetic fields. The MFLI offers an advanced set of built-in data processing and visualization tools, leading to simpler setups where less time is spent on home-built data analysis solutions.
The Benefits Of Choosing Zurich Instruments
- The low input noise of the MFLI unlocks better sensitivities and reduces the measurement time on your experiment.
- Perform fast measurements for scanning and imaging applications or take advantage of long integrations to detect weak signals: the large range of time constants of the MFLI allows you to adapt to different experimental needs.
- The integrated PID controller of the MFLI can feed back on the bias field modulation parameters in response to changes in the detected fluorescence signal, thus helping you to track the MW transitions and reducing the overall complexity of your experimental setup.
- With the LabOne control software you can save time on developing data processing and visualization tools: modules such as the Scope and Plotter can display and log the demodulated data, and the DAQ tool can use the demodulated data to trigger measurements.