Working meeting at Hahn-Schickard on the microwave

A working meeting on the topic of microwave excitation of diamonds took place on 4 April.

Microwave excitation is necessary to be able to measure optically detectable magnetic resonance (ODMR). In addition to a microwave generator, this also requires a microwave structure that emits the microwave and couples it into the diamond. In our previous test setups, a discrete resonator was used for this purpose.

This consists of two high-frequency sockets (SMA) soldered together, the centre pins of which were shortened and to which a 1 pF ceramic capacitor and a 2 mm diameter conductor loop made of copper wire were then soldered. The illuminated diamond sits in its centre. Together with a 50 Ω terminating resistor, the structure forms an almost ideal resonator, which we use to bring the electrons of the nitrogen defect in the diamond into the ms = ±1 state.

This resonator works so well that a microwave generator with very little microwave power is required (< 8 dBm). It actually sounds good, but we build these resonators by hand and trim each resonator individually to a resonance frequency of around 2.87 GHz using the stretched surface of the conductor loop. Measurements prove how good the resonator is. The lower the peak at resonance, the better and the less microwave power we need.

A hand-built resonator is naturally unsuitable for our QOOOL Kit Magneto, which is why we are looking for solutions that are small, can be manufactured cost-effectively and reproducibly and also work as well as our discrete design. That's why we started with an omega resonator and have designed, manufactured and tested various resonators on very different substrates over the past few months with the help of simulations.

Resonator on special high-frequency material (RO4003C)

This resonator behaves very similarly in reality (right) to the simulation (left), but is still too large for our QOOOL Kit Magneto and does not really come close to our discrete resonator.

Resonator on 100 µm thick PEEK film

These resonators are small, but the peak at the resonant frequency does not go down far enough in reality (right) compared to the simulation (left). We suspect that the simulation parameters do not quite correspond to reality.

Resonator as flex conductor on polyimide

They are cheap to produce, but ...

... simulation and reality don't really match up here either.

Conclusion:

Unfortunately, none of the resonators we have realised so far have met our requirements. Will we find a solution for this? Find out in the next update on the microwave.