Near-field magnetic loop for 800 MHz - 1 GHz

[atomtrapper]

Hi all,

I am wondering if anyone can share any wisdom regarding the construction or purchase of a small magnetic loop antenna for use at 800 to 1000 MHz. The application is a little unusual: the point of the antenna is to drive spin-flip transitions in trapped lithium atoms inside a vacuum chamber, for a series of science experiments. Currently I have a homemade antenna made with a copper strip bent into a loop of about 3 inches diameter and held in a 3D-printed PLA jig, with the ends overlapped to form an air-gap capacitor for very approximate tuning. The atoms are about 1.5 inches away on the other side of a glass window, and the whole antenna assembly is inside an 8-inch-diameter stainless steel bucket which is part of the vacuum chamber. Because of this, I only care about near-field properties: basically I want the maximum possible current to be flowing in the loop as I drive it in this frequency range, which will give rise to the maximum alternating magnetic field at the location of the atoms (I don't care about the electric field either). Far-field transmission is pretty minimal because of all the shielding from the conducting materials in the vacuum chamber and nearby, which is fine (good, even-- I obviously don't want to interfere with other users of this band).

The current antenna is almost certainly not ideal, and I am wondering if there are tricks I can do to enhance its performance or particular references I should be reading to understand the design process better. I understand the theory of electromagnetism well, but am a total amateur in antenna design. In addition to wanting to improve the performance in general so I can drive the atoms harder, I also noticed that (after years of use) the PLA holder is charred near the airgap capacitor; either PLA is not the ideal material here and is absorbing some of the signal, or the antenna is generating a good bit of heat.

Any thoughts, resources, or ideas would be very welcome. Thanks in advance for your time, and as a newcomer I apologize if I have posted this in the wrong thread or forum.

 

[RFI-EMI-GUY]

Holy cow, you lost me at "spin-flip transitions in trapped lithium atoms".

Having said that: I saw something like this years ago at open house at the the National High Magnetic Field Laboratory facility at Florida State University Tallahassee. It was part of a magnetic resonance experiment.

You might want to contact Mag Lab to see if they have any expertise in making such an antenna.

The great thing about antennas are that they are scalable from one frequency to another and there are modeling tools like NEC for simulation.

Maximizing current requires minimizing resistive loses by using a heavy conductor like silver or copper.

Because there is skin effect you may want to silver plate the copper. Any unwanted heat is either caused by resistive loss or dielectric heating of the PLA. You might use a no contact IR thermometer to see where heat is generated in your existing antenna and the connections including cables and connectors.

You have to match the antenna to the RF generator to maximize current. A return loss bridge is used to tune the antenna for the minimum reflected power. Your RF generator might want to be used with a ferrite circulator which is strictly for protection of the generator at this point. However, it will incur some insertion power loss. The important part is to minimize impedence mismatch of the antenna itself.

As far as heat damage and heating loss, a machined TEFLON insulator would be more suitable than the PLA. If you can minimize the insulator to simply an air gap, it would be best.

The steel jug, could also be affecting the resonance so you may have to modify its size or shape to tune out any power reflections caused by it. Sweeping the assembly with a return loss bridge and VNA should reveal those resonances. There are absorptive (Eccozorb) materials that can minimize the reflections, but they will create heat and absorb power. But might be helpful in the analysis.

I am going to step back and let the antenna experts weigh in.

 

[atomtrapper]

Thanks for your message! Oh, interesting re teflon, we could certainly do that-- we used the PLA for prototyping speed and then basically just kept using it out of laziness. We do have a circulator in the system, for the reasons you say. You're definitely right that all the neighboring conductors mess up the resonance-- a practical problem we have is that we can only tune the antenna outside the system (coarsely, by adjusting the geometry), then when we put it in the bucket the tuning changes. Talking to the mag lab, silver plating, and looking with an IR thermometer are also good ideas we can try-- thanks again!