Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
I think a mica capacitor would probably be best but you would need a larger value mica capacitor than I have seen. (For a reactance of 1 ohm you would need 5.3 nF but you would probably need much less than 1 ohm reactance so for 0.1 ohm you would need 53 nF.) If it did not need to operate over a wide frequency range you could cancel out the capacitve reactance with a series inductor (Forming a series tuned circuit.) With 5.3 nF capacitor you would need an inductor of 5.3 nH (With a standard value of 4.7 nF you would need a 6.0 nH inductor.)
My suggestion is definitely no good for that frequency range. I've not done any work with RF amplifiers on the HF bands The only reasonably high power amplifirers I have built have been for 70 cm (400W) and 23 cm (100W)
Back when I was doing VHF power amps, we typically had to put shunt capacitors right on the collector of the NPN power transistor and these caps took a lot of current, as I can remember burning up a few. We usually used multilayer mica caps in metal surface mount packages but not the kind you see today. They looked like this: https://www.rfparts.com/capacitors/capacitors-metalcased/capacitors-j602/j602-43.html
I expect that this sort of cap is obsolete now. Later on, we changed over to multilayer plastic capacitors but I can't remember what kind of plastic. They were from only one vendor ( a Japanese company) and could stand a lot of RF current, and were a vivid green colour, but that's not much help. I think that porcelain caps might be the way to go nowadays, especially those with the larger soldering tabs as these offer very good heat sinking and are compatible with microstrip layout. For example, the ATC series 100 caps include RF current graphs in their spec, and 4.5 amps is within their range.
that is a lot of current, especially at such a high frequency (6.36 A peak).
Disk ceramic capacitors would have a nice low ESR but would introduce distortion in the waveform. You would also need many disk ceramic capacitors in parallel, to handle the current and keep the total effective series inductance down, off the top of my head around 20.
But, foil or even film capacitors may be worth a look: MKP, MFP, MKT, MFT. I would still be inclined to put some foil/film capacitors in parallel though, and perhaps a few, lower value, disk ceramics for good measure.