Please read the question on security stack exchange which I've linked above. I'm not talking about the main battery. Tinfoil does not work for every phone. Regarding your last suggestions please read security.stackexchange.com/questions/59093/…. security.stackexchange.com/questions/12740/…

the correct answer on that question was "No, even with any internal capacitors or small secondary batteries, there isn't enough power to get off even a small amount of transmission. Transmission simply requires too much power, even for relatively short range given the type of transmitter in the phone. "

A Faraday cage is basic physics, if it doesn't work you haven't built a good one.

If you factor in improvements in the capacity of internal batteries it will become possible. Can you provide references, that transmission is not possible with the power an internal battery provides? I'm not saying a Faraday cage does not work, obviously it does. However wrapping in tinfoil does not necessarily work. And the Faraday cage bags you can buy wear out pretty easily so they might function for a few days / weeks but fail to do so for longer periods of time.

Any secondary internal battery is going to be obvious to anyone opening the device, so no phone you buy in a shop is going to have one. Espionage agencies may modify the occasional phone for spying on a specific target, but unless you're in that trade or a high-ranking government official, getting your phone swapped out for a rigged one is not a credible threat.

"As you can see in the picture above, the speaker is about the size of a dime and the microphone is no larger than the watch battery beside it." electronics.howstuffworks.com/cell-phone6.htm

I feel the discussion is steering into the wrong direction. Given the question, the discussion should not be about what constitutes a credible threat but rather if there exist any, preferably non-intrusive methods, to degrade the capacity of the (hypothetical) internal battery.

Any "methods, to degrade the capacity of the (hypothetical) internal battery" depend on the implementation of that battery. A battery installed for espionage purposes isn't going to be reachable by regular phone applications

It's far easier to hack the phone via software and use the existing microphone and radio than to steal your phone, insert tiny custom circuitry and get you your phone back.

I believe the NSA or other three letter agencies have already implemented a function to power your mobile phone even when your main battery is removed. The could have done this using National security letters. (en.wikipedia.org/wiki/National_security_letter) Its no secret that they already used similar measures to get your email, browsing history etc. (see theguardian.com/world/2014/mar/19/…)

By using National security letters a producer can be forced to implement a certain functionality. For example a small chip powered by the internal battery which broadcastes your location any x hours.

That would be trivially detectable. Millions of phones are opened up every year, radio scanners etc.

How is that trivially detectable? If they used a frequency domain normally not used? Maybe many phones are opened but how many people actually know what the chipset is capable of?

" a frequency domain normally not used" - there aren't any. There is a very limited number of frequencies available. You can get scanners for all of them.

Any electronics engineer with some experience in mobile engineering can recognize a battery and transmitter. There are lots of engineers capable of this, and many are incredibly cirious so they'll open any device they buy.

And you only need one of them to make the discovery.

Regarding your objection that there are scanners for all the frequencies. That could be true. However it would be difficult to test, because such functionality would only very rarely be used by our favorite three letter agencies. I have not looked into a phone myself but I will definitely open my old phone when a buy a new one. The article I linked above (electronics.howstuffworks.com/cell-phone6.htm) claims that there is indeed a second battery - at least in some phones.

Also there is no need for additional transmitters, just design the circuit as such that the internal battery can also power GPS.

I'm not sure if such wiring would be detected by even the most curious engineer

Probably correct at the moment. It would be nice to have a reference to back up the claim that conventional clock batteries in mobile phones do not have enough capacity to run the transmitter. In the future however that might change with new battery technology