Now you have me wondering if that faulty code from the cat is just from the leaky manifold. Although if logic follows, once the sensors first sensor in the cats detect more air in the system, it would send a signal to the injectors to provide a richer fuel mix, thus causing unburned fuel into the cats which may damage the cats.
Does anyone have any other scenarios you may think of.
Yes, it can be the case. The thing is that the catalyst diagnosis is not "an exact science" but is a functionality in the engine management system.
If I'm not totaly wrong on the configuration of the 355 it uses the signals from the oxygen sensors (one in front and one after the catalyst) to judge the performance of the catalyst. In order to be legal you have to turn on MIL (malfunction indication light a.k.a. check engine) if the emissions are higher than the corresponding emission standard levels.
From that point of view, as an engineer at Bosch (which I believe did the job on the 355) you want to be sure that you turn on MIL for all catalysts that generates higher emissions than are legal in order to avoid for example recall on sold cars which is about the most expensive thing in the world. On the other hand you don't want to turn on MIL if you are not sure that the catalyst actually is malfunctioning. If you do, then you have a customer that has to have his/her cars at the shop and also pay for a new cat and that's no good will at all. Of course it would be even worse if they were to find out that it actually was unnecessary...
Anyway this is a bit tricky because when you do the calibration you do not have a lot of real world aged catalysts to do the calibration on. This leads to the calibration not being exact but as said a judgement of where to draw the line between a good and a bad cat so to speak. If you throw in the uncertainty of a leakage in the manifold you really can't tell if the diagnosis are working properly and is able to make the right decision if a cat is good or bad. That's why I suggested that you get the leakage sorted out first, the have your car reseted and then see if the cat code comes back.
Just some more reasoning: As you write if the sensor gets an extra dose of air then the EMS should react with injecting more fuel so that the sensor will detect the target lambda. That would then mean that the engine would run richer than wanted but the sensor (and thus the cat due to the extra air) would see the target mixture and it would probably not directly damaged the cat. The problem is if you have unburned fuel AND extra air coming into the cat and the combustion takes place inside the cat. Then the high temperatures would most likely damaged the cat and that may well be the case here.
even more blablabla: As someone else wrote in the thread you could expect to have a higher pressure inside the exhaust than outside and then no air would enter it. Unfortunately it in most cases is not as simple as that. If you run on idle or low speed and loads (which is where the most diagnoses runs) you do not have that high pressure and for sure you can have swirls in your gas flow that creates local areas with very low pressure giving air the possibility to enter.
Finally I'm not sure if I have made any sense or help to you. I apologize for being so fuzzy in my attempt to explain my view. :-( I guess the bottom line is this that I wrote earlier:
If you throw in the uncertainty of a leakage in the manifold you really can't tell if the diagnosis are working properly and is able to make the right decision if a cat is good or bad. That's why I suggested that you get the leakage sorted out first, the have your car reseted and then see if the cat code comes back.