I'm going to argue that 8.5.1 is also where shelf life for production raw materials belongs.
While I agree that 8.4 can control how much remaining shelf life a purchased material must have
at the time of purchase, it does not control monitoring of that remaining shelf life while the raw material sits in our building waiting to be used, or in-between processes. I also agree that 8.5.4 specifies outputs, rather than raw material inputs. But in the OPs case, "The left over mixed ink is stored and reused..." so an output of this process is a raw material for some later process.
Specifically, 8.5.1b ("availability and use of suitable monitoring and measuring resources") and 8.5.1c ("implementation of monitoring and measurement activities at appropriate stages") would seem to apply to tracking and acting upon shelf life expiration dates, whether it be for purchased raw materials, or 'subassemblies' such as the mixed ink.
As to the OP's question ("Can I state that we have identified expired ink as an internal risk but not to the degree that it would impact the product and require mitigation (based on there being no customer complaints or internal scrap caused by expired ink) and include this in our risk analysis process?"), I would say 'Yes...but', where the 'but' is three parts:
- Is the data (lack of customer complaints or scrap) based on known expired ink? If you don't know if the ink was expired at the time it was used, then you don't really know if your data is good evidence of the lack of impact of expired ink.
- Any potential concern over latent nonconformances, which the customer may not experience now, but could experience over the next months/years. E.g., the product for the customer is fine until 2 years go by and then the ink starts to peel off, etc. due to the ink being expired at the time of production. So, I would include in any risk analysis any evidence you have of no latent issues due to the ink (e.g., "there have been no ink-related nonconformances even after the product has been in the field for 5 years").
- I presume you already have some kind of controls in your processes to prevent 'bad ink' (whatever that means) from being used? E.g., ink that has hardened, separated, etc., is somehow already detected and prevented from being used?