---

but after the reaction when everything should have been converted to meth then would'nt it be considered a byproduct or maybe just an impurity?

---

I think the correct answer is neither in this case. Side-rxns take place with that substance that in turn create the impurities. When the reaction is stopped prematurely and worked UP, the resulting product will contain impurities. Any Iodo's will be left behind during the work-up.
Free Iodine from post-rxn solution can be oxidized & removed with sodium thiosulfate.
I would suspect Iodoeph to be result of an under-reduced rxn.
Yield reports confirm a lower yield of meth when these intermediates are present. Instead of meth xtals obtained....nothing is obtained because the Iodo's stay in the basified solution.
Heavy emulsion during work-up also substantiates this occurance.
Yes...less than clean feedstock will also create this but the longer cooks usually provide relief to a small degree.
That tells me that full reduction of dirty feedstock can only occur given the time necessary to counter the buffering effect that gaaks play in rxn.
Ibee's done rxns at all stages(Superclean feed) and stopped them purposely during the early LWR rxn experiments to record the yields and meth/pfed percentages.
Most bees at the time thought Ibee flipped his wig cause who in their right mind would create failure scenarios in order to achieve meth???
Well the cats outta the bag on that note and Ibee will never stop that line of investigation in order to reveal what works best under any condition.
The strength mystery sparked these early experiments.
When Iodo's are present they account for expected meth weight and yields can be used to back this up.
This is part of what compelled Ibeeware to push the time/temp & hydration envelope to absurd extremes to unheard of levels.
To this day it's difficult to expect bees to adopt that tradeoff when they can have product in a fraction of the time it takes for a full reduction.
I think you'll agree that impurities are created by side- rxns in the less controlled faster cooks.
The old journals contain references and instruction that pointed to lengthier, milder rxns producing higher yields and in higher purity.
I guess the arguement will always exist between that which can be done in half the time, producing some of what bees are after in a potent enough form to pass for the real thing.
But evidence and experimentation leads me to believe that less impurity exists in a fully reduced product because the time was taken for the rxn to proceed to completion, cooking off or fully converting those intermediates instead of stopping the rxn before that takes place.
The ancient alchemist deluxe undoubtably has been performing this rxn for longer than most and has incorporated steps and procedures that minimize the iodo's presense in the end-result.
But for anyone that performs a curbshot or a superassman style rxn....those impurities run rampant.
I guess the answer is this....
What most think is fully reduced at the first sign of meth, does not exclude that product from higher levels of impurities that when ingested/consumed have similar stimulating effects on the CNS!

Allowing the rxn to proceed past those suggested cook times has always produced a superior product with less health related cramping and impurity associated jawclenching hellrides!
Again....not answering for the moose...these are just my observations based on Ibee's personal experimentation!