I think I've read description of this experiment somewhere. Now all I can find is this other recollection on Redd dit (sorry for the extra space, as the original spelling is banned by this site... the url, also banned so you'll have to remove the blank and copy/paste in your browser, is https://www.redd it.com/r/bikecommuting/comments/4nor1d/some_questions_on_bike_chain_cleaning/), where the conclusion is that there were no significant differences.

Frankly, I am not sure what should be expected from this kind of experiment -- I mean, can we really expect that one half of the chain can stretch more than the other??

We can find all sorts of chain mileage, and this thread is no exception. I would argue that the expected life of a chain in average to adverse conditions can be expected to be in the vicinity of 2 500 kms for 0.5% stretch. This happens to jive with my experience and is coincidentally almost precisely the conclusion of the Wipperman-sponsored study.

I am not aware of any meaningful and systematic comparison of chain life under different maintenance schedules. When I wrote "anecdotal evidence", I mean that if you read about ultrasonic cleaning, you'll find that many posts claim important improvements in chain life. These claims may well be biased (if you go to such extreme trouble, it better do something, right?). But I seem to be able to see that chain cleaning with a solvent does reduce stretch, so I'd tend to believe that it does. (and if you read threads about ultrasonic whistles that purportedly annoy dogs, you'll find many testimonials saying that users have tried and that it doesn't work reliably -- so maybe there is no confirmation bias after all).

----

Regarding the relevance of this, while traveling... it is certainly not worth spending energy for short to medium-length tours (i.e. of less than, say, 5 000 kms). It is not so clear for longer tours, however. Two observations:

(1) I've perused repair logs that long-distance tourers post on CGOAB and other places, and the average chain life is under 6 000 kms. I suppose that chains were more likely than not to be changed after they created issues. I've also met long-term tourers and they spontaneously mentioned having had to to replace chain and cassette during their (7 5000+ kms) trip, which was not a found memory. First there is the annoyance of mechanical failures. Second there is the time wasted finding replacement parts, that may or may not be fully compatible or matching the quality of your setup.

(2) I am presently rotating chains in order to increase drivetrain's life. Zero trouble with Connex's quicklinks that do not require any sort of tool. So chain cleaning while traveling simply means carrying the alternate chains in a bottle half-filled with some solvent, placed in the under-the-downtube bottle cage (which I do not typically use). Really not that complicated.

---

Actually, I think that I've settled the question with myself, and that the answer is kerosene. Apparently, white gas isn't available in China, whereas kerosene is, and MSR international stove burns kerosene, among other fuels. So the next question is whether one can use kerosene that has been used to clean a chain as camp stove fuel