Hello all:
I'm sorry I haven't been participating for some time. Let me see if I can catch up on some of the conservation threads that have been ongoing.
I've always been rather circumspect regarding benzotriazole and its effectiveness on coins. Of course, I've only been working with benzotriazole since 1990, and perhaps Hannes' enthusiasm for it is a result of longer experience with it? It is true that a number of museum conservators have used it when conserving large copper-alloy statuary, but as I've mentioned before, what works for statuary doesn't always work on coins (like electrolytic reduction).
Benzotriazole is controversial within the museum conservation crowd because of its possible long-term ineffectiveness. The ineffectiveness doesn't seem to result, however, from where the benzotriazole is, but from where it ain't. Its permeation rate into a copper alloy flan is not going to be as effective as other wet-based processes, and while it will form a fairly effective chelate where it is present, if there are significant parts of the flan (containing chlorides) that it doesn't reach, then its long-term effectiveness is compromised. There are other assertions by some conservators that there are more problems with the treatment, but of course there are others that swear by it, and so the debate goes on.

This is one of the reasons why I don't like benzotriazole as a first treatment, and most assuredly not like Hannes suggests it, simply rubbed on the surface for a few minutes. Longer impregnation times would seem to be important (i.e. weeks), and of course you could then have possible complications from the ethanol vehicle as Hannes correctly points out. Besides there being better treatments available for coins, I've never recommended benzotrizole to the list because it is simply not a safe substance to handle (maybe Hannes has better lawyers than I do!) Carcinogenic qualities aside, it's a dangerous chemical for people to handle on a casual basis.

Now, I have used benzotriazole in cases where a flan has had a significant amount of chlorides present, but only as a secondary process. After removal of chlorides by an initial process, I have sealed coins with Incralac (a lacquer containing benzotriazole). This has appeared to be useful in several instances. The mechanics of applying Incralac to a coin, so that it doesn't end up looking like a cough drop, are a considerable challenge. I usually use an air-brush specifically adapted for the Incralac, but please understand that there is airborne benzotriazole as a result (i.e. kids don't try this at home).

I'm going to stick with my original recommendation that people will still do best by using a sesquicarbonate treatment for treating chlorides. With regards to patina loss, the least-popular answer is, as usual, the correct one. The possibility of loss of oxidation/other layers from a flan from any treatment depends on the composition of those layers. That means a 5% sesquicarbonate solution might do nothing to the visual appearance of one coin, and a 1% solution might strip three-quarters of the layers of another.

If you have access to "Studies in Conservation", please look at Ian MacLeod's seminal article "Conservation of corroded copper alloys":
A comparison of new and traditional methods for removing chloride ions" (vol. 32 - 1987, pp. 25-40). This is the article that prompted me to pursue this particular line of insanity, and to me remains (along with Plenderleith's volume) the two best introductions to the subject. There's much that has been published more recently, but they are a good start. As I mentioned before to this frustrated list, there is never a single answer or approach to all conservation issues, and sometimes we just have to be patient and think things through.

I have two main areas of information that led me to my comments. The first is a study group of coins I put together around 1991 or so specifically to test out the various methods in MacLeod's article. The big weakness in my test group, as I now painfully understand, is that the alloy composition is not consistent throughout the group, rendering results of mixed value (of course, who has twenty or so identical flans all with bronze disease, anyway? Well, aside from Turkish museums...)

All coins had chlorides removed with sodium dithionite, and tested negative for chlorides (well, below the detection limit I could measure)

One of the coins in the group then received only a cursory coating of benzotriazole - sort of as a surface sealant.

In testing the coin for chlorides every two years or so subsequently ( I haven't done it lately), I noticed by 1998 or so that I got a chloride reading above my detection limit (~25ppm or so). Now, does that mean the benzotriazole failed or did I do a poor job of removing the chlorides in the first place? Also, other coins with other surface treatments didn't do well either, so all my test really shows is that benzotriazole failure is simply one possible cause.

The second piece of information I had was a portion of a working paper testing benzotriazole on shipwreck items ( I don't have the paper, I think it was from somewhere within the Texas A&M facility, sorry for no information), which raised questions similar to MacLeod's comments. I don't think I ever saw it published, though, so take it for what it's worth.

Clearly, a treatment process like benzotriazole is head and shoulders above most of what's currently practised out there, so I shouldn't sound too negative about it.

I tend to be rather cautious about this sort of thing, and don't want to lead people to think any one treatment is some sort of panacea for everything. If it were less dangerous to handle, I might suggest people try it on some coins as well.

Oh yes, the effect of sesquicarbonate on "patina". I agree that a 5% solution will, if not completely remove the oxidation layers, will certainly change the aethetics of a coin! The problem is that 1% solutions are in most cases ineffective, being functionally almost no different from simply using distilled water (for three to five years!) There's never a simple answer to these things, and I think that is what fascinates (and frustrates) so many of us - sort of a metallurical version of golf.

Ah, the scourge of declining mental capacity. I mispoke in my reply to Hannes when I said all coins were treated with sodium dithionite first. Not quite half of the test coins were for the purpose of testing sealants after treatment, with Incralac, benzotriazole alone, various microcrystalline waxes and some sort of polymer I forget the name of (doesn't matter, it was definitely cough-drop material). The rest of the group were unconserved coins to test out the treatment processes; if you didn't mind losing the "patina", the sodium dithionite was the most effective, with various forms of sesquicarbonate coming in second in my highly subjective evaluation.

I never have had good success with citric acid (MacLeod sort of liked it), and please don't ever handle acetonitrile if you don't know what it is. Interesting experiment.

Of course, around the same time (1991 or so) I coated several identical copper alloy blanks with "Deller's Darkener"; if you remember the thread on it, it's basically some sort of sulfide compound in a petrolatum base. At last examination, the solid black, soft, absolutely disgusting surfaces are continuing to profilerate, and I suspect that any year now they will grow arms and legs and take their leave of me.