Expert: Dr. Henry Boyter, Jr. - 11/27/2005

Question
Hello,

My hobby or side job is art restoration; there is a problem in art conservation/restoration related to aniline dyes used in Japanese prints of the 19th century. These were synthesized in Europe in the 1850's - 1860's and arrived in Japan around 1870.

The new colors changed the style of Japanese prints, most art critics would agree that it was for the worse because the traditional natural dyes such as indigo, safflower, and turmeric were superceded by "loud" and harshly contrasting synthetic colors.

Typical Edo-period print by Kikugawa Eizan,  published 1815
http://www.datazap.net/sites/saboten7/ukiyoe/eizan.jpg

Typical Meiji-period print by Toyohara Chikanobu, published 1888
http://www.datazap.net/sites/saboten7/ukiyoe/u593.jpg

So I would concur that, aesthetically, this was not a change for the better but there was a fascination with modernization and "progress" in Meiji-period Japan and that was simply what the public wanted to buy - it's like as if color television had just been invented, of course everyone would want to get it.

The problem arises from two notoriously water-soluble aniline dyes, one red and one purple, which had a tendency to bleed and to cause transfer stains from face-to-face contact of prints if the prints were ever exposed to moisture (Tokyo is a high-humidity area similar to Florida).

___________________________________________
"As far as a solvent, water is the obvious choice, BUT that depends on what you want to do with it. Are you trying to save a fabric or ID the dye? If IDing the dye, a more organic solvent may be needed for the analysis that will be used. If you are going to do new dyeings, water is the only choice."
________________________________________________


The solvent is applied to the stain on a vacuum suction table using a brush, pipette, or airbrush, depending on the situation. The stain is removed as the solvent is drawn through the paper by the vacuum and into a sheet of blotter or chromatography paper below.

So, if you have the choice of any solvent or combination of solvents which can be prepared as liquids at room temperature (not limited to the 'OTC' armamentarium) what would you use? I would like to know first from the point of view of theory "What will these dissolve best in" and then secondarily I can consider whether it will actually be a feasible method.

Example: you suggest water as the solvent because it is plentiful, cheap, and safe. But those are only practical considerations which are already known. And since it's known that methyl violet is more soluble in alcohol than in water, I would choose alcohol instead.

Or, I ask you "would aniline itself be a good choice", and you might say "No, because aniline is toxic, expensive, and might leave an oily residue".  Again that would be only a pragmatic consideration that doesn't go to the heart of the solubility question.

I'm aware of what the practical and safety issues are - I guess I am strong on practice but weak on theory because there are a lot of factors which influence solubility, and the very technical parameters like 'Hildebrand' and 'Hansen' solubility are way beyond the scope of my knowledge.

_______________________________________________
"Is your restoration to strip and redye, just redye, recreate, or to clean. It gets to be a very complicated process depending on your goal. If stripping, do you want to destroy the dye and/or prep for redyeing?"
_______________________________________________

There are unintentional transfer stains which are the result of direct contact of one print with another, even if the prints never got "wet" this still would tend to occur by slow leaching of the dyes from one sheet to another. If two prints were stored like that in a folder for 100 years or so it would be pretty much inevitable given the high-humidity environment.    

Regarding "destroying" the dyes, I know that both the red and the purple dyes are susceptible to oxidative bleaching with hypochlorite and to reductive bleaching with sodium borohydride, and sodium hydrosulfite (an ingredient in most commercial 'color remover' products).

In some cases it's possible to just hit an isolated stain with borohydride (if the paper is already very white, because it will bleach the paper too). But that is rarely encountered and it's always preferable to remove as much of the staining as possible by solvation before considering any kind of bleaches.

Here are some sequential photos of a damaged print which originally had severe red and purple transfer stains. It is shown getting gradually better from left to right as I "restore" the print on the vacuum table :

http://www.datazap.net/sites/saboten7/misc/restore40a.jpg

_______________________________________________
"Lastly, why do you think you have aniline dyes? There are different types also, do you know the type (acid, direct, etc)?"
_______________________________________________

They are known to be early aniline dyes because the first synthetic dye of any kind was mauveine synthesized by Perkin in England in 1856.

http://en.wikipedia.org/wiki/Mauveine

So there is a very short list of candidates because there were simply very few synthetic dyes in existence at that time. They had to be in mainstream commercial production in England or Germany during the decade of the 1860's in order to have reached Japan in the early 1870's and adopted as the colors of choice.


My basic chemical testing of these leads me to suspect that the purple is one of the methyl violets and the red is one of the fuchsines.

http://en.wikipedia.org/wiki/Methyl_violet

http://en.wikipedia.org/wiki/Fuchsine

That's my working hypothesis at this point but I don't have access to any real analytical equipment so can't prove it. I found an old paper written by the French chemist Joffre' in the 1880's about how to ID aniline dyes and my own primitive tests are so far consistent with his results. I can you more about the specifics later (example - the red dye really "likes" to dissolve in 10% ammonium hydroxide so this behavior is probably a clue to its identity - it 'likes' strong aqueous base or it 'likes' the amine group, or both)

So, this is getting to be a long email - if the question is of interest to you, next time would it be acceptable to write you directly at your email instead of submitting it through this form on AllExperts?  Everyone that I asked about this before admitted that it was an interesting intellectual puzzle but didn't have a good answer, so that's why I feel compelled to consult "the pros"

Thank you !

Andrew  

Answer
Thanks for the info.  Having the whole picture helps.  I would contact the two experts I mentioned.  I think it will be a matter of trial and error however, because the solubility of the dye may not have anything to do with the solubility of the dye on the weathered print since you may have oxidized dye also and fiber-dye interaction.