I referred your question to my friend and chemistry wiz Robert Carter, a painter and reader of this blog. The following is his answer. Robert has the ability to make the complex understandable to laymen like me.
To answer this definitively, I pulled up the Manufacturer’s Data Sheets (MSDS) for turpentine (Utrecht), Gamsol (an odorless mineral spirit, OMS) and Eco-House Natural Orange Turpene #915. I have no experience with the Orange Turpene product, but it’s interesting to see that it is being promoted as a safer solvent that will dissolve damar. The principal ingredient is food grade orange turpene oil, which has an FDA GRAS rating (generally regarded as safe). GRAS is applied to most foodstuffs on the basis of long experience (e.g., spices are on the GRAS list), but there is no presumption of rigorous testing.
The health concerns of any solvent are acute toxicity on the one hand, and long-term health risks on the other. On both counts, turpentine is definitely the worst, Gamsol (and other OMS products) are better, and Orange Turpene is the (presumably) most benign. As the Merk Index notes, turpentine is absorbed through the skin, lungs, and intestine. It causes acute skin and mucous membrane irritations, skin eruptions, gastrointestinal irritation, delirium, ataxia (loss of muscle coordination), kidney damage, and coma. Inhalation causes palpitation, dizziness, nervous disturbances, chest pain, bronchitis, and nephritis (kidney irritation). Chronic contact can cause benign skin tumors. All in all, we’re better off not using it, except when necessary (e.g., damar-based mediums). OMS has similar risks, but they are generally regarded as less acute. In part, this is because it is less volatile (has a lower vapor pressure) than turpentine, so the build-up of vapor in the studio over time is lower. Certainly the narcotic effect is less. But among OMS products, there is a great deal of variation in the composition of the hydrocarbons present. Gamsol’s claim to superiority is that it is very low in aromatic hydrocarbons (less than 0.02%), which potentially reduces the long-term safety concerns. Chemists these days are very concerned about the use of volatile organic compounds (VOCs) in synthesis and manufacturing, but most especially about aromatic hydrocarbons. Aromatic hydrocarbons are composed of planar, six-member rings of carbon atoms, with benzene (C6H6) being the parent compound. Years ago, we thought nothing of doing reactions in benzene, but today you have to jump through hoops just to buy a bottle. The problem is that benzene is carcinogenic. So anything that minimizes benzene and other aromatic compound content is preferred. The Orange Turpene product, having ingredients on the GRAS list, is presumed to be safe. But we should throw in a word of caution here – GRAS just means we don’t know of any problems. Just because something is natural does not mean it is necessarily benign. In the absence of data, one should still take reasonable handling precautions. Beyond this, the only other thing I spot in the MSDSs is a difference in the flash point temperatures: 91o F for turpentine, 113o F for Orange Turpene, and 145o F for Gamsol. This is the same order as the boiling points. It means that the potential of starting a fire in the solvent is least with Gamsol. But this is really a minor concern.
The principle to apply in evaluating for longevity is that simple is better than complex. The process of forming a paint film (“drying”) is actually an oxidative polymerization process. Polymerization is the joining together of smaller molecular units (monomers) to make an infinitely large structure. The strongest polymer, and hence the strongest film, would be formed from a single vehicle (e.g., linseed oil), because only the same kinds of monomers would be joined. With a variety of alternative monomers present, the network building frequently ends in dead ends or less strong linkages. Now that’s the theory, but in practice it may not make a big difference. For example, white paints frequently have a mixture of linseed oil and safflower oil to reduce the yellowing tendency, but I do not know of any data that says these make weaker films than a white mixed with, say, pure linseed oil. So, how does this apply to solvents? Well, we need to think about the residue they leave. Looking again at the MSDS data, Gamsol is 100% volatile, turpentine is 99.5% volatile, and Orange Turpene is 99% volatile. The latter two, then, leave a nonvolatile residue in the paint. We have hundreds of years experience with turpentine, so we know that the oily residue it leaves does not interfere with the polymerization of linseed oil, and may actually be incorporated into the film. At least in theory, Gamsol should be even better, because it leaves no residue to interrupt polymer formation. In other words, your questioner is misinformed to think OMS compromises longevity. The Orange Turpene leaves 1% residue, so the question is what effect if any does it have on the strength of the film? The manufacturer claims that this material is archival, but it is a new product that does not have the lengthy record of turpentine, or even OMS. They may be right, but they could be wrong. As far as I know, orange oil (which is 90% d-limonene) is not a drying oil, which means it does not readily polymerize on exposure to oxygen. If that is the residue, it could be a problem. But to be fair to the product, I am speculating here.
Speaking personally, I tend to be very conservative about materials with respect to longevity. (By contrast, painters like Fairfield Porter loved to mess around with odd materials, and now their paintings are employing legions of restorers trying to hold them together.) I avoid turpentine, except when necessary, only because of the health issues, but certainly not because of longevity concerns. I prefer OMS because of health concerns, and I am confident it poses no compromise to longevity. (As an aside, if someone with the technical expertise of Robert Gamblin doesn’t have a problem with it, then I don’t.) Personally, I would be reluctant to take a chance on this new orange product. I know, for example, that Turpenoid Natural should not be used as a painting medium solvent, so I guess that prejudices me on this stuff. Maybe I’m just reacting to the knowledge that if I really wanted to mess up my paint film, I’d add orange oil. That’s like adding Goo Gone. I guess if you like the product and you’re willing to take the manufacturer’s word on its archival nature, then go for it.
Hope this helps.
Robert L. Carter, Chair
Department of Chemistry
University of Massachusetts Boston
Thank you Robert, That was great!.....................Stape