In the whole debate about whether or not acetone can improve your fuel mileage, there is some conversation about whether or not acetone will damage your fuel system, particularly seals, hoses, and gaskets. I think everyone who comments is really just speculating, and I am going to chime in on the basis that I have no idea whether or not damage is actually occurring. However, there are several, potentially mistaken although somewhat logical, hypotheses being floated around. Here I’m just addressing what those ideas are and why they aren’t necessarily reliable.
Here, off the top of my head, are the 4 things that I’ll focus on.
1.) I’ve used acetone for years and never had a problem.
2.) I put fuel system parts in a jug of pure acetone for an extended time (years?) and there has been no problem.
3.) Gasoline is a solvent. Acetone is a solvent. Gasoline doesn’t destroy a fuel system, ergo acetone won’t, either.
4.) Acetone comes in plastic bottles, ergo it doesn’t destroy polymers.
Let’s address these items:
1.) One person giving their experience is just an anecdote. Even if we assume they are telling the truth as they see it, the result isn’t rigorous. I seriously doubt they took their systems apart, analyzed and measured all the components, reassembled them, used the gas/acetone mix, re-analyzed everything, and then come to their conclusion.
To say that years of experience have led to no problem is, in my view, like saying “well I have smoked cigarettes for 20 years and I don’t have cancer, therefore smoking doesn’t cause cancer.”
2.) Dunking parts in acetone for any period of time is only of limited value in the discussion. The main problem I have with this experiment is that it is a comparison of a static system with a dynamic system. In any textbook or lecture on chemical reactions or mixture formations, there is the explicit statement or underlying assumption that the reaction/mixture is somehow well stirred. The motion can have a serious effect on the outcome of the reaction/mixture.
As a common example, consider iced tea. If you have iced tea and add sugar, you find that the sugar sinks to the bottom and doesn’t dissolve if left alone. As such, it may take hours or days for the sugar to dissolve. Or dissolution can be achieved in a few minutes by stirring the tea after the sugar is added.
Other factors in this situation may be temperature, or the existence of other solvents. Higher temperatures often speed of reactions. I think it’s safe to say that a fuel system and high engine components operate at a temperature higher than the ambient temperature of someone’s garage. And I could also hypothesize that acetone, in combination with other components of gasoline could accelerate a negative reaction.
3. & 4.) Items 3 and 4 are closely related, so I’m going to group them together. A little background in polymer chemistry is needed to understand why these ideas aren’t necessarily correct. Fortunately I have some experience with polymer chemistry and am somewhat qualified to comment on it. In particularly I’ve worked with poly(dimethylsiloxane), or PDMS, which is a silicon based synthetic rubber. You are probably familiar with in some of its uses as it is a primary ingredient in some caulks as well as silly putty. Given this experience, my examples will relate solely to it, even though it probably isn’t in any fuel systems.
So here’s a little polymer chemistry:
- When it comes to a particular polymer, not all solvents are equal.
- A good solvent doesn’t necessarily dissolve a polymer. If the polymer is cross-linked, the solvent will cause it to swell.
- A good solvent may take up to 24 hours or more to fully swell a polymer (although some solvents can act in a few seconds).
To speak very generally, when a polymer and a solvent have what might be called “similar properties” the solvent tends to be a good one for that particular polymer. Other solvents may still work, but not nearly as well.
For example, acetone is a solvent for PDMS but not a very good one. A solvent can be measured by the amount of swelling it causes. For PDMS, triethylamine is a very good solvent which causes swelling of about 40-50%. (By swelling I mean that any linearly measured feature increases in length by 40-50%, thus the volumetric swelling may be 96-238%). Acetone, on the other hand, causes swelling of about 5%. The amount of swelling due to acetone is easy to miss. What’s more, PDMS tends to de-swell very quickly when removed from its solvent bath. Thus it can be very difficult just to measure the amount of swelling.
Let’s take this information and re-consider the case of putting parts into a jug of acetone. If you performed such an experiment and looked for parts to completely dissolve, it isn’t surprising that they didn’t. It also isn’t surprising that someone might think to look for some small amount of swelling. Furthermore, when items are unconstrained, a little bit of swelling won’t cause any problems. That might not be true if they are constrained.
Therefore, if a part of the fuel system is swelled by acetone (or some other solvent) the problem would likely arise in areas where there are physical constraints such as clamps on the end of a hose or some gasket. The problem is that the polymer will want to expand in a place where it can’t. This can cause excess stress within the polymer that could possibly cause it to rip itself apart. Researchers at Harvard, adhered PDMS to glass, then put the assembly into strong solvents. They noted “the glass inhibits the PDMS from reaching equilibrium swelling, since glass itself does not expand or swell in the presence of these solvents. The PDMS is therefore under stress; stress is relieved when the polymer deseals from glass, a times even breaking the glass or tearing the PDMS.”
Having said all that, I figured I would look at some numbers. Above when I said that polymers and solvents need to have “similar properties,” I was referring to what is called the solubility parameter. When the solubility parameter of the solvent is similar to that of the polymer, the solvent usually acts as a good solvent. [Some discussion here] The solubility parameter of acetone is ~9.8. Here are the parameters for some other common components of gasoline:
CONCLUSION: Now, I’d like to repeat that this little exercise of mine was to address the logic behind the conclusions of others. In general I consider their reasoning to be substantially flawed. What I’ve written here, are the types of considerations I would make if my job was to determine if acetone could be safely added to gasoline. How exactly proper testing turns out is unknown to me.
Were I to guess, based on solubility parameters, whether or not a <1% concentration of acetone would damage a fuel system, I would say that it wouldn’t while also reminding everyone that it is a non-definitive guess.
 Lee, J. N.; Park, C.; Whitesides, G. M. “Solvent Compatibility of Poly(dimethylsiloxane)-Based Microfluidic Devices,”Anal. Chem. 2003, 75, 6544-6554