Packard Forum Index » General

do you see an aqueous phase in all the ones to which you added water? 1% is well over the solubility of water in most hydrocarbons (typically 100-300 ppm if I recall correctly), so they should all be at equilibrium.

Last weekend my neighbor asked me to look at his pressure washer which would not start. After taking the carb apart found that the brass jet was corroded to a nice green color and plugged closed after cleaning with small wire all brass parts it started on first try. He had drained all the gas the last time he used it. Must have been small amount in the jet and tube leading to throttle bore. I assume that this is a problem we will have with the new gas in older carb parts not designed with alcohol in mind

Only added water to one. Ethanol and water are infinitely soluble so the ethanol solubilizes some of the water and vice versa so I wanted that 2 phase effect. Copper needs water to corrode. The copper is in that "aqueous" phase, not protected by the organic, that's why I added so much. There should be more water in the organic phase than normal(saturated).
The aqueous phase is colorless. So the yellow is not water nor ethanol soluble. Without analysis can't tell what's causing, but could be just the hydrocarbon/base stock color.

Brass should not corrode in tap water and often used in gas fittings, that's why it's a control, it's only seeing what's in the gasoline.
This test may take a long time to produce any results, if at all. That's why I mentioned I might have to expose to air, if no observations capped. Even air may not produce negative results as the oxidative potential will be changed - color from ions is sometimes oxidation state dependent.

I noticed today all the samples with metal and the E10 control have darkened slightly and equally. ie,deeper or darker yellow.
The ethanol free control did not darken and the ethanol free with copper did not darken as much by only slightly. This is subjective eyeballing, not colorimetric analyses. So there's some photochemical and/or oxidative process at work. This glass jar static test won't mimic a car in heat history, light exposure, or liquid velocity.
{It's been quite hot, mid90s, here this week. Garage is uninsulated, so maybe slightly hotter during the heat of day.}

I found a study*pubs.acs.org/doi/ipdf/10.1021/acs.energyfuels.7b01682 that looks at all ranges of ethanol and steel, brass, copper and 304 stainless. Only stainless 304 was resistant to the highest corrosive mixture(E60+6%H2O). Corrosion rate as about equal (~40 micron/yr) at that concentration for all non-304 samples. However, evidence of pitting corrosion is shown in some samples. The corrosion rates of dry low ethanol mixtures is quite low, ~3 microns/year, as suspected. {Brass<steel<copper<aluminum}
What I didn't like about this study was they added water up to the soluble amt for each of the blends 10-100% ethanol, ie [water} is variable in some but constant in others. The trend was clear, if you get water in your ethanol gasoline system, it starts some corrosion.
So keeping water out is key to protecting the fuel system when employing ethanol fuels. That means draining and drying on storage(long term) and turning over the fuel(driving). Using corrosion resistant materials, not copper or aluminum tubing.
* - requires ACS membership to view for free. I'd post, but copyright.

Update:
The ethanol ones continue to slowly darken.
The ethanol control(no metal) is the darkest but only slightly.
The one with water has formed some fines that accumulated in between the phases, what we call a "rag" layer. I suspect* they're partly reduced(Cu+) copper oxides/hydroxides of some mixture. When I shook it up, the gas was very cloudy and remained so - stable emulsion, majority of water slowly separates but more stays with the gas now. The water layer picked up a gold yellow color.
Ethanol free control remains as it started.
Not much change in the ethanol free w/ copper. Slightly more color than control.

Early result/conclusion: If you store with ethanol gas in system, make sure you get no water in and/or better yet, fill up with non-ethanol. It appears to be much more stable.
ps: I'll post some pictures soon.

* - I need to check the lid underside to make sure it's not corroded.

I shall be most interested if either of your copper combinations turns neon green.

I suspect copper plus oxygen may have been involved as the green color did not show in the fresh gas I put in the tank, only once it had passed through the copper line. Perhaps as the fuel gradually evaporated from the lines these elements reacted with the heavier organic molecules, the latter having remained un-evaporated longest. Accordingly, when I ran fresh gas through, the residue was dissolved, hence the neon green color in sediment bowl.

PS - I have good reason to believe the engine had not been run since before the widespread use of ethanol in gasoline which began in the late '90s as I recall. It is quite possible the old gasoline contained MTBE.

From a chemistry standpoint, there's many green colored copper compounds but not that many that you'd ascribed to organic corrosion of a copper fuel line from gas. Hydration also affects the color, so it well may have just been evaporated empty, exposed to air, and water/humidity got in it. Then the gas flushed the surface carbonate or oxide hydrate compounds into the fuel bowl. It was pretty cloudy so I think water was a factor.
MBTE - methyl tertiary butyl ether is fairly stable but also contains oxygen, but in a form more stable than the hydrolyzable OH group of the ethanol that also hydrogen bonds to water.
You're also talking sitting around for years, like the copper roofs that develop that greenish patina, it takes decades, so I might not find anything relevant in a shorter period.
I'm using Shell fuels because that's what I had, and as I said may not show the same color but I'm also looking at the stability. I guess I could study how to turn gas green from copper, but I'd have to have a sample of your stuff analyzed to match and that's not probable.
You can do your own test with your own gas and a sample of the old fuel line. Use a small glass jar with tight lid.
If you kept the sample, it could be analyzed for copper very easily but the exact color compound(s) is a bit more difficult to isolate. Several methods would be employed.

Pictures as of today.
non-ethanol: Std on rt., copper left
std has developed a slight greenish tint and maybe darker.
copper made darker.
Ethanol; l to R: - w/ copper, w/ water, std w/ brass, std.
Water shows the turbidity of the shaken sample. It settles out but slowly.
All much darker than any of non ethanol

In Sept Hydrocarbon Processing article on testing gasoline for corrosion. I can't post it, it's copyright..
They were testing butanol as a replacement. (Refineries struggle with C4 cuts because of VP stds. Normal amt as butane is about 2% as an ignition promoter)
Data showed rate much higher (2x) with water above solubility.
They report bio-degradation in oxygenated gas.
ASTM test method D130 for corrosion, similar to what I'm doing but iron.
Interestingly, MBTE had increased corrosion rate.

Attach file:

---

  (101.89 KB)


  (173.77 KB)

Could it just be a case where the older copper line oxidized internally when the fuel line had dried out from sitting, and then the application of new fuel later started to dissolve that internal copper corrosion and tinted the gas?

If so, then just submerging the non-corroded copper sample in fuel may never replicate the corrosion that the copper would see in a non-submerged environment.

To me, the test would be to take a piece of heavily corroded copper and place that into modern fuel and see if that causes the fuel to dissolve the corrosion and tint the fuel.

Perhaps this was already one of your scenarios that I just missed in the thread above.

Been a while.
Each brain will rationalize something to explain observations to itself. How do you design an experiment to replicate something that no one really knows what happened? Soft copper is also not a recommended fuel line material normally. I hope we agree that what he had was no longer "gasoline".
Copper oxides aren't soluble in gas normally, certainly not enough to cause color. They may have some in ethanol(?), but EtOH is soluble in the gas until it takes on enough water to partition aka phase separate. There's also other stuff in there they don't disclose. Wet ethanol gas is corrosive to a degree as I showed in the study summary.
I looked at the jars tonight and it's been around 30-50ish in the garage lately and the ethanol ones are still developing darker color, yellow brown, and some precipitate. I was surprised to see some in the reference, but it disappeared on swirling. (artifact of temperature)
The ethanol free look the same, nice light yellow green color and clear. So there's definitely issues with ethanol gas and storage - even closed in glass, but the guy thought it might be old MBTE oxygenated gas in one of his later posts. I just didn't have any to test.
In order to make some sort of copper organo complex it is possible but I think it would need some acid component to ionize the copper oxide and the Cu oxidation state would determine the color. Not many of them are bright green. In my experience, bright green is more indicative of nickel, but let's not go there. It would just take a quick UV/vis and IR analysis of the green gas to tell more - info we don't have. I'm retired out of the lab for 10+ years now - no access either.
I urge anyone to try their own tests if they think they know more.
It's merely a passive exercise for me in visualizing what I had gleaned from the literature rather than a rigorous test. Much hype and mythology in this subject. Like I surmised early on, it's a slow process. So for the most part, if you use it quickly, not much issue, but they don't/can't formulate for you to store in every possible manner in your classic for long periods and remain perfectly like new... I doubt advertised stabilizer additives would do much, but what do I know? I bought some once decades ago for one of my boats that was sitting awaiting a buyer, and I think it's still on the shelf and the bottle has kinda shrunk. i need to throw it out.