Phosphoric Acid for rust cleaning / inhibitor: Mixture and Procedure

[tmassey]

I just got a few tanks back from hydro today. Inspired by a recent thread on tank whips and a post recommending phosphoric acid instead by @rjack321 (and confirmed by @rob.mwpropane ), I decided to give it a try.

It seems that many people are using the mixture and proportions recommended in this post:

O2-Safe Rust Inhibitor for Steel Tanks

So I'm planning on tumbling and cleaning my OMS/Faber LP85s. I've been looking at using rust inhibitor, but I don't really want to spend the $50+ and shipping to get a gallon of Compound O (Diethanolamine/DEA/DEOA). Looking for alternatives, I came across Evapo-Rust and also Cocamide DEA...

scubaboard.com

Seemed like a good starting point, so I gave it a try.

There were a few issues with this recipe, however: this recipe starts by mixing up a 5-gallon batch, and then dilutes it even further, to the point where you could have as much as 45 gallons of the stuff. That's way more than I need or want. No problem: I simply used the description to mix a smaller batch.

It turns out that the textual description doesn't even remotely match the numeric mixtures -- at *all*. Because I followed the text descriptions, I ended up with something about twice as concentrated as what the numeric description would have been. So I decided to sit down and figure out the actual concentrations, and create much simpler mixing instructions. And hopefully my time and effort will help someone else in the future.

In case I made any mistakes, I'm going to break down the math of the original instructions, and then how I put the back together. If you just *love* math (or want to make sure that my information is actually correct), please feel free to review. For all the normal people, just skip this section and go to the Mixing Instructions section below.

Breakdown of original instructions.

The original instructions tell you to make an initial 5-gallon mix using 107 oz of 85% Phosphoric Acid (PA) and 533 oz Water (H20), yielding 5 gallons. This is called the "20% solution". The volumes are correct: 107 PA + 533 H20 = 640 oz = 5 Gal. But there's *nothing* "20%" about this at all. The initial mix is 5 parts water and 1 part Acid (107 * 5 ~= 533) which means the acid is one part in six *total* parts, so it's 16.7% of the total, not 20; and of course we started with 85% acid, so the final acid concentration is lower: 107 oz @ 85% ~= 91 oz of 100% acid; 91 oz / 640 total volume = 14.2%

The rest of the instructions don't give any kind of percentage or anything, just a ratio of "20%" and water: 2:1, 4:1 and 8:1. My guess is he was trying to make it "half strength", "quarter strength" and "1/8 strength", but that's *NOT* what his instructions actually do. These end up being 1/3, 1/5 and 1/9 the concentration of the original mixture (it's 1 part "20%" to 3 parts *total*, 5 parts *total* and 9 parts *total* to determine final concentration).

In the end, it doesn't matter if the numbers actually make any logical sense: it matters if the mixtures do the job they're supposed to do. Seeing as others have used them and been successful, they seem to do their job. But given that I don't want between 15 and 45 *gallons* of this stuff lying around, and given that it's real easy to make a mistake when most of the textual description is either wrong or misleading, I've done the math to create similar concentrations in smaller (and similar-sized) batches without making multiple mixtures.

I've included direct mixing instructions for all concentrations in the original instructions. But remember that the original "20%" concentrated mix is never used directly: it's just used to mix the other concentrations. And seeing as we have direct mixing instructions, there really isn't a reason to mix it that way. But I'm including it anyway, and you can still use the X parts water to 1 part concentrated mix instructions from the original if you so choose.

Mixing Instructions

A note on mixing these. Rather than giving you a volume of 85% and a volume of water, this chart will list a total volume and an amount of acid. The idea is that you will simply top up with water until you get to the final volume. The advantage is that if you have the right size container (like a 1 gallon bottle), you only need to measure one relatively small volume and just fill up the rest of the way. However, please note: you should always add acid to water, not water to acid! To be safe, fill your final container 2/3 full with water, then add the properly-measured amount of acid, then finish filling up with water. And it would be wise to take the proper precautions: eye protection and glasses.

For further information about mixing acids and water, including additional safety details: Adding Acid to Water Safely

Also, in the "Easy measure" lines, the percentage listed is how much the easy measure is off from the exact measurements. Given that none of this is very exact, these should be just fine, and the worst error is only 11%; but if you're worried about it, either go a touch light or heavy where you need to -- or just do it the hard way with the exact ounces.

	Concentrated Mix: 14% Acid (Never used directly!)	Heavy Scale: 4.7% Acid	Mildly Rusty: 2.9% Acid	Flash Rust: 1.6% Acid
1 Gallon: Exact ounces	21.4 oz	7.1 oz	4.3 oz	2.4 oz
1 Gallon: Easy measure	3 Cups (+10%)	1 Cup (+10%)	1/2 Cup (-7%)	1/3 Cup (+11%)
2.5 Gallons: Exact ounces	53.5 oz	17.8 oz	10.7 oz	5.9 oz
2.5 Gallons: Easy measure	6-1/2 Cups (-3%)	2-1/4 Cups (+1%)	1-1/4 Cups (-8%)	3/4 Cups (+1%)
5 Gallons: Exact ounces	107 oz	35.7 oz	21.4 oz	11.9 oz
5 Gallons: Easy measure	3-1/2 Quarts (+5%)	4-1/2 Cups (+1%)	2-1/2 Cups (-7%)	1-1/2 Cups (+1%)

I used a 1-gallon water jug and a Mason jar that has gradations on the side. I filled up the water jug 2/3 of the way, measured out my 85%, poured it into the water jug and topped it up to complete the mixture. Of course, you could use a liquid measuring cup, but I like to embrace my country roots...

 

[tmassey]

Cleaning Procedure

Here's how I used the mixture. My tanks had what I would consider heavy flash rust. Honestly, they were fine without treating them, but I'm setting up to clean some much rustier LP72's that I'm putting *way* too much time and effort into, so I wanted to use this as a bit of a test run. My initial intention was to mix up one gallon total of the "quarter-strength" mixture for mildly rusty, but based on the textual description I mixed an actual 5% acid solution: that means I was really using the heavy scale mixture, so my timing and results may not be too representative. But here's what I did and what I experienced.

First step: put the mixture into the tank. Second step: put a scrap valve on the tank. (I should have used tank plugs, but I forgot until after I started rolling the tank around. Fortunately, it *was* an old scrap valve.) Third step, put it on my wheelie board (see the picture in the first post of the LP72 tank thread linked above). Fourth step: roll the tank around as a poor substitute for a tank tumbling rig. (The wheelie board made this straightforward, but not as easy as one might hope. Turns out a 50-pound combination of steel and water still takes some forearm oomph to slowly rotate even when it's on a bed of wheels.)

Given that I was doing this by hand, there was no way on God's green earth I was doing this for an hour, or even 20 minutes. I did it for 15. Also, because I was using such a small volume of mixture, I wasn't going to get the entire bottom of the cylinder. I was doing this with PST HP133's, and my favorite scuba tank specs chart says they have a water volume of 4.39 gallons, so 1 gallon will only fill it less than 1/4 of the way. That means that even with rotating the tanks I won't get the very bottom or top of the tank. To address that, I also allowed it sit there upright with a bit of side-to-side agitation for about 3 minutes. (None of this will get the very top of the crown of the tank, but I'm going to live with that for these tanks.)

I then put a wide-mouth container in the sink (the utility tub next to the washing machine), and emptied the tank into the container. The liquid was a bit grey (like dishwater), but nothing like the black mess I've seen come out of rusty tanks when I've tried this before. So I will re-use this liquid for the second tank. But first, let's finish this one. I had previously screwed a 5' segment of garden hose onto the end of the faucet. With the acid out, I stuck the hose all the way in and ran straight hot water into the tank for several minutes, until the tank became a little uncomfortable to the touch. It was not too hot to handle, but if you kept your hand on it you knew it. With the tank now rinsed and heated, it was time to dry it. To do this, I had previously taken a 4' length of 1/4" stainless steel tubing, and pushed about 2" of vinyl tubing on the end of it (to protect the tank), and about 5' of vinyl tubing on the other end. That tubing was put onto a shop air trigger blower thingy, which was then hooked up to a BC hose of a regulator connected to a SCUBA tank of air. (I don't have a nitrogen tank, and I'm not getting one just for this.). ETA: Just in case you might misunderstand: Don‘t use shop air. It it Is either oily or wet or both. SCUBA air is specially dried and oil-free, and it’s what‘s supposed to be in there. Remember, you’ll be breathing from this tank when you’re done: don’t mess it up now.

I then rotated the tank vertically (with the hole at the bottom), stuck the vinyl-tipped steel tube all the way into the tank and pressed the blower trigger. I basically had to hold the tank, with the crown kind of resting on the edge of the sink, but still holding a bunch of the weight of the tank. A tank holder thingy would be *very* nice: another project for the future. In any case, I blew clean, dry SCUBA air into the tank. At first, I was getting warm, moist air out of the tank. This lasted for what felt like 2 or so minutes. Then, it all of a sudden switched to cool air. I'm not sure if the air was moist or dry, but it was definitely cool. And oddly, the transition was fairly sudden: warm, then cool. Not a gradual change. I'm *hoping* that means I've got a dry tank at the same time; but to be sure, I blew air through it for at least 60 seconds more, maybe even 90 or so.

So, hopefully at this point I've got a dry tank. I put my scrap valve back on it and laid it down while I did the second tank.

The second tank went just like the first tank, with the pre-used mixture. The rest of the procedure was the same.

Next: haul the two tanks out to the garage. The moment of truth: what do they look like?

They don't look like shiny, bright metal -- of course: we cleaned them, we didn't polish them. But the metal is fairly uniformly grey. There are some patches of white in some places, but no orange to be seen. In addition, the bottoms look just like the walls (so standing them up for a few minutes seems to have done the trick). Also, both tanks look the same: I can't tell which one got the fresh mix and which got the re-used. So that part seems fine, too.

In addition, I see no evidence of moisture anywhere in the tank. I think (or at least certainly hope!) that they are completely dry.

So at this point I put my manifold and bands on the tank, and then flushed the tank three times with EAN32. That's just how I like to handle a tank that's had a valve off: I fill it up to 200 PSI or so and dump it most of the way down three times: then it's ready for the final fill. That way, any extra humidity that gets in from the valve being off is flushed out as well.


Now that I've had a good walkthrough with a set of mostly-clean tanks, I think I'm ready for the real reason I set all this up: my gnarly 53-year-old LP72's. I just finished painting them today, so if the paint is hard enough to take the abuse I'll start this process on them tomorrow. At least now I'll be able to mix up the correct mixture!

 

[tmassey]

I moved on to my project LP72's today and got a chance to see how the phosphoric acid worked with tanks in rougher condition.

I started with the left-over mixture that I had used to clean out the two HP130's I worked on yesterday. This was the 5% solution made from food-grade 85% and water, which then cleaned two previous tanks. After the second tank it looks a bit darker than dishwater now, but it's still rather clean-looking. So I started with this solution and added to it some Klean Strip phosphoric acid I used for a prevous project, diluted to the same 5%. In total, I had 3.5 gallons of (nominally) 5% phosphoric acid, and I put half of it in each of the two LP72's. One went on the wheelie board, and one went next to it. One got rotated about 1/6 of a turn every 3 minutes for 30 minutes, and one got rolled around for about 15 minutes and then stepped every 3 minutes for 15 more: both had 30 minutes of acid time.

I then dumped the tanks one at a time back into my big container and rinsed with lots of hot water. By the way, this is the container I've been using:


I bought it for my wife to sous vide in (she'll do tens of pounds at a time), but it's working *really* nice for this, too. Anyway, after the healthy hot-water rinse, a quick peek in with the light showed a nicely grey/white tank, with minimal streaks of orange. Probably not quite as clean as yesterday's 130's, but *way* better than I would expect a tank to come back from hydro, and there was no fluffy orange at all. (Except Faber tanks: I do love the passivated steel hot-dipped galvanized Fabers. Maybe the perfect steel tank?)

Because I used the Klean Strip, I was always going to do one more round with these with just food-grade acid and water. Given there was only a hint of orange, I decided to go simply with the flash rust concentration: 1/3 of a cup diluted to a gallon total. I didn't roll those around for very long: 7 minutes or so? Then dumped the residue out into my big container and then dried them as per the description above. I only took a very brief look inside: I was in the middle of a lot of stuff at that moment, but they looked great, and seemed quite dry.

By the way, at a glance I'm not sure if I could see the difference between the two tanks: one rolled and then rotated slowly, the other only rotated slowly. There *might* have been a bit more rust on the non-rolled tank, but I'm not certain. I didn't have a lot of time for critical looking, but I noted which tank was the non-rolled one and I'll look more carefully either tonight or tomorrow afternoon.

The reason I didn't have time was because I was in the middle of trying to get the one set of LP72's clean and dry, and a second set of LP72's into the cleaning process. I just picked up a second pair today that were at least outwardly much cleaner than the LP72's I've been working on. I hadn't been able to look inside of them before I bought them, so there was some suspense as I popped the tops off them when I got them home. What would I find?

I found that the inside seemed actively moist, and they were coated with an even, solid patina of orange-brown. Sigh. But no pitting: very nice and smooth throughout. I can't tell if the color I'm seeing is from rust, or if they're epoxy-coated: I've never had the 'pleasure' of coming across any lined tanks yet. So, I'm going to throw some acid in there and see what happens. Well, at least I get to get some more experience with the phosphoric acid cleaning process...

You may have wondered why I was keeping the used 5% acid solution: that's why. I collected the waste acid and put it into the newly-acquired bottles, and then filled them carefully up to the tippy-top with water. So it won't be 5% (probaly more like 3%), and it's already been used, but I'm sure it's got plenty of oomph left, and I'm going to take advantage of it! I'll let them sit for a while, until either tonight or (more likley) tomorrow afternoon. I'll dump them (and discard the solution) and take a look. Hopefully I see some grey metal and not some really clean and 'nice' brown-orange epoxy...

The adventures of old SCUBA tanks continues...

 

[tmassey]

In the above post I linked to a post trying to get some more info on those tanks I started on tonight. Turns out that they're probably PST's made for US Divers, and that according to the catalog, in 1967 the yellow tanks were vinyl-coated on the outside, and epoxy-coated on the inside.

Last I wrote above, I had left them with a reasonably strong acid mixture to sit awhile. When I got home tonight, armed with the new information, I dumped the acid to see what I might learn. It was quite clear that these are lined tanks. So that thing I thought was a moist, rusty layer was actually the epoxy liner. Note that it feels moist even when it's dry: probably because it's smooth.

(As an aside, now that I know what it looks like, I know that the PST LP121's from 1971 I also own are lined too: and here I thought only LP72's were ever lined...)

So I dumped the acid and did a full rinse and dry. That's it for them for now. I'll get them hydroed hopefully this week and go from there. If they pass, they get used as-is: I'm not trying to remove either liner. I'm very confident in the dryness of my fills (I've got 2 x 30-inch filter stacks for a 6CFM compressor...), so I'm really not worried about corrosion on the inside of these tanks. And this is the very definition of why the VIP program was started: to take an annual glance inside and make *sure* that nothing's changed. And that's exactly how I'll handle these.

Thus ends my rust removal journey. So far, I've used phosphoric acid to clean up flash rust from two modern HP steel tanks, longer-term surface corrosion from a pair of vintage plain steel tanks, , and clean (not remove) the epoxy lining from a pair of vintage coated steel tanks. All of it seemed to go about as well as I could hope. The rinsing with hot water until the tank is hot and then blowing out with SCUBA air also seemed to work well. I used a single HP100 cylinder as my drying gas, and it dried all six tanks and still has pressure left in it, which is worth knowing for the future.

I hope this helps someone else to be able to take advantage of phosphoric acid to clean some steel tanks in the future. Now it's time to assemble these guys so I can take advantage of all this hard work!

 

[Johnoly]

I hope this helps someone else..

I can tell you that I'll be referring back to your excellent writeup when my tanks are ready.
Thank you for all the details and exact recipe. Even though not alot of readers may post, I can assure you that they will remember it in the future when doing a search on the key word "rust"

 

[captain]

I use Ospho, it is pretty much a standard in the marine industry.

Ospho

 

[SubseaSports]

Excellent write up.
I have a done a metric sh*& tonne of phosphoric tank treatment.

Yes, it is ideal to have a tumbler to do this, but I have done it both ways.
When no tumbler, a small sheet of plywood propped at a slight angle.
Kick the tank up, and it will come right back...most times. Repeat. After 30 mins, I usually grab a beer.
No more than 50% for wicked tanks and I adjust down from there. I use a measuring cup and only put in enough to cover the bottom of the tank up to the straight wall. I never found the need to put any more that that. This way you are really only dealing with the weight of the tank. Also, you are saving money.
Let stand for 20-30 mins, then tumble. Invert plugged tank if the bell is bad. Typically no more than an hour total wash if you have a tumbler, a little longer if not.

I have used Ospho and Kleenstrip versions. I find the Ospho a tad stronger, but I get a discount at Lowes, so Kleenstrip it is.

 

[tmassey]

To put a bow on this thread for now:

I finished assembling my non-lined LP72's this evening. See them in all their restored-vintage glory...

I examined the two tanks I did, one by rotating it basically continuously, the other by rotating it by 1/6 a turn and letting it sit for 2 minutes between increments, both 1/3 or so full. I could not identify a difference between the tanks: they were both equally and completely cleaned. I had said in the other thread that they weren't perfect, but after re-examining them tonight, they really are: there's not a bit of rust anywhere. There *is* a thin and fixed coating of white on the steel. A quick Google shows that this is *exactly* what phosphoric acid *does* to corroded steel. For more details, check out the Wikipedia article on metal pickling.

@SubseaSports suggested kicking it up an inclined board, which would make a lot of sense. I actually used my wheelie board for the 1/6 turns, and just rolled the other tank back and forth on top of a towel on top of my washing machine (seeing as I was right there using the utility tub). Worked well, and I didn't have to bend over like I would if it were on the floor. But both ways worked equally well in the end, with the strength and volume of acid I used.

A couple of people in the thread (such as @captain) mentioned Ospho, and @SubseaSports seems to enjoy a bargin like I do and mentioned Klean Strip. Some mentioned that they simply use it straight (which is 35-45% phosphoric acid). As I suspected, the percentage isn't real picky... I'm certain that all of them will do the job of cleaning the steel just wonderfully -- and the Ospho no doubt will work better because of its "Surfactants, and Extenders" (from the data page), but I worry about their use in tanks I'm going to breathe from, and that might see elevated O2 levels... Others seem to have done it and lived to tell the tale, but for me, I'll use at least a final cycle with food-grade acid and a good rinse. I don't do enough tanks for the expense to be noticeable.

So now you have no excuses for leaving any rust -- even flash rust -- in your tanks. You have the exact materials, recipes and procedures you need to get those tanks sparkling clean -- or at least flat-gray clean. Have at it!

 

[captain]

Thanks, I was put on to Ospho by a charter boat captain who swore by it. I do do a hot water rinse after the Ospho has dried.

 

[happy-diver]

Magnificent thread the best of the best!
I use phosphoric for everything except teeth unless they're false or yours
No more vinegar for me either
Ha ha ha ha ha ha ha ha!