D.I. Water

Here's all I know. Most ready to use antifreeze is a 50/50 mix of antifreeze and DI water.

My assumption is that it would be ok, but I'd wait for an expert to drop by.

My instinct says you should be just fine. But, Like Lyn, I'm not an expert in this field.

>Most ready to use antifreeze is a 50/50 mix of antifreeze and DI water.

A major brand who also private labels other automotive antifreeze brands in the US does not use DI water for the 50/50 mix. The water blended with the ethylene glycol (and additives) for 50/50 is city water.

It's been a few years but if I remember correctly, We used DI (Deonized) Water to cool Lasers used to engrave serial numbers in steel casings. We never used automotive antifreeze, but purchased industrial glycol in 50 gallon drums. It was not diluted. I believe that we used propylene glycol rather than ethelyne glycol because it is non toxic and breaks down in the environment in a short period of time. We never had an issue with pipes, tubing or other items in contact with the DI water. I do remember checking the cooling water periodically to be sure it never exceeded a predetermined level of electrical conductance. If I can find any remaining data that could clarify this, I will return and post it. Hope this helps at least a bit.

http://www.keytometals.com/articles/art160.htm

You might want to check this site out. Metal in contact with water, especially when there are different metals (because of different potentials), undergoes galvanic corrosion. Glycol may prevent or slow this but DI water is a strong solvent- do you have access to a materials scientist? Does the manufacturer of the MRI provide recommendations?

From the link above:

"Galvanic processes occur between different metals and between different areas of the same metal in the water environment. Water is an electrolyte, a poorly conductive one at the low dissolved solids content of fresh waters, and a highly conductive one at the high dissolved solids content of sea water.

When two different metals are immersed in an electrolyte and connected through a metallic path, current will flow. Oxidation occurs at the anode and reduction (normally oxygen reduction) occurs at the cathode".

"Galvanic corrosion occurs between weld metal and base metal, between different areas of the same metal and between different metals in water."

The glycol concentration may need to be greater than 30%

I worked in materials research and Lynn is right. DI is no longer DI once it comes into contact with a metal like copper or iron. There will be ion transfer and corrosion if you use both materials in your plumbing. I would limit the material to one type of pipe, copper or iron, or use Stainless steel throughout.

The glycol can buffer the reaction but will not prevent it. Iron may have a bad effect on an MRI machine too. It's magnetic, and you are ruining your DI water by passing it through Iron pipe! This may cause loading of the MRI.

Check with the manufacturer of the MRI machine. I'll bet they specify copper only or stainless only.

DI Water, has had all the free Ions removed. It's those pesky little Ions that do all the damage. DI (in theory) would not cause any corrosion, in fact you can take your computer, leave it on and plugged in, and go ahead and submerse it completely in a tank of (pure) deionized water, and guess what will happen? if you guessed nothing, then you'd be right. People have submersed their motherboards in tanks of DI water to keep them cool, and they have done this successfully. Now, on the other hand, if you tried this with any other form of water that does contain those free radical Ions, then you would be in trouble. Deionized water is NOT electrically conductive, and therefore will NOT cause Bi-Metallic corrosion. (until it has been contaminated with foreign particles which do conduct electricity.)

I'm not saying you wont see corrosion eventually, but in the initial few seconds the DI water remains pure (before atoms of other substances are dissolved into the water) it won't directly cause the corrosion (although inevitably it will occur). But in the end, you cannot keep the DI water from becoming contaminated, therefore the other mixtures, and corrosion preventatives will save the day.

But in the end, technically speaking, DI water is NOT an electrolyte until it has been mixed up with other things that will conduct electricity.

RVZ717

Although galvanic cell does not form in a poor electrolyte like DI water, I do not agree with this statement at all. DI water is devoid or very low in conductivity but it is considered very aggressive. If you look up any of the saturation indices calculators available on the web and enter values for DI water you will find that the water will be very aggressive. DI is under-saturated in minerals and as a result will be corrosive to all metals it comes in contact. A computer will be destroyed by such water and relatively quick. I did a calculation using a conductivity of 1μS/cm, hardness of 1 mg/L ,d Alkalinity of 1 mg/L and temperature of 0 degrees C. If you use the calculator indicated it will be minus 5.5. That is very aggressive water.

As a guide to using the index consider values less than zero as aggressive, greater as scale forming, and zero as neutral. Anything less than -1 is considered very aggressive. At -5.5 the water is extremely aggressive. Conversely, water above 1.0 is considered very scale forming and if it was at +5.5 that would be extreme scale forming. Most water would be in a range of -0.5 to + 0.5 and even then can be problematic at the upper or lower ends. You are correct it is not galvanic corrosion.

Hey, I have no calculations to run, but I have seen with my own eyes, a motherboard suspended in a tank of DI water, RUNNING, and running well. It was left that way for quite some time, plugging away at it's calculations the whole time with no ill effects. I cannot say what effects it may have long term, but I know people use DI water in their water cooled computers as well. It was just a thought.

Water in any form is the universal solvent.

I am definitely no chemistry expert, and most likely should leave topic like this alone to avoid the foot-in-mouth syndrome.

No, it won't. Corrosion resistance is achieved by having wetted materials that are compatible with the fluid in question.

Old salt and PWSlack have both given good answers. You may want to consider propylene glycol from Nu-Galgon. It seems to have all the qualities you need. (not sure of price)

Dielectric unions where copper meets iron will help reduce electrolysis, whatever conductive coolant you use. One possible option would be transformer oil, such as Shell Diala AX. This hydrocarbon oil is an excellent insulator, as well as heat transfer medium. The flow rate for a given heat flux would be greater, as the specific heat is lower than water/glycol.

Your belief is spot on about mineral depleted water. I have experienced piping failure in a centerless grinding operation. The grinder coolant was mixed to proper concentration w/D.I . water . Within 6 mo. of the test start,we had massive Cu pipe failures.

A bit of a bad choice of words - purity content? That's almost an oxymoron. There is no content. That's is the reason as you stated for the aggressive nature of the DI. Pure water with no ion content is not happy. As soon as you add the glycol, there are ions and the DI is no longer DI. We build many heat exchanger skids per year, and most of these are run on a glycol/DI water mix through both carbon steel and copper tubing. DI also creates other strange problems that is not associated with water. Any conductivity based sensor, such as a point level sensor, or magnetic flow meters do not function with DI. We caught ourselves the other day trying to put a water sensor in an area where DI might leak. It was based on conduction between two points. Will not work.

Since you are using this in a chiller, the effects should be slow, but glycols break down at higher temperatures and form acids in closed loop systems. That is why antifreezes for automobiles have inhibitors added to the mix. Also, as elsewhere stated, the water will slowly increase ionic concentration from reaction with the metals and set up a galvanic action. I would find a commercial antifreeze package that prevents (or technically slows) corrosion.

Thanks you all for the input and advice, we are going to use 50% Glycol supplied by the chiller manufacture, the product has inhibitors to buffer the DI.