DI Water for Humidity Chambers

[PurewaterGuy]

Hi to everyone, new here, looks like a great site.

I have a question pertaining to the level of DI Water feeding Humidity chambers. The chambers require DI Water of 0-2 Megohm quality. In the past I've provided 18 Meg through mixed bed deionizers but now this seems to be too agressive for the new equipment. I know RO water will achieve the 1 meg I'm looking for but is there an alternative to using RO?

 

[PurewaterGuy]

Does anyone deal with this? Or maybe I'm in the wrong forum?

 

[bimr]

You do not have many options. The RO water would surely be suitable. However RO water is probably more aggressive than DI since the RO pH is typically lower than DI water.

The aggressive is probably caused by the oxygen in the water, something that you can not change.

You might try to increase the pH slightly.

Better materials of construction would help as well.

 

[PurewaterGuy]

I don't know of any way to raise the PH in water and get a continued constant. Even the labs I service use buffers to raise the water PH, however they do this on a very small quantity sample. I've used acid injection prior to RO's to bring the PH down. I've used Limestone to raise the PH after deionizers but that again still needs buffering.

This one has me baffled. I have a customer looking to use approximately 100 US Gallons per day to feed his Humidity Chambers. The only information he has given me thus far is he needs a water spec of 1 meg, information he recently obtained from the Chamber manufacturer. Since 1995 I have been providing him with 18 meg withthe following setup for his Chambers: Feed source tap: (1) - 20" 5 micron Polyspun prefilter - (1) 1.2 CFT of Bituminous coal Carbon Tank 8"x44" (2) 1.2 CFT Mixed Bed Deionizer Tank 8"x44" (1) 20" .35 micron Pleated filter -

Any advice or discussion here will be greatly appreciated.

 

[sbush]

Replace the two mixed bed tanks with a cation tank followed by an anion tank followed by another cation tank (all in series). This won't work without the cation polisher, so don't omit it.

In addition to meeting all of you effluent objectives, if you use the same size exchange vessels, much greater capacity will acheived over the previous mixed bed experience.

When it's time for service-exchange, rotate the existing cation polisher to the primary cation position and put the new cation unit in the polisher position.

S. Bush

http://www.water-eg.com

 

[bimr]

The effluent pH of a DI unit is typically about 10 pH units because the effluent is NaOH. The OH is alkaline.

The pH prior to an RO unit is lowered to prevent scaling. However, carbon dioxide will pass through the RO membrane causing the pH to be lower on the effluent side of the RO, typically about 6 pH units.

The simplest and least expensive method to reduce the corrosivity of water is to raise the pH. That is done with an alkaline agent such as Soda Ash. Since the water is pure, only a slight amount of alkaline agent is required.

If this humidity chamber is open to the atmosphere, the water will also absorb CO2 from the air, causing a lowering of the pH.

Corrosion is also caused by oxygen in the water. You may think about adding an oxygen scavenger.

You have not said what materials this humidity chamber is constructed from. Is further information available?

 

[sbush]

Bimr,

Are you sure about the pH of deionized (DI) water being about 10 units, because my experience has proven that the pH of 18 meg water is neutral. When DI water is exposed to the atmoshpere or comes into contact with other foreign materials, then pH will change as electrolytes are introduced into the otherwise "pure" solution.

I've been told by chemists that the pH of DI water is extremely difficult to measure accurately and therefore resistivity or conductivity are the only accurate measurements of purity. When I have seen people attempt to measure pH of DI water, the readings are always low. This is due to the carbon dioxide that is quickly introduced when the samples are exposed to the atmosphere while being tested.

I'm no expert in lab analysis or water testing, so I could be wrong about this.

S. Bush

http://www.water-eg.com

 

[sbush]

Bimr,

Oh. I see what you're saying...as sodium leakage occurs, in theory, pH will rise.

S. Bush

http://www.water-eg.com

 

[PurewaterGuy]

To the best of my knowledge the chambers are closed chambers. It has been my experience as well that the reading of PH in 18 meg (purewater) is extremely difficult to obtain a reading on due to absorption of co2 from the air. I have sent a email asking for the material makeup of the chambers (although I believe they are stainless but not sure of what level)

This is what I'm dealing with for information as stated in their reply from my last email. (I have removed their names
...........................................................
...........................................................
We think the 1 meg number is reliable information, since the source was a company that rented some chambers to us, and they advised us what the chambers need. We're no longer using those rented chambers, but the chambers we own are very similar. In any case, your research seems to confirm that it's right. We don't have any better information.

Concerning PH, we know even less. We simply want water that is not too "aggressive". We may want to stay with a setup similar to what you did for us previously, so we have some filtering out of chlorine or other chemicals that might act aggressively.

(name removed) will make the decision on how we proceed. He may add to or revise my comments in this email.

 

[PurewaterGuy]

This is a response to the question regarding the Chamber material:
...........................................................
...........................................................
the chamber material is stainless steel. The drawings don’t specify type or finish.
...........................................................
...........................................................

So, not knowing the material is making it rather difficult to properly quote the right materials. How should I proceed?

 

[bimr]

Leakage from DI units whether they are cation or mixed bed units is always sodium since sodium is the most difficult cation to remove.

In water, the cations and anions always are balance. If you have 100mg/l (as CaCO3) of magnesium and sodium, you must have 100mg/l (as CaCO3) of anions (Cl, MO3...)

What happens in a DI unit is that the sodium leakage causes OH leakage from the anion resin, with the resulting leakage being in the form of NaOH. The OH ion causes the slight elevation in pH. This should be obvious since NaOH (caustic) is a commonly used alkaline chemical.

For similar reasons, as you correctly stated, as CO2 is absorbed from the air, the pH should drop to a 5-6 pH unit range.

CO2) aq + H2O == H2CO3

The free hydrogen cation H2 causes the pH to drop.




This is somewhat academic since as you correctly state that DI water is unbuffered so that a slight amount of acid or caustic causes a large change in pH.

 

[bimr]

Stainless steel such as the 304 grades is typically used in DI water service without problems.

What is the type of corrosion that you are having?

 

[PurewaterGuy]

In most stainless applications I use it's 316 SS Electropolished passivated.

I'm not having any corrosive problems at the moment. What I'm up against is what I stated previously. The customer is now looking for 1 meg water as opposed to 18 meg water without justification, other than what they were told by a previous chamber manufacturer they rented chambers from. I have done some research on this and found it true that most if not all Humidity chamber manufacturers are stating 1 meg. BUT, like I've asked the customer, Why were we employing 18 meg for all these years previously? The chambers have shown no adverse effects by using 18 meg water. I know this due to the fact they are currently installing these same chambers in their new facility. (The company has moved and is using the same equipment)

So, in the end it comes down to the old saying "The customer is always right" ,,, and I want to give them what they want.

Recap - The customer needs 100 US Gallons per day - 1 meg quality - feeding Humidity Chamber Stainless Steel -

PH range - They don't know - unknown?

Past System feeding chambers:
Feed source tap: (1) - 20" 5 micron Polyspun prefilter - (1) 1.2 CFT of Bituminous coal Carbon Tank 8"x44" (2) 1.2 CFT Mixed Bed Deionizer Tank 8"x44" (1) 20" .35 micron Pleated Post-filter -

Recommended changing of system by Sbush:
Replace the two mixed bed tanks with a cation tank followed by an anion tank followed by another cation tank (all in series).

My questions:
What is the expected meg quality? What is the expected PH range? What type of meter should be used following the Post-Filter for reading meg quality? (such as the Myron L meter)

 

[sbush]

PurewaterGuy,

This technique is sometimes called "the poor man's mixed bed".

You will most likely acheive a range from 1 - 5 megohm depending on the influent water chemistry.

You should install a meter or quality light after the anion unit and prior to the cation polisher. This will determine the need for service on the primary cation and anion tanks. The service setpoint should be whatever you normally use for separate bed systems (100K is pretty typical). If the customer needs to continuously monitor the final effluent quality, then an in-line resistivity/conductivity meter can be installed after the cation polisher. If continuous isn't necessary, then spot checking with a meter is fine. You might even find a reliable 1 megohm light, if they exist (I'm not sure).

So long as you service the primary tanks within a reasonable period of time, and rotate the cation polisher with each service exchange cycle; then the effluent should always remain above 1 megohm.

I have no idea what the pH of the effluent would be...and I really don't see that it matters. DI water is going to be aggressive at any pH. The lower the resistivity; however, the less aggressive it becomes. As Bimr has pointed out, stainless should be okay.

Does this application involve a machine known as the "Weather-O-Meter"? I've seen these in textile mills. The ones I've seen used a single (.25 cf) mixed bed exchange tank on the make-up water line.

S. Bush

http://www.water-eg.com

 

[GarySCWSVI]

18 meg water is just fine for the application. However, since you feel obligated to give them water they want then you may want to just blend back some raw water to decrease the resistivity of the MB water. Blend it in after the point where the DI water quality probe is so that you know when the MB is exhausting. Install another 1 meg DI water quality indicator after the blend point so that the Customer knows to call you when it drops below 1 meg. Funny way to do it. 18 meg water, again, is just fine. I believe the system supplier requirement of 1 meg water is a minimum requirement. They should not object if the quality is higher.

Gary Schreiber, CWS VI
The Purolite Co.

 

[sbush]

GarySCWSVI,

Blending 18 meg water with raw water to 1 meg target quality: seems like it might be extremely difficult to acheive and maintain. For instance, at a flow rate of 3 gpm/cubic foot of resin and a raw water quality of 100 ppm of total dissolved solids (as CaCO3), what would be the required bypass flow? How critical are changes in system flow rate, pressure, temperature, raw water quality, and resin bed output quality to maintaining accuracy of the blend? How do you stabalize or restrict these potential changes?

I know it's easier in a batch process (e.g., add 1 ounce of raw water to a vat containing 100 gallons of 18 meg water), but I think PurewaterGuy is describing a continuous process with intermittent flow periods in an industrial setting where the conditions of the water supply change frequently.

Maybe I'm way off-course, but blending applications have always been like trying to hit a moving target for me, so I'm always anxious to learn more about them.

S. Bush

http://www.water-eg.com

 

[bimr]

Why don't you just tell the customer that you will supply water with a minimum 1 meg quality and leave it at that.

Isn't that what is referred to as under promise and over acheive?

Your application is too small to justify any extra work.

Blending may be a fine theoretical concept, but it makes little economic sense to try to blend maybe a 1/2 gal per day of water into 100 gals/ per day.

It is not going to make much sense using multiple DI bottles either. Just supply a single MB with the minimum 1 meg quality.

 

[PurewaterGuy]

A lot of good advice here. I think what I'll do is contact a couple of chamber manufacturers. I think I should ask them if the 1 meg is indeed the "minimum" requirement. I'll be back to let you all know how I make out with that.

I'll still have to go with at least three tanks. The carbon tank is necessary to remove the chlorine and control the PH, otherwise the resin tank will take a real beating, organics, bead blinding, chloride, If I use only one tank after the carbon with a quality meter, the entire system will be exhausted without warning leaving them with zero quality water to use.

 

[PurewaterGuy]

I got the chamber info and contacted the manufacturer. Turns out the 1 meg is a minimum requirement. The literature says 1-15 meg although the rep stated 18 meg isn't going to damage the chamber. So, I need a spec just below 18 meg if I want to sell this off to the customer.

Any final thoughts and suggestions just as a recap?