Corrosion Control: Calcite for RO Permeate

[wright44]

Hello,

Background:
I am working on a potable project for some folks who are looking to add an RO unit to their potable plant. The plant has an existing RO that was built and permitted some time ago for Arsenic removal. They are adding a new unit to increase throughput as demands are increasing. There are some fussy complications with the project, but essentially we are stepping in to help fill gaps and get it permitted, but to do so we need to validate the design.

The Issue:
Naturally, RO can produce some aggressive water and we anticipate this one is no different. Due to the current spotlight on Lead and Copper rule violations around the country, our regulator is very concerned about treatments that lead to aggressive water. The proposed RO unit utilizes a Calcite filter to attempt to raise pH and adjust the final alkalinity. The water sits in a storage tank open to atmosphere before final delivery, so I also have to assume there will be an equilibrium with air before distribution. The issue I am having is properly validating that the calcite filter is sufficient.

I have been using this spreadsheet to try and simulate various water conditions and the effect it may have on the finished water quality. Based on what we know, the permeate from the RO will have incredibly low alkalinity and calcium (nearly 0). The pH typically ranges from 7.0 to 7.4. Water temperature around 20 Celsius. Resulting Langelier Index is -7.27 and Calcium Carbonate Precipitation Potential (CCPP) is -12.6 aka VERY AGGRESSIVE.

From my calculations, it seems like no matter what dose of Calcite (as a function of bed depth) we use will lead to a neutral or non corrosive end point. Hoping some folks with experience on these things could help provide insight. I have seen recommendations to dose CO2, which I simulated and it does lead to a better end point. If that is the only solution, I will need to make recommendation but that will complicate things for the client and operator.

Data

INPUTS

Water Parameters
pre-Calcite pH: 7 to 7.4
pre-Calcite Alkalinity: 0.007 to 5 mg/L as CaCO3
pre-Calcite Calcium: 0.143 to 5 mg/L as Ca
Water Temperature: 20 Celsius

Calcite Filter
Bed Depth: 6 feet
Superficial Velocity: 8gpm/ft^2 (Notably higher than recommended 1 to 2 gpm/ft^2)

Finished Goals
pH: 8 to 8.5
Langelier Index: -0.3 to 0.3
CCPP: >0
Generally needs to be sufficient to avoid Lead & Copper corrosion.

Potential Solutions?

If calcite is not suitable for this water, will CO2 or some other means be sufficient?

References

I have been using the following references:

Limestone Contactors as a Corrosion Control Technology for Small Public Water Systems
Optimal Corrosion Control Treatment Evaluation Technical Recommendations for Primary Agencies and Public Water Systems

Any thoughts or recommendations are greatly appreciated.

 

[ashtree]

Wright 44,

Do you have any details of the Limestone that you propose to use. Porosity, diameter, roundness?

With regards to the water do you know the permeate TDS and the dissolved inorganic carbon

As a general requirement you typically need to have an empty bed contact time of at least 20 minutes but maybe as long as 45mins and i am guessing that you will have a couple of minutes at the most. 6foot bed depth is the minimum. I severely doubt that you will be able to achieve your goal of 8 to 8.5pH with just calcite. The relationship between EBCT and pH becomes every steep above about 7.6pH meaning that extra bed volumes beyond a certain amount have very little increase in pH. Somewhere about pH 7.7 is the practical limit and you will need very long contact times and deep bed to achieve this.

In any respect you are going to have to think about how you are going to manage the regular loading of the contactors in the design as the amount used will be significant.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[EdStainless]

Being open to atmosphere and absorbing CO2 will push you more acidic.
Ashtree is correct, with calcite you can get to 7.5 or so easily, but then the reactions are so slow that the time required will be very long.
What other additives do you have available to use?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[wright44]

Sounds like the filter is too small for our flowrates. I had concerns about that but I wasn't sure if there were more resources on dissolution rates. As it is, the plant doesn't have much else to work with. The problem is they bought this system on the advice of their original engineer, who then retired before they got it permitted and running.

So no other additives are used in the plant yet (besides chemicals they use for Clean In Place for the RO). Looks like a heavy dose of CO2 prior to the filter would increase dissolution rate and help bump up the final alkalinity, but that will complicate things for the operator and probably won't be cheap. Maybe soda ash or other additive in the final storage tank would help?

The final pH is important, but even more so the water needs to have some alkalinity/calcium present so it does not dissolve protective scaling in the distribution system. If they turn this thing on and it dissolves scaling, they are liable to dissolve copper plumbing and trigger a violation.

 

[ashtree]

Why is finished water pH important beyond the obvious. If you are worried about corrosion, high pH is not required if you have the appropriate hardness, alkalinity, TDS etc.
Higher pH is one way to get the LSI to an appropriate value but its also a balance with calcium and alkalinity.
What this means is that if you have an appropriate sized contactor you may achieve your overall goal of a near neutral LSI with a pH of 7.5. A small trim dose of sodium hydroxide , soda ash if you prefer, or sodium bicarbonate, after the contactor might also be used as an additional control measure.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[bimr]

The original system with the calcite filter probably never worked.

You haven't provided enough information for a reasonable person to respond.

You may know that the carbon dioxide passes through the RO membrane. Because of this, the RO effluent has a lower pH, and excess carbon dioxide. Some RO systems are followed by degasification to remove/reduce the carbon dioxide and raise the pH.

You haven't provided any capacity of the RO system. A calcite system will only be appropriate for a very small system or a residential unit. In addition, the calcite filter (known as a neutralizing filter) will not raise the effluent pH above neutral, maybe a pH of 7.2, which is probably an inadequately low pH. So a calcite filter will not work for your application.

Adding carbon dioxide will not work either as carbon dioxide (a weak acid) lowers the pH.

If you want an answer, post the raw water quality and the flow rate.
.
Here is a link to a paper that discusses remineralization:

Remineralization

 

[wright44]

Thanks for the responses. Below is a table indicating raw and post-RO parameters (from RO vendor's modeling). The final RO pH says 7.22, operators have noted permeate pH of between 6.8 and 7.4 from the existing RO they have. There is no CO2, and there is very little alkalinity. This is indeed a very small system, the RO is anticipated to run at ~60 gpm (permeate production). I will take a deeper dive into the paper @bimr linked this evening, but it looks like they show CO2 pre-dosing like I mentioned in Figure 3.

@bmir "In addition, the calcite filter (known as a neutralizing filter) will not raise the effluent pH above neutral, maybe a pH of 7.2."

Is this just an issue with contact time? My understanding is that in a pressure vessel with no water and air interactions, calcite can actually result in a quite high pH. If the water has very low minerals, it can dissolve a lot of carbonate and equilibrium could go well above neutral until it comes in contact with air where pH drops back towards 7 to 8.

@ashtree You are right the final pH is not necessarily the most important. It seems to me after modeling it, that the water cannot get enough alkalinity from calcite alone to fix the issue. May be no way around using other chemicals to fix the issue.

Water Quality Parameters

Carbonate Chemistry Equilibrium Solutions

Thanks folks

 

[wright44]

Just wanted to link this study done in Texas which has some additional insights. Their conclusion is as follows:

"In any calcite dissolution reaction, though it is possible to achieve finished water calcium hardness and alkalinity goals, final polishing by caustic soda dosing is required to elevate the LSI and CCPP indices above zero and 4 mg/L as CaCO3, for anti-corrosion purposes.​

So seems without a polishing step, it is not easy to fix the LSI/CCPP. The paper also discusses CO2 dosing as a function of feed water characteristics.

 

[bimr]

The water analysis is a little suspect. You are showing 98% RO rejection which is on the high side. Is this a 2 pass RO? Do you need 100% RO throughput to meet your water quality objectives or can some of the raw water bypass the RO? If some of the raw water bypasses the RO, it will help with the remineralization.

Calcite is a carbonate mineral and the most stable polymorph of calcium carbonate (CaCO[sub]3[/sub]). Calcite is not lime and won't raise the pH by itself. Calcite will neutralize CO[sub]2[/sub], not significantly raise the pH. You would have to add an alkali in addition to the calcite. So you would add CO[sub]2[/sub] to lower the pH to dissolve the calcite, then add an alkali to raise the pH to your desired level.

Under normal conditions, CaCO[sub]3[/sub] is only slightly soluble in water. However, its dissolution in water can be enhanced in acidic conditions. The use of CO[sub]2[/sub] for this purpose is advantageous because it will both lower the pH and convert to alkalinity as the calcite raises the pH of the water. When CO[sub]2[/sub] is added to water, the acid equivalent of CO[sub]2[/sub] in water, carbonic acid (H[sub]2[/sub]CO[sub]3[/sub]) is formed, which is regularly used in soft drinks.

The proportion of available H[sub]2[/sub]CO[sub]3[/sub] as a function of pH in the water. In calcite dissolution by CO[sub]2[/sub], the first step is the carbonic acidification of the water, shown in:

CO[sub]2[/sub] + H[sub]2[/sub]O = H[sub]2[/sub]CO[sub]3[/sub] (aq)

Calcite will then react with the carbonic acid:
CaCO[sub]3[/sub] + H[sub]2[/sub]CO[sub]3[/sub] = Ca[sup]2[/sup]+ + 2HCO[sub]3[/sub]

At that point in the reaction, you are still left with a neutral pH, probably too low to prevent corrosion in the distribution system. That is why you will need to add an alkali.

The table below provides some other alternatives that can be used. The capacity of the system and the skill of the operators should be used to select the appropriate system.

The 3rd alternative is probably the best, because it adds magnesium as well. Dolomite filters are probably more common in the Middle East where desalination is more common.

Lenntech

Dolimite Filter

 

[wright44]

Thanks for the insight I think this is exactly what I need. We will have to take a step back, take a look at the existing process and see about making some modifications. I appreciate the resources and discussion.