Since the solubility of oxygen, nitrogen, and carbon dioxide is zero at the boiling point of water, cold water can be deaerated by reducing the pressure over it until it boils. The amount of vacuum required will be a function of the deaerated water temperature. The solubility of gases in water is described in Henry’s law. The solubility decreases with increased temperatures and decreasing pressure.

As stated by bimir, the presure needs to be reduced until you reach the vapour pressure of the solution, at which point it will start boil.  This dictates the pessure you need to operate at. You need to maintain the solution at the bubble point. Perry has vapour pressure data for water at given temperatures. I recently specified some vacuum degassing columns to remove O2 from a sulphate solution at approx 55 oC, and we had to operate at about 15 kPa a.

Mark McLean

WorleyParsons

Thanks bimr.
Thanks MMclean.

I shall then find out the vapor pressure at the desired boiling temperature which is permissible in my treatment plant design. Thanks again.

lin sh  

greenwater,

Are you doing any pH adjustment?  You should familurize yourself with the relationship between carbonate, bicarbonate and CO2.  Once you return the water to "normal" pressure the ratio will be reestablished and you can end up with CO2 being created from the carbonate/bicarbonate.  I have found that using pH adjustment(lower), a traditional degasifier (atmospheric with blower) and pH adjustment (higher) can remove almost all the CO2.  If TDS and pH are not a concern just increasing the pH will convert CO2 to (bi)carbonate and eliminate the degasifier altogether.

marcusmcgee,

There are three reactions that govern the chemistry of CO2 in water:

1. CO2 + H2O<--->H2CO3 (carbonic acid)
2. H2CO3 <---> H+ + HCO3- (bicarbonate ion)
3. HCO3- <---> H+ + CO32- (carbonate ion)

When gaseous CO2 is dissolved in water, a portion is hydrated to form carbonic acid. Carbonic acid then dissociates into bicarbonate and hydrogen ions. At a pH 4.3, very little of the carbonic acid is dissociated or carbon dioxide is formed. At a pH of 6.2, the molar concentration of carbonic acid equals that of the bicarbonate and hydrogen ions. At a pH of 8.2, there is no CO2 present in the water. Above this pH, the bicarbonate ion is converted to carbonate.

It is thus best to operate the degasifier at pH between 4 to 5. I am dosing in HCl to reduce the incoming water pH prior to the degasifier tower.

Degasifiers are normally used to remove free carbon dioxide.

Typically, pH adjustment is not part of the degasification process. The effluent from a degasifier always has a higher pH than the influent because of the loss of free carbon dioxide in the water.

If an acid water is aerated under the exact same conditions as an appreciably alkaline water, the acid water will contain a greater residual of free carbon dioxide than the alkaline water. That is due to the tendency of the bicarbonates to part readily with some of their half-bound carbon dioxide.

Acidification with HCl acid destroys alkalinity according to the following equation and increases the amount of free carbon dioxide in the water.

CaCO3(s) + 2HCl(aq) ==> CaCl2(aq)+ H2O(l) + CO2(g)

This acidication does not in itself improve the efficiency of degasifier. However, you are increasing the quantity of free carbon dioxide to be removed.

Not sure that exchanging the alkalinity with chloride is a useful treatment process in water treatment. In water treatment, you are usually trying to remove salts from water.

If you have a degasifier following a cation demineralizer, then you are removing alkalinity (that has been converted to soluble carbon dioxide gas)) according to the equation.

CaHCO3  + H2Z ==> CaZ+ 2H2O + 2CO2(g)

Destroying alkalinity with acid is commonly done in a boiler or cooling tower to control the level of alkalinity. Not sure for what purpose that you are trying to do this in a degasifier.

bimr,

I am actually designing an ultrapure water treatment plant (WTP) for a wafer plant. The objective of installing a degasifier is to remove the alkalinity in water in the form of CO2. If alkalinity is removed, this will prevent the hardness scaling in the downstream equipment i.e resins or membranes.

If content of CO2 is high, as mentioned by you, efficiency of degasifier tower will decrease. Thus, a bigger tower, longer contact time and higher air flow is required. Due to this, I would like to explore the option of vacuum degasifier which is not common in ultrapure water treatment. Is it because it is really not reliable?

I could not obtain much informations on this as water engineers are not as strong in chemical process design. Thus they prefer to use more expensive options such as softeners and gas transfer membranes.

Wafer manufacturers are struggling hard in reducing the cost of production and ultrapure water is one of the major contributions.

If a good vaccum degasifier design can replace a gas transfer membrane or the degasifier or the softeners, this will greatly reduce the cost of the wafer production and it's an added advantage as a selling point. But I still need to know why it's not a common practice.





 

 

In years past, the need for a vacuum degasifier in ultrapure water was greater.

Prior to the invention of the RO process, it was common to have a degasifier of some sort between the cation and anion demineralizer units. The RO unit effectively removes 90% of the process load to the ion exchange unit. The cation unit will convert 100% of the alkalinity into CO2.

A degasifier was used because it was less expensive than using the ion exchange resin in the anion unit to remove the CO2. Less caustic was used and the anion exchange unit was reduced in size.

So, the degasifiers were used to remove CO2 that was converted from alkalinity in the cation unit units.

At the present time, almost all of the alkalinity is now removed with the RO unit. The RO unit rejects 90% of the alkalinity but none of the CO2.

I suspect the reason that vacuum degasifiers are no longer used is that:

1. They are expensive.
2. Most of the alkalinity (that was formerly CO2 in the Ion Exchange plant) is rejected by the RO unit.
3. The anion resin can remove the residual CO2.
4. Some UPW plants are even using double pass RO units for higher rejection.

The typical simplified UPW process scheme is generally:

1. RO Unit
2. Degasifier if required for high CO2 (vacuum degasifier is preferable over a forced air degasifier)
3. IX

Vacuum degasifiers are reliable and still in use in power plants for oxygen removal.

If you are concerned about scaling in the RO unit, the generally accepted practice is to install a cation exchange water softener prior to the RO unit. The water softener will be less expensive than a degasifier.

From your post, I understand that you are considering the following process:

1. Acidification/ Vacuum Degasification
2. RO
3. Ion exchange

First the acidification/vacuum degasification step. If you use HCl as your acid, you will be simply converting the calcium carbonate water to calcium chloride water and not removing any of the TDS. You are just substituting the chloride ion for the alkalinity ion.

If you use H2SO4 as your acid, you will be simply converting the calcium carbonate water to calcium sulfate water and not removing any of the TDS. The water will be more scaling that than the calcium carbonate water.

In this process application, the use of a vacuum degasifier does not allow the following process equipment to be reduced in size and is therefore not economical.

You also must recognize that all waters are not the same. Some waters contain carbonate and some do not.

Why don't you post a complete water analysis and someone can provide you a process recommendation.

Dear bimr,

To use a softener upfront of RO unit is common. However, to use a degasifier in from of RO unit is also common. Reason being that removing either the hardness (Mg or Ca) or the alkalinity will then prevent the RO membrane from scaling. I do not agree that a forced draft degsifier is more expensive than a softener. But I am not sure if we were to compare between a vacuum degasifier and a softener.

When HCl is dosed, the acidic condition will convert the carbonate or bicarbonate into carbon dioxide which is then removed via degasification. We will get calcium chloride eventually and this will be removed by RO membrane. The key point is, we must get rid of either the hardness or the alkalinity prior to the reverse osmosis process. This is to prevent hardness scaling let be calcium or magnesium carbonate.

We are not afraid to load the membrane with water of high concentration in salt or dissolved solids, but to load the membrane with water which will cause fouling or scaling to the membrane. Once the membrane is scaled, the plant needs shutdown for chemical cleaning. Furthermore, the amount of acid required to bring down the pH from 7 to about 5.6 is negligible. Removing of the alkalinity is also reducing the total dissolved solids in the water, thus, this is a void.

Hope I did not confused anyone from my post. Thanks for the valuable reply and we shall learn more when more discussion goes on.

Regards,
Greenwater

  

1. I have been in the water business since the 70's and have never seen an application with a degasifier in front of an RO unit. A degasifier in front of an RO unit would be considered an unusual application, not a common application.

(Degasifiers are more commonly used after a RO unit to remove the CO2 that passes through the RO membrane.)

One simple problem with the use of a degasifier preceding a RO unit. You do not want to blow air containing contaminants through the water. The contaminants may foul the RO membranes and degrade the water quality.

2.  I have never seen anyone try to take out all of the alkalinity out with a degasifier.

If you started off with 40 mg/l of calcium bicarbonate, that equates to a TDS of 162 mg/l. If you add 100 mg/l of sulfuric acid, you will decrease the alkalinity by 100 mg/l, but increase the TDS by 36 mg/l.

If you use hydrochloric acid, the TDS will decrease by about the same amount.

The net result is that the acidification step has an insignificant effect on reducing TDS. Why would anyone do it?

3. The optimum operating rejection range for RO membranes is not in the low pH ranges and generally occurs at a pH range of 6.5-7.0. So you will get poorer salt rejection at a low pH from your RO membranes.

If you had an acidification process, you would have to raise the pH prior to the RO process to get the best rejection from the membranes.

4. Use of acidification will not reduce the size or cost of the treatment equipment that are necessary after acidification to produce UPW. You will not reduce the size of the RO or any other process with the use of acidification.

In summary, here are the problems with acidification prior to the RO unit:

1. Insignificant reduction in TDS.
2. Potential contamination of water from airborne contaminants.
3. Poorer rejection from RO membrane when operating at low pH.
4. No cost savings in equipment that follow the acidification step.

There exists simpler and less expensive methods to reduce the scaling potential in RO influent water.

1. Partial acidification.
2. Reducing RO recovery.
3. Partial water softening prior to RO.

You also must recognize that all waters are not the same. Some waters contain carbonate and some do not.

Why don't you post a complete water analysis and someone can provide you a process recommendation.

Dear bimr,

Thanks for the valuable comments. The water contains a total alkalinity of 215 ppm as CaCO3. Calcium and Magnesium is 195 and 95 ppm as CaCO3 respectively.

Cation is(Ca-3.9, Mg-1.9, Na-0.61, K-0.03, Ba-0.001) and anion (HCO3-4.3, Cl-0.77, F-0.01, NO3-0.1, PO4-0.01, SO4 0.78) meq/l respectively. With this high level of alkalinity an hardness, the probability of scaling the RO is very high.

Of course using a two pass RO with acid injection to pH 5.6 in the first pass (at this pH, alkalinity is in the form of CO2 thus will not react with the harness to form carbonate scaling) and NaOH injection to 8.2 prior to second pass (alkalinity in form of carbonate/bicarbonate which then can be removed by RO) might seem to work, but antiscalant is required because hardness is not fully removed. Using this method, permeate quality is not consistent as it depends on the accuracy of the dosing pumps and pH analyser. The permeate still contains CO2.

Thus you are right, degasifer should be used after the RO unit, but you are right too that this will contaminate the RO permeate if not properly designed.

Thus, I tried to put the degasifier upfront instead. Purpose is to remove the alkalinity in the form of CO2 at low pH (5.6). The water is then pumped straight through a RO prefilter and a two pass RO system. When alkalinity is removed, the scaling of RO membrane will not occur.

The RO permeate tank must be blanketted with Nitrogen. Theorectically this seems to work.

On the other side, as explained earlier, if I use a softener, it will remove the Ca and Mg. Thus, with such high alkalinity, scaling will not occur with the absence of of harness. Running the RO unit at higher pH (8.3) will get rid of the alkalinity and in a way removing the CO2 from the water.


For your information, in a conventional demineralisation plant which consists of Cation, Anion and Mixed Bed ion exchanger, which is commonly used in Power Plant for their boiler make up water, a degasifier is commonly used. It is located in between the cation and anion exchanger. The theory is to benefit from the acidic water produced after the cation exchanger whereby the alkalinity is in the form of CO2 to remove by degasification. This will reduce the size of the anion exchanger and thus save on the regeneration chemicals.

I hope my theory is correct as I have difficulties in understanding it till now. Please correct me if I am wrong as I have designed demineralisation plant for power plant using degasifier. I would like to adopt this theory in my ultrapure water treatment system as well. However, as mentioned, the contamination is a problem and thus, I am considering the use of vacuum degasifier.

To conclude, see the following flow system (this is applicable only to the above mentioned water quality)

option 1
degasifier ==> 1st Pass RO ==> +NaOH ==> 2nd Pass RO

option 2
softener ==> 1st Pass RO ==> +NaOH ==> 2nd Pass RO

option 3
+Acid ==> 1st Pass RO ==> +NaOH + antiscalant ==> 2nd Pass RO ( permeate not constant and still contains high CO2)

Note:
1. We need to remove the alkalinity or hardness before the 1st Pass RO
2. We cannot run the 1st Pass RO at high pH without removing either alkanility or hardness. Instead, we need to run the 1st Pass RO at low pH where all alkalinity is in the form of CO2.
3. After the 1st Pass RO, most of the hardness is removed.
4. In the 2nd Pass RO, pH is raised to remove the alkalinity

Using Filmtec ROSA program, with option 1 and 2, I am able to get good results of the permeae with CO2 less than 0.09 ppm.





 

If you are set on doing this acidification process, then you should seriously investigate a dealkalizer unit. The dealkalizer will remove most of the cations as well as the alkalinity.

The dealkalizer is applicable for waters in which the alkalinity anions constitute a high percentage of the total anions as your water analysis shows. In this system, a degasifier is always employed. In the first step, most of the water passes through a hydrogen cation exchanger unit which is fitted with a bypass so that a small controlled flow of raw water may bypass the unit and mix with its effluent. The flow of raw water so used should be just sufficient so that its alkalinity content will neutralize the relatively small amount of mineral acicity in the hydrogen cation echanger effluent.

For the most efficient operation, this mixture should be practically zero in both mineral acidity and alkalinity. Tee mixture then passes through a degasifier which reduces the total carbon dioxide content (that formed in the hydrogen cation exchanger plus the carbon dioxide formed in the neutralization and the free carbon dioxide content of the raw water) to a low residual.


>>>>>>>>>>Cation Exchanger>>>>>>>>Degasifier>>>>>RO
     |                        |
     |                        |                 
      >>>>>Bypass Flow>>>>>>>>

Hi to all,

I am also evaluating the installation of CO2 degasser upstream of an existing reactor clarifier. Caustic solution is injected to the water stream in the reactor clarifier in order to precipitate Calcium Carbonate, Magnesium Carbonate and Silica at about 140°F. Harness is reducd from 1000 to 20 ppm, but the consumption of caustics is high.

Produced water has been sampled and analyzed for pH and free CO2 content before the caustic injection showing free CO2. Water pH changes from about 6 to 7 when sample is exposed to atmospheric conditions, after CO2 is released.

I certainly don’t recommend reducing pH for displacing the equilibrium and generating additional CO2 unless the harness is really low and total removal of CO2 is feasible. Acidification facilities were installed in one of the water treatment plant located in California, but that operation was abandoned due to poor performance and increases in operation cost.

My intention with the degasser is to minimize the use of caustics, theoretically used to neutralize carbonic acid promoted by the free CO2. This operation was also tried before in another plant and abandoned due to packing fouling with carbonates and Silica. I am not considering packed degasser to avoid fouling. I am thinking into buble nitrogen.

Also, if you have an economical way of increasing the operating temperature (typically not exceeding 200 °F), the solubility of carbonates will decrease and water treatment may be less expensive.

David

dochoa,

You have raised several issues:

1. Why do you use caustic instead of lime, which is traditionally used for this purpose. If you have a significant flowrate, lime will be much less expensive.

2.  If you installed a simple cascade aerator on top of your reactor clarifier, you would release the CO2, probably down to 5 mg/l. You would not need a packed bed desgasifier to do this, a simple slat tray aerator would suffice.

Thank you bimr

1. The capacity of the plant where I am proposing the degasser is 50,000 BPD. This plant has been running for many years. Similar processes using caustic solution are installed in other two treatment plants. The total installed capacity may be around 400,000 BPD, all treated with caustics. I will investigate opportunities if using lime.

2. My intention is installing a simple degasser at atmospheric pressure, bubbling Nitrogen to minimize the retention time required to release free CO2.

I agree with you that bubbling Nitrogen through water will aid the removal of CO2. But having said that, CO2 will be released quite readily so that Nitrogen is not required. I have never seen anyone try to do it either.

Lab measurement of pH changes at atmospheric pressure indicated that free CO2 will be released at about 1 minute. When you translate that time to degasser retention time, it may result in large vessel depending of plant design flow. The idea of bubbling Nitrogen is minimizing required time by inducing the gas phase.

You can also use other gases for improving the process (air, hydrocarbon gases), but Nitrogen is clean and it is available at site.

The only time I have considered a decarbonator in the RO feed stream was a WAC followed by a decarbonator with caustic feed to the product tank. I was doing this because I was lucky enough to have the right type of water (i.e. sufficient hardness and alkalinity) and unlucky enough to have way to much silica and a tight water balance so that I needed to have a feed pH > 8.5 so that I could use the higher SiO2 solubility to my advantage.

The one downside to a forced draft decarbonator is you really have to filter the air well if you don't want a new crop of particles in your RO feed.

Now that I have been through all of that discourse. Why not use a Liqui-Cel membrane contactor for doing the CO2 removal ahead of RO?

The water flood system dochoa is doing is completely different critter.     

Labolsh,

We build both types, and the membrane contactor type degasifiers are much less expensive than their equivalent packed tower type.  In addition, membrane contactors take up much less space.  You should get another quote.

S. Bush
www.water-eg.com