Silica 3

[Fritzfrederix]

Hi there,

At our plant we've got three demiwater installations and since we've started up we have problems with the amount of silica in the water which we pump up from a well.
The supplier told us that is was normal at the start of a new well, but the problem still exists after 1 year.
Does anyone else have the same problem and if yes what have yoou done the solve it?

many thanks Rudi

 

[GarySCWSVI]

Are your "demiwater" installations "Deminineralizers" using ion exchange resins? If so, what type of Anion resin is being used? After 1 year if the silica level in the untreated well water is still the same then that is the level you have to deal with. When you answer my questions above I can give you a more detailed solution for silica removal.

 

[Fritzfrederix]

Hi GarySCWSVI,

Yes, our "demiwater" installations "Deminineralizers" use ion exchange resins?

We buy it by Rohm and Haas.

For the weak Acid Cation :
AMERLITE IRC86RF Gel polyacrylic copolymer -COO¯

For the Strong Acid Cation :
AMBERJET 1200 Na Styrene divinylbenzene copolymer -SO3-

And for the Strong Base Anion :
AMBERJET 4600 Cl Styrene divinylbenzene copolymer
N+(CH3)2CH2CH2OH

I hope this is sufficient, if you need some more information just let me know.

Thanks in advance

Rudi

 

[bimr]

Are you having problems with your well? Or, are you having problems with your demineralizer? Not clear from your post.

 

[Fritzfrederix]

BIMR,

The silica is in the water which we pump up out of a well. The silica generates two problems.

First problem the silica breaks of our resins and secondly the silica get attached on the walls of the pipes of our steamboiler, which implements that you've got a bad temperature transmission in the boiler itself.

Rudi

 

[bimr]

Thanks for the prompt response.

Do you have a heated regeneration system to take the silica off the resins. In the power industry where I work, the DI systems use a hot water tank to heat water to 120 Deg F to elute the silica off.

 

[GarySCWSVI]

Though a Type 2 SBA resin like 4600 will remove silica we highly recommend a Type 1 since the Type 1 can be regenerated at higher temperatures to better elute silica from the resin. As previously mentioned by bimr higher temp regeneration is necessary. The temp limit for a Type 2 is 95 degrees F. A Type 1 temp limit is typically 140 degrees F.

Silica will break through any SBA resin sooner than other ions and since it is not conductive that break through will not be detected with standard demin system monitors unless one has a recorder or is watching the monitor on a continuous basis. There will be an upward spike in conductivity when the silica breaks. Most systems that are used primarily for silica removal will have a mid-bed conductivity probe and monitor that will look for that spike and cause regeneration on that basis rather than using a probe and monitor on the outlet of the demin system.

Since silica removal is important to you I suggest changing to a Type 1 or a Type 3 SBA resin and installing a mid-bed silica monitoring system. As far as I know, The Purolite Company is the only resin manufacturer that offers a Type 3 SBA resin. All other resin manufacturers, including Purolite, offer Type 1 and Type 2 SBA resins.

 

[bimr]

You might also consider adding flow totalizing to your DI system. Many DI users regenerate based on throughput rather than effluent quality. Each run of the demineralizer should have the same run length (total gallonage).

It is important to note that the effluent guarantees from DI equipment suppliers is an average quality that is averaged across the entire run length. This gallonage includes the middle of the run where the quality is best and a portion of the run at the beginning and end of the run where the quality is not as good. If you want to improve the effluent quality, you regenerate before the effluent quality deteriorates at the end of the run

 

[Fritzfrederix]

Hi,

I would like to thank you both for the help.

Bimr,

The regeneration can start on 2 conditions, the first condition is when the max flow reaches 480 m3 or when the conductivity reaches 1 microSiemens.Or depending on the measurement we decide if the DI should go in regeneration.

GarySWESVI,

We've got a silica measurement that measures every DI and the storage tank every each halve hour. So each DI get's measured ones every two hours. Depending on the measurement we decide if the DI should go in regeneration.
Is it possible to explain what you mean by the following sentence?
Type 3 SBA resin?
and installing a mid-bed silica monitoring system?

Thanks in advance

Rudi

 

[cub3bead]

Rudi,

A couple of questions that I haven't seen asked yet.

1. What method of regneration do you employ, co-current, countercurrent or is it a fluidized or packed bed design?

2. Does your regeneration sequence include a bed preheat step? A bed preheat step will pass one bed volume of warm water though the bed via the regenerant inlet header but without caustic. This step will greatly enhance silica eleution from the resin. If your system doesn't have this step it is realively easy to add to the programming sequence.

 

[GarySCWSVI]

Rudi,

A Type 3 SBA resin has the good quality of a Type 1 in that it can handle higher regeneration temperatures especially those when silica is the main issue. It also has the good quality of greater capacity that a Type 2 has but, again, a Type 2 cannot handle temperature greater than 95 degrees F especially during regeneration.

Installing a "mid-bed monitoring system": It sounds like you are monitoring for silica at or after the outlet of your demin unit. That is too late when silica is your primary issue. From my previous posting above: "Most systems that are used primarily for silica removal will have a mid-bed conductivity probe and monitor that will look for that spike and cause regeneration on that basis rather than using a probe and monitor on the outlet of the demin system.". That probe is installed in the part of the resin bed where silica breakthrough is likely to occur. When that breakthrough happens there will be an upward spike in the conductivity. That is the time to regenerate the unit.

 

[yulinio]

Hi there

We need to regenerate resins when the conductivity reachs 0.01 uS/cm, automatic trip was also included to protect the demin storage water tank to avoid any contamination.

Maybe you need to change resins type, calculating the capacity with the manufacter manual curves to take off the silica, actually do you have trends capacity of removal salts of the resins vs time operation?, maybe your resins are saturated in a big percentage.

Good Luck

 

[GarySCWSVI]

Yulinio,

Good suggestion on the cause of the problem and correction.

However, I have a question for you. In you first sentence you say "regenerate resins when the conductivity reachs 0.01 uS/cm...". I hope that is a typo. Maybe you meant 0.1? As far as I know, there is no water that can have a conductivity as low as 0.01 unless you died and went to heaven and saw it there.

 

[yulinio]

Hi There
Yes, you right, it was my type mistake, 0.1 is our value to regenerate the resins. If you took a determination to solve the problem I'll thank you if you let me know about it.
Good Luck

 

[GarySCWSVI]

yuinio,

See my Oct 30 and Nov 1 responses for recommended solutions to the problem of silica leakage.

 

[luissabate]

Consider this and debate,

A silica leakeage (eg 25 ppm SiO2)from a silica saturated resin may be less than a 0.1 uS/cm. Then the SiO2 mesurement is important.

On the other hand we have the non ionic colloidal silica. It needs some kind of filtration. This one will not be stopped by resin, and could be hydrolized in the boiler to form MgSiO3 and CaSiO3.

Only reactive silica will be detected on SiO2 determanation.

 

[cub3bead]

To get rid of the colloidal silica you will most likely need a membrane process ahead of your IX equipment.

Colliodal silica can be determined using AA or ICP instruments; one sample is filtered through a 0.45 micron filter and the other is unfiltered. The difference in the results is colloidal.

Acid disgestion of the sample with boiling will also convert the colloidal silica into the reactive form so that you can measure total silica using the molbdate blue test. A second, undigested sample is run and the difference in results is the colloidal silica.

 

[GarySCWSVI]

iussabate, Good point. Silica is non-conductive and will not be detected by conductivity monitoring per se. However, one can detect silica breakthough leakage by noting a spike in the reading of the conductivity monitor. Water quality, by the monitor, will be shown by the spike to have improved when in fact, from a silica standpoint, it has degraded.

Your point about "colloidal silica" is also true in that standard Strong Base Anion resins will not remove that form. cub3bead is also making a good recommendation on using membrane pretreatment of the water before the IX system. In addition, most resin manufacturers offer a macroporous or macroreticular resin that is highly porous and capable of removing/reduced colloidal silica. That is more of a filtering resin, in addition to being an ion exchange process, by trapping the colloidal silica within the pores of the resin.

Gary Schreiber, CWS VI
The Purolite Co.