Greensand Leaching Mn 2

[ozfish]

I have a question regarding Greensand leahing further Mn into fitrate water. I am running a trial using a greensand filter to remove 2ppm of Mn in groundwater. I have followed the manufacturers instructions and am suprised to note the Mn is increasing once passing through the filter! Mn after filtration is reaching levels of 5ppm and over.

I have charged the filter by backwshing several times with KmNO4 and no change. The manufacturer advised that the KmNO4 should be fed upstream of the filter continuosly? I find this also suprising as a was of the impression the greensand only need to be recharged as it became less effective.

The pH of the ground water entering into the filter is ranging from 5.4 to 6.5.

 

[bimr]

The greensand process is usually limited to not over 1 mg/l of manganese because of the relatively high cost of potassium permanganate.

It is advisable to neutralize the carbon dioxide in the water prior to the greensand bed to eliminate the possibility of manganese pickup from an exhausted bed.

Inline oxidation of manganese by potassium permanganate would be expected to be a slow process at the pH that you are operating. For example, the pH needs to be at least 10 if you were oxidizing manganese with chlorine in order for the oxidation reaction to be completed within 15 minutes. Feeding potassium permanganate inline would probably not add anything to the process.

It may be possible to introduce a limited amount of air into the water under pressure, preceding the manganese treatment. This would reduce the amount of potassium permanganate required.

 

[semo]

I agree w/ Bimr that you need to raise the pH of the water to speed the reaction rate. This may be due to excess CO2 that can be released with Aeration. 2mg/l of Manganese is not excessive for a greensand filter, just realize that without prior oxidation/sedimentation, you will have shorter filter run times (i.e. more backwashing).

As far as the greensand filters, there are two methods of operating greensand filters.

1. Intermitent Regeneration (IR) or Batch Method. The filter is charged with Permanganate and placed on line. As the manganese is oxidized, the filter loses its manganese oxide coating (charge) and have to be re-charged again. This process continues.

2. Continuous Regeneration (CR). Once the filter is initially charged, permanganate is feed prior to the filter to oxidize the manganese and maintain a charge on the filter media.

Most larger plants and those I deal with use the CR process.

That being said, overfeeding the permanganate prior to the filter will result in pass through of manganese. The permanganate contains soluble manganese, therefore, overfeeding adds manganese to the water. You only need a slight pink tint to the water at the top of the filter. You can perform a jar test to determine your feed rate.

As it stands, I would cut back on your permanganate feed rate. With the filter being charged, you can probably even eliminate the feed for a short period. After a while you will see breakthrough of manganese though so you should keep an eye on it.

I'm assuming this is a new filter so as the filter ages, your feed rates will also need to be adjusted.

 

[ozfish]

Thank you both for your information. Semo, I tried a few jar tests using the media without charging and also found an increase in the Mn over the initial value. Would I be correct to assume that this may be due to the CO2 in the water stripping the Mn coating off the media?

 

[bimr]

yes

 

[semo]

Like Bimr said, yes.

You will probably need to look at raising the pH through aeration and/or the introduction of a chemical such as Caustic Soda.

A simple test I've done at the well head to see the effect of aeration on pH is to take a sample, test the pH, then cap and shake the sample vigorously to remove any free CO2 then test the pH again.

If you don't have room or capital funds for aeration, you can look at chemical feed. In the long run it will cost more; but, it is less expensive up front.

 

[ozfish]

Thanks again for the replies. I have some capacity to aerate prior to filtration. As the source water also contains Fe2+ I would like to investigate sedimentation prior to filtration. Can you recommend some ideas and possibly some literature for sedimentation or prefiltration that is practicle to maintain and manage. Finally, if greensand and Zeolite both remove Mn and both require charging, what parameters sway the choice?

 

[semo]

A lot depends on the quantity of Iron (and also manganese) you are talking about. As I mentioned, if the levels get high you will be backwashing very frequently if no sedimentation is available.

If you begin seeing levels above 5-7 mg/l, I think a better option than a settling tank is a solids contact clarifier. I've found large settling tanks act more as reaction vessles than settling tanks. Iron floc is light and sticky and not easily settled without an aid; therefore, it tends to pass through a common holding tank and be captured by the filter (ie. short filter runs).

These reaction vessels work ok on small quantities of Fe/Mn as they allow the aeration and chlorine to react with the Fe/Mn and save on other chemical costs. pH is still a major factor however.

A solids contact clarifier will save on space; but, require the addition of a polymer to help bind and settle the particles. At high Fe/Mn concentrations, this will remove a great deal of the oxidized Fe/Mn prior to the filters and extend the run times.

The zeolite process does not use sedimentation as it is an ion exchange process. If the iron is oxidized prior to the softener, it can tend to coat the media and prevent the process from working. It also typically exchanges sodium for the Fe which is not always wanted for some of the users (high blood pressure). Zeolite process is not typically used for high concentrations of Fe/Mn either (over 5 to 7 mg/l).

A good all around water treatment book is "Water Quality and Treatment: Handbook of Community Water Supplies". A pretty good book on Fe/Mn removal is "Iron and Manganese Removal Handbook". Both can be found at the AWWA bookstore. I'm not sure; but, I doubt there are any online copies.

A quick search on Iron and Manganese Removal in this forum or on yahoo search engine will also provide a good deal of information.

 

[chemeng01]

Certainly the increase in Mn is from KMnO4 overfeed.

Usually these things are operated with either the oxidant regen OR the oxidant prefeed; not both.

Question: are you sure the regeneration calls for KMnO4 in the backwash water? I would have expected an untreated backwash followed by downflow KMno4 regen. with an endpoint verification of some kind(filtered DPD?).

Question: how did you jar test this process?

Your source water is really ugly. Have you looked at the cost of running a line to the closest pws?

If you have to treat this water I would suggest lime softening be considered. All of the iron and manganese will be removed and the chemistry can be set to remove from 1/3 to 2/3 of the hardness and alkalanity depending on the full water analysis.

I'm interested in hearing from you.

Thanks

 

[ozfish]

Hi Chemeng01. The Mn increase was noted even without the addition of KmNO4. The posts above have indicated the increase is most likely due to CO2 rich ground water. In this situation the ground water has C02 levels above 150ppm.

With regard to the recharge, I was adding the KmNO4 after typical backwash and rinse. We added the KmNO4 feed into clean water supply and backwashed again.

Jar test was simply taking untreated water, and passing through a greensand in a bucket with holes in the bottom allowing the water to pass through into a second bucket of which we then sampled.


In the current situation there is no PWS. I consult to many aquaculture operations in this region and the groundwater sources are usually rich in FE2+ and Mn. We have found the use of Lime excellent for the removal of both, however we do not want to soften the water or remove alkalinity, or change the Mg Ca ratio to radically.