Ask A Scientist Chemistry Archive

name         Ian M.
status       other
age          30s

Question -   I work for a water department on Long Island.We are having difficulty with some rusty 
water. No one seems to know why this is happening.We add Cl2 to the water for disinfection purposes 
and caustic soda to bring up the pH. We also add polyphosphate to help coat the pipes.
I know you may need to know the feed rates for these but, before we get to that, our problem happens 
when it is very cold, these 3 chemicals not mix. I know Cl2 breaks down slower in the cold weather. 
this is driving us crazy. It only happens in the part of town where our iron filtration plant is. We 
do not see the problem anywhere else in town and we flush hydrants regularly.any help would be 
greatly appreciated.
---------------------------------------------------------------
I would be reluctant to offer a suggestion, even if I had a good idea, in a case like this because who 
knows what the legal liability might be. This cannot be a unique set of circumstances (cold weather / 
high iron content). I would suggest the inquirer be referred to the American Water Works Association 
(AWWA) which is a professional society of public and municipal managers etc. Their home page is: 
www.awwa.org

Vince Calder
=====================================================
Hi Ian -
      This is sort of a "troubleshooting" question.  I wouldd love to help with such a thing.   But it is hard to answer without being there, or 
	  asking many questions first.
After all, iron rusts, unless one happens do something clever enough, right enough, to stop it.  The iron phosphate skin that stops the iron 
hydroxide formation (first stage of rust) is pretty thin and shaky.  Many little wrongs can upset that.


 You already seem to know that adding Chlorine gas (CL2) to the water makes acidity:   Cl2  + H2O    -->   ClO-  +  Cl-  +  2 H+
  and the caustic soda (NaOH) is added to cancel the acid:      NaOH + H2O -> Na+ + OH- + H2O  ,        and    H+ +  OH-  ->  H2O.
I presume your "polyphosphate" is Sodium tri-poly-phosphate, Na5P3O10 or Na3H2P3O10, or maybe a higher poly-phosphate.


Some things you mentioned, I do not quite get: what specifically do you mean by "chemicals not mix"?  That sounds important.
What's the whole idea of your "iron filtration plant"?  Is it there to catch rust flakes, or to remove dissolved iron?  Is it there because of 
this problem?
    Is the problem rust happening upstream or downstream of that plant?   Is that the same plant where you add chlorination and polyphosphate?


So I cannot know, but I can guess like crazy:    (please take no offense; I realize you probably do these things already...)


  - Are all the chemicals be completely dissolved  before they leave the plant? change what you need, to do that in winter like in summer.  List below.
         The only thing that is "not mixing", is your polyphosphate not dissolving.
  - Learn your series-of-phosphates chemistry:
         tri-poly-phosphate -> pyro-phosphate -> ortho-phosphate,   and it stops there.   This change makes acid.    Which one stops rust?
  - Consider that rusting can be a history-dependent condition.   Once it starts, you may have to shock it with stronger-than-normal anti-rust conditions to stop it.
          Higher pH, more phosphates, less chlorine for a time.  Then back to normal mix.
          Then if your mixing has one big hiccup, that could start it again.   Despite doing the magic "right thing" most of the time.
  - Measure your pH downstream, and compare with the plant's output.  If the tap is more acidic than the plant, that difference needs to be stopped.
  - Measure your chlorine downstream, where users get it.  Use only as much as needed to make that level right.  that will be less in winter.
          If you need more to suppress bacteria, where is the bacteria coming from?
  - Find out if the pipes are really ragged with rust.  Polyphosphate may never be enough to stop that.
  - Find out if there are any breaks in the pipe.   If road salt or ocean salt-water is bathing the outside of your pipes, rusting them through.
  - Find out if there are any dead branches hidden underground.   If flow cannot go in one end and out the other,
             it cannot ever get flushed, and any degradation inside that dead-end leg could gain momentum.
  - At any break that does not have strong outflow, salty water could be diffusing in, even if the pressure is higher inside the pipe than outside.
  - Swim down the main pipes with some little remote camera, scraping the floor as it goes.  Could some persistent sediment be there, despite flushes?
            Sediment tends to smother a surface and dominate the pH, despite the careful chemical balance flowing by above it.
            You'd see whether the rust starts at some particular spot, or happens evenly down a whole length of pipe.
            If it is the whole length, it is bad chemistry.   If it is one spot, maybe it is a break, or a rust-ruined spot, or other flaw.
  - Is there unnoticed positive electric current running into the pipes, maybe from the "iron filtration plant", driving the corrosion?
          Can you find it and stop it, or deliberately put on some negative current to try to protect the pipes instead?
          If the water is made negative and the pipes are made positive, rust goes way faster.
          And sometimes, if the pipes are negative and the water is positive, it goes much slower.   Sometimes that takes more current 
		  than you can afford.
          The water, for this point, can be the water inside the pipe or the groundwater outside the pipe.
          Although, it is hard to have tap water conductive enough to carry rust-inducing currents very far inside a pipe.
          And corrosion outside a pipe doesn't make rust inside,  until it breaks through the pipe somewhere.


I wonder how many of these things somebody around your place already knows?
Any detail that nobody knows, eventually gets to be the problem.
**************************************
series-of-phosphates chemistry:
               poly pyro               ortho           acid 
>              ortho         more acid
           (P3O10)5-   --> +(H2O, time)-->   P2O7(4-) + (PO4)3- + 2 H+  --> +(H2O, time)-->    3 (PO4)3-  + 4 H+


           Which of these three phosphates is the rust protectant that does the job?   I'm not an expert in that, yet.
                The concentrations added to city water are so low, that maybe only the most insoluble is good enough to do the job.
                Well, iron pyro-phosphate Fe4(P2O7)3 is more insoluble than iron ortho-phosphate FePO4, in my old CRC book.
                        No data on poly-phosphate.


           If ortho-phosphate is good enough, one might prefer to finish the conversion in the plant, so one can finish neutralizing in the plant,
                   so pipes do not get a time-delayed dose of acidity.
           If pyro-phosphate is needed, then converting that far before neutralizing and releasing would at least kill half the stored acidity.
                   If this is what works best, it would be an arcane recipe, easily lost or messed up.  How does one do "invisible soup, 
				   cooked half-way"?
           If undissolved solids remain, it is probably all polyphosphate, and all of the acid release will happen somewhere down the pipe,
                  sometime after they finally dissolve.
           If it is all dissolved, it is hard to tell what proportions the different types have in the clear water.  It can take time to change.
                I need to look up how slow it is.    Slower when cold...?  Faster with chlorination or fluoridation?



I do not think that the 3 chemicals refuse to mix in cold water, I think that the polyphosphate, all by itself, 
either dissolves or it does not.
   In the long run, in pure water, your sodium poly-phosphate always dissolves, but:
    - it takes time and stirring       
	(your plant has a fixed design and procedure)
    - it dissolves faster in warm water and slower in cold.               (winter makes it cold)
    - it dissolves faster in alkaline and slower in acid (I think)       (chlorine makes it acid, soda makes it alkaline)
    - it helps to have more mixing water or lower concentrations.   (your plant has a fixed design and procedure)
    - it tries to precipitate with Calcium or Magnesium (or Iron) in hard water, really slowing down its eventual dissolution.


After it dissolves, it always converts from Poly-Phosphate to Ortho-Phosphate, both being invisible,  dissolved in the water.
One poly-phosphate molecule generates 3 or more ortho-phosphate molecules, each with its own acidity.
This may be enough to lower the pH a point, and accelerate rusting.   It requires more soda, but if the chemicals "did not mix",
and someone let clouds of precipitate flow out of the plant, the acid would evolve way downstream after
your last soda-addition point, and after your pH testing point, and you would never get to finish neutralizing it.
If it is not all dissolved, you can neutralize the pH right now without neutralizing all its, uh, "potential acidity".


I think the plant needs to change whatever is can to make "it all" dissolve, if there are solids flowing out.
Even reducing the poly-phosphate flow to 50%, if that's all you can do.
Anything un-dissolved lowers the pH later, like a time-delayed acid bomb.   That's a hypothesis, anyway.
Improving things so you can dissolve a100% flow would be better, when you can do it.


Things which might help mixing and dissolution:
   - use more mixing water,
   - make the mixing water warmer, or better stirred, or mix longer.
   - get the best mixing order:      add most of your usual soda, then add polyphosphate and mix, then check it's all dissolved,
           then add chlorine, maybe check dissolution again (turbidity?),  then add a little more soda and measure pH before release.
   -try using a more soluble type of phosphate: sodium  ortho-phosphate is much more soluble than sodium tri-poly-phosphate 
   or sodium pyro-phosphate.
              (if ortho-phosphate is good  enough to suppress rust; I am not an expert at that.)
  - Minimize the chlorine, taking advantage of winter, if your pipes are clean enough to allow it:
           If the plant chlorination can be set to get the downstream levels constant, instead of plant-output levels,
           this will automatically require a little less chlorine in winter, and that might make polyphosphate 
		   dissolution a little easier.


Just maybe, keeping it together longer with some chlorine, before it is diluted by merging with the main water flow, will make 
it convert to ortho-phosphates faster.
I do not know whether that is better or worse, and I do not know which way your plant now has it, but it might be significant.


hates them non-working situations:

Jim Swenon
=====================================================