Chlorination, coagulation, and pH adjustment are very different activities, but they are often accomplished simultaneously using the same equipment: a chemical feed pump with its reservoir of concentrated chemicals, followed by a contact/mixing tank. If the oxidant/disinfectant is liquid chlorine bleach (5.25-18% sodium hypochlorite, NaOCI), the coagulant is sodium aluminate (NaAIO2), and the pH reagent is sodium hydroxide (NaOH, “caustic soda”) or sodium carbonate (Na2CO3. “soda ash”), they can all be mixed together and fed with a single pump.

The preferred process is a variation of standard practice known as “super-chlorination-dechlorination” because a large excess of chlorine is used (about 10 times the usual dosage) and then removed before drinking. It is based on the concept of the “CT envelope,” which states that the arithmetic product of one combination of the chlorine concentration in ppm, C, times the contact time in minutes, T, is equivalent in disinfection ability to any other combination that produces the same total. For example, 30 minutes of contact time with 0.5 ppm FAC is equivalent to 5 minutes at 3 ppm, because they both total a CT of 15, which is the minimum for less-than-ideal conditions. Large waterworks have the time and space to use low chlorine concentrations with long contact times, but it is more economical for small systems to use relatively small contact tanks with ten times as much chlorine as usual, and then waste 95% of it. At a flow rate of 5 gal/min., a 5-minute contact time would only require a 25-gallon tank (if the mixing were perfect), while a 30-minute contact time would require a 150-gallon tank. (Correction for mixing efficiency changes those figures to 89 gallons and 536 gallons.)

Sodium hypochlorite solutions are made by bubbling chlorine gas up through a solution of sodium hydroxide:
CI2 + NaOH -> H+ + CI + NaOCI (not an equilibrium reaction)

The resulting solution is pH 10-12, depending on the local producer, and the free chlorine is present as the sodium salt hypochlorous acid, which is a “weak” acid with a KEq = 10-7.5. However, NaOCI itself is “strong” – a simple salt like NaCI that is 100% ionized the instant it is dissolved:

NaOCI + H2O -> Na+ + OCI (not an equilibrium reaction)

Only the acid, HOCI, is weak:

OCI + H+ <=> HOCI       KEq = 10-7.5

That implies two things: first, that when the chlorine bleach is injected into the water, it will instantly change from being all hypochlorine ion, OCI, in the bleach to being mostly in the form of the un-ionized hypochlorous acid molecule, HOCI, in the water. Second, hypochlorous acid’s KEq is close enough to the H+ concentration in neutral water to interfere with the pH of the water into which it is being injected. Actually, because of the small amounts involved, it’s the other way round: the pH of the water interferes with the hypocholorous acid equilibrium. This is extremely important, because only the HOCI molecule disinfects. All forms of free chlorine are equally effective as oxidizing agents for iron and manganese and hydrogen sulfide, and OCI ion can kill some viruses, but if the pH is top high to allow most of the HOCI to exist as a neutral molecule that can penetrate cell membranes, there will be little disinfection of bacteria. The 50/50 point in the dissociation of hypochlorous rises to 8.5 the HOCI will be 95% dissociated and useless as a bactericide. Therefore it is important not to raise the pH any higher than necessary if the chlorination system is expected to disinfect as well as oxidize.

Sodium Aluminate is a source of “alum,” or aluminum salts that form a fluffy precipitate called “floc” when dissolved in water at ordinary water pH. If the concentration is high enough, the pH is right, and there is enough gentle mixing to promote agglomeration of the precipitate into larger clumps which entrap particles of dirt and turbidity (called coagulation), the water can be greatly clarified without the use of fine-filters. In large waterworks there is lots of space and time, and after flocculation and coagulation the floc is allowed to settle, and virtually all of the suspended particles sink to the bottom enmeshed in a solid, jelly-like mass. The clear water is then sent to large granular media filters, but the filters only have to remove the occasional clump of floc that gets stirred up. Everything else has already been removed as sludge. Small, individual water supply systems do no have the luxury of acres of space for water treatment, so there is no settling phase, and the final filters are more important.

Aluminum is somewhat unusual in its ability to precipitate as a hydroxide and then to re-dissolve as a “complex ion” with the addition of more strong base:

AI+++ + 3OH <=> AI(OH)3 (solid aluminum hydroxide “floc”)
+ OH <=> AIO2 (soluble aluminate ion) + H+ + H2O

Thus, if a solution “normal” pH (near pH 7) or lower, the aluminate ion will convert to the AI(OH)3 form and precipitate, or flocculate as a fluffy, gooey solid material that entraps colloids and particles. Sodium aluminate can be purchased as a 10% solution or prepared by dissolving any aluminum salt in water and then adding NaOH until it first precipitates and then redissolves.

Note that both the chlorine bleach (sodium hypochlorite) and the flocculant (sodium aluminate) are produced as sodium hydroxide (NaOH) solutions. Both chemicals will affect the pH only slightly, because they are added only at low ppm levels. However, more NaOH can be added to either at any time. That means that the final pH of any water supply that is treated with either or both chemicals can be raised to any desired level by adding more NaOH, and they can all be mixed together and fed as a single solution. As an alternative, sodium carbonate (Na2CO3, “soda ash”) may be used to raise the pH instead of NaOH. This is sometimes a good idea, since NaOH is so dangerous to use, while sodium carbonate is relatively innocuous. However, since it is not as strongly alkaline. Na2CO3 may not be strong enough to keep sodium aluminate from precipitating too early, before it is fed, thus clogging the pump. This is seldom a problem except when the brand of bleach locally available is deficient in NaOH. So, if addition of sodium aluminate to chlorine bleach causes immediate precipitation. NaOH is the only recourse. Preparing such a mixed chemical feed solution is unfortunately rather complicated. The best way is to treat each ingredient separately in a “jar test” as described below, and then combine them in the indicated ratio. The setting on the feeder pump is then determined by trial and error.

Determining the Chlorine Dose: You will need a gallon jug, an ordinary bucket, an eye-dropper, and a chlorine test kit that measures free chlorine (not “total chlorine”). Put a gallon of the water to be treated into the bucket and start adding chlorine bleach drop by drop, counting the drops. After 5 drops, mix and wait 5 minutes, and then test for free chlorine with the test kit. The 5-minute wait is to allow time for the chlorine to react with any iron, sulfide, ammonia, organics, whatever, that may be present and acting as “oxidant demand” or “chlorine demand: The demand, if any, must be destroyed before any extra chlorine left over (called the “residual”) can be made available to do any disinfection. If the demand is not destroyed, the chlorine that would be measured with a total chlorine test would include any “combined chlorine” consisting of monochloramine, dichloramine, and organic chloramines derived from amino acids, etc., which are all very poor disinfectants and also very weak oxidizers. (They are very stable and last a long time in the pipes, which is one reason monochloramine is sometimes intentionally produced by large water works with large distribution systems. But that is done only after effective disinfection with free chlorine has already been completed.) The goal in the small chlorinator systems used for private wells is a concentration of 3-5 ppm free chlorine. If it’s already higher than that after 5 drops, dump the bucket and start over with a diluted bleach solution. Repeat the process of adding drops of bleach, mixing, waiting 5 minutes, and testing as many times as it takes to find the total number of drops of bleach needed to destroy the demand and produce a residual of 3-5 ppm free chlorine after a five-minute contact time. Record the number of drops.

Determining the Sodium Aluminate Dose: Using the same gallon jug, bucket and eye-dropper (cleaned up, of course!), add and count the drops of sodium aluminate needed to coagulate all of the turbidity or color in the water. After every 5 drops, stir very strongly for about 30 seconds, then back-stir for a moment to quiet the solution, and then let it stand undisturbed for several minutes, until the floc begins to settle. As soon as the clear solution above the cloudiness can be seen, judge for yourself whether it is clear or color-less enough. If not, add more sodium aluminate, mix violently, wait, and look again. Repeat until you find the total number of drops needed per gallon to achieve the effect you want.

Putting it All Together: The total number of drops of chlorine bleach and sodium aluminate represent the ratio of the volumes of solution to mix together. For example, if the chlorine dosage is 10 drops and the alum dosage is 20 drops, mix the solutions in the ratio of 1:2. Then try out the mixed solution in the actual installation: set the chemical feed pump at half-speed, tum on the water, and take samples. Adjust the setting as needed to produce a free chlorine residual of 3-5 ppm. But first, we need to discuss the rest of the system.

Designing a Chemical Feed Svstem: The starting point is the desired or expected peak flow rate. For most homes and small businesses, 5 gallons per minute or 19 liters per minute is about right. The size of the contact tank or mixing tank required is calculated by multiplying the flow rate times the desired contact time and then dividing by a mixing efficiency factor: (5 gal/min.) (5 min.) = 25-gal. tank; + 28% = 89-gallon tank.

The efficiency of mixing in ordinary pressure tanks is very poor. Treatment chemicals and treated water “short-circuit” from influent to effluent in contact tanks quickly, providing much less actual contact time than expected. The best that can be done is about 28% efficiency and even that requires filling the bottom third of an up-flow tank with pea-sized gravel. Thus, at a flow rate of 5 gal/min., a standard 82-gallon (310 L) galvanized steel tank with pea-gravel will provide about 4.6 minutes of contact time; a 120-gal. (450 L) tank, 6.7 min.