Here is a sample of some of the text from an onBalance handout:

Introduction - The chemistry of chlorination is a subject that is well understood and has been extensively documented over the course of the last century. G. Clifford White's Handbook of Chlorination, for example, is considered a seminal work on this subject. In the swimming pool environment, chlorine is the preferred method of sanitation, used as the primary sanitizer in nearly 90% of all swimming pools. Many types of chlorine compounds are used, primarily elemental (gaseous) chlorine, sodium trichlor, sodium dichlor, calcium hypochlorite, and calcium hypochlorite. Each has its own characteristic chemistry which is involved in the introduction of HOCl/OCl- to the water, but the end effect (chlorine species formed) after the dissolving or blending of a chlorine compound into pool water is the same, to the point where it is virtually impossible to distinguish, after the fact, which type of chlorine was originally introduced to a pool. The actual mechanics of the introduction of various chlorine compounds to pool water is as varied as its chemistry. Methods include direct injection, broadcasting, and various injectors, feeders, floaters, etc. Little or no published data exists documenting the initial distribution of chlorine throughout a swimming pool, or the pH dynamics of the water during the blending period. Since each chlorine compound has a different pH, this study was undertaken to "track" the flow of each chlorine compound during the initial blending process, noting both the effect on pH in the various parts of the pool as well as the length of time needed to entirely blend into the pool (as defined by the effect on pH). Alkalinity, pH, and buffer dynamics of the various systems are outside the scope of this study, except as relates to the "tracking" of the chlorine blending process.
The Equipment - With the advent of microcomputers and electronic data-logging systems, it was decided to apply modern equipment to the tracking process, and document the results. A special testing meter was designed and assembled for the project by Jerald Eckels, of Kruger and Eckels in Santa Ana California, a manufacturer of pool and spa chemical meters and controllers. The equipment included a 4 channel meter capable of simultaneously monitoring three pH probes and one ORP probe. These probes were wide-range combination electrodes with silver/silver chloride references, and were submersible with long connecting leads - allowing the probes to be placed virtually anywhere in the pool.. The probes were connected to a Model 600 Electronic pH recorder and data-logger. The data-logger was set to sample pH at 30 second intervals, and values were directly downloaded into a computer graphing software package designed for the data-logger.
The Pools - Approximately 60 - 70 pools were tested for this study. The balance included initial pH levels between 7.8 and 8.2, and alkalinity levels between 120 and 140 ppm. In each case, lab analysis of the pool water was performed prior to the chlorine addition to insure that the water was properly prepared for the reception of the chlorine and its attendant effect on pH. Aside from the pH meter testing, other tests included a poolside temperature reading, and a sample of pool water (previous to injection) being drawn and tested in the lab for pH, total hardness, total calcium, total alkalinity, total dissolved solids, iron, copper, chlorine, and cyanuric acid.
Note that times on the charts reflect two separate phenomena: the amount of time required for chlorine to dissolve into the water from its gaseous, powder, tablet, or liquid states, and the amount of time required for affected water to dilute with less affected water. Differing processes are required for dissolving liquid, granular, and tablet compounds of chlorine into the pool water, and the time required will vary. The actual chemical reaction for dissolving gas chlorine into pool water (Cl2 + H2O à HOCl + HCl) is remarkably fast, and is referred to as being "substantially complete in less than one second even at 1ºC" (White 1972, p. 183). The process of dilution of the affected areas to the rest of the pool is referred to as "blending" or "mixing", and is dependent on flow (for tablets), circulation (for tablet, powder and liquid compounds), and the action of the injection method (for gas).

SODIUM HYPOCHLORITE - For this profile, the top pH probe was placed 6 inches below the water surface, the middle probe was placed 3 feet below water surface, and the bottom probe was placed 6 inches from the bottom of the pool. With the circulation system running, 1 gallon of sodium hypochlorite was poured over the top probe. The graph reflects both the amount of time required for the chemical to affect the pH of the water (see the difference between the top and middle pH levels), as well as the widening area of coverage as the material descended to the bottom of the pool (see the difference between the middle and bottom pH levels). The circulation system carried the liquid around the pool, lessening its pH impact. A stratification of the pH in the pool lasted 30 to 35 minutes before blending was complete.

CALCIUM HYPOCHLORITE - For this profile, the top pH probe was placed 6 inches below the water surface, the middle probe was placed 3 feet below water surface, and the bottom probe was placed 6 inches from the bottom of the pool. With the circulation system running, 2 pounds of calcium hypochlorite was broadcast over the top probe. The graph reflects both the amount of time required for the chemical to dissolve and affect the pH of the water (see the difference between the top and middle pH levels), as well as the widening area of coverage as the material descended to the bottom of the pool (see the difference between the middle and bottom pH levels). The circulation system carried the powder around the pool, encouraging dissolution of the solid material and lessening its pH impact. All three probes showed a leveled-off pH 7 to 8 minutes after the addition of the chlorine.

TRICHLOR TABLETS - For this profile, the first pH probe was placed on the bottom of the pool, the middle probe was placed over the wall outlet to the pool, and the bottom probe was placed 4" downstream from the chlorinator. A backflow preventer was installed between the chlorinator and the filter, and a PVC loop was installed downstream from the chlorinator to prevent seepage down the wall of the pool when the system was off. With the circulation system running, the bottom of the pool was unaffected, the wall outlet showed a .2 pH reduction, and the probe 4" past the chlorinator showed a .7 pH reduction. When the system was turned off, the pH by the chlorinator dropped to pH 3.2 within 2 minutes, and remained there until the system was turned on again. When the system started back up, the chlorinator probe jumped back to -0.7 of the overall pH within 15 seconds, and the wall return was affected by the passage of the acidic solution for about a 15 second period, dropping approximately .7 of a pH unit.

Poolside Gas Chlorine Injection - Unlike the mixing action of solid or liquid chemicals, which tend to sink to the bottom and then mix upward, gas injection primarily affects the center and surface of the pool - which is logical since gas bubbles rise rather than sink in pool water. Also, gas injection differs in that the diffusion method physically moves the water, resulting in the blending of the chlorine, as opposed to the solid, powder, and liquid compounds of chlorine which are circulation dependent for dissolution and blending. The testing for gas chlorine consisted of injecting two pounds of chlorine. The pH probes were initially placed vertically above the gas injector: one six inches above the point of injection, the second six inches below water surface directly above the first probe, and the third midway between the first two. The probes remained in the pool for 24 hours after injection. In subsequent tests the probes were moved one foot to the side of the point of injection, then two feet, three feet, and so on until finally tests were done with the probes resting against the pool wall. Follow-up testing was conducted with the probes in the shallow end, under the injector, and in the skimmer. The accompanying diagrams result from the "mapping out" of the process. The actual two pound chlorine injection lasted four minutes, releasing chlorine at a rate of ½ pound per minute. During the injection, the pH immediately above the injection point drops relatively little. Within a few feet, the pH reaches its lowest level, as the chlorine and water reaction generates HOCl (free active chlorine) and HCl (hydrochloric acid). By the time the flow reaches the surface, blending has already begun, due to the venturi-like flow of water created by the gas flow, diluting the acid and allowing the pH to rise. As the water flows across the surface and then back down, the distribution and the dilution of the chlorine and acid continues. Immediately after the injection is stopped, the pH "jumps" to the 7.0+ levels. In this particular method of application the time required for full blending was under 30 minutes.

References

• Cardall, John T. and Jon Powell. Documentation for EPA Product Registration. April 1980.
• George Clifford White. Handbook of Chlorination. New York: Van Nostrand Reinhold Company, 1972).
• A review of portions of the testing (Part 4 - Poolside Gas Chlorine Injection) appears on pages 42 - 46 of the April 19, 1993 issue of Pool and Spa News.
  

(This paper was provided to the Swimming Pool Water Treatment Professionals (SPWTP) for its use by its authors, Que Hales, Doug Latta and Kim Skinner of onBalance. SPWTP has approved this material for use in its Educational Seminar Series. It may be purchased through the SPWTP web site.)