One area where an environmentally friendly technology has been applied with economic benefits and superior performance pertains to electronic water treatment in cooling towers. The conventional or standard method for the treatment of recirculating water in cooling towers is with chemical additives. The concentrated forms of these chemicals are often hazardous, and the residual chemicals released by periodic blowdown, evaporation, and drift is problematic. This article describes a non-chemical treatment technology (the Dolphin System, manufactured by Clearwater Systems of Essex, CT) on cooling tower water in a commercial office building in Pittsburgh, Pennsylvania. The water chemistry and the biological activity resulting from this treatment are also discussed, with appropriate data provided.
Comfort Cooling Systems
The most energy-efficient method to remove waste heat for comfort cooling is by evaporating water in a cooling tower. Under a typical arrangement, a chiller using a hydrochlorofluorocarbon (HCFC) or hydrofluorocarbon (HFC) refrigerant will extract heat from a closed-loop water system (chilled water used to cool the building) and dump that heat into an open-loop system. The heat is removed from the open-loop system by evaporation in a cooling tower, with approximately 1% of the recirculation flow being evaporated per 10oF heat rejection.
For example, a 600-ton chiller needs to dump 9,000,000 BTUs per hour at full utilization. This requirement is accomplished by raising 1,800 gallons per minute (gpm) of water 10oF. This warm water is then run through a cooling tower where 18 gpm are evaporated, lowering the remaining water by 10oF for reuse in the chiller.
While this method is extremely efficient in removing heat and conserving water, there remains an ongoing need to control water chemistry with respect to microbial growth, scaling, corrosion, and fouling. Traditionally, chemical treatment programs have been employed to accomplish this control.
Periodic or continuous blowdown is required to remove soluble and semi-soluble salts to control water chemistry within the range of effective chemical treatment. Environmental concerns about this blowdown limit the type and strength of chemical treatment that can be used. The previously described 600-ton chiller, operating at 4 cycles of concentration, would require 6 gpm of blowdown, which would discharge to a publicly owned treatment works (POTW) or receiving stream, subject to discharge approvals.
Standard Method of Water Treatment
The standard chemical treatment method for open-loop systems includes oxidizing biocides (chlorine or bromine), non-oxidizing biocides, corrosion inhibitors, dispersants, and scale inhibitors. Several of these chemicals are often combined in proprietary blends. Care must be taken with the particular chemicals used since they can work at cross-purposes. These chemicals are added continually or periodically to the open-loop water. Residual levels of these chemicals found in the blowdown, air emissions, and drift loss may trigger federal and state discharge regulations. These discharge regulations are a "moving target" of increasing scrutiny with continual pressure by regulators to improve the quality of the discharge and reduce negative environmental impacts. Historically, chemical treatment has been considered to be an accepted and relatively effective way-until recently, perhaps the only way-to control cooling tower water chemistry.
Electronic Water Treatment Using Pulsed-Power
Over the years there has been a plethora of magnetic devices alleging to control water chemistry with respect to scaling, corrosion, etc. Many of these items have been permanent magnetic devices with a fixed-field strength. Generally speaking, these devices have been judged as being an untenable, illogical, "black box" approach to water treatment. Most engineers and scientists dismiss these devices as just another gimmick.
In support of the technology discussed in this article, Maxwell Technologies employs and licenses a pulsed-power device for the cold pasteurization of food products called CoolPure, which meets or exceeds Federal and Drug Administration (FDA) requirements. The pulsed-powered Dolphin System described in this article is an adaptation of similar technology to open loop recirculating cooling water systems, but it utilizes much lower power draw.
The Dolphin System imparts pulses of a high frequency, electromagnetic field into the circulating water. What distinguishes this system from magnetic water treatment devices is that the Dolphin unit sets up a pulsing, coil resonating (or "ringing") harmonic field across the flow gradient as it passes through the system to which it is installed. The Dolphin's coils are located externally to a pipe in a special, self-contained spool piece, thus providing a complete non-contact treatment method. The coils operate at a low voltage of less than 45 volts.
Applications of the this device to date have demonstrated that the pulsing electromagnetic field alters water chemistry as the recirculating minerals begin to concentrate in the recycle loop. Specifically, it has been demonstrated that this system does not permit calcium carbonate to supersaturate; a condition that normally occurs in chemically based treatment programs. The chemical treatment programs typically control deposition by altering the crystal structure to inhibit scaling which would normally occur on pipe walls as a hard, smooth deposit. The pulsed-power system operates at or slightly above the saturation point of calcium carbonate by promoting an in-situ precipitation of this mineral into the bulk solution.
Photomicrographs of the two minerals are shown on the following magnifications:
The morphology of these two photos displays the typical layered structure of adherent, surface-nucleating scale, and the amorphous, almost spherical appearance of the pulsed-power induced, bulk-solution precipitate. These samples were collected from cooling towers receiving chemical or pulsed-power treatment, respectively.
The amorphous type scale does not adhere to the pipe wall but remains with the bulk solution and is removed via blowdown and/or side-stream filtration. Further, as the precipitate forms, it encapsulates particulates and immobilizes bacteria, which act as the nucleating sites for crystal growth. Additionally, bacteria are further destroyed by the process of electroporation, which is similar to the pasteurization technique employed in the CoolPure process. Essentially, the pulsed-power system "shoots holes" through the bacteria, thereby promoting bacterial disintegration through cell lysis.
Corrosion inhibition with the Dolphin System is accomplished indirectly by maintaining sufficient cycles of concentration to force the system into the alkaline mode at the saturation point of calcium carbonate, which favors scaling over corrosion. This principle is essentially the same one employed in today's chemical treatment programs, since low/neutral pH chromate treatment is no longer allowed. Further, the pulsed-power system reduces total dissolved solids (TDS), (TDS causes increased corrosion activity), by precipitating calcium salts with alkalinity at the saturation point. Chemical systems inherently maintain higher TDS levels under supersaturated conditions.
The Installation in Pittsburgh
A pulsed-power system was installed at a major corporate headquarters in Pittsburgh, Pennsylvania, in May of 1998, on the condenser water loop of their 600-ton cooling tower. This facility was a brand new office building and cooling system. An 8" diameter pulsed-power unit was installed on the condenser line after the chillers and prior to the cooling tower. A second 2" diameter unit was installed on the make-up water line. The driving force for the installation was the elimination of chemical handling in the facility and the subsequent discharge via blowdown into the local POTW.
The system has two chillers, a 400-ton and a 200-ton unit, with a full-flow centrifugal separator prior to each of the chillers. A conductivity meter controls blowdown. A side-stream bag filter on each separator filters approximately 10% of the water flow and returns it to the circulating pump. For the first year of operation, the bag filter inlet was located at the bottom of the separator. Later it was moved to the side of the separator, and the bottom discharge was connected to the sewer system and manually flushed once per day. Later still, the conductivity controlled blowdown, which was originally located on the combined line prior to the cooling tower, was moved to the bottom of the separators. These steps were taken to reduce the frequency for cleaning the filter bags and proved to be very successful.
Originally a 50-micron bag needed to be cleaned every few days; however, with the final setup, 25-micron bags easily last over one month between cleanings. A schematic of the final system is shown in Figure 1.
The system is operated for eight months of the year, from early spring to late fall. The pulsed-power system has been in operation through three complete cooling seasons.
Chemical and Biological Results
Biological samples were taken every month of operation and analyzed for Heterotrophic Plate Counts (HPC) using the EPA standard method of analysis (SMEWW 9215) at a Pennsylvania-certified laboratory. Related data are plotted in Figure 2.
For the last two years the conductivity meter has been set at 1250 mSiemens/cm. A detailed chemical analysis of the recirculating water was conducted in November of 1999 and in August of 2000. Related data are listed in Tables 1 and 2, respectively.
Examining the Results
It is difficult with chemical controls to maintain low microbiological activity on a consistent basis. Chemical control requires maintaining a residual level of biocide in the water and/or slug treatment at higher dosages. The loss of the biocide varies with temperature, pH, heat demand, blowdown, and biological activity. Bacteria often develop resistant strains to the particular biocide being used. Some amount of biofilm will develop, providing an ecosystem for bacteria to flourish and periodically releasing variable quantities of bacteria into the water. Because of these circumstances, excursions of HPC in the 100,000 CFU/ml level are not uncommon in a chemically controlled tower.
The biological data for the pulsed-power system reveal an exceptionally well-maintained cooling loop. The highest HPC measurement of 2,600 CFU/ml was made on the system immediately after the new system was flushed and cleaned, before it began to operate. All values since then have been lower. The Cooling Technology Institute (CTI) target value for recirculating water is less than 10,000 CFU/ml. After the initial start-up, the highest HPC value was only 25% of the CTI's target value. As seen in Tables 1 and 2, the HPC is 1,000 CFU/ml for both test periods.
Tables 1 and 2 show that the detailed water chemistry results are very consistent from year to year. Several points on this issue are noted as follows.
1. Using SO4 and Cl concentrations more accurately predicts the cycles of concentration, since conductivity and/or TDS are reduced by the in-situ precipitation of calcium carbonate. The recycle system appears to be operating at around 5 cycles, based on SO4 and Cl levels.
2. Alkalinity is running at less than 2 cycles and calcium hardness is running between 3 and 4 cycles. This evidence points to a precipitation reaction of calcium carbonate.
3. The bag filter sediment shows an HPC of between 270,000 to 510,000 CFU/ml, thus indicating encapsulation of bacteria. This result is further evidenced by the supporting high TOC and LOI (180°C) values in the residue.
4. Turbidity of the recycled water is actually less than the city water make-up indicating the "polishing" effect of the in-situ precipitation process. Microscopic particles (including bacteria) act as nucleating sites for precipitation of calcium carbonate, which is removed in the side-stream filters and blowdown.
5. In addition to the recycle water analysis reported in Tables 1 and 2, corrosion coupons were installed and analyzed on an ongoing basis. A 58-day copper test showed <0.1 mpy and a 485-day test showed <0.01 mpy corrosion rates. Both of the copper coupons exhibited no localized attack and general corrosion rates less than the accuracy of the test. A 58-day mild steel test showed 5.2 mpy, a 278-day mild steel showed 1.7 mpy, and a 421-day mild steel showed 1.75 mpy corrosion rate. All of the mild steel coupons also exhibited uniform, general-etch type corrosion with no localized attack.
6. Since a precipitate is forming in the bulk solution, the Langelier Saturation Index, as expected, is slightly greater than 0.0. The Ryznar and Puckorius indices, which are normally favored in predicting the scale forming/dissolving tendencies of recycled water, do not seem to be applicable in evaluating pulsed-power treated water. Both of these indices indicate a corrosive, non-precipitate-forming water chemistry, which is definitely not the case as shown in point 5 above.
7. One significant inference that can be drawn from the test results is the potential to operate the pulsed-power treated system (and other similar systems) at higher cycles of concentration than would be allowed with a chemical treatment regime, thus reducing make-up and blowdown requirements. Since calcium carbonate, alkalinity, and TDS are being controlled or limited by the pulsed-power system, it would seem that the only limitation to ramping up the cycles would be factors such as chloride level (<500 ppm max preferred) and the product of magnesium and silica (<25,000 preferred). These indicated levels should not be exceeded for reasons relating to corrosion and scaling, respectively.
The system in general has been very easy to operate. The routine maintenance consists of cleaning the filter bags and conductivity probe (about once a month) and washing out the tower with a garden hose twice per year. The chillers have been opened during each of the three years in operation. Over the entire three year period, there was no slime layer or biofilm on the inside of the chiller. The first year there was a small quantity of soft deposit in the chillers. This deposit was removed using a power washer with nylon brushes. During the second year, after the blowdown was moved to the bottom of the separators to reduce the frequency of cleaning of the bag filter, there was less soft deposit and again it was removed with nylon brushes. When the chillers were opened after the third year, operating for the entire season with the blowdown at the bottom of the separators, there were almost no deposits. For the entire three years of operation under potentially scaling conditions, no adherent scale was formed. Chiller efficiency is continually monitored and has shown no measurable degradation of heat- transfer properties. The facility operators are very pleased with the results.
The tower has remained free of slime (biofilm) on all wetted surfaces. There has been no odor associated with the system and the water is exceptionally clear.
Pulsed-power is a very effective technology for the treatment of open loop cooling water systems where effective control of biological and mineral precipitation is paramount. The use of pulsed-power completely eliminates the discharge of residual biocides and scaling/corrosion inhibitors into the blowdown or into the air while maintaining an exceptionally clean recycle system.
There has been no evidence of a slime layer on cooling towers under pulsed-power treatment towers. Work being done at Quinnipiac University indicates that pulsed-power not only prevents biofilm formation but also, under most conditions, will destroy a pre-existing biofilm and the entire biological ecosystem that the biofilm supports.
Under pulsed-power treatment, mineral laden water forced into scaling conditions by heat or evaporation will form bulk-solution precipitates rather than adherent surface scale. Modification of mineral precipitation without chemical addition can be a very powerful tool in many industrial and commercial applications. Work currently being done with Yale University's Colloids and Complex Fluids Group suggests that the pulsed-power fields suppress nucleation on surfaces while encouraging growth on particles.
These capabilities of pulsed-power-controlling biological activity and modifying mineral precipitation-have shown additional promise on unit operations other than cooling towers. This technology is also being used successfully on heat exchangers, steam boilers, and once-through hot water systems.
John Lane is the Director of Technology for Clearwater Systems in Essex, Connecticut. (E-mail address: firstname.lastname@example.org).
Mr. Lane holds a B.S. in Metallurgy and Material Science from the Massachusetts Institute of Technology. His industrial experience includes: Laboratory Manager at UNC Resources, Uncasville, CT; Research Metallurgist at ASARCO, South Plainfield, NJ; and VP Technology and President at Aerospace Metals, Inc, Hartford, CT.
David F. Peck, P.E., is Principal Engineer - Water and Wastewater Treatment for Eichleay Engineers & Constructors, in Pittsburgh, Pennsylvania. (E-mail address: email@example.com).
Mr. Peck holds a B.S. degree in Chemical Engineering from Carnegie Institute of Technology and an M.S. degree in Civil/Sanitary Engineering from the University of Pittsburgh. He has over 35 years of consulting and design experience in industrial, municipal, and institutional water and wastewater treatment.
Appendix - Cycles of Concentration at Pittsburgh Case Study
A shorthand method for measuring cycles of concentration is to use the ratio of the recirculating water conductivity to incoming water conductivity. With a chemically controlled system this will be reasonably accurate though slightly over stating the true cycles. Chemical control is usually designed to prevent any precipitation so that all the minerals in the incoming water stay in solution until they are removed via blowdown. It overstates the true cycles since the chemicals themselves will add to the recirculating water conductivity and hence the apparent cycles of concentration but without any real water savings.
The shorthand method using the ratio of conductivities is NOT appropriate with the Dolphin System. The Dolphins allow a bulk solution precipitate to form, relieving mineral concentration from the water. Looking at the ratio of soluble ions such as chloride and sulfate is a better method for determining cycles than using the conductivity ratio.
Applying this technique to the case study (using the data from 8/31/2000), the true cycles of concentration is seen to be 4.7, not the 3.7 that the ratio of conductivity (using the measured value of 1290) gives. The chart below illustrates why the ratios for calcium and alkalinity are so different than those for magnesium, sulfate, and chloride.
Calcium carbonate is being precipitated out of solution at a rate of 147 ppm thus reducing the mineral content and conductivity of the recirculating water. (Precipitated solids do not add to conductivity).
For a chemically
controlled system to have the same water use as that measured at case study
installation in the end of August, the conductivity of the recirculating water
would need to exceed 4.7 X 352 = 1654 mSiemens/cm. Thus in this instance, a
reading of 1290 conductivity for a Dolphin System conductivity is equivalent
of a reading in excess of 1650 for a chemical system.