5(6($5&+ :+,7( 3$3(56 $1' 7+,5' Papers & Evaluati… · S. Frenette Sali Group, Ann Arbor, MI D. Koontz, W. Wang VRTX Technologies, LLC 5850 Corridor Parkway Schertz, TX 78154 ABSTRACT

RESEARCHWHITEPAPERS

AND

THIRDPARTY

EEVALUATIONS

Controlling Cooling Water Quality by Hydrodynamic Cavitation

W. A. Gaines, B. R. Kim, A. R. Drews, C. Bailey, T. Loch Ford Motor Company, Dearborn, MI

S. Frenette

Sali Group, Ann Arbor, MI

D. Koontz, W. Wang VRTX Technologies, LLC

5850 Corridor Parkway Schertz, TX 78154

ABSTRACT An independent field study was conducted on a cooling tower system at an automotive manufacturing facility to evaluate the performance of a VRTX hydrodynamic cavitation device for disinfection, scaling, corrosion and heat transfer efficiency. The VRTX treatment enabled the cooling tower system to operate at higher cycles of concentration than those obtained while on the chemical program without adversely affecting scaling, corrosion and heat-transfer efficiency. On average, water consumption was reduced ~ 10% and blowdown was reduced over 30%. The test results also showed that the VRTX treatment performed as well as the chemical treatment program that it replaced with regard to bacteria control without adding any chemicals. The bacteria population was maintained at equal to or less than bacterial levels obtained while on chemical treatment. No Legionella was detected during the study period. KEYWORDS Cooling water treatment, cooling tower, hydrodynamic cavitation, bacteria control, Legionella, corrosion, scaling, heat transfer, water conservation, water saving INTRODUCTION

Cooling water must be treated to control microbial growth, to prevent scale formation and to limit corrosion of the system components. Due to performance issues at some facilities and increasingly stringent environmental regulations, Ford Motor Company decided to evaluate non-chemical devices for cooling water treatment. After extensive investigation of applications at non-Ford facilities, the VRTX system was selected as one of the non-chemical technologies to be evaluated. The evaluation team consisted of an independent consultant from the Sali Group and engineers with wide ranges of expertise from different Ford

Copyright ©2007 Water Environment Federation. All Rights Reserved

Industrial Water Quality 2007

538

divisions, such as the Environmental Quality Office, Research & Advanced Engineering, Powertrain Operations, and Manufacturing Executive Office. The performance of VRTX treatment on the bacterial activity, corrosion, scale, heat transfer and water consumption was closed monitored during the evaluation period. BACKGROUND INFORMATION 1. Overview of Test Cooling System The test site was located at a Ford facility near Detroit, MI. This facility provides prototype assembly, tooling, and testing analysis of transmissions under development. Test stands and machines are cooled by three closed loop systems. One of the loops was selected for this study. The test cooling system removes heat from the closed loop system through two plate and frame heat exchangers to an open loop cooling system. It consists of a 5,000 gallon cold well, two parallel heat exchangers, and a cooling tower. The total system volume is ~ 11,000 gallons. The cooling tower is a Marley NC series counter-flow two-cell with PVC honeycomb film fill, designed to cool 1,400-gal/min recirculation from 105 oF to 85 oF at 78 oF ambient wet bulb temperature. The heat exchangers are Type A15-BFL plate heat exchangers manufactured by Alfa-Laval Thermal Co. with total heat transfer surface area of 1,792 ft2. Cooling water flowrate, temperature and pressure across the heat exchangers are continuously displayed and recorded. Detroit city water is used as makeup for this cooling system. The water quality is summarized in Table 1. Blowdown is regulated via a conductivity controller connected to a solenoid valve.

Table 1. Summary of Makeup Water Quality

pH 7.6 Alkalinity (mg/L as CaCO3) 75 Total Hardness (mg/L as CaCO3) 108 Ca Hardness (mg/L as CaCO3) 75 Mg Hardness (mg/L as CaCO3) 33 Chloride (mg/L) 15 Conductivity (μS/cm) 220

2. Chemical Treatment Program The chemical treatment program for the cooling water system consisted of an oxidizing biocide (12.5% sodium hypochlorite solution), a non-oxidizing biocide (Isothiazolinone), and a proprietary scale/corrosion inhibitor from the vendor. Each chemical was dosed independently to deliver timed dosing of chemicals via LMI metering pumps in a side-stream loop. Bacteria population was manually monitored by ATP analyzer and the timer set-points



539

for biocides were adjusted based on these readings. The upper limit for bacteria count was set at 100,000 cfu/mL. 3. VRTX Treatment The hydrodynamic cavitation unit utilized in this study was supplied by VRTX Technologies LLC (Schertz, TX). The VRTX unit consists of a pressure equalizing chamber and a cavitation chamber. Inside the cavitation chamber, nozzles are positioned opposite each other at specific distances, lengths and angles. Water is pumped into the pressure-equalizing chamber at ~94 psig and then channeled into the cavitation chamber. Inside the cavitation chamber, water is forced to rotate at high velocities through the nozzles. The rotation creates a high vacuum (~ -28.5 mm Hg). The high vacuum causes micro-sized bubbles to form and grow in the water streams. The water streams in the nozzles are greatly accelerated and rotate in opposite directions. Upon exiting each nozzle, the opposing streams collide at the mid-point of the cavitation chamber where the pressure increases dramatically causing the spontaneous implosion of the micro-bubbles. Cavitation chemistry appears to be effective against numerous strains of bacteria. Of particular note, Stout (2002) has demonstrated the efficacy of VRTX against Legionella pneumophila serogroup 1 of both laboratory and naturally grown cultures in a controlled laboratory setting. Cavitation can also shift the chemical reaction to form CaCO3 colloidal particles. These CaCO3 particles act as the growth sites for dissolved calcium ions to precipitate, instead of forming on equipment surfaces (Koestler et al., 2003). Corrosion is controlled by maintaining water at alkaline pH, controlling bacterial activity, and eliminating corrosive chemicals. The treatment capacity of the tested VRTX unit was 60 gallons per minute. It was connected to the cold well as a side-stream treatment system. A filtration unit was also installed at the cold well in an independent side-stream. The filtration unit was designed to remove CaCO3 and other suspended solids from the recirculating cooling water. Daily monitoring and sampling commenced about 10 days before the VRTX unit was activated to obtain baseline data under chemical treatment. Then, non-oxidizing biocide and scale/ corrosion inhibitor feeds were turned off, followed by reducing conductivity blowdown set point from 1,100 to 450 μS/cm to purge residual treatment chemicals from the system. Additionally, the cold well tank was drained by one-third of its volume. After five days of excess blowdown, the VRTX unit along with the filtration unit was activated and the sodium hypochlorite feed was terminated. The blowdown valve was closed and the conductivity set point was changed back to 1,100 μS/cm two days later. Free residual chlorine measurement showed that its concentration was below the detection limit (0.1 mg/L). After a month of steady-state operation, the conductivity blowdown set point was increased to 1,250 μS/cm to determine if the cycles of concentration could be increased without adversely affecting performance. Such conductivity setting would yield cycles of concentration around 6.0.



540

EVALUATION METHODS

During the evaluation period, the Sali Group conducted daily visits to collect water samples and monitor the system operation, including VRTX system operation parameters, heat exchanger operation condition, and water chemistry data by on-line meters. Water samples collected were shipped overnight for analysis. Wet chemistry and bacterial enumeration were conducted by Paragon Laboratories, Inc. Wet chemistry analyses included pH, Ca, Mg, alkalinity, conductivity, residual chlorine (total and free), total suspended solids, volatile suspended solids (VSS) and total dissolved solids (TDS). Legionella analyses were performed by Pure Earth Environmental Lab, Inc. Information recorded from heat exchanger monitors was used to calculate the overall heat- transfer coefficient (Uo). The overall heat-transfer coefficient is directly related to any fouling of the heat exchangers. Any scale or deposit buildup inside a heat exchanger will inhibit heat transfer and reduce the overall heat-transfer coefficient. Following formula was used for calculation: Uo = Q / (Ao ΔTlm) 1 with Uo = overall heat-transfer coefficient (BTU/hour-ft2-oF);

Q = heat-transfer rate (BTU/hour); Ao = heat-transfer area (ft2);

ΔTlm = log mean temperature difference (oF) = ([T2in – T2out] – [T1in – T1out]) / ln ([T2in – T2out]/[ T1in – T1out]); T2in = heat exchanger process water inlet temperature (oF); T2out = heat exchanger tower water outlet temperature (oF); T1in = heat exchanger process water outlet temperature (oF); T1out = heat exchanger tower water inlet temperature (oF).

Corrosion was monitored using weight-loss coupons made of mild steel, galvanized steel, stainless steel and copper. The coupons were placed in a coupon rack following their galvanic series in seawater: galvanized steel, mild steel, stainless steel, and finally copper in the direction of flow. The coupons were exposed to the system water for 65 days and sent to Garrat-Callahan Company for metallographic analysis. PERFORMANCE OF VRTX TREATMENT 1. Bacterial Control Figure 1 summarizes the heterotrophic plate counts and dip slide results from samples taken from the cooling tower bulk water. Before starting the VRTX study, the baseline bacterial concentration fluctuated daily between 1,000 to 10,000 cfu/mL, with the exception of days when the cooling water was chemically shocked. Right after the VRTX treatment, the bacterial concentration increased by several orders of magnitude to about 1,000,000 to 10,000,000 cfu/mL. This initial increase had been commonly observed during previous pilot



541

tests. A similar phenomenon was also observed when using ultrasonic treatment on suspensions of Bacillus subtilis and attributed it to de-clumping which breaks up bacterial clumps into a greater number of individual bacteria in a suspension (Mason et al., 2003). Therefore, the high bacteria counts here were largely caused by the dispersion of bacteria cells, not an indication of increased bacterial activities. This was further supported by test data. The bacterial concentration declined gradually to the baseline level over the next few weeks. The total heterotrophic plate counts showed that the VRTX unit was able to maintain the bacterial concentration consistently ~10,000 cfu/mL in the absence of any disinfectant addition. The dip slides data showed even lower bacteria population. After the initial "de-clumping" period, most dip-slide data showed levels at or below the detection limit of 1,000 cfu/mL. Figure 1. Bacteria test results using both heterotrophic plant counting (solid squares) and dip

slides (open diamonds).

All Legionella tests showed that the Legionella population was below the method detection limit (1 cfu/mL). 2. Scaling Control CaCO3 scale predominantly forms on the heated surface in cooling water system. A very thin layer of scale deposit inside heat exchanger can significantly reduce the heat transfer rate and, consequently, reduce the overall heat transfer coefficient. For a given system, the tendency of CaCO3 scaling increases with cycles of concentration.

Date



542

Figure 2 illustrates the changes in overall heat-transfer coefficient of heat exchanger during the study period. Despite increased cycles of concentration, the calculated heat transfer efficiencies show marginal improvement trends with VRTX treatment. More data points are needed to see if such change was statistically significant. Nonetheless, these results suggest that the VRTX treatment, at least, is as effective as the previous chemical treatment in deposit control, even at increased cycles of concentration. Figure 2. Calculated overall heat transfer coefficient for heat exchangers #1 (circles) and #2 (squares)

3. Corrosion Control Table 2 lists the coupon test results along with data from two prior tests during the chemical program for comparison. The corrosion rate of copper was equivalent to those obtained during the chemical program; and the corrosion rate of mild steel was much better. The corrosion rates observed under VRTX program fall into the "Negligible or Excellent" category commonly used in industry (Boffardi, 2000). Corrosion rate for galvanized steel was high, as expected, because coupon was not pre-treated. Fresh zinc metal usually has high corrosion rate at pH > 8.5. VRTX recommends all new galvanized systems should be passivated initially. All coupons showed uniform general corrosion. No localized corrosion was observed (see Photo 1). 4. Water Consumption During the first half of the study period, the average makeup water consumption was reduced ~ 25% compared to chemical treatment. The water consumption was reduced another ~30% during the second half of study period when the blowdown conductivity set point was further

Date



543

increased. Part of the reduction in water consumption was contributed by changes in evaporation rate due to changes in heat load and weather condition. The estimates based on cycles of concentration show ~ 10% reduction in makeup consumption and ~ 40% reduction in blow-down discharge during test period.

Table 2. Results of the Corrosion Coupons Test

Corrosion Rate (mils per year) Date Coupon Retrieved

Days Exposed Stainless

Steel Copper Galvanized Steel Mild Steel

03/2000 23 <0.1 1.3 06/2004 61 <0.1 0.5* 12/2004 65 <0.1 <0.1 4.3 0.3

* Coupon was pre-treated. Photo 1. Coupons retrieved after a 65-day exposure (From top to bottom: mild steel, copper, galvanized steel, and 316L stainless steel). DISCUSSION It is known that microorganisms in cooling water can become resistant over time when subjected to a single biocide. Periodic alternating of biocides is often used to attempt to mitigate this. However, it is difficult to determine the optimum product/dosage/frequency to alternate within a given system. Resistance is often first noted by a failure of the process – namely elevated bacterial levels following routine disinfection. VRTX treatment eliminates the potential for this resistance to develop.

Before Cleaning After Cleaning



544

Eliminating biocides also improves morale, health, and safety of working environment. Exposed over extended time periods to biocides, some employees develop hypersensitivity and exhibit "flare-up" dermal reactions. VRTX treatment eliminates the potential for sensitizing to occur. In addition, it eliminates inventory control, ordering, material handling, storage, and training associated with hazardous chemicals. Applying VRTX treatment simplifies plant operation. While biocides must be managed by federally licensed applicators that generally perform testing and instrument calibration, any skilled individual can maintain VRTX system. Moreover, reducing water consumption and wastewater discharge improve corporate image by supporting “green” and sustainable manufacturing. VRTX treatment is also cost effective. Cost analysis indicated the total annual saving was over $15,000 for this tested cooling system. CONCLUSIONS The field study was conducted to evaluate the performance of VRTX treatment with respect to disinfection, corrosion and scaling control. With respect to disinfection, the VRTX unit performed as well as the chemical program that it replaced in terms of total heterotrophic plate count without adding any chemicals. With respect to scaling control, the heat transfer efficiency of heat exchangers was not adversely affected by VRTX treatment even at increased cycles of concentration; in fact, there was evidence of a marginal improvement. With respect to corrosion control, results from weight-loss coupon tests revealed very low corrosion rates except for un-passivated fresh galvanized steel. The observed corrosion rates of copper and mild steel were either equivalent or better than those obtained during the chemical program. Besides eliminating hazardous chemicals and simplifying operation, VRTX treatment can reduce water consumption and waste discharge. Additionally, the blowdown water could be captured and further used for other applications, such as irrigation and urinal flushing, because there are no toxic chemicals in it. The use of VRTX technology can also aid in attaining LEED certification. REFERENCES Koestler, P.; Wang, W.; and Kelsey, R. (2003) New Technology Saves Cooling Water

Demands. Industrial Water Conference, Las Vegas, NV. Mason, T.J.; Duckhouse, H.; Joyce, E.; Lorimer, J.P. (2003) Uses of Ultrasound in the

Biological Decontamination of Water. World Congress on Ultrasonics, Paris, France. Stout, J. (2002) New and Emerging Technologies for Legionella Control: A Multi-Step

Approach to Evaluating Efficacy. 63rd International Water Conference, Pittsburgh, PA.



545

PBMP – Cooling Systems Koeller and Company By James Riesenberger

Commercial - Industrial Cooling Water Efficiency 1.0 Background Commercial and industrial (mechanical) cooling systems have become commonplace throughout the United States and the world over the last 60 years. Improved indoor environments due to air conditioning systems enhance the productivity of millions of workers around the world. In addition, commercial and industrial refrigeration systems lengthen the shelf life of perishable foods, minimizing exposure to harmful bacteria and spoilage, thereby allowing the transportation of these foods over vast distances. Industrial cooling systems help make many processes and products possible we normally take for granted. Since the advent of the ammonia and vapor compression refrigeration cycles, these systems have become an increasingly more important in numerous aspects of our daily lives even though we are largely unaware of their presence. The largest and most efficient cooling systems use water to cool the refrigerant in ammonia and vapor compression cycles. Evaporated water carries away the heat necessary to do this. However, much water is wasted in the controlling of solids concentrated in the cooling water left behind. Water supplies vary in quality from site to site, primarily in the amount and type of dissolved solids which, in turn, require a customized chemical selection and water treatment strategy. In certain areas, chemical treatment against scale formation only seems to have a marginal effect, thus necessitating frequent cleaning of heat exchange surfaces. This is true even though the science of chemical water treatment has improved. Leaving the application of chemicals to the discretion of the water treatment specialists, who may or may not have the knowledge to treat a condenser water systems, rarely maximizes water efficiency. This paper discusses several available technologies that improve the treatment of condenser water, thereby reducing the amount sent to the drain (sewer). Technologies include chemical water treatment, which is the traditional and accepted approach to condenser water treatment, and some newer, non-chemical technologies that approach this important area quite differently than traditional methods. We focus on those technologies that have shown the best track record for long-term water efficiency and successful operation. Appendix A provides additional background information regarding cooling systems, which will help readers to become acquainted with the various technologies and industry terminology. Appendix B concentrates on water treatment technologies and terminologies.

-1-


1.1 History As early as 1882, companies such as Frick and Vilter were manufacturing large steam driven reciprocating ammonia systems for industry that replaced ice as the primary refrigerant. In 1928, Frigidaire discovered a new class of refrigerants, halocarbons or chlorofluorocarbons (CFC’s), which were not as dangerous as ammonia and allowed for smaller compressor design. However, it was not until the introduction of the electrically driven hermetic centrifugal water chiller introduced by the Trane Company in 1938, with its compact design, that cooling of larger, multi-story buildings and industrial processes became feasible. This was previously not practical with smaller chillers with reciprocating compressors or with steam driven reciprocating chillers. Centrifugal chillers are still the mainstay in today’s larger systems, with tonnages ranging from about 300 tons (1 ton = 12,000 btuh) up to about 4,000 tons in a single chiller unit. (One ton cools approximately 350 square feet of conventional office or other open space.) Larger systems are comprised of several chillers piped in parallel. Absorption chillers range in size from about 100 tons up to approximately 2,000 tons and use steam or hot water to drive the refrigeration process, but are less common because they are relatively inefficient and require specialized service and maintenance. Absorption chillers are always used in conjunction with cooling tower technology. Helirotor technology was originally introduced by Frick (now York), Vilter and Dunham-Bush. This newer generation of chillers became very common in the 1980’s when Trane, and later Carrier in the 1990s, introduced their own versions of this technology. The helical rotary, or “screw” compressor chillers, are available in both air-cooled and water cooled configurations, ranging in tonnage from about 100 tons through 500 tons. Since the helirotor compressors are a positive displacement compressor, they work very well in low temperature/process ammonia applications. Frick and FES, a German Company, are two of the larger suppliers of this technology in low temperature applications. Water-cooled systems below 100 tons in size use older reciprocating technology, although this is a dated and relatively inefficient technology. Beginning in the mid-1990s, reciprocating technology has steadily been replaced with scroll compressor technology primarily developed and employed by Trane. Scroll chillers range from 20 tons and up. Although chillers are made in sizes smaller than 20 tons, they are very rare and constitute a very small percentage of the total market. Water Source Heat Pump (WSHP) and Self Contained Variable Air Volume (SCVAV) systems, which are condenser water loop systems, primarily utilize either type of compressor technologies mentioned above. The history of condenser water treatment is dominated by chemical treatment. Chemical treatment has done an adequate job over the years, but the quality of the treatment is subject to human error, chemical limitations, reliability of the treatment equipment, exactness and continuity of control, and the diligence and integrity of the individual(s) maintaining the system. It is estimated that about 99 percent of all chilled water and condenser water (those systems employing cooling towers or closed circuit coolers) are treated chemically. In the past, operators and chemical treatment specialists have mainly focused on keeping the wetted surfaces clear of scale, biological growth and corrosion, not on water efficiency. This is understandable given that in the best of circumstances, proper water treatment is a very difficult task requiring

-2-


constant attention by the chemical treatment specialist, and because water was also relatively inexpensive. Consequently, all decisions regarding water use were left to the chemical treatment specialist who had no concern for or stake in efficiency. 1.2 HVAC Systems

Generally speaking, buildings (with their heating-ventilation-air conditioning – HVAC system) are constructed with the lowest cost system available. There are exceptions, particularly when the HVAC decision is driven by the owner who demands higher reliability, quality, lower sound levels, lower energy consumption or any of a list of reasons to move up to a better-performing system. Chilled water and other water-cooled systems are generally preferred among all available choices. Additional information regarding HVAC systems is provided in Appendix A. DX Air Cooled Applications

In most homes, low-rise apartments and small commercial buildings, some version of a direct expansion (DX) system is found. These systems are generally smaller; a vapor compression refrigerant runs through the refrigeration circuit with the product being cold air that is blown by an internal fan system into the room. These systems are usually rooftop “packaged” units, typically found on a strip mall or small office building (one or two story building), or a split system, similar to what is installed in the typical California home, low rise apartment houses and in many smaller commercial buildings. Some of the largest rooftop and split system units in the 100-ton range can serve buildings up to six stories. Another type of DX system is the packaged terminal air conditioner (PTAC), which are those compact units that are typically located beneath the window. The product of each of these system types is cold air – no chilled water is involved. There is no water consumption in any of these system types so they do not enter into consideration for this analysis. They are mentioned simply to acquaint the reader with the various systems and where they are generally used in different types of building applications. Water-Cooled Applications

As the building square footage increases, particularly in the vertical direction, the system choices available to the system designer become more limited. When a building is larger than 25,000 square feet and almost always when it is taller than the five or six stories, the only practical solution to central air conditioning is to run a chilled water or condenser water loop throughout the building. Many industrial processes require chilled water for cooling the space where the product is manufactured and sometimes this chilled water is also used to cool the equipment in the process itself. There are two types of water-cooling systems in use today:

1) Condenser water cooling (loop) systems where the coolant is sent to a cooling tower and heat is removed through the process of evaporation. The temperature attained is a function of the ambient wet bulb condition and loop temperatures can be maintained between 60° and 90°F.

-3-


2) Chilled water systems, which are further broken down into air-cooled and water-cooled systems. Loop temperatures typically range between 39° and 50°F. Only water-cooled chilled water systems are analyzed in this report since air cooled chillers utilize an air-cooled condenser to remove refrigerant heat even though its product is chilled water. Appendix A provides more detail about this subject.

Water-cooled systems are much more efficient than air-cooled systems, including the air-cooled DX systems, and they can be made in much larger sizes while consuming minimal space. A 500-ton air-cooled chiller might be 11 feet wide and 45 feet long, the maximum dimensions that can be shipped on the back of a flat bed truck. The largest DX air-cooled packaged rooftop unit is around 120 tons in capacity and is about the same dimensions as the 500-ton chiller. Water-cooled systems also require better-trained technicians to maintain them and better subsystem technologies to assist in the maintenance and proper care of these systems. Water-cooled systems typically cost more than DX systems of equivalent tonnage. Condenser Water Systems

Condenser water systems appear in two types of configurations, one where the compressor bearing units are inside the building ceiling plenum, the other when the units are in individual mechanical rooms next to the working space. They are (1) the water source heat pump (WSHP) system that was very popular in the 1980s and early 1990s and (2) a new concept called the vertical self-contained VAV (SCVAV) system. The WSHP system typically has very small air conditioning units placed above the ceiling in two- through five-ton sizes, whereas the SCVAV systems have units sized in the 40- to 100-ton range. Condenser water loop temperatures range between 60° and 90°F. Chilled Water Systems

Chilled water systems (CWS) are designed to deliver water to another device called an air handler or a fan coil unit. These units accept the cold water into an internal coil with water passing through the inside of the tubes. Room air is recirculated from the room and back to the return side of the air handler or fan coil where it is then filtered and passed through the exterior of the coil. The cold chilled water in the tubes absorbs heat from the air passing around it, thus cooling the air as heat is given up to the water. Typically, the chilled water temperature is between 38° and 47°F, easily driving the temperature of the air delivered to the space to 55°F from a 70° to 75°F return temperature. 1.3 Applications

All water-cooled systems have advantages and disadvantages, but the important thing to remember is that they all require cooling towers or closed loop coolers for heat dissipation. The building applications where chilled water or condenser water systems are found will be larger structures of at least 25,000 square feet (requiring 75 tons of capacity). It is not until buildings reach this size that it is economically feasible to invest in the more expensive water-cooled system types.

-4-


Figure 1 illustrates that as building size grows, the likelihood of finding a CWS increases, especially on buildings of 100,000 square feet (300 tons) and above.59 The data shown account for both air- and water-cooled systems but does not account for those buildings with condenser water systems (with WSHP and SCVAV systems) and industrial process cooling loops. Data were not available for these systems types. However, these systems fit into smaller buildings ranging in tonnage from about 75 tons (25,000 square feet) up to about 600 tons (200,000 square feet), which would have the effect of making the three middle columns of Figure 1 somewhat larger.

Figure 2 shows an estimated breakdown of CWS by building type expressed in terms of square footage.60 Again, these figures do not include square footage for condenser water systems due to lack of data. Condenser water systems are used almost entirely in commercial office building applications, which, if included, would enlarge the office column in Figure 2 well above the 5 billion square foot level shown for chilled water systems.

-5-

59 E-SOURCE, Chilled Water Systems, Chapter 8 60 E-SOURCE, Chilled Water Systems, Chapter 8


1.4 Current Condenser Water Treatment Methodology

As mentioned earlier, approximately 99 percent of all cooling towers and closed circuit coolers are treated chemically. The conventional water treatment system in the United States utilizes chemical additives that are typically administered through an automated system. The basic automation includes a TDS (total dissolved solids) meter to monitor the concentration of solids in the water. When the upper limit for TDS concentration (commonly referred to as cycles of concentration) is reached, it triggers an automated blow down or bleed valve to open. When this occurs, blow down water is sent from the cooling tower to the sanitary drain (sewer). As the level of water in the cooling tower sump is lowered, the blow down valve is closed and a valve in the make-up (municipal) water line is automatically opened. The municipal make-up water is of relatively low TDS. As it is introduced back into the system it thus dilutes the condenser water loop when it is mixed with the water remaining in the sump. The system also utilizes metering pumps for the administration of various chemicals taken from stored drums. The higher the TDS is allowed to concentrate in solution, the greater the potential water efficiency. However, at the same time, the higher the TDS concentration, the greater the risk of scale, biological growth and corrosion. The delivery system and the treatment strategy devised by the chemical supplier are designed to control the three areas of concern just mentioned. For the vast majority of customers, chemical treatment has been the only treatment methodology available in the past. The system owner usually accepts without question the results that the chemical supplier is able to achieve, good or bad. Unfortunately, most owners cannot tell the difference.

-6-


The maximum concentration of TDS or cycles of concentration vary from geographic area to geographic area and must be understood by the people administering the chemical treatment. A treatment strategy devised and administered by the chemical supplier must take into account the solubility of the various scale-forming constituents, the pH of the water, and water temperature. Initially, this water treatment specialist must form the strategy from empirical data and knowledge of the local water supply. Later, after a few months of operating experience, visual results and laboratory analysis, adjustments are made as judged necessary by the chemical supplier. The success of any program is dependent upon the quality of the chemical itself, the diligence and knowledge of the individual administering the program, and the integrity of that individual. 1.5 Evolution Toward New Technology Condenser Water Treatment

The U.S. EPA and state environmental agencies have slowly been prohibiting the use of the most dangerous chemicals used in the treatment of cooling water. Hexavalent chromates have been banned since the early 1990s as they were found to be carcinogenic. Other chemicals continue to come under scrutiny with many requiring double containment as a minimum safety precaution. As this trend of chemical elimination continues, the chemical industry must continue the costly process to find new substitutes as replacements for banned substances. This fact alone may drive the industry toward non-chemical treatment systems. However, at this time, it appears as though it will take many years for this to happen. Chemical costs have contributed to the move away from this method of treatment as well. Generally, as a rule of thumb, chemicals cost about $1.00 per installed ton per month. A 1,000-ton system would therefore cost about $12,000 per year for a traditional chemical treatment program. Some non-chemical systems enjoy payback periods fewer than 3 years based upon chemical cost savings alone. Water and sewer costs have also continued to rise, primarily over the past 15 years. As operators realize that these costs play an increasingly important role in building operating costs, there has been increasing attention given to water efficiency in condenser water cooling systems by conservation-minded owners and operators. Finally, poor overall results, including scaled condenser tubes, biological growth in the cooling tower, and/or high corrosion rates with chemical water treatment have also contributed to the switch to non-chemical systems with the promise of better results and often, greater water efficiency. It is clear that some non-chemical treatment systems work very well in selected applications with defined water conditions. Two technologies appear to work well in all applications regardless of water conditions, VRTX’s and Dolphin’s. Both provide superior operation in the three required areas of scale, biological and corrosion control while providing a much greater degree of water savings than the typical chemical program. It is believed that, at a minimum, the concentration of TDS (cycles of concentration) can be doubled with these new technologies. A given program need not be tied to these technologies alone, but all technologies can be considered if results are

-7-


measured and the program incentive is tied to field-verified water savings and to measured scale, biological and corrosion results. Appendix B discusses all of the available technologies including magnetic (permanent magnets), electromagnetic (using DC current), electrostatic, AC Induction, electro-ionization, ozone and depressurization/kinetic energy. 1.6 Effect of Cycles of Concentration (CoC) Upon Water Efficiency

As water evaporates, it leaves behind dissolved solids that are brought into the condenser water system. These solids are indistinguishable from the water itself as they are “in solution”, i.e., part of the water. Particulates in solution are sub-micron sized and cannot be removed through common filtration means but require some means of membrane filtration for their removal. Membrane filtration wastes anywhere from 20 to 50 percent of the water it filters as effluent, depending upon the inlet pressure to the membranes. By itself, membrane filtration is not considered to be a water-saving technology for condenser water systems. As the CoC goes up, certain molecules of minerals begin to precipitate as they reach their individual solubility level at a given condenser water temperature. Precipitation (dropping out of solution) is only bad with certain types of common molecules such as those containing calcium, magnesium, and those two together with silica, which are the primary constituents of scale. Scale formation leads to poor heat transfer in the chiller condenser, thereby requiring more energy to perform at a given air conditioning or refrigeration load. Additionally, scale can assist in the formation of biological fouling and corrosion of metallic surfaces. These subjects are discussed further in Appendix B. For a water-efficient system, the overall goal should be to increase CoC, delivering a lower blowdown rate and conserving water, while, at the same time, not causing long-term deterioration of the condenser water system. Generally speaking, but not in all cases, water in the northern part of California is relatively clear of scale-forming minerals. On the other hand, water taken from the Colorado River and certain underground well sources tends to contain much higher levels of these materials. As a result, CoC’s in the northern parts of California tend to run higher than in the southern parts of the state and are believed to average around 3.0 for the average application in California. This is based upon experience in the field where CoC’s were seen as low as 2.0 in San Diego applications, while known to reach as high as 5.0 with effective chemical treatment programs in the north.. Figure 3 shows the effect of cycles of concentration upon overall water consumption and blowdown. The chart looks at a typical 100 ton-hour system61.

-8-

61 A ton-hour is defined as one ton of load per hour for 100 hours, 100 tons for one hour, or something in between.


Evaporation

Evaporation of one pound of water removes 1,000 BTUs of thermal energy. One condenser water ton is defined as 15,000 BTUs (includes compressor heat). Water weighs 8.34 pounds per gallon. Therefore, one ton-hour requires 15,000 BTUs per ton / 1,000 BTUs per pound / 8.34 pounds per gallon = 1.8 gallons of evaporation is required to dissipate one ton of heat. Blowdown and total water consumption

Blowdown (also termed bleed) is defined as evaporation (gallons) / (CoC – 1.0). It can be seen in Figure 3 that as the CoC is increased, the blowdown rate decreases. At a CoC of 1.5, the blowdown is 360 gallons for a 100 ton-hour system; the evaporated water is 180 gallons, thus totaling 540 gallons of water consumption. By doubling the CoC to 3.0, the evaporated water volume remains the same at 180 gallons. However, by concentrating the water solids to this level, the blowdown is now only 90 gallons and the total water consumption drops to 270 gallons, or one-half the total consumption at 1.5 CoC. By doubling the CoC again to 6.0 cycles of concentration, the evaporated water remains the same at 180 gallons while the blowdown is further reduced to 36 gallons. Total consumption drops to 217 gallons. This constitutes a 60 percent water savings when compared to the 1.5 CoC application and a 20 percent savings over the 3.0 CoC application. Savings continue to increase up to about 8.0 CoC, beyond which point the additional savings attributed to increasing CoC becomes very marginal. It is generally believed that the CoC can be doubled (in most applications) from the average of 3.0 to the higher 6.0 by utilizing one of the newer, more prominent technologies such as VRTX or Dolphin.

-9-


1.7 Size of Water-Cooled Chilled Water and Condenser Water Market

To estimate the total market in California, information was obtained from the Trane Company about the number and dollar value of large tonnage chilled water system (CWS) shipments (all chillers over 100 tons and all vendors) into California during the 26-year period from 1978 to 200362. Total shipments for this period of time were 74,201 chillers at an average tonnage of 371 tons each, equaling an estimated 27.5 million total tons. This 26-year period approximates the real life expectancy of a centrifugal, helirotor or absorption chiller. Because most of these units are likely to still be in service, this is considered a reliable estimate of the current size of the market in California. Shipping information regarding WSHP systems was also obtained from the Trane Company and spanned the ten years from 1995 through 2004 for all vendor shipments into California63. The total shipments for the 10-year period amounted to 76,547 units. The average unit size was 3 tons (these fit in the ceiling plenum) making the total tonnage approximately 230,000 tons installed during this period of time. Extrapolating this amount for the prior 10-year period would double the numbers to 153,094 units and about 460,000 tons installed in California. The life expectancy for one of these units is approximately 20 years.

Information regarding SCVAV unit shipments was not available. However, this system type constitutes a very small portion of shipments when compared to CWS and WSHP systems. An educated guess based upon local knowledge in the San Diego market is that SCVAV shipments constitute approximately 5 percent of the total WSHP shipments for the same period of time, or about 23,000 tons statewide. Total installed tonnage in California represented by this shipment data and estimates, then, amounts to approximately 28 million tons of water-cooled and condenser water systems. 2.0 Water Savings Estimates and California Potential Estimating water savings for non-chemical treatment technologies considered in this paper requires the use of conservative assumptions regarding (1) the estimated average statewide cycles of concentration under current practice and (2) the equivalent full load hours64 of operation by building type. The metric for equivalent full load ton-hours is used quite often in basic energy comparisons between systems or equipment types. For the purpose of this analysis, tonnage of each system type was provided by Trane. Tons installed by building type was calculated by applying the percentages derived from the information shown in Figure 265 and multiplying each by the total tonnage for each system type (CWS, WSHP and SCVAV). Ton-hours was calculated next by multiplying the equivalent full load hours by the tonnage in each category.

-10-

62 Trane Company, ARI statistics of large tonnage chillers into California between 1978 and 2003. 63 Trane Company, ARI statistics of WSHP shipments into California between 1995 and 2004. 64 Equivalent full load hours is computed based upon taking all of the loads in a building over the course of a year and then equating them to hours where the chiller is assumed to run at full load with no part load operation. This metric is expressed in ton-hours. This conversion is done to keep the calculations as simple as possible. 65 Assuming that California is similar to the U.S. statistics in Figure 2 in terms of percent building type of total tonnage.


Next, the comparison of water consumption was performed at 3.0 (a conservative estimate of the statewide average) and 6.0 CoC. This was calculated by multiplying the ton-hours of operation for each building type by the gallons per hour for the specific CoC. The water consumption per ton-hour at 3.0 cycles is 2.7 gallons and 2.16 gallons at 6.0 cycles of concentration. The numbers can be derived from Figure 3, where the effect of CoC on water consumption is shown. The difference between the two represents potential statewide savings that could be achieved with the application of improved or new technologies. Based upon these calculations (detailed in Table 1 below), we estimate that over 10,000 acre-feet of annual water savings are available in California with full implementation of the new technologies to existing systems.

-11-

Table 1: Estimate of Total Yearly Water Savings For All Systems

Building

Type

System

Type

Equiva-

lent Full

Load

Hours

Percent

Ton-

nage of

Total

Tons

Ton-Hours of

Operation Per

Year

Annual

Water

Consump-

tion at 3

Cycles

(Acre-Feet)

Annual

Water

Consump-

tion at 6

Cycles

(Acre-Feet)

Water

Savings

(Acre-

Feet)

Office 1310 0.47 1,318,446 1,727,691,481 14,313 11,451 2,863

Education 1344 0.23 645,197 867,144,660 7,184 5,747 1,437

Healthcare 4368 0.13 364,677 1,592,907,039 13,197 10,557 2,639

Mercantile 2184 0.11 308,572 673,922,209 5,583 4,467 1,117

Lodging 1835 0.06 168,312 308,778,903 2,558 2,047 512

2,805,204 5,170,444,293 42,836 34,269 8,567

Office 1310 0.47 85,897 112,559,691 933 746 187

Education 1344 0.23 42,035 56,494,771 468 374 94

Healthcare 4368 0.13 23,759 103,778,438 860 688 172

Mercantile 2184 0.11 20,104 43,906,262 364 291 73

Lodging 1835 0.06 10,966 20,117,051 167 133 33

182,760 336,856,214 2,791 2,233 558

Office 1310 0.90 413352 541,491,120 4,486 3,589 897

Lodging 1835 0.10 45928 84,277,880 698 559 140

459,280 625,769,000 5184 4147 1037

Office SCVAV 1310 1.00 23,000 30,130,000 250 200 50

TOTAL 51,061 40,848 10,212

SUB-TOTAL

SUB-TOTAL

SUB-TOTAL

Centrifugal,

Helirotor

Absorption

WSHP


3.0 System Cost System costs in the year 2000 for the VRTX system were made available by the manufacturer. Since 2000, most manufacturers have instituted price increases of approximately 15 percent and such an increase is included in the system cost for VRTX estimates that follow. In addition to the system cost, installation labor and materials must be included in the total cost, along with a 25 percent contractor installation margin. Overall, typical installed costs range from $40 per ton on larger systems up to over $160 per ton on smaller systems, with the average for most commonly sized systems at approximately $70 per air conditioning ton. For the average chiller size of 371 tons as noted in section 1.7, the installed system cost would therefore be approximately $26,000 in 2005 dollars. Current (2005) system costs for the Dolphin system were obtained from the manufacturer along with approximations of installation costs, which were based upon their installation experience. The costs assume a retrofit installation that will be more costly than a typical new construction scenario. Again, a 25 percent contractor margin is added to reflect actual installation costs. A 160-ton system was quoted at approximately $16,000 including installation and contractor margin. This equates to $100 per ton. An actual price quotation on a recent 2,400-ton project, again including installation and margin, was approximately $100,000 or about $42 per ton. For the average system of 371 tons, the cost would be about $21,000 installed, or about $57 per refrigeration ton.

Conventional chemically treated systems likewise have a cost associated with installation. These systems require a conductivity controller66; chemical pumps (required to pump the various treatment chemicals from their holding containers into the condenser water system); a make-up water meter, which is used to control the amount of acid (pH) of the condenser water system; and an array of small bore piping used to deliver the various treatment chemicals into the condenser water system. Installed system costs for these systems range from $12,000 for small systems to over $30,000 on larger systems, depending upon the complexity of the installation. If we were to assume a $20,000 installed cost for our 371-ton system, this would equate to an approximate cost of $54 per nominal ton.

VRTX and Dolphin systems yield additional savings not associated with water and sewer cost savings. The chemical costs associated with the treatment of conventional condenser water systems are ongoing and must be added to the operation costs of those systems. However, VRTX and Dolphin require little or no addition of chemicals in most applications. Chemical costs vary by provider, quality of the chemical program being provided, and by the quality of the water being treated. Low-maintenance programs can cost as little as $0.50 per ton per month. In this case, a 1000-ton system would cost 1,000 x 12 months x $0.50 or $6,000 per year. More sophisticated programs can cost three times this amount or about $18,000 per year for the same 1,000-ton system. Generally speaking, the more sophisticated programs tend to run higher cycles of concentration and protect the equipment from degradation for a longer period of time. Using our same 371-ton system as an example, and an average chemical cost of $1.00 per ton per month, this expected chemical cost would be estimated at 371 x 12 x $1.00 = $4,450 per year.

-12- 66 A conductivity controller is a device that measures the conductivity of the water and is set to trigger blow-down of the water and the addition of chemical when the conductivity (high TDS and Cycles) limit is reached.


4.0 Cost Effectiveness The anticipated system life for the VRTX system is estimated at 20 to 25 years, approximately the same as the chilled water systems that they serve. The weighted average equivalent full load ton-hours for all building types is 1,843 ton-hours. Therefore, the water savings on a 371 ton chilled water system would be 1,843 equivalent full load ton-hours x 371 tons, or 684,000 ton-hours. At a 3.0 CoC, this system would consume approximately 5.7 acre-feet of water annually67. At 6.0 CoC, the water consumption would drop to approximately 4.5 acre-feet annually68. The net water savings would amount to 1.14 acre-feet per year for the estimated lifetime of 20 years, 22.8 acre-feet of water. At a net initial cost of approximately $26,000, the cost of the total system would calculate to approximately $1,140 per acre foot saved. This does not take into account any long-term energy savings due to clean condenser tubes or chemical cost savings to the owner. From the standpoint of the owner, the annual savings become significant when the savings related to utilities (water and sewer) are combined with chemical savings. Figure 4 illustrates the owner’s savings potential for the typical 371-ton system (initial cost of $26,000) at various combined water-sewer rates. We expect that Dolphin treatment systems will last the same 20 to 25 years, although the oldest known systems are only about two years old at this date. Using the same weighted average equivalent full load, 1843 ton-hours and 584,000 ton hours of operation for the 371-ton system, the same annual savings of 1.14 acre-feet of savings would be expected. Lifetime savings of 22.8 acre feet would also be expected. Using the initial installed cost of $20,000 for the 371-ton chiller plant as stated above, the cost of this system would be approximately $1,030 per acre foot saved. Again, Figure 4 illustrates the expected benefit to the owner. Because each such system is custom designed to meet the specific requirements of the building cooling system, we recommend that a water agency or municipality considering an incentive program directed at encouraging system upgrades to either a VRTX or Dolphin system incorporate it into a process water program. Process water programs generally provide for variable subsidies for system upgrades and replacements based upon engineering calculations and subsequent measurements of the actual system designed for the purpose. That is, the subsidy is different for each installation, instead of being based upon pre-determined equipment-based amounts.

-13-

67 Computed as follows: 684,000 ton-hours x 2.7 gallons per ton-hour = 5.67 AFY. 68 Computed as follows: 684,000 ton-hours x 2.16 gallons per ton-hour = 4.53 AFY.


-14-


Appendix A

Commercial – Industrial Cooling Principles and Systems History As early as 1882, companies such as Frick and Vilter were manufacturing large steam driven reciprocating ammonia systems for industry, replacing ice as the primary refrigerant. In 1928, Frigidaire discovered a new class of refrigerants, halocarbons or chlorofluorocarbons (CFC’s) that were not as dangerous as ammonia and allowed for smaller compressor design. However, it was not until the introduction of the electrically driven hermetic centrifugal water chiller introduced by the Trane Company in 1939, with its compact design, that cooling of larger, multi-story buildings and industrial processes became possible. Centrifugal chillers are still the mainstay in today’s larger systems in tonnages ranging from about 300 tons to about 3500 tons per single chiller unit. A new generation of chillers entered the market in the 1980s, the helical rotary or screw compressor chiller, in both air-cooled and water-cooled configurations, ranging in tonnage from about 100 tons though 500 tons. Frick, Vilter and Dunham-Bush introduced this technology and Trane, Carrier and others have since followed suit. Since the helirotor compressors are a positive displacement compressor, they work very well in low temperature/process ammonia applications.

Air Conditioning Heat Exchange Process Figure 1 below shows the entire air conditioning process for a water-cooled chilled water system. Five loops with four heat exchanges are necessary for the system to work. Each heat exchange process can be thought of as a sponge, each of which is used for absorbing and rejecting heat.

Figure 1: Air Conditioning Heat exchange Process - Water Cooled System

90 90909090

90 9090 90

90

Return Air 73F

Supply Air 55F

Chilled Water

Return 53F

Chilled Water

Supply 41F

Refrigerant

Saturated Evap

Temp 39F

Refrigerant

Saturated Cond

Temp 92F

Condenser

Water Return

90F

Condenser

Water Supply

80F

Chiller Condenser

Shell and Tube HXChiller Evaporator

Shell and Tube HXCooling Tower

Air Handler (Fan

Filter and Coil)

Atmospheric

Air

Condenser

Water

Chilled

WaterIndoor Air Refrigerant

A-1


Referring to Figure 1, moving from left to right and beginning inside the building, chilled water at 38° to 45°F is pumped through devices called air handlers or fan coil units in order to transfer heat from the interior room into the chilled water loop. The air handler and fan coil units have internal coils through which the cold water passes and cools the return air (from the building space) passing through the external part of the coil. Essentially, the chilled water loop absorbs heat from the air passing through it, dropping the air temperature to approximately 55°F. Air is then recirculated back to the room where it absorbs heat from the room. The air then recirculates back to the air handler and the process starts over again. As the chilled water cools the room air, and absorbs its heat, it rises in temperature and must be re-cooled so that it can repeat the process. To do this, the chilled water loop (which is a closed loop with no evaporation) sends water back to the chiller.

Typically, the chilled water supplied to the air handler is at a temperature between 38° and 47°F, easily driving the air temperature delivered to the building space down to 55°F from a 70 to 75°F return temperature. In turn, the chilled water loop rises in temperature 10 to 20°F, depending upon the design, before being sent back to the evaporator shell and tube heat exchanger of the chiller.

The chilled water loop is cooled by refrigerant inside the chiller, which is hermetically sealed from the atmosphere. Refrigerant in the evaporator is at a temperature of approximately 39°F in order to yield an exiting water temperature of 41°F. The refrigerant absorbs heat from the chilled water, and the refrigerant returns back to the compressor as a gas. In a vapor compression cycle, the refrigerant is compressed and then passes into the condenser where heat is removed from the refrigerant by the condenser water loop. Typically, condenser water returns to the chiller condenser heat exchanger between 80° and 85°F depending upon atmospheric conditions (wet bulb temperature). As it passes through the condenser it absorbs heat from the hot refrigerant and will typically gain 10 to 20 degrees before it exits and returns to the cooling tower. By absorbing heat from the refrigerant, the refrigerant is cooled and passes back into the evaporator, thus completing its loop. The condenser water is pumped into the water boxes on the top of the cooling tower where it is distributed through the media. Within the cooling tower, the water is broken up into smaller and smaller particles as it is exposed to the surrounding air. Because the air is dry, relative to the molecules next to the water droplets, a certain percentage of the water flashes into a vapor where it is essentially absorbed into the air outside of the building. As this process takes place, the remaining condenser water is sub-cooled as air absorbs its heat. Approximately 1000 BTUs69 of heat are removed from the condenser water for each pound of water evaporated. The cooling tower is sized so that the total drop in condenser loop temperature is approximately 10 to 20°F before it returns to the chiller condenser heat exchanger. Cooling Systems Description All mechanical (vapor compression) cooling systems compress a gas to a state of high pressure and high temperature at which point the gas must be cooled to a high pressure, low temperature

A-2

69 BTU – British Thermal Unit; a measure of heat energy


liquid prior to entering an expansion device, where it becomes a low pressure and temperature liquid, ready to absorb heat from the process in the evaporator. At this point, the refrigerant gains heat, converts back to a low pressure and temperature gas and reenters the compressor where the process starts all over again. During the cooling phase, while the refrigerant is in the condenser, two mediums are used for this cooling: air or water. Air-cooled systems, although described below, are not of any significant interest to water conservation professionals.

Water-cooled systems are further broken into two categories: once-through and evaporative systems. Once-through systems use municipal water, lake or river water to cool the refrigerant. They generally take in water, use it in one pass through the condenser, and then send it to drain or back to the original river or lake source. These systems will not be discussed in this analysis. Only cooling systems where the water is used in an evaporative process to cool the refrigerant are considered and addressed in this analysis. In these systems, a portion of the water is vaporized, taking advantage of the principle of latent heat of vaporization of water where heat is absorbed in the process, thus cooling the remaining liquid. This allows the water to be sent back into the condenser for additional cooling where the process repeats itself over and over again.

System Application DX air-cooled applications

Generally speaking, buildings (and their heating-ventilating-air conditioning – HVAC) are constructed at the lowest cost system available. DX (direct expansion) systems fit this requirement best, and is therefore the most prevalent system found in California. Several versions of DX systems are available and they come in tonnages from about one-half ton (6,000 BTU-hours) up to about 125 tons in the largest rooftop style units. There are exceptions as to when an owner or engineer will drive the HVAC decision to a higher-end system, namely, when the occupant demands higher quality, lower sound levels, and lower energy consumption. The decision might be made for any combination of these reasons or a host of other reasons to move up to a better performing system. Generally, on homes and small commercial buildings, some type of a DX system is installed, which is either a rooftop “package” units, typically found on a strip mall (one or two story building) or a split system, similar to what is installed in the typical California home and in many smaller commercial buildings. Some of the largest rooftop units, up to the 125-ton range, can serve buildings of up to six stories. Another type of DX system is the packaged terminal air conditioner (PTAC), which are those units typically installed beneath a window. There is no water consumption in any of these DX system types. Because these are all air-cooled systems, they are not considered in this analysis. Water-cooled applications

As the building square footage increases, particularly in the vertical direction, the system choices become more limited to the system designer. When a building is larger than 25,000 square feet and almost always when it is taller than five or six stories, the only practical solution to building air conditioning is to run a chilled water or condenser water loop throughout the building. Once this decision is made, the building will have either a cooling tower or closed loop cooler, with the one exception being when an air cooled chiller is employed. Chilled water is also required in many industrial processes applications.

A-3


The main advantage of water-cooled systems is that they are much more efficient than air-cooled systems. They also tend to last much longer, 30 years or more in some cases. On the negative side, water cooled systems require better trained maintenance personnel and better subsystem technologies to assist in the maintenance and proper care of these systems. Water-cooled systems typically cost more than air-cooled systems of equivalent tonnage. Finally, water-cooled systems consume water. The other types of evaporative water loops include condenser water loops. Condenser water systems are primarily found in two types of air conditioning systems and in some industrial processes that do not require the temperatures of chilled water. Condenser loop temperatures generally range from 60° to about 95°F depending upon the application, and upon whether the cooling device is a cooling tower or a closed loop cooler. Lower temperatures are always attainable in cooling towers when compared with closed loop coolers, but, in some systems, open loop cooling towers are not advised because they introduce dirt, scale and biological growth in the condenser piping system and, more importantly, into the HVAC equipment being cooled. The two types of systems where condenser water is used are the water source heat pump system (WSHP) and self -contained variable air volume (SCVAV) system. The WSHP system typically has very small air conditioning units placed above the ceiling in 2- through 5-ton sizes (for zoning), whereas the vertical self contained systems have units in the 40- through 100-ton size range. Zoning is accomplished with small variable air boxes for temperature control. Typically, these systems do not employ open loop cooling towers but rather closed loop evaporative coolers. These are both popular systems in California and are seen throughout the major metropolitan office building inventory. Figures showing closed loop coolers and cooling towers are included later in this appendix. Chilled Water System Classifications Figure 2 illustrates that as building size grows, the likelihood of finding a chilled water system increases, especially on buildings of 100,000 square feet (300 tons) and above.70 The figure accounts for both air and water cooled systems but does not account for those buildings with condenser water loops with WSHP and SCVAV systems and industrial process cooling loops.

A-4

70 E-SOURCE, Chilled Water Systems, Chapter 8


Air-Cooled Chilled Water Systems Refrigerant cooling is accomplished in the condenser in one of two ways, air cooling or water cooling. Smaller systems, i.e., 40-50 tons of refrigeration (TR) and below, are overwhelmingly designed as air-cooled systems because of first cost considerations and, in some cases, because of the lack of a reliable water supply for cooling. However, the first cost advantages of air-cooled systems over water-cooled systems are often offset by huge increases in energy consumption, especially as systems approach 300 to 500 tons. Secondly, the space requirements (footprint) for systems over 500 tons begin to become impractical in many applications. Figure 3 shows a basic air-cooled chilled water system configuration. Because the means of cooling is air, this is not one of the system types to be analyzed within this report. Water-Cooled Chilled Water Systems Water-cooled systems are typically larger than air-cooled systems, with the smallest systems starting at 5 tons, but more typically, a small water-cooled system would be thought of as being in the 20-30 ton range. Naturally, as the size of the building and/or cooling load increases, so does the probability of finding a water-cooled system as opposed to any of the previously mentioned system configurations. By the time the building size reaches approximately 50,000 square feet (75 tons), water-cooled systems start to become more prevalent. When the building size reaches 200,000 square feet (300 tons), most chilled water systems are water-cooled. When properly designed, water-cooled systems consume roughly one-half the energy of an equivalent

A-5


air-cooled system operating at full load on “design day”. Figure 4 shows the major components of a water-cooled chilled water system, which matches up with the air conditioning heat exchange process described earlier in this appendix.

Figure 3: Air Cooled Chilled Water System

Air Handlers

`

Chilled Water Pump

Building Envelope

Chilled Water Supply From Chiller

42 FChilled Water Return to

Chiller 52 F

Air Cooled Chiller

Evaporator

Air Cooled Condenser


Condenser Water

to Chiller 85 F

Air Handlers

`

Condenser

Evaporator

Cooling Tower

Condenser Water Pump

Chilled Water Pump

Building Envelope

Condenser Water

From Chiller 95 F

Chilled Water Supply

From Chiller 42 F

Chilled Water Return

to Chiller 52 F

Chiller

Figure 4: WATER COOLED CHILLED WATER SYSTEM

A-7


Condenser Water Systems (WSHP & SCVAV Systems) Figure 5 illustrates a combined water source heat pump and a self contained variable air volume system. Typically, they do not appear in the same building but are shown here for simplicity. Both systems are similar in that they use condenser water with temperatures ranging between 60° and 90°F. They are almost always served by a closed loop cooler in order to keep internal water to the refrigerant heat exchangers clean on the water side. The closed loop is not exposed to the atmosphere but the open spray loop is. The heat exchange is not as efficient as in open cooling towers but this is a necessary compromise.

Condenser

Waterto Building 85 F

`

Closed Loop Cooler

Building

Envelope

Figure 5: WATER SOURCE AND SELF CONTAINED SYSTEMS

Condenser WaterFrom Building 95 F

Condenser Water Pump

5-Ton WSHP's

50-Ton SC VAV

Evaporative Spray Loop

A-8


Heat Rejection Devices Heat rejection equipment is broken down to two main categories: (1) cooling towers as shown in Figure 6, where the condenser water loop serving the chillers is exposed to the atmosphere and is directly evaporated and (2) closed loop evaporative coolers as shown in Figure 7, where the condenser water loop is closed and is not directly exposed to the atmosphere.

Fill

Water Box

Condenser Water to Tower

100 F

Air & Water Vapor

(Rejected Heat to

Atmosphere)

Condenser Water Breaks Into

Droplets Exchanges Heat with

Air and Gravitates to the Sump

City Water

Make-Up

Condenser Water

to Chillers 85 F

Figure 6: COOLING TOWER

Cooling

Tower Sump

Blowdown

to Drain

A-9


Condenser Water to

Tower 100 F

Closed

Loop Coil

Air & Water Vapor

(Rejected Heat to

Atmosphere)

City Water

Make-Up

Figure 7: Closed Loop Evaporative Cooler

Condenser Water

to Chillers 87 F

Blowdown to Drain

Spray Pump

Spray Nozzles

Forced Draft

Fan

Cooling towers are configured in a draw-through (induced draft) configuration, as shown in Figure 6, or are in a blow-through (forced draft) configuration where the fans are positioned near the cooling tower sump and are turned on to push air through the fill which is located above the fans near the top of the tower. When the cooling load is low and the air temperatures are low, induced draft towers take advantage of their natural “stack effect” and provide a fair amount of cooling without turning on the fan motor. Blow-through configurations cannot provide the same amount of cooling without fan operation.

A-10


In an open loop cooling tower, the water is evaporated directly from the condenser water itself. Thus, dissolved solids in the water begin to concentrate. If not controlled, some of the minerals will precipitate out of solution and form scale. In the closed loop system shown in Figure 7, water is sprayed over a heat exchanger located inside the closed loop cooler. This secondary loop (evaporative spray loop) of water is subject to evaporation and the water within this loop undergoes mineral concentration as water is evaporated. However, the closed loop water, i.e., the water that runs out to the water source heat pump or self contained unit air conditioners, never evaporates because it is never exposed to the atmosphere. It is completely free of the scale and dirt that is prevalent in open condenser water loop systems. As a result, the heat exchange surfaces of the individual air conditioners remain very clean. Especially on the smaller water source heat pumps, it is very difficult to clean the coaxial tube condenser coils. If they are not kept clean, the units become very unreliable and maintenance can become a very troublesome problem. This method of indirect evaporation creates an “approach” temperature of a few degrees between the closed loop temperature and the temperature of the open “secondary” loop, where water is sprayed over the closed loop cooler coil mentioned earlier and shown in Diagram 6 above. The air conditioning equipment loses some efficiency with higher condensing temperatures as a result.

Both systems operate on the principle of water evaporation. For every pound of water evaporated, 1000 BTUs (8,700 BTUs per gallon of water) of heat energy is absorbed into the vapor that is created, thus reducing the temperature of the condenser water. The ability of the air to absorb water and convert it to a vapor depends on what is called “wet bulb depression” or the difference between the dry bulb temperature and the (lower) wet bulb temperature. The lower the wet bulb temperature, the more water that can be absorbed, and the lower the temperature attained by the condenser water. This temperature differential of the condenser water to the design wet bulb temperature is called the approach temperature. As this mixture of air, which is now saturated with moisture, leaves the envelope of the cooling tower, it is forced into the open atmosphere. Fans, either induced draft (draw-through) or forced draft (blow-through) provide the means for air flow. Generally speaking, the lower the condenser water temperature, the more efficient the heat exchange within the condenser of the HVAC equipment and the lower the energy consumption. Cooling Tower Operation As seen in Figures 5 and 6, cooling tower and closed loop cooling equipment is relatively simple in design. Condenser water enters the cooling tower, after removing heat from the chiller, having gained 10° to 20°F in the process. The water is pumped into the water boxes in cooling towers and into the closed circuit coil bundle in the closed circuit cooler. The water boxes collect and distribute the water over the fill or media which breaks the water up into discretely sized droplet’s, sized for the optimal contact with air and therefore absorption of heat. This water continues to drop or gravitate through the media as air also moves across this media in a cross flow direction until the water reaches the sump at the bottom of the tower. This is the element of the process whereby heat is removed and absorbed into the vapor which has been

A-11


created by the heat gain. Ideally, the water vapor is removed from the tower confines by the high velocity of air flow and is distributed into the atmosphere above the tower to be swept away by prevailing winds. The sump is simply a basin or collecting place in the bottom of the tower. Water levels can vary, depending upon tower design, from six inches to as much as several feet deep. From here, the water travels through a suction screen (prohibits larger debris from entering the condenser water piping) and into the return piping to the chillers where the process starts all over again. The sump contains blowdown piping that is used to flush out condenser water to the sanitary sewer drain connection. It also has the make-up water connection to the municipal water system, which supplies fresh water when required. A valve located on the make-up line automatically opens when the float valve (similar to the float valve in a household toilet) or electronic level sensor calls for additional water. The initial stage of cooling may simply involve having the condenser water pump run and distribute water to the water boxes. The fan may not need run at low operating loads in this type of tower since there is a natural convective air flow, sometimes called “stack effect” through the tower. The tower fan is turned on when the convective air flow can no longer maintain the condenser water temperature within the desired set points. Most cooling tower fans will either have a two-speed fan motor or a variable speed drive (frequency inverter) to match the tower cubic feet per minute (CFM) of airflow and corresponding capacity to the air conditioning or refrigeration load. In a closed loop evaporative cooling system, the condenser water loop never “sees” the atmosphere, therefore staying very clean and not experiencing any water evaporation. This has the net effect of keeping the heat exchange surfaces of the air conditioning or refrigeration equipment very clean and at optimal energy efficiency. Control is staged in a similar manner to a cooling tower in that the first stage is to simply run the condenser water through the internal coil. Convective air flow removes enough heat to satisfy the heat transfer requirement when a low operating loads. The second stage would be to turn on the fan at low speed. The third stage then initiates the spray pump which takes water from the sump and sprays it over the closed loop coil. This is the water that evaporates and removes heat from the condenser water inside the coil. The final stage of cooling would be to turn the fan on high speed. Because the condenser water must transfer heat through the internal coil of the closed loop cooler, an additional approach temperature is introduced thus elevating its temperature slightly over the corresponding temperature from a cooing tower with no coil. Cooling Tower Manufacturers The market place for cooling tower and closed loop coolers is populated with many participants from around the world. The dominant manufacturers in the United States are: Baltimore Aircoil, Evapco, and Marley. These companies produce a wide range of products that are pre-engineered, “packaged” off the shelf designs with certification of performance. They are supported by a large network of dealers and service technicians throughout the United States. Smaller manufacturers of packaged cooling towers and closed loop coolers are Recold and Delta. Another group of specialty tower manufacturers are the group that includes Tower Engineering,

A-12


Inc., Ceramic Cooling Tower, Tower Performance, Inc.., Cooling Tower Systems, Inc. and Cooling Technology, Inc. This latter group custom designs towers to meet the individual needs of a particular facility or customer to meet site space, architectural, sound or other requirements. Tower construction materials vary from the most common, galvanized steel, to stainless steel, concrete, wood, fiberglass, and an assortment of plastics. Tower tonnages range from small 5 to 10 tons up to several thousand tons.

A-13


Appendix B Condenser Water Treatment Systems

Background

The history of condenser water treatment has primarily been that of chemical treatment. Chemical treatment has done a modestly adequate job over the years, but the quality of the treatment is subject to human error, the quality of the chemicals, the diligence and integrity of the individual technician or specialist performing the work. The main concern of the operators and chemical treatment specialists has been to keep the wetted surfaces clear of scale, biological growth and corrosion. This is a very difficult task that depends upon the constituencies inherent in the water and requires constant attention by the chemical treatment specialist. In general, very little attention has been paid to water efficiency. The main focus has been on the condition of the wetted surfaces of the system equipment and only occasionally on system efficiency. Because it has been the general belief that water is inexpensive, and when coupled with a lack of understanding of water treatment by system operators, the subject of water efficiency is never raised. Consequently, all decisions regarding water have been left to the chemical treatment specialist who has no stake in the potential savings or operating costs. Today, several technologies are available that promise or ensure greater water savings when compared to the traditional chemical treatment approaches. In some cases, these technologies also do not require the constant attentiveness and thoroughness required of a comprehensive chemical treatment program. A technology review and brief description of available treatment systems follows, together with an overview of the advantages and disadvantages of each such technology. It is important to note that these technologies must not only perform all of the basic water treatment requirements (the elimination or prevention of scale, biological growth, and corrosion) but also perform as water efficient technologies. All requirements must be satisfied in order to be recommended as a viable water-efficiency measure or practice. Condenser Water Treatment Program Requirements

The most basic requirements of any water treatment system are three fold:

Scale Control – prohibit the buildup of scale (typically calcium or silica scale) on heat exchanger surfaces. The heat transfer coefficient “k” of calcium carbonate is k=0.4625 (Btu/Hr ft °F) and for silica k= 0.04625 (Btu/Hr ft °F). This is a very low heat transfer coefficient when compared to copper which is k = 227.2 (Btu/Hr ft °F).71 Calcium and silica scale are effective insulators. Scale precipitates as molecules bond to one another or to colloidal particles in the condenser water. This occurs through natural bonding of opposite charges within the atomic structure of

B-1

71 Drexel University, 1999. Overview of Electronic Descaling Technology, Professor Young I. Cho, Department of Mechanical Engineering, July.


the various molecules. As these particles grow, they move from the state of being in solution to one of being in suspension and finally to precipitation where the water can no longer hold them and they either attach themselves to wetted surfaces or drop from gravitational force to low areas of relatively low water velocity. Larger particulate can be filtered, with the finest filters being around 5 microns. However, colloidal particles can be in the 0.01 to 5 micron size range and, as such, settling times for such particles are very slow and they are not filterable without the use of membrane technology. A 0.01 micron sized particle will contain approximately 7 million calcium carbonate molecules and a 1.0-micron particle will contain around 10 billion molecules of calcium carbonate.72 Corrosion Control – prohibit the corrosion of metal surfaces, such as: copper tubes in the chiller condenser tube bundle; steel piping throughout the system; and zinc in the galvanized steel of the cooling tower. Corrosion rates of 2 – 5 mils per year (mpy) ( 1 mil = 1 one thousandth of an inch) are considered acceptable and a rate of 0 – 2 mils is excellent. Corrosion generally occurs when oxygen attacks the surface of iron, copper or zinc and oxidizes it, releasing ferrous and hydroxyl ions into the surrounding water. A secondary reaction then occurs when further (secondary) oxidation takes place, changing the ferrous ion into a ferric ion. Similar reactions occur in copper and zinc. Corrosion can take place in several ways:

General Corrosion - Uniform loss of metal across the entire exposed surface.

Pitting Corrosion – Localized corrosion over a small portion of the exposed surface. Crevice Corrosion - Also a localized form of corrosion where certain anions such as chlorides promote a chemical reaction with hydrogen (H+) ions or hydroxide (OH) ions in water (hydrolysis). Underdeposit Corrosion – One of the most prevalent in condenser water-cooling systems. Scale formed on the surface of metal traps dissolved oxygen behind the scale. Oxygen oxidizes the metal atoms on the surface. The deposit prohibits scale inhibitors from gaining access to the corrosion site.

Galvanic Corrosion – Associated with dissimilar metals connected by an electrolytic solution. Generally speaking, the greater the difference in metal “nobility” the greater the galvanic corrosion rate. Hydrogen ions in the vicinity of the metal surface accelerates the removal of electrons from the metal surface causing a weakening of the atomic bonds of the atoms on the metal surface which break away.73

Biological Control – prohibit the growth of living organisms in the condenser water system. The two main areas of concern are: 1) plant life, such as “biofilm”, “bioslime” or algae, and 2) bacteria. Algae can generally be seen by the naked eye and is most prevalent in the recesses of the cooling tower structure, especially in areas exposed to direct sunlight. Biofilm or bioslime may be

B-2 72 McLachian, David, PhD, 2002. Fundamentals of Water Treatment, Electrostatic Technologies, Inc., Fluid Treatment Solutions, Inc. 73 Gulf Coast Chemical Commercial, Inc., 1999. Corrosion, Revised, June 29..


invisible to the naked eye and grows on any wetted surface of the condenser water system. It is believed that it acts as a kind of glue for the attachment of scale crystals, thus accelerating scale accumulation and/or can form an anaerobic layer of biofouling covered by an aerobic layer of biofouling. As a result, little oxygen and nutrients reach the anaerobic layer accelerating the growth of anaerobic bacteria and upon the death of the bacteria, localized acids promote corrosion in metals, particularly steel and iron.74 Bacteria found in condenser water systems can be very harmful to people. The Cooling Technology Institute (CTI) target value for controlling bacteria is 10,000 CFU/ml (colony forming units). It is important to keep bacteria under control for the maintenance personnel working on or around the cooling tower(s). Legionnaires’ Disease (Legionella Pneumophelia), a deadly pneumonia contracted by the inhalation of water droplets containing the bacteria, has gained much notoriety since the first known outbreak in Philadelphia in 1976 when 29 people died from this single incident. The Center for Disease Control (CDC) in Atlanta estimates that the disease infects 10,000 to 15,000 persons annually in the United States and that approximately 39 people die per week from the disease without anyone knowing that the cause of the death was Legionella.75 (some citations here would be very useful) Although Legionella can, and most often does come from the water in domestic plumbing systems, condenser water systems present a perfect habitat for legionellae bacteria where it is warm (80° to 120°F) and where surfaces covered with biofilm and scale are protected from biocides. Although the U.S. Government and the State of California do not require testing for bacterial levels, including testing for legionellae bacteria, liability does exist for the owner/operator should someone become infected with the disease that could be traced back to their cooling tower site. A comprehensive water treatment program keeps all three aspects, i.e., scale control, corrosion control and biological control, under control and the cooling tower in good working order while, at the same time, maximizing water efficiency. Traditional Chemical Treatment The conventional water treatment system for cooling towers in the United States utilizes chemical additives that are typically administered through an automated system. The basic automation includes a TDS (total dissolved solids) meter to monitor the concentration of solids in the water. When the upper limit for TDS concentration is reached, it triggers an automated blow down or bleed valve to open. When that occurs, blow down water is sent from the cooling tower to the sanitary drain (sewer). As the level of water in the cooling tower sump is lowered, a valve in the make-up water line is automatically opened. Fresh municipal water of relatively low TDS is introduced back into the system, thus diluting the condenser water loop remaining in the cooling tower sump. The system also utilizes metering pumps for the administration of the various chemicals taken from stored drums. The delivery systems and the treatment strategy devised by the chemical supplier are designed to control the three important areas of concern

B-3

74 Zeta Corporation website, 2005. Biofouling and Biocorrosion taken from Role of Bacterial Adhesion in Biofilm Formation and Biocorrosion, Marshall, K.C. and Blainey, B.L. (1991), in Biofouling and Biocorrosion in Industrial Water systems, eds, Fleming H. and Geesey, G.G., Berlin, Springer-Vertag. 75 PM Engineer, 2000. Legionnaires’ Disease, by Matthew R. Freije, July.


discussed earlier. For the vast majority of customers, chemical treatment has been the only treatment methodology available through the years. The system owner usually accepts without question the results that the chemical supplier is able to achieve, good or bad. Unfortunately, most owners cannot tell the difference.

The term “cycles of concentration” (CoC) or “cycles” is often used when referring to condenser water systems. CoC refers to how many times the solids in the municipal make-up water can be concentrated without incurring damage (scale, corrosion and/or biofouling) on the wetted surfaces of the system components. In basic terms, the higher the CoC, the higher the water efficiency of the system. This translates into less water being sent to drain for any corresponding delivery of air-conditioning tons and/or ton-hours of cooling. The maximum possible CoC varies from area to area and must be understood by the specialists administering the chemical treatment. As an example, one CoC would mean that the municipal water level of dissolved solids is maintained in the condenser water loop. Stated another way, the amount of water evaporated equals the amount drained to the sanitary sewer system. For example, a water supply contains 30 PPM (parts per million) of a given mineral, and at that level, the water might allow concentration (with chemical additives) up to 5 times that concentration (five cycles) without risk of scale accumulation, resulting in a CoC of 5. In another geographic area, the water supply may have 125 PPM of the same mineral. Utilizing the same chemical treatment program, the condenser water can only be “cycled up” to a little over 1 cycle while still providing the same degree of scale protection (i.e., a CoC of 1). The constraining scale formation, or in some areas the biological constituencies, determine the allowable concentration levels. A treatment strategy properly developed and administered by the chemical supplier takes into account the solubility of the various scale-forming constituents, the pH of the water, and water temperature, each one initially determined from empirical data, but later from actual experience with the facility and its cooling tower. The success of any program is dependent upon the quality of the chemical itself, the diligence and knowledge of the specialist administering the program, and the integrity of that person.

Scale Control The chemical treatment strategy is fashioned around the worst constituent. Essentially, chemical additives keep the oppositely charged particles from bonding to one another and precipitating out of solution. In many parts of California, especially Southern California, calcium carbonate (CaCO3) is the main ingredient in scale formation, and is the biggest concern of water treatment specialists. Other scale-forming molecules containing silica and magnesium are also prevalent but not to the same degree as calcium in its various forms, primarily calcium carbonate. Therefore, to save water, higher concentration levels of calcium and other scale-forming constituencies must be dealt with in the condenser water system by the chemical specialist.

To do this, pH control (lowering of pH) is administered and is accomplished with sulfuric (lower cost), hydrochloric or citric acid. When acid control is utilized correctly, attention is shifted

B-4


from scale control to the other two concerns, i.e., corrosion and biological control that each become more complex and difficult as a result of low pH. Because of the dangers of handling and administering acid chemicals, the trend today is away from such programs. Instead, phosphonates (organic phosphates) are used when pH (acid) control is not available.76 Phosphonates are often supplemented with polymers or co-polymers that are added to permit the concentration of calcium and other scale forming constituencies to levels above traditional chemical treatment strategy levels. These chemicals keep the calcium carbonate and other ions dispersed in the water, thus prohibiting attraction and the resulting crystalline growth and precipitation. An acceptable average for concentration levels in the State of California is 2.5 to 3.0 cycles with a well-run chemical program. The TDS controller is set so that all dissolved solids never exceed a predetermined level. The worst constituent is typically hardness measured as calcium carbonate, CaCO3. By keeping the TDS below a predetermined level, calcium is kept under the critical concentration level, thus prohibiting it from precipitating out of solution and forming scale. As the limiting TDS is reached, the bleed or blow-down valve is opened for a specific period of time thus diluting the concentration of all solids. This process repeats itself over and over again. Saturated concentrations of calcium to higher levels will force calcium out of solution (precipitation) that will likely result in scale formation. It is important to note that the solubility of calcium goes down as the temperature of the condenser water rises, making the most likely location for scale in the chiller condenser tubes, the warmest point in the system. Heat energy increases the motion of water molecules, which, in turn, increases the motion of ions (charged particles), which collide thus increasing the chance of the formation of insoluble particulate matter (scale). As this process continues, the size and concentration of the calcium crystals increases to the point of insolubility where they no longer can stay in solution and precipitate out of solution into suspension. This is especially dangerous to system operation in that the most important heat exchange in a chilled water system is the one in the condenser. Small changes (additions) in scale formation in the condenser lead to large reductions in overall chiller efficiency and capacity. No single chemical process has totally eliminated scale formation without side effects. Chemical treatment is a “balancing act” between the three basic needs of any program. As ecological constraints on the use of many chemicals continues to narrow the available choices, the effectiveness of chemical programs becomes more and more difficult, especially as they relate to scale related problems and how they affect operating costs, including energy and water consumption. Corrosion Control Since May of 1990, chromates used in many inorganic polyphosphate compounds have been banned from use in treating cooling water in comfort air conditioning systems. These chemicals were particularly effective in corrosion control and actually assisted in biological control because

B-5 76 Harfst, William F., 1999. Water Wars, Water treatment Strategies to Fight Contaminants in Facility Chillers and Boilers, July 16.


of their high toxicity of the chromates.77 Corrosion control in copper based systems is primarily accomplished with alternative chemicals such as triazoles. These chemicals are used primarily to keep surfaces from pitting due to attack from oxidizing biocides. If triazoles are maintained at too low a level, the surface of the copper will not be sufficiently covered and corrosion will occur. Other inhibitors include molybdinate compounds, phosphates, nitrite salts and silicate compounds that are used to coat the wetted surfaces and thus pacify them from oxygen attack. Biofouling, accumulated on the wetted surfaces can also cause corrosion. The treatment program also must effectively kill living organisms, thus prohibiting their growth in the system. Biological Control

Microbiological growths such as algae, biofilm or bioslime, bacteria and mold can find suitable habitats for growth on the wetted surfaces of any condenser water system. This is likely to occur in the recesses of the cooling tower fill and in the spiral grooves of “enhanced” condenser tubes found in today’s chillers. As noted earlier, chromates have been banned from use since 1990. Now, the traditional biocides are oxidizing chlorides and bromides that are effective but sometimes lose their effectiveness against certain strains of living organisms. Additionally, when biofilm is allowed to build up for any lengthened period of time or where wetter surfaces are exposed to direct sun light, these chemicals are sometimes non-effective. Non-oxidizing biocides are now being used such as glutaraldehyde and isothiazolinone with some success. A relatively new technology, chlorine dioxide, has shown excellent results as a biocide but must be produced insitu with an independent system. Chlorine dioxide levels are maintained through direct real time monitoring to ensure effectiveness. Chlorine dioxide is a very strong oxidizer and is selective in the sense that it only oxidizes living matter such as bacteria and biofilm but will not go after metals such as zinc, copper and iron. The administration of these chemicals at the time when they are needed and in the right proportion is somewhat of a balancing act. It takes the diligent care of a knowledgeable chemical treatment specialist to maintain a meaningful and effective program. Cost constraints do not often allow this luxury. The chemical supplier must devote their time to multiple accounts and attempt to keep everything running perfectly. It is important to keep in mind that changes in treatment systems happen quickly, but the deleterious results often occur gradually and take months or years to be discovered. Because of this, they are rarely noticed by the building owner. When lapses in the treatment program occur, the results are not manifested until years later and may require that the cooling tower be replaced.

Chemical Treatment Programs and Savings Verification Because of the hit and miss nature of chemical treatment programs, consideration should be given to excluding this technology from a conservation program unless a very rigorous measurement and verification program is instituted and maintained by an independent third party. The chemical treatment company’s incentive compensation needs to be tied to the verified long-term efficiency of the cooling tower and the resulting water savings. This can be

B-6 77 McLachian, David, PhD, ibid.


accomplished with the addition of meters on the tower make-up line and blowdown line. The ratio of the two readings gives the average cycles of concentration maintained. Periodic physical examinations of the cooing tower and chiller readings must be included in the program. The main areas to be specified are corrosion control held to within strict limits of metal depletion. Accurate logs should be kept of all readings and be readily available for the viewing of the third party inspectors. The best and most accurate way to keep logs is to do it automatically through an electronic system. As mentioned earlier, a chemical treatment program is only as effective as the individual assigned to the given account and the time that is spent administering the adopted chemical program. Even with the most conscientious of specialists administering a chemical program, things can go wrong. If the specialist is not present during changes in the source water quality and/or constituency, the chemical program can be rendered ineffective in a matter of a few hours. In addition, empty chemical drums, inoperable chemical feed pumps, broken TDS meters, or valves, or a malfunctioning controller are all possible impediments to a successful program with sustainable water savings.

Non-Traditional Treatment Technologies

History

Traditional water treatment has been the domain of the chemical industry for the past 70 years in the HVAC industry. Results vary widely based upon the source of make-up (municipal) water and the chemicals used, but results also have a lot to do with the individual specialist supplying the chemicals and the profitability of the company providing the services. In some cases, the results of chemical water treatment have been marginal and even poor in the extreme, encouraging customers to look for better, less costly methods of water treatment. The history of non-chemical water treatment is long and very controversial.78 Magnets were the first non-chemical approach to water treatment to be tried. Benefits of magnetic lodestones were understood and used to decrease the formation of scale in cooking and laundry applications. This technology has been investigated and adapted for use in various water treatment applications since the turn of the 19th century. However, it was not until the availability of rare-earth element magnets, solid state electronics, and advanced ceramic materials, that non-chemical treatment systems became commercialized during the environmental movement starting in the 1970’s79 Technologies utilizing magnetic, electromagnet and then electrostatic principles, dominate the non-chemical suppliers of water treatment technologies. Other technologies, not utilizing these principles, have also been introduced in more recent years and are also be discussed in this paper. In response to this trend away from the traditional treatment of water through chemistry, the

B-7

78 Huchier, Loraine A., P.E. MarTech Systems, Inc., Lawrenceville, NJ, no date. Non-Chemical Water Treatment Systems: Histories, Principles and Literature Review. 79 U.S. Department of Energy (DOE), 1998. Non-Chemical Technologies for Scale and Hardness Control, DOE/EE-0162.


chemical suppliers have responded with advanced chemical treatment programs that have demonstrated a better ability to deal with scale and biological fouling, thus allowing higher levels of water savings through reduced blow down. Since the 1970s, scores of suppliers of non-chemical “systems” have come and gone. The claims for the effectiveness of their systems were matched only by the claims against them! Early systems, due to misunderstanding and misapplication of the technology, failed or the results did not meet the claims of the seller or expectations of the buyer. This, then, led to the general dismissal of this approach to the treatment of water. Chemical supply companies, who were temporarily displaced by the new technology, became very vocal in attacking all non-chemical technologies for understandable reasons.80 This has been the primary resistance factor impeding a more rapid growth in non-chemical technologies. New technologies entering the marketplace must always overcome serious obstacles. For the most part, newer technologies are developed and marketed by small under capitalized companies that do not have the backing necessary to fully develop the technology’s efficacy with supporting documentation and testing. Most of the claims by the manufacturers are not substantiated with scientific field-testing and this naturally leads to skepticism on the part of knowledgeable building owners and engineers. The success of both laboratory controlled conditions and field testing have been unpredictable.81 For the most part, testing (if any) is performed by the manufacturers themselves, which has added to the skepticism about their claims. Comprehensive testing is very expensive and, for the most part, has been non-scientific. Because cooling system operation is very dynamic, fluctuating cooling loads, varying water conditions and temperatures make controlled testing very difficult. Additionally, manufacturers have tended to focus their attention on a particular area of performance and not on all aspects of water treatment necessary for a complete and comprehensive treatment program (i.e., scale, biological, and corrosion control). Results, then, are limited in scope and incomplete, often little better than anecdotal information. Additionally, most of the manufacturers and the people conducting their testing seem to have little or no knowledge of heat transfer equipment, particularly chillers and chiller performance. Their assessment of overall chiller efficiency and corresponding energy calculations appear to be limited and do not hold up with the chiller manufacturers. None of the technologies surveyed have successfully engaged the support of a major chiller manufacturer and thereby gained the support and understanding that they could bring in calculating energy consumption and verification of savings. Unfortunately, most of the firms engaged in this relatively new industry either do not know how to go about comprehensive testing or do not have the financial backing to run scientifically based testing to measure and validate results of their product. Further, without the staying power to learn and then fix a given problem, the easiest path for many was to walk away form the problem or give the money back to the buyer.

B-8

80 West, Burke A., P.E., no date. Non-Chemical Water Treatment Technologies for Cooling Towers. 81 Huchler, Loraine A. P.E., ibid.


Almost all of the manufacturers surveyed are one-dimensional and focus on one aspect of water treatment, usually scale control or biological control. Several made reference to water conservation or water efficiency but did not elaborate on how it was accomplished or how much efficiency could be anticipated. The emerging water treatment technologies vary from magnetic (permanent magnets), electromagnetic (using DC current), electrostatic, AC Induction, electro-ionization, ozone and depressurization/kinetic energy. Scores of companies have entered into the market as suppliers of water treatment technologies for condenser water systems. Unfortunately, many, if not most, have gone out of business. The reasons for their failure are many. However, the primary reasons are that many overstated savings and performance claims, mainly because they either misunderstood the technology they were selling or they unknowingly misapplied the technology. When water conditions and chemistry, which vary greatly from location to location, were right, the project worked. When the same system was applied to another location with different characteristics, it might not have worked because the necessary analyses of site conditions were not taken into account at the new location. All water treatment technologies, including chemical, work best in recirculating systems with sufficient dosage and contact time.82 The second reason for systems success occurs when they naturally precipitate aragonite (calcium) vs. calcite (calcium) crystals, the aragonite being softer and looser than calcite and therefore less prone to form monolithic sheet scale on wetted surfaces. It is obvious that some of the companies continue to conduct business after many years and, as such, have been successful in their selling effort. That appears to some to be the best evidence of their success and the efficacy of their technology. Even Steven Lower, a retired professor from Fraser University in Vancouver, Canada states that his criticism is not to condemn the technologies discussed, but to warn the buyer that they should be extremely cautious and insist on strong measurable performance guarantees.83 It is also true that the scientific community does not even agree as to how or why the technologies work and under what conditions the technology will work repeatedly on a long-term basis. However, there is wide recognition that there is interaction between magnetic, electromagnetic and electrostatic fields and crystallization of matter that affects scaling. Magnetic and Electromagnetic Technologies Magnetic technologies are the oldest of all water treatment categories to be discussed in this paper. The first patent on an electromagnetic water treatment device was awarded in 1890. Most of the literature found regarding this subject devotes itself to the treatment of scale and its prevention. Very little attention is devoted to that of corrosion and the treatment of biofouling

B-9

82 West, Burke A. P.E., ibid. 83 West, Burke A. P.E., no date. Non-Chemical Water Treatment Technologies for Cooling Towers, includes quotation by Steven Lower, Department of Chemistry, Simon Fraser University.


and bacterial control. However, by virtue of the fact that the control of scale formation is the restraining factor to water efficiency in cooling towers, the technology is worth further investigation and discussion. The application of magnetic water treatment requires the installation of a permanent or electrically induced magnetic force field (flux) in or around a non-magnetic pipe material through which the system water flows. The magnets can be placed on the makeup water line where the water is in contact with the magnetic flux one time but, generally, the magnets are instead placed on a recirculating line for maximum contact and maximum effect. The magnets can be either invasive (mounted inside the vessel or pipe) or non-invasive (wrapped around the pipe). The general operating principle for magnetic technology is a result of a moving ionized fluid (flowing water which causes a minute electric current) through a magnetic field. When an ion passes through the magnetic field, a force (called the Lorentz Force) is exerted on each ion. These ionic forces within the cooling tower system ultimately result in the precipitation of colloidal particles. The attraction to these colloids, which act as naturally favored sites for nucleation (coagulation) and precipitation, is very favorable in comparison to formation of scale on the piping walls. Since a percentage of calcium carbonate is removed from the soluble state, the ionic equilibrium balance is such that the water is now able to reabsorb some existing scale. The second phenomenon is that the calcium precipitates out of solution as aragonite, a soft, loosely bonded crystal that is easily suspended and removed in blowdown. This is preferred over calcite, which has less desirable characteristics and is the primary constituent of surface scale.84 (Some scientists believe the calcite precipitate to be more desirable than the aragonite version which is in direct opposition to other findings and the statement above.) It is believed that magnetic fields somehow break up the hydrogen bonds in water, releasing dissolved ions that promote nucleation of the smaller, more desirable aragonite crystals. There appears to be consensus on the patterns observed in magnetic water conditioning which are as follows:85

• Hydrogen bonding affects the surface tension or “wetness” of water. By breaking a very small percentage of these hydrogen bonds, its reactivity is increased.

• By flowing water through a strong magnetic or electrostatic force field, hydrogen bonds are broken.

• Multi-pass recirculating systems have better anti-scaling characteristics than single pass systems.

• Scale tends not to form where there is higher water velocity or turbulence. • Calcium or magnesium will scale in water producing calcium carbonate once its

saturation level is exceeded. • The structure of calcium carbonate scale without other competing metals such as iron,

copper, magnesium, and zinc will precipitate as calcite. With sufficient levels of these other metal ions, the precipitate will be the more desirable aragonite form.

• Aragonite is a less stubborn scale form than calcite and precipitates as a suspended solid or easily crumbles if deposited on a system surface.

B-10

84 U.S. Department of Energy (DOE), ibid. 85 West, Burke A. P.E., no date. Non-Chemical Water Treatment Technologies for Cooling Towers.


• Silica must also be present for the calcium to precipitate in bulk solution versus on the boundary surfaces.

• The technology is not effective on silica scale. • There is a “memory” characteristic in this treatment that keeps its anti-scaling properties

for up to 143 hours after treatment • No mention is made in how corrosion or biological fouling are dealt with. • It is possible that in water containing “appreciable” levels of iron, the technology is not

effective and is therefore not applicable. The success of the technology requires several preexisting conditions and installation requirements to be met.86

• The magnetic field strength and intensity must be sufficient. • The water flow rate and exposure time of the field on the water are critical. • The properties of the water must include other minerals such as iron, magnesium and

copper must be present to form the lesser tenacious scale, aragonite.. • The water must have sufficient silica to promote bulk solution precipitation. • The magnet or electromagnet must be located away from high voltage sources

Manufacturers (Magnetic)

Magnatech Corporation, Superior Manufacturing Division has been in business since the 1960s and is probably the best known of all magnetic system manufacturers. They support magnetic technology for several types of applications including condenser water treatment systems. They mention the formation of aragonite scale versus calcite by use of a cobalt alloy permanent magnet that has dual polarity. The Superior device is installed intrusively as a spool piece in a water recirculation line.

Manufacturers (Electromagnetic)

Triangular Wave Technologies is a multifaceted manufacturer of water treatment systems and devices including those used in condenser water treatment applications. Their device is a non-intrusive magnet wrapped around the pipe, typically of a recirculation line of the cooling tower. The literature addresses scale control very extensively; however, very little attention is given to biological or corrosion control. Electronic/ AC Induced Electric Field Technologies Electronic water conditioning technology was first seen in the late 1990s, making this a fairly new technology with respect to the others being considered. This technology uses a time-varying electronic current in a non-intrusive solenoid wrapped around a pipe to create an induced electromagnetic field inside the condenser water pipe. The magnetic field is grown, then shrunk to nothing, and then reversed. This process occurs over and over every few milliseconds. This changing magnet field, in turn, induces an electric current which is circumferential within the pipe as quantified by Faraday’s law of induction. The electric current is directly proportional to

B-11

86 West, Burke A. P.E., ibid.


the rate of change of the magnetic flux and its direction reverses with the reversal of the magnet field. Similar to that of magnetic or electromagnetic treatment systems, induced electric field technologies increase the repulsion and attraction intensity of ionic constituencies in the water, thus increasing the nucleation and precipitation occurrences. This occurs in bulk solution as aragonite crystals. These crystals are then easily removed in blow down or filtration. Manufacturers

ED 2000. Originally developed by Dr. Cho, a professor at Drexel University, the technology appears to actually be a combination of an electromagnetic and an electrostatic device utilizing much of the same science and attaining similar (expected) results. This technology is sometimes called ED (electronic descaling) or EAF (electronic anti-fouling) technology. This manufacturer does not appear to be very active in the pursuit of new customers, at least not on the west coast. The system will be explained in the next paragraphs.

Clearwater Technologies Corporation, Dolphin System. This system is very similar to the ED 2000 in most all respects. In fact, a very worthwhile technical paper about the Dolphin technology makes reference to Dr. Cho’s contribution.87 This same paper goes into considerable depth regarding the technology, water efficiency as it relates to cycles of concentration, and energy efficiency. An important discussion relating to the measurement of cycles of concentration is given in this paper and explains how the traditional measurement of the ratios of “condenser water concentration to that of the make up water” is appropriate for chemically controlled systems, but it is not germane to a Dolphin (and other) system(s) that remove calcium in bulk solution. A better method for determining cycles of concentration and water efficiency is by metering or by measuring the ratio of other soluble ion constituents such as chlorides or sulfates. This is by far one of the best papers regarding the efficacy of the technology as it relates to water efficiency while discussing all other necessary parts of a comprehensive water treatment program. In brief, the Dolphin system works by the frequent (500 times per second) pulsing and reversals of a magnetic field that parallels the flow of water. As described above, this magnetic field, in turn, induces a circumferential electric field inside the pipe that grows and shrinks with the alternating magnetic field. Clearwater claims that there is an interaction between the two fields which enhances surface charge and precipitation behavior of charged particles although this interaction is not discussed in the literature. The Dolphin literature does discuss biological control stating that the technology is “bacteriostatic” in nature, not a true bactericide that kills bacteria. Instead, the bacteria are not immediately killed but are “controlled”. Dolphin claims that pulsed-power is the basis of low temperature pasteurization in the dairy industry as approved by the FDA in technology developed by Maxwell Laboratories in San Diego.

B-12

87 Lane, John and Peck, David F., 2003. Condenser Water Treatment Using Pulsed Power, Cooling Technology Institute, February.


Corrosion control is attained indirectly in the Dolphin system by maintaining sufficient cycles of concentration to force the system into an alkaline state at the saturation point of calcium carbonate, which is a natural corrosion inhibitor. This technology appears to be well-supported by the manufacturer and they take a total system approach to water treatment and savings. Clearwater does have a very large distribution network in the United States and has over 2,000 installations. This is a very popular technology at the present time and it is believed that there are several installations on commercial and industrial condenser water systems in California.

Electrostatic Water Treatment

The difference between magnetic or electromagnetic water treatment and electrostatic treatment is that instead of a magnetic field, an electric field is imposed on the water flow. The electric field, in turn, generates a magnetic field. The science as it affects the results of water treatment are similar to that of magnetic water treatment. This technology is typified by an insulated cylindrical electrode placed intrusively into the water stream with an externally grounded outer metal housing. Water flows between the two components. A very high voltage, low amperage (micro-amps) electric current is placed on the electrode. The device can be placed in side stream or full flow water streams. The system design is such that the water will run through the device several times a day. Although the science is different, electromagnetic water treatment is very similar to that of magnetic water treatment in the way that ionic structures are affected. For the most part, the results are the same as stated in the previous section. Two systems, the Ion Stick and the Zeta Rod are all coated electrodes inserted into the condenser water pipe. Water flows between the electrode and the outer walls of the pipe. Similar to the configuration cited above, all technologies create an electrostatic field between the electrode and the walls of the pipe. The main discussion of this family of systems deals with scale control and removal of existing scale on wetted surfaces. Although some manufacturers claim bio-fouling control as well, very little information is given as to how this is actually accomplished.

It is believed that as hardness and alkalinity are driven to high limits with increased cycles of concentration, the piping and other metal surfaces are protected from corrosion.

Although this science is less established than magnetic water treatment and few technical papers are available for review, it is widely believed that this technology works best when suspended matter, and especially colloidal particles are kept in suspension. These particles act as nucleation sites for homogeneous nucleation of scale in bulk solution. It is this reaction in the bulk solution of the water stream that lessens the tendency to form scale on the heat transfer and other wetted surfaces of the condenser water system.88

B-13

88 Huchler, Loraine A., ibid.


Manufacturers

York Energy Conservation, Ion Stick. The Ion Stick was first marketed in 1978, primarily as a scale control device. The literature goes into some detail about the electrostatic field that is generated and its ability to enhance and intensify the electrical charge of positive and negative ions in the water, causing similar charges to repel with greater force and dissimilarly charged particles to attract with greater force. This intensified ionic activity is supposed to prevent scale formation on wetted surfaces but also allows some precipitation that is removed through blowdown or through a 10-micron side stream filtration system. The Ion Stick is nominally sized for full condenser water system flow for best results. The literature mentions it being mounted in the cooling tower sump or in an elbow of the condenser water piping. Either 115 VAC or 230 VAC is supplied to the system power pack where it is converted to DC voltage. The only reference in product literature to biological and corrosion control is that the system is kept clean of scale, which lessens the effects of treating these concerns. The literature also states that a conductivity meter is needed to keep cycles of concentration from becoming too great. It therefore recommends blowdown or bleed as the remedy for this occurrence. The literature also states that this technology may not be the complete replacement for chemical treatment. No specific mention is made about water savings potential. Instead, most of the discussion concentrated on scale prevention, which exposes its focus on and predisposition toward energy conservation. Zeta Corporation, Zeta Rod. The Zeta Rod was first marketed in 1990. The literature goes into great depth and explanation of the science behind the technology, particularly about the zeta potential, the force of the surface charge and the resulting repulsive forces on all wetted surfaces, particularly between colloidal particles in aqueous suspension. The literature speaks of a double layer of ionic particles surrounding colloidal particles that is essentially balanced in charge causing no further flocculation of particles and thus a stable dispersion. In water supplies with high concentrations of polyvalent ions such as calcium or iron, the system voltage must be increased from 10,000 VDC to the 30,000 VDC range for effective water treatment. Biofouling, as addressed in the literature, is achieved in a similar manner to the way in which scale is prevented. That is accomplished primarily by the ability of the system to surround biological matter suspended in the water with layers of ions; those ions repel ions surrounding other biological matter thus keeping them from bonding with one another and forming a biofilm layer on wetted surfaces. Without a biofilm layer, the habitat and food for microbe reproduction is removed and prevented. Corrosion control is discussed extensively in the literature, mentioning that the two most prevalent types of corrosion are galvanic and biocorrosion. Biocorrosion is addressed primarily by the prevention and elimination of the biofilm layer on wetted surfaces, thus preventing

B-14


metabolic action and oxygen concentration on and under the biofilm surface. Secondly, as the depth of the biofilm increases, the lower layers become anaerobic as oxygen can no longer reach these lower layers of matter. Corrosion is accelerated in these areas as anaerobes are created which are able to metabolize metals. Their waste products contain acids, such as nitric, sulfuric and other organic acids, that further accelerate corrosion in these localized areas. The technology is mounted in the recirculation pipe of the condenser water system but it also can be mounted in tanks or other locations as long as it is subjected to a constant flow of condenser water. The technology accepts 115 VAC or 240 VAC. No mention is made as to the extent or potential for water savings although this must be assumed with higher cycles of concentration. Again, we see an overall focus on scale prevention and energy savings, not water efficiency. Electrolysis (Ionization)/Electrostatic Water Treatment This unique technology incorporates two technologies into one system. By passing water though an electrode cell, it imparts benefits much the same as electrostatic or magnetic water conditioners, making colloidal particles favored nucleation sites for crystalline growth and precipitation in bulk solution like that mentioned earlier. It is believed that biological matter is also destroyed as it passes through the relatively high electrical energy inside the electrode cell. However, this technology has enhanced biological control because it electrolytically generates and releases ionic copper and silver (or other metals such as zinc or magnesium) in order to attain a residual effect from these metals as biocides throughout the entire condenser water system. Biological Control: The bactericidal effects of electrolytically generated copper and silver ions on bacteria and algae are well documented. Released copper and silver ions remain in solution and act continuously to kill biological materials at concentrations of 0.2 to 0.4 ppm (parts per million) of copper, and 20 to 40 ppb (parts per billion) (0.002 to 0.004 ppm) of silver. Copper/silver treatment is biocydal. This is in sharp contrast to typical chlorine/bromine-based treatments which are only bio inhibiting to much of the living matter in the condenser water. Chlorine concentrations degrade over time, requiring constant additions and changes to the strategy as organisms develop immunities to given chemical strategies. In systems that are shut down over the winter, chlorine levels drop, and the system lies unprotected from algae (and harmful pathogens) growth for months. This allows biological matter to form in chiller tubes, and elsewhere in the system, which encourages scale growth. Copper and silver remain in solution indefinitely, which prevents biological growth, even in systems that are shut down for very long periods of time.

Scale Control: The mechanisms employed to remove minerals are electro-coagulation, mineral precipitation, crystallization and filtration. The first phase of the ionization process is initiated by the electrode flow cell, where an electrical charge is placed across the electrode, causing a simultaneous release of copper/silver ions which are discharged into the water. As with previous technologies discussed in this paper, the electric field causes an increased molecular activity and

B-15


intensity creating a much greater number of collisions between positive and negatively charged particles, particularly colloidal particles and positive ions. These relatively large colloidal particles now act as nucleation sites for calcium and other positively charged (cation) particles having passed through the electric field of the flow cell. One of the manufacturers believes that the same principles associated with electro-coagulation and colloidal particles also apply to the copper and silver ions as well. It is claimed that these also act as nucleation sites, calcium crystallization causing a very pronounced and rapid reduction of calcium carbonate in solution. As the crystalline structures grow they are either filtered out of the water in the side stream filter or become so large that they fall to the bottom of the cooling tower sump as they can no longer remain suspended in the relatively low velocity water found there. It is believed that most of the calcium crystals formed are those of aragonite, which are easily removed through backwash of the filter, blowdown or manual vacuuming of the sump. Small crystals of aragonite scale form on the cathode of the flow cell, and are removed by the high velocity of the flowing water and by the reversal of polarity of the flow cell electrodes. Because all of this activity is done in bulk solution, the scale is formed around the nucleation crystals rather than on the condenser tubes or other wetted system surfaces. Because the system water is demineralized of much of the scale forming calcium carbonate previously in solution, the water is such that it can now reabsorb calcium (existing scale) back into solution much the same as previously technologies discussed. This phenomenon eventually cleans all of the wetted surfaces of the system reducing corrosive effects and improving heat transfer. Corrosion Protection: Corrosion occurs in a system due to several phenomena discussed earlier in this paper. Two of the primary reasons for corrosion are either galvanic or under deposit corrosion. System pH is normally maintained between 8.0 and 9.2. Under these conditions, metals establish a thin layer of natural protection, which is not penetrated by oxygen and which also minimizes electrolytic action. Because the wetted surfaces are kept very clean, problems associated with under deposit corrosion are also minimized. Manufacturers

Baker Hydro Filtrations, Inc., Pure Treat, has been marketed and sold since the early-1990s as a complete system solution. Besides the flow cell technology and controls, this manufacturer offers the other technologies necessary for a complete installation, including the filtration, various media types, pumps, strainers, piping specialties and the local site support needed for first time installations. In addition to the basic technology, Baker has been a leader in filtration media technology using a natural zeolite in many of their installations. The zeolite actually enhances the calcium removal process, unlike the other manufacturers that use sand or some variation of a centrifugal filter. The main difference advantage of this approach is that the zeolite does basic filtration down to 5 microns, automatically backwashes and serves as an additional nucleation site for calcium crystallization.

B-16


The micro environments immediately surrounding the individual zeolite media particles approach a supersaturated condition as additional cations are introduced and natural absorption and desorption occurs within the condenser water system. The media, selected for both its ion exchange capabilities and its unique surface structure, encourages precipitation of the scale forming species on its surface. As calcium carbonate crystalline growth occurs upon the surface of the media, the scale forming species within the water are precipitated within the vessel and begin to adhere to the media. Violent and frequent backwash is necessary to “fluidize” the bed, resulting in dislodgement of the calcium crystals from the media and removal. The Brinell hardness of the zeolite media, approximately 5.0, is greater than that of either calcite or aragonite so the mechanical dislodgement is easily achieved with frequent and highly agitated backwash cycles. Some areas such as the San Francisco Bay area will not allow copper effluent to be discharged to sanitary sewer, therefore making this technology inappropriate for use there. This is the case even though the concentrations of copper and silver ions in these systems are well below EPA levels recommended for drinking water. These limits are 1.2 ppm for copper and 100 ppb for silver. The system operates at a maximum of 0.4 ppm copper and 300 ppb silver. Cooling tower discharges that contain these trace elements do not have any measurable negative effect on the environment. Overlooked unfortunately, is the positive effect of elimination of normal chemicals used in cooling tower treatment, which are very detrimental to the environment.

Oxion Water Technologies is mentioned in several literature pieces and shows a robust web site with much product information. The technology appears to be very similar to that of Baker’s Pure Treat system with the exception that this system uses copper and zinc instead of copper and silver. The literature states that the results are good with an entire section devoted to savings of chemical, water, energy, and maintenance costs.

Ozone Water Treatment Ozone water treatment systems were first introduced in the 1890s and are gaining general acceptance in a variety of applications pertaining to the water industry. The power of ozone to disinfect water is well known in the industry. Today, ozone is primarily used for purification of drinking water at water treatment facilities and is in wide use internationally. Ozone has also been gaining acceptance as a viable means of treating condenser water systems since the mid 1980s. Ozone works best in ambient water temperatures around the 65° to 70°F range and becomes increasingly less effective as temperature rises. Most condenser water systems are designed to operate in the 80° to 95° range where ozone is still potent. However, ozone becomes totally ineffective at 135°F and above. Many companies selling ozone systems for cooling applications have gone out of business over the last 15 to 20 years. It is probably true that although their owners may have understood ozone technology, they probably did not understand commercial cooling systems operations and probably did not have the financial backing to see themselves through problem jobs.

B-17


Historically, new companies that have introduced this technology, have overstated potential savings to the customer and they did not understand the importance of mineral constituencies in the water supply, which led to a range of scale and other problems. The principle behind ozone treatment is that it is a very strong oxidant, O3, that works well in most condenser water system temperature ranges. Ozone is a powerful oxidant that works 3,000 times faster and is 150 percent more effective and powerful than chlorine but without chlorine’s dangerous properties. The ozone molecule is composed of three oxygen atoms that are very unstable (very reactive). Ozone (O3) ranks second in oxidizing potential (fluorine gas is number one but is extremely toxic), whereas chlorine (bleach) is number 16 on the scale of oxidizing potential. Oxygen normally prefers to be in its stable diatomic state of two oxygen atoms, so when it is allowed to react, the third atom in an ozone molecule breaks away very quickly in the oxidation reaction. Ozone cannot be transported so it is manufactured on site with a fairly simple system consisting of an air dryer, ozone generator, venturi injector, side stream pump and a control system to measure the level of ozone in the system water. The system takes in ambient air to produce the ozone; this is done through a corona discharge in larger systems or UV (ultraviolet) process in very small systems. Biological control is accomplished in a process called cell lysing, taking only seconds, whereas chlorine must be ingested by the cell, which takes up to 16 minutes for that process to occur. Residual ozone levels greater than or equal to 0.4 ppm or mg/L have been effective in achieving a 100 percent kill rate in two to three minutes for Pseudomonas fluorescens a biofilm producer. It has also been found that residuals as low as 0.1 ppm or mg/L will remove 70-80 percent of biofilm in a 3-hour exposure. Ozone levels less than 0.1 ppm or mg/L will reduce populations of Legionella pneumophila in condenser water systems by 80 percent.89 There is no question regarding ozone’s excellent ability to kill all forms of biological matter including bacteria, biofilm, fungi, viruses and the other airborne organic matter that enter the condenser water system from the municipal water supply or via the surrounding air. Because ozone degrades to O2 or CO2 once the third oxygen atom has oxidized another molecule, there are no residual hazardous or toxic chemicals to deal with in the wastewater stream. Corrosion control can be tricky with the use of ozone. The phenomenal characteristics of ozone and its ability to oxidize can work against the system operator. If the cooling tower is subjected to too much ozone for too long a period of time, ozone will begin to oxidize the zinc in the galvanized metal of the cooling tower, copper in the condenser tubes of the chiller, and iron in the steel piping of the condenser water system. Generally, the operator controls the system to keep the pH above the 8.5 level where a natural pacification and ionic protection of the wetted surfaces takes place, thereby lowering the risk of oxidation to the metallic surfaces.

Scale prevention is the most difficult area for ozone to handle, especially in areas with high TDS, particularly calcium, but other mineral constituencies as well. The documentation for this process does refer to scale removal but only in cases where the scale is bound and interwoven with biofilm as the binder. Ozone does act as a mile coagulant which means that there will be

B-18

89 Federal Technology Alert, Ozone Treatment for Cooling Towers, December 1995.


some increase in the size of filterable particulate which is of some benefit. Ozone is not a good system choice when heavy levels of calcium and other scale forming constituencies are present. Manufacturers Many ozone manufacturers have come and gone over the years. However, several are active today. PuroTek has been a leader in the ozone industry for many years. They appear to have extensive experience in the treatment of condenser water systems with the use of ozone. However, there is no distribution network or local service network for this company. Ozonia North America and Clear Water Technology, Inc. are prominent manufacturers, both of which make a very wide range of ozone equipment used in multiple applications for multiple industries. Unfortunately, neither company supports the technology at the system level specific to commercial cooling applications. Literature addressing energy, chemical and water savings with the use of ozone was found on various websites. Hydrodynamic Cavitation (HDC) Water Treatment HDC technology has been around for approximately 15 years. It is one of the most innovative technologies employed today and is unlike all other technologies previously discussed. The technology essentially consists of two side-stream water loops typically connected into the sump of the cooling tower. One loop acts as a side stream filter. The second side stream loop is where condenser water passes through a pair of vortices and is accelerated to a very high velocity at the discharge. At the point of discharge, the two opposing water streams whose internal rotation is opposite from one another, collide, creating hydrodynamic cavitation, shear force and vacuum. This sudden lowering of pressure into a vacuum state forces the release of dissolved carbon dioxide (CO2) from the water. This release of CO2 in turn, causes calcium carbonate to immediately drop out of solution into suspension where it is removed by filtration. Unlike many of the other technologies previously discussed, HDC is a somewhat more “bulletproof” technology because is impervious to alkaline or acidic source water, high or low pH, hardness, TSS or TDS, all of which are the parameters that render many of the other technologies less desirable. Additionally, it is relatively easy for operational personnel to see if this system is working, because it can be either remotely monitored or monitored on-site by physical observation of the pump status and the vacuum in the vortex chamber. These characteristics also make this technology one of the better ones to ensure long-term savings of water.

B-19


Manufacturers

VRTX is the only known manufacturer of this technology. This manufacturer has extensive literature posted on their website and other papers were found relating to the technology and potential for savings in energy, water and chemical.90 This manufacturer takes the systems approach and does not leave the installation in the hands of the customer or a representative. They take an active roll in the sizing, installation and post-installation monitoring of the system performance, a definite plus in guaranteeing of water savings.

System Operation: Water is pumped into the pressure-equalizing chamber from the cooling tower sump. It is then channeled into precision-manufactured nozzles (vortices) that are configured in pairs to impart a specific rotation and velocity to the water streams. The circular motion of the water is accelerated as the stream from the first nozzle feeds into the second nozzle. The resultant discharge from the second stage is a conical stream. The opposing cones collide in the low-pressure stage (stabilizing chamber) to form a circular zone of very high shear force and high vacuum that is caused by the collapse of micrometer-sized bubbles and cavities. Essentially, the pressure change causes hydrodynamic cavitation with locally high temperature at the point of collision. This cavitation crushes solid particles, and the rapid change in pressure to a vacuum causes the cell walls of microorganisms to break, thus killing the cell. Finally, the hydrogen-bonding molecular arrays of water are broken down, thereby allowing entrapped gasses, such as CO2, to be released and off-gassed to atmosphere. The remaining energy dissipates as turbulent flow, and the treated water exits the unit at ambient pressure. Suspended matter is removed from the cooling tower sump via the second “side stream” loop that is designed to sweep the debris from the floor of the sump into the “zero gravity” filter or centrifugal separator. The filter is automatically backwashed to remove solid matter on a timed basis and is then sent to the sanitary waste system in the building. Biological Control: VRTX technology is in sharp contrast to chlorine or bromine treatments that are bio-inhibiting – chlorine concentrations degrade over time, requiring constant additions. As mentioned earlier in this paper, for the cell to die, the cell must ingest chlorine or bromine bio-inhibitors. This often takes up to 30 minutes if and when the cell comes in contact with a chlorine molecule, and is therefore not always 100 percent effective. On the other hand, VRTX technology causes a combination of physical changes to take place in the water that, together, disrupt the cell membranes of biological matter, ultimately destroying the cell. Every cell pumped through the system is subjected to vacuum, high pressure, kinetic energy, high velocity collision, shear energy, and high localized temperature. The pressure of the fluid inside the cell wall is in balance with ambient water pressure prior to its entrance into the VRTX. However, the pressure differential becomes relatively high once the cell enters the low-pressure stage that is in vacuum, resulting in a pressure imbalance between the inside and

B-20 90 Kitzman, Kevin A., Maziarz, Edward F., and Padgett, Bobby, Alcoa; Blumenschein, Charles D., U.S. Filter; and Smith, Alan, 2003. Chemical vs. Non-Chemical Cooling Water Treatments – a Side-by-Side Comparison, November.


outside of the cell. The cell wall cannot withstand the pressure differential and the cell wall ruptures, dispersing the cell cytoplasm. After the low-pressure stage, localized high temperature and high pressure at the intersection point of the vortices also kills additional bacteria and cell life. Scale and Hardness Control: Cooling systems build up scale over time due to the addition and concentration of soluble calcium often in the form of calcium bicarbonate. Calcium bicarbonate can decompose to yield insoluble calcium carbonate and carbonic acid with any changes in temperature and pressure. Carbonic acid can further decompose to carbon dioxide and water. Ca (HCO3)2 ⇔ CaCO3+CO2+H2O (Calcium bicarbonate) ⇔ (Calcium carbonate) + (carbon dioxide) + (water) At a given temperature and pressure, this equation is in equilibrium with no chemical reaction taking place. Once the pressure is lowered to a vacuum in the low pressure stage, the CO2 equilibrium is shifted between aqueous and gas phase, causing dissolved CO2 to release to the gas phase. This phenomena, together with the high localized temperature created by the collision of the conical water streams, decreases the solubility of calcium in water and a simultaneous elevation of water pH, which, in turn, causes a massive formation of calcium carbonate precipitate. Soluble calcium carbonate species concentrations are thus depleted (by design) both via desorption of CO2 and the precipitation of CaCO3. As the water stream leaves the Vortex unit it enters the sump of the cooling tower where the water pressure is stabilized (at atmospheric) and the velocity of the water slows down. Sub-micron particles of calcium carbonate called colloids are formed and flow with the water. These become thermodynamically favored to grow crystals composed of Ca2+ and HCO3

- ions versus metal surfaces in the system. As the molecules coagulate, they become heavy and sink to the sump floor. At this point, the resultant calcium scale is removed via the side stream filter or centrifugal separator and collection system. This device is periodically backwashed to remove entrapped calcium and other suspended matter. Since the bleed or blow down has been eliminated from the cooling tower treatment requirement, the only water to leave the system other than evaporated water or drift, is the very small quantity of water used to backwash the filter or centrifugal separator. Corrosion Protection: Corrosion occurs in a system due to several phenomena mentioned earlier in this report. All waters are corrosive to some degree; however, the level of corrosive tendency will depend upon its physical and chemical characteristics. The materials that a given water supply will negatively affect may differ. Water that is corrosive to galvanized pipe may not be corrosive to copper. Corrosion inhibitors that protect one material may have no effect or may even be detrimental to other materials. Biological growth in a piping system can also cause corrosion by providing an environment in which physical and chemical interactions can occur. Several types of system level problems can occur if the condenser water systems are left untreated.

B-21


A major source of corrosion is the addition of the bio-inhibiting chemicals themselves. This is true especially when “shocking” is required as chemicals lose there effectiveness over time and a chemical alternative is administered. By the time this new chemical, usually bromine- or chlorine-based, is put into service, the biological growth in the system has gotten out of control such that a “super concentration” of biocide is required. These chemicals tend to be very corrosive.

As mentioned earlier in this paper, a layer of biofouling on any surface in the condenser water system acts as a haven for aerobic and anaerobic activity, bacteria formation, scale accumulation and potential corrosion. By eliminating biofouling, the potential for corrosive activity is greatly diminished.

The pH level is elevated to a level above 9.0. At values higher than 9.0, both iron and copper are protected from oxidation corrosion which cannot operate in an alkaline state. Under these conditions, metals are allowed to establish a thin layer of natural protection, which is not penetrated by system water or dissolved oxygen.

B-22

Paper presented at the EPRI/PIER Advanced Cooling Strategies/Technologies Conference on June 2, 2005 in Sacramento CA

Independent Assessment of the Energy Savings, Environmental Improvements and Water Conservation of Emerging Non-Chemical Water Treatment Technologies

By: Michael Gravely, Technical Consultant, California Energy Commission PIER Program Bruce La Belle, Ph.D. California Environmental Protection Agency Dr. John Balachandra, California State University at Sacramento Abstract: This paper will discuss the results of a California Energy Commission Public Interest Energy Research (PIER) funded project to complete an independent assessment of the energy savings, environmental improvements and water conservation capabilities of the new and emerging non-chemical water treatment technologies. The project was completed by a team from California State University at Sacramento and included a technical review of several of the emerging technologies and a detailed assessment of two of the emerging non-chemical water treatment technologies. Clearwater Systems Corp. and VRTX Technologies, LLC produced the two technologies assessed. The research was focused on gathering information from industrial field customers who had purchased and installed these systems and had actual experience with their operational characteristics from several months to several years. The team completed a telephone survey with approximately 15 end user customers and made site visits to ten sites. Some limited independent water testing was also completed. The results of these phone surveys and site visits were consolidated and placed in an interim report. Even though only a small number of end user customers were actually surveyed or visited, the research indicated that several hundred systems have been successfully installed in California and throughout the United States. Both these emerging technologies provide non-chemical treatment for cooling tower and evaporative condenser system water. All the information collected and results derived from this effort will be made available to the public later this year in the form of a PIER Technical Report. A Project Advisory Committee that included representatives from CalEPA, the Energy Commission PIER Program and local utilities supported this team. Disclaimer: This technical paper is a result of work sponsored by the California Energy Commission and does not necessarily represent the views of the Energy Commission, its employees or the State of California. This technical paper has not been approved or disapproved by the California Energy Commission nor has the Energy Commission passed upon the accuracy or adequacy of the information in this technical paper. Introduction: The California Energy Commission's Public Interest Energy Research (PIER) Program Industrial/Agricultural/Water (IAW) Team recently became aware of non-chemical water treatment technologies that provide potential energy efficiency improvements in industrial and commercial cooling tower and evaporative condenser applications. In addition to energy efficiency improvements, these technologies when applied provided “chemical free” water treatment of these systems. This “chemical free” environment eliminates the requirement to add treatment chemicals into the system water to control scaling, corrosion and biological agents that routinely form in these applications (these additive chemicals are considered toxic and environmentally undesirable). These technologies also report to provide increased water conservation and improve the quality of the water released into the ground water recovery systems at these industrial and agricultural applications. The PIER IAW Team formed a Project Advisory Committee for this effort which included representatives from several PIER program areas (IAW, Buildings End-Use Energy Efficiency, and Energy-Related

Page 1 of 15


Environmental Research), California Environmental Protection Agency and several utilities from Northern California.). The primary purpose of this work effort was to perform an independent assessment of the capabilities and performance of emerging non-chemical water treatment technologies. This PAC was used to help review the progress of the effort and provide guidance on the completion of the research. Background: Makeup waters for cooling towers and evaporative condensers contain a wide range of impurities: dissolved solids, dissolved gasses, organic compounds, suspended solids and microorganisms. These impurities cause problems in cooling water systems as they increase in concentration as the water evaporates in the cooling process. The most common problems are scale, corrosion, and bacterial growth. To control these problems, industry has historically utilized traditional methods of chemical treatment. A proper chemical treatment program needs to be adjusted, monitored and controlled. In practice this is difficult to accomplish as chemical treatment programs often consists of a series of compromises; and, a properly managed program requires constant adjustments due to changes in the system water as well as changes in make-up water. Typically, scale formation is inhibited by the use of acids and threshold inhibitors, including polymers, polyphosphates, phosphonates or a combination of these chemicals, along with managed cycles of concentration. Anodic, cathodic, or combinational corrosion inhibitors are added to slow down corrosion by forming a protective layer on metal surfaces. Zinc, copper, molybdates and other heavy metals ore often added in formulations to increase the effectiveness of the chemical additives. Corrosive oxidizing and toxic non-oxidizing biocides are added to control the growth of microorganisms. The use of some treatment chemicals can create additional problems and interfere with each other’s performance. Acids are frequently used to control scale. Oxidizing biocides are used to control microorganisms. Low pH conditions enhance the effectiveness of biocides. Acids and biocides attack metal because of their oxidizing nature. To neutralize the corrosiveness of water, and to a degree to offset chemicals added for scale and bacterial control, corrosion inhibitors are added. Some corrosion and scale inhibitors may react with divalent cations in the system water to form pseudo scale. Phosphate based chemicals can be used by bacteria as a food source. Oxidizing biocides can breakdown certain types of scale/corrosion inhibitors; and, corrosion inhibitors require elevated pH levels in order to inhibit corrosion; yet, higher operating pH conditions adversely impact oxidizing biocides and their half-life. Additionally, a growing number of water treatment chemicals are coming under even stricter laws relating to their use and handling. Added to this reality is the fact that the chemical costs, utility costs and training costs for the personnel who must handle the chemicals, store and dispose of the chemicals, are steadily increasing along with associated maintenance costs. And finally, many industrial companies have corporate mandates to lower water usage and reduce chemical usage corporate wide. These issues and the operational problems associated with traditional chemical treatment have collectively caused many cooling tower and condenser users to re-evaluate alternative, non-chemical methods. During this effort, several emerging technologies that reported to provide non-chemical water treatment were reviewed. A criterion was developed to determine which technologies were to be surveyed and researched in more detail. The following criteria were used to select the emerging technologies to be research:

1. The emerging technology reported the ability to provide non-chemical water treatment for the make up water in cooling towers and evaporative condensers. Techniques that recommended the use of chemical treatment for scale, corrosion, or bacterial growth were not considered.

2. The emerging technology representative was able to provide a list of end use customers who had previously purchased and installed their technology at the end user facility. No PIER funding was used to purchase or test this technology in this effort.

3. The end use customers were willing to discuss the performance of their non-chemical treatment system with the research team and end user was willing to host a tour of their facility if requested

Page 2 of 15


Based on these criteria, Clearwater Systems Corp. and VRTX Technologies, LLC were the two technologies assessed in this effort. Technology Overview: 1. Description of Dolphin System™ Technology by Clearwater Systems Corp. The Dolphin System™ is PULSED POWERED. It consists of two primary components:

a) A high-frequency pulse generator (controller) b) A Coil-Pipe Assembly (Sized for full system flow)

The controller uses proprietary electronic circuitry to induce a high frequency, time varying electromagnetic field into the circulating water via the coil-pipe assembly. The coil-pipe assembly is essentially a section of PVC pipe outside of which a set of custom wound solenoid coils is fastened. Stainless steel pipe is used for boiler applications. There is no direct contact between the coils and the water. The coil-pipe assembly is applied to the full flow of the recirculating process water, and as result, water molecules are completely exposed to the pulsating electromagnetic field. It is preferably located between the discharge side of the condenser water circulation pump and the chiller. It may also be installed between the chiller and the cooling tower. The Dolphin Pulsed Power System™, through the pulsed electromagnetic field, imparts a circumferential electric field in the bulk solution in opposition to the magnetic field, creating a complex electric field in the bulk solution. This affects the way minerals precipitate out of solution by altering the electrical charge of the colloidal particles so that they form “Colloidal nucleating powder” or crystals (at a microscopic level). The technique causes minerals to “CLUMP TOGETHER” (Nucleate) rather than depositing onto the equipment surfaces. The mineral crystals are subsequently carried off in the discharge blow-down. There is no velocity essential to this process, and the system is operable in all types of water chemistry without special construction or signal tuning. A full description of “Operating Principles: Method of Action” of Dolphin system is attached herewith with report. Facts and figures of the Clearwater Systems Dolphin:

Removes existing scale (depending on scale properties) Controls microbe (bacteria population) by limiting their ability to reproduce. Controls corrosion indirectly. Significant Water Savings compared to chemically treated tower. Financial savings Reduced maintenance More environmentally friendly It can treat the process waters of cooling towers, chiller systems, heat exchangers, direct

evaporative air-coolers (“swamp coolers”), steam boilers, some hot water systems and fountains. a) Scale Control: The Dolphin System™ allows Calcium Carbonate and other dissolved minerals in the water to precipitate on suspended colloidal particles in bulk water rather than on the heat transfer surfaces. These colloids are available in the system bulk solution primarily from scrubbed in solids and secondarily those entering through make-up water. The unique spectrum and polarity of electromagnetic fields produced by pulsed power shifts the CaCO3 equilibrium chemistry to favor formation of stable crystal nuclei in the bulk solution. Thus crystal growth and precipitation will occur in solution and accumulate as a loose powder instead of on a surface as a scale. (Details of Scale Control including picture is attached herewith in “Operating Principles: Method of Action- serial B”).

Page 3 of 15


b) Bacteria Control: With the Dolphin system, although the bacteria are not immediately killed, they are controlled. The Pulsing electromagnetic magnetic field from the Dolphin system generates a very low frequency, non-ionizing electromagnetic radiation. This pulse has the same shape and frequency as those generated by a pulsed-electric-field device developed by Maxwell Technologies and approved by the FDA for food Pasteurization. A Dolphin device uses a fraction of the energy that the Maxwell devices use on a per pulse basis. However, because cooling towers involve recirculation water, a Dolphin device on a cooling tower will “see” each bacterium many times before the bacterium exits the system. In a typical cooling tower setup, each bacterium will see over 5000 pulses, exposing the bacteria to over 50% on the total amount of radiation as with the Maxwell device. This exposure over a few hours does not have the same biological effect as a single dose obtained with the Maxwell device; however, there is recurring sub-lethal damage to the microorganisms. This damage is sufficient to inhibit reproduction but not sterilize the system. Some bacteria may recover in few days, but while they are being recirculated through the Dolphin they are inactive. (Details of Bacteria Control including picture is attached herewith in “Operating Principles: Method of Action- serial C”). A second method of control is through the bulk-solution crystal growth described in the section on scale control. The formation of the crystals in the bulk solution will agglomerate bacteria and help inhibit their growth. The bacterial agglomerated in the process are removed from nutrient sources and have no path for waste product to exit. The result is similar to the flocculation process commonly applied for potable water supplies to clarify suspended solids from the raw water and in the process remove bacteria as a byproduct of the clarification. Essentially, the Dolphin System™ could be compared to this process while using a physical means of flocculation rather than a chemical means. Field testing results offered by Clearwater Systems Corp has shown Heterotrophic Plate Counts (HPC by SMEWW pour plate method 9215B) of consistent planktonic control below 5,000 CFU/ml, with systems frequently operating below 1,000 CFU/ml. The Dolphin System™ has proven efficacy over biofilm, or slime layers, that exist in most cooling water systems. The benefit of biofilm eradication is the elimination of microbial influenced corrosion (MIC) which causes systems to develop metal failure in chiller tubes and piping systems. Such extensive biological control eliminates the addition of oxidizing biocides - further reducing the corrosion of systems. c) Corrosion Control: Corrosion inhibition with the Dolphin System is accomplished indirectly by maintaining sufficient cycles of concentration to force the system into and above the saturation point of calcium carbonate. Above the point of saturation, calcium carbonate provides excellent cathodic corrosion protection. Additionally, operation of the bulk solution in higher alkalinity and in pH ranging from pH 7.5 to pH 9 provides a rather benign water chemistry which is naturally less corrosive to metals used in cooling system construction. (Details of Scale Corrosion is attached herewith in “Operating Principles: Method of Action- serial D”).

Page 4 of 15


2. Description of operation of the VRTX technology As illustrated below, the system is a side-stream treatment application. It includes two parts: a mechanical unit and a separation/filtration system. The separation/filtration system unit is used to remove the precipitated calcium carbonate and other suspended solids from the circulating cooling water.

VRTX

To drain

Return from process

To proces

s The unit works primarily on the principals of Controlled Hydrodynamic Cavitation (CHC). Cavitation is the dynamic process of the formation, growth, and collapse of micro-sized bubbles in a fluid. When pressure falls below a critical value, cavities are formed in the liquid. When pressure increases, the cavity cannot sustain the surrounding pressure. Consequently, they collapse catastrophically. Studies have shown that when a liquid moves fast enough, gas bubbles will form and collapse creating a process called cavitation. Studies have indicated that in turbulent liquid flows, and notably at high velocity, hydrodynamic cavitation will occur. The patented chamber, (depicted below), contained within the unit, consists of a pressure equalizing chamber and a cavitation chamber. Water is pumped from the sump into the pressure-equalizing chamber at a constant pre-determined pressure. The water is then channeled through opposing nozzles that impart a specific rotation and velocity to the water streams. The flow and rotation of the water streams creates a high vacuum, usually 27.5” Hg to 29.0” Hg.

Water out

Water in The water streams in the opposing nozzles rotate in opposite directions as they move forward at increasing velocity. Upon exiting from the nozzles, the water streams collide at the mid-point of the cavitation chamber. When the streams collide a region of high pressure is momentarily established causing vapor bubbles to collapse (implode) and the water chemistry to shift and chemical reactions to proceed. At the moment of bubble collapse an intensive shock wave and extremely high temperatures are generated. When this occurs, CO2 and other dissolved gasses are released. These combined effects are used by the technology to force the precipitation of calcium carbonate, control corrosion and eradicate microorganisms.

Page 5 of 15


a) Scale Control:

By facilitating the precipitation of calcium carbonate, at lower temperatures, the mechanical unit continuously removes calcium carbonate, from the re-circulating cooling water. When water passes through the chamber, calcium carbonate is forced to precipitate out. The following equation describes the reaction that occurs within the chamber.

Ca(HCO3)2 CaCO3 + CO2 + H2O In the chamber the chemical equilibrium of the carbonate species is shifted, driving the above reaction to the right. As long as CO2 is removed the equation tends to stay to the right. As a result, the soluble calcium bicarbonate converts into insoluble calcium carbonate (from solution to suspension) and, carbon dioxide gas is removed. The calcium carbonate colloids steadily grow and are easily removed from the water stream through the use of a filtration system.

Like any precipitation process, the formation of CaCO3 in water is limited by the nucleation step. In order for the dissolved ions to react and form stable nuclei, extra energy is required to overcome the surface tension. This extra energy can be significantly reduced if nucleation takes place on existing surfaces such as equipment surfaces in cooling water systems or preferably on newly formed, pure colloidal CaCO3 crystals. With the effect of hydrodynamic cavitation, dissolved calcium and carbonate ions are forced to form within the chamber. The acceleration of nucleation via cavitation has been well documented and the small-sized CaCO3 colloids act as growth sites for other dissolved ions. Continued crystal growth is thermodynamically favored over the formation of new nuclei (less required energy). Normally, when cooling water travels to various hot spots, more dissolved calcium and bicarbonate tend to precipitate out because of the decrease in solubility causing scale build-up. Conversely, the mechanically treated water behaves differently. Instead of forming nuclei on the equipment surface, dissolved ions will redirect their growth on the newly formed colloids. The Calcium Carbonate crystals continue to grow. As they grow in size, coagulation increases due to greater mass attraction. Crystal growth accelerates and the larger particles precipitate out and are removed by the filtration system.

b) Bacteria Control

The effectiveness of the unit in eradicating bacteria has been well documented. Several years of data has repeatedly shown nearly total eradication of E-Coli. Current testing has shown nearly identical results with various heterotropic bacteria. Field tests, independent laboratories and third party evaluations have confirmed these laboratory findings. Figures 3 and 4 summarized typical laboratory test results. High vacuum, high pressure, high temperature, collision and mechanical sheer are believed to contribute to the killing of various microorganisms. Microscopic examination reveals that VRTX technology kills bacteria by physically rupturing the cell membrane. The exact mechanism is still under investigation; however, one or the combination of the following two actions may account for the major factors in eradicating bacteria:

• Dramatic changes in pressure and vacuum – When water passes through the chamber, it experiences dramatic changes in pressure/vacuum/pressure. The membrane wall of bacteria is permeable and fragile. Under such dramatic pressure changes over such a short period (seconds), it is believed that the cell membrane is ruptured. The direct impact of shear and collision forces created by the collision of water streams would also contribute to cell wall destruction. Once the

Page 6 of 15


membrane is broken, vital liquid components inside the cell leak out, thereby causing the death of the bacteria.

• Hydrodynamic cavitation – When cavity bubbles collapse, an extremely high, localized temperature and pressure wave is momentarily created. The cell membrane of bacteria can easily be ruptured under these extreme conditions. Additionally, the cavitation process momentarily forms highly reactive oxidizing radicals and hydrogen peroxide, which are very effective in killing bacteria.

Under the action of cavitation, water molecules decompose into hydroxyl free radicals and hydrogen atoms:

H2O OH- + H

The dissociation of water is thought to be produced by electrical discharges resulting from hydrodynamic cavitation or by thermal dissociation due to adiabatic compression of the collapsing bubbles (12). Hydroxyl radicals can combine to form hydrogen peroxide:

OH- + OH- H2O2

Hydrogen peroxide is also formed by the reaction between hydrogen atoms and dissolved oxygen in water:

H + O2 H2O2 H2O H2O2 + O2

It is believed that both actions contribute to the eradication of microorganisms.

c) Corrosion Control:

Corrosion, triggered by the pH decrease of the process water, is a common problem for cooling tower maintenance and equipment life. Because the circulating water in a cooling water system contains oxygen, iron (or other metals) is oxidized when it comes in contact with the circulating water. The following equation illustrates this anodic reaction:

Fe(s) Fe2+

(aq) + 2e-

These electrons are then consumed by the hydrogen ion (H+) in one of the following cathodic reactions:

2H+(aq) + 2e- H2(g)

4H+

(aq) + O2 (aq) + 4e- 2H2O(aq) Because the concentration of hydrogen ions is exponentially proportional to pH, a small decrease in pH will substantially increase the hydrogen concentration. This makes pH an important factor in controlling corrosion.

Other important factors that affect the rate of corrosion are the chemical composition of the water, the concentration of dissolved gases, flow rates, temperature, microorganisms, and type of metal exposed to the circulating water. Dissolved gases such as oxygen, carbon dioxide, sulfur dioxide, and sulfur trioxide build up in cooling circulating water as a result of continuous aeration. These gases which are absorbed into the cooling system water from the air (much like a scrubber) also contribute to the oxidation of metal as shown in the above reactions.

Page 7 of 15


The technology minimizes corrosion reactions by reducing the corrosiveness of cooling circulating water. Eliminating corrosive chemicals, maintaining relatively high pH levels, and keeping the growth of microorganisms under control achieve this goal. Consequently, the corrosion reaction rates are reduced even at relatively high TDS levels. Research Project Activities The research team for this effort was selected from California State University at Sacramento. The research team completed the following activities in the completion of the project:

1. Assess several non-chemical treatment emerging techniques to determine if they meet the criteria for this project.

2. Obtained a list of at least twenty end use customers from the two emerging technologies selected for further research.

3. Complete telephone surveys for a selected number of end user customers who had installed the technology.

4. Complete on site visits to several of the end users sites to gain better insight into the operation of the assessed technologies

5. Summarize the result of the entire effort in a final report for the PIER IAW Program. Project Survey Results The following is a summary report on the information gained from the telephone and site surveys. The same questions were presented to end users and a consolidated summary of their responses is listed below by technology: The Dolphin System™ Technology by Clearwater Systems Corp survey results: 1) Type of Applications:

a) Cooling Tower - 8 b) Evaporative Condenser - 2 c) Boiler - 0 d) Others - 0

2) Number of systems installed at your site or location:

a) One - 1 b) Two - 0 c) Three - 1 d) Four - 3 e) More than Four - 2

3) How long have you been using (Years) :

a) Less than a year - 4 b) (1-2) yrs - 5 c) (3- 4)Yrs - 1 d) (4- 5)Yrs - 0 e) (5-6) Yrs - 0

4) Is the system still in use:

a) Yes - 10 b) No - 0

Page 8 of 15


5) If not, what were the reasons for removing the system (s)? N/A a) Performance b) Financial c) Maintenance d) Others

7) Are you satisfied with the overall performance of the system (s)?

a) Yes - 10 b) No - 0 c) Any Comments - It is removing scale, lowering head pressure and most

of all it is maintenance free.

8) Do you have any future plans to install additional systems at your current or any other company facilities?

a) Yes - 8 b) No - 1 (Currently they don’t have anymore facilities

to install) c) Don’t know - 1

10) Did you use chemical water treatment in your system before?

a) Yes - 10 b) No - 0

10) How long were you using chemical treatment prior to installation of system (s)?

a) Less than 10 Yrs - 1 b) (11-20) Yrs - 3 c) (21-25) Yrs - 3 d) (26-30) Yrs - 1 e) > 30 Yrs - 2

11) Are you using chemical water treatment in other plant applications where this technology could be

used as well: a) Yes - 8 b) No. - 2

12) How many sites or other company locations are you using chemical water treatment in other plant

applications where this technology could be used as well? How many sites are currently using this technology?

a) 65 - 1 b) 63 - 1 c) 10 - 1 d) 6 - 1 e) 2 - 1 f) 1 - 1

13) How did you learn about the _________________ system?

a) Sales call by vendor. - 6 b) Referred from others - 2 d) Trade show - 2

14) What is the main reason you decided to purchase the system?

a) Economic - 1 b) Water saving and safety - 1 c) Environmental friendly and Economical - 4 d) Environmental friendly and remove scaling - 3 e) Reliability - 1

Page 9 of 15


15) How long did it take you to decide to install the system?

a) No time - 2 b) (0 - 3) months - 2 c) (3 - 6) months - 5 d) > 6 months - 1

15) Does the system require routine maintenance, if so, how often?

a) Routine Maintenance not required - 8 b) Every week - 2

16) What type of routine maintenance does the system need?

a) Clean strainers - 1 b) Connection and visual check - 1

17) Do you regularly perform any type of water sample test on your system?

a) Yes - 9 b) No - 1

18) What water quality tests do you perform?

a) PH - 5 b) Conductivity - 8 c) Alkalinity - 3 d) Hardness - 3 e) Silica - 2 f) Corrosion Test - 1 g) Bacteria test - 2

19) How often do you perform this test (s)?

a) Every week - 3 b) Every two week - 3 c) Every month - 1 d) Quarterly - 2

20) How long do you keep the record of the test results?

a) Yes, forever - 8 b) One year - 1 c) No response - 1

21) Are you satisfied with the test results you have received to date?

a) Yes - 9 b) No. - 0 c) No response - 1

22) Have you observed any changes in scaling or befouling?

a) Yes - 8 b) No - 2

23) If you did observe any changes in scaling or biofouling, could you briefly

describe what you saw. a) Scale Control - 7 b) Less Biofouling - 1 c) Less Algae - 3 d) Bacteria Control - 1

Page 10 of 15


24) Did you complete any economic analysis of the system (s) when you made the decision to install the system? If so, what was the estimated pay back time or other expected value provided by the new system?

a) No - 3 b) (0- 1) Yrs - 0 b) (1-2) Yrs - 3 c) (2-3) Yrs - 2 d) > 3 Yrs - 2

25) Can you estimate the savings provided by the system(s) provided by the following:

a) Reduction of Chemical used - 9 b) Water savings - 5 c) Energy saved - 5 d) less maintenance - 9

26) Can you quantify the total annual savings (in dollars) expected from the complete system?

a) $(0-10) K - 7 b) $(20-50)k - 1 ($ 48k) c) No - 2

27) Have you recommended the system to others?

a) Yes - 9 b) No - 1

28) Would you recommend the system to others?

a) Yes - 10 b) No - 0

29) Do you have any additional comments or suggestions for our evaluation team?

a) Satisfied with performance - 8 b) Still observing, so far positive - 1 c) No Comments - 1

VRTX Technology Survey Results: 1) Type of Applications:

a) Cooling Tower - 5 b) Evaporative Condenser - 3 c) Boiler - 0 d) Others - 0

2) Number of systems installed:

a) One - 1 b) Two - 1 c) Three - 4 d) Four - 1

3) How long have you been using (Years):

a) (1- 2) Yrs - 3 b) (3- 4) Yrs - 2 c) (4- 5) Yrs - 2 d) (5-6) Yrs - 1

Page 11 of 15


4) Is the system still in use:

a) Yes - 7 b) No - 0

5) If not, what were the reasons for removing the system (s)? N/A 6) Are you satisfied with the overall performance of the system (s)?

a) Yes - 7 b) No - 0

7) Do you have any future plans to install additional systems at your current or any other company facilities?

a) Yes - 3 b) No - 4

8) Did you use chemical water treatment in your system before?

a) Yes - 6 b) No - 1

9) How long were you using chemical treatment prior to installation of system (s)?

a) (20-25) Yrs - 1 b) (25-30) Yrs - 2 c) (30-35) Yrs - 1 d) > 35 Yrs - 2

10) Are you using chemical water treatment in other plant applications where this technology

could be used as well: a) Yes - 3 b) No. - 3

11) How many sites or other company locations are you using chemical water treatment in other plant

applications where this technology could be used as well? How many sites are currently using this technology?

a) 4 - 1 b) 2 - 1 c) Zero - 3

12) How did you learn about the system?

a) Sales call by vendor. - 5 b) Referred from others - 1 c) Trade show - 1

13) What is the main reason you decided to purchase the system?

a) Economic - 2 b) Water saving and safety - 2 c) Environmental friendly and Economical - 2 d) Easy to use - 1

14) How long did it take you to decide to install the system?

a) No time - 1 b) (0 - 3) months - 2 c) (3 - 6) months - 3 d) > 6 months - 1

Page 12 of 15


15) Does the system require routine maintenance, if so, how often? a) Every day - 1 b) Every week - 4 c) Every month - 1 d) Every three month - 1

16) What type of routine maintenance does the system need?

a) Clean strainers - 5 b) Power on - 1 c) Change bag filters - 1 d) Inspection - 1

17) Do you regularly perform any type of water sample test on your system?

a) Yes - 7 b) No - 0

18) What water quality tests do you perform?

a) PH ,Bacteria test and Conductivity - 1 b) Standard water analysis - 4 c) Standard analysis and PH - 1 d) Water analysis and Bacteria test - 1

19) How often do you perform this test (s)?

a) Every week - 2 b) Every month - 5

20) How long do you keep the record of the test results? Forever. 21) Are you satisfied with the test results you have received to date?

a) Yes - 7 b) No. - 0

22) Have you observed any changes in scaling or befouling?

a) Yes - 7 b) No - 0

23) If you did observe any changes in scaling or biofouling, could you briefly

describe what you saw. a) Scale, bacteria and corrosion control - 4 b) Corrosion control and bacteria control - 1 c) Less fouling - 2

24) Did you complete any economic analysis of the system (s) when you made the decision to install

the system? If so, what was the estimated pay back time or other expected value provided by the new system?

a) (0-1) Yrs - 1 b) (1-2) Yrs - 2 c) (2-3) Yrs - 2 d) > 3 Yrs - 2

25) Can you estimate the savings provided by the system (s) provided by the following:

a) Reduction of Chemical used - 7 b) Water savings - 7 c) Energy saved - 7 d) Less maintenance - 7

Page 13 of 15


26) Can you quantify the total annual savings (in dollars) expected from the complete system? a) $(0-50) K - 1 b) $(50-100) K - 1 c) $(100-150) K - 1 d) > $150k - 2

27) Have you recommended the system to others?

a) Yes - 7 b) No - 0

28) Would you recommend the system to others?

a) Yes - 7 b) No - 0

29) Do you have any additional comments or suggestions for our evaluation team?

a) Satisfied with performance - 5 b) No comments - 2

Key Take Ways from the Survey Efforts:

1. The project's qualitative survey found that respondents have had positive experiences with the alternative water treating technologies, indicating that growing market acceptance appears likely

2. As with any emerging technology, there are still many obstacles to obtaining complete commercial

acceptance • Broader market acceptance • Increased number of fielded systems • Additional examples of clear economic paybacks • More published case studies by independent groups such as utilities, government agencies

and others

3. Both technologies assessed are building a portfolio of case studies, independent reports and technical papers to aid in the market acceptance transition.

4. Both technologies are substantially increasing their installed base of operating systems

Next Steps 1. PIER final report will be published in next few months. The report will be available for public download on the Energy Commission PIER Web Site. The PIER Final Report will include more details on the energy savings, water savings and environmental benefits determine in this effort. This information will be reviewed by the Project Advisory Committee prior to being approved for publishing. 2. The Energy Commission and Cal EPA Staffs will complete a detailed assessment of the project results determine the recommended next steps.

a) Additional research and analysis. b) Assessing the potential for consideration the technologies for energy savings, water savings and

other government incentive programs.

Page 14 of 15


Acknowledgement This paper provided information gained through a research effort funded by the California Energy Commission's Public Interest Energy Research (PIER) Program. The PIER Program supports energy research, development and demonstration (RD&D) projects that will help improve the quality of life in California by bringing environmentally safe, affordable and reliable energy services and products to the marketplace. The PIER Program annually awards up to $62 million to conduct the most promising public interest energy research by partnering with RD&D organizations including individuals, businesses, utilities, and public or private research institutions. PIER brings new energy services and products to the marketplace and creates state-wide environmental and economic benefits. PIER funding efforts are focused on the following RD&D program areas: Buildings End-Use Energy Efficiency Energy Innovations Small Grant Program Energy-Related Environmental Research Energy Systems Integration Environmentally Preferred Advanced Generation Industrial/Agricultural/Water End-Use Energy Efficiency Renewable Energy Technologies Additional information on the PIER program can be obtained by visiting the California Energy Commission web site at http://www.energy.ca.gov/pier/index.html. Disclaimer: This technical paper is a result of work sponsored by the California Energy Commission and does not necessarily represent the views of the Energy Commission, its employees or the State of California. This technical paper has not been approved or disapproved by the California Energy Commission nor has the Energy Commission passed upon the accuracy or adequacy of the information in this technical paper.

Page 15 of 15

Industrial Water Conference 2003, Las Vegas, NV, December 9 – 11, 2003

New Technology Saves Cooling Water Demands

Peter Koestler The Gambrinus Company

San Antonio, TX

Wiley Wang and Robert Kelsey VRTX Technologies, LLC

San Antonio, TX Abstract A non-chemical cooling water treatment technology was evaluated versus a conventional chemical treatment program in a cooling tower application at Spoetzl Brewery. Water consumption data was collected over a three-year period. A water consumption baseline was established while on chemical treatment. This baseline was compared with water consumption on a non-chemical treatment system over a two-year period. Water consumption data is based upon metered readings of both makeup and blowdown. Data was monitored and collected on a daily basis and then a weekly basis. The study an annual water saving of approximately 2 million gallons with the non-chemical treatment program. Introduction The beverage industry has been challenged to find ways to reduce water use. The brewing industry has an especially vital interest in the preservation and quality of water. Not only does its product consist of over 90% water but during the brewing process an additional 4-7 barrels (each barrel equals 31 US gallons) of water are used for every barrel of finished beer. The Spoetzl Brewery, Inc. in Shiner, Texas, serves as an example of how this concern can be successfully addressed. The brewery, established in 1909, is Texas’ oldest independent brewery. It produces the Shiner brand family of beers enjoyed by many in the Southern half of the United States, with Shiner Bock its most famous brew. The brewery operates a refrigeration system with anhydrous ammonia as the primary coolant. The recompressed ammonia is being cooled down in three evaporative condensers with a total cooling capacity of ~ 1,500 tons and a water volume of ~ 3,000 gallons. Municipal water is used as makeup for this cooling system. On average, the makeup water contains 130 ppm (as CaCO3) of total hardness, 100 ppm (as CaCO3) of calcium hardness and 400 ppm (as CaCO3) of total alkalinity. The pH is around 7.7 with a TDS of approximate 700 ppm. The original chemical water treatment system was designed with a makeup water holding tank into which additives were automatically injected, based on water conductivity. From this holding tank, the cooling water was pumped up to the towers which are located on the roof.

Chemical Treatment During chemical treatment, three different types of chemicals were applied in an effort to maintain the proper operation of the system. Sulfuric acid was added to prevent calcium carbonate buildup; Chlorine and a non-oxidizing biocide were used to control microorganism growth; and a corrosion inhibitor, based on molybdenum, was applied to reduce corrosion rates. The control range for pH was 8.0 – 8.2; the residual molybdenum and free chlorine were set at 1.5 to 2.5 ppm and 0.3 to 0.5 ppm, respectively. In conjunction with the chemical treatment, cycles of concentration were kept low with repeated bleed-off to assure water quality. The cycles of concentration were 2 – 2.5 during the summer months and below 2 for the rest of the year, resulting in an average of around 2. Despite these efforts, significant calcium carbonate deposit accumulated around the condenser tubes and inside the condensers. Considering that the thermal conductivity of calcium carbonate is many times less than that of carbon steel, the formation of inorganic scale on condenser tubes significantly reduces the heat transfer rate across condenser tubes; and, consequently, excess electric energy is required to run the ammonia compressors. Under chemical treatment bacteria counts averaged in excess of 10,000 cfu/ml. Non-Chemical Treatment In August 2001, the traditional chemical treatment program was replaced with a VRTX system, a patented non-chemical technology for cooling water treatment. The VRTX system consists of a VRTX unit and a filtration/separation unit. The VRTX unit is a mechanical device using the principle of kinetic energy, chemical equilibrium and hydrodynamic cavitation to treat fluids. When the cooling water is pumped through the VRTX chamber, dramatic changes in velocity and, consequently, static pressure lead to the destruction of microbial cell walls and the conversion of dissolved calcium and bicarbonate ions into calcium carbonate (CaCO3) colloids:

++ Carbon dioxide gas Water+ CaCO3

colloidsDissolved Ca

ions Dissolved

bicarbonate ions Energy

Ca2+ + 2HCO3-

CaCO3 + CO2 + H2O Energy

The CaCO3 colloids formed under this condition are in the form of aragonite, shown by microscopic analysis (Photo 1). Aragonite has the same chemical composition as calcite, but their molecular structures are quite different. The rhombohedra calcite crystal has six flat surfaces; these flat surfaces lead to significant adhesion sites and readily attach to other surfaces and form a hard deposit. Conversely, the shape of aragonite is needle-like. This unique shape substantially reduces the adhesion sites to other surfaces, and tends to form soft deposits. Deposits and formation of aragonite CaCO3 are more stable upon heating and can be carried throughout a heating or cooling system while causing no apparent damage and no appreciable build-up. This transport property allows the CaCO3 particles to be removed by the filtration unit, where they are physically separated from the recirculating water.


Photo 1. Aragonite Crystals Formed in VRTX Treated Water

Based on the water volume in the system, makeup feed rates and water chemistry, a 60 GPM VRTX unit and a 50 GPM disc filter were installed. Both the VRTX unit and filter draw and return water to three separate sumps. Globe valves were installed on the intake lines to adjust flow rates for equal treatment of all three sumps.

The disc filter combines the advantages of centrifugal separator and media filter. The spinning incoming water creates a centrifugal cleaning action, spiraling heavier particulate away from the disc stack. This dramatically reduces the backflush frequency. The media is comprised of injection molded polypropylene disks, which form a three-dimensional filter media. As cooling water passes through the depth of the disks, the disks capture the suspended particulates. As these particulates accumulate, it causes a pressure differential across the filter. When the differential pressure reaches a predetermined level, an automatic back-flushing cycle is initiated. The disk caps lift hydraulically to allow the disc stack to open and separate. Backflush nozzles spray filtered water uniformly through the disc stack. When the backflush cycle is complete, the disc caps re-compress the disc stack and normal filtration is resumed. The water required for the backflush is a fraction of the amount used by traditional sand filters. With the reduced backflush frequency and water volume, water consumption and discharge are reduced. Performance Since the installation of the VRTX system, water samples were taken on an regular basis for water chemistry and bacteria analysis. Condenser tubes and the inside of the condensers are periodically inspected for scale buildup. In addition, corrosion coupons were installed. The performance of VRTX treatment can be summarized as follows: Old scale was gradually removed and no new scale has formed: Two weeks after

installation, it was noticed that the hard scale on the condenser tubes started to soften. Most


of the scale was removed after 2 months of operation. A recent inspection shows that the condenser tubes are clean (Photo 2), and scale buildup on the inside walls of the condensers have been significantly reduced.

Very low bacteria counts: Total bacteria counts in cooling water normally ranges from 400

to 2,500 CFU/ml (Chart 1), measured by standard Pour Plate Method (Standard Method 9215 B). Over nearly a 2 ½ year period the bacteria counts averaged ~ 1,000 CFU/ml.

Excellent corrosion control: Corrosion rates of less than 2.2 mpy for galvanized steel and

2.5 mpy for carbon steel and less than 0.3 mpy for copper alloy. The latest coupon test indicates that the corrosion rate is less than 0.15 mpy for copper alloy and less than 0.80 mpy for carbon steel (Chart 2).

Photo 2. A Recent Photo Showing Clean Condenser Tubes

Water Savings The brewery installed water meters on both makeup and blowdown lines 47 days prior to the installation of the VRTX system. Readings from both meters have been recorded regularly ever since. Chart 3 summarizes the results. During the 47 days prior to VRTX treatment (6/22/2001 – 8/8/2001), the blowdown was 395,600 gallons and makeup consumption was 901,600 gallons. Since the installation of VRTX system, the blowdown rate has decreased significantly. It took 311 days to reach the same amount of blowdown, as opposed to 47 days with chemical treatment. Over the 28 months period of VRTX treatment, total makeup water consumption is 7.52 million gallon and total blowdown volume is 1.09 million gallons.


0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

07/28/01 11/05/01 02/13/02 05/24/02 09/01/02 12/10/02 03/20/03 06/28/03 10/06/03

Date

Bac

teria

Cou

nt (C

FU/m

l)Chart 1. Bacteria Test Results Since VRTX Treatment

Chart 2. Summary of Corrosion Test Results Since VRTX Treatment

2.34 2.27

0.735

1.982.18

0.25 0.22 0.14

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Year 2001 Year 2002 Year 2003

Cor

rosi

on R

ate

(mpy

)

C1010

HD GALV

CDA 110

Based on the above data, the makeup and blowdown rates were calculated, as shown in Chart 4. The makeup consumption rate and blowdown rate were 19,115 and 8,387 gallons per day, respectively, prior to VRTX treatment. With VRTX treatment, the average blowdown rate was reduced to 1,323 gallons per day and makeup rate was 7,624 gallons per day. Chart 4 also shows the changes of makeup consumption and blowdown rates with seasons, as evaporation rate changes with plant production and weather conditions. Consequently, the makeup consumption and blowdown rates are higher during the summer season and lower during the winter season.


-

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

7,000,000

8,000,000

6/22/01 9/30/01 1/8/02 4/18/02 7/27/02 11/4/02 2/12/03 5/23/03 8/31/03

Date

Gallo

nsChart 3. Recorded Makeup Water Consumption and Blowdown Data

-

5,000

10,000

15,000

20,000

25,000

30,000

6/22/01 9/30/01 1/8/02 4/18/02 7/27/02 11/4/02 2/12/03 5/23/03 8/31/03

Date

Gallo

ns/da

y

VRTX Treatment

Makeup

Blowdow

Chart 4. Makeup and Blowdown Rates Calculated from The Metered Data

To eliminate the effect of different seasons on evaporation, makeup consumption and blowdown are charted to provide a direct comparison of water consumption during the same period (June 22 – Aug. 8) with chemical treatment for the Year 2001 and VRTX treatment for the Year 2002 and 2003 (see Charts 5 – 8). In the summer months the daily makeup water declined from an average of 19,115 gallons to 12,324 gallons, representing a 36% reduction. The daily blowdown declined from an average of 8,387 gallons to 1,929 gallons, representing a 77% reduction. Total water savings during this 47-day period is over 325,500 gallons.


Cycles of concentration are also calculated using the recorded data. Prior to VRTX installation the cycles were 2.3. For 28 months since VRTX installation, the average cycles of concentration are 6.9, as shown in Chart 9. Based on the estimated annual average evaporation rate from the recorded data, the water savings are over 1.8 million gallons per year with VRTX treatment, compared to previous chemical treatment.

-

100,000

200,000

300,000

400,000

500,000

600,000

700,000

800,000

900,000

1,000,000

6/22 6/27 7/2 7/7 7/12 7/17 7/22 7/27 8/1 8/6

Date

Gal

lons

Year 2002

Year 2003

Year 2001

Chart 5. A Direct Comparison of Makeup Consumption During The Same Period

-

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

6/22 6/27 7/2 7/7 7/12 7/17 7/22 7/27 8/1 8/6

Date

Gal

lons

Year 2002Year 2003

Year 2001

Chart 6. A Direct Comparison of Blowdown During The Same Period


5,000

10,000

15,000

20,000

25,000

30,000

35,000

6/22 6/27 7/2 7/7 7/12 7/17 7/22 7/27 8/1 8/6

Date

Mak

eup

Rat

e (g

allo

n/da

y)

VRTX Treatment

Prior to VRTX

Chart 7. A Direct Comparison of Makeup Rate During The Same Period

-

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

6/22 6/27 7/2 7/7 7/12 7/17 7/22 7/27 8/1 8/6

Date

Blo

wdo

wn

Rat

e (g

allo

n/da

y)

VRTX Treatment

Prior to VRTX

Chart 8. A Direct Comparison of Blown Rate During The Same Period


2.3

7.0

7.9

5.4

6.9

1

2

3

4

5

6

7

8

9

Prior to VRTX VRTX (2001) VRTX (2002) VRTX (2003) VRTX (Av erage)

Cyc

les

of C

once

ntra

tion

Chart 9. Comparison of Cycles of Concentration Conclusion Industrial growth, population growth, evermore stringent worldwide environmental regulations, aquifer pollution and water shortage are resulting in increased restrictions and rising costs on the use of water for a thirsty, water-consuming world, whose past demands have been insatiable and whose past practices have ill treated this vital resource. Water is a scarce resource in many parts of the world. Water is becoming a scarce resource in many parts of the United States. Industries are now challenged to reduce water consumption. The VRTX system provides a reliable and cost-efficient alternative to traditional methods of chemical treatment, with the added benefit of substantial reductions in water consumption. With the installation of the VRTX system, discharge from this cooling system has been reduced. Consequently, the makeup water consumption also decreased. The annual water saving is over 1.8 million gallons. In addition, the discharge water now contains no hazardous chemicals. Along with water savings, field observations and laboratory tests have indicated that the VRTX system is superior in controlling problems, such as scale, corrosion and bacterial activities, when compared to the previous chemical treatment program.



Acknowledgements The authors wish to thank Mr. Carlos Alvarez, owner of the Spoetzl Brewery, for his permission to publish the data referenced in this article. About the authors: Peter Koestler is the brewery engineer at The Gambrinus Company in San Antonio, TX. Wiley Wang, Ph.D., is Senior Chemist at VRTX Technologies, where he is responsible for all research and development activities. Robert Kelsey is the Executive Vice President/General Manager of VRTX Technologies, LLC.

1

Chemical vs. Non-chemical Cooling Water Treatments – a Side-by-Side Comparison

KEVIN A. KITZMAN, Alcoa, Alcoa Center, PA and EDWARD F. MAZIARZ, Alcoa, Pittsburgh, PA and BOBBY PADGETT, Alcoa, Goose Creek, SC CHARLES D. BLUMENSCHEIN, USFilter, Pittsburgh, PA ALAN SMITH, Consultant, Pittsburgh, PA IWC - 03 - 22

KEYWORDS: cooling towers, non-chemical treatment technologies, chemical treatment, scale control, monitoring, micro-biological control, corrosion control ABSTRACT: Two non-chemical cooling water treatment technologies, a pulsed power system, and a hydrodynamic cavitation device, were evaluated against conventional chemical treatment in a detailed six-month study. The comparison was conducted at Alcoa’s Mt. Holly Works in Goose Creek, SC. Three identical, non-contact, evaporative coolers used to cool air compressor systems were used as the test sites. The three technologies were installed on the external 600-gpm spray water loop in these towers. The towers were operated at a range of setpoints representing between 4 – 20 cycles of concentration. Water quality parameters, corrosion, scaling indices, microbiological (aerobic and anaerobic planktonic and sessile, Legionella), and aquatic toxicity data/results will be presented for the different operating conditions. Operating cost comparisons will also be shared. Conclusions and Recommendations will be presented.

BACKGROUND

With the rising cost of chemical treatment additives for cooling water systems, and Alcoa’s announced goal of reducing water usage and wastewater discharges by 60% by 2008, locations were eager to explore alternatives for reducing their water treatment costs and achieving the Company’s conservation goals. Spurred by the perceived successes of two non-chemical, mechanical/electrical cooling water treatment devices at the Alcoa Corporate Center, and an extrusion plant in Louisiana, other locations were anxious to install their own systems. These devices were reported to successfully control scaling, corrosion, and microbiological growth in recirculating cooling water systems. Ten more units were installed in 2001, and approximately the same number installed in 2002.

These non-chemical devices were being installed at such a rapid pace, that there was little opportunity to gather historical performance data beyond the claims and testimonials provided by the manufacturers. Because of the great interest in these technologies corporate-wide and the potential for significant cost savings and environmental benefits, it was decided to perform a side-by-side evaluation of two technologies. These were an electromagnetic pulsed power system (herein after called PP), and a hydrodynamic cavitation unit (herein after called HDC). Both were compared to conventional chemical treatment.

Mt. Holly Works was selected for the test site because they had three identical, but segregated (i.e. separate influent and effluent water lines with no interconnection between the basins) cooling towers

serving their air compressor systems, and they were interested in evaluating these non-chemical technologies.

PROJECT OBJECTIVE(S)

The objectives of this study were to evaluate the effectiveness of commercially available mechanical/electrical treatment devices for controlling corrosion, scale formation, and microbiological growth in recirculating cooling water systems, and to substantiate the following performance claims made by both non-chemical treatment suppliers: • Eliminates essentially all chemical additives with the

associated cost savings • Eliminates most of the employee health, safety, and

environmental issues typically associated with the storage and use of chemicals

• Reduces water consumption through operation at higher cycles of concentration. Blowdown reduced 80% and Makeup by 40% when cycles increase from 2 to 6

• Reduces or eliminates the toxicity of blowdown to aquatic life in receiving streams

• Degrades and removes old scale encrustations and bio-film from wetted pipes and tower surfaces

• Overall lower operating costs including savings in water and sewer charges, maintenance, chemicals, and labor.

2

PROJECT APPROACH

A. MT. HOLLY’S COOLING TOWERS. Mt. Holly has three, Baltimore Aircoil (BAC)

Model F1463 – PR evaporative coolers serving their air compressor system (see Figure 1). A closed internal loop cools the main air compressors and air dryers. The internal loop is comprised of 10 pass cooling coils. Each coil is a 1.05 in. external diameter tube with 2 in. spacing between the coils. Hot internal loop water from the combined compressor system is returned and distributed across the three segregated, independently functioning, BAC towers.

Figure 1. BAC Towers (back)

Water in each 1,200 gallon basin is recirculated to the top of the tower at 610 gpm, distributed through a piping manifold, and then sprayed through a series of nozzles over the closed loop coils. Fans (3) at the base of each tower, blow air upwards against the falling spray water. Drift eliminators placed above the pipe manifold trap water droplets and return them to the tower basin. This comprises the outer loop cooling circuit, and the chemical characteristics and microbiological content of this spray water was the subject of this six-month study.

The towers were installed in 1996, with design specifications to cool 343 gpm of internal loop water from 116 0F to 90 0F at an 80 0F wet bulb. Each tower has a design heat load of 4.45 MM BTU/hr. or approximately 297 Tons of cooling. At design conditions of a 26-degree change in temperature, and 343 gpm flow rate the estimated evaporation rate of the recirculated water is approximately 12,800 gpd. Design drift loss is 0.001% or 0.006 gpm (~9.0 gpd). The basins of all three towers were flushed and thoroughly cleaned before the test program began. B. TREATMENT ALTERNATIVES. Chemical Treatment. Chemical treatment is presently used in the external loop spray water in all three towers. The following control ranges for the chemical treatment program are maintained:

pH = 6.5 – 9.0 Conductivity = 400 – 1,000 umhos

Molybdate = 2.0 – 5.0 mg/l A corrosion/scale inhibitor is continuously fed to

maintain molybdate levels in the desired control range. In addition to sodium molybdate and a phosphate compound for mild steel corrosion control, the corrosion inhibitor contains a triazole for copper corrosion protection, and an acrylic polymer for scale control. Chemical treatment also includes two non-oxidizing biocides, isothiazolin and glutaraldehyde, on an alternating, batch dosage basis to maintain microbiological control in the three towers. Under this control program the towers normally operate at 3 to 5 cycles of concentration with a typical discharge (blowdown) rate of 3,200 – 6,400 gallons per day.

As part of this investigation the middle tower was left on the chemical treatment program while the West and East towers were converted to the mechanical/electrical treatment approaches. The performance of the two non-chemically treated towers was measured against that of the conventional, chemically treated middle tower.

Pulsed Power. The pulsed power (PP) system was installed on the West tower (see Figure 2). The PP imparts a pulsed, high frequency (100,000 hz) electromagnetic energy into the circulating water by inducing varying magnetic fields 60 times per second. The system is normally sized according to the pipe diameter of the recirculating cooling water. The coils operate at a low voltage of less than 45 volts rms1.

Figure 2. 3” PP with Insulated Cyclonic Separator

The PP supplier prefers to size their units to turn over the system volume once every 15 minutes for good microbiological control. The initial 110 gpm flow rate through the 3 in. unit created a 10-minute turnover frequency. When a larger 6-inch unit was installed, the turnover frequency increased to once every 2 minutes.

A centrifugal pump draws water from a 3 in. drain connection on the bottom of the tower, and directs it through the 3 inch PP unit, and then to a cyclonic separator for solids removal before returning the treated

3

water back to the tower basin. The reason for installing the separator was to achieve results that could be compared to the hydrodynamic cavitation system (discussed next). The hydrodynamic cavitation system typically is installed with a solids separation device.

Hydrodynamic Cavitation. The hydro-dynamic cavitation (HDC) system was installed on the East tower. The HDC unit is based on the principle of controlled hydrodynamic cavitation. Cavitation is the dynamic process of the formation, growth, and collapse of micro-sized bubbles in a fluid. Studies have shown that when a liquid moves fast enough, gas bubbles will form and collapse creating a process called cavitation. In turbulent liquid flows, and notably at high velocity, hydrodynamic cavitation will occur2.

Figure 3. HDC System with Bag Filters and Cyclonic Separator

A 20 gpm unit was sized and selected by the supplier for Mt. Holly. HDC units are sized to treat the entire system volume seven times a day plus an allowance for makeup water. Based on the size of this tower and expected heat load this would have calculated to a 15 gpm unit, but the HDC supplier elected to increase to the next larger size in their product line, a 20 gpm unit, rather than use their smallest 10 gpm model.

The HDC unit was installed with 2 process loops. One circuit draws water from the tower basin through a 3 in. strainer, and pumps it through a cyclonic separator and then through duplex bag filters before returning it through a piping manifold into the bottom of the tower basin. The second process water loop draws water from the tower basin through a 2-in. strainer before entering the HDC unit. C. SAMPLING AND MONITORING.

The following parameters were monitored: • Water meter readings were recorded at least 5 times

per week and included make-up and blowdown. • Water temperatures were monitored continuously

from various points in each tower. • pH and conductivity were measured from makeup

supply, and outer loop spray water from all three towers on a daily basis.

• Corrosion was measured using a rack that contained slots for five (0.5 in. x 3.0 in.) coupons. A mild steel coupon occupied one slot, an aluminum coupon another, and a steel electrode used in conjunction with a Corrater®3 meter a third. Corrater® readings were taken once per week.

• Microbiological analysis included a variety of techniques to measure planktonic and sessile organisms, both aerobic and anaerobic. Two perforated steel coupons were used to provide a growth substrate for sessile organisms.

• Aquatic Toxicity of the tower blowdowns was also measured during the study.

D. OPERATING SET POINTS Before the trial, the three towers were operating

between 3-5 cycles of concentration as determined by comparing the conductivities of the water in the recirculating spray loop with that of the makeup water.

The intent of the trial was to operate the three towers at increasingly higher cycles of concentration to see how the different treatment technologies would perform under increasing scale-forming potentials. Corrosion rates and microbiological control for both planktonic and sessile organisms were also measured during these different operating periods. The different operating settings were controlled by adjusting the conductivity setpoint on the continuous conductivity monitor in each tower. When the conductivity of the recirculating spray water reached the setpoint, the conductivity controlled blowdown valve would automatically open and some high conductivity water would be purged from the tower, and sent to drain. Since the blowdown setpoint (based on conductivity) was set the same for all three towers, it was considered an independent variable of this test.

Conductivity controllers were initially set at 1,000 umhos to approximate 5.0 cycles of concentration. The following table lists the dates and setpoints for the operating periods used during this study.

Table 1. Conductivity Setpoints Dates 2002

Conductivity Setpoint (umhos)

Intended Cycles of Concentration

6/24 – 8/4 1,000 5.0 8/5 – 8/26 1,250 6.0 8/27 – 9/12 1,650 8.0 9/13 – 9/23 2,150 8.0 9/24 – 12/5 2,7501 10.0 12/6 – 12/20 4,0002 16.0 12/21 2,150 Test over Notes: 1. The Chemical tower was set back to 2,150 umhos on

November 1, due to excessive buildup of suspended solids in the recirculating spray water loop.

2. Only the PP and HDC non-chemical towers were operated at 4,000 umhos

4

Figure 4. History of Makeup Conductivity

In early September, the conductivity of the makeup water increased dramatically as shown in Figure 4 necessitating an increase in the conductivity setpoints of the three towers on September 13, in order to maintain 8.0 cycles of concentration. The 2,750 umhos operating period lasted the longest - 73 days – as a variety of tests and observations were made during that time frame. The last operating period was used to try to “stress” the non-Chemical towers to see how they would perform under high scaling conditions.

TECHNICAL RESULTS OR FINDINGS A. CYCLES OF CONCENTRATION.

Cycles initially based on water meter readings were calculated by dividing the makeup water by the blowdown + solids separator purge + drift losses. Only the PP and HDC towers had solids separators, which contributed small daily purges of 26 and 73 gpd respectively. Drift losses were calculated as 9 gpd for all towers, which was the designed drift loss specified by the manufacturer. While it was initially thought that cycles based on water meter readings would be the most accurate, unmeasured leakage and observed drift losses that appeared higher than calculated likely contributed to higher values using this method.

When the conductivity settings were moved to 1,250 umhos during the next operating period, cycles based on magnesium, sodium, sulfate, and potassium concentrations were also calculated and compared against cycles based on conductivity for each of the operating periods. These are plotted in Figure 5. The similarity amongst towers of cycles based on salts verified that this was the best way to measure true cycles during this test.

Note that potassium values from the Chemical tower were not used in the calculation on Figure 5 since potassium was a component of the chemical additives used in this tower.

The quick and easy method of using the ratio of conductivities was not appropriate for calculating cycles, particularly as the test program moved into potentially higher scaling regimes. It was determined that for all of the operating periods starting with 1,250 umhos that cycles based on magnesium, potassium (not including Chemical tower), sodium, and sulfate were the most representative of the true operating concentration factors in the towers.

Note that as the conductivity settings are increased in later operating periods, the cycles determined from the constituent values increase progressively higher than the conductivity values. This was taken as an indication that calcium carbonate and other compounds were precipitating at higher chemical saturation levels, thereby lowering conductivity values, and yielding lower than actual cycles of concentration determinations. This precipitation might have been via nucleation in the recirculating water as the non-chemical vendors claim, or simply scaling on the heat exchangers. Additional discussion on scale is provided later in this text. Two values are plotted for the Chemical Tower during the 2,150 umhos operating period. As was mentioned previously, the Chemical Tower was only operated at the 2,750 umhos setpoint for a limited time, and was dropped back to 2,150 umhos on November 1. B. WATER METER READINGS.

Figure 6 shows the average daily blowdowns for each of the three towers during the various operating periods. This blowdown includes discharge from the

Makeup Conductivity History

190

210

230

250

270

290

310

330

7/3 8/2 9/1 10/1 10/31 11/30 12/30

umho

sStartup 1250 1650 4000

Operating Periods (umhos)

2150 2750

5

solids separator devices from the non-chemical units. The only statistical difference (95% confidence) was between the PP and HDC tower during the startup period. It is not clear why this might have occurred, however it can be assumed that the three towers were not completely in

control during this time period so some anomaly during startup caused the difference. Similarly there was no statistical difference in make-up water usage between the three towers.

Figure 5. Cycles of Concentration History

Figure 6. Average Daily Blowdowns

1,50

01,

359

1,31

8

1,14

11,

009

1,06

6

943

885 883

745

692

716

346 42

6

339

173

181

0

200

400

600

800

1,000

1,200

1,400

1,600

Aver

age

Blow

dow

n (g

pd)

Startup 1250 1650 2150 2750 4000

Conductivity Periods (umhos)

Average Daily Blowdown for Three Towers

PP Chem HDC

Cycles of Concentration Average of Na, SO4, Mg, K

vs Conductivity

6.9

6.7 8

14.9

23.7

6.7

5.8

8.7

14.2

6.8

6.2

8

15.4

22.8

11.1

8

5.7 6.3 7.

1

10.0

16.0

0

5

10

15

20

25

1250 1650 2150 2750 4000

Operating Periods (umhos)

Aver

age

Cycl

es o

f Con

cent

ratio

n

PP Chem I HDC Chem II Hand Conductivity

Note: Potassium Cycles were not used for Chem Tower

6

C. WATER QUALITY DATA. In a further attempt to identify any performance

differences between the three towers, water samples were taken an average of twice per week from the makeup source as well as the three towers and the following parameters were measured: alkalinity, total dissolved solids (TDS), total hardness, calcium hardness, and turbidity.

The average values for all parameters showed a steady increase across the operating periods as the cycles of concentration were raised. TDS, total alkalinity, total hardness, calcium hardness and conductivity were scrutinized statistically and generally showed no difference between all three towers for each operating period respectively. A few exceptions are as follows. First, during the 2,750 umhos time frame the alkalinity for the Chemical tower was higher than that of either non-chemical tower. The reason for this is likely due to the scale inhibitors keeping scale forming elements in solution for the Chemical tower, while scale or other precipitates were forming in the non-chemical towers.

Second, while the turbidity levels track fairly closely across the towers at average values below 10 NTU over the first two periods, the Chemical tower began to show some elevated spikes during the 1,650 umhos period. As the conductivity controlled blowdown volumes continued to decrease with increasing cycles, the adverse affect of not having a sidestream solids separator became more pronounced in the Chemical tower. Finally at the 2,750 umhos control point while both non-chemical towers were still operating below 10 NTU, the Chemical tower recorded spikes above 120 NTU with an average

over 70 NTU. After five weeks of operation at these elevated cycles (~14), the chemical supplier recommended reducing the setpoint back to 2,150 umhos so that their chemical treatment program would not be compromised. Within days of returning to this lower setpoint, the affect of larger volume blowdowns successfully reduced turbidity levels below 20 NTU where they stayed for the remainder of the study. The ambient air scrubbing affect of the cooling towers, and the close proximity of the three towers to a carbon silo may have exaggerated the high turbidity levels observed in the Chemical tower due to lack of side stream filtration on this unit. D. CORROSION CONTROL.

Weekly Corrater® readings were used to supplement the corrosion results determined by weight loss of the mild steel coupons. The mild steel coupons were removed and replaced three times at various intervals during the study period. The aluminum coupons were only removed once, after 77 days in the rack. Unfortunately, the last set of coupons, steel and aluminum, were lost in transit and never recovered.

Figure 7 is a graphical presentation of all of the corrosion rate data obtained during the study. The Chemical tower exhibited excellent control averaging < 1.3 mils/year over the entire study period based on Corrater® readings. The rate exhibited a gradual decline from approximately 1.3 mils/year at the start of the study to < 0.8 mils/year during the final month. This extraordinary rate was partly attributable to the fact that the chemical supplier did not adjust the feed rate of their

Figure 7. Corrater Corrosion History

7

corrosion inhibitor for most of the study. Thus as blowdown volumes decreased, the concentration of corrosion inhibitor increased. While the stated control range was 3-5 mg/l of molybdate, they ranged from 3.0 – 18.5 mg/l concentration with an average of 8.5 mg/l.

Corrater® readings for the two non-chemical towers also exhibited an overall decrease over the study period, although the decline was not as consistent. Initial readings were at the 3.7 to 3.6 mils/year for the PP and HDC towers respectively, and they dropped to 1.4 and 1.9 mils/year respectively at the conclusion of the study. However, the HDC tower spiked to almost 5.0 mils/year for two weeks midway through the study. The overall decrease in corrosion rates could be attributed to either or both of two mechanisms: 1) further passivation from operation in moderate to severe calcium carbonate scaling regimes over the course of the study, and 2) cycling up of the ortho-phosphate levels in the makeup water which acted as a passivating anodic inhibitor and provided another measure of corrosion protection. Corrosion rates between 1 - 3.0 mils/year for mild steel are considered very good control. Excellent control is < 1.0 mils/year. Corrosion coupon results are also shown on the Figure 7. These results are summarized below on Table 2.

Table 2 – Corrosion Coupon Results

PP Chem HDC Mild steel @ 28 days

1.84 0.72 0.83

Mild steel @ 72 days

1.50 0.22 1.71

Note: All results in mils/year

E. SCALE CONTROL. Precipitated solids form both soft and hard scale

deposits on the heat transfer surfaces, which increases the resistance to heat transfer, thereby decreasing the thermal efficiency of the equipment. One of the claims of the non-chemical water treatment manufacturers is that their devices prevent the formation of scale on heat transfer surfaces, allowing the dissolved solids (calcium carbonate) to be precipitated in the bulk water, and then purged from the systems via blowdown or in sidestream solids removal devices. To be effective from the standpoint of corrosion control, and in some cases microbiological control too, non-chemical cooling water treatment systems must operate in an alkaline pH range, which encourages scale formation.

A convenient place to start when evaluating scale is a scaling index. For purposes of this paper, only the Practical Scaling Index (PSI4) will be discussed, although both the Langelier and Ryznar indices were also calculated and produced similar results as the PSI.

The PSI showed steady movement into increasingly severe scaling regimes until about the third week in October or approximately midway through the 2,750 umhos operating period. At this time a change in direction toward lower scaling conditions is noted (see Figure 8). This shift could be somewhat explainable for the Chemical tower, because that operating setpoint was reduced back to 2,150 umhos on November 1st due to high turbidity resulting from reduced blowdown rates as described above. Lower operating cycles of concentration created by this shift, should translate into lower scaling indices, but it initially was puzzling why the two non-chemical towers followed the same pattern.

Figure 8. Practical Scaling Index

8

A possible explanation may be found by looking at the alkalinity and calcium hardness levels, which both took a dramatic downward turn at the same time. These variables are believed to have precipitated the changes in the PSI. Figures 9 and 10 plot the history of alkalinity and calcium hardness levels in the three towers over the study period.

Alkalinity levels in the makeup water remain fairly steady at 60 mg/l from the middle of September until dropping to 50 mg/l after December 5th. Makeup calcium hardness peaks on October 24, at 73 mg/l and then begins a steady drop to 44 mg/l by the end of the study. These changes in the make-up water, particularly for calcium hardness, could partially account for the observed drop in their respective concentrations in the tower spray water. Another possible explanation for the drop in these values for the spray water in the two non-chemical towers was the fact that the treatment devices were turned OFF from November 6 – 21 to determine what, if any, influence they were having on the water chemistry in those two towers. Without any “treatment” from the non-chemical devices, and with the recirculating spray water at saturation levels for calcium carbonate as determined by the scaling indices, its possible that calcium carbonate may have precipitated from the bulk water onto heat transfer surfaces. This is evidenced by a reduction in carbonate and alkalinity as shown in Figures 9 and 10 during the period when the non-chemical devices were shut OFF. When the non-chemical devices were turned back ON, their respective “treatments” kept the calcium carbonate in solution and the levels of alkalinity and calcium hardness began to increase slightly again for the PP and HDC towers. However when the towers were increased to the 4,000 umhos operating mode, the calcium levels increased substantially while the alkalinity levels remained steady (Figure 9). This indicates that calcium

carbonate was being formed and remained in the recirculating water where it would be measured by the test for calcium, but not measured in the standard titration test for alkalinity. The alkalinity test most likely did not dissolve all of the calcium carbonate in the recirculating water, thus resulting in a lower alkalinity measurement.

The significance of this finding is that it has been hypothesized that certain non-chemical treatment devices accelerate the coagulation-flocculation of solid particles suspended in water, and increase the crystal formation in the bulk solution instead of deposition as scale on heat-transfer surfaces. 5 F. TEMPERATURE OBSERVATIONS.

Cooling efficiency is presented in Figure 11, which plots the Return (to tower) minus the Supply (to process) water temperatures for the study period. This differential is a good measure of the cooling efficiency of each tower. All three towers delivered 17-21 degrees of cooling until late in October when the PP tower started to have a drop-off in efficiency. Water supplied back to the compressors from the PP tower was 3 to 9 degrees warmer compared to the HDC tower, while the Chemical tower remained in the middle of the pack at 2 to 4 degrees warmer than the HDC tower. In an attempt to correct what we interpreted as an indication of scale formation on the heat exchange tubing in the PP tower, a larger (6 inch) PP unit was installed on the Spray Water loop on November 26. Nothing else was done to clean the tubes at that time. The temperature differential for the PP tower began to improve, even when the operating setpoint was increased to 4,000 umhos. Water temperature monitoring was continued after the test program was concluded, and a noticeable although somewhat erratic improvement for the PP tower continued. This could be construed that the larger, greater powered PP unit removed some of the previously formed scale on the heat exchange surfaces.

Figure 9. Alkalinity History

9

Figure 10. Calcium Hardness History

Figure 11. Return – Supply Water Temperature Differential

10

The HDC tower consistently had the best cooling performance across the entire study, and their differential increased to 20 – 22 degrees during the last six weeks. G. SCALE OBSERVATIONS.

After the test program was completed, the towers were opened to observe the presence of any scale accumulation. Both the PP and HDC towers appear to have more tube scale than the Chemical tower, perhaps an indication that the scale inhibitor in the chemical treatment program was performing as intended. Alternatively this could be due to the fact that the non-chemical towers were operated in severe scaling regimes for longer time periods. However, close scrutiny of the sidewall scale shows distinctive differences between the three towers. The sidewall scale in both the PP and HDC towers had a sparkling, crystalline appearance compared to the dull, flat white scale in the Chemical tower. See figures 12, 13, and 14.

Figure 12. PP Tower – Scale in Sidewall.

Figure 13. HDC Tower – Scale on Sidewall

Figure 14. Chemical Tower – Scale on Sidewall

Mt. Holly Maintenance cleaned the towers shortly after the test, and an effort was made to qualify the cleaning procedures required to remove the scale in the three towers. The entire top layer and some of the bottom layer of “sparkling – crystalline” scale in the PP and HDC towers were removable by water under “garden hose” pressure. However, there were some spots of tenacious scale that required “water blast” pressure for removal and not all scale was accessible for removal. The Chemical tower didn’t have as much scale, but it was more difficult to remove and required water blasting. The smaller amount of scale in this tower was expected, due to both the scale inhibitor included in the chemical treatment additive, and the fact that this tower did not operate at the length and severity of the scaling regimes selected for the PP and HDC towers. No acid cleaning was done on any of the towers, and according to plant maintenance staff, some scale had been present on the heat exchanger tubes and sidewalls of all three towers before the study began (although all three towers where cleaned and flushed prior to the study).

Samples of scale were obtained from both of the non-chemical towers and analyzed by X-ray diffraction (XRD) at the Alcoa Technical Center. The results show that the majority of scale was calcite, which is a typical form of calcium carbonate in cooling water. Calcite is characterized by its crystal formation rather than chemical composition. There was a significant amount of carbon in the sample, which was confirmed visually by its blackish-gray color. It’s possible that carbon was part of the particulate that was “trapped” in the scale matrix. H. MICROBIOLOGICAL CONTROL.

Cooling water systems, particularly open recirculating systems, provide a favorable environment for the growth of microorganisms. Microbial growth on wetted surfaces leads to the formation of biofilms. If uncontrolled, such films cause biofouling, which can adversely affect equipment performance (i.e. increased pump pressures, heat transfer problems, etc.), promote metal corrosion (i.e. microbial induced corrosion or MIC), accelerate wood deterioration, and cause pathogen

11

concerns (e.g. Legionella). There have also been increasing federal and state regulatory restrictions regarding chemical use in cooling towers, aimed at reducing the aquatic toxicity of effluent discharges to receiving waters. This has made microbiological control in cooling towers more difficult, and encouraged searches for alternatives to current treatment methods.

As was mentioned earlier, the PP and HDC technologies claim different mechanisms for controlling microbiological growth. Both technologies were compared against a conventional chemical approach involving alternating two non-oxidizing biocides - isothiazolin and glutaraldehyde.

Analytical Methods. There are a variety of techniques available for measuring microbiological activity in cooling towers. For this investigation the bulk recirculating spray water was analyzed for aerobic free floating or planktonic organisms by an FDA approved procedure (FDA Bacteriological Analytical Manual, 8th Ed. Chapter 3) which utilizes "Standard Plate Count Agar," with a 48 hr. incubation period @ 35C. Anaerobic planktonic bacteria were also measured using Standard Methods “Standard Plate Count Agar” with a 48 hr. incubation period in an anaerobic environment, @ 35C. All results were reported in colony forming units per milliliter (CFU/ ml). Planktonic samples were generally taken every two weeks.

Attached, slime forming, or sessile organisms colonize wetted surfaces and are primarily responsible for biofouling. While some information was available on the effectiveness of non-chemical treatment devices against planktonic organisms, very little information had been collected on their ability to control sessile organisms. Perforated steel coupons were included in the coupon racks and one additional coupon was hung inside the

cooling loop coil pack to encourage sessile growth. The coupons were routinely removed, after approximately four weeks of exposure to the flowing spray water and shipped to an independent laboratory. Upon receipt by the laboratory, both sides of the coupon were swabbed and the swabs returned to the buffer solution used for shipment. A vortex mixer was used to remove the organisms from the swabs. This “inoculated” solution was then tested using Standard Methods procedure for Heterotrophic Plate Count. The results were expressed as CFU/cm2. The sessile samples were also analyzed for sulfate-reducing bacteria with their presence reported as a simple Positive or Negative result.

Planktonic Results. Figure 15 provides a graphic summary of the planktonic aerobic results for all three towers over the course of the study.

The chemical supplier was alternating two non-oxidizing biocides (isothiazolin and glutaraldehyde) each week. Note that both non-chemical towers were able to control aerobic planktonic organisms with PP averaging approximately 65,000 CFU/ml and HDC averaging approximately 95,500 CFU/ml. Plate counts < 100,000 CFU/ml are considered good control. In contrast the Chemical tower suffered through swings in aerobic counts, ranging from a low of 7 CFU/ml to a high of 8,000,000 CFU/ml. The average count for the Chemical tower was approximately 1,270,000 CFU/ml. The type and dosage of chemical biocides is also recorded on Figure 15. The chemical supplier applied some extremely high dosages (>1,200 mg/l) in an effort to get the tower under control, but the correlation of plate counts with biocide doses was mixed. Generally, one would expect to see lower counts shortly after biocide addition. However, isothiazolin is a slow acting product,

Figure 15. Planktonic Counts – Aerobic History

Planktonic - Aerobic

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

8/2 8/22 9/11 10/1 10/21 11/10 11/30 12/20

CFU

/ml

0

200

400

600

800

1000

1200

1400

PP ChemHDC Biosperse 250 - IsothiazolinBiosperse 254 - Glutaraldehyde

8,000,000

Biocide D

osage (mg/L)

GlutaraldehydeIsothiazolin

12

and a significant bacterium kill usually required 16-20 hours of contact. Glutaraldehyde is a faster acting product, and will provide a significant reduction in the bacteria population within 2-6 hours. So sampling time relative to specific microbiocide addition has a significant impact on the measured results. It is important to alternate non-oxidizing biocides as the organisms can develop immunity to a single product over time.

Note that all of these results were cultured from unfiltered samples, which put the Chemical tower at somewhat of a disadvantage, because their higher turbidity levels due to lack of a sidestream solids removal device. These particulates gave bacteria a place to “hide-out” as neither of the biocides are considered to be effective to penetrate the interstices of the particles. This certainly became more evident as cycles of concentration were increased with a concurrent reduction in blowdown rates. Shortly after the operating set point was increased to 2,750 umhos on September 24, equivalent to approximately 14 cycles in the Chemical tower, turbidity levels steadily increased to over 100 NTU while planktonic counts soared reaching a peak of 8,000,000 CFU/ml on October 28. The following table compares the average turbidity levels with average planktonic counts during the 2,750 umhos operating period for the three towers.

Table 3 - Average Turbidity and Planktonic Values @ 2,750 umhos Operating Setpoint

Tower Turbidity

(NTU) Planktonic Aerobes (CFU/ml)

PP 10.7 70,000 Chemical 71.1 2,600,000 HDC 14.7 80,000

With conditions in the Chemical tower steadily

worsening, the decision was made to drop the control point back to 2,150 umhos to increase the blowdown rates and help purge the tower. Conditions showed a steady improvement from that point to the end of the study, but this experience served to emphasize the importance of a sidestream solids removal device for maintaining tower cleanliness.

Figure 16 shows a similar history for planktonic anaerobic results. Once again the non-chemical towers provided more consistent control averaging 85 and 87 CFU/ml respectively for the PP and HDC towers. The Chemical tower exhibited wide swings for anaerobes ranging from <1 to 1,200 CFU/ml with an average of 290 CFU/ml for the study period.

Figure 16. Planktonic Counts – Anaerobic History

13

Sessile Results. The following table summarizes the sessile aerobic results for all three towers over the course of the study. All results are presented as colony forming units per square centimeter (CFU/cm2). The area of the 0.5” x 3.0” perforated, mild steel coupon used to attract sessile growth was 21.8 cm2 (including both sides and edges minus perforations). The coupons designated as “cold” were located in the coupon racks with the corrosion coupons through which a sidestream of recirculating spray water (~ 5.0 gpm) was sent. The “hot” coupons were suspended approximately mid-way in the cooling coil pack at the top of the tower. These coupons were difficult to locate, and toward the end of the study, they were subjected to alternating “wet” and “dry” conditions as the spray water pumps and fans turned “OFF” and “ON” based on temperature settings during cold mornings to minimize potential icing problems. These set points were eventually lowered to permit more continuous water flow, but the “hot” coupons were not analyzed during this period.

The results indicate that with only a few exceptions, the non-chemical devices provided effective biofilm control. The HDC system delivered the best results of all three technologies with all of their coupons measuring less than 1x 106 CFU/cm2 (Average = 1.9 x 105 CFU/cm2) which is an accepted industry standard for control of slime forming organisms6. The PP system performed almost as well with only two coupons measuring over 106 organisms / cm2. However, the last sample is difficult to explain, as it is two orders of magnitude higher than another coupon removed from the same rack after an identical four-week exposure period. The average results for the PP tower with and without the last sample were 2.7 x 106 and 5.9 x 105 CFU/cm2 respectively. By contrast the majority (62%) of the coupons from the Chemical tower were above the target 1x 106 CFU/cm2 control point with a calculated average = 2.5 x 106 CFU/cm2.

All coupons were tested for sulfate reducing bacteria (SRBs). The presence of SRBs is an indication of anaerobic conditions, which can promote corrosion of fouled metal surfaces. As shown in the above table, seven

coupons tested positive for SRBs, no indications of MIC were observed. Four of the seven coupons were from the Chemical tower.

Analyses conducted by Ecole Polytechnique for planktonic, heterotrophic organisms indicated that only a moderate level of disinfection was occurring across all of the towers with a significant proportion of the population remaining viable/active. This is consistent with PP claims that their process is more bacteriostatic than bactericidal. However, HDC claim of destroying microbial populations by high temperature, pressure, and vacuum is not as evident 7,8

Seven samples taken from each of the three towers

tested for Legionella. None tested positive (> 1.0 CFU/50 ml.). However, the study protocols were not specifically designed to determine the efficacy of these chemical and non-chemical technologies toward controlling Legionella organisms9.

Algae Control. Although these Baltimore Aircoil Towers were closed on four sides, the exposed moist drift eliminators on the top of the tower afforded a perfect haven for algae growth. The non-oxidizing biocides added to the Chemical tower and carried in the drift were effective algaecides as indicated by the stark black color of the drift eliminators on top of the Chemical tower. The PP and HDC towers on the other hand showed patches of green algae, with the PP showing the most extensive coverage. Although neither non-chemical supplier claims efficacy for algae control, algae growth will require additional maintenance (e.g. physical cleaning or periodic use of an algaecide). I. WHOLE EFFLUENT TOXICITY (WET) TESTS.

As effluent discharge limits become more stringent, and facilities strive toward greater water conservation on the way to “zero liquid discharge” the quality of our cooling tower blowdown waters will play a more significant role in attaining “zero non-compliance.” One of the anticipated advantages of using non-chemical cooling water treatments is the elimination of aquatic toxicity concerns, and a test program was developed to validate this premise.

Table 4 – Summary of Sessile Results

Date PP Chemical HDC Removed Cold Hot Cold Hot Cold Hot

7/9 9.5x104 1x103 8x104 8/5 2.6x105 4.1x105 1.8x105 9/5 8x104 2x105** 1.6x104 9/23 6.8x105 1.2x105 1x106 9x105 1.6x105 1.5x105

10/21 2x106 1.1x106 7.5x105** 10/28 4.5x105 1.3x104 2.1x106** 1.6x105 1.3x105 7x104 11/25 4.5x105** 8.1x106** 1.4x105 11/25 1.7x107 6.8x106** 8.6x104**

Concentration < 1x106 = Biofilm Control - EFFECTIVE 1x106 < Conc.< 1x107 = Biofilm Control - CONSIDER MODIFICATION for IMPROVEMENTConcentration > 1x107 = Biofilm Control - IMPROVEMENT REQUIRED

** Sulfate Reducing Bacteria tested positive in these samples

14

Whole effluent toxicity (WET) procedures were run using static acute definitive toxicity tests (96 hours for Pimephales promelas, and 48 hours for Ceriodaphnia dubia) at the following concentrations of blowdown: 6.25%, 12.5%, 25%, 50%, and 100%. With the Chemical tower the time of sampling relative to the time of biocide application has a great impact on blowdown toxicity. Three samples were sent for testing. Two were taken on the same day after biocide addition (isothiazolin), and the last one six days after biocide addition (glutaraldehyde). The first sampling event was corrupted for the PP tower. The results of the last two tests are summarized in Table 5 below.

Note that both non-chemical treatment systems were shut down during the last sample taken on November 19, and had been off since November 6. This is not expected to impact the results of blowdown toxicity. Blowdowns from all three towers exhibited toxicity to Ceriodaphnia in all three samples. This can be explained by the effect of salinity on freshwater organisms. The improved survival of Pimephales Promelas (i.e. fathead minnow) from the Chemical tower is evident on the last test. This is due to taking the sample 6-days after biocide addition, compared to taking the sample right after addition on October 28th. A value of >100.0 means that all organisms survived at all blowdown concentrations. J. OPERATING AND MAINTENANCE COST COMPARISONS.

Another important aspect of this study, in addition to a performance evaluation of the three technologies, was to determine the costs to install, operate, and maintain them. A number of cost categories were established to provide a basis for comparison. These included: depreciation, supplies, makeup water, blowdown water (assuming treatment / surcharge costs), power consumption, R&M materials, labor charges (i.e. O/M, supervisory, contracted services, and administrative oversight), and costs to address toxic blowdown. Table 6 below summarizes the results of the net costs (+) or savings (-) to operate the two non-chemical devices compared to conventional chemical treatment, all at 4 cycles of concentration.

A few notes on the tables. Recall that this study involved a 1,600-gallon system. At this size, in this location, neither non-chemical device is competitive with chemical treatment, because the capital costs for the chemical system were fully depreciated. For the extrapolations starting at a 2,500-gallon system, capital costs for all three systems are included and depreciated over a 10-year period. All costs for the extrapolation were based on this study except that capital costs were based on vendor input. Installation costs were extrapolated from the cost of the Mt. Holly system based on six-tenths rule applied to the ratio of model sizes (e.g. square of diameters for PP and flow rates for HDC). Comparison was made for oxidizing and non-oxidizing biocides in order to capture costs or savings based on this

Table 5 – Summary of Toxicity Test

% of blowdown corresponding to 50% mortality (LC50) October 28, 2002 November 19, 2002

2,750 umhos 2,750 umhos

2,150 umhos

2,750 umhos

PP Chemical HDC PP Chemical HDC Pimephales Promelas (96 hour LC 50)

>100.0 70.8 >100.0 >100.0 >100.0 >100.0

Ceriodaphnia Dubia (48 hour LC50)

36.6 11.3 35.4 29.0 9.5 21.0

Table 6 – Cost Comparison at 4 Cycles of Concentration

Size System

(gallons)

Chemical System

Comparison

Pulsed Power

HDC

Annual Cost (+) or Savings (-)

Simple Payback (yrs)

Annual Cost (+)

2,500 gal Oxidizing $161 NA $3,049 2,500 gal Non-Oxidizing -$114 295 $2,774 10,000 gal Oxidizing -$1,749 24 $2,754 10,000 gal Non-Oxidizing -$2,851 15 $1,652 50,000 gal. Oxidizing -$5,432 12 $10,904 50,000 gal. Non-Oxidizing -$10,931 6 $14,567 100,000 gal. Oxidizing -$10,884 7 $21,541 100,000 gal. Non-Oxidizing -$21,899 4 $10,525

15

variable. It must be noted however that although oxidizing biocides (such as bleach) are generally cheaper, they require much more containment, ventilation and controls. The costs for this infrastructure were included in the comparison. Finally maintenance labor and materials costs for the HDC unit for this test were high due to the presence of the carbon particles and their effect on the pumps and the HDC hardware. Subsequent to our field trial, the HDC manufacturer made improvements to their design, which eliminated the need for one of their high maintenance filtration devices. For the cost analysis above, essentially similar maintenance costs were used for both non-chemical systems. A significant portion of the maintenance expenses for the chemical system was assumed to be included in the unit cost of the chemicals.

Non-chemical suppliers claim they can run at 6 to 8 cycles of concentration, which this study shows is feasible. The chemical systems typically run at 3 to 5 cycles, which was where the chemically treated towers where running prior to this test. Cost comparison at 8 cycles for non-chemical and 4 cycles for chemical treated systems is shown in Table 7. Recall from Figure 6 that going from 8 to 4 cycles for these towers resulted in an approximate 50% blowdown reduction.

As shown on the tables, the HDC does not seem to have the economies of scale that the PP unit does. This is due to the significant capital cost increases required by the HDC suppliers to move into the larger systems. In addition to capital cost, the power costs are also higher for the HDC units to run the extra pumps required. Both non-chemical systems would compare even more favorably if consideration was given to potential fines levied for violations of WET test standards associated with toxic blowdowns discharged from chemically treated cooling towers. Another potential benefit of non-chemical treatment is a reduction in amount of reportable chemicals into the environment as reported in the annual Toxic Release Inventory (TRI) report (e.g.chlorine from bleach). Lastly, the better heat transfer results exhibited by the

HDC system during this trial, while difficult to quantify, could also favorably impact their overall operating costs.

SUMMARY AND CONCLUSIONS

Based upon the results and findings of this twenty-six week study we offer the following conclusions:

A general observation was that fugitive dusts emanating from a nearby carbon silo were readily drawn into the three cooling towers by the air scrubbing action of the tower fans. While the abrasive carbon fines caused some operating difficulties for pumps associated with the PP and HDC systems, the lack of a sidestream solids removal device severely compromised operation of the Chemical tower at elevated cycles creating unacceptable (>120 NTU) turbidity levels.

The make-up (MU) and blowdown (BD) volumes across all three towers were statistically the same (95% CI) at all setpoints. This is expected, as conductivity setpoints were the same across all three towers.

Conductivity is not reflective of true cycles of concentration in higher cycles (greater than 7 cycles at this site). Additionally, make-up and blowdown in this study proved to be an inaccurate measure of cycles of concentration due to possible unquantified leaks and drift loss. The best way to track cycles of concentration at higher cycles is by comparing salt (Mg, Na, K, SO4, etc) contents of the make-up and blowdown.

TDS, total alkalinity, total hardness, calcium hardness and conductivity were scrutinized statistically, and showed no difference between all three towers for each operating period respectively, with a couple exceptions. During the 2,750 umhos time frame the alkalinity for the Chemical tower was higher than that of either non-chemical tower. The reason for this is likely due to the scale inhibitors keeping scale forming elements in solution for the Chemical tower, while scale or other precipitates were forming in the non-chemical towers. Turbidity was higher in the Chemical tower at higher cycles due to the lack of a solids separation device.

Table 7 – Cost Comparison at 8 Cycles for Non-Chemical and 4 Cycles for Chemical Treated Systems

Size System (gallons)

Chemical System

Comparison

Pulsed Power HDC





2,500 gal Oxidizing -$8,168 4 -$5,280 9 2,500 gal Non-Oxidizing -$8,443 4 -$5,556 8 10,000 gal Oxidizing -$22,253 2 -$17,750 4 10,000 gal Non-Oxidizing -$23,354 2 -$18,851 4 50,000 gal. Oxidizing -$34,897 2 -$18,570 9 50,000 gal. Non-Oxidizing -$40,404 2 -$24,078 7 100,000 gal. Oxidizing -$49,328 2 -$16,903 17 100,000 gal. Non-Oxidizing -$60,343 1 -$27,919 10

16

The Chemical tower provided the best corrosion control with rates averaging 1.1 mils/year over the entire study period. Corrosion rates for the non-chemical devices were higher, but still within industry standards. This may have been due to the action of the non-chemical devices in operating in moderate to severe calcium carbonate scaling regimes, or simply cycling up of the ortho-phosphate levels in the makeup water, which acted as an anodic inhibitor.

Both the PP and HDC towers showed a greater potential for scale formation when these non-chemical devices were turned OFF from November 6-21. This is inferred from the reduction in both alkalinity and calcium hardness during that time (i.e.presumably scale was forming on the surfaces when the devices were OFF). However when the devices were turned back on and moved up to 4,000 umhos calcium hardness increased, while alkalinity stayed low. A reasonable explanation of this result is that calcium carbonate was forming in the water column (very small particulate) as suppliers had claimed. Additional evidence for scale control could be inferred by the effect a larger PP unit had on improving heat transfer of its respective tower.

The sidewall scale in both the PP and HDC towers had a sparkling, crystalline appearance compared to the dull white scale of the Chemical tower. The buildup in the PP scale may have been an example of the “ripening” affect whereby the crystal size of the precipitate increases. The entire top layer and some of the bottom layer of “sparkling – crystalline” scale in the PP and HDC towers was removable by water under “garden hose” pressure. However, there were some spots of tenacious scale that required “water blast” pressure to remove. The Chemical tower had less scale accumulation, probably as a result of operating at lower cycles of concentration for extended periods, and the affects of the scale inhibitor additive. However, this scale was difficult to remove and required water blasting.

Both of the non-chemical towers delivered better and more consistent microbiological control for both aerobic and anaerobic planktonic organisms compared to the Chemical tower. There was no statistical difference in the microbiological control performance between the PP and HDC technologies. Despite alternating two non-oxidizing biocides, microbiological control performance in the Chemical tower suffered due to higher turbidity levels created by lack of a sidestream solids removal device. While there was some correlation between these high turbidity levels and high microbiological counts, this does not appear to fully explain all the biological variability in the Chemical tower.

Microbiological control for slime forming, sessile organisms was better and more consistent for the non-chemical systems. There were only a few instances where the non-chemical treatment systems did not provide effective control, and both instances occurred in the PP tower. The HDC technology delivered the best results with all of their test coupons measuring < 1x106

CFU/cm2, which is considered the standard for effective biofilm control.

Seven of thirty (23%) sessile coupons tested positive for sulfate reducing bacteria (SRB), which is an indicator of anaerobic conditions that can promote corrosion of fouled metal surfaces. Of these seven positive coupons, four were from the Chemical tower, two were from the HDC and one was from the PP.

The non-oxidizing biocides added to the Chemical tower appeared to also serve as effective algaecides as evidenced by the stark, black surface of the drift eliminators on this tower. The top of the drift eliminators on the PP and HDC towers exhibited scattered patches of green algae with the largest coverage on the PP tower.

The results showed a clear cost savings advantage for the PP systems against the other two technologies across all system sizes.

RECOMMENDATIONS

As locations implement water conservation measures by operating their cooling towers at higher cycles of concentration, they should strongly consider installing side stream solids removal devices to improve system cleanliness at reduced blowdown levels.

Locations considering new applications of these non-chemical approaches as a water conservation measure should carefully evaluate their source of makeup water and the overall system economics. Makeup water with reasonable levels of hardness (40 – 50 mg/l) and ortho-phosphate (0.5-1.0 mg/l) will result in better corrosion control than soft water without any additives. Addition of a sidestream solids removal device is also highly recommended. In very abrasive environments (i.e. lots of dust/sand in the air) the standard sidestream removal may not be enough to protect mechanical non-chemical treatment devices. Locations currently experiencing algal control problems should be aware that these non-chemical technologies may not provide effective algal control.

17

REFERENCES

1. Condenser Water Treatment Results Under Pulsed-Power Technology by J. Lane (PP Systems), and D.F. Peck (Hatch Mott MacDonald). Presented at the Cooling Tower Institute Conference 2002

2. An Innovative and Alternative Method for Cooling Water Treatment by R. Kelsey, D. Koontz, and W. Wang (HDC Technologies). Presented at the International Water Conference, October 2001

3. Rohrback Aquamate Corrater User Manual, Corrater is a registered trademark™ of Rohrback Cosasco

4. Cooling Water Scale & Scaling Indices: What They Mean – How to Use Them Effectively – How They Can Cut Treatment Costs by P. R. Puckorius and G. R. Loretitsch (Puckorius and Associates, Inc). Presented at the International Water Conference, October 1999

5. Physical Water Treatment for the Mitigation of Mineral Fouling in Cooling-Tower Water Applications, by Y.I. Cho Ph.D, S. Lee, and W. Kim presented at ASHRAE Conference January 2003

6. Dr. Benoit Barbeau, Personal Communication, Date Feb 17, 2003

7. Alternative Methods of Microbiological Control – Review of Major Rapid Techniques by Petat etal, STP Pharma Pratiques 6:6, pp. 449-464 1996

8. BacLight: Application of a New Rapid Staining Method for Direct Enumeration of Viable and Total Bacteria in Drinking Water by B. Barbeau etal, Journal of Microbiological Methods, Vol. 37, pp 77-86, 1999

9. Guideline: Best Practices for Control of Legionella by Cooling Tower Institute, February 2000.

Chemistry Induced by Hydrodynamic Cavitation

Kenneth S. Suslick,* Millan M. Mdleleni, andJeffrey T. Ries

Department of ChemistryUniVersity of Illinois at Urbana-Champaign

601 South Goodwin AVenue, Urbana, Illinois 61801

ReceiVed July 1, 1997

Cavitation (the formation, growth, and implosive collapse ofgas or vapor-filled bubbles in liquids) can have substantialchemical and physical effects. While the chemical effects ofacousticcavitation (i.e., sonochemistry and sonoluminescence)have been extensively investigated during recent years,1-5 littleis known about the chemical consequences ofhydrodynamiccavitation created during turbulent flow of liquids. Hydrody-namic cavitation is observed when large pressure differentialsare generated within a moving liquid and is accompanied by anumber of physical effects, erosion being most notable from atechnological viewpoint.6,7 In contrast, reports of hydrodynami-cally induced chemistry or luminescence and direct comparisonsto sonochemistry or sonoluminescence have been extremelylimited.8,9

In aqueous liquids, acoustic cavitation leads to the formationof reactive species such as OH•, H•, and H2O2. These short-lived species are capable of effecting secondary oxidation andreduction reactions. For example, iodide can be sonochemicallyoxidized to triiodide by OH• radicals or H2O2 produced duringcavitation. From aqueous solutions containing chlorocarbons,Cl• and Cl2 are also liberated in high yields and this increasesrates of iodide oxidation.10 The rate of triiodide formation iseasily monitored spectrophotometrically. For many years, thisso-called Weissler reaction has remained the standard dosimeterfor sonochemical reactions.The recent advent of commercially available high-pressure

jet fluidizers capable of pressure drops as high as 2 kbar andjet velocities approaching 200 m/s has led to numerousapplications in the physical processing of liquids, for emulsifica-tion, cell disruption, etc. Chemical consequences of suchprocessing, however, have received little examination. Oneimportant exception comes fromW. R. Moser and co-workers,11

who have shown that such a device can be utilized to preparenanostructured catalytic materials. Moser speculated that theunusual properties of his catalysts resulted from hydrodynamiccavitation within the fluidizer.11 We describe here conclusiveexperimental evidence for chemical reactions caused by hydro-dynamic cavitation within a jet fluidizer.

In a typical run,12a 60 mL of 1.0 M KI in purified watersaturated with carbon tetrachloride was introduced at a constantflow rate into the Microfluidizer with a liquid pressure of 1.24kbar. The reaction solution temperature increased 10 to 12°Cwithin 90 s and stabilized at the temperatures reported herein.Aliquots (4 mL) of the processed solution were periodicallyextracted from the reaction system by airtight syringes, analyzedspectrophotometrically, and returned to the reservoir afteranalysis. The rate of I3- formation was calculated from thechange in absorbance at 353 nm (ε ) 26 400 M-1 cm-1)12b asa function of reaction time. Initial studies conducted with Ar-sparged water gave relatively low rates of I3

- production;saturation of the Ar sparged H2O with CCl4 resulted in a 20-fold increase in I3- production, as has been typically observedfor ultrasonic cavitation.10 This is attributed to ready formationof Cl• and Cl2 from CCl4 under cavitation conditions.The effect of upstream liquid pressure on the rate of I3

-

production was investigated over the range 100-1500 bar. Thereaction rate increases linearly with liquid pressure (Figure 1),but with a threshold pressure of 150 bar. Below 150 bar ofhydrostatic pressure, no chemical reactions were observed; thisprobably represents the minimum jet velocity necessary toinduce cavitation. The resistance of a turbulent flow tocavitation is given by its cavitation numberσ, as defined in eq1:6

wherepd, pu, andpv are the downstream, upstream, and vaporpressures, respectively, and the approximation holds whenpu. pd . pv, as they do under our experimental conditions. Anincrease in upstream pressure should decreaseσ and increasethe number of cavitation events. This in turn should increasethe rate of I3- formation, if the chemistry is cavitation driven,consistent with our observations.The conditions formed duringacousticcavitation and con-

sequently sonochemical rates are known to be affected both bythe polytropic ratio of the dissolved gas (i.e.,γ ) Cp/Cv,) andby the thermal conductivity of the dissolved gas. The formerparameter determines the temperature achieved during bubblecompression, and the latter is responsible for heat dissipationfrom the collapsing bubble to the surrounding solution. In thepresent study using Ar/He mixtures, theγ of the dissolved gaswas fixed at 1.67, while the thermal conductivity was variedfrom 0.017 to 0.142 W m-1 K-1. As shown in Figure 2, theI3- formation rate decreases exponentially as the thermalconductivity of the dissolved gas increases. This observationis best explained in terms of the hot-spot model for cavitationwhich suggests that the maximum temperature (Tmax) realized

(1) Suslick, K. S.MRS Bull.1995, 20, 29(2) Mason, T. J., Ed.AdVances in Sonochemistry; JAI Press: New York,

1990-1994; Vols. 1-3.(3) Price, G. J., Ed.Current Trends in Sonochemistry; Royal Society of

Chemistry: Cambridge, U.K., 1992.(4) Suslick, K. S.Science1990, 247, 1439.(5) Suslick, K. S., Ed.Ultrasound: Its Chemical, Physical, and Biological

Effects; VCH: New York, 1988.(6) Knapp, R. T.; Daily, J. W.; Hammitt, F. G.CaVitation; McGraw-

Hill: Inc.: New York, 1970.(7) (a) Young, F. R.CaVitation; McGraw-Hill: Inc.: New York, 1989.

(b) Brennen, C. E.CaVitation and Bubble Dynamics; Oxford UniversityPress: Oxford, U.K., 1995.

(8) Anbar, M.Science1968, 161, 1343.(9) (a) Verbanov, V. S.; Margulis, M. A.; Demin, S. V.; Korneev, Y.

A.; Klimenko, B. N.; Nikitin, Y. B.; Pogodaev, V. I.Russ. J. Phys. Chem.1990, 64, 1807. (b) Margulis, M. A.; Korneev, Y. A.; Demin, S. V.;Verbanov, V. S.Russ. J. Phys. Chem.1994, 68, 828.

(10) (a) Weissler, A.; Cooper, N. W.; Snyder, S.J. Am. Chem. Soc. 1950,72, 1769. (b) Ibisi, M.; Brown, B.J. Acoust. Soc. Am.1967, 41, 568. (c)Clark, P. R.; Hill, C. R. J. Acoust. Soc. Am.1970, 47, 649. (d) Chendke, P.K.; Fogler, H. S.J. Phys. Chem. 1983, 87, 1362.

(11) (a) Moser, W. R.; Marshik, B. J.; Kingsley, J.; Lemberger, M.;Willette, R.; Chan, A.; Sunstrom, J. E.; Boye, A.J. Mater. Res.1995, 10,2322. (b) Moser, W. R. Personal communication.

(12) (a) Reagent grade KI and CCl4 were obtained from Aldrich ChemicalCo. and used as received. High-purity water was prepared with a BarnsteadNANOpure. Spectrophotometric measurements were obtained with a HitachiU3300 UV-vis double-monochromator spectrophotometer. MKS mass flowcontrollers 247C were used to adjust the composition of the Ar/He spargegas. Hydrodynamic cavitation studies were performed with an air-drivenmodel M-110Y Microfluidizer from Microfluidics International Corp., 30Ossipee Rd., Newton, MA 02164. The reaction solutions were exposed onlyto stainless steel or glass. The reaction solution was first sparged with high-purity argon or Ar/He mixtures and light-proofed to prevent CCl4 photo-decomposition and then injected into the pressurizing reservoir through aself-sealing septum. A portion of the reaction solution was pressurized bya large pneumatically driven pump into an interaction chamber, where twopulsed flows were redirected at each other through jewel orifices withvelocities of≈190 m/s11bcontrolled by a back-pressure regulator. Cavitationcan occur when there is sufficient turbulence upon liquid jet impact or whenthere exists a sufficient pressure drop as the streams pass through the orifices.High-velocity pumping is also accompanied by bulk heating of the flowingliquid. The reaction chamber, pump, and plumbing were therefore immersedin a thermally equilibrated water bath. The processed stream was returnedto the solution reservoir for recirculation and analysis. (b) Awtrey, A. D.;Connick, R. E.J. Am. Chem. Soc.1951, 73, 1842.

σ )pd - pvpu - pd

≈ pdpu

(1)

9303J. Am. Chem. Soc.1997,119,9303-9304

S0002-7863(97)02171-9 CCC: $14.00 © 1997 American Chemical Society

in a collapsing bubble decreases linearly with increasing thermalconductivity of the entrapped gas.13 This inverse relationshipof Tmax to the thermal conductivity of dissolved gases shouldlead to an exponential decrease in the I3

- formation rate(assuming Arrhenius behavior) with increasing thermal con-ductivity, as observed.The influence of bulk solution temperature on the I3

-

production rate was investigated to probe the effect of vaporpressure. The I3- production rates decrease sharply withincreasing temperature, as shown in Figure 3. This samebehavior has been reported for many previousacousticcavitationstudies.10b,14,15 Figure 3 also demonstrates that the observedrates decrease exponentially with increasing solvent vapor

pressure. This same dependence is seen in sonochemicalreactions and is attributed to the increase in polyatomic vaporinside the bubble before collapse, which decreasesγ andcushions the collapse of the cavitating bubble.15 While it isdifficult to make direct comparisons of energy efficiency (e.g.,moles of product per kilowatt hour of electrical energy), acousticcavitation provides significantly higher rates for the Weisslerreaction, at least for the specific source of hydrodynamiccavitation tested here.In summary, we have demonstrated that the chemical effects

of hydrodynamic cavitation and acoustic cavitation respondidentically to experimental parameters, notably the bulk tem-perature and the nature of the dissolved gas. In particular, therates decrease with increasing solution temperature, due to theincreased solvent vapor pressure inside the bubble; increasingsolvent vapor pressure attenuates the efficacy of cavitationalcollapse, the maximum temperature reached during such col-lapse, and, consequently, the rates of cavitational reactions. Byreducing the adiabaticity of bubble collapse, the thermalconductivity of the dissolved gas also has a substantial effecton the maximum temperature achieved inside a cavitatingbubble; increased thermal conductivity decreased rates ofcavitation reactions.

Acknowledgment. This work was supported by the NationalScience Foundation (CHE 94-20758) and in part by the DOE. TheMicrofluidizer used in this study was provided on loan from Catalytica,Inc., and Microfluidics International Corp. We thank Dr. David L. Kingand Prof. W. R. Moser for useful discussions.

JA972171I

(13) Young, F. R.J. Acoust. Soc. Am.1976, 60, 100.

(14) (a) Sehgal, C.; Sutherland, R. G.; Verrall, R. E. J. Phys. Chem.1980, 84, 525. (b) Didenko, Y. T.; Nastich, D. N.; Pugach, S. P.; Polovinka,Y. A.; Kvochka, V. I.Ultrasonics1994, 32, 71.

(15) (a) Suslick, K. S.; Gawienowski, J. J.; Schubert, P. F.; Wang, H.H.Ultrasonics1984, 22, 33. (b) Suslick, K. S.; Gawienowski, J. J.; Schubert,P. F.; Wang, H. H.J. Phys. Chem. 1983, 87, 2299.

(16) (a) Washington, C.; Davis, S. S.Int. J. Pharm. 1988, 44, 169. (b)Lidgate, D. M.; Fu, R. C.; Fleitman, J. S.Biopharm1989, 2, 28. (c) Lidgate,D. M.; Trattner, T.; Shulz, R. M.; Maskiewicz, R.Pharm. Res. 1992, 9,860.

Figure 1. Dependence of triiodide formation rate on the hydrodynamicpressure used to induce cavitation. Conditions: 60 mL of 1 M KIsolution in CCl4-saturated water was recycled under static Ar atmo-sphere at a constant reaction temperature of 12°C.

Figure 2. Dependence of triiodide formation rate on the nature of thedissolved gas during hydrodynamic cavitation. Conditions: 60 mL of1 M KI solution in CCl4-saturated water was recycled under static Aror Ar/He atmosphere at a constant reaction temperature of 12°C andliquid pressure of 1.24 kbar.

Figure 3. Dependence of triiodide formation rate on bulk temperatureand on solvent vapor pressure during hydrodynamic cavitation. Condi-tions: 60 mL of 1 M KI solution in CCl4-saturated water was recycledunder static Ar atmosphere at liquid pressure of 1.24 kbar.

9304 J. Am. Chem. Soc., Vol. 119, No. 39, 1997 Communications to the Editor

Documents

5(6($5&+ :+,7( 3$3(56 $1' 7+,5' Papers & Evaluati… · S. Frenette Sali Group, Ann Arbor, MI D. Koontz, W. Wang VRTX Technologies, LLC 5850 Corridor Parkway Schertz, TX 78154 ABSTRACT