CA1081604A - Cooling tower scaling control - Google Patents

Abstract

FOR COOLING TOWER SCALING CONTROL

ABSTRACT
A method for controlling the level of scale forming species in open recirculating cooling tower systems comprising:
A) maintaining in the cooling tower water from 0.5 to 300 ppm by weight of a scale inhibitor; and B) continuously removing a portion of the cooling tower water to form a sidestream, treating the sidestream in a lime-soda softening process to remove at least a portion of the softenable species thereby forming a softened sidestream and a sedimentation sludge; and C) returning the softened sidestream to the cool-ing tower water; and D) collecting and disposing of the sedimentation sludge.

Description

108~0~

The use of water as a heat transfer medium has historically been practiced in a wide variety of industrial applications. In the past, the relative ease with which large amounts of good quality water could be obtained favored the installation and general accep-tance of "once-through" cooling water system designs. However, recent enactment of strict water pollution control legislation along with the shortage of water available for cooling purposes in many geographic areas, have prompted the design and installation of open recirculating cooling tower systems to serve the cooling needs of many industries.
Although the use of open recirculating cooling systems re-sults in a significant reduction in water usage compared to "once-through" cooling, increasingly stringent effluent discharge require-ments for cooling towers will render this form of cooling impractical unless means are available to improve blowdown quality or to reduce or eliminate blowdown altogether. Our invention is addressed to the second of these alternatives.
Our studies of zero blowdown systems have entailed research employing pilot cooling tower systems equipped with bypass lime-soda softening equipment. This work has demonstrated the feasibility of a zero blowdown approach to recirculating cooling tower systems.
We have been able to identify the degree of sidestream softening required for a given system. More importantly we have identified the maximum safe concentration and softenability of certain dissolv-ed chemical species at steady state levels of non-softenable ions.
Finally, we have established the softenability of scale inhibited water thereby establishing the feasibility of the utilization of alkaline, sidestream-softened cooling towers.
The alkaline, sidestream-softened operation which we de-scribe herein constitutes an important contribution to the ~ .

10~60~

developing art of zero blowdown recirculating cooling tower systems. This new method avoids the destruction and replenish-ment of carbonate which is inherent in pH controlled, sidestream-softened operations thereby affording substantial reductions in soda ash and acid usage. It also avoids the significant corrosion problems inherent in pH controlled systems. Other advantages of our invention will be apparent from the discussion below.
In conventional recirculating cooling tower systems, the rate of blowdown is regulated to maintain a concentration ratio of about 2 to 5 cycles. This rate of blowdown is necessary in order to control the concentration of dissolved and suspended solids in the system. When the concentration ratio exceeds about 2 to 5 cycles, the increasing concentration of the dissolved and suspended solids creates an extremely corrosive recirculating water which, due to the high concentration of scale-forming ions, also causes sig-nificant scale deposition problems.
Unfortunately, cooling towers maintaining 2 to 5 cycles of concentration by constantly bleeding off recirculating water, may not be able to comply with the Water Pollution Control Act Amendments of 1972 which established a national goal of achieving zero blowdown discharge of pollutants by 1985. The need to comply with the requirements of this law as well as the desire to achieve water savings has resulted in the development of zero blowdown water systems which operate at extremely high cycles of concentration. We refer to these systems as zero blowdown systems because they elim-inate the traditional bleed off of the recirculating water.
In these zero blowdown systems the major source of water loss, besides evaporation, is wind drift. It is contemplated that, some of the low wind drift systems presently being developed will be able to achieve a steady state with recirculating water at over 50 cycles of concentration and with total dissolved solids values exceeding 50,000 ppm.

1~)81604 Some of the zero blowdown systems already on the drawing board or in operation utilize ion removal systems such as lime-soda sidestream softening. Other ion removal systems which may be used include reverse osmosis, electrodialysis, freeze crystallization, distillation, ion exchange and zeolite softening.
Actual operating experiences with zero blowdown recirculat-ing cooling tower operations are limited. One such operation was reported by M. E. Parry and J. B. Matson in a paper entitled "Complete Reuse of Cooling Tower Blowdown," which was presented at the A~CHE Convention, Houston, Texas, 1975. This system used side-stream lime softening to limit hardness levels and carbon dioxide for pH control to limit hardness levels by maintaining the pH below the calcium carbonate scaling threshold.
OBJECTS
An important object of our invention is to provide to the art a new and practical method for preventing scale formation in zero blowdown recirculating cooling tower systems.
Another object of our invention is to provide a scale pre-vention method which will make possible the operation of open cooling tawer systems at very high cycles of concentration.
Yet another object of our invention is to provide a method for operating zero blowdown cooling tower systems at alkaline pH
without significant scaling problems.
A further object of our invention is to teach a new method for controlling calcium sulfate scaling in very highly concentrated recirculating cooling tower water.
Other objects will appear hereinafter.
BRIEF DESCRIPTION OF DRAWINGS
The drawings are intended to clarify the discussion below.

Figure 1 is a schematic representation of an evaporative cooling tower system operated with limi~ed water discharge and sidestream softening.
Figure 2 is a graphic representation of the calcium sul-fate scaling data generated in Example 1 below. This figure is intended to facilitate interpretation of the data generated in Example 1.
THE INVENTION
Our invention is directed toward the treatment of open recirculating cooling tower systems operating without blowdown.
According to the method of our invention, the level of scale-forming species in the alkaline recirculating cooling tower water will be controlled by:
A) maintaining in the cooling tower water from 0.5 to 300 ppm by weight of a scale inhibitor; and B) continuously removing a portion of the cooling tower water to form a sidestream, treating the sidestream in a softening process to remove at least a portion of the softenable species thereby forming a softened sidestream and a sedimentation sludge; and C) returning the softened sidestream to the cool-ing tower water; and D) collecting and disposing of the sedimentation sludge.
The alkalinity of the recirculating cooling tower water should be maintained between pH 8 and pH 10; preferably, it should be maintained between pH 8 and pH 9.
Scale Inhibition The methods employed for controlling scaling in conven-tional open recirculating cooling tower systems generally entailed one of the following:

:

A) softening makeup water to remove scaling material, B) adding acid to maintain recirculating water pH in a range in which scale-forming matter remains dissolved, or C~ treating the cooling tower water with spe-cific scale inhibitors to prevent scale deposition.
~e have found that in the highly concentrated recirculating water of zero blo~down systems, softening makeup water alone is an im-practical method for controlling the concentration of scale-forming chemical species. Similarly, continuously softening the recirculat-ing ~ater is also insufficient to satisfactorily control the levels of scale-forming species.
The addition of sufficient acid to maintain system pH in which scale forming materials reMain dissolved has been suggested as an alternative to softening. This approach is impractical because of the excessive acid dosages which are required to achieve a pH
sufficientl~ acidic to control the high concentrations of such scale-~0 forming species as calcium sulfate. Also, the acidic corrosivity of the recirculating water poses a significant problem.
Finally, we have determined that the use of scale inhibitors to control the extremely high concentrations of scale-forming species present in the highly concentrated recirculating water of zero blow-down cooling tower systems is impractical. The heavy dosages of scale inhibitors required would rende~ the zero blowdown concept uneconomical. Further, such high levels of scale inhibitors would interfere with other treatment chemicals present in the system, such as microbiocides and corrosion inhibitors.
The combination of sidestream softening and acid pH have been suggested for controlling scale-forming ions in zero blowdown ;
. -6-, . .
. . .

1~81604 recirculating water. Unfortunately, there are serious drawbacks associated with this combination. Corrosive amounts of acid would s~ill be required to control non-softenable species present in the recirculating water. Also, alkalinity contributed to the system by the sidestream softener would further increase acid requirements and hence the expense and inconvenience associated therewith.
Our invention is concerned with a zero blowdown cooling tower system which operates at alkaline pH. We have combined side-stream softening with the use of scale inhibitors. The success of our method is unexpected because conceptually water softening and the scale inhibition operate at cross purposes. That is, it is the function of the water softener to precipitate scale-forming species whereas it is the purpose of the scale inhibitor to prevent the pre-cipitation of scale. Further, it would be expected that excessive removal of the scale inhibitor by the water softener would shutout the usefulness of the combination of a water softener and scale inhibitors. Nevertheless, we have found that the predicted exces-sive interference between softening and scale inhibition does not materialize.
While, as we have just noted, the combination of scale inhibitors and sidestream softening offers great promise for control-ling scaling in zero blowdown cooling tower systems, this combination will not control the high levels of SO4 found in these systems.
This sulfate problem confronts designers of zero blowdown systems whether they contemplate control of scaling by water softening, acid controlJ scale inhibitors, or by any practical, presently available scale inhibiting system.
Within the range of normal cooling tower operating pH, the solubility of calcium sulfate is not greatly dependent on the system pH. Therefore, pH control, which is commonly used in preventing CaCO3 deposition, is ineffective against calcium sulfate. Thus, ~08~604 the common practice of adjusting tower water to a low enough ptl to convert carbonate to bicarbonate is impractical with respect to sulfate--an equivalent conversion of sulfate to bisulfate would require the pH to be reduced below pH l' Traditionally) the control limit for calcium sulfate has been determined based on the solubility product relationship of calcium and sulfate. We have found that this approach leads to substantial underestimation of the maximum tolerable concentration of Ca and SO4 in recirculating systems operating at high levels of concentration such as those found in zero blowdown systems. Hill and Wills in a paper in the Journal of the American Chemical Society, Volume 60, page 1647 (1938) demonstrated the shortcomings of the ion solubility product approach for predicting the concentration limita-tion of the Ca and the SO4= ions. These data were confirmed in later studies by Yeats and Marshall reported in J. Chem. Eng. Data 17:163 (1972). This early work, in which aqueous solutions of the desired composition were equalibrated with solid calcium sulfate was concerned with static systems quite unlike cooling towers. Never- !
theless, the results obtained by tlill and others provided the jumping-off point for a significant discovery which constitutes an important aspect of the method of our invention.
We have discovered that the commonly used ion solubility product is inapplicable in the presence of high carbonate concentra-tions. Thus, as long as the Ca concentration is maintained below certain threshold levels, calcium sulfate deposition will not occur.
In other words, scaling due to the high levels of sulfate (probably as high as 30,000 ppm or more as Na2SO4) expected in zero blowdown cooling tower recirculating water can be controlled by controlling Ca levels.
Examination of the data generated in Example l indicates that with sulfate concentrations below 30,000 ppm (Na2S04) a calcium limit 1~)81604 of 800 ppm (CaCO3) will preclude scaling by calcium sulfate. Furthermore, since it is well known that the presence of non-common ions such as magnesium and chloride can substantially increase the solubility of calcium sulfate, the calcium limits indicated by these results can be considered minimum values.
Minimum calcium limits in actual zero blowdown systems would commonly be at least 1000 ppm. Of course, due to the wide possible variation in makeup waters and other operationg parameters, calcium limits below 800 ppm might be encountered. Therefore, a very safe calcium limit would be 500 ppm. Cal-cium maximums, where desired, may be determined on a case-by-case basis through the application of techniques similar to those employed in Example 1.
Preferred scale inhibiting compounds useful in the practice of our invention include amino phosphonic acids, diphosphonic acids, phosphono tri-carboxylic acids, polyphosphoric acids, polyol phosphate esters, amino phosphonates, maleic anhydride copolymers and acrylic polymers. Especially preferred scale inhibiting compounds include the phosphate ester of ethoxy-lated glycerine and glycerine phosphate esters.
Scale Inhibitor Dosage Useful dosages of the scale inhibitor range from 0.5 to 300 ppm by weight. These dosages are maintained in the recirculating cooling water by continually replenishing inhibitor lost to the sidestream softening and elsewhere. More preferred dosages of the scale inhibitors will range from 2 to 100 ppm and the most preferred dosages will range from 5 to 40 ppm.
Scale Inhibiting Compounds AMINO PHOSPHONIC ACIDS
Certain organophosphorous compounds including aminomethylene phosphonic acid, N-substituted aminomethylenephosphonic acids, and _ g 108160~

both N- and C-substituted aminomethylenephosphonic acids may be employed as scale inhibitors in the practice of our invention.
These compounds may be prepared according to the teaching of United States Patent 3,288,846.
Generally, these compounds can be characterized as con-taining at least one N-C-P linkage in their molecules, and as having the formula:

Rl~
N _ C -- P - O
R2 _R4 Oll Wherein R3 and R4 can be like or unlike, and are either hydrogen or ?
organic radicals; and Rl and R2 can be like or unlike, and can be hydrogen, hydroxyl, amino, organic radicals, or alkylene phosphonic acid radicals (such as that within the bracketed portion of Formula 1). Salts of the above compounds may also be employed. An espe-cially useful organophosphorous compound is aminotri (methylene-phosphonic acid and the potassium salt of hexamethylenediamine tetra (methylene phosphonic acid).
DIPHOSPHONIC ACIDS
Certain hydroxyalkane-l,l-diphosphonic acids such as those described in U.S. Re-28,553 have been found to be useful scale in-hibitors in the practice of our invention. These compounds may be obtained by reacting phosphorous acid with acid anhydrides and/or acid chlorides, especially those of acetic propionic, butyric, valeric and caproic acid. When both the anhydride and the chloride are used simultaneously, they must be derived from the same acid, e.g. the anhydride and the chloride of acetic acid can be used simultaneously, but not acetic anhydride together with propionic chloride. In lieu of phosphorous asid and one of the acid chlorides ~ -named above, phosphorus trichloride can be rsacted with one of the - 1 0 - "

. . , :

1~81604 carboxylic acids themselves. Particularly readily available are the reaction products of phosphorus acid with acetic anhydride, with acetyl chloride or with a mixture thereof. The reactions opportunely are carried out at elevated temperatures, preferably between 50 and 200C.
The acylation products of phosphorus acid, depending upon the process whereby they are manufactured, are obtained in pure form, but frequently in the form of mixtures. As has been ascer-tained by the reactions described above, all products obtained contain at least two phosphorus atoms in their molecules. Of the products whose constit~tion is established, the following represen-tative formula is of especial importance as a scale inhibitor in the practice of our invention:

f Rl 1 HO - Pl - C - IP - OH
OH OH OH
Wherein R denotes a low alkyl radical having 1 to 5 carbon atoms.
When mixtures are obtained, the products also have the above formula wherein the OH groups are partia]ly esterified. The acyl group, in that case, corresponds to the carboxylic acid component used in the reaction. Furthermore, two or more molecules of the above formula may convert into the corresponding intermolecule anhydrides while splitting off water and thus may be present together with the com-pound conforming to the formula given. Pure or refined compounds as ~ell as the above described mixtures can be employed. An es-pecially useful diphosphonic acid is l-hydroxyethylidene 1, l-diphos-phonic acid.
PHOSPHONOTRICARBOXYLIC ACIDS
Certain 2-phosphono-butane-1,2,4-tricarboxylic acids such as thosedescribed in U. S. Patent No. 3,886,205 can be advantageously ~08~604 employed as scale inhibitors in the practice of our invention.
These compounds are generally described by the formula:

R Rl 0 Cll - CH - CO - CH
HO ll ¦
P - C - CO - CH

HO

in which R is hydrogen, lower alkyl or a carboxyl group, and Rl is hydrogen or methyl, as well as their alkali metal, ammonium or amine salts, which compounds have been found to exhibit a strong complex-forming effect on alkaline earth metal ions. An especially effec-tive 2-phosphono-butane-1,2,4-tricarboxylic acid is represented by the formula:
: ' HO~¦¦ I ' :' P - C - CO - C~l ' HO

. .' ' .
POLYPHOSPHORIC ACIDS
Useful polyphosphoric acid scale inhibitors are disclosed in United States Patent No. 2,358,222. These include various poly-phosphoric acid compounds, particularly those which are soluble in water and still more particularly the alkali polyphosphates. These polyphosphoric acid compounds include pyrophosphates, the meta-phosphates and complex phosphates. The polyphosphates, such as the pyrophosphates (Na4P207 and Na2H2P207), the triphosphate ~Na P30lo~, the tetraphosphate (Na6P4013), the hexa~etaphosphate lC)B1604 (NaP03)6, and the complex phosphate (NagP7022), are genetically derived by molecular dehydration of orthophosphoric acid compounds, and are therefore sometimes called "molecularly dehydrated phos-phates." Of these polyphosphates, the polymetaphosphates are glassy and are therefore sometimes called "phosphate glasses."
The polyphosphates may be employed alone or with a protective compound, including the tannins, gelatin, starch, lignin which is a soluble lignin derivative. Lignin and tannin are organic dispersive agents which probably have dispersive effects upon calcium orthophosphate, and related materials: it is found that the efficiency of the combination is increased not simply by the additive effects of the two, but by their combined action.

~081~04 POLYOL PHOSPHATE ESTERS
Scale inhibiting compounds comprising certain phosphate esters of polyols containing one or more 2-hydroxyethyl groups and one or more of the group O O
Il 11 - O - P - OH and - 0 - P - O -OH OH
and salts thereof, are used to inhibit scale deposits by adding them to water containing scale-forming chemicals. These compounds and their preparation are described in more detail in United States Patent No. 3,462,365. Preferred such phosphate esters are the glycerine phosphate esters. Closely related to the above-described scale inhibitors are the phosphated mixed esters of non-surface active polyols containing at least one hydroxyethyl group and mono-hydric surface active compounds containing oxyethylene groups des-cribed in United States Patent No. 3,728,420. Preferred among these ~ -compounds are phosphated mixed esters of: (A) either oxyethylated or oxypropylated-terminally oxyethylated polyols, e.g. polyoxy-ethylated glycerol, ethylene glycol, hexylene glycol, sorbitol, mannitol or trimethylolpropane, or oxyethylated or oxypropylated-terminally oxyethylated erythritol, arabitol, xylitol, quercitol, inositol, and mono-, di-, or tri-pentaerythritol and (B) oxyalkyl-ated mono-hydroxy surface active compounds, e.g. oxyethylated nonyl phenol, oxyethylated tridecyl alcohol, and oxyethylated normal alcohol mixtures containing six or more carbon atoms, and (C) oxyalkylated dihydroxy surface active compounds, e.g. ethoxylated glycerine.
AMINO PHOSPHONATES
Certain amino phosphonates useful as scale inhibitors in the practice of our invention are described by the formula:

~8~04 R" - N
R' where R is o CH - P - OM
OM
R~ is R or --CH2CH2)H~ and R" is R~ --CH2cH20H~ or R
( 2)n where M is H, NH4, alkali metal, or combination thereof, and "n" is 1 to 6. These compounds are described in United States Patent No.
3,336,221. Preferably compounds from among those described in the general formula include (CH2)6N2ICH2PO~OM)2~4, N[CH2PO(OM)2]3 and 10(CH2CH20H)2NCH2PO(OM)2 where M is H, NH4, alkali metal or combina-tion thereof.
Other useful amino phosphonates are described in United States Patent No. 3,434,969. These are of the following general formula:

R ~ ~ ~ R ~
N - ~ CH2 - CH2 - N ~ R :
R n/

wherein each R is independently selected from the group consisting of hydrogen and ~0 - C~2 - P - OM
OM

provided, however, that at least half of the radicals represented by R are O

- CH - P - OM
OM

~0816V4 and n is an integer ef 2 to 14 and M indicates that the inhibitor is in water-soluble form. Typically, M will be independently selected from the group consisting of hydrogen, alkali metals, ammonium, alkaline earth metals, and zinc. Preferred scale`inhi-bitors of this group are N-methylene phosphonated diethylene triamines of ~he general formula:

X\ /X ~ -/ 2 2ICH2CH2N \ ~ -wherein each X is independently selected from the group consisting of hydrogen and o OH
and at least four of the radicals represented by X are - CH2 ~ OH
0~1 and water-soluble salts thereof.
~~LEIC ANHYDRIDE COPOLYMERS
Other useful scale inhibitors include polymeric compounds characterized by having two carboxylic acids adjacent to one an-other and spaced along the polymeric chain. A convenient way to prepare such polymers is to copolymerize maleic anhydride with ' other polymerizable mono-ethylenic compounds such as methyl vinyl ether, ethyl vinyl ether, styrene, alpha-methyl styrene, vinyl acetate, methyl methacrylate, isopentene, amylene, diisobutylene, isoheptene, nonene, dipentene, ethyl cinnamate or abietic acid. The resultant polymeric compound need not contain the free carboxylic acid groups.

~0~31604 A large number of copolymers of maleic anhydride and other monoethylenic compounds have been prepared but for the purpose of the present invention most effective copolymers are the reaction products of maleic anhydride with another polymerizable monoethylen-ic compounds in molecular ratios of 1:1 to 2:1. Although substan-tially equimolecular ratios may be used, a slight excess of maleic anhydride is preferred.

Yet another group of compounds considered useful as scale inhibitor in the practice of our invention are the water-soluble, acrylic polymers with molecular weights under 5000. These acrylic polymers include polyacrylic acid, polymethacrylic acid, and the alkali metal salts thereof, and mixtures and/or copolymers of such compounds. Such po]yacrylic and polymethacrylic acids, and mixtures and/or copolymers thereof may thus be defined as a water-soluble alkenecarboxylic acid polymer, the alkenecarboxylic radical of such acid having not more than 4 carbon atoms and not more than 3 of such carbon atoms from a straight chain. Preferably, the acrylic polymer used ~ill contain repeating units of the formula:
1CH3 1l R IC - C - O X
wherein R is H or CH3 and X is H, Na or K. This structure contem-plates the use therein of mixtures and/or copolymers of acrylyl and methacrylyl derived groups, as well as the partial or complete re-placement of the acid H's by Na or K or any combination thereof.
Sidestream Softening The present invention contemplates the use of lime-soda sidestream softening. Lime-soda softening is generally employed to precipitate and remove calcium and magnesium salts through the use of lime ~Ca~OH)2) and soda ash (Na2CO3). In the present application, 1~ !31604 a portion of the zero blowdown cooling tower recirculating water is continu-ously removed to form a sidestream that is subjected to lime-soda softening.
Lime-soda softening may be carried out at normal makeup tem-peratures ~"cold process") or it may be carried out at temperatures near or `!
above the boiling point of water ("warm process"). The chemistry of lime-soda softening entails the reduction of hydrated lime with carbonates and --bicarbonates of calcium and magnesium thereby forming the insoluble precipi-tates CaC03 and which form a sedimentation sludge that may be conveniently collected and removed from the softening apparatus. Hydrated lime also reacts with non-carbonate magnesium salts to form insoluble Mg(OH)2 which is removed by sedimentation. Where it is desirable to selectively reduce total calcium ion concentrationJ lime softening is used to reduce calcium carbonate hardness. System parameters may be adjusted in order to selec-tively increase calcium removal.
Although many known water softening systems may be applied in the practice of our invention, we have found that warm process softening is a very desirable softening system in zero blowdown cooling tower systems.
The apparatus employed in warm process softening generally entails a source of heat, a sedimentation tank, feeding means for introducing lime, soda ash and coagulants and sludge collecting and rcmoval means. Common retention times in such units are about one hour over all. Warm process lime-soda softener is used to remove or control the levels of calcium and magnesium salts, to remove silica, to deaerate the makeup water and to remove suspended matter.
It is contemplated that other softening processes may be sub-stituted for lime-soda softening or used in conjunction therewith. These include caustic softening, aluminate lime-soda softening, zeolite softening, ion exchange, reverse osmosis and others. The usefulness of these systems will be a function of: 1) the nature of the makeup water, 2) the efficiency of the treatment, and 3) the economics of the overall softening system. Sodium aluminate-lime-soda softening, for example, may be used in conjunction with lime-soda softening for treating very high sulfate water.
Other Cooling Tower Treatments Although the present invention is concerned with scale pre-vention in zero blowdown cooling tower systems, it must be recognized that these systems will experience problems similar to those found in conventional bleed off cooling tower systems, in-cluding corrosion of metals found in the cooling tower system, microbiological growth and other problems. Although, generally these "non-scaling-type" problems may be dealt with in the tradi-tional manner, it will be necessary to determine the compatibility of the corrosion inhibitors, microbiocides and other treatments with the treatments employed in the practice of our method. For example, we have found that chromate-zinc corrosion inhibitors do not interfere with the operation of our method whereas phosphate based corrosion inhibitors tend to inhibit the softening process.
Ne have also found that when sodium hypochlorite is used as a bio-cide, it ~ill not significantly interfere with the practice of our invention. Of course, it will be possible to determine a broader class of acceptable corrosion inhibitors, microbiocides and other treatments which are compatible with our scale inhibition method.
Determination of Minimum Sidestream ~ Scale Inhibitor Requirements A systematic analysis of evaporative cooling systems oper-ating with limited water discharge using sidestream softening to control the levels of scale-forming species was generated for use in the practice of our invention. A schematic representation of the sy~tem contemplated appears in Figure 1.

.

~0~1604 In this figure, the symbols used have the following meanings:
E = Evaporation rate - determined by heat load.
M = Makeup rate - defined by evaporation rate and system losses.
B = Blowdown, drift and incidental losses - partially determined bysystem characteristics but may be augmented as dictated by waste treatment faciIities and system characteristics.
Si = Sidestream rate = So with minimal softener blowdown.
Sc = Rate of addition of softening chemicals.
The size of stream E, the evaporation rate, is generally fixed for a given situation, since it is determined by the heat load. If the total losses of liquid water from the system ~B) are small compared to E, the makeup stream (M) will be approximately equal to E. The minimum size of the loss stream (B) is fixed by the drift, softener blowdown and incidental losses by the system.
It may be desirable, depending on the quality o the makeup stream, to raise B above this value.
The size of the sidestream is a major independent variable in the design of a zero blowdown system. For reasons of economy, it is desirable to make this stream as small as possible. However, it is this variable, coupled with the makeup rate, the makeup com-position, and the degree to which a given species can be removed in the softener which determine the steady state level of a species in the system. If this species can contribute to scale or deposi-tion, its steady state level becomes highly important with regard to the reliability of operation. Therefore, the three chemical parameters which determine the minimum size of the sidestream rela-tive to the makeup rate are:

108~1604 A) the composition of the makeup water;
B) the level to which potential scalan~s can be softened; and, C) the maximum level of the scalant which can be tolerated in the cooling system without serious maintenance problems.
The following example will illustrate these points. A species which is contained in the makeup water in a concentration A can be softened under the projected operating conditions to the level As~ A pilot cooling tower study has shown that this species can be tolerated in the projected recirculating water at the concentration Amax without scaling. In order to calculate the minimum softener stream necessary to limit species A to safe levels, the following exercise is done:
At Steady State Ain = AoUt Ain = M x A + SO x As AoUt = Si x Amax + B x Amax Assuming So = Si and M = B
Ain = M x A + So x As AoUt = So X AmaX
So = M x A
Amax - As

2~ This calculation should be done for all softenable species in the makeup water. The species which generates the highest value of So represents the cons,traint against which So can be minimized. The steady state concen-trations of non~limiting softenable ions can be determined by a similar exercise which yields the relationship.

AsS = A -S M + As ~)816(~4~

When the steady state concentrations of each non-limiting softenable species has been determined, the appropriate dosage of softening chemicals can be determined by standard softening form-ulas.
Non-softenable species concentrations can be worked out by similar methods. While these concentrations do not directly enter into the analysis of system design and operation, they exert an important secondary effect. Non-softenable ions, particularly sulfate, can have major effects both on some softening reactions and on the maximum tolerable cooling system concentration of some species.
Particularly influenced by this effect are the calcium carbonate maximum (as reflected by the alkalinity limit) and the reduction of calcium and alkalinity by lime-soda softening. There-fore, it is desirable to know the concentrations of these species at which the system will equilibrate. If these levels are exces-sive in terms of the inhibition of softening reactions, it may be economical to provide for an increase in the blowdown.
The non-softenable ion concentrations (N) may be calculat-ed as follows:
By mass balance Nin = N ut NoUt = B x N s N = N k x M + (Nso - Nsi) So t Ntreatment trt So:
Nm x M + (Nso ~ Nsi) So t t Using this algorithim ~which is similar to the more simplified method used for softenable ions) one may calculate the steady state concentrations of all non-consumed species in the system.

1(~81604 Chemical treatment dosages are determined in a similar way. A
chemical residual (CsS) of a material which is reduced to concentration Cs by the sidestream softener and suffers no other attrition in the system requires chemical feed such that:
F = SOX (Css - Cs) + B x Css A material which undergoes degradation other than removal by the softening process will, of course, not obey this relationship.
The systematic analysis of a zero blowdown problem consists of the application of the simple mass balances described above in a logical man-ner. As each problem tends to be unique, there is no guaranteed "set" for-mula which may be applied. Often the best approach is iterative, that is, the application or a set of approximations, checking and refining them and using the improved approximation to redo the calculation until the approxi-mation is sufficiently accurate.
EXAMPLES
Example 1 Determinations of the solubility limits of scaling species were done using non-sidestream softened pilot cooling towers at slightly acid pH. These experiments used synthetic makeup waters with conductivity-controlled blowdown. The composition of the makeup was such that the re-circulating water would have the desired composition upon concentration to the estimated solubility maximum. In cases where these estimates were inaccurate, the experiments were repeated with an adjustment in the makeup water composition.
The main reason for using the regular cooling towers for these experiments rather than the sidestream softened unit was the difficulty of detecting chemical imablance in the softened unit. The composition of the water in the regular units was also more easily controlled.

.~ . ., The laboratory tests of solubility limits of scaling species were carried out in laboratory pilot cooling towers such as that described in a paper by D. T. Reed and R. Nass entitled "Small-Scale Short-Term Methods of Evaluating Cooling Tower Treatments...Are They Worthwhile?" presented at the 36th Annual Meeting of the INTERNATIONAL WATER CONFERENCE, Pittsburgh, Pennsylvania, November 4-6, 1975, available from Nalco Chemical Company -Reprint 233.
Calcium Carbonate -The tendency of a system to scale with calcium carbonate under alkaline operation has been shown to be highly dependent on the total alka-linity. The effect of high levels of sulfate, chloride and silica on the ability of a scale inhibitor containing the phosphate ester of ethoxylated glycerine to stabilize calcium carbonate was determined by observing the effect of various levels of these species on the alkalinity limit. Chicago tap water was enriched with various amounts of these materials and the alka-linity limit determined by studying chemical imbalance in the tower water.
Calcium Sulfate Scaling The operating constraints imposed by calcium sulfate solubility limits were determined in pilot cooling tower tests which were run at 100F.
pH was maintained at 6.5 to 7.0 in order to avoid thc high levels of calcium carbonate which would otherwise be obtained upon concentration of alkaline recirculating water. The makeup water used in the tests included Chicago tap water containing varying amounts of sodium sulfate (0, 350 ppm, 1200 ppm, 1800 ppm and +3000 ppm). The concentration of the recirculating water was slowly increased using conductivity control of the blowdown. The solubility limit of calcium sulfate was indicated by a decrease in the concentration ratio of calcium relative to that of the system. The data generated appears in Table I and in Figure 2.

..... . ~ ~

~.~)8~604 TABLE I

CALCIUM SULFATE -- MAXIMUM CONCENTRATIONS

Makeup Added Test Run Ca* S04**Naaso4 Ca x S04 X 106 1 1060 2~10~ 0 2~2 2 1100 2~750 0 3~0

3 92033 ~ 5003 ~ 000 30 ~ 8

4 90031 ~ 5003 r 000 28 ~ 4 900 5~550 350 5~0 6 8705r 650 350 ~ 9 7 92023r5001~800 ~1~6 8 86019 ~ 8001 ~ 800 17 r 0 9 9S018~0001~200 17~1 93015~8001~200 14~7 * as CaC03 ** as NaSO4 Examination of this data indicates that with sulfate concentrations below 30,000 ppm (Na2S04) a calcium limit of 800 ppm (CaCO3) will preclude scaling by calcium sulfate. It is well known that the presence of non-common ions such as magnesium and chloride can substantially increase the solubility of calcium sulfate.
Since, the makeup water used in these experiments was not enriched in these materials ~as a zero blowdown recirculating water would be), the calcium limits indicated by these results can be consid-ered minimum values. Minimum calcium limits in actual zero blow-down systems would commonly be at least 1000 ppm. Of course, due to the wide possible variation in makeup waters cmd other operating parameters, calcium limits below 800 ppm might be encountered.
Therefore, a very safe calcium limit would be 500 ppm. Calcium maximums will be determined on a case-by-case basis through the application of techniques well known to those skilled in the art.
Example 2 Jar tests were run to determine the feasibility of lime-soda softening of highly concentrated cooling tower water.
Specifically, the relationship of sulfate and chloride ion levels to the efficiency of softening was studied. The results of the experiments indicated that, as sulfate and chloride levels in-creased, the efficiency of calcium removal decreased; high solids concentrations affected the removal of calcium more than they affected the removal of magnesium. Nevertheless, the results show-ed that as the total dissolved solids increased, the water became only slightly more difficult to soften. Finally, these tests sug-gested that, in very high solids waters, changes in the ability of the lime-soda softening system to soften may not be specific to an ion but rather may be generally dependent upon total dissolved solids.

1o8l6o~

Five simulated cooling tower blowdown waters were examined in these softening tests. Makeup of each water is given in Table II. All five of the waters had the same degree of hardness and alkalinit~, but differed in ~he amounts of sulfate and chloride.

~V~604 TABLE I I

PPM AS CaCO3 Water No. CA 2 Mg 2 OH CO3 HC03 Cl 2 608 468 0 0 2541,000470 3 515 485 0 0 7228035,Q00 4 575 515 0 0 7528,00035,000 600 490 0 0 662~80~4,900 108~604 Each of the five waters listed in Table II were treated with 90 mg/l of a corrosion inhibitor containing chromate, phosphate and zinc; each had an initial pll between 6 and 7 which became alka-line after treatment with water softening chemicals. We found that the phosphate based corrosion inhibitor reduced the effectiveness of the lime-soda softening: higher dosages of the corrosion inhibitor corresponded to poorer softening.
Dosages for each water were calculated to meet the require-ments for: (a) select calcium removal, (b) complete softening and (c) excess lime softening. Also, water No. 3 was treated for select calcium removal plus 20% magnesium removal, and select calcium re-moval plus 4~% magnesium removal.
The jar tests were carried out by adding the appropriate dosage to each sample at the beginning of an initial five minute mix which was run at 100 RPM, followed by a thirty minute mix at 50 RPM and fifteen minutes for settling. A sample from each jar was then filtered and analyzed for total hardness (TH), calcium hardness ~CaH), phenolphthalein alkalinity (P), total alkalinity ~M) and pH. Selected samples were also analyzed for CrO4 , PO4 , Zn 2, Cl and S04 . The results of these tests are summarized in Table III.

108~604 _l ~1 ~
~ ~ , <~ U~ o , o ~ ~ , o~ o o~ ~ , I~ ~ o .,, t, ~.
CY
I , . . , I
\ ~ , ~ ~ o ~ ~ ~ O

N ~D U~ I ~ O O a~ ~1 O 1~
~,1 `D C~) O O I a, o o ~ o ~1 ~1 ~o 1` o o o ~ h O ~ O O O O O
h ~

O O O O o I~ O o~ oo o ~1 N O t~l N O

H ~ 1~
Z O O O O O
U~ ll ~O 1~ 0 0 O
H E--~_ :t ~D ~ O O O
H O O
¢ ~ t~
¢ O ~ d' ~D 1~ Il~ ~ I~ I~ N N ~1 ~O 1~ 0 Il~
E--'V~ Ul ~ ~ N N cn 1~ cn 1~ Il') O Ol 1~ ~ O N ~ O 00 0 H _~ O t~ N O ~ t~ N O 00 t ~ O O O 1/~ 0 CO 0 :1 _I _I N ~I r-l N N _I N
. , .
¦ o ~ ~ ~ `D O11~~1 0 0 N N

¦ o ul ~ 0~ 00 O ~ ~ Ll~ N 11~ 0 0 U~ 11~ 11~ ~ a'l o ~ ~ _~ o a~ D ~tOtl N N ~I t~ t~ Ln ~) C~11~ H ~D N ~H _I~t \D 11) d' Ir~ ~ _I _I
_ tl ~Cl~ O IJ~ N ~1 1~ 1~ 0t~ N ~ t~) 1~) tt) 00 ~--1 O O ~t O O
Z

~> ~`
,_ OOOO OOOO OOOO OOOO OOOO
~d O'J O N N ~ ~IJ7 ~I t~) ~D Lr~ Lt~ `.D ~t ~0 ~ ~_el ~t t ~ I N ~t ~t E-~ ~
h E- Z
1 .-1 N 1 O
.~1 O
o ~ - 30 -.
- ,: ., , ~0~1~04 ~: I , ~ o , oo ~ ~ ` ~ ~ _, U~
o\o ~d ~, ~r P~ ~ Crl ~ O~ O O O O O O O O

O O
r~
tn ,~
h ,1 O
h~
,_ ~ ~ I O
~ ~ O
~0 ~ oo C~ N
H,_~ ¦ g H E-- H O a~
H O ~ V~ ~t U~ U~ ~ . ' ¢ ;~ V~ ~ _I ~ ~ N 0~ N O 0~ ~ ~ 11 E- O ,1 ~ o o _~ o ~
V~ I N N N H ~ I H
H ¢ O CJ~ 00 1~ N ~ 1~ _I O ~ O

O N ~ ~J ~ ~ O N ~0 N O --I
~:C cr 00 ~ 1~ 00 1~ ~o 1~ N t~ O _I
O ~ ~ ~ ~1 ~ ~
_~ .
~ 1 o ~ ~ a~
O N H N 0~ ~ 00 01 11~ It~ ~t ~ ~D N N N

_l O OOOOOOOOOOOO
C~ ~ I~ 1~ CO 00 ~1 ~1 a~ a~ N N

~! z ." ." u~ a o o o o o ~ -I
~:: ::::

~d O Lt~ I~ Ln ~1 ~ t~ t" ~1 ~ t~) t E--~_ h E- zo $~ ~ ~ ', ~1 ~ ' .
o 3 .~

- 30a -.. . .

- :

1~816()4 The relative efficiencies of the various softening combinations in relation to the makeup water being treated were then determined by comparing the percent of calcium hardness and sulfate hardness. It was found that calcium removal became more difficult as total dissolved solids increase, as indicated in Table IV.

10~1604 TABLE IV

As CaC03 (ppm) ~% Residual -Water[so47 [Cl~ Slope ~ppm Soda*
1 660 270 5.55 x 10 2 2 470 41,000 7.73 x 10-2 335,000 280 15.3 x 10-2 435,000 28,000 14.8 x 10-2

5 4,900 2,800 5.70 x 10-2 *that is the slope of a graph showing soda ash concentration vs. determined percent residual hardness as in Table III.

108~604 Nevertheless, the degree of calcium removal necessary to achieve the very lenient Ca levels suggested for the practice of our invention (1000 ppm, 800 ppm or 500 ppm) readily could be attained.
Example 3 Jar tests similar to those described in Example 2 were run for aluminate-lime-soda softening of the cooling tower waters. In several tests, lime was added to these cooling tower waters as a source of calcium in order to determine the effects of aluminate-lime-soda softening on sulfate removal.
These tests were conducted in two stages. Stage 1 entailed the addition of aluminate and lime at the beginning of a five minute rapid mix; sodium hydroxide was added at the end of the mix for pH
control. The cycles of stage 1 were as follows: five minutes at 100 RPM (rapid mix), 90 minutes at 50 RPM, and fifteen minutes of settling.
Following stage 1, the jar contents were subjected to vacuum filtration, the filtrate was recarbonated and the jar contents were again filtered.
Analysis was carried out on some samples after stage 1 and on other samples analysis was done after stage 2. The data obtained in these experiments appears in Table V.

~,`f, :`, ~08~604 .--1 ~ I O O I oO 00 I Gl CJ~ I I I ~ I ~D I e:t I
~d O I CO ~ I r~ I` I ~0 ~D I I I N I ~ D r-l :~ U) .,1 ~) N O
~ ~I O ~ ~ , I ~ O O
o\

~ h O O
Il~ ~rl h ~rl , , , , , , , , , , , , , O
h ,C
O ~
H
a) o cr~ ~ ~ a I ~ o I ~ c~ I I I I o o o~ oo O ~ ~ O O O O cn ~ o o o o u~
~ N N N N O O 00 ~ ~ 1~ N I ~1 --I O O N
V~ ~ N et N
n 0~ ~CI ~In 1~N C~ ~0 O IJ'~ O
I ~ t'~ DI 00 0 NI 1~ I N O

~:) .--1 N cn 1~N ~1 ';t 0) ~ O U~ O
Z~1I CO I OI ~~~ I 00 I N O
1--1 ~ I t~I t'~ I N I t~ N ~ I O I H
Z ~
~v~ o o o o u~ u~O O In o o a~ ~ o u~ O n I
O ~ N N ~ N O ~N 0tl In 1~ O) N ~ N
~ ¢ ~ t~
;L~ U~ ¢ ~
c4 ~ ~o o o o u~ u~O O In o o ~ ~ u~ ~ u~ ~ ~ , ¢ 5 L/~ Ol H 1.~ ~ 00 U~ C~) N H ~ _I ~ N :
E--' H ¦N N CJI N O ~ N 0~ N ~ _I Lr~ Ln ~ -Z¢t~ . '.
~ O 00 0 0 0 00 0 0 a~ ~, , ,, ,, , O O , ,, ~ ~d , , , I I I ~ co ¢ Z

6 o a~ a~

H N ~1 O O O O I CJ~ O I O~ 0 1 1 1 0 1 0 tll .' ' ' ., .
~ ~ o ~ ¦ ~ ~ " o u~ o u o ~ Z~- I I I I I
t~ ~`
h ~_ O 1-~ 0 0 0 d O m O O OO O
E- ~o Io Icn I o~ I I o o o O~ I~ IN I N I
tl~ ~_l O
~ ~ .
o lo ~o ~o cn o o o o o o o u~ I_~ I~D I 1` 1 1 n In U~
Z
C~
01O H HO H H O H H O H ¢ C~ O H O ¢
H H HH 1~ 1 H
H H
U~

~ - 34 -1()~3~604 cd O I CJ> I O~ ~
U~

~1 1 1`
cd I ~ I
o\o ~, g h 0~ o I o I I
~'~ ~n I I I

1 ' ~ ' ooooo o ~ o o el- 1~
O U~
t~ t~ N
I O I Lr~ O
E~
~ _I ~) C~ U~ U~ O
_~ I ~ I U~ U) ~1 ~ I 0~ I a N
~~1 oo .
O .~1 v~ ~ o L~ o n ~~ E-~ C~
O ~ t~
U) ~ ~ .
¢~ ooooo `' '' ~_1 ~1 ~ `D et ~
U~
H O O O

:~
o~ O O
C~ O l l l ~ ~ 1 11 , o - 1 u~ o L
~d h ,_O o o O o E- Oo o o ~ -~ ~ lo o~ o o o o o : :
_~~D ~ ~D
~¢d N
Z ~D " ' w~lo H O ¢ ~
t~ ¦H H
V~
- 34a -Analysis of the data obtained in the aluminate-lime-soda softening tests indicate that in this softening method, sulfate and calcium are removed as follows;
A) where Ca 2 ~ S04 2 the removal of Ca 2 non-carbonate hardness is good; and, B) where Ca 2 C S0~ 2 the efficiency of sulfate removal improves with increased sulfate concentrations.
Where the sulfate concentration is greater than the calcium concentra-tion, lime could be used as a source of calcium to precipitate the sul-fate. Aluminate-lime-soda softening may have use as a partial treatment for very high sulfate waters. At present, it should be noted that the use of aluminate-lime-soda softening as the sole sidestream softening method is a zero blowdown system would be prohibitively expensive. How-ever, partial sulfate removal with resultant improved efficiency of sulfate removal per dollar of treatment may be practical.
Example 4 The experiments described in this example were carried out in order to determine the effect of various scale inhibitors on sidestream softening reactions. Two types of experiments were carried out. The first entailed jar tests similar to those described in Example 2. These tests were used on synthetic cooling tower waters to predict the inter-action of softening reactions with various softening programs.
The second type of experiment carried out entailed the opera-tion of a pilot cooling tower such as that described in the above-referenced paper of Reed and Nass, and which contained sidestream lime-soda softening apparatus. The sidestream was drawn from the basin of the unit by a variable rate Masterflex* pump into the first chamber of the horizontal clarifier.

*Trade Mark .. ~ .
.

10~1~04 There the water was rapidly mixed by an electric mixer with lime, soda and clarification chemicals. The clarification chemicals were added by a Harvard Apparatus syringe pump. The lime and soda ash (as a slurry) were added by a fixed speed Masterflex* pump. The lime soda slurry was stored in a stirred 800 ml jar. The water overflowed a weir and entered the second chamber where it was slowly agitated to provide coagulation. After over-n owing a second weir and a baffle, the water was allowed to go into a set-tling chamber where the precipitate fell to the bottom. The precipitate was intentionally allowed to build up in this chamber for extensive solids contact.
After this chamber, the water entered a final chamber where it was recarbonated or pH adjusted by sulfuric acid. From there the softened water flowed by gravity back into the basin of the unit.
The dosage of lime and soda applied was varied by changing the composition of the slurry or by using a different fixed-speed Masterflex.
Treatment chemicals were added to the system on a constant basis or in slugs. Daily analysis was done on both the recirculating water and the softened sidestream.
A number of scale inhibitors were tested in the pilot zero blow-down system just described. These included the phosphate ester of ethoxy-lated glycerine, potassium hexamethylenediaminotetramethylene phosphate esters, hydroxyethylidine diphosphonic acid and aminotrimethylene phosphonic acid, in dosages sufficient to prevent scale formation.
We found that all of these scale inhibitors inhibited softening reactlons to varying extents. Due to the constant state of flux of the chemistry of the recirculating water, direct comparison of the inhibitors were difficult to make. However, we did find that the degree of softening inhibition by these materials was related to their activity as scale in-hibitors.

*Trade Mark ~' .. .... ~ ~

( 1081~i04 Jar tests were then run in order to further explore the extent of softening inhibition. Jar tests were done on potassium hexamethylenediaminotetramethylene phosphate, hydroxyethylidine diphosphonic acid and aminotrimethylene phosphonic acid at levels which would give roughly equivalent activity. The inhibitor do-sages were roughly what would be required to stabilize the given water in a normal cooling tower environment. The composi.tion of the synthetic cooling water is set forth below Ca = 700 ppm Mg = 700 ppm Alk - 275 ppm sio2 = 95 ppm :
" S04 = 15,000 ppm Cl- = 10,000 ppm .
CrO4 = 60 ppm The dosages were as follows:

316(~4 ~ABLE VI

Inhibitors Dosage Phosphate ester of ethoxylated glycerine22 ppm Glycerine phosphate esters 22 ppm Aminotrimethylene phosphonic acid 4 ppm Potassium hexamethylenediamino- 15 ppm ,tetramethylene phosphonate Hydroxyethylidine diphosphonic acid 3 ppm :

~ ~ ( lV~31604 The basic water was softened at 100F using ;ime and soda_ Residual hardnesses and alkalinities were determined by standard techniques. The results, which appear in Table VII verify the ¦ hypothesis that the degree of softener inhibition varies with the scale inhibitor activity since the inhibitors gave nearly equal interferences.

lV1~16i04 ~1 o u~ ~ o ~ P- ~ P~
CO U~
~ V ~ o a~
_ U
~d U
sz E~ ~ ~r o oo ., o ~ ~ .

H ~ ~ t~ ~ ~ ~ ~1 ~
~ ~ Id OD CD ~
l ~n o o o o o 14 E-l ~ ,1 ~1 ~1 -I 'I .

~ .~ . .

H C.) ~
E~ ~ ,1 H ~J ~ ~>
o O~
P,~ ~ ~ ~ O
~i a) ~
~J ~ rl O
~ O ~ rî ~ h X ~ (n ~ ~ a~
O ~ O C
O U~
~0 o a) a a) ,~
~ rlX ~: O

O
rl ~rl X ~4 0 0 U~
,~ ~ ul Fl h t~ ~ U
.C '~ O rl ~ l) ~ :>~
~: '-I ~ El ~ O a~
H m .

~0~ 4 Example 5 The next question which was examined concerned the removal of scale inhibitors by lime softening. This factor is critical to determining the proper inhibitor application in a zero blowdown syStem since it constitutes the primary mechanism for removal of these chemicals. This softener removal effect was studied in both jar tests and in a sidestream pilot cooling tower unit. The jar tests provided a batch wise evaluation of the effects of the softening reactions while the pilot unit with its continuous re-cycle, concentration and softening, gave useful chemical consump-tion information.
On the basis of jar testing, removal of the scale inhibitor based upon the phosphate ester of ethoxylated glycerine was found to vary between 33 and 65%. In the pilot testing system, the degree of absorption of this scale inhibitor was also found to be highly variable. This variation in scale inhibitor removal makes application of this scale inhibitor rather difficult to determine.
Hence, it will be necessary to replenish removed inhibitor at a rate equal to the highest predicted removal rate.
Other scale inhibitors were tested in the pilot unit.
These included a glycerine phosphate ester, potassium hexamethylene-diaminotetramethylene phosphonate, aminotrimethylene phcsphonic acid and hydroxyethylidine diphosphonic acid. The glycerine phos-phate ester exhibited a variation of removal somewhat similar to that shown by the phsophate ester of ethoxylated glycerine, tho~gh the effect was much less pronounced. The initial removal of or-ganic phosphate was 50 to 75% of the recirculating residual. This fell off somewhat as the test progressed over 50 days.
Potassium hexamethylenediaminotetramethylene phosphonate was very similar to phosphate ester of ethoxylated glycerine in both inhibition of softening and initial softener attrition.

~lowever, the rate of removal across the softener remained far moreconstant at 40 to 60% than did either phosphate ester of ethoxylated glycerine.
Aminotrimethylene phosphonic acid and hydroxyethylidine di-phosphonic acid were found to be quite similar in behavior. N0ither exhibited build-up as did phosphate ester of ethoxylated glycerine.
Both gave severe inhibition of softening at very high levels.
Exam~e 6 A number of experiments were run to demonstrate the practical feasibility of the operation of a cooling tower system using scale inhibiting chemicals in conjunction with sidestream softening and cor-rosion inhibitors. The apparatus employed was the pilot cooling tower system referenced at lines 1 to 7, page 24.
(a) This experiment entailed the use of sodium hydroxide as the softening chemical to convert bicarbonate into carbonate. The very high softener effluent alkalinity was reduced by a continuous HCl feed and recarbonation. The scale inhibitor employed was the phosphate ester of ethoxylated glycerine. A chromate based corrosion inhibitor was also used. Scale prevention was found to be adequate and variable re-moval of the scale inhibitor was noted.
(b) This experiment was similar to (a) except that lime slurry was used in place of sodium hydroxide as the primary softening chemical.
Acid feed was not necessary to reduce alkalinity. Scale prevention was satisfactory.
(c) This experiment parallelled experiment (b) except that the scale inhibitor used was potassium hexamethylenediaminotetramethy-lene phosphonate. Neither acid nor recarbonation were required for treatment of the softener sidestream return. Scaling of the cooling tower tubes was noted, but this was believed to be due to system mal-functions.

~ ( 10~1604 ( (d) This experiment was similar to experirnents (b) and ~c~
except that the scale inhibitor used was glycerine phosphate ester.
Very little deposit on the cooling tower tubes was noted.
(e) The makeup water in this experiment was obtained in Southern Cali~ornia. The scale inhibitor employed was a glycerine phosphate ester and the corrosion inhibitor was chromate. Since there was insufficient alkalinity in the makeup water to allow lime softening to give sufficient hardness removal, supplemental soda ash was added to the softener. The sidestream was sized at 22~ of the evaporation rate in order to limit the alkalinity to 300 ppm.
Although there was minor silica scale formation, no s:ignificant calciurn carbonate scaling problems were observed.
,- (f) Further experiments were run with aminotrimethylene phosphonic acid, hydroxyethylidine diphosphonic acid and 2-phos -phonotabutane 1, 2, 4, tricarboxylic acid sodium salt as the scale inhibitors. These inhibitors were found to function ade~uately r although replenishment was necessary in order to co~pensate for softener removal.

We Clairn:

Claims (8)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. In an open aqueous alkaline recirculating cooling tower system operating with zero blowdown, a method for controlling the level of scale forming species in the recirculating cooling tower water comprising:
A) maintaining in the cooling tower water from 0.5 to 300 ppm by weight of a scale inhibitor; and B) continuously removing a portion of the cooling tower water to form a sidestream, treating the sidestream in a softening process to remove at least a portion of the softenable species thereby forming a softened sidestream and a sedimentation sludge; and C) returning the softened sidestream to the cooling tower water;
and D) collecting and disposing of the sedimentation sludge.

2. The method of Claim 1 wherein the softening process is lime-soda softening.

3. The method of Claim 2 wherein the lime-soda softening process is adjusted to maintain the level of Ca++ in the cooling tower water be-low 1000 ppm.

4. The method of Claim 2 wherein the lime-soda softening process is adjusted to maintain the level of Ca++ in the cooling tower water be-low 800 ppm.

5. The method of Claim 2 wherein the lime-soda softening process is adjusted to maintain the level of Ca++ in the cooling tower water be-low 500 ppm.

6. The method of Claim 1 or 2 wherein the scale inhibitor is chosen from the group consisting of amino phosphonic acids, diphosphonic acids, phosphono tricarboxylic acids, polyphosphonic acids, polyol phosphate esters, amino phosphonates, maleic anhydride copolymers and acrylic polymers.

7. The method of Claim 1 or 2 wherein the scale inhibitor is a phosphate ester of ethoxylated glycerine.

8. The method of Claim 1 or 2 wherein the scale inhibitor is main-tained in the cooling tower water in a concentration from 10 to 200 ppm.