Desalination, 93 (1993) 287-3 10 Elsevier Science Publishers B.V., Amsterdam

287

Scale formation and fouling problems effect on the performance of MSF and RO desalination plants in Saudi Arabia

Malik Al-Ahmad and Farag Abdul Aleem

chemical Engineering Department, College of Engineering, King Saud University, PO Box 800, Riyadh 11421 (Saudia Arabia)

SUMMARY

Most people working in water desalination recognize that scale formation and fouling problems are facts of their practical life. Seawater and brackish water always have the tendency for scale formation and fouling problems due to dissolved salts and finely suspended solids. Elaborate models are available to simulate these scale and fouling problems to help the design engineer to predict the effects of such problems on the performance of the desalination plants. From the practical point of view, main types of scales in MSF plants are CaCO,, Mg(OH), and CaS04 while in the RO plants biofouling of the membranes is a major problem. The present paper investi- gates the various models and mechanisms of scale formation and biofouling processes in water desalination plants and discusses proper pretreatment methods and scale control techniques to minimize the drastic effects on the performance of desalination plants in Saudia Arabia. The interaction between scale formation, biofouling and corrosion problems in both MSF and RO desalination plants are discussed and recommendations are given for both design engineers and operators for overcoming such scale and fouling problems in desalination plants.

OOll-9164/93/$06.00 0 1993 Elsevier Science Publishers B.V. All rights reserved.

288

INTRODUCTION

The main technologies used for desalination are thermal processes, reverse osmosis and electrodialysis. All suffer from scale formation and fouling problems.

In a typical desalination plant, about 40% of the heat transfer area is provided to allow for fouling and scale problems which is equivalent to about 10% increase of the whole capital cost of the desalination plant. Furthermore, in the fouled condition the plant usually needs between lo- 15 % increase in the fuel consumption compared to its clean conditions [ 11.

Seawater is a chemical system with considerable complexity, containing the scale-forming ion types shown in Table I.

TABLE I

Main seawater constituents (mg/l)

Sodium Na+ Magnesium Mg+ + Calcium Ca+ + Potassium K+ Chloride Cl- Sulfate Sob-- Bicarbonate HCOs- Bromide Br- Other Solids TDS Density (20°C) Water

10561 1272

380 18980 142 142 65 34 34483 1.0243 965517

Total hardness: = 1952 mg/L CaC03

Certain combinations of these ions form components which have a low solubility in water. Once the solubility has been exceeded, the compound precipitates as a solid. These precipitated solids may either remain in suspen- sion in the water or form a coherent scale on the surface of the equipment.

SCALING TENDENCIES OF SEAWATER AND MAIN TYPES OF SCALES IN

DESALINATION PLANTS

The salt content of seawater is of critical importance in the determination of its scaling tendency. Wherever the maximum solubility of a certain salt has been surpassed, its scale deposits will eventually form.

289

The three main scale forming constituents of seawater are calcium bicarbonate, magnesium salts and calcium sulphates [2]. Calcium bicarbon- ate is present in seawater in concentration from about (1 lo-140 ppm) (as CaCOs) (3). On heating above about 5O“C calcium carbonate is formed, which can precipitate and carbondioxide is liberated.

Above about !Xl°C, the carbonate ions can hydrolyse to hydroxyl ions which can combine with magnesium ions to produce magnesium hydroxide scale. Thus, the calcium carbonate and magnesium hydroxide are known in practice as the alkaline scales. The following alkaline scale forming reac- tions occur upon heating seawater [1,2]:

2 HCO, - CO, + CO, + H,O

CO, + H,O - 2OH- + CO,

Mg++ + 20H- - Mg(OH), (alkaline scale)

Ca++ + co, - CaCOs (alkaline scale)

Whether CaCO, or Mg(OH), or both are formed depends on the opera- tional parameters of the plant. Higher temperatures and higher pH values favor the formation of Mg(OH), [ 1,2]. Solubility calculations should be used to predict the scaling tendency of the seawater processed. For example, a method suggested by Stiff and Davis is used for the prediction of CaCO, scale from certain saline water. A brief description of this method is given here below.

Calcium carbonate scaling index [4,5]

s1 = PH*,,al - PHsatumtion PH saturation = K + PCs + PALK

where:

SI = scaling index pH = the usual hydrogen ion concentration PCs = Ca++ ions concentration expressed as pH form PALK = alkalinity of saline water expressed as pH form K = constant depends on the ionic strength of water

It is also noticed that the scaling tendency of water due to CaCOs increases with increasing temperature, pH and by decreasing the partial pressure of CO, in its solution.

290

4

\

t

3

CF 2

1

0 c 20 40 60 80 100 120 140 160

tn1 -

Fig. 1. Solubility of different forms of calcium sulfate in sea water.

1.2 h I I I

0’ 1.0 cn 0

; 0.8

0

0 80 120 160

Temperature [“Cl

200 240

Fig. 2. Solubility of gypsum (CaSOJ, or-8-hemihydrate and anhydrite in pure water.

The other important type of scale in sea water is the sulphate scale (namely, calcium sulphate). Concerning, this scale, one can notice that there are three chemical formulae of calcium sulphate which can be formed in sea water, i.e. CaS04 (anhydrite), CaSO,.1/2 Hz0 (hemihydrate) and CaS04. 2H,O (dihydrate). The sea water solubility of these compounds as a function

291

of temperature is given in Fig. 1 [2] where as their solubilities in pure water are given in Fig. 2 [2].

In genal gypsum (CaSO, l 2H,O) is the stable form at low temperatures while anhydrite is formed at higher temperature. Even though anhydrite would be expected above 40°C due to its lower solubility while gypsum (CaSO,.2H,O) can be found at temperatures up to 100°C.

As a result of the limited solubility of calcium sulphate, the distillation temperature in desalination plants is restricted to 120°C at the most and the concentration factor (CF), (with reference to normal sea water) is less than 2. But by removing the calcium and sulphate ions from seawater, consider- ably higher evaporatoration temperature can be attained, as well as higher concentration factors [2].

Skillman, McDonald and Stiff [5] suggested a calculation method for the scaling tendency of water with respect to calcium sulphate as follows:

s=looo[pTiEz]

where S=solubility of gypsum, x=excess common ion concentration, the difference between Ca+ + and SOL concentrations and K=constant depend- ing on the ionic strength of water and its temperature.

MECHANISM OF SCALE FORMATION AND BIOFOULING IN DESALINATION

PROCESSES

Scale formation problems are widely encountered in MSF plants while biofouling is the major problem in RO plants. The driving force for scale formation is the super saturation of scale forming agents. The stages of scale formation (as well as fouling formation) are given below.

Mechanism of scale formation

These mechanisms can be grouped into five steps as follows:

Initiation offouling. During this initial delay period, the surface is being continued for the fouling which will take place later. This step which is usually observed in precipitation and crystallization fouling lasts to the order of hours. For example, Ritter [6,8] observed an induction period of 20 hours while studying the deposition of calcium and lithium sulphate in crystalliza- tion fouling. After this period has been observed, the fouling resistance starts to increase with time in some fashion.

292

Transport to the surjke. This transport results from a variety of process- es including: (i) diffusion, (ii) sedimentation, (iii) turbulent downsweep and (iv) thermophoresis. Diffusion plays an important role in fouling especially, in the transport of both gaseous and particulate species. Sedimentation operation, as might be expected, has significant importance in fouling where particles are dense and the fluid velocities are low. It is generally considered in liquid fouling side.

The next transport process considered here is that due to turbulent down- sweep. Cleaver and Yates [9,10] found that, eddies in the fluid stream are able to penetrate the laminar sublayer and transport solid material to the surface. They also obsrved that; turbulent bursts are an efficient removal mechanism.

The final transport mechanism considered here is the thermophoresis. This mechnism is defined as the movement of small particles in the fluid stream under the influence of a temperature gradient. This thermophoresis mechanism is important for particles below 5 micron and becomes dominant at about 0.1 micron.

Attachment to the surjbce. Not all of the materials transported to the surface actually sticks. The forces acting on the particles as they approach the surface play an important role. Also, the properties of the particles such as density, elasticity, surface conditions and state are important. Finally, the nature of the surface such as roughness and its type material plays an important role in sticking of various particles to its surface.

Removalfrom the su@-xe. Material can be removed from the deposit by several mechanisms, including spalling (caused by the shear forces and turbulent bursts), i.e. resolution and errosion. The velocity gradient, velocity of the fluid and the roughness of the surface can also play an important role in the removal mechanism. Resolution of the deposit material can occur, for example, if the pH of the liquid stream is changed by additives or some other means. Errosion by particulate matter or by liquid impengement can remove material from the fouling layer.

Ageing of the deposit. Once a deposit layer is formed on a surface, it does not remain static. Usually, the deposit thickness grows up with time until it reaches a stable value. In addition, the mechanical strength of the deposit can change with time due to chanage in crystal structure or the chemical composition of the deposition. Hence, ageing may strengthen or weaken the scale deposit.

293

l%e biofouling mechanism of RO plants

In principle RO is just a mechanical process for extracting the fresh water from saline one under application of high pressure as shown Fig. 3. This mechanism is illustrated by its steps of formation as shown in Fig. 4. During the first few hours of operation, concentration polarization increases the salt concentration at the surface of the membrane. This increases the osmotic pressure of the feed water thereby reducing water flux. Concurrently bacteria in the feed water are attaching to the membrane. It appears that, only certain bacteria can attach at this stage and a limited number of attachment sites (about 15 % of the membrane area) can be detected. During the next couple of weeks the bacteria either grow on the membrane, or else some other mechanism (like charge and/or hydrophobic changes on the membrane surface) allows the full membrane surface to be colonized by bacteria. By day 16 (Fig. 4c) the entire membrane is coated with bacteria several micrometers thick [ 111.

As the biofilm matures, more bacterial growth and extracellular polymers are added to the biofilm. Biofilm’s growth eventually becomes limited by the shear force of the bulk feed stream (Fig. 4d). It appears that the outermost layer of the biofilm is the least compact. It may be that with time, the older biofilm layers compress (Fig. 4-e) forming a more impermeable barrier; and/or salt concentration increases at the membrane to back diffusion produced by the growing and/or compressing biofilm. While the exact mechanism is not known, the reduction in water flux and the increase in salt flux throughout the life of the membrane element is well known.

Fig. 3. RO principle of operation.

294

1 I I I I Fig. Ib. (’ **3 %f

I I Fig. Lc.

I Fig. Cd.

FF*w .. .Q - ..a .:. .o.. ,. .‘_.. ..I

,ime;) ycor

1 x

Fig. be.

Fig. 4. Mechanisms of membrane fouling.

In addition to biological fouling, the RO desalination plants suffer also from the following: suspended solids, colloids, scale formation, oil and grease deposits and metal oxides.

It was also noticed that when the membrane surface is coated with the biological foulants, it results in the formation of slimes, and the rejected salts by the membrane are trapped in these slimes layer and will not swept away by the feed water. Hence scale formed within this layer will blind the membrane surface and unless chemical cleaning is done the membrane may fail completely. Both chlorine and copper sulphate are used for disinfection of RO plants depending on the membrane type used.

The main types of scales usually present in the RO plants are carbonates, sulphates, fluorides and silica. In practice, the pretreatment steps are necessary for RO plants. These steps include acid treatment, pH adjustment, addition of antiscalants and disinfection processes.

Concerning the particulate fouling, prefiltration with multi-media filters (five micron catridge filter) will generally control this problem. But, the cooloidal foulings (less than 1 micron) are usually removed by coagulation and filtration. Coagulants used are alum, ferric chloride and polyelectrolytes [12]. In practice, the fouling guidelines are useful to reduce the fouling and scale problems in RO plants.

1 Reduce the product water recovery ratio to avoid exceeding the solubility limits.

295

Soften the water by ion exchange to remove polyvalent metal ions Acid dozing to reduce the pH and bicarbonate/carbonate level Add scale inhibitors as polyphosphates Use the suitable disinfection dose (of either chlorine or copper sulphate).

SIMPLIFIED MODEL FOR SCALE FORMATION

The basic scale model is usually given in the form of [13]:

(1)

where dRJ&=rate of change in fouling resistance, $d=deposition rate, and & = removal rate.

For water basically

4d = Cl Pd Qn exp -E t ) RTS

4, = c2 z 5

(2)

(3)

where

Cl, c2 : constant 7 : shear stress

X : deposit thickness * : strength of deposit

‘d : deposition of factor (related to velocity and stickiness of deposit)

tY : water quality factor exp (-E/RT,) : arrhenius reaction rate function

E : activation energy T, : absolute surface temperature R’ : gas constant

Integration of Eqn. (1) with substitute of Eqns. (2) and (3) leads to

Rr = clpd*n 5 = $1 J’

exp[$j[ 1 -exp( -“i ‘jr)]

= @‘d 7 [I-exp(-4,0] r

(4)

296

As time becomes large, then

In integration the values of T,, Pd, V, 7 and ?I! are assumed constant, i.e. applied to one scale, one water quality and one set of flow conditions. Therefore these constants can be gathered in one cond C,

5= cexp(-E)

The basic asymptolic model has been practically proved to be successful in describing the fouling relationship in MSF desalination plants [14].

TECHNIQUES OF SCALE CONTROL

lheoretical aspects

The driving force for scale formation is super saturation of scale-forming agents. In the case of sulphate scales, this saturation is caused by a reduction in the solubility of the calcium sulphate hemihydrate in water as the tempera- ture increase (inverse-solubility behaviour). But in the case of alkaline scale, super saturation is caused firstly by the inverse solubility behaviour and also by the following two reactions:

2 HCO, - CO, + CO, + H,O H,O + CO, - 20H- + CO,

which produce CO, and OH- required for Ca++ and Mg++ ions to form the precipitates of the alkaline scales (CaCOs) and Mg(OH),. Hence the chemical methods used for scale control are either the acid treatment (to neutralize the alkalinity of water) or the addition of scale inhibitors (antiscal- ing), which can prevent the precipitation of these scale forming salts. Other mechanical methods like cleaning (especially ball cleaning) are also applied.

For reverse osmosis processes, biofouling and suspended solids are the main fouling problems and hence disinfection and proper pretreatment steps are necessary in this case.

Acid addition (acid dosing)

The addition of a strong acid (usually sulphuric acid) to seawater contain- ing bicarbonate ions results in the following reaction:

297

HCO, + H+ - CO, + H,O

In MSF desalination plants, the carbon dioxide formed is removed from the feed water by passing it through a degassing tower before it is used in the plant. In theory, sufficient acid must be added to the seawater to react all the bicarbonate ions (and also hydroxyle ions) present in water in order to prevent alkaline scale formation. Acid addition is a simple and effective means of preventing alkaline scale formation and was widely advocated during the 1960s and 1970s as the only practical method of alkaline scale control in MSF plants operating at temperatures upto 120°C [15].

It is worth mentioning that the addition of sulphuric acid has led to the following problems:

1. Corrosion problems due to lower pH values. In many cases, the addition of acid is followed by treatment with sodium hydroxide solution to control the pH value precisely.

2. Transport and storage problems. 3. Safety problems due to the special care needed during its handling to

avoid the probable danger of accidents in this case. Hence, the other method of scale control (chemical additives) is widely used.

Chemical additives (or antiscalants)

The addition of chemicals to retard the formation of scale has been used by the water treatment industry for many years. Organic materials such as starch and tannin were used in many of the old submerged tube evaporators and were claimed to give rise to scale which did not readily adhere to the heat transfer surface.

Currently acceptable methods of preventing alkaline scale formation utilize chemicals which are generally known as threshold agents. This term “threshold agent” arose from the research carried out by a number of workers [16,18]. They found that; the addition of less than stochiometric quantities of certain polyphosphates to supersaturated solutions of various salts, particularly calcium carbonate, would prevent precipitation for substantial period of time. This property was described as a threshold effect because of the small quantities of chemicals which were used and chemicals which possessed this property were called threshold agents.

It is known that polyphosphates can prevent the precipitation of CaCO,; however, their use as scale inhibitors is restricted to operating temperature bel.ow 90°C, since they are hydrolyzed at higher temperature and from calcium phosphate which is difficult to dissolve [2].

In the last 20 years, new threshold inhibitors have been developed, primarily on the basis of polycarboxylic and phosphoric acids. The selection of the suitable inhibitor should be accompanied by the plant operator on the basis of practical tests under the actual conditions prevailing in the plant itself. All threshold inhibitors require a flow velocity more than 1.5 m/s, deposits must be expected at lower values of velocity. In many MSF plants, the use of these inhibitors is backed up with the Taprogge [ 151 system which permits continuous mechanical cleaning of the heat exchanger during the operation.

In principle, threshold inhibition can have an effect on nucleation and/or crystal growth.

Nucleation. It may be assumed that scale forming takes place in a hetero- geneous manner. The surfaces inside the plant as well as the particles suspended in the solution may serve as nucleation sites. An inhibitor may affect nucleation by lining these surfaces.

Crystal growth. The further growth of crystals takes place on the crystal surface at particularly reactive points in which the arriving ions are incorpo- rated. An inhibitor can take effect here by occupying these active centers. With this in mind it is possible to explain the effect of inhibitors even in sub- stochiometric quantities. If the active centers are practically blocked, this is associated with a change in the habit of the crystals being formed, this change may possibly prevent caking on the heat exchanger surfaces. Thus the precipitation are held in suspension due to the altered habit and can be removed in the blow down.

Mechanical cleaning

Two basic cleaning systems are currently used in desalination plants for scale removal. The first which is relatively old is the off-cleaning system (by brushes) while the plant is off. A second system is the on-line cleaning (by balls) with continuous plant operation.

Since scale formation problem is an unsteady state, both recording and analysis of actual plant data are of significant importance. An example of these data is given in Fig. 5 which indicates the fluctuations in fouling resistance, brine heater temperature, water production rate and flow velocity. It is clear that during the unsteady state period of plant (Region A), all these variables are oscillating depending on the thickness of the scale deposited, e.g a sudden increase in scale at point (1) in the graph is reflected directly in a sudden drop in both water productivity and its brine heater

299

+- Region A -- Region 8 1

6.0 - Fouling Resistance(Wlsq.m K)-‘xld4

x nw -

0 Outpol Cu.mlhr

E /

500 - 2 400 - dr I

- -u Top temperature ‘C

c .,o .E

z

90 VW 7sL/

.- ;i ‘;

x 5 -%

._ 80 - 3 2.5- Brine velocity mlsec

.c iz L- .r

5 I I

220 300 400

Days

Fig. 5. Performance profile. Steady-state conditions.

I I 500 600

temperature as well as the actual velocity of the flow (which is expected due to increase in pressure drop). Further mechanical cleaning (off-line cleaning using brushes) could again adjust the plant pefromance at the designed level. Region (B) of the graph which is quazi steady state of smooth operation was really achieved by preciese on-line cleaning. Recently the off-line cleaning has been replaced by a proper on-line cleaning program.

Chemical cleaning

This includes the use of dilute acids or EDTA, Even liquid ion exchan- gers can be used in certain situations [6].

iThe practical application and the present situation of scale control tech- niques in actual desalination plants in Saudi Arabia

The choice of the particular method to be used for scale control depends largely on the plant design, effectiveness of the method to be used and its costs [19]. Fig. 6 shows the cost of chemical additives at various brine temperatures.

Acid dosing

In this treatment a narrow range of pH (7.7-8.0) should be maintained to keep low fouling rate [13]. Acid treatment was considered economic. The use of antiscale additives was found to be 1.8 times that with acid in Jed III WI.

2 % ._ a

9-

a- l-

6-

5-

c-

3-

2-

1-

HTA 0 510=4.5

HTA (L SIOr3

._ 3 O

I

51 O.&O- .r a35 - t 0 0.30 - LTA 8, S/O:6 u Tj 0.25 - LTA C S/0:4.5

.g o.zo- -o LTA I SIOz3 o

2 0.15 - v n Ill

d0 160 Ii0 Top Brine Temperature, ‘C

li0

Fig. 6. Chellating additive chemical cost versus top bring temperature.

Antiscale additives

There are many types available additives based on operation temperature (high or low), water salinity and type of scale formed. Dosage of antiscale (around 3 ppm) increases with temperature and might reach 7 ppm (which means higher cost) at high temperature (lOg°C) [21].

It should always be remembered that with antiscale use, brine velocity should be kept 1.8 m/s [ 131. The velocity in the range of 1.5 - 1.8 m/set was widely recommended as a minimum value to insure minimum scale.

In practice, two types of additives are known: (1) Low temperature addi- tives (LTA) and (2) High temperature additives (HTA). Each has its advan- tages. Usually the most important parameter in correct selection is the performance ratio which is higher in HTA than in LTA. Examples are ALJ II with PR of 3.44 and 4.09 respectively [22] and with dosages of 5 and 7 ppm; and 4.33 and 4.82 with dosages of 3.73 and 6.19 respectively [23].

Meanwhile the higher perfromance ratio recorded with HTA is paid for by the higher cost of additives due to higher dosage. Therefore optimization procedure should be followed to determine the suitable type of additive. Some investigators recommended LTA (Aquagil) [24] for ALK II. Others recommended HTA for its lower capital cost [23].

fouling rate (~10% hr*ClkCal ) BRINE HEATER

3-O 4-- design fouling factor 3.0~10~‘m’ hr°C/Kcal

2.0 louling rate 2.389x10” rnz hr .C/kCal/day

301

HEAT RECOVERY SECTION

2.0 + design fouling factor 2.0~10% hrbltcal

fouling mte 2.867~10~ rr? hr ‘ClkCalktay

1.0

0

I 6 I ‘I , I I I c days 2 4 6 8 IO 12 14 16 18 20 22 24 26 20 30

Fig. 7. Heat transfer fouling rate-polyphosphate #2 Al Jubail 1.

It is worth mentioning that polyphosphate as low temperature additive (LTA) has proved to be widely successful in scale control as in Jub I and II, ALK II with the correct ball cycles of cleaning [20,23]. Actual field test of Al-Jubail I [25] during one month of operation showed that polyphosphate dosing could reduce the fouling factor by about 50% of its designed value as shown in Fig. 7. It is the most common LTA (below 91OC). At higher temperature hydrolysis or polyphosphate is expexcted [21].

It is usually added in a form of mixture (Aquagil) containing dispersing agent to prevent sludge formation (Mg C03, FqO,, organics, etc.) plus antifoam. As previously mentioned polyphosphate function is to prevent precepitation of CaCOs.

Another additive known as polycarboxylate (Belgard EV) which is a high temperature additive (HTA) was applied in Jed III bust was not promising because of high scale formed. This results is similar to Jed I, ALK I and Jed oil Ref. and Jed port.

It is found that Belgard EV can be effective only in small units as LTA [26]. But in Jub I Belgard RB gave good results as LTA (TBT=90.6OC) and close to other additive tested [27]. Meanwhile its effect may vary from one unit to another giving different perfromance ratio (PR38.969.55) [27].

An interesting field study was carried out in ALJl (Fig. 8) to find the comparative effect of various types of additives used on the scale formation and control. It .was found that except Shuwaikh additive, all other additives have almost the same effect on the scale composition formed (Table II). Major scale components were calcium carbonates, magnesium hydroxide, and organics .

302

RGE ,

DESUPERNEATER

I A CONDLHSAlE C J A!‘!!_. I I I 1

Fig. 8. Flow diagram of Al Jubail 1.

TABLE II

Scale component sampled at water box of brine heater outlet for each additive used

Component Belgard Albrivap Falcon Albrivap Shuwaikh EVN B 247 A Mix unit 1 Unit 2 unit 4 unit 5 Unit 6

CaC03 30.0 26.5 23.7 28.9 25.0 CaS04 4.4 2.2 Nil 3.7 6.2

Mg(OH), 26.4 42.0 31.4 11.4 18.4 NaCl 4.1 0.3 0.5 0.5 21.8 SiO, Nil 2.6 2.1 2.5 2.6 FeS04 1.7 4.7 0.52 2.0 2.0 cue Nil 0.3 0.2 0.1 0.2 Oreanics 25.8 24.7 36.5 32.45 33.2

Ball cleaning

As previously mentioned on-line sponge ball cleaning is essential especially with additives treatment in order to keep low fouling coefficient and to increase running period before acid cleaning [22,24]. It should be utilized in heat recovery section and brine heater to minimize heat transfer fouling [22]. Optimum dosing of balls varies with capacity of unit and chemicals used. Examples are ALK II with 12 balls/day/tube [24] and Jub I and II with 8 balls/day/tube [23] as average frequency of balls increases with the decrease in brine velocity in tubes [28].

Based on the data published in literature as well as the research and technical activities of SWCC [29] a comparison could be made between various methods of scale control techniques applied in Saudi Arabia as shown in Table III.

303

TABLE III

Comparison between various additives

Acid cleaning equipment cost Ball cycles PR Dosing Ball cleaning Scale fouling Dearetor Decarborator

LTA HTA Acid

Essential Essential Not required Medium Medium LOW High High Very low High High Medium LOW High Not applicable Must Must Not necessary Very low Very low LOW Essential Essential Essential Not required Not required Essential

The desalination plants in Saudi Arabia which uses those additives together with the ball cleaning and/or acid dosing are also summarised in Table IV.

TABLE IV

Present situation of scale control techniques currently used in some of the desalina- tion plants in Saudia Arabia

Cap/MGD LTA + dosage HTA Acid Ball + dosage cleaning

JED I 5 II 11 III 23 IV 58

Yan I II

29 20

MEC I 50 ASSER I 25 KHAP II 5

JOB I 36

JOB II 259

ALK I ALK II

7.5 NA NA Yes No 52 Polyphosphate BelgardEV Yes

NA NA NA

NA NA Yes Yes

Yes No Yes No Yes No Yes No

NA NA Yes

YeS Yes

Polyphosphate Polycarboxylate Polymeric BelgardEV Yes

PRETREATMENT PROCESSES IN REVERSE OSMOSIS DESALINATION PLANTS

No doubt that raw water pretreatment is necessary for all membranes currently used in desalination processes so as to attain a more reliable and economic operation. In the reverse osmosis desalination plants, fouling is the main problem and it is mainly ascribed to the presence of suspended solids and microorganisms. For the membrane materials most currently used (polyamide and cellulose accetate), the feed water to the plant must meet the following requirements (according to membrane manufacturers).

A) Modules with Polyamide Membranes

Hollow fiber module: pH value Temperature Free chlorine SD1 (silt density index)

4-11 Max 45OC not detectable Max. 3% min-’

Tubular, spiral-wound aud plate aud frame module: pH value 4-11 Temperature Max 40°C Free chlorine not detectable Turbidity Max. 1.0 TE/F SD1 Max. 3-5% min-’

B) Modules With Cellulose Acetate Membranes

Hollow fiber module: pH value Temperature Free chlorine SD1 (silt density index)

3-8 Max 40°C Max. 1 mg/liter Max. 4% min-’

Tubular, spiral wound and plate and frame module: pH value 3-8.5 Temperature Max 45OC Free chlorine Max. 1 mg/liter Turbidity Max. 1 TE/F SD1 Max. 4% min-’

305

TABLE V

RO plant performance

Product water quality Umm Lujj Jedaah Al-Biik

2-m @pm) >200 850 > 225 PI-I 8.0 Blended 8.9

Total Capacity

Design (m3/d) Initial (m3/d) Present, Sept 88 (m3/d) Average annual decline from design

Salt Rejection

Initial (from 2 stag=) Present (Sept 88) (from 2 stages)

Average decline (annual)

Recovery

Initial Present

12cNxl 2275 4520 12000 2272 3940 6()OODl 1500[4’ 3.48% 2.5% 7.13%

99.99% 99.49%

98.5% 92.2% 99.45%

0.16 0.78%

25.7% 24.21%

30.5% 28.4%

Energy

22.6 max Without energy recovery (kWh/m3) 17.0%

With energy recovery (kWh/m3) 14.5%

Saving 15%

Plant Availability

Downtime > 95% g-day

94.1-95.236

306

TABLE VI

Water analysis and composition (PPM) of water feed to the three SWRO plants

Al-Birk Jeddah Umm Lujj

Na+ 12000 12600 13957 K+ 580 Ca+ + 473 500 484 Mg++ 1591 1423 1598 Fe++ 0.03 0 0.01 Cl- 22200 22048 21493 5334 3000 3216 3100 HC03 95 43 132 SiO, 0.1 0.4 Cl TDS 39000 41492 41157 Conductivity @cm 60540 58000 PH 8.2 8.1 8.2 Turbidity (NTU) 0.25 0.5 0.2

1. Salinity of feed water (or TDS) 2. Hardness of water used. 3. Colloidal suspended matter 4. Presence of organic or bacterial matter 5. Presence of oxidizing agents 6. Presence of metallic ions in solution 7. Presence of hydrogen sulphide 8. Temperature of feed water 9. Volumetric flow to the plant and output required from the plant.

The major factors which determine the type of pretreatment required are:

According to a recent study conducted by the saline water conversion corporation [30] (SWCC in Saudi Arabia) on three RO plants (Jeddah, Umm Lujj and Al-Birk Plants) the main pretreatment steps used are:

1. Disinfection. Either by copper sulphate or by chlorine, but residual chlorine have to be removed completely by sodium bisulphite.

2. Coagulation, jloculation andflltration. Alum and magnifloc 537°C were used as coagulants while sand filters were used for filtration followed by fine cartridge filters (5-25 pm). The filter water then has SD1 range of 3- 4% min-‘.

307

3. Softening and antiscaling. Sulphuric acid is dosed in the feed at a concentration range of 30 - 100 ppm (depending on the analysis of feed water used in each plant) while sodium hexa-meta phosphate at a level of about 2 ppm is also used to prevent Ca++ or Mg++ sulphate scaling.

The performance of the above-mentioned plants (Table V) are acceptable, but only fouling problems were detected in Al-Birk plant [30]. The sea water feed analysis of these plants is given in Table VI.

Fig. 9 gives the flow sheet of one of these RO plants (Al-Birk).

REDOX

METER-

+ IO train zoo 160m)l h

--- SDI 501 *

16Om)lh TDS400OOPPm

lo sea dutlall 109tn'fh

Fig. 9. Schematic flow diagram of Al-Birk SWRO plant.

THE INTERACTION BETWEEN CORROSION AND SCALE OR FOULING PROBLEMS

Corrosion and corrosion products especially in the thermal desalination plants play an important role in foulding and scale formation processes. For example in the acid dosing systems used in the pretreatment sections, corrosion of metalic surfaces occurs (due to the low pH value of the medium) and the corrosion products will roughen the surfaces of these equipments and will be considered as active sites for precipitation of more scale deposits at that places. In addition, these corroded surfaces can also become a good place for microbiological growth leading to biofouling problems. Therefore a cycle of corrosion and fouling will develop. This

308

TABLE VII

Common scale and fouling problems in desalination plants and means of treatment

Constituent Difficulties Caused Means of Treatment

HardneSS Chief source of scale in ther- Softening demineralixation

Alkalinity

mal units and pipelines surface active agents

Main reason of alkaline scale Acid treatment lime and lime-

PH

Oxygen

Low pH causes corrosion while high pH cause scale

Corrosion problems

Cloride Main source of corrosion problems

Sulfate

Free mineral acids

Combines with calcium to form calcium sulfate scale

Corrosion

Biological and organic impurities

Disolved solids

(TDS)

Biofouling in R. 0. plants

High TDS values (Arabian Gulf) will limit high tem- perature of plant and reduce the recovery

Silica Scale formation

Suspended solids Cause particulate fouling

Iron Scale fouling problems

Turbidity Deposit in water lines and process equipment

soda softening

Precise control of pH

Sodium sulfite corrosion inhibitors

Use of proper material of construction

Use of scale inhibitors

Neutralization with alkaliies

Pretreatment and disinfection

Optimum selection of temper- ature and performance ratio

Pretreatment system

Filtration and coagulation

Corrosion inhibitors and/or ball cleaning

Coagulation, settling and filtration

cycle can be started by a small amount of particulate fouling followed by crevice corrosion and/or biological corrosion fouling [6].

309

In order to avoid corrosion in desalination plants, more attention has to be given to the proper selection of the material of construction in each part of the plant and oxygen removal has to be done completely either by use of dearators or adding of the oxygen scavengers (like Na$O,) in the pretreat- ment section of the flowsheet. Removal of hardness from feed water and use of corrosion inhibitors like polyphosphates or pilgard are also promising to reduce the corrosion problems in desalination plants.

In view of the widespread literature on scale problems and its drawbacks as well as the present practical situation in desalination field either by thermal processes (MSF) or membrane processes (RO), the present work pointed out the main problems associated with both scale and fouling tendencies of saline water together with its recommended suitable remedies as shown in Table VII.

CONCLUSIONS

1. Precise control on the opeating conditions like temperature, pH, solids content and flow velocity is of significant importance to achieve smooth operation with minimum scale problems.

2. Removal of hardness from feed water (in the pretreatment section) can help not only in reducing the scale problems but also to increase the plant recovery and economy.

3. More attention has to be given to the pretreatment steps in desalination plants especially for the RO systems.

4. Disinfection of feed water and removal of any residual chlorine are essential to avoid fouling problems of RO units.

5. Mechanical cleaning compete with chemical additives (antiscalants) for scale control in MSF plants and hence actual field studies are necessary for each plant to achieve the optimum cleaning program for each case (level of additive dosage plus cycle of ball cleaning).

6. Selection of the proper materials and alloys as well as oxygen removal from feed water can help greatly in reducing corrosion fouling problems in desalination plants.

RECOMMENDATIONS

1. More theoretical work is needed to understand the actual mechanisms of scale and fouling formation and also to correlate it with both the operating process conditions (T, pH, flow velocity, TDS) and the type of feed water processes.

310

2. More improvement is required in membranes to handle wider ranges of operating conditions (pH, free chorine suspended solids, and high water fluxes).

3. Precise monitoring of the process variables (temperature, flow rates, pH) are required to detect any scale formation deposits at each part of the plant in order to tackle this problem at its early stages.

4. Precisecontrol equipment for pH, oxygen content, surface temperature and suspended solids as well as free chlorine.

REFERENCES

1 J.G. Knudsen, Fouling in Heat Exchangers, in: Hemisphere Handbook of Heat Exchangers Design, G.F. Hewitt, ed., Hemisphere, 1990.

2 H.G. Heitmann, H.G., Saline Water Processing, VCH, Weinheim, 1990. 3 N.M. Wade, Desalination, 31 (1979) 309. 4 H.A. Stiff, Jr. and I.E. Davis, Petroleum Transactions, (1952) 195. 5 H.L. Skillman, J.P. McDonald, Jr. and H.A. Siff, Jr. H.A., A Simple Accurate Fast

Method for Calculating CaSO4 solubihty in Oil Field Brine, Spring Meeting of the Southwestern District Division of Production, Lubbock, Texas, 1969.

6 M.I. Alahmad and F.A. Abd El Aleem, Third Saudi Engineering Conference, Vol. I, Nov. 1991.

7 N. Epstein, Heat Transfer Engineering, 4(l) (1983) 43. 8 R.B. Ritter, R.B., Heat Transfer, 105 (1983) 374. 9 J. W. Cleaver and B. Yates, Chem. Eng. Sci., 30 (1975) 983.

10 J.W. Cleaver and B. Yates, Chem. Eng. Sci., 30 (1976) 147. 11 D.H. Paul and A.M. Abanmy, Ultra Pure Water, 7(3) (1990) 25. 12 K.O. Talal, M. Lutfi and M. Bakheat, Topics in Desalination, Report of Research Activities

of SWCC, Saudi Arabia (1986), KISR, Kuwait, 1986. 13 J.G. Knudsen, J.G., Analysis of the fouling processes, Part 3.17.3, in: Handbook of Heat

Exchangers Design, Hemisphere, Weinheim, 1990. 14 K. Cooper et al., Desalination, 47 (1983) 37-47. 15 A. Porteous, Desalination Technology, Applied Science, London, 1983. 16 0. Rice and E.P. Partridge, Ind. Eng. Chem., 31 (1939) 58-63. 17 T.F. Buchrer and C.F. Reitemeier, J. Phys. Chem., 44 (1940) 535-551. 18 G.B. Hatch and 0. Rice, Ind. Eng. Chem., 31 (1939) 51-63. 19 Y.I. El-Akeel and H.I. Amer, Desalination, 23 (1977) 255-262. 20 A.M.A. Al-Mudaiheem and R.M. Sxoslak, R.M., Desalination. 21 M. Al-So& M.S.F. Chemical and Fuel Consumption, in: Topics in Desalination, SWCC,

Saudi Arabia, 1986. 22 M. Al-Sofi et al., Desalination, 55 (1985) 357-371. 23 N. Nada, 2nd Saudi Engineering Conference, Dhahran, Nov., 1985. 24 D.S. Khumayyis, Desalination, 55 (1985) 43-44. 25 N. Nada, Desalination, 50 (1984) 83-86. 26 N. Nada, SWCC Seminar on H.T.A., Jeddah, April, 1981. 27 D.S. Khumayyis and M. Ohtani, Desalination, 45 (1983) 155-165. 28 A.M.A. Al-Mudaiheem and H. Miyamora, Desalination, 55 (1985) l-11. 29 Topics in Desalination, Report of Research Activities of the Saudi Saline Water Conversion

Corporation, Riysdh, 1986. 30 A.M. Hsssan et al., Desalination, 74 (1989) 37-51.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具

http://www.xuebalib.com/cloud/http://www.xuebalib.com/http://www.xuebalib.com/cloud/http://www.xuebalib.com/http://www.xuebalib.com/vip.htmlhttp://www.xuebalib.com/db.phphttp://www.xuebalib.com/zixun/2014-08-15/44.htmlhttp://www.xuebalib.com/

Scale formation and fouling problems effect on the performance of MSF and RO desalination plants in Saudi Arabia学霸图书馆link:学霸图书馆