Scale Control in Saline and Wastewater Desalination

David Hasson* and Raphael Semiat

GWRI Rabin Desalination Laboratory, Department of Chemical Engineering, Technion—Israel Institute of Technology,
Haifa 32000, Israel

(Received 28 November 2004 and in revised form 12 October 2005)

Abstract. This paper briefly reviews current understanding of basic phenomena

involved in scale formation, outlines current scale control techniques, and indicates
some promising research directions.

Introduction Scale Formation Mechanisms

Scale deposition is a difficulty encountered with water
containing ions of sparingly soluble salts and is one of
the obstacles hampering the further progress of desalinat-
tion. Scale precipitation occurs whenever process condit-
tions lead to the creation of supersaturation with respect
to one or more of the sparingly soluble salts. This diffic-
culty is encountered in both evaporation and membrane
desalination processes. Unless preventive measures are
undertaken, the concentration effect involved in every
desalination process induces precipitation of sparingly
soluble salts such as CaCO3 and CaSO4.
Scale deposition cannot be tolerated because of its
highly deleterious effects on production capacity and
specific energy consumption. It is therefore essential to
incorporate in every desalination process scale control
measures that can be rationally designed and are cost
effective. The difficulties lie not in the lack of adequate
scale control techniques but in the constraint of a narr-
row cost margin and in the frequently encountered situa-
ation of insufficient know-how. The aim of this paper
is to review current understanding of basic phenomena
involved in scale formation, to outline current scale cont-
trol techniques, and to indicate some promising research
directions.
Scale deposition is a crystallization process involvi-
ing four stages:1–5
• Attainment of supersaturation
• Nucleation
• Crystal growth around the nucleus
• Growth of small crystals into larger ones; scale
layer thickening
Creation of a Supersaturation Potential
The most important single factor determining the
intensity of scaling is the supersaturation level of the
deposit-forming species. Supersaturation conditions are
achieved when a solution is concentrated beyond the
solubility limits of one or more of its constituents by
evaporation (as in thermal desalination processes) or by
separation of pure water (as in membrane processes).
Supersaturation conditions can also be achieved by
a change in temperature. Most of the frequently enc-
countered scaling salts—CaCO3, Mg(OH)2, CaSO4, and
Ca3(PO4)2—exhibit inverse solubility characteristics,
i.e., solubility decreases with increasing temperature.
A solution of an inverted solubility salt in contact with
*Author to whom correspondence should be addressed. E-mail:
hasson@techunix.technion.ac.il

Israel Journal of Chemistry Vol. 46 2006 pp. 97–104

98

a hot surface can attain supersaturation by the inverted

solubility effect and deposit scale on the hot surface,
even when undersaturation conditions prevail in the
bulk of the solution.
Induction Period
A supersaturated solution is at an unstable equilibr-
rium and relieves its supersaturation by precipitating
a solid phase. Attainment of supersaturation alone is
not sufficient for a system to begin precipitation. At
relatively moderate supersaturations, the solution can
remain stable without precipitation for a certain time,
which is termed the delay or induction period. Only at
a sufficiently high supersaturation level is the induction
period virtually absent and precipitation instantaneous.
The induction period phenomenon is of utmost practic-
cal importance in scale control efforts. The widely used
technique discussed below, of scale control by dosage of
an anti-scalant, is based on the increase of the induction
period induced by the presence of the anti-scalant. Fig. 1. Linear correlation of induction time data for CaSO4
The physical mechanisms governing the induction deposition in the presence of anti-scalants.7 SRw denotes satur-
period are nucleation processes. To create a solid phase, ration ratio at the membrane wall.
an energy barrier has to be overcome. An embryo cons-
sisting of a cluster of molecular aggregates can survive at two levels.7 The inhibitory effect exerted by anti-scala-
as a stable solid nucleus only if it reaches a certain ants is also reflected in the increase in surface energy
critical size. Nuclei are of visually unobservable micros- for deposit formation in the presence of anti-scalants.
scopic dimensions. Their formation is facilitated by the Surface energy values extracted from data similar to
presence of a deposition surface so that in all practical those of Fig. 1 were around 15 mJ/m2 in the absence of
systems nucleation can be assumed to be of a heterogen- an anti-scalant, increasing to the range of 20–30 mJ/m2
neous nature. in the presence of anti-scalants dosed at concentrations
Induction time data may be correlated with the supers- of 6 to 12 ppm.
saturation level S by assuming that τ, the extent of the
Scale Layer Growth
induction period, is inversely proportional to the heterog-
Once an initial scale layer is formed, subsequent
geneous nucleation rate:6
deposition is facilitated. Growth of a crystal layer on a
(1) flow surface involves several consecutive processes:3
1. Diffusional transport of the crystal-forming ions
where T is the absolute temperature prevailing on the
towards the crystallizing layer;
deposition surface, σ is the scale crystal surface energy,
2. Incorporation of the ions on growth sites of the
θ is the contact angle of the crystallite on the surface,
crystal lattice;
and C1 and C2 are constants of the system. Induction
3. Adhesion and removal processes.
times are seen to be highly dependent on the supersatur-
ration level. A long induction period is induced by a low The interfacial process of crystallization (step 2) is
supersaturation level, a low surface temperature, and a viewed as a “surface reaction process”. In the absence
non-wettable surface (low contact angle θ). Induction of removal effects (to be discussed later), the deposition
periods can, therefore, be characterized by measuring flux r kg/s m2 (corresponding to a scale growth velocity
the slope of the straight line correlating lnτ with 1/(ln S)2 of dx/dt m/s) is obtained by equating the transport rate of
and extracting from the slope the value of the surface the scale constituents to the crystallization surface with
energy parameter σ. the rate of the crystallization reaction:1,6
Equation 1 can be used to evaluate the effectiveness
(2)
of anti-scalants in delaying scale precipitation. Figure 1
shows induction times for CaSO4 scale deposition meas- where x is the layer thickness, ρs is the scale bulk dens-
sured in a recycling membrane system at various supers- sity, kD is the mass transfer coefficient, kR is the surface
saturation levels in the presence of an anti-scalant dosed reaction kinetic coefficient, CB is the solution concent-

Israel Journal of Chemistry 46 2006


tration of the scaling salt, Ci is its concentration at the
solution–scale crystal interface, and CS is the solubility
of the scaling salt. For most crystallizing salts, the value
of the power coefficient is p = 2.
When the scaling rate is relatively high, the deposit-
tion process is governed by the diffusional transport
process. In this case the increase in scaling rate with
temperature is moderate and the main effect is exerted
by the flow velocity. Since the mass transfer coefficient
kD increases with the velocity V (kD ~ V 0.33 in laminar
flow and kD ~ V 0.8–0.9 in turbulent flow), an increased
velocity facilitates transport of the scaling species to the
flow surface, thus acting to augment the scaling rate.
In most industrial systems (notably in anti-scalant
treated feeds), the scaling rate is relatively low and scali-
ing is governed by the rate of the crystallization surface
reaction. In this case the velocity effect is secondary.
An increase in temperature sharply increases the scali-
ing rate, since the crystallization rate coefficient kR foll-
lows an exponential dependence given by the Arrehnius
relationship:
(3)
where E is the crystallization reaction activation energy
and R is the universal gas constant.
Data on the kinetic coefficient are scant, but since
values of kD can be readily estimated, it is possible to
evaluate an upper limit of the scaling rate from the exp-
pression:
(4)
The mass transfer coefficient kD for typical flow cond-
ditions in spiral-wound modules is of the order of 10–5
to 10–6 m/s. With this mass transport coefficient and a
scale density of about 2000 kg/m3, the maximum posss
sible scaling rate that can occur with a supersaturation
driving force of ∆C ppm is:
(5)
with M having a value in the range of 0.04 to 0.4.
Impurity Effects
A tenacious well-bonded scale deposit can form only
when the water precipitates a pure crystalline deposit.
Fortunately, most practical systems contain impurities
to some extent. Impurities tend to adsorb on active sites
of a crystallizing deposit and interfere with the format-
tion of a well-ordered solid structure. Such processes
can induce a significant decrease in both the nucleation
rate and the crystal growth rate, often accompanied by
a change in precipitate morphology. Deposits formed
from mixed salts and in the presence of impurities have
a weakened adherence and the net scaling rate will be
99
lowered due to partial removal of the deposit by the
shear force acting on the flow surface.
The deposition rate of a scale layer composed of
a pure crystalline substance such as pure calcite will
usually be governed by the rate of diffusional transport
discussed above. The layer thickness will grow linearly
with time and an increase in flow velocity will augment
the rate of scaling. On the other hand, a weakly adhere-
ent scale deposit formed in the presence of impurities
usually experiences a decrease in the scaling rate with
time due to effects such as removal of weakly adherent
crystals, accumulation of adsorbed impurities on act-
tive crystal sites, or buildup of a repelling electrostatic
charge.
The above phenomena are illustrated in Fig. 2, which
shows the effect of an anti-scalant on the growth of a
calcite scale layer depositing from a falling film of a
supersaturated hot Ca(HCO3)2 solution flowing down
the outer surface of a vertical tube.8 As discussed below,
anti-scalants are in effect impurities that adsorb on act-
tive crystal sites and hinder the crystallization process.
Figure 2 shows that a dosage of 0.5 ppm of a polyp-
phosphate anti-scalant (SHMP) results in an increase
in the induction period and suppression of further layer
growth after an asymptotic thickness is reached.
Metal Impurities
A literature review focusing on the effect of Zn+2
ions on CaCO3 precipitation provides credible evidence
that Zn+2 and other ions can generate a useful scale supp-
pression effect. The experimental findings of various
researchers (e.g., refs 9–11) indicate that various metal
ions can exert retarding effects on nucleation and crystal
growth and can also affect the crystal habit. It is evident
that both Zn+2 and Cu+2 ions can exert an inhibitory act-
tion on scale formation and that their full potential for
scale suppression has been very scantily explored. The
Fig. 2. Asymptotic fouling induced by the presence of an anti-
scalant.8
Hasson and Semiat / Scale Control
100
Fig. 3. Scale inhibition in a membrane system by low zinc ion
dosage. Lp = membrane permeability; Lpo = initial membrane
permeability; Tb = bulk temperature; TA = total alkalinity;
Qconc = concentrate flow rate.
utilization of Zn+2 ions appears to be preferable to that of
Cu+2 ions, as the permissible Zn+2 level in drinking water
is 5 mg/L compared to 1 mg/L for Cu+2.12
Figure 3 shows results of exploratory experiments
undertaken with scaling waters at two supersaturation
levels (LSI = 1.1 and 1.5) in a recycling membrane syst-
tem. It is seen that the permeate flow decline of 25 to
30% observed in 16 h in the untreated water could be
arrested by a dosage of 2 ppm Zn+2. We are studying the
possibility of replacing all or part of the currently used
organic anti-scalants by trace Zn ion dosage.
Scale Control Techniques
The severity of a potential scaling difficulty in a des-
salination process depends of course on the nature and
concentration of the scaling species present in the raw
feed water. The main scale control strategies are:3,4
• Removal or reduction of scale-forming species;
• Inhibition of scale deposition by dosage of anti-
scalants;
• Controlled scale deposition with periodic cleaning
schedules.
Prevention of Supersaturation Conditions
The most effective way for eliminating a scaling
problem is by water treatment measures designed to
eliminate the scaling constituents present in the feed
water. For instance, acidifying the raw water and degass-
sing the CO2 released by carbonate decomposition will
prevent alkaline scale formation. This technique is often
adopted but usually in combination with anti-scalant
dosage at an economically optimized pH level.
Israel Journal of Chemistry 46 2006
Ion exchange columns can readily remove scale-
forming ions. However, attempts to develop an ion
exchange process based on cheap regenerating chemic-
cals have so far been unsuccessful.13 Lime softening
processes, which are widely used for hardness removal
in water supply systems, are too expensive for desalin-
nation processes. Water hardness can also be reduced
by nanofiltration (NF) membranes. Because of their
negative charge, NF membranes preferentially retain
divalent ions and thus reduce the concentration of the
ions involved in the precipitation of CaCO3 and CaSO4.
Nanofiltration softening is constrained by economic
considerations, but has potential in integrated memb-
brane systems.
Electrolytic scale removal is a relatively new
technique4,14,15 that is mainly used for Ca hardness rem-
moval from cooling tower waters. An electrolytic scale
removal device consists of a stainless steel vessel acting
as a cathode and a central rod acting as an anode. The
scale removal principle of the electrolytic technique is
the generation of a high pH environment around the
cathode. Scale deposition is due to the alkaline environm-
ment created by the cathodic reactions:
2 H2O + 2e– ⇑ + 2OH– (6)
O2 + 2H2O + 4e– ⇒ 4OH– (7)
The hydroxyl ions promote removal of the calcium
hardness by the precipitation of CaCO3 according to the
reaction:
Ca2+ + HCO3– + OH– ⇒ CaCO3 ⇓ + H2O (8)
The scale precipitating on the cathodic walls of the
electrolytic system is removed either by periodic piston
scraping or by periodic charge reversal, which loosens
the adhering scale deposit.
Electrolytic scale removal has been scantily treated
in the literature. Very little quantitative information is
available on mechanisms of the electrode reactions and
on the influence of geometrical and hydrodynamic par-
rameters.
Figure 4 shows the effect of flow rate on scale rem-
moval by deposition on the cathodic surface under lamin-
nar flow conditions. The data show that the scaling rate
increases roughly with Re1/3. As previously noted, the
mass transfer coefficient in laminar flow also increases
with Re1/3, indicating that scale removal in an electrol-
lytic cell is mass transfer controlled.
The specific energy data displayed in Fig. 5 show that
the electrical energy consumption (kWh per kg CaCO3)
in electrolytic scale removal can be significantly red-
duced by operating at high Reynolds numbers. Another
important parameter is the solution conductivity. Energy

Fig. 4. Effect of the Reynolds number on the scale removal
rate in an electrolytic cell.
consumption is reduced at high salinities. The energy
consumption in experiments carried out with solutions
simulating desalination plant concentrates was as low as
2.2 kWh/kg CaCO3.
The exploratory data so far obtained indicate that
the electrolytic method holds considerable promise for
beneficial integration in reverse osmosis desalination
processes.
Suppression of Scale Precipitation by Anti-scalants
The most widely used technique for scale control is
by dosage of an anti-scalant. This technique is cost eff-
fective because in many cases scaling can be suppressed
with only a few ppm of the anti-scalant, typically less
than 10 ppm. Since such low dosages are only a small
fraction of the stoichiometric concentration of the aff-
fected species, the inhibition phenomena do not involve
chemical reactions. The inhibition is by physical rather
than chemical mechanisms and involves adsorption
processes.2, 16
Inhibition mechanisms
Adsorption of the inhibitor on the crystallization
surface acts to delay nucleation, reduce the precipitat-
tion rate, and distort the crystal structure such that the
deposit tenacity to the flow surface is weakened, as was
illustrated in Fig. 2. By adsorbing on embryos, inhibitor
molecules prohibit their growth beyond the critical size
and thereby retard the rate of nucleation. Since low-mol-
lecular-weight inhibitor molecules are more mobile, the
extension of the induction period is more pronounced
with the lower molecular weight molecules. In the cryst-
101
Fig. 5. Effect of the flow rate on the specific energy consumpt-
tion in electrolytic scale removal
tal growth stage, the inhibitor molecules can retard the
growth process in several ways. If the ions of the inhibi-
iting molecule are small and are preferentially adsorbed
on active sites, a relatively small crystal surface covera-
age is sufficient for effective retardation of the growth
process. Most inhibitor compounds possess large ions,
and effective crystal growth retardation probably occurs
through relatively large crystal surface coverage.
Anti-scalant compounds
Commonly used anti-scalants are derived from three
chemical families: condensed polyphosphates, organop-
phosphonates, and polyelectrolytes. Polyphosphates,
the cheapest anti-scalants, can display effective thresho-
old effects as well as dispersive and corrosion protection
properties. Organophosphonates are superior to the ino-
organic polyphosphates in their marked thermal and hyd-
drolytic stabilities over a wide range of pH and temperat-
ture conditions. Effective polyelectrolyte inhibitors are
mostly polycarboxylic acids. Some of the more familiar
polycarboxylates are polyacrylic acid, polymethacrylic
acid, and polymaleic acid. A noteworthy development
is the trend for designing copolymers having different
functionalities to achieve specific goals and for blendi-
ing different polymers so as to extend their anti-scaling
capabilities.2
Numerous studies have shown that all inhibitor polym-
mers are effective only within a relatively narrow range
of molecular weights, typically around 1000 to 5000 Da.
Howie-Meyers et al.17 investigated the role of calcium
and polymer MW on the crystal growth inhibition of
hydroxyapatite by polycarboxylates. Maximum inhibit-
Hasson and Semiat / Scale Control
102
tion activity occurred in a narrow MW range—around
1000–2000 Da for acrylate–maleate copolymers and
around 5000–6000 Da for polyacrylates.
It is presumed that the size of the inhibiting molecule
should be sufficiently large so that when adsorbed on the
scaling species, it occupies a sufficient area for exerting
its effects, but it should not be so large that it lacks suff-
ficient mobility and that its adsorption rate is too slow.
A low-molecular-weight fraction of a given inhibitor
is adsorbed on the scaling species more rapidly than a
high-molecular-weight fraction, but the adsorbed quant-
tity is larger with the high-molecular-weight fraction.
Adsorption increases with the degree of dissociation
of phosphonic and polycarboxylic acids. The enhanced
scale suppression activity encountered at high pH is att-
tributed to the increase in the degree of deprotonation
with increasing pH.
Anti-scalant performance
Since the fundamentals of inhibition mechanisms,
particularly from their quantitative aspects, are poorly
understood, the effects of operating parameters on inh-
hibition effectiveness are largely unpredictable. The
useful guidance available from anti-scalant companies,
based on service experience of their products, must
often be complemented by extensive field tests. Labor-
ratory procedures we have developed can serve to red-
duce the laborious and costly efforts involved in field
tests.7,18–22
When properly applied, anti-scalants can provide
Fig. 6. Illustration of the stability imparted by an anti-scalant
to a highly supersaturated CaCO3 solution.
Israel Journal of Chemistry 46 2006
excellent service for most but not all scaling species.
The striking inhibitory effectiveness that an anti-scalant
can provide is illustrated in Fig. 6. The figure shows
the change with time of the ion concentration product
.
[Ca2+] [CO32–] of a highly supersaturated CaCO3 solut-
tion, simulating the concentrate composition of the Gat
desalination pilot plant (near Kiryat Gat, Israel) at 88%
recovery. In this beaker experiment, the solution had an
initial LSI of 2.5 and was dosed with the anti-scalant
Permatreat 191 at three concentration levels: 5, 10, and
25 ppm. The high stability imparted to this highly sup-
persaturated solution is reflected by the extremely long
induction times, exceeding two months, required for
reaching the equilibrium value of the calcite solubility
product.
Anti-scalant problems
Some of the problems associated with the use of anti-
scalants are that they can induce fouling under certain
conditions. For instance, in the presence of calcium,
calcium phosphonates can precipitate.23 Virtually all
anti-scalants tend to promote biofouling in the presence
of assimilable organic carbon.24
Currently, the major uncertainties are in the control
of silica scales and calcium phosphates by available
anti-scalants. The difficulty in inhibiting silica scales
stems from its protean nature. We have applied a fundam-
mentally based procedure for assessing the relative eff-
fectiveness of various silica anti-scalants at comparable
supersaturation conditions prevailing on the membrane
surface.22
Comparative tests of four commercial anti-scalants
tested, dosed at recommended dosage levels, revealed
that there was no significant difference in their inhibit-
tory effectiveness. Data similar to those illustrated in
Figs. 7 and 8 indicated that the mitigating effect prov-
vided by all anti-scalants tested was to increase the
threshold limit for the onset of scaling from a membrane
surface supersaturation level of 1.7–1.9 in the absence
of an anti-scalant, to the level of 2.1–2.3 in the presence
of the anti-scalant. In all cases, the degree of inhibition
achieved was below that indicated in the commercial
literature. Scale precipitation occurred below the limiti-
ing solution bulk concentration of 250–300 ppm silica
suggested in the product manuals.
The calcium phosphate scaling difficulty imposes a
severe limitation on water recovery from wastewaters.
The increasing water scarcity can be alleviated by usi-
ing treated sewage effluents for irrigation purposes.
Secondary treated wastewater purified by an MF/UF
(microfiltration/ultrafiltration) process has restricted irr-
rigation usages. Unrestricted irrigation usage requires
further purification by an RO process.
Au: Need permission for figs 7 and 8.
Fig. 7. Permeability decline as a function of water recovery.22
Fig. 8. Silica supersaturation level on the membrane.22
Since wastewaters usually contain phosphate and
calcium ions, they are prone to precipitate sparingly-
soluble calcium phosphate salts. At the present time
there are no known reliable anti-scalants that can inhibit
calcium phosphate scale deposition. The literature ind-
dicates increasing efforts of anti-scalant manufacturers
to develop more effective calcium phosphate inhibit-
tors.25,26
Experiments carried out with simple model solut-
tions containing only CaCl2 and KH2PO4 at a NaOH-
controlled pH illustrate the calcium phosphate scaling
103
EXP. No. Ca+2 ppm PO4–3 ppm Cl– ppm log(s/s)bulk
60 321 26.2 587 1.48
61 324 26.6 587 1.46
62 318 27.2 587 1.49
Fig. 9. Flux decline as function of time in the presence of a
calcium-phosphate anti-scalant. Ex60 = experiment 60; log(s/s
wall) = log of the supersaturation ratio at the wall.
difficulty. Figure 9 shows the flux decline occurring in
the presence of an anti-scalant recommended by a well-
known company. Scale prevention under such situat-
tions calls for a lowered pH and a reduction in permeate
flux.
Concluding remarks
The scaling problem poses a unique challenge in desalin-
nation, due to the severe economic constraints imposed
by the very low cost of the water-product. Scale control
has been and continues to be one of the overriding cons-
siderations in the development of improved desalination
processes. Further reductions in the cost of desalinated
water can be expected from R & D efforts oriented tow-
wards better understanding of currently practiced scale
control techniques (notably the anti-scalant dosage
method) and by the development of novel techniques
(such as the electrolytic scale removal method).
References and notes
(1) Hasson, D. Precipitation Fouling—A Review. In Foulis
ing of Heat Transfer Equipment; Somerscales, E.F.C.;
Knudsen, J.G., Eds.; Hemisphere Publishing Corporat-
tion: Bristol, PA, 1981; pp 527–568.
(2) Hasson, D. In Understanding Heat Exchanger Fouling
and its Mitigation; Bott, T.R. Ed.; Begell House: New
York, 1999; pp 67–90.
(3) Hasson, D. In Scaling in Seawater Desalination; Glade,
Hasson and Semiat / Scale Control
104
H.; Ulrich, J., Eds.; Shaker Verlag: Aachen, 2001; pp
49–68.
(4) Sorber, C.A.; Valenzuela, S.R. Evaluation of an Elects
trolytic Water Conditioning Device for the Elimination
of Water-Formed Scale Deposits in Domestic Water
Systems; Tech. Rep.—Univ. Texas, Austin, Cent. Res.
Water Resources, 1982, CRWR 186.
(5) Perez, L.A. In Calcium Phosphates in Biological and
Industrial Systems; Amjad, Z., Ed.; Kluwer Academic:
Norwell, MA, 1998; pp 395–415.
(6) Mullin, J.W. Crystallization. 3rd ed.; Butterworth-
Heinemann: Oxford, 1997.
(7) Hasson, D.; Drak, A.; Semiat, R. Desalination 2003,
157, 193–207.
(8) Hasson, D.; Semiat, R.; Bramson, D.; Busch, M.; Lim-
moni-Relis, B. Desalination 1998, 118, 285–296.
(9) Wada, N.; Yamashita, K.; Umegaki, T. J. Cryst. Growth
1995, 148, 297–304.
(10) Pernot, B.; Euvrard, M.; Remy, F.; Simon, P.J. Water
SRT—Aqua 1999, 48 (1), 16–23.
(11) Abouali, E.; Jean, O.; Lédion, J. J. Européen d’Hydro­
logie 1996, 27, 109–126.
(12) De Juane, J. Drinking Water Quality, 2nd ed.; John
Wiley: New York, 1997.
(13) Harris, A. In Desalination Technology: Developments
and Practice; Porteous, A., Ed.; Applied Science Publ.:
1983, pp 31–55.
(14) Nair, M.K.V.; Misra, B.M. Desalination 1978, 27,
59–64.
(15) Elgressy Engineering Services, Commercial Catalogue.
Israel Journal of Chemistry 46 2006
Electrochemical System for Scale Precipitation and
Removal; POB 13390, Netanya, Israel.
(16) Sangwall, K. Prog. Cryst. Growth Charact. Mater.
1996, 32, 3–43.
(17) Howie-Meyers, C.L.; Yu, K.; Eilliott, D.; Vasudevan,
T.; Aronson, M.P.; Ananthapadmanabhan, K.P.; Somas-
sundaran, P. In Mineral Scale Formation and Inhibits
tion; Amjad, Z., Ed.; Plenum press: New York, 1995;
pp 169–182.
(18) Semiat, R.; Hasson, D.; Bramson, D. Proceedings of
AWWA Membrane Technology Conference; New Orl-
leans, Louisiana, 1997; pp 1013–1027.
(19) Drak, A.; Glucina, K.; Busch, M.; Hasson, D.; Laine,
J.M.; Semiat, R. Desalination 2000, 132, 233–242.
(20) Hasson, D.; Drak, A.; Semiat, R. Desalination 2001,
139, 73–81.
(21) Semiat, R.; Sutzcover, I.; Hasson, D. Desalination
2001, 140, 181–193.
(22) Semiat, R.; Sutzkover, I.; Hasson, D. Desalination
2003, 159, 11–19.
(23) Jasbir, S.G. Desalination 1999, 124, 43–50.
(24) Vrouwenvelder, J.S.; Manolarakis, S.A.; Veenendaal, H.
R.; van der Kooij, D. Desalination 2000, 132, 1–10.
(25) Betz, G.E. Betz Dearborn Australia Pty. Kwinana Water
Recycling Project: Solutions for Biofouling, Organic
Fouling and Calcium Phosphate Scaling on Wastewater
Recycling Membrane Plants. Report GENTD001, July
2002.
(26) Hendel, R.; Bandick, K. IDA Desalination Conference;
Bahamas, September 2003, Paper BAH03-091.