Desalination 408 (2017) 60–69

Contents lists available at ScienceDirect

Desalination

journal homepage: www.elsevier.com/locate/desal

Effects of substitution degree and molecular weight of carboxymethyl

starch on its scale inhibition☆
Yawen Wang, Aimin Li, Hu Yang ⁎
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210023, PR China

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Carboxymethyl starch (CMS) as “Green”

antiscalant was prepared simply.
• The effects of the structural factors of
CMS on scale inhibition were evaluated.
• CMS with higher substitution degree
owns higher scale-inhibition efficiency.
• Decrease of the molecular weight
would evidently reduce the required
CMS dose.
• CMS can disturb CaCO3 crystal growth
via chelating and complexing effects.

a r t i c l e i n f o a b s t r a c t

Article history: “Green” antiscalants are gaining increased interest because of their environment-friendliness. In this work, var-
Received 16 August 2016 ious samples of carboxymethyl starch (CMS), with different substitution degrees of carboxymethyl groups and
Received in revised form 15 December 2016 molecular weight were designed and prepared. The structures of these CMS samples were characterized by Fou-
Accepted 5 January 2017
rier transform infrared spectroscopy, 1H nuclear magnetic resonance, and viscosity. CMS was used as a green
Available online 13 January 2017
antiscalant for the inhibition of the growth and formation of calcium carbonate (CaCO3) scale, as evaluated by
Keywords:
a static test in laboratory scale. Apart from the environmental parameters, effects of structural factors of CMS, in-
Carboxymethyl starch cluding the substitution degree of carboxymethyl groups and molecular weight, on its scale-inhibition perfor-
Scale inhibition mance were extensively studied. Results showed that increased substitution degree of carboxymethyl groups
Substitution degree of carboxymethyl groups and decreased molecular weight of CMS samples favored the distortion of CaCO3 crystal growth through chelat-
Molecular weight ing effects and improved the scale-inhibition efficiency. The morphology and crystal form of CaCO3 scale were
Inhibition mechanism characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD), respectively, to further inves-
tigate the scale-inhibition mechanisms of CMS.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction

Scale deposition that mainly includes insoluble salts of calcium and

☆ Supported by the Natural Science Foundation of China (grant nos. 51378250 and
magnesium, such as CaCO3, Ca3(PO4)2, CaSO4, MgCO3 and Mg(OH)2,
51438008), the Natural Science Foundation of Jiangsu Province (grant no. BK20161405),
and Six Talent Peaks Project in Jiangsu Province of China (grant no. 2015-JNHB-003).
occurs frequently in circulating cooling water system, resulting in the
⁎ Corresponding author. reduction of heat transfer efficiency and the aggravation of corrosion
E-mail address: yanghu@nju.edu.cn (H. Yang). of the operation equipment. Moreover, the scale in drinking water can

http://dx.doi.org/10.1016/j.desal.2017.01.006
0011-9164/© 2017 Elsevier B.V. All rights reserved.
 Y. Wang et al. / Desalination 408 (2017) 60–69 61

do harm to human health, causing gastrointestinal diseases [1–2]. An ef- studied in detail, in addition to external parameters. To further investi-
fective approach for controlling scale formation is to add some gate the scale-inhibition mechanism of CMS, scanning electron micro-
antiscalants [3–6]. On the basis of widely accepted scale-inhibition scope (SEM) and X-ray diffraction (XRD) were respectively used to
mechanisms, which are the effects of chelation, dispersion, crystal dis- characterize the CaCO3 crystal morphology and structure. Accordingly,
tortion, and threshold, respectively [7,8], many inorganic phosphates the structure–activity relationship of CMS was well established and ef-
and organic phosphorus compounds have been employed as efficient fectively exploited.
antiscalants for years. However, those phosphorus-containing materials
can be harmful to the environment, especially eutrophication in water 2. Experimental section
[9]. In addition, polyphosphate antiscalants are usually unstable and
easily suffer hydrolysis, causing the phenomenon of calcium phosphate 2.1. Reagents and instruments
deposition [2]. Thus, the development of non-phosphorus antiscalants
is of great significance in scientific research and practical applications. Starch, of which weight–average molecular weight is around
Many studies have been done on scale inhibition using synthetic poly- 1.5 × 105 g/mol, was obtained from Jinhui Corn Development Co. Ltd.,
mers without phosphorus, such as poly(ethylene glycol), poly(acrylic Binzhou. Monochloroacetic acid (Lushuo Economic Trade Co., Ltd.,
acid), poly(citric acid) and poly(maleic acid), as antiscalants [10–13]. Zibo), calcium chloride (Xilong Chemical Reagent Co., Ltd., Shantou), so-
However, these polymeric antiscalants are usually non-biodegradable. dium bicarbonate (Lingfeng Chemical Reagent Co., Ltd., Shanghai), and
Moreover, many of reported polymeric antiscalants are ter- or even sodium borate (Lingfeng Chemical Reagent Co., Ltd., Shanghai) were
tetra- copolymers, which are complicated and costly for obtaining used without further treatment. All other chemicals that were reagent
higher scale-inhibition efficiency [9,14–16]. grade and directly used as received were from Chemical Reagent Co.
“Green” antiscalants, such as polyaspartic acid, polyalkylepoxysuccinic Ltd., Nanjing.
acid, and polyepoxysuccinic acid, have been receiving increasing atten- Instruments used in this research mainly included a Bruker model
tion owing to their non-phosphorus, low toxic, and biodegradable fea- IFS 66/S FTIR, a Bruker AVANCE model DRX-500 NMR spectrometer, a
tures [17–21]. These “green” antiscalants usually include abundant FEI Quanta 250 FEG SEM, and a Shimadzu model XRD-6000 X-ray dif-
hydroxyl and carboxyl groups, which play critical roles in scale control be- fractometer. The detected wavelength range in FTIR measurement is
cause of their excellent chelation, dispersion, and crystal distortion effects. from 600 cm−1 to 4000 cm−1. D2O is the solvent in 1H NMR measure-
Furthermore, polysaccharides, a type of very important natural polymers, ment at 500 MHz. SEM was carried out under a 5-kV acceleration volt-
such as starch, cellulose, chitosan, pectin etc., possess many advantages, age. XRD was operated at a voltage of 40 kV and a current of 30 mA,
i.e. not only the various active groups they contain, including hydroxyl using Cu Kα radiation (λ = 0.15418 nm).
and carboxyl groups, but also their some distinct characteristics, such as
environment-friendliness, widespread availability, biodegradability, 2.2. Preparation of CMS
and low cost. Also, due to containing abundant oxygen-containing
groups, these polysaccharides can be chemically modified and intro- CMS was synthesized based on the following method. Firstly, 8.0 g of
duced with various functional groups onto their backbones easily, starch and 4.0 g of NaOH were added into a 100 mL of 95% ethanol solu-
using graft, etherification, esterification, oxidation etc., to further im- tion, then they were incubated under mechanical stirring for 1.0 h in a
prove their application performance [22,23]. Some of works related to water bath at 50 °C to make the starch swelled and alkalized. A certain
polysaccharides and their derivatives that act as antiscalants have amount of monochloroacetic acid was dropwise added into the reaction
been reported [5,16,20,24–26]. phase. After a 4-h reaction at 50 °C, the mixture was adjusted to neutral
The molecular structure is known to account for the final application pH with hydrochloric acid aqueous solution, and then deposited in 95%
performance of materials [27–30]. On the basis of structural features ethanol. The obtained product was purified by filtration and rinse to re-
and final application performance of materials, the relationship be- move extra salt and unwanted byproducts, then vacuum-dried in an
tween structure and activity can be properly built. Accordingly, a proper oven at 60 °C for 48 h. Finally, the target product, CMS, was successfully
material can be selected or designed by precise structural control to synthesized and stored at room temperature [33]. By adjusting the
achieve the optimal application performance. As for polymeric molar feeding ratio of monochloroacetic acid to starch according to
antiscalants, their structural factors, such as the various types of func- Table 1, six CMS samples with different substitution degrees of
tional groups and their substitution degrees, as well as the molecular carboxymethyl groups were obtained, which were named accordingly
weight, are important for their scale-inhibition properties [18,21,27– as CMS(1) to CMS(6).
29,31]. However, until now, little work concerning the aforementioned
strategy has been reported on scale inhibition, especially for polysac- 2.3. Ultrasonic degradation of CMS samples
charide-based antiscalants.
Starch is a kind of popular and high-performance polysaccharides, Ultrasonic irradiation treatment of CMS(3) and CMS(5) was con-
which was recognized as a potential green antiscalant [32]. After suit- ducted using a JY 99-IID ultrasonic reactor (200 W, Ningbo Xinzhi Bio-
able modification, the scale-inhibition performance of starch-based ma- technology Co., Ltd.) to obtain two series of CMS samples with the
terials could be evidently improved. Among them, carboxymethyl
starch (CMS) is one of simplest starch derivatives, which contains abun- Table 1
dant hydroxyl and carboxyl groups on the chain backbone. Thus CMS Summary of the preparation conditions, structural parameters, and the scale-inhibition ef-
ficiency of various CMS samples with different substitution degrees of carboxymethyl
may own good scale-inhibition performance and have significant appli-
groups.
cation potentials as a commercial scale inhibitor. However, little work
concerning CMS used as a scale inhibitor has been reported and studied Samples Molar feeding ratio Substitution degree [η] (L/g) Scale-inhibition
of choroacetic acid of carboxymethyl efficiencya (%)
systematically until now. In this current work, a series of CMS samples
to starch groups
with different substitution degrees of carboxymethyl groups and molec-
CMS(1) 0.1:1 0.10 0.0277 13.03
ular weight was obtained. These well-prepared starch-based samples
CMS(2) 0.2:1 0.20 0.1046 17.39
were then characterized by Fourier transform infrared spectroscopy CMS(3) 0.5:1 0.48 0.1446 35.71
(FTIR), 1H nuclear magnetic resonance (1H NMR), and viscosity. The CMS(4) 0.9:1 0.63 0.2679 44.05
scale-inhibition performance of CMS was evaluated by a static test CMS(5) 1:1 0.74 0.3812 72.81
against calcium carbonate (CaCO3) in the synthetic water sample. The CMS(6) 2:1 0.95 0.4655 89.80

effects of structural factors of CMS on scale control efficiency were a

Scale-inhibition efficiency of various CMS samples at dose of 60 mg/L.
62 Y. Wang et al. / Desalination 408 (2017) 60–69

CMS(6)
same substitution degree of carboxymethyl groups but different molec- (a)
ular weight, through the adjustment of ultrasonic time. CMS solution
was thermostated at 20 °C in a 50-mL beaker during the whole irradia- CMS(5)
tion process. In order to reduce the experimental errors probably caused
by uneven transfer of power, the ultrasound source was dipped into the CMS(4)
CMS sample solution at an approximately 20-mm depth. Each irradiated
sample was prepared in three replicates.
CMS(3)
2.4. Viscosity measurement
CMS(2)
An Ubbelohde-type capillary viscometer with 0.413 mm of diameter
and 102.1 mm of length was employed for viscosity measurements. A CMS(1)
circulating water bath was used to control the measured temperature
at 25 ± 0.1 °C. The efflux time of various CMS aqueous solutions (tu) Starch
was detected, and that of solvent, pure water, (tv) was also measured
carefully. The relative viscosity of polymer solution (ηr) was roughly COO
-

equal to tu/tv. One-point method [34] was employed to estimate the in-
trinsic viscosity ([η]), which is an index for indirectly appraising the
4000 3500 3000 2500 2000 1500 1000 500
degradation degree and the molecular weight of the series of CMS sam-
ples with the same substitution degree of carboxymethyl groups but dif- -1
ferent ultrasonic time, as mentioned in Section 2.3 [35].
Wavenumber (cm )
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
h ffi
 i CH2COOH (b)
2  ηsp − ln ηr
½η ¼ ð1Þ
C

in which C (g/L) is the CMS concentration and ηsp is specific viscosity

equal to (ηr − 1). In all viscosity measurements, distilled water was
CMS(6)
used.
CMS(5)
2.5. Evaluation of scale-inhibition performance of CMS against the CaCO3
scale CMS(4)

On the basis of the National Standard of China (GB/T 16632-2008),
the static scale-inhibition test method was adopted to evaluate the abil-
ity of scale inhibition for CMS against CaCO3 precipitation. CaCl2 and
NaHCO3 aqueous mixtures were well prepared as the target water in a
250-mL Erlenmeyer flask, in which the molar concentrations of Ca2+
and HCO–3 are both 5 mmol/L. The resultant solution was consistently
mingled with a certain amount of CMS antiscalants. The specimens
were heated at 70 °C in a water bath for 10 h. After the reaction solution
cooled to room temperature, EDTA standard solution was used to titrate
and determine the concentration of Ca2 + in the supernatant, and a
blank test was conducted simultaneously to calculate the scale-inhibi-
tion efficiency of CMS against CaCO3 scale, as shown in formula (2) [18].
V 2 −V 1
Inhibition efficiency ð%Þ ¼  100%
V 0 −V 1
in which V0 (mL) is the consumed volume of EDTA solution for titration
of the synthetic water sample neither addition of scale inhibitors nor in-
cubation treatment, while V1 and V2 (mL) are the consumed volumes of
EDTA solution in the absence and in the presence of scale inhibitors after
a 10-h incubation at 70 °C respectively. The inhibition tests were all re-
peated at least 3 times and the final results were the average of three
runs. The relative errors of experimental results were below 5.0%.
3. Results and discussion
3.1. Preparation and characterization of CMS
CMS was prepared smoothly by etherification of starch, and the
etherifying agent was chloroacetic acid. The detailed preparation pro-
cess was described in the experimental section. The FTIR and 1H NMR
spectra of various CMS samples and starch are shown in Fig. 1. Com-
pared with the FTIR spectrum of starch in Fig. 1(a), the new appeared
CMS(3)
CMS(2)
CMS(1)
Starch
ppm 5 4 3 2 1 0
Fig. 1. (a) FTIR and (b) 1H NMR spectra of various starch-based samples.
characteristic peak at 1598 cm−1 in those of CMS samples was due to
carboxymethyl groups of CMS [36].
1
H NMR was also carried out to further confirm the structure of CMS,
ð2Þ
as shown in Fig. 1(b). Besides the peaks of starch from 3.3 ppm to
3.9 ppm and 5.3 ppm, corresponding to the various protons on the
starch backbone, a new proton signal at 3.97 ppm appeared in the
NMR spectra of all CMS samples, which belonged to the protons of the
methylene on the carboxymethyl group [36,37]. Characterization and
analysis of FTIR and 1H NMR spectra both confirmed that the CMS sam-
ples were successfully obtained. Besides, the peak area of the aforemen-
tioned new appeared NMR signal turns larger and larger with increasing
amount of monochloroacetic acid fed in the preparation process. More-
over, the substitution degrees of carboxymethyl groups in the CMS were
further calculated and obtained from their respective 1H NMR spectra
[37], as listed in Table 1.
Moreover, the intrinsic viscosity ([η]) of the various CMS samples
was also measured by one-point method [34] using the Ubbleholde vis-
cometry shown in Table 1. Based on Table 1, [η] of CMS samples in-
creased steadily from 0.0277 L/g to 0.4655 L/g, corresponding to the
growth of the substitution degree of carboxymethyl groups in various
CMS samples, from 0.10 of CMS(1) to 0.95 of CMS(6). [η] intrinsically re-
flects the hydrodynamic size of a polymer in a solution, which is
 Y. Wang et al. / Desalination 408 (2017) 60–69
influenced not only by the molecular weight of the polymer but also by
many other factors, such as distinct molecular structure and solvent
quality [34,38]. In this case, various CMS samples were obtained under
very similar preparation conditions but using different amounts of
monochloroacetic acid, which may have a similar molecular weight.
However, the hydrodynamic size of CMS increases with increased sub-
stitution degree of carboxymethyl groups in the measured range be-
cause of the enhanced intramolecular electrostatic repulsion in water,
resulting in a more extended molecular structure, higher hydrodynamic
size, and intrinsic viscosity in water [34,38].
3.2. Scale inhibition of CMS
3.2.1. Effects of dose
The series of CMS samples with varied substitution degrees of
carboxymethyl groups were used as antiscalants. The effects of CMS
dose on the scale-inhibition efficiency of CaCO3 were studied and pre-
sented in Fig. 2. According to Fig. 2, the dose of CMS antiscalants had
an important impact on the growth and formation of CaCO3 scale.
When the CMS dose increased from 5 mg/L to 60 mg/L, the scale-inhibi-
tion efficiency of CMS(5) and CMS(6) both increased dramatically; but
that of CMS(1), CMS(2), CMS(3) and CMS(4) increased quite slightly,
which may be due to relatively poor inhibition performance of these
compounds for their low substitution degree of carboxymethyl groups.
However, after the dose over 60 mg/L, the inhibition efficiency of all
CMS samples did not change or declined slightly with further increased
CMS dose. It meant that the optimal dose of various CMS samples was all
around 60 mg/L, at which their inhibition efficiency was listed in Table
1. The increase in scale-inhibition efficiency of the antiscalants, with
dose concentrations ranging from 5 mg/L to 60 mg/L, was due to the in-
creased addition of CMS, resulting in more carboxylate groups that can
bond more Ca2+ ions [14]. Another reason may be that the increase in
calcium binding to CMS with increasing polymer concentration promot-
ed the dissolution equilibrium of CaCO3 and hence decreased the possi-
bility of calcite formation [39]. On the other hand, the reduction of scale
inhibition at overdose (above 60 mg/L) may be due to the flocculation
effect of polymeric inhibitor [1,18,40]. Moreover, the enhancement of
interactions among polymers, especially that of H-bonding of CMS,
when the dose of CMS was high, reduced the formation possibility of
calcium-CMS complexation, which in effect may decrease scale-inhibi-
tion efficiency [41].
3.2.2. Effects of substitution degree of carboxymethyl groups
In addition to the antiscalant dose, which is an external factor, as
discussed above, structural factors of materials are also known to play
CMS(6)
100 CMS(5)
Inhibition efficiency (%)
CMS(4)
CMS(3)
80 CMS(2)
CMS(1)
60
40
20
0
0 20 40 60 80
Dose (mg/L)
Fig. 2. Effects of dose on the scale-inhibition efficiency of various CMS samples at 70 °C and
pH 8.0.
63
more important effects on application performance [27–30]. Substitu-
tion degree of carboxymethyl groups is a vital structural factor of CMS.
According to Fig. 2, the inhibition efficiency of various CMS samples
with different substitution degree generally increased with increased
substitution degree at the same dose.
Moreover, on the basis of Fig. 2 and the aforementioned discussion
about dose effects, various CMS samples had a similar optimal dose of
approximately 60 mg/L, at which their inhibition efficiency was sum-
marized in Table 1. The inhibition efficiency of different CMS samples
increased with increased substitution degree from 0.10 to 0.95. This
trend may reflect the important relationship between scale-inhibition
efficiency and the density of carboxymethyl groups on the CMS chains.
Higher substitution degree of carboxymethyl groups in CMS indicated
more carboxylic groups and greater anionic charge density, thus leading
to more possibility and efficiency of chelation and complexation effects
between CMS and calcium ions via electrostatic interaction [42]. This re-
sult was also consistent with that reported by other researchers [27,29].
Qiang et al. [27] investigated the scale-inhibition performance of a series
modified collagens and found that polymers with higher substitution
degree of carboxyls exhibited higher inhibitory performance. Verraest
et al. [29] also concluded that carboxymethyl inulin with a high substi-
tution degree (substitution degree N 1.0) was the most effective crystal-
lization inhibitor.
3.2.3. Effect of molecular weight
In addition to the substitution degree of carboxymethyl groups, the
molecular weight is another important structural factor of CMS. To in-
vestigate the effects of molecular weight of CMS, CMS(3) and CMS(5)
were selected and degraded by ultrasound treatment, to prepare two
series of CMS samples with different molecular weight but the same
substitution degree of carboxymethyl groups. Because each series of
CMS had similar substitution degree of carboxymethyl groups and mo-
lecular structure, [η] reflects the molecular weight of polymers here [34,
38]. The changes in [η] with different ultrasonic time are presented in
Fig. 3. From Fig. 3, [η] of the two ultrasonic treated samples decreased
rapidly in the early 10 min, then gradually tended to a constant with fur-
ther increased ultrasonic time. The macromolecular chains of CMS(3)
and CMS(5) were inferred to be broken in the process of ultrasound,
resulting in the continuous decline of intrinsic viscosity and molecular
weight of CMS antiscalants.
The scale-inhibition performance of the two series of degraded CMS
samples, CMS(3) and CMS(5), was then tested at various antiscalant
doses, and the relevant results are presented in Fig. 4. According to
Fig. 4, the dose effects of various CMS samples with different molecular
weights on scale-inhibition efficiency had a similar tendency to those of
CMS with different substitution degrees of carboxymethyl groups, as
0.40
CMS(3)
0.38 CMS(5)
0.36
[η] (L/g)
0.10
0.05
100 0.00
0 20 40 60 80 100 120
Ultrasonic time (min)
Fig. 3. Ultrasonic degradation of CMS(3) and CMS(5).
64 Y. Wang et al. / Desalination 408 (2017) 60–69

50 50 100

Inhibition efficiency (%)

CMS(3) CMS(3) (a) 90
(a) 40 80

Optimal dose (mg/L)

Inhibition efficiency (%)

40
70
30 60
30 50
20 40
20 CMS(3) [η]=0.1444
30
CMS(3) [η]=0.1281
CMS(3) [η]=0.1251 10 20
CMS(3) [η]=0.1085
10 10
CMS(3) [η]=0.0918
CMS(3) [η]=0.0800 0 0
CMS(3) [η]=0.0729 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15
0 [η] (L/g)
0 20 40 60 80 100
Dose (mg/L) 100 100

Inhibition efficiency (%)

100 90 CMS(5) (b) 90

Optimal dose (mg/L)

80 80
90 CMS(5) (b)
Inhibition efficiency (%)

70 70
80
60 60
70
50 50
60
40 40
50 30 30
CMS(5) [η]=0.3812
40 CMS(5) [η]=0.3671 20 20
CMS(5) [η]=0.3643
30 10 10
CMS(5) [η]=0.3620
20 CMS(5) [η]=0.3600 0 0
0.345 0.350 0.355 0.360 0.365 0.370 0.375 0.380 0.385
CMS(5) [η]=0.3590
10
CMS(5) [η]=0.3501 [η] (L/g)
0
0 20 40 60 80 100
Fig. 5. the scale-inhibition efficiency of various (a) CMS(3) and (b) CMS(5) samples with
Dose (mg/L) different molecular weight at their optimal doses based on Figs. 3 and 4.

Fig. 4. Effects of molecular weight of (a) CMS(3) and (b) CMS(5) on their scale-inhibition
efficiency.
shown in Fig. 2. The scale-inhibition efficiency also exhibited an up–cli-
max–down variation, and an optimal antiscalant dose was ascribed to
the mentioned flocculation effects [1,18,40] and H-bonding formation
in CMS [41].
Fig. 5 also summarizes both the maximal scale-inhibition efficiency
and the corresponding optimal dose of various CMS(3) and CMS(5) ul-
trasonic samples respectively, which had different molecular weights
but similar substitution degrees of carboxymethyl groups. Different
from the effects of substitution degree of carboxymethyl groups, the
maximal scale-inhibition performance of each series of CMS samples
had almost no change before and after ultrasonic treatment. The scale-
inhibition efficiency of a series of CMS(3) fluctuated between 33.33%
and 38.10%, whereas that of CMS(5) was between 68.09% and 72.81%.
Interestingly, the optimal dose of CMS(3) and CMS(5) both declined
dramatically from 60 mg/L to 10 mg/L after 5 min or 10 min ultrasonic
degradation. This result indicated that both structural factors of CMS,
substitution degree of carboxymethyl groups and molecular weight,
can significantly affect scale control ability, but by different mechanisms
and various apparent effects. Based on the above discussion, CMS with
higher substitution degree of carboxymethyl groups had enhanced che-
lating effects with Ca2+, resulting in higher inhibition efficiency (Fig. 2
and Table 1). As for the effects of molecular weight, the lower the molec-
ular weight of CMS was, with the same density of carboxymethyl groups
on the starch backbone, the greater the mobility of antiscalants was.
Antiscalants are facile to collision and complex with scale ions via elec-
trostatic interaction. Moreover, since the CaCO3 crystal itself possesses a
positive charge in the solution, whereas carboxymethyl groups of CMS
have a negative charge, CMS with a smaller molecular weight can be
more easily adsorbed onto the active growth sites of the CaCO3 crystal
surface and enter the crystal lattices, further disturbing the normal
growth of CaCO3 crystal [41]. These effects resulted in a higher inhibitor
utilization efficiency and the lower optimal dose of CMS with a smaller
molecular weight.
3.3. Investigation of scale-inhibition mechanism
3.3.1. SEM observation
To further investigate the scale-inhibition mechanisms, the mor-
phologies of CaCO3 scale deposits were directly observed by SEM. Fig.
6 shows the SEM images of CaCO3 scales obtained in the absence and
in the presence of various CMS inhibitors. From Fig. 6(a), CaCO3 crystals
not treated by antiscalants showed a calcite structure with regular
rhombohedrum shape, glossy surface, and large size approximately 6–
8 μm. On the other hand, CaCO3 obtained in the presence of scale inhib-
itors showed irregular shapes, rough surfaces, and small particles size
around 1–3 μm based on Fig. 6(b–g). The morphological investigation
on CaCO3 scale crystals with the existence of the CMS samples indicated
that the normal growth process of CaCO3 crystals appeared to be dis-
turbed or blocked by the antiscalants, and that the mechanism for con-
trolling scale growth and formation may be due to the irreversible
adsorption of CMS at the active growth site of crystals [43]. More de-
tailedly, according to Figs. 6(b–g), with increased substitution degree
of CMS samples, the shapes of CaCO3 scales, in the presence of CMS(1)
and CMS(2) with low substitution degree of carboxymethyl groups,
mostly showed relatively regular spherical, in which still contained a
 Y. Wang et al. / Desalination 408 (2017) 60–69 65

rhomb
m ic
rhombic

rhombic

(a) Blank: prepared CaCO3 (b) Substitution degree=0.1

rhombic
rhomb
m ic
irregular

spherical

spherical

(c) Substitution degree=0.2 (d) Substitution degree=0.48

irregular spherical irregular

spherical
al

(e) Substitution degree =0.63 (f) Substitution degree =0.74

irregular
ar

(g) Substitution degree =0.95

Fig. 6. SEM images of CaCO3 prepared from CaCl2 and NaHCO3 aqueous mixtures (a) in the absence of antiscalants and in the presence of various CMS samples with different substitution
degrees of caboxymethyl groups: (b) 0.1, (c) 0.2, (d) 0.48, (e) 0.63, (f) 0.74, and (g) 0.95, at the dose of 60 mg/L.

few rhombohedrum ones (Figs. 6(b–c)); But CaCO3 scale with regular
rhombohedrum shape was never observed but more irregular scales ap-
peared in Figs. 6(d–e); Moreover, most of CaCO3 scales, with the exis-
tence of CMS(5) and CMS(6) owning high substitution degree of
carboxymethyl groups, turned irregular and amorphous (Figs. 6(f–g)).
It could qualitatively conclude that the obtained CaCO3 scales became
more and more irregular, coarse, and amorphous with the increase of
substitution degree of CMS, indicating that CMS with higher density of
carboxymethyl groups was more effective in chelating with Ca2+ and
distorting crystal growth. This result was fully consistent with previous
inhibition test shown in Fig. 2.
Figs. 7 and 8 exhibit the SEM images of CaCO3 scales obtained in the
presence of two series of CMS antiscalants, CMS(3) and CMS(5), with
the same substitution degree of carboxymethyl groups but different
molecular weight at their respective optimal dose, as in Fig. 5. When
various CMS samples were added into the CaCl2 and NaHCO3 aqueous
mixtures, the shapes of the CaCO3 scales became irregular and spherical,
as presented in Figs. 7(b–h) and 8(b–h), because of the crystal distortion
effect caused by antiscalants [18]. Namely, the inhibitor adsorbed onto
the active growth sites of the CaCO3 crystal surfaces and inhibited the
normal and regular growth of the crystal lattice. Based on Figs. 7(b–h)
and 8(b–h), the spherical shapes of the CaCO3 crystal scales were
66 Y. Wang et al. / Desalination 408 (2017) 60–69

(a) Blank: prepared CaCO3 (b) T=0 min [η]=0.1444 L/g

(c) T=3 min [η]=0.1281 L/g (d) T=5 min [η]=0.1251 L/g

(e) T=10 min [η]=0.1085 L/g (f) T=30 min [η]=0.0918 L/g

(g) T=60 min [η]=0.0800 L/g (h) T=120 min [η]=0.0729 L/g
Fig. 7. SEM images of CaCO3 prepared from CaCl2 and NaHCO3 aqueous mixtures (a) in the absence of antiscalants and in the presence of various CMS(3) samples with different intrinsic
viscosity previously ultrasonic treated at various time: (b) 0 min, (c) 3.0 min, (d) 5.0 min, (e) 10 min, (f) 30 min, (g) 60 min, and (h) 120 min, at respective optimal dose based on Fig. 5.

roughly similar to each other in the presence of various CMS
antiscalants with different molecular weight. It may be due to the fact
that each series of CMS ultrasonic samples, which have different molec-
ular weights but similar substitution degrees of carboxymethyl groups,
had very close scale-inhibition efficiency (Fig. 5).
3.3.2. XRD analysis
Apart from the direct SEM observations, characterization of CaCO3
scale crystals was also studied by XRD to identify the crystal phases
and further investigate the scale-inhibition mechanism. Figs. 9, 10, and
11 show the XRD spectra of CaCO3 crystals prepared under various con-
ditions. CaCO3 has three types of polymorphic forms, namely, calcite,
aragonite and vaterite, among which calcite is the most thermodynam-
ically stable form whereas vaterite is the least [18,19,42,44,45]. Figs.
9(a–b), 10(a–b), and 11(a–b) show that the blank CaCO3 samples
mainly exist in the form of calcite due to the evident diffraction peaks
at 23.04°, 29.47°, 36.05°, 39.48°, 43.19°, 47.54° and 48.53° based on re-
ported literatures [18,19,42,44,45]. On the other hand, the crystal
phase changed into the most unstable form of vaterite largely after the
addition of various CMS antiscalants owing to the different diffraction
peaks at 24.90°, 27.05°, 32.78°, 43.85°, and 50.08° [18,19,42,44,45], as
shown in Figs. 9(c–h), 10(c–i), and 11(c–i). The change of crystal
forms demonstrated that CMS contributed to the distortion effect of
the CaCO3 crystal growth.
As in Figs. 9(c–h), with the steady increase in substitution degree of
CMS antiscalants, from CMS(1) to CMS(6), the strength of the diffrac-
tion peaks corresponding to the calcite form of CaCO3 scale became in-
creasingly weaker, whereas the strength of the peaks due to the
vaterite form simultaneously became more evident. It indicated that
the CaCO3 scales obtained in the presence of CMS with higher density
 Y. Wang et al. / Desalination 408 (2017) 60–69 67

(a) Blank: prepared CaCO3 (b) T=0 min [η]=0.3812 L/g

(c) T=3 min [η]=0.3671 L/g (d) T=5 min [η]=0.3643 L/g

(e) T=10 min [η]=0.3620 L/g (f) T=30 min [η]=0.3600 L/g

(g) T=60 min [η]=0.3590 L/g (h) T=120 min [η]=0.3501 L/g
Fig. 8. SEM images of CaCO3 prepared from CaCl2 and NaHCO3 aqueous mixtures (a) in the absence of antiscalants and in the presence of various CMS(5) samples with different intrinsic
viscosity previously ultrasonic treated at various time: (b) 0 min, (c) 3.0 min, (d) 5.0 min, (e) 10 min, (f) 30 min, (g) 60 min, and (h) 120 min, at respective optimal dose based on Fig. 5.

of carboxymethyl groups were more unstable and more difficultly de-
posited, showing higher scale-inhibition efficiency (Fig. 2). Similarly,
as in Figs. 10(c–i) and 11(c–i), after the addition of various CMS(3)
and CMS(5) samples, the strength of the diffraction peaks ascribed
to calcite forms of CaCO3 all became weaker while that owing to the
vaterite form was stronger. However, the change tendency in the
strength of the aforementioned diffraction peaks showed the indepen-
dence of molecular weight of CMS, since each series of CMS ultrasonic
samples, with different molecular weights but similar substitution de-
grees of carboxymethyl groups, had very close scale-inhibition effi-
ciency (Fig. 5). The aforementioned results were fully consistent
with the scale-inhibition performance of various CMS antiscalants
(Figs. 2 and 4) and further confirmed that these starch-based mate-
rials were efficient antiscalants.
Based on the analysis of SEM images and XRD spectra (Figs. 6–11), ad-
dition of CMS scale inhibitors may disturb the crystal growth habits and
distort the crystal lattice, resulting in the changes in the crystal morphol-
ogy of the scale deposits. Therefore, the scale became floppy instead of
being in the original form of a hard scale, and was easily removed in
water. Moreover, CMS with higher substitution degree of carboxymethyl
groups and lower molecular weight owned an enhanced distortion of the
crystal growth and improved scale control properties.
68 Y. Wang et al. / Desalination 408 (2017) 60–69

C-Calcite C-Calcite
V-Vaterite V-Vaterite
CMS(5)
VV C V V V
C
(h) CMS(6) VV V C C V V
(i) T=120 min [η]=0.3501 L/g
(g) CMS(5) (h) T=60 min [η]=0.3590 L/g
(f) CMS(4) (g) T=30 min [η]=0.3600 L/g
(e) (f) T=10 min [η]=0.3620 L/g
CMS(3) T=5 min [η]=0.3643 L/g
(d) (e)
CMS(2) (d) T=3 min [η]=0.3671 L/g
(c) CMS(1)
C C C C C C
(b) Prepared CaCO3 (c) C T=0 min [η]=0.3812 L/g
Standardized CaCO 3 C C C C C
(a) (b) Prepared CaCO3
(a) Standardized CaCO3
0 20 40 60 80
2θ (°) 0 20 40 60 80
2θ (°)
Fig. 9. The XRD patterns of the (a) standardized CaCO3 crystals and CaCO3 prepared from
CaCl2 and NaHCO3 aqueous mixtures (b) in the absence of antiscalants and in the presence
Fig. 11. The XRD patterns of the (a) standardized CaCO3 crystals and CaCO3 prepared from
of various CMS samples with different substitution degrees of caboxymethyl groups: (c)
CaCl2 and NaHCO3 aqueous mixtures (b) in the absence of antiscalants and in the presence
0.1, (d) 0.2, (e) 0.48, (f) 0.63, (g) 0.74, and (h) 0.95, at the dose of 60 mg/L.
of various CMS(5) samples with different intrinsic viscosity previously ultrasonic treated
at various time: (c) 0 min, (d) 3.0 min, (e) 5.0 min, (f) 10 min, (g) 30 min, (h) 60 min,
4. Conclusion and (i) 120 min, at respective optimal dose based on Fig. 5.

Various carboxymethyl starch antiscalants with different substitu- of the distortion of the crystal lattice via chelating and complexing ef-
tion degrees of carboxymethyl groups and molecular weight were ob- fects between antiscalants and scale ions, resulting in the changes in
tained by the carboxymethylation method and ultrasound treatment. the crystal morphology of the scale deposits from a stable and hard cal-
The structural effects of CMS, i.e., substitution degree of caboxymethyl cite form to an unstable and floppy vaterite one. Thus CMS with a higher
groups and molecular weight, on the scale-inhibition performance substitution degree of carboxymethyl groups shows higher scale-inhibi-
were investigated in detail by a static test. With increased substitution tion efficiency. However, the starch-based antiscalant with a lower mo-
degree of caboxymethyl groups, the scale-inhibition efficiency of CMS lecular weight has a greater mobility and more easily enters the crystal
greatly increased, whereas the optimal dose incurred almost no change, lattices, further disturbing the normal growth of CaCO3 crystal, resulting
at approximately 60 mg/L. Moreover, the reduction of scale inhibition at in its higher inhibitor utilization efficiency and the lower optimal dose.
overdose may be due to the nature of the flocculation behavior of CMS However, in comparison with some conventional antiscalants, such
and the enhancement of interactions among polymers. As for the effects as phosphates and phosphonic acids, of which optimal doses are usually
of molecular weight, two series of CMS samples with the same substitu- around 2–10 mg/L with inhibition efficiency above 90% [46], those
tion degree of carboxymethyl groups but different molecular weight, conventional antiscalants own higher performance in scale inhibition
CMS(3) and CMS(5), were obtained by ultrasound degradation. than CMS reported in this work. But, from the perspective of ecological
Interestingly, with decreased molecular weight of CMS, the scale-inhibi- environment and sustainable development, the utilization of “green”
tion efficiency had no evident improvement, whereas the required antiscalants with non-phosphorus, low toxic, and biodegradable fea-
antiscalant dose was evidently reduced from 60 mg/L to 10 mg/L after tures, such as CMS, must be more cost-effective and have wide applica-
5–10 min of ultrasound treatment. The aforementioned two structural tion prospects. Thus, for improvement of the scale-inhibition properties
factors of CMS may obey different inhibition mechanisms. SEM observa- of the “green” antiscalants, on the one hand, the “green” antiscalants
tion and XRD analysis were further employed to have a better under- could be composited with conventional ones to reduce the phosphorus
standing of the scale-inhibition mechanisms of CMS. The CaCO3 crystal discharge; On the other hand, more advanced modification technolo-
growth was disturbed after the addition of CMS scale inhibitors because gies should be developed to further optimize and improve the perfor-
mance of those “green” antiscalants for finally abandoning the use of
C-Calcite phosphorus-containing antiscalants completely.
V-Vaterite CMS(3)
References
C
(i) T=120 min [η]=0.0729 L/g [1] P. Shakkthivel, R. Sathiyamoorthi, T. Vasudevan, Development of acrylonitrile copol-
VV V C C C C
(h) T=60 min [η]=0.0800 L/g ymers for scale control in cooling water systems, Desalination 164 (2004) 111–123.
(g) T=30 min [η]=0.0918 L/g [2] B.R. Zhang, L. Zhang, F.T. Li, W. Hu, P.M. Hannam, Testing the formation of Ca–phos-
(f) T=10 min [η]=0.1085 L/g phonate precipitates and evaluating the anionic polymers as Ca–phosphonate pre-
(e) T=5 min [η]=0.1251 L/g cipitates and CaCO3 scale inhibitor in simulated cooling water, Corros. Sci. 52
(d) T=3 min [η]=0.1281 L/g (2010) 3883–3890.
[3] H.Y. Li, W. Ma, L. Wang, R. Liu, L.S. Wei, Q. Wang, Inhibition of calcium and magne-
(c) T=0 min [η]=0.1444 L/g sium-containing scale by a new antiscalant polymer in laboratory tests and a field
C C CC C C
(b) trial, Desalination 196 (2006) 237–247.
Prepared CaCO 3
(a) [4] P. Shakkthivel, T. Vasudevan, Acrylic acid-diphenylamine sulphonic acid copolymer
Standardized CaCO3 threshold inhibitor for sulphate and carbonate scales in cooling water systems, De-
salination 197 (2006) 179–189.
0 20 40 60 80 [5] K. Chauhan, R. Kumar, M. Kumar, P. Sharma, G.S. Chauhan, Modified pectin-based
2θ (°) polymers as green antiscalants for calcium sulfate scale inhibition, Desalination
305 (2012) 31–37.
[6] B. Nowack, Environmental chemistry of phosphonates, Water Res. 37 (2003)
Fig. 10. The XRD patterns of the (a) standardized CaCO3 crystals and CaCO3 prepared from 2533–2546.
CaCl2 and NaHCO3 aqueous mixtures (b) in the absence of antiscalants and in the presence [7] D. Lisitsin, Q.F. Yang, D. Hasson, R. Semiat, Inhibition of CaCO3 scaling on RO mem-
of various CMS(3) samples with different intrinsic viscosity previously ultrasonic treated branes by trace amounts of zinc ions, Desalination 183 (2005) 289–300.
at various time: (c) 0 min, (d) 3.0 min, (e) 5.0 min, (f) 10 min, (g) 30 min, (h) 60 min, [8] E.G. Darton, Membrane chemical research: centuries apart, Desalination 132 (2000)
and (i) 120 min, at respective optimal dose based on Fig. 5. 121–131.
 Y. Wang et al. / Desalination 408 (2017) 60–69
[9] X.R. Guo, F.X. Qiu, K. Dong, X.S. Rong, K.C. He, J.C. Xu, D.Y. Yang, Preparation and ap-
plication of copolymer modified with the palygorskite as inhibitor for calcium car-
bonate scale, Appl. Clay Sci. 99 (2014) 187–193.
[10] M. Sedlak, M. Antonietti, H. Colfen, Synthesis of a new class of double-hydrophilic
block copolymers with calcium binding capacity as builders and for biomimetic
structure control of minerals, Macromol. Chem. Phys. 199 (1998) 247–254.
[11] Z. Amjad, P.G. Koutsoukos, Evaluation of maleic acid based polymers as scale inhib-
itors and dispersants for industrial water applications, Desalination 335 (2014)
55–63.
[12] Y.Z. Zhao, L.L. Jia, K.Y. Liu, P. Gao, H.H. Ge, L.J. Fu, Inhibition of calcium sulfate scale by
poly (citric acid), Desalination 392 (2016) 1–7.
[13] M. Dietzsch, M. Barz, T. Schüler, S. Klassen, M. Schreiber, M. Susewind, N. Loges, M.
Lang, N. Hellmann, M. Fritz, K. Fischer, P. Theato, A. Kühnle, M. Schmidt, R. Zentel, W.
Tremel, PAA-PAMPS copolymers as an efficient tool to control CaCO3 scale forma-
tion, Langmuir 29 (2013) 3080–3088.
[14] X.R. Guo, F.X. Qiu, K. Dong, X. Zhou, J. Qi, Y. Zhou, D.Y. Yang, Preparation, character-
ization and scale performance of scale inhibitor copolymer modification with chito-
san, J. Ind. Eng. Chem. 18 (2012) 2177–2183.
[15] Y.H. Gao, L.H. Fan, L. Ward, Z.F. Liu, Synthesis of polyaspartic acid derivative and
evaluation of its corrosion and scale inhibition performance in seawater utilization,
Desalination 365 (2015) 220–226.
[16] X.R. Guo, F.X. Qiu, K. Dong, K.C. He, X.S. Rong, D.Y. Yang, Scale inhibitor copolymer
modified with oxidized starch: synthesis and performance on scale inhibition,
Polym.-Plast. Technol. Eng. 52 (2013) 261–267.
[17] Y. Xu, L.L. Zhao, L.N. Wang, S.Y. Xu, Y.C. Cui, Synthesis of polyaspartic acid–melamine
grafted copolymer and evaluation of its scale inhibition performance and dispersion
capacity for ferric oxide, Desalination 286 (2012) 285–289.
[18] J.X. Chen, L.H. Xu, J. Han, M. Su, Q. Wu, Synthesis of modified polyaspartic acid and
evaluation of its scale inhibition and dispersion capacity, Desalination 358 (2015)
42–48.
[19] D. Liu, W.B. Dong, F.T. Li, F. Hui, J. Lédion, Comparative performance of
polyepoxysuccinic acid and polyaspartic acid on scaling inhibition by static and
rapid controlled precipitation methods, Desalination 304 (2012) 1–10.
[20] H.X. Zhang, F. Wang, X.H. Jin, Y.C. Zhu, A botanical polysaccharide extracted from
abandoned corn stalks: modification and evaluation of its scale inhibition and dis-
persion performance, Desalination 326 (2013) 55–61.
[21] D. Hasson, H. Shemer, A. Sher, State of the art of friendly “Green” scale control inhib-
itors: a review article, Ind. Eng. Chem. Res. 50 (2011) 7601–7607.
[22] F. Zhu, Composition, structure, physicochemical properties and modifications of cas-
sava starch, Carbohydr. Polym. 122 (2015) 456–480.
[23] W. Shen, S.Y. Chen, S.K. Shi, X. Li, X. Zhang, W.L. Hu, H.P. Wang, Adsorption of Cu(II)
and Pb(II) onto diethylenetriamine-bacterial cellulose, Carbohydr. Polym. 75 (2009)
110–114.
[24] H.X. Zhang, D.X. Sun, Y.C. Zhu, Preparation of carboxymethyl-quaternized oligochitosan
and its scale inhibition and antibacterial activity, J. Water Reuse Desalination 4 (2014)
65–75.
[25] H.P. Zhang, X.G. Luo, X.Y. Lin, P.P. Tang, X. Lu, M.J. Yang, Y.H. Tang, Biodegradable
carboxymethyl inulin as a scale inhibitor for calcite crystal growth: molecular
level understanding, Desalination 381 (2016) 1–7.
[26] L.Z. Lakshtanov, N. Bovet, S.L.S. Stipp, Inhibition of calcite growth by alginate,
Geochim. Cosmochim. Acta 75 (2011) 3945–3955.
69
[27] X.H. Qiang, Z.H. Sheng, H. Zhang, Study on scale inhibition performances and inter-
action mechanism of modified collagen, Desalination 309 (2013) 237–242.
[28] C.E. Fu, Y.M. Zhou, H.T. Xie, L. Cang, Research progress in the studies on the struc-
ture-activity relationship for scale inhibitors in circulating cooling water systems,
Ind. Water Treat. 30 (2010) 30–37.
[29] D.L. Verraest, J.A. Peters, H.V. Bekkum, G.M.V. Rosmalen, Carboxymethyl inulin: a
new inhibitor for calcium carbonate precipitation, J. Am. Oil Chem. Soc. 73 (1996)
55–62.
[30] H.G. Elias, Macromolecules, Structure and Properties, vol. 1, Plenum Press, New
York, 1984.
[31] A. Jada, R. Ait Akbour, C. Jacquemet, J.M. Suau, O. Guerret, Effect of sodium
polyacrylate molecular weight on the crystallogenesis of calcium carbonate, J.
Cryst. Growth 306 (2007) 373–382.
[32] E.R. McCartney, A.E. Alexander, The effect of additives upon the process of crystalli-
zation I. crystallization of calcium sulfate, J. Colloid Sci. 13 (1958) 383–396.
[33] M. Huang, Y.W. Wang, J. Cai, J.F. Bai, H. Yang, A.M. Li, Preparation of dual-function
starch-based flocculants for the simultaneous removal of turbidity and inhibition
of Escherichia coli in water, Water Res. 98 (2016) 128–137.
[34] R.S. Cheng, Extrapolation of viscosity data and calculation of intrinsic viscosity from
one concentration of solution viscosity, Polym. Bull. Chin. 3 (1960) 159–163.
[35] C.S. Zhou, H.L. Ma, Ultrasonic degradation of polysaccharide from a red algae
(Porphyra yezoensis), J. Agric. Food Chem. 54 (2006) 2223–2228.
[36] H. Song, D. Wu, R.Q. Zhang, L.Y. Qiao, S.H. Zhang, S. Lin, J. Ye, Synthesis and applica-
tion of amphoteric starch graft polymer, Carbohydr. Polym. 78 (2009) 253–257.
[37] H.J. Li, T. Cai, B. Yuan, R.H. Li, H. Yang, A.M. Li, Flocculation of both kaolin and hema-
tite suspensions using the starch-based flocculants and their floc properties, Ind.
Eng. Chem. Res. 54 (2015) 59–67.
[38] P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, 1953.
[39] M.F. Butler, N. Glaser, A.C. Weaver, M. Kirkland, M. Heppenstall-Butler, Calcium car-
bonate crystallization in the presence of biopolymers, Cryst. Growth Des. 6 (2006)
781–794.
[40] B. Senthilmurugan, B. Ghosh, S.S. Kundu, M. Haroun, B. Kameshwari, Maleic acid
based scale inhibitors for calcium sulfate scale inhibition in high temperature appli-
cation, J. Pet. Sci. Eng. 75 (2010) 189–195.
[41] L.C. Wang, K. Cui, L.B. Wang, H.X. Li, S.F. Li, Q.L. Zhang, H.B. Liu, The effect of ethylene
oxide groups in alkyl ethoxy carboxylates on its scale inhibition performance, Desa-
lination 379 (2016) 75–84.
[42] S. Kırboga, M. Öner, The inhibitory effects of carboxymethyl inulin on the seeded
growth of calcium carbonate, Colloids Surf. B 91 (2012) 18–25.
[43] G. Li, S.H. Guo, J.W. Zhang, Y. Liu, Inhibition of scale buildup during produced-water
reuse: optimization of inhibitors and application in the field, Desalination 351
(2014) 213–219.
[44] C. Wang, S.P. Li, T.D. Li, Calcium carbonate inhibition by a phosphonate-terminated
poly(maleic-co-sulfonate) polymeric inhibitor, Desalination 249 (2009) 1–4.
[45] H.C. Wang, Y.M. Zhou, Q.Z. Yao, S.S. Ma, W.D. Wu, W. Sun, Synthesis of fluorescent-
tagged scale inhibitor and evaluation of its calcium carbonate precipitation perfor-
mance, Desalination 340 (2014) 1–10.
[46] D.R. Li, An Overview of Water Treatment Chemicals, Chemical Industry Press, Bei-
jing, 2005 147–169.