Nanosilica modified with polyaspartic acid as an industrial circulating water scale inhibitor

Given the special performance of nanosilica with its small size, large specific surface area and high surface activity, nanosilica containing reactive amino group (denoted as SiO2–NH2) and polysuccinimide were allowed to take part in polymerization reaction to afford SiO2–NH2 modified polyaspartic acid (denoted as SiO2–NH2/PASP), a potential polymer scale inhibitor with good water solubility for industrial circulating water. The scale inhibition performance of the as-prepared SiO2–NH2/PASP was evaluated by static scale inhibition test; and its scale inhibition mechanism was explored by means of scanning electron microscopy, X-ray diffraction, and X-ray photoelectron spectroscopy. Results indicated that SiO2–NH2/PASP exhibits excellent scale inhibition performance against CaSO4 and CaCO3 at very low concentrations (optimum scale inhibition rate of 100% and 68%, respectively), and the presence of 5 mg/L of SiO2–NH2/PASP greatly increases the inhibition efficiency of CaSO4 and CaCO3 scale by 21% and 53%, obviously higher than that of pure PASP.


INTRODUCTION
The pipeline inner walls of circulating cooling water system often face the challenge of inorganic scale deposition upon exposure to a large amount of poorly soluble inorganic salts, which could cause huge economic losses and operational problems in facilities such as reverse osmosis desalination system, multi-stage flash desalination system, and multi-effect distillation system [1][2][3] .
Therefore, it is imperative to introduce a scale inhibitor to get rid of the scale formation on the inner walls of the pipelines of industrial circulating water. Some plant extracts and phosphorous compounds are of particular significance for inhibiting scale formation 4,5 . For example, Gambier extract as a green inhibitor can efficiently inhibit the scale formation of calcium carbonate; and the natural extract of tobacco rob can function as a scale-and corrosion-inhibitor in artificial seawater 6,7 . The reason lies in that the plant extracts contain a large amount of amino and carboxyl functionalities which have the strong complexing ability with Ca 2+ . However, the huge amount of plant extracts could cause potentially deteriorating the ecological environment 8 ; and phosphorous compounds might lead to serious environmental risks of water eutrophication 9,10 . Therefore, the exploitation of environmentally friendly scale inhibitors for water treatment is of vital importance.
Polyaspartic acid (denoted as PASP) as an environmentally friendly additive is nontoxic, biodegradable, and free of phosphorus [10][11][12] . However, its scale inhibition performance is poor, which limits its application in the industry. Since the scale inhibition performance of polymer scale inhibitors is highly dependent on their carboxyl, hydroxyl, and acylamido groups with strong ionic chelating ability [13][14][15] , we introduce these functional groups into the PASP molecule through the copolymerization of L-aspartic acid or the ring-opening modification of polysuccinimide (denoted as PSI) to obtain functionalized PASP with improved scale inhibition performance. The copolymerization modification mainly involves the reaction of aspartic acid with other monomers, which, in combination with the introduction of other functional groups into the molecular chain of aspartic acid, could contribute to improving the scale inhibition performance to some extent. For example, Dong et al. 16 reported that curcumin, a fluorescent monomer, can be introduced into the PASP molecular chain in the presence of L-aspartic acid and citric acid; and the assynthesized scale inhibitor exhibits superior fluorescent-tagged properties and excellent scale inhibition performance. However, L-aspartic acid is expensive and faces challenges with industrial production. Fortunately, highly active linear PSI can be easily cyclized by amino to form PASP with side chains 17 , which provides a feasible pathway to yielding functionalized PASP. Chen et al. 18 reported an interesting exploration in this respect: they synthesized a scale inhibitor with PSI and histidine as the starting materials. Thanks to the introduction of an amino group and carboxyl group to inhibit the reaction crystallization of Ca-scale crystals, the as-synthesized PASP-based scale inhibitor has a promising anti-scaling ability. Similarly, Xu et al. 19 introduced hydroxyl into the side chain of PASP with PSI and 2-amino-2methyl-1-propanol (AMP) as raw materials. The as-obtained PASP/ AMP exhibits improved scale-and corrosion-inhibition performances, and its biodegradation rate is acceptable in terms of the Convention for the Protection of the Marine Environment of the Northeast Atlantic 20 . These researches demonstrate that introducing strong chelating groups into PASP molecular chain through ring-opening reaction is favorable for improving the scale inhibition performance; and the ring-opening reaction, with desired environmental acceptance, could be of special significance for the development of polymer-matrix scale inhibitor.
Viewing the abovementioned perspectives, here we attempt to introduce amino-functionalized silicon dioxide into the ringopening reaction of PSI, hoping to obtain a modified PASP scale inhibitor with improved scale inhibition performance. We pay special attention to amino-functionalized nano-silica (nano-SiO 2 ) with small size, large specific surface area, high surface activity, and multiple reaction sites, because it has been found to be intriguing in the petroleum industry, electronics industry and coating industry and so on [21][22][23][24] . The surface-capped nano-SiO 2 has little steric hindrance and contains many reaction sites, and it can react with PSI to generate SiO 2 -NH 2 /PASP containing organic functional groups that can chelate with calcium ions, thereby effectively preventing scaling. This article deals with the preparation of SiO 2 -NH 2 /PASP, a potential green scale inhibitor applicable to the industrial circulating water system; and it also reports its scale inhibition performance and mechanism.
RESULTS AND DISCUSSION Synthesis and structural characterization of SiO 2 -NH 2 /PASP The TEM image of SiO 2 -NH 2 nanoparticles is shown in Fig. S1. It can be seen that SiO 2 -NH 2 nanoparticles appear as aggregates and have a size of about 10-20 nm. The synthetic route of SiO 2 -NH 2 /PASP scale inhibitor is shown in Fig. 1. There is a large amount of amino group on the surface of SiO 2 -NH 2 ; and PASP is connected to the surface of nano-silica by CO-NH covalent bond. As a result, the as-synthesized SiO 2 -NH 2 /PASP product contains a large amount of amide group and carboxyl group that can inhibit the formation of scale.
The FTIR spectra of SiO 2 -NH 2 /PASP and PASP are shown in Fig.  2a. The stretching vibration absorbance peaks of N-H, C=O, and C-N in the amide bond appear at 3417 cm −1 , 1624 cm −1 , and 1402 cm −1 , respectively; and these FTIR data prove that PSI successfully participates in ring-opening reaction to generate PASP 25 . The asymmetric stretching vibration and stretching vibration peaks of Si-O-Si can be detected at 1053 and 872 cm −1 , respectively 26 . This indicates that SiO 2 -NH 2 is involved in the polymerization of PSI yielding SiO 2 -NH 2 /PASP. Figure 2b shows the 1 H NMR spectra of SiO 2 -NH 2 /PASP in D 2 O. The signals at 4.5 and 2.8 ppm are assigned to -CH-and -CH 2 -of SiO 2 -NH 2 / PASP, respectively. The FTIR and 1 H NMR data give good evidence of the successful synthesis of SiO 2 -NH 2 /PASP.
The EDS characterization and Si element mapping of SiO 2 -NH 2 / PASP are shown in Fig. S2. The Si peak is assigned to SiO 2 -NH 2 in the as-synthesized SiO 2 -NH 2 /PASP scale inhibitor (Fig. S2a). The uniform Si element distribution in SiO 2 -NH 2 /PASP indicates that SiO 2 -NH 2 nanoparticles do not agglomerate therein while they participate in the polymerization reaction (Fig. S2b).
To further verify the successful synthesis of SiO 2 -NH 2 /PASP scale inhibitor, we synthesized PASP under the same conditions for synthesizing SiO 2 -NH 2 /PASP. The as-obtained PASP was then mixed with SiO 2 -NH 2 under 10 min of mechanical stirring. The mixed dispersion of PASP and SiO 2 -NH 2 tends to sediment at the bottom of the beaker; and that of the as-synthesized SiO 2 -NH 2 / PASP is homogeneous, possibly due to the bonding of a large amount of PASP molecular chains on the surface of SiO 2 -NH 2 . This further proves that the ring-opening reaction of PSI in the presence of SiO 2 -NH 2 indeed gives rise to SiO 2 -NH 2 /PASP.
One milliliter of PASP and SiO 2 -NH 2 /PASP solutions of 0.0230 and 0.0166 g/mL were taken into two beakers, respectively, acidified to pH ≈2 with 0.1 mol/L HCl solution, then diluted and shaken with 10 mL distilled water, titrated with 0.104 mol/L NaOH standard solution. When 0.108 mL was added to each drop, the primary conductivity value was recorded. The titration curve is shown in Fig. S3. The carboxyl group contents of PASP and SiO 2 -NH 2 /PASP were calculated to be 15.06% and 15.59%, respectively.  Effect of SiO 2 -NH 2 /PASP against CaSO 4 and CaCO 3 scales Figure 3 shows the variations of scale inhibition efficiency with inhibitor concentration as well as static inhibition test temperature and time. Generally, the as-synthesized SiO 2 -NH 2 /PASP exhibits better scale inhibition performance than PASP; and its scale inhibition efficiency varies diversely with varying concentration as well as static inhibition test temperature and time. The inhibitory performance of SiO 2 -NH 2 /PASP with 5 mg/L was 21% higher than that of the unmodified PASP (Fig. 3a). Particularly, the scale inhibition efficiency of SiO 2 -NH 2 /PASP reaches the maximum of nearly 100% at a concentration of 6 mg/L. We further explore the influence of the dosage of PASP and SiO 2 -NH 2 /PASP on the inhibition rate of CaSO 4 , as is shown in Fig. S4. In the absence of a scale inhibitors, a large amount of calcium sulfate precipitate is formed. After the addition of PASP inhibitor, there is a little scale at the bottom of the conical flask; and there is nearly no scale after the introduction of 5 mg/L SiO 2 -NH 2 /PASP. This indicates that SiO 2 -NH 2 /PASP can effectively prevent the formation of the CaSO 4 scale. The reason might lie in that the carboxylic group of SiO 2 -NH 2 /PASP can coordinate with calcium ions to adsorb free calcium ions in the solution, and thus reduce the combination of Ca 2+ and sulfate ions, to prevent the generation of CaSO 4 scale 27 . As can be seen from Fig. S3, compared with PASP, the carboxyl group content in the molecular structure of SiO 2 -NH 2 /PASP is higher, indicating that SiO 2 -NH 2 /PASP exhibits more excellent scale inhibition performance.
Static inhibition test temperature also plays an important role in the scale inhibition performance. As shown in Fig. 3b, the scale inhibition efficiency of SiO 2 -NH 2 /PASP decreases rapidly with rising test temperature; and that of SiO 2 -NH 2 /PASP remains at nearly 100% in the temperature range of 70-75°C. Besides, SiO 2 -NH 2 /PASP with the optional concentration of 6 mg/L always exhibits much higher scale inhibition efficiency than PASP in the whole temperature range of static inhibition test; and the scale inhibition efficiency tends to decline at the higher static inhibition test temperature. Yang reported that "the nucleation rate of CaSO 4 scale can be influenced by temperature and supersaturation ratio (S) 28,29 . The temperature has little effect on S. When S remains constant, the higher the temperature is, the higher the nucleation rate is, and the easier scaling is 29,30 .
where A is a frequency factor, β is a geometric (shape) factor of 16 π/3 for the spherical nucleus, and ƒ(θ) is a correction factor, when heterogeneous nucleation occurs ƒ(θ) = 0.01 31 . V m is the molar volume of the phase forming, γ is the surface energy (J/m 2 ). S is the supersaturation ratio, k is the Boltzmann constant, T is the absolute temperature 29 ." Moreover, SiO 2 -NH 2 /PASP (at the optimal dosage of 6 mg/L) has excellent inhibition performance against the CaSO 4 scale over a long test time; and its inhibition efficiency stays nearly at 100% after 25 h of static scale inhibition test (Fig. 3c). It is possible that the chelation stability between PASP and calcium ions becomes worse with the increase of test time, leading to the decrease of scale inhibition efficiency. This demonstrates that SiO 2 -NH 2 /PASP exhibits excellent scale inhibition performance and could have promising potential for industrial water treatment. Figure 4 shows the static scale-inhibition performance of SiO 2 -NH 2 /PASP and PASP against calcium carbonate precipitate. The inhibition efficiency of SiO 2 -NH 2 /PASP on the CaCO 3 scale is significantly higher than that of pure PASP. Particularly, at a concentration of 5 mg/L, the complexation and solubilization effect of SiO 2 -NH 2 /PASP for CaCO 3 is increased by 53% as compared with those of the PASP scale inhibitor (Fig. 4a). This is caused by the chelation of the carboxyl group in the scale inhibitor molecule and calcium ions. The results in combination with Fig. S3 show that SiO 2 -NH 2 /PASP has more carboxyl group content and is easier to combine with the free calcium ions in the solution compared with PASP, thus inhibiting the formation of calcium carbonate scale. The scale inhibition efficiency initially increases with increasing concentration of SiO 2 -NH 2 /PASP; and the inhibition efficiency of SiO 2 -NH 2 /PASP is close to 68% under an inhibitor concentration of 30 mg/L. After that, the inhibition efficiency remains stable with increasing concentration, showing a so-called "threshold effect". This is because, on the one hand,  the solubilization performance of SiO 2 -NH 2 /PASP on calcium carbonate precipitate is enhanced with the increasing concentration of scale inhibitors. On the other hand, the adsorption capacity of the scale inhibitor and scale crystal is enhanced under increased concentration to reduce the formation of precipitate 32 .
The static scale inhibition test temperature and time also have some effects on the inhibition efficiency against the CaCO 3 scale. The inhibition performance of SiO 2 -NH 2 /PASP with 30 mg/L tends to gradually decrease with the increase of the test temperature (Fig. 4b); and its inhibition efficiency maintains at 100% in the test temperature range of 50-70°C. As the test temperature rises from 70 to 80°C, the inhibition efficiency of SiO 2 -NH 2 /PASP decreases from 100 to 68%, but it is still much higher than that of PASP. This indicates that the introduction of SiO 2 -NH 2 contributes to improving the scale inhibition performance of PASP at elevated static scale inhibition temperature. With the increase of temperature, the solubility of calcium carbonate decreases, and the formation of calcium carbonate crystals is easier. Meanwhile, temperature also affects the nucleation rate of calcium carbonate, the higher the temperature, the higher the nucleation rate 29 .
The effect of static inhibition test time on the inhibition efficiency against the CaCO 3 scale is shown in Fig. 4c. The inhibition efficiencies of SiO 2 -NH 2 /PASP decrease with the increase of time; and it is as much as 76% after 8 h of static scale inhibition test, higher than 50% of PASP under the same test condition. Therefore, it can be inferred that SiO 2 -NH 2 /PASP has a much stronger chelating ability with Ca 2+ than PASP.

Characterization of CaSO 4 and CaCO 3 scales
The crystal morphology exploration tests were repeated three times. The morphology of CaSO 4 scales formed under different concentrations of SiO 2 -NH 2 /PASP and PASP are shown in Fig. S5. The calcium sulfate formed in the presence of SiO 2 -NH 2 /PASP is of disordered, irregular, and patchy morphology; and that formed after the addition of PASP has a bar-shape morphology as well as increased particle size and length. Figure 5 shows the SEM images of calcium sulfate crystals in the absence and presence of SiO 2 -NH 2 /PASP scale inhibitor. The CaSO 4 crystals formed in the absence of antiscalants have a regular rod-shaped as well as smooth surface and compact crystal structure (Fig. 5a, a'). When 2 mg/L SiO 2 -NH 2 /PASP is added, CaSO 4 crystal with irregular shape and relatively rough surface is formed (Fig. 5d, d'); and CaSO 4 macrocrystals in good dispersion state are formed after the addition of 4 mg/L SiO 2 -NH 2 /PASP (Fig. 5e, e') 4,33 . As the concentration of PASP increases from 2 to 4 mg/L, strip-like CaSO 4 crystal with increased size is formed (Fig. 5b, c, b', c'). This demonstrates that the introduction of antiscalants, especially SiO 2 -NH 2 /PASP, leads to significant changes in the morphology as well as average length and diameter of CaSO 4 crystals, which could be because the carboxylic group not only can chelate with Ca 2+ to block the active growth point of CaSO 4 crystal but also can dwell on CaSO 4 microcrystal surfaces via adsorption to disperse the microcrystal via charge repulsion 10,34 . In other words, the inhibition effect of SiO 2 -NH 2 /PASP against the CaSO 4 scale is dominated by the chelating ability of SiO 2 -NH 2 /PASP and its adsorption on CaSO 4 microcrystal surfaces.
The SEM photographs of CaCO 3 crystals formed under different static scale inhibition test conditions are shown in Fig. 6. The CaCO 3 crystal obtained without scale inhibitor has a regular shape and smooth surface. With the increase of PASP concentration, the morphology of calcium carbonate becomes disordered and its particle size increases. With 10 mg/L SiO 2 -NH 2 /PASP, irregular CaCO 3 crystal forms, and its sharp edges and acute corners disappear. When the concentration of SiO 2 -NH 2 /PASP increases to 30 mg/L, the cubic shape of the calcium carbonate scale is destroyed, and its surface is relatively rough; and its shape and size are increased. This could be attributed to two antiscaling mechanisms. On the one hand, the antiscalants can prevent more scale-forming positive ions in the solution from being precipitated through complexation action. On the other hand, the antiscaling chemicals can interact with mineral nuclei to disrupt the crystallization process and keep the crystal particles dispersed in aqueous suspension, thereby inhibiting their sedimentation or adhesion 35,36 . Therefore, it can be inferred that multiple inhibition effects, including lattice distortion, chelation, and dispersion, work together to significantly inhibit the formation of the CaCO 3 scale.
The crystal structures of CaSO 4 and CaCO 3 scales are further analyzed by X-ray diffraction (XRD), and corresponding XRD patterns are shown in Fig. 7. In the absence of the scale inhibitors, the major XRD peaks at 2θ = 11.53°, 20.69°, 23.31°, and 29.02° (Fig.  7a) are assigned to gypsum (CaSO 4 ·2H 2 O). After the addition of SiO 2 -NH 2 /PASP or PASP, the resultant CaSO 4 does not undergo crystal transition, but its crystallization behavior and particle size change to some extent. This further confirms that SiO 2 -NH 2 /PASP functions to inhibit the formation of CaSO 4 scale mainly through chelation and adsorption, which is consistent with relevant SEM results. As to the aqueous solution of CaCO 3 , calcite and aragonite are the primary crystals formed in the absence of the scale inhibitors (Fig. 7b) 37 . After the addition of SiO 2 -NH 2 /PASP, the XRD signals of the (104) and (110) planes of calcite disappear, and the strong diffraction peak of vaterite emerges at 2θ = 29.21 o . Calcite is the most thermodynamically stable, whereas vaterite is the least stable 38,39 . This indicates that SiO 2 -NH 2 /PASP can inhibit the growth of calcium carbonate crystals and completely block the conversion path from amorphous calcium carbonate to calcite as compared with PASP. In other words, the SiO 2 -NH 2 /PASP scale inhibitor functions to inhibit the formation of CaCO 3 scale through chelating reaction in association with lattice distortion effect.

Scale inhibition mechanism of SiO 2 -NH 2 /PASP against CaSO 4 and CaCO 3
To further explore the adsorption characteristics of SiO 2 -NH 2 / PASP on scale surface, we conducted X-ray photoelectron spectroscopy (XPS) analyses of the CaSO 4 and CaCO 3 scales formed under static scale inhibition tests. The high-resolution Ca 2p-XPS spectra of CaSO 4 and CaCO 3 scales formed without the scale inhibitors and in the presence of PASP and SiO 2 -NH 2 /PASP are given in Figs. 8 and 9. In the absence of the scale inhibitors, the CaSO 4 scale shows the distinctive peaks of Ca 2p 1/2 and Ca 2p 3/2 at 351.42 and 347.84 eV, respectively (Fig. 8). After the addition of 4 mg/L PASP, the Ca 2p 1/2 and Ca 2p 3/2 peaks shift to 351.53 and 347.93 eV, which corresponds to a change in the chemical surroundings of Ca 2+ ion upon the introduction of PASP.
The Ca 2p 1/2 and Ca 2p 3/2 peaks further shift to 351.58 and 347.99 eV after the addition of 4 mg/L SiO 2 -NH 2 /PASP, which gives further evidence to the uptake of scale inhibitors on the surface of calcium sulfate scales. Similarly, the introduction of PASP and SiO 2 -NH 2 /PASP leads to shifting in the Ca 2p peaks of the CaCO 3 scale by 0.17 and 0.42 eV, respectively (Fig. 9). These findings imply that the SiO 2 -NH 2 /PASP scale inhibitor can be adsorbed on the surface of CaSO 4 and CaCO 3 scales to affect the growth, morphology, and size of scales, thereby effectively retarding their formation in aqueous solutions.
The inhibition mechanism of SiO 2 -NH 2 /PASP against CaSO 4 and CaCO 3 scales is shown in Fig. 10. In the absence of the scale inhibitors, calcium sulfate and calcium carbonate form stable scales on the pipe surface, as shown in Fig. 10a, b. When SiO 2 -NH 2 /PASP is dissolved in water, the negatively charged carboxylic group of SiO 2 -NH 2 /PASP coordinates with positively charged Ca 2+ ions to prevent them from forming calcium carbonate and sulfate. Meanwhile, SiO 2 -NH 2 /PASP can be   adsorbed on the surface of newly generated scale crystal to increase the inter-particle repulsive force and prevent intercrystalline bonding, thereby accommodating good dispersion of the scales and retarding their deposition on the inner walls of the pipes (Fig. 10c, d). As a result, SiO 2 -NH 2 /PASP can effectively inhibit the formation of CaSO 4 and CaCO 3 scales in aqueous solutions.
In summary, SiO 2 -NH 2 /PASP, an organic-inorganic hybrid as a potential green scale inhibitor for industrial circulating water, is synthesized via a facile liquid-phase polymerization route. Static scale inhibition tests demonstrate that SiO 2 -NH 2 /PASP exhibits much better inhibition performance against CaSO 4 and CaCO 3 scales than PASP, and has good environmental adaptability. This is because, as evidenced by SEM, XRD, and XPS analyses, the carboxylic group of SiO 2 -NH 2 /PASP can chelate with Ca 2+ to block the active growth points of the inorganic scales while its adsorption on scale crystal surface also contributes to retarding the formation of the scales. Furthermore, SiO 2 -NH 2 /PASP can completely change the conversion path from amorphous calcium carbonate to calcite, which could also account for its desired inhibition ability against the CaCO 3 scale. The as-prepared SiO 2 -NH 2 /PASP scale inhibitor with good environmental acceptance could be promising in preventing the formation of CaSO 4 and CaCO 3 scales on the inner walls of the pipes for industrial circulating water.

METHODS Materials
Reactive nanosilica (SiO 2 -NH 2 ) surface modified with amino (adsorption aperture: 19.2 nm) was provided by Henan Wangwu Nanotechnology Company Limited (Jiyuan, China). Industrial grade polysuccinimide (PSI, M w = 7000) was purchased from Wuhan Yuancheng Gongyi Technology Company Limited (Wuhan, China). Anhydrous calcium chloride, disodium edetate, anhydrous sodium sulfate, borax, and potassium hydroxide were purchased from Tianjin Komi Chemical Reagent Company Limited (Tianjin, China). Anhydrous sodium hydrogencarbonate was commercially obtained from Tianjin Deen Chemical Reagent Company Limited (Tianjin, China). Anhydrous ethanol was bought from Anhui Ante Food Company Limited (Suzhou, China). Deionized water (DI) was used as the solvent and for rinsing as well.

Preparation of SiO 2 -NH 2 /PASP
A certain amount of SiO 2 -NH 2 was dispersed in the mixed solution of ethanol and DI. The as-obtained dispersion (0.013 g) was placed into a round bottom flask, followed by the addition of 0.984 g of PSI under stirring at 60°C for 1 h. The resultant mixed dispersion was allowed to react for 24 h thereat while its pH was adjusted to be 8-9. Upon completion of the reaction, the product was precipitated with absolute ethanol for 12 h and dried at 60°C for 24 h to afford SiO 2 -NH 2 /PASP solid product. PASP was prepared in the same manner while no SiO 2 -NH 2 was used 19 .
The carboxyl content of PASP and SiO 2 -NH 2 /PASP was determined by conductance titration. First, excess HCl solution was added to acidize PASP and SiO 2 -NH 2 /PASP, then titrated with NaOH standard solution, and the volume and conductivity values of the consumed NaOH standard solution were recorded 40 . The conductivity curve is shown in Fig. S6. The carboxyl group content (W) can be calculated by Eq. (2) 40 M is the concentration of the NaOH solution, mol/L; V 1 is the volume of the NaOH standard solution consumed by excess hydrochloric acid, mL; V 2 is the volume of the NaOH standard solution consumed by excess hydrochloric acid and the carboxyl group contained in the polymer, mL; V 2 -V 1 is the volume of the standard solution of NaOH consumed by the carboxyl group in the polymer, mL; W is the mass of the sample, g 40 .

Static scale inhibition tests
The antiscalant performance of SiO 2 -NH 2 /PASP against calcium sulfate was tested by the static scale inhibition method. Briefly, proper amounts of CaCl 2 (6800 mg/L Ca 2+ ) and Na 2 SO 4 (7100 mg/L of SO 4 2− ) were mixed at 70°C for 6 h; and the filtrate was titrated with ethylene diamine tetraacetic acid (EDTA) standard solution to determine the concentration of calcium ion. The scale inhibition efficiency (E, %) against CaSO 4 is calculated as where X 0 (mg/L) is the initial concentration of Ca 2+ in the water samples before heating, X 2 and X 1 (mg/mL) are the concentration of Ca 2+ after reaction in the presence and absence of scale inhibitor, respectively. The static inhibition tests were repeated three times, and the average of the repeat tests was cited to minimize data scattering. An aqueous solution containing 240 mg/L Ca 2+ and 732 mg/L HCO 3 − was prepared and heated with a water bath at 80°C for 10 h. At the end of heating, the solution was cooled to room temperature; and the Ca 2+ concentration in the supernatant was determined by EDTA titration. The scale inhibition efficiency (η, %) against CaCO 3 is calculated as where C 0 (mg/mL) is the concentration of Ca 2+ in the solution before the static inhibition test, C 1 and C 2 (mg/mL) are the concentration of Ca 2+ in the absence and presence of scale inhibtor, respectively. All the static inhibition tests were conducted in triplicate to ensure reproducibility.