Green and High Effective Scale Inhibitor Based on Ring-Opening Graft Modiﬁcation of Polyaspartic Acid

: Polyaspartic acid (PASP)-based green scale inhibitor has great potential application in water treatment. Here, we ﬁrst synthesized PASP in ionic liquid. Then, an effective PASP-based green scale inhibitor was synthesized by ring-opening graft modiﬁcation of PASP with both aspartic acid (ASP) and monoethanolamine (MEA). Its chemical composition was characterized by gel chromatography (GPC), Fourier transform infrared spectroscopy (FTIR), and 1H nuclear magnetic resonance ( 1 H NMR). Scale inhibition efﬁciency was measured by static scale inhibition tests. The results showed that the new PASP-based scale inhibitor has high scale inhibition to both CaCO 3 and Ca 3 (PO 4 ) 2 . When the concentration was increased to 2 mg/L, the inhibition efﬁciency of the new PASP-based scale inhibitor was 99% for CaCO 3 , while when the concentration was raised to only 4 mg/L, its inhibition efﬁciency increased to 100% for Ca 3 (PO 4 ) 2 . Scanning electronic microscopy (SEM) and X-ray diffraction (XRD) were used to analyze the changes of crystal structure for CaCO 3 and Ca 3 (PO 4 ) 2 after adding the new PASP-based scale inhibitor. The crystal size of CaCO 3 and Ca 3 (PO 4 ) 2 became smaller and the crystal form became amorphous after adding the modiﬁed PASPs compared with adding pure PASP. Moreover, the modiﬁed PASP showed good biodegradation performance. for the crystals of CaCO 3 and Ca 3 (PO 4 ) 2 , while the hydroxyl group may result in distorting crystal lattices. In general, the synergy effect of the double-functional-group-modiﬁed PASP showed high scale inhibition performance for CaCO 3 and Ca 3 (PO 4 ) 2 , and it also showed good biodegradation performance. g, 0.044 g, 0.027 g, 0.011 g) and MEA (0.025 g, 0.020 g, 0.013 g, 0.005 g) were added to the mixture together. The reaction was carried out at 0 ◦ C for 24 h under stirring. The reddish-brown viscous solid was obtained after the precipitate was washed with ethanol. The relevant synthetic reaction is expressed in Scheme 3. High performance liquid chromatography (HPLC) was used to detect the surplus amount of ASP, and gas chromatography (GC) was used to detect the surplus amount of MEA.


Introduction
In recent years, with the increase of worldwide population and expansion of modern society, the consumption and pollution of industrial water increases shortage of water resources significantly [1][2][3]. Water treatment is an important technology to deal with the shortage of water resources [4][5][6]. Among various methods explored for water treatment, the circulating cooling water system is widely utilized in industrial processes [7,8]. However, scale deposition is one of the major problems in this system [9]. Scales degrade the performance of heat exchangers by increasing the resistance to heat transfer and eventually result in tremendous economic loss due to energy waste by increasing the power requirements of pumps. In order to control scale formation, a variety of scale inhibitors have been explored and used widely in cooling water systems. Scale inhibitors are usually nonpolymeric phosphonates (ATMP, HEDP, PBTC) and polymers with functional groups. The most common polymers are phosphonate, carboxylate, and sulfonate polymers. However, phosphonate and sulfonate polymers are being restricted due to environmental legislation [10][11][12]. With the increase of environmental awareness, environmentally friendly scale inhibitors have been drawing broad attention all over the world [13][14][15][16][17].
Polyaspartic acid (PASP) is a promising environmentally friendly scale inhibitor with nontoxicity and good biodegradability. It has excellent scale inhibiting performance for CaCO 3 . However, its scale inhibiting performance for Ca 3 (PO 4 ) 2 is poor. In order to improve the comprehensive scale inhibition performance of PASP, it is essential to modify prove the comprehensive scale inhibition performance of PASP, it is essential to by introducing functional groups into its side chain, such as hydroxyl groups, c groups, sulfonic groups, and/or phosphonyl groups [18,19].
Many studies have already reported on PASP modification. For example, al. synthesized PASP/Urea by introducing an acylamino group into the side PASP. It was shown that the modified PASP has better scaling inhibition to Ca Chen et al. synthesized Ser-PASP via introducing a hydroxylic group into the s of PASP. When Ser-PASP was added, the scale crystals became irregular, and inhibition to CaSO4 was achieved [21]. Xu et al. obtained PASP-melamine graft ymer via polysuccinimide (PSI) ring-opening by melamine [22]. Xu et al. syn PASP/aminobenzenesulfonic acid (ABSA) copolymer using sulphanilic acid an was found that the PASP/ABSA copolymer is able to efficiently inhibit CaCO3 hough single-functional-group-modified PASP has shown improved scale inh CaCO3 and CaSO4, it is still a challenge to achieve high scale inhibition to Ca3(PO Here, we prepared a new green and highly effective PASP-based scale inh water treatment (Figure 1). PASP was first synthesized in ionic liquid. Ionic liqu subset of molten salts with melting points at or below 100 °C [24]. Spurred by t chemistry movement, ionic liquids are considered as promising alternative so replace traditional volatile organic compounds (VOCs) due to their low volat Then, a ring-opening graft modification strategy was adopted to modify the syn PASP with both aspartic acid (ASP) and monoethanolamine (MEA). The ASP a provide a carboxyl and hydroxyl group to PASP, respectively. The introduced can enhance the calcium chelation ability and solubilization for the crystals of Ca Ca3(PO4)2, while the hydroxyl group may result in distorting crystal lattices. In the synergy effect of the double-functional-group-modified PASP showed high hibition performance for CaCO3 and Ca3(PO4)2, and it also showed good biodeg performance.

Ring-Opening Graft Modification of PASP
We first synthesized PASP in ionic liquid. Ionic liquid has unique advan cluding good thermal stability, strong solubility, nonvolatility, free design of a anion, etc. It can replace traditional organic solvents and catalysts to achieve gr thesis of PASP [26]. Then, a ring-opening graft modification strategy was adopted ify PASP with both carboxyl and hydroxyl groups. Here, we put emphasis on d the graft modification process on the grafting ratio of ASP and MEA. The graftin both molecules are largely determined by the molar ratio of reagent (PSI-ASP-M action time, and reaction temperature ( Figure 2). First, the grafting ratio showed icant increase with the increase of molar ratio of ASP and MEA (Figure 2a). W molar ratio of the reagent (PSI-ASP-MEA) was fixed at 1:1:0, the grafting ratio of 74.0%, while when the molar ratio of the reagent (PSI-ASP-MEA) was fixed at

Ring-Opening Graft Modification of PASP
We first synthesized PASP in ionic liquid. Ionic liquid has unique advantages, including good thermal stability, strong solubility, nonvolatility, free design of anion and anion, etc. It can replace traditional organic solvents and catalysts to achieve green synthesis of PASP [26]. Then, a ring-opening graft modification strategy was adopted to modify PASP with both carboxyl and hydroxyl groups. Here, we put emphasis on discussing the graft modification process on the grafting ratio of ASP and MEA. The grafting ratio of both molecules are largely determined by the molar ratio of reagent (PSI-ASP-MEA), reaction time, and reaction temperature ( Figure 2). First, the grafting ratio showed a significant increase with the increase of molar ratio of ASP and MEA (Figure 2a). When the molar ratio of the reagent (PSI-ASP-MEA) was fixed at 1:1:0, the grafting ratio of ASP was 74.0%, while when the molar ratio of the reagent (PSI-ASP-MEA) was fixed at 1:0:1, the grafting ratio of MEA was 89.8%. The difference of the grafting ratio for the two molecules may be ascribed to the difference of the steric hindrance effect between them. When successively increasing the molar ratio of ASP or MEA, the grafting ratio of ASP and MEA obviously increased. This can be ascribed to the increased reaction probability during the whole reaction process. Typically, when the molar ratio of the reagent (PSI-ASP-MEA) was fixed at 1:0.5:0.5, the grafting ratio of ASP and MEA was probably 36.22% and 43.3%, respectively. Then, when increasing the reaction time from 12 h to 24 h, the increase first brought a significant enhancement of the grafting ratio for ASP from 32.8% to 36.2% and for MEA from 39.8% to 43.4%. This may be because there were more reactive sites generated during the initial reaction from 12 h to 24 h (Figure 2b). However, little change to the grafting ratios of either ASP or MEA occurred when further increasing the reaction time from 24 h to 60 h ( Figure  2b). This indicates that the reaction reached equilibrium at 24 h. As shown in Figure 2c, the grafting ratio of both ASP and MEA decreased significantly when increasing the reaction temperature from 0 • C to 40 • C. Specifically, the grafting ratio of ASP decreased from 38.8% to 31.9%, and the grafting ratio of MEA decreased from 43.2% to 24.2%. As is well known, higher temperature usually leads to an increased number of active molecules, increased effective collision, and higher reaction rate. It should theoretically lead to an increased grafting ratio of both ASP and MEA during the fixed time. However, because the ring-opening graft modification is an exothermic reaction, the increased temperature inhibited the reaction process [27] and led to hydrolysis of the formed graft bond. Thus, the grafting ratio of both ASP and MEA inevitably decreased.
Catalysts 2021, 11, 802 3 of 12 grafting ratio of MEA was 89.8%. The difference of the grafting ratio for the two molecules may be ascribed to the difference of the steric hindrance effect between them. When successively increasing the molar ratio of ASP or MEA, the grafting ratio of ASP and MEA obviously increased. This can be ascribed to the increased reaction probability during the whole reaction process. Typically, when the molar ratio of the reagent (PSI-ASP-MEA) was fixed at 1:0.5:0.5, the grafting ratio of ASP and MEA was probably 36.22% and 43.3%, respectively. Then, when increasing the reaction time from 12 h to 24 h, the increase first brought a significant enhancement of the grafting ratio for ASP from 32.8% to 36.2% and for MEA from 39.8% to 43.4%. This may be because there were more reactive sites generated during the initial reaction from 12 h to 24 h ( Figure 2b). However, little change to the grafting ratios of either ASP or MEA occurred when further increasing the reaction time from 24 h to 60 h ( Figure 2b). This indicates that the reaction reached equilibrium at 24 h. As shown in Figure 2c, the grafting ratio of both ASP and MEA decreased significantly when increasing the reaction temperature from 0 °C to 40 °C. Specifically, the grafting ratio of ASP decreased from 38.8% to 31.9%, and the grafting ratio of MEA decreased from 43.2% to 24.2%. As is well known, higher temperature usually leads to an increased number of active molecules, increased effective collision, and higher reaction rate. It should theoretically lead to an increased grafting ratio of both ASP and MEA during the fixed time. However, because the ring-opening graft modification is an exothermic reaction, the increased temperature inhibited the reaction process [27] and led to hydrolysis of the formed graft bond. Thus, the grafting ratio of both ASP and MEA inevitably decreased.

Chemical Characterization of the Modified PASP
From Table 1, we can obtain the conclusion that the molecular mass of PASP synthesized in the ionic liquid was M n 4.87 kDa (PDI~1.63). Compared with the pure PASP, GPC results show that the molecular mass of the modified PASP increased with the increase of MEA or ASP proportion in the monomer solution. Particularly, when the molar ratio of PSI-ASP-MEA was 1:0.5:0.5 (at reaction temperature of 0 • C and reaction time of 24 h), the molecular mass of the modified PASP was increased to M n 4.90 kDa. The corresponding grafting ratios of ASP and MEA were calculated to be increased to 36.2% and 43.3%, respectively. This is reasonable, because increasing the molar ratio of ASP or MEA increases the grating ratio of the two molecules, inevitably enhancing the ring-opening graft modification. Thus, graft modification obviously increased the molecular weight of PASP. We then verified that the ASP and MEA were conjugated onto the PASP side chain by analyzing the molecular structure using FTIR ( Figure 3). The absorption peaks at 3422 cm −1 ,~1598 cm −1 and~1401 cm −1 are attributed to the stretching vibration of N-H and C=O in -CONH and the absorption peak at C-N [28,29]. This indicates that the PASP was successfully synthesized in the ionic liquid. In addition, for the modified PASP, a peak corresponding to the stretching vibration of the C=O group was observed at~1703 cm −1 , which indicates that the ASP was successfully grafted onto the side chain of PASP. Moreover, the stretching vibrations peak of the -C-O-group at~1164 cm −1 and~1064 cm −1 indicate that the MEA was successfully introduced to the side chain of PASP. In addition, 1H NMR was also used to verify the graft modification process ( Figure S1). Proton NMR results show that for the PASP main chain, the peaks corresponding to -CH-and -CH 2were observed at 4.45 ppm and 2.74 ppm, respectively. The peaks at 3.60~3.66 ppm and 3.16 ppm are attributed to -CH-and -CH 2 -of the side chain for carboxyl group-modified PASP (PASP-ASP). Similarly, the peaks at 3.8 ppm and 3.12 ppm are attributed to -CHand -CH 2 -of the side chain for hydroxyl group-modified PASP (PASP-MEA). Furthermore, we obtained all of the above characteristic peaks for the product after both the carboxyl and hydroxyl group modifications to PASP, which indicates that the PASP-ASP-MEA was successfully synthesized.

Scale Inhibition Performance of the Modified PASP
To investigate the scale inhibition properties of the modified PASP, we chose CaCO 3 and Ca 3 (PO 4 ) 2 as the two typical crystals. Figure 4a,b presents the scale inhibition performance of PASP-ASP-MEA against CaCO 3 and Ca 3 (PO 4 ) 2 scales. When its concentration was 1 mg/L and 2 mg/mL, the inhibition efficiency of PASP-ASP-MEA against CaCO 3 scale reached 95% and 99%, respectively, while its inhibition efficiency against Ca 3 (PO 4 ) 2 scale reached 75% and 89%, respectively. The maximum inhibition efficiency (100%) to both CaCO 3 and Ca 3 (PO 4 ) 2 was achieved at a scale inhibitor dosage of 4 mg/L. Compared to PASP, PASP-ASP, and PASP-MEA, PASP-ASP-MEA had better inhibition performance against CaCO 3 and Ca 3 (PO 4 ) 2 scales. The reasons may be listed as follows. Firstly, the sparingly soluble salts, such as CaCO 3 , Ca 3 (PO 4 ) 2 , and gypsum, are crystallized on the nano/microdust impurities in the aqueous solution [30]. Those nano/microdusts are not uniform and consist of different ingredients. The antiscalants PASP and their derivatives can block the nano/microdust crystallization centers of the salts. Therefore, the rate of crystallization of sparingly soluble salts decreases. Secondly, the grafted polymer has a better sorption on nano/microdust surface than a non-grafted one [31]. Because the nano/microdusts are not uniform, it is reasonable that some fractions of the dusts are better blocked by PASP-ASP and others by PASP-MEA. Thus, synergy becomes inevitable due to PASP-ASP-MEA combining the properties of PASP-ASP and PASP-MEA.
sponding to the stretching vibration of the C=O group was observed at ~1703 cm −1 , whic indicates that the ASP was successfully grafted onto the side chain of PASP. Moreove the stretching vibrations peak of the -C-O-group at ~1164 cm −1 and ~1064 cm −1 indica that the MEA was successfully introduced to the side chain of PASP. In addition, 1H NM was also used to verify the graft modification process ( Figure S1). Proton NMR resul show that for the PASP main chain, the peaks corresponding to -CH-and -CH2-were ob served at 4.45 ppm and 2.74 ppm, respectively. The peaks at 3.60~3.66 ppm and 3.16 ppm are attributed to -CH-and -CH2-of the side chain for carboxyl group-modified PASP (PASP ASP). Similarly, the peaks at 3.8 ppm and 3.12 ppm are attributed to -CH-and -CH2-of th side chain for hydroxyl group-modified PASP (PASP-MEA). Furthermore, we obtained a of the above characteristic peaks for the product after both the carboxyl and hydroxyl grou modifications to PASP, which indicates that the PASP-ASP-MEA was successfully synth sized.

Scale Inhibition Performance of the Modified PASP
To investigate the scale inhibition properties of the modified PASP, we chose CaCO3 and Ca3(PO4)2 as the two typical crystals. Figure 4a,b presents the scale inhibition performance of PASP-ASP-MEA against CaCO3 and Ca3(PO4)2 scales. When its concentration was 1 mg/L and 2 mg/mL, the inhibition efficiency of PASP-ASP-MEA against CaCO3 scale reached 95% and 99%, respectively, while its inhibition efficiency against Ca3(PO4)2 scale reached 75% and 89%, respectively. The maximum inhibition efficiency (100%) to both CaCO3 and Ca3(PO4)2 was achieved at a scale inhibitor dosage of 4 mg/L. Compared to PASP, PASP-ASP, and PASP-MEA, PASP-ASP-MEA had better inhibition performance against CaCO3 and Ca3(PO4)2 scales. The reasons may be listed as follows. Firstly, the sparingly soluble salts, such as CaCO3, Ca3(PO4)2, and gypsum, are crystallized on the nano/microdust impurities in the aqueous solution [30]. Those nano/microdusts are not uniform and consist of different ingredients. The antiscalants PASP and their derivatives can block the nano/microdust crystallization centers of the salts. Therefore, the rate of crystallization of sparingly soluble salts decreases. Secondly, the grafted polymer has a better sorption on nano/microdust surface than a non-grafted one [31]. Because the nano/microdusts are not uniform, it is reasonable that some fractions of the dusts are better blocked by PASP-ASP and others by PASP-MEA. Thus, synergy becomes inevitable due to PASP-ASP-MEA combining the properties of PASP-ASP and PASP-MEA. The CaCO3 and Ca3(PO4)2 scale deposits were observed by scanning electron microscope (SEM). As shown in Figures 5 and 6, in the absence of PASP and the modified PASPs, CaCO3 scale deposits show a calcite structure with regular shape and glossy surface, and Ca3(PO4)2 scale deposits have an irregular polygon-shaped structure, not parallel to the surface to bulk. In comparison, when PASP and the modified PASPs were introduced into the solution, the shapes of CaCO3 and Ca3(PO4)2 scale deposits were irregular, and their crystalline grain tended to be finer. Both PASP and the modified PASPs (PASP-ASP, PASP-MEA and PASP-ASP-MEA) showed extremely good inhibition effects on CaCO3 and Ca3(PO4)2 scales. Particularly, PASP-ASP-MEA showed the smallest crystal scale deposits, and it had the best inhibition effects on CaCO3 and Ca3(PO4)2 scales. The synergy of the introduced hydroxyl and carboxylic groups of PASP-ASP-MEA can enhance the blocking of the nano/microdust crystallization centers of CaCO3 and Ca3(PO4)2. Therefore, the rate of crystallization of CaCO3 and Ca3(PO4)2 decreased, which finally resulted in the smaller crystals [30][31][32]. of the nano/microdust crystallization centers of CaCO 3 and Ca 3 (PO 4 ) 2 . Therefore, the rate of crystallization of CaCO 3 and Ca 3 (PO 4 ) 2 decreased, which finally resulted in the smaller crystals [30][31][32].   Figure 7a, which are characteristic peaks of aragonite, and there are also diffraction peaks of 29.4° and 43.2° for calcite. These results indicate that in the absence of PASP and the modified PASPs, the calcium carbonate precipitate is the mixture of aragonite, which is the main crystal form, and some calcite [33]. In the other spectra (Figure 7b Figure 7a, which are characteristic peaks of aragonite, and there are also diffraction peaks of 29.4° and 43.2° for calcite. These results indicate that in the absence of PASP and the modified PASPs, the calcium carbonate precipitate is the mixture of aragonite, which is the main crystal form, and some calcite [33]. In the other spectra (Figure 7b-e), diffraction peaks (24.9°, 27.1°, 32.8°, 43.9°, and 50.1°) corresponding to vaterite are very strong, which demonstrates that vaterite is the main crystal form in the presence of PASP and the modified PASPs [34]. Obviously, the vaterite peaks are the weakest for PASP-ASP-MEA. The change of crystal forms indicates that PASP-ASP-MEA contributed to the distortion of CaCO3 crystals and showed the best inhibition effect on CaCO3 scales.  Figure 7a, which are characteristic peaks of aragonite, and there are also diffraction peaks of 29.4 • and 43.2 • for calcite. These results indicate that in the absence of PASP and the modified PASPs, the calcium carbonate precipitate is the mixture of aragonite, which is the main crystal form, and some calcite [33]. In the other spectra (Figure 7b [35], and the peak at 29.3° is obviously reduced in spectra b, c, d, and e. The addition PASP and the modified PASPs (PASP-ASP, PASP-MEA, and PASP-ASP-MEA) showed large influence on the crystal structure, and the diffraction peaks became quite weak aft the addition of PASP and the modified PASPs, which implies that the surface morpholog and particle size changed in the presence of the inhibitor. The diffraction peaks are th weakest for PASP-ASP-MEA, which shows that PASP-ASP-MEA had the best inhibitio effect on Ca3(PO4)2 scales.
According to the above SEM images and XRD analysis, the introduction of the mod ified PASP disturbed the crystal growth habits and distorted the lattice, which resulte changes in the crystal morphology of the precipitates. The synergy of the introduced fun tional groups of PASP-ASP-MEA made the scale crystal become much smaller compare with the effects of other inhibitors. Consequently, the scales become floppy and can b removed easily.      According to the above SEM images and XRD analysis, the introduction of the modified PASP disturbed the crystal growth habits and distorted the lattice, which resulted changes in the crystal morphology of the precipitates. The synergy of the introduced functional groups of PASP-ASP-MEA made the scale crystal become much smaller compared with the effects of other inhibitors. Consequently, the scales become floppy and can be removed easily.

Biodegradation Performance of the Modified PASP
The biodegradation property of the modified PASP was investigated, and the results are shown in Figure 9. Compared with the pure PASP, the modified PASP showed a decreased biodegradation rate. This may be attributed to the formed cross-linked chemical bonds in the modified PASP [36,37]. However, it still experienced an easy degradation process; more than 60 and 70 wt% was lost after 21 and 28 days of biodegradation, respectively. When the polymers were catalyzed by the bacteria, the cross-linking chain and their backbone were both degraded and ruptured. They were first degraded to oligomers and smaller molecules and then biocatalyzed to carbon dioxide and water [38,39]. The modified PASP showed good biodegradable performance; it is indubitably an environmentally friendly scale inhibitor.

Biodegradation Performance of the Modified PASP
The biodegradation property of the modified PASP was investigated, and the result are shown in Figure 9. Compared with the pure PASP, the modified PASP showed a de creased biodegradation rate. This may be attributed to the formed cross-linked chemica bonds in the modified PASP [36,37]. However, it still experienced an easy degradatio process; more than 60 and 70 wt% was lost after 21 and 28 days of biodegradation, respec tively. When the polymers were catalyzed by the bacteria, the cross-linking chain an their backbone were both degraded and ruptured. They were first degraded to oligomer and smaller molecules and then biocatalyzed to carbon dioxide and water [38,39]. Th modified PASP showed good biodegradable performance; it is indubitably an environ mentally friendly scale inhibitor.

Characterization
Fourier transform infrared spectrometry (FTIR) was carried out on a Perkin-Elme SpectrumGX Fourier transform infrared spectrometer (Waltham, MA, USA). Proton NMR spectra were recorded on a Bruker Ascend 400 MHz nuclear magnetic resonance (NMR spectrometer (Bruker, Billerica, MA, USA) using D2O as solvent. Gel chromatograph

Biodegradation Performance of the Modified PASP
The biodegradation property of the modified PASP was investigated, and th are shown in Figure 9. Compared with the pure PASP, the modified PASP show creased biodegradation rate. This may be attributed to the formed cross-linked c bonds in the modified PASP [36,37]. However, it still experienced an easy deg process; more than 60 and 70 wt% was lost after 21 and 28 days of biodegradation tively. When the polymers were catalyzed by the bacteria, the cross-linking ch their backbone were both degraded and ruptured. They were first degraded to ol and smaller molecules and then biocatalyzed to carbon dioxide and water [38, modified PASP showed good biodegradable performance; it is indubitably an mentally friendly scale inhibitor.

Characterization
Fourier transform infrared spectrometry (FTIR) was carried out on a Perkin-Elmer SpectrumGX Fourier transform infrared spectrometer (Waltham, MA, USA). Proton NMR spectra were recorded on a Bruker Ascend 400 MHz nuclear magnetic resonance (NMR) spectrometer (Bruker, Billerica, MA, USA) using D 2 O as solvent. Gel chromatography (GPC) (PL-GPC50, UK) was used to determine the molecular mass of polymers. Scanning electron microscopy (SEM, HITACHI TM3000, Tokyo, Japan) was used to capture the crystal structure of CaCO 3 and Ca 3 (PO 4 ) 2 . X-ray diffraction (XRD) was used to analyze the crystal form of CaCO 3 and Ca 3 (PO 4 ) 2 .

Synthesis of Polyaspartic Acid (PASP) in Ionic Liquid
L-aspartic acid (5 g) was dissolved in a three-necked flask. Ionic liquid 1-ethyl-3methylimidazol dihydrogen phosphate (EmimH 2 PO 4 ) (15 mL) was slowly added into the flask, and the mixture was reacted at 180 • C for 3 h. Afterwards, the mixture was poured into anhydrous ethanol to form the precipitate. Then, light yellow solid product was obtained. The product polysuccinimide (PSI) was filtrated and dry.
PSI (5 g) and 10 wt% hydroxide solution were added into a round bottom flask together. The mixture was stirred at 40 • C for 1 h. Then, the solution was filtered, washed with absolute alcohol, and dried. Finally, the reddish-brown product was obtained. The relevant synthetic reaction is expressed in Scheme 2.

Synthesis of Polyaspartic Acid (PASP) in Ionic Liquid
L-aspartic acid (5 g) was dissolved in a three-necked flask. Ionic liquid 1-eth thylimidazol dihydrogen phosphate (EmimH2PO4) (15 mL) was slowly added flask, and the mixture was reacted at 180 °C for 3 h. Afterwards, the mixture wa into anhydrous ethanol to form the precipitate. Then, light yellow solid product tained. The product polysuccinimide (PSI) was filtrated and dry.
PSI (5 g) and 10 wt% hydroxide solution were added into a round bottom gether. The mixture was stirred at 40 °C for 1 h. Then, the solution was filtered, with absolute alcohol, and dried. Finally, the reddish-brown product was obtai relevant synthetic reaction is expressed in Scheme 2.

Ring-Opening Graft Modification of PASP
PSI (2 g) and NH3-NH4Cl buffer solution (pH = 10, 20 mL) were mixed at 0 ° ASP (0.055 g, 0.044 g, 0.027 g, 0.011 g) and MEA (0.025 g, 0.020 g, 0.013 g, 0.005 added to the mixture together. The reaction was carried out at 0 °C for 24 h under The reddish-brown viscous solid was obtained after the precipitate was washed w anol. The relevant synthetic reaction is expressed in Scheme 3. High performan chromatography (HPLC) was used to detect the surplus amount of ASP, and gas tography (GC) was used to detect the surplus amount of MEA. The grafting ratio (GD, %) was calculated by the following Equation (1)

Ring-Opening Graft Modification of PASP
PSI (2 g) and NH 3 -NH 4 Cl buffer solution (pH = 10, 20 mL) were mixed at 0 • C. Then, ASP (0.055 g, 0.044 g, 0.027 g, 0.011 g) and MEA (0.025 g, 0.020 g, 0.013 g, 0.005 g) were added to the mixture together. The reaction was carried out at 0 • C for 24 h under stirring. The reddish-brown viscous solid was obtained after the precipitate was washed with ethanol. The relevant synthetic reaction is expressed in Scheme 3. High performance liquid chromatography (HPLC) was used to detect the surplus amount of ASP, and gas chromatography (GC) was used to detect the surplus amount of MEA.

Synthesis of Polyaspartic Acid (PASP) in Ionic Liquid
L-aspartic acid (5 g) was dissolved in a three-necked flask. Ionic liquid 1-ethy thylimidazol dihydrogen phosphate (EmimH2PO4) (15 mL) was slowly added i flask, and the mixture was reacted at 180 °C for 3 h. Afterwards, the mixture was into anhydrous ethanol to form the precipitate. Then, light yellow solid product w tained. The product polysuccinimide (PSI) was filtrated and dry.
PSI (5 g) and 10 wt% hydroxide solution were added into a round bottom f gether. The mixture was stirred at 40 °C for 1 h. Then, the solution was filtered, w with absolute alcohol, and dried. Finally, the reddish-brown product was obtain relevant synthetic reaction is expressed in Scheme 2.

Ring-Opening Graft Modification of PASP
PSI (2 g) and NH3-NH4Cl buffer solution (pH = 10, 20 mL) were mixed at 0 °C ASP (0.055 g, 0.044 g, 0.027 g, 0.011 g) and MEA (0.025 g, 0.020 g, 0.013 g, 0.005 g added to the mixture together. The reaction was carried out at 0 °C for 24 h under s The reddish-brown viscous solid was obtained after the precipitate was washed w anol. The relevant synthetic reaction is expressed in Scheme 3. High performance chromatography (HPLC) was used to detect the surplus amount of ASP, and gas c tography (GC) was used to detect the surplus amount of MEA. The grafting ratio (GD, %) was calculated by the following Equation (1): where m 1 is the feed amount of ASP or MEA and m 2 is the surplus amount of ASP or MEA.

Measurement of the Efficiency of Static Scale Inhibition
Static scale inhibition tests were performed according to Chinese National Standards GB/T 16632-2008 and GB/T 22626-2008. The experimental condition of scale inhibition to CaCO 3 was CaCl 2 (240 mg·L −1 ) mixed with NaHCO 3 (732 mg·L −1 ). The experimental condition of scale inhibition of Ca 3 (PO 4 ) 2 was CaCl 2 (100 mg·L −1 ) mixed with KH 2 PO 4 (5 mg·L −1 ). Borax buffer solution was used as the initial solution (pH = 9.0, 0.01 mol·L −1 ). Both brines passed filtration (220 nm filter membrane) before use. The reaction was carried out at 80 • C for 10 h with various amounts of scale inhibitors. When the reaction was finished, it was cooled to room temperature and filtered by filter paper. The filtrate was titrated with ethylene diamine tetraacetic acid (EDTA) standard solution to determine the concentration of Ca 2+ (CaCO 3 ).
The inhibition efficiency of CaCO 3 was calculated by the following Equation (2): where ρ 0 is the concentration of Ca 2+ before experiment, ρ 1 is the concentration of Ca 2+ in the absence of scale inhibitor in the solution, and ρ 2 is the concentration of Ca 2+ in the presence of scale inhibitor in the solution. The PO 4 3− concentration in the filtrate was detected by the ammonium molybdate spectrophotometric method. The inhibition efficiency of Ca 3 (PO 4 ) 2 was calculated by the following Equation (3): where ρ 0 is the concentration of PO 4 3− before experiment, ρ 1 is the concentration of PO 4 3− in the absence of scale inhibitor, and ρ 2 is the concentration of PO 4 3− in the presence of inhibitor.

Biodegradation of the Modified PASP
The biodegradability of the modified PASP was estimated based on Chinese National Standard GB/T 21803-2008. The sample was treated with the standard activated sludge, which was obtained from Nanjing High Tech University Biological Technology Research Institute Co. Ltd., at 30 ± 1 • C for 28 days. The concentration of the activated sludge was 3~5 g/L. The concentration of the modified PASP was 100 mg/L. Sodium acetate was used as the standard molecule to estimate the activity of the standard activated sludge. The chemical oxygen demand (COD) detection was used to evaluate the biodegradability of the modified PASP. The COD was measured using a thermostat, DRB200 (Hach Co. Ltd., Loveland, CO, USA) and a spectrophotometer, DR1010 (Hach Co. Ltd., Loveland, CO, USA).

Conclusions
A new PASP-based green scale inhibitor (PASP-ASP-MEA) was obtained by ringopening graft modification of PASP with both aspartic acid (ASP) and monoethanolamine (MEA). The modified PASP had excellent inhibition efficiency against CaCO 3 and Ca 3 (PO 4 ) 2 scales. When the concentration of PASP-ASP-MEA was increased to 2 mg/L, its inhibition efficiency increased to 99% against CaCO 3 and 89% against Ca 3 (PO 4 ) 2 . Inhibition efficiency of 100% against both CaCO 3 and Ca 3 (PO 4 ) 2 was achieved at a scale inhibitor dosage of 4 mg/L. The functional groups of PASP-ASP-MEA have good sorption on nano/microdust surface. PASP-ASP-MEA can therefore block the nano/microdust crystallization centers of CaCO 3 and Ca 3 (PO 4 ) 2 and significantly decrease the crystallization rate of these sparingly soluble salts. Compared with PASP-ASP and PASP-MEA, the synergy of the introduced groups (-OH and -CO 3 ) of PASP-ASP-MEA inevitably led better blocking of nano/microdusts. Thus, the scale crystals of CaCO 3 and Ca 3 (PO 4 ) 2 became much smaller, increasing the solubility of these calcium salts in water. Moreover, the modified PASP also shows good biodegradable performance. In short, it is a promising green scale inhibitor.