1.Introduction

Industrial wastewater as worldwide environmental concerns has attracted substantial attention due to their negative impact on the environment and human health[1-4].It is important to explore more effective approaches to remove the contaminants in environmental media[5-7].One of the most useful methods for wastewater primary treatment is flocculation,which has been widely used to remove suspended particles by aggregating small or fine particles into larger-sized ones because of its simplicity,high efficiency and low cost[8,9].However,the flocculation efficiency and stability need to be improved owing to the increasing wastewater treatment demand[10,11].Bridging flocculation can be much stronger than those coagulated by simple salts or electrolyte via charge neutralization[12,13].Therefore,the aggregates formed by polymeric flocculants seem to be significantly more resistant to breakage[14].

Currently,compared with inorganic flocculants such as polymer aluminum chloride(PAC),polymeric ferric sulfate(PFS),organic flocculants including polyacrylamide(PAM),cationic polyacrylamide(CPAM),and some natural polymer modified flocculants such as starch and chitosan have been commonly used in wastewater treatment[15].Among them,natural polymer flocculants have aroused much attention because of their biodegradable property and no secondary pollution[16-18].Cellulose,the most abundant biomolecule on the planet,is a promising alternative to other biodegradable flocculants[19-21].Targeted modification of cellulose can be employed to introduce the desired functional groups to prepare the effective flocculants[22,23].Yan et al.[24]employed microcrystalline cellulose to react with glycidyltr imethyl ammonium chloride to obtain quaternized celluloses as eco-friendly flocculants.Nourani et al.prepared anionic cellulose sulfate flocculant by hydrolysis and sulfonation of cellulose with chlorosulfonic acid[25].

In this work,microcrystalline cellulose was exploited as the skeleton ofthe flocculant.The conventional cellulosecontaining flocculant with ammonium ceric nitrate or sodium periodate to open six-ring structure is likely to damage cellulose skeleton[26,27].Herein,we report a new method to adopt a reactive silane coupling agent bearing vinyl groups as the linkage between the cellulose and grafted cationic polymers which was synthesized via the copolymerization of cationic monomer with the reactive silane.The resulting cellulosebased flocculants allowed the covalent bonding between cationic polymer chains and cellulose fibers as the core,thus creating a novel star-like flocculants tolerant to high shearing or turbulence during the application.Three flocculants with different length and quantity of cationic branch chains have been synthesised.To assess the flocculation performance of the as-synthesized flocculants,the suspension of fine clay particles was selected as a model contaminant system,and the flocculation efficiency was determined mainly based on the turbidity measurements under various conditions.The flocculation performance was also compared and contrasted with commercial polyaluminium chloride(PAC)[28-30].

2.Experimental

2.1.Materials

γ-methacryloxypropyl trimethoxy silane(KH-570)was purchased from Shanghai Ryon Biological Technology Co.,Ltd.Microcrystalline cellulose and methyl acryloyl oxygen ethyl trimethyl ammonium chloride(DMC)were provided by Aladdin Industrial Corporation.Toluene and potassium persulfate were supplied by Kemiou(Tianjin,China)and used without further purification.Polyaluminium chloride was also obtained from Kemiou.Commercial grade Kaolin clay(1-4 microns)was procured from Fuchen(Tianjin,China).The pH of kaolin suspension was adjusted using NaOH and HCl purchased from Kemiou.All other chemicals used in this study were of analytical grade.

2.2.Graft copolymerization of DMC onto cellulose

The detailed synthesis process is shown in Text S1.The preparation of C-g-P(DMC)is illustrated schematically in Scheme 1.Tables S1 and S2 display the ratios of reagents in the reaction.In summary,three flocculants with different length and quantity of cationic branch chains have been synthesised.Compared with C-g-P(DMC)1,C-g-P(DMC)2 has shorter cationic branch chains while C-g-P(DMC)3 has less cationic branch chains.

2.3.Materials characterization and instruments

Scheme 1.Synthetic rout of C-g-P(DMC)via two steps:1)inverse emulsion polymerization with γ-methacryloxypropyl trimethoxy silane(KH-570);2)double bond addition reactions with DMC.

Fig.2(c)shows that with the extension of standing time,the lowering of turbidity itself of kaolin suspensions was observed,implying the clay suspension was unstable to some extent.In contrast,the flocculation of clay,induced by C-g-P(DMC)at optimal dosage,led to dropping of turbidity sharply between 2 and 5 min,followed by slowly dropping after 10 min.The turbidity reduce percentage of C-g-P(DMC)1 and C-g-P(DMC)2 was around 96%,while the PAC was 39.58%.In such a short period,it can be settled completely,which can reduce the time of the industrial wastewater in the initial sink,and increase the efficiency and decrease the cost.C-g-P(DMC)1,C-g-P(DMC)2 and C-g-P(DMC)3 have the same tendency to settle with the time.

2.4.Flocculation experiments

Kaolin suspension with the concentration of 2.5 g L-1was employed as simulated wastewater to evaluate the flocculation performance of C-g-P(DMC).The initial turbidity of simulation wastewater,which was taken from the upper portion of kaolin suspension,is between 2000 and 3000 NTU.As the most commonly-used flocculant for wastewater treatment,PAC was selected as a control to compare the flocculation ability with C-g-P(DMC)in this study.The flocculants in solution form were added into 500 mL of kaolin suspension.The suspension was then immediately started to stir at a constant speed of 200 rpm for 1 min,followed by a slow stirring at 40 rpm for 5 min.Thereafter,the sample was left to settle down for 30 min.At the end,the supernatant of kaolin suspension was collected and the turbidity of the samples was measured using 2100N turbidimeter from HACH.To reveal the effect of pH on the turbidity of the system,0.1 mol L-1of hydrochloric acid and sodium hydroxide were added to adjust the pH of the solution so as to find the optimal pH.Along with the determination of optimal standing time of turbidity measurement,the optimal dosage quantity of C-g-P(DMC)was identified.The calculation method of turbidity reduce percentage was recorded in Text S2.In addition,for a better explanation of the independent variables and their interactive effects on the turbidity reduction of kaolin clay suspension,response surface methodology(RSM)was used to analyze the effect of pH,PAC dosage and C-g-P(DMC)dosage on the removal of turbidity.The actual values were presented in Table S3.The Design-Expert 7 software was used to design experiments and evaluate results.

3.Results and discussion

3.1.Characterizations of C-g-P(DMC)

To con firm the structures of C-g-P(DMC)1 flocculants,the characterizations were carried out.Fig.1(a)shows the FT-IR spectra of microcrystalline cellulose and the product of graft copolymerization(i.e.,C-g-P(DMC)1).Compared with Fig.1(a)I,the peaks in Fig.1(a)II at 983 cm-1were ascribed to Si-O-C group stretching vibration,and the peak at 870 cm-1 was assigned to Si-OH,which was hydrolyzate of Si-O-CH3[31].The bending vibration absorption peaks of-CH2-N+(CH3)3group of DMC appeared at 1485 cm-1[25,32].The absorption peaks at 1729 cm-1correspond to the carboxyl groups(-COO-)in the silane and polymer grafted,suggesting that the silane coupling agent had reacted successfully with the-OH groups of cellulose.Moreover,the absorption peak at around 3400 cm-1was broadened,which means that the reaction of KH-570 grafting cellulose destroyed part of the hydroxyl in cellulose[33].The FT-IR spectra of the reference flocculants C-g-P(DMC)2 and C-g-P(DMC)3 are in Figure S1.

For further confirmation of the molecular structures,the XRD patterns of cellulose and C-g-P(DMC)1 were determined;and the results are shown in Fig.1(b).It can be seen that cellulose has 3 crystalline peaks around 2θ =15°,2θ =22° and 2θ =37°,respectively.However,in XRD profiles of C-g-P(DMC)1,the two peaks decreased drastically around 2θ = 15°,22° and almost disappeared around 2θ =37°,suggesting a decrease in crystallinity[34].The loss in cryst allinity indicated that the introduction of DMC onto cellulose backbone reduced the hydrogen bonding ability of cellulose and destroyed original structure[35,36].XRD spectra also show DMC has been successfully grafted onto cellulose,which is consistent with the infrared results.

The TGA results also confirmed the occurrence of graft copolymerization.As shown in Fig.1(c)and(d),the thermal behavior of C-g-P(DMC)1 is different from cellulose.Cellulose had weight losses in two stages as shown in the TGA pro file.The first weight loss occurred between 30°C and 100°C,which might correspond to the loss of absorbed and bound water.The second weight loss began at 300°C,which may be due to the decomposition of cellulose.The differential TGA cellulose curve shows 65 °C and 350 °C are the temperature of quickest weight loss in two stages[37-40].By contrast,the TGA curve of C-g-P(DMC)1 shows three stages of weight losses,which occurred at 69 °C,272 °C and 430 °C,respectively.Compared with cellulose,C-g-P(DMC)1 showed two different temperatures of quickest weight losses,which one was lower and the other one was higher than the quickest weight loss temperature of cellulose,indicating that grafting of the DMC to the cellulose backbone obviously changed the thermal stability of cellulose.Surface morphologies of cellulose,C-KH-570 and C-g-P(DMC)1 were imaged by SEM.Fig.1(e-1)exhibits the surface morphology of the original cellulose.Its surface was rough and rod-like with the width of 7-20 microns.As shown in Fig.1(e-2)and 1(e-3),the surface of cellulose appeared amorphous status for the graft of KH-570 or DMC.The fibrils were separated from each other.This is mainly due to the destruction of crystalline regions on cellulose molecular chains[41].

Furthermore,the CP/MAS13CNMR spectra of an omeric ring carbon of cellulose and C-g-P(DMC)are illustrated in Figure S2.The peak at 105 ppm is assigned to carbon nuclei C-1 of glucose in cellulose.The resonances around peaks at 70 and 80 ppm are assigned to C-2,C-3 and C-5.The peaks at 88 ppm represents the crystalline C-4 and 85 ppm indicates an amorphous C-4.The peaks at 66 and 62 ppm attribute to the C-6 of cellulose.Compared with cellulose,the peak of the C-g-P(DMC)1 at 66 ppm is obviously improved,which can prove the success of grafting.In order to further prove the accuracy of the synthesis,elemental analysis was used to determine the mass ratio of the elements in the sample,and the results were shown in Table S4.From the results,it can be seen that the mass ratio of N element increases to some extent due to the access of DMC to the grafted sample.

Zeta potential(ZP)can be evaluated by measuring the mobility distribution of charged particles as they are subjected to an electric field(Figure S3).The pH dependence of ZP for the standard kaolin suspension and C-g-P(DMC)1 were exhibited in Figure S3.It can be seen that the kaolin particles were negatively charged over all the pH range.Furthermore,the ZP of C-g-P(DMC)1 was almost close to zero in the pH range from 3 to 4.When the environmental pH was weakly acidic,an isoelectric point at pH 4 appeared due to charge neutralization.When the pH value was below 4,the ZP of C-g-P(DMC)1 exhibited negatively charged.

3、能力不足，农民受教育程度普遍较低，综合素质较差。市场对非熟练工人的需求是有限的，这就限制了农村劳动力非农就业机会的获取和工资的提高。

Web服务器采用Web服务多态化技术[15]，通过动态轮换多个虚拟Web服务器给入侵者呈现不断变换的系统攻击面，增大入侵者利用系统脆弱性的难度.在其中1台虚拟Web服务器上设置远程溢出漏洞(CVE-2015-1635)，入侵者可利用该脆弱性远程获取Web服务器root权限.下面根据式(12)仿真分析脆弱性变换周期interval 和变换空间|W|对asp的影响.图3中interval= (0,100τ)，|W|=(0,100).

3.2.Flocculation of fine clay induced by C-g-P(DMC)flocculants

BMD提供了支持定理1中对函数进行A变换的算术运算[12]，如A(xi∧xj)通过算术与实现，A(xi⊕xj)通过算术异或实现，并且可以实现定义6中的“在A变换过程中利用幂等律抑制随机变量的指数成分”.对函数f的逻辑表达式实施相应的算术运算即可得到A(f)对应的BMD，将随机变量Xi的值代入，进行BMD求值[12]即可得到f的信号概率.BMD可以较高效率地表示和操纵类似于概率表达式的代数表达式，并且BMD求值的复杂度为线性复杂度[12]，因此本文在底层使用BMD表示概率表达式.

Fig.1.(a)FT-IR spectra of I)Cellulose and II)C-g-P(DMC)1.(b)XRD patterns of cellulose and the C-g-P(DMC)1.(c)TG curves of Cellulose and C-g-P(DMC)1.(d)DTG curves of Cellulose and C-g-P(DMC)1.(e)SEM images of e-1)cellulose;e-2)C-KH-570;e-3)C-g-P(DMC)1.

Fig.2.(a)Effect of dosage of flocculants on the turbidity reduce percentage of a kaolin suspension(5 g L-1).Standing time:30 min.(b)Comparison flocculation efficiency of different pH in coagulation- flocculation kaolin suspension(5 g L-1)in the optimal flocculants dosage of C-g-P(DMC).(c)Comparison of flocculation efficiency at different standing times in coagulation- flocculation kaolin suspension(5 g L-1)in the optimal flocculants dosage of C-g-P(DMC).

The dosage of flocculant is one of the important factors for lowering the turbidity of wastewater for most flocculation processes.Fig.2(a)shows the effect of dosage on the turbidity reduce percentage of kaolin suspension(5 g L-1)under the standing times of 30 min.These four samples showed different flocculation efficiency under different dosages.The optimum dosage of C-g-P(DMC)1,C-g-P(DMC)2,C-g-P(DMC)3 and PAC was 0.6 mg L-1,0.8 mg L-1,0.8 mg L-1and 2 mg L-1,respectively.The corresponding percentages of turbidity reduction were 96.90%，95.65%,94.35%and 86.06%,respectively.It can be seen in Fig.2(a)that compared with C-g-P(DMC)2,the C-g-P(DMC)1 has better removal efficiency,which suggests that the longer cationic polymer chains grafted in C-g-P(DMC)1 has better effect on flocculation.In addition,the more cationic polymer chains grafted in C-g-P(DMC)1 enhanced the removal efficiency compared with the less cationic polymer chains grafted in C-g-P(DMC)3 as shown in Fig.2(a).For PAC at lower dosages,it was difficult to execute electrical neutralization because of insufficient charge,leading to weak aggregation.For C-g-P(DMC)1,the optimum dosage was lower(0.6 mg L-1),which might be associated with the star-like structure of the flocculant,which is useful in improving the bridging adsorption and charge-neutralizing ability for robust flocculation.However,the high concentration of clay suspension tended to decrease the flocculation efficiency.The main reason was that residual positive charge after neutralizing the negative charge of kaolin suspension regenerates the repulsion forces to lead to system instability,which was exceedingly hard to sever.The optimum dosage of PAC was 2 mg L-1,because PAC did not possess bridging ability as cationic polymer chains do.Therefore,from the aspects of additive amount and turbidity reduce percentage;the effect of C-g-P(DMC)was superior to PAC.

Fig.3.Surface plots and its corresponding contour plots of turbidity reduction showing the effect of variables.(a)PAC dosage-pH at C-g-P(DMC)2 dosage of 0.8 mg L-1(b)PAC dosage-C-g-P(DMC)2 dosage at pH of 7(c)C-g-P(DMC)2 dosage-pH at PAC dosage of 2 mg L-1.

As reported previously,acid or alkali has an important influence on flocculation performance of the flocculant.Namely,acid or alkali might affect the charge characteristics of flocculants,thus leading to the different interactions between flocculant and clay particles.Fig.2(b)displays the influence of all kinds of flocculants on turbid ity change of the kaolinsuspension underdifferent pHvalues;whereas the dosage of flocculantswas based on the optimal ones determined previously.The results showed that C-g-P(DMC)was superior to PAC in lowering the turbidity of clay suspension,regardless of pH range.PAC in alkaline condition generated anionic-charge Al(OH)3,which seriously dramatically declined the ability of flocculation due to the electrostatic repulsion between PAC and mainly anioniccharge clay particles.Fig.2(b)also shows the effects of different chain lengths and chain numbers on turbidity reduction under different pH conditions,respectively.C-g-P(DMC)1 has better removal efficiency thanC-g-P(DMC)2andC-g-P(DMC)3 in a wide pH range,which is consistant with dosage effect results.Moreover,using cellulose as core for floccul ants lowers their own environment impact due to the biodegradable features of cellulose as well as relatively short cationic polymer chains grafted,compared to the conventional cationic polyacrylamide with extremely high molecular weights.

这首七律写在小溪边生活的游禽，描写它们自由自在、无拘无束的生活状态，写它们美丽的羽毛和如云的溪花、如练的溪水混成一色，写它们快若闪电、隐现无常的飞翔姿态，写它们翻动羽毛自我炫耀的神情面貌，都刻画得栩栩如生，入木三分。最后两句曲终奏雅，点出主旨，提醒鸟儿们，不要以为这个偏僻之地就没有了罗网，到时被罗网网住，就只能被人类豢养在樊笼中了，那时虽然有吃有喝，但这种生活哪有自由飞翔时的生活值得羡慕。这首诗就把形象和哲理结合得非常好。

3.2.2.Effect of pH on turbidity of kaolin suspensions

2015年5月25日—29日，由ICMI主席Fernaidando Arzarello、副主席Angel Ruiz和办公室主任Lena Koch组成的考察团对上海市申办ICME-14进行了现场考察．ICMI秘书长Abraham Arcavi原定来沪，但因身体原因临时取消行程．

The structures of the cellulose and C-g-P(DMC)were characterized using a Nicolet Corporation AVATAR-360 FT-IR spectrophotometer(USA)to elucidate the reaction mechanisms.XRD patterns were obtained with an X-ray diffractometer (XRD-6100,Shimadzu,Japan) with graphite monochromatized Cu Kα radiation(lambda=1.54056 Å).The thermogravimetric analyses(TGA)was performed using a Germany Netzsch Corporation STA409PC simultaneous thermal analysis instrument under N2and the temperature was raised from 25 °C up to 800 °C with the heating rate of 10°C min-1.The scanning electron microscope(SEM)images were recorded by a ZEISS SUPRA 40 SEM system using the acceleration voltage of 30 kV.The 29C MAS spectra were recorded on a Bruker AVANCE III 400 WB spectrometer equipped with a 4 mm standard bore probe heat whose X channel was tuned to 79.50 MHz for 29C,pulse sequence is onepulse,using a magnetic field of 9.39T at 297 K.The dried and finely powdered samples were packed in the ZrO2rotor closed with Kel-F cap which were spun at 12 kHz rate.A total of 2000 scans were recorded with 3 s recycle delay for each sample.All 29C CP MAS chemical shifts are referenced to the resonances of 3-(trimethylsilyl)-1-propanesulfonic acid sodium salt(DSS)standard(δ=0.0).The elemental analysis was performed using ELEMENTAR vario EL III.

3.2.4.Three-dimension response surface and contour plots

3.2.3.Effect of standing time on turbidity of kaolin suspensions

（2）若（W′-W1）Q1<2PRQ3，则买家会选择“购买套装化妆品”，此时卖家的收益为>2P（1-R）3r。

To get insight into the independent variables and their interactive effects on the turbidity reduction of kaolin clay suspension,3D plots and their corresponding contour plots of PAC and C-g-P(DMC)2 formulation experiments are represented in Fig.3.Therefore,the optimal value was obtained through the response surface plot and the contour of each pair of independent factors.

Fig.4.The distribution of particles size of(a)kaolin particles(b)kaolin particles adding PAC(c)kaolin particles adding C-g-P(DMC)2(d)kaolin particles adding C-g-P(DMC)1.

Scheme 2.Proposed flocculation mechanism of C-g-P(DMC)for treating Kaolin suspension.

The authors declare no conflict of interests.

3.2.5.Particle size analysis

等间隔非平稳信号的处理，小波方法[11]较为适用。由于隧道位移序列是按照相同监测频率测得的数据，因此为等间隔序列，小波变换对于等间隔序列具有非常好的适配性，通过小波分解将位移序列分解成多个子序列，子序列分为高频序列和低频序列。令小波基函数为ψ(t)，经过连续变换函数为ψ(a)，则有ψ(a)为：

Fig.4 shows the distribution of particles size.As can be seen from Fig.4(a),the distribution of kaolin particle size is mainly between 0 and 4 microns[42,43].After adding PAC,the second peak occurred with mean particle size approximately 8 microns(see Fig.4(b)),which is attributed to the kaolin flocs induced by PAC.In Fig.4(c),it can be seen that the flocculation of kaolin particles was improved significantly after adding C-g-P(DMC)2 compared to PAC;and the floc size ranged distributed between 5 and 26 microns with peak size around 8 micros.After adding C-g-P(DMC)1,the flic size was further increased(see,Fig.4(d)).The majority offloc size was distributed between 7 and 26 microns with peak size close to 14 micros,thus demonstrating that the C-g-P(DMC)1 is superior to C-g-P(DMC)2 due to more branched chains present in C-g-P(DMC)1 which facilitates the flocculation of clay particles.

3.3.The flocculation mechanism

The proposed flocculation mechanism of the cellulose-based flocculants is illustrated in Scheme 2. Overall, the different flocculation performance among three types of flocculants assessed in the current work is attributed to their distinct structural characteristics. The modified cellulose-based flocculants possess different chain architectures and grafted chain lengths, mainly depending on the monomer ratio to cellulose over the grafting polymerization. The resulting C-g-P (DMC) has branched pendent chains bearing a number of quaternary ammonium groups which are highly cationic. The extended cationic chains and relatively rigid core of cellulose createmore robust flocs after the primary particles of clay are flocculated. The adsorption of cellulose-based flocculant on the surface of clay particles is mainly driven by charge association between the anionic-charge clay top/bottomsurfaces and cationic chains grafted on cellulose. The electrostatic attraction between the positively charged polymer and the negatively charged particles promotes the adsorption [44,45]. Furthermore, the cellulose-based flocculant with relatively long and more grafted polymer chains, i.e., C-g-P (DMC) 1, could be readily extended to be beyond the thickness of the electric double layer of clay, thus creating effective bridging. As a result, C-g-P (DMC) 1 is superior to C-g-P (DMC) 2 with shorter chains and C-g-P (DMC) 3 with less chains and much better than PAC in inducing clay flocculation. Whereas charge neutralization is the dominant mechanism for PAC induced the aggregation of clay particles [46,47].

3.2.1.Effect of dosage of flocculants on turbidity of kaolin suspensions

4.Conclusions

A renewable and biodegradable flocculant C-g-P(DMC)has been investigated through a new synthetic method for f l occulation of kaolin suspension.Compared with other flocculant PAC,C-g-P(DMC)had the advantages of low dosing quantity and acid-alkali environmental adaptation.The dosage of C-g-P(DMC)can be reduced to 0.6 mg L-1,while the turbidity reduce percentage can reach to 96.5%.In addition,the response surface methodology(RSM)study con firms that PAC and C-g-P(DMC)have synergy in turbidity reduction with a higher efficiency of 98.32%.These green materials could also avoid the secondary pollution,which makes it an excellent flocculant for further systematic research.

Conflict of interests

Fig.3(a)shows that the optimum dosage of PAC varies independent of pH and the optimum dosage of PAC is 1.5 mg L-1at pH 7.Besides,the illustration reveals the dosage of PAC has significant interaction with pH,suggesting that PAC in alkaline condition generated anionic-charge Al(OH)3 which reduced the turbidity reduce percentage.On this basis,the different dosage of PAC was put into use according to the different pH of industrial wastewater.Fig.3(b)shows that the highest turbidity reduce percentage of 98.32%was attained with the dosage of PAC 1.5 mg L-1and C-g-P(DMC)2 0.6 mg L-1.Moreover,with PAC and C-g-P(DMC)2 increased to a certain amount,the turbidity reduce percentage reached to a maximum and then decreased.It indicates that f l oc breakup occurs due to charge reversal and dispersion when there is an extreme dosage of coagulant and flocculant.Besides,the PAC and C-g-P(DMC)2 together can improve the turbidity reduce percentage and decrease the processing cost in dealing with industrial wastewater.Fig.3(c)shows that the influence of pH on turbidity reduces percentage is very low as the dosage of C-g-P(DMC)2 is between 0.6 mg L-1to 0.8 mg L-1.It can be seen in Fig.3(c)that the turbidity reduction is more dependent on the coagulant dosage,which suggests colloidal particles neutralization will not occur when the coagulant dosage is not sufficient.The C-g-P(DMC)2 is not classified as an acidic or basic flocculant and it is cation in the wide range of pH.Variation of pH has no effect on its charge density,indicating insignificant interaction between pH and C-g-P(DMC)2.

Acknowledgements

This work was supported by the National Natural Science Foundation of China(51379077,21607044),the Fundamental Research Funds for the Central Universities(2016MS108)and Natural Science Foundation of Hebei Province(B2017502069).

Appendix A.Supplementary data

Supplementary data related to this article can be found at https://doi.org/10.1016/j.gee.2017.11.005.

③治理工程透水性。治理工程实施后河道径流和地下水之间应顺利完成交换，使地下水能够在丰水期得到补充，并在枯水期对河道径流予以逆向补给。

References

[1]M.Yao,X.W.Zhang,L.C.Lei,Sci.China Chem.55(2012)416-425.

[2]L.H.Andrade,F.D.S.Mendes,J.C.Espindola,M.C.S.Amaral,Braz J.Chem.Eng.32(2015)735-747.

[3]Y.Z.Sang,H.N.Xiao,J Colloid Interface Sci.326(2008)420-425.

[4]L.Lu,Z.Pan,H.Nan,Water Res.57(2014)304-312.

[5]C.Overden,H.N.Xiao,Colloid Surf.A 197(2002)225-234.

[6]Y.Su,H.Du,Y.Huo,Y.Xu,J.Wang,L.Wang,S.Zhao,S.Xiong,Int.J.Boil Macromol.87(2016)34-40.

[7]H.Luo,X.A.Ning,X.Liang,Y.Feng,J.Liu.Bioresour.Technol.139(2013)330-336.

[8]T.Cai,H.Li,R.Yang,Y.Wang,R.Li,H.Yang,A.Li,R.Cheng,Cellulose 22(2015)1439-1449.

[9]K.E.Lee,N.Morad,T.T.Teng,B.T.Poh,Chem.Eng.J.203(2012)370-386.

[10]A.Hanc,M.Dreslova,Bioresour.Technol.127C(2016)112-118.

[11]C.Gadipelly,A.Pérez-González,G.D.Yadav,O.I.Inmaculada,R.Ibáñez,V.K.Rathod,K.V.Marathe,Ind.Eng.Chem.Res.53(2014)11571-11592.

[12]J.Ji,J.Qiu,N.Wai,F.S.Wong,Y.Li,Water Res.44(2010)1627-1635.

[13]J.S.Li,P.R.Modak,H.N.Xiao,Colloid Surf.A 289(2006)172-178.

[14]W.Zhang,P.Xiao,Y.Liu,S.Xu,F.Xiao,D.Wang,C.W.K.Chow,Sep.Purif.Technol.132(2014)430-437.

[15]X.Lin,X.Li,S.Lu,F.Wang,T.Chen,J.Yan,Environ.Sci.Pollut.R.22(2015)14629-14636.

[16]G.L.Xu,Y.B.Fan,Y.Yan,W.Yang,D.Yuan,G.Wu,Fresen Environ.Bull.19(2010)1591-1598.

[17]H.Zhu,Y.Zhang,X.Yang,H.Liu,L.Shao,X.Zhang,J.Yao,J.Hazard Mater 296(2015)1.

[18]W.Hu,Z.Liu,Y.Hu,A.Li,Water Res.96(2016)126-135.

[19]L.Yan,J.G.Huang,Sci.China Chem.57(2014)1672-1682.

[20]R.Yang,H.Li,M.Huang,H.Yang,A.Li,Water Res.95(2016)59-89.

[21]M.B.Kurade,K.Murugesan,A.Selvam,S.M.Yu,J.W.C.Wong,Bioresour.Technol.217(2016)179.

[22]J.W.Wong,K.Murugesan,A.Selvam,B.Ravindran,M.B.Kurade,S.M.Yu,Bioresour.Technol.213(2016)31-38.

[23]Y.J.Sun,H.L.Zheng,Z.Xiong,Y.Wang,X.Tang,W.Chen,Y.Ding,Desalin Water Treat.56(2014)1-11.

[24]L.F.Yan,H.Y.Tao,P.R.Bangal,Clean.Soil Air Water 37(2009)39-44.

[25]M.Nourani,M.Baghdadi,M.Javan,G.N.Bidhendi,J.Environ.Chem.Eng.4(2016)1996-2003.

[26]K.E.Lee,B.T.Poh,N.M.T.T.Teng,J.Macromol.Sci.A46(2009)240-249.

[27]J.Yang,C.R.Han,J.F.Duan,M.G.Ma,X.M.Zhang,F.Xu,R.C.Sun,X.M.Xie,J.Mater Chem.22(2012)22467-22480.

[28]X.Liu,H.Seki,H.Maruyama,Sep.Purif.Technol.93(2012)1-7.

[29]Z.H.Yang,J.Huang,G.M.Zeng,M.Ruan,C.S.Zhou,L.Li,Z.G.Rong,Bioresour.Technol.100(2009)4233-4239.

[30]Y.X.Zhao,B.Y.Gao,Y.Wang,H.K.Shon,X.W.Bo,Q.Y.Yue,Chem.Eng.J.183(2012)387-394.

[31]L.Dong,H.Hu,S.Yang,F.Cheng,Ind.Eng.Chem.Res.53(2014)6221-6229.

[32]H.L.Zheng,Y.J.Sun,J.S.Guo,F.Li,W.Fan,Y.Liao,Q.Guan,Ind.Eng.Chem.Res.53(2014)2579-2582.

[33]V.F.Kurenkov,E.V.Shatokhina,H.G.Hartan,F.I.Lobanov,Russ.J.Appl.Chem 78(2005)1872-1875.

[34]D.Wang,T.Zhao,L.Yan,Z.Mi,Q.Gu,Y.Zhang,Int.J.Biol.Macromol.92(2016)761-768.

[35]J.P.Wang,Y.Z.Chen,X.W.Ge,H.Q.Yu,Chemosphere 66(2007)1752-1757.

[36]H.Liu,X.Yang,Y.Zhang,H.Zhu,J.Yao,Water Res.59(2014)165-171.

[37]M.Huang,Y.Wang,J.Cai,J.Bai,H.Yang,A.Li,Water Res.98(2016)128-137.

[38]S.Matsuoka,H.Kawamoto,S.Saka,Carbohydr.Res.346(2011)272-279.

[39]K.Babeł,Fuel Process Technol.85(2004)75-89.

[40]S.Vyas,S.D.Pradhan,N.R.Pavaskar,A.Lachke,Appl.Biochem.Biotech.118(2004)177-188.

[41]J.Long,Y.Zhang,L.Wang,X.Li,Sci.China Chem.59(2016)557-563.

[42]J.P.Wang,S.J.Yuan,Y.Wang,H.Q.Yu,Water Res.47(2013)2643-2648.

[43]G.Z.Kyzas,P.I.Siafaka,E.G.Pavlidou,K.J.Chrissafis,D.N.Bikiaris,Chem.Eng.J.259(2015)438-448.

[44]H.N.Xiao, “Fine clay flocculation”,in:A.Hubbard(Ed.),the “encyclopedia of surface and colloid science”,second ed.,Taylor&Francis Group,USA,2006,pp.2197-2206.

[45]S.Barua,J.Ramos,T.Potta,D.Taylor,H.C.Huang,G.Montanez,K.Rege,Comb.Chem.High.T Scr 14(2011)908-924.

[46]D.Dihang,P.Aimar,J.Kayem,S.N.Koungou,Chem.Eng.Process 47(2008)1509-1519.

[47]H.N.Xiao,N.Cezer,J.Colloid Interf.Sci.283(2005)406-413.