外文翻譯-- 用表面活性劑改性的涂氧化鐵砂對(duì)天然有機(jī)物的去除（英文）

1、Journal of Hazardous Materials 174 (2010) 567–572Contents lists available at ScienceDirectJournal of Hazardous Materialsjournal homepage: www.elsevier.com/locate/jhazmatRemoval of natural organic matter using surfactant-

2、modified iron oxide-coated sandChunli Ding a, Xin Yang a, Wei Liu b, Yujung Chang c, Chii Shang a,?a Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Clear Water Bay,

3、 Kowloon, Hong Kongb Department of Environmental Science and Engineering, Sun Yat-Sen University, Guangzhou, Chinac HDR Engineering, Inc, 500 108th Avenue NE, Bellevue, WA, 98004-5549, USAa r t i c l e i n f oArticle his

4、tory:Received 20 August 2009Received in revised form15 September 2009Accepted 15 September 2009Available online 23 September 2009Keywords:AdsorptionIron oxide-coated sandNatural organic matterSurfactantsa b s t r a c tIr

5、on oxide-coated sand (IOCS) was modified with hexadecyltrimethyl ammonium (HDTMA) and testedas an adsorbent for the removal of natural organic matter (NOM) from water. The modification did notchange the physical properti

6、es of the IOCS but coated HDTMA onto its surface. The HDTMA-modified IOCSdisplayed a faster initial NOM adsorption and substantially higher capacity than the unmodified IOCSover a wide pH range in both batch and column a

7、dsorption. The enhancement was more pronounced athigher pH. Compared to unmodified IOCS, the HDTMA-modified IOCS removed more hydrophobic andlarger NOM molecules and its NOM adsorption was less sensitive to the changes i

8、n ionic strength. Theadsorption capacity of the modified IOCS was regenerated in-situ with NaOH solution and ex-situ withHDTMA solution. HDTMA-modified IOCS adsorption may be a promising alternative technology for NOMrem

9、oval.© 2009 Elsevier B.V. All rights reserved.1. IntroductionNatural organic matter (NOM) is present in all natural watersources. It is a complex mixture of organic compounds contain-ing both hydrophilic (phenolic a

10、nd carboxylic) functional groupsand hydrophobic (aromatic, aliphatic) moieties [1]. NOM itself isnot harmful, but it can react with disinfectants to form disinfec-tion by-products [2]. In addition, NOM reduces the effect

11、ivenessof water treatment by interfering with flocculation processes, foul-ing membranes and adsorbents, and interfering with oxidation andprecipitation of dissolved iron and manganese. As a result, NOMoften increases th

12、e dose requirements for coagulants, oxidants,adsorbents and the frequency of membrane cleaning.To remove NOM during water treatment, a great variety ofprocesses have been designed or modified, such as enhancedcoagulation

13、, membrane filtration, ozonation/biofiltration, andadsorption. Among these processes, adsorption on low-cost mediais particularly an attractive option in many situations. Iron oxide-coated sand (IOCS) with a layer of iro

14、n oxides coated on the sandsurface has been demonstrated as an effective NOM adsorbent[3–5]. IOCS has several advantages for NOM removal over otheradsorbents. It is easy to prepare, convenient to apply in filtrationsetup

15、, easy to separate from treated water, and easily regenerated? Corresponding author. Tel.: +852 2358 7885; fax: +852 2358 1534.E-mail address: cechii@ust.hk (C. Shang).[3]. Additionally, IOCS can be developed in groundwa

16、ter treat-ment plants where sand filters are used for removing iron fromgroundwater [6]. This is especially important in developing coun-tries where high-cost absorbents are not generally accessible oraffordable. For all

17、 of these reasons, IOCS is an adsorbent worth-developing for NOM removal from water.One disadvantage of IOCS is that its NOM adsorption is stronglypH dependent and often works best at a rather acidic pH (usuallyaround 4–

18、5) but less well in neutral and basic conditions [3,5].Adjusting pH to the acidic range may not be cost-effective withwater that requires the addition of large amounts of acid. Modi-fying IOCS to function well over a wid

19、er pH range would make itmore competitive with other adsorbents for NOM removal.Surface modification with cationic surfactants or polymers canenhance adsorbents’ capacity for organic solutes by creating anorganic surface

20、 layer. Quaternary ammonium compounds (QACs)are frequently used. QAC-modified clays and zeolites have beeninvestigated extensively and proven to be effective for removal oforganic contaminants from water [7–10]. IOCS coa

21、ting with QACshas not been evaluated before, but it should be promising. Thehypothesis is that coating the IOCS surface with QACs containinghydrophobic tails and pH-independent cationic head groups mayimprove the hydroph

22、obicity and/or positive charge density of theIOCS surface. This should strengthen surface interactions betweenNOM and the IOCS. Such tailored IOCS shall then have better NOMadsorption performance, stability and durabilit

23、y, and it possiblyadsorb NOM over a larger pH range.0304-3894/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.jhazmat.2009.09.089C. Ding et al. / Journal of Hazardous Materials 174 (2010)

24、 567–572 569Fig. 1. FTIR spectra of (a) HDTMA, (b) HDTMA-modified and (c) unmodified IOCS.3.2. NOM removal in batch tests3.2.1. Adsorption kineticsFig. 2 shows the kinetics of NOM adsorption on the unmodifiedand modified

25、 IOCS at pH 4.5 ± 0.2 and 9.0 ± 0.1, along with curvespredicted from a pseudo-second-order kinetic model. Under bothacidic and basic conditions, NOM adsorption on the unmodifiedor modified IOCS was fast at the

26、very beginning, but slowed downsoon after the initial uptake, and the amount of NOM adsorbed thenincreased slowly to reach a plateau in approximately 20 h. Suchrapid initial NOM adsorption followed by slow adsorption lat

27、er haspreviously been reported for iron oxides [22,23]. The rapid initialadsorption has been attributed to physical adsorption mechanismssuch as electrostatic interactions, which determine the attachmentof NOM to the sol

28、id surface at the beginning stage. The decreasedadsorption rate thereafter has been assigned to ligand exchange.The pseudo-second-order rate model has been widely appliedto adsorption of pollutants from aqueous solutions

29、 [24]. The advan-tage of using this model is that the model calculates the rateconstant and the equilibrium capacity so that there is no needFig. 2. Adsorption of NOM onto the unmodified and HDTMA-modified IOCS at pH4.5

30、± 0.2 and pH 9.0 ± 0.1. Initial concentration of 10 mg/L NOM, 0.1N NaNO3 ionicstrength, 2.5 g IOCS/L.Fig. 3. Adsorption isotherms of NOM (a) at pH 5.0 ± 0.1, 7.0 ± 0.1 and 10.0 ± 0.1,with an ioni

31、c strength of 0.01N and (b) with an ionic strength of 0.01 and 0.1N, atpH 7.5 ± 0.1 onto the unmodified and HDTMA-modified IOCS, with an equilibrationtime of 24 h.to obtain the latter from isotherm experiments [25].

32、 The pseudo-second-order rate model is expressed as [24]:dqt dt = k(qe ? qt)2 (1)ortqt = 1kq2 e + 1qe t (2)where qe and qt are the amount (mg/g) of the adsorbate adsorbedat equilibrium and time t, respectively, and k is

33、the adsorptionrate constant (g/mg/min). Fig. 2 shows that the experimental dataagreed well with such model. Table S1 presents the kinetic param-eters of NOM adsorption derived for both the unmodified andmodified IOCS. Th

34、e rate constants of NOM adsorption on the mod-ified IOCS were higher than those on the unmodified IOCS at bothacidic and basic pH, suggesting a stronger driving force for NOMadsorption on the HDTMA-modified IOCS.3.2.2. A

35、dsorption isothermsNOM adsorption isotherms were developed at pH 5.0 ± 0.1,7.0 ± 0.1 and 10.0 ± 0.1. The pH 10 was above the PZC of the modi-fied IOCS. The ionic strengths tested were 0.01 or 0.1N NaNO3. T

36、heexperimental data plotted with curves predicted from a Langmuirmodel are shown in Fig. 3a and b. The Langmuir adsorption modelis expressed as:qe = qmbCe1 + bCe (3)where Ce is the equilibrium concentration in mg/L, qe i

37、s the amountadsorbed at equilibrium in mg/g, qm is the maximum amount ofadsorbate that can be adsorbed up to monolayer coverage, and b isthe Langmuir constant, which is related to the binding strength. Asshown, the exper