Adsorption of emerging sodium p-perfluorous nonenoxybenzene sulfonate(OBS)onto s

Wei WANG,Xue ZHANG,Yuhui ZHANG,Xin MI,Run WANG,Huilan SHI,Chunli LI,Ziwen DU and Youming QIAOState Key Laboratory of Plateau Ecology and Agriculture,Qinghai University,Xining 8006(China)

2College of Eco-environmental Engineering,Qinghai University,Xining 810016(China)

3College of Environmental Science and Engineering,Beijing Forestry University,Beijing 100083(China)

ABSTRACT The widespread use of sodium p-perfluorous nonenoxybenzene sulfonate(OBS),a typical alternative to perfluorooctane sulfonate,has resulted in potential threats to the environment,but the adsorption behavior of OBS in soils has not yet been reported.In this study,the adsorption behaviors of OBS on five soils with different physicochemical properties were investigated.The rate of OBS adsorption was fast,and most of the OBS uptake was completed within 12 h.The good model fit of OBS adsorption to the pseudo-second-order and Elovich models indicated the occurrence of chemical adsorption.The adsorption isotherms of OBS on the soils were better described by the Freundlich model than by the Langmuir model,suggesting that the OBS adsorption sites on the soils were heterogeneous.This is possibly associated with various adsorption mechanisms including hydrophobic,π-π,hydrogen bonding,and electrostatic interactions,further confirmed by the good model fit to the D-R isotherm.Adsorption of OBS occurred on the soils,and the adsorption process was spontaneous and endothermic.In addition,the soils were more suitable for OBS adsorption at lower pH values due to the stronger electrostatic adsorption.The OBS adsorption on the soils decreased with the increase of soil depth from 0 to 30 cm.Moreover,the presence of organic matter and ammonia nitrogen in the soils was favorable for OBS adsorption,and these parameters decreased with increasing soil depth,making OBS adsorption less prominent in the deeper soil.This study indicates that OBS is easily enriched in surface soils,and that soil organic matter and ammonia nitrogen significantly affect OBS migration in soil.Key Words:adsorption model,environmental behavior,perfluorooctane sulfonate alternative,soil properties,thermodynamic parameters

INTRODUCTION

Perfluorooctane sulfonate(PFOS)is one of the most typical perfluoroalkyl and polyfluoroalkyl substances(PFASs)(Qianet al.,2017)that is stable,persistent,and highly bioaccumulative(Parsonset al.,2008;Ghisiet al.,2019).Most importantly,PFOS has been widely detected in various environmental matrices,such as surface water,soil,and animal tissues(Gellrichet al.,2012;Princzet al.,2018).Therefore,various international organizations have issued regulations to limit the production and use of PFOS(Gellrichet al.,2012;Fenget al.,2017).For instance,PFOS was listed in Annex B of the UNEP Stockholm Convention in May 2009(UNEP,2009).

However,on the market,the currently available alternatives to PFOS are also PFASs.As one of the most common alternatives to PFOS,sodiump-perfluorous nonenoxybenzene sulfonate(OBS)is gradually being applied in various fields such as film-forming fluoroprotein foams,alcohol-resistant foams,and as a surfactant for improving oil production in China(Xuet al.,2017).Previous studies have demonstrated the possibility of OBS detection and have shown that the peak of OBS concentration is 3.2×103ng L-1in surface water(Xuet al.,2017).However,OBS is likely more bioaccumulative than PFOS because of its hydrophobicity(Conderet al.,2008).The lethal concentration required to kill 50% of the population over 96 hours of exposure to OBS,for zebrafish and tadpoles,is less than 100 mg L-1but greater than 10 mg L-1,which indicates that OBS has acute toxicity to aquatic animals,similar to PFOS(UN,2007;Xuet al.,2017).According to the Globally Harmonized System of Classification and Labeling of Chemicals(GHS)definition,OBS should be classified as a category III chemical(UN,2007).Even so,almost no study has been conducted regarding its behavior in the environment,and its further spread in the soil is not clear.

Soil is a complex mixture of inorganic particles and organic matter(OM),and it is one of the important environmental media carrying OBS(Ishiguro and Koopal,2016).It is hypothesized that OBS will be spread widely in surface water and soil(Xuet al.,2017).Adsorbed on soil,OBS can reflect its fate and environment behavior in soil(Liet al.,2019).The components of soil can greatly influence their capture ability for target organic pollutants.The organic matter in soil can promote the adsorption of PFOS,while ferric oxides have the opposite effect(Qianet al.,2017).Apart from the chemical composition of soil,the capture capacity can be greatly influenced by its particle size(Qiet al.,2014).Therefore,it is important to consider different soil properties for studying the environmental behavior of OBS in soil.

An OBS molecule consists of a hydrophobic C-F chain(except one oxygen atom in the carbon chain)and a hydrophilic sulfonate ion end,similar to PFOS.It is shown that hydrophobic tendencies and phase transfer are the primary mechanisms of PFOS adsorption by soil(Zhanget al.,2015).Moreover,ion exchange also contributes to the adsorption of PFOS because of its ionic form.Crops can also adsorb PFOS when PFOS is released into the soil through water irrigation,and this can influence human health(Wanget al.,2010).The adsorption of PFOS on the soil in the plain area can reach an equilibrium state within 7 d,and the adsorption isotherm is non-linear with adsorption amounts ranging from 13.1 to 66.8 μg g-1(Zhanget al.,2015).However,OBS has shorter fluorocarbon chains and contains ether bonds,benzene rings,and carbon-carbon double bonds.Therefore,differences may exist between OBS and PFOS in the adsorption process,making it very important to study the environmental behavior of emerging PFASs in soils.

The objective of this study was to comprehensively investigate the adsorption behavior of OBS onto five selected kinds of soils.The adsorption kinetics,isotherms,and the influences of soil environment parameters including OM,total nitrogen(TN),ammonia nitrogen(AN),nitrate nitrogen(NN),depth,pH,and temperature were also studied.Most importantly,the potential interaction mechanisms of OBS adsorption on soils were then inferred from the findings.

MATERIALS AND METHODS

Chemicals and materials

Sodiump-perfluorous nonenoxybenzene sulfonate(C15H4F17O4SNa,Fig.1)was purchased from Guangzhou Wengjiang Reagent Co.Ltd.(Guangzhou,China).Potassium phosphate monobasic and methanol were of analytical grade and obtained from Sinopharm Group Chemical Reagent Co.Ltd.(Shanghai,China).The other chemicals used in the experiment were of analytical reagent grade or better.The ultrapure water that was used was produced by a Milli-Q integral water purification system(Millipore,USA).Five soils(0–30 cm)including a Kastanozem(S1),a Calcisol(S2),a Leptosols(S3),a Histosol(S4),and a Chernozem(S5)were obtained from Qinghai Province,China.The soils at 0–20 cm were selected for the major adsorption experiments.To further explore the changes in OBS adsorption behavior at different soil depths,the soil with a depth of 0–30 cm was divided into 0–10,10–20,and 20–30 cm samples for the experiment.All soil samples were air-dried at room temperature and ground in a mortar,and then they were sieved through a 2-mm plastic sieve.

Fig.1 Molecular structure of sodium p-perfluorous nonenoxybenzene sulfonate(OBS).

Soil characterization

The particle size distribution of each soil sample was measured by laser particle size distribution(Dandong Baxter Instruments Co.,China).The point of zero charge(pHpzc)of soils was obtained following three steps:the soil was suspended in 0.1 mol L-1NaCl solution,which was previously adjusted with NaOH or HCl solution to a specific initial solution pH,and shaking for 24 h in an orbital shaker.The final pH,which was consistent with the initial pH,was regarded as pHpzc.The soil organic carbon content was determined using potassium dichromate oxidation with external heating(Xueet al.,2019).The total nitrogen was measured using the Kjeldahl procedure,and this was followed by colorimetric analysis(Luoet al.,2019).The main physicochemical properties of the five soils are shown in Table SI(see Supplementary Material for Table SI).

Adsorption experiments

Initially,the OBS solutions were blended using ultrasound for 20 min.Adsorption kinetic tests of OBS on different types of soils were carried out in a flask containing 2.5 mg L-1OBS solution and 10 g L-1adsorbent.The experiments were conducted in an orbital shaker at the rotating speed of 150 r min-1and temperature of 25°C(except for the temperature effect experiment).Samples of 1 mL were taken at predetermined time intervals of 1,3,6,12,24,and 36 h.The adsorption isotherm experiments were conducted with 500 mg adsorbent in 40 mL of 2–15 mg L-1OBS solution,and the adsorption time used was 24 h.To investigate the influence of the complex soil environment on OBS adsorption,the effects of environmental temperature,soil depth,and solution pH were investigated.The effect of environmental temperature was examined at 25 and 50°C.Except for the different experimental temperatures,the other conditions were the same as the adsorption isotherm experiment.The effect of pH was examined in 40 mL of 2.5 mg L-1OBS solution and 10 g L-1of soil at pH ranging from 4.0 to 9.0.The effect of soil depth(0–10,10–20 and 20–30 cm)was also investigated with 10 g L-1of soil in 40 mL of 2.5 mg L-1OBS solution.All measurements were performed twice.The adsorption kinetic values were fitted with four kinetic models including pseudo-first-order,pseudo-second-order,Elovich,and intra-particle diffusion models(Wuet al.,2009;Wanget al.,2018a):

whereqt(μmol g-1)andqe(μmol g-1)are the OBS adsorption amount at timet(h)and equilibrium,respectively,k1(h-1),k2(g μmol-1h-1),andkn(μmol g-1h-0.5)are the rate constants of the pseudo-first-order,pseudo-secondorder,and intra-particle diffusion models,respectively,v0(μmol g-1h-1)=k2q2e,cn(μmol g-1)is the boundary layer thickness coefficient,anda(μmol g-1h-1)andb(g μmol-1)are the Elovich model adsorption rate constants.The Langmuir,Freundlich,Temkin and D-R isotherm models were employed to analyze the OBS adsorption isotherms(Saltalıet al.,2007;Wanget al.,2018b):

whereqm(μmol g-1)is the maximum OBS adsorption capacity,Ce(μmol L-1)is the equilibrium OBS concentration in solution,KL(L μmol-1)is the Langmuir constant,Kfis the Freundlich constant(μmol g-1(μmol L-1)-1/n),nis the index of heterogeneity,Bis the constant related to the heat of adsorption,KTis the equilibrium binding constant(L μmol-1),βis the constant of the adsorption energy(mol2J-2),ε(J mol-1)is the Polanyi potential(ε=RTln(1+1/Ce)),andE(kJ mol-1)is the value of mean sorption energy when 1 mol of adsorbate is adsorbedon the surface of adsorbent

The thermodynamic parameters were calculated according to the following formulae:

whereΔG0is the change in the free energy of the reaction system(kJ mol-1),Ris the ideal gas constant(8.314 kJ(mol K)-1),Tis the temperature in Kelvin(K),ΔH0is the change in enthalpy(kJ mol-1),andΔS0is the change in entropy(kJ(mol K)-1).

Analytical methods

After adsorption,the collected sample was placed in a 2-mL centrifuge tube and centrifuged at 10 000 r min-1for 10 min.Then 0.5 mL of the centrifuged sample was collected for detection.Residual OBS in the filtrate was measured using high-pressure liquid chromatography(Agilent 1260 Infinity II,Agilent Technologies,USA)with a UV/Vis detector and a TC-C18 column(4.6 mm×250 mm inner diameter,particle size of 5 μm;Agilent Technologies,USA)at a maximum absorption wavelength of 220 nm.The mobile phase consisted of methanol and 0.2 mol L-1potassium phosphate at a flow rate of 1.0 mL min-1.The total analysis time for a sample was about 7 min.

Quality control

The contribution of the plastic sieve to soil PFASs and OBS adsorption by the flask were tested before all experiments,and the results showed that their influences were negligible.The ultrapure water was tested as the control for each of these tests to detect soil PFASs and OBS.The same samples with different concentrations were used to make the standard curve for each test.The sampling needle was rinsed with methane before each injection and the chromatographic column was washed with methane at least 20 min after determination.All measurements were performed twice.

Statistical analysis

The adsorption kinetics and isotherms were examined by above models using software Origin 8.0.The coefficient of determination(R2)was used to determine the accuracy of these fitting results.To investigate the effects of soil chemical properties,the normalization method was used to standardize the contents of OM,TN,AN,NN in the soils.We calculated the ratios between the contents of soil chemical properties and their corresponding maximum values of the five soils to obtain dimensionless standardized values.Similarly,the maximum adsorption capacities(qm)of the five soils obtained from the Langmuir model fitting were also normalized with the maximum value(qmof S4)as the standard.Then,the normalized values of soil chemical properties(CN)and the normalizedqmwere analyzed by linear regression analysis through Origin 8.0.

RESULTS AND DISCUSSION

Adsorption kinetics

The adsorption rate of OBS on five soils changed over time,as shown in Fig.2.The plots implied that most of the OBS uptake was completed within 12 h,indicating a fast adsorption process(Yueet al.,2017).To further study the adsorption kinetics,four kinetic models including pseudo-first-order,pseudo-second-order,Elovich,and intraparticle diffusion models were used to describe the adsorption kinetics data.The obtained fitting parameters are shown in Table I.

According to the correlation coefficient(R2)values,the pseudo-second-order model is more suitable than the pseudofirst-order model for describing the OBS adsorption kinetics process on soils,suggesting that chemical interaction may exist between OBS and soils in the adsorption process.A previous study(Wuet al.,2018)suggested that valence forces could be produced in the soil adsorption process by sharing or exchanging electrons of the active groups.The Elovich model is usually used for describing adsorption derived from chemical interactions on heterogeneous surfaces.The good fit of the five kinetic curves by the Elovich model(Fig.2c)according to theR2values(Table I)further indicated the chemical interaction between the active adsorption sites of soils and OBS molecules.To further clarify the OBS adsorption process on soils,the intra-particle diffusion model was used to describe the OBS adsorption kinetics(Fig.2d).The OBS adsorption process onto soils can be divided into two stages including the boundary layer and intra-particle diffusion.The adsorption began with diffusion of OBS from the aqueous solution to the out-surface of soils,and then OBS passed across the boundary layer into the inside surface of soils and was adsorbed on the activate sites.The lack of a straight line through the origin suggests that the boundary layer and intra-particle diffusion were all rate-limiting steps.The boundary layer thickness coefficient(cn)of the first stage was significantly smaller than the second stage.Therefore,the boundary layer diffusion step played a more important role in limiting the rate in the initial adsorption process,while the intra-particle diffusion step was the key rate-limiting step in the second adsorption stage.When the adsorption reached equilibrium,intra-particle diffusion was the main factor limiting further OBS adsorption.

Fig.2 Adsorption kinetics of sodium p-perfluorous nonenoxybenzene sulfonate(OBS)on five soils as well as modeling using pseudo-first-order(a),pseudo-second-order(b),Elovich(c),and intra-particle diffusion(d)models.qt is the OBS adsorption amount at adsorption time t.S1=Kastanozem;S2=Calcisol;S3=Leptosol;S4=Histosol;S5=Chernozem.

Adsorption isotherms

The adsorption capacities of OBS on five soils were determined using the equilibrium adsorption tests.Figure 3 shows the adsorption isotherms of OBS on soils,and the OBS adsorption amounts on soils increased sharply with the increase in OBS concentration.Soil with the highest OM content(S4)showed higher adsorption than the other soils,and S1 and S2 showed lower adsorption amounts for OBS possibly due to their low OM content.Previous study(Wuet al.,2018)has indicated that the addition of OM into soils can significantly enhance their adsorption capacity for aromatic organic pollutants.Moreover,previous research has indicated that aromatic organic pollutants bind to soil organic matter stronger than to soil minerals(Ahmedet al.,2014).The Langmuir,Freundlich,Temkin and D-R models were employed to analyze the OBS adsorption isotherms.

The fitting curves are shown in Fig.3 and the recorded parameters of adsorption are showed in Table II.The Langmuir model was based on the monolayer adsorption with homogeneous adsorption sites,while the Freundlich model assumed that the adsorption sites are not equivalent.In terms of theR2values,the Freundlich model is more suitable than the Langmuir model to fit the adsorption data,suggesting the adsorption sites on the soils for OBS adsorption were heterogeneous.This is possibly associated with various adsorption mechanisms including hydrophobic,π-π,hydrogen bonding,and electrostatic interactions(Liet al.,2019).The adsorption capacity of OBS on soils decreased in the order of S4＞S3＞S5＞S1＞S2 based on the Langmuir model,and S4 exhibited the highest adsorption capacity of 7.291 μmol g-1.The Freundlich isotherm constant 1/ncan beused to calculate the adsorption intensity and to explore the favorability of the adsorption process.The 1/nvalue indicates the type of adsorption process(i.e.,irreversible(1/n=0),favorable(0＜1/n＜1),and unfavorable(1/n＞1))(Saltalıet al.,2007).The values of 1/nshown in Table II were lower than 1,indicating a stronger interaction between adsorbate and adsorbent(Boglioneet al.,2019).According to the fitting results of the Temkin model,the obtainedBvalue(a higher value leads to a stronger interaction between adsorbate and adsorbent)followed the order of S4＞S3＞S5＞S1＞S2,confirming their different adsorption capacity for OBS.The D-R model is usually used to identify the occurrence of chemical interaction in the adsorption,andEvalue of＜8 kJ mol-1indicates a physical adsorption,while＞8 kJ mol-1implies a chemical adsorption(Salehet al.,2017).As expected,all of theEvalues were higher than 8 kJ mol-1(Table II),supporting the theory outlined above of the existence of various chemical interactions in OBS adsorption.

TABLE ICalculated parameters of the kinetic models for sodium p-perfluorous nonenoxybenzene sulfonate(OBS)adsorption on five soilsa)

Fig.3 Adsorption isotherms of sodium p-perfluorous nonenoxybenzene sulfonate(OBS)on five soils fitted using Langmuir(a),Freundlich(b),Temkin(c),and D-R(d)models.S1=Kastanozem;S2=Calcisol;S3=Leptosol;S4=Histosol;S5=Chernozem;qe=the OBS adsorption amount at equilibrium;Ce=the equilibrium OBS concentration in solution;ε=the Polanyi potential.

TABLE IICalculated parameters of the isotherm models for sodium p-perfluorous nonenoxybenzene sulfonate(OBS)adsorption on five soilsa)

Adsorption thermodynamics

To analyze the adsorption mechanism between OBS and soils,the adsorption of OBS on two typical soils(S1 and S4)were performed at two temperatures(25 and 50°C)(Fig.4).The thermodynamic parameters were calculated and presented in Table SII(see Supplementary Material for Table SII).The negative values ofΔG0indicated that OBS adsorbed on soils was a spontaneous and feasible process.The positive values ofΔH0indicated that the adsorption was endothermic,which means that higher temperatures will accelerate the adsorption process.The potential adsorption mechanism can be obtained from the value ofΔH0.The low value of about 5 kJ mol-1indicated a major hydrophobic interaction,and the value ranged from 10 to 40 kJ mol-1suggested that the H-bonding was the main adsorption interaction(Kaur and Kaur,2018).According to the magnitude ofΔH0,hydrogen bonding may be considered as the main interaction involved in OBS adsorption on both S1 and S4 due to theirΔH0values of 28.605 and 32.662 kJ mol-1,respectively(Yueet al.,2017).The values ofΔS0were positive,suggesting the affinity of selected soils to OBS and the increase of randomness at the solid/solution interface after OBS adsorption.

Fig.4 Adsorption isotherms of sodium p-perfluorous nonenoxybenzene sulfonate(OBS)on five soils at temperatures of 25 and 50 °C,respectively.qe=the adsorption amount at equilibrium;Ce=the equilibrium OBS concentration in solution;S1=Kastanozem;S2=Calcisol;S3=Leptosol;S4=Histosol;S5=Chernozem.

Effects of solution pH and soil depth

The pH changes both the properties of the soil surface and the organic pollutants in soil,affecting their adsorption behaviors.The effect of pH on OBS adsorption was studied.As shown in Fig.5a,the adsorption of OBS on two soils exhibited a similar tendency,decreasing with increasing solution pH.Some similar tendencies were also found in other studies,suggesting that the change of positive or negative charge of soil surfaces may dominate their adsorption difference for organic pollutants in solutions with different pH(Yueet al.,2017;Huet al.,2019).Since the OBS exists as an anion form in water,a lower solution pH would increase the number of positive surface charges in soil(Irielet al.,2018),resulting in higher adsorption amounts of OBS at lower solution pH.Indeed,the pHpzcvalues of S1 and S4 were found to be about 6.7 and 7.0,respectively(Fig.S1,see Supplementary Material for Fig.S1).At a solution pH of 4 and 6,the soil surfaces were positively charged,and the enhanced adsorption amounts could be attributed to the electrostatic interaction.At a pH of 9,weak electrostatic repulsion was generated,leading to lower adsorption amounts.Soil fertility and pH vary at different depths,influencing the transport of organic pollutants at different depths.As exhibited in Fig.5b,adsorption amounts decrease with the increase of soil depth from 0 to 30 cm.Indeed,the contents of OM,TN and AN decreased with increasing soil depth(Fig.S2,see Supplementary Material for Fig.S2).These fertility parameters of soils exhibited great influence on the soil adsorption capacity for target pollutants,and higher OM and nitrogen-containing organic compounds in soils can enhance their adsorption capacity(Liet al.,2019;Yanget al.,2019).Furthermore,it was reported that the interaction of pollutants with OM depended more on the chemical composition of OM than on the OM content(Ahmedet al.,2014,2015).In this study,the decline of OM,TN and AN contents in the deeper soil layers may cause the weaker affinity of soil for OBS due to the loss of active sites for OBS adsorption.Additionally,the solution pH increased slightly with increase in soil depth(Fig.S3,see Supplementary Material for Fig.S3),further decreasing the adsorption amount of OBS on the soil in the deeper soil layers.It was also found that the soil particles were different at the three soil layers especially at the range of 0.05–0.2 mm(Fig.S4,see Supplementary Material for Fig.S4),and the deeper soil layer had a smaller particle size.However,the soil particle size only showed a slight influence on OBS adsorption on soils(Fig.S5,see Supplementary Material for Fig.S5),indicating that the different particle size of the three soil layers is not the key factor affecting OBS adsorption.

Fig.5 Effects of solution pH(a)and soil depth(b)on sodium p-perfluorous nonenoxybenzene sulfonate(OBS)removal by and adsorption amount on two soils,respectively.Error bars are standard deviation of means(n=2).S1=Kastanozem;S4=Histosol.

Effects of soil chemical properties

The factors affecting the binding of pollutants include the physical and chemical properties of the pollutants,the chemical composition of the soil,and type and strength of the interactions between the pollutant and soil components(Ahmedet al.,2012).To identify the soil properties influencing the OBS adsorption onto soils,different soils with different chemical properties including OM,TN,AN and NN were selected to analyze their influence on OBS adsorption.The correlations between the normalized soil chemical properties(OM,TN,AN and NN)and normalizedqmobtained from the Langmuir model are presented in Fig.6.TheR2value for the OM was 0.58,indicating that the content of OM could influence the OBS adsorption on soils.Some similar results were also found for the adsorption of PFOS on different soils(Qianet al.,2017;Weiet al.,2017).For the short PFASs like PFBS,the influence of OM on PFASs adsorption was negligent(Liet al.,2019).The reason could be the hydrophobic interaction between OM and PFASs(Ahmedet al.,2015),and the fact that hydrophobic PFASs like OBS and PFOS more easily enable strong hydrophobic adsorption.The influence of TN content on the OBS adsorption on soils was also examined,and theR2value of the linear regression analysis was 0.57,suggesting that the TN in soils is a factor of the OBS adsorption onto soils.Some nitrogen-containing groups could be protonated,improving the positive potential of the soil surface(Niuet al.,2010).Our previous study indicated that the electrostatic interaction between the protonated nitrogen/oxygen-containing groups and anionic PFASs is the key adsorption force for the high adsorption amount of anionic PFASs on adsorbents(Wanget al.,2019).These protonated amino or amide groups in soils were involved in the adsorption of anionic OBS.This force can significantly increase the adsorption performance of soils due to the occurrence of electrostatic interaction.To further analyze the effect of soil nitrogen on the adsorption of OBS on soils,two typical soil nitrogen forms including AN and NN were examined,and the relationship between the content of these two types of soil nitrogen and the normalizedqmare shown in Fig.6.Apparently,there was a relatively good linear plot between AN and the normalizedqm(R2=0.68),while the influence of NN on OBS adsorption was not obvious(R2=0.1),indicating the important role of AN in OBS adsorption on soils.Generally,soil has a strong ability to absorb and fix ammonia nitrogen(Liuet al.,2008).The fixed NH+4in soils can be regarded as an effective active adsorption site due to its positive charge,which can adsorb OBSviaan electrostatic interaction.Apart from NH+4,other amino groups like amino and amide are also easily pronated and can enhance OBS adsorption efficiency through electrostatic interaction(Duet al.,2015;Liet al.,2019).

Fig.6 Correlations between the normalized values of soil chemical properties(CN)and the normalized sodium p-perfluorous nonenoxybenzene sulfonate(OBS)maximum adsorption capacities(qm)obtained from the Langmuir model.OM=organic matter;TN=total nitrogen;AN=ammonia nitrogen;NN=nitrate nitrogen.

According to the above discussion of the OBS adsorption on soils,possible mechanisms were proposed(Fig.7).The OBS could be adsorbed quickly on the soil possiblyviahydrophobic,electrostatic and π-π interactions.Organic matter and AN were also important factors involved in OBS adsorption on soil.The contents of OM and AN decreased with increasing soil depth,leading to decrease in the OBS adsorption amount on the soil with increasing soil depth.

Fig.7 Schematic diagram for the adsorption of sodium p-perfluorous nonenoxybenzene sulfonate(OBS)on soil.

CONCLUSIONS

In this study,the adsorption behaviors of OBS on soils with different physicochemical properties were investigated.The adsorption kinetic data were fitted better using pseudosecond-order and Elovich models than using the pseudofirst-order model,suggesting that a chemical interaction may be involved in the OBS adsorption on soils.Boundary layer diffusion played an important role in limiting the early OBS adsorption process,and intra-particle diffusion gradually became the key rate-limiting step when approaching the adsorption equilibrium.Apart from the traditional hydrophobic interaction involved in PFASs adsorption on soils,the π-π,hydrogen bonding,and electrostatic interactions were possibly involved in the OBS adsorption process.The top surface soils exhibited a higher OBS adsorption amount than the deep soils due to higher contents of OM and AN and lower pH,which are favorable for the generation of new active sites and stronger adsorption force.Furthermore,OBS adsorbed on soil was a spontaneous endothermic process.This study is very helpful for understanding the transport of OBS and its interfacial behavior on solid surface.

ACKNOWLEDGEMENT

This study was supported by the Fun damental Research Project of Qinghai Province(No.2019-ZJ-935Q),China,and the Beijing Natural Science Foundation(No.8184079),China.

SUPPLEMENTARY MATERIAL

Supplementary material for this article can be found in the online version.