Journal of Sciences, Islamic Republic of Iran 31(4): 305 - 319 (2020) http://jsciences.ut.ac.ir University of Tehran, ISSN 1016-1104

305

Natural Gas Condensate Desulfurization via Polyacrylonitrile/Ag Nanocomposite Nanofibers:

Optimization and Kinetics/isotherms Studies

R. Dadashvand-nigjeh1, A. Mollahosseini2*, M. Alimoradi1, M. Ramezani1

1 Department of Chemistry, Faculty of Science, Islamic Azad University, Arak, Islamic Republic of Iran 2 Research Laboratory of Spectroscopy & Micro and Nano Extraction, Department of Chemistry, Iran

University of Science and Technology, Tehran, Islamic Republic of Iran

Received: 16 June 2020 / Revised: 27 September 2020 / Accepted: 24 October 2020

Abstract In the present study for the first time, a novel polyacrylonitrile-silver nanocomposite

nanofiber was synthesized by electrospinning for adsorbing sulfur compounds in natural gas condensate. The synthesized material was characterized by SEM, EDAX, and XRD. The presence of C, N, O, and Ag proved that the composite was synthesized successfully. In order to achieve the best adsorption conditions, the effect of contact time, adsorbent dosage and initial sulfur compound concentration was examined. Under optimal values, efficiency of greater than 90% was found. In addition, different isotherms and kinetics models were tested to describe the sorption process. It was found that Freundlich (0.9900) was superior to Langmuir (0.9688), Temkin (0.9648) and Dubinin-Radushkevich (0.8273) models, revealing that sulfur compounds tend to for, multilayers on the heterogeneous surface of polyacrylonitrile-silver nanocomposite nanofiber. The energy value of the adsorption was 23.57 kJ/mol, indicating chemisorption reactions. Based on kinetics studies, the desulfurization by nanofibers followed Pseudo-second-order and Elovich kinetics. Finally, the desulfurization function of nanocomposite was studied and validated using adsorbent columns. The obtained results demonstrate polyacrylonitrile-silver nanocomposite nanofiber as a promising material in the field of desulfurization. Keywords: Adsorption; Nanofiber; Natural gas condensate; Sulfur compounds.

* Corresponding author; Tel: +982173228342; Fax: +982173021578; Email: amollahosseini@iust.ac.ir

Introduction Gas condensate is considered as one of the most

valuable products derived from natural gas reservoirs, the price of which is slightly higher than those in crude oil on global markets. Cuts 5 and 6 are the major

components of gas condensate [1]. In addition, sulfur compounds are one of the most important contaminants of gas condensate which release toxic gases by combustion, damage metals, catalysts in the engine, and fuel cells [2, 3]. In this regard, the existing general methods for removing sulfur compounds of fuels

Vol. 31 No. 4 Autumn 2020 R. Dadashvand-nigjeh, et al. J. Sci. I. R. Iran

306

include: oxidation methods, adsorption via a solvent, surface absorption in stationary phase bed and hydrodesulphurization [4]. However, there are some disadvantages for the above-mentioned methods: such as the requirement for difficult operating conditions such as high temperature and high pressure; the necessity for expensive catalysts; high energy consumption and failure to remove some sulfur compounds. Therefore, providing a new, low-cost, and practical method is needed to remove these compounds from gas condensate to enhance its quality.

Traditional materials for adsorptive removal of sulfur are zeolites, activated carbon, and alumina, which are regarded as porous materials and perform desulfurization mainly via adsorption. The implementation of transition metal ions incorporated in these materials leads to reactive adsorption via σ–π bonding between metal ions and adsorbates, which results in enhancing their performance [5]. Bakhtiari et al. [6] reported the optimization of sulfur adsorption over Ag-zeolite nano-adsorbent with a thiophene adsorption efficiency close to 50 ppm for 5.53% metal content. Another study, the combination of electrochemical oxidation and solvent extraction approaches was suggested to reduce the sulfur content in condensate gasoline down to 13.1 μg/g with a desulfurization efficiency of 99.62% [7]. Furthermore, Adedibu et al. [8] synthesized zig-zag 1D coordination polymer of copper(II) [Cu(4-mba)2(H2O)3] and revealed that this complex had the potential for desulfurization of fuel with an observed adsorption capacity of 9.6 mg/g at 32 0C for 6 h. More recently, the polystyrene nanofibrous membrane loaded by Ag+ cations can be employed for deep desulfurization [9]. Further, the desulfurization of gasoline was conducted by acrylonitrile electrospunned nanofibers, including lead nanoparticles [10].

Nanocomposites and nanofibers are among the candidates which are used for filtration and purification as well as composites strengthening, fuel cells, electronic and tissue engineering [10-12]. The high surface-to-volume ratio, high flexibility in functionalizing of surfaces, and excellent mechanical properties are considered as some of the important features related to nanofiber composites compared to conventional fibers. Chemical vapor deposition, hydrothermal technique, mechano-chemical methods and electrospinning are of the primary composite preparation methods, that among them electrospinning has gained attention of many scholars [13–15]. Electrospinning is regarded as a simple and cost-effective approach for the synthesis of nanofibers [16, 17]. In the recent years, many researchers have tried to

synthesize fibrous composite for the removal/adsorption of nickel [18], chromate [19], methylene blue [20], congo red [15], methyl orange [21], protein [22], and mercury [23]. In view of sulfur compound removal, few number of studies have carried out. Desulfurization of gasoline using acrylonitrile electrospun nanofibers and lead nanoparticles led to removal efficiency and adsorption capacity of 93.31% and 31.4 mg S/g, respectively [10].

Utilization of noble metals with nanomaterials, including Pt, Ag and Au could improve the capacity of the composite, and many scholars suggested them as a promising material in the field of environmental technology and remediation. ZnO/Ag/GO nanocomposite was employed for the removal of naphthalene, and resulted in 92% degradation of the pollutant [24]. Yang et al. [25] proposed ZnO/Ag nanocomposites for dye elimination. Ammonia adsorption in Ag/AgI-coated hollow waveguide was also reported by Jinyi et al. [26]. Another nanocomposite namely, biogenic Ag@Fe was utilized for methyl red removal and resulted in maximum adsorption capacity of 125 mg/g [27]. In the case of gases, dimethylsulfide and t-butylmercaptan removal from pipeline natural gas fuel by means of Ag-zeolite was studied [28]. Nair and J. Tatarchuk (2011) attempted to remove species from hydrocarbon fuels by supported silver adsorbents [29]. Desulfurization of hydrocarbon fuels was also conducted by Ag/TiO2 [30]. These reports are a clear demonstration of the fact that Ag could have superior adsorption capacity in the adsorption process to remove the contaminant such as sulfur species.

In the current study for the first time, electrospun polyacrylonitrile-silver nanocomposite was synthesized and utilized for desulfurizing condensate. The prepared nanocomposite was characterized, and then employed to investigate the effect of time and adsorbent dosage. The whole process was studied by kinetics and isotherm models.

Materials and Methods Materials

The PAN powder, DMF, AgNO3 and all other chemicals were all purchased from Sigma Aldrich (with high purity) and were used without further purification. Distilled water was used during the study. Characterization

First, a scanning electron microscope (FE-SEM; LEO 440i) was used to evaluate the morphology and surface structure of the synthesized nanocomposite. Then, the

XRD patterresearch wadiffractometethe sample w(TS3000, Th Synthesis of

In order tappropriate a80000) powdand stirred ftemperature. powder was hours by usiAgNO3 soluDMF.

For electrointo a syringdiameter attsyringe pumpthe needle tisupply (ES3aluminum shThe infusionneedle to corespectively. Measuring t

In order synthesized mL of gaconcentrationsulfur in theanalyzer afteof adsorptioequation (1):

where qesynthesized adsorbent, concentrationrespectively, liters, and m grams.

Studies relat

In an Erlecondensate poured. Thenplaced insideminutes, andThen, the e

Natural Gas C

rn of syntheas provided er. The concenwas measuredhermo).

f nanocomposto prepare 11amount of poder was weighfor 2 hours b

Then, the added to thi

ing a magnetiution toward t

ospinning, thege with a staitached to. Tp (KDS 200, ip was conne30P, Gammaheet (20×20 cn rate, appliedllector were

the sample adto obtain t

nanocomposias condensatn of 100 ppm sample was

er stirring for n was calcul: =e represents nanofibers inC0 and C n of sulfur and V indiis regarded a

ted to adsorptenmeyer flaskcontaining 4

n, different ame the flask, thd the amount equilibrium a

Condensate D

esized nanomby the S

ntration of remd using a tota

site % w/w solut

olyacrylonitrilhed and addedby using a strequired amois solution anic stirrer to pthe total weig

e prepared solinless steel ne

The syringe KD Scientifi

ected to a higa High Vo

cm) was used d voltage, and1.5 ml/h, 15

dsorption capthe adsorptioites, they werte with a m and the am

measured usi10 min. Fina

lated by usin

∗ the adsorpti

n mg of sulfare the in

in the saicates the samas the mass of

tion isothermsk, the amount 400 mg/L tomounts of nanhe mixture wof total sulfu

adsorption ca

Desulfurization

material in Siemens D50maining sulfu

al sulfur analy

ion of PAN, le (PAN, MWd to DMF solvtirrer at ambiount of AgNnd stirred for repare 45% wght of PAN

lution was draeedle with 1 mwas pushed c Inc., MA),

gh voltage powltage, FL) as the collec

d distance of KV, and 15

acity on capacities re poured int

primary sumount of residing a total sually, the capacng the follow

(1) ion capacity fur per gram

nitial and fiample in mmple volumef the adsorben

s of 30 mL of otal sulfur wocomposite w

was stirred forur was measurapacity (qe) w

n via Polyacryl

307

this 000

ur in yzer

the W = vent ient

NO3 r 24 w/w and

awn mm

by and wer and

ctor. the cm,

of to 5 ulfur dual ulfur city

wing

of m of final

mg/L e in nt in

gas was

were r 30 red. was

cal

(mcon(mand

Kin

conpouinsamintwa

Ch

motheAsstrademsucthenanAgrevin ttheTh

Fn

lonitrile/Ag Na

lculated from

In this equati

mg/L), Ce incentration o

mg/L), V repred W is the ma netic studiesIn each two sndensate incluured. Then, 0side each of t

mount of sulervals and th

as calculated f

haracterizationSEM images

orphology of te SEM image s shown, the aight strips monstrates thccessfully duee EDAX analynocomposite. g were detectvealing that pothe prepared m

e synthesized he characteri

Figure 1. SEnanocomposite

anocomposite

equation (2): = ∗ on, C0 indicatis considere

of sulfur comesents the voass of the adso

eparate Erlenmuding 150 and0.02 g of the nthe flasks, thefur was mee adsorption

from the equat

Results andn s were usedthe synthesizeof polyacrylosynthesized

with a fairlyhat the fibroue to its uniformysis of electroAccordingly,

ted in the syolyacrylonitrilmaterial. In ad

nanocomposistic peak

EM images nanofiber.

Nanofibers …

(2

tes the initial ed as the

mpounds afterolume of soluorbent in gram

meyer flasks, d 400 mg/L tonanocomposit

he mixture waeasured at dcapacity at dtion (2).

d Discussion

d to observeed material. Fionitrile /Ag nafibers are in

y uniform dius structure wm structure. Fiospun polyac, the elementsynthesized nale and Ag areddition, the XRsite is shown of the pol

of polyacrylo

…

2)

concentrationequilibrium

r 30 minutesution in liters,ms.

10 mL of gasotal sulfur waste was placedas stirred, thedifferent timeifferent times

n

the surfaceigure 1 showsanocomposite.

the form ofistribution. Itwas preparedigure 2 showsrylonitrile/Ag C, N, O, and

anocomposite,e both presentRD pattern ofin Figure 3.

lyacrylonitrile

onitrile-silver

n m s ,

s s d e e s

e s . f t d s g d , t f . e

Vol. 31 No.

crystalline strelated to silv(311) are lorespectively, in the nanoclearly indicAg were bonanocomposbeneficial in Optimization

The curreFirstly, the esulfur-composecond part,

4 Autumn 20

tructure is locver phase incl

ocated at 2θ’s which confirocomposite.

cative of the fth presented ite has a desulfurizatio

n ent study wasffect of contaound concen the obtained

Fi

F

020

cated at 2θ oluding (111), s of 38, 45,

rms the presenThese chara

fact that polyin the materifibrous struc

on process.

s carried out act time, adsorntration was d results wer

igure 2. EDAX

Figure 3. XRD

R. Dad

of 17. The pe(200), (220), 65.5, and 7

nce of two phaacterizations yacrylonitrile ial, and also cture which

in two sectiorbent dosage, studied. In

re modeled w

X analysis for el

analysis for ele

ashvand-nigje

308

eaks and 8.5, ases are and the

is

ons. and the

with

varme

Eff

of peran adeprosulconpernanfin

lectrospun poly

ectrospun polya

eh, et al.

rious isothermechanism of su

ffect of contacLike any chem

the most rformance of efficient bala

equate contacocesses. Accolfur removalncentration leriod of 0-150 nofiber comp

ndings are exh

acrylonitrile /A

acrylonitrile /Ag

m and kinetiulfur-species e

ct time and sumical reactionimportant pthe whole sysance betweent time must b

ordingly, the l was examvels (150 andmin. The dosa

posite was mibited in Figur

Ag nanocomposi

g nanocomposit

J.

ics models telimination.

ulfur concentrns, time is addparameters, astem. Generaln adsorbent anbe considered

effect of conmined at twd 400 ppm) anage of polyac

maintained at ure 4. Based on

ite.

te.

Sci. I. R. Iran

o clarify the

ration dressed as oneaffecting thelly, to achievend adsorbate,in adsorption

ntact time onwo differentnd in the timerylonitrile/Ag0.02 g. The

n Figure 4a, it

n

e

e e e ,

n n t e g e t

indicates thremoval effimaximum vconcentrationFigure 4b shtime intervalconcentrationinitial concen

A closer initial step oincreased rap

Natural Gas C

at by increaiciency tends value of 44ns of 150 and

hows the concls, and it cann of sulfur rntrations of 15look at the cf adsorption ppidly, and the

Figure

Condensate D

asing the coto improve,

4% and 22d 400 ppm, recentration of sn be figured oreached 84 an50 and 400 ppcurve (Fig. 4aprocess the reen experienced

4. Effect of con

Desulfurization

ontact time, and reached .5% for su

espectively. Asulfur at differout that the find 310 ppm pm, respectivea) reveals thamoval efficied a smooth sl

ntact time (a), a

n via Polyacryl

309

the the

ulfur Also,

rent final

for ely. at at ency lope

to thethetimattrcomadsremoccremsitu

and sulfur conce

lonitrile/Ag Na

reach an eque reaction tome performanceme was founributed to themposite has absorption, leadmoval efficiecupied by sumoval efficienuation in wh

entration in diff

anocomposite

uilibrium at 90me from 90 to e of the systnd 90 min.

fact that polybundant vacanding to a quency [31]. Hulfur moleculncy slope wouhich the rate

ferent time inter

Nanofibers …

0 min. Furtheo 150 min hadtem. So, opti

These resulyacrylonitrile/nt sites at inituick enhance

However, theles through

ould decrease e of the ad

rvals (b).

…

er increase ind no effect onimum contactlts could be/Ag nanofibertial step of theement in theese sites aretime, so theuntil reach a

dsorption and

n n t e r e e e e a d

Vol. 31 No.

desorption fidentical (ocalso reported

Figure 4concentrationaffects the rsulfur removgreater than concentrationtechnique. Swas constantnumber of logically lowremoval effadsorption caand is showadsorption ctime contact,33 mg/g for greater adconcentration400 ppm thango toward adadsorption ca Effect of ads

Adsorbentthat is requefficiency. T

Figure 5. polyacrylon

4 Autumn 20

from the surcurred at 90 m

d by previous 4a also demn is vitally iremoval efficval efficiencythat of 400 pn is more ince polyacryt during the eavailable sit

wer sulfur conficiency. Moapacity by timwn in Figurecapacity was , and the maxi150 ppm an

sorption can is due to gren 150 ppm whdsorbent surfaapacity [32].

sorbent dosagt dosage is deired to reach

The importan

The fluctuationitrile/Ag nanof

020

rface of themin). Similar scholars.

monstrates thaimportant becciency of they at 150 ppmppm, proving

beneficial ylonitrile/Ag nexperiments, tes on the centration mig

oreover, the me variations e 5. At bothincreased by

imum adsorptd 45 mg/g fo

apacity at eater gradienthich forces suace, and conse

ge efined as the h the desirednce of this p

ons of adsorpfiber composite

R. Dad

adsorption deductions h

at initial sucause it direc pollutants. T

m was 1.95 timthat lower sufor adsorpt

nanofiber dosthere are limiadsorption, ght end in hig

fluctuations were calculat

h concentratiy increasing tion capacity wor 400 ppm. T

higher sut concentrationulfur moleculeequently a grea

mass of mated sulfur remoarameter is a

ption capacity .

ashvand-nigje

310

are have

ulfur ctly The mes

ulfur tion sage

mited and

gher of

ated, ion, the

was The

ulfur n at

es to ater

erial oval also

attrdeseffsulinvFigto figdotheby frodoeffattrdoadsthaeff

46froeffowpolto mograsul

by time cha

eh, et al.

ributed to itssulfurization. fect of polyaclfur removal vestigated. Thgure 6. The ad0.4g, and su

gure clearly psage, the reme removal effi

increasing om 0.02 to 0.4sage of adsorficiency. The mributed to thesage, more sorbate molecat desulfurizficiency by incIn contrast, t.50 to 23.33 m

om 0.02 to ficiency, it ca

wns higher lyacrylonitrilethe gradient b

olecules. In otadient betweenlfur molecu

anges for sulf

s direct influeAccordingly,

crylonitrile/Agand also

e experimentadsorbent dosalfur concentrroves that by

moval efficienciency improvpolyacrylonitr40 gr, respectrbent is benefmain reason fe fact that by

vacant sitesules [33]. Zhe

zation perforcreasing the Athe adsorptionmg/g by increa

0.4 g, respean be observe

values ae/Ag nanofibebetween adsorther words, at n adsorbent anles are fo

fur adsorption

J.

ence on the , in the curreg nanofiber dadsorption c

al findings areage was rangiration was 40y increasing tncy excelled. ved from 7.75trile/Ag nanotively. It stateficial in termfor such behay increasing ts become aeng et al. [34]rmance exhi

Ag-based adson capacity deasing the adsoectively. Un

ed that adsorpat lower er. This obserrbent dosage at low adsorbennd adsorbate i

orces to m

(150 and 40

Sci. I. R. Iran

final cost ofent study the

dosage on thecapacity wase illustrated ining from 0.0200 mg/L. Thethe adsorbentNumerically,

5% to 77.75%ofiber dosagees that higher

ms of removalvior might bethe adsorbentavailable for also reportedibited better

orbent. ecreased fromorbent dosagelike removal

ption capacitydosage of

rvation is dueand adsorbatent dosage, theis of high that

move toward

00 ppm) by

n

f e e s n 2 e t ,

% e r l e t r d r

m e l y f e e e t d

polyacrylonitadsorbent capadsorbate mothat are somthe accompli Isotherm of

The isothindicate the which play aprocess. In aas one of thewhich descriand adsorbaattempts to and adsorbattemperature have been pLangmuir, Temkin are k

Langmuir adsorptions hand Freunmultilayer suLinear formsexpressed as

where K

Natural Gas C

trile/Ag surapacity. By incolecules face a

mehow redundishment of the

adsorption hermal equati

adsorption a significant roaddition, adsoe important faibes the interaate moleculesprovide a rete in the equand pH. Up

proposed by rFreundlich,

known as the mr equation ison a homoge

ndlich isotheurface adsorpts of Langmui[11,12]: = 1ln =

KL (L/mg) is

Figure

Condensate D

rface, resulticreasing the aa plenty numbdant, and reme adsorption p

ions of a spproperties of

ole in designinorption isotheractors in desigactions betwes. In fact, islationship be

uilibrium stateto now, cou

researchers thDubinin-Rad

most commons employed feneous surfac

erm model tion in heterogir and Freund1 + 1+ 1

s the Langm

6. the effect of

Desulfurization

ing in higadsorbent dosaber of empty smain vacant a

rocess [32].

pecific adsorbf the adsorbng the adsorptrm is considegning the syst

een the adsorbotherm equattween adsorbe under constuntless isotherhat among thdushkevich, n ones. for single lace, on the otis suitable geneous systedlich models

muir equilibri

adsorbent dosa

n via Polyacryl

311

gher age, sites after

bent ent, tion ered tem bent tion bent tant rms

hem, and

ayer ther for

ems. are

ium

conwitcapengtheandme

arereswawhsatcomcanthaappfacpolsurof thetenadshigelupar[35

age on the remo

lonitrile/Ag Na

nstant which dth the adsorbpacity referregross all the e Freundlich d n is the heasure of intenThe constant

e summarizedspectively. Thas selected by hich as its valtisfactorily demparing the Rn be stated than Langmuir propriateness ct that sulfulyacrylonitrilerface acts as aFreundlich m

e whole procend to estabsorptions. Altgh R2 value oucidate the rameter (RL) 5]:

oval efficiency a

anocomposite

describe the dbate and qmax d to the amoavailable siteconstant relat

heterogeneity nsity of adsorpparameters anin Table 1 a

he most approthe coefficien

lue gets closeescribe the eR2 value of that Freundlicequation to of Freundlichr molecules e/Ag surface ta heterogeneou

model was greaess was favoralish chemicathough Freunof Langmuir

process, dican be empl

= 1 +

and adsorption c

Nanofibers …

dependence of(mg/g) is the

ount of sulfues of nanocomted to adsorpcoefficient a

ption. nd linear plotsand illustratedopriate adsorpnt of determiner to one, theexperimental the examined ch assumptionexpound the

h model is indtend to ag

to form multilous media. That than one, s

rable, and sulfal bonds v

ndlich was thecould not beimensionless loyed which 1+

capacity.

…

f binding sitese single layerur required tomposite. Kf isption capacityand 1/n is a

s of isothermsd in Figure 7,ption isothermnation (R2) inmodel couldfindings. Byisotherms, it

ns was better results. The

dicative of thegregate overlayers and thee n parameterignifying thatfur moleculesvia chemicale best model,e ignored. To

equilibriumis written as

s r o s y a

s ,

m n d y t r e e r e r t s l ,

o m s

Vol. 31 No.

Based onadsorption irreversible. composite shindicating thasuggest thatFreundlich athe adsorbenis expected.

For furthe

Table

4 Autumn 20

n the RL valis favorableSulfur adsor

howed the Rat the processt the whole assumptions, nt monolayer d

er investigatio

1. Calculated isIsotherm

Langmuir

Freundlich

DR

Temkin

Figure 7. Th

020

lue it can be, unfavorabrption by poly

RL value of los is favorable.

adsorption however, in disposition of

on, Dubinin-R

sotherms paramm

r

h

e linear plots if

R. Dad

be said that ble, linear, yacrylonitrile/ower than unThus, the resis governed some regionssulfur molecu

adushkevich w

meters for sulfur

f isotherm mode

ashvand-nigje

312

the or

/Ag nity, sults

by s of ules

was

utistuadspotRaonlconRa[11

r species adsorpParameter

qm KL R2 RL

Kf n

1/n R2

qm β R2

b A R2

els: (a) Langmu

eh, et al.

lized that haudies. Micropsorptive mechtential theo

adushkevich, ily for homognstant adso

adushkevich (D1]:

ption by electro

uir, (b) Freundli

s been emplopore volume hanism and thry. Of adt can be menti

genous surfaceorption potD-R) linear eq=

ospun polyacryl

00

00

00

100

ich, (c) D-R, an

J.

oyed in manyfilling is ass

the model is dvantages otioned that thises, and also dtential. Thequation could−

lonitrile/Ag comValue 59.88

0.0068 0.9688

< 1

2.71 2.11

0.4746 0.9900

41.93

0.0009 0.8273

87.0071 0.04461 0.9648

nd (d) Temkin.

Sci. I. R. Iran

y equilibriumsumed as thebased on the

of Dubinin-s model is notdo not intende Dubinin-

d be written as

mposite.

n

m e e -t d -s

Natural Gas Condensate Desulfurization via Polyacrylonitrile/Ag Nanocomposite Nanofibers …

313

= 1 + 1 where qe (mg/g) is the amount of sulfur compounds

adsorbed per unit mass of polyacrylonitrile/Ag at equilibrium; qmax (mg/g) is the maximum adsorption capacity of polyacrylonitrile/Ag; β (mol2/kJ2) is a constant related to the adsorption energy; and ε (kJ/mol) is the adsorption potential. The constant parameters are obtained from linear plot of lnqe versus ε2.

Temkin isotherm model was proposed based on the adsorption energy in which states that adsorption energy tends to decreased linearly by occupation of the empty sites of polyacrylonitrile/Ag by sulfur [36]. This model is written as: = +

where b is called Temkin constant, relating to the heat of adsorption process, and A is Temkin isotherm constant. The parameters are calculated from slope and intercept of linear plot of qe versus lnCe.

Figure 7 and Table 1 show the linear plot and constants of the D-R and Temkin isotherms, respectively. According to the obtained R2 values, it can be stated that Temkin was more suitable than D-R model. The appropriateness of Temkin isotherm over D-R model could also be figured out by the semi-large difference between the maximum experimental adsorption capacity and D-R qm value. Furthermore, D-R isotherm could provide substantial information regarding the adsorption energy, which is calculated by: = 12

where E is the adsorption energy (kJ/mol). Generally, according to E value adsorption isotherm is categorized into physical adsorption and chemical adsorption. Physical sorption processes have low adsorption energy that is usually lower than 8 kJ/mol [37]. In the current study for sulfur adsorption by polyacrylonitrile/Ag composite, the E value was found 23.57 kJ/mol,

demonstration chemical interactions between adsorbent and adsorbate. Considering all four isotherms, it can be deduced that the order of isotherm fitting is as follows:

Freundlich > > > − Therefore, it can be concluded that sulfur adsorption

follows Freundlich assumptions. Finally, as shown in Table 2, a comparison was made between the desulfurization performances of the nanocomposites developed in the present study with those of the previous reports. As shown, the adsorption capacity of nanofiber described in this work is considerably higher than those in previous reports, which demonstrates its higher adsorbent ability for desulfurizing the gas condensate. Kinetic studies

The rate of chemical reactions states how fast the system is working, and could highly affect the in-situ operational parameters. Accordingly, kinetic models are categorized into two different groups, providing different insights regarding the reactions. The first group, called solute concentration-based, considers the fluctuation of the adsorbate concentration regardless of the adsorbent presence. This approach intends the initial and final concentration of the sulfur compound in the solution. In another perspective, the mass of the adsorbent and sulfur concentration changes is considered, known as adsorbent dosage-based models.

Among solute concentration-based models, first order and second order equations have been utilized in many investigations. First and second order rate equations are expressed as follow [38]: = − 1 = + 1

where Ct (mg/L) and C0 (mg/L) refer to concentration of sulfur at time t and initial sulfur concentration, respectively. K1 and K2 are the constant parameters of

Table 2. Comparison of the adsorption capacity of thesynthesized PAN/Ag nanofibers with other works in desulfurization proccess

Adsorbent Adsorption capacity (mg/g adsorbent)

Reference

30 wt% Ni/SBA-15 with a low-sulfur diesel 1.7 [44] Nickel nanoparticles inside the mesopores of MCM-41 b 1.67 [45] Pelletized Ce–Y zeolites 0.86 [46] 30 wt% Ni/SBA-15 1.7 [47] Ni-based adsorbent 14.2 [48] ACFH–Cu+ 19.0 [49] Graphene-like boron nitride 28.17 [50] PAN–ABS–Pb 31.4 [10] PAN/Ag nanofiber 102 In this work

Vol. 31 No.

the model, anvs. t for firResults are su

Accordingthat at low ssecond modvalues of grcompound coorder model removal prosuggested polyacrylonitmodel.

Up to nowmodels havconsidering pseudo-first (PSO) are thto approve thgenerally exp

where qpolyacrylonitequilibrium ((min-1) and Kthe pseudo-s

The kinetkinetic mode

Initial Sulconcentrat150 400

4 Autumn 20

nd found by lrst order andummarized ing to the obtaisulfur compoudels are almoreater than 0.oncentration (

could not bocess mechanthat sulfur trile/Ag comp

w, various ade been prop

different aorder (PFO

e most populahe adsorptionpressed as [39−=qe (mg/g) trile/Ag comp(min) and at K2 (g mg-1 mecond order ratic parameterels were calcu

Tablfur tion

020

linear plot of d second ordn Table 3 and ined results, und (150 mg/ost well sinc9. However, (400 mg/L), ie appropriatenism. Hence,

compoundposite follows

dsorbent dosagposed by maassumptions.

O) and pseudar equations a

n kinetic. This9]: = − +1 + 1

and qt (mposite adsorpttime t (min),

min-1) are the ate constants. s for the PFulated from t

Figure 8. L

ble 3. The value

R2 0.93520.6971

R. Dad

lnC vs. t andder, respectivFigure 8. it could be sL) both first e they have at higher su

t seems that f to expound

it is generad removal

the second or

ge-based kineany research

Among thdo-second orand were selecs two models

mg/g) are tion capacitierespectively.pseudo-first O and the Pthe intercept

Linear plot of fi

es of constant pFirst order m

2 1

ashvand-nigje

314

d vely.

seen and R2

ulfur first the

rally by

rder

etics hers, hem, rder cted are

the s at K1 and

PSO and

sloconof vieverinshowfacequexh0.9adsexpeno obwa17of ThadsPSarrrpmocccomppmovepolexpste

irst order (a) an

parameters of fir

model K1

0.0043 0.0012

eh, et al.

ope of linear nstant paramethe models co

ew of R2 valury high valuestance, for 40wever, of PFct that PSOuilibrium datahibited high R95. So, more sorption capaerimental) result large differe

served. In casas 45.10 mg/g.47 and 45.87

qe, experimentalherefore, it cousorption by pO assumption

rival [40], andm was 1.47 ticurrence maympound molem is higher ter PFO firlyacrylonitrilepected since P

eps of adsorpt

nd second order

rst order and se

plots preseneters are summould be verifiues, it shows e close to uni0 ppm, R2 vO it was 0.96 is more s

a. It must notR2 values whidiscussion co

acities obtaineand kinetic mnce between se of 400 rpmg, and qe, kineti7 mg/g, respe

- qe, kinetic wuld be conclupolyacrylonitrn. K2 value is d in the currenimes higher thy attribute tecules in the han 150 ppmr sulfur coe/Ag compoPFO is frequetion, and it is

(b) models.

econd order modSecond o

R2 0.9617 0.9471

J.

nted in Figurmarized in Tabied by R2 valu that PSO deity compared

value for PSO656. It is indsuitable to t be ignored tich were bothould be carrieed from expe

models (qe, kinetithese two va

m, for instancetic for PFO anectively. Clearwas negligibuded that sulfrile/Ag compa sign of equnt study, K2 vhan that of 1to the numb

solution in wm. Appropriateompound adosite has aently applied s good almost

dels. order model

K5E3E

Sci. I. R. Iran

re 9, and theble 4. Validityues and qe. Inemonstrated a

to PFO. ForO was 0.9998;

icative of thedescribe the

that PFO alsoh greater thaned out by theerimental (qe,ic). Generally,alues must bee, qe, experimentalnd PSO wererly, the value

ble for PSO.fur compoundposite obeyeduilibrium statevalue for 40050 ppm. This

ber of sulfurwhich at 400eness of PSOdsorption byalready beenfor the initialt for physical

K2 -05 -06

n

e y n a r ; e e o n e , , e l

e e .

d d e 0 s r 0 O y n l l

adsorption [3on polyacrychemical bchemisorptio

Elovich ikinetic mode[42]. This mo

Figure 9. Lpolyacryloni

Table

Pseu

do-fi

rst

orde

r Ps

eudo

-sec

ond

orde

r El

ovic

h Natural Gas C

32,41]. It is bylonitrile/Ag bonds withon. is also knowel to describe odels implies

Linear plots of pitrile/Ag.

e 4. The constanKinetic mo

Eq

qeqe, ex

Eq

qeqe, ex

Eq

Condensate D

believed that fare capable

h sulfur c

wn as a pracchemical reaa multilayer a

pseudo-first or

nt parameters oodel

quation K1

e, kinetic xperimental R2

quation K2

e, kinetic xperimental R2

quation α β

R2

Desulfurization

functional groe of providcompounds

ctical adsorptction mechanadsorption, an

der (a), pseudo

f examined kin

n via Polyacryl

315

oups ding

via

tion nism nd is

exp

thesho

o-second order

netics models fo

150 ppm= 0.0585 +0.058520.1433.100.9943= 0.0292 +0.007234.2433.100.9982

= 4.954 +46.4370.20180.9823

lonitrile/Ag Na

pressed as [43= +where (g/m

e initial adsoown in FigureFor both con

(b), and Elovi

or sulfur compom + 3.00335

3 + 0.11722

2

+ 11.787 8 3

anocomposite

3]: +

mg) is the conorption. The e 9 and Table 4ncentration, th

ch (c) for sulfu

und removal by= −

= 0

= 5

Nanofibers …

nstant and β (plot and pa

4. he R2 value w

fur compound a

y polyacrylonitr400 ppm −0.0552 + 2.80.0552 17.47 45.10 0.9656 0.0218 + 0.040.0106 45.87 45.10 0.9998

5.5381 + 22.4318.182 0.1805 0.9531

…

(mg/g min) isarameters are

was extremely

adsorption by

rile/Ag 8613

448

435

s e

y

Vol. 31 No.

high (0.982respectively)well-fitted thall three kinePSO and Eprocess. It dcreate chemipolyacrylonit Efficiency st

The experto investigatcomposite n

Figure 10. nanocomposi

4 Autumn 20

23 and 0.95) suggesting the sulfur competics models, lovich are ab

demonstrates ical bonds wtrile/Ag.

tudies rimental adsote the efficinanofibers f

(a) Glass coluite nanofiber (a

020

531 for 150that Elovich apound adsorpt

it is fair to sble to descrithat sulfur m

with heterogen

orption columency of poly

for the rem

umn with 0.5 adsorbent= 20 g

R. Dad

0and 400 ppassumptions mtion. Considersuggest that bibe the sorpt

molecules tendneous surface

mn was employacrylonitrile/

moval of su

cm diameter g, initial sulfur c

ashvand-nigje

316

pm, may ring both tion d to e of

oyed /Ag

ulfur

compurwit20 glacon12mLof 1, be Fig

and (b) adsorpcontent= 1250 m

eh, et al.

mpounds frorpose, a glassth a glass wog of the synth

ass column, ndensate was 50 mg/L withL/min as showgas condensaindicating tharecovered, th

gure 10b. In order to

ption of sulfurmg/L, flow rate

om natural gcolumn of 0.

ool placed at thesized nanoc

a tank conplaced with

h a regulator wn in Figure 1ate, the C/C0 rat the adsorbenhe plot of th

conduct retr

r using colume= 6 ml/min).

J.

gas condensa.5 cm in diamthe bottom an

composite. At ntaining 5 lthe total sulfvalve at a fl0a. After pass

ratio approachnt is saturatedhis results are

rieval studies

mn packed with

Sci. I. R. Iran

ate. For thismeter was used

nd filled withthe top of theiters of gas

fur content oflow rate of 6sing 2160 mLhed a value ofd and needs toe depicted in

, 0.02 gr of

h synthesized

n

s d h e s f 6 L f o n

f

Natural Gas Condensate Desulfurization via Polyacrylonitrile/Ag Nanocomposite Nanofibers …

317

polyacrylonitrile/Ag nanocomposites was placed in 10 mL of gas condensate with the total sulfur 150 mg/L and the total sulfur content was reached to 90 mg/l (40% adsorption) after 30 minutes. Then, the nanocomposites were placed under hydrogen gas purge for 10 minutes and again the same nanocomposites were inserted into 10 mL of gas condensate with 150 mg/L total sulfur content, and the total sulfur content was reached to 110 mg/L (26.7% adsorption) after 30 minutes. Based on the results, the nanofibers were recovered and the sulfur compounds were adsorbed again by the nanofibers under the hydrogen gas purge, indicating their good recoverable behavior.

Finally, it is fair to suggest that the prepared material in the current study for desulfurization, has an appropriate removal/adsorption efficiency as well as the synthesis procedure is simple and cost-effective. Conclusion

In this work, polyacrylonitrile/silver nanoparticles nanocomposite was prepared for the first time to adsorb sulfur contents in gas condensates. XRD peaks revealed that Ag nanoparticles were available in the composite and SEM images showed that the fibres were synthesized very well. These nanofibers are connected to each other at different points, which are not completely separate. It was found that by increasing the contact time and adsorbent dosage, desulfurization enhanced and got higher performance. Furthermore, lower concentration of sulfur compounds was beneficial in terms of adsorption/removal. Based on the results in adsorption isotherms, the adsorption of sulfur compounds was followed by Freundlich isotherm. Such behavior was ascribed to heterogeneous nature of the polyacrylonitrile/silver nanoparticles that tend to establish chemical bonds with the sulfur compounds. The energy of the adsorption calculated by Dubinin-Radushkevich equation also stated identical deduction by showing the process to be chemical-based sorption. Further, kinetic studies indicated that desulfurization by using nanocomposite follows pseudo-second-order kinetics. Furthermore, the adsorbent column studies confirmed the usability of desulfurization process with polyacrylonitrile/Ag nanocomposite with a good recovery performance. Finally, the unique adsorbent ability and simple synthesis makes it a good candidate for industrial applications.

Acknowledgment The present work was supported by the South Pars

Gas Complex. We would like to thank the SPGC for providing laboratory facilities.

References 1. Dong S., Li M., Firoozabadi A. Effect of salt and water

cuts on hydrate anti-agglomeration in a gas condensate system at high pressure. Fuel, 210: 713–20 (2017).

2. Ke T., Xin H. Deep desulfurization of model gasoline by adsorption on mesoporous CeMCM-41. Pet. Sci. Technol., 28: 573–81 (2010).

3. Samadi M., Zarenezhad B. Extension of gas adsorption models to liquid diesel fuel desulfurization: Validity and predictability. Pet. Sci. Technol., 34: 1826–32 (2016).

4. Leong M., Chian K., Mhaisalkar P., Ong W., Ratner B. Effect of electrospun poly(D,L-lactide) fibrous scaffold with nanoporous surface on attachment of porcine esophageal epithelial cells and protein adsorption. J. Biomed. Mater. Res. A, 89: 1040–8 (2009).

5. Yang RT., Hernández-Maldonado AJ., Yang FH. Desulfurization of transportation fuels with zeolites under ambient conditions. Science (80-. )., 301: 79–81 (2003).

6. Bakhtiari G., Abdouss M., Bazmi M., Royaee SJ. Optimization of sulfur adsorption over Ag-zeolite nanoadsorbent by experimental design method. Int. J. Environ. Sci. Technol., 13: 803–12 (2016).

7. Tang X., Hu T., Li J., Wang F., Qing D. Deep desulfurization of condensate gasoline by electrochemical oxidation and solvent extraction. RSC Adv., 5: 53455–61 (2015).

8. Tella AC., Owalude SO., Olatunji SJ., Oloyede SO., Ogunlaja AS., Bourne SA. Synthesis, crystal structure and desulfurization properties of zig-zag 1D coordination polymer of copper (II) containing 4-methoxybenzoic acid ligand. J. Sulfur Chem., 39: 588–606 (2018).

9. Li C-J., Li Y-J., Wang J-N., Zhao L., Cheng J. Ag+-loaded polystyrene nanofibrous membranes preparation and their adsorption properties for thiophene. Chem. Eng. J., 222: 419–25 (2013).

10. Allafchian AR., Gholamian M., Mohammadi J. Desulfurization of gasoline using acrylonitrile electrospun nanofibers and lead nanoparticles. Int. J. Environ. Sci. Technol., 14: 1489–96 (2017).

11. Khadir A., Negarestani M., Mollahosseini A. Sequestration of a non-steroidal anti-inflammatory drug from aquatic media by lignocellulosic material (Luffa cylindrica) reinforced with polypyrrole: study of parameters, kinetics, and equilibrium. J. Environ. Chem. Eng., 8: 103734 (2020).

12. Khadir A., Negarestani M., Ghiasinejad H. Low-cost sisal fibers/polypyrrole/polyaniline biosorbent for sequestration of reactive orange 5 from aqueous solutions. J. Environ. Chem. Eng., 8: 103956 (2020).

13. Shu C., Yanwei W., Hong L., Zhengzheng P., Kangde Y. Synthesis of carbonated hydroxyapatite nanofibers by mechanochemical methods. Ceram. Int., 31: 135–8 (2005).

14. Yu J., Yu H., Cheng B., Zhao X., Zhang Q. Preparation and photocatalytic activity of mesoporous anatase TiO2 nanofibers by a hydrothermal method. J. Photochem. Photobiol. A Chem., 182: 121–7 (2006).

15. Malwal D., Gopinath P. Efficient adsorption and antibacterial properties of electrospun CuO-ZnO composite nanofibers for water remediation. J. Hazard. Mater., 321: 611–21 (2017).

Vol. 31 No. 4 Autumn 2020 R. Dadashvand-nigjeh, et al. J. Sci. I. R. Iran

318

16. Huang Z-M., Zhang Y-Z., Kotaki M., Ramakrishna S. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos. Sci. Technol., 63: 2223–53 (2003).

17. Mollahosseini A, Rastegari M., Hatefi N. Electrospun Polyacrylonitrile as a New Coating for Mechanical Stir Bar Sorptive Extraction of Polycyclic Aromatic Hydrocarbons from Water Samples. Chromatographia, 83: 549–58 (2020).

18. Habiba U., Afifi AM., Salleh A., Ang BC. Chitosan/(polyvinyl alcohol)/zeolite electrospun composite nanofibrous membrane for adsorption of Cr6+, Fe3+ and Ni2+. J. Hazard. Mater., 322: 182–94 (2017).

19. Zhang S., Shi Q., Korfiatis G., Christodoulatos C., Wang H., Meng X. Chromate removal by electrospun PVA/PEI nanofibers: Adsorption, reduction, and effects of co-existing ions. Chem. Eng. J., 387: 124179 (2020).

20. Lee C-G., Javed H., Zhang D., Kim J-H., Westerhoff P., Li Q., Alvarez PJJ. Porous electrospun fibers embedding TiO2 for adsorption and photocatalytic degradation of water pollutants. Environ. Sci. Technol., 52: 4285–93 (2018).

21. Habiba U., Siddique TA., Li Lee JJ., Joo TC., Ang BC., Afifi AM. Adsorption study of methyl orange by chitosan/polyvinyl alcohol/zeolite electrospun composite nanofibrous membrane. Carbohydr. Polym., 191: 79–85 (2018).

22. Dou X., Wang Q., Li Z., Ju J., Wang S., Hao L., Sui K., Xia Y., Tan Y. Seaweed-Derived Electrospun Nanofibrous Membranes for Ultrahigh Protein Adsorption. Adv. Funct. Mater., 29: 1905610 (2019).

23. Thielke MW., Bultema LA., Brauer DD., Richter B., Fischer M., Theato P. Rapid Mercury (II) removal by electrospun sulfur copolymers. Polymers (Basel)., 8: 266 (2016).

24. Mukwevho N., Gusain R., Fosso-Kankeu E., Kumar N., Waanders F., Ray SS. Removal of naphthalene from simulated wastewater through adsorption-photodegradation by ZnO/Ag/GO nanocomposite. J. Ind. Eng. Chem., 81: 393–404 (2020).

25. Yang Y., Li H., Hou F., Hu J., Zhang X., Wang Y. Facile synthesis of ZnO/Ag nanocomposites with enhanced photocatalytic properties under visible light. Mater. Lett., 180: 97–100 (2016).

26. Li J., Yang S., Du Z., Wang R., Yuan L., Wang H., Wu Z. Quantitative analysis of ammonia adsorption in Ag/AgI-coated hollow waveguide by mid-infrared laser absorption spectroscopy. Opt. Lasers Eng., 121: 80–6 (2019).

27. Zaheer Z., AL-Asfar A., Aazam ES. Adsorption of methyl red on biogenic Ag@Fe nanocomposite adsorbent: Isotherms, kinetics and mechanisms. J. Mol. Liq., 283: 287–98 (2019).

28. Satokawa S., Kobayashi Y., Fujiki H. Adsorptive removal of dimethylsulfide and t-butylmercaptan from pipeline natural gas fuel on Ag zeolites under ambient conditions. Appl. Catal. B Environ., 56: 51–6 (2005).

29. Nair S., Tatarchuk BJ. Supported silver adsorbents for selective removal of sulfur species from hydrocarbon fuels. Fuel, 89: 3218–25 (2010).

30. Sun X., Tatarchuk BJ. Photo-assisted adsorptive desulfurization of hydrocarbon fuels over TiO2 and

Ag/TiO2. Fuel, 183: 550–6 (2016). 31. Mirjavadi ES., M.A.Tehrani R., Khadir A. Effective

adsorption of zinc on magnetic nanocomposite of Fe3O4/zeolite/cellulose nanofibers: kinetic, equilibrium, and thermodynamic study. Environ. Sci. Pollut. Res., 26: 33478–33493 (2019).

32. Mollahosseini A., Khadir A., Saeidian J. Core–shell polypyrrole/Fe3O4 nanocomposite as sorbent for magnetic dispersive solid-phase extraction of Al+3 ions from solutions: investigation of the operational parameters. J. Water Process Eng., 29: 100795 (2019).

33. Ghenaatgar A., M.A.Tehrani R., Khadir A. Photocatalytic degradation and mineralization of dexamethasone using WO3 and ZrO2 nanoparticles: Optimization of operational parameters and kinetic studies. J. Water Process Eng., 32: 100969 (2019).

34. Zheng M., Hu H., Ye Z., Huang Q., Chen X. Adsorption desulfurization performance and adsorption-diffusion study of B2O3 modified Ag-CeOx/TiO2-SiO2. J. Hazard. Mater., 362: 424–35 (2019).

35. Khadir A., Mollahosseini A., Tehrani RMA., Negarestani M. A Review on Pharmaceutical Removal from Aquatic Media by Adsorption: Understanding the Influential Parameters and Novel Adsorbents BT - Sustainable Green Chemical Processes and their Allied Applications. (Inamuddin , Asiri Abdullah ed.), Springer International Publishing, Cham, pp.207–65 (2020).

36. Nimibofa A., Ebelegi A., Donbebe W. Modelling and Interpretation of Adsorption Isotherms. Hindawi J. Chem., Volume 201: 11 pages (2017).

37. Mohammadi A., Khadir A., Tehrani RMA. Optimization of nitrogen removal from an anaerobic digester effluent by electrocoagulation process. J. Environ. Chem. Eng., 7: 103195 (2019).

38. Khadir A., Negarestani M., Motamedi M. Optimization of an electrocoagulation unit for purification of ibuprofen from drinking water: effect of conditions and linear/non-linear isotherm study. Sep. Sci. Technol., (2020).

39. Khadir A., Motamedi M., Negarestani M., Sillanpää M., Sasani M. Preparation of a nano bio-composite based on cellulosic biomass and conducting polymeric nanoparticles for ibuprofen removal: Kinetics, isotherms, and energy site distribution. Int. J. Biol. Macromol., 162: 663-677 (2020).

40. Turk Sekulic M., Boskovic N., Milanovic M., Grujic Letic N., Gligoric E., Pap S. An insight into the adsorption of three emerging pharmaceutical contaminants on multifunctional carbonous adsorbent: Mechanisms, modelling and metal coadsorption. J. Mol. Liq., 284: 372–82 (2019).

41. Tran HN., You S-J., Hosseini-Bandegharaei A., Chao H-P. Mistakes and inconsistencies regarding adsorption of contaminants from aqueous solutions: A critical review. Water Res., 120: 88–116 (2017).

42. Wu F-C., Tseng R-L., Juang R-S. Characteristics of Elovich equation used for the analysis of adsorption kinetics in dye-chitosan systems. Chem. Eng. J., 150: 366–73 (2009).

43. Li C., Chen D., Ding J., Shi Z. A novel hetero-exopolysaccharide for the adsorption of methylene blue from aqueous solutions: Isotherm, kinetic, and mechanism

Natural Gas Condensate Desulfurization via Polyacrylonitrile/Ag Nanocomposite Nanofibers …

319

studies. J. Clean. Prod., 265: 121800 (2020). 44. Park JG., Ko CH., Yi KB., Park J-H., Han S-S., Cho S-H.,

Kim J-N. Reactive adsorption of sulfur compounds in diesel on nickel supported on mesoporous silica. Appl. Catal. B Environ., 81: 244–50 (2008).

45. Samadi-Maybodi A., Teymouri M., Vahid A., Miranbeigi A. In situ incorporation of nickel nanoparticles into the mesopores of MCM-41 by manipulation of solvent–solute interaction and its activity toward adsorptive desulfurization of gas oil. J. Hazard. Mater., 192: 1667–74 (2011).

46. Montazerolghaem M., Seyedeyn-Azad F., Rahimi A. The performance of pelletized Ce-Y and Ni-Y zeolites for removal of thiophene from model gasoline solutions. Korean J. Chem. Eng., 32: 328–34 (2015).

47. Ko CH., Park JG., Park JC., Song H., Han S-S., Kim J-N. Surface status and size influences of nickel nanoparticles

on sulfur compound adsorption. Appl. Surf. Sci., 253: 5864–5867 (2007).

48. Ma X., Velu S., Kim JH., Song C. Deep desulfurization of gasoline by selective adsorption over solid adsorbents and impact of analytical methods on ppm-level sulfur quantification for fuel cell applications. Appl. Catal. B Environ., 56: 137–47 (2005).

49. Moosavi E., Dastgheib S., Karimzadeh R. Adsorption of Thiophenic Compounds from Model Diesel Fuel Using Copper and Nickel Impregnated Activated Carbons. Energies, 5: 4233–50 (2012).

50. Xiong J., Zhu W., li H., Ding W., Chao Y., Wu P., Xun S., Zhang M., Li H. Few-layered graphene-like boron nitride induced a remarkable adsorption capacity for dibenzothiophene in fuels. Green Chem., 17: 1647-1656 (2014).