j o ur nal ho me pa ge: www.elsev ier .com/ locate / jhazmat
odeling of sulfonamide antibiotic removal by TiO2/high-silicaeolite HSZ-385 composite
huji Fukahoria,c,∗, Taku Fujiwarab,c,1
New Paper Industry Program Center, Faculty of Agriculture, Ehime University, JapanResearch and Education Faculty, Natural Sciences Cluster, Agriculture Unit, Kochi University, JapanJapan Science and Technology Agency, CREST, Japan
i g h l i g h t s
TiO2/high-silica zeolite composite was applied to remove sulfamethazine (SMT).We made simple model to evaluate the mechanism of SMT degradation in the composite.The existence of synergistic reaction in the composite was suggested.Some of the adsorbed SMT were degraded in the composite without being released to water.The synergistic reaction played a significant role in total removal of SMT.
r t i c l e i n f o
rticle history:eceived 10 October 2013eceived in revised form 15 February 2014ccepted 19 February 2014vailable online 28 February 2014
eywords:hotocatalystdsorbentompositeodeling
ynergistic reaction
a b s t r a c t
TiO2/high-silica zeolite composite synthesized by a sol–gel method was applied for the removal ofsulfamethazine (SMT) antibiotic from water, and simple models including both adsorption and photocat-alytic decomposition were developed. In this study, two types of models were constructed: a synergisticmodel that included the interaction between the zeolite and TiO2 in the composite, and an individualmodel, which did not include the interaction. We obtained rate constants for adsorption, desorption andphotocatalytic decomposition experimentally, and compared them with the results calculated using thesynergistic and individual models. The individual model predicted that ca. 55% of SMT would be removedfrom the system after 6 h of treatment; however, our experiments showed that 80% of the SMT wasremoved, suggesting the existence of another reaction pathway. Therefore, a synergistic model was con-structed, in which, part of the SMT was adsorbed onto the zeolite within the composite, desorbed fromthe zeolite and migrated to the TiO2, and was then photocatalytically decomposed. Experiments werecarried out with varying amounts of the TiO2-zeolite composite, and the synergistic model was validated.We estimated that 10% of the desorbed SMT was photocatalytically decomposed without being releasedinto the water. When TiO2-zeolite composite concentrations were 0.04, 0.12 and 0.20 g/L, and the treat-
ment time was 6 h, the proportions of the total decomposition of SMT that occurred via this synergisticreaction pathway were calculated as 52.2%, 58.6% and 66.7%, respectively. In other words, over half ofthe SMT was decomposed through the synergistic reaction, which played a very significant role in theoverall removal of SMT (the remainder of the SMT was decomposed through simple photocatalysis onthe TiO2).
1 Research and Education Faculty, Natural Sciences Cluster, Agriculture Unit,ochi University, 200 Monobe Otsu, Nankoku, Kochi 783-8502, Japan.el.: +81 88 864 5163; fax: +81 88 864 5163.∗ Corresponding author at: New Paper Industry Program Center, Faculty of Agri-
Pharmaceuticals and compounds derived from personal careproducts are increasingly found in the environment and haveattracted much attention [1–4]. The concentration of pharma-
ceuticals detected in the environment is quite low (ng/L–�g/L);however, ecotoxicity of pharmaceuticals at �g/L levels has beenreported [5]. For the sustainable use of water, suitable treat-ment methods for the removal of discharged pharmaceuticals are
2 S. Fukahori, T. Fujiwara / Journal of Hazardous Materials 272 (2014) 1–9
Nomenclature
C concentration of sulfamethazine in the aqueousphase (mg/L)
C′ concentration of sulfamethazine after desorption(mg/L)
C0 initial concentration of sulfamethazine (mg/L)Cc concentration of TiO2-zeolite composite (g/L)Ce equilibrium concentration of sulfamethazine
(mol/L)CP25 concentration of P25 (g/L)kad rate constant of adsorption of sulfamethazine
(L/mg/min)kde rate constant of desorption of sulfamethazine
(1/min)kp1 pseudo-first-order rate constant of photocatalytic
decomposition (1/min)kp2 rate constant of decomposition of sulfamethazine
desorbed from zeolite within TiO2-zeolite compos-ite (1/min)
kc-SMT pseudo-first-order rate constant of photocatalyticdecomposition of sulfamethazine treated by TiO2-zeolite composite (1/min)
kc-SA pseudo-first-order rate constant of photocatalyticdecomposition of sulfanilic acid treated by TiO2-zeolite composite (1/min)
kP25-SMT pseudo-first-order rate constant of photocatalyticdecomposition of sulfamethazine treated by P25(1/min)
kP25-SA pseudo-first-order rate constant of photocatalyticdecomposition of sulfanilic acid treated by P25(1/min)
KL Langmuir constant (L/mg)mp1 amount of sulfamethazine decomposed through Rp1
per minute (mg/g)mp2 amount of sulfamethazine decomposed through Rp2
per minute (mg/g)Mw amount of sulfamethazine remaining in water (mg)Mc amount of sulfamethazine remaining in TiO2-
zeolite composite (mg)Ms amount of sulfamethazine remaining in system
(mg)q amount of adsorbate on the adsorbent (mg/g)qe amount of sulfamethazine adsorbed onto zeolite
or TiO2-zeolite composite after reaching adsorptionequilibrium (mg/g)
qm maximum adsorption capacity (mg/g)Rad adsorption of sulfamethazine onto TiO2-zeolite
compositeRde desorption of sulfamethazine from TiO2-zeolite
compositeR′
de simple transfer of sulfamethazine from TiO2-zeolitecomposite to aqueous phase after desorption
Rp1 photocatalytic decomposition of sulfamethazine onTiO2-zeolite composite
Rp2 photocatalytic decomposition of sulfamethazinedesorbed from zeolite onto TiO2 in TiO2-zeolitecomposite
S the sum of the squares of the differences betweenexperimental and calculated results for the removalof sulfamethazine using TiO2-zeolite composite
vde desorption rate of sulfamethazine from TiO2-zeolitecomposite (mg/g/min)
v′de simple transfer rate of sulfamethazine from TiO2-
zeolite composite to aqueous phase after desorption(mg/g/min)
vp1 photocatalytic decomposition rate on TiO2-zeolitecomposite (mg/L/min)
vp2 photocatalytic decomposition rate of sulfamet-hazine which has been transferred to the surface of
(mg2)V volume of the sulfamethazine solution (L)vad adsorption rate of sulfamethazine on TiO2-zeolite
composite (mg/g/min)
TiO2 particles on the composite (mg/g/min)w amount of the adsorbent (g)
required, and research on water purification methods is currentlybeing conducted in scientific and industrial circles [6,7].
Recently, the removal of pharmaceuticals using photocata-lysts and adsorbents such as zeolites and activated carbons hasbeen investigated [8–11]. We tested high-silica Y-type zeolite,a relatively hydrophobic zeolite, for adsorption of sulfonamideantibiotics, and revealed that the pH of the sulfonamide solu-tion greatly affected the removal efficiencies [12]. In addition,high-silica Y-type zeolite could quickly and selectively removesulfonamides, even if coexisting materials were present at highconcentrations [13]; however, it cannot adsorb sulfonamides afterreaching saturation. Conversely, TiO2 photocatalyst has high oxi-dizing power, and can decompose recalcitrant organic compoundsunder ultraviolet (UV) irradiation, although it typically takes afew hours to photocatalytically remove pollutants from water[9–11]. To overcome the disadvantages of each material, thesynthesis of photocatalyst/adsorbent composites and their appli-cation to removal of pharmaceuticals or chemicals in water havebeen reported [14–17]. We have also reported the synthesis ofTiO2-zeolite composites and applied them to remove sulfonamideantibiotics from secondary effluent [18]. The composites couldremove sulfonamide from secondary effluent more effectively thanTiO2 alone. However, the mechanism of the interaction between thephotocatalyst and adsorbent has not yet been thoroughly exam-ined.
In this study, we applied the TiO2-zeolite composite to theremoval of sulfamethazine (SMT), which is one of sulfonamideantibiotics, and its analog. Sulfonamide antibiotic are popular activeantimicrobial agents used in animal food production due to theirrelatively low cost [19,20]. However, some of sulfonamide antibi-otics are not completely removed by conventional wastewatertreatment systems, such as activated sludge processes, due totheir high resistance to biodegradation and detected in naturalriver or effluent of wastewater treatment plant [21]. We quanti-tatively determined the rate constants for adsorption, desorptionand photocatalytic decomposition. Based on these parameters, weconstructed simple models containing adsorption, photocatalyticdecomposition and photocatalytic decomposition after desorption,and evaluated the contributions of each reaction to the removal ofSMT. The details of our models are described in next chapter.
2. Development of TiO2-zeolite composite model
Two simple models were constructed in this study, one is theindividual model in which there is no interaction between the TiO2and zeolite, and adsorption (Rad), desorption (Rde) and photocat-alytic decomposition (Rp1) of SMT occur (Fig. 1(a)). In this work,
we use “adsorption” to mean adsorption of SMT onto zeolite in thecomposite, and are not referring to any short-term adsorption ontothe surface of TiO2. Another is the synergistic model, in which partof the SMT adsorbed on the zeolite transfers to the surface of the
S. Fukahori, T. Fujiwara / Journal of Hazardous Materials 272 (2014) 1–9 3
iO2-ze
TApsppbi
ssrfimRStmrpabrtrFsTktttt
3
3
SpASfu
3
2
Fig. 1. Proposed decomposition mechanisms of SMT on T
iO2, and is subsequently photocatalytic decomposed (Fig. 1(b)).s well as Rad, Rde and Rp1, two reactions after Rde are suggested:hotocatalytic decomposition of SMT that has transferred to theurface of the TiO2 particles within the composite (Rp2) and sim-le transfer of SMT from the TiO2-zeolite composite to the aqueoushase after desorption (R′
de). For simplicity of the model, only theehavior of SMT in water with TiO2-zeolite composite was treated
n this study.In our previous studies, kinetic parameters for adsorption of
ulfonamides onto high-silica zeolite and photocatalytic decompo-ition using TiO2 were investigated and it was clarified that theseeactions could be expressed as a Langmuir model and pseudo-rst-order model, respectively [10,12]. To construct a synergisticodel, we first determined experimental rate constants for Rad,
de and Rp1 when SMT was treated by TiO2-zeolite composite. TheMT solution was mixed with the composite under dark condi-ions, and adsorption and desorption rate constants (kad, kde) were
easured. On the other hand, it is impossible to determine theate constant for Rp1 directly because adsorption (desorption) andhotocatalytic decomposition occur simultaneously. Therefore, annalog compound of SMT (sulfanilic acid: SA), which is not adsorbedy HSZ-385 zeolite was treated by TiO2-zeolite composite and theate constant (kc-SA) was determined. Similarly, SMT and SA solu-ions were treated by TiO2 powder (P25) and the pseudo-first-orderate constants for each reaction (kP25-SMT, kP25-SA) were determined.rom the value of kc-SA and the ratio of kP25-SMT/kP25-SA, the rate con-tant for Rp1 (kc-SMT) was estimated (kP25-SMT/kP25-SA = kc-SMT/kc-SA).hen, the individual model was constructed using kad, kde andc-SMT and the calculated results were compared to the experimen-al results. We assumed that the difference between the results ofhe individual model and experiments was because of the synergis-ic reaction (Rp2), and the rate constant Rp2 was set so as to minimizehe difference.
. Experimental
.1. Chemicals
High-silica Y-type zeolite (HSZ-385, mean particle size 4 �m,iO2/Al2O3 = 100) was purchased from TOSOH Ltd., Japan. TiO2owder (P25, surface area 50 m2/g) was kindly provided by NipponEROSIL Ltd. SMT (purity > 99%) was purchased from Aldrich andA (purity > 99%) and titanium tetraisopropoxide were purchasedrom Kanto Chemicals, respectively. The details of other chemicalssed in this study are given in the Appendice.
.2. Synthesis of TiO2-zeolite composite
TiO2-zeolite composite was synthesized as follows: 22.5 mL of-propanol was added to 2.5 mL of titanium tetraisopropoxide.
olite composite; individual (a) and synergistic model (b).
Subsequently, 2.5 mL of an aqueous high-silica zeolite suspensioncontaining 670 mg of high-silica zeolite was added to the titaniumtetraisopropoxide solution. After stirring for 1 h, the precipitateswere recovered by filtration, washed with distilled water, driedat 105 ◦C, and then calcined at 700 ◦C for 3 h. The ratio of TiO2 tozeolite in the composite was 1:1. TiO2 powder was prepared ina similar manner. Characterization of the TiO2-zeolite compositewas performed using scanning electron microscopy (SEM: JSM-5510LV, JEOL, Ltd.), energy dispersive X-ray spectroscopy (EDX:Genesis, EDAX) and X-ray diffraction (XRD, Ultima IV, Rigaku Co.Ltd.). The SEM and EDX images and XRD patterns are shown inour previous study [18]. The surface areas of high-silica zeolite andTiO2-zeolite composite were measured by BET adsorption method(BELSORP-mini, BEL Japan, Inc.).
3.3. Quantitative analyses
The concentrations of SMT and SA were measured by ultraperformance liquid chromatography (UPLC: ACQUITY, Waters).The UPLC analysis was performed using a BEH C18 column(2.1 × 150 mm; Waters) with a linear gradient from 10% acetoni-trile in 0.05% formic acid (isocratic for 0.5 min) to 90% (0.5–7 min)at a constant flow rate of 0.3 mL/min. A photodiode array detectorwas placed after the analytical column and the wavelength was setto 254 nm. The coefficient of variance for the LC analysis was <5%based on three measurements.
3.4. Adsorption experiment
The adsorption rate was evaluated by changing the amount ofTiO2-zeolite composite (dosages 2, 4, 6, 8 and 10 mg; concentra-tions 0.04, 0.08, 0.12, 0.16 and 0.20 g/L) added to SMT solution(10 mg/L and 50 mL) under dark condition. The pH was adjustedto 7 using sodium hydroxide. After addition of TiO2-zeolite com-posite powder into the reaction vessel, aliquots (1 mL) were takenat designated times, filtered through a membrane filter (DISMIC;pore size, 0.20 �m; ADVANTEC, Ltd.) and subjected to UPLC.
After stirring for 24 h, SMT concentrations were measured andan adsorption isotherm was constructed. Each adsorption experi-ment was repeated three times, and the error bars represent thestandard deviation.
3.5. Photocatalytic decomposition of SMT and SMT analog (SA)
To evaluate the photocatalytic activity of TiO2-zeolite compos-
ite, an SMT analog which was not adsorbed by the compositewas used as the target compound. SA solution (10 mg/L, 50 mL)and TiO2-zeolite composite (dosages 2, 4, 6, or 10 mg, concentra-tions 0.04, 0.08, 0.12, or 0.20 g/L) were placed in a glass vessel.
4 of Hazardous Materials 272 (2014) 1–9
pw3atfidw
3T
vtrs
bitrr1scis
(a
M
M
M
V
4
4
tcaltttbvaa
q
woteLi1p
0
5
10
0 10 20 30 40 50 60
Time (min)
SMT
conc
entra
tion
(mg/
L)
Fig. 2. Adsorption of SMT on TiO2-zeolite composite as a function of contact time.Composite concentrations: 0.04 (diamonds), 0.08 (squares), 0.12 (triangles), 0.16(crosses) and 0.20 g/L (circles).
0.00 0
0.00 5
0.01 0
0.01 5
0.020
0.025
0.00 0.50 1.00 1.501/C (L /mg)
1/q e
(g/m
g)
S. Fukahori, T. Fujiwara / Journal
H and temperature were adjusted to 7.0 and 25 ◦C. The solutionas stirred and irradiated with a UV lamp (maximum wavelength
65 nm; UV intensity 1.0 mW/cm2). After UV irradiation, a filteredliquot was subjected to UPLC. Similarly, 10 mg/L of SMT or SA werereated by TiO2 powder and the pseudo-first-order rate constantor photocatalytic decomposition of SMT by TiO2-zeolite compos-te was estimated. As a control, the rate constants of photocatalyticecomposition of SMT and SA by P25 (a commercial TiO2 powder)ere obtained.
.6. Photocatalytic and adsorptive removal of SMT usingiO2-zeolite composite
The SMT solution (10 mg/L and 50 mL) was poured into a glassessel. TiO2-zeolite composite (dosages 2, 6, or 10 mg, concen-rations 0.04, 0.12, or 0.20 g/L, respectively) was placed into theeaction vessel, and the pH was adjusted to 7.0. The solution wastirred at 25 ◦C and irradiated with a UV lamp set to 1.0 mW/cm2.
After UV irradiation, a designated aliquot that had been mem-rane filtered was subjected to UPLC, and the SMT concentration
n the aqueous phase (C) was quantified. In addition, we measuredhe amount of SMT contained in the composite. We had previouslyeported that the adsorption of SMT onto high-silica zeolite waseversible, and that no adsorption of SMT occurred at a pH above0 [12]. The pH of a designated aliquot of SMT-TiO2-zeolite suspen-ion was adjusted to greater than 10 and the SMT adsorbed onto theomposite desorbed. After desorption of SMT, TiO2-zeolite compos-te was removed by filtration, and the concentration of SMT in theupernatant (C′) was measured using UPLC.
The amounts of SMT in the water Mw (mg), the whole system Ms
mg) and on the TiO2-zeolite composite Mc (mg) were calculateds follows:
w = C · V (1)
s = C ′ · V (2)
c = Ms − Mw (3)
(L) is the solution volume.
. Results and discussion
.1. Adsorption of SMT on TiO2-zeolite composite
Fig. 2 shows the time course of concentration of SMT solutionsreated by different amounts of TiO2-zeolite composite under darkonditions. We had confirmed that our synthesized TiO2 could notdsorb SMT (Fig. A2) and adsorbed SMT was released under alka-ine condition and all of adsorbed SMT could be recovered [18];herefore, the decrease in SMT concentrations must be attributedo adsorption on the composite. Rapid adsorption occurred, withhe solutions reaching equilibrium within 30 min, which is quite ait faster than that observed for activated carbon [7]. In our pre-ious study, we found that adsorption behavior of sulfonamidentibiotics on high-silica zeolite HSZ-385 could be expressed by
Langmuir model:
e = KLqmCe
1 + KLCe(4)
here qe (mg/g), qm (mg/g), KL (L/mg) and Ce (mg/L) are the amountf SMT adsorbed onto the TiO2-zeolite composite at equilibrium,he maximum adsorption capacity, the Langmuir constant and thequilibrium concentration of the SMT, respectively. Therefore, a
angmuir adsorption isotherm was constructed using the exper-mental data: after 24 h stirring, SMT concentrations were 5.9, 2.7,.7, 1.1 and 0.8 mg/L when the concentrations of TiO2-zeolite com-osite were 0.04, 0.08, 0.12, 0.16 and 0.20 g/L, respectively (Fig. 3).
e
Fig. 3. Langmuir isotherm of SMT on TiO2-zeolite composite.
The values of qm, KL and the coefficient of determination (R2) for theLangmuir plot were 141 mg/g, 0.59 L/mg and 0.994, respectively.The qm value for the TiO2-zeolite composite was about half thatreported in our previous study for high silica zeolite HSZ-385 [12].The BET surface areas of high-silica zeolite and TiO2-zeolite com-posite were 424, 222 m2/g, respectively. The zeolite content of thecomposite was ca. 50%; this measured qm value is plausible and theadsorption capacity of zeolite does not seem to decrease when thezeolite is incorporated into a composite with TiO2. We have alreadyreported the value of KL for the adsorption of SMT on high-silicazeolite in our previous study (0.62 L/mg) [12]. It is similar to thevalue of KL for the adsorption of SMT on the composite (0.59 L/mg),indicating adsorption rate of SMT on zeolite is close to that on thecomposite.
In a Langmuir model, the adsorption rate vad (mg/g/min) anddesorption rate vde (mg/g/min) are expressed as follows:
vad = kadC(qm − q) (5)
vde = kdeq (6)
where kad (L/mg/min) and kde (1/min) are rate constants of adsorp-tion and desorption, and C (mg/L) and q (mg/g) are adsorbateconcentration and the amount of adsorbate on the adsorbent
of Hazardous Materials 272 (2014) 1–9 5
(f
r
−
ViL
K
ocwr
4
sSodgafi
wastsot(lsSkPaacti(
k
k
k
ooe
k
0
5
10
0 30 60Time (min )
SA co
ncen
tratio
n (m
g/L)
(b)
0
5
10
0 30 60Time (min )
SAco
ncen
tratio
n (m
g/L)
(c)
0
5
10
0 30 60Time (min)
SMT
conc
entra
tion
(mg/
L)
(a)
Fig. 4. Time courses (under UV irradiation) of (a) SMT and (b) SA concentrationsduring treatment with P25; (c) SA concentration during treatment with TiO2-zeolitecomposite. P25 or composite concentrations: 0.04 (diamonds), 0.08 (squares), 0.12(triangles) and 0.20 g/L (crosses).
S. Fukahori, T. Fujiwara / Journal
TiO2-zeolite composite). The variation in q can be expressed as aunction of vad and vde:
dq
dt= vad − vde = kadC(qm − q) − kdeq (7)
During the adsorption process, the amount of adsorbateemoved from solution is equal to that adsorbed on the adsorbent:
V · dC
dt= w · dq
dt(8)
dC
dt= −w
V· dq
dt= −w
V[kadC(qm − q) − kdeq] (9)
(L) and w (g) are the volume of solution and mass of compos-te adsorbent. The values of kad and kde can be expressed usingangmuir constant KL:
L = kad
kde(10)
We selected the values of kad and kde to minimize the sumf the squares of the differences between experimental and cal-ulated results. The values of kad and kde obtained in this studyere 3.00 × 10−2 and 5.20 × 10−2/min, respectively, and calculated
esults are shown as solid lines in Fig. 2.
.2. Photocatalytic decomposition of SMT and SMT analog
In our previous study, sulfonamides were adsorbed onto high-ilica zeolite through hydrophobic interaction, therefore, we usedMT analog which had similar structure but was not adsorbednto the composite to determine rate constant for photocatalyticecomposition. SA was selected as SMT analog because SA has sulforoup and has negative charge under all pH range. We checkeddsorption behavior of SA onto TiO2-zeolite composite and con-rmed SA was not adsorbed (Fig. A3).
The time courses of the concentrations of SA solutions treatedith irradiation in the presence of TiO2-zeolite composite, and SMT
nd SA solutions irradiated in the presence of P25 and the relation-hip between ln(C/C0) and UV irradiation time (where C0 and C arehe concentrations of SMT or SA at times 0 and t, respectively) arehown in Fig. 4 and 4A. We have confirmed that no adsorption of SAr SMT on P25 or synthesized TiO2 occurred (Fig. A2). A linear rela-ionship between ln(C/C0) and UV irradiation time was confirmedFig. A4). Hence, a pseudo-first-order reaction model expressed asn(C/C0) = −kt (where k (1/min) is the pseudo-first-order rate con-tant) was suitable to describe the photodegradation of SMT andA. In addition, the relationships between the rate constants kc-SA,P25-SMT and kP25-SA and the concentrations of either composite or25, are shown in Fig. 5. The expected value of kc-SMT is also given, as
function of composite concentration. The values of kc-SA, kP25-SMTnd kP25-SA increased logarithmically rather than linearly with bothoncentrations of composite and P25. This is because of light scat-ering caused by the TiO2 particles; similar results were obtainedn our previous study [10]. The values of kc-SA, kP25-SMT and kP25-SA1/min) can be expressed as:
Cc and CP25 (g/L) are concentrations of TiO2-zeolite compositer P25. The calculated ratio of kP25-SMT to kP25-SA (0.04 < CP25 < 0.20)btained using Eqs. (12) and (13) was 3.17 ± 0.18; thus, we
6 S. Fukahori, T. Fujiwara / Journal of Hazardous Materials 272 (2014) 1–9
Fig. 5. Relationships between (a) P25 concentration and reaction rate constantskP25-SMT (diamonds) and kP25-SA (crosses); (b) TiO2-zeolite composite concentrationa
4T
daawdpSoSpatcw
SnoaTzs
0
0.1
0.2
0.3
0.4
0.5
0 60 12 0 18 0 240 30 0 360
Mw,Mc,Ms
(mg)
Time (min )
(b)
0
0.1
0.2
0.3
0.4
0.5
0 60 120 18 0 24 0 30 0 36 0
Mw,Mc,Ms
(mg)
Time (min )
(a)
0
0.1
0.2
0.3
0.4
0.5
0 60 120 180 24 0 30 0 36 0
Mw,Mc,Ms
(mg)
Time (min )
(c)
Fig. 6. Time courses of amounts of SMT in water (Mw , blue squares), TiO2-zeolitecomposite (Mc , red diamonds) and total system (Ms , green triangles) obtained exper-imentally with composite concentrations of (a) 0.04, (b) 0.12 and (c) 0.20 g/L. Linesshow the calculated results obtained using the individual model with the same com-posite concentration. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of the article.)
nd kc-SA (triangles). Dashed line shows the expected value of kc-SMT.
.3. Adsorption and photocatalytic decomposition of SMT byiO2-zeolite composite
Fig. 6 shows the removal of SMT by TiO2-zeolite composite atifferent composite concentrations. In all conditions, the amount ofdsorbed SMT drastically increased in the initial stage of treatment,nd then decreased quickly. Without UV irradiation, adsorbed SMTas retained [18]; therefore, the decrease is due to photocatalyticecomposition. When P25 was used, it still took a few hours tohotocatalytically remove SMT from water. Such quick removal ofMT when the composite was used is the greatest advantage ofur composite for water purification. In addition, the amounts ofMT adsorbed onto the composite gradually decreased as a result ofhotocatalysis and 52%, 59% and 73% of the SMT was decomposedfter 3-h treatment when the TiO2-zeolite composite concentra-ions were 0.04, 0.12 and 0.20 g/L, respectively. Therefore, theomposite was regenerated, and could adsorb SMT continuouslyithout reaching adsorption equilibrium.
To model the adsorption and photocatalytic decomposition ofMT, we first suggested an ‘individual model’, in which there waso interaction between TiO2 and zeolite, and the reactions thatccurred were: adsorption (Rad), desorption (Rde) and photocat-lytic decomposition (Rp1) of SMT (Fig. 1(a)). The concentrations of
iO2-zeolite composite were set as 0.04, 0.12 and 0.20 g/L (TiO2-eolite composite dosages of 2, 6 and 10 mg per 50 mL of SMTolution, respectively). The rates of Rad (vad) and Rde (vde) are
of Hazardous Materials 272 (2014) 1–9 7
av
v
m
MhtesitahiTa
v
v
wtb
toTtwTddcuo
4S
adpb
m
0
0.1
0.2
0.3
0.4
0.5
0 60 12 0 18 0 24 0 30 0 36 0
Mw,Mc,Ms
(mg)
Time (min )
(b)
0
0.1
0.2
0.3
0.4
0.5
0 60 120 18 0 240 30 0 36 0
Mw,Mc,Ms
(mg)
Time (min)
(a)
0
0.1
0.2
0.3
0.4
0.5
0 60 120 180 240 300 36 0
Mw,McMs
(mg)
Time (min )
(c)
Fig. 7. Time courses of amounts of SMT in water (Mw , blue squares), TiO2-zeolitecomposite (Mc , red diamonds) and total system (Ms , green triangles) obtained exper-imentally with composite concentrations of (a) 0.04, (b) 0.12 and (c) 0.20 g/L. Linesshow the calculated results obtained using the synergistic model with the samecomposite concentration. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of the article.)
S. Fukahori, T. Fujiwara / Journal
lready defined in Eqs. (5) and (6). The rate of Rp1 is defined asp1 and expressed as follows:
p1 = kc-SMTC (15)
From Eqs. (7) and (9), the variation of C and q in the individualodel can be expressed as:
dq
dt= vad − vde = kadC(qm − q) − kdeq (16)
dC
dt= −w
V[vad − vde] − vp1 = −w
V[kadC(qm − q) − kdeq] − kc-SMTC
(17)
The calculated results are shown in Fig. 6. The overall trends inw, Mc and Ms are similar to those of the experimental results;
owever, the values of Mc and Ms were different. After 6-hreatment, Ms obtained by calculation was much larger than thexperimental Ms, indicating that there must be another decompo-ition mechanism. Therefore, we suggested a ‘synergistic model’,n which a portion of the SMT adsorbed on the composite transferso the surface of the TiO2. As well as Rad, Rde and Rp1, two reactionsfter Rde were suggested: photocatalytic decomposition of SMT thatas transferred onto the surface of the TiO2 particles in the compos-
te (Rp2) and simple transfer to the aqueous phase (R′de) (Fig. 1(b)).
he rates of Rp2 and R′de are defined as vp2 and v′
de, and expresseds follows:
p2 = kp2q (18)
′de = vde − vp2 = (kde − kp2)q (19)
here kp2 (1/min) is the rate constant for reactions Rp2. The varia-ion in C and q during treatment using TiO2-zeolite composite cane expressed as:
dC
dt= −w
V[vad − v′
de] − vp1
= −w
V[kadC(qm − q) − (kde − kp2)q] − kc-SMTC
(20)
dq
dt= vad − vde
= kadC(qm − q) − kdeq
(21)
The values of kad, kde and kc-SMT are calculated in previous sec-ions. We simulated C and q in the presence of different amountsf TiO2-zeolite composite (w = 2, 6, 10 mg, Cc = 0.04, 0.12, 0.20 g/L).he values of kc-SMT obtained using Eq. (14) when the concen-rations of TiO2-zeolite composite were 0.04, 0.12, and 0.20 g/Lere 0.57 × 10−2, 2.23 × 10−2 and 3.00 × 10−2/min, respectively.
he value of kp2 which minimized the sum of the squares of theifferences between experimental and calculated results (S) wasetermined (kp2 = 0.48 × 10−2, see Fig. A5). The experimental andalculated results are shown in Fig. 7(a)–(c). By comparing the val-es of kp2 and kde (5.20 × 10−2/min), we can determine that ca. 10%f desorbed SMT was photocatalytically decomposed through Rp2.
.4. Contribution of synergistic effect to overall decomposition ofMT
To quantitatively evaluate the synergistic effect between TiO2nd zeolite in the composite, we compared the amounts of SMTecomposed through Rp1 (mp1 (mg/min)) and Rp2 (mp2 (mg/min))
er minute using the synergistic model. mp1 and mp2 are definedased on vp1 and vp2:
p1 = vp1V = kc-SMTCV (22)
8 S. Fukahori, T. Fujiwara / Journal of Haz
0
0.1
0.2
0.3
0.4
0.5C
umul
ativ
e am
ount
sof
deco
mpo
sed
SMT
(mg)
Time (min )3600 120 240
(a)
0
0.1
0.2
0.3
0.4
0.5
Cum
ulat
ive
amou
nts o
fde
com
pose
d SM
T (m
g)
Time (min )3600 120 240
(b)
0
0.1
0.2
0.3
0.4
0.5
Cum
ulat
ive
amou
nts o
fde
com
pose
d SM
T (m
g)
Time (min )3600 120 240
(c)
Fac
m
pcIatca
ig. 8. Cumulative amounts of SMT decomposed through Rp1 (blue) and Rp2 (red),s calculated using the synergistic model. (For interpretation of the references toolor in this figure legend, the reader is referred to the web version of the article.)
p2 = vp2w = kp2qw (23)
From Eqs. (22) and (23), the cumulative amounts of SMT decom-osed through Rp1 and Rp2 when the concentrations of TiO2-zeoliteomposite were 0.04, 0.12 and 0.20 g/L were calculated (Fig. 8).n the initial stage (t < 5 min), where the concentrations of SMT in
queous media were relatively high, SMT was mainly decomposedhrough Rp1. Then, the contribution of Rp2 became higher as theoncentrations of SMT in the aqueous media were lowered throughdsorption onto the zeolite in the composite. The proportions of
ardous Materials 272 (2014) 1–9
the amounts of SMT decomposed through Rp2 to the total of SMTdecomposed after 6-h treatment when TiO2-zeolite compositeconcentrations were 0.04, 0.12 and 0.20 g/L were 52.2%, 58%.6and 66.7%, respectively, indicating adsorption-photocatalyticdecomposition played a substantial role in total removal of SMT.
In previous studies [15–18], though advantages ofphotocatalyst-adsorbent composites were reported, the mecha-nism of the interaction between the photocatalyst and adsorbentwas not thoroughly discussed. Therefore, in this work we havemeasured and deduced the reaction rate of each reaction, andquantitatively evaluated the contribution of the synergisticreaction to the total decomposition of SMT. For the modelingof photocatalytic reaction, photon absorption, irradiance andreactor geometry also affect the rate constant therefore Grcicet al. constructed promising model with considering them [22].In this study, empirical equations were used for explaining theeffect of composite concentration and photon adsorption was notincluded in our proposed model. In our future plan, the modelthat includes the factor of UV intensity or photon absorption willbe developed. Photocatalyst-adsorbent composites can be widelyapplied to other pollutants by choosing the appropriate adsorbent,and our analytical approach should be useful for understandingother composite-pollutant systems.
5. Conclusion
A model of removal of SMT from solutions using TiO2-zeolitecomposite was constructed based on measured rate constants ofadsorption and desorption of SMT onto zeolite and photocatalyticdecomposition of SMT on TiO2. The experimental results differedfrom those calculated using our “individual model”, indicating thatthere must be another decomposition mechanism in addition tosimple photocatalytic decomposition. Thus, we proposed a syner-gistic model, in which a portion of the adsorbed SMT transfers tothe surface of the TiO2 in the composite and is subsequently photo-catalytically decomposed. The good agreement between the resultscalculated using the synergistic model and our experimental dataimplies that there is a synergistic effect between the TiO2 and thezeolite in our composite.
Supporting information
Details of the chemicals used in this study, adsorption of SMT orSA on synthesized TiO2 or P25, adsorption of SA onto high-silicazeolite and TiO2-zeolite composite, plots of ln(C/C0) vs reactiontime for photocatalytic decomposition of SMT and SA, and the curveused to determine the value of kp2 are shown in the Appendice.
Acknowledgment
This research was financially supported by the Core Research forEvolutional Science and Technology (CREST) program of the JapanScience and Technology Agency.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.jhazmat.2014.02.028.
References
[1] C.A. Toles, W.E. Marshall, M.M. Johns, Granular activated carbons from nutshellsfor the uptake of metals and organic compounds, Carbon 35 (1997) 1407–1414.
[2] T. Heberer, Occurrence, fate, and removal of pharmaceutical residues in theaquatic environment: a review of recent research data, Toxicol. Lett. 131 (2002)5–17.
[3] M.D. Hernando, M. Mezcua, A.R. Fernández-Alba, D. Barceló, Environmental riskassessment of pharmaceutical residues in wastewater effluents, surface watersand sediments, Talanta 69 (2006) 334–342.
[4] M. Kuster, M.J. López de Alda, M.D. Hernando, M. Petrovic, J. Martín-Alonso,D. Barceló, Analysis and occurrence of pharmaceuticals, estrogens, progesto-gens and polar pesticides in sewage treatment plant effluents, river water anddrinking water in the Llobregat river basin (Barcelona, Spain), J. Hydrol. 358(2008) 112–123.
[5] K. Fent, A.A. Weston, D. Caminada, Ecotoxicology of human pharmaceuticals,Aquat. Toxicol. 76 (2006) 122–159.
[6] T.A. Ternes, J. Stüber, N. Herrmann, D. McDowell, A. Ried, M. Kampmann, B.Teiser, Ozonation: a tool for removal of pharmaceuticals, contrast media andmusk fragrances from wastewater? Water Res. 37 (2003) 1976–1982.
[7] I. Kim, N. Yamashita, H. Tanaka, Performance of UV and UV/H2O2 processesfor the removal of pharmaceuticals detected in secondary effluent of a sewagetreatment plant in Japan, J. Hazard. Mater. 166 (2009) 1134–1140.
[8] A.S. Mestre, J. Pires, J.M.F. Nogueira, A.P. Carvalho, Activated carbons for theadsorption of ibuprofen, Carbon 45 (2007) 1979–1988.
[9] L. Prieto-Rodriguez, S. Miralles-Cuevas, I. Oller, A. Agüera, G. Li Puma, S. Malato,Treatment of emerging contaminants in wastewater treatment plants (WWTP)effluents by solar photocatalysis using low TiO2 concentrations, J. Hazard.Mater. 211–212 (2012) 131–137.
10] S. Fukahori, T. Fujiwara, R. Ito, N. Funamizu, Photocatalytic decomposition ofcrotamiton over aqueous TiO2 suspensions: determination of intermediatesand the reaction pathway, Chemosphere 89 (3) (2012) 213–220.
11] H. Cao, X. Lin, H. Zhan, H. Zhang, J. Lin, Photocatalytic degradation kinetics and
mechanism of phenobarbital in TiO2 aqueous solution, Chemosphere 90 (2013)1514–1519.
12] S. Fukahori, T. Fujiwara, R. Ito, N. Funamizu, pH-Dependent adsorption of sulfadrugs on high silica zeolite: modeling and kinetic study, Desalination 275(2011) 237–242.
[
ardous Materials 272 (2014) 1–9 9
13] S. Fukahori, T. Fujiwara, N. Funamizu, K. Matsukawa, R. Ito, Adsorptive removalof sulfonamide antibiotics in the livestock urine using high-silica zeolite, WaterSci. Technol. 67 (2) (2013) 319–325.
14] C. Zhao, H. Deng, Y. Li, Z. Liu, Photodegradation of oxytetracycline in aqueousby 5A and 13X loaded with TiO2 under UV irradiation, J. Hazard. Mater. 176(2010) 884–892.
15] P.-S. Yap, T.-T. Lim, Solar regeneration of powdered activated carbonimpregnated with visible-light responsive photocatalyst: factors affecting per-formances and predictive model, Water Res. 46 (2012) 3054–3064.
16] P.-S. Yap, Y.-L. Cheah, M. Srinivasan, T.-T. Lim, Bimodal N-doped P25-TiO2/ACcomposite: preparation, characterization, physical stability, and synergisticadsorptive-solar photocatalytic removal of sulfamethazine, Appl. Catal. A: Gen.427–428 (2012) 125–136.
17] J. Matos, J. Laine, J.-M. Herrmann, Synergy effect in the photocatalytic degra-dation of phenol on a suspended mixture of titania and activated carbon, Appl.Catal. B: Environ. 18 (1998) 281–291.
18] M. Ito, S. Fukahori, T. Fujiwara, Adsorptive removal and photocatalyticdecomposition of sulfamethazine in secondary effluent using TiO2–zeolitecomposites, Environ. Sci. Pollut. Res. 21 (2014) 834–842.
19] R. Hirsch, T. Ternes, K. Haberer, K.-L. Kratz, Occurrence of antibiotics in theaquatic environment, Sci. Total Environ. 225 (1999) 109–118.
20] J.V. Holm, K. Ruegge, P.L. Bjerg, T.H. Christensen, Occurrence and distributionof pharmaceutical organic compounds in the groundwater downgradi-ent of a landfill (Grindsted, Denmark), Environ. Sci. Technol. 29 (1995)1415–1420.
21] T.A. Ternes, Occurrence of drugs in German sewage treatment plants and rivers,
Water Res. 32 (1998) 3245–3260.
22] I. Grcic, G. Li Puma, Photocatalytic degradation of water contaminants in multi-ple photoreactors and evaluation of reaction kinetic constants independent ofphoton absorption, irradiance, reactor geometry, and hydrodynamics, Environ.Sci. Technol. (2013), http://dx.doi.org/10.1021/es403472e.
