Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15

Contents lists available at ScienceDirect

Journal of Photochemistry and Photobiology B: Biology

journal homepage: www.elsevier .com/locate / jphotobiol

Factors affecting UV/H2O2 inactivation of Bacillus atrophaeus spores indrinking water

http://dx.doi.org/10.1016/j.jphotobiol.2014.03.0221011-1344/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel./fax: +86 21 65986313.E-mail address: Angelina-Zhou@163.com (L. Zhou).

Yongji Zhang a, Yiqing Zhang a, Lingling Zhou b,⇑, Chaoqun Tan a

a Key Laboratory of Yangtze River Water Environment, Ministry of Education, Tongji University, Shanghai 200092, PR Chinab State Key Laboratory of Pollution Control and Resources Reuse, College of Environmental Science & Engineering, Tongji University, Shanghai 200092, PR China

a r t i c l e i n f o a b s t r a c t

Article history:Received 28 August 2013Received in revised form 26 February 2014Accepted 26 March 2014Available online 13 April 2014

Keywords:UVH2O2

Bacillus atrophaeus sporesInactivation

This study aims at estimating the performance of the Bacillus atrophaeus spores inactivation by the UVtreatment with addition of H2O2. The effect of factors affecting the inactivation was investigated, includ-ing initial H2O2 dose, UV irradiance, initial cell density, initial solution pH and various inorganic anions.Under the experimental conditions, the B. atrophaeus spores inactivation followed both the modified HomModel and the Chick’s Model. The results revealed that the H2O2 played dual roles in the reactions, whilethe optimum reduction of 5.88 lg was received at 0.5 mM H2O2 for 10 min. The inactivation effect wasaffected by the UV irradiance, while better inactivation effect was achieved at higher irradiance. Anincrease in the initial cell density slowed down the inactivation process. A slight acid condition at pH5 was considered as the optimal pH value. The inactivation effect within 10 min followed the order ofpH 5 > pH 7 > pH 9 > pH 3 > pH 11. The effects of three added inorganic anions were investigated andcompared, including sulfate (SO4

2�), nitrate (NO3�) and carbonate (CO3

2�). The sequence of inactivationeffect within 10 min followed the order of control group > SO4

2� > NO3� > CO3

2�.� 2014 Elsevier B.V. All rights reserved.

1. Introduction

The existence of microorganisms has threatened the safety ofdrinking water. A wealth of waterborne diseases caused by micro-organisms has attracted broad attention. Among these microorgan-isms, Bacillus atrophaeus, once known as Bacillus subtilis [1], is akind of non-pathogenic, Gram-positive bacteria, and extremelyresistant to traditional disinfection methods [2,3]. It can surviveunder extreme environment with the help of a tough, protectiveendospore.

Advanced oxidation processes (AOPs), especially UV-basedtechnologies, have been wildly investigated, such as UV/TiO2 [4],UV/O3 [5], and UV/persulfate [6]. Among various AOP methods,the application of UV/H2O2 was proposed as an extremely promis-ing technology in organic pollutants degradation [7,8]. However, itis still uncertain that whether this method is effective in microor-ganism inactivation. Mamane et al. [9] found that hardly did UV/H2O2 exhibit any effect in B. subtilis spores inactivation, for theprotection of spore coat layers. Gardner and Shama [10], on thecontrary, confirmed that the inactivation of B. subtilis sporestreated with UV/H2O2 process was 5.3 times higher than with UV

treatment alone. The difference may be concerned with differentkinds of UV sources, diverse UV or H2O2 doses and dissimilarmicroorganism initial concentrations. As a result, confirming thereal performance of the B. atrophaeus spores inactivation by theUV treatment with addition of H2O2 proved to be significant.

Furthermore, some experimental parameters could make signif-icant contributions to affect B. atrophaeus spores inactivation. Lar-son and Mariñas [11] indicated that the fastest inactivation ratewith ozone or monochloramine were observed at pH 10 and 6,respectively. Wang et al. [12] demonstrated that vacuum-UV treat-ment alone at 254 nm or 222 nm was much better than at 172 nm.Zhang et al. [13] investigated the effect of O3 dose with micro-bub-ble ozonation system. Some researches have also reported theeffect of parameters on various organic pollutants degradation byUV/H2O2, such as ametryn [7], diethyl phthalate [14], and N-nitros-amines [15]. However, there has not been investigation into theeffect of parameters affecting inactivation of microorganisms suchas B. atrophaeus spores by UV/H2O2.

This study aims at determining the kinetics of B. atrophaeusspores inactivated by UV/H2O2 under various factors. In addition,the effects of initial H2O2 dose, UV irradiance, initial cell densityand initial solution pH on B. atrophaeus spores inactivation wereinvestigated. In addition, since inorganic anions may react with�OH, which played a dominant role in UV/H2O2 process, it is essen-tial to consider the existence of the anions [16]. The experiments

10 Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15

were carried out in compliance with the rules of changing singlevariable at each time and keeping other parameters constant.

2. Materials and methods

2.1. Experimental apparatus

The facility used in the experiments was a Collimated BeamApparatus containing a low-pressure mercury lamp (Philips TUV36 T5 SP 40W, Netherland), as illustrated in Fig. 1. The monochro-matic UV radiation emitting by this lamp was directed to thesurface of the test samples. The average irradiance at 254 nmmeasured by a UV-M radiometer (Beijing Normal UniversityExperiment Company, China) was 113.0, 56.5, and 28.3 lW/cm2,respectively, based on the Bolton and Linden protocol [17].

2.2. B. atrophaeus spores culture and enumeration

Pure cultured B. atrophaeus spores (ATCC 9372), provided byChina General Microbiological Culture Collection Center, wererehydrated aseptically with Nutrient Broth (Peptone 10 g/L, NaCl5 g/L, Beef extract 3 g/L). The bacterial suspension was incubatedfor 24 h and sporulation medium (Yeast extract 0.7 g/L, Glucose1 g/L, Peptone 1 g/L, MgSO4�7H2O 0.2 g/L, (NH4)2SO4 0.2 g/L),respectively, for 48 h at 37 �C in a shaker. After that, the sporeswere centrifuged (6000 rpm, 10 min) and redissolved in 10% NaClsolution. Then the bacterial suspension was placed in a water bathto kill the remaining vegetative cells (80 �C, 10 min). The ultimatecell density was approximately 106–108 colony forming units permilliliter (CFU/mL).

The viable spore suspension was serially diluted depending onthe order of magnitudes. Then 0.1 mL of the suspension wasinjected onto nutrient agar medium. Each dilution was plated intriplicate, and incubated with nutrient agar medium (37 �C, 24 h)to enumerate the B. atrophaeus spores [18].

2.3. Materials

The solutions were adjusted to the desired pH by addition ofHCl or NaOH. Since the spores were redissolved in 10% NaCl solu-tion and H2O contained large amount of OH�, the added inorganicanions Cl� and OH� had no inhibition influence on inactivationeffect. Solutions with tested inorganic anions, such as SO4

2�, NO3�

and CO32� were obtained via adding Na2SO4, NaNO3 and Na2CO3

Fig. 1. Schematic diagram of Collimated Beam Apparatus.

into certain amount of deionized water, respectively. All thereagents, offered by Sinopharm Chemical Reagent Company Lim-ited (China), were analytical reagent grade. Distilled water for ana-lytical use was from Direct-Q3 (MilliPore, USA). Reagents andmaterials used in this experiment were sterilized by autoclavingfor 20 min at 120 �C.

2.4. Experimental methods

Petri dishes (90 mm diameter) with 40 mL samples wereexposed to the UV in the Collimated Beam Apparatus and stirredgently by a magnetic stirring apparatus. H2O2 (30%) was dilutedand added to the B. atrophaeus spores samples to achieve variousfinal concentrations. As soon as the designed exposure time wasfinished, 1 mL Na2SO3 was added to cease further oxidation pro-cess. Then 1 mL sample was taken out, serially diluted and incu-bated on culture medium in the dark. Samplings were performedat various intervals from 0 to 10 min. An adjustable volume pipettewas used to transfer the liquid.

2.5. Data presentation

In order to evaluate the effect of disinfection, the inactivationeffect of B. atrophaeus spores is usually analyzed via various models[19–22]. This study selected two different models to fit the exper-imental results.

2.5.1. The Hom ModelThe Hom Model is given as

lgðN0=NÞ ¼ k0th ð1Þ

where N0 and N are microbial concentrations (CFU/mL) before andafter disinfection, t refers to the exposure time (min), k0 refers tothe constant rate and h refers to a second parameter. This equationexists an initial shoulder delay when h > 1, or a final trailing curvewhen h < 1. Since it is incapable to illustrate the simultaneous pres-ence of both phenomena, the modified Hom Model was testedinstead, shown as follows:

lgðN0=NÞ ¼ k1½1� expð�k2tÞ�k3 ð2Þ

This three-parameter model is suitable to fit the disinfectionprocesses with a shoulder delay at the beginning, a log-linearregion in the middle and a tail behavior in the end [24].

2.5.2. The Chick’s ModelThe Chick’s Model conveyed the primary principles of conven-

tional disinfection processes [23]. It is described as linear relation-ship between the inactivation effect and the exposure time [25]:

lgðN0=NÞ ¼ kt ð3Þ

where k, the slope of the line, is the pseudo-first order rate constant(min�1). The inactivation of B. atrophaeus spores shows a pseudo-first-order kinetic behavior, while the goodness of fit is presentedby correlation coefficient (R2) and the slope of the line is stated aspseudo-first order rate constant (k). The Chick’s Model has beenwidely utilized in the literatures to compare the inactivation effectthrough the values of k. In addition, other parameters, such as stan-dard deviation and root mean square error, were also calculated.

3. Results and discussion

3.1. Effect of initial H2O2 dose

The experiments were carried out to investigate the effect ofinitial H2O2 dose on B. atrophaeus spores inactivation at irradiance

Table 1Fitting equation at different initial H2O2 dosages (irradiance = 113.0 lW/cm2; initialcell density = 107 CFU/mL and initial solution pH = 7).

H2O2 dose (mM) Fitting equation

0 lgðN0=NÞ ¼ 12:08½1� expð�0:059tÞ�1:43

0.1 lgðN0=NÞ ¼ 6:209½1� expð�0:216tÞ�2:22

0.3 lgðN0=NÞ ¼ 13:26½1� expð�0:056tÞ�1:151

0.5 lgðN0=NÞ ¼ 10:96½1� expð�0:107tÞ�1:458

0.7 lgðN0=NÞ ¼ 12:39½1� expð�0:081tÞ�1:306

0.9 lgðN0=NÞ ¼ 8:724½1� expð�0:122tÞ�1:430

Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15 11

of 113.0 lW/cm2, initial cell density of 107 CFU/mL and initial pH 7.As shown in Fig. 2, at a contact time of 10 min, the inactivationeffect increased from 3.85 to 5.88 lg with the H2O2 dosageincreased from 0 to 0.5 mM, indicating that UV/H2O2 was moreefficient than UV treatment alone. However, the inactivation effectreduced to 5.28 lg when H2O2 concentration increased to 0.9 mM,referring that the exorbitant H2O2 dose exhibited a negative effect.The fitting results of the modified Hom Model were illustrated inFig. 2(a), while the fitting equations were presented in Table 1. Itis no doubt that the modified Hom Model successfully fit theexperimental data, since the correlation coefficients were close to1. However, the three-parameters-model made it complicated tocompare the inactivation effect. In order to give a striking contrast,the Chick’s Model was proposed in Fig. 2(b) and the pseudo-first-order rate constants (k) of various initial H2O2 dose were exhibitedin the inset of Fig. 2(b). It was apparent that k increased from 0.365to 0.597 min�1 with the increasing H2O2 concentration from 0 to0.5 mM, while dropped to 0.544 min�1 at 0.9 mM H2O2. As a result,an optimum value at about 0.5 mM for the H2O2 concentration wasobserved in this study. Although this conclusion is valid only underthe reported experimental conditions, it can be used to guide theH2O2 dose selection in inactivating B. atrophaeus spores by UV/H2O2 to some extent. The root mean squares for the inactivationeffects shown in Fig. 2 were presented in Table SI 1.

This phenomenon could be explained by dual roles H2O2 playedin the reactions, both promoter and scavenger of �OH [27–29]. Atlow H2O2 concentration, the quantity of �OH is the critical factorinfluencing photolysis rate. Through trapping photogenerated elec-trons and photolysis, increasing dosage of H2O2 was able toimprove quantities of �OH (Eq. (4)). However, terminal reactionsmay also be involved to reduce inactivation efficiency at high

0

1

2

3

4

5

6

0 mM, R2=0.9940.1mM, R2=0.9920.3 mM, R2=0.9940.5 mM, R2=0.9990.7 mM, R2=0.9980.9 mM, R2=0.997

time (min)

/lg

(N

0N

)

(a)

0

1

2

3

4

5

6

0 mM, R2=0.9840.1mM, R2=0.9790.3 mM, R2=0.9930.5 mM, R2=0.9950.7 mM, R2=0.9960.9 mM, R2=0.993

time (min)

/lg

(N

0N

)

0.3

0.5

0.7

H2O2 dose

k (m

in-1

)

(mM)

(b)

0 2 4 6 8 10

0 2 4 6 8 10

0.0 0.3 0.6 0.9

Fig. 2. Effect of initial H2O2 dosages on B. atrophaeus spores inactivation: (a) themodified Hom Model, (b) the Chick’s Model (irradiance = 113.0 lW/cm2; initial celldensity = 107 CFU/mL and initial solution pH = 7). Inset of (b) shows the variation ofpseudo-first-order rate constant with respect to H2O2 dose. The plot represents themean value, and the error bar represents the standard deviation value. The numberof data for each plot was 3.

H2O2 concentration. H2O2 in excess acts as a scavenger of highlyreactive �OH to form �HO2 and O2 with less oxidizing capacity(Eqs. (5) and (6)). Moreover, high concentration of �OH will readilyself-recombine to produce H2O2 (Eq. (7)), which is undesirable.

H2O2 þ hm���! 2�OH ð4Þ

H2O2 þ �OH���! H2Oþ �HO2 ð5Þ

�OHþ �HO2 ���! H2Oþ O2 ð6Þ

�OHþ �OH ���! H2O2 ð7Þ

Some results have been reported to prove the existence of H2O2

optimum concentration using UV/H2O2. Chang et al. [30] discov-ered that highest azo dye decomposition rate was obtained atH2O2 concentration of 8.42 mM with azo dye concentration of20 mg/L. Zhang et al. [27] investigated the rapid photocatalyticdecolorization of methylene blue, indicating that 2.94 mM wasfound to be the optimum H2O2 concentration. He et al. [31] provedthe threshold concentration of H2O2 to be 0.882 mM under0.27 mW/cm2 irradiance and 1 lM microcystin-LR in microcy-stin-LR degradation. The variation could be ascribed to differentmaterial characteristics. Commonly, microorganisms required lessH2O2 than organic compounds, since they possessed comparativelylarger sizes and more feasible to be attacked [26].

3.2. Effect of UV irradiance

In order to investigate the effect of irradiance on B. atrophaeusspores inactivation, three reaction solutions were prepared andexposed to irradiance of 113.0, 56.5, 28.3 lW/cm2, respectively,while initial cell density was 107 CFU/mL and initial pH was 7.The chosen H2O2 concentration was 0.5 mM, which was the opti-mum H2O2 concentration under reported experimental condition.Fig. 3 illustrates that better inactivation effect was received athigher irradiance. Taking exposure time of 10 min as an example,the inactivation effect of B. atrophaeus spores decreased from5.88 to 3.08 lg with the irradiance decreased from 113.0 to28.3 lW/cm2, namely a substantial 47.6% decrease. The fittingcurves of the modified Hom Model were demonstrated inFig. 3(a), while the fitting equations were shown in Table 2. Obvi-ously, the modified Hom Model matched the experimental resultsvery well. However, it is unable to make a comparison of the inac-tivation effect. As a result, the Chick’s Model was utilized (Fig. 3(b))and the pseudo-first-order rate constants (k) were presented in theinset of Fig. 3(b), which showed that k decreased from 0.597 to0.308 min�1 with the irradiance decreased from 113.0 to28.3 lW/cm2. The root mean squares for the inactivation effectsshown in Fig. 3 were presented in Table SI 2.

This could be attributed to the increasing amount of radiationphoton by the improvement of irradiance. According to Eq. (2),the quantity of �OH was closely related to the number of radiationphoton. Increasing UV irradiance improved the �OH produced in

0

1

2

3

4

5

6

113.0 µW/cm2, R2=0.999 56.5 µW/cm2, R2=0.988 28.3 µW/cm2, R2=0.989

time (min)

/lg

(N

0N

)

(a)

0

1

2

3

4

5

6

113.0 µW/cm2, R2=0.995 56.5 µW/cm2, R2=0.971 28.3 µW/cm2, R2=0.985

time (min)

/lg

(N

0N

)

0 50 100 1500.0

0.3

0.6

0.9

UV irradiance

k (m

in-1

)

(µW/cm2)

(b)

0 2 4 6 8 10

0 2 4 6 8 10

Fig. 3. Effect of UV irradiances on B. atrophaeus spores inactivation: (a) the modifiedHom Model, (b) the Chick’s Model (initial H2O2 dose = 0.5 mM; initial celldensity = 107 CFU/mL and initial solution pH = 7). Inset of (b) shows the variationof pseudo-first-order rate constant with UV irradiance. The plot represents themean value, and the error bar represents the standard deviation value. The numberof data for each plot was 3.

Table 2Fitting equation at different UV irradiances (initial H2O2 dose = 0.5 mM; initial celldensity = 107 CFU/mL and initial solution pH = 7).

UV irradiance (lW/cm2) Fitting equation

113 lgðN0=NÞ ¼ 10:96½1� expð�0:107tÞ�1:458

56.5 lgðN0=NÞ ¼ 5:326½1� expð�0:254tÞ�2:146

28.3 lgðN0=NÞ ¼ 6:001½1� expð�0:1tÞ�1:466

0

1

2

3

4

5

6

7

108 CFU/mL, R2=0.993107 CFU/mL, R2=0.999106 CFU/mL, R2=0.963

(a)

time (min)

/lg

(N

0N

)

0

1

2

3

4

5

6

7

108 CFU/mL, R2=0.962107 CFU/mL, R2=0.995106 CFU/mL, R2=0.980

time (min)

/lg

(N

0N

)

105 106 107 108 1090.1

0.3

0.5

0.7

cell density (CFU/mL)

k (m

in-1

)

(b)

0 2 4 6 8 10

0 2 4 6 8 10

Fig. 4. Effect of initial cell densities on B. atrophaeus spores inactivation: (a) themodified Hom Model, (b) the Chick’s Model (initial H2O2 dose = 0.5 mM, irradi-ance = 113.0 lW/cm2 and initial solution pH = 7). Inset of (b) shows the variation ofpseudo-first-order rate constant with initial cell density. The plot represents themean value, and the error bar represents the standard deviation value. The numberof data for each plot was 3.

Table 3Fitting equation at different initial cell densities (initial H2O2 dose = 0.5 mM,irradiance = 113.0 lW/cm2 and initial solution pH = 7).

Cell density (CFU/mL) Fitting equation

106lgðN0=NÞ ¼ 7:94½1� expð�0:171tÞ�1:427

107lgðN0=NÞ ¼ 10:96½1� expð�0:107tÞ�1:458

108 –a

a Ambiguous.

12 Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15

the UV/H2O2 system, resulting in enhanced inactivation efficiency[32]. As a result, higher irradiance gave rise to better inactivationeffect [14,26]. It has been reported that irradiance plays a signifi-cant role in microcystin-LR removal [31], DNA photorepair of Esch-erichia coli [33] and dimethyl phthalate degradation [34] by UV/H2O2. Benabbou et al. [26] found that the required time for totallyinactivating E. coli by UV treatment increased from 90 to 180 min,when the irradiance decreased from 3.85 to 0.48 mW/cm2. Mean-while, the irradiance trend remained the same on E. coli inactiva-tion with addition of 1.5 g/L TiO2. Wu et al. [35] reported that UVirradiation plays a significant character removal in UV/Ag–TiO2/O3 system. The inactivation effect of Dunaliella salina improvedfrom 0.98 to 1.17 lg with the irradiance increased from 1.2 to6.5 mW/cm2, within an exposure time of 15 s. Although other bio-chemical responses may occur, since D. salina is a photosyntheticeukaryote, it still demonstrates the importance of UV irradiationto some extent.

3.3. Effect of initial cell density

The influence of the initial cell density was shown in Fig. 4, withH2O2 dose of 0.5 mM, irradiance of 113.0 lW/cm2 and pH 7. Theresults showed that the inactivation effect within 10 min resulted

in 5.88 lg and 2.78 lg, respectively, with the initial cell densityincreased from 107 to 108 CFU/mL, namely a 53% reduction. How-ever, reducing initial cell density from 107 to 106 CFU/mL did notenhance the inactivation efficiency of B. atrophaeus spores. Theobserved inactivation effect within 10 min was 5.99 lg for106 CFU/mL, indicating that the increase was only 2%.

As demonstrated in Fig. 4(a), the modified Hom Model perfectlymatched the experimental results, with the fitting equationsshown in Table 3. While in the light of the Chick’s Model(Fig. 4(b)), at initial cell densities from 106 to 108 CFU/mL, all thereactions exhibited pseudo-first-order kinetic behaviors. Thesepseudo-first-order rate constants decreased from 0.641 to0.253 min�1 with the initial density increased from 106 to108 CFU/mL, as shown in the inset of Fig. 4(b). As expected, anincrease in the initial cell density slowed down the inactivationefficiency. The root mean squares for the inactivation effectsshown in Fig. 4 were presented in Table SI 3.

The phenomenon could be ascribed to two reasons [30,36]:

(1) The limited �OH generated in the UV/H2O2 processdominates the reaction rate. As high initial cell density wasable to arouse a high internal optical density of solution, it

Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15 13

prevented UV treatment from penetrating towards the solu-tion immediately and slowed down the generation of �OH. Asa result, the improvement of initial cell density would resultin a negative impact on inactivation effect, and vice versa.

(2) As initial cell density increased, the ratio of H2O2 to celldecreased. Accordingly, the amount of cell inactivated bythe �OH was relatively lower against a higher initial density.This could give rise to the deceleration of the inactivationeffect as well.

The various effects of the initial density of treated material werealso reported previously. Haji et al. [36] observed a high degrada-tion rate of methyl orange dye at high dye concentration usingUV–H2O2, since the possibility of collisions between OH and thestuff increased at higher density. Li et al. [37] believed that themicrocystin-LR degradation rate decreased with its concentrationincreased using UV–H2O2. Sontakke et al. [38] showed the initialcell concentration exhibited a significant effect on the E. coli andPichia pastoris inactivation using UV–Ag/TiO2, while decreasingthe initial cell concentration of both microorganisms led to anincrease in the inactivation rate. The variations could also beattributed to diverse nature of materials.

3.4. Effect of the initial solution pH

Experiments were performed at five initial solution pH valuevaried from 3 to 11 at 0.5 mM H2O2, irradiance of 113.0 lW/cm2

and initial cell density of 107 CFU/mL to investigate the effect ofinitial pH. As shown in Fig. 5, at an exposure time of 10 min, theinactivation effect increased from 4.40 to 5.88 lg with pH increas-ing from 3 to 5, then decreased to 3.62 lg with pH increasing to 11.

0

1

2

3

4

5

6

7

pH=7, R2=0.999pH=9, R2=0.992pH=11, R2=0.986

pH=3, R2=0.990pH=5, R2=0.999

time (min)

/lg

(N

0N

)

(a)

0.3

0.5

0.7

0

1

2

3

4

5

6

7

pH=7, R2=0.995pH=9, R2=0.990pH=11, R2=0.972

pH=3, R2=0.983pH=5, R2=0.985

time (min)

/lg

(N

0N

)

pH

k (m

in-1

)

(b)

0 2 4 6 8 10

3 5 7 9 11

0 2 4 6 8 10

Fig. 5. Effect of initial solution pH on B. atrophaeus spores inactivation: (a) themodified Hom Model, (b) the Chick’s Model (initial H2O2 concentration = 0.5 mM,irradiance = 113.0 lW/cm2 and initial B. atrophaeus spores density = 107 CFU/mL).Inset of (b) shows the variation of pseudo-first-order rate constant with initialsolution pH. The plot represents the mean value, and the error bar represents thestandard deviation value. The number of data for each plot was 3.

The modified Hom Model method was applied to the experi-mental data as demonstrated in Fig. 5(a), resulting in the equations(Table 4) closely fit the data. While according to the fitting curvesof the Chick’s Model (Fig. 5(b)) and the pseudo-first-order rateconstants (k) (inset of Fig. 5(b)), it was obviously that the rate con-stants increased from 0.461 to 0.657 min�1 with the pH increasedfrom 3 to 5. Above the threshold pH value, rate constants variedsignificantly within the range of 0.657–0.389 min�1 at pH 5–11.Pseudo-first-order rate constants followed the sequence: pH5 > pH 7 > pH 9 > pH 3 > pH 11. It may give an instruction to thepH selection in B. atrophaeus spores inactivation by UV/H2O2. Theroot mean squares for the inactivation effects shown in Fig. 5 werepresented in Table SI 4.

The opposite effect of pH may be ascribed to the following rea-sons [14,34,39,40]. According to Eqs. (8) and (9), more �OH weregenerated under alkaline conditions, while �OH is more reactivethan H2O2. However, according to Eqs. (10) and (11), the interme-diate product hydroperoxide anion (HO�2 ) acted as a scavenger of�OH. Moreover, high pH may also strongly impel the self-decomposi-tion of H2O2 through Eq. (12). In conclusion, a slight acid condition atpH 5 was considered as the optimal pH value.

H2O2 ���! HO�2 þHþ ð8Þ

HO�2 þHþ þ hm ���! 2�OH ð9Þ

HO�2 þ �OH ���! �HO2 þ OH� ð10Þ

HO�2 þH2O2 ���! H2Oþ O2 þ OH� ð11Þ

2H2O2 ���! 2H2Oþ O2 ð12Þ

In previous researches, Aleboyeh et al. [39] reported that thehighest decolorization rates of Acid Blue 74 were found at pH rangebetween 3.5 and 5.5 using UV/H2O2. Xu et al. [34] observed that theoptimum degradation effect for dimethyl phthalate removal byUV/H2O2 was at pH 4.0. Pelaez et al. [41] determined the highestinitial reaction rate of microcystin-LR under visible light-activatedTiO2 was achieved at pH 3.0. In conclusion, both organic contami-nant and microcystin being treated with UV/H2O2 showed greatsimilarity with this work.

3.5. Effect of different inorganic anions

By adding 1 mM anions into the system respectively, the effectsof three inorganic anions were investigated and compared, includ-ing sulfate (SO4

2�), nitrate (NO3�) and carbonate (CO3

2�). As illus-trated in Fig. 6, the sequence of inactivation effect within 10 minfollowed the order of control group (5.88 lg) > SO4

2� (5.30 lg) > NO�3(3.52 lg) > CO3

2� (2.97 lg).Compared with control group, only a slight reduction of 0.58 lg

was observed with the addition of SO2�4 , since SO2�

4 was unable tocompete with B. atrophaeus spores for �OH [7].

In contrast, NO�3 resulted in a negligible decrease of 2.36 lgwithin 10 min, namely a 40.1% reduction. The inhibition effect ofNO�3 is attributed to its ability to react with �OH during the photoly-sis. In fact, NO�3 showed a dual character in the process [42]. On onehand, it contributes to the production of additional �OH under UVtreatment (Eqs. (13)–(17)). On the other hand, it also acts as an innerfilter and cuts down the absorbed radiation photon account.Whether character dominates relies heavily on the reaction condi-tions. Apparently the negative effect triumphs over the positiveeffect in this research.

NO�3 þ hm ���! NO�2 þ O ð13Þ

NO�3 þ hm ���! �O� þ �NO2 ð14Þ

Table 4Fitting equation at different initial solution pH (initial H2O2 concentration = 0.5 mM,irradiance = 113.0 lW/cm2 and initial B. atrophaeus spores density = 107 CFU/mL).

pH Fitting equation

3 lgðN0=NÞ ¼ 6:37½1� expð�0:162tÞ�1:627

5 lgðN0=NÞ ¼ 7:594½1� expð�0:229tÞ�1:921

7 lgðN0=NÞ ¼ 10:96½1� expð�0:107tÞ�1:458

9 lgðN0=NÞ ¼ 13:16½1� expð�0:053tÞ�1:094

11 lgðN0=NÞ ¼ 4:526½1� expð�0:226tÞ�1:911

0

1

2

3

4

5

6

Control, R2=0.999SO4

2-, R2=0.988NO3

-, R2=0.997

CO32-, R2=0.984

time (min)

/lg

(N

0N

)

(a)

0 2 4 6 8 10

0 2 4 6 8 100

1

2

3

4

5

6

Control, R2=0.995SO4

2-, R2=0.981NO3

-, R2=0.972

CO32-, R2=0.980

time (min)

/lg

(N

0N

)

Contro

lSO 4

2-

NO 3-

CO 32-

0.2

0.4

0.6

k (m

in-1

)

(b)

Fig. 6. Effect of inorganic anions on B. atrophaeus spores inactivation: (a) themodified Hom Model, (b) the Chick’s Model (initial H2O2 dose = 0.5 mM; irradi-ance = 113.0 lW/cm2; initial cell density = 107 CFU/mL; initial solution pH = 7 andinorganic anions concentrations of 2 mM). Inset of (b) shows the variation ofpseudo-first-order rate constant with respect to inorganic anions. The plotrepresents the mean value, and the error bar represents the standard deviationvalue. The number of data for each plot was 3.

Table 5Fitting equation with inorganic anions (initial H2O2 dose = 0.5 mM; irradi-ance = 113.0 lW/cm2; initial cell density = 107 CFU/mL; initial solution pH = 7 andinorganic anions concentrations of 2 mM).

Inorganic anions Fitting equation

Control lgðN0=NÞ ¼ 10:96½1� expð�0:107tÞ�1:458

SO42�

lgðN0=NÞ ¼ 8:613½1� expð�0:115tÞ�1:27

NO3�

lgðN0=NÞ ¼ 4:066½1� expð�0:275tÞ�2:011

CO32�

lgðN0=NÞ ¼ 6:157½1� expð�0:099tÞ�1:492

14 Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15

2�NO2 þH2O���! NO�2 þ NO�3 þ 2Hþ ð15Þ

OþH2O ���! 2�OH ð16Þ

�O� þH2O ���! �OHþ OH� ð17Þ

In addition, CO2�3 exhibited a more significant inhibiting effect

and slowed down the inactivation efficiency evidently. About

2.91 lg of decrease was observed within 10 min, namely a 49.5%reduction. Since the addition of CO2�

3 would quickly form HCO�3and reach the equilibration of carbonate system (Eq. (18)), both ofthe anions act as scavengers of �OH (Eqs. (19)–(22)) Accordingly, itmakes an inhibition effect by producing radicals with less oxidizingcapacity [15,37].

CO2�3 þH2O ���! HCO�3 þ OH� ð18Þ

�OHþHCO�3 ���! �HCO3 þ OH� ð19Þ

�HCO3 ���! �CO�3 þHþ ð20Þ

�OHþ CO2�3 ���! �CO�3 þ OH� ð21Þ

�CO�3 þH2O2 ���! �HO2 þHCO�3 ð22Þ

As in the above case, the modified Hom Model well fitted theexperimental data as illustrated in Fig. 6(a) and the results wereshown in Table 5. While on the basis of the Chick’s Model(Fig. 6(b)), the constant rates gave a striking comparison. It indi-cated that the rates of control group, SO2�

4 , NO�3 and CO2�3 were

0.597, 0.558, 0.389 and 0.306 min�1, respectively. The rates ofCO2�

3 and NO�3 were far below that of control group, referring thatthe reaction between the two anions and �OH slowed down the inac-tivation effect. The results are in agreement with other publishedinvestigations [7,15,37]. The root mean squares for the inactivationeffects shown in Fig. 6 were presented in Table SI 5.

4. Conclusions

The inactivation of B. atrophaeus spores followed both the mod-ified Hom Model and the Chick’s Model. Inactivation effect of B.atrophaeus spores was affected by various parameters, such as ini-tial H2O2 dose, UV irradiance, initial cell density, initial solution pHand various inorganic anions. H2O2 played both promoter andscavenger of �OH in the reactions. The optimum H2O2 concentra-tion was found to be 0.5 mM, which resulted in 5.88 lg reductionwithin 10 min. The inactivation effect exhibited a substantial47.6% decrease when irradiance decreased with 113.0–28.3 lW/cm2. Reducing initial cell density made a contribution to the decel-eration of inactivation effect. The highest inactivation effect wasobserved at pH 5, following the order: pH 5 > pH 7 > pH 9 > pH3 > pH 11. Inorganic anions exhibited a negligible inhibiting effectand slowed down the inactivation efficiency to different degrees.The inhibiting effect on B. atrophaeus spores inactivation followedthe order: SO4

2� < NO3� < CO3

2�.

Acknowledgements

This work was supported by the National Natural ScienceFoundation of China (Grant No. 51178323 and 51108329), ChinaPostdoctoral Science Foundation funded project (No.2012T50413), and the Fundamental Research Funds for the CentralUniversities (No. 0400219205).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jphotobiol.2014.03.022.

References

[1] D. Fritze, R. Pukall, Reclassification of bioindicator strains Bacillus subtilis DSM675 and Bacillus subtilis DSM 2277 as Bacillus atrophaeus, Int. J. Syst. Evol.Microbiol. 51 (2001) 35–37.

Y. Zhang et al. / Journal of Photochemistry and Photobiology B: Biology 134 (2014) 9–15 15

[2] A. Driedger, E. Staub, U. Pinkernell, B. Mariñas, W. Köster, U.V. Gunten,Inactivation of Bacillus subtilis spores and formation of bromate duringozonation, Water Res. 35 (2001) 2950–2960.

[3] K. Warriner, G. Rysstad, A. Murden, P. Rumsby, D. Thomas, W. Waites,Inactivation of Bacillus subtilis spores on packaging surfaces by UV excimerlaser irradiation, J. Appl. Microbiol. 88 (2000) 678–685.

[4] A.L. Santos, N.C.M. Gomes, I. Henriques, A. Almeida, A. Correia, Â. Cunha,Contribution of reactive oxygen species to UV-B-induced damage in bacteria, J.Photochem. Photobiol., B 117 (2012) 40–46.

[5] J. Kim, H.-K. Cho, J. Mok, H.D. Yoo, N. Cho, Effects of ozone and aerosol onsurface UV radiation variability, J. Photochem. Photobiol., B 119 (2013) 46–51.

[6] M.G. Antoniou, A.A. de la Cruz, D.D. Dionysiou, Degradation of microcystin-LRusing sulfate radicals generated through photolysis, thermolysis and e�

transfer mechanisms, Appl. Catal., B 96 (2010) 290–298.[7] N. Gao, Y. Deng, D. Zhao, Ametryn degradation in the ultraviolet (UV)

irradiation/hydrogen peroxide (H2O2) treatment, J. Hazard. Mater. 164(2009) 640–645.

[8] E. Timchak, V. Gitis, A combined degradation of dyes and inactivation ofviruses by UV and UV/H2O2, Chem. Eng. J. 192 (2012) 164–170.

[9] H. Mamane, H. Shemer, K.G. Linden, Inactivation of E. coli, B. subtilis spores, andMS2, T4, and T7 phage using UV/H2O2 advanced oxidation, J. Hazard. Mater.146 (2007) 479–486.

[10] D. Gardner, G. Shama, The kinetics of Bacillus subtilis spore inactivation onfilter paper by U.V. light and U.V. light in combination with hydrogen peroxide,J. Appl. Microbiol. 84 (1998) 633–641.

[11] M.A. Larson, B.J. Mariñas, Inactivation of Bacillus subtilis spores with ozone andmonochloramine, Water Res. 37 (2003) 833–844.

[12] D. Wang, T. Oppenlander, M.G. El-Din, J.R. Bolton, Comparison of thedisinfection effects of vacuum-UV (VUV) and UV Light on Bacillus subtilisspores in aqueous suspensions at 172, 222 and 254 nm, Photochem. Photobiol.86 (2010) 176–181.

[13] F. Zhang, J. Xi, J.-J. Huang, H.-Y. Hu, Effect of inlet ozone concentration on theperformance of a micro-bubble ozonation system for inactivation of Bacillussubtilis spores, Sep. Purif. Technol. 114 (2013) 126–133.

[14] B. Xu, N.-Y. Gao, X.-F. Sun, S.-J. Xia, M. Rui, M.-O. Simonnot, C. Causserand, J.-F.Zhao, Photochemical degradation of diethyl phthalate with UV/H2O2, J. Hazard.Mater. 139 (2007) 132–139.

[15] C. Zhou, N. Gao, Y. Deng, W. Chu, W. Rong, S. Zhou, Factors affecting ultravioletirradiation/hydrogen peroxide (UV/H2O2) degradation of mixed N-nitrosamines in water, J. Hazard. Mater. 231–232 (2012) 43–48.

[16] A. Riga, K. Soutsas, K. Ntampegliotis, V. Karayannis, G. Papapolymerou, Effectof system parameters and of inorganic salts on the decolorization anddegradation of Procion H-exl dyes. Comparison of H2O2/UV, Fenton, UV/Fenton, TiO2/UV and TiO2/UV/H2O2 processes, Desalination 211 (2007) 72–86.

[17] J.R. Bolton, K.G. Linden, Standardization of methods for fluence (UV dose)determination in bench-scale UV experiments, J. Environ. Eng. 129 (2003)209–215.

[18] W. Liu, Y. Zhang, Effects of UV intensity and water turbidity on microbialindicator inactivation, J. Environ. Sci. 18 (2006) 650–653.

[19] F. Chen, X. Yang, H.K. Mak, D.W. Chan, Photocatalytic oxidation forantimicrobial control in built environment: a brief literature overview, Build.Environ. 45 (2010) 1747–1754.

[20] R.A. Fenner, K. Komvuschara, A new kinetic model for ultraviolet disinfectionof greywater, J. Environ. Eng. 131 (2005) 850–864.

[21] M. Iannotti, R. Pisani Jr, Inactivation of Bacillus atrophaeus spores in healthcarewaste by UV light coupled with H2O2, Braz. J. Chem. Eng. 30 (2013) 507–519.

[22] R. Lambert, M. Johnston, Disinfection kinetics: a new hypothesis and modelfor the tailing of log-survivor/time curves, J. Appl. Microbiol. 88 (2000) 907–913.

[23] O.K. Dalrymple, E. Stefanakos, M.A. Trotz, D.Y. Goswami, A review of themechanisms and modeling of photocatalytic disinfection, Appl. Catal., B 98(2010) 27–38.

[24] J. Marugán, R. van Grieken, C. Sordo, C. Cruz, Kinetics of the photocatalyticdisinfection of Escherichia coli suspensions, Appl. Catal., B 82 (2008) 27–36.

[25] W.A.M. Hijnen, E.F. Beerendonk, G.J. Medema, Inactivation credit of UVradiation for viruses, bacteria and protozoan (oo)cysts in water: a review,Water Res. 40 (2006) 3–22.

[26] A. Benabbou, Z. Derriche, C. Felix, P. Lejeune, C. Guillard, Photocatalyticinactivation of Escherischia coli: Effect of concentration of TiO2 andmicroorganism, nature, and intensity of UV irradiation, Appl. Catal., B 76(2007) 257–263.

[27] Q. Zhang, C. Li, T. Li, Rapid photocatalytic decolorization of methylene blueusing high photon flux UV/TiO2/H2O2 process, Chem. Eng. J. 217 (2013) 407–413.

[28] A. Afzal, P. Drzewicz, J.W. Martin, M. Gamal, El-Din, Decomposition ofcyclohexanoic acid by the UV/H2O2 process under various conditions, Sci.Total. Environ. 426 (2012) 387–392.

[29] C. Wu, K.G. Linden, Degradation and byproduct formation of parathion inaqueous solutions by UV and UV/H2O2 treatment, Water Res. 42 (2008) 4780–4790.

[30] M.-W. Chang, C.-C. Chung, J.-M. Chern, T.-S. Chen, Dye decomposition kineticsby UV/H2O2: initial rate analysis by effective kinetic modelling methodology,Chem. Eng. Sci. 65 (2010) 135–140.

[31] X. He, M. Pelaez, J.A. Westrick, K.E. O’Shea, A. Hiskia, T. Triantis, T. Kaloudis,M.I. Stefan, A.A. de la Cruz, D.D. Dionysiou, Efficient removal of microcystin-LRby UV–C/H2O2 in synthetic and natural water samples, Water Res. 46 (2012)1501–1510.

[32] J. Zhao, V. Krishna, B. Hua, B. Moudgil, B. Koopman, Effect of UVA irradiance onphotocatalytic and UVA inactivation of Bacillus cereus spores, J. Photochem.Photobiol., B 94 (2009) 96–100.

[33] Z. Bohrerova, K.G. Linden, Standardizing photoreactivation: comparison ofDNA photorepair rate in Escherichia coli using four different fluorescent lamps,Water Res. 41 (2007) 2832–2838.

[34] B. Xu, N. Gao, H. Cheng, S. Xia, M. Rui, D. Zhao, Oxidative degradation ofdimethyl phthalate (DMP) by UV/H2O2 process, J. Hazard. Mater. 162 (2009)954.

[35] D. Wu, H. You, J. Du, C. Chen, D. Jin, Effects of UV/Ag–TiO2/O3 advancedoxidation on unicellular green alga Dunaliella salina: implications for removalof invasive species from ballast water, J. Environ. Sci. 23 (2011) 513–519.

[36] S. Haji, B. Benstaali, N. Al-Bastaki, Degradation of methyl orange by UV/H2O2

advanced oxidation process, Chem. Eng. J. 168 (2011) 134–139.[37] L. Li, N. Gao, Y. Deng, J. Yao, K. Zhang, H. Li, D. Yin, H. Ou, J. Guo, Experimental

and model comparisons of H2O2 assisted UV photodegradation of Microcystin-LR in simulated drinking water, J. Zhejiang Univ. Sci. A 10 (2009) 1660–1669.

[38] S. Sontakke, J. Modak, G. Madras, Photocatalytic inactivation of Escherischia coliand Pichia pastoris with combustion synthesized titanium dioxide, Chem. Eng.J. 165 (2010) 225–233.

[39] A. Aleboyeh, Y. Moussa, H. Aleboyeh, The effect of operational parameters onUV/H2O2 decolourisation of acid blue 74, Dyes Pigm. 66 (2005) 129–134.

[40] C.-H. Wu, Effects of operational parameters on the decolorization of C.I.Reactive red 198 in UV/TiO2-based systems, Dyes Pigm. 77 (2008) 31–38.

[41] M. Pelaez, A.A. de la Cruz, K. O’Shea, P. Falaras, D.D. Dionysiou, Effects of waterparameters on the degradation of microcystin-LR under visible light-activatedTiO2 photocatalyst, Water Res. 45 (2011) 3787–3796.

[42] M. Sörensen, F.H. Frimmel, Photochemical degradation of hydrophilicxenobiotics in the UVH2O2 process: influence of nitrate on the degradationrate of EDTA, 2-amino-1-naphthalenesulfonate, diphenyl-4-sulfonate and 4,40-diaminostilbene-2,20-disulfonate, Water Res. 31 (1997) 2885–2891.