Environ Monit Assess (2017) 189:452 DOI 10.1007/s10661-017-6157-2
Benzene-contaminated groundwater remediation using calcium peroxide nanoparticles: synthesis and process optimization Hamid Mosmeri & Ebrahim Alaie & Mahmoud Shavandi & Seyed Mohammad Mehdi Dastgheib & Saeideh Tasharrofi
Received: 31 October 2016 / Accepted: 1 August 2017 # Springer International Publishing AG 2017
Abstract Nano-size calcium peroxide (nCaO2) is an appropriate oxygen source which can meet the needs of in situ chemical oxidation (ISCO) for contaminant remediation from groundwater. In the present study, an easy to handle procedure for synthesis of CaO2 nanoparticles has been investigated. Modeling and optimization of synthesis process was performed by application of response surface methodology (RSM) and central composite rotatable design (CCRD) method. Synthesized nanoparticles were characterized by XRD and FESEM techniques. The optimal synthesis conditions were found to be 5:1, 570 rpm and 10 °C for H2O2:CaSO2 ratio, mixing rate and reaction temperature, respectively. Predicted values showed to be in good agreement with experimental results (R2 values were 0.915 and 0.965 for CaO2 weight and nanoparticle size, respectively). To study the efficiency of synthesized nanoparticles for benzene removal from
: M. Shavandi : S. Tasharrofi
H. Mosmeri Ecology and Environmental Pollution Control Research Group, Environment and Biotechnology Research Division, Research Institute of Petroleum Industry, Tehran, Iran H. Mosmeri e-mail:
[email protected] E. Alaie (*) Environment and Biotechnology Division, Research Institute of Petroleum Industry, Tehran, Iran e-mail:
[email protected] S. M. M. Dastgheib Microbiology and Biotechnology Group, Environment and Biotechnology Research Division, Research Institute of Petroleum Industry, Tehran, Iran
groundwater, batch experiments were applied in biotic and abiotic (chemical removal) conditions by 100, 200, 400, and 800 mg/L of nanoparticles within 70 days. Results indicated that application of 400 mg/L of CaO2 in biotic condition was able to remediate benzene completely from groundwater after 60 days. Furthermore, comparison of biotic and abiotic experiments showed a great potential of microbial stimulation using CaO2 nanoparticles in benzene remediation from groundwater. Keywords Chemical oxidation . Nanoparticle synthesis . Design expert . Benzene remediation . Groundwater
Introduction Groundwater contamination by petroleum hydrocarbons is one of the most important issues in developed and developing countries. This problem may have occurred because of leaking in underground tanks and pipelines or other accidents (Kvenvolden and Cooper 2003). Benzene, toluene, ethylbenzene, and xylene (BTEX) are considered dangerous monoaromatics because of their high water solubility and carcinogenic impacts. Among the BTEX group, due to higher water solubility (1790 mg/L at 15 °C) and lower LC50 benzene is an important water contaminant (Gillis et al. 2007; Van Raalte and Grasso 1982). Various physical, chemical, and biological methods have been investigated for benzene remediation from groundwater, such as pump and treat, chemical oxidation, and bioremediation (Vignola
452
Environ Monit Assess (2017) 189:452
Page 2 of 14
et al. 2008; Northup and Cassidy 2008; Yeh et al. 2010). Two main mechanisms are proposed for in situ benzene degradation: chemical oxidation and biological remediation (Volkamer et al. 2002; Fritsche and Hofrichter 2008). Hydroxyl radical is a strong oxidant which plays a key role in chemical oxidation. Attacking of monoaromatic rings by hydroxyl radicals primarily generates hydroxyl cyclohexadienyl (HCHD) radicals. In addition, different possible pathways proposed for reaction of HCHD radicals, but more than 80% of these radicals converted (or further oxidized) to phenol (Volkamer et al. 2002). It has been demonstrated that phenol can be degraded in presence of OH− and OH•, easier than benzene (J. Zhang et al. 2009). Hydrogen peroxide (H2O2) has been commonly used in chemical oxidation. Liquid H2O2 is unstable and release oxygen into aquatic environment only for several minutes to few hours (Schumb et al. 1955). Accordingly, short lifetime of H2O2 makes it inappropriate candidate for groundwater remediation. To overcome this problem, solid peroxides such as calcium peroxide, magnesium peroxide and sodium percarbonate have been successfully applied in groundwater remediation (Yehia et al. 2012; Schumb et al. 1955; Qian et al. 2013; Lee et al. 2014; Xin et al. 2013; Lin et al. 2012; Careghini et al. 2015; Odencrantz et al. 1996; Nykänen et al. 2012). Among these materials, CaO2 is synthesized in high purity with low water solubility (Chevalier and McCann 2008; Lewis et al. 2009). By exposure of CaO2 to water, hydrogen peroxide generated (Eq. 1) and depending on the physicochemical environmental conditions, H2O2 is further proceed to Eq. 2 or Eq. 3. CaO2 þ 2H2 O→H2 O2 þ Ca ðOHÞ2
ð1Þ
2H2 O2 →2H2 O þ O2
ð2Þ
H2 O2 þ e− →OH• þ OH−
ð3Þ
H2 O2 þ Fe2þ →OH• þ OH− þ Fe3þ
ð4Þ
which degrades benzene by generation of hydroxyl radicals in neutralized condition (Vignola et al. 2008; Lewis et al. 2009; Yehia et al. 2012). Several researches have shown that activation of CaO2 with Fe2+ and Fe3 helps to better control the pH and elevates the OH· generation rate (Zhou et al. 2017; X. Zhang et al. 2016; Xue et al. 2016). Recent studies have focused on improvement of CaO2 and nano-CaO2 (Qian et al. 2013) application for groundwater remediation. It has been shown that BTEX can be remediated efficiently at pH 6–7 (Qian et al. 2013). In addition, capsulation of CaO2 was studied as a novel method to control releasing of oxygen from synthesized beads (Lee et al. 2014; Xin et al. 2013). Application of calcium peroxide for remediation of groundwater by MF reaction has shown more detrimental environmental impacts than biostimulation of microbial population (Lin et al. 2012). Groundwater contains a wide range of microorganisms which play a key role in pollution degradation (Careghini et al. 2015; Mumford et al. 2015; Lien et al. 2016; Chen et al. 2010; Suja et al. 2014). Several commercial oxygen-releasing products are available in the market for environmental field applications such as PermeOx®Ultra,Klozur®CR,PermeOx®Plus,IXPER®, and Cool-Ox™ (Lu et al. 2017), but higher purity and more injectability of nano-sized CaO2 into groundwater, make the nano-CaO2 more suitable for field applications. The objective of this study was to propose an efficient and economically feasible method for synthesis of calcium peroxide nanoparticles and to investigate its application in the bioremediation of benzenecontaminated groundwater. Effects of H2O2 molar ratio, temperature, and mixing on purity and size of the nanoparticles in the synthesis process were studied. Response surface methodology (RSM) was applied to optimize the synthesis process and to improve the purity and performance of the synthesized nanoparticles. In addition, the effects of microbial stimulation on benzene remediation from groundwater by CaO2 nanoparticles were investigated in the batch experiments.
Materials and methods Chemicals
In the last decade, in situ chemical oxidation (ISCO) has been applied to remediate contaminated groundwater. ISCO typically uses modified Fenton (MF) reaction. In MF reaction, Fe2+ ions are added as catalyst in Eq. 4,
Chemicals benzene (99.9%) were prepared from SigmaAldrich and H2O2 (30%) and R2A agar purchased from Merck. All other chemicals were of analytical grade and obtained from commercial sources.
Environ Monit Assess (2017) 189:452
Page 3 of 14 452
Synthesis of nanoscale calcium peroxide
Nanoparticle synthesis process optimization
Calcium peroxide nanoparticles were synthesized by reaction of calcium sulfate and hydrogen peroxide. Five grams of calcium sulfate dispersed in 50 ml deionized (DI) water and 20 ml KOH (1 M) in a 250-ml beaker. Hydrogen peroxide was added to the reaction beaker in a molar ratio of 1:7 (CaSO4:H2O2) and the mixture maintained at room temperature (25 °C). The reaction was completed by 2 h vigorous mixing (Eq. 5). In this method, KOH was used to neutralize H2SO4 generated during CaO2 synthesis (Eq. 6).
Response surface methodology
CaSO4 þ H2 O2 →CaO2 ðhydrateÞ þ H2 SO4
ð5Þ
H2 SO4 þ KOH→K2 SO4 þ H2 O
ð6Þ
Where y is the response of the process, Xi is the variable and k is the number of variables. A secondorder model in RSM results the equation presented below:
Response surface methodology (RSM) is a technique commonly used for modeling and analyzing process. Optimization of the response surface of the influenced process parameters is the objective of RSM (Aslan 2007). By assumption of all variables to be measurable, the response surface can be written as Eq. 7. y ¼ f ðX 1 ; X 2 ; X 3 ; … X k Þ
k y ¼ β 0 þ∑ki¼1 β i X i þ∑ki¼1 βii X 2i þ∑k−1 i¼1 ∑ j¼1 β ij X i X j
Nanoparticles were separated from solution by centrifugation at 5000 rpm for 5 min (Eppendorf, model 5810R). The nanoparticles were washed two times by DI water and ethanol, respectively. Final precipitant was dried in Eppendorf concentrator, model 5301. Mira-3 FESEM (TESCAN) was used to determine the size distribution of nanoparticles. XRD analyses were carried out using a Philips X-ray diffraction system (PW1729) and the size of the nanoparticles was determined using Scherrer formula (Patterson 1939). Moreover, the purity of synthesized CaO2 was evaluated using titration with KMnO4 in acidic condition. Dissolved oxygen (DO) of an aquatic system as an indicator of oxygen-releasing potential of the nanoparticles was measured using HQ40d multimeter (HACH). Benzene analysis Benzene concentration was measured by highperformance liquid chromatography (HPLC) (SPDM10A, Shimadzu, Japan). The HPLC column length and internal diameter were 150 and 4.6 mm, respectively. Columns packed with the octadecyl group bonded type silica gel (C18). HPLC analysis performed at 1 ml per minute flow tests for 10 min with 65:35 of acetonitrile: water as mobile phase. Spectra was recorded at wavelength 254 nm.
ð7Þ
ð8Þ
Where Xi and Xj are the variables which influence the response, β0 is the intercept, βii represents the quadratic effect of Xi, and βij is the interaction between Xi and Xj (Aslan 2007, 2008).
Central composite rotatable design To achieve a statistical model, it is needed to design a series of experiments, then develop a statistical model based on second-order response surface and predicting the coefficients in the model and finally to check the model applicability. The central composite rotatable design (CCRD), which developed by Box and Wilson (Box and Wilson 1951) and improved by Box and Hunter (Box and Hunter 1957) is an applicable method to achieve a model. To develop calcium peroxide nanoparticle synthesis, 5-level-3-factor CCRD and response surface methodology (RSM) analysis were applied. To design the experiments, a statistical package DesignExpert version 7 was employed and analysis of results performed by analysis of variance (ANOVA). The value of each variable was defined as follows: molar ratio of H2O2 to CaSO4 of 5–10, mixing rate of 300–1000 rpm, and reaction temperature of 5–17.5 °C. The levels of variables and coded factors are shown in Table 1. Particle size and purified CaO2 weight were considered as the responses of the designed experiments.
452
Environ Monit Assess (2017) 189:452
Page 4 of 14
Table 1 Variables and their levels for CCRD Factors
Results and discussion
Symbol Actual level of coded factors − 1.68 1
0
+1
+ 1.68
H2O2: CaSO4 (M) X1
3.3
7.5
10
11.7
Mixing rate (rpm) X2
61
300 650
Temperature (°C)
5
10
X3
5
1000 1238
17.5 25
30
Benzene remediation in batch experiments by CaO2 nanoparticles The efficiency of nanoparticles for chemical benzene removal and bioremediation from groundwater were studied in batch experiments. The experiments designed by addition of 100, 200, 400, and 800 mg/L of nanoparticles to the 100 ml vials without any headspace containing benzene-contaminated groundwater (50 mg/L). The bioremediation test was conducted in the presence of native microbial community of the groundwater. Additionally, to study the ability of hydroxide radicals in chemical benzene removal (abiotic tests), different concentrations of calcium peroxide nanoparticles were added to sterilized groundwater in presence of sodium azide (1% w/v)to prevent any microbial activity and growth. The groundwater obtained from a well located in the Iranian Research Institute of Petroleum Industry (35° 69′ 61 North, 51° 42′ 31 East). The effect of nanoparticles on pH, DO, microbial population, and benzene concentration was studied to find out the efficiency of the nanoparticles in chemical benzene remediation. Furthermore, standard plate count (SPC) method was applied to study the microbial population on R2A agar using serial dilutions of the water samples in buffered sterile water. Fig. 1 X-ray diffraction (XRD) patterns of synthesized CaO2 and reference
Characteristics of synthesized nanoparticles In order to identify the obtained powder, XRD analysis was performed on the synthesized nanoparticles (Fig. 1). Four dominant peaks (2θ) at about 35.5°, 41.6°, 55.5°, and 62.8° strongly match to XRD of CaO2 (card number 03-0865).
Optimization of nano-CaO2 synthesis According to designed experiments, eighteen run were performed based on three factors and two levels of fractional and axial points and four replicates at the center point to optimize the synthesis process. The molar ratio of H2O2 to CaSO4 (X1), mixing rate (X2) and reaction temperature (X3) considered as independent variables and studied to predict y responses (size and pure weight of nanoparticles). The variable design matrix and the results of variables are presented in Table 2. Needless to say, the required experiments, calculated as follows (Box and Behnken 1960): 2k 23 ¼ 8; star points þ 2k ð2 3 ¼ 6; axial pointsÞ þ 4 ðcenter points; 4 replicatesÞ ¼ 18 run Analysis of variance (ANOVA) results for the model obtained from CCRD employed in pure weight and size of nanoparticles optimization are listed in Table 2. The prediction results showed good agreement with measured data, indicating accuracy of the proposed model. Probability value (P value) was statistically
Environ Monit Assess (2017) 189:452
Page 5 of 14 452
Table 2 Central composite rotatable design experiments and experimental and predicted results H2O2: CaSO4X1
Mixing rate X2
Temperature X3
Particle size (nm)
Pure CaO2 weight (g)
molar ratio
(rpm)
(°C)
Actual
Predicted
Actual
Predicted
1
11.7 (+ 1.68)
650 (0)
17.5 (0)
23.00
24.58
2.86
2.90
2
7.5 (0)
61 (− 1.68)
17.5 (0)
33.00
35.95
2.44
2.51
3
10 (+ 1)
300 (− 1)
10 (− 1)
62.00
59.25
4.24
3.86
4
7.5 (0)
650 (0)
5 (− 1.68)
34.00
36.69
3.94
4.06
5
7.5 (0)
650 (0)
30 (+ 1.68)
11.70
13.59
3.28
3.25
6
7.5 (0)
650 (0)
17.5 (0)
14.40
17.97
3.31
3.15
7
7.5 (0)
1238 (+ 1.68)
17.5 (0)
14.70
16.33
0.99
1.03
Run
8
10 (+ 1)
300 (− 1)
25 (+ 1)
22.40
19.99
2.86
3.16
9
10 (+ 1)
1000 (+ 1)
25 (+ 1)
13.40
11.55
2.54
2.24
10
5 (− 1)
300 (− 1)
10 (− 1)
25.00
23.61
3.23
3.46
11
7.5 (0)
650 (0)
17.5 (0)
17.00
17.97
3.14
3.15
12
5 (− 1)
1000 (+ 1)
25 (+ 1)
21.00
20.51
2.05
2.35
13
7.5 (0)
650 (0)
17.5 (0)
15.60
36.69
3.06
3.15
14
7.5 (0)
650 (0)
17.5 (0)
21.00
17.97
3.10
3.15
15
3.3 (− 1.68)
650 (0)
17.5 (0)
13.00
11.35
2.59
2.65
16
5 (− 1)
1000 (+ 1)
10 (− 1)
20.50
19.67
2.99
2.61
17
10 (+ 1)
1000 (+ 1)
10 (− 1)
20.70
18.51
1.61
1.86
18
5 (− 1)
300 (+ 1)
25 (+ 1)
15.10
14.05
2.44
2.12
significant (P value
