- Email: [email protected]
Application of ozone micro-nano-bubbles to groundwater remediation
Journal of Hazardous Materials 342 (2018) 446–453
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Application of ozone micro-nano-bubbles to groundwater remediation Liming Hu ∗ , Zhiran Xia State Key Laboratory of Hydro-Science and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, China
h i g h l i g h t s • • • •
This paper presents the high efficiency of gas supply and mass transfer in water using MNBs. The ozone MNBs technique was applied for in situ groundwater remediation of an organics-contaminated site. Ozone MNBs show considerable advantages in contaminant cleanup and time efficiency. Ozone MNBs potentially represent an innovative technology for in situ remediation of organics-contaminated groundwater.
a r t i c l e
i n f o
Article history: Received 25 February 2017 Received in revised form 3 July 2017 Accepted 12 August 2017 Available online 18 August 2017 Keywords: Micro- nano-bubbles (MNBs) Ozone Groundwater remediation Mass transfer Field test
a b s t r a c t Ozone is widely used for water treatment because of its strong oxidation ability. However, the efficiency of ozone in groundwater remediation is limited because of its relatively low solubility and rapid decomposition in the aqueous phase. Methods for increasing the stability of ozone within the subsurface are drawing increasing attention. Micro-nano-bubbles (MNBs), with diameters ranging from tens of nanometres to tens of micrometres, present rapid mass transfer rates, persist for a relatively long time in water, and transport with groundwater flow, which significantly improve gas concentration and provide a continuous gas supply. Therefore, MNBs show a considerable potential for application in groundwater remediation. In this study, the characteristics of ozone MNBs were examined, including their size distribution, bubble quantity, and zeta potential. The mass transfer rate of ozone MNBs was experimentally investigated. Ozone MNBs were then used to treat organics-contaminated water, and they showed remarkable cleanup efficiency. Column tests were also conducted to study the efficiency of ozone MNBs for organics-contaminated groundwater remediation. Based on the laboratory tests, field monitoring was conducted on a trichloroethylene (TCE)-contaminated site. The results showed that ozone MNBs can greatly improve remediation efficiency and represent an innovative technology for in situ remediation of organics-contaminated groundwater. © 2017 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction Soil and groundwater contamination are major environmental problems; thus, numerous technologies have been developed to remediate such contaminants [1]. In situ chemical oxidation is one method commonly used to remediate polluted sites. Oxidants such as Fenton’s reagent, peroxydisulfate, and permanganate show remarkable efficiency in the oxidation of organic contaminants and are generally used for site remediation [2]. However, the efficiency of the Fenton process is strongly dependent on the pH [3]. A considerable amount of oxygen is formed during the Fenton process,
∗ Corresponding author. E-mail address: [email protected] (L. Hu).
which may cause the blockage of pore channels and can limit the area affected by Fenton’s reagent [2]. Peroxydisulfate tends to be relatively stable at ambient temperatures (∼20 ◦ C) and must be activated to be used in site remediation [4]. A large amount of sulfate is produced as a by-product [5], however, which results in secondary contamination. In situ chemical oxidation with permanganate produces MnO2 , which also may result in pore plugging and can lower the remediation efficiency [6]. Ozone is widely used for oxidation of pharmaceuticals in drinking water [7,8]. Because of its strong oxidation ability, ozone also has high potential in the treatment of wastewater [9,10]. Hydrogen peroxide can be used to accelerate the oxidation of contaminants by ozone [11,12]. However, the efficiency of ozone oxidation is limited by the rapid decomposition rate of dissolved ozone in water, which is much faster than that in the gas phase. Methods used to prolong
http://dx.doi.org/10.1016/j.jhazmat.2017.08.030 0304-3894/© 2017 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
the reactivity of aqueous-phase ozone are thus urgently needed, and some stabilisers have been used to increase the stability of ozone in groundwater [13]. Micro-nano-bubbles (MNBs) are tiny bubbles with diameters ranging from tens of nanometres to several tens of micrometres [14,15]. Owing to their small diameter, MNBs present high internal pressures and rapid mass transfer rates, which can significantly improve gas solubility. Compared with normal bubbles, MNBs have lower rising velocity in the liquid phase. Nano-bubbles can persist in water for long periods [16,17]. Those with radii of 150–200 nm have been shown to remain stable for two weeks [18], and clusters of nano-bubbles could further increase their stability [19]. Owing to their long existence in water, MNBs can migrate with the water flow and provide continuous gas supply for the dissolution phase. In our previous work, the properties and mass transfer efficiency of MNBs were studied [20–22]. One remarkable proven characteristic of MNBs is that their large specific surface area leads to a considerable capacity for pollutant adsorption on the bubble surface [23]. In recent years, the potential application of MNBs in environmental engineering has become a research focus [24–27]. Such studies include the use of MNBs in surface water treatment owing to their special characteristics of large specific surface area, negatively charged surface, and high mass transfer efficiency [28–30]. Although the MNB technique has remarkable advantages in environmental cleanup, its application to groundwater remediation has not been systematically investigated thus far [20–22]. The purpose of this study is to investigate the feasibility and efficiency of applying ozone MNBs to groundwater remediation. The physic-chemical characteristics of ozone MNBs, such as the size distribution and zeta potential, were studied experimentally, and the mass transfer behaviour was investigated by model tests. The remediation efficiency for contaminated water and groundwater was examined under laboratory conditions using methyl orange as a representative organic contaminant. A field test on a trichloroethylene (TCE)-contaminated site was also conducted to study the efficiency of in situ remediation by ozone MNBs. 2. Methods and materials 2.1. Experimental facilities 2.1.1. Micro-nano-bubble generator The MNBs used in this research were produced by a spiral liquid flow-type [31] MNB generator (Eco-20, Taikohgiken Ltd., Nishi-ku, Kumamoto, Japan). Water was pumped into the generator, and a maelstrom-like cavity was formed by the high-speed rotation of the liquid flow. Gas was injected into the generator and was reduced to MNBs by the centrifugation effect. The MNB size is affected by the injection rates of water and gas. During the generation of MNBs, a generator was placed inside the water, and a pipe was used to inject gas. In this research, the flow rates of ozone and water were 4 L/min and 270 L/min, respectively. 2.1.2. Millimetre-bubble generation facility A millimetre-bubble generation facility was used to generate millimetre bubbles for ozone mass transfer tests and treatment efficiency tests. Ozone was generated by the ozone generator and was injected into the water by the air compressor through a pipe having a diameter of 7 mm. A gas flow meter was used to control the flow rate of the ozone. 2.1.3. Ozone generators Two ozone generators were used in this research. Both are based on the corona discharge method, and oxygen was transformed into ozone by high-voltage discharge. In this study, the output rate of
447
gas was 4 L/min. The generator (RQ-30, Ruiqing Ltd., Jinan, Shandong, China) which was used to produce ozone for all laboratory tests including size distribution, gas mass transfer, and treatment of methyl orange; the supplied ozone had a concentration of approximately 50 mg/L. As a result, the mass percentages of ozone and oxygen in the supplied gas were 3.5% and 96.5%, respectively. A second generator (S4-R02, Ecodesign Inc., Ogawa, Saitama, Japan), which was used to produce ozone in the field tests, supplied ozone with a concentration of approximately 100 mg/L. The mass percentages of ozone and oxygen in the generated gas were 6.8% and 93.2%, respectively. 2.1.4. Size distribution analyser The size distribution and number of the ozone MNBs were measured by a nanoparticle tracking analyser (NanoSight LM10, Malvern Instruments Ltd., Malvern, Worcestershire, UK). For nanoparticles in liquids, the rate of Brownian motion was not affected by the particle density and was related only to the viscosity and temperature. MNBs in liquid were illuminated by a laser, and the analyser used a charge-coupled device to capture the MNB movement. The nanoparticle size was calculated according to the rate of nanoparticle movement by using the Stokes–Einstein equation. The measurement range for the size distribution was from 10 to 1000 nm, and the measurement range for the number of MNBs was 106 –109 bubbles per mL. 2.1.5. Zeta potential analyser The zeta potential of ozone MNBs in solution was measured by using a zeta potential analyser (Delsa-nano C, Beckman Coulter Inc., Brea, California, US). The interfacial charge characteristics of the ozone MNBs were measured by calculating the electrophoretic mobility. The measurement range of the zeta potential was from −200 mV to +200 mV. 2.1.6. Dissolved ozone monitor The concentration of dissolved ozone in water was measured by a dissolved ozone monitor (Q45H/64, Analytical Technology Inc., Collegeville, Pennsylvania, US). The monitor used a polarographic membrane sensor to accurately determine the concentration of the dissolved ozone. The display range of the monitor was 0–20.00 mg/L, and the accuracy was ±0.1 mg/L. 2.1.7. UV spectrophotometer The concentration of methyl orange was measured by UV spectrophotometry (DR5000, Hach, Loveland, Colorado, US). Based on the Beer-Lambert law, at a wavelength of 462 nm, the intensity of light absorbed by the methyl orange solution was measured to determine the concentration of methyl orange. 2.1.8. Gas chromatograph The concentration of TCE was measured by a gas chromatograph (GC-310C, SRI Instruments, Torrance, California, US). Samples were heated in a water bath, and the TCE in the headspace was measured to determine the TCE concentration in the samples. Each of the groundwater samples was measured at least twice, and the deviation was less than 0.005 mg/L. 2.2. Setup of laboratory tests 2.2.1. Size distribution and zeta potential analysis tests A Perspex tank with internal dimensions of 0.8 m (length) × 0.2 m (height) × 0.2 m (width) was used to perform the experiments. The MNB generator was placed inside 20 L of deionised water to generate ozone MNBs under 20 ◦ C. Ozone MNBs were generated for 30 min, and water samples were then taken to measure the MNB size distribution and quantity over time
448
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Fig. 2. Relative locations of wells.
Fig. 1. Column tests setup.
with the size distribution analyser. The effect of salinity on size distribution and zeta potential was also studied. Sodium chloride was used to adjust the salinity of the deionised water, and the size distribution and zeta potential of ozone MNBs generated under different salinity conditions were measured. For this analysis, filtration equipment was used to remove bubbles larger than 1 m because micro-bubbles significantly affect the measurement of nano-bubbles. 2.2.2. Gas mass transfer tests The gas mass transfer efficiency of bubbles was studied by measuring the dissolved ozone concentration. The MNB generator was placed inside the tank with 20 L of deionised water to generate ozone MNBs. For comparison, the millimetre-bubble generation facility was used to generate millimetre bubbles in 20 L of deionised water. The dissolved ozone concentration was measured during and after the generation process. The ozone MNB tests were divided into four groups, in which ozone MNBs were generated for 8 min, 12 min, 17 min, 30 min, respectively. Millimetre-sized bubbles were generated for 30 min. Each test was conducted twice to ensure the repeatability of the data. 2.2.3. Organics-contaminated water treatment tests The treatment efficiency of ozone MNBs was studied. Methyl orange, a commercial dye which can be degraded by ozone [32], was used as the target contaminant to investigate the treatment efficiency by ozone MNBs. 20 L of methyl orange solution at a concentration of 10 mg/L was treated by ozone MNBs, ozone millimetre bubbles, and oxygen MNBs, respectively. The ozone used in MNB tests and millimetre-bubble tests was controlled to the same amount to study whether MNBs can increase the treatment efficiency by ozone. Because the collapse of MNBs may produce free radicals to react with methyl orange [33], oxygen MNBs were used to show the effects for comparison. Each test was conducted twice to ensure the repeatability of the data. 2.2.4. Column tests Column tests were conducted in a Perspex cylinder (Fig. 1) with an internal diameter of 12 cm to study the treatment of organics-contaminated groundwater by ozone MNBs. Glass beads
were mixed with methyl orange solution at an initial concentration of 10 mg/L to simulate organics-contaminated soil and groundwater. The specific gravity of the glass beads was 2.43, the average diameter was 60 m, and the dry density was 1.43 g/cm3 . The hydraulic conductivity was 1.3 × 10−4 m/s because the glass beads were poorly graded, which represents medium sand. The height of the soil sample was 76 cm, and four sampling ports #1, #2, #3, and #4 were distributed at 16 cm, 32 cm, 48 cm, and 64 cm from the bottom, respectively. Ozone MNBs were generated in deionised water continuously to ensure that the quantity of ozone MNBs was maintained at the peak value. Water containing ozone MNBs was fed upwards to the column from the bottom with a hydraulic gradient i of 0.368; deionised water without MNBs was used for comparison testing. Each test was conducted twice to ensure the repeatability of the data. 2.3. Setup of field tests 2.3.1. Site condition The field tests were conducted at an organics-contaminated site in Niiza, Japan. The contaminated site, which is the former location of an electronic components factory, has an area of 1100 m2 . The main contaminant was TCE, which according to the Japanese standard has a concentration limit of 0.03 mg/L in groundwater. The groundwater at this site is mainly distributed in a confined sandy aquifer at depths of 12 m to 16 m, and the maximum concentration of TCE in the groundwater is approximately 10 mg/L. The hydraulic conductivity of the sandy aquifer is approximately 10−6 m/s, whereas the upper and lower strata of the aquifer are clay layers with a hydraulic conductivity of approximately 10−8 m/s. The groundwater level is 6.5 m below the ground surface. 2.3.2. Extracted groundwater treatment Experiments were conducted onsite to study the effect of H2 O2 on the treatment efficiency of ozone MNBs. 100 L of groundwater was extracted and stored in an airtight reaction tank having a capacity of 500 L. Ozone MNBs, H2 O2 , and ozone MNBs plus H2 O2 were used to treat the TCE-contaminated water. H2 O2 reagent was added during the generation of ozone MNBs at a mass ratio of ozone:H2 O2 of 1:1. The concentration of TCE during the treatment was measured. 2.3.3. Wells One extraction well, one injection well, and five monitoring wells, the locations of which are shown in Fig. 2, were used for in situ remediation. The initial groundwater flow direction was from the injection well to the extraction well. The depth of each well was 16 m, and the screened intervals were from 12 m to 16 m. The diameter of the extraction well was 25 cm, and a pump with an extraction rate of 36 L/min was placed inside the well at a depth of 15 m. The diameter of the injection well and the monitoring wells was 5 cm. The injection well was connected to the injection unit, and the injection rate was 15 L/min. Because the injection rate was limited by site and facility conditions, a high extraction rate was applied to create stronger groundwater flow. The residual water was treated and discharged far from the tested area. Groundwater
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
449
Fig. 3. In situ remediation facility.
samplers were installed in the monitoring wells to collect samples at a depth of 15 m. 2.3.4. In situ remediation An in situ remediation facility (Fig. 3) was developed for the field test. The remediation system contained three parts: the extraction unit, the reaction unit, and the injection unit. Contaminated groundwater was extracted from the site and was filtered to remove soil particles. After filtration, the groundwater was treated by an air sparging facility to remove the TCE. Subsequently, the treated groundwater was injected into the reaction tank. The MNB generator was located inside the tank to generate in situ ozone MNBs. H2 O2 reagent was added during the generation of ozone MNBs at a mass ratio of ozone:H2 O2 of 1:1. After the ozone MNBs and H2 O2 were added, the water was injected back into the ground by the injection unit. The flow rate from reaction tank to the injection unit was 15 L/min, and the residence time was 33.3 min in order to achieve the best MNB generation. In situ groundwater remediation was conducted for six days from 09:00 to 18:00 local time each day. Groundwater was sampled to monitor the TCE concentration.
Fig. 4. Ozone MNBs size distribution.
3. Results and discussions Knowledge of the physic-chemical characteristics of ozone MNBs and gas mass transfer is essential for designing a system for site remediation. 3.1. Size distribution and zeta-potential of ozone MNBs The size and concentration of MNBs are main factors affecting the mass transfer efficiency. Smaller bubbles present larger specific surface area and higher inner pressure, which result in higher mass transfer efficiency. The MNB concentration affects the influencing area of each bubble and therefore affects the concentration of dissolved ozone and the remediation efficiency. Zeta potential describes the charge characteristics of MNBs and affects their dispersion [20]. High absolute value of zeta potential prevents MNBs from coalescence and improves their stability. The size distribution of ozone MNBs generated in deionised water for 30 min is shown in Fig. 4. The diameters of the ozone MNBs mainly ranged from 32 nm to 460 nm, and the Sauter mean diameter was 247 ± 9 nm. The quantity of ozone MNBs was measured continuously after the generation was stopped; the results are shown in Table 1. Despite fluctuation caused by measurement error, the quantity of ozone MNBs showed no significant decrease within 3 h; thus, it is reasonable to conclude that ozone MNBs remain stable in water. Therefore, ozone MNBs can stay relatively stable and can continuously provide dissolved ozone when injected into groundwater. The migration ability of ozone MNBs will be studied further through column tests.
Fig. 5. Zeta-potential of MNBs under various salinity conditions.
The zeta potential and Sauter mean diameter of ozone MNBs under various salinity conditions are shown in Fig. 5 and Table 2, respectively. The salinity showed a slight effect on the zeta potential, and the ozone MNBs remained negatively charged under various salinity conditions; however, the salinity showed no obvious effect on the diameter of ozone MNBs. Thus, the ozone MNBs are stable and can be applied for remediation of groundwater with various salinity levels. 3.2. Gas mass transfer and fate of MNBs MNBs have a high internal pressure and a large specific surface area, which greatly enhances the ozone mass transfer. The gas mass transfer characteristics of ozone MNBs determine the treatment efficiency of organic contaminants in both surface water and groundwater. The dissolved ozone concentration during and after the generation process is shown in Fig. 6. For the group in which ozone MNBs were generated for 30 min, the dissolved ozone concentration reached a peak value of 10.09 ± 0.09 mg/L at 22 min. For the millimetre-bubbles test, millimetre-sized bubbles were generated for 30 min, and a peak value of 0.64 ± 0.00 mg/L was reached at 6 min.
450
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Table 1 Quantity of ozone MNBs. Time after generation was stopped (h)
0
1
2
3
Quantity in 1 mL (107 bubbles)
4.55 ± 0.23
4.63 ± 0.27
4.44 ± 0.27
4.35 ± 0.18
Fig. 6. Mass transfer process for ozone MNBs and millimetre bubbles.
Table 2 Sauter mean diameter of ozone MNBs under various salinity conditions.
Fig. 7. Degradation rates of methyl orange in laboratory tests. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3.3. Contaminated water treatment efficiency by ozone MNBs
Salinity (mg/L)
0
9
18
Sauter mean diameter (nm)
247 ± 9
226 ± 25
265 ± 27
The mass transfer tests results are shown in Table 3. The average increasing rate of dissolved ozone concentration from the beginning of generation to the first achievement of the peak value was used to evaluate the dissolved ozone provision ability. Compared with the millimetre bubbles, the MNBs significantly enhanced the dissolution process of ozone. The high internal pressure and large specific area of MNBs significantly improved the increasing rate. Ozone MNBs can persist in water and can reach high concentration levels, which greatly increase the peak value of the dissolved ozone concentration. Similar results were reported by Li et al. [21], who showed that MNBs significantly increased the mass transfer efficiency of oxygen. After the generation was stopped, the decrease of dissolved ozone obeyed the first-order reaction model, and the half-life time of the dissolved ozone was calculated. The average half-life time for the MNB system was 10.51 min, whereas that for the millimetre-bubble system was 0.70 min. Compared with the millimetre bubbles, MNBs can persist in water for long periods and can provide ozone constantly, thereby significantly prolonging the active time of the ozone.
Methyl orange was treated by ozone MNBs and millimetre bubbles for 30 min to study whether MNBs can increase the treatment efficiency by ozone. The tests were conducted twice, and the deviation was less than 0.03 mg/L. The concentration of methyl orange after the treatment is shown in Table 4, and the degradation rates over time are shown in Fig. 7. Oxygen MNBs were not shown to oxidise methyl orange, which means that free radicals produced through the collapse of MNBs had little effect on methyl orange in this case. After treatment for 30 min, the concentration of methyl orange in ozone MNBs and millimetre-bubble tests decreased to 0.16 mg/L and 9.08 mg/L, respectively. The MNBs greatly enhanced the decomposition efficiency of methyl orange by ozone. At 10 min, 2.25% methyl orange was degraded in the ozone millimetre-bubble test, whereas 92.95% methyl orange was degraded in the ozone MNBs test. The treatment efficiency of the MNBs was 40 times faster than that of the millimetre bubbles. After 10 min of treatment, the degradation rate of methyl orange slowed obviously because its concentration decreased. Similar results were reported by Chen [32], in which the concentration of methyl orange decreased to a minimum of 0.5 mg/L and remained unchanged during further injection of ozone. In the present study, the concentration of methyl orange further decreased to 0.18 mg/L after the 30 min treatment by MNBs. The high concentration of dissolved ozone by the MNBs was speculated to be the main reason. Furthermore, methyl orange is a non-volatile compound. For volatile organic compounds, it can be inferred that the contami-
Table 3 Mass transfer tests results. Bubble type
MNBs
MNBs
MNBs
MNBs
Millimetre bubbles
Generation duration (min) Peak value of dissolved ozone concentration (mg/L) Increasing rate of dissolved ozone concentration (mg/L min) Half-life time of dissolved ozone (min) R2 for dissolved ozone half-life time
30 10.09 ± 0.09 0.46 10.61 0.980
17 8.38 ± 0.07 0.49 10.62 0.978
12 6.95 ± 0.08 0.58 10.70 0.983
8 5.33 ± 0.06 0.67 10.12 0.972
30 0.64 ± 0.00 0.11 0.70 0.980
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
451
Table 4 Concentration of methyl orange after 30 min of treatment. Bubble type
Ozone MNBs
Ozone millimetre bubbles
Oxygen MNBs
Concentration of methyl orange after 30 min treatment (mg/L)
0.18 ± 0.02
9.10 ± 0.02
9.99 ± 0.01
Fig. 8. Concentrations of methyl orange in column tests over time. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 9. Concentrations of TCE during treatment of ozone MNBs, hydrogen peroxide, and ozone MNBs plus hydrogen peroxide.
nants can transfer into MNBs and react with ozone in the gas phase. As a result, the treatment efficiency of ozone MNBs will be further increased. 3.4. Column tests of ozone MNBs In the organics-contaminated water treatment tests, ozone MNBs showed remarkable treatment efficiency. It was reported by Li et al. [20] that water with MNBs presented nearly the same hydraulic conductivity as water without MNBs. Column tests were conducted to study the treatment of organics-contaminated groundwater by ozone MNBs. The concentrations of methyl orange at the sampling ports are shown in Fig. 8. For all the sampling ports, the concentration of methyl orange in the ozone MNBs test decreased significantly faster than that in the deionised water tests. The transport of MNBs in groundwater can be described as colloid transport [21]; it can be inferred that the dispersion of ozone MNBs significantly increased the migration velocity. In contaminated sites, dispersion with groundwater flow can expand the influencing area of ozone MNBs. When the concentration of methyl orange decreased from 10 mg/L to 1 mg/L, ozone MNBs were 64% faster than deionised water; when it decreased from 1 mg/L to 0.2 mg/L, the ozone MNBs were 82% faster. Methyl orange was slightly adsorbed by soil. At low concentrations, the removal of methyl orange by deionised water slowed, whereas ozone MNBs remained more efficient. Ozone MNBs showed strong potential for application in organicscontaminated groundwater remediation. 3.5. Field test of ozone MNBs Based on the laboratory tests, field tests were conducted at a TCE-contaminated site to study the remediation ability of ozone MNBs in groundwater. Ozone is a practical decontamination method for TCE, and hydrogen peroxide can significantly accelerate the oxidation [11].
Fig. 10. TCE concentrations in monitoring wells over time.
Experiments were conducted to study the effect of H2 O2 on the treatment of TCE by ozone. The concentration of TCE during the treatment is shown in Fig. 9. Hydrogen peroxide alone showed low treatment efficiency for TCE, although it improved the treatment efficiency by ozone MNBs by a factor of approximately 3. Therefore, the combination of ozone and H2 O2 was used for the field tests. In situ remediation of organics-contaminated groundwater was conducted; the initial and final TCE concentrations in the groundwater are shown in Table 5. During the in situ remediation by MNBs, the TCE concentration in the monitoring wells showed remarkable decrease; an overall removal rate of 99% was reached. The final TCE concentration in the entire treated area decreased to less than 0.03 mg/L, which is the standard limit in Japan. The concentration of TCE over time is shown in Fig. 10. Ozone MNBs migrated with the groundwater for a constant supply of dissolved ozone. Previous research indicated that an ozone–peroxide system forms OH radicals, which presents high treatment efficiency on TCE; moreover, TCE can be treated also
452
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Table 5 Initial and final TCE concentrations in groundwater. Well Number
#1
#2
#3
#4
#5
Initial TCE concentration (mg/L) Final TCE concentration (mg/L)
3.529 ± 0.002 0.010 ± 0.004
4.502 ± 0.005 0.003 ± 0.001
10.130 ± 0.004 0.018 ± 0.014
0.264 ± 0.002 0.007 ± 0.002
2.129 ± 0.004 0.003 ± 0.000
by direct reaction with ozone [11]. The reaction between ozone plus H2 O2 and TCE was described by pseudo-first-order kinetics, in which the decreasing rate of TCE was related to the concentrations of ozone and H2 O2 [11]. For monitoring wells #1, #2, #3, and #5, 90% of TCE was removed during the first three or four days, and the decreasing rate from high to low occurred in #3, #2, #5, and #1. Well #3 was in closest proximity to the injection well and presented the highest concentration of oxidants, which resulted in the fastest decrease of TCE. In the field tests, the groundwater flow was affected by the injection and extraction wells. Well #2, located along the injection-extraction direction, with groundwater presented a higher flow rate; thus, the concentration of oxidants at #2 was higher than at #5. Well #1, located farthest from the injection well, presented the lowest oxidants concentration and resulted in the slowest decreasing rate of TCE. The initial concentration of monitoring well #4 was 0.26 mg/L, which is lower than that in other areas. However, the concentration of TCE showed apparent decreases during the treatment, which indicates that ozone MNBs are still effective for treating groundwater with relatively low concentrations of contaminants. During the first three or four days, the decreasing rate of TCE increased over time, which indicates that the concentration of oxidants in the site increased during the remediation. According to the results of extracted groundwater treatment, the degradation of TCE was attributed mainly to ozone. It can be concluded that ozone MNBs are stable in groundwater and that the constant injection increased the concentration of ozone MNBs at the site. When the concentration of TCE was lower than 0.01 mg/L, its degradation rate obviously slowed. Similar results were obtained in the treatment of methyl orange; thus, it can be assumed that the low concentration of organics-contaminant was the main reason for the limited remediation efficiency. 4. Conclusion In this study, the characteristics of ozone MNBs were studied. MNBs can reach a high unit quantity and remain stable in water for long periods. Ozone MNBs stay negatively charged under varied salinities, which means that they are stable and can be applied in the remediation of groundwater with varying salinities. MNBs significantly increase the mass transfer efficiency of ozone and can remain stable in water to constantly supply ozone. The half-life of ozone in the MNB system is significantly longer than that in the millimetre-bubble system. Under laboratory conditions, ozone had a significant effect on organics-contaminants in the surface water and groundwater, and the MNBs greatly enhanced the treatment efficiency. An in situ remediation facility was developed, and field tests were performed in a TCE-contaminated site in Japan. An overall removal rate of 99% was reached after six days of treatment. Ozone MNBs showed a sound effect on the remediation of TCE-contaminated groundwater and potentially represent an innovative technology for in situ remediation of organics-contaminated groundwater. Novelty statement As a widely used oxidant, the application of ozone in groundwater remediation is limited by its migration ability. The purpose
of this manuscript is to study the feasibility and efficiency of applying ozone to groundwater remediation in the form of micronano bubbles (MNBs). Basic characteristics and mass transfer behaviour of ozone MNBs were studied, and the remediation efficiency of organics-contaminants by ozone MNBs were studied through laboratory and field tests. Ozone MNBs represent an innovative technology for in-situ remediation of organics-contaminated groundwater, and this manuscript will surely attract attention of geo-environmental engineers interested in technologies for organics-contaminated groundwater remediation.
Acknowledgements The financial support from National Natural Science Foundation of China (Project No. 41372352, 51323014, 51661165015) and Tsinghua University (Project No. 2015THZ02-2-20161080101, 2015THZ01-1-20161080079) are gratefully acknowledged. The support for the testing facility from State Key Laboratory of Hydro Science and Engineering (SKLHSE-2016-d-03) is also acknowledged. The authors thank the assistance of Dr. D Song at IS Solution Company for field monitoring on MNB remediation. The two anonymous reviewers are also thanked for giving valuable comments that improved the overall quality of this paper.
References [1] How to Evaluate Alternative Cleanup Technologies for Underground Storage Tank Sites, A Guide for Corrective Action Plan Reviewers, United States Environmental Protection Agency, Washington, DC, 2004. [2] S.G. Huling Pivetz, In-situ Chemical Oxidation. No. EPA/600/R-06/072, Environmental Protection Agency Washington DC Office of Water, 2006. [3] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced oxidation technique, J. Hazard. Mater. 98 (2003) 33–50. [4] L. Hou, H. Zhang, X. Xue, Ultrasound enhanced heterogeneous activation of peroxydisulfate by magnetite catalyst for the degradation of tetracycline in water, Sep. Purif. Technol. 84 (2012) 147–152. [5] D.A. House, Kinetics and mechanism of oxidations by peroxydisulfate, Chem. Rev. 62 (1962) 185–203. [6] L.X. David, F.W. Schwartz, Using phosphate to control the Mn oxide precipitation during in situ chemical oxidation of chlorinated ethylenes by permanganate, In ACS symposium series. Oxford University Press 904 (2005) 82–95. [7] M.M. Huber, S. Canonica, G.Y. Park, U. von Gunten, Oxidation of pharmaceuticals during ozonation and advanced oxidation processes, Environ. Sci. Technol 37 (2003) 1016–1024. [8] T.A. Ternes, M. Meisenheimer, D. McDowell, F. Sacher, H.J. Brauch, B. Haist-Gulde, G. Preuss, U. Wilme, N. Zullei-Seibert, Removal of pharmaceuticals during drinking water treatment, Environ. Sci. Technol. 36 (2002) 3855–3863. [9] M.M. Mitani, A.A. Keller, C.A. Bunton, C.A. Bunton, R.G. Rinker, O.C. Sandall, Kinetics and products of reactions of MTBE with ozone and ozone/hydrogen peroxide in water, J. Hazard. Mater. 89 (2002) 197–212. [10] H. Barndõk, L. Cortijo, D. Hermosilla, C. Negro, A. Blanco, Removal of 1: 4-dioxane from industrial wastewaters: routes of decomposition under different operational conditions to determine the ozone oxidation capacity, J. Hazard. Mater. 280 (2014) 340–347. [11] W.H. Glaze, J.W. Kang, Advanced oxidation processes for treating groundwater contaminated with TCE and PCE: laboratory studies, J. (Am. Water Works Assoc.) (1988) 57–63. [12] C. Tizaoui, L. Bouselmi, L. Mansouri, A. Ghrabi, Landfill leachate treatment with ozone and ozone/hydrogen peroxide systems, J. Hazard. Mater. 140 (2007) 316–324. [13] A. Dettmer, R. Ball, T.B. Boving, N.A. Khan, T. Schaub, N. Sudasinghe, C.A. Fernandez, K.C. Carroll, Stabilization and prolonged reactivity of aqueous-phase ozone with cyclodextrin, J. Contam. Hydrol. 196 (2017) 1–9. [14] P. Attard, M.P. Moody, J.W.G. Tyrrell Nanobubbles, The big picture, Physica. A: Stat. Mech. Appl. 314 (2002) 696–705.
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453 [15] L.B. Chu, X.H. Xing, A.F. Yu, Y.N. Zhou, X.L. Sun, B. Jurcik, Enhanced ozonation of simulated dyestuff wastewater by microbubbles, Chemosphere 68 (2007) 1854–1860. [16] J.R.T. Seddon, D. Lohse, W.A. Ducker, V.S.J. Craig, A deliberation on nanobubbles at surfaces and in bulk, ChemPhysChem 13 (2012) 2179–2187. [17] F.Y. Ushikubo, T. Furukawa, R. Nakagawa, M. Enari, Y. Makino, Y. Kawagoe, T. Shiina, S. Oshita, Evidence of the existence and the stability of nano-bubbles in water, Coll. Surf. A 361 (2010) 31–37. [18] A. Azevedo, R. Etchepare, S. Calgaroto, J. Rubio, Aqueous dispersions of nanobubbles: generation, properties and features, Min. Eng. 94 (2016) 29–37. [19] J.H. Weijs, J.R.T. Seddon, D. Lohse, Diffusive shielding stabilizes bulk nanobubble clusters, ChemPhysChem 13 (2012) 2197–2204. [20] H. Li, L. Hu, Z. Xia, Impact of groundwater salinity on bioremediation enhanced by micro-nano bubbles, Materials 6 (2013) 3676–3687. [21] H. Li, L. Hu, D. Song, A. Al-Tabbaa, Subsurface transport behavior of micro-Nano bubbles and potential applications for groundwater remediation, Int. J. Environ. Res. Public Health. 11 (2014) 473–486. [22] H. Li, L. Hu, D. Song, F. Lin, Characteristics of micro-nano bubbles and potential application in groundwater bioremediation, Water Environ. Res. 86 (2014) 844–851. [23] A. Yoshida, O. Takahashi, Y. Ishii, Y. Sekimoto, Y. Kurata, Water purification using the adsorption characteristics of microbubbles, Jpn. J. Appl. Phys. 47 (2008). [24] L. Hu, X. Wu, Y. Liu, J.N. Meegoda, S. Gao, Physical modeling of air flow during air sparging remediation, Environ. Sci. Technol. 44 (2010) 3883–3888. [25] J.N. Meegoda, L. Hu, A review of centrifugal testing of gasoline contamination and remediation, Int. J. Environ. Res. Public Health 8 (2011) 3496–3513.
453
[26] L. Hu, J.N. Meegoda, J. Du, S. Gao, X. Wu, Centrifugal study of zone of influence during air-sparging, J. Environ. Monit. 13 (2011) 2443–2449. [27] D. Song, A. Katayama, Approach for estimating microbial growth and the biodegradation of hydrocarbon contaminants in subsoil based on field measurements: 1. Model development and verification, Environ. Sci. Tech. 44 (2009) 767–773. [28] M. Takahashi, Base and technological application of micro-bubble and nano-bubble, Mater. Integr. 22 (2009) 2–19. [29] A. Agarwal, W.J. Ng, Y. Liu, Principle and applications of microbubble and nanobubble technology for water treatment, Chemosphere 84 (2011) 1175–1180. [30] L.B. Chu, X.H. Xing, A.F. Yu, Y.N. Zhou, X.L. Sun, B. Jurcik, Enhanced ozonation of simulated dyestuff wastewater by microbubbles, Chemosphere 68 (2007) 1854–1860. [31] K. Terasaka, A. Hirabayashi, T. Nishino, S. Fujioka, D. Kobayashi, Development of microbubble aerator for waste water treatment using aerobic activated sludge, Chem. Eng. Sci. 66 (2011) 3172–3179. [32] L.C. Chen, Effects of factors and interacted factors on the optimal decolorization process of methyl orange by ozone, Water Res. 34 (2000) 974–982. [33] P. Li, M. Takahashi, K. Chiba, Degradation of phenol by the collapse of microbubbles, Chemosphere 75 (2009) 1371–1375.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Application of ozone micro-nano-bubbles to groundwater remediation Liming Hu ∗ , Zhiran Xia State Key Laboratory of Hydro-Science and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, China
h i g h l i g h t s • • • •
This paper presents the high efficiency of gas supply and mass transfer in water using MNBs. The ozone MNBs technique was applied for in situ groundwater remediation of an organics-contaminated site. Ozone MNBs show considerable advantages in contaminant cleanup and time efficiency. Ozone MNBs potentially represent an innovative technology for in situ remediation of organics-contaminated groundwater.
a r t i c l e
i n f o
Article history: Received 25 February 2017 Received in revised form 3 July 2017 Accepted 12 August 2017 Available online 18 August 2017 Keywords: Micro- nano-bubbles (MNBs) Ozone Groundwater remediation Mass transfer Field test
a b s t r a c t Ozone is widely used for water treatment because of its strong oxidation ability. However, the efficiency of ozone in groundwater remediation is limited because of its relatively low solubility and rapid decomposition in the aqueous phase. Methods for increasing the stability of ozone within the subsurface are drawing increasing attention. Micro-nano-bubbles (MNBs), with diameters ranging from tens of nanometres to tens of micrometres, present rapid mass transfer rates, persist for a relatively long time in water, and transport with groundwater flow, which significantly improve gas concentration and provide a continuous gas supply. Therefore, MNBs show a considerable potential for application in groundwater remediation. In this study, the characteristics of ozone MNBs were examined, including their size distribution, bubble quantity, and zeta potential. The mass transfer rate of ozone MNBs was experimentally investigated. Ozone MNBs were then used to treat organics-contaminated water, and they showed remarkable cleanup efficiency. Column tests were also conducted to study the efficiency of ozone MNBs for organics-contaminated groundwater remediation. Based on the laboratory tests, field monitoring was conducted on a trichloroethylene (TCE)-contaminated site. The results showed that ozone MNBs can greatly improve remediation efficiency and represent an innovative technology for in situ remediation of organics-contaminated groundwater. © 2017 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction Soil and groundwater contamination are major environmental problems; thus, numerous technologies have been developed to remediate such contaminants [1]. In situ chemical oxidation is one method commonly used to remediate polluted sites. Oxidants such as Fenton’s reagent, peroxydisulfate, and permanganate show remarkable efficiency in the oxidation of organic contaminants and are generally used for site remediation [2]. However, the efficiency of the Fenton process is strongly dependent on the pH [3]. A considerable amount of oxygen is formed during the Fenton process,
∗ Corresponding author. E-mail address: [email protected] (L. Hu).
which may cause the blockage of pore channels and can limit the area affected by Fenton’s reagent [2]. Peroxydisulfate tends to be relatively stable at ambient temperatures (∼20 ◦ C) and must be activated to be used in site remediation [4]. A large amount of sulfate is produced as a by-product [5], however, which results in secondary contamination. In situ chemical oxidation with permanganate produces MnO2 , which also may result in pore plugging and can lower the remediation efficiency [6]. Ozone is widely used for oxidation of pharmaceuticals in drinking water [7,8]. Because of its strong oxidation ability, ozone also has high potential in the treatment of wastewater [9,10]. Hydrogen peroxide can be used to accelerate the oxidation of contaminants by ozone [11,12]. However, the efficiency of ozone oxidation is limited by the rapid decomposition rate of dissolved ozone in water, which is much faster than that in the gas phase. Methods used to prolong
http://dx.doi.org/10.1016/j.jhazmat.2017.08.030 0304-3894/© 2017 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
the reactivity of aqueous-phase ozone are thus urgently needed, and some stabilisers have been used to increase the stability of ozone in groundwater [13]. Micro-nano-bubbles (MNBs) are tiny bubbles with diameters ranging from tens of nanometres to several tens of micrometres [14,15]. Owing to their small diameter, MNBs present high internal pressures and rapid mass transfer rates, which can significantly improve gas solubility. Compared with normal bubbles, MNBs have lower rising velocity in the liquid phase. Nano-bubbles can persist in water for long periods [16,17]. Those with radii of 150–200 nm have been shown to remain stable for two weeks [18], and clusters of nano-bubbles could further increase their stability [19]. Owing to their long existence in water, MNBs can migrate with the water flow and provide continuous gas supply for the dissolution phase. In our previous work, the properties and mass transfer efficiency of MNBs were studied [20–22]. One remarkable proven characteristic of MNBs is that their large specific surface area leads to a considerable capacity for pollutant adsorption on the bubble surface [23]. In recent years, the potential application of MNBs in environmental engineering has become a research focus [24–27]. Such studies include the use of MNBs in surface water treatment owing to their special characteristics of large specific surface area, negatively charged surface, and high mass transfer efficiency [28–30]. Although the MNB technique has remarkable advantages in environmental cleanup, its application to groundwater remediation has not been systematically investigated thus far [20–22]. The purpose of this study is to investigate the feasibility and efficiency of applying ozone MNBs to groundwater remediation. The physic-chemical characteristics of ozone MNBs, such as the size distribution and zeta potential, were studied experimentally, and the mass transfer behaviour was investigated by model tests. The remediation efficiency for contaminated water and groundwater was examined under laboratory conditions using methyl orange as a representative organic contaminant. A field test on a trichloroethylene (TCE)-contaminated site was also conducted to study the efficiency of in situ remediation by ozone MNBs. 2. Methods and materials 2.1. Experimental facilities 2.1.1. Micro-nano-bubble generator The MNBs used in this research were produced by a spiral liquid flow-type [31] MNB generator (Eco-20, Taikohgiken Ltd., Nishi-ku, Kumamoto, Japan). Water was pumped into the generator, and a maelstrom-like cavity was formed by the high-speed rotation of the liquid flow. Gas was injected into the generator and was reduced to MNBs by the centrifugation effect. The MNB size is affected by the injection rates of water and gas. During the generation of MNBs, a generator was placed inside the water, and a pipe was used to inject gas. In this research, the flow rates of ozone and water were 4 L/min and 270 L/min, respectively. 2.1.2. Millimetre-bubble generation facility A millimetre-bubble generation facility was used to generate millimetre bubbles for ozone mass transfer tests and treatment efficiency tests. Ozone was generated by the ozone generator and was injected into the water by the air compressor through a pipe having a diameter of 7 mm. A gas flow meter was used to control the flow rate of the ozone. 2.1.3. Ozone generators Two ozone generators were used in this research. Both are based on the corona discharge method, and oxygen was transformed into ozone by high-voltage discharge. In this study, the output rate of
447
gas was 4 L/min. The generator (RQ-30, Ruiqing Ltd., Jinan, Shandong, China) which was used to produce ozone for all laboratory tests including size distribution, gas mass transfer, and treatment of methyl orange; the supplied ozone had a concentration of approximately 50 mg/L. As a result, the mass percentages of ozone and oxygen in the supplied gas were 3.5% and 96.5%, respectively. A second generator (S4-R02, Ecodesign Inc., Ogawa, Saitama, Japan), which was used to produce ozone in the field tests, supplied ozone with a concentration of approximately 100 mg/L. The mass percentages of ozone and oxygen in the generated gas were 6.8% and 93.2%, respectively. 2.1.4. Size distribution analyser The size distribution and number of the ozone MNBs were measured by a nanoparticle tracking analyser (NanoSight LM10, Malvern Instruments Ltd., Malvern, Worcestershire, UK). For nanoparticles in liquids, the rate of Brownian motion was not affected by the particle density and was related only to the viscosity and temperature. MNBs in liquid were illuminated by a laser, and the analyser used a charge-coupled device to capture the MNB movement. The nanoparticle size was calculated according to the rate of nanoparticle movement by using the Stokes–Einstein equation. The measurement range for the size distribution was from 10 to 1000 nm, and the measurement range for the number of MNBs was 106 –109 bubbles per mL. 2.1.5. Zeta potential analyser The zeta potential of ozone MNBs in solution was measured by using a zeta potential analyser (Delsa-nano C, Beckman Coulter Inc., Brea, California, US). The interfacial charge characteristics of the ozone MNBs were measured by calculating the electrophoretic mobility. The measurement range of the zeta potential was from −200 mV to +200 mV. 2.1.6. Dissolved ozone monitor The concentration of dissolved ozone in water was measured by a dissolved ozone monitor (Q45H/64, Analytical Technology Inc., Collegeville, Pennsylvania, US). The monitor used a polarographic membrane sensor to accurately determine the concentration of the dissolved ozone. The display range of the monitor was 0–20.00 mg/L, and the accuracy was ±0.1 mg/L. 2.1.7. UV spectrophotometer The concentration of methyl orange was measured by UV spectrophotometry (DR5000, Hach, Loveland, Colorado, US). Based on the Beer-Lambert law, at a wavelength of 462 nm, the intensity of light absorbed by the methyl orange solution was measured to determine the concentration of methyl orange. 2.1.8. Gas chromatograph The concentration of TCE was measured by a gas chromatograph (GC-310C, SRI Instruments, Torrance, California, US). Samples were heated in a water bath, and the TCE in the headspace was measured to determine the TCE concentration in the samples. Each of the groundwater samples was measured at least twice, and the deviation was less than 0.005 mg/L. 2.2. Setup of laboratory tests 2.2.1. Size distribution and zeta potential analysis tests A Perspex tank with internal dimensions of 0.8 m (length) × 0.2 m (height) × 0.2 m (width) was used to perform the experiments. The MNB generator was placed inside 20 L of deionised water to generate ozone MNBs under 20 ◦ C. Ozone MNBs were generated for 30 min, and water samples were then taken to measure the MNB size distribution and quantity over time
448
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Fig. 2. Relative locations of wells.
Fig. 1. Column tests setup.
with the size distribution analyser. The effect of salinity on size distribution and zeta potential was also studied. Sodium chloride was used to adjust the salinity of the deionised water, and the size distribution and zeta potential of ozone MNBs generated under different salinity conditions were measured. For this analysis, filtration equipment was used to remove bubbles larger than 1 m because micro-bubbles significantly affect the measurement of nano-bubbles. 2.2.2. Gas mass transfer tests The gas mass transfer efficiency of bubbles was studied by measuring the dissolved ozone concentration. The MNB generator was placed inside the tank with 20 L of deionised water to generate ozone MNBs. For comparison, the millimetre-bubble generation facility was used to generate millimetre bubbles in 20 L of deionised water. The dissolved ozone concentration was measured during and after the generation process. The ozone MNB tests were divided into four groups, in which ozone MNBs were generated for 8 min, 12 min, 17 min, 30 min, respectively. Millimetre-sized bubbles were generated for 30 min. Each test was conducted twice to ensure the repeatability of the data. 2.2.3. Organics-contaminated water treatment tests The treatment efficiency of ozone MNBs was studied. Methyl orange, a commercial dye which can be degraded by ozone [32], was used as the target contaminant to investigate the treatment efficiency by ozone MNBs. 20 L of methyl orange solution at a concentration of 10 mg/L was treated by ozone MNBs, ozone millimetre bubbles, and oxygen MNBs, respectively. The ozone used in MNB tests and millimetre-bubble tests was controlled to the same amount to study whether MNBs can increase the treatment efficiency by ozone. Because the collapse of MNBs may produce free radicals to react with methyl orange [33], oxygen MNBs were used to show the effects for comparison. Each test was conducted twice to ensure the repeatability of the data. 2.2.4. Column tests Column tests were conducted in a Perspex cylinder (Fig. 1) with an internal diameter of 12 cm to study the treatment of organics-contaminated groundwater by ozone MNBs. Glass beads
were mixed with methyl orange solution at an initial concentration of 10 mg/L to simulate organics-contaminated soil and groundwater. The specific gravity of the glass beads was 2.43, the average diameter was 60 m, and the dry density was 1.43 g/cm3 . The hydraulic conductivity was 1.3 × 10−4 m/s because the glass beads were poorly graded, which represents medium sand. The height of the soil sample was 76 cm, and four sampling ports #1, #2, #3, and #4 were distributed at 16 cm, 32 cm, 48 cm, and 64 cm from the bottom, respectively. Ozone MNBs were generated in deionised water continuously to ensure that the quantity of ozone MNBs was maintained at the peak value. Water containing ozone MNBs was fed upwards to the column from the bottom with a hydraulic gradient i of 0.368; deionised water without MNBs was used for comparison testing. Each test was conducted twice to ensure the repeatability of the data. 2.3. Setup of field tests 2.3.1. Site condition The field tests were conducted at an organics-contaminated site in Niiza, Japan. The contaminated site, which is the former location of an electronic components factory, has an area of 1100 m2 . The main contaminant was TCE, which according to the Japanese standard has a concentration limit of 0.03 mg/L in groundwater. The groundwater at this site is mainly distributed in a confined sandy aquifer at depths of 12 m to 16 m, and the maximum concentration of TCE in the groundwater is approximately 10 mg/L. The hydraulic conductivity of the sandy aquifer is approximately 10−6 m/s, whereas the upper and lower strata of the aquifer are clay layers with a hydraulic conductivity of approximately 10−8 m/s. The groundwater level is 6.5 m below the ground surface. 2.3.2. Extracted groundwater treatment Experiments were conducted onsite to study the effect of H2 O2 on the treatment efficiency of ozone MNBs. 100 L of groundwater was extracted and stored in an airtight reaction tank having a capacity of 500 L. Ozone MNBs, H2 O2 , and ozone MNBs plus H2 O2 were used to treat the TCE-contaminated water. H2 O2 reagent was added during the generation of ozone MNBs at a mass ratio of ozone:H2 O2 of 1:1. The concentration of TCE during the treatment was measured. 2.3.3. Wells One extraction well, one injection well, and five monitoring wells, the locations of which are shown in Fig. 2, were used for in situ remediation. The initial groundwater flow direction was from the injection well to the extraction well. The depth of each well was 16 m, and the screened intervals were from 12 m to 16 m. The diameter of the extraction well was 25 cm, and a pump with an extraction rate of 36 L/min was placed inside the well at a depth of 15 m. The diameter of the injection well and the monitoring wells was 5 cm. The injection well was connected to the injection unit, and the injection rate was 15 L/min. Because the injection rate was limited by site and facility conditions, a high extraction rate was applied to create stronger groundwater flow. The residual water was treated and discharged far from the tested area. Groundwater
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
449
Fig. 3. In situ remediation facility.
samplers were installed in the monitoring wells to collect samples at a depth of 15 m. 2.3.4. In situ remediation An in situ remediation facility (Fig. 3) was developed for the field test. The remediation system contained three parts: the extraction unit, the reaction unit, and the injection unit. Contaminated groundwater was extracted from the site and was filtered to remove soil particles. After filtration, the groundwater was treated by an air sparging facility to remove the TCE. Subsequently, the treated groundwater was injected into the reaction tank. The MNB generator was located inside the tank to generate in situ ozone MNBs. H2 O2 reagent was added during the generation of ozone MNBs at a mass ratio of ozone:H2 O2 of 1:1. After the ozone MNBs and H2 O2 were added, the water was injected back into the ground by the injection unit. The flow rate from reaction tank to the injection unit was 15 L/min, and the residence time was 33.3 min in order to achieve the best MNB generation. In situ groundwater remediation was conducted for six days from 09:00 to 18:00 local time each day. Groundwater was sampled to monitor the TCE concentration.
Fig. 4. Ozone MNBs size distribution.
3. Results and discussions Knowledge of the physic-chemical characteristics of ozone MNBs and gas mass transfer is essential for designing a system for site remediation. 3.1. Size distribution and zeta-potential of ozone MNBs The size and concentration of MNBs are main factors affecting the mass transfer efficiency. Smaller bubbles present larger specific surface area and higher inner pressure, which result in higher mass transfer efficiency. The MNB concentration affects the influencing area of each bubble and therefore affects the concentration of dissolved ozone and the remediation efficiency. Zeta potential describes the charge characteristics of MNBs and affects their dispersion [20]. High absolute value of zeta potential prevents MNBs from coalescence and improves their stability. The size distribution of ozone MNBs generated in deionised water for 30 min is shown in Fig. 4. The diameters of the ozone MNBs mainly ranged from 32 nm to 460 nm, and the Sauter mean diameter was 247 ± 9 nm. The quantity of ozone MNBs was measured continuously after the generation was stopped; the results are shown in Table 1. Despite fluctuation caused by measurement error, the quantity of ozone MNBs showed no significant decrease within 3 h; thus, it is reasonable to conclude that ozone MNBs remain stable in water. Therefore, ozone MNBs can stay relatively stable and can continuously provide dissolved ozone when injected into groundwater. The migration ability of ozone MNBs will be studied further through column tests.
Fig. 5. Zeta-potential of MNBs under various salinity conditions.
The zeta potential and Sauter mean diameter of ozone MNBs under various salinity conditions are shown in Fig. 5 and Table 2, respectively. The salinity showed a slight effect on the zeta potential, and the ozone MNBs remained negatively charged under various salinity conditions; however, the salinity showed no obvious effect on the diameter of ozone MNBs. Thus, the ozone MNBs are stable and can be applied for remediation of groundwater with various salinity levels. 3.2. Gas mass transfer and fate of MNBs MNBs have a high internal pressure and a large specific surface area, which greatly enhances the ozone mass transfer. The gas mass transfer characteristics of ozone MNBs determine the treatment efficiency of organic contaminants in both surface water and groundwater. The dissolved ozone concentration during and after the generation process is shown in Fig. 6. For the group in which ozone MNBs were generated for 30 min, the dissolved ozone concentration reached a peak value of 10.09 ± 0.09 mg/L at 22 min. For the millimetre-bubbles test, millimetre-sized bubbles were generated for 30 min, and a peak value of 0.64 ± 0.00 mg/L was reached at 6 min.
450
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Table 1 Quantity of ozone MNBs. Time after generation was stopped (h)
0
1
2
3
Quantity in 1 mL (107 bubbles)
4.55 ± 0.23
4.63 ± 0.27
4.44 ± 0.27
4.35 ± 0.18
Fig. 6. Mass transfer process for ozone MNBs and millimetre bubbles.
Table 2 Sauter mean diameter of ozone MNBs under various salinity conditions.
Fig. 7. Degradation rates of methyl orange in laboratory tests. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3.3. Contaminated water treatment efficiency by ozone MNBs
Salinity (mg/L)
0
9
18
Sauter mean diameter (nm)
247 ± 9
226 ± 25
265 ± 27
The mass transfer tests results are shown in Table 3. The average increasing rate of dissolved ozone concentration from the beginning of generation to the first achievement of the peak value was used to evaluate the dissolved ozone provision ability. Compared with the millimetre bubbles, the MNBs significantly enhanced the dissolution process of ozone. The high internal pressure and large specific area of MNBs significantly improved the increasing rate. Ozone MNBs can persist in water and can reach high concentration levels, which greatly increase the peak value of the dissolved ozone concentration. Similar results were reported by Li et al. [21], who showed that MNBs significantly increased the mass transfer efficiency of oxygen. After the generation was stopped, the decrease of dissolved ozone obeyed the first-order reaction model, and the half-life time of the dissolved ozone was calculated. The average half-life time for the MNB system was 10.51 min, whereas that for the millimetre-bubble system was 0.70 min. Compared with the millimetre bubbles, MNBs can persist in water for long periods and can provide ozone constantly, thereby significantly prolonging the active time of the ozone.
Methyl orange was treated by ozone MNBs and millimetre bubbles for 30 min to study whether MNBs can increase the treatment efficiency by ozone. The tests were conducted twice, and the deviation was less than 0.03 mg/L. The concentration of methyl orange after the treatment is shown in Table 4, and the degradation rates over time are shown in Fig. 7. Oxygen MNBs were not shown to oxidise methyl orange, which means that free radicals produced through the collapse of MNBs had little effect on methyl orange in this case. After treatment for 30 min, the concentration of methyl orange in ozone MNBs and millimetre-bubble tests decreased to 0.16 mg/L and 9.08 mg/L, respectively. The MNBs greatly enhanced the decomposition efficiency of methyl orange by ozone. At 10 min, 2.25% methyl orange was degraded in the ozone millimetre-bubble test, whereas 92.95% methyl orange was degraded in the ozone MNBs test. The treatment efficiency of the MNBs was 40 times faster than that of the millimetre bubbles. After 10 min of treatment, the degradation rate of methyl orange slowed obviously because its concentration decreased. Similar results were reported by Chen [32], in which the concentration of methyl orange decreased to a minimum of 0.5 mg/L and remained unchanged during further injection of ozone. In the present study, the concentration of methyl orange further decreased to 0.18 mg/L after the 30 min treatment by MNBs. The high concentration of dissolved ozone by the MNBs was speculated to be the main reason. Furthermore, methyl orange is a non-volatile compound. For volatile organic compounds, it can be inferred that the contami-
Table 3 Mass transfer tests results. Bubble type
MNBs
MNBs
MNBs
MNBs
Millimetre bubbles
Generation duration (min) Peak value of dissolved ozone concentration (mg/L) Increasing rate of dissolved ozone concentration (mg/L min) Half-life time of dissolved ozone (min) R2 for dissolved ozone half-life time
30 10.09 ± 0.09 0.46 10.61 0.980
17 8.38 ± 0.07 0.49 10.62 0.978
12 6.95 ± 0.08 0.58 10.70 0.983
8 5.33 ± 0.06 0.67 10.12 0.972
30 0.64 ± 0.00 0.11 0.70 0.980
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
451
Table 4 Concentration of methyl orange after 30 min of treatment. Bubble type
Ozone MNBs
Ozone millimetre bubbles
Oxygen MNBs
Concentration of methyl orange after 30 min treatment (mg/L)
0.18 ± 0.02
9.10 ± 0.02
9.99 ± 0.01
Fig. 8. Concentrations of methyl orange in column tests over time. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 9. Concentrations of TCE during treatment of ozone MNBs, hydrogen peroxide, and ozone MNBs plus hydrogen peroxide.
nants can transfer into MNBs and react with ozone in the gas phase. As a result, the treatment efficiency of ozone MNBs will be further increased. 3.4. Column tests of ozone MNBs In the organics-contaminated water treatment tests, ozone MNBs showed remarkable treatment efficiency. It was reported by Li et al. [20] that water with MNBs presented nearly the same hydraulic conductivity as water without MNBs. Column tests were conducted to study the treatment of organics-contaminated groundwater by ozone MNBs. The concentrations of methyl orange at the sampling ports are shown in Fig. 8. For all the sampling ports, the concentration of methyl orange in the ozone MNBs test decreased significantly faster than that in the deionised water tests. The transport of MNBs in groundwater can be described as colloid transport [21]; it can be inferred that the dispersion of ozone MNBs significantly increased the migration velocity. In contaminated sites, dispersion with groundwater flow can expand the influencing area of ozone MNBs. When the concentration of methyl orange decreased from 10 mg/L to 1 mg/L, ozone MNBs were 64% faster than deionised water; when it decreased from 1 mg/L to 0.2 mg/L, the ozone MNBs were 82% faster. Methyl orange was slightly adsorbed by soil. At low concentrations, the removal of methyl orange by deionised water slowed, whereas ozone MNBs remained more efficient. Ozone MNBs showed strong potential for application in organicscontaminated groundwater remediation. 3.5. Field test of ozone MNBs Based on the laboratory tests, field tests were conducted at a TCE-contaminated site to study the remediation ability of ozone MNBs in groundwater. Ozone is a practical decontamination method for TCE, and hydrogen peroxide can significantly accelerate the oxidation [11].
Fig. 10. TCE concentrations in monitoring wells over time.
Experiments were conducted to study the effect of H2 O2 on the treatment of TCE by ozone. The concentration of TCE during the treatment is shown in Fig. 9. Hydrogen peroxide alone showed low treatment efficiency for TCE, although it improved the treatment efficiency by ozone MNBs by a factor of approximately 3. Therefore, the combination of ozone and H2 O2 was used for the field tests. In situ remediation of organics-contaminated groundwater was conducted; the initial and final TCE concentrations in the groundwater are shown in Table 5. During the in situ remediation by MNBs, the TCE concentration in the monitoring wells showed remarkable decrease; an overall removal rate of 99% was reached. The final TCE concentration in the entire treated area decreased to less than 0.03 mg/L, which is the standard limit in Japan. The concentration of TCE over time is shown in Fig. 10. Ozone MNBs migrated with the groundwater for a constant supply of dissolved ozone. Previous research indicated that an ozone–peroxide system forms OH radicals, which presents high treatment efficiency on TCE; moreover, TCE can be treated also
452
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453
Table 5 Initial and final TCE concentrations in groundwater. Well Number
#1
#2
#3
#4
#5
Initial TCE concentration (mg/L) Final TCE concentration (mg/L)
3.529 ± 0.002 0.010 ± 0.004
4.502 ± 0.005 0.003 ± 0.001
10.130 ± 0.004 0.018 ± 0.014
0.264 ± 0.002 0.007 ± 0.002
2.129 ± 0.004 0.003 ± 0.000
by direct reaction with ozone [11]. The reaction between ozone plus H2 O2 and TCE was described by pseudo-first-order kinetics, in which the decreasing rate of TCE was related to the concentrations of ozone and H2 O2 [11]. For monitoring wells #1, #2, #3, and #5, 90% of TCE was removed during the first three or four days, and the decreasing rate from high to low occurred in #3, #2, #5, and #1. Well #3 was in closest proximity to the injection well and presented the highest concentration of oxidants, which resulted in the fastest decrease of TCE. In the field tests, the groundwater flow was affected by the injection and extraction wells. Well #2, located along the injection-extraction direction, with groundwater presented a higher flow rate; thus, the concentration of oxidants at #2 was higher than at #5. Well #1, located farthest from the injection well, presented the lowest oxidants concentration and resulted in the slowest decreasing rate of TCE. The initial concentration of monitoring well #4 was 0.26 mg/L, which is lower than that in other areas. However, the concentration of TCE showed apparent decreases during the treatment, which indicates that ozone MNBs are still effective for treating groundwater with relatively low concentrations of contaminants. During the first three or four days, the decreasing rate of TCE increased over time, which indicates that the concentration of oxidants in the site increased during the remediation. According to the results of extracted groundwater treatment, the degradation of TCE was attributed mainly to ozone. It can be concluded that ozone MNBs are stable in groundwater and that the constant injection increased the concentration of ozone MNBs at the site. When the concentration of TCE was lower than 0.01 mg/L, its degradation rate obviously slowed. Similar results were obtained in the treatment of methyl orange; thus, it can be assumed that the low concentration of organics-contaminant was the main reason for the limited remediation efficiency. 4. Conclusion In this study, the characteristics of ozone MNBs were studied. MNBs can reach a high unit quantity and remain stable in water for long periods. Ozone MNBs stay negatively charged under varied salinities, which means that they are stable and can be applied in the remediation of groundwater with varying salinities. MNBs significantly increase the mass transfer efficiency of ozone and can remain stable in water to constantly supply ozone. The half-life of ozone in the MNB system is significantly longer than that in the millimetre-bubble system. Under laboratory conditions, ozone had a significant effect on organics-contaminants in the surface water and groundwater, and the MNBs greatly enhanced the treatment efficiency. An in situ remediation facility was developed, and field tests were performed in a TCE-contaminated site in Japan. An overall removal rate of 99% was reached after six days of treatment. Ozone MNBs showed a sound effect on the remediation of TCE-contaminated groundwater and potentially represent an innovative technology for in situ remediation of organics-contaminated groundwater. Novelty statement As a widely used oxidant, the application of ozone in groundwater remediation is limited by its migration ability. The purpose
of this manuscript is to study the feasibility and efficiency of applying ozone to groundwater remediation in the form of micronano bubbles (MNBs). Basic characteristics and mass transfer behaviour of ozone MNBs were studied, and the remediation efficiency of organics-contaminants by ozone MNBs were studied through laboratory and field tests. Ozone MNBs represent an innovative technology for in-situ remediation of organics-contaminated groundwater, and this manuscript will surely attract attention of geo-environmental engineers interested in technologies for organics-contaminated groundwater remediation.
Acknowledgements The financial support from National Natural Science Foundation of China (Project No. 41372352, 51323014, 51661165015) and Tsinghua University (Project No. 2015THZ02-2-20161080101, 2015THZ01-1-20161080079) are gratefully acknowledged. The support for the testing facility from State Key Laboratory of Hydro Science and Engineering (SKLHSE-2016-d-03) is also acknowledged. The authors thank the assistance of Dr. D Song at IS Solution Company for field monitoring on MNB remediation. The two anonymous reviewers are also thanked for giving valuable comments that improved the overall quality of this paper.
References [1] How to Evaluate Alternative Cleanup Technologies for Underground Storage Tank Sites, A Guide for Corrective Action Plan Reviewers, United States Environmental Protection Agency, Washington, DC, 2004. [2] S.G. Huling Pivetz, In-situ Chemical Oxidation. No. EPA/600/R-06/072, Environmental Protection Agency Washington DC Office of Water, 2006. [3] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced oxidation technique, J. Hazard. Mater. 98 (2003) 33–50. [4] L. Hou, H. Zhang, X. Xue, Ultrasound enhanced heterogeneous activation of peroxydisulfate by magnetite catalyst for the degradation of tetracycline in water, Sep. Purif. Technol. 84 (2012) 147–152. [5] D.A. House, Kinetics and mechanism of oxidations by peroxydisulfate, Chem. Rev. 62 (1962) 185–203. [6] L.X. David, F.W. Schwartz, Using phosphate to control the Mn oxide precipitation during in situ chemical oxidation of chlorinated ethylenes by permanganate, In ACS symposium series. Oxford University Press 904 (2005) 82–95. [7] M.M. Huber, S. Canonica, G.Y. Park, U. von Gunten, Oxidation of pharmaceuticals during ozonation and advanced oxidation processes, Environ. Sci. Technol 37 (2003) 1016–1024. [8] T.A. Ternes, M. Meisenheimer, D. McDowell, F. Sacher, H.J. Brauch, B. Haist-Gulde, G. Preuss, U. Wilme, N. Zullei-Seibert, Removal of pharmaceuticals during drinking water treatment, Environ. Sci. Technol. 36 (2002) 3855–3863. [9] M.M. Mitani, A.A. Keller, C.A. Bunton, C.A. Bunton, R.G. Rinker, O.C. Sandall, Kinetics and products of reactions of MTBE with ozone and ozone/hydrogen peroxide in water, J. Hazard. Mater. 89 (2002) 197–212. [10] H. Barndõk, L. Cortijo, D. Hermosilla, C. Negro, A. Blanco, Removal of 1: 4-dioxane from industrial wastewaters: routes of decomposition under different operational conditions to determine the ozone oxidation capacity, J. Hazard. Mater. 280 (2014) 340–347. [11] W.H. Glaze, J.W. Kang, Advanced oxidation processes for treating groundwater contaminated with TCE and PCE: laboratory studies, J. (Am. Water Works Assoc.) (1988) 57–63. [12] C. Tizaoui, L. Bouselmi, L. Mansouri, A. Ghrabi, Landfill leachate treatment with ozone and ozone/hydrogen peroxide systems, J. Hazard. Mater. 140 (2007) 316–324. [13] A. Dettmer, R. Ball, T.B. Boving, N.A. Khan, T. Schaub, N. Sudasinghe, C.A. Fernandez, K.C. Carroll, Stabilization and prolonged reactivity of aqueous-phase ozone with cyclodextrin, J. Contam. Hydrol. 196 (2017) 1–9. [14] P. Attard, M.P. Moody, J.W.G. Tyrrell Nanobubbles, The big picture, Physica. A: Stat. Mech. Appl. 314 (2002) 696–705.
L. Hu, Z. Xia / Journal of Hazardous Materials 342 (2018) 446–453 [15] L.B. Chu, X.H. Xing, A.F. Yu, Y.N. Zhou, X.L. Sun, B. Jurcik, Enhanced ozonation of simulated dyestuff wastewater by microbubbles, Chemosphere 68 (2007) 1854–1860. [16] J.R.T. Seddon, D. Lohse, W.A. Ducker, V.S.J. Craig, A deliberation on nanobubbles at surfaces and in bulk, ChemPhysChem 13 (2012) 2179–2187. [17] F.Y. Ushikubo, T. Furukawa, R. Nakagawa, M. Enari, Y. Makino, Y. Kawagoe, T. Shiina, S. Oshita, Evidence of the existence and the stability of nano-bubbles in water, Coll. Surf. A 361 (2010) 31–37. [18] A. Azevedo, R. Etchepare, S. Calgaroto, J. Rubio, Aqueous dispersions of nanobubbles: generation, properties and features, Min. Eng. 94 (2016) 29–37. [19] J.H. Weijs, J.R.T. Seddon, D. Lohse, Diffusive shielding stabilizes bulk nanobubble clusters, ChemPhysChem 13 (2012) 2197–2204. [20] H. Li, L. Hu, Z. Xia, Impact of groundwater salinity on bioremediation enhanced by micro-nano bubbles, Materials 6 (2013) 3676–3687. [21] H. Li, L. Hu, D. Song, A. Al-Tabbaa, Subsurface transport behavior of micro-Nano bubbles and potential applications for groundwater remediation, Int. J. Environ. Res. Public Health. 11 (2014) 473–486. [22] H. Li, L. Hu, D. Song, F. Lin, Characteristics of micro-nano bubbles and potential application in groundwater bioremediation, Water Environ. Res. 86 (2014) 844–851. [23] A. Yoshida, O. Takahashi, Y. Ishii, Y. Sekimoto, Y. Kurata, Water purification using the adsorption characteristics of microbubbles, Jpn. J. Appl. Phys. 47 (2008). [24] L. Hu, X. Wu, Y. Liu, J.N. Meegoda, S. Gao, Physical modeling of air flow during air sparging remediation, Environ. Sci. Technol. 44 (2010) 3883–3888. [25] J.N. Meegoda, L. Hu, A review of centrifugal testing of gasoline contamination and remediation, Int. J. Environ. Res. Public Health 8 (2011) 3496–3513.
453
[26] L. Hu, J.N. Meegoda, J. Du, S. Gao, X. Wu, Centrifugal study of zone of influence during air-sparging, J. Environ. Monit. 13 (2011) 2443–2449. [27] D. Song, A. Katayama, Approach for estimating microbial growth and the biodegradation of hydrocarbon contaminants in subsoil based on field measurements: 1. Model development and verification, Environ. Sci. Tech. 44 (2009) 767–773. [28] M. Takahashi, Base and technological application of micro-bubble and nano-bubble, Mater. Integr. 22 (2009) 2–19. [29] A. Agarwal, W.J. Ng, Y. Liu, Principle and applications of microbubble and nanobubble technology for water treatment, Chemosphere 84 (2011) 1175–1180. [30] L.B. Chu, X.H. Xing, A.F. Yu, Y.N. Zhou, X.L. Sun, B. Jurcik, Enhanced ozonation of simulated dyestuff wastewater by microbubbles, Chemosphere 68 (2007) 1854–1860. [31] K. Terasaka, A. Hirabayashi, T. Nishino, S. Fujioka, D. Kobayashi, Development of microbubble aerator for waste water treatment using aerobic activated sludge, Chem. Eng. Sci. 66 (2011) 3172–3179. [32] L.C. Chen, Effects of factors and interacted factors on the optimal decolorization process of methyl orange by ozone, Water Res. 34 (2000) 974–982. [33] P. Li, M. Takahashi, K. Chiba, Degradation of phenol by the collapse of microbubbles, Chemosphere 75 (2009) 1371–1375.