- Email: [email protected]
Real-time monitoring of arsenic filtration by granular ferric hydroxide
ARTICLE IN PRESS Applied Radiation and Isotopes 68 (2010) 821–824
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Real-time monitoring of arsenic filtration by granular ferric hydroxide David E.B. Fleming a,, Isadel S. Eddy a, Mihai R. Gherase a, Meaghan K. Gibbons b, Graham A. Gagnon b a b
Physics Department, Mount Allison University, Sackville, NB, Canada Civil and Resource Engineering, Dalhousie University, Halifax, NS, Canada
a r t i c l e in fo
Keywords: X-ray fluorescence Arsenic Drinking water Granular ferric hydroxide
abstract Contamination of drinking water by arsenic is a serious public health issue in many parts of the world. One recent approach to this problem has been to filter out arsenic by use of granular ferric hydroxide (GFH), an adsorbent developed specifically for the selective removal of arsenic from water. Previous studies have documented the efficiency and high treatment capacity of this approach. We present a novel X-ray fluorescence method to monitor the accumulation of arsenic within a specially designed GFH column, as both a function of time (or water volume) and location along the column. Using a miniature X-ray tube and silicon PiN diode detector, X-ray fluorescence is used to detect characteristic X-rays of arsenic excited from within the GFH. Trials were performed using a water flow rate of approximately 1.5 L per hour, with an added arsenic concentration of approximately 1000 mg per litre. In this paper, trial results are presented and potential applications described. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction Contamination of drinking water by arsenic is a significant health issue for certain populations throughout the world. Normally, contamination results when naturally occurring geologic stores allow arsenic to reach groundwater (Vatutsina et al., 2007). The issue is especially critical in Bangladesh and parts of India and China (Chowdhury et al., 2000; Mo et al., 2006). Chronic exposure to arsenic through drinking water has been associated with skin effects and skin cancer (Karim, 2000), as well as other types of cancer (Chen et al., 1985; Morales et al., 2000). To reduce arsenic exposure, many water filtering approaches have been introduced. Granular adsorptive materials represent the most common filtering approach for small community systems and homes (Jing et al., 2008). One relatively new (Driehaus et al., 1998) filter of this type is granular ferric hydroxide (GFH). GFH is highly efficient at removing arsenic from water and has a high treatment capacity (Driehaus et al., 1998; Sperlich et al., 2005). Consequently, GFH has been implemented in more than 20 European water systems (Driehaus, 2002). An area of ongoing research seeks to understand and model the accumulation of arsenic in GFH. This is relevant as a matter of basic science and also for practical applications relating to issues such as filter replacement. Studies of GFH often make use of adsorption isotherms and breakthrough curves to better understand arsenic adsorption capacity. An adsorption isotherm is a relationship between arsenic concentration in GFH and its Corresponding author. Tel.: + 1 506 364 2584; fax: +1 506 364 2583.
E-mail address: dfl[email protected] (D.E.B. Fleming). 0969-8043/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2009.09.048
concentration in water under a given set of conditions (Sperlich et al., 2005). A breakthrough curve plots effluent over influent arsenic concentration as a function of filtered water volume. For GFH, existing models describing breakthrough are not in accord with experimentally observed arsenic breakthrough curves (Sperlich et al., 2005). The systematic differences between model and experiment may be reduced by lowering the modeled surface diffusion coefficient and introducing a time-dependent aspect to the coefficient (Sperlich et al., 2005). One explanation for this could relate to the formation of a surface iron arsenate precipitate (Hongshao and Stanforth, 2001). The development of a real-time method of monitoring arsenic concentration in GFH would be of considerable benefit to future research. In this paper, we demonstrate such a method using X-ray fluorescence (XRF). This novel approach has the additional advantages of being non-invasive and non-destructive; the system under study is not disturbed in any fashion. As such, XRF is ideally suited to the monitoring of arsenic accumulation in an experimental GFH column as a function of time (or water volume) and location.
2. Methods For the purpose of this pilot study, tap water from Sackville, New Brunswick, Canada was spiked with arsenic and circulated through a column of GFH (described below) over a period of 30 days. Each day, tap water was measured into a 20 L influent tank, and prepared to an arsenic concentration of approximately 1000 mg A s L 1 by adding measured quantities of a 10030 730 mg A s mL 1 atomic absorption spectrometry standard solution
ARTICLE IN PRESS 822
D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
(Sigma-Aldrich; Oakville, Ontario), using an adjustable volume pipettor. A concentration of 1000 mg A s L 1, while extreme, falls within the range observed from drinking water in parts of Bangladesh and India (Chowdhury et al., 2000). The prepared water was run continuously through the GFH column at a flow rate of approximately 1.5 L h 1 for 12 h each day using a Masterflex 7553-70 peristaltic pump (Cole-Parmer; Mississauga, Ontario). The flow rate and configuration were selected not to reproduce a residential treatment system, but rather to allow a controlled determination of arsenic accumulation within GFH as a function of time and location. The average flow rate was 1.477 L h 1, although day-to-day variations were observed within the range of 1.458–1.625 Lh 1. A 12 h resting period was observed between daily pumping sessions. Influent and effluent water samples were monitored for pH using a Beckman 300 series pH meter (Beckman Instruments; Fullerton, CA) calibrated with standard buffer solutions. Influent and effluent samples were also tested for arsenic concentration using an atomic absorption spectrometer (PerkinElmer AAnalyst 200) equipped with a graphite furnace (PerkinElmer HGA 900), or GFAAS system. Arsenic standards were used for calibration and quality control procedures used to estimate recovery. The minimum detection limit (MDL) for arsenic in water using the GFAAS system was determined to be 1.2 mg A s L 1. This MDL is characteristic of a sensitive system; by comparison, the World Health Organization provisional guideline value for water is 10 mg A s L 1. The GFH column was 97 cm in height and constructed from polyvinyl chloride (PVC) piping. The PVC piping had an inner diameter of 3.9 cm and was filled with GFH (Siemens Water Technologies; Warrendale, PA) on top of a base layer of approximately 4 cm of gravel. At regular spacing, holes were cut into the piping and Plexi-glass windows attached to allow arsenic measurement via X-ray fluorescence (see below). Windows were 2 mm in thickness, 2.4 cm in width, and 10.7 cm in height, and were placed above the influent end of the column (Fig. 1). For the purpose of the current study, arsenic measurements were predominantly made from the bottommost window, at three locations labelled Bottom, Middle, and Top (Fig. 1). Arsenic adsorbed by the GFH was measured by inducing X-ray fluorescence and detecting the characteristic Ka X-rays of arsenic (Fleming and Gherase, 2007). This was accomplished using an Innov-X Alpha-4000S handheld unit containing an X-ray tube and radiation detector (Innov-X Systems Canada; Mississauga, Ontario). The unit contained a tungsten X-ray target, with 2 mm of aluminum for beam filtration. A silicon PiN diode was used as the radiation detector. Measurements were made by placing the aperture of the X-ray tube/detector assembly flush against the window of the GFH column (Fig. 2), with each measurement lasting 180 s (real time). Initially, measurements were made only at the Middle location of the bottommost window (Window 1) every four hours. When the active pumping time reached 60 h, readings were also introduced for the Bottom and Top locations. Monitoring was performed only occasionally for Windows 2–5, with the Window 2 measurements made approximately 23 cm above the gravel layer. When fluoresced, arsenic emits high intensity characteristic Ka X-rays at 10.5 keV (Deslattes et al., 2003). The detected energy spectrum (count rate vs. energy) was imported into Origin 7.5 software (OriginLab Corporation; Northampton, MA) and isolated over the range 10–11 keV. A Gaussian function was then fit to the detected Ka X-ray peak, assuming a constant ‘‘background’’ count rate over the energy range of interest. For each measurement, the area of the peak was calculated and taken as an indicator of arsenic at the location. The resulting areas were used to compare the accumulation of arsenic in GFH as a function of time (or water volume) and location along the column. Finally, these areas were
Fig. 1. Schematic of the experimental set-up. The three locations of measurement in Window 1 are labeled: Bottom, Middle, and Top. Distances are from the top of the gravel layer.
Fig. 2. The handheld X-ray fluorescence device being used to measure arsenic within the column of granular ferric hydroxide.
converted to concentrations of arsenic, using a relation established from a series of calibration trials. For these trials, eight GFH standards were prepared, with arsenic concentrations ranging from 0 to 140 mg A s g 1 GFH. The standards were measured through the same 2 mm thick Plexi-glass as used in the experimental trials. In this way, a relation between peak area and arsenic concentration was established.
ARTICLE IN PRESS D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
Fig. 3. Energy spectra obtained from the three measurement locations along the column of granular ferric hydroxide after 360 h (532 L) of water flow. The best fit Gaussian function is provided for each of the three measurements.
823
Fig. 4. Peak area of the characteristic arsenic Ka X-ray as a function of water volume for the three measurement locations.
3. Results The median pH of the prepared influent water was 6.86, while the median pH of the effluent water was 6.53. Atomic absorption spectrometry confirmed high concentrations of arsenic in the influent water, with samples ranging between 880 mg A s L 1 and 934 mg A s L 1. In every case, concentrations of arsenic in the effluent water were below the MDL. Over the course of this pilot study, XRF measurements clearly demonstrated the accumulation of arsenic at the various locations within the bottommost window (Window 1) of the GFH column. Notably, a fluorescent peak was evident at the time of the very first measurement (t = 4 h), made at the Middle location. Examples of Gaussian function fits to experimental data are shown in Fig. 3 for measurements performed at t = 360 h from the three locations. For each measurement, a Gaussian peak area was calculated from the fit to the associated energy spectrum. To account for the slight variability in water flow rate during the experiment, arsenic signal was examined as a function of filtered water volume (rather than as a function of time) and location. Results corresponding to measurements made at 12 h (or 18 L) intervals are shown in Fig. 4. When converted from a peak area to a concentration of arsenic, using the calibration approach described above, the final set of readings (corresponding to the data shown in Fig. 3) indicated arsenic concentrations of 5652 7 176 mg A s g 1 GFH, 4518 7 141 mg A s g 1 GFH, and 3455 7 108 mg A s g 1 GFH. For measurements made on Window 2 and above, the arsenic signal never exceeded the method MDL (11 mg A s g 1 GFH).
4. Discussion In a novel application of X-ray spectrometry, the accumulation of arsenic within a GFH column was successfully monitored using a handheld XRF device. The excellent arsenic filtration properties of GFH were confirmed, with no arsenic detected in the effluent water despite very high influent concentrations. An X-ray fluorescent arsenic signal was clearly evident from the GFH early in the experiment, and the time and spatial evolution of this signal was described. Large quantities of arsenic accumulated within 12 cm of the influent end of the GFH but, significantly, no arsenic was detected at further distant locations of measurement.
The progression of arsenic accumulation with time in the GFH column indicated a rapid rise in concentration near the influent end. Accumulation then slowed as the arsenic concentration within GFH approached a plateau. This effect was most noticeable for measurements made 5.2 cm from the influent end. In contrast, measurements made 12.0 cm from the influent end demonstrated an initial low concentration of arsenic, followed by a relatively constant increase. This may be understood by considering the concentration of arsenic in water flowing through this location. Initially, water reaching this location contained almost no arsenic since effective adsorption had occurred near the base of the column. Over time, as these GFH sites near the base became heavily laden with arsenic, the amount of adsorption decreased, allowing higher water concentrations to reach points further downstream. It is evident that this XRF approach can yield important information in real-time about arsenic adsorption as a function of time (or water volume) and location along a column of GFH. Future work should explore the use of this technique to help refine and validate quantitative models of arsenic adsorption under various experimental conditions. Practical applications could also relate to monitoring water treatment filters to optimize the timing of media replacement, and the development of a compact GFH-containing unit for rapid field assessment of arsenic in water without the need for chemical testing. Further analytical improvements to the described method are possible. Consideration could be given to the use of Kb arsenic characteristic X-rays (11.7 keV) in addition to the Ka X-rays (10.5 keV) analyzed here. Since, however, the Kb X-rays are less intense and, in the current application, located on a higher background component of the energy spectrum, their inclusion would be of minor consequence. Perhaps more significant would be the development of a normalization approach comparing arsenic fluorescent peaks to those of iron (Ka: 6.4 keV; Kb: 7.1 keV) as a function of arsenic concentration within GFH. Finally, calibration standards should be developed over a wider concentration range than those presented here. Nonetheless, results from this pilot study indicate that the use of XRF to monitor arsenic accumulation in GFH represents a rapid, convenient, and informative technique with considerable potential for future application.
ARTICLE IN PRESS 824
D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
Acknowledgments This work was supported by the Natural Sciences and Engineering Research Council of Canada. The Canada Research Chair program is gratefully acknowledged. Equipment used in this research was partially funded by the RURAL Centre for Research Development of the Canadian Institutes of Health Research. References Chen, C.J., Chuang, Y.-C., Lin, T.-M., Wu, H.-Y., 1985. Malignant neoplasms among residents of a blackfoot disease-endemic area in Taiwan: high-arsenic artesian well water and cancers. Cancer Res. 45, 5895–5899. Chowdhury, U.K., Biswas, B.K., Chowdhury, T.R., Samanta, G., Mandal, B.K., Basu, G.C., Chanda, C.R., Lodh, D., Saha, K.C., Mukherjee, S.K., Roy, S., Kabir, S., Quamruzzaman, Q., Chakraborti, D., 2000. Groundwater arsenic contamination in Bangladesh and West Bengal, India. Environ. Health Perspect. 108, 393–397. Deslattes, R.D., Kessler Jr., E.G., Indelicato, P., de Billy, L., Lindroth, E., Anton, J., 2003. X-ray transition energies: new approach to a comprehensive evaluation. Rev. Mod. Phys. 75, 35–99. Driehaus, W., 2002. Arsenic removal—experience with the GEH process in Germany. Water Supply 2, 275–280.
Driehaus, W., Jekel, M., Hildebrandt, U., 1998. Granular ferric hydroxide—a new adsorbent for the removal of arsenic from natural water. J. Water SRT-Aqua 47, 30–35. Fleming, D.E.B., Gherase, M.R., 2007. A rapid, high-sensitivity technique for measuring arsenic in skin phantoms using a portable X-ray tube and detector. Phys. Med. Biol. 52, N459–N465. Hongshao, Z., Stanforth, R., 2001. Competitive adsorption of phosphate and arsenate on goethite. Environ. Sci. Technol. 35, 4753–4757. Jing, C., Liu, S., Meng, X., 2008. Arsenic remobilization in water treatment adsorbents under reducing conditions: Part I. Incubation study. Sci. Total Environ. 389, 188–194. Karim, M.M., 2000. Arsenic in groundwater and health problems in Bangladesh. Water Res. 34, 304–310. Mo, J., Xia, Y., Wade, T.J., Schmitt, M., Le, X.C., Dang, R., Mumford, J.L., 2006. Chronic arsenic exposure and oxidative stress: OGG1 expression and arsenic exposure, nail selenium, and skin hyperkeratosis in Inner Mongolia. Environ. Health Perspect. 114, 835–841. Morales, K.H., Ryan, L., Kuo, T.-L., Wu, M.-M., Chen, C.-J., 2000. Risk of internal cancers from arsenic in drinking water. Environ. Health Perspect. 108, 655–661. Sperlich, A., Werner, A., Genz, A., Amy, G., Worch, E., Jekel, M., 2005. Breakthrough behavior of granular ferric hydroxide (GFH) fixed-bed adsorption filters: modeling and experimental approaches. Water Res. 39, 1190–1198. Vatutsina, O.M., Soldatov, V.S., Sokolova, V.I., Johann, J., Bissen, M., Weissenbacher, A., 2007. A new hybrid (polymer/inorganic) fibrous sorbent for arsenic removal from drinking water. React. Funct. Polym. 67, 184–201.
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Real-time monitoring of arsenic filtration by granular ferric hydroxide David E.B. Fleming a,, Isadel S. Eddy a, Mihai R. Gherase a, Meaghan K. Gibbons b, Graham A. Gagnon b a b
Physics Department, Mount Allison University, Sackville, NB, Canada Civil and Resource Engineering, Dalhousie University, Halifax, NS, Canada
a r t i c l e in fo
Keywords: X-ray fluorescence Arsenic Drinking water Granular ferric hydroxide
abstract Contamination of drinking water by arsenic is a serious public health issue in many parts of the world. One recent approach to this problem has been to filter out arsenic by use of granular ferric hydroxide (GFH), an adsorbent developed specifically for the selective removal of arsenic from water. Previous studies have documented the efficiency and high treatment capacity of this approach. We present a novel X-ray fluorescence method to monitor the accumulation of arsenic within a specially designed GFH column, as both a function of time (or water volume) and location along the column. Using a miniature X-ray tube and silicon PiN diode detector, X-ray fluorescence is used to detect characteristic X-rays of arsenic excited from within the GFH. Trials were performed using a water flow rate of approximately 1.5 L per hour, with an added arsenic concentration of approximately 1000 mg per litre. In this paper, trial results are presented and potential applications described. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction Contamination of drinking water by arsenic is a significant health issue for certain populations throughout the world. Normally, contamination results when naturally occurring geologic stores allow arsenic to reach groundwater (Vatutsina et al., 2007). The issue is especially critical in Bangladesh and parts of India and China (Chowdhury et al., 2000; Mo et al., 2006). Chronic exposure to arsenic through drinking water has been associated with skin effects and skin cancer (Karim, 2000), as well as other types of cancer (Chen et al., 1985; Morales et al., 2000). To reduce arsenic exposure, many water filtering approaches have been introduced. Granular adsorptive materials represent the most common filtering approach for small community systems and homes (Jing et al., 2008). One relatively new (Driehaus et al., 1998) filter of this type is granular ferric hydroxide (GFH). GFH is highly efficient at removing arsenic from water and has a high treatment capacity (Driehaus et al., 1998; Sperlich et al., 2005). Consequently, GFH has been implemented in more than 20 European water systems (Driehaus, 2002). An area of ongoing research seeks to understand and model the accumulation of arsenic in GFH. This is relevant as a matter of basic science and also for practical applications relating to issues such as filter replacement. Studies of GFH often make use of adsorption isotherms and breakthrough curves to better understand arsenic adsorption capacity. An adsorption isotherm is a relationship between arsenic concentration in GFH and its Corresponding author. Tel.: + 1 506 364 2584; fax: +1 506 364 2583.
E-mail address: dfl[email protected] (D.E.B. Fleming). 0969-8043/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2009.09.048
concentration in water under a given set of conditions (Sperlich et al., 2005). A breakthrough curve plots effluent over influent arsenic concentration as a function of filtered water volume. For GFH, existing models describing breakthrough are not in accord with experimentally observed arsenic breakthrough curves (Sperlich et al., 2005). The systematic differences between model and experiment may be reduced by lowering the modeled surface diffusion coefficient and introducing a time-dependent aspect to the coefficient (Sperlich et al., 2005). One explanation for this could relate to the formation of a surface iron arsenate precipitate (Hongshao and Stanforth, 2001). The development of a real-time method of monitoring arsenic concentration in GFH would be of considerable benefit to future research. In this paper, we demonstrate such a method using X-ray fluorescence (XRF). This novel approach has the additional advantages of being non-invasive and non-destructive; the system under study is not disturbed in any fashion. As such, XRF is ideally suited to the monitoring of arsenic accumulation in an experimental GFH column as a function of time (or water volume) and location.
2. Methods For the purpose of this pilot study, tap water from Sackville, New Brunswick, Canada was spiked with arsenic and circulated through a column of GFH (described below) over a period of 30 days. Each day, tap water was measured into a 20 L influent tank, and prepared to an arsenic concentration of approximately 1000 mg A s L 1 by adding measured quantities of a 10030 730 mg A s mL 1 atomic absorption spectrometry standard solution
ARTICLE IN PRESS 822
D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
(Sigma-Aldrich; Oakville, Ontario), using an adjustable volume pipettor. A concentration of 1000 mg A s L 1, while extreme, falls within the range observed from drinking water in parts of Bangladesh and India (Chowdhury et al., 2000). The prepared water was run continuously through the GFH column at a flow rate of approximately 1.5 L h 1 for 12 h each day using a Masterflex 7553-70 peristaltic pump (Cole-Parmer; Mississauga, Ontario). The flow rate and configuration were selected not to reproduce a residential treatment system, but rather to allow a controlled determination of arsenic accumulation within GFH as a function of time and location. The average flow rate was 1.477 L h 1, although day-to-day variations were observed within the range of 1.458–1.625 Lh 1. A 12 h resting period was observed between daily pumping sessions. Influent and effluent water samples were monitored for pH using a Beckman 300 series pH meter (Beckman Instruments; Fullerton, CA) calibrated with standard buffer solutions. Influent and effluent samples were also tested for arsenic concentration using an atomic absorption spectrometer (PerkinElmer AAnalyst 200) equipped with a graphite furnace (PerkinElmer HGA 900), or GFAAS system. Arsenic standards were used for calibration and quality control procedures used to estimate recovery. The minimum detection limit (MDL) for arsenic in water using the GFAAS system was determined to be 1.2 mg A s L 1. This MDL is characteristic of a sensitive system; by comparison, the World Health Organization provisional guideline value for water is 10 mg A s L 1. The GFH column was 97 cm in height and constructed from polyvinyl chloride (PVC) piping. The PVC piping had an inner diameter of 3.9 cm and was filled with GFH (Siemens Water Technologies; Warrendale, PA) on top of a base layer of approximately 4 cm of gravel. At regular spacing, holes were cut into the piping and Plexi-glass windows attached to allow arsenic measurement via X-ray fluorescence (see below). Windows were 2 mm in thickness, 2.4 cm in width, and 10.7 cm in height, and were placed above the influent end of the column (Fig. 1). For the purpose of the current study, arsenic measurements were predominantly made from the bottommost window, at three locations labelled Bottom, Middle, and Top (Fig. 1). Arsenic adsorbed by the GFH was measured by inducing X-ray fluorescence and detecting the characteristic Ka X-rays of arsenic (Fleming and Gherase, 2007). This was accomplished using an Innov-X Alpha-4000S handheld unit containing an X-ray tube and radiation detector (Innov-X Systems Canada; Mississauga, Ontario). The unit contained a tungsten X-ray target, with 2 mm of aluminum for beam filtration. A silicon PiN diode was used as the radiation detector. Measurements were made by placing the aperture of the X-ray tube/detector assembly flush against the window of the GFH column (Fig. 2), with each measurement lasting 180 s (real time). Initially, measurements were made only at the Middle location of the bottommost window (Window 1) every four hours. When the active pumping time reached 60 h, readings were also introduced for the Bottom and Top locations. Monitoring was performed only occasionally for Windows 2–5, with the Window 2 measurements made approximately 23 cm above the gravel layer. When fluoresced, arsenic emits high intensity characteristic Ka X-rays at 10.5 keV (Deslattes et al., 2003). The detected energy spectrum (count rate vs. energy) was imported into Origin 7.5 software (OriginLab Corporation; Northampton, MA) and isolated over the range 10–11 keV. A Gaussian function was then fit to the detected Ka X-ray peak, assuming a constant ‘‘background’’ count rate over the energy range of interest. For each measurement, the area of the peak was calculated and taken as an indicator of arsenic at the location. The resulting areas were used to compare the accumulation of arsenic in GFH as a function of time (or water volume) and location along the column. Finally, these areas were
Fig. 1. Schematic of the experimental set-up. The three locations of measurement in Window 1 are labeled: Bottom, Middle, and Top. Distances are from the top of the gravel layer.
Fig. 2. The handheld X-ray fluorescence device being used to measure arsenic within the column of granular ferric hydroxide.
converted to concentrations of arsenic, using a relation established from a series of calibration trials. For these trials, eight GFH standards were prepared, with arsenic concentrations ranging from 0 to 140 mg A s g 1 GFH. The standards were measured through the same 2 mm thick Plexi-glass as used in the experimental trials. In this way, a relation between peak area and arsenic concentration was established.
ARTICLE IN PRESS D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
Fig. 3. Energy spectra obtained from the three measurement locations along the column of granular ferric hydroxide after 360 h (532 L) of water flow. The best fit Gaussian function is provided for each of the three measurements.
823
Fig. 4. Peak area of the characteristic arsenic Ka X-ray as a function of water volume for the three measurement locations.
3. Results The median pH of the prepared influent water was 6.86, while the median pH of the effluent water was 6.53. Atomic absorption spectrometry confirmed high concentrations of arsenic in the influent water, with samples ranging between 880 mg A s L 1 and 934 mg A s L 1. In every case, concentrations of arsenic in the effluent water were below the MDL. Over the course of this pilot study, XRF measurements clearly demonstrated the accumulation of arsenic at the various locations within the bottommost window (Window 1) of the GFH column. Notably, a fluorescent peak was evident at the time of the very first measurement (t = 4 h), made at the Middle location. Examples of Gaussian function fits to experimental data are shown in Fig. 3 for measurements performed at t = 360 h from the three locations. For each measurement, a Gaussian peak area was calculated from the fit to the associated energy spectrum. To account for the slight variability in water flow rate during the experiment, arsenic signal was examined as a function of filtered water volume (rather than as a function of time) and location. Results corresponding to measurements made at 12 h (or 18 L) intervals are shown in Fig. 4. When converted from a peak area to a concentration of arsenic, using the calibration approach described above, the final set of readings (corresponding to the data shown in Fig. 3) indicated arsenic concentrations of 5652 7 176 mg A s g 1 GFH, 4518 7 141 mg A s g 1 GFH, and 3455 7 108 mg A s g 1 GFH. For measurements made on Window 2 and above, the arsenic signal never exceeded the method MDL (11 mg A s g 1 GFH).
4. Discussion In a novel application of X-ray spectrometry, the accumulation of arsenic within a GFH column was successfully monitored using a handheld XRF device. The excellent arsenic filtration properties of GFH were confirmed, with no arsenic detected in the effluent water despite very high influent concentrations. An X-ray fluorescent arsenic signal was clearly evident from the GFH early in the experiment, and the time and spatial evolution of this signal was described. Large quantities of arsenic accumulated within 12 cm of the influent end of the GFH but, significantly, no arsenic was detected at further distant locations of measurement.
The progression of arsenic accumulation with time in the GFH column indicated a rapid rise in concentration near the influent end. Accumulation then slowed as the arsenic concentration within GFH approached a plateau. This effect was most noticeable for measurements made 5.2 cm from the influent end. In contrast, measurements made 12.0 cm from the influent end demonstrated an initial low concentration of arsenic, followed by a relatively constant increase. This may be understood by considering the concentration of arsenic in water flowing through this location. Initially, water reaching this location contained almost no arsenic since effective adsorption had occurred near the base of the column. Over time, as these GFH sites near the base became heavily laden with arsenic, the amount of adsorption decreased, allowing higher water concentrations to reach points further downstream. It is evident that this XRF approach can yield important information in real-time about arsenic adsorption as a function of time (or water volume) and location along a column of GFH. Future work should explore the use of this technique to help refine and validate quantitative models of arsenic adsorption under various experimental conditions. Practical applications could also relate to monitoring water treatment filters to optimize the timing of media replacement, and the development of a compact GFH-containing unit for rapid field assessment of arsenic in water without the need for chemical testing. Further analytical improvements to the described method are possible. Consideration could be given to the use of Kb arsenic characteristic X-rays (11.7 keV) in addition to the Ka X-rays (10.5 keV) analyzed here. Since, however, the Kb X-rays are less intense and, in the current application, located on a higher background component of the energy spectrum, their inclusion would be of minor consequence. Perhaps more significant would be the development of a normalization approach comparing arsenic fluorescent peaks to those of iron (Ka: 6.4 keV; Kb: 7.1 keV) as a function of arsenic concentration within GFH. Finally, calibration standards should be developed over a wider concentration range than those presented here. Nonetheless, results from this pilot study indicate that the use of XRF to monitor arsenic accumulation in GFH represents a rapid, convenient, and informative technique with considerable potential for future application.
ARTICLE IN PRESS 824
D.E.B. Fleming et al. / Applied Radiation and Isotopes 68 (2010) 821–824
Acknowledgments This work was supported by the Natural Sciences and Engineering Research Council of Canada. The Canada Research Chair program is gratefully acknowledged. Equipment used in this research was partially funded by the RURAL Centre for Research Development of the Canadian Institutes of Health Research. References Chen, C.J., Chuang, Y.-C., Lin, T.-M., Wu, H.-Y., 1985. Malignant neoplasms among residents of a blackfoot disease-endemic area in Taiwan: high-arsenic artesian well water and cancers. Cancer Res. 45, 5895–5899. Chowdhury, U.K., Biswas, B.K., Chowdhury, T.R., Samanta, G., Mandal, B.K., Basu, G.C., Chanda, C.R., Lodh, D., Saha, K.C., Mukherjee, S.K., Roy, S., Kabir, S., Quamruzzaman, Q., Chakraborti, D., 2000. Groundwater arsenic contamination in Bangladesh and West Bengal, India. Environ. Health Perspect. 108, 393–397. Deslattes, R.D., Kessler Jr., E.G., Indelicato, P., de Billy, L., Lindroth, E., Anton, J., 2003. X-ray transition energies: new approach to a comprehensive evaluation. Rev. Mod. Phys. 75, 35–99. Driehaus, W., 2002. Arsenic removal—experience with the GEH process in Germany. Water Supply 2, 275–280.
Driehaus, W., Jekel, M., Hildebrandt, U., 1998. Granular ferric hydroxide—a new adsorbent for the removal of arsenic from natural water. J. Water SRT-Aqua 47, 30–35. Fleming, D.E.B., Gherase, M.R., 2007. A rapid, high-sensitivity technique for measuring arsenic in skin phantoms using a portable X-ray tube and detector. Phys. Med. Biol. 52, N459–N465. Hongshao, Z., Stanforth, R., 2001. Competitive adsorption of phosphate and arsenate on goethite. Environ. Sci. Technol. 35, 4753–4757. Jing, C., Liu, S., Meng, X., 2008. Arsenic remobilization in water treatment adsorbents under reducing conditions: Part I. Incubation study. Sci. Total Environ. 389, 188–194. Karim, M.M., 2000. Arsenic in groundwater and health problems in Bangladesh. Water Res. 34, 304–310. Mo, J., Xia, Y., Wade, T.J., Schmitt, M., Le, X.C., Dang, R., Mumford, J.L., 2006. Chronic arsenic exposure and oxidative stress: OGG1 expression and arsenic exposure, nail selenium, and skin hyperkeratosis in Inner Mongolia. Environ. Health Perspect. 114, 835–841. Morales, K.H., Ryan, L., Kuo, T.-L., Wu, M.-M., Chen, C.-J., 2000. Risk of internal cancers from arsenic in drinking water. Environ. Health Perspect. 108, 655–661. Sperlich, A., Werner, A., Genz, A., Amy, G., Worch, E., Jekel, M., 2005. Breakthrough behavior of granular ferric hydroxide (GFH) fixed-bed adsorption filters: modeling and experimental approaches. Water Res. 39, 1190–1198. Vatutsina, O.M., Soldatov, V.S., Sokolova, V.I., Johann, J., Bissen, M., Weissenbacher, A., 2007. A new hybrid (polymer/inorganic) fibrous sorbent for arsenic removal from drinking water. React. Funct. Polym. 67, 184–201.