Grantee Research Project Results
Final Report: Use of Bone Char for the Removal of Arsenic and Uranium from Groundwater at the Pine Ridge Reservation
EPA Grant Number: SU835069Title: Use of Bone Char for the Removal of Arsenic and Uranium from Groundwater at the Pine Ridge Reservation
Investigators: Werth, Charles J , Llewellyn, Alex , Parker, Kimberly M , Salvatore, Michelle , Becraft, Jacob , Genchanok, Yana , Dam, Emily Van , Freeck, Jason , Wang, Hanting , Nell, Marika , Feeney, Connor , Nguyen, Tien-Hung , Llewellyn, Brett , Marcinkevicius, Algimantas , Benson, Nora , Wisniewski, Alexander , Hou, Serena , DeMarco, Vanessa , Bollinger, Drew , Mosiman, Daniel , Knaizer, Brendon , Michelson, Kyle , Choe, Jong Kwon , Bergquist, Allison
Institution: University of Illinois Urbana-Champaign , Oglala Lakota College
EPA Project Officer: Page, Angela
Phase: II
Project Period: August 15, 2011 through August 14, 2013 (Extended to August 14, 2014)
Project Amount: $75,000
RFA: P3 Awards: A National Student Design Competition for Sustainability Focusing on People, Prosperity and the Planet - Phase 2 (2011) Recipients Lists
Research Category: Pollution Prevention/Sustainable Development , P3 Awards , Sustainable and Healthy Communities
Objective:
In August of 2009, our team traveled for the first time to the Pine Ridge Reservation in South Dakota, home of the Oglala Lakota Tribe. The purpose of the trip was to verify a U. S. Geological Survey (USGS) report from the 1990s, which indicated that much of the groundwater on the reservation was contaminated with arsenic (As) and uranium (U) above the Environmental Protection Agency (EPA) maximum containment limits (MCLs). Samples of private wells, municipal sources, and spring sources were taken, and it was confirmed that 35% of private wells contained As above the MCL, while 6% contained U above the MCL. While some residents have access to contaminant-free municipal water sources, many prefer the taste of their non-chlorinated private well water. This trip established the motivation for our Phase I and Phase II USEPA P3 grants.
The goal of our project was to develop a cost-effective filter that could remove both arsenic and uranium focusing on people, prosperity, and planet. The Pine Ridge Reservation is home to 28,000 people, many of whom do not know the dangers of drinking contaminated water. Efforts were taken to inform their community of the issues and develop an educational program in partnership with the Oglala Lakota College (OLC) for further under-standing of the need for clean drinking water on the Reservation. This project was almost entirely led and performed by undergraduate students, with the exception of several graduate students who provided guidance and laboratory assistance at various times. Of special mention is Kyle Michelson, who was supported for one semester to help the undergraduate students and evaluate biological removal of U.
Phase I objectives were partially completed in the first year of the project, and involved: 1) evaluating As and U groundwater contamination at the Pine Ridge Reservation 2) characterizing bone char (BC), 3) assessing U and As removal capacity and kinetics using actual BC, and 4) developing a BC filtration device. Phase II objectives, and the focus of this report, built upon Phase I objectives and involved 1) further testing of drinking water used by the residents of the Reservation, 2) designing and testing a bone char filter, 3) helping to establish a small business for water treatment on the reservation, and 4) educating students at Illinois and the OLC about the importance of safe drinking water. Phase II objectives were addressed over the last three years.
Phase II continued the work on As and U removal capacity and kinetics using bone char synthesized at Illinois by the undergraduate students. Although the mechanisms of the removal process are not known, removal capacity and kinetics were evaluated in order to design a prototype filter. Bone char successfully removed U from the Reservation groundwater in our initial studies. However, the Reservation water contained As(III), which, unlike As(V), proved difficult to remove with bone char in our initial efforts. Hence, we then focused on using zero valent iron to remove As(III), and our batch results suggested successful removal. We then focused on finding the optimal mix of bone char and zero valent iron for plausible implementation in the field.
In order to evaluate removal of uranium from Reservation groundwater with bone char, we flushed the water through a bone char filled column with a 30 minute retention time. The column was 5 cm long and 1 cm in diameter. Water contained 80 ppb U, and was flushed through the column for 53 days before U breakthrough occurred. We then evaluated a 15 minute retention time column, and a similar breakthrough period was observed. This suggests that U sorption is fast, and mass transfer is not a concern for filter design. We used these results to design a prototype filter, and column study results indicate that breakthrough will not occur for over one year in the full-size reactor.
Summary/Accomplishments (Outputs/Outcomes):
Groundwater Contamination on the Pine Ridge Reservation
As noted above, undergraduate students took two trips to the Pine Ridge Reservation during Phase II of our study. During each of these trips, they sampled from a variety of sources including indoor faucets, outdoor wells, and an outdoor pipe that was continuously flowing. The locations of 23 groundwater sampling locations are shown in Figure 1. Numbers in red indicate locations where As was above the maximum contaminant level (MCL) in groundwater. Numbers in green indicate locations where U was above the MCL in groundwater. The groundwater is being ingested without treatment, so residents of the Pine Ridge Reservation are consuming As and U at unsafe levels.
Figure 1. Groundwater sampling locations at the Pine Ridge Indian Reservation in South Dakota.
Bone Char Synthesis
Bones were obtained from the meat laboratory at the University of Illinois. They were boiled in water to remove attached meat, and then pulverized in a shatter box. After pulverization, the bones were charred at 500ºC for 7 hours, and then sieved to obtain a size range of 300-2000 μm.
Initial Column Studies Analyzing Uranium Breakthrough
Figure 2 shows the results of a bench-scale column experiment. The column used in this experiment was 5 cm in length, 1 cm in diameter, and made of stainless steel. The column contained approximately 2.16 g of bone char that ranged between 300 uM and 2 mm in particle size. Throughout the experiment, effluent samples were analyzed using Ion Coupled Plasma - Mass Spectrometry. The full length of the experiment was 69 days in total, and breakthrough began around ~60 days. However, there were small gaps in continuous flow during the experiment due to student conflicts with breaks, exams, etc. Because the students recorded all of the �non-flow� periods, they were able to accurately calculate the full run time of the column, which was 52.6 days. Using an estimate of bone char porosity n = 0.5, column volume V = 3.93 cm3, and flow rate Q = 4.00 ml/h, the number of pore volumes that flowed through the column over the length of the experiment was calculated to be VP = 2527.
Figure 2. Normalized breakthrough curve for the removal of U (30 minute retention time).
Following the experiment shown in Figure 2, a new column was started with a retention time of 15 minutes. This column was stopped after 30 days, but breakthrough did not occur during this period. Hence, breakthrough for a 15 minute retention time exceeded 30 days, and breakthrough for a 30 minute retention time was approximately 40 to 50 days. Based on the retention time expected in the field, the column breakthrough results suggests that a full-scale column will be in operation for one year or more.
Batch Studies Analyzing Arsenic Removal
Different mass ratios of zero valent iron to bone char were tested for removal of both arsenic and uranium from water. Uranium was initially at 35 ppb, and arsenic as As(III) was initially at 12 ppb. This composition mimics the recorded maximum levels of As and U in well water collected at the Pine Ridge Reservation. Six different mixtures of bone char and zero valent iron were created, and approximately 5g of each mixture were placed into 250 mL beakers containing 100 ml of contaminated water. The mass ratios of bone char to zero valent iron evaluated were: 100:0, 80:20, 70:30, 50:50, 30:70, 0:100.
The beakers were agitated on a shaker table at 400 rpm for the length of the experiments. Approximately 5 mL samples were taken after 5 minutes, 10 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, and 24 hours. The samples were filtered through 0.45 μm PTFE filters and analyzed. All experiments were run in duplicate, and results are shown in Figures 3 and 4.
Figure 3. The dependence of As(III) adsorption on the ratio of bone char to ZVI.
Figure 4. The dependence of U adsorption on the ratio of bone char to ZVI.
The results in Figure 3 show that having a higher ratio of zero valent iron to bone char leads to greater removal of As(III). When only bone char is present, As(III) removal drops in half. We note that in Phase I we found that bone char did not remove any As(III) from Pine Ridge Reservation groundwater. This suggests that something in the Pine Ridge groundwater inhibited As(III) adsorption to bone char. The results in Figure 3 also show that the rate of removal increases with increasing amounts of zero valent iron. Thus, the presence of zero valent iron is critical for efficient and complete arsenic removal.
We note that other research has found that zero valent iron is effective in removing As(III) via adsorption. Other research has also indicated that HCO3-, H4SiO4, and H2PO42- can interference with As(III) adsorption. Hence, it is possible that As(III) removal by zero valent iron from natural waters will be hindered by the presence of competing ions.
The results in Figure 4 show that U is effectively removed at all ratios of zero valent iron to bone char. However, the results also show that the kinetics of removal slow down for the three highest ratios of zero valent iron to bone char. However, the trend is not clear for these three samples, possibly due to experimental error. Regardless, the results indicate that bone char is important for uranium removal.
One side effect of using the zero valent iron is that it rusted and turned the aqueous solution a deep orange. The color was removed when samples were filtered for analysis, indicating particulate ferric iron. It is not clear if this color change will be an issue for a flow through filter.
Column Studies Analyzing Arsenic and Uranium Removal
This last spring and summer (2014), glass columns 1 cm in diameter and 1 cm in length were used instead of the stainless steel columns so that the uniformity of column packing could be directly observed. Two sets of column experiments were performed. In the first set, two columns were used to evaluate removal of 13.8 ppb As(III) in water. One column (Column A) contained 5.147 g of zero valent iron with a 1 minute retention time. The other column (Column B) contained 5.524 g of zero valent iron with a 15 minute retention time. The results are shown in Figure 5.
Figure 5. The removal of As(III) by ZVI in a column.
At both retention times, zero valent iron effectively removed As(III) to levels that were below the detection limit, even after 7 days of continuous flow. For the 1 minute retention time column, this represented over 10,000 pore volumes. For the 15 minute retention time column, this represented 670 pore volumes. The results suggest that As(III) uptake occurred fast relevant to the 1 minute retention time, and mass transfer limitations were not an issue.
In the second set of experiments, two columns were used to evaluate removal of 18.5 ppb As(III) and 125 ppb U in water. One column (Column C) contained 1.155 g of zero valent iron and 1.155 g of bone char, with a 1 minute retention time. The other column (Column D) contained 1.367 g of zero valent iron and 1.366 g of bone char, with a 15 minute retention time. The first column was run for 4 days before it shattered (due to a malfunction in the gripping mechanism of a clamp). The second column was run for 7 days. The results are shown in Figure 6 and 7.
Figure 6. The removal of U by bone char and ZVI in a column.
Figure 7. The removal of As(III) by bone char and ZVI in a column.
In the 1 minute retention time column, the effluent U concentration initially dropped, stayed below 5 ppb for 50 hours, and then started to increase (indicating initial breakthrough). The effluent As(III) concentration also dropped initially, but then started to rise almost immediately. In contrast, in the 15 minute retention time column, the effluent U and As(III) concentrations dropped and stayed below several ppb for the experiment duration. At the end of the column experiments, the effluent As(III) and U concentrations in the 1 minute retention time column were approximately 67% of their respective MCL values, whereas these concentrations in the 15 minute retention time column were less than 10% of MCL values. Approximately 672 pore volumes of water were treated in the 15 minute retention time column. For the prototype filter we are considering (i.e., 6" diameter, 12" length), and a typical water use rate of 10 gallons per day, treatment can be sustained for at least 72 days, and likely longer since breakthrough was not observed in the 15 minute retention time column experiment.
Analysis:
In order to calculate breakthrough times for the prototype column to be used for water treatment in homes, we calculated retardation factors for As(III) and U from laboratory column studies, and then used these to calculate breakthrough times of the prototype column assuming 10 gallons per day of water treatment and a retention time of at least 1 hour. We note that 10 gallons per day does not represent total water use in a home, but only the water used at the kitchen tap for drinking and cooking. To calculate retardation factors, we used the governing equation:
where V=volume of the column, C =the concentration of the contaminant, Q= the flow rate, 
where 
where 
The following values were used in the equations above to calculate a breakthrough time of roughly 180 days for the column that was running U contaminated water with a retention time of 30 min.
Table 1. Model parameters.
| Variable | Value | Units |
| | 549 | g BC/ L vessel |
| | 0.50 | L water/ L vessel |
| | 2.965 | L water/ g BC |
| | 1.136 | L water/ g BC |
| | 3257 | N/A |
| | 1248 | N/A |
| V | 3.93 | cm^3 |
| Q | 4.0 | mL/hour |
We used the calculated retardation factor, with the prototype flow rate and volume, in order to calculate breakthrough times for in-home use. These were in excess of two months, and in-home tests are needed to verify these results.
Life Cycle Assessment:
A life cycle assessment (LCA) was performed to compare environmental impacts of a bone char filter to those of a reverse osmosis filter. The latter is an accepted albeit expensive option to remove arsenic and uranium from contaminated groundwater. The LCA boundary condition was chosen to only include treatment for As and U, not other constituents in the water. Material inputs were considered, as well as consumables such as replacement bone char. The life cycle inventory data for each material (i.e., resources and emissions associated with extraction of raw materials, manufacturing, and transportation processes) were obtained from EcoInvent database (v2.2) implemented in SimaPro (v7.3; PRé Consultants; The Netherlands). The environmental impacts of each treatment technology were assessed using the Tool for the Reduction and Assessment of Chemicals and their environmental Impacts (TRACI) v2.0, which uses the following midpoint environmental impact categories: global warming (kg CO2-equivalent), acidification (H+ moles-eq.), human health (kg benzene-eq., kg toluene-eq., and kg PM2.5-eq. for carcinogenicity, non-carcinogenicity, and respiratory effects, respectively), eutrophication (kg N-eq.), ozone depletion (kg CFC-11-eq.), ecotoxicity (kg 2,4-dioxane-eq.), smog formation (g NOx-eq.), and fossil fuel depletion (MJ-surplus). Results are shown in Figure 8. The bone char filter outperformed RO filters in 8 of 9 categories after a projected 10 years of use.
Prototype Column System:
Figure 8. Life Cycle Assessment of a bone char filter versus a reverse osmosis fitler for heavy metal removal
Figure 9. Gravity fed column filter.
Based on our results to date, a simple design for a bone char / zero valent iron filter is a gravity fed system like the one shown in Figure 9. A similar filter design was used successfully in the Rift Valley region of eastern Africa where bone char is used for defluoridation of drinking water. Contaminated water is deposited into a reservoir above the filter and enters the filter column via the pipe at the top of the column. The water then flows through the piping to the bottom of the column, and then up through the bone char and zero valent iron mix. The water exits the filter through the small tube shown at the top on the right side
The filter can be constructed using PVC pipe that is 6" in diameter and 12" long. This reduces the cost to between $30 and $50 depending on the source of the bone char and zero valent iron materials. The piping size can be altered to change the flow rate and retention time. Although our column studies suggest that filter can effectively remove As and U for at least two months, further studies are needed with location specific groundwater to determine the frequency of replacement.
We initially performed batch studies with mixures of bone char and zero valent iron to remove both As(III) and U. Arsenic removal was more effective with the mixture than with bone char alone. We then focused on varying the ratio of zero valent iron to bone char in batch experiments to identify optimal removal conditions for both As(III) and U. These experiments showed that most ratios of zero valent iron and bone char are capable of effectively removing both As(III) and U.
Our next step was to evaluate As(III) and U removal in a flow through column, conditions which mimic the prototype filter we expected to be implemented. A column containing 70% bone char and 30% zero valent iron was tested with water containing 40 ppb U and no As(III) with a 15 minute retention time. The mixture of zero valent iron and bone char effectively removed U, indicating the iron did not negatively affect U removal. Finally, this last spring and summer (2014), reciprocating syringe pumps were purchased which allowed two sets of column experiments to be run. In the first set of experiments, both columns contained 100% zero valent iron and they were purged with water containing only As(III). One column had a 1 minute retention time and the other had a 15 minute retention time. In the second set of experiments, both columns contained a mix of 50% bone char and 50% zero valent iron (on a mass basis), and both were purged with water containing As(III) and U. Similar to the earlier experiment, one column had a 1 minute retention time and the other had a 15 minute retention time. Both As(III) and U removal were effective for the 15 minute retention time columns, but early breakthrough was observed for the 1 minute retention time column. These results were used to further optimize our prototype filter design.
Team members developed an educational program for elementary students in the Oglala Lakota tribe. The program focuses on water purification ideas and techniques, and can serve as a bridge for implementing actual reactor design on the Reservation. It also increases awareness in a nonintrusive manner. In order to ensure that the plans are appropriate for elementary students, the program materials were reviewed by a practicing primary education teacher.
To support our experimental efforts during Phase II, a UIUC undergraduate student team visited the Pine Ridge Reservation twice to collect additional water samples and survey the contamination. At the Reservation, samples were collected from 13 additional locations across the Reservation. Samples were collected from a variety of sources including indoor faucets, outdoor wells, and an outdoor pipe that was continuously flowing. All of these locations were of frequent use for the residents of the Pine Ridge Reservation, and covered a wide representative area.
While collecting samples, the team spoke with the residents and local Oglala Lakota College leaders on the reservation to discuss water quality concerns, and a prototype of the filter design. The latter helped the students understand if their perceived filter design would actually be of use in Reservation homes. The team collected valuable design feedback. They also discussed technical feasibility of bone char production on the Reservation with science leaders at the college.
Currently, one of our team members is continuing to engage the residents of the Pine Ridge Reservation. Serena Hou is both a member of KOLA, a group of masters of business administration students dedicated to economic development of the reservation, and a member of the Oglala Lakota Water Project. She worked with KOLA on the reservation in December 2013, and she is continuing to find more people on the reservation who may be interested in working on the business plan for the project and the implementation of the educational program mentioned above.
Conclusions:
Bone char filtration is capable of sufficiently removing U but not As from contaminated groundwater. Zero valent iron filtration is capable of sufficiently removing As from contaminated groundwater, even as the difficult to remove As(III). A mixture of the two sorbents is capable of sufficiently removing both contaminants from contaminated groundwater. However, additional work is needed to make the technology usable on the Pine Ridge Reservation. Specifically, additional tests are needed using Pine Ridge Reservation groundwater to determine the removal efficiency of arsenic and uranium over many weeks to months of continuous removal in a column system. Also, in-home tests are needed to determine if the prototype column effectively removes arsenic and uranium under real water use conditions, and to determine if the filters are easy to use and socially acceptable. Lastly, further work on the business plan is needed to identify a path forward to synthesize the bone char and assemble the column systems on the Reservations at a reasonable cost. These technical steps will need to be combined with community education and marketing to promote acceptance of the filters in the Pine Ridge Reservation community.
Journal Articles:
No journal articles submitted with this report: View all 4 publications for this projectSupplemental Keywords:
bone char, hydroxylapatite, hydroxyapatite, arsenic, uranium, human health, Oglala Lakota, Pine Ridge ReservationProgress and Final Reports:
Original AbstractP3 Phase I:
Use of Bone Char for the Removal of Arsenic and Uranium from Groundwater at the Pine Ridge Reservation | Final ReportThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.