https://orcid.org/0000-0002-3417-4143
Figures
Abstract
In rural agricultural regions characterized by historical fumigant use and in industrial areas, groundwater contamination by 1,2,3-trichloropropane (TCP) poses a significant environmental and health concern due to its potential as a carcinogen. This study evaluates the effectiveness of commercially available water pitchers equipped with carbon filters and almond biochar for point-of-use TCP treatment. The study found that the filters were able to remove TCP (>98%) from untreated groundwater during their lifespan, and different filter brands with varying flow rates showed no significant difference in TCP removal. These results suggest that these pitchers may provide a simple and efficient short-term solution. Furthermore, the study explored the feasibility of low-cost, locally sourced biochar derived from almond shells as a sustainable alternative to traditional carbon feedstocks. Batch isotherm tests, BET analysis, X-ray photoelectron spectroscopy (XPS) analysis, and scanning electron microscopy (SEM) imaging were used for biochar studies. The study found that the almond biochar used had a low surface area and total pore volume in comparison to commercial Granular Activated Carbons (GACs) and that more than half of the total area was composed of micropores (< 2 nm), while XPS surveys revealed the presence of Calcium, Phosphorus, and Potassium on the char’s surface. Finally, batch isotherm studies show that almond biochar exhibits lower TCP absorption efficiency compared to commercially available granulated carbon. However, further research into biochar produced under varied pyrolysis conditions is needed to determine its potential as a substitute for coconut shells. These findings can provide affected communities with information on efficient and cost-effective treatment technologies of TCP at the domestic well and household levels.
Citation: Hauptman BH, Harmon TC, Nasef Z, Rosales AA, Naughton CC (2024) Evaluation of point-of-use treatments and biochar to reduce 1,2,3-trichloropropane (TCP) contamination in drinking water. PLOS Water 3(7): e0000244. https://doi.org/10.1371/journal.pwat.0000244
Editor: Guillaume Wright, PLOS: Public Library of Science, UNITED KINGDOM
Received: December 5, 2023; Accepted: June 7, 2024; Published: July 5, 2024
Copyright: © 2024 Hauptman et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data that support the findings of this study are available through Dryad at: https://doi.org/doi:10.5061/dryad.d51c5b09f.
Funding: • The University of California Multicampus Research Program Initiative (Award #M21PR3417), CCN. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. • UC Merced Climate Action Grant, CCN, TCH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. • Rotary International District 5220 Scholarship Award for advanced research, BHH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. • The Achievement Rewards for College Scientists (ARCS) Foundation, BHH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. • The American Association of University Women (AAUW), BHH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
In 2010, the United Nations General Assembly passed a resolution recognizing the Human Right to Water and Sanitation [1]. However, in 2020 approximately one in four people lacked safe drinking water access in their homes [2]. Rural communities with limited resources and small water systems are more likely to have unsafe drinking water [3]. These communities often lack access to safe and affordable alternatives [4]. Moreover, rural communities located in or near agricultural areas face additional challenges, such as water contamination from pesticides and agricultural amendments [5].
The San Joaquin Valley (SJV) in California, USA is a prime example of this issue. The SJV’s eight counties are home to some of California’s lowest-income communities and the highest number of drinking water violations [6]. One of the most persistent chemical contaminants found in the SJV’s groundwater is 1,2,3-trichloropropane (TCP), a suspected human carcinogen [7]. TCP initially entered the ecosystem as an impurity in a soil fumigant, a pesticide employed to combat root-attacking nematodes [7], thus introducing a non-point source contamination issue in the regions in which it was applied. Furthermore, TCP had widespread industrial usage as a solvent and degreaser, resulting in point source releases from industrial operations [7].
In California, the maximum contaminant level (MCL) of TCP in drinking water has been set at a low 5 ng/L due to the potential risk of cancer [8]. In carcinogenic dose-response studies with rodents, the oral slope factor which measures the “incremental lifetime risk of cancer by oral intake of the chemical” for TCP is 30 mg/kg-day which is three orders of magnitude greater than the oral slope factor for similar chlorinated organic contaminants such as TCE (4.6 x 10−2 mg/kg-day) [9, 10]. About 8% of California’s domestic wells have TCP levels above the MCL compared to 5% of municipal wells [11].
Water utilities use granular activated carbon (GAC) as the best available treatment technology to remove TCP and achieve low MCLs [7]. Previous studies have investigated the viability of Granular Activated Carbon (GAC) treatment through Rapid Small-Scale Column Tests (RSSCTs) and batch isotherm experiments [12–14]. RSSCTs assess GAC’s contaminant removal efficiency with reduced water volumes and shorter contact times, while batch isotherm tests determine sorbate equilibrium concentration with GAC. Of the GACs tested in wells in Hawaii and California, a coal-derived carbon sold by Calgon called Filtrasorb 400 (F400) and Calgon Coconut Shell Carbon (OLC 12 × 40) sorbed the greatest amount of TCP before breakthrough to the MCL concentration (5 ng/L), 677 and 676 ng/kg, respectively [12, 14]. Despite all GAC types reducing TCP to the MCL, no single GAC proved universally effective across all water sources. Variability in GAC performance was attributed to variation in carbon feedstocks and the unique chemistry of source water, including organic matter presence [12, 14].
Carbon has been utilized for centuries to eliminate impurities and improve the taste and odor of drinking water [15]. Modern carbon filters are composed of finely ground particles of wood, coal, or nutshells that have been treated chemically or heated to create surfaces with high sorption capacity [16]. However, water consumers generally lack evidence-based information on the effectiveness of sustainable carbon sources and point-of-use treatments to remove TCP. To address this issue, a batch isotherm analysis was conducted in this study to determine the adsorptive capacity of biochar made from almond shells.
Biochar, similar to activated carbon (GAC), is cheaper and requires less energy to produce due to abundant biomass feedstocks, simpler pyrolysis production processes, and lower operating temperatures, making it a more cost-effective alternative for various adsorption purposes [17]. Multiple types of biochar were found to remove both organic and inorganic contaminants from water in a recent review [18]. Previous studies have shown that biochar derived from agricultural byproducts such as pecan and almond shells could remove volatile contaminants like benzene and chloroform, but did not test for TCP [16]. Although shell waste has been used for cogeneration facilities in the past, this practice is being phased out in California due to strict air quality regulations, resulting in increased shell mass in the waste stream. As approximately 80% of the world’s almonds are grown in California [19], using almond shell waste as a source of carbon to remove pesticide contaminants such as TCP from groundwater could be a more locally sustainable solution than importing coconut or coal-based carbon GAC feedstocks [13].
In addition to the lack of knowledge about the utility of using local nutshell-based biochar, there is a lack of knowledge concerning the efficacy of point-of-use (POU) filters in reducing TCP levels in drinking water. Tap water consumers in California’s Central Valley are concerned about the safety of their drinking water supply, specifically about the potential presence of pesticide contaminants, which is a grave concern for domestic well owners [20]. Since the Safe Drinking Water Act (SDWA) does not cover private wells, domestic well users are responsible for both arranging to monitor and for paying the treatment costs [21]. Moreover, a 2019 study by the USGS concluded that shallow private domestic wells have significantly higher TCP concentrations than public supply wells due to their proximity to agricultural fields and relatively shallow depths [11]. Using low-cost POU filters may empower domestic well users who are at an elevated risk for TCP exposure by allowing households to play a direct role in the safety of their water supply while reducing the reliance on costly water bottled in plastic [22].
There are several studies documented in the literature that used carbon-based POU filters for the removal of metals such as lead and arsenic and trace organic compounds [23–26]. A study by Brown et al. (2017), notes that POU filters may be effective to mitigate health risks from exposure to contaminants and that this may be especially true for substances not treated by typical water treatment approaches [27]. Both the NSF (National Sanitation Foundation) International and ANSI (American National Standards Institute) certify some POU filters in their ability to reduce volatile organic contaminants with an NSF/ANSI 53 certification. Since a contaminant reduction test method for 1,2,3-trichloropropane (TCP) has been added as NSF/ANSI 53 as of 2022, POU treatment certification for TCP may soon follow [28].
The overall goal of this research is to conduct a treatment analysis for TCP in drinking water supplies. Specific objectives include to evaluate: (1) the efficiency of pitcher point-of-use (P-POU) treatments for their ability to lower TCP levels and (2) almond shell-derived biochar for TCP sorption. This study’s results will inform communities, water utilities, and almond producers with updated treatment guidelines that prioritize safe drinking water at the household level while making use of agricultural byproducts and avoiding the need to import additional carbon from coal and coconuts for TCP extraction.
2. Materials and methods
The methods are divided into subsections for P-POU tests and biochar isotherm batch tests. The same source water and procedure for spiking groundwater was used for both tests. Source water was collected from the tap at the University of California, Merced in California (Fig 1) as non-potable, untreated groundwater. Samples were kept in high-density polyethylene (HDPE) containers at room temperature throughout the experiment. Water samples were spiked with TCP to a concentration close to 200 ppt. This concentration was chosen for comparison purposes, as it was also used in a 2014 study conducted by the University of California Davis, which tested GAC from coconut shells and coal to reduce TCP levels [14].
Sustainability Research and Engineering Building, University of California, Merced, California, USA. June, 2023.
To create a spiking solution, volatile organics are sometimes diluted in methanol. However, an alternative method to dilute TCP was used in this study that would better mimic how households use filters and avoid any possible chemical interference from methanol. A 120 mL glass amber bottle fitted with a mininert cap was filled with tap water then an air-tight Hamilton syringe was used to inject 1 ml of pure TCP into the bottle. Since TCP is a dense non-aqueous phase liquid (DNAPL) [29], the injected TCP sank to the bottom of the bottle where it was visible. The bottle remained undisturbed in a fume hood for 48 hours to ensure a saturated TCP solution. The theoretical concentration of the solution was 1.75 g/L TCP, which is the solubility of TCP at 20°C [30]. This saturated solution was then used to create the diluted solutions of approximately 0.2 μg/L needed for the P-POU and isotherm tests. For P-POU tests, the diluted solutions were stored in a 50-liter HDPE plastic carboy. For the biochar isotherm, spiked solutions were made directly in the glass reaction flasks. The protocol is published and can be found on the open-access platform protocols.io at the following link https://www.protocols.io/view/evaluation-of-point-of-use-treatments-and-biochar-c2hfyb3n.
2.1 Pitcher point-of-use filters
Three types of pitcher point-of-use (P-POU) water filters were evaluated for their ability to remove TCP from tap water (Table 1). Filters 1 and 2, from different brands, use activated carbon combined with cation exchange resins, while Pitcher 3 contains a cation-anion exchange resin along with an oxidation-reduction alloy [25], showcasing diverse filtration technologies. The manufacturer’s expected lifetime (MELs) for Pitcher filters 1 and 2 is 151 L, while for Pitcher 3, it is 85 L. Each filter was tested in triplicate.
Before testing, each P-POU device was pre-conditioned according to the manufacturer’s instructions (S1 Table). Spiked water was added to P-POU filters in volumes of one liter and allowed to equilibrate for 30–60 seconds between additions. The experiments were carried out over ten working days with no water being added to pitchers over the weekend. Tests were conducted at room temperature (21.5 °C) and experiments were conducted in triplicate for each filter type. Filtered samples were collected when the percentage of water passed through the filters equaled 0, 25, 50, 75, 100, and 125% of the MEL for the filter. Additionally, the approximate flow rates for each of the three filters were measured during the experiment. One-liter samples were poured into the filter and timed until they passed through.
Control samples were taken when the filtered water was sampled. The control group consisted of pitchers without filters with the same contact time with spiked groundwater as the experimental groups. Because the filters vary in terms of lifetime and performance, two criteria were used to compare the efficacy of the different P-POU devices to attenuate TCP levels: 1) the individual removal efficiency (IRE) defined as the removal of TCP at a given sampling point along the MEL, and 2) the lifetime individual removal efficiency (LIRE) defined as the average removal of TCP throughout the MEL (0–100% MEL). The removal efficiency percentage was calculated using the following equation where Cin represents the initial influent concentration and Cout represents the effluent concentration of TCP.
To determine if the plastic on the container walls adsorbed TCP, tests were conducted using three Teflon-lidded glass containers and three polyurethane containers filled with tap water spiked with TCP. Since drinking water cups, bottles, and storage vessels are commonly made from HDPE plastics, understanding if and to what extent TCP sorbs to HDPE is important since TCP could also desorb and remobilize from HDPE surfaces. Additionally, the pitchers used were all made of HDPE plastic, and the reported TCP removal percentages may be due, in part, to plastic sorption. The contact time was four days, after which each container was sampled and analyzed for TCP concentration. All effluent samples were capped with no headspace and sent to an environmental analytical laboratory certified to test for TCP at low levels called BSK Analytical Laboratory in Fresno, California, USA for analysis. An unpaired two-sample t-test [31] was used to compare the mean of water spiked with TCP to understand if there is a significant difference in the amount of TCP sorbed to the plastic walls of a sealed container compared to the walls of glass containers.
To determine if there are any significant differences in the removal efficiency among various filter types, a Shapiro normality test and a one-way ANOVA analysis were conducted using R software. The ANOVA test is parametric, requires the data to follow a normal distribution, allows for smaller sample sizes and is suitable for comparing three or more groups. Unpaired t-tests were also used to determine if there was a significant reduction in the concentration of the effluent from a specific filter compared to a control in which no filter was present.
2.2 Baseline water sample analysis
Sample water pH and conductivity were tested with a Mettler Toledo Seven Excellence S400 Benchtop pH Meter. Total dissolved solids concentration (TDS) was determined using the TDS meters provided with one of the tested pitcher systems. Sample water was analyzed for nitrate levels and total organic carbon (TOC) by BSK Analytical Laboratory in Fresno, California USA. Nitrate concentration and TOC concentration were determined using EPA method 300.0 and TOC method SM 5310C.
2.3 Almond biochar production and size sampling
The biochar was produced from almond shell feedstock by Dr. Hugh McLaughlin from NextChar for the University of California Merced. Biochar production occurred in a prototype continuous-feed rotary reactor called a mobile pyrolysis reactor. The temperature was kept at 350 °C with high residence time of approximately two hours in a slow pyrolysis configuration. Biochar samples were sampled for particle size using a WS Tyler model RX-29 sieve shaker. Biochar that passed through a US sieve size 18 (1 mm) but was retained on a US sieve size of 35 (500 μm) was used. This size range was selected because it corresponded with the median size of particles obtained from a P-POU GAC filter (S1 Fig). After selecting the appropriate size for the biochar, the sample was divided into smaller portions using a riffle splitter to ensure that the isotherm tests were conducted on representative samples.
2.4 Biochar physical characteristics
The physical characteristics of biochar were examined at the Imaging and Microscopy Facility (IMF) at UC Merced. The Zeiss Gemini 500 Field Emission scanning electron microscope (FE-SEM) is a high-resolution FE-SEM that was used to analyze the size and distribution of macropores on the surface of the biochar from multiple high-resolution images. The mean distance between macropores was determined using image J software developed by the National Institutes of Health from twelve different images [32]. A Thermo Scientific Nexsa G2 X-Ray Photoelectron Spectrometer (XPS) was used to determine which if any elements and functional groups are present on the surface of the char. Five different biochar samples were scanned for surface chemistry features. XPS scans were made in both survey mode and high-resolution mode. The XPS ‘survey’ setting scans for all surface elements while higher-resolution scans provide data at a narrower range of binding energies. The binding energies of photoelectrons emitted from the surface of a sample are used to identify elements present. Binding energy signals were identified using the Thermo Scientific Material Science Data System for XPS [33].
2.5 BET analysis
A Brunauer—Emmett—Teller (BET) analysis was carried out for the almond shell biochar using a surface analyzer (model Micromeritics Gemini VII 2390a) to produce N2 BET-specific surface area results. First, the sample was dried at 110°C for 2 hours prior to degassing. Degassing was performed with a Micromeritics FlowPrep 060 at 200°C for 6 hours. N2 gas was supplied as an adsorptive at 18 psi pressure using a two-stage regulator into a tube containing the sample and into a balance tube (static volumetric technique). Both tubes were submerged in cryogenic liquid N2 and kept under identical conditions to achieve isothermal conditions. In a Gemini VII 2390a, a separate servo valve coupled to a differential pressure transducer regulates the flow rate of analysis gas into the balance tube. The adsorption of the analysis gas onto the sample causes the pressure difference between the sample and balance tubes and is detected by the differential pressure transducer. The servo valve repeatedly adjusts for the pressure drop by allowing additional gas into the sample tube. Gemini VII Version 5.03 software provided the isotherm and BET surface area using user-defined points.
2.6 Isotherm adsorption of TCP using almond-derived carbon
Isotherms are commonly used to determine the capacity of a sorbent for a particular sorbate [34]. Tap water with TCP was mixed with various amounts of prepared almond shell biochar, ranging from 0.03 grams– 10 grams and allowed to equilibrate to determine the maximum amount of TCP removed at a given temperature. The quantity of contaminant adsorbed per unit mass of sorbent versus the concentration of contaminant remaining in solution at equilibrium is used to compare carbon sources and optimize adsorbent use. Results from this analysis were compared to a previous study of TCP sorption efficiency with coal and coconut shell-derived GACs [14]. For the batch isotherm tests, the American Society for Testing and Materials (ASTM) Standard Practice for Determination of Adsorptive Capacity of Activated Carbon by Aqueous Phase Isotherm Technique protocol was modified to minimize the loss of TCP via volatilization while handling, transferring, and storing solutions during the experiment [35]. These modifications include the use of filled glass bottles with no headspace and sealed with airtight Teflon caps, the agitation of glass bottles with glass beads for 48 hours, and substituting filtration with a 48-hour settling period following agitation. Additionally, the volatilization losses that do occur using control samples with no GAC are accounted for in removal calculations. The procedure was performed at room temperature (21.5°C).
The test flasks, caps, and glass beads were acid-washed (2 M HCl) and then allowed to dry completely. Selected carbon amounts were added to each flask, capped, and massed. An approximate range of carbon mass estimated from published pure solute isotherm data required to bracket the adsorption of the TCP concentration used in this study was determined by selecting masses from the lower, middle, and top suggested masses from the ASTM. Three glass beads were added to each flask to increase agitation and the bottles were massed again. TCP spiked water from the HDPE container was then delivered into each flask to the brim so that there is no headspace and immediately covered with a Teflon cap. The volume of the water in the flask was determined gravimetrically. The flasks were then placed on a platform agitator at 250 rpm for 48 hours and then allowed to settle for 48 hours. Solutions were then loaded into 40 mL amber vials containing a 1 M hydrochloric acid preservative, capped with no headspace, and refrigerated at 4 °C until transported on ice for analysis. Samples were analyzed at BSK Analytical Laboratory in Fresno, California, USA using the Department of Public Health’s Sanitation and Radiation Laboratories (SRL) method SRL524.2 liquid-chromatography and mass spectrometry [36].
2.7 Isotherm calculations
The chemical concentration of TCP which remains in solution after equilibration with biochar, and the mass of TCP adsorbed per unit mass of carbon (x/M) was determined. The mass of TCP adsorbed per unit mass of activated carbon was calculated by subtracting the final corrected effluent concentration using the equation:
The equation x/M represents the mass of TCP adsorbed per unit mass of activated carbon, expressed in milligrams per gram. Co stands for the initial TCP concentration in milligrams per liter, while Ce represents the TCP concentration in the solution after adsorption, also in milligrams per liter. V denotes the volume of water in the sample bottle, measured in liters, and M refers to the mass of activated carbon added per sample, measured in grams. Lastly, x represents the chemical mass adsorbed per sample, measured in milligrams.
The Freundlich isotherm was used to analyze the sorption data and determine the distribution of TCP between the adsorbed phase and solution phase at equilibrium. This model is ideal for biochar and other adsorbents with a non-uniform surface, as it considers various adsorption energy sites and multiple layers of adsorption [14]. The Freundlich isotherm equation is as follows:
The equation describes the equilibrium mass of adsorbate per mass of adsorbent, qe, expressed in micrograms per milligram. Kf stands for the adsorption capacity, measured in liters per milligram. Ce denotes the aqueous concentration of the adsorbate at equilibrium, measured in micrograms per liter. The parameter n represents the adsorption intensity, which is dimensionless.
3. Results
3.1 P-POU results
During the MEL, a slight decrease in the flow rates of Pitcher 1 and Pitcher 2 filters was observed as indicated in Fig 2 and S2 Table. Pitcher 3 had inconsistent flow rates due to frequent backups. To address this issue, the manufacturer’s recommendation was followed by shaking the filter contents vigorously between influent additions after removing the filter from the pitcher base. This approach temporarily improved the flow rate before it slowed down again (up to 2 hours). Despite efforts to enhance the flow rates through shaking, Pitcher 3 was unusable for the final assessment at 125% MEL since it took over an hour to drain one-liter additions after the 100% MEL.
Error bars represent standard deviation (n = 3).
Although individual removal percentages showed a slight lifetime decrease for Pitchers 1 and 2, and a greater decline for Pitcher 3, all filters still maintained relatively high removal rates at the end of their lifespan (S3 Table). At 125% of the MEL, Pitcher 1 still removed over 95% of TCP and Pitcher 2 removed over 90%. At 100% of the MEL (the last sampling point) Pitcher 3 removed over 85% (Fig 3). The mean percentage of TCP removed during the lifetime of all three filters exceeds 98% (Fig 4). After conducting an unpaired ANOVA test, no significant difference (p = 0.8) was found between the average removal percentages of the three filters.
Error bars represent one standard deviation from the mean (n = 3).
Error bars represent standard deviation (n = 3).
A cost analysis of various pitcher point-of-use (P-POU) devices was conducted, considering replacement filter costs and annual ownership expenses (Table 2). The calculation method for annual pitcher ownership costs factored in replacement filter expenses, assuming a family of four would use the filter for its entire Maximum Expected Lifetime (MEL). A daily water consumption rate of two liters per person was used to mirror realistic usage patterns. All P-POU devices had annual costs below $0.18 per liter of treated water. Pitcher 1 had the lowest annual expense at $153, closely followed by Pitcher 2 ($211). In contrast, Pitcher 3 had the highest annual ownership cost ($497) due to its lower filter lifetime volume (85 L) compared to Pitchers 1 and 2 (151 L), necessitating more frequent filter replacements.
3.2 Plastic sorption results
To assess the potential sorption of TCP onto plastic container walls, an unpaired two-sample t-test was employed to compare the mean concentrations of water spiked with TCP at a concentration of 200 ng/L stored in plastic and glass-sealed containers. The results revealed that the mean concentration of TCP at equilibrium in water sampled from plastic containers was approximately 19% lower than that in glass containers. Statistical analysis using an unpaired t-test demonstrated a significant difference in mean TCP concentration between the two groups (p = 0.0008) (Fig 5). Furthermore, analysis applying a linear isotherm model to the HDPE plastic equilibrium data indicated a distribution coefficient, Kd, value of 1.3 L/g (S4 Table and S2 Fig). This suggests a measurable sorption of TCP onto the walls of the plastic containers, indicating potential interactions between TCP and the plastic material.
Error bars represent standard deviation, with a sample size of n = 3.
3.3 Source water analysis
The pH of the source water used in this study was slightly basic, measuring 7.5, with a conductivity of 251 μs/cm at a temperature of 21.5°C. Nitrate levels were found to be near the federal maximum contaminant level, at 9.2 mg/L; the US Environmental Protection Agency (EPA) deems nitrate levels exceeding 10 mg/L (measured as NO3-N) in drinking water as harmful [37]. Total organic carbon (TOC) levels were detected at 0.215 mg/L, and the total dissolved solids measurement was recorded at 104 mg/L (S5 Table).
3.4 Almond biochar surface features and functional groups
The mean length of macropores on the char surface was measured from multiple SEM images at 400–1,000x magnification for seven different char fragments. The mean macropore length, 19 microns, was measured with Image J software with 25 macropores measured per sample (Fig 6). As the macropores are not circular, the largest diameter across any given pore was selected for the measurement. Differences between the groups, as indicated with a one-way ANOVA statistic, indicate significant (p<0.05) differences in the mean length of macropores from sample to sample (p<0.5).
Macropores (darker ovals) were measured to estimate the mean length of pores (n = 25) from seven different images.
Using XPS plots for Binding Energy (eV) on the x-axis and measured photoelectron counts on the y-axis, the following groups and elements were identified given the binding energies: C = O (a carbonyl group), C-O (alcohol or an ester), C-NH2 (amine), Calcium (Ca), and Potassium (K). XPS data are in the form of spectral plot data in either survey mode or high-resolution mode (Fig 7 and S4 Fig).
Five total scans were conducted. Scans for specific peaks seen in the survey scan (for the elements C, N, O, Ca, K, and P) can be found in the supplementary materials (S4 Fig).
3.5 BET analysis results
Table 3 shows the N2 BET-specific surface area and pore volume for the almond biochar used in the batch isotherm tests. The surface area value of 1.21 m2/g is low but the micropore area (less than 2 nm pore size) comprises 58.3% of the total area. The single-point adsorption total pore volume is also low at 0.00028 cm3/g. In a study of GAC adsorption using multiple commercial coal and coconut-based granular activated carbons, the surface area ranged from 780–1390 m2/g for coal and from 970–1050 for coconut; total pore volume ranged from 0.391–0.728 cm3/g for coal and 0.461–0.464 for coconut shell carbon GACs [38]. The N2 adsorption isotherm can be found in Fig 8.
3.6 Almond biochar TCP isotherm
Thorough testing of various sorption models, including linear, Langmuir, and Freundlich, was performed to evaluate their effectiveness in fitting this study’s data. The Freundlich isotherm provided the most suitable fit for this study’s results. Notably, this model is well-suited for heterogeneous sorbents [14]. The prior application of the Freundlich isotherm in a study involving granular activated carbon (GAC) and TCP [14] allowed for a meaningful comparison with the data in this study.
The isotherm data were plotted on a log-log scale (Fig 9). The TCP concentrations (μg/L) remaining in solution after equilibration with activated carbon (Ce) were plotted on the horizontal (x) axis; the masses of TCP adsorbed per unit mass of activated carbon (μg/g) were plotted on the vertical (y) axis. The resulting Kf value is 0.032 (Table 4).
The error band represents a 95% confidence interval.
A 95% confidence interval is denoted within parentheses as two standard deviations from the mean.
A comparison of the isotherm at low concentrations was conducted for the three sorbents shown in Fig 10 as complete experimental data collected for the almond shell biochar was not achieved due to challenges in making TCP solution concentration in the desired range due to the volatile nature of TCP. For solutions at 0.2 μg/L, Freundlich parameters were employed to estimate the distribution coefficient, Kd, representing the extent of TCP sorbed to the carbon relative to the TCP concentration in the solution. This measurement helps determine the affinity of a chemical for solid versus aqueous materials. The resulting Kd values for almond biochar, coal, and coconut were 0.029, 0.092, and 0.27 respectively. Notably, almond biochar exhibited an estimated Kd value one order of magnitude lower than that of coconut shell GACs.
4. Discussion
This study examined the efficacy of common P-POU filters and almond shell biochar for treating TCP-contaminated well water. The dearth of information available to households regarding the removal efficiency and associated costs of mitigating TCP at the household level underscored the need for this study. P-POU filters were effective at removing TCP and almond biochar emerged as a potentially sustainable and easily accessible alternative to conventional GAC feedstocks, particularly in the Central Valley of California.
This research tested the lifetime removal efficiency of three P-POU filters employing GAC. Across all three tested P-POU filters, the mean lifetime removal efficiency consistently exceeded 98% (Fig 4). This finding holds particular relevance for households reliant on domestic wells without filtration systems. Despite the cost-effectiveness of P-POU filters, it’s pertinent to note that larger families may encounter challenges due to slow flow rates (Fig 2 and S3 Table) and the expense of filter replacements. Nevertheless, filters present a more economical and environmentally friendly alternative to bottled water, as underscored by the substantial cost disparity between purchasing bottled water and the annual filter replacement cost for a family of four [39].
Moving forward, future research is recommended to include under-the-sink and refrigerator filters to ascertain their TCP removal as well as incorporate their annual replacement costs for comparative economic analysis. Additionally, while our cost analysis (Table 2) provides a basic cost overview, it does not include detailed economic or environmental assessments. Further research is encouraged, including life cycle assessments and cost comparisons with bottled water and under-the-sink filtration units. Furthermore, we recommend a comprehensive analysis of almond GAC on a broader municipal and household private-well scale to assess its cost-effectiveness and sustainability.
While carbon-based P-POU filters have demonstrated efficacy in reducing the concentration of various organic contaminants [16], it’s important to acknowledge their inherent limitations. These filters are not specifically engineered to remove bacteria or lower nitrate concentrations. Consequently, bacterial proliferation within the filter can occur, leading to heightened bacterial levels in the filtered water [40]. GAC filters are susceptible to biofilm formation, primarily due to the adsorption of microbes from water. Over time, these microbes adhere to the GAC surface, gradually forming biofilms [41]. Within the porous structure of activated carbon, microorganisms find ideal conditions for growth, fueled by the organic matter they absorb. Moreover, periods of water stagnation during regular use can exacerbate biofilm growth [41], especially when filtering water from private wells, which may harbor higher concentrations of microbial contaminants, including pathogens [23]. Biofilms pose a significant risk to water quality in POU Systems, as they can release bacteria if filters with attached biofilms are not replaced regularly, resulting in elevated bacterial concentrations in the filtered water. Additionally, the adsorption capacity of carbon point-of-use filters for nitrate and nitrite is relatively low [42]. Given that wells contaminated with TCP often also contain nitrates due to agricultural activities, P-POU filters may require supplementation with reverse osmosis or ion exchange units to effectively reduce nitrate and nitrite levels [42].
In the investigation of almond biochar batch tests, a TCP-biochar Kf value lower than that of coal and coconut shell GACs was noted. However, this variance is not unexpected given the absence of steam or carbon dioxide activation typically utilized in the production of commercial GACs. Refinement of almond biochar through chemical or physical activation shows potential for boosting its effectiveness, possibly positioning it as a feasible substitute or complement to commercial carbon GACs. Further evaluations, such as Rapid Small Scale Column Tests (RSSCT), akin to scaled-down pilot versions of carbon columns, would provide valuable insights into the efficiency of almond biochar as a carbon material in pitcher filters [17].
This study also highlighted the heterogeneity of almond biochar’s topography, suggesting further exploration into char production under varied conditions to identify optimal sorption profiles. Additionally, XPS analysis revealed surface characteristics facilitating intermolecular attraction between TCP and biochar, warranting investigation into alternative metal coatings such as zinc, known for its efficacy in TCP reduction [43]. Furthermore, the properties of biochar are contingent on various factors, including pyrolysis temperature and biomass feedstock characteristics. Continued research is needed to optimize biochar production processes and enhance its sorption capabilities to use as a sustainable solution for water treatment.
5. Conclusions
This study tested almond biochar and pitcher filters for the removal of TCP from groundwater spiked with TCP. The outcomes of this research can provide valuable insights to almond producers, utilities, and communities regarding household treatments while trying to minimize the reliance on imported carbon for TCP removal. The research objectives support UN Sustainability Goals 3 and 6 targeting safe and affordable drinking water for all and a reduction in illnesses from pollution.
Of the three pitcher filters tested, there was no significant difference between the different filters’ ability to remove TCP from sample groundwater. All filters maintained relatively high removal rates for the lifetime of their use. Given these results, future research is needed to determine if removal rates remain high with different drinking water sources such as those that have higher TOC levels, co-contaminant concentrations, and different water chemistry. This understanding will help target locations where filter pitchers may be considered for use by water consumers. Additionally, the results of this work should be considered by pitcher filter certifiers. Given the widespread contamination of drinking water sources by TCP, NSF/ANSI certification could be expanded to include TCP.
Batch tests indicate that almond biochar sorption of TCP is less effective than granular activated carbon sorbents made from coconut shells or coal. Future studies may use char made under different conditions, such as longer heat times and higher heating temperatures, to test whether this affects char sorption efficiency. Additional tests may also include coating the char with elements such as zinc, which has been shown in other studies to effectively degrade TCP via chemical reduction.
Supporting information
S1 Fig. Particle size analysis for A) Pitcher point-of-use (P-POU) filter material and B) Almond biochar samples.
https://doi.org/10.1371/journal.pwat.0000244.s001
(DOCX)
S2 Fig. Linear isotherm of an high-density polyethylene (HDPE) plastic container and TCP solutions at equilibrium.
Error band represents a 95% confidence interval. Performed at 21.5°C.
https://doi.org/10.1371/journal.pwat.0000244.s002
(DOCX)
S3 Fig. Almond biochar macropore length as visible with scanning electron microscope (SEM) (n = 25).
https://doi.org/10.1371/journal.pwat.0000244.s003
(DOCX)
S4 Fig.
A-J) X-ray photoelectron spectroscopy (XPS) survey scans indicating functional groups on the surface of almond biochar samples. In total five scans were conducted, f-j) Scans for specific peaks seen in survey scans for the elements C, N, O, Ca, K, P.
https://doi.org/10.1371/journal.pwat.0000244.s004
(DOCX)
S1 Table. Preparing pitcher point-of-use (P-POU) filters according to manufacturer’s recommendations.
https://doi.org/10.1371/journal.pwat.0000244.s005
(DOCX)
S2 Table. Flow rates in seconds for each filter type along different checkpoints of the manufacturer’s estimated filter lifetime (MEL).
https://doi.org/10.1371/journal.pwat.0000244.s006
(DOCX)
S3 Table. Percent reduction in 1,2,3-trichloropropane (TCP) concentration along various (0, 25, 50, 75, 100, 125%) manufacturer’s estimated lifetime checkpoints.
Percent reduction = (Cin − Cout)/Cin.
https://doi.org/10.1371/journal.pwat.0000244.s007
(DOCX)
S4 Table. 1,2,3-trichloropropane (TCP) removal in plastic containers and glass containers at equilibrium.
https://doi.org/10.1371/journal.pwat.0000244.s008
(DOCX)
S5 Table. pH, conductivity, nitrate as NO3-, total organic carbon (TOC), and total dissolved solids (TDS) of source water samples (n = 3).
https://doi.org/10.1371/journal.pwat.0000244.s009
(DOCX)
Acknowledgments
The Achievement Rewards for College Scientists (ARCS) Foundation, the American Association of University Women (AAUW), the Rotary International District 5220 Scholarship Award for Advanced Research, the University of California Merced Climate Action Grant, the University of California Multicampus Research Program Initiative Labor and Automation in California Agriculture (LACA) (Award #M21PR3417) provided financial support for this research. The authors thank UC Merced IMF lab which provided training for the SEM and XPS analysis, Dr. Gerardo Diaz for providing the biochar, and Drs. Sarina Ergas and Teamrat Ghezzehei for general discussions and guidance.
References
- 1.
United Nations General Assembly. Resolution 64/292: The human right to water and sanitation. 64th Session. 2010. https://documents-ddsny.un.org/doc/UNDOC/GEN/N09/479/35/PDF/-N0947935.pdf?OpenElement.
- 2.
Progress on household drinking water, sanitation, and hygiene 2000–2020: five years into the SDGs. Geneva: World Health Organization (WHO) https://data.unicef.org/resources/progress-on-household-drinking-water-sanitation-and-hygiene-2000-2020/.
- 3. Pace C, Balazs C, Bangia K, Depsky N, Renteria A, Morello-Frosch R, et al. Inequities in drinking water quality among domestic well communities and community water systems, California, 2011–2019. American Journal of Public Health. 2022; 112(1):88–97.
- 4.
London J, Fencl A, Watterson S, Jarin J, Aranda A, King A, et al. The Struggle for Water Justice in California’s San Joaquin Valley: A Focus on Disadvantaged Unincorporated Communities. Davis, CA: UC Davis Center for Regional Change. 2018. https://regionalchange.ucdavis.edu/report/the-struggle-for-water-justice
- 5.
Ravenscroft P, Lytton L. Seeing the Invisible: A Strategic Report on Groundwater Quality. World Bank, Washington, DC. 2022. https://openknowledge.worldbank.org/entities/publication/0737c4ba-14e6-59d3-b772-321a5e453481
- 6.
Balazs C. Just Water? Social Disparities in Nitrate Contaminated Drinking Water in California’s Central Valley [dissertation]. Berkeley, CA: University of California at Berkeley. 2010. https://digitalassets.lib.berkeley.edu/etd/ucb/text/Balazs_berkeley_0028E_11939.pdf.
- 7. Hauptman BH, Naughton CC. Legacy 1, 2, 3-trichloropropane contamination: a systematic review of treatments. Journal of Water, Sanitation and Hygiene for Development. 2021;11(4):515–534.
- 8.
State Water Resources Control Board (SWRCB). 2017. Division of Water Quality GAMA Program. Groundwater Information Sheet, 1,2,3-Trichloropropane (TCP). 2023. https://www.waterboards.ca.gov/drinking_water/certlic/drinkingwater/123TCP.html
- 9.
USEPA Integrated Risk Information System (IRIS). Chemical Assessment Summary National Center for Environmental Assessment. 1,2,3-trichloropropane. 2009. https://iris.epa.gov/static/pdfs/0200_summary.pdf
- 10.
USEPA Integrated Risk Information System (IRIS). Chemical Assessment Summary National Center for Environmental Assessment. Trichloroethylene. 2011. https://iris.epa.gov/static/pdfs/0199_summary.pdf
- 11. Burow KR, Floyd WD, Landon MK. Factors affecting 1,2,3-trichloropropane contamination in groundwater in California. Science of the Total Environment. 2019; 672:324–334. pmid:30959299
- 12.
Harada BK. Comparative evaluation of six different granular activated carbon for TCP removal using rapid small scale column test (Doctoral dissertation, [Honolulu]: [University of Hawaii at Manoa], [December 2014]). https://scholarspace.manoa.hawaii.edu/items/bcc73e59-a03b-489a-bb53-a280a246fd86
- 13. Babcock RW Jr, Harada BK, Lamichhane KM, Tsubota KT. Adsorption of 1, 2, 3-trichloropropane (TCP) to meet a MCL of 5 ppt. Environmental pollution. 2018; 233:910–5. pmid:29128248
- 14.
Mital J. Granular Activated Carbon Treatment of 1,2,3-Trichloropropane. University of California, Davis; 2014. https://www.waterboards.ca.gov/drinking_water/certlic/drinkingwater/documents/123-tcp/sbddw17_001/doc%2027.pdf
- 15.
National Research Council (NRC). Drinking water and health, Volume 2. In: Drinking Water and Health: An Evaluation of Activated Carbon for Drinking Water Treatment. The National Academies Press. 1980.
- 16. Bansode RR, Losso JN, Marshall WE, Rao RM, Portier RJ. Adsorption of volatile organic compounds by pecan shell-and almond shell-based granular activated carbons. Bioresource Technology. 2003; 90(2):175–184. pmid:12895561
- 17. Huggins TM, Haeger A, Biffinger JC, Ren ZJ. Granular biochar compared with activated carbon for wastewater treatment and resource recovery. Water Research. 2016; 94:225–232. pmid:26954576
- 18. Palansooriya KN, Yang Y, Tsang YF, Sarkar B, Hou D, Cao X, et al. Occurrence of contaminants in drinking water sources and the potential of biochar for water quality improvement: A review. Critical Reviews in Environmental Science and Technology. 2020; 50(6):549–611.
- 19. Kendall A, Marvinney E, Brodt S, Zhu W. Life cycle—based assessment of energy use and greenhouse gas emissions in almond production, part I: analytical framework and baseline results. Journal of Industrial Ecology. 2015; 19(6):1008–1018.
- 20. Fernandez-Bou AS, Ortiz-Partida JP, Dobbin KB, Flores-Landeros H, Bernacchi LA, Medellín-Azuara J. Underrepresented, underserved, understudied: gaps and opportunities for advancing justice in disadvantaged communities. 2021.
- 21.
U. S. Safe Drinking Water Act (SDWA) of 1996. Committee on Environment and Public Works. U.S. Government Printing Office. 1996. https://www.govinfo.gov/content/pkg/CPRT-106SPRT67528/pdf/CPRT-106SPRT67528.pdf.
- 22. Sobsey MD, Stauber CE, Casanova L, Brown JM, Elliott MA. Point of use household drinking water filtration: a practical, effective solution for providing sustained access to safe drinking water in the developing world. Environmental Science & Technology. 2008; 42(12):4261–4267. pmid:18605542
- 23. Mulhern R, MacDonald Gibson J. Under-Sink Activated Carbon Water Filters Effectively Remove Lead from Private Well Water for over Six Months. Water. 2020; 12(12):3584.
- 24. Gulson BL, Sheehan A, Giblin AM, Chiaradia M, Conradt B. The efficiency of removal of lead and other elements from domestic drinking waters using a bench-top water filter system. Science of the Total Environment. 1997; 196(3):205–216. pmid:9141715
- 25. Anumol T, Clarke BO, Merel S, Snyder SA. Point-of-use devices for attenuation of trace organic compounds in water. Journal-American Water Works Association. 2015; 107(9): E474–85.
- 26. Doré E, Formal C, Muhlen C, Williams D, Harmon S, Pham M, et al. Effectiveness of point-of-use and pitcher filters at removing lead phosphate nanoparticles from drinking water. Water Research. 2021; 201:117285. pmid:34107361
- 27. Brown KW, Gessesse B, Butler LJ, MacIntosh DL. Potential effectiveness of point-of-use filtration to address risks to drinking water in the United States. Environmental Health Insights. 2017; 11. pmid:29270018
- 28.
Andrew, R. 1,2,3-Trichloropropane Reduction Under NSF/ANSI 53 2022. [Cited 2023 Dec 4] https://wcponline.com/2022/03/14/123-trichloropropane-reduction-under-nsf-ansi-53/
- 29.
USEPA Technical Fact Sheet. 2014. https://19january2017snapshot.epa.gov/sites/production/files/2014-03/documents/ffrrofactsheet_contaminant_tcp_january2014_final.pdf
- 30.
USEPA CompTox Chemicals Dashboard. 2021. https://comptox.epa.gov/dashboard/.
- 31.
Helsel DR, Hirsch RM. Statistical methods in water resources. Elsevier; 1992. ISBN: 9780080875088.
- 32. Schneider CA, Rasband WS, & Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nature Methods. 2012; 9(7), 671–675. pmid:22930834
- 33.
Thermofisher. Table of Elements. 2021. [Cited 2023 December 14]. https://www.thermofisher.com/us/en/home/materials-science/learning-center/periodic-table/non-metal/carbon.html.
- 34. Brusseau ML, Chorover J. Chemical processes affecting contaminant transport and fate. In Environmental and Pollution Science. 2019; 8, 13–130. Academic Press.
- 35.
ASTM D3860-98(2020). Standard practice for determination of adsorptive capacity of activated carbon by aqueous phase isotherm technique. ASTM International, West Conshohocken, PA, USA. 2020.
- 36.
Okamoto HS, Steeber WR, Remoy J, Perera SK. Determination of 1, 2, 3-trichloropropane in drinking water by purge and trap gas chromatography/mass spectrometry. CDHS Division of Drinking Water and Environmental Management. Sanitation and Radiation Laboratories Branch. 2002. https://www.waterboards.ca.gov/drinking_water/certlic/drinkingwater/documents/123-tcp/tcp_by_pt_gcms.pdf.
- 37.
USEPA Regulated Drinking Water Contaminants: Inorganic Chemicals. https://www.epa.gov/ground-water-and-drinking-water/table-regulated-drinking-water-contaminants.
- 38. Crincoli K.R., Jones P.K. and Huling S.G., 2020. Fenton-driven oxidation of contaminant-spent granular activated carbon (GAC): GAC selection and implications. Science of The Total Environment, 734, p.139435. pmid:32470665
- 39. Cohen A, Ray I. The global risks of increasing reliance on bottled water. Nature Sustainability. 2018;1(7):327–329.
- 40. Wu C.C., Ghosh S., Martin K.J., Pinto A.J., Denef V.J., Olson T.M. et al., 2017. The microbial colonization of activated carbon block point-of-use (PoU) filters with and without chlorinated phenol disinfection by-products. Environmental Science: Water Research & Technology, 3(5), pp.830–843.
- 41. Shi X., Liu D., Chen L., Lin Y., Fu M.L., Sun W. et al., 2023. Challenges of point-of-use devices in purifying tap water: The growth of biofilm on filters and the formation of disinfection byproducts. Chemical Engineering Journal, 462, p.142235.
- 42. Mazarji M., Aminzadeh B., Baghdadi M. and Bhatnagar A., 2017. Removal of nitrate from aqueous solution using modified granular activated carbon. Journal of Molecular Liquids, 233, pp.139–148.
- 43. Sarathy V, Salter AJ, Nurmi JT, O’Brien Johnson G, Johnson RL, Tratnyek PG. (2010). Degradation of 1,2,3-trichloropropane (TCP): hydrolysis, elimination, and reduction by iron and zinc. Environmental Science & Technology. 44(2):787–793. pmid:20000732