Advanced Drinking Groundwater As Phytofiltration by the Hyperaccumulating Fern Pteris vittata

Abstract

The reuse of Pteris vittata plants for multiple phytofiltration cycles is a main issue to allow an efficient phytoremediation of arsenic (As)-contaminated groundwater. Here, we assessed the capacity of phytofiltration of P. vittata plants grown for two cycles on naturally As-contaminated drinking water (collected in Central Italy), spaced by a growth cycle on non-contaminated water (N cycle). P. vittata young plants, with extensive frond and root development, were suspended individually in 15 L of water with initial As of 59 µg/L, without any additional treatment or water refilling. During cycle 1, in 45 days P. vittata plants reduced As concentration below 10 µg/L, the allowed EU limits for drinking water. During the subsequent 30 day-N cycle on non-contaminated water, no leaching of As from the roots was observed, while the water pH increased 0.9 Units, but is within the allowed limits. During cycle 2, under the same conditions as cycle 1, As concentration decreased below 10 µg/L in less than seven days. These results show that P. vittata young plants, previously used for the phytofiltration of As, do not extrude As and, when reused, remove As much more rapidly. No additional treatments were required during phytofiltration and thus this represents a sustainable, efficient, and scalable strategy.

1. Introduction

Arsenic (As) is a toxic carcinogenic metalloid ubiquitously distributed in the environment [1]. Arsenic contamination could be due to natural geological events or to human industrial or agricultural activities and could affect both soil and water quality [2]. Arsenic contamination of drinking and irrigation water is an important threat to public health [3,4] and current European legislation imposes a limit of 10 µg/L for As concentration in water for human consumption. In several volcanic areas this limit is highly exceeded because of natural mineral release from soil to underground aquifers [5,6,7]. As a consequence of the excess of As concentration, in Viterbo and the surrounding area (Lazio, Italy), de-arsenification of drinking water is performed by means of filtering devices. In particular, the most used filtration systems used in this area are based on iron oxide adsorbents and constitute a high cost for the community (EGATO1-Vt sources). In fact, these absorbents work more efficiently in an acid environment and therefore water pre-treatment is required; in addition, adsorbents should be replaced cyclically. An alternative adsorption method could be the activated alumina, which is easy to manage and relatively low cost for the removal of arsenic and fluoride ions from water [8,9]. The downside is the possible release of aluminium into water.

Phytoremediation techniques, i.e., the use of plants to clean contaminated water and soils are gaining an increasing interest for As remediation of drinking water. Indeed, this environmentally sustainable strategy could be a valid solution upstream from the use of filters.

The fern Pteris vittata is a hyperaccumulator, which is capable of taking up and accumulating in fronds up to 22.000 mg/Kg As from the soil and thus can be used to clean up As-contaminated soils [10].

Increasing evidence suggests that P. vittata is able to grow hydroponically and to accumulate As when grown on contaminated water [11,12,13,14]. In addition, it was shown that P. vittata is more efficient than the non-hyperaccumulator fern Nephrolepis exaltata [15].

Hydroponical growth of P. vittata plants allows the development of the root system that is essential for As uptake [16,17,18,19]. It has been shown that adventitious and young roots, located in the first 10–20 cm of the underground tissue, seem to be responsible for active absorption in water, using low strength nutrition solution [12,14]. Most of these studies were conducted by growing P. vittata on water supplemented with As salts, or on As-contaminated water supplemented with other elements such as P and N [15,20,21,22]. In contrast, few hydroponic experiments were performed on water without further additions; in this case, water collected in Florida was probably anthropically polluted by the use of pesticides [17,18].

Limited information is available on possible effects of P. vittata growth on water pH. It was shown that water pH could change after P. vittata growth, as Stamps et al. [14] showed an increased pH value in hydroponic culture medium. However, no measurements were done on water pH after phytofiltration by P. vittata. This is a major issue for using phytofiltration on water for human consumption, as the pH value cannot exceed the allowed limits of 9.5 (98/83/CE).

Different groups investigated the possibility of reusing the same plants in multiple cycles and obtained opposite results: A decrease in As uptake during the second cycle was shown by Tu et al. [23] and Huang et al. [15], whereas an increased in As uptake during subsequent cycles was shown by Natarajan et al. [17,18,24].

The objective of this study, by utilizing As-contaminated drinking water collected in Nepi, a town near Viterbo (Lazio, Italy), was to evaluate: (i) The efficiency of P. vittata young plants to remove As during cycle 1; (ii) the release of As from P. vittata roots, and possible changes in pH values during the subsequent N cycle (on non-contaminated water); (iii) the rapidity of As uptake when the same P. vittata plants are reused for cycle 2.

2. Materials and Methods

2.1. Arsenic-Contaminated Groundwater

Naturally contaminated groundwater for the experiments, kindly provided by EGATO1-Vt, and Talete Spa (https://www.taletespa.eu/ (accessed on 30 May 2019)), was collected from Varano waterwell, located in Nepi (Viterbo, Lazio, Italy, 42°15’29.2” N 12°18’24.2” E, Figure 1) and was stored in polyethylene tanks at room temperature. Groundwater was collected twice and the As concentration was, respectively, 59.7 μg/L and 46.04 μg/L, while pH was 7.3 in both.

2.2. Plant Growth

A Chinese fern brake (Pteris vittata) collected from the Botanical Garden of La Tuscia University (Viterbo, Italy) was moved to a thermostated greenhouse for propagation [25]. In addition to natural sunlight irradiation, greenhouse illumination was supplied with 100 µmol/m²s white light lamps set to a photoperiod of 16 h light and 8 h dark. Greenhouse temperature and humidity were set to 28 °C and 70%, respectively. Mature spores were collected from fertile fronds, resuspended in water and sown in pots filled with soil and expanded clay and covered with a thin plastic film. Spores germinated and gametophytes arose after 4 weeks (Figure S1). The reproduction process occurred quickly and 4 weeks later the young sporophytes reached a height of about 2 cm, showing at least two small fronds. Young sporophytes were moved into individual pots with a soil mixture (pet/perlite/expanded clay) for about 8–10 weeks, without adding fertilizers. At this stage, all the sporophytes showed 4–6 fully developed leaves and an expanded radical system, and thus were moved to the hydroponic system.

Here, mature sporophytes were supplied with fertilizers and water electroconductivity (EC) was set to 2 (mS/cm), while water pH was naturally established at 8.4 by ferns themselves. In this condition, ferns needed about 4 weeks of hydroponic growth to be considered ready for subsequent experiments. EC values were also monitored during P. vittata growth on non-contaminated water with a portable Conductivity Meter (Milwaukee EC60 model, (Milwaukee Instruments, Inc, Szeged, Hungary) to check the mineral content of the water. Results show a stable EC during the first 14 days of growth on contaminated water with a slight but significant decrease after 2 days (Figure S2).

2.3. Hydroponic Growth Systems

For plant growth and phytofiltration experiments, we adopted two hydroponic growth systems (https://www.idroponica.it/ (accessed on 9 August 2021)):

• A Deep Water Culture (DWC) hydroponic system, based on a 15 L bucket housing a single fern and equipped with an air pump to oxygenate water (Figure 2A,B). This system was used for phytofiltration experiments.
• A Recirculating Deep Water Culture (RDWC) hydroponic system, composed by 4 individual deep water culture systems (DWC) connected together and with a central control tank that allows the simultaneous control of pH and nutrients (Figure 3A,B). The control tank is equipped with a water pump that moves the water constantly from the tank to each bucket (outward water flow) and back to the control tank (inward water flow); each bucket contains an air pump that increases water oxygenation. This system was used for sporophytes growth, prior to experiments.

2.4. Phytofiltration Experimental Set Up

A schematic workflow of phytofiltration experiments is shown in Figure 3C: Each fern was moved to a Deep Water Culture (DWC) hydroponic system composed by individual buckets provided with air pumps and filled with 15 L of naturally As-contaminated groundwater (Figure 3B,C, Supplementary Figure S1). No pre-treatment to adjust the pH conditions, neither fertilizers were added during the experiments. Evapotranspiration (ET) was evaluated and it was chosen not to be compensated. At the end of the first phytofiltration cycle, ferns were moved to uncontaminated and unfertilized water for 30 days for further analysis. Next, the same ferns were re-used for a second phytofiltration cycle, using a second batch of naturally As-contaminated groundwater.

2.5. Chemical Analysis

Water samples (50 mL) were collected before and during phytofiltration cycles at different times. They were then acidified to 1% with fuming nitric acid, and analyzed with a Perkin–Elmer NexION 300D ICP mass spectrometer (PerkinElmer Inc., Shelton, CT, USA) operating with an ESI autosampler (PerkinElmer Inc., Waltham, MA, USA) and an ultrasonic nebulizer (Teledyne Cetac, Omha, NE, USA) U6000AT+ Teledyne Cetac Technologies. ICP-MS operating conditions (that is, argon flowrates into nebulization chamber and torch for sample nebulization and plasma generation and containment as well as plasma generation potential, quadrupole and detector potentials) were optimized on a daily basis using a setup solution containing Be, Ce, Fe, In, Li, Mg, Pb and U at a concentration of 1 mg/L for each element and 1% of HNO₃. Quantitative determination of As was performed operating under standard mode conditions, without activating the collision/reaction cell, as the polyatomic interference (⁴⁰Ar³⁵Cl⁺) was not detected at the low chloride concentrations present in the examined samples.

Internal calibration method was applied for quantitative determination using a blank solution (HNO₃ 1%), five external standard solutions in the range of 1–50 µg/L and 20 µg/L ¹¹⁵Indium internal standard, the latter added to samples and external standards by peristaltic pump. Every standard solution was prepared by diluting the corresponding 1000-mg/L solution (AccuTrace Reference Standard) and acidified with extra-pure fuming nitric acid.

3. Results and Discussion

3.1. Hydroponic Growth Systems

To monitor the concentration of As in contaminated water over time, in the absence of P. vittata plants and in the presence of an air pump, preliminary analysis have been conducted. Samples of water at T0, 15 days and 30 days were collected and As concentration measured by ICP-MS. As shown in Supplementary Table S1 there was only a negligible decrease in As concentration over time.

Young ferns used for phytofiltration experiments were initially grown 4 months in soil and subsequently 4 weeks in the RDWC hydroponic system (Figure S1, Figure 3A,B), which allows to control and monitor ferns growth, exposing plants to the same water pH, nutrient density and oxygenation.

This ensures a more homogenous growth for all plants providing optimal conditions for phytofiltration experiments. Furthermore, the RDWC system is easily scalable. Pre-vious works have shown that the fern growth stage, defined as the number and age of the fronds and the length of the root system, is the most critical parameter for an efficient phytofiltration of As [12,15,17,18,23,24]. Hence in this study we focused on the homogeneity of plant growth prior to experiments, in order to increase the efficiency of phytofiltration cycles and obtain reproducible data.

3.2. Phytofiltration Cycle 1

Three young ferns (#1, #2, #3) which showed 6–8 fronds and a 20–25 cm long root system were chosen for the first cycle (cycle 1). Ferns #1, #2 and #3 were placed in single pots, filled with 15 L of naturally As-contaminated groundwater with an initial concentration of 59.705 µg/L and aerated by individual air pumps (DWC system, Figure 2A,B, Figure 4A). Water sampling was performed at different times and As concentration measured by ICP-MS. As shown in Figure 4B, ferns were able to reduce As concentration about 15%, 20%, 50% and 98%, in 24 h, 48 h, 30 days and 60 days, respectively. These results are in good agreement with the experiments conducted by Natarajan et al. [17,24] with hydroponically grown ferns, under similar phytofiltration conditions. The authors transferred two ferns per tank in 30 L of As-contaminated water (130 μg/L As) amended with modified 0.25 strength Hoagland’s solution and found that ferns needed 5 weeks to decrease As concentration below 10 µg/L. In contrast, further experiments performed by the authors used completely different phytofiltration conditions, such as 8 plants per tank in 45 L [18], while other authors added growth elements to contaminated water [15,22]. Our results suggest that young ferns grown in soil and subsequently for a short period in a low-nutrient RDWC hydroponic system, when transferred to naturally As-contaminated ground water, show an arsenic uptake similar to that obtained with hydroponically grown plants supplied with nutrients.

3.3. Arsenic Excretion and pH Changes during N Cycle Performed after Phytofiltration Cycle 1

After cycle 1, ferns were moved to buckets filled with non-contaminated water to evaluate two parameters: (i) Possible excretion of the accumulated As, (ii) water pH after P. vittata growth.

Several papers reported an efflux of arsenite and arsenate from root cells to the external medium, in different plant species, while little is known about As efflux from Pteris vittata [26,27]. Thus, to evaluate whether ferns used in cycle 1 could excrete the accumulated As, ferns #1, #2 and #3 were grown for one month in single pots, filled with 15 L of non-contaminated water and aerated by individual air pumps (DWC system, Figure 2A,B) and water sampling was performed at 0, 7, 15 and 30 days. As shown in Figure 5A, As concentration remains below 3.2 μg/L in all samples with no increase over time. These results show that ferns do not excrete or excrete negligible amounts of accumulated As.

As P. vittata is able to secrete hydroxide ions (OH⁻) and modifies pH values when grown in soils [28], we evaluated if this could also happen when ferns are grown in water. Thus, we measured water pH at different times, 0, 2, 4, 10 and 14 days. As shown in Figure 5B water pH increases from 7.3 to 8.4 in four days without any further subsequent increase. This is consistent with previous studies [14], which showed that P. vittata grown at different nutrient concentrations is able to modify and increase the pH value above 8 in hydroponic cultures. The limited increase in the pH value and the negligible release of As, following P. vittata growth, indicate that the ferns can be reused for subsequent phytofiltration cycles.

3.4. Phytofiltration Cycle 2

Conflicting results were obtained on the efficiency of As uptake for ferns reused in multiple phytofiltration cycles [15,18,19,20,21,22,23,24]. To investigate the capacity of reused ferns to uptake As, we used the same plants to perform a second phytofiltration cycle (cycle 2), under the same conditions as for cycle 1. At this time, ferns showed 10–12 fronds and a 30–35 cm long and expanded root system (Figure 6A). Ferns #1, #2 and #3 were placed in single pots, filled with 15 L of naturally As-contaminated groundwater and aerated by individual air pumps (DWC system, Figure 2A,B, Figure 6A,B). As shown in Figure 6B, ferns during cycle 2 removed 99.5% of As in less than seven days, indicating an increase in As uptake of about 45-fold, from cycle 1 to cycle 2. These results show that reused ferns are more efficient in As removal from natural contaminated water, in agreement with Natarajan [18,19,20,21,22,23,24], who reused ferns on water samples contaminated by pesticides. The increase in As uptake showed by reused ferns is possibly due to a larger size of the aerial and root system [23] or could be related to an increased capacity of As uptake due to the previous growth in the presence of As. In addition, our results show that to achieve an efficient and fast dearsenification for drinking water, the homogeneity of plant growth is an additional parameter to consider. This can be improved through advanced systems such as RDWC as well as the use of micro and macro elements before plants transfer to As-contaminated water.

4. Conclusions

Here we show that young P. vittata plants, grown in soil and subsequently for a short period in in a low-nutrient RDWC hydroponic, can be used for at least two cycles, resulting in a faster As uptake during the second cycle. Furthermore, we demonstrate that P. vittata, when transferred into non-contaminated water after cycle 1, does not excrete the absorbed As but causes a limited increase in water pH, possibly due to OH⁻ excretion. Altogether our results indicate that P. vittata phytofiltration is a promising technology to remove As from drinking water. Future studies aimed at optimizing P. vittata growth conditions will be necessary to increase initial efficiency in phytofiltration and allow the fern reuse in multiple consecutive cycles.