Extraction of Rare Earth Elements from Idaho-Sourced Soil Through Phytomining: A Case Study in Central Idaho, USA

Abstract

Environmentally friendly and low-emission extraction methods are needed to meet worldwide rare earth element (REE) demand. Within a greenhouse setting, this study aims to investigate the REE hyperaccumulation ability of four plant species (e.g., Phalaris arundinacea, Solanum nigrum, Phytolacca americana, and Brassica juncea) and the impact of amending REE-rich soil with biochar or fertilizer and watering with citric acid solution. Harvested samples were pyrolyzed, and the resulting bio-ores were acid-digested and underwent elemental analysis to determine REE content. Amending soil with fertilizer and biochar increased bio-ore production, while plant species explained the most variation in bioaccumulation factor. The results indicate that Phalaris arundinacea achieved the highest average REE concentration of 27,940 µg/g for the targeted REEs (comprising cerium, lanthanum, neodymium, praseodymium, and yttrium) and 37,844 µg/g for total REEs. It is also found that soil amendment and plant species are critical parameters in the design and implementation of Idaho-based REE phytomining operations. The life cycle assessment study estimated that the electricity demand of the greenhouse contributed the most to GHG emissions during the greenhouse study. Within the field study, electricity demand of the pyrolysis reactor was determined to be the largest producer of GHGs. The techno-economic analysis estimated that the total cost of growing P. arundinacea for six weeks on a one-acre field area is USD 6213, including 39%, 22%, 21%, and 18% of that cost derived from cultivation, biomass processing, soil treatment with fertilizer, and pyrolysis, respectively. It is concluded that the proposed low-emission extraction pathway, which combines phytomining, drying, and pyrolysis, is a promising sustainable approach for REE extraction, especially from REE-rich soil sourced in Idaho.

1. Introduction

Across the globe there is a growing demand for products and technologies that rely on the use of rare earth elements (REEs). China and the United States of America (USA) are at the forefront of REE-based technology development, with demand for the elements spanning across medical devices, aircraft components, and communication systems, as well as renewable energy, rechargeable batteries, and smartphones [1,2,3]. Mining of REEs within the USA is largely limited to a single California site, and most of the country’s supply is dependent on negotiations with China, which currently supplies the USA with most of the REEs it uses [4,5].

The traditional mining process results in undesired consequences such as environmental degradation and the production of toxic mining waste that is often not processed further [6]. In response to the adverse nature of conventional mining, novel methods have been developed to minimize impacts and improve sustainability. Electrokinetic mining, bioleaching, and phytomining are some approaches relevant to sustainable REE mining [7,8].

Idaho (a state in the northwestern USA) is endowed with a high abundance of REEs dispersed within soil that has not yet been targeted by any commercial mining operations [2,9]. REEs comprise over 1% (11,601 µg/g) of the total soil composition, especially across central Idaho in Lemhi Pass and Diamond Creek before crossing the Idaho–Montana state line (Supplementary Table S1). For comparison, soil from northern China contains only 0.024% (241 µg/g) REEs and regolith originating in the southeastern USA is 0.105% (1048 µg/g) REEs [10,11]. Idaho REE-rich soil can be an ideal candidate for phytomining, an alternative mining process used to recover REEs for processing and commercial use. Large-scale phytomining operations in this region can address national needs for rare earth metal (REM) production, along with job creation and other economic benefits for the region and nation. In addition, phytomining is a relatively cheap solution that can be integrated with traditional mining and improve soil quality for future crop production, forestation, and seeding of metal-intolerant native species to increase native biodiversity and avoid erosion [12,13].

REEs can enter vascular plants by forming complexes with organic macromolecules that facilitate adsorption to the root surface via transport proteins. They are transported via both symplastic and apoplastic pathways to the xylem, where they travel up the plant to the aboveground tissues via the transpiration stream and are then accumulated intracellularly, associating with multiple organelles and systems [14]. Hyperaccumulator plant species tolerate high concentrations of REEs and other metals in their cells without suffering from metal toxicity [14]. The bioaccumulation factor (BF) represents the concentration of target elements (e.g., REEs) in dried plant tissue versus the growth substrate (i.e., soil), with higher BF values indicating greater hyperaccumulation ability [15]. A plant is typically considered a hyperaccumulator when it displays a BF value greater than one [13]. These plants may be grown specifically for their REE storage abilities and then harvested and processed via pyrolysis (heating in the absence of oxygen) to convert plant biomass into bio-ore, rich in concentrated minerals, such as REEs [13]. The 91–99% of minerals retained in bio-ore after pyrolysis can then be extracted and refined for commercial use via organic acid leaching (bioleaching) and metallurgy [13,16,17,18].

Due to the wide variety of plant species, substrates, growing treatments, and targeted elements, it is not fully understood which combinations of these factors yield the best phytomining results. To help motivate and direct the design of our experiments, we reviewed past phytomining research (

Table 1. Overview of phytomining studies from the past two decades.

Study	Objective		Elements	Plant Species	Substrate Type	Treatment
	1	2
[19]	✗	✓	La, Ce, Pr, Nd	Dicropteris dichotoma	Soil from REE ore deposit	Histidine, citric acid, malic acid
[20]	✗	✓	Al, Fe, K, Ca, Mg, Mn Ni, Zn, Cr, Pb, Co, Cu, Cd	Plantago almogravensis	Podzol	-
[21]	✗	✓	Lanthanides (Atomic # 57-71)	Achillea millefolium, Artemisia vulgaris, Papaver rhoeas, Taraxacum officinale, Tripleurospermum inodorum	Roadside-sourced soil	-
[22]	✗	✓	Cd, Ce, La, Nd, Sr, Y	Trifolium pratense, Helianthus annuus	Phosphogypsum compost mix	Bacillus cereus
[23]	✗	✓	Lanthanides	Dicranopteris dichotoma	Soil from rare earth mines	-
[24]	✗	✓	Lanthanides	Phytolacca americana	Soil with REE mine tailings	Manure and sawdust mixture, biochar
[25]	✗	✓	Cu, Pb, Cr, Zn, Cd, Ni	Ludwigia stolonifera, Sphaeranthus gomphrenoides, Leersia hexandra, Commelina benghalensis, Sphaeranthus kirkii, Typha capensis, Cyperus articulantus, Fuirena umbellate, Agave sisalana, Cyperus exaltatus, Crinum papilosum, Hoslundia opposita, Pluchea dioscoridis, Hygrophylla auricultata, Ipomoea batata	Heavy metal contaminated soil	-
[26]	✗	✓	Sc, Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu	Phytolacca americana, P. acinose, P. clavigera, P. bogotensis, P. isosandra	Murashige and Skoog medium, quarter-strength Hoagland nutrient solution	H2SO4, REE tri-chloride salt
[27]	✗	✓	La, Y, Nd, Dy, Ce, Tb	Salix myrsinfolia, S. shwerinii	Hydroponically grown	REE-enriched tap water
[28]	✓	✓	As, Cd, Cu, Ni, Pb, Zn	Helianthus annuus	Soil with various heavy metal concentrations	none
[29]	✗	✓	Sc, Y, La, Ce, Pr, Nd, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu	Dicranopterus dichotoma, D. linearis, Melastoma malabathricum, Cyperus difformis, C. kyllingia. C. distans, C. rotundus	Soil from a former mining area	-
[30]	✓	✓	As, Cu, Mo, Ni, Zn, Re	Arundo donax, Tamarix ramosissima, Salsola kali, Cynodon dactylon, Chenopodium album, Atriplex leucoclada, Zygophyllum fabago	Soil with mine tailings	-
[31]	✗	✓	La, Ce, Nd, Sm, Eu, Tb, Yb, Lu, Sc	Citrus limonia	Commercial substrate	Superphosphate fertilizer
[32]	✓	✓	Lanthanides, Ge	Lupinus albus, Brassica napus, Zea mays	Soil containing trace REEs and Ge	Fertilizer, lime
This study	✓	✓	Ce, La, Nd, Y, Pr	P. arundinaceae, S. nigrum, P. americana, B. juncea	REE-rich Idaho-sourced soil	Fertilizer, biochar, citric acid

1: Economic; 2: Environmental; ✓: investigated the objective; ✗: did not investigate the objective; #: number.

).

In this study, we focused on the hyperaccumulation ability of four plant species due to their high BF values displayed during our primary experimental studies and in previous literature (Supplementary Table S2). The selected plant species are Phalaris arundinacea (common name reed canary grass), Solanum nigrum (black nightshade), Phytolacca americana (pokeweed), and Brassica juncea (brown mustard) [24,32,33,34] (

Table 2. Detailed summary of experimental design.

Objective	Evaluate REE Uptake Across Plant Species and Soil/Water Treatments
Species	Phalaris arundinacea, Solanum nigrum, Phytolacca americana, Brassica juncea
Experimental units	72 pots total, 18 per species, 3 replications of each species–soil–water combination
Soil treatment	Non-treated vs. fertilizer vs. biochar
Water treatment	Non-treated vs. citric acid
Greenhouse conditions	18–32 °C controlled temperature, 16:8 h light:dark cycle with minimum 308 µmol/m2/s light intensity
Duration	6 weeks from germination to harvest
Sample processing	Biomass washed, dried, pyrolyzed, and acid-digested
Sample analysis	Via ICP-MS
Elements analyzed	REEs including Ce, La, Nd, Pr, Y
Postanalysis evaluation	Life cycle analysis and techno-economic analysis

).

To better understand how different environmental conditions impact hyperaccumulator performance, each plant species was subject to control conditions, soil treatment with fertilizer or biochar, and water treatment with citric acid to determine the best combination with REE-rich soil substrate. Biomass then underwent drying and pyrolysis to concentrate REEs for elemental analysis and assess the effectiveness of these steps (Figure 1). Results from the experiment were used to conduct a life cycle assessment (LCA) and techno-economic assessment (TEA) to evaluate environmental impacts and monetary costs.

2. Materials and Methods

2.1. Soil Preparation

Soil was collected from the Diamond Creek project site (located near Salmon, Idaho) in the summer of 2022 (Figure 2). Rocks were removed from the soil with a #5 mesh (0.157 inches or 4 mm) and triturated into finer grains using a mixer and loose metal ball bearings. The resulting material was then homogenized back into the fine soil using the same mixer.

Table S1 provides the soil characteristics and elemental results (analytical methods detailed in Supplementary Table S3). Planting pots were filled with approximately 10.5 kg of soil and characterized using an X-ray fluorescence (XRF) spectrometer to determine and record the average composition of targeted REEs, which are REEs with the highest concentrations within the soil. The targeted REEs comprised cerium (Ce), lanthanum (La), neodymium (Nd), praseodymium (Pr), and yttrium (Y). The soil in each pot was either treated with 67 g of fertilizer, 525 g of biochar, or left non-treated (

Table 3. Averaged characteristics of collected soil from the Diamond Creek project site near Salmon, Idaho.

Parameters	Values	Parameters	Values
pH	7.6	Calcium (meq/100 g)	12.4
Sand %	47.3	Magnesium (meq/100 g)	3.4
Silt %	39.5	Sulfate-S (µg/g)	3.0
Clay %	13.2	Zinc (µg/g)	1.3
Salts (mmhos/cm)	0.6	Iron (µg/g)	9.2
Chlorides (µg/g)	7.0	Manganese (µg/g)	1.5
Sodium (meq/100 g)	0.2	Copper (µg/g)	0.1
CEC (meq/100 g)	16.3	Boron (µg/g)	0.2
Excess lime (%)	3.2	Cerium (µg/g)	2889
Organic matter (%)	2.8	Lanthanum (µg/g)	2071
Organic N (lb/Acre)	55.0	Neodymium (µg/g)	1690
Ammonium-N (µg/g)	2.6	Praseodymium (µg/g)	435
Nitrate-N (µg/g)	5.0	Yttrium (µg/g)	808
Phosphorus (µg/g)	5.0	Total mixed target REEs (µg/g)	7893
Potassium (µg/g)	97.0

).

2.2. Seed Preparation

Packaged, commercially available seeds of the four chosen plant species Phalaris arundinacea (reed canary grass), Solanum nigrum (black nightshade), Phytolacca americana (pokeweed), and Brassica juncea (brown mustard) were obtained and used in this study. Prior to sowing, seeds of P. americana were subjected to cold-stratification by storing in a humid environment at 5 °C for 3–4 weeks, followed by incubation in a growth chamber at 26.5 °C until germinated (approximately seven days). Seeds of B. juncea were similarly incubated in a growth chamber at 26.5 °C for seven days until germination. After preparing P. americana and B. juncea seeds, all four species were sown in the prepared pots. Each pot received 11 seeds or seedlings of the singular species randomly assigned to it.

2.3. Experimental Design, Growing, and Treatments

The experiment was established in a greenhouse as a randomized complete block design with three blocks (replications) and three factors: plant species (four levels: P. arundinacea, S. nigrum, P. americana, and B. juncea), soil treatment (three levels: non-treated, fertilizer-treated, and biochar-treated), and water treatment (two levels: non-treated and citric-acid-treated). Overall, 72 pots were prepared for planting with 18 assigned to each species. Each unique combination of species, soil treatment, and water treatment was replicated three times. Seeds were planted in pots containing REE-rich soil from the Diamond Creek site that was non-treated or treated with either Osmocote^® (a slow-release fertilizer with a nitrogen:phosphorus:potassium ratio of 14:14:14) or biochar (nutrient-rich pyrolyzed material from pinewood). The biochar used in this study was sourced from ponderosa pine (Pinus ponderosa) feedstock and had chemical composition C_cH_hN_nO_oS_s, pH of 8.47, was less than 100 microns in size, had nitrogen content of <1.75 μg/g, and phosphorous content of 1200 μg/g. More details about the basic properties of biochar are provided in [35]. Each pot was subjected to either non-treated tap water or tap water treated with 1% citric acid. Laboratory results indicated the water had a pH of 7.6 and hardness of 14.76 grains per gallon. Pots were bottom-watered (water poured into a tray beneath the pot) with 300 mL and top-watered (water poured on top of the soil in the pot) with 300 mL of the assigned water solution three times per week with at least one full day in between waterings. Plants were grown in a greenhouse that was consistently kept between 18 °C and 32 °C. Grow lights were used in conjunction with natural light to maintain a minimum of 308 µmol/m²/s for 16 h a day, followed by 8 h of darkness.

2.4. Plant Harvesting

After six weeks of growth, plants that survived were harvested. Individual plants were gently removed from pots by hand to recover as much underground root biomass as possible. Specimens were washed with tap water to remove all excess soil. Roots were separated from the aerial parts and gently agitated in an ultrasonic cleaner to ensure the removal of as many soil particles as possible. Following washing, the total biomass collected from each pot was weighed and recorded. Tissues were left to dry for 24 h in a 95 °C oven and then finely ground using a mortar and pestle. Plant biomass was weighed once dried to estimate initial moisture content.

2.5. Pyrolysis and Elemental Analysis

Dry biomass was placed into a small reactor and pyrolyzed at 400 °C for 30 min while nitrogen gas was pumped into the reaction chamber at 1.5–3.0 LPM to displace any oxygen gas. Three distinct products were formed in this process: pyrolysis oil, gas, and REE-rich solid (bio-ore). Pyrolysis oil was weighed but not collected or analyzed. Pyrolysis gas was released without being collected. Bio-ore samples were weighed and prepared using the United States EPA 3050B acid digestion method to concentrate REEs into a liquid suspension for detection [36]. The 5 mL resulting product was run through a 45-micrometer filter via vacuum filtration to remove solids and diluted with ultrapure water to 100 mL. The diluted product then underwent inductively coupled plasma mass spectrometry (ICP-MS) to detect concentrations of REEs that the plants in each pot took up collectively. Depending on the bio-ore amount, up to three digestions were made for each sample, and each digestion was analyzed twice to maximize measurement repetition.

2.6. Statistical Analysis

Once REE composition in µg/L was obtained via ICP-MS characterization, the results were converted to µg/g. The target REE bioaccumulation factor (BF) was calculated for each sample by dividing the total µg/g of targeted REEs in the dried plant tissue by the total µg/g of targeted REEs in the Diamond Creek site soil (Equation (1)). BF represents the efficiency of a species and treatment combination in extracting and storing the target REEs from the soil and retaining these materials through processing via pyrolysis, with higher numbers indicating higher extraction and retention ability.

$$Target REE Bioaccumulation Factor={\displaystyle \frac{mixed target REEs in dry plant tissue (\mathsf{\mu }\mathrm{g}/\mathrm{g})}{mixed target REEs in soil (\mathsf{\mu }\mathrm{g}/\mathrm{g})}}$$

(1)

Statistical analysis was performed, using R (an open-source programming language) [37]. BF was normalized and modeled as a normally distributed variable, while bio-ore mass was modeled as a gamma-distributed variable. A two-tailed analysis of variance (ANOVA) was performed to analyze the significance of the key parameters (i.e., plant species, soil treatment, and water treatment) on BF. In the case of bio-ore mass analysis, a two-tailed generalized linear model with a log link function was used. The significance level was drawn at α = 0.05 for both approaches.

2.7. Life Cycle Assessment

Two LCA studies were conducted to understand the environmental impacts and the most influential steps and factors of growing and processing the biomass of P. arundinacea for phytomining purposes. One study examines the impacts of the controlled environment greenhouse study, while the other considers a scaled-up theoretical outdoor field study one acre (1233.5 m³) in size. An LCA study consists of four distinct phases: goal and scope definition, life cycle inventory, life cycle impact assessment (LCIA), and interpretation. OpenLCA (version 2.4.0), an open-source program, was used for these analyses.

2.7.1. Goal and Scope Definition

The scope of both studies considered a gate-to-gate system boundary divided into four steps: (a) treatment of soil with fertilizer, (b) plant cultivation (planting seeds, growing plants, harvesting), (c) biomass processing (washing and drying), and (d) pyrolysis of biomass. The reference flow unit was defined as 1 kg of P. arundinacea-sourced REE-rich bio-ore for the greenhouse study and 460.9 kg for the scaled field study.

2.7.2. Life Cycle Inventory

Inputs and outputs for flows were sourced from OpenLCA databases, the AGRIBALYSE database, and data from this greenhouse phytomining experiment. When appropriate, values from the greenhouse analysis were scaled up by a factor of 460.9 for the field analysis.

2.7.3. Life Cycle Impact Assessment

The TRACI 2.1 baseline method, developed by the U.S. Environmental Protection Agency, was used to conduct the LCIA. Equations (2) and (3) were used by the program to calculate emission factors and GWP of the processes. Nomenclature is included in the Supplementary Materials section.

G_EF = ER_CO2 × G_EFCO2 + ER_CH4 × G_EFCH4 + ER_N2O × G_EFN2O

(2)

G_GWP = M_BC × G_EF

(3)

The main assumptions are as follows:

• The greenhouse climate control system and grow lights used 24,200 kWh and 64,500 kWh, respectively, for a total of 88,700 kWh of electricity consumed during controlled-environment cultivation.
• A diesel tractor is used to fertilize, sow seeds, and harvest P. arundinaceae biomass in the field study.
• A growth period of six weeks is enough for P. arundinacea to hyperaccumulate REEs optimally.
• P. arundinacea biomass is 90% water by weight and oven drying it for 24 h at 95 °C removes all possible moisture content.
• Biomass is pyrolyzed directly after drying (no grinding).
• Dried biomass weight is reduced by 50% when converted to bio-ore.
• Nitrogen gas pumped through the pyrolysis reaction chamber remains unchanged after it is emitted into the atmosphere.
• Emissions from pyrolysis of biomass at 350 °C have the following chemical contents: 69% CO₂, 25% CO, 2.5% H₂, 1% CH₄, and 2.5% other mixed hydrocarbon gases [38].
• All GHG emissions produced are emitted directly into the atmosphere.
• All electricity used was supplied by Western Electricity Coordinating Council (WECC) Northwest. Emission factors from this provider are as follows: 602.1 lb/MWh CO₂, 0.056 lb/MWh CH₄, and 0.008 lb/MWh N₂O [39].
• Energy used to power the biomass drying and pyrolysis processes was considered, but energy used to power the facility itself (e.g., lighting, heating) was not considered.
• Transport between the field and biomass processing and pyrolysis facilities was not considered.

2.7.4. Interpretation

Impact categories were evaluated by TRACI 2.1 within OpenLCA. Of all categories, global warming is by far the most affected by both the greenhouse and field studies. Results provide information on the environmental consequences associated with using phytomining to produce P. arundinacea bio-ore containing mixed REEs extracted from naturally REE-enriched soil over a single growing period.

2.8. Techno-Economic Assessment

The scaled field study was additionally explored through techno-economic assessment to examine economic viability if executed as a commercial process. The total cost of a six-week growing period was estimated using Equation (4) and includes the cost of materials, using Equation (5), and the cost of electricity and water utilities, using Equation (6).

C_AT = C_M + C_U

(4)

C_M = C_F + C_S + C_N + C_D

(5)

C_U = C_E + C_W

(6)

3. Results

3.1. Growing Success and Bio-Ore Mass Production

Out of the 18 individual pots planted per species, a pot was considered a success if enough biomass was produced to be harvested for pyrolysis and characterization (>0.1 g bio-ore). The overall success rates were as follows: 22% for P. arundinacea, 50% for P. americana, and 61% for S. nigrum and B. juncea (

Table 4. Summary of plant accumulation of REEs according to species and treatment.

Plant Species	Success Rate (out of 18)	Treatment #	Soil Treatment	Water Treatment	Avg Bio-ore Mass (g)	Ce	La	Nd	Pr	Y	Total Target REE	Total Mixed REE	Target REE BF
Phalaris arundinacea (reed canary grass)	22% (4)	1	Non-treated	Non-treated	0.3	2820	74	1567	420	1895	6676	9121	0.79
		2		1% CA	-	-	-	-	-	-	-	-	-
		3	Fertilizer	Non-treated	4.5	14,930	392	7236	1858	11,623	36,041	47,830	4.72
		4		1% CA	4.2	15,977	391	7800	2009	13,726	39,904	52,910	4.77
		5	Biochar	Non-treated	1.6	11,170	245	6460	1564	9100	29,138	41,513	4.40
		Average			2.7	11,224	276	5766	1463	9086	27,940	37,844	3.67
Solanum nigrum (black nightshade)	61% (11)	1	Non-treated	Non-treated	0.5	11,950	282	5245	1428	6020	24,924	32,309	2.6
		2		1% CA	-	-	-	-	-	-	-	-	-
		3	Fertilizer	Non-treated	3.6	2653	84	1390	367	2890	7384	9349	1.00
		4		1% CA	6.0	3362	78	1569	419	2521	7948	10,204	1.32
		5	Biochar	Non-treated	6.4	5017	104	2577	668	3322	11,690	16,072	1.20
		Average			4.1	5746	137	2695	721	3688	12,987	16,984	1.53
Phytolacca americana (pokeweed)	50% (9)	1	Non-treated	Non-treated	0.4	1731	30	1060	276	1467	4562	6165	0.53
		2		1% CA	7.4	1702	68	994	251	1579	4594	6163	0.61
		3	Fertilizer	Non-treated	1.1	507	0	277	76	769	1625	2072	0.19
		4		1% CA	7.1	1523	540	738	210	1487	4500	5429	0.60
		5	Biochar	Non-treated	0.8	1644	70	922	235	1091	3962	5382	0.57
		Average			3.4	1421	142	798	210	1279	3849	5042	0.5
Brassica juncea (brown mustard)	61% (11)	1	Non-treated	Non-treated	3.5	266	6	142	37	270	720	950	0.08
		2		1% CA	-	-	-	-	-	-	-	-	-
		3	Fertilizer	Non-treated	8.9	566	28	318	76	675	1666	2200	0.29
		4		1% CA	10.9	1099	28	553	147	980	2808	3629	0.51
		5	Biochar	Non-treated	3.7	288	64	163	41	284	840	1108	0.16
		Average			6.8	555	32	294	75	552	1508	1972	0.26

#: number; CA: citric acid; Target REE (i.e., Ce, La, Nd, Pr, Y) BF: bioaccumulation factor of target REEs (target REE concentration in dry plant tissue/target REE concentration in soil); dashes indicate missing data (due to no biomass or insufficient biomass production); plants were grown in pots with 10.5 kg REE-rich soil that was either non-treated, treated with slow-release fertilizer (67 g or 0.63%), or treated with biochar sourced from ponderosa pine feedstock (525 g or 5%); element concentrations are in µg/g bio-ore; Treatments: (1) non-treated soil and non-treated water, (2) non-treated soil and CA-treated water, (3) fertilizer-treated soil and non-treated water, (4) fertilizer-treated soil and CA-treated water, and (5) biochar-treated soil and non-treated water.

). Overall, 35 out of 72 pots were counted as successes. Multiple pots contained seeds that did not grow or failed to produce enough biomass for characterization. Fertilizer use was positively correlated with pot success rate (p < 0.007) and biomass (p < 0.005). Non-treated water was also associated with a higher success rate.

During the initial growing period, individuals from all species except for B. juncea successfully grew from seeds into mature plants in the greenhouse (Figure 3).

B. juncea individuals that had been planted as seeds directly into the 10.5 kg pots during the first round of planting did not germinate. It is assumed that the seeds did not receive sufficient warmth and light. To address this assumption, seeds were planted in growth trays containing the same soil type and stored in a warmer, brighter growth chamber before being transferred to 10.5 kg pots in the greenhouse, as detailed in the Seed Preparation subsection in the Materials and Methods section. This vastly improved germination rate and may have contributed to the species’ considerable biomass production.

Treatment of soil with fertilizer or biochar increased bio-ore yield in grams (p < 0.001) (Figure 4). Supplementary Table S4 presents the two-tailed generalized linear model results of bio-ore mass analysis. We found that soil treatment with fertilizer and water treatment with citric acid significantly affected bio-ore yield. Plant species did not significantly affect bio-ore mass production, except in the case of B. juncea. The REE concentration detected in the bio-ore did not considerably affect its production. Fertilizer-treated soil resulted in significantly higher bio-ore yields on average than the other soil treatments. Overall, B. juncea (brown mustard) produced the most bio-ore while P. arundinacea (reed canary grass) produced the least.

3.2. Hyperaccumulation Ability

Results indicate that P. arundinacea (reed canary grass) was the most effective hyperaccumulator of the four species tested, accumulating approximately 27,940 micrograms of REEs per gram of bio-ore (µg/g) of mixed targeted REEs (Ce, La, Nd, Pr, Y) and exhibiting an average BF of 3.67. Supplementary Table S5 provides plant accumulation of the targeted REEs, as well as the total accumulation of all present REEs. The second most effective hyperaccumulator was S. nigrum (black nightshade), accumulating approximately 12,987 µg/g mixed targeted REEs and displaying an average BF of 1.53. P. americana (BF of 0.5) and B. juncea (BF of 0.26) displayed weaker REE accumulation abilities.

Soil and water treatments increased bio-ore production. However, these factors do not appear to play a significant role in BF. The ability to accumulate REEs is notably different among species. Supplementary Table S6 provides the ANOVA result of BF analysis. Results from the ANOVA analysis performed indicated that species selection and the two-way interaction between species and soil treatment had significant effects on the uptake and storage of all target REEs except for La (Supplementary Table S7).

3.3. Life Cycle Assessment and Techno-Economic Analysis Results

Table 5. Environmental impact assessment results of the phytomining process in this greenhouse study and a theoretical scaled field study.

Impact Category	Unit	Greenhouse Study	Field Study
Global warming	kg CO2 eq.	69.11	14.28 × 102
Smog	kg O3 eq.	1.41 × 10−2	6.48
Respiratory effects	kg PM2.5 eq.	8.89 × 10−5	4.10 × 10−2

displays the environmental impacts of the greenhouse and field studies. Most impacted was the global warming potential, with lesser impacts in smog and respiratory effects. Impacts in all other categories (e.g., acidification, carcinogenics, non-carcinogenics, fossil fuel depletion, eutrophication, ozone depletion, and ecotoxicity) were indicated as negligible. The electricity demand of the greenhouse contributed the most to GHG emissions during the greenhouse study. Within the field study, electricity demand of the pyrolysis reactor was determined to be the largest producer of GHGs.

The total cost of growing P. arundinacea for six weeks on a one-acre field area is USD 6213, with 39% of that cost derived from cultivation, 22% from biomass processing, 21% from soil treatment with fertilizer, and 18% from pyrolysis. A breakdown of costs according to products and utilities is included in Supplementary Table S8.

4. Discussion

This study evaluated four plant species for their survival and ability to hyperaccumulate REEs (especially Ce, La, Nd, Pr, and Y) when grown in Idaho-sourced REE-rich soil. Plants were subjected to different soil (non-treated, fertilizer, and biochar) and water (non-treated and citric acid) treatments to additionally investigate the effects of differing conditions on bio-ore yield and bioaccumulation factor (BF). Fertilizer and biochar were selected as soil treatments due to evidence of their ability to promote plant tissue growth and increase hyperaccumulator yield via mobilization of target elements [32,40]. Citric acid was chosen as a water treatment based on previous studies that show that small doses enhance REE uptake via the root system, aid in mobilization for transfer from the roots to aboveground tissues, decrease plant stress, and boost biomass production [13,41,42]. Collected data were used to inform LCA and TEA studies.

Results show that plant species had a significant effect on BF, with Phalaris arundinacea displaying the highest value of all tested species. Brassica juncea produced the most bio-ore, which may have been due to the different germination conditions where seeds sprouted in a growth chamber rather than directly in the greenhouse pots. The insignificant impact of soil and water treatments on BF conflicts with the results of Turra et al., who found that fertilizer application and a decrease in pH (in this study via citric acid application) result in higher plant tissue REE concentration [31]. Regarding bio-ore mass production, soil and water treatments appeared to play a more significant role than species type in our study. To maximize the effectiveness and efficiency of a phytomining operation, it is important to find a balance between hyperaccumulation ability and bio-ore (and biomass) yield. A high BF is essential for extracting target REEs, but bio-ore yield must also be considerable as it is what ultimately determines how much of the resource will be recovered from the soil. Since REE hyperaccumulators are not well described as compared to those of other metals (e.g., plants are considered nickel hyperaccumulators when they concentrate at least 1000 µg/g in their dried tissues), this study acts as a valuable contribution to the limited existing body of knowledge [43]. The research into the impact of different treatments additionally aids in the understanding of how to increase bio-ore production during the growing process. Evaluation of the data through an LCA and TEA context provided information on the environmental impacts of the process and the costs that must be considered for a scaled-up operation.

Some of the obtained results in this study are comparable to the earlier studies in this field. Accumulation values from the experiments in this study are higher than those from Wiche et al. (2017) and Yuan at al. (2018), which reported 524 ng/g (0.524 μg/g) REEs in P. arundinacea tissue and 1040 mg/kg (1040 μg/g) REEs in P. americana leaves, respectively [44,45]. There is little evidence that suggests S. nigrum and B. juncea effectively accumulate REEs, with the former used primarily for cadmium extraction and the latter for gold recovery, and this is supported by the species’ poor performance in this study.

A major challenge presented in this experimental study was the failure of multiple plant individuals to germinate or to gain biomass after sprouting. Previous research has shown that an overabundance of metals in the soil can cause a decline in root development, interference with mitosis, and a decrease in biomass production, among other problems [46]. Soil pH that maintains plant health tends to fall between 6.2 and 6.8, and the REE-rich soil exhibits a far more basic average pH of 7.6, which could have impacted plant health, reducing biomass yield and survival rate [47]. Fertilizer significantly increased pot success rate and bio-ore mass production. These factors pose the question of what amendments (e.g., fertilizer, biochar, or citric acid) should be made to this substrate to make it more hospitable for future use.

5. Conclusions

This study demonstrates the potential of phytomining as a sustainable and environmentally friendly method that uses hyperaccumulator plants to extract rare earth elements (REEs) from Idaho-sourced, REE-rich surface soil and ore deposits. Our results show that soil amendments and plant species selection play a crucial role in determining the bioaccumulation factor and bio-ore production, with Phalaris arundinacea achieving the highest average REE concentration. This study highlights the importance of soil treatments and plant species when designing and implementing sustainable REE phytomining operations and the proposed low-emission extraction pathway of phytomining, drying, and pyrolysis offers a promising approach for REE extraction from Idaho-sourced soil. Further research is needed to explore the effectiveness of hyperaccumulator species and performance-enhancing treatments on this soil and other various substrates.