Journal of Geochemical Exploration 221 (2021) 106698

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

journal homepage: www.elsevier.com/locate/gexplo

Combined procedure of metal removal and recovery of technology elements

from fertilizer industry effluents
Ricardo Millán-Becerro *, Carlos R. Cánovas, Rafael Pérez-López, Francisco Macías, Rafael León
Department of Earth Sciences & Research Center on Natural Resources, Health and the Environment, University of Huelva, Campus ‘El Carmen’, 21071 Huelva, Spain

A R T I C L E I N F O A B S T R A C T

Keywords: This study focuses on the search for a sustainable treatment and metal recovery system for acid discharges from a
Alkaline treatment fertilizer industry in SW Spain. The methodology proposed involves the addition of two types of alkaline ma­
Phosphogypsum leachates terials (an industrial waste and a commercial reagent) to neutralize the acidity and remove dissolved elements. In
Pollutants removal
the first case, the treatment consisted on batch reactions between biomass ashes and phosphogypsum leachates at
Valuable metals
different solid-liquid ratios (i.e. 1:2.5, 1:5, and 1:10). On the other hand, a 0.01 M solution of Ca(OH)2 was used.
The experiment with biomass ashes at a solid:liquid ratio of 1:2.5 showed a high effectiveness, reaching removal
percentages close to 100% for F, Fe, Zn, Al, Cr, U, Cu and Cd. The depletion of contaminants from solutions
during the alkaline treatments occurred mainly by co-precipitation and/or adsorption onto phosphate phases, in
addition to precipitation of fluorides. Moreover, the solids precipitated during the alkaline treatments contain
elements of high economic interest such as rare earth elements plus Y (353–3992 mg/kg), Sc (21–164 mg/kg), Be
(5.0–7.0 mg/kg), V (1036–2974 mg/kg), Ga (16–40 mg/kg) or U (721–2963 mg/kg). These values could make
this by-product a promising source of technology metals. This study proposes an environmentally-friendly so­
lution for these industrial leachates, removing selectively impurities and target elements and producing a
promising exploitable metal concentrate. This research could lay the foundations for an effective and sustainable
treatment system for acid leachates from phosphogypsum stacks. Furthermore, the costs related with the
treatments could be covered by recovering the valuable elements contained in the newly-formed precipitates
during the treatments.

1. Introduction manufacturing process (Bolívar et al., 2009; Lottermoser, 2010). The PG

The production of phosphoric acid (H3PO4) in the phosphate fertil­
izer industry generates a waste known as phosphogypsum (PG, mainly
composed of gypsum, CaSO4⋅2H2O). This by-product is generated dur­
ing the wet chemical digestion of phosphate ore (i.e. fluorapatite,
Ca5(PO4)3F) with sulphuric acid (H2SO4). The industrial process is
summarized in the following reaction (Eq. (1)):
Ca5 (PO4 )3 F + 5H2 SO4 + 10H2 O→3H3 PO4 + 5CaSO4 ⋅2H2 O + HF
The raw phosphate ore contains high amounts of metallic impurities
and radionuclides from the decay series of 238U such as uranium, radium
and radon, which are released during the industrial process and subse­
quently transferred to the generated by-products, according to their
chemical behavior (Rutherford et al., 1994; Pérez-López et al., 2010;
Macías et al., 2017). The PG is highly acidic due to the presence of re­
sidual acids (i.e. phosphoric-sulphuric-hydrofluoric) from the
is usually piled in stacks near coastal areas exposed to weathering
conditions (Tayibi et al., 2009). Nevertheless, this is not the only waste
derived from the manufacture of phosphoric acid. The industrial process
requires a large volume of water, known as process water (PW), which is
employed for slurry and transport the PG to the stack in a closed-circuit
system. Thus, the continuous reuse of PW in the closed-loop system
yields a wastewater characterized by extremely low pH values and very
high concentrations of fluoride, sulfate, phosphate, radionuclides, and
(1)
metal/loids (Lottermoser, 2010). Such waters require storage and
isolation within the waste repository in central or lined ponds, and must
be neutralized before their discharge to the surrounding water bodies.
The neutralization process, for example with Ca(OH)2, implies that
contaminants pass from their soluble forms to an insoluble solid waste
composed of phosphates and to a lesser extent fluorides (Fuleihan and
Werner, 2011; Millán-Becerro et al., 2019). Therefore, this highly-
polluted sludge requires isolation, usually in the PG stack (Ericson

* Corresponding author.
E-mail address: ricardo.millan@dct.uhu.es (R. Millán-Becerro).

https://doi.org/10.1016/j.gexplo.2020.106698
Received 26 May 2020; Received in revised form 16 September 2020; Accepted 11 November 2020
Available online 14 November 2020
0375-6742/© 2020 Elsevier B.V. All rights reserved.
R. Millán-Becerro et al.
et al., 1997). However, PW may also contain elevated concentrations of
elements of high economic interest, so-called technology metals, such as
rare earth elements (REE), Y, Sc, U, Mo, V, Be or Ga (Cánovas et al.,
2017). The behavior of these elements during the neutralization is not
well known, and their presence in the newly-formed solid precipitates
may convert this sludge into a promising source of critical raw materials.
In this sense, there are authors (e.g. Cánovas et al., 2017) who highlight
the recovery of technology metals from PG leachates by the fertilizer
industry as a good option to achieve a sustainable and effective man­
agement of these wastes.
On the other hand, the use of conventional alkaline reagents such as
Ca(OH)2 (Fuleihan and Werner, 2011; Millán-Becerro et al., 2019) im­
plies not only high economic costs but also environmental impacts
associated with mineral exploitation and processing. Therefore, it is
necessary to search low-cost and environmentally-friendly alkaline
materials to replace the conventional alkaline reagents, improving the
sustainability of the treatment. These materials should be generated
close to the treatment plant in order to avoid environmental impacts and
costs associated with the transport. In this sense, this work investigates
the mobility of elements of economic interest (i.e. REE, Y, Sc, Be, V, Ga
and U) during the treatment of PW with two different types of alkaline
materials (i.e. low-cost alkaline waste and conventional reagent), the
latter for comparison purposes. Furthermore, the recovery of these ele­
ments of economic interest from the resulting solid precipitates was
evaluated, which could help to offset the costs associated with the
neutralization. As far as we know, this is the first attempt to assess the
solids generated during the alkaline treatment of highly acid and
polluted PG leachates as a possible source of critical raw materials. The
results obtained in this study could also serve to improve the sustain­
ability and effectiveness of the fertilizer industry worldwide.
2. Materials and methods
2.1. Study area
A singular PG stack is located in the estuary formed by the
Fig. 1.. Location map of the PG stack in the Estuary of Huelva.
2
Journal of Geochemical Exploration 221 (2021) 106698
confluence of the Tinto and Odiel rivers, close to the urban area of the
city of Huelva in southwestern Spain (geographical coordinates of
37◦ 16′ N; 6◦ 55′ W) (Fig. 1). The phosphate fertilizer industry has been
producing PG during almost 4 decades (from 1968 to 2010). As a result,
this stack accumulates around 100 Mt of PG wastes directly on 1200 ha
of marshlands in the estuary, less than 1 km away from the city center.
The PG stack also stores in a surface pond a huge volume of PW, around
10 Mm3, due to the implantation of a closed-circuit system for PW in
1997. The PW is characterized by extremely high acidity and mean
dissolved concentrations of elements such as PO4 (59,528 mg/L), F
(710 mg/L), Fe (909 mg/L), Zn (321 mg/L), Cr (195 mg/L), U (167 mg/
L), As (154 mg/L), Cu (64 mg/L), Cd (59 mg/L) and Ni (36 mg/L),
among others (Pérez-López et al., 2015). Presently, the stored PW is
subjected to evaporation as a measure to reduce its volume, which ex­
plains, along with its continuous reuse in the closed system, the extreme
levels of contamination. Therefore, it is urgently needed to treat theses
leachates, especially considering that this industrial water is discharged
uncontrollably to the estuarine environment through edge leakages
(Pérez-López et al., 2016; Papaslioti et al., 2018). A thorough descrip­
tion of the Huelva PG stack can be found in Pérez-López et al. (2016).
A promising alkaline material could be the biomass ashes (BA)
derived from the combustion of agroforestry biomass to produce elec­
tricity. These wastes are generated at high rates (aprox. 56,000 tons/yr)
in a nearby power plant (50 MW). Although the primary management
strategy for BA is their disposal in landfills, there are numerous studies
dealing with different potential applications for BA such as acid soil
amendment and fertilizing material in agriculture (Maeda et al., 2017;
Qin et al., 2017) or as alkaline reagent for the treatment of acid mine
drainages (Heviánková et al., 2014; Qureshi et al., 2016; Bogush et al.,
2019).
2.2. Sampling and alkaline mixing experiments
A sample of approximately of 5 L of PW was collected from the
surface pond of the PG stack in July 2019, when the highest concen­
trations of pollutants are found due to intense evaporation and the
R. Millán-Becerro et al.
absence of rainfalls. This acid leachate was taken manually with a sterile
60 mL-syringe and stored in a sterile and high-density polyethylene
bottle. The water sample was immediately put into a polystyrene ice box
and subsequently transferred to the laboratory, where the mixing ex­
periments were carried out within 24 h after sampling. Lab-scale
neutralization experiments were performed as a first step to optimize
an effective and sustainable treatment system at full scale. At industrial
scale, the neutralization of wastewaters is commonly accomplished on
raw waters rather than in filtered ones. For this reason, to simulate real
conditions, initial unfiltered solutions were employed in the
experiments.
The experiments involved the addition of two alkaline materials (a
low-cost waste and a conventional reagent) to neutralize the PW and
reduce the concentration of pollutants in solution. In the first case, the
experiments consisted of batch reactions between BA and PW at
different solid-liquid ratios (S:L) (i.e. 1:2.5, 1:5, and 1:10) with agitation
at 500 rpm on a shaker. The removal of elements was studied at different
reaction times: 5 and 15 min, 1, 3, 6, 12, 24, 48 and 96 h. The pH values
of the solutions resulting from the batch reactions ranged from 1.07 to
5.11. The BA was selected as alkaline reagent due to its high neutrali­
zation capacity and its low cost, since it is generated by a nearby in­
dustry (<10 km from the PG stack). On the other hand, Ca(OH)2 was
selected as this reagent is commonly employed to treat acid solutions on
industrial scale (Fuleihan and Werner, 2011). A 0.01 M solution of Ca
(OH)2 was added drop by drop and samples were collected at increasing
pH intervals of 0.5 up to reach a final pH of around 5. This pH value was
selected for comparison reasons between both alkaline treatments. A
detailed description of the neutralization experiments with Ca(OH)2 can
be seen in Millán-Becerro et al. (2019).
2.3. Analytical methodology
2.3.1. Liquid samples characterization
The pH, oxidation-reduction potential (ORP), electrical conductivity
(EC) and temperature values of solutions were measured with a CRISON
multi-parameter portable equipment MM 40+. A calibration at three-
points was performed for both EC and pH (147 μS/cm, 1413 μS/cm,
and 12.88 mS/cm and 4.01, 7.00, and 9.21, respectively), while ORP
was controlled using two different standard solutions (220 and 470 mV).
Values of ORP were corrected to obtain Eh values, referenced to the
standard hydrogen electrode (Nordstrom and Wilde, 1998). The solu­
tions were filtered at 0.45 μm with Teflon filters and divided into two
different aliquots; one non-acidified for the determination of anions and
another acidified to pH < 1 with ultrapure HNO3 (2%) for the determi­
nation of cations, and subsequently stored refrigerated until their
chemical analysis.
The concentration of anions in samples was analysed by high per­
formance liquid chromatography (HPLC) using a Metrohm 883 plus
basic ion chromatograph (IC) equipped with Metrosep columns. The
cations were determined in solutions by optical emission spectroscopy
with inductively coupled plasma (ICP-OES, Jobin Yvon Ultima 2) and by
mass spectrometry with inductively coupled plasma (ICP-MS; Agilent
7700) for major and trace elements, respectively. The analyses were
carried out in the Central Research Services of the University of Huelva.
The detection limits were: 0.2 mg/L for S; 0.1 mg/L for Na; 0.05 mg/L
for Fe, K and Mg; 0.02 mg/L for Al, Ca, Mn and P; and 0.1 μg/L for trace
elements. In the analyses of ICP-OES and ICP-MS, the regression co­
efficients for calibration curves exceeded 0.999. Three laboratory stan­
dards with concentrations within the range of the samples and blank
solutions with the same acid matrix were included in each analysis
sequence. In addition, dilutions from 1:10 to 1:200 were performed to
ensure that the concentration of samples was within the concentration
range of the standards. The electrical balance in samples was calculated
using PHREEQC 3.0 geochemical code (Parkhurst and Appelo, 2005),
being the average value below 5%. The quality of analysis has been
verified with a NIST-1640 certified reference material. A triplicate
3
Journal of Geochemical Exploration 221 (2021) 106698
analysis was performed in order to evaluate the precision, being better
than 5% in all cases.
2.3.2. Solid samples characterization
The solid samples (i.e. newly-formed precipitates obtained in the
experiments and the original BA) were mineralogically characterized by
X-ray diffraction (XRD), via a Bruker D8 Advance X-ray diffractometer
with Cu Kα radiation. Diffractometer settings were 20 mA, 40 kV and a
step size of 2◦ 2θ/min (3–65◦ ). In addition, the solids were examined by
scanning electron microscopy (SEM-EDS) with an environmental scan­
ning electron microscope coupled with a dispersive energy detector
(ESEM; FEI, QEMSCAN 650F). Samples were run at an accelerating
voltage of 5.00 kV for 120 s. The solids were imaged at 5–100 μm with
the SEM backscattered electron (BSE) detector.
Solid samples were also chemically characterized by digestion with
aqua regia as follows: 5 mL of aqua regia (12 mol/L HCl and 15.8 mol/L
HNO3 in a 3:1 ratio) were added to 0.5 g of sample in Teflon reactors,
allowed to react for 24 h in a gas hood, and then, on a heating plate for
2 h at 130 ◦ C. In addition, the BA was subjected to two international
standard leaching tests in order to classify it according to its potential
risk: the European standard leaching test (EN 12457-2, 2002) and the
USEPA TCLP test (US EPA, 1992).
2.3.3. Data treatment
The saturation indices (SI) of all solutions with respect minerals that
may control the solubility of elements during the experiments were
calculated with the PHREEQC 3.0 geochemical code (Parkhurst and
Appelo, 2005) using the thermodynamic database provided by Law­
rence Livermore National Laboratory (llnl.dat). Negative, zero, or pos­
itive SI values indicate undersaturation, equilibrium, or oversaturation
between the mineral phases and the solution, respectively. However,
these values indicate the tendency of a certain mineral to precipitate,
therefore the occurrence of this mineral was confirmed by mineralogical
evidences (i.e., SEM-EDS, XRD).
3. Results and discussion
3.1. Biomass ash characterization
The BA material used in this study corresponds to bottom ash, i.e. the
heavy residual fraction that remains at the bottom of the boiler after
combustion. The sample is constituted mainly by angular particles with
a median grain size between 250 and 500 μm, mainly composed of Al2O3
(11.3 wt%), CaO (8.45 wt%), K2O (4.83 wt%), Fe2O3 (4.69 wt%), MgO
(1.47 wt%), and to a lesser extent of other components such as Na
(2.49 g/kg), P (2.03 g/kg), Ba (0.21 g/kg), S and Sr (0.18 g/kg), among
others. The XRD patterns of BA only detected the presence of quartz, due
probably to the low crystallinity of other phases with respect to quartz
(Fig. S1a). According to the EU and US environmental regulations, the
BA must be considered as an inert waste since sulfate and metal/loids
concentrations in leachates did not exceed threshold values (Tables S1
and S2 of the Supporting Information).
3.2. Chemical evolution during the alkaline treatments
The PW collected for this study exhibited an extreme acidity
(pH 1.07) and dissolved solids content (EC of 54.4 mS/cm), with high
concentrations of pollutants: 65,220 mg/L of PO4, 5304 mg/L of SO4,
1134 mg/L of F, 116 mg/L of Fe, 105 mg/L of Zn, 60 mg/L of Al, 48 mg/
L of As, 48 mg/L of Cr, 33 mg/L of U, 19 mg/L of Cu and 14 mg/L of Cd.
These waters also contain high concentrations of elements of high eco­
nomic interest such as 66 mg/L of V, 334 μg/L of Be, 279 μg/L of Sc,
60 μg/L of Ga or REY (REE + Y); with concentrations of 743 μg/L and
737 μg/L of light (ΣLREE; La–Sm) and heavy rare earth elements
(ΣHREE; Eu–Lu, plus Y), respectively.
The neutralization with BA consisted on batch experiments with
R. Millán-Becerro et al.
solid to liquid ratios of 1:2.5, 1:5, and 1:10. The chemical composition of
solutions at each reaction time can be seen in Table S3. As can be seen in
Fig. 2, a gradual increase of pH was recorded until reaching values
around 3.1 (ratio 1:10), 4.2 (ratio 1:5) and 5.1 (ratio 1:2.5) at 96 h of
interaction time. The rise in pH values promoted a significant decrease
in the concentration of most dissolved pollutants (Fig. 2), a near total
removal of F, Fe and Cr was recorded for all ratios used. In addition, a
meaningful reduction in Al concentration was accomplished, despite an
initial increase due likely to the dissolution of Al2O3 contained in the BA,
reaching elimination percentages between 87 and 99% with respect to
the maximum concentration. However, different removal rates were
observed for U (removal percentages of around 58–100%), Zn
(20–100%), Cu (15–99%), Cd (13–92%) and PO4 (19–73%) (Fig. 2),
depending on the BA:PW ratio. Regarding SO4, the alkaline treatments
displayed removal values close to 50% for the ratios 1:10 and 1:5 at 96 h
of experiment time. In the case of ratio 1:2.5, SO4 concentration suffered
a strong initial decrease, reaching removal rates around 50% during the
first 15 min of treatment; however, this removal values decreases as the
increment of contact time between the PW and the BA. On the other
hand, As showed a more conservative behavior in solution, reaching
elimination values of only around 9–13% (Fig. 2). These removal per­
centages of dissolved contaminants are generally much higher than the
elimination rates obtained for the same pH values using Ca(OH)2
(Table S4). In fact, the mobility of some contaminants such as PO4, SO4,
F, As, Cd or Sb have hardly been altered during the treatment of PW with
Ca(OH)2 at pH values below 5 (Fig. 3). In addition, a slight reduce in
concentration of metals such as Zn and Cu was obtained, reaching
removal percentages of 22–37% at the end of treatment (Fig. 3). In this
sense, it is noteworthy that the removal values achieved during the PW
treatment with BA, despite not exceed pH values above 5.1, are similar
to those obtained during the treatment with Ca(OH)2 at pH values close
to 10 (Millán-Becerro et al., 2019). These remarkable results could be
due to the fact that during the alkaline treatment with BA, contaminants
can be intensively adsorbed both onto the surface of the BA grains and
Fig. 2.. Evolution of the concentration of main pollutants dissolved in the PW during the alkaline treatments with BA at different solid:liquid ratios.
4
Journal of Geochemical Exploration 221 (2021) 106698
on the newly-formed precipitates (Bogush et al., 2019).
Concerning the behavior of elements of high economic interest
during the treatment of PW with BA for a ratio 1:10, the removal per­
centages reached values close to 100% for REY after 15 min (pH ≈ 2.15),
100% for Ga (1 h, pH ≈ 2.34) and 100% for Sc (3 h, pH ≈ 2.46) (Fig. 4a,
b, c). In addition, other valuable metals such as Be and V reached
elimination percentages of around 60–69%, after 96 h of treatment and
at pH values close to 3 (Fig. 4c). The removal of these metals was favored
by increasing the solid proportion in the batch experiments. At a ratio of
1:5, the total removal of REE and Ga was observed 5 min after the
contact of BA with PW (Table S3) coinciding with pH values below 2.57.
A delay in the total removal was observed for Sc (at 15 min, pH < 2.88),
and Be (3 h, pH < 3.59). In the case of V, this metal was almost removed
from solution at 3 h (pH < 3.59), with removal values higher than 99%
for the rest of the experiment. However, the fastest removal for all ele­
ments took place by increasing the addition of BA until a ratio of 1:2.5;
REE, Be, Ga and Sc were totally removed during the first 5 min of the
batch experiment at pH values below 3.67, while the removal of V was
around 99% at this time (Table S3).
In comparison with the removal rates observed for BA, the neutral­
ization with Ca(OH)2 also achieved the total removal of elements of
economic interest such as REY, Be, Sc, and Ga from PW (Fig. 4d, e, f),
while the removal percentages of V were around 60% (Fig. 4f). Most REE
and Ga were completely removed at pH values close to 4, however Tm,
Eu (pH 1.48), Ho (pH 2.98), Be, Y and Sc (pH 4.99) deviated from this
trend. In the case of REY, the results obtained in this study provide a
different solubility pH range than others polluted acid waters, such as
acid mine drainage (AMD), where these elements seem to behave
conservatively at pH below 5 (Sahoo et al., 2012; Ayora et al., 2016).
3.3. Mineral saturation based on the thermodynamic modeling
According to the calculations performed with the PHREEQC code,
the solutions obtained during batch experiments with BA were
R. Millán-Becerro et al. Journal of Geochemical Exploration 221 (2021) 106698

Fig. 3.. Evolution of the concentration of main pollutants dissolved in the PW during the alkaline treatments with the solution of Ca(OH)2. Concentrations are
corrected according to dilution factor. The graphs show the relative concentrations of the contaminants in solution, which results from dividing the total volume of
solution (leachate + alkaline solution) by the starting volume of each leachate.

Fig. 4.. Evolution of the concentration of LREE, HREE plus Y and other valuable metals dissolved in the PW during the treatments with both alkaline materials; i.e.
with BA at solid:liquid (S:L) ratio of 1:10 (a, b, c), and with the solution of Ca(OH)2 (d, e, f) (concentrations are corrected according to dilution factor), respectively.

oversaturated (depending on ratio and reaction time) with respect to
several phosphate minerals (i.e. fluorapatite (Ca5(PO4)3F), strengite
(FePO4⋅2H2O) and berlinite (AlPO4)) (Fig. 5a, b, c; Tables S5, S6 and
S7). Specifically, for fluorapatite at ratios of 1:5 and 1:2.5 and reaction
times >3 h and 5 min, respectively (Fig. 5b and c), the undersaturation is
reached, which may be related with the strong removal of fluoride at
these steps of the experiments (Table S3). The precipitation of these
phosphate phases could also control the solubility of the remaining
contaminants such as Zn, Cr, U, Cu and Cd by co-precipitation and/or
adsorption processes. In this same sense, the decrease of dissolved REY,
Sc, Be, V and Ga concentrations in PW could be also attributed to the
precipitation of phosphate minerals (Alemrajabi et al., 2017; Jin et al.,
2019). On the other hand, gypsum (CaSO4⋅2H2O) was close to equilib­
rium during the experiments (Fig. 5a, b, c), which may have caused a
decrease in the dissolved concentrations of sulfate. Therefore, it could
not be ruled out that gypsum precipitation could also control the
mobility of some of these elements of high economic interest during the
experiments.
According to PHREEQC, the mobility of REY and other valuable el­
ements during the alkaline treatment with Ca(OH)2 could be also asso­
ciated with the precipitation of phosphate minerals. In fact, the solutions
obtained after Ca(OH)2 addition were oversaturated with respect to
strengite and fluorapatite (Fig. 5d). Moreover, these solutions were
oversaturated with respect to LREE phosphates (i.e. LREE-PO4⋅10H2O)
(Table S8). Nevertheless, other phases could be also controlling the
solubility of REE in the neutralization experiment with Ca(OH)2.
According to PHREEQC calculations, these solutions were oversaturated
with respect to LREE fluorides (i.e. LREE-F3⋅0.5H2O). In addition, fluo­
rite (CaF2) was oversaturated throughout the experiment (Fig. 5d).
Therefore, adsorption and/or co-precipitation processes of REY and
other elements of economic interest during the precipitation of fluorite
could not be ruled out (Fig. 5d). In fact, the role of fluorite precipitation
on REE solubility has been previously observed in other geological
matrices (e.g. Deng et al., 2014; Pingitore et al., 2014; Ayora et al.,
2016).
Overall, the newly-formed phosphate, fluoride and sulfate phases
could be acting as sinks of REE, controlling their mobility during the
neutralization experiments. In this sense, Cánovas et al. (2019) previ­
ously suggested that phosphate and fluoride minerals are probably the
main carriers of REE in PG, and instead of gypsum, based on leaching
experiments using commercial acids. These findings disagree with those
found by Borges et al. (2016), who claimed that sulfate minerals are the
main hosts of REE in PG.
3.4. Mineralogical characterization of the newly-formed phases
In order to confirm the data provided by PHREEQC, mineralogical
information was obtained. The mineralogical study by XRD showed that
the newly-formed solids during the alkaline treatment of PW with BA
had very low crystallinity (Fig. S1b). Thus, no minerals could be iden­
tified due to their amorphous or poorly crystalline lattices, with absence
of diffraction peaks. However, a detailed examination of the solid phases

5
R. Millán-Becerro et al. Journal of Geochemical Exploration 221 (2021) 106698

Fig. 5.. Saturation indices (SI) of the calculated oversaturated phases according to PHREEQC for the solutions derived from the alkaline treatments of PW with BA in
different solid:liquid (S:L) ratios; i.e. S:L ratio of 1:10 (a), 1:5 (b) and 1:2.5 (c), and with Ca(OH)2 (d).

by SEM-EDS revealed the existence of sub-rounded granular aggregates
composed mainly of P and Al, indicative of berlinite (Fig. 6), which was
predicted to precipitate by the PHREEQC code (Fig. 5a, b, c). In addition,
the EDS analysis detected the occurrence of other pollutants such as F,
Ca and Fe with P in the same newly-formed aggregates, suggesting the
precipitation of fluorapatite, fluorite and strengite, or co-precipitation
processes with other phosphate phases. Moreover, the SEM images
also showed the presence of tabular aggregates composed of sulfate and
calcium, typical of gypsum (Fig. 6). In this sense, P was abundantly
detected in gypsum particles suggesting co-precipitation with this sul­
fate mineral.
On the other hand, the mineralogical characterization by SEM per­
formed to the solids precipitated during the alkaline treatment of the PW
with Ca(OH)2 showed angular granular aggregates mainly composed of
P and metals such as iron and calcium (Fig. 7). In addition, the EDS
analysis showed also lower amounts of other elements such as Al, Mg, Cr
and V (Fig. 7). A detailed mineralogical study performed to these newly-
formed solids can be found in Millán-Becerro et al. (2019).
During the SEM-EDS examination, the presence of REE could not be
identified in the newly-formed solids probably due to the low amount of
REE precipitates with respect to phosphate phases.
3.5. Content of elements of high economic interest in precipitates
generated during alkaline treatments
As above commented, the neutralization of PW with BA and Ca(OH)2
successfully removed the elements of economic interest (i.e. REE, Y, Sc,
Be, V, Ga and U) contained in PW. Table 1 shows the concentration of

Fig. 6.. SEM images and EDS spectra of the newly-formed solids during the treatment of PW with BA.

6
R. Millán-Becerro et al. Journal of Geochemical Exploration 221 (2021) 106698

Fig. 7.. SEM images and EDS spectra of the newly-formed solids during the treatment of PW with Ca(OH)2.

total content of the valuable elements. Therefore, it is of critical

Table 1
importance to accurately adjust the neutralizing conditions (i.e. S:L
Estimation in mg/kg of the reserves of elements of high economic interest
ratio, amount of alkaline reagent, pH of solid separation) to optimize the
contained in the solids generated during the treatments with biomass ash (S-BA)
and with Ca(OH)2 (S-Ca(OH)2), and calculation of its economic value (MUSD). removal of target elements while reducing the level of impurities. As can
The metals valuation has been calculated according to the prices established by be seen in Table 1, the solids generated during the alkaline treatment of
the Shanghai Metals Market (SMM, 2020), the London Metal Exchange (LME, PW could be a great source of elements of high economic interest if
2020) and USGS (USGS, 2020). selective recovery schemes are developed to concentrate these target
S-BA S-Ca(OH)2 Price S-BA S-Ca(OH)2
elements. For comparison purposes, the sum of REY concentrations
(mg/kg) (mg/kg) (USD/kg) MUSD MUSD present in the S-BA (i.e. 353 mg/kg; Table S9) were of the same order of
magnitude as those found in other industrial wastes such as red mud
Elements of high economic interest
LREE 190 1935 31.1a 1.94 4.19 (789 mg/kg of REY; Borra et al., 2015), generated during the production
HREE 163 2057 5257a 43.5 117 of alumina, or fly ash (533 mg/kg of REY; Pan et al., 2018) derived from
Be 7.09 4.99 813 28.8 3.98 the combustion of coal in thermal power plants, which are considered
V 1036 2974 368 1906 1073 potential secondary sources of REY. The sum of concentrations of REY
Sc 20.9 164 986 103 158
Ga 15.7 40.0 137 10.8 5.37
found in S-Ca(OH)2 (i.e. 3992 mg/kg; Table S9) were from 5 to 7 times
U 721 2963 61.0 220 177 higher than the reported values by these authors.
a
Average values.
4. Environmental, economic and technical implications

the main elements of high economic interest in the solids generated
during the treatment of PW with both alkaline materials at pH 3. This
target pH value was selected based on two different criteria: a high
elimination of dissolved REY from PW and a low concentration of im­
purities. In addition, the newly-formed solids contain different amounts
of other target metals (i.e. Sc, Be, V, Ga or U) depending on the solubility
of each elements during the neutralization of PW (Tables S3 and S4).
Significant amounts of precipitates were collected at this pH value;
around 50 and 9.8 g of solid for each 1 L of PW treated with BA and Ca
(OH)2, respectively. However, given the difficulty of separating the
newly-formed solids during the alkaline treatment with the BA from the
original solid waste, it could not be ruled out that part of the solid
recovered after treatment was unreacted alkaline waste used in the
process. As can be seen in Table 1, the newly-formed solids after the BA
treatment (S-BA) contain high concentrations of elements of high eco­
nomic interest such as V (1036 mg/kg), U (721 mg/kg), Sc (21 mg/kg),
LREEs and HREEs (sum of values of 190 mg/kg and 163 mg/kg,
respectively), among others. However, the concentrations in the solids
originated during the Ca(OH)2 treatment (S-Ca(OH)2) were between 3
and 12 times greater than the values found in S-BA for these valuable
elements (e.g. 2974 mg/kg of V, 2963 mg/kg of U, 164 mg/kg of Sc,
1935 mg/kg of LREE and 2057 mg/kg of HREE respectively; Table 1).
This could be due to the difficulty in separating the solids precipitated
during the BA treatment, which contain elements of high economic in­
terest, from the inert waste (which may not retain these elements).
Therefore, the alkaline material would be causing a dilution effect on the
PW stored in PG stacks exhibit extreme acidity and metal concen­
trations, and in some cases like the Huelva PG stack, this water is an
important vector of pollution in water bodies (Pérez-López et al., 2016).
In this sense, this study proposes an environmentally-friendly solution
for these leachates, removing selectively impurities and target elements
and producing a promising exploitable metal concentrate. In addition,
this procedure follows the guidelines proposed for a circular economy.
On the one hand, a waste stream from the fertilizer industry (i.e. process
waters) is used as source of elements of economic interest (i.e. REE, Y,
Sc, Be, V, Ga and U). On the other hand, the ashes generated in a biomass
power plant can be used as raw material to treat highly-contaminating
PG leachates. These synergies between different industrial wastes or
by-products could be extended to other areas worldwide affected by PG
deposits due to growing production of BA in the world. In this sense, the
BA generation worldwide was estimated at 480 Mt/yr (Vassilev et al.,
2013). In the case of the PG stack of Huelva, the PW reservoirs stored on
the piles have a huge volume with around 10 Mm3. According to the
results obtained in this study, around 4 Mt of BA would be necessary to
effectively treat the total volume of PW, although the costs associated
with this treatment may be primarily due to the transport from the
nearby power plant. Considering the use of Ca(OH)2, around 60 g would
be required for each 1 L of PW. That is, considering a market price of
60–100 USD/t (INAP, 2020), the economic cost associated with the
purchase of Ca(OH)2 could range between 36 and 60 MUSD to treat all
the PW stored in the PG stack. In addition to these already high costs, the
costs associated with the construction and maintenance of an active

7
R. Millán-Becerro et al.
treatment plant should be also considered.
Table 1 shows an estimation of the value of the reserves of elements
of high economic interest contained in the newly-formed precipitates.
For this purpose, the prices of the target elements published by the
London Metal Exchange, the Shanghai Metals Market and USGS (LME,
2020; SMM, 2020; USGS, 2020) have been used (Table 1). Therefore,
considering the amount of target metals in the solids and their market
values, a total economic value of 4628 and 15,699 USD/t could be
estimated for the S-BA and S-Ca(OH)2, respectively. Considering the
potential tonnage of precipitated solids after the treatment of all PW
stored in the stack, these metals would have a total economic value of
2314 and 1539 MUSD for the S-BA and S-Ca(OH)2, respectively. The
elements of highest economic importance in the solids (i.e. S-BA and S-
Ca(OH)2) were V (with a valuation of 1906 and 1073 MUSD, respec­
tively), U (220 and 177 MUSD) and Sc (103 and 158 MUSD). Lower
valuations were calculated for other elements such as HREE (43.5 and
117 MUSD), Be (28.8 and 3.98 MUSD) and Ga (10.8 and 5.37 MUSD),
among others (Table 1). However, these figures could not be considered
as realistic, because market prices for high purity elements and total
recovery of elements were considered. However, some reported cases
(Smith et al., 2013; Kulczycka et al., 2016) claimed that a correction
factor needs to be applied on these valuations. This correction factor
may vary depending on certain criteria associated with the reserves or
their subsequent recovery, such as purity, quantity, technical limitations
of extraction, availability of necessary industry and economic factors,
among others. Therefore, other studies would be necessary to develop
more realistic estimations.
In this sense, a key point in the cost-effective recovery of these
valuable metals from the solids generated in the treatments is the
development of selective recovery methods for each element. In this
study, mineralogical and chemical evidences suggest that these valuable
metals are incorporated in phosphates. In some cases, like for REE, the
recovery processes of metals from phosphate have been extensively
studied. For instance, Sadri et al. (2017) reviewed different procedures
to recover REE from different mineral concentrates. A feasible solution
for the treatment precipitates would be the dissolution of REE in a weak
acid solution and subsequently the recovery of REE by neutralization,
precipitation or solvent extraction methods. A technical barrier to
recover REE from these minerals concentrates is the low selectivity of
processes, where Ca, Fe or P may lead to further problems during the
obtaining a saleable metal concentrate. However, the occurrence of
these impurities may be not an insuperable obstacle as reported by
Alemrajabi et al. (2019), who optimized a REE extraction method from a
matrix quite similar to that found in our study (Al, Fe, Ca and REE
phosphates). For this reason, more detailed studies about metal recovery
from PG leachates should be performed in the fertilizer industry.
5. Conclusions
This paper proposes the use of BA to treat leachates from a PG stack,
with the aim of designing an effective treatment system that is eco­
nomic- and environmentally-friendly. In addition, the solids generated
during this treatment were explored as a potential source of valuable
elements.
In general, the treatment of PG waters with BA generated an
improvement in the chemical quality of these highly polluted waters,
being the solid:liquid ratio of 1:2.5 the most efficient in the removal of
dissolved contaminants, reaching removal percentages close to 100% for
F, Fe, Zn, Al, Cr, U, Cu and Cd. However, treatment with the alkaline
waste was not as effective for other contaminants such as PO4 (removal
of 73%), SO4 (55%) and As (13%). On the other hand, this treatment
achieved a high removal of the elements of high economic interest (i.e.
REE, Y, Sc, Be, V, Ga and U) dissolved in the acid leachates. These results
achieved during the treatment of the PG leachates with the BA are
comparable to those obtained during the treatment with a conventional
reagent (i.e. Ca(OH)2).
8
Journal of Geochemical Exploration 221 (2021) 106698
Furthermore, the solids generated during the experiments with both
alkaline materials (i.e. S-BA and S-Ca(OH)2) host high concentrations of
elements of high economic value such as V (1036 and 2974 mg/kg,
respectively), U (721 and 2963 mg/kg), Sc (21 and 164 mg/kg), LREEs
(sum of values of 190 and 1935 mg/kg) and HREEs (163 and 2057 mg/
kg), among others. In addition, an economic estimate of the metal re­
serves contained in both solids suggests a total valuation of 2314 and
1539 MUSD for the S-BA and S-Ca(OH)2, respectively.
In overall, this study lays the foundations for the development of a
future successful and environmentally-friendly treatment system, which
uses an alkaline waste to treat acid leachates from PG stacks. The pro­
posed treatment could be extended to other areas of the world affected
by these environmental problems, given the high generation of PG and
BA worldwide. Moreover, this system would have a positive impact on
society, which must frequently assume the costs incurred by industrial
activities which lead to environmental impacts. Furthermore, solids
generated during treatments could be considered as a promising sec­
ondary source of REY, V, U, Sc, Ga and Be, which could be recovered,
helping to offset the costs associated with the treatments. However, it
would be important to emphasize that the economic value estimated for
these solids could probably be overestimated, since this estimation as­
sumes market prices for high purity elements and a total recovery of
such elements enclosed in the precipitates. Therefore, additional studies
are necessary to make a correct estimation of the potential economic
value contained in the solids originated during alkaline treatments of PG
leachates.
CRediT authorship contribution statement
Ricardo Millán-Becerro: Investigation, Data Curation, Writing-
Original Draft, Visualization. Carlos Ruiz Cánovas: Funding acquisi­
tion, Conceptualization, Writing - review & editing. Rafael Pérez-
López: Funding acquisition, Project administration, Writing - review &
editing. Francisco Macías: Funding acquisition, Methodology, Writing -
review & editing. Rafael León: Investigation, Data Curation.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
This work was supported by the Spanish Ministry of Economy and
Competitiveness, the European Regional Development Fund and the
University of Huelva under the research projects CAPOTE (MINECO;
CGL2017-86050-R), VALOREY (MINECO; RTI2018-101276-J-I00) and
EMFHA-SIST (FEDER; UHU-1253533). F. Macías was funded by the
R&D FEDER Andalucía 2014-2020 call through the project RENOVAME
(FEDER; UHU-1255729). Authors thanks ENCE Energía y Celulosa
Company for providing the biomass ashes. The authors are very grateful
to the funding support for the Committee of Experts on “The environ­
mental diagnosis and the proposal of measures of restoration of the
phosphogypsum stacks of Huelva”, appointed by the City Hall of Huelva.
We would also like to thank Dr. Stefano Albanese (Editor-in-Chief), Dr.
Annika Parviainen (Guest Editor) and two anonymous reviewers for the
support and comments that significantly improved the quality of the
original paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.gexplo.2020.106698.
R. Millán-Becerro et al. Journal of Geochemical Exploration 221 (2021) 106698

References Macías, F., Cánovas, C.R., Cruz-Hernández, P., Carrero, S., Asta, M.P., Nieto, J.M., Pérez-
López, R., 2017. An anomalous metal-rich phosphogypsum: characterization and
classification according to international regulations. J. Hazard. Mater. 331, 99–108.
Alemrajabi, M., Rasmuson, Å.C., Korkmaz, K., Forsberg, K., 2017. Recovery of rare earth
https://doi.org/10.1016/j.jhazmat.2017.02.015.
elements from nitrophosphoric acid solutions. Hydrometallurgy 169, 253–262.
Maeda, N., Katakura, T., Fukasawa, T., Huang, A.N., Kawano, T., Fukui, K., 2017.
https://doi.org/10.1016/j.hydromet.2017.01.008.
Morphology of woody biomass combustion ash and enrichment of potassium
Alemrajabi, M., Rasmuson, Å.C., Korkmaz, K., Forsberg, K., 2019. Processing of a rare
components by particle size classification. Fuel Process. Technol. 156, 1–8.
earth phosphate concentrate obtained in the nitrophosphate process of fertilizer
Millán-Becerro, R., Pérez-López, R., Macías, F., Cánovas, C.R., Papaslioti, E.M.,
production. Hydrometallurgy. 189, 105144. https://doi.org/10.1016/j.
Basallote, M.D., 2019. Assessment of metals mobility during the alkaline treatment
hydromet.2019.105144.
of highly acid phosphogypsum leachates. Sci. Total Environ. 660, 395–405.
Ayora, C., Macias, F., Torres, E., Lozano, A., Carrero, S., Nieto, J.M., Perez-Lopez, R.,
Nordstrom, D.K., Wilde, F.D., 1998. Reduction-Oxidation Potential (Electrode Method).
Fernandez-Martinez, A., Castillo-Michel, H., 2016. Recovery of rare earth elements
National Field Manual for the Collection of Water Quality Data. U.S. Geological
and yttrium from passive-remediation systems of acid mine drainage. Environ. Sci.
Survey Techniques of Water-Resources Investigations (Book 9, chapter 6.5).
Technol. 50, 8255–8262. https://doi.org/10.1021/acs.est.6b02084.
Pan, J., Zhou, C., Liu, C., Tang, M., Cao, S., Hu, T., Ji, W., Luo, Y., Wen, M., Zhang, N.,
Bogush, A.A., Dabu, C., Tikhova, V.D., Kim, J.K., Campos, L.C., 2019. Biomass Ashes for
2018. Modes of occurrence of rare earth elements in coal fly ash: a case study.
Acid Mine Drainage Remediation. Waste Biomass Valori. 1-13, 467. https://doi.org/
Energy Fuel 32 (9), 9738–9743. https://doi.org/10.1021/acs.energyfuels.8b02052.
10.1007/s12649-019-00804-9.
Papaslioti, E.M., Pérez-López, R., Parviainen, A., Macías, F., Delgado-Huertas, A.,
Bolívar, J.P., Martin, J.E., García-Tenorio, R., Pérez-Moreno, J.P., Mas, J.L., 2009.
Garrido, C.J., Nieto, J.M., 2018. Stable isotope insights into the weathering processes
Behaviour and fluxes of natural radionuclides in the production process of a
of aphosphogypsum disposal area. Water Res. 140, 344–353. https://doi.org/
phosphoric acid plant. Appl. Radiat. Isot. 67, 345–356. https://doi.org/10.1016/j.
10.1016/j.watres.2018.04.060.
apradiso.2008.10.012.
Parkhurst, D.L., Appelo, C.A.J., 2005. PHREEQC-2 version 2.12: a hydrochemical
Borges, R.C., Fávaro, D.I.T., Caldas, V.G., Lauria, D.C., Bernedo, A.V.B., 2016.
transport model. http://wwwbrr.cr.usgs.gov2005.
Instrumental neutron activation analysis, gamma spectrometry and geographic
Pérez-López, R., Nieto, J.M., López-Coto, I., Aguado, J.L., Bolívar, J.P., 2010. Dynamics
information system techniques in the determination and mapping of rare earth
of contaminants in phosphogypsum of the fertilizer industry of Huelva (SW Spain):
element in phosphogypsum stacks. Environ. Earth Sci. 75, 705. https://doi.org/
from phosphate rock ore to the environment. Appl. Geochem. 25 (5), 705–715.
10.1007/s12665-016-5468-x.
Pérez-López, R., Nieto, J.M., de la Rosa, J.D., Bolívar, J.P., 2015. Environmental tracers
Borra, C.R., Pontikes, Y., Binnemans, K., Van Gerven, T., 2015. Leaching of rare earths
for elucidating the weathering process in a phosphogypsum disposal site:
from bauxite residue (red mud). Miner. Eng. 76, 20–27. https://doi.org/10.1016/j.
implications for restoration. J. Hydrol. 529, 1313–1323. https://doi.org/10.1016/j.
mineng.2015.01.005.
jhydrol.2015.08.056.
Cánovas, C.R., Pérez-López, R., Macías, F., Chapron, S., Nieto, J.M., Pellet-Rostaing, S.,
Pérez-López, R., Macías, F., Cánovas, C.R., Sarmiento, A.M., Pérez-Moreno, S.M., 2016.
2017. Exploration of fertilizer industry wastes as potential source of critical raw
Pollutant flows from a phosphogypsum disposal area to an estuarine environment:
materials. J. Clean. Prod. 143, 497–505. https://doi.org/10.1016/j.
an insight from geochemical signatures. Sci. Total Environ. 553, 42–51. https://doi.
jclepro.2016.12.1083.
org/10.1016/j.scitotenv.2016.02.070.
Cánovas, C.R., Chapron, S., Arrachart, G., Pellet-Rostaing, S., 2019. Leaching of rare
Pingitore, N., Clague, J., Gorski, D., 2014. Round Top Mountain rhyolite (Texas, USA), a
earth elements (REEs) and impurities from phosphogypsum: a preliminary insight
massive, unique Y-bearing-fluorite-hosted heavy rare earth element (HREE) deposit.
for further recovery of critical raw materials. J. Clean. Prod. 219, 225–235.
J. Rare Earths 32 (1), 90–96. https://doi.org/10.1016/S1002-0721(14)60037-5.
Deng, X.H., Chen, Y.J., Yao, J.M., Bagas, L., Tang, H.S., 2014. Fluorite REE-Y (REY)
Qin, J., Hovmand, M.F., Ekelund, F., Rønn, R., Christensen, S., de Groot, G.A., Krogh, P.
geochemistry of the ca. 850 Ma Tumen molybdenite–fluorite deposit, eastern
H., 2017. Wood ash application increases pH but does not harm the soil mesofauna.
Qinling, China: constraints on ore genesis. Ore Geol. Rev. 63, 532–543. https://doi.
Environ. Pollut. 224, 581–589.
org/10.1016/j.oregeorev.2014.02.009.
Qureshi, A., Jia, Y., Maurice, C., Öhlander, B., 2016. Potential of fly ash for neutralisation
EN 12457-2, 2002. Characterization of Waste, Compliance Test for Leaching of Granular
of acid mine drainage. Environ. Sci. Pollut. Res. 23 (17), 17083–17094.
Wastes Materials and Sludges, Part 2: One Stage Batch Test at a Liquid to Solid Ratio
Rutherford, P.M., Dudas, M.J., Samek, R.A., 1994. Environmental impacts of
of 10 l/kg− 1 for Materials with Particle Size Below 4 mm (Without or With Size
phosphogypsum. Sci. Total Environ. 149, 1–38. https://doi.org/10.1016/0048-9697
Reduction). European Committee of Standardization, p. 28. CEN/TC 292,12/02.
(94)90002-7.
Ericson, W.A., Patel, S.K., Marek, A.C., Brown, C.M., 1997. Proposed design for fertilizer
Sadri, F., Nazari, A.M., Ghahreman, A., 2017. A review on the cracking, baking and
gypsum stack and acid pond closure. In: Tailings and Mine Waste’97, pp. 69–77.
leaching processes of rare earth element concentrates. J. Rare Earths 35 (8),
Fuleihan, N.F., Werner, R.J., 2011. Evaluation of Lime Treatment Sludge Alternative
739–752. https://doi.org/10.1016/S1002-0721(17)60971-2.
Disposal Methodologies Including Utilization in Closure of Phosphogypsum Stacks.
Sahoo, P.K., Tripathy, S., Equeenuddin, S.M., Panigrahi, M.K., 2012. Geochemical
Florida Industrial and Phosphate Research Institute (FIPR). http://fipr.state.fl.us/
characteristics of coal mine discharge vis-à-vis behavior of rare earth elements at
wp-content/uploads/2014/12/03-143-247Final.pdf. (Accessed 26 April 2020).
Jaintia Hills coalfield, northeastern India. J. Geochem. Explor. 112, 235–243.
Heviánková, S., Špakovská, B., Klimko, T., Kyncl, M., Bílská, Z., Kučerová, L., 2014. Acid
Smith, K.L., Figueroa, L.A., Plumlee, G.S., 2013. Can treatment and disposal costs be
mine drainage treatment by ash from wooden chip combustion: study of mine water
reduced through metal recovery? In: Wolkersdorfer, Brown, Figueroa (Eds.),
composition in dependence on the ash dose and duration of mutual interaction.
Proceedings of the Annual International Mine Water Association Conference-
Carpath. J. Earth Environ. Sci. 9 (2), 159–170.
Reliable Mine Water Technology. Golden, Colorado (USA).
INAP (The International Network for Acid Prevention), 2020. Drainage treatment. htt
SMM, 2020. Shangai metals market. https://www.metal.com/. (Accessed 28 April 2020).
p://www.gardguide.com/index.php?title=Chapter_7. (Accessed 16 April 2020).
Tayibi, H., Choura, M., López, F.A., Alguacil, F.J., López-Delgado, A., 2009.
Jin, S., Hu, Z., Huang, Y., Hu, Y., Pan, H., 2019. Evaluation of several phosphate
Environmental impact and management of phosphogypsum. J. Environ. Manag. 90,
amendments on rare earth element concentrations in rice plant and soil solution by
2377–2386. https://doi.org/10.1016/j.jenvman.2009.03.007.
X-ray diffraction. Chemosphere 236, 124322. https://doi.org/10.1016/j.
US EPA, 1992. (US Environmental Protection Agency), TCLP, Method 1311, Rev 0. InSW-
chemosphere.2019.07.053.
846: Test Methods for Evaluating Solid Waste, Physical/Chemical Methods. Office of
Kulczycka, J., Kowalski, Z., Smol, M., Wirth, H., 2016. Evaluation of the recovery of rare
Solid Waste, Washington, DC.
earth elements (REE) from phosphogypsum waste - case study of the WIZ_OW
USGS, 2020. Mineral Commodity Summaries 2020. U.S. Geological Survey, 2020,
chemical plant (Poland). J. Clean. Prod. 113, 345–354. https://doi.org/10.1016/j.
Mineral Commodity Summaries 2020. U.S. Geological Survey (200 pp.).
jclepro.2015.11.039.
Vassilev, S.V., Baxter, D., Andersen, L.K., Vassileva, C.G., 2013. An overview of the
LME, 2020. London metal exchange official prices for metal global market. https://lme.
composition and application of biomass ash. Part 1. Phase–mineral and chemical
com/. (Accessed 28 April 2020).
composition and classification. Fuel 105, 40–76.
Lottermoser, B.G., 2010. Mine Wastes: Characterization, Treatment and Environmental
Impacts, third ed. Springer-Verlag, Berlin, Heidelberg.