Sustainable Chemistry for the Environment 6 (2024) 100103

Contents lists available at ScienceDirect

Sustainable Chemistry for the Environment

journal homepage: www.editorialmanager.com/scenv

Pit-lake remediation by chemically activated Chlorocardium rodiei:

Simultaneous metal ion removal from acidic waters
Hamant E. France, O.L.K. Strong, Kevin M. Scotland, Tyler M. Roy, Andrew J. Vreugdenhil *
Inorganic Materials Research Laboratory, Department of Chemistry, Trent University, Peterborough, ON K9J 7B8, Canada

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

Keywords: The current study investigates the valorization of waste wood from a tropical hardwood found in Guyana
Chemically activated carbon (Greenheart – Chlorocardium rodiei) to high surface area activated carbons (AC). We subsequently deployed these
Tropical waste wood adsorbents to sequester Mn 2+, Fe3+, and Al3+ from the acidic waters of a recreational bauxite pit-lake. We
Acidic lake
studied the impact of activation parameters such as temperature, impregnation ratio and acid concentration on
Antagonistic interaction
Greenheart
the texture and surface chemistry of ACs and demonstrate that optimized low pH at point-of-zero charge (pHpzc)
ACs are efficient adsorbents for the target ions. A mesoporous AC with specific surface area of 2208 m2/g, 11 %
surface oxygen and pHpzc of 1.98 was produced under optimized conditions. ACs removed 93–100 % of target
ions from pit-lake waters at a native pH of 3.1. Al3+ exerted an antagonistic effect on Mn2+ adsorption in syn­
thetic binary ion systems reducing adsorption by as much as 56 %. The Sips Model fitted the adsorption data best
predicting maximum adsorption capacities for Mn2+, Fe3+ and Al3+ of 17.8 mg/g, 23.7 mg/g, and 6.12 mg/g for
these low pHpzc optimized materials. These materials show great promise in removing heavy metals from acidic
waters.

1. Introduction from aqueous systems [14–17]. Marrakchi et al., 2017 fabricated ACs
from chitosan flakes and cross-linked chitosan/sepiolite and successfully
Pit-lakes, an environmental legacy of open pit mining, are formed by applied them in methylene blue and reactive orange 16 removal [18,
the accumulation of ground and surface waters in abandoned pit-mines 19]. Khalid et al., 2023 used pinecone to produce activated carbons
[1–3]. These lakes are aesthetically pleasing and are an important rec­ which were subsequently used in dibenzothiophene from model
reational resource for some communities [1,2]. Bauxite pit-lakes in gasoline-based adsorbates [20]. Metal ions such as Cu2+, Pb2+, Zn2+ and
particular are known to have very low pH and contain elevated levels of Cd2+ have also been removed from aqueous systems by various ACs
metal ions, particularly Al3+, Mn2+, and Fe3+ [4,5]. Elevated levels of [21]. Some ACs may however, be unsuitable to remove metal ions from
these ions pose health risks to swimmers, particularly vulnerable waters with low pH. The challenge arises from the pH at point-of-zero
swimmers such as children [1,6]. Levels of heavy metals in natural charge of the AC being higher than the solution pH [22,23]. This re­
waters can be reduced by low-cost carbonaceous adsorbents [7–10]. sults in unfavorable electrostatic interactions and poor adsorption
Low-cost adsorbents provide significant advantages in pollutant typically results below pH 4. This challenge can be overcome by fabri­
removal such as versatility, economically feasible fabrication, ease of cating adsorbents with low pHpzc.
application and potential for regeneration [11–13]. Activated carbons in In this study, we produce high surface area, low pHpzc ACs by a two-
particular have gained widespread attention in aquatic remediation staged phosphoric acid activation of a novel precursor; waste wood from
applications. Chlorocardium rodiei (greenheart). The activant phosphoric acid was
Activated carbon is an adsorbent material with extensive specific used because it typically produces ACs with yields approaching 50 %
surface area, well-developed pore structure and distinctive surface and up to 96 % of the acid is recyclable [24]. We used ACs to remove
chemistry [14]. These properties of ACs make them well suited to Al3+, Mn2+ and Fe3+ from the waters of an acidic bauxite pit lake (pH
removing pollutants from aquatic media and their versatility has been 3.1) in Guyana widely used for recreational purposes. Other works have
demonstrated in their use to remove both inorganic and organic species addressed the use of various adsorbents to remove these cations from

* Corresponding author.
E-mail address: [email protected] (A.J. Vreugdenhil).

https://doi.org/10.1016/j.scenv.2024.100103
Received 8 February 2024; Received in revised form 22 April 2024; Accepted 22 April 2024
Available online 24 April 2024
2949-8392/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

water, however, these typically have been done using model solutions Analyte concentration was again determined by MP-AES. Experiments
with pH >3 or typically report poor adsorption at low pH [25–28]. were conducted at the native pH of lake water; 3.10 at 20 ℃. Solution
Moreover, a ubiquitous approach in adsorption experiments involving pH was measured before and after the adsorption equilibrium studies.
metal ions is to optimize the pH at which adsorption occurs and conduct
batch tests at this pH. This approach, however, has largely left metal ion 2.6. Isotherm modelling
adsorption from strongly acidic waters unexplored and hence materials
which can efficiently function under such conditions undeveloped. To Many adsorption models have been developed to describe adsorption
the best of our knowledge no study has been reported on the fabrication phenomena. These models are derived based on assumptions which may
of super activated carbons from Chlorocardium rodiei waste wood and be applicable to the system under consideration. The Langmuir model
their subsequent use to simultaneously remove Al3+, Mn2+ and Fe3+ (Eq. 1), one of the more ubiquitously applied models, assumes mono­
from the waters of a bauxite pit-lake with a pH of 3. layer coverage of adsorbate on a homogeneous adsorbent. Further, it is
We characterized the texture and surface chemistry and investigated assumed that there are no lateral interactions between adsorbed species
adsorption behaviors of ACs using gas adsorption, TGA, XPS, Raman, pH [13].
drift and batch adsorption experiments. To obtain estimates of their
qm KL Ce
maximum adsorption capacities, we modelled the adsorption of ions Langmuir Adsorption Modelqe = (1)
1 + KL Ce
onto ACs using Langmuir, Freundlich and Sips adsorption models.
The model parameters qe and qm are the amount of adsorbate
2. Materials and methods adsorbed at equilibrium and the maximum monolayer adsorption ca­
pacity (mg/g) respectively. Equilibrium concentrations of adsorbate and
2.1. Materials the Langmuir constant are denoted by Ce (mg/l) and KL (L/mg).The
Freundlich Model (Eq. 2) is empirical in nature and assumes multilayer
Phosphoric Acid, 85 %, was obtained from Caledon Laboratory adsorption on a heterogeneous surface. Here, the Freundlich constants
Chemicals, while nitric acid, 70 %, and all 1000 ppm standards of Al are KF ((mg/g)/(mg/l)n) and n. [29].
(NO3)3, Mn(NO3)2, Fe(NO3)3, NaCl, NaOH, and HCl were purchased
Freundlich Modelqe = KF Cne (2)
from Sigma Aldrich.
The Sips Model (Eq. 3) is a three-parameter model and a hybrid of the
2.2. Sample collection and sizing of greenheart precursor Langmuir and Freundlich Models. It assumes monolayer coverage of
adsorbate on the adsorbent surface. Further, its hybrid nature allows for
Greenheart sawdust was obtained from a local sawmill in Parika application to both homogeneous and heterogeneous surfaces. In this
(Guyana) and sun-dried for 2 h. Samples were then placed into sealed model, the Sips parameter denoting maximum adsorption capacity is
sample bags and stored at room temperature. A 5 g sample of the pre­ qms , while Ks (L/mgns) and ns are the sips constant and heterogeneity
cursor was sized using a manual Keck sieve shaker (Cole Parmer) fitted factor respectively, with 0 ≤ ns ≥ 1 [29].
with OPN 40, 26, 20 and 9 sieves. Most of the particles ranged in size
qms Ks Cne s
from 1–0.2 mm, with the major size fraction being 0.5 mm. Conse­ Sips Model qe = (3)
1 + Ks Cne s
quently, further size reduction of the precursor was unnecessary.
25 mL of synthetic Al, Mn and Fe single ion solutions of concentra­
2.3. Preparation of greenheart phosphoric acid activated carbon (GH- tions ranging from 0.5–50 ppm were prepared from 1000 ppm stock
PAC-DIW) and base-washed activated carbon (GH-PAC-BW) solutions and equilibrated (24 h) with 50 mg of GH-PAC-BW. Suspen­
sions were allowed to settle for 10 min after which aliquots of the su­
The method used for activation of greenheart was a modified version pernatant were filtered with 0.45 µm syringe filters and made up to
of that developed by Strong et al. [24]. Key modifications included 25 mL with 5 % HNO3. Residual ion concentrations in the filtrate were
additional activations at 500 ℃ and 600 ℃ and digestion and carbon­ determined by MPAES. Experiments were conducted at pH 5 at 20 ℃.
ization for times ranging from 0.5–20 h at 110 ℃. In addition, Adsorption data were fitted to the Langmuir, Freundlich and Sips
base-modified ACs were prepared by washing GH-PAC-DIW with 0.1 or adsorption isotherm models.
0.4 M NaOH solutions with 1 or 2 h of stirring at 400 rpm.
2.7. Characterization techniques
2.4. Single, binary and ternary metal ion batch tests
pH at pzc of ACs was measured by the pH drift method in the range
Single ion standard solutions of 10 ppm Al(NO3)3, 2 ppm Mn (NO3)2, 1–11. Specific surface area, pore volume and pore size distribution were
and 2 ppm Fe(NO3)3 were prepared from 1000 ppm stock solutions and obtained using a Micromeritics Tristar II+ analyzer equipped with
buffered at pH 5. 25 mL of each solution was placed into separate Density Functional Theory (DFT) pore modelling software at 77 K [24].
125 mL sealable plastic cups and 50 mg of AC added. Suspensions were XPS was done using a Thermofisher Scientific K(alpha) spectrometer
shaken at 200 rpm for 24 h (Digital Orbital Shaker Table, Ultident) at 20 with a monochromatic Al K(alpha) source (15 mA, 15 kV) and pass
℃. Aliquots of the supernatant were removed and filtered with a energy of 50 eV. XPS data analysis was done using CasaXPS (version
0.45 µm syringe filter and made up to 25 mL with 5 % HNO3. Microwave 2.3.1). Spectra were corrected to the main line of the carbon 1 s peak
Plasma Atomic Emission Spectroscopy (MP-AES) was used to determine observed at 284.85 eV. Changes in the greenheart sawdust sample with
residual ion concentrations in filtrates. Binary and ternary ion batch changes in temperature were measured by a TGA Q500 (TA Instruments,
tests were conducted similarly to single ion tests but with multi-ion Newcastle, DE, USA). The sample was heated on a platinum pan from 25
solutions. ℃ to 600 ℃at 10 ℃/min under nitrogen flowing at 60.0 mL/min.
Raman characterization was done using a method published by Scotland
2.5. Lake water batch tests et al. [30]. Metal ion concentration in solutions were measured by the
Agilent MP-AES 4200 at wavelengths of 371.9 nm, 403.1 nm and
25 mL of lake water samples were added to 50 mg of GH-PAC-DIW or 396.2 nm for Fe, Mn, and Al respectively. Calibration curves were
GH-PAC-BW and then agitated at 200 rpm (Digital Orbital Shaker Table, plotted from 6 multi-ion standards of Al(NO3)3, Mn(NO3)2, Fe(NO3)3 of
Ultident) for 24 h. Suspensions were filtered with a 0.45 µm membrane concentrations 1 ppm, 5 ppm, 10 ppm, 25 ppm, 50 ppm and 100 ppm. A
syringe filter and aliquots were made up to 25 mL with 5 % HNO3. 5 % HNO3 solution was used as the blank.

2
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

3. Results and discussion characteristics of AC for samples labelled C1-C13. Increases in impreg­
nation ratio correlated with increases in specific surface area up to a
3.1. BET surface area, pore volume and pore size distribution of ACs critical maximum (Fig. 2a); 1:1 ratio. Surface area decreased thereafter.
When higher ratios of acid are used, pyrophosphates are more likely to
AC production from greenheart was optimized at 600 ℃, 1:1 form and occupy more spaces in the material. Subsequent removal of
impregnation ratio with 20 h of digestion (Fig. 1a–c). This AC was pyrophosphates by washing engenders larger mesopores and macro­
predominantly mesoporous and had a trimodal pore size distribution pores in AC [24]. These larger pores may be unstable and consequently
(Fig. 1a - inset) with about 80 % of the pore volume contribution being collapse leading to a decrease in surface area. Pore collapse is mitigated
due to mesopores. Hysteresis, caused by the phenomenon of capillary by using lower concentrations of acid and optimized impregnation ratios
condensation, is observed on the isotherm at 600℃ (Fig. 1a) and pro­ as this reduces the likelihood of the formation of polymeric acid species.
vides strong evidence for mesoporosity [31]. The extensive surface area Increases in digestion time allowed for improved permeation of the
and mesoporous texture of GH-PAC-DIW likely resulted from the dual activant into the natural porosity of the precursor which resulted in
effects of adequate heat energy to mediate bond rearrangement in the better cleavage of aryl-ether and glycosidic bonds in lignin and cellulose
material and the presence of sufficient phosphoric acid molecules to act respectively. Higher activation temperatures provided more energy for
as templating agents. GH-PAC-DIW texture compared favorably with the elimination of volatiles and the reorganization of carbon centers.
ACs produced from other wood-based feedstock under similar activation Since this reorganization leaves spaces on the order of nanometers in the
conditions including cedar, chestnut, cannabis, prosopis and mixed material, a network of porosity and increased surface area result
hardwood (Fig. 1d) [32–34]. These results suggest the viability of [35–37]. These results show that moderate temperature and minimal
greenheart waste wood as a suitable and sustainable feedstock for chemicals can generate ACs with good texture from greenheart waste
fabricating high surface area activated carbons with well-developed wood. They further suggest that this feedstock is a good candidate for
porosity. efficient and economical valorization to ACs.

3.2. Impact of activation parameters on material texture 3.3. TGA and Raman analysis

Table 1 summarizes the impact of activation parameters on textural The initial portion of the TGA scan under nitrogen of the raw

Fig. 1. - BET Isotherm Plots for ACs produced at 400–600 ℃, (b) 0.5–2 impregnation ratio (c) 0.5–20 h digestion, and (d) comparison of GH-PAC-DIW texture with
other ACs.

3
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

Table 1
Specific surface areas, micropore and total pore volumes of ACs produced under various conditions.
AC Carbonization Time (h) Activation Temperature (℃) Acid Conc. (wt%) Impregnation Ratio SBET (m2/g) Vt (cm3/g) Vmic (cm3/g)

C1 0.5 400 25 1:1 1140 ± 35 0.54 ± 0.02 0.24 ± 0.00

C2 3.0 400 25 1:1 1201 ± 34 0.55 ± 0.03 0.22 ± 0.02
C3 20 400 25 1:1 1841 ± 52 0.84 ± 0.04 0.28 ± 0.03
C4 20 600 25 1:1 1986 ± 159 1.01 ± 0.07 0.19 ± 0.04
C5 20 400 50 1:1 1743 ± 111 0.84 ± 0.04 0.29 ± 0.09
C6 3.0 400 25 2:1 750.0 ± 67 0.34 ± 0.02 0.10 ± 0.02
C7 5.0 600 50 1:1 1814 ± 139 0.85 ± 0.07 0.34 ± 0.13
C8 20 500 50 1:1 1888 ± 37 0.83 ± 0.02 0.46 ± 0.01
C9 20 600 50 0.5:1 873.6 ± 59 0.36 ± 0.02 0.28 ± 0.01
C10a 20 600 50 1:1 2208 ± 91 1.01 ± 0.03 0.20 ± 0.02
C11b 20 600 50 1:1 1740 ± 22 1.92 ± 0.38 0.48 ± 0.01
C12 20 600 50 1.5:1 2165 ± 143 1.11 ± 0.07 0.35 ± 0.04
C13 20 600 50 2:1 1065 ± 145 1.18 ± 0.11 0.17 ± 0.01
a
GH-PAC-DIW
b
GH-PAC-BW-0.1

Fig. 2. - Impact of (a) Impregnation Ratio, (b) Activation Temperature with 1:1 impregnation ratio (c) digestion time with 1:1 impregnation ratio on ACs texture and
(d) relationship between ID/IG ratio and total pore volume with impregnation ratio for ACs produced with 50% H3PO4 at 600 ℃.

greenheart precursor (Fig. 3a) indicates a loss of mass of ~5 % at an decomposition ranges of the major constituents of wood. The amor­
onset temperature of 57 ℃. This change is most likely due to loss of phous texture of the materials was confirmed by characteristic Raman D
adsorbed moisture on the precursor surface. On the DTG curve, a clear and G-bands (Fig. 3b) at ~1350 cm− 1 and ~1580 cm− 1 respectively
peak at 354 ℃ with a subtle shoulder at ~260 ℃ is observable. These [41,42]. The relative intensities of these bands yield the ID/IG ratio. This
features most likely owe their appearance to decomposition of the major metric provides a measure of the degree of disorder in the sample. It was
macromolecular components of the wood precursor. As described pre­ observed that the ID/IG ratio was proportional to the phosphoric acid
viously in the literature, decomposition of the three macromolecular impregnation ratio as well as total pore volume of the ACs (Fig. 2d).
components of wood take place in narrow temperature ranges which
overlap [38,39]. Hemicellulose and cellulose decompose within the
3.4. pH at point-of-zero of ACs
temperature ranges 220–315 ℃ and 315–400 ℃, respectively [40].
TGA and DTG analysis confirms the suitability of the activation
GH-PAC-DIW had a low point of zero charge pH of 1.98; a conse­
temperatures used in this work as they overlap with the major
quence of the presence of protonated residual phosphoric acid species on

4
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

Fig. 3. - (a) TGA and DTG scans of greenheart under N2 (60 mL/min) at a scan rate of 10 ℃/min and (b) Raman spectra of ACs produced at impregnation ratios in
the range 0.5–2.

the AC surface. Control of this parameter was critical to the adsorption of 3.6. Lake water quality and batch adsorption tests analysis
target ions [43]. Base-wash treatments produced GH-PAC-BW-0.1 and
GH-PAC-BW-0.4 with increased pHpzc of 2.95 and 7.08. These treat­ 3.6.1. Lake water quality
ments deprotonated acidic functions on the material surface. Conse­ Target ion levels were respectively one (Mn2+ and Fe3+) or two
3+
quently, batch tests in this work were carried out using GH-PAC-DIW (Al ) orders of magnitude above the guideline values (GVs) for recre­
and GH-PAC-BW-0.1 at pH 3.10 (the native pH of lake water samples) ational water quality as outlined by the Australian and New Zealand
and at pH 5 for buffered synthetic solutions. Base wash treatments Fresh and Marine Water Quality guidelines (ANZECC & ARMCANZ-
therefore provide a means of tailoring the surface charge of ACs. 2000) [48] (Table 2). Comparison to drinking water reference guide­
lines may be justified when the possibility of frequent and protracted use
of pit lakes for swimming is considered.
3.5. Surface chemistry of ACs and impact of base wash treatment
3.6.2. Batch adsorption tests with lake and synthetic waters
Greenheart precursor was predominantly composed of carbon and The initial pH of lake water samples was 3.10 (Table 2) with no
oxygen with modest amounts of silicon and nitrogen (Fig. 4a and d). significant alteration in pH after batch experiments. Both GH-PAC-DIW
Activation simultaneously increased the percentages of C and decreased and GH-PAC-BW demonstrated excellent removal efficiencies of Fe3+,
the percentages of heteroatoms [37]. The resultant adsorbents both Mn2+ and Al3+ (Fig. 5), with removal efficiencies >80 %. Lobo-Recio
contained ~97 % carbon and oxygen. As expected, GH-PAC-BW con­ et al. [26] reported removal efficiencies for Fe3+, Mn2+ and Al3+ at
tained sodium on its surface from the base washing process (Fig. 4d). pH 4 of 15 %, 43 % and 54 % on a zeolite. Enhanced removal of Al3+ and
Interestingly, this AC also showed reduced specific surface areas. Mn2+ was observed in the lake water batch tests with GH-PAC-BW
GH-PAC-BW had ~21 % smaller surface area than GH-PAC-DIW which had pHpzc of 2.95 (Fig. 5b). The complementary surface charge
(Table 1). This reduction in surface area may be largely due to pore of greenheart ACs at the native pH of lake water and the presence of
widening, and pore collapse because of the etching effect of sodium sufficient OFG adsorption sites most likely facilitated excellent perfor­
hydroxide on the walls of pores; particularly micropore walls [44,45]. mance of these tropical hardwood adsorbents.
Sodium adsorption on AC surface may have also played a role in surface In single ion batch tests, GH-PAC-DIW, completely removed Fe3+ and
area loss. This may have occurred in two ways. Firstly, the additional Mn2+. Here too, extensive surface area and complementary charge
mass of sodium would reduce the specific surface of AC without facilitated adsorption. Additionally, the comparatively low initial con­
contributing to the surface area. Secondly, sodium salt precipitates may centration of these two species (2 ppm) did not exhaust the available
have filled otherwise accessible pores. The need to moderate base adsorption sites. Moreover, these two cations had comparatively large
washing so as to preserve material texture is made apparent from these Sips Ks constants: 1.32 and 3.46 for Mn2+ and Fe3+ respectively
results. (Table 3). Generally, the Langmuir affinity factor KL is used to infer the
Residual phosphorous peaks (Fig. 4a) were present at ~134 eV. Base strength of the interaction between adsorbate and adsorbent [13–49]. In
wash treatment reduced the proportions of phosphorous in GH-PAC-BW our modelling analysis, the Sips constant (ns) approaches unity for all
(Fig. 4d). While base wash treatment reduced the proportions of Lewis adsorption systems (Table 3) implying that the Sips model reduces to the
base functionalities, the impact of increasing negative surface charge Langmuir model. Consequently, Ks can be used in place of KL to infer the
was sufficient to enhance the adsorption behaviour of GH-PAC-BW to strength of adsorbate-adsorbent interaction [49]. The percentage
target ions in lake water (Fig. 5b). removal for Al3+ was ~77.0 ± 3.21 % in all systems. Additionally, of the
Deconvolution of the O1s peak (Fig. 4b and e) showed the presence three cations, Al3+ had the smallest Ks constant (0.26) implying rela­
of ethers and esters (~533 eV), phenolic groups (~532 eV) and amides tively weaker interaction with greeheart ACs. These results suggest that
and aromatic ketones (~531 eV), on ACs. C1s greenheart ACs are promising materials for remediating acidic waters
deconvoluted high resolution spectra provided complementary evi­ containing heavy metals. Furthermore, since their fabrication involves
dence for these functionalities (Fig. 4c and f). Oxygen containing func­ the valorization of greenheart waste wood with minimal chemical input,
tional groups (OFGs) were critical in facilitating removal of target they allow for a green approach to remediating polluted waters. More­
cations. It is known that ACs with high proportions of OFGs show good over, these materials may have value in assisting efforts to recover
cationic adsorption performance [46]. This is generally ascribed to valuable metals from waters impacted by acidic mine drainage partic­
increased columbic attractions between the resulting negatively charged ularly where other methods are costly or inefficient.
AC surface and cationic species [47].

5
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

Fig. 4. - Typical XPS wide scans, O1s and C1s high resolution scan of (a-c) GH-PAC-DIW and (d-f) GH-PAC-BW.

3.7. Metal ion interactions Secondly, antagonistic interactions (IR < 1) may occur where decreased
adsorption of one or more species results. And thirdly, there may be
In the binary and ternary systems, ion interaction appeared to in­ noninteractions (IR = 1) [50].
fluence the adsorption behaviour of ions. This gave rise to system Antagonistic interactions may have occurred between Mn2+ and Al3+
dependent adsorption of Mn2+. These effects were most clearly observed in the binary and ternary ion systems where the IRs were 0.44 and 0.35
in the systems containing both Mn2+ and Al3+ (Fig. 5a). Generally, the respectively (Fig. 5a). These effects may be due to differences in ionic
adsorption interactions of ions in multi-ion systems may be of three strengths of the aluminum-containing systems where Al3+ concentration
types. Firstly, interactions may be synergistic with interaction ratio (IR) was 10 ppm. It is known that increasing ionic strength reduces adsorp­
> 1. Such interactions are characterized by increased adsorption of one tion of some cations on adsorbents due to decreased activity coefficients.
or more species in the mixture when compared to single ion adsorption. Lower activity coefficients result in retardation of species transfer to

6
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

Fig. 5. - Removal Efficiencies and interaction ratios of (a) ACs with Single, Binary and Ternary-ion Synthetic Systems and (b) Lake water samples.

Lobo-Recio et al. [26] reporting Qmax of 13.9 mg/g and Goher et al. [28],
Table 2
106.5 mg/g. These experiments were however done in the pH range
Comparison of lake water chemistry with literature and water standards.
4–7. Al-Muhtaseb et al. [53] reported maximum adsorption of Al3+ of
Lake Water Ion Guideline Values for Guideline Values for 0.676 mg/g on a date-pit AC at pH 3. The maximum adsorption capac­
Concentration (ppm) Recreational Water Drinking Water
ities of Mn2+ and Fe3+ on GH-PAC was generally comparable with other
(ppm) (ppm)
adsorbents but Al3+ Qmax was typically lower (Table 4.)
This Study Williams ANZECC WHO WHO (2021)
Excellent model fit was seen for all ions as indicated by R2 values
2020 (2000) (2021)
approaching unity (Table 3). The extent of this correlation may have
pH 3.10 ± 3.4 5–9 - 6.5–9.5 been improved by the inherent advantages of model fit due to the use of
0.01
Fe 1.47 ± 3.69 0.3 NA NA
a three-parameter model. Furthermore, ns values and the nearly iden­
0.15 tical KL and Ks values show that the Sips model reduced to the Langmuir
Mn 1.23 ± 2.40 0.1 8 0.4 model for all systems considered.
0.12 Modelling analysis confirms that greenheart ACs perform well in
Al 7.49 ± 9.02 0.2 18 0.9
acidic environments and therefore may be deployed as remediation
0.75
materials for acidic pit-lakes and waters impacted by acid mine
drainage. Further, these low pHpzc materials may additionally be suit­
adsorbent sites [51]. A greater retardation in Mn2+ removal was seen in able candidates for isolating valuable metals from acidic waters.
the ternary system (35 % removal) compared to the binary system (44 %
removal). This was likely due to increased competition for adsorption 4. Conclusions
sites from the two trivalent cations.
This work demonstrates the valorization of a waste lignocellulosic
feedstock produced by industry proximate to the metal contaminated pit
3.8. Isotherm modelling analysis
lakes. The valorization is accomplished by conversion to low-pH-at-pzc
activated carbons which remove environmentally relevant levels of
To better understand the interactions between ACs and analyte sys­
important metal contaminants from low pH waters. Our results show
tems, the adsorption data were fitted to the Langmuir, Freundlich and
that low pHpzc ACs with surface areas of 2200 m2/g are achievable with
Sips Models (Fig. 6a–c).
minimal chemical inputs, at low activation times and moderate tem­
Model parameters are shown in Table 3. The Sips Model fitted the
peratures. Since moderate temperatures and minimal chemical inputs
adsorption data best. Maximum adsorption capacities (Qmax) of
were used to produce these high surface area materials, their fabrication
17.8 mg/g, 23.7 mg/g, and 6.12 mg/g for Mn2+, Fe3+ and Al3+
is both economically feasible and sustainable. Fabrication of these ACs
respectively were predicted (Table 3). Qmax for adsorption of Mn2+, on
derive the dual benefits of diverting a copious local waste stream and
greenheart AC was greater than that reported by Goher et al. (0.176 mg/
converting it, via a largely environmentally friendly process, to efficient
g) for adsorption on a commercial AC and Nunez-Gomez et al. [28–52]
environmental remediation materials.
(3.87 mg/g) for adsorption on shrimp shells. Adsorption on a zeolite at
Evaluation of the AC processing parameters demonstrated that SBET
pH 4 was however greater (50.11 mg/g) [26]. GH-PAC adsorption ca­
and total pore volume increased with increasing impregnation ratio.
pacity for Al3+ was lower than reported in the literature, with

Table 3
Model parameters and statistical evaluation of model fit for Langmuir, Freundlich and sips isotherms.
Metal Ions Langmuir Parameters Freundlich Parameters Sips Parameters

Qmax (mg/g) KL (L/mg) R2 KF (L/g) n R2 Qmax (mg/g) Ks ns R2

Mn 17.8 ± 3.05 3.46 ± 3.30 0.839 13.7 ± 0.06 9.44 ± 0.19 0.772 17.8 ± 1.31E-16 3.46 ± 1.94E-16 1.00 ± 5.73E-17 1.00
Fe 23.7 ± 1.76 1.32 ± 0.33 0.976 11.3 ± 1.21 3.00 ± 0.55 0.955 23.7 ± 3.60E-16 1.32 ± 5.08E-17 1.00 ± 4.10E-17 1.00
Al 6.01 ± 0.24 0.27 ± 0.04 0.991 2.21 ± 0.20 3.74 ± 0.04 0.930 6.12 ± 5.29 E-1 0.26 ± 7.11E-2 0.96 ± 1.61E-1 0.99

7
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

Fig. 6. - Langmuir, Freundlich and Sips Adsorption Model Isotherms for (6a) Al, (6b) Fe and (6c) Mn, in single ion systems.

Table 4
Langmuir maximum adsorption capacities for Mn2+, Al3+ and Fe3+ for various adsorbents.
Adsorbent pH Isotherm Model Maximum Adsorption Capacity - Qmax (mg/g) Reference

Mn Al Fe

Limonia acidissima shell AC 5 Langmuir - - 48.5 [54]

Type A Zeolite 4 Sips 50.1 25.9 3238.8 [26]
Zeolite 5 Langmuir 10.8 - - [55]
LTA Zeolite 3.5 - 1.4 15.2 - [56]
Chitosan/MMT Nanocomposite 5.5 Langmuir - - 7.0 [57]
MWCNT 10 Langmuir - 393.5 - [52]
Greenheart-AC 3 Sips 17.8 ± 3.05 3.46 ± 3.30 23.7 ± 1.76 This study

TGA and Raman spectroscopy were used to confirm the amorphous Author contributions
nature of the ACs. Control and characterization of the AC surface
functionality demonstrated a pHpzc of 1.98 with primarily oxygen rich H.E. France and A.J. Vreugdenhil proposed the concept. H.E. France
functionality with a surface atomic percent of 11 %. Sips modeling performed the experiments and wrote the manuscript. Oliver K.L. Strong
analysis of the adsorption behaviour of Mg2+, Fe3+ and Al3+ showed assisted with XPS data curation and analysis, K.M. Scotland assisted with
capacities of 17.8, 23.7 and 6.12 mg/g respectively. Raman analysis and Tyler M. Roy assisted with modelling analysis. All
Moreover, with removal efficiencies in real aquatic matrices of authors contributed to manuscript generation and revision.
>90 %, these materials are promising candidates for efficient metal ion
decontamination applications such as heavy metal removal from waters Declaration of Competing Interest
impacted by acid mine drainage. Additionally, these materials are likely
to prove useful in the harvesting of valuable metal ions from acidic mine The authors declare the following financial interests/personal re­
waters. lationships which may be considered as potential competing interests
Hamant E. France reports financial support was provided by Sustainable
Guyana Program. If there are other authors, they declare that they have

8
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

no known competing financial interests or personal relationships that [19] F. Marrakchi, W.A. Khanday, M. Asif, B.H. Hameed, Cross-linked chitosan/sepiolite
composite for the adsorption of methylene blue and reactive orange 16, Int. J. Biol.
could have appeared to influence the work reported in this paper.
Macromol. vol. 93 (2016) 1231–1239, https://doi.org/10.1016/j.
ijbiomac.2016.09.069.
Data Availability [20] R.Y. Khalid, A.B. Fadhil, Adsorptive removal of dibenzothiophene from model fuel
over activated carbon developed by KOH activation of pinecone: equilibrium and
kinetic studies, Asia-Pac. J. Chem. Eng. vol. 19 (1) (2024), https://doi.org/
Data will be made available on request. 10.1002/apj.2976.
[21] I.K. Tetteh, I. Issahaku, A.Y. Tetteh, Recent advances in synthesis, characterization,
Acknowledgements and environmental applications of activated carbons and other carbon derivatives,
in: Carbon Trends, vol. 14, Elsevier Ltd, 2024, https://doi.org/10.1016/j.
cartre.2024.100328.
The authors would like to acknowledge funding support from the [22] L.S. Kostenko, I.I. Tomashchuk, T.V. Kovalchuk, O.A. Zaporozhets, Bentonites with
Sustainable Guyana Program, a partnership between Trent University grafted aminogroups: synthesis, protolytic properties and assessing Cu(II), Cd(II)
and Pb(II) adsorption capacity, Appl. Clay Sci. vol. 172 (2019) 49–56, https://doi.
and the University of Guyana funded by CGX Energy Inc. and Frontera org/10.1016/j.clay.2019.02.009.
Energy Corporation. [23] H. Ghasemi, M. Afshang, T. Gilvari, B. Aghabarari, S. Mozaffari, Rapid and
effective removal of heavy metal ions from aqueous solution using nanostructured
clay particles, Results Surf. Interfaces vol. 10 (2023), https://doi.org/10.1016/j.
References rsurfi.2023.100097.
[24] O.K.L. Strong, H.E. France, K. Scotland, K. Wright, A.J. Vreugdenhil, Selenite
[1] A.L. Hinwood, J. Heyworth, H. Tanner, C. McCullough, Recreational use of acidic adsorption and reduction via iron(II) impregnated activated carbon produced from
Pit Lakes—human health considerations for post closure planning, J. Water Resour. the phosphoric acid activation of construction waste wood, Arch. Environ. Contam.
Prot. vol. 04 (12) (2012) 1061–1070, https://doi.org/10.4236/ Toxicol. (2023), https://doi.org/10.1007/s00244-023-01032-y.
jwarp.2012.412122. [25] M. A. Akl, Removal of iron and manganese in water samples using activated carbon
[2] C.D. McCullough, M. Schultze, J. Vandenberg, Realizing beneficial end uses from derived from local agro-residues, J. Chem. Eng. Process Technol. vol. 04 (04)
abandoned pit lakes, Minerals vol. 10 (2) (2020), https://doi.org/10.3390/ (2013), https://doi.org/10.4172/2157-7048.1000154.
min10020133. [26] M.Á. Lobo-Recio, C. Rodrigues, T. Custódio Jeremias, F.R. Lapolli, I. Padilla,
[3] M. Søndergaard, T.L. Lauridsen, L.S. Johansson, E. Jeppesen, Gravel pit lakes in A. López-Delgado, Highly efficient removal of aluminum, iron, and manganese ions
Denmark: chemical and biological state, Sci. Total Environ. vol. 612 (2018) 9–17, using Linde type-a zeolite obtained from hazardous waste, Chemosphere vol. 267
https://doi.org/10.1016/j.scitotenv.2017.08.163. (2021), https://doi.org/10.1016/j.chemosphere.2020.128919.
[4] D. Núñez-Gómez, A.A. de A. Alves, F.R. Lapolli, M.A. Lobo-Recio, Aplication of the [27] W.K. Siabi, E.D.-J. Owusu-Ansah, H.M.K. Essandoh, N.Y. Asiedu, Modelling the
statistical experimental design to optimize mine-impacted water (MIW) adsorption of iron and manganese by activated carbon from teak and shea charcoal
remediation using shrimp-shell, Chemosphere vol. 167 (2017) 322–329, https:// for continuous low flow, Water-Energy Nexus vol. 4 (2021) 88–94, https://doi.org/
doi.org/10.1016/j.chemosphere.2016.09.094. 10.1016/j.wen.2021.02.001.
[5] M.S. Williams, T.D.T. Oyedotun, D.A. Simmons, Assessment of water quality of [28] M.E. Goher, A.M. Hassan, I.A. Abdel-Moniem, A.H. Fahmy, M.H. Abdo, S.M. El-
lakes used for recreational purposes in abandoned mines of Linden, Guyana, Geol. sayed, Removal of aluminum, iron and manganese ions from industrial wastes
Ecol. Landsc. vol. 4 (4) (2020) 269–281, https://doi.org/10.1080/ using granular activated carbon and Amberlite IR-120H, Egypt J. Aquat. Res. vol.
24749508.2019.1633220. 41 (2) (2015) 155–164, https://doi.org/10.1016/j.ejar.2015.04.002.
[6] G.K. Sharma, et al., Evaluating the geochemistry of groundwater contamination [29] J. Wang, X. Guo, Adsorption isotherm models: Classification, physical meaning,
with iron and manganese and probabilistic human health risk assessment in application and solving method, in: Chemosphere, vol. 258, Elsevier Ltd, 2020,
endemic areas of the world’s largest River Island, India, Environ. Toxicol. Pharm. https://doi.org/10.1016/j.chemosphere.2020.127279.
vol. 87 (2021), https://doi.org/10.1016/j.etap.2021.103690. [30] K.M. Scotland, S. Shetranjiwalla, A.J. Vreugdenhil, Curable hybrid materials for
[7] F. Xu, et al., Fabrication of mesoporous lignin-based biosorbent from rice straw and corrosion protection of steel: development and application of UV-cured 3-metha­
its application for heavy-metal-ion removal, J. Environ. Sci. (China) vol. 53 (2017) cryloxypropyltrimethoxysilane-derived coating, J. Coat. Technol. Res vol. 17 (4)
132–140, https://doi.org/10.1016/j.jes.2016.03.026. (2020) 977–989, https://doi.org/10.1007/s11998-019-00317-z.
[8] T. Wajima, K. Murakami, T. Kato, K. Sugawara, Heavy metal removal from aqueous [31] M. Thommes, et al., Physisorption of gases, with special reference to the evaluation
solution using carbonaceous K2 S-impregnated adsorbent, J. Environ. Sci. vol. 21 of surface area and pore size distribution (IUPAC Technical Report), Pure Appl.
(12) (2009) 1730–1734, https://doi.org/10.1016/S1001-0742(08)62480-8. Chem. vol. 87 (9–10) (2015) 1051–1069, https://doi.org/10.1515/pac-2014-1117.
[9] W. Zeng, et al., Simultaneous removal of Cd(II) and As(V) by ferrihydrite-biochar [32] J. de Celis, N.E. Amadeo, A.L. Cukierman, In situ modification of activated carbons
composite: enhanced effects of As(V) on Cd(II) adsorption, J. Environ. Sci. (China) developed from a native invasive wood on removal of trace toxic metals from
vol. 139 (2024) 267–280, https://doi.org/10.1016/j.jes.2023.04.020. wastewater, J. Hazard. Mater. vol. 161 (1) (2009) 217–223, https://doi.org/
[10] T.S. Anirudhan, S.S. Sreekumari, Adsorptive removal of heavy metal ions from 10.1016/j.jhazmat.2008.03.075.
industrial effluents using activated carbon derived from waste coconut buttons, [33] M.A. Díaz-Díez, V. Gómez-Serrano, C. Fernández González, E.M. Cuerda-Correa,
J. Environ. Sci. vol. 23 (12) (2011) 1989–1998, https://doi.org/10.1016/S1001- A. Macías-García, Porous texture of activated carbons prepared by phosphoric acid
0742(10)60515-3. activation of woods. Applied Surface Science, Elsevier, 2004, pp. 309–313, https://
[11] A.B. Fadhil, H.N. Saeed, L.I. Saeed, Polyethylene terephthalate waste-derived doi.org/10.1016/j.apsusc.2004.05.228.
activated carbon for adsorptive desulfurization of dibenzothiophene from model [34] I. Lupul, J. Yperman, R. Carleer, G. Gryglewicz, Tailoring of porous texture of
gasoline: kinetics and isotherms evaluation, Asia-Pac. J. Chem. Eng. vol. 16 (2) hemp stem-based activated carbon produced by phosphoric acid activation in
(2021), https://doi.org/10.1002/apj.2594. steam atmosphere, J. Porous Mater. vol. 22 (1) (Feb. 2015) 283–289, https://doi.
[12] Q.M. Abdulhamid, E.T.B. Al-Tikrity, A.B. Fadhil, P.J.S. Foot, Thermal cracking of org/10.1007/s10934-014-9894-4.
Al-Dora asphalt for the simultaneous production of light fuel and activated carbon [35] A. Kumar, H.M. Jena, Preparation and characterization of high surface area
for desulfurization process, J. Anal. Appl. Pyrolysis vol. 173 (2023), https://doi. activated carbon from Fox nut (Euryale ferox) shell by chemical activation with
org/10.1016/j.jaap.2023.106072. H3PO4, Results Phys. vol. 6 (2016) 651–658, https://doi.org/10.1016/j.
[13] A.A. Hussein, A.B. Fadhil, Kinetics and isothermal evaluations of adsorptive rinp.2016.09.012.
desulfurization of dibenzothiophene over mixed bio-wastes derived activated [36] S. Yorgun, D. Yildiz, Preparation and characterization of activated carbons from
carbon, Energy Sources, Part A: Recov. Util. Environ. Eff. (2021), https://doi.org/ Paulownia wood by chemical activation with H3PO4, J. Taiwan Inst. Chem. Eng.
10.1080/15567036.2021.1895372. vol. 53 (2015) 122–131, https://doi.org/10.1016/j.jtice.2015.02.032.
[14] L. Hynes, et al., Melamine adsorption on carbon materials: impact of carbon ́
[37] Harry Marsh, F. Rodriguez-Reinoso, Activated carbon, Elsevier, 2006.
texture and surface chemistry, Mater. Adv. vol. 1 (2) (2020) 262–270, https://doi. [38] T. Sebio-Puñal, S. Naya, J. López-Beceiro, J. Tarrío-Saavedra, R. Artiaga,
org/10.1039/d0ma00097c. Thermogravimetric analysis of wood, holocellulose, and lignin from five wood
[15] C. Patawat, K. Silakate, S. Chuan-Udom, N. Supanchaiyamat, A.J. Hunt, species, J. Therm. Anal. Calorim. (2012) 1163–1167, https://doi.org/10.1007/
Y. Ngernyen, Preparation of activated carbon fromDipterocarpus alatusfruit and its s10973-011-2133-1.
application for methylene blue adsorption, RSC Adv. vol. 10 (36) (2020) [39] M. Poletto, A.J. Zattera, M.M.C. Forte, R.M.C. Santana, Thermal decomposition of
21082–21091, https://doi.org/10.1039/d0ra03427d. wood: influence of wood components and cellulose crystallite size, Bioresour.
[16] H. Hwang, O. Sahin, J.W. Choi, Manufacturing a super-active carbon using fast Technol. vol. 109 (2012) 148–153, https://doi.org/10.1016/j.
pyrolysis char from biomass and correlation study on structural features and biortech.2011.11.122.
phenol adsorption, RSC Adv. vol. 7 (67) (2017) 42192–42202, https://doi.org/ [40] H. Yang, R. Yan, H. Chen, D.H. Lee, C. Zheng, Characteristics of hemicellulose,
10.1039/c7ra06910c. cellulose and lignin pyrolysis, Fuel vol. 86 (12–13) (2007) 1781–1788, https://doi.
[17] S.A.C. Carabineiro, T. Thavorn-Amornsri, M.F.R. Pereira, J.L. Figueiredo, org/10.1016/j.fuel.2006.12.013.
Adsorption of ciprofloxacin on surface-modified carbon materials, Water Res. vol. [41] S.-K. Seo, J.-S. Roh, E.-S. Kim, S.-H. Chi, S.-H. Kim, S.-W. Lee, Thermal emissivity of
45 (15) (2011) 4583–4591, https://doi.org/10.1016/j.watres.2011.06.008. a nuclear graphite as a function of its oxidation degree (2) - effect of surface
[18] F. Marrakchi, M.J. Ahmed, W.A. Khanday, M. Asif, B.H. Hameed, Mesoporous- structural changes, Carbon Lett. vol. 10 (4) (2009) 300–304, https://doi.org/
activated carbon prepared from chitosan flakes via single-step sodium hydroxide 10.5714/cl.2009.10.4.300.
activation for the adsorption of methylene blue, Int. J. Biol. Macromol. vol. 98
(2017) 233–239, https://doi.org/10.1016/j.ijbiomac.2017.01.119.

9
H.E. France et al. Sustainable Chemistry for the Environment 6 (2024) 100103

[42] G. Lota, et al., The application of activated carbon modified by ozone treatment for [50] V.C. Srivastava, I.D. Mall, I.M. Mishra, Antagonistic competitive equilibrium
energy storage, J. Solid State Electrochem. vol. 20 (10) (2016) 2857–2864, https:// modeling for the adsorption of ternary metal ion mixtures from aqueous solution
doi.org/10.1007/s10008-016-3293-5. onto bagasse fly ash, Ind. Eng. Chem. Res vol. 47 (9) (2008) 3129–3137, https://
[43] B. Li, K. Li, X. Li, Fabrication of abundantly functionalized dendritic biochar doi.org/10.1021/ie0709842.
composites as adsorbents for the high-efficiency removal of heavy metal ions and [51] G.P. Rao, C. Lu, F. Su, Sorption of divalent metal ions from aqueous solution by
dyes, Sep Purif. Technol. vol. 337 (2024) 126368, https://doi.org/10.1016/j. carbon nanotubes: a review, Sep Purif. Technol. vol. 58 (1) (2007) 224–231,
seppur.2024.126368. https://doi.org/10.1016/j.seppur.2006.12.006.
[44] Y. Zhang, X. Song, P. Zhang, H. Gao, C. Ou, X. Kong, Production of activated [52] A.O. Basheer, M.M. Hanafiah, M.A. Alsaadi, W.Z.W. Yaacob, Y. Al-Douri,
carbons from four wastes via one-step activation and their applications in Pb2+ A. Bouhemadou, Synthesis and characterization of low-cost carbon nanotubes by
adsorption: Insight of ash content, Chemosphere vol. 245 (2020), https://doi.org/ chemical vapor deposition for aluminum removal from aqueous solution, Desalin.
10.1016/j.chemosphere.2019.125587. Water Treat. vol. 214 (2021) 440–451, https://doi.org/10.5004/dwt.2021.26749.
[45] Y. Gao, et al., Preparation of high surface area-activated carbon from lignin of [53] S.A. Al-Muhtaseb, M.H. El-Naas, S. Abdallah, Removal of aluminum from aqueous
papermaking black liquor by KOH activation for Ni(II) adsorption, Chem. Eng. J. solutions by adsorption on date-pit and BDH activated carbons, J. Hazard. Mater.
vol. 217 (2013) 345–353, https://doi.org/10.1016/j.cej.2012.09.038. vol. 158 (2–3) (2008) 300–307, https://doi.org/10.1016/j.jhazmat.2008.01.080.
[46] X. Song, H. Liu, L. Cheng, Y. Qu, Surface modification of coconut-based activated [54] S. Das, S. Mishra, Insight into the isotherm modelling, kinetic and thermodynamic
carbon by liquid-phase oxidation and its effects on lead ion adsorption, exploration of iron adsorption from aqueous media by activated carbon developed
Desalination vol. 255 (1–3) (2010) 78–83, https://doi.org/10.1016/j. from Limonia acidissima shell, Mater. Chem. Phys. vol. 245 (2020), https://doi.
desal.2010.01.011. org/10.1016/j.matchemphys.2020.122751.
[47] C. Duan, T. Ma, J. Wang, Y. Zhou, Removal of heavy metals from aqueous solution [55] J.P. Vistuba, et al., Adsorption behaviour of the zeolite, Controll M.F. 574® in
using carbon-based adsorbents: A review, in: Journal of Water Process Engineering, removing iron and manganese from natural water, Desalin. Water Treat. vol. 55 (6)
vol. 37, Elsevier Ltd, 2020, https://doi.org/10.1016/j.jwpe.2020.101339. (2015) 1523–1533, https://doi.org/10.1080/19443994.2014.927333.
[48] Australian and New Zealand Environment and Conservation Council. and [56] C.L. Chostak, A. López-Delgado, I. Padilla, F.R. Lapolli, M.Á. Lobo-Recio, Use of a
Agriculture and Resource Management Council of Australia and New Zealand., waste-derived linde type-a immobilized in agarose for the remediation of water
Australian and New Zealand guidelines for fresh and marine water quality 2000. impacted by coal acid mine drainage at pilot scale, Materials vol. 16 (11) (2023),
Australian and New Zealand Environment and Conservation Council, 2000. https://doi.org/10.3390/ma16114038.
[49] V. Kocherbitov, T. Arnebrant, Hydration of lysozyme: the protein-protein interface [57] A.M. Shehap, R.A. Nasr, M.A. Mahfouz, A.M. Ismail, Preparation and
and the enthalpy-entropy compensation, Langmuir vol. 26 (6) (2010) 3918–3922, characterizations of high doping chitosan/MMT nanocomposites films for
https://doi.org/10.1021/la903210e. removing iron from ground water, J. Environ. Chem. Eng. vol. 9 (1) (2021),
https://doi.org/10.1016/j.jece.2020.104700.

10