Applied Clay Science 157 (2018) 31–40

Contents lists available at ScienceDirect

Applied Clay Science

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

Research paper

Development of Mn or Fe sulfides in the interlayer space of raw and Al- T

pillared bentonite
⁎
Luis A. Galeanoa, , Helir J. Muñoza, Ana M. Garcíaa, Antonio Gilb, Miguel Á. Vicentec
a
Research Group on Functional Materials and Catalysis, Department of Chemistry, Universidad de Nariño, Calle 18, Cra 50 Campus Torobajo, 520002 Pasto, Colombia
b
Department of Applied Chemistry, Universidad Pública de Navarra, Los Acebos Building, Campus of Arrosadía, 31006 Pamplona, Spain
c
GIR QUESCAT, Department of Inorganic Chemistry, Chemical Sciences Faculty, Universidad de Salamanca, Plaza de la Merced, s/n, 37008 Salamanca, Spain

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

Keywords: Preparation of metal sulfide nanoclusters (Mn or Fe) confined in the interlayer space of a bentonite either with or
Al-pillared clay without previous Al2O3-pillaring was investigated by H2S-sulfidization of Mn2+- or Fe2+-exchanged Al2O3-
Sulfidization pillared bentonite as potential strategy to control growth and mean final size of interlayered, MnS or FeS-like
Interlayered Mn sulfide polynuclear nanoclusters. The materials were characterized by elemental analysis, cation exchange capacity,
Interlayered Fe sulfide
low-angle powder X-ray Diffraction, N2 adsorption isotherms, simultaneous thermal analysis (DTA/TGA), diffuse
Bentonite
reflectance UV-vis spectroscopy and high-resolution X-ray photoelectron spectroscopy. Interlayered metal sul-
fides were successfully obtained in pretty short time of reaction (only ~12 h). Optimal set of preparation con-
ditions was established promoting stabilization of the interlayered metal sulfides, whereas preserving typical
layered stacking of the starting clay: (i) ratio H2S/M2+
(incorporated) = 50 mol/mol (M = Mn
2+
or Fe2+); (ii) tem-
perature of sulfidization = 100 °C; (iii) 200 °C for final thermal treatment, irrespective the type of atmosphere
used (inert-N2 or oxidizing-air) and (iv) acid-base neutralization prior to thermal treatment did not enhance
thermal stability. The most efficient Al-pillaring was achieved with 20 meq Al3+/g clay in terms of both higher
XRD 001 basal reflection and narrower distribution of alumina-pillars. HR-XPS and DR-UV-vis analyses of the
best materials with every transition metal evidenced formation of polysulfide groups and blue-shift of the ab-
sorption signal as a consequence of the modification performed on the layered aluminosilicate.

1. Introduction
Transition metal aggregates with low number of atoms have at-
tracted great interest, as constituting a very particular kind of new state
of matter with very special physicochemical properties (Galeano et al.,
2011; Viswanatha et al., 2008; Wang et al., 2009; Xia et al., 2003). The
preparation of metal nanoclusters is very thermodynamically unstable
because of their characteristic trend of unlimited growth. Recent studies
have proposed that it could be controlled by preparing them inside
confined spaces of well-defined dimensions. In this sense, it has been
suggested (Khaorapapong et al., 2009; Khaorapapong et al., 2008a,
2010, 2011) that the confinement of metal nanoclusters within the
interlayer space of clay minerals followed by annealing might lead to a
new type of expanded clay minerals with similar final structure as the
conventional pillared clays (Barrault et al., 2000). Likewise, it may
allow higher stability as well as control on the size of the grown metal
aggregates, thanks to the tailored dimensions of the interlayer space in
either raw or previously modified hosting aluminosilicate (Pillared In-
terlayered Clay, PILC) (Iacomi et al., 2006). Thus, the immobilization of
metal sulfides within restricted spaces including those featured by clay
minerals has become very attractive. Hence, incorporation of metal
sulfides in silica (Hirai et al., 1999; Zhang et al., 2001), clay minerals
(Stramel et al., 1986) and polymers (Antolini et al., 2005; Hirai et al.,
2001) has been reported. Thin-films of metal sulfides have also been
prepared by several methods for high-performance optical devices
(Chavhan and Sharma, 2005; Öznülüer et al., 2006; Su and Choy,
2000). The metal sulfides at the nanoscale have been reported to dis-
play unique optical and electrical properties, as well as a variety of
potential applications in electroluminescence (Lou et al., 2003) and
non-linear optical devices (Chin et al., 2001). It is well-known that
physical and optical properties of these metal sulfides depend in a great
extent on the particle size; therefore, their successful trapping within
confined spaces may indirectly control their final properties and ap-
plications (Khaorapapong et al., 2009).
As long as the presence of sulfur in fossil-based fuels is still one of
the main causes of environmental pollution in many countries, the
development of catalytic systems able to deplete such contents in fuels
up to levels environmentally acceptable has become very important. In

⁎
Corresponding author.
E-mail address: alejandrogaleano@udenar.edu.co (L.A. Galeano).

https://doi.org/10.1016/j.clay.2018.02.022
Received 1 December 2017; Received in revised form 13 February 2018; Accepted 14 February 2018
0169-1317/ © 2018 Elsevier B.V. All rights reserved.
L.A. Galeano et al. Applied Clay Science 157 (2018) 31–40

this sense, some transition metal sulfides, mainly cobalt, nickel, mo- Table 1
lybdenum and/or tungsten-based have been successfully employed as Physicochemical properties of manganese-modified solids and starting mineral: SiO2-
normalized contents of Al2O3 and MnS, cationic exchange capacity, compensation of CEC
very active phases of hydrotreating catalysts (Infantes-Molina et al.,
and XRD-basal d-value.
2012). It undoubtedly constitutes another field where nanosized metal
sulfides might find interesting catalytic application. Likewise, Romero Sample Content (wt%)a CECb CCc (%) d001d
(Romero-Pérez et al., 2012) prepared supported RuS2 on Al-pillared (meq/ (nm)
clays in order to study their performance as catalysts of the di- Al2O3 MnS 100 g)

benzothiophene hydrodesulfurization and the influence of the added Al BV-E 31.82 0.04 82 – 1.56
on the catalytic activity. So far, the preparation of several polynuclear AlP10 39.75 0.08 52 37 D
metal sulfides on variety of supports including montmorillonites, zeo- AlP20 41.35 0.12 37 55 1.83
lites and silicas has been developed using various experimental ap- AlP30 47.42 0.13 8 90 1.72
AlP20-Mn 40.75 7.28 45 45 1.63
proaches. It is worth mentioning the solid-state reaction as reported by
AlP20-Mn(5)S100 – – 15 66 1.53
Khaorapapong et al. (Khaorapapong et al., 2009; Khaorapapong et al., AlP20-Mn(50)S100 – – 7 84 1.60
2008a, 2010, 2011; Khaorapapong et al., 2008b) and Kabilaphat et al. AlP20-Mn(500)S100 – – 29 35 D
(Kabilaphat et al., 2015), who introduced the targeted metal by cation AlP20-Mn(5)S100-350 – – 37 18 1.52
N2
exchange on the montmorillonite as a first step, followed by hand
AlP20-Mn(50)S100-350 – – 9 80 –
grinding with solid sodium sulfide in agate mortar, yielding the sulfide N2
product after four months. Meanwhile, Iacomi et al. (Iacomi et al., AlP20-Mn(500)S100- – – 4 91 D
2005) carried out almost the same reaction but in solid-liquid system 350 N2
using zeolites as supports, metal ion exchange at 23 °C and final sulfi- AlP20-MnS50-200 N2 40.70 3.63 9 80 1.49
AlP20-MnS100-200 N2 41.09 2.70 8 90 1.59
dization with dissolved Na2S at room temperature, employing a solid to
AlP20-MnS-150-200 N2 37.78 3.78 10 88 1.57
liquid ratio (s/l = 1/50) for 12 h. A little bit later, the same authors AlP20-MnS400 41.17 3.58 17 79 D
(Iacomi et al., 2006) reported formation of MnS clusters within lau- AlP20-MnS100-200Air 44.71 3.03 7 91 1.61
montite holes by using almost the same procedure but slightly changing AlP20-MnS100-400 N2 38.86 2.89 14 82 D
AlP20-MnS100-400Air 44.27 3.29 15 82 D
the sulfidization step to H2S (40 cm3/min) at 77 °C along 12 h. This
AlP20-MnS100-n-200 N2 38.58 3.80 13 84 D
procedure was later adapted (Galeano et al., 2011) introducing poly- AlP10-MnS100-200 N2 44.44 2.55 54 34 D
nuclear aggregates of MnS in raw calcium-rich bentonite. High metal AlP30-MnS100-200 N2 50.38 2.31 68 17 1.58
uptake efficiency was achieved, with high specific location of the metal BV-Mn 30.17 4.92 29 64 D
clusters in the clay's interlayer space and interesting performance of BV-MnS100-200 N2 28.63 3.91 60 27 D

final solids activating catalytic wet peroxide oxidation of aqueous me- a

thyl orange with hydrogen peroxide, at room temperature and pressure.
Accordingly, this work was devoted to study the influence of the
following experimental parameters on the final structure of sulfided
Fe2+ or Mn2+-exchanged bentonite, with or without previous pillaring
with Keggin-like (Al13)7+ ions: (i) molar ratio H2S/M2+(incorporated)
(M = Mn2+or Fe2+ (incorporated) in the range 5–500); (ii) temperature of
sulfidization (50–400 °C); (iii) temperature and atmosphere of thermal
treatment (200–400 °C; N2 or air); (iv) acid-base neutralization prior to
thermal treatment; (v) different density of pillars on the starting ex-
panded clay and (vi) use of a common ramp for H2S streaming and final
thermal treatment.
2. Experimental section
2.1. Materials
A Colombian bentonite previously refined twice by particle size
(≤2 μm) separation, with cation exchange capacity, CEC, of 82 meq/
100 g (dry-basis) was used as starting mineral (denoted BV-E). The
elemental composition (X-ray Fluorescence, XRF, dry-basis) was de-
termined as explained later. The material was modified using
AlCl3.6H2O (99%, Sigma-Aldrich®), Na2S·9H2O (technical grade, TC
Químicos), MnCl2.4H2O (98%, Sigma-Aldrich®), FeSO4.7H2O (99.5%,
Panreac®), HCl (37%, Panreac®), N2 (99.999%, Cryogas®) and synthetic
air (mixture 79.0% of N2 99.999% and 21% of O2 99.6 +/− 0.5% of
absolute accuracy, Cryogas®), all used as received.
2.2. Preparation of modified materials
In order to assess the effect of the pillar density, the starting mineral
(BV-E) was interlayered and pillared (thermal treatment) by Al using
three nominal loadings of 10 (AlP10), 20 (AlP20) or 30 (AlP30) meq
Al3+/g clay, mainly represented as (Al13)7+ Keggin-like polycations
(the intercalated materials, prior of the final heating step, were denoted
as AleI10, AleI20 and AleI30). The preparation of the pillared clays
was carried out following a standard procedure described elsewhere
All contents are normalized to the SiO2 content in the starting clay BV-E in order to
rule-out dilution effect produced by incorporation of either Al, S and Mn upon mod-
ification. Determined by atomic absorption spectroscopy (AAS), dry basis.
b
CEC: Cationic Exchange Capacity (dry-basis).
c
CC: Compensation of CEC.
d
Obtained from oriented specimens. D: Delaminated material.
(Galeano et al., 2010). First of all, the intercalating agent was prepared
by hydrolysis of 0.2 mol/dm3 AlCl3.6H2O at 70 °C under gentle stirring
with 0.2 mol/dm3 NaOH poured dropwise to get a final ratio OH−/
Al3+ = 2.4. The mixture was then further aged under stirring at the
same temperature for 2 h. Afterwards, the solution was dropped on
2.0% w/v dispersion of BV-E bentonite previously swollen in distilled
water under stirring for 2 h at room temperature. The resulting inter-
layered material was washed in dialysis tubing cellulose membrane
(average flat width 43 mm, Sigma®) until the conductivity of the
washing water decreased below 35 μS/cm, dried at 60 °C and finally
heated at 500 °C for 2 h.
The starting clay as well as the three pillared clays were then em-
ployed as supports to grow and immobilize either the Mn (yielding
AlPD-MnSX-Tatm series of materials, where D = pillar density;
X = temperature of sulfidization; T = temperature of final thermal
treatment; atm = atmosphere used in final thermal treatment) or the Fe
sulfide clusters (yielding AlPD-FeSX-Tatm series of materials) (see
Tables 1 and 2, respectively). In the case of Mn-series, an extra group of
samples was first prepared in order to determine the optimal molar
ratio H2S/(Mn2+ or Fe2+)(incorporated); for instance, sample AlP20-Mn
(5)S100-350 N2 was prepared from AlP20-pillared bentonite, sulfidized
at 100 °C under H2S/(Mn2+)(incorporated) molar ratio = 5.0 and finally
thermally treated at 350 °C under nitrogen as inert atmosphere.
The Mn2+ and Fe2+ ions were introduced in the minerals (either,
starting or pillared) by cation exchange. The procedure was adapted
from the one originally reported by Iacomi et al. (Iacomi et al., 2005) to
develop clusters trapped inside zeolite structures. It began with cation
exchange of every solid with aqueous 0.5 mol/dm3 MnCl2.4H2O or
FeSO4.7H2O (5.0 cm3 solution/g clay) throughout 8 h under constant
stirring at room temperature. In this research, the starting

32
L.A. Galeano et al. Applied Clay Science 157 (2018) 31–40

Table 2 Fe2+)(incorporated), several fractions of the materials were treated under

Physicochemical properties of iron-modified solids: SiO2-normalized content of FeS, ca- constant flow of H2S (as a function of the metal content introduced in
tionic exchange capacity, compensation of CEC and XRD-basal d-value.
every case) at various temperatures of 50, 100 or 150 °C throughout 8 h.
Sample FeS contenta (wt CECb (meq/ CCc (%) d001d (nm) In order to establish the effect of the one-pot sulfidization and final
%) 100 g) thermal treatment, another sample was sulfidized from RT to 400 °C
under a ramp of 6.0 °C/min; the solid stood two extra hours at such final
BV-E 2.56 82 – 1.56
temperature under constant flow of N2 (AlP20-MnS400; AlP20-
AlP20-Fe 5.83 36 56 1.63
AlP20-FeS50-200 N2 5.01 2 98 1.56 FeS400). The constant flow of hydrogen sulfide generated as explained
AlP20-FeS100-200N2 7.88 2 98 1.51 above was first passed through a CaCl2 trap in order to remove any
AlP20-FeS-150-200 5.35 6 93 1.40 moisture present before reaching the solid. A fraction of every sulfided
N2 solid was then treated under either oxidizing (synthetic air) or inert
AlP20-FeS400 7.70 11 87 D
atmosphere (N2) at 200 °C or 400 °C/2 h. Another fraction of every
AlP20-FeS100- 9.61 6 93 D
200Air sulfidized solid (previously Al-interlayered) was also treated by basic
AlP20-FeS100-400 6.64 6 93 D neutralization to establish the effect of the acid-base neutralization
N2 before the final thermal treatment, as follows: a dispersion of the solid
AlP20-FeS100- 7.61 6 93 D
(2.0% w/v) in 0.2 mol/dm3 of NaOH was kept under constant stirring
400Air
AlP20-FeS100-n-200 7.40 55 33 1.38 along 12 h. The materials were then filtered, repeatedly washed with
N2 distilled water and finally dried at 60 °C. Later, such materials were also
AlP10-Fe 5.81 53 36 D heated at 200 °C under a nitrogen atmosphere for 2 h; final samples are
AlP30-Fe 6.62 7 92 1.70 denoted as AlP20-MnS100-n-200 N2 and AlP20-FeS100-n-200 N2.
AlP10-FeS100-200 5.90 54 34 D
N2
AlP30-FeS100-200 4.50 6 93 1.60 2.3. Physicochemical characterization
N2
BV-Fe 5.31 36 56 1.51 The elemental analyses for Al, Si, Mn and Fe were made using
BV-FeS100-200 N2 5.34 4 95 D
atomic absorption spectroscopy (AAS) in a Perkin Elmer 2380 appa-
a ratus, on samples previously digested in a volumetric mixture (1.0 cm3/
Fe content normalized to the SiO2 content in the starting clay BV-E in order to rule-
out dilution effect produced by incorporation of Al, S and Fe upon modification. 4.0 cm3) of HF (40% v/v, Panreac®)/HNO3 (65% v/v, JT Baker®) per
Determined by atomic absorption spectroscopy (AAS), dry basis. every 10.0 mg of dry solid sample. Starting material BV-E was further
b
CEC: Cationic Exchange Capacity (dry-basis). analyzed by X-ray Fluorescence in a Bruker S8 TIGER equipment of
c
CC: Compensation of CEC. 4 kW dispersive wavelength, Rh anode as source of X-rays provided
d
Obtained from oriented specimens. D: Delaminated material. with scintillation (heavy elements from Ti to U) and flow (light ele-
ments from Na to Sc) detectors. The materials AlP20, AlP20-MnS100
aluminosilicate was homoionized with two different precursor salts and AlP20-FeS100 were analyzed by wavelength dispersive X-ray
(MnCl2.4H2O or FeSO4.7H2O) which in solution contained the Mn2+ or fluorescence spectroscopy (WDXRF) in AXIOS mAX PANalytical appa-
Fe2+ ions, respectively; the oxidation state (2+) was then assumed to ratus equipped with 4 kW Rh anode as source of X-rays; elements from
remain for the ions intercalated in the interlayer space of the alumi- Na to U were detected and semi-quantitative analyses performed by
nosilicate. In the case of iron this assumption was supported in the using software SuperQ versión 5.0 L – OMNIAN database standards, in
results of the HR-XPS analysis displayed latter, but also taking into order to check effective ratios S/Mn and S/Fe stabilized in these ma-
account the data available in the literature about the thermodynamics terials. UV–vis diffuse reflectance absorption spectra of powdered
(Beverskog and Puigdomenech, 1996) and kinetics (Morgan and Lahav, samples of AlP20-MnS100 and AlP20-FeS100 (60 mesh) were also
2007) of the very slow transformation of Fe2+ (aq.) ➔ Fe(aq.) under the pH
3+
measured in a Shimadzu UV–2600 spectrophotometer using the ISR-
of the used solution (2.78). Finally, the dispersion was left to stand and 2600Plus integrating sphere.
the supernatant removed; the cycle was repeated once and the treat- The CEC of the materials was determined by saturation of every
ment finished with a third cycle where the cation exchange was made at solid under reflux with 2.0 mol/dm3 ammonium acetate solution (97%,
60 °C for 2 h. Final exchanged materials were then washed again in Carlo Erba®), 45 cm3 solution/g solid, followed by several cycles of
dialysis tubing until no presence of chlorides (Ag+ testing, in the case of washing with distilled water and centrifugation to remove any excess of
the Mn-derived material) or sulfates (Ba2+ testing, in the case of Fe- adsorbed ammonium ions. The amount of NH4+ cations retained was
derived material) was evidenced. The effect of the used molar ratio then determined by micro-Kjeldahl analysis.
H2S/Mn2+ (incorporated) was assessed in values of 5, 50 and 500, keeping Powder X-ray diffraction patterns were determined in a Bruker D8
constant the following conditions of reaction: temperature of H2S - Advance diffractometer at 40 kV and 30 mA under a scanning rate of
sulfidization = 100 °C (Hydrogen sulfide was generated by slow addi- 2.29° 2θ/min, using filtered Cu Kα radiation (λ = 0.15416 nm).
tion under continuous stirring of an excess of concentrated HCl (37%) Modified clays were analyzed in the range 2–30°2θ, whereas BV-E
on the saturated solution of Na2S·9H2O, reaction 1); final thermal starting mineral was analyzed from 2 to 70°2θ.
treatment at 350 °C/2 h under N2 flow. As long as the treated solids Textural analyses were performed on N2 adsorption isotherms at
exhibited a clear tendency towards delamination upon sulfidization at −196.15 °C in ASAP 2010 HD Micromeritics Sorptometer in a broad
350 °C, probably because of the aggressive acid conditions of treatment range of relative pressures on samples (100–150 mg) previously de-
(see 001 reflection at Fig. S2), one higher (400 °C) and one lower gassed at 200 °C/12 h. The specific surface areas (SBET) were de-
(200 °C) temperatures were also tested. These samples were prepared termined by the multipoint BET model; the external surfaces (SExt),
only for Mn and denoted as AlP20-Mn(5/50/500)S100-350 N2 and the micropore volumes (Vμp) and micropore surfaces (Sμp) by means of the
best molar ratio extrapolated for the Fe system. Output stream of gases t-plot.
was bubbled through a NaOH 10 mol/dm3 trap solution before the exit Simultaneous thermal analyses (DTA/TGA) were recorded in a SDT
to the environment in order to remove any fraction of not reacted Q600 TA apparatus for the samples pillared with 20 meq Al3+/g clay,
H2S(g). then exchanged with either Mn2+ or Fe2+ and sulfidized at 100 °C
Na2S(s) + 2HCl (l) → 2NaCl (l) + H2 S(g) (1) (without final thermal treatment). Around 10 mg of sample (dry basis)
were heated at 10 °C/min under a constant flow of 20 cm3/min of
Once determined the best molar ratio H2S/(Mn2+ or oxygen (Air Liquid, 99.999%) from room temperature to 1000 °C.

33
L.A. Galeano et al. Applied Clay Science 157 (2018) 31–40

Finally, high-resolution X-ray photoelectron spectroscopy was used Table 3

to analyze samples AlP20-MnS100-200 N2 and AlP20-FeS100-200N2, in Textural properties of starting clay (BV-E) and selected Mn- or Fe- sulfide interlayered
materials.
a Physical Electronics 548 XPS – Auger Spectrometer, with a mono-
chromatic Mg Kα (1253.6 eV) source. Deviations caused by electric Sample Textural properties
charges on the samples were corrected employing the line C 1 s at
284.8 eV as internal standard. The elemental abundances were de- SBETa (m2/ SExtb (m2/ Vμpc (cm3/ Sμpd (m2/g)
termined from the areas below every relevant peak in the high-re- g) g) g)

solution spectra, operating in large area analysis with incident radiation BV-E 85 ND – –
of 100 W, 200 μm of scanning diameter on a 1400 × 200 μm cross AlP20 112 21 0.042 90
section of sample. Both spectra were obtained with a pass energy of AlP20-MnS50-200 N2 88 20 0.032 68
23.5 eV. AlP20-MnS100-200 N2 92 11 0.036 78
AlP20-MnS-150-200 N2 35 12 0.011 23
AlP20-FeS50-200 N2 50 23 0.013 27
3. Results and discussion AlP20-FeS100-200N2 60 11 0.023 49
AlP20-FeS-150-200 N2 41 10 0.015 31
3.1. Physicochemical properties of the materials and effect of al-pillar
a
density SBET: Specific surface area determined by multipoint BET model.
b
Sext: External specific surface area.
c
Vμp: Micropore volume.
The XRF full elemental composition of the starting BV-E material is d
Sμp: Micropore surface area. ND: Not determined.
in terms of oxides (% w/w): MgO 2.10%; CaO 0.30%; K2O 0.40%; Na2O
0.20%; Al2O3 20.3%; SiO2 46.0% and MnO 0.03. This composition was materials was achieved in solid AlP20-Mn(50)S100 (basal d-
consistent with the typical range widely reported for montmorillonites value = 1.60 nm), showing that under that ratio the stabilized metal
(Galeano et al., 2010; Newman, 1987) as well as the diffraction d060 nanoclusters allowed still stable layered stacking in the host inorganic
signal at 61.90° (2θ) (d = 0.150 nm) characteristic of dioctahedral 2:1 mineral. Apparently, under the lowest H2S/Mn2+ (incorporated) molar ratio
clay minerals (Thorez, 1976) and also the CEC of this material (82 meq/ (5.0) there was not enough hydrogen sulfide to promote significant
100 g) was within the typical interval reported for smectites like growth of the interlayered nanoclusters (basal d-value = 1.53 nm) but
montmorillonites (70–130 meq/100 g) (Moreno et al., 1999; Newman, strongly affected the stacking of the already pillared structure (Fig. S2).
1987). The pillared AlP20 material showed 0.06% CaO, 0.42% K2O and The highest H2S/Mn2+ (incorporated) molar ratio (500) also affected very
0.14% Na2O, indicating that the modification was mainly carried out significantly the host layered structure, the 001 reflection almost
via cation exchange of the starting content of Ca2+ and Na+ by Keggin- completely disappeared, in this case probably because of the extremely
like polycations (Al13)7+, while K could be in the solid taking part of high abundance of the acid sulfidizing agent. Regarding this, it seems
non-swellable clayey phases, probably K-feldspars (Kloprogge, 1998). that the growth of the clusters by reaction with H2S(g) is accompanied
Regarding the effect of the Al-pillar loading, the AAS elemental by release of protons, which get mostly retained at the interlayer of the
composition of the pillared solids showed that Al uptake was propor- aluminosilicate. It was suggested in our previous work, that such a
tional to the amount of this element in the solutions (Table 1). The same proton content may destabilize the aluminosilicate layers probably
trend was observed for the compensation of the cation charge (CC) after leading to delamination of the clay, in turn giving rise to the so-called
pillaring, that is, the fraction of the CEC effectively compensated by the house of cards structure (Galeano et al., 2011). The acid treatment is
polycations: AlP10 showed the lowest CC and AlP30 the highest. Al- probably able to leach out metals from the octahedral sheets of smec-
though all used pillaring loadings of Al in solution were in clear excess tites leading to progressive loss of the stacking pattern, as reported also
(2.19 times for AlP10; 4.38 times for AlP20 and 6.56 times for AlP30) in by other authors (Steudel et al., 2009; Toranzo et al., 1998; Vicente
comparison with the experimental CEC of the starting material BV et al., 1996). AlP20-Mn(50)S100 also displayed 80 and 84% of CC upon
when taking into account a mean nominal positive charge carried by sulfidization and thermal treatments respectively, whereas H2S/
every atom of the metal in the most probable pillaring Keggin-like Mn2+(incorporated) ratios of either 5 or 500 led to 66 and 35% upon sulfi-
adduct ((7+)/13 Al atoms = ~0.54+), apparently the largest excess dization and 18 and 91% after thermal treatment, respectively. This
did not longer enhance the stability of the interlayered clay upon confirmed that H2S/Mn2+ (incorporated) = 50 provided the best balance be-
heating, but instead led to a disorganization in the Al2O3 interlayer, tween H2S required for the growing of the transition metal nanoclusters
conducting in turn to a basal reflection poorly defined. Therefore, a at the interlayer of the clay and the amount of sulfidizing agent that the
loading of 20 meq Al3+/g clay provided the best conditions to stabilize hosting PILC might support without undergoing significant lack of ty-
the Al-pillared form of the mineral, in good agreement with several pical layer stacking. A change in the color of the solids from brown to
literature reports (Galeano et al., 2011; Salerno et al., 2001; Vicente dark brown was also observed upon sulfidization as an indirect evi-
et al., 2004); indeed, this material displayed 55% of CC (Table 1), SBET dence of change in the oxidation state of the interlayered Mn ions
of 112 m2/g, Vμp of 0.042 cm3/g and SExt relatively low of 21 m2/g (Khaorapapong et al., 2008a). It is noteworthy that such interlayer
(Table 3). phases of MnS may resemble some crystal structures well known as
magnetic semiconductors of either rock-salt type (α-MnS: stable, octa-
3.2. Effect of the molar ratio H2S/Mn2+
(incorporated) hedral-coordinated, polymorph at room temperature), zinc blende (β-
MnS) or wurtzite (ϒ-MnS); the last two ones being metastable, tetra-
First of all, the most proper amount of hydrogen sulfide per unit hedral-coordinated phases (Pathan and Lokhande, 2004; Skromme
mass of incorporated Mn2+ promoting the growth of the metal clusters et al., 1995; Takahashi and Mitsui, 1996).
was established using AlP20 by the reasons above explained and also
since such a pillar-density exhibited high efficiency of the Mn2+ ions
incorporation (Table 1) (Galeano et al., 2011). Regarding this feature, it 3.3. Effect of the temperature of sulfidization on the growth of interlayered
can be seen that AlP20 achieved significant uptake of Mn2+ ions by MnS/FeS nanoclusters
means of the homoionization treatment (see MnS content in sample
AlP20-Mn at Table 1); indeed, the amount of the metal exchanged in The temperature of sulfidization is a key parameter in this proce-
this material exceeded the nominal CEC available in the AlP20 pillared dure, because it may govern in great extent the rate of diffusion of the
clay, even assuming the lower oxidation state Mn2+ being stabilized in reacting gas within the interlayer space of the solids where Mn2+ or
the solid material. The highest layer expansion within this series of Fe2+ were mostly hosted, thus either promoting or inhibiting the

34
L.A. Galeano et al. Applied Clay Science 157 (2018) 31–40

growth of the metal sulfur clusters (Galeano et al., 2011). In general,

the Mn-interlayered materials sulfidized at different temperatures did
not display significant differences in their final basal reflections; but a
strong effect was observed over the pillar distribution by comparing
AlP20-MnS100 and AlP20-MnS150 solids (Fig. S3), since the material
AlP20-MnS50 showed a higher full width at half maximum-fwhm index
(related with more widely size-distributed interlayered species). This
suggests that temperatures of sulfidization between 100 and 150 °C
promoted more successful diffusion of H2S within the interlayer space
of the clay without significant negative effect on the stability of either
the aluminosilicate layers or Al2O3 pillars. Therefore, the solid sulfi-
dized at 100 °C exhibited the highest stability upon final thermal
treatment, whereas the Mn-material sulfidized at 400 °C underwent full
delamination. Finally, irrespective the nature of the interlayered tran-
sition metals here studied (Mn or Fe), the highest cation compensation
upon sulfidization was found at 100 °C (Tables 1 and 2). However, it
must be stressed that series of materials modified with Fe (Fig. S4)
resulted significantly more affected in the host-layered structure within
the full range of temperatures evaluated, effect that could be ascribed to
stronger acid properties of Fe2+ in comparison to Mn2+ (Li et al.,
2017). The contents of S, Mn, Fe and other elements of the samples
upon sulfidization (AlP20-MnS100 and AlP20-FeS100) were analyzed
by X-ray fluorescence (XRF). The atomic percentage of Mn in sample
AlP20-MnS100 was 0.89, whereas the atomic percentage of Fe in
AlP20-FeS100 was 3.67. It led to effective atomic ratios of 4.21 (3.75/
0.89) and S/Mn and 2.37 (8.71/3.67) for S/Mn and S/Fe, respectively.
First of all, it demonstrates the materials fixed sulfur in their structure
upon the sulfidization treatment here applied. In addition, the effective
atomic ratios obtained where within the range of 0.06–20 reported by
Iacomi et al. (Iacomi et al., 2006), although such authors modified a
zeolite with 10 times higher S/Mn molar ratio (molar ratio S/
Mn = 500) compared to the ratio used in this research. In spite of S/M
atomic ratios expected for both bulk metal chalcogenides were sig-
nificantly lower than those obtained, the extra amount of S fixed by the
structures can be attributed to other metals present in the materials
forming highly insoluble sulfides like those of Ti, Zn and Zr and others
only slightly soluble including those of Ca or Mg. Fig. 1. Thermal diagrams (DTA/TGA) of (a) AlP20-MnS100 and (b) AlP20-FeS100.

DTA signals on simultaneous thermal analyses of these materials

(Fig. 1) showed strong exothermic peaks at Tmax in the range
210–240 °C for both metals; accompanied by mass loss (air used as
carrier gas), reason by which it could be attributed to oxidation of the
MnS or FeS aggregates, mostly interlayered, probably involving evo-
lution of sulfur oxides (SO2 and/or SO3), as reported in previous studies
(Crabtree, 2005). The mass loss in the range 195–268 °C for AlP20-
MnS100 (2.0%) was exactly 3 times higher against that expected for the
transformation of MnS to bixbyite (Mn2O3), computed on the basis of
the Mn content measured in this solid. On the other hand, the mass loss
in the range 138–274 °C for AlP20-FeS100 was fairly higher (> 100
times) than that expected for the transformation of FeS to Fe2O3 or
Fe3O4 (Chen and Yan, 1997), as before taking into account the ex-
perimental Fe content (Table 2). Moreover, the exothermic peak at
Tmax = 346 °C in the thermal diagram of AlP20-FeS100, accompanied
by a slight mass loss should correspond to a phase transition from the
precursor oxide formed around 200 °C (hematite-like nanoparticulate
aggregates, Fe2O3) to the Fe3O4 phase (magnetite), which was reported
taking place close to 350 °C at low partial pressures of oxygen (Ketteler
et al., 2001). The exothermic peaks at Tmax = 908 °C or Tmax = 916 °C
for modified solids with Mn and Fe, respectively, occurred without
change in mass, and can be attributed to loss of the layered pattern of
montmorillonite, to form mullite and silica; in this sense, the thermal
diagram of solid AlP-20 displayed the same exothermic peak centered
at 909 °C (not shown), which is closer to the value exhibited by the Mn-
sulfided Al-PILC than for the Fe-sulfidized one, suggesting higher in-
fluence of the Fe nanoclusters on the stability of the layered stacking
featuring the hosting aluminosilicate. The mass loss involved in the first
thermal event of the Fe-modified material (AlP20-FeS100, 4.1%) was
around twice that observed for the Mn-modified material (AlP20-
MnS100, 2.0%); however, that probably was more related to the con-
tribution by the evolved water overlapping such mass loss, given the
lower temperature from which the thermal event was measured in
sample AlP20-FeS100 (138 °C in comparison to 198 °C for the Mn-solid)
than to differences in the metal contents (Tables 1 and 2). It must be
realized the tiny difference in the molar masses of the starting and final
iron phases between which the phase transition apparently took place.
Finally, all the textural properties significantly decreased in the
sulfidized materials for both metals in comparison with the starting Al-
PILC (AlP20), irrespective the temperature of sulfidization (Table 3).
This can be due to the blocking of the porosity previously developed
during Al-pillaring by the metal sulfides. As it can be seen, the micro-
porous network featured by PILCs was seriously affected, but clearly
less significantly when the sulfidization was carried out at 100 °C.
Moreover, it was remarkable that previous pillaring of the layered
structure allowed to preserve a significant fraction of the micropores
still after the sulfidization step, which did not occur when the same type
of layered structures were sulfidized without prior pillaring, as reported
in advance (Galeano et al., 2011). Therefore, the H2S-sulfidization at
100 °C was the most proper treatment in order to achieve the growth of
either Mn or Fe interlayered aggregates without significant loss of
layered structure and textural properties featured by PILCs. As before,
textural properties got more affected in Fe sulfide materials, although
showing the same above-mentioned trend as a function of the tem-
perature of reaction. Accordingly, from now on the solids modified with
Mn or Fe were sulfidized at 100 °C (AlP20-MS100, M: Mn o Fe) and then

35
L.A. Galeano et al.
Fig. 2. Powder X-ray diffraction patterns from materials modified with Mn2+ as a
function of temperature/atmosphere of final treatment and basic neutralization pre-
treatment.
employed as starting materials for preparation of the remaining samples
in order to estimate: (i) The effect of temperature and type of atmo-
sphere used in the final thermal treatment, and (ii) The effect of basic
neutralization prior to thermal treatment.
3.4. Effect of temperature and type of atmosphere used in the final thermal
treatment
The final thermal treatment may be important in the purpose of
anchoring the metal sulfide nanoparticles to the hosting clay. Final
thermal treatment of the sulfidized solids was assessed under either
inert (N2) or oxidizing atmosphere (synthetic air) at 200 or 400 °C.
According to powder X-ray diffraction patterns, the Mn-solids modified
at 200 °C (Fig. 2) preserved their layered structure with basal reflections
a little bit lower in comparison with the starting Al-PILC (close to
1.60 nm), irrespective the type of atmosphere used, as well as CC values
very similar in both cases (Table 1). Therefore, taking also into account
the thermal analysis diagram discussed above (Fig. 1a), it can be in-
ferred that MnS-like nanoclusters developed within the layered host
phase were stable up to around 200 °C, even in the presence of oxidizing
atmosphere. Meanwhile, the solid modified with Fe at 200 °C showed
almost full delamination of its structure even under inert, nitrogen at-
mosphere (AlP20-FeS100-200N2, Fig. 3). Meanwhile, the materials BV-
E, AlP20-MnS100-200 N2, AlP20-FeSS100-200 N2 were analyzed by
DR-UV-vis spectroscopy and compared with bulk-MnS and bulk-FeS
phases. First of all, the spectrum at Fig. 4a shows that bulk-MnS pre-
sents a wide absorption associated with high metallic sulfide con-
centration and vibrational coupling (Lu et al., 2001; Trindade et al.,
36
Applied Clay Science 157 (2018) 31–40
Fig. 3. Powder X-ray diffraction patterns from materials modified with Fe2+ as a function
of final treatment temperature (inert N2 atmosphere).
1997), as well as wide particle size distribution. The absorption edge
reported for MnS ranges between 325 and 340 nm as a function of the
particle size, with a blue-shift for nanoparticles (Lu et al., 2001),
whereas the absorption edge for FeS is entered around 226 nm (Wood,
1979). In the other hand, montmorillonite has shown intense absorp-
tion peak close to 250 nm (Yu and Yang, 2010), in good agreement with
the one observed in starting clay BV-E. As it can be seen, both Mn- and
Fe sulfidized AlP20 materials displayed blue-shift in the absorption over
around 340 nm and 300 nm respectively, which could be ascribed to
incorporation of very small particles of the metal sulfides in the struc-
ture (Fig. 4). In addition, recorded absorption increase over these re-
gions is clearly more pronounced in the case of AlP20-FeS, as expected
from the higher final loading of Fe in comparison to Mn as shown by the
elemental analyses.
The Mn- and Fe-materials once heated at 400 °C got completely
delaminated irrespective the type of atmosphere used, as suggested by
the almost total loss of the 001 reflection (see Fig. 2 and 3). This can be
explained by the partial disaggregation/decomposition of the metal
sulfidized nanoclusters by thermal effect, but may also be related to
acid-attack by the protons that probably remained in the structure after
sulfidization; as described before, it is expectable that residual CEC in
final materials was compensated by protons provided by the thorough
H2S streaming, effect that could get reinforced at high temperature.
This loss of stability occurred at lower temperatures in Fe-modified
solids, although also accompanied by a smaller loss in the fraction of
compensated layer-charge between 200 and 400 °C as compared with
the Mn-materials (Tables 1 and 2). The main difference in the pattern of
charge compensation between the Mn-sulfidized raw mineral (not pil-
lared in advance) reported before (Galeano et al., 2011) and the solid
previously pillared here reported should be related with the absolute
L.A. Galeano et al.
Fig. 4. Diffuse reflectance UV–vis absorption spectra of (a) BV-E, Bulk MnS and AlP20-MnS100-200 N2 and (b) BV-E, Bulk FeS and AlP20-FeS100-200N2.
final CEC being significantly lower in the pillared starting solid, since
the CEC compensated by Al2O3-pillars in AlP20 (about 37 meq/100 g)
must be added to that compensated by the sulfide metal aggregates.
However, in both cases the CEC of the materials increased slightly as
the temperature used in the final thermal treatment increased from 200
to 400 °C; moreover, this effect was clearly less noticeable in the Fe-
modified materials (Table 2).
X-ray photoelectron spectroscopy (XPS) of AlP20-MnS100-200 N2
and AlP20-FeS100-200N2 samples allowed to infer the preferential
position of the grown nanoclusters of each metal. In the Mn-material,
the presence of the metal sulfide was not detected (Mn 2p3/2 in MnS
characteristic signal absent, Fig. 5a), which suggested that MnS poly-
nuclear aggregates were preferentially stabilized within the interlayer
of the clay through the insertion strategy via cation exchange of the
metal, without affecting the structure of the Al2O3-pillar or its dis-
tribution (as shown above by XRD analyses). Meanwhile, the signal Fe
2p3/2 for sample AlP20-FeS100-200N2 (Fig. 5b) allowed to infer that at
least certain fraction of the incorporated iron was fixed on the external
surface of the clay mineral layers and then, probably for that reason, it
was more easily detected by this surface-technique. However, another
portion of such detected Fe content should also correspond to extra-
structural iron already present in the starting clay (BV-E), in which
1.63% Fe was found compared to Fe-modified material with 5.00% Fe
(7.88% when expressed as FeS, Table 2). The net incorporation of Fe2+
(3.37% w/w) was more successful as compared with the incorporation
of Mn2+ (1.70% w/w), possibly due to the higher polar affinity of Fe2+
cations for the aluminosilicate layers and then, higher ability of this
metal to replace the exchangeable cations originally present in the
mineral. But in the other hand, the HR-XPS S2p spectrum of AlP20-
MnS100-200 N2 shows a signal in the range of 162–166 eV accom-
panying a couple of stronger ones more easily identified as sulfate
groups, which is not observed in its Fe-modified counterpart (Fig. 6).
Although signals in this region are related in most of the available
databases with elemental sulfur, whereas sulfide usually appear more
displaced to lower energies around 161 eV, it has been recently re-
ported that either central or terminal S in polysulfide groups become
slightly shifted to higher binding energies in the range found for our
Mn-modified material (Fantauzzi et al., 2015). Then, it could be good
evidence of polysulfided Mn-nanoclusters formed in material AlP20-
MnS100-200 N2. Of course two extra questions could be then raised:
why Mn was not detected in this sample? And why no signal was re-
corded for S2p in the HR-spectrum of AlP20-FeS100-200N2? First one
could be due to significantly lower sensitivity of XPS in comparison to
either AAS or XRF determinations (Table 1) which confirmed in-
corporation of Mn in such material; regarding second one, for the same
reason of the first question probably the fraction of interlayered Fe was
not recorded and then S detected in AlP20-FeS100-200N2 corresponded
entirely to external sulfide particles of the metal that were easily
37
Applied Clay Science 157 (2018) 31–40
oxidized to the corresponding sulfate under the heating conditions,
even in the presence of the inert atmosphere, since such external,
contaminant particles of iron present in natural minerals use to be
mixed oxides. Same sulfate species appearing in S2p spectrum of AlP20-
MnS100-200 N2 can be then safely attributed to above-mentioned Fe
impurities present in the starting clay mineral. Finally, it is noteworthy
that a ratio S/Fe only slightly lower to unity (0.86) was found in survey
spectrum of AlP20-FeS100-200N2 (Fig. 5), suggesting that Fe was not
oxidized under conditions of sulfidization and final thermal treatment
used, considering the 1:1 ratio expected for either FeS or FeSO4.
Probably the significantly higher effective ratio S/Fe measured by XRF
in the same solid's precursor AlP20-FeS could be due to formation of
sulfides of other metals also detected in the materials in not negligible
amounts, including Ti, Ca, Zr and Zn. The effective atomic ratio S/
Mn = 4.21 determined by XRF for AlP20MnS100 confirmed that the
MnS-clusters were specifically formed and incorporated in the inter-
layer space of the aluminosilicate.
3.5. Effect of basic neutralization prior to final thermal treatment
As explained above, in order to ensure high yields in formation of
MnS or FeS polynuclear aggregates, a significant excess of H2S was used
along the sulfidization process. However, it has been proposed that high
susceptibility shown by the layered structure of the clay mineral upon
final thermal treatment could be a direct consequence of the presence of
residual protons as interlayer cations in the final sulfidized material
(Galeano et al., 2011). Solids submitted to basic neutralization prior to
thermal treatment for each metal (AlP20-MnS100-n-200 N2 and AlP20-
FeS100-n-200 N2, Tables 1 and 2) clearly showed a structural destabi-
lization with shifting of the 001 reflection towards larger angles in
comparison to solids without such a prior treatment (Fig. 2). This could
be due to the strongly basic pH in the NaOH dispersion (13.3), which
probably caused decomposition of the interlayered metal sulfide na-
noclusters, promoting formation of either iron or manganese hydro-
xides, very stable species at pH values over 7.0 and 4.6 for Mn and Fe,
respectively (Ahenach et al., 1998; Ketteler et al., 2001). Therefore, it
seems that acid/base neutralization directly affected the crystalline
structure of the materials, at the same time increasing back again the
CEC of the materials (Tables 1 and 2); it probably arose to dis-
aggregation of the metal sulfide nanoparticles in the modified mineral
as compared to both materials treated under same conditions, but ex-
cluding previous neutralization. As it can be seen, this effect of layer
charge increasing was much more evident for the Fe-material, conse-
quently with the expected greater susceptibility of the nanoclusters of
this metal sulfide against the pH increase. Furthermore, considering the
amphoteric behavior of the aluminum, alumina pillars could be also
partially dissolved under such a high pH of treatment (pH > 12.5, used
in this preparation) (Ahenach et al., 1998), explaining the partial
L.A. Galeano et al.
Fig. 5. XPS survey spectra of (a) AlP20-MnS100-200 N2 (expected position for signal Mn 2p3/2 is shown) and (b) AlP20-FeS100-200N2 (expected and observed position of signal Fe 2p3/2
is shown). Compositions ignore C, O and N measured contents.
delamination undergone by the layered host under these conditions,
clearly more stressed in the case of the solid AlP20-MnS100-n-200 N2,
probably due to the lower acid neutralizing nature displayed by Mn.
3.6. Effect of pillar-density on the incorporation of the MnS/FeS
polynuclear aggregates in the interlayer space of the clay
A contrast between the diffraction patterns of the Mn sulfide ma-
terials without prior pillaring (BV-MnS100-200 N2) and pillared with
20 or 30 meq Al3+/g clay enabled to find out that materials from pil-
lared precursors exhibited more stable basal reflection in comparison
with the non-sulfidized precursor AlP20-Mn (Fig. 2 and Table 1).
However, the increasing of pillar-density from 20 to 30 meq Al3+/g
clay (AlP30-MnS100-200 N2) drastically reduced the compensated layer
charge (CEC) from 90 to 17%. It might be attributed to very low re-
sidual CEC of AlP30 versus AlP20, leading the amount of exchanged
Mn2+ in AlP30-Mn to be pretty much lower (not shown) and in turn,
the H2S streaming to be more specifically able to attack the alumina
pillars instead than growing interlayered nanoclusters of the transition
metal (Table 1). Conversely, formation of sulfidized iron aggregates was
38
Applied Clay Science 157 (2018) 31–40
more favorable on the clay with pillar-density of 30 meq Al3+/g clay
(AlP30-FeS100-200 N2), probably related to greater structural stability
against the more acid interlayered Fe2+ (Khaorapapong et al., 2009).
Indeed, this material exhibited the highest basal d-value (1.70 nm) as
compared to either pillared solids with 10 or 20 meq Al3+/g clay fol-
lowed by Fe homoionization (AlP10-Fe and AlP20-Fe), as well as
against the homoionized solid without prior pillaring BV-Fe (Table 2),
which besides remained pretty stable even after final thermal treat-
ment.
Therefore, for both transition metals the previous pillaring of the
clay clearly favored the interlayer growth of MnS and FeS nanoclusters
(higher final basal d-values), being this effect fairly more important in
the case of Fe which required a higher pillar-density to keep the ex-
panded pattern of the clay mineral layers. It could be also influenced by
the higher amount of Fe2+ ions than Mn2+ ions incorporated in the Al-
pillared clay, proportionally double amount of metal sulfide na-
noclusters formed in the interlayer. In the case of the Mn-modified set
of materials, the highest stability was attained when using the inter-
mediate density of alumina pillars (20 meq Al3+/g clay, Fig. 7).
L.A. Galeano et al. Applied Clay Science 157 (2018) 31–40

Fig. 6. High-resolution XPS spectra for Fe 2p3/2 and S 2p in AlP20-MnS100-200 N2 and AlP20-FeS100-200N2 solids (Mn was not detected in any case).

4. Conclusions
The effects exerted by main parameters in the rapid preparation of
metal sulfide nanoclusters of Mn or Fe growth in the interlayer space of
either raw or Al-pillared bentonite as a novel strategy in order to con-
trol their final particle size have been established. The condensation of
this type of metal sulfide nanoparticles in the interlayer space of the
clay, without considerably affecting the crystalline properties of the
host inorganic matrix got favored under following set of conditions: (i)
pillar-density of 20 meq Al3+/g clay for MnS and 30 meq Al3+/g clay
for FeS nanoclusters; (ii) sulfidization temperature with H2S at 100 °C;
(iii) inert atmosphere for final thermal treatment in order to delay de-
composition of the nanoclusters towards the metal oxides and (iv)
temperature of final thermal treatment not exceeding 200 °C. Higher
temperatures and/or oxidizing atmospheres used in final heating may
promote formation of interlayered nanoparticles of the corresponding

Fig. 7. General proposed scheme for the in-situ growth of interlayered MnS or FeS nanoclusters in Al-PILC with different optimal loading of Al-pillars for each metal. BV-E: Particle-size
refined bentonite; AleI20 and AleI30 solids denote interlayered solids with different Al-pillar density; AlP20 and AlP30 are the pillared bentonites with 20 or 30 meq Al3+ (Al13)7+/g
clay nominal pillar loading. AlP20-Mn: Cation-exchange incorporated Mn2+ in AlP20; AlP30-Fe: Cation-exchange incorporated Fe2+ in AlP20.

39
L.A. Galeano et al.
metal oxides that could be useful for some specific applications. The
sulfidization of the interlayered transition metals was successfully
achieved by means H2S streaming in times of reaction around 250 times
shorter than Na2S solid state, method widely reported in recent litera-
ture; however, stable final expanded basal reflection showed to strongly
depend on carefully tuning the synthesis conditions for every particular
set of interlayered metal, starting clay and pillar-density. HR-XPS and
DR-UV-vis analyses of the best materials of every metal evidenced
formation of polysulfide groups and blue-shift of the absorption signal
as a consequence of the modification performed on the layered alumi-
nosilicate.
Acknowledgment
Authors thank financial support provided by VIPRI - Universidad de
Nariño (project code 557; A149/2013). XPS measurements performed
by Prof. Carlos Ostos at CATALAD Group - Antioquia University are
gratefully appreciated. The Spanish authors thank the Spanish Ministry
of Economy and Competitiveness (MINECO) and the European Regional
Development Fund (ERDF) for jointly financial support (grant
MAT2013–47811–C2–R and MAT2016-78863-C2-R). AG is also
grateful for financial support from Santander Bank through the
Research Intensification Program.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.clay.2018.02.022.
References
Ahenach, J., Cool, P., Vansant, E.F., 1998. Acid/base treatment of Al-PILC in KCl solution.
Microporous Mesoporous Mater. 26, 185–192.
Antolini, F., Pentimalli, M., Di Luccio, T., Terzi, R., Schioppa, M., Re, M., Mirenghi, L.,
Tapfer, L., 2005. Structural characterization of CdS nanoparticles grown in poly-
styrene matrix by thermolytic synthesis. Mater. Lett. 59, 3181–3187.
Barrault, J., Abdellaoui, M., Bouchoule, C., Majesté, A., Tatibouët, J.M., Louloudi, A.,
Papayannakos, N., Gangas, N.H., 2000. Catalytic wet peroxide oxidation over mixed
(Al–Fe) pillared clays. Appl. Catal. B Environ. 27, L225–L230.
Beverskog, B., Puigdomenech, I., 1996. Revised pourbaix diagrams for iron at 25–300 °C.
Corros. Sci. 38, 2121–2135.
Chavhan, S., Sharma, R.P., 2005. Growth, structural and optical transport properties of
nanocrystal Zn1−xCdS thin films deposited by solution growth technique (SGT) for
photosensor applications. J. Phys. Chem. Solids 66, 1721–1726.
Chen, K., Yan, Q., 1997. CO hydrogenation over zirconia supported iron catalysts pro-
moted with rare earth oxides. Appl. Catal. A Gen. 158, 215–223.
Chin, A.H., Calderon, O.G., Kono, J., 2001. Extreme midinfrared nonlinear optics in
semiconductors. Phys. Rev. Lett. 86, 3292–3295.
Crabtree, R., 2005. The Organometallic Chemistry of Transition Metals. Wiley,
Connecticut.
Fantauzzi, M., Elsener, B., Atzei, D., Rigoldi, A., Rossi, A., 2015. Exploiting XPS for the
identification of sulfides and polysulfides. RSC Adv. 5, 75953–75963.
Galeano, L.A., Gil, A., Vicente, M.A., 2010. Effect of the atomic active metal ratio in al/
Fe-, al/cu- and Al/(Fe–Cu)-intercalating solutions on the physicochemical properties
and catalytic activity of pillared clays in the CWPO of methyl orange. Appl. Catal. B
Environ. 100, 271–281.
Galeano, L.A., Gil, A., Vicente, M.A., 2011. Strategies for immobilization of manganese on
expanded natural clays: catalytic activity in the CWPO of methyl orange. Appl. Catal.
B Environ. 104, 252–260.
Hirai, T., Okubo, H., Komasawa, I., 1999. Size-selective incorporation of CdS nano-
particles into mesoporous silica. J. Phys. Chem. B 103, 4228–4230.
Hirai, T., Saito, T., Komasawa, I., 2001. Stabilization of CdS nanoparticles immobilized on
thiol-modified polystyrene particles by encapsulation with Polythiourethane. J. Phys.
Chem. B 105, 9711–9714.
Iacomi, F., Vasile, A., Pavlidou, E., Paraskevopoulos, K.M., Varoutzis, N., Lioutas, C.,
Polychroniadis, E.K., 2005. Formation of MnS clusters into laumontite zeolite. J.
Optoelectron. Adv. Mater. 7, 859–863.
Iacomi, F., Vasilescu, M., Simon, S., 2006. Studies of MnS cluster formation in laumontite
zeolite. Surf. Sci. 600, 4323–4327.
Infantes-Molina, A., Romero-Perez, A., Eliche-Quesada, D., Merida-Robles, J., Jimenez-
Lopez, A., Rodriguez, E., 2012. Transition Metal Sulfide Catalysts for Petroleum
Upgrading – Hydrodesulfurization Reactions.
Kabilaphat, J., Khaorapapong, N., Saito, K., Ogawa, M., 2015. Preparation of metal sul-
fide mixtures in montmorillonite by solid–solid reactions. Appl. Clay Sci. 115,
248–253.
40
Applied Clay Science 157 (2018) 31–40
Ketteler, G., Weiss, W., Ranke, W., Schlögl, R., 2001. Bulk and surface phases of iron
oxides in an oxygen and water atmosphere at low pressure. Phys. Chem. Chem. Phys.
3, 1114–1122.
Khaorapapong, N., Ontam, A., Ogawa, M., 2008a. Formation of MnS particles in the in-
terlayer space of montmorillonite. Mater. Lett. 62, 3722–3723.
Khaorapapong, N., Ontam, A., Youngme, S., Ogawa, M., 2008b. Solid-state intercalation
and in situ formation of cadmium sulfide in the interlayer space of montmorillonite.
J. Phys. Chem. Solids 69, 1107–1111.
Khaorapapong, N., Ontam, A., Khemprasit, J., Ogawa, M., 2009. Formation of MnS- and
NiS-montmorillonites by solid-solid reactions. Appl. Clay Sci. 43, 238–242.
Khaorapapong, N., Ontam, A., Ogawa, M., 2010. Formation of ZnS and CdS in the in-
terlayer spaces of montmorillonite. Appl. Clay Sci. 50, 19–24.
Khaorapapong, N., Ontam, A., Ogawa, M., 2011. Very slow formation of copper sulfide
and cobalt sulfide nanoparticles in montmorillonite. Appl. Clay Sci. 51, 182–186.
Kloprogge, J.T., 1998. Synthesis of smectites and porous pillared clay catalysts: a review.
J. Porous. Mater. 5, 5–41.
Li, K., Lei, J., Yuan, G., Weerachanchai, P., Wang, J.-Y., Zhao, J., Yang, Y., 2017. Fe-, Ti-,
Zr- and Al-pillared clays for efficient catalytic pyrolysis of mixed plastics. Chem. Eng.
J. 317, 800–809.
Lou, Y., Chen, X., Samia, A.C., Burda, C., 2003. Femtosecond spectroscopic investigation
of the carrier lifetimes in Digenite quantum dots and discrimination of the electron
and hole dynamics via ultrafast interfacial electron transfer. J. Phys. Chem. B 107,
12431–12437.
Lu, J., Qi, P., Peng, Y., Meng, Z., Yang, Z., W, Y., Q, Y., 2001. Metastable MnS crystallites
through Solvothermal synthesis. Chem. Mater. 13, 2169–2172.
Moreno, S., Kou, R.S., Molina, R., Poncelet, G., 1999. Al-, Al,Zr-, and Zr-pillared mon-
tmorillonites and Saponites: preparation, characterization, and catalytic activity in
heptane Hydroconversion. J. Catal. 182, 174–185.
Morgan, B., Lahav, O., 2007. The effect of pH on the kinetics of spontaneous Fe(II) oxi-
dation by O2 in aqueous solution – basic principles and a simple heuristic description.
Chemosphere 68, 2080–2084.
Newman, A.C.D., 1987. The Chemical Constitution of Clays, Chemistry of Clays and Clay
Minerals. Longman-Mineralogical Society, New York.
Öznülüer, T., Erdoǧan, İ., Demir, Ü., 2006. Electrochemically induced atom-by-atom
growth of ZnS thin films: a new approach for ZnS co-deposition. Langmuir 22,
4415–4419.
Pathan, H.M., Lokhande, C.D., 2004. Deposition of metal chalcogenide thin films by
successive ionic layer adsorption and reaction (SILAR) method. Bull. Mater. Sci. 27,
85–111.
Romero-Pérez, A., Infantes-Molina, A., Jiménez-López, A., Jalil, E.R., Sapag, K.,
Rodríguez-Castellón, E., 2012. Al-pillared montmorillonite as a support for catalysts
based on ruthenium sulfide in HDS reactions. Catal. Today 187, 88–96.
Salerno, P., Asenjo, M.B., Mendioroz, S., 2001. Influence of preparation method on
thermal stability and acidity of Al–PILCs. Thermochim. Acta 379, 101–109.
Skromme, B.J., Zhang, Y., Smith, D.J., Sivananthan, S., 1995. Growth and characteriza-
tion of pseudomorphic single crystal zinc blende MnS. Appl. Phys. Lett. 67,
2690–2692.
Steudel, A., Batenburg, L., Fischer, H., Weidler, P., Emmerich, K., 2009. Alteration of non-
swelling clay minerals and magadiite by acid activation. Appl. Clay Sci. 44, 95–104.
Stramel, R.D., Nakamura, T., Thomas, J.K., 1986. Cadmium sulfide on synthetic clay.
Chem. Phys. Lett. 130, 423–425.
Su, B., Choy, K.L., 2000. Synthesis, microstructure and optical properties of ZnS films
formed by electrostatic assisted aerosol jet deposition. J. Mater. Chem. 10, 949–952.
Takahashi, M., Mitsui, K., 1996. Single-site approximation for the s-f model in ferro-
magnetic semiconductors. Phys. Rev. B 54, 11298–11304.
Thorez, J., 1976. Practical Identification of Clay Minerals: A Handbook for Teachers and
Students in Clay Mineralogy, 1st. ed. G. Lelotte, Belgium.
Toranzo, R., Vicente, M.A., Bañares-Muñoz, M.A., Gandı́a, L.M., Gil, A., 1998. Pillaring of
saponite with zirconium oligomers. Microporous Mesoporous Mater. 24, 173–188.
Trindade, T., O'Brien, P., Zhang, X.-m., 1997. Synthesis of CdS and CdSe nanocrystallites
using a novel single-molecule precursors approach. Chem. Mater. 9, 523–530.
Vicente, M.A., Suárez, M., López-González, J.D.D., Banares-Munoz, M.A., 1996.
Characterization, surface area, and porosity analyses of the solids obtained by acid
leaching of a saponite. Langmuir 12, 566–572.
Vicente, M.A., Belver, C., Trujillano, R., Rives, V., Álvarez, A.C., Lambert, J.F., Korili,
S.A., Gandı́a, L.M., Gil, A., 2004. Preparation and characterisation of Mn- and Co-
supported catalysts derived from Al-pillared clays and Mn- and Co-complexes. Appl.
Catal. A Gen. 267, 47–58.
Viswanatha, R., Battaglia, D.M., Curtis, M.E., Mishima, T.D., Johnson, M.B., Peng, X.,
2008. Shape control of doped semiconductor nanocrystals (d-dots). Nano Res. 1,
138–144.
Wang, D.S., Zheng, W., Hao, C.H., Peng, Q., Li, Y.D., 2009. A synthetic method for
transition-metal chalcogenide nanocrystals. Chemistry 15, 1870–1875.
Wood, B.J., 1979. Diffuse reflectance spectra and optical properties of some sulphides and
related minerals. Mineral. Mag. 43, 509–518.
Xia, Y., Yang, P., Sun, Y., Wu, Y., Mayers, B., Gates, B., Yin, Y., Kim, F., Yan, H., 2003.
One-dimensional nanostructures: synthesis, characterization and applications. Adv.
Mater. 15, 353–389.
Yu, J., Yang, Q.-X., 2010. Magnetization improvement of Fe-pillared clay with application
of polyetheramine. Appl. Clay Sci. 48, 185–190.
Zhang, Z., Dai, S., Fan, X., Blom, D.A., Pennycook, S.J., Wei, Y., 2001. Controlled
synthesis of CdS nanoparticles inside ordered mesoporous silica using ion-exchange
reaction. J. Phys. Chem. B 105, 6755–6758.