Nitroarene and dye reduction with 2:1 Co/Al layered double hydroxide catalysts – Is gold still necessary?

Article abstract of DOI:10.1016/j.ica.2021.120336

A 2:1 Co/Al layered double hydroxide (LDH) material was synthesized with Cl^– anions intercalated and then anion exchanged with deprotonated methionine. This allowed further immobilization of Au nanoparticles. The catalytic performance of the set of three LDH materials for comparative benchmarking was assessed in the aqueous reduction of nitroaromatic compounds and of rhodamine dyes in the presence of NaBH₄. All catalysts were found to be very active for the nitro-to-amine reduction. The same was observed for the dyes, where reduction rhodamine 6G was faster than that of rhodamine B, for all catalysts. While the Au-containing catalyst was apparently the one showing the best catalytic activity the LDH with Cl^– was found to follow it closely in terms of catalytic performance. In addition, while the LDH catalyst with Au was prone to deactivation by amine products in recycling experiments, the LDH catalyst with Cl^– was almost insensitive to that, which is a large advantage. These results showed that using a catalyst based on first-row metals for efficient processes is possible and addresses current sustainable and environmental concerns.

Full text of DOI:10.1016/j.ica.2021.120336

Inorganica Chimica Acta 521 (2021) 120336
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Nitroarene and dye reduction with 2:1 Co/Al layered double hydroxide
catalysts – Is gold still necessary?
a
a
a
b
,
,e
Sonia R. Leandro , Ines J. Marques , Ruben S. Torres , Tiago A. Fernandes c, Pedro D. Vazd
,
´
ˆ
,
*
Carla D. Nunesa
a
ˆ
Centro de Química Estrutural, Faculdade de Ciencias da Universidade de Lisboa, 1749-016 Lisboa, Portugal
b
c
´
Centro de Química Estrutural, Instituto Superior Tecnico, Universidade de Lisboa, 1049-001 Lisboa, Portugal
ˆ
Centro de Química e Bioquímica, Faculdade de Ciencias da Universidade de Lisboa, 1749-016 Lisboa, Portugal
d CICECO – Aveiro Institute of Materials, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
e Champalimaud Foundation, Champalimaud Centre for the Unknown, 1400-038 Lisboa, Portugal
A R T I C L E I N F O
A B S T R A C T
A 2:1 Co/Al layered double hydroxide (LDH) material was synthesized with Cl– anions intercalated and then
anion exchanged with deprotonated methionine. This allowed further immobilization of Au nanoparticles. The
catalytic performance of the set of three LDH materials for comparative benchmarking was assessed in the
aqueous reduction of nitroaromatic compounds and of rhodamine dyes in the presence of NaBH4. All catalysts
were found to be very active for the nitro-to-amine reduction. The same was observed for the dyes, where
reduction rhodamine 6G was faster than that of rhodamine B, for all catalysts.
Keywords:
Gold nanoparticles
Hydrotalcite
Dyes
Nitroaromatic compounds
Reduction catalysis
Sustainability
While the Au-containing catalyst was apparently the one showing the best catalytic activity the LDH with Cl–
was found to follow it closely in terms of catalytic performance. In addition, while the LDH catalyst with Au was
prone to deactivation by amine products in recycling experiments, the LDH catalyst with Cl– was almost
insensitive to that, which is a large advantage. These results showed that using a catalyst based on first-row
metals for efficient processes is possible and addresses current sustainable and environmental concerns.
1. Introduction
development is compulsory [6].
Layered double hydroxides (LDHs) are among the inorganic mate-
Developing sustainable and efficient catalytic processes is a major
challenge that is still taking its early steps. This quest is a mandatory
mission due to environmental concerns focused on the development of
cheap catalysts (including green reaction media) while aiming at a lower
consumption of natural resources [1–3].
rials that have experienced a high development in recent years, owing to
their applications in many fields, namely heterogeneous catalysis [7].
These materials are known for their ion exchange capacity, making
possible the introduction of adequately functionalized guests [7,8]. The
importance of these materials is based on their ability to retain chemical
species with electrical charges compatible to those of the layers. LDHs
are known for their extreme versatile customization, arising not only
from their composition but also from their anion-exchange capacity,
where a huge range of possibilities, both organic and inorganic, is
virtually infinite. Anion exchange can be achieved by either (i) anion
exchange, (ii) co-precipitation and (iii) reconstruction. Independently of
the route chosen, contamination from CO2 must be prevented, due to the
ready and persistent incorporation of carbonate anions within the
interlayer of LDHs intercalated with anions other than carbonate [7].
Consequently, it is mandatory to use decarbonated and deionized water
A way to attain this goal is to mimic Nature’s processes and so far, it
has never stopped to be an inspiration. Nature designs its activity and
provides an incredible range of metal-based materials that are readily
available and usually cheap. These are commonly metals from the first-
row transition series or others such a Mg, Ca, Zn or Al [4,5]. Exploring
such field will result in reducing the use of precious metals, which so far
are still unrivaled in catalytic performance, but present huge drawbacks
from economic and environmental points of view.
Thus, the number of catalysts and catalytic systems based on the
cheaper metals is still limited, fostering new frontiers in their
ˆ
* Corresponding author at: Centro de Química Estrutural, Departmento de Química e Bioquímica, Faculdade de Ciencias da Universidade de Lisboa, Campo Grande,
Ed. C8, 1749-016 Lisboa, Portugal.
E-mail address: cmnunes@fc.ul.pt (C.D. Nunes).
https://doi.org/10.1016/j.ica.2021.120336
Received 26 October 2020; Received in revised form 15 February 2021; Accepted 2 March 2021
Available online 9 March 2021
0020-1693/© 2021 Elsevier B.V. All rights reserved.

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
under inert atmosphere.
2.2. Catalytic reduction of nitroaromatic compounds
Concerning catalytic applications, the use of LDHs can still go much
further in this field meeting the sustainability requirements. The
versatility of these materials is not restricted to the ion exchange ca-
pacity. In fact, the use of LDHs as catalysts is not restricted to their
composition or bulk form (powder) either [9–13]. LDH catalysts were
engineered as colloidal nanosheets and more recently, as nanoparticles,
as monoliths with hierarchical porosity, or as gelled nanoparticles with
high concentration, which defined new frontiers for applications of such
materials [11,12,15–24].
The reduction of nitroaromatic compounds by the HTCo/Al-Cl, HTCo/
Al-met and HTCo/Al-met-Au catalysts was conducted in aqueous solution
in the presence of NaBH4. We chose to test 4-nitroaminophenol, 2,6-
dinitrophenol and 4-nitroaniline as substrates. The catalytic reactions
were carried out using a quartz cuvette with an optical pathlength of 1
cm. Typically, the catalytic reactions were conducted at room temper-
ature (298–303 K) by mixing 200 μL of an aqueous solution of the
relevant nitroaromatic substrate (1 mmol/L) with 1.0 mg of the solid
catalyst (which dispersed neatly) in a quartz cuvette, and then adding
2.5 mL deionized water to the quartz cuvette and sonicating to mix them
Increasing research in recent years led to new non-noble metal cat-
alysts capable of carrying out reactions at lower temperatures and
pressures with higher selectivity towards the desired products [25].
Cobalt materials have contributed to this quest as demonstrated by the
number of published works describing their use in reduction reactions.
Several reports have demonstrated the efficient use of Co in the reduc-
tion of nitroaromatic compounds to their amine counterparts. For
instance, the chemoselective reduction of a set of nitro compounds was
reported by Rai and coworkers using a Ni and Co-based nanocatalyst
[26]. The reduction reactions were accomplished strictly in aqueous
phase using hydrazine as reducing agent and under very mild conditions
(room temperature). Although some of the reactions were not very fast,
they reached outstanding performance, especially in selectivity. In other
reported works, LDH materials have been used as the catalyst or catalyst
support towards reduction of 4-nitrophenol [27,28]. In both works the
results were excellent with the catalysts being robust enough to be used
across several catalytic cycles. More recently, the group of Zhang has
reported a similar system where they replaced Mg by Ni with improved
results relatively to their previously reported system [29].
evenly (ca. 10 s). Subsequently, 200
μL of freshly prepared NaBH4
aqueous solution (0.1 mol/L) were added to the above reaction mixture
to start the reduction, being monitored using UV–Vis spectroscopy over
a scanning range of 250–500 nm. The kinetics of the reactions were
followed for 60 min and measured using the absorbance maxima of each
tested substrate (see Fig. S3 in SI material for the UV–Vis spectra of neat
compounds) by recording the mixture absorbance at regular time in-
tervals with a UV–Vis spectrophotometer from Shimadzu (UV-2450)
equipped with a Peltier cell for temperature control. Initial reaction
rates were determined by fitting the absorbance vs. time data to Eq. (1):
(
)
(
)
C
C0
C
C0
ln
= ꢀ kt + ln
(1)
t
∞
(
)
(
)
C
C
where
and
are the relative quantities of the substrate at
C0
C0
t
∞
time t and at the end of reaction, respectively and k representing the
calculated rate.
The reduction of dye compounds catalyzed by non-noble metals has
also been reported. For instance, Kundu and coworkers reported the
synthesis of CoO nanowires with excellent activity [30]. Our group has
also reported the efficient use of nanoclustered Co/Al LDH materials for
the reduction of rhodamine B dye and benchmarked the results against
the bulk LDH equivalent [14]. The results showed that while the stan-
dard bulk Co/Al LDH reached 48% substrate conversion at 15 min re-
action time, the LDH nanocluster reached 98%. The results were even
more dramatic only after 1 min of reaction with the conversion reaching
3.4% and 91% for the bulk and nanocluster catalysts, respectively. The
measured reaction rate for the nanocluster LDH was 47-fold higher than
that of the bulk, while its surface being only 14 times higher [14].
In this work we report a study on the versatility of Co/Al LDH-based
catalysts and how they compared with the engineered version holding
Au nanoparticles for the aqueous reduction of nitroaromatic compounds
and of rhodamine dyes. As will be discussed throughout this work we
found that although the catalyst with Au was the best performing one,
one of the Co/Al LDH catalysts followed it closely. This evidenced that
despite the presence of Au is beneficial it can be avoided with very good
results for catalytic applications in the reduction of the above-
mentioned substrates.
2.3. Catalytic reduction of rhodamine dyes
The protocol followed for the reduction of dyes was similar to that
described above for the nitroaromatic compounds. Rhodamine B and 6G
were chosen as substrates using 20 ppm solutions. In a typical catalytic
experiment, 10.0 mg of the solid catalyst were added to 2.5 mL of the 20-
ppm aqueous solution of the chosen dye and sonicated for homogeneous
dispersion (ca. 10 s). After that, 200
μL of freshly prepared NaBH4
aqueous solution (0.1 mol/L) were added to the above mixture deter-
mining the start of the reduction. The reaction was monitored for 60 min
using UV–Vis spectroscopy and scanning across a range between 400
and 700 nm at regular time intervals. The kinetics of the reactions were
measured using the absorbance maxima of each substrate tested (see
Fig. S4 in SI material for the UV–Vis spectra of neat compounds). The
reaction rates were calculated as described above using Eq. (1).
3. Results and discussion
3.1. Synthesis of LDH materials
2. Experimental
In this work we have used a combination of co-precipitation and
anion exchange methods to prepare clay-based hybrid materials with L-
methionine (met) as evidenced in Scheme 1 [9]. The first step comprised
the synthesis of the Co/Al LDH material (denoted as HTCo/Al) synthe-
sized by the co-precipitation method following a reported procedure, as
outlined in Scheme 1.
2.1. General
All reagents were purchased from Aldrich and used as received. In
this work we have used deprotonated L-methionine as the intercalating
ligand and will be denoted as met. HTCo/Al-Cl, and HTCo/Al-met mate-
rials were synthesized as reported recently by us [9,15]. A biomimetic
methodology was used to synthesize the Au nanoparticles (AuNP) as
already reported by us as well [9,15]. Detailed experimental and
computational procedures alongside XRD powder patterns (Fig. S1) and
FTIR (Fig. S2) for precursor materials can be found in the SI material.
The synthesis produces the carbonate-intercalated material (HTCo/
Al-CO3), which was then ion-exchanged with chloride anions to yield
HTCo/Al-Cl [7]. The latter was easier to undergo further ion exchange
(due to lower charge density of the Cl– anions). Elemental analysis
allowed estimation of the chemical composition Based on this, the
estimated
chemical
formula
was
[Co0.67Al0.33(OH)2]
[(met)0.326⋅0.99H2O] with an experimental Co/Al molar ratio compo-
sition of 2.07. The deprotonated met amino acid was introduced by ion
exchanging the Cl– anions [9], yielding the HTCo/Al-met material. of the
2

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
Scheme 1. Synthetic procedure for preparation of clay materials.
LDH material.
Further reaction of HTCo/Al-met with the gold nanoparticles (AuNP
)
yielded the new HTCo/Al-met-Au hybrid material. The Au loading was
found to be 0.116 wt-% corresponding to 5.9 × 10-3 mmolAu⋅gꢀ 1 (or 1.2
mgAu⋅gꢀ 1) for this material.
Electronic spectroscopy (using both absorption and diffuse reflec-
tance techniques) was used to assess formation of the Au-containing
hybrid material. HTCo/Al-met-Au, whose measured spectra are shown
in Fig. 1.
The measured spectrum showed that the HTCo/Al-met-Au material
displayed the surface plasmon resonance band at comparable wave-
lengths to those observed for the as-synthesized AuNP, indicating the
successful formation of the HTCo/Al-met-Au material [9,15]. Moreover,
the measured absorbance maxima of the hybrid material also changed,
which probed the interaction between the AuNP and the host HTCo/Al
-
met material [9,15]. The surface plasmon resonance band shifted to 565
nm and the methioninate bands blue-shifted from 631 and 671 nm (in
neat methionine) to 611 and 665 nm, respectively, confirming specific
interactions between AuNP and methionine in the HTCo/Al-met-Au
material.
Fig. 2. XRD powder patterns of HTCo/Al-met and HTCo/Al-met-Au. The peak at.
2θ = 5.74◦ was assigned to the 003 diffraction peak corresponding to the
interlayer space containing the methionine guests. Diffraction peaks arising
from AuNP in HT-met-Au are highlighted with an arrow at 2θ = 38.2◦, 44.4◦
and 64.6◦. Diffraction patterns have been offset vertically for clarity.
Fig. 2 shows the powder XRD patterns of the HTCo/Al-met and HTCo/
Al-met-Au materials. Both materials displayed a series of well-developed
Bragg reflections in the range 4◦ < 2θ < 70◦, which could be indexed to
the hexagonal layered LDH structure with rhombohedral symmetry
[31]. No peaks from impurities were detected, indicating the high purity
of the products.
structure (ca. 4.8 Å) plus the thickness of the interlayer region [32]. The
calculated d parameter was calculated from the second 2θ value corre-
sponding to the 0 0 3 plane and found to be 15.30 and 15.20 Å for both
the HTCo/Al-met and HTCo/Al-met-Au materials, respectively. Assuming
HT layers (brucite layers) to be 4.8 Å thick [33,34], the gallery height for
HTCo/Al-met was estimated to be 10.5 Å, indicating that the guest met
molecules were most probably in a tilted position.
A set of intense peaks due to basal 0 0 l reflections at low 2θ angles
allowed us to estimate directly the normal basal spacing to the 0 0 1
plane, which is equal to the thickness of one layer of the brucite
In addition, other relevant lattice parameters to describe the unit cell
were calculated from the 0 0 3 and 1 1 0 diffraction peaks for both
materials, as shown in Table 1 [35,36].
The shortest distance between two cations in one hydroxide layer, a
= 2d110, is also an important lattice parameter, with both materials
displaying the same value (a = 3.04 Å). From the d003 value, the unit
cells (c = 3d003) could be calculated showing values of 45.90 and 45.60
Table 1
Structural parameters for HTCo/Al-met and HTCo/Al-met-Au materials deter-
mined by XRD in solid state.
Material
Peak positiona (◦)
Structural parameters (Å)
003
110
db003
db110
ac
cd
υe
HTCo/Al-met
5.80
5.80
60.85
60.90
15.30
15.20
1.52
1.52
3.04
3.04
45.90
45.60
119.04
133.89
HTCo/Al-met-Au
aMeasured on the XRD powder pattern; d-spacing, calculated with Bragg
equation, nλ = 2dsinθB, where n is an integer (in general 1), λ is the wavelength
of the radiation (Cu Kα = 1.5405 Å), d is the lattice spacing and θB is the Bragg
b
Fig. 1. Electronic spectra of gold nanoparticles (AuNP), methionine (met), and
of the hybrid clay material HTCo/Al-met-Au. The spectrum of the hybrid ma-
terials was obtained as diffuse reflectance (DRUV) while the remaining were
obtained from aqueous solutions in transmission mode. Spectra have been offset
vertically for clarity.
angle; c Shortest distance between two cations in the layer and a = 2d110
;
d Unit
e
cell width, c = 3d003
;
Average particle thickness, calculated with Scherrer
equation,
υ = (Kλ)/(βcosθB), where K is the shape factor (0.89 was used in the
calculation), λ is the wavelength of the radiation (Cu Kα = 1.54 Å) and β is the
110 peak full width at half maximum in radian.
3

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
Å, respectively, for HTCo/Al-met and HTCo/Al-met-Au materials. More-
material allowed identification of a set of bands assigned to the
νC–H,
over, the average thickness of the particles (
ν
) along the c axis was found
ν
C–O and C=O modes of the methionine amino acid moiety (Fig. 3b).
ν
to be 119.04 and 133.89 Å for HTCo/Al-met and HTCo/Al-met-Au,
respectively. Given that a brucite layer is 4.8 Å thick, this suggested that
the particles were constituted by ca. 24–28 brucite-like hydroxide
layers. These results confirmed that the adopted synthesis procedure of
HTCo/Al-met and HTCo/Al-met-Au materials did not alter the primitive
structure of the layered double hydroxide.
Unfortunately, bands corresponding to νN-H vibrational modes, ex-
pected in the 3300–3500 cmꢀ 1 range could not be observed due to
overlapping. By the same token, νC–N modes, expected between 1020
and 1220 cmꢀ 1 were not observed as well due to overlap of the
νC–O
modes. The characteristic band of lamella hydroxyl groups bound to the
the metals,
band.
ν
M–O modes, appears near 600 cmꢀ 1 in the form of a broad
Peaks due to the presence of the AuNP structures in the HTCo/Al-met-
Au material were identified at 2θ = 38.2◦, 44.4◦ and 64.6◦ (Fig. 2, arrow
markers).
Thermogravimetric analysis (TGA) also confirmed the intercalation
of met in the interlayer spacing of the HT LDH material. According to
Fig. 4, severe decomposition of met was observed at temperatures in the
500–650 K range. Afterwards, decomposition continues at a slower pace
till an overall mass loss of ca. 96.8% at 1073 K.
Diffuse reflectance infrared spectroscopy (DRIFT) was used to char-
acterize all the materials, as shown in Fig. 3.
The FTIR spectra of HTCo/Al-Cl, HTCo/Al-met and HTCo/Al-met-Au
(Fig. 3a) showed similar profiles. All materials showed a band centered
In the case of HTCo/Al-met its decomposition profile evidenced a mass
loss of 12% till ca. 493 K, which was probably due to both adsorbed and
intercalated water. The material further decomposed in a two-step way
till 1073 K, yielding a total weight loss of 25%. One could assume that
met was the sole contributor for the observed weight loss. However, one
should still bear in mind that at higher temperature some dehydrox-
ilation was likely to occur and therefore the observed weight loss may
arise from decomposition of both moieties. The decomposition profile
should be commented as well; in this case the profile was different from
that observed for neat met, which could be a proof that met was indeed
intercalated within the interlayer spacing of the HT host and therefore
protected against thermal decomposition due to specific host–guest in-
teractions [9,15]. The thermal decomposition profile observed for the
HTCo/Al-met-Au material was found to superimpose that of the precursor
HTCo/Al-met material. This evidenced that the introduction of the AuNP
moieties did not induce further changes to the HTCo/Al-met material.
at ca. 3440 cmꢀ 1 being assigned to the
νO-H mode of hydroxyl groups,
while the presence of a shoulder at ca. 3150 cmꢀ 1 suggested that water
molecules were interacting with other species by hydrogen bonding.
Remaining bands below 800 cmꢀ 1 may be associated with metal–oxygen
bonds in the structure of the clay [15]. The spectrum of the HTCo/Al-met
3.2. Catalytic studies on reduction of nitro- to amino-aromatic
compounds
It is known that Au and/or Co catalysts can be used as efficient
catalysts in reduction reactions [21–24,30]. In this way we tested the set
of three clays reported here in the reduction of nitro-aromatic com-
pounds to their amine counterparts. To accomplish this the chosen
substrates were p-nitrobenzylalcohol, p-nitroaniline and 2,6-
dinitrophenol.
The set of three nitro-aromatic compounds was chosen as to contain
model substrates to test the efficiency of the LDH materials in the cat-
alytic reduction of those compounds to their corresponding amine
products, in the presence of NaBH4. The set of substrates also allowed to
Fig. 3. DRIFT spectra of HTCo/Al-met and HTCo/Al-met-Au materials. In (a) the
full spectra are shown. The νCH bands of met in the spectra of HTCo/Al-met and
HTCo/Al-met-Au materials were marked for clearer identification. Spectra in (b)
display the fingerprint region (modes below 2000 cmꢀ 1) of the hybrid mate-
rials; the spectrum of met was included for comparison. Spectra in both figures
have been offset vertically for clarity.
Fig. 4. TGA profiles obtained under N2 atmosphere at 10 K/min of met, HTCo/
Al-met and HTCo/Al-met-Au.
4

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
screen the catalyst for its chemoselectivity and scope of the reaction. The
set of LDH materials allowed benchmarking of their catalytic activity to
check whether the presence of the AuNP moieties could be avoided.
Furthermore, control reactions with only NaBH4 in the absence of any
catalyst showed almost no effect (after ca. 2 h) on any substrate con-
centration, confirming that the reduction reaction requires a catalyst.
The results were found to be promising. The reaction kinetics, with
reactions being carried out for 60 min, matched [28] or outperformed
[26] other systems found in the literature. The use of HTCo/Al-met-Au as
catalyst yielded the desired amine products under green chemistry con-
ditions: room temperature (298 K) and water as the reaction medium.
The use of NaBH4 is also environmentally friendly as its decomposition
yields NaBO4 [24].
Fig. 5 shows the UV–Vis kinetics with the substrate concentration of
the different nitro-aromatic compounds during the catalytic reaction. It
is worth noting that for all substrates there was a dramatic concentration
decrease as the reduction reaction progressed in the presence of the Au
catalyst.
However, the Au-containing catalyst was followed closely by the
HTCo/Al-Cl counterpart, whereas the HTCo/Al-met catalyst was the worst
performing one in terms of substrate conversion. In fact, a plausible
reason for this may arise from the fact that the met moiety in the HTCo/
Al-met catalyst was able to establish H-bond interactions with the sub-
strates and delay the reaction kinetics. This situation was minimized or
mitigated in the HTCo/Al-met-Au catalyst because in this case met was
interacting with AuNP and could not establish further interactions. In the
case of the HTCo/Al-Cl catalyst the H-bonding situation was not possible
and therefore the observed catalytic activity was not biased.
The pseudo first-order kinetic model was applied to evaluate the rate
constants for nitro-aromatic reduction, allowing to confirm the above
statements, because the concentration of NaBH4 was much higher (100-
fold) than those of the substrates and could be considered constant
throughout the reaction. The concentration of the substrates at time t
was denoted as Ct, while its initial concentration at t = 0 was defined as
C0. The Ct/C0 ratio was measured from the relative absorbance intensity
(At/A0). The rate constants for nitroarene reduction were calculated to
be 0.129 minꢀ 1 for 2,6-dinitrophenol, 0.122 minꢀ 1 for p-nitro-
benzylalcohol and 0.040 minꢀ 1 for p-nitroaniline using the HTCo/Al-met-
Au catalyst (Fig. 5, Table 2). The performance of the catalyst in a
recycling experiment with 2,6-dinitrophenol displayed an initial reac-
tion rate of 0.046 minꢀ 1. This value was around a third of the reaction
rate calculated for the fresh catalyst, which indicated that the catalyst
suffered deactivation. On the other hand, the HTCo/Al-met catalyst
despite being the one displaying the lowest substrate conversion pre-
sented the lowest initial rate constants overall as well. Surprisingly, the
HTCo/Al-Cl one showed rate constant values close to those calculated for
the HTCo/Al-met-Au as catalyst (Table 2). Despite that the HTCo/Al-Cl
catalyst did not reach such a good substrate conversion as displayed by
the latter.
The reduction of 2,6-dinitrophenol did not occur through formation
of the 2-amino-6-nitrophenol intermediate, as previously observed for
the gold-catalyzed reduction of 2,4-dinitrophenol [22]. In that work by
Lin and Doong, reduction of 2,4-dinitrophenol was found to occur by
sequential reduction of each nitro group, i.e., with 2-amino-4-nitrophe-
nol as intermediary compound, which was evidenced by the appearance
of a new band at early times of the reaction, which vanished with
increasing time [22]. In our study we did not observe this behavior
according to the measured spectra (not shown). The reason behind this
may be that 2,6-dinitrophenol may undergo parallel reduction of both
nitro groups to form directly 2,6-diaminophenol in the presence of the
Au catalyst and NaBH4. This may arise from the favorable geometrical
arrangement of its nitro groups, which can bind at the same time to the
catalyst’s surface.
Fig. 5. Time-dependent UV–Vis changes of a) p-nitroaniline (375 nm), b) p-
nitrobenzylalcohol (272 nm) and c) 2,6-dinitrophenol (342 nm). The inset in c)
shows the kinetics for the recycling experiment using the HTCo/Al-met-Au
catalyst. Fig. S5 shows the linearized plots of the reaction kinetics data.
The mechanism for this Au-catalyzed process should follow that re-
ported in the literature for a related system [23]. The mechanism
assumed that the first step comprises the adsorption of BH–4 ions at the
5

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
Table 2
Catalytic aqueous reduction of nitroaromatic compounds using the set of HT
materials as catalysts in the presence of NaBH4.
Entry
Substrate
Catalysta
Conversionb
(%)
kc (×103
minꢀ 1
)
1
2
HTCo/Al-Cl
69
50
14
HTCo/Al
met
-
< 5
3
HTCo/Al
met-Au
-
100
40
4
5
HTCo/Al-Cl
60
21
64
19
HTCo/Al
-
met
6
HTCo/Al
met-Au
-
100
122
7
8
HTCo/Al-Cl
71
43
78
46
HTCo/Al
-
-
met
9
HTCo/Al
met-Au
100
129/46d
aAll reactions were carried out in the presence of 1.0 mg of catalyst; b Conversion
Fig. 6. DFT optimized structures of Au(0) complexes with aniline (top) and
c
calculated after 60 min. reaction time; Rates calculated according to Eq. (1)
borohydride anion (bottom).
d
(Experimental section); Rate values for the regular (1st) and recycling (2nd
)
experiments, respectively.
3.3. Catalytic studies on reduction of dyes
surface of the Au nanoparticles yielding Au-hydride complex. The sec-
ond step comprises adsorption of the nitroaromatic substrate at the
surface of the Au nanoparticles as well. Both steps are reversible thus
competing with NH2 groups (when present) confirming the observation
of lower reaction rates in p-nitroaniline, as already discussed before
[22,23].
To evaluate the scope of action of this set of three catalysts, we also
tested them in the reduction of dyes. For this study rhodamine B and
rhodamine 6G were chosen as the candidates. Overall, the results fol-
lowed the same trend as that discussed above for the catalytic reduction
of the nitro aromatic compounds. According to Fig. 7 the reaction pro-
gressed spontaneously as evidenced by the control experiments on both
substrates.
After the adsorption, hydrogen transfer takes place sequentially,
forming nitroso and hydroxylamine (intermediaries) to finally yield the
aniline product. According to literature data [37], the reaction is fast till
formation of the hydroxylamine intermediary, but the final step
(involving product desorption) to originate the aniline is the rate
determining step. The aniline product then desorbs regenerating the
catalyst. We believe that a similar mechanism could be used to explain
the catalytic activity of the remaining HTCo/Al-Cl and HTCo/Al-met cat-
alysts, by hypothesizing that BH–4 ions adsorb at the surface of the mixed-
metal oxide sheets, especially at Co sites. In this case the reducing agent
was probable to reduce Co(III) loci to Co(II) or Co(0), which in turn will
make the catalyst active for the reduction reaction.
However, in the presence of any of the catalysts the reactions were
much faster. The observed catalytic performance followed the HTCo/Al
-
met < HTCo/Al-Cl < HTCo/Al-met-Au trend. These results were valid for
both the substrate conversion as well as for the initial reaction rates as
shown in Table 3.
As can be seen, the HTCo/Al-met counterpart was the worst per-
forming. An explanation for this may be the same discussed earlier in
this work, where the existence of intermolecular H-bonds between the
met moiety in the HTCo/Al-met catalyst and the substrates hindered the
reaction kinetics.
Comparing the reaction kinetics between the dye substrates for a
given catalyst it was observed that reduction of the rhodamine 6G
substrate was always faster than the kinetics for reduction of rhodamine
B. This may arise from the slight structural differences presented by both
substrates (Table 3). Rhodamine 6G has less bulky amine groups (sec-
ondary amine groups) than rhodamine B (tertiary amine groups), which
could be more prone for interaction with the reactive sites. This was
observed across all catalysts studied in this work.
It is known that Au catalysts are environment-sensitive to reaction
conditions, such as solvent or functional groups – e.g. amine or thiol
groups – from other molecules in solution [20]. In this case, we observed
that reduction of 2,6-dinitrophenol and p-nitrobenzylalcohol displayed
higher rate constants than p-nitroaniline. This was also evidenced in the
recycling experiment of HTCo/Al-met-Au in the reduction of 2,6-dinitro-
phenol (Table 2). According to the literature data, the dramatic decrease
in rate constants for p-nitrophenol reduction is attributed to the presence
of amine products, which then bind to the surface of Au nanoparticles
with concomitant deactivation.
Most interestingly, the results from the recycling experiments were
more exciting. As observed for the reduction of nitroaromatic com-
pounds, the HTCo/Al-met-Au catalyst experienced a dramatic decrease of
the reaction rate due to deactivation by the presence of the amine
moieties. In this case, for rhodamine B the reaction rate was found to
decrease from 0.076 to 0.040 minꢀ 1 (47% decrease; Table 3, entry 4),
while for rhodamine 6G the rate dropped from 0.168 to 0.087 minꢀ 1
(48% decrease; Table 3, entry 9).
To confirm these assumptions, we carried out simple DFT calcula-
tions to explore both the amine (using aniline as model compound) and
borohydride coordination to gold. The optimized geometries for the
resulting gold complexes are shown in Fig. 6.
The formation of both species was predicted to be favorable with
their enthalpies being –6.0 and –8.4 kcal molꢀ 1 for the aniline and
borohydride complexes, respectively. This result showed that the
borohydride complex is the most stable and, therefore, there should be
no problem associated with the recycling experiments concerning ki-
netics. However, the energy difference was found to be small enough to
allow both species to coexist at the reaction temperature (ca. 300 K) and
compete for Au coordination. This may certainly account for the
observed slower kinetics effect when starting with p-nitroaniline and in
the recycling experiment of 2,6-dinitrophenol.
However, the HTCo/Al-Cl counterpart did not evidence major
changes in the reaction rate, showing that this catalyst is not prone to
poisoning by the amine moieties, which was a pleasant surprise. With
this catalyst, the reaction rate for rhodamine B decreased from 0.069 to
0.053 minꢀ 1 (23% decrease; Table 3, entry 1), while for rhodamine 6G
the rate changed from 0.182 to 0.116 minꢀ 1 (36% decrease; Table 3,
entry 5). Overall, these results evidenced that, although achieving
similar substrate conversion, the HTCo/Al-Cl catalyst was the best per-
forming one given that it experienced less deactivation in the recycling
6

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
Fig. 7. Time-dependent UV–Vis changes of rhodamine B (554 nm, a and c) and rhodamine 6G (526 nm, b and d) in the presence of the HTCo/Al-Cl, HTCo/Al-met and
HTCo/Al-met-Au catalysts. The plots in c) and d) show the kinetics for two consecutive catalytic reduction reactions of rhodamine B (c) and rhodamine 6G (d) in the
presence of HTCo/Al-Cl evidencing its little sensitivity to deactivation. Fig. S7 shows the linearized plots of the reaction kinetics data.
Table 3
Catalytic aqueous reduction of rhodamine B and 6G dyes using the set of HT materials as catalysts in the presence of NaBH4.
Entry
Substrate
Catalysta
Conversionb (%)
kc (×103 minꢀ 1
)
1
2
3
4
HTCo/Al-Cl
100
100d
86
69/53e
2100d
64
nanoHTCo/Al-Cl
HTCo/Al-met
HTCo/Al-met-Au
100
76/40e
5
6
7
HTCo/Al-Cl
100
88
182/116e
76
HTCo/Al-met
HTCo/Al-met-Au
100
168/87e
b
c
aAll reactions were carried out in the presence of 1.0 mg of catalyst; Conversion calculated after 60 min. reaction time; Rates calculated according to Eq. (1)
(Experimental section); d results from ref. [14] after 15 min; e Rate values for the regular (1st) and recycling (2nd) experiments, respectively.
7

S.R. Leandro et al.
Inorganica Chimica Acta 521 (2021) 120336
experiment.
(projects UIDB/00100/2020 and UIDP/00100/2020). TAF thanks CQB
and FCT for a fellowship (PEst-OE/MULTI/00612/2013 and CEECIND/
02725/2018).
Compared to literature data for a related nanosized catalyst (nano-
HTCo/Al) recently reported by us, the one reported here (bulk form) is
much less active concerning the reaction kinetics (by ca. 30-fold,
Table 3, entries 1 and 2) [14]. However, the substrate conversion was
also found to be the same, reaching reaction completeness with the
difference that the reaction times were 15 and 60 min., respectively for
the nanoHTCo/Al and HTCo/Al-Cl catalysts.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ica.2021.120336.
Finally, we have also studied eventual structural changes suffered by
the catalysts after catalysis. In this way, we measured the XRD powder
pattern for a HTCo/Al-met-Au catalyst sample recovered after two
consecutive Rh6G reduction reactions and found that the overall profile
was kept (Fig. S8 in SI material), demonstrating that these catalysts were
robust. It should be worth mentioning that Au loading after these two
reactions decreased by 19%.
References
[1] D. Astruc, Nanoparticles and Catalysis, Wiley, Weinheim, 2008.
[2] R.J. White, R. Luque, V.L. Budarin, J.H. Clark, D.J. Macquarrie, Chem. Soc. Rev. 38
(2009) 481–494.
[3] L. Mai, F. Yang, Y. Zhao, X. Xu, L. Xu, B. Hu, Y. Luo, H. Liu, Mater. Today 14 (2011)
346–353.
[4] [4] European Cluster on Catalysis, Science and Technology Roadmap on Catalysis
for Europe, http://www.catalysiscluster.eu/wp/wp-content/uploads/2017/01/
Science-and-Technology-Roadmap-on-Catalysis-2016_Edited-Version-with-ISBN-1.
pdf, 2016 (accessed 15 February 2021).
4. Conclusions
[5] G. Arrabito, R. Pezilli, G. Prestopino, P.G. Medaglia, Crystals 10 (2020) 602.
[6] [6] SPIRE, SPIRE Roadmap, www.spire2030.eu/sites/default/files/pressoffice/
spire-roadmap.pdf, 2019 (accessed 15 February 2021).
In this work we reported the preparation of a set of Co/Al LDH
materials with increasing complexity. All of them were successfully
tested as catalysts for the reduction of nitroaromatic compounds and
dyes. The best performance was obtained with the one holding Au
nanoparticles. The DFT calculations demonstrated that there was
competition between amine moieties and borohydride, which explained
the deactivation of the Au catalyst in the recycling experiments. How-
ever, the simplest catalyst, HTCo/Al-Cl presented catalytic activity close
to the Au-containing counterpart. This opened a window of opportunity
to take further steps in exploring this system for those reactions. In
addition, this also demonstrated that efforts to prepare sustainable and
environmentally friendly catalysts such as those reported here are
promising viable options. Therefore, as exhibited by the results obtained
for HTCo/Al-Cl, based on first-row d-block elements, it was demonstrated
that noble and expensive metals can be avoided and replaced conve-
niently. Moreover, the reactions herein reported were conducted in
aqueous phase, which showed that these systems did not pose any
environmental pressure concerns in terms of residues.
[7] C.I. Fernandes, C.D. Nunes, P.D. Vaz, Curr. Org. Synth. 9 (2012) 670–694.
[8] A.R. Sotiles, L.M. Baika, M.T. Grassi, F. Wypychm, J. Am. Chem. Soc. 141 (2019)
531–540.
[9] S.R. Leandro, C.I. Fernandes, A.S. Viana, A.C. Mourato, P.D. Vaz, C.D. Nunes, Appl.
Catal. A: Gen. 584 (2019), 117155.
[10] Y. Tokudome, N. Tarutani, K. Nakanishib, M. Takahashia, J. Mater. Chem. A 1
(2013) 7702–7708.
[11] N. Tarutani, Y. Tokudome, M. Fukui, K. Nakanishi, M. Takahashi, RSC Adv. 5
(2015) 57187–57192.
[12] Y. Tokudome, T. Morimoto, N. Tarutani, P.D. Vaz, C.D. Nunes, V. Prevot, G.B.
G. Stenning, M. Takahashi, ACS Nano 10 (2016) 5550–5559.
[13] J. Wang, L. Zhao, H. Shi, J. He, Angew. Chem. Int. Ed. 5 (2011) 9171–9176.
[14] D. Kino, Y. Tokudome, P.D. Vaz, C.D. Nunes, M. Takahashi, J. Asian Ceram. Socs. 5
(2017) 466–471.
´
[15] S.R. Leandro, A. Mourato, U. Łapinska, O.C. Monteiro, C.I. Fernandes, C.D. Nunes,
P.D. Vaz, J. Catal. 358 (2018) 187–198.
[16] L. Li, L. Dou, H. Zhang, Nanoscale 6 (2014) 3753–3763.
[17] L. Wang, J. Zhang, X. Meng, D. Zheng, F. Xiao, Catal. Today 175 (2011) 404–410.
[18] J. Wang, X. Lang, B. Zhaorigetu, M. Jia, J. Wang, X. Guo, J. Zhao, ChemCatChem 6
(2014) 1737–1747.
[19] X. Wang, G. Wu, F. Wang, K. Ding, F. Zhang, X. Liu, Y. Xue, Catal. Commun. 28
(2012) 73–76.
Recycling experiments also demonstrated that although the Au
catalyst was prone to deactivation by amine products in recycling ex-
periments, the HTCo/Al-Cl catalyst seemed to be almost insensitive to
that. It is a remarkable feature of this catalyst.
[20] S. Nakagaki, M. Halma, A. Bail, G.G.C. Arizaga, F. Wypych, J. Colloid Interf. Sci.
281 (2005) 417–423.
[21] X. Bai, Y. Gao, H. Liu, L. Zheng, J. Phys. Chem. C 113 (2009) 17730–17736.
[22] F. Lin, R. Doong, J. Phys. Chem. C 115 (2011) 6591–6598.
[23] K. Layek, M.L. Kantam, M. Shirai, D. Nishio-Hamane, T. Sasaki, H. Maheswaran,
Green Chem. 14 (2012) 3164–3174.
Therefore, excellent perspectives for future developments of this
system, which are being endeavored by us, can be foreseen.
[24] H.K. Kadam, S.G. Tilve, RSC Adv. 5 (2015) 83391–83407.
[25] D. Formenti, F. Ferretti, F.K. Scharnagl, M. Beller, Chem. Rev. 119 (2019)
2611–2680.
CRediT authorship contribution statement
[26] R.K. Rai, A. Mahata, S. Mukhopadhyay, S. Gupta, P.-Z. Li, K.T. Nguyen, Y. Zhao,
B. Pathak, S.K. Singh, Inorg. Chem. 53 (2014) 2904–2909.
[27] L. Dou, H. Zhang, J. Mater. Chem. A 4 (2016) 18990–19002.
[28] N. Arora, A. Mehta, A. Mishra, S. Basu, Applied Clay Sci. 151 (2018) 1–9.
[29] Z. Wei, Y. Li, L. Dou, M. Ahmad, H. Zhang, ACS Appl. Nano Mater. 2 (2019)
2383–2396.
´
ˆ
Sonia R. Leandro: Investigation, Formal analysis. Ines J. Marques:
Investigation, Formal analysis, Writing - review & editing. Ruben S.
Torres: Investigation, Formal analysis. Tiago A. Fernandes: Investi-
gation, Formal analysis, Writing - review & editing. Pedro D. Vaz:
Conceptualization, Investigation, Data curation, Writing - review &
editing. Carla D. Nunes: Conceptualization, Methodology, Supervision,
Project administration, Writing - review & editing.
[30] S. Kundu, M.D. Mukadam, S.M. Yusuf, M. Jayachandran, CrystEngComm 15 (2013)
482–497.
[31] C. Taviot-Gueho, F. Leroux, C. Payen, J.P. Besse, Appl. Clay Sci. 28 (2005)
111–120.
[32] J.C. Dupin, H. Martinez, C. Guimon, E. Dumitriu, I. Fechete, Appl. Clay Sci. 27
(2004) 95–106.
[33] S.P. Newman, W. Jones, New J. Chem. 22 (1998) 105–115.
[34] S.P. Newman, S.J. Williams, P.V. Coveney, W. Jones, J. Phys. Chem. B 102 (1998)
6710–6719.
Declaration of Competing Interest
[35] C.I. Fernandes, P.D. Vaz, T.G. Nunes, C.D. Nunes, Applied Clay Sci. 190 (2020),
105562.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[36] M. Pavlovic, F. Dits, Z. Gu, M. Adok-Sipiczki, I. Szilagyi, RSC Adv. 6 (2016)
16159–16167.
[37] A. Corma, P. Concepcion, P. Serna, Angew. Chem. 119 (2007) 7404–7407.
Acknowledgments
The authors would like to acknowledge FCT for financial support
8