Multifunctionality and cytotoxicity of a layered coordination polymer

Article abstract of DOI:10.1039/c9dt04211c

This work reports the synthesis and multifunctionality of 2D layered coordination polymers formulated as [Ln₂(H₃nmp)₂]·xH₂O (1, where Ln = Sm³⁺, Eu³⁺, Tb³⁺, Dy³⁺, Ho

Full text of DOI:10.1039/c9dt04211c

Dalton
Transactions
View Article Online
View Journal
PAPER
Multifunctionality and cytotoxicity of a layered
coordination polymer†
Cite this: DOI: 10.1039/c9dt04211c
a
a
b
a
Patrícia Silva,‡ Ricardo F. Mendes, ‡ Carlos Fernandes, Ana C. Gomes,
a,c
d
b
Duarte Ananias,
Anabela A. Valente
Fernando Remião,
and Filipe A. Almeida Paz
Fernanda Borges,
a
a
*
This work reports the synthesis and multifunctionality of 2D layered coordination polymers formulated as
3
+
3+
3+
3+
3+
3+
3+
[
2 3 2 2
Ln (H nmp) ]·xH O (1, where Ln = Sm , Eu , Tb , Dy , Ho , Er and Y ) (x = 1 to 4). We describe
detailed synthesis of the materials using various methods [typical hydrothermal reaction (HT), microwave-
assisted synthesis (MWAS) and one-pot method (OP)], while discussing the various crystal morphologies
which can be fine tuned by varying systematically the conditions. We further explore the multifunctionality
of this material by studying its heterogeneous catalytic activity in the ring opening of styrene oxide, its
photoluminescence behaviour and its cytotoxicity. A conversion of 88%/100% yield at 4 h/24 h reaction
respectively, with excellent selectivity towards 2-methoxy-2-phenylethanol product (100%) was observed.
2 3 2 2
Photoluminescence properties of the optically-active [Eu (H nmp) ]·xH O (1Eu) and the mixed-lantha-
nide [(Gd0.95Eu0.5 (H nmp) ]·xH O (1GdEu), on the other hand, show potential use for UV-to-visible light
)
2
3
2
2
Received 30th October 2019,
Accepted 11th December 2019
converters, with lifetimes of 2.31 ± 0.01 and 2.61 ± 0.01 ms at ambient temperature for 1Eu and 1GdEu
samples. Preliminary cytotoxic studies showed no effects on metabolic activity of both in vitro human epi-
thelial kidney (HK-2) and human hepatocellular carcinoma (HepG2) cell lines. A reduction of NR uptake
was, however, observed indicating some cytotoxic effect on lysosomal activity.
DOI: 10.1039/c9dt04211c
rsc.li/dalton
6
–15
out for their use in several technological fields.
Even
Introduction
though particular attention has been devoted to gas storage
Metal–Organic Frameworks (MOFs) or Coordination Polymers and separation processes, research efforts have been spread
1
6–19
14,20,21
(
CPs) have grown from the infancy related to their general and over different fields: biomedicine,
magnetism,
in
2
2–25
26–29
sporadic reporting to a more adult perspective in which these the fabrication of membranes
compounds are now used as functional materials. Still, catalysis,
or thin films,
3
0–32
33
34,35
ion exchange, electrodes for batteries,
or
thousands of these new crystalline and regular compounds, even in the preparation of light-conversion devices (using their
6
,11,36
obtained from the self-assembly of metal oxide clusters with peculiar photoluminescent properties as optical sensors).
organic linkers, can be found in the literature, some also
Our research group focused in the last decade on the prepa-
1
–5
based on peculiar structural architectures.
As expected, ration of MOFs and CPs based lanthanides self-assembled
these hybrid (organic–inorganic) materials have been pointed with organic linkers bearing phosphonic acid groups. By fine
tuning the experimental conditions and designing new
organic linkers we were able to obtain a myriad of 1-, 2- and
a
3-dimensional materials with interesting properties. In fact, we
were able to obtain materials with potential application in
Departamento de Química, CICECO, Aveiro Institute of Materials, Universidade de
Aveiro, 3810-193 Aveiro, Portugal. E-mail: filipe.paz@ua.pt; Fax: (+351) 234 401470;
Tel: (+351) 234 401418
3
7
38,39
heterogeneous catalysis, photoluminescence
recently as excellent proton conductors.
and more
One of these
b
CIQUP – Centro de Investigação em Química, Departamento de Química e
40
Bioquímica, Faculdade de Ciências, Universidade do Porto, 4169-007 Porto, Portugal
materials, a positively charged 2D layered material based on
c
Departamento de Física, CICECO, Aveiro Institute of Materials, Universidade de
3
+
Gd and the H
6
nmp organic linker [H
6
nmp = nitrilo(tri-
Aveiro, 3810-193 Aveiro, Portugal
d
3
7
UCIBIO-REQUIMTE, Laboratório de Toxicologia, Departamento de Ciências
methylphosphonic)acid], [Gd(H
4
nmp)(H O)
2
2
]Cl·2H
2
O,
pre-
Biológicas, Faculdade de Farmácia, Universidade do Porto, 4050-313 Porto, Portugal
sented an interesting single-crystal to single-crystal transform-
ation into another 2D layered material. This transformation is
followed by a considerable increase of proton conduction
†
Electronic supplementary information (ESI) available: Experimental conditions
and additional characterization of compound 1. Additional catalytic and photo-
luminescence data. Electron microscopy studies: EDS mapping for the europium
and the mixed-lanthanide materials. See DOI: 10.1039/c9dt04211c
−
1
reaching 0.51 S cm , being one of the best reported materials
4
1
‡
These authors contributed equally to the present work.
up to date. While in that work we solely focused on the
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
41
2 3 2 2
Fig. 1 Schematic representation of the crystal packing of [Gd (H nmp) ]·xH O (1Gd) (x = 1 to 4).
proton conduction process, herein we present
a
Several conclusions can be drawn (please note: all prepared
detailed characterization of the transformed material, materials have been inspected using a combination of various
3
+
3+
[
2 3 2 2
Ln (H nmp) ]·xH O (x = 1 to 4) (1, where Ln = Sm , Eu , characterisation techniques):
3
+
3+
3+
3+
3+
Tb , Dy , Ho , Er and Y ) (Fig. 1), in terms of synthesis
optimization. In fact, this new crystalline 2D structure has been nide cations with small atomic radius (Sm , Eu , Tb , Dy ,
prepared by alternative synthetic methods based on typical Ho , Er and Y ). The use of Yb promotes the formation of
hydrothermal reaction (HT), microwave-assisted synthesis an amorphous compound.
(i) Isotypical materials are preferentially isolated for lantha-
3
+
3+
3+
3+
3
+
3+
3+
3+
(MWAS) and one-pot method (OP), either using the respective
(ii) HS leads to “poorly” crystalline 1 that could be isolated
lanthanide chloride salt or its oxide precursor. This optimiz- after just a few hours of reaction (e.g., 2 h at 140 °C or 1 h at
ation allowed us to synthesize 1 in just 1 minute (using MWAS), 190 °C); crystallinity could be slightly improved using longer
which is an important improvement for the rapid preparation reaction periods (e.g., 4 days at 140 °C or 2 days at 190 °C).
of potential functional MOF compounds. This paper further
(iii) Using HS, the molar ratio composition of the initial
explores the potential multifunctionality of this material as a reaction mixture does not influence neither the isolated
heterogeneous catalyst in the ring opening of styrene oxide and product nor its overall crystallinity.
as a light-converting compound. We have also evaluated its
(iv) Using MWAS, for temperatures lower than ca. 100 °C
cytotoxic effect in human epithelial kidney (HK-2) and human amorphous materials are typically isolated; at ca. 140 °C and
hepatocellular carcinoma (HepG2) cell lines.
100 W of power, physical mixtures of 1 with another (known)
layered material, [Gd(H nmp)], are isolated. For this latter
3
4
2
temperature, decreasing the irradiation power to, e.g. 50 W, 1
is isolated as a pure phase (please note: a careful combination
of irradiation power and the time for the heating ramp needs
to be adjusted).
Results and discussion
Synthesis optimization
Optimization of the synthetic conditions revealed that
(v) Contrasting with HS, in the MWAS the reaction time
[
Ln
2 3 2 2
(H nmp) ]·xH
O (1) (x = 1 to 4) could readily be isolated plays a decisive role in the isolation of 1. Mixtures with either
4
3
42
using various synthetic approaches: static hydrothermal syn- [Gd(H
nmp)]·1.5H
O
or [Gd(H nmp)] are easily obtained:
3
3
2
thesis (HS) based on convection heating, microwave-assisted the optimal conditions to isolate 1 within just 10 minutes
synthesis (MWAS) and one-pot procedures (OP) at atmospheric correspond to a temperature of ca. 100–120 °C and an
pressure (Fig. S1†). It was additionally found that the material irradiation power of ca. 100 W.
can also be isolated using a wide range of composition for the
(vi) The one-pot method (OP) produces again “poorly” crys-
reactive mixtures. We note that throughout this manuscript talline 1 at lower temperatures (e.g., refluxing at 80 °C for
each sample can be identified as T P t (with T standing for 12 h). Slightly larger crystals (isolated as aggregated) of 1 were
m
n p
temperature, P for pressure, and t for reaction time) with the obtained using the OP method by replacing the metal source
subscripts indicating each set of specifically employed experi- (chloride for the oxide) and including HCl in the reaction
mental conditions.
medium. The optimal conditions were found to be: 100 °C for
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
2 h using a volume ratio of HCl/H
Paper
1
2
O of about 1/9, or 120 °C different shapes (Fig. 2 and 3). Crystals are, thus, very regular
for 48 h and HCl/H O of about 1/1. Please note: with the thin (in the nanometer range) plates which grow uniformly
2
increase of acid concentration we can observe the formation of without the presence of precipitates (or other defects). This
yet another 2D layered material also previously reported by us: was observed using TEM studies, but because of the sensitivity
3
7
[
Gd(H nmp)(H O) ]Cl·2H O.
of 1 towards the electron beam (most likely due to the presence
of a large amount of water molecules), more detailed high-
resolution TEM studies could not be performed.
4
2
2
2
Crystal morphology and particle size
Crystals of [Ln (H nmp) ]·xH O (1) (x = 1 to 4) present a plate-
Considering the average particle size, we observed vari-
2
3
2
2
like morphology for a wide range of synthetic conditions, with ations according to the employed reaction conditions, specifi-
individual crystallites being stacked into aggregates with cally the heating method. A significant decrease in the particle
Fig. 2 SEM images of bulk [Ln
2 3 2 2
(H nmp) ]·xH O (1) (x = 1 to 4) obtained using hydrothermal reaction conditions. The effect on the crystallite mor-
phology and size when varying the reaction time and the temperature is emphasised.
2 3 2 2
Fig. 3 SEM images of the bulk material [Ln (H nmp) ]·xH O (1) (x = 1 to 4) obtained using microwave-assisted hydrothermal synthesis. The effect
on the crystallite morphology and size when varying the reaction time and the temperature is shown.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
size happens when we change from the typical hydrothermal having dozens of micrometres), in addition to the concomitant
approach to the microwave-assisted synthesis (Fig. 3). This increase in the crystallite thickness.
effect was expected considering that microcrystalline powders
Having in mind the latter result, we have transposed the
or nano-sized crystals are most often isolated under microwave knowledge of using HCl in this chemical system to the typical
heating due to the fast kinetics of crystal nucleation and hydrothermal conditions in order to unveil if large crystals of 1
4
4
growth. The particle size is, thus, a feature which can be could be obtained by this route. Despite this objective was
somehow controlled by changing the synthesis approach readily achieved, unlike before, the purity of the final product
between microwave heating and typical hydrothermal is dependent on the cooling rate of the reaction mixture. For
conditions.
this case, when quenching the reaction vessel 1 was always
For the case of the hydrothermal approach the increase in obtained as a phase-pure compound, while mixtures of 1 with
temperature yielded plate-like units with larger thickness and the aforementioned unknown material were observed when
overall size, while no differences were observed when increasing slow cooling was used.
the reaction time. Furthermore, the differences that occasionally
could be observed in the dimension of the aggregates when
Solid-state structural characterization
3
1
increasing the time of reaction are not significant and similar
Solid-state NMR. Fig. S2 (in the ESI†) shows the P MAS
sizes have been found for a wide range of experimental con- solid-state NMR spectrum of the isotypical [Y (H nmp) ]·xH O
2
3
2
2
ditions. Noteworthy, the aggregation of crystallites usually gave material (1Y) (x = 1 to 4). The isotropic region of the spectrum
rise to well defined spherical agglomerates (Fig. 2).
is composed of two main sets of resonances. One is formed of
Using MWAS it is possible to note a similar increase on the three convoluted resonances peaking at ca. −1.10, −0.50 and
average particle size when increasing the temperature, though −0.17 ppm, and a more resolved peak at ca. 1.35 ppm. The
this behaviour is not so clear as before. For this case, and more deshielded resolved peak is, in fact, composed of two
unlike the typical hydrothermal reaction, the agglomeration of resonances peaking at ca. 9.07 and 9.44 ppm. Peak deconvolu-
particles gave rise to clusters with irregular shapes (Fig. 3).
tion and integration, including the spinning sidebands, yields
The one-pot method was the one that showed more overall a ratio of ca. 1 : 1 : 1 : 1 : 1 : 1 for the six resonances, which
differences. Even when the purity of the final product has not agrees well with the presence of six crystallographically inde-
3
1
been compromised for different experimental conditions, this pendent P sites composing the asymmetric unit of 1Y.
method was able to yield particles with distinct sizes and
FT-IR. Vibrational spectroscopy studies also support the
shapes. In fact, when using the same molar ratio of the reac- structural features unveiled by the X-ray diffraction studies.
tants, the increase in temperature (from 100 °C to 120 °C), in Fig. S3† presents the FT-IR spectra of [Gd (H nmp) ]·xH O
2
3
2
2
reaction time (from 12 h to 48 h) and in ratio of solvent (1Gd) (x = 1 to 4), in the 3800–400 cm− region, including
1
2
mixture (from HCl/H O 1/10 to 1/1) resulted in a significant assignments for each main observed band. 1Gd contains a
increase on the crystallite dimensions (Fig. 4). The reactions in broad band centred at about 3500 cm− attributed to the
which the quantity of water was much higher than that of ν(O–H) stretching vibrational modes from water molecules.
hydrochloric acid, resulted in 1 having particle morphology The typical ν(N–H) and ν(C–H) stretching vibrational modes of
1
−
1
alike to that previously observed in the first two synthetic the N–C–H group appear in the 3100–2600 cm region. In the
methods. Nevertheless, the agglomeration of the crystallites is 1500–1300 cm− range, a number of very weak bands can be
1
rather similar to that obtained from MWAS, though the crystal- attributed to ν(C–H) modes characteristic of P–CH moieties.
2
lite size can have in some cases a slightly smaller size. On the The typical P–OH stretching modes were also observed
other hand, and for higher contents of HCl we observed the between 2500 and 2200 cm− , as very faint and broad bands.
presence of rectangular and larger particles of 1 (with length In the range of 1500–600 cm− , it is possible to notice the
1
1
Fig. 4 SEM images of the bulk material [Ln
phology and size when varying the temperature, the reaction time and the molar ratio of the HCl/H
2
(H
3
nmp)
2
]·xH
2
O (1) (x = 1 to 4) obtained using the one-pot method. The effect on the crystallite mor-
O mixture is emphasised.
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
typical ν(C–N) stretching vibrational modes of tertiary amines (Fig. 5 – inset). With the increase in temperature and the early
assigned to the intense bands pointing at about 1099 and release of the water moieties, the first reflection [corres-
1
1
070 cm− . Also, in this region, the stretching modes of ponding to the (001) plane] of the material is shifted to higher
−
1
ν(PvO) are noticed from ca. 1250 to 1150 cm , and those 2theta values while getting significantly broader. This obser-
of ν(P–O) from ca. 1080 to 900 cm . We observe as well the vation suggests a reduction of the unit cell dimensions, par-
−
1
ν(P–C) stretching vibrational modes, in particular between ca. ticularly along the [001] direction, which further indicates a
70–600 cm−
1 45
.
closer proximity between the 2D layers. The thermodiffracto-
7
Thermal stability. The thermal stability of bulk metry study clearly shows a decrease in the overall crystallinity
Gd (H nmp) ]·xH O (1Gd) (x = 1 to 4) was investigated of the bulk material. Between 140 °C and 350 °C no additional
[
2
3
2
2
between ambient temperature and ca. 800 °C providing weight losses are observed, including changes of the crystalline
additional information on the overall hydration level of the phase or overall crystallinity. At higher temperatures (above ca.
compound (Fig. 5). Thermogravimetric analysis revealed three 350 °C) we may consider three subsequent weight losses, glob-
main weight losses, with the first occurring between ambient ally corresponding to ca. 1.6% (350–425 °C range), ca. 4.4%
temperature and ca. 140 °C attributed to the early release of (425–620 °C range) and ca. 3.3% (620–800 °C range), respect-
crystallization water molecules. Indeed, in this temperature ively. These latter three weight losses are attributed to the full
range, the compound loses approximately 4.7% of total decomposition of the organic component, with the thermo-
weight, which agrees with the release of 1.4 water molecules diffractometry indicating the existence of an almost amor-
per formula unit (calculated 3.6% for the release of a single phous residue at the end. Even though the final residue could
water molecule). We note that thermogravimetric analysis of not be identified using powder X-ray diffraction, it is reason-
samples obtained with different experimental conditions can able to assume that it is composed of a mixture of phospho-
lead to slightly different, ultimately corresponding to the nate and oxides of the corresponding lanthanide employed.
release between one and up to four water molecules (which
Heterogeneous catalysis
also induces to changes in elemental analysis). More details of
the structure are explained elsewhere.
4
1
2 3 2 2
[Gd (H nmp) ]·xH O (1Gd) (x = 1 to 4) was investigated as a
Thermodiffractometry shows that the overall crystalline heterogeneous catalyst in the reaction of the ring opening of
structure of the material only remains intact up to ca. 50 °C styrene oxide. 2-Methoxy-2-phenylethanol was the sole product
2 3 2 2
Fig. 5 Thermogram and variable-temperature powder X-ray diffraction studies of [Gd (H nmp) ]·xH O (1Gd) (x = 1 to 4).
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
a
Table 1 Ring opening of styrene oxide with methanol, at 55 °C
b
c
Sample
None
Reaction time (h)
Conversion (%)
Selectivity (%)
4
2
100
1
Gd
4/24
4/24
1
88/100
65/94
100
100/100
75/82
100
GdCl
H
3
·6H
2
O
d
6
nmp
a
Reaction conditions: Styrene oxide (0.83 mmol), catalyst (36.5 mg
Gd, or 0.08 mmol of GdCl or H nmp), methanol (2 mL), in a closed
batch reactor at 55 °C. Conversion of styrene oxide. Selectivity to
1
3
6
b
c
d
2
-methoxy-2-phenylethanol. The same catalytic results were obtained
using 100 times less initial amount of H nmp.
6
formed in 88%/100% yield at 4 h/24 h reaction, respectively,
exhibiting excellent regioselectivity towards 2-methoxy-2-phe-
nylethanol. Under similar experimental conditions, but in the
absence of the catalyst, the conversion of styrene oxide was
only of ca. 2%/10% at 4 h/24 h of reaction, respectively
(Table 1).
The primary building units of 1Gd, i.e., the ligand (H nmp)
6
and the lanthanide source (GdCl
catalysts in the methanolysis of styrene oxide at 55 °C (using
3
·6H
2
O), were also tested as
Fig. 6 Catalytic reaction of the ring opening of styrene oxide with
methanol, at 55 °C and three different reuse runs, in the presence of the
equivalent molar quantities as to those found in 1Gd). For the heterogeneous catalyst [Gd (H nmp) ]·xH O (1Gd) (x = 1 to 4).
2
3
2
2
two precursors the reaction mixture was homogeneous in
nature. In the case of H nmp, 2-methoxy-2-phenylethanol was
formed in quantitative yields within 1 h of reaction, which
6
st
rd
The catalysts recovered from the 1 and 3 runs were
indicates that this molecule is a highly active and selective characterized in the solid state by FTIR spectroscopy, PXRD
homogeneous Brønsted acid organocatalyst (Table 1). A and microscopy imaging (Fig. S4–S6 in the ESI†). The FTIR
hundred-fold decrease in the initial amount of H nmp gives spectra of the re-used catalysts are similar to that of the as-pre-
6
1
00% conversion at 1 h of reaction, further strengthening the pared material, with only a handful of small differences in the
high catalytic activity of the ligand precursor. In the case of bands assigned to the P–O and P–OH vibrational modes
the lanthanide precursor, GdCl ·6H O, the reaction of styrene (Fig. S5†). The PXRD data show a small decrease in crystalli-
3
2
oxide with methanol gives 2-methoxy-2-phenylethanol in 86% nity, with SEM images further revealing a gradual disaggrega-
yield at 48 h reaction (89% selectivity). In addition, two minor tion of the crystallites. We thus infer that the small decrease in
products were also detected and identified as the catalytic activity may be attributed to the destruction of some
α-(chloromethyl)-benzyl alcohol and (chlorophenyl)ethanol active sites. Nevertheless, TEM investigations also showed that
isomers, formed in ca. 3% and 8% selectivity, at 98% conver- after catalysis 1Gd retains its uniform nature with no precipi-
sion. A major limitation of these homogeneous catalysts is tates (or other defects) being observed in the crystal faces
their difficult recyclability which makes their potential appli- (Fig. S4†).
cation unattractive.
Photoluminescence
The recyclability of catalyst 1Gd was investigated in three
3
+
consecutive batch runs. Prior to reuse, the solid catalyst The isotypical pure Eu material, [Eu (H nmp) ]·xH O (1Eu)
2
3
2
2
was separated from the reaction mixture by centrifugation, (x = 1 to 4) emits the characteristic red colour when irradiated
washed with n-hexane and dried at ambient temperature over- under UV light. A mixed-lanthanide phase was also prepared
3
+
night. 2-Methoxy-2-phenylethanol remained as the sole reac- in which a stoichiometric amount of Eu was dispersed in the
3
+
tion product. A comparison of the kinetic curves for consecu- Gd -based matrix of the network: [(Gd0.95Eu ) (H nmp) ]·
0.05 2
3
2
tive batch runs shows a significant decrease in the reaction xH O (1GdEu) (x = 1 to 4). Both materials were characterized
2
st
nd
rate from the 1 to the 2 run; this is less pronounced from by SEM microscopy and EDS mapping showing unequivocally
the 2 to the 3 run (Fig. 6). Leaching tests performed for the presence of Eu in the mixed-lanthanide matrix, being
nd
rd
3+
the as-prepared material gave ΔLT(1–4 h) of 5%, which is distributed in a homogenous way (Fig. S7 and S8 in the ESI†).
minor in comparison to 33% conversion for the same interval
The excitation spectrum of 1Eu recorded at ambient temp-
of reaction time in the presence of 1Gd. Leaching tests per- erature (298 K) and 12 K (Fig. 7) displays the characteristic
st
rd
3+
6
7
formed for the solid catalysts recovered from the 1 and 3
runs also gave negligible values of ΔLT(1–4 h). These
results suggest that the catalytic reaction is heterogeneous in spectra exhibits at ambient temperature a broad band attribu-
sharp lines of the Eu intra-4f transitions, namely F0,1 →
5
5
5
5
5
D4–1, L , G , H and F1–5. In the high energy region the
6
2–6
3–7
3
+
2−
3+
nature.
table to a ligand-to-Eu (O → Eu ) charge transfer (CT),
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 7 Excitation spectra of [Eu
2
(H
3
nmp)
2 2
]·xH O (x = 1 to 4) (1Eu,
Fig. 8 Emission spectra of (bottom) [Eu
2
(H
3 2 2
nmp) ]·xH O (x = 1 to 4)
bottom) and [(Gd0.95Eu0.05
)
2
(H
3
nmp)
2
]·xH O (x = 1 to 4) (1GdEu, top)
2
(
1Eu) and (top) [(Gd0.95Eu0.05 (H nmp) ]·xH
)
2
3
2
2
O (x = 1 to 4) (1GdEu), with
recorded at ambient temperature (black line) and at 12 K (red line), while
monitoring the emission at 611 nm. The intensity is comparable for the
variation of the temperature in each sample, but not between different
samples.
excitation at 393, recorded at ambient conditions (black line; 298 K and
pressure of ca. 1 bar), with a high vacuum (blue line; 298 K and pressure
−6
−6
of ca. 5 × 10 mbar), and at 12 K (red line; pressure of ca. 5 × 10
mbar). Please note: the intensity is only comparable for the variation of
the temperature and pressure in each sample, not between different
samples.
peaking at 249 nm (4.98 eV; FWHM of 0.34 eV), which agrees
4
6
well with reported values for related materials. At low temp-
erature (12 K) an additional and more complex broad band times of 2.31 ± 0.01 and 2.61 ± 0.01 ms, at ambient temperature
appears with, at least, two peaks at ca. 275 and 282 nm. This and 12 K, and 2.61 ± 0.01 and 2.63 ± 0.01 ms, at ambient temp-
temperature-dependent broad band is partially superimposed erature and 12 K, respectively, for the 1Eu and 1GdEu materials,
3
+
6
3+
with some of the Eu intra-4f sharp lines. For 1GdEu the respectively. Assuming the presence of two distinct Eu
3
+
7
excitation spectra are dominated by the Gd intra-4f lines, environments, as called by the crystallographic determination
such as, the S7/2 → I7/2–11/2 and P3/2–7/2 (Fig. 7), showing the and model of the material, the time-resolved emission spectra
occurrence of an energy transfer from the Gd to the Eu
8
6
6
41
3
+
3+
3+
and the decay curves indicates that the two Eu local sites must
ions, which is more effective at ambient temperature.
be very similar, almost identical. In the present case, the emis-
Fig. 8 shows the emission spectra of 1Eu (bottom) and sion spectra of both 1Eu and 1GdEu samples are similar, even
GdEu (top) excited at 393 nm as a function of pressure and with distinct excitation wavelengths (data not shown).
1
3
+
temperature. The sharp lines are assigned to transitions
Assuming that the two Eu sites have identical emission
5
between the first excited non-degenerate
0
D state and the properties (they cannot be distinguished), we can follow the
7
3+
3+
F0–4 levels of the fundamental Eu septet. Except for the small modifications on the first Eu coordination sphere,
5
7
0 1
D → F lines, which have a predominant magnetic-dipole such as those promoted by high vacuum exposure, analysing
3
+
character independent of the Eu crystal site, the observed the changes observed in the emission spectra. The ratio
7
5
7
5
transitions are mainly of electric-dipole nature. The F levels between the integrated intensities of the D → F and D →
0
–2
0
2
0
7
have a local-field splitting into one, three and, at least, six
F
1
transitions, I
(
5
D →
7
F2)/I
(
5
D →
F1), known as the asymmetric
7
0
0
Stark sublevels, respectively. Considering the maximum split- ratio (R), gives values at ambient pressure and high vacuum of
ting by the crystal-field effect in 2J + 1 Stark sublevel for a 2.24 and 2.34, and 2.30 and 2.43 for 1Eu and 1GdEu, respect-
3
+
unique Ln local site, the identification of at least six com- ively. These values, typical of moderately symmetrical local
5
7
ponents for the D
0 2
→ F transition in both 1Eu and 1GdEu environments, point to a slightly more distorted environment
3
+
samples is a clear indication of the presence of more than one of the Eu coordination polyhedron with the vacuum exposure
single Eu environment in the materials. Additionally, the in both samples (a low value indicates a lower distortion of
third Stark component of the
3
+
5
7
3+
0 1
D → F do not has a sym- Eu local environment, approaching to the ideal case of an
5
7
metrical shape which can be due to the superposition of mul- inversion centre for which the D → F transition is absent).
0
2
tiple levels. The time-resolved emission spectra of 1Eu (Fig. S9 This is probably due to the release of some solvent water mole-
3
+
in the ESI†) for very distinct resolution times gives almost cules. The small distortion on the Eu coordination polyhe-
5
3+
3+
identical results. Also, the D lifetime of Eu of the pure and dron also justifies the decrease of the associated Eu emission
0
3
+
diluted (5%) Eu samples (Fig. S10 in the ESI†), indepen- lifetime from 2.31 ± 0.01 to 2.01 ± 0.01 ms, when the 1Eu
dently of the selected excitation and emission wavelengths, are sample is exposed to the high vacuum at ambient temperature
always well fitted by single exponential functions, yielding life- (Fig. S9 in the ESI†).
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
In vitro cytotoxicity studies
metallic course (Gd
2
O
3
) were also tested at concentrations
calculated by the amount of each component in the CP
structure. The organic linker (H nmp) and the gadolinium
metallic course (Gd ) were also tested at concentrations cal-
culated by the amount of each component in the CP structure.
Cell viability was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) and neutral red (NR)
assays. The results are presented in Fig. 8 and expressed as
mean (% of control) ± SEM of three independent experiments
The multifunctionality of MOFs and CPs, as well as the con-
tinuous change from academic research to a more industrial
approach, gives arguments to the growing concern regarding
both the potential environmental and health impact. Although
most of these compounds are not meant to be used in bio-
medical applications, the increase use of MOFs and CPs can
lead to bioaccumulation through contamination of, for
example, water sources. For this reason, we believe that is
necessary to further investigate the potential cytotoxicity of
6
2 3
O
(
n = 3).
After 24 hours of exposure, it was observed a slight (but sig-
The cytotoxic profile of [Gd (H nmp) ]·xH O (x = 1 to 4) nificant) decrease in metabolic activity of HK-2 cells when
4
7
such materials.
2
3
2
2
(
1Gd) in human epithelial kidney (HK-2) and human hepato- treated with 1Gd, the H
cellular carcinoma (HepG2) cell lines was evaluated in order to concentration. The treatment of HepG2 cells with 1Gd (all
estimate the potential harmful effects of bioaccumulation of tested concentrations), H nmp ligand (5.2 and 10.4 µM) and
this material in cells that play a crucial role in metabolism and Gd (12.4 µM) also led to a slight decreasing of formazan
For that, both cell lines were exposed to 1Gd at formation. However, in both cell lines, metabolic activity was
6 2 3
nmp ligand and Gd O at the lowest
6
2 3
O
4
8,49
excretion.
concentrations of 100, 200 and 400 µg mL for 24 hours. Cell higher than 88.2%, which is not considered as a cytotoxic
viability was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe- effect of materials.
nyltetrazolium bromide (MTT) and neutral red (NR) assays
The treatment of cells with Gd
−
1
5
0
O caused a decrease in NR
2 3
(
Fig. 9). The organic linker (H
6
nmp) and the gadolinium dye uptake in HK-2 cells (10.4 µM, 84.8 ± 3.4%; 20.8 µM, 77.2 ±
O (x = 1 to 4) (1Gd) (black data; 100, 200 and 400 µg mL− ), H
1
nmp ligand (red data; 5.2, 10.4
Fig. 9 Cytotoxicity evaluation of [Gd
and 20.8 µM) and Gd
2
(H
3
nmp)
2
]·xH
2
6
2
O
3
(blue data; 3.1, 6.2 and 12.4 µM) in HepG2 and HK-2 cell lines after 24 hours of exposure determined by MTT reduction
and the NR uptake assays. Results are expressed as mean % cell viability ± SEM (n = 3) Statistical comparisons were made using two-way ANOVA
and in all cases, p values lower than 0.05 were considered significant (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. control data).
6 2 3
Legend: M – 1Gd, L – H nmp, G – Gd O .
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
3
.0%) and in HepG2 cells (20.8 µM, 82.1 ± 3.4%). These are regarded as potentially toxic and harmful compounds, 1
harmful effects were also observed for the highest concen- showed no cytotoxic effects in metabolic activity of both
tration tested after treatment with 1Gd, which led to a decrease in vitro human epithelial kidney (HK-2) and human hepatocel-
of NR uptake percentage to 71.8 ± 2.1% and 72.7 ± 4.5% for lular carcinoma (HepG2) cell lines. It caused, however, a
HK-2 and HepG2 cells, respectively.
The data from cytotoxicity evaluation suggested that 1Gd, mal activity of the cells. The data obtained in this study
as well as the H nmp ligand and Gd , did not exert any cyto- reflects the importance of understand the MOF cytotoxic
toxic effects in metabolic activity of the two types of in vitro profile as well as the mechanisms that led to cell viability
cell models. However, 1Gd and Gd caused a reduction of decrease, since due to their inherent insolubility, MOFs might
reduction of NR uptake suggesting a cytotoxic effect in lysoso-
6
2 3
O
2 3
O
NR uptake in both cell lines. Despite the lack of information be subject to bioaccumulated in organisms. Further studies
in the literature related with a typical MOF cytotoxic profile, are currently being carried out.
Tamames-Tabar et al. reported that after cellular internaliz-
ation, the MOF particles are localised mainly in cytoplasmatic
5
1
Conflicts of interest
organelles, such as lysosomes. They further suggested that
lysosomal enzymes might degrade the polymeric structure of
There are no conflicts to declare.
3
+
MOF releasing the content, which in our case contains Gd
ions. In the alignment of this, materials containing gadoli-
nium could led to lysosomal rupture in concentration and Acknowledgements
5
2
time-dependent manners. This combined information allow
This work was developed within the scope of the project
CICECO – Aveiro Institute of Materials, FCT Ref. UID/CTM/
us to suggest that both 1Gd and Gd O3 presented cytotoxic
2
effects in lysosomal activity of the cells in a concentration-
dependent manner.
5
0011/2019, financed by national funds through the FCT/
MCTES. The work was also supported by UID/MULTI/04378/
019 and UID/QUI/00081 with funding from FCT/MCTES
2
through national funds. CF grant is supported by FCT, POPH
and FEDER/COMPETE. FCT is also gratefully acknowledged
for the Junior Research Position CEECIND/00553/2017 (to
RFM), for the Assistant Research Position CEECIND/02128/
2017 (to ACG) and for the doctoral research grant No. SFRH/
BD/46601/2008 (to PS).
Conclusions
In this work we report the synthesis and multifunctionality of
the 2D layered coordination polymer [Ln
2
(H
3
nmp)
2
]·xH
2
O (1,
where Ln = Sm , Eu , Tb , Dy , Ho , Er and Y ) (x = 1 to
). This material was recently obtained by us during a single-
2
crystal to single-crystal transformation of [Gd(H nmp)(H O) ]
3
+
3+
3+
3+
3+
3+
3+
4
4
2
Cl·2H O at high temperature and relative humidity. The trans-
2
References
formation lead to an increase in proton conductivity, reaching
values never reported for other MOF materials (up to 0.51
S cm− ). Here, the heterogeneous catalytic activity, photo-
luminescent behaviour and cytotoxicity of 1 is reported.
1 K. J. Gagnon, H. P. Perry and A. Clearfield, Chem. Rev.,
2012, 112, 1034–1054.
2 H. Gliemann and C. Wall, Mater. Today, 2012, 15, 110–116.
3 M. O’Keeffe and O. M. Yaghi, Chem. Rev., 2012, 112, 675–
702.
4 N. Stock and S. Biswas, Chem. Rev., 2012, 112, 933–969.
5 V. Guillerm and D. Maspoch, J. Am. Chem. Soc., 2019,
16517–16538.
6 Y. J. Cui, Y. F. Yue, G. D. Qian and B. L. Chen, Chem. Rev.,
2012, 112, 1126–1162.
7 P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin,
P. Couvreur, G. Ferey, R. E. Morris and C. Serre, Chem. Rev.,
2012, 112, 1232–1268.
8 L. E. Kreno, K. Leong, O. K. Farha, M. Allendorf, R. P. Van
Duyne and J. T. Hupp, Chem. Rev., 2012, 112, 1105–1125.
9 M. Yoon, R. Srirambalaji and K. Kim, Chem. Rev., 2012,
112, 1196–1231.
1
1
showed good catalytic activity in the ring opening of
styrene oxide, with the formation of 2-methoxy-2-phenyletha-
nol as the sole product with 88%/100% yield at 4 h/24 h of
reaction. The recyclability was also tested in three different
st
runs, showing a decrease in the reaction rate from the 1 to
nd
nd
rd
the 2 run, being less pronounced from the 2 to the 3 run.
This may be attributed to the destruction of some active sites,
since we observe a small loss in crystallinity and the crystals
reveal
a gradual disaggregation of the crystallites. The
inclusion of optically active lanthanide centers, on the other
hand, yielded photoluminescent materials. Clearly, the
absence of aromatic component of the organic linker accounts
3
+
for lower emission of the Eu and the mixed-lanthanide-
based material. Nonetheless, 1 has proved to be effective in
3
+
transferring energy to Eu with lifetimes of 2.31 ± 0.01 and 10 W. Zhang and R.-G. Xiong, Chem. Rev., 2012, 112, 1163–
2
0
.61 ± 0.01 ms, at ambient temperature and 12 K, and 2.61 ±
.01 and 2.63 ± 0.01 ms, at ambient temperature and 12 K, 11 J. Rocha, L. D. Carlos, F. A. A. Paz and D. Ananias, Chem.
Soc. Rev., 2011, 40, 926–940.
A preliminary cytotoxic study was also performed for 1. 12 O. Shekhah, J. Liu, R. A. Fischer and C. Woll, Chem. Soc.
1195.
respectively, for the 1Eu and 1GdEu samples.
While MOFs and CPs, especially lanthanide-based materials,
Rev., 2011, 40, 1081–1106.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
1
1
1
3 A. Morozan and F. Jaouen, Energy Environ. Sci., 2012, 5, 35 J. P. Zhu, X. H. Wang and X. X. Zuo, R. Soc. Open Sci., 2019, 6,
269–9290. 190634.
4 D. Dusan, H. Youcef, C. L. Lin, Y. Tan, L. Kunhao and 36 W. T. Koo, J. S. Jang and I. D. Kim, Chem, 2019, 5, 1938–
L. Jing, J. Appl. Phys., 2012, 111, 07E335. 1963.
5 P. Silva, S. M. F. Vilela, J. P. C. Tomé and F. A. A. Paz, Chem. 37 R. F. Mendes, M. M. Antunes, P. Silva, P. Barbosa,
9
Soc. Rev., 2015, 44, 6774–6803.
F. Figueiredo, A. Linden, J. Rocha, A. A. Valente and
6 Z. Q. Jiang, B. Yuan, N. X. Qiu, Y. J. Wang, L. Sun,
F. A. A. Paz, Chem. – Eur. J., 2016, 22, 13136–13146.
Z. N. Wei, Y. Y. Li, J. J. Zheng, Y. H. Jin, Y. Li, S. Y. Du, J. Li 38 D. Ananias, A. D. G. Firmino, R. F. Mendes, F. A. A. Paz,
and A. G. Wu, Nano-Micro Lett., 2019, 11, 61.
7 I. A. Lazaro and R. S. Forgan, Coord. Chem. Rev., 2019, 380,
M. Nolasco, L. D. Carlos and J. Rocha, Chem. Mater., 2017,
29, 9547–9554.
1
1
1
2
2
30–259.
39 R. F. Mendes, D. Ananias, L. D. Carlos, J. Rocha and
F. A. A. Paz, Cryst. Growth Des., 2017, 17, 5191–5199.
40 A. D. G. Firmino, R. F. Mendes, M. M. Antunes,
P. C. Barbosa, S. M. F. Vilela, A. A. Valente,
F. M. L. Figueiredo, J. P. C. Tome and F. A. A. Paz, Inorg.
Chem., 2017, 56, 1193–1208.
8 Z. D. Luo, S. R. Fan, C. Y. Gu, W. C. Liu, J. X. Chen, B. H. Li
and J. A. Liu, Curr. Med. Chem., 2019, 26, 3341–3369.
9 L. Xinyi, W. Zhiyong, Z. Yongtai and C. Yongming, Prog.
Chem., 2019, 31, 783–790.
0 D. W. Ryu, W. R. Lee, J. W. Lee, J. H. Yoon, H. C. Kim,
E. K. Koh and C. S. Hong, Chem. Commun., 2010, 46, 8779– 41 R. F. Mendes, P. Barbosa, E. M. Domingues, P. Silva,
8
781.
F. Figueiredo and F. A. A. Paz, Nat. Commun., 2019,
Submitted.
2
2
2
2
2
2
2
2
2
1 H. Tian, Q. X. Jia, E. Q. Gao and Q. L. Wang, Chem.
Commun., 2010, 46, 5349–5351.
2 T. H. Bae, J. S. Lee, W. L. Qiu, W. J. Koros, C. W. Jones and
S. Nair, Angew. Chem., Int. Ed., 2010, 49, 9863–9866.
42 P. Silva, F. Vieira, A. C. Gomes, D. Ananias, J. A. Fernandes,
S. M. Bruno, R. Soares, A. A. Valente, J. Rocha and
F. A. A. Paz, J. Am. Chem. Soc., 2011, 133, 15120–15138.
3 A. S. Huang, W. Dou and J. Caro, J. Am. Chem. Soc., 2010, 43 L. Cunha-Silva, L. Mafra, D. Ananias, L. D. Carlos, J. Rocha
32, 15562–15564. and F. A. Almeida Paz, Chem. Mater., 2007, 19, 3527–3538.
4 Y. X. Hu, X. L. Dong, J. P. Nan, W. Q. Jin, X. M. Ren, 44 J. Klinowski, F. A. A. Paz, P. Silva and J. Rocha, Dalton
N. P. Xu and Y. M. Lee, Chem. Commun., 2011, 47, 737–739. Trans., 2011, 40, 321–330.
5 M. Kadhom and B. L. Deng, Appl. Mater. Today, 2018, 11, 45 G. Socrates, Infrared and Raman Characteristic Group
1
2
19–230.
6 A. Betard and R. A. Fischer, Chem. Rev., 2012, 112, 1055–
083.
Frequencies: Tables and Charts, John Wiley & Sons Ltd,
Baffins Lane, Chichester, 2nd edn, 1994.
46 P. Dorenbos, J. Phys.: Condens. Matter, 2003, 15, 8417–8434.
1
7 S. M. Han, R. A. Ciufo, M. L. Meyerson, B. K. Keitz and 47 S. Benfeito, C. Oliveira, C. Fernandes, F. Cagide, J. Teixeira,
C. B. Mullins, J. Mater. Chem. A, 2019, 7, 19396–19406.
8 Z. L. Liao, T. F. Xia, E. Y. Yu and Y. J. Cui, Crystals, 2018,
R. Amorim, J. Garrido, C. Martins, B. Sarmento, R. Silva,
F. Remião, E. Uriarte, P. J. Oliveira and F. Borges,
Eur. J. Med. Chem., 2019, 167, 525–545.
8(9), 338.
9 J. R. Luo, Y. Li, H. C. Zhang, A. L. Wang, W. S. Lo, Q. Dong, 48 C. Lopez-Chaves, J. Soto-Alvaredo, M. Montes-Bayon,
N. Wong, C. Povinelli, Y. C. Shao, S. Chereddy, S. Wunder,
U. Mohanty, C. K. Tsung and D. W. Wang, Angew. Chem.,
Int. Ed., 2019, 15313–15317.
0 D. B. Dang, P. Y. Wu, C. He, Z. Xie and C. Y. Duan, J. Am.
Chem. Soc., 2010, 132, 14321–14323.
J. Bettmer, J. Llopis and C. Sanchez-Gonzalez, J. Nanomed.
Nanotechnol., 2018, 14, 1–12.
49 A. Kermanizadeh, S. Vranic, S. Boland, K. Moreau,
A. Baeza-Squiban, B. K. Gaiser, L. A. Andrzejczuk and
V. Stone, BMC Nephrol., 2013, 14, 96.
3
3
3
3
3
1 J. Kim, H.-Y. Cho and W.-S. Ahn, Catal. Surv. Asia, 2012, 16, 50 D. Chavarria, C. Fernandes, V. Silva, C. Silva, E. Gil-
1
06–119.
2 A. Dhakshinamoorth, A. M. Asiri and H. Garcia, Adv. Mater.,
019, 31, 1900617.
3 Z. R. Ranjbar and A. Morsali, Inorg. Chim. Acta, 2012, 382,
71–176.
Martins, P. Soares, T. Silva, R. Silva, F. Remião, P. J. Oliveira
and F. Borges, Eur. J. Med. Chem., 2019, 111770.
51 C. Tamames-Tabar, D. Cunha, E. Imbuluzqueta, F. Ragon,
C. Serre, M. J. Blanco-Prieto and P. Horcajada, J. Mater.
Chem. B, 2014, 2, 262–271.
2
1
4 L. Zhang, H. W. Liu, W. Shi and P. Cheng, Coord. Chem. 52 Y. Jin, S. Chen, J. Duan, G. Jia and J. Zhang, J. Inorg.
Rev., 2019, 388, 293–309. Biochem., 2015, 146, 28–36.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020