Robust covalent organic frameworks with tailor-made chelating sites for synergistic capture of U(vi) ions from highly acidic radioactive waste

Article abstract of DOI:10.1039/d1dt00186h

A synergistic strategy for enhancing U(vi) capture under highly acidic conditions (2 M HNO₃) by radiation resistant phosphonate-functionalized two-dimensional covalent organic frameworks with tailor-made binding sites bearing a strong affinity was described. The combination of the radiation resistant characteristic with a strong acid-resistant property endows COFs with practical capabilities for actinide capture from real radioactive liquid waste.

Full text of DOI:10.1039/d1dt00186h

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
Robust covalent organic frameworks with tailor-
made chelating sites for synergistic capture of U(VI)
ions from highly acidic radioactive waste†
Cite this: Dalton Trans., 2021, 50,
3792
Received 19th January 2021,
Accepted 19th February 2021
Jipan Yu,‡a Jianhui Lan,‡a Shuai Wang,a Pengcheng Zhang,a Kang Liu,a
a
Liyong Yuan, *a Zhifang Chaia,b and Weiqun Shi
*
DOI: 10.1039/d1dt00186h
rsc.li/dalton
A synergistic strategy for enhancing U(VI) capture under highly U(VI) from aqueous solution. With increasing pH from 2 to 6,
acidic conditions (2 M HNO3) by radiation resistant phosphonate- the adsorption ability toward U(VI) improved from 60 to
functionalized two-dimensional covalent organic frameworks with 100 mg g−1, whereas the adsorption value fell to only 30 mg
tailor-made binding sites bearing a strong affinity was described. g−1 at pH 1.0. Ion-imprinted PAF6 could achieve nearly com-
The combination of the radiation resistant characteristic with a plete removal of U(VI) at pH 6 while reducing the uranyl ion
strong acid-resistant property endows COFs with practical capa- concentration from the ppm to the ppb level, but because of
bilities for actinide capture from real radioactive liquid waste.
protonation, the uptake decreases considerably at lower pHs.
However, the one thing we should always bear in mind is that
most of the HLLW is a highly acidic atmosphere throughout.
Thus, seeking stable materials with high adsorption capacities,
outstanding selectivity, excellent repeatability, and radiation
stability toward actinides in a highly acidic environment
remains a huge challenge.
The environmental issues arising from the development of
nuclear energy have attracted wide attention due to the high
radioactivity of nuclear waste.1 Uranium (U), as the main fuel
for nuclear reactors and the predominant radiotoxic element
in the spent fuel, possesses a long-term threat to the environ-
ment.2 Conventional separation processes, based on liquid–
liquid extraction, possess common disadvantages in several
aspects such as large amounts of toxic and/or flammable
organic solvent consumption, formation of emulsions, and
production of large volumes of radioactive organic wastes. For
this reason, designing of tailor-made adsorption materials as
an alternative technique has been at the frontier of this field
to overcome these severe shortcomings while holding the
advantages in U separation from high level liquid waste
(HLLW).
Covalent organic frameworks (COFs) are an emerging class
of crystalline porous polymers with periodic structures based
on the precise integration of organic building blocks.7 Because
of their regular pore structure, high surface area and tunable
chemistry, the COF materials are explored in various fields
such as energy storage,8 optoelectronics,9 catalysis,10 and a
number of others.11 Compared with MOFs, the important
characteristics that COFs possess are because they are inge-
niously constructed using organic building blocks containing
light-weight elements via strong covalent bonds, allowing
them to be excellent candidates for enhancing the sorption
capacities when facing harsher conditions. Over the past
decade, a large number of functionalized COFs with structural
diversity of skeletons and pore walls have been synthesized.
However, few COFs have been obtained for addressing radio-
active waste issues. Recently, several groups12–19 have made
outstanding contributions to the removal of UO22+ or TcO4− by
COFs, respectively. In 2019, our group20 had designed functio-
nalized COFs with phosphonate as the side arms located in
Varieties of solid adsorption materials, including functiona-
lized MOFs,3 mesoporous carbon,4 graphene oxide (GO),1,5
and porous aromatic frameworks (PAFs),6 have been reported
to show a high affinity toward uranium. However, these sor-
bents exhibit very poor uptakes in highly acidic medium (pH <
1). For instance, Wang et al.5 reported GOs for the extraction of
aLaboratory of Nuclear Energy Chemistry, Institute of High Energy Physics, Chinese
Academy of Sciences, Beijing 100049, China. E-mail: yuanly@ihep.ac.cn,
shiwq@ihep.ac.cn
bEngineering Laboratory of Advanced Energy Materials, Ningbo Institute of Industrial
Technology, Chinese Academy of Sciences, Ningbo, 315201, China
†Electronic supplementary information (ESI) available: Experimental details,
additional figures, tables, etc. See DOI: 10.1039/d1dt00186h
‡These authors contributed equally to this work.
2+
the channel for adsorption of UO2 and Pu4+ under strong
acid conditions. Even so, irradiation resistant COFs with
strong acid resistance for the removal of actinide ions still
remains a largely unexplored virgin territory.
As our interest in this field increases,20,21 in response to the
demanding requirements for U(VI) removal from highly acidic
3792 | Dalton Trans., 2021, 50, 3792–3796
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Communication
Scheme 1 Synthesis of COF-JLU4, COF-IHEP10 and COF-IHEP11 by condensation reaction.
solutions, we developed a three-component reaction system
consisting of 2,4,6-trihydroxybenzene-1,3,5-tricarbaldehyde
(TBTA) with a mixture of tetraethyl (((2,5-di(hydrazinecarbo-
nyl)-1,4-phenylene)bis(oxy))-bis(ethane-2,1-diyl))bis(phospho-
nate) (TBBP) and 2,5-dimethoxyterephthalohydrazide (DMPA)
at various molar ratios (X = [TBBP]/([TBBP] + [DMPA]) × 100 =
0, 50, and 100) for the synthesis of three COFs with different
phosphonate contents on their edges (Scheme 1). Herein, the
non-phosphonate substituted COF, named COF-JLU4, was pre-
pared by TBTA and DMPA using mesitylene/dioxane as the
solvent in the presence of 6 M acetic acid with subsequent
heating at 120 °C for 7 days according to the literature
report.22 COF-IHPE10 (partially anchoring phosphonates) and
COF-IHEP11 (fully chelating phosphonates) were constructed
by a similar operating procedure. The Fourier transform infra-
red (FT-IR) spectrum of COFs exhibited a peak at ∼1628 cm−1
corresponding to a lower energy characteristic of the carbonyl
stretching vibration in the β-keto-enamine bond, which is
attributed to the 2D π-conjugated sheet and intramolecular
hydrogen bonding. The characteristic stretching vibration
peaks for CvC (1591 cm−1) and N–H (3293 cm−1) confirmed
the occurrence of a condensation reaction (Fig. S1–S3, ESI†).
The high throughput of the phosphonate transformation was
determined by the gradual disappearance of the peaks
assigned to the methyl groups at around 57 ppm in the solid-
state 13C cross-polarization magic-angle spinning (CP/MAS)
Fig. 1 Experimental (red), simulated (blue) and difference plot (black)
PXRD patterns of COF-IHEP10 (a) and COF-IHEP11 (b). Simulated PXRD
patterns with an eclipsed structure (AA stacking mode) for COF-IHEP10
(c) and COF-IHEP11 (d). (C, gray; N, blue; P, purple; O, red. H atoms are
omitted for clarity).
The crystalline structures of COF-IHEP10 and COF-IHEP11
NMR spectrum of COF-IHEP10 and COF-IHEP11 (Fig. S4–S7, were determined by powder X-ray diffraction (PXRD) analysis
ESI†). Meanwhile, the concomitant emergence of strong peaks with Cu Kα radiation in conjunction with the Pawley refine-
at 64.4, 27.2, and 17.1 ppm was ascribed to the phosphonate ment and computational approaches. COF-IHEP10 exhibits a
species. These results confirm the successful construction of strong PXRD peak at 3.60°, a weaker peak at 6.81°, and low-
COF-IHEP10 and COF-IHEP11.
intensity broad features around 15.0° and 26.2°, corres-
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3792–3796 | 3793

View Article Online
Communication
Dalton Transactions
ponding to the (100), (220), (600) and (001) facets, respectively
(Fig. 1a). To elucidate the atomic information of this frame-
work, Pawley refinements of the PXRD patterns were per-
formed in combination with the initially built unit cell from
molecular mechanics calculations. The details are provided in
Fig. S35 and 36.† The simulated structure of COF-IHEP10 exhi-
bits a PXRD pattern fitting well with the experimentally
measured one (Fig. 1c). A hexagonal primitive unit cell (P1)
with the lattice parameters of a = b = 50.03 Å, c = 3.60 Å, α = β =
Fig. 3 (a) Stability of COF-IHEP11 after being immersed in an aqueous
90°, and γ = 120° was deduced with residuals of Rp = 1.35% solution with different acidic environments (pH 1 to 14 and 1 to 3 M
HNO3) for 24 h as demonstrated by PXRD patterns. (b) Radiation stability
of COF-IHEP11 after different doses of gamma irradiation.
and Rwp = 1.86%. COF-IHEP11 has a strong PXRD peak at
3.53°, some weaker peaks at 5.83°, 6.93°, 9.24°, and 12.10° and
low-intensity broad features around 26.6°, corresponding to
the (100), (200), (210), (220) and (001) facets, respectively
(Fig. 1b). The simulated PXRD pattern fits relatively well with characterizations. The PXRD patterns confirmed that
the experimental one compared to COF-IHEP10. The hexag- COF-IHEP11 retained its original crystallinity (Fig. 3b).
onal primitive unit cell (P6) with the lattice parameters of a = b Undoubtedly, the excellent irradiation stability of COFs has
= 28.47 Å, c = 3.60 Å, α = β = 90°, and γ = 120° was obtained two key factors that must be considered to overcome the chal-
with residuals of Rp = 4.42% and Rwp = 3.44%. The complete lenges of the intense irradiation environment: the first is the
structures are given in Fig. S35 and S36† for clarity. The intra-layer infinitely extended π-conjugated system and the
Brunauer–Emmett–Teller (BET) surface areas of COF-IHEP10 second is the adjacent interlayer π–π stacking. This assembling
and COF-IHEP11 were calculated as 386 and 167 m2 g−1
,
architecture could effectively transfer and dissipate the radi-
respectively (Fig. 2a and b). The nonlocal density functional ation energy between the intra-layer and interlayer of COFs,
theory (NLDFT) gave rise to a pore size distribution with an thus avoiding the breaking of the covalent bonds of COFs.
average pore width of ∼2.0 Å and 1.3 Å for COF-IHEP10 and Meanwhile, FT-IR results also suggested an unchanged struc-
COF-IHEP11, respectively (Fig. S8–S10, ESI†), which agrees well ture, which further confirms the excellent radiation stability
with the calculated values based on the eclipsed structure (Fig. S17, ESI†). COF-JLU4 and COF-IHEP10 show similar
(Fig. 1d; Table S1†).
chemical stability and radiation stability to COF-IHEP11 since
To confirm the chemical stability, we dispersed no discriminable changes of their conjugated structure were
COF-IHEP11 in aqueous solution with different acidities observed following irradiation and strong acid treatment
adjusted by HNO3 for 24 h. As shown in Fig. 3a, the PXRD pat- (Fig. S11–16, ESI†). Thermogravimetric analysis indicates that
terns of the immersed materials at the pH range 1–14 and 1 M our COFs are thermally stable up to 250 °C (Fig. S18–20, ESI†).
HNO3 are almost the same as that of the pristine COF sample, In all, these COFs maintain reliable chemical stability and
which indicates that COF-IHEP11 could hold its original excellent radiation resistance, which are the prerequisite for
crystal structure over this acidity range. At 3 M HNO3, the the real-time application potential as the trap material for acti-
proton starts to enter into the interlayer, and the layered struc- nide capture from highly acidic radioactive waste.
ture became swollen. Despite this, the COF-IHEP11 still
After confirming the conjugated structure, porosity with a
retained its original crystallinity partly. In addition, a key con- high density of phosphonate groups as the side arm of
sideration is that, facing high intensity irradiation of the COF-IHEP10 and COF-IHEP11, U(VI) capture by these COFs
highly acidic radioactive waste, the COFs should be robust to was performed in detail to evaluate their capture ability toward
high radiation doses. To further evaluate the irradiation stabi- U(VI) and their potential for HLLW treatment and environ-
lity of the COFs, dry COF-IHEP11 samples as well as the mental remediation. The U(VI) adsorption kinetics of COFs was
samples soaked in water were exposed to a γ-irradiation studied with an initial concentration of 100 ppm at pH 1.0. As
environment using a 60Co source, followed by PXRD and FT-IR shown in Fig. 4a, the concentration of U(VI) in the aqueous
solution decreases rapidly as the contact proceeded. In the
initial 20 min, the uptake of U(VI) is 75 mg g−1 for COF-IHEP10
and 80 mg g−1 for COF-IHEP11. After about 3 h, the adsorption
process reaches an equilibrium with an adsorption capacity of
∼110 mg g−1 for the two COFs. Although the COF-JLU4 shows
a similar kinetic process, the adsorption capacity is less than
70 mg g−1 because of the lack of a chelating site in the skel-
eton of the COFs. In order to clarify the adsorption process of
U(VI), pseudo-first-order and pseudo-second-order models were
applied to analyse the kinetic data (Table S2, ESI†). It is found
that the pseudo-second-order model fits well the experimental
Fig. 2 (a) N2 adsorption and desorption isotherms of COF-IHEP10 at
77 K. (b) N2 adsorption and desorption isotherms of COF-IHEP11 at 77 K.
kinetic data with a much better correlation coefficient, and the
3794 | Dalton Trans., 2021, 50, 3792–3796
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Communication
Cycle performance of COFs is critical for real applications.
After the treatment of U/COF-IHEP11 with sodium carbonate
aqueous solution, it is obvious that COF-IHEP11 can be fully
regenerated. Notably, COF-IHEP11 retained 92% of the orig-
inal capacity even after four cycles (Fig. 4d). This outcome con-
firmed the recycle use of COF-IHEP11 for the facile removal of
U(VI). Selectivity is another important factor for practical U(VI)
removal from water. It is apparent that at pH 1.0, COF-IHEP11
can effectively remove uranyl ions at a capacity of 113 mg g−1
,
but it is not active for other metal ions, such as Zn(II), Co(II), Ni
(II), Cd(II), Sr(II), Yb(III), Sm(III), Nd(III) and La(III), while for
other metal ions it is less than 6 mg g−1 (Fig. 5). Therefore,
COF-IHEP11 is able to remove toxic and radioactive U(VI) ions
in a highly selective manner.
Fig. 4 (a) U(VI) adsorption kinetics of COFs at pH 1.0. (b) U(VI) adsorption
kinetics of COFs (pH 1.0, contact time = 4 h). (c) U(VI) adsorption onto
COFs at a wide range of acidity. (d) Recycle use of COFs for U(VI) uptake
at pH 1.0.
To explore the binding mechanism of uranyl ions, we
selected COF-IHEP11 as a representative model and performed
density functional theory calculations based on the Born–
Oppenheimer approximation. Fig. 6 displays the optimized
favorable binding mode of the uranyl ion with COF-IHEP11.
result obviously indicates that the U(VI) adsorption by the
COFs is dominated by chemical adsorption. These results
show that COFs could capture U(VI) from lower acidic solution
rapidly, and the relatively fast kinetics can meet the require-
ment of spent fuel reprocessing.
To further understand the mode of U(VI) adsorption on the
COFs, Langmuir and Freundlich models were selected to
match the experimental data. The results indicated that the
adsorption isotherm fitted well to the Langmuir model com-
pared to the Freundlich model because of a better correlation
coefficient for phosphonate-decorated COFs (Table S3, ESI†).
The saturated adsorption capacities for U(VI) are 102, 127, and
147 mg g−1 for COF-JLU4, COF-IHEP10 and COF-IHEP11,
respectively (Fig. 4b). To further investigate the influence of
system acidity on the U(VI) adsorption by the COFs, pH values
varying from 1 to 5 and 1 to 3 M HNO3 were tested. As shown
in Fig. 4c, the U(VI) uptake was affected by the solution acidity
during the adsorption. In particular, the U(VI) uptake of
COF-IHEP11 is 203 mg g−1 with a removal percentage of 81.2%
at pH 5.0, which is slightly higher than that of COF-IHEP10.
When the acidity of the solution was increased from pH 5.0 to
3.0, the U(VI) uptake of COF-IHEP11 decreased to 118 mg g−1
due to competitive protonation. At pH 1.0, COF-IHEP11
achieved a U(VI) removal percentage of 45.2%, corresponding
to a capacity of 113 mg g−1. Such a high adsorption capacity
reveals a remarkable attraction of COFs toward U(VI). More sig-
nificantly, when highly acidic media were used, as is typical for
irradiated and high discharge waste liquid treatment and dis-
posal, the U(VI) uptakes of 92 mg g−1 in 1 M HNO3 and 82 mg
g−1 in 2 M HNO3 were associated with COF-IHEP11. In light of
this, the adsorption capability of COF-IHEP11 toward U(VI)
Fig. 5 Selective sorption of U(VI) with COF-IHEP11 from a solution con-
taining metal ions at pH 1.0.
from strongly acidic solution is breath-taking and the value Fig. 6 The favorable binding mode of uranyl with COF-IHEP11 (left
panel) and the charge density distribution within the plane consisting of
the uranium atom and two carbonyl oxygen atoms (right panel) in RGB
mode. The bond lengths between uranium and its surrounding neigh-
bors are displayed in angstroms. The H-bonds are also shown in dotted
clearly meets the demands for spent fuel reprocessing. In the
entire acidity range tested, especially in highly acidic medium,
what is really worth noting is that the U(VI) uptake onto the
three COFs follows the order of COF-IHEP11 > COF-IHEP10 >
COF-JLU4.
blue lines. Gray, blue, red, pink, and yellow spheres represent C, N, O, U,
and P atoms, respectively.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3792–3796 | 3795

View Article Online
Communication
Dalton Transactions
We can see that the uranium atom of the uranyl ion prefers to
be five-coordinated by four oxygen atoms respectively from the
PvO moiety, the backbone, one water molecule, and one
nitrogen atom on the backbone. The calculated bond lengths
range from 2.267 to 2.830 Å, respectively. The relatively long
interaction distance, 2.830 Å, between uranium and water
would be attributed to the steric effect, which inhibits the
closer approaching of water to uranium. In contrast, the carbo-
nyl oxygen atoms play a more important role in attracting the
uranyl ion with much shorter U–O bonds of 2.267 and 2.433 Å.
The charge density distribution within the plane consisting of
the uranium atom and two carbonyl oxygen atoms is also
shown in Fig. 6. Obviously, the electron density overlaps
around the U–O and U–N bonds with the O/N atoms from the
backbone, which indicates the formation of coordination
bonds. Our theoretical calculations suggested that the remark-
able performance of COF-IHEP11 can be attributed to the
3 M. Carboni, C. W. Abney, S. Liu and W. Lin, Chem. Sci.,
2013, 4, 2396–2402.
4 G. Tian, J. X. Geng, Y. D. Jin, C. L. Wang, S. Q. Li, Z. Chen,
H. Wang, Y. S. Zhao and S. J. Li, J. Hazard. Mater., 2011,
190, 442–450.
5 Y. B. Sun, S. B. Yang, Y. Chen, C. C. Ding, W. C. Cheng and
X. K. Wang, Environ. Sci. Technol., 2015, 49, 4255–4262.
6 B. Y. Li, Q. Sun, Y. M. Zhang, C. W. Abney, B. Aguila,
W. B. Lin and S. Q. Ma, ACS Appl. Mater. Interfaces, 2017, 9,
12511–12517.
7 C. S. Diercks and O. M. Yaghi, Science, 2017, 355, 923–930.
8 C. R. DeBlase, K. E. Silberstein, T. T. Truong, H. D. Abruna
and W. R. Dichtel, J. Am. Chem. Soc., 2013, 135, 16821–
16824.
9 M. Dogru, M. Handloser, F. Auras, T. Kunz, D. Medina,
A. Hartschuh, P. Knochel and T. Bein, Angew. Chem., Int.
Ed., 2013, 52, 2920–2924.
synergistic effect of carbonyl functional groups on the COF 10 X. Hang, J. J. Huang, C. Yuan, Y. Liu and Y. Cui, J. Am.
skeleton and phosphonate active sites as side arms in the COF
channels.
In summary, we herein report an operational strategy for
enhancing U(VI) removal from highly acidic radioactive waste
Chem. Soc., 2018, 140, 892–895.
11 Q. R. Fang, J. H. Wang, S. Gu, R. B. Kaspar, Z. B. Zhuang,
J. Zheng, H. X. Guo, S. L. Qiu and Y. S. Yan, J. Am. Chem.
Soc., 2015, 137, 8352–8355.
by
a radiation resistant phosphonate-functionalized COF 12 Y. Li, X. H. Guo, X. F. Li, M. C. Zhang, Z. M. Jia, Y. Deng,
through the synergistic action of the skeleton and side arm.
Considering the acid and radiation-resistant nature of
Y. Tian, S. J. Li and L. J. Ma, Angew. Chem., Int. Ed., 2020,
59, 4168–4175.
COF-IHEP11, it can be a promising sorbent candidate for 13 Q. Sun, B. Aguila, L. D. Earl, C. W. Abney, L. Wojtas,
intense radiation actinide capture from wastewater. Further
P. K. Thallapally and S. Q. Ma, Adv. Mater., 2018, 30,
work to fabricate more effective COF sorbents by varying the
1705479.
skeleton and lateral side techniques, and also to emerge as an 14 L. W. He, S. T. Liu, L. Chen, X. Dai, J. Li, M. X. Zhang,
excellent scaffold for applications in the nuclear fuel cycle, is
in progress.
F. Y. Ma, C. Zhang, Z. X. Yang, R. H. Zhou, Z. F. Chai and
S. Wang, Chem. Sci., 2019, 10, 5183–5184.
15 X. H. Xiong, Z. W. Yu, L. Gong, Y. Tao, Z. Gao, L. Wang,
W. H. Yin, L. Yang and F. Luo, Adv. Sci., 2019, 6, 1900547.
16 W.-R. Cui, C.-R. Zhang, W. Jiang, F.-F. Li, R.-P. Liang, J. Liu
and J.-D. Qiu, Nat. Commun., 2020, 11, 436.
Conflicts of interest
There are no conflicts to declare.
17 Y. Yuan, Q. H. Meng, M. Faheem, Y. J. Yang, Z. N. Li,
Z. Y. Wang, D. Deng, F. X. Sun, H. M. He, Y. H. Huang,
H. Y. Sha and G. S. Zhu, ACS Cent. Sci., 2019, 5, 1432–1439.
18 X. F. Wang, Y. Y. Chen, L. P. Song, Z. Fang, J. Zhang,
F. N. Shi, Y. W. Lin, Y. K. Sun, Y. B. Zhang and J. Rocha,
Angew. Chem., Int. Ed., 2019, 58, 18808–18812.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (No. U2067212, 21806167, 21777161), 19 M. Y. Xu, T. Wang, P. Gao, L. Zhao, L. Zhou and D. B. Hua,
and Youth Innovation Promotion Association of CAS
(21017020).
J. Mater. Chem. A, 2019, 7, 11214–11222.
20 J. P. Yu, L. Y. Yuan, S. Wang, J. H. Lan, L. R. Zheng, C. Xu,
J. Chen, L. Wang, Z. W. Huang, W. Q. Tao, Z. R. Liu,
Z. F. Chai, J. K. Gibson and W. Q. Shi, CCS Chem., 2019, 1,
286–295.
21 Y. Wang, M. S. Xie, J. H. Lan, L. Y. Yuan, J. P. Yu, J. Q. Li,
J. Peng, Z. F. Chai, J. K. Gibson, M. L. Zhai and W. Q. Shi,
Chem., 2020, 6, 2796–2809.
Notes and references
1 Z. Li, F. Chen, L. Yuan, Y. Liu, Y. Zhao, Z. Chai and W. Shi,
Chem. Eng. J., 2012, 210, 539–546.
2 C. Degueldre, J. Bertsch, G. Kuri and M. Martin, Energy 22 Y. W. Zhang, X. C. Shen, X. Feng, H. Xia, Y. Mu and
Environ. Sci., 2011, 4, 1651–1661.
X. M. Liu, Chem. Commun., 2016, 52, 11088–11091.
3796 | Dalton Trans., 2021, 50, 3792–3796
This journal is © The Royal Society of Chemistry 2021