Redox-Active Two-Dimensional Covalent Organic Frameworks (COFs) for Selective Reductive Separation of Valence-Variable, Redox-Sensitive and Long-Lived Radionuclides

Article abstract of DOI:10.1002/anie.201916360

We report the first example of 2D covalent organic framework nanosheets (Redox-COF1) for the selective reduction and in situ loading of valence-variable, redox-sensitive and long-lived radionuclides (abbreviated as VRL nuclides). Compared with sorbents ba

Full text of DOI:10.1002/anie.201916360

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Redox-active Two-dimensional Covalent Organic Frameworks
for Selective Reductive Separation of Valence-variable, Redox-
sensitive and Long-lived Radionuclides
Authors: Yang Li, Xinghua Guo, Xiaofeng Li, Meicheng Zhang, Zhimin
Jia, Yun Deng, Yin Tian, Shoujian Li, and Lijian Ma
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201916360
Angew. Chem. 10.1002/ange.201916360
Link to VoR: http://dx.doi.org/10.1002/anie.201916360
http://dx.doi.org/10.1002/ange.201916360

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Redox-active Two-dimensional Covalent Organic Frameworks for
Selective Reductive Separation of Valence-variable, Redox-
sensitive and Long-lived Radionuclides
Yang Li,[a] Xinghua Guo,[a] Xiaofeng Li,[a] Meicheng Zhang,[a] Zhimin Jia,[a] Yun Deng,[b] Yin Tian*,[b]
Shoujian Li* [a] and Lijian Ma* [a]
Abstract: Selective separation of valence-variable, redox-sensitive
and long-lived radionuclides (abbreviated as VRL nuclides) produced
by nuclear fission chain reaction in the reactor is difficult but important
not only for the sustainable development of nuclear energy but also
for the protection of ecological environment. Previously reported
adsorbents for VRL nuclides were mainly based on chemical
adsorption and/or physical adsorption, and the adsorption selectivity
and capacity for target metal ions need to be further improved. Herein,
we report the first example of 2D covalent organic framework
nanosheets (Redox-COF1) for selective reduction and in situ loading
of VRL nuclides with superior adsorption selectivity and capacity.
Compared with sorbents based on chemical adsorption and physical
adsorption, the redox adsorption mechanism of Redox-COF1 can
effectively reduce the impact of functional group protonation under
usual high-acidity conditions in chemisorption, and raise the
adsorption efficiency from the monotonous capture by pores in
Presently, more than 443 commercial nuclear power plants
in operation worldwide account for 15% of the total global power
generation, while discharging about 10,000 tHM (tons of heavy
metals) of spent fuel each year. Fission products produced by the
nuclear fission chain reaction in the reactor are important
components of spent fuel and include about 300 nuclides of about
40 elements.[1] Some of the elements have several valence-
variable,
redox-sensitive
and
long-lived
radionuclides
(abbreviated as VRL nuclides). The VRL nuclides mainly include
some actinide nuclides, such as uranium (U-238 and U-235),
plutonium (Pu-239 and Pu-240) and neptunium-237, existing
usually in the form of actinyl ions, accounting for ~ 95%, ~ 1% and
~0.1% of the total respectively.[2] The half-lives of the VRL
nuclides are all longer than 104 years. And the dissolving capacity,
migration ability and biological toxicity of them in environmental
media are generally stronger, which could cause potential harm
to earth organisms and human ecological environment. Therefore,
the separation and recovery of these VRL nuclides is one of the
important tasks for spent fuel reprocessing.
Because VRL nuclides have long half-life, dynamic chemical
properties, and they are more easily being bioaccumulated, if the
high-level liquid waste containing these nuclides (except uranium
and plutonium) would be directly treated with virtrification and then
geological disposal, it would bring great environmental risk and
high cost, which would not come up to the pursuit goal of the spent
fuel reprocessing. Conventionally, selective separation of VRL
nuclides from high-level liquid waste would not only require a
number of additional separation processes, but also further
increase the amount of secondary waste. Moreover, the industrial
plutonium uranium redox extraction (PUREX) process currently
used for spent fuel reprocessing still has the following main
problems during the operation: 1) addition of extractant and
stripping agent will generate a large volume of secondary liquid
wastes; 2) The extractant TBP will decompose after strong
irradiation, resulting in decreasing extraction efficiencies or even
a failure of the extractant; 3) as is well known, the TBP molecule
contains phosphorus element, so the spent extractant could not
be completely burned due to the formation of phosphorus oxides,
which would increase the volume of solid waste. These problems
have been research hotspots in the field of spent fuel
reprocessing.[3]
2+
physisorption. Impressively, the adsorption selectivity for UO2
reaches up to unprecedented ~ 97 % at pH = 3, more than 20 percent
of selectivity increment over any analogous adsorbing materials
reported so far. Moreover, the elaborate nanosheet structure of
Redox-COF1 can increase the diffusion rate of the target ions in COF
pores and expose more active redox sites, resulting in decreased
sorption equilibrium time by 4 times and increased sorption capacity
by 2 times compared to its rodlike analogous Redox-COF2. This study
provides an optional approach for preparing high-performance novel
sorbing materials with prospective applications in selective separation
of target metal ions. Besides, by selective reduction and in situ loading
of noble metal ions, this study also provides an extensible way for the
rational design and fabrication of high performance supported
catalysts.
Introduction
[a]
[b]
Dr. Y. Li, Dr. X. Guo, Dr. X. Li, Dr. M. Zhang, Dr. Z. Jia, Prof. S. Li,
Prof. L. Ma
College of Chemistry, Sichuan University, Key Laboratory of
Radiation Physics & Technology, Ministry of Education
No. 29 Wangjiang Road, Chengdu, 610064, P. R. China
E-mail: ma.lj@hotmail.com. (Prof. L. Ma); E-mail:
sjli000616@scu.edu.cn. (Prof. S. Li)
Thus, our solution for the problems is to separate, at first, the
VRL nuclides from fuel reprocessing feed solutions and then
transfer them to the TBP extraction process. Meanwhile, other
fission products remaining in the feed liquid can be directly treated
and disposed of as high level radioactive wastes. This can
effectively lighten the burden of the PUREX process, simplify the
Prof. Y. Deng, Prof. Y. Tian
School of Pharmacy, Chengdu University of Traditional Chinese
Medicine
Chengdu 611137, P.R. China
E-mail: ytian227@outlook.com. (Prof. Y. Tian)
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
subsequent separation system for the VRL nuclides, and can also
greatly reduce the radiation dose and lower volume of secondary
wastes to obtain a larger volume reduction ratio. To achieve this
tentative idea, solid phase extraction (SPE) with its own distinctive
characteristics of flexible operation, high separation factor, high
volume reduction ratio, basically using no organic solvents, etc. is
considered as a practical candidate auxiliary method.[4]
In principle, solid phase extraction as an auxiliary method for
enhanced efficiency of the PUREX process should be in line with
the goal pursued by the aqueous reprocessing process of spent
nuclear fuel. Unfortunately, the interactions between solid phase
extractants (hydrothermal carbon,[5] porous organic polymers,[6]
metal-organic frameworks (MOFs),[7] porous silica,[8] graphene,[9]
etc.) and metal ions reported so far are mostly chemical
adsorption (coordination/complexation by functional groups) and
adsorbent. The reduction of high-valent metal ions into low-valent
metal oxides can not only be conducive to increase the selectivity
of the material to target VRL ions and facilitate subsequent solid-
liquid separation, but also effectively reduce the adverse impact
of acidity environment on the material's efficiency of nuclide
separation. In addition, the redox solid phase extractant designed
should be free of metal ions. That is, the SPE adsorbent should
be preferably composed of completely combustible elements
such as C/N/O. In this way, the volume of spent sorbent would be
greatly reduced through incineration, which thereby could
effectively reduce the burden of deep geological disposal of solid
waste.[11]
Up to now, the common redox-type SPE sorbents reported
include mainly nano-zero-valent iron[12] and its composites, metal-
containing quantum dots[13] and MOFs.[14] However, these
sorbents are usually not stable enough, especially under acidic
conditions, the metal ions will be dissolved out of the sorbents,
resulting in the generation of secondary pollution.
Covalent organic frameworks (COFs) are a class of porous
crystalline polymeric materials mainly composed of light elements
such as C/H/O/N. This type of materials has been regarded to
possess its inherent characteristics such as rich structural units,
easy to functionalize with different groups, variety of chemical and
topological connections, relatively high physical and chemical
stability and high specific surface area.[10a, 15] Although there are
some examples of applying COF materials to the separation or
recovery of metal ions, as with other types of SPE sorbents, the
current reported interaction between COF-based sorbents and
metal ions is dominantly based on the above mentioned physical
adsorption and functional group complexation.[10a, 15]
physical adsorption (trapping via pores/channels, or van der
7c, 7f, 10]
Waals adsorption) or both.[5a,
Chemical adsorption is
mainly based on coordination interaction between electron-rich
(donor) atoms, such as O/N/S, and electron-deficient metal
cations, but at higher acidity (usually pH<2), the separation effect
will be lost due to the protonation of ligands.[7f, 10b] Meanwhile, the
selectivity of adsorbing materials to the target nuclide ion is also
easily affected by the charge of the target ion and the competition
of coexisting metal ions. Physical adsorption conventionally has
no specific selectivity for metal ions, and it is also very difficult to
separate a specific metal ion with high selectivity by virtue of the
size screening effect of the pores in materials. Furthermore, the
intragranular diffusion effect in the porous material will also
significantly reduce the adsorption rate.[10a] Thus, to date, no SPE
material has been used to improve the separation efficiency of the
PUREX process.
From the above, we synthesized a novel 2D redox-active
COF material using monomers containing hydroquinone (2,5-
dihydroxyterephthalaldehyde, DHPA) and hydrazide (benzene-
1,3,5-tricarbohydrazide, BTCH) components respectively, which
In this study, we tried to design a new solid phase extractant
based on redox reaction, hoping to effectively separate all/most
of the important VRL nuclides from the feed solution originating
from dissolved spent fuel and load them in situ on the SPE
Figure 1. Structures of monomers and the obtained Redox-COF1 as well as the synthetic conditions of the aqueous acetic acid-CH2Cl2 biphasic system.
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 2. SEM images of Redox-COF1 obtained at different reaction time.
are common organic reductant,[16] and explored the reduction-
loading ability of this material for VRL nuclides. The structures of
the monomers selected and the resulting COF are shown in
Figure 1. The product is named Redox-COF1.
obtained at different growth periods (Figure 2) reveals that the
growth process of Redox-COF1 may firstly generate a network
structure in which one-dimensional COF nanowires are cross-
connected with each other. As the time went on, the network
structure gradually grew denser, and finally formed a smooth COF
film after 7 days. The magnification of the upper row in Figure 2 is
5000 times while the lower row is 30000 times. Redox-COF1
obtained at different reaction time were also characterized by
TEM after ultrasonic treatment. The results are shown in Figure
3a-d. It can be found that, when the reaction time is 2 days, the
nanowires will be broken into quantum dots with the size of about
5 nm (Figure 3a). As the reaction time was extended to 5 days,
both COF quantum dots and nanosheets could be observed in the
TEM image (Figure 3b). When the reaction was further extended
to 7 days, only large COF nanosheets could be observed (Figure
3c). High-resolution TEM pattern of Redox-COF1 in Figure 3d
demonstrates the well-ordered stacking of the layers, which,
combined with the selected-area electron diffraction, proves that
Redox-COF1 has a good crystallinity. The above findings may
also provide a simple way for the controllable preparation of COF
from quantum dots to nanosheets. The FT-IR spectra showed that,
after the reaction, the characteristic absorption peak of the
aldehyde group disappeared at 2892 cm-1, and correspondingly,
the vibration absorption peak of C=N appeared at 1550 cm-1 and
1247 cm-1 (Figure S3). Meanwhile, the amide-oxygen C=O double
bond still exists at 1662 cm-1, which preliminarily indicates that the
synthesis is successful.[19] We then characterized the framework
structure of Redox-COF1 by 13C solid-state NMR spectroscopy
and XPS. From the magic-angle rotation 13C NMR spectrum, it
can be seen that the amide (~ 165 ppm) and the imine (~ 160
ppm) are present in the structure of Redox-COF1. And the signal
located at ∼ 150 ppm is ascribed to the C-O unit (Figure 3e and
Results and Discussion
Presently, the pore sizes of most two-dimensional COFs
reported are in the range of 1-3 nm,[17] while the size of hydrated
metal ions is about 0.6-1.1 nm.[18] Considering the size difference,
we think that the metal ions adsorbed by the functional groups on
the pore wall will affect the diffusion of metal ions followed, leading
to problems such as slow mass transfer rate and inadequate
exposure of active functional sites during the adsorption process,
which will affect the sorption efficiency of sorbents for target metal
ions. It is worth trying to make the COF sizes smaller
(nanoparticles/nanosheets) to shorten the ion diffusion paths and
increase the amount of functional groups exposed. So Redox-
COF1 is prepared by an aqueous acetic acid-CH2Cl2 biphasic
system. Unlike common solvothermal methods, the morphology
of COFs prepared by the interfacial method is usually 2D
nanosheets, which is beneficial to the diffusion of metal ions and
increasing the adsorption performance of COFs. After a certain
period of time, a yellow COF layer was gradually formed at the
interface, and the newly formed thin COF layer emitted red
fluorescence under UV lamp irradiation (Figure S1). For batch
production of Redox-COF1, a 250 mL beaker was used as the
reaction vessel, the preparation process and results were shown
in the SI (Figure S2 and Section 5).
The growth of Redox-COF1 at the acetic acid-CH2Cl2
interface is a slow process. SEM characterization of samples
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. TEM images of Redox-COF1 obtained at the time of 2 days (a), 5 days (b) and 7 days (c) after ultrasonic treatment; (d) HR-TEM image and
selected-area electron diffraction pattern of Redox-COF1 obtained at the time of 7 days; (e-i) Structural characterization of Redox-COF1. (e) 13C solid-state
NMR spectroscopy. The typical XPS survey spectra (f) and high resolution XPS spectra of O1s (g), N1s (h) and C1s (i); (j-l) Comparison of experimental and
simulated PXRD patterns of Redox-COF1 (j) and the top (k) and side view (l) of AA-stacked Redox-COF1.
Figure S4).[20] The typical XPS survey spectra show that Redox-
COF1 is mainly composed of C/N/O (Figure 3f). In O1s signal
(Figure 3g), there are two types of oxygen, which respectively are
C-O (532.6 eV) and C=O (531.2 eV).The XPS peak of N1s (Figure
3h) indicates that the product contains two types of N, which could
Redox-COF1 with an average d100 = 2.06 nm was directly
observed under high resolution mode (Figure S6). The d100 value
given by TEM is similar to the pore diameter of AA stacking mode
measured in MS, which further indicates that Redox-COF1 is AA-
stacked. The results of N2 adsorption isotherm of Redox-COF1 is
be ascribed to amide (400.4 eV) and imine (399.8 eV) respectively. shown in Figure S7. From the aperture distribution given by BJH
In contrast, the C1s peak (Figure 3i) is quite complex and can be
roughly divided into three categories: C=O and C=N (288.2 eV),
C-O and C-N (285.8 eV) and C=C (284.8 eV).[21] Besides, the
experimental and theoretical values obtained from elemental
analysis (Table S2) are also well matched. PXRD patterns of
Redox-COF1, DHPA and BTCH are shown in Figure S5. The
complete disappearance of the characteristic peaks of DHPA and
BTCH in the PXRD pattern of Redox-COF1 reveals the thorough
consumption of the monomers. The crystal structure of Redox-
COF1 is obtained by Materials Studio (MS) simulation (Figure 3j-
l). As shown in Figure 3j, the experimental PXRD pattern is more
similar to the calculated value of the AA stacking mode. The pore
size and interlayer spacing of AA-stacking structure is about 2.3
nm (Figure 3k) and 0.34 nm (Figure 3l) respectively. In addition,
in order to further verify that Redox-COF1 is in AA-stacking mode,
we characterized Redox-COF1 with TEM again and hoped to
obtain more structural information. Fortunately, the (100) plane of
desorption, the pore size distribution of Redox-COF1 exhibited
maxima of 2.3 nm, which is also in line with the predicted value of
AA eclipsed geometries of the frameworks. Thermogravimetric
experiment under nitrogen atmosphere showed that Redox-COF1
was basically stable within 350 °C (Figure S8). And the infrared
spectra of the samples were basically unchanged after soaked in
dilute nitric acid of pH 1-7 for 3 days (Figure S9), indicating that
Redox-COF1 had good acid stability.
Redox-COF1 was then used for selective removal and
response of uranyl ions. The as-synthesized Redox-COF1 is
yellow powder (Figure 4a), which turns brown after adsorption of
uranyl ions (Figure 4b). The color changes of COF dispersions
under sunlight and ultraviolet lamp before and after UO22+ sorption
were shown in Figure 4c and Figure 4d. Generally speaking, the
color change of sorbents are mostly related to the color of metal
salts after sorption. The yellow Redox-COF1 abnormally turns
brown when it absorbs UO22+, which is distinctly different from that
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
2+
2+
Figure 4. (a-d) Color changes of Redox-COF1 before and after UO2 sorption; (e-f) Binding energies of O1s before (e) and after (f) UO2 sorption; (g)
Changes in XPS binding energy of U after uranyl sorption by Redox-COF1 and Reference-COF; (h) EPR signal of Redox-COF-U.
of uranyl nitrate. This anomalous color change indicates that
uranyl ions may oxidize the hydroquinone structure to
benzoquinone during adsorption. To verify this conjecture, we
firstly estimated the initial oxidation potential of Redox-COF1 by
cyclic voltammetry. The results are shown in Figure S10. As we
can see, the onset oxidation potential of Redox-COF1 was about
-0.78 V, which is much lower than the reduction potential of
UO22+/UO2 (0.411 V vs. SHE). The result indicates that the
photos before and after the adsorption of uranyl ions by
Reference-COF is shown in Figure S14. As we can see, the color
of Reference-COF dispersions does not change under sunlight
and UV lamp after sorption, which indicates that the interaction
between Reference-COF and uranyl is much lower than that of
Redox-COF1. For convenience, the samples of Redox-COF1 and
Reference-COF after uranium loading were named as Redox-
COF-U and Reference-COF-U, respectively. From the XPS
reduction of UO22+ by Redox-COF1 is thermodynamically feasible. spectra of uranium, it can be found that the binding energy of
We then employed XPS to characterize the changes of COF
Redox-COF-U significantly shifted to lower binding energies,
which indicates that U(VI) is reduced to U(IV) during the
adsorption process (Figure 4g). In addition, uranyl ion is
diamagnetic and has no electron paramagnetic resonance (EPR)
signal. However, Redox-COF-U has an EPR signal with g = 1.99,
which is consistent with the g value of uranium dioxide (Figure 4h
and Figure S15).[22] To verify whether the reduced UO2 is
monodisperse or aggregated, Redox-COF1 samples with
different U loadings were prepared and characterized by TEM and
XRD. From the results shown in Figure S16-18, it can be seen
that the electrically neutral UO2 is easily aggregated into
nanoparticles for the reason that the coordination ability of
2+
samples before and after UO2 adsorption, the results were
shown in Figure 4e-f and Figure S11. From O1s (Figure 4e and f)
and C1s (Figure S11a and b) binding energy in XPS spectra, it
can be seen clearly that after sorption of uranyl ions, the
proportion of C-O decreases, while that of C=O increases, and
the ratio of C=O to C-O changed from 1:1 to 2:1 after sorption by
semi-quantitative analysis of O1s XPS spectrum. The apparent
change in the ratio of C=O to C-O strongly confirms the
conversion of hydroquinone structure to benzoquinone structure.
In order to obtain stronger experimental evidence for the reduction
of uranyl ions by Redox-COF1, we then designed and
synthesized
a
COF with similar structure but without
electrically neutral UO2 with functional groups is greatly weakened
2+
hydroquinone component and named it as Reference-COF. It
should be noted that the adsorption of uranyl by COF of this non-
redox functional component does not involve redox process. The
structure of Reference-COF is shown in Figure S12. PXRD
analysis showed that Reference-COF is in AA-stacking mode.
And the results of XPS and elemental analysis were consistent
with the structure of Reference-COF (Figure S13). The digital
compared with that of the charged UO2 .
In order to gain deeper insights into this in situ redox
adsorption mechanism, density functional theory (DFT) was
utilised to explore and analyze this reaction progress. The species
involved in the calculation and their relative free energy profiles
are shown in Figure 5a and Figure S19-20. As we can see, the
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 5. (a) The species involved in the calculation and the relative free energy profiles (in kcal/mol) for the complexing and redox reaction between Redox-
COF1 and uranyl ions obtained by the DFT calculations; (b) The optimized structures of the stationary points for adsorption structure of the complexing state
and the reduction state. Furthermore, the spin density of uranium atom and important bond lengths (angstroms) are also shown in this figure.
relative free energy profiles of Redox-COF1 decreased from 0
kcal/mol to -15.6 kcal/mol after the removal of two protons and
energy of the system reduces by 23.4 and 138.3 kcal/mol with the
electrons of N2 atom in the O3=C8-N2H-N1=C7-structure flowing to
the π* orbital of C7=N1 and n*-orbital of C8. Therefore, due to the
presence of N2 atoms, the system tends to form a large
conjugated structure. On the one hand, the large conjugated
structure greatly reduces the energy of the system, and helps to
effectively disperse the charge during the coordinated hydrogen
dissociation process, so that the protons are more easily to be
removed (Table S5 and Figure S23); and in the electron transfer
reduction process, when electrons are lost, the large conjugate
system is also easier to be stabilized. In addition, the presence of
N2 atom also leads to an increase in the negative charge of N1
and O3 atoms in the C7=N1 and C8=O3 structures, enhancing their
coordination ability with UO22+, which facilitates the transfer of
electrons from oxygen atoms to uranium.
2+
coordination with UO2 to form the complex state (Figure 5a).
Next, two electrons of complex state transfer from oxygen to
uranium to form reduction state (Figure 5a), the process would
further lower the relative free energy from -15.6 kcal/mol to -23.8
kcal/mol, indicating that Redox-COF1 could be spontaneously
oxidized by UO22+. The optimized structures of complex state and
reduction state are shown in Figure 5b. It can be seen, the spin
density of U changed from 0 to 1.114, which corresponds to 5f-
orbital occupancy (the lower right corner of Figure 5b), and
suggests that electrons transfer from O atom to U5f orbitals. The
spin density result of U atom shows good agreement with the
electron paramagnetic resonance signal in Figure 4h. After
electron transfer, bond lengths of C1-C2, C3-C4, C4-C5 and C1-C6
become longer, while that of C2-C3, C5-C6, C1-O1 and C4-O2
become shorter. The results are well consistent with the fact that
hydroquinone is oxidized by uranyl ions to form benzoquinone
structure. In addition, the same conclusion can be drawn from
Wiberg bond indices (WBI) of the selectedchemical bonds
obtained by natural bond orbital (NBO) analysis (Table S3). As
For comparison, we also calculated the adsorption process
of Reference-COF by DFT. Density functional theory calculation
reveals that, after sorption, the Gibbs free energy of Reference-
COF decreases by 6.2 kcal/mol in aqueous solution (Figure S24).
However, after adsorption of uranyl ion by Redox-COF1, the
Gibbs free energy decreased by 23.8 kcal/mol, which was much
larger than that of Reference-COF (Figure 5a). This also indicates
that the interaction between Reference-COF and uranyl ions is far
less strong than that of Redox-COF1. More detailed DFT
calculation results can also be found in Section 7 of SI.
we should point out, the hydroquinone and hydrazide monomers
2+
can’t be spontaneously oxidized by UO2
in the control
experiments (Figure S21), then why does the redox reactions
happen when they reacted to form Redox-COF1? Through
theoretical calculations, we believe the electron delocalization
effect of N2 atom in the O3=C8-N2H-N1=C7-structure may play a
key role (Table S4 and Figure S22), in which the negative charge
is transferred from the nitrogen lone pair to the π*-orbital of C7=N1
and n*-orbital of C8. The charge transfer can be evaluated by the
second-order interaction energy between donor and acceptor
orbitals, the results are listed in Table S4. As we can see, the total
As mentioned above, we hope to design a novel redox-active
COF for the purpose of selective reduction and in-situ loading of
VRL nuclides. So firstly, we chose several monovalent, divalent,
and trivalent nuclides which are commonly used in the nuclear
industry as competitive ions to test the selective uranium
(accounting for ~ 95% of the total) separation ability of Redox-
COF1. It can be seen from Figure 6a that the color change of
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 6. (a) Color changes of Redox-COF1 dispersions after adding different metal ions; (b) Effect of pH on the adsorption of metal ions in a multi-ion system
2+
n+
and (c) effect of pH on the adsorption selectivity of UO2 in a multi-ion system (CM ≈ 1.0 mmol/L, t = 120 min, v = 25 mL, T =298 K，w = 10 mg );(d)
Comparison of adsorption performance of Redox-COF1 and Reference-COF in multi-ion-mixed systems (CMn+ ≈ 1.0 mmol/L, t = 120 min, v = 25 mL, T =298
K, w = 10 mg, pH = 4.5).
uranyl ions is most obvious when adding the same amount of
metal nitrate.
shown in Figure 6d, Redox-COF1 has a total sorbing capacity of
1.04 mmol/g for all metal ions, which is 2.1 times higher than that
of Reference-COF. Particularly, the sorbing capacity for target
uranyl ion of Redox-COF1 (0.991 mmol/g) is higher than that of
Reference-COF (0.065 mmol/g) by 15 times. Moreover, the
uranium selectivity of Redox-COF1 is ~ 96%, which is far superior
to that (~13%) of Reference-COF. The experimental results
strongly confirm that the redox-active sorbing materials can make
full use of the redox characteristics of the target element for the
purpose of selective reduction and in situ loading of VRL nuclides.
We also prepared Redox-COF materials by solvothermal
method using the same monomers (See SI, Section 5). The
resulting product was named Redox-COF2. PXRD pattern
demonstrates that Redox-COF2 is crystalline (Figure S26). SEM
characterization shows that Redox-COF2 has rod-like structures
(Figure S27). Batch adsorption experiments showed that Redox-
COF1 has higher sorbing capacity, faster sorbing rate (Figure
S28) and better uranium sorbing selectivity (Figure S29) than that
of Redox-COF2. The experimental results are consistent with our
initial design that nanosheet morphology is beneficial to
increasing the adsorption ability.
Subsequently, we also tested the interaction of Redox-COF1
with several common cations and oxyacid anions with higher
oxidation potentials. The results are shown in Figure S25. As we
can see, Redox-COF1 has a macroscopic response to all of these
oxidized ions. The batch adsorption experiment results (Figure
6b) showed that, even at pH = 2, the adsorption capacity (qe, see
the SI, Section S8) of Redox-COF1 for uranyl ions could reach up
to 60 mg/g. Generally speaking, due to the protonation of
functional groups, most of sorbents based on functional group
[5]
coordination will lose their adsorption capacity at pH ≤ 2. This
indicates that adsorbent based on redox sorption mechanism can
alleviate the adverse effect of protonation on adsorption to a
certain extent. It should be noted that the concentration of each
metal ion in the multi-ion-mixed solution used in batch adsorption
experiments was approximately the same (CMn+ ≈ 1.0 mmol/L) in
order to fairly compare the selectivity of the adsorbent material to
various ions. As we can see, the sorbing capacity of uranyl ions
is much higher than that of the other coexisting 11 competing ions
at pH 2-4.5. And the uranium sorbing selectivity (Figure 6c) is
about 90% (SU, SI, Section 8) at all pH values selected. Especially,
the SU could reach up to ~ 97% at pH around 3.0, which is
significantly better than that of previously reported uranium
sorbing materials (Table S7). We then compared the adsorption
performance of Redox-COF1 and Reference-COF at pH = 4.5. As
Conclusion
In summary, we have prepared a novel nanosheet-shape
Redox-COF1 for selective reductive separation of VRL
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
198-213; d) L. A. Berrueta, B. Gallo, F. Vicente, Chromatographia 1995, 40,
474-483.
radionuclides by acetic acid-organic solvent interface method.
Study of the growth process revealed that the resulting 2D
nanosheets grow from intertwined nanowires gradually to flat
nanosheets. After ultrasonic treatment of the products in different
growth stages, COF quantum dots, nanosheets or their mixtures
can be obtained optionally, which provides a valuable avenue for
the controllable synthesis of COF quantum dots and nanosheets.
The results of batch adsorption experiments show that the
nanosheet-structured COF can increase the diffusion rate of the
target ions in COF pores and expose more functional sites, which
is beneficial to increasing the adsorption efficient of the sorbents.
In addition, the redox adsorption mechanism can effectively
[5] a) H. Wang, L. Ma, K. Cao, J. Geng, J. Liu, Q. Song, X. Yang, S. Li, J.
Hazard. Mater. 2012, 229, 321-330; b) B. Li, L. Ma, Y. Tian, X. Yang, J. Li,
C. Bai, X. Yang, S. Zhang, S. Li, Y. Jin, J. Hazard. Mater. 2014, 271, 41-
49.
[6] a) Q. Sun, B. Aguila, J. Perman, A. S. Ivanov, V. S. Bryantsev, L. D. Earl,
C. W. Abney, L. Wojtas, S. Ma, Nat. Commun. 2018, 9, 1644; b) B. Li, Q.
Sun, Y. Zhang, C. W. Abney, B. Aguila, W. Lin, S. Ma, ACS Appl. Mater.
Interfaces 2017, 9, 12511-12517.
[7] a) W. Yang, Q. Pan, S. Song, H. Zhang, Inorg. Chem. Front. 2019, 6, 1924-
1937; b) C. Xiao, M. A. Silver, S. Wang, Dalton Trans. 2017, 46, 16381-
16386; c) J. Li, X. Wang, G. Zhao, C. Chen, Z. Chai, A. Alsaedi, T. Hayat,
X. Wang, Chem. Soc. Rev. 2018; d) M. Feng, P. Zhang, H. C. Zhou, V. K.
Sharma, Chemosphere 2018, 209, 783-800; e) L. Chen, Z. Bai, L. Zhu, L.
Zhang, Y. Cai, Y. Li, W. Liu, Y. Wang, L. Chen, J. Diwu, J. Wang, Z. Chai,
S. Wang, ACS Appl. Mater. Interfaces 2017, 9, 32446-32451; f) C. W.
Abney, R. T. Mayes, T. Saito, S. Dai, Chem. Rev. 2017, 117, 13935-14013.
[8] a) P. J. Lebed, J.-D. Savoie, J. Florek, F. Bilodeau, D. Larivière, F. Kleitz,
Chem. Mater. 2012, 24, 4166-4176; b) C. Gunathilake, J. Górka, S. Dai, M.
Jaroniec, J. Mater. Chem. A 2015, 3, 11650-11659.
alleviate the protonation of functional groups under high acidity
2+
conditions and greatly improve the adsorption selectivity of UO2
.
This manuscript provides an optional approach for preparing
novel high-performance sorbing materials with prospective
applications in selective separation of target metal ions. Besides,
by selective reduction and in situ loading of noble metal ions, the
methods can also be extended to provide an extensible way for
the rational design and fabrication of high performance supported
catalysts.
[9] Y. Tian, J. Fu, Y. Zhang, K. Cao, C. Bai, D. Wang, S. Li, Y. Xue, L. Ma, C.
Zheng, Phys. Chem. Chem. Phys. 2015, 17, 7214-7223.
[10] a) S. Zhang, X. Zhao, B. Li, C. Bai, Y. Li, L. Wang, R. Wen, M. Zhang, L.
Ma, S. Li, J. Hazard. Mater. 2016, 314, 95-104; b) C. Bai, M. Zhang, B. Li,
Y. Tian, S. Zhang, X. Zhao, Y. Li, L. Wang, L. Ma, S. Li, J. Hazard. Mater.
2015, 300, 368-377.
Experimental Section
[11] a) T. Tsujino, T. Aochi, T. Hoshino, J. Nucl. Sci. Technol. 1976, 13, 321-
326; b) M. Nilsson, S. Andersson, F. Drouet, C. Ekberg, M. Foreman, M.
Hudson, J. O. Liljenzin, D. Magnusson, G. Skarnemark, Solvent Extr. Ion
Exch. 2006, 24, 299-318; c) C. Musikas, Sep. Sci. Technol. 2006, 23, 1211-
1226; d) C. Musikas, Miner. Process. Extr. Metall. Rev. 1997, 17, 109-142.
[12] a) S. Tsarev, R. N. Collins, E. S. Ilton, A. Fahy, T. D. Waite, Environ. Sci.
Nano 2017, 4, 1304-1313; b) M. Qiu, M. Wang, Q. Zhao, B. Hu, Y. Zhu,
Chemosphere 2018, 201, 764-771; c) L. Ling, W. X. Zhang, J. Am. Chem.
Soc. 2015, 137, 2788-2791; d) L. Chen, S. Feng, D. Zhao, S. Chen, F. Li,
C. Chen, J. Colloid Interface Sci. 2017, 490, 197-206.
Preparation of Redox-COF1. Firstly, 25 mg BTCH was dissolved in 50
mL 6M aqueous acetic acid to give the amine solution (A); and aldehyde
solution (B) was prepared by dissolving 25 mg DHPA in 50 mL CH2Cl2.
Next, the solution A was added dropwise to the surface of the solution B.
After a period of time, the product was collected and washed thoroughly
with dichloromethane, ethanol, acetone and DMF in turn, and put aside for
further investigation. The obtained product was named Redox-COF1.
[13] a) W. Li, L. D. Troyer, S. S. Lee, J. Wu, C. Kim, B. J. Lafferty, J. G. Catalano,
J. D. Fortner, ACS Appl. Mater. Interfaces 2017, 9, 13163-13172; b) S. S.
Lee, W. Li, C. Kim, M. Cho, B. J. Lafferty, J. D. Fortner, J. Mater. Chem. A
2015, 3, 21930-21939; c) S. S. Lee, W. Li, C. Kim, M. Cho, J. G. Catalano,
B. J. Lafferty, P. Decuzzi, J. D. Fortner, Environ. Sci. Nano 2015, 2, 500-
508.
Acknowledgements
The financial support from Science Challenge Project TZ2016004
and the National Natural Science Foundation of China (Grants
21771128, 21976125 and 21671140
acknowledged.
)
are gratefully
[14] a) H. Zhang, W. Liu, A. Li, D. Zhang, X. Li, F. Zhai, L. Chen, L. Chen, Y.
Wang, S. Wang, Angew. Chem., Int. Ed. 2019, 58, 16110-16114; b) L. Xu,
D. Zhang, F. Ma, J. Zhang, A. Khayambashi, Y. Cai, L. Chen, C. Xiao, S.
Wang, ACS Appl. Mater. Interfaces 2019, 11, 21619-21626.
Keywords: covalent organic frameworks • nuclear energy •
uranium • sorption • redox
[15] a) Q. Sun, B. Aguila, J. Perman, L. D. Earl, C. W. Abney, Y. Cheng, H. Wei,
N. Nguyen, L. Wojtas, S. Ma, J. Am. Chem. Soc. 2017, 139, 2786-2793; b)
Q. Sun, B. Aguila, L. D. Earl, C. W. Abney, L. Wojtas, P. K. Thallapally, S.
Ma, Adv. Mater. 2018, 30, 1705479; c) L. Merí-Bofí, S. Royuela, F. Zamora,
M. L. Ruiz-González, J. L. Segura, R. Muñoz-Olivas, M. J. Mancheño, J.
Mater. Chem. A 2017, 5, 17973-17981; d) Z. Li, Y. Zhang, H. Xia, Y. Mu,
X. Liu, Chem. Commun. 2016, 52, 6613-6616; e) Y. Li, M. Zhang, X. Guo,
R. Wen, X. Li, X. Li, S. Li, L. Ma, Nanoscale Horiz. 2018, 3, 205-212; f) N.
Huang, L. Zhai, H. Xu, D. Jiang, J. Am. Chem. Soc. 2017, 139, 2428-2434;
g) S. Y. Ding, M. Dong, Y. W. Wang, Y. T. Chen, H. Z. Wang, C. Y. Su, W.
Wang, J. Am. Chem. Soc. 2016, 138, 3031-3037.
[1] a) K. Maher, J. R. Bargar, G. E. Brown, Jr., Inorg. Chem. 2013, 52, 3510-
3532; b) H. Kleykamp, Nucl. Technol. 2017, 80, 412-422; c) S. Demir, N. K.
Brune, J. F. Van Humbeck, J. A. Mason, T. V. Plakhova, S. Wang, G. Tian,
S. G. Minasian, T. Tyliszczak, T. Yaita, T. Kobayashi, S. N. Kalmykov, H.
Shiwaku, D. K. Shuh, J. R. Long, ACS Cent. Sci. 2016, 2, 253-265.
[2] R. C. Ewing, Nat. Mater. 2015, 14, 252-257.
[3] a) K. J. Parikh, P. N. Pathak, S. K. Misra, S. C. Tripathi, A. Dakshinamoorthy,
V. K. Manchanda, Solvent Extr. Ion Exch. 2009, 27, 244-257; b) A. P. Paiva,
P. Malik, J. Radioanal. Nucl. Chem. 2004, 261, 485-496; c) M. Muthukumar,
S. Kumar, P. K. Sinha, U. K. Mudali, R. Natarajan, J. Radioanal. Nucl.
Chem. 2011, 288, 819-821; d) J. M. McKibben, Radiochim. Acta 1984, 36,
3-15.
[16] a) R. J. Taylor, I. May, V. S. Koltunov, S. M. Baranov, V. I. Marchenko, E.
A. Mezhov, V. G. Pastuschak, G. I. Zhuravleva, O. A. Savilova, Radiochim.
Acta 1998, 81, 149-156; b) D. R. Prabhu, N. Kumari, A. S. Kanekar, P. N.
Pathak, J. Radioanal. Nucl. Chem. 2013, 298, 691-698; c) P. G. Kulkarani,
K. K. Gupta, P. N. Pathak, D. R. Prabhu, A. S. Kanekar, D. K. Pant, G. A.
Chaugule, P. B. Gurba, V. K. Manchanda, Desalin. Water Treat. 2012, 12,
57-61.
[4] a) C. S. K. Raju, M. S. Subramanian, Sep. Purif. Technol. 2007, 55, 16-22;
b) S. Kesava Raju Ch, M. S. Subramanian, J. Hazard. Mater. 2007, 145,
315-322; c) B. Buszewski, M. Szultka, Crit. Rev. Anal. Chem. 2012, 42,
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
[17] a) J. L. Segura, M. J. Mancheno, F. Zamora, Chem. Soc. Rev. 2016, 45,
5635-5671; b) M. S. Lohse, T. Bein, Adv. Funct. Mater. 2018, 28, 1705553.
[18] L. Wang, X. Guo, K. Cao, B. Li, Y. Li, M. Zhang, R. Wen, X. Li, S. Li, L. Ma,
J. Mater. Chem. A 2017, 5, 8051-8061.
[19] a) F. J. Uribe-Romo, C. J. Doonan, H. Furukawa, K. Oisaki, O. M. Yaghi, J.
Am. Chem. Soc. 2011, 133, 11478-11481; b) L. Stegbauer, K.
Schwinghammer, B. V. Lotsch, Chem. Sci. 2014, 5, 2789-2793; c) W. Liu,
Q. Su, P. Ju, B. Guo, H. Zhou, G. Li, Q. Wu, ChemSusChem 2017, 10, 664-
669; d) D. N. Bunck, W. R. Dichtel, J. Am. Chem. Soc. 2013, 135, 14952-
14955.
[20] a) H. L. Qian, C. Dai, C. X. Yang, X. P. Yan, ACS Appl. Mater. Interfaces
2017, 9, 24999-25005; b) S. Kandambeth, D. B. Shinde, M. K. Panda, B.
Lukose, T. Heine, R. Banerjee, Angew. Chem., Int. Ed. 2013, 52, 13052-
13056.
[21] a) T. I. T. Okpalugo, P. Papakonstantinou, H. Murphy, J. McLaughlin, N. M.
D. Brown, Carbon 2005, 43, 153-161; b) A. P. Dementjev, A. de Graaf, M.
C. M. van de Sanden, K. I. Maslakov, A. V. Naumkin, A. A. Serov, Diamond
Relat. Mater. 2000, 9, 1904-1907.
[22] a) C. Miyake, T. Nakase, Y. Sano, J. Nucl. Sci. Technol. 1993, 30, 1256-
1260; b) S. D. McLaughlan, Phys. Rev. 1966, 150, 118-120.
This article is protected by copyright. All rights reserved.

10.1002/anie.201916360
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents (Please choose one layout)
Layout 2:
RESEARCH ARTICLE
Yang Li,[a] Xinghua Guo,[a] Xiaofeng Li,[a]
Meicheng Zhang,[a] Zhimin Jia,[a] Yun
Deng,[b] Yin Tian*,[b] Shoujian Li* [a] and
Lijian Ma* [a]
Page No. – Page No.
Redox-active Two-dimensional
Covalent Organic Frameworks for
Selective Reductive Separation of
Valence-variable, Redox-sensitive
and Long-lived Radionuclides
This study provides an optional approach for preparing high-performance novel
sorbing materials with prospective applications in selective separation of target
metal ions. Besides, by selective reduction and in situ loading of noble metal ions,
this study also provides an extensible way for the rational design and fabrication of
high performance supported catalysts.
This article is protected by copyright. All rights reserved.