A Porous π-Stacked Self-Assembly of Cup-Shaped Palladium Complex for Iodine Capture

Acquiring adsorbents capable of effective radioiodine capture is important for nuclear waste treatment; however, it remains a challenge to develop porous materials with high and reversible iodine capture. Herein, we report a porous self-assembly constructed by a cup-shaped PdII complex through intermolecular π···π interactions. This self-assembly features a cubic structure with channels along all three Cartesian coordinates, which enables it to efficiently capture iodine with an adsorption capacity of 0.60 g g−1 for dissolved iodine and 1.81 g g−1 for iodine vapor. Furthermore, the iodine adsorbed within the channels can be readily released upon immersing the bound solid in CH2Cl2, which allows the recycling of the adsorbent. This work develops a new porous molecular material promising for practical iodine adsorption.


Introduction
As a non-greenhouse energy source, nuclear energy is most likely to replace traditional fossil fuels [1,2]. Currently, nuclear energy is widely applied in many areas related to human life [3]. With the rapid development of the nuclear energy industry, the safe disposal of nuclear waste containing radioactive species, especially radioactive iodine, has become a significant concern [4][5][6][7][8]. Both 129 I and 131 I, which are the main radioisotopes for iodine, are harmful to its ecological surroundings and human health. 129 I is extremely dangerous because it has a long half-life (1.57 × 10 7 years) and can be accumulated in the human thyroid gland, causing serious diseases [6]. As for 131 I, it is often combined with hydrocarbons, giving rise to harmful organic compounds such as methane iodide [9][10][11][12]. Among various possible radioactive iodine species, molecular iodine (I 2 ) is the main pollutant in nuclear waste disposal and the nuclear accident [13,14]. Therefore, acquiring adsorbents for effective capture of I 2 is on demand.
Recently, macrocycle-based supramolecular assemblies have emerged as a class of adsorbents for I 2 capture [37,41,42]. For example, Huang's group reported perethylated pillar [6] arene, which acts as a candidate for I 2 capture [41], while Zhang and co-workers directly observed the ambiguous binding sites for I 2 in a mesoporous assembly of aluminum molecular rings [42]. Recently, we have successfully obtained a series of π-stacked porous assemblies based on metal complexes of tripodal tris(2-benzimidazolylmethyl) amine or tris(2-naphthimidazolemethyl) amine [43,44]. These achievements promoted us to synthesize porous assemblies based on metal complexes of tripodal ligands to explore high-performance adsorbents for I 2 capture.
In this work, we report a porous π-stacked self-assembly based on a cup-shaped Pd II complex. Due to the channels in the structure, this material permits the capture of both dissolved I 2 and I 2 vapor. Furthermore, the present adsorbent can be reused several times without significant loss of I 2 uptake capacity.

Structure Characterizations of the π-Stacked Self-Assembly
The self-assembly of (2,2 -bipyridine) dichloropalladium (II)([Pd(bipy)]Cl 2 ) with tris(2naphthimidazolemethyl) amine (H 3 L) in a mixture of MeOH/acetone (v/v: 1/3) with a trace of triethylamine affords yellow crystals of [Pd 3 (bipy) 3 L] Cl 3 ·solvent (1). Singlecrystal X-ray analysis (Table S1) reveals a cup-shaped trinuclear Pd II complex in which three [Pd(bipy)] 2+ cations are bridged by the naphthimidazolemethyl arms of L, giving rise to a macrocycle ( Figure 1a). Driven by the coordination mentioned above, L is fixed into an unusual cup-shaped conformation [39,40] and the three [Pd(bipy)] 2+ cations act as the cup holder. In the crystal structure, each [Pd 3 (bipy) 3 L] 3+ associates with its six neighbors ( Figure 1b) through π···π interactions between bipy and L, forming a porous non-symmetric cubic supramolecular assembly (Figure 1c). This porous structure possesses two kinds of channels along all three crystallographic axes, which are filled with Cl − and solvent molecules. Determined by PLATON, the void volume is 8658 Å 3 per unit cell, which is 48.3% of the unit volume. In the view of topology, treating [Pd 3 (bipy) 3 L] 3+ as a node and the π···π interaction between bipy and L as a linker ( Figure S1a), the porous assembly can be simplified as a pcu network with a Schläfli symbol of 4 6 ·6 9 ( Figure S1b). Thermogravimetric (TG) analysis with the sample heated under an N 2 stream revealed a weight loss of~15% between 30 and 200 • C, which can be attributed to the removal of solvent molecules (Figure 2a). After desolvation, the framework structure of the porous assembly collapses, as indicated by powder X-ray diffraction (PXRD) studies ( Figure 2b).

Iodine Adsorption Study
The poor thermostability of compound 1 prohibits us from investigating its iodine adsorption performance at high temperatures. Therefore, the adsorption performances of compound 1 on both gaseous and dissolved iodine were investigated at room temperature. Exposing compound 1 to iodine vapor at room temperature led to a gradual color change from yellow to black ( Figure S2a). The iodine uptake also gradually increased with time and attained an uptake of 1.37 g g -1 after 240 h without saturation (Figure 3a). The gaseous iodine adsorption profile can be well described by the pseudo-first-order kinetic model (R 2 = 0.996), which gives an adsorption rate k = 1.0 × 10 -4 g min -1 and an equilibrium adsorption capacity Qe = 1.81 g g -1 (Table S2).

Iodine Adsorption Study
The poor thermostability of compound 1 prohibits us from investigating its iodine adsorption performance at high temperatures. Therefore, the adsorption performances of compound 1 on both gaseous and dissolved iodine were investigated at room temperature. Exposing compound 1 to iodine vapor at room temperature led to a gradual color change from yellow to black ( Figure S2a). The iodine uptake also gradually increased with time and attained an uptake of 1.37 g g -1 after 240 h without saturation ( Figure 3a). The gaseous iodine adsorption profile can be well described by the pseudo-first-order kinetic model (R 2 = 0.996), which gives an adsorption rate k = 1.0 × 10 -4 g min -1 and an equilibrium adsorption capacity Qe = 1.81 g g -1 (Table S2).

Iodine Adsorption Study
The poor thermostability of compound 1 prohibits us from investigating its iodine adsorption performance at high temperatures. Therefore, the adsorption performances of compound 1 on both gaseous and dissolved iodine were investigated at room temperature. Exposing compound 1 to iodine vapor at room temperature led to a gradual color change from yellow to black ( Figure S2a). The iodine uptake also gradually increased with time and attained an uptake of 1.37 g g −1 after 240 h without saturation (Figure 3a). The gaseous iodine adsorption profile can be well described by the pseudo-first-order kinetic model (R 2 = 0.996), which gives an adsorption rate k = 1.0 × 10 −4 g min −1 and an equilibrium adsorption capacity Q e = 1.81 g g −1 (Table S2). We then examined the adsorption performance of compound 1 for iodine dissolved in cyclohexane. A crystalline sample of compound 1 (0.05 g) was immersed in a 3 mM iodine-cyclohexane solution. UV-Vis spectroscopy was used to evaluate the iodine adsorption rate (Figures 3b,c and S3). With the adsorption going on, the color of the iodinecyclohexane solution gradually faded ( Figure S2b). The color of the sample of compound 1 gradually deepened and turned black when the adsorption equilibrium was reached ( Figure S2c). The monitoring data revealed a fast adsorption rate in the first 6 h, and then the adsorption gradually slowed down until equilibrium (Figure 3b). The experimental data can be well described by the pseudo-second-order kinetic model (R 2 = 0.977), which gives an adsorption rate k2 = 3.0 × 10 -3 g min -1 and an equilibrium adsorption capacity of 0.60 g g -1 (Figure 3b, Table S2). The gaseous I2 and dissolved I2 uptake capacities of compound 1 are comparable to those of some promising I2 adsorbents (Table S3) [45][46][47][48]. Furthermore, the adsorbed iodine can be released from I2@1 by soaking I2@1 in CH2Cl2. When 0.50 g of solid I2@1 was immersed in CH2Cl2, the solution gradually changed from colorless to dark brown in 36 h, indicating a large amount of I2 was released ( Figure S4). Therefore, this adsorbent for iodine capture can be recycled. In the third adsorption-desorption cycle, ~70% of the I2 adsorption capability can be retained (Figure 3d).
To give insights into the I2 adsorption mechanism, we conducted Fourier transform infrared (FT-IR) spectroscopy ( Figure 4a) and X-ray photoelectron spectroscopy (XPS) (Figure 4b-d) studies on compound 1 before and after I2 uptake. After I2 loading, the characteristic band at ∼1634 cm −1 assigned to the C=N stretching vibration decreases significantly [14,19,29,33,34]. A pair of I 3d signals can be seen from the XPS of the sample after I2 uptake (Figure 4a,b). The signals at 617.84 and 629.37 eV can be attributed to I 3d5/2 and I 3d3/2, respectively. After I2 loading, the two N 1s signals shift from 397.94 and 399.14 eV We then examined the adsorption performance of compound 1 for iodine dissolved in cyclohexane. A crystalline sample of compound 1 (0.05 g) was immersed in a 3 mM iodinecyclohexane solution. UV-Vis spectroscopy was used to evaluate the iodine adsorption rate (Figures 3b,c and S3). With the adsorption going on, the color of the iodine-cyclohexane solution gradually faded ( Figure S2b). The color of the sample of compound 1 gradually deepened and turned black when the adsorption equilibrium was reached ( Figure S2c). The monitoring data revealed a fast adsorption rate in the first 6 h, and then the adsorption gradually slowed down until equilibrium (Figure 3b). The experimental data can be well described by the pseudo-second-order kinetic model (R 2 = 0.977), which gives an adsorption rate k 2 = 3.0 × 10 −3 g min −1 and an equilibrium adsorption capacity of 0.60 g g −1 (Figure 3b, Table S2). The gaseous I 2 and dissolved I 2 uptake capacities of compound 1 are comparable to those of some promising I 2 adsorbents (Table S3) [45][46][47][48]. Furthermore, the adsorbed iodine can be released from I 2 @1 by soaking I 2 @1 in CH 2 Cl 2 . When 0.50 g of solid I 2 @1 was immersed in CH 2 Cl 2 , the solution gradually changed from colorless to dark brown in 36 h, indicating a large amount of I 2 was released ( Figure S4). Therefore, this adsorbent for iodine capture can be recycled. In the third adsorption-desorption cycle,~70% of the I 2 adsorption capability can be retained (Figure 3d).
To give insights into the I 2 adsorption mechanism, we conducted Fourier transform infrared (FT-IR) spectroscopy ( Figure 4a) and X-ray photoelectron spectroscopy (XPS) (Figure 4b-d) studies on compound 1 before and after I 2 uptake. After I 2 loading, the characteristic band at ∼1634 cm −1 assigned to the C=N stretching vibration decreases significantly [14,19,29,33,34]. A pair of I 3d signals can be seen from the XPS of the sample after I 2 uptake (Figure 4a,b). The signals at 617.84 and 629.37 eV can be attributed to I 3d 5/2 and I 3d 3/2 , respectively. After I 2 loading, the two N 1s signals shift from 397.94 and 399.14 eV to 398.29 and 399.43 eV, respectively (Figure 4c). The two Pd 3d signals also shift from 336.31 and 341.51 eV to 337.88 and 343.78 eV, respectively (Figure 4d). These results indicate that the N and Pd atoms on compound 1 interact with the captured iodine [49]. This interaction may be rationalized in terms of that polarized bound iodine molecules favor interaction with the partly negatively charged N lone pairs, while the cylindrical electron surface of the I−I bond would favor interaction with the positively charged Pd atoms [45]. The PXRD of I 2 @1 is significantly different from that of compound 1, indicating a possible significant structural change upon iodine adsorption. However, the poor crystallinity of I 2 @1 prohibits us from directly observing the I 2 binding sites by single-crystal X-ray analysis. The recycled sample of compound 1 that lost crystallinity probably implies good dispersion of the adsorbed iodine molecules around the cup-shaped molecules (Figure 2b). to 398.29 and 399.43 eV, respectively (Figure 4c). The two Pd 3d signals also shift from 336.31 and 341.51 eV to 337.88 and 343.78 eV, respectively (Figure 4d). These results indicate that the N and Pd atoms on compound 1 interact with the captured iodine [49]. This interaction may be rationalized in terms of that polarized bound iodine molecules favor interaction with the partly negatively charged N lone pairs, while the cylindrical electron surface of the I−I bond would favor interaction with the positively charged Pd atoms [45]. The PXRD of I2@1 is significantly different from that of compound 1, indicating a possible significant structural change upon iodine adsorption. However, the poor crystallinity of I2@1 prohibits us from directly observing the I2 binding sites by single-crystal X-ray analysis. The recycled sample of compound 1 that lost crystallinity probably implies good dispersion of the adsorbed iodine molecules around the cup-shaped molecules (Figure 2b).

Iodine Adsorption Study
The ligand H3L is synthesized according to the previously reported method [50]. [Pd(bipy)] Cl2 and 2,3-diaminonaphthalene were purchased from bidepharmatech. All other reagents were purchased from Adamas (Shanghai, China) and used directly, without purification.

Iodine Adsorption Study
The ligand H 3 L is synthesized according to the previously reported method [50].

Characterization
Fourier-transform infrared (FTIR, Nicolet iS 50, Thermo Fisher, Waltham, MA, USA) spectra were recorded on a Thermo Fisher Nicolet iS 50 in the range 500-4000 cm −1 at room temperature. Powder X-ray diffraction (PXRD, Miniflex 600, Akishima, Rigaku, Tokyo, Japan) patterns were obtained on a Miniflex 600 diffractometer using Cu-Kα radiation with flat plate geometry. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, Waltham, MA, USA) studies were performed on an AXIS SUPRA Kratos system, and the C 1s line at 284.8 eV was used as the binding energy reference. TGA was performed using a thermo plus EVO2 system at a rate of 10 • C/min in the range of 30-800 • C (TGA/DSC 1, Mettler Telodo, Zurich, Switzerland). UV-Vis spectra were recorded on an Agilent Cary 5000 spectrophotometer (UV-Vis, Agilent, Santa Clara, CA, USA).

Crystallography
Single-crystal X-ray data were harvested on a Bruker D8 Venture diffractometer with Mo-Kα radiation at 200 K. Structures were solved using a direct method and refined by the full-matrix least-squares technique on F2 with the SHELXTL 2014 program [51]. All the H atoms are geometrically generated and refined using a riding model. The PLATON/SQUEEZE procedures [52] were used to treat the highly disordered solvents in the void of the porous structure. The X-ray crystallographic coordinates for structures reported in this article have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition number CCDC 2245193. The data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/ cif (accessed on 28 February 2023). Details of the crystallographic data are listed in Table S1.

Iodine Adsorption Experiments
Both the gaseous iodine and dissolved iodine uptake behaviors of compound 1 were studied at room temperature.

Iodine Vapor Adsorption
Air-dried compound 1 (0.050 g) was loaded into an uncapped glass vial, which was located in a sealed container with excess solid iodine kept at the bottom. After certain time intervals, the vial was taken out and weighed, and then reloaded into the vapor of iodine to continue adsorption. The iodine uptake at a certain time was calculated using Equation (1): where Q t represents the iodine uptake at a certain time and m 1 and m 2 are the masses of the sample of compound 1 before and after iodine uptake, respectively. The pseudo-first-order model (Equation (2)) was used to fit the gaseous iodine adsorption profile, giving a set of parameters with k 1 = 1.0 × 10 −4 g min −1 , Q e = 1.81 g g −1 , and R 2 = 0.996.

Iodine Adsorption in Solution
Air-dried compound 1 (0.050 g) was immersed in a 50 mL solution of iodine in cyclohexane (3 mM). The iodine adsorption process was monitored by UV-Vis spectroscopy. The iodine uptake was calculated using Equation (3): where Q t represents the iodine uptake at a certain time, C 0 and C t represent the concentration of iodine before and after adsorption, respectively, m represents the mass of compound 1, and V represents the volume of the solution. The pseudo-second-order model (Equation (4)) was used to fit the dissolved iodine adsorption profile, giving a set of parameters with k 2 = 3.0 × 10 −3 g min −1 , Q e = 0.60 g g −1 , and R 2 = 0.977.
3.5.3. Iodine Release and Recyclability of Compound 1 I 2 @1 was immersed in CH 2 Cl 2 to release the adsorbed iodine. Here, I 2 @1 (0.050 g) was immersed in CH 2 Cl 2 (100 mL). When the release was deemed essentially complete, the resulting solid was recycled and analyzed by PXRD. Then the recycled solid of compound 1 was added to the I 2 /cyclohexane solution again. After four cycles,~70% of the I 2 adsorption capability can be retained.

Conclusions
In summary, we have developed a porous self-assembly of a cup-shaped Pd II complex. This porous structure is constructed through intermolecular π···π interactions. The channels along all three crystallographic axes within the self-assembly allow for efficient reversible iodine capture, either from the vapor or solution source phases. These results demonstrate that porous crystalline materials assembled through weak intermolecular interactions can serve as a new type of promising adsorbent for I 2 capture.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules28072881/s1, Figure S1: (a) The simplification of the network of compound 1, (b) The simplified pcu network of the supramolecular framework of compound 1; Table S1: Crystallographic data of compound 1; Figure S2: (a) The setup for I 2 vapor adsorption, (b) Photographs showing color changes of the I 2 /cyclohexane solution as a function of time when 0.050 g of compound 1 was immersed in the solution, (c) Photographs showing the color change of the crystals of compound 1 before and after dissolved I 2 adsorption, (d) Photographs showing the release of I 2 from I 2 @1 in CH 2 Cl 2 ; Table S2: Fitting the iodine adsorption kinetics of compound 1; Figure S3: Standard plot between absorbance (λ = 523 nm) and I 2 concentration of the solution of I 2 in cyclohexane; Table S3: The comparison of I 2 adsorption capacities for various adsorbents.