Confinement of polysulfides within bi-functional metal–organic frameworks for high performance lithium–sulfur batteries

Abstract

A lithium–sulfur (Li–S) battery is regarded as the most promising candidate for next generation energy storage systems, because of its high theoretical specific capacity (1675 mA h g⁻¹) and specific energy (2500 W h kg⁻¹), as well as the abundance, low cost and environmental benignity of sulfur. However, the soluble polysulfides Li₂S_x (4 ≤ x ≤ 8) produced during the discharge process can cause the so-called “shuttle effect” and lead to low coulombic efficiency and rapid capacity fading of the batteries, which seriously restrict their practical application. Using porous materials as hosts to immobilize the polysulfides is proved to be an effective strategy. In this article, a dual functional cage-like metal–organic framework (Cu-MOF), Cu-TDPAT, combining the Lewis basic sites from the nitrogen atoms of the ligand H₆TDPAT with the Lewis acidic sites from Cu(II) open metal sites (OMSs), was employed as the sulfur host in a Li–S battery for lithium ions and polysulfide anions (S_x²⁻). In addition, the size of nano-Cu-TDPAT was also optimized by microwave synthesis to reduce the internal resistance of the batteries. The electrochemical test results showed that the optimized Cu-TDPAT material can efficiently confine the polysulfides within the MOF, and the resultant porous S@Cu-TDPAT composite cathode material with the size of 100 nm shows good cycling performance with a reversible capacity of about 745 mA h g⁻¹ at 1C (1C = 1675 mA g⁻¹) after 500 cycles, to the best of our knowledge, which is higher than those of all reported S@MOF cathode materials. The DFT calculation and XPS data indicate that the good cycling performance mainly results from the dual functional binding sites (that is, Lewis acid and base sites) in nanoporous Cu-TDPAT, providing the comprehensive and robust interaction with the polysulfides to overcome their dissolution and diffusion into the electrolyte. Clearly, our work provides a good example of designing MOFs with suitable interaction sites for the polysulfides to achieve S@MOF cathode materials with excellent cycling performance by multiple synergistic effects between nanoporous host MOFs and the polysulfides.

---

Introduction

Efficient energy storage systems are more and more desired with the development of portable electronic devices, electric vehicles, smart grids and other large energy storage equipment.^1,2 Traditional lithium-ion batteries (LIBs) are powerless to face the requirements of the market for power supply, transportation and portable energy storage because of their low energy densities and specific capacities based on the theoretical capacity limits of currently used cathode materials such as LiCoO₂ and LiFePO₄.³ Lithium–sulfur (Li–S) batteries are regarded as the next-generation energy storage systems because of their theoretically very high energy density of 2500 W h kg⁻¹ and specific capacity of 1675 A h kg⁻¹, as well as the abundance, low cost and environmental benignity of sulfur.^2,4 However, to realize their practical application, there are several key problems that should be overcome, and one of which is the solubility of polysulfides Li₂S_x (4 ≤ x ≤ 8) produced during the discharge process which can cause the so-called “shuttle effect” and lead to the low coulombic efficiency and rapid capacity fading of Li–S batteries.^2,5–9 Finding ways to fix the polysulfides at the cathode and effectively inhibit their dissolution and diffusion into the electrolyte is the key to solve this issue.

Using porous materials such as active carbon and porous oxides as the host for fastening sulfur is reported as an effective strategy. In order to improve the interaction between porous materials and polysulfides, and more effectively limit their dissolution, the porous materials are usually modified with binding sites that can interact with the polysulfides. Generally, metallic elements such as the transition metals Ti, V, Mn, Co, and the rare metal Nb and so on are used to build the binding sites for the polysulfide (S_x²⁻) anion, while non-metallic elements, such as N, O, P, S, are employed to interact with the lithium ion.^10–20 On the other hand, Metal–Organic Frameworks (MOFs), as a new kind of crystalline porous materials assembled by metal ions/clusters (as nodes) and organic ligands (as linkers) in infinite arrays, possess the advantages of being easy to design and functionalize owing to their pore structures and chemical environment compared with the traditional porous materials, which enable their potential applications in gas storage and separation, catalysis, detection, electrochemistry etc.^21–29 In recent years, using MOFs as the host of immobilizing sulfur in Li–S batteries gradually has attracted more and more interest.^30–35 Compared with the traditional porous carbon/metal oxides, from well designed assemblies, the metal nodes (Lewis acidic sites) and the functional groups from the organic linkers (Lewis basic sites) can provide effective binding sites for the lithium polysulfides and strongly confine them within the MOFs, especially those nano-sized porous framework complexes with rich cage-like structures, which offer a platform for scientists to design materials for effectively restraining the dissolution and diffusion of polysulfides at the molecular level. Until now, most of the reported S@MOF cathode materials only took advantage of the Lewis acid OMSs as the active binding sites for polysulfides, or controlled the sizes of the windows of pores by changing the organic ligands to limit their dissolution to a certain extent.^31,32,35 Nevertheless, the study of S@MOF composite cathode materials is just beginning and the developed S/MOF cathodes have still not attained the electrochemical performance of the top S/carbon cathodes. In order to fully explore the potential of MOF hosts to achieve superior battery performance, the pore structures and properties of MOFs need to be synergistically investigated.

Based on the above discussion, herein ligand 2,4,6-tris(3,5-dicarboxylphenyl-amino)-1,3,5-triazine (H₆TDPAT) derived from N-rich melamine is used to construct the cage-like MOF, Cu-TDPAT, which is rich in OMSs (namely, Cu(II)) and nitrogen functional sites, and used as the host for fixing sulfur/polysulfides in Li–S batteries (Scheme 1).³⁶ At the same time, the size of nano-Cu-TDPAT was also optimized by microwave synthesis to reduce the internal resistance of the batteries.^37–39 The electrochemical test results show that Cu-TDPAT can efficiently confine the polysulfides within the MOFs, and the S@Cu-TDPAT composite cathode material with the size of 100 nm exhibits good cycling performance, with a reversible capacity of about 745 mA h g⁻¹ at 1C after 500 cycles, which is higher than those of all reported S@MOF cathode materials.^30–35 The DFT calculation and XPS data indicate that the good cycling performance results from the dual functional binding sites (that is, Lewis acidic and basic sites) in nanoporous Cu-TDPAT providing the comprehensive and robust interaction with the polysulfides to overcome their dissolution and diffusion into the electrolyte. Clearly, our work provides a good example of designing MOFs with suitable interaction sites for the polysulfides to achieve S@MOF cathode materials with excellent cycling performance.

Scheme 1 Crystal structure of Cu-TDPAT constructed by H₆TDPAT and Cu²⁺ ions containing three types of cages.

Experimental

Preparation of Cu-TDPAT nanoparticles

The ligand 2,4,6-tris(3,5-dicarboxylphenylamino)-1,3,5-triazine (H₆TDPAT) was synthesized according to the ref.36 In the typical procedure, a solution of Cu(NO₃)₂·3H₂O (2.416 g, 10 mmol) and H₆TDPAT (1.236 g, 2 mmol) in 150 mL N,N-dimethyl formamide (DMF) was added in a 250 mL round-bottom flask, which was placed in a microwave synthesizer with the power of 400 W and allowed to react for a certain period. After cooling down to room temperature, the product was collected by centrifugation and washed with DMF several times. The reaction times of nano-Cu-TDPAT with the sizes of 100 nm, 200 nm, 500 nm and 1 μm were 2 min, 3 min, 4 min and 5 min, respectively.

Preparation of S@Cu-TDPAT

In order to remove the guest solvent in Cu-TDPAT, the powder of Cu-TDPAT was soaked in MeOH for 2 days and then filtered and dried at 60 °C which was further heated at 120 °C under vacuum conditions for 24 h to obtain active Cu-TDPAT. Active Cu-TDPAT and sulfur with the mass ratio of 1 : 1 were mixed by hand-milling and heated through the melt-diffusion method at 155 °C under an Ar atmosphere for 2 h. After cooling to room temperature, the S@Cu-TDPAT cathode materials were collected.

Battery assembly

80 wt% S@Cu-TDPAT composite, 10 wt% super-P and 10 wt% poly(vinylidene fluoride) (PVDF) binder were mixed in N-methyl-2-pyrrolidone (NMP) to form a slurry. Sonication was used to improve the dispersion. The slurry was then coated on aluminium current collectors as the cathode and dried at 60 °C over 24 h. Coin cells 2032 were assembled in an Ar-filled glovebox (O₂, H₂O < 0.1 ppm) with an electrolyte of 1.0 M LiTFSI and 0.1 M LiNO₃ in DOL/DME (v : v, 1 : 1), a Celgard 2400 membrane, and a Li foil reference anode. For all of the electrochemical tests, the sulfur mass loading was about 1.2 mg cm⁻².

Characterization

The X-ray powder diffraction patterns were measured on a Bruker D8 Advance diffractometer at 40 kV and 40 mA with a Cu target tube and a graphite monochromator. Thermogravimetric analyses (TGA) were performed on a Netzsch TG 209 F3 Tarsus from room temperature to 800 °C with a heating rate of 10 °C min⁻¹ under flowing nitrogen. The nitrogen physisorption was performed using automatic volumetric adsorption equipment (Belsorp-max); the surface area analysis was based on the Brunauer–Emmett–Teller (BET) theory. SEM morphology and EDS mapping were performed using TESCAN Brno, s.r.o. MIRA3 LMH field-emission scanning electron microscopy (SEM) with an Oxford Instruments NanoAnalysis INCA Energy 150 X-ACT energy dispersive spectrometer (EDS). X-ray photoelectron spectroscopy (XPS) analysis was performed on a Thermo Scientific Escalab 250Xi instrument with Mg Kα radiation (1253.6 eV) at a scan step of 0.1 eV. Galvanostatic charge–discharge cycles were studied using a LAND CT2001A instrument (Wuhan Jinnuo Electronic Co. Ltd) at different C rates (1C = 1680 mA g⁻¹) between 1.8 and 2.8 V (vs. Li⁺/Li) at room temperature.

Computational methods

The preferred sorption locations were searched through Grand Canonical Monte Carlo (GCMC) simulations by using the fixed loading task and the Metropolis method in the sorption calculation module. For all the GCMC simulations, the simulation box was set with 4 (2 × 2 × 1) unit cells while the framework and the polysulfide (Li₂S_x, 4 ≤ x ≤ 8) molecules were described using the universal force field (UFF). Both the framework and the polysulfide (Li₂S_x, 4 ≤ x ≤ 8) molecules were regarded as rigid models and the Q_Eq partial charges and Q_Eq charges were applied to the atoms of the framework and guest molecules, respectively. The cut off distance was set to 18.5 Å for the Lennard–Jones (LJ) interactions, and the electrostatic interactions and the van der Waals interactions were handled using the Ewald and Atom based summation methods, respectively. The loading steps, equilibration steps and the production steps were all set to 1.0 × 10⁷. The binding energies between MOF and polysulfide (Li₂S_x, 4 ≤ x ≤ 8) were calculated using the density functional theory (DFT) within the Perdew–Burke–Ernzerhof generalized gradient approximation (GGA-PBE), as implemented in the Dmol3 package. The double numerical plus polarization (DNP) basis sets with effective core potential were employed to express atomic potentials. Considering the large size of the Cu-TDPAT cell with 960 atoms, as shown in Fig. S10a,† a segment of Cu₂(TDPAT)₂ containing 104 atoms (Fig. S10b†) was used to model its interaction with polytellurides. The optimized binding configurations with the lowest energy were obtained based on adsorption conformations obtained from GCMC simulation. Self-consistent field calculations were carried out until the SCF tolerance was below 1 × 10⁻⁶. The adsorption energy between the segment of Cu₂(TDPAT)₂ and adsorbed polysulfide (Li₂S_x, 4 ≤ x ≤ 8) is determined using E_a = E_seg. + E_Li2Sx − E_Li2Sx@_seg., where E_seg., E_Li2Sx and E_Li2Sx@_seg. denote the total energies of the segment of Cu₂(TDPAT)₂, the adsorbed polysulfide, and the polysulfide–segment system, respectively.

Results and discussion

As shown in Scheme 1, Cu-TDPAT is constructed using ligand H₆TDPAT and the paddle-wheel-like node Cu₂(COO)₄ with three different kinds of cages in which the nitrogen functional group on ligand H₆TDPAT and the axial position of Cu₂(COO)₄ can provide the adsorption sites for lithium polysulfides (Li₂S_x). In Cu-TDPAT, cage A is built by the Cu₂(COO)₄ nodes and m-phthalic acid, which is similar to the cage structure of HKUST-1; cage B is a tetrahedral cage using four ligands as the wall with the size of 14 Å, while cage C is the largest one adopting eight ligands as the wall to form the octahedral cage with the size of 20.3 Å (Scheme 1). The cage structure with the confined nanospace effect, abundant Cu(II)-OMSs or/and the nitrogen functional sites provides strong assurance for confining the polysulfides and indicates its application as the sulfur carrier in Li–S batteries.

Nano-Cu-TDPAT was prepared through microwave synthesis because most of the MOFs perform poorly in electronic and ionic conduction, while reducing the MOF particles may shorten the transport length of both electron and lithium ions and increase the electrode/electrolyte contact area, so as to decrease the internal resistance.^37–39 Octahedral nano-Cu-TDPAT with the sizes of ∼100 nm, ∼200 nm, ∼500 nm and ∼1 μm were obtained after the microwave reaction of the mixture of Cu(NO₃)₂ and H₆TDPAT in a DMF solution at the power of 400 W for 2 min, 3 min, 4 min and 5 min respectively (Fig. 1a and S4†). Powder X-ray diffraction indicated that nano-Cu-TDPAT samples with different sizes were all in their pure phase (Fig. 1e). The adsorption isotherms of N₂ at 77 K for different sizes of nano-Cu-TDPAT all show typical type I curves for micropores (Fig. S2†). In order to fully confine the sulfur in the MOFs’ pores, it is reasonable to adjust the sulfur loading level according to the micropore volume of the Cu-TDPAT samples, and the sulfur contents of loading for Cu-TDPAT samples with different sizes are all about ∼50% (Table S1†). Here all the Cu-TDPAT samples were loaded with the same quality of sulfur for better comparison with each other. In order to load the sulfur, after being soaked in methanol for 2 days and then activated at 120 °C for 24 h, the Cu-TDPAT samples were manually blended with the sulfur powder at the mass ratio of 1 : 1 and then heated at 155 °C for 2 h to obtain S@Cu-TDPAT. The XRD results (Fig. 2a and S1†) show that the structure of Cu-TDPAT was still maintained well after loading with sulfur while the characteristic peak at about 23.1 degrees of sulfur disappeared in S@Cu-TDPAT indicating that the sulfur was fully adsorbed into the pores of Cu-TDPAT, which was further proved by the N₂ adsorption data of Cu-TDPAT and S@CuTDPAT at 77 K (Fig. S2†). Meanwhile, the SEM of S@Cu-TDPAT revealed that after loading with sulfur, S@Cu-TDPAT was still in its octahedral shape and there were no obvious sulfur particles on the surface of MOFs (Fig. S3†). The EDS mapping of S@Cu-TDPAT also indicated that the sulfur was equably loaded in Cu-TDPAT (Fig. S3†). In order to calculate the total loading amount of sulfur, the thermogravimetric analysis of Cu-TDPAT, S and S@Cu-TDPAT was carried out under a nitrogen atmosphere from room temperature to 800 °C. As shown in Fig. 2b, S@Cu-TDPAT experienced two obvious weight loss processes: the first one was from about ∼205 °C to ∼327 °C with a weight loss of about 50% which can be attributed to the sublimation of sulfur in S@Cu-TDPAT; the second one occurred from ∼327 °C to 800 °C which corresponded to the decomposition of MOFs. It is found that the sublimation temperature of sulfur in S@Cu-TDPAT was higher than that of pristine sulfur (about 190 °C) by about 25 °C which implies the interaction between Cu-TDPAT and sulfur in S@Cu-TDPAT. To further illustrate this interaction, the XPS characterization of pristine sulfur and S@Cu-TDPAT was executed and it revealed the obvious chemical shift of S_2p spectra in S@Cu-TDPAT compared with that of pristine sulfur, that is from original 164.8 eV to 164.4 eV of S_2p1/2 and 163.6 eV to 163.2 eV of S_2p3/2 (Fig. 2c). All of these confirm the good interaction between Cu(II)-OMSs in Cu-TDPAT and sulfur, and Cu-TDPAT is a suitable carrier for sulfur to fix it in the cathode so as to improve the cycling performance of Li–S batteries.

Fig. 1 SEM of nano-Cu-TDPAT with different sizes: (a) ∼100 nm; (b) ∼200 nm; (c) ∼500 nm; and (d) ∼1 μm and (e) their corresponding XRD.

Fig. 2 (a) PXRD patterns of sulfur and Cu-TDPAT after being ground (Cu-TDPAT + S) and heated (S@Cu-TDPAT); (b) TGA curves of sulfur, Cu-TDPAT and S@Cu-TDPAT under a N₂ atmosphere; and (c) XPS S_2p spectra of Cu-TDPAT + S and S@Cu-TDPAT-100 nm.

It was reported that the size of nano-MOFs have great impact on the performance of Li–S batteries.³⁴ Hence, Cu-TDPAT samples with different sizes were chosen to prepare S@Cu-TDPAT cathode materials for electrochemical tests. All of S@Cu-TDPAT were assembled into button batteries with lithium as the anode and were galvanostatically charged and discharged at 0.5C (837 mA g⁻¹). As shown in Fig. S7,† the cyclic voltammetry curves of S@Cu-TDPAT composites with different particle sizes all present two cathodic peaks at around ∼2.35 and ∼2.05 V corresponding to the reduction of S₈ and soluble long chain polysulfides, respectively, while the two close anodic peaks at around ∼2.35 and ∼2.45 V in the anodic sweep result from the progressive transition of Li₂S to lower order polysulfides and higher order polysulfides to elemental sulfur, respectively. Two typical voltage platforms of all S@Cu-TDPAT in Fig. 3a also indicated that the sulfur in S@Cu-TDPAT was discharged from S₈ to soluble long chain polysulfides and to Li₂S rather than from short sulfur of S_2∼4 to Li₂S in some reported microporous carbons.^40,41 It can be seen clearly that the discharge capacity of the first cycle increased with the decrease of the size of nano-Cu-TDPAT, from 632 mA h g⁻¹ to 995 mA h g⁻¹ under the current of 0.5C. On further exploring the discharge behaviour, it is found that the capacity of Q_H (the discharge process occurs at a higher voltage corresponding to the reduction of S₈ to soluble Li₂S₄) gradually increases from 186 mA h g⁻¹ to 269 mA h g⁻¹ with the decrease of the size of nano-Cu-TDPAT that is the percentage of sulfur participating in the reaction increases from 44% to 64% and the ratio of the discharge capacity between low voltage and high voltage, Q_L/Q_H, was enhanced from 2.40 to 2.68 (Fig. 3a and b). The combination of these phenomena indicates that with the reduction of the size of nano-Cu-TDPAT, the internal sulfur loaded in Cu-TDPAT can be reacted more sufficiently leading to the enhancement of the charge–discharge capacity of the first cycle which is proved by Zhou³⁴ as well. Unlike the case of the conductive hosts such as porous carbon, the electrochemical process would only occur on the external surfaces of the MOF particles, where the active material can get access to both the electron delivering phase (carbon black) and the Li ion delivering phase (electrolyte). The inner sulfur of the MOF particles can only be used after the outer one is reacted completely. So the size of the MOF particles can directly affect the utilization of the sulfur loading in MOFs. The internal sulfur loaded in Cu-TDPAT with a small size can be reacted sufficiently leading to the enhancement of sulfur utilization.

Fig. 3 Galvanostatic tests of the S@Cu-TDPAT composites at 0.5C. Voltage profiles (a) and their normalized forms (the inset figure shows the dependence of maximum discharge capacity and Q_L/Q_H on the particle size) of the first cycle (b). (c) Long term cycling performance at 0.5C. (d) Rate capability of S@Cu-TDPAT composites. (e) Cycling stability of a S@Cu-TDPAT-100 nm cathode at 1 C for 500 cycles.

The long-term cycling performance of 300 cycles of S@Cu-TDPAT with different sizes under 0.5C is presented at Fig. 3c. It can be seen that the first cycle capacity and the degree of attenuation over the 300 cycles differed with the size of S@Cu-TDPAT, and the reversible capacities at 0.5C after 300 cycles are 352 mA h g⁻¹ (1 μm), 436 mA h g⁻¹ (500 nm), 631 mA h g⁻¹ (200 nm) and 831 mA h g⁻¹ (100 nm), respectively. Among them, the 100 nm-size S@Cu-TDPAT shows the relatively stable cycling performance and the highest capacity after 300 cycles, which represents the best performance of the reported S@MOF cathode for Li–S batteries so far, compared with the S@NiMOF (about 500 mA h g⁻¹ at 0.2C after 200 cycles), S@ZIF-8 (about 600 mA h g⁻¹ at 0.5C after 250 cycles) and S@MOF-525(Cu) (about 704 mA h g⁻¹ at 0.5C after 200 cycles) (Fig. S9†).^30–35 Besides, the ratio performance of different sizes of S@Cu-TDPAT cathodes were studied as well and the 100 nm-size S@Cu-TDPAT also shows better. It is demonstrated that the 100 nm-size S@Cu-TDPAT cathode exhibits capacities of about ∼1000 mA h g⁻¹ at 0.1C, ∼900 mA h g⁻¹ at 0.5C, ∼820 mA h g⁻¹ at 1C, ∼720 mA h g⁻¹ at 2C, and even ∼523 mA h g⁻¹ at 5C, while being restored to ∼948 mA h g⁻¹ as the C-rate switches from 5C to 0.1C. What's more, the 100 nm-size S@Cu-TDPAT cathode delivers good long-term cycling performance even at the current as high as 1C with its capacity stabilizing at about 745 mA h g⁻¹ (298 mA h g⁻¹ based on the whole electrode including binder and electrolyte) after 500 cycles (Fig. 3e and S6†).

The electrochemical test of the S@Cu-TDPAT cathode indicated that Cu-TDPAT can effectively confine the polysulfides within the pores leading to the high capacity and cycling stability. It is reported that the microporous properties and the polarity of pores in the materials usually could lead to the good adsorption for polysulfides.^42,43 For Cu-TDPAT, the type I N₂ adsorption isotherms in Fig. S2† confirm its microporosity which contributes to its high adsorption affinity for guest polysulfides. What's more, the OMSs and the N-rich functional group of ligand H₆TDPAT render the pore surface rather polar, which also increase the polar–polar interactions with polysulfides in MOFs.^31,35,43 In order to understand the binding states of different kinds of polysulfides in Cu-TDPAT, the GCMC and DFT calculations were employed to study the adsorption sites and energy. As shown in Fig. 4, the optimum sorption sites of all the polysulfides are located in the cages containing N-rich ligands (either cage B or cage C) with a similar behavior in Cu-TDPAT. For the polysulfide anions S_x²⁻ (2 ≤ x ≤ 8), in all of the cages the terminal sulfur atom of the chain binds to the Cu²⁺ ions of the Cu(II)-OMSs which is also reported in other MOFs (Fig. 4 and S7†).³² Regarding the lithium ion, it interacted with the nitrogen functional group in cage B or cage C and we can find that the lithium ion prefers to interact with the secondary amines which may be due to the better coordination between the lone pair electrons of the secondary amine and the lithium ion. What's more, the adsorption energy between polysulfides and Cu-TDPAT was calculated at the same time, and the result showed that Li₂S₄ and Cu-TDPAT had the strongest interaction (Fig. S12†).

Fig. 4 The optimum atomic model configurations showing the interactions between Cu-TDPAT and polysulfide anions. (a) Cu-TDPAT and Li₂S₈; (b) Cu-TDPAT and Li₂S₆; (c) Cu-TDPAT and Li₂S₄; and (d) Cu-TDPAT and Li₂S₂.

To provide more evidence for the strong interaction between Cu-TDPAT and polysulfides, the activated Cu-TDPAT powder was soaked in the Li₂S₄/DME solution and it is clear that Cu-TDPAT can completely adsorb the Li₂S₄ (Fig. 5a). Then Cu-TDPAT adsorbed with Li₂S₄, Li₂S₄@Cu-TDPAT, was tested by XPS. The chemical shifts of the Cu_2p spectra of Li₂S₄@Cu-TDPAT to the lower binding energy is clearly observed in Fig. 5b because of the interaction between polysulfide anions and Cu²⁺ ions. Fig. 5c shows the spectra of N_1s. Because there are only two types of nitrogen in Cu-TDPAT, namely secondary-amine nitrogen and pyridine nitrogen, the higher binding energy of 400.4 eV is attributed to the secondary-amine nitrogen and 398.8 eV to the pyridine nitrogen. As shown in Fig. 5c, there are clear N_1s chemical shifts of Li₂S₄@Cu-TDPAT to the lower energy while the chemical shift of secondary-amine nitrogen (0.6 eV) is larger than that of pyridine nitrogen (0.4 eV) which imply that the interaction between secondary-amine nitrogen and lithium is stronger and the result is in accordance with the DFT calculation. The Li_1s and S_2p chemical shifts to the lower energy in Li₂S₄@Cu-TDPAT also prove the strong interaction (Fig. 5d and e).

Fig. 5 (a) Li₂S₄/DME solution and the solution after contact with Cu-TDPAT; (b) XPS Cu2p spectra of Cu-TDPAT and Li₂S₄@Cu-TDPAT; (c) XPS N1s spectra of Cu-TDPAT and Li₂S₄@Cu-TDPAT; (d) XPS Li1s spectra of Li₂S₄ and Li₂S₄@Cu-TDPAT; and (e) XPS S2p spectra of Li₂S₄ and Li₂S₄@Cu-TDPAT.

In summary, ligand H₆TDPAT with the N-rich functional groups as linkers and the paddle-wheel-like Cu₂(COO)₄ including Cu(II) open metal sites (OMSs) as nodes were used to develop a bi-functional nano-sized MOF (namely Cu-TDPAT) containing three types of cages A, B, and C for carrying sulfur. Because of the synergistic effect from Lewis acidic sites (central unsaturated Cu^II ions) binding with polysulfide anions and Lewis basic sites (N atoms in ligand) coordinating to lithium ions as well as the microporous properties, the resultant porous Cu-TDPAT shows excellent confining ability of the polysulfides and effectively inhibits their dissolution and diffusion into the electrolyte. Thus, a high reversible capacity of about 745 mA h g⁻¹ at 1C after 500 cycles was achieved. To the best of our knowledge, it represents the best performance in the currently reported S@MOF cathode materials for MOF-based Li–S batteries.^30–35 Clearly, our work provides a good example of designing MOFs with suitable synergistic interaction sites for sulfur and polysulfides to achieve S@MOF cathodes with excellent cycling performance.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are grateful for the financial aid from the National Natural Science Foundation of P. R. China (Grant No. 21471061 and 21671071), the National Natural Science Foundation of Guangdong Province (Grant No. 2014A030311001 and 2016A030310435), Scientific Research Foundation of Graduate School of South China Normal University (Grant No. 2016lkxm06), Applied Science and Technology Planning Project of Guangdong Province, Guangzhou, China (No. 2015B010135009), Innovation team project of Guangdong Ordinary University (No. 2015KCXTD005), and the great scientific research project of Guangdong Ordinary University (No. 2016KZDXM023).

Notes and references

1. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J.-M. Tarascon, Nat. Mater., 2012, 11, 19–29.
2. Z. W. Seh, Y. Sun, Q. Zhang and Y. Cui, Chem. Soc. Rev., 2016, 45, 5605–5634.
3. J. B. Goodenough and Y. Kim, Chem. Mater., 2010, 22, 587–603.
4. A. Manthiram, S.-H. Chung and C. Zu, Adv. Mater., 2015, 27, 1980–2006.
5. W. Hua, Z. Yang, H. Nie, Z. Li, J. Yang, Z. Guo, C. Ruan, X. Chen and S. Huang, ACS Nano, 2017, 11, 2209–2218.
6. H. Lin, L. Yang, X. Jiang, G. Li, T. Zhang, Q. Yao, G. W. Zheng and J. Y. Lee, Energy Environ. Sci., 2017, 10, 1476–1486.
7. Y. Pan, Y. Zhou, Q. Zhao, Y. Dou, S. Chou, F. Cheng, J. Chen, H. K. Liu, L. Jiang and S. X. Dou, Nano Energy, 2017, 33, 205–212.
8. C. Zhang, H. B. Wu, C. Yuan, Z. Guo and X. W. Lou, Angew. Chem., Int. Ed., 2012, 51, 9592–9595.
9. H. Wang, W. Zhang, H. Liu and Z. Guo, Angew. Chem., Int. Ed., 2016, 55, 3992–3996.
10. Z. Xiao, Z. Yang, L. Zhou, L. Zhang and R. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 18845–18855.
11. Q. Pang, D. Kundu, M. Cuisinier and L. F. Nazar, Nat. Commun., 2014, 5, 4759.
12. T. Zhou, W. Lv, J. Li, G. Zhou, Y. Zhao, S. Fan, B. Liu, B. Li, F. Kang and Q.-H. Yang, Energy Environ. Sci., 2017, 10, 1694–1703.
13. Z. Sun, J. Zhang, L. Yin, G. Hu, R. Fang, H.-M. Cheng and F. Li, Nat. Commun., 2017, 8, 14627.
14. X. Liu, J. Q. Huang, Q. Zhang and L. Mai, Adv. Mater., 2017, 29, 1601759.
15. Z. Li, J. Zhang and X. W. Lou, Angew. Chem., Int. Ed., 2015, 54, 12886–12890.
16. K. Liao, P. Mao, N. Li, M. Han, J. Yi, P. He, Y. Sun and H. Zhou, J. Mater. Chem. A, 2016, 4, 5406–5409.
17. Q. Pang and L. F. Nazar, ACS Nano, 2016, 10, 4111–4118.
18. Z. Xiao, Z. Yang, L. Zhang, H. Pan and R. Wang, ACS Nano, 2017, 11, 8488–8498.
19. H. Wang, T. Zhou, D. Li, H. Gao, G. Gao, A. Du, H. Liu and Z. Guo, ACS Appl. Mater. Interfaces, 2017, 9, 4320–4325.
20. K. Mi, S. Chen, B. Xi, S. Kai, Y. Jiang, J. Feng, Y. Qian and S. Xiong, Adv. Funct. Mater., 2017, 27, 1604265.
21. X. Chen, Z. Wang, R. Zhang, L. Xu and D. Sun, Chem. Commun., 2017, 53, 10560–10563.
22. F. Wang, H.-Y. Zhuo, X. Han, W.-M. Chen and D. Sun, J. Mater. Chem. A, 2017, 5, 13079–13085.
23. P.-Q. Liao, N.-Y. Huang, W.-X. Zhang, J.-P. Zhang and X.-M. Chen, Science, 2017, 356, 1193–1196.
24. J. Liu, L. Chen, H. Cui, J. Zhang, L. Zhang and C. Y. Su, Chem. Soc. Rev., 2014, 43, 6011–6061.
25. Z. Hu, B. J. Deibert and J. Li, Chem. Soc. Rev., 2014, 43, 5815–5840.
26. J. Liu, D. Zhu, C. Guo, A. Vasileff and S.-Z. Qiao, Adv. Energy Mater., 2017 DOI:10.1002/aenm.201700518.
27. Z. Bai, Y. Zhang, Y. Zhang, C. Guo, B. Tang and D. Sun, J. Mater. Chem. A, 2015, 3, 5266–5269.
28. X. Han, W.-M. Chen, X. Han, Y.-Z. Tan and D. Sun, J. Mater. Chem. A, 2016, 4, 13040–13045.
29. W.-M. Chen, X.-L. Meng, G.-L. Zhuang, Z. Wang, M. Kurmoo, Q.-Q. Zhao, X.-P. Wang, B. Shan, C.-H. Tung and D. Sun, J. Mater. Chem. A, 2017, 5, 13079–13085.
30. R. Demir-Cakan, M. Morcrette, F. Nouar, C. Davoisne, T. Devic, D. Gonbeau, R. Dominko, C. Serre, G. Ferey and J. M. Tarascon, J. Am. Chem. Soc., 2011, 133, 16154–16160.
31. J. Zheng, J. Tian, D. Wu, M. Gu, W. Xu, C. Wang, F. Gao, M. H. Engelhard, J. G. Zhang, J. Liu and J. Xiao, Nano Lett., 2014, 14, 2345–2352.
32. J. Zhou, R. Li, X. Fan, Y. Chen, R. Han, W. Li, J. Zheng, B. Wang and X. Li, Energy Environ. Sci., 2014, 7, 2715.
33. Z. Zhao, S. Wang, R. Liang, Z. Li, Z. Shi and G. Chen, J. Mater. Chem. A, 2014, 2, 13509–13512.
34. J. Zhou, X. Yu, X. Fan, X. Wang, H. Li, Y. Zhang, W. Li, J. Zheng, B. Wang and X. Li, J. Mater. Chem. A, 2015, 3, 8272–8275.
35. Z. Wang, B. Wang, Y. Yang, Y. Cui, Z. Wang, B. Chen and G. Qian, ACS Appl. Mater. Interfaces, 2015, 7, 20999–21004.
36. B. Li, Z. Zhang, Y. Li, K. Yao, Y. Zhu, Z. Deng, F. Yang, X. Zhou, G. Li, H. Wu, N. Nijem, Y. J. Chabal, Z. Lai, Y. Han, Z. Shi, S. Feng and J. Li, Angew. Chem., Int. Ed., 2012, 51, 1412–1415.
37. J. Zhu, Y. K. Sharma, Z. Zeng, X. Zhang, M. Srinivasan, S. Mhaisalkar, H. Zhang, H. H. Hng and Q. Yan, J. Phys. Chem. C, 2011, 115, 8400–8406.
38. J. G. Kang, Y. D. Ko and J. G. Park, Nanoscale Res. Lett., 2008, 3, 390–394.
39. C.-T. Hsieh, J.-S. Lin, Y.-F. Chen and H. Teng, J. Phys. Chem. C, 2012, 116, 15251–15258.
40. Z. Li, Y. Jiang, L. Yuan, Z. Yi, C. Wu, Y. Liu, P. Strasser and Y. Huang, ACS Nano, 2014, 8, 9295–9303.
41. Z. Li and L. Yin, ACS Appl. Mater. Interfaces, 2015, 7, 4029–4038.
42. F. Hippauf, W. Nickel, G.-P. Hao, K. Schwedtmann, L. Giebeler, S. Oswald, L. Borchardt, S. Doerfler, J. J. Weigand and S. Kaskel, Adv. Mater. Interfaces, 2016, 3, 1600508.
43. G.-P. Hao, C. Tang, E. Zhang, P. Zhai, J. Yin, W. Zhu, Q. Zhang and S. Kaskel, Adv. Mater., 2017, 29, 1702829.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nr07118c

---