One-dimensional lead iodide hybrid stabilized by inorganic hexarhenium cluster cations as a new broad-band emitter

A novel one-dimensional (1D) hybrid {[Re6S8(PzH)6][Pb3I8(DMF)2]}·6(DMF) with hexarhenium cluster cations has been synthesized and characterized by means of single-crystal X-ray diffraction. Two DMF oxygen atoms bridge three lead iodides to form a set of lead iodides, {[PbI4/2I(ODMF)1/2][PbI4/2(ODMF)2/2][PbI4/2I(ODMF)1/2]}2−, and these sets of lead iodide share edges to form a 1D lead iodide chain, [Pb3I8(DMF)2]2− which has never been reported before and is different from the typical edge sharing of octahedral PbI6 units. 1D lead iodide chains are stacked along the a axis, and [Re6S8(PzH)6]2+ cations with H-bonded DMF molecules to pyrazole N–H reside between 1D lead iodide chains. This 1D lead iodide hybrid shows strong broad-band emission with a peak at 634 nm. The excellent photoluminescent properties of the new lead iodide hybrid exhibit great potential for optoelectronic applications in photonic devices with broad-band emission and stability. This study introduces a new class of lead iodide hybrid compounds having new inorganic cluster cations rather than the organic amine cations that have been used in numerous studies to date. This work opens a promising path to overcome the instability of perovskites including of organic amine cations.


Introduction
In recent years, research on organometal halide perovskites for optoelectronic and photovoltaic use has grown exponentially. [1][2][3][4] The use of inorganic-organic hybrid perovskites is currently receiving a great deal of attention due to their superior light harvesting, tunable band gap, high color purity and low-cost solution processability. [5][6][7][8] These perovskites can be classied depending upon the connectivities of their inorganic networks. The general formula for a three-dimensional (3D) perovskite is ABX 3 , where A is methylammonium (MA + ), formamidinium (FA + ), or Cs + ; B is Pb 2+ , Sn 2+ , or Ge 2+ ; and X is a halide (Cl À , Br À , or I À ). 3D perovskites consist of corner-sharing metal halide octahedra, and A + cations reside in the cavities of 3D inorganic frameworks. When larger organic cations than MA + are employed, it is possible to form lowdimensional perovskite structures in which the connectivity of the inorganic network is reduced to two-dimensional (2D) sheets, onedimensional (1D) chains, or zero-dimensional (0D) clusters. 4,9 Lowering the dimensionality to 2D gives rise to increased structural diversity. The connectivity modes of the [BX 6 ] 4À make the formula repeating unit of one-dimensional chains vary when the dimensionality is reduced from 3D or 2D to 1D. To date, facesharing connectivity has been the most common connectivity in the 1D perovskite family of structures, followed by corner-sharing, and edge-sharing connectivity is the rarest structure in the 1D group. In general, the metal halide octahedra in 1D perovskites are connected in chain structures and these chains are wrapped by organic cations. The reduction of dimensionality in these perovskites greatly affects their optoelectronic and physical properties, opening a door to new realms of materials science.
In the near future, the ideal replacement technology for conventional incandescent and uorescent lamps is lightemitting diodes (LEDs), because of their energy savings, high luminous efficiency, and long lifetime. Recently, organometal halide perovskites have emerged as a new class of materials for LED applications. [10][11][12] In 2014, the rst LED based on a perovskite material was reported by Friend and coworkers. 13 This device showed a maximum external quantum efficiency (EQE) of 0.1% in the green region and 0.76% in the infrared region. In 2018, four different groups achieved EQEs greater than 20% in LEDs based on perovskite materials. [10][11][12]14 Thus, perovskites have great potential for lighting applications. White lightemitting diodes are of special interest, and LEDs have numerous merits over conventional light sources. [15][16][17] Diverse device architectures have been researched toward the goal of pure white emission. These architectures consist of (I) a blue LED chip, (II) a monochromatic blue LED, and (III) a single broad-band white phosphor LED. [18][19][20] Poor color rendition, efficiency loss, and instability of emission color over time are serious drawbacks of approaches (I) and (II). Compared with these approaches, approach (III) has emerged as a rising star because it can overcome these difficulties. Low-dimensional perovskites are garnering more interest because of their structural tunability, leading to controllable band structure and tunable color. Compared with the research focus on optimization for 2D perovskites, less research attention has been paid to the 1D perovskites even though 1D perovskite chains also feature intrinsic broad-band light emissions. For example, Yuan and coworkers synthesized new 1D lead bromide perovskites that showed broad-band emission with a photoluminescence characteristic peak around 480 nm. 21 1D perovskites usually demonstrate stronger broad-band emission than 2D perovskites. However, most 1D perovskites to date have exhibited hypsochromically shied white-light emission, with emission peaks ranging from 450 to 550 nm. 22 Therefore, the synthesis of various bathochromically shied 1D perovskite materials is desirable to prepare candidates for UV LEDs employing a single broadband white phosphor.
Until now, the ligands used to prepare low-dimensional perovskites have been large organic ammonium ligands such as C 4 H 9 NH 3 + , CH 3 (CH 2 ) 8 NH 3 + , and 3-(aminomethyl)piperidinium (3AMP + ). 23 Herein we report a crystalline 1D lead halide containing a very new inorganic cation for this purpose, a hexarhenium cluster. Rhenium cluster compounds are known to exhibit broad emission windows from 550 to 1000 nm. These compounds are luminescent both in the solid state and in solution, and can be phosphorescent in the red and near-infrared regions, with long-lived microsecondscale emission upon excitation by UV-Vis light. [24][25][26][27][28][29][30][31][32][33] Although perovskite materials are desired for light-emitting device applications, their poor stability, high sensitivity toward water, and inherent moisture due to the use of organic ammonium cations has restricted their commercialization. Interestingly, rhenium cluster cations are much heavier than organic ammonium cations, and thus using a rhenium cluster as a cation could improve the longterm stability of hybrid materials. Furthermore, hybrid crystals containing rhenium cluster cations and inorganic anions can exhibit synergetic broad-band optical properties. Electronic interactions between the anionic inorganic lead halide framework and emissive rhenium cluster cations can give rise to unusual photophysical properties. As shown in this report, this unique 1D structure results in strong broad-band bathochromically shied emission, peaking at 634 nm with a large Stokes shi.

Preparation of Cs 4 [Re 6 S 8 Br 6 ]$2H 2 O
The starting cluster compound Cs 4 [Re 6 S 8 Br 6 ]$2H 2 O was synthesized from Re 6 S 8 Br 2 as described previously. 34 In detail, stoichiometric quantities of metallic rhenium, sulfur, and cesium bromide were ground, thoroughly mixed, and loaded into a quartz ampoule. The ampoule was evacuated, and bromine liquid was added. The mixture was frozen with liquid nitrogen, and the ampoule was evacuated and sealed. The sealed ampoule was heated to 850 C at the rate of 0.5 C min À1 , held at this temperature for 72 h, and then cooled to room temperature at 6 C min À1 . The product was washed with water and then with acetone, and then dried in an oven at 80 C for 24 h.

Preparation of [Re 6 S 8 (PzH) 6 ]Cl 2 (I)
Base on the synthesis of [Re 6 S 8 (PzH) 6 ]Cl 2 where PzH is pro-ligandspyrazole described previously, 35 1H-pyrazole-1-carboxamidine hydrochloride was used instead of pyrazole as a source of pyrazole ligands and chloride anions. In the details, Cs 4 [Re 6 S 8 Br 6 ]$2H 2 O of 0.2 g (0.083 mmol) and excess 1H-pyrazole-1-carboxamidine hydrochloride of 0.2 g (1.36 mmol) were ground in a mortar, and the resulting mixture was allowed to react by heating it in a sealed glass ampoule at 200 C for two days. The ampoule was then slowly cooled to room temperature and opened. The reaction mixture was washed with diethyl ether to remove excess ligand and then washed with water to remove unreacted starting cluster compounds and the reaction by-product cesium bromide. The remaining solid was dried in an oven at 80 C for 24 h to give the corresponding yellowish rhenium cluster complex in a yield of 80%. Crystals suitable for X-ray structure determination were obtained directly from the mixture aer the reaction. Energy Dispersive X-ray Spectroscopy (EDS) showed a consistent Re : S : Cl ratio of 6.0 : 6.4 : 2.1 (Fig. S1 †). Anal. calc. for C 18

Instrumentation
Absorption spectra were obtained by means of UV-vis spectrometry. Powder X-ray diffraction (PXRD) measurements were collected using a Bruker D8-Focus Bragg-Brentano X-ray powder diffractometer with a Cu Ka radiation source (l ¼ 1.5418Å). PXRD data were acquired over the range 5-60 . The emission spectra of solid {[Re 6 S 8 (PzH) 6 ][Pb 3 I 8 (DMF) 2 ]}$6(DMF) was measured on a Perkin Elmer LS 55 uorescence spectrometer equipped with a xenon lamp at 298 K. Emission spectra were recorded over the range 400-900 nm with an excitation wavelength of 340 nm. EDS measurements were recorded on a SEM-EDS (JEOL, Tokyo, Japan). LC-MS (liquid chromatography-mass spectrometry) was taken on Ultimate 3000 RSLC (Thermo scientic) and Q-exactive orbitrap plus MS (Thermo scientic) in negative condition.

Time-resolved photoluminescence (TR-PL) experiments
Time-resolved PL decay data for all crystal dispersion samples were recorded using a home-built TR-PL decay measurement system. It utilizes an EKSPLA tunable nanosecond pulse laser (NT-342), a Hamamatsu microchannel plate PMT (R3809U-50) detector, Jovin-Yvon H10VIS monochromator, and Tektronix TDS 3052B 500 MHz digital oscilloscope. The typical pulse width of the excitation beam is about 3 ns, and the power is managed to be less than $100 mJ per pulse. The instrument response function (IRF) appeared to be almost a delta function in the current experiments, therefore no deconvolution was necessary for date tting processes. TR-PL decay data were acquired from a home-built soware developed using Labview (National Instruments) and all the solution samples were prepared in a 1 cm quartz cuvette by dissolving crystal samples in dimethylformamide (DMF, anhydrous 99.8%, Sigma Aldrich) at ambient conditions. X-ray crystallography X-ray diffraction data for I and II were collected on a Bruker APX-II diffractometer equipped with a monochromator in an Mo Ka (l ¼ 0.71073Å) incident beam. Each crystal was mounted on a glass ber, and the data for I and II were collected at 223 and 170 K, respectively. The CCD data were integrated and scaled using the Bruker-SAINT soware package, and the structure was solved and rened using SHEXL-2014. All hydrogen atoms were placed in the calculated positions. Table 1 lists the crystallographic data for I and II. Tables S1 and S2 † list selected bond distances and angles for compounds I and II, respectively. Structural information was deposited at the Cambridge Crystallographic Data Centre; the CCDC reference numbers are 2034482 for I and 2034483 for II. †

Crystal structures
In the preparation of [Re 6 S 8 (PzH) 6 6 ] 2+ . During reaction process at high temperature, 1Hpyrazole-1-carboxamidine can be decomposed to release pyrazole ring. However, according to stoichiometry the reaction mixture contains 0.498 mmol of Br À -anions and 1.36 mmol of Cl À -anions. Thus the excess of Cl À -anions leads to the formation of   6 ] 2+ is shown in Fig. 1. The cationic complex [{Re 6 S 8 }(PzH) 6 ] 2+ contains {Re 6 S 8 } 2+ which consists of a Re 6 octahedron residing inside a S 8 cube; that is to say, each face of the Re 6 octahedron is coordinated by a m 3 -S-bridged ligand. This is typical for 24e Re 6 cluster complexes like [Re 6 (m 3 -Q) 8 L 6 ] having 12 two-electron metal-metal bonds. Each Re is additionally coordinated by the nitrogen atom of a pyrazole. Average interatomic distances within the cluster unit are 2.5926(5)Å, 2.412(2)Å, and 2.166(8)Å for the Re-Re, Re-S, and Re-N bonds, respectively (Table S1 †).
The 1D lead iodide hybrid compound {[Re 6 S 8 (PzH) 6 ][Pb 3 I 8 (-DMF) 2 ]}$6(DMF) (II) was prepared by dissolving specic stoichiometric quantities of PbI 2 , MAI, and the rhenium cluster [Re 6 S 8 (PzH) 6 ]Cl 2 (I) in DMF at 80 C for 24 h with vigorous stirring under owing argon gas. A typical wet chemical synthetic method for preparing low-dimensional perovskites involves dissolution of the organic ammonium salt and lead halide salt in a suitable solvent by heating. In the present work, the emissive and air-stable rhenium cluster complex was used for the rst time as the cation, instead of using an organic halide salt, and both the rhenium cluster and the lead iodide were dissolved in DMF. Diethyl ether was added to the clear DMF solution to provide single crystals of II for X-ray crystallographic analysis. All crystals were dried under vacuum and used for further characterizations.
The structure of II is depicted in Fig. 2, and crystallographic data are given in Table 1 Fig. 2). This 1D lead iodide chain has never been reported before and is different from the typical edge-sharing octahedral PbI 6 units. The 1D lead iodide chains are stacked along the a axis, and [Re 6 S 8 (PzH) 6 ] 2+ cations and free DMF molecules reside between the 1D chains. Each pyrazole N-H of [Re 6 S 8 ] 2+ is connected to DMF oxygen atom through H-bonding (N PzH -H/O DMF 1.82-1.88 A) as shown in green dotted lines (Fig. 2).
The structures of both the prepared rhenium cluster (I) as a cation precursor and the 1D lead iodide hybrid (II) were again conrmed by means of powder X-ray diffraction analysis. The  PXRD patterns of the as-prepared I and II were in good agreement with the simulated PXRD patterns (Fig. 3 for I and Fig. 4 for II). The disagreement in the intensity between some PXRD peaks can be attributed to differences in preferred orientation due to the 1D structure.
Scanning electron microscopy (SEM) images of the 1D lead iodide hybrid are shown in Fig. 5. A large number of microrods were obtained in II. This crystal shape is totally different from the microplate shape of the starting rhenium cluster compound (I) (Fig. S4 †). The length of the microrods ranged from several to tens of micrometers.
TGA analysis of crystals of II showed three stages of weight losses (Fig. 6). In the range from 90 to 150 C, a weight loss of about 10% was attributed to a loss of solvate DMF molecules (calc. 10.8%), and a loss about 3% was attributed to loss of coordinated DMF ligands (calc. 3.59%).

Optical properties
Both rhenium cluster complex (I) and 1D lead iodide hybrid (II) dissolved in DMF were excited with laser pulses at the wavelength of 340 nm and optical measurements were carried out at 298 K. Emission spectra were recorded over the wavelength range 400-900 nm. Absorption, excitation and emission of I are shown in Fig. S5. † The emission spectra of compounds I and II in both solution and solid state are similar as shown in Fig. S6. † Brown needle-shaped crystals of II are shown in Fig. 7c. Under UV irradiation (l ¼ 365 nm), the II compound in DMF exhibited a bright orange emission (Fig. 7a). To further characterize the optical properties of II, its UV-vis absorption and photoluminescence (PL) spectra were acquired (Fig. 7b). The UV-vis absorption spectrum displayed sharp absorption ranging from 200 to 450 nm and a weak broad absorption band at about 330 nm, possibly caused by an intramolecular p electron transition in conjugated organic ligand attached hexarhenium cations. 36 Upon excitation at 340 nm, the 1D lead iodide hybrid (II) had a broad band that peaked at 634 nm and spanned from 550 nm to 850 nm, covering the entire visible range (Fig. 7). The luminescence spectra for both I and II were measured in solid     state (Fig. 8). Characteristic emission of the {Re 6 S 8 } core with maximum at 648 nm (blue line in Fig. 8) was observed in compound (I) which was assigned to the rhenium cluster, and the maximum of compound (II) at 634 nm (red line in Fig. 8) has been bathochromically shied. This result indicates that 1D lead iodide contained hybrid compound (II) displays broadband light emission similar to that of rhenium cluster core and a larger bathochromically shied emission due to lead iodide anions overlapped with that of rhenium cluster in 550 range shiing maximum peak of Re cluster to shorter wavelength from 648 nm to 634 nm as shown Fig. 8.
Absorption and emission spectra of the 1D lead halide hybrid showed a large Stokes shi of 310 nm. In 2017, Yuan and coworkers reported one-dimensional organic lead bromide perovskites (C 4 N 2 H 14 PbBr 4 ) that showed broad emission with a maximum at 475 nm. 21 In 2019, Biswas and coworkers synthesized a one-dimensional perovskite [(H 2 O) (C 6 H 8 N 3 ) 2 -Pb 2 Br 10 ] that also gave an efficient bluish white emission at around 560 nm. 37 More recently, a series of one-dimensional face-sharing perovskites were reported by Feng and coworkers; these exhibited broad-band bluish white light emissions with maximum peaks around 500 nm. 36 In the present work, the emission spectrum of a 1D lead halide chain wrapped by rhenium cluster cations showed a strongly enhanced bathochromically shied emission compared with reported 2D and 1D lead halide perovskites including ammonium cations, which usually exhibit characteristic peaks in the violet to green region. 22,23,38 To test the moisture tolerance of compound (II), a solution of II in DMF was exposed to air at room temperature for 3 months.  2 ]}$6(DMF) was also conrmed by powder X-ray diffraction as shown in Fig. 9. This evidence shows that the 1D lead iodide hybrid is stable in ambient atmosphere.
The time-resolved 650 nm PL decay curve of compound (II) collected at 300 K was tted by a double exponential function, giving the average PL lifetime of 5.25 ms, and this was an increased value from 3.77 of compound (I) (Fig. 10). Finally, the band gap was calculated by using Tauc's function on both indirect and direct transition models. Tauc plots gave the best tting for the direct band gap model; thus, the band gap of compound (II) obtained by the direct transition model is about 2.7 eV (Fig. 11). This band gap is characteristic for lowdimensional perovskite materials 39 and the excitonic absorption was stable at room temperature. The band gap was similar to that reported for other 1D perovskites such as CsPbI 3 (3.05 eV) 40

Conclusions
Using for the rst time a rhenium cluster cation as a structuredirecting spacer and a luminophore for an organometallic lead halide, we were able to design and synthesize the novel lead iodide hybrid {[Re 6 S 8 (Pz) 6 ][Pb 3 I 8 (DMF) 2 ]}$6(DMF) with structural architectures based on 1D chains. This 1D lead iodide contained hybrid compound (II) displays broad-band light emission similar to that of rhenium cluster core and a larger bathochromically shied emission due to lead iodide anions overlapped with that of rhenium cluster in 550 range shiing maximum peak of Re cluster to shorter wavelength from 648 nm   to 634 nm. Specically, the emission peaked near 634 nm, an unprecedentedly long wavelength for the lead halide hybrid. This work veries that there is a strong synergetic interaction between anions and cation clusters favorable for exciton selftrapping to produce highly efficient broad-band luminescence. This compound's highly efficient and stable broad-band light emissions are promising for use in light emitters for optoelectronic devices.
This report opens a new path for the synthesis of optically useful materials combining optically active inorganic cations and anions. For the rst time, the rhenium cluster cation was used as a cation to stabilize lead iodide chains, and this material will pave the way to a new class of compounds with great potential for application in photonic devices.

Author contributions
The original idea was conceived by G. T. L. and S.-J. K.; experiments and data analysis were performed by G. T. L., J. C., Y. K., S. K. and S.-H. Y.; the crystal structure of the 1D lead iodide hybrid was analyzed by Y. K.; and the manuscript was draed by G. T. L., J. C., Y. K. and S.-J. K. All authors analyzed the data and reviewed and contributed to the manuscript.

Conflicts of interest
There are no conicts to declare.