Multi-Element Topochemical-Molten Salt Synthesis of One-Dimensional Piezoelectric Perovskite

Summary

One-dimensional perovskites are an interesting material for energy and optoelectronic applications. However, exploring the full wealth of architectures these materials could allow, through multi-element doping of A-sites and B-sites, is still a challenge. Here, we report a high-yield synthetic strategy for 1D perovskites via a two-step method based on a multi-element topochemical-molten salt method. Typically, a high yield of 1D multicomponent perovskite niobates (Li_0.06Na_0.47K_0.47)(Nb_0.94Sb_0.06)O₃ (LNKNS2) is rapidly achieved from as-synthesized 1D K₂(Nb_0.94Sb_0.06)₈O₂₁ with multi-element B-sites. In this process, 1D K₂(Nb_0.94Sb_0.06)₈O₂₁ has been first achieved, and the proportion of the ions in A-sites is affected by the radius and molar ratio of ions. The z axis direction of K₂(Nb_0.94Sb_0.06)₈O₂₁ rod is transformed into the x axis direction of LNKNS2 rod. Furthermore, the output voltage of the 1D niobates-based flexible piezoelectric device (FPD) was nearly 600% compared with that of the isotropic niobates-based FPD. This work also allows convenient fabrication of other 1D multicomponent perovskites.

Graphical Abstract

Highlights

• •A kind of Multi-element Topochemical-Molten Salt method was proposed
• •Large-scale 1D perovskites with multi-doping of A-sites and B-sites was realized
• •The formation and structure of the 1D multi-perovskite niobate were clearly deduced
• •The output voltage of the prepared device was 600% of that of the control sample

Physics; Materials Science; Materials Chemistry; Electronic Materials; Energy Materials

Introduction

One-dimensional (1D) micro-nanomaterials have fascinated wide interest owing to their 2D confinement structure and exotic mechanical, electrical, and optical properties (You et al., 2017, Ning et al., 2018, Duan et al., 2003, Li et al., 2016, Lee et al., 2017, Niu et al., 2017, Arellano et al., 2018, Lu et al., 2018, Yang et al., 2012a, Yang et al., 2012b, Yang et al., 2012c, Yang et al., 2011). For flexible electronic devices (Deutz et al., 2017, Wang, 2018, Wang et al., 2019, Yang et al., 2012a, Yang et al., 2012b, Yang et al., 2012c), which exhibit great significance for viable economic growth and the enhancement of human quality of life, the 1D materials are highly desirable, and a wide range of 1D materials-based flexible devices such as field-effect transistors (Qing et al., 2014), flexible display devices (Wang et al., 2017), flexible photodetectors (Zhou et al., 2018), flexible sensors (Deutz et al., 2017, Rim et al., 2016), and flexible energy storage devices (Gao et al., 2016, Park et al., 2017, Deng et al., 2018, Guo et al., 2012) have been demonstrated to show superior performance than their counterparts with arbitrary shapes.

Perovskite materials, owing to their applications in piezoelectric, ferroelectric, solar cells, and other fields, arouse a research boom (Burschka et al., 2013, Abdi-Jalebi et al., 2018, Becker et al., 2018, Christians et al., 2018, Domanski et al., 2018, Zhang et al., 2018a, Zhang et al., 2018b, Yang et al., 2019, Wu et al., 2012a, Wu et al., 2012b). The structure of perovskite materials is mainly ABO₃-type of cubic or pseudo cubic phase. Generally, the shape of crystalline particles depends on their intrinsic structure, which means that materials with cubic phase usually form isotropic particles (Pribosic et al., 2005). Meanwhile, owing to the demand for various properties, A-sites and B-sites of perovskites are doped with multi-elements mainly via solid-state method (Zheng et al., 2018, Wu et al., 2016, Qin et al., 2016, Li et al., 2018, Liu et al., 2018). 1D perovskites materials, because of their unique electronic, nonlinear optical probe and mechanical properties, have potential in various fields (Ren et al., 2010, Nakayama et al., 2007, Gao et al., 2018, Yang et al., 2012a, Yang et al., 2012b, Yang et al., 2012c, Wu et al., 2012a, Wu et al., 2012b). To further improve performance, it is critical to achieve 1D multi-perovskite materials, which can incorporate 1D structural design into perovskites (Zhai et al., 2018, Sun et al., 2016, Meng et al., 2017, Li et al., 2009, Cheng et al., 2013); however, the preparation of 1D perovskites with multi-doping of A-sites and B-sites at large scale is still a great challenge. At present, only a few synthesis methods of 1D ABO₃ perovskites are reported, which mainly focus on the preparation of A-sites or B-sites as a single element, such as solvothermal method (Zhai et al., 2018), hydrothermal method (Nakayama et al., 2007), reprecipitation method (Sun et al., 2016), sol-gel method (Meng et al., 2017), and molten salt method (Li et al., 2009, Cheng et al., 2013), whereas it is difficult to control the multicomponent composition of perovskites at A-sites and B-sites with these methods.

Perovskite niobate is considered as one of the most competitive lead-free materials to replace lead-bearing perovskite because of its excellent piezoelectric and ferroelectric properties and suitable Curie temperature (Saito et al., 2004, Zhang et al., 2018a, Zhang et al., 2018b, Xu et al., 2016). Herein, we proposed a simple two-step method to synthesize 1D perovskite-type niobate based on a kind of Multi-element Topochemical-Molten Salt (MTMS) method. The key to this approach was that, using 1D anisotropic non-perovskite-type niobate with multi-“B-sites” element as template, large quantities of 1D morphology of perovskite materials at the boundary of quasi-isomorphic phase with controlling multi-doping A-sites and B-sites could be achieved. Accordingly, for the first time, we synthesized the rod-like K₂(Nb_0.94Sb_0.06)₈O₂₁ and proposed the mechanism of its derivation into perovskite-type multicomponent niobate products. Furthermore, flexible piezoelectric device (FPD) with planar orientation of rod-like product was prepared. At the same degree of bending, the output voltage was nearly 600% compared with that of the granular material of similar component. This strategy could be extended to the synthesis of large-scale other 1D multi-perovskite materials, which were expected to be widely used in flexible electronics, sensor devices, and energy storage.

Results and Discussion

The two-step strategy was processed to synthesize 1D multi-perovskite as shown in Figure 1A. Salts were common ionic crystals that could ionize in the molten state. First, salts mixed with a certain proportion of transition metal oxides and 1D anisotropic non-perovskite-type oxides with multi-“B-sites” element was rapidly synthesized via its intrinsic crystal growth in molten salts, and large scale of anisotropic non-perovskite-type oxides was obtained (Figure 1B). Then 1D and multi-dimensional ABO₃ perovskite oxides could be synthesized in large quantities (Figure 1C) by using the anisotropic non-perovskite-type oxides as template, mixed with other reactors and salts, via the MTMS method. The MTMS method is a kind of the multi-element, environmentally friendly, and mild way to prepare pure and morphologically controllable samples at a moderate temperature in a short soaking time, and it is developed from the molten salt method. The morphology of the products can inherit that of the major solid-state raw materials via the MTMS method, that is, the shape and size of the as-synthesized compounds can be controlled by an appropriate choice of raw materials, salts, sintering temperature, and reaction time. The MTMS method has the advantage of combining the molten salt method and the topochemical method, associated with the use of localized solid-state compound transformations via the exchange, deletion, or insertion of different individual ions. In the typical experiment, the non-perovskite-type niobate K₂(Nb_0.94Sb_0.06)₈O₂₁ rods with multi-“B-sites” was prepared by calcinations of corresponding mole ratio of Nb₂O₅ and Sb₂O₃ in equal weight molten KCl salts at 1,000°C for 3 h. The composition could be confirmed by the energy dispersive spectrometry (EDX) pattern (Figure S1), and the reaction could be as follows.

$$3.76{\text{Nb}}_2{\text{O}}_5+0.24{\text{Sb}}_2{\text{O}}_3+2\text{KCl}+0.74{\text{O}}_2\underset{{\text{Na}}_2{\text{CO}}_3}{\overset{\text{KCl}}{\to }}{\text{K}}_2{\left({\text{Nb}}_{0.94}{\text{Sb}}_{0.06}\right)}_8{\text{O}}_{21}+{\text{Cl}}_2\uparrow$$ (Equation 1)

Figure 1.

Synthesis Scheme of 1D Multi-Perovskite via Two Steps of Multi-Element Topochemical-Molten Salt Method

(A) Schematic of the MTMS synthesis of 1D multi-perovskite.

(B) Image of 1D K₂(Nb_0.94Sb_0.06)₈O₂₁ powder with large scale after one time synthesis of first step (see also Figure S1).

(C) Image of 1D (Li_0.06Na_0.47K_0.47) (Nb_0.94Sb_0.06)O₃ powder with large scale after one time preparation (see also Figure S2).

See also Figures S1 and S2.

Then, the multi-niobate (Li_0.06Na_xK_0.94-x)(Nb_0.94Sb_0.06)O₃, marked as LNKNSm, was synthesized from an MTMS reaction by mixing the 0.25 mol K₂(Nb_0.94Sb_0.06)₈O₂₁ rods and 0.06 mol Li₂CO₃ with different amounts of Na₂CO₃ (m mol = 0.5, 1, 1.5, 2, and 2.5 mol) soaked in KCl melts at 800°C for 100 min. The resulting powders were washed several times with distilled water and then dried at 80°C. x is the amount of Na⁺ ions in LNKNSm. The reaction involving the LNKNSm formation was proposed as follows.

$$0.25{\text{K}}_2{\left({\text{Nb}}_{0.94}{\text{Sb}}_{0.06}\right)}_8{\text{O}}_{21}+x{\text{Na}}_2{\text{CO}}_3+0.06{\text{Li}}_2{\text{CO}}_3+\left(1.38-2x\right)\text{KCl}+\left(0.94-x\right){\text{O}}_2\underset{{\text{Na}}_2{\text{CO}}_3}{\overset{\text{KCl}}{\to }}2\left({\text{Li}}_{0.06}{\text{Na}}_x{\text{K}}_{0.94-x}\right)\left({\text{Nb}}_{0.94}{\text{Sb}}_{0.06}\right){\text{O}}_3+\left(0.69-x\right){\text{Cl}}_2\uparrow$$ (Equation 2)

In Equation 2, KCl salt took part in the reaction and the K⁺ ions (radius: 1.33 Å) mainly competed with Na⁺ ions (radius: 1.02 Å) to occupy the A-sites of LNKNSm but had almost no effect on Li⁺ ion (radius: 0.76 Å). This phenomenon might be caused by their different radii and could be confirmed by the X-ray diffraction (XRD) patterns, EDX patterns, and inductively coupled plasma (ICP) measurements of LNKNSm with different value of m (Figures 2A and S2, and Table S1). XRD patterns of LNKNSm indicated the phase evolution with the change of m. Table S2 listed the crystal structure and the lattice parameters of LNKNSm indexed from the XRD patterns. LNKNS0.5, LNKNS1, and LNKNS1.5 were indexed in an orthorhombic symmetry (space group Amm2), LNKNS2 ((Li_0.06Na_0.47K_0.47) (Nb_0.94Sb_0.06)O₃) was indexed in double phases with different orthorhombic symmetries (space groups Amm2 and Pmmm), and LNKNS2.5 was indexed in an orthorhombic symmetry (space group Pmmm). From the lattice parameter changes of the orthorhombic, it could be found that the substitution of some of the Na⁺ for K⁺ results in decreasing lattice constants, since the ionic radius of K⁺ was much larger than that of Na⁺. A phase transition of LNKNSm occurs from an orthorhombic phase (Amm2) to another orthorhombic phase (Pmmm) with an increase in the Na content. It was likely that there was a morphotropic phase boundary (MPB), at which the property of the material could be enhanced largely (Saito et al., 2004), near m = 2 for LNKNSm.

Figure 2.

Structure of the Non-perovskite Niobate and Perovskite Niobate

(A) X-ray diffractograms of the perovskite niobates LNKNSm [m = 0.5, 1.0, 1.5, 2.0, 2.5] (see also Figure S2, Tables S1 and S2).

(B) Rietveld refinement profiles for first-step synthesized non-perovskite niobate K₂(Nb_0.94Sb_0.06)₈O₂₁. Data were refined in the space group Pbam.

(C) Rietveld refinement profiles for the second-step synthesized perovskite niobates (Li_0.06Na_0.47K_0.47)(Nb_0.94Sb_0.06)O₃ (LNKNS2). Data were refined in double space groups Amm2 and Pmmm (for data refined in single space group Amm2, see also Figure S3).

(D) Raman spectra of K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2.

(E) Crystal structure of atomic structure of K₂(Nb_0.94Sb_0.06)₈O₂₁ (see also Tables S3 and S4 and Data S1).

(F) Crystal structure of atomic structure of LNKNS2 (see also Tables S5–S7 and Data S2).

See also Figures S2 and S3, Tables S1–S7, Data S1 and S2.

For further analysis of the detailed structural evolution, the atomic structures of K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2 were calculated and discussed. Figure 2B showed the XRD pattern of the as-synthesized non-perovskite K₂(Nb_0.94Sb_0.06)₈O₂₁, indicating pure phase of the product. The XRD Rietveld method with FULLPROF Suite software package was used to refine the structure, and the phase of K₂(Nb_0.94Sb_0.06)₈O₂₁ was orthorhombic (Pbam) with lattice parameters of a = 37.498 Å, b = 12.509 Å, and c = 3.963 Å, which had similar structure of K₂Nb₈O₂₁ phase (JCPDS 76–977). The detailed structure and atomic parameters of the K₂(Nb_0.94Sb_0.06)₈O₂₁ crystal were shown in Tables S3 and S4 and Data S1 file in Supplemental Information, and the CIF and HKL documents of the K₂(Nb_0.94Sb_0.06)₈O₂₁ crystal was deposited with the deposition number CSD 1914187.

Meanwhile, for further clarification, the Rietveld refinement was done for LNKNS2 with a single (space group Amm2) (Figure S3) and double phase (space groups Amm2 and Pmmm) (Figure 2C). From the better fit of the double phase, compared with that of single phase, it revealed that LNKNS2 was in double phases with different orthorhombic symmetries (Amm2 and Pmmm), and it was confirmed that LNKNS2 was near the MPB. The detailed crystal cell data, atomic parameters, and crystal plane parameters of LNKNS2 with double phase were shown in Tables S5–S7, respectively, the Data S2 file of the LNKNS2 structure was shown in Supplemental Information, and the CIF and HKL documents of the (Li_0.06Na_0.47K_0.47)(Nb_0.94Sb_0.06)O₃ crystal was deposited with the deposition number CSD 1915240.

The structures of K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2 are shown in Figures 2E and 2F, respectively. Both the rods of K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2 had a similar structure of NbO₆ octahedron, which shared corners in the z axis direction for K₂(Nb_0.94Sb_0.06)₈O₂₁ and in the x axis direction for (Li_0.06Na_xK_0.94-x)(Nb_0.94Sb_0.06)O₃ shown in the oblong loop. For the template of K₂(Nb_0.94Sb_0.06)₈O₂₁, the NbO₇ decahedron and NbO₆ octahedron shared with edges and NbO₆ octahedron shared with corners in the xy plane. It only required part of Nb-O bonds breaking and recombination via the entrance of A-sites ions to form the perovskite structure of LNKNS2 with only corner-sharing NbO₆ octahedron in the yz plane. The similar structure of NbO₆ octahedron share with corners was also found in the oblong loop in the xy plane of K₂(Nb_0.94Sb_0.06)₈O₂₁ and the yz plane of LNKNS2. From the above-mentioned analysis, it could be speculated that the z axis direction and xy plane of K₂(Nb_0.94Sb_0.06)₈O₂₁ were transformed into the x axis direction and yz plane of LNKNS2, respectively. The structural evolution was further confirmed by the Raman patterns of K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2 (Figure 2D). The bands observed in the 700–1,000 cm⁻¹ region correspond to the longitudinal optical modes of the Nb-O stretching associated with NbO₆ octahedra. The corresponding transverse optical modes were observed around the 650 cm⁻¹ regions. The weak bands observed in the 350–560 cm⁻¹ region were attributed to be the T_2g mode. The strong peaks observed in the range 200–300 cm⁻¹ were assigned to the T_2u modes. The major band at 700 cm⁻¹ of K₂(Nb_0.94Sb_0.06)₈O₂₁ was the characteristic band for the structure consisting of NbO₆ and NbO₇ octahedra-sharing corners (Blasse and Vandenhe, 1972). All of these results were in agreement with the earlier analysis and calculation of the structure K₂(Nb_0.94Sb_0.06)₈O₂₁ and LNKNS2.

The morphology of the as-synthesized template of K₂(Nb_0.94Sb_0.06)₈O₂₁ rods was examined by scanning electron microscopy (SEM) image, as shown in Figure 3A. A large quantity of rods with diameters of several hundred nanometers and length of several tens of micrometers were observed. The selected area electron diffraction (SAED) patterns taken from the rod of K₂(Nb_0.94Sb_0.06)₈O₂₁ (Figures 3B and 3C) indicated that they were single-crystalline in nature. It was further proved by HRTEM image of the same rod (Figure 3D), and the growth direction of the rods was determined to be along [001]. Meanwhile, the 1D structure of LNKNS2 could be maintained from K₂(Nb_0.94Sb_0.06)₈O₂₁ rods via the MTMS method shown in the SEM and transmission electron microscopy (TEM) (Figures 3E and 3F). The SAED pattern (Figure 3G) taken from the LNKNS2 rod depicted the single-crystalline nature of the obtained perovskite niobate via the MTMS method, which was also proved by HRTEM image of the same rod (Figure 3H). The growth direction of LNKNS2 rods was determined to be along [100], which was also consistent with the earlier speculation of the structural transformation, that is, z axis direction of K₂(Nb_0.94Sb_0.06)₈O₂₁ was transformed into the x axis direction of LNKNS2. The two-step MTMS strategy could be a general method to synthesize a wide range of 1D multi-perovskites, including other multi-perovskite niobate materials with high electric properties (Table 1), which used to be isotropy shape synthesized by the solid-state method (Wu et al., 2016, Hao et al., 2019). The XRD patterns, SEM images, and EDX data of these multi-perovskite niobate rods have been supplied in Figures S4 and S5.

Figure 3.

Morphology of the Non-perovskite Niobate K₂(Nb_0.94Sb_0.06)₈O₂₁ and Perovskite Niobate LNKNS2

(A) The SEM image of the K₂(Nb_0.94Sb_0.06)₈O₂₁ rods.

(B) The TEM image of a K₂(Nb_0.94Sb_0.06)₈O₂₁ rod along the [001] direction.

(C) The SEAD pattern and the crystal plane data of the diffraction point are determined.

(D) The HRTEM image confirmed the growth direction of the rod.

(E) The SEM image of LNKNS2 rods.

(F) The TEM image of an LNKNS2 rod along the [100] direction.

(G) The SEAD pattern and the crystal plane data of the diffraction point are determined by calculation.

(H) The HRTEM image confirmed the growth direction of the LNKNS2 rod.

Table 1.

Morphology and Structure of Some Multi-Perovskite Niobate Materials

Morphology of the Compositions	This Worka	Previous Work
0.96[0.93(K0.5Na0.5)NbO3-0.07LiNbO3]-0.04CaZrO3	1D rodsa	Isotropic particles (Zhang et al., 2015)
0.9625(K0.48Na0.52)(Nb0.94Sb0.06)O3-0.0375Bi0.5(Na0.82K0.18)0.5ZrO3	1D rodsa	Isotropic particles (Qin et al., 2016)
0.96(K0.48Na0.52)(Nb0.95Sb0.05)O3-0.04Bi0.5Na0.5ZrO3	1D rodsa	Isotropic particles (Zheng et al., 2015)
0.9625(K0.45Na0.55)(Nb0.96Sb0.04)O3-0.0375Bi0.5Na0.5Zr0.85Hf0.15O3	1D rodsa	Isotropic particles (Tao and Wu, 2016)
0.964(K0.4Na0.6) (Nb0.955Sb0.045)O3-0.006BiFeO3-0.03Bi0.5Na0.5ZrO3	1D rodsa	Isotropic particles (Wu et al., 2016)

More details in Figures S4 and S5.

This work via the MTMS method, and previous work via the solid-state method.

^aThe XRD patterns, SEM images, and EDX data of the compositions have been supplied in Figures S4 and S5.

To further reveal the property of the as-synthesis perovskite niobate rods, the LNKNS2 rods-based FPD (LNKNS2-FPD) was prepared. The schematic fabrication process of LNKNS2-FPD is shown in Figure S6 (detailed information is given in the Transparent Methods supplemental file). Figure 4A demonstrated the cross-sectional SEM image of the film of LNKNS2 rods with the poly(dimethylsiloxane) (PDMS) matrix via spin-coating, and a magnified cross-sectional SEM image (Figure 4B) showed that the LNKNS2 rods adopt preferential orientations in PDMS. The output of the FPD during the periodic bending/unbending tests was carried out, employing a bending stage executed at a horizontal displacement of 2 cm with a moving speed of one-side movable fixture of 30 mm/s (see Video S1). Figure 4C showed that the output of open-circuit voltage of the LNKNS2-FPD was 600% compared with that of FPD with the isotropic niobates (I-LNKNS-FPD), which had a similar composition [(Li_0.06Na_0.47K_0.47)(Nb_0.94Sb_0.06)O₃] obtained via the solid-state method, for the same bending angle. The XRD pattern, SEM images, and EDX image of the isotropic niobates are shown in Figures S7–S9, respectively. It was interesting that the output voltage of the FPD was remarkably enhanced. The voltage generated by the FPD, V_out, could be calculated from the Equation 3 (Xu et al., 2013, Hwang et al., 2014).

$$V_{\text{out}}=\int g_{33}\cdot \varepsilon \left(l\right)\cdot Ydl$$ (Equation 3)

where l is the perpendicular distance between the adjacent electrodes, ɛ(l) is the function of the strain along the direction of l, Y is the Young's modulus, and g₃₃ is the piezoelectric voltage coefficient. Accordingly, the first merit of LNKNS2 rods-based FPD for piezoelectric applications was their high piezoelectric constant (d₃₃ = 340 pC/N), which increases near 25% of that obtained from isotropic niobates prepared in this work (d₃₃ = 270 pC/N) and that of regular particles with similar composition synthesized via the solid-state method reported in previous study (d₃₃ = 267 pC/N) (Zang et al., 2006). So, the piezoelectric constant corresponded to a high g₃₃ following g₃₃ = d₃₃/ɛ₀·K, where ɛ₀ was the permittivity of free space and K was the relative dielectric constant of the FPD. Meanwhile, the high anisotropic LNKNS2 rods along [100] made the FPD a little textured (Figure 4B), and the identity of the nanowires' longitudinal direction led to an enlargement of the nanowires' deformation (Deutz et al., 2017, Chen et al., 2017), thus resulting in exaggerated dipole displacement and piezoelectric property. For further clarification, a corresponding simulation was carried out via a finite element method, and the numerical modeling result was approximately in agreement with the experimental result. The calculated output of open-circuit voltage of the LNKNS2-FPD and I-LNKNS-FPD was presented in Figure 4D, where piezopotential was depicted by color code. Furthermore, to investigate the effect of the ratio of length to width of the perovskite-type niobates, the niobates (S-LNKNS2) rods with short length were achieved by reducing the amount of salt in first step via the MTMS method (see Figures S10–S12, and Transparent Methods), and the statistical data on the ratio between length and width of the LNKNS2, S-LNKNS, and I-LNKNS were about 1:1, 1:6, and 1:12, respectively. The output of the FPD with S-LNKNS was obtained under the same test condition of LNKNS2-FPD and I-LNKNS-FPD (see Figure S13). It was found that, with the increasing ratio between length and width, the open circuit voltage of the FPD increases. The simulation of piezoelectric voltage output of FPD with particles of different ratios of length to width (1:1, 1:6, and 1:12) was also provided (see Figure S14), which was consistent with the experimental result. Meanwhile, Figure S15 exhibited the clear increase of the output voltage of the LNKNS2-FPD with increasing bending angle. This property is of benefit in the wearable area and other fields.

Figure 4.

The Electric Properties of FPD Based on the As-synthesis Perovskite Niobate Rods

(A) Cross-sectional SEM image of the LNKNS2 rods/PDMS film.

(B) Magnified cross-sectional SEM photograph of the LNKNS2 rods/PDMS film.

(C) The measured open-circuit voltage of an LNKNS2-FPD and I-LNKNS-FPD. The insets are optical images of an FPD in its bending and releasing states.

(D) Simulation results of voltage output of an LNKNS2-FPD and an I-LNKNS-FPD.

Also see Figures S6–S15 and Video S1.

In summary, we proposed a strategy based on MTMS method to achieve large-scale 1D perovskite materials with controlling multi-doping A-sites and B-sites. The formation, structural evolution, quality, and scalability of the 1D multi-perovskite niobate LNKNS2 were clearly highlighted. The z axis direction and xy plane of K₂(Nb_0.94Sb_0.06)₈O₂₁ were transformed into the x axis direction and yz plane of LNKNS2, respectively. The piezoelectric constant (d₃₃) of LNKNS2 increases near 25% of that obtained from the isotropic particles. The output of open-circuit voltage of the LNKNS2-FPD is 600% of that of the I-LNKNS-FPD. Furthermore, the two-step MTMS strategy can be applied generally to a wide range of 1D multi-perovskites, including multi-perovskite niobate materials with high electric properties, to establish a library of 1D multi-perovskites with diverse functionalities. Our study thus provides a robust pathway to the scalable production of 1D multi-perovskites for electronics, piezoelectrics, self-generating, and energy-related fields.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Published: July 26, 2019

Data and Software Availability

The Accession Numbers for the crystallographic entries deposited in the CCDC database: CSD 1914187 for K₂(Nb_0.94Sb_0.06)₈O₂₁ crystal and CSD 1915240 for (Li_0.06Na_0.47K_0.47) (Nb_0.94Sb_0.06)O₃.