Conversion Efficiency Enhancement for - Wiley Online Library

COMMUNICATION Perovskite Solar Cells

www.advmat.de

Slow-Photon-Effect-Induced Photoelectrical-Conversion Efficiency Enhancement for Carbon-Quantum-DotSensitized Inorganic CsPbBr3 Inverse Opal Perovskite Solar Cells Shujie Zhou, Rui Tang, and Longwei Yin* the component of the perovskite materials and fabrication techniques, organic−inorganic hybrid PSCs still suffer from poor stability due to the inferior thermal and moist stability of the mostly used methylammonium lead triiodide (MAPbI3) and formamidinium lead triiodide (FAPbI3).[7] One possible way to solve this problem is to replace the organic cation with an inorganic cation such as cesium.[8] In that case, great effort has been made on inorganic PSCs such as CsPbBr3 and CsPbI3 based PSCs with excellent stability against heat and humidity.[9] In addition, the structure of PSCs usually consists of flat and uniform perovskite films working as the light absorption layer.[2,10] However, the mostly used planar perovskite films greatly limit its light harvesting performance. Therefore, designing uniquely structured perovskite films such as nanowires and nanosheets should be a promising strategy to improve the photoelectronic performance that can integrate the electrical properties such as charge separation rate and emission wavelength.[11] Recently, textured organic–inorganic hybrid perovskite films with inverse opal (IO) morphology have attracted increasing attention for their novel optical features of tunable photonic stop bands and vivid color to naked eyes.[12] The IO structure refers to a 3D ordered photonic crystals consisting of periodically close-packed arrays of voids that are surrounded by solid materials with different refractive index.[13] This periodical change of refractive index results in a photonic band gap (PBG) that affects the propagation of electromagnetic waves and thus influences the optical properties of IO films. Furthermore, IO structure exhibits distinctive slow-photon effect, which would slow down the group velocity of photons at a certain frequency near the PBG, thereby improve the light absorption performance.[14] It is reported that the organic−inorganic hybrid halide perovskites IOs can inherit the advantages of IO such as large surface area, short diffusion length, and sufficient transport tunnels for charge carriers while retain the special optical properties of IO.[15] Also, according to the modified Bragg’s law,[16] by altering the composition, incident angles, and template diameter, the position

All-inorganic cesium lead halide perovskite is suggested as a promising candidate for perovskite solar cells due to its prominent thermal stability and comparable light absorption ability. Designing textured perovskite films rather than using planar-architectural perovskites can indeed optimize the optical and photoelectrical conversion performance of perovskite photovoltaics. Herein, for the first time, this study demonstrates a rational strategy for fabricating carbon quantum dot (CQD-) sensitized all-inorganic CsPbBr3 perovskite inverse opal (IO) films via a template-assisted, spin-coating method. CsPbBr3 IO introduces slow-photon effect from tunable photonic band gaps, displaying novel optical response property visible to naked eyes, while CQD inlaid among the IO frameworks not only broadens the light absorption range but also improves the charge transfer process. Applied in the perovskite solar cells, compared with planar CsPbBr3, slow-photon effect of CsPbBr3 IO greatly enhances the light utilization, while CQD effectively facilitates the electron–hole extraction and injection process, prolongs the carrier lifetime, jointly contributing to a double-boosted power conversion efficiency (PCE) of 8.29% and an increased incident photon-to-electron conversion efficiency of up to 76.9%. The present strategy on CsPbBr3 IO to enhance perovskite PCE can be extended to rationally design other novel optoelectronic devices.

In recent years, organic−inorganic hybrid halide perovskites have attached great attention for the application in photovoltaic devices.[1,2] The power conversion efficiency (PCE) of organic– inorganic perovskite solar cells (PSCs) has exhibited extraordinary boost to more than 20%.[3] The excellent performance of PSCs is mainly attributed to the long carrier lifetime,[4] suitable diffusion length,[5] and high light absorption coefficient.[6] In spite of the rapid increase in PCE closely related to S. Zhou, Dr. R. Tang, Prof. L. Yin Key Laboratory for Liquid–Solid Structural Evolution and Processing of Materials Ministry of Education School of Materials Science and Engineering Shandong University Jinan 250061, P. R. China E-mail: [email protected] The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adma.201703682.

DOI: 10.1002/adma.201703682

Adv. Mater. 2017, 29, 1703682

1703682 (1 of 9)

© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com

www.advmat.de

Scheme 1. Preparation process of CQD/CsPbBr3 IO perovskite solar cells.

of the PBG can be effectively tuned so as to regulate the optical properties of perovskite IOs. However, the previous study related to perovskite IOs is mainly focused on the synthesis method of organic−inorganic hybrid halide perovskite IOs. In addition, how can the slow-photon effect caused by IO structure take effect on the PCE performance of PSCs and how to integrate the optical features of inorganic perovskite IO such as CsPbBr3 IO still remain unclear. Therefore, it is of great importance and challenge to explore inorganic perovskite IOs and find out the mechanism of slow-photon effect on the PCE performance, which is also very meaningful for the future application of wearable or other smart photovoltaic appliances. Recently, carbon quantum dot (CQD) has been attracting great attention due to its novel optical properties,[17] fluorescence features,[18] excellent charge transfer performance,[19,20] and environmental-friendly and nontoxic features.[21] In light of these beneficial features, CQDs have been applied in some solar water splitting and photocatalytic systems to optimize optical and charger transfer properties of semiconductors. For instance, Ye reported CQDs as a visible light sensitizer in bismuth vanadate photoanodes that efficiently broaden the light absorption range and facilitate charge carrier transfer, leading to a dramatically improved solar water splitting performance.[20] In respect of PSCs, Li and co-workers reported the significant enhancement in PCE of CQD-coupled TiO2 as electron transport layer for efficient organic−inorganic hybrid perovskite solar cells, resulting in a boosted PCE owing to the facilitated charge extraction process of CQD.[22] However, most of the enhancement caused by CQDs is applied in planar structures rather than multidimensional structures and the influence of CQDs on inorganic PSCs still remains vague but of great importance. Hence, investigating the photoelectrical and optical properties of CQD-sensitized IO-structured inorganic

Adv. Mater. 2017, 29, 1703682

PSCs can probably fill in the blank of CQD-incorporated inorganic perovskites. Herein, we for the first time synthesized CQD-embedded CsPbBr3 IO structure via a template-assisted spin-coating method and fabricated efficient PSCs based on the structure of FTO/TiO2/CQD-CsPbBr3 IO/spiro-OMeTAD/Ag. The CsPbBr3 IO exhibits tunable optical response performance with iridescent colors to naked eyes arisen from the PBG change of IO structures. Slow-photon effect originated from the periodically ordered frameworks effectively optimizes the light absorption ability by virtue of decelerating the photon group velocity near the PBG. By means of adjusting the composition of CsPbBr3 IO and the incidence angles, tuning PBG position can be evidenced by UV–vis reflectance and transmittance spectrum. CQD effectively helps to broaden the light absorption range of CsPbBr3 and increase the efficiency of charge extraction and injection. Working as light absorption layer of inorganic PSCs, CQDs embedded CsPbBr3 IO exhibited improved light harvesting and efficient charge separation ability that results in obvious enhanced PCE of 8.29% in ambient environment, which is more than two times higher than that of the planar CsPbBr3film-based PSCs. We believe that our work provides a new standpoint in all-inorganic perovskite film for PSCs, which can also be expanded and applied in other photovoltaic devices. As illustrated in Scheme 1 the fabrication process of CQD/ CsPbBr3 IO perovskite solar cells can be divided into six main steps and the detailed experimental process is provided in the Supporting information. Firstly, bl-TiO2 and meso-TiO2 are respectively spin-coated on the FTO substrate, followed by an annealing process. Secondly, double-layered polystyrene (PS) template is assembled on the FTO/bl-TiO2/meso-TiO2 substrate. Thirdly, CsPbBr3 precursor is infiltrated into the

1703682 (2 of 9)



www.advmat.de

Figure 1. a) XRD patterns of CsPbBr3 IO and CQD/CsPbBr3 IO. SEM images of b) PS template, c) CsPbBr3 IO, d) CQD/CsPbBr3 IO, and e) crosssectional SEM image of CQD/CsPbBr3 IO.

voids of PS template and CsPbBr3 is then crystallized in the voids. Fourthly, CQD is dispersed in toluene in advance, followed by immersing the CsPbBr3 infiltrated template into the dispersion for PS removal and CQD loading. After that, hole transport materials (HTM) (spiro-MeOTAD) are spin-coated on the CQD/CsPbBr3 IO, followed by evaporating Ag as the

Adv. Mater. 2017, 29, 1703682

counterelectrode. The fabrication process is concise and universal, which can also be applied to prepare other perovskite inverse opal materials. Figure 1a demonstrates X-ray diffraction (XRD) patterns of CsPbBr3 IO and CQD/CsPbBr3 IO. It can be seen that both CsPbBr3 IO and CQD/CsPbBr3 IO samples show good

1703682 (3 of 9)



www.advmat.de

crystallizability. The characteristic peak around 30° reveals that the obtained CsPbBr3 IO and CQD/CsPbBr3 IO belong to cubic perovskite phase, which can be indexed to (200) plane of cubic CsPbBr3. In addition, the CQD/CsPbBr3 IO sample shows similar XRD pattern to CsPbBr3 IO with no additional peaks, which indicates that coupling of CQD does not change the phase composition of CsPbBr3. The other eight peaks generated by the CsPbBr3 can be indexed to (100), (110), (111), (210), (211), (220), (222), and (400) plane of cubic CsPbBr3 (PDF 54-0752).[23] In addition, due to the poor crystallization of CQD, no characteristic peak of CQD can be witnessed in the XRD pattern, but the existence of CQD can be confirmed by Raman spectra and transmission electron microscopy (TEM) results. Raman spectra are tested to confirm the existence of CQD, as is shown in Figure S1 in the Supporting information. Two obvious peaks located at 1353 and 1584 cm−1 can be assigned to the D-band and G-band peaks of CQD, respectively. As for the CsPbBr3 IO sample, no peaks in the provided region can be witnessed. In addition, the CQD/CsPbBr3 IO sample inherits the D-band and G-band characteristic peaks of CQDs, which can confirm the successful loading of CQD on the CsPbBr3 IO. Figure 1b–e depicts the scanning electron microscopy (SEM) images of PS template, CsPbBr3 IO, and CQD/CsPbBr3 IO. It can be seen from Figure 1b that PS spheres with uniform dia meter of 300 nm orderly arrange on the FTO substrate with a face-centered cubic structure. As is demonstrated in Figure 1c, CsPbBr3 IO inherits the highly ordered macroporous IO structure with clear hexagon structural unit, indicating that the infiltration and template removal method applied here does not destroy the ordered structure arisen from the PS template. As for CQD/CsPbBr3 IO sample in Figure 1d, after coupling with CQD, the surface of the CsPbBr3 IO becomes rough, accom panied by small nanoparticles decorating around the IO frameworks. From the cross-sectional SEM image of CQD/CsPbBr3 IO in Figure 1e, double-layered CQD/CsPbBr3 IO can be distinctly distinguished, which is uniformly and consecutively covered on the substrate. In addition, the elemental mapping images of CQD/CsPbBr3 IO are exhibited in Figure S2 in the Supporting information, showing a homogeneous distribution of Cs, Pb, Br, and C elements. Also, SEM image of CsPbBr3 planar sample is provided in Figure S3 in the Supporting information, demonstrating general surface coverage with some cracks near the boundaries. TEM images in Figure S4 in the Supporting information confirm the morphology and polycrystallinity of CQD and CsPbBr3 IO. It can be seen from Figure S4a in the Supporting information that the as-prepared CQD disperses homogeneously with a diameter less than 10 nm. In addition, the macroporous structure can be witnessed in Figure S4b in the Supporting information with pore diameter of around 300 nm, which is consistent with SEM results. From the high-resolution transmission electron microscopy (HRTEM) image of CQD/CsPbBr3 IO in Figure S4c in the Supporting information, coexistence of CQD and CsPbBr3 IO can be confirmed with clear lattice fringes. The lattice spacing of 0.32 nm can be assigned to the (002) of CQD, while the lattice spacing of 0.29 nm can be attributed to (200) of CsPbBr3. The diffraction rings in Figure S4d in the Supporting information show good agreement with (321), (310), (211), (210), (200), (110) planes of CsPbBr3 and (002) plane of CQD.

Adv. Mater. 2017, 29, 1703682

The as-prepared CsPbBr3 IO exhibits vivid iridescent colors to naked eyes, which is a visualized reflection of the presence of PBG. The close-packed PS template results in a periodic contrast of the refractive index with a periodicity on the order of half a wavelength of visible light, which results in a photonic stop band in the visible light, according to the modified Bragg’s law[24]

λ=2

2 2 2 D n filler f + n voids (1 − f ) − sin 2θ 3

(1)

where λ is the wavelength of the PBG.[25] Photographs of CsPbBr3 IO, CsPbBr2I IO, CsPbBrI2 IO, and CsPbI3 IO are provided in Figure S6 in the Supporting information, displaying different vivid colors from yellow to dark brown visible to eyes. Iridescent color variation from different angles can also be witnessed, which is a direct evidence of PBG. It can be seen in Figure 2b that all the perovskite IOs exhibit an intrinsic absorption edge and an additional absorption peak, the intrinsic absorption edges go through a red shift from 530 nm of CsPbBr3 IO to 700 nm of CsPbI3 IO. The PBGs of the iodinedoped inorganic IOs also exhibit a red shift, from 680 to about 750 nm of CsPbI3 IO. Therefore, it indicates that iodine doping can obviously regulate the optical properties of inorganic perovskite IOs while it also leads to a red shift of PBG. Also, UV–vis spectra of CsPbBr3 planar, CsPbBr2I planar, CsPbBrI2 planar, and CsPbI3 planar are presented in Figure S7 in the Supporting information for comparison, in which only one absorption edge can be witnessed respectively, consistent with the results of the corresponding IO samples. Also, angle-dependent transmittance spectra of CsPbBr3 IO are studied so as to make out the influence of different incident light angles on the PBG positions of CsPbBr3 IO. Figure 2c depicts the angle-dependent UV–vis transmittance spectra of CsPbBr3 IO. It can be seen that the PBG of CsPbBr3 IO is located around 670 nm at normal incidence (θ = 0). There exists a blue shift from 670 to 640 nm (θ = 60°) in the dips of the transmittance spectra when different angles are applied in the transmittance measurement. It can be concluded from the above results that the stop band position of inorganic perovskites can be effectively tuned by adjusting the component and the incidence angle. To understand the comprehensive influence of IO structure and CQD modification on the optical performance of CsPbBr3, the reflectance spectra of CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO samples are tested. As is shown in Figure 2d, it can be seen that CsPbBr3 shows an obvious absorption edge located around 530 nm, which is in accordance with the intrinsic band gap of cubic CsPbBr3. In addition, compared with CsPbBr3 planar sample, absorption edge of CQD/CsPbBr3 planar sample exhibits a slight red shift to about 550 nm, showing that coupling of CQD can broaden the light absorption region of planar CsPbBr3. As for CsPbBr3 IO sample, it possesses an additional absorption peak located at about 670 nm, which can be assigned to the PBG of CsPbBr3 IO. The presence of PBG brings about the slow-photo effect, in that case, group velocity of light can be decreased when incident light wavelength approaches the PBG from the red side, then the path length of light increases,[26] and thus the

1703682 (4 of 9)



www.advmat.de

Figure 2. a) Schematic diagram of the reflection experiment, inset is a digital photograph of CsPbBr3 photonic crystal film. b) UV–vis reflectance spectra of CsPbBr3 IO, CsPbBr2I IO, CsPbBrI2 IO, and CsPbI3 IO. c) Angle-dependent UV–vis transmittance spectra of CsPbBr3 IO. d) UV–vis reflectance spectra of CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO.

light utilization is strongly improved. Importantly, obviously enhanced light absorption intensity can be observed in the IO-structured samples comparing with planar CsPbBr3, which should be attributed to the enhanced light interaction and multiple incident light scattering among the highly periodically 3D macro porous IO structures. In addition, CQD/CsPbBr3 IO sample exhibits similar reflectance curve to CsPbBr3 IO, implying that the loading of CQD does not influence the PBG of inverse opal structure. Meanwhile, after coupling with CQD, the absorption edge of CQD/CsPbBr3 IO also exhibits a minor red shift from 530 to 550 nm. It can be concluded that CQD/ CsPbBr3 IO can effectively enhance the light harvesting ability as well as introduce the slow-photon effect, in turn improving the PCE performance of perovskite solar cells. To investigate the facilitation ability of carrier separation caused by the coupling of CQD, steady-state photoluminescence (PL) spectra and time-resolved photoluminescence (TRPL) spectra are tested. In general, the steady-state PL intensity depends on the recombination of excited electrons and holes, and the lower PL intensity represents more efficient electron–hole separation and lower carrier recombination rate.[27] PL spectra of FTO/TiO2/CsPbBr3 planar, FTO/TiO2/CsPbBr3 IO, and FTO/TiO2/CQD/CsPbBr3 IO are tested at an excitation wavelength of 380 nm to investigate the charge transfer process between the interface of perovskite layer and TiO2 (Figure 3a). Apparently, all the samples exhibit an obvious peak centered at 538 nm with different intensity. It is worth noticing that after coupling with CQD, PL intensity obviously decreases, which

Adv. Mater. 2017, 29, 1703682

is the lowest compared with the other two samples, indicating an efficient separation and injection process of photogenerated carriers. Also, FTO/TiO2/CsPbBr3 IO exhibits lower PL intensity than FTO/TiO2/CsPbBr3 planar, which can be attributed to application of IO structure because well-organized macroporous structure provides abundant tunnels for the interfacial carrier transport. The TRPL spectra of CsPbBr3 planar, CsPbBr3 IO, CQD/ CsPbBr3 IO directly on glass substrate are shown in Figure 3b to determine the carrier lifetime of the photogenerated carriers within the perovskite, which are fitted to a triexponential function and the parameters are listed in Table S1 in the Supporting information. The average carrier lifetime is calculated by the formula[28]

τ avg =

τ 12B1 + τ 22B2 + τ 32B3 τ 1B1 + τ 2B2 + τ 3B3

(2)

where B1, B2, and B3 are relative amplitudes of carrier lifetime, and τ1, τ2, and τ3 are the corresponding fluorescence lifetimes. The average time constant (τavg) can reflect the excited-state decay and free carrier recombination dynamics in the perovskite layers. CQD/CsPbBr3 IO sample has the highest τavg up to 17.47 ns, which is more than seven times higher than that of planar CsPbBr3, indicating photogenerated carriers possess a longer lifetime and higher opportunity to be transferred and collected by TiO2. Coupling CQD can effectively facilitate the carrier extraction process thereby the recombination

1703682 (5 of 9)



www.advmat.de

Figure 3. a) Steady-state PL spectra of FTO/TiO2/CsPbBr3 planar, FTO/TiO2/CsPbBr3 IO, and FTO/TiO2/CQD/CsPbBr3 IO. b) Time-resolved PL spectra of glass/CsPbBr3 planar, glass/CsPbBr3 IO, and glass/CQD/CsPbBr3 IO. c) J–V curves of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO. d) IPCE spectra of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO. e) Partially amplified IPCE spectra of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/ CsPbBr3 IO. f) EIS of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO.

opportunity will be reduced, leading to a relatively longer carrier lifetime.[29] It is also worth noticing that τavg of CsPbBr3 IO (5.96 ns) is longer than planar CsPbBr3 (2.36 ns). In addition, it is reported that τ1 and τ2 represent the fast-decay component, which should be attributed to the bulk recombination in the perovskite crystals, while τ3 stands for the slow-decay component, which should result from the recombination of free carriers in the radiative channels.[30] Notably, for CsPbBr3 IO and CQD/CsPbBr3 IO samples, τ3 is pronouncedly higher than planar sample, which should be ascribed to construction of the 3D inverse opal structures. A schematic graph in Figure S8 in the Supporting information can clearly explain the reason why IO structure can effectively make it easier for charge transfer by providing adequate transfer tunnels. It must be noted that multiscattering process happens in the IO structure, which guarantees an enhanced light harvesting ability compared with

Adv. Mater. 2017, 29, 1703682

planar structures, thereby it can generate more photoinduced carriers. In addition, it can be seen that in planar structures, charge carriers go through chaotic tunnels, which can be easily recombined either at the grain boundaries or consumed by other free carries. While for the CsPbBr3 IO samples, electrons can go through the consecutive periodically arrays and be successfully injected into ETL, which effectively improves the transfer process of charges. Also, after coupling with CQD, the carrier lifetime further increases, which should be ascribed to the effective charge extraction and injection of CQD. PL measurements show that on coupling CQD with CsPbBr3 IO, an efficient carrier transfer process can be obtained with lower carrier recombination rate and longer carrier lifetime. The current density–voltage (J–V) curves of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO have been tested and shown in

1703682 (6 of 9)



www.advmat.de

Table 1. Detailed data of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO. Sample

Voc [V]

Jsc [m A cm−2]

FF

η [%]

CQD/CsPbBr3 IO

1.06

11.34

0.69

8.29

CQD/CsPbBr3 planar

0.92

9.42

0.63

5.45

CsPbBr3 IO

0.93

9.25

0.61

5.25

CsPbBr3 planar

0.82

7.46

0.57

3.48

Figure 3c, and detailed data on the PCE performance of the perovskite solar cells is provided in Table 1. In addition, the cross-sectional SEM image of perovskite solar cells based on CQD/CsPbBr3 IO is shown in Figure S9 in the Supporting information. Clear boundary of bl-TiO2, meso-TiO2, perovskite layer, and HTM layer can be seen. Since the macroporous CQD/ CsPbBr3 IO layer is filled with HTM, it is difficult to figure out the IO structure in the cross-sectional SEM image of the solar cells. Notably, CQD/CsPbBr3 IO exhibits the highest PCE up to 8.29%, which is more than two times higher than the CsPbBr3 planar sample. The boosted PCE comes from an enhanced Voc of 1.06 V and Jsc of 11.34 mA cm−2, which can be ascribed to the promoted light absorption ability arisen from IO structure as well as the facilitated carrier transfer process caused by sensitization of CQD. In addition, CsPbBr3 IO and CQD/CsPbBr3 planar sample shows a PCE of 5.45% and 5.25%, respectively, higher than the CsPbBr3 planar sample, confirming that both IO structure and CQD can work individually on the enhancement of PCE performance of CsPbBr3-based perovskite solar cells. As is illustrated in Figure 3d, incident photon-to-electron conversion efficiency (IPCE) spectra of perovskite solar cells based on CsPbBr3 planar, CQD/CsPbBr3 planar, CsPbBr3 IO, and CQD/CsPbBr3 IO are provided with an inset partial amplified IPCE from 550 to 850 nm. All samples exhibit a wide absorption range from 300 to about 540 nm, which is consistent with the UV–vis spectra. In addition, CQD/CsPbBr3 IO displays the strongest IPCE value up to 76.9%, much higher than CsPbBr3 planar sample. Apparently, obvious additional peaks of CsPbBr3 IO and CQD/CsPbBr3 IO samples can be seen in the partial amplified IPCE spectra (Figure 3e), centered at around 680 nm, which can be assigned to the PBG of IO structure and in good agreement with the UV–vis reflectance results. The additional IPCE peaks once again confirm the enhancement of light harvesting ability resulted from the IO structure. The additional peak of CQD/CsPbBr3 IO is slightly higher than the CsPbBr3 IO sample, owing to the promoted charge extraction and injection process caused by CQD. J–V and IPCE results suggest that the optimization between structure and component can indeed guarantee the enhancement of PCE performance. In addition, in light of the angle-dependent features of the PBG, regional IPCE spectra from 600 to 750 nm with different incident angles are measured (Figure S10, Supporting Information). An obvious blue shift from 680 to 650 nm can be seen of the IPCE peak with the increase of incident angle from 0° to 60° in good agreement with the UV–vis spectra. Also, the IPCE values exhibit a decline with the increase of irradiation angles, which should be attributed to the cosinoidal loss of light

Adv. Mater. 2017, 29, 1703682

intensity. Figure S11 in the Supporting information depicts the calculated integrated current of CQD/CsPbBr3 IO according to the IPCE, with a Jsc of 11.13 mA cm−2, slightly lower than the Jsc tested in J–V curves. Reasons for the minor difference should come from the deviation of the irradiance intensity from solar simulator and IPCE instruments. The electrochemical impedance spectroscopy (EIS) is performed to further investigate the charge transfer performance of the obtained solar cells. The equivalent circuits shown in the inset figure of Figure 3f are fitted with Z-view software, and the fitting data are shown in Table S2 in the Supporting information. The Nyquist plots usually consist of two arcs, the first semicircle denotes the charge exchange process (Rct1) at the counterelectrodes interface, and the second semicircle represents the charge recombination process (Rct2) at the perovskite–TiO2 interface, while Rs refers to the resistance of the substrate.[31] It can been seen from the fitted results that the CQD/CsPbBr3 IO possesses the smallest Rct1 and Rct2 among the four samples, which stands for an efficient charge extraction and injection process at both perovskite–TiO2 interface and counterelectrodes interface. Also, for CQD/CsPbBr3 planar and CsPbBr3 IO samples, Rct1 and Rct2 are lower compared with CsPbBr3 planar samples because of the enhanced light utilization ability of IO structure and facilitated carrier transport process aroused by CQD. The EIS results show good agreement with the PL results. The charge carrier density of the four samples can be analyzed by the Mott–Schottky plot (Figure S12, Supporting Information). The donor density (Nd) can be calculated by the slope of the Mott–Schottky plots via the equation[32] 1 2 = × (Vbi − V ) C 2 A 2qεε 0Nd

(3)

where A is electrode area, q is electronic charge, ε is dielectric constant of the sample, ε0 is permittivity of the vacuum, Nd is donor density, Vbi is built-in potential, and V is applied voltage. The carrier concentration of CQD/CsPbBr3 IO is estimated to be 7.91 × 1018 cm−3, higher than that of CsPbBr3 planar sample (3.16 × 1018 cm−3). In addition, the built-in potential can be estimated using the intercept of the linear regime with the x-axis of Mott–Schottky curves to about 1.02 V,[33] which is in accordance with Voc results in the J–V curves, suggesting that the photogenerated carriers can be separated efficiently by the presence of highly built-in field. The relatively higher built-in potential can not only suppress the back transfer of electrons from the ETL to the perovskite layer, but also benefit the charge collection and transfer of the photogenerated carriers.[34] In order to investigate the energy level relation of the perovskite cells, UPS spectra are used to figure out the valence band (VB) of CsPbBr3 and CQD/CsPbBr3 (Figure S13, Supporting Information). In addition, the band gap of CsPbBr3 and CQD/CsPbBr3 can be calculated by the Kubelka–Munk function-converted plots (Figure S14, Supporting Information). The energy level diagram accompanied by a simplified schematic of the perovskite solar cells is described in Figure 4. An obvious rise in both conductive band (CB) and VB can be witnessed, resulting in an increased driving force of the electron injection from CB of CQD/CsPbBr3 to CB of TiO2. A

1703682 (7 of 9)



www.advmat.de

Figure 4. Schematic energy diagram of the perovskite solar cells based on CQD/CsPbBr3 IO.

facilitated electron–hole extraction and injection process arisen from the matched band structure and increased charge transfer thermodynamic driving force guarantees the promoted PCE performance of the CQD/CsPbBr3 IO-based perovskite solar cells. It is also worth noticing that CQD/CsPbBr3 IO-based PSCs can remain 85% PCE within 45 days in ambient environment (Figure S15, Supporting Information), much stable than the commonly reported MAPbI3-based PSCs.[35] Also, we have tested the normalized PCE under different ambient humidity, as is shown in Figure S16 in the Supporting information. It can be seen that the obtained CQD/CsPbBr3 IO perovskite solar cells can maintain up to 88% PCE under a relative humidity (RH) of 30% and about 60% residual efficiency under 70% RH. Despite an accelerating attenuation in PCE is witnessed with the increase of environment humidity, the stability of CQD/ CsPbBr3 IO perovskite solar cells is still much better than the commonly reported organic perovskite solar cells (MAPbI3 or MAPbBr3).[36] The outstanding performance of stability makes it a promising candidate for the application of PSCs in the future. In summary, we for the first time fabricated carbon quantum dot sensitized all-inorganic CsPbBr3 perovskite inverse opals via a facile template-assisted, spin-coating method. Tunable optical response performance visible to naked eyes comes from the novel optical properties of photonic band gaps of CsPbBr3 IO are witnessed, which can be effectively regulated by means of adjusting the composition and the irradiation angles. Slow-photon effect originated from PBG efficiently strengthens the light utilization efficiency by slowing down the group velocity of photons near the location of PBG. Also, carbon quantum dot embedded among the IO frameworks displays enhanced light absorption ability and facilitates charge transfer process. Perovskite solar cells based on CQD/CsPbBr3 IO exhibit excellent stability and greatly improved PCE of 8.29%, more than two times higher than the planar CsPbBr3, which can be attributed to the optimized light harvesting ability, prolonged carrier lifetime, and facilitated electron–hole extraction and injection process, which can be attributed to the synergetic effects of CQD and CsPbBr3 IO. The present strategy on CsPbBr3 IO to enhance perovskite PCE can be potential to extend to rational design other novel optoelectronic devices.

Adv. Mater. 2017, 29, 1703682

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements The authors acknowledge support from the State Key Program of National Natural Science of China (No.: 51532005), the National Nature Science Foundation of China (No.: 51472148, 51272137), and the Tai Shan Scholar Foundation of Shandong Province.

Conflict of Interest The authors declare no conflict of interest.

Keywords CsPbBr3, carbon quantum dots, inverse opals, perovskite solar cells, photonic band gap Received: July 2, 2017 Revised: August 14, 2017 Published online: October 10, 2017

[1] a) J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin, M. Grätzel, Nature. 2013, 499, 316; b) H. Cho, S.-H. Jeong, M.-H. Park, Y.-H. Kim, C. Wolf, C.-L. Lee, J. H. Heo, A. Sadhanala, N. Myoung, S. Yoo, Science. 2015, 350, 1222. [2] M. Liu, M. B. Johnston, H. J. Snaith, Nature. 2013, 501, 395. [3] a) G. Hodes, Science. 2013, 342, 317; b) W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo, S. I. Seok, Science. 2015, 348, 1234. [4] C. Wehrenfennig, G. E. Eperon, M. B. Johnston, H. J. Snaith, L. M. Herz, Adv. Mater. 2014, 26, 1584. [5] S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza, H. J. Snaith, Science. 2013, 342, 341. [6] W. J. Yin, T. Shi, Y. Yan, Adv. Mater. 2014, 26, 4653. [7] a) J. Liang, C. Wang, Y. Wang, Z. Xu, Z. Lu, Y. Ma, H. Zhu, Y. Hu, C. Xiao, X. Yi, J. Am. Chem. Soc. 2016, 138, 15829; b) C. F. J. Lau,

1703682 (8 of 9)



www.advmat.de

X. Deng, Q. Ma, J. Zheng, J. S. Yun, M. A. Green, S. Huang, A. W. Ho-Baillie, ACS Energy Lett. 2016, 1, 573; c) J. S. Niezgoda, B. J. Foley, A. Z. Chen, J. J. Choi, ACS Energy Lett. 2017, 2, 1043. [8] a) G. E. Eperon, G. M. Paternò, R. J. Sutton, A. Zampetti, A. A. Haghighirad, F. Cacialli, H. J. Snaith, J. Mater. Chem. A. 2015, 3, 19688; b) M. Kulbak, S. Gupta, N. Kedem, I. Levine, T. Bendikov, G. Hodes, D. Cahen, J. Phys. Chem. Lett. 2015, 7, 167. [9] a) B. Li, Y. Zhang, L. Zhang, L. Yin, J. Power Sources. 2017, 360, 11; b) M. Kulbak, D. Cahen, G. Hodes, J. Phys. Chem. Lett. 2015, 6, 2452; c) A. Swarnkar, A. R. Marshall, E. M. Sanehira, B. D. Chernomordik, D. T. Moore, J. A. Christians, T. Chakrabarti, J. M. Luther, Science. 2016, 354, 92; d) Q. A. Akkerman, M. Gandini, F. Di Stasio, P. Rastogi, F. Palazon, G. Bertoni, J. M. Ball, M. Prato, A. Petrozza, L. Manna, Nat. Energy. 2016, 2, 16194. [10] a) N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu, S. I. Seok, Nat. Mater. 2014, 13, 897; b) W. Qiu, T. Merckx, M. Jaysankar, C. M. de la Huerta, L. Rakocevic, W. Zhang, U. Paetzold, R. Gehlhaar, L. Froyen, J. Poortmans, Energy Environ. Sci. 2016, 9, 484. [11] a) A. A. Petrov, N. Pellet, J.-Y. Seo, N. A. Belich, D. Y. Kovalev, A. V. Shevelkov, E. A. Goodilin, S. M. Zakeeruddin, A. B. Tarasov, M. Grätzel, Chem. Mater. 2017, 29, 587; b) P. Li, B. Shivananju, Y. Zhang, S. Li, Q. Bao, J. Phys. D: Appl. Phys. 2017, 50, 094002; c) J. Song, L. Xu, J. Li, J. Xue, Y. Dong, X. Li, H. Zeng, Adv. Mater. 2016, 28, 4861; d) J. Liu, J. Leng, K. Wu, J. Zhang, S. Jin, J. Am. Chem. Soc. 2017, 139, 1432; e) D. Zhang, Y. Yu, Y. Bekenstein, A. B. Wong, A. P. Alivisatos, P. Yang, J. Am. Chem. Soc. 2016, 138, 13155. [12] S. Schünemann, K. Chen, S. Brittman, E. Garnett, H. Tüysüz, ACS Appl. Mater. Interfaces. 2016, 8, 25489. [13] C. I. Aguirre, E. Reguera, A. Stein, Adv. Funct. Mater. 2010, 20, 2565. [14] a) J. Liu, H. Zhao, M. Wu, B. Van der Schueren, Y. Li, O. Deparis, J. Ye, G. A. Ozin, T. Hasan, B. L. Su, Adv. Mater. 2017, 29, 1605349; b) E. Eftekhari, P. Broisson, N. Aravindakshan, Z. Wu, I. S. Cole, X. Li, D. Zhao, Q. Li, J. Mater. Chem. A. 2017, 5, 12803. [15] K. Chen, H. Tüysüz, Angew. Chem., Int. Ed. 2015, 54, 13806. [16] L. Zhang, C. Baumanis, L. Robben, T. Kandiel, D. Bahnemann, Small. 2011, 7, 2714. [17] a) X. Zhang, F. Wang, H. Huang, H. Li, X. Han, Y. Liu, Z. Kang, Nanoscale. 2013, 5, 2274; b) Z. Wang, J. Liu, W. Wang, Z. Wei, F. Wang, P. Gong, J. Wang, N. Li, B. Liu, Z. Zhang, J. Mater. Chem. B. 2017, 5, 4679.

Adv. Mater. 2017, 29, 1703682

[18] Q. Zeng, D. Shao, X. He, Z. Ren, W. Ji, C. Shan, S. Qu, J. Li, L. Chen, Q. Li, J. Mater. Chem. B. 2016, 4, 5119. [19] Y. Zhu, X. Ji, C. Pan, Q. Sun, W. Song, L. Fang, Q. Chen, C. E. Banks, Energy Environ. Sci. 2013, 6, 3665. [20] K. Ye, Z. Wang, J. Gu, Y. Yuan, Y. Zhu, Y. Zhang, W. Mai, S. Yang, Energy Environ. Sci. 2017, 10, 772. [21] Q. Tang, W. Zhu, B. He, P. Yang, ACS Nano. 2017, 11, 1540. [22] H. Li, W. Shi, W. Huang, E.-P. Yao, J. Han, Z. Chen, S. Liu, Y. Shen, M. Wang, Y. Yang, Nano Lett. 2017, 17, 2328. [23] L. Zhou, K. Yu, F. Yang, H. Cong, N. Wang, J. Zheng, Y. Zuo, C. Li, B. Cheng, Q. Wang, J. Mater. Chem. C. 2017, 5, 6224. [24] L. Zhang, C. Y. Lin, V. K. Valev, E. Reisner, U. Steiner, J. J. Baumberg, Small. 2014, 10, 3970. [25] J. Zhou, F. Huang, H. Lin, Z. Lin, J. Xu, Y. Wang, J. Mater. Chem. C. 2016, 4, 7601. [26] X. Chen, J. Ye, S. Ouyang, T. Kako, Z. Li, Z. Zou, ACS Nano. 2011, 5, 4310. [27] a) S. Zhou, L. Yin, J. Alloys Compd. 2017, 691, 1040; b) R. Tang, Z. Xie, S. Zhou, Y. Zhang, Z. Yuan, L. Zhang, L. Yin, ACS Appl. Mater. Interfaces. 2016, 8, 22201. [28] J. H. Heo, H. J. Han, D. Kim, T. K. Ahn, S. H. Im, Energy Environ. Sci. 2015, 8, 1602. [29] D.-Y. Guo, C.-X. Shan, S.-N. Qu, D.-Z. Shen, Sci. Rep. 2014, 4, 7469. [30] P. Zhao, W. Yin, M. Kim, M. Han, Y. J. Song, T. K. Ahn, H. S. Jung, J. Mater. Chem. A. 2017, 5, 7905. [31] R. Tang, R. Yin, S. Zhou, T. Ge, Z. Yuan, L. Zhang, L. Yin, J. Mater. Chem. A. 2017, 5, 4962. [32] M. H. Kumar, S. Dharani, W. L. Leong, P. P. Boix, R. R. Prabhakar, T. Baikie, C. Shi, H. Ding, R. Ramesh, M. Asta, Adv. Mater. 2014, 26, 7122. [33] O. Almora, C. Aranda, E. Mas-Marzá, G. Garcia-Belmonte, Appl. Phys. Lett. 2016, 109, 173903. [34] a) W. A. Laban, L. Etgar, Energy Environ. Sci. 2013, 6, 3249; b) G. Yang, C. Wang, H. Lei, X. Zheng, P. Qin, L. Xiong, X. Zhao, Y. Yan, G. Fang, J. Mater. Chem. A. 2017, 5, 1658. [35] M. Saliba, T. Matsui, J.-Y. Seo, K. Domanski, J.-P. Correa-Baena, M. K. Nazeeruddin, S. M. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt, Energy Environ. Sci. 2016, 9, 1989. [36] T. A. Berhe, W.-N. Su, C.-H. Chen, C.-J. Pan, J.-H. Cheng, H.-M. Chen, M.-C. Tsai, L.-Y. Chen, A. A. Dubale, B.-J. Hwang, Energy Environ. Sci. 2016, 9, 323.

1703682 (9 of 9)
