Cs2AgInCl6 Double Perovskite Single Crystals: Parity

Article pubs.acs.org/journal/apchd5

Cs2AgInCl6 Double Perovskite Single Crystals: Parity Forbidden Transitions and Their Application For Sensitive and Fast UV Photodetectors Jiajun Luo,†,‡ Shunran Li,†,‡ Haodi Wu,‡ Ying Zhou,‡ Yang Li,‡ Jing Liu,‡ Jinghui Li,‡ Kanghua Li,‡ Fei Yi,§ Guangda Niu,‡ and Jiang Tang*,‡ ‡

Wuhan National Laboratory for Optoelectronics (WNLO) and School of Optical and Electronic Information, Huazhong University of Science and Technology (HUST), 1037 Luoyu Road, Wuhan, Hubei 430074, China § School of Optical and Electronic Information, Huazhong University of Science and Technology (HUST), Wuhan, 430074, China S Supporting Information *

ABSTRACT: Double perovskite Cs2AgInCl6 is newly reported as a stable and environmentally friendly alternative to lead halide perovskites. However, the fundamental properties of this material remain unexplored. Here, we first produced high-quality Cs2AgInCl6 single crystals (SCs) with a low trap density of 8.6 × 108 cm−3, even lower than the value reported in the best lead halide perovskite SCs. Through systematical optical and electronic characterization, we experimentally verified the existence of the proposed parity-forbidden transition in Cs2AgInCl6 and identified the role of oxygen in controlling its optical properties. Furthermore, sensitive (dectivity of ∼1012 Jones), fast (3 dB bandwidth of 1035 Hz), and stable UV photodetectors were fabricated based on our Cs2AgInCl6 SCs, showcasing their advantages for optoelectronic applications. KEYWORDS: double perovskite, single crystals, parity forbidden transitions, surface treatment, UV detection

O

limited to wavelengths shorter than 400 nm, making it attractive for UV light detection. Unlike other indirect band gap double perovskite materials including Cs2B(I)BiX6 and Cs2B(I)SbX6 (B(I) = Ag, Cu, Na),15,16,21 Cs2AgInCl6 exhibits a direct band gap through both first-principles calculations and experimental observations as well as an ultralong carrier lifetime (6 μs),19,21 which makes it a promising competitor to Pb-based halide perovskites for photodetectors. Recently, Yan and co-workers theoretically demonstrated the large difference between experimentally measured optical bandgap of 3.2 eV and photoluminescence emission energy of 2.1 eV is caused by the parity-forbidden transitions, which are due to the centrosymmetry of double perovskite and the lowest conduction band derived from the unoccupied In 5s orbitals.26 Given the new interesting physical and chemical phenomena of Cs2AgInCl6, the optical and electronic properties deserve further investigation. High-quality perovskite single crystals (SCs), which are free of the grain boundaries and noncrystalline domains, are commonly regarded as the ideal platform to study the fundamental properties as well as the surface properties. Furthermore, the low trap density (toward 108 cm−3) renders them more advantageous for applications in the field of solar

rganic−inorganic hybrid lead halide perovskites have emerged as one family of the most promising optoelectronic materials, due to their unique properties such as high absorption coefficients, long carrier diffusion lengths, and low processing cost.1−3 Various lead halide perovskitebased optoelectronic devices have been successfully achieved, including solar cells,4 light emitting diodes,5 and lasers,6 as well as photodetectors for ultraviolet, visible, and near-infrared light detections.7−11 Despite all these excellent properties, Pb-based hybrid halide perovskites suffer from two main issues, the high toxicity of lead12 and the intrinsic instability.13 All these halide perovskites are highly soluble in water and could cause brain related symptoms when Pb poisoned. It is of great significance if Pb could be substituted yet their outstanding performance could be preserved. Currently, Sn2+ has been used to replace Pb2+ in perovskite.14 However, the Sn2+ cations tend to undergo oxidation due to the high-energy-lying 5 s orbitals, rendering the corresponding perovskite extremely unstable in ambient atmosphere. Over the past one year, double perovskites (A2B(I)B′(III)X6) have been proposed as stable and environmental friendly alternatives to lead halide perovskites, with one B(I) and one B′(III) to substitute two toxic Pb2+.15−26 Cs2AgInCl6 is newly reported as a lead-free stable double perovskite (Figure 1a) semiconductor with a wide band gap of 3.2 eV at room temperature.19−21 The absorption of Cs2AgInCl6 is mainly © 2017 American Chemical Society

Received: July 26, 2017 Published: October 30, 2017 398

DOI: 10.1021/acsphotonics.7b00837 ACS Photonics 2018, 5, 398−405

ACS Photonics

Article

Figure 1. Synthesis and optimization of Cs2AgInCl6 SCs. (a) Structure of the ordered double perovskite Cs2AgInCl6. Yellow, pink, cyan, and blue spheres represent In, Ag, Cs, and Cl atoms, respectively. (b, c) Current−voltage curves for the Cs2AgInCl6 SC grown from 0.067 and 0.05 M precursor solution, respectively. Linear and quadratic fitting are applied according to the space charge-limited current (SCLC) model. Top inset in panel b: device structure for SCLC measurement where Cs2AgInCl6 SC was sandwiched by two Au electrodes deposited by thermal evaporation on opposite sides. Insets at the bottom of panels b and c: the photographs of corresponding SCs for SCLC measurements. (d) Absolute transparency of single crystals synthesized with precursor concentration of 0.067 M and 0.05 M, respectively. (e) XRD and Rietveld refined results of Cs2AgInCl6 powders. (f) XRD of Cs2AgInCl6 SCs. The inset is the high-resolution XRD of (222) diffraction.

cells and photodetectors. Especially for UV light detection, the trap states, despite introduce gain and enhance sensitivity, are often responsible for the sluggish response. Traditionally, oxides such as zinc oxide (ZnO) and tin oxide (SnO2) exhibit slow time response of well over one second due to the easy formation of massive oxygen vacancy.27−31 In contrast, the expected low trap density, limited absorption to wavelengths shorter than 400 nm, as well as ultralong carrier lifetime, all make Cs2AgInCl6 SCs an ideal candidate for UV detections such as fire and missile flame detection and optical communications. In this article, we prepared high-quality Cs2AgInCl6 single crystals (SC), achieving an ultralow trap-density as low as (8.6 ± 1.9) × 108 cm−3, even lower than the highest-quality Pbbased perovskites ((1.80 ± 1.07) × 109 cm−3).33 We experimentally verified the existence of parity-forbidden transition in Cs2AgInCl6 and identified the role of oxygen in controlling its optical properties. Furthermore, fast (3 dB cutoff frequency of 1035 Hz), sensitive (ON−OFF ratio of ∼500, measured detectivity of ∼1012 Jones), nontoxic, and stable UV detectors were fabricated based on Cs2AgInCl6 SCs, which compared favorably to most UV detectors reported in the literature.10,11,29−32 Here, we introduced a hydrothermal method to grow highquality Cs2AgInCl6 SCs in a confined space of Teflon autoclave (Figure S1a). The temperature was set to 180 °C to ensure the complete dissolution of precursors. Subsequently, the solution was slowly cooled at 0.5 °C per hour to promote crystal growth. The hydrophobic nature of the Teflon autoclave makes Cs2AgInCl6 nucleate in the solution rather than at the beaker wall, thus reducing the number of nuclei and assuring the crystal large enough for further characterization. As-prepared SCs possessed truncated octahedral shape with an average size of 2.88 mm × 2.81 mm × 1.95 mm and shiny surfaces (Figure

1b,c, inset). Interestingly, all the crystals show light yellow color on the surface but colorless interior, which will be explained later (Figure S1b,c). The crystal quality was optimized through tuning the precursor concentration (0.050 and 0.067 M) due to ultralow solubility of the precursor materials in concentrated HCl (AgCl has a solubility limit of ∼0.02 M at 298 K). The trap density (ntrap) and carrier mobility (μ) was quantitatively studied through space charge limited currents (SCLC) method on the freshly prepared samples2. It should be noted that valid SCLC measurements require: (i) at least one electrode should be ohmic; (ii) the carrier mobility and dielectric constant should be independent of applied voltage; and (iii) surface conducting path should be absent. At low bias voltages, we could clearly see a linear Ohmic region (red line). With increasing bias voltage, the current transited to a trap-filled limit (TFL) region (n > 3), and finally evolved into the square region (n = 2). The onset voltage VTFL (1.5 V for 0.067 M, 6.5 V for 0.050 M) was used to calculate the trap density. The single crystals derived from 0.067 M solution showed an extremely low trap density of (8.6 ± 1.9) × 108 cm−3, much lower than that from 0.050 M solution ((7.3 ± 1.7) × 109 cm−3). It should be noted that the trap density of as-grown Cs2AgInCl6 was even lower than that in most of lead trihalide perovskite SCs.29 Furthermore, from the quadratic Child region (n = 2), we can extract the carrier mobility of Cs2AgInCl6 SCs. For the crystals grown from 0.067 and 0.05 M precursor solution, the mobility was calculated to be 3.31 cm2 V−1 s−1 and 2.29 cm2 V−1 s−1, respectively. The change of crystal quality could also be reflected by the crystal transparency (Figure 1d). The absolute transparency (T2/T1), calculated based on the reflection, scattering, and transmittance of Cs2AgInCl6 SCs, is expected high in a selected wavelength window for a clear crystal with less scattering and absorption (Figure S3). Here we measured the absolute transparency by a spectrophotometer with an 399


ACS Photonics

Article

Figure 2. Identification of the parity-forbidden transitions in Cs2AgInCl6 experimentally. (a) Schematic illustration of the parity-forbidden transitions of Cs2AgInCl6. (b) Absorption spectrum of Cs2AgInCl6 SCs. Weak absorption was observed around 595 nm. Inset: Tauc plot showing the characteristics of a direct band gap of 3.2 eV (384 nm). (c) Photoluminescence spectra (excited by 325 nm wavelength laser) of Cs2AgInCl6 crystal in the range of 350 to 800 nm. (d, e) Time-resolved room-temperature PL decay of Cs2AgInCl6 SCs monitored at 425 and 595 nm.

Figure 3. Effects of oxygen in controlling optical properties on the surface of Cs2AgInCl6 SC. It should be noted that we first conducted UV−O3 treatment on pristine Cs2AgInCl6 SC, and after that we conducted vacuum treatment on UV−O3 treated sample. (a) Absorption spectra of the same Cs2AgInCl6 SC (wavelength between 450 and 600 nm) before and after different treatments. (b) Photoluminescence spectra (excited by 532 nm laser) of Cs2AgInCl6 SC before and after different treatments. (c) XPS spectra of Cs2AgInCl6 SC for O 1s before and after etching. (d) Biasdependent photoconductivity of the same Cs2AgInCl6 SC after different treatments.

integrating sphere, and the measuring method was described elsewhere.29 Clearly, the ratio of T2/T1 of SC from 0.067 M solution was ∼75%, higher than that from 0.050 M solution (∼60%), confirming the crystal quality. All these results show that growing Cs2AgInCl6 SCs in 0.067 M precursor solution yields better crystals with improved optical and electronic properties.

X-ray diffraction (XRD) of the optimized SCs showed a strong diffraction peak of (222) plane, and three weak peaks of (111), (333), and (444) planes, revealing obviously that the single crystal was orientated along the (111) direction. The full width half-maximum (fwhm) of (222) diffraction peak reached 0.05° (180 arcsec), indicating its high single crystallinity (Figure 1e, inset), which is further confirmed by single crystal 400


ACS Photonics

Article

Figure 4. Photoelectric performance of our photoconductive detector based on Cs2AgInCl6 SC in air and vacuum. (a) The current−voltage characteristic of the detector in dark condition. Inset: the device configuration for photoconductive planar detector. (b) Light-to-dark current− voltage (I−V) curves under 365 nm monochromatic illumination generated by a LED. The light power density is 2.36 mW/cm−2. (c) Normalized response of the detector vs the input signal frequency (light intensity of 1.27 mW cm−2) at a bias voltage of 5 V. The −3 dB point is marked with the dash line. (d) Wavelength-dependent responsivity under 5 V bias. The light source is the bromine tungsten lamp modulated by optical grating to generate monochromatic light with a minimum step of 10 nm. (e) Measured dark current noise at various frequencies under the bias of 5 V for device in vacuum. The instrument noise floor, calculated shot noise and thermal noise limit are also included for reference. (f) Light intensitydependent responsivity and normalized detectivity under the bias of 5 V in vacuum.

X-ray diffraction analysis (Figure S4). As shown in Figure 1f, the experimental XRD values and the corresponding Rietveld refined results of Cs2AgInCl6 powder grinded from crystals showed a typical perovskite structure with lattice parameter of a = b = c = 10.478 Å, consistent with literature report.19,21 X-ray photoelectron spectroscopy (XPS) results exhibited the existence of the four elements in as-prepared SCs (Figure S6). We further measured absorption and photoluminescence (PL) of the optimized Cs2AgInCl6 SCs. Based on Yan’s calculation, there are parity-forbidden transitions from valence band maximum (VBM) to conduction band maximum (CBM) due to the same even parity of VBM and CBM, and the same goes for the CBM and VBM-1 located between VBM and VBM-2.26 However, transitions from a lower level VBM-2 to CBM are parity-allowed, which is ∼1.10 eV larger than the fundamental bandgap between VBM and CBM (Figure 2a). As shown in Figure 2b, we indeed observed two absorption edges, with a sharp absorption starting at 384 nm (3.2 eV) while an extremely weak one near 595 nm (2.1 eV). The PL spectrum was obtained using a 325 nm excitation laser through a confocal Raman system (Ocean Optics USB 2000). When we focused the excitation beam on the SC surface, only one PL peak centered at 595 nm was observed. In contrast, when we intentionally focused a high power laser (∼25 mJ cm−2) inside the SC, a strong peak at 595 nm and a relatively weak satellite peak at 425 nm appeared (Figure 2c). To confirm the two transition types, time-resolved roomtemperature PL spectrum was conducted by monitoring 425 and 595 nm peaks. With an excitation laser wavelength of 365 nm, the PL decay curve at 425 nm composed of an initial fast component with a lifetime of 0.9 ns (64.3%), an intermediate component of 2.8 ns (31.8%), and a slow component of 9.8 ns (3.9%). Such short lifetimes are indicative of radiative

recombination between electrons in CBM and holes in VBM2; they also suggested that the photoexcited holes underwent a rapid nonradiative relaxation process from VBM-2 to VBM (Figure 2d). On the contrary, the PL lifetime of Cs2AgInCl6 SCs at 595 nm was much longer (Figure 2e): a fast component of 65.3 ns (97.0%) and a slow component of 566.9 ns (3.0%). The relatively long lifetime was attributed to the parityforbidden transitions from CBM to VBM, which was similar to emission in rare earth ions with PL lifetime in μs to ms scale.35 The different colors and emitting behaviors from the surface and interior of the crystals indicate their different compositions. If soaking the light yellow colored crystal into heated concentrated HCl solution for several hours, the crystal became totally colorless and transparent. In contrary, UV−O3 treatment could deepen the yellow color on surface. As shown in Figure 3a, after UV−O3 treatment, absorption from 450 to 600 nm significantly increased, showing the relaxation of parityforbidden transitions from VBM to CBM. When the same sample was vacuumed further for 90 min, the absorption declined back to some degree. Similarly, 532 nm wavelength laser was used to excite the transitions from VBM to CBM, the PL peak at 595 nm also increased after UV−O3 treatment, and decreased back with further vacuum treatment (Figure 3b), consistent with the absorption results. Furthermore, the photoresponse toward 530 nm wavelength light was also tested on the same crystal, and the light-to-dark current ratio showed a significant increase after the UV−O3 treatment and decreased back with further vacuum treatment (Figure S7). The above phenomenon suggest oxygen or oxygen-containing functional groups altered the surface composition and thus optical properties, which was commonly found in hybrid perovskite34,36,37 and traditional oxides.27 401


ACS Photonics

Article

such increase was from the defects induced gain of photocurrent. The −3 dB cutoff frequency was applied to evaluate the response time of the devices, and the obtained −3 dB point in air and vacuum were 473 Hz (2.11 ms) and 1035 Hz (0.97 ms), respectively (Figure 4c). In addition, when the vacuum device exposed to a rapid repetitive switching of 365 nm light, we extracted a rise and decay time as 0.8 and 1.0 ms, respectively (Figure S9), consistent with the −3 dB cutoff frequency measurement. Such fast response time in vacuum was comparable to previously reported optimal MAPbCl3-based UV detectors (∼1 ms)11 and much faster than the traditional oxides-based devices (∼1 s).27−30 Wavelength-dependent responsivity of our devices were shown in Figure 4d. The responsivity (R) can be calculated as

X-ray photoelectron spectroscopy (XPS) was used to characterize the chemical composition of Cs2AgInCl6 SCs surface. As shown in Figure 3c, the surface of Cs2AgInCl6 SCs showed a strong oxygen 1s peaks compared to the Ar-ions etched sample, indicating high-vacuum environment in XPS measurements could not remove the chemically attached oxygen species completely. The oxygen was from hydroxyl group and oxygen anions considering the fitted binding energy at 531.5 and 533.0 eV, respectively. Moreover, after UV−O3 treatment, Ag 3d and In 3d peaks shifted to lower binding energy (Figure S8), also indicating the formation of Ag−O/ Ag−OH and In−O/In−OH species.38,39 Theoretically, oxygen (O) substitution of chlorine (Cl) either in bulk or on surface could break the local structure of double perovskites and create defect states in the bandgap, which influences the optical properties of Cs2AgInCl6. Besides optical properties, the electronic properties of SCs surface were also crucial for high quality SCs.34,36 To evaluate the influence of adsorbed oxygen on electronic properties, surface recombination velocity s was measured by the photoconductivity method. The measured photocurrent was fitted to a modified Hecht equation:34,40

I=

(

L2

I0μτV 1 − exp − μτV 2

L

1+

Ls Vμ

R = (Ip − Id)/(A ·P)

(2)

Where Ip is the photocurrent, Id is the dark current, A is the active area of the device, and P is the incident light power density. The device demonstrated negligible photoresponse in the wavelength range of 400 to 550 nm wavelength and then presented a steep rising edge located at ∼390 nm. Under vacuum, we could completely suppress the response in the range of 410 to 550 nm, which was from the weak parity forbidden transition. The calculated responsivity toward 365 nm UV light is ∼0.013 A/W at light power density of 10.5 μW/ cm2. Based on the optimal performance in vacuum, the normalized detectivity, used for evaluating the ability to detect weak light of various photodetectors, was calculated by the following equation:

) (1)

where L is the thickness, I0 is the saturated photocurrent, τ is the carrier lifetime, and V is the applied bias. The device structure was the same as SCLC characterization and the illumination light used here was 365 nm LED. The bias dependent photoconductivity of pristine, UV−O3 treated, vacuum treated Cs2AgInCl6 SC are shown in Figure 3d. The surface recombination velocity s of pristine Cs2AgInCl6 was calculated to be 1002 cm s−1, and the UV−O3 treated and vacuum treated Cs2AgInCl6 SC have a s value of 1758 and 1580 cm s−1, respectively, indicating that the oxygen species attached on the SCs surface introduce surface recombination centers. Finally, Cs2AgInCl6 SC-based visible-blind UV detectors with a photoconductive planar structure were fabricated. Two Au electrodes with interval of 200 μm are deposited by thermal evaporation onto the same side of crystal (Figure 4a, inset). Please note our current Cs2AgInCl6 SCs and resultant devices are self-standing. They could, in principle, be integrated with various substrates providing these substrates are acid solution resistive. To evaluate the effect of oxygen on device performance, we conducted all the measurements on the same device in air and vacuum, respectively. As shown in Figure 4a, the current−voltage curves in dark exhibited features expected from good Ohmic contacts. The resistance of Cs2AgInCl6 in vacuum was calculated to be 5.7 × 1011 Ω and decreased to 2.7 × 1010 Ω under air exposure, and such conductivity change was attributed to the oxygen-induced surface conductance. Under vacuum condition, the dark current under 5 V bias reached 10 pA, resulting from the low carrier concentration, wide band gap, and low recombination rates. Such ultralow dark current guarantees detection of weak optical signals and high detectivity. Figure 4b showed the dynamic current−time (I−t) photoresponse under the repetitive switching of 365 nm monochromatic illumination at a bias of 5 V both in air and vacuum. The Ilight and Idark were stable as the reversible and rapid switch from illumination to dark. The photocurrent increased from 5 nA in vacuum to 8 nA in air, and

D* = (Af )1/2 R /in

(3)

where A is the effective area of the detector, f is the electrical bandwidth, and in is the noise current. The noise current, which was associated with the dark current, bandwidth, temperature and various other noise sources, such as shot noise and thermal noise, cannot be derived simply from the dark current of the devices. Here we directly obtained the noise current by a spectrum analyzer. As shown in Figure 4e, the measured noise current of the Cs2AgInCl6 single crystal is approximately 1.55 × 10−15 A/Hz1/2, barely sensitive to frequency, indicating a negligible 1/f noise of our devices due to the low trap density and minimized grain boundaries of our Cs2AgInCl6 SCs. Another two common noise sources, shot noise (in,s) and thermal noise (in,T), were calculated to be 1.27 × 10−15 A Hz−1/2 and 1.29 × 10−16 A Hz−1/2, respectively. in,s =

2eidf

in,T =

4kBTf R

(4)

(5)

where id is the dark current, e is the elementary charge, f is the electrical bandwidth, kB is the Boltzmann constant, T is the temperature, and R is the resistance of the device. Clearly, large resistivity resulted in low dark current and small shot noise. The total noise (in,t) can be calculated by the equation below: in,t =

in,s 2 + in,T 2

(6)

in,t, including contributions from both shot noise and thermal noise, was calculated as 1.40 × 10−15 A Hz−1/2, close to the 402


ACS Photonics

Article

Table 1. Comparison of Device Performance of Our Cs2AgInCl6 UV Photodetectors with Literature Reported Photodetectors photodetectors

dark current (nA)

ZnO SnO2 TiO2 GaN MAPbCl3 film MAPbCl3 SC Cs2AgInCl6 SC

∼0.2 ∼105 1.9 ∼0.01 ∼1 415 0.01

detectivity (Jones) 1.3 × 1013

9.3 × 1011 ∼1012, estimated 1.2 × 1010 ∼1012

response time (ms)

detectivity/response time

ref

>2.0 × 104 >5.0 × 104 6.0 × 103

6.5 × 108

∼1.0 62 0.97

∼1012 1.9 × 108 ∼1012

29 30 31 32 11 10 this work

Figure 5. Stability studies. (a) Thermogravimetric differential thermal analysis of Cs2AgInCl6 powder. (b) PXRD patterns of Cs2AgInCl6 powder after exposure to humidity (55% RH). (c) PXRD patterns of Cs2AgInCl6 powder after exposure to light (5.02 mW/cm−2). (d) Device storage stability of Cs2AgInCl6 devices. Aging condition: ∼25 °C, ∼55% humidity, devices without encapsulation.

measured noise. At last, the calculated D* of our device was 9.60 × 1011 Jones (Figure 4f). Besides, our device performance is compared with a few representative results of UV photodetectors adapted from literatures, as compiled in Table 1. Sensitivity and response speed are the two key parameters for photodetectors, and the product of detectivity and 3 dB bandwidth is often used to compare the performance among different photodectors. Since in most literature papers only response times were reported, we hence use the detectivity-toresponse-time-ratio to evaluate the performance of photodetectors. Traditional oxides, such as ZnO, SnO2, and TiO2, suffer from slow response time. Obviously, our photodector exhibited the highest calculated detectivity-to-response-timeratio among all these devices, showcasing the advantage of using Cs2AgInCl6 SC for UV detection. The chemical and thermal stability of perovskite is another important criterion to assess their potential for practical applications. Thermogravimetric analysis (TGA) showed that Cs2AgInCl6 was stable until 507 °C (Figure 5a), much higher than organic−inorganic hybrid perovskites,41 and no phase transition was observed before decomposition referring to the differential thermal analysis (DTA). The stability against moisture and light was investigated by storing either in the dark at 55% relative humidity for 5 months or irradiated at 50 °C with 365 and 530 nm LED at an intensity of 5.02 mW/cm−2

for 48 h. As shown in Figure 5b,c, no decomposition was found according to the PXRD patterns. Furthermore, we investigated the durability of the Cs2AgInCl6 device, without encapsulation, by recording the photocurrent every 5 days. The device performance maintained 90% of the initial value after storage in ambient (∼25 °C, 55% relative humidity) for 60 days, demonstrating excellent device stability (Figure 5d). The oxygen contamination may influence the stability of devices, however, such degradation could be largely remedied by storing the devices in vacuum to eliminate surface contaminants. In summary, high-quality Cs2AgInCl6 SCs with a low trapstate density ((8.6 ± 1.9) × 108 cm−3) were successfully grown by one-pot hydrothermal method. The existence of parityforbidden transition in Cs2AgInCl6 was experimentally verified, and oxygen was found effective in controlling its optical properties. By eliminating oxygen contamination on the SCs surface in vacuum, we fabricated a Cs2AgInCl6 SC based visibleblind UV detector, with high ON−OFF ratio (∼500), fast photoresponse (∼1 ms), low dark current (∼10 pA at 5 V bias), and high detectivity (∼1012 Jones). Overall, our studies showed that Cs2AgInCl6 double perovskite SCs are promising for UV photodetection because of their nontoxicity, outstanding sensitivity, response time, as well as excellent stability. 403


ACS Photonics

■

Article

(10) Maculan, G.; Sheikh, A. D.; Abdelhady, A. L.; Saidaminov, M. I.; Haque, M. A.; Murali, B.; Alarousu, E.; Mohammed, O. F.; Wu, T.; Bakr, O. M. CH3NH3PbCl3 single crystals: inverse temperature crystallization and visible-blind UV-photodetector. J. Phys. Chem. Lett. 2015, 6, 3781−3786. (11) Adinolfi, V.; Ouellette, O.; Saidaminov, M. I.; Walters, G.; Abdelhady, A. L.; Bakr, O. M.; Sargent, E. H. Fast and Sensitive Solution Processed Visible Blind Perovskite UV Photodetectors. Adv. Mater. 2016, 28, 7264−7268. (12) Babayigit, A.; Ethirajan, A.; Muller, M.; Conings, B. Toxicity of organometal halide perovskite solar cells. Nat. Mater. 2016, 15, 247− 251. (13) Conings, B.; Drijkoningen, J.; Gauquelin, N.; Babayigit, A.; D’Haen, J.; D’Olieslaeger, L.; Ethirajan, A.; Verbeeck, J.; Manca, J.; Mosconi, E. Intrinsic thermal instability of methylammonium lead trihalide perovskite. Adv. Energy Mater. 2015, 5, 1500477. (14) Noel, N. K.; Stranks, S. D.; Abate, A.; Wehrenfennig, C.; Guarnera, S.; Haghighirad, A.; Sadhanala, A.; Eperon, G. E.; Pathak, S. K.; Johnston, M. B. Lead-free organic−inorganic tin halide perovskites for photovoltaic applications. Energy Environ. Sci. 2014, 7, 3061−3068. (15) Slavney, A. H.; Hu, T.; Lindenberg, A. M.; Karunadasa, H. I. A bismuth-halide double perovskite with long carrier recombination lifetime for photovoltaic applications. J. Am. Chem. Soc. 2016, 138, 2138−2141. (16) Filip, M. R.; Hillman, S.; Haghighirad, A. A.; Snaith, H. J.; Giustino, F. Band Gaps of the Lead-Free Halide Double Perovskites Cs2BiAgCl6 and Cs2BiAgBr6 from Theory and Experiment. J. Phys. Chem. Lett. 2016, 7, 2579−2585. (17) Wei, F.; Deng, Z.; Sun, S.; Xie, F.; Kieslich, G.; Evans, D. M.; Carpenter, M. A.; Bristowe, P. D.; Cheetham, A. K. The synthesis, structure and electronic properties of a lead-free hybrid inorganic− organic double perovskite (MA)2KBiCl6. Mater. Horiz. 2016, 3, 328− 332. (18) Wei, F.; Deng, Z.; Sun, S.; Zhang, F.; Evans, D. M.; Kieslich, G.; Tominaka, S.; Carpenter, M. A.; Zhang, J.; Bristowe, P. D. Synthesis and Properties of a Lead-Free Hybrid Double Perovskite:(CH3NH3) 2AgBiBr6. Chem. Mater. 2017, 29, 1089−1094. (19) Volonakis, G.; Haghighirad, A. A.; Milot, R. L.; Sio, W. H.; Filip, M. R.; Wenger, B.; Johnston, M. B.; Herz, L. M.; Snaith, H. J.; Giustino, F. Cs2InAgCl6: A new lead-free halide double perovskite with direct band gap. J. Phys. Chem. Lett. 2017, 8, 772−778. (20) Tran, T. T.; Panella, J. R.; Chamorro, J. R.; Morey, J. R.; McQueen, T. M. Designing Indirect-Direct Bandgap Transitions in Double Perovskites. Mater. Horiz. 2017, 4, 688. (21) Zhou, J.; Xia, Z.; Molokeev, M. S.; Zhang, X.; Peng, D.; Liu, Q. Composition design, optical gap and stability investigations of leadfree halide double perovskite Cs2AgInCl6. J. Mater. Chem. A 2017, 5, 15031. (22) Xiao, Z.; Meng, W.; Wang, J.; Mitzi, D. B.; Yan, Y. Searching for promising new perovskite-based photovoltaic absorbers: the importance of electronic dimensionality. Mater. Horiz. 2017, 4, 206. (23) Xiao, Z.; Yan, Y. Progress in Theoretical Study of Metal Halide Perovskite Solar Cell Materials. Adv. Energy Mater. 2017, 1701136. (24) Xiao, Z.; Du, K. Z.; Meng, W.; Wang, J.; Mitzi, D. B.; Yan, Y. Intrinsic Instability of Cs2In(I)M(III)X6 (M = Bi, Sb; X = Halogen) Double Perovskites: A Combined Density Functional Theory and Experimental Study. J. Am. Chem. Soc. 2017, 139, 6054. (25) Xiao, Z.; Du, K. Z.; Meng, W.; Wang, J.; David, B. M.; Yan, Y. Chemical Origin of the Stability Difference between Copper(I)- and Silver(I)-Based Halide Double Perovskites. Angew. Chem. 2017, 129, 12275. (26) Meng, W.; Wang, X.; Xiao, Z.; Wang, J.; Mitzi, D. B.; Yan, Y. Parity-Forbidden Transitions and Their Impacts on the Optical Absorption Properties of Lead-Free Metal Halide Perovskites and Double Perovskites. J. Phys. Chem. Lett. 2017, 8, 2999. (27) Li, Y.; Della Valle, F.; Simonnet, M.; Yamada, I.; Delaunay, J. Competitive surface effects of oxygen and water on UV photoresponse of ZnO nanowires. Appl. Phys. Lett. 2009, 94, 023110. (28) Liu, K.; Sakurai, M.; Aono, M. Sensors 2010, 10, 8604−8634.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsphotonics.7b00837. Experimental methods; details of all the characterization techniques; photographs; and additional data figures and analysis (PDF).

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Jiang Tang: 0000-0003-2574-2943 Author Contributions †

J.L. and S.L. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the Major State Basic Research Development Program of China (2016YFB0700700, 2016YFA0204000), the National Natural Science Foundation of China (91433105), and the HUST Key Innovation Team for Interdisciplinary Promotion (2016JCTD111). The authors also thank the Analytical and Testing Center of HUST and the facility support of the Center for Nanoscale Characterization and Devices, WNLO. In addition, the authors thank Dr. Zewen Xiao for useful discussions, and the authors also thank Prof. Xing Lu for single crystal X-ray diffraction analysis.

■

REFERENCES

(1) Green, M. A.; Ho-Baillie, A.; Snaith, H. J. The emergence of perovskite solar cells. Nat. Photonics 2014, 8, 506−514. (2) Dong, Q.; Fang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.; Huang, J. Electron-hole diffusion lengths > 175 μm in solution-grown CH3NH3PbI3 single crystals. Science 2015, 347, 967−970. (3) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J. Electron-hole diffusion lengths exceeding 1 micrometer in an organometal trihalide perovskite absorber. Science 2013, 342, 341− 344. (4) Saliba, M.; Matsui, T.; Domanski, K.; Seo, J.; Ummadisingu, A.; Zakeeruddin, S. M.; Correa-Baena, J.; Tress, W. R.; Abate, A.; Hagfeldt, A. Incorporation of rubidium cations into perovskite solar cells improves photovoltaic performance. Science 2016, 354, 206−209. (5) Kim, Y. H.; Cho, H.; Heo, J. H.; Kim, T. S.; Myoung, N.; Lee, C. L.; Im, S. H.; Lee, T. W. Multicolored Organic/Inorganic Hybrid Perovskite Light Emitting Diodes. Adv. Mater. 2015, 27, 1248−1254. (6) Zhu, H.; Fu, Y.; Meng, F.; Wu, X.; Gong, Z.; Ding, Q.; Gustafsson, M. V.; Trinh, M. T.; Jin, S.; Zhu, X. Y. Lead halide perovskite nanowire lasers with low lasing thresholds and high quality factors. Nat. Mater. 2015, 14, 636. (7) Dou, L.; Yang, Y. M.; You, J.; Hong, Z.; Chang, W.; Li, G.; Yang, Y. Solution-processed hybrid perovskite photodetectors with high detectivity. Nat. Commun. 2014, 5, 5404. (8) Dong, R.; Fang, Y.; Chae, J.; Dai, J.; Xiao, Z.; Dong, Q.; Yuan, Y.; Centrone, A.; Zeng, X. C.; Huang, J. High Gain and Low Driving Voltage Photodetectors Based on Organolead Triiodide Perovskites. Adv. Mater. 2015, 27, 1912−1918. (9) Saidaminov, M. I.; Adinolfi, V.; Comin, R.; Abdelhady, A. L.; Peng, W.; Dursun, I.; Yuan, M.; Hoogland, S.; Sargent, E. H.; Bakr, O. M. Planar-integrated single-crystalline perovskite photodetectors. Nat. Commun. 2015, 6, 8724. 404


ACS Photonics

Article

(29) Weng, W. Y.; Chang, S. J.; Hsu, C. L.; Hsueh, T. J. ZnO-based ultraviolet photodetectors. ACS Appl. Mater. Interfaces 2011, 3, 162− 166. (30) Chen, H.; Hu, L.; Fang, X.; Wu, L. A ZnO-nanowire phototransistor prepared on glass substrates. Adv. Funct. Mater. 2012, 22, 1229−1235. (31) Xue, H.; Kong, X.; Liu, Z.; Liu, C.; Zhou, J.; Chen, W.; Ruan, S.; Xu, Q. General Fabrication of Monolayer SnO2 Nanonets for HighPerformance Ultraviolet Photodetectors. Appl. Phys. Lett. 2007, 90, 201118. (32) Chang, P.; Yu, C.; Chang, S.; Lin, Y. C.; Liu, C. H.; Wu, S. L. TiO2 based metal-semiconductor-metal ultraviolet photodetectors. IEEE Sens. J. 2007, 7, 1270−1273. (33) Lian, Z.; Yan, Q.; Gao, T.; Ding, J.; Lv, Q.; Ning, C.; Li, Q.; Sun, J. Perovskite CH3NH3PbI3(Cl) Single Crystals: Rapid Solution Growth, Unparalleled Crystalline Quality, and Low Trap Density toward 108 cm−3. J. Am. Chem. Soc. 2016, 138, 9409−9412. (34) Wei, H.; Fang, Y.; Mulligan, P.; Chuirazzi, W.; Fang, H.; Wang, C.; Ecker, B. R.; Gao, Y.; Loi, M. A.; Cao, L. Sensitive X-ray detectors made of methylammonium lead tribromide perovskite single crystals. Nat. Photonics 2016, 10, 333−339. (35) Aitasalo, T.; Dereń, P.; Hölsä, J.; Jungner, H.; Krupa, J.; Lastusaari, M.; Legendziewicz, J.; Niittykoski, J.; Stręk, W. Persistent luminescence phenomena in materials doped with rare earth ions. J. Solid State Chem. 2003, 171, 114−122. (36) Fang, H.; Adjokatse, S.; Wei, H.; Yang, J.; Blake, G. R.; Huang, J.; Even, J.; Loi, M. A. Ultrahigh sensitivity of methylammonium lead tribromide perovskite single crystals to environmental gases. Sci. Adv. 2016, 2, e1600534. (37) Murali, B.; Yengel, E.; Yang, C.; Peng, W.; Alarousu, E.; Bakr, O. M.; Mohammed, O. F. The Surface of Hybrid Perovskite Crystals: A Boon or Bane. ACS Energy Lett. 2017, 2, 846−856. (38) Kaushik, V. K. XPS core level spectra and Auger parameters for some silver compounds. J. Electron Spectrosc. Relat. Phenom. 1991, 56, 273−277. (39) Fan, J. C.; Goodenough, J. B. X ray photoemission spectroscopy studies of Sn doped indium oxide films. J. Appl. Phys. 1977, 48, 3524− 3531. (40) Stoumpos, C. C.; Malliakas, C. D.; Peters, J. A.; Liu, Z.; Sebastian, M.; Im, J.; Chasapis, T. C.; Wibowo, A. C.; Chung, D. Y.; Freeman, A. J. Crystal growth of the perovskite semiconductor CsPbBr3: a new material for high-energy radiation detection. Cryst. Growth Des. 2013, 13, 2722−2727. (41) Baikie, T.; Fang, Y.; Kadro, J. M.; Schreyer, M.; Wei, F.; Mhaisalkar, S. G.; Graetzel, M.; White, T. J. Synthesis and crystal chemistry of the hybrid perovskite (CH3NH3)PbI3 for solid-state sensitised solar cell applications. J. Mater. Chem. A 2013, 1, 5628− 5641.

405
