Jiayu Di1, Haojin Li2, Jie Su1, Haidong Yuan1, Zhenhua Lin1, Kui Zhao1, Jingjing Chang1,3, Yue Hao1. 1. State Key Discipline Laboratory of Wide Band Gap Semiconductor Technology, School of Microelectronics, Xidian University, Xi'an, 710071, China. 2. Key Laboratory of Applied Surface and Colloid Chemistry, National Ministry of Education, Shaanxi Key Laboratory for Advanced Energy Devices, Shaanxi Engineering Lab for Advanced Energy Technology, Institute for Advanced Energy Materials, School of Materials Science and Engineering, Shaanxi Normal University, Xi'an, 710119, China. 3. Advanced Interdisciplinary Research Center for Flexible Electronics, Academy of Advanced Interdisciplinary Research, Xidian University, Xi'an, 710071, China.
Abstract
Generally, growing phase pure CsPbBr3 single crystals is challenging, and CsPb2 Br5 or Cs4 PbBr6 by-products are usually formed due to the different solubilities of CsBr and PbBr2 in the single solvent. Herein, the growth of high-quality phase pure CsPbBr3 perovskite single crystals at room temperature by a humidity controlled solvent evaporation method is reported first. Meanwhile, the room temperature phase transition process from three dimensional (3D) cubic CsPbBr3 to two dimensional (2D) layered tetragonal CsPb2 Br5 and the detailed mechanism induced by humidity are revealed. Moreover, compared with the organic-inorganic perovskite, the prepared CsPbBr3 single crystals are much more stable under high humidity, which satisfies the long-term working conditions of X-ray detectors. The X-ray detectors based on CsPbBr3 single crystals show a high sensitivity and a low detection limit of 1.89 μGyair s-1 , all of which meet the needs of medical diagnosis.
Generally, growing phase pure CsPbBr3 single crystals is challenging, and CsPb2 Br5 or Cs4 PbBr6 by-products are usually formed due to the different solubilities of CsBr and PbBr2 in the single solvent. Herein, the growth of high-quality phase pure CsPbBr3 perovskite single crystals at room temperature by a humidity controlled solvent evaporation method is reported first. Meanwhile, the room temperature phase transition process from three dimensional (3D) cubic CsPbBr3 to two dimensional (2D) layered tetragonal CsPb2 Br5 and the detailed mechanism induced by humidity are revealed. Moreover, compared with the organic-inorganic perovskite, the prepared CsPbBr3 single crystals are much more stable under high humidity, which satisfies the long-term working conditions of X-ray detectors. The X-ray detectors based on CsPbBr3 single crystals show a high sensitivity and a low detection limit of 1.89 μGyair s-1 , all of which meet the needs of medical diagnosis.
The application of lead halide perovskites in the field of optoelectronics has attracted extensive research interest in recent years. As a new semiconductor photoelectric conversion material, it has plenty of advantages, such as high light absorption coefficient,[
,
] long carrier lifetime,[
,
,
,
] low defect state density[
] and high carrier mobility,[
,
] etc. Perovskite single crystals can be prepared by the low‐temperature solution method. Compared with the traditional semiconductor preparation technology, it has the advantages of the simple fabrication method, the controllable cost and the reduced complexity of device integration.[
] Moreover, single crystals of perovskites have better air stability than the thin films prepared by the spin‐coating method, which is more suitable for the long‐term application of devices.[
,
,
]As a conversion device to convert an X‐ray photon signal into an electrical signal, the X‐ray detector is a key part of the X‐ray application system and widely used in medical diagnostic radiotherapy, industrial flaw detection, safety detection, aerospace navigation and material analysis of scientific research, et al.[
,
] The direct‐detection semiconductor X‐ray detector has a better sensitivity and energy resolution than the one using a scintillator.[
] Generally, high‐performance direct‐detection semiconductor X‐ray detectors need to have high carrier mobility and charge carrier lifetime product for ensuring the holes and electrons have sufficient drift length, and high resistivity at room temperature.[
] Furthermore, it also has a large average atomic number (Z) to ensure good absorption of radiation. In the case of certain X‐ray photon energy (E), the absorption coefficient (α) of X‐ray is determined by Z4/E3.[
] The adjustability of elements in perovskite makes it have a large material system and also helps to adjust the average atomic number for X‐ray detection.As for the direct detection X‐ray detector, the lead‐based perovskite semiconductor has made a series of progress due to its advantages of the high X‐ray absorption coefficient and the high carrier collection efficiency, which proves the great potential of the perovskite in the field of X‐ray detection.[
] The nature of all‐inorganic perovskite CsPbBr3 that without the volatile organic molecules, which migrate easily under the electric field, makes its long‐term stability much enhanced.[
,
,
] Meanwhile, the average atomic number of the all‐inorganic perovskite is larger compared to that of the organic–inorganic hybrid perovskite, which means the all‐inorganic perovskite has a higher absorption coefficient to X‐ray.[
] Therefore, all‐inorganic perovskites like CsPbBr3 have been extensively studied for X‐ray detection.[
] Amid them, the sensitivity of the X‐ray detector using CsPbBr3 thin film with the vertical structure of Au/film/ITO can reach 1700 µC Gyair
–1 cm–2,[
] while the highest sensitivity of the vertical Au/CsPbBr3 single crystal/Au X‐ray detector can reach 918 µC Gyair
–1 cm–2.[
] Though CsPbBr3 based X‐ray detectors have achieved much progress, most CsPbBr3 single crystals are grown by the high‐temperature melting method[
,
] and many by‐products (CsPb2Br5 or Cs4PbBr6) appear during the growth process of the low‐temperature solution method,[
,
] which lead to some difficulties in single crystal growth. Thus, changing the factors that affect its growth to simplify the growth method and controlling the by‐product formation is worth investigating by researchers.In this work, we utilize a humidity controlled solvent evaporation method to synthesize various CsPbBr3 perovskite single crystals under the room temperature condition. The experimental results show that CsPbBr3 single crystals of different phases can be obtained from the precursor solution under different environmental humidity conditions. Meanwhile, phase pure transparent 2D tetragonal CsPb2Br5 single crystals could be obtained in the solution by this method under extremely high humidity conditions. While for yellow hexagonal Cs4PbBr6 single crystal (emitting green light under 365 nm light‐emitting diode (LED)), it can be prepared at the preparation temperature of 90–100°C. Finally, the planar X‐ray detector based on orthorhombic CsPbBr3 perovskite single crystals shows a high sensitivity of 6021.99 µC Gyair
–1 cm–2, a low detection limit of 1.89 μGyair s–1, all of which meet the needs of medical diagnosis. Comparing with the organic–inorganic perovskite, CsPbBr3 single crystals are much stable under high humidity, which is convenient for the material to satisfy the long‐term working conditions of X‐ray detectors.
Results and Discussion
CsPbBr3 3D perovskite single crystals were synthesized by a simple solvent evaporation method.[
,
]
Figure
illustrates the two stages of crystal growth during solution evaporation. Similar to the crystal growth process of other materials, the nucleation rates on both sides of the solution‐air interface are different, and the crystal nucleus is initially formed on the solution surface.[
,
,
] When the process continues, the crystal nucleus overcomes the buoyancy, sinks to the bottom of the solution and keeps growing. After 15 d, the CsPbBr3 single crystal of mm‐sized was obtained. It is worth noting that the ambient humidity could affect the crystal growth behavior of CsPbBr3 single crystal. To investigate this strange phenomenon, XRD tests were performed on two shapes of CsPbBr3 single crystals. Under the condition of low humidity (RH ≈ 20%), the single crystal that grows toward the strip shape is cubic CsPbBr3 (the insert picture of Figure 1B), while the single crystal that grows toward the square plane (the insert picture of Figure 1C) is orthorhombic CsPbBr3 growing under the high humidity condition (RH ≈ 60%). Further to increases the humidity (RH > 80%), the orange CsPbBr3 single crystals gradually turn into opaque white crystals (shown in Figure S1A, Supporting Information). Finally, transparent square CsPb2Br5 single crystals will appear in the solution (the insert picture of Figure 1D). The same solution, which contains CsPbBr3 crystals, was placed in an oven with the humidity of 20% as a control experiment. However, the CsPbBr3 orange crystals in the control experiment are not transformed, indicating that the transformation process requires the participation of water. The opaque white crystals, on the other hand, return to orange color immediately after heating on a hot plate at 40 °C (Figure S1B, Supporting Information). Sequently removing it from the hot plate and placing it in an air environment of RH 80%, the orange color of the crystal lightens and then becomes orange and white crystals after a few days (Figure S1C, Supporting Information).
Figure 1
The growth model and crystal type of materials. Schematic diagrams of the growth process of CsPbBr3 perovskite single crystals. A) Primary stage: nucleation process due to the solution surface tension, and growth stage: single crystal sinks to the bottom of the precursors and continue to grow. B) XRD pattern of a typical cubic CsPbBr3 single crystal. The upper left inset is the photograph of a ≈1 × 1 × 5 mm3 orange CsPbBr3 single crystal. C) XRD pattern of the orthorhombic CsPbBr3 single crystal. The upper left inset is the photograph of a ≈3 × 3 × 2 mm3 orange CsPbBr3 single crystal. D) XRD pattern of the CsPb2Br5 single crystal. The upper left inset is the photograph of the transparent ≈1 × 1 × 0.3 mm3 CsPb2Br5 single crystal.
The growth model and crystal type of materials. Schematic diagrams of the growth process of CsPbBr3 perovskite single crystals. A) Primary stage: nucleation process due to the solution surface tension, and growth stage: single crystal sinks to the bottom of the precursors and continue to grow. B) XRD pattern of a typical cubic CsPbBr3 single crystal. The upper left inset is the photograph of a ≈1 × 1 × 5 mm3 orange CsPbBr3 single crystal. C) XRD pattern of the orthorhombic CsPbBr3 single crystal. The upper left inset is the photograph of a ≈3 × 3 × 2 mm3 orange CsPbBr3 single crystal. D) XRD pattern of the CsPb2Br5 single crystal. The upper left inset is the photograph of the transparent ≈1 × 1 × 0.3 mm3 CsPb2Br5 single crystal.Figure 1B shows the X‐ray diffraction (XRD) pattern of the strip shape CsPbBr3 single crystal under RH ≈ 20% condition, containing a series of well‐defined periodically distributed diffraction peaks (101). Moreover, none of the other peaks are detected in the XRD patterns, which confirms its high phase purity for the fabricated cubic CsPbBr3. According to the thermodynamic stability properties of the phase structure reported by Stoumpos et al., the crystal structure of CsPbBr3 is orthorhombic at room temperature, and it could convert to tetragonal phase at 88 ℃ and cubic phase at 130 °C.[
] However, the single‐crystal XRD result of orange CsPbBr3 shows that it belongs to a typical cubic system, with the centrosymmetric space group Pm‐3m (unit cell dimensions a = b = c = 5.857 Å, Table S1, Supporting Information for more details). This result is different from the previous cognition that cubic CsPbBr3 single crystals could also be grown at a low temperature. Figure 1C shows the XRD pattern of orthorhombic CsPbBr3 single crystals growing under RH ≈ 60% condition. Its diffraction peak is consistent with the standard card (PDF#97851), and the crystal direction is along (001) and (110) directions. The single‐crystal XRD result of square CsPbBr3 shows that it belongs to the orthorhombic system, with the centrosymmetric space group Pnma (unit cell dimensions a = 8.2890 Å, b = 11.7075 Å, c = 8.1066 Å, Table S1, Supporting Information for more details). Figure 1D shows the XRD pattern of transparent square single crystals growing under RH ≈ 80% condition, which is perfectly indexed to the diffraction peak of the CsPb2Br5 crystal (PDF#25‐0211). The single‐crystal XRD result of the CsPb2Br5 crystals shows that it belongs to the tetragonal system, with the centrosymmetric space group I4/mcm (unit cell dimensions a = b = 8.4546 Å, c = 15.0987 Å, Table S1, Supporting Information for more details). The corresponding unit cell structures are concluded from the single crystal XRD results and depicted in Figure
with the sequence of cubic CsPbBr3, orthorhombic CsPbBr3, tetragonal CsPb2Br5 and hexagonal Cs4PbBr6.
Figure 2
DFT calculations of phase transitions during the crystal growth. A) The unit cell structures of cubic CsPbBr3, orthorhombic CsPbBr3, tetragonal CsPb2Br5 and hexagonal Cs4PbBr6. B) The formation energies (ΔH) of cubic CsPbBr3 and orthorhombic CsPbBr3 under different conditions. C) The potential barriers (ΔE
p) need to be overcome for the interconversion.
DFT calculations of phase transitions during the crystal growth. A) The unit cell structures of cubic CsPbBr3, orthorhombic CsPbBr3, tetragonal CsPb2Br5 and hexagonal Cs4PbBr6. B) The formation energies (ΔH) of cubic CsPbBr3 and orthorhombic CsPbBr3 under different conditions. C) The potential barriers (ΔE
p) need to be overcome for the interconversion.In order to further explore the differences of the three structures, their band structures were calculated by Perdew–Burke–Ernzerhof functional within the generalized gradient approximation (GGA‐PBE), as shown in Figure S2 (Supporting Information). The bandgaps of cubic CsPbBr3, orthorhombic CsPbBr3 and CsPb2Br5 are 1.83, 2.07, and 3.05 eV, respectively. Besides, the three structures have similar hybrid orbits that the conduction band minimum (CBM) is contributed by Pb 5s orbit and the valence band maximum (VBM) is contributed by Br 5p orbit. Nevertheless, different from the direct band structures of the CsPbBr3 (cubic and orthorhombic phases), CsPb2Br5 shows an indirect band structure, which indicates a lower absorption at the wavelength of around 350 nm. There are also many reports on the crystal formation and degradation of perovskites under the influence of water.[
,
,
,
] Density functional theory (DFT) calculation is used to evaluate the formation energy to further verify the phase transition of crystal growth. The CsBr and PbBr2 unit cells are set as the initial system, and the total energy is 0. First, the formation energies (ΔH) of the cubic CsPbBr3 (ΔH
c) and the orthorhombic CsPbBr3 (ΔH
o) are calculated to be −0.22 and −0.20 eV, respectively. It shows that the initial system can proceed spontaneously in the direction of forming CsPbBr3, which corresponds to the process (i) and (ii) in Figure 2B. According to the experiments, CsBr and PbBr2 are dissolved in dimethyl sulfoxide (DMSO) to grow single crystals. The ΔH
c and ΔH
o that dissolved in DMSO are calculated to be −0.76 and −0.68 eV, respectively. Clearly, ΔH
c is smaller than ΔH
o, indicating that the cubic CsPbBr3 is easier to form in DMSO. This result corresponds to processes (iv) and (v) in Figure 2B. However, when the water (H2O) molecule is added in this process, the ΔH
o is the lowest (−0.93 eV, corresponding to (iii)). It shows that the crystal growth process involving H2O could greatly affect the crystal phase, which reduces the ΔH
o of the orthorhombic CsPbBr3 in DMSO and promotes the reaction to proceed in this direction.In order to explore the transformation process of orange crystals in DMSO to white opaque crystals, the conversion barriers between cubic CsPbBr3, orthorhombic CsPbBr3 and tetragonal CsPb2Br5 are further calculated by DFT. The potential barrier (ΔE
p) for the direct phase transition from cubic CsPbBr3 to orthorhombic CsPbBr3 is 0.78 eV (process Ι in Figure 2C). With the participation of H2O, the ΔE
p is reduced to 0.08 eV (process II). The reason is that Cs+ and Br– combine with OH– and H+ respectively after absorbing H2O, thereby changing the bond length.[
] Without the participation of H2O, two processes are involved when the orthorhombic CsPbBr3 transforms to tetragonal CsPb2Br5:[
] one is the decomposition of CsPbBr3 (ΔE
p = 0.37 eV, process III), and the other is the combination of CsPbBr3 and PbBr2 in the DMSO (ΔE
p = 0.17 eV, process IV). However, with the participation of H2O, this process uses CsPbBr3 as a catalyst for the photolysis of water[
,
,
] to produce H3O+ and OH– to replace part of the Cs+ and Br– in the crystal structure, thereby extracting hydrates and producing CsBr.[
] The ΔE
p required for this process is only 0.03 eV (process V). This result also corresponds to the experimental result that no phase transition occurred in the control group in the oven. Namely, in the process of transforming CsPbBr3 to CsPb2Br5, it is first transformed into orthorhombic CsPbBr3 with the participation of H2O, and then into CsPb2Br5. In addition, the transparent CsPb2Br5 single crystal will not decompose like a white opaque crystal when being heated at 40–120 ℃, which can be explained by the inverse process of (III), (IV), and (V) in Figure 2C. The white opaque substance is the reverse process of (V), in which it only needs to overcome the ΔE
p of 0.03 eV (corresponding to the electron temperature of 0.026 eV at room temperature (300k)). It is easy to overcome and can occur with mild heating. The CsPb2Br5 single crystal undergoes the reverse process of (IV), and the energy required to overcome the ΔE
p is far from being achieved under heating conditions of 40–120 ℃. Therefore, the humidity and heat stability of CsPb2Br5 is good. The reaction process and formula involved in Figure 2B,C are detailed in the supporting information of the formula part.Meanwhile, millimeter‐sized yellow single crystals could appear when the growth temperature is raised to 90–100 °C. The yellow crystal was proved to be Cs4PbBr6 by single‐crystal XRD test (Table S1, Supporting Information for more information). The Cs4PbBr6 belongs to the hexagonal system, with the centrosymmetric space group R‐3c (unit cell dimensions a = b = 13.685 Å, c = 17.279 Å). LED with a wavelength of 365 nm is used to illuminate hexagonal Cs4PbBr6 single crystal, which emits strong green light, while cubic CsPbBr3, orthorhombic CsPbBr3 and tetragonal CsPb2Br5 single crystals have no luminescence phenomenon under the same condition (shown in Figure S3, Supporting Information). CsPb2Br5 consists of inorganic octahedral layers, each of which is neutralized and separated by Cs+. Connectivity modes of octahedron PbI8– in the inorganic layer are connected by face‐sharing. CsPb2Br5 is a 2D layered structure with good humidity and heat stability. However, the size problem of growth still needs to be solved further. Considering the small crystal size of CsPb2Br5 and the instability of the cubic CsPbBr3 at room temperature, a more stable orthorhombic CsPbBr3 single crystal at room temperature and a certain humidity is used for further device fabrication.In the absorption spectrum, the absorption edge of orthorhombic CsPbBr3 powder is approximately at the wavelength of 553 nm, corresponding to an optical bandgap of ≈2.24 eV (Figure
). Moreover, the absorption edge of CsPb2Br5 powder (white opaque crystals) is approximately at the wavelength of 390 nm, corresponding to an optical bandgap of ≈3.18 eV (Figure 3A,C). It is worth noting that CsPb2Br5 has a very low absorption edge at 563 nm, which may be due to the heat generated by grinding CsPb2Br5 crystals, resulting in its decomposition to produce CsPbBr3. The photoluminescence (PL) peak position of orthorhombic CsPbBr3 single crystal is located at 512 nm under a 375 nm laser excitation (Figure 3D), showing a blue shift relative to the absorption edge. Figure 3E shows the time‐resolved PL (TR‐PL) spectroscopy of orthorhombic CsPbBr3 single crystal with the same excited laser. The TR‐PL curve decays into a double exponential form, and the average carrier life (t
av) is 16.96 ns, in which the fast decay lifetime (t
1) and the slow decay life (t
2) are 24.3 ns and 2.3 ns, respectively, accounting for 66.6% and 33.4%. The SEM images of orthorhombic CsPbBr3, cubic CsPbBr3, CsPb2Br5 and Cs4PbBr6 single crystals are shown in Figure 3F‐I. Figure 3G are scanning electron microscope (SEM) images of cubic CsPbBr3 at the scales of 5 µm and 1 mm, respectively, confirming that the CsPbBr3 single crystals grown by the solvent evaporation method at the low temperature have high quality. The size of the whole CsPb2Br5 crystal is small, but its quality is better, as shown in SEM (Figure 3H).
Figure 3
Characterizations of crystal materials. A) UV–vis absorption spectra of the orthorhombic CsPbBr3 powder and CsPb2Br5 powder. B) The Tauc plot for orthorhombic CsPbBr3. (C) The Tauc plot for CsPb2Br5. D) PL spectrum of the orthorhombic CsPbBr3 single crystal excited at 375 nm (FWHM = 27.4 nm). E) Time‐resolved PL (TR‐PL) spectrum of the orthorhombic CsPbBr3 single crystal at 512 nm. The excitation laser beam wavelength is 375 nm. F) SEM image of the orthorhombic CsPbBr3 single crystal with a scale bar of 5 µm. The upper right inset is the SEM image of the orthorhombic CsPbBr3 with a scale bar of 250 µm. G) SEM image of the cubic CsPbBr3 with a scale bar of 5 µm. The upper right inset is the SEM image of the cubic CsPbBr3 with a scale bar of 1 mm. H) SEM image of the CsPb2Br5 single crystal with a scale bar of 15 µm. The upper right insert is the SEM image of the CsPb2Br5 with a scale bar of 250 µm. I) SEM image of the Cs4PbBr6 single crystal with a scale bar of 50 µm. The upper right inset is the SEM image of the Cs4PbBr6 with a scale bar of 500 µm.
Characterizations of crystal materials. A) UV–vis absorption spectra of the orthorhombic CsPbBr3 powder and CsPb2Br5 powder. B) The Tauc plot for orthorhombic CsPbBr3. (C) The Tauc plot for CsPb2Br5. D) PL spectrum of the orthorhombic CsPbBr3 single crystal excited at 375 nm (FWHM = 27.4 nm). E) Time‐resolved PL (TR‐PL) spectrum of the orthorhombic CsPbBr3 single crystal at 512 nm. The excitation laser beam wavelength is 375 nm. F) SEM image of the orthorhombic CsPbBr3 single crystal with a scale bar of 5 µm. The upper right inset is the SEM image of the orthorhombic CsPbBr3 with a scale bar of 250 µm. G) SEM image of the cubic CsPbBr3 with a scale bar of 5 µm. The upper right inset is the SEM image of the cubic CsPbBr3 with a scale bar of 1 mm. H) SEM image of the CsPb2Br5 single crystal with a scale bar of 15 µm. The upper right insert is the SEM image of the CsPb2Br5 with a scale bar of 250 µm. I) SEM image of the Cs4PbBr6 single crystal with a scale bar of 50 µm. The upper right inset is the SEM image of the Cs4PbBr6 with a scale bar of 500 µm.In general, the higher the average atomic number of a material, the more X‐ray absorbability it has. The X‐ray absorption coefficients of CsPbBr3 and several other common semiconductor materials for photons with energy from 10 keV to 10 MeV were calculated by the NIST X‐COM application. It can be seen from Figure
that CsPbBr3 has absorption coefficient much higher than that of MAPbBr3, MAPbCl3 and silicon, indicating that CsPbBr3 has better X‐ray detection capability, although it absorbs fewer X‐rays at about 26–90 keV than commercially available CdTe. Figure 4B is used to determine the thickness of CsPbBr3 single crystal applied to the X‐ray detection of 40 KeV, and it shows that when the crystal thickness reaches 1.7 mm, more than 99% of the X‐ray photons can be absorbed. The carrier mobility and mean lifetime (µτ) product is also one of the factors affecting the X‐ray detection capability of devices, and the value reflects the electrons extraction ability of materials. By using the modified Hecht equation (supporting information equation 1), the photoconductivity curve of CsPbBr3 single‐crystal vertical device can be fitted that its µτ product value is about 5.14 × 10–3 cm2·V–1, as shown in Figure 4C. This value can reach the level of CsPbBr3 single crystals prepared by melt growth method, which reflects the high quality of orthorhombic CsPbBr3 single crystal grown by low‐temperature solvent evaporation method is realized.[
,
] The planar device structure is utilized for the measurement to obtain the X‐ray response characteristics. Figure S4A (Supporting Information) is the I
dark
–V diagram of reverse sweep from 50 to −50 V in the dark state of the device. The I
dark
–t diagram of the device under 5 V bias shows that the I
dark of the device does not drift with the time change, and its value is stable around 21 pA, as shown in Figure S4B (Supporting Information). Figure S5 (Supporting Information) shows the I–V curves of the CsPbBr3 single‐crystal X‐ray detector measured with different dose rates from 103.14 to 5293 μGyair s−1. The I–V curves of the device are smooth under different X‐ray dose rate irradiations and increase gradually with the augments of X‐ray dose rate (with a positive correlation).
Figure 4
Electrical tests of the orthorhombic CsPbBr3 single‐crystal X‐ray Detector. A) Absorption coefficients of CsPbBr3, CdTe, MAPbBr3, MAPbCl3, and silicon as a function of photon energy. B) Attenuation efficiency of CsPbBr3, CdTe, MAPbBr3, MAPbCl3, and silicon for 40 keV X‐ray photons versus thickness. C) Photoconductivity measurement of the CsPbBr3 single crystal device. D) ON/OFF current responses under various bias voltages (from 50 to 5 V) and dose rates (from 852.75 to 103.14 μGyair s–1) of the CsPbBr3 single crystal X‐ray detector. E) X‐ray generated current density versus dose rate under different bias voltages. F) X‐ray sensitivity of the CsPbBr3 single‐crystal X‐ray detector as a function of applied voltage. G) Signal to noise ratio under different X‐ray dose rates of the CsPbBr3 single crystal X‐ray detector (5 V bias voltage). H) The signal‐to‐noise ratio (SNR) of CsPbBr3 single‐crystal device under different bias voltages and dose rates (4.3092, 6.507, and 8.8245 μGyair s–1, respectively). I) CsPbBr3 single‐crystal device responses to X‐ray when turning the X‐ray source on and off. The voltage bias is 5V and the dose rate is 5292.9 μGyair s–1.
Electrical tests of the orthorhombic CsPbBr3 single‐crystal X‐ray Detector. A) Absorption coefficients of CsPbBr3, CdTe, MAPbBr3, MAPbCl3, and silicon as a function of photon energy. B) Attenuation efficiency of CsPbBr3, CdTe, MAPbBr3, MAPbCl3, and silicon for 40 keV X‐ray photons versus thickness. C) Photoconductivity measurement of the CsPbBr3 single crystal device. D) ON/OFF current responses under various bias voltages (from 50 to 5 V) and dose rates (from 852.75 to 103.14 μGyair s–1) of the CsPbBr3 single crystal X‐ray detector. E) X‐ray generated current density versus dose rate under different bias voltages. F) X‐ray sensitivity of the CsPbBr3 single‐crystal X‐ray detector as a function of applied voltage. G) Signal to noise ratio under different X‐ray dose rates of the CsPbBr3 single crystal X‐ray detector (5 V bias voltage). H) The signal‐to‐noise ratio (SNR) of CsPbBr3 single‐crystal device under different bias voltages and dose rates (4.3092, 6.507, and 8.8245 μGyair s–1, respectively). I) CsPbBr3 single‐crystal device responses to X‐ray when turning the X‐ray source on and off. The voltage bias is 5V and the dose rate is 5292.9 μGyair s–1.Detection sensitivity is one of the important parameters to evaluate the performance of an X‐ray detector. Under the condition of the same X‐ray radiation dose rate, the higher the gain current density of the device means the higher detection sensitivity and the better device performance. Figure 4D shows the ON/OFF current responses of the CsPbBr3 single‐crystal X‐ray detector under different voltage biases (from 50 to 5 V) and dose rates (from 852.75 to 103.14 Gyair s–1) to explore the device sensitivity. The ON/OFF current response of CsPbBr3 single crystal devices at 50 V bias voltage is shown in Figure S6 (Supporting Information). The dose rate is changing from 50.364 to 419.85 μGyair s–1. The device continues to switch at 50 V high voltage, maintaining good switching characteristics. The relationship between the gain current density and the radiation dose rate of the CsPbBr3 single‐crystal X‐ray detector, tested at 5, 10, 20, 30, 40, 50 V bias voltage, is shown in Figure 4E. It can be clearly seen that there is an approximately linear relationship between the gain current density of the device and the X‐ray dose rate. According to the slope of the curve, the detection sensitivity of the CsPbBr3 single‐crystal X‐ray detector device at a low voltage bias of 5 V can reach 2549.36 µC Gyair
–1 cm–2. When the applied voltage is increased to 50 V, the sensitivity can reach 6021.99 µC Gyair
–1 cm–2. The high sensitivity of CsPbBr3 is not only related to the structural symmetry of lead‐based perovskite itself, but also due to the unique electronic configuration ns2np0 of Pb2+.[
] Furthermore, compared with vertical or horizontal structure devices based on different materials, this parameter should be among the highest values (Table
is supported for summary comparisons). In addition, the detection sensitivity of the devices at different voltages is shown in Figure 4F.
Table 1
Comparison of the X‐ray detector sensitivity of various materials and device structure response to X‐rays
Comparison of the X‐ray detector sensitivity of various materials and device structure response to X‐rays*SC is the abbreviation of single crystal.The X‐ray radiation can cause radiation damage to creatures. Therefore, another requirement for commercial X‐ray detectors is to pursue extremely low detection limits to reduce radiation loss. Many pieces of literature employ signal‐to‐noise ratio (SNR) = 3 recommended by IUPAC as the signal detection limit.[
,
,
] Based on this standard, SNR can reach above the detection limit at different voltages and different dose rates when testing the CsPbBr3 single‐crystal X‐ray detector, as shown in Figure 4G. Figure 4H shows the SNR of devices with a particular 5 V bias and different dose rates. It is difficult to measure at a lower dose rate. The detection limit of the device can reach 1.89 μGyair s–1 by extending the fitting curve to the direction of low dose rate. This value for CsPbBr3 single‐crystal plane X‐ray detector is only one‐tenth of that of the CsPbBr3 vertical X‐ray detector,[
] which is also lower than the value required for regular medical diagnostics (5.5 μGyair s–1).[
,
] (A comparison of the lowest dose rates of different devices is provided in Table 1.) As is known to all, humidity, oxygen and other factors could affect the stability of perovskite materials and devices. Light also causes instability of perovskite materials, which influences the service life and performance stability of devices. Figure 4I presents the radiation response of a CsPbBr3 X‐ray detector under 5 V bias voltage in ambient air with I
on and I
off corresponding to ≈1.3 nA and ≈20 pA, respectively. It indicates that the device has a large ON/OFF ratio (≈50) and a steady current under a fast switching test at low bias voltage (5 V) and high dose rate (5292.9 μGyair s–1). Figure S7 (Supporting Information) shows the CsPbBr3 single‐crystal X‐ray detector has good operating stability and current remains steady at 5293.9 μGyair s–1 with ± 5 V bias voltage. These two figures show that the device performance is not degraded and still maintains good performance under high‐dose X‐ray irradiation. To sum up, CsPbBr3 is a promising perovskite material for X‐ray detectors.
Conclusion
High‐quality 3D CsPbBr3 perovskite single crystals were synthesized by a simple solvent evaporation method at room temperature. Under high humidity (80%) condition, 3D cubic CsPbBr3 perovskite can be transformed into CsPb2Br5 hydrate during growth, which is thermally unstable. 2D tetragonal CsPb2Br5 single crystal grown directly under RH 80% condition has good humidity and heat stability. Yellow hexagonal Cs4PbBr6 single crystal (emitting green light under 365nm LED) can be prepared at the preparation temperature of 90–100 °C. Most importantly, the planar‐structured CsPbBr3 perovskite single crystal X‐ray detector shows high sensitivity (more than 300 times higher than the state‐of‐the‐art commercial α‐Se vertical counterparts) and a low detection limit of 1.89 μGyair s–1 (need for medical diagnosis is below 5.5 μGyair s–1). Compare with the organic–inorganic perovskite, CsPbBr3 single crystals are still not decomposed under high humidity and have better air stability, which is convenient for the material to satisfy the long‐term use conditions of X‐ray detectors.
Experimental Section
CsPbBr3 Single Crystal Synthesis
CsPbBr3 3D perovskite single crystals were synthesized by a simple solvent evaporation method. 1.5975g CsBr (Advanced Electron Technology Co., Ltd, 99.999%) and 2.7525g PbBr2 (Advanced Electron Technology Co., Ltd, 99.9%) were dissolved in 30 ml DMSO (Sigma‐Aldrich, 99.9%, anhydrous) at a molar ratio of 1:1 to generate 0.25 m CsPbBr3 precursor. The precursors were stirred constantly in the atmospheric environment for 24 h at room temperature (25 °C). After filtration with a 0.45 µm organic filter head, the precursor solution was clarified and transferred to a precleaned crystallizing dish. The crystallizing dish was covered to prevent dust or impurities from falling contamination and to reduce the evaporation rate of the solution to obtain higher quality crystals. After that, the whole unit was placed in the oven and kept at 27 °C. Several days later, CsPbBr3 3D perovskite single crystals were obtained with a size of a few millimeters. The cubic CsPbBr3 single crystals (tended to be a long strip) were grown in the low humidity growing environment. However, with increasing humidity, the growth tended to be an orthorhombic phase crystal structure (growing along a square plane).
Material Characterizations
XRD was performed with a Bruker D8 Discover X‐ray diffractometer with a conventional Cu target X‐ray tube set to 40 kV and 40 mA. Cubic CsPbBr3, orthorhombic CsPbBr3, CsPb2Br5, and Cs4PbBr6 single‐crystal XRD measurements were performed with Bruker D8 Venture with Mo Kα X‐rays. SEM images of cubic CsPbBr3, orthorhombic CsPbBr3, CsPb2Br5, and Cs4PbBr6 were collected with a tungsten filament scanning electron microscope (HITACHI SU‐3500). Absorption spectra were measured using a Perkin–Elmer Lambda 950 UV–vis–NIR spectrophotometer. Steady‐state and time‐resolved PL measurements of cubic CsPbBr3 and orthorhombic CsPbBr3 were taken using a PicoQuant FT‐300 and FT‐100, with 375 nm excitation wavelength. All material characterizations were measured in the air without encapsulation.
Device Fabrication and Characterization
The orthorhombic CsPbBr3 grown in a planar shape was simpler than the cubic CsPbBr3 grown in a stripe shape to make a planar X‐ray detector electrode. The X‐ray detector prepared later used the orthorhombic CsPbBr3. Two kinds of X‐ray devices were prepared by gold electrodes deposition on orthorhombic CsPbBr3 perovskite single crystal for testing. Au/CsPbBr3/Au vertical structure devices (Au thickness is 100 nm, effective area of the device is 0.07 cm2 and crystal thickness is 2 mm, device structure diagram is as shown in Figure S8A, Supporting Information) measured µτ and horizontal structure devices (CsPbBr3 is 3 × 3 × 2 mm3, Au is 100 nm, effective area of the interpolating device is 0.012 mm2, device structure diagram is as shown in Figure S8B, Supporting Information) measured the X‐ray response characterization. Tungsten anode X‐ray tube (DX‐DS2901/24) was used as the source. A Keysight B2902A source table provided the bias voltage and recorded the response current. The X‐ray source operated at a constant voltage of 40 kV. The current was adjusted from 40 to 5 mA to tune the dose rate of the emitted X‐rays. Several pieces of 2 mm thick aluminum foils were inserted between the source and the CsPbBr3 single‐crystal X‐ray detector as attenuators. The X‐ray dose rate was carefully measured using the Fluke Si diode (RaySafe X2 R/F) dosimeter. All X‐ray response characterizations were performed directly in the dark air with optical and electrical shielding to reduce electromagnetic and ambient light interference. All measurements were performed at room temperature.
First‐Principles Calculation
All calculations were carried out by using density functional theory based on the projector‐augmented wave method implemented in the VASP code.[
] The exchange‐correction functional was described by the Generalized Gradient Approximation with the Perdew–Burke–Ernzerhof functional.[
] The transformation barriers were calculated using the climbing image nudged elastic band (CI‐NEB) method through the VTST tools.[
,
] The plane‐wave cutoff energy was set to be 400 eV. The Monkhorst−Pack k‐point mesh is sampled with a separation of 0.05 and 0.015 Å−1 in the Brillouin zone. All structures were relaxed until the residual force on each atom was less than 0.01 eV Å−1. The self‐consistent convergence accuracy was set at 10–5 eV per atom in the structural calculation, and 10–7 eV per atom in the CI‐NEB calculation. The formation energies were obtained from the total energy variations of the chemical processes listed in the Supporting Information.
Conflict of Interest
The authors declare no conflict of interest.Supporting InformationClick here for additional data file.
Authors: Long Zhou; Jie Su; Zhenhua Lin; Xing Guo; Jing Ma; Tao Li; Jincheng Zhang; Jingjing Chang; Yue Hao Journal: Research (Wash D C) Date: 2021-06-28