Perovskites’ ability to change light into electricity has recently gotten a lot better, hitting up to 25.5% in solar cells. Their optoelectronic properties, such as their high absorption coefficient, long carrier diffusion lengths, high charge carrier mobility, and ability to create free charge carriers by photoexcitation, make them good for photodetectors, which are devices that sense light by turning optical signals into electrical signals. Next-generation perovskite photodetectors are becoming more and more common because they can be used for spectroscopy, picture sensing, biological/chemical/radiation sensing, tracking the environment, and fiber-optic communication. These photodetectors have great features, like a quick response time, a high signal, a large linear response range, low noise, and a response that can be controlled and depends on wavelength.
Why Perovskites for Photodetectors
Silicon is the most common type of visible light photodetector used in business, but it is also very sensitive to light that isn’t visible. On the other hand, perovskite photodetectors have higher absorption coefficients and don’t react to IR light. This means that they can sense light without the need for extra optical filters. They take in a lot more photogenerated charge carriers per unit of thickness. This increases the external quantum efficiency (EQE) and lets a film that is only a few hundred nanometres thick almost completely capture light that hits it.
The electrical output per visual input is what responsiveness (R) measures. It tells you how well a photodetector reacts to different colours of light. A photodetector works best when it is measuring light at or near the wavelength with the highest responsivity. This gives it the biggest signal-to-noise ratio (SNR) and sensitivity. Most of the time, photon energies just above the bandgap of the photoactive material equate to maximum responsivity.
Because they have a straight bandgap, perovskites don’t have a soft absorption edge like silicon does. Instead, they have clear areas of high or low responsivity. Perovskite photodetectors also have a response range that can be easily changed. Changing the ABX3 composition changes the bandgap and moves the response range from ultraviolet to near infrared (nIR).
Charge carriers must quickly get to the electrodes and join again. How quickly they can do this depends on how mobile they are and how far they can travel. These are limited by the defect/trap density. Perovskites are hard to dope because they are good at balancing the charges of point flaws. This means that their performance is almost the same as it would be without doping. Also, they can handle a lot of defects. The few natural flaws that are present are mostly shallow traps that stop charge transport from working well instead of helping it happen.
The reaction time (r) is the amount of time it takes for the minority carriers (p) and majority carriers (n) that are created by light to reach their respective terminals and join back together. This is set to the longer of the rise time (rise) or the fall time (fall). There are three main things that affect response time: the RC time constant (𝼏RC), the drift component (𝼏drift), and the diffusion component (𝼏diff).
Electronic noise is made when charge carriers are trapped, released, and recombined, with deep-trap-mediated recombination being the most common when the light is on. This change in current caused by the trap shows up as the 1/f noise, or flicker noise. Jones–Nyquist (thermal) noise and dark noise are two other major types of noise.
Specific detectivity (D) is often used as a standard way to compare how well different photodetectors work. D can be thought of as the signal-to-noise ratio (SNR) of a photodetector with a 1 cm2 incident area and 1 W of incident power when the electrical frequency (f) is 1 Hz. This normalises performance based on the photodetector’s active area and spectral power density (Δf).
Types of Perovskite Photodetectors
There are three types of photodetectors: photoconductors, phototransistors, and
photodiodes. Each of these devices has their own benefits for different applications
and their own shortcomings
Photodiodes
A lot of study has been done on photodiodes in perovskite because they work well with little or no voltage bias and have a shape that is similar to solar cells. They are used a lot in medical images, household goods, and fiber-optic transmission. The shape of perovskite photodiodes is backwards, with a vertical p–i–n junction. They focus on low noise and fast charge collection when the bias is turned around. They respond linearly to light and can’t have a photoconductive gain greater than 1. These features come from interfacial charge transport layers that stop charge from entering the perovskite, protect defect-filled surfaces, lower flicker noise, and lower dark current. A substance called bathocuproine (BCP) blocks holes and moves electrons around, and a substance called PTAA does the same. As an intermediate cushion layer, C60 keeps charges from building up.
Broadband Photodiodes
In their study, Dou et al. made three devices: one with a bathocuproine (BCP) hole-blocking layer (HBL), one with a poly(9,9-bis(30-(N,N-dimethylamino)propyl)-2,7-fl(9,9-dioctylfluorene) HBL, and one without an HBL. The control unit that didn’t have an HBL had a high dark current density of 10−7 A/cm2 and a low specific detectivity of 3 × 1011 Jones at 550 nm. When either BCP or PFN HBLs were used, the specific detectivity went up to 2 × 1012 Jones and 8 × 1013 Jones, respectively. This was because more electrons were injected, which greatly reduced the leaking current. The device also had a big LDR of over 100 dB, which is about the same as the 120 dB LDR of Si photodetectors. Its rise and fall times were also very fast, at 180 and 160 ns, respectively.
Gong discovered that a fullerene derivative called PCBM helps charge carriers move more efficiently while also blocking the TiO2. This results in a lot less leaking current when the bias is turned around. More than 4 × 1012 Jones of specificity was achieved, and more than 80% of the EQE was met at 375 nm to 800 nm. Teng et al. mixed polymethyl methacrylate (PMMA) with PCBM to lower the work function and make it better at covering holes. This photodetector constantly had high specific detectivities of about 1013 Jones, which is about the same as Si photodetectors in the visible range.
Fang et al. proved that a C60 buffer layer cuts dark current by a large amount. It was possible to lower the reverse bias dark current to 1.6 × 10−7 A/cm2 by using PCBM/C60 (20 nm/20 nm) as the electron transport layer (ETL) and PEDOT:PSS (poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) as the hole transport layer (HTL). This finding is a million times less than what Dou et al. found when they didn’t use C60.
Using OTPD instead of PEDOT:PSS (N4,N4′-bis(4-(6-(3-ethyloxetan-3-yl)methoxy)hexyl)pheny-diphenylbiphenyl-4,4′ diamine)) By making the C60 layer thicker from 20 to 80 nm, the dark current was slowed down to 9.1 × 10−9 A/cm2 at -2 V, which is more than 50 times less than the best perovskite photodetector that had been reported before. This record photodetector got a responsivity of 0.21 A/W, which means that charge extraction is not slowed down.
Perovskite photodetectors have shown promise for use in images, which is what the work is mostly about. The detector can pick up 0.64 pW/cm2 of light and has a specific detectivity greater than 1 × 1012 Jones from 330 nm to 790 nm, with a high point of 7.4 × 1012 Jones at 680 nm. Lin et al. made four devices with different amounts of PCBM and C60 buffer layers. The one with the lowest dark current (5 × 10−10 A/cm2) was the one they made. This leads to a high LDR of about 170 dB at −0.5 V, a low NEP of about 200 fW/Hz1/2, a specific detectivity of over 1012 Jones in the UV–Vis band, and a totally blind response in the nIR, showing that it could be used for imaging.
Bao et al. used space-confined inverse-temperature crystallisation to make single crystals that were only a few to tens of microns thick. There was only about 2 fA/Hz of noise in the detectors at 8 Hz, which led to a 6 ×104 CH3NH3+PbI4+I4+ type.
The study also talks about the different kinds of perovskite photodetectors, such as hybrid photodetectors with various PDPPTDTPT:PCBM ratios at a bias of -0.2 V, the ability of hybrid photodetectors to detect different wavelengths at a bias of -0.1 V, and the noise currents of hybrid photodetectors when they are dark and when they are lit by 𝼆 = 890 nm at different intensities.
It talks about making hybrid photodetectors with Pb-based perovskite, which has a wide bandgap and can only receive light wavelengths below 800 nm. A team led by Shen and others created a hybrid MAPbI3 photodetector using a PDPPTDTPT:PCBM polymer. This polymer had a narrow bandgap and was used as a light-absorbing layer, a charge transport layer, and a blocker to charge input from the cathode. It had a NEP of 5 pW/cm2 and a specific detectivity of 1 × 1011 Jones at 900 nm. With an LDR of 95 dB, the reaction range was widened to about 1000 nm.
A self-powered photodetector (SPPD) was described by Su et al. It uses both photoelectric and triboelectric effects. There is an open-circuit voltage between the two electrodes when an outside force repeatedly presses the top copper electrode against the perovskite layer. This is because the contacts have triboelectric charges of opposite signs. This triboelectric device responds very strongly to the amount of light that hits it, which opens up a new way to detect perovskite light.
Gao et al. used an ITO/PEDOT:PSS/PbS QD/MAPbI3/PCBM/Al device structure to get a broad response from 375 to 1100 nm. Quantum dots (QDs) made of PbS work as a broad light sensitiser in the Vis/nIR range, and MAPbI3 works as both a visible-light sensitiser and a trap state passivator of the QD surfaces. Zhang et al. showed that adding a different photoactive layer of PbS QDs greatly increases absorption in the UV–Vis range and also gives a noticeable reaction in the near infrared range. The responsivities and specific detectivities that Liu et al. found were 0.302 A/W and 1.2 × 1013 Jones for 𝜆 = 500 nm and 0.132 A/W and 5.1 × 1012 Jones for 900 nm.
Researchers have made perovskite photodetectors that can find and measure different kinds of light. The exciton peak has grown 60 times in the wideband PbS/CH3NH3PbI3 hybrid photodetector, which means it has a wider reaction range. Ning et al. showed that the perovskite (QDiP) solution can make a poly-heterocrystalline film that is well-passivated and very easy to control. Clifford et al. created a MAPbI2.5Br0.5:PbS liquid QD device in which excitons are created by light and held in place by the PbS quantum dots. The excitons split apart and the charge carriers are thrown out when the electric field is strong enough. One way this is shown is by the quick 45-fold rise in responsivity at a 0.5 V shift.
Habisreutinger et al. showed that single-walled carbon nanotubes (SWCNTs) work well as an HTL for perovskite solar cells. They did this by lowering the number of grain boundaries and pinholes in the perovskite, which allowed photocurrents of up to 20.8 mA/cm2. Li et al. built on this work by adding a nanonet of carbon nanotubes to the perovskite film. This made it easier to remove charges quickly, lowered the film’s total resistance, and raised its responsiveness and equalisation potential (EQE). For a UV-to-IR photodetector, Alwadai et al. used Gd-doped ZnO nanorods (NRs) that were arranged vertically in a MAPbI3 film.
Perovskites make it possible for photodetectors to be very sensitive and quick. There are different kinds of photodetectors. A lot of people use artificial, organic, and quasi-2D (PEA)2(MA)n−1PbnI3n+1 perovskite photodiodes, where n ranges from 1 to ∞. In terms of specific detectivity and responsivity, these detectors work with a 0 V bias and low dark currents. Both n = 60 and n = ∞ show incredibly similar performance.
Narrowband Photodiodes
We have seen that Fang et al. and Lin et al. have made filterless narrowband perovskite photodetectors. These can be tuned by changing the halide makeup to allow for gaps in bands from blue to red. You can change the main absorption start, which lets you use bandgaps for blue to red wavelengths. Above-bandgap photogenerated charges are mostly made near the surface, where they are slowed down and stuck. Below-bandgap charge carriers, on the other hand, are made deeper in the crystal and help make the high, longer-wavelength EQE. If the bias is big enough, the collection efficiency of photogenerated charges with shorter wavelengths rises much more than that of charge carriers with longer wavelengths. This results in a wideband reaction in the end.
A group of scientists led by Li showed how to make self-filtering narrowband photodetectors by depositing a perovskite filter on the top of a material that doesn’t conduct electricity. There is a minimum width at half maximum (FWHM) of 28 nm that can be made bigger in these narrowband photodetectors. They keep the better specific detectivity, LDR, and reaction time of perovskite detectors, which are 2.65 × 10−12 Jones, 190 dB, and 100 ns, respectively.
The photodetectors made by Lin et al. had low highest EQEs of about 12%. This was probably because the organic molecules didn’t allow charge carriers to move around very easily. The reaction time of the photodetectors that Fang et al. made is bound by how thick the crystal is. A photodetector with two active layers was made by Qin et al. They used a polymer active layer and a perovskite electronically cooled, optically active layer. Charge carriers are created by light in the perovskite and then quickly stopped by the thick, slow-moving polymer. The narrowband reaction range is set by the difference in the absorption edges of polymers and perovskite. With an FWHM of 70 nm and rise and fall times of 3.9 and 4.0 μs, the best detection showed an EQE of about 20%.
Photoconductors
The two ends of a photoconductor are p-n junctions, and the gadget can be orientated either vertically or laterally. Vertical photoconductors have short carrier transit lengths and small electrode spacings, which makes them respond quickly and need low drive voltages. On the other hand, it is easier to make horizontal photoconductors and they can handle bigger photocurrents. In some situations, photoconductors can show a gain (₺ph), which is useful for uses with small signals, like single-photon uses.
A very important part of photoconductors is counting photons, which makes it possible for them to have high photocurrent and high responsivity. But it gives up the benefit of low dark current, which leads to a high NEP and low LDR and specific detectivity. Commercial photodetectors with high gain are usually very expensive and hard to make. A solution-processed device could make things a lot better.
In short, photoconductors are just two-terminal p-n junctions that can be arranged either vertically or laterally. Vertical photoconductors have short carrier transit lengths and small electrode spacings, which makes them respond quickly and need low drive voltages. On the other hand, it is easier to make horizontal photoconductors and they can handle bigger photocurrents.
Vertical Photoconductors
It was shown by Moehl et al. that photoconductive gain can be reached with a vertical frame and either a porous or compact TiO2 HBL. The porous HBL let MAPbI3 come into direct contact with the fluorine-doped tin oxide (FTO) electrode. This caused charge to be injected into the perovskite when the bias was turned around. This created a gadget with a high dark current and photocurrent, and at low light levels, it reached a gain of about 150. A third device that didn’t have any charge blocking layers showed a photocurrent increase of about 14 times. This shows that perovskite photoconductors can work without charge blocking layers.
Chen et al. made a gadget that could change from photovoltaic to photoconductive modes by altering the direction of the bias. When the photodiode was run at 0 V bias, it showed excellent accuracy. For low light levels, it was possible to increase the detecting range even more by running it in the photoconductive mode. When the bias was turned around, the photocurrent changed a lot linearly with the bias size.
Ishii et al. made it work better by adding a Eu-terpy complex molecular layer between the MAPbI3 nanocrystal absorption layer and the cathode. This used charge tunnelling injection with the help of a trap at a mixed interface to make photomultiplication possible by causing a big leaking current. The photodetector had an EQE of more than 2 × 105% at 400≦≤750 nm with a bias of −0.5 V and a maximum of 9 × 105% when light intensity was 0.76 mW/cm2. It kept working the same way when light intensity dropped to 10 μW/cm2.
Using a MAPbI3 perovskite, Dong et al. made a device with a high responsivity that didn’t have an intermediate electron-blocking layer. They found that Pb2+ cations trap holes at the surface, which lets electrons move in and make a big gain. Yang et al. created an MSM photoconductor that doesn’t have any intermediate layers. It achieved an EQE of over 10,000% at 300 nm to 800 nm and a maximum of 60,000% at 380 nm.
Lateral Photoconductors
Hu et al. created the first perovskite-based photodetector by putting MAPbI3 on a bendable PET layer and connecting it to metallic ITO contacts. It was sensitive across a wide range of wavelengths and had EQEs and responsivities of 1.19 × 103% and 3.49 A/W at = 365 nm and 5.84% and 0.0367 A/W at 𝼆 = 780 nm with a -3 V bias. The local electric field at the perovskite/ITO Schottky barriers effectively splits charge carriers, which lowers the rate of recombination and raises the conductivity.
In their study, Wang et al. used MAPbCl3 to make an MSM UV photoconductor. This device had a high responsivity of 7.56 A/W at = 360 nm, a fast response speed of less than 50 ms, and a low dark current. Zhou and others used an MSM structure with n = 1, 2, and 3 layers of (CH3(CH2)3NH3)2(CH3NH3)n−1PbnI3n+1 perovskite, which showed that it could pick up light from different areas.
An all-inorganic CsPbI3 nanocrystal (NC) photodetector was made by Ramasamy et al. It had an on/off ratio of about 105 and rise and decay times of 24 and 29 ms, respectively. Dong et al. showed that the Au plasmonic effect can change performance depending on direction and improve photodetecting skills.
X-ray diffraction (XRD) tests revealed that centrifugal casting makes denser and more ordered films than drop-coated films. This makes the photocurrent go from 0.67 µA to 2.77 µA because the particles can move around more easily. The Au-NC device had a much higher on/off ratio and a photocurrent rise from 245.6 to 831.1 μA at a -2 V bias and a wavelength of 532 nm. Horváth et al. made the first nanowire-based photodetector with slip-coated MAPbI3, which is good at blocking light at night.
Perovskite photodetectors have been made for many uses, such as detecting light with a wavelength between 300 and 900 nm. Some researchers have used a mask to make patterns and tweaked evaporation-induced self-assembly to line up nanowires (NWs) for the NW active layer. This led to a responsivity of 0.85 A/W at -1 V bias and a light intensity of 4 μW/cm2. Deng et al. created a MAPb(I1−xBrx)3 NW-array photoconductor with smooth surfaces, a narrow width distribution, and a high level of directional regularity. It worked amazingly well, with 12500 A/W and 150 dB LDR.
Later photoconductors focus on light with a wavelength between 300 and 900 nm. Ka et al. used single-wall carbon nanotubes (SWCNTs) to increase the reaction range to around 1600 nm and make charge transfer more efficient. A lower responsivity is seen when more light hits the material, which suggests that the photoconductive gain is smaller because there are fewer empty traps. Pan et al. combined erbium ytterbium silicate (EYS) nanosheets with MAPbI3 perovskites to make a hybrid 2D-3D IR (1530–≜≤ 1565 nm) photodetector that works at the wavelength range for optical communication and has a reaction time of 900 μs.
Qian et al. showed that flexible, lead-free detectors are possible and can work pretty well compared to other flexible photoconductors that do the same thing. These devices use 2D (PEA)2SnI4 perovskites that have 30 mol% SnF2 added to them to make them more stable and lower the number of empty spaces. They got a responsivity of 16 A/W, which was higher than the previous best of 3 A/W for bendable perovskite detectors with 𝼆 = 470 nm. Lan et al. came out with more research on 2D MAPbI3 photodetectors and found that they could respond to 405 nm light with 40 A/W.
Single crystals, which are usually a few millimetres thick, work well as a top active layer in lateral devices. Saidaminov et al. grew a single crystal of MAPbBr3 on top of an ITO-covered material that had been carved. They were able to achieve a better responsivity of 4000 A/W and a fall time of 25 μs. Liu et al. showed that 2D single-crystal (PEA)2PbI4 can be used to identify visible light. They found that the material’s response was different based on the plane where it was made.
Phototransistor
A phototransistors has three terminals, the same as a field-effect transistor: a source terminal, a drain terminal, and a gate terminal. Through photoexcitation, they can change the conductivity and charge carrier density to control the flow of current. Through the photogating effect, they can also show a lot of gain.
Hybrid heterojunction structures, like the perovskite-graphene type, have done amazingly well. A perovskite-graphene heterojunction was used to make a broadband UV-Vis phototransistor with a high responsivity of 180 A/W. Graphene added electrons to the perovskite valence band to fill the gaps left by photoexcited electrons in the conduction band. The efficiency was greatly improved when perovskite nanocrystals and nanowires were added to the graphene layer.
Because graphene is very conductive and doesn’t have a bandgap, it causes a lot of dark current in electronics. Transition metal dichalcogenides (TMDs) can be used as charge carrier layers to cut down on noise. Photodetectors made with WS2, WSe2, and MoS2 have reached responsivities of 17, 950, and 1.94 × 106 A/W. With less dark current, the MoS2 phototransistor was able to achieve a high specific detectivity of 1.29 × 1012 Jones at 𝼆 = 520 nm.
Zhang et al. recently added WS2 nanosheets to the perovskite layer to make a perovskite–graphene heterojunction phototransistor. This made the phototransistor more sensitive to light by 678.8 A/W and able to detect 4.99 × 1011 Jones.
Conclusion
This article talks about the basic parts of photodetectors and how they work, as well as why perovskites are important in photodetector uses. It talks about recent progress in perovskite photodetectors, which have mostly been used in solar systems. Despite this, there are still big chances to do study in the area. People are becoming more and more interested in perovskites because they are very good at many electrical tasks. It is believed that the study of perovskites will continue to grow.
Leave a Reply