Organic Photodetector (OPD)

1. Theory

• Definition: An organic photodetector (OPD) is a thin-film photodiode whose active layer is an organic donor-acceptor blend, a bulk heterojunction of small molecules (for example ZnPc:${\mathrm{C}}_{60}$) or of a polymer and a fullerene (for example PCDTBT:$\mathrm{P}{\mathrm{C}}_{70}\mathrm{BM}$), between a transparent contact (ITO) and a metal contact, with electron-blocking and hole-blocking layers on both sides.
• Charge generation: A photon creates a tightly bound exciton, the exciton diffuses to a donor-acceptor interface, forms a charge-transfer state, which separates into free carriers, and the carriers drift to the contacts, so that the external quantum efficiency is the product $EQE={\eta }_A{\eta }_{ED}{\eta }_{CT}{\eta }_{CC}$.
• Thickness: The absorption of the blend is strong, so the layer is a few hundred nanometres thick; a junction thicker than about 600 nm follows the Beer-Lambert law and has a flat spectral response, whereas a thin junction acts as a low-finesse cavity.
• Dark current: The dark current of an ideal device is the thermal generation across the charge-transfer gap, $J_0\approx J_{SC}\exp (-q{V}_{OC}/k_{B}T)$, but the measured reverse current is at least three orders of magnitude higher, because traps near the middle of the effective gap generate carriers, and because carriers are injected over the blocking layers at larger reverse bias.
• Noise: Above a corner frequency of 0.3 Hz to 150 Hz the noise approaches the shot noise of the dark current, and the specific detectivity of the best visible devices is about $10^{12}$ to $10^{13}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/W$, against a radiative limit of the order of $10^{15}$ for a gap of 0.8 eV at 300 K (computed).
• Speed: The mobilities of the blends are $10^{-4}$ to $10^{-3}\mathrm{c}{\mathrm{m}}^2/Vs$, so the transit time through a few hundred nanometres is of the order of microseconds, and the response times reported range from 35 μs to the millisecond range.
• Gain: A photomultiplication OPD contains a small fraction of one component, so that trapped carriers lower the injection barrier for the other carrier and the external quantum efficiency exceeds 100%.
• Extensions: The charge-transfer absorption extends the response into the near and short-wave infrared with a lower detectivity, and tandem devices divide the voltage between serially connected junctions to suppress the injection.
• Figures of merit: The external quantum efficiency, the dark current density at a reverse bias of -0.1 V to -1 V, the noise current, $D^{\ast }$, the linear dynamic range and the $-3\mathrm{dB}$ bandwidth.

2. Working Principle

• Absorption and generation: The exciton generated in a domain of the blend reaches the interface if the domain is comparable to its diffusion length, and the charge-transfer state separates in the field of about $10^5$ V/cm at a reverse bias of -1 V.
• Collection: The carriers drift through the blend, electrons in the acceptor and holes in the donor, and are extracted through the blocking layers; the transit time $t_{tr}=d^2/[\mu (V_{bi}-V)]$ grows with the square of the thickness.
• Reverse bias: A reverse bias raises the speed and the collection, but also the dark current, so that the operating bias is a compromise; the review advises to keep it small (-0.1 V to -0.5 V) when the thermal generation through traps is studied.
• Blocking layers: A thin electron-blocking layer (5 nm of MeO-TPD) and a fullerene hole-blocking layer reduce the injection; the ${\mathrm{C}}_{60}$ layer lowered the dark current by about an order of magnitude and a thick junction by almost two orders.
• Traps: A trap density of $10^{15}$ to $10^{16}\mathrm{c}{\mathrm{m}}^{-3}$ raises the dark current, and the deep states act through the generation rate $G_{SRH}$, which grows as the trap level approaches the middle of the gap.
• Photomultiplication: Under illumination the electrons trapped in isolated acceptor domains narrow the hole injection barrier of the electrode, so that many holes are injected for each absorbed photon at a bias of several volts.
• Dynamic range: The response is linear up to a photocurrent density of about 0.03 $A/c{\mathrm{m}}^2$, where bimolecular recombination sets in, which was suppressed about 15 times relative to the Langevin rate.

3. Applications

• Large-area and flexible photosensing: Monolithic thick-junction devices of 16 $\mathrm{c}{\mathrm{m}}^2$ on a plastic substrate had a dark current below 1 $nA/c{\mathrm{m}}^2$.
• Wearable low-light sensing: An organic photodiode with a 500 nm active layer had a noise current of tens of femtoamperes and a noise equivalent power of a few hundred femtowatts at low bandwidth, and its noise does not grow with the area as that of silicon does.

Name	Formula	Related Formulae	Variables/Constants
External quantum efficiency chain	$EQE\left(\lambda \right)={\eta }_A{\eta }_{ED}{\eta }_{CT}{\eta }_{CC}$	$\mathfrak{R}=EQE\frac{q\lambda }{hc}$ $J_{ph}=q\Phi EQE$	$\begin{array}{r}{\eta }_A:fractionofthephotonsabsorbedin\\ theactivelayer\end{array}$ $\begin{array}{r}{\eta }_{ED}:fractionoftheexcitonsthatreacha\\ donor-acceptorinterface\end{array}$ $\begin{array}{r}{\eta }_{CT}:fractionofthecharge-transferstates\\ thatseparateintofreecarriers\end{array}$ ${\eta }_{CC}:chargecollectionefficiency$ $\mathfrak{R}:responsivity$ $\Phi :incidentphotonfluxdensity$ $\begin{array}{r}(a700nmPCDTBT:\mathrm{P}{\mathrm{C}}_{70}\mathrm{BM}junctiongaveEQE=\\ 70\%at-1Vacrossthevisible,anda\\ thin100nmjunction80\%fallingto60\%)\end{array}$
Thickness and optical absorption	${\eta }_A=\left(1-R\right)\left(1-e^{-\alpha d}\right)$	$t_{tr}=\frac{d^2}{\mu \left(V_{bi}-V\right)}$ $P_{def}=e^{-A{D}_D}$	$R:reflectance$ $\alpha :absorptioncoefficientoftheblend$ $d:thicknessoftheactivelayer$ $t_{tr}:transittime$ $\mu :mobility$ $V_{bi}:built-involtage$ $\begin{array}{r}{P}_{def}:probabilitythatadeviceofareaA\\ isfreeofdefects\end{array}$ $\begin{array}{r}{D}_D:defectdensity,theBeer-Lambertform\\ holdsforjunctionsthickerthanabout600\\ nm,thinnerjunctionsbehaveas\\ low-finessecavities\end{array}$ $\begin{array}{r}(at-1Vthedarkcurrentwas315nA\\ withoutand17nAwitha{\mathrm{C}}_{60}layerfor100\\ nm,and6.6nAand0.8nAfor700nm)\end{array}$
Ideal dark current and effective gap	$J_0\approx J_{SC}\exp \left(-\frac{q{V}_{OC}}{k_{B}T}\right)$	$J_D=J_0\left[\exp \left(\frac{qV}{n{k}_{B}T}\right)-1\right]$ $V_{OC}\approx \frac{E_{CT}}{q}-\frac{k_{B}T}{q}\ln \left[\frac{2\pi q}{h^3{c}^2}\frac{E_{CT}-{\lambda }_{CT}}{J_{SC}EQ{E}_{EL}}{f}_{CT}\right]$	$J_0:idealdarksaturationcurrentdensity$ $J_{SC}:short-circuitphotocurrent$ $V_{OC}:open-circuitvoltage$ $\begin{array}{r}{E}_{CT}:energyofthecharge-transferstate\\ (effectivegap)\end{array}$ ${\lambda }_{CT}:reorganizationenergy$ $f_{CT}:oscillatorstrengthfactor$ $EQ{E}_{EL}:electroluminescencequantumefficiency$ $n:idealityfactor(1.5to1.8)$ $k_B:Boltzmannconstant$ $\begin{array}{r}(themeasuredreversedarkcurrentwasat\\ leastthreeordersofmagnitudeabove{J}_0,\\ withvaluesnear10^{-7}mA/c{\mathrm{m}}^{2}at-1Vfor{E}_{CT}\\ =1.58eV;theactivationenergyof\\ PCDTBT:PCBMwas0.80to0.92eV)\end{array}$
Trap-assisted generation	$G_{SRH}=\frac{{\beta }_{SRH}{N}_t{n}_i}{2\cosh \left[\left(E_t-E_i\right)/k_{B}T\right]}$	$E_{t,eff}=E_t-\sqrt{\frac{q^{3}F}{\pi ϵ{ϵ}_0}}$ $J_D\propto G_{SRH}\propto N_t,S_n\propto \sqrt{J_D}\left(\mathrm{shot}\mathrm{noise}\right)$	$G_{SRH}:generationratethroughtraps$ ${\beta }_{SRH}:capturecoefficient$ $N_t:trapdensity$ $n_i:intrinsicdensityoftheeffectivegap$ $E_t,E_i:traplevelandintrinsiclevel$ $\begin{array}{r}{E}_{t,eff}:trapdepthloweredbythe\\ Poole-FrenkeleffectinthefieldF\end{array}$ $S_n:noisecurrentdensity$ $k_B:Boltzmannconstant$ $\begin{array}{r}(N_t=10^{15}to10^{16}\mathrm{c}{\mathrm{m}}^{-3}inalldevices;\\ loweringtheTPDPcontentfrom26.7to6.0\\ mol\%reduced{N}_{t}from3.5\times 10^{15}to1.0\times 10^{15}\\ \mathrm{c}{\mathrm{m}}^{-3}and{J}_{D}byoneorder)\end{array}$
Noise and specific detectivity	$D^{\ast }=\frac{q\lambda \sqrt{A}}{hc}\frac{EQE}{S_n}$	$S_n^2=2q{I}_D+\frac{4k_{B}T}{R_{sh}}$ $D_{shot}^{\ast }=\frac{\mathfrak{R}}{\sqrt{2q{J}_D}}$	$A:area$ $S_n:noisecurrentspectraldensityinA/\sqrt{Hz}$ $I_D,J_D:darkcurrentanditsdensity$ $\begin{array}{r}{R}_{sh}:shuntresistance,theshot-noiseform\\ assumesthatthenoiseisonlytheshot\\ noiseofthedarkcurrentand\\ overestimates{D}^{\ast }whenothersources\\ contribute\end{array}$ $k_B:Boltzmannconstant$ $\begin{array}{r}(adetectivityof4.22\times 10^{13}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/Wfrom\\ thedarkcurrentwas8.4\times 10^{10}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/W\\ fromthemeasurednoisecurrent,afactor\\ 5.0\times 10^2,computed)\end{array}$
Radiative limit of the detectivity	$D_{rad}^{\ast }=\frac{q\lambda }{hc}\sqrt{\frac{EQE}{2q{J}_{0,rad}}}$	$J_{0,rad}=q\int {\varphi }_{BB}\left(E\right)EQE\left(E\right)dE$ ${\varphi }_{BB}\left(E\right)=\frac{2\pi }{h^3{c}^2}\frac{E^2}{e^{E/k_{B}T}-1}$	$\begin{array}{r}{J}_{0,rad}:darkcurrentofradiativeoriginset\\ bythethermalphotonfluxatthegap\end{array}$ $\begin{array}{r}{\varphi }_{BB}:blackbodyphotonfluxperunitenergy\\ intoahemisphere\end{array}$ $\begin{array}{r}E:photonenergy,astepresponse(EQE=1\\ abovethegap)isassumed\end{array}$ $k_B:Boltzmannconstant$ $\begin{array}{r}(foragapof0.8eVat300Ktheintegral\\ gives{J}_{0,rad}=1.0\times 10^{-11}A/c{\mathrm{m}}^{2}and{D}_{rad}^{\ast }=\\ 6.9\times 10^{14}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/W,computed,againstabout\\ 10^{15}quoted;9.0\times 10^{15}for0.95eV)\end{array}$
Bandwidth	$\frac{1}{f_{-3dB}^2}=\frac{1}{f_t^2}+\frac{1}{f_{RC}^2}$	$f_t=\frac{3.5}{2\pi t_{tr}}$ $f_{RC}=\frac{1}{2\pi RC}$	$f_t:transit-limitedcutoff$ $f_{RC}:cutoffofthecircuit$ $t_{tr}:transittimeoftheslowercarrier$ $R:loadplusseriesresistance$ $\begin{array}{r}C:capacitanceofthedevice(3.7nF/c{\mathrm{m}}^2\\ fora700nmjunction)\end{array}$ $\begin{array}{r}(f_{-3dB}=120kHzat0Vandabout1MHzat\\ -5Vfora700nmjunction;f_{RC}=2.2MHz\\ for100\Omega and0.2\mathrm{c}{\mathrm{m}}^2,computed;t_{tr}=18\\ \mu \mathrm{s}for300nm,\mu =10^{-4}\mathrm{c}{\mathrm{m}}^2/Vsand0.5V)\end{array}$
Trap-assisted photomultiplication	$EQE=\frac{\chi \tau }{T}$	$\mathfrak{R}=\frac{\left(I_l-I_d\right)}{P_{in}}$ $D^{\ast }=\frac{\mathfrak{R}}{\sqrt{2e{J}_d}}$	$\chi :fractionofthetrappedelectrons$ $\tau :lifetimeofthetrappedelectrons$ $T:transporttimeoftheholes$ $I_l,I_d:currentinthelightandinthedark$ $\begin{array}{r}{P}_{in}:incidentpower,thetrappedelectrons\\ lowertheholeinjectionbarrierofthe\\ electrode(about0.9eVforAl/P3HT)\end{array}$ $\begin{array}{r}(anAl/P3HT:\mathrm{P}{\mathrm{C}}_{70}\mathrm{BM}(100:1)devicegaveEQE=\\ 1.06\times 10^5\%and\mathfrak{R}=344A/Wat405nmand\\ -17V;theratio\mathfrak{R}hc/(q\lambda )gives1.05\times 10^3,\\ i.e.thesameEQE,computed)\end{array}$
Linear dynamic range	$LDR=20{\log }_{10}\left(\frac{j_{max}}{j_{min}}\right)$	$j_{max}\approx 3\times 10^{-2}A/c{\mathrm{m}}^2$ $j_{min}\approx 2\mathrm{to}3\times 10^{-11}A/c{\mathrm{m}}^2$	$LDR:lineardynamicrangeindB$ $\begin{array}{r}{j}_{max}:photocurrentdensityatwhichthe\\ responsedepartsfromlinearity(setby\\ bimolecularrecombination)\end{array}$ $\begin{array}{r}{j}_{min}:lowerlimitofthemeasurement,both\\ inA/c{m}^2\end{array}$ $\begin{array}{r}(20pA/c{m}^{2}gives183.5dBand30pA/c{m}^2\\ gives180.0dB,computed,against180dB\\ quoted;theresponsewaslinearupto0.1\\ W/c{\mathrm{m}}^{2}at532nmand-1V)\end{array}$

Important Questions

Ques: Why is a thick junction of 700 nm preferred to a thin junction of 100 nm, and what does it cost?

The 700 nm PCDTBT:$\mathrm{P}{\mathrm{C}}_{70}\mathrm{BM}$ junction had a flat EQE of about 70% across the visible, because it follows the Beer-Lambert law, whereas a 100 nm junction had an EQE that fell from 80% to 60% as cavity interference changed with wavelength. The dark current at -1 V was 6.6 nA without and 0.8 nA with a ${\mathrm{C}}_{60}$ layer for 700 nm, against 315 nA and 17 nA for 100 nm, and a large device of 16 $\mathrm{c}{\mathrm{m}}^2$ had less than 1 $nA/c{\mathrm{m}}^2$, because the probability of a defect in a thin film grows as $e^{-A{D}_D}$ with the area. The cost is the speed: the transit time grows with $d^2$, so that the bandwidth was 120 kHz at zero bias and about 1 MHz at -5 V. Reference: Armin et al., Thick junction broadband organic photodiodes, Laser Photonics Rev. 8, 924 (2014).

Ques: Why is the measured reverse dark current of an organic photodiode at least three orders of magnitude above the ideal value?

The ideal current $J_0\approx J_{SC}\exp (-q{V}_{OC}/k_{B}T)$ is the thermal generation across the charge-transfer gap, and the measured reverse current exceeded it by at least three orders. The excess is attributed to trap states near the middle of the effective gap, which generate carriers at the rate ${\beta }_{SRH}{N}_t{n}_i/[2\cosh ((E_t-E_i)/k_{B}T)]$, with trap densities of $10^{15}$ to $10^{16}\mathrm{c}{\mathrm{m}}^{-3}$ in all devices; lowering the fullerene content from 26.7 to 6.0 mol% reduced the density from $3.5\times 10^{15}$ to $1.0\times 10^{15}\mathrm{c}{\mathrm{m}}^{-3}$ and the current by one order. The review adds that the activation energy of 0.80 to 0.92 eV lies 0.3 eV to 0.5 eV below the effective gap, and that injection adds to the current at larger reverse bias. Reference: Kublitski et al., Reverse dark current in organic photodetectors and the major role of traps as source of noise, Nat. Commun. 12, 551 (2021); Ma, Janssen and Gelinck, Trap-Assisted Charge Generation and Recombination in State-of-the-Art Organic Photodetectors, Adv. Mater. Technol. 8, 2300234 (2023).

Ques: Is the noise of an organic photodiode equal to the shot noise of the dark current?

For the low-noise devices of the trap study the noise approached the shot-noise level $\sqrt{2qI}$ above a corner of 0.3 Hz to 150 Hz, and was proportional to $\sqrt{J_D}$, whereas a perspective letter found a detectivity of $4.22\times 10^{13}$ from the dark current and only $8.4\times 10^{10}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/W$ from the measured noise, a factor 500 (computed), and the review states that shot-noise values typically overestimate $D^{\ast }$. The thick-junction diodes showed a noise a few times above the shot noise. The more likely statement is that the shot noise is reached only by devices with a low trap density, so that a detectivity must be derived from a measured noise current, preferably at the bandwidth of the application and not at about 1 Hz. Reference: Kublitski et al., Reverse dark current in organic photodetectors and the major role of traps as source of noise, Nat. Commun. 12, 551 (2021); Detectivity analysis for organic photodetectors, Org. Electron. 57, 89 (2018) (KAIST record, abstract); Ma, Janssen and Gelinck, Trap-Assisted Charge Generation and Recombination in State-of-the-Art Organic Photodetectors, Adv. Mater. Technol. 8, 2300234 (2023); Armin et al., Thick junction broadband organic photodiodes, Laser Photonics Rev. 8, 924 (2014).

Ques: Why do the bandwidths quoted for organic photodiodes range from 120 kHz to 1 MHz and from tens of microseconds to milliseconds?

The transit time $t_{tr}=d^2/[\mu (V_{bi}-V)]$ gives 18 μs for 300 nm, $\mu =10^{-4}\mathrm{c}{\mathrm{m}}^2/Vs$ and 0.5 V (computed), close to the 10 μs given by the review, which quotes 1 ms to 10 ms as typical for response times. For the 700 nm junction with ${\mu }_h=3\times 10^{-4}$ and ${\mu }_e=3\times 10^{-3}\mathrm{c}{\mathrm{m}}^2/Vs$ at -5 V, $f_t=3.5/(2\pi t_{tr})$ is 0.17 MHz and 1.7 MHz (computed), and $f_{RC}$ is 2.2 MHz, so that the measured 1 MHz lies between the two values, not at the hole value that the authors identify as limiting. A mobility that rises with field is the more likely reason for the difference, so that the bandwidth depends on the junction, the bias and the definition of the response time. Reference: Armin et al., Thick junction broadband organic photodiodes, Laser Photonics Rev. 8, 924 (2014); Ma, Janssen and Gelinck, Trap-Assisted Charge Generation and Recombination in State-of-the-Art Organic Photodetectors, Adv. Mater. Technol. 8, 2300234 (2023); Fuentes-Hernandez et al., Science 370, 698 (2020), as reported by Physics World, 19 November 2020.

Ques: Is a detectivity of $10^{13}$ for a photomultiplication organic photodetector with an EQE above $10^5\%$ meaningful?

The photomultiplication device reached EQE = $1.06\times 10^5\%$ and 344 A/W at 405 nm and -17 V, consistent with its quantum efficiency, since $\mathfrak{R}=EQEq\lambda /hc$ gives 345 A/W (computed). The detectivity was calculated as $\mathfrak{R}/\sqrt{2e{J}_d}$ with the argument that the noise in the dark is shot noise, without a noise measurement; with 344 A/W and $1.87\times 10^{13}$ it corresponds to $J_d=1.1\times 10^{-3}A/c{\mathrm{m}}^2$ (computed). Gain multiplies the noise as well, and traps add generation-recombination noise, so the value is an upper bound. The Results assign $2.32\times 10^{13}$ to the long-wavelength range of the 325 nm device and $1.87\times 10^{13}$ to the short-wavelength range of the 205 nm device, whereas the Conclusions assign $1.87\times 10^{13}$ to the longer wavelengths, so that the Results are the more likely. Reference: Wang et al., Improving Performance of Organic Photodetectors based on Trap Assisted Photomultiplication through Adjusting Photogenerated Carrier Distributions (preprint, Taiyuan University of Technology); Ma, Janssen and Gelinck, Trap-Assisted Charge Generation and Recombination in State-of-the-Art Organic Photodetectors, Adv. Mater. Technol. 8, 2300234 (2023).

Ques: What limits the detectivity of an organic photodetector in principle?

The radiative limit follows from the thermal photon flux at the gap: for a step response above 0.8 eV at 300 K the flux is $6.3\times 10^7\mathrm{c}{\mathrm{m}}^{-2}{\mathrm{s}}^{-1}$ and $J_{0,rad}=1.0\times 10^{-11}A/c{\mathrm{m}}^2$, which gives $D_{rad}^{\ast }=6.9\times 10^{14}\mathrm{cm}\mathrm{H}{\mathrm{z}}^{1/2}/W$ (computed), close to the ceiling of about $10^{15}$ quoted by the review; the limit rises steeply with the gap, to $9.0\times 10^{15}$ at 0.95 eV (computed). The best visible devices are at $10^{12}$ to $10^{13}$, a few orders below the limit, because of the trap-assisted current, and short-wave infrared devices are lower: $2.96\times 10^{10}$ at 1300 nm and 0 V, and $1.7\times 10^{13}$ at 1100 nm for the noise-based value of a device with an EQE of 58.9%. Reference: Ma, Janssen and Gelinck, Trap-Assisted Charge Generation and Recombination in State-of-the-Art Organic Photodetectors, Adv. Mater. Technol. 8, 2300234 (2023); Specific detectivity-oriented low-noise management in organic photodetectors, Nanoscale (2025) (RSC article, text read through a summary).

Ques: Why do the blocking layers lower the noise of the photodiode?

A reverse bias drives carriers from the contacts into the blend, and a blocking layer raises the injection barrier, so that only the thermal and trap-assisted generation remains. The review lists self-assembled electron-blocking layers that bind to the ITO and passivate traps, with a shunt resistance of 21.4 TΩ against 4.52 GΩ for PEDOT:PSS and 113 GΩ for 2PACz, and a median noise current of 852 aA, the lowest of the review; a 5 nm MeO-TPD layer removed the hole extraction barrier in the trap study. The mechanism is thus a reduced injection and a reduced trap density at the interface, and the quoted shunt resistances come from different measurements, which makes the ranking of the layers qualitative. Reference: Specific detectivity-oriented low-noise management in organic photodetectors, Nanoscale (2025) (RSC article, text read through a summary); Kublitski et al., Reverse dark current in organic photodetectors and the major role of traps as source of noise, Nat. Commun. 12, 551 (2021).

Ques: Why is the spectral response of a thick junction flat while that of a thin junction is not?

In a thin junction the optical field is shaped by interference between the reflections at the electrodes, as in a low-finesse cavity, so that the absorbed fraction depends on the wavelength and on the exact thickness; the EQE of the 100 nm junction fell from about 80% at 300 nm to about 60% at 450 nm. Above about 600 nm of thickness the simulation enters the Beer-Lambert regime, where the interference averages out, and the absorbed fraction $1-e^{-\alpha d}$ is nearly one over the band, so that the response is flat at about 60% (70% at -1 V with the ${\mathrm{C}}_{60}$ layer) and insensitive to variations of the thickness. The internal quantum efficiency was assumed independent of wavelength in the simulation, which is the basis of the comparison. Reference: Armin et al., Thick junction broadband organic photodiodes, Laser Photonics Rev. 8, 924 (2014); Tian et al., Mechanism of Organic Photovoltaics, Encyclopedia (MDPI entry adapted from Energies 16, 4494 (2023)).

Related pages in this vault: Perovskite Photodetector (Photodiode and Photoconductor), Colloidal Quantum-Dot Photodetector (CQD), Photoconductor, Graphene and 2D-Material Photodetector, Ultraviolet Photodetector (AlGaN, SiC, Ga2O3, Diamond), Organic Solar Cell (Bulk Heterojunction), Organic Light Emitting Diode (OLED), Light Emitting Diode (Homojunction and Heterojunction LED).