† Corresponding author. E-mail:
Project supported by the National Natural Science Foundation of China (Grant No. 61106043), the Natural Science Foundation of Shaanxi Province, China (Grant No. 2015JM6267), and Program of Xi’an University of Technology, China (Grant No. 103-4515013).
Photomultiplication (PM) structure has been widely employed to improve the optoelectronic performance of organic photodetectors (OPDs). However, most PM-type OPDs require a high negative operating voltage or complex fabrication. For obtaining high-efficiency OPDs with low voltage and easy process, here the bulk heterojunction (BHJ) structure of high exciton dissociation efficiency combined with the method of trap-assisted PM are applied to the OPDs. In this paper, we investigate the operating mechanism of OPDs based on poly(3-hexylthiophene) (P3HT): (phenyl-C61-butyric-acid-methyl-ester) (PC61BM), and poly-{[4,8-bis[(2-ethylhexyl)oxy]-benzo[1,2-b:4,5-b]dithiophene-2,6-diyl]-alt-[3-fluore-2-(octyloxy)carbonyl-thieno[3,4-b]thiophene-4,6-diyl]} (PBDT-TT-F):PC61BM doped with C60 as active layer. Furthermore, the influence of C60 concentration on the optoelectronic performances is also discussed. With 1.6 wt.% C60 added, the P3HT:PC61BM:C60 OPD exhibits a 327.5% external quantum efficiency, a 1.21 A·W−1 responsivity, and a 4.22 × 1012 Jones normalized detectivity at −1 V under 460 nm (0.21 mW·cm−2) illumination. The experimental results show that the effective electron traps can be formed by doping a small weight of C60 into BHJ active layer. Thus the PM-type OPDs can be realized, which benefits from the cathode hole tunneling injection assisted by the trapped electrons in C60 near the Al side. The efficiency of PM is related to the C60 concentration. The present study provides theoretical basis and method for the design of highly sensitive OPDs with low operating voltage and facile fabrication.
Organic photodetectors (OPDs) in the image sensing, communications, ultraviolet (UV)–visible–near infrared (NIR) region, and chemical/biological sensing show a wide range of commercial and scientific applications.[1–5] Conventionally, most of the OPDs are photodiodes with photoresponse originating from the collection of photogenerated charges.[6,7] Due to the limitation of photon harvesting efficiency, exciton dissociation efficiency, and photogenerated charge carrier collection efficiency, the external quantum efficiency (EQE) of photodiode-type OPDs is lower than 100%.[8,9] In spite of the photodiode-type OPDs possessing relatively low working voltage and noise current, the responsivity (R) is not too high due to the EQE values of <100%. In order to achieve highly sensitive optical detection, environmental noise should be generally reduced by complex equipment.[10,11] Therefore, the OPDs with facile fabrication, high responsivity R, and normalized detectivity (D*) are in urgent need.[12–15] The critical steps for improving normalized detectivity of OPDs are to enhance responsivity.[16] In recent years, photomultiplication (PM) structure has been widely applied to the OPDs so as to further improve the optoelectronic performance. The efforts on trap-assisted PM have always been attempted in OPDs and become a hot topic recently.[12–19]
Wang et al. proposed a solution processed PM-type OPD based on P3HT:PC61BM blend film with PEIE modified ITO electrode, which exhibits a R of 14.25 A·W−1 under 550 nm (3.09 μW/cm2) at −1.0 V bias.[16] However, the D* is not improved correspondingly. Luo et al. reported PM-type OPDs with a 5 nm C60 as hole block layer between ITO and active layer which show a R of 1.33 A·W−1 under 850 nm (0.782 mW/cm2) at −3.0 V bias.[17] However, a higher evaporation temperature of C60 will be complex to fabricate. Nie et al. utilized a photon-controllable “valve" formed by PBDTT-DPP: PC71BM/MoO3 to accumulate and release external circuit carriers, realizing EQE>100%.[18] However, this structure needs donor material with its electron trap levels lower than the lowest unoccupied molecular orbital (LUMO) of PC71BM, so the materials selection will be limited. Recently, Wang et al. reported that an EQE of 53500% was achieved by trap-assisted PM via a small weight of PC71BM doped in P3HT.[19] Whereas this large EQE value brought defect of too high operating voltage (−60 V). From the perspective of fabrication process, trap doping is an easy method to realize PM.
In order to solve the shortcomings of the above-mentioned methods, such as special materials and too high working voltage, the bulk heterojunction (BHJ) structure of high exciton dissociation efficiency combined with the method of trap-assisted PM is applied to the OPDs. In this paper, we investigate the PM-type OPDs prepared with P3HT:PC61BM (and PBDT-TT-F:PC61BM) blends doped with low LUMO level of C60 (4.5 eV, lower than PC71BM LUMO level) using the one-step solution process. The PM mechanism of this OPD is investigated systematically, especially for the varying relation between PM efficiency and C60 concentration. Additionally, the device performance of the two donor–acceptor combinations is also compared. Therefore, our works may provide an easy method for obtaining highly sensitive OPDs with low working voltage, high R and D* simultaneously.
In this paper, the two donor–acceptor combinations are P3HT:PC61BM and PBDT-TT-F:PC61BM respectively, and the trap dopant is C60. The C60 concentrations in the P3HT:PC61BM (1:1 wt/wt) active layer were 0, 0.4 wt.%, 1.6 wt.%, and 4.8 wt.%, corresponding to device A, B, C, and D, respectively. For C60-doped PBDT-TT-F:PC61BM (1:1.5 wt/wt) devices, the C60 concentrations were 0, 0.3 wt.%, 1.3 wt.%, and 3.8 wt.% and these represent device E, F, G, and H correspondingly. Then the P3HT (or PBDT-TT-F): PC61BM:C60 blends were dissolved in 1 ml chlorobenzene, respectively. These solutions were all stirred for 12 hours at room temperature with a magnetic stirrer. All devices were fabricated on indium tin oxide (ITO)-coated substrates (sheet resistance: 15 Ω/□). The ITO substrates were cleaned with sequential sonication in deionized water, acetone, ethanol, and then dried by a nitrogen gun. The PEDOT:PSS was firstly spin-coated on the top of ITO with 3500 rpm for 1 min and was subsequently dried in ambient at 100 °C for 10 min. Then, the active layers were spin-coated on top of ITO/PEDOT: PSS layer with 500 rpm for 1 min and were also annealed at 100 °C for 10 min. Finally, a 100 nm cathode Al was thermally deposited on top with 0.3 nm·s−1 deposition rate at a vacuum degree of 5 × 10−4 Pa. The deposition rates and thickness were monitored by oscillating quartz monitors. The effective area of the devices fabricated was 1 cm2. Device structure reported here is ITO/PEDOT: PSS(35 nm)/P3HT:PC61BM (and PBDT-TT-F:PC61BM):C60(160 nm)/Al(100 nm), as shown in Fig.
The current density–voltage (J–V) characteristics and capacitance–voltage (C–V) characteristics of devices were measured by Keithley 2636B Semiconducting System and Keithley PCT-CVU Multi-frequency CV meter (frequency: 10 kHz), respectively. The light source is a λ = 460 nm commercially available diode (Edson, 5050) powered by different bias so as to vary light intensity (Pin). The light intensity was calibrated by a Newport 818-UV power meter. The photoluminescence spectra were recorded using a Transient Steady-state Fluorescence Spectrometer (Edinburgh FLS9, laser wavelength: 510 nm). The absorption spectra and thickness of the active layer films were measured using a UV–Vis spectrophotometer (PerkinElmer, Lambda 950) and Ellipsometer (VB-400), respectively. An atomic force microscopy (AFM, Dimension Icon) was used to analyze the morphology of the films. All the measurements were carried out at room temperature in the ambient condition.
In this paper, the responsivity (R), normalized detectivity (D*), and external quantum efficiency (EQE) of the OPDs were calculated by the following equations:[15,18]
![]() |
![]() |
![]() |
Figure
![]() |
![]() |
It can be clearly seen that the light current density (Jlight) of C60-doped devices exhibits an obvious improvement compared with their respective undoped devices. For example, device A shows only Jlight = 5.01 × 10−5 A·cm−2 under −1.0 V, however device B, C, and D exhibit larger Jlight (1.99 × 10−4, 2.55 × 10−4, and 1.68 × 10−4 A·cm−2 correspondingly). To give a more intuitive comparison, we calculate the EQE values based on light J–V curves according to Eq. (
The dispersed C60 in the active layer can be considered as electron traps, which has been discussed in the SCLC analysis above. Given that the electron traps will also influence the transportation of charge carriers generated by incident photons, we further do photoluminescence (PL) study and light-assisted capacitance measurements subsequently. PL is also a strong mean to prove the electron trapping ability of C60 doped in active layer,[13] which can trap electrons generated in P3HT and PC61BM based on its energy level, shown in Fig.
As is well known, light-assisted capacitance measurements have been used to study the device capacitance, which is determined by the charges at the electrode interfaces.[16,24] Figure
![]() |
![]() |
![]() | Fig. 4. Schematic diagram of the operating mechanism of typical photovoltaic OPDs (a) and PM-type OPDs (b). |
Incident photons absorbed in P3HT (PBDT-TT-F) and PC61BM firstly generate electron–hole pairs, which separate at the interface of P3HT (PBDT-TT-F)/PC61BM into free electrons and holes, for undoped devices, holes eventually end up on P3HT (PBDT-TT-F), and electrons end up on PC61BM, finally, these free charges transport in their respective channels and are collected by anode/cathode, which is sketched in Fig.
In order to exclude the other factors that may influence the device performance to some extent, such as the morphology and the light absorption of active layer film,[27–29] next we investigate the difference of films after doping C60 based on their AFM images and absorption spectra. Figures
![]() | Fig. 5. AFM images of films for (a) P3HT:PC61BM and (b) P3HT:PC61BM:1.6 wt.% C60. (c) Absorption spectra of P3HT:PC61BM:C60 (0 and 1.6 wt.%) and PBDT-TT-F:PC61BM:C60 (0 and 1.3 wt.%), respectively. |
It is known that the responsivity (R) and normalized detectivity (D*) of the OPDs strongly depend on the voltage and photon wavelength.[16] Tables
![]() | Table 1. Optoelectronic performances for P3HT:PC61BM:C60 (A, B, C, and D) and PBDT-TT-F:PC61BM:C60 (E, F, G, and H) OPDs under −1.0 V and comparison with other reported PM-type OPDs. . |
It is noteworthy that the C60-doped P3HT:PC61BM devices show better optoelectronic performance comparing with C60-doped PBDT-TT-F:PC61BM devices according to Table
Table
In summary, we have demonstrated a one-step solution processed PM-type OPD based on C60-doped bulk heterojunction structure at low bias. Using the advantages of BHJ with high exciton dissociation efficiency as active layer, large PM under low bias can be achieved by doping a small weight of C60 as traps. A schematic energy-level model combined with experimental measurements is capable of explaining the operating mechanism of the PM-type OPDs proposed here. The PM originates from cathode hole tunneling injection assisted by the trapped electrons in C60 near the Al side. The efficiency of PM is related to the C60 concentration in the active layer. In order to further improve the device performance, the donor material should have higher LUMO level and shallower HOMO level. On the basis, the 1.6 wt.% C60-doped P3HT:PC61BM device exhibits EQE = 327.5%, R = 1.21 A·W−1, and D* = 4.22 × 1012 Jones at −1 V under 460 nm (0.21 mW·cm−2) illumination. This paper provides a universal method of designing high PM OPDs with low operating voltage and facile fabrication.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] | |
[19] | |
[20] | |
[21] | |
[22] | |
[23] | |
[24] | |
[25] | |
[26] | |
[27] | |
[28] | |
[29] | |
[30] | |
[31] |