Photomultiplication type quasi-planar all-polymer photodetectors (PM-QAPDs) are fabricated with P3HT/PY-IT as active layers by employing a sequential spin-coating method. The part of PY-IT can penetrate into a P3HT layer to emerge isolated electron-traps formed with PY-IT surrounded by P3HT. The trapped electron distribution near an Al electrode will determine the spectral response range of PM-QAPDs. Broadband PM-QAPDs can be achieved with a 0.25 μm thick P3HT layer and a ultra-thin PY-IT layer prepared from 1 mg/ml solution, exhibiting a broad response from 320 to 870 nm. An external quantum efficiency (EQE) value of optimal PM-QAPDs approaches 16 000% at 360 nm under −12 V bias. When the thickness of the P3HT layer is increased to 2.4 μm, the PM-QAPDs exhibit a narrowband response from 630 to 870 nm, which can be well explained according to the Beer–Lambert law. The work may provide a smart strategy to adjust response range of PM-QAPDs by alerting the thickness of the donor layer.
Photodetectors are indispensable due to their great potential applications in electronic components, such as image-sensing, optical-communication, environmental, biological monitoring, and night vision imaging, especially for the photodetectors with near-infrared light response ability.1–4 Polymer photodetectors (PPDs) attract great attention during the recent years due to the rapid development of organic semiconducting materials with tunable optical bandgap.5,6 An External quantum efficiency (EQE) value of photodiode type PPDs is lower than 100%, which is commonly determined by the efficiencies of charge-collection, exciton-dissociation, and photon-harvesting.7–9 The relatively low EQE of photodiode type PPDs will limit its application in the field of weak light detection. In 2015, Li et al. reported a trap-assisted photomultiplication type polymer photodetector (PM-PPDs) by employing poly(3-hexylthiophene-2,5-diyl) (P3HT):6,6]-phenyl-C71-butyric acid methyl ester (PC71BM) with 100:1 weight ratio as active layers.10 The special active layers contain a continuous hole transport channels formed by P3HT and plenty of isolated electron-trap formed by P3HT/PC71BM/P3HT. The EQE values of PM-PPDs are much larger than 100% under −10 V bias, also keeping rather low dark current density (JD).11 The PM-PPDs exhibit great potential in weak light detection due to the large photo-induced current density (JPI), the JPI of PM-PPDs originates from the charge tunneling-injection rather than the photoelectric conversion process. Chung et al. reported a photomultiplication type quasi-planar polymer photodetectors by employing a relatively thick P3HT layer and A super-thin PC71BM layer, part of PC71BM can permeate into P3HT layer to emerge some isolated electron-traps on the top surface of donor layers.12 The quasi-planar structure can offer an efficient hole transport channel in the donor layer without electron traps, leading to the well suppressed charge recombination in the donor layer for better performance of quasi-planar P3HT/PC71BM based photodetectors.13 The previously reported quasi-planar PM-PPDs exhibit a spectral response range in the visible light range. It can be envisioned that electron traps formed with polymer acceptor surrounded by the polymer donor will improve its stability due to the tangled polymer side chains, which should be beneficial to further optimize the performance of PM-QAPDs. In fact, there is great challenge to adjust the spectral response range of PM-PPDs. Yang et al. proposed a dual-band PM-PPDs with one photon harvesting layer and one PM layer. The optimal dual-band PM-PPDs exhibit EQE of 2800% at 350 nm with FWHM of 80 nm under +10 V bias, achieving the EQE values of 7500% at 430 nm and 5200% at 600 nm under −10 V bias.14 Liu et al. reported a ternary strategy by incorporating an ultranarrow bandgap material as the third component, achieving broad spectral response from 350 to 950 nm.15 It should be highlighted that a quasi-planar structure can provide a simple and efficient method to adjust spectral response range of PM-QAPDs by altering the thickness of the donor layer.
In this study, PM-QAPDs were fabricated by employing the structure of indium tin oxide (ITO)/poly(3,4-ethylenedioxythiophene):poly (styrene sulfonate) (PEDOT:PSS)/P3HT/Poly[2,2′-((2Z,2′Z)–((12,13-bis(2-octyldodecyl)-3,9-diundecyl-12,13dihydro[1,2,5]thiadiazolo[3,4e]thieno[2′′,3′′:4′,5′]thieno[2′,3′:4,5]pyrrolo[3,2-g]thieno[2′,3′:4,5]thieno[3,2-b]-indole-2,10-diyl)bis(methanylylidene))bis(5-methyl-3-oxo-2,3-dihydro-1H-indene-2,1-diylidene)) dimalononitrile-co-2,5-thiophene] (PY-IT)/Al, P3HT layer thickness was adjusted from 0.25 to 2.4 μm by spin-coating chlorobenzene solutions with the concentrations of 50, 60, or 80 mg/ml, and the ultra-thin PY-IT layer was prepared from chloroform solution with the concentrations from 0.5 to 2 mg/ml. The device schematic diagram of PM-QAPDs, energy level, and chemical structures are shown in Figs. 1(a)–1(c). PY-IT is deliberately selected as the acceptor due to its suitable energy levels matched with P3HT to form 0.8 eV deep electron traps. The depth of electron traps is determined by the lowest unoccupied molecular orbital (LUMO) level of P3HT and PY-IT. Meanwhile, polymer PY-IT with a narrow optical bandgap of 1.8 eV has strong photon harvesting ability in long wavelength range, leading to the near infrared light response of PM-QAPDs. The relatively deep highest occupied molecular orbital (HOMO) level (−5.6 eV) of PY-IT can also hamper hole injection under dark conditions. The absorption spectra of films were obtained by a Shimadzu UV-2600 PC spectrophotometer. The absorption spectra of PY-IT, P3HT, and P3HT/PY-IT films are shown in Fig. 1(d), and the quasi-planar structure should be formed due to the co-existence of two absorption peaks (one for P3HT and the other one for PY-IT) in the absorption spectrum of P3HT/PY-IT film. In fact, one donor layer/one acceptor layer structure has been commonly reported by a sequential spin-coating method in the field of polymer solar cells (PSCs), the quasi-planar PSCs also exhibit excellent performance in comparisons with the corresponding bulk heterojunction PSCs with the same donor and acceptor materials.16,17 The EQE values of PM-QAPDs with 0.25 μm thick P3HT layer arrives to 9900% under a reverse bias of −10 V, benefiting from efficient hole tunneling-injection from an external-circuit induced by interfacial trapped electron near an Al electrode.18 The response of PM-QAPDs covers from 320 to 870 nm, which can be well explained by absorption spectra of PY-IT and P3HT films. When P3HT layer thickness is increased from 0.8 to 2.4 μm, the response of PM-QAPDs could be markedly narrowed by suppressing short-wavelength response. The narrowed response spectra of PM-QAPDs can be sufficiently explained from the principle of interfacial trap-assisted charge tunneling-injection.19 The quasi-planar strategy can offer a smart strategy to realize narrowband PM-QAPDs by adjusting the thickness of the donor layer according to the Beer–Lambert law.20
(a) Device schematic diagram of PM-QAPDs. (b) Chemical structures of PY-IT and P3HT. (c) Energy levels of PY-IT and P3HT. (d) Normalized absorption spectra of neat P3HT, PY-IT, and P3HT/PY-IT films.
(a) Device schematic diagram of PM-QAPDs. (b) Chemical structures of PY-IT and P3HT. (c) Energy levels of PY-IT and P3HT. (d) Normalized absorption spectra of neat P3HT, PY-IT, and P3HT/PY-IT films.
Series of ultra-thin PY-IT layers were prepared onto 0.25 μm thick P3HT layers by spin-coating chloroform solution with the concentrations of 0.5 to 2 mg/ml to investigate the effect of acceptor layer thickness on performance of PM-QAPDs. The J–V curves of PM-QAPDs were investigated by using a Keithley 2400 source meter under −10 to 1 V bias in the dark or at 1.5 mW/cm2 white-light illumination, as exhibited in Fig. 2(a). Obviously, the JD of PM-QAPDs is slightly increased with the increasing of PY-IT solution concentration, which should be ascribed to partial electron percolation pathway in the P3HT layer due to the sufficient permeation of PY-IT. The light-current density (JL) of PM-QAPDs is increased and then decreased dependence on PY-IT solution concentration increasing. The JL of PM-QAPDs originates from hole tunneling-injection assisted by trapped-electron near the Al electrode.16 The thickness of PY-IT layer is increased along with PY-IT solution concentration increasing, which will result in the P3HT/PY-IT interface far-away from the Al electrode. If trapped electrons are far-away from the Al electrode, which can hardly induce sufficient interfacial band-bending for hole tunneling-injection.21,22 The increased and then decreased JL of PM-QAPD dependence on PY-IT layer thickness can be well explained on the basis of the interfacial trap-assisted hole injection theory.23 The JPI is stipulated as the distinction of JL and JD for evaluating photodetection ability. The JPI vs applied bias (JPI–V) curves of PM-QAPDs are exhibited in Fig. 2(b). The JPI of PM-QAPDs are increased and then decreased along with the increased thickness of PY-IT layer preparing from different concentration PY-IT solutions. The optimal JPI can be obtained in the PM-QAPDs with PY-IT (1 mg/ml) as an acceptor layer, also preserving JD as low as possible.
(a) J−V curves of the PM-QAPDs at distinct conditions and (b) JPI−V curves and the signal-to-noise ratio curves of the PM-QAPDs.
(a) J−V curves of the PM-QAPDs at distinct conditions and (b) JPI−V curves and the signal-to-noise ratio curves of the PM-QAPDs.
Signal-to-noise ratio (SNR) of PM-QAPDs is achieved from the ratios of JPI to JD for evaluating its photodetection ability from intrinsic noise signal.24–26 The SNR of PM-QAPDs with distinct PY-IT layer thickness dependence on bias are shown in Fig. 2(b). The PM-QAPDs based on PY-IT (1 mg/ml) as acceptor layers exhibit larger SNR than that of PM-QAPDs with PY-IT (0.5, 1.5, and 2 mg/ml) as acceptor layers, resulting from relatively low JD. The PM-QAPDs with PY-IT (1 mg/ml) as acceptor layers show the maximal SNR values when the bias is larger than −4 V, benefiting from the well suppressed JD. The suppressed JD is one of efficient strategies to increase SNR of PM-QAPDs. After well-contrasting the JPI and SNR of PM-QAPDs, optimal concentration of PY-IT solution should be 1 mg/ml for achieving highly sensitive PM-QAPDs.
(a) EQE spectra and (b) D* of optimal PM-QAPDs at distinct bias. (c) Normalized TPC curves measured at 803 nm light condition. (d) LDR of PM-QAPDs at −10 V bias.
(a) EQE spectra and (b) D* of optimal PM-QAPDs at distinct bias. (c) Normalized TPC curves measured at 803 nm light condition. (d) LDR of PM-QAPDs at −10 V bias.
The thickness of P3HT layer is adjusted from 0.8 to 2.4 μm to investigate the P3HT/PY-IT interfacial trapped-electron distribution on the EQE spectral shape of PM-QAPDs. The J–V curves of PM-QAPDs with distinct P3HT layer thickness are displayed in Fig. 4(a). It is obvious that JD of PM-QAPDs is markedly suppressed by increasing P3HT layer thickness. Meanwhile, the JL values of PM-QAPDs are also decreased by increasing P3HT layer thickness, originating from weakened hole tunneling-injection due to the decreased trapped-electron density near the Al electrode.36 It is envisaged that photogenerated charge distribution in P3HT layer is easily adjusted by changing active layer thickness, which will influence trapped-electron density near the Al electrode.37,38 The EQE spectra of PM-QAPDs with the distinct P3HT layer thickness were investigated at −10 V bias, as shown in Fig. 4(b). It is obvious that the EQE values of PM-QAPDs are decreased as the P3HT layer thickness increasing, especially in the 400–600 nm wavelength range related to photon-harvesting of P3HT films. Thick P3HT layer can harvest most of incident photon in the short wavelength range, and more photogenerated excitons in P3HT layers hardly diffuse to P3HT/PY-IT interface and dissociate into the free carrier.39 The less trapped-electron density near the Al electrode is difficult to induce sufficient interfacial band-bending for hole tunneling-injection, resulting in the decreased EQE in the short wavelength range of PM-QAPDs with the thick P3HT layer. When P3HT layer thickness is increased to 2.4 μm, PM-QAPDs exhibit relatively narrow response covering 630–870 nm. The EQE spectra of PM-QAPDs with 2.4 μm thick P3HT layer were investigated at distinct reverse bias, as shown in Fig. 4(c). The EQE values of PM-QAPDs are markedly increased along with reverse bias increasing, which should be on account of the more titled interfacial band-bending and the enhanced hole transport in active layers at larger bias.15,40 The EQE values of PM-QAPDs are increased to 970% at 780 nm at −40 V bias. The EQE spectral shape of PM-QAPDs with 2.4 μm thick P3HT layer can be well explained from absorption spectra of corresponding P3HT/PY-IT layers, as displayed in Fig. 4(d). The P3HT/PY-IT layers have sufficient photon-harvesting-ability in short wavelength range and weak photon-harvesting-ability in the range from 630 to 870 nm. Most of the photons with short wavelength are effectively harvested by P3HT and these photogenerated excitons are hardly dissociated into free carrier, leading to the rather weak photo-response of PM-QAPDs in short wavelength range. The photons with wavelength from 630 to 870 nm are sufficiently harvested by PY-IT and these photogenerated excitons are dissociated into free carrier at P3HT/PY-IT interface. These photogenerated-electrons will be trapped in PY-IT surrounded by P3HT, and trapped-electron will induce the interfacial band-bending for hole tunneling-injection, leading to relatively remarkable photon-response of PM-QAPDs in the long wavelength range.
PM-QAPDs with different P3HT layer thicknesses. (a) J−V curves, (b) EQE spectra, and (c) EQE spectra of the optimized PM-QAPDs at distinct bias. (d) Absorption spectrum of 2.4 μm P3HT/PY-IT layer, and the inserted image is the zoomed-in absorption spectrum.
PM-QAPDs with different P3HT layer thicknesses. (a) J−V curves, (b) EQE spectra, and (c) EQE spectra of the optimized PM-QAPDs at distinct bias. (d) Absorption spectrum of 2.4 μm P3HT/PY-IT layer, and the inserted image is the zoomed-in absorption spectrum.
In summary, quasi-planar PM-QAPDs were fabricated by sequential spin-coating P3HT and PY-IT layers. The part of PY-IT can penetrate into P3HT layer to emerge isolated electron-traps with PY-IT surrounded by P3HT. The optimal PY-IT solution concentration is about 1.0 mg/ml by evaluating the JPI and SNR of PM-QAPDs. The spectral response of PM-QAPDs covers from 320 to 870 nm, well corresponding with photon-harvesting range of P3HT/PY-IT layers. The EQE of PM-QAPDs with 0.25 μm thick P3HT layer arrives to 9900% at −10 V bias, which is due to efficient hole tunneling-injection from external-circuit induced by interfacial trapped electrons near the Al electrode. When P3HT layer thickness is increased to 2.4 μm, the optimal PM-QAPDs exhibit relatively narrow response from 630 to 870 nm, resulting from the adjusted trapped-electron distribution near P3HT/PY-IT interface. The quasi-planar strategy should provide an efficient method for achieving PM-QAPDs with tunable spectral response range by alerting the thickness of the donor layer.
The work was financially supported by the Fundamental Research Funds for the Central Universities (No. 2022JBZY012), the Natural Science Foundation of Beijing (No. 4232073), the National Natural Science Foundation of China (Nos. 61975006 and U22A6002), and the National Natural-Science-Foundation of China-Swedish Foundation for International Cooperation in Research and Higher education (No. 62211530056).
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
H. Zhang and F. Zhang conceived idea and designed experiments. H. Zhang, M. Liu and X. Zhao fabricated PM-QAPDs. H. Zhang, J. Li and G. Yuan carried out the characterization and analysis on PM-QAPDs. H. Zhang and F. Zhang wrote this manuscript. This manuscript has been fully discussed and revised by authors.
Haolan Zhang: Conceptualization (supporting); Data curation (supporting); Formal analysis (equal); Methodology (equal); Validation (equal); Writing – original draft (supporting); Writing – review & editing (equal). Ming Liu: Formal analysis (equal); Methodology (equal); Supervision (equal). Xingchao Zhao: Formal analysis (equal); Methodology (equal); Supervision (equal). Xiaoling Ma: Methodology (equal); Project administration (equal); Supervision (equal). Guangcai Yuan: Formal analysis (equal); Validation (equal). Junming Li: Formal analysis (equal); Validation (equal). Fujun Zhang: Funding acquisition (supporting); Project administration (equal); Resources (supporting); Software (supporting); Supervision (equal); Writing – original draft (equal); Writing – review & editing (equal).
DATA AVAILABILITY
Data sharing is not applicable to this article as no new data were created or analyzed in this study.