Colloidal quantum dot (CQD) light-emitting diodes (LEDs) deliver a compelling performance in the visible, yet infrared CQD LEDs underperform their visible-emitting counterparts, largely due to their low photoluminescence quantum efficiency. Here we employ a ternary blend of CQD thin film that comprises a binary host matrix that serves to electronically passivate as well as to cater for an efficient and balanced carrier supply to the emitting quantum dot species. In doing so, we report infrared PbS CQD LEDs with an external quantum efficiency of ~7.9% and a power conversion efficiency of ~9.3%, thanks to their very low density of trap states, on the order of 1014 cm−3, and very high photoluminescence quantum efficiency in electrically conductive quantum dot solids of more than 60%. When these blend devices operate as solar cells they deliver an open circuit voltage that approaches their radiative limit thanks to the synergistic effect of the reduced trap-state density and the density of state modification in the nanocomposite.
Near infrared and short-wave infrared light-emitting diodes serve a rather broad range of applications, including night vision1, surveillance2, remote sensing3, biological imaging4 and spectroscopy5. Recent progress in on-chip and wearable infrared spectroscopy for quality inspection, health and process monitoring also requires the development of highly efficient, complementary metal–oxide semiconductor-compatible and low-cost near infrared and short-wave infrared LEDs6,7,8. In contrast to other high-performance solution-processed materials, such as polymers and dyes, whose bandgaps are mainly limited in the visible, CQDs offer a unique opportunity as they readily provide access to both the visible9,10,11,12 and the infrared parts of the spectrum13. In view of this, several efforts that employ core–shell structures14,15, interdot spacing engineering16,17, chemical passivation with perovskite18 and organic–inorganic hybrid19 approaches have been made to develop efficient CQD infrared-emitting LEDs. The use of core–shell CQD structures14,15 to increase the photoluminescence quantum efficiency (PLQE) has reached external quantum efficiencies (EQEs) in excess of 4%. Alternatively, the use of appropriate host matrices has been considered as a means to suppress PLQE quenching in close-packed CQDs due to energy transfer. Initial reports have employed polymer host matrices20,21,22,23, yet with limited EQEs mainly due to the polymers’ poor electron transport properties. Recently, an alternative matrix was reported, based on perovskite materials epitaxially connected to the CQD-emitting species, that serves both as a chemical passivant of the quantum dot (QD) surface and as an efficient carrier transport matrix, which leads to an EQE of 5.2% and a power conversion efficiency (PCE) of 4.9% (ref. 18). The PCE in LEDs is defined as the ratio of the optical output power over the electrical input power, and it is of paramount importance when the power consumption of the device is considered.
We posited that, instead of relying solely on the chemical passivation of the CQD-emitting species, the use of a remote charge-passivation mechanism induced from an appropriate matrix would be more robust and efficient in reducing the trap-state density in the CQDs24,25. Unlike prior approaches18,22, our implementation is based entirely on CQD materials. In doing so, we exploited the advances made in QD solids in terms of mobility and carrier diffusion length thanks to the progress in photovoltaic (PV) devices26,27,28, in which mobilities and carrier diffusion lengths in excess of ~10−2 cm2 V−1 s−1 and 230 nm, respectively, were reported, which fulfils the needs for an efficient carrier transport in the typical thinner-than-solar-cells LED devices.
LED architecture and performance
We have considered two LED architectures, one that comprises a binary blend of small PbS QDs with a large bandgap serving as the carrier supplier for the large PbS QDs with a smaller bandgap that acts as the carrier acceptor and emitting species (Fig. 1a). The second case comprises a ternary blend formed by the binary blend with the addition of ZnO nanocrystals (NCs), which serve as a high-bandgap electron-rich transporting medium, and employed to further balance carrier injection in the active region as well as to further passivate remotely the traps of the PbS QDs24 (Fig. 1b). The ligand-exchange scheme employed for the active layers was based on a mixture of zinc iodide and 3-mercaptopropionic acid (MPA) as it has previously delivered solar cells with long carrier diffusion lengths and high open circuit voltage (VOC) values29 (Methods gives the details of the device fabrication). Both structures also employ a ZnO front layer as an electron injecting, hole-blocking layer and a 1,2-ethanedithiol-treated small PbS QD layer on top that facilitates hole injection and electron blocking at the back interface. The thickness of each of the layers used in the high-performance devices considered in this study is illustrated in the cross-sectional focused ion beam scanning electron microscopy images in Fig. 1c,d for the binary and ternary blends, respectively. Typically, the optimal thicknesses of the electron-injecting, active and hole-injecting layers are around 80 nm, 60 nm and 70 nm, respectively. The transmission electron microscopy (TEM) images in Fig. 1 illustrate the effective blending of these QD species at the nanoscale and support the nature of the nanocomposite active layer. The corresponding band diagrams of the constituent materials used in these devices are shown in Fig. 1e, taken from ultraviolet photoelectron spectroscopy (UPS) measurements28,30. According to this, both ZnO NCs and small PbS QDs serve as a type I heterostructure with the emitting large PbS QDs. The small PbS QD matrix forms a marginal type I heterostructure with the large PbS QD emitters in which the band offset confinement for both electrons and holes is between 0.1 and 0.2 eV. Based on the band diagram, electron transport and injection take place within the ZnO NC and the small PbS QD matrix given their matched conduction band levels, whereas hole transport is facilitated largely via the small PbS QD matrix.
Figure 2a shows the radiance of the binary and ternary blend-based LED devices with applied bias voltages. The control device, which comprises only large PbS QDs as the active layer, is also plotted for comparison. All the devices showed a very high radiance of ~9 W sr−1 m−2 at 3.5 V, which is more than 50% higher than previously reported PbS QD-based infrared LEDs16,18. Note that the turn-on voltage for both binary and ternary blend devices was around 0.6 V, that is, below the bandgap of emission, whereas the turn-on voltage of the control device was 0.87 V, that is, matching closely the bandgap of emission (note that we considered 1 nW radiance as the turn-on power). The electroluminescence spectra with different values of voltage bias for the ternary device are shown in Fig. 2b, with a clear band-edge electroluminescence emission at sub-bandgap voltages. Our PbS QD-based LED thus shows a notably low turn-on voltage. Although this below-bandgap turn-on value is not thermodynamically favourable in the absence of multicarrier processes, it has been reported previously for polymer-based31 and QD-based LEDs11, attributed to Auger-assisted charge-injection processes. According to this, the low-energy barrier for electron injection results in electron accumulation at the active layer and hole-transporting layer interface which can fulfil the condition for Auger-assisted charge injection11,31,32,33. In our case, we attribute this partially to the improvement of mobility and trap passivation with the mixed ligand treatment employed herein as well as to the use of bulk heterojunctions. To examine the hypothesis about the role of the ligand passivation, we fabricated a ternary blend active layer device using only MPA as the ligand (the MPA ligand exchange yields carrier mobility lower than the ZnI2–MPA treatment). The device showed a much higher turn-on voltage (2.5 V) and lower radiance compared to the devices reported in this manuscript (Supplementary Fig. 2). An additional plausible mechanism can be assigned to Auger-assisted recombination in the small PbS QD matrix in which hole transport takes place via midgap delocalized states34,35 and the recombination with electrons transfers the energy to the remaining holes in those states, which enables them to move to the valence band and subsequently inject into the emitting QDs. We consider that further studies are needed to shed light on this interesting effect.
Despite the similar radiance measured across those three LED devices, a large difference in their driving currents was recorded, with the control (single) device yielding a very high leakage current that progressively decreased in the binary and ternary blend cases, as shown in Fig. 2c. This has a significant effect on the EQE of the LEDs, as shown in Fig. 2d. The peak EQE of the ternary device reached 7.87% (average, 7.11 ± 0.30% (Supplementary Fig. 3)) compared to 0.38% (0.3 ± 0.07%) for the single and 4.5% (4.12 ± 0.29%) for binary QD-based devices. The highest EQE found here, for CQD LEDs that emitted at a 1,400 nm wavelength, can be compared with previously reported devices that showed peak EQEs around 5% for a similar spectral range14,15,18. We attribute this to the trap passivation and the reduction of leakage current in the blend devices. The high EQE recorded in the binary and ternary blend devices, taken together with the low turn-on voltage, also results in a high PCE. Figure 2e plots the PCE of the LED devices, from which the device based on the ternary blend performs with a peak PCE as high as 9.3%, a nearly twofold improvement over values in ref. 18. The synergistic use of both PbS QD matrix dots and ZnO NCs was instrumental in reaching the high EQE. Binary control devices that comprise PbS QD emitters with ZnO NCs reached an optimum EQE of 1.8% (Supplementary Fig. 4).
In the course of device optimization and to demonstrate the versatility of this approach, we fabricated and measured ternary devices with an optimized blend ratio by varying the bandgap of the PbS QD matrix as well as the bandgap of the PbS QD emitters, the results of which are summarized in Table 1. The electroluminescence spectra of various emitting PbS QD bandgaps are shown in Supplementary Fig. 5a. The variation of the PbS QD matrix bandgap (Table 1 and Supplementary Fig. 5b) revealed that optimized EQEs are achieved with increasing the bandgap of the PbS QD matrix, probably due to a more efficient charge transfer to the emitting PbS QDs and an increased confinement in the emitting QDs due to the larger band offset with the surrounding PbS QD matrix. Given that our devices are based entirely on CQD components with a reported high stability28,36, we tested the stability of the best-performing device under a constant applied current over a period of 48 hours. The radiance and the EQE of the device are highly stable throughout the course of the test, as shown in Fig. 2f, especially given that the devices were fabricated and characterized in ambient air conditions without any encapsulation.
The origin of this high EQE lies in the very high PLQE of the QD films employed, and we therefore explored in more depth the role of the host matrix on the PLQE of the emitting QD species. Figure 3a shows the photoluminescence (PL) spectra for the ligand-exchanged binary blend films with varying loadings of the emitting QDs in the small PbS QD matrix. We first note that the position of the PL peak of the emitter QDs does not alter with varying concentration, which suggests the absence of exciton transfer among them, consistent with the Förster radius value of 3.5 nm for the energy transfer among the emitter QDs and their much larger interdot spacing in the blends (Supplementary Section 6). The absence of PL emission from the matrix QDs on blending corroborates the highly efficient carrier transfer from the matrix QDs to the emitting QDs and suggests that energy transfer between these species is not present, given the small Förster radius of 1.6 nm for the matrix–emitter QD system (Supplementary Section 6). This finding is in accordance with prior reports in which such a high charge transfer has been ascribed to the increased coupling strength enabled in ligand-exchanged electronically coupled QD solids37. Here we made use of this effect to develop QD solids with a high PLQE under the following rationale: for a given density of poorly emitting (defective) large PbS QDs and a given QD film volume, we envision a much higher PLQE when the large PbS QDs are introduced into a small PbS QD matrix, due to the proportionally lower number of defective large PbS QDs existent in the same volume. This hypothesis is valid when the matrix is absorbing at the optical excitation wavelength and possesses long carrier diffusion lengths to supply the emitting QD species with electrons and holes. The schematic representation of this mechanism is illustrated in Fig. 3d,e. PLQE measurements of such binary blends are in agreement with the proposed mechanism. Figure 3b plots the PLQE values of the binary blends against varying the emitter QD loading in the host PbS QD matrix. An optimum PLQE of 60% is recorded for a loading of 10% of emitter PbS QDs in a donor PbS QD host matrix, drastically increased over the 2.3% PLQE of the neat emitter PbS QD film. This is a remarkably high PLQE given that it refers to electronically conductive ligand exchanged QD solids. Such high PLQE values for the binary case should also express in long PL lifetimes. The transient PL of the 7.5% loaded binary blend, shown in Fig. 3c, yields a bi-exponential decay with corresponding lifetimes of 632 ns and 2.99 μs (fitting values are summarized in Supplementary Section 7). The first decay is attributed to the intrinsic PL lifetime of the emitter QDs, whereas the longer component is associated with the carrier supply time of photogenerated carriers in the matrix of the binary blend to diffuse and excite the emitter QDs. The shorter component of 632 ns is in agreement with the recorded PLQE values given that the ideal radiative lifetime of PbS was reported as on the order of 1 μs (refs 38,39). Increasing the amount of emitter QDs in the matrix reduces the PLQE as a result of increasing the probability of non-radiative recombination in the defective emitter QDs, whereas decreasing the amount of emitter QDs in the matrix reduces the PLQE as a result of increasing the probability of non-radiative recombination in the host matrix, given a carrier diffusion length of around 70 nm in the small PbS QD matrix37. Although the 10% emitter QD-based binary blend yields the best PLQE, it does not yield the best EQE among the different binary blends (Supplementary Fig. 10). We attribute this to the trade-off between an efficient balanced charge injection and the PLQE.
The synergistic use of ZnO NCs, in addition to improving the balance of charge injection (as shown in Fig. 2c), offers the benefit of remotely passivating electron traps in the PbS QD emitters. The electrons of the n-type ZnO NCs, when in the vicinity of PbS QDs with electron traps available for population, are used to electrically passivate those electron traps and result in higher radiative recombination efficiencies24,25 (Fig. 3f). The PLQE of the ternary blend films as a function of PbS QD emitter loading in the PbS QD host matrix at a given ZnO loading of 40% is presented in Fig. 3b. According to this, a further increase in PLQE approaching a value of 80% for a loading of 20% of the emitting QD species is observed. The PLQE of the ternary blends with different PbS QD and ZnO ratios are shown in Supplementary Fig. 11. In all types of PbS QD combinations, a 40% ZnO loading in the ternary blend gives the best PLQE and also the best EQE in LED devices, as shown in Supplementary Fig. 12. Note that the loading of emitter QDs in the donor matrix or the addition of ZnO NCs does not cause any spectral shift in the PL (Supplementary Fig. 13), which thus suggests that the observed increase in the PLQE of the emitting PbS QDs is not due to the suppression of energy transfer among them. Similar to the binary case, a transient PL measurement of the ternary system exhibits a bi-exponential PL lifetime: the first component, associated with the radiative lifetime, is on the order of 827 ns, and the second one is longer-lived, associated with the carrier diffusion from the matrix PbS QDs to the emitter QDs, on the order of 3 μs.
These high PLQE values are also in line with the quantum efficiency values reported in the LED devices. In Supplementary Fig. 14, we plot the internal quantum efficiency (IQE) of the LED devices defined as the ratio of photons generated inside the device to the injected charge carriers. The ternary blend devices yield a peak IQE value of 33.3%, whereas the binary blend shows an IQE of 17.5%. The improvement in EQE and thereby IQE is nearly twofold, that is, much higher than the improvement in the PLQE. In Table 2 we summarize the different efficiency factors that determine the EQE of an LED according to the equation18:
where the carrier supply efficiency (ηcarrier supply) measures how the efficiently balanced carrier injection takes place and is transported at the active layer of the LED. IQE is estimated as per Supplementary Fig. 14 using the experimentally determined EQE and the optical loss model. The experimentally determined values of IQE and PLQE allow us to estimate ηcarrier supply for the three classes of devices. Although the single LED yields the highest carrier supply efficiency, its performance is limited by the low PLQE of the active layer. The binary LED improves on the PLQE factor, yet suffers from a low carrier supply to the emitter QDs. The ternary system yields a more efficient carrier supply over the binary along with a higher PLQE, which leads to an overall higher EQE. This shows that our current LED devices are carrier-supply limited, and further improvement in the EQE can be within reach through device optimization. It also supports the hypothesis that the use of ZnO serves a twofold role: it improves the trap passivation and PLQE, and also improves the balance of carrier injection at the optimal LED configuration (40% ZnO loading).
Performance of blend devices as solar cells
A highly performing LED material, that is, one having a very high PLQE, is expected to be the ideal material for solar cell applications in demonstrating a VOC very close to the radiative limit40,41,42. Such high PLQE values recorded in the electrically conductive QD solids led us to test this expectation. We therefore constructed PV cells based on the architectures considered previously for the LED devices, essentially mimicking the LED structures but with a much thicker active layer to facilitate a high optical absorption. The thicknesses of the electron-transporting, active and hole-transporting layers were around 40 nm, 220 nm and 30 nm, respectively, and the PV structure followed a typical CQD solar cell43. The dark current density–voltage (J–V) characteristics, plotted in Fig. 4a, provide the initial features of a suppressed recombination on blending, as evidenced by the significantly lower reverse current of the blended-layer diodes compared to the single-QD-layer diode case. The J–V curves of those cells under simulated AM1.5 solar illumination are shown in Fig. 4b. The VOC of the binary device increased to 0.59 V from 0.39 V of the device based on a single QD, which is a typical VOC value for PbS QD solar cells of the same bandgap28. The large VOC deficit (defined as the deficit of VOC from the bandgap voltage) for PbS QD-based solar cells has been believed to be a result of a significant presence of non-radiative, in-gap traps44. The increased VOC (and subsequent decrease in VOC deficit) can therefore stem from a reduced trap-state density, in accordance with the drastically improved PLQE values recorded for the binary blend films (Fig. 3b). The addition of ZnO further improves VOC to a value of 0.69 V for the ternary blend device. This is a notably large value of VOC that approaches the radiative limit40 given that the solar cell harnesses photons with energies down to 0.92 eV. We calculated the radiative VOC limit of the PV devices following a standard analysis (Supplementary Section 13)40,41. Figure 4c summarizes the radiative limit and non-radiative loss of the VOC. Non-radiative losses decrease with binary blending compared to the single QDs and further decrease in ternary blend devices with ZnO loading. The EQE spectra of these solar cell devices (Fig. 4d) demonstrate their spectral reach down to 1,400 nm, determined by the low-bandgap PbS QDs. Naturally, the EQE in the infrared is reduced in the case of the binary blend due to the low loading of the large PbS QDs in the blend and its consequent lower absorption. In the case of the ternary blend, we observe a more drastic reduction in EQE across the spectrum, which is ascribed to the poor electron transport through the ZnO NC network24,25.
Theoretically, Gong et al. showed the influence of trap-state reduction on the enhancement of radiative recombination18. To further characterize and quantify the degree of trap-state passivation on blending, we employed thermal admittance spectroscopy (TAS). TAS allows us to obtain a quantitative picture of the in-gap trap distribution of the photoactive material, as previously demonstrated on different types of PV devices45,46. The detailed analysis of the employed method is described in Supplementary Section 14. The trap distribution as a function of Et (the position of the trap state with respect to the band edge) deduced from the TAS analysis gives a clearer picture of the trap reduction on QD blending, as shown in Fig. 4e. Et decreases only slightly from 0.254 eV for a single-QD-based device to 0.237 eV for binary blending and further reduces to 0.174 eV for ternary blends. The position of Et follows a similar trend to that of trap activation energy (EA) as determined from the Arrhenius plot (Supplementary Fig. 18). The overall trap-state density decreased from 1016 cm−3 for single-QD-based device to 4.8 × 1015 cm−3 for binary QDs and further down to 6.5 × 1014 cm−3 for a ternary QD-based device. These results from TAS also corroborate our hypothesis on the passivation mechanism at play on binary and ternary blending. In the case of binary blending there is an overall effective reduction of trap states without much alteration of the energetic value of the traps, that is their nature; this is in accordance with the model described in Fig. 3e. On ternary blending, in addition to the reduction of the trap density, there is also a significant lowering of the trap-state depth due to the remote passivation facilitated by the ZnO NCs (Fig. 3f). Note that the TAS technique measures the overall trap-state density of the blend and cannot distinguish between traps in the emitter QDs and the matrix QDs. However, in the case of the ternary blend, the resultant trap-state density per emitter QD is one order of magnitude lower than that of the single QD device, which provides strong evidence of the trap-state reduction on the use of ZnO NCs (Supplementary Section 15).
To thoroughly examine whether the trap-state reduction is solely responsible for this notable increase in VOC, we performed solar cell capacitance simulator (SCAPS) simulations of a single-layer device in which we varied the trap-state density across the experimentally obtained range of values. According to SCAPS, the expected VOC increase from single to binary and then to ternary devices is 40 mV and 100 mV, respectively, whereas the observed increase in VOC from single to binary is around 200 mV. SCAPS simulations confirm that on trap-state reduction from single to binary, VOC increases but to a lesser extent than we experimentally measured, whereas the further increase in VOC from the binary to ternary is accounted for by trap-state reduction. We therefore sought additional mechanisms at play in the binary blends that may contribute to this high value of Voc. The binary blend devices comprise ~7.5% of lower bandgap emitter QDs embedded in a larger bandgap matrix of QDs. Therefore, the density of states (DOS) of the low bandgap QDs in a binary device is reduced over the DOS of the single-layer counterpart. In Supplementary Section 16, we present a more detailed model that takes into account the effective reduction of the DOS of the binary blend. This reduction of the DOS yields an improvement in VOC of ~120 mV when only the emitter QDs are excited and ~160 mV on the excitation of both the matrix and emitter QDs. During the revision of this manuscript, Sun et al. also reported the effect of DOS reduction on the VOC of CQD blended solar cells, yet in a different bandgap mixing regime47. To further corroborate this hypothesis, we plot in Fig. 4f the intensity and wavelength dependence of the VOC for the three classes of devices. In agreement with the SCAPS modelling (Supplementary Fig. 22), the VOC of the single device is independent of the photon energy for the same intensity, whereas in the case of the blend devices, lower-energy photons of 1,310 nm (that excite only the low bandgap QDs) yield a lower VOC than the higher-energy photons of 637 nm (that excite both the low bandgap and matrix QDs). These measurements support our hypothesis of the combined effect of trap-state reduction and DOS modification on the VOC improvement with blend structures compared to the emitter-only devices.
In summary, we report a new approach to the engineering of QD solids at the supra-nanocrystalline level that has led to a low trap-state density, high PLQE values in solid-state conductive QD films and thereby highly efficient LEDs. The use of different bandgap QDs also offers a leverage to tune the DOS in quantum-confined nanocomposite solids, which has allowed us to reach a high VOC when these devices are operated as solar cells. This work offers new insights into engineering the energetic potential landscape of QD solids with important implications towards higher-performance light emitters and solar cells.
Synthesis of PbS QDs
The Schlenk technique was used to synthesize PbS QDs. PbS QDs based on the 830 nm and 940 nm excitonic peaks were synthesized following the standard recipe. Lead oxide (2 mmol), oleic acid (4.7 mmol) and 1-octadecene (ODE) (9.4 mmol) were pumped overnight at 95 °C. Then 15 ml of ODE was added and the temperature of the reaction adjusted to 75 °C or 100 °C for the 830 nm and 940 nm PbS QDs, respectively. When the temperature point is reached, 1 mmol hexamethyldisilane mixed with 10 ml of ODE was immediately injected. The heating was stopped (without removing the heating mantle) and the sample allowed to cool gradually (~1 h). The NCs were isolated by adding acetone and then centrifuged, purified by dispersion/precipitation with toluene/acetone three times and finally dispersed in anhydrous toluene (30 mg ml−1) before using them for device formation.
PbS QDs based on the 1,300 nm excitonic peak were synthesized by a previously reported multi-injection method with modifications48. Typically, lead oxide (0.45 g), oleic acid (3.8 ml) and ODE (50 ml) were mixed together at 95 °C under vacuum for 12 h. Then, the temperature of the reaction was raised to 100 °C. The solution of 90 μl of hexamethyldisilane in 3 ml of ODE was injected, and an additional three injections (25 μl of hexamethyldisilane in 3 ml of ODE for each injection) were sequentially followed at a fixed time. When the injections were finished, the heating was stopped immediately and the sample was allowed to cool gradually to room temperature under constant stirring. QDs were precipitated by adding acetone, followed by centrifugation and purified in air by using toluene/acetone as the solvent/antisolvent. The final QDs were dispersed in toluene with a concentration of 30 mg ml−1 for the device fabrication.
ZnO NC preparation
ZnO NCs were prepared following a previously reported method24. Zinc acetate dihydrate (2.95 g) was dissolved in 125 ml of methanol under vigorous stirring and the temperature of the solution was set at 60 °C. At the same time, in a separate vial, 1.48 g of KOH (90%) was dissolved in 65 ml of methanol solution. The prepared KOH solution was then added dropwise to the zinc acetate solution for a period of 4 min with the temperature kept at 60 °C with constant stirring. The reaction conditions were left unaltered for the next 2.5 h. After completion of the reaction, the heating source was removed and the solution was allowed to cool down slowly to room temperature. The solution was then centrifuged at 3,500 r.p.m. for 5 min. The supernatant was discarded and an equal amount of methanol was added and the centrifugation repeated. After three rounds of purification, the NCs were dispersed in a solution of 2% butylamine in chloroform for the base layer formation and in 5% butylamine in toluene for the ternary blend formation.
LED device preparation
LEDs were prepared on cleaned indium tin oxide (ITO)-coated glass. The electron-transporting layer (ZnO) was prepared by spin-coating ZnO NCs in chloroform (40 mg ml−1) with a spin speed of 4,000 r.p.m. The procedure was repeated once more to have a thicker film of approximately 80 nm. The active emitting layer was grown on top of the ZnO layer. Before mixing, all the QD and ZnO NC solutions were prepared in separate vials with the same concentration (30 mg ml−1). For binary blends, the emitter PbS QDs were mixed with donor PbS QDs of different volume ratios. Ternary blends were formed by mixing the ZnO NC solution with the binary blend at different volume ratios. During the film formation, QDs were treated with the ZnI2 and MPA mixed ligand, as described in our previous report29. The mixed ligand was prepared by mixing 25 mM ZnI2 in methanol and 0.015% MPA in methanol solutions. The ZnO substrates were covered with 50 μl of QD solutions and spun immediately at 2,500 r.p.m. for 15 s. Then, the spin-coater was stopped to add a few drops of mixed ligand to treat for 5 s. After that, the spin-coater was started again to dry the film, which was then washed with a few drops of methanol. The procedure was repeated thrice to get an average thickness of 60 nm. The hole-transporting layer was formed by using small diameter PbS QDs treated with 0.02% ethanedithiol in acetonitrile solution. The back electrode was formed with Au deposition through a pre-patterned shadow mask in thermal evaporator (Nano 36 Kurt J. Lesker) at a base pressure of 10−6 mbar. The active area for each device was 3.14 mm2.
LED performance characterization
All the devices were fabricated and characterized in ambient air conditions. The J–V characteristics were recorded using a computer-controlled Keithley 2400 source measurement unit. To calculate the EQEs, electroluminescence from the front face of the device was detected using a calibrated Newport 918D-IR-OD3 germanium photodetector connected to a Newport 1918-C power meter in parallel to the J–V measurements. A Shadow mask of 3 mm in diameter was placed in front of the device to minimize the waveguide effect from the ITO-coated glass. Lambertian emission was assumed. The thickness of the glass substrate was considered during the solid angle measurement. The radiance was further verified with a NIST (National Institute of Standards and Technology)-certified 818-IG InGaAs photodetector with a calibrated DB15 module by Newport.
PL, PLQE and PL decay measurements
PL measurements were performed using a Horiba Jobin Yvon iHR550 Fluorolog system coupled with a Hamamatsu RS5509-73 liquid-nitrogen cooled photomultiplier tube and a calibrated Spectralon-coated Quanta-phi integrating sphere. All the reported steady-state PL spectra and PLQE measurements were collected using a continuous-wave Vortran Stradus 637 laser diode as the excitation source (λ = 637 nm, maximum power = 80 mW) and all the reported steady-state PL spectra were corrected for the system-response function. Both binary and ternary blend films used in the PL measurements were prepared via the aforementioned layer-by-layer deposition using the mixed-ligand-exchange procedure (five layers) onto (3-mercaptopropyl)trimethoxysilane-functionalized glass. This treatment was used to improve the adhesion of the QDs on glass to obtain continuous and uniform QD films. The treatment was performed prior to QD deposition by inserting 1 × 1 cm glass substrates into a Petri dish that contained a (3-mercaptopropyl)trimethoxysilane toluene solution (10% by volume), then removed after 24 h and gently dried under a flux of nitrogen.
The PLQE measurements were carried out as follows. Initially, the PLQE of a 100 nm thick film of PbS QDs (reference film) was measured inside the calibrated integrating sphere using the procedure proposed by de Mello et al.49 (λPL = 1,125 nm; the film was fabricated following the method described above) (Supplementary Section 17) to yield a PLQE of 24 ± 3%. The emission intensity of the film was then measured in the sample chamber of the fluorimeter (that is, outside the integrating sphere) using predetermined and fixed excitation and collection conditions. By comparing the emission intensities measured outside and inside the integrating sphere, and the optical absorption at the excitation wavelength obtained inside the integrating sphere, we obtained a correction factor (that is, the PLQE of the reference film calculated using the number of emitted photons measured in the fluorimeter sample chamber and the number of absorbed photons measured in the integrating sphere was 48 times higher than the real value). Afterward, all the films were measured using the same procedure (number of absorbed photons measured inside the sphere and number of emitted photons measured in the fluorimeter sample chamber) using the predetermined and fixed excitation and collection conditions employed for the reference film, and the correction factor was applied to obtain the PLQE value.
The above method was employed for all the PLQE values reported in this article as it was not possible to use the Quanta-phi integrating sphere for λPL > 1,200 nm as the fibre bundles that interfaced the Quanta-phi integrating sphere to the Horiba Jobin Yvon iHR550 Fluorolog system are not transparent in this spectral range. The obtained correction factor from the reference film accounts for the response function of the whole fluorimeter and detector. Yet, given the different emission wavelength of the reference film (λPL = 1,100 nm) and the blended films (λPL = 1,371 nm), a deviation of the correction factor value can arise from the different spectral reflectivity of the sphere, although Spectralon presents a small reflectance variation in this spectral region (less than 1%). The PL decay measurements were performed using a Horiba SpectraLED S-625 excitation source (λ = 637 nm) with a time resolution of 300 ns and Fluorolog hub time-correlated single counting photon card Hamamatsu RS5509-73 liquid-nitrogen cooled photomultiplier tube.
Spectral electroluminescence measurements were also performed using a Horiba Jobin Yvon iHR550 Fluorolog system coupled with a Hamamatsu RS5509-73 liquid-nitrogen cooled photomultiplier tube. The voltage bias to the device was applied with a Keithley 2400 source measurement unit. The acquired spectra were corrected using the system response factor provided by the manufacturer.
PV device preparation and characterizations
The PV device preparation follow a similar procedure as that for the LED described above other than the thickness of the respective layers. The ZnO base layer was deposited more thinly compared to the LED device (~40 nm). The active layer was prepared much thicker to absorb sufficient photons (~200–220 nm). Finally, two layers of ethanedithiol-treated PbS (~30–35 nm) were used as the electron-blocking layer. Approximately 100 nm of Au was deposited as the back electrode. The active area of the device was 3.14 mm2. All the PV characterizations were performed in ambient conditions. The device current–voltage responses were collected using a Keithley 2400 source meter. The illumination intensity of AM1.5 was maintained using a class AAA solar simulator (Oriel sol3A, Newport Corporation). The accuracy of the measurement was determined as ±4%. EQE measurements were performed with an in-house built experimental set-up by using chopped (220 Hz, Thorlab) monochromatic illumination. The power was measured with a calibrated Newport-UV power meter. The device response of the chopped signal was measured using a Stanford Research system lock-in amplifier (SR830), which was fed by a Stanford Research system low noise current pre-amplifier (SR570). The final EQE spectra were obtained with the help of the LabVIEW program. The intensity-dependent VOC measurements were performed with Vortran (637 nm) and Superk Extreme (1,310 nm (NKT Photonics)) lasers.
The measurements were performed with the PV devices in a Lakeshore four-probe cryogenic chamber controlled by a Lakeshore-360 temperature controller. The frequency-dependent capacitance was measured with an Agilent B1500 connected to an external capacitance measurement unit. The temperature was varied from 220 K to 320 K to acquire the frequency-dependent capacitance variation. The voltage-dependent capacitance was measured with the same instrument to obtain the value of depletion width and built-in voltage. The detailed data analysis procedure is described in the Supplementary Section 14.
The bright-field TEM images of the films were obtained with a JEOL JEM-2100 (LaB6 electron gun) transmission electron microscope, which was operated at 200 kV. The samples were prepared by spin-coating the QD solutions onto a 300-mesh carbon-coated copper grid at 2,500 r.p.m. Then, the ligand exchange with ZnI2–MPA was performed in line with the aforementioned device fabrication procedure.
The data that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The authors acknowledge financial support from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 725165), the Spanish Ministry of Economy and Competitiveness (MINECO) and the ‘Fondo Europeo de Desarrollo Regional’ (FEDER) through grant TEC2017-88655-R. The authors also acknowledge financial support from Fundacio Privada Cellex, the program CERCA and from the Spanish Ministry of Economy and Competitiveness through the ‘Severo Ochoa’ Programme for Centres of Excellence in R&D (SEV-2015-0522). F.D.S. and S.C. acknowledge support from two Marie Curie Standard European Fellowships (NANOPTO, H2020-MSCA-IF-2015-703018 and NAROBAND, H2020-MSCA-IF-2016-750600).