Charge transfer between lead halide perovskite nanocrystals and single-walled carbon nanotubes

A charge transfer study between lead halide-based perovskite nanocrystals and single-walled carbon nanotubes (PNC@CNT nanocomposite) was performed. Solution-processed MAPbX3 PNCs displayed very bright luminescence, but it quenched in the presence of CNTs. This was attributed to the electron transfer from PNCs to CNTs. The detailed changes in fluorescence lifetime were investigated through time-correlated single-photon counting (TCSPC), which suggested mixed static and dynamic quenching along with a decrease in the lifetime. Morphological changes were investigated via transmission electron microscopy (TEM) and attributed to the incorporation of PNCs on long CNTs. Also, the PNC@CNT nanocomposite was explored for photoinduced current response, which indicated an ∼3 fold increase in photoconductivity under light illumination (with a 1 mV bias). This electron transfer study between PNCs and CNTs contributes to the exploration of charge dynamics.


Introduction
Metal halide perovskites 1 have emerged among the superior materials for optoelectronic devices 2-4 including light-emitting diodes, 4 solar cells, 5,6 photodetectors, lasers, and in light harvesting. 2,3,[8][9][10][11][12] The efficacy of these applications relies on good absorption characteristics, easy charge carrier formation and transport in these materials. 2,[13][14][15] Besides, the low-cost solution synthesis of these perovskite nanocrystals (PNCs) results in intense uorescence and nearly unity quantum yield. 2,[16][17][18][19][20][21] On the other hand, the strong connement of carbon nanotubes (CNTs) to one dimension has also attracted the attention of the scientic community in the areas of charge transport, eld emission, gas storage, photodetectors and other photovoltaic applications. 22,23 Advancement in this research eld has been probed through various photo-responsive CNT-based nanocomposites with organic uorophores and inorganic moieties including conventional quantum dots (e.g., CdSe/ZnS). [24][25][26][27][28] The mechanistic investigation of CNTs and quantum dots revealed that the excited states of uorescent quantum dots donate either excited electrons or holes to CNTs, which leads to the quenching of the photoluminescence. 25 PNCs have size-dependent properties 7,17 because of the quantum connement effect, whereas CNTs have high electron transfer properties because of their nanoscale size, large surface area and direct path for charge transport. [29][30][31] The formation of nanocomposites by combining the properties of both materials will lead to new avenues of opportunities in the eld of nanotechnology. The photo-excited electron can be transferred from electron donor PNCs to electron acceptor CNTs. Earlier investigations on electron migration through static uorescence quenching have been performed by Nair et al. using FAPbBr 3 and fullerene C 60 . 32 Kuang and coworkers studied the improvements in the charge transfer and light-harvesting properties of mixed cesium lead halide PNCs and CNTs. 33 Chen and colleagues recently reported graphene sandwich stable PNC light-emissive ultrasensitive and ultrafast broadband phototransistors. 34 Herein, we report a detailed electron transfer study using single-walled CNTs and methylammonium lead halide PNCs. The preference of single-walled CNTs over C 60 fullerene is due to their superior electron transfer properties. Also, we have investigated the steady-state photoluminescence and the corresponding time-resolved lifetime, through which we have found a mixture of static and dynamic quenching (Fig. 1). The   17 Absorption peaks of MAPbBr 3 solution were observed at 462 nm and at 563 and 604 nm for MAPbI 3 in the presence of CNTs. MAPbX 3 PNCs are highly uorescent and their optical properties can be easily modied by size variation. The addition of CNTs to PNC solution leads to the adhering of quantum dots to the walls of CNTs because of van der Waals interactions. In our case, we sonicated the CNT solution for a longer time to separate the CNT bundles and increase the surface area of CNTs so that electron transfer would take place more efficiently. Pristine CNTs (not functionalized) were used, and the presence of oleic acid and oleylamine helped the adhesion of PNCs to the walls of the CNTs. To monitor the effect of solvent (i.e. CHCl 3 ), we performed the emission studies in the presence of CHCl 3 (without the addition of CNT) and with respect to time as shown in Fig. S3, † indicating the negligible effect of the solvent, which conrmed that the quenching is due to the adhesion of perovskite nanocrystals to CNTs.
As shown in Fig. 3a-c, there was almost 100% uorescence quenching of blue, green and red luminescent PNCs with CNTs, which is clear evidence that the electron transfer takes place from the electron donor PNCs to the acceptor CNT bundles. The quenching experiment was performed with the sequential addition of CNT solution (0.01% w/v in chloroform) from 10 mL to 300 mL. We observed blue shis in all PNC peaks as compared to previous reports 17 due to the dilution of samples before quenching studies. In Fig. 3a-c, the emission peaks of the blue luminescent MAPbBr 3 , green luminescent MAPbBr 3 and red luminescent MAPbI 3 (without CNTs) were observed at 443 nm, 510 nm and 596 nm, respectively.
We observed almost 100% quenching with the addition of 300 mL of CNT solution for all three PNCs. Apart from the quenching of the emission peaks, we also observed blue shis in the spectra (initially) on increasing the amount of CNT (predominantly for blue luminescent MAPbBr 3 PNCs). The rate of PL quenching was nonlinear as shown in Fig. 3d-f. A possible reason for these PNC@CNT nanocomposites to give nonlinear Stern-Volmer plots is the presence of both static and dynamic quenching. 35 Fig. 3d-f shows curves along with exponential equations and rate constants for blue, green and red luminescent PNCs, respectively.
The most efficient way to determine whether quenching is static, dynamic or both is through PL decay lifetime measurements. If the lifetime values remain unchanged in the presence of CNTs, it might be due to static quenching, as previously reported with FAPbBr 3 and fullerene. 28 However, in the case of MAPbX 3 PNCs there was a decrease in the lifetime of PNCs with increasing the amount of CNTs, which was attributed to the presence of static and dynamic quenching as shown in Fig. 3g-i. The average lifetimes along with the details of parameters for blue, red and green luminescent PNCs in the presence of CNT solution are listed in Table 1. The average lifetime was calculated by the following formula: where A 1 , A 2 and A 3 are weighing parameters, s 1 , s 2 and s 3 are the corresponding lifetimes and s avg is the average lifetime. The X-ray diffraction (XRD) pattern of MAPbBr 3 @CNT is provided in Fig. S4, † showing the presence of perovskite peaks. TEM images of blue, green and red luminescent MAPbX 3 PNCs are given in Fig. S5. † From Fig. S5, † the tunable morphology is directly evidenced by the TEM images, where the morphology changes from quantum dots to nanoplates to quantum dots by shiing from blue-green-red luminescent MAPbX 3 PNCs, which leads to adhesion to CNTs as shown in Fig. 4. These interactions are due to the electron migration or transfer from electron donor halide perovskites to electron acceptor CNTs. TEM images of PNC@CNT nanocomposites provide direct evidence of PNCs adhering to CNTs as shown in Fig. 4. The typical Fourier transform-infrared spectroscopy (FT-IR) vibration modes are presented in Fig. S6. † XPS analysis of green luminescent MAPbBr 3 in the presence of CNT conrms the presence of the elemental composition of PNC with the main peaks of Pb 4f, C 1s, N 1s and Br 3d in narrow scans as shown in Fig. S7. †

Photoresponse study
A prototype device was fabricated to investigate the conductive properties of the PNC@CNT nanocomposite. A schematic illustration is provided in Fig. 5 for measuring the photoconductivity. The device was made by two silver electrodes separated by 0.5 mm on a silica substrate and drop-casting the concentrated MAPbI 3 @CNT nanocomposite sample to connect the two electrodes. The photoinduced current was measured on This journal is © The Royal Society of Chemistry 2020 Nanoscale Adv., 2020, 2, 808-813 | 809

Paper
Nanoscale Advances a Keithley 2612B source meter while irradiating the sample with a daylight lamp having a brightness of 1200 lumens. Fig. 5a shows the current-voltage curve of the nanocomposite in the dark and in the presence of light, displaying excellent ohmic responses. The resistance in the presence of light was 14.2 U, whereas in the dark it was 15.37 U, which conrms that irradiation in light leads to an increase in current (Fig. 5a). Photoinduced charge was generated in this device system upon irradiation with light. 21 It was found that the photoconductivity increased $3 fold under light illumination at 1 mV voltage (Fig. 5b) and the photocurrent was basically constant over time as shown in Fig. 5b, indicating the good stability of the nanocomposite over a number of cycles. It has been conrmed that the presence of CNT alone does not lead to the change in current under light irradiation. 36,37 As such, the increase in the current ( Fig. 5a and b) was due to the presence of PNCs and that  Nanoscale Advances Paper led to the charge transfer between CNTs and PNCs and completed the circuit. For multiple current cycle investigations, the light was turned on and off with a 15 s interval. Photocurrent (I ph ) was evaluated by I ph ¼ |I l À I d | where I d is the current in the dark and I l is the current under illumination. As shown in Fig. 5b, the PNC@CNT nanocomposite exhibited a photocurrent which led to a difference in current under light irradiation. The device exhibited a photocurrent of 2.92 mA because of the adhesion of PNCs on CNTs through van der Waals interactions. Applied bias led to the generation of charge carriers that migrated from the electron donor PNCs to electron acceptor CNTs, leading to the enhancement of the current. These results indicate that charge transfer between PNCs and CNTs produced a noticeable photocurrent, which paves the way towards new optoelectronic and nanosensor-based applications.

Conclusions
In summary, we studied the charge transfer behaviour between lead halide based perovskites and carbon nanotubes. Fluorescence quenching and lifetime studies revealed the presence of static as well as dynamic quenching. TEM morphological studies and PL quenching also indicated signicant interaction enhancement between PNCs and CNTs. Photoconductive perovskite nanocomposites have opened new avenues for solar cells, sensor-based nanodevices and light-driven optoelectronics.

Conflicts of interest
There are no conicts of interest to declare.  This journal is © The Royal Society of Chemistry 2020 Nanoscale Adv., 2020, 2, 808-813 | 811

Paper
Nanoscale Advances