⇦Chinese Physics Letters, 2018, Vol. 35, No. 2, Article code 028401 Ultrafast Terahertz Probes of Charge Transfer and Recombination Pathway of CH$_{3}$NH$_{3}$PbI$_{3}$ Perovskites * Hui-Jie Yan(严慧婕)1, Zhi-Liang Ku(库治良)2, Xue-Feng Hu(胡雪峰)1, Wan-Ying Zhao(赵婉莹)1, Min-Jian Zhong(钟敏建)1, Qi-Biao Zhu(朱绮彪)1, Xian Lin(林贤)1, Zuan-Ming Jin(金钻明)1**, Guo-Hong Ma(马国宏)1** Affiliations 1Department of Physics, Shanghai University, Shanghai 200444 2State Key Laboratory of Advanced Technology For Materials Synthesis and Processing, Wuhan University of Technology, Wuhan 430070 Received 21 September 2017 *Supported by the National Natural Science Foundation of China under Grant Nos 11674213, 11604202 and 61735010, the Young Eastern Scholar at Shanghai Institutions of Higher Learning under Grant No QD2015020, the Universities Young Teachers Training Funding Program under Grant No ZZSD15098, and the 'Chen Guang' Project of Shanghai Municipal Education Commission and Shanghai Education Development Foundation under Grant No 16CG45.
**Corresponding author. Email: physics_jzm@shu.edu.cn; ghma@staff.shu.edu.cn Citation Text: Yan H J, Ku Z L, Hu X F, Zhao W Y and Zhong M J et al 2018 Chin. Phys. Lett. 35 028401 Abstract We use transient terahertz photoconductivity measurements to demonstrate that upon optical excitation of CH$_{3}$NH$_{3}$PbI$_{3}$ perovskite, the hole transfer from CH$_{3}$NH$_{3}$PbI$_{3}$ into the organic hole-transporting material (HTM) Spiro-OMeTAD occurs on a sub-picosecond timescale. Second-order recombination is the dominant decay pathway at higher photo-excitation fluences as observed in neat CH$_{3}$NH$_{3}$PbI$_{3}$ films. In contrast, under similar experimental conditions, second-order recombination weakly contributes the relatively slow recombination between the electrons in the perovskite and the injected holes in HTM, as a loss mechanism at the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD interface. Our results offer insights into the intrinsic photophysics of CH$_{3}$NH$_{3}$PbI$_{3}$-based perovskites with direct implications for photovoltaic devices and optoelectronic applications. DOI:10.1088/0256-307X/35/2/028401 PACS:84.60.Jt, 78.47.-p, 07.57.Ty © 2018 Chinese Physics Society Article Text Organic–inorganic hybrid systems based on lead halide compound have recently achieved considerable success as light harvesters in solid-state solar cells.[1-6] Several groups have reported power conversion efficiencies (PCEs) from 10% to 20% for solution processed organometal halide perovskite (OMHP)-based solar cells in the past five years. Slow electron–hole recombination and persistent high mobility are essential features for an efficient solar cell.[7-10] The fundamental photophysical processes underlying solar cell function need to be understood to fully utilize the properties of these materials. Further work is needed to obtain clearer correlations among spectroscopic absorbance, transport, sample structure, interface, defect, excitation condition, and so on.[11,12] The charge transport in ordered crystalline semiconductors is largely delocalized. The photocarrier lifetime is defined by the mechanisms of photocarrier energy dissipation.[13-15] In contrast, the recovery kinetics in perovskites is modelled by the rate equation,[16] $-\frac{dn(t)}{dt}=k_1n+k_2n^2+k_3n^3$, where $n$ is the photogenerated carrier density, and $k_{1}$–$k_{3}$ represent the monomolecular process (trap-assisted recombination), bimolecular process (free carrier recombination) and trimolecular process (nonradiative Auger recombination), respectively. Indeed, to elucidate the role of free carriers and the carrier transfer between OMHP and electron- or hole-acceptor materials would provide a deeper understanding of the mechanisms that give rise to the high performance of hybrid perovskite-based solar cell devices.[17] Generally, by using various time-resolved spectroscopic techniques, such as transient absorption and photoluminescence spectroscopy, the quenching of the photo-induced absorption or suppression of the fluorescence of OMHP in the presence of Spiro-OMeTAD (extensively used as a hole transport material (HTM)) has been observed qualitatively to investigate the hole injection dynamics.[18-23] In this work, we applied time-resolved terahertz spectroscopy to focus on the carrier dynamics of all occurring photo-induced charge transfer processes and to derive the effective mechanism for charge separation at the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD interface. Our quantitative information is important for understanding the mechanism of the devices and eventually for the improvement of their photovoltaic performances. To prepare the CH$_{3}$NH$_{3}$PbI$_{3}$ film, the PbI$_{2}$ film was firstly deposited on the quartz slide by spin-coating a PbI$_{2}$ solution (462 mgml$^{-1}$ in DMF) at 3000 rpm for 30 s. After drying at 70$^\circ\!$C, the as-prepared PbI$_{2}$ film was dipped in a solution of CH$_{3}$NH$_{3}$I in 2-propanal (10 mgml$^{-1}$ for 20 s and rinsed with 2-propanal, followed by annealing at 100$^{\circ}\!$C for 1 min in a glovebox. For the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD film, the Spiro-OMeTAD was deposited on top of the as-prepared CH$_{3}$NH$_{3}$PbI$_{3}$ film by spin-coating a Spiro-MeOTAD solution containing 0.17 M Spiro-MeOTAD, 0.198 M 4-tert-butylpyridine (TBP) and 0.064 M lithium bis(trifluoromethylsul-phonyl)imide (LiTFSI) in chlorobenzene at 3000 rpm for 30 s. In our measurements, the films of CH$_{3}$NH$_{3}$PbI$_{3}$ and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD were deposited on fused silica substrates. Our deposited films had an average thickness of 400 nm$\pm$100 nm. The samples were characterized by UV-visible absorption spectrum, scanning electron microscopy (SEM) and x-ray diffraction (XRD) pattern, as shown in Fig. 1. SEM revealed micron-sized crystalline domains of CH$_{3}$NH$_{3}$PbI$_{3}$ on fused silica. Room temperature absorption revealed a band gap onset at around 770 nm and a distinct absorption peak at around 480 nm. XRD revealed 14.1$^{\circ}$ and 28.2$^{\circ}$ peaks for (110) and (220), respectively, which are characteristics for the pristine perovskite with tetragonal crystal structure at room temperature, in agreement with earlier works.[3]
cpl-35-2-028401-fig1.png
Fig. 1. The characterization of the thin film: (a) SEM image, (b) absorption spectra and (c) XRD pattern of the samples for THz measurements.
Our THz setup was driven by a Ti:sapphire femtosecond amplifier, operating at repetition rate of 1 kHz, and generating 120 fs pulses at central wavelength 800 nm. A portion of the laser output was split as a pump beam, after second harmonics with a $\beta$-BaB$_{2}$O$_{4}$ (BBO) crystal for the above band gap of the CH$_{3}$NH$_{3}$PbI$_{3}$ excitation at 400 nm.[24] The residual beam was used to generate and detect the THz radiation by optical rectification and electro-optical sampling using ZnTe nonlinear crystals with a frequency window of 0.3–2.0 THz. Both optical pump and terahertz probe pulses were incident on the sample at normal incidence. The in-plane conductivity of the photoexcited sample was probed with the terahertz pulse.[25-27] The terahertz spectroscopy measurement enclosure was purged with $N_{2}$ to minimize photo-oxidization and water vapor interference effects. No significant degradation was observed during the experiments. The change in peak amplitude of the THz waveform is proportional to the real part of the photo-induced conductivity, mostly arising from the influence of mobile charge carriers.[28-30] By varying the time delay between the 400 nm optical pump and THz probe pulses, we acquired the dynamics of the real-valued sheet photoconductivity $\Delta \sigma _{\rm 2D}$, which can be reconstructed from the photo-induced peak change in the THz electric-field amplitude as a function of the pump-probe time delay, using a thin-film approximation known as the Tinkham equation[31] $$\begin{align} \Delta \sigma _{\rm 2D}=\frac{1+n}{Z_0}\Big(\frac{1}{1+\Delta E/E}-1\Big),~~ \tag {1} \end{align} $$ where $n=1.95$ is the THz refractive index of the fused silica substrate, and $Z_{0}=377{\it \Omega}$ is the impedance of free space.
cpl-35-2-028401-fig2.png
Fig. 2. Characteristic evolution of the pump-induced photoconductivity dynamics obtained by OPTP on (a) the CH$_{3}$NH$_{3}$PbI$_{3}$ neat film and (b) the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD film. The two samples were excited at wavelength of 400 nm with fluences of 27 and 72 μJ/cm$^{2}$. [(c), (d)] Normalized photoconductivity transients measured at two fluences for the two samples, respectively. Thick solid lines represent bi-exponential fits of the experimental data.
Figures 2(a) and 2(b) show the photo-induced sheet real-valued conductivity $\Delta \sigma _{\rm 2D}$ of the two samples studied here for two selected pump fluences, 27 and 72 μJ/cm$^{2}$. The dynamics of $\Delta \sigma_{\rm 2D}$ for both the CH$_{3}$NH$_{3}$PbI$_{3}$ neat film and the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD film exhibit a sharp rise immediately after the excitation within the instrumental time resolution ($\sim$120 fs). Under very similar excitation conditions, the initial value of $\Delta \sigma_{\rm 2D}$ for CH$_{3}$NH$_{3}$PbI$_{3}$ is around three times larger than that of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD. As shown in Figs. 2(c) and 2(d), the maximum signal normalized to unity allows us to compare the evolution of the photoconductivity decay dynamics with exponential fits. It is important to note the difference between the two samples. For the CH$_{3}$NH$_{3}$PbI$_{3}$ neat film with pump fluence of 27 μJ/cm$^{2}$, the photoconductivity drops by ca. 48% within first $\tau_{1}=31.7\pm1.2$ ps and then 28% within $\tau_{2}=263.8\pm11$ ps, in contrast to a single-exponential decay of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD film: 49% within $\tau_{1}=963\pm84$ ps. This observation shows that the decay pathways of the photo-induced mobile charges in the neat CH$_{3}$NH$_{3}$PbI$_{3}$ and that in the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD must be different.
cpl-35-2-028401-fig3.png
Fig. 3. (a) The normalized dynamical traces of neat CH$_{3}$NH$_{3}$PbI$_{3}$ sample at various pump fluences (9–81 μJ/cm$^{2}$). (b) Reciprocal of the dynamical traces shown in (a).
We assume that the carrier mobilities do not change significantly with the pump-probe time delay, hence the decay in THz conductivity mainly reflects the temporal evolution of charge carrier density. The observed decay becomes increasingly faster as the excitation fluence increases from 9 μJ/cm$^{2}$ ($N_{\rm ph}\sim1.57\times10^{17}$ photon/m$^{2}$) to 81 μJ/cm$^{2}$ ($N_{\rm ph}\sim13.9\times10^{17}$ photon/m$^{2}$), as shown in Fig. 3(a). The time constants for CH$_{3}$NH$_{3}$PbI$_{3}$ obtained from the bi-exponential fits for pump fluences are shown in the inset of Fig. 2(c). The first-$\tau_{1}$ and second-exponential $\tau_{2}$ decay processes become faster upon increasing the pump fluence. In our experiment, the second exponential time constant $\tau_{2}$ decreases from 264 ps for 9 μJ/cm$^{2}$ to 130 ps for 72 μJ/cm$^{2}$, which can be assigned to the second-order bimolecular electron–hole recombination.[32] It is also evidenced by the linearity of ($\Delta \sigma_{\rm 2D})^{-1}$ as a function of time over the entire range of excitation fluences, as shown in Fig. 3(b). Furthermore, the first fast exponential time constant $\tau_{1}$ decreases with the increasing pump fluence, from 32 ps for 9 μJ/cm$^{2}$ to 10 ps for 72 μJ/cm$^{2}$, which can be reasonably attributed to the trimolecular Auger recombination.[32,33] On the other hand, THz conductivity transient for the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD at low fluence (27 μJ/cm$^{2}$) is almost flat and fits well to a monoexponential ($\tau =963\pm80$ ps). In this case, the second-order recombination is almost eliminated. The recombination dynamics of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD is nearly pump fluence independent. Only at high pump fluence (72 μJ/cm$^{2}$), the data can be fitted better to a biexoponential decay with decay time constants $\tau_{1}=62\pm5$ ps and $\tau_{2}=426\pm16$ ps, which is dominantly assigned to the recombination between holes injected into Spiro-OMeTAD and mobile electrons in the perovskite. Knowing the conductivity is defined as $\sigma =eN\mu$, where $e$ is the elementary charge, $N$ is the carrier density, and $\mu=e\tau_{\rm s}/m^{*}$ is the mobility dependent on carrier momentum scattering time $\tau_{\rm s}$ and effective mass $m^{*}$. In our case, therefore the product of the quantum yield $\eta$ and the sum of electron and hole mobility $\mu_{\rm tot}=\mu_{\rm e}+\mu_{\rm h}$ is expressed as the effective mobility $\mu_{\rm eff}$ by $$\begin{align} \mu _{\rm eff} =\eta \times \mu _{\rm tot}=\frac{\Delta \sigma _{2{\rm D}}}{eN_{{\rm exc}} }.~~ \tag {2} \end{align} $$ Figure 4(a) shows $\Delta \sigma_{\rm 2D}$ scaled to the $e$ times excited electron density $N_{\rm exc}$ as a function of the pump-probe delay. In the case of 400 nm optical excitation at a fluence of 27 μJ/cm$^{2}$ corresponded to $N_{\rm ph}=4.7\times10^{17}$ and 4.8$\times$10$^{17}$ photon/m$^{2}$ for CH$_{3}$NH$_{3}$PbI$_{3}$ and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, respectively. Assuming that each absorbed photon is converted into an electron–hole pair, the initial carrier density $N_{\rm exc}$ is calculated as $N_{\rm exc}=2N_{\rm ph}$.
cpl-35-2-028401-fig4.png
Fig. 4. (a) The photoconductivity signal expressed as the product of the quantum yield and the carrier mobility (inset: normalized to unity). (b) Schematic diagram of energy levels and electron relaxation and transfer processes in a perovskite/HTM sample. Electrons left in the conduction band of MAPbI$_{3}$ after photo-excitation (i), and then recombine with both the dark holes in the valence band of CH$_{3}$NH$_{3}$PbI$_{3}$, including the radiative/photoluminescence (ii) and/or the non-radiative electron-hole recombination (iii) and the photo-generated holes injected into Spiro-OMeTAD (iv) and back charge transfer at the interface between perovskite/HTM (v).
We find that the effective mobility $\mu_{\rm eff}$ (i.e., the product of the quantum yield and total mobility) of the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD film shows a very different photo-response dynamics at the pump fluence of 27 μJ/cm$^{2}$, compared with the neat perovskite film. The peak value of $\mu_{\rm eff}$ for the neat CH$_{3}$NH$_{3}$PbI$_{3}$ is 18.5 cm$^{2}$/Vs, very similar to that reported by Wehrenfennig et al.[21] and Carlito et al.[33] For the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, the initial value of $\mu_{\rm eff}$ is about 3 times smaller ($\sim$6.5 cm$^{2}$/Vs) than the neat CH$_{3}$NH$_{3}$PbI$_{3}$ film. The reduction of the effective mobility can be well interpreted as charge transfer. In our case of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, where there is ultrafast injection of holes from the perovskite into Spiro-OMeTAD, as shown in the schematic of photo-physical processes in Fig. 4(b), then the total mobility will be $\mu_{\rm tot}=\mu_{\rm e, perovskite}+\mu_{\rm h, Spiro-OMeTAD}$. As previously reported, the hole mobility in Spiro-OMeTAD is $\sim$1$\times$10$^{-4}$ cm$^{2}$/Vs,[34] which is four orders smaller than that in CH$_{3}$NH$_{3}$PbI$_{3}$. Hence, the overall lower mobility $\mu_{\rm tot}$ obtained in the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD is dominated by the mobility of electrons only staying in perovskite. It should be noted that the reduction in both the THz photoconductivity and the effective mobility response happens within the instrument response function, as soon as the charge carriers are photogenerated. As shown in Fig. 4(b), the offset in the valence band of Spiro-OMeTAD with respect to CH$_{3}$NH$_{3}$PbI$_{3}$ is around 0.21 eV, which is enough to drive this ultrafast transfer.[35] This ultrafast injection of holes also suggests the perfect electronic coupling between the Spiro-OMeTAD and the perovskite via the oxygen-containing groups.[20] Recently, such an ultrafast hole transfer was also observed in perovskite/NiO (inorganic hole extractor) interface.[16]
cpl-35-2-028401-fig5.png
Fig. 5. Excitation fluence dependence of (a) $\Delta \sigma_{\rm 2D}$ and (b) $\mu_{\rm eff}$ taken at zero-delay time for neat CH$_{3}$NH$_{3}$PbI$_{3}$ (full circles) and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD (open circles), respectively.
Figure 5(a) shows the peak value of real-valued photoconductivity as a function of the excitation fluence for the two samples. We observe a very weak sublinear fluence dependence of the initial photoconductivity. The best fits by a power law of the form: $\Delta \sigma_{\rm 2D, peak}=AF^{B}$, where $F$ is the pump fluence, and $B=0.85\pm0.06$ and 0.98$\pm$0.08 for neat CH$_{3}$NH$_{3}$PbI$_{3}$ and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, respectively. Compared with the conjugated polymers-based organic solar cells, of which $B\approx0.5$, indicating that the instantaneous rise of the real conductivity at zero time can be attributed to the exciton–exciton annihilation upon photoexcitation.[36,37] In P3HT:PC$_{70}$BM, after exciton splitting, charge transfer states or polaron pairs are formed at the donor–acceptor interface on the sub-ps time scale, in which the charges still attract each other coulombically.[36] In contrast, unbound electrons and holes play a principal role in the excited-state dynamics of OMHP under photoexcitation. The efficiency of dissociation of excitons into free electrons and holes depends on the exciton binding energy ($E_{\rm b}$). The value of $E_{\rm b}$ of CH$_{3}$NH$_{3}$PbI$_{3}$ is rather small (estimated $ < $10 meV), and is of the order of thermal energy at room temperature, indicating that a significant fraction of exciton will dissociate into free electron and holes right after photoexcitation. Our experimental results indicate that the rise of THz conductivity mainly comes from the free charges rather than exciton.[38-42] Figure 5(b) shows the dependence of effective mobility on the incident pump fluence. We see an initial (pump fluence $ < $27 μJ/cm$^{2}$) increase in the effective mobility $\mu_{\rm eff}$, up to 18.5 cm$^{2}$/Vs and 6.5 cm$^{2}$/Vs for the neat CH$_{3}$NH$_{3}$PbI$_{3}$ and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, respectively. As the pump fluence increases further above 27 μJ/cm$^{2}$, $\mu_{\rm eff}$ then gradually flattens off for the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, while it decreases on increasing pump fluence for the neat CH$_{3}$NH$_{3}$PbI$_{3}$, which can also be assigned to the second order recombination process.[32,33] In addition, another important observation is shown by the normalized absolute values for the two samples in Figs. 2(c) and 2(d) and in the inset of Fig. 4(a), the rise time of the THz photoconductivity and effective mobility of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD are very similar to those of the neat CH$_{3}$NH$_{3}$PbI$_{3}$ film, which suggests that the mechanism of photo-carriers generation is not altered by the presence of Spiro-OMeTAD. In contrast, $\mu_{\rm eff}$ of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD decays much more slowly than that for neat CH$_{3}$NH$_{3}$PbI$_{3}$ within the first ns, indicating a different mechanism of recombination. There are several processes that could contribute to the decay path of photoexcited carriers, as shown in Fig. 4(b). For the neat CH$_{3}$NH$_{3}$PbI$_{3}$, the only possible pathway of photo-energy is electron–hole recombination, either through radiative or non-radiative process (processes ii and iii). Once the Spiro-OMeTAD is introduced into the CH$_{3}$NH$_{3}$PbI$_{3}$, hole injection (process iv) occurs as a primary charge separation step. The slow decay of $\mu_{\rm eff}$ is assigned to the back charge transfer at the CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD interface (process v).[17] In summary, we have studied the transient THz conductivities of CH$_{3}$NH$_{3}$PbI$_{3}$ and CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD using time-resolved THz spectroscopy. Free charges are directly photo-generated within sub-picosecond for the two samples. Compared with CH$_{3}$NH$_{3}$PbI$_{3}$, the effective mobility is significantly suppressed in CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD, which demonstrates that the efficient holes transfer from perovskite into Spiro-OMeTAD on the sub-picosecond time scale, driven by the energy difference between the two materials. Rapid collection of holes by Spiro-OMeTAD helps to avoid the second-order recombination in the neat CH$_{3}$NH$_{3}$PbI$_{3}$ film. The slow recombination channel of CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD is dominated by the holes injected into Spiro-OMeTAD and the mobile electrons in the perovskite, named as back charge transfer at CH$_{3}$NH$_{3}$PbI$_{3}$/Spiro-OMeTAD interface. Our findings open the opportunity to design perovskite/HTM interfaces in devices. References Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide PerovskitesThe emergence of perovskite solar cellsEfficient planar heterojunction perovskite solar cells by vapour depositionHigh-efficiency solution-processed perovskite solar cells with millimeter-scale grainsIntriguing Optoelectronic Properties of Metal Halide PerovskitesElectro-optics of perovskite solar cellsElectron-Hole Diffusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite AbsorberLong-Range Balanced Electron- and Hole-Transport Lengths in Organic-Inorganic CH3NH3PbI3Improved Understanding of the Electronic and Energetic Landscapes of Perovskite Solar Cells: High Local Charge Carrier Mobility, Reduced Recombination, and Extremely Shallow TrapsPhotocarrier Recombination Dynamics in Perovskite CH 3 NH 3 PbI 3 for Solar Cell ApplicationsHybrid organic—inorganic perovskites: low-cost semiconductors with intriguing charge-transport propertiesUnusual defect physics in CH 3 NH 3 PbI 3 perovskite solar cell absorberLow trap-state density and long carrier diffusion in organolead trihalide perovskite single crystalsLow surface recombination velocity in solution-grown CH3NH3PbBr3 perovskite single crystalHigh Charge Carrier Mobilities and Lifetimes in Organolead Trihalide PerovskitesUltrafast Dynamics of Hole Injection and Recombination in Organometal Halide Perovskite Using Nickel Oxide as p-Type Contact ElectrodeUnravelling the mechanism of photoinduced charge transfer processes in lead iodide perovskite solar cellsComparison of Recombination Dynamics in CH 3 NH 3 PbBr 3 and CH 3 NH 3 PbI 3 Perovskite Films: Influence of Exciton Binding EnergyBand filling with free charge carriers in organometal halide perovskitesUltrafast charge carrier dynamics in CH 3 NH 3 PbI 3 : evidence for hot hole injection into spiro-OMeTADHomogeneous Emission Line Broadening in the Organo Lead Halide Perovskite CH 3 NH 3 PbI 3– x Cl x High Photoluminescence Efficiency and Optically Pumped Lasing in Solution-Processed Mixed Halide Perovskite SemiconductorsPhotoexcitation dynamics in solution-processed formamidinium lead iodide perovskite thin films for solar cell applicationsFirst-principles study on the electronic and optical properties of cubic ABX3 halide perovskitesCarrier dynamics in semiconductors studied with time-resolved terahertz spectroscopyTemperature-Dependent Charge-Carrier Dynamics in CH 3 NH 3 PbI 3 Perovskite Thin FilmsElucidating the role of disorder and free-carrier recombination kinetics in CH3NH3PbI3 perovskite filmsCharge-carrier dynamics in vapour-deposited films of the organolead halide perovskite CH 3 NH 3 PbI 3−x Cl xIntrinsic femtosecond charge generation dynamics in single crystal CH 3 NH 3 PbI 3Organometal Halide Perovskite Solar Cell Materials Rationalized: Ultrafast Charge Generation, High and Microsecond-Long Balanced Mobilities, and Slow RecombinationUltrafast carrier relaxation in radiation-damaged silicon on sapphire studied by optical-pump–terahertz-probe experimentsTerahertz Conductivity within Colloidal CsPbBr 3 Perovskite Nanocrystals: Remarkably High Carrier Mobilities and Large Diffusion LengthsMechanism of Charge Transfer and Recombination Dynamics in Organo Metal Halide Perovskites and Organic Electrodes, PCBM, and Spiro-OMeTAD: Role of Dark CarriersElectron and Hole Transport through Mesoporous TiO2 Infiltrated with Spiro-MeOTADPerovskite as Light Harvester: A Game Changer in PhotovoltaicsUltrafast Terahertz Photoconductivity of Photovoltaic Polymer–Fullerene Blends: A Comparative Study Correlated with Photovoltaic Device PerformanceSignature of exciton annihilation in the photoconductance of regioregular poly(3-hexylthiophene)Exciton versus Free Carrier Photogeneration in Organometal Trihalide Perovskites Probed by Broadband Ultrafast Polarization Memory DynamicsExcitons versus free charges in organo-lead tri-halide perovskitesImpact of microstructure on local carrier lifetime in perovskite solar cellsOptical-Electrical Characteristics and Carrier Dynamics of Semi-Insulation GaAs by Terahertz Spectroscopic Technique
[1] Lee M M, Teuscher J, Miyasaka T et al 2012 Science 338 643
[2] Green M A, Ho-Baillie A and Snaith H J 2014 Nat. Photon. 8 506
[3] Liu M, Johnston M B and Snaith H J 2013 Nature 501 395
[4] Nie W, Tsai H, Asadpour R et al 2015 Science 347 522
[5] Manser J S, Christians J A and Kamat P V 2016 Chem. Rev. 116 21
[6] Lin Q, Armin A, Nagiri R C R et al 2015 Nat. Photon. 9 106
[7] Stranks S D, Eperon G E, Grancini G et al 2013 Science 342 341
[8] Xing G, Mathews N, Sun S et al 2013 Science 342 344
[9] Oga H, Saeki A, Ogomi Y et al 2014 J. Am. Chem. Soc. 136 13818
[10] Yamada Y, Nakamura T, Endo M et al 2014 J. Am. Chem. Soc. 136 11610
[11] Brenner T M, Egger D A and Kronik L 2016 Int. Mater. Rev. 1 15007
[12] Yin W J, Shi T and Yan Y 2014 Appl. Phys. Lett. 104 063903
[13] Shi D, Adinolfi V, Comin R et al 2015 Science 347 519
[14] Yang Y, Yan Y, Yang M et al 2015 Nat. Commun. 6 7961
[15] Wehrenfennig C, Eperon G E, Johnston M B et al 2014 Adv. Mater. 26 1584
[16] Corani A, Li M H, Shen P S et al 2016 J. Phys. Chem. Lett. 7 1096
[17] Marchioro A, Teuscher J, Friedrich D et al 2014 Nat. Photon. 8 250
[18] Yang Y, Yang M, Li Z et al 2015 J. Phys. Chem. Lett. 6 4688
[19] Manser J S and Kamat P V 2014 Nat. Photon. 8 737
[20] Brauer J C, Lee Y, Nazeeruddin M K et al 2016 J. Mater. Chem. C 4 5922
[21] Wehrenfennig C, Liu M, Snaith H J et al 2014 J. Phys. Chem. Lett. 5 1300
[22] Deschler F, Price M, Pathak S et al 2014 J. Phys. Chem. Lett. 5 1421
[23] Fang H, Wang F, Adjokatse S et al 2016 Light: Sci. Appl. 5 e16056
[24] Lang L, Yang J H, Liu H R et al 2014 Phys. Lett. A 378 290
[25] Ulbricht R, Hendry E, Shan J et al 2011 Rev. Mod. Phys. 83 543
[26] Milot R L, Eperon G E, Snaith H J et al 2015 Adv. Funct. Mater. 25 6218
[27] Chan L O V, Salim T, Kadro J et al 2015 Nat. Commun. 6 7903
[28] Wehrenfennig C, Liu M, Snaith H J et al 2014 Energy Environ. Sci. 7 2269
[29] Valverde-Chávez D A, Jr. Ponseca C S, Stoumpos C C et al 2015 Energy Environ. Sci. 8 3700
[30] Jr. Ponseca C S, Savenije T J, Abdellah M et al 2014 J. Am. Chem. Soc. 136 5189
[31] Lui K P H and Hegmann F A 2001 Appl. Phys. Lett. 78 3478
[32] Yettapu G R, Talukdar D, Sarkar S et al 2016 Nano Lett. 16 4838
[33] Jr. Ponseca C S, Hutter E M, Piatkowski P et al 2015 J. Am. Chem. Soc. 137 16043
[34] Snaith H J and Grätzel M 2007 Adv. Mater. 19 3643
[35] Samrana K, Mohammad K N, Michael G et al 2014 Angew. Chem. Int. Ed. 53 2812
[36] Jin Z, Gehrig D, Dyer-Smith C et al 2014 J. Phys. Chem. Lett. 5 3662
[37] Dicker G, Haas M P and Siebbeles L D A 2005 Phys. Rev. B 71 155204
[38] Sheng C, Zhang C, Zhai Y et al 2015 Phys. Rev. Lett. 114 116601
[39] D'Innocenzo V, Grancini G, Alcocer M J et al 2014 Nat. Commun. 5 3586
[40] de Quilettes D W, Vorpahl S M, Stranks S D et al 2015 Science 348 683
[41]Han X W, Hou L, Yang L et al 2016 Chin. Phys. Lett. 33 120701
[42]Fan Z F, Tan Z Y, Wan W J et al 2017 Acta Phys. Sin. 66 087801 (in Chinese)