Ultrafast transient absorption spectroscopy investigations of charge carrier dynamics of methyl ammonium lead bromide (CH 3 NH 3 PbBr 3 ) perovskite nanostructures Hamzeh Telfah 1 ; Abdelqader Jamhawi 1 ; Meghan B. Teunis 3 ; Rajesh Sardar 3 ; Jinjun Liu 1, 2 1 University of Louisville - Department of Chemistry, Louisville, KY; 2 University of Louisville Conn Center for renewable energy, Louisville, KY; 3 Indiana University Purdue University, Indianapolis, IN June 21 st 217
Outline Motivation: why we want to study excited-state dynamics of perovskites. Perovskite structure Sample synthesis. Transient absorption (TA) measurements Experimental Results Pump fluence dependence measurements Proposed kinetic model Global fitting Conclusions.
Motivation High demand for alternative energy sources of higher efficiency Perovskites are promising light absorbers for photovoltaic devices [1] as well as materials in light-emitting diodes [2], photodetectors [3], and lasers [4]. Ultimately, increasing efficiency relies on optimization of photo-induced processes in photovoltaic materials. Transient absorption (TA) spectroscopy is a powerful tool for the study of photo-induced processes in photovoltaic materials. 1. Nature Photon. 8, 133 138 (214). 2. Nature Nanotech. 9, 687 692 (214). 3. Nature Photon. 9, 679 686 (215). 4. Nature Mater. 14, 636 642 (215).
Photovoltaic Efficiency Chart Advantages of perovskites: Strong absorption across the solar spectrum, Low exciton binding energy (~16 mev), High charge-carrier mobility, Long diffusion length (>1 µm), Band gap tunability. Higher efficiencies http://www.nrel.gov/ncpv/images/efficiency_chart.jpg
Samples, Synthesis and Characterization Nano-Crystals (NCs) Nano-Wires (NWs) Nano-Plates (NPs) Chem. Mater. 216, 28, 543 554
Synthesis, NCs and NWs Samples and NPs Chem. Mater. 216, 28, 543 554
Samples Transmission Electron Microscope (TEM) TEM images of NCs, NWs and NPs. Dimensions: NCs: ~2.4 nm, NWs: 3 9 nm length and ~3.8 nm diameter, NPs: 25 4 nm edge length. Chem. Mater. 216, 28, 543 554
Intinsity Normalized Steady-state absorption & photoluminescence(pl) spectra 1. (A) NCs (B) NWs (C) NPs Absorption PL.5. 2.25 2.4 2.55 2.7 NCs 2.25 2.4 2.55 2.7 NWs Energy 2.25 2.4 2.55 2.7 NPs Abs. Peak PL Peak Φ FL radiative life time (ns) Chem. Mater. 216, 28, 543 554 (nm) (nm) (%) τ fast τ slow NCs 499 513 16 ± 2 9 ± 2 5 ± 5 NWs 522 529 59 ± 7 11 ± 1 89 ± 13 NPs 497 515 44 ± 3 1 ± 1 63 ± 8
Pump-probe TA spectroscopy experiment, concept Sample probe I : Unpumped Probe laser transmission pump probe I*: Pumped Probe laser transmission Δt I*( t, ) OD( t, ) -log I( )
Pump-probe TA spectroscopy experiment: possible signal contributions S n S 1 Ground-state bleach (GSB) Stimulated Emission (SE) Excited-state absorption (ESA) S
Pump-probe TA spectroscopy experiment: setup Multi-Channel Camera Beam dump Sample Doubling crystal λ = 388 nm Pulse Duration< 15 fs fs Ti:Sapphire Amplifier λ=775 nm E pulse.75 mj r.r.=1 khz Pulse Duration<15 fs Pump Δt Motorized Delay Stage Mechanical Chopper 5 Hz OPA λ = 45-65 nm Pulse Duration 3 fs Continuum (White light probe) White light Crystal
TA Spectra, results 3.2 ev (388 nm) Pump, 2 μj/cm 2 Pump Flounce E 1 (fast, positive) E 2 (slow, negative) E 3 (fast, negative) E 4 (slow, positive)
mod TA Spectra, Results Suggested Kinetics E τ th < τ IRF E 1 : photo-induced absorption due to bandgap renormalization, E 2 : unresolved state filling and stimulated emission (nongemenate recombination), E 3 : state filling due to hot carriers, E 4 : photo-induced intraband absorption. 3 E 4 2 τ c 1-1 -2 E pump E 1 E 3 E 2 E * g E g τ r -3-4 E 1 E 2 E 3 τ th < τ IRF τ c -5 1 1 1 1 Pump-Probe Delay time (ps) τ th : thermalization time τ IRF : instrument response time τ c : carrier cooling time τ r : carrier recombination half time E 4
Recombination pathways based on charge carrier density ~2 x 1-18 cm -3 Acc. Chem. Res. 216, 49, 294 32
TA Spectra, Results Pump Fluence Dependency, 3.2 ev (388 nm) E g BM in the CB E E g E g = E g + E g BM E g BM in the VB Carrier Density Burstein-Moss effect
Normalized TA at E 3 (Unresolved SE and GSB) Cancellation of Burstein-Moss effect and band gap renormalization.
Pump Fluence Dependency, 3.2 ev (388 nm) Recombination times [n(t)] = [n ] 1 + [n ] t τ r 7 t 1/2 (ps) 6 5 4 3 NCs NWs NPs Error bars = 5s 2 1 2 4 6 8 1 12 [n ] x 1-18 (cm -3 )
k r /1-21 (cm 3 ps -1 ) 388 nm Pump Flounce Dependency Recombination Rates (k r ) of NWs does not dependent on [n ], while those of NCs and NPs do. k r = 1 [n ] τ r 5 4 3 NCs NWs NPs error bar=5s k k n () r r [ ] r k r () Χ r 2 (1-21 cm 3 ps -1 ) (1-3 cm 6 ps -1 ) NCs 2.21 (1) 3.7 (31) 1 2 4 6 8 1 12 NWs 1.81 (4) [n ]/1 18 (cm -3 ) NPs 1.3 (6).84 (14)
TA Spectra, Results Global Fitting, results Decay of E 1 and E 3 3 (a) NCs t =.4 ps t = 1 ps OD(t, ħω) = A ( ) i A (2 ) ( 2) ( ) i i=1,3 A i (ħω i ) e t τ c + A i (ħω i ) e t 1 τ c + 1 + t i=2,4 τ r Growth, Decay of E 2 and E 4 i i 2 2 i Lorentzian line shape used to describe the TA spectra in the frequency domain Y IRF( t) s IRF IRF e 2 IRF 1 t 2 s Instrument response function IRF mod 2-3 -6 3-3 -6 t = 1 ps t = 4 ps 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.2 2.3 2.4 2.5 2.6 2.7 2.8 Prope Photon Energy (ev) Expt. Simulation
TA Spectra, Results Global Fitting, results (b) NWs (c) NPs 9 6 t =.4 ps t = 1 ps -9-6 -18 t =.4 ps t = 1 ps -12 mod -27 9 6 t = 1 ps t = 4 ps -9-6 -18 t = 1 ps t = 4 ps -27 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.2 2.3 2.4 2.5 2.6 2.7 2.8 Probe Photon Energy (ev) -12 Expt. Simulation 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.2 2.3 2.4 2.5 2.6 2.7 2.8 Prope Photon Energy (ev)
TA Spectra, Results Global Fitting results, time domain hν 1 γ 1 hν 2 γ 2 hν 3 γ 3 hν 4 γ 4 τ c τ 1/2,r (ev) (ev) (ev) (ev) (ev) (ev) (ev) (ev) (ps) (ps) NCs 2.36.13 2.46.13 2.53.13 2.62.19.28 139 NWs 2.35.13 2.37.12 2.38.13 2.47.19.23 264 NPs 2.35.16 2.46.16 2.52.16 2.62.18.34 318 3 (a) NCs 1 (b) NWs 5 (c) NPs mod -3-6 -1-2 -5-1 -9-12 Simulation 2.35 ev 2.44 ev 2.54 ev 2.7 ev 1 2 1 1 1-3 Simulation 2.3 ev 2.38 ev 2.42 ev 2.51 ev 1 2 1 1 1-15 -2 Simulation 2.35 ev 2.44 ev 2.54 ev 2.7 ev 1 2 1 1 1 Pump-Probe Delay Time (ps)
Conclusions TA spectra were collected for the three nanostructures: NCS, NWs and NPs. Sub-picosecond charge carrier cooling is observed. Decay of the ta signals is best described by a second-order reaction dynamics and is attributed to non-geminate recombination of charge carriers. The effective rate constants of the NCs and NPs are linearly proportional to the density of the charge carriers with different linearity coefficients, while that of the NWs remains constant. Differences between recombination rates of the nanostructures suggest strong influence of the quantum confinement on the recombination processes. Charge carrier thermalization and recombination processes of perovskite nanostructures strongly dependent on their size, shape and morphology which can be utilized for improvement of power conversion efficiency.
Acknowledgements Indiana university Purdue university Dr. Rajesh Sardar; Meghan Teunis Department of Energy Advisor Dr. Jinjun Liu and lab members at the University of Louisville. Conn center for renewable energy - University of Louisville