Coulomb Screening and Coherent Phonon in MethylammoniumLead Iodide PerovskitesHe Wang,†,‡,§ Leonas Valkunas,∥,⊥ Thu Cao,†,‡,§ Luisa Whittaker-Brooks,# and Graham R. Fleming*,†,‡,§

†Department of Chemistry, University of California, Berkeley, California 94720, United States‡Molecular Biophysics and Integrated Bioimaging Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720,United States§Kavli Energy NanoSciences Institute at the UC California, Berkeley, and the Lawrence Berkeley National Laboratory, Berkeley,California 94720, United States∥Theoretical Physics Department, Vilnius University, Vilnius 10222, Lithuania⊥Molecular Compound Physics Department, Center for Physical Sciences and Technology, Vilnius 10222, Lithuania#Department of Chemistry, University of Utah, Salt Lake City, Utah 84112, United States

*S Supporting Information

ABSTRACT: Methylammonium lead iodide (CH3NH3PbI3) hybrid perovskite in thetetragonal and orthorhombic phases have different exciton binding energies and demonstratedifferent excitation kinetics. Here, we explore the role that crystal structure plays in thekinetics via fluence dependent transient absorption spectroscopy. We observe strongersaturation of the free carrier concentration under high pump energy density in theorthorhombic phase relative to the tetragonal phase. We attribute this phenomenon to smalldielectric constant, large exciton binding energy, and weak Coulomb screening, which resultsin difficult exciton dissociation under high light intensity in the orthorhombic phase. Athigher excitation intensities, we observe a coherent phonon with an oscillation frequency of23.4 cm−1 at 77 K, whose amplitude tracks the increase of the first-order lifetime.

The methylammonium lead iodide (CH3NH3PbI3) hybridperovskite has attracted tremendous attention in the

photovoltaic field. CH3NH3PbI3, when incorporated into solarcells, has shown an efficiency of around 20%.1 The highefficiency and interest in hybrid perovskites stem from theirunique mechanical, optical, and electronic properties. Theseinclude small and tunable band gaps,2 broad light absorptionspectrum,3 crystal formation,4 exciton binding energy com-parable to or smaller than thermal energy at room temper-ature,5−7 high ambipolar charge transport mobility,8−11 limitedcharge recombination,12−17 easy processability, and potentiallow-cost fabrication.18 CH3NH3PbI3 hybrid perovskite under-goes various structural phase transitions depending on thetilting and rotation of the PbI6 polyhedra as well as theorientation of methylammonium cation in the lattice upon theintroduction of an external perturbation such as temper-ature.19,20 The disordered methylammonium cation in thetetragonal phase and the ordered cation in an antiferroelectricway in the orthorhombic phase lead to a variation of dielectricconstant,19 which results in different exciton binding energies21

in these two phases. In our previous work, we utilizedtemperature dependent transient absorption spectroscopy tofind that the crystal structure of a CH3NH3PbI3 thin film affectsexciton and free carrier dynamics.22 The recovery of the groundstate bleach in the orthorhombic phase is faster than that of thetetragonal phase,22 due to a larger fraction of excitons in the

orthorhombic phase. In addition to exciton and free carrierdynamics, transient absorption spectroscopy and quantitativeanalysis have revealed other aspects of the physics of theexcitations in CH3NH3PbI3.

13,15,23−26 The observed saturationof free carrier concentration under high-intensity excitation andthe increase of the band edge transition energy with free carrierconcentration suggests the existence of a band filling effect.15

The emergence of a sub-band gap signal has been attributed toband gap renormalization and hot carrier distribution.23,24 Theslow cooling of hot carriers in CH3NH3PbI3 has also beenobserved via transient absorption spectroscopy.13,23−25

In this work, we study the role that crystal structure plays inthe kinetics via fluence dependent transient absorptionspectroscopy. By probing at the center of ground state bleachand fitting the experimental data, we find stronger saturation offree carrier concentration at high-intensity excitation in theorthorhombic phase. At 77 K at higher fluencies, we observe anoscillation with a frequency of 23.4 cm−1 due to a coherentphonon, which may contribute to the increase of the first-orderdecay lifetime.Figure 1a displays transient absorption kinetic traces (a

semilog plot of ΔT/T as a function of delay time) at the band-

Received: June 28, 2016Accepted: August 2, 2016Published: August 2, 2016

Letter

pubs.acs.org/JPCL

© 2016 American Chemical Society 3284 DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

edge transition (760 nm) for CH3NH3PbI3 in the tetragonalphase at room temperature under various pump intensitiesfrom 1.8 μJ/cm2 to 76.6 μJ/cm2 with an excitation wavelengthof 650 nm. The decay of the signal originates from the recoveryof the ground state bleach as well as the stimulated emissiondue to radiative decay of excitons and from electron−holerecombination. In order to understand the kinetic mechanism,the traces are modeled by a generic rate equation, shown in eq1.15,27 This equation describes the multiple-order processes ofthe time evolution of exciton and free carrier density.

= − − −nt

An Bn Cndd

2 3(1)

where n is the density of free carriers or excitons, and t is thepump probe delay time. A, B, and C are the coefficients for thefirst-, second-, and third-order terms, respectively. The first-order term may originate from trap state assisted recombinationor from exciton relaxation. The second-order term mayoriginate from free carrier recombination and from exciton−exciton annihilation. The third-order term may originate from athree body effect, Auger recombination (the energy releasedfrom electron−hole recombination exciting a third carrier to ahigher energy state). As per eq 1, the form of the dynamicsshould reflect the type of carrier recombination or excitonrelaxation. The solution for the first-, second-, and third-orderkinetics are shown in eqs 2.1, 2.2, and 2.3.

= −n t n At( ) (0) exp( ) (2.1)

=+

n tn

Bn t( )

(0)1 (0) (2.2)

=+

n tn

Cn t( )

(0)

1 2 (0)2(2.3)

where n(0) is the maximum carrier concentration intensity. Alinear relationship of ln(n(t)) as a function of delay time tsuggests a first-order kinetic mechanism for the decays of theexciton or free carrier. A linear relationship of n(0)/n(t) as afunction of delay time t suggests a second-order kinetic

mechanism, while a linear relationship of [n(0)/n(t)]2 as afunction of delay time t suggests a third-order kineticmechanism.We first used the above description to understand the

dynamics of CH3NH3PbI3 in the tetragonal phase (roomtemperature) at different light intensities. The nonlinearrelationship at high light intensity of Figure 1a suggests thekinetics is not solely governed by a first-order process. We thenplot n(0)/n(t) (inverse of normalized ΔT/T) versus delay time,as shown in Figure 1b. The linear relationship at most lightintensities (shown by the dash-dot lines) indicates the second-order exciton or carrier dynamics. Given that the excitonbinding energy for CH3NH3PbI3 at room temperature iscomparable to or much smaller than the thermal energy −kT,5,6excitons can easily dissociate into free electrons and holes. Thesecond-order term therefore originates from electron−holerecombination rather than exciton−exciton annihilation.Finally, we plot [n(0)/n(t)]2 (inverse of the square ofnormalized ΔT/T) versus delay time, as shown in Figure 1c.We notice that, although the nonlinear relationship stronglydepends on the specific sample, the Auger recombination isweak even at high intensity excitation. Consistent with previouswork, we observe that electron−hole recombination is thedominant process in the tetragonal phase at room temperature.We fit the experimental data at all pump intensities accordingto the sum of eqs 2.1 and 2.2, the fit shown in the solid lines inFigure 1a and fitting parameters listed in the SupportingInformation. For example, at the highest pump fluence (76.6μJ/cm2), we extracted a lifetime 1/A from the first order fit tobe 55 ps with a weight percentage of 18% and the product ofthe second-order recombination rate constant and initial freecarrier concentration intensity Bn(0) to be 0.0024 ps−1 with aweight percentage of 82%. We used the analytical results for thephysical insight they provide. An extended study of theexcitation dynamics based on a more sophisticated theoreticalmodel will be presented in a future publication.We then utilized a similar method to analyze the transient

absorption spectra of CH3NH3PbI3 thin films in theorthorhombic phase. Figure 2a displays the semilog plot ofΔT/T versus delay time for perovskite in the orthorhombicphase at 77 K at the band edge transition (740 nm) under avariety of pump energy densities from 1.8 μJ/cm2 to 76.6 μJ/cm2. Figure 2b displays the magnification of Figure 2a from−0.6 to +6 ps. The wavelength of the band-edge transition (740nm) in the orthorhombic phase is smaller than that in thetetragonal phase (760 nm), due to the smaller band gap in thelatter phase.2,20,22,28−30 Both near-infrared and visible probes intransient absorption spectroscopy have shown that the excitonlifetime is of the order of ps and the free carrier lifetime is of theorder of ns.31 The exciton has been observed in transientabsorption spectra at room temperature15,23−26 and in theorthorhombic phase via ultraviolet−visible absorption spec-troscopy.5 The obvious existence of two regimes, a fast and aslow decay, likely can be attributed to exciton decay and freecarrier decay, respectively, given that the lifetime of excitons inCH3NH3PbI3 is much shorter than that of the free carriers.22,31

The linear dependence of ΔT/T versus t for the fast decay(∼1−6 ps at seven high light intensities above 14.9 μJ/cm2 and<1 ps at five low light intensities below 7.3 μJ/cm2) shown inFigure 2b suggests the relaxation of excitons. The trap densityhas been generally accepted to be low,8 and carrier lifetime ison the order of ns; we therefore expect that carrier trapping willnot make a significant contribution to the picosecond decay.

Figure 1. (a) ΔT/T plotted on a semilog scale. (b) Inverse ofnormalized ΔT/T. (c) Inverse of square of normalized ΔT/T as afunction of delay time for perovskite in the tetragonal phase at roomtemperature at various pump energy densities. The arrow indicatesincreasing of pump energy density. The circles and solid linesrepresent the experimental data and fit, respectively. The dash-dottedlines in panels b and c are the linear fits. The pump and probewavelengths are 650 and 760 nm, respectively.

The Journal of Physical Chemistry Letters Letter

DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

3285

The slow decay does not follow the linear relationship well.The plot of n(0)/n(t) (inverse of normalized ΔT/T) versusdelay time shown by the dash-dotted lines in Figure 2c followsa linear relationship, suggesting electron−hole recombination.We did not observe a deviation of linear relationship in such aplot, suggesting that Auger recombination is not prevalent inthe orthorhombic phase. By fitting the experimental data at thehighest pump energy density (76.6 μJ/cm2) according to thesum of eqs 2.1 and 2.2, we extracted a lifetime 1/A from thefirst-order component to be 3.6 ps with a weight percentage of79% and the product of the second-order recombination rateconstant and initial free carrier concentration intensity Bn(0) tobe 0.0023 ps−1 with a weight percentage of 21%. The details ofthe fitting parameters are listed in the Supporting Information.In addition to the dynamics, we also studied the maximum

signal ΔT/T as a function of pump energy density, whichshows different trends in the tetragonal and orthorhombicphases. Figure 3a displays the maximum ΔT/T for thetetragonal phase as a function of pump energy density viafitting the experimental data shown in Figure 1a. The maximumΔT/T follows a linear relationship with pump energy density inthe tetragonal phase, suggesting a linear dependence of freecarrier concentration on pump energy density. Figure 3bdisplays the maximum ΔT/T as a function of the square root ofthe pump energy density in the orthorhombic phase. Incontrast to the tetragonal phase, the maximum ΔT/T follows asquare root relationship with pump energy density in theorthorhombic phase. Given that the signal ΔT/T can originatefrom both excitons and free carriers for perovskite in theorthorhombic phase, we decoupled the two contributions byfitting the experimental data to the sum of eqs 2.1 and 2.2, thefit shown in Figure 2a. Figures 3c and 3d display the maximumΔT/T as a function of pump energy density extracted from thefast decay and slow decay, respectively. The fast decay (excitondecay) follows a linear relationship. In contrast, the slow decay(free carrier decay) follows a saturation trend with pumpenergy density when pump energy density values are aboveabout 10 μJ/cm2, suggesting that higher pump energy densitiesdo not generate larger concentrations of free carriers. We

modeled the maximum of ΔT/T for the slow decay as afunction of the pump energy density according to thephenomenological expression15 shown in eq 3.

=+

IaI

b I cmaxpump

pump (3)

where Ipump is the pump energy density in μJ/cm2, Imax is themaximum ΔT/T, and a, b, and c are the floating parameters.The fit gives a = 0.26, b = 61.4, and c = 1.31. The saturation offree carrier concentration can be affected by both band fillingand the ease of exciton dissociation at high light intensity. Theband filling effect has been observed in semiconductors32 and inCH3NH3PbI3 hybrid perovskites,15 in which the lowest energystate in the conduction band is filled under high-intensityexcitation, which drives the nominal band gap to higher energy.Band filling depends on the effective mass. When the effectivemass is small, it is easy to fill the band, resulting in a nominallarger band gap, and free carrier saturation at high-intensityexcitation. Previous studies have demonstrated the effectivemass in the orthorhombic phase is comparable to or larger thanthat in the tetragonal phase.33,34 Therefore, we suspect the bandfilling caused by the effective mass cannot be the dominantorigin of the saturation effect. In the tetragonal phase, highfluence excitation can create a large concentration of freecarriers due to the large dielectric constant and small excitonbinding energy. The quadratic relationship of maximumfluorescence amplitude under high excitation intensity hasbeen suggested to indicate the nongeminate recombination offree carriers at high light intensity at room temperature.35 Thelarge concentration of free carriers induces strong Coulombscreening, which further increases the dielectric constant anddecreases the exciton binding energy.24 However, in theorthorhombic phase, the dielectric constant is small andexciton binding energy is large. The neutral exciton ratherthan a positive or negative charge leads to weak Coulombscreening even at high light intensity. We suggest that this weakCoulomb screening at high light intensity is responsible for thesaturation of free carriers in the orthorhombic form, as this will

Figure 2. (a) ΔT/T plotted on a semilog scale. (b) Magnification ofpanel a from −0.6 to +6 ps. (c) Inverse of normalized ΔT/T as afunction of delay time for perovskite in the orthorhombic phase at 77K at various pump energy densities. The arrow indicates increasing ofpump energy density. The circles and solid lines represent theexperimental data and fit, respectively. The dash-dotted lines in panel care a linear fit. The pump and probe wavelengths are 650 and 740 nm,respectively.

Figure 3. (a) Maximum transient absorption signal ΔT/T as afunction of pump energy density extracted from the perovskite in thetetragonal phase at room temperature. (b) Maximum transientabsorption signal ΔT/T as a function of square root of pump energydensity in the orthorhombic phase at 77 K. Maximum ΔT/T as afunction of pump energy density extracted from the fit of (c) fastdecay and (d) slow decay for perovskite in the orthorhombic phase at77 K. Blue circles andred dash-dotted lines represent the data and fit,respectively.

The Journal of Physical Chemistry Letters Letter

DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

3286

make it more difficult for excitons to dissociate into freecarriers.Interestingly, we observe oscillatory behavior in the

orthorhombic phase at 77 K, especially at high pumpintensities, as shown in Figure 2b. Figure 4a displays a

representative decay trace probed at 740 nm for theorthorhombic phase at 77 K with the pump energy density of46.4 μJ/cm2. The oscillation is obvious. To improve theresolution of the oscillations, we used a lock-in amplifier andoptical chopper to increase the signal-to-noise ratio in all themeasurements reported here. We have verified that thisoscillation is not from a satellite pulse by cross-correlationand it is very reproducible. After fitting the experimental dataaccording to the sum of equations of 2.1 and 2.2, we extractedthe residual, shown in Figure 4b. Fourier transformation givesan oscillation frequency of around 23.4 cm−1, corresponding to0.70 THz. According to the terahertz spectroscopy, a peakaround 0.8 THz is associated with a phonon in theorthorhombic phase.36 This phonon has been attributed tothe vibration of Pb−I bonds. Furthermore, via first-principleslattice dynamics, the 1-THz phonon mode has been assigned tothe buckling of the Pb−I−Pb angles.37 Although coherentphonons have been observed in inorganic perovskitematerials,38 so far as we know, this is the first observation ofa coherent phonon in CH3NH3PbI3. The relaxation of thecoherent phonon is determined by elastic and inelasticscattering. The latter is typically the dominant process in adielectric material. A very interesting phenomenon is that whenthe oscillation of the coherent phonon is prominent at sevenhigh light intensities (above 14.9 μJ/cm2), the excitonrelaxation time is long, shown in Figure 2b. The fit underthese high light intensities gives a first-order lifetime of 2.8 ±0.5 ps. The fit at five low light intensities (below 7.3 μJ/cm2)gives a first-order lifetime of 0.38 ± 0.03 ps. Figure 4d displaysthe lifetime of first-order decay and maximum Fouriertransform power around 23.4 cm−1 at different pump energydensities. The first-order decay time and the Fourier transformpower are clearly correlated. Whether there is a causal

connection between the two is not clear, but below we offersome suggestions as to how coherent phonons could increasethe first order decay time. Coherent phonons can increase thefirst-order lifetime by increasing the exciton lifetime and/orslowing of the carrier cooling, which both happen on the orderof ps. For example, coherent phonons can directly increase theexciton lifetime before relaxation by affecting the dielectricconstant or refractive index which ultimately influences theexciton binding energy.38 The thermalization of free carriers hasbeen observed to be slow in CH3NH3PbI3 at room temperatureby frequency resolved transient absorption spectroscopy.23,24,39

Besides increasing the exciton lifetime, the carrier thermal-ization may be slowed by the coherent phonons if they arestrongly mixed with the excitons. In order to disentangle thecontributions of coherent phonons to exciton and free carrierdecay at low temperature, frequency resolved transientabsorption spectroscopy with high signal-to-noise is planned.Quantitative analysis and a theoretical model of the coherentphonon is needed and will be addressed separately. Phononsplay significant roles in energy transport, carrier mobility, andthermal properties. For example, the high mobility observed inCH3NH3PbI3 may originate from the weak scattering ofphonons and slow thermalization of hot carriers.23,24,39 Howthe coherent phonon affects energy transport, carrier mobility,and thermal properties is a topic that merits further study.In summary, we studied exciton and free carrier dynamics in

CH3NH3PbI3 perovskite thin films in the tetragonal andorthorhombic phases using fluence dependent transientabsorption spectroscopy. We observe a saturation of freecarrier at high-intensity excitation in the orthorhombic phase,which we attribute to the weak Coulomb screening. We alsoobserve a coherent phonon with an oscillation frequency of23.4 cm−1 at 77 K. The existence of a coherent phonon iscorrelated with an increase in the lifetime of the first-orderdecay. Studying the coupling between phonons and excitons orfree carriers should increase our understanding of optoelec-tronic and thermal properties of hybrid organic inorganicperovskites.

■ EXPERIMENTAL METHODSFabrication of CH3NH3PbI3 Polycrystalline Thin Film. Thesynthesis of methylammonium iodide (CH3NH3I) wasreported previously.22 The synthesized CH3NH3I and PbI2(Alfa Aesar) were dissolved in anhydrous dimethylformamide(Sigma-Aldrich) at 3:1 molar ratio with a total molarconcentration of 3.52 mol/L and the solution was stirredovernight. The CH3NH3PbI3 film was formed by spin coatingthe solution at 2000 rpm for 1 min onto a precleaned glasssubstrate, followed by thermal annealing at 135 °C for 45 minin air. The film thickness was 550 ± 20 nm, obtained byprofilometry. X-ray diffraction was utilized to check thesuccessful formation of a polycrystalline film, with an averagecrystal size ∼37 nm, derived from Scherrer Equation. The laserpump spot size diameter is ∼300 μm, much larger than thecrystal size. So we are averaging over quite a few grains andgrain boundaries.Fluence-Dependent Transient Absorption Spectroscopy. Tran-

sient absorption spectra were collected in the pump probegeometry based on an amplified Ti:sapphire laser with arepetition rate of 250 kHz. The pump beam, generated by acollinear optical parametric amplification of the 805 nm output,was centered at 650 nm with a full width at half-maximum of 32nm and of about 40 fs duration after compression by a pair of

Figure 4. (a) Representative transient absorption signal ΔT/T at 77 Kin the orthorhombic phase with a pump energy density of 46.4 μJ/cm2.The red line and blue line represent the experimental data and fit,respectively. (b) Residual (blue circle) of the transient absorptionsignal obtained by the subtraction of the fit from the experimentaldata. The red line is the interpolation of data every 50 fs from theexperimental data. (c) Fourier transformation of interpolated data. (d)Lifetime of the fast decay (left y axis, blue) and maximum Fouriertransform power (right y axis, green) due to the coherent phonon as afunction of pump energy density.

The Journal of Physical Chemistry Letters Letter

DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

3287

prisms. The pump beam was focused at the sample positionwith a diameter of 300 μm. The pump beam was chopped withan optical chopper and pump fluence was controlled by neutraldensity filters. Data were acquired at 12 different pump energydensities: 76.6, 58.1, 46.4, 36.9, 29.4, 18.8, 14.9, 7.3, 5.7, 4.6,3.6, and 1.8 μJ/cm2. A white light supercontiunuum, created byfocusing 805 nm output onto a 1 mm-thick sapphire plate, wasused as the probe beam. The polarization angle between pumpand probe was set at 55°. The probe beam was spectrallyresolved by a spectrometer and detected by a Si photodiodewith a lock-in amplifier in all the transient absorptionspectroscopy measurements. The cross correlation betweenpump and probe pulses was around 100 fs. The data wereacquired with pump−probe delay times from −600 to 3000 fswith a step size of 20 fs, from 3.1 to 10 ps with a step size of 50fs (at 77 K) or 100 fs (at room temperature), from 11 to 100 pswith a step size of 1 ps, and from 110 to 800 ps with a step sizeof 10 ps. The sample was placed in a vacuum chamber of anOxford Instruments liquid N2 cryostat, with a temperature of77 K for the orthorhombic phase and 295 K for the tetragonalphase. The sample was stored in vacuum. The sample wasexposed to air during spin coating and transport from samplepreparation site to the laser lab.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.jpclett.6b01425.

Details of fitting parameters (PDF)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: grfleming@lbl.gov.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by the National Science Foundation(NSF) under Awards CHE-1012168 and CHE-1362830. L.V.was supported by the Research Council of Lithuania (LMTGrant No. MTP-080/2015).

■ REFERENCES(1) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D.Solar Cell Efficiency Tables (Version 47). Prog. Photovoltaics 2016, 24,3−11.(2) Xing, G.; Mathews, N.; Lim, S. S.; Yantara, N.; Liu, X.; Sabba, D.;Gratzel, M.; Mhaisalkar, S.; Sum, T. C. Low-Temperature Solution-Processed Wavelength-Tunable Perovskites for Lasing. Nat. Mater.2014, 13, 476−480.(3) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith,H. J. Efficient Hybrid Solar Cells Based on Meso-SuperstructuredOrganometal Halide Perovskites. Science 2012, 338, 643−647.(4) Wang, Q.; Lyu, M.; Zhang, M.; Yun, J.-H.; Chen, H.; Wang, L.Transition from the Tetragonal to Cubic Phase of OrganohalidePerovskite: The Role of Chlorine in Crystal Formation ofCH3NH3PbI3 on TiO2 Substrates. J. Phys. Chem. Lett. 2015, 6,4379−4384.(5) D’Innocenzo, V.; Grancini, G.; Alcocer, M. J. P.; Kandada, A. R.S.; Stranks, S. D.; Lee, M. M.; Lanzani, G.; Snaith, H. J.; Petrozza, A.Excitons Versus Free Charges in Organo-Lead Tri-Halide Perovskites.Nat. Commun. 2014, 5, 3586.

(6) Lin, Q.; Armin, A.; Nagiri, R. C. R.; Burn, P. L.; Meredith, P.Electro-Optics of Perovskite Solar Cells. Nat. Photonics 2015, 9, 106−112.(7) Frost, J. M.; Butler, K. T.; Brivio, F.; Hendon, C. H.; vanSchilfgaarde, M.; Walsh, A. Atomistic Origins of High-Performance inHybrid Halide Perovskite Solar Cells. Nano Lett. 2014, 14, 2584−2590.(8) Oga, H.; Saeki, A.; Ogomi, Y.; Hayase, S.; Seki, S. ImprovedUnderstanding of the Electronic and Energetic Landscapes ofPerovskite Solar Cells: High Local Charge Carrier Mobility, ReducedRecombination, and Extremely Shallow Traps. J. Am. Chem. Soc. 2014,136, 13818−13825.(9) Wehrenfennig, C.; Eperon, G. E.; Johnston, M. B.; Snaith, H. J.;Herz, L. M. High Charge Carrier Mobilities and Lifetimes inOrganolead Trihalide Perovskites. Adv. Mater. 2014, 26, 1584−1589.(10) Wehrenfennig, C.; Liu, M.; Snaith, H. J.; Johnston, M. B.; Herz,L. M. Charge-Carrier Dynamics in Vapour-Deposited Films of theOrganolead Halide Perovskite CH3NH3PbI3‑xClx. Energy Environ. Sci.2014, 7, 2269−2275.(11) Ponseca, C. S.; Savenije, T. J.; Abdellah, M.; Zheng, K.; Yartsev,A.; Pascher, T.; Harlang, T.; Chabera, P.; Pullerits, T.; Stepanov, A.;et al. Organometal Halide Perovskite Solar Cell Materials Rationalized:Ultrafast Charge Generation, High and Microsecond-Long BalancedMobilities, and Slow Recombination. J. Am. Chem. Soc. 2014, 136,5189−5192.(12) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.;Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J.Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in anOrganometal Trihalide Perovskite Absorber. Science 2013, 342, 341−344.(13) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Gratzel,M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- andHole-Transport Lengths in Organic-Inorganic CH3NH3PbI3. Science2013, 342, 344−347.(14) Sum, T. C.; Mathews, N. Advancements in Perovskite SolarCells: PhotophysicsBehind the Photovoltaics. Energy Environ. Sci.2014, 7, 2518−2534.(15) Manser, J. S.; Kamat, P. V. Band Filling with Free ChargeCarriers in Organometal Halide Perovskites. Nat. Photonics 2014, 8,737−743.(16) Stamplecoskie, K. G.; Manser, J. S.; Kamat, P. V. Dual Nature ofthe Excited State in Organic-Inorganic Lead Halide Perovskites. EnergyEnviron. Sci. 2015, 8, 208−215.(17) Marchioro, A.; Teuscher, J.; Friedrich, D.; Kunst, M.; van deKrol, R.; Moehl, T.; Gratzel, M.; Moser, J.-E. Unravelling theMechanism of Photoinduced Charge Transfer Processes in LeadIodide Perovskite Solar Cells. Nat. Photonics 2014, 8, 250−255.(18) Green, M. A.; Ho-Baillie, A.; Snaith, H. J. The Emergence ofPerovskite Solar Cells. Nat. Photonics 2014, 8, 506−514.(19) Onoda-Yamamuro, N.; Matsuo, T.; Suga, H. Dielectric Study ofCH3NH3PbX3 (X = Cl, Br, I). J. Phys. Chem. Solids 1992, 53, 935−939.(20) Baikie, T.; Fang, Y.; Kadro, J. M.; Schreyer, M.; Wei, F.;Mhaisalkar, S. G.; Graetzel, M.; White, T. J. Synthesis and CrystalChemistry of the Hybrid Perovskite (CH3NH3)PbI3 for Solid-StateSensitised Solar Cell Applications. J. Mater. Chem. A 2013, 1, 5628−5641.(21) Yamada, Y.; Nakamura, T.; Endo, M.; Wakamiya, A.; Kanemitsu,Y. Photoelectronic Responses in Solution-Processed PerovskiteCH3NH3PbI3 Solar Cells Studied by Photoluminescence and Photo-absorption Spectroscopy. IEEE J. Photovolt. 2015, 5, 401−405.(22) Wang, H.; Whittaker-Brooks, L.; Fleming, G. R. Exciton andFree Charge Dynamics of Methylammonium Lead Iodide PerovskitesAre Different in the Tetragonal and Orthorhombic Phases. J. Phys.Chem. C 2015, 119, 19590−19595.(23) Yang, Y.; Ostrowski, D. P.; France, R. M.; Zhu, K.; van deLagemaat, J.; Luther, J. M.; Beard, M. C. Observation of a Hot-PhononBottleneck in Lead-Iodide Perovskites. Nat. Photonics 2016, 10, 53−59.

The Journal of Physical Chemistry Letters Letter

DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

3288

(24) Price, M. B.; Butkus, J.; Jellicoe, T. C.; Sadhanala, A.; Briane, A.;Halpert, J. E.; Broch, K.; Hodgkiss, J. M.; Friend, R. H.; Deschler, F.Hot-Carrier Cooling and Photoinduced Refractive Index Changes inOrganic-Inorganic Lead Halide Perovskites. Nat. Commun. 2015, 6,8420.(25) Trinh, M. T.; Wu, X.; Niesner, D.; Zhu, X.-Y. Many-BodyInteractions in Photo-Excited Lead Iodide Perovskite. J. Mater. Chem.A 2015, 3, 9285−9290.(26) Simpson, M. J.; Doughty, B.; Yang, B.; Xiao, K.; Ma, Y.-Z.Separation of Distinct Photoexcitation Species in FemtosecondTransient Absorption Microscopy. ACS Photonics 2016, 3, 434−442.(27) La-o-vorakiat, C.; Salim, T.; Kadro, J.; Khuc, M.-T.;Haselsberger, R.; Cheng, L.; Xia, H.; Gurzadyan, G. G.; Su, H.;Lam, Y. M.; et al. Elucidating the Role of Disorder and Free-CarrierRecombination Kinetics in CH3NH3PbI3 Perovskite Films. Nat.Commun. 2015, 6, 7903.(28) Geng, W.; Zhang, L.; Zhang, Y.-N.; Lau, W.-M.; Liu, L.-M. First-Principles Study of Lead Iodide Perovskite Tetragonal andOrthorhombic Phases for Photovoltaics. J. Phys. Chem. C 2014, 118,19565−19571.(29) Zhu, X.; Su, H.; Marcus, R. A.; Michel-Beyerle, M. E. Computedand Experimental Absorption Spectra of the Perovskite CH3NH3PbI3.J. Phys. Chem. Lett. 2014, 5, 3061−3065.(30) Wehrenfennig, C.; Liu, M.; Snaith, H. J.; Johnston, M. B.; Herz,L. M. Charge Carrier Recombination Channels in the Low-Temperature Phase of Organic-Inorganic Lead Halide PerovskiteThin Films. APL Mater. 2014, 2, 081513.(31) Sheng, C.; Zhang, C.; Zhai, Y.; Mielczarek, K.; Wang, W.; Ma,W.; Zakhidov, A.; Vardeny, Z. V. Exciton versus Free CarrierPhotogeneration in Organometal Trihalide Perovskites Probed byBroadband Ultrafast Polarization Memory Dynamics. Phys. Rev. Lett.2015, 114, 116601.(32) Bennett, B. R.; Soref, R. A.; Del Alamo, J. A. Carrier-InducedChange in Refractive Index of InP, GaAs and InGaAsP. IEEE J.Quantum Electron. 1990, 26, 113−122.(33) Feng, J.; Xiao, B. Crystal Structures, Optical Properties, andEffective Mass Tensors of CH3NH3PbX3 (X = I and Br) PhasesPredicted from HSE06. J. Phys. Chem. Lett. 2014, 5, 1278−1282.(34) Miyata, A.; Mitioglu, A.; Plochocka, P.; Portugall, O.; Wang, J.T.-W.; Stranks, S. D.; Snaith, H. J.; Nicholas, R. J. Direct Measurementof the Exciton Binding Energy and Effective Masses for ChargeCarriers in Organic-Inorganic Tri-Halide Perovskites. Nat. Phys. 2015,11, 582−587.(35) Zheng, K.; Zhu, Q.; Abdellah, M.; Messing, M. E.; Zhang, W.;Generalov, A.; Niu, Y.; Ribaud, L.; Canton, S. E.; Pullerits, T. ExcitonBinding Energy and the Nature of Emissive States in OrganometalHalide Perovskites. J. Phys. Chem. Lett. 2015, 6, 2969−2975.(36) La-o-vorakiat, C.; Xia, H.; Kadro, J.; Salim, T.; Zhao, D.; Ahmed,T.; Lam, Y. M.; Zhu, J.-X.; Marcus, R. A.; Michel-Beyerle, M.-E.; et al.Phonon Mode Transformation Across the Orthohombic-TetragonalPhase Transition in a Lead Iodide Perovskite CH3NH3PbI3: ATerahertz Time-Domain Spectroscopy Approach. J. Phys. Chem. Lett.2016, 7, 1−6.(37) Brivio, F.; Frost, J. M.; Skelton, J. M.; Jackson, A. J.; Weber, O.J.; Weller, M. T.; Goni, A. R.; Leguy, A. M. A.; Barnes, P. R. F.; Walsh,A. Lattice Dynamics and Vibrational Spectra of the Orthorhombic,Tetragonal, and Cubic Phases of Methylammonium Lead Iodide. Phys.Rev. B: Condens. Matter Mater. Phys. 2015, 92, 144308.(38) Liu, Y.; Frenkel, A.; Garrett, G. A.; Whitaker, J. F.; Fahy, S.;Uher, C.; Merlin, R. Impulsive Light Scattering by Coherent Phononsin LaAlO3: Disorder and Boundary Effects. Phys. Rev. Lett. 1995, 75,334−337.(39) Zhu, X.-Y.; Podzorov, V. Charge Carriers in Hybrid Organic-Inorganic Lead Halide Perovskites Might Be Protected as LargePolarons. J. Phys. Chem. Lett. 2015, 6, 4758−4761.

The Journal of Physical Chemistry Letters Letter

DOI: 10.1021/acs.jpclett.6b01425J. Phys. Chem. Lett. 2016, 7, 3284−3289

3289