Pump-Power Dependence of Coherent Acoustic Phonon Frequencies

Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Communication

Pump-Power Dependence of Coherent Acoustic Phonon Frequencies in Colloidal CdSe/CdS Core/Shell Nanoplatelets Dong Shuo, Jie Lian, Mark H. Jhon, Yinthai Chan, and Zhi-Heng Loh Nano Lett., Just Accepted Manuscript • Publication Date (Web): 24 Apr 2017 Downloaded from http://pubs.acs.org on April 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Pump-Power Dependence of Coherent Acoustic Phonon Frequencies in Colloidal CdSe/CdS Core/Shell Nanoplatelets

Shuo Dong,1 Jie Lian, 2 Mark Hyunpong Jhon,3* Yinthai Chan,2,4* and Zhi-Heng Loh1,5* 1

Division of Chemistry and Biological Chemistry, and Division of Physics and Applied Physics,

School of Physical and Mathematical Sciences, Nanyang Technological University, 21 Nanyang Link, Singapore 637371, Singapore 2

Institute of Materials Research & Engineering, A*STAR, 2 Fusionopolis Way, Innovis #

08-03, Singapore 138634, Singapore 3

Institute of High Performance Computing, A*STAR, 1 Fusionopolis Way, #16-16 Connexis,

Singapore 138632, Singapore 4

Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore

117543, Singapore 5

Centre for Optical Fibre Technology, The Photonics Institute, Nanyang Technological

University, 50 Nanyang Avenue, Singapore 639798, Singapore

Corresponding Authors *E-mail:

[email protected]

(M.H.J.),

[email protected]

[email protected] (Z.-H.L.).

1

ACS Paragon Plus Environment

(Y.C.),

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

Abstract

Femtosecond optical pump-probe spectroscopy resolves hitherto unobserved coherent acoustic phonons in colloidal CdSe/CdS core/shell nanoplatelets (NPLs). With increasing pump fluence, the frequency of the in-plane acoustic mode increases from 5.2 to 10.7 cm–1, whereas the frequency of the out-of-plane mode remains at ~20 cm–1. Analysis of the oscillation phases suggests that the coherent acoustic phonon generation mechanism transitions from displacive excitation to sub-picosecond Auger hole trapping with increasing pump fluence. The measurements yield Huang-Rhys parameters of ~10–2 for both acoustic modes. The weak electron-phonon coupling strengths favor the application of NPLs in optoelectronics.

Keywords: vibrational wave packet, acoustic phonon, 2D semiconductor nanostructure, Auger hole trapping, femtosecond transient absorption spectroscopy,

TOC graphic

2


Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Acoustic phonons play a dominant role in heat transport and in determining the thermoelectric properties of low-dimensional semiconductor nanostructures.1 At low temperatures, acoustic phonons significantly impact the photophysics of semiconductor nanostructures2 through deformation potential3,4 and piezoelectric coupling5–7 mechanisms. Confined acoustic phonons provide the required angular momentum to render dark states bright.2,8 In addition, the properties of acoustic phonons are also intrinsically related to the mechanical properties of nanomaterials.9,10 The size-dependent phonon frequency provides a way to determine the dimensions9,11–13 and the elastic moduli14–16 of nanomaterials. Manipulation of acoustic phonon modes in spatially confined nanoscale materials can be effected by controlling their characteristic geometry, surface termination, core-shell structures, and chemical composition.5,17–20 These various approaches to tuning the properties of acoustic phonons have been applied to the engineering of piezoelectric devices.21,22 As a result of their low frequencies, the investigation of acoustic phonons by frequency-domain Raman spectroscopy can be challenging. An alternative approach is to employ time-domain optical pump-probe spectroscopy. Ultrashort laser pulses generate coherent acoustic phonons either by displacive excitation or by impulsive heating.23,24 Vibrational wave packet motion induced by a pump pulse leads to amplitude and/or spectral modulations of the probe pulse as a function of pump-probe time delay. Fourier transformation of the probe signal along the time axis yields the frequency of the coherent phonon. The specific implementation of pump-probe spectroscopy to investigate the behavior of phonons, known as coherent phonon 3


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

spectroscopy,23,25 has been applied to a multitude of nanostructures, such as carbon nanotubes,26-28

semiconductor

nanocrystals,29,30

and

two-dimensional

transition

metal

dichalcogenides.31 Low-frequency acoustic phonon modes have also been elucidated in semiconductor quantum wells32,33 and nanocrystals,12,34 as well as metallic nanoparticles.35,36 Compared to frequency-domain measurements, coherent phonon spectroscopy directly furnishes the electron-phonon coupling strengths37,38 and provides insight into the mechanism by which the coherent phonons are generated.5,39,40 Here, we employ optical pump-probe spectroscopy to resolve the coherent optical and acoustic phonons of CdSe/CdS core/shell semiconductor nanoplatelets (NPLs). NPLs have recently emerged as a new class of II-VI nanomaterials that have so far comprised of 0D and 1D nanocrystals.41,42 Analogous to semiconductor quantum wells, colloidal semiconductor NPLs exhibit one-dimensional carrier quantum confinement due to their nanometer-scale thicknesses.43 The ability to control the thicknesses of NPLs with atomic precision leads to samples with uniform thickness.44,45 The narrow absorption and photoluminescence spectra,46 along with their suppressed fluorescence intermittency,47 favor the application of NPLs as efficient light emitters and optical gain media.48,49 Previous time-resolved studies have focused on the carrier dynamics of NPLs,50-54 while little is known about the phonon dynamics and the exciton-phonon coupling strengths. Steady-state photoluminescence and Raman measurements have unraveled the in-plane LO phonon,55-58 in addition to the out-of-plane LO phonon58 and surface optical (SO) phonons.57 Moreover, the out-of-plane acoustic phonon has recently been observed via 4


Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

low-frequency Raman scattering.59 Complementary to these frequency-domain studies, we employ time-domain measurements to elucidate the coherent phonon dynamics of both optical and acoustic modes of CdSe/CdS core/shell NPLs. Beyond the previously observed phonon modes, our femtosecond optical pump-probe measurements reveal the existence of an in-plane coherent acoustic phonon, whose frequency exhibits a pronounced dependence on excitation fluence. Analysis of the oscillation phases provides insight into the coherent phonon generation mechanisms.

The

measurements

furnish

Huang-Rhys

factors

that

characterize

the

exciton-acoustic phonon coupling strengths. The preparation of zinc-blende CdSe/CdS core/shell NPLs follows the procedure reported in ref. 44. Transmission electron microscopy (TEM) yields lateral dimensions of (28.8  3.1)  (7.1  1.1) nm2 and a thickness of 2.5  0.3 nm (Figure 1a). The measured thickness is consistent with 4.5 monolayers of CdSe (core) sandwiched between two monolayers of CdS (shell), which yields a theoretical thickness of 2.4 nm based on the lattice constants of CdSe and CdS.60 Femtosecond optical pump-probe measurements employ a narrowband pump pulse centered at 2.03 eV (33 meV FWHM) to resonantly excite the band-edge heavy-hole (HH) to conduction band (CB) transition (Figure 1b). The fluence of the pump pulse varies from 0.02 – 0.6 mJ/cm2, corresponding to an initial exciton population

of 1.1 – 13.4. The probe pulse spans 1.65 –

2.25 eV and is compressed to a transform-limited duration of 6 fs FWHM. The overall instrument response is 81 fs FWHM.

5


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

Photoexcitation of the HH-CB transition of the CdSe/CdS core/shell NPL sample leads to a combination of ultrafast carrier dynamics and coherent phonon dynamics. The mixture of carrier and phonon dynamics is evident from the time-dependent spectral first moment 〈

〉

∆

,

∆

,

, where

and

define the limits of the

HH-CB transition over which the first moment is computed and ∆ absorbance at probe photon energy

,

and pump-probe time delay

is the differential

(Figure 2). Because the

carrier dynamics define the baseline on top of which signatures of coherent phonons are found, accurate knowledge of the carrier dynamics is essential to reliably extract information on coherent phonons from the data. This is especially critical for the low-frequency acoustic phonon modes observed in this work. Previous fluence dependent measurements performed on the same sample with identical 2.03-eV photoexcitation established that the observed sub-picosecond dynamics originates from Auger mediated hole trapping,61 which is characterized by an effective second-order rate constant of 3.5  1.0 cm2/s. In the 〈

〉 traces, Auger trapping manifests

itself as a blue shift due to concomitant Auger heating of the holes. In addition, a spectral red shift rising from hot electron cooling is apparent at low pump intensities. The residual time traces 〈

〉 obtained following the removal of the carrier

dynamics baseline encodes coherent phonon dynamics. The observable 〈

〉 is particularly

well-suited for investigating vibrational wave packet dynamics because wave packet motion results in modulations of the probe transition energy.62 The 〈

6


〉 time trace obtained for

Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

1.1 (Figure 3a) reveals the presence of multiple frequency components. Quantitative 〉 is performed by fitting it to a sum of damped cosine functions

analysis of 〈

〈

where,

,

,

and

〉

cos

exp

⁄

,

(1)

correspond to amplitude, frequency, initial phase, and damping time

of oscillation component i, respectively. The best fit requires four components, whose parameters are summarized in Table 1. Figure 3b shows the contributions of the individual components that are reconstructed from the fit parameters. The high-frequency mode at 209  1 cm–1 corresponds to the in-plane LO phonon, in good agreement with the bulk CdSe LO phonon frequency of ~210 cm–1,60 as well as the frequencies for the LO phonon in CdSe NPLs obtained previously from frequency-domain measurements.55–58 A second high-frequency mode at 191  1 cm–1 is assigned to the SO phonon mode, given its proximity to the previously reported SO phonon frequency of 188 cm–1 for CdSe NPLs.57 Finally, the oscillation frequency at 20.1  0.4 cm–1 is consistent with the out-of-plane acoustic phonon mode that was previously observed by low-frequency Raman scattering.59 The FFT power spectrum of the 〈

〉 time trace

computed over the 12-ps time interval is shown in Figure 3c, verifying the existence of two low-frequency and two high-frequency components. the SO phonon appears only in the FFT power spectrum computed over a 3-ps interval (see Figure 3c inset) due to its short dephasing

7


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

time (~0.6 ps); in this case, however, the FFT resolution is insufficient to distinguish the two acoustic phonon modes. Table 1. Parameters obtained from fitting 〈 (cm–1)

Component i

〉 for

1.1 to Eq. 1.

(meV)

(

rad)

(ps)

1

5.2

0.2

0.013

0.002

0.90

0.07

—

2

20.1

0.4

0.082

0.009

0.95

0.03

1.66

0.22

3

191

1

0.183

0.030

0.40

0.05

0.60

0.07

4

209

1

0.153

0.020

0.70

0.02

1.60

0.15

Note: Component 1 is undamped in our experiments, i.e., the damping time is much longer than the maximum time delay of 12 ps employed in our experiments.

While the use of 16 fitting parameters in Eq. (1) –

,

,

, and

for

1

4 –

might raise questions regarding the possible over-parameterization of the fitting model, we note that three of the four measured frequencies have previously been observed and that the standard errors of the fits are small. These factors, along with the good agreement between the extracted oscillation frequencies and the fast Fourier-Transform (FFT) power spectrum (Figure 3c), lend credence to reliability of the fit parameters. Moreover, analysis of the signal in the time domain presents several advantages over the FFT approach. First, since the pump-probe data is acquired over a scan length of 12 ps (limited by the back-reflection from a 1-mm-thick beam splitter used for our few-cycle probe pulses), the FFT resolution is only 2.8 cm–1. Such an FFT resolution 8


Page 9 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

impedes our ability to accurately define the frequency of the low-frequency acoustic phonon modes. Second, phonons that dephase rapidly, such as the SO phonon, will not appear in the FFT, hence giving the false impression of their absence (see inset of Figure 3c). Third, the time-domain approach directly furnishes the oscillation amplitudes and damping times, which provide useful information regarding the exciton-phonon coupling strengths and coherent phonon dephasing times, respectively. Beyond the two optical phonon modes, two low-frequency components observed at 5.2  0.2 and 20.1  0.4 cm–1 can be assigned to coherent acoustic phonons. The first-moment modulations of both acoustic modes exhibit initial phases  0.07)

and (0.95  0.03)

that closely approach

rad: (0.90

rad for the 5.2- and 20.1-cm–1 modes, respectively, indicative of

wave packet motion in the excited state triggered by displacive excitation (Figure 4a). In this process, vibrational wave packet motion along the electronically excited-state potential energy curve starts from one extremum of the lattice displacement coordinate, hence giving rise to an initial oscillation phase of either 0 or peak with time delay, implied by

~

rad. The initial blue shift of the differential absorption rad, rules out the encoding of the wave packet motion

in the stimulated emission signal, which should exhibit an initial red shift with time delay to give ~ 0 rad. Furthermore, because ground-state bleaching and excited-state absorption produce out-of-phase spectral shifts, i.e.,

(see Supporting Information), the retrieved

phase of the excited-state absorption is ~0 rad, implying that the excited-state absorption transition exhibits an initial red shift. This red shift in turn suggests that the excited-state 9


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

absorption accesses a final state (ES*) that is displaced further away from the ground state (GS) than the excited state (ES), as shown in Figure 4a. This trend is consistent with ES* having formally

1 excitons for an ES with

the experimental data, we refer to

excitons. In order to simplify the interpretation of

in the remainder of this text.

Coherent phonons are detected in optical spectroscopy due to exciton-phonon coupling: coherent phonons modulate the energy levels of the excitonic states, which in turn give rise to the observed oscillations of the probe transition energy 〈

〉.18,23,26,30,62 Coupling of optical

phonons to excitons involve the Fröhlich interaction,29,30 whereas acoustic phonons are coupled via the deformation and piezoelectric potentials.12,34 Theoretical studies on II-VI semiconductor nanocrystals10 and quantum wells63 suggest that the deformation potential interaction dominates exciton-acoustic phonon coupling. It is important to note that displacive excitation launches only coherent phonons that conserve both the lattice symmetry and the point symmetry of the NPL.23 Even though NPLs in solution are randomly oriented relative to the light source, only symmetric eigenmodes are excited, which allows the identification of the observed vibrations. Similar analysis has been performed to distinguish the radial breathing and extensional phonon modes of gold nanorods in solution.24 These constraints suggest that the 5.2- and 20.1-cm–1 vibrations can be ascribed to in-plane and out-of-plane acoustic breathing modes, respectively. The assignment of the 20.1-cm–1 oscillation to the out-of-plane acoustic phonon mode is consistent with recent results obtained from frequency-domain Raman measurements,59 as well

10


Page 11 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

as the theoretical fundamental frequency for a classical longitudinal vibration through the thickness of the NPL, given by

1 2 where

is the speed of light,

stiffness tensor, and

is the thickness of the NPL,

is the mass density. Employing

for zinc-blende CdSe,60 and

(2)

,

is an element of the elastic 66.7 GPa and

5664 kg m–3

2.5 nm for the NPL yields a vibrational frequency of 22.9 cm–1.

In this calculation, the NPL is assumed to be a monolithic CdSe structure due to the similar elastic constants and mass densities for CdSe and CdS. This estimate neglects the effect of adsorbed ligands on the surface, which are expected to be small due to the low molecular weight of the 5-amino-1-pentanol ligands. A conservative estimate of these effects, assuming a close packed layer of ligands modeled as a concentrated mass at the surface59 reduces the fundamental 20.2 cm–1.

frequency to

At higher pump fluences, the frequency of the in-plane mode ( fluence dependence, increasing from 5.2 – 10.7 cm–1 over the range of whereas that of the out-of-plane mode (

) exhibits a pronounced ~ 1 – 13 (Figure 5a),

) fluctuates in the range of ~19 – 25 cm–1 without

any discernible fluence dependence (Figure 5b). The initial phase of the in-plane component (

) undergoes a

component (

-phase jump at

≳ 4 (Figure 5c), whereas that of the out-of-plane

) deviates from ~0 rad at

≳ 4 (Figure 5d). Finally, the oscillation

11


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

amplitudes of both in-plane (

) and out-of-plane (

Page 12 of 40

) components decrease with

(Figures 5e and 5f). The increase of the in-plane acoustic phonon frequency

with

rules out the

electronic softening of phonon modes64,65 as its origin. Electronic softening occurs when an increasing fraction of bonding electrons that is photoexcited into anti-bonding states, hence weakening the lattice and reducing the phonon frequency. Instead, the observed increase in with

can be intuitively rationalized in terms of a carrier density distribution with a larger

number of nodes coupling more favorably to higher-frequency phonons. With increasing fluence, more excitons

are delocalized along the lateral dimension of the NPL,53 resulting in an

exciton density distribution with a higher spatial frequency. Note that a similar scaling of coherent LA phonon frequencies with the inverse-periodicity of carrier distributions has previously been observed in semiconductor quantum wells.32,66 On the other hand, the out-of-plane frequency

is relatively independent of

because the addition of excitons to

NPLs occurs preferentially along the lateral dimension than the thickness direction, where the energy penalty is higher as a result of quantum confinement.67 The abrupt

-phase jump in

appear for initial exciton numbers

and the pronounced deviation of

from ~0 rad

≳ 4. Both observations suggest that a mechanism other

than displacive excitation is responsible for the generation of coherent acoustic phonons at high values. We consider the possibility that the sub-picosecond Auger-mediated trapping of holes in NPLs,61 which becomes operative nominally at

2, could provide the impulse for

12


Page 13 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

driving coherent phonons in NPLs. Given the previously measured Auger trapping rate constant and our experimental conditions, the time constant for Auger trapping decrease from ~0.3 ps to ~0.1 ps with increasing

is expected to

. The sub-picosecond trapping times, being

shorter than the acoustic phonon oscillation periods

2 ⁄

, satisfy the condition for the

impulsive excitation of coherent acoustic phonons by Auger-mediated trapping; in the present work,

fluctuates in the range of 1.4 – 1.7 ps and

increasing

decreases from 6.4 ps to 3.0 ps with

. Auger hole trapping transfers the excited-state (ES) population and its associated

vibrational coherence to the hole-trapped state (TS) on a timescale of

. Immediately following

Auger trapping, the first-moment of the trap-state absorption oscillates with an initial phase ( small

in, out). In the case of the in-plane acoustic phonon, the vanishingly

term compared to

phonon, however,

implies

(Figure 5c). For the out-of-plane

is smaller than, but not negligible compared to

the finite Auger trapping time at

≳ 4, and hence, the time needed for the out-of-plane

vibrational wave packet to appear on TS, recovers obtained at (

. Correction for

4 (Figure 5d). Note that the

~ 0 rad, similar to the

-phase jump in

values

upon Auger trapping

≳ 4) points to initial wave packet motion along opposite directions on the ES and TS

potential energy curves for the in-plane acoustic phonon mode (Figure 4b; see Supporting Information). On the other hand, the similar out-of-plane phases measured before ( 4) and after Auger trapping (

≳ 4) suggests that the initial motions of the

for

13

for


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 40

out-of-plane wave packet on both ES and TS occur in the same direction (Figure 4c; see Supporting Information). The manner in which Auger-mediated hole trapping initiates coherent acoustic phonons in NPLs is analogous to the previously reported mechanism of coherent LA phonon generation in CdSe nanocrystals by surface trapping.7 The exciton wave function, initially delocalized along the lateral plane of the NPL, becomes localized upon trapping. The emission of transient electric fields that accompanies hole trapping sets off coherent acoustic waves in the NPLs via piezoelectric coupling to the CdSe lattice. The creation of in-plane coherent acoustic phonons by this mechanism is particularly favorable because the lateral plane of the NPLs contains the [110] direction, along which the piezoelectric coupling coefficient for the zinc-blende lattice is the largest.68 Similar launching of the coherent LO phonon in CdSe nanocrystals by ultrafast charge transfer to an electron acceptor has also been observed.5,40 In these systems, coupling of the acoustic and optical phonons to the charge migration event occurs via the piezoelectric7 and Fröhlich interactions,5,40 respectively. Another well-established mechanism for coherent phonon generation, impulsive lattice heating induced by ultrafast carrier cooling,69 is expected to contribute negligibly in this case because the heating of the lattice is estimated to be

Pump-Power Dependence of Coherent Acoustic Phonon Frequencies

Recommend Documents