Exciton Localization and Dissociation Dynamics in CdS and CdSâPt

Subscriber access provided by YORK UNIV

Article

Exciton Localization and Dissociation Dynamics in CdS and CdS-Pt Quantum Confined Nanorods: Effect of Non-uniform Rod Diameters Kaifeng Wu, William E Rodríguez-Córdoba, and Tianquan Lian J. Phys. Chem. B, Just Accepted Manuscript • Publication Date (Web): 19 Jun 2014 Downloaded from http://pubs.acs.org on June 24, 2014

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.

The Journal of Physical Chemistry B 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 22

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

The Journal of Physical Chemistry

Exciton Localization and Dissociation Dynamics in CdS and CdS-Pt Quantum Confined Nanorods: Effect of Non-uniform Rod Diameters

Kaifeng Wu, William Rodríguez-Córdoba†, Tianquan Lian*

Department of Chemistry, Emory University, Atlanta, GA 30322, U.S.A. *Corresponding author: [email protected] †Present address, Escuela de Física, Universidad Nacional de Colombia Sede Medellín, A.A. 3840, Medellín, Colombia.

Abstract One-dimensional colloidal multicomponent semiconductor nanorods, such as CdSe-CdS dot-in-rod, have been extensively studied as a promising class of new materials for solar energy conversion because the possibilities of using the band alignment of component materials and rod diameter dependent quantum confinement effect to control the location of electrons and holes and to incorporate catalysts through the growth of Pt tips. Here we used CdS nanorods as an example to study the effect of non-uniform diameters along the rod on the exciton localization and dissociation dynamics in CdS and (platinum tipped) CdS-Pt nanorods. We showed that in CdS nanorods grown by seeded growth, the presence of a bulb with larger diameter around the CdS seed resulted in additional absorption band lower in energy than the exciton in the CdS rod. As a result, excitons generated in CdS rod could undergo ultrafast localization to the bulb region in addition to trapping on the CdS rod. We observed that the Pt tip lead to fast exciton dissociation by electron transfer. However, excitons localized on the CdS bulb showed a slower average ET rates than those localized in the rod region. Our findings suggested that the effect of rod morphology should be carefully considered in designing multicomponent nanorods for solar energy conversion applications.

1 ACS Paragon Plus Environment


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 22

KEYWORDS : CdS Nanorods, morphology inhomogeneity, exciton localization, exciton dissociation, light harvesting, semiconductor-metal heterostructures, transient absorption spectroscopy

Introduction Recent development of synthetic methodologies for controlling the size and shape of nanocrystals has led to colloidal one-dimensional (1D) quantum confined semiconductor nanostructures, such as nanorods (NRs) and tetrapods.1-3 In 1D NRs with diameters of a few nanometers and lengths of 10-100 nanometers, excitons are quantum confined in the radial direction but bulk-like along the long axis of the rod.

4-5

Compared to the zero-dimensional

quantum dots (QDs), these nanorods have enhanced chemical and photo stabilities,6-7 larger absorption cross sections,8-10 longer multi-exciton lifetimes,11-14 and linearly-polarized emission,15-20 making them promising materials for light emitting,21-22 optical gain,12, 23-24 and solar energy conversion applications.25-30 More importantly, 1D morphology facilitates the formation of various colloidal semiconductor/semiconductor heterostructures (such as dot-in-rod NRs, 14-15 tetrapods16-17, 31 and nanobarbells32-33), in which fast charge separation and long lived charge separated states can be achieved by controlling the band alignment of the constituent materials through their radii and compositions. For example, in CdSe/CdS dot-in-rod NRs or tetrapods, the quasi-type II band alignment can lead to the formation of a long lived charge transfer exciton state with the valence band (VB) hole in the CdSe core and the conduction band (CB) electron extending into the CdS rod.34-35

Further functionalization

of the NRs with metal particles at the tip (such as platinum and gold tipped nanorods36-38) can couple these long-lived charge separate states with catalysts to drive photocatalytic reactions.25, 27, 39-40 Because precise control of size and shape distributions of colloidal nanocrystals remains challenging,1, 41 it is important to understand the effects of these heterogeneities on exciton dynamics. For 1D nanorods, in addition to size and shape variation among different rods, heterogeneous distributions of rod diameters along the nanorod axis lead to a spatial 2 ACS Paragon Plus Environment

Page 3 of 22

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


distribution of the quantum confinement energies, which have significant effects on exciton dynamics.42-43 In a recent study of CdSe/CdS dot-in-rod NRs grown by a seeded growth technique, we show by TEM images that there exists a bulb region surrounding the CdSe seed with a larger rod diameter.

35

This morphology is likely a common feature of many

seeded-growth NRs,34, 44-48 although the spatial variation of rod diameters depends critically on the growth conditions and varies among samples. We showed that excitons generated in the CdS rod can be transported into the CdSe seed region, although this process competes with hole trapping-driven localization on the CdS rod, leading to the presence of three spatially separated long-lived exciton species that have different charge separation properties in the presence of electron acceptors.

35, 49

In these rods, the conduction and valence band

offsets between bulb and rod regions are contributed by the bulk band edge differences of the CdSe and CdS as well as the diameter differences of the rod and seed/bulb regions.35 To investigate the effect of the distribution in rod diameter alone, in this paper, we investigated the exciton dynamics in three CdS nanorods with different extents of rod diameter inhomogeneities. These CdS nanorod samples were grown from the same CdS seeds but with different growth times. We investigated the effects of the extent of morphology uniformity on the steady state absorption and emission spectra, as well as the exciton transport dynamics. We showed the presence of two distinct long lived excitons localized in the bulb and the rod regions, respectively. In the presence of Pt tip, these spatially distinct excitons exhibited different dissociation dynamics.

Methods

Synthesis of CdS NRs. Three CdS NRs were synthesized by a seeded-growth procedure starting with the same CdS seeds under the same growth conditions.17, 31

CdS QD

seeds with the lowest energy exciton band at 375 nm and estimated diameter of 2.65 nm were synthesized according to a published procedure,50 and described in the Supporting Information. 0.06 g Cadmium Oxide (CdO), 3 g tri-n-octylphosphine oxide (TOPO), 0.29 g octadecylphosphonic acid (ODPA), and 0.08 g hexylphosphonic acid (HPA) were degassed 3 ACS Paragon Plus Environment


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 22

under vacuum for 1 hour at 150 oC. After heating to 350 oC under nitrogen for half an hour, the mixture turned into clear solution, indicating the dissolution of CdO. At this point, 1.8 mL Trioctylphosphine (TOP) was injected into the solution. In a separated container, sulfur injection solution (0.12g S in 1.8 mL TOP) was mixed with CdS QDs. The amount of seeds in a typical synthesis was 8×10-8 mol. When the temperature of the Cd-containing solution was stabilized at 350 oC, the seed-containing sulfur injection solution was quickly injected. The temperature dropped down upon the injection and recovered to 350 oC in ~1 min. The solution was allowed to grow for 2, 4, and 8 min, respectively, for synthesizing NR1, NR2, and NR3 samples, respectively. Synthesis of platinum tipped NRs (CdS-Pt).

The platinum tipped NR1 (NR1-Pt)

sample was prepared according to a procedure in the literature.36 In a typical synthesis, a mixture of 0.2 mL oleic acid, 0.2 mL oleylamine, 43 mg 1,2-hexadecanediol and 10 mL diphenyl ether was degassed under vacuum for 1h at 80 oC. The temperature of the solution was raised to 190 oC under argon. 10 mg platinum (II) acetylacetonate was added to a suspension of NRs in dichlorobenzene and sonicated for 5 minutes to dissolve the Pt precursor. This solution was swiftly injected into the diphenyl ether solution. The reaction was allowed to proceed for 9 minutes. The product was washed twice by precipitation with ethanol followed by centrifugation, and finally dispersed in chloroform.

Visible femtosecond transient absorption spectroscopy.

A regeneratively amplified

Ti:sapphire laser system (Coherent Legend, 800 nm, 150 fs, 3 mJ/pulse, and 1 kHz repetition rate) and Helios (Ultrafast Systems LLC) spectrometers was used for the transient absorption measurements. The 800 nm output pulse from the regenerative amplifier was split into two parts with a 50% beam-splitter. One part was used to pump a Coherent Opera Optical Parametric Amplifier (OPA) which generates the signal and idler beams. The former was used to generate the 480 nm excitation beam by sum frequency mixing with the 800 nm fundamental beam in a BBO crystal. 400 nm pump beam was made by frequency-doubling of one part of 800 nm beam in the BBO crystal. A series of neutral-density filters were used to adjust the power of the pump beam. The pump beam was focused at the sample with a beam waist of about 300 µm. A white light continuum (WLC) from 420 to 800 nm was generated 4 ACS Paragon Plus Environment

Page 5 of 22

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


by attenuating and focusing ∼10 µJ of the 800 nm pulse into a sapphire window. The WLC was split into a probe and reference beam. The probe beam was focused with an Al parabolic reflector onto the sample (with a beam waist of 150 µm at the sample). Both the reference and probe beams were focused into a fiber optics-coupled multichannel spectrometer with complementary metal-oxide-semiconductor (CMOS) sensors and detected at a frequency of 1 kHz. The intensity of the reference beam was used to correct the pulse-to-pulse fluctuation of the white-light continuum. The pump beam was chopped by a synchronized chopper to 500 Hz; the difference between probe intensities of the pumped and unpumped sample was used to calculate the pump induced absorbance change. The delay time (up to ~ 1ns) between the pump and probe pulses was controlled by a motorized delay stage. The polarization of the pump beam was set at the magic angle of 54.7°with respect to the probe beam by a waveplate. The instrument response function (IRF) of this system was determined to be ~150 fs by measuring solvent responses under the same experimental conditions (with the exception of a higher excitation power). The sample was held in an 1 mm quartz curvette and stirred constantly by a magnetic stirrer during the measurements. Nanosecond Pump-probe Transient Absorption Spectrometer. Nanosecond TA measurement from 0.5 ns to microseconds was performed with the EOS spectrometer (Ultrafast Systems LLC). The pump beam at 400 nm was generated in the same way as the femtosecond TA experiments described above. The white light continuum (380-1700 nm, 0.5 ns pulse width, 20 kHz repetition rate) used here was generated by focusing a Nd:YAG laser into a photonic crystal fiber. The delay time between the pump and probe beam was controlled by a digital delay generator (CNT-90, Pendulum Instruments). The probe and reference beams were detected with the same multichannel spectrometers used in femtosecond TA experiments. The IRF of this system was measured to be ~280 ps. To construct TA spectra and kinetics over the entire delay time window from femtosecond to microseconds, the nanosecond TA sigals were scaled to match those of the femtosecond result in the overlapping delay time window (from 500 -1000 ps).

Results and discussions 5 ACS Paragon Plus Environment


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

Nanorod Morphology.

Representative Transmission Electronic Microscopy (TEM)

images of NR1, NR2, and NR3 (Figure 1a, b, c, respectively) showed that many CdS rods contain a bulb-like structure near one end with a diameter larger than the rod. This morphology was especially pronounced for NR1, as indicated by red dashed circles in Figure 1a. Similar structures have been observed in seeded-grown CdS NRs, including pure CdS NRs,44-46 CdSe/CdS NRs,34, 45-46 and ZnSe/CdS NRs.47-48 Typically, these bulbs surround the seeds, which likely results from non-negligible isotropic growth in addition to the prevalent anisotropic growth along the [001] crystal axis.34 For simplicity, we approximately represent the nonuniform morphology by a structure consisting of a rod region with uniform diameter and a bulb region with a larger diameter, as shown in Figure 2a, although it does not accurately describe those NRs with continuously shrinking diameter from the bulb to the end (see NR1 in Figure 1a). The exciton localization and dissociation dynamics in NR1 could be complicated by this distribution, which will be discussed later. The average NR lengths (and diameters) are 13.8 ±1.7, 18.1±1.8, and 26.9±2.1 nm, (and 3.75 ± 0.48, 3.79 ± 0.31 and 3.82 ± 0.27) for NR1, NR2, and NR3, respectively. Due to the presence of bulbs surrounding the seed, these diameters were measured at the relatively more uniform rod region. As a qualitative measure of the percentage of NRs with a pronounced bulb feature, we adopted the following criterion: if the rod diameter varies more than 50% along the NR length, a NR was considered to have a bulb. This value was chosen because the average diameters in the rod region are ~3.8 nm and 1.5 times of this value is roughly the Bohr exciton diameter for bulk CdS (~5.6 nm51-53). Bulbs of this size have absorption bands that can be clearly differentiated from the quantum confined nanorod absorption in the absorption spectra. Using this criterion, the percentages of NRs with bulbs were found to decrease from NR1 to NR3, as summarized in Table 1 and shown in Figure S1. This result likely indicates that most of the bulbs formed at early growth stage were digested and transformed into more uniform NRs. This shape distribution “focusing” effect is similar to the size distribution “focusing” effect observed in QD growth, which has been well explained by considering the kinetics of nanocrystal growth within the framework of Gibbs-Thomson theory.54


Page 6 of 22

Page 7 of 22

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


Figure 1. Transmission Electronic Microscopy (TEM) images of a) NR1, b) NR2, and c) NR3. Bulb-like structures can be observed, as indicated by the red dashed circles in a).

-‐

X2

a)

+ X1

-‐

-‐

b)

B2

X1

X2

B1

+

+

1σe

-‐

c)

1πe 1σe

1σh

-‐ Ef

X2 +

X1 +

Pt

1σh

Figure 2. a) Scheme of a nanorod with a bulb structure. The initial photo-generated excitons can be trapped in rod region or localized in to the bulb to form excitons X1 and X2, respectively. b) Energy level diagram of a nanorod, showing quantum confined levels of the rod and band edge of the bulb (solid lines) and trap states (gray dotted line). X1 and X2 7 ACS Paragon Plus Environment


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 22

indicate long live exciton states. B1 and B2 are lowest energy exciton bands in the rod and bulb regions, respectively. c) Energy level diagram of a bulb NR with a platinum (Pt) nanoparticle attached at the end. Two green arrows indicate dissociation of X1 and X2 by electron transfer to Pt.

Absorption and Emission Spectra.

The static absorption spectra of NRs dispersed in

chloroform are displayed in the upper panels of Figure 3. They are similar to previous reported spectra of CdS NRs.17,

39, 55

In these NRs with cylindrical symmetry, quantum

confinement in the radial direction leads to discrete electron and hole levels labeled as 1σ, 1π,…….43, 56-57 Because of the carrier motion in the radial direction is much faster than the axial direction, electron-hole interaction can be described by an effective 1D Coulomb potential that depends on their separation along the long axis of the NR. This 1D potential between the 1σ (π) electron and hole forms a manifold of bound 1Σ (Π) exciton states with the oscillator strength largely concentrated on the lowest energy exciton state, 1Σ0(1Π ).43, 0

56-57

Following these models, the peaks at ~390 nm (3.12 eV) and 450 nm (2.76 eV) can be

attributed to the lowest energy 1Π0 and 1Σ0 transitions, respectively, as shown in Figure 2b.39, 43

In addition, the absorption spectra also show an absorption tail that extends from the 1Σ0 band to ~500 nm, the absorption onset for CdS bulk crystals.58 To quantify the contribution of this feature, we fit this part of the absorption spectrum by two Gaussian functions, with the high energy B1 (the 1Σ0 exciton) band at ~2.76 eV and lower energy B2 band at ~2.64 eV. The fitting parameters for NR1, NR2 and NR3 are tabulated in Table S1. The ratio between the peak intensities of B2 and B1 bands, reflecting the relative contribution of the B2 feature, decreases from NR1 to NR3. This trend is positively correlated with the percentage of bulbs in these NRs determined from TEM image, as shown in Table 1, which suggests that the bulb features in the nanorod are responsible for the B2 band. As shown in Figure 2b, the larger size of the bulb feature reduces the confinement energies of both the electron and hole, which leads to a lower exciton energy.


Page 9 of 22

Abs (a.u.)

a) NR1 Abs Band Fit

b) NR2

c) NR3

Abs Band Fit

Abs Band Fit

B1 B2

PL Intensity (a.u.)

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


PL Band Fit

PL Band Fit

1.8

2.1

2.4

2.7

Energy (eV)

3.0

1.8

2.1

2.4

PL Band Fit

2.7

3.0

Energy (eV)

1.8

2.1

2.4

2.7

3.0

Energy (eV)

Figure 3. Static absorption (upper panel) and photoluminescence (lower panel) spectra (red open circles) of a) NR1, b) NR2, and c) NR3. Also shown are fits (black dashed lines) to these spectra and Gaussian components (blue solid lines) used in the fit.

The photoluminescence (PL) spectra of NR1, NR2 and NR3 measured under 400 nm excitation are shown in the lower panels of Figure 3. These PL spectra show three distinct bands and can be fit by the sum of three Gaussian bands centered at ~2.70 eV, 2.59 eV, and 1.95 eV, respectively. The fitting parameters are tabulated in Table S1. The higher energy bands can be assigned to emissions from B1 (~2.70 eV) and B2 (~2.59 eV), which are red shifted from their absorption peaks by 60 and 50 meV, respectively. The ratios between integrated areas of B2 and B1 emission bands, a measure of their relative emission intensities are summarized in Table 1. The relative intensity of B2 emission decreases from NR1 to NR3, correlating well with percentage of bulb features determined by TEM images and the static absorption intensity of B2 band. The correlation of these features is shown in Figure S1. The correlation suggests that B2 band cannot be attributed to the Urbach tail, which is commonly observed in distorted and impure bulk semiconductors, where the defect states interact with each other and form a sub-band below (above) the conduction (valence) band edge and therefore narrowing the band gap.59-61 The lowest energy emission band (~1.95 eV), red shifted by ~ 0.6 eV from the bulk band gap, can be attributed to the recombination between conduction band electrons and trapped holes, indicated as X1 and X2 in Figure 2b. Similar 9 ACS Paragon Plus Environment


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 22

trap emission bands have been observed in many CdS nanocrystals.39, 55 This assignment is also based on our previous study of CdS NRs,39 which shows that the conduction band electrons in CdS NRs are long-lived, whereas the valence band holes are trapped at electron-rich centers on the surface.62-63

The broad width of this feature likely contains the

contribution of the distributions of CB electron energy levels in the bulb and rod region, hole trap levels and strong hole-lattice coupling.64-65

Table 1. Percentage of bulb structures and absorption/PL intensity of B2 in NRs Bulb Percentage

Absorption Intensity

PL Intensity

NR1

32.6%

0.237

0.512

NR2

11.4%

0. 0901

0.336

NR3

3.3%

0.0344

0.116

Excition relaxation and localization dynamics on NRs. To examine how the presence of the bulb feature affects the exciton dynamics in nanorods, we first study the transient absorption spectra of NR1 with 480 nm excitation, which selectively excite the B2 band. The TA spectra (Figure 4a) show the bleach of B2 band with negligible contribution of the B1 feature. It has been shown that TA bleach of exciton bands in CdS nanocrystals is caused by the state filling of the CB electron level.39 Thus, the lack of B1 bleach feature indicates that B2 and B1 exciton bands involve different electron levels, consistent with assignment shown in Figure 2b. Transient kinetics of the bleach feature (Figure 4b) shows an instantaneous formation of the bleach, as indicated by a rise time of

Exciton Localization and Dissociation Dynamics in CdS and CdSâPt

Recommend Documents

Exciton Localization and Dissociation Dynamics in CdS and CdSâPt