Charge Transfer in CdSe Nanocrystal Complexes with an

Article pubs.acs.org/JPCC

Charge Transfer in CdSe Nanocrystal Complexes with an Electroactive Polymer Megan H. J. Oh,† Ming Chen,†,‡ Chi-Hung Chuang,§ Gerard J. Wilson,‡ Clemens Burda,§ Mitchell A. Winnik,† and Gregory D. Scholes*,† †

Department of Chemistry, Institute of Optical Sciences and Center for Quantum Information and Quantum Control, University of Toronto, 80 St. George Street, Toronto, Ontario, M5S3H6 Canada ‡ CSIRO Molecular Science and Engineering, Bag 10, Bayview Avenue, Clayton South, 3169, Victoria, Australia § Center for Chemical Dynamics and Nanomaterials Research, Department of Chemistry, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio 44106, United States S Supporting Information *

ABSTRACT: A better understanding of the essential interaction of semiconductor nanocrystals with their adsorbates and surrounding media can be used for controlling and optimizing the energy or charge transfer reactions found in these systems and also for comparing a diverse assortment of photoexcited charge transfer systems found in the latest materials research. In this study, the photoinduced interfacial charge transfer system, a complex of CdSe semiconductor nanocrystals and electroactive polymers containing ruthenium(II) tris(bipyridine), is characterized and then investigated in a comprehensive range of dielectric solvents from toluene to water. The effects of the solvent on the nanocrystal and adsorbate/ligand as well as the charge transfer dynamics of the system are explored through the fluorescence lifetime of the nanocrystal. Through this investigation it was found that fluorescence decays showed the presence of two decay components, which are influenced by solvent dielectric contributions on nanocrystal passivation and surface traps, and charge transfer processes present in the system. The results of the fluorescence decays were put into perspective using Marcus theory, providing some general insight on QD-based charge transfer systems and the effects of solvent polarity.

1. INTRODUCTION Photoinduced interfacial electron or charge transfer (ET or CT) between semiconductor nanocrystals (NCs), also known as quantum dots (QDs), and inorganic or organic adsorbates is a subject of great interest,1−10 especially for applications in nanocrystal-based photovoltaic and optoelectronic devices.11−17 Semiconductor nanocrystals, such as cadmium chalcogenides, have been extensively studied due to their unique quantum confinement effects and size-dependent optical and electronic characteristics.18−27 Many applications involve pairing the nanocrystal to other species, utilizing the versatile sizedependent properties of the individual nanocrystal, and correspondingly tuning the collective properties in the larger composite system. These systems have attracted considerable attention due to their potential use toward realization of artificial light-harvesting systems and other typical nanophotonic systems from a fundamental perspective. Numerous composite systems have been produced and investigated in recent years, including CdSe QD/TiO 2 nanoparticle systems by Kamat et al.,3−5 various pairings of semiconductor nanoparticles and dyes by Lian and co-workers © 2013 American Chemical Society

(e.g., CdSe QD/Rebipyridyl complexes, TiO2 nanoparticle/ Rebipyridyl complexes, CdS QD/Rhodamine B, etc.),6−10 CdSe QD/Ru−polypyridine complexes by Sykora et al.,28,29 and others,30−33 all of which provide excellent opportunities for improved understanding of charge transfer dynamics through their respectively unique systems. The studies focus on matters ranging from tuning the QD band gap for obtaining maximal energetics, broadening the spectral range of productivity through the use of various sensitizers, improving charge separation while suppressing charge recombination, to improving donor−acceptor coupling to enhance transfer efficiencies, all providing creative approaches for designing better QD-based electron transfer systems. Despite these innovative and important ideas, efficiencies will be limited by surface chemistry. Generally, a change in the solvent environment can affect the nanocrystal functionality, solubility and hence stability, as well as optoelectronic properties via Received: June 23, 2013 Revised: August 14, 2013 Published: August 23, 2013 18870

dx.doi.org/10.1021/jp406196a | J. Phys. Chem. C 2013, 117, 18870−18884

The Journal of Physical Chemistry C

Article

Scheme 1. Synthesis of PDMAEMA−Ru(bpy)32+−PDMAEMA (RuPDMAEMA) Block Copolymer

in an extensive range of media, ranging from nonpolar to polar aprotic and protic solvents, not originally available to the typical TOPO (trioctylphosphine oxide) passivated QD. Multifunctional use of the polymer affords a unique prospect of looking at our QD-based system under the influence of a diverse selection of dielectric media. Here, we describe the synthesis of the electroactive polymer and preparation of the complex, characterize its basic properties, conduct fluorescence decay experiments with respect to solvents with ranging polarities, and attempt to understand what important factors need to be considered when dealing with such systems.

nanocrystal surface passivation and traps. Hence, the essential interaction of the QD with the substrate/adsorbates and solvent and how it affects electron transfer will require reexamination, which will offer us a better understanding of the consequences of working with semiconductor nanocrystalbased systems. In this paper, we present an investigation on a CdSe QDbased electron transfer system using an electroactive polymer, [(PDMAEMA)−Ru(bpy)32+−(PDMAEMA)]. The polymer is a block copolymer composed of poly(N,N-dimethylaminoethyl methacrylate) (PDMAEMA) and ruthenium(II) tris(bipyridine) (Ru(bpy)32+). PDMAEMA is a polymer which incorporates repeating tertiary amino groups that allow for binding of the polymer to a variety of materials, such as CdSe semiconductor nanocrystals. Ru(bpy)32+ is an organometallic dye, which forms the main “electroactive” component in the polymer for this particular system, and has been used as a model photosensitizer in many dye-sensitized solar cell (DSSC) investigations due to many of its favorable properties.12−14,17 It has both oxidative and reductive capabilities, making it useful for promoting photochemical and electron or energy transfer reactions due to its long excited-state lifetime.34 Through the use of the multidentate polymer and a simple ligand-exchange procedure, not only does it offer passivation of the QD but also secure attachment of the functional electroactive component onto the QD surface is possible35−37 while solubilizing the QD

2. CDSE NANOCRYSTAL COMPLEXES WITH ELECTROACTIVE POLYMER (QD−RUPDMAEMA) Materials. Commercially available technical-grade chemical reagents and spectrophotometric-grade solvents were purchased from Aldrich. Trioctylphosphine oxide, trioctylphophine, and 100 mesh selenium powder were used without further purification. Acenaphthylene (Rütgers) was recrystallized three times from methanol. N-Dimethylaminoethyl methacrylate (DMA) was purified by passing a short column of basic alumina oxide. High-purity dimethylcadmium was purchased from STREM and used as received. 2,2-Azobis(2methylbutyronitrile) (AMBN) was obtained from Wako 18871



Article

Figure 1. Illustration of QD−RuPDMAEMA complex preparation via ligand-exchange reaction.

Precipitated polymer was finally dried under vacuum at room temperature to a constant mass (50% conversion). In the removal of the dithiobenzoyl end group from polymer,39 a mixture of RAFT−PDMAEMA−Ru(bpy)32+− PDMAEMA−RAFT, 20 mol equiv of AMBN, and toluene was degassed through three freeze−pump−thaw cycles, sealed under vacuum, and heated at 80 °C for 2 h. Polymer was then precipitated three times by addition into a rapidly stirred excess of n-hexane. Precipitated polymer was finally dried under vacuum to a constant mass. Synthesis of CdSe Semiconductor Nanocrystals. CdSe semiconductor nanocrystals were synthesized by a wellestablished organometallic method.18 A coordinating solvent, trioctylphosphine oxide (TOPO), was heated to a temperature of approximately 300 °C under an inert atmosphere. Colloid nucleation was initiated by a quick injection of 10.00 mmol of organometallic precursor (dimethylcadmium) and 13.35 mmol of chalcogenide (Se) precursors into the reaction vessel; then time was allowed for growth. After a desired size had been reached, further reaction was ceased by removal of heat. Once the QDs had been synthesized, size-selective precipitation was carried out by dissolving the dots in toluene and then adding a small amount of nonsolvent (methanol) to precipitate out the QDs. The step was repeated three times to narrow the sample size distribution and clean the sample of any excess residue, coordinating solvent, and contaminants. Synthesized QDs showed quantum yields of 0.5−1% and emission spectra fwhm (full width at half-maximum) of 35−40 nm. The surfaces of the QDs were capped by a monolayer of TOPO, allowing solubility in various types of organic solvents. Preparation of QD−RuPDMAEMA Complex. CdSe nanocrystals are prepared with TOPO,40−42 a monodentate organic ligand, which plays an important role in the nucleation, growth, and shape control of the nanocrystal during synthesis. Subsequent to its formation, the ligands, which are labile and in

Chemie Co. and used as received. Also, the molecular weights of the polymer were characterized by gel permeation chromatography (GPC) performed in tetrahydrofuran with 2% triethylamine (V/V, 0.6 mL/min) at 25 °C with a Waters 515 HPLC pump, a Viscotek VE3580 Refractive Index Detector, a VE3210 UV/vis Detector, and a Waters Styragel HR 4E column. GPC was calibrated with a narrow polydispersity polystyrene standards (MW from 580 to 377 400), and molecular weights are reported as polystyrene equivalents. Synthesis of RuPDMAEMA Samples. PDMAEMA samples (Mn = 27K, PDI = 3.9) were synthesized by a conventional solution polymerization of DMA in toluene at 95 °C, initiated with an azo-type free-radical initiator, AMBN. PDMAEMA(18)−Ru(bpy)32+−PDMAEMA(18) block copolymer with a functional Ru(bpy)32+ complex was synthesized using RAFT (reversible addition−fragmentation chain transfer) polymerization. A Ru(bpy) 3 2+ dye RAFT−acid, bis(dithiobenzoyl)-functionalized Ru(II) complex (Ru−diRAFT) was prepared as reported,38 forming the initial RAFT agent. Copolymerization of the DMA monomers with the RAFT agent, followed by a RAFT end group termination procedure, produced the final desired polymer. Scheme 1 outlines the synthesis, and the following discussion describes the procedures in more detail. For the synthesis of Ru(II) complex functionalized Ndimethylaminoethyl methacrylate polymer (RAFT−PDMAEMA−Ru(bpy)32+−PDMAEMA−RAFT), a mixture of DMA (6.37 M), AMBN (0.021M), Ru−di-RAFT (0.106 M), and toluene in ampules attached to a vacuum were degassed through three freeze−pump−thaw cycles and heated in a 70 °C oil bath for 8 h. Polymerization was terminated by rapid cooling in cold water. The polymer was then precipitated into a rapidly stirred excess of n-hexane. The solid was purified twice by redissolving in toluene and reprecipitated in n-hexane. 18872



Article

Figure 2. Absorption and emission spectra for CdSe QD, RuPDMAEMA, QD−RuPDMAEMA, and QD−PDMAEMA (left). Emission spectra of QD−RuPDMAEMA complex formation, taken at increasing time intervals, immediately after mixing the stock solutions to after 2 days of equilibration (right).

3. CHARACTERIZATION Scanning transmission electron microscopy (STEM) images were obtained in high-resolution annular dark-field (ZC) and transmission electron microscopy (TEM) modes using a Hitachi HD-2000 instrument. Absorption spectra were obtained on a CARY BIO UV−vis spectrophotometer. Steady-state photoluminescence (PL) spectra were measured using a Cary Eclipse fluorescence spectrophotometer. Fluorescence decay profiles were recorded in solution using a nanosecond time-correlated single-photon-counting system with a 456 nm NanoLED from IBH as the excitation source. Femtosecond transient absorption (TA) measurements were conducted using a Clark MXR 2001 fs laser system producing 780 nm, 150 fs pulses from a regenerative amplifier. The laser pulse train was split to generate a white light continuum probe pulse in a sapphire crystal and a tunable pump pulse. The tunable pump pulse was generated using an optical parametric amplifier (TOPAS, Lightconversion). All femtosecond laser experiments were carried out in a 2 mm quartz cuvette at room temperature. Instrumental time resolution was determined to be ∼150 fs via a pump−probe cross-correlation analysis. Reported spectra were all collected from fresh samples at room temperature. STEM Images, UV−vis Absorption, and Fluorescence Spectroscopy. In order to obtain high-resolution annular dark-field images from the STEM, significantly diluted QD− RuPDMAEMA samples were prepared in toluene. The images (see Supporting Information Figure S3) show that the complexes were well dispersed and do not aggregate following ligand exchange, demonstrating solution stability. The TEM images obtained, which do not resolve organic matter (i.e., PDMAEMA polymer, TOPO ligand), also confirm the dispersion of the complexes and show that there are no observable physical changes between the original QD−TOPO and the newly formed QD−RuPDMAEMA samples. Absorption and emission spectra of the CdSe QDs, RuPDMAEMA polymer, QD−PDMAEMA complex, and QD−RuPDMAEMA complex are shown in Figure 2 (left). QDs used in formation of the complex exhibit the first absorption peak at 530 nm (2.34 eV) and fluorescence peak at 542 nm (2.29 eV). A QD average diameter of approximately 2.7

dynamic equilibrium with the surrounding medium, also prevent aggregation, determine the dispersion interactions of the particles, and passivate (electronically) the surface defects found on the nanocrystal. Thus, the surface of the nanocrystal and the type of ligands associated with the surface strongly influence many of the physical and chemical behaviors and interactions of the nanocrystals. Ligand exchange is a procedure that allows for substitution of the existing surface-associated ligands with another type through prolonged exposure of the nanocrystals (in solution) to an excess amount of competing ligand of choice. Generally, as the new incoming ligand molecules bind more strongly to the nanoparticle surface, the existing ligands will be exchanged in a process driven by mass action. The degree of exchange will strongly depend on the ligand properties and its affinity to the nanocrystal surface. The QD−RuPDMAEMA complex exploits this convenient ligand-exchange procedure (Figure 1). The repeating units of the dangling dimethylaminoethyl group, or more specifically the tertiary amine group, off the main methacrylate backbone of the PDMAEMA polymer have an affinity to the Cd2+ ions on the surface of the QD, similarly to the TOPO. Owing to the multidentate binding power of PDMAEMA, the original TOPO ligands are displaced and the block copolymer binds strongly to the nanocrystal surface. The procedures are as follows. QD− RuPDMAEMA complexes, and similarly QD−PDMAEMA complexes (without Ru(bpy)32+), were prepared by addition of an aliquot of a size-precipitated QD stock solution in toluene (∼7.17 × 10−7 mol/L) to 1 mg of RuPDMAEMA copolymer (or PDMAEMA polymer) to form a mixture of solution with a concentration of approximately 5.859 × 10−5 mol/L. The mixture was left stirring for 2 days at room temperature in the dark to allow the ligand-exchange reaction between the block copolymer and the original TOPO ligands to take place. To transfer the complex into the various solvents, the newly formed complexes are precipitated using hexane (a nonsolvent) and collected through centrifugation. Subsequently, the precipitant was redissolved in a solvent of choice via sonication. It is important to note that both the excess polymer and the complex are soluble and present in the solvents used. 18873



Article

Figure 3. TA measurements of QD (green), RuPDMAEMA (yellow), and QD−RuPDMAEMA (red).

nm can be estimated from the energy of the lowest exciton band (see Supporting Information) and roughly confirmed by the STEM images. The RuPDMAEMA polymer shows an absorption maximum at 458 nm (2.71 eV) and emission peak at 627 nm (1.98 eV). QD−RuPDMAEMA complex absorption and fluorescence spectra are comprised of the same peaks and

peak locations as its constituents. Also, the QD−PDMAEMA complex, similar to the QD−RuPDMAEMA complex but in the absence of Ru(bpy)32+, has an absorption spectrum that is nearly identical to the CdSe QD samples, within the visible wavelength range of interest. 18874



Article

Table 1. Analysis of Kinetic Parameters of QD, RuPDMAEMA, and QD−RuPDMAEMA Complexes in Toluenea τ1 RuPDMAEMA QD QD−RuPDMAEMA

τ2

excitation Φ (mW)

excitation λ (nm)

probe λ (nm)

constant (ps)

percentage (%)

constant (ps)

percentage (%)

τave (ps)

0.9 0.06 0.06 0.9 0.9

458 458 458 458 458

465 532 532 532 465

8.19 4.79 5.74 0.73

56 59 63 82

267 223 157 15.2

44 41 37 18

1207 122 94.3 61.7 3.33

Φ is power, λ is wavelength, τ is the decay time constant, and τave is the particle-averaged lifetime, τave = ΣAnτn/ΣAn. Decays were fit using a biexponential function. Contributions of each exponential are represented as pre-exponential factors, A1 and A2. Hence, the percentage values for τ1 and τ2 are obtained as A1/(A1 + A2) and A2/(A1 + A2), respectively. a

The QD−RuPDMAEMA complex formation following mixing of the initial components was systematically monitored through the emission intensity at increasing time intervals (Figure 2, right) up to approximately after 2 days. QD (∼542 nm) and RuPDMAEMA (∼630 nm) PL peaks were found to decrease in intensity over time. Simultaneous quenching of both peaks is not characteristic of resonance energy transfer (RET), since no signs of donor emission quenching are followed by an acceptor peak sensitization. RET cannot be fully discounted due to a partial donor and acceptor spectral overlap present in the system, but it does not appear to be the dominant photophysical process. Sykora et al.28 also reported similar quenching in their investigation with CdSe QDs and ruthenium polypyridine complexes. They explained that the quenching occurred upon gradual adsorption of the complexes onto the QD surface and attributed it to electronic interactions between the two components, most likely a charge transfer reaction. On closer inspection, the emission spectra reveal stronger quenching of the QD emission peak compared to the RuPDMAEMA polymer. The initial ratio of the QD:RuPDMAEMA peak intensities at the start of mixing is 0.28:1 and changes to 0.17:1 after 2 days. The concentration of the components remains fixed during this equilibriation or complex formation time, and results are reproducible with the same starting ratio concentrations. The stronger quenching of the QD PL suggests that additional RuPDMAEMA polymers adsorb onto the QD surface over time, which can allow for the possibility of multiple, concurrent charge transfer reactions to occur, resulting in a more rapid decrease in the QD emission intensity compared to the RuPDMAEMA PL band. Since the polymer was added in excess, the presence of the free polymer in the mixture is expected to contribute significantly to the steady-state RuPDMAEMA emission peak intensity. The quenching in the emission intensity of the RuPDMAEMA also seems to be in association with the polymer’s increasing fractional binding to the QD surface over time. Sykora also reported similar quenching trends for their system using different concentrations; however, the QD peak eventually fully quenched in their studies after 39 h. We did not further explore the effect of various concentration ratios on the PL quenching of QD−RuPDMAEMA due to the limited amount of RuPDMAEMA available for the study. Oxygen can also act as a quencher and influence the luminescence efficiency of the CdSe nanocrystals and RuPDMAEMA. However, all measurements were performed from “freshly” made batches of QD− RuPDMAEMA samples, and effects of oxygen are assumed to be minor and relatively stable for the system throughout the duration of the experiments, since no visible changes could be noticed in the absorption and PL spectra.

Dynamic light scattering (DLS) techniques were employed to see if we could acquire size, size distribution, and diffusion coefficient information for the QD−RuPDMAEMA complex; however, due to the interference of the RuPDMAEMA polymer used in excess, non-negligible multiple scattering contributions limited the useful information that could be obtained. The average number of polymer units per QD cannot be straightforwardly extracted from these ratios due to the excess of free polymer, although an estimate can be obtained from previous work,37 which illustrates a method of isolating a similar QD−polymer complex from the free polymer and quantifying this information using size-exclusion chromatography and steady-state spectroscopic analysis. It was found that on average approximately 4−5 PDMAEMA polymer molecules (Mn = 6.7K, labeled at one end with a pyrene molecule) were bound to a single CdSe QD of ∼2.5 nm in diameter. These approximate figures may be applied to our system given the similarities in the components (CdSe QD with d = 2.7 nm, Ru(bpy)32+-labeled PDMAEMA polymer with Mn = 6.68K). Transient Absorption. Figure 3a shows the femtosecond transient absorption (TA) spectra for the CdSe QD, RuPDMAEMA polymer, and QD−RuPDMAEMA complex in toluene following ∼150 fs fwhm excitation at 458 nm. We begin with a typical description of the observed features, and then further insight into these features will be presented in the following paragraphs.43 Considering first the QD, transient bleach signals at ∼530 nm (Figure 3a, green) corresponding to the nanocrystal exciton transition are apparent. In the QD measurements, the excitation power was controlled to ensure that the average number of excitations was below one. The long-lived transient bleach signal of the RuPDMAEMA polymer is seen at ∼455 nm (Figure 3a, yellow). The location of the bleach band in the TA and the absorption maximum in the steady-state spectra are consistent. The TA spectrum of the QD−RuPDMAEMA complex in toluene is shown in red (Figure 3a). The bleach signals near 530 and 455 nm correspond to the QD and RuPDMAEMA states, respectively (Figure 3a, 0.9 mW). For the QD−RuPDMAEMA complex, a high excitation power was required to obtain bleaching signals in both spectral positions of the CdSe and RuPDMAEMA transitions. Figure 3b (left) and Table 1 show the temporal signal decays at 465 nm for the RuPDMAEMA polymer and QD− RuPDMAEMA complex. According to steady-state absorption, the kinetic traces measured at this probe wavelength should represent a sum of both the ground-state (ground-state bleaching) and the excited-state (stimulated emission) signals. Due to the strong ground-state absorbance of Ru(bpy)32+ of the RuPDMAEMA, the temporal profile typically seen would represent ground-state recovery. The kinetic trace of the free 18875



Article

RuPDMAEMA is long lived with no decay in the first 100 ps. In the absence of any reactions (i.e., charge transfer), it is apparent that GSR takes a long time to occur because the excited state is long lived.44 However, in the QD−RuPDMAEMA complex, an extremely fast decay (in comparison) is observed. A biexponential fit was used to obtain two components, 0.73 (82%) and 15.2 (18%) ps, yielding an average decay time constant of 3.33 ps. Here, the bleach is very short lived, and it is difficult to distinguish whether GSR or excited-state dynamics is dominating the signal measured; however, there seems to be some unique process(es) present for the polymer-bound complex not present in the free polymer, which is responsible for the exceptionally faster dynamics that are occurring on a picosecond time scale. The mechanistic details of the earliest moments following photoexcitation are influential in processes associated with, and subsequent to, excited-state formation (i.e., excited-state reactivity). In general, it is believed44−49 that upon excitation at ∼450 nm of Ru(bpy)32+ an electron from the ruthenium metal is promoted to one of the bipyridyl ligands, resulting in an excited MLCT singlet-state (1 MLCT), leaving the ruthenium metal oxidized (Ru3+) and the excited electron localized on one of the ligands (bpy•−), as shown in Scheme 2.

absorption feature at time zero leads to the possibility that the localized electron on one of the ligands may already have transferred (i.e., with the possibility of ET to the QD conduction band) within the 150 fs of the excitation pulse. Thus, the transient bpy•− feature remains absent for the duration of the measurement (1399 ps). The diminishing Ru(bpy)33+ feature shows the disappearance of the transient species over time, which suggests that Ru(bpy)33+ is being converted back to Ru(bpy)32+ (i.e., ground-state recovery) through some other process. This other process may be a charge recombination process, where the transferred electron from the QD conduction band recombines with the Ru(bpy)33+, or a side reaction between the strongly oxidizing Ru3+ and the excited electron from the QD conduction band. Since a high excitation power was used to obtain bleach signals in both the QD and the polymer spectral positions for the QD−RuPDMAEMA complex, simultaneous excitation of the QD and polymer species in the complex may likely have induced a side redox reaction between the Ru3+ form and the excited QD, causing the ruthenium species to recover to its Ru(bpy)32+ state.44,46,50−55 Kinetic traces at 532 nm were also measured for the free QD and QD−polymer-bound complex (Figure 3b, right). Owing to the large near degeneracy of valence band (VB) edge states compared to the conduction band (CB) edge states in the CdSe QD, a difference of 0.05 eV or 20 nm red shift between the 1S absorption band and the PL maximum, it has been stated that the TA profile measured at this transition provides information about the depopulation rate of the electrons from the conduction band edge states.28,56,57 The TA data for QD samples were obtained using a low excitation power which made for a single-exciton scenario in the CdSe QDs. An average decay time of 122 ps was obtained using a biexponential fit (8.19 (56%) and 267 ps (44%)), which is a typical value assigned for electron relaxation from the conduction band to the defect surface states. Measurements were obtained for two excitation powers in the QD−RuPDMAEMA complex (Table 1). For data gathered at 0.06 mW, the polymer-bound complex had an average decay time of 94.3 ps (4.79 (59%) and 223 ps (41%)), while at 0.9 mW, the complex had an average decay time of 61.7 ps (5.74 (63%) and 157 ps (37%)). The higher excitation power (0.9 mW) was used to observe both the QD and the RuPDMAEMA bleaching; hence, there may have been the possibility of multiple excitations generated in the QDs for the QD−RuPDMAEMA samples. There is an overall faster decay rate measured for the complex compared with the decay of the free QD in either the low- or the high-excitation power scenarios; however, due to the possibility of multiexciton effects generated in the complex for the high-power case, the faster decay may be partially accredited to this side effect. Single excitons mainly decay through slow radiative recombination which occurs on a nanosecond time scale, while multiexcitons mainly decay through fast nonradiative Auger recombination which occurs on a picosecond time scale. Also, the redox reaction resulting from simultaneous excitation of the QD and polymer species will contribute to the faster excited-state decay observed. If the lower excitation power scenario is considered, where multiexciton and simultaneous excitation effects are less likely, there seems to be some process that is repopulating the ground state at a faster rate compared to the free QD case. Similar systems consisting of ruthenium polypyridine dyes and CdSe QDs were also investigated by Schaller et al.56 using TA and

Scheme 2

However, the early time excited-state dynamics which lead to this emissive state are still in question. It is believed that photoexcitation of Ru(bpy)32+ in the 1MLCT manifold produces an initial Franck−Condon state that is delocalized on all ligands.44 Then in tens of femtoseconds, the promoted electron becomes localized on one of the ligands, due to local solvent dipole coupling. Intersystem crossing (ISC) takes place afterward within about 100 fs, producing a localized triplet state. Finally, on a time scale of 10 ps, the 3MLCT becomes randomized through interligand hopping (thermalized 3MLCT excited state). Since our work has been carried out with ∼150 fs resolution, insight into these initial processes cannot be assessed, but based on the literature, the data that has been acquired largely reflect the dynamics on the 3MLCT surface (since ISC is presumed to have already occurred within our excitation pulse).44−49 The RuPDMAEMA TA spectrum obtained is similar to that of the difference absorption spectrum in H2O for Ru(bpy)32+ taken from the literature.50 Interpretation of the difference spectrum reveals intense positive absorption bands in the 360− 380 nm region and positive low-amplitude residual absorption at wavelengths greater than 500 nm due to transitions localized on the bpy•− ligand, while a negative absorption band in the 400−450 nm region is representative of Ru(bpy)33+. Considering the TA spectra for the RuPDMAEMA polymer and QD− RuPDMAEMA complex (Figure 3a, yellow and red, 0.9 mW), based on the above interpretation, the polymer spectra contain both features, the bpy•− ligand (>550 nm, pronounced at 690 nm) and Ru(bpy)33+ (∼455 nm), and are relatively constant with respect to time, suggesting that these particular transient species of the excited state are long lived, while in the QD− RuPDMAEMA spectra the Ru(bpy)33+ feature is seen diminishing over time and the bpy•− ligand feature seems to be almost nonexistent (near 690 nm ΔA is near 0) for all times. The time resolution of 150 fs and the lack of a bpy•− ligand 18876



Article

Figure 4. Illustration of the possible processes available to QD−RuPDMAEMA complex. Green and yellow colors represent CdSe QD and Ru(bpy)32+, respectively.

mass)17,58−60 and the first absorption peak energy of the QD (E1s) using Brus’ equation

time-resolved up-conversion experiments. In their studies, they determined that a hole transfer from the CdSe QD to the ruthenium polypyridine dye occurred (in ∼5 ps) in their systems, with the QDs serving as sensitizers. This explanation would complement the faster GSR dynamics that is observed for the QD−RuPDMAEMA samples. A hole transfer from the QD (VB) to the RuPDMAEMA (ground state) has an equivalent net effect as an electron transfer from the RuPDMAEMA to the QD. Therefore, the transferred electron could replenish the bleached valence band state of the QD; hence, a faster GSR may result for the QD−polymer-bound complex.

⎛ m* ⎞ h ⎟ ECB(QD) =ECB(bulk) + ΔEconfinement⎜ ⎝ mh* + me* ⎠ = ECB(bulk) + (E1s(QD) − Eg (bulk)) ⎛ m* ⎞ h ⎟ ⎜ ⎝ mh* + me* ⎠

(1)

The calculated CB edge is then put on a vacuum scale using the bulk CdSe electron affinity, χs, value (χs = −4.95 eV) (see Figure S4a, Supporting Information). Due to the redox versatility of Ru(bpy)32+, both the ground- and the excitedstate redox potentials for RuPDMAEMA should be considered to determine the full range of processes which may be compatible with the energies of the QD. The excited-state redox potentials were also estimated using the ground-state reduction potentials of Ru(bpy)32+ (1.26 vs SCE for Ru(bpy)33+/2+ and −1.28 vs SCE for Ru(bpy)32+/+, in aqueous

4. POSSIBLE PROCESSES OF THE PHOTOEXCITED QD−RUPDMAEMA COMPLEX The relative offset of the valence/conduction band edge states for the QD and the reduction potentials of the RuPDMAEMA (i.e., Ru(bpy)32+) must be considered. The energy of the QD CB edge was estimated from bulk CdSe values (Eg = 1.74 eV, ECB = 4.95 eV, and effective electron and hole masses of me* = 0.13m0 and mh* = 1.14m0, where m0 is the electron rest 18877



Article

Table 2. Calculated Average Lifetimes of QD−PDMAEMA and QD−RuPDMAEMA Complexes in Various Solvents, along with Selected Solvent and CdSe QD Physical Propertiesa

a

solvents

μ (D)

n

ε

⟨τQD−PDMAEMA⟩ (ns)

⟨τQD−RuPDMAEMA⟩ (ns)

toluene THF DMF water CdSe QD

0.36 1.63 3.82 1.85 41

1.496 1.407 1.431 1.333 2.67

2.38 7.52 38.25 80.10 10.2

25.09 20.88 18.64 9.39 ⟨τQD⟩ (in toluene) = 12.06

40.20 76.24 171.36 230.56

μ is the permanent dipole moment in debyes, n is the refractive index, and ε is the static dielectric constant.

solution)28,44 along with the energy of the emission peak of the RuPDMAEMA (Eem0−0)

observed kinetics of this complex system. It is important to mention that different mechanisms have been proposed for similar nanocrystal−dye systems,3−10,28−33 and although the systems look seemingly identical in composition and construct, the different mechanisms put forth may all be valid. A general mechanism is not possible for such systems due to the versatility of the QD band gap. A slight difference in properties can shift the relative energies of the nanocrystal in relation to the dye, which can considerably change the likelihood of the processes available, hence altering the dominant mechanism of the system. Thus, it is imperative to consider the relative energies on a case by case scenario.

0−0 E1/2(*Ru(bpy)33 + /2 + ) = E1/2(Ru(bpy)33 + /2 + ) − Eem

(2) 0−0 E1/2(*Ru(bpy)32 + / + ) = E1/2(Ru(bpy)32 + / + ) + Eem

(3)

SCE stands for the saturated calomel electrode (with standard electrode potential, E = +0.244 V saturated). Converting SCE to NHE, which stands for normal hydrogen electrode (E ≈ 0.000 V), and using the value with respect to a vacuum level (with reference or vacuum electrode potential of Eref = −4.50 eV), the energies of the QD and redox potentials of the RuPDMAEMA can be compared on a relative redox potential energy diagram (for a diagrammatic illustration of how the band edge energies and redox potentials are put together, see Figure S4c, Supporting Information). Several energy and charge transfer processes, resulting from single excitations of either the QD or the polymer, are potentially available to the QD−RuPDMAEMA system, and Figure 4 illustrates these processes. Given the spectral overlap of donor emission and acceptor absorption spectra, resonance energy transfer (RET) from the photoexcited QD to the RuPDMAEMA or the photoexcited RuPDMAEMA to the QD are both possible. However, given the small spectral overlap, in the case of RET from the photoexcited QD to the RuPDMAEMA, unlikely to be significant. In the case of RET from the photoexcited RuPDMAEMA to the QD there is a fairly good spectral overlap for the process to be plausible; however, this particular process will result in a quenching of the polymer emission followed by sensitization of the QD emission, which is not apparent in the steady-state complex formation emission spectra. Hence, RET from the polymer to the QD does not seem to be the dominant process. On the basis of the energetics of the system, there are also two charge transfer processes which are likely. The first involves a hole transfer from the QD to the RuPDMAEMA polymer, following QD photoexcitation, and the second involves an electron transfer from the RuPDMAEMA to the QD, following RuPDMAEMA photoexcitation. Despite the mechanism, the final state will be the same in either case, with the extra electron residing on the QD conduction band (QD−) and a deficit in electron density on the ruthenium metal ion (Ru(bpy)33+ ≡ [RuIII(bpy)3]3+). The charged QD− is said to be in a dark nonemissive state because the additional electron in the conduction band decays through high-efficiency, nonradiative Auger recombination, while the oxidized Ru(bpy)32+ of the RuPDMAEMA polymer will also be in a nonemissive Ru(III) form.28 It is difficult to deduce a clear dominant mechanism, and most likely a combination of these processes contributes to the

5. SOLVENT STUDIES Generally, the local environment can have a substantial effect on the ground- and excited-state energies of molecules and consequently their optical properties, especially with regard to the excited-state and emissive characteristics.61−72 Molecules are small and hence closely embedded in the polarizable medium. Therefore, the locally surrounding solvent, provided that they are to some degree polar, can experience polarization effects from the molecule, enabling them to reorganize and result in an overall energy stabilization of the system. This is one of the origins of the Stokes shift.73,74 As the solvent polarity increases, this effect becomes more pronounced and results are reflected in lower emission energies. Due to the larger size of QDs in comparison with molecular systems and their significantly large dipole moments (e.g., ∼41 D for QD with an approximate diameter of 2.7 nm, compare with solvent dipole moments listed in Table 2),75−77 QDs are expected to induce a very strong polarization on its environment. However, the surrounding solvent molecules will not be able to reciprocate an effect of a similar magnitude on the QDs (because of the large distance from the medium to the center of the QD). Consequently, the bath reorganization energy is expected to be nominal, along with the energy stabilizing role of the solvent. Analogous to solvent effects on emission properties of molecules, solvation plays a crucial role in the rates of charge transfer reactions. A change in solvent polarity can affect the energetics of the donor and/or acceptor, their electronic coupling, and the activation barrier of the reaction, all of which affect the rate. These effects in conjugate semiconductor nanoparticle and molecular adsorbate systems and on the charge transfer processes found within have not been adequately addressed. We investigated solvent contributions on the QD−RuPDMAEMA system using nanosecond TCSPC measurements to obtain further insight on QD-based CT systems. The photoluminescence decays of both the QD−PDMAEMA and the QD−RuPDMAEMA complexes were measured in four sovents, toluene (Tol), tetrahydrofuran (THF), dimethyl18878



Article

PDMAEMA lifetimes can be attributed to changes in surface trap states with respect to the changing dielectric environment. It has been shown that surface trap states can quench photoluminescence on short time scales.57,79 Trapped holes in CdSe QDs are short lived, detrapping on a picosecond time scale (∼3−50 ps), faster than Auger decay, and can result in both radiative and nonradiative recombination. Trapped electrons are long lived and detrap on a nanosecond time scale (∼1−100 ns), competitive with PL time scale, and typically decay via nonradiative recombination. This insinuates that as the QD−PDMAEMA are solvated in higher dielectric solvents, more surface traps seem to be available, trapping the excited electron or hole and reducing the excited-state population, hence decreasing the average PL lifetimes.80−84 Then what is causing the increase in availability of surface traps? It has been previously reported that the QY of CdSe QDs are dependent on the type of ligands adsorbed onto the surface of the nanocrystal, which changes the passivation and in turn affects the efficiency of the surface trapping of charge carriers.85−87 Consider the possible role of surface passivation of the QD. While the experiment was performed with freshly prepared samples of QD−RuPDMAEMA and QD−PDMAEMA that looked visibly clear and stable and formal timedependent sample stability studies were not conducted, it was noted that PDMAEMA- or RuPDMAEMA-coated QDs seemed to be most stable in toluene and the least in water. Small amounts of scatter were noticed in the absorption spectra for water (see Figure S2, Supporting Information), and varying degrees of aggregation in the samples, in vacuum sealed tubes, were observed over time in DMF and water. Sample aggregation in water occurred after approximately several weeks after sample preparation, while sedimentation was noticed for DMF after a few months. From this observation we propose that as higher dielectric solvents are used, the attractive polymer−solvent interaction diminishes, making the system unstable and prone to aggregation over time. The attractive polymer−solvent interaction can cause the polymer to shrink or collapse, reducing the number of available binding sites. With a reduction in the number of binding sites, the probability of adsorption onto the QD surface decreases and the polymer tends to desorb, hence also exposing more surface traps (Figure 7). The decay profiles of the QD−RuPDMAEMA samples show the presence of a fast and slow decay component. These components signify that at least two distinctive mechanisms are responsible for deactivation of the excited state. The initial 50 ns emphasizes the fast component and shows decay trends similar to the QD−PDMAEMA complex. It is likely resulting from higher exposure to surface traps with respect to the solvent polarity. Aside from this decay trend, these fast components show a subtly faster PL decay as a function of the increasing solvent polarity, compared to the QD− PDMAEMA samples (Figure 6, insets). The additional faster decay, or quenching, may be attributed to the charge transfer reaction occurring between the QD and the RuPDMAEMA. Since, the nanosecond measurements cannot resolve these ultrafast dynamics, below ∼22 ns we are most likely capturing the results following the CT reaction. Focusing on times greater than 100 ns, the additional slow component defining the tail of the decay profile shows opposite solvent trends in comparison with the fast component. The slow component may be due to a delayed radiative

formamide (DMF), and water, at the QD emission (550 nm) using a 456 nm excitation source. Absorption spectra of the QD−RuPDMAEMA in different solvents are shown in Figure 5. The PL decays obtained were fitted using a standard

Figure 5. Absorption spectra of QD and RuPDMAEMA in toluene and QD−RuPDMAEMA in toluene, THF, DMF, and water.

nonlinear, least-squares method with a multiexponential decay function (eq 4). The chosen fit is, to some extent, arbitrary; however, it is a typical model used to characterize decay kinetics of QD systems.78 Also, for convenience we compare average fluorescence lifetimes (⟨τ⟩) in the analysis, which is the average time a molecule spends in the excited state before emitting a photon and returning to the ground state, hence the mean time taken for a photon emission (eq 5). I (t ) =

n

⟨τ ⟩ =

t⎞ ⎟ ⎝ τn ⎠ ⎛

∑ αn exp⎜−

(4)

∑n αnτn2 ∑n αnτn

(5)

The two sets of decays measured show opposing trends in the calculated average lifetimes (Table 2). The average lifetimes of the QD−PDMAEMA complexes decrease with increasing dielectric constant of the solvent, from toluene to water. Conversely, the average lifetimes of the QD−RuPDMAEMA complexes increase with the same increase in solvent dielectric constant. Figure 6 compares the two complexes in each of the various solvents separately. Subsequent to excitation of the CdSe nanocrystal, many processes can be responsible for relaxation of the innate QD excited-state population, which includes radiative and nonradiative band edge exciton recombination, radiative and nonradiative surface-assisted exciton recombination, and nonradiative Auger recombination. The rates of these radiative and nonradiative processes will affect the PL lifetime of the QD. Since all of the samples were prepared from the same batch of QDs and PDMA polymer and the same experimental conditions were imposed during these measurements, the main factor for the variations that are observed in the QD− 18879



Article

Figure 6. PL decay of QD−PDMAEMA and QD−RuPDMAEMA in various solvents. (Insets) Close up of the fast decay components.

transfer from the QD to the RuPDMAEMA polymer (induced by QD photoexcitation), followed by BET from the QD to the polymer, replacing the hole and restoring the exciton on the QD, allowing for subsequent radiative charge recombination may be a possible mechanism that would result in this observation. This would result in a delayed emission from the QD represented by the slow component in the PL decay profile. If the slow component does indeed represent the BET reaction, the solvent trend seen for this component appears consistent with expectations. Once charge transfer has occurred, the polar solvent molecules are able to reorganize around the system to accommodate for the new charge distribution, stabilizing the products of the ET reaction. In order for BET to occur, sufficient energy is required to overcome this stability. In cases where higher polarity solvents are used, a greater solvent stabilization will result, which will likely further impede the BET reaction.

Figure 7. Illustration of solvent dielectric effects on QD passivation.

recombination resulting from a back electron transfer (BET) reaction. To be able to observe BET at the QD emission, it would seem that a charge transfer processes involving a hole 18880



Article

reorganization energy is a modified equation for use at the planar semiconductor−liquid interface. It is calculated using a dielectric continuum model given by the dielectric constant of the solvent and semiconductor.7,32,33 The inner-sphere (λi) contributions to the reorganization energy are expected to remain relatively constant over the different solvents; hence, the overall reorganization energy was calculated using only the outer-sphere (λ = λo) solvent contributions. As shown by our analysis summarized in Table 3, as solvent polarity increases, λo also increases because higher dielectric solvents exhibit greater interaction with the charge distribution changes of a system. When λo and ΔG°ET (−0.75 eV) values are compared, the ET reaction is considered to be in the Marcus inverted region (λ < |ΔG°|) for all solvents. A greater difference between λo and |ΔG°ET| in the inverted region will cause the reaction rate to diminish. This can be seen in our calculated kET values with water having the highest ET rate and toluene having a negligible rate (N.B. the calculated rates will be overestimated, leading to faster rates, using this model in the inverted region). This predicted trend is consistent with the fast component of our QD−RuPDMAEMA PL decays. Water is a highly polar solvent, and it has a high λo value near |ΔG°ET|. As λo and |ΔG°ET| values become similar, the reaction converts from the inverted to the optimal region, characterized by a barrierless (ΔG‡ET = 0) transition, resulting in an optimal ET rate. Conversely, toluene is a nonpolar organic solvent and thus has a very low λo. Hence, it is expected that toluene will aid very little in the ET process, as can be seen in the relatively high ΔG‡ET and ΔG‡BET values, resulting in unusually slow calculated rates, since the model breaks down at this limit. This is also reflected in the relatively similar PL decay curves obtained for the QD−RuPDMAEMA and QD−PDMAEMA in toluene. BET rates were initially calculated using eq 9 to obtain values for the BET activation energy (ΔG‡BET). However, with the reaction being in the Marcus inverted region, the calculated BET rates also came out with the same solvent trend as the forward ET rates, which does not fit the slow component trend of the QD−RuPDMAEMA PL decays. The incongruence between the model and the experimental results indicates that certain factors have been overlooked. As suggested in the experimental results of the QD−PDMAEMA samples, solvent effects on QD passivation and surface states seem to play an important role in ET dynamics of the QD−RuPDMAEMA system. Since hole-trapping events seem to occur rapidly and with relative ease in CdSe QDs,57 this may provide a rationale for surface-assisted hole transfer back from the RuPDMAEMA to the QD. Surface-assisted BET will decrease the difference in Gibbs free energy for the reverse reaction states, and a suitable solvent trend can be produced, complementing the experimental data. A smaller ΔG°BET value pushes the reaction into the Marcus normal region (λ > |ΔG°|), with the exception of toluene, which increases the BET rates for the lower dielectric solvents. Thus, in the case of the forward ET reaction in the Marcus inverted region a faster rate is observed for higher dielectric solvents since λo is shifted closer to |ΔG°ET|, resulting in a low barrier for the ET transition.97 As for the BET reaction, in the normal region, due to the high λo associated with the higher dielectric solvents, this factor will dominate over ΔG°BET (in regard to ΔG‡BET); thus, more energy is required to reorganize this solvent. The polar solvent molecules reach an equilibrium with the product state and are now relatively more stabilized

In order to better understand the solvent trends with respect to the charge transfer reaction observed in the complex, we turn to Marcus theory to model our system.88−92 Marcus theory allows us to predict the energetics and kinetics of the photoinduced ET and BET reactions for the various solvents. For the purpose of obtaining driving force (ΔG°ET), initial and final states of the ET reaction need to be defined. As stated above, an exciton dissociation pathway involving a hole transfer from the QD to the Ru(bpy)32+ group following QD photoexcitation will be used based on interpretation of the QD−RuPDMAEMA PL decays. Values for the free energy difference of the ET reaction (ΔG°ET, ΔG°BET), reorganization energy (λ), barrier energies for ET and BET processes (ΔG‡ET, ΔG‡BET), as well as rate Table 3. Calculated Values for the Solvent Reorganization Energy (λo), Activation Energies for ET and BET Processes (ΔG‡ET, ΔG‡BET), and ET and BET Rate Constants (kET, kBET)a solvents

λo (eV)

ΔG‡ET (eV)

toluene THF DMF water

0.04 0.41 0.52 0.62

3.26 0.07 0.02 0.01

kET (1/s) (1.4 1.3 5.5 9.4

× × × ×

10−41) 1013 1013 1013

ΔG‡BET (eV) 0.75 0.31 0.32 0.34

kBET (1/s) (2.8 9.3 4 1.7

× × × ×

102) 108 108 108

a

ET and BET activation energies and rate constants are calculated using ΔG°ET = −0.7453 eV and ΔG°BET = 0.30 eV.

constants (kET, kBET) have been calculated (Table 3) using eqs 6−10 ◦ ◦ ◦ ΔG ET = e(EOX − E RED ) − E00

λo =

(6)

⎡ ⎛1 1 1 1⎞ e2 ⎜ × ⎢ − ⎟ 2 ⎝ ⎢ + 2 4πεo 2 ⎣ rRu(bpy)3 n ε⎠ −

2 2 ⎤ εQD − ε 1 ⎛⎜ nQD − n 1 1 ⎟⎞⎥ × − × 2 2R ⎜⎝ nQD εQD + ε ε ⎟⎠⎥⎦ n2 + n2

‡ ΔG ET =

◦ (ΔG ET + λ)2 4λ

(8)

‡ ◦ ‡ ◦ ‡ ΔG BET = ΔG BET + ΔG ET = ( −ΔG ET ) + ΔG ET

2π kET(BET) = |HDA|2 ℏ

(7)

⎛ −ΔG ‡ ⎞ 1 ET(BET) ⎟ exp⎜⎜ ⎟ kBT 4πλkBT ⎝ ⎠

(9)

(10)

where E°OX and E°RED are the oxidation and reduction energy of the electron donor and acceptor pair and E00 is the excitation energy put into the system to form the initial photoexcited state of the QD which is equal to the energy from the absorption edge of the first exciton band. R and rRu(bpy)32+ represent the donor−acceptor separation distance and the radius of the molecular adsorbate (0.58 nm for Ru(bpy)32+),93 respectively. The refractive index (n) and static dielectric constants (ε) refer to the respective solvents used, unless explicitly specified via subscript (i.e., QD). |HDA|, the electronic coupling between the initial and final states, is on the order of 0.05 eV.94 The ET process was modeled using nonadiabatic Marcus theory for weak donor−acceptor coupling.91,95,96 The solvent 18881



Article

passivation, exposing surface trap states and ultimately influencing the PL lifetime of the complex. In addition, with the additional ET and BET reactions, the overall excited state or more specifically radiative lifetime of the QD could be lengthened and also greatly affected by the polarity of the solvent. Marcus theory allowed us to put our results into perspective, illustrating the high complexity of the QD− RuPDMAEMA system and providing general insight on QDbased CT systems.

compared to the reactant state; thus, the system resists BET, and sufficient energy is required to overcome this stability. This product stabilization effect becomes more pronounced as the solvent polarity increases, and slower BET rates result for the higher dielectric solvents. The model is not a perfect fit, and there are still many significant factors that have been overlooked (i.e., conformation of adsorbate, Coulombic attraction, and solvation factors in the calculation of ΔG°), along with a margin of error associated with accurately predicting ΔG° due to ambiguities in the actual energy levels and reorganization processes of the intrinsic surface states. However, the general model manages to explain some of the basic trends that we observed in our experimental data. This solvation experiment allowed us to demonstrate the complete ligand exchange and association of the PDMAEMA or RuPDMAEMA polymer with the QD through the solubility of the complex in an extensive range of dielectric solvents, which were not possible with the TOPO-passivated QD. This, in turn, allowed us to explore the changes in surface passivation of the QD with respect to the various solvents through PL lifetime. Through the use of the electroactive polymer, we were also able to see that forward and backward charge transfer processes, in combination with trap states, can alter the ideal kinetic scheme of QD photophysics, introducing intricate, alternative pathways, which can extend the overall excited-state lifetime of the QD, especially with respect to various solvent polarities. Since QD-based CT systems are not normally soluble and have not been studied in such an extensive range of dielectric solvents, this gave us a unique opportunity to examine the degree of influence different dielectric media has on such systems.

■

ASSOCIATED CONTENT

* Supporting Information S

Preparation of QD−RuPDMAEMA complex; STEM images of QD−RuPDMAEMA; UV−vis absorption and fluorescence of QD−RuPDMAEMA and QD−PDMAEMA in various solvents; relative redox potential energy diagram of CdSe QD and Ru(bpy)32+. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the Natural Sciences and Engineering Research Council (NSERC) of Canada for financial support.

■

6. CONCLUSION This article introduced the properties of CdSe nanocrystal complexes passivated with an electroactive block copolymer, RuPDMAEMA. Synthesis of the RuPDMAEMA polymer and preparation of QD−RuPDMAEMA complexes, via a simple ligand-exchange procedure, were presented. QD−polymer complex characterization was constructed using STEM images, steady-state UV−vis absorption and fluorescence spectroscopy, and transient absorption measurements. It was found that adsorption of the RuPDMAEMA polymer onto the QD surface produced PL quenching in the complex. It was deduced that the quenching was most likely due to a charge transfer reaction between the QD and the polymer. TA measurements lead to fairly complex interpretations with mixed results. Through some refinement, the relative energy level positions of the QD and RuPDMAEMA were calculated to reflect upon which photoinduced processes of the QD−RuPDMAEMA complex were available. Solvent polarity effects on the QD−RuPDMAEMA system were explored using nanosecond TSCPC measurements. First, we were able to verify the adsorption of the RuPDMAEMA polymer onto the QD surface through the QD solubilization in a wide range of solvents. QD−RuPDMAEMA PL decays showed the presence of two decay components, fast and slow. It was suggested that the fast component may have had contributions from surface trap and ET effects found in the system, while the slow component may have been a result of BET, resulting from the initial hole transfer from the QD to the RuPDMAEMA, a possible CT mechanism for the QD− RuPDMAEMA system. From this experiment it was found that solvent polarity played an important role on QD surface

REFERENCES

(1) Miller, R. J. D.; McLendon, G. L.; Nozik, A. J.; Schmickler, W.; Willig, F. Surface Electron Transfer Processes; John Wiley & Sons: New York, 1995. (2) Michael Grätzel Festschrift, A Tribute for His 60th Birthday, Dye Sensitized Solar Cells. Coord. Chem. Rev. 2004, 248, 1195−1213. (3) Robel, I.; Subramanian, V.; Kuno, M.; Kamat, P. Quantum Dot Solar Cells. Harvesting Light Energy with CdSe Nanocrystals Molecularly Linked to Mesoscopic TiO2 Films. J. Am. Chem. Soc. 2006, 128, 2385−2393. (4) Robel, I.; Kuno, M.; Kamat, P. Size-Dependent Electron Injection from Excited CdSe Quantum Dots into TiO2 Nanoparticles. J. Am. Chem. Soc. 2007, 129, 4136−4137. (5) Kongkanand, A.; Tvrdy, K.; Takechi, K.; Kuno, M.; Kamat, P. Quantum Dot Solar Cells. Tuning Photoresponse Through Size and Shape Control of CdSe−TiO2 Architecture. J. Am. Chem. Soc. 2008, 130, 4007−4015. (6) Boulesbaa, A.; Issac, A.; Stockwell, D.; Huang, Z.; Huang, J.; Guo, J.; Lian, T. Ultrafast Charge Separation at CdS Quantum Dot/ Rhodamine B Molecule Interface. J. Am. Chem. Soc. 2007, 129, 15132−15133. (7) She, C.; Guo, J.; Lian, T. Comparison of Electron Injection Dynamics from Re-bipyridyl Complexes to TiO2 Nanocrystalline Thin Films in Different Solvent Environments. J. Phys. Chem. B 2007, 111, 6903−6912. (8) Huang, J.; Stockwell, D.; Huang, Z.; Mohler, D.; Lian, T. Photoinduced Ultrafast Electron Transfer From CdSe Quantum Dots to Re-bipyridyl Complexes. J. Am. Chem. Soc. 2008, 130, 5632−5633. (9) Issac, A.; Jin, S.; Lian, T. Intermittent Electron Transfer Activity From Single CdSe/ZnS Quantum Dots. J. Am. Chem. Soc. 2008, 130, 11280−11281. (10) Jin, S.; Lian, T. Electron Transfer Dynamics From Single CdSe/ ZnS Quantum Dots to TiO2 Nanoparticles. Nano Lett. 2009, 9, 2448− 2454. 18882



Article

(11) Ginger, D. S.; Greenham, N. C. Photoinduced Electron Transfer From Conjugated Polymers to CdSe Nanocrystals. Phys. Rev. B 1999, 59, 10622−10629. (12) O’Regan, B.; Grätzel, M. A Low-Cost, High-Efficiiency Solar Cell Based on Dye-Sensitized Colloidal TiO2 Films. Nature 1991, 353, 737−740. (13) Nazeeruddin, M.; Kay, A.; Rodicio, I.; Humphrey-Baker, R.; Müller, E.; Liska, P.; Vlachopoulos, N.; Grätzel, M. Conversion of Light to Electricity by Cis-X2bis(2,2′-bipyridyl-4,4′-dicarboxylate)ruthenium(II) Charge-Transfer Sensitizers (X = Cl-, Br-, I-, CN-, and SCN-) on Nanocrystalline Titanium Dioxide Electrodes. J. Am. Chem. Soc. 1993, 115, 6382−6390. (14) Cahen, D.; Hodes, G.; Grätzel, M.; Guillemoles, J.; Riess, I. Nature of Photovoltaic Action in Dye-Sensitized Solar Cells. J. Phys Chem. B 2000, 104, 2053−2059. (15) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Hybrid NanorodPolymer Solar Cells. Science 2002, 295, 2425−2427. (16) Coe, S.; Woo, W.-K.; Bawendi, M.; Bulović, V. Electroluminescence From Single Monolayers of Nanocrystals in Molecular Organic Devices. Nature 2002, 420, 800−803. (17) Grätzel, M. Photochemical Cells. Nature 2001, 414, 338−344. (18) Murray, C. B.; Norris, D. J.; Bawendi, M. G. Synthesis and Characterization of Nearly Monodisperse CdE (E = Sulfur, Selenium, Tellurium) Semiconductor Nanocrystallites. J. Am. Chem. Soc. 1993, 115, 8706−8715. (19) Hines, M. A.; Guyot-Sionnest, P. Synthesis and Characterization of Strongly Luminescing ZnS-Capped CdSe Nanocrystals. J. Phys. Chem. 1996, 100, 468−471. (20) Scholes, G. D. Controlling the Optical Properties of Inorganic Nanoparticles. Adv. Funct. Mater. 2008, 18, 1157−1172. (21) Klimov, V. I. Spectral and Dynamical Properties of Multilexcitons in Semiconductor Nanocrystals. Annu. Rev. Phys. Chem. 2007, 58, 635−673. (22) Efros, A. L.; Rosen, M. The Electronic Structure of Semiconductor Nanocrystals. Annu. Rev. Mater. Sci. 2000, 30, 475− 521. (23) Norris, D. J.; Efros, A. L.; Rosen, M.; Bawendi, M. G. SizeDependence of Exciton Fine Structure in CdSe Quantum Dots. Phys. Rev. B 1996, 53, 16347−16354. (24) Nirmal, M.; Murray, C. B.; Norris, D. J.; Bawendi, M. G. Surface Electronic Properties of CdSe Nanocrystallites. Z. Phys. D 1993, 26, 361−363. (25) Woggon, U.; Gaponenko, S. V. Excitons in Quantum Dots. Phys. Status Solidi B 1995, 189, 285−343. (26) Alivisatos, A. P. Perspectives on the Physical Chemistry of Semiconductor Nanocrystals. J. Phys. Chem. 1996, 100, 13226−13239. (27) Burda, C.; Chen, X. B.; Narayanan, R.; El-Sayed, M. A. Chemistry and Properties of Nanocrystals of Different Shapes. Chem. Rev. 2005, 105, 1025−1102. (28) Sykora, M.; Petruska, M.; Alstrum-Acevedo, J.; Bezel, I.; Meyer, T.; Klimov, V. Photoinduced Charge Transfer Between CdSe Nanocrystal Quantum Dots and Ru-Polypyridine Complexes. J. Am. Chem. Soc. 2006, 128, 9984−9985. (29) Koposov, A. Y.; Szymanski, P.; Cardolaccia, T.; Meyer, T. J.; Klimov, V. I.; Sykora, M. Electronic Properties and Structure of Assemblies of CdSe Nanocrystal Quantum Dots and Ru-Polypyridine Complexes Probed by Steady State and Time-Resolved Photoluminescence. Adv. Funct. Mater. 2011, 21, 3159−3168. (30) Zaban, A.; Mićić, O.; Gregg, B.; Nozik, A. Photosensitization of Nanoporous TiO2 Electrodes with InP Quantum Dots. Langmuir. 1998, 14, 3153−3156. (31) Plass, R.; Pelet, S.; Kruegar, J.; Grätzel, M. Quantum Dots Sensitization of Organic-Inorganic Hybrid Solar Cells. J. Phys. Chem. B 2002, 106, 7578−7580. (32) Hyun, B.-R.; Bartnik, A.; Lee, J.-K.; Imoto, H.; Sun, L.; Choi, J.; Chujo, Y.; Hanrath, T.; Ober, C.; Wise, F. Role of Solvent Dielectric Properties on Charge Transfer From PbS Nanocrystals to Molecules. Nano Lett. 2010, 10, 318−323.

(33) Cui, S.-C.; Tachikawa, T.; Fujitsuka, M.; Majima, T. SolventPolarity Dependence of Electron-Transfer Kinetics in a CdSe/ZnS Quantum Dot-Pyromellitimide Conjugate. J. Phys. Chem. C 2010, 114, 1217−1225. (34) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. Handbook of Photochemistry, 3rd ed.; CRC Press: Boca Raton, 2006. (35) Wang, X.-S.; Dykstra, T. E.; Salvador, M. R.; Manners, I.; Scholes, G. D.; Winnik, M. A. Surface Passivation of Luminescent Colloidal Quantum Dots with Poly(dimethylaminoethyl methacrylate) Through a Ligand Exchange Process. J. Am. Chem. Soc. 2004, 126, 7784−7785. (36) Wang, M.; Oh, J. K.; Dykstra, T. E.; Lou, X.; Scholes, G. D.; Winnik, M. A. Surface Modification of CdSe and CdSe/ZnS Semiconductor Nanocrystals with Poly(N,N-dimethylaminoethyl methacrylate). Macromoecules 2006, 39, 3664−3672. (37) Wang, M.; Dykstra, T. E.; Lou, X.; Salvador, M. R.; Scholes, G. D.; Winnik, M. A. Colloidal CdSe Nanocrystals Passivated by a DyeLabeled Multidentate Polymer: Quantitative Analysis by SizeExclusion Chromatography. Angew. Chem. 2006, 118, 2279−2282. (38) Chen, M.; Ghiggino, K. P.; Launikonis, A.; Mau, A. W. H.; Rizzardo, E.; Sasse, W. H. F.; Thang, S. H.; Wilson, G. J. RAFT Synthesis of Linear and Star-Shaped Light Harvesting Polymers Using Di- and Hexafunctional Ruthenium Polypyridine Reagents. J. Mater. Chem. 2003, 13, 2696−2700. (39) Perrier, S.; Takolpuckdee, P.; Mars, C. A. Reversible Addition− Fragmentation Chain Transfer Polymerization: End Group Modification for Functionalized Polymers and Chain Transfer Agent Recovery. Macromolecules 2005, 38, 2033−2036. (40) Alivisatos, A. P. Perspectives on the Physical Chemistry of Semiconductor Nanocrystals. J. Phys. Chem. 1996, 100, 13226−13239. (41) Katari, J.; Colvin, V.; Alivisatos, A. X-ray Photoelectron Spectroscopy of CdSe Nanocrystals with Applications to Studies of the Nanocrystal Surface. J. Phys. Chem. 1996, 100, 13226−13239. (42) Aldana, J.; Wang, Y. A.; Peng, X. Photochemical Instability of CdSe Nanocrystals Coated by Hydrophilic Thiols. J. Am. Chem. Soc. 2001, 123, 8844−8820. (43) Howe, L.; Zhang, J. Z. Ultrafast Studies of Excited-State Dynamics of Phthalocyanine and Zinc Phthalocyanine Tetrasulfonate in Solution. J. Phys. Chem. A. 1997, 101, 3207−3213. (44) Campagna, S.; Puntoriero, F.; Nastasi, F.; Bergamini, G.; Balzani, V. Photochemistry and Photophysics of Coordination Compounds: Ruthenium. Top. Curr. Chem. 2007, 280, 117−214. (45) Damrauer, N. H.; Cerullo, G.; Yeh, A. T.; Boussie, T. R.; Shank, C. V.; McCusker, J. K. Femtosecond Dynamics of Excited-State Evolution in [Ru(bpy)3]2+. Science 1997, 275, 54−57. (46) Yeh, A. T.; Shank, C. V.; McCusker, J. K. Ultrafast Electron Localization Dynamics Following Photoinduced Charge Transfer. Science 2000, 289, 935−938. (47) Cannizzo, A.; van Nourik, F.; Gawelda, W.; Zgrablic, G.; Bressler, C.; Chergui, M. Broadband Femtosecond Fluorescence Spectroscopy of [Ru(bpy)3]2+. Angew. Chem., Int. Ed. 2006, 45, 3174− 3176. (48) Webb, M. A.; Knorr, F. J.; McHale, J. L. Resonance Raman Spectrum of [Ru(bipyridine)3]2+ in Water, Acetonitrile and Their Deuterated Derivatives: The Possible Role of Solvent in Excited-State Charge Localization. J. Raman Spectrosc. 2001, 32, 481−485. (49) Wallin, S.; Davidsson, J.; Modin, J.; Hammarströ m, L. Femtosecond Transient Absorption Anisotropy Study on [Ru(bpy)3]2+ and [Ru(bpy)(py)4]2+. Ultrafast Interligand Randomization of the MLCT State. J. Phys. Chem. A 2005, 109, 4697−4704. (50) Yoshimura, A.; Hoffman, M. Z.; Sun, H. An Evaluation of the Excited State Absorption Spectrum of Ru(bpy)32+ in Aqueous and Acetonitrile Solutions. J. Photochem. Photobiol., A: Chem. 1993, 70, 29−33. (51) Damrauer, N. H.; Cerullo, G.; Yeh, A.; Boussie, T. R.; Shank, C. V.; McCusker, J. K. Femtosecond Dynamics of Excited-State Evolution in [Ru(bpy)3]2+. Science 1997, 275, 54−57. 18883



Article

(52) McCusker, J. K. Femtosecond Absorption Spectroscopy of Transition Metal Charge-Transfer Complexes. Acc. Chem. Res. 2003, 36, 876−887. (53) Müller, P.; Brettel, K. [Ru(bpy)3]2+ as a Reference in Transiet Absorption Spectroscopy: Differential Absorption Coefficients for Formation of Long-Lived 3MLCT Excited State. Photochem. Photobiol. Sci. 2012, 11, 632−636. (54) Tarnovsky, A. N.; Gawelda, W.; Johnson, M.; Bressler, C.; Chergui, M. Photoexcitation of Aqueous Ruthenium(II)-tris-(2,2′bipyridine) with High-Intensity Femtosecond Laser Pulses. J. Phys. Chem. B 2006, 110, 26497−26505. (55) Lachish, U.; Infelta, P. P.; Grätzel, M. Optical Absorption Spectrum of Excited Ruthenium Tris-bipyridyl (Ru(bpy)32+). Chem. Phys. Lett. 1979, 62, 317−319. (56) Schaller, R. D.; Sykora, M.; Jeong, S.; Klimov, V. I. HighEfficiency Carrier Multiplication and Ultrafast Charge Separation in Semiconductor Nanocrystals Studied Via Time-Resolved Photoluminescence. J. Phys. Chem. B 2006, 110, 25332−25338. (57) Knowles, K. E.; McArthur, E. A.; Weiss, E. A. A Multi-Timescale Map of Radiative and Non-Radiative Decay Pathways for Excitons in CdSe Quantum Dots. ACS Nano 2011, 5, 2026−2035. (58) Rajeshwar, K. Fundamentals of Semiconductor Electrochemistry and Photoelectrochemistry. In Encyclopedia of Electrochemistry; Bard, A., Stratmann, M., Licht, S., Eds.; Wiley-VCH: Weinheim, 2007; Vol. 6 (Semiconductor Electrodes and Photoelectrochemistry). (59) In Properties of Narrow Gap Cadmium-Dased Compounds; Capper, P., Ed.; INSPEC: London, 1994. (60) In Handbook of Electrochemistry; Zoski, C. G., Ed.; Elsevier Science Publishers: Amsterdam, 2007. (61) Marini, A.; Muñoz-Losa, A.; Biancardi, A.; Mennucci, B. What is Solvatochromism? J. Phys. Chem. B 2010, 114, 17128−17135. (62) Oh, M. H. J.; Salvador, M. R.; Wong, C. Y.; Scholes, G. D. Three-Pulse Photon-Echo Peak Shift Spectroscopy and Its Application for the Study of Solvation and Nanoscale Excitons. ChemPhysChem. 2011, 12, 88−100. (63) Ooshika, Y. Absorption Spectra of Dyes in Solution. J. Phys. Soc. Jpn. 1954, 9, 594−602. (64) McRae, E. G. Theory of Solvent Effects on Molecular Electronic Spectra. Frequency Shifts. J. Phys. Chem. 1957, 61, 562−572. (65) Mataga, N.; Kaifu, Y.; Koizumi, M. Solvent Effects Upon Fluorescence Spectra and the Dipole Moments of Excited Molecules. Bull. Chem. Soc. Jpn. 1956, 29, 465−470. (66) Mataga, N. Solvent Effects on the Absorption and Fluorescence Spectra of Naphthylamines and Isomeric Aminobenzoic Acids. Bull. Chem. Soc. Jpn. 1963, 36, 654−662. (67) Lippert, E. Dipolmoment und Elektronstrukturen von Angeregten Molekülen. Z. Naturforsch. A 1955, 10, 541−545. (68) Suppan, P. The Influence of the Medium on the Energy of Electronic States. J. Photochem. Photobiol. A 1990, 50, 293−330. (69) Tomasi, J.; Persico, M. Molecular Interactions in Solution: An Overview of Methods Based on Continuous Distributions of the Solvent. Chem. Rev. 1994, 94, 2027−2094. (70) Tomasi, J.; Mennucci, B.; Cammi, R. Quantum Mechanical Continuum Solvation Models. Chem. Rev. 2005, 105, 2999−3094. (71) Cramer, C. J.; Truhlar, D. G. Implicit Solvation Models: Equilbiria, Structure, Spectra, and Dynamics. Chem. Rev. 1999, 99, 2161−2200. (72) Reichardt, C. Solvatochromic Dyes as Solvent Polarity Indicators. Chem. Rev. 1994, 94, 2319−2358. (73) Bagchi, B.; Oxtoby, D. W.; Fleming, G. R. Theory of the Time Development of the Stokes Shift in Polar Media. Chem. Phys. 1984, 86, 257−267. (74) Maroncelli, M.; Fleming, G. R. Comparison of Time-Resolved Fluorescence Stokes Shift Measurements to a Molecular Theory of Solvation Dynamics. J. Chem. Phys. 1988, 89, 875−881. (75) Shim, M.; Guyot-Sionnest, P. Permanent Dipole Moment and Charges in Colloidal Semiconductor Quantum Dots. J. Chem. Phys. 1999, 111, 6955−6964.

(76) Rabani, E.; Hetényi, B.; Berne, B.; Brus, L. E. Electronic Properties of CdSe Nanocrystals in the Absence and Presence of a Dielectric Medium. J. Chem. Phys. 1999, 110, 5355−5369. (77) Colvin, V. L.; Alivisatos, A. P. CdSe Nanocrystals with a Dipole Moment in the First Excited State. J. Chem. Phys. 1992, 97, 730−733. (78) Fisher, B. R.; Eisler, H.-J.; Stott, N. E.; Bawendi, M. G. Emission Intensity Dependence and Single-Exponential Behavior in Single Colloidal Quantum Dot Fluorescence Lifetimes. J. Phys. Chem. B 2004, 108, 143−148. (79) Jones, M.; Lo, S. S.; Scholes, G. D. Quantitative Modeling of the Role of Surface Traps in CdSe/CdS/ZnS Nanocrystal Photoluminescence Decay Dynamics. Proc. Natl. Acad. Sci. 2009, 106, 3011−3016. (80) Klimov, V. I.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M. G. Electron and Hole Relaxation Pathways in Semiconductor Quantum Dots. Phys. Rev. B 1999, 60, 13740−13749. (81) Klimov, V. I. Optical Nonlinearities and Ultrafast Carrier Dynamics in Semiconductor Nanocrystals. J. Phys. Chem. B 2000, 104, 6112−6123. (82) Efros, A. L.; Rosen, M. Random Telegraph Signal in the Photoluminescence Intensity of a Single Quantum Dot. Phys. Rev. Lett. 1997, 78, 1110−1113. (83) Ispasoiu., R. G.; Lee, J.; Papadimitrakopoulos, F.; Goodson, T., III. Surface Effects in the Fluorescence Ultra-Fast Dynamics from CdSe Nano-Crystals. Chem. Phys. Lett. 2001, 340, 7−12. (84) Frantsuzov, P. A.; Marcus, R. A. Explanation of Quantum Dot Blinking Without the Long-Lived Trap Hypothesis. Phys. Rev. B 2005, 72, 155321. (85) Bullen, C.; Mulvaney, P. The Effect of Chemisorption on the Luminescence of CdSe Quantum Dots. Langmuir 2006, 22, 3007− 3013. (86) Kalyuzhny, G.; Murray, R. W. Ligand Effects on Optical Properties of CdSe Nanocrystals. J. Phys. Chem. B 2005, 109, 7012− 7021. (87) Liu, I.-S.; Lo, H.-H.; Chien, C.-T.; Lin, Y.-Y.; Chen, C.-W.; Chen, Y.-F.; Su, W.-F.; Liou, S.-C. Enhancing Photoluminescence Quenching and Photoelectric Properties of CdSe Quantum Dots with Hole Accepting Ligands. J. Mater. Chem. 2008, 18, 675−682. (88) Marcus, R. A. Electron Transfer Reactions in Chemistry: Theory and Experiment (Nobel Lecture). Angew. Chem., Int. Ed. Engl. 1993, 32, 1111−1121. (89) Kavarnos, G. J. Fundamentals of Photoinduced Electron Transfer; John Wiley & Sons: New York, 1993. (90) In Photoinduced Electron Transfer; Fox, M. A., Chanon, M., Eds.; Elsevier Science Publishers: Amsterdam, 1988. (91) Paddon-Row, M. Electron and Energy Transfer. In Stimulating Concepts in Chemistry; Vögtle, F., Stoddart, J. F., Shibasaki, M., Eds.; Wiley-VCH: Weinheim, 2000. (92) Sutin, N. Electron Transfer Reaction in Solution: A Historical Perspective. In Advances in Chemical Physics; Jortner, J., Bixon, M., Eds.; John Wiley & Sons: New York, 1999; Vol. 107, Part 1. (93) Otsuka, T.; Iwamura, M.; Kaizu, Y. Effect of Ligand Congeniality on Energy Transfer Reaction Between Photo-Excited Tris(bipyridine)ruthenium(II) and Chromate(III) Complexes in Aqueous Solutions. Inorg. Chim. Acta 2006, 359, 1351−1356. (94) Scholes, G. D.; Jones, M.; Kumar, S. Energetics of Photoinduced Electron-Transfer Reactions Decided by Quantum Confinement. J. Phys. Chem. C 2007, 111, 13777−13785. (95) Vos, J. G.; Forster, R. J.; Keyes, T. E. Theoretical Framework for Electrochemical and Optical Processes. In Interfacial Supramolecular Assemblies; John Wiley & Sons: Chichester, UK, 2003. (96) Walker, G. C.; Bertan, D. N. Electron transfer reactions. In Encyclopedia of Chemical Physics and Physical Chemistry; Moore, J. H., Spencer, N. D., Eds.; Institute of Physics: Bristol, 2001; Vol. 3 (Applications). (97) Suppan, P. The Marcus Inverted Region. Top. Curr. Chem. 1992, 163, 95−130.

18884


Charge Transfer in CdSe Nanocrystal Complexes with an

Recommend Documents