photophysical properties of cdse/cds core/shell quantum ...s 2p 3/2 and 1/2 (approximately 162 and...

Chemical Physics 471 (2016) 24–31

Contents lists available at ScienceDirect

Chemical Physics

journal homepage: www.elsevier .com/locate /chemphys

Photophysical Properties of CdSe/CdS core/shell quantum dots withtunable surface composition

http://dx.doi.org/10.1016/j.chemphys.2015.09.0100301-0104/� 2015 Elsevier B.V. All rights reserved.

⇑ Corresponding author at: Department of Chemistry, University of Rochester,Rochester, NY 14627, United States.

E-mail address: [email protected] (T.D. Krauss).

Kelly L. Sowers a, Zhentao Hou a, Jeffrey J. Peterson a, Brett Swartz a, Sougata Pal c, Oleg Prezhdo c,Todd D. Krauss a,b,⇑aDepartment of Chemistry, University of Rochester, Rochester, NY 14627, United Statesb Institute of Optics, University of Rochester, Rochester, NY 14627, United StatescDepartment of Chemistry, University of Southern California, Los Angeles, CA 90089, United States

a r t i c l e i n f o a b s t r a c t

Article history:Available online 28 September 2015

Keywords:NanocrystalsSurface chemistryNonradiative relaxationPhotoluminescenceTime-resolvedSynthesis

We report the synthesis and optical characterization of core/shell CdSe/CdS quantum dots (QDs) withcontrolled surface composition. Using secondary phosphine chalcogenide and cadmium carboxylate pre-cursors in an alternating layer-by-layer synthetic approach, the elemental surface composition of thequasi-type-I CdSe/CdS core/shell QDs can be repeatedly tuned from predominately cadmium to predom-inantly sulfur, as measured by X-ray photoelectron spectroscopy (XPS). Similar to CdS and CdSe core-onlyQDs, the surface composition has a significant effect on the photoluminescence (PL) quantum yield: sul-fur terminated QDs exhibit quenched PL, while cadmium terminated QDs have relatively bright PL.Density-functional tight-binding calculations on CdSe/CdS core/shell clusters suggest that PL quenchingfor sulfur-rich surfaces is the result of a high density of hole surface states in the QD bandgap. Time-re-solved PL measurements confirm the QDs’ nonradiative recombination rates are strongly sensitive to thesurface composition.

� 2015 Elsevier B.V. All rights reserved.

1. Introduction

Colloidal semiconductor nanocrystals, commonly known asquantum dots (QDs), have been the source of much study overthe past two decades, owing to a wide range of customizable fea-tures including size-dependent optical properties [1–3]. Controlledmodifications of surface composition and surface passivation canbe used to tune QD photophysical properties, such as photolumi-nescence energy and intensity [4–7]. Since particle size, composi-tion, and surface passivation can be controlled in the initial QDsynthesis reaction [8] and post-synthetic workup, QDs can poten-tially be useful in a wide range of applications such as light har-vesting for solar fuels [9,10] and solar cells [11,12], displays andLEDs [13,14], and biological imaging [15], labeling [16], and cellstimulation [17].

Due to their large surface to volume ratio, the surface of QDs hasbeen historically recognized as integral in influencing the opticalproperties of colloidal QDs [6,18–26]. For example, early syntheticbreakthroughs that produced highly-fluorescent QDs wereenabled, in part, by the discovery that phosphine ligands provided

effective passivation of dangling bonds at the nanoparticle surface[27]. Subsequent development of core/shell structures (e.g. CdSe/ZnS), dramatically enhanced QD photoluminescence (PL), PL quan-tum yield (QY), and particle photostability [24,28]. Such efforts sig-nificantly advanced the potential for QDs in various technologicalapplications and helped enable new levels of understanding inQD photophysics. Current research seeks even greater tunabilityand optimization of QD optical properties through surface engi-neering [29,30], such as synthesis of ‘‘giant” shells [31,32] andalloyed QD cores and shells [33–35] to minimize single particlePL blinking. Photophysical properties can be tuned through wavefunction engineering provided by type-I and type-II heterostruc-tures [36–38], and metallo-ligands to tune PL over broad spectralregions [39,40].

One historical challenge in engineering the QD surface hasbeen achieving precise compositional control of the surfacechemistry. For metal chalcogenide QDs, such control is difficultdue to a limited understanding of the specific molecular reactivespecies of metal and chalcogenide precursors, and the mecha-nisms involved in QD nucleation and growth. However, recentreports suggest that by employing secondary phosphine sulfideprecursors in colloidal QD synthesis (such as diphenyl phosphinesulfide (DPPS)) [41,42] the surface composition can be tuned

http://crossmark.crossref.org/dialog/?doi=10.1016/j.chemphys.2015.09.010&domain=pdfhttp://dx.doi.org/10.1016/j.chemphys.2015.09.010mailto:[email protected]://dx.doi.org/10.1016/j.chemphys.2015.09.010http://www.sciencedirect.com/science/journal/03010104http://www.elsevier.com/locate/chemphys

K.L. Sowers et al. / Chemical Physics 471 (2016) 24–31 25

from all metal to all sulfur. Furthermore, the composition of aQD surface can also have dramatic effects on the QD opticalproperties; for core-only CdSe or CdS [43] QDs, when the surfaceis cadmium terminated the QD exhibits bright PL, however whenthe surface is S- or Se-terminated the PL is quenched. Theseeffects are believed to originate from mid-gap hole states thatare associated with surface S or Se atoms and that effectivelytrap photoexcited holes [43].

We sought to extend these studies to core/shell CdSe/CdS QDheterostructures and investigate the effects of surface compositionon their photophysical properties. In particular, such studies canprovide information about the relative extent of the electron andhole wavefunctions in these nanoparticles. The simplest thinkingsuggests that due to the relative energy band offsets for CdSeand CdS, CdSe/CdS core/shell QDs are expected to localize the holewavefunction in the core of the particle [44,45]. If holes are limitedin sampling the surface in a core/shell QD and the surface is theorigin of the PL quenching, it is possible that changing QD surfacechemistry will not significantly affect PL. Contrarily, the presenceof a significant number of hole energy states in the bandgap forCdS could drive the hole toward localization on the CdSe/CdS QDsurface, even though it would have to tunnel through a smallenergy barrier (�0.88 eV) [46] due to the CdS layer. In this case,the PL intensity should rise and fall with cadmium or sulfur QDtermination.

Here, we describe the synthesis of core/shell CdSe/CdS QDswith controlled surface composition using exclusively secondaryphosphine chalcogenide precursors. By adopting an alternatinglayer-by-layer synthetic approach [47], the surface compositionof the CdSe/CdS QD can be reversibly tuned from predominantlycadmium to predominantly sulfur [43], as verified with X-rayphotoelectron spectroscopy (XPS). Similar to previous studieson bare core QDs [43,48], the surface composition is shown tohave a significant effect on the PL QY. Sulfur terminated QDshave dramatically quenched PL and longer fluorescence lifetimes,while cadmium terminated QDs have significantly brighter PLand shorter fluorescence lifetimes. Density-functional tight-binding (DFTB) calculations on CdSe/CdS QDs suggest the pres-ence of a high density of hole surface states in the QD bandgapquenches PL, despite the quasi-type-I electronic heterostructurethought to effectively trap the hole in the core.

2. Methods

2.1. CdSe QD synthesis

CdSe QDs were synthesized using adaptations of an air-free,hot injection method as previously reported [49]. 1.0 mmol ofdiphenylphosphine selenide (DPPSe; 0.266 g) was dissolved in asmall volume of toluene (�1–3 mL) at 60 �C in an inert atmo-sphere. Concurrently, in a separate reaction flask, 3.0 mmol of0.2 M cadmium oleate (Cd(oleate)2; 15 mL) and a small volumeof 1-octadecene (ODE; 10 mL; degassed under vacuum overnightat room temperature) were heated to 270 �C. At temperature, theDPPSe/toluene mixture was rapidly injected and the temperaturewas dropped to 220 �C. Small aliquots (0.1–0.5 mL) wereremoved every minute and growth was monitored by UV–visiblespectrophotometry. QDs were allowed to continue growing at220 �C until the first exciton absorbance peak ceased shifting,approximately 2–5 min. When QD growth had ceased, the reac-tion was quenched by removing the flask from heat and apply-ing a cool air flow to the outside of the flask. The resulting QDproduct was kept under a positive nitrogen flow for the shellingreactions.

2.2. CdS/CdSe core/shell QD synthesis

CdS shells were grown using variations of a successive ioniclayer absorption and reaction (SILAR) method [47]. Volumes ofcadmium and sulfur precursors for each full monolayer were calcu-lated for a given diameter [50] and surface area of the core QD,accounting for increased surface area with each subsequent shell.As-prepared CdSe QD cores were kept in a flask under continuousnitrogen flow during the shelling procedure, at a temperature of235 �C. All shelling reactions began by introducing a layer of cad-mium to the CdSe cores through swift injection of Cd(oleate)2 at235 �C, and reacting for 15 min to produce cadmium terminatedCdSe-QDs. We denote such species as CdSe–Cd QDs. In the sameflask, a monolayer of sulfur was grown by promptly injecting eithera diphenylphosphine sulfide (DPPS) or di-isobutylphosphine sul-fide (DiBPS) solution, and reacting for 15 min to produce sulfur ter-minated CdSe–CdS (sample A), completing the first full shell.Subsequent injections were performed, alternating Cd and S until3 full shells were grown (samples B–E), followed by an additionalcadmium layer, producing QDs with the composition CdSe–CdS/CdS/CdS/Cd (sample F). Fractions of core–shell QDs were with-drawn (0.1–0.5 mL aliquots) at the end of each 15 min reactiontime (before the next injection) and diluted with hexanes. Sulfurterminated QDs synthesized using DPPS were observed to rapidlyagglomerate and precipitate in hexanes, whereas the same cad-mium terminated QD fractions were stable. In contrast, both sulfurand cadmium terminated QDs synthesized using DiBPS formed astable colloid in hexanes.

2.3. Optical characterization

Absorbance data was taken using a Perkin–Elmer Lambda 35UV/Vis spectrophotometer, and PL data were taken on a modularActon fluorometer system. Samples were dissolved in hexanesand placed in a 1 cm square quartz cuvette. Optical densities werecontrolled to be between 0.01 and 0.1.

2.4. Fraction preparation

In preparation for transmission electron microscopy (TEM) andXPS experiments, aliquots in hexanes were flocculated using a1:4:1:1:3 ratio of QDs (hexanes): acetone: ethanol: butanol:methanol. Samples were centrifuged, the supernatant discarded,and the resultant QD pellet was resuspended in hexanes.

2.5. Transmission electron microscopy

TEM images were taken with an FEI Tecnai F-20 field emissionelectron microscope, operated with an accelerating voltage of200 keV. Samples were prepared as above, then dropcast ontoultrathin carbon film grids for larger QDs, or silicon nitride gridsfor smaller QDs.

2.6. X-ray photoelectron spectroscopy (XPS)

XPS data were taken with Al Ka as the source of incident X-rays.QDs were purified as described for the TEM measurements above.Thick QD films were drop cast on O2 plasma-cleaned silicon wafersfor XPS measurements. Gaussian functions were used to fit XPSdata individually, to determine relative signals from surface orinner atoms. Due to the unique electron binding energies of Cd3d 5/2 and 3/2 (approximately 405 and 412 eV respectively) andS 2p 3/2 and 1/2 (approximately 162 and 163 eV respectively)[51], the relative chemical composition of the QD can be deter-mined straightforwardly assuming a spherical QD with acore radius determined from the first exciton absorption [50].

26 K.L. Sowers et al. / Chemical Physics 471 (2016) 24–31

Subsequent shell radii can be extrapolated by the addition of unitcell dimensions of the CdS crystal [52] as confirmed by TEM anal-ysis (Fig. S2). The Cd-to-S ratios were obtained by XPS survey scansand corrected for atomic sensitivity factors [48,53,54]. If theamount of Cd or S atoms in a QD sample is known, based on parti-cle size the numbers of surface atoms for Cd or S can be determinedfrom the integrated elemental region signals from the surface ver-sus interior XPS peaks [43,48]. By comparing the total atoms of Cdor S to the percent of surface atoms, relative surface compositioncan be inferred.

2.7. PL lifetime measurements

A drop (approximately 6 lL) of the QD solution in 1-octadecenewas deposited onto a glass coverslip. A 488-nm excitation beamfrom a picosecond pulsed diode laser (PicoQuant PDL 800-D) fil-tered by a 482/18 nm band pass filter and circularly polarized bya quarter-wave plate was reflected by a 488-nm dichroic mirrorto a Nikon 10 � 0.30 N.A. Plan Fluor objective. The emissioncoming out of the microscope (Nikon TE-2000) passed through a100-lm pinhole and a 488-nm long pass filter, and imaged on anavalanche photodiode (Micro Photon Devices PDM Series 5CTB)by a pair of lenses. Photon statistics were acquired with a PicoHarp300 Time-Correlated Single Photon Counting (TCSPC) module toconstruct the fluorescence intensity trajectory over many cycles.Fluorescence lifetimes were determined by fitting the TCSPC datawith a double exponential function. Amplitude-weighted lifetimeswere calculated using the s ¼ P

iAisi, where s is the amplitude-

weighted lifetime, si are the lifetime components, and Ai are theircorresponding normalized coefficients.

Fig. 1. XPS elemental analysis of two core/shell samples: (a) CdSe/CdS/CdS/CdS (samplesignals are shown on the left and right, respectively. Black dots represent raw XPS data,corresponding to interior atoms, and red lines are Gaussian curve fits corresponding treferences to colour in this figure legend, the reader is referred to the web version of th

2.8. Computational methodology

To elucidate the role of composition on the PL intensity, we per-formed self-consistent charge density-functional tight-binding(SCC-DFTB) calculations. The SCC-DFTB method is based on a sec-ond-order expansion of the total Kohn–Sham energy in density-functional theory (DFT) with respect to charge density fluctuations.The one and two center matrix elements of the associated Hamil-tonian are calculated in an explicit minimal valence basis, wherethe repulsive energy part is approximated by universal pair poten-tials, which are fitted using the data of self-consistent field calcu-lations. The SCC-DFTB has a minimal number of input parametersby retaining the efficiency of TB approaches and has beendescribed in detail elsewhere [55–58]. This work uses the SCC-DFTB parameter set, developed recently and extensively testedfor cadmium-chalcogenide systems [59].

3. Results and discussion

3.1. X-ray photoelectron spectroscopy analysis of particle surfacetermination

X-ray photoelectron spectroscopy (XPS) data from a series ofCdSe/CdS core/shell QDs with alternating Cd or S surface passiva-tion is shown in Fig. 1. XPS is a widely-used technique for provid-ing information about the surface composition of solid statematerials [53,54], due to the slightly higher electron binding ener-gies of surface atoms compared to respective bulk values [60].Peaks assigned to interior atoms of Cd 3d 5/2 and 3/2 and S 2p3/2 and 1/2 electrons are observed to occur at 405 and 412 eVand 162 and 163 eV respectively [43,51]. Notably, for each interior

E), and (b) CdSe/CdS/CdS/CdS/Cd (sample F). XPS spectra corresponding to S and Cdblack lines are total summed Gaussian curve fits, blue lines are Gaussian curve fitso the higher-energy offsets attributed to surface atoms. (For interpretation of theis article.)

Table 1XPS compositional analysis of samples A–F.

Sample Cd (%) Se (%) S (%) Cd/S Corrected D (nm) Cds/Cdi Ss/Si Surface (%)

Cd S

A 5.15 0.18 4.01 1.28 1.10 3.0 0.21 2.42 21 79B 8.66 0.20 6.57 1.32 1.14 3.3 0.75 0.24 72 28C 9.04 0.26 8.35 1.08 0.93 3.7 0.16 2.14 16 84D 9.50 0.21 6.45 1.47 1.27 4.0 0.44 0.35 60 40E 16.61 0.38 16.72 0.99 0.85 4.4 0.43 2.28 27 73F 8.98 0.17 6.66 1.35 1.17 4.7 0.42 0.33 58 42

Fig. 2. (Main) Photoluminescence spectra of a set of shelled QDs, normalized toeach sample’s individual absorbance at the excitation wavelength to account forvariations in concentration. Samples are CdSe–Cd, A, B, C, D, E, and F as indicated inthe legend. In this particular example, CdSe–Cd had the largest QY, while for otherbatches sample B was found to be the brightest of the set. (Inset) Photograph of thesame QDs, of similar optical density, illuminated under UV lamp, demonstrating thevisibly quenched and subsequently restored fluorescence within a single reaction.


peak there exists a peak shifted approximately 1.5 eV to higherenergies due to contributions from surface atoms [60]. Generallyspeaking, the SILAR process worked as expected: QDs that had aterminal layer of S applied had a strong XPS signal from surface Satoms, and a minimal signal from surface Cd atoms (sample E(CdSe–CdS/CdS/CdS), as shown in Fig. 1(a). The opposite was alsoobserved, cadmium terminated surfaces showed strong XPS datafrom primarily surface Cd, and not S atoms (sample F (CdSe–CdS/CdS/CdS/Cd) as shown in Fig. 1(b).

Quantitative analysis of the XPS data yielded the relativeamount of cadmium or sulfur atoms on the surface of the QD asdescribed in Table 1. The XPS data reveals that surfaces nominallyterminated with sulfur have a high percentage of surface sulfuratoms (>70% S for all cases). While the Cd terminated surfaces doshow a preponderance of surface Cd atoms (60–70% Cd), the frac-tion is not nearly as great as for the sulfur termination. This latterresult is in contrast to small, similarly prepared CdS QDs (i.e. usingsecondary phosphine sulfur and Cd-carboxylate precursors), whichshowed greater than 90% of the surface atoms are Cd for a cad-mium terminated CdS QD [43]. One source of error that could pos-sibly account for the difference in the XPS data may be due to thesize of the QD. For CdSe/CdS QDs, Cd atoms are located throughoutthe QD down to the very core. The attenuation of the XPS signalsfor interior atoms, and variable escape depths for different typesof atoms are accounted for through an atomic sensitivity factorof the instrument, and by additional calibration to account forthe differences in electron mean free paths of Cd and S [61,62].For these experiments, the inelastic mean free paths (k) were cal-culated to be �19.8 Å for Cd, and �23.3 Å for S. Thus, in the anal-ysis for larger QDs (i.e. the ones with thicker shells where the Cdsurface percentage seems ‘‘low”) the interior Cd atom XPS signalmay not accurately reflect the true amount of Cd present. Anotherpossible explanation is that the Cd or S precursors, although addedsequentially, do not react to completion and addition of the otherelement increases the number of stoichiometric CdS units on thesurface. Finally, the increasingly non-spherical particle morphol-ogy with increasing shell thickness (Fig. S2) will also contributeto uncertainties in the quantitative analysis of the surface atompercentage. Particle size analysis of TEM images shown in Fig. S2shows agreement with predicted average particle size as deter-mined by core size and CdS lattice parameters. While shelledQDs do demonstrate less uniform shapes as the shell thicknessincreases, this non-uniformity will result in only small perturba-tions to PL energies or linewidths, as quantum confinement ofthe charges in the CdSe core remains largely unaffected for suffi-ciently thick CdS shells.

3.2. Effects of surface composition on photoluminescence

The specific surface composition of CdSe/CdS core/shell QDs hasa significant impact on the photoluminescence quantum yield(QY), as shown in Fig. 2. When the QD is cadmium-terminated,strong PL is observed; when the QD is sulfur-terminated, the PLis quenched (Table S1). The quenching of PL with sulfur is a rever-

sible effect: application of Cd to a sulfur terminated surface resultsin fluorescence recovery, while subsequent application of S to thecadmium terminated surface quenches the PL (Fig. 2). Controlexperiments showed that the quenched PL from the sulfur-terminated QDs is not due to a visible degradation of the structuralor optical absorbance properties of the QDs, as evidenced by absor-bance (Fig. S1) and TEM (Fig. S2) data. Additionally, this reversiblePL quenching effect does not depend on the specific identity of sec-ondary phosphine sulfide: di-isobutyl phosphine sulfide (DibPS)and diphenyl phosphine sulfide (DPPS) showed similar results(Table S1).

This alternating bright-quenched PL behavior with alternatingCd or S layers is very similar to that observed in CdS [43] and CdSe[48] core-only QDs. For CdS core-only QDs, the quenching of the PLwith sulfur termination was hypothesized to arise from an abun-dance of hole-trap states in the energy gap, similar to the effectof capping QDs with small molecules containing thiol groups[18,63,64]. Interestingly, it was recently shown that thiol cappedCdSe/CdS QDs could have a QY > 30%, which is extremely high[65], and may at first contradict the idea that sulfur terminatedQDs have quenched PL. However, even for these CdSe/CdS QDs,which were of extremely high quality, exchanging the native phos-phine caps with PEGylated-thiol ligands resulted in significant PL


quenching [65]. Some PL quenching was attributed to imperfectsurface passivation [65], which must also be considered as a sourceof quenching in our case as well. However, for CdS QDs, DFT calcu-lations clearly showed the appearance of mid-gap states in sulfur-rich particles, which act as trap states to quench PL. We suspectedmid-gap trap states could similarly act to quench the PL in CdSe/CdS core/shell QDs. Further, it is possible that the formation ofmid-gap states due to the sulfur rich surface also played a role inthe PL quenching for the CdSe/CdS thiol capped QDs, and that ahigher fraction of surface sulfur would have resulted in QDs dis-playing even more quenching. DFT calculations were performedto quantify the presence of such trap states in these particles.

Fig. 3. (Left panel): Projected density of states (DOS) of Cd(blue), Se(magenta) and S(reCd40S38and (c) S-rich (Cd17Se17)Cd38S40 core–shell QDs. The zero of energy is set equal to tgeometry and [Bottom of right panel] charge density distribution of HOMO, LUMO and mCd, Se and S atoms, respectively. (For interpretation of the references to colour in this fi

3.3. Calculations of CdSe/CdS surface trap states

To represent CdSe/CdS core/shell QDs with a nanostructure thatis computationally tractable, we chose a fixed core size of Cd17Se17,and a shell with fixed thickness and variable surface compositionof Cd39+/�XS39�/+X. First, the stoichiometric core/shell QD Cd17Se17/Cd39S39 was constructed from the bulk wurtzite structure and fullyrelaxed to its lowest energy configuration. Subsequently, wereplaced one, two, and four Cd (S) atoms with S (Cd) atoms inthe outer shell layer to simulate S(Cd)-rich QDs, and relaxed thesestructures to their lowest energy conformations. DFT calculationson larger cluster sizes, or on clusters with ligands, both of which

d) lines respectively of (a) stoichiometric (Cd17Se17)Cd39S39 (b) Cd-rich (Cd17Se17)he energy of the middle of the HOMO–LUMO gap. (Top of right panel) The optimizedid-gap states of the respective core–shell QDs. Blue, magenta and red balls representgure legend, the reader is referred to the web version of this article.)


better approximate the size of the measured QDs, would be desir-able. However, in the current case the system of over 110 atoms isso large that computationally it is intractable to include signifi-cantly more atoms in the calculation. We recognize that for theCd or S rich surfaces, maintaining charge neutrality without includ-ing ligands introduces additional reconstruction of surface atomsthat may not be present on the larger QDs.

As could be seen from the projected density of states (PDOS)of Fig. 3 for a stoichiometric cluster, the highest occupied molec-ular orbital (HOMO) is dominated by S atoms with very littlecontribution from Se atoms, while the lowest unoccupied molec-ular orbitals (LUMO) is dominated by Cd atoms. The charge den-sity plots show that the HOMO and LUMO states are localized inthe opposite surfaces of the cluster (Fig. 3(a)). CdSe/CdS QDs arequasi-type-I particles, wherein the electron is relatively mobilethroughout both core and shell, while the hole sees a much lar-ger potential energy barrier at the CdSe/CdS interface and, thus,is more localized in the core. A HOMO with significant contribu-tions from the sulfur atoms in the shell is an interesting findingand presents a contrasting picture to this traditional particle-in-a-finite potential well model for a ‘‘core–shell” QD. Similarresults are obtained for QDs with Cd-rich surfaces, althoughthe HOMO–LUMO gap is reduced marginally (Fig. 3). Narrowingof the gap is due to both decreased quantum confinement (as aresult of the increased size of Cd atoms compared to S atoms)

Fig. 4. PL decay dynamics. Lines and symbols in red indicate cadmium-terminated Qindicated. (a) Lifetime decay curves for select samples shown in Fig. 2. Samples A and Eapplications of cadmium, thus cadmium terminated. (b) Amplitude weighted lifetimesRadiative and non-radiative rate constants as a function of cadmium or sulfur terminatiinterpretation of the references to colour in this figure legend, the reader is referred to

and appearance of defect states near the edges of the valenceand conduction bands.

The situation is qualitatively different in QDs with S-rich sur-faces. The PDOS plot reveals appearance of localized defect statesin the mid-gap region, 0.6 eV below the conduction band edge(Fig. 3(c)). Further increasing the concentration of surfaces S atomsin the nonstoichiometric clusters increases the number and widthof the mid-gap surface states, as shown in Fig. S4.

To understand the origin of the mid-gap states in the sulfur-richQDs, we analyzed the bonding patterns and Mulliken charge distri-bution of these nonstoichiometric clusters. Due to nonmetalliccharacter, S atoms require formation of covalent bonds. In contrast,Cd atoms form metallic bonds that are less directional than cova-lent bonds and can accommodate stoichiometric defects muchmore easily. S atoms are more mobile on the QD surface, creatingdangling bonds in S-rich nonstoichiometric QDs. Mulliken chargedistributions reveal that the charges on the S atoms responsiblefor the defects deviate from the charges of bulk S atoms as wellas from the charges of surface S atoms in stoichiometric QDs. Inparticular, the defect S atoms are less anionic, indicating thatchemical bonds are missing. On the other hand, even in highly non-stoichiometric Cd-rich QDs, structural reorganization prevents cre-ation of dangling bonds.

The localized nature of the mid-gap states is particularly impor-tant and can play a crucial role in QD photophysical processes.

Ds, whereas black indicates sulfur-terminated QDs, or the seed CdSe core wheredenote full shells, thus sulfur terminated. Samples B and F denote the subsequentas a function of shell growth. (c) Average PL quantum yields for each shell. (d)

on. Relative uncertainties in (b) are contained within the width of the symbol. (Forthe web version of this article.)


Localized trap states can separate electrons and holes, causing areduction in radiative decay rates. Further, localized states exhibithigher electron–phonon coupling [66,67], which are shown to pro-vide efficient nonradiative recombination pathways for electronsor holes [67–69]. Thus, we may predict that QDs with a higher Sconcentration in the shell layer efficiently quench fluorescencevia both a reduction in the radiative rate (due to decreased elec-tron–hole overlap) and an increased nonradiative recombinationrate (due to mid-gap surface trap states).

3.4. Effect of surface composition on fluorescence dynamics

Photoluminescence decay dynamics were measured for CdSecore-only QDs and for CdSe/CdS core–shell QDs for each subse-quent application of cadmium or sulfur to test these predictions.As expected, and consistent with similar systems [70], theobserved lifetimes were found to be multi-exponential and compa-rable to literature values [33]. The multi-exponential nature ofthese ensemble measurements are indicative of surface trappingand recombination on many timescales [71], as opposed to singleexponential data, which signifies high quality, high quantum yieldQDs with near-perfect surfaces.

Several interesting trends are observed in the PL lifetime data assubsequent layers of Cd or S are applied. In general, as the CdS shellgrows larger, the PL lifetime increases (Fig. 4). This effect has beenobserved previously and attributed to reduced electron–hole over-lap in the CdSe/CdS QD due to the electron delocalizing over theentire nanoparticle [33]. Upon application of each successive sulfurlayer, the PL lifetime shortens compared to the previous cadmiumterminated value (Fig. 4). This decrease in PL lifetime is accompa-nied by a drop in PL QY (Table S2). The opposite effect occurs withapplication of each successive Cd layer: the PL lifetime gets longerwhich is accompanied by an increase in PL QY (Table S2). We notethat the application of the first Cd layer on a CdSe core QD (formingCdSe/Cd QDs) does not follow the overall trend seen as the CdSshell gets thicker. In general application of the first Cd-shell ledto a small change in QY (Fig. 2), with a proportional drop in thePL lifetime (Fig. 4). Thus, the change in QY is almost exclusivelydue to an increase in the non-radiative rate with the applicationof an additional Cd layer. As discussed in the previous theoreticalsection, the increase in non-radiative rate could arise from anincrease in defect states near the conduction and valence bandedges.

Using the definition of QY, QY = kr/(kr + knr), the impact of sur-face composition on the QD’s radiative and nonradiative recombi-nation rates can be characterized (Fig. 4(d)). The drop in QY insulfur terminated QDs is due to both an increase in the non-radiative rate and a decrease in the radiative rate (compared tocadmium terminated surfaces). Both effects are consistent withthe predictions of the DFTB calculations. The nonradiative rateincrease originates from the increase in the number mid-gap holestates. Similar studies have found comparable patterns in CdTenanoparticles with increasing telluride concentration on surfaces[19], where hole traps and less passivated telluride surfaces con-tribute to increased nonradiative rates. Further, selenium rich sur-faces on CdSe cores also produce insufficiently passivated particlesurfaces [54] and comparable changes in fluorescent lifetimes[21]. The radiative rate decrease is consistent with localization ofholes on surface sulfur atoms, which causes increased electron–hole separation. The PL lifetime data for cadmium terminatedQDs is also consistent with the DFTB calculations. Although theDFTB calculations indicate the LUMO wavefunctions are localizedon the Cd atoms, Cd-termination does not lead to a high densityof surface trap states and only a weak sensitivity of kr and knr tothe QD size are observed. Thus, we conclude that the mid-gap trap

states (which are only observed in the sulfur terminated particles)are critical in mediating the PL QY of the core/shell particles.

4. Conclusions

CdSe/CdS QDs were synthesized using secondary phosphinechalcogenide and cadmium carboxylate precursors using a SILAR-type method. This method allowed for well-controlled surfacecompositions that can be continually tuned from predominantlycadmium to predominantly sulfur termination. As seen in CdSeand CdS QDs, surface composition correlates with sample fluores-cence intensity: chalcogen termination quenches PL intensity whilemetal termination restores QD PL. This surface termination photo-physical effect is propagated into lifetime data, where nonradiativedecay components drastically increase and dominate lifetimes,especially of sulfur terminated particles. DFTB analysis indicatesthe presence of midgap states arising from sulfur termination, inagreement with previous studies of bare cores. Despite the quasi-type-I electronic heterostructure of these core/shell particles, thesemidgap states are shown by DFTB to act as hole traps on the surface,similarly to CdS coreQDs. In conjunctionwith the theory, the drasticand reversible quenching of fluorescence on chalcogen terminatedsamples suggests that CdS is not an effective hole-confining shellfor CdSe, as surface hole traps remain accessible in spite of the shell.

A notable implication of these findings is that sulfur terminatedsamples may provide more direct access to surface holes than theircadmium terminated counterparts. The inefficiencies of CdS as ahole confining shell, combined with the accessible hole trap stateson the surface, make CdSe/CdS systems with sulfur terminationviable candidates for QD-sensitized photocatalytic hydrogen gen-eration systems [9,10,72], where hole quenching in QDs by a sacri-ficial donor has been shown to be the rate limiting step [73].

Conflict of interest

The authors have no conflicts of interest regarding any aspectsof this work.

Acknowledgements

This work was supported by the National Science Foundationthrough CHE-1307254: (B.S., T.D.K., K.S., J.P.) and CHE-1300118(O.P.) and the Department of Energy Office of Basic Energy Sciencesthrough Grant DE-FG02-06ER15821 (Z. H.).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.chemphys.2015.09.010.

References

[1] A.I. Ekimov, A.A. Onushchenko, Zh. Eksp. Teor. Fiz. 34 (1981) 363.[2] A.P. Alivisatos, Science 271 (5251) (1996) 933.[3] R. Rossetti, S. Nakahara, L.E. Brus, J. Chem. Phys. 79 (1983) 1086.[4] C.M. Evans, L. Guo, J.J. Peterson, S. Maccagnano-Zacher, T.D. Krauss, Nano Lett.

8 (9) (2008) 2896.[5] M. Nirmal, B.O. Dabbousi, M.G. Bawendi, J.J. Macklin, J.K. Trautman, T.D. Harris,

L.E. Brus, Nature 383 (6603) (1996) 802.[6] M.T. Frederick, E.A. Weiss, ACS Nano 4 (6) (2010) 3195.[7] S. Jin, R.D. Harris, B. Lau, K.O. Aruda, V.A. Amin, E.A. Weiss, Nano Lett. 14 (9)

(2014) 5323.[8] C.B. Murray, C.R. Kagan, M.G. Bawendi, Annu. Rev. Mater. Sci. 30 (1) (2000)

545.[9] Z. Han, F. Qiu, R. Eisenberg, P.L. Holland, T.D. Krauss, Science 338 (6112) (2012)

1321.[10] K. Wu, H. Zhu, T. Lian, Acc. Chem. Res. 48 (3) (2015) 851.[11] D.S. Ginger, N.C. Greenham, Phys. Rev. B 59 (1999) 10622.

http://dx.doi.org/10.1016/j.chemphys.2015.09.010http://dx.doi.org/10.1016/j.chemphys.2015.09.010http://refhub.elsevier.com/S0301-0104(15)00298-0/h0005http://refhub.elsevier.com/S0301-0104(15)00298-0/h0010http://refhub.elsevier.com/S0301-0104(15)00298-0/h0015http://refhub.elsevier.com/S0301-0104(15)00298-0/h0020http://refhub.elsevier.com/S0301-0104(15)00298-0/h0020http://refhub.elsevier.com/S0301-0104(15)00298-0/h0025http://refhub.elsevier.com/S0301-0104(15)00298-0/h0025http://refhub.elsevier.com/S0301-0104(15)00298-0/h0030http://refhub.elsevier.com/S0301-0104(15)00298-0/h0035http://refhub.elsevier.com/S0301-0104(15)00298-0/h0035http://refhub.elsevier.com/S0301-0104(15)00298-0/h0040http://refhub.elsevier.com/S0301-0104(15)00298-0/h0040http://refhub.elsevier.com/S0301-0104(15)00298-0/h0045http://refhub.elsevier.com/S0301-0104(15)00298-0/h0045http://refhub.elsevier.com/S0301-0104(15)00298-0/h0050http://refhub.elsevier.com/S0301-0104(15)00298-0/h0055


[12] I. Robel, V. Subramanian, M. Kuno, P.V. Kamat, J. Am. Chem. Soc. 128 (7) (2006)2385.

[13] S. Coe, W.-K. Woo, M. Bawendi, V. Bulovic, Nature 420 (6917) (2002) 800.[14] S. Coe-Sullivan, W.-K. Woo, J.S. Steckel, M. Bawendi, V. Bulović, Org. Electron. 4

(2–3) (2003) 123.[15] J.K. Jaiswal, H. Mattoussi, J.M. Mauro, S.M. Simon, Nat. Biotechnol. 21 (1)

(2003) 47.[16] M. Bruchez Jr., M. Moronne, P. Gin, S. Weiss, A.P. Alivisatos, Science 281 (5385)

(1998) 2013.[17] T.C. Pappas, W.M.S. Wickramanyake, E. Jan, M. Motamedi, M. Brodwick, N.A.

Kotov, Nano Lett. 7 (2) (2007) 513.[18] V.V. Breus, C.D. Heyes, G.U. Nienhaus, J. Phys. Chem. C 111 (50) (2007) 18589.[19] B. Omogo, J.F. Aldana, C.D. Heyes, J. Phys. Chem. C 117 (5) (2013) 2317.[20] K.B. Subila, G.K. Kumar, S.M. Shivaprasad, K.G. Thomas, J. Phys. Chem. Lett. 4

(2013) 2774.[21] Y. Gao, X. Peng, J. Am. Chem. Soc. 137 (12) (2015) 4230.[22] O. Chen, Y. Yang, T. Wang, H. Wu, C. Niu, J. Yang, Y.C. Cao, J. Am. Chem. Soc. 133

(43) (2011) 17504.[23] M. Kuno, J.K. Lee, B.O. Dabbousi, F.V. Mikulec, M.G. Bawendi, J. Chem. Phys. 106

(23) (1997) 9869.[24] M.A. Hines, P. Guyot-Sionnest, J. Phys. Chem. 100 (2) (1996) 468.[25] J.S. Steckel, J.P. Zimmer, S. Coe-Sullivan, N.E. Stott, V. Bulović, M.G. Bawendi,

Angew. Chem. Int. Ed. 43 (16) (2004) 2154.[26] V.V. Protasenko, M. Kuno, A. Gallagher, D.J. Nesbitt, Opt. Commun. 210 (1–2)

(2002) 11.[27] C.B. Murray, D.J. Norris, M.G. Bawendi, J. Am. Chem. Soc. 115 (1993) 8706.[28] B.O. Dabbousi, J. Rodriguez-Viejo, F.V. Mikulec, J.R. Heine, H. Mattoussi, R.

Ober, K.F. Jensen, M.G. Bawendi, J. Phys. Chem. B 101 (46) (1997) 9463.[29] G.E. Cragg, A.L. Efros, Nano Lett. 10 (1) (2010) 313.[30] Q. Lin, N.S. Makarov, W.-K. Koh, K.A. Velizhanin, C.M. Cirloganu, H. Luo, V.I.

Klimov, J.M. Pietryga, ACS Nano 9 (1) (2015) 539.[31] Y. Chen, J. Vela, H. Htoon, J.L. Casson, D.J. Werder, D.A. Bussian, V.I. Klimov, J.A.

Hollingsworth, J. Am. Chem. Soc. 130 (15) (2008) 5026.[32] B. Mahler, N. Lequeux, B. Dubertret, J. Am. Chem. Soc. 132 (3) (2010) 953.[33] W.K. Bae, L.A. Padilha, Y.-S. Park, H. McDaniel, I. Robel, J.M. Pietryga, V.I.

Klimov, ACS Nano 7 (4) (2013) 3411.[34] A.M. Smith, L.A. Lane, S. Nie, Nat. Commun. 5 (2014).[35] X. Zhong, M. Han, Z. Dong, T.J. White, W. Knoll, J. Am. Chem. Soc. 125 (28)

(2003) 8589.[36] H. Zhu, T. Lian, Energy Environ. Sci. 5 (11) (2012) 9406.[37] S. Kim, B. Fisher, H.-J. Eisler, M. Bawendi, J. Am. Chem. Soc. 125 (38) (2003)

11466.[38] S.A. Ivanov, A. Piryatinski, J. Nanda, S. Tretiak, K.R. Zavadil, W.O. Wallace, D.

Werder, V.I. Klimov, J. Am. Chem. Soc. 129 (38) (2007) 11708.[39] M.V. Kovalenko, M.I. Bodnarchuk, J. Zaumseil, J.-S. Lee, D.V. Talapin, J. Am.

Chem. Soc. 132 (29) (2010) 10085.[40] N.C. Anderson, M.P. Hendricks, J.J. Choi, J.S. Owen, J. Am. Chem. Soc. 135 (49)

(2013) 18536.

[41] C.M. Evans, M.E. Evans, T.D. Krauss, J. Am. Chem. Soc. 132 (32) (2010) 10973.[42] J.S. Steckel, B.K.H. Yen, D.C. Oertel, M.G. Bawendi, J. Am. Chem. Soc. 128 (40)

(2006) 13032.[43] H.H.-Y. Wei, C.M. Evans, B.D. Swartz, A.J. Neukirch, J. Young, O.V. Prezhdo, T.D.

Krauss, Nano Lett. 12 (9) (2012) 4465.[44] N.J. Borys, M.J. Walter, J. Huang, D.V. Talapin, J.M. Lupton, Science 330 (6009)

(2010) 1371.[45] D. Steiner, D. Dorfs, U. Banin, F. Della, Nano Lett. 8 (9) (2008) 2954.[46] A. Pandey, P. Guyot-Sionnest, J. Chem. Phys. 127 (10) (2007) 104710.[47] J.J. Li, Y.A. Wang, W. Guo, J.C. Keay, T.D. Mishima, M.B. Johnson, X. Peng, J. Am.

Chem. Soc. 125 (41) (2003) 12567.[48] J. Jasieniak, P. Mulvaney, J. Am. Chem. Soc. 129 (10) (2007) 2841.[49] W.W. Yu, X. Peng, Angew. Chem. Int. Ed. 41 (13) (2002) 2368.[50] W.W. Yu, L. Qu, W. Guo, X. Peng, Chem. Mater. 15 (14) (2003) 2854.[51] C.D. Wagner, W.M. Riggs, L.E. Davis, J.F. Moulder, G.E. Muilenberg, Handbook

of X-ray Photoelectron Spectroscopy. (Physical Electronics Inc, 1978).[52] A.R. West, Basic Solid State Chemistry, 1988.[53] J. Nanda, B.A. Kuruvilla, D.D. Sarma, Phys. Rev. B 59 (1999) 7473.[54] J.E.B. Katari, V.L. Colvin, A.P. Alivisatos, J. Phys. Chem. 98 (1994) 4109.[55] D. Porezag, T. Frauenheim, T. Köhler, G. Seifert, R. Kaschner, Phys. Rev. B 51

(19) (1995) 12947.[56] M. Elstner, D. Porezag, G. Jungnickel, J. Elsner, M. Haugk, T. Frauenheim, S.

Suhai, G. Seifert, Phys. Rev. B 58 (11) (1998) 7260.[57] T.A. Niehaus, S. Suhai, F. Della Sala, P. Lugli, M. Elstner, G. Seifert, T.

Frauenheim, Phys. Rev. B 63 (8) (2001) 085108.[58] G. Seifert, J. Phys. Chem. A 111 (26) (2007) 5609.[59] S. Sarkar, S. Pal, P. Sarkar, A.L. Rosa, T. Frauenheim, J. Chem. Theory Comput. 7

(7) (2011) 2262.[60] M.S. Setty, A.P.B. Sinha, Thin Solid Films 144 (1) (1986) 7.[61] S. Tanuma, C.J. Powell, D.R. Penn, Surf. Interface Anal. 20 (1) (1993) 77.[62] S. Tanuma, C.J. Powell, D.R. Penn, Surf. Interface Anal. 21 (3) (1994) 165.[63] Y. Liu, M. Kim, Y. Wang, Y.A. Wang, X. Peng, Langmuir 22 (14) (2006) 6341.[64] S.F. Wuister, C. de Mello Donegá, A. Meijerink, J. Phys. Chem. B 108 (45) (2004)

17393.[65] O. Chen, J. Zhao, V.P. Chauhan, J. Cui, C. Wong, D.K. Harris, H. Wei, H.-S. Han, D.

Fukumura, R.K. Jain, M.G. Bawendi, Nat. Mater. 12 (5) (2013) 445.[66] B.F. Habenicht, H. Kamisaka, K. Yamashita, O.V. Prezhdo, Nano Lett. 7 (11)

(2007) 3260.[67] B.F. Habenicht, C.F. Craig, O.V. Prezhdo, Phys. Rev. Lett. 96 (18) (2006) 187401.[68] B.F. Habenicht, O.N. Kalugin, O.V. Prezhdo, Nano Lett. 8 (8) (2008) 2510.[69] B.F. Habenicht, O.V. Prezhdo, J. Phys. Chem. C 113 (32) (2009) 14067.[70] A.V. Malko, Y.-S. Park, S. Sampat, C. Galland, J. Vela, Y. Chen, J.A. Hollingsworth,

V.I. Klimov, H. Htoon, Nano Lett. 11 (12) (2011) 5213.[71] K.E. Knowles, E.A. McArthur, E.A. Weiss, ACS Nano 5 (3) (2011) 2026.[72] A. Vaneski, J. Schneider, A.S. Susha, A.L. Rogach, APL Mater. 2 (1) (2014)

012104.[73] K. Wu, Z. Chen, H. Lv, H. Zhu, C.L. Hill, T. Lian, J. Am. Chem. Soc. 136 (21) (2014)

7708.

http://refhub.elsevier.com/S0301-0104(15)00298-0/h0060http://refhub.elsevier.com/S0301-0104(15)00298-0/h0060http://refhub.elsevier.com/S0301-0104(15)00298-0/h0065http://refhub.elsevier.com/S0301-0104(15)00298-0/h0070http://refhub.elsevier.com/S0301-0104(15)00298-0/h0070http://refhub.elsevier.com/S0301-0104(15)00298-0/h0075http://refhub.elsevier.com/S0301-0104(15)00298-0/h0075http://refhub.elsevier.com/S0301-0104(15)00298-0/h0080http://refhub.elsevier.com/S0301-0104(15)00298-0/h0080http://refhub.elsevier.com/S0301-0104(15)00298-0/h0085http://refhub.elsevier.com/S0301-0104(15)00298-0/h0085http://refhub.elsevier.com/S0301-0104(15)00298-0/h0090http://refhub.elsevier.com/S0301-0104(15)00298-0/h0095http://refhub.elsevier.com/S0301-0104(15)00298-0/h0100http://refhub.elsevier.com/S0301-0104(15)00298-0/h0100http://refhub.elsevier.com/S0301-0104(15)00298-0/h0105http://refhub.elsevier.com/S0301-0104(15)00298-0/h0110http://refhub.elsevier.com/S0301-0104(15)00298-0/h0110http://refhub.elsevier.com/S0301-0104(15)00298-0/h0115http://refhub.elsevier.com/S0301-0104(15)00298-0/h0115http://refhub.elsevier.com/S0301-0104(15)00298-0/h0120http://refhub.elsevier.com/S0301-0104(15)00298-0/h0125http://refhub.elsevier.com/S0301-0104(15)00298-0/h0125http://refhub.elsevier.com/S0301-0104(15)00298-0/h0130http://refhub.elsevier.com/S0301-0104(15)00298-0/h0130http://refhub.elsevier.com/S0301-0104(15)00298-0/h0135http://refhub.elsevier.com/S0301-0104(15)00298-0/h0140http://refhub.elsevier.com/S0301-0104(15)00298-0/h0140http://refhub.elsevier.com/S0301-0104(15)00298-0/h0145http://refhub.elsevier.com/S0301-0104(15)00298-0/h0150http://refhub.elsevier.com/S0301-0104(15)00298-0/h0150http://refhub.elsevier.com/S0301-0104(15)00298-0/h0155http://refhub.elsevier.com/S0301-0104(15)00298-0/h0155http://refhub.elsevier.com/S0301-0104(15)00298-0/h0160http://refhub.elsevier.com/S0301-0104(15)00298-0/h0165http://refhub.elsevier.com/S0301-0104(15)00298-0/h0165http://refhub.elsevier.com/S0301-0104(15)00298-0/h0170http://refhub.elsevier.com/S0301-0104(15)00298-0/h0175http://refhub.elsevier.com/S0301-0104(15)00298-0/h0175http://refhub.elsevier.com/S0301-0104(15)00298-0/h0180http://refhub.elsevier.com/S0301-0104(15)00298-0/h0185http://refhub.elsevier.com/S0301-0104(15)00298-0/h0185http://refhub.elsevier.com/S0301-0104(15)00298-0/h0190http://refhub.elsevier.com/S0301-0104(15)00298-0/h0190http://refhub.elsevier.com/S0301-0104(15)00298-0/h0195http://refhub.elsevier.com/S0301-0104(15)00298-0/h0195http://refhub.elsevier.com/S0301-0104(15)00298-0/h0200http://refhub.elsevier.com/S0301-0104(15)00298-0/h0200http://refhub.elsevier.com/S0301-0104(15)00298-0/h0205http://refhub.elsevier.com/S0301-0104(15)00298-0/h0210http://refhub.elsevier.com/S0301-0104(15)00298-0/h0210http://refhub.elsevier.com/S0301-0104(15)00298-0/h0215http://refhub.elsevier.com/S0301-0104(15)00298-0/h0215http://refhub.elsevier.com/S0301-0104(15)00298-0/h0220http://refhub.elsevier.com/S0301-0104(15)00298-0/h0220http://refhub.elsevier.com/S0301-0104(15)00298-0/h0225http://refhub.elsevier.com/S0301-0104(15)00298-0/h0230http://refhub.elsevier.com/S0301-0104(15)00298-0/h0235http://refhub.elsevier.com/S0301-0104(15)00298-0/h0235http://refhub.elsevier.com/S0301-0104(15)00298-0/h0240http://refhub.elsevier.com/S0301-0104(15)00298-0/h0245http://refhub.elsevier.com/S0301-0104(15)00298-0/h0250http://refhub.elsevier.com/S0301-0104(15)00298-0/h0265http://refhub.elsevier.com/S0301-0104(15)00298-0/h0270http://refhub.elsevier.com/S0301-0104(15)00298-0/h0275http://refhub.elsevier.com/S0301-0104(15)00298-0/h0275http://refhub.elsevier.com/S0301-0104(15)00298-0/h0280http://refhub.elsevier.com/S0301-0104(15)00298-0/h0280http://refhub.elsevier.com/S0301-0104(15)00298-0/h0285http://refhub.elsevier.com/S0301-0104(15)00298-0/h0285http://refhub.elsevier.com/S0301-0104(15)00298-0/h0290http://refhub.elsevier.com/S0301-0104(15)00298-0/h0295http://refhub.elsevier.com/S0301-0104(15)00298-0/h0295http://refhub.elsevier.com/S0301-0104(15)00298-0/h0300http://refhub.elsevier.com/S0301-0104(15)00298-0/h0305http://refhub.elsevier.com/S0301-0104(15)00298-0/h0310http://refhub.elsevier.com/S0301-0104(15)00298-0/h0315http://refhub.elsevier.com/S0301-0104(15)00298-0/h0320http://refhub.elsevier.com/S0301-0104(15)00298-0/h0320http://refhub.elsevier.com/S0301-0104(15)00298-0/h0325http://refhub.elsevier.com/S0301-0104(15)00298-0/h0325http://refhub.elsevier.com/S0301-0104(15)00298-0/h0330http://refhub.elsevier.com/S0301-0104(15)00298-0/h0330http://refhub.elsevier.com/S0301-0104(15)00298-0/h0335http://refhub.elsevier.com/S0301-0104(15)00298-0/h0340http://refhub.elsevier.com/S0301-0104(15)00298-0/h0345http://refhub.elsevier.com/S0301-0104(15)00298-0/h0350http://refhub.elsevier.com/S0301-0104(15)00298-0/h0350http://refhub.elsevier.com/S0301-0104(15)00298-0/h0355http://refhub.elsevier.com/S0301-0104(15)00298-0/h0360http://refhub.elsevier.com/S0301-0104(15)00298-0/h0360http://refhub.elsevier.com/S0301-0104(15)00298-0/h0365http://refhub.elsevier.com/S0301-0104(15)00298-0/h0365

Photophysical Properties of CdSe/CdS core/shell quantum dots with tunable surface composition1 Introduction2 Methods2.1 CdSe QD synthesis2.2 CdS/CdSe core/shell QD synthesis2.3 Optical characterization2.4 Fraction preparation2.5 Transmission electron microscopy2.6 X-ray photoelectron spectroscopy (XPS)2.7 PL lifetime measurements2.8 Computational methodology

3 Results and discussion3.1 X-ray photoelectron spectroscopy analysis of particle surface termination3.2 Effects of surface composition on photoluminescence3.3 Calculations of CdSe/CdS surface trap states3.4 Effect of surface composition on fluorescence dynamics

4 ConclusionsConflict of interestAcknowledgementsAppendix A Supplementary dataReferences

photophysical properties of cdse/cds core/shell quantum ...s 2p 3/2 and 1/2 (approximately 162 and...

Documents