Description of the Adsorption and Exciton Delocalizing Properties of p

Subscriber access provided by University Libraries, University of Memphis

Article

A Description of the Adsorption and Exciton Delocalizing Properties of p-Substituted Thiophenols on CdSe Quantum Dots Kenneth O. Aruda, Victor A. Amin, Christopher M. Thompson, Bryan Lau, Alexander B Nepomnyashchii, and Emily A. Weiss Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b00689 • Publication Date (Web): 22 Mar 2016 Downloaded from http://pubs.acs.org on March 26, 2016

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

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

Page 1 of 32

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

Langmuir

A Description of the Adsorption and Exciton Delocalizing Properties of p-Substituted Thiophenols on CdSe Quantum Dots Kenneth O. Aruda, Victor A. Amin, Christopher M. Thompson, Bryan Lau, Alexander B. Nepomnyashchii, and Emily A. Weiss* Department of Chemistry, Northwestern University, 2145 Sheridan Rd., Evanston, IL 60208-3113 *corresponding author. Email: [email protected] Abstract This paper describes the quantitative characterization of the interfacial chemical and electronic structure of CdSe quantum dots (QDs) coated in one of five para-substituted thiophenolates (XTP, X= NH2, CH3O, CH3, Cl, NO2), and the dependence of this structure on the para-substituent X. 1H NMR spectra of mixtures of CdSe QDs and X-TPs yield the number of X-TPs bound to the surface of each QD. The binding data, in combination with the shift in the energy of the first excitonic peak of the QDs as a function of the surface coverage of X-TP and Raman and NMR analysis of the mixtures indicate that X-TP binds to CdSe QDs in at least three modes, two modes that are responsible for exciton delocalization, and a third mode that does not affect the excitonic energy. The first two modes involve displacement of OPA from the QD core, whereas the third mode forms cadmium-thiophenolate complexes that are not electronically coupled to the QD core. Fits to the data using the dual-mode binding model also yield the values of ∆r1, the average radius of exciton delocalization due to binding of the X-TP in modes 1 and 2. A 3D parameterized particle-in-a-sphere model enables the conversion of the measured value of ∆r1 for each X-TP to the height of the potential barrier that the ligand presents for tunneling of excitonic hole into the interfacial region. The height of this barrier increases from 0.3 eV to 0.9 eV as the substituent, X, becomes more electron-withdrawing.

1 ACS Paragon Plus Environment

Langmuir

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

INTRODUCTION This paper describes the quantitative characterization of the interfacial chemical and electronic structure of CdSe quantum dots (QDs) coated in mixed monolayers of octylphosphonate (OPA) and one of five para-substituted thiophenolates (X-TP, X= NH2, CH3O, CH3, Cl, NO2). Thiophenolates are exciton-delocalizing ligands; substitution of an insulating native ligand like OPA for an X-TP results in extension of the excitonic wavefunction beyond the inorganic core and into the mixed inorganic-organic interfacial region, and a bathochromic shift of the first excitonic absorption of the QD, Figure 1A. We denote the apparent increase in the radius of the exciton “∆R”.1 Delocalization of the exciton into the organic portion of the QD, without sacrificing the narrow linewidth of the absorption and emission or chemical passivation offered by the ligands2,3, is one strategy for increasing the electronic coupling for charge or exciton extraction, a necessary step in any solar energy conversion or photodetection application of QDs.4-10 The binding affinity and geometries of thiophenolates and other delocalizing ligands are inextricably tied to the response of the QD bandgap (measured by ∆R) to adsorption of these ligands because: (i) the observed value of ∆R is a convolution of its value per bound ligand and the surface coverage of the ligand, and (ii) the binding modes of the ligands determine the availability of mixed-character (QD/ligand) orbitals for delocalization of excitonic electron and/or hole, Figure 1B.1,11 We demonstrated previously12 an NMR method to quantify the coverages of methylthiophenolate (CH3-TP) on the surfaces of solution-phase CdSe QDs of a series of core sizes. In that manuscript, we also describe a model to simultaneously fit (i) the binding isotherms derived from that data, and (ii) the dependence of ∆R on the average surface coverage of CH3-TP for the ensemble, for each size. This model unambiguously showed the presence of (at least) two binding modes for CH3-TP on CdSe QDs, yielded the value of the equilibrium adsorption constant 2 ACS Paragon Plus Environment

Page 2 of 32

Page 3 of 32

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

Langmuir

for each of these modes, and revealed that only the tighter-binding mode induces exciton delocalization and contributes to the observed ∆R. The model also yielded the value of ∆r: the exciton delocalization length per bound CH3-TP. This value, in conjunction with a simple particlein-a-sphere model of the predicted excitonic energies as a function of surface coverage of adsorbed delocalizing ligand, allowed us to determine the absolute height of the tunneling barrier for the excitonic hole presented by CH3-TP when it binds to the surface of a CdSe QD. Importantly, the combination of the NMR and optical absorption data provided both evidence for the existence of two binding geometries for CH3-TP and quantitative information about the electronic structure of the QD/CH3-TP interface that was not available by analyzing either dataset alone. In the previous study12, we only examined the effects of one delocalizing ligand, CH3-TP (for a series of sizes of CdSe QDs), and we did not describe chemically the multiple detected binding modes of CH3-TP on the QD. Here, we use quantitative NMR and the previously developed model to correlate the binding affinities of a series of exciton-delocalizing para-substituted thiophenols (X-TP, where X ranges from electron-donating to electron withdrawing) to the barrier heights they present for the excitonic hole. This analysis of a series of X-TP molecules interacting with a single size of CdSe QD allows us to (i) determine the dependence of the strength of electronic interaction between X-TP and the QD on X, and thereby propose a model for orbital mixing at the inorganic/organic interface of the QD; and (ii) determine the dependence of the binding constants for QD/X-TP system on X, and, with the help of calculations of adsorption enthalpy as a function of X, and measurements of the solubility of the X-TPs in our solvent, CHCl3, thereby propose chemical descriptions of the detected binding modes. We support our chemical description of the system with Raman and NMR spectra of the QDs coated with both their native phosphonate and X-TP ligands. These data indicate that X-TPs adsorb onto CdSe QDs in at least three binding


Langmuir

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

Figure 1. A) Ground state absorption spectra of 9.2µM CdSe QDs in CHCl3 before (black) and after (blue) addition of 4 mM NO2-TP. The value ΔR is defined as the apparent increase in excitonic radius that corresponds to the observed shift in the energy of the first excitonic absorption of the QDs upon ligand exchange from octylphosphonate (OPA) to X-TP. B) Energies of the relevant states of the isolated QD, including core states (determined by UPS spectroscopy) and a state corresponding to undercoordinated Cd2+ on the surface (placed an arbitrary energy below the CB edge),energies of the HOMOs of two of the isolated X-TP ligands in their solvated, anionic form and of the solvated Cd(XTP)2 complexes (determined from electrochemistry), see also Table 3. Upon electron donation from XTP- to Cd2+ and mixing with VB orbitals, a new interfacial HOMO forms that is closer to the VB edge than is the state formed by bound OPA. The gap between the VB edge of the QDs and the new interfacial states is the barrier for delocalization of the excitonic hole, EB. For the X-TPs, this barrier is calculated from ∆r1, as described in the text. For simplicity, unoccupied interfacial orbitals are not shown in this diagram. We calculate the HOMO of OPA- (solvated) with DFT. We cannot measure the oxidation potential of Cd(OPA)2, so we draw its oxidation potential, and the energy of the interfacial state, as degenerate with that of OPA-, which is the upper limit of these energies.


Page 4 of 32

Page 5 of 32

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

Langmuir

modes, two involving Cd2+ in the core of the QD, and one involving Cd2+ complexes that are not electronically coupled to the core.

MATERIALS AND METHODS Materials. All reagents were used as-received from Sigma-Aldrich, with the exception of cadmium stearate, which was used as-received from MP Biomedicals, Inc. Synthesis and Purification of QDs. We added 90% technical grade trioctylphosphine oxide (1.94 g, 5.02 mmol), 90% technical grade hexadecylamine (1.94 g, 8.03 mmol), and cadmium stearate (0.112 g, 0.165 mmol) to a dry 50-mL three-neck round bottom flask and dried the reaction mixture for 20 minutes at 150 °C under N2 flow. We heated the mixture to 320 °C with stirring under positive N2 pressure. At 320 °C, we rapidly injected 1 mL of 1 M trioctylphosphine selenide (99.99% Se 100 mesh powder in 97% trioctylphosphine, prepared and stored in air free conditions) and immediately reduced the temperature to 290 °C. The size of the QD was monitored by ground state absorption spectroscopy and when the desired size was reached, the reaction was cooled to 120 °C and quenched with chloroform. To purify the QDs, they were first precipitated with methanol (2:1 methanol:QDs) and separated by centrifugation. The resultant QD pellets were dispersed in a minimal amount of hexanes and sat in the dark overnight, during which time excess reactants precipitated. We then centrifuged the dispersions to separate the precipitant, decanted the QD supernate, removed the solvent, and redispersed the QDs in deuterated chloroform (CDCl3). Ligand Exchange Procedure. For each X-TP, we created two stock solutions in CDCl3 (~10 mM and ~100 mM). In between spectroscopic measurements, we added varying quantities (5-40 µL) of the X-TP stock solutions with a calibrated micropipette to 400 µL of 9.2 µM solutions of QDs in CDCl3. After each addition, the QDs were vortexed for 15 seconds to ensure homogeneity.


Langmuir

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

We allowed the ligand exchange to proceed for at least 15 minutes before measuring a sample (a length of time determined by a kinetic experiment, see Figure S6). Synthesis of Model Compounds. We synthesized Cd(CH3-TP)2 (used as a model compound for analysis of Raman spectra) by the addition of CdCl2 (14 mM in methanol) to a solution containing two molar equivalents of MBT (223 mM in methanol) and 2 equivalents triethylamine. We collected the white precipitate and washed with methanol twice, and then dried. We synthesized CdxOPAy complexes (used as a model compounds for analysis of NMR spectra) by adding an aqueous CdCl2 solution to a solution of two equivalents of octylphosphonic acid in a water/methanol mixture. A white precipitate formed immediately. The pH of the solution was adjusted to 12 using a 0.2 M solution of NaOH. The precipitate was collected, washed with methanol and dried. Ground State Absorption Measurements. We performed ground state absorption measurements on a Varian Cary 5000 spectrometer. We placed the samples, still in their NMR tubes, inside a 1-cm pathlength quartz cuvette and ensured that they were centered by using a plastic spacer. We corrected the baselines of all spectra with neat solvent (in the same NMR tube cuvette holder) prior to measurement. All QD spectra were taken in CDCl3. In order to estimate the radius of the QDs (and determine ΔR upon exchange with CH3-TP) from their band-edge absorption energy, we used the calibration curves (R vs. εabs) from Yu et al.13 Geometry Optimization and Orbital Energy Calculations. All geometries and orbital energies were calculated with density functional theory (DFT) using QChem 4.114, with the m06 functional15 and a def2 basis set.16 We performed a preliminary geometry optimization with an MMFF94 force field to provide a starting geometry for the DFT calculation. After geometry optimization, DFT single point energies were calculated with the def2-SVPD basis set and a larger-


Page 6 of 32

Page 7 of 32

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

Langmuir

than-default grid (XC_GRID 000128000302).17 We calculate ∆Hsolv using DFT with a COSMO solvation model,18 where the solvent is treated as a dielectric continuum with the static dielectric constant of CHCl3 (ε0 = 4.8). Electrochemistry. All compounds used for synthesis or electrochemical measurements were from Sigma-Aldrich (Milwaukee, WI) and used as received without further purification. Electrochemical investigations were performed in an oxygen-free atmosphere glovebox using nitrogen as a purged gas (Vacuum Atmospheres, CA). We made solutions with cadmiumthiophenol complexes and 0.1 M tetra-n-butylammonium hexafluorophosphate (TBAPF6 in dimethylformamide (DMF) inside the glovebox.

We synthesized Cd-X-TP complexes by

dissolving X-TPs in methanol and adding the solutions dropwise to cadmium acetate hydrate dissolved in water.2 The concentrations used were 2.3 mM for OCH3-TP, 2.7 mM for CH3-TP, 2.9 mM for Cl-TP and 3.0 mM for NO2-TP. The collected precipitate was dried under vacuum. The final solutions were sealed with a Teflon cap (ACE glass, Vineland, NJ). Rods were drilled through the cap to assure electrical connections with working, reference and counter electrodes. Glue was used to prevent oxygen leakage. A platinum 0.0314 cm2 disc electrode (CH instruments, Austin, TX) was applied as a working electrode, and platinum and silver wires were used as counter and reference electrodes. The platinum disc working electrode was polished prior cyclic voltammetry investigations with a saturated aqueous dispersion of 0.3 µm alumina and vigorously washed with water afterwards. We calibrated the potential for the cyclic voltammetry experiments using a ferrocene solution in acetonitrile and assuming the potential for the ferrocene/ferrocenium couple to be +0.342 V vs. SCE.19 We carried out cyclic voltammetry investigations using a Versa Stat potentiostat (Princeton Applied Research, Berwyn, PA).


Langmuir

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

1H

Page 8 of 32

NMR Spectroscopy and Analysis. We performed 1H NMR spectroscopy on a Bruker

Avance III 600 MHz spectrometer in identical NMR tubes. All QD sample concentrations were 9.2 μM in CDCl3. We acquired 8 scans per sample, with a relaxation time of 1 s, and a pulse angle of 45°. Although longer relaxation times are often required for analyzing molecules in bound to, or in exchange with, the surface of a colloid, the signals we analyze quantitatively are those of freely diffusing small molecules, so all of our measurements of X-TP concentration have ≤2% error (see the Supporting Information, Table S2). For a series of concentrations of added X-TP, we measured the NMR intensity of free X-TP relative to a QD-free standard and defined the number adsorbed to QDs as the difference between the two values. In some samples another pair of doublets appears, which we attribute to formation of bis(p-X phenyl) disulfide; in this case we subtract the integrated disulfide signal from the total integration of the free X-TP in the QD-free standard. Raman Measurements and Fitting. We prepared CdSe QDs with a first excitonic absorption at 540 nm for Raman spectroscopy. We dried the quantum dots in glass vials and resuspended them in solutions of methylbenzenethiol (MBT) in chloroform, such that the final concentration of QDs was always 20 µM and the final concentration of MBT corresponded to 450, 200, 100, 80, 50, and 25 equivalents per QD (±1%). We allowed the solutions to mix for 1 hour with vigorous stirring, verified that the absorption spectra of the QDs showed the appropriate bathochromic shift, and then deposited and dried them on clean glass slides. We collected the dried material with a razor blade and a spread a thick film onto aluminum foil for Raman. Raman spectra were collected on a Labram HR Evolution Raman microscope (Horiba) using 633 nm excitation using a 50x long working distance objective.

The incident power was

approximately 20 mW. No damage was evident in the samples during measurement. Collection


Page 9 of 32

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

Langmuir

time and incident laser power were consistent from sample to sample. We baseline corrected Raman spectra of quantum dots with methylbenzenethiol treatment and normalized the each spectrum by the area of the CH-twist feature of the phosponate alkyl group at 1300 cm-1.

RESULTS AND DISCUSSION We perform all of the following measurements on a single synthetic batch of CdSe QDs (diameter = 2.7 nm), in order to eliminate batch-to-batch variation in average size or surface chemistry of the QDs. We add X-TPs to 9.2 µM CdSe QDs at a series of molar equivalents (0 ~700 for X = CH3O, CH3, Cl, NO2 and 0 - ~2500 for X = NH2) and observe a reduction in the optical band gap of the QDs as X-TPs displace the QD’s native ligands, which, as a result of our synthetic procedure (see the Supporting Information) are primarily octylphosphonate and cadmium-octylphosphonate complexes.20-22 We monitor the shift in the first excitonic peak of the QDs with UV-visible absorption spectroscopy and report the shift as the increase in the apparent excitonic radius, ΔR, Figure 1A, using a previously measured calibration curve of radius vs. bandgap for CdSe QDs.11 We note that the parameter ΔR should not be interpreted as a physical delocalization length, since it is calculated as if the exciton were extending into CdSe, when, in fact, it is extending into a n interfacial region of mixed QD-ligand character. ΔR is rather an indicator of the change in confinement energy induced by adsorption of X-TP. We use the 1H NMR spectra of the QD-ligand mixtures to calculate the surface coverage of XTP ligands on CdSe QDs. NMR is a convenient tool for this purpose because, unlike thermogravimetric analysis, which is extremely useful for estimating the total amount of organic content in a QD sample, but requires the sample to be dried and often involves some educated guesswork about the decomposition temperatures of closely related molecules,23 we examine our


Langmuir

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

Page 10 of 32

systems in solution, and thereby maintain the binding equilibrium of the various ligands. The NMR signals of ligands broaden when the ligands bind to QDs because (i) the chemical environment at the surface of nanocrystals is chemically and structurally heterogeneous, and (ii) being bound to a (relatively) large object constrains the rotational degrees of freedom of the molecule.12,24-37 X-TP ligands bound to QDs are especially susceptible to both broadening mechanisms because of the sensitivity of the aromatic protons to the chemical environment of the thiolate, the small size and rigidity of the ligand, the large binding affinity of the thiolate group for surface metal ions.38,39 The signals from all of the protons on bound X-TP are therefore indistinguishable from the baseline in the NMR spectra we acquired.23,40 Figures 2A-E show the NMR spectra of two equimolar solutions of X-TP. One solution contains 2.4 mM X-TP dissolved in CDCl3, and the other contains 2.4 mM X-TP plus 9.2 μM CdSe QDs dissolved in CDCl3. Much of the signal from X-TP protons disappears in the presence of QDs; the residual signal in the QD sample (with a linewidth equal to that of the peak for freely diffusing X-TP) corresponds to unbound X-TP molecules within the sample. We observe no additional broad signals from bound X-TP. The Supporting Information (Figure S1) contains 1H NMR spectra for all of the free X-TP molecules for more convenient comparison, as well as a kinetic study confirming that the samples are equilibrated with respect to adsorption of each type of ligand to the QDs (Figure S2). For each QD/X-TP mixture, we calculate the number of X-TP bound per QD by subtracting the number of free X-TP per QD within the mixture (using the residual signal from X-TP within the NMR spectrum of the mixture) from the number of X-TP per QD that we add (using the signal from the spectrum of a QD-free standard sample with the same concentration of added X-TP). Figure 2F shows that, for the mixture of, for example, Cl-TP with QDs, the signal from aromatic protons on Cl-TP are absent at low concentrations of added Cl-TP because all of the


Page 11 of 32

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

Langmuir

1

Figure 2. A-E) H NMR spectra in the aromatic region of 2.4 mM X-TP in CDCl3 (black lines) and in mixtures with 9.2 μM CdSe QDs in CDCl3 (red lines). The spectra are to scale within each tile (the y-axes of all tiles are NMR intensity in arbitrary units), but not relative to each other, and are focused on the aromatic protons of free X-TP (blue shading), and those of the corresponding disulfides (red shading). Each X-TP has a pair of doublets from the two protons adjacent to the thiol and the two protons adjacent to the para substituent, with the exception of Cl-TP, for which the chemical shift of all four protons is identical. The 1H NMR spectrum of NO2-TP (which is purchased as technical grade, purity = 70%) has a sizeable impurity, which we believe is the Na+ NO2-TP- salt. Dashed brackets highlight the decrease in peak height upon addition of QDs to the free X-TP. The Supporting Information contains the spectra of the free X-TPs in direct comparison to one another. F) For each QD sample, we titrate increasing amounts of X-TP (shown here for Cl-TP) into a solution of QDs, and compare the integrated signal strength of the 7.21-ppm singlet (for Cl-TP, or doublet for the other X-TPs) to that for a QD-free solution of the same concentration of11 X-TP. We define the concentration of bound XTP as the difference between these integrated ACStwo Paragon Plus signals. Environment

Langmuir

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

Page 12 of 32

molecules bind to the QDs, and increases linearly as Cl-TP fills the binding sites on the QD surfaces. In the spectra of some mixtures, a pair of doublets appears at 7.27 and 7.28 ppm. This pair corresponds to bis(p-X-phenyl) disulfide.41 The conversion of X-TP to disulfide is not reversible under these conditions, so in cases where we observe the disulfide, we define the “amount of X-TP added” as the amount of X-TP added minus the amount of X-TP converted to disulfide. Figure 3A contains plots of the average number of adsorbed X-TP per QD, measured by 1H NMR as described above, versus the concentration of free X-TP in the sample. Figure 3B shows plots of ΔR, the change in the apparent excitonic radius of the QD upon ligand exchange from OPA to X-TP, measured from the absorption spectrum of the sample, versus the average number of X-TP bound per QD in the ensemble. Note that both Figure 3A and 3B have fewer data points in the plots for NH2-TP than in the plots for the other X-TPs because, at the high concentrations of NH2-TP required to observe appreciable binding of NH2-TP to the QDs, the ligand etches their surfaces and causes precipitation of the particles.


Page 13 of 32

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

Langmuir

Figure 3. A) Plots of the number of bound X-TP per QD versus the concentration of free X-TP, measured by 1H NMR. B) Plots of ΔR vs. the number of bound X-TP per QD. The solid lines correspond to fits of the data in A to the two-site Langmuir equation, and the data in B to eq 3, where the values of all fitting parameters are shared across both datasets. The data for NH2-TP are not included in the fits since they do not span an adequate range of surface coverages. Uncertainties in [X-TP]Bound (

Description of the Adsorption and Exciton Delocalizing Properties of p

Recommend Documents