Project topics and materials
Get Complete Project Material File(s) Now! »
Synthesis of CdSe/CdS Quantum Dots
Several synthesis protocols have been developed over the last twenty years to grow stable colloidal Quantum Dots with low size dispersion, narrow emission band and high quantum yield.
Setup used for the synthesis
The synthesis of colloidal Quantum Dots is performed in a round-bottomed flask in organic solvents. Because of the low reactivity of the precursors, the reaction is carried out at high temperature (~300°C); and the temperature needs to be controlled, in particular to separate efficiently the nucleation and the growth of the NCs (see I.4). To limit the oxidation of the QDs during the synthesis, the reaction is performed under inert atmosphere (argon gas) after evacuating oxygen and water from the medium and the solvent by degassing under vacuum. Different protocols are available for the synthesis of NCs. For the protocols requiring an injection of one or several precursors, a syringe can be used, whose injection rate can be controlled by a syringe pump if necessary. The setup is illustrated in Figure II.1.
Characterization methods
Absorption
Absorbance measurement is an easy way to obtain crucial information about a colloidal suspension of NCs (Figure II.2). It gives the excitonic structure of the Quantum Dots, allows a rough estimation of the emission wavelength, informs on the amount of secondary nucleation during the reaction thanks to a comparison with photoluminescence excitation spectrum (PLE, see II.2.3), but can also be used to extract information on the size and concentration of the QDs in the suspension.
The determination of the size and the concentration was developed by Leatherdale et al.134 and Yu et al.135, and further confirmed by other works.136,137 The analysis of transmission electron microscopy images (see II.2.5) and the correlation to the position of the first excitonic peak yields a formula that gives the size of the QD:
= (1.6122 × 10−9) 4 − (2.6575 × 10−6) 3 + (1.6242 × 10−3) 2 − 0.4277 + 41.57
where is the diameter of the QD and the wavelength at the first excitonic peak, both in nm. This equation is valid for CdSe QDs, but similar equations can be found for CdS and CdTe.135,138
A similar approach leads to the molar extinction coefficient dependence with size. However, when studied at the position of the first excitonic peak, quantum confinement effects require taking into account the effective band-gap of the material. Leatherdale et al. worked at high excitation energies, far from the band edge, at 350 nm to ensure that the quantum confinement effects are negligible: the material behaves as if it was bulk and the linear absorption coefficient, which defines the extent of light absorption through the material, is independent of its size. The particle cross-section ( ) or the molar extinction coefficient ( ) can further be calculated:
(CdSe at 350 nm) = 1.438 ∙ 1026 ∙ 3 M−1 ∙ cm−1
(CdSe at 350 nm) = 5.501 ∙ 105 ∙ 3 cm2
where is the particle radius in cm.
The concentration C (in M−1 = mol ∙ l−1) of CdSe QDs can therefore be estimated using Beer-Lambert’s law where is the absorbance and the length of the optical path length in the absorbing medium (width of the measurement cuvette, in cm)
With the position of the first excitonic peak in the absorbance spectrum, and the absorbance at 350 nm, a simple absorbance spectrum gives the size and the concentration of the CdSe particles in colloidal suspension.
Fluorescence emission
Fluorescence measurement is also an easy-to-implement characterization to follow the evolution of nanocrystal growth during the synthesis. Several information can be extracted from a fluorescence spectrum, otherwise called photoluminescence (PL) spectrum (Figure II.2).
First, the relative intensity collected on a PL spectrum informs on the quantum yield of the QDs. Indeed, the comparison, at the same relative absorption, between the emission intensity of a sample of QDs and that of a reference sample of known QY allows to determine the QY of the QDs (see II.2.3).
As seen previously (see I.3.1.2), the QDs have a narrow emission band, typically around 20 to 30 nm of full-width at half maximum (FWHM). Thus, the broadness of the band can inform on the inhomogeneity of the NC growth. Indeed, the FWHM can be correlated with the size dispersion of the sample. The inhomogeneities can come from the presence of different sizes of QDs or from the heterogeneous growth of a shell of core QDs. The larger the size dispersion, the broader the band. This correlation is however limited as it assumes the NCs of different sizes have the same QY.
The appearance in the spectrum of another emission peak than the one originating from the band edge recombination informs on the existence of either secondary nucleation or emission from traps. Typically, secondary nucleation occurs when the synthesis parameters are not optimized and when the precursors that should deposit on existing particles and make them grow, nucleate in solution and form new, smaller seeds, emitting at a lower wavelength. Emission from traps can be seen mostly in the case of the smallest NPs for which the surface-to-volume ratio is the highest: not all the surface bonds are well passivated, forming traps for the charge carriers within the band gap, which yields emission at lower energy. Those peaks can be seen on the PL spectrum only if the secondary nuclei are fluorescent or the recombinatons from traps is radiative.
The PL spectra are acquired on a F900 spectrometer from Edinburgh Instruments. It is equipped with several detectors: the R928P in the 200-870 nm region; an MCP-PMT in the same region, but with a response time ~12 times faster; the R2658P for the NIR (200-1010 nm); and a NIR-PMT up to 1700 nm. For CdSe/CdS QDs, the R928P provides the good spectral range, adequate sensitivity and a response time fast enough (600ps) for typical CdSe/CdS NCs lifetime measurements.
Determination of the Quantum Yield
As presented previously (see I.3.2.4), the quantum yield can be defined at the single particle level, but it can also be measured on an ensemble of NCs. The idea here is to compare, for the same relative absorption, the number of photons emitted by the sample of interest (QD) and by a reference sample of already known QY. The reference needs to be carefully chosen: its emission wavelength needs to be close to the emission wavelength of the studied QDs, and it has to absorb at the same wavelength (which should not be a problem for QDs as they have a wide range of absorption wavelengths). For CdSe/CdS Quantum Dots, we usually use Rhodamine 6G, which emits around 550 nm and has a QY of 95% in ethanol.
To measure the quantum yield, a series of dilute solutions are prepared (absorbance at the excitation wavelength < 0.1 to avoid reabsorption effects). Their absorbance and PL are measured, and the integrated PL are plotted vs the absorbance. This curve can be fitted by a straight line whose slope is then used to determine the QY where is the QY of the reference used, and are the slopes of the integrated PL vs absorbance curves respectively for the sample and for the reference, and and are the refractive indices of the solvents used respectively for the sample and for the reference.139
The precision of this ensemble quantum yield measurements is relatively low and a 10% error is commonly accepted. It represents an average over the whole sample, including the low-QY, or non-emitting nanoparticles if present, and consequently leading to an underestimation of the QY of the brighter NCs.
Excitation (PLE)
The excitation (or Photoluminescence Excitation, PLE) spectrum is based on the detection of the PL intensity at a given wavelength with a changing excitation wavelength (Figure II.2). Given the fact that the PL efficiency, or QY, for CdSe NCs is independent from the excitation wavelength,140 this allows to probe the absorbance of one population of NCs, the one that emits at the selected wavelength. If no secondary nuclei are present, the absorption is not increased by NCs that absorb but do not emit, or emit at a different wavelength, and the PLE and absorbance spectra should be very similar. Besides, as only one population of QDs is probed, the excitonic peak on the PLE spectra is narrower than the one on the absorbance spectra: the size dispersion is masked by the fact that only one wavelength is observed.
PLE is therefore a good tool to investigate the growth of QDs and to verify the absence of secondary nucleation (Figure II.3). But it is also useful to monitor the growth of a shell. For example, if CdS is grown on CdSe, due to its larger band-gap (around 515 nm), the intensity of the PLE spectrum below 515 nm will increase showing that the CdS shell absorbs and is involved in the emission at the probed wavelength.
Finally, the narrow peak observed on the PLE spectrum is due to the diffusion through the sample of the excitation beam when its wavelength is the same as the probed one, i.e. the wavelength at which the detector is set.
Transmission Electron Microscopy
The structural characterization of QDs can be performed by transmission electron microscopy (TEM, also Transmission Electron Microscope). This technique allows to directly visualize the nanoparticles and to get information on their size, size distribution, morphology, crystallinit, etc. (Figure II.4)
When accelerated under 200 kV, electrons can be transmitted through a thin sample. When they hit the detector, they produce an image whose contrast is correlated with the atomic mass (and therefore with the electron density) of the elements in the material and with its thickness. The resolution of a TEM is around 1 nm, or even a few tenth of nanometer: individual atoms and their organization in the crystal can be seen on high-resolution images (magnification of 500,000 to 1,000,000). Finally, TEM allows to carry out electronic diffraction which is particularly useful to determine the crystal structure of the QDs.
The sample is prepared as follows: several washing steps in ethanol are performed to get rid of the excess of organic ligands in solution (they do not stand a long exposure to the electron beam, and degrade rapidly decreasing the contrast of the image). The QDs are then redispersed in hexane at a concentration of around 5 µM and a 10-µL drop is deposited on a standard carbon grid. The grid is left to dry at atmospheric pressure and then put under vacuum overnight to remove as much ligands and solvent as possible.
The observations are performed on a Jeol 2010-F for the TEM or on a FEI Titan Themis for the Scanning Transmission Electron Microscopy (STEM) in collaboration with Gilles Patriarche from Laboratoire de Photonique et de Nanostructures, Marcoussis.
Energy dispersive X-ray spectrometry (EDX)
Elemental composition of the sample can be performed thanks to energy dispersive X-ray spectroscopy, or EDX. When QDs are exposed to a high-energy electron beam (electrons are accelerated by a voltage between 5 and 20 kV), electrons from inner atomic energy levels can be ejected out of the electron cloud. For an atom to relax, an electron from a higher energy level goes down in energy to take the place of the ejected electron, giving its excess energy to the environment in the form of a photon in the X-ray range. This energy is characteristic of the atom; thus, by detecting the number and the energy of all these emitted X-ray photons, the elemental composition of the sample can be deduced.
The EDX experiments were performed in collaboration with Gilles Patriarche from Laboratoire de Photonique et de Nanostructures, LPN, Marcoussis, France, on a Titan Themis.
X-ray Diffraction (XRD)
To determine the crystal structure of the synthesized NCs, X-Ray Diffraction (XRD) is a particularly useful technique. Most commonly produced by the Kα emission of copper at a wavelength of 1.54 Å, the X-rays are diffracted in different directions by the electrons of the atoms that compose the crystal. In the powder XRD method, the incident beam hits an ensemble of randomly-oriented nanocrystals, and a detector collects the diffracted signal at a 2 angle compared to the incident beam. The obtained diffractogram presents peaks at given angles that are characteristic of the crystal structure of the material. By comparing to the diffractograms present in the crystallographic databases, the crystal structure of the studied sample can be determined.
This however requires the crystals to be large enough. Indeed, the width of the peaks on the diffractogram can be related to the size of the crystals through the Scherrer formula:0.91 (2 ) = cos( ) where is the full-width at half maximum of the peak, is the incident beam wavelength (which in our case corresponds to the Kα emission line of copper, = 1.5418 Å) , is the average size of the crystals and is the diffraction angle. The smallest the particles, the largest the FWHM. Zinc-blende (ZB) and wurtzite (W) structures, the two possible structures for CdSe and CdS crystals, have characteristic peaks at 2 angles close to each other, which makes it difficult to distinguish between small (< 3 nm) ZB or W nanocrystals as the peaks are broad for such small sizes.
A typical sample is prepared by drop-casting several drops of a concentrated dispersion of QDs on a silicon wafer. The diffractograms are acquired on a Philips X’Pert diffractometer.
Table of contents :
Introduction
Chapter I. Colloidal semiconductor nanocrystals – Physics and chemistry of Quantum Dots
I.1. Introduction
I.1.1. Description
I.1.2. Crystallinity
I.1.3. Historical background
I.2. Electronic properties of Quantum Dots
I.2.1. Bulk material
I.2.2. Quantum confinement theory
I.2.2.1. Different confinement regimes
I.2.2.2. Particle-in-a-sphere model
I.2.2.3. Optimization of the model
I.2.3. First excited state : 1Se – 1S3/2
I.2.3.1. Degeneracy of first excited state
I.2.3.2. Dark and bright excitons
I.3. Optical properties of Quantum Dots
I.3.1. On an ensemble of Quantum Dots
I.3.1.1. Absorption
I.3.1.2. Fluorescence Emission
I.3.2. On a single Quantum Dot
I.3.2.1. Creation of an exciton and thermal relaxation of charge carriers
I.3.2.2. Exciton recombination and lifetime
I.3.2.3. Auger recombination
I.3.2.4. Quantum yield
I.4. Synthesis of colloidal semiconductor nanocrystals
I.4.1. Nucleation-growth – LaMer theory
I.4.2. Ostwald ripening
I.4.3. Mechanistic approach
I.4.4. Core/shell system
I.4.4.1. Advantages of a core/shell system
I.4.4.2. Different types of core/shell structures
I.4.4.3. Band-gap engineering
I.4.5. Synthesis methods
I.4.5.1. Synthesis of the cores – with precursor injection
I.4.5.2. Synthesis of the cores – one pot
I.4.5.3. Synthesis of the shell – SILAR
I.4.5.4. Synthesis of the shell – dropwise addition
I.5. Blinking of the Quantum Dots emission
I.5.1. The blinking phenomenon
I.5.2. Quantum Dots blinking
I.5.2.1. Characterization of blinking
I.5.2.2. Causes of blinking
I.5.3. Reduction of blinking
I.5.3.1. Addition of molecules
I.5.3.2. Compositions gradient between the core and the shell
I.5.3.3. Thick-shell Quantum Dots
I.6. Two-dimensional systems – Nanoplatelets
I.6.1. Nanoplatelets : atomically flat nanocrystals
I.6.2. Quantum wells & unique optical features
I.7. Conclusion
Chapter II. Synthesis of CdSe/CdS Quantum Dots
II.1. Setup used for the synthesis
II.2. Characterization methods
II.2.1. Absorption
II.2.2. Fluorescence emission
II.2.3. Determination of the Quantum Yield
II.2.4. Excitation (PLE)
II.2.5. Transmission Electron Microscopy
II.2.6. Energy dispersive X-ray spectrometry (EDX)
II.2.7. X-ray Diffraction (XRD)
II.2.8. Magneto-optical measurements
II.3. Synthesis of CdSe cores
II.3.1. Preparation of precursors
II.3.2. Zinc-blende CdSe cores
II.3.2.1. One-pot synthesis
II.3.2.2. Injection synthesis
II.3.3. Wurtzite CdSe cores
II.4. Synthesis of CdSe/CdS Quantum Dots – Dropwise addition
II.4.1. Growth of a thick CdS shell on CdSe one-pot zinc-blende cores (studied in III.2):
II.4.2. Growth of a thick CdS shell on CdSe injection-synthesized zincblende cores:
II.4.3. Growth of a thick CdS shell on CdSe wurtzite cores:
II.4.4. Growth of a composition gradient and thick CdS shell on CdSe wurtzite cores:
II.5. Summary of the Quantum Dots synthesized for the study
II.6. Conclusion
Chapter III. CdSe/CdS Quantum Dots with 100% quantum yield in air, at room temperature
III.1. Setup used for spectroscopic studies
III.1.1. Epifluorescence microscope
III.1.2. Confocal microscope and acquisition setup
III.1.3. Time-resolved acquisitions
III.1.3.1. Photoluminescence trace
III.1.3.2. Photoluminescence decay
III.1.3.3. Photoluminescence intensity autocorrelation
III.1.3.4. Preparation of the sample for single particle studies
III.2. Thick-shell CdSe/CdS Quantum Dots
III.2.1. Zinc-blende CdSe core with a thick CdS shell
III.2.1.1. Optical and Structural characterization
III.2.1.2. Two emissive states at room temperature
III.2.1.3. One grey state under vacuum
III.2.1.4. 100% quantum yield at cryogenic temperature
III.2.1.5. Thermal activation of Auger processes
III.2.2. Other samples
III.2.3. Conclusion
III.3. Bulky-gradient QDs – thick-shell and gradient composition
III.3.1. Optical and Structural characterization
III.3.2. Correlative optical/electron microscopy
III.3.3. 100% quantum yield in air, at room temperature
III.3.3.1. Low excitation regime
III.3.3.2. 100% quantum yield
III.3.3.3. High excitation power
III.3.4. Evolution under vacuum
III.3.5. Quantum yield of the biexciton
III.3.6. White-light emitting Quantum Dots
III.3.7. Lifetime measurements
III.3.8. Evolution at cryogenic temperature
III.3.9. Magneto-optical measurements
III.3.10. Slow recombination dynamics
III.3.11. Amplified spontaneous emission
III.3.12. Conclusion
III.4. Golden-QD – hybrid Quantum Dots/gold nanoshell nanoparticles
III.4.1. Surface plasmons and Purcell effect
III.4.2. Description of Golden-QDs and synthesis
III.4.4. First observations with thin-shell CdSe/CdS QDs
III.4.5. Golden-QDs with thick-shell CdSe/CdS QDs
III.4.6. Effect on blinking behavior
III.4.7. Increased photostability…
III.4.7.1. …with time
III.4.7.2. …with power
III.4.8. Correlative light-electron microscopy
III.4.9. Conclusion
III.5. Conclusion
Chapter IV. Spectroscopic studies of thick-shell CdSe/CdS nanoplatelets
IV.1. Synthesis of nanoplatelets
IV.1.1. Synthesis of CdSe NPLs
IV.1.2. Synthesis of CdSe/CdS core/shell NPLs
IV.2. Optical properties of CdSe nanoplatelets
IV.2.1. Photoluminescence
IV.2.2. Time-resolved photoluminescence
IV.2.3. Blinking behavior
IV.3. Optical properties of CdSe/CdZnS core/shell nanoplatelets synthesized at room temperature
IV.3.1. Photoluminescence
IV.3.2. Time-resolved photoluminescence
IV.3.3. Blinking behavior
IV.4. Optical properties of new generation of core/shell nanoplatelets
IV.5. Conclusion
Chapter V. Quantum Dots as probes for biological imaging
V.1. Fluorescent probes for biology
V.2. Quantum Dots vs organic fluorophores
V.2.1. Photobleaching
V.2.2. Excitation and emission ranges
V.2.3. Lifetime
V.2.4. Functionalization and surface chemistry
V.2.5. Multimodal imaging
V.2.6. Cytotoxicity of Quantum Dots
V.3. Ligand exchange on the surface of Quantum Dots
V.3.1. Necessity of a ligand exchange
V.3.2. Choice of proper ligand
V.4. Targeting of Voltage Dependent Calcium Channels in C.elegans
V.4.1. C.elegans – anatomy and interest
V.4.1.1. A model organism…
V.4.1.2. … widely used in biology
V.4.2. Voltage Dependent Calcium Channels in C.elegans
V.4.2.1. Genetic modification of C. elegans
V.4.2.2. Quantum Dots as bimodal probes for targeting of VDCCs
V.4.2.3. Quantum Dots for biological applications
V.4.3. New polymeric ligand
V.4.4. Preparation of antiGFP-QDs
V.4.4.1. Ligand exchange on the QDs
V.4.4.2. Coupling of anti-GFP antibody to an amino group
V.4.4.3. Coupling of anti-GFP antibody to a carboxylic group
V.4.4.4. Coupling of anti-GFP antibody to a thiol group
V.4.4.5. In vitro evaluation of the coupling specificity of the antiGFP-QDs
V.4.5. Microinjection of antiGFP-QD
V.4.6. High-pressure freezing
V.4.7. Macroscopic targeting in living worms
V.4.8. Nanometer resolution analysis
V.4.9. Conclusion
V.5. DNA nanocage as a functional biocompatible scaffold
V.5.1. DNA nanocage: a versatile scaffold
V.5.2.1. Preparation of Quantum Dots
V.5.2.2. Encapsulation of QDs in the nanocage
V.5.2.3. Verification of encapsulation
V.5.3. Monofunctionalization of DNA icosahedron
V.5.4. Probing the Shiga toxin endocytosis pathway
V.5.5. Conclusion
Conclusion
References
