Imaging Quantum Dots on Buckyballs and Nanotubes
University of Rochester, Rochester, NY

by Caleb Whittier

OPT307/407 Spring 2016
Individual Project


Proposal
  1. Introduction
  2. Methods
Results
  1. Overview
  2. Electron Flight Simulation
  3. TEM Imaging
  4. X-Ray Microanalysis
  5. STEM Elemental Mapping
  6. Conclusions
  7. Summary






Proposal

1. Introduction

Quantum dots (QDs) are semiconducting nanocrystals with unique optoelectronic properties, with potential applications in LEDs and biological imaging. Carbon nanotubes (CNTs) are graphene layers rolled into a tube formation, and are well-known for their high tensile strength and nature as a lightweight material, lending them use in strengthening other materials. Buckyballs are hollow carbon spheres comprised of 60 carbons, and have not yet been widely implemented.

hhhh


Fig. 1. From left to right, see a core/shell quantum dot, carbon nanotube, and buckyball.

It is possible to use chemical techniques to connect QDs to both CNTs and buckyballs. By doing so, it may be possible to partially combine the properties of these structures. In this project, simplistic techniques were used to connect the structures, and resulting samples were imaged in the TEM to confirm the project's success.


2. Methods

CdSe/CdS core/shell QDs were obtained from Kelly Sowers of the Krauss Group. The QDs were diluted in toluene (~10mL), and a small amount (0.2mL) of them were were bath sonicated in the presence of single-walled CNTs or buckyballs (<1mg) in a solution of 1-methyl-2-pyrrolidone (~10mL) to break up large clumps of buckyballs/CNTs. The resultant sample was then probe sonicated to create full dispersion and connection between nanoparticles. Three drops of each sample were then placed on a gold TEM grid and allowed to air dry. Data collection was acquired on the University's TEM system, and electron flight simulations were performed on the SEM computer.




Results

1. Overview

Electron flight simulation was performed for CdSe and CdS to allow for a projection of interactions between the electron beam and sample. TEM imaging was performed to evaluate the chemical treatment. Using the knowledge that the QDs were ~7nm in diameter and that the buckyballs were ~20nm in diameter, areas were searched for showing a connection between particles of roughly these sizes. For QD/CNT samples, areas were searched for that showed QDs attached to tubular structures, and lattice was searched for in the tubes to differentiate between organic matter and CNTs. Images were colorized using the ImageJ program. X-Ray microanalysis was performed to confirm the chemical nature of the samples. The TEM's STEM setting was utilized to allow for elemental mapping of the QD and buckyball sample.

2. Electron Flight Simulation

Electron flight simulation shows how the electrons scatter through the sample upon entrance, and as such helps to understand how certain microscope conditions affect the imaging. Bars on the right side represent x-ray emissions from the sample’s chalcogenide component (Se or S).

CdSe E flight

Fig. 2. Electron flight simulation of a spherical CdSe particle atop bulk CdSe.


CdS E flight
Fig. 3. Electron flight simulation of a spherical CdS particle atop bulk CdS.

Due to the sample's nature and the utilized imaging techniques, the resultant electron flight simulations required some tweaking. For example, the spherical particles could not be simulated at their proper size, as it would be too small for the simulator to yield legible data for. Furthermore, if using a 200kV beam in the simulator as was used in the TEM, the resultant data barely showed the particle. In order to receive some data rather than no data, parameters were adjusted to be a 20kV beam in the simulator, a 70nm CdSe particle (Fig. 2), and a 700nm CdS particle (Fig. 3). These values were chosen as they were different from the actual values by only an order of ten (in the case of CdS, an order of 100).

3. TEM Imaging

TEM was utilized at 200kV for all samples, which were deposited on gold TEM grids. Samples imaged were a composite of QDs and buckyballs, as well as composites of QDs and CNTs. Images were colorized using the ImageJ program.

QD BB OverheadQD BB Overhead Zoomed 

Fig. 4. Overhead of a conglomerate of QDs and buckyballs, and a closeup image of a QD on a buckyball found on the conglomerate's edge. 


QD BB Zoomed In QD BB Closeup 

Fig. 5. QD connected to a buckyball, and a closeup of this region to show detail. 

As is seen in Fig. 4, QDs and buckyballs tend to aggregate together into large clumps, making it necessary to skirt the edges to find clear images. The zoomed in image of this region shows some areas where there may be a dot on top of a buckyball, but overlaying of other buckyballs and dots makes it challenging to evaluate these areas, aside from the clear portion hanging off the edge. Fig. 5 shows an ideal case of this, where the aggregate has a buckyball-QD connection clearly jutting out from the edge.


CdSe QD on NTs

Fig. 6. CdSe/CdS core/shell QDs on CNTs. 


PbS QD on NT

Fig. 7. PbS QDs on a CNT.

Fig. 6 yields an image of several CdSe/CdS QDs on top of an array of CNTs. An issue similar to the QD/buckyball sample arises, where it is easiest to evaluate results when some particles are jutting off the edge. This is more prevalent in Fig. 7, where many small PbS QDs can be seen attached to the CNT. Despite not being properly focused, the image still shows several interesting results. For example, the CNT has several areas where dots seem to be hanging off the edge. Furthermore, if looking at the top right corner of the image it can be seen that some dots might be inside the tube.


4. X-Ray Microanalysis

X-Ray data allows us to discern the elements present in the samples, which assists in supporting the theory of dots connected to CNTs or buckyballs.

EDS of overhead region

Fig. 8. X-Ray data corresponding to the overhead image in Figure 4. 



EDS of zoomed in region 

Fig. 9. X-Ray data corresponding to the zoomed in image in Figure 5.

Peaks on the leftmost side of each graph are expected to be carbon. Ni peaks stem from leftover Ni catalyst when using the QDs for hydrogen generation. In the data from Fig. 8, it can be seen that there is a significant Ni peak, and much smaller S and Cd peaks. The seemingly large Se peak in Fig. 8 is actually a shoulder on an Au peak, which is generated by the grid. This data shows that while there is a significant amount of Ni catalyst remaining in the conglomerate, there is also a very large amount of carbon and some CdSe/CdS observed. In Fig. 9, a much bigger S peak is seen. This implies that the zoomed in area does truly relate to CdSe/CdS, with a non-negligible carbon peak. Such data points towards a connection having formed between the QD and the buckyball, as the area of interest yields data expected from both QDs and buckyballs.


5. STEM Elemental Mapping
Raw STEM Image Ni-K Cd-LSe-K S-K

Fig. 10. STEM elemental composition map of a QD on buckyball sample. From left to right, images correspond to the raw STEM image, Ni, Cd, Se, and S.

STEM mapping shows the elemental composition of a QD on buckyball sample. By showing these images side-by-side, it is possible to see which areas of the sample are comprised of which element. The brightest regions correspond to Ni, Se is found throughout the sample, and Cd and S are concentrated on the fringes. Although the Se appears throughout the entire sample, this may be a misattribution of Au, which overlaps with Se x-ray emissions. Rather than the data showing several QDs attached to one another, it instead implies that the QDs are attached to some other surface. This shows that the QDs are not connected to one another, but are more likely attached to buckyballs instead.


6. Conclusions

Results confirm that QDs could be attached to buckyballs and NTs using this simplistic method of sonication, albeit to a very small yield only useful for imaging purposes. TEM imaging yields multiple locations in the samples where connections occur. It is probable that there are more connections than those observed, but these likely appear within the formed conglomerates. If connections are formed in the center of the conglomerates, it becomes impossible to make concrete conclusions regarding the chemical treatment, and as such data is limited to the outer fringes of the aggregates. In order to more efficiently create nanocomposites of this nature, further chemical processes need to be performed, which could potentially result in more connections and less aggregated samples.

7. Summary

QDs are connected to buckyballs or CNTs using the simple method of sonication. Results are monitored via TEM, which shows that connections between the nanoparticles do occur. Electron flight simulation, x-ray data, and STEM elemental mapping are used as supporting information for this conclusion.


Acknowledgments

Thank you to Kelly Sowers for providing CdSe/CdS core/shell QDs, Amanda Amori for assisting with sonication, Amanda Preske for providing guidance in PbS synthesis and chemistry in general, and the entire Krauss Lab for their support. Thank you to Hanyuan Zhang for being a patient TA, and to Brian McIntyre for all of his assistance in using the TEM and his guidance in this project.





Please contact me if you have any questions, criticism or suggestions:

Caleb Whittier
Department of Chemistry
University of Rochester
Rochester, NY 14627
cwhittie@u.rochester.edu

Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address: