Binary superlattices of semiconductor nanocrystals : a path towards possible high-temperature superconductivity

This thesis provides a platform to experimentally test Dr. Daniel Mattis' theoretical work on the possible superconductive behavior of nanostructured intrinsic semiconductors1-2. The theoretical work hinged on the nano-scale architecture of semiconductors. Therefore, using various types of semiconductor nanocrystals, i.e., quantum dots, to correspond to Dr. Mattis' single electron models, nanoscale structures were formed by self-assembly methods based on the architecture suggested. In detail, various nanocrystal lattices and binary superlattice combinations were studied. In this work, superlattice combinations of palladium (Pd), gamma iron (III) oxide (y-Fe2O3), lead selenide (PbSe) and cadmium selenide (CdSe) nanocrystals were studied as possible material platforms. These nanocrystals were chosen based on their properties as well as their reported ability to form the desired superlattices. The nanocrystals were synthesized through various metal-organic colloidal nucleation-and-growth-based methods. The superlattice formation, in terms of ordering range and uniformity, was studied by transmission electron microscopy. The best results were obtained with the combination of cadmium selenide and lead selenide nanocrystals in an approximate 12 to 1 molar ratio. The optimization of this superlattice then allows for a sufficiently adequate experimental evidence for Dr. Mattis' theory.

BINARY SUPERLATTICES OF SEMICONDUCTOR NANOCRYSTALS: A PATH TOWARDS POSSIBLE HIGH-TEMPERATURE SUPERCONDUCTIVITY by Carlos Alberto Burga A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment of the Requirements for the Honors Degree in Bachelor of Science In Chemistry Approved: Dr. Michael H. Bartl Supervisor Dr. Henry S. White Chair, Department of Chemistry Dr. Thomas G. Richmond Department Honors Advisor Dr. Sylvia D. Torti Dean, Honors College May 2013 ABSTRACT This thesis provides a platform to experimentally test Dr. Daniel Mattis’ theoretical work on the possible superconductive behavior of nanostructured intrinsic semiconductors1-2. The theoretical work hinged on the nano-scale architecture of semiconductors. Therefore, using various types of semiconductor nanocrystals, i.e., quantum dots, to correspond to Dr. Mattis’ single electron models, nanoscale structures were formed by self-assembly methods based on the architecture suggested. In detail, various nanocrystal lattices and binary superlattice combinations were studied. In this work, superlattice combinations of palladium (Pd), gamma iron (III) oxide (y-Fe2O3), lead selenide (PbSe) and cadmium selenide (CdSe) nanocrystals were studied as possible material platforms. These nanocrystals were chosen based on their properties as well as their reported ability to form the desired superlattices. The nanocrystals were synthesized through various metal-organic colloidal nucleation-and-growth-based methods. The superlattice formation, in terms of ordering range and uniformity, was studied by transmission electron microscopy. The best results were obtained with the combination of cadmium selenide and lead selenide nanocrystals in an approximate 12 to 1 molar ratio. The optimization of this superlattice then allows for a sufficiently adequate nanostructure to be formed and will therefore be the starting point for future work on experimental evidence for Dr. Mattis’ theory. ii ACKNOWLEDGMENTS I would like to thank my advisor Dr. Michael H. Bartl for giving me the opportunity to work in his research group and for introducing me to the wonderful and fascinating world of nanoscience. His guidance, support and encouragement have been invaluable to the development and progress of this project. I would also like to thank Julian Cedric Porsiel; a graduate student from the Braunschweig University of Technology who did an internship in Dr. Bartl’s group and who worked with me at the beginning of this project. His help in the early stages of the research were vital and helped define the overall direction of the project. I would be remiss not to thank the amazing members of the Bartl group, whose helpful discussions allowed me overcome the difficulties inherent to any research endeavor. Special thanks go to Michael R. Dahlby, a graduate student in the Bartl group, who was invaluable in his assistance with the transmission electron microscopy and in his encouragement of the project; to Dr. Jacqueline T. Siy-Ronquillo, former member of the Bartl group, whose expertise in cadmium selenide nanocrystals was incredibly helpful; and to Adam Hollerbach from the Muller group for his help with infrared spectroscopy. Finally, I want to thank my family for their support and encouragement. My parents and two sisters have always been a source of support for me and I can honestly say that they have, on more than one occasion, been single handedly responsible for keeping me going. Lastly, I want to thank the newest member of my family, my niece, who without knowing it has brightened some of my most despairing days. All of you have my utmost gratitude. lll TABLE OF CONTENTS Abstract...................................................................................................................ii Acknowledgements.............................................................................................. iii Introduction............................................................................................................ 1 Experimental Section............................................................................................. 4 Results & Discussion.......................................................................................... 14 Conclusions..........................................................................................................25 References............................................................................................................26 iv 1 1. INTRODUCTION Since the discovery of superconductivity there has been a great deal of effort put forth to develop materials that allow for superconductive properties to manifest at more manageable temperatures (i.e. temperatures around liquid nitrogen, 77 K, instead of around liquid helium, 4.2 K). Although some extremely interesting materials have indeed been made that allow for superconductivity at temperatures around liquid nitrogen, these materials tend to be highly complex cuprate-based compounds that require an assortment of other atoms to stabilize its structure and whose synthesis, unsurprisingly, is quite involved, making the need for simpler high-temperature superconductors quite significant. Thus, in 2006, Dr. Daniel C. Mattis, Emeritus Professor at the Physics Department of the University of Utah, published a theoretical paper demonstrating that semiconductors can exhibit anti-ferromagnetic, ferromagnetic and, more importantly for the present work, superconductive phases depending on the electron concentration in very specific nanoscale architectures1. He studied the electrons in the conduction band of semiconductors in the proposed geometry, an anti-dot lattice (Figure 1), and determined that several “sub-bands” formed from the main conduction band allowing for quite interesting properties to develop. 2 He d is c o v e r e d ^ A f tM latttcceirparaJ^ur^J?rojfo^ d ;b= D 'd M ^ also Figure 1), the lowest of the sub-bands formed have an electron probability distribution that rest primarily, and symmetrically, on the intersections between the holes in the lattice. This spatial distribution of the electrons in the nano-cell allowed for the semiconductor’s conductive properties to change significantly based on their electron density. Thus, he showed that at 1 electron per nano-cell the semiconductor is expected to exhibit an anti­ ferromagnetic insulating phase, which changes back to its normal conductivity at 2 electrons per nano-cell and finally changes to a ferromagnetic phase at 3 electrons per nano-cell. However, at somewhere between 1 and 2 electrons per cell, the semiconductor is expected to exhibit superconductive properties analogous to the high-temperature superconductors that have cuprate (CuO2) planes in their crystal structures (Figure 2). He discusses the symmetry of the geometry and electronic properties of his proposed anti-dot lattice and the CuO2 lattice, noting that the copper oxide planes also exhibit an anti­ ferromagnetic phase depending on the electron density on the CuO2 cells. 3 Figure 2. (Left) Crystal structure depiction of the YBa2Cu3O7high-temperature superconductor2. (Right) Isolated depiction of CuO2 planes in the superconductor2. In a subsequent and more detailed publication , Dr. Mattis was able to calculate the optimal parameters of the proposed anti-dot lattice and the resulting critical temperature for superconductive properties. He calculated that a lattice parameter of a = 15 nm would give rise to superconductive properties in the range of around 80 K, similar to the temperatures accomplished with more complex cuprate-based superconductors. He also states, however, that due to the assumptions made for the calculations, the optimal parameters could be closer to 10 nm, affording room temperature superconductive effects if all the other parameters of the lattice are also maintained in the nanometer range. Based on Dr. Mattis’ assumptions of electrons in independent particle models, the decision was taken to use semiconductor nanocrystals as starting materials. Semiconductor nanocrystals, i.e., quantum dots, have been used as independent particle models for a number of years now and their choice is vindicated by the properties that have been uncovered through their research4. Thus, several different semiconductor nanocrystals were utilized in the present project. These included the more recently developed gamma iron (III) oxide and palladium nanocrystals as well as the cadmium 4 selenide and lead (II) selenide nanocrystals, whose properties have been well characterized in the years since Dr. Mattis’ work. More significantly, these nanocrystals have also been reported to form self-assembling binary superlattices depending on the molar ratios and the coalescing conditions used5-6, lending themselves perfectly for the formation of the nano-scale architectures required by Dr. Mattis’ calculations. 2. EXPERIMENTAL SECTION 2.1 Syntheses of Nanocrystals 2.1.1 Palladium (Pd) The synthesis of palladium nanocrystals was adapted from work by the Murray group5. In a typical synthesis, palladium (II) chloride (25.4 mg, 0.14 mmol), toluene (10 mL, 99.8%) and didodecyldimethylammonium bromide, DDAB, (0.16 g, 0.34 mmol) were added to a three-neck round bottom flask. The solution was then sonicated for 2 hours or until all the black palladium chloride powder dissolved to yield a solution of a deep orange color. While the palladium precursor was being sonicated a fresh 9.3 M solution of sodium borohydride was made by dissolving 1.75 g (46.5 mmol) of sodium borohydride powder in 5 mL of deionized water. Once the palladium was completely dissolved in the toluene/DDAB solution, 4 p,L of the 9.3 M sodium borohydride solution was added while vigorously stirring the palladium precursor. Finally, after 20 minutes of 5 stirring - during which time the solution turned from a deep orange to an almost black color - dodecanthiol (10 mL, >98%) was added to solution. At this point the palladium nanocrystals were purified with ethanol (200 proof), which was added to the solution, allowing the palladium nanocrystals to precipitate out. The solution was then centrifuged and the supernatant was discarded. After repeating the purifying steps two more times, the still moist nanocrystals were fully dried in vacuum using a Schlenk line system (Figure 3). Once the palladium nanocrystals were completely dry, they were redispersed in a toluene/dodecanthiol solution. The resulting solution was then refluxed at 130 °C under a standing argon atmosphere for 30 minutes. Finally, the nanocrystals were purified with 2-propanol (>99.5%) via a method similar to the one described above for ethanol. After the second purification, the nanocrystals were redispersed in toluene. a Figure 3. Schematic representation of the apparatus used to synthesize nanocrystals7. 2.1.2 Gamma Iron (III) Oxide (Y-Fe2O3) 6 The synthesis for the y-Fe2O3 nanocrystals was adapted from the work by Hyeon o et al . In a typical synthesis, iron (0) pentacarbonyl (0.2 mL, >99.99%) was injected into 50 mL three-neck round bottom flask containing a solution of octyl ether (10 mL, 99%) and oleic acid (1.44 mL, 90%) that had been brought to 100 °C under argon atmosphere in a Schlenk line (Figure 3). The resulting yellow-orange solution was then brought to reflux at 2o5 °C via a heating mantle and was allowed to reflux at that temperature for 1 hour, causing the solution to turn black. After 1 hour of reflux the solution was allowed to cool to room temperature and trimethylamine N-oxide (0.38 g, 5.1 mmol) was added. The solution was subsequently heated to 130 °C under an argon atmosphere for 2 hours, after which time the solution turned a deep brown metallic color. The temperature was increased to 275 °C and the solution was allowed to reflux once more for 1 hour, returning the solution to its previous black color. At this point the solution was allowed to cool to room temperature before proceeding with the purifying steps. The solution was purified by precipitating the nanocrystals with ethanol (200 proof) and centrifuging the resulting mixture. The supernatant was discarded and the whole process was repeated twice more before finally redispersing the nanocrystals in hexane. A secondary procedure, also adapted from the Hyeon group8, was also used. In this procedure, iron (0) pentacarbonyl (0.2 mL, >99.99%) was injected into a 50 mL three-neck round bottom flask containing solution of lauric acid (0.95 g, 4.74 mmol), octyl ether (7 mL, 99%) and trimethylamine N-oxide (0.55 g, 7.32 mmol) at 100 °C under an argon atmosphere (Schlenk line). After the injection, which turned the clear solution a dark-red color, the temperature was increased to 120 °C and stirred at that temperature for 1 hour, during which time the solution turned black. After this time, the temperature was 7 increased to its reflux temperature, which after some difficulty was found to be 285 °C. The solution was then allowed to reflux for 1 more hour. After that the solution was allowed to cool to room temperature and was purified with ethanol (200 proof) with the method described above. The purified nanocrystals were then redispersed in toluene. 2.1.3 Lead (II) Selenide (PbSe) The synthesis of the PbSe nanocrystals proved to be quite exciting as it was observed that they not only form the commonly observed spherically-shaped crystals, but that they can also form nanocubes 10-11 and that the synthesis of the nanocubes was quite straight-forward. The synthesis for the spherical PbSe nanocrystals was adapted from the method described by the Murray group5, while, after serendipitously observing the existence of PbSe nanocubes, its synthesis was adapted from the work by Fang et.aln . In a typical synthesis of spherical PbSe nanocrystals, lead (II) acetate trihydrate (0.55 g, 1.45 mmol), oleic acid (1.8 mL, 90%) and squalane (12 mL, 99%) were added to a three-neck 50 mL round bottom flask and set up in a Schlenk line with a vacuum pump and an argon line. Based on a paper by Houtepen et al.10 on the role of acetate in forming the PbSe nanocubes, the flask was heated with an oil bath to 85 °C under vacuum for 1.5 hours to ensure the complete removal of the acetic acid formed in the solution. While the lead precursor dried, the selenium precursor, a 1 M trioctylphosphine selenide solution, was made by adding selenium powder (0.40 g, 5.0 mmol) to trioctylphosphine (5 mL, 90%) and vigorously stirring, sonicating and vortexing until all the black selenium powder dissolved in the clear solution. 8 Once the lead precursor was dried, the vacuum was turned off and a standing argon atmosphere was allowed to form throughout the Schlenk line. The temperature of the lead precursor flask was then raised to 180 °C using a heating mantle. Once the solution reached this temperature, the heating mantle was removed and 4.5 mL of the trioctylphosphine selenide solution was injected into the flask as the temperature dropped to 170 °C. The solution was then vigorously stirred for approximately 3 minutes and promptly cooled to room temperature by immersing the outside of the reaction flask into a water bath. This prevented the lead selenide nanocrystals from forming bigger clusters. Although the synthesis of the PbSe nanocubes is similar in the composition of the precursors, it differs vastly in the way it brings them together. In a typical synthesis of nanocubes, lead (II) acetate trihydrate (1.1 g, 2.9 mmol), oleic acid (3 mL, 90%) and phenyl ether (15 mL, 99%) were added to a three-neck 50 mL round bottom flask attached to a Schlenk line and heated to 160 °C for 30 minutes under an argon atmosphere. While the lead precursor was heating, a 1 M trioctylphosphine selenide solution was made by the same method described above for the spherical PbSe nanocrystals and another two-neck 50 mL flask with just phenyl ether (15 mL, 99%) was set up in the Schlenk line and heated to 230 °C under argon atmosphere with a heating mantel. After holding the lead precursor at 160 °C for 30 minutes, it was allowed to cool to 40 °C and 4.0 mL of the selenide precursor were injected into it under vigorous stirring. From this solution, 10 mL were taken and rapidly injected into the 230 °C phenyl ether flask. Subsequent 2 mL portions of the lead selenide low-temperature solution were then injected every 5 minutes to the phenyl ether flask for a total of 5 injections and a 9 total growth time of 25 minutes at ~ 230 °C. The solution was then allowed to cool to room temperature by itself, providing slightly more time for growth as it cooled. Once at room temperature, the nanocrystal solutions, both the spherical and the cubic PbSe, were purified. The spherical nanocrystals were precipitated by adding a solution of 1:3 hexanes (95%) to ethanol (200 proof), while the nanocubes were precipitated with just ethanol (200 proof). To facilitate the precipitation the mixtures were vortexed. The mixtures were then centrifuged and the supernatants discarded. This process was repeated twice more for both mixtures and the nanocubes were redispersed in toluene while the spherical nanocrystals were dispersed in tetrachloroethylene (+99%). 2.1.4 Cadmium Selenide (CdSe) The synthesis of the CdSe nanocrystals was adapted from methods described by Donega et al15 and by the Peng group16. In a typical synthesis, cadmium acetate dihydrate (80 mg, 0.30 mmol) and stearic acid (3 g, 0.01 mol) were added to a three-neck 50 mL round bottom flask to make the cadmium precursor. The flask was set up in the same Schlenk line as described above. The cadmium precursor was then heated under vacuum to 130 °C by means of an oil bath for approximately 40 minutes. This ensured the complete removal of all the water in the solution. Subsequently, the vacuum was turned off and the solution was kept under an argon atmosphere. The reaction was left at 130 °C for another 30 minutes. Lastly, the cadmium precursor was removed from heat and allowed to cool to room temperature. This mixture was then kept under in the argon 10 atmosphere for at least 12 hours to release any tension build up in the alkyl chain ligands16. The selenide precursor was prepared by adding selenium powder (80 mg, 1.01 mmol), trioctylphosphine (4.5 mL, 90%) and toluene (0.4 mL, 99.8%) into a two-neck 25 mL round bottom flask under an argon atmosphere. The solution was then stirred vigorously for at least 30 minutes, or until the black selenium powder was no longer visible in the clear solution. While the selenide precursor stirred, trioctylphosphine oxide (3 g, 7.7 mmol) was added to the cadmium precursor and the solution was heated via a heating mantle to 330 °C under an argon atmosphere. Finally, after the cadmium precursor reached 330 °C the heating mantled was removed and the selenide precursor (3.5 mL) was injected when the temperature dropped to 300 °C. The cadmium selenide solution was allowed to cool to room temperature without any heating after the injection. Once cool, the nanocrystals were precipitated out of solution with ethanol (200 proof) and centrifuged to pack the nanocrystals to the bottom of the centrifuge tube. Once the nanocrystals were centrifuged, the supernatant was discarded and the purifying process was repeated two more time before finally redispersing the nanocrystals in tetrachloroethylene (+99%). 2.2 Binary Nanocrystal Superlattices The assembly of binary superlattices was adapted from the work by the Murray and Vanmaekelbergh groups5-6. In a typical synthesis, the two nanocrystals were brought together in a ratio of between 10 and 20-fold excess of the smaller one to the larger one, in order to correspond to the geometric studies performed on the superlattices5-6. In order 11 for the nanocrystals to form a superlattice, there needs to be a driving force that brings them together while doing so slowly enough that they are able to arrange themselves instead of simply aggregating into a large clump. This is accomplished by slowly evaporating the solvent in the system by reducing the pressure above the solution while at the same time heating the solution at a low temperature to maintain the solubility of certain nanocrystals, i.e. palladium, in the evaporating solvent5. Since the vacuum created by the standard oil pump is far too strong for most of the solvents used (i.e. the rate of evaporation would be too high), a water aspirator was used instead (Figure 4). The suction created by the water aspirator is the result of the narrowing of the tube through which the water flows. The narrowing of the tube then causes the water to flow faster, which in turn creates a decrease in pressure that is harnessed to draw air through the side arm from the system there attached. The nature of this effect means that it is limited by the vapor pressure of the liquid flowing through the tube, in this case water. This then yields a vacuum of around 20 Torr that is far better suited for removing the solvent at a slow pace necessary than the 10 Torr vacuum created by a standard oil vacuum pump. Therefore, by means of this method 3 types of binary superlattices were attempted to be created: a lead selenide-palladium superlattice, an iron oxide-lead selenide and a lead selenide-cadmium selenide. 12 Figure 4. Schematic representation of a water aspirator18. Finally, although the concentration of the lead selenide and cadmium selenide nanocrystals can be calculated to some extent through their absorbance spectra and the experimentally determined equations (see sections 3.1.3 and 3.1.4), this is not the case for the iron oxide and palladium nanocrystals. Consequently, the molar ratios needed for the superlattices cannot be quantitatively calculated and a certain amount of experimental optimization is required. 2.3 Transmission Electron Microscopy (TEM) Sample Characterization All the nanocrystals synthesized were characterized using a FEI Tecnal T12 transmission electron microscope operating at 120 kV. As sample holder a 400 mesh PELCO Ultrathin Carbon Type-A copper grid was used. The nanocrystal samples were prepared by allowing a drop from a diluted dispersion solution to evaporate on the grid, making sure that the grid was protected from possible air contaminants by a glass casing. For the binary superlattices samples, the lattices were formed onto the grid themselves by submerging the grid into the binary solution and evaporating it with the grid inside. In order to maximize the likelihood of the lattice forming on the grid itself, 13 the vial containing the binary solution and the grid was inclined in a 30° angle with respect to the horizontal plane (Figure 5). Thus allowing for the nanocrystals to orderly arrange themselves onto the grid as the solvent slowly evaporates and pulls them downwardly along the surface of the grid. Figure 5. Schematic depiction of the sample preparation methods for characterizing the binary superlattices. The cartoon depicts a side cross-section view of a vial holding a TEM grid (red line) with an evaporating solvent (blue triangle) 19 2.4 Optical Characterization 2.4.1 UV-Vis Absorbance Sample Characterization Aside from being characterized by TEM, the cadmium selenide nanocrystals were also able to be characterized by UV-Vis absorption since they have a strong absorption peak is in that region 17. The UV-Vis absorption spectra was thus taken with an Ocean Optics USB2000 UV/Vis spectrophotometer. To ensure that the absorption peak was within the upper detection limits of the spectrophotometer, the already redispersed cadmium selenide nanocrystals were diluted again in tetrachloroethylene. Finally, quartz cuvettes were used as sample holders due to their ability to withstand tetrachloroethylene as well as to avoid attenuating the signal to the spectrophotometer. 2.4.2 Infrared Absorbance Sample Characterization 14 Similarly to cadmium selenide, lead selenide nanocrystals also exhibit optical properties that allow for their characterization through their infrared absorption spectra 12 14 Thus, their IR absorbance spectra was taken with a Hitachi U-4100 UV/Vis/Near-IR spectrophotometer. Once again, to ensure that the absorbance peak was within the upper detection limits of the spectrophotometer, the already redispersed lead selenide nanocrystals were diluted once more in tetrachloroethylene and quartz cuvettes were also used to take the IR absorption spectra of the lead selenide nanocrystals for the same reasons given above. 3. RESULTS & DISCUSSION 3.1 Nanocrystal Characterization 3.1.1 Palladium (Pd) 15 As stated above, the palladium nanocrystals were characterized via TEM to determine their approximate size and uniformity. The images obtained by TEM (Figure 6) show the quite satisfactory size uniformity that was expected for the synthesis method chosen. Although size selection methods can be used to increase the size uniformity, it was believed to be unnecessary to use those methods as the samples observed showed an acceptable size uniformity for the purposes of the project. Figure 6. TEM image of the palladium nanocrystals synthesized. 3.1.2 Gamma Iron (III) Oxide (y-Fe2O3) Similarly to the images obtained with palladium, the TEM images obtained for the y-Fe2O3 nanocrystals showed the same level of high size uniformity that was expected 16 from the synthesis method chosen and that was needed the for purposes of the project (Figure 7). Figure 7. TEM image of the y-Fe2O3 nanocrystals synthesized. Curiously, an interesting find was made during the TEM characterization of a sample of the y-Fe2O3 synthesized, when it was observed that some of the nanocrystals had a cubic shape instead of the expected, and previously-observed, spherical shape (Figure 8). After careful review of the sample preparation that lead to this sample, it was determined that the cubical nanocrystals observed were the result of the initial difficulty that was encountered when determining the reflux temperature for the secondary method used for the synthesis of the iron oxide nanocrystals. 17 Figure 8. TEM image of a sample of iron oxide nanocrystals in question. The sample observed (see also Figure 8), was an early sample under this new method and in the search for the correct reflux temperature the solution had been heated for far longer than originally required and had been held at various temperatures for an unspecified length of time, making the exact replication of the synthesis quite unlikely. Lastly, while a procedure for the specific synthesis of these a-Fe2O3 nanocubes was ultimately found in the work by Qin et a l , it was quite an involved synthesis and as such was not pursued. 3.1.3 Lead (II) Selenide (PbSe) As stated above, lead selenide absorbs quite strongly in the infrared region 12-13 and its absorption spectrum was quite useful in characterizing the nanocrystals synthesized. Based on the wavelength, X, in nanometers of the first absorption peak observed in the IR spectra, the diameter, D, also in nanometers, of the lead selenide nanocrystals in the solution was able to be determined based on equation (1) determined 18 experimentally by Yu et al.14. Similarly, once the size was determined, the molar absorptivity, s, in 105 M-1cm-1 units, was calculated with another equation (2)14. Lastly, from this piece of information and the relationship established by the Beer-Lambert Law (3), the concentration of the lead selenide nanocrystals was finally calculated. Where A is the absorbance of the sample in absorbance units, l is the path length of the sample cell in centimeters and C is the concentration in moles per liter. D = (X - 143.75) / 281.85 (1) s = (0.03389)D2-53801 (2) A = s ■C ■l (3) Figure 9. IR absorbance spectra of the spherical PbSe nanocrystals. Based on the equations from Yu et al.14, this sample has an average diameter of 9.2 nm, an extinction coefficient of 9.6x105 M-1cm-1 and a concentration of 0.95 ^M. Aside from the IR spectra obtained, the lead selenide nanocrystals, both the spherical as well as the nanocubes were also characterized via TEM (Figures 10-12). Once again the samples showed quite a high size uniformity and the spherical nanocrystals even arranged themselves in a hexagonally uniform lattice (Figure 12). 19 Figure 10. TEM image of the unexpectedly observed PbSe nanocubes. Review of the sample preparation revealed that they had been accidently synthesized at a higher temperature. Figure 11. TEM image of the purposefully synthesized PbSe nanocubes. Figure 12. TEM image of the spherically shaped PbSe nanocrystals. 20 3.1.4 Cadmium Selenide (CdSe) Similarly to the PbSe nanocrystals, the Peng group experimentally determined equations (3-4) that correlate the diameter, D, of the CdSe nanocrystals in nanometers to the wavelength of its first absorption peak, X, also in nanometers14. Likewise, with the UV-Vis absorbance spectra taken and through the Beer-Lambert Law, stated above, the concentration in moles per liter of the cadmium selenide nanocrystals was calculated. D = (1.6122 x 10-9)X4 - (2.6575 x 10-6)X3 + (1.6242 x 10-3)X2 - (0.4277)X + (41.57) e = (5857)D ,2.65 (3) (4) Figure 13. Absorbance of the CdSe nanocrystals. Based on the equations described above from the Peng group17, this sample has an average diameter of 3.3 nm, an extinction coefficient of 1.4x105 M1cm-1 and a concentration of 2.46 ^M. Lastly, aside from the UV-Vis absorbance characterization, the CdSe nanocrystals were also characterized via TEM. The images obtained then showed the great size uniformity that was sought for (Figure 14). 21 Figure 14. TEM image of the CdSe nanocrystals synthesized 3.2 Binary Superlattices 3.2.1 Iron Oxide-Lead Selenide (y-Fe2O3/PbSe) The first superlattice studied was the one formed by combining y-Fe2O3 nanocrystals with a 20-fold excess of PbSe nanocrystals. After a few attempts at optimizing the molar ratios needed for the superlattice, a range of a few hundred nanometers of tightly packed nanocrystals could be observed under TEM imaging (Figure 15). Figure 15. TEM image of a y-Fe2O3/PbSe superlattice sample showing a few hundred nanometers long section of tightly packed nanocrystals. However, closer inspection reveals that the formed arrangement showed strong disorder. Although the nanocrystals are in close contact, they do not appear to form any 22 ordered structured, but rather seem to be nonuniformly distributed across the TEM grid (Figure 16). Similarly, although the possibility that the two nanocrystals did not interact can be discarded by the absence of a hexagonally ordered lattice, the more geometrically ordered lattice expected of the proposed superlattice is also unobserved. While a hexagonally packed lattice is expected for spherical nanocrystals of similar size, a more cubic oriented lattice is expected of the binary superlattice whose spherical components differ in diamter5. Figure 16. TEM image of a y-Fe2O3/PbSe superlattice sample showing a more detailed view of the features of the arrangement of the nanocrystals. 3.2.2 Lead Selenide-Palladium (PbSe/Pd) The next superlattice studied was the one formed by combining Pd nanocrystals in an approximate 20-fold excess to PbSe nanocrystals. Once again, a few optimization steps of the molar ratios of the component nanocrystals were necessary in order to observe signs of the formation of the desired lattice. These presented themselves by forming pockets of the expected cubic arrangements of the nanocrystals in the binary lattice (Figure 17). These pockets can confidently be surmised to come from an optimized 23 local molar ratio that is not found in the entire sample. Even though the optimal conditions for the formation of the lattice are not found globally throughout the sample, a range of a couple of hundred nanometers long of the cubically arranged lattice was still able to be observed (Figure 18). Figure 17. TEM image of a PbSe/Pd superlattice sample showing rows of nanocrystals that indicate the expected cubic arrangement of the nanocrystals, as opposed to the hexagonal arrangement that would be expected in the ab sence of a lattice. 24 Figure 18. TEM image of the PbSe/Pd superlattice showing a range of a few hundred nanometers of the binary superlattice. 3.2.3 Lead Selenide-Cadmium Selenide (PbSe/CdSe) The last binary superlattice studied was the one formed by combining CdSe nanocrystals in a 12 to 1 molar ratio with PbSe nanocrystals. Although the nominal concentration of both the cadmium and lead selenide nanocrystals dispersion solutions can be found through their absorbance measurements (see above). The samples based on those concentration did not give rise to the long-range-ordered lattices that were expected. Therefore, a few optimization steps were required to obtain ordered superlattices. Importantly, the binary PbSe/CdSe superlattices (Figure 19) showed a distinctively better cubic arrangement than the samples obtained with the PbSe/Pd nanoparticles. 25 Figure 19. TEM image of the PbSe/CdSe superlattice sample showing the detailed cubic arrangement of the binary latticed formed. Another gratifying aspect of the PbSe/CdSe superlattice samples was that it not only ordering did not only occur in a range of a few nanometers, but over an extended 3D lattice (Figure 20). The 3D nature of the lattice can be interpreted from the increasingly darkness of the interior of the sample (towards the right of Figure 20). The darkness in the TEM image implies a sample thickness too large for the electrons to penetrate through. Therefore, in the image below where we can see the beginning of the lattice starting at the left of the image but it becomes increasingly difficult to make out as one moves to the right of it, it can be understood that the lattice continues and that its thickness increases to the point where it surpasses the penetrating capacity of the electron beam. The same effect can also be seen to a smaller extent in the PbSe/Pd superlattice sample (Figure 18), where different areas of the lattice depicted are darker than other, indicating an increased thickness of the sample. 26 Figure 20. TEM image of the PbSe/CdSe superlattice sample showing an ordered range of a couple hundred nanometers. Note that the borderline seen on the left is the border of the sample. 4. CONCLUSIONS Based on a proposal on the possibility of high-temperature superconductivity through nanoscale architecture by Dr. Mattis1, a nanocrystal-based platform has been developed through which experimental evidence for this exciting idea can now be pursued. The ability of nanocrystals to self-assemble into binary superlattices that meet all the parameters of Dr. Mattis’ calculations enables them to be used for an adequate and meaningful experimental test of his ideas. Nanocrystals from different compounds and with different sizes were successfully prepared using solution chemistry based colloidal crystals nucleation and growth 27 techniques. All nanocrystals showed uniform sizes and well defined shapes (spherical and cubic, depending on synthesis conditions). Various nanocrystal combination (compounds and sizes) were combined at different molar ratios to form binary aggregates. Several combinations were identified that lead to successful formation of ordered superlattices with cubic geometries. Current work is focused on achieving higher uniformity and increased ordering ranges (from hundreds of nanometers to tens of micrometers). The work of this thesis paves the path to experimentally test the superconductive properties of nanoscale superlattices of semiconductors supported by Dr. Mattis’ theoretical calculations. The next phase towards achieving the goal of superconductivity will include integration of the superlattices into device structures and carrying out temperature-controlled conductivity measurements on these structures. 5. REFERENCES 1. Mattis, Daniel C. “Magnetism and superconductivity in nanoarchitectures”. Physica B. 2006, 384, 239-243. 2. Crystal Structure Gallery. Hiroi Research Group, Institute of Solid State Physics, University of Tokyo. < http://hiroi.issp.u-tokyo.ac.jp/saito/Gallery.html> (Accessed March 13, 2013). 3. Sjostrom, Travis; Mattis, Daniel C.; Yin, Wei-Guo; Ku, Wei. “Electronic Properties of Thin Film Periodic Structures”. Journal o f Computational and Theoretical Nanoscience. 2008, 6, 1-15. 4. Hyeon, Taeghwan; Jang, Youngjin; Kwon, Soon Gu; Park, Jongnam. “Synthesis of Monodisperse Spherical Nanocrystals”. Angew. Chem. Int. Ed. 2007, 46, 4630-4660. 5. Murray, Christopher M.; Shevchenko, Elena V; Talapin, Dmitri V; O’Brien, Stephen. “Polymorphism in AB13Nanoparticle Superlattices: An Example of Semiconductor-Metal Metamaterials”. J. Am. Chem. Soc. 2005, 127, 8741-8747. 28 6. Vanmaekelbergh, Daniel; Overgaag, Karin; Evers, Wiel; de Nijs, Bart; Koole, Rolf; Meeldjik, Johannes. “Binary Superlattices of PdSe and CdSe Nanocrystals”. J. Am. Chem. Soc. 2008, 130, 7833-7835. 7. Carion, Olivier; Mahler, Benoit; Pons, Thomas; Dubertret, Benoit. “Synthesis, encapsulation, purification and coupling of single quantum dots in phospholipid micelles for their use in cellular and in vivo imaging”. Nature Protocols. 2007, 2(10), 2383-2390. 8. Hyeon, Taeghwan; Lee, Su Seong; Park, Jongnam; Chung, Yunhee; Na, Hyon Bin. “Synthesis of Highly Crystalline and Monodisperse Maghemite Nanocrystallites without Size-Selection Process”. J. Am. Chem. Soc. 2001, 123, 12798-12801. 9. Qin, Wenqing; Yang, Congren; Yi, Ran; Gao, Guanhua. “Hydrothermal Synthesis and Characterization of Single-Crystalline a-Fe2O3 Nanocubes”. Journal o f Nanomaterials. vol. 2011, Article ID 159259, 5 pages, 2011. doi: 10.1155/2011/159259. 10. Houtepen, Arjan J.; Koole, Rolf; Vanmaekelbergh, Daniel; Meeldjik, Johannes; Hickey, Stephen G. “The Hidden Role of Acetate in the PbSe Nanocrystal Synthesis”. J. Am. Chem. Soc. 2006, 128, 6792-6793. 11. Fang, Jiye; Lu, Weigang; Ding, Yong; Wang, Zhong Lin. “Formation of PdSe Nanocrystals: A Growth towards Nanocubes”. J. Phys. Chem. B. 2005, 109, 19219­ 19222. 12. Bawendi, Moungi G.; Bulovic, Vladimir; Steckel, Jonathan S. “ 1.3 p,m to 1.55 p,m Tunable Electroluminescence from PbSe Quantum Dots Embedded within an Organic Device”. Adv. Mater. 2003, 15(21), 1862-1866. 13. Murray, Christopher B.; Wise, Frank W.; Hyen, Byung-Ryool; Bartnik, A.C.; Koh, Weon-kyu; Agladze, N.I.; Wrubel, J.P.; Sievers, A.J. “Far-Infrared Absorption of PbSe Nanorods”. Nano Lett. 2011, 11, 2786-2790. 14. Yu, William W.; Dai, Quanqin; Wang, Yingnan; Li, Xinbi; Zhang, Yu; Pellegrino, Donald J.; Zhao, Muxun; Zou, Bo; Seo, JaeTae; Wang, Yiding. “Size-Dependent Composition and Molar Extinction Coefficient of PbSe Semiconductor Nanocrystals”. ACS Nano. 2009, 3(6), 1518-1524. 15. Donega, Celso de Mello; Hickey, Stephen G.; Wuister, Sander F.; Vanmaekelbergh, Daniel; Meijerink, Andries. “Single-Step Synthesis to Control the Photoluminescence Quantum Yield and Size Dispersion of CdSe Nanocrystals”. J. Phys. Chem. B. 2003, 107, 489-496. 16. Peng, Xiaogang; Wang, Y. Andrew; Aldana, Jose. “Photochemical Instability of CdSe Nanocrystal Coated by Hydrophilic Thiols”. J. Am. Chem. Soc. 2001, 123, 8844-8850. 29 17. Yu, William W; Qu, Lianhua; Guo, Wenzhuo; Peng, Xiaogang. “Experimental Determination of the Extinction Coefficient of CdTe, CdSe, and CdS Nanocrystals”. Chem. Mater. 2003, 15, 2854-2860. 18. Zubrick, James W. “The Water Aspirator: A Vacuum Source”. What-When-How.com <http://what-when-how.com/organic-chemistry-laboratory-survivalmanual/recrystallization-part-2-laboratory-manual/> (Accessed March 16, 2013). 19. Vanmaekelbergh, Daniel; Overgaag, Karin; Evers, Wiel; de Nijs, Bart; Koole, Rolf; Meeldjik, Johannes. “Binary Superlattices of PdSe and CdSe Nanocrystals Supporting Information”. Pubs.ACS.org <http://pubs.acs.org/doi/suppl/10.1021/ja802932m/suppl_file/ja802932mfile001.pdf> (Accessed March 16, 2013).