First-principles Theoretical Studies of Carbon Nanomaterials and New Data Analysis Methods for Catalytic Platinum Microswimmers

Open Access
- Author:
- Byun, Young-moo
- Graduate Program:
- Physics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 22, 2012
- Committee Members:
- Vincent Henry Crespi, Dissertation Advisor/Co-Advisor
Vincent Henry Crespi, Committee Chair/Co-Chair
Milton Walter Cole, Committee Member
Renee Denise Diehl, Committee Member
Eric M Mockensturm, Committee Member
Richard Wallace Robinett, Committee Member - Keywords:
- Fullerene
Fluorographene
Density Functional Theory
GW Approximation
Microswimmer
Chemotaxis - Abstract:
- This dissertation has two main topics: carbon nanomaterials and platinum microswimmers. First, carbon nanomaterials such as C$_{60}$, carbon nanotubes, and graphene have extraordinary physical properties that can be exploited for a wide range of applications, and thus have been studied extensively by a variety of experimental and theoretical techniques. In this dissertation, we study doping, defect, and many-body effects on the structural and electronic properties of carbon nanomaterials using a first-principles (\textit{ab initio}) approach (i.e. without using empirical parameters). Second, self-propelled microswimmers such as catalytically driven platinum microrods have a potential to be used as novel drug delivery vehicles. Many experiments on microswimmers have been done, but there has been no quantitative analysis on experimental data so far because different motions such as self-propelled and Brownian motions take place at the same time, leading to wrong interpretations of experimental results often. In this dissertation, we present novel data analysis methods for self-propelled microswimmers. In Chapter 3, we study the doping effects of alkali and alkaline-earth atoms and C$_{8}$H$_{8}$ molecules on the structural, electronic, and superconducting properties of solid C$_{60}$ using density functional theory (DFT). We show that there is no charge transfer between C$_{8}$H$_{8}$ and C$_{60}$ molecules in a C$_{8}$H$_{8}$/C$_{60}$ co-crystal using band-structure calculations. Using total-energy calculations, we also show that alkali atoms can fill the tetrahedral voids in a C$_{8}$H$_{8}$/C$_{60}$ co-crystal, leading to the formation of A$_{2}$(C$_{8}$H$_{8}$)C$_{60}$ (A = K, Rb, Cs). Finally, we show that ABa(C$_{8}$H$_{8}$)C$_{60}$ and ALi$_{2}$(C$_{8}$H$_{8}$)C$_{60}$ (A = K, Rb, Cs) -- hypothetical three-electron-doped fcc (C$_{8}$H$_{8}$)C$_{60}$ co-crystals with the same crystal structure and doping level as superconducting A$_{3}$C$_{60}$ (A = K, Rb) -- might have superconducting transition temperatures ($T_{c}$) of up to $\sim$50 K using McMillan formula. Recently synthesized fluorographene, fully fluorinated graphene in a chair configuration, is a wide band-gap ($E_{g}$) semiconductor with an optical band gap of $\sim$3 eV. However, accurate first-principles calculations have shown that pristine fluorographene should have $E_{g}$ of 5.4--7.5 eV. To explain this discrepancy, in Chapter 4, we study the defect effects of F vacancies, a Stone-Wales (SW) defect, C single vacancies and C double vacancies on $E_{g}$ of fluorographene using density functional theory and many-body perturbation theory within the $GW$ approximation. We show that F vacancies and a SW defect are not likely to affect $E_{g}$ of fluorographene, whereas a C single vacancy with a doubly fluorinated C atom, a C double vacancy, and a C double vacancy with two doubly fluorinated C atoms, which are energetically more favorable than other C vacancies at a wide range of chemical potential of F ($\mu_{F}$), lead to a \textit{GW} band gap of 3.8--4.8 eV, which is in good agreement with the optically measured $E_{g}$ ($\sim$3 eV). The remaining 0.8--1.8 eV difference is attributed to strongly bound excitons, which are neglected in the $GW$ approximation. Whereas photoemission and inverse photoemission spectra (PES and IPES) for C$_{60}$ highest-occupied-molecular-orbital (HOMO)- and lowest-unoccupied-molecular-orbital (LUMO)-derived bands of pristine and alkali-intercalated C$_{60}$ indicate that the mixing of C$_{60}$ and alkali electronic states takes place, DFT band-structure calculations show that alkali atoms simply act as electron donors into the mostly rigid C$_{60}$ LUMO-derived band. To resolve this discrepancy, in Chapter 5, we study the correlation effects in A$_{3}$C$_{60}$ and A$_{6}$C$_{60}$ (A = K, Rb, Cs) using the $GW$ approximation. Our $GW$ quasiparticle band structures show that C$_{60}$ HOMO- and LUMO-derived bands broaden by $\sim$10\% and $\sim$30\%, respectively, upon alkali intercalation in both fcc and A15 A$_{3}$C$_{60}$ as compared to fcc and sc C$_{60}$ with the same lattice constants, which is qualitatively consistent with the PES and IPES experiments. Our $GW$ calculations using experimental results show that a Mott-Hubbard metal-insulator transition takes place when the C$_{60}$ LUMO-derived band width is $\sim$0.43 and $\sim$0.75 eV for fcc and A15 A$_{3}$C$_{60}$, respectively, in good agreement with quantum Monte Carlo (QMC) calculations. %we calculate quasiparticle energies of face-centered-cubic (fcc), simple-cubic (sc), and body-centered-cubic (bcc) C$_{60}$, fcc and A15 A$_{3}$C$_{60}$ (A = K, Rb, Cs), and bcc A$_{6}$C$_{60}$ Recently, the first example of non-biological chemotaxis -- a directed movement towards or away from chemicals -- was reported: $\sim$2-$\mu$m-long catalytically-driven colloidal platinum/gold rods moved towards a hydrogen peroxide (H$_{2}$O$_{2}$)-soaked hydrogel, which was assumed to be a region of high H$_{2}$O$_{2}$ concentration. To find the mechanism for chemotaxis of Pt/Au microrods, in Chapter 6, we develop and implement new quantitative data analysis algorithms. Using our data analysis algorithms, we (i) decompose the Pt/Au microrod's motion into self-propelled, drift, and Brownian components successfully, (ii) obtain correct and accurate values for self-propulsion speed, drift speed, and diffusion coefficients, and (iii) find any possible correlation between self-propelled and drift motions. Our data analysis results show that the Pt/Au microrods' drift towards the H$_{2}$O$_{2}$-soaked hydrogel is not active chemotaxis but passive convection by a bulk flow towards the hydrogel, which is verified by new experiments. This proves that our data analysis algorithms are a reliable and accurate tool in the microswimmers research.