Theoretical investigation of solid hydrogen and deuterium
View/ Open
Date
29/11/2016Author
Magdau, Ioan-Bogdan
Metadata
Abstract
Solid hydrogen forms at extreme conditions, under high pressures. Although
the hydrogen atom is easy to understand theoretically, when interacting in the
solid state it becomes complicated. Up to now, five different solid phases have
been confirmed experimentally and theory has predicted numerous competing
crystal candidates. The goal is to obtain solid metallic hydrogen which has been
predicted theoretically eighty years ago and has since been considered the holy
grail of high pressure science. In nature, this form of matter is believed to exist
at the core of large planets like Jupiter and Saturn, being responsible for the
planets' large magnetic fields. Understanding the different phases of hydrogen is
a test for our most advanced theories of quantum mechanics in condensed matter
and it is fundamentally important for both planetary and material science.
Recently discovered solid phase IV is stabilized by entropy and therefore only
exists at relatively high temperatures. Using molecular dynamics (MD) I studied
the room temperature behavior of phase IV starting with the ground state
candidate structures reported in the literature. Additionally, I devised a velocity
projection method for extracting Raman spectra from MD in light of direct
comparison to experiment. My results helped establish the true nature of phase
IV and validated the structure against experimental data. Applying the same
method to the previously proposed C2=c crystal structure, I obtained results
that confirm this structure is the best candidate for phase III.
Within the last year, a new phase V of solid hydrogen was discovered in Raman
experiments. While attempting to identify the crystal structure associated with
this new phase, I discovered a manifestation of solid hydrogen in the form of
long polymeric chains that could be stabilized by a charge density wave. Here I
discuss the possibility of such a state of matter as an intermediate on the path
to molecular dissociation of hydrogen. Chains could, however, be a spurious
structure - the effect of a subtle non-convergence problem in the MD, which could
indicate serious issues with many previous studies reported in the literature. A
far more likely candidate for phase V is a structure similar to that of phase IV
with a subtle dynamical modification. I will present Raman and phonon results
from both static and dynamic calculations to support this claim. I conclude my
work on pure solid hydrogen with an instructive model that could explain the
entire phase diagram based on simple thermodynamic considerations. All of the
assumptions were extracted from our previous ab initio studies through analysis
and observations. This model encodes a comprehensive summary of the current
understanding of solid hydrogen at high pressures.
Raman and infrared spectroscopy have been the methods of choice in most
hydrogen studies. Another way to look at the problem is to analyze the
behavior of isotopic mixtures: hydrogen-deuterium binary alloys. Using isotopic
substitutions, I revealed a textbook effect in hydrogen: phonon localization
by mass disorder. The effect might be unique to this element, owing to the
large mass ratio between hydrogen and deuterium. Phonon localization explains
the complicated Raman spectra obtained experimentally in hydrogen-deuterium
mixtures at various concentrations. More recent experimental results claim an
unexpected phase transition in mixtures at low temperatures based on splittings
in the infrared spectra. Here I will show that the infrared splitting seen
experimentally could be induced by mass disorder in phase III and does not
necessarily indicate a structural transformation.