Olivier C. Gagné
Banting and Carnegie Postdoctoral Fellow
Earth and Planets Laboratory, Carnegie Institution for Science
Current research
Modeling semiconductor doping and quantifying dopant viability
Stronger than ever is the link between humankind’s welfare and prosperity and the state of scientific research on semiconductors – those simple solids at the core of solar cells, battery electrodes, electronics, and many more technologies we've come to depend on. Next-generation semiconductor design has become an evident solution to Earth’s energy crisis and worsening climate, allowing us to safely harness and store the sun’s energy without depleting Earth’s finite resources. How quickly these problems come to pass will depend on their allocated research intensity, and of course, serendipity. How do we maximize our chances for breakthroughs in semiconductor design, and where are these breakthroughs most likely to come from?
With rather simple bulk compositions, the bottleneck in designing efficient semiconductors rests in their doping by foreign ions, i.e. the process which enables their useful properties. The problem is that the process of doping is severely clouded by our incomplete understanding of chemical bonding in solids, which lags well behind the theoretical breakthroughs made in first-principles calculations over the past decades. Today, doping studies rely exclusively on lengthy first-principles calculations, with little guidance available to computationalists on where to direct their resources. An atomistic understanding of the factors underlying ion substitutions in solids would remedy this problem, allowing proactive calculations to be made for host/dopant combinations prior to their synthesis, as opposed to spending costly resources on rationalizing why certain combinations simply cannot work.
In this project, my goal is to investigate the various factors underlying ion substitutions in solids, and to quantify their relative importance to build a universal, straightforward, atomistic model of ion substitution that will be used to quantitatively assess dopant viability for any type of inorganic semiconductor. To this end, I bring intimate familiarity with the near entirety of stereochemical bonding behaviors observed across the periodic table (see Published Work), and extensive experience dealing with highly complex combinations of compositions, crystal structures and external factors from the mineral world. The end goal of this project is no less than to make silicon technologically obsolete.
Elucidating asymmetric bonding behavior in solids
The design of new materials with targeted properties is a fundamental driver of innovation in the materials sciences. In solid-state chemistry, rational design begins at the atomic level with the elucidation of structure-property relationships which depend on atomic arrangement and chemical bonding. Because a great deal of material properties arise from ions with asymmetric bonding behavior (e.g. ferroelectricity, piezoelectricity, non-linear optical behavior), it is crucial to understand the fundamental reasons underlying anomalous bonding environments in solids.
My interest lies in the identification, quantification and particularly, prediction of symmetry-breaking bonding phenomena in inorganic solids. An important component of my research is the systematization of chemical-bonding behavior for whole classes of compounds (e.g., oxides, nitrides) via bond-length dispersion analysis, which facilitates tracing back anomalous bonding behavior to structural, electronic and/or bond-topological effects. Gaining a higher understanding of the origin(s) of polyhedral distortion allows one to pinpoint the cause of material properties, and to determine if, and how, optimization of these properties may be undertaken. Furthermore, understanding the extent (and precedence) for which symmetry-breaking phenomena materialize into bond-length variations is crucial to maximize the harnessing of these effects within the constraints of physically realistic crystal structures. Following extensive work on oxide and nitride materials, I am currently working on a large-scale bond-length dispersion analysis for chalcogenide materials (with S2-, Se2- and/or Te2- as the main anion), with emphasis on the compositional and structural aspects underpinning their functional properties.
Right: Vibronic mixing (pseudo Jahn-Teller effect) between relevant HOMO-LUMO orbitals resulting in energetically favorable displacement of Ti4+ toward a vertex (O2-), possibly affording parent structures functional properties associated with non-centrosymmetric behavior.
Exploring uncharted chemical spaces
In the past decade, the landscape of new materials discovery evolved from one of trial-and-error and serendipity to one of data-driven, high-throughput computational exploration. While such methods play a critical role in fast-tracking materials discovery via the identification of “missing” compounds and the calculation of their properties, they are limited to processing data derived from known chemical and structural spaces. Today, rapid increase in available computational power has by-and-large transformed the problem of in silico exploration from one of computational feasibility to one of a priori identifying compositional spaces of interest, for which crystal-chemical intuition and know-how are valuable skills.
My interest in this matter lies in the identification of anomalously bonded coordination units bearing functional properties, and their transposing into new chemical spaces for the exploitation and optimization of their properties. Such functional coordination units may include cations prone to vibronic mixing (leading to the Jahn-Teller and pseudo Jahn-Teller effects), lone-pair stereoactive electrons, multiple (double or triple) metal-ligand bonds, etc., while new chemical spaces may feature new chemical environments, altered functional coordination units not limited to anion identity, coordination number and ligand type, new crystal structures, etc. In essence, the challenge is to reduce the categorically intractable totality of compositional and structural spaces into a handful of promising ones bearing functional properties, using chemical foresight. I am currently exploring new phenomenological spaces in inorganic nitrides, with potential material properties spanning the catalysis of nitrogen fixation, ultra-low thermal conductivity, ferroelectricity, ferromagnetism, magnetic-dielectric bistability, switchable catalytic behavior, improved electrochemical performance via “opening” of diffusion channels, etc. I am further implicated in the exploration of new chemical spaces in the thriving class of chalcogenide materials.
Left: Scheme of promising, uncharted chemical spaces in inorganic nitrides.