Publications

Recombinant retroviruses encoding human adenosine deaminase (ADA; adenosine aminohydrolase, EC 3.5.4.4) have been used to infect murine hematopoietic stem cells. In bone marrow transplant recipients reconstituted with the genetically modified cells, human ADA was detected in peripheral blood mononuclear cells of the recipients for at least 6 months after transplantation. In animals analyzed in detail 4 months after transplantation, human ADA and proviral sequences were detected in all hematopoietic lineages; in several cases, human ADA activity exceeded the endogenous activity. These studies demonstrate the feasibility of introducing a functional human ADA gene into hematopoietic stem cells and obtaining expression in multiple hematopoietic lineages long after transplantation. This approach should be helpful in designing effective gene therapies for severe combined immunodeficiency syndromes in humans.

The existence of new compounds is often postulated by solid state chemists by replacing an ion in the crystal structure of a known compound by a chemically similar ion. In this work, we present how this new compound discovery process through ionic substitutions can be formulated in a mathematical framework. We propose a probabilistic model assessing the likelihood for ionic species to substitute for each other while retaining the crystal structure. This model is trained on an experimental database of crystal structures, and can be used to quantitatively suggest novel compounds and their structures. The predictive power of the model is demonstrated using cross-validation on quaternary ionic compounds. The different substitution rules embedded in the model are analyzed and compared to some of the traditional rules used by solid state chemists to propose new compounds (e.g., ionic size).

Based on our findings from an ongoing pilot with shared computers in rural primary schools in Uttar Pradesh, India since 2001 and two follow-up field studies in the same schools in 2004, we propose some principles for designing educational technology for developing countries as hypotheses. They include constructivist, small-group collaborative learning through digital storytelling and integrating paper- with computer-based practices. We also discuss the need to balance pedagogy, usability and mobility.

We have applied CZE and CEC in microfluidic devices to a number of analytical problems including the analysis of chemical warfare agents, polycyclic aromatic hydrocarbons, and proteins. In addition to developments in the areas of separations, we are also heavily involved in the "front-end" of analytical systems particularly in the area of sample preconcentration. We have a particular interest in the concentration of a wide range of molecules including small molecules, biopolymers, and cells. Our current toolbox includes a microfluidic platform known as Liquid Phase ChemLab developed by a team at Sandia for protein analysis. This platform has been demonstrated with protein biotoxins, CW agent degradation products, protein preconcentration, and is suitable for other targets. Additionally, electrodeless dielectrophoresis technology developed at Sandia is an attractive alternative to filters for cell concentration.

Conversion of carbon dioxide to C₂-C₄ hydrocarbons is a major pursuit in clean energy research. Despite tremendous efforts, the lack of well-defined catalysts in which the spatial arrangement of interfaces is precisely controlled hinders the development of more efficient catalysts and in-depth understanding of reaction mechanisms. Herein, we utilized the strategy of tandem catalysis to develop a well-defined nanostructured catalyst CeO₂-Pt@mSiO₂-Co for converting CO₂ to C₂-C₄ hydrocarbons using two metal-oxide interfaces. C₂-C₄ hydrocarbons are found to be produced with high (60%) selectivity, which is speculated to be the result of the two-step tandem process uniquely allowed by this catalyst. Namely, the Pt/CeO₂ interface converts CO₂ and H₂ to CO, and on the neighboring Co/mSiO₂ interface yields C₂-C₄ hydrocarbons through a subsequent Fischer-Tropsch process. In addition, the catalysts show no obvious deactivation over 40 h. The successful production of C₂-C₄ hydrocarbons via a tandem process on a rationally designed, structurally well-defined catalyst demonstrates the power of sophisticated structure control in designing nanostructured catalysts for multiple-step chemical conversions.

With the growing adoption of short-form video by social media platforms, reducing the spread of misinformation through video posts has become a critical challenge for social media providers. In this paper, we develop methods to detect misinformation in social media posts, exploiting modalities such as video and text. Due to the lack of large-scale public data for misinformation detection in multi-modal datasets, we collect 160,000 video posts from Twitter, and leverage self-supervised learning to learn expressive representations of joint visual and textual data. In this work, we propose two new methods for detecting semantic inconsistencies within short-form social media video posts, based on contrastive learning and masked language modeling. We demonstrate that our new approaches outperform current state-of-the-art methods on both artificial data generated by random-swapping of positive samples and in the wild on a new manually-labeled test set for semantic misinformation.

Achieving stable operation of photoanodes used as components of solar water splitting devices is critical to realizing the promise of this renewable energy technology. It is shown that p-type transparent conducting oxides (p-TCOs) can function both as a selective hole contact and corrosion protection layer for photoanodes used in light-driven water oxidation. Using NiCo2O4 as the p-TCO and n-type Si as a prototypical light absorber, a rectifying heterojunction capable of light driven water oxidation was created. By placing the charge separating junction in the Si using a np(+) structure and by incorporating a highly active heterogeneous Ni-Fe oxygen evolution catalyst, efficient light-driven water oxidation can be achieved. In this structure, oxygen evolution under AM1.5G illumination occurs at 0.95 V vs RHE, and the current density at the reversible potential for water oxidation (1.23 V vs RHE) is >25 mA cm(-2). Stable operation was confirmed by observing a constant current density over 72 h and by sensitive measurements of corrosion products in the electrolyte. In situ Raman spectroscopy was employed to investigate structural transformation of NiCo2O4 during electrochemical oxidation. The interface between the light absorber and p-TCO is crucial to produce selective hole conduction to the surface under illumination. For example, annealing to produce more crystalline NiCo2O4 produces only small changes in its hole conductivity, while a thicker SiOx layer is formed at the n-Si/p-NiCo2O4 interface, greatly reducing the PEC performance. The generality of the p-TCO protection approach is demonstrated by multihour, stable, water oxidation with n-InP/p-NiCo2O4 heterojunction photoanodes.

We investigated the structural stability and electrochemical properties of LiMnBO3 in the hexagonal and monoclinic form with ab initio computations and, for the first time, report electrochemical data on monoclinic LiMnBO3 . In contrast to the negligible Li-storage capacity in the hexagonal LiMnBO3 , a second cycle discharge capacity of 100mAh/g was achieved in the monoclinic LiMnBO3 , with good capacity retention over multiple cycles. Elevated temperature cycling indicates that the capacity of monoclinic LiMnBO3 is kinetically limited, and further improvement may be expected by addressing the Li ion and/or electron transport limitations.

A chiral phosphate ion plays a dual role, coaxing a fluorinating agent into solution and then catalyzing its reaction there.