Mechanistic Implications of the Varying Susceptibility of PAHs to Pyro-Catalytic Treatment as a Function of Their Ionization Potential and Hydrophobicity.

Transition metal catalysts in soil constituents (e.g., clays) can significantly decrease the pyrolytic treatment temperature and energy requirements for efficient removal of polycyclic aromatic hydrocarbons (PAHs) and, thus, lead to more sustainable remediation of contaminated soils. However, the catalytic mechanism and its rate-limiting steps are not fully understood. Here, we show that PAHs with lower ionization potential (IP) are more easily removed by pyro-catalytic treatment when deposited onto Fe-enriched bentonite (1.8% wt. ion-exchanged content). We used four PAHs with decreasing IP: naphthalene > pyrene > benz(a)anthracene > benzo(g,h,i)perylene. Density functional theory (DFT) calculations showed that lower IP results in stronger PAH adsorption to Fe(III) sites and easier transfer of π-bond electrons from the aromatic ring to Fe(III) at the onset of pyrolysis. We postulate that the formation of aromatic radicals via this direct electron transfer (DET) mechanism is the initiation step of a cascade of aromatic polymerization reactions that eventually convert PAHs to a non-toxic and fertility-preserving char, as we demonstrated earlier. However, IP is inversely correlated with PAH hydrophobicity (log Kow), which may limit access to the Fe(III) catalytic sites (and thus DET) if it increases PAH sorption to soil OM. Thus, ensuring adequate contact between sorbed PAHs and the catalytic reaction centers represents an engineering challenge to achieve faster remediation with a lower carbon footprint via pyro-catalytic treatment.

Combined Surface-Enhanced Raman and Infrared Absorption Spectroscopies for Streamlined Chemical Detection of Polycyclic Aromatic Hydrocarbon-Derived Compounds.

Polycyclic aromatic hydrocarbons (PAHs) constitute a class of universally prevalent carcinogenic environmental contaminants. It is increasingly recognized, however, that PAHs derivatized with oxygen, sulfur, or nitrogen functional groups are frequently more dangerous than their unfunctionalized counterparts. This much larger family of chemicals─polycyclic aromatic compounds─PACs─is far less well characterized than PAHs. Using surface-enhanced Raman and IR Absorption spectroscopies (SERS + SEIRA) combined on a single substrate, along with density functional theoretical (DFT) calculations, we show that direct chemical detection and identification of PACs at sub-parts-per-billion concentration can be achieved. Focusing our studies on 9,10-anthraquinone, 5,12-tetracenequinone, 9-nitroanthracene, and 1-nitropyrene as model PAC contaminants, detection is made possible by incorporating a hydroxy-functionalized self-assembled monolayer that facilitates hydrogen bonding between analytes and the SERS + SEIRA substrate. 5,12-Tetracenequinone was detected at 0.3 ppb, and the limit of detection was determined to be 0.1 ppb using SEIRA alone. This approach is straightforwardly extendable to other families of analytes and will ultimately facilitate fieldable chemical detection of these dangerous yet largely overlooked environmental contaminants.

Pyro-Catalytic Degradation of Pyrene by Bentonite-Supported Transition Metals: Mechanistic Insights and Trade-Offs with Low Pyrolysis Temperature.

Transition metal catalysts can significantly enhance the pyrolytic remediation of soils contaminated with polycyclic aromatic hydrocarbons (PAHs). Significantly higher pyrene removal efficiency was observed after the pyrolytic treatment of Fe-enriched bentonite (1.8% wt ion-exchanged content) relative to natural bentonite or soil (i.e., 93% vs 48% and 4%) at the unprecedentedly low temperature of 150 °C with only 15 min treatment time. DFT calculations showed that bentonite surfaces with Fe3+ or Cu2+ adsorb pyrene stronger than surfaces with Zn2+ or Na+. Enhanced pyrene adsorption results from increased charge transfer from its aromatic π-bonds to the cation site, which destabilizes pyrene allowing for faster degradation at lower temperatures. UV-Vis and GC-MS analyses revealed pyrene decomposition products in extracts of samples treated at 150 °C, including small aromatic compounds. As the pyrolysis temperature increased above 200 °C, product distribution shifted from extractable compounds to char coating the residue particles. No extractable byproducts were detected after treatment at 400 °C, indicating that char was the final product of pyrene decomposition. Tests with human lung cells showed that extracts of samples pyrolyzed at 150 °C were toxic; thus, high removal efficiency by pyrolytic treatment does not guarantee detoxification. No cytotoxicity was observed for extracts from Fe-bentonite samples treated at 300 °C, inferring that char is an appropriate treatment end point. Overall, we demonstrate that transition metals in clay can catalyze pyrolytic reactions at relatively low temperatures to decrease the energy and contact times required to meet cleanup standards. However, mitigating residual toxicity may require higher pyrolysis temperatures.

Multi-Task Network Representation Learning.

Networks, such as social networks, biochemical networks, and protein-protein interaction networks are ubiquitous in the real world. Network representation learning aims to embed nodes in a network as low-dimensional, dense, real-valued vectors, and facilitate downstream network analysis. The existing embedding methods commonly endeavor to capture structure information in a network, but lack of consideration of subsequent tasks and synergies between these tasks, which are of equal importance for learning desirable network representations. To address this issue, we propose a novel multi-task network representation learning (MTNRL) framework, which is end-to-end and more effective for underlying tasks. The original network and the incomplete network share a unified embedding layer followed by node classification and link prediction tasks that simultaneously perform on the embedding vectors. By optimizing the multi-task loss function, our framework jointly learns task-oriented embedding representations for each node. Besides, our framework is suitable for all network embedding methods, and the experiment results on several benchmark datasets demonstrate the effectiveness of the proposed framework compared with state-of-the-art methods.

3D protein structure prediction of influenza A virus based on optimization genetic algorithm.

The 3D structure of close polymer is constituted by the interaction of close contact couples among amino acid residues. In this paper, 3D protein structure of influenza A virus was predicted. Twenty kinds of amino acid residues were divided into four categories according to the number of close contact couples. The stable structure with minimum energy was obtained by using optimization genetic algorithm. The HNXP 3D lattice model was established to predict the 3D protein structure. It can be concluded that the two kinds of structures are significantly similar by computing the similarity.